Influence of nuclear and cytoplasmic factors in callus initiation in anther culture of wheatby Philip William Becraft

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science inAgronomyMontana State University© Copyright by Philip William Becraft (1987)

Abstract:Low success frequency limits the use of anther culture in crop breeding. Increasing the capacity ofcrops to undergo the anther culture procedure is one approach to improving success rates. Nuclear andcytoplasmic factors and cytoplasmic male sterility (cms) have been reported to affect tissue cultureresponse in wheat.

The objective of Part I was to assess the relative importance of nuclear and cytoplasmic factors anddetermine if cms enhanced callus yields in anther culture of wheat. Three nuclear genotypes, each innormal cytoplasm and two alien cytoplasms conferring cms, were evaluated for callus initiationfrequency. Cytoplasmic male sterility did not increase callus induction. The nuclear genotype of 'Chris'outperformed 'Butte' and 'Coteau' across all cytoplasms. Overall, normal Triticum aestivum andTriticum timopheevi cytoplasms outperformed Aeailops speltoides cytoplasm. Although nucleargenotype had the greatest effect, cytoplasm and nucleus X cytoplasm interaction were also significant.This suggests the best strategies for improving the capacity of wheat to initiate callus in anther cultureinvolve improving the nuclear capacity for callus initiation.

Heritability studies have suggested that variation in callus inducibility should respond to selection. Theobjective of Part II was to determine if crosses designed for improving callus initiation would producesegregation from which genotypes with improved callus inducibility could be selected. The winterwheat cultivars 'Centurk 78', 'Norstar' and 'Norwin' were chosen as parents based on their high callusinitiation frequencies and diverse genetic backgrounds. F2 populations from crosses between each ofthese parents were evaluated for callus initiation frequencies and callus vigor.

Heterosis was detected for callus initiation in the Norwin/Norstar cross and for callus vigor in allcrosses, but is believed to be of minor importance. Heritability estimates of .25 and .37 for callusinitiation in the Centurk 78/Norstar and Norwin/Norstar crosses, respectively, suggest substantialgenetic variation is available from these crosses for improving this trait. Heritability estimates forcallus vigor were less promising. 

INFLUENCE OF NUCLEAR AND CYTOPLASMIC FACTORS ON CALLUS INITIATION IN ANTHER CULTURE OF WHEAT

byPhilip William Becraft

A thesis submitted in partial fulfillment of the requirements for the degree

> ofMaster of Science

inAgronomy

MONTANA STATE UNIVERSITY Bozeman, Montana

May 1987

MAIN lib ./V37? Cop- oL

ii

APPROVAL

of a thesis submitted by

Philip William Becraft

This thesis has been read by each member of the thesis committee and has been found to be satisfactory regarding content, English usage, format, citations, bibliographic style, and consistency, and is ready for submission to the College of Graduate Studies.

Date Chairperson ,Graduate t/ommittee

Approved for the Major Department

Date Head, Major Department

Approved for the College of Graduate Studies

- /7Date Graduate Dean

iii

STATEMENT OF PERMISSION TO USE

In presenting this thesis in partial fulfillment of the requirements for a master's degree at Montana State University, I agree that the Library shall make it available to borrowers under rules of the Library. Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made.

Permission for extensive quotation from or reproduction of this thesis may be granted by my major professor, or in his absence, by the Director of Libraries when, in the opinion of either, the proposed use of the material is for scholarly purposes. Any copying or use of the material in this thesis for financial gain shall not be allowed without written permission.

Signature.Date

iv

ACKNOWLEDGMENTS

Special thanks goes to Dr. G. Allan Taylor for his friendship, support, guidance and wisdom while serving as my major advisor.

I also wish to express sincere thanks to the following:

Drs. J . M . Martin, T . K . Blake and E. R. Vyse for the help and advice they offered me as friends and members of my graduate committee.. Drs. S. S. Maan, K . A. Lucken and D . J. Cox of North Dakota State University, Fargo, and Dr. J. Gellner of South Dakota State University, Brookings, for kindly providing seed for this research.

Drs. M . D. Lazar, T . H. H. Chen and The Alberta Research Council for their hospitality and helpful technical advice.

USDA, which supported this research through a grant.Special thanks also goes to Diane Luth, soon to be my

wife, whose strength, understanding and encouragement made graduate school more tolerable.

TABLE OF CONTENTSPage

APPROVAL............... iiSTATEMENT OF PERMISSION TO USE....................... iiiACKNOWLEDGMENTS.......... ivTABLE OF CONTENTS........................... VLIST OF TABLES.................... viLIST OF FIGURES........................... viiABSTRACT.......................... ViiiINTRODUCTION............ IPART I. EFFECTS OF NUCLEUS, CYTOPLASM AND CYTOPLASMIC

MALE STERILITY ON CALLUS INITIATION IN ANTHER CULTURE OF WHEAT.......... ........ *....... 4

INTRODUCTION................ 5MATERIALS AND METHODS........................... 8RESULTS......... 10DISCUSSION......... 14

PART II. GENETIC VARIATION FOR ANTHER CALLUSINDUCIBILITY IN CROSSES OF HIGHLY CULTURABLE WINTER WHEATS ................................ 19

INTRODUCTION. ... ........................... 20MATERIALS AND METHODS............. 22RESULTS......................................... 25DISCUSSION...................................... 32

SUMMARY........ 36REFERENCES,.......................................... 39

V

. vi

LIST OF TABLES '

Table Page1.1. Mean callus induction percentages for

environments, nuclear genotypes, and cytoplasms. 111.2. Analysis of variance for nuclear and cytoplasmic

effects with data pooled across environments.... 121.3. Results of Nucleus-Cytoplasm combinations...... 12

11.1. Mean callus initiation frequencies of winterwheat cultivars tested for use as parents...... 25

11.2. Mean callus initiation frequencies and callus vigor ratings of the parent .and F2 populationsover all environments............. 26

11.3. Estimates of heterosis for callus initiation andcallus vigor in the F2 populations.......... 27

11.4. Broad sense heritability estimates for callusinitiation frequency and callus vigor........... 28

vii

LIST OF FIGURES

Figure PageI.I Performance of nuclear genotypes across

cytoplasms for callus initiation............... 1311.1 Percentage distribution of callus initiation

frequencies in the parent and F2 populations ofthe Centurk78/Norwin cross....... ............... 29

11.2 Percentage distribution of callus initiation frequencies in the parent and F2 populations ofthe Centufk78/Norstar cross... ................. 30

11.3 Percentage distribution of callus initiation frequencies in the parent and F2 populations ofthe Norwin/Norstar cross...... ...... -.......... 31

viii

ABSTRACT

Low success frequency limits the use of anther culture in crop breeding. Increasing the capacity of crops to undergo the anther culture procedure is one approach to improving success rates. Nuclear and cytoplasmic factors and cytoplasmic male sterility (cms) have been reported to affect tissue culture response in wheat.

The objective of Part I was to assess the relative importance of nuclear and cytoplasmic factors and determine if cms enhanced callus yields in anther culture of wheat. Three nuclear genotypes, each in normal cytoplasm and two alien cytoplasms conferring cms, were evaluated for callus initiation frequency. Cytoplasmic male sterility did not increase callus induction. The nuclear genotype of 'Chris' outperformed 'Butte' and 'Coteau' across all cytoplasms. Overall, normal Triticum aestivum and Triticum timopheevi cytoplasms outperformed Aeailops speltoides cytoplasm. Although nuclear genotype had the greatest effect, cytoplasm and nucleus X cytoplasm interaction were also significant. This suggests the best strategies for improving the capacity of wheat to initiate callus in anther culture involve improving the nuclear capacity for callus initiation.

Heritability studies have suggested that variation in callus inducibility should respond to selection. The objective of Part II was to determine ii: crosses designed for improving callus initiation would produce segregation from which genotypes with improved callus inducibility could be selected. The winter wheat cultivars 'Centurk 78', 'Norstar' and 'Norwin' were chosen as parents based on their high callus initiation frequencies and diverse genetic backgrounds. F2 populations from crosses between each of these parents were evaluated for callus initiation frequencies and callus vigor.

Heterosis was detected for callus initiation in the Norwin/Norstar cross and for callus vigor in all crosses, but is believed to be of minor importance. Heritability estimates of .25 and .37 for callus initiation in the . Centurk 78/Norstar and Norwin/Norstar crosses, respectively, suggest substantial genetic variation is available from these crosses for improving this trait. Heritability estimates for callus vigor were less promising.

I

INTRODUCTION

Anther culture provides plant breeders with a means of. producing homozygous lines rapidly. This could decrease the time required to produce true breeding lines from at least five years with conventional inbreeding, to one year using anther culture, and result in an overall savings of two to three years for variety development (Schnell et al., 1980). Somaclonal variation produced by anther culture has also been cited as a possible source of novel genetic variation for crop improvement (Evans et al., 1984).

The low frequency of success presently associated with anther culture makes it cost and labor intensive to obtain a representative array of the genetic variation available from a cross, or to insure a high probability of isolating favorable somaclonal variants. This is limiting the application of anther culture in crop breeding programs (Fouroghi-Wehr et al., 1982).

Success frequency is influenced by environmental, genetic, (Lazar et al., 1984a) physiological (Schaeffer et al., 1979) and developmental factors (Ding-gang and Jun- wen, 1984). Environment includes both the growing conditions of the donor plant (Lazar et al., 1984a) and culture conditions (Ouyang et al., 1983). Culture media

2

have been studied extensively, with limited recent improvements in success frequency (Kao, 1981; Chien and Kao, 1983; Liang et al., 1982; Chuang et al., 1978). The optimum developmental stage for culturing is well established at the mid to late uninucleate stage of microspore development (Ding-gang and Jun-wen, 1984). The physiological factors in the donor plant which' promote anther culture success are poorly understood, but generally, more vigorous plants perform better (Schaeffer et al., 1979). Lazar et al. (1984a) found that field grown wheat plants performed better than greenhouse or growth chamber plants, but that donor plant environment had a less signifigant effect than genotype. Since little headway has been achieved through recent investigations into anther culture technique, researchers have begun exploring the possibility of improving the capacity of cereals to undergo the anther culture process.

It has been reported that normal microspores will not respond to culture and only functionally impotent microspores become embryogenic (Heberle-Bors and Reinert, 1980). Heberle-Bors and Odenbach (1985) reported that cytoplasmic male sterility, which blocks normal pollen development, can produce large numbers of responsive microspores and enhance callus yields.

3

Genetic factors include both nuclear and cytoplasmic elements. Nuclear genotypic differences for tissue culture response have been reported extensively (Schaeffer et al., 1979). Heritability studies suggest that anther culture responsiveness is a heritable trait which should be amenable to breeding (Lazar et al, 1984b; Miah et al.,1985; Charmet and Bernard, 1984). Alien cytoplasms have been reported to increase callus production from wheat and triticale anther culture (Heberle-Bors and Odenbach, 1985; Charmet and Bernard, 1984) and enhance in vitro characteristics of wheat embryo cultures (Mathias et al., 1986). . ’ . ■ .

In Part I of this thesis, three nuclear genotypes of wheat, each in three cytoplasms were studied. The two alien cytoplasms conferred cytoplasmic male sterility. The objective was to evaluate the effects of nucleus, cytoplasm, nucleus X cytoplasm interaction and cytoplasmic male sterility on callus initiation in anther culture of wheat.

In Part II, three highly culturable winter wheats were chosen as parents. F2 populations from crosses between these parents were evaluated for callus initiation frequencies in anther culture to determine the feasibility of improving callus initiation rates through selection.

4

PART I

EFFECTS OF NUCLEUS, CYTOPLASM AND CYTOPLASMIC MALE

STERILITY ON CALLUS INITIATION IN ANTHER CULTURE OF WHEAT

5

INTRODUCTION

Anther culture provides a rapid means of producing homozygous lines in many crops. Anther culture of alloplasmic lines could expedite the breeding of restorer lines for hybrid breeding programs (Ling et al.f 1978) or of varieties in programs using cytoplasmic male sterile (cms) facilitated crossing. Picard and de Buyser (1973) reported successful anther culture and haploid plant regeneration from the cms wheat lChamplein(T. timopheevi). Meiosis and pollen development in cms wheat with 2Ls_ timopheevi cytoplasm appear normal through the first mitotic division (Joppa et al., 1966) which is beyond the critical stage for anther culture (Ding-gang and Jun- wen, 1984). Ling et al. (1978) reported increased callus induction frequencies from anther cultures of certain cms lines of rice.

Normal tobacco flowers were found to produce dimorphic pollen (Heberle-Bors and Reinert, 1980). P-grains are aberrant pollen grains which stain weakly with acetocarmine, lack starch grains, and are functionally sterile in situ, but potentially embryogenic in culture (Heberle-Bors, 1982). They reasoned that cytoplasmic male sterility may block normal pollen formation such that high

6

levels of P-grains are produced. Pollen from 88 cms wheat lines was studied, and 13, including lines in Ti timopheevi

' ' -"Xand Ae. speltoides cytoplasms, were identified as high producers of P-grains. Two of these lines in Ti timopheevi cytoplasm were anther cultured and formed embryogenic structures, while corresponding fertile lines did not (Heberle-Bors and Odenbachf 1985). Ti timopheevi cytoplasm has also been reported to increase embryogenesis in anther cultures of triticale (Charmet and Bernard, 1984), although no mention was made of male sterility.

The effect of nuclear genotype on tissue culture response in wheat has been well documented (Schaeffer et al.f 1979). Alien cytoplasms of TriticUm and Aeailops species produce many phenotypic effects in wheat (Mukai et al.f 1978). Recently, several investigators studied effects of alien cytoplasms on tissue culture of wheat. Kinoshita and Mikami (1984) measured the growth of callus derived from seeds of a euplasmic and 17 alloplasmic lines of wheat with the nuclear genotype of cv. 'Chinese Spring1. They reported that euplasmic Chinese Spring produced the most prolific callus while several alloplasmic lines, including one in Aeailops speltoides cytoplasm, had reduced growth rates.

Mathias et al. (1986) compared euplasmic and seven alloplasmic Chinese Spring lines for response to immature

7

embryo culture. The euplasmic line was near average for growth rate and all but one alloplasmic line had higher plant regeneration frequencies. The line in Triticum timopheevi cytoplasm exceeded the euplasmic line in all parameters.

Although effects of nuclear and cytoplasmic factors on the tissue culture of wheat have been demonstrated, nucleo- cytoplasmic interactions are known to occur in the expression of several phenotypic traits (Maan, 1979). This has not been studied in regard to tissue culture. Environment can also affect the expression of cytoplasmic factors (Maan, 1979) as well as the anther culture response of wheat (Lazar et al., 1984a). The objective of this study was to evaluate nuclear, cytoplasmic, nucleus X cytoplasm and cms effects on callus initiation frequencies in anther cultures of wheat grown in differentenvironments.

8

MATERIALS AND METHODS

Three euplasmic and six alloplasmic wheat lines were evaluated for callus initiation ability.in anther culture. Euplasmic lines included the spring wheat (Triticum aestivum L .) cultivars 'Chris', 'Butte', and 'Coteau'. Alloplasmic lines consisted of the same nuclear genotypes, each in the cytoplasms of Triticum timopheevi Zhuk. and Aeailops speltoides Tausch. Seed was kindly provided by Drs. S. S. Maan and K . A. Lucken of North Dakota State University, Fargo. The alloplasmic Chris lines were derived from over twenty generations of backcrossing. The alloplasmic Butte and Coteau lines resulted from a minimum of four backcrosses to alloplasmic Chris (Lucken, 1987).

Plants were grown in three environments. The first set was grown spaced on benches in the glasshouse during early spring 1985. Lighting was natural and temperature was unregulated but not allowed to drop below I C . The maximum temperature reached was 23 C . The second set of material was space planted in the field under dryland conditions during the summer of 1985. The third set was started in 5 X 25 cm conetainers in a growth chamber with 20 C day, 10 C night temperatures, a 12 hr daylength and cool white fluorescent light source. After six weeks, the

9

plants were transferred to 8 inch pots in the glasshouse and grown at 20 +/- 5 C constant temperature with 14 hour daylength. Natural light was supplemented about two hours in the evening with sodium halide lamps.

Spikes for cytological study were fixed in a 3:1 solution of 95% ethanol:glacial acetic acid for 24 hours and then stored in 70% ethanol. Anther squashes were stained with acetocarmine to study pollen development (Smith, 1947). Some pollen preparations were stained for starch with IKI.

Spikes were harvested for culturing at the mid to late uninucleate microspore stage, wrapped in foil and prechilled at 4 C for 7-10 days. Anthers were cultured on 60 X 15 mm petri dishes containing 12.5 ml potato medium (Schaeffer et al., 1979) made with cv. 'Kennebec1 potatoes and supplemented with Img/I thiamine HCl. Each plate consisted of the anthers cultured from one spike. At least 15 anthers, with a mean of 22, were cultured from each spike. Cultures were kept in the dark at 26 C for eight weeks at which time macroscopic calli were counted.

Frequencies of callus induction were determined for each spike as the number of callusing anthers/antherscultured. Frequencies were weighted according to the .

1/2number of anthers cultured and transformed by arcsin X prior to statistical analysis.

10

RESULTS

Cytological examination confirmed that pollen development in cms-(3% timopheevi) wheat appeared normal through the first pollen mitosis (Joppa et al., 1966). Pollen development in cms-(Ae.. speltoides) wheat was very similar to cms-(T» timophOevi). Microsporogenesis appeared normal through meiosis to the late uninucleate stage, producing microspores with sparse cytoplasms. In some grains, the cytoplasm degenerated at this stage, while others continued development beyond the first pollen mitosis. No pollen was observed to undergo the second mitotic division. IKI produced no stain reaction indicating a lack of starch deposits in cms-(Ae. speltoides) pollen.

Mean callus induction frequencies for environments, nuclear genotypes, and cytoplasms are shown in Table I.I. Nucleus and cytoplasm both had significant effects on callus initiation. The nuclear genotype of Chris had a significantly higher callus initiation frequency than Butte or Coteau. Euplasmic and T.timopheevi cytoplasms produced the highest induction frequencies t and Ae. speltoides was significantly lower. Although main environmental effects were nonsignificant, there was a significant environment X

11

nucleus interaction. The large number of factors known to affect anther culture (Vasil, 1980) make it impossible to determine a specific cause of the interaction. There was no detectable environment X cytoplasm interaction.

Table I.I. Mean callus induction percentages forenvironments, nuclear genotypes, and cytoplasms.

Environment Nucleus Cytoplasm

I 0.97 A* Chris 3.5 A T. aestivum 2.0 A2 1.3 A Butte 0.46 B T . timopheevi 2.1 A3 2.1 A Coteau 0.75 B Ae. speltoides 0.75 B* Means followed by different letters are significantly

different at the .05 level by Tukey's Studentized Range.

. To facilitate the analysis of nuclear and cytoplasmic effects, data were pooled across environments. The mean squares from analysis of variance (Table 1.2) indicate the nuclear genotype has a greater effect than cytoplasm. A nucleus X cytoplasm interaction was also significant.

Means of nucleus-cytoplasm combinations are presented in Table 1.3. Chris had the highest initiation frequency followed closely by Chris-(Ti timopheevi). Callus initiation was low for all lines in Ae. speltoides cytoplasm. Although Butte and Coteau had low callus initiations in all cytoplasms, Butte performed best in its normal cytoplasm while Coteau performed best in Ti

12

timopheevi. The relative performances are apparent in Figure I.I.

Table 1.2. Analysis of variance for nuclear and cytoplasmic effects with data pooled across environments.

Source DF Mean Square F-value

Nucleus 2 21.01 39.63 **Cytoplasm 2 4.72 8.90 **Nuc.X Cyt. 4 2.67 5.04 **Residual 663 0.53** significant at .01.

Table 1.3. Results of nucleus-cytoplasm combinations.

Line Number Number Number PercentSpikes Anthers Calli Initiation

Chris 89 1969 83 4.2 AChris-(T. timooheevi) 86 1936 79 4.1 BChris-(Ae. speltoides) 66 1449 . 22 1.5 CButte 67 1353 12 0.89 DButte-fT. timooheevi) 55 1180 3 0.25 FButte-(Ae. speltoides) 76 - 1554 5 0.32 FCoteau 80 1752 10 0.57 ECoteau-(T. timopheevi) 83 1937 20 1.0 CDCoteau-(Ae. speltoides) 70 1545 8 0.52 EI. Means followed by different letters are significantly different at the .05 level by Tukey's Studentized Range.

-BH*.

13

5-1

1-Butte

— a •-©

Ae. speltoides I . timopheevl Cytoplasm

T. aestlvum

Figure I.I Performance of nuclear genotypes across cytoplasms for callus initiation.

14

DISCUSSION

The development of pollen in cms-(Ae. speltoides) wheat closely resembles that reported for cms-(T. timopheevi) (Joppa et al., 1966). No anther sectioning was done but it is likely that a failure of tapetal breakdown and/or reduced vasculature of the anthers, as reported for cms-(T\ timopheevi) wheat, also occurs in cms-(Ae. speltoides).

The absence of environmental main effects is somewhat surprising considering the diverse growing conditions. Plants were generally vigorous in all environments which may be the major consideration. A genotype X environment interaction has previously been reported for callusing of Wheat anthers (Lazar et al., 1984a). Many factors differed between environments, making it impossible to assign any specific cause to the interaction. Qian et al. (1986) reported a temperature X genotype interaction on pollen development in wheat. Temperature may be the most important variable in the three environments since the first greenhouse environment was generally quite cool, while temperatures in the field often exceeded 32 C . Photoperiod and light intensity can also affect androgenetic response (Vasil, 1980) and may have

15

contributed to the interaction.Although nuclear genotype produced the greatest

overall effect on callusing, Ae. speltoides cytoplasm caused a large reduction in the initiation frequency, of Chris. This indicates responsive genotypes will not perform well in unfavorable cytoplasmic combinations. The Chris-(Ae5. speltoides) plants were the least vigorous of any line, which lends credence to the suggestion that cytoplasms may affect tissue culture indirectly via the physiology of the donor plant (Mathias et al., 1986). The results from Butte and Coteau show that different genotypes respond better to culture in different cytoplasms. The Coteau nuclear genotype performed significantly better in T. timopheevi cytoplasm than in normal Ti aestivum cytoplasm.

Overall, wheats in T. timopheevi cytoplasm performed as well as euplasmic wheats. Since Ti timpheevi cytoplasm is the most common source of cms in wheat (Lucken., 1973) , it should be possible to implement anther culture in cms facilitated breeding programs. It.would be important to evaluate the anther culture performance of specific nucleus-cytoplasm combinations in the breeding lines to avoid the wasted effort of culturing nonresponsive types.

Cytoplasmic male sterility did not enhance callus initiation frequencies. The low frequency of callus

16

induction from cms-(Ae. speltoides) lines shows that there are factors influencing androgenetic competence which can not be observed cytologically as described by Heberle-Bors(1982) . It is possible that the increased anther culture response reported in two cms wheat lines by Heberle-Bors and Odenbach (1985) was due to a favorable nucleus X cytoplasm interaction and was not a direct result of cytoplasmic male sterility.

Mathias et al. (1986) reported that related cytoplasms react similarly in culture but the results of Kinoshita and Mikami (1984) do not support a relationship between cytoplasmic phylogeny and in vitro reaction. There is strong indication that the cytoplasmic donor of T_s_ aestivum is Ae ̂ longissima (Tsunewaki and Ogihara > 1983). As a diploid, Ae. speltoides is considered to be its own cytoplasm donor (Ogihara and Tsunewaki, 1982). The origin of the Ti timopheevi cytoplasm is more confusing. On the basis of restriction endonuclease fragment length polymorphisms in chloroplast DNA, Tsunewaki and Ogihara(1983) believe Ti aucheri is the donor and that the three cytoplasms utilized in the current study of separate plasma types. Mitochondrial genome analysis also indicates that all the current cytoplasms are distinct (Vedel et al.,1978). Based on morphological characteristics of eu- and alloplasmic lines of different wheats, Maan and Lucken

17

(1971) propose that Ae. speltoides donated the T« timopheevi cytoplasm. This is further complicated by the fact that the alloplasmic (Ae. speltoides) lines developed by Tsunewaki are morphologically indistinguishable from common wheat, while those of Maan cause complete male sterility (Mukai et al., 1978). It appears these researchers may be reporting on different cytoplasms.Vedel et al. (1978) made no mention of whether the Ae. speltoides cytoplasm they studied confers cms to wheat.

The lines used in this study were developed by Drs. Maan and Lucken, so it is perhaps better to assume their classificatioh (Maan and Lucken, 1971). That being the case, Ae. speltoides and timopheevi would be closely related but behave differently in culture, while Tjl aestivum and 3L*. timopheevi would be less closely related, but behave more similarly. Whichever classification is correct, it is difficult to draw conclusions about how . cytoplasmic relationships affect wheat anther culture from this small sample of cytoplasms. The nucleus X cytoplasm interaction further complicates any generalizations. _

The results of this study demonstrate that cytoplasmic male sterility per se does not increase callus initiation from anther culture of wheat, although one cms line did perform better than its respective euplasmic line. Nuclear genotype had the greatest effect on callus initiation

18

frequency. Significant effects due to cytoplasm and nucleus X cytoplasm interaction were also observed. This emphasizes the interrelationship between cellular components in the response of plant tissues to culture. There was no clear relationship between cytoplasmic phylogeny and callus initiation in these lines. These results suggest that efforts for improving the anther culture capacity of wheat would best be directed toward the nuclear genotype.

19

PART II

GENETIC VARIATION FOR ANTHER CALLUS INDUCIBILITY

IN CROSSES OF HIGHLY CULTURABLE WINTER WHEATS

20

INTRODUCTION

Anther culture technique has been studied extensively with limited recent progress. Investigators have begun exploring the possibility of improving the genetic capacity of cereals to undergo the anther culture procedure. Successful anther culture depends on callus induction, and subsequent green plant formation (Charmet and Bernard,1984; Lazar et al ., 1984b), although some authors report the involvement of up to five separate steps (Henry and DeBuyser, 1985; Keyes et al., 1980). Success rates for each step are genotype dependent (Schaeffer et al., 1979; Foroughi-Wehr et al., 1982), and independently inherited (Foroughi-Wehr et al., 1982; Lazar et al., 1984b; Charmet and Bernard, 1984) .

Control of these traits appears to be due mainly to additive gene action, but significant dominance effects have also been reported (Charmet and Bernard, 1984; Lazar et al., 1984b; Miah et al., 1985). In wheat, narrow sense heritability estimates range from 0.62 to 0.71 for callus initiation and 0.60 to 0.72 for plant regeneration frequencies (Lazar et al., 1984b). Broad sense heritabilities in triticale have been estimated at 0.66 for embryogenesis and 0.54 for green plant regeneration

21

(Charmet and Bernard # 1984). Narrow sense heritability for callus induction in rice was estimated at 0.92 and broad sense at 0.97 (Miah et al., 1985). These reports suggest that cereals should respond to selection for betterculturability. Alfalfa with increased ability to

-

regenerate plants from callus has been successfully bred using recurrent selection (Bingham et al., 1975). Transgressive segregation for androgenesis has been demonstrated in crosses of wheats with low embryogenic rates (Henry and DeBuyser, 1985), but there have been no reports in which highly culturable parents were used.

Since subsequent steps in anther culture are dependent on successful callus initiation, it was decided to study the possibility of breeding for improved callus inducibility. This study's objective was to determine if crosses between highly callus inducible parents would produce segregation for callus inducibility in F2populations from which superior genotypes could be selected.

22

MATERIALS AND METHODS

Seven field grown winter wheat (Triticum aestivum L.) lines were evaluated for ability to form callus in anther culture. lCenturk1, lNorstar1 and lNorwin1 were selected as highly culturable cultivars for use as parents. lCenturk 781 replaced Centurk because of heterogeneity in Centurk. These cultivars are of diverse genetic backgrounds. Centurk 78 is a five head selection from Centurk (Schmidt et al.f 1981) which has the following pedigree: 'Kenya 58'/2/lNewhatch'/3/'Hope'/2*'Turkey'/4/'Cheyenne'/5/'Parker' (Schmidt et al., 1973). Norstar originated from a 'Winalta'/'Alabaskaya' cross (Grant,.1980). Norwin is derived from a 'Froid'/'Minoka'//MT6928/ 'Trader' cross (Taylor et al., 1986).

Parent and F populations from crosses between each of 2these three cultivars were grown in four environments. The first environment was an early spring planting in dryland field conditions. Plants in environments 2, 3 and 4 were started in 2.5 X 15 cm conetainers and vernalized for 9, 10 arid 11 weeks respectively in a controlled environment chamber under 15 C day temperature, 2 C night temperature, and 13 hour daylength with a sodium halide light source. Plants were transplanted into 15 cm pots and placed in a

23

glasshouse with 20 +/- 5 C constant temperature with a 14 hour daylength and natural light supplemented with sodium halide lamps. Population 2 of each cross inadvertently received a severe moisture stress during tillering. .

Individual plants were sampled with two spikes perplant as subsamples. Spikes were harvested at the mid tolate uninucleate stage of microspore development, wrappedin foil and prechilled at 4 C for 7-12 days. Anthers fromthe middle of each spike were aseptically plated on a 60 X15 mm petri dish containing 12.5 ml potato medium(Schaeffer et al., 1979) supplemented with I mg/1 thiamineHCl. At least 18 anthers were cultured from each spikewith a mean of 28.4. Plates were incubated in the dark at26 C for 8 weeks at which time macroscopic call! werecounted and scored. Callus initiation frequency wascalculated for each spike as the number of callusinganthers/anthers cultured. Spike callus initiationfrequencies were weighted according to the number of

. 1/2anthers cultured and transformed by arcsin X prior to statistical analyses. " Calli were visually scored for vigor. Calli up to approximately I mm in diameter were given a score of I, from I to 4 mm scored 2 and over 4 mm scored 3. Watery or necrotic calli were dropped a score and very small watery or necrotic calli were not counted.

Heterosis was calculated from population means using

24

the following formula: (F -mp)/mp (Mather and Jinks,21982). Between plant variances were petitioned out foreach population and used to calculate broad senseheritabilities using the following equation: (F -

2(P1+P2)/2)/F (Mather and Jinks, 1982).2

25

RESULTS

The lines surveyed for ability to initiate callus in anther culture showed a wide range of mean callus induction frequencies. These results are shown in Table II.I.Norstar, Centurk and Norwin had the three highest means and were chosen as parents for crosses designed to improve callus initiation ability, with Centurk 78 replacing Centurk.

Table II.I.. Mean callus initiation frequencies of winter wheat cultivars tested for use as parents.

Line % Callus Initiation

Redwin 0.004 A*MT8003 0.005 AMT7811 1.1 AWinalta 1.6 A BNorstar 3.0 BCenturk CO BNorwin CO B

* Frequencies followed by the same letter are notsignificantly different at the .05 level by Tukey's Studentized Range Test

26

In the parent and F populations, analysis of variance2showed significant effects due to pedigree and pedigree Xenvironment interaction on both callus initiation frequencyand callus vigor rating. There was no significant effectof donor plant environment on either trait. Means for theparents and F populations over all environments are

2presented in Table II.2. The Norwin/Norstar and Centurk78/Norstar F populations had the highest mean callus

2initiation frequencies. The Centurk 78/Norstar F2population also had the highest mean callus vigor rating.

Table II.2. Mean callus initiation frequencies and callus vigor ratings of the parent and F2 populations over all environments.

% Callus CallusPedigree N Initiation Vigor

Norwin 98 3.6 A* 1.7 ACenturk 78 82 5.5 A B 1.8 A BNorstar 86 6.2 B 1.9 BCenturk 78/Norwin 219 5.4 A B : 2.0 C BNorwin/Norstar 193 7.0 B 2.0 C BCenturk 78/Norstar 186 7.0 B 2.1 C* Means followed by the same letter are not significantly

different at the .05 level by Tukey's Studentized Range Test.

Comparison of the F population means to midparent2

means produced the estimates of heterosis presented in

27

Table II.3. Only the Norwin/Norstar cross showed significant heterosis for callus initiation in the F

2population. All three crosses showed significant heterosis for callus vigor.

Table II.3. Estimates of heterosis for callus initiation and callus vigor in the F2 populations.

Cross % Callus Initiation Callus Vigor

Centurk 78/Norwin .10 .11 **Norwin/Norstar .20 ** .10 **Centurk 78/Norstar .10 .15 **** Significant at .01 according to F2 vs. midparent

contrasts.

Between plant variances for callus initiationfrequency and callus vigor were petitioned out for eachparent and F population. These were used to calculate

2broad sense heritability estimates (Table II.4). for eachcross. The Norwin/Norstar cross produced the highestheritabilities for both callus initiation and callus vigor.The heritability estimate of -0.15 for callus initiationability in the Centurk 78/Norwin cross was considered 0.

Figures II.I - II.3 show the percentage distributionsof callus initiation frequency in each F population

2compared to its parents. There is a substantial decreasein the proportion of zeros in all the F populations

2

28

compared to their parents. However, the Centurk 78/NorwinF population distribution shows little shift toward higher 2

callus initiation frequencies, while in the Centurk 78/Norstar and Norwin/Norstar F populations there are shifts

2toward higher callus initiation frequency as compared to the parents.

Table II.4. Broad sense heritability estimates for callus initiation frequency and callus vigor.

Cross % Callus Initiation Callus Vigor

Centurk 78/Norwin O .13Centurk 78/Norstar .25 .11Norwin/Norstar .37 .24

Linear regression showed a weak correlation between2callus initiation frequency and callus vigor with an rvalue of .26.

29

PopulationE53 CENTURK 78 CZI NORWIN

Percent Callus Initiation (upper bound)

Figure II.I Percentage distribution of callus initiationfrequencies in the parent and F2 populationsof the Centurk 78/Norwin cross.

30

PopulationESI NORSTAR □ CENTURK 78

Percent Callus Initiation (upper bound)

Figure II.2 Percentage distribution of callus initiationfrequencies in the parent and F2 populationsof the Centurk 78/Norstar cross.

31

PopulationESI NORSTAR CD NORWIN

Percent Callus Initiation (upper bound)

Figure II.3 Percentage distribution of callus initiationfrequencies in the parent and F2 populationsof the Norwin/Norstar cross.

32

DISCUSSION

The genetic control of in vitro traits andheritability for these traits have been well documented inmany small grain crops (Foroughi-Wehr et al., 1982; Lazaret al., 1984b; Charmet and Bernard, 1984; Miah et al.,1985) . This suggests that in vitro traits should respondto selection. The heritability studies, however, have allbeen limited to diallel analyses, and involved crossesbetween poorly and highly cultutable genotypes. Breedingstrategy generally involves selecting parents of superiorgenotypes and preferably of diverse genetic backgrounds.In this study, segregation of callus inducibility andcallus vigor were examined in F populations from crosses

2designed for improving callus initiation.

The broad sense heritability estimates of .25 and .37 for callus inducibility in the current study would generally be considered low, but since the parents were phenotypically similar, and superior performers for this trait, these values represent substantial genetic variation for selecting improved genotypes. These are broad sense heritability estimates of which only the additive genetic variance is fixable. The heterosis effects imply dominance genetic variance is also involved, but Figures II.I - II.3

33

suggest that the distributions of these populations forcallus initiation frequency approximate a Poissondistribution. Since the mean equals the variance in aPoisson distribution, it is impossible to determine if theincrease in population mean of the F over the midparent is2due to an increased genetic variance produced bysegregation, or if the increased variance is an artifact ofheterosis. Data from F or more advanced generations wouldIbe required to definitely determine the importance of each factor. However, heterotic effects are generally limited in wheat (Hayes and Foster, 1976), and would be less in F

2generations than F . Since previous reports indicate a1predominance of additive gene action (Lazar et al., 1984b),the shift in the F population distributions relative to

2their parents is likely due to segregation for callus inducibility.

Callus vigor produced low estimates of heritability and heterosis in each cross. The heritability estimate of .24 for the Norwin/Norstar cross suggests some potential for improvement, but the other estimates are not encouraging, especially since the parents were not chosen on the basis of callus vigor. Heterosis has previously been reported for callus growth (Keyes and Bingham, 1979), but in this case it is surprising because the callus is presumably haploid or doubled haploid and thus of a

34

completely inbred nature. This effect could be explainedby gametic selection as suggested by Wenzel and Uhrig(1981). In the virtually homozygous parent cultivars,microspores would have little selective advantage over oneanother in culture. Genetic segregation within thesporogenous tissues of the F plants, creates a

2heterogeneous population of microspores. Those genetically more suited to the tissue culture environment could outcompete less suited ones and produce more vigorous calli. Gametic selection probably does not explain the heterotic effect for callus initiation which is thought to be under sporophytic control (Raquin, 1982; Pei-Iun et al., 1978) .

The correlation between callus initiation frequency and callus vigor suggests the possibility of some linkage or pleiotropy between these traits. Selection for high callus initiation may produce a slight trend toward higher callus vigor.

The lack of significant environmental main effects was unexpected. Effects of donor plant environment are well documented (Lazar et al., 1984a; Sunderland, 1978) and the growing conditions of the different sets of plant material were quite diverse. There was a significant difference in the parent means between the preliminary survey and the other four environments. That, and the significant

35

pedigree X environment interaction emphasize the influence of environment on the expression of quantitative characters.

These results indicate that breeding efforts could increase the callus initiation rates of wheat above those currently present in the best lines. It is likely that plant regeneration ability would respond similarly and that breeding for both traits could produce lines with outstanding in vitro characteristics. Since these traits have never been selected for, it is probable that initial advances would be rapid. Highly culturable lines would enhance the efficiency of anther culture research and breeding. While placing a major emphasis on improving culturability would not be practical for most breeding projects, it should be feasible to develop some elite lines with excellent in vitro characteristics and good agronomic traits. These lines could be used to confer culturability to nonresponsive lines and eventually high culturability could be incorporated into most breeding lines, making it feasible to routinely use anther culture in wheat breeding.

36

SUMMARY

In Part I, callus initiation frequency was evaluated in three euplasmic and six alloplasmic wheat lines.Nuclear genotypes of the spring wheat varieties 'Chris', 'Butte' and 'Coteau' were each present in normal Triticum aestivum cytoplasm and the alien cytoplasms of Triticum timopheevi and Aeailops speltoides. The alien cytoplasms both conferred cytoplasmic male sterility.

Breakdown of pollen development occurred after the critical stage for anther culture in all cms lines. Cytoplasmic male sterility per se did not enhance callus initiation, although one cms line outperformed its respective euplasmic line. Nuclear genotype had the greatest influence on callus initiation frequency, but cytoplasm and nucleus X cytoplasm interaction effects were also significant. . Chris was the superior nuclear genotype in all cytoplasms, and overall, the euplasmic and cms-(-T. timopheevi) lines outperformed the cms-(Ae. speltoides). These results suggest that efforts to improve the capacity of wheat to initiate callus in anther culture would best be directed toward the nuclear genotype.

In Part II, crosses were designed for improving callus initiation frequency in anther culture. The three winter

37

wheat cultivars, lCenturk 78', lNorstar1 and .'Norwin1, wereselected as parents on the basis of high callus initiationfrequencies. F populations from crosses between each of2these parents were evaluated for callus initiation frequency and callus vigor.

Significant heterosis was detected for callus initiation in the Norwin/Norstar cross, and for callus vigor in all crosses. Heterosis for callus vigor was attributed to possible in vitro microspore selection since heterosis is theoretically impossible in haploid or doubled haploid tissue.

Heritability estimates of 0, .25 and .37 were obtainedfor callus initiation in the Centurk 78/Norwin, Centurk78/Norstar and Norwin/Norstar crosses, respectively. Theseestimates would normally be considered low, but since bothparents were superior performers, substantial geneticvariation may be available in the Centurk 78/Norstar andNorwin/Norstar crosses for improving callus initiationrates. Due to the callus initiation distributions in theseF populations it is impossible to distinguish between

2effects of heterosis and segregation. However, based on previous knowlege of gene action in callus induction in anther culture of wheat, it is believed that segregation is the predominant contributor. These results suggest that transgressive segregation can occur in crosses of highly

38

culturable wheats and that, through breeding, callus initiation frequencies could be improved oyer those currently present in the best lines.

39

REFERENCES

40

REFERENCES

Bingham, E. T., L. V. Hurley, D. M. Kaatz and J. W.Saunders. 1975. Breeding alfalfa which regenerates from callus tissue in culture. Crop Sci. 15: 719-721

Charmet, G. and S. Bernard. 1984. Diallel analysis ofandrogenetic plant production in hexaploid Triticale (X . tritosecale, Wittmack). Theor. Appl. Genet. 69: 55-61

Chien, Y. C . and K. N. Kao. 1983. Effects of osmolality,cytokinin and organic acids on pollen callus formation in Triticale anthers. Can. J. Bot. 61: 639-641

Chuang, C. C., J. W. Ouyang, H. Chia, S. M . Chou and C. K. Ching. 1978. A set of potato media for wheat antherculture. In: Proc. Symp. Plant Tissue Culture,Science Press, Peking. p. 51-56

Ding-gang, H. and 0. Jun-wen. 1984. Callus and plantletformation from cultured wheat anthers at different developmental stages. Plant Sci. Lett. 33: 71-

Evans, D . A., W. R. Sharp and H. P. Medina-Filho. 1984.Somaclonal and gametoclonal variation. Amer. J. Bot. 71: 759-774

Foroughi-Wehr, B., W. Freidt and G. Wenzel. 1982. On the genetic improvement of androgenetic haploid formation in Hordeum vulaare L. Theor. Appl. Genet. 62: 233-239

Grant, M. N. 1980. Registration of lNorstar1 wheat. Crop Sci. 20: 552

Hayes, J. D. and C . A. Foster. 1976. Heterosis in self-pollinating crops, with particular reference to barley. In: A. Janossy and F . G. H. Lupton (eds.) Heterosis in Plant Breeding. Elsevier Scientific Pub. Co., NY. p. 239-256

Heberle-Bors, E. and J. Reinert. 1980. Isolated pollen cultures and pollen dimorphism. Naturwissenshaften 67: 311

41

Heberle-Bors, E . 1982. In vitro pollen embryqgenesis inNicotiana tabacum L. and its relation to pollen sterility, sex balance, and floral induction of the pollen donor plants. Planta 156: 396-401

Heberle-Bors, E. and W. Odenbach. 1985. %n vitro pollenembryogenesis and cytoplasmic male sterility in Triticum aestivum. Z. Pflanzenzuchtg. 95: 14-22

Henry, Y . and J. DeBuyser. 1985. Effect of the 1B/1Rtranslocation on anther culture ability in wheat (Triticum aestivum L.). Plant Cell Reports 4: 307-310

Joppa, L. R., F. H. McNeal and J. R. Welsh. 1966. Pollen and anther development in cytoplasmic male sterile wheat. Crop Sci. 6: 296-297

Kao, K. N . 1981. Plant formation from barley anthercultures with Ficoll media. Z. Pflanzenphysiol. 103: 437-443

Keyes, G. J., G. B. Collins and N . L. Taylor. 1980.Genetic variation in tissue cultures of red clover. Theor. Appl. Genet. 58: 265-271

Keyes, G. J. and E. T. Bingham. 1979. Heterosis andploidy effects on the growth of alfalfa callus. Crop Sci. 19: 473-476

Kinoshita, T . and T . Mikami. 1984. Alloplasmic effects oncallus proliferation and streptomycin resistance in common wheat. Seiken Ziho 32: 31-38

Lazar, M. D., G. W. Schaeffer and P . S. Baenziger. 1984a.Cultivar and cultivar X environment effects on the development of callus and polyhaploid plants from anther cultures of wheat. Theor. Appl. Genet. 67: 273-277

Lazar, M . D., P. S. Baenziger and G. W. Schaeffer. 1984b. Combining abilities and heritability of callus formation and plantlet regeneration in wheat (Triticum aestivum L.) anther cultures. Theor. Appl. Genet. 68: 131-134

Liang, G. H., N. Sangduen, E. G. Heyne and R. G. Sears.1982. Polyhaploid production through anther culture in common wheat. J. Heredity. 73: 360-364

42

Ling, T., H. Huang, C . Liang and P . Chun. 1978. The study of anther culture in "three lines" breeding and utilization of heterosis in Oryza sativa subsp. Hsein. In: Proc. Symp. of Plant Tissue Culture, SciencePress, Peking p. 213-222.

Lucken, K . A. 1973. Comparative use of cytoplasmic male sterility-fertility restoration systems in hybrid- wheat breeding. In: Proc. 4th Intl. Wheat GeneticsSymp. p . 361-366

Lucken, K . A. 1987. Agronomy Dept., North Dakota State University, Fargo. Personal communication.

Maan, S. S. and K . A. Lucken. 1971. Nucleo-cytoplasmicinteractions involving Aeailops cytoplasms and Triticum genomes. J. Hered. 62: 149-152

Maan, S. S. 1979. Specificity of nucleo-cytoplasmicinteractions in Triticum and Aegilops species (a review). Wheat Inf. Serv. 50: 71-79

Mather, K . and J. L. Jinks. 1982. Biometrical Genetics (3rd ed.). Chapman and Hall, New York.

Mathias, R. J-., K . Fukui and C . N. Law. 1986. Cytoplasmic effects on the tissue culture response of wheat (Triticum aestivum) callus. Theor. Appl. Genet. 72: 70-75 '

Miah, M. A. A.,.E. D . Earle and G. S. Khush. 1985.Inheritance of callus formation ability in anther cultures of rice, Oryza sativa L. Theor. Appl. Genet. 70: 113-116

Mukai,'Y., S. S. Maan, I. Panayotov and K. Tsunewaki.1978. Comparative studies of the nucleus-cytoplasm hybrids of wheat produced by three research groups. Proc. 5th Int. Wheat Genet. Symp. I: 282-292

Ogihara, Y. and K . Tsunewaki. 1982. Molecular basis ofthe genetic diversity of the cytoplasm in Triticum and Aeailops♦ I. Diversity of the chloroplast genome and its lineage revealed by the restriction pattern of ct- DNAs. Jpn. J. Genet. 57: 371-396

Ouyang, J. W., S. M. Zhou and S. E. Jia. 1983 Theresponse of anther culture to culture temperature in Triticum aestivum. Theor. Appl. Genet. 66: 101-109

43

Pei-Iunf S., C. Chi-Iiang and P. Wen-chung. 1978. The induction effect of somatic tissues in the anther cultured in vitro on androgenesis and its control.In: Proc. Symp. Plant Tissue Culturef Science Press, Peking p . 143-150

Picardf E. and J. de Buyser. 1973. Obtention de plantuleshaploides de Triticum aestivum L. a partir de culture d'antheres la vitro. C . R. Acad. Sc. Paris 277-D: 1463-1466

Qian/ C.M., A. Xu and G. H. Liang. 1986. Effects of low temperatures and genotypes on pollen development in wheat. Crop. Sci. 26: 43-46

Raquinf C . 1982. Genetic control of embryo production andembryo quality in anther culture of petunia. Theor. Appl. Genet. 63: 151-154

Schaeffer, G. W.f P. S. Baenziger and J. Worley. 1979. Haploid plant development from anthers and in vitro embryo culture of wheat. Crop Sci. 19: 697-702

Schmidtf J. W.f V. A. Johnsonf P. J. Mattern and A. F.Dreier. 1973. Registration of 'Centurk' wheat. CropSci. 13: 776

Schmidtr J. W., V. A. Johnsonf P. J. Matternf A. F. Dreier and D . V. McVey. 1981. Registration of 'Centurk .78'wheat. Crop Sci. 21: 989

Schnellf R. J.f IIf E. A. Wernsman and L. G. Burk. 1980. Efficiency of single-seed-descent vs. anther-derived dihaploid breeding methods in Tobacco. Crop Sci. 20: 619-622

Smith, L . 1947. The acetocarmihe smear technique. StainTech. 22: 17-31

Sunderlandf N. 1978. Strategies in the improvement of yields in anther culture. In: Proc. Symp. of Plant Tissue Culture, Science Pressf Peking p. 65-86

Taylorf G. A.f G. H. Spitlerf C. F. McGuire, J. W. Bergman, A. L. Dubbsf G. Carlson, G. F . Stallknecht and V. R, Stewart. 1986. Registration of 'Norwin' wheat. Crop Sci.'26: 1086-1087

44

Tsunewaki, K . and Y . Ogihara. 1983. The molecular basis ofgenetic diversity among cytoplasms of Triticum and Aeailops species. II. on the origin of polyploid wheat cytoplasms as suggested by chloroplast DNA restriction fragment patterns. Genetics 104: 155-171

Vasil, I. K. 1980. Androgenetic haploids. In: I. K.Vasil (ed.) Perspectives in Plant Cell and Tissue Culture. Intl. Rev. Cytol. Suppl. IlA. Academic Press, NY

Vedel, F., F. Quetier, F. Dosba and G. Doussinault. 1978.Study of wheat phylogeny by EcoRI analysis of chlbroplastic and mitocondrial DNAs. Plant Sci. Lett. 13: 97-102

Wenzel, G. and H. Uhrig. 1981. Breeding for nematode andvirus resistance in potato via anther culture. Theor. Appl. Genet. 59: 333-340