www.elsevier.com/locate/ijfoodmicro

International Journal of Food Microbiology 90 (2004) 237–248

High efficiency transformation of Penicillium nalgiovense with

integrative and autonomously replicating plasmids

Francisco Fierroa,b, Federico Laicha, Ramon O. Garcıa-Ricoa, Juan F. Martına,b,*

a Instituto de Biotecnologıa de Leon (INBIOTEC), Parque Cientıfico de Leon, Avda. del Real, no. 1, 24006 Leon, SpainbArea de Microbiologıa, Facultad de Ciencias Biologicas y Ambientales, Universidad de Leon, 24071 Leon, Spain

Received 2 January 2003; received in revised form 19 May 2003; accepted 30 May 2003

Abstract

Penicillium nalgiovense is a filamentous fungus that is acquiring increasing biotechnological importance in the food industry

due to its widespread use as starter culture for cured and fermented meat products. Strains of P. nalgiovense can be improved by

genetic modification to remove the production of penicillin and other potentially hazardous secondary metabolites, to improve

its capacity to control the growth of undesirable fungi and bacteria on the meat product, and other factors that contribute to the

ripening of the product in order to get safer and better quality foods. Genetic manipulation of P. nalgiovense has been limited by

the lack of molecular genetics tools that were available for this fungus, particularly for ‘‘self-cloning’’ avoiding the use of

exogenous DNAs. In this article we describe a series of vectors, selectable markers and transformation methods that can be used

for efficient transformation of P. nalgiovense, gene cloning and expression. A uridine auxotrophic P. nalgiovense mutant with

an inactive pyrG gene has been isolated. The P. nalgiovense wild-type pyrG gene was cloned and sequenced, and vectors

carrying the gene were shown to complement the pyrG mutant. Autonomously replicating plasmids carrying the AMA1 region

from Aspergillus nidulans transformed P. nalgiovense very efficiently; these plasmids were shown to be maintained as stable

extrachromosomal elements in P. nalgiovense and could be rescued in Escherichia coli. The mitotic stability of self-replicative

AMA1 plasmids in P. nalgiovense was higher than that reported for Penicillium chrysogenum.

D 2003 Elsevier B.V. All rights reserved.

Keywords: Food starters; Penicillium nalgiovense; Transformation; Autonomous replication

1. Introduction the product (Grazia et al., 1986; Lucke, 1986). How-

In the elaboration of dry sausages (salami) and dry

curedmeat products, a layer of fungi develops naturally

on the surface of the product during the curing/ripening

process, which is considered to be beneficial due to

their positive effects on the flavour and appearance of

0168-1605/$ - see front matter D 2003 Elsevier B.V. All rights reserved.

doi:10.1016/S0168-1605(03)00306-4

* Corresponding author. Instituto de Biotecnologıa de Leon

(INBIOTEC), Parque Cientıfico de Leon, Avda. del Real, no. 1,

24006 Leon, Spain. Fax: +34-987-210388.

E-mail address: degjmm@unileon.es (J.F. Martın).

ever, many of these fungi are mycotoxigenic (Sutic et

al., 1972; Leistner and Pitt, 1977; Leistner, 1984;

Pestka, 1995) and their presence should be avoided to

get a product suitable for human consumption. The use

of the filamentous fungus Penicillium nalgiovense as

starter for cured and fermented meat products is be-

coming a routine in the food industry as a way to

prevent growth of undesirable microbiota (Fink-Grem-

mels et al., 1988; Leistner, 1990; Berwall and Dincho,

1994). Some of the features that make P. nalgiovense

suitable as a starter are (i) it contains enzymatic

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248238

activities like proteases (Geisen, 1993a), which con-

tribute to the ripening of the product, and (ii) it has not

been found to produce known mycotoxins. Geisen

(1993b) described the requirements that a given strain

should fulfill to be used as starter, among them the lack

of production of antibiotics and mycotoxins and the

ability to antagonize the growth of other undesirable

microorganisms on the product. Despite its widespread

use, P. nalgiovense does not totally fulfill Geisen’s

requirements, as it produces the antibiotic penicillin

(Andersen and Frisvad, 1994; Farber andGeisen, 1994;

Laich et al., 1999), and other secondary metabolites

like isocoumarins (Larsen and Breinholt, 1999, and

references therein), whose potential risk on humans has

not been tested yet. In addition, P. nalgiovense is not

able to prevent completely the development of other

fungi and bacteria on some cured products like Spanish

Cecina (Laich et al., unpublished results), though it

causes a reduction in their number.

P. nalgiovense strains can be improved in many

ways to be used as starter. Strains should be obtained

that do not produce penicillin and, if proven toxic, other

secondary metabolites (Laich et al., 2003). Proteases,

lipases and other activities that contribute to the ripen-

ing and development of the organoleptic features

typical of the product can be enhanced or introduced

by transformation with plasmids containing adequate

genes. This method can also be used to introduce genes

that help to inhibit the development of other undesir-

able microorganisms.

In this article, we describe the development of

several genetic tools for P. nalgiovense, namely aux-

otrophy and antibiotic resistance markers, pyrG host

strains and integrative as well as self-replicative

vectors. These tools will greatly facilitate the research

at the molecular level on this fungus in order to

characterize secondary metabolite biosynthetic path-

ways and will also be useful to obtain improved

strains that can function as better and safer starters

than the currently used ones.

2. Materials and methods

2.1. Fungal strains

P. nalgiovense 16a, a biotype 6 strain isolated from

Cecina (Laich et al., unpublished results), was used as a

source of DNA for the genomic library, as recipient for

transformation, and in mutagenesis experiments to

obtain the pyrG mutant.

2.2. Mutagenesis of P. nalgiovense 16a with UV light

Thirty milliliters of a suspension of 1.5� 107 co-

nidia/ml were submitted to UV light radiation of 30 W

and 253.7 nm wavelength from a UV source placed at

30 cm above the conidia on a petri dish (without cover),

inside a laminar flow hood. Samples of 1 ml were taken

at initial time (t = 0) and every 5 s for 2.5min. Irradiated

conidia were kept in the dark. From each sample, 100

Al were taken, and serial dilutions up to 10� 6 were

made, which were plated on petri dishes with MEA

medium (containing in g/l: malt extract, 20; peptone, 1;

glucose, 20; agar, 20; pH 5.6) and incubated at 25 jCfor 5 days in the dark. The number of colonies in each

plate was counted and the percentage of survival with

respect to t = 0 was calculated for each irradiation time.

Samples showing an 80% ofmortality were selected for

screening of pyrG mutants.

2.3. Screening of the irradiated conidia for pyrG

mutants

An auxotroph enrichment method was used, mod-

ified from Bos and Stadler (1996). A suspension

containing 1�107 viable conidia was inoculated in

100 ml Czapeck (Cz) minimal medium in flasks and

incubated in the dark at 25 jC, 250 rpm, for 48 h. The

culture was then filtrated through nylon membranes

(20-Am pore size), which allowed retention of germi-

nated conidia. Nongerminated conidia were collected

by centrifugation at 8000� g for 15 min, resuspended

in 0.8% NaCl at a concentration of 1�103 viable

conidia/ml, and 100 Al of this suspension were plated

on petri dishes with Cz solid medium supplemented

with uridine (140 Ag/ml final) and incubated in the dark

at 25 jC for 7 days. The colonies obtained were picked

individually with a toothpick on Cz and Cz + uridine

solid media, respectively, and the plates were incubated

at 25 jC for 5–7 days. Uridine auxotrophs were chosen

and streaked on Cz + uridine + 5-fluoroorotic acid (5-

FOA, 1 mg/ml final concentration) to select those

clones resistant to 5-FOA, which would lack the

orotidine-5V-phosphate decarboxylase, encoded by

the pyrG gene.

of Food Microbiology 90 (2004) 237–248 239

2.4. Screening of the genomic library

Lysis plaques obtained after infection of Escher-

ichia coli LE392 with a P. nalgiovense 16a genomic

library (constructed in the vector Lambda GEMR-12,Promega, Madison, WI, USA) were transferred onto

nitrocellulose membranes (Protan BA 85, Schleicher

and Schuell, Dassel, Germany). The membranes were

hybridized by standard procedures (Sambrook et al.,

1989) with a 1.5-kb HindIII DNA probe containing

the pyrG gene of Penicillium chrysogenum (Fierro et

al., 1996). The probe was labeled by the nick trans-

lation system (BioRad, Hercules, CA, USA) with

[32P]-dCTP. Positive phage plaques were taken out

from the plate as plugs for a second and third

screening, after which, pure clones with DNA frag-

ments containing the pyrG gene were isolated.

2.5. Extraction of DNA from recombinant phages

Recombinant phages isolated after the screening of

the genomic library were used to infect E. coli LE392

at a proportion of 1�108 pfu per 5� 109 E. coli cells,

and incubated at 37 jC for 20–30 min. The infection

mixture was transferred to a flask containing 22 ml of

TY (2� ) medium (containing in g/l: bacto-triptone,

20; yeast extract, 10; pH 7.2) and incubated at 37 jC,250 rpm, until the bacterial culture was totally lysed

by the phage (approximately 5 h). Then, 5 Ag/ml

RNAse, 10 Ag/ml DNAse and 1 Ag/ml lysozyme were

added to the lysate, which was incubated at 37 jC and

100 rpm for 30 min and centrifuged at 11,000� g at 4

jC for 10 min. The supernatant was transferred to

CorexR tubes (Corning, Big Flats, NY, USA) and the

phages were precipitated by adding 1.4 g NaCl and

6.25 ml of 50% PEG 6000 (w/v), incubated for at least

60 min in ice and centrifuged 7500� g, 4 jC for 30

min. The precipitated phages were then resuspended

in 3 ml of TE buffer and extracted three times with

chloroform–isoamyl alcohol (CIA, 24:1). Three milli-

liters of 4% SDS (w/v) were added and the mixture

was incubated for 20 min at 70 jC; afterwards, 3 ml

of 2.5 M potassium acetate, pH 4.8, was added, mixed

thoroughly and incubated in ice for 10 min. The

mixture was centrifuged at 26,500� g, 4 jC, for 10min and the supernatant filtered through a nylon

membrane (30 Am in diameter) to eliminate the debris.

The released DNA was precipitated with 1 volume of

F. Fierro et al. / International Journal

isopropanol (30 min at room temperature) and centri-

fuged at 7500� g for 30 min. The DNA pellet was

washed with 70% ethanol, resuspended in 500 Al TEbuffer and treated with RNAse (100 Ag/ml final) at 37

jC for 60 min. Finally, the RNAse-treated DNA

solution was extracted successively with 1 volume

phenol–CIA (1:1) and 1 volume CIA, precipitated

with ethanol at � 20 jC, washed with 70% ethanol

and resuspended in 50 Al TE buffer.

2.6. Plasmid constructions

Plasmids for P. nalgiovense transformation were

constructed following the standard methods. Compe-

tent cells of E. coli DH5a were used for high

efficiency transformation and isolation of plasmid

DNA.

2.7. Sequencing of DNA

Sequencing clones were generated by unidirection-

al deletions using the Erase-a-baseR system (Prom-

ega) following the manufacturer’s instructions.

Sequencing reactions and automatic sequencing were

performed with the AutoReadk system (Pharmacia,

Uppsala, Sweden).

2.8. Fungal transformation

Protoplasts of P. nalgiovense 16a were obtained as

described by Wang et al. (1999). Plasmid DNA (1 Ag)was mixed with 100 Al STC buffer (1 M sorbitol, 50

mM CaCl2, 2.5 mM Tris–HCl, pH 7.5) containing

108 protoplasts/ml and 10 Al of 66% polyethylene

glycol (PEG) 3350 solution in STC buffer; this

mixture was incubated for 20 min in ice. Then, 500

Al of 66% PEG 3350 in STC buffer were added, and

the transformation mixture was incubated at room

temperature for 20 min and mixed with 600 Al of

STC buffer afterwards. Serial dilutions (up to 10� 4)

of the transformation mixture were made with KTC

buffer (0.6 M KCl, 50 mM CaCl2, 2.5 mM Tris–HCl

pH 7.5), and 1 ml of each dilution was mixed with 10

ml of molten Cz–KCl (Cz medium supplemented

with 0.6 M KCl) and plated onto petri dishes con-

taining 10 ml Cz–KCl as base medium.

When resistance to antibiotics (phleomycin) was

used as selective marker, the transformation mixture

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248240

was incubated with 5 ml of CM medium (in g/l: yeast

extract, 5; malt extract, 5; glucose, 5) containing 1 M

sorbitol at 25 jC for 8 h with gentle shaking. Aliquots

of 1 ml were then mixed with 9 ml of molten CM agar

(1.5% w/v) containing 1 M sorbitol and 15 Ag/ml

phleomycin (Cayla, Toulouse, France) and plated onto

petri dishes containing 10 ml of the same medium.

2.9. Plasmid recovery in E. coli

One microgram of total DNA from P. nalgiovense

transformants was used to transform competent E. coli

DH5a cells (Hanahan, 1986). Extraction and analysis

of DNA from the E. coli transformants was performed

by standard procedures.

2.10. Plasmid stability studies

Plasmid stability analysis of P. nalgiovense trans-

formants was performed essentially as described pre-

viously (Fierro et al., 1996), but using MEA (see

composition above) instead of POWER as sporulation

medium. MEA does not support growth of P. nalgio-

vense uridine auxotrophs. Transformants were grown

on MEA under selective pressure conditions (without

supplement of uridine) or without selective pressure

(MEA supplemented with 140 Ag/ml uridine). Conid-

ia were then collected, diluted to a suitable concen-

tration and plated onto Cz minimal medium with or

without uridine, and the colonies growing on each

condition were counted.

2.11. Southern blotting and hybridization

Total DNA from the transformants was extracted

and Southern blotting to Hybond N membranes

(Amersham, Little Chalfont, Buckinghamshire, Eng-

land) were performed as described by Fierro et al.

(1996). The probe was labeled with the DIG DNA

labeling Mix (Boehringer Mannheim, Mannheim,

Germany) according to the manufacturer’s protocol.

Prehybridization and hybridization were done with

40% formamide standard buffer (Sambrook et al.,

1989) at 42 jC. After hybridization, the membrane

was washed for 15 min at room temperature with 2�SSC, 0.1% sodium dodecyl sulfate (SDS), 15 min at

42 jC with 0.1� SSC, 0.1% SDS, and 3 min at 65

jC with 0.1� SSC, 0.1% SDS. The signals were

visualized with a chemiluminescent substrate for al-

kaline phosphatase, according to the manufacturer’s

protocol (CDP-Star, Roche, Mannheim, Germany).

3. Results

3.1. Isolation of a P. nalgiovense pyrG mutant

Transformation of auxotrophic mutants using the

wild-type P. nalgiovense pyrG gene offers clear

advantages over transformation with vectors based

on antibiotic resistance markers for use in the food

industry (see Discussion). Therefore, our first aim was

to get a P. nalgiovense pyrG mutant to establish a

transformation system for this fungus based on aux-

otrophy complementation.

The possibility that more than one copy of the

pyrG gene was present in the genome of P. nalgio-

vense 16a was tested by Southern analysis, but the

results indicated that there was one single copy of

the gene (data not shown). If there is only one copy

of the pyrG gene, it should be feasible to get uridine

auxotrophs mutated in this gene. Using a method

based on an enrichment of auxotrophs after UV light

treatment (see Material and Methods), six uridine

auxotrophs were isolated from 1800 colonies

screened. One of these auxotrophs was resistant to

5-FOA (Fig. 1), suggesting that it was mutated in the

pyrG gene. The pyrG auxotrophy was finally con-

firmed by transformation with plasmid pAMPF9L

(Fierro et al., 1996), containing the pyrG gene of P.

chrysogenum, which was able to complement the P.

nalgiovense mutant. This mutant, named P. nalgio-

vense 16a pyrG-1, had the same morphology as the

parental P. nalgiovense 16a strain and did not show

any phenotypic difference with respect to the paren-

tal strain other than the auxotrophy of uridine.

3.2. Cloning of the P. nalgiovense pyrG gene

The pyrG gene of P. nalgiovense 16a was cloned

by screening of a genomic library with a probe

containing the P. chrysogenum pyrG gene. Positive

clones were isolated and purified, and their DNAs

analyzed by Southern to locate the gene among the

different restriction fragments obtained (Fig. 2). Fi-

nally, an EcoRI 3.1 kb and an XhoI 2.3 kb fragments

Fig. 2. Southern blot hybridization with a pyrG probe of the DNA of

recombinant phages Y6 and Y10, isolated after screening of the

genomic library, digested with different restriction enzymes. A 1.5 -

kb HindIII fragment containing the P. chrysogenum pyrG gene was

used as probe. The enzyme NotI was used to release the

recombinant DNA from the arms of the phage vector. DNAs

digested with enzymes SalI, XbaI, SacI and BamHI gave two

hybridizing bands, whereas digestion with enzymes EcoRI and XhoI

gave only one, suggesting that the hybridizing EcoRI or XhoI DNA

fragments contained the whole P. nalgiovense pyrG gene.

Fig. 1. Growth on Czapeck medium containing 140 Ag/ml uridine

and 1 mg/ml 5-FOA of six P. nalgiovense mutants auxotrophs of

uridine. After 10 days of incubation at 28 jC, only one of the

mutants, named P. nalgiovense 16a pyrG-1 (strain 1 in the

photograph), was able to grow in presence of 5-FOA, therefore

indicating that it is defective in orotidine 5V-phosphate decarboxy

lase activity, encoded by the pyrG gene.

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248 241

apparently containing the whole P. nalgiovense pyrG

gene were subcloned in the vector pBluescript-SK+

(Stratagene), generating, respectively, the plasmids

pBPnYE12 and pBPnYX19.

The nucleotide sequence of the XhoI 2.3 kb DNA

fragment was determined. This fragment contained an

ORF of 892 bp that showed 94.3% identity at the

nucleotide level with the P. chrysogenum pyrG gene

and 69.5% with the Aspergillus niger pyrG gene. The

ORF contained one intron in the same position as in

the P. chrysogenum gene and encoded a polypeptide

of 276 amino acids with a deduced molecular mass of

29,924 Da. Comparisons of the amino acid sequence

with proteins in databases showed that the protein

encoded by the putative P. nalgiovense pyrG gene is

unequivocally an orotidine 5V-phosphate decarboxyl-

ase. The functionality of the cloned P. nalgiovense

pyrG gene was confirmed by transformation of the

strain of P. chrysogenum Wis54-1255 pyrG1 (Dıez et

al., 1987), with the plasmids pBPnYE12 and

pBPnYX19. Both plasmids were able to complement

the mutation of the P. chrysogenum pyrG strain and

reverted its auxotrophy, thus confirming the function-

ality of the cloned P. nalgiovense pyrG gene.

The sequence of the P. nalgiovense pyrG gene has

been deposited in GenBank, accession number

AF510725.

3.3. Transformation of P. nalgiovense with integrative

and autoreplicative plasmids by uridine auxotrophy

complementation

The uridine auxotroph P. nalgiovense 16a pyrG-1

was used as recipient strain for transformation with

two different constructions containing the P. nalgio-

vense pyrG gene: plasmid pBPnYSX15 (integrative),

which was a pBluescript-SK+ derivative carrying a

XhoI–EcoRI 2.2 kb fragment with the cloned pyrG

Fig. 3. Integrative and autoreplicative plasmid constructions to transform strain P. nalgiovense 16a pyrG-1. The P. nalgiovense pyrG gene is

shown with an arrow, the AMA1 fragment (see text) is indicated with a black box, and the grey boxes correspond, respectively, to pBluescript-

SK+ sequences (in pBPnYSX15) and pBC-KS+ (Stratagene) sequences (in pAMPF2-H and pAMPn2). Plasmid pAMPF2-H is a derivative of

pAMPF2 (Fierro et al., 1996) obtained by HindIII digestion and auto-religation of pAMPF2. The following restriction sites are shown: ApaI (A),

BamHI (B), BglII (Bg), EcoRI (E), EcoRV (EV), HindIII (H), KpnI (K), SacI (Sc), SalI (S) and XhoI (X). Steps to construct plasmid pAMPn2

were as indicated in the figure. The designation AM is used in constructions or transformants containing the autonomous replication AMA-1

sequence.

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248242

Table 1

Transformation efficiencies of P. nalgiovense with integrative and

autoreplicative plasmids

Strain Selection

marker

Plasmid Transformants/

Ag DNA per

107 protoplasts

P. nalgiovense pyrG gene pBPnYSX15 749 (F 216)

16a pyrG-1 pAMPn2 45,000 (F 7000)

P. nalgiovense ble gene pULJ43 104 (F 16)

16A pAMPF21 768 (F 134)

The figures are the average of three independent experiments; the

standard deviation is shown in parentheses.

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248 243

ORF and its promoter region, and plasmid pAMPn2,

which contained the same pyrG fragment and the

AMA1 region of Aspergillus nidulans (Fig. 3). The

AMA1 region (Gems et al., 1991) has been shown to

confer autonomous replication capacity to plasmids in

several filamentous fungi (Fierro et al., 1996; Alek-

senko and Clutterbuck, 1997).

Transformation frequencies were very different for

each of these plasmids (Table 1). With pBPnYSX15, a

transformation efficiency typical of an integrative

Fig. 4. Plasmid pAMPn2 is self-replicative and plasmid pBPnYSX15 is in

strains and different transformants electrophoresed in an agarose gel. L

nalgiovense 16a pyrG-1 (parental untransformed strain); 3, transformant A

marker (E-phage DNA digested with AccI); 11, plasmid pAMPn2 extracted

marker (E-phage DNA digested with HindIII). (B) Hybridization of the gel

kb fragment consisting of plasmid pAMPF2-H linearized by digestion with

on the hybridized membrane, the sizes of the different bands are indicated

plasmid (Cantoral et al., 1987; Dıez et al., 1987)

was obtained, but when pAMPn2 was used, the

efficiency was about 60-fold higher, in the range of

the efficiencies described for AMA1-based autono-

mously replicating plasmids in different fungi (Gems

et al., 1991; Fierro et al., 1996). The percentage of

protoplast regeneration was in all experiments be-

tween 26% and 38%, as measured by the number of

colonies growing on nonselective minimal medium

with respect to the initial number of protoplasts in the

transformation reaction.

3.4. Transformation of P. nalgiovense with integrative

and autoreplicative plasmids by resistance to

phleomycin

The use of the antibiotic phleomycin and the ble

gene conferring resistance to this antibiotic is an

alternative transformation method for several fila-

mentous fungi (Kolar et al., 1988; Austin et al.,

1990). It was of interest to compare the transfor-

mation efficiencies of P. nalgiovense by phleomycin

tegrative in P. nalgiovense. (A) Total, undigested DNA from control

anes: 1, size marker (E-phage DNA digested with HindIII); 2, P.

M1; 4, AM2; 5, AM3; 6, AM4; 7, AM5; 8, AM6; 9, AM7; 10, size

from E. coli; 12, transformant YSX1; 13, YSX2; 14, YSX3; 15, size

in panel A transferred to a nylon membrane. The probe was an 8.6-

NotI. The size marker in lane 15 was pre-stained and thus it is visible

in kilobases at the right of the panel. See text for details.

Fig. 5. Hybridization of a nylon-blotted agarose gel containing SalI-

digested DNAs from control strains and different transformants with

plasmids pAMPn2 and pBPnYSX15. The probe was the same as in

Fig. 4. Lanes: 1, P. nalgiovense 16a (parental untransformed strain);

2, transformant AM1; 3, AM2; 4, AM3; 5, AM4; 6, AM5; 7, AM6;

8, AM7; 9, plasmid pAMPn2 extracted from E. coli and digested

with SalI; 10, transformant YSX1; 11, YSX2; 12, YSX3; 13, size

marker (E-phage DNA digested with HindIII). The sizes of the size

marker bands are indicated in kilobases at the right of the panel. See

text for details.

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248244

resistance, with those obtained using the pyrG

system. Two different plasmids, pULJL43 and

pAMPF21 (Fierro et al., 1996), were used for

transformation of strain P. nalgiovense 16a. Both

plasmids contain the ble gene expressed under the

promoter of the pcbC gene of P. chrysogenum

(Barredo et al., 1989); plasmid pAMPF21 contains

in addition the AMA1 fragment. The transformation

efficiency obtained with plasmid pAMPF21 (700–

800 transformants/Ag of DNA) was about eightfold

higher than that obtained with the integrative vector

pULJL43 (Table 1), which clearly suggests that

plasmid pAMPF21 functions as an autonomously

replicating plasmid in P. nalgiovense. In general,

the plasmids with the ble gene as selection marker

showed a lower transformation efficiency than those

with the pyrG, especially in the case of the AMA1-

carrying plasmids.

3.5. Southern analysis of P. nalgiovense transform-

ants with the autoreplicative plasmids

To confirm that AMA1-carrying plasmids are

maintained extrachromosomally and replicate auton-

omously in P. nalgiovense, total DNA was extracted

from seven transformants obtained with plasmid

pAMPn2 and three transformants obtained with plas-

mid pBPnYSX15, and Southern analysis was per-

formed both with total undigested DNA (Fig. 4) and

with SalI-digested DNA (Fig. 5). The results of the

hybridizations showed that plasmid pAMPn2 is

maintained as an extrachromosomal circular DNA

in all transformants (AM1 to AM7); no rearrange-

ments were found in at least five of the seven

transformants analyzed (transformants AM3 to

AM7), as shown by the same SalI restriction pattern

in the transformants with respect to the original

plasmid pAMPn2 (Fig. 5). In transformants AM1

and AM2, an additional weaker hybridization signal

of a different size in each case appeared in the

hybridization of the SalI-digested DNAs, which

suggests that a rearrangement of the plasmid DNA

has taken place in at least some of the nuclei in the

mycelia of those transformants; this was confirmed

by the results of the hybridization of the undigested

DNAs (Fig. 4), where transformants AM1 and to a

lesser extent AM2 showed a different band pattern

from that of the rest of transformants and from that

of the pure plasmid. In all the transformants with

plasmid pAMPn2, with the exception of transformant

AM1, the pattern of bands in the hybridization of the

undigested DNAs is very similar to that of pure

pAMPn2 (extracted from E. coli by alkaline lysis).

This result indicates that in the fungus, plasmid

pAMPn2 adopts topological conformations similar

to those in E. coli, although there are also differ-

ences, like the absence in the fungus of a plasmid

form present in E. coli, and vice versa. Another

difference is the higher intensity of the high molec-

ular weight form (upper band) in the fungus with

respect to E. coli; this upper band appears to be a

multimeric form of the plasmid; the presence of

multimers of pAMPn2 seems to be more frequent

in the fungi than in E. coli.

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248 245

In the area where total undigested DNA is

located in the gel, a very faint signal appears in

some of the transformants with plasmid pAMPn2.

This can be interpreted as a very low rate of

integration of the plasmid in the genome of the

fungus, though the possibility that some multimeric

form co-migrates with the total DNA and causes

that signal cannot be excluded. In contrast, trans-

formants obtained with plasmid pBPnYSX15

(YSX1, YSX2 and YSX3) showed a strong hybrid-

ization signal in the area of the total undigested

DNA (Fig. 4, lanes 12–14), indicating that plasmid

pBPnYSX15 integrates into the genome. However,

faint hybridization signals of much smaller size are

also visible in the three transformants; the position

of these bands does not agree with the expected

size of the different topological forms of plasmid

pBPnYSX15.

Fig. 6. Nature of the plasmids rescued in E. coli from P. nalgiovense tr

extracted from 28 E. coli clones was double-digested with the enzymes Xho

membrane was then hybridized with a 2.9-kb pBluescript probe (panel A) o

digestion with XhoI and EcoRI) (Panel B). Lane C (control) corresponds to

indicate that a few rescued plasmids (clones 11, 13 and 24) seem to contain

similar to that plasmid. The rest of the rescued plasmids show sizes betw

sequences and one or two fragments that hybridize with the pyrG probe, an

pBPnYSX15. The rest of the clones contain pBluescript sequences plus u

two transforming pBPnYSX15 plasmids (not involving pyrG sequences) o

In the filter with the SalI-digested DNAs (Fig. 5),

the three transformants with plasmid pBPnYSX15

show different hybridization patterns that are indica-

tive of different locations and number of copies

integrated in each case.

3.6. Recovery of the plasmid in E. coli

The definitive confirmation that plasmid pAMPn2

is maintained extrachromosomally in P. nalgiovense

was provided by transformation of E. coli with total

DNA extracted from transformants AM4 and AM6.

About 550–650 transformant colonies per microgram

of DNA were obtained. The analysis of plasmids

present in the E. coli transformants showed that the

restriction pattern in all clones tested was the same as

that of the original plasmid (not shown), indicating

that no rearrangements had occurred in any of them.

ansformants with plasmid pBPnYSX15 (see text). Plasmidic DNA

I +EcoRI, electrophoresed and blotted onto a nylon membrane. The

r with a 2.2-kb pyrG probe (obtained from plasmid pBPnYSX15 by

plasmid pBPnYSX15 digested with the same enzymes. The results

exclusively pBluescript sequences and have a size identical or very

een 3 and 4.5 kb. Clones 2, 12, 18, 21 and 22 contain pBluescript

d therefore they most likely represent truncated forms of the original

nidentified DNA, and might have arisen by recombination between

r between plasmid pBPnYSX15 and the chromosome of the fungus.

Table 2

Mitotic stability of plasmid pAMPn2 in P. nalgiovense

AM5 (%) AM7 (%) AM9 (%) YSX2 (%)

With selective

pressure

93.6

(F 5.9)

81.8

(F 8.2)

88

(F 7.6)

95.4

(F 4.2)

Without selective

pressure

84.3

(F 6.9)

70.6

(F 3.5)

67.9

(F 7.9)

96.9

(F 3.1)

Transformant YSX2 was obtained with the integrative plasmid

pBPnYSX15 (see text). Nonselective or selective conditions during

the development of the cultures for one generation were established

by adding or not adding uridine to the MEA sporulation medium.

The percentages represent the number of colonies growing on Cz

minimal medium with respect to the total number growing on Cz

supplemented with uridine.

The figures are the average of five independent experiments;

standard deviations are shown in brackets.

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248246

The same experiment performed with total DNA

from the P. nalgiovense transformants with plasmid

pBPnYSX15 (YSX1, YSX2 and YSX3) yielded 25–

35 E. coli transformants per microgram of DNA.

When plasmid DNA extraction was attempted from

these E. coli transformants, no plasmid molecules

could be observed in some of them, while in other

transformants, plasmids around 3–4.5 kb in size were

obtained, a size which is smaller than the 5.2 kb of

plasmid pBPnYSX15. This range of sizes is in accor-

dance with the size of the bands present in the

hybridized undigested DNAs (Fig. 4). These results

suggested that the small bands most likely represent

truncated forms of plasmid pBPnYSX15, containing

essentially pBluescript sequences, which are able to

replicate autonomously in P. nalgiovense and con-

serve sequences that allow their recovery in E. coli. To

test this hypothesis, 28 rescued plasmids extracted

from the E. coli clones were analyzed by Southern,

with plasmid pBluescript or with a 2.2-kb pyrG-

containing fragment as probes (Fig. 6). The results

indicated that a few of the rescued plasmids contained

exclusively pBluescript sequences, while others

contained also fragments 0.3–1 kb in size, comprising

pyrG sequences. Therefore, all these plasmids repre-

sent truncated/rearranged forms of pBPnYSX15. In

other plasmids, unidentified sequences were present

linked to the 3.0-kb pBluescript fragment, and they

might have arisen from recombination processes be-

tween two pBPnYSX15 molecules or between

pBPnYSX15 and the chromosome of the fungus.

3.7. Mitotic stability of autoreplicative plasmids in P.

nalgiovense

The mitotic stability of the autoreplicating plasmid

pAMPn2 in P. nalgiovense was studied, as indicated

in Material and Methods, using conidia from trans-

formants AM5, AM7 and AM9, obtained after one

generation of growth in MEA medium with or without

selective pressure. The results (Table 2) showed that

plasmid pAMPn2 is highly stable in P. nalgiovense:

between 82% and 94% of conidia kept the plasmid

after one generation under selective pressure. This

percentage is roughly indicative of the number of

mitosis in which the plasmid is distributed between

the phialide and the generated conidia. After one

generation without selective pressure, between 68%

and 84% of conidia conserved the plasmid. The

integrative plasmid pBPnYSX15 showed a higher

stability, 96.9% after one generation without selective

pressure, which correlated well with the fact that it

integrates into the genome.

4. Discussion

Genetic manipulation of P. nalgiovense has been

limited, so far, to the introduction of the lysostaphin

gene from Staphylococcus staphylolyticus (Geisen et

al., 1990) and the glucose oxidase gene from A. niger

(Geisen, 1995, 1999) with the aim of increasing the

antibacterial properties of the fungus. The amdS gene

of A. nidulans was used as selection marker in all

cases (Geisen and Leistner, 1989).

The purpose of this work was to develop various

molecular tools useful for efficient transformation and

gene cloning in P. nalgiovense, in order to facilitate

the molecular studies about secondary metabolite

biosynthesis and other metabolic aspects related to

the use of this fungus as starter in the food industry.

The availability of a pyrG mutant strain and a

transformation system based on uridine auxotrophy

complementation with the homologous pyrG gene

described in this article offers several advantages over

transformation based on antibiotic resistance or on

acetamidase utilization using the amdS gene (Geisen

and Leistner, 1989; Geisen et al., 1990; Geisen, 1995):

(1) The transformation efficiency is much higher

when using the pyrG gene than with either of the

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248 247

other systems; (2) the use of the homologous pyrG

gene as selection marker allows counter-selection

using 5-FOA, which is very useful to perform suc-

cessive transformations on the same strain, for in-

stance, to disrupt several genes, eliminating the

previously used pyrG gene by recombination and 5-

FOA selection (d’Enfert, 1996); (3) any construction

can be introduced in a single copy at the pyrG locus

by gene targeting using a mutant pyrG gene (contain-

ing a point mutation) in the transforming vector

(Gouka et al., 1995; Kosalkova et al., 2000); (4) there

is much concern about the use of heterologous genes

to get new improved strains of microorganisms used

in the food industry, and in many countries, strict

regulations on this issue have been implemented; the

use of a homologous gene as selection marker, such as

the pyrG, overcomes this problem.

The use of autonomously replicating vectors in P.

nalgiovense is also of great interest. The transforma-

tion efficiency achieved with autonomous-replicating

vectors and the possibility of recovering these plasmids

in E. coli make them excellent tools for constructing P.

nalgiovense genomic libraries to clone genes by com-

plementation, which will help the molecular studies on

secondary metabolite pathways and genes of interest in

this fungus. The behaviour of AMA1-based plasmids in

P. nalgiovense is similar to what has been previously

reported for P. chrysogenum (Fierro et al., 1996).

However, there are differences in the mitotic stability

of the plasmids and in the multimeric forms that occur

in both Penicillium species. Mitotic stability is higher

in P. nalgiovense, and the multimeric forms are clearly

predominant in P. chrysogenum, whereas in P. nalgio-

vense, monomeric and multimeric forms appear at a

similar frequency (Fig. 4). In both fungi, AMA1-based

plasmids show a low degree of reorganization and can

be easily rescued in E. coli.

An interesting result is the presence of small

plasmids, apparently derived from the integrative

plasmid pBPnYSX15, which are maintained auton-

omously in P. nalgiovense. Plasmid pBPnYSX15

integrates in the genome of the fungus, as shown

by the results of the Southern analysis (Figs. 4 and

5) and by its mitotic stability (Table 2), but small-

size DNA molecules also appear when undigested

DNA is hybridized (Fig. 4). These small plasmids

correspond to truncated forms of pBPnYSX15 that

conserve E. coli sequences enabling them to be

rescued in E. coli at low rate (Fig. 6). There are

reports on plasmids lacking known replication ori-

gins that replicate autonomously in other fungi, for

instance, plasmid pUT737 in Botrytis cinerea

(Santos et al., 1996). Autonomous plasmid replica-

tion occurs much more frequently in mucoraceous

fungi than in other groups of filamentous fungi

(Wostemeyer et al., 1987).

In summary, novel plasmids containing homolo-

gous DNA sequences and P. nalgiovense host strains

have been developed and are now available for

modification of this filamentous fungus widely used

in the food industry.

Acknowledgements

This work was supported by grants of the

Diputacion de Leon (DLE 01/97) and Proyecto

Generico of the ADE (Junta de Castilla y Leon,

Valladolid, Spain) (10-2/98/LE/0003). F. Laich re-

ceived a fellowship of the Agencia Espanola de

Cooperacion con Iberoamerica (AECI) of the Ministry

of Foreign Affairs (Madrid, Spain). We thank J.

Merino, B. Martın and M. Corrales for excellent

technical assistance.

References

Aleksenko, A., Clutterbuck, A.J., 1997. Autonomous plasmid rep-

lication in Aspergillus nidulans: AMA1 and MATE elements.

Fungal Genetics and Biology 21, 373–387.

Andersen, S.J., Frisvad, J.C., 1994. Penicillin production by Penicil-

lium nalgiovense. Letters in AppliedMicrobiology 19, 486–488.

Austin, B., Hall, R.M., Tyler, B.M., 1990. Optimized vectors and

selection for transformation of Neurospora crassa and Aspergil-

lus nidulans to bleomycin and phleomycin resistance. Gene 93,

157–162.

Barredo, J.L., Cantoral, J.M., Alvarez, E., Dıez, B., Martın, J.F.,

1989. Cloning, sequence analysis and transcriptional study of

the isopenicillin N synthase of Penicillium chrysogenum AS-P-

78. Molecular and General Genetics 216, 91–98.

Berwall, J.S., Dincho, D., 1994. Molds as protective cultures for

raw dry sausages. Journal of Food Protection 57, 928–930.

Bos, C.J., Stadler, D., 1996. Mutation. In: Bos, C.J. (Ed.), Fungal

Genetics: Principles and Practice. Marcel Dekker, New York,

pp. 13–42.

Cantoral, J.M., Dıez, B., Barredo, J.L., Alvarez, E., Martın, J.F.,

1987. High-frequency transformation of Penicillium chrysoge-

num. Bio/Technology 5, 494–497.

d’Enfert, C., 1996. Selection of multiple disruption events in As-

F. Fierro et al. / International Journal of Food Microbiology 90 (2004) 237–248248

pergillus fumigatus using the orotidine-5V-decarboxylase gene,

pyrG, as a unique transformation marker. Current Genetics 30,

76–82.

Dıez, B., Alvarez, E., Cantoral, J.M., Barredo, J.L., Martın, J.F.,

1987. Selection and characterization of pyrG mutants of Pen-

icillium chrysogenum lacking orotidine-5-phosphate decarboxy-

lase and complementation by the pyr4 gene of Neurospora

crassa. Current Genetics 12, 277–282.

Farber, P., Geisen, R., 1994. Antagonistic activity of the food-re-

lated filamentous fungus Penicillium nalgiovense by the produc-

tion of penicillin. Applied and Environmental Microbiology 60,

3401–3404.

Fierro, F., Kosalkova, K., Gutierrez, S., Martın, J.F., 1996. Auton-

omously replicating plasmids carrying the AMA1 region in Pen-

icillium chrysogenum. Current Genetics 29, 482–489.

Fink-Gremmels, J., El-Banna, A.A., Leistner, L., 1988. Developing

mould starter cultures for meat products. Fleischwirtschaft 68,

1292–1294.

Geisen, R., 1993a. Cloning of a protease gene from Penicillium

nalgiovense by expression in Escherichia coli. Letters in Ap-

plied Microbiology 16, 303–306.

Geisen, R., 1993b. Fungal starter cultures for fermented foods:

molecular aspects. Trends in Food Science and Technology 4,

251–256.

Geisen, R., 1995. Expression of the Aspergillus niger glucose ox-

idase gene in Penicillium nalgiovense. World Journal of Micro-

biology and Biotechnology 11, 322–325.

Geisen, R., 1999. Inhibition of food-related pathogenic bacteria by

god-transformed Penicillium nalgiovense strains. Journal of

Food Protection 62, 940–943.

Geisen, R., Leistner, L., 1989. Transformation of Penicillium nal-

giovense with the amdS gene of Aspergillus nidulans. Current

Genetics 15, 307–309.

Geisen, R., Standner, L., Leistner, L., 1990. New mould starter cul-

tures by genetic modification. Food Biotechnology 4, 497–504.

Gems, D., Johnstone, I.L., Clutterbuck, J., 1991. An autonomously

replicating plasmid transforms Aspergillus nidulans at high fre-

quency. Gene 98, 61–67.

Gouka, R.J., Hessing, J.G.M., Stam, H., Musters, W., van den

Hondel, C.A.M.J.J., 1995. A novel strategy for the isolation

of defined pyrG mutants and the development of a site-specific

integration system for Aspergillus awamori. Current Genetics

27, 536–540.

Grazia, L., Romano, P., Bagni, D., Roggiani, D., Guglielmi, G.,

1986. The role of moulds in the ripening process of salami.

Food Microbiology 3, 19–25.

Hanahan, D., 1986. Techniques for transformation of E. coli In:

Glover, D.M. (Ed.), DNA Cloning, vol. 1. IRL Press, Oxford,

pp. 109–135.

Kolar, M., Punt, P.J., van den Hondel, C.A.M.J.J., Schwab, H.,

1988. Transformation of Penicillium chrysogenum using domi-

nant selection markers and expression of an Escherichia coli

lacZ fusion gene. Gene 62, 127–134.

Kosalkova, K., Marcos, A.T., Fierro, F., Hernando-Rico, V., Gutier-

rez, S., Martın, J.F., 2000. A novel heptameric sequence

(TTAGTAA) is the binding site for a protein required for high

level expression of pcbAB, the first gene of the penicillin bio-

synthesis in Penicillium chrysogenum. Journal of Biological

Chemistry 275, 2423–2430.

Laich, F., Fierro, F., Cardoza, R.E., Martın, J.F., 1999. Organ-

ization of the gene cluster for biosynthesis of penicillin in

Penicillium nalgiovense and antibiotic production in cured

dry sausages. Applied and Environmental Microbiology 65,

1236–1240.

Laich, F., Fierro, F., Martın, J.F., 2003. Isolation of Penicillium

nalgiovense strains impaired in penicillin production by disrup-

tion of the pcbAB gene and application as starters on cured meat

products. Mycological Research 107 (6), 717–726.

Larsen, T.O., Breinholt, J., 1999. Dichlorodiaportin, diaportinol,

and diaportinic acid: three novel isocoumarins from Penicillium

nalgiovense. Journal of Natural Products 62, 1182–1184.

Leistner, L., 1984. Toxinogenic penicillia occurring in feeds and

foods: a review. Food Technology (Austria) 36, 404–406.

Leistner, L., 1990. Mould-fermented foods: recent developments.

Food Biotechnology 4, 433–441.

Leistner, L., Pitt, J.I., 1977. Miscellaneous Penicillium toxins. In:

Rodricks, J.V., Hesseltine, C.W., Mehlmann, M.A. (Eds.), My-

cotoxins in Human and Animal Health, Park Forest South, IL.

Pathotox Publishers, IL, pp. 639–653.

Lucke, F.K., 1986. Microbiological processes in the manufacture of

dry sausage and raw ham. Fleischwirtschaft 66, 1505–1509.

Pestka, J., 1995. Fungal toxins in raw and fermented meats. In:

Campbell-Platt, G., Cook, P.E. (Eds.), FermentedMeats. Blackie

Academic and Professional, London, pp. 194–216.

Sambrook, J., Fritsh, E.F., Maniatis, T., 1989. Molecular Cloning:

A Laboratory Manual. Cold Spring Harbor Laboratory Press,

Cold Spring Harbor, New York.

Santos, M., Vallejo, I., Rebordinos, L., Gutierrez, S., Collado, I.G.,

Cantoral, J.M., 1996. An autonomously replicating plasmid

transforms Botrytis cinerea to phleomycin resistance. FEMS

Microbiology Letters 137, 153–158.

Sutic, M., Ayres, J.C., Koehler, P.E., 1972. Identification and afla-

toxin production of molds isolated from cured hams. Applied

Microbiology 23, 656–658.

Wang, H.L., Kim, S.H., Siu, H., Breuil, C., 1999. Transfor-

mation of sapstaining fungi with hygromycin B resistance

plasmids pAN7-1 and pCB1004. Mycological Research 103,

77–80.

Wostemeyer, J., Burmester, A., Weigel, C., 1987. Neomycin

resistance as a dominantly selectable marker for transforma-

tion of the zygomycete Absidia glauca. Current Genetics 12,

625–627.