prof. dr. ir. guido van huylenbroeck prof. anne de paepe · use of rnai to evaluate the role of...
TRANSCRIPT
Promotor: Prof. dr. ir. Erik Van Bockstaele
Department of Plant Production, Faculty of BioScience Engeneering
Ghent University
Dean FBW: Prof. dr. ir. Guido Van Huylenbroeck
Rector UGent: Prof. Anne De Paepe
Use of RNAi to evaluate the role of isoflavones in resistance of red clover to
Sclerotinia trifoliorum and Sclerotium rolfsii
Khosro Mehdi Khanlou
Thesis submitted in fulfillment of the requirements for the degree of Doctor (PhD) in Applied
Biological Sciences (Biotechnology)
To refer to the present thesis:
Mehdi Khanlou K. 2013 Use of RNAi to evaluate the role of isoflavones in resistance of red
clover to Sclerotinia trifoliorum and Sclerotium rolfsii. PhD Thesis, Faculty of Bioscience
Engineering, Ghent University, Belgium.
ISBN-number: 978-90-5989-656-7
The author and promotor give the authorisation to consult and to copy parts of this work for
personal use only. Any other use is limited by the Laws of Copyright. Permission to
reproduce any material contained in this work should be obtained from the author.
Acknownledgement
When my time in Ghent is going to get over, everything which had happened to me during my
stay comes back to me. I have realized that I had a pleasant and happy stay in a peaceful city
in Ghent surrounded by supportive and friendly people. Ghent would be my second
motherland. It becomes in my soul when I am leaving. Here I would like to express my
deepest appreciation to the people without whom this endeavor would not have been
accomplished.
First of all I would like to thank my promotor, Prof. dr. ir. Erik Van Bockstaele, for his
support during this scientific survival trip. Particularly for the freedom I received and the
possibility I got to start this new line of research with RNAi to evaluate the role of isoflavones
in resistance of red clover to Sclerotinia trifoliorum and Sclerotium rolfsii. I also enjoyed and
learned a lot from research activities in the lab and the labs around.
I would like to thank my dear colleagues ing. Sabine Van Glabeke, Laurence Desmet, Nancy
Mergan, Katrien Liebaut, Carina Pardon, Ronald Van Den Oord, Kristien Janssens, Ariane
Staelens, ir. Kris Van Poucke, ir. Joost Baert, dr. Gerda Cnops, dr. ir. Kurt Heungens who all
together created a fantastic working atmosphere in the lab. By sharing the joyful and the
stressful moments, I had an unforgettable time with you. My special thanks to Miriam
Levenson, who spent a lot of time to correct my papers and thesis.
I would like to thank the secretaries for all their kindness and help, especially Edwine De
Bruycker, Caroline Claeys, Cindy Boonaert, Connie Verliefde and Ilse De Vuyst. They all
helped me in these years of study.
My gratitude also goes to all members of examination committee, Prof. dr. ir. Guy Smagghe
(Chairman), Prof. dr. ir. Monica Höfte (Secretary), Prof. dr. Danny Geelen, Em. Prof.
Margaret Essenberg (USA), dr. lic. Marc De Loose (ILVO), dr. Mansour Karimi (VIB) for
their efforts in reading my thesis and giving their precious suggestions and comments.
This work would never be possible without valuable support from Iranian government during
my stay in Ghent. My gratitude goes to Iranian government to provide the possibility to go
abroad to study.
I would like to express my gratitude to the staff of OBSG, Father Charle, Marleen, Isabelle
and Annemie for their support during my stay in Ghent. OBSG was my home away from
home and I spent my joyful and precious moments with the students from all over the world.
I would like to thank all my best friends, dr. Thuy Tran Thi, dr. Katrien Vandepitte, dr.
Hossein Hosseini Moghaddam, dr. Asad Maroufi, ir. Saman Kamangar, dr. Mansour Karimi,
ir. Mary Shalet, ir. Siamak Ghafaripour, ir. Panagiotis Theocharis, ir.Vasiliki Pantoula, ir.
Linda Zamariola, ir. Giacomo Bellandi, dr. Luca Pipino, ir. Wannes Voorend, dr. Lina
Maloukh, ir. Prabhu Lakshmanan with whom I experienced an unforgettable time and really
enjoyed. I never forget their kindness and appreciate their honest friendship, my living in
Belgium could not be so nice and wonderful without them.
To my family: my father, mother, sisters (Negar, Leila and Sarvenaz), brother (Shahram),
they deserve the greatest gratitude. Thank you for your love, inspiration, endless support, and
understanding throughout my study abroad.
This end means a new beginning for me. A new path in science and life will be opened like
spring time, I would love it. I will keep this moment in my mind. I would like to thank you all
who made my time in Ghent.
Ghent, October 2013
Khosro Mehdi Khanlou
Table of Contents
Table of Contents
List of abbreviations ................................................................................................................ I
Thesis outline ........................................................................................................................ V
Chapter 1: Introduction ....................................................................................... IX
1.1. Red clover........................................................................................................................ 3
1.1.1. Importance ................................................................................................................ 3
1.1.2. Taxonomy ................................................................................................................. 3
1.1.3. Genetics .................................................................................................................... 4
1.1.4. Morphology .............................................................................................................. 5
1.1.5. Breeding methods ..................................................................................................... 6
1.1.6. Red clover breeding objectives ................................................................................ 8
1.2. The pathogen Sclerotinia trifoliorum Erikss. .................................................................. 8
1.2.1. Taxonomy ................................................................................................................. 9
1.2.2. Life cycle of Sclerotinia trifoliorum Erikss. .......................................................... 10
1.2.3. Disease management .............................................................................................. 11
1.3. The pathogen Sclerotium rolfsii Sacc. ........................................................................... 12
1.3.1. Taxonomy ............................................................................................................... 12
1.3.2. Life cycle of Sclerotium rolfsii Sacc. ..................................................................... 13
1.3.3. Disease management .............................................................................................. 14
1.4. Secondary Metabolism .................................................................................................. 15
1.4.1. Phenylpropanoid pathway ...................................................................................... 16
Table of Contents
1.4.2. Flavonoid biosynthesis ........................................................................................... 19
1.4.3. The flavonoid subclasses ........................................................................................ 20
1.4.4. Isoflavonoid biosynthesis ....................................................................................... 21
1.4.5. Health benefits of isoflavonoids in plants .............................................................. 25
1.4.6. Genetic engineering of isoflavonoid biosynthesis ................................................. 27
1.5. Role of secondary metabolites in plant defense response ............................................. 29
1.5.1. Constitutive and induced antifungal compounds ................................................... 30
1.5.2. Antifungal compounds: definition and classification ............................................. 30
1.5.3. Phytoalexins in non-Fabaceous plants ................................................................... 32
1.5.4. Phytoalexins in legumes ......................................................................................... 33
1.5.5. Strengthening plant defense by phytoalexin engineering....................................... 34
1.6. Application of biotechnology in plant breeding ............................................................ 37
1.6.1. Tissue culture as a pre-requisite for gene transformation ...................................... 38
1.6.2. Development of tissue medium in red clover ......................................................... 39
1.6.3. Tissue culture response .......................................................................................... 40
1.6.4. Breeding regenerable varieties in forage legumes ................................................. 42
1.6.5. Gene transformation ............................................................................................... 43
1.6.6. Biological gene transfer ......................................................................................... 44
1.6.7. Reporter genes ........................................................................................................ 44
1.7. Functional genomics: a window opened by gene transformation ................................. 45
1.7.1. Over-expression ..................................................................................................... 46
1.7.2. Gene silencing ........................................................................................................ 46
Table of Contents
1.8. Real-time PCR technique, a valuable tool for functional genomic studies ............... 50
1.8.1. Absolute quantification .......................................................................................... 50
1.8.2. Relative quantification ........................................................................................... 51
Chapter 2: Regeneration and genetic transformation of red clover .................... 53
2.1. Introduction ................................................................................................................... 55
2.2. Material and Methods .................................................................................................... 57
2.2.1. Genotype screening ................................................................................................ 58
2.2.2. Optimizing plant growth conditions ....................................................................... 60
2.2.3. Optimization of tissue culture media for regeneration ........................................... 61
2.2.4. Optimizing transformation parameters ................................................................... 61
2.2.5. Kanamycin sensitivity test ..................................................................................... 62
2.2.6. Bacterial strains and vectors ................................................................................... 62
2.2.7. Bacterial culture ..................................................................................................... 63
2.2.8. Inoculation and co-cultivation ................................................................................ 63
2.2.9. PCR analysis .......................................................................................................... 64
2.2.10. Histochemical GUS assay .................................................................................... 65
2.2.11. Statistical analysis ................................................................................................ 65
2.3. Results ........................................................................................................................... 66
2.3.1. Genotype screening ................................................................................................ 66
2.3.2. Optimization of plant growth conditions in vitro ................................................... 67
2.3.3. Optimization of tissue culture media for regeneration ........................................... 68
2.3.4. Determination of kanamycin concentration as a selection agent ........................... 70
Table of Contents
2.3.5. Effect of Agrobacterium strains ............................................................................. 71
2.3.6. Effect of callus induction period prior to Agrobacterium infection ....................... 72
2.3.7. Verification of transformation by PCR analyses .................................................... 72
2.3.8. Assay of GUS Activity ........................................................................................... 73
2.4. Discussion ..................................................................................................................... 74
Chapter 3: Identification of reliable reference genes in red clover ..................... 81
3.1. Introduction ................................................................................................................... 83
3.2. Materials and Methods .................................................................................................. 86
3.2.1. Plant materials ........................................................................................................ 86
3.2.2. Total RNA isolation and cDNA preparation .......................................................... 86
3.2.3. Primer design and verification of amplified products ............................................ 86
3.2.4. Determination of primer efficiency ........................................................................ 88
3.2.5. qRT-PCR conditions .............................................................................................. 88
3.2.6. Data analysis .......................................................................................................... 89
3.3. Results ........................................................................................................................... 90
3.3.1. Identification of red clover candidate reference genes ........................................... 90
3.3.2. PCR efficiency and amplification specificity ......................................................... 90
3.3.3. Descriptive statistics and relations of the raw Ct values ........................................ 91
3.3.4. Determining the stability of the candidate reference genes ................................... 94
3.4. Discussion ..................................................................................................................... 98
Chapter 4: RNA interference silencing of isoflavone synthase gene in red clover
........................................................................................................................... 109
4.1. Introduction ................................................................................................................. 111
Table of Contents
4.2. Materials and methods................................................................................................. 113
4.2.1. Genes of interest ................................................................................................... 113
4.2.2. RNAi vector construction ..................................................................................... 113
4.2.3. DNA/RNA concentration measurement ............................................................... 113
4.2.4. Primers design ...................................................................................................... 113
4.2.5. Development of the RNAi construct .................................................................... 113
4.2.6. The BP reaction .................................................................................................... 114
4.2.7. Transformation of E. coli ..................................................................................... 114
4.2.8. Plasmid DNA preparation .................................................................................... 114
4.2.9. Restriction endonuclease digestion of pEN-L1-IFS-L2 vector ............................ 115
4.2.10. Agarose gel electrophoresis ............................................................................... 115
4.2.11. DNA sequencing ................................................................................................ 115
4.2.12. The LR reactions ................................................................................................ 116
4.2.13. Selection of a clone with the correct orientation ................................................ 116
4.2.14. Restriction endonuclease digestion of pXK7-IFS-I-SFI-2 vector ...................... 117
4.2.15. Transformation into Agrobacterium tumefaciens............................................... 117
4.2.16. Verification of pXK7-IFS-I-SFI-2 in A. tumefaciens by digestion .................... 117
4.2.17. Plant transformation ........................................................................................... 118
4.2.18. PCR analysis ...................................................................................................... 119
4.2.19. qRT-PCR analysis of transgenic plants .............................................................. 119
4.3. Results ......................................................................................................................... 120
4.3.1. Hairpin binary vector ........................................................................................... 120
Table of Contents
4.3.2. Generation of expression vector ........................................................................... 122
4.3.3. Genetic transformation and plant regeneration .................................................... 124
4.3.4. PCR analyses ........................................................................................................ 125
4.3.5. qRT-PCR analysis of IFS-RNAi in silenced red clover lines .............................. 126
4.4 Discussion .................................................................................................................... 126
Chapter 5: Evaluation of the role of isoflavones in resistance of red clover to
Sclerotinia trifoliorum and Sclerotium rolfsii ................................................... 131
5.1. Introduction ................................................................................................................. 133
5.2. Materials and methods................................................................................................. 134
5.2.1. Plant material ........................................................................................................ 134
5.2.2. Fungal isolates ...................................................................................................... 135
5.2.3. General procedure used in pathogenicity assays .................................................. 135
5.2.4. Identification of proper reference genes and determination of suitable sampling
position for wild type plants via expression study using qPCR ..................................... 136
5.2.5. Determination of suitable reference genes and assessment of IFS expression level
in transgenic lines ........................................................................................................... 136
5.2.6. Collection of mycelia samples for qPCR ............................................................. 137
5.2.7. Total RNA isolation and cDNA preparation ........................................................ 137
5.2.8. qRT-PCR conditions ............................................................................................ 137
5.2.9. Analysis of qPCR results ...................................................................................... 137
5.2.10. Pathogenicity assays of S. trifoliorum and S. rolfsii .......................................... 137
5.2.11. Statistical analysis .............................................................................................. 138
5.3. Results ......................................................................................................................... 138
Table of Contents
5.3.1. Identification of proper reference genes for S. trifoliorum and S. rolfsii in wild type
plants .............................................................................................................................. 138
5.3.2. Identification of suitable position of sampling with qPCR .................................. 140
5.3.3. Determination of appropriate reference genes for S. trifoliorum and S. rolfsii
infected transgenic lines ................................................................................................. 141
5.3.4. Evaluation of IFS expression in RNAi-silenced transgenic and wild type plants 143
5.3.5. Pathogenicity tests ................................................................................................ 143
5.4 Discussion .................................................................................................................... 146
Chapter 6: General discussion & future perspectives ....................................... 153
6.1. Introduction ................................................................................................................. 155
6.2. Regeneration and transformation ................................................................................ 156
6.3. Gene expression .......................................................................................................... 158
6.4. RNA interference silencing ......................................................................................... 160
6.5. Pathogenicity assay ..................................................................................................... 161
6.6. Future perspectives ...................................................................................................... 165
Summary ............................................................................................................................ 171
Samenvatting ...................................................................................................................... 175
Reference List..................................................................................................................... 177
Curriculum vitae ................................................................................................................. 201
List of abbreviations
I
List of abbreviations
2,4-D: 2,4-dichlorophenoxyacetic acid
4CL: 4-coumaroyl CoA ligase
6-BAP: 6-benzylamino purine
ACT: actin
AD: adenine
ANOVA: analysis of variance
B5: Gamborg B5
bp: base pair
C4H: cinnamate 4-hydroxylase
CaMV: cauliflower mosaic virus
ccdB: gene encoding the cdB protein controlling cell death B gene
cDNA: complementary DNA
CHI: chalcone flavanone isomerase
CHR: chalcone reductase
CHS: chalcone synthase
Coff. Var.: coefficient of variation
Cp: crossing cycle number
CRD: completely randomized design
CRG: candidate reference gene
Ct: cycle threshold
DF: degrees of freedom
dNTPs: deoxyribonucleotides (dATP, dCTP, dGTP and dTTP)
dsDNA: double-stranded DNA
E: efficiency
EF-1a: elongation factor-1alpha
EIF-4a: translation initiation factor
EST: expressed sequence tags
GADPH: glyceraldehyde-3-phosphate-dehydrogenase
GUS: β-glucuronidase reporter gene
gusint: intron region
hpRNA: hairpin RNA
IAA: indole-3-acetic acid
IDH: isoflavone dehydratase
IFS: isoflavone synthase
List of abbreviations
II
KIN: kinetin
LB: left border of T-DNA
M value: gene expression stability measure
miRNAs: microRNA
mRNA: messenger RNA
MS: Murashige and Skoog (macro and micro elements)
MSB: intergroup variance
MSE: mean square error
MSW: intragroup variance
NAA: naphthalene acetic acid
NF: normalization factor
ng: nanogram
nM: nanomolar
NPTII: neomycin phosphotransferase-II
NTC: no-template control
OD: optical density
Oligo dT: oligodeoxythymidylic acid
PAL: phenylalanine ammonia-lyase
PCR: polymerase chain reaction
PIC: picloram
pmol: picomole
P-NOS: nopaline synthase promoter
PTGS: post transcriptional gene silencing
qRT-PCR: quantitative reverse transcription real-time polymerase chain reaction
R2: correlation coefficient
RB: right border of T-DNA
RG: reference gene
RISC: RNA-induced silencing complex
RNAi: RNA interference
rRNA: ribosomal RNA
RT: reverse transcription
S.D: standard deviation
SAND: sand family protein
siRNA: small interfering RNA
sRNA: small RNA
ssDNA: single-stranded DNA
TGS: transcriptional gene silencing
List of abbreviations
III
T-NOS: nopaline synthase terminator
T-OCS: octopine synthase terminator
UBC2: ubiquitin-conjugating enzyme E2
UBQ10: polyubiquitin 10
X-gluc: 5-bromo-4-chloro-indolylglucuronide
YEB medium: yeast extract and beef
YLS8: yellow-leaf-specific protein 8
µl: microliter
Thesis outline
V
Thesis outline
A large body of literature reports that secondary metabolite isoflavones have critical roles in
plant-antifungal activity in planta. Red clover is an abundant source of isoflavones. This
makes them an ideal subject for exploring the possibility of strengthening red clover plants
using the plants’ own defense arsenals. In this study, we evaluated possible significant roles
of isoflavones in creating resistance against Sclerotinia trifoliorum and Sclerotium rolfsii by
altering the isoflavone synthase gene using new molecular techniques. Our work is the first
attempt to investigate the antifungal activity of isoflavones using genetic engineering in red
clover. The thesis contains six chapters, as follows:
Chapter One: This chapter is a literature review about phytoalexins, which are derived from
secondary metabolism. This chapter focuses on isoflavonoids and their antifungal activity of
role in plant defense systems. Also the application of new molecular techniques such as gene
transformation and RNA interference in genetic engineering is described.
Chapter Two: This chapter describes the optimization of regeneration and transformation of
red clover. This chapter presents results on investigation of different media sequences in order
to achieve a sequence that could exploit all of the potential regeneration capacity of explants
used for gene transformation. We also evaluated different Agrobacterium strains to obtain the
best one for gene transformation.
Chapter Three: This chapter discusses the potential pitfalls arising from uncritical use of the
available statistical algorithms. This chapter presents our suggested new stability index for
ranking candidate reference genes. We also systematically validated a panel of putative
reference genes to identify the useful genes for data normalization in future gene expression
studies among different tissues in red clover.
Chapter Four: In this chapter, we present and discuss the construction of double-stranded
RNA (dsRNA) from the coding region of the isoflavone synthase gene and the transformation
Thesis outline
VI
of red clover using an Agrobacterium tumefaciens strain containing a binary hairpin vector.
We screened transgenic lines using PCR. A preliminary analysis was also conducted to
investigate expression level of isoflavone synthase gene in transgenic and non-transgenic
lines.
Chapter Five: This chapter represents the first in vivo attempt to evaluate the possibly
significant role of isoflavones in disease resistance via genetic manipulation of red clover.
RNAi-silenced transgenic lines in a wild type were subsequently exposed to Sclerotinia
trifoliorum and Sclerotium rolfsii. An unbalanced analysis of variance (ANOVA) technique
was used to investigate possible statistically significant differences between transgenic lines
and the wild type when exposed to the fungi.
Chapter Six: This chapter presents a general discussion of the thesis with recommendations
for future research.
Chapter 1: Introduction
Introduction
3
1.1. Red clover
1.1.1. Importance
Globally, grasslands (including savanna, shrubland, temperate grassland and tundra) cover
about 40% of the earth’s landmass (White et al., 2000). Red clover (Trifolium pratense L.) is
an important component of temperate grasslands, and covers about 8% of the landmass
worldwide. Red clover, a perennial forage legume of limited persistence, is capable of
adaptation to a wide range of soil types and environmental conditions. Red clover is able to
adapt to annual rainfalls ranging from 310 to 1910 mm (Smith et al., 1985). Red clover is
cultivated in most of North and South America. It is also found in parts of New Zealand,
Australia, China and Japan (Taylor and Quesenberry, 1996). Currently, red clover is found in
most of Europe, from the Mediterranean region to northern Scandinavia. In Belgium, red
clover is mostly grown in association with cool season grasses. A 2012 report showed that a
mixture of red clover and grass was cultivated in 4466 hectares in Belgium. Due to its ability
to fix nitrogen, red clover is an effective soil-improver. It has been demonstrated that the
amount of N-fixation of red clover varies from 76 to 373 kg ha-1
in grass-clover mixtures in
various environments (Boller and Nosberger, 1987; Crush et al., 1983). Moreover, red clover
generally increases the total dry matter yield of pastures and meadows by up to 2 t ha-1
and
may improve forage crude protein content by 1 to 3% when mixed with grasses as compared
to pure grass stands (Taylor and Quesenberry, 1996).
1.1.2. Taxonomy
The family of Fabaceae is classified into three subfamilies: Mimosoideae, Caesalpinioideae
and Faboideae (Papilionoideae). Among the subfamilies of Fabaceae, Faboideae has drawn
most attention. Not only does it have more members than other subfamilies of Fabaceae, also
most of its members are economically important crops including red clover (Trifolium
Chapter 1
4
pratense L.), soybean (Glycine max Merrill.), pea (Pisum sativum L.), and alfalfa (Medicago
sativa L.). The subfamily Faboideae contains several tribes: the tribe Trifoliaea comprises the
genus Trifolium and the genus Medicago and several other genera (Figure 1.1). The genus
Trifolium is divided into eight sections; each section contains several species. The species red
clover is placed in the section Trifolium, together with species such as diffuse clover
(Trifolium diffusum Ehrh.) and zigzag clover (Trifolium medium L.). Although several
varieties of the species Trifolium pratense (T. p.) have been identified, cultivated red clover is
generally referred to as the variety T. p.
Figure 1.1: Taxonomic relationship of the family of Fabaceae (Choi et al., 2004).
1.1.3. Genetics
Red clover is a perennial, diploid (2n = 2x = 14) and outcrossing species with a gametophytic
self incompatibility; in other words, it needs to be cross-fertilised before it can set seed. The
gametophytic self incompatibility system in red clover is controlled by one locus (Taylor and
Quesenberry, 1996). This S locus is characterized by up to 200 different alleles, which is
Introduction
5
higher than the number of S-alleles in most other self-incompatible species (Lawrence, 1996).
One of these alleles, the Sf allele, induces self-fertility and is dominant over the other S-alleles
(Sn). When heterozygous, self-fertile plants (Sf SX) are test crossed to unrelated self-
incompatible plants (SX SY), offspring segregate approximately 1:1 for self-fertility and self-
incompatibility as expected (Taylor and Quesenberry, 1996).
1.1.4. Morphology
The plant grows from a crown consisting of an accumulation of basal buds at or slightly
above the soil’s surface. From the crown, the stems elongate and become erect, usually in a
Figure 1.2: Roots, stems, leaves and flowers of red clover (Öhberg 2008).
branched form, reaching a height of 20–80 cm. The leaves are alternate and trifoliate (with
three ovate leaflets), green in color, and often characterized by a pale “V” in the center. The
inflorescence is most commonly pink or purple, and forms a capitilum or head with up to 300
florets at the ends of the main and axillary stems (Figure 1.2). The florets are pollinated by
bumblebees and honey bees. Red clover plants usually have a well-developed taproot, which
Chapter 1
6
can extend to a depth of more than 1 m. The taproot is somewhat branched; adventitious or
lateral roots emanating from the crown also exist (Taylor and Quesenberry, 1996).
1.1.5. Breeding methods
Due to the open pollinated nature of red clover, breeders should use the method that can
increase accumulation of suitable genes to achieve any improvement. Most of the methods
which are being used for red clover breeding have been first used for many years to develop
new cultivars in corn (Zea mays), and use spaced plant field populations (Taylor and
Quesenberry, 1996).
Mass selection is a very simple method for red clover. Its use is quite efficient for breeders
wanting to select for a small number of traits with high heritability, such as general
adaptation, pathogen resistance and cold tolerance (Vleugels, 2013). In this method,
individual plants are screened on the basis of their phenotypes and the seed yield. The
production of seeds is under open-pollinated condition which means that maternal plants are
selected but the pollen (male parent) source is unknown. The next generation is sown from a
composite of equal amounts of seed from each selected individual plant.
Family selection is the most widely used breeding program for red clover (Boller et al., 2010).
Selection is based on phenotypic scores (also leading to the alternative name for this method,
“recurrent phenotypic selection”). It is very similar to cyclic mass selection. The main
difference is that seeds of half-sib families are maintained individually rather than bulked.
Similar to mass selection, parent plants are not progeny tested (Taylor and Quesenberry,
1996). In this method, seeds of a number of plants with desirable phenotype are cultivated in
rows in a way that each row contains multiple plants of one family. After two to three years of
selection, the plants with desirable characteristics are selected from the best families. The
selected plants are pollinated with various other desirable plants to generate half-sib families.
Then seeds of each half-sib family are harvested separately. In the next selection cycle, a new
Introduction
7
field trial is established with rows of multiple plants from each family. Again, selection is
conducted for two to three years and the most desirable plants from the best families are
selected for pollination in isolation. After the last selection cycle, the seeds of most desirable
plants and families are harvested to be tested in a yield trial. Many variations of recurrent
phenotypic selection are possible. For instance, selection can be conducted based on a halfsib
family or combining the best plants from the best family. Although recurrent phenotypic
selection is most effective for simple inherited characters, it can also improve traits with lower
heritability (Vleugels 2013).
Polycross is another method used for breeding red clover. In this method, plants are
propagated by grafting. The clones obtained from vegetative propagation are evaluated for
their general combining ability (GCA). The suitable clones with high GCA are crossed. The
seeds obtained from polycross are used for progeny testing to identify the combination of
clones that is best to form a synthetic cultivar. The main advantage of the polycross method is
that breeding materials are evaluated in the progeny tested. Therefore the method is
appropriate for selection for complex characteristics. Using this method in the breeding of red
clover for persistence appears to have been effective in view of the genetic gain made in the
selected materials (Taylor et al., 1968). However, this method is time consuming and
expensive (Taylor, 2008).
Strain building is similar to the polycross method with a few important differences. In the
polycross method, the parents are maintained by cloning, whereas in strain building the
parents are kept by seed populations. This makes strain building less labor-intensive than the
polycross method. All populations are developed and kept separately and after testing, the
best populations are mixed to form a synthetic variety. A disadvantage is that intra-population
crossing can happen and genetic gains are likely to be based only on additive genetic variance
(Taylor and Quesenberry, 1996).
Chapter 1
8
Single- and double-cross hybrids have been produced in red clover with genetic control of
crossing by using inbreds with homozygous S-alleles (Anderson et al., 1972). In double cross,
the parental lines are kept vegetatively on inbred clones, and single cross in which the
parental lines are kept by alternate selfing and sibbing (Taylor and Quesenberry, 1996). The
hybrid cultivars exploit both the general and specific combining ability that exists in red
clover (Cornelius et al., 1977). Although hybrid red clover has great potential, isolating rare
lines or clones with a highly specific combining ability is expensive. The development of
semi-hybrid populations can be an alternative approach to capture partial heterotic gains in
red clover for developing superior cultivars (Brummer, 1999).
1.1.6. Red clover breeding objectives
Increasing the yield is the first major goal of red clover breeding programs. The second major
breeding aim is to improve persistence (individual plant longevity) because red clover suffers
from insufficient persistence (Hejduk, 2011). Although the life of clover is mostly controlled
by programmed senescence (Taylor, 2008), fungal diseases contribute considerably to the lack
of persistence of red clover (Pokorny et al., 2003). Therefore, to improve persistence, red
clover disease resistance needs to be increased. Selection for plant longevity in older stands of
red clover and the incorporation of genes for resistance to fungal pathogens are the most
practical approaches to achieve higher persistence (Smith and Hay, 1983). Persistence is
related to resistance to Sclerotinia trifoliorum Erikss and other pathogens that cause rot in red
clover, including Sclerotium rolfsii Sacc (Vleugels, 2013). Resistance to these pathogens
should lead to improved persistence.
1.2. The pathogen Sclerotinia trifoliorum Erikss.
The most destructive disease of red clover in temperate regions with cool seasons, including
North America and Europe, was first described by Eriksson in the early 1800's along with its
pathogen, named Sclerotinia trifoliorum Eriksson (clover rot, clover cancer or Sclerotinia
Introduction
9
crown and stem rot). The fungus also attacks other economically important forage legumes
such as alfalfa (Medicago sativa L.) and white clover (Trifolium repens L.) (Delclos et al.,
1997). The most obvious early symptom of Sclerotinia diseases is the appearance of white
mycelia on the infected plant. When mycelia encounter nutrient-limited environments, soon
afterward mycelia produce the compact resting bodies (sclerotia) (Rollins and Dickman,
1998) with size of 2 to 20mm (Kohn, 1979a). The damage caused by clover rot varies
according to the year because the weather conditions play a key role in disease development;
i.e., dry weather slows disease progression. Also flooding for long periods could reduce the
fungus population in the soil by inducing starvation, lack of oxygen, or desiccation (Agrios,
2005). The amount of germinated ascospores increases in warm winters with short periods of
frost leading to increased loss, while the mycelia grow very slowly in cold and dry winters,
resulting in a reduction of the risk of disease loss (Vleugels et al., 2010).
1.2.1. Taxonomy
Ascomycota is the largest phylum of Fungi, with over 64,000 species. The characteristic
feature of the Ascomycota phylum is that sexual ascospores are produced in groups of eight
within a sac-like structure called the ascus. Leotiomycete is a class of ascomycete fungi.
Many of the Leotiomycetes cause serious plant diseases. Leotiomycetes consist of several
orders. Helotiale, which possesses 10 families, is one of the Leotiomycete orders. The
Sclerotiniaceae family, which belongs to the order of Helotiales, produce apothecia. An
apothecium, a sexual fruiting body, has concave or flat tops 6 mm or more in diameter (Hao et
al., 2003). The genus Sclerotinia, which belongs to the family Sclerotiniaceae, has several
species. Kohn (1979a) used cultural, biological, macroscopic, and microscopic characteristics
to separate species in the genus Sclerotinia. All species of this genus produce dense white
bodies on their mycelia. These bodies, which later become hard and black, are called
sclerotia. Some of the species of this genus are considered plant pathogens: Sclerotinia
Chapter 1
10
homeocarpa F.T. Bennett, Sclerotinia minor Jagger, Sclerotinia sclerotiorum (Lib.) de Bary
and Sclerotinia trifoliorum Erikss. Taxonomy is well understood in Sclerotinia trifoliorum
Erikss. (Kohn, 1979b) (Figure 1.3).
Kindom : Fungi
Phylum : Ascomycota
Class : Leotiomycetes
Subclass : Leotiomycetidae
Order : Helotiales
Family : Sclerotiniaceae
Genus : Sclerotinia
Species : Sclerotinia trifoliorium Erikss.
1.2.2. Life cycle of Sclerotinia trifoliorum Erikss.
In late summer and autumn, when the temperature drops and relative humidity increases,
sclerotia with a broken dormancy are induced to form apothecia. Each sclerotium can
generate up to 10 apothecia and each apothecium can produce up to a million ascospores per
day for a period of three to four weeks (Delclos and Raynal, 1995). Ascospores are deposited
onto stems and leaves and then germinate and penetrate directly into healthy tissues. Soon
after penetration, brown spots appear on stems and leaves. The infection stays in this stage
until weather conditions become favorable to the fungus, namely between late autumn and
winter. Then the mycelia that develop from ascospores systemically infect the entire plant by
growing through the petioles and stems into the crown and finally into the taproot (Öhberg
and Bang, 2010) (Figure 1.4). Disease progression appears on dead plants during early spring
after snow melt (Delclos et al., 1997).
Figure 1.3: Taxonomy of Sclerotinia trifoliorum Erikss. (Eriksson, 1880). TKimbrough,
1978)).
Introduction
11
1.2.3. Disease management
The disease can mostly be controlled using cultural, biological and chemical methods.
Symptoms caused by S. trifoliorum can be controlled by cultural practices including early
sowing and deep ploughing after clover rot outbreaks. Also plants should be planted in well-
drained soils, and should not be planted too close together to promote the passage of air. The
sclerotia remain viable in the soil for at least three years, therefore non-susceptible crops such
as small grains should be planted to remove the fungus from infected fields (Agrios, 2005).
Numerous species of fungi have been reported to parasitize Sclerotinia spp. by using the
nutrients stored in sclerotia as carbon source (Huang et al., 2000). One of the most important
fungi for biological control of many Sclerotinia spp. (e.g., Sclerotinia minor, Sclerotinia
sclerotiorum, Sclerotium cepivorum and Sclerotinia trifoliorum) is Coniothyrium minitans
which destroys existing sclerotia or inhibits the formation of new sclerotia and leads to
reduction of fungus population in the soil (Whipps et al., 2008). Coniothyrium minitans has
been found to be effective when red clover is grown in monoculture (Taylor and Quesenberry,
Primary local
Infection on leaves
Late autumn, winter
During late autumn and winter, the
pathogen starts to grow systemically,
finally entering the taproot
Spring
Sclerotia
Ascospores
Autumn
Formation of
apothecia from
sclerotia
Asc
Figure 1.4: The Life cycle of clover rot (Öhberg, 2008).
Chapter 1
12
1996). Another way to control this disease is by using chemical products such as vinclozolin
which can control 80 to 100% of the disease during sporulation of the fungus (Sauer, 1984).
Also the chemicals carbendazin and iprodione were found to be most effective in inhibiting
the growth of S. trifoliorum in a growth chamber (Raynal and Picard, 1985). Due to the huge
variation in the pathogen, completely resistant red clover varieties have yet to be developed.
1.3. The pathogen Sclerotium rolfsii Sacc.
In the beginning of the 20th
century, Peter Henry Rolfs found an unnamed fungus which
caused tomato blight in Florida. He sent a sample of the fungus to Italian mycologist Pier
Andrea Saccardo. The fungus was described and placed in the genus Sclerotium and named
Sclerotium rolfsii by Pier Andrea Saccardo in 1911. Mario Curzi discovered in 1932 that the
teleomorph (spore-bearing state) was a corticioid fungus and accordingly placed the species in
the genus Corticium. In 1978, the name of Corticium rolfsii was changed to Athelia rolfsii.
Sclerotium rolfsii is anomorph state of Athelia rolfsii. The diseases caused by S. rolfsii are
often referred to as southern blight (Fery and Dukes, 2002). The fungus Sclerotium rolfsii has
a broad host range; at least 600 species in 100 families are susceptible, including almost all
the agricultural and horticultural crops (Rivard, 2010). As a soil borne pathogen, S. rolfsii
grows, survives, and attacks plants at or near the soil line. Therefore the stems of hosts are the
first place to be attacked by S. rolfsii. Under suitable conditions, the fungus may infect other
parts of hosts including roots, fruits, petioles, leaves, and flowers. Weather conditions can
strongly influence the development of S. rolfsii. The regions with warm weather are optimal
for growth of Sclerotium rolfsii. Especially periods of temperature and moisture fluctuation
promote the growth of the fungus (Punja, Z. K. 1985).
1.3.1. Taxonomy
Basidiomycota (colloquially basidiomycetes) is the second largest phylum with over 30,000
species (Webster and Weber, 2007). Basidiomycota are filamentous and reproduce sexually
Kingdom: Fungi
Phylum: Basidiomycota
Class: Agaricomycetes
Order: Atheliales
Family: Atheliaceae
Genus: Athelia
Species: Athelia rolfsii (Curzi) C.C. Tu & Kimbr
Kingdom: Fungi
Phylum: Basidiomycota
Class: Agaricomycetes
Order: Atheliales
Family: Atheliaceae
Genus: Athelia
Species: Athelia rolfsii (Curzi) C.C. Tu & Kimbr
Introduction
13
via the formation of specialised club-shaped structures (basidium). Usually four basidiospores
were observed per basidium but in some cases there are one, two or more than four
basidiospores per basidium (Mehrotra and Aneja, 1990). One of the most well-known classes
of this phylum is Agaricomycetes, which contains the monotypic order Atheliales.
Kindom : Fungi
Phylum : Basidiomycota
Class : Agaricomycetes
Order : Atheliales
Family : Atheliaceae
Genus : Athelia
Species : Athelia rolfsii (Curzi) C.C. Tu & Kim br.
The Atheliaceae is a family of fungi in the Atheliales order. The Atheliaceae include 22
genera (Kirk et al 2008). Athelia is a genus in the family Amylocorticiaceae with 28 species.
The best known species in this genus is Athelia rolfsii (anamorph Sclerotium rolfsii) (Xu et al
2010) (Figure 1.5).
1.3.2. Life cycle of Sclerotium rolfsii Sacc.
In spring, during favorable weather conditions, the fungus life cycle begins with the
germination of sclerotia which are the survival structure of the fungus (Punja and Grogan
1981). When the fungal mycelia have grown out from sclerotia and make contact with plants,
direct penetration occurs. S. rolfsii rarely produces basidiospore (the sexual stage of
reproduction) at the lesions and under humid conditions.
Figure 1.5: Taxonomy of Athelia rolfsii (anamorph Sclerotium rolfsii) (Tu and Kimbrough, 1978).
Chapter 1
14
However the produced basidiospores could cause secondary infection (or a secondary cycle)
by spreading on air currents and infection of other plants. In the beginning of summer, as the
disease develops, a cottony, white mass of mycelium covers the stem of host plants. The
speed of the pathogen in covering whole plant with mycelia highly depends on the amount of
moisture present (Agrios, 2005). The infected mature plants suddenly wilt and turn yellow
due to production of chemicals by the fungus that cause systematic wilting and plant death.
Soon after plant death, numerous small roundish sclerotia form on all infected tissues to
overwinter (Agrios, 2005). Sclerotia have diameters of 0.5 - 2 mm and may survive and
remain infective up to 7 years in field soil or on plant debris depending on experimental
conditions (Xu, 2008) (Figure 1.6).
1.3.3. Disease management
Generally, the control of Sclerotium diseases is difficult (Agrios, 2005).
Nevertheless,infection of the pathogen can be reduced by changes in cultural practices using
resistant varieties and application of fungicides. Because S. rolfsii has a large number of hosts
The sclerotia germinate
Sclerotia are
moved in wind
The fungus infect
stem and roots
Spores, but they
are rarely formed
The fungus forms sclerotia on
the stem at the soli surface
Figure 1.6: The life cycle of Sclerotium rolfsii based on Mullen (2001).
Introduction
15
and can overwinter on many types of crop debris, the application of non‐hosts for crop
rotation is difficult (Punja, 1985). Maize and cabbages are little affected by the fungus,
however. Cultural control is difficult when inoculum levels are high and environmental
conditions are appropriate for disease development. To reduce infection, plant remains should
be removed and burned after harvest. Although the sclerotia can survive up to 7 years in soil
near the surface (Aycock, 1966), they do not survive for more than 45 days if buried 20-30
cm. Deep ploughing of the soil can be one way to minimize infection. The application of
ammonium-type fertilizers is also useful to control the disease (Agrios, 2005). Using resistant
varieties is also a way to control the fungus. The search for varieties with resistance to this
fungus has encountered little success, however. The application of fungicides such as PCNB
has also been recommended for the control of S. rolfsii but for the soil-borne pathogens, use
of fungicides is not practical because of the exorbitant costs and environmental hazards
involved. Moreover, numerous reports have been made on successful application of fungi and
bacteria as antagonists for S. rolfsii through competition for nutrients and production of
antibiotics (Abd-Allah, 2005; Yaqub and Shahzad, 2005; Zmora-Nahum et al., 2008).
1.4. Secondary Metabolism
The term “plant metabolism” is generally used for a closely co-ordinated series of enzyme-
mediated chemical reactions in a living plant (Mann, 1987). Groups of reactions together form
metabolic pathways to produce a range of molecules such as sugars, amino acids and fatty
acids. From these, polymers can be derived, such as polysaccharides, proteins, lipids, RNA
and DNA, etc. Plant metabolism is divided into primary and secondary metabolism. Because
distinction between primary and secondary metabolism is difficult on the basis of molecules,
chemical structures, or biosynthetic origin, many authors prefer a functional definition of
primary and secondary metabolism. Primary metabolites participate in nutrition and essential
metabolic processes, and secondary metabolites (also known as low-molecular-mass natural
Chapter 1
16
products) have no direct functions in growth and development and are primarily involved in
interaction between the plant and its environment (Buchanan et al., 2000). Approximately 450
million years ago, when plant species spread out from sea to dry land, they had to adapt to a
wide array of new environmental challenges including live microorganisms in land and
increased UV radiation. Therefore, the first land plants generated new metabolic metabolites
(secondary metabolites) to adapt to the new environmental conditions in land (Ferrer et al.,
2008). The secondary metabolites are differentially distributed among taxonomic groups
within the plant kingdom. The various pathways of secondary metabolism generate an
enormous structural diversity, yielding an estimated 140,000 different metabolites. However,
taking into account that until now only 20–30% of the 350,000 known plant species have been
phytochemically studied, the real number of secondary metabolites present in the plant
kingdom could exceed 200,000 compounds (Wink, 2010). In the past, secondary metabolites
have been considered as waste products for the plant (Mothes et al., 1985). In recent years,
plant scientists have become aware of the impact that secondary metabolites can have through
their interaction with the biotic and abiotic environment.
1.4.1. Phenylpropanoid pathway
The phenylpropanoid pathway serves as a starting point for the production of an enormous
array of secondary metabolites (Figure 1.7). The phenylpropanoid derived metabolites are
involved in plants defense against infection caused by microorganisms (Vogt, 2010). Due to
advances in molecular biology techniques in recent years, various excellent studies have been
performed to advance the knowledge of the enzymology of phenylpropanoid synthesis, the
cellular and subcellular sites of synthesis, and the roles of phenylpropanoid compounds in the
life of the plant. The first three enzymes involved in phenylpropanoid pathway are
phenylalanine ammonia-lyase (PAL), cinnamate 4-hydroxylase (C4H) and 4-coumaroyl CoA
ligase (4CL). Phenylalanine ammonia-lyase (PAL), a key enzyme in phenylpropanoid
Introduction
17
metabolism, catalyses the gateway metabolic step from a primary metabolite (amino acid
phenylalanine) into phenylpropanoid metabolism. The isolation and partial purification of this
enzyme was first described in by Koukol and Conn (1961). The PAL is the most extensively
studied enzyme in the phenylpropanoid pathway, perhaps even in the entire secondary
metabolism. This enzyme belongs to the family of lyases, specifically ammonia lyases, which
cleave carbon-nitrogen bonds. In some plants, the PAL is encoded by a single gene, whereas
in other plants, it is the product of a multigene family. The PAL does not need any cofactor
for activity. The PAL converts phenylalanine to cinnamic acid, which in turn is the precursor
of various phenylpropanoids such as lignins, flavonoids, coumarins, and many small phenolic
molecules which have multiplicity of functions in structural support, pigmentation, defense
and signaling. This enzyme has been isolated from a variety of plants such as Populus
trichocarpa (poplar), Oryza sativa (rice) and Arabidopsis thaliana (Hamberger et al., 2007).
Figure 1.7: Phenylpropanoid biosynthetic pathways. Enzymes and representative reactions of the
flavonoid pathway and related biosynthesis of various antimicrobial phytoalexins are shown. The
color indicates highly represented enzyme classes:cytochrome P450-catalyzed reactions are
highlighted in red, short-chain dehydrogenase/reductase (SDR) reactions are green, and O-
methyltransferase (OMT) reactions are lavender. “R” represents –H or –OH. (see Figure 1.7
continued) (Ferrer et al., 2007).
Chapter 1
18
Figure 1.7 continued.
HO
HO
OH
OH
OH
OH OH
~ i '"'' jX.t(fööii;'''·~~ t ~ob~en~j .........................
OH
(·······c~·~~~ri~ .. ·· · ··'\~
i Cortdem &:i Tannns : ! {Proan:l'locy:r.kllns) i \., ,, ,_ ,,, .............. _,,,,,_;
..
CHS+
/
OFR --
FS'H -
Dm3MAT3
c .. s ~- r'YH ~~·I· Co.\
CHS~
CHI~
HO
NO
OH OH
ANS~
HO
UFCT l
OH
Dlhytlro'l&:eo:b ~F'ow:mo1:.;
A'
R'
R'
FNS I FNSII /
IFS
~
OH
" HO
H~ll ,., " 0 1",
12'H~ OCH,
,., I I H~o I"
~o "' oe.,
HO
VR ~
OU:IO!
HO
OCH,
OOH,
Introduction
19
The second enzyme active in general phenylpropanoid metabolism is cinnamate 4-
hydroxylase (trans-cinnamate 4-mono-oxygenase). Oxidation of cinnamic acid by cinnamate
4-hydroxylase (C4H), a cytochrome P450, NADPH-dependent enzyme, results in p-coumaric
acid (or 4-coumaric acid). In some monocots, PAL accepts tyrosine as a substrate to directly
produce p-coumaric acid, bypassing the requirement for C4H (Yu and Jez, 2008). In the next
step, 4-coumaroyl CoA ligase (4CL) joins a CoA molecule to p-coumaric acid generating p-
coumaroyl-CoA (or 4-coumaroyl CoA) (Ferrer et al., 2008). p-Coumaroyl-CoA provides
numerous active metabolites with the basic C6-C3 carbon skeleton of phenylalanine through a
series of hydroxylation, methylation, and dehydration reactions (Bernards, 2010). These
activated intermediates are used in the biosynthesis of diverse compounds via specific branch
pathways, such as those leading to lignin and flavonoid biosynthesis. Phenylpropanoid units
derived from the shikimate pathway are common structural elements of all flavonoid
compounds.
1.4.2. Flavonoid biosynthesis
The flavonoid biosynthetic pathway begins with the condensation of one molecule of 4-
coumaroyl-CoA and three molecules of malonyl-CoA, resulting in naringenin chalcone
(tetrahydroxychalcone). This reaction is performed by the enzyme chalcone synthase (CHS)
(Orlikova et al., 2011) (Figures 1.7 and 1.8). CHS does not need any cofactor. In some
species, the coordinated action of the CHS and chalcone reductase (CHR) produces
trihydroxychalcone (isoliquiritigenin). By enzyme chalcone flavanone isomerase (CHI),
chalcones (tetrahydroxychalcone and trihydroxychalcone) are isomerised to the flavanones
naringenin (5,7, 4-trihydroxy flavanone) and liquiritigenin (5,7-dihydroxyflavanone), which is
a branching point in flavonoid metabolism (Ralston et al., 2005). From this point, the pathway
diverges into several side branches, resulting in a different class of flavonoids.
Chapter 1
20
1.4.3. The flavonoid subclasses
The term “flavonoid” is generally used to define a part of phenylpropanoid pathway products
that have C6-C3-C6 carbon framework, or more specifically phenylbenzopyran functionality.
Based on the position of the linkage of the aromatic ring to the benzopyrano (chromano)
moiety, one can classify them into three classes: the flavonoids (2-phenylbenzopyrans),
isoflavonoids (3-benzopyrans), and the neoflavonoids (4-benzopyrans). These classes usually
possess a common chalcone precursor. As a result, they are biogenetically and structurally
related.
Figure 1.8: Flavonoid biosynthesis. Figure based on Andersen and Markham (2006).
Gouraud et al. 2002 ).
Introduction
21
Depending on the degree of oxidation and saturation present in the heterocyclic C-ring, the
flavonoids may be classified into the following groups (Figure 1.9):
1.4.4. Isoflavonoid biosynthesis
The enzyme isoflavone synthase (IFS) acts as the key metabolic entry point for the formation
of all isoflavonoid compounds. The IFS is a cytochrome P450 enzyme and a NADPH-
dependent enzyme (CYP93C1). IFS has been cloned from soybean, alfalfa, red clover and
several other Fabaceae plants (Jung et al., 2000; Steele et al., 1999). It is suggested that the
IFS gene has arisen independently during evolution in taxonomically distinct families (Dixon
and Sumner, 2003). The IFS hydroxylates the C2 carbon of the substrate, and early migration
results in movement of the phenyl ring from C-2 to C-3, producing 2,5,7,4′-tetrahydroxy-
isoflavanone (2-hydroxyisoflavanone), an unstable reaction intermediate (Sawada et al.,
Figure 1.9: Different classes of flavonoids. Figure based on Grotewold (2006).
Chapter 1
22
2002). Isoflavone dehydratase (IDH) changes the intermediate 2-hydroxyisoflavanone into
isoflavone genistein (Akashi et al., 2005). There is a parallel pathway which uses an
alternative IFS substrate to produce daidzein (5-deoxyisoflavones). This pathway starts with a
Fabaceae-specific enzyme called chalcone reductase (CHR). CHR collaborates with CHS to
remove one hydroxyl group from the tetraketide intermediate of CHS, giving rise to 5-
deoxyisoflavones (or isoliquiritigenin). The CHI in Fabaceae catalyses the cyclization of 5´-
deoxychalcone to liquiritigenin (Ralston et al., 2005) from which IFS and IDH generate the
isoflavone daidzein. Additional methylation of genistein and daidzein leads to production of
the isoflavones Biochanin A and formononetin, respectively (Figure 1.10).
After production of isoflavones, these metabolites could either undergo glycosylation and
then be stored in vacuoles or go through some additional modifications such as hydroxylation,
methylation and prenylation, which lead to generation of a variety of isoflavonoid compounds
depending on plant species such as pterocarpans (glyceollin in soybean), isoflavanones
(vestitone in alfalfa, kievitone in french bean), and isoflavans (vestitol and sativan in alfalfa)
(Yu, O. and Jez, J. M. 2008) (Figure 1.11).
Liquiritigenin Naringenin
Isoflavone synthase
2, 7, 4'-trihydroxy-isoflavanone 2, 5,7, 4'-tetrahydroxy-isoflavanone
Isoflavone dehydratase
Daidzein Genistein
Isoflavone 4'-O-methyltransferase
Formononetin Biochanin A
Figure 1.10: Isoflavonoid biosynthesis. Figure based on Andersen and Markham (2006).
Introduction
23
Red clover has been demonstrated to contain high levels of the isoflavones daidzein,
genistein, formononetin, biochanin A (Sivesind and Seguin, 2006). Further modification of
red clover isoflavones results in production of pterocarpans such as medicarpin and maackiain
(McMurray et al., 1986, Vetter, 1995) (Figure 1.12).
Daidzein
Formononetin
Isoflavone 4'-O-methyltransferase
3'-hydroxyformononetin
Isoflavone 2'-hydroxylase
2'-hydroxyformononetin
Pseudobaptigenin synthase
Pseudobaptige
2'-hydroxyPseudobaptigenin
Isoflavone reductase
(-)-Sophorol
(-)-Maackiain
Isoflavone 2'-hydroxylase
Isoflavone reductase
Vestitone
Vestitone reductase
7,2'-dihydroxy-4'-methoxyisoflavanol
2',7-dihydroxy-4'-methoxyisoflavanol dehydratase
Medicarpin
2'- isoflavone O-methyltransferasePterocarpan synthase
(-)-Vestitol
(-)-Sativan
Pterocarpan reductase
2'-hydroxydaidzein
2'-hydroxydaidzein reductase
2'-hydroxydihydrodaidzein
Pterocarpan synthase
(6aR, 11aR)-3,9-dihydroxypterocarpan
3,9-dihydroxypterocarpan 6a-monooxygenase
(-)-Glycinol
Trihydroxypterocarpan dimethylallyltransferase
(6αS,11αS)-dimethylallyl-3,6α,9-trihydroxypterocarpan
Glyceollin synthase
Glyceollin
Genistein
Isoflavone 2'-hydroxylase
2'-hydroxygenistein
2,3-dehydrokievitone synthatase
2,3-dehydrokievitone
Kievitone
Kievitone synthase
Biochanin A
Isoflavone 4'-O-methyltransferase
Isoflavone 3'-hydroxylase
Pratensein
A
B
Figure 1.11: Further biosynthesis of daidzein (A) and genistein (B). Figure based on
Andersen and Markham (2006).
Chapter 1
24
a)
b)
c)
d)
e) f)
As mentioned above, isoflavones could be conjugated in which isoflavone aglycones
(genistein, daidzein, Biochanin A and formononetin) could be converted to conjugate forms in
two steps.
Figure 1.12: Structures of biochanin A (a), formononetin (b), genistein (c), daidzein (d),
medicarpin (e) and maackiain (f) (Harborne et al., 1999).
Figure 1.13: Schematic of phenylpropanoid enzymes involved in the isoflavonoid
pathway in a plant cell. IFS, located in endoplasmic reticulum (ER), produces the
isoflavone aglycone (1) which then converts into isoflavone glycosides (2), followed
by malonylation of the isoflavone glycosides generated isoflavone malonyl
glycosides (3). The isoflavone conjugates go to the vacuole for storage (Dhaubhadel
et al., 2008).
Introduction
25
First the aglycones undergo a glycosylation reaction which is catalysed by a
glycosyltransferase to produce isoflavone glycosides. Then the respective malonyl derivatives
are generated by malonyltransferase (Du et al., 2010). Generally, conjugated glycosyl- and
malonyl- forms of the aglycones are more soluble and stable leading to easier transportation
to vacuoles or any other site of accumulation (Figure 1.13).
1.4.5. Health benefits of isoflavonoids in plants
When plants colonized land, they encountered a new situation that was characterized by more
UV light than in the sea, as well as animals (herbivores and insects) and deleterious
microorganisms (bacteria and fungi) which did not exist in their oceanic environment.
Obviously, plants cannot run away when challenged by an herbivore or exposed to more light.
Plants had to find ways to cope with these challenges to their survival. They thus began to
generate secondary metabolites to ensure their health and fitness. Isoflavonoids, as secondary
metabolites, have diverse functions in plant health including anti-microbial activity, plant-
microbial interactions and UV protection.
1.4.5.1. Isoflavonoids in nitrogen fixation
Fabaceae have a symbiotic association with nitrogen–fixing soil bacteria of the genus
Rhizobium. This symbiosis leads to the formation of root nodules, where biological nitrogen
fixation happens (Pueppke, 1996; Subramanian et al., 2007). In this two-way interaction,
isoflavonoids act as chemoattractants and regulate gene expression in the rhizobia. Legume–
rhizobial nodulation starts with a signal exchange between the symbiotic partners initiated by
passing isoflavonoid compounds through the plant roots (Peters et al., 1986). The rhizobial
symbionts distinguish specific isoflavonoid signals produced by compatible host Fabaceae,
leading to the production of lipo-chitooligosacharides (LCOs). These LCOs act as signals
between the bacterium and the plant and cause expression of many genes which are
responsible for nodulation (Long, 1989). The type of isoflavonoid molecules exuded by the
Chapter 1
26
plant and the ability of the rhizobial species to recognise these molecules start nod-factor
biosynthesis. This constitutes the earliest step in determining host specificity (Fisher and
Long, 1992). On the roots of a compatible host plant, the rhizobia’s nod signals induce a
series of physiological and developmental responses that result in the formation of a
functional nodule. One of these responses is cell division in cortex cells to begin the nodule
primordium. In some Fabaceae, cell division in the nodule primordia is preceded by an
inhibition in auxin transport that is caused by isoflavonoids. This may indicate an important
role played by isoflavonoids in regulating auxin transport during nodulation (Mathesius et al.,
1998). Biochemical analysis of soybean roots confirmed that isoflavone levels increased in
response to Bradyrhizobium japonicum treatment (Cho and Harper, 1991a). It was suggested
that the nod signalling pathway regulated increases in isoflavone levels, because higher
isoflavone content was found in hypernodulating soybean mutants (Cho and Harper, 1991b).
In soybean, it has been also proven that isoflavones are necessary for nodulation of roots due
to their ability to induce the nodulation genes of Bradyrhizobium japonicum (Brechenmacher
et al., 2010). The expression of isoflavone synthase was shown to be particularly induced by
B. japonicum. When IFS was silenced using RNA interference in soybean hairy roots, these
plants had severely reduced nodulation. Even pre-treatment with B. japonicum or exogenous
application to the root system of either of the major soybean isoflavones, daidzein or
genistein, was unsuccessful to restore normal nodulation (Subramanian et al., 2006).
1.4.5.2. Anti-microbial activity of isoflavonoids
In 1950s, a simple test with plant extracts showed for the first time that plants have ability to
inhibit the growth of plant-destroying fungi (Virtanen et al., 1957). It was suggested that there
are some anti-microbial substances in these plants, which can be used to breeding varieties
showing higher resistance to fungi and bacteria. In alfalfa, plant extracts from non–flowering
plants absolutely inhibit the growth of Fusarium nivale. The anti-microbial factor of red
Introduction
27
clover was isolated in pure form in 1955, as 7-hydroxy-4’-methoxyisoflavone. The leaf
extraction of red clover can inhibit Sclerotinia trifoliorum and Fusarium nivale. After these
first studies, numerous publications on the anti-microbial activity of isoflavones and their
mode of action appeared (Bredenberg, 1961; Kramer et al., 1984; Vanetten, 1976).
1.4.5.3. Isoflavonoids in UV protection
UV irradiation on plants has harmful effects such as DNA damage resulting in reduced
growth and photosynthesis (Germ et al., 2005; Ruhland et al., 2005). Plants show a range of
protective mechanisms when exposed to UV radiation (Stapleton, 1992). These protective
mechanisms comprise the following: increasing the photo-repair activity of DNA
(Bredenberg, 1961; Pang and Hays, 1991), enhancing the synthesis of UV-absorbing
compounds such as isoflavonoids by inducing genes involved in isoflavonoid biosynthesis
(Hahlbrock and Scheel, 1989; Li et al., 1993; Lois, 1994); accumulating phytoalexins (Creasy
and Coffee, 1988); and enhancement of various types of defence-related gene expression.
Among these, the accumulation of UV-absorbing compounds is the most important
mechanism in shielding the plant from UV radiation (Robberecht and Caldwell, 1983;
Schmelzer et al., 1988). The accumulation occurs in the critical range of 230-380 nm, where
damage happens from UV irradiation. Isoflavonoids, as major protective agents, accumulate
in the vacuoles of epidermal cells and hypodermal layers of leaves and stems to strongly
absorb UV irradiation in vivo as well as in vitro (Matern et al., 1983; Schnabl et al., 1986).
1.4.6. Genetic engineering of isoflavonoid biosynthesis
Soon after the discovery of the advantages of isoflavonoid compounds in plant defense,
nodulation and human health, the question arose whether these benefits could be enhanced by
increasing the amount of isoflavonoids through metabolic engineering of isoflavonoid
biosynthesis. Some studies were then performed to address this question. A preliminary report
of over-expression of IFS genes in soybean showed that independent transgenic lines had
Chapter 1
28
significantly higher levels of seed isoflavones compared to controls (Jung et al., 2003). In
another study, the phenylpropanoid pathway-activating transcription factors C1 and R
belonging to maize were introduced to soybean. As a result, the total isoflavone level
increased in soybean seeds (Yu et al., 2003). In another attempt, the IFS gene, isolated from
Medicago truncatula Gaertn., was expressed in alfalfa (Medicago sativa L.). The result
indicated that the transgenic lines which were exposed to UV-B or pathogens produced
additional quantities of isoflavones, including formononetin and daidzein (Deavours and
Dixon, 2005). Several attempts have been undertaken to introduce isoflavonoid biosynthesis
in non-Fabaceous crops. Over-expression of IFS in Arabidopsis, tobacco and maize results in
genistein accumulation in these non-Fabaceous plants (Liu et al., 2002a). However, the
quantity of isoflavonoids present is generally low compared to that in soybean. Even
introduction of putatively rate-limiting enzymes of the pathway, such as PAL, CHS and IFS,
has a limited effect on isoflavonoid yields in soybean (Lozovaya et al., 2007). There are
several factors that limit isoflavonoid accumulation in a heterologous target plant. These
include the limited IFS activity itself, limitation of precursor pools, and, most importantly,
competition between IFS and flavanone 3-hydroxylase, which both use naringenin as
substrate (Liu et al., 2002b). Once these limiting factors can be controlled, optimization of
isoflavonoid biosynthesis in non-Fabaceous plants should be possible to expand the delivery
of dietary isoflavonoids and to achieve new sources for the more complex bioactive
isoflavonoids. Researchers have applied different approaches to conquer these limitations.
One approach is to physically link the proteins in the pathway. Expression of an IFS–CHI
fusion protein enhanced production of isoflavonoids in Arabidopsis. This approach suggests
that joining these sequential enzymes to each other improves pathway flux (Tian and Dixon,
2006). Over-expression of a synthetic maize transcription factor that triggers the expression of
multiple genes that encode flavonoid biosynthetic enzymes in soybean seeds resulted in lower
Introduction
29
isoflavonoid levels (Yu et al., 2003). In these plants, the activation of the flavonoid
metabolizing enzymes that use naringenin and a lack of activation of downstream isoflavone-
biosynthetic enzymes led to decreased genistein and increased daidzein levels. Co-expression
of the transcription factor with a silencing construct targeting flavanone 3b-hydroxylase
(F3H), the competitor of IFS for naringenin, resulted in increased total isoflavone content (Yu
et al., 2003). Silencing of F3H reduced flavonoid biosynthesis and increased isoflavonoid
accumulation.
1.5. Role of secondary metabolites in plant defense response
In the past, secondary metabolites were considered to be waste products resulting from
mistakes of primary metabolism (Calsamiglia et al., 2007), and were therefore assigned little
importance to plant metabolism and growth. It has since become clear that such views are
largely inaccurate and misguided, as the close relationship between plant secondary
metabolism and defence response had been widely recognized (Bennett and Wallsgrove,
1994). Plants respond to attack by pathogens, and other biotic and abiotic stresses, by
activating an array of defence mechanisms. Biosynthesis of secondary metabolites is one of
the plants’ mechanisms to defend themselves against pathogens (Iriti and Faoro, 2009). The
phytoalexins could not only kill the invading pathogens, but also delayed their propagation,
thereby enabling the plant to launch more sophisticated yet slower dynamic defense responses
(Grosskinsky et al., 2011). Many different secondary metabolites are associated with the
phytoalexin phenomenon. The analyses of plant–pathogen interactions, incorporating new
techniques, revealed that the most important classes of phytoalexins are derived from the
phenylpropanoid pathway, the isoprenoid pathway and from different alkaloid forming
pathways (Iriti and Faoro, 2009).
Chapter 1
30
1.5.1. Constitutive and induced antifungal compounds
Fungal spores that come into contact with a plant surface germinate only under the right
microclimatic conditions (temperature, humidity, light conditions, etc.). A germinated fungal
spore has to break several lines of the plant’s defences before reaching a living cell. The first
line is barrier such as cuticles that hamper pathogen penetration into the host. The next are
chemical barriers such as exudate compounds which inhibit spore germination and germ tube
elongation. These components are called constitutive (also named preformed or
phytoanticipins) (Grosskinsky et al., 2012). If all these barriers are not sufficient to prevent
germination of the fungal spore and penetration of the hyphae through the epidermis, the plant
still has several responses to the intrusion. After penetration of pathogen, the plant primary
and secondary metabolism are induced by changes in sugar levels which result in changes in
photosynthesis rates. As a consequence, antimicrobial peptides or enzymes are induced
(Roitsch et al., 2003). These antimicrobial compounds mainly belong to two classes of
substances of either proteinogenic origin or derived from secondary metabolism (Grosskinsky
et al., 2012). The antimicrobial compounds which derived from secondary metabolism (so-
called phytoalexins) activated only in the presence of the pathogen in living host cells close to
the infection site.
1.5.2. Antifungal compounds: definition and classification
The antifungal compounds derived from secondary metabolism can be classified into two
major groups: constitutive antifungal substances, which are preformed in the plant, and
induced antifungal constituents, or phytoalexins, that are generated after infection. Both
groups can be either classified according to their taxonomic distribution or their chemical
structures. When constitutive antifungal substances were first described, they were called
‘prohibitins’ (Schmidt, 1933). Later the word prohibitin was restricted to those pre-infectional
plant metabolites which are normally present in concentrations high enough to inhibit most
Introduction
31
fungi (Ingham, 1973). In other plant species, the concentration of an antifungal substance may
normally be low, but may rise dramatically after infection in order to combat attack by micro-
organisms; this type of constituent is named ‘inhibitin’. A third type of constitutive
compound, ‘post-inhibitin’, is described as “an antimicrobial metabolite generated by plants
in response to infection, but whose formation does not involve the elaboration of a
biosynthetic pathway within the tissues of the host”. In 1994, the new term “phytoanticipin”
was suggested to substitute for all terms which had been used for any types of low molecular
weight antifungal metabolites that are present in plants before challenge by micro-organisms.
Phytoanticipins are assumed to have a role in disease resistance of plants (Van Etten et al.,
1994). Phytoanticipins are normally present in the plant in an inactive, bound form, but are
converted into the active antifungal substance after infection by means of a short and simple
biochemical reaction. This process of activation only takes a short time, because the
enzyme(s) needed for the reaction are already present in the uninfected plant but in a different
compartment. Damage or infection of the plant brings together the enzyme and inactive form
of the compound to produce the active phytoanticipin. In contrast, phytoalexin production
may take two or three days. The term phytoalexin was proposed to describe a chemical
molecule generated by potato in response to infection by Phytophthora infestans (Müller and
Börger, 1940). With more work on the common bean, the term phytoalexin was defined as an
antibiotic compound that inhibits the growth of micro-organisms pathogenic to plants (Müller,
1956). The phytoalexin can also be induced in plants by means of abiotic factors, e.g. UV
irradiation. The definition of phytoalexin was thus redefined to “anti-microbial compounds
that are synthesized and accumulated in plants through a metabolic sequence after exposure to
either micro-organism or chemical or environmental factors” (Van Etten et al., 1994).
Determination whether a compound is constitutive or induced is sometimes difficult because
the same compound may be a preformed antifungal substance in one species and a
Chapter 1
32
phytoalexin in another. For example, the flavanone sakuranetin is a constitutive antifungal
compound in the leaves of blackcurrant (Ribes nigra L., Grossulariaceae) (Atkinson and
Blakeman, 1982) and of Hebe cupressoides (Scrophulariaceae) (Perry and Foster, 1994), but
induced in the leaves of rice (Oryza sativa L., Gramineae) (Kodama et al., 1992).
Furthermore, some compounds may be a phytoalexin in one organ and constitutive in another
part of the same plant, such as momilactone A, which occurs constitutively in rice husks
(Kato et al., 1973) and rice stems (Lee et al., 1999) but is a phytoalexin in rice leaves
(Cartwright et al., 1981). Thus, phytoalexins are identified by the dynamics of biosynthesis
and function, not by the chemical, structural, or taxonomic distribution, or the biosynthetic
pathway through which they were formed (Grayer and Kokubun, 2001). Classification of
phytoalexins can be based on their taxonomy or their chemical structures.
1.5.3. Phytoalexins in non-Fabaceous plants
Soon after the discovery of the first phytoalexin, many other phytoalexins were identified in
both Monocotyledoneae and Dicotyledoneae. Some of the types of phytoalexin found in the
monocotyledons are the same as those in the dicotyledons. However, there are many cases
where the phytoalexin generated is exclusively characteristic of a particular monocot genus.
For example, infected bulbs of the onion Allium cepa L. produce two aliphatic diones, 5-
hexylcyclopenta-1,3-dione and the 5-octyl analogue (Tverskoy et al., 1991). Another
structurally unique phytoalexin is yurinelide, a benzodioxin-2-one produced by the monocot
Lilium maximowczii in the bulb (Monde and Takasugi, 1992). Phytoalexins can also be unique
at the family level. For instance, sulfur-containing indoles are found uniquely in the family
Cruciferae. However, in some other families, a variety of phytoalexins belonging to different
compound groups are produced, e.g., in the Moraceae (Grayer and Harborne, 1994). In
Asteraceae, although the antifungal compounds identified are all phenolic, they belong to
different chemical subclasses. Even in the two species of the Helichrysum genus, two types of
Introduction
33
phenolics were recorded: phloroglucinol isolated from H. decumbens (Tomás-Lorente et al.,
1989) and methylated flavonoids from H. nitens (Tomás-Barberán et al., 1988). In the
Gramineae, the range of phytoalexins is even wider, e.g., saponins in oats (Crombie et al.,
1984), an alkaloid in barley (Wippich and Wink, 1985), fatty acids in rice (Kato et al., 1973),
phenolics in sorghum (Jambunathan et al., 1986) and alkadienals in wheat (Spendley et al.,
1982).
1.5.4. Phytoalexins in legumes
Phytoalexins are probably more extensively researched in Fabaceae than any other plant
family. During the 1970s and 1980s alone, more than one hundred different phytoalexins were
isolated and identified from over 500 species of the Fabaceae subfamily Papilionoideae.
Isoflavones are one of the major groups of phytoalexins which mainly occur in species of
Fabaceae. The isoflavones daidzein, genistein, biochanin A, formononetin and glycitein are
regarded generally as precursors of phytoalexins. For instance, formononetin is a precursor of
the phytoalexins maackiain and medicarpin in red clover and glycitein is a precursor of the
phytoalexin glyceollin in soybean. However, it has been shown that daidzein is capable of
inhibiting the growth of Fusarium culmorum, while glycitein and formononetin can reduce
mycelial development in Aspergillus ochraceus (Kramer et al., 1984). Therefore, they may
also be functionally classified as phytoalexins. Another study found that formononetin was a
good inhibitor of mycelial growth in Aphanomyces euteiches (Van Etten et al., 1994).
In red clover, biochanin A and genistein were found to exhibit antifungal activity on
Rhizoctonia solani (Bredenberg, 1961) and Sclerotium rolfsii (Weidenborner et al., 1990). In
red clover, maackiain was found in leaves which had been inoculated with Helminthosporium
turcicum (Higgins and Smith, 1972). It is found that treating red clover roots with the elicitor
CuCl2 increased the amount of maackiain and formononetin within 24 h after the start of
treatment (Tebayashi et al., 2001). It has also been reported that two greenhouse-grown red
Chapter 1
34
clover cultivars (‘Azur’ and ‘Start’) were used to induce isoflavone production in the foliage
through the use of the elicitors acetic acid (50,100, 250 and 500 mM), yeast extract (1, 2, 3
and 4 g/l), and chitosan (125, 250, 500, and 1000 mg/l) at the late vegetative stage. The leaves
were collected 2 and 8 days after spraying to measure the amount of isoflavone. HPLC did not
reveal any differences between the various concentrations of each elicitor nor between the
various elicitors (Sivesind and Seguin, 2006). Table 1.1 lists some phytoalexins that have
been identified in Fabaceae.
1.5.5. Strengthening plant defense by phytoalexin engineering
Soon after phytoalexins in plants were discovered, a question was raised about phytoalexins:
can manipulation of the amount or type of phytoalexins enhance resistance in plants? No
reports on phytoalexin engineering were published until 1990, when it was shown that
Species Chemical type Phytoalexins
Trifolium pratense
Pterocarpan
Medicarpin
Trifolium pratense Pterocarpan Maackiain
Glycine max Pterocarpan
Glyceollin II
Vicia faba
Furanoacetylene
Wyerone
Laburnum anagyroides
Isoflavone
Wighteone
Arachis hypogaea Stilbene
Resveratrol
Lotus uliginosus Isoflavan
Vestitol
Medicago sativa Isoflavanone
Sativanone
Phaseolus vulgaris Isoflavanone
Kievitone
Vigna anguiculata Benzofuran
Vignafuran
Phaseolus vulgaris Isoflavan Phaseollinisoflavan
Table 1.1: Phytoalexins in legumes.
Introduction
35
expression in tobacco of a gene from peanut (Arachis hypogeae L.) resulted in synthesis of
the stilbene phytoalexin, resveratrol. This resulted in increased resistance to infection by
Botrytis cinerea in tobacco (Hain et al., 1990). Resveratrol was a good choice for such
pioneering work, because it is produced in a single enzymatic step from p-coumaryl-CoA and
malonyl-CoA, precursors that exist in all plants. Subsequently, the resveratrol synthase gene
from grapevine was introduced into other plants. In tomato (Solanum lycopersicum L.), the
level of resistance to Phytophthora infestans increased, but there was no effect on resistance
to Alternaria solani and Botrytis cinerea (Thomzik et al., 1997). In alfalfa (Medicago sativa
L.), transgenic plants containing the resveratrol synthase gene, showed higher resistance to
Phoma medicaginis (Hipskind and Paiva, 2000; Thomzik et al., 1997). Further evidence has
revealed that induction of phytoalexins is an important mechanism of plants to defend
themselves against fungal pathogens. These findings have encouraged scientists to explore the
possibility of strengthening crop plant defences by manipulating phytoalexin metabolism. The
following approaches have been proposed to increase resistance using phytoalexins: (1)
altering the regulation of existing phytoalexin systems, (2) effecting single-enzyme changes in
existing pathways to change phytoalexins into more effective forms, and (3) introduction of
heterologous or genetically modified biosynthetic enzymes so as to generate phytoalexins that
are new to the host plant. However, in order to design transgenic approaches for the
modulation of phytoalexin levels, we need comprehensive knowledge about the different
biosynthetic pathways involved and possible cross-regulation and also competition between
different enzymatic branches (Grosskinsky et al., 2012).
1.5.5.1. Altered regulation of existing phytoalexins
Theoretically, over-expressing an endogenous gene involved in phytoalexin biosynthesis can
enhance accumulation of phytoalexin more quickly and to higher levels than in the parent
plant. In order to test this idea, hamster 3-hydroxy-3-methylglutaryl coenzyme A reductase
Chapter 1
36
(HMGR) was introduced into tobacco (Chappell et al., 1995). Although the level of
cycloartenol, a sterol precursor, increased more than 100-fold, total sterol accumulation
increased only 3- to 10-fold. It was suggested that in wild-type tobacco, HMGR could be the
rate-limiting step in sterol synthesis. Also in the cotton cultivars, resistance to bacterial blight
is only controlled by single major genes. Resistance of these plants is not as strong as that of
cotton plants which have multiple resistance genes. Although the same phytoalexins
accumulate in cotton by both single resistance genes and multiple resistance genes, in plants
having multiple resistance genes, accumulation is quicker and stronger than in plants with
single resistance genes (Shevell, 1985). Phytohormones also play an important role in
regulation of phytoalexin biosynthesis. For instance, an increasing in the amount of cytokinin
could lead to higher accumulation of different phenolic acids which can act as potential
phytoalexin precursors in tobacco (Grosskinsky et al., 2011).
1.5.5.2. Altered pathways
Certain single genes are able to enhance the anti-microbial activities of phytoalexins.
Manipulation of these genes can increase resistance to pathogens. For example, the species of
cotton (Gossypium) vary in the proportion of methylated to non-methylated sesquiterpenoid
phytoalexins. Non-methylated phytoalexins have been shown to be more toxic to some fungal
(Mace et al., 1985) and bacterial pathogens (Essenberg et al., 1990) than methylated
phytoalexins. Interspecific crosses increased the degree of O-methylation of desoxyhem-
igossypol. As a result, cotton plants became more susceptible to disease caused by
Verticillium dahliae (Bell et al., 1994; Essenberg et al., 1990). An O-methyltransferase that is
specific for 6-O-methylation of desoxyhemigossypol has been purified and a cDNA clone for
it has been isolated and functionally expressed (Liu et al., 2002b). Use of this clone for RNAi
or anti-sense silencing of the gene may increase phytoalexin activity in the cultivated species
of cotton.
Introduction
37
1.5.5.3. Introduction of new phytoalexins to plants
It was proposed that phytopathogens may have evolved mechanisms for detoxifying or
tolerating the phytoalexins of their hosts through natural selection. Since host–pathogen
interactions are frequently very specific, introduction of a new type of phytoalexins which is
unfamiliar phytoalexins for pathogen could increase resistance (Grosskinsky et al., 2012).
Due to the great progress in gene cloning over the past decade, manipulation of phytoalexin
biosynthetic pathways through introduction of heterologous genes has become possible. There
are numerous reports on the introduction of heterologous genes involved in phytoalexin
biosynthetic pathways, mainly in isoflavonoid phytoalexins of legumes. One report which
surveyed the toxicities of both enantiomers of maackiain and pisatin to 36 fungal isolates
representing 19 species showed that production of non-host phytoalexins in transgenic plants
has potential to increase resistance of red clover and garden pea to some fungal pathogens
(Delserone et al., 1992). Moreover, the opposite enantiomer of the host phytoalexin was often
more inhibitory to a pathogenic fungus than the naturally-occurring enantiomer. Since some
legumes, such as pea, produce (+) enantiomers, and others, such as alfalfa, synthesise (-)
enantiomers, interest has been shown in the re-direction of isoflavonoid biosynthesis to the
opposite enantiomer (Van Etten et al., 1989). The following should be noted when
introducing foreign phytoalexins: genetic manipulations to introduce new phytoalexins will be
successful only if the biosynthetic enzymes are provided with enough precursors. If precursor
levels are too low, not only the quantities of phytoalexins will be unacceptable, but
competition of the phytoalexin pathway with other pathways that require the same precursor
may have ill effects on the plant's health.
1.6. Application of biotechnology in plant breeding
Traditional breeding has been greatly successful in improving the nutritional value of food
and feed. However, limitation in available genetic resources has obstructed progress in
Chapter 1
38
development of new cultivars via traditional plant breeding methods. The boost in world
population and climate change have rapidly increased the demand for food. Traditional plant
breeding methods are too time-consuming to keep up with the dramatically increasing food
demands. To avoid a food crisis, new techniques are needed to aid conventional plant
breeding for crop improvement. Biotechnology is considered to be a powerful tool to broaden
genetic resources and speed up the breeding process. The plant biotechnology era began in the
early 1980s with the reports of generating transgenic plants via Agrobacterium (Bevan et al.,
1983; Fraley et al., 1983). The discovery of gene transformation techniques has opened a
gateway for further investigation of the genetic mechanisms of growth and development of
plants leading to facilitated plant breeding.
1.6.1. Tissue culture as a pre-requisite for gene transformation
A great deal of effort and attention has been paid to the genetic mechanisms of tissue culture
response and regeneration capacity in forage legumes to take advantage of gene
transformation techniques. Forage legumes have a poor ability for plant regeneration in tissue
culture. Therefore genetic transformation of these plants has faced limited success. In red
clover, studies showed that the regeneration of red clover in vitro could not be accomplished
using protocols developed for other tissue cultures. Numerous researchers have tried to
develop a specific protocol for red clover regeneration from callus culture. They confirmed
that all phases of tissue culture and plant regeneration can be performed. But the capability of
plant regeneration in red clover is low and most of the protocols to regenerate red clover are
genotype-dependent. Moreover, the studies showed that different genotypes have different
responses to a particular media system. Therefore, finding the appropriate combination of the
genotype which has a superior regeneration capacity and media system that allows the
genotype to have the highest capacity for regeneration in red clover is vital. It has also been
proven that the capability of regeneration in red clover callus is controlled by additive genetic
Introduction
39
effects (Myers et al., 1989). This trait can be heritable, along with its response to recurrent
selection (Quesenberry and Smith, 1993). Understanding the genetic mechanism of tissue
culture response and regeneration capacity can assist in the selection and breeding of
genotypes with high regeneration capacity.
1.6.2. Development of tissue medium in red clover
The major barrier for genetic transformation experiments is low efficiency of red clover
regeneration in both callus and cell suspensions. Over the last 30 years, the barrier has
challenged the researchers who have focused on red clover genetics transformation. In 1979,
the first mention appeared of successful whole plant regeneration from tissue cultures taken
from callus cultures of Trifolium pratense L. and Trifolium incarnatum L. (Beach and Smith,
1979). Soon after, more protocols including a series of different media for callus initiation,
embryo induction, embryo development or shoot development, and plant regeneration have
been developed for different parts of red clover used as explants (Maclean and Nowak, 1989;
McGee et al., 1989; Phillips and Collins, 1979). As mentioned above, Beach and Smith
introduced the first regeneration protocol for red clover (1979). They developed a protocol for
callus induction and plant regeneration from seedling hypocotyls and excised pistils of two
cultivars on B5 medium (Gamborg et al., 1968). The callus was induced with 10 µM each of
2, 4-dichlorophenoxy acetic acid (2, 4-D), α-naphthalene acetic acid (NAA), and 6-furfuryl
amino purine (kinetin). Hypocotyls produced more calli than pistils did. After four weeks,
calli were transferred to B5 medium supplemented with twice the normal concentration of
thiamine, 10 µM NAA, and 15 µM adenine. Many shoots appeared. Rooting of shoots was
achieved on B5 medium containing the higher concentration of thiamine and 1.1 µM NAA.
Producing callus from seedling sections was investigated by Phillips and Collins (1979).
Different combinations of the plant regulators (α-naphthalene acetic acid (NAA), indole-3-
acetic acid (IAA), p-chlorophenoxy acetic acid (CPA), 2,4-dichlorophenoxy acetic acid (2,4-
Chapter 1
40
D), 4-amino-3,5,6-trichloropicolinic acid (picloram), 6-furfuryl amino purine (kinetin), 6-
(gamma, gamma-dimethylallylamino) purine (2iP), and 6-benzylamino purine (BAP)) on the
basal media SH (Schenk and Hildebrandt, 1972), B5 and MS (Murashige and Skoog, 1962)
were conducted to find the combination which has higher efficiency in plant regeneration.
Based on these results, they concluded that the combination of 0.25 µM of picloram and 0.44
µM of BAP was the best for callus initiation and cell proliferation. However, they concluded
that the basal media were not sufficient for red clover regeneration. An improved basal
medium, L2, was developed by Phillips and Collins (1979). In the medium L2, the
concentration of NH4+ is lower than that in MS, whereas the concentrations of K
+, Mg
2+, and
Ca2+
higher than those in MS. Vigorous calli were produced on L2 medium by explants.
Maclean and Nowak (1989) evaluated regeneration of red clover using media sequences
which were suggested by Beach and Smith (1979) and Collins and Phillips (1982), and a
combination of media from both sequences. They reported that the best plant regeneration
happened when callus tissues derived from B5 callus initiation medium (Beach and Smith,
1979) were transferred onto L2 which contained 0.002 mg per liter picloram and 0.2 mg per
liter BAP medium using regenerative genotypes. A more rapid, one-step regeneration system
was developed by McGee et al. (1989) for Trifolium petiole segment of shoots which cultured
in vitro. They suggested that petioles placed on regeneration medium (L2 basal medium
containing 0.015 mg per liter picloram and 0.06 mg per liter) provided an optimal system for
regeneration.
1.6.3. Tissue culture response
There are numerous reports about the role of genetic components in plant response to tissue
cultures. Many publications have shown that nuclear genes control the in vitro response in
various plant species (Frankenberger et al., 1981; Komatsuda et al., 1989; Quimio and Zapata,
1990; Rakoczy-Trojanowska and Malepszy, 1993; Zhou and Konzak, 1992). In forage
Introduction
41
legumes, genetic studies have been reported for in vitro response, especially in alfalfa
(Bingham et al., 1975), red clover (Keyes et al., 1980) and alyceclover (Wofford et al., 1992).
In alfalfa, regeneration ability was found to be controlled by genetic components (Bingham et
al., 1975). Afterward, by investigation of diploid and tetraploid alfalfa cultivars, it was shown
that the genetic control of somatic embryogenesis was under the control of at least two
independent loci and this trait is highly heritable (Reisch and Bingham, 1980; Wan et al.,
1988). Later on, it was reported that the regeneration was genotype-specific and only a few
genotypes in certain cultivars have been found with the ability to regenerate plants through
tissue cultures (Arcioni et al., 1990). Moreover, to confirm the previous study, another
research was conducted to evaluate the genetic control of somatic embryogenesis in the alfalfa
cultivar ‘Adriana’. The study showed that the somatic embryogenesis was under genetic
control in the cv. ‘Adriana’ (Chappell et al., 1995). In red clover, genetic studies were
conducted to evaluate variation in callus cultures. The callus was induced with L2 medium
which was supplemented with Beach and Smith (1979) callus growth regulators. For
regeneration, callus was transferred to the L2 media which contained 0.001 mg per liter
picloram and 0.2 mg per liter BAP medium. The results showed that additive genetic effects
were most important in controlling somatic embryogenesis and this trait was highly heritable
(Myers et al., 1989). Also it is reported that regeneration traits could be transmitted by
crossing to the next generation (Maclean and Nowak, 1989). Additional research was
performed to investigate genetic analysis of regeneration in red clover. That experiment was
conducted with a three-step tissue culture protocol (callus induction, embryo induction, and
plant development) based on B5 medium supplemented with NAA and 2, 4-D as auxins, and
kinetin and adenine as cytokinins. Genotypes which are capable of regenerating in vitro have
been investigated for their ability to sexually transmit the regeneration trait and by conducting
Chapter 1
42
analyses of segregating populations. The result demonstrated that regeneration was highly
heritable in red clover (Quesenberry and Smith, 1993).
1.6.4. Breeding regenerable varieties in forage legumes
Numerous reports confirmed that regeneration in forage legumes is under the control of a
genetic component, which indicates that the trait is highly heritable. The trait could therefore
be responsive to selection programs, meaning that cultivars with a higher regeneration ability
can be achieved. Moreover a low capacity for regeneration may cause problems for use in in
vitro selection and gene transformation programs. Therefore, researchers decided to develop
varieties with high regeneration capacities. In alfalfa, the high heritability of embryogenesis
allowed an increase in the regeneration capacity from 12 to 67% of the cultivar ‘Saranac’ by 2
cycles of recurrent selection for regeneration (Bingham et al., 1975) and the Regen-S
germplasm, which is useful for transformation system, was released from the cultivar Saranac
(Bingham et al., 1975). Another report demonstrated that selection for regeneration capacity
increased regeneration from 3% to 50% in cv. ‘Adriana’ (Arcioni et al., 1990). In red clover,
despite several reports of whole plant regeneration (Beach and Smith, 1979; Mace et al.,
1985; Maclean and Nowak, 1989; Phillips and Collins, 1979), advanced cultivars of red
clover did not have a high regeneration capacity. Since then, it has been proven that
improvement in the frequency of plantlet regeneration from callus of red clover could
effectively be achieved by breeding and selection for embryogenic types (Keyes et al., 1980).
Therefore it seemed that developing germplasm lines of red clover with high regeneration
capacity was possible and was also needed to achieve high efficiency for in vitro selection and
transformation programs. It was demonstrated that crossing regenerable clones with non-
regenerable clones produced a progeny of which 29% was regenerable (Maclean and Nowak,
1989). Afterward, it was demonstrated that five cycles of recurrent selection for plant
regeneration of the red clover cultivar ‘Arlington’ from callus tissue culture was an efficient
Introduction
43
approach which increased regeneration from 4 to 72% (Quesenberry and Smith, 1993). Later,
the NEWRC cultivar, which had a high regeneration capacity, was developed from Arlington
(Smith and Quesenberry, 1995). Although there are other cultivars which have regeneration
ability such as F49 (Maclean and Nowak, 1989) and also the Chilean cultivars and
experimental lines Quiñequeli-INIA, Redqueli-INIA, Syn-IV and Syn-VI (Carrillo et al.,
2004), NEWRC has higher regeneration capacity compared with other cultivars and lines and
also NEWRC is the only red clover cultivar which has been successfully subjected to a
transformation program (Quesenberry and Smith, 1993; Sullivan and Hatfield, 2006). The
variety NEWRC was chosen for our study based on two reasons mentioned above: first, for
this variety’s regeneration trait, which yields higher efficiency of regeneration; and second, it
has been successfully used in a gene transformation experiment. NEWRC, like other red
clover varieties, is a highly heterozygous population. Thus, genotypes within NEWRC vary in
some traits. One of the traits which demonstrate variation is response to tissue culture. For
that reason, before conducting a transformation experiment, it is necessary to perform
preliminary tests with the purpose of finding highly regenerable genotypes within the
NEWRC variety.
1.6.5. Gene transformation
In many crops conventional breeding methods have been obstructed by the limited variation
in germplasm. The development and improvement of gene transfer is thus widely viewed as
the only way to develop new varieties. The methods used for gene transformation can be
classified into two groups: (a) mechanical and/or chemical gene transfer, including
microprojectile bombardment of explants and polyethylene glycol (PEG), electroporation and
protoplast-mediated transformation; and (b) biological gene transfer based on the use of
Agrobacterium tumefaciens. Given that biological gene transfer was successfully performed
Chapter 1
44
in red clover during this study, mechanical and/or chemical gene transfer methods will not be
discussed further.
1.6.6. Biological gene transfer
Agrobacterium tumefaciens is a soil bacterium capable of transferring DNA (the T-DNA) to
the genome of higher plants, where it is then stably integrated (Gheysen et al., 1991). The
discovery of the capacity of Agrobacterium tumefaciens has provided efficient tools to
introduce new genes into plants. During the gene transformation procedure, T-DNA, a
specific fragment of a Ti (tumor-inducing) plasmid, is transferred from the Agrobacterium to
the host plant cells. The transferred DNA (called T-DNA) is inserted into the plant nuclear
DNA (Binns and Thomashow, 1988). Normally, the genes located in T-DNA encode plant
growth hormones and cause the production of a plant tumor called a crown gall. However,
none of these genes are needed for T-DNA transfer from the Agrobacterium to the host plant:
the virulence genes (or Vir genes), located in the Ti-plasmid of Agrobacterium, are those
essential to gene transfer. Vir genes are not normally transferred to the plant, but rather
encode proteins required for the processing of the T-DNA movement from the Ti-plasmid to
plant. Those proteins form the channels in bacterial walls through which T-DNA exits and
enter to the plant genome (Gelvin, 2003). Thus, any fragment of DNA placed into the T-DNA
can be transferred to the plant and inserted into the plant nuclear DNA. Using recombinant
DNA methods, a selectable marker and/or genes of interest can be substituted between the T-
DNA borders.
1.6.7. Reporter genes
A reporter gene can encode an enzyme with an easily assayable activity. The reporter gene
can, in turn, report on the transcriptional activity of a gene of interest. In most reporter gene
constructs, the original promoter of the reporter gene is eliminated and is substituted with the
promoter of the gene to be studied. Then, the new gene is introduced into an organism
Introduction
45
(bacterium or animal or plant) using a gene transformation method. The expression of the
gene of interest is monitored by assaying for the reporter gene product. Use of a reporter gene
facilitates the study of expression of a gene for which the gene product is not known or is not
easy to identify. Some reporter genes that are well characterized and employed in plants
include cat (chloramphenical acetyl transferase), gus (β-glucuronidase) (Jefferson et al.,
1986), gfp (green fluorescent protein) (Morise et al., 1974), and lux (firefly luciferase) (Ow et
al., 1986). One reporter gene often used in plants is gusA, which encodes the enzyme β-
glucuronidase (GUS). This enzyme catalyses the cleavage of the chromogenic (color-
generating) substrate X-gluc (5-bromo-4-chloro-3-indolyl β-D-glucuronic acid), and produces
an insoluble blue color in the plant cells which have received the gusA through gene
transformation. Plant cells themselves do not normally have any GUS activity, so the
observation of a blue color when stained with X-gluc indicates the activity of the promoter
that drives the transcription of the gusA-chimeric gene in that particular cell. The GUS assay
is easy to carry out, sensitive, relatively inexpensive, and highly reliable. It does not require
specialised equipment and is highly visual (Jefferson et al., 1987). The introduction of an
intron into the gus gene avoids any gus expression in Agrobacterium, thus allowing direct
GUS assays in plant transformation (Vancanneyt et al., 1990).
1.7. Functional genomics: a window opened by gene
transformation
The innovation of the gene transformation technique enables us to investigate the function of
the gene of interest by manipulation of its expression through over-expression and gene
silencing approaches. An exhaustive amount of data is being generated from functional
genomics which could significantly assist breeding efforts to select a superior combination of
genes for higher performance in newer cultivars.
Chapter 1
46
1.7.1. Over-expression
Gene over-expression as a means to identify plant gene function has been used almost since
the first plant transformation protocols became available. The aim of application of over-
expression method is to investigate any phenotypic changes that could happen by increasing
the expression of a gene of interest. Over-expression of a gene is an appropriate alternative
when the knockout mutation of that gene results in plant death. To over-express a gene, the
upstream activation sequence is driven by a strong promoter such as the cauliflower mosaic
virus 19S (CaMV 19S) and 35S (CaMV 35S) which lead to constitutive transcription and then
high steady-state protein levels of the given gene.
1.7.2. Gene silencing
The term gene silencing is used to describe one of the regulatory mechanisms of gene
expression in eukaryotes (Kasai and Kanazawa, 2012). Gene silencing has been classified into
two classes in plant research: transcriptional gene silencing (TGS) and post-transcriptional
gene silencing (PTGS) (Kishimoto et al., 2013). TGS is associated with DNA methylation
which leads to prevention of transcription. Generally TGS happens when the genes involved
share sequence homology in their promoter regions. There are two kind of TGS in plants,
including cis-TGS and trans-TGS. During gene transformation, multiple copies of the T-DNA
can insert at the same chromosomal site. These T-DNAs can be arranged `head-to-tail' as a
direct repeat (DR), and `head-to-head' or `tail-to-tail' as an inverted repeat (IR). Transgenes of
T-DNAs that are organized as IRs often show low expression (Stam et al., 1997). This gene
silencing which is associated with the number of transgenes per integration site is named cis-
TGS (Vaucheret and Fagard, 2001). The trans-TGS occurs by either crossing or secondary
transfected (supertransformed) when the newly introduced transgene is homologous to a pre-
existing silenced promoter which already inserted at a different locus (Kishimoto et al., 2013).
The both types of TGSs cause the prevention of transcription if the genes involved share
Introduction
47
sequence homology in their promoter regions. However the PTGS, also known as RNA
interference (RNAi), acts through sequence-specific degradation of transcripts when there is
sequence homology within transcribed regions of the genes involved (Kishimoto et al., 2013).
1.7.2.1. History of RNAi
The RNAi phenomenon was first observed when scientists decided to generate a petunia with
deep purple flowers. To achieve the goal, they over-expressed the chalcone synthase (CHS)
gene (a pigment producing gene) by transferring the gene under the control of a 35S promoter
into petunia genotypes. However, the introduction of CHS unexpectedly caused the co-
suppression of the both endogenous and transgenic CHS, leading to a loss of flower
pigmentation instead of an increase. Although over-expression of CHS was expected to result
in the darkening of the flower petals, some transgenic lines had mosaic flowers with
unpigmented white sectors (Napoli et al., 1990). This phenomenon was initially coined “co-
suppression” to describe the loss of mRNAs of both the endogenous gene and the transgene
(Geley and Muller, 2004). Two years later, a similar phenomenon was observed in the fungus
Neurospora crassa (Romano and Macino, 1992). In order to boost the synthesis of orange
pigment, they introduced extra copies of a gene encoding a carotenoid synthesizing enzyme
into the fungus. Unexpectedly the engineered fungus became pigmentless. With more
investigation, they realized that the introduction of the transgene led to reduction of mRNA
levels of the endogenous genes. They named this phenomena quelling. Gene silencing was
first reported in animals by Guo and Kemphues (1995) who tried to knocked out expression of
the par-1gene using antisense RNA to investigate the gene function using Caenorhabditis
elegans as a model organism. At the time, a common method for studying gene function was
to inject large amounts of antisense single-stranded RNA (ssRNA) (Nellen and Lichtenstein,
1993). It was thought the pairing of antisense RNA with endogenous sense RNA leaded to
block translation. As expected, the injection of antisense RNA disrupted expression of par-1.
Chapter 1
48
Surprisingly, the injection of the sense-strand which was used as control also blocked par-1
mRNA expression. Later on, Fire et al. (1998) sought to provide an explanation for the
previously reported silencing of endogenous genes by co-suppression, quelling and sense
mRNA. They realized that the ssRNA of both sense and antisense samples was often
contaminated with double-stranded RNA (dsRNA). This led them to hypothesize that perhaps
dsRNA caused silencing. To test their hypothesis, they injected both ssRNA and dsRNA of a
gene into worm gonads. They found that dsRNA was over 100-fold more effective at
silencing than antisense ssRNA. They termed the phenomena RNA interference (RNAi) and
an entire new field of molecular biology was born. Soon after, extensive studies have been
performed to reveal the mechanism of RNAi.
1.7.2.1. Mechanism of RNAi
Although the RNAi pathway exists in eukaryotes, it has functionally and mechanistically
diversified among them. The process of RNAi can be moderated by either microRNAs
(miRNA) or small interfering RNAs (siRNA) in plants (de Faria et al., 2013). The siRNA and
miRNA pathways are partially overlapped (Swevers et al., 2011). The miRNAs originate from
endogenous non-coding RNA, while siRNAs are usually produced from long double-stranded
RNAs (dsRNAs) that are derived from inverted repeats, transgenes, transposable elements,
and viruses in plant cells (Bolognesi et al., 2012). Both miRNA and siRNA are generated in
several steps. The RNA polymerse II transcribes pri-miRNAs from endogenous genes. The
nuclear RNase III processes pri-miRNA into 70-nucleotide hairpin shaped structures, called
precursor miRNAs (pre-miRNAs). The plant exportin-5 orthologue hasty (HST) transported
the pre-miRNAs from the nucleus to the cytoplasm (Wu, 2013). Pre-miRNAs and dsRNAs
are digested into short miRNA duplex and small interference RNA (siRNA) by Dicers (Luo et
al., 2013). Dicers are large proteins which possess several domains including classical
dsRNA-binding motifs, helicase, PAZ, DUF283, and RNase III. These domains allow Dicers
Introduction
49
to be multifunctional enzymes (de Faria et al., 2013). Each of the miRNAs and siRNAs will
be unwound into two strands, namely the sense (passenger) strand and the antisense (guide)
strand by argonaute proteins. The passenger strand will be degraded, and the guide strand is
incorporated into the RNA induced silencing complex (RISC) in siRNA pathway or
microRNA-induced silencing complex (miRISC, also known as a microRNA
ribonucleoprotein complex (miRNP)) in miRNA pathway. The RISC or miRISC trigger the
target RNAs for translational repression or cleavage of fully complementary target mRNAs,
resulting in gene silencing (Manavella et al., 2012) (Figure 1.14).
Figure 1.14: The process of RNAi can be happened via either microRNAs (miRNA) or small
interfering RNAs (siRNA) in plants. Pathways of RNAi originate from dsRNA or endogenous hairpin
pre-miRNAs (Modified from Lingel and Sattler, 2005).
Chapter 1
50
1.8. Real-time PCR technique, a valuable tool for functional
genomic studies
The real-time, fluorescence-based quantitative reverse transcription real-time polymerase
chain reaction (qRT-PCR) is a powerful technique for detecting differences in gene
expression caused by genetic manipulation. qRT-PCR is more sensitive and specific for
identification of differences in expression of a target gene compared to conventional methods
of gene expression analysis such as Northern blot, ribonuclease protection assay and reverse
transcriptase PCR assay. In real time PCR, the data is collected by the detection of a
fluorescent labeling dye incorporated into newly synthesized PCR product during the early
phase of the reaction where the fluorescent signal is greater than the background fluorescence
signal. This point is called the cycle of threshold (Ct) or crossing point (Cp). However, like
conventional methods, qRT-PCR still faces challenges from errors caused by variation in the
quantity and integrity of the RNA template, RNA recovery, efficiency of cDNA synthesis and
differences between tissues or cells in overall transcriptional activity. To gain trustworthy
quantification of gene expression levels, these errors should be minimised by normalization.
Among numerous methods suggested to normalize qRT-PCR data, the use of reference genes
is the most widespread method. Absolute quantification and relative quantification are two
types of real-time quantification.
1.8.1. Absolute quantification
Although called absolute quantification, the quantification of an unknown sample is
calculated based on an external standard curve which is generated by a standard dilution
series with a known concentration sample (Bustin, 2000). A linear regression model that is
obtained from the standard curve allowed calculation of the absolute copy number of any
unknown sample relative to that standard curve. The slope of the standard curve presents the
amplification efficiency. In this method, the amplification efficiencies of all standards and
Introduction
51
samples need to be approximately equal to produce accurate results. Moreover, the
concentration of the dilution series should stay within the range of measurable levels of both
machine and assay. The external standard can be a reference sample, constructed from PCR
fragments, in vitro T7-transcribed RNA, or single-stranded sense-strand oligodeoxy-
ribonucleotides.
1.8.2. Relative quantification
Measuring an mRNA level of a given gene across multiple samples and expressing it relative
to the levels of another RNA is called relative quantification. Unlike absolute quantification,
relative quantification does not need a standard curve with known concentrations. As the
collected samples usually vary in tissue mass or cell number, experimental treatment, or RNA
quantity, correction of these sample-to-sample variations is necessary. Numerous methods
exist to correct (normalize) these errors in qRT-PCR data, but the most widespread is the use
of an internal control gene. The control gene can be co-amplified in the same tube in a
multiplex assay (as endogenous controls) or amplified in a separate tube (as exogenous
controls) (Livak, 1997; Morse et al., 2005; Wittwer et al., 2001). During the pre-genomic era,
the genes that have housekeeping roles in basal cell activities were chosen as control genes in
transcript expression studies. But further investigations proved that the expression level of
some of the best known and commonly used control genes were found to be unstable in
expression level across different treatments, different biological processes and even different
tissues. In other words, being a housekeeping gene does not always guarantee that the gene
has a stable expression level in cells. The use of unstably expressed housekeeping genes for
normalization purposes leads to severe misinterpretation of results (Remans et al., 2008).
Thus, the stability of the expression level of housekeeping genes should be validated carefully
prior to normalization. Accurate quantification of the expression level offered by qRT-PCR
can only be achieved by use of an experimentally validated housekeeping gene (also called a
Chapter 1
52
“reference gene”) (Lovdal and Lillo, 2009). Another concern is the number of reference genes
that should be used for normalization. In the past, the quantification of gene expression assays
was normalized by only one reference gene. However, further studies demonstrated that
multiple reference genes should be applied for normalization to avoid erroneous measurement
of the gene expression level (Suzuki et al., 2000; Vandesompele et al., 2002). To address
these concerns about normalization, the statistical algorithms including geNorm, NormFinder
and BestKeeper have been suggested. Depending on which one is being used, the obtained
result can be different.
Chapter 2: Regeneration and genetic
transformation of red clover
This chapter is adapted from:
Mehdi Khanlou K, Karimi M, Maroufi A, Van Bockstaele E (2011). Improvement of plant
regeneration and Agrobacterium-mediated genetic transformation efficiency in red clover
(Trifolium pratense L.). RESEARCH JOURNAL OF BIOTECHNOLOGY. 6(3). p.13-21.
Regeneration and genetic transformation of red clover
55
2.1. Introduction
In forage legumes, the main barriers to progress when developing new cultivars via
conventional breeding methods are the narrow genetic variation in their germplasm and
sexual incompatibility with related and wild species (Jaiwal and Singh 2003). In such cases,
genetic engineering must be applied to create new properties. Gene transferring, the widely
used in vitro technique to introduce foreign genes, has been viewed as the only way to enrich
germplasm with desirable genes across taxonomic boundaries when conventional breeding
was obstructed by limited variation in germplasm. The totipotency of single cells to develop
into whole plants (“regeneration ability”) is needed for successful application of gene
transformation techniques. Forage legumes are not acquiescent to in vitro manipulation and
have low frequencies of regeneration. Increasing the regeneration capability of forage
legumes has therefore become a goal in order to increase efficiency of gene transformation.
Two strategies have been applied to enhance regeneration capacity in forage legumes: the first
is to determine media composition as well as plant growth regulators to meet the requirement
for in vitro culture and the second is to screen for germplasm with high capacity of
regeneration, such as the highly regenerative alfalfa (Medicago sativa L.) germplasm Regen-S
(Bingham et al 1975).
Some additional factors can highly affect the efficiency of transformation. In legumes,
Agrobacterium-mediated transformation is the preferred method for gene introduction, and
strains of Agrobacterium tumefaciens vary in their transformation efficiency in legume crops
(Akbulut et al 2008; Surekha et al 2007). Manipulation of explants is another important factor
that might impact efficiency of gene transformation. For example, pre-culturing explants can
increase the ability for cell dedifferentiation and division leading to competence for
transformation (Ghorbel et al 2000).
Chapter 2
56
Red clover (Trifolium pratense L.) is one of the major forage legumes. It originated in
southern Europe and has acclimatized to most temperate regions. Evaluation of red clover
germplasm has revealed a low level of genetic variability for resistance to diseases
(Quesenberry et al 1996). In fact, despite all advantages that are offered by red clover, it is
susceptible to fungal diseases which results in poor persistence of red clover (Skipp and
Christensen 1990). Taking into account the low level of genetic variability for resistance to
diseases, as mentioned above, the genetic variation must be broadened via newly innovated
molecular techniques to develop new resistant varieties. An efficient regeneration protocol is
prerequisite for any genetic manipulations via modern molecular techniques. However,
regeneration protocols developed for other species have failed to regenerate red clover in vitro
(Beach and Smith 1979). Most primary studies have focused on developing specific protocols
for red clover regeneration using different combinations of media and plant regulators with
different sources of explants. However, all of the developed regeneration protocols were
genotype-specific and suffered from low frequencies of regeneration (Beach and Smith 1979;
Phillips and Collins 1979). Low regeneration frequency could be problematic for in vitro
selection or transformation. Taylor et al. (1989) declared that red clover lines did not
regenerate after a two-year selection for acetic acid tolerance. The importance of plant
regeneration for the application of in vitro techniques has encouraged researchers to develop
regenerative germplasm for red clover. Recurrent selection was applied to increase
regeneration capacity and as a result, the NEWRC red clover germplasm with high frequency
regeneration was released (Smith and Quesenberry, 1995). Soon afterward,
an Agrobacterium-mediated transformation protocol for red clover was reported
(Quesenberry, Wofford, Smith, Krottje, and Tcacenco 1996). To our knowledge this is the
only reported transformation protocol for red clover. In this protocol, a three-step-
regeneration system (Beach and Smith 1979) was used as the media sequence for
Regeneration and genetic transformation of red clover
57
regeneration, the strain EHA101 was utilized as carrier for a GUS reporter gene, and NEWRC
germplasm was employed as source of explants. Some studies suggested that there are
regeneration protocols which yield superior frequency for regeneration efficiency compared
with that of the regeneration protocol which was used in the existing transformation protocol
of red clover (Maclean and Nowak 1989; Phillips and Collins, 1979). Unpublished studies in
our laboratory indicated that all the potential ability of regeneration has not been employed in
the present transformation protocol. Also, factors that are reported to be effective on
efficiency of gene transformation for other plant species in the literature have not been
explored enough in the present transformation protocol for red clover. In this study, we
investigated the possibility of beneficial modification to the published transformation
protocol. Our strategy was as follows: first, to examine different media sequences to find one
that could exploit all of the potential regeneration capacity of explants used for gene
transformation; second, to assess different Agrobacterium strains to find the best one for gene
transformation; and third, to evaluate the effect of pre-culture of explants on transformation
efficiency. Our established protocol for transformation and regeneration was demonstrated to
be efficient for transformation with the pTJK136 binary vector containing the GUS (β-
glucuronidase) reporter gene by PCR and histochemical GUS assay on transformed red clover
plants.
2.2. Material and Methods
Seeds of NEWRC red clover germplasm were supplied by Kenneth H. Quesenberry,
Agronomy Department, University of Florida. The media compositions and plant growth
regulators used in all experiments are illustrated in Table 2.1. The basal salt B5 (Gamborg et
al 1968) and L2 (Phillips and Collins 1979) were used in this study. In all experiments tissue
culture media were gelled with 0.3% w/v gelrite (Duchefa, the Netherlands) and cultures were
kept in a growth chamber at 22 ± 2 °C with 16 h per day 40 μmol/m2/s illumination. In
Chapter 2
58
experiments for optimizing plant growth, some media were gelled with 0.7% Agar (Lab M).
Also, some plants were kept in an incubator whose environmental conditions were set at 24 ±
1°C under cool white fluorescent light with illumination of 170 μmol/m2/s in 16/8 h
(light/dark) photoperiod. The pH of all media was adjusted to 5.8 with a solution of 1 N of
KOH. All media were autoclaved at 121°C for 20 min. Plant growth regulators and antibiotics
were filter-sterilized through 0.2 µm filters and added to the cooled autoclaved medium.
Media were poured into either 15 × 90-mm sterile petri dishes or 9-cm-diameter glass jars for
all experiments.
2.2.1. Genotype screening
Seeds were surface-sterilized in 70% (v/v) ethanol for 2 minutes, then immersed in 50% (v/v)
sodium hypochlorite (Solvay, Belgium) supplemented with 0.5% (v/v) Tween 20 (Duchefa,
the Netherlands) for 20 minutes. Seeds were washed three times in sterile deionized water and
placed in petri dishes containing 20 ml of GM (Table 2.1). After 7 days, 120 seedlings
(original plants or genotypes) were transferred to glass jars (one seedling per jar) containing
GM and grown for 2 more months. A regeneration protocol developed by Beach and Smith
was employed to evaluate regeneration capacities of genotypes (Beach and Smith, 1979). For
each genotype, 18 petiole sections (approximately 5 mm) of young leaves were placed in 3
sterile petri dishes (6 explants per plate) containing 20 ml B5C (Table 2.1) and were
incubated in the growth chamber for 30 days. Subsequently, explants were transferred to B5E
(Table 2.1) and maintained for an additional 30 days. After a total of 60 days after initiation of
culture, formation of somatic embryos from calli was scored as following: 1 = no or limited
callus formation with no green embryos, 2 = good callus formation with few green embryos, 3
= excellent callus formation with abundant green embryos.
Medium Amount of sucrose Basal salt
Growth Regulators (mg L-1)
Proposed Use
Auxin Cytokinin
GM
% 2.0
B5
-
-
Germination
L2C
% 2.5
L2
0.06 mg l-1
PIC
0.1 mg l-1
BAP
Callus induction
SEL
% 2.5
L2
0.01 mg l-1
2,4-D
2 mg l-1
Adenine
Embryo induction
SPL
% 2.5
L2
0.002 mg l-1
PIC
0.2 mg l-1
BAP
Shoot induction
KB2
% 2.5
L2
2 mg l-1
2,4-D; 2 mg l-1
NAA
2 mg l-1
KIN
Callus induction
B5C
% 2.0
B5
2 mg l-1
2,4-D; 2 mg l-1
NAA
2 mg l-1
KIN
Callus induction
B5E
% 2.0
B5
2 mg l-1
NAA
2 mg l-1
Adenine
Embryo induction
B5S
% 2.0
B5
0.2 mg l-1
NAA
-
Shoot induction
Table 2.1: Composition of plant media and plant growth regulator combination used in this study.
Chapter 2
60
2.2.2. Optimizing plant growth conditions
After all 120 original plants showed stunted growth during the genotype screening test, an
experiment was conducted to investigate optimal growth conditions to keep the original red
clover plants as a resource of explants for a long term in vitro experiment. In this experiment,
2 plant growth media (B5 and L2), 2 types of gelling agents (gelrite and agar) and 2 growth
conditions (growth chamber and incubator) were tested. To find the best growth conditions,
48 out of the 120 original plants which had been used as source of explants for the genotype
screening test were removed from the glass jars, their roots were washed with distilled water,
dead leaves were removed, and they were incubated under different conditions (Table 2.2).
B5 (24 Plants)
Agar (12 Plants)
Gelrite (12 Plants)
Growth Chamber (6 Plants)
Incubator (6 Plants)
Growth Chamber (6 Plants)
Incubator (6 Plants)
L2 (24 Plants)
Agar (12 Plants)
Gelrite (12 Plants)
Growth Chamber (6 Plants)
Incubator (6 Plants)
Growth Chamber (6 Plants)
Incubator (6 Plants)
Basal salt Gelling agent Maintenance condition
Table 2.2: Schematic plan to find the best growth condition.
Regeneration and genetic transformation of red clover
61
2.2.3. Optimization of tissue culture media for regeneration
After 60 days, all plants were evaluated based on their appearance to investigate effects of
conditions on plant growth. Among genotypes which were found to be regenerative in the
genotype screening test, the genotypes of NEWRC7, NEWRC8, NEWRC19, NEWRC65 and
NEWRC68 were compared on three different media sequences (Table 2.3).
Five-millimeter sections of petioles from the original plants were placed on the B5C, L2C and
KB2 callus induction media and kept in the growth chamber for 30 days. The fresh weight of
callus from all genotypes was recorded and all calluses were also visually rated based on their
color at the end of callus induction period. Subsequently, petioles with a developed callus
were transferred onto somatic embryo induction media B5E, SEL and SPL and maintained in
the growth chamber for an additional 30 days. The numbers of embryos per explant were then
counted, and all cultures were also visually recorded for their color and condition. The
cultures were then transferred to shoot B5S, SPL and SPL induction media for another 30
days. At the end this culture period the numbers of developed shoots of each genotype were
recorded, and the appearances of developed shoots were rated.
2.2.4. Optimizing transformation parameters
Based on results of the optimization of tissue culture media test, the highly regenerable
genotype (NEWR65) and optimal media sequence (media sequence number 3) were chosen
for optimization of the gene transfer protocol. Plants of NEWR65 were propagated by
splitting off pieces of crown with an attached root and maintaining them under sterile
Number Callus induction Embryo induction Shoot induction Described by
1 B5C → B5E → B5S Beach and Smith (1979)
2 L2C → SEL → SPL Collins and Phillips(1982)
3 KB2 → SPL → SPL Proposed in this study
Table 2.3: Sequences of media used for the regeneration test.
Chapter 2
62
conditions in the growth chamber in glass jars containing L2 medium, 2.5% sucrose and 0.3%
gelrite without growth regulators.
2.2.5. Kanamycin sensitivity test
The petioles of young leaves from healthy plants which had been clonally propagated from
genotype NEWR65 were sectioned into small pieces of approximately 5 mm. Petiole sections
were transferred on KB2 containing 0, 25, 50, 75 and 100 mg l-1
of kanamycin. Five replicate
plates containing 6 explants were included for each kanamycin concentration. All explants
were incubated in a growth chamber for 30 days to induce calli. Then all explants were
transferred onto SPL containing the same concentrations of kanamycin and kept under the
same conditions for an additional 30 days. Subsequently all explants were placed on new SPL
media with the same concentrations of kanamycin and maintained under the same conditions.
We recorded the number of explants which had developed shoots after 90 days of culture.
2.2.6. Bacterial strains and vectors
Agrobacterium tumefaciens C58C1 RifR (pMP90) and EHA101 strains were used for plant
transformation. The pTJK136 binary vector (Kapila et al 1997) contained the nptII (neomycin
3'-O- phosphotransferase) selectable marker at the left border and the GUS (β-glucuronidase)
reporter gene (Jefferson et al 1987) at the right border (Figure 2.1). The GUS reporter gene
and neomycin phosphotransferase gene are under the control of the Cauliflower Mosaic Virus
“CaMV” 35S and nopaline synthase promoter, respectively. The reporter GUS gene is
interrupted by an intron which prevents any leaky expression of the reporter gene inside the
Agrobacterium cells. The binary vector pTJK136 was transferred to Agrobacterium
tumefaciens C58C1 RifR (pMP90) and EHA101 strains using a standard freeze-thaw method
(Chen et al 1994).
Regeneration and genetic transformation of red clover
63
2.2.7. Bacterial culture
Both Agrobacterium C58C1 RifR (pMP90) and EHA101 strains were cultured in 100 ml of
YEB liquid medium supplemented with 50 mg l−1
rifampicin and 100 mg l−1
spectinomycin.
Agrobacterium cultures were harvested by centrifuging at 4000 rpm for 20 min and then
pellets were resuspended in sterile water to an optical density (OD) of 1.5 at 600 nm.
2.2.8. Inoculation and co-cultivation
For transformation, petioles of young leaves from the plants which had been in vitro clonally
propagated from genotype NEWR65 and maintained aseptically in glass jars were used.
Petioles were cut into sections of approximately 5 mm and were pre-cultured on KB2 (15
petiole sections per plate) for 3 days prior to Agrobacterium infection. Petiole sections were
also freshly prepared on the infection day. The explants were inoculated by a drop of
overnight Agrobacterium cultures (C58C1 RifR (pMP90) or EHA101 strain) using a pipette.
All infected petioles as well as additional, uninfected petiole sections in 4 petri dishes (kept as
controls) were sealed with 3M micropore tape and co-cultivated for 48 h in the growth
chamber. Subsequently the explants were removed from the co-cultivation medium (KB2)
and rinsed 3 times with sterile water. The explants were transferred to KB2 medium
supplemented with 500 mg l−1
of timentin to remove excess bacteria and also with 75 mg l−1
of kanamycin sulphate (based on our previous study) for selection of transformed tissue. Half
of the uninfected petioles were transferred to KB2 medium with 500 mg l−1
of timentin
without kanamycin for plant regeneration. The remaining uninfected controls were transferred
Figure 2.1: A schematic diagram of the T-DNA region from the binary vector pTJK136. RB: the right
border of T-DNA; T-NOS: nopaline synthase terminator; GUS: β-glucuronidase gene; gusint: intron
region, P-35S: cauliflower mosaic virus 35S promoter, P-NOS: nopaline synthase promoter, nptII:
neomycin phosphotransferase gene; T-OCS: octopine synthase terminator, LB: left border.
Chapter 2
64
to KB2 medium with 75 mg l−1
kanamycin sulphate as a control for kanamycin selection. The
cultures were sealed and incubated for 30 days in the growth chamber. After 30 days of
incubation on KB2 callus induction medium, the percentages of petioles that had formed calli
were recorded to evaluate the efficiency of Agrobacterium strains. Afterwards, infected
explants were transferred to SPL medium with 500 mg l−1
of timentin and 75 mg l−1
of
kanamycin sulphate for regeneration under the same environmental conditions. Necrotic
petioles were discarded. Uninfected explants cultured on selection medium were transferred
to SPL medium with kanamycin sulphate for subculture. Also, uninfected explants without
selection were transferred to SPL medium with 500 mg l−1
timentin for regeneration. To
verify the transformation events, kanamycin-resistant embryos were transferred to SPL
medium containing 500 mg l−1
of timentin and 75 mg l−1
of kanamycin sulphate for shoot
production.
2.2.9. PCR analysis
Genomic DNA from putative transformants and non-transformed, control regenerants was
extracted from young leaves using DNeasy® Plant Mini Kit (Qiagen, USA). DNA
concentration and purity was measured with a NanoDrop® ND-1000 Spectrophotometer
(NanoDrop Technologies, USA). Two sets of primers designed from the 35S promoter and
the nptII coding region were used for PCR. Primers from the 35S promoter were as follows,
35SF: 5’- GAACGTCTTCTTTTTCCACGATGCT-3’ and 35SR: 5’-
CAGGACTAACTGCATCAAGAACACA-3’; the expected size of amplicon was about 554
bp long. Primers from the nptII coding region were as follows; nptIIF: 5’-
AGCGGCGATACCGTAAAGCACGA-3’ and nptIIR: 5’-AAGGGACTGGCTGCTATTGG
GC-3’, with an expected amplicon of 481 bp. The PCR reaction was carried out in 50µL
containing 100 ng of plant DNA, 20 pmol of each primer, dNTPs mix that was 0.25 mM of
each dNTP (Invitrogen, USA), 5µL of 10x reaction buffer (Applied Biosystems, USA) and
Regeneration and genetic transformation of red clover
65
1.25 unit of Taq DNA polymerase (Applied Biosystems). PCR was performed in a thermal
cycler (Applied Biosystems) with the following program: initial denaturation for 2 min at 94
°C, followed by 30 cycles of 15 s denaturation at 94 °C, 30 s annealing at 61 °C for nptII or
57 °C for 35S primers, 30 s elongation at 72 °C. The program ended with a final elongation
for 5 min at 72 °C. PCR products were separated by 1.5% agarose gel and detected under UV
light after ethidium bromide staining.
2.2.10. Histochemical GUS assay
GUS assay was carried out according to the procedure of Jefferson (Jefferson 1987) , with the
following modifications. The explants, including pieces of leaves, were immersed in 20 ml of
GUS substrate buffer (5 mM potassium ferrocyanide, 0.1M sodium phosphate buffer pH 7.0,
20 l of 0.3% v/v Triton X-100) at 37 °C for 25 minutes, and the substrate 5-bromo-4-chloro-
indolylgluc-uronide (X-gluc) was added. The explants were vacuum-infiltrated for 10
minutes, then incubated at 37 °C for 1 to 4 hours. The stained tissues were transferred into
95% ethanol for 2 h for removal of excess color. GUS-expressing regions in explants were
examined under a microscope (Leica DM IRB, Germany).
2.2.11. Statistical analysis
To find a highly regenerable genotype and optimal media sequence, a factorial experiment
with two factors (5 genotypes and 3 media sequences) in a completely randomized design
(CRD) with 5 replicates (6 explants per petri dish each) was carried out. Respective data for
callus fresh weight gain, number of embryos and number of developed shoots were recorded
after 30, 60, and 90 days of culture, respectively. Prior to analysis, data were square root
transformed for both number of embryos per explant and number of shoots per explant to
satisfy assumptions of the analysis of variance. To evaluate effects of 2 Agrobacterium strains
(C58C1 RifR (pMP90) and EHA101) and also of pre-culturing on efficiency of
transformation, a factorial experiment with 2 factors (Agrobacterium strains and pre-cultured
Chapter 2
66
explants) in completely randomized design with 9 replicates (15 explants per petri dish each)
was conducted. The number of petioles which developed a callus in each petri dish was
divided by the total number of explants in each petri dish to compute proportion of calli that
formed explants for each replication with each strain. Prior to analyzing data, all proportions
were arcsine-transformed to satisfy the assumption of normality. In all experiments,
each plate was considered as one replicate. Duncan's multiple range test was used to compare
means in experiments. All statistical analyses were performed using SAS 9.2 (SAS Institute
Inc., Cary, NC, USA).
2.3. Results
2.3.1. Genotype screening
The regeneration test was conducted for genotype screening to select those with higher
regeneration frequencies among 120 genotypes. According to the rates of the plates,
genotypes were classified into 4 groups according to regeneration capability: no, low,
moderate and high. Among the 120 genotypes used for the regeneration test, about 70% had
regeneration capacities. The genotypes NEWRC7, NEWRC8, NEWRC19, NEWRC65 and
NEWRC68, which varied in regeneration ability, were selected for further experiments. The
genotypes NEWRC65 and NEWRC68 showed high frequency of regeneration and genotypes
NEWRC7, NEWRC8 and NEWRC19 had moderate capacity of regeneration (Table 2.4).
Genotype number Rate in Petri 1 Rate in Petri 2 Rate in Petri 3 Regeneration ability
NEWRC7 2 3 2 Moderate NEWRC8 3 2 3 Moderate
NEWRC19 2 3 3 Moderate NEWRC65 3 3 3 High NEWRC68 3 3 3 High
Table 2.4: The score of 5 selected genotypes out of 120 genotypes (original plants) in genotype
screening test. Petri with low, medium and high regeneration ability were scored 1,2 and 3
respectively. These genotypes were employed for further experiment.
Regeneration and genetic transformation of red clover
67
2.3.2. Optimization of plant growth conditions in vitro
To determine optimum growth conditions for the original plants in vitro, an experiment was
conducted by growing red clover plants on different media, gelling agents and environmental
conditions. After 2 months, all plants were evaluated based on their appearance. All 24 red
clover plants that were placed on B5 medium, regardless of gelling agent or growing
conditions, were stunted and their appearance was unchanged as compared to before the test.
In contrast, all 24 plants that were transferred onto L2 medium managed to recover normal
growth; the yellow leaves were replaced with new green leaves and the rooting system
developed well (Figure 2.2). The result convincingly showed that neither gelling agent nor
suboptimal growth conditions cause stunting in red clover growth. L2 basal salt mixture was
consistently superior to that of B5; the mineral needs of the red clover variety NEWRC cannot
be supported by B5 for long-term growth during in vitro culture. Also the gelling agent,
gelrite, which has never been reported for solidification of media for red clover, was found to
be suitable in this experiment.
Figure 2.2: Appearance of the red clover variety NEWRC65 in vitro. (A)
B5 medium, (B) L2 medium.
Chapter 2
68
2.3.3. Optimization of tissue culture media for regeneration
Petiole sections from 5 different genotypes were placed on callus induction for 30 days. No
significant difference was seen in callus fresh weight (Table 2.5).
Table 2.5: Genotypes and media sequences: analysis of variance for fresh callus weight, number of
embryos per explant in 60 days, and number of developed shoots from embryos per explant in 90
days.
** significant at the 0.01 level.
ns: non-significant.
M.S.: mean squares. 1 Data underwent square root transformation.
According to the Duncan test, even those genotypes that showed variation in regeneration
capacity were found to be the same in callus proliferation, and all genotypes were classified
into one group (Table 2.6).
Table 2.6: Genotypes: mean values for fresh callus weight, number of embryos per explant in 60 days,
and number of developed shoots per explant in 90 days.
1 Data underwent square root transformation.
2 Figures in parentheses are non-square root-transformed mean values.
Values within a column followed by different letters are significantly different at the 0.05 level using
the Duncan test.
Values within a column followed by the same letters are not significantly different using the Duncan
test.
Among callus induction media, all genotypes had the best callus production on B5C and KB2;
in contrast, explants had lowest callus production on L2C (Table 2.7). Higher callus
proliferation in B5C and KB2 resulted from using 2, 4-D in combination with the second
auxin (NAA) compared to L2C, which contained BAP and PIC. There was no significant
Source of variance M.S. of fresh callus
weight
M.S. of number of embryos
per explant in 60 days1
M.S. of number of developed
shoots per explant in 90 days1
Genotypes 0.0004 ns 11.81 ** 106.10 **
Media sequences 0.004 ** 7.16 ** 21.46 ** Genotypes×Media sequences 0.0005 ns 0.20 ns 0.21 ns
Genotype number Fresh callus weight Number of embryos
per explant in 60 days1
Number of developing embryos
per explant in 90 days1
NEWRC7 0.194a 6.987 (49.152)
2 b 9.136 (83.756) d
NEWRC8 0.191a 7.0133 (49.476)
b 9.219 (85.289) cd
NEWRC19 0.197a 6.945 (48.518)
b 9.319 (87.122) c
NEWRC65 0.195a 7.704 (59.614)
a 11.370 ( 129.524) a
NEWRC68 0.196a 7.637 (58.57)
a 11.237 ( 126.463) b
Regeneration and genetic transformation of red clover
69
interaction between genotypes and culture protocols for callus weight (Table 2.5). The result
indicated that the response of genotypes was similar on the three culture protocols. Across all
genotypes, calluses in all media were evaluated visually and all calluses were white and
compact (data not shown).
Table 2.7: Media sequences: mean values for fresh callus weight, number of embryos per explant in
60 days and number of developed shoots per explant in 90 days.
Media sequences Fresh callus weight Number of embryos per
explant in 60 days1
Number of developed shoots
per explant in 90 days1
B5C-B5E- B5ED 0.197a 7.000 ( 49.352)
2 b 9.685 (95.000)
c
L2C-SEL-SPL 0.188b 7.365 ( 54.637)
a 10.033 ( 101.970)
b
KB2-SPL-SPL 0.199a 7.417 (55.377)
a 10.472 ( 110.755)
a
1 Data underwent square root transformation.
2 Figures in parentheses are non-square-root-transformed mean values.
Values within a column followed by different letters are significantly different at the 0.05 level using
the Duncan test.
Values within a column followed by the same letter are not significantly different using the Duncan
test.
After 60 days of incubation of callus of various genotypes on different media sequences,
significant differences were found in number of embryos per explant among media sequences
and genotypes. No significant interactions occurred between genotypes and media sequences
in number of embryos per explant (Table 2.5). Compared to other genotypes, NEWRC65 and
NEWRC68 exhibited the highest number of embryos per explant, whereas NEWRC 7 showed
the lowest number (Table 2.6). Although the highest number of embryos per explant was
observed in culture media sequence 3 (KB2-SPL), it did not differ significantly from culture
media sequence 2 (L2C-SPE). Differentiated calluses from all genotypes that had followed
media sequences 2 and 3 were white and friable, while some brown spots observed on callus
which had been placed on the media sequence number 1 (B5C-B5E) (data not shown).
After 90 days of culture, the number of developed shoots per explant were counted and then
subjected to analysis of variance. A highly significant difference was seen among genotypes
and media sequences. No significant interactions were detected between genotypes and media
Chapter 2
70
sequences (Table 2.5). The highest and lowest numbers of developed shoots per explant were
obtained from media sequences number 3 (KB2-SPL-SPL) and number 1 (B5C-B5E-B5S),
respectively. Genotype NEWRC65 had the highest number of developed shoots per explant,
while NEWRC7 had the lowest number of developed shoots per explant (Table 2.6). Shoots
derived in media sequence numbers 2 (L2C-SPE-SPL) and 3 (KB2-SPL-SPL) showed a
normal appearance, whereas shoots in media sequence number 1 showed mineral deficiency
and the necrotic spots, which had already appeared at the end of the embryo induction stage
(B5E), turned from brown to black.
2.3.4. Determination of kanamycin concentration as a selection agent
To find the suitable concentration of kanamycin for selection of transformed explants, the
petioles of young leaves were cultured on media sequence number 3. Numbers of shoots were
counted at the end of the shoot induction stage (Figure 2.3).
Figure 2.3: Kanamycin sensitivity test. Petiole explants in regeneration medium with
(A) No kanamycin; (B) 50 mg/l kanamycin; and (C) 75 mg/l kanamycin.
As illustrated in Table 2.8, calli were completely degenerated when the kanamycin
concentration was 75 and 100 mg l-1
. To avoid negative effects of kanamycin on growth, the
concentration of 75 mg l-1
was chosen.
Regeneration and genetic transformation of red clover
71
Table 2.8: Effect of different concentrations of kanamycin on regeneration in 90 day- cultures.
Kanamycin (mg/L) Observation at end of shoot induction stage
0 Excellent callus formation and numerous shoots developed
25 Callus formation and several shoots developed
50 Limited callus formation and few embryos appeared, none of which developed into shoots
75 No or little callus formation without any embryos
100 No or little callus formation without any embryos
2.3.5. Effect of Agrobacterium strains
C58C1 RifR (pMP90) and EHA101 stains were compared in search of a highly infective
Agrobacterium. The arcsine-transformed proportions of petiole explants with developed
calluses (putative transformants) were analyzed using ANOVA. We observed a significant
difference between Agrobacterium strains in the number of petioles with developed calli
(Table 2.9).
Table 2.9: Agrobacterium strains and pre-culture of explants: analysis of variance of callus
development.
Source of variance M.S. of the number of
petioles with developed callus1
Agrobacterium strains 0.160**
Pre-culture of explants 0.008 ns
Agrobacterium strains × pre-treatment of explants 0.006 ns
** significant at the 0.01 level.
ns: non-significant.
M.S.: mean squares. 1 Data underwent arcsine transformation.
The Agrobacterium C58C1 RifR (pMP90) strain had a higher mean percentage of callus
formation (83%), whereas EHA101 showed a lower mean percentage of callus formation
(73%). The Duncan test clearly separated C58C1 RifR (pMP90) and EHA101 into two groups
(Table 2.10).
Table 2.10: Mean values for Agrobacterium strains
Agrobacterium strains The number of petioles with developed callus1
EHA101 1.019 (72%)2 b
C58C1 RifR (pMP90) 1.153 (83%)
a
1 Data underwent arcsine transformation.
2 Figures in parentheses are non-arcsine-transformed mean values.
Values within a column followed by different letters are significantly different at the 0.05 level using
the Duncan test.
Chapter 2
72
2.3.6. Effect of callus induction period prior to Agrobacterium infection
In red clover, the petiole explants incubated on L2C for 2 days (pre-cultured explants) before
Agrobacterium co-cultivation were compared with non-incubated petioles (freshly prepared
explants). The result indicated a slight increase in transformation efficiency upon pre-
incubation, but the difference was not significant (Table 2.11).
Table 2.11: Mean values for pre-treatment of explants
1 Data were transferred to arcsine transformation.
2 Figures in parentheses are non- arcsine -transformed mean values.
Values within column followed by the same letters are not significantly
different using the Duncan test.
2.3.7. Verification of transformation by PCR analyses
T-DNA integration in the putative transformed plants was investigated using PCR. The
genomic DNA was extracted from kanamycin resistant and untransformed (control) plants
and used as template for PCR. The amplified PCR products were separated by agarose gel
electrophoresis. PCR gave the predicted band of 481 bp for the nptII gene in plants
transformed by the C58C1 RifR (pMP90) strain (Figure 2.4) and the EHA101 strain (Figure
2.5).
Figure 2.4: PCR assay of nptII gene in transgenic red clover plants for strain C58C1. Lane L: 1 kb
DNA ladder; 1-8: Putative transgenic plants; -: An untransformed control plant; +: Positive control.
L L
L 1 2 3 4 5 6 7 8L 1 2 3 4 5 6 7 8
L 1 2 3 4 5 6 7 8L 1 2 3 4 5 6 7 8
Pre-treatment of explants The number of petioles with developed calli1
Pre-cultured petioles 1.10 (78%)2 a
Freshly prepared petioles 1.07 (76%) a
Regeneration and genetic transformation of red clover
73
A band of 554 bp was seen from 35S promoter in red clover transformed by both strains (data
not shown). No amplification of either gene was detected from untransformed (control) plants
(Figures 2.4 and 2.5).
Figure 2.5: PCR assay of nptII gene in transgenic red clover plants for strain EHA101. Lane L: 1 kb
DNA ladder; 1-8: Transgenic plants; -: An untransformed control plant; +: Positive control.
2.3.8. Assay of GUS Activity
β-glucuronidase activity was used to verify transformation and gene expression in kanamycin-
resistant plantlets which were positive in PCR analysis. The tissues of kanamycin-resistant
plantlets stained positively for GUS in the histochemical assay and indicated that these tissues
were transformed (Figure 2.6). None of the untransformed control plants stained positive with
the X-gluc substrate. The presence of an intron in the GUS gene prevents GUS expression in
bacteria, thus false GUS activity was not possible; the PCR analysis results and GUS activity
confirmed each other.
Figure 2.6 : Histochemical GUS assay for putative transformed (left) and wild type (right) red clovers.
Chapter 2
74
2.4. Discussion
Because greenhouse-collected explants have caused microbial contamination of in vitro
cultures, microorganisms must be removed from explants using chemical agents. But these
disinfectants can cause damage to the explant tissue and negatively impact subsequent
regeneration and transformation steps. To avoid toxic effects of disinfectants to explants, we
eliminated the sterilization stage by keeping the original plants (120 genotypes) under in vitro
conditions. During the genotype screening test, the original plants which had already been
maintained for 4 months in vitro looked unhealthy. Growth had slowed dramatically, the
leaves had begun to turn yellow, and few, short roots were formed. Three reasons for this
aberrant growth phenotype could be postulated: mineral deficiency in the medium, unsuitable
gelling agent, or suboptimal growth conditions. Previous studies showed that B5 medium is
suitable for NEWRC germplasm (Quesenberry and Smith, 1993; Sullivan and Hatfield, 2006).
However, in all previous studies, the red clover plants used as resource for explants were kept
in a greenhouse. We hypothesized that B5 might not be an appropriate basal salt medium for
long-term maintenance of red clover in vitro. The second possible reason for growth
inhibition could be the gelling agent. It has been demonstrated that the type of gelling agent
can influence growth (Foroughi-Wehr et al., 1993) and the availability of mineral salts
(Nadolska-Orczyk and Orczyk, 2000). Since the gelling agent phytagar (Gibco BRL), which
was used in previous studies, is no longer available, gelrite was used in this study. Some
reports have demonstrated negative effects of gelrite on root growth in plants. For example,
gelrite caused poor rooting of the legume mung bean (Arthur et al., 2004) and reduced root
length of cultured tobacco shoots (Ozel et al., 2008). Similar symptoms were observed in the
growth of red clover regenerants on gelrite (this study). We hypothesized that gelrite might
cause poor growth in roots, leading to deficient mineral absorbance and consequently
unhealthy leaves. Finally, suboptimal growing conditions could also cause growth retardation.
Regeneration and genetic transformation of red clover
75
Ideal in vitro conditions for red clover are 24 °C, with a 16-h photoperiod under cool white
fluorescent light at intensity of 160-250 μmol/m2/s illumination (Maclean and Nowak, 1989).
Lower temperature and light intensity could be a third possible reason for growth inhibition of
red clover. However, our results indicated that neither gelling agent nor suboptimal growth
conditions caused aberrant growth. B5 medium was not suitable for long-term maintenance of
red clover plants. L2 medium was demonstrated to be appropriate for this purpose. This
confirms previous studies which have demonstrated the superiority of L2 to other medium
including basal salts of B5 for red clover (Carrillo et al., 2004; Phillips and Collins, 1979).
Several genotypes showed regeneration capacities and were therefore selected for future use.
Interactions between regenerative genotypes and media have been reported in the highly
regenerative alfalfa (Medicago sativa L.) germplasm Regen-S (Kris and Bingham, 1988).
Also in other legumes such as soybean, similar interactions between genotypes and media
have been reported (Tiwari et al., 2004). Therefore, to account for the possibility of
interaction between medium and regenerable genotypes of red clover achieved from the
genotype screening test, we hypothesized that regenerable genotypes might have different
performance in different media sequences. We decided to test not only highly regenerative
genotypes (NEWRC65 and NEWRC68), but also the genotypes with moderate regeneration
ability (NEWRC7, NEWRC8 and NEWRC19) in optimizing tissue culture media for
regenerative genotypes.
Three media sequences were used in this experiment. Media sequence number 1 (B5C-B5E-
B5S) was employed by Beach and Smith (1979) for regeneration of red clovers using petioles
as source of explants. The basal salt mixture B5, which was used for media sequence number
1, had been shown in our preliminary growth condition test to be inappropriate as growth
medium for long-term growth of red clover. However it is included as the standard media
system for comparison of callus formation and plant regeneration response with other media.
Chapter 2
76
Media sequence number 2 (L2C-SPE-SPL) was developed by Collins and Phillips, 1982. Its
basal salt mixture L2 was established by Phillips and Collins (1979). They found L2 to be
superior for the culture of certain legumes, which is consistent with the result of our long-term
growth test reported above. To our knowledge, there is no report of a regeneration test with
media sequence number 2 using petioles as source of explants. The media sequence number 3
(KB2- SPL-SPL) proposed in this study was based on an earlier study on red clover
regeneration which indicated efficient regeneration achieved with the media sequence (B5C-
SPL-SPL), using petioles of the red clover variety ‘Florex’, which has low capacity for
regeneration (Maclean and Nowak, 1989). The sequence media number 3 and the sequence
media described by Maclean and Nowak (1989) differ in that Maclean and Nowak used a
different callus induction medium. The callus induction medium KB2 which was developed
by Keyes et al. (1980) contain L2 basal salt whereas the callus induction medium B5C which
described by Beach and Smith (1979) consists of B5 basal salts. KB2 and B5C have the same
growth regulators, however. We decided to substitute KB2 for B5C to avoid the basal salt B5
due to the results of the preliminary experiment which indicated that it is not an appropriate
medium for red clover. To our knowledge, KB2 was used once for callus proliferation with
hypocotyls as explants from the red clover Arlington variety (Keyes et al., 1980). But it has
never been reported for callus induction using the NEWRC germplasm, nor has it been
compared with other callus induction media for red clover.
The highest callus proliferation was achieved with those callus induction media that used 2, 4-
D in combination with the second auxin NAA. This result is in accordance with other reports
that showed high callus production in the presence of 2, 4-D in combination with NAA
(Beach and Smith, 1979).
Although it is a well-known general opinion that the more callus mass is produced, the more
regenerated plants will develop, the results of this study suggested that callus mass may not be
Regeneration and genetic transformation of red clover
77
significantly associated with regeneration response among these genotypes. It should also be
taken into account that all genotypes used here were from the same germplasm and have no
variation in callus fresh weight (Table 2.6). A similar result reported by Myers et al. (1989)
showed that regeneration capacity is not related to cell growth or protoplast division
frequency in red clover.
Our result showed that the critical stage for regeneration is during culture on embryo
induction medium. MacLean and Nowak (1989) reported similar results. They reported that
somatic embryogenesis occurred when calluses were placed on embryo induction medium,
and that the embryonic induction stage is vital for regeneration (Maclean and Nowak, 1989).
In this study, some brown spots were observed on cultures incubated on B5E. The source of
auxin in B5E was NAA, which could be the cause of the dark spots on the calli. Because
NAA is incorporated in the basal medium, it could lead to the dark colour on the red clover
calli, perhaps by stimulated production of phenolics or related compounds (Phillips and
Collins, 1979). In the shoot development stage, the color of the necrotic spot changed to
black. Since the source of auxin in B5S also is NAA, the change in color to black might also
be caused by the presence of NAA in culture. Regeneration never happened in these dark
spots, which explains why cultures on media sequence number 1 had the lowest number of
developing embryos. No significant interaction was observed between media sequences and
genotypes for shoot development. This result was in agreement with the previous report that
interaction of media and number of shoots in several genotypes did not have any significant
differences (Carrillo et al., 2004).
The various levels of regeneration of the genotypes among the different media indicated that
regeneration was partially dependent upon the media sequence. These results lead to the
conclusion that the appropriate media sequence could increase the efficiency of regeneration.
In other words, the suitable media sequence could maximize the potentiality of a regenerable
Chapter 2
78
genotype. According to the result of optimization of tissue culture media, it indicated that the
media sequence number 3, proposed for the first time in this study, was the best media
sequence for regeneration of NEWRC red clover germplasm. Using media sequence number 3
with a highly embryogenic genotype such as NEWRC65 could provide a platform to develop
a robust red clover regeneration system and be a potential base for future genetic
transformation.
Prior to gene transformation, an experiment for determination of selection agent concentration
was performed; the result indicated that a concentration of 75 mg l-1
of kanamycin was
suitable. Our result was contrary to that of a previous report in which red clover petiole
explants were put onto B5C (solidified with agar) containing 0, 12.5, 25, 50 and 100 mg l-1
of
kanamycin. Their result indicated that growth reduction happened at 12.5 mg l-1
of
kanamycin, and total growth was inhibited at 25 mg l-1
of kanamycin. The fact that we used
gelrite, which reduces efficiency of kanamycin, can explain the discrepancy between our data
and those of the previous report (Quesenberry et al., 1996).
Agrobacterium EHA101 and C58C1 RifR (pMP90) strains, which both harbored the pTJK136
binary vector, were chosen to determine the Agrobacterium strain with higher efficiency for
red clover transformation. EHA101 has been used for several legumes, such as fava bean
(Vicia faba L.) (Bottinger et al., 2001), soybean (Glycine max Merr.) (Olhoft et al., 2003), and
narbon beans (Vicia narbonensis L.) (Czihal et al., 1999). In red clover, EHA101 was found
to be superior to the strain A208 (Quesenberry et al., 1996). There are several reports of
successful use of the C58C1 strain in legumes such as tepary bean (Phaseolus acutifolius A.
Gray) (De Clercq et al., 2002), pea (Pisum sativum L.) (Nadolska-Orczyk and Orczyk, 2000),
and peanut (Arachis hypogaea L.) (Sharma and Anjaiah, 2000). There are no reports on use of
C58C1 RifR (pMP90) in red clover. Our result showed the superiority of C58C1 Rif
R
(pMP90) over EHA101 for transformation of red clover. In chickpea (Cicer arietinum L.), the
Regeneration and genetic transformation of red clover
79
efficiency of C58C1 RifR (pMP90) and EHA101 strains was investigated; results were poor
and no significant difference was found between them, however, over the cultivars tested, the
C58C1 RifR (pMP90) had a higher mean percentage of gene transformation (Krishnamurthy
et al., 2000). Compared with a previous report of using EHA101 in red clover, in which the
maximum mean percentage of gene transformation was 63% (Quesenberry et al., 1996), our
transformation efficiency was considerably greater (Table 2.10). It demonstrated that using a
superior media sequence plus Agrobacterium C58C1 RifR (pMP90) strain in this study caused
higher transformation and regeneration efficiency compared with the previous gene
transformation protocol in red clover.
Pre-culture of explants in callus induction medium prior to inoculation has been shown to
enhance efficiency of gene transformation in plant species (Birch, 1997). For instance, in river
red gum (Eucalyptus camaldulensis Dehnh.), the maximum average putative transformation
rate was achieved by a pre-culturing for three days followed by two days of co-cultivation
(Ho et al., 1998). Our result indicated that two days of pre-culturing of petioles slightly
increased efficiency of transformation. The result was similar to a study by Agarwal et al.,
2004. They reported that the transformation frequency in white mulberry (Morus alba L.)
increased after 2 days of pre-culturing .
An improved protocol for Agrobacterium-mediated transformation of red clover was
developed by modifications including using a new media sequence (KBC-SPL-SPL) for
regeneration and employing a superior Agrobactrium strain (C58C1 RifR (pMP90)). This
exhibited significantly higher efficiency of gene transformation compared to the previously-
described protocol. It could facilitate introduction of useful foreign genes such as disease
resistance and other desirable traits to develop new red clover varieties.
Chapter 3: Identification of reliable
reference genes in red clover
This chapter is adapted from:
Mehdi Khanlou K, Van Bockstaele E (2012). A critique of widely used normalization
software tools and an alternative method to identify reliable reference genes in red clover
(Trifolium pratense L.). PLANTA. 236(5). P. 1381-1393.
Identification of reliable reference genes in red clover
83
3.1. Introduction
In the previous chapter, we described the optimization of regeneration and transformation of
red clover. Before any bioengineering manipulation of the metabolic pathway of the
secondary metabolites in red clover can be implemented, a tool is needed which enables
careful monitoring of changes in the expression level of the manipulated genes. Quantitative
reverse transcription real-time polymerase chain reaction (qRT-PCR) is the most robust tool
available for detection and quantification of expression of studied genes. In comparison with
conventional methods of gene expression analysis such as Northern blot, ribonuclease
protection assay and reverse transcriptase PCR assay, qRT-PCR has the advantages of high
sensitivity, specificity, dynamic range and throughput capacity (Vandesompele et al., 2002).
However, like conventional methods, the measured gene expression variation in qRT-PCR
analysis consists of true biological variation generated by the phenotype or under-investigated
phenomena (Vandesompele et al., 2009) as well as non-biological variations caused by
differences in the quantity and integrity of the RNA template, RNA recovery, and efficiency
of cDNA synthesis (Andersen et al., 2004; Exposito-Rodriguez et al., 2008). These
experimental errors (non-biological variations) should be eliminated as much as possible to
quantify gene expression at a trustworthy level. Numerous computational methods (so-called
normalization methods) have been introduced to remove or minimize experimentally caused
errors, leaving only the true variants. Among the suggested normalization methods, the use of
the internal control gene is most acknowledged. A prerequisite for using qRT-PCR is
determining a suitable internal control gene for red cover.
During the pre-genomic era, the genes that have housekeeping roles in basal cell activities
were chosen as internal control for expression profile studies. Later on, however, further
investigations proved that the expression level of some of the best-known and most
commonly used control genes, which were supposed to be uniformly expressed, were actually
Chapter 3
84
found to be unstable in expression level across different treatments, biological processes and
even across different tissues (Hu et al., 2009; Radonic et al., 2004; Schmittgen and Zakrajsek,
2000). In other words, having a housekeeping function does not always mean that the gene
has stable expression level in the cell. For normalization purposes, the use of an unstably
expressed housekeeping gene can lead to severe misinterpretation of the results (Remans et
al., 2008). This has convinced researchers to carefully validate the stability of the expression
level of housekeeping genes prior to using them for normalization in qRT-PCR data analysis.
The identification of experimentally validated control genes (so-called reference genes) that
guarantee the accurate quantification of expression level of the target gene(s) (Lovdal and
Lillo, 2009) has been a challenge. Several statistical algorithms including geNorm
(Vandesompele et al., 2002), BestKeeper (Pfaffl et al., 2004) and NormFinder (Andersen et
al., 2004) have been developed to address this conundrum. geNorm is one of the most
frequently used algorithms. It is based on the criterion that two ideal reference genes (RGs)
should have a minimal expression ratio across the investigated sample set, regardless of cell
type or condition. The geNorm software defines the gene expression stability value (M),
which is calculated as the mean standard deviation of the logarithmic transformed expression
ratios across samples for the particular gene relative to other genes under investigation. Then
geNorm eliminates the gene with the highest M value (i.e., the least stable gene) from the
panel and recalculates new M values for remaining genes until reaching the last two genes
with the smallest M value (i.e., the most stable genes). The BestKeeper program is also an
Excel-based tool to select best-suited RGs by performing a pair-wise correlation analysis of
candidate reference genes (CRGs) (Pfaffl et al., 2004). First, descriptive statistics of the raw
cycle threshold (Ct) values including standard deviation (SD) and coefficient of variance (CV
%) are computed for all the RGs. In the next step, Pearson’s coefficients of correlation are
calculated for each CRG pair. The highly correlated CRGs, which have a low standard
Identification of reliable reference genes in red clover
85
deviation (SD < 1), are then combined into index value (i.e., normalization factor) using the
geometric mean of their Ct values. Finally, BestKeeper determines the correlation coefficient
of each CRG with the index value (called the “BestKeeper index”), along with the probability
(p) value, which indicates the significance of the obtained correlation coefficient. In this way,
it can present the contribution of each CRG over BestKeeper index. The CRG that is best
correlated to the BestKeeper index is considered to be the most stable. The NormFinder
program is another algorithm that allows ranking CRGs according to their expression
stability. NormFinder calculates the stability index using analysis of variance on log-
transformed expression values. Using NormFinder, the proportion of both components of
overall variance (intra- and intergroup variation) can be taken into account to calculate the
normalization factor (NF). To do so, the NormFinder package considers intra- and intergroup
variations in determination of the expression stability in a given set of CRGs. However, the
accuracy of these algorithm outputs rests on the assumption that none of the tested CRGs
shows any systematic variation in expression. The existence of CRGs with systematic
variations among investigated CRGs could lead to false selection of RGs, which would result
in erroneous normalization. We believe that it is unlikely that all investigated CRGs would
have unbiased expression, because experimental conditions could influence CRG expressions.
The stability of expression of individual CRGs cannot be presumed prior to being subjected to
the condition under investigation. To overcome this problem, it is necessary to use a statistical
method which could evaluate each CRG independent of the assumption of whether the
expressions of studied CRGs have systematic variation or not. For this purpose, we have
created a new stability index based on the analysis of variance (ANOVA) model.
The aims of the present study were (1) to discuss the potential pitfalls arising from uncritical
use of the abovementioned statistical algorithms when the existence of variation in the
expression of CRGs violates the underlying assumption, (2) to propose a novel stability index
Chapter 3
86
for ranking CRGs, and (3) to systematically validate a panel of putative RGs and identify the
useful genes for data normalization in future gene expression studies among different tissues
in red clover.
3.2. Materials and Methods
3.2.1. Plant materials
Regeneration ability is a prerequisite for any genetic manipulation study. The capacity for
plant regeneration in red clover is under genetic control, and not all red clover cultivars have
capability of regeneration. We therefore used NEWRC red clover germplasm, which is a
highly regenerable population. The seeds were sown in pots containing a mixture of sand,
expanded clay and soil (1:1:1) and were kept under greenhouse conditions at 25 °C with 16 h
per day illumination of 400 μmol m-2
sec-1
. The leaves, stems and roots of 10 plants were
separately sampled at the late vegetative stage and then immediately frozen at -80 °C.
3.2.2. Total RNA isolation and cDNA preparation
Frozen tissue samples were ground with a Retsch Tissuelyser (Qiagen). Total RNA was
isolated using the RNeasy Plant Mini Kit (Qiagen) using the manufacturer's instructions.
Residual DNA was removed with TURBO™ DNase (Ambion) following the manufacturer’s
instructions. The quantity and quality of total RNA was determined using a NanoDrop ND-
1000 spectrophotometer (NanoDrop Technologies). Hereafter, equal amounts of the different
tissue samples of total RNA (50 ng) were reverse transcribed into cDNA in a total reaction
volume of 20 µl using the Superscript VILO cDNA synthesis kit (Invitrogen) according to the
manufacturer’s protocol and then the cDNA was diluted 1:10 for quantitative real-time PCR.
3.2.3. Primer design and verification of amplified products
Sequences of RGs of Arabidopsis were BLASTed against the red clover EST database from
the National Center for Biotechnology Information (NCBI) to identify putative orthologues.
Identification of reliable reference genes in red clover
87
Primers were designed according to the red clover EST sequences which had presented high
similarity to the Arabidopsis RGs using Vector NTI Advance™ version 11 (Invitrogen). The
GC content of the primers ranges from 45% to 55% and the melting temperatures (Tm value)
are between 59 °C and 60 °C. Hairpin structures and primer-dimer formation of all the
primers were evaluated using Vector NTI Advance™ version 11. All primer pairs were
custom-ordered from a commercial supplier (Invitrogen, Merelbeke, Belgium). High Fidelity
PCR was performed in 50-µl volumes that contained 100 ng cDNA, 1× High Fidelity PCR
buffer, 0.2 mM dNTP mix, 0.25 µM each primer, 2 mM MgSO4 , and 1 U Platinum Taq High
Fidelity (Invitrogen). Amplification was carried out in a GeneAmp 9600 thermocycler
(Applied Biosystems) with the programme 94 °C for 2 min, 30 cycles of 94 °C for 30 sec, 60
°C for 30 sec, 68 °C for 2 min, with a final extension at 68 °C for 10 min. Running an aliquot
of the PCR products on a 1.5% agarose gel revealed the presence of single products of
appropriate size, which confirmed the specificity of all primers (figure not shown). To verify
the sequences of amplification products, the remaining PCR products were cloned using the
TOPO TA Cloning Kit for sequencing (Invitrogen) according to the manufacturer’s
instructions. Plasmids isolated from the possible positive colonies were purified using a
plasmid mini kit (Qiagen). The purified plasmids were directly sequenced using an ABI Prism
BigDye Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems) with the M13 Forward
and Reverse universal primers following the manufacturer’s instructions. The sequencing
PCR was conducted as follows: 1 min at 94 °C followed by 25 cycles of 94 °C for 10 s, 58 °C
for 5 s, 60 °C for 1 min with a final extension at 60 °C for 4 min in the above-mentioned
thermocycler. The sequencing reactions were subjected to ethanol precipitation to remove
excess dye-terminators. The sequencing products were dissolved in formamide and then heat-
denatured for 8 min before loading onto a capillary ABI3130 Genetic analyzer system. Bases
were determined, and sequences assembled using Sequencing Analysis software v.5.2
Chapter 3
88
(Applied Biosystems). The nucleotide sequences obtained were aligned against the reference
sequences of red clover EST database using Vector NTI Advance™ version 11.
3.2.4. Determination of primer efficiency
The TOPO-cloned amplicons with identical sequences linearized with the restriction enzyme
EcoRV. The digested plasmids were separated on 1% agarose gel, isolated and then purified
by Qiagen Gel extraction kit in accordance with the manufacturer’s instructions. The primer
efficiencies (E) of each primer pair were calculated from 10-fold dilution series of linearized
plasmid containing the respective sequence-verified amplicons over 5 dilution points that
were measured in duplicates. The dilution series was used to generate a standard curve by
plotting the quantification cycle for each dilution point against log-transformed of dilution
series of input template (linearized plasmid). The slope produced by the standard curve used
to calculate the primer efficiency was according to the formula E = 10(-1/slope)
, as described by
Pfaffl (2001).
3.2.5. qRT-PCR conditions
Quantitative real-time PCRs (qRT-PCR) were carried out in 384-well plates using the
LightCycler 480 (Roche). The reactions were performed in duplicate and consisted of 5 μL
LightCycler 480 SYBR Green I Master, 1 μl deionized water, 1 μl of each of the 10 μM
forward and reverse gene-specific primers and 2 μl 10-fold diluted cDNA as template in a
final volume of 10 μl. To ensure absence of DNA contamination in the reaction mixture, no-
template control (NTC) was included in each run. Cycling parameters were 95 °C for 10
minutes to denature the DNA templates and activate polymerase enzyme, followed by 45
cycles of 95 °C for 10 s, 60 °C for 12 s and a final extension of 72 °C for 10 s. Detection of
fluorescence was performed at the end of the annealing period of each cycle. Immediately
after the amplification, the specificity of each amplicon was verified via generation of a
melting curve by heating to 95 °C for 5 s, cooling to 65 °C for 1 m and slowly heating to 97
Identification of reliable reference genes in red clover
89
°C with a continuous fluorescence data collection of 10 acquisitions per °C. Also, qRT-PCR
assays were performed on equivalent amounts of total RNA without reverse transcription (as
template) for each sample in accordance to the abovementioned reaction mixture and the
thermal profile to verify absence of genomic DNA contamination in the extracted total RNA.
3.2.6. Data analysis
Expression levels of the eight housekeeping genes were determined by the number of cycles
required for the amplification-related fluorescence to reach a particular threshold level of
detection. The raw Ct values were exported into qBASE software version 1.3.5 (Hellemans et
al., 2007), corrected by PCR efficiency, then transformed into relative expression. The
relative expressions were then imported into geNorm version 3.5 (Vandesompele et al., 2002)
and NormFinder version 0.953 (Andersen et al., 2004) as described in their manual. Unlike
geNorm and NormFinder, raw Ct values were imported directly into BestKeeper version 1
(Pfaffl et al., 2004) as inputs. To examine the stability of each CRG independently, the
relative expression ratios (R) were calculated for each CRG using the formula R = E – Ct sample
/
E − Ct calibrator
where E is the efficiency of the gene amplification of corresponding CRG and the
calibrator is the sample with the highest expression (or lowest Ct value) within samples of the
corresponding CRG. The relative expression ratios were natural log transformed and then
subjected to a one-way analysis of variance (ANOVA). The mean square of groups (MSB)
which represent intergroup variance caused by difference among tissue samples and mean
square errors (MSW), which represents intragroup variance, were computed. Then the MSB
and MSW of each CRG were divided by (-1/ ̅) to calculate intergroup variation index (VB)
and intragroup variation index (VW) of each gene, respectively, where ̅ is average of natural
logarithmic transformed relative expression ratios for the corresponding CRG. A stability
index was subsequently generated by multiplying VB by VW for each gene. The lower the
stability value, the more stable the gene is across the investigated tissues. The statistical
Chapter 3
90
analysis including correlation analysis and ANOVA were performed using SAS 9.2 (SAS
Institute Inc., Cary, NC, USA).
3.3. Results
3.3.1. Identification of red clover candidate reference genes
In the present study, eight frequently-used CRGs (i.e., actin (ACT), glyceraldehyde-3-
phosphate-dehydrogenase (GADPH), elongation factor-1alpha (EF-1α), translation initiation
factor (EIF-4a), ubiquitin-conjugating enzyme E2 (UBC2), polyubiquitin (UBQ10), sand
family protein (SAND), and yellow-leaf-specific protein 8 (YLS8) were chosen to identify the
most constitutively expressed RGs. Except for ACT (Kim et al., 2003; Sullivan, 2009a) and
EIF-4a (Sullivan, 2009b), which have already been reported as being used to normalize
relative expression of genes of interest in red clover, BLAST was used to determine the
sequences of other CRGs. The genes tested were the most commonly used RGs which were
found to be stably expressed in a genome-wide study in Arabidopsis (Czechowski et al.,
2005) and a collection of ESTs obtained from three-week-old red clover plants. The genes
with a similarity higher than 1e-92 (E-value) were considered as being putative orthologous to
the Arabidopsis genes (Table 3.1).
3.3.2. PCR efficiency and amplification specificity
High-quality total RNA from 30 samples was isolated from different tissues including leaves,
stem and roots. The obtained RNA was reverse transcribed and then used to assess the
expression stability of the CRGs. To confirm specificity of transcript amplification, the PCR-
generated products were analyzed using agarose gel electrophoresis after ethidium bromide
staining. All primer pairs generated a single band of the desired size. The identities of all PCR
products were also verified by TOPO T/A cloning and subsequent sequencing (data not
shown). Furthermore, the presence of a single peak in the melting curve, without formation of
primer-dimers, also confirmed the good specificity of amplicons generated from the primer
Identification of reliable reference genes in red clover
91
sets used in qRT-PCR (figures not shown). To determine the amplification efficiency for each
primer pair, a standard curve was generated by five serial 10-fold dilutions of the cloned
amplicons in a qRT-PCR assay. The threshold cycles (Ct) obtained for duplicates of each
dilution were plotted against the log of the arbitrary numbers representing the 10-fold
dilutions difference were used, as the concentration is unknown and not relevant. The slope of
the line linking all points was then used to calculate the efficiency. The efficiency values of
selected primers ranged from 1.85 for UBC2 to 1.99 for UBQ10, which meets standard
guidelines of acceptability (Bookout et al., 2006) (Table 3.1).
3.3.3. Descriptive statistics and relations of the raw Ct values
In our study, the Ct values obtained varied for each CRG (Figure 3.1). Among the selected
genes, the gene encoding UBQ10 had the most abundant transcript level (Ct ranged between
21.57 and 25.45 across all tissues studied). In comparison, the gene encoding SAND had the
least abundant transcript level, (Ct between 29.49 and 35.09 across all tissues considered). Of
the achieved Ct values of all of the CRGs studied, none had an invariant level of gene
expression.
Table 3.1: Red clover candidate reference genes and comparison with Arabidopsis orthologues.
Symbol GenBank
accession number Primer sequences
Arabidopsis
orthologous locus
Amplicon
length (bp)
Amplification
efficiency
Similarity
(E-value)
ACT AY372368
ATGAGCTTCCTGATGGACAG
CCAGCAGCTTCCATTCCAAT
- 100 1.93 -
EF-1α BB903271 GCGTGTGATTGAGAGGTTTG
ACGTTCAGCCTTGAGCTTGT
AT5G60390 94 1.98 2e-166
EIF- 4a BB919542 GACCTGTTGGCTCGTGGTAT
TCAGGTTGGGTAGGCAAATC
- 74 1.95 -
GAPDH BB911051 CGGAATCGTTGAGGGTCTTA
TTCCACCTCTCCAGTCCTTG
AT1G13440 95 1.90 2e- 157
SAND BB902537 TTATGCAACAAGGCAAGCTG
TCTGAGCGCCAACAAGACTA
AT2G28390 113 1.86 3e-92
UBC2 BB916552 TCCAAACCCAAACTCTCCAG
CTGCTCAACAATCTCGCGTA
AT2G02760 97 1.85 2e-129
UBQ10 BB903576
ACCTTGTCTTGCGTCTTCGT
TCTTGGATCTTGGCCTTGAC
AT4G05320 121 1.99 1e-169
YLS8 BB917276
CCCTTCCTTGCACCTCTGTA
CACCGGAAACAACAACAAGA
AT5G08290 93 1.88 2e-121
Chapter 3
92
Figure 3.1: Distribution of Cycle threshold (Ct) values for the candidate reference genes in red clover.
Error bars show the range of Ct values of leaf (left), stem (center) and root (right) for each gene.
This highlighted the importance of seeking appropriate RGs via statistical approaches to
normalize the relative expression of genes of interest for red clover. To describe the degree of
relationship, the raw Ct values of investigated CRGs were also used to perform Pearson
pairwise correlation tests among genes in each tissue. A highly significant correlation (p <
0.001, r = 0.91) between YLS8 and UBQ10 in leaf tissue was observed. A highly significant
correlation (p < 0.001, r = 0.92) was found between EF-1α and ACT in stem tissue. Also,
UBC2 and YLS8 had a highly significant correlation (p < 0.001, r = 0.90) in root tissue (Table
3.2).
ACT EF-1α EIF- 4a GAPDH SAND UBC2 UBQ10 YLS8
32.5
30.0
27.5
25.0
22.5
CT
val
ues
Identification of reliable reference genes in red clover
93
Table 3.2: Pairwise correlation among candidate reference genes in leaf (A), stem (B) and root (C).
Correlation analyses were carried out based on the Ct values of the eight candidate reference genes.
* Indicates correlations below the significance threshold of p>0.05.
** Indicates correlations below the significance threshold of p>0.01.
B ACT EF-1α EIF- 4a GAPDH SAND UBC2 UBQ10 YLS8
ACT 1
EF-1α 0.92** 1
EIF- 4a 0.77** 0.74** 1
GAPDH 0.82** 0.82** 0.77** 1
SAND -0.13 -0.08 -0.48* 0.03 1
UBC2 0.66** 0.41 0.57** 0.32 -0.31 1
UBQ10 0.64** 0.49* 0.30 0.39 0.17 0.56* 1
YLS8 0.75** 0.66** 0.51* 0.46* 0.15 0.66** 0.59** 1
A ACT EF-1α EIF- 4a GAPDH SAND UBC2 UBQ10 YLS8
ACT 1
EF-1α 0.79** 1
EIF- 4a 0.78* 0.71** 1
GAPDH 0.62** 0.64** 0.49* 1
SAND 0.44 0.32 0.04 0.05 1
UBC2 0.39 0.37 0.06 0.15 0.84** 1
UBQ10 0.78** 0.87** 0.66** 0.69** 0.51* 0.66** 1
YLS8 0.82** 0.79** 0.66** 0.47* 0.70** 0.68** 0.91** 1
C ACT EF-1α EIF- 4a GAPDH SAND UBC2 UBQ10 YLS8
ACT 1
EF-1α 0.71** 1
EIF- 4a 0.54* 0.67** 1
GAPDH 0.68** 0.83** 0.50* 1
SAND 0.34 0.40 0.11 0.52* 1
UBC2 0.62** 0.82** 0.48* 0.59** 0.67** 1
UBQ10 0.40 0.69** 0.41 0.55* 0.30 0.68** 1
YLS8 0.59** 0.72** 0.32 0.56** 0.55* 0.90** 0.83** 1
Chapter 3
94
3.3.4. Determining the stability of the candidate reference genes
3.3.4.1. Analysis of gene expression stability using geNorm
The relative expression values calculated by qBASE for the eight CRGs were subjected to the
geNorm algorithm to evaluate their reliability as references (Figure 3.2).
Figure 3.2: Average expression stability values (M) of red clover candidate reference genes. M values
of the candidate reference genes were calculated using the geNorm algorithm. Ranking of the stability
was carried out on leaf (A), stem (B) and root (C). Lower M values indicate the more stably expressed
genes.
Av
era
ge
exp
ress
ion
sta
bil
ity
M
<−−− Least stable genes Most stable genes −−−>
A
B
C
Identification of reliable reference genes in red clover
95
The M values for all CRGs across tissues were less than 1; this fell below the default limit of
M=1.5 (Vandesompele et al., 2002). The UBQ10 and YLS8 genes ranked highest in the
samples gathered from leaves, with M values of 0.29 and 0.33, respectively (Figure 3.2A). In
stems, the ACT and EF-1α genes proved to be the best candidates for normalization, with M
values of 0.14 and 0.18, respectively (Figure 3.2B). We determined that UBC2 and YLS8
genes were most stably expressed in the red clover root samples, with an M value of 0.28 and
0.31, respectively (Figure 3.2C). Notably, using geNorm, SAND was found to be the least
stable among the genes examined in any of the red clover tissues.
3.3.4.2. Analysis of gene expression stability using BestKeeper
The raw Ct value generated by a real-time PCR platform was used as input file in the Excel-
based spreadsheet of BestKeeper applet to determine the stability of eight CRGs. The result of
the correlation of each eight CRG with BestKeeper Index is presented in Table 3.3.
Table 3.3: The expression stability values for red clover reference genes calculated by the BestKeeper
algorithm.
These results showed that UBQ10 (r =0.96) in leaves, ACT (r = 0.94) in stems and YLS8 (r
=0.91) in roots presented the highest coefficient of correlation with the BestKeeper index (i.e.,
the highest stability). SAND demonstrated the highest variation in all tissues, especially in
Tissue BestKeeper vs. ACT EF-1α EIF- 4a GAPDH SAND UBC2 UBQ10 YLS8
Leaf coeff. of corr. [r] 0.88 0.86 0.66 0.65 0.66 0.68 0.96 0.95
p value 0.001 0.001 0.001 0.002 0.002 0.001 0.001 0.001
Stem coeff. of corr. [r] 0.94 0.89 0.73 0.86 0.13 0.61 0.71 0.81
p value 0.001 0.001 0.001 0.001 0.603 0.005 0.001 0.001
Root coeff. of corr. [r] 0.74 0.89 0.56 0.84 0.72 0.91 0.75 0.87
p value 0.001 0.001 0.011 0.001 0.001 0.001 0.003 0.001
Chapter 3
96
leaves and roots with a standard deviation of greater than 1 (SD >1). SAND was therefore
excluded from the BestKeeper index calculation in leaf and root tissue.
3.3.4.3. Analysis of gene expression stability using NormFinder
The NormFinder algorithm was performed on the data where the sample sets are considered
as three groups; the sample set was divided into three groups based on the tissue type (leaves,
stems and roots). Plotting inter- and intragroup variations for each gene showed that ACT had
the highest intergroup variation and SAND presented the highest intragroup variation (Figure
3.3).
Figure 3.3: Plotting the inter- and intragroup variances calculated by NormFinder for candidate
reference genes in three tissues of red clover. For each gene, the intergroup variations of leaf (left),
stem (center) and root (right) is shown as histogram bars and intragroup variation as error bars.
The stability index for each gene was calculated according to corresponding inter- and
intragroup variations. The obtained result suggested that YLS8 (stability index of 0.115) and
EF-1α (stability index of 0.172) were the best combination of two RGs (Table 3.4).
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
ACT EF-1α EIF- 4a GAPDH SAND UBC2 UBQ10 YLS8
Identification of reliable reference genes in red clover
97
Table 3.4: Stability values of candidate reference genes according to their expression profiles across
tested tissues (leaf, stem and root) of red clover including as calculated by NormFinder.
3.3.4.4. Analysis of gene expression stability using ANOVA
Each investigated gene was evaluated independently for possible differences in expression
among tissues using one-way ANOVA. UBQ10 had the highest F value (or lowest p value) at
18.33, indicating a significant difference in gene expression among tissues. SAND had the
lowest F value (0.22), indicating no difference among tissues tested (Table 3.5). Also, both
intra and intergroup variances were broken down into their component parts across tissues.
The results show that UBC2 had the lowest variance (0.070 and 0.011 in leaf and stem,
respectively). EIF-4a had the lowest variance in root (0.022). SAND had the highest variances
in all three tissues. The stability indexes were calculated as described in “Materials and
Methods” below. A CRG with a low stability index is considered to be most stable across all
tissues tested. Among the CRGs under study, YLS8 (stability index of 0.028) was the most
stable; UBC2 was second most stable, (stability index of 0.07) and EIF-4a was third most
stable (stability index of 0.12) (Table 3.5).
Candidate reference gene Stability index
ACT 0.612
EF-1α 0.172
ELF- 4a 0.268
GAPDH 0.487
SAND 0.428
UBC2 0.224
UBQ10 0.385
YLS8 0.115
Chapter 3
98
Table 3.5: Determination of the stability of candidate reference genes based on analysis of variance
technique (ANOVA). Intergroup variance (MSB) and intragroup variance (MSW) are broken down into
the contributing components (tissues).
1 group variance divided by mean
2 within-group variance divided by mean
3.4. Discussion
Numerous sources of variation, including the total RNA content of the sample, the number of
cells in the starting material, the RNA extraction efficiency, differential enzymatic
efficiencies, and transcriptional activity resources, could introduce bias in evaluating the
expression profile of the investigated genes. Therefore, a strategy is required to minimize the
effect of any non-biological variation as much as possible in order to monitor the actual
intrinsic variation of desirable genes across a given sample set. For this purpose, several
strategies (so-called normalization strategies) were proposed to obtain accurate data using
qRT-PCR. It appears that the use of RGs could offer more reliability and accuracy compared
to other strategies; for this reason, this is now the most preferred way of normalization
Genes
Source of variation
MS (Variance) components Total MS
(Variance)
F
Sig.
-1/𝑿
VB 1
VW 2
Stability
index Leaves Stems Roots
ACT
Between Groups 0.066 1.201 1.826 3.093 15.00** 0.000 0.959 3.22 0.21 0.692
Within Groups 0.104 0.043 0.059 0.206
EF-1α
Between Groups 0.337 0.011 0.463 0.811 2.86 0.074 0.704 1.15 0.40 0.461
Within Groups 0.165 0.042 0.076 0.283
EIF-4a
Between Groups 0.411 0.161 0.058 0.630 3.21 0.056 1.009 0.62 0.19 0.120
Within Groups 0.102 0.072 0.022 0.196
GAPDH
Between Groups 0.916 0.477 2.711 4.104
12.70** 0.000 0.574 7.13 0.56 4.011
Within Groups 0.167 0.062 0.092 0.323
SAND Between Groups 0.079 0.106 0.001 0.188
0.22 0.799 0.629 0.29 1.32 0.394
Within Groups 0.384 0.089 0.356 0.831
UBC2 Between Groups 0.118 0.164 0.559 0.841
7.67** 0.002 1.142 0.73 0.09 0.070
Within Groups 0.070 0.011 0.029 0.110
UBQ10 Between Groups 0.766 0.361 2.183 3.310
18.33** 0.000 0.741 4.46 0.24 1.087
Within Groups 0.079 0.033 0.069 0.181
YLS8 Between Groups 0.183 0.020 0.081 0.284
1.43 0.255 1.364 0.20 0.14 0.028
Within Groups 0.113 0.019 0.066 0.198
Identification of reliable reference genes in red clover
99
(Vandesompele et al., 2009). However, internal RGs should have invariant levels of gene
expression across a sample set. The use of unstably expressed RGs expressed constitutively
could increase errors introduced by the technical variations, and could also add more bias to
the result. Therefore, prior to using internal control genes for normalization, their expression
stability should be statistically validated across all of the treatments/categories being studied.
Several statistical algorithms such as geNorm, BestKeeper and NormFinder have been
developed to identify consistently expressed genes. Each of them has pitfalls; the strength of
each algorithm lies in the circumstances of its use. These algorithms also do not
independently assess CRGs for their expression stability. Therefore, we believe that in
addition to commonly-used algorithms such as geNorm, NormFinder and BestKeeper,
statistical methods which can evaluate CRGs independently are needed to provide credence to
the selection of the "best" suited CRGs for superior transcript normalization of gene
expression.
geNorm, developed by Vandesompele et al. (2002), is the most commonly used normalization
algorithm. geNorm has two main advantages: 1) the raw data are used as an input file and
does not need to be normality distributed, and 2) geNorm does not need a large sample size to
provide reliable results (Serrano et al., 2011). The analysis relies on the principle that the
expression ratio of two ‘ideal’ RGs should be identical in all samples, regardless of the
experimental condition or cell type (Vandesompele et al., 2002). geNorm top-ranks those
genes with the highest degree of similarity in their expression profile, rather than with
minimal variation (Andersen et al., 2004), because geNorm investigates expression ratio
candidate genes and does not take the variation across sample sets into account. It is therefore
crucial that the two ‘ideal’ RGs top ranked by geNorm should be independently regulated and
not co-regulated, because co-regulated genes have high a similarity in their expression profile.
To avoid selection of co-regulated genes, the tested CRGs are chosen from distinct biological
Chapter 3
100
processes and metabolic pathways. However, it is also possible that two CRGs show a strong
bias towards either down- or up-regulation during the experiment, regardless of their origins
in independent cellular functions. In other words, the experimental condition could regulate
them in a way that results in a high similarity of expression. As a result, geNorm erroneously
top ranks these genes, regardless of whether they have a high level of variation or not.
However, based on the generally accepted criterion, RGs should show invariant levels of gene
expression in a given test set and experimental design. Because it is not currently possible to
presume any systematic variation of CRGs which can be caused by experimental conditions
before performing the experiment, we believe that independent evaluation of each CRG is
necessary to reduce the risk of artificial selection of variant CRGs. Upon comparison of the
results of a correlation study (Table 3.2) with geNorm results (Figure 3.2), we observed that
the CRGs with a high correlation are chosen as most stable RGs by geNorm. However, when
the within-group variations of CRGs for each tissue are taken into to account (Table 3.5), we
observed that in leaves, UBQ10 (variation of 0.079) and YLS8 (variation of 0.113) ranked as
the second and fifth out of eight CRGs. In stem tissue, ACT and EF-1a (correlation of 0.92)
(Table 2) are found to be most stable by geNorm (Figure 3.2). However, EF-1a (0.042) and
ACT (0.043) ranked as the fourth and fifth among CRGs. The situation is better in root tissue
because the selected CRGs by geNorm, UBC2 and YLS8 (correlation of 0.90) (Table 3.2),
ranked as the second and third invariant CRGs. BestKeeper uses Ct values (instead of relative
expressions) as input to identify the best normalizers among the investigated candidate
reference. For this purpose, BestKeeper employs Pearson correlation analysis, which is a
parametric method. Therefore, the Pearson correlation coefficient is only valid for normally
distributed data with a homogeneous variance. If the data do not meet these assumptions,
using a parametric test can thus lead to false results. Although Pfaffl et al. (2004 stated that
the Ct values seem to be best estimators of the expression levels, some reports warn of
Identification of reliable reference genes in red clover
101
subjecting Ct values to parametric methods (Schmittgen et al., 2000) due to Ct values being
logarithmic in nature (Pelch et al., 2010), if ever normally distributed without a linear scale
(Kortner et al., 2011). BestKeeper was developed based on the assumption of normal
distribution of input data (Ct values). To calculate the BestKeeper index in the original paper,
Pfaffl et al. (2004) used three CRGs, UBQ, GAPD and β-ACT. When we subjected the Ct
values of those three CRGs to Shapiro-Wilk’s normality test, we realized that UBQ (p values
of 0.016) and GAPD (p values of 0.006) do not have normal distribution and β-ACT (p values
of 0.052) has only a poor normal distribution. Violating normal distribution assumptions
raises questions about the accuracy of the results obtained by parametric statistical methods in
BestKeeper. Although the authors promised to develop a new version of BestKeeper software
using Spearman and Kendall Tau correlations for non-parametric data, those versions are not
yet available. The determination of standard deviation for each investigated CRG is
considered as the major advantage of BestKeeper. BestKeeper then eliminates the CRGs
which show high standard deviation (SD≥ 1) from further calculation. However, Schmittgen
et al. (2000) believe that presentation of statistical data calculated from the raw Ct values
falsely represents the error and should be avoided. As earlier mentioned, BestKeeper employs
pairwise correlation analysis to determine suitable RGs, BestKeeper top ranks the RGs which
display similar expression patterns across samples as well as geNorm (Ponton et al., 2011). In
our study, the results indicated that the CRGs were chosen as highly stable RGs in different
tissues by geNorm. They were also top ranked by BestKeeper despite the existence of CRGs
with less variation in the tested tissues, but they were considered to be less stable CRGs when
using geNorm and BestKeeper. This result was in agreement with the previous report that
GAPDH, which showed the lowest overall CV of all genes, was ranked second to last by
geNorm and last by BestKeeper (Kortner et al., 2011).
Chapter 3
102
The NormFinder algorithm uses a model-based approach which takes variations across
subgroups into account and avoids artificial selection of co-regulated genes (Andersen et al.,
2004). NormFinder, which is less sensitive to co-regulation of the RGs, enables ranking of all
reference gene candidates based on intra- and intergroup variations. Users should understand
that the intergroup variation calculated by NormFinder does not show systematic variation
across the sample subgroups of a single CRG. In reality, the intergroup variation shows to
which degree the data from group 1 of a particular CRG varies compared to the mean
variation of data from group 1 for all tested CRGs (Kortner et al., 2011). Therefore, the
accuracy of the result achieved by NormFinder is still based on the assumption that the
average of the tested candidate genes should show no systematic variation. In the original
paper, the stability of expression of the CRGs subjected to NormFinder were already proved
by microarray analysis (Andersen et al., 2004). In practice, such microarray expression data is
usually not available. If the tested CRGs show a similar systematic variation, the mean
expression will be biased toward the variant genes. In this case, the top ranked genes will not
be the most stable; in reality, they are top ranked because of their similar expression to the
biased mean expression. In our study, the NormFinder algorithm ranked YLS8 and EF-1α as
the best combination of RGs across tissues (Table 3.4). However, a visual inspection of the
descriptive plot of Ct values (Figure 3.1) and also the result of the stability index achieved by
ANOVA (Table 3.5) revealed that EF-1α occupied the fifth position out of eight CRGs, thus
it cannot be considered as stable reference gene. ACT ranks as the least stable CRG across
tissues (Table 3.4) with the highest intergroup variation by NormFinder (Figure 3.3).
However, Figure 3.1and Table 3.5 show that ACT was not the least stable gene. This result is
in accordance with the study of Kortner et al. (2011), who reported that GAPDH (the lowest
FCRGs) vary systematically, but this assumption does not always reflect reality. CRGs must
therefore be tested with statistical methods which are free from this assumption. In previous
Identification of reliable reference genes in red clover
103
studies, the application of standard parametric statistical techniques (i.e., t-test, regression,
ANOVA) for independent assessment of expression of CRGs were reported that used Ct or
relative expression ratios as input data. The most notable examples of these previous reports is
the study performed by Brunner et al. (2004) who proposed a stability index calculated by
multiplication of the coefficient of variation (CV) with the slope of regression (b). To
determine intragroup variation, the raw Ct values were subjected to ANOVA to calculate CV
via division of MSE by the mean, and then were multiplied by 100. However as mentioned
above, Schmittgen et al. (2000) disagreed with calculation of CV from the raw Ct values. To
account for intergroup variation, Brunner et al. (2004) calculated the slope of regression for
each gene across different groups based on their Ct values. However, the regression slope (b)
presents only the linear part of intergroup variation. But intergroup variation consists of linear
variation and nonlinear variation. Thus we believe that the application of the linear regression
slope (b) could cause inaccurate estimation of intergroup variation. Also, although plotting Ct
is informative, the raw Ct values do not have a linear scale and also are not corrected for
primer efficiency (Kortner et al., 2011). Therefore, the direct use of raw Ct values for
ANOVA and linear regression could lead to poor estimation and erroneous results. A recent
study conducted by Kortner et al. (2011) is another example of independent evaluation of
CRGs using relative expression ratios as input file. They proposed total CV as the stability
index. They argued in cases where two CRGs have the same total CV, the gene with lower
intergroup variation is a suitable reference gene for normalization. However, it is unknown to
which degree the intra- and intergroup variation contribute to the total variation (CV). They
therefore suggest another stability index, the F value, which is the ratio between the
intergroup and the intragroup variation. They claimed that stable RGs should have small total
CV and F values. However this cannot always be correct. Imagine two CRGs (x and y), in
which the intergroup variation of gene x is two times bigger than its intragroup variation. The
Chapter 3
104
F value of gene x thus equals 2. The intergroup variation of gene y is two times bigger than
the intergroup variation of gene x, and the intragroup variation of gene y is four times bigger
than intergroup variation of gene x. Therefore, the F value of gene y equals 1. Although gene
y has a smaller F value, its total CV will be much higher than gene x. That is why Kortner et
al. (2011) were faced with discrepancies when ranking genes with total CV and F values.
Most importantly, they performed parametric statistical methods on relative expression ratios
which have no symmetrical scale (Gorte et al., 2011). Subjecting Ct values or relative
expression ratios to standard parametric statistical techniques is an egregious error that has
also occurred in previous studies. It is now important to ask what kind of data can be
subjected to statistical methods. When taking the formula used to calculate the M value in
geNorm, one sees that the relative expression ratios were log2 transformed. By default,
NormFinder uses the logarithmic transformation of the raw relative quantities (Rytkonen et
al., 2010). We therefore decided to perform natural log-transformation on the relative
expression ratios to model fold changes in an additive way (Szabo et al., 2004) and to satisfy
the assumption of normality for ANOVA . We then propose a novel stability index based on
ANOVA using the natural logarithm of the relative expression ratios as input data. We
divided MSW into its components (different tissues) which enabled us to distinguish the
suitable CRGs with the lowest variation in each tissue. Moreover, when observing the
components of MSB, one can identify the tissue which has the lowest or highest contribution
in intergroup variation in the corresponding CRG. Such information can help researchers in
preliminary identification of candidate genes. If one wants to conduct a new qRT-PCR
experiment for expression analysis in some of the tissues tested in the current study, the genes
showing a lower contribution in intergroup variation can be appropriate CRGs in
corresponding tissues. Also, because the component of stability index (VB and VW ) are
known in our study (Table 3.5), if one encounters two CRGs with very close or the same
Identification of reliable reference genes in red clover
105
stability index, the gene with the lower VB can be simply chosen as proper RG. Another
advantage of the proposed stability index is that a limited number of CRGs can be evaluated
in all algorithms, whereas any number of CRGs can be evaluated by our proposed stability
index. Also, unlike NormFinder, inter- and intragroup variations do not have the same weight
when computing its stability index. In our proposed stability index, these two components are
equally weighted.
Selecting reliable RGs is of crucial importance for qRT-PCR studies. Several available
algorithms, including geNorm, BestKeeper and NormFinder, have been developed to identify
reliable RGs among the CRGs being studied. All of the algorithms are based on the
assumption that none of the tested CRGs show variability in their expression across all of the
samples being considered. Observation in practice disproves this assumption. In our study, we
revealed that in cases where the underlying assumption is violated, the application of geNorm,
BestKeeper and NormFinder for selecting the best RG(s) could greatly heighten the risk of
artificial selection. We propose an alternative, model-based method for computing stability of
CRGs expression which is free from the assumption used in these analysis tools. Simplicity of
calculation and accuracy are the major advantages of our model-based stability index as
compared with the previously developed model-based methods, because their application is
either too intricate (Serrano et al., 2011; Szabo et al., 2004) or they do not take the efficiency
of PCR into consideration (Brunner et al., 2004). In this study, we have attempted to alert
users to the assumption and limitations of the available algorithms. Of course, if the
assumption is met, the outputs of these algorithms can be reliable.
To the best of our knowledge, in previous expression studies of red clover, only ACT or EIF-
4a have been used as a normalizer in leaf samples. However, neither ACT nor EIF-4a have
been statistically evaluated for their expression stability. The present study is the first
systematic comparison of potential RGs to determine the best-performing genes for accurate
Chapter 3
106
normalization in red clover tissues (i.e., leaves, stems and roots). For leaf tissue, UBC2 and
UBQ10 showed the least variation of the eight CRGs and were thus identified as being the
most stable RGs in leaf. For stem tissue, UBC2 and YLS8 presented the lowest variations and
were thus determined to be the proper RGs for stems. For red clover root tissue, EIF-4a and
UBC2 showed the lowest variability; we therefore believe these to be suitable RGs for the
root of red clover. Keep in mind that the red clover plants used for this study were kept under
greenhouse conditions (see Materials and Methods, above). Sampling was performed in the
late vegetative stage of the tested plants without any treatment applied. The abovementioned
CRGs, which were found to be the best RGs for their corresponding tissues, can therefore
only be used as normalizers for expression studies of any genes of interest needing to be
evaluated under the same particular conditions. For mRNA expression normalization of the
combined plant tissues (leaf, stem and root combined), YLS8 and UBC2 showed the highest
stability for red clover kept under the abovementioned conditions and sampling protocol.
YLS8 and UBC2 were therefore the most appropriate RGs for normalization of any genes
needed to be assessed across tissues of red clover grown under these conditions. In this study,
we sampled three tissue types from 10 non-treated red clover plants to evaluate the expression
stability of eight CRGs using the commonly used statistical algorithms geNorm, BestKeeper
and NormFinder as well as a new method proposed in this study. Our proposed method can be
also used to identify reliable RGs among CRGs of a tissue (e.g., leaves) that has been
subjected to different treatments. For this purpose, similar to the way that each tissue was
considered as a group in this study, each treatment can be considered as a group. The CRGs
with low within-group variations in a treatment are considered as suitable RGs in the
corresponding treatment. To determine the most stable RGs across treatments, the stability
index should be calculated for each CRG in the same way that we calculated the stability
Identification of reliable reference genes in red clover
107
index across tissues in this study. The genes with a low stability index are considered to be the
best RGs across the treatments applied.
Chapter 4: RNA interference silencing of
isoflavone synthase gene in red clover
RNA interference silencing of isoflavone synthase gene in red clover
111
4.1. Introduction
In Chapter 2 above, we presented an optimized regeneration and gene transformation
methods for red clover. We then suggested suitable reference genes for normalization of
qPCR data for comparative transcriptomics analyses in red clover in Chapter 3. In this
chapter, we discuss silencing of the IFS gene in red clover using RNA interference (RNAi),
a post-transcriptional gene-silencing technique. RNAi has been recognized as part of the
immune response against viruses in plants by RNA degradation mechanism (Nakahara et
al., 2012). Shortly after the discovery of RNAi, researchers realized that by taking
advantage of this method, a target endogenous gene can be also silenced by introduction
of a hairpin R N A construct of the targeted gene via a gene transformation technique
(Walti et al., 2006). In this way, the function of the silenced target gene can be evaluated.
Such loss of function allows functional analyses of hundreds genes by RNAi technique. To
use RNAi method in functional studies, we need to make a hairpin construct which harbors
a gene specific sequence tag (GST) in an inverted repeat separated by a spacer (intron). The
construct should be driven by a strong constitutive promoter such as the Cauliflower mosaic
virus promoter (CaMV35SP). Then the binary vector containing RNAi construct should be
transferred into Agrobacterium tumefaciens and is subsequently used for plant
transformation. After transcription of the transgene in a transformed plant, the intron from
mRNA of the inserted construct is spliced out and the GST sequences folds around on itself,
leading to production of a hairpin structure which triggers RNAi degradation mechanism
(Mello and Conte, 2004). The traditional method to create an R N A i construct of an
interested gene is to use restriction digestion and ligation of insert into T-DNA of a binary
vector, but this is time-consuming and laborious. Invitrogen (Carlsbad, CA, USA) has
introduced an alternative method to overcome this technical difficulty: the GatewayTM
cloning system. This system was designed based on interaction of bacteriophage lambda
Chapter 4
112
with its host E. coli (Traore and Zhao, 2011). The use of site-specific recombination system
of the bacteriophage lambda eliminates the need for the restriction enzyme digestion and
ligation steps. This system is thus easier and faster than traditional cloning methods.
Gateway cloning is performed via site-specific recombination using the attachment sites of
bacteriophage lambda (attP) and E. coli (attB). The procedure of integration of the
bacteriophage lambda into the E. coli chromosome and excising the bacteriophage lambda
from E. coli has been adapted for use with the Gateway system. Naturally, the attP site
(242 bp) from bacteriophage lambda and the attB site (25 bp) of the host E. coli
recombines (BP reaction) and, upon integration of phage lambda into E. coli, new
recombination sites called attachment Left (attL-100bp) and attachment Right (attR- 168
bp) are formed inside E. coli. To excise the phage DNA from the E. coli chromosome, a
reverse recombination reaction occurs between attL and attR sites (LR reaction). In
practice, the first step in the Gateway cloning method is to amplify the DNA of interest
with specific primers with additional sequences (attB sites) fused to the 5’ end of the
primers. This is accomplished by means of the polymerase chain reaction (PCR) with a
high-fidelity polymerase. The attB-flanked PCR product can be recombined with an attP
containing vector by BP Clonase enzyme mix, which contains two enzymes: the phage
integrase (Int) and the E. coli integration host factor (IHF) (Fu et al., 2008). The result of
BP recombination is the generation of a pENTRY clone that contains attLs recombination
site flanking the insert. Subsequently, the at tL-fragment from the pENTRY clone can be
recombined to the attR sites of the destination vector during incubation of a pENTRY
vector and a destination vector in presence of LR Clonase enzyme mix (Earley et al.,
2006; Fu et al., 2008). Plant destination vectors are available for a variety of
applications including gene knockdown via RNA interference (Karimi et al., 2002). In this
chapter, we present and discuss: 1) the construction of double-stranded RNA (dsRNA) from
RNA interference silencing of isoflavone synthase gene in red clover
113
the coding region of IFS, 2) the transformation of red clover using Agrobacterium
tumefaciens strain C58C1 RifR (pMP90) containing a binary hairpin vectors, 3) screening
transgenic lines using PCR, and 4) the preliminary analysis of IFS expression level in
transgenic and non-transgenic lines.
4.2. Materials and methods
4.2.1. Genes of interest
According to the NCBI gene database, isoflavone synthase (IFS) with GenBank accession
numberAY253284 is responsible for expression of isoflavone in red clover. In this study, IFS
was subjected to the RNAi technique.
4.2.2. RNAi vector construction
In this study a RNAi construct was developed to suppress the isoflavone synthase gene. The
IFS silencing construct was generated using the Gateway system as described below.
4.2.3. DNA/RNA concentration measurement
Nucleotide concentrations were measured using a Nanodrop ND-1000 spectrophoto-
meter (Isogen, IJsselstein, the Netherlands). The DNA/RNA elusion buffer was used as blank
value. For the measurement, 1.5 μl of DNA/RNA solution was loaded into the instrument.
4.2.4. Primers design
Primers were designed from the genomic sequence of the IFS gene using Vector NTI
AdvancedTM
10 (Invitrogen).
4.2.5. Development of the RNAi construct
To create the RNAi construct, a 500 base pair from position of 1284 to 1783 in the coding
region of IFS gene was selected. The fragment was amplified in a PCR reaction using attBs-
attached to the gene specific primers attB1F-IFS-Forward (GGGGACAAGTTTGTAC
AAAAAAGCAGGCTTAAGAGCCATGGTGAAAGAGGT) and attB2R-IFS-Reverse (GGG
Chapter 4
114
GACAGCTTTCTTGTACAAAGTGGGAGAAAGGAGTTTAGCTGCACCAG). Platinum
Taq DNA Polymerase High Fidelity (Invitrogen) was used for amplification of IFS as
described in the manufacturer’s protocol. The cycling program was as follows: initial
denaturation at 94°C for 2 min, 30 cycles of denaturation at 94°C for 30 s, annealing at 54°C
for 30 s, extension at 68°C for 2 min and a final extension step at 68°C for 10 min.
4.2.6. The BP reaction
To generate a pENTRY plasmid, the amplified fragment was recombined with pDONR 221
(Invitrogen, Carlsbad, CA, USA) in a BP recombination reaction. Briefly, 50 ng DNA of the
donor vector (pDONR 221) and same amount of the attB-containing PCR fragment were
mixed with 1 µl of BP ClonaseTM
Enzyme Mix and 1 ul from stock of 5X BP Clonase
Reaction Buffer, then water was added to a total volume of 5 µl. The mixture was incubated
at 25 °C overnight. Prior to E. coli transformation, the BP clonase enzymes were destroyed by
adding 1ul proteinase K to the BP reaction mixture followed by 10 minutes of incubation at
37 °C.
4.2.7. Transformation of E. coli
A vial of E. coli DH5α competent cells (50 μl) stored at -80°C were thawed on ice and mixed
with an aliquot of the BP reaction mixture. The cells were incubated on ice for 10 minutes,
then heat shock transformation was done in a water bath at 42 °C for 90 seconds. After the
heat shock treatment, the cells were put back on ice for 1 minute and then diluted with 350μl
SOC medium (preheated to 37 °C). The cells were incubated for 1 hour at 37 °C on a shaker
at 200 rpm. The cultures were plated at 150 μl onto LB medium supplemented with 50 μg ml-1
kanamycin and plates were placed at 37 °C overnight to obtain transformed single colonies.
4.2.8. Plasmid DNA preparation
A few E. coli colonies were picked from LB-kanamycin plates using a sterile pipette tip
which was used to inoculate 2 ml liquid LB (50 μg ml-1 kanamycin) into a sterile 15 ml
RNA interference silencing of isoflavone synthase gene in red clover
115
Falcon tube. Culture tubes were placed at 37 °C on a rotary shaker at 200 rpm overnight. To
extract the plasmid DNA of selected colonies, a GeneElute™ Plasmid Miniprep kit was used
according to the manufacturer’s instructions (Sigma-Aldrich). The extracted plasmid DNA
was analyzed by digestion and sequenced to verify successful fragment insertion.
4.2.9. Restriction endonuclease digestion of pEN-L1-IFS-L2 vector
To confirm the presence of the fragment of interest in pEN-L1-IFS-L2 vector, an aliquot of
extracted plasmid DNA was used for digestion with EcoRV restriction enzyme. The
restriction digestion reaction was performed in a total volume of 40 μl containing 1 µg DNA
of pEN-L1-IFS-L2, 1 µl (5 units) EcoRV enzyme 4 μl digestion buffer and sterile water up to
40 µl. Reactions were kept at 37 ºC for 1 h.
4.2.10. Agarose gel electrophoresis
The digested fragments were separated by horizontal agarose gel electrophoresis. Agarose
(1.5% w/v) was suspended in 1x TAE buffer by melting. After cooling down to
approximately 60 °C, the gel was poured into a clean gel-casting frame. Combs with the
appropriate number and size of the teeth were used to make the loading slots. After the gel
solidified, digested DNA samples along a molecular size marker containing 1x loading dye
(0.25% bromophenol blue, 0.25% Xylencyanol and 30% glycerin in water) were loaded into
the gel slots. The electrophoresis adjusted at 100 volts and the agarose gel ran for 30 minutes.
The gel was scanned using UV irradiation and photographed.
4.2.11. DNA sequencing
To confirm the recombination event and whether the recombined fragment in entry clones had
no error in the sequences, the plasmids were sequenced using the BigDyeTM
Deoxy
Terminator Sequencing Kit v.1.1 (Applied Biosystems) according to the manufacturer’s
instructions. M13 forward and reverse universal primers, which are located outside of the
attL1/attL2 region in pEN-L1-IFS-L2, were used for the sequencing reaction. Cycle
Chapter 4
116
sequencing conditions were 94 °C for 1 min and 25 cycles of 94 °C for 10 s, 58 °C for 5 s, 60
°C for 1 min with a final extension at 60 °C for 4 min. The sequencing reactions were
subjected to ethanol precipitation to remove excess dye-terminators. The sequencing products
were dissolved in formamide and then heat-denatured for 8 min before loading onto a
capillary ABI3130 genetic analyzer system. Bases were determined, and sequences were
assembled using Sequencing Analysis software v.5.2 (Applied Biosystems). The nucleotide
sequences obtained were aligned against the reference sequence provide by NCBI using
Vector NTI AdvancedTM
10 (Invitrogen).
4.2.12. The LR reactions
Following sequence verification of pEN-L1-IFS-L2 , an aliquot of plasmid DNA from the
sequenced entry clone was used to recombine the fragment into the chosen destination vector
pK7GW1WG2 (Karimi et al., 2002) for RNA interference using the LR clonase reaction
enzyme mix (Invitrogen, USA). Briefly, 50 ng DNA of pEN-L1-IFS-L2 clone was mixed
with 50 ng DNA of destination vector and 1 µl of 5 x LR reaction buffer and 1µl of LR
ClonaseTM
Enzyme Mix then filled up to 5 µl with water. The result of LR recombination was
generation of an expression clone was named pXK7-IFS-I-SFI-2.
4.2.13. Selection of a clone with the correct orientation
The LR reaction was incubated overnight at 25 °C. After inactivation of LR clonase enzymes
by proteinase K (see 4.2.7 above) an aliquot of the reaction mix was used to transform E. coli
competent cells. The transformed E. coli cells were plated overnight on solidified LB plates
containing 100 mg/l spectinomycin (Sp). A few grown clones were cultured overnight in
liquid LB with 100 mg/l spectinomycin and they subjected to plasmid extraction as described
above. All the culture media used to grow E. coli cells containing pXK7-IFS-I-SFI-2 vector
were supplemented with 100 μg ml-1
spectinomycin.
RNA interference silencing of isoflavone synthase gene in red clover
117
4.2.14. Restriction endonuclease digestion of pXK7-IFS-I-SFI-2 vector
To confirm the presence of the fragment of interest and also to select a clone with the correct
orientation of the intron, an aliquot of extracted plasmid DNA was used for digestion. A
restriction digestion reaction was set up as following: in a total volume of 40 µl, 1 µg DNA of
pXK7-IFS-I-SFI-2 plus 1 µl (5 units) Xba1 restriction enzyme (Fermentas) and 4 µl
appropriated buffer was added in an Eppendorf tube then sterile water was added up to 40 µl.
The reaction incubated for 1 hour at 37 ºC. The generated fragments were then analyzed in a
1.5% agarose electrophoresis gel (see 4.2.10 above).
4.2.15. Transformation into Agrobacterium tumefaciens
Transformation of the plasmid into Agrobacterium requires competent Agrobacterium
tumefaciens cells. To perform the transformation, an aliquot of Agrobacterium C58C1 RifR
(pMP90) was cultured overnight in 10 ml YEB medium (5 g/l yeast extract, 10 g/l tryptone, 5
g/l sodium chloride, 5 g/l sucrose, 0.5 g/l MgSO4 supplemented with 50 μg ml-1
rifampacin) at
28 °C with 150 rpm shaking. The following day, the culture was cooled on ice for 10 min and
was pelleted by centrifugation at 5000 rpm for 1 min at 4 ºC. The bacteria pellet was re-
suspended in one-half volume of ice-cold 20 mM CaCl2 solution and kept on ice. Then 50 ng
of the extracted plasmid (pXK7-IFS-I-SFI-2) was mixed gently with 50 μl of competent
Agrobacterium tumefaciens cells. The mixture was frozen in liquid nitrogen for 3 minutes and
then thawed in a water bath at 37 ºC for 5 minutes. The culture was incubated at 28 ºC on a
shaker at 150 rpm for 1 h after adding 350 µL of SOC medium. Subsequently the culture was
spread on YEB medium-agar plates with 50 mg l−1
rifampicin and 100 mg l−1
spectinomycin.
The colonies became visible after two days.
4.2.16. Verification of pXK7-IFS-I-SFI-2 in A. tumefaciens by digestion
Several grown colonies of A. tumefaciens were picked and cultured overnight in YEB
medium supplemented with 50 μg ml-1
rifampacin at 28 °C with shaking at 150 rpm. The
Chapter 4
118
plasmids DNA were extracted (see 4.2.8) and transferred to E. coli DH5α competent cells and
then cultured overnight (see 4.2.7). The plasmid DNA of several colonies was extracted and
digested with Xba1 (see 4.2.14). The resulting DNA fragments were separated by horizontal
agarose gel electrophoresis (see 4.2.10)
4.2.17. Plant transformation
All transformation procedures were performed as described in Chapter 2 above. Briefly,
petioles of young leaves from red clover plants (genotype NEWR65) which had been clonally
propagated and aseptically maintained in glass jars, were used for transformation. Freshly
prepared petioles were cut into sections of approximately 5 mm and placed on KB2. The
explants were inoculated by overnight culture of Agrobacterium C58C1 RifR (pMP90)
harboring the RNAi construct (pXK7-IFS-I-SFI-2). All infected petioles and four petri dishes
of uninfected petiole sections (controls) were placed on co-cultivation medium (KB2) for 48 h
in the growth chamber. Subsequently, the explants were removed from the KB2 and rinsed
three times with sterile water. The explants were transferred to KB2 medium supplemented
with 500 mg l−1
of timentin to remove excess bacteria and 75 mg l−1
of kanamycin sulphate
(based on our previous study) for selection of transformed tissue. Half of the uninfected
petioles were transferred to KB2 medium with 500 mg l−1
of timentin without kanamycin for
plant regeneration. The remaining uninfected controls were transferred to KB2 medium
containing 75 mg l−1
kanamycin sulphate for kanamycin selection. The cultures were sealed
and incubated for 30 days in the growth chamber. Afterwards, infected explants were
transferred to shoot induction (SPL) medium containing 500 mg l−1
timentin and 75 mg l−1
kanamycin sulphate for regeneration under the same environmental conditions. Uninfected
explants were subcultured on SPL medium with kanamycin sulphate. Also, uninfected
explants without selection were transferred to SPL medium with 500 mg l−1
timentin for
regeneration. Then kanamycin-resistant embryos were transferred to SPL medium containing
RNA interference silencing of isoflavone synthase gene in red clover
119
500 mg l−1
timentin and 75 mg l−1
kanamycin sulphate and were kept for 30 days in the
growth chamber for shoot production. For root induction, all produced shoots were placed for
another 30 days on SPL medium supplemented with 500 mg l−1
of timentin and 75 mg l−1
of
kanamycin sulphate until roots were appeared. Rooted shoots were transferred to soil for
further growth and analysis.
4.2.18. PCR analysis
The existence of transgenes in putative transformed plants was confirmed by polymerase
chain reaction (PCR). Briefly, genomic DNA from putative transformants and untransformed
control plants (wild type) was extracted from young leaves using DNeasy®
Plant Mini Kit
(Qiagen, USA). DNA concentration and purity was measured with a NanoDrop® ND-1000
Spectrophotometer. A pair of primers was designed for this propose. The forward primer
(TCGACAATTTGACTTTCAAGAGTAG) was from intron sequence of the pXK7-IFS-I-
SFI-2 expression vector, and the reverse primer (ACCAAAGG GCTATTGAGACT TT) was
from IFS region. The expected size of amplicon was 1512 bp long. The PCR reaction was
carried out in 50µL and was contained of 100 ng plant DNA, 20 pmol of each primer pair,
0.25 mM dNTPs (Invitrogen, USA), 5µL of 10x reaction buffer (Applied Biosystems, USA)
and 1.25 unit of Taq DNA polymerase (Applied Biosystems). PCR was performed in a
thermal cycler (Applied Biosystems) with the following program: initial denaturation at 94 °C
for 2 min, followed by 30 cycles of denaturation at 94 °C for 15 s, annealing at 55 for 30 s,
elongation at 72 °C for 90 s and a final elongation at 72 °C for 10 min. PCR products were
separated on 1.5% agarose gel and detected under UV light after ethidium bromide staining.
4.2.19. qRT-PCR analysis of transgenic plants
The suppression level of the studied gene (IFS) was verified using qRT-PCR technique. For
this purpose, a pair of primer (forward; 5′- GGTACGT GATCCCAGAAGGA-3′ and reverse;
5′-CCTCTCAGGACGAAATTCCA-3′) was designed from IFS according to the primer
Chapter 4
120
design method described in Chapter 3 above. Then three leaves from each independent
putative transformants (RNAi_1 to RNAi_5) and untransformed control plants (wild type,
WT) were collected. The RNA isolation, cDNA preparation and qRT-PCR conditions and
data processing (normalization) were performed as mentioned in Chapter 3.
4.3. Results
4.3.1. Hairpin binary vector
A 500 bp fragment from the coding sequence of IFS chosen to generate a hairpin construct,
the fragment was amplified using gene specific forward and reverse primers flanked by attBs
sequences (Figure 4.1).
1 GCCACACACA AAAACCACAG CAACATAAAA TTTGCCATTT CTCATCATTT AGCAGATACC
61 AAAGAATCTG CAACCATGTT GTTAGAAATT GCAGTTGCTT TATTGGTCAT AGCTTTGTTC
121 ATTTATTTGC GTCCAACACC CACCGCTAAA TCAAAAGCAC TTCGTCACCT TCCAAATCCA
181 CCAAGTCCAA AACCTCGTTT ACCATTCATT GGTCATCTTC ATCTTTTAGA TCATCCTCTT
241 CTTCACATAT CTTTGATCCG TTTAGGTGAA CGTTATGGCC CTTTATACTC TCTTTACTTT
301 GGTTCTATGC CAACTGTTGT AGCTTCAACC CCTGATTTGT TCAAACTCTT TCTTCAAACT
361 CATGAAGCTA CTTCTTTTAA CACAAGGTTT CAAACCTCTG CTATTAGACG TTTAACTTAT
421 GATAACTCTG TTGCTATGGT ACCCTTTGGA CCATACTGGA AATTTGTTAG GAAACTTATT
481 ATGAATGATC TTCTTAATGC TACTACTGTT AACAAGTTGA GACCTTTGAG GAGTAAAGAG
541 ATTCGTAAGG TTCTTAATGT TATGGCTAAC AGTGCTGAAA CTCAACAACC TTTGAATATC
601 ACTGTTGAGT TACTTAAGTG GACAAATAGT ACTATTAGTA CTATGATGTT GGGTGAAGCT
661 GAAGAGGTTA GAGATATTGC TCGTGATGTT CTTAAGATCT TTGGGGAGTA TAGTGTTACT
721 GATTTCATTG GGCCTTTGAA GATTTTCAAG AAGTTTGGAA ACTATGAGCA GAGGATTGAT
781 GCTATTTTTA ATAAGTATGA TCCTATTATT GAAAGGGTTA TCAAGAAAAG ACAAGGGATT
841 GTGAACAAAA GAAAGAATGG AGAAGTTCTA GTTGGTGAGG AGGAGAATGT AGTTTTTCTT
901 GATACTTTGC TTGAATTTGC TCAGGATGAG ACCATGGAGA TCAAAATTAC TAAGGAACAA
961 ATCAAAGGTC TTGTTGTGTC AAAGGTCTTG TTGTGGTAAG TTTCTTTTTG TTCTAATTGA
1021 TTTGATATTA AATACTTGTT TTGCGACTCT GGTCTTTAGT AGAAGAAAAA ATTTACTTTT
1081 TAGATTCATT GAATAATTGT TGTATTTGTT TGCAGGATTT CTTCTCTGCT GGTACAGACT
1141 CAACGGCTGT GGCAACCGAA TGGACTTTGG CGGAGCTTAT CAATAATCCG AGGGTGTTGA
RNA interference silencing of isoflavone synthase gene in red clover
121
1201 AGAAAGCTCG CGAGGAGGTT GAATCAGTTG TGGGAAAAGA TAGACTTGTT GACGAATCAG
1261 ATATTCAGAA TCTTCCTTAC ATTAGAGCCA TGGTGAAAGA GGTATTCCGC TTGCACCCAC
1321 CGCTACCTGT GGTTAAGAGA AAATGTACCG AAGAGTGTGA GATCAACGGG TACGTGATCC
1381 CAGAAGGAGC ATTGATACTT TTCAATGTAT GGCAAGTTGG AAGAGACCCA AAATATTGGG
1441 AGAAACCATT GGAATTTCGT CCTGAGAGGT TCTTAGAAAA TGCTGGTGTA GGTGAAGGTG
1501 AAGCGTCTTC AATTGATCTT AGGGGTCAAC ATTTCACACT TCTACCATTT GGGTCCGGAA
1561 GAAGAATGTG TCCTGGAGTC AATTTGGCTA CCGCCGGAAT GGCCACACTA CTTTCATCTA
1621 TTATCCAATG CTTTGATCTA CAAGTACCTG GTCCTAATGG ACAAATATTG AAAGGTAGTG
1681 ATGCTAAGGT TACCATGGAT GAGAGACCTG GTCTCTCTGT TCCAAGAGCA CAAAATCTTG
1741 TGTGTGTTCC ACTTGCAAGA GCTGGTGCAG CTAAACTCCT TTCCTCTTAA ATCATATTTG
1801 GAACAAAAGA AATGGTTGAC ATTAGTCATG GTTCATGTTT TTATATTGTA GTAAAAATCC
1861 TTTGCAATAA GATATTATTG AGAGACGGTG TGAATCTGAT GAACATTGTC TCTCTTTGGT
1921 ATGATGTAAT TTTGTTCTTT TCCTTTTTGT TCTTGTCACA AAAGTGTTGT AATTGCG
Figure 4.1: Sequence of IFS gene (AY253284). Arrows indicate the position of primers. The
amplified fragment is shown in red.
The recombination of the PCR product containing attB sites with pDONR 221 vector
containing attP sites was performed by a BP reaction to generate pEN-L1-IFS-L2 clone
(Figure 4.2). To confirm the insertion of the fragment in pEN-L1-IFS-L2 clone, restriction
enzyme digestions were performed. Miniprep plasmid DNA from eight positive clones was
digested with EcoRV and digests were fractionated on 1.5% agarose gel (Figure 4.3). The
results confirmed existence of the desired fragment in pEN-L1-IFS-L2 clone. For further
verification, the clones with proper size were sequenced. The obtained sequences were
aligned to the sequence of IFS (NCBI). The alignment result showed 100% identity between
sequenced of clones and IFS sequence. The sequence-validated pEN-L1-IFS-L2 clones were
used in the LR reaction.
Chapter 4
122
Figure 4.2: Physical map of pEN-L1-IFS-L2 clone containing a 500 bp fragment of the red clover
isoflavone synthase (IFS) gene from BP reaction. Vector contains: attL1/attL2: Gateway
recombination sites; Kan (R): Kanamycin resistance gene.
Figure 4.3: DNA of pEN-L1-IFS-L2 clones digested by EcoRV. Lanes 1-8 represent linearized
vectors with correct size of 3044 bp.
4.3.2. Generation of expression vector
The pK7GW1WG2 destination vector and validated pEN-L1-IFS-L2 clone were used to
perform a Gateway LR recombination reaction to generate the expression vector pXK7-IFS-I-
SFI-2 (Figure 4.4).
pEN-L1-IFS-L2
3047 bp
IFS
attL1
attL2
Kan(R)
T7 promoter
ApaI (2433)
AvaI (2426)
EcoRV (103)
NruI (342)
Esp3I (701)
NcoI (2541)
NcoI (2946)
AflII (2420)
AflII (2533)
L 1 2 3 4 5 6 7 8L 1 2 3 4 5 6 7 8
A B
RNA interference silencing of isoflavone synthase gene in red clover
123
Figure 4.4: Map of the pXK7-IFS-I-SFI-2 expression vector. The construct was designed to silence
IFS gene in red clover. Vector contains: LB/RB: Left and Right border of Agrobacterium T-DNA;
Kan (R): Kanamycin resistance gene; T35S: CaMV35S terminator; attB1/attB2: Gateway
recombination sites; P35S: CaMV35S promoter; Sm/SpR: Sterptomycin/ Spectinomycin resistance
marker.
LR recombination resulted in an expression construct that contained IFS fragments in an
inverted repeat configuration. During LR recombination, there is a high possibility that the
intron flips, which results in failure of intron splicing of the expression vector. Therefore,
restriction analysis was performed on five colonies to choose the clones with the correct
intron orientation.
Figure 4.5: An illustration of restriction patterns of expression vectors cut by XbaI enzyme. The XbaI
digestion of expression vector with correct orientation of inserts generated two fragments of 10270
and 1003 bp. Lanes 2, 3,4 and 5 are clones with correct intron orientation; lane 1 indicates a clone with
a flipped intron.
Using Vector NTI software, virtual restriction analysis was carried out on expression RNAi
clones using the XbaI restriction enzyme. The result of the XbaI restriction analysis showed
that the digestion of expression vector with the right orientation generated two fragments
pXK7-IFS-I-SFI-2
11293 bp
IFS
IFS
intron
attB2
attB1
attB1
attB2
Sm/SpR
KanR
p35S
pNOS T35S
Tnos
RB
LB
XbaI (1)
XbaI (1013)
1 2 3 4 5 L 6 7 8 9 10 1 2 3 4 5 L 6 7 8 9 10 1 2 3 4 5 L 6 7 8 9 10 1 2 3 4 5 L 6 7 8 9 10
Chapter 4
124
(10270 pb and 1003 pb). Compared to the virtual restriction analysis, the clones 2, 3, 4 and 5
showed the correct restriction patterns and clone 1 showed a different pattern of digestion,
indicating that this clone contained an insert with a flipped intron (Figure 4.5). The verified
expression vector (clone 5) was transferred to Agrobacterium tumefaciens C58C1 RifR
(pMP90) strain. DNA from five transformed Agrobacteria were cut by XbaI and checked for
the presence of the pXK7-IFS-I-SFI-2 binary vector (Figure 4.6).
Figure 4.6: The DNA of transformed Agrobacterium with hairpin vector was isolated and digested by
XbaI. The result of digestion indicated that the selected vector (clone 5) was presented in all
Agrobacterium clones.
4.3.3. Genetic transformation and plant regeneration
The procedure of red clover transformation using Agrobacterium system consists of three
basic steps: generation of transgenic callus, shoot differentiation and root differentiation. The
protocol used to produce transgenic plants was reported in the “Materials and Methods” (see
4.2.16). Petiole segments from young leaves of red clover (genotype NEWR65) were infected
with Agrobacterium C58C1 RifR (pMP90) harboring the expression vector pXK7-IFS-I-SFI-
2. As mentioned in Chapter 2, use of the media sequence KBC-SPL-SPL significantly
increased the efficiency of somatic embryogenesis of red clover. Therefore, we used this
media sequence for genetic transformation of red clover. The explants after co-cultivation
with Agrobacterium were transferred onto callus induction medium (KB2) supplemented with
500 mg l−1
timentin and 75 mg l−1
kanamycin sulphate for selection of transformed tissue.
1 2 3 4 5 L 6 7 8 9 101 2 3 4 5 L 6 7 8 9 10
1 2 3 4 5 L 6 7 8 9 101 2 3 4 5 L 6 7 8 9 10
1 2 3 4 5 L 6 7 8 9 101 2 3 4 5 L 6 7 8 9 101 2 3 4 5 L 6 7 8 9 101 2 3 4 5 L 6 7 8 9 101 2 3 4 5 L 6 7 8 9 101 2 3 4 5 L 6 7 8 9 10
RNA interference silencing of isoflavone synthase gene in red clover
125
After a total of 30 days incubation of tissues on callus induction medium (KB2), putative
transformed calli were transferred onto SPL medium with 500 mg l−1
timentin and 75 mg l−1
kanamycin sulphate and they placed in culture room for two months for further shoot
production and root development. The regenerated plantlets then were transferred to soil for
further growth and analysis.
4.3.4. PCR analyses
The presence of the transgenes in the genome of the putative T0 transformant red clover
plants via Agrobacterium C58C1 (pMP90: pXK7-IFS-I-SFI-2) was investigated using PCR.
Figure 4.7: A view of agarose gel electrophoresis of PCR amplified products from DNA of individual
putative transgenic plants. Lanes 1-5 corresponded to the individual transgenic lines. PCR product of
untransformed red clover severed as negative control (-) and amplified product of the pXK7-IFS-I-
SFI-2 expression construct used as positive control (+).
The genomic DNA of kanamycin resistant and untransformed (negative control) red clover
plants was extracted. The obtained genomic DNA and plasmid DNA of the pXK7-IFS-I-SFI-2
expression vector (positive control) were used as template for PCR. The amplified PCR
products were separated by agarose gel electrophoresis. PCR gave a predicted band of 1512
bp in transformed red clover plants and pXK7-IFS-I-SFI-2 expression vector (positive
Chapter 4
126
control). No amplification was detected from untransformed (negative control) plants (Figure
4.7).
4.3.5. qRT-PCR analysis of IFS-RNAi in silenced red clover lines
A total of five independent red clover transformants which generated the predicted band for
the CaMV35S promoter in PCR were chosen to determine whether the inserted RNAi
construct is functional in transformants. For this purpose, independent putative transformants
and untransformed control lines (wild type) were subjected to real time quantitative PCR
analysis to detect the level of expression of IFS (Figure 4.8). The level of the IFS transcript
was considerably reduced (approximately 30-fold) in five transformed lines (RNAi_1,
RNAi_2, RNAi_3, RNAi_4 and RNAi_5) containing the IFS hairpin constructs compared to
the control line (WT). This result suggested that the hairpin construct is efficiently silenced
the target gene.
Figure 4.8: The relative expression levels of IFS in five RNAi silenced red clover lines (RNAi_1,
RNAi_2, RNAi_3, RNAi_4 and RNAi_5) and wild type (WT).
4.4 Discussion
Due to the great advances in sequencing technology, the genomes of numerous organisms
have already been sequenced. Nowadays one of the major challenges is functional
characterization of the sequenced genes. Although the undirected methods such as chemical
rela
tive
exp
ress
ion
level
0
5
10
15
20
25
30
35
40
RNAi_1 RNAi_2 RNAi_3 RNAi_4 RNAi_5 WT
RNA interference silencing of isoflavone synthase gene in red clover
127
mutagens, transposons and T-DNA tagging have been very useful for gene functional studies
(Wesley et al., 2001), these methods are time-consuming and laborious. For instance, to
obtain a single insert in a particular gene in Arabidopsis, approximately 350 000 independent
transformants need to be screened (Krysan et al., 1999). The direct method for gene functional
studies such as RNAi technique overcomes some of the limitations of undirected methods. In
this part of our study, we used the RNAi technique to evaluate function of the IFS in red
clover. RNAi is a natural cellular immune mechanism to response to exogenous pathogenic
nucleic acids (Davidson and Mccray, 2011). To apply this phenomenon in gene factional
studies, an inverted repeat sequence of the interested gene should be produced and cloned into
T-DNA regions of binary vectors and then transferred into the plant genome. The expression
of the integrated T-DNA generates dsRNAs of interested gene in transgenic plants. The
dsRNAs trigger a cascade of events that cause the degradation of corresponding mRNA of the
target gene in the cytoplasm, leading to a reduction of the level of the corresponding protein
(Anower 2012). The efficiency of the RNAi method can be influenced by the length of the
fragment chosen to build the hairpin construct (Sayaka and Kodama, 2008). In animals, long
dsRNAs could induce the interferon pathway, leading to cell death (Elbashir et al., 2001a).
Studies on plants have revealed that the appropriate length of selected fragments which enable
to induce efficient silencing can vary from 98 to 853 in different species (Gavilano et al.,
2006; Wesley et al., 2001). However, an effective silencing in various plant species was
achieved when the chosen length was about 200 to 500 bp (Heilersig et al., 2006). In our
study, we chose a 500 bp fragment from the coding sequence of IFS to generate the hairpin
construct. The construction of an RNAi vector via conventional cloning method is
cumbersome and time-consuming. We therefore used the gateway RNAi destination vector
pK7GW1WG2 (Karimi et al., 2002) to generate a hairpin construct for IFS gene. The
pK7GW1WG2 possesses two independent and inverted destination gateway cassettes
Chapter 4
128
separated by an intron (attR1-ccdB-attR2-intron-attR2-ccdB-attR1). During LR reaction, the
two inverted cassettes allow two copies of the target fragment to be positioned head to head in
the resulting RNAi expression vector (pXK7-IFS-I-SFI-2). The intron spacer which is
functional and splices out in pre-mRNA processing greatly enhances the stability of inverted-
repeat sequences in E. coli, facilitating the cloning procedure (Bao and Cagan, 2006).
Moreover, the functional intron increases the number of silenced transgenic plants
(Stoutjesdijk et al., 2002). However, caution should to be taken when selecting the RNAi
expression vector because the orientation of the intron is not correct in 25-30 percent of RNAi
expression vectors generated in LR recombination (M. Karimi, personal communication). The
reverse intron (or flipped intron) which is non-splicing dramatically decreases the number of
independent transgenic plants showing silencing (Smith et al., 2000). In our study, we have
screened five expression vectors to find a clone with the correct intron orientation. The result
obtained from Xba1 digestion of pXK7-IFS-I-SFI-2 expression vectors revealed that 20
present of clones (1 out 5 clones) contained a flipped intron. The clone number 5 that
contained inverted repeat of IFS in a right intron orientation was used in Agrobacterium for
plant transformation. Next, the collected explants from the red clover plant were successfully
transformed using the Agrobacterium harboring the hairpin transgene. The putative
transformant lines were achieved via regeneration. Transgenic lines were selected on
kanamycin containing culture medium. To eliminate any doubt, a second screening in selected
transgenic plants was done based on the presence of the hairpin construct using PCR
technique against CaMV 35S promoter (see Materials and Methods, above). PCR analysis
showed that all selected kanamycin-resistant red clover plants were indeed transformed with
an RNAi construct. An RT-qPCR measurement of the endogenous expression of IFS was
performed with the proper reference genes (see Chapter 3). Our result showed that the average
expression level of IFS in the T0 generation of the transgenic lines was approximately 30-fold
RNA interference silencing of isoflavone synthase gene in red clover
129
lower compared to the controls. This result indicates that the efficiency of RNAi in silencing
of the target gene was higher than 95%. It has been shown the best results of silencing
occurred when a loopless hairpin is formed (Wesley et al., 2001). Indeed, the results of our
experiment showed the same range of silencing. In addition, our results are in accordance
with Sullivan and Hatfield (2006). They stated that RNAi is highly effective in red clover
both in terms of the number of plants displaying silencing and in the extent of silencing of
polyphenol oxidase (PPO) gene. In their study, expression of the PPO gene in leaf extracts of
plants transformed with the PPO-silencing vector showed more than a 90% reduction
compared to the wild type. Their results and our study indicate that RNAi is the most robust
technology for silencing genes at the post transcriptional level in red clover. The five chosen
transgenic lines which showed IFS down-regulation and the wild type plants were transferred
to the greenhouse for propagation and pathogenicity testing.
Chapter 5: Evaluation of the role of
isoflavones in resistance of red clover to
Sclerotinia trifoliorum and Sclerotium rolfsii
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
133
5.1. Introduction
Red clover (Trifolium pratense L.) is an important source of nutrition for sheep and cattle.
Due to its ability to fix elemental nitrogen, red clover is also a good soil improver. Moreover,
red clover is a valuable source of isoflavonoids, an important class of secondary metabolites
derived from the phenylpropanoid pathway. Isoflavones, a subclass of isoflavonoids, are
produced by a limited number of higher plants. Although the production of isoflavones has
been reported in at least 22 plant families, the synthesis of isoflavones are mostly limited to
the subfamily Papilionoideae of the Fabaceae (previously known as Leguminosae) (Yu and
McGonigle, 2005). The isoflavones daidzein, genistein, formononetin and biochanin A are
forms of isoflavonoids abundantly found in different parts of red clover including flowers,
leaves, stems and roots (Saviranta et al., 2008). Several enzymes are involved in the
biosynthesis of isoflavones. The key reaction is catalyzed by isoflavone synthase (IFS). IFS is
an NADPH-dependent cytochrome P450 enzyme (CYP93C1). It catalyzes both the migration
of the B-ring to the 3-position and hydroxylation of the C-2 of flavanones to generate 2-
hydroxyisoflavanones. These are unstable; under conditions of spontaneous dehydration, they
result in the production of isoflavones. IFS has been cloned from soybean, alfalfa, red clover
and several other leguminous plants (Steele et al., 1999; Jung et al., 2000). Because the
enzyme isoflavone synthase (IFS) acts as the key metabolic entry point for the formation of
all isoflavonoid compounds, IFS has been used in most genetic modifications studied within
the isoflavone pathway.
The isoflavones are well known for their role in plant disease resistance. Isoflavones are
considered as preformed antibiotics and also as precursors for the defense-related
phytoalexins (Subramanian et al., 2005). After discovery of the antifungal functions of
phenylpropanoid pathway-derived compounds in plant defense, early studies were conducted
in vitro to evaluate the role of these natural products in specific plant diseases. However, in
Chapter 5
134
cases where natural products showed antifungal activity in vitro, researchers are now
questioning the conclusion that these natural products have antifungal activity in vivo. Direct
evidence to support the phytoalexin hypothesis in vivo is lacking (Lozovaya et al., 2007).
Genetic engineering offers new possibilities to study the role of phytoalexins in plant defense
in vivo. These studies can be categorized into three classes: genetic modification of the
pathogen to disrupt the mechanisms which enable it to detoxify phytoalexins genetic
manipulation of the host to enhance or reduce levels of a specific natural product; and
introduction of exogenous genes which could produce a novel antimicrobial compound in the
host (Dixon et al., 2002).
To the best of our knowledge, all the studies performed with the goal of evaluating the
activity of existing isoflavones of red clover to particular pathogens have been carried out in
vitro. This work represents the first in vivo attempt to evaluate a possible significant role of
isoflavones in disease resistance via genetic manipulation of red clover. For this purpose,
among genes involved in the isoflavone pathway, IFS was chosen for suppression due to its
vital function in production of isoflavones. RNAi-silenced transgenic lines were subsequently
exposed to Sclerotinia trifoliorum and Sclerotium rolfsii, which are important fungal
pathogens of red clover.
5.2. Materials and methods
5.2.1. Plant material
A highly regenerable red clover genotype (NEWRC68) derived from a population of NEWRC
germplasm was used as the source of explants for gene transformation via Agrobacterium-
mediation. The seeds were sown in Among the obtained transgenic plants bearing the IFS-
RNAi gene construct, five lines (RNAi_1, RNAi_2, RNAi_3, RNAi_4 and RANi_5) which
derived from independent events plus a wild type control were chosen for further steps. All
chosen lines (transgenic lines and the wild type control) were propagated in vitro. Then in
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
135
vitro-raised plants (transgenic lines and the wild type control) were transferred to pots
containing a mixture of sand, expanded clay and soil (1:1:1) and were kept under greenhouse
conditions at 25 °C with 16 h per day illumination of 400 μmol m-2
sec-1
. All the leaf samples
were collected for the plants which were acclimated in greenhouse for two months.
5.2.2. Fungal isolates
Sclerotia of the Sclerotium rolfsii isolate used in the inoculation test were kindly provided by
R. G. Pratt (USDA, ARS, MS). Sclerotia of the Sclerotinia trifoliorum were collected by Tim
Vleugels from the red clover field experimental sites at the Institute for Agriculture and
Fisheries Research (ILVO), Merelbeke, Belgium.
5.2.3. General procedure used in pathogenicity assays
Assays for fungal pathogenicity were conducted on detached leaves of selected lines
including transgenic and wild type. First, the sclerotia of Sclerotinia trifoliorum or Sclerotium
rolfsii were transferred to the center of a potato dextrose agar (PDA) petri dish and cultured at
28 or 15 ºC, respectively, for 5 days. On the day of infection, young healthy leaves were
collected and placed on petri dishes containing 20 ml of agar. The plugs of fungal mycelium
were then removed with a sterile cork borer from the actively growing margin of the colony.
Sterile plugs were also taken from uninoculated Petri dishes containing 20 ml of PDA to use
for mock inoculation. As adhesive agent, 15µL of 10% low melting point agarose (Invitrogen)
was used to ensure full contact between leaves and plugs. A 5-mm plug (containing mycelium
or sterile PDA) was placed on the tip of the middle leaflet of each leaf. On the third day of
infection, the plates of both pathogenicity tests (Sclerotinia trifoliorum and Sclerotium rolfsii)
were sealed with parafilm and incubated in the growth chambers with a relative humidity of
95% at 28 and 15 ºC respectively, with a 16/8 h photoperiod.
Chapter 5
136
5.2.4. Identification of proper reference genes and determination of suitable
sampling position for wild type plants via expression study using qPCR
In total, nine wild type red clover plants were chosen for the pathogenicity assay for each
fungus (Sclerotinia trifoliorum and Sclerotium rolfsii). From each plant, three leaves were
collected as biological replications. Inoculated leaves were incubated for three days. Three
types of sampling were performed on each leaf: (1) entire infected (brown) part of middle
leaflet; (2) uninfected, green part of middle leaflet ≤1 cm away from the infection front; and
(3) the proximal 1 cm of the uninfected, green part of the right leaflet. Samples were
immediately frozen in liquid nitrogen and stored at −80°C. The expression profiles of
candidate reference genes were investigated and the most stable ones were chosen. Then the
expression level of IFS were determined for the three sampling types in both fungi.
5.2.5. Determination of suitable reference genes and assessment of IFS
expression level in transgenic lines
Three leaves each from transgenic lines (RNAi_1, RNAi_2, RNAi_3, RNAi_4, RNAi_5) and
a wild type line (WT) were collected in the way that each leaf sampled was from a separate
plant which was a clone of the corresponding line. All the leaves were placed on petri dishes
containing agar. Plugs of fungal mycelium as well as plugs of sterile PDA for mock-
inoculated controls were placed on the leaves, which were incubated for three days at a
suitable temperature. The uninfected, green part of middle leaflet ≤1 cm away from the
infection front of the leaves was sampled. In case of mock-inoculated controls, samples were
collected from ≤1 cm away from the plugs. Then samples were quickly frozen in liquid
nitrogen and stored at −80°C. The expression levels of candidate reference genes were studied
and their stabilities were calculated. The expression levels of IFS were studied in the leaf
samples. These procedures were carried out for both fungi separately.
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
137
5.2.6. Collection of mycelia samples for qPCR
Both fungi were cultured on PDA for 3 days. The actively growing mycelia were scraped
from the surface of PDA using a sterile scalpel. Mycelia were immediately frozen in liquid
nitrogen and stored at −80 C for RNA extraction.
5.2.7. Total RNA isolation and cDNA preparation
Frozen leaves were ground with a Retsch Tissuelyser (Qiagen). The freeze-dried mycelia
were ground to a fine powder using a mortar and pestle. We used the RNeasy Plant Mini Kit
(Qiagen), which provided the same protocol for extraction of total RNA from plants and
fungi. After extraction of total RNA of leaves and mycelia, the residual DNA in the extracts
was destroyed with TURBOTM DNase (Ambion). Subsequently, 50 ng of total RNA (50 ng)
of each sample was reverse transcribed into cDNA using the Superscript VILO cDNA
synthesis kit (Invitrogen), and then the cDNA was diluted 1:10 for qRT-PCR (see Chapter 3).
5.2.8. qRT-PCR conditions
Quantitative real-time PCRs were performed in 384-well plates using the LightCycler 480
(Roche). The reaction mixtures and thermal profile were the same as described in Chapter 3.
Also, qRT-PCR assays were carried out on equivalent amounts of total RNA without reverse
transcription (as template) for each sample to verify the absence of genomic DNA
contamination in the extracted total RNA.
5.2.9. Analysis of qPCR results
The analysis of expression data obtained from qPCR was carried out for determination of
appropriate reference genes for each fungus using the method suggested by Mehdi Khanlou
and Van Bockstaele (Chapter 3 above).
5.2.10. Pathogenicity assays of S. trifoliorum and S. rolfsii
Pathogenicity assays of both fungi were performed according to the above-mentioned general
procedure (see 5.2.3). After three days of infection, the inoculated and mock-inoculated
Chapter 5
138
leaves were photographed with a digital camera and pictures were scanned using the image-
analysis software Assess® (American Phytopathological Society, Saint Paul, USA) to estimate
the infected area. The severity of disease in each picture was determined by dividing number
of "diseased pixels" by number of "diseased pixels"+"healthy pixels", followed by multiple
100. The pathogenicity experiment for each studied fungus was conducted three times.
5.2.11. Statistical analysis
The infected areas as measured by Assess were analyzed using one-way analysis of variance
(ANOVA) with unbalanced completely randomized design. ANOVA was carried out using
the data recorded from each of the three independent infection assays for each fungus. Each
leaf was considered as a replicate of each treatment. Orthogonal contrasts were used to
compare transgenic line and wild type clones. Duncan's multiple range tests were used to
compare means in experiments. All statistical analyses were performed using SAS 9.2 (SAS
Institute Inc., Cary, NC, USA).
5.3. Results
5.3.1. Identification of proper reference genes for S. trifoliorum and S. rolfsii
in wild type plants
To determine the most appropriate reference genes in both fungi, the stability indexes of all
candidate reference genes were calculated across the three different samples (groups): (1)
entire infected (brown) part in the tip of infected the middle leaflet (2) uninfected, green part
of middle leaflet ≤1 cm away from infected front; (3) proximal 1 cm of uninfected, green part
of right leaflet. The calculation of stability indexes for each candidate reference gene was
performed using the methods we proposed to identification of the best-performing genes for
accurate normalization in red clover tissues (Chapter 3). The candidate reference genes EIF-
4a (0.471), UBC2 (1.109) and GAPDH (2.125) for (Table 5.1) and the candidate reference
genes UBQ10 (2.391), EIF-4a (2.422), GAPDH (2.560) were found to be the proper reference
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
139
genes to determine amount of IFS in the study samples (Table 5.2). The candidate reference
genes SAND (66.745) in S. trifoliorum and ACTIN (54.569) in S. rolfsii showed the lowest
stability in our assays.
Table 5.1: Stability of candidate reference genes based on analysis of variance for S. trifoliorum-
infected wild type red clover.
Genes
Source of variation
MS (Variance) components Total MS
(Variance) -1/𝑿
VB 1
VW 2
Stability index for S.
trifoliorum 1 2 3
ACTIN Between Groups 12.192 2.220 4.006 18.418
0.821 22.433 0.1096 2.459 Within Groups 0.064 0.025 0.001 0.090
EF-1α Between Groups 14.359 3.489 3.691 21.539
0.588 36.584
0.4871 17.710 Within Groups 0.194 0.022 0.069 0.285
EIF-4a Between Groups 0.069 1.023 1.622 2.714
0.954 2.843
0.166 0.471 Within Groups 0.007 0.116 0.035 0.158
GAPDH Between Groups
1.585 1.843 0.009 3.437 0.808 4.251
0.499 2.125
Within Groups 0.052 0.044 0.308 0.404
SAND Between Groups 21.845 4.375 6.667 32.887 0.565 58.201 1.146 66.745
Within Groups 0.606 0.031 0.011 0.648
UBC2
Between Groups 9.282 2.203 2.436 13.921 0.826 16.860 0.065 1.109
Within Groups 0.024 0.011 0.018 0.053
UBQ10
Between Groups 7.313 2.013 1.651 10.977 0.681 16.117 0.232 3.746
Within Groups 0.014 0.039 0.105 0.158
YLS8
Between Groups 9.315 2.232 2.427 13.974 0.759 18.393 0.142 2.614
Within Groups 0.048 0.037 0.023 0.108
Chapter 5
140
Table 5.2: Stability of candidate reference genes based on analysis of variance for S. rolfsii-infected
wild type red clover
5.3.2. Identification of suitable position of sampling with qPCR
After identification of suitable reference genes across the above-mentioned samples, we
studied the expression of IFS in the wild type samples. As illustrated in Figure 5.1, the
expression level of IFS in infected tissues caused by both fungi was the lowest compared with
other sampled parts of leaves. A large increase of IFS next to the infection front revealed that
the IFS expression was induced by both fungi with a similar amount of expression. This
confirmed that the observed IFS expression was due to the IFS of red clover plant and that
studied fungi do not express IFS by themselves. We investigated the possible expression of
IFS in the both fungi in the way that, along with red clover leaf samples, the expression of IFS
was measured in cDNA which made from total RNA of both studied fungi. We observed no
IFS expression in either fungus.
Genes
Source of variation
MS (Variance) components Total MS
(Variance) -1/𝑿
VB 1
VW 2
Stability index
S. rolfsii 1 2 3
ACTIN Between Groups 19.630 3.425 6.655 29.711
0.560 53.053 1.028 54.569 Within Groups 0.532 0.020 0.024 0.576
EF-1α Between Groups 16.971 4.690 3.817 25.478
0.646 39.440
0.813 32.052 Within Groups 0.497 0.017 0.011 0.525
EIF-4a Between Groups 3.630 2.227 0.170 6.027
0.898 6.711
0.361 2.422 Within Groups 0.274 0.040 0.010 0.324
GAPDH Between Groups 0.810 3.515 0.949 5.274 0.762 6.921
0.370 2.560
Within Groups 0.256 0.016 0.010 0.282
SAND Between Groups 19.741 3.470 6.658 29.869 0.579 51.587 0.854 44.103
Within Groups 0.466 0.018 0.011 0.495
UBC2
Between Groups 11.511 2.373 3.430 17.314 0.776 22.311 0.349 7.795
Within Groups 0.236 0.025 0.009 0.270
UBQ10 Between Groups 7.313 2.013 0.665 6.772 1.002 6.758 0.353 2.391
Within Groups 0.014 0.039 0.021 0.351
YLS8
Between Groups 5.606 1.612 1.205 8.423 0.925 9.105 0.369 3.356
Within Groups 0.313 0.025 0.003 0.341
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
141
Figure 5.1: The relative expression level of IFS was shown in infected part of middle leaflet (1),
uninfected part of middle leaflet till one cm away from infected front (2) and one cm from bottom to
tip of right leaflet (3) when wild type red clover leaves were exposed to S. trifoliorum (I) and S. rolfsii
(II). The leaves of nine wild type red clover plants (three leaves from each plant) were collected for
each pathogen. IFS expression was not detected in both fungi. Standard deviations are marked in
columns by black lines.
5.3.3. Determination of appropriate reference genes for S. trifoliorum and S.
rolfsii infected transgenic lines
The suitable reference genes were identified as outlined in Chapter 3 above. The stability
indexes of all reference genes were calculated for the transgenic lines infected with each of
the fungi. Visual comparison of estimated stability indexes revealed that the reference genes
UBQ10 (0.001), YLS8 (0.047) and EIF-4a (0.054) were suitable for the pathogenicity assay
performed on leaves of red clover using S. trifoliorum (Table 5.3). Also, as mentioned in
Table 5.4, the most suitable reference genes for use for the normalization of data in
quantitative real-time RT-qPCR assays conducted on leaves infected with S. rolfsii were EIF-
4a (0.015), YLS8 (0.019) and UBQ10 (0.025). As recommended in a previous study
(Vandesompele et al., 2002), we selected the three most stable internal control genes for
calculation of an RT-PCR normalization factor. These selected reference genes were
statistically proven not to be regulated by infection with either fungus and therefore are fit for
a reliable normalization procedure to quantify the transcript levels of IFS in transgenic lines.
Chapter 5
142
Table 5.3: Stability of candidate reference genes based on analysis of variance for
S. trifoliorum-infected transgenic red clover.
Table 5.4: Stability of candidate reference genes based on analysis of variance for
S. rolfsii-infected transgenic type red clover.
Genes
Source of variation
MS (Variance)
components Total MS
(Variance) -1/𝑿
VB 1
VW 2
Stability index
for S. rolfsii Infected Mock
ACTIN
Between Groups 8.119 8.119 16.238 1.238 16.18 0.19 3.216
Within Groups 0.183 0.015 0.199
EF-1α
Between Groups 1.344 1.344 2.688 1.176 2.28 0.10 0.233
Within Groups 0.063 0.056 0.120
EIF-4a
Between Groups 0.276 0.276 0.552 1.955 0.28 0.05 0.015
Within Groups 0.080 0.028 0.108
GAPDH
Between Groups 6.303 6.303 12.607 1.024 12.30 0.15 1.949
Within Groups 0.044 0.117 0.162
SAND Between Groups 5.728 5.728 3.450
1.765 1.95 0.05 0.108
Within Groups 0.090 0.007 0.097
UBC2 Between Groups 2.867 2.867 5.735
1.452 3.948 0.03 0.156
Within Groups 0.046 0.011 0.057
UBQ10 Between Groups 0.454 0.454 0.908
1.713 0.53 0.04 0.025
Within Groups 0.033 0.049 0.082
YLS8 Between Groups 0.847 0.847 1.695
1.801 0.94 0.02 0.019
Within Groups 0.017 0.019 0.037
Genes
Source of variation
MS (Variance)
components Total MS
(Variance) -1/𝑿
VB 1
VW 2
Stability index
for S. trifolorum Infected Mock
ACTIN
Between Groups 5.947 5.947 11.894 0.892 13.32 0.27 3.598
Within Groups 0.140 0.100 0.240
EF-1α
Between Groups 1.201 1.201 2.403 1.361 1.76 0.14 0.253
Within Groups 0.124 0.070 0.195
EIF-4a
Between Groups 0.541 0.541 1.083 1.009 0.61 0.08 0.054
Within Groups 0.110 0.044 0.154
GAPDH
Between Groups 3.142 3.142 6.285 0.898 6.99 0.24 1.746
Within Groups 0.062 0.162 0.224
SAND Between Groups 5.728 5.728 11.456
0.945 12.12 0.18 2.298
Within Groups 0.145 0.033 0.179
UBC2 Between Groups 1.016 1.016 2.033
1.514 2.03 0.08 0.072
Within Groups 0.036 0.045 0.081
UBQ10 Between Groups 0.017 0.017 0.035
1.825 0.01 0.05 0.001
Within Groups 0.043 0.053 0.096
YLS8 Between Groups 0.513 0.513 1.026
1.681 0.61 0.07 0.047
Within Groups 0.036 0.093 0.129
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
143
5.3.4. Evaluation of IFS expression in RNAi-silenced transgenic and wild
type plants
After the proper reference genes were identified, the expression levels of IFS were studied in
transgenic lines. Figure 5.2 illustrates the expression of IFS in different transgenic lines and
wild types following infection with both fungi.
Figure 5.2: The relative expression level of IFS in infected (I) and mock-inoculated (M) leaf samples
of transgenic lines (RNAi_1, RNAi_2, RNAi_3, RNAi_4 and RNAi_5) and wild type (WT) when
their leaves were exposed to S. trifoliorum (I) and S. rolfsii (II). Standard deviations are marked in
columns by black lines.
The average expression level of IFS in infected wild type was more than 100-fold higher
compared to expression in infected transgenic lines in both fungi. There was no considerable
differences between infected and mock-inoculated transgenic lines in the expression of IFS.
These results proved that the RNAi construct worked to down-regulate the expression of IFS
in transgenic lines.
5.3.5. Pathogenicity tests
We carried out pathogenicity assays on RNAi-silenced lines (RNAi_1, RNAi_2, RNAi_3,
RNAi_4 and RNAi_5) and wild type (WT) for both studied fungi using the detached leaf
technique (Figure 5.3).
Table 5.5: The result of ANOVA and orthogonal contrasts of three pathogenicity assays with S.
trifoliorum on detached leaves collected from transgenic lines (RNAi_1, RNAi_2, RNAi_3, RNAi_4
and RNAi_5) and wild type (WT).
Type of Analysis Number and type of lines studied
Assay 1 Assay 2 Assay 3
F value P value F value P value F value P value
ANOVA All lines (five transgenic lines and a wild type line) 2.52 0.0 327 2.38 0.0 418 2.71 0.0 226
Orthogonal contrast Five transgenic lines VS a wild type line 4.89 0.0 287 5.40 0.0 217 4.98 0.0 272
Chapter 5
144
Three experiments were performed for each fungus, and data were subjected to a one-way
ANOVA to identify statistically significant differences among studied lines. Then orthogonal
contrasts were used to compare transgenic lines and wild type. The ANOVA result showed
significant mean differences in studied lines in all assays, and also orthogonal contrast tests
indicated that differences between transgenic lines and wild type were significant (p < 0.05)
for S. trifoliorum (Table 5.5). The statistical analysis using ANOVA presented the existence
of highly significant differences (p < 0.001) in investigated lines in the three assays.
Furthermore a highly significant difference was seen in orthogonal comparison of transgenic
lines and wild type in three assays (p < 0.001) for S. rolfsii.
Table 5.6: The result of ANOVA and orthogonal contrasts of three pathogenicity assays with S. rolfsii
on detached leaves collected from transgenic lines (RNAi_1, RNAi_2, RNAi_3, RNAi_4 and
RNAi_5) and wild type (WT).
We also illustrated the average percentage of severity for both pathogens with histograms
(Figure 5.4). We used the pooled data of three assays which were conducted on RNAi lines
and wild type. The figure confirms the ANOVA and orthogonal contrasts result.
(a) (b) (e) (f)
(c) (d) (g) (h)
Figure 5.3: Pathogenicity assays were conducted on RNAi-silenced lines (RNAi_1, RNAi_2,
RNAi_3, RNAi_4 and RNAi_5) and wild type (WT) using the detached leaf technique. On the third
day of infection, the leaves were photographed with a digital camera and then pictures were scanned
using image-analysis software to estimate the infected area. The figure presents the leaves of mock-
inoculated wild type (a), inoculated wild type (b), mock-inoculated RNAi-silenced line (c) and
inoculated RNAi-silenced line (d) which were exposed to S. trifoliorum. Also it illustrates the leaves
of mock-inoculated wild type (e), inoculated wild type (f), mock-inoculated RNAi-silenced line (g)
and inoculated RNAi-silenced line (h) which were exposed to S. rolfsii.
Type of Analysis Number and type of lines studied
Assay 1 Assay 2 Assay 3
F value P value F value P value F value P value
ANOVA All lines (five transgenic lines and a wild type line) 17.45 < 0.0001 17.00 < 0.0001 13.83 < 0.0001
Orthogonal contrast Five transgenic lines VS a wild type line 64.01 < 0.0001 68.67 < 0.0001 47.31 < 0.0001
Figure 5.4: The percentage of severity of the RNAi lines ( RNAi_1, RNAi_2, RNAi_3, RNAi_4, RNAi_5) and wild type (WT), infected by Sclerotinia
trifoliorum (I) and S. rolfsii (II) using a detached leaf screening technique. The severity was determined by dividing number of "diseased pixels" by number
of "diseased pixels"+"healthy pixels", followed by multiple 100. Each graph represents the pool data of three assays which were conducted for each pathogen.
Number of investigated samples when the samples of three assays were pooled for each pathogen were RNAi_1= 63, RNAi_2=52, RNAi_3=48, RNAi_4=49, RNAi_4=41 and WT=50 for Sclerotinia trifoliorum. Number of pooled samples for S. rolfsii were RNAi_1= 73, RNAi_2=46, RNAi_3=58, RNAi_4=65,
RNAi_4=58 and WT=63. Columns sharing the same lowercase letters are not significantly different by Duncan's multiple range tests. Standard deviations are
marked in columns by black lines.
Chapter 5
146
5.4 Discussion
Red clover is widely used as forage for animal grazing and pastures. Red clover has the ability
to grow in different environmental conditions across wide geographical areas. Because it is an
abundant source of isoflavonoids, red clover has recently received considerable interest.
Therefore it may be an inexpensive and valuable resource for many applications. Isoflavones
are the most abundant of the sub- classes of isoflavonoids in red clover. The main isoflavones
are formononetin and Biochanin A, but red clover also contains daidzein and genistein as well
as their glycosidic conjugates. Several studies have been conducted to evaluate the anti-fungal
activity of isoflavonoids in red clover. The first attempt was conducted by Virtanen et al.
(1957) who reported antifungal activity of red clover extract. Due to the significant progress
that has been made in techniques for the separation, identification and characterization of
these substances, it was discovered that the antifungal compounds which exist in red clover
extracts are isoflavonoids. Moreover, investigation of the biosynthetic pathway of
isoflavonoids revealed that the isoflavones are precursors for the defense related pterocarpan
phytoalexins, which are another sub-class of isoflavonoids in red clover (Ebel, 1986). More
studies have been conducted to apply the red clover-extracted isoflavonoids to various fungi.
The isoflavones biochanin A and genistein were found to exhibit antifungal activity on
Rhizoctonia solani (Bredenberg, 1961). In another study, the red clover isoflavones tested
were effective against the phytopathogenic fungus Drechslera teres and could inhibit growth
of the low temperature fungus Microdochium nivale (Saviranta et al., 2008). However, all
abovementioned assays were performed in vitro. This is the first attempt to evaluate
isoflavones’ antifungal activity in vivo through genetic engineering of red clover. Given that
the sub-classes of isoflavonoids including isoflavones and pterocarpans demonstrated
antifungal activity, choosing the right genes to be subjected to genetic engineering to generate
antifungal-free red clover is critical. According to the current knowledge of relevant structural
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
147
genes involved in the isoflavonoid pathway, there are at least two approaches for this purpose.
First, using a gene silencing approach to knock down IFS, which is a branch point enzyme to
the isoflavonoid biosynthetic pathway; and second, the combination of overexpression of the
CRC transcription factor and gene silencing of chalcone reductase enzyme to simultaneously
reduce both the isoflavones genistein and daidzein. As the latter approach faces the technical
difficulty of multiple gene transformation, we chose the first one for our study. In previous
chapters, we explained that the RNAi vector for IFS was made using the gateway cloning
system, and transformation of red clover was performed with Agrobacterium harboring the
IFS-RNAi gene construct. Healthy looking transgenic lines were transferred to the
greenhouse. We hypothesized that if the IFS expression in wild type plants remained the same
before and after inoculation with the studied fungi, this would show that de novo isoflavonoid
synthesis is not part of defense system of red clover when faced with the studied fungi.
Therefore we needed to confirm that the expression of IFS is influenced by the studied fungi
before conducting any pathogenicity assays on transgenic lines. For this purpose, an
expression assay needed to be performed to evaluate expression profile of IFS when lines are
infected. Note that although wild type red clovers contain both constitutive and induced
isoflavone concentrations, we assumed that the amount of constitutive isoflavone in
transgenic red clover leaves is too low to be detectable. As mentioned above, in a two-step
hydrolysis, any conjugated forms present in red clover leaves are converted to the free
aglycones (isoflavones) following a pathogen’s attack (Graham et al., 1990). Therefore it is
important that the RNAi-silenced transgenic lines have no constitutive isoflavone which could
cause resistance by production of isoflavone in IFS-RNAi silenced red clover by hydrolysis
(Subramanian et al., 2005). IFS enzyme has no role in production of isoflavones obtained
from hydrolysis of the conjugated forms. Therefore, in our expression study, we measured
only induced capacity for de novo isoflavone synthesis in response to pathogen attack.
Chapter 5
148
Prior to the evaluation of expression of IFS, we need to address two questions; 1) which
reference genes are suitable for this evaluation and 2) where should they be sampled.
Regarding the appropriate reference genes, to our knowledge, there is no report of evaluation
of stability of candidate reference genes in red clover following inoculation with our
investigated fungi. Based on our review of literature, in two existing expression studies
performed in red clover, only ACT or EIF-4a have been used as internal controls in non-
treated leaf samples (Kim et al. 2003; Sullivan 2009b). However, none has been statistically
evaluated for their expression stability. In this study, the determination of the best reference
genes was performed as outlined in Chapter 3. The assay for identification of proper position
of sampling revealed that both fungi influenced the expression level of IFS in wild type red
clover. This result is in accordance with another report that showed high expression of
isoflavones and their conjugates when soybean leaves were inoculated with Sclerotinia
sclerotiorum (Wegulo et al., 2005). Also the result indicated that due to the death of most
cells in the infected part of the leaflet, the expression of IFS cannot be accurately detected
there. Therefore, samples should be collected from the uninfected part of the same leaflet ≤1
cm away from infected front. This result was in agreement with the previous report that genus
Aspergillus caused intense elicitation of phytoalexin synthesis close to the infected area of a
peanut kernel under suitable conditions to both the host and the fungus (Sobolev, 2008).
In the next step of this study, we determined proper reference genes for transgenic lines.
These suitable reference genes were then used for evaluation of expression of IFS in
transgenic lines to test whether elicitation caused by fungi overwhelms the effect of the
silencing construct. It would be possible that the strength of the IFS promoter itself or
elicitation by fungi during infection causes such high transcriptional activity that it might
overwhelm the RNAi. If that happened, it would be likely that transgenic and wild type lines
would not exhibit much or any difference in resistance to the fungi. Normally the isoflavones
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
149
are present at relatively low levels in healthy wild type tissues (Subramanian et al., 2005).
However, our infection assays with both fungi revealed that the expression of IFS in
transgenic lines was much lower (100-fold less) even during elicitation compared to wild
types. The expression study of our mock-inoculated controls revealed that the used agar plugs
did not influence IFS expression caused by the pathogens. Our result confirmed the efficiency
of using the post-transcriptional gene silencing technique RNAi in red clover. The RNAi was
previously found to be effective in silencing the endogenous red clover polyphenol oxidase
(PPO) gene, where more than 90% reduction was achieved in PPO enzyme activity (Sullivan
and Hatfield, 2006). Also, IFS has already been subjected to RNAi technique in soybean
roots, leading to a nearly 90% reduction in total isoflavone pools (Subramanian et al., 2005).
Moreover the application of RNAi technique in Lotus corniculatus L., which is a forage
legume, caused significant reduction in the amount of isoflavonoids in hairy root cultures
(Colliver et al., 1997). These successes show the usefulness of RNAi technology to improve
crops.
The result obtained from evaluation of IFS expression in transgenic lines gave us enough
confidence to use these lines in pathogenicity assays. To analyze the results, we used a single
degree-of-freedom orthogonal contrast, which is a powerful means of perfectly partitioning
the ANOVA model to gain greater insight into our data. The difference in the orthogonal
contrasts of wild type line versus transgenic lines were found to be significant (p < 0.05) for
S. trifoliorum, whereas the orthogonal comparison of wild type line against transgenic in S.
rolfsii demonstrated highly significant differences (p < 0.001). We believe that since S.
trifoliorum is able to detoxify the isoflavones, it can grow faster than S. rolfsii in the wild type
line. Thus the differences between wild type line and transgenic lines in S. trifoliorum are
smaller than those in S. rolfsii. Previous study also showed Nectria haematococca detoxified
the pterocarpans medicarpin and maackiain, which are isoflavonoid end-products (Wu and
Chapter 5
150
Vanetten, 2004). Interestingly, our result also indicated that although S. trifoliorum can
detoxify the isoflavones, the speed of spreading infection in leaves of the wild type line is
slower than those of transgenic lines. This finding led us to the conclusion that the present of
isoflavones in leaves which are infected even with the fungus with detoxification ability can
retard spreading of the fungus. It seems that the time and energy used for the detoxification
process decelerate the pathogen’s speed of infection. The highly significant difference which
was observed in the S. rolfsii assay showed us the importance of isoflavones in the defense
system of red clover against S. rolfsii. This finding should be taken into account by breeders
who wish to develop new resistant varieties against S. rolfsii.
The pathogenicity assays were performed using unbalanced completely randomized design
because the numbers of new and healthy leaves available in clones of each line were different
during each harvesting time. The unbalanced ANOVA test enabled us to analyze the variance
of groups with unequal sample sizes without losing test power.
Our result demonstrated that in both fungal assays, the transgenic lines which possessed low
levels of isoflavone synthase were susceptible. However the difference in disease
susceptibility between IFS-RNAi transgenic plants and wild types was significant (p < 0.05)
in Sclerotinia trifoliorum and highly significant (p < 0.001) in Sclerotium rolfsii. The active
result can be due to the ability of Sclerotinia trifoliorum in detoxification of isoflavones and
pterocarpans. Our result was in agreement with the previous report that the soybean cultivars
which were low in isoflavones were susceptible to S. sclerotiorum in detached leaf assays
(Wegulo et al., 1998). Our result is also consistent with the previous study carried out by Wu
and VanEtten (2004) who reported that the reduction of isoflavonoid accumulation in cultured
root tissues of pea (Pisum sativum L.) using the RNAi technique resulted in enhanced
susceptibility to the fungus Nectria haematococca. Also RNAi silencing of IFS in soybean
Glycine max Merr. led to enhanced susceptibility to the fungus Phytophthora sojae
Evaluation of the role of isoflavones in resistance of red clover to S. trifoliorum and S. rolfsii
151
(Subramanian et al., 2005). Our results added to the evidence from pea and soybean that
RNAi is a reliable technique in genetic engineering which can open up new possibilities for
studying the role of genes in plants. We showed that wild type lines are more resistant to two
fungal pathogens leading to the conclusion that isoflavones are involved in the defense
response of red clover. Although there is a high variation in the amount of isoflavones in red
clover (Ramos et al., 2012), higher amounts of isoflavones do not always mean higher
resistance. It was reported that the isoflavone concentration in red clover before or after
infection did not correlate with their Sclerotinia resistance. As a consequence, isoflavone
concentration cannot be used as an index of disease resistance (Debnam and Smith, 1976).
The stated that the studied resistant cultivars consistently accumulated more of the
pterocarpans maackiain and medicarpin when they inoculated with Sclerotinia. They
concluded that the pterocarpans medicarpin and maackiain were clearly the most important
components in the development of antifungal activity after infection. In fact, two different
cultivars with the same amount of isoflavones could show different resistance. In soybean,
although a susceptible variety (‘Williams’) and a resistant one (‘Williams 82’) have high
levels of the preformed conjugates of daidzein, analytical analysis showed that only the
resistant variety used its conjugated isoflavones when exposed to a pathogen (Subramanian et
al., 2005). Therefore breeders should look for red clover varieties with high concentrations of
the pterocarpans medicarpin and maackiain phytoalexins and most importantly, varieties
which have the ability to use those pterocarpans when faced with attack by a pathogen.
Genetic engineering of red clover is an alternative way to enhance the amount of medicarpin
and maackiain leading to resistance to S.trifoliorum and S. rolfsii.
Chapter 6: General discussion & future
perspectives
General discussion & future perspectives
155
6.1. Introduction
Plant diseases threaten the productivity of world agriculture. Most agricultural and
horticultural crop species suffer from significant yield losses due to pathogen attack. More
than 70% of crop diseases are caused by fungi (Wani et al., 2010). In the past, there have been
two commercial approaches to avoid fungal diseases: 1) the application of fungicides and 2)
developing new resistant cultivars through traditional breeding methods. Although the
application of chemicals is still an effective way to control plant diseases, there is increasing
concern that chemical residues in soil and water may be harmful to human health and the
environment (Manczinger et al., 2002). The conventional breeding methods have been
successful to release resistant cultivars. However, these methods will soon reach their limits.
New alternative methods to improve crops are clearly needed. Advances in genetic
engineering technologies have broadened the breeding possibilities. These recently available
tools enable us to overcome the limitations of traditional methods in disease resistance
breeding. Along with the innovation of modern molecular techniques, the knowledge about
plant defense mechanisms have been also increased, giving us ways us to achieve higher
levels of resistance in crops. In red clover, one of the key goals of most breeding programs is
increased persistence because of its association with yield. Although the life of red clover is
basically controlled by programmed senescence (Taylor, 2008), fungal diseases could
contribute to the lack of persistence of red clover (Pokorny et al., 2003). A large body of
literature reports that secondary metabolite isoflavones have critical roles in plant-antifungal
activity in planta. The exploration of the possibility of strengthening plants by the plants’ own
defense arsenals could be considered a potent approach for improving red clover protection.
Red clover is an abundant source of isoflavones (Saviranta et al., 2008). In this study, we
decided to evaluate possible significant role of isoflavones in resistance against Sclerotinia
trifoliorum and Sclerotium rolfsii by altering the isoflavone synthase gene using the new
Chapter 6
156
molecular techniques. We would like to emphasize that most previous studies which reported
antifungal activity of isoflavones were based on the in vitro evaluation of mycelial growth of
different fungi on media contained red clover isoflavones in petri dishes. Our work is the first
attempt to investigate by genetic engineering the antifungal activity of isoflavones in the red
clover plants that are exposed to fungi. Since all in vitro evaluation of isoflavones confirmed the
antifungal activity of isoflavones, we also expected to observe antifungal activity in our work.
However, the main goal of this study was to address whether the isoflavones’ antifungal activity is
statistically significant in planta so as to be recommendable for commercial red clover breeding.
To achieve this goal, we generated isoflavone-null red clover using gene silencing and plant
transformation techniques. Then the collected leaves of IFS-RNAi transgenic red clover and
parental wild type, serving as control, were exposed to the two soil-borne fungal pathogens
Sclerotinia trifoliorum and Sclerotium rolfsii in pathogenicity assays using a detached leaf
method.
6.2. Regeneration and transformation
Limited genetic variation of red clover germplasm has hindered progress in traditional
breeding for disease resistance. Genetic engineering can be used to overcome traditional
breeding method limitations in red clover improvement. In fact, using genetic engineering
enables us to elucidate the function of the genes which are involved in plant defense. The
genes which contribute to resistance can then be used to develop new resistant cultivars. The
prerequisites for genetic engineering are efficient plant regeneration and efficient
transformation. In previous studies, B5 medium was found a suitable for NEWRC germplasm
for tissue culture work (Quesenberry and Smith, 1993; Sullivan and Hatfield, 2006).
However, the red clover plants used as resource for explants were kept in a greenhouse. In our
study, to avoid the deleterious effects of surface-sterilization of explants, we decided to keep
the plants used as resource for explants in sterile condition. After four months of living in B5
General discussion & future perspectives
157
medium, however, the plants looked unhealthy. We then sought optimal in vitro growth
conditions for red clover. The results indicated that L2 medium was suitable for long-term
maintenance of red clover plants in vitro. Although there were some reports that the gelling
agent Gelrite caused poor rooting of the legume mung bean (Arthur et al., 2004) and reduced
root length of cultured tobacco shoots (Ozel et al., 2008), we did not observed Gelrite causing
any negative effect on growth of red clover. In order to establish an efficient regeneration
system in red clover, highly regenerative genotypes were screened and utilized for evaluation
of different media sequences for their suitability to utilize potentiality of regeneration
capacity. Media sequence KBC-SPL-SPL significantly increased the efficiency of somatic
embryogenesis of red clover. There was no relationship between regeneration ability and fresh
callus weight. Also, no interaction was observed between genotypes and media sequences for
fresh callus weight, number of embryos and number of developed shoots. No significant
interaction was observed between media sequences and genotypes for shoot development.
This result was in agreement with the previous report that interaction of media and number of
shoots in several genotypes of red clover did not have any significant differences (Carrillo et
al., 2004). To develop an efficient gene transfer system, some parameters were investigated,
including evaluation of suitable concentration of selection agent, examination of the effect of
pre-culturing of explants on efficiency of transformation, and comparison of different
Agrobacterium strains for efficiency of gene transformation. Based on the selection agent test,
75 mg/l of kanamycin sulfate was an appropriate concentration for transformant selection.
Our result was contrary to that of a previous report in that total growth was inhibited at 25 mg
L-1
of kanamycin (Quesenberry et al., 1996). The discrepancy between results is because the
Gelrite reduces the efficiency of kanamycin. Pre-culture of explants in callus induction
medium prior to inoculation has been reported to considerably increase the efficiency of gene
transformation in plant species (Birch, 1997). Pre-culturing explants only slightly (not
Chapter 6
158
significantly) increased the frequency of gene transfer. In order to identify the Agrobacterium
strain with higher efficiency for red clover transformation, Agrobacterium EHA101 and
C58C1 strains, which both harbored the pTJK136 binary vector, were compared. Our result
showed the superiority of C58C1 over EHA101 for transformation of red clover. Highly
infective Agrobacterium strain C58C1 can increase the transformation efficiency for red
clover to 83%.
Several reports of successful use of the C58C1 strain in legumes have been published.
Examples include tepary bean (Phaseolus acutifolius A. Gray) (De Clercq et al., 2002), pea
(Pisum sativum L.) (Nadolska-Orczyk and Orczyk, 2000), and peanut (Arachis hypogaea L.)
(Sharma and Anjaiah, 2000). PCR and histochemical GUS assay of transformed plants
confirmed successful integration of the T-DNA into the red clover genome. The system
described in this study can be used as a platform for genetic manipulation of red clover.
6.3. Gene expression
To confirm the successful gene manipulation done by biosynthesis engineering, the express of
the gene is question can be tested. This requires a tool that enables careful monitoring of
changes in the expression level of the manipulated gene. Quantitative reverse transcription
real-time polymerase chain reaction (qRT-PCR) is the most robust tool available for the
detection and quantification of expression of studied genes (Vandesompele et al. 2002). But
successful qRT-PCR first requires determination of the appropriate reference genes (internal
control genes). This prevents biased results in qRT-PCR studies. For this reason, prior to
attempting bioengineering manipulation of the metabolic pathway of the secondary
metabolites, we chose to first identify the best reference genes for red clover. Several
statistical algorithms such as geNorm, BestKeeper and NormFinder have been developed for
this purpose. Each of them has pitfalls; the strength of each algorithm lies in the
circumstances of its use. geNorm, the most commonly used algorithm, has two main
General discussion & future perspectives
159
advantages: 1) the raw data are used as an input file and do not need to be normally
distributed, and 2) geNorm does not need a large sample size to provide reliable results
(Serrano et al. 2011). However, geNorm top ranks those genes with the highest degree of
similarity in their expression profile, rather than with minimal variation (Andersen et al.
2004). That is why, in our study, geNorm chose the CRGs with a high correlation as being the
most stable. BestKeeper uses Pearson correlation analysis to identify the best RGs from Ct
values. False results can occur if the Ct values subjected to BestKeeper do not meet
assumptions of normal distribution. Although Pfaffl et al. (2004) stated that the Ct values
seem to be the best estimators of the expression levels, one report warns against subjecting Ct
values to parametric methods (Schmittgen et al., 2000). Those authors argue that because Ct
values are logarithmic in nature (Pelch et al. 2010), if they are ever normally distributed, Ct
values do not have a linear scale (Kortner et al., 2011). A new version of BestKeeper software
using Spearman and Kendall Tau correlations for non-parametric data has been promised but
is not yet available. BestKeeper employs pairwise correlation analysis to determine suitable
RGs and therefore top-ranks the RGs which display similar expression patterns across
samples as well as geNorm (Ponton et al. 2011). In our study, this led the results obtained by
geNorm and BestKeeper to be similar. The NormFinder algorithm uses a model-based
approach which takes variations across subgroups into account and avoids artificial selection
of co-regulated genes (Andersen et al. 2004). However, the intergroup variation calculated by
NormFinder does not show systematic variation across the sample subgroups of a single
CRG. In reality, the intergroup variation shows to which degree the data from group 1 of a
particular CRG vary compared to the mean variation of data from group 1 for all tested CRGs
(Kortner et al. 2011). In this case, the top-ranked genes will not be the most stable. The results
of our study show that the top-ranked RGs by NormFinder were not the most stable genes.
We have demonstrated that the application of geNorm, BestKeeper and NormFinder for
Chapter 6
160
selecting the best RG(s) could greatly heighten the risk of artificial selection. We therefore
propose an alternative, model-based method for computing the stability of CRG expression
which is free from the assumption used in these analysis tools. Simplicity of calculation and
accuracy are the major advantages of our model-based stability index as compared with the
previously developed model-based methods, because their application is either too intricate
(Serrano et al. 2011; Szabo et al. 2004) or they do not take the efficiency of PCR into
consideration (Brunner et al., 2004). Furthermore, the present study is the first systematic
comparison of potential RGs to determine the best-performing genes for accurate
normalization in red clover tissues (i.e., leaves, stems and roots). Our proposed method can be
also used to identify reliable RGs among CRGs of a tissue that has been subjected to different
treatments. To determine the most stable RGs across treatments, the stability index should be
calculated for each CRG in the same way that we calculated the stability index across tissues
in this study. The genes with a low stability index are considered to be the best RGs across the
treatments applied.
6.4. RNA interference silencing
Traditional breeding has been successful to develop new varieties of crops over the centuries.
The main disadvantages of the traditional breeding approach is the time-consuming and labor-
intensive nature of the work. Even more importantly, the potential progress via traditional
plant breeding is limited by the genetic resources available (Tang et al., 2007). Over the past
few decades, the innovation of genetic engineering techniques has opened many possibilities
to overcome the obstacles of traditional plant breeding. Prior to genetic manipulation, a
detailed understanding of functions of the genes involved in metabolic pathway in plants is
needed. Post-transcriptional gene silencing (PTGS) is a powerful tool to elucidate the
functions of genes (Sen and Blau, 2006), and has been frequently used to down-regulate genes
in plants (Schnettler et al., 2012). In our study, the function of the isoflavone synthase gene
General discussion & future perspectives
161
(IFS) in plant defense system was evaluated by the RNAi technique. To achieve this goal, an
IFS-RNAi hairpin construct was made and then transformed into Agrobacterium. Then, gene
transformation was successfully carried out by the Agrobacterium harboring the hairpin
transgene. The putative transformant lines were achieved via the regeneration protocol.
Transgenic lines were screened using PCR and were based on presence of the hairpin
transgenes in transgenic lines. Subsequently, the expression of the transformed IFS-RNAi
construct was validated by RT-qPCR. The result showed that the average expression level of
IFS in the T0 generation of the five transgenic lines (RNAi_1, RNAi_2, RANi_3, RANi_4
and RNAi_5) was much lower (approximately 30-fold) than in the controls (WT). This result
indicates the high efficiency of RNAi in silencing of the target gene. Our result is in
accordance with the report of Sullivan and Hatfield (2006). They reported that RNAi is highly
effective in red clover both in terms of number of plants displaying silencing and in extent of
silencing. In their study all studied transformant lines showed more than 90% reductions in
the expression of the target gene. All previous studies indicate that RNAi is a useful technique
to analyze gene function. Our result also showed that RNAi was efficient to silence the
isoflavone synthase gene (IFS) in red clover. The five chosen transgenic lines which showed
IFS down-regulation and the wild type plants were transferred to the greenhouse for
propagation and pathogenicity testing.
6.5. Pathogenicity assay
Two strategies can be used to study the role of isoflavones in a plant defense system. The first
is to increase the amount of isoflavones present, and then conduct corresponding assays to
discover whether higher levels of isoflavones enhance the resistance of red clover to given
pathogens. The second strategy is to decrease the isoflavone levels to investigate whether low
volume of isoflavones enhance the susceptibility of red clover to the studied pathogens.
Regarding the enhancement of isoflavone levels in red clover, one can assume that the
Chapter 6
162
overexpression of isoflavone synthase gene (IFS) could be the best strategy to increase the
production of isoflavonoids in hosts. However, numerous studies have shown that the
overexpression of only isoflavone synthase gene (IFS) results in only small changes in the
amounts of isoflavones. An investigation revealed that a high level of the intermediate 2-
hydroxyisoflavanone caused by the overexpression of IFS induced a negative feedback
inhibition of IFS. The latter problem might be overcome by overexpression of the enzyme
isoflavanone dehydratase (Akashi et al. 2005), which dehydrates 2-hydroxyisoflavanones to
produce isoflavones. Moreover, literature states that to have higher production of isoflavones,
the transcription factor genes need also to be overexpressed along with overexpression of IFS.
Also in order to increase the flux in the isoflavone pathway, the competing pathway of
flavonol biosynthesis needs to be impaired to provide more substrates for IFS. Therefore, to
enhance the amount of IFS, several genes needed to be manipulated. At the time we began
this work, there had been three approaches to introduce several genes into a plant. The
simplest way is that a plant containing one transgene is crossed with another plant harboring
other transgenes. This method is not applicable to red clover because of its self-
incompatibility. Another strategy is re-transformation. This method is slow and labor
intensive, however (Halpin, 2005). The last method is co-transformation with several vectors.
Stuitje et al. (2003) have reported the disadvantage that most inserts co-transformed into
Arabidopsis yielded very complex loci consisting of multiple tandem or inverted T-DNA
repeats. The enhancement of isoflavone levels requires transformation of multiple genes,
which is time-consuming and bears technical limitations. Although nowadays the application
of a multiple gene expression vector permits easy genetic engineering of several genes, we
started our research, the method of transgene stacking into a single vector was not yet
available. We therefore decided to choose the second strategy, to suppress isoflavone levels to
investigate the consequence of this manipulation on the plant defense system of red clover.
General discussion & future perspectives
163
At least two approaches to produce isoflavone-null plants have been published. The first is
traditional breeding through an extensive mutant isolation screen to identify isoflavone
depleted lines (Yu and McGonigle, 2005). The second is gene silencing to decrease isoflavone
levels. The mutant screening method is expensive and time-consuming, and mutant plants
exhibited lack of vigor and slow growth even under optimal growth conditions (Misyura et
al., 2013). Moreover, the RNAi technique was reported to be effective in silencing an
endogenous gene of red clover (Sullivan and Hatfield, 2006). We were therefore convinced
that the RNAi silencing approach was a better choice in our study. In our study, it was
important to produce transgenic plants that lack any antifungal compounds related to
isoflavonoid biosynthesis. Although the pterocarpans medicarpin and maackiain, which are
isoflavonoid end-products, are considered phytoalexins (Akashi et al., 2009), the isoflavones
themselves also exhibit antifungal activity (Graham and Graham, 1996). Thus the branch
point enzyme 2-hydroxyisoflavanone synthase (IFS), which introduces the isoflavonoid
biosynthetic pathway, was chosen to be subjected to the RNAi technique to produce
isoflavonoid-null plants.
It is now clear that total disease resistance is divided into specific resistance and basal
resistance in a homologous, incompatible interaction whereas a compatible interaction
consists only of basal resistance. The antifungal components may or may not be a part of
either specific resistance or basal resistance. In incompatible interactions, a disease resistance
(R) locus or its gene product recognizes the corresponding avr (avirulence) gene product and
then triggers a chain of signal-transduction events leading to fast activation of defense
mechanisms (Frey et al., 1997; Dangl and Jones, 2001). Therefore, if the production of
antifungal components are rapidly produced at the infection site to cease pathogen growth in
an incompatible interaction, one can conclude that the antifungal components are part of
specific resistance and if no production is observed, that means the antifungal components are
Chapter 6
164
not part of specific resistance. For example, no correlation between pathogen avirulence
(Pseudomonas syringae) and the phytoalexin camalexin accumulation was observed in A.
thaliana (Glazebrook and Ausubel, 1994). If antifungal components are part of basal
resistance, the antifungal components will be observed at infection sites of compatible
interactions with lower amount and slower induction speed compared to an incompatible
interaction and if not, no antifungal components will be detected. The silencing of the
antifungal components when they are part of specific resistance could cause severe infection.
The most notable example is the study performed on soybean by Subramanian et al. (2005).
They reported that the silencing of isoflavone accumulation severely compromised this R
gene-mediated specific resistance in transformed tissues. Also in another study, maize which
lacked phytoalexins was extremely susceptible to the fungal disease northern corn leaf blight
(Frey et al., 1997). Our pathogenicity assays demonstrated that the difference in disease
susceptibility between IFS-RNAi transgenic plants and wild types was relatively higher
toward Sclerotium rolfsii and relatively lower toward Sclerotinia trifoliorum, possibly due to
the latter pathogen’s isoflavone detoxification ability. However, the difference between IFS-
RNAi transgenic plants and wild type in both studied fungi was statistically significant,
especially with Sclerotium rolfsii. This result was in agreement with the previous report that
suppression of 2-hydroxyisoflavanone synthase (isoflavone synthase, IFS) and chalcone
reductase (CHS) in soybean resulted in decreased 5-deoxyisoflavonoid levels and reduced
resistance to pathogen infection (Du et al., 2010). Numerous other studies have shown that
reduced phytoalexin biosynthesis can lead to increased susceptibility of plants to pathogens
(Ferrari et al., 2003; Ferrari et al., 2007; Nafisi et al., 2007; Ren et al., 2008).
A large body of literature shows that overexpression of phytoalexin biosynthesis could
improve plant resistance. In rice, the enhancement of two phytoalexins, momilactone A and
sakuranetin, caused higher resistance against the blast fungus Magnaporthe grisea (Mori et
General discussion & future perspectives
165
al., 2007). More resistance to infection by Botrytis cinerea was obtained when the phytoalexin
stilbene synthase gene is expressed at a high level in tobacco (Hain et al., 1993). The
overexpression of isoflavone O-methyltransferase gene which mediates a key reaction in the
biosynthesis of the phytoalexin medicarpin display resistance to Phoma medicaginis in alfalfa
(He and Dixon, 2000). When taken together, the results of the present study and previous
findings show that enhancing phytoalexin levels by screening for cultivars with high levels of
phytoalexins through conventional breeding systems or genetic engineering in red clover may
increase resistance to S. trifoliorum and S. rolfsii.
6.6. Future perspectives
In the beginning, the concept of phytoalexins as antimicrobial compounds originated from the
fact of rapid accumulation of these products only in the presence of the pathogen living host
cells close to the infection site. Later on, further investigations showed phytoalexins
accumulates at infection sites could prevent further growth of the invading microorganism
(Essenberg et al., 1996; Keen and Holliday, 1982; Mansfield, 1982). The). However in that
time, the main focus of plant scientists was on production of pesticides or the genetic
engineering of defense mechanisms other than phytoalexins (Gurr and Rushton, 2005). The
reasons that phytoalexins drew less attention for a long time were that 1) phytoalexins contain
a huge diversity and their biosynthesis is complex, making it difficult to understand their
general regulatory mechanisms (Grosskinsky et al., 2012); 2) some phytoalexins are
detoxified by pathogens, and 3) during that time, suitable techniques were lacking which
could allow the manipulation of multiple genes on the same, or interconnected, pathways to
achieve effective metabolic engineering. Nowadays, however, new advances in genetic
engineering enable us to identify phytoalexin biosynthetic genes and their regulation in plants
and also understand the process of phytoalexin detoxification in pathogens. The recent
technique, multiple gene expression vector, enables us to control several genes in a pathway.
Chapter 6
166
There is a trend to reduce pesticide use and only little room is left for genetic manipulation of
other defense components. Taken together, phytoalexins have become a promising target for
improving crops (Gurr and Rushton, 2005). However, we need to use caution in genetic
engineering of phytoalexins, because all precursors of secondary metabolism pathways are
derived from primary metabolism, and any imbalances between them result in limit crop
growth and consequently yield (Aharoni and Galili, 2011).
To achieve the red clover with higher resistance to Sclerotinia trifoliorum and Sclerotium
rolfsii, the amount of phytoalexins needs to be increased. Nevertheless we should bear in
mind that the plant secondary metabolism pathways are normally divided into many branches
and in most of cases there is competition for the same substrates between the investigated flux
pathways and other pathways. That is why, in most cases, low levels of isoflavones were
achieved when legumes or non-legumes were subjected to IFS gene transformation. We may
speculate that such poor results were caused by the competition between different upstream
substrates of IFS. A notable example is the study performed by Liu et al. (2002), which
demonstrated such a competition between IFS and flavanone-3-hydroxylase (F3H). They
reported the introduction of a soybean IFS into a double mutant of Arabidopsis thaliana
(tt6/tt3: blocked in flavonol and anthocyanin synthesis) resulted in a high level of the
isoflavone genistein compared to transgenic non-mutant. This indicates IFS was not the rate
limiting step in the isoflavone genistein production, and the major bottleneck for isoflavone
engineering is the partition of flux between flavonol and isoflavone biosynthesis at the level
of naringenin, the substrate for both IFS and F3H (Kaurt and Murphy, 2012). Therefore an
effective metabolic engineering can only be achieved by manipulation of multiple genes on
the same, or interconnected, pathways (Halpin, 2005). If red clover has the same bottleneck
for biosynthetic engineering of isoflavones, we need to divert the flux to the isoflavone
pathway by the RNAi technique, which is the best strategy to block the other pathway.
General discussion & future perspectives
167
The constitutive CaMV 35S promoter has been used in most previous biosynthesis
engineering studies that have been performed to increase disease resistance. Constitutive
overexpression of defense components can lead to increased disease resistance but it often can
reduce plant size (Chen and Chen, 2002) and alter morphology (Li et al., 2004). Therefore,
instead of constitutive overexpression, the expression of transgenes should be restricted to
infection sites and also occur rapidly enough to prevent the growth of pathogens. In other
words, to avoid the downsides of constitutive overexpression, we need a suitable promoter to
express in the right place at the right time. Due to the modular nature of plant promoters,
appropriate synthetic promoters can be constructed by putting together needed elements in the
same way that one might put Lego blocks together (Gurr and Rushton, 2005), taking into
account that specific transcription factors are the main regulators of temporal and spatial gene
expressions. Therefore the best strategy is to use a synthetic promoter with multiple binding
sites for multiple transcription factors that are inducible by the key players in defense
signaling. Such a strategy enables one to regulate coordinately several genes participating in
the same pathway for phytoalexin biosynthesis and could lead to a more robust increase in
phytoalexin levels. MYB transcription factors can be used in building an appropriate synthetic
gene in red clover, because it has been proved that MYB transcription factors are involved in
defense signaling (Mengiste et al., 2003) and also regulation of the flavonoid biosynthetic
pathway (Boddu et al., 2005). Another alternative way to increase phytoalexin levels is the
application of phytohormones, which are important regulators of phytoalexin biosynthesis. It
is reported that increasing the amount of cytokinin could enhance accumulation of different
phenolic acids, which are known as potential phytoalexin precursors (Schnablova et al.,
2006). Interestingly and importantly, the induction of phytoalexins with phytohormones could
happen independently of salicylate, jasmonate and ethylene (Grosskinsky et al., 2011). The
application of a synthetic pathogen inducible promoter for the isopentenyl transferase gene,
Chapter 6
168
which encodes a critical enzyme in cytokinin biosynthesis, caused high levels of the
phytoaexins of tobacco, leading to enhanced resistance against Pseudomonas syringae
(Grosskinsky et al., 2011). Cytokinin allows gentle increases in phytoalexin, leading to
efficient pathogen resistance without the induction of HR. Avoiding HR formation protects
tissue integrity and reduces the ecological cost of loss in photosynthetic tissue (Kliebenstein
and Rowe, 2008). In addition to altering endogenous phytoalexins, the introduction of
‘foreign’ antifungal compounds into host plants can be a useful strategy to increase resistance
(Hain et al., 1993). Nevertheless the foreign antifungal compounds need to be to fit with the
plant’s endogenous defense compounds. The most notable example was the transfer of a gene
with antifungal activity to carrot and cucumber, which led to an increase of resistance in
carrot, but not in cucumber, even when the same pathogen was used to infect both crops
(Punja and Raharjo, 1996).
Finally, we would like to stress two common mistakes in phytoalexin studies. First, no one
should expect full inhibition of pathogens by plant extraction or synthetic phytoalexins in
vitro. In fact, phytoalexins are only a part of the complex pathogen defense responses (Iriti
and Faoro, 2009). Therefore, if less inhibition was observed in vitro, we cannot conclude that
the studied phytoalexins have no role in plant defense responses. In fact, phytoalexins buy
time by delaying pathogen propagation and allow the plant to launch other defense responses
(Grosskinsky et al., 2011). Second, there is no relation between the amount of existing
precursors and their conjugates with the resistance of the studied plants. In another words,
higher isoflavones don’t always result in higher resistance. The most notable example is the
study conducted by Debnam and Smith on red clover in 1976. They reported that the
isoflavone concentration in red clover before or after infection did not correlate with their
Sclerotinia resistance. Therefore, they concluded that isoflavone concentration cannot be used
as an index of disease resistance. Also Subramanian et al. (2005) claimed that two different
General discussion & future perspectives
169
cultivars of soybean with the same amount of isoflavones could show different resistance.
They demonstrated that although susceptible (Williams) and resistant varieties (Williams 82)
have high level of the preformed conjugates of daidzein, analytical analysis showed that only
resistant variety enabled to hydrates its conjugated isoflavones when faced with a pathogen.
As a conclusion, we cannot judge based on the existing amount of isoflavones and their
conjugated forms.
Numerous studies have shown that the manipulation of phytoalexin levels by transgenic
expression of genes involved in biosynthesis was successful for improved crop resistance. We
think that increasing the amount of infection-induced phytoalexin may lead to enhanced
resistance of red clover against fungi including Sclerotinia trifoliorum and Sclerotium rolfsii,
which are the most economically important diseases of red clover. To increase phytoalexin
levels, we suggest that the combination of using synthetic promoters containing suitable
transcription factors and suppression of the competing pathway using the RNAi technique
could provide a successful method to enhance accumulation of phytoalexin in red clover
leading to more resistant cultivars. For those breeders who are not allowed to apply gene
transformation methods in their disease resistance breeding programs against Sclerotinia
trifoliorum and Sclerotium rolfsi, we suggest that they conduct a detached leaf assay in
parallel to their breeding program for each generation. Then they should measure the
infection-induced levels of the pterocarpans medicarpin and maackiain by analytical analysis
equipment such as LC-MS. Only plants which show high level of medicarpin and maackiain
in should be used to build the next generation’s population.
Summary
171
Summary
Red clover (Trifolium pratense L.) is one of the important forage crops in most temperate
agricultural regions. In addition to the importance of red clover in nitrogen fixation and
animal feeding, it has also drawn attention because of high levels of the secondary metabolite
isoflavones which possess antifungal activity (Saviranta et al., 2008). The main disadvantage
of red clover is its insufficient persistence (Hejduk, 2011). Therefore, persistence (individual
plant longevity) is the main goal of most of red clover breeding programs (Taylor, 2008).
Because fungal diseases are one contributing factor to the lack of persistence of red clover,
improving fungal disease resistance should help to improve persistence. Using isoflavones as
the plants’ own defense arsenals can be an environmentally friendly strategy to strengthen
plant resistance to fungal pathogens. In this study, we decided to evaluate the possibly
significant role of isoflavones in a red clover defense system against the fungal pathogens
Sclerotinia trifoliorum and Sclerotium rolfsii using biosynthesis engineering of the
corresponding gene. The first step toward achieving this goal is to set up effective
regeneration and transformation systems, which are the prerequisite any biosynthesis
engineering. In order to establish an efficient regeneration system in red clover, highly
regenerative genotypes were used to evaluate the suitability of different media sequences to
maximize the regeneration capacity of the plant. The media sequence KBC-SPL-SPL
significantly increased the efficiency of somatic embryogenesis of red clover. To develop an
efficient gene transfer system, some parameters were investigated, including evaluation of a
suitable concentration of the selection agent, examination of the effect of pre-culturing of
explants on the efficiency of transformation and comparison of different Agrobacterium
strains for the efficiency of gene transformation. Based on the selection agent test, 75 mg/l of
kanamycin sulfate was an appropriate concentration for transformant selection. The effect of
pre-culturing explants was increased slightly by the frequency of gene transferring. The
Summary
172
highly infective Agrobacterium strain C58C1 can increase the transformation efficiency for
red clover to 83%. PCR and histochemical GUS assay of the transformed plants confirmed
successful integration of the T-DNA into the red clover genome. In addition to an efficient
gene transformation system, a tool is needed to enable careful monitoring of changes of the
manipulated gene conducted by biosynthesis engineering. Quantitative reverse transcription
real-time polymerase chain reaction (qRT-PCR) is the most robust tool available to detect any
changes in expression of studied genes (Vandesompele et al., 2002). However, before qRT-
PCR can be used in this way, one must determine the appropriate reference genes for
normalization of gene expression. To identify the best-suited reference gene(s) for
normalization, several statistical algorithms such as geNorm, BestKeeper and NormFinder
have been developed. All of these algorithms are based on the key assumption that none of the
investigated candidate reference genes show systematic variation in their expression profile
across the samples being considered. But this assumption is likely to be violated in practice.
We therefore suggested a simple and novel stability index based on the analysis of variance
model, which is free from the assumption made by the algorithms. We identified several
reference genes with a high level of stability using our own model-based method. Then an
IFS-RNAi hairpin construct was made using the gateway cloning system. Subsequently, gene
transformation was successfully performed with an Agrobacterium harboring the hairpin
transgene. The putative transformant lines were achieved using the regeneration protocol.
Transgenic lines with a functional construct were screened using RT-qPCR. The five chosen
transgenic lines which showed IFS down-regulation, plus the wild type plants, were
transferred to the greenhouse for propagation. Then the collected leaves of IFS-RNAi
transgenic red clover and wild type, which served as control, were exposed to the two soil-
borne fungal diseases S. trifoliorum and S. rolfsii in pathogenicity assays using a detached-
leaf method. The result obtained from orthogonal contrasts demonstrated that the difference
Summary
173
between wild type line versus transgenic lines (isoflavone-null lines) were found to be
significant for S. trifoliorum and highly significant for S. rolfsii. The present study confirms
the significant role of isoflavones in disease resistance. The use of isoflavones is suggested in
resistance breeding programs to improve red clover persistence.
Samenvatting
175
Samenvatting
Rode klaver (Trifolium pratense L.) is een belangrijk voedergewas in veel landbouwsystemen.
Naast zijn stikstoffixatie en hoge kwaliteit als veevoeder beschikt rode klaver over hoge
gehalten van de secondaire metaboliet isoflavonen, die een antischimmel activiteit bezit
(Saviranta et la., 2008). Een nadeel van rode klaver is zijn beperkte persistentie (Hejduk,
2011). Langleefbaarheid is dan ook een belangrijk selectiecriterium bij rode klaververedeling
(Taylor, 2008). Vooral grondgebonden ziekten spelen een belangrijke rol bij de
langleefbaarheid en dus zal een betere ziekteresistentie ook de persistentie verbeteren. Eigen
plantisoflavonen zou de plant op een milieuvriendelijke wijze kunnen wapenen tegen ziekten.
In dit doctoraat werd nagegaan in hoeverre isoflavonen een significante rol kunnen spelen in
het afweermechanisme tegen de pathogenen Sclerotinia trifolium en Sclerotium rolfsii gebruik
makend van genetische modificatie van de betreffende genen.
De eerste stap was een effectief regeneratie en transformatie protocol op te zetten. Genotypes
met een hoge regeneratiecapaciteit werden gebruikt om de geschiktheid van diverse media uit
te testen teneinde een maximale regeneratie te bekomen. Hieruit bleek dat de
mediumopvolging KBC-SPL-SPL de somatische embryogenese bij rode klaver significant
verhoogde. Teneinde een efficiënt transformatiesysteem te ontwikkelen werden diverse
parameters onderzocht zoals: concentratie van de selectie agent, rol van precultuur en
vergelijking van verschillende Agrobacterium stammen. De resultaten toonden dat 75 mg/l
kanamycinsulfaat een geschikte concentratie was voor selectie van transformanten en dat
precultuur slechts een beperkte invloed had op transformatiesucces. De Agrobacterium stam
C58C1 daarentegen verhoogde de transformatie efficiëntie met 83%. PCR en GUS testen
bevestigden de succesvolle integratie van T-DNA in het rode klavergenoom.
Veranderingen in expressie van de bestudeerde genen gebeurde via qRT-PCR (Vandesompele
et al. 2002). Hiervoor zijn geschikte referentiegenen nodig. Om deze te identificeren zijn
Samenvatting
176
meerdere statistische algoritmen ontwikkeld zoals GeNorm, BestKeeper en NormFinder. Al
deze algoritmen zijn gebaseerd op het vermoeden dat geen van de onderzochte kandidaat
referentiegenen variatie vertoond in hun expressieprofiel, maar deze stelling geldt niet steeds.
Daarom werd een eenvoudig en nieuwe stabiliteitsindex ontwikkeld gebaseerd op een
variantieanalysemodel. Zo konden meerdere referentiegenen bepaald worden met een hoge
stabiliteit. Vervolgens werd een ISF-RNAi hairpin construct gemaakt waarna de
gentransformatie succesvol kon gebeuren met een Agrobacterium die het hairpin transgen
bevatte. De vermoedelijke transformanten werden geregenereerd via het ontwikkelde
regeneratie protocol. Transgene lijnen met een functioneel construct werden gescreend door
middel van RT-qPCR.
Vijf transgene lijnen die IFS down regulatie vertoonden en de wild type (uitgangsmateriaal)
planten werden in serres vermeerderd. Bladeren van de IFS-RNAi transgene rode klaver en
van de wild type planten werden blootgesteld aan de schimmels S. Trifoliorum en S. rolfsii.
De resultaten toonden dat de verschillen in aantasting tussen wild type bladeren en de
bladeren van de transgene (isoflavone nul ) planten significant waren voor S. trifoliorum en
zeer significant varen voor S. rolfsii. Deze studiea bevestigt de significante rol van
isoflavonen in ziekteresistentie. Isoflavonen zouden dus een rol kunnen spelen in
resistentieveredeling met het oog op een verbetering van de rode klaverpersistentie.
Reference List
177
Reference List
Abd-Allah,E.F. (2005). Effect of a Bacillus subtilis isolate on southern blight (Sclerotium
rolfsii) and lipid composition of peanut seeds. Phytoparasitica 33(5), 460-466.
Agarwal,S., Kanwar,K., Saini,N., and Jain,R.K. (2004). Agrobacterium tumefaciens mediated
genetic transformation and regeneration of Morus alba L. Scientia Horticulturae 100(1-4),
183-191.
Agrios,G.N. (2005). Plant pathology. (Elsevier Academic Press).
Aharoni,A. and Galili,G. (2011). Metabolic engineering of the plant primary-secondary
metabolism interface. Current Opinion in Biotechnology 22(2), 239-244.
Akashi,T., Aoki,T., and Ayabe,S. (2005). Molecular and biochemical characterization of 2-
hydroxyisoflavanone dehydratase. Involvement of carboxylesterase-like proteins in
leguminous isoflavone biosynthesis. Plant Physiology 137(3), 882-891.
Akashi,T., Sasaki,K., Aoki,T., Ayabe,S., and Yazaki,K. (2009). Molecular cloning and
characterization of a cDNA for pterocarpan 4-dimethylallyltransferase catalyzing the key
prenylation step in the biosynthesis of glyceollin, a soybean phytoalexin. Plant Physiology
149(2), 683-693.
Akbulut,M., Yucel,M., and Oktem,H.A. (2008). Analysis and optimization of DNA delivery
into chickpea (Cicer arietinum L.) seedlings by Agrobacterium tumefacience. African Journal
of Biotechnology 7(8), 1011-1017.
Andersen,C.L., Jensen,J.L., and Orntoft,T.F. (2004). Normalization of real-time quantitative
reverse transcription-PCR data: A model-based variance estimation approach to identify
genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Research
64(15), 5245-5250.
Andersen,Ø.M. and Markham,K.R. (2006). Flavonoids: Chemistry, Biochemistry and
Applications. (CRC Press Book).
Anderson,M.K., Taylor,N.L., and Kirthavi,R. (1972). Development and performance of
double-cross hybrid red clover. Crop Science 12(2), 240-242.
Anower Jabed, M. D. (2012). RNAi-mediated reduction of a major ruminant specific milk
allergen using a transgenic mouse model. The University of Waikato. Ref Type:
Thesis/Dissertation.
Arcioni,S., Damiani,F., Pupilli,F., and Pezzotti,M. (1990). Somatic embryogenesis and
somaclonal variation in Medicago sativa L. Journal of Genetics and Breeding 43(4): 223-230.
Arthur,G., Stirk,W.A., and van Staden,J. (2004). Screening of aqueous extracts from gelling
agents (Agar and Gelrite) for root-stimulating activity. South African Journal of Botany 70(4),
595-601.
Reference List
178
Atkinson,P. and Blakeman,J.P. (1982). Seasonal occurrence of an antimicrobial flavanone,
sakuranetin, associated with glands on leaves of Ribes nigrum L. New Phytologist 92(1), 63-
74.
Aycock, R. (1966). Stem rot and other diseases caused by Sclerotium rolfsii.
N.C.Agr.Expt.St.Tech.Bul. No. 174.
Bao,S. and Cagan,R. (2006). Fast cloning inverted repeats for RNA interference. Rna-A
Publication of the Rna Society 12, 2020-2024.
Beach,K.H. and Smith,R.R. (1979). Plant regeneration from callus of red and crimson clover.
Plant Science Letters 16(2-3), 231-237.
Bell,A.A., Stipanovic,R.D., Mace,M.E., and Kohel,R.J. (1994). Genetic manipulation of
terpenoid phytoalexins in gossypium: Effects on disease resistance. Genetic Engineering of
Plant Secondary Metabolism 28, 231-249.
Bennett,R.N. and Wallsgrove,R.M. (1994). Secondary metabolites in plant defense
mechanisms. New Phytologist 127(4), 617-633.
Bernards,M.A. (2010). Plant natural products: a primer. Canadian Journal of Zoology-Revue
Canadienne de Zoologie 88(7), 601-614.
Bevan,M.W., Flavell,R.B., and Chilton,M.D. (1983). A chimaeric antibiotic resistance gene
as a selectable marker for plant cell transformation. Nature 304, 184-187.
Bingham,E.T., Hurley,L.V., Kaatz,D.M., and Saunders,J.W. (1975). Breeding alfalfa which
regenerates from callus tissue in culture. Crop Sci 15(5), 719-721.
Binns,A.N. and Thomashow,M.F. (1988). Cell biology of Agrobacterium infection and
transformation of plants. Annual Review of Microbiology 42, 575-606.
Birch,R.G. (1997). Plant transformation: Problems and strategies for practical application.
Annual Review of Plant Physiology and Plant Molecular Biology 48, 297-326.
Boddu,J., Svabek,C., Ibraheem,F., Jones,A.D., and Chopra,S. (2005). Characterization of a
deletion allele of a sorghum Myb gene, yellow seed1 showing loss of 3-deoxyflavonoids.
Plant Science 169(3), 542-552.
Boller,B., Posselt,U.K., and Veronesi,F. (2010). Handbook of Plant Breeding. (Dordrecht,
Germany: Springer).
Boller,B.C. and Nosberger,J. (1987). Symbiotically fixed nitrogen from field-grown white
and red clover mixed with ryegrasses at low levels of 15
N-fertilization. Plant and Soil 104(2),
219-226.
Bolognesi,R., Ramaseshadri,P., Anderson,J., Bachman,P., Clinton,W., Flannagan,R.,
Ilagan,O., Lawrence,C., Levine,S., Moar,W., Mueller,G., Tan,J.G., Uffman,J., Wiggins,E.,
Heck,G., and Segers,G. (2012). Characterizing the mechanism of action of double-stranded
RNA activity against western corn rootworm (Diabrotica virgifera virgifera LeConte). Plos
One 7(10).
Reference List
179
Bookout,A.L., Cummins,C.L., Mangelsdorf,D.J., Pesola,J.M., and Kramer,M.F. (2006). High-
throughput real-time quantitative reverse-transcription PCR. In Current protocols in molecular
biology, F.M.Ausubel, R.Brent, R.E.Kingston, D.D.Moore, J.G.Seidman, J.A.Smith, and
K.Struhl, eds. (New York: John Wiley, Sons, Inc.), 15.8.1-15.8.28.
Bottinger,P., Steinmetz,A., Schieder,O., and Pickardt,T. (2001). Agrobacterium-mediated
transformation of Vicia faba L. Molecular Breeding 8(3), 243-254.
Brechenmacher,L., Lei,Z.T., Libault,M., Findley,S., Sugawara,M., Sadowsky,M.J.,
Sumner,L.W., and Stacey,G. (2010). Soybean metabolites regulated in root hairs in response
to the symbiotic bacterium Bradyrhizobium japonicum. Plant Physiology 153(4), 1808-1822.
Bredenberg,J.B. (1961). Identification of an antifungal factor in red clover as biochanin A.
Suomen Kemistilehti B 34, 23-26.
Brummer,E.C. (1999). Capturing heterosis in forage crop cultivar development. Crop Sci
39(4), 943-954.
Brunner,A., Yakovlev,I., and Strauss,S. (2004). Validating internal controls for quantitative
plant gene expression studies. Bmc Plant Biology 4, 14.
Buchanan,B.B., Gruissem,W., and Jones,R.L. (2000). Biochemistry and molecular biology of
plants. (River Street: Wiley, John & Sons).
Bustin,S.A. (2000). Absolute quantification of mRNA using real-time reverse transcription
polymerase chain reaction assays. Journal of Molecular Endocrinology 25, 169-193.
Calsamiglia,S., Busquet,M., Cardozo,P.W., Castillejos,L., and Ferret,A. (2007). Essential oils
as modifiers of rumen microbial fermentation. Journal of Dairy Science 90(6), 2580-2595.
Carrillo,J.C., Ojeda,V.A., Quiroz,H.A.C., and Ortega,F.M. (2004). Optimization of a protocol
for direct organogenesis of red clover (Trifolium pratense L.) meristems for breeding
purposes. Biological Research 37, 45-51.
Cartwright,D.W., Langcake,P., Pryce,R.J., Leworthy,D.P., and Ride,J.P. (1981). Isolation and
characterization of two phytoalexins from rice as momilactones A and momilactones B.
Phytochemistry 20(3), 535-537.
Chappell,J., Wolf,F., Proulx,J., Cuellar,R., and Saunders,C. (1995). Is the reaction catalyzed
by 3-hydroxy-3-methylglutaryl coenzyme A reductase a rate-limiting step for isoprenoid
biosynthesis in plants? Plant Physiology 109(4), 1337-1343.
Chen,C.H. and Chen,Z.X. (2002). Potentiation of developmentally regulated plant defense
response by AtWRKY18, a pathogen-induced Arabidopsis transcription factor. Plant
Physiology 129(2), 706-716.
Chen,H., Nelson,R.S., and Sherwood,J.L. (1994). Enhanced recovery of transformants of
Agrobacterium tumefaciens after freeze-thaw transformation and drug selection.
Biotechniques 16(4), 664-668, 670.
Reference List
180
Cho,M.J. and Harper,J.E. (1991a). Effect of inoculation and nitrogen on isoflavonoid
concentration in wild-type and nodulation-mutant soybean roots. Plant Physiol 95(2), 435-
442.
Cho,M.J. and Harper,J.E. (1991b). Root isoflavonoid response to grafting between wild-type
and nodulation-mutant soybean plants. Plant Physiol 96(4), 1277-1282.
Choi,H.K., Mun,J.H., Kim,D.J., Zhu,H.Y., Baek,J.M., Mudge,J., Roe,B., Ellis,N., Doyle,J.,
Kiss,G.B., Young,N.D., and Cook,D.R. (2004). Estimating genome conservation between
crop and model legume species. Proceedings of the National Academy of Sciences of the
United States of America 101(43), 15289–15294.
Collins,G.B., Phillips,G.C., Earle,E.D., and Demarly,Y. (1982). In vitro tissue culture and
plant regeneration in Trifolium pratense L. (New York: Praeger).
Colliver,S.P., Morris,P., and Robbins,M.P. (1997). Differential modification of flavonoid and
isoflavonoid biosynthesis with an antisense chalcone synthase construct in transgenic Lotus
corniculatus. Plant Molecular Biology 35(4), 509-522.
Cornelius,P.L., Taylor,N.L., and Anderson,M.K. (1977). Combining ability in I1 single
crosses of red clover. Crop Sci 17(5), 709-713.
Creasy,L.L. and Coffee,M. (1988). Phytoalexin production potential of grape berries. Journal
of the American Society for Horticultural Science 113(2), 230-234.
Crombie,L., Crombie,W.M.L., and Whiting,D.A. (1984). Isolation of avenacins A-1, A-2, B-1
and B-2 from oat roots:Structures of their 'aglycones', the avenestergenins. Journal of the
Chemical Society-Chemical Communications 4, 244-246.
Crush,J.R., Cosgrove,G.P., and Brougham,R.W. (1983). Nitrogen fixation during 1979-81 in
two pastures on the Manawatu plains. New Zealand Journal of Experimental Agriculture
11(1), 17-20.
Czechowski,T., Stitt,M., Altmann,T., Udvardi,M.K., and Scheible,W.R. (2005). Genome-
wide identification and testing of superior reference genes for transcript normalization in
Arabidopsis. Plant Physiology 139(1), 5-17.
Czihal,A., Conrad,B., Buchner,P., Brevis,R., Farouk,A.A., Manteuffel,R., Adler,K.,
Wobus,U., Hofemeister,J., and Baumlein,H. (1999). Gene farming in plants: Expression of a
heat-stable Bacillus amylase in transgenic legume seeds. Journal of Plant Physiology 155(2),
183-189.
Dangl,J.L. and Jones,J.D.G. (2001). Plant pathogens and integrated defence responses to
infection. Nature 411, 826-833.
Davidson,B.L. and Mccray,P.B. (2011). Current prospects for RNA interference-based
therapies. Nature Reviews Genetics 12, 329-340.
De Clercq,J., Zambre,M., Van Montagu,M., Dillen,W., and Angenon,G. (2002). An optimized
Agrobacterium-mediated transformation procedure for Phaseolus acutifolius A. Gray. Plant
Cell Reports 21(4), 333-340.
Reference List
181
de Faria, I. J., Olmo, R. P., Silva, E. G., and Marques, J. T. (2013). dsRNA sensing during
viral infection: Lessons from plants, worms, insects, and mammals. Journal of Interferon &
Cytokine Research 33(5), 239-253.
Deavours,B.E. and Dixon,R.A. (2005). Metabolic engineering of isoflavonoid biosynthesis in
alfalfa. Plant Physiology 138(4), 2245-2259.
Debnam, J. R. and Smith, I. M. (1976). Changes in the isoflavones and pterocarpans of red
clover on infection with Sclerotinia trifoliorum and Botrytis cinerea. Physiological Plant
Pathology 9(1), 9-23.
Delclos,B., Mousset-Declas,C., and Raynal,G. (1997). A simple method for the evaluation of
red clover (Trifolium pratense L) resistance to Sclerotinia trifoliorum. Euphytica 93(2), 173-
179.
Delclos,B. and Raynal,G. (1995). Comparison of techniques for the production of Sclerotinia
trifoliorum ascospores in the laboratory for forage legumes resistance tests. Journal of
Phytopathology-Phytopathologische Zeitschrift 143(6), 345-348.
Delserone,L.M., Matthews,D.E., and Vanetten,H.D. (1992). Differential toxicity of
enantiomers of maackiain and pisatin to phytopathogenic fungi. Phytochemistry 31(11), 3813-
3819.
Dhaubhadel,S., Farhangkhoee,M., and Chapman,R. (2008). Identification and characterization
of isoflavonoid specific glycosyltransferase and malonyltransferase from soybean seeds.
Journal of Experimental Botany 59(4), 981-994.
Dixon,R.A. (2001). Natural products and plant disease resistance. Nature 411, 843-847.
Dixon,R.A., Achnine,L., Kota,P., Liu,C.J., Reddy,M.S.S., and Wang,L.J. (2002). The
phenylpropanoid pathway and plant defence - a genomics perspective. Molecular Plant
Pathology 3(5), 371-390.
Dixon,R.A. and Sumner,L.W. (2003). Legume natural products: understanding and
manipulating complex pathways for human and animal health. Plant Physiology 131(3), 878-
885.
Du,H., Huang,Y.B., and Tang,Y.X. (2010). Genetic and metabolic engineering of
isoflavonoid biosynthesis. Applied Microbiology and Biotechnology 86(5), 1293-1312.
Earley,K.W., Haag,J.R., Pontes,O., Opper,K., Juehne,T., Song,K.M., and Pikaard,C.S. (2006).
Gateway-compatible vectors for plant functional genomics and proteomics. Plant Journal
45(4), 616-629.
Ebel,J. and Cosio,E.G. (1994). Elicitors of plant defense responses. International Review of
Cytology - A Survey of Cell Biology 148, 1-32.
Elbashir,S.M., Harborth,J., Lendeckel,W., Yalcin,A., Weber,K., and Tuschl,T. (2001).
Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells.
Nature 411, 494-498.
Reference List
182
Eriksson, J. (1880). Kungliga Landtbraksakoemiens handlingar och tidskrift. Akad.Handi
Tidskr. 19, 28.
Essenberg,M., Grover,P.B., and Cover,E.C. (1990). Accumulation of antibacterial
sesquiterpenoids in bacterially inoculated gossypium leaves and cotyledons. Phytochemistry
29, 3107-3113.
Exposito-Rodriguez,M., Borges,A.A., Borges-Perez,A., and Perez,J.A. (2008). Selection of
internal control genes for quantitative real-time RT-PCR studies during tomato development
process. Bmc Plant Biology 8.
Ferrari,S., Galletti,R., Denoux,C., De Lorenzo,G., Ausubel,F.M., and Dewdney,J. (2007).
Resistance to Botrytis cinerea induced in Arabidopsis by elicitors is independent of salicylic
acid, ethylene, or jasmonate signaling but requires PHYTOALEXIN DEFICIENT3. Plant
Physiology 144, 367-379.
Ferrari,S., Plotnikova,J.M., De Lorenzo,G., and Ausubel,F.M. (2003). Arabidopsis local
resistance to Botrytis cinerea involves salicylic acid and camalexin and requires EDS4 and
PAD2, but not SID2, EDS5 or PAD4. Plant Journal 35, 193-205.
Ferrer,J.L., Austin,M.B., Stewart,C., and Noe,J.P. (2008). Structure and function of enzymes
involved in the biosynthesis of phenylpropanoids. Plant Physiology and Biochemistry 46,
356-370.
Fery,R.L. and Dukes,P.D. (2002). Southern blight (Sclerotium rolfsii Sacc.) of cowpea: yield-
loss estimates and sources of resistance. Crop Protection 21, 403-408.
Fire,A., Xu,S.Q., Montgomery,M.K., Kostas,S.A., Driver,S.E., and Mello,C.C. (1998). Potent
and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature
391, 806-811.
Fisher,R.F. and Long,S.R. (1992). Rhizobium - plant signal exchange. Nature 357, 655-660.
Foroughi-Wehr,B., Wenzel,G., Hayward,M., Bosemark,N.O., and Romagosa,I. (1993).
Andro- and parthenogenesis, 261-277.(London: Chapman and Hall).
Fraley,R.T., Rogers,S.G., Horsch,R.B., Sanders,P.R., Flick,J.S., Adams,S.P., Bittner,M.L.,
Brand,L.A., Fink,C.L., Fry,J.S., Galluppi,G.R., Goldberg,S.B., Hoffmann,N.L., and Woo,S.C.
(1983). Expression of bacterial genes in plant cells. Proceedings of the National Academy of
Sciences of the United States of America-Biological Sciences 80, 4803-4807.
Frankenberger,E.A., Hasegawa,P.M., and Tigchelaar,E.C. (1981). Diallel analysis of shoot-
forming capacity among selected tomato genotypes. Zeitschrift fur Pflanzenphysiologie 102,
233-242.
Frey,M., Chomet,P., Glawischnig,E., Stettner,C., Grun,S., Winklmair,A., Eisenreich,W.,
Bacher,A., Meeley,R.B., Briggs,S.P., Simcox,K., and Gierl,A. (1997). Analysis of a chemical
plant defense mechanism in grasses. Science 277, 696-699.
Fu,C.L., Wehr,D.R., Edwards,J., and Hauge,B. (2008). Rapid one-step recombinational
cloning. Nucleic Acids Research 36.
Reference List
183
Gamborg,L., Miller,R.A., and Ojima,K. (1968). Nutrient requirements of suspension cultures
of soybean root cells. Exp. Cell Res. 50, 151-158.
Gavilano,L.B., Coleman,N.P., Burnley,L.E., Bowman,M.L., Kalengamaliro,N.E., Hayes,A.,
Bush,L., and Siminszky,B. (2006). Genetic engineering of Nicotiana tabacum for reduced
nornicotine content. Journal of Agricultural and Food Chemistry 54, 9071-9078.
Geley,S. and Muller,C. (2004). RNAi: ancient mechanism with a promising future.
Experimental Gerontology 39, 985-998.
Gelvin,S.B. (2003). Agrobacterium-mediated plant transformation: the biology behind the
"gene-jockeying" tool. Microbiol. Mol. Biol. Rev. 67, 16-37.
Germ,M., Kreft,I., and Osvald,J. (2005). Influence of UV-B exclusion and selenium treatment
on photochemical efficiency of photosystem II, yield and respiratory potential in pumpkins
(Cucurbita Pepo L.). Plant Physiology and Biochemistry 43, 445-448.
Gheysen,G., Villarroel,R., and VanMontagu,M. (1991). Illegitimate recombination in plants:
A model for T-DNA integration. Genes & Development 5, 287-297.
Ghorbel,R., Dominguez,A., Navarro,L., and Pena,L. (2000). High efficiency genetic
transformation of sour orange (Citrus aurantium) and production of transgenic trees
containing the coat protein gene of citrus tristeza virus. Tree Physiology 20, 1183-1189.
Glazebrook,J. and Ausubel,F.M. (1994). Isolation of phytoalexin-deficient mutants of
Arabidopsis thaliana and characterization of their interactions with bacterial pathogens.
Proceedings of the National Academy of Sciences of the United States of America 91, 8955-
8959.
Gorte,M., Horstman,A., Page,R.B., Heidstra,R., Stromberg,A., and Boutilier,K. (2011).
Microarray-based identification of transcription factor target genes. Methods in Molecular
Biology 754, 119-41.
Graham,T.L. and Graham,M.Y. (1996). Signaling in soybean phenylpropanoid responses
(dissection of primary, secondary, and conditioning effects of light, wounding, and elicitor
treatments). Plant Physiol 110, 1123-1133.
Graham,T.L., Kim,J.E., and Graham,M.Y. (1990). Role of constitutive isoflavone conjugates
in the accumulation of glyceollin in soybean infected with Phytophthora megasperma.
Molecular Plant-Microbe Interactions 3, 157-166.
Grayer,R.J. and Harborne,J.B. (1994). A survey of antifungal compounds from higher plants,
1982-1993. Phytochemistry 37, 19-42.
Grayer,R.J. and Kokubun,T. (2001). Plant-fungal interactions: the search for phytoalexins and
other antifungal compounds from higher plants. Phytochemistry 56, 253-263.
Grosskinsky,D.K., Naseem,M., Abdelmohsen,U.R., Plickert,N., Engelke,T., Griebel,T.,
Zeier,J., Novak,O., Strnad,M., Pfeifhofer,H., van der Graaff,E., Simon,U., and Roitsch,T.
(2011). Cytokinins mediate resistance against Pseudomonas syringae in tobacco through
increased antimicrobial phytoalexin synthesis independent of salicylic acid signaling. Plant
Physiology 157, 815-830.
Reference List
184
Grosskinsky,D.K., van der Graaff,E., and Roitsch,T. (2012). Phytoalexin transgenics in crop
protection-Fairy tale with a happy end? Plant Science 195, 54-70.
Grotewold,E. (2006). The Science of Flavonoids. (Springer).
Guo,S. and Kemphues,K.J. (1995). par-1, a gene required for establishing polarity in C.
elegans embryos, encodes a putative Ser/Thr kinase that is asymmetrically distributed. Cell
81, 611-620.
Gurr,S.J. and Rushton,P.J. (2005). Engineering plants with increased disease resistance: how
are we going to express it? Trends in Biotechnology 23, 283-290.
Hahlbrock,K. and Scheel,D. (1989). Physiology and molecular biology of phenylpropanoid
metabolism. Annual Review of Plant Physiology and Plant Molecular Biology. 40, 347–369.
Hain,R., Bieseler,B., Kindl,H., Schroder,G., and Stocker,R. (1990). Expression of a stilbene
synthase gene in Nicotiana tabacum results in synthesis of the phytoalexin resveratrol. Plant
Molecular Biology 15, 325-335.
Hain,R., Reif,H.J., Krause,E., Langebartels,R., Kindl,H., Vornam,B., Wiese,W.,
Schmelzer,E., Schreier,P.H., Stocker,R.H., and Stenzel,K. (1993). Disease resistance results
from foreign phytoalexin expression in a novel plant. Nature 361, 153-156.
Halpin,C. (2005). Gene stacking in transgenic plants - the challenge for 21st century plant
biotechnology. Plant Biotechnology Journal 3, 141-155.
Hamberger,B., Ellis,M., Friedmann,M., Souza,C.D.A., Barbazuk,B., and Douglas,C.J. (2007).
Genome-wide analyses of phenylpropanoid-related genes in Populus trichocarpa,
Arabidopsis thaliana, and Oryza sativa: the Populus lignin toolbox and conservation and
diversification of angiosperm gene families. Canadian Journal of Botany-Revue Canadienne
de Botanique 85, 1182-1201.
Hao,J.J., Subbarao,K.V., and Duniway,J.M. (2003). Germination of Sclerotinia minor and S.
sclerotiorum sclerotia under various soil moisture and temperature combinations.
Phytopathology 93, 443-450.
Harborne,J.B., Baxter,H., Moss,G.P., eds. (1999). Phytochemical Dictionary: A Handbook of
Bioactive Compounds from Plants, 2nd ed. London. (Taylor & Francis).
He,X.Z. and Dixon,R.A. (2000). Genetic manipulation of isoflavone 7-O-methyltransferase
enhances biosynthesis of 4 '-O-methylated isoflavonoid phytoalexins and disease resistance in
alfalfa. Plant Cell 12, 1689-1702.
Heilersig,H.J.B., Loonen,A., Bergervoet,M., Wolters,A.M.A., and Visser,R.G.F. (2006). Post-
transcriptional gene silencing of GBSSI in potato: Effects of size and sequence of the inverted
repeats. Plant Molecular Biology 60(5), 647-662.
Hejduk, S. (2011). Differences in persistence in red clover cultivars as an important trait for
their use in permanent grasslands. Grassland Science in Europe 16, 76-78.
Reference List
185
Hellemans,J., Mortier,G., De Paepe,A., Speleman,F., and Vandesompele,J. (2007). qBase
relative quantification framework and software for management and automated analysis of
real-time quantitative PCR data. Genome Biology 8(2), R19.
Higgins,V.J. and Smith,D.G. (1972). Separation and identification of two pterocarpanoid
phytoalexins produced by red clover leaves. Phytopathology 62, 235-238.
Hipskind,J.D. and Paiva,N.L. (2000). Constitutive accumulation of a resveratrol-glucoside in
transgenic alfalfa increases resistance to Phoma medicaginis . Molecular Plant-Microbe
Interactions 13, 551-562.
Ho,C.K., Chang,S.H., Tsay,J.Y., Tsai,C.J., Chiang,V.L., and Chen,Z.Z. (1998).
Agrobacterium tumefaciens-mediated transformation of Eucalyptus camaldulensis and
production of transgenic plants. Plant Cell Reports 17, 675–680.
Hu,R.B., Fan,C.M., Li,H.Y., Zhang,Q.Z., and Fu,Y.F. (2009). Evaluation of putative
reference genes for gene expression normalization in soybean by quantitative real-time RT-
PCR. Bmc Molecular Biology 10.
Huang,H.C., Bremer,E., Hynes,R.K., and Erickson,R.S. (2000). Foliar application of fungal
biocontrol agents for the control of white mold of dry bean caused by Sclerotinia
sclerotiorum. Biological Control 18, 270-276.
Ingham,J.L. (1973). Disease resistance in higher plants the concept of pre-infectional and
post-infectional resistance. Journal of Phytopathology 78, 314-335.
Iriti,M. and Faoro,F. (2009). Chemical diversity and defence metabolism: How plants cope
with pathogens and ozone pollution. International Journal of Molecular Sciences 10, 3371-
3399.
Jaiwal,P.K. and Singh,R.P. (2003). Applied genetics of leguminosae biotechnology.
(Dordrecht: Kluwer Academic Publishers).
Jambunathan,R., Butler,L.G., Bandyopadhyay,R., and Mughogho,L.K. (1986). Polyphenol
concentrations in grain, leaf, and callus tissues of mold-susceptible and mold-resistant
sorghum cultivars. Journal of Agricultural and Food Chemistry 34, 425–429.
Jefferson,R. (1987). Assaying chimeric genes in plants: The GUS gene fusion system. Plant
Molecular Biology Reporter 5, 387-405.
Jefferson,R.A., Burgess,S.M., and Hirsh,D. (1986). Beta-glucuronidase from Escherichia coli
as a gene-fusion marker. Proceedings of the National Academy of Sciences of the United
States of America 83, 8447-8451.
Jefferson,R.A., Kavanagh,T.A., and Bevan,M.W. (1987). Beta-glucuronidase (GUS) as a
sensitive and versatile gene fusion marker in higher plants. EMBO J. 6, 3901-3907.
Jung,W., Chung,I.M., and Heo,H.Y. (2003). Manipulating isoflavone levels in plants. Plant
Biotechnology Journal 5, 149-155.
Reference List
186
Jung,W., Yu,O., Lau,S.M.C., O'Keefe,D.P., Odell,J., Fader,G., and McGonigle,B. (2000).
Identification and expression of isoflavone synthase, the key enzyme for biosynthesis of
isoflavones in legumes. Nature Biotechnology 18, 208-212.
Kapila,J., DeRycke,R., VanMontagu,M., and Angenon,G. (1997). An Agrobacterium-
mediated transient gene expression system for intact leaves. Plant Science 122, 101-108.
Karimi,M., Inze,D., and Depicker,A. (2002). Gateway (TM) vectors for Agrobacterium-
mediated plant transformation. Trends in Plant Science 7, 193-195.
Kasai,M. and Kanazawa,A. (2012). RNA silencing as a tool to uncover gene function and
engineer novel traits in soybean. Breeding Science 61, 468-479.
Kato,T., Kabuto,C., Sasaki,N., Tsunagaw,M., Aizawa,H., Fujita,K., Kato,Y., and Kitahara,Y.
(1973). Momilactones, growth inhibitors from rice, Oryza sativa L. Tetrahedron Letters, 39,
3861-3864.
Kaurt, N. and Murphy, J. B. (2012). Enhanced isoflavone biosynthesis in transgenic cowpea
(Vigna unguiculata l.) callus. Plant Molecular Biology and Biotechnology 3(1), 1-8.
Keen,N.T. and Holliday, M.J. (1982) Recognition of bacterial pathogens by plants. In Mount,
M.S. and Lay, G. eds, Phytopathogenic Prokaryotes, Vol 2. (New York: Academic Press), pp
179-215.
Keyes,G.J., Collins,G.B., and Taylor,N.L. (1980). Genetic variation in tissue cultures of red
clover. Theoretical and Applied Genetics 58, 265-271.
Kim,B.G., Kim,S.Y., Song,H.S., Lee,C., Hur,H.G., Kim,S.I., and Ahn,J.H. (2003). Cloning
and expression of the isoflavone synthase gene (IFS-Tp) from Trifolium pratense. Molecules
and Cells 15(3), 301-306.
Kishimoto,N., Nagai,J., Kinoshita,T., Ueno,K., Ohashi,Y., and Mitsuhara,I. (2013). DNA
elements reducing transcriptional gene silencing revealed by a novel screening strategy. Plos
One 8(1).
Kliebenstein,D.J. and Rowe,H.C. (2008). Ecological costs of biotrophic versus necrotrophic
pathogen resistance, the hypersensitive response and signal transduction. Plant Science 174,
551-556.
Kodama,O., Li,W.X., Tamogami,S., and Akatsuka,T. (1992). Oryzalexin S, a novel
stemarane-type diterpene rice phytoalexin. Bioscience Biotechnology and Biochemistry 56,
1002-1003.
Kohn,L.M. (1979a). A monographic revision of the genus Sclerotinia. Mycotaxon 2, 364-
444.
Kohn,L.M. (1979b). Delimitation of the economically important plant pathogenic Sclerotinia
species. Phytopathology 69, 881-886.
Komatsuda,T., Enomoto,S., and Nakajima,K. (1989). Genetics of callus proliferation and
shoot differentiation in barley. Journal of Heredity 80, 345-350.
Reference List
187
Kortner,T.M., Valen,E.C., Kortner,H., Marjara,I.S., Krogdahl,A., and Bakke,A.M. (2011).
Candidate reference genes for quantitative real-time PCR (qPCR) assays during development
of a diet-related enteropathy in Atlantic salmon (Salmo salar L.) and the potential pitfalls of
uncritical use of normalization software tools. Aquaculture 318(3), 355-363.
Koukol,J. and Conn,E.E. (1961). The metabolism of aromatic compounds in higher plants.
IV. Purification and properties of the phenylalanine deaminase of Hordeum vulgare . J. Biol.
Chem. 236, 2692-2698.
Kramer,R.P., Hindorf,H., Jha,H.C., Kallage,J., and Zilliken,F. (1984). Antifungal activity of
soybean and chickpea isoflavones and their reduced derivatives. Phytochemistry 23, 2203-
2205.
Kris,M.H.S. and Bingham,E.T. (1988). Interactions of highly regenerative genotypes of
alfalfa (Medicago sativa L.) and tissue culture protocols. In Vitro Cellular & Developmental
Biology 24, 1047-1052.
Krishnamurthy,K.V., Suhasini,K., Sagare,A.P., Meixner,M., de Kathen,A., Pickardt,T., and
Schieder,O. (2000). Agrobacterium mediated transformation of chickpea (Cicer arietinum L.)
embryo axes. Plant Cell Reports 19, 235-240.
Krysan,P.J., Young,J.C., and Sussman,M.R. (1999). T-DNA as an insertional mutagen in
Arabidopsis. Plant Cell 11, 2283-2290.
Lawrence,M.J. (1996). Number of incompatibility alleles in clover and other species.
Heredity 76, 610-615.
Lee,C.W., Yoneyama,K., Takeuchi,Y., Konnai,M., Tamogami,S., and Kodama,O. (1999).
Momilactones A and B in rice straw harvested at different growth stages. Bioscience
Biotechnology and Biochemistry 63, 1318–1320.
Li,J., Brader,G., and Palva,E.T. (2004). The WRKY70 transcription factor: A node of
convergence for jasmonate-mediated and salicylate-mediated signals in plant defense. Plant
Cell 16, 319-331.
Li,J., Ou-Lee,T.M., Raba,R., Amundson,R.G., and Last,R.L. (1993). Arabidopsis flavonoid
mutants are hypersensitive to UV-B irradiation. Plant Cell 5, 171-179.
Lingel,A. and Sattler,M. (2005). Novel modes of protein-RNA recognition in the RNAi
pathway. Current Opinion in Structural Biology 15, 107-115.
Liu,C.J., Blount,J.W., Steele,C.L., and Dixon,R.A. (2002a). Bottlenecks for metabolic
engineering of isoflavone glycoconjugates in Arabidopsis. Proceedings of the National
Academy of Sciences of the United States of America 99, 14578-14583.
Liu,J., Benedict,C.R., Stipanovic,R.D., Magill,C.W., and Bell,A.A. (2002b). Cloning and
expression of desoxyhemigossypol-6-O-methyltransferase from cotton (Gossypium
barbadense). Journal of Agricultural and Food Chemistry 50(11), 3165-3172.
Livak, KJ. (1997). ABI Prism 7700 Sequence detection System User Bulletin #2 Relative
quantification of gene expression. ABI company publication. Ref Type: Report.
Reference List
188
Lois,R. (1994). Accumulation of UV-absorbing flavonoids induced by UV-B radiation in
Arabidopsis thaliana L .1. Mechanisms of UV-resistance in Arabidopsis . Planta 194, 498-
503.
Long,S.R. (1989). Rhizobium-legume nodulation: life together in the underground. Cell 56,
203-214.
Lovdal,T. and Lillo,C. (2009). Reference gene selection for quantitative real-time PCR
normalization in tomato subjected to nitrogen, cold, and light stress. Analytical Biochemistry
387, 238-242.
Lozovaya,V.V., Lygin,A.V., Zernova,O.V., Ulanov,A.V., Li,S.X., Hartman,G.L., and
Widholm,J.M. (2007). Modification of phenolic metabolism in soybean hairy roots through
down regulation of chalcone synthase or isoflavone synthase. Planta 225, 665-679.
Luo,D.H., Kohlway,A., and Pyle,A.M. (2013). Duplex RNA activated ATPases (DRAs)
platforms for RNA sensing, signaling and processing. Rna Biology 10, 111-120.
Mace,M.E., Stipanovic,R.D., and Bell,A.A. (1985). Toxicity and role of terpenoid
phytoalexins in verticillium wilt resistance in cotton. Physiologial Plant Pathology 26, 209-
218.
Maclean,N.L. and Nowak,J. (1989). Plant regeneration from hypocotyl and petiole callus of
Trifolium pratense L. Plant Cell Reports 8, 395-398.
Manavella,P.A., Koenig,D., and Weigel,D. (2012). Plant secondary siRNA production
determined by microRNA-duplex structure. Proceedings of the National Academy of Sciences
of the United States of America 109, 2461-2466.
Manczinger, L., Antal, Z., and Kredics, L. (2002). Ecophysiology and breeding of
mycoparasitic Trichoderma strains. Acta Microbiol Immunol Hung. 49(1), 1-14.
Mann,J. (1987). Secondary Metabolism. (Oxford: Oxford Science Publications).
Mansfield, J.W. (1982). The role of phytoalexins in disease resistance. In Bailey, J.A. and
Mansfield, J.W., eds, Phytoalexins. (New York: John Wiley & Sons) pp 253-288.
Matern,U., Heller,W., and Himmelspach,K. (1983). Conformational changes of apigenin 7-O-
(6-O-Malonylglucoside), a vacuolar pigment from parsley, with solvent composition and
proton concentration. European Journal of Biochemistry 133, 439-448.
Mathesius,U., Schlaman,H.R., Spaink,H.P., Of,S.C., Rolfe,B.G., and Djordjevic,M.A. (1998).
Auxin transport inhibition precedes root nodule formation in white clover roots and is
regulated by flavonoids and derivatives of chitin oligosaccharides. Plant J. 14, 23-43.
McGee,J.D., Williams,E.G., Collins,G.B., and Hildebrand,D.F. (1989). Somatic
embryogenesis in Trifolium: Protein profiles associated with high-frequency and low-
frequency regeneration. Journal of Plant Physiology 135, 306-312.
McMurray,C.H., Laidlaw,A.S., and McElroy,M. (1986). The effect of plant development and
environment on formononetin concentration in red clover (Trifolium pratense L.). Journal of
the Science of Food and Agriculture 37, 333–340.
Reference List
189
Mehrotra,R.S. and Aneja,K.R. (1990). An Introduction To Mycology. (New Delhi: New age
international (P) Ltd., publishers).
Mello,C.C. and Conte,D. (2004). Revealing the world of RNA interference. Nature 431, 338-
342.
Mengiste,T., Chen,X., Salmeron,J., and Dietrich,R. (2003). The botrytis susceptible1 gene
encodes an R2R3Myb transcription factor protein that is required for biotic and abiotic stress
responses in Arabidopsis. Plant Cell 15, 2551-2565.
Misyura,M., Colasanti,J., and Rothstein,S.J. (2013). Physiological and genetic analysis of
Arabidopsis thaliana anthocyanin biosynthesis mutants under chronic adverse environmental
conditions. Journal of Experimental Botany 64, 229-240.
Monde,K. and Takasugi,M. (1992). Studies on stress metabolites: XIV. High-performance
liquid chromatographic analysis of cruciferous phytoalexins using complex ternary mobile
phase gradients. Journal of Chromatography 598(1), 147-152.
Mori,M., Tomita,C., Sugimoto,K., Hasegawa,M., Hayashi,N., Dubouzet,J.G., Ochiai,H.,
Sekimoto,H., Hirochika,H., and Kikuchi,S. (2007). Isolation and molecular characterization
of a Spotted leaf 18 mutant by modified activation-tagging in rice. Plant Molecular Biology
63, 847-860.
Morise,H., Shimomur.O, Johnson,F.H., and Winant,J. (1974). Intermolecular Energy Transfer
in Bioluminescent System of Aequorea. Biochemistry 13(12), 2656-2662.
Morse,D.L., Carroll,D., Weberg,L., Borgstrom,M.C., Ranger-Moore,J., and Gillies,R.J.
(2005). Determining suitable internal standards for mRNA quantification of increasing cancer
progression in human breast cells by real-time reverse transcriptase polymerase chain
reaction. Analytical Biochemistry 342, 69-77.
Mothes,K., Schütte,H.R., and Luckner,M. (1985). Biochemistry of alkaloids. (Berlin:
Deutscher Verlag der Wissenschaften).
Mullen, J. (2001). Southern blight, Southern stem blight, White mold. The Plant Health
Instructor.
Müller,K.O. (1956). Einige einfache versuche zum nachweis von phytoalexinen.
Phytopathologische Zeitschrift 27, 237–254.
Müller,K.O. and Börger,H. (1940). Experimentelle untersuchungen über die Phythophthora-
resistenz der kartoffel. Arbeiten aus der biolgischen Bundensanstalt fur Land- und
Forstwirtschaft 23, 189-231.
Murashige,T. and Skoog,F. (1962). Revised medium for rapid growth and bioassays with
tobacco tissue cultures. Physiol. Plant. 15, 473-497.
Myers,J.R., Grosser,J.W., Taylor,N.L., and Collins,G.B. (1989). Genotype-dependent whole
plant regeneration from protoplasts of red clover (Trifolium pratense L.). Plant Cell Tissue
and Organ Culture 19(2), 113-127.
Reference List
190
Nadolska-Orczyk,A. and Orczyk,W. (2000). Study of the factors influencing Agrobacterium-
mediated transformation of pea (Pisum sativum L.). Molecular Breeding 6, 185-194.
Nafisi,M., Goregaoker,S., Botanga,C.J., Glawischnig,E., Olsen,C.E., Halkier,B.A., and
Glazebrook,J. (2007). Arabidopsis cytochrome P450 monooxygenase 71A13 catalyzes the
conversion of indole-3-acetaldoxime in camalexin synthesis. Plant Cell 19, 2039-2052.
Nakahara,K.S., Masuta,C., Yamada,S., Shimura,H., Kashihara,Y., Wada,T.S., Meguro,A.,
Goto,K., Tadamura,K., Sueda,K., Sekiguchi,T., Shao,J., Itchoda,N., Matsumura,T.,
Igarashi,M., Ito,K., Carthew,R.W., and Uyeda,I. (2012). Tobacco calmodulin-like protein
provides secondary defense by binding to and directing degradation of virus RNA silencing
suppressors. Proceedings of the National Academy of Sciences of the United States of
America 109(25), 10113-10118.
Napoli,C., Lemieux,C., and Jorgensen,R. (1990). Introduction of a chimeric chalcone
synthase gene Into petunia results in reversible co-suppression of homologous genes in trans.
Plant Cell 2, 279-289.
Nellen,W. and Lichtenstein,C. (1993). What makes an mRNA anti-sensi-itive? Trends in
Biochemical Sciences 18, 419-423.
Öhberg, H. (2008). Studies of the persistence of red clover cultivars in Sweden with particular
reference to Sclerotinia trifoliorum Erikss. Swedish University of Agricultural Sciences
Umeå. Ref Type: Thesis/Dissertation.
Öhberg,H. and Bang,U. (2010). Biological control of clover rot on red clover by
Coniothyrium minitans under natural and controlled climatic conditions. Biocontrol Science
and Technology 20, 25-36.
Olhoft,P.M., Flagel,L.E., Donovan,C.M., and Somers,D.A. (2003). Efficient soybean
transformation using hygromycin B selection in the cotyledonary-node method. Planta 216(5),
723-735.
Orlikova,B., Tasdemir,D., Golais,F., Dicato,M., and Diederich,M. (2011). Dietary chalcones
with chemopreventive and chemotherapeutic potential. Genes and Nutrition 6, 125-147.
Ow,D.W., Wood,K.V., Deluca,M., Dewet,J.R., Helinski,D.R., and Howell,S.H. (1986).
Transient and stable expression of the firefly luciferase gene in plant-cells and transgenic
plants. Science 234(4778), 856-859.
Ozel,C.A., Khawar,K.M., and Arslan,O. (2008). A comparison of the gelling of isubgol, agar
and gelrite on in vitro shoot regeneration and rooting of variety Samsun of tobacco (Nicotiana
tabacum L.). Scientia Horticulturae 117(2), 174-181.
Pang,Q. and Hays,J.B. (1991). UV-B-inducible and temperature-sensitive photoreactivation
of cyclobutane pyrimidine dimers in Arabidopsis thaliana . Plant Physiol 95(2), 536-543.
Pelch,K.E., Schroder,A.L., Kimball,P.A., Sharpe-Timms,K.L., Davis,J.W., and Nagel,S.C.
(2010). Aberrant gene expression profile in a mouse model of endometriosis mirrors that
observed in women. Fertility and Sterility 93, 1615-1627.
Reference List
191
Perry,N.B. and Foster,L.M. (1994). Antiviral and antifungal flavonoids, plus a triterpene from
Hebe cupressoides. Planta Med. 60(5), 491-492.
Peters,N.K., Frost,J.W., and Long,S.R. (1986). A plant flavone, luteolin, induces expression
of Rhizobium meliloti nodulation genes. Science 233(4767), 977-80.
Pfaffl,M.W. (2001). A new mathematical model for relative quantification in real-time RT-
PCR. Nucleic Acids Research 29(9).
Pfaffl,M.W., Tichopad,A., Prgomet,C., and Neuvians,T.P. (2004). Determination of stable
housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper -
Excel-based tool using pair-wise correlations. Biotechnology Letters 26, 509-515.
Phillips,G.C. and Collins,G.B. (1979). In vitro tissue culture of selected legumes and plant
regeneration from callus cultures of red clover. Crop Sci 19, 59-64.
Pierce, M. L., Cover, E. C., Richardson, P. E., Scholes, V. E., and Essenberg M. (1996).
Adequacy of cellular phytoalexin concentrations in hypersensitively responding cotton leaves.
Physiological and Molecular Plant Pathology 48, 305-324.
Pokorny, R., Andersson, B., Nedèlnk, J., and Riha, P. (2003). Current state of red clover
breeding for resistance in central and northern Europe. Czech J.Genet.Plant Breed. 39, 82-85.
Ponton,F., Chapuis,M.P., Pernice,M., Sword,G.A., and Simpson,S.J. (2011). Evaluation of
potential reference genes for reverse transcription-qPCR studies of physiological responses in
Drosophila melanogaster. Journal of Insect Physiology 57, 840-850.
Pueppke,S.G. (1996). The genetic and biochemical basis for nodulation of legumes by
rhizobia. Crit Rev. Biotechnol. 16(1), 1-51.
Punja,Z.K. (1985). The Biology, Ecology, and Control of Sclerotium rolfsii. Annual Review
of Phytopathology 23, 97-127.
Punja,Z.K. and Raharjo,S.H.T. (1996). Response of transgenic cucumber and carrot plants
expressing different chitinase enzymes to inoculation with fungal pathogens. Plant Disease
80, 999-1005.
Quesenberry,K.H. and Smith,R.R. (1993). Recurrent selection for plant regeneration from red
clover tissue culture. Crop Sci 33(3), 585-589.
Quesenberry,K.H., Wofford,D.S., Smith,R.L., Krottje,P.A., and Tcacenco,F. (1996).
Production of red clover transgenic for neomycin phosphotransferase II using Agrobacterium.
Crop Sci 36(4), 1045-1048.
Quimio,C.A. and Zapata,F.J. (1990). Diallel analysis of callus induction and green-plant
regeneration in rice anther culture. Crop Sci 30(1), 188-192.
Radonic,A., Thulke,S., Mackay,I.M., Landt,O., Siegert,W., and Nitsche,A. (2004). Guideline
to reference gene selection for quantitative real-time PCR. Biochemical and Biophysical
Research Communications 313, 856-862.
Reference List
192
Rakoczy-Trojanowska,M. and Malepszy,S. (1993). Genetic factors influencing regeneration
ability in rye (Secale cereale L.). I. Immature inflorescences. Theoretical and Applied
Genetics 86, 406-410.
Ralston,L., Subramanian,S., Matsuno,M., and Yu,O. (2005). Partial reconstruction of
flavonoid and isoflavonoid biosynthesis in yeast using soybean type I and type II chalcone
isomerases. Plant Physiology 137, 1375-1388.
Ramos,G.P., Dias,P.M.B., Morais,C.B., Dall'Agnol,M., and Zuanazzi,J.A.S. (2012). Genetic
variability of isoflavones in the USDA red clover core collection. Revista Brasileira de
Farmacognosia-Brazilian Journal of Pharmacognosy 22, 1241-1252.
Raynal, G. and Picard, J. (1985). Laboratory production of the sexual stage of Sclerotinia
trifoliorum, the winter crown-rot agent of red clover, consequences for improving control
methods. Proc.Intern.Grassl.Cong. 15, 782-783.
Reisch,B. and Bingham,E.T. (1980). The genetic control of bud formation from callus
cultures of diploid alfalfa. Plant Science Letters 20, 71-77.
Remans,T., Smeets,K., Opdenakker,K., Mathijsen,D., Vangronsveld,J., and Cuypers,A.
(2008). Normalisation of real-time RT-PCR gene expression measurements in Arabidopsis
thaliana exposed to increased metal concentrations. Planta 227, 1343-1349.
Ren,D.T., Liu,Y.D., Yang,K.Y., Han,L., Mao,G.H., Glazebrook,J., and Zhang,S.Q. (2008). A
fungal-responsive MAPK cascade regulates phytoalexin biosynthesis in Arabidopsis.
Proceedings of the National Academy of Sciences of the United States of America 105, 5638-
5643.
Rivard, G. L. (2010). Grafting for open-field and high tunnel tomato production. North
Carolina State University. Ref Type: Thesis/Dissertation.
Robberecht,R. and Caldwell,M.M. (1983). Protective mechanisms and acclimation to solar
ultraviolet-B radiation in Oenothera stricta . Plant Cell and Environment 6, 477-485.
Roitsch,T., Balibrea,M.E., Hofmann,M., Proels,R., and Sinha,A.K. (2003). Extracellular
invertase: key metabolic enzyme and PR protein. Journal of Experimental Botany 54, 513-
524.
Rollins,J.A. and Dickman,M.B. (1998). Increase in endogenous and exogenous cyclic AMP
levels inhibits sclerotial development in Sclerotinia sclerotiorum. Applied and Environmental
Microbiology 64, 2539-2544.
Romano,N. and Macino,G. (1992). Quelling: Transient inactivation of gene expression in
Neurospora Crassa by transformation with homologous sequences. Molecular Microbiology
6, 3343-3353.
Ruhland,C.T., Xiong,F.S., Clark,W.D., and Day,T.A. (2005). The influence of ultraviolet-B
radiation on growth, hydroxycinnamic acids and flavonoids of Deschampsia antarctica during
springtime ozone depletion in antarctica. Photochemistry and Photobiology 81, 1086–1093.
Reference List
193
Rytkonen,K.T., Renshaw,G.M.C., Ashton,K.J., Williams-Pritchard,G., Leder,E.H., and
Nikinmaa,M. (2010). Elasmobranch qPCR reference genes: a case study of hypoxia
preconditioned epaulette sharks. Bmc Molecular Biology 11, 27.
Sauer, R. (1984). On the importance of clover rot Sclerotinia trifoliorum Erikss on red clover
and possibilities for control with vinclozolin. Mededelingen van de Faculteit
Landbouwwetenschappen.Rijksuniversiteit Gent. 49, 277-282.
Saviranta,N.M., Anttonen,M.J., von Wright,A., and Karjalainen,R.O. (2008). Red clover
(Trifolium pratense L.) isoflavones: determination of concentrations by plant stage, flower
colour, plant part and cultivar. Journal of the Science of Food and Agriculture 88, 125-132.
Sawada,Y., Kinoshita,K., Akashi,T., Aoki,T., and Ayabe,S. (2002). Key amino acid residues
required for aryl migration catalyzed by the cytochrome P450 2-hydroxyisoflavone synthase.
Plant J. 31, 555–564.
Sayaka, H. and Kodama, H. (2008). RNAi vectors for manipulation of gene expression in
higher plants. The Open Plant Science Journal 2, 21-30.
Schenk,R.U. and Hildebrandt,A.C. (1972). Medium and techniques for induction and growth
of monocotyledonous and dicotyledonous plant cell cultures. Canadian Journal of Botany 50,
199-204.
Schmelzer,E., Jahnen,W., and Hahlbrock,K. (1988). In situ localization of light-induced
chalcone synthase mRNA, chalcone synthase, and flavonoid end products in epidermal cells
of parsley leaves. Proceedings of the National Academy of Sciences of the United States of
America 85, 2989-2993.
Schmidt,M. (1933). Zur entwicklung sphysiologie von cladosporium fulvum und über die
widerstandsfähigkeit von solanum racemigerum gegen diesen parasiten. Planta 20(3), 407-
439.
Schmittgen,T.D., Zakrajsek,B.A., Mills,A.G., Gorn,V., Singer,M.J., and Reed,M.W. (2000).
Quantitative reverse transcription-polymerase chain reaction to study mRNA decay:
Comparison of endpoint and real-time methods. Analytical Biochemistry 285, 194-204.
Schnabl,H., Weissenbock,G., and Scharf,H. (1986). In vivo-microspectrophotometric
characterization of flavonol glycosides in Vicia Faba guard and epidermal cells. Journal of
Experimental Botany 37, 61-72.
Schnablova,R., Synkova,H., Vicankova,A., Burketova,L., Eder,J., and Cvikrova,M. (2006).
Transgenic ipt tobacco overproducing cytokinins overaccumulates phenolic compounds
during in vitro growth. Plant Physiology and Biochemistry 44, 526-534.
Schnettler, E., Sterken, M. G., Leung, J. Y., Metz, S. W., Geertsema, C., Goldbach, R. W.,
Vlak, J. M., Khromykh, A. A., and Pijlman, G. P. (2012). Noncoding flavivirus RNA displays
RNAi suppressor activity in insect and mammalian cells. Journal of Virology 86(24), 13486-
13500.
Sen,G.L. and Blau,H.M. (2006). A brief history of RNAi: the silence of the genes. Faseb
Journal 20, 1293-1299.
Reference List
194
Serrano,M., Moreno-Sanchez,N., Gonzalez,C., Marcos-Carcavilla,A., Van Poucke,M.,
Calvo,J.H., Salces,J., Cubero,J., and Carabano,M.J. (2011). Use of maximum likelihood-
mixed models to select stable reference genes: a case of heat stress response in sheep. BMC
Molecular Biology 12, 36.
Sharma,K.K. and Anjaiah,V. (2000). An efficient method for the production of transgenic
plants of peanut (Arachis hypogaea L.) through Agrobacterium tumefaciens-mediated genetic
transformation. Plant Science 159(1), 7-19.
Shevell, J. L. (1985). Phytoalexin production in congenic cotton lines challenged with races of
Xanthomonas campestris pv. malvacearum . MS thesis. Stillwater, Oklahoma State Univ. Ref
Type: Thesis/Dissertation.
Sivesind,E. and Seguin,P. (2006). Effects of foliar application of elicitors on red clover
isoflavone content. Journal of Agronomy and Crop Science 192(1), 50-54.
Skipp,R.A. and Christensen,M.J. (1990). Selection for persistence in red clover: Influence of
root disease and stem nematode. New Zealand Journal of Agricultural Research 33, 319-333.
Smith,N.A., Singh,S.P., Wang,M.B., Stoutjesdijk,P.A., Green,A.G., and Waterhouse,P.M.
(2000). Gene expression: Total silencing by intron-spliced hairpin RNAs. Nature 407, 319-
320.
Smith, R. R. and Hay, V. W. (1983). Breeding for disease resistance in red clover.
Proceedings of the XIV International Grassland Congress, Lexington, Kentucky , 110-113.
Ref Type: Conference Proceeding
Smith,R.R. and Quesenberry,K.H. (1995). Registration of NewRC red clover germplasm.
Crop Sci 35,295.
Smith,R.R., Taylor,N.L., and Bowley,S.R. (1985). Red clover. N.L.Taylor, ed. (Madison:
American society of agronomy).
Sobolev,V.S. (2008). Localized production of phytoalexins by peanut (Arachis hypogaea)
kernels in response to invasion by Aspergillus species. Journal of Agricultural and Food
Chemistry 56, 1949-1954.
Spendley,P.J., Bird,P.M., Ride,J.P., and Leworthyt,D.P. (1982). Two novel antifungal alka-
2,4-dienals from Triticum aestivum. Phytochemistry 21, 2403–2404.
Stam,M., Mol,J.N.M., and Kooter,J.M. (1997). The silence of genes in transgenic plants.
Annals of Botany 79, 3-12.
Stapleton,A.E. (1992). Ultraviolet radiation and plants: burning questions. Plant Cell 4, 1353-
1358.
Steele,C.L., Gijzen,M., Qutob,D., and Dixon,R.A. (1999). Molecular characterization of the
enzyme catalyzing the aryl migration reaction of isoflavonoid biosynthesis in soybean.
Archives of Biochemistry and Biophysics 367, 146-150.
Reference List
195
Stoutjesdijk,P.A., Singh,S.P., Liu,Q., Hurlstone,C.J., Waterhouse,P.A., and Green,A.G.
(2002). hpRNA-mediated targeting of the Arabidopsis FAD2 gene gives highly efficient and
stable silencing. Plant Physiology 129, 1723-1731.
Stuitje,A.R., Verbree,E.C., van der Linden,K.H., Mietkiewska,E.M., Nap,J.P., and
Kneppers,T.J.A. (2003). Seed-expressed fluorescent proteins as versatile tools for easy
(co)transformation and high-throughput functional genomics in Arabidopsis. plant
biotechnology journal 1, 301-309.
Subramanian,S., Graham,M.Y., Yu,O., and Graham,T.L. (2005). RNA interference of
soybean isoflavone synthase genes leads to silencing in tissues distal to the transformation site
and to enhanced susceptibility to Phytophthora sojae. Plant Physiology 137, 1345-1353.
Subramanian,S., Stacey,G., and Yu,O. (2006). Endogenous isoflavones are essential for the
establishment of symbiosis between soybean and Bradyrhizobium japonicum . Plant Journal
48, 261-273.
Subramanian,S., Stacey,G., and Yu,O. (2007). Distinct, crucial roles of flavonoids during
legume nodulation. Trends in Plant Science 12, 282-285.
Sullivan,M.L. (2009a). Phenylalanine ammonia lyase genes in red clover: Expression in
whole plants and in response to yeast fungal elicitor. Biologia Plantarum 53, 301-306.
Sullivan,M.L. (2009b). A novel red clover hydroxycinnamoyl transferase has enzymatic
activities consistent with a role in phaselic acid biosynthesis. Plant Physiology 150, 1866-
1879.
Sullivan,M.L. and Hatfield,R.D. (2006). Polyphenol oxidase and o-diphenols inhibit
postharvest proteolysis in red clover and alfalfa. Crop Sci 46, 662–670.
Surekha,C., Arundhati,A., and Seshagiri Rao,G. (2007). Differential response of Cajanus
cajan varieties to transformation with different strains of Agrobacterium. Journal of
Biological Sciences, 7, 176–181.
Suzuki,T., Higgins,P.J., and Crawford,D.R. (2000). Control selection for RNA quantitation.
Biotechniques 29, 332-337.
Swevers,L., Liu,J.S., Huvenne,H., and Smagghe,G. (2011). Search for limiting factors in the
RNAi pathway in silkmoth tissues and the Bm5 cell line: The RNA-binding proteins R2D2
and translin. Plos One 6(5).
Szabo,A., Perou,C.M., Karaca,M., Perreard,L., Quackenbush,J.F., and Bernard,P.S. (2004).
Statistical modeling for selecting housekeeper genes. Genome Biology 5, R59.
Tang,G.L., Galili,G., and Zhuang,X. (2007). RNAi and microRNA: breakthrough
technologies for the improvement of plant nutritional value and metabolic engineering.
Metabolomics 3, 357-369.
Taylor,N.L. (2008). A century of clover breeding developments in the United States. Crop Sci
48(1), 1-13.
Reference List
196
Taylor,N.L., Kendall,W.A., and Stroube,W.H. (1968). Polycross-progeny testing of red clover
(Trifolium pratense L). Crop Sci 8(4), 451-454.
Taylor,N.L. and Quesenberry,K.H. (1996). Red clover science. (Dordrecht, Netherlands:
Kluwer Academic Publishers).
Taylor,S.G., Shilling,D.G., and Quesenberry,K.H. (1989). In vitro selection for 2,4-D
tolerance in red clover. TAG Theoretical and Applied Genetics 78, 265-270.
Tebayashi,S., Ishihara,A., and Iwamura,H. (2001). Elicitor-induced changes in isoflavonoid
metabolism in red clover roots. Journal of Experimental Botany 52(357), 681-689.
Thomzik,J.E., Stenzel,K., Stocker,R., Schreier,P.H., Hain,R., and Stahl,D.J. (1997). Synthesis
of a grapevine phytoalexin in transgenic tomatoes (Lycopersicon esculentum Mill.) conditions
resistance against phytophthora infestans. Physiological and Molecular Plant Pathology 51(4),
265-278.
Tian,L. and Dixon,R.A. (2006). Engineering isoflavone metabolism with an artificial
bifunctional enzyme. Planta 224(3),496-507.
Tiwari,S., Shanker,P., and Tripathi,M. (2004). Effects of genotype and culture medium on in
vitro androgenesis in soybean (Glycine max Merr.). Indian Journal of Biotechnology 3, 441-
444.
Tomás-Barberán,F.A., Msonthi,J.D., and Hostettmann,K. (1988). Antifungal epicuticular
methylated flavonoids from Helichrysum nitens. Phytochemistry 27(3), 753-755.
Tomás-Lorente,F., Iniesta-Sanmartin,E., Tomás-Barberán,F.A., Trowitzsch-Kienast,W., and
Wray,V. (1989). Antifungal phloroglucinol derivatives and lipophilic flavonoids from
Helichrysum decumbens. Phytochemistry 28(6), 1613-1615.
Traore,S.M. and Zhao,B.Y. (2011). A novel Gateway (R)-compatible binary vector allows
direct selection of recombinant clones in Agrobacterium tumefaciens. Plant Methods 7(1), 42.
Tu, C. C. and Kimbrough, J. W. (1978). Systematic and phylogeny of fungi in the Rhizoctonia
complex. Bot Gaz 139, 454-466.
Tverskoy,L., Dmitriev,A., Kozlovsky,A., and Grodzinsky,D. (1991). Two phytoalexins from
Allium cepa bulbs. Phytochemistry 30(3), 799-800.
Van Etten,H.D., Mansfield,J.W., Bailey,J.A., and Farmer,E.E. (1994). Two classes of plant
antibiotics: phytoalexins versus phytoanticipins. Plant Cell 6(9), 1191-1192.
Van Etten,H.D., Matthews,D.E., and Matthews,P.S. (1989). Phytoalexin detoxification:
Importance for pathogenicity and practical implications. Annual Review of Phytopathology
27, 143-164.
Vancanneyt,G., Schmidt,R., Oconnorsanchez,A., Willmitzer,L., and Rochasosa,M. (1990).
Construction of An Intron-Containing Marker Gene: Splicing of the intron in transgenic
plants and its use in monitoring early events in Agrobacterium-mediated plant transformation.
Molecular & General Genetics 220, 245-250.
Reference List
197
Vandesompele J, Kubista M, and Pfaffl MW (2009). Reference Gene Validation Software for
Improved Normalization. In Real-time PCR: current technology and applications, Logan J,
Edwards K, and Saunders N, eds. (Norfolk, UK: Caister Academic Press), 64.
Vandesompele,J., De Preter,K., Pattyn,F., Poppe,B., Van Roy,N., De Paepe,A., and
Speleman,F. (2002). Accurate normalization of real-time quantitative RT-PCR data by
geometric averaging of multiple internal control genes. Genome Biology 3(7).
Vanetten,H.D. (1976). Antifungal activity of pterocarpans and other selected isoflavonoids.
Phytochemistry 15(5), 655-659.
Vaucheret,H. and Fagard,M. (2001). Transcriptional gene silencing in plants: targets, inducers
and regulators. Trends in Genetics 17, 29-35.
Vetter,J. (1995). Isoflavones in different parts of common Trifolium species. Journal of
Agricultural and Food Chemistry 43(1), 106-108.
Virtanen,A.I., Hietala,P.K., and Wahlroos,O. (1957). Antimicrobial substances in cereals and
fodder plants. Arch. Biochem. Biophys. 69, 486-500.
Vleugels, T. (2013). Breeding for resistance to clover rot (Sclerotinia spp.) in red clover
(Trifolium pratense L.). Ghent University. Ref Type: Thesis/Dissertation.
Vleugels,T., Baert,J., De Riek,J., Heungens,K., Malengier,M., Cnops,G., and Van
Bockstaele,E. (2010). Diversity study on Sclerotinia trifoliorum Erikks., the causal agent of
clover rot in red clover crops (Trifolium pratense L.). Commun Agric Appl Biol Sci. 75, 649-
653.
Vogt,T. (2010). Phenylpropanoid Biosynthesis. Molecular Plant 3, 2-20.
Walti,M.A., Villalba,C., Buser,R.M., Grunler,A., Aebi,M., and Kunzler,M. (2006). Targeted
gene silencing in the model mushroom Coprinopsis cinerea (Coprinus cinereus) by
expression of homologous hairpin RNAs. Eukaryotic Cell 5, 732-744.
Wan,Y., Sorensen,E.L., and Liang,G.H. (1988). Genetic control of in vitro regeneration in
alfalfa (Medicago sativa L.). Euphytica 39, 3-9.
Wani, S. H., Sanghera, G. S., and Singh, N. B. (2010). Biotechnology and plant disease
control-role of RNA interference. American Journal of Plant Sciences 1, 55-68.
Webster,J. and Weber,R.W.S. (2007). Introduction to Fungi. (Cambridge, UK: Cambridge
University Press).
Wegulo,S.N., Yang,X.B., and Martinson,C.A. (1998). Soybean cultivar responses to
Sclerotinia sclerotiorum in field and controlled environment studies. Plant Disease 82, 1264-
1270.
Wegulo,S.N., Yang,X.B., Martinson,C.A., and Murphy,P.A. (2005). Effects of wounding and
inoculation with Sclerotinia sclerotiorum on isoflavone concentrations in soybean. Canadian
Journal of Plant Science 85, 749-760.
Reference List
198
Weidenborner,M., Hindorf,H., Jha,H.C., Tsotsonos,P., and Egge,H. (1990). Antifungal
activity of isoflavonoids in different reduced stages on Rhizoctonia solani and Sclerotium
rolfsii. Phytochemistry 29(3), 801-803.
Wesley,S.V., Helliwell,C.A., Smith,N.A., Wang,M.B., Rouse,D.T., Liu,Q., Gooding,P.S.,
Singh,S.P., Abbott,D., Stoutjesdijk,P.A., Robinson,S.P., Gleave,A.P., Green,A.G., and
Waterhouse,P.M. (2001). Construct design for efficient, effective and high-throughput gene
silencing in plants. Plant Journal 27, 581-590.
Whipps,J.M., Sreenivasaprasad,S., Muthumeenakshi,S., Rogers,C.W., and Challen,M.P.
(2008). Use of Coniothyrium minitans as a biocontrol agent and some molecular aspects of
sclerotial mycoparasitism. European Journal of Plant Pathology 121, 323-330.
White,R.P., Murray,S., and Rohweider,M. (2000). Pilot analysis of global ecosystems:
Grassland Ecosystems. (Washington DC: World Resources Institute).
Wink,M. (2010). Annual Plant Reviews, Biochemistry of Plant Secondary Metabolism.
Wiley-Blackwell).
Wippich,C. and Wink,M. (1985). Biological properties of alkaloids. Influence of
quinolizidine alkaloids and gramine on the germination and development of powdery mildew,
Erysiphe graminis f.sp. hordei . Experientia 41(11), 1477-1479.
Wittwer, CT, Herrmann, MG, Gundry, CN, and Elenitoba-Johnson, KS. (2001). Real-time
multiplex PCR assays. Methods 25(4), 430-442.
Wofford,D.S., Baltensperger,D.D., and Quesenberry,K.H. (1992). In vitro culture responses
of alyceclover genotypes on four media systems. Crop Sci 32(1), 261-265.
Wu, G. (2013). Plant MicroRNAs and Development. Journal of Genetics and Genomics 40,
217-230.
Wu,Q.D. and Vanetten,H.D. (2004). Introduction of plant and fungal genes into pea (Pisum
sativum L.) hairy roots reduces their ability to produce pisatin and affects their response to a
fungal pathogen. Molecular Plant-Microbe Interactions 17, 798-804.
Xu, Z. (2008). Overwinter survival of Sclerotium rolfsii and S. rolfsii var. delphinii, screening
hosta for resistance to S.rolfsii var. delphinii, and phylogenetic relationships among
Sclerotium species. Iowa State University. Ref Type: Thesis/Dissertation.
Yaqub,F. and Shahzad,S. (2005). In vitro evaluation of microbial antagonists against
Sclerotium rolfsii. Pakistan Journal of Botany 37, 1033-1036.
Yu,O. and Jez,J.M. (2008). Nature's assembly line: biosynthesis of simple phenylpropanoids
and polyketides. Plant Journal 54(4), 750-762.
Yu,O. and McGonigle,B. (2005). Metabolic engineering of isoflavone biosynthesis. Advances
in Agronomy 86, 147-190.
Yu,O., Shi,J., Hession,A.O., Maxwell,C.A., McGonigle,B., and Odell,J.T. (2003). Metabolic
engineering to increase isoflavone biosynthesis in soybean seed. Phytochemistry 63(7), 753-
763.
Reference List
199
Zhou,H.P. and Konzak,C.F. (1992). Genetic control of green plant regeneration from anther
culture of wheat. Genome 35(6), 957-961.
Zmora-Nahum,S., Danon,M., Hadar,Y., and Chen,Y. (2008). Chemical properties of compost
extracts inhibitory to germination of Sclerotium rolfsii. Soil Biology & Biochemistry 40(10),
2523-2529.
Curriculum vitae
201
Curriculum vitae
PERSONALIA
First Name Khosro
Last Name Mehdi Khanlou
Address Kortrijksesteenweg 502
9000 Ghent , Belgium
Telephone Number +32 485171075
E-mail [email protected]
Date of Birth February 23, 1973
Place of Birth Iran-Tehran
EDUCATION
PhD student in “Applied Biological Science”, Faculty of Bioscience engineering, department
of plant production, Ghent University Ghent, Belgium.
M.Sc. in “Plant Breeding” from Razi University, Iran, (2000).
The title of Master thesis “Evaluation of yield stability and drought resistance in
tetraploid & hexaploid lines of wheat”.
B.Sc. in “Agronomy and Plant Breeding” from Tehran University, Iran, (1996).
JOB EXPERIENCE
Working for Crops Department in Agriculture research center of Khuzestan province for two
years.
Teaching “plant breeding and statistics” to undergraduate students.
Curriculum vitae
202
BOOKS
Translating MSTAT-C user’s guide. 2003. Dibagaran Publication. ISBN: 964-354-429-X.
(From English into Persian)
Assistant in writing “Statistical designs in Agricultural Research”. 2001.
Dibagaran Publication. ISBN: 964-6966-85-3.
ARTICLES
- A critique of widely used normalization software tools and an alternative method to
identify reliable reference genes in red clover (Trifolium pratense L.). Planta 2012.
- Regeneration ability and genetic transformation of root type chicory (Cichorium intybus
var. sativum). AJB (2012).
- Improvement of Plant Regeneration and Agrobacterium-mediated Genetic
Transformation Efficiency in Red Clover (Trifolium pratense L.). JBR (2011).
- Towards an optimal sampling strategy for assessing genetic variation within and among
white clover (Trifolium repens L.) cultivars using AFLP. Genet Mol Biol. (2011).
- Evaluation drought resistance in tetraploid & hexaploid lines of wheat by principle
component analysis method. 7th “Crop Production & Plant Breeding” Congress in Iran
(2002).
- Drought resistance evaluation in tetraploid & hexaploid lines of wheat in vitro condition
with poly ethylene glycol 6000. 7th “Crop Production & Plant Breeding” Congress in
Iran (2002).
- Study on wheat lines adaptation by Russell and Eberhart method. 7th “Crop Production &
Plant Breeding” Congress in Iran (2002).
- Selection Superior lines using selection index method. 7th “Crop Production & Plant
Breeding” Congress in Iran (2002).
Curriculum vitae
203
ATTENDING AT SYMPOSIUMS, CONFERENCES AND WORKSHOPS
- Lab work experience (Workshop gene manipulation) in Rice Research Center of Rasht,
1998. Rasht, Iran.
- Workshop on DNA markers. in Rice Research Center of Rasht ,1998. Rasht, Iran.
- 6th Iranian Congress of Crop Production & Plant Breeding. Sep. 4-7. 2000. Karaj, Iran.
- 7th Iranian Congress of Crop Production & Plant Breeding. Sep. 3-6. 2002. Kara, Iran.
- Symposium of Agriculture in North-west of Iran at Islamic Azad University of
Mahabad, 2004. Mahabad, Iran.
- GMO Workshop, 7th February 2006, ILVO, Caritasstraat 21, 9090 Melle, Belgium.
- 12th Symposium on Applied Biological Sciences, September 21st, 2006. Ghent
University, Belgium.
- European Technical Seminar Tour, June 7, 2007, BIO-RAD. Mechelen, Belgium.
- Workshop on bluetongue virus real time RT-PCR, 16 March 2007. Department virology,
Section Development of Diagnostic Tools for Epizootic Diseases, Brussels, Belgium.
- RNA symposium18 October 2007, Applied Biosystems, Gosselies, Belgium. PCR/Real
Time workshop, 27 November 2007. Applied Biosystems, The Netherlands.
- qPCR user meeting, 11 April 2008. Roche Applied Sience, Vilvoorde, Belgium.
- First Benelux qPCR Symposium, 2008, Poster presentation Ghent, Belgium.
- 60nd International Symposium on Crop Protection, May 2008. Faculty of Bioscience
Engineering, Ghent University, Ghent, Belgium.
- The First European Conference of Iranian Scientists in Agriculture and Natural
Resources. October 2008, Paris, France.
- The First Biology Conference for Iranian Scholars in Europe, December 6-7, 2008
Hamburg, Germany.
- AgriGenomics World Congress Conference. 2-3 July 2009, Poster presentation, London,
UK.
- 62nd International Symposium on Crop Protection, May 2010. Faculty of Bioscience
Engineering, Ghent University, Ghent, Belgium.