Genet. Sel. Evol. 33 (Suppl.

1) (2001) S353-S364 8353

© INRA, EDP Sciences, 2001

Original article

Use of specific DNA probes for the rapid

characterization of yeasts isolated
from complex biotopes
Anne-Marie DAVILA, Monique DIEZ, Mauricio CORREDOR,
Yves PAGOT, Michele WINKLER, Claude GAILLARDIN,
Serge CASAREGOLA *

aCollection de levures d'interet biotechnologique, Genetique moleculaire et

cellulaire, Cnrs URA1925, Inra UMR216, Institut national agronomique
Paris-Grignon, 78850 Thiverval-Grignon, France

Abstract - We describe a rapid method developed to identify yeast species com-

monly found during cheese ripening. It is based on the isolation of species-specific
sequences and their hybridization with coarse preparations of genomic DNA. Sev-
eral strategies were followed for the construction of probes: PCR amplification of
sequences available in databases, random cloning of genomic DNA fragments, specific
RAPD fragments and PCR-amplified ribosomal DNA fragments. After validation, the
probes were applied to the characterization of 400 yeast strains isolated from various
French cheeses. Since the strains had been previously identified with classical diag-
nostic tests, we were able to compare molecular and conventional identification. In
addition, the specific probes for one of the species, Debaryomyces hansenii, were used
successfully in colony hybridization experiments. The probes developed here proved
to be very useful for the screening of large yeast collections and for the assessment of
biodiversity within the yeast flora in cheese.
yeast / cheese / molecular identification / hybridization / biodiversity

Resume - Utilisation de sondes ADN specifiques pour la caracterisation

rapide de levures isoIees d'ecosystemes complexes. Au cours de ce travail,
nous avons developpe une methode a la fois rapide et sensible pour l'identification
des levures couramment isolees pendant l'affinage du fromage. Son principe re-
pose sur l'hybridation de sequences specifiques sur les preparations brutes d' ADN
genomique des souches a identifier. Pour la construction des sondes, nous avons du
suivre plusieurs strategies differentes: amplification de sequences disponibles dans les
bases de donnees, clonage au hasard de sequences d' ADN genomique, isolement de
sequences specifiques obtenues par amplification RAPD d' ADN genomique ou par
amplification d' ADNr. Apres leur validation, les son des ont pu etre mises en ceuvre

* Correspondence and reprints

E-mail: serge@grignon.inra.fr
S354 A.-M. Davila et al.

dans differentes applications. Elles ont ete utilisees pour cribler 400 souches de levures
isolees de differents from ages fran«ais. Les souches ayant ete prealablement identi-
fiees par des tests classiques, nous avons pu comparer les deux types de methodes
d'identification, moleculaire et conventionnelle. La son de specifique de D. hansenii a
He testee avec succes dans des essais d'hybridation sur colonie. Au total, ces son des
se sont revelees etre de bons outils pour le suivi et l'inventaire de la biodiversite des
levures fromageres.
levures / fromage / identification moleculaire / hybridation / biodiversite

1. INTRODUCTION

Cheese making is probably one of the most ancient means used for milk
preservation. It proceeds in 3 steps: curdling, draining of whey and ripening.
The latter step consists of an enzymatic digestion of the curd constituents and it
confers on cheeses their characteristic textures and flavors. Enzymes are already
present in the milk and also provided by a complex flora comprising bacteria,
molds and yeasts [17]. Yeasts stimulate the growth of bacteria by metabolizing
lactic acid, producing NH3 and releasing vitamins and nutrients [10]. They
contribute directly to the ripening process by providing proteolytic and lipolytic
enzymes. They facilitate the growth and penetration of molds by opening up
the texture of blue-veined cheeses. They are thought to contribute to the taste
of cheese by providing specific flavors [9,12,17,20]. It is also reported that
yeasts such as Debaryomyces hansenii inhibit the growth of spoilage bacteria
of the genus Clostridium [8,11].
Many yeast species are encountered in milk, dairies and brine but only a
few of them are able to adjust to the cheese ecosystem. Yeast maintenance and
development are very likely due to their ability to grow in the presence of salt
at low temperature and/or to metabolize lactic and citric acids.
Several yeast species are associated with the ripening of cheese: D. hansenii,
Kluyveromyces lactis, Kluyveromyces marxianus, Candida zeylanoides,
Yar'rOwia lipolytica, Geotrichum candidum, Saccharomyces cerevisiae, Torulas-
pora delbrueckii, Pichia fermentans, and Pichia menbranaefaciens [5]. Their
prevalence depends on the type of cheese considered. Texture and flavor devel-
opment in cheese is linked to the proportion of each species present at a given
time in connection with the biochemical activities of these microorganisms that
develop at the surface or within the cheese.
Previous studies on yeast populations present in cheese were based on the
identification of the species with conventional diagnostic testing [2,3,20,25,
26]. The use of morphological, physiological and biochemical criteria is time-
consuming. At least three weeks are required to identify yeast species following
this approach, which also necessitates painstaking manipulation and laborious
computing steps. It thus cannot be used for the systematic study of complex
flora in cheese. Here, we constructed specific molecular probes that allow the
species identification of hundreds of strains belonging to the predominant yeast
 Specific DNA probes for yeast characterization S355

species involved in cheese ripening in less than 2 weeks. The construction,

validation and characteristics of these probes are reviewed. The probes were
applied to the screening of a set of yeast strains isolated from various cheeses
during processing.

2. MATERIALS AND METHODS

2.1. Yeast strains, bacterial strains and growth conditions

The reference strains used in this study are listed in Table 1. Isolates were
identified using the morphological and biochemical tests described [1,14]. Yeast
cells were routinely grown in YPD (1% yeast extract, 1% peptone, 1% glucose)
at 28 QC. Escherichia coli DH5a cells were routinely grown in LB (2% LB
Broth base, 0.5% NaCl) at 37 QC.

2.2. DNA techniques

Common DNA manipulations were as described previously [24]. Genomic

DNA extractions for cloning were performed as described previously [23]. Rapid
genomic DNA extraction was derived from a published method [13]. After 48-h
cultivation in 10 ml of YPD liquid medium, cells were washed and resuspended
in 0.2 ml of cell lysis buffer (10 mM Tris pH 8, 1 mM EDTA, 100 mM NaCl, 2%
Triton, 1% SDS). 0.3 ml phenol/CHCh and 0.3 g of glass beads were added
to the mixture and vortexed for 4 min. Then 0.2 ml TE (10 mM Tris pH
8, 1 mM EDTA) were added before 5 min centrifugation. Supernatant (1.6 ml)
was precipitated once in presence of 1 ml 100% (V/V) ethanol. After 2 min
centrifugation, the DNA pellet was washed with 1 ml 70% (V/V) ethanol and
dried quickly. Then it was resuspended in 200 ILl TE and incubated for 15 min
at 37 QC with 1 ILl of a 10 mg/ml RNase solution. For cloning, pBluescript®
SK- (Stratagene) was digested to completion with EcoRV and its ends were de-
phosphorylated with alkaline phosphatase (Boehringer-Mannheim, Germany).
The fragments were treated with a combination of the Klenow fragment of poly-
merase I and T4 DNA polymerase, to generate blunt ends, and were inserted
into pBluescript SK-.

2.3. Electroporation

DH5a E. coli cells in the late exponential phase were prepared by washing
3 times in ice-cold H 2 0. Glycerol (10% final concentration) was added for the
storage of the cells at -80 QC. Cells were transformed by electroporation using
a Bio-Rad Gene Pulser n® under the following conditions: 2,500 V, 100 n
S356 A.-M. Davila et al.

and 25 ILF. Cells were recovered with 1 ml LB, incubated for I-h at 37 QC and
plated out on 80 jLg.ml- 1 ampicillin containing LB media.

2.4. PCR amplification

PCR amplifications with NL1 and NL4 primers were performed as described
previously [4]. NL1 and NL4 primers, described in [21], are universal primers
that are used to amplify the 5'end of the large subunit of the ribosomal RNA
gene spanning the variable domains D1 and D2 (D1/D2 rDNA). For RAPD,
we used conditions similar to those described [27] with the deca-primers of
Operon Technology (USA). For the other PCR amplifications, yeast genomic
DNA (50 ng) was used as template under the following conditions: 4 min
at 94 QC, 25/35 cycles of 30 s at 94 QC, 30 s at the Tm required, 1 min per
kb to be amplified at 72 QC followed by 5 min at 72 QC, with 2.5 units of
thermostable polymerase (Appligene, France) and 1.5 units of Pfu polymerase
(Biolabs, USA) in the recommended buffer.

2.5. Purification of PCR products

PCR products for hybridization were separated in 0.6% Seaplaque (FMC,

USA) low-melting agarose gels, excised in gel bands, and labeled. PCR prod-
ucts for sequencing were purified using the Qiaquick DNA purification kit (Qi-
agen, Germany), according to the manufacturer's recommendations.

2.6. DNA/DNA hybridization

Genomic DNA was deposited on GeneScreen nylon membrane (DuPont,

USA). Membranes were incubated on top of a 0.1 M NaOH/1.5 M NaCI so-
lution for 1 min, immersed in a 2x SSC/0.2 M Tris HCI pH 7.5 solution for
1 min and air dried. DNA was linked to the membrane with UV-irradiation.
DNA probes were prepared from plasmids by digestion or amplified by specific
primers and purified after gel separation. They were labeled with (a- 32 P)dCTP
using the Megaprime labeling kit (Amersham, UK). DNA/DNA hybridizations
were carried out overnight at 65 QC in 0.5 M Na 2HP0 4/1 mM EDTA/7% SDS
buffer. Washing steps were 3 x 5 min in 40 mM Na2HP04/1 % SDS followed
by 2 x 15 min in 0.1 x SSC/0.1% SDS buffer [6].

2.7. Colony hybridization

Colonies were transferred with toothpicks onto a Hybond H+ membrane

which was placed DNA side up on YPD agar medium. After 48-h cultivation,
the membrane was treated with (10 mM Tris pH 8, 1 mM EDTA, 1 M sorbitol,
3 mg/ml DTT) buffer for 10 min. Protoplasts were made with a mixture of
 Specific DNA probes for yeast characterization S357

zymolyase and cytohelicase in the presence of sorbitol. Cells were lysed with
1 M NaOH for 10 min. Membranes were neutralized with 0.5 M Tris pH 7.5,
1.5 M NaCI solution. DNA was linked to the membrane with UV-irradiation.
DNA/DNA hybridizations were carried as described above. The probe was
labeled with dioxigenin-dUTP according to the manufacturer's recommenda-
tions. Hybridization was revealed with CSPD®.

2.8. Sequencing

DNA was sequenced on both strands essentially as described previously [19]

with 200 ng of PCR product and 4 pmols of primers. NL1 and NL4 primers
are described in [21]. Sequences were compiled and analyzed using the Staden
package [7] and FASTA [22] in the GCG environment (Genetics Computer
Group, Madison, Wisconsin, USA).

3. RESULTS AND DISCUSSION

3.1. Construction and validation of the DNA probes

We constructed specific probes for the identification of the major yeast

species present in cheese during the ripening process [5]. The strains utilized
are listed in Table 1. Except for Geotrichum candidum, the type strains of the
other species were used.
Various strategies were adopted for the construction of probes. When-
ever possible, probes were deduced from sequences available in databases.
This was the case for Saccharomyces cerevisiae, Kluyveromyces marxianus,
Kluyveromyces lactis and G. candidum. We designed specific primers and
PCR-amplified parts of various genes: CDC15 for S. cerevisiae (CDC15 is
an essential gene and cannot be deleted), MIGl for K. lactis and K. marxi-
anus (these genes are orthologs and are sufficiently divergent in sequence to be
species-specific) and LIP2 for G. candidum (one of the few available genes of
this species).
For most species, DNA sequences were not available. Probes were obtained
after enzymatic digestion of genomic DNA and random cloning of 2 to 3 kb
fragments. This method was utilized for Candida zeylanoides, Pichi a guil-
liermondii, Pichia jermentans, Pichia anomala, Pichia menbmnaejaciens and
Zygosaccharomyces rouxii. A large number of D. hansenii strains were typed by
RAPD. Some primers gave major amplification fragments of respectively 1.0 kb
and 1.8 kb which were common to all the strains tested. These two fragments
were isolated and cloned. They proved to be highly specific for D. hansenii
when used as probes.
Yarrowia lipolytica and Torulaspom delbrueckii failed to hybridize with the
probes developed by the methods described above. This could be due to the
 iJJ
~
Table I. List of strains used for the construction of probes. Q1
(f)

Strain Species Other Origin Country

number collection
CLIB 199 Candida zeylanoides T CBS 619 Blastomycotic
macroglossia
CLIB 197 Debaryomyces hansenii T var. CBS 767 Cherry Denmark
hansenii

CBS 109-12 Geotrichum candidum Milk USA

CLIB 196 Kluyveromyces lactis NT CBS 683 Cheese United Kingdom >
CLIB 282 Kluyveromyces marxianus T CBS 712 ~
CLIB 284 Pichia anomala NT CBS 5759 USA t:I
~
CLIB 198 Pichia jermentans T CBS 187 Buttermilk The Netherlands :5.
~
CLIB 186 Pichia guilliermondii T CBS 566 Sputum Hungary ~
CLIB 212 Pichia membranaejaciens T CBS 107 ~
CLIB 227 Saccharomyces cerevisiae NT CBS 1171 Beer The Netherlands
CLIB 230 Torulaspora delbrueckii NT CBS 1146
CLIB 183 Yarrowia lipolytica T CBS 6124 Maize-processing USA
CBS 732 Zygosaccharomyces rouxii NT Concentrated plant Italy
grape must
(T): type strain
(NT): nea-type strain.
 Specific DNA probes for yeast characterization S359

Table 11. Strategies for the construction and the validation of specific probes.

Strain Type of probe Type of validation

Candida zeylanoides 2 A and B
Debaryomyces hansenii 3 A and B
Geotrichum candidum 1 A and B
Kluyveromyces lactis 1 A and B
Kluyveromyces marxianus 1 A and B
Pichia anomala 2 A and B
Pichia fermentans 2 A and B
Pichia guilliermondii 2 A and B
Pichia membranaefaciens 2 A and B
Saccharomyces cerevisiae 1 A and B
Torulaspora delbrueckii 4 A and B
Yarrowia lipolytica 4 A and B
Zygosaccharomyces rouxii 2 A
(1): sequence available in databases
(2): random cloning of 2 to 3 kb fragments
(3): characteristic fragment from RAPD amplification
(4): rDNA amplification (NTS or NTS2)
(A): validation of the inter-species specificity
(E): validation against strains belonging to the same species

DNA extraction procedure since we used a very crude and rapid method which
enables us to treat a large number of strains. For Y. lipolytica and T. del-
brueckii, we therefore PeR-amplified the NTS (Non-Transcribed Sequence) and
NTS2 of ribosomal DNA, respectively. These sequences are naturally repeated
in the genome and result in efficient probes whatever the concentration or pu-
rity of DNA preparations for hybridization.
Probes were then validated (Tab. II) by hybridization with genomic DNA
deposited on filters (Materials and Methods). The probes were classified as
specific when no positive signal was detected in hybridization experiments with
the DNA of the other yeast species commonly found in cheese. Most of the
probes were also tested against several strains of the same species isolated from
various ecosystems (Tab. II). D. hansenii was submitted to a careful validation
which also covered other species of the Debaryomyces genus.
We developed 14 DNA probes specific for the yeast species commonly iso-
lated during cheese ripening. Their utilization combines rapidity, sensitivity
and simplicity since 96 DNA samples can be deposited on the same membrane
and several menbranes can be treated together. It is therefore possible to carry
out the identification of hundreds of strains at a time. Nevertheless, genomic
S360 A.-M. Davila et al.

DNA preparation remains the limiting step of this method. The range of DNA
concentrations for optimal hybridization conditions also has to be established.

3.2. Screening of yeast strains isolated from the cheese and dairy
environment

The probes described above were used to screen 400 yeast strains isolated
between the 70's and the 80's [2,20,25,26]. This set of strains covers various
cheese productions from distinct regions of France. Most isolates came from
milk, from cheese at different steps of the processing or from the dairy envi-
ronment. Some strains were isolated from the environment i.e. on cattle and
feeds. Strains were previously identified according to Lodder [18] or following
a simplified set of classical tests [3].
Genomic DNA of the 400 strains was prepared and deposited on Genescreen
membrane. The membranes were successively hybridized with the radiolabeled
probes developed in this work. The results are reported in Table Ill. 311 strains
were unambiguously identified by hybridization. No cross-hybridization was
observed with the different probes, indicating that the method is very specific.
A large database of the variable regions D 1 and D2 of the DNA encoding the
large subunit of the ribosomal RNA (Dl/D2 rDNA) sequences from over 500
yeast species including 200 Candida spp., has been recently published [15,16].
This database is now widely used and was applied to identify the strains that
gave either no signal or a very faint signal after hybridization. We used NLl
and NL4 primers to PCR-amplify over 550 bp of the Dl/D2 rDNA [21] as
described in Materials and Methods. The sequences of the PCR products were
then compared to sequences present in the database. Dl/D2 rDNA sequencing
confirmed the identification of the strains that hybridized weakly to one of
the probes (Tab. Ill). In addition, sequencing was systematically performed on
both strands to avoid ambiguity.
We also unambiguously identified 17 strains that did not give any signal in
hybridization. They belong to species not commonly found during cheese matu-
ration [5]: Candida oeleophila, Candida sake, Candida pseudoglaebosa, Candida
boidinii, Saccharomyces castellii, Debaryomyces castellii and Williopsis ealifor-
nica. We did not construct probes for these species since they are rarely present
in cheese. Indeed, these strains, which belong to 7 species, represent only 4%
of the strains screened, indicating that sequencing was more suitable for the
analysis of this small sample. 27 strains remained unidentified by Dl/D2 rDNA
sequencing, indicating that these species are neither referenced nor described
for the moment. Among them, four groups of respectively 2, 2, 3 and 6 strains
exhibited the same Dl/D2 rDNA sequence. The other sequences were unique.
Interestingly, 60% of these unidentified strains were not directly isolated from
cheese during processing. Sequence analysis will be described in a future report
(Davila et al., manuscript in preparation).
 Specific DNA probes for yeast characterization S361

Table Ill. Identification of yeasts isolated from cheese and dairy at the species level.

Strain Number of strains Number of strains

detected by identified with
hybridization D1/D2 sequencing
Associated with cheese ripening
Candida zeylanoides 23 0
Debaryomyces hansenii 61 0
Geotrichum candidum 0 0
Kluyveromyces lactis 83 17(2)
Kluyveromyces marxianus 63 12(2)
Pichi a anomala 0 0
Pichia fermentans nd 13
Pichia membranaefaciens 0 0
Saccharomyces cerevisiae 54 3(2)
Torulaspora delbrueckii 5 0
Yarrowia lipolytica 13 0
Zygosaccharomyces rouxii 0 0
Opportunistic
Candida boidinii nd 1
Candida oeleophila nd 3
Candida pseudoglaebosa nd 1
Candida sake nd 2
Debaryomyces castellii nd 2
Pichi a guilliermondii 9 0
Saccharomyces castelli nd 6
Williopsis californica nd 2
Undetermined 89(1) 27
Total 400 89
(1): weak signal or no signal in hybridization.
(2): weak signal in hybridization confirmed by D1/D2 rDNA sequencing [15,16,21].
(nd): not determined.

The comparison of molecular and conventional identification methods re-

vealed some discrepancies. D. hansenii and S. cerevisiae gave a good con-
vergence since C. famata is the anamorph of D. hansenii and S. italicus is a
variety of S. cerevisiae. Most of the discrepancies were, however, due to the
use of a simplified method of identification at the early stage of this project.
It was shown that the method can be imprecise in some cases (Davila et al.,
8362 A.-M. Davila et al.

A B

Figure 1. Colony hybridization with a probe specific for D. hansenii A: colonies

on YPD agar medium; B: Hybond H+ replicate, dioxigenin-dUTP labeling, CSPD@
detection; 1: D. hansenii colonies; 2: S. cerevisiae colonies; 3: K. lactis colonies.
Note that the film was largely over-exposed to ensure that no aspecific hybridization
could be detected.

manuscript in preparation). For instance, 12% of the K. marxianus strains

were mixed up with other species of the Kluyveromyces genus in conventional
identification. Real discrepancies, evidenced by a complete conventional identi-
fication [1,18], were in fact very rare and could not be solved. This is a problem
due to the set-up of this new molecular method. Universal and systematic use
of D1/D2 sequencing will help to redefine species.
A relevant number of strain species identified by hybridization were tested
with D1/D2 sequencing. Their identification was always confirmed, indicating
the accuracy of our method.

3.3. Colony hybridization

One of the two specific probes for D. hansenii [71 was tested in colony hy-
bridization experiments. Thirty six D. hansenii, S. cerevisiae and K. lactis
colonies were deposited on solid medium, grown and transferred onto Hybond
membrane. Cells were lysed and DNA was bound to the membrane as described
in Materials and Methods. As shown in Figure 1, the probe hybridized to all
D. hansenii replicates but not to S. cerevisiae or K. lactis ones. The colony
hybridization methodology seems even more suitable for the identification of
large number of strains as it eliminates the genomic DNA preparation step.

3.4. Concluding remarks

The study of complex flora by classical diagnostic tests is very painstak-

ing and time-consuming. In this work, we review the development of specific
probes for an identification of yeasts which combines simplicity, rapidity and
 Specific DNA probes for yeast characterization S363

accuracy. These probes were successfully utililled to identify at the species level
400 yeast strains. Furthermore, D. hansenii-specific probes could be used in
colony hybridization experiments. These probes had proved to be efficient tools
for the analysis of yeast species biodiversity. This method could be improved
by using a mixture of probes labeled with different fluorescent dyes to reduce
the number of hybridizations.

ACKNOWLEDGEMENTS

Annie Auger and Andree Lepingle are gratefully acknowledged for sequenc-
ing. This work was supported by a grant from the Bureau des ressources
genetiques. Mauricio Corredor was supported by a fellowship from the French
ministere des Relations Exterieures.

REFERENCES

[1] Barnett J.A., Payne R.W., Yarrow D., Yeasts: characteristics and identification,
2nd ed. Cambridge University Press, Cambridge, 1990.
[2] Baroiller C., Schmidt J.L., Contribution A l'etude de l'origine des levures du
from age de Camembert, Lait 70 (1990) 67-84.
[3] Baroiller C., Schmidt J.L., Mise au point d'une grille simplifiee d'identification
des principales especes de levures presentes dans les fromages, Lait 64 (1984)
16-28.
[4] Casaregola S., Nguyen H.V., Lepingle A., Brignon P., Gendre F., Gaillardin C.,
A family of laboratory strains of Saccharomyces cerevisiae carry rearrangements
involving chromosomes I and Ill, Yeast 14 (1998) 551-564.
[5] Choisy C., Desmazeaud M., Gueguen M., Lenoir J., Schmidt J.L., Tourneur C.,
L'affinage du fromage. Les phenomenes microbiens, in: Le fromage, 3e ed. Tec.
& Doc., Lavoisier, Paris, 1997, pp. 377-446.
[6] Church G.M., Gilbert W., Genomic sequencing, Proc. Natl. Acad. Sci. USA 81
(1984) 1991-1995.
[7] Corredor M., Davila A.-M., Gaillardin C., Casavegola S., DNA probes specific
for the yeast species Debaryomyces hansenii: useful tools for rapid identification,
FEMS Microbiol. Lett. 193 (2000) 171-177.
[8] Dear S., Staden R., A sequence assembly and editing program for efficient man-
agement of large projects, Nucleic Acids Res. 19 (1991) 3907-3911.
[9] Deiana P., Fatichenti F., Farris G.A., Mocquot G., Lodi R., Todesco R., Cecchi
L., Metabolization of lactic and acetic acids in Pecorino Romano cheese made
with a combined starter of lactic acid bacteria and yeast, Lait 64 (1984) 380-394.
[10] Devoyod J.J., Yeasts in cheese-making, in: Spencer J.F.T., Spencer D.M. (eds.),
Yeast technology, Springer-Veriag, Berlin, 1990, pp. 228-240.
[11] Devoyod J.J., Desmazeaud M., Les associations microbiennes dans le from age
de Roquefort. Ill. Action des enterocoques et des levures ferment ant le lactose
vis-A-vis des lactobacilles, Lait 51 (1971) 399-415.
8364 A.-M. Davila et al.

[12] Fatichenti F., Bergere J.L., Deiana P., Farris G.A., Antagonistic activity of De-
baryomyces hansenii towards Clostridium tyrobutyricum and Cl. butyricum, J.
Dairy Res. 50 (1983) 449-457.
[13] Fleet G.H., Yeasts in dairy products, J. Appl. Bacteriol. 68 (1990) 199-211.
[14] Hoffman C.S., Winston F., A ten-minute DNA preparation from yeast efficiently
releases autonomous plasmids for transformation of Escherichia coli, Gene 57
(1987) 267-272.
[15] Kreger-van Rij N.J.W., The yeasts: a taxonomic study, 3rd ed., Elsevier Science
Publishers, Amsterdam, 1984.
[16] Kurtzman C.P., Robnett C.J., Identification and phylogeny of ascomycetous
yeasts from analysis of nuclear large sub unit (26S) ribosomal DNA partial se-
quences, Antonie Van Leeuwenhoek 73 (1998) 331-371.
[17] Kurtzman C.P., Robnett C.J., Identification of clinically important ascomycetous
yeasts based on nucleotide divergence in the 5' end of the large-subunit (26S)
ribosomal DNA gene, J. Clin. Microbiol. 35 (1997) 1216-1223.
[18] Lenoir J., Lamberet G., Schmidt J.L., Tourneur C., La maitrise du bioreacteur
from age , Biofutur 41 (1985) 23-50.
[19] Lodder J., The yeasts. A taxonomic study, 2nd ed., North Holland Publishing
Company, Amsterdam-London, 1970.
[20] Maftahi M., Nicaud J.M., Levesque H., Gaillardin C., Sequencing analysis of
a 15.4 kb fragment of yeast chromosome XIV identifies the RPD3, PAS8 and
KREl loci, five new open reading frames, Yeast 11 (1995) 567-572.
[21] Nahabieh F., Schmidt J.L., Contribution a l'etude de la flore levure de quelques
grands types de from ages de chevre, Lait 70 (1990) 325-343.
[22] O'Donnell K, Fusarium and his near relatives, in: Reynolds D. J., Taylor W.
(eds), The fungal holomorph: mitotic, meiotic and pleiomorphic speciation in
fungal systematics, C. A. B. international, Wallingford, 1993, pp. 225-233.
[23] Pearson W.R., Lipman D.J., Improved tools for biological sequence comparison,
Proc. Natl. Acad. Sci. USA 85 (1988) 2444-2448.
[24] Romano A., Casaregola S., Torre P., Gaillardin C., Use of RAPD and mitochon-
drial DNA RFLP for typing of Candida zeylanoides and Debaryomyces hansenii
yeast strains isolated from cheese, System. Appl. Microbiol. 19 (1996) 255-264.
[25] Sambrook J., Fritsch E., Maniatis T., Molecular cloning: a laboratory manual.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1989.
[26] Schmidt J.L., Lenoir J., Contribution a l'etude de la flore levure du from age de
Camembert [Il], Lait 60 (1980) 272-282.
[27] Schmidt J.L., Lenoir J., Contribution a l'etude de la flore levure du from age de
Camembert. Son evolution au cours de la maturation, Lait 58 (1978) 355-370.
[28] Williams J.G., Kubelik A.R., Livak K.J., Rafalski J.A., Tingey S.V., DNA poly-
morph isms amplified by arbitrary primers are useful as genetic markers, Nucleic
Acids Res. 18 (1990) 6531-6535.