Transfer of Genetic Material between Pathogenic and Food-Borne Yeasts

Many pathogenic yeast species are asexual and therefore notinvolved in intra- or interspecies mating. However, high-frequencytransfer of plasmid DNA was observed when pathogenic and food-borneyeasts were grown together. This property could play a crucialrole in the spread of virulence and drug resistance factorsamong yeasts.

INTRODUCTION

Top

Introduction

For millennia, yeasts belonging to the genus Saccharomyces havebeen widely used in preparation of food and beverages. Historically,natural isolates of Saccharomyces species have been consideredto be harmless, nonpathogenic saprophytes. However, in the pastfew decades these yeasts have been found to be causative agentsof human infections (18, 19). How these originally inoffensiveyeasts evolved into pathogenic forms is not understood. Paradoxically,one of the reasons behind the emergence of novel human fungalpathogens is the success of modern medical care, which leadsto the survival of immunocompromised patients (23).

Candida glabrata, formerly classified as Torulopsis glabrata,can be found as a commensal yeast in healthy individuals, butit is also a human opportunistic pathogen (9). C. glabrata isa close relative of Saccharomyces species, and recently it wasadded together with other Saccharomyces yeasts to the jointSaccharomyces clade (13). C. glabrata diverged from the commonancestor of Saccharomyces sensu stricto species after whole-genomeduplication occurred, and the C. glabrata genome subsequentlyunderwent reductive evolution, possibly associated with theemergence of this yeast as a human pathogen (6, 26). The widespreaduse of immunosuppressive therapy, chemo- and radiotherapy forcancer, and broad-spectrum antimycotic therapy has increasedthe frequency of both systemic and mucosal infections causedby C. glabrata (7). Candida species are some of the most commonbloodstream pathogens in the United States, and a substantialshift in the epidemiology of hematogenous candidiasis to non-Candidaalbicans species, such as C. glabrata, has been observed recently.Fluconazole treatment may have played a role in this shift (1).A nationwide survey of candidemia from 1991 to 2000 in Swisstertiary care hospitals showed that C. glabrata was the majornon-C. albicans cause of candidiasis, with an incidence of 15%(14). C. glabrata was the second most frequently occurring fungusand the dominant non-C. albicans species causing candidal vulvovaginitisin pregnant women in China (25). In the Flemish population inBelgium the incidence of infection placed C. glabrata thirdamong the organisms identified in symptomatic vulvovaginal candidiasispatients after fluconazole treatment (5). Vulvovaginal candidiasisis an example of a common mucosal infection that also occursin healthy, immunocompetent women.

Could intra- and interspecies exchange of genetic material beresponsible, at least in part, for the spread of virulence factorsfrom pathogenic species to nonpathogenic species? It has beenproven previously that in experimental populations Saccharomycesspecies can form hybrids and thereby create novel combinationsof genetic material (15). These results suggest that Saccharomycesyeasts have the potential to exchange genetic material in nature,and this view is supported by the fact that several naturalisolates are indeed hybrids between different species (8, 16).It may well be that some Saccharomyces isolates that are humanpathogens are naturally occurring interspecific hybrids (4,17). However, mating per se has not been observed yet for C.glabrata (12), and therefore it is not thought that C. glabratacan be involved in interspecies mating. To evaluate the possibilityof gene transfer between pathogenic and food-borne yeasts species,we studied possible plasmid transfer between C. glabrata andthe Saccharomyces species S. cerevisiae and S. bayanus.

Plasmid transfer from S. cerevisiae to C. glabrata.

Top

Plasmid transfer from s....

To develop a potential acceptor of plasmids, C. glabrata typestrain Y475 was mutagenized with ethyl methanesulfonate as describedpreviously (15). Auxotrophic ura3 mutants were selected on platescontaining 5-fluoroorotic acid (5-FOA) at a concentration of1.00 mg/ml. 5-FOA-resistant mutants were tested for growth ona minimal medium with uracil, on a minimal medium without uracil,and on a synthetic complex medium without uracil, and they werechecked for the ability to revert using standard yeast techniques.Only nonreverting strains were used in experiments. Plasmidsto be used in the plasmid transfer experiment (P158, P159, andP199) (Table 1) were first transformed into C. glabrata ura3mutants (Table 2). Only the C. glabrata mutant strains (Y718and Y719) which were able to grow under selective conditionswhen they were transformed by plasmids were used as plasmidrecipients in the plasmid transfer experiments. S. cerevisiaestrain Y391 was used as a potential donor of genetic materialin the plasmid transfer experiment and was transformed for thispurpose with the P158, P159, and P199 plasmids. Overnight culturesof C. glabrata ura3 mutant strains Y718 and Y719 were mixedwith the same volume (300 µl) of overnight cultures ofS. cerevisiae plasmid donor strains Y745, Y746, and Y747. OnYPD plates 5-µl portions of mixed cultures were spottedand replica plated onto the selective minimal medium the followingday. Neither the S. cerevisiae plasmid donor nor the C. glabrataplasmid acceptor could grow on minimal medium plates withouturacil (Fig. 1), but after a few days small colonies appearedin the "mixed" spots. The frequencies of plasmid transfer betweenplasmid donor strains and plasmid recipient strains were calculatedby determining the fractions of the plasmid recipient strainclones possessing the plasmids based on all of the plasmid recipientstrain colonies plated. In the case of P199 the frequency was1.2 x 10^–4. Several colonies (referred to below as "resultingstrains") (Table 2) were checked for their karyotype patterns,the presence of exchanged plasmid, and the ability to lose theacquired plasmid. Chromosomes were prepared as described byPetersen et al. (20) and were separated by pulsed-field gelelectrophoresis using a five-step program, as follows: step1, 240-s pulse for 6 h; step 2, 160-s pulse for 13 h; step 3,120-s pulse for 10 h; step 4, 90-s pulse for 10 h; and step5, 60-s pulse for 3 h. The angle was 60°, and the potentialwas 150 V (4.5 V/cm). All but one of the resulting strains exhibitedthe same karyotype pattern as the C. glabrata plasmid recipient(Fig. 2A). Thus, C. glabrata cells acquired the plasmid fromthe donor S. cerevisiae strain. The karyotype of Y732 displayedboth S. cerevisiae and C. glabrata sets of chromosome bands,but this strain was later shown to consist of cells of bothparents. The presence of a plasmid in C. glabrata recipientstrains Y718 and Y719 was confirmed by PCR with total DNA isolatedfrom strains resulting from plasmid transfer (designated strainsY727 to Y732) using primers Amp^r12.01 (5'-CAA ATA TGT ATC CGCTCA TGA GAC A-3') and Amp^r12.02 (5'-GTA AAC TTG GTC TGA CAGTTA CC-3'). For amplification of the plasmid ampicillin resistance-encodinggene, the following conditions were used: 94°C of initialdenaturation for 3 min and then 35 cycles of 30 s at 94°C,1 min at 55°C, and 1.5 min at 72°C, followed by 72°Cfor 3 min. Total DNA from Y718 and Y719 served as a negativecontrol. The possibility of plasmid loss from strains resultingfrom plasmid transfer was tested by growing these strains undernonselective conditions. The resulting strains were grown overnightin liquid YPD medium, and then the cells were spread on solidYPD medium. After 3 days colonies were replica plated onto minimalmedium without uracil. Only the colonies which had not lostthe plasmid could grow under selective conditions. The percentageof the P199 plasmid lost from C. glabrata strain Y729 was 84%.In conclusion, C. glabrata indeed acquired the plasmid at arelatively high frequency when it was grown in the presenceof S. cerevisiae cells.

View this table:
[in this window]
[in a new window]
TABLE 1. Shuttle plasmids (E. coli-yeast) used in the experiments and their characteristics

View this table:
[in this window]
[in a new window]
TABLE 2. Strains used in this study and their characteristics

View larger version (62K):
[in this window]
[in a new window]
FIG. 1. C. glabrata auxotrophic ura3 strain Y718 was grown together with S. cerevisiae strains Y745, Y746, and Y747 containing different plasmids carrying the URA3 gene on YPD plates. The colonies were replica plated onto the selective minimal medium the following day, and the images show the growth on the minimal medium after 1 week. Four controls (parental strains Y745, Y746, Y747, and Y718) were also grown on the minimal medium.

View larger version (139K):
[in this window]
[in a new window]
FIG. 2. (A) Karyotypes of C. glabrata recipient strains Y718 and Y719, S. cerevisiae donor strains Y745, Y746, and Y747, and the four resulting strains (lane 1, Y729; lane 2, Y732; lane 3, Y719/Y606 no. 3; lane 4, Y719/Y748 no. 4). All but one strain had the same karyotype pattern as the C. glabrata plasmid recipient. The karyotype of resulting strain Y732 (lane 2) had both S. cerevisiae and C. glabrata chromosome bands. (B) Karyotypes of S. bayanus recipient strain Y244, C. glabrata donor strains Y781, Y782, and Y784, and the four resulting prototrophs (lane 1, Y837; lane 2, Y832; lane 3, Y840; lane 4, Y841). All four resulting strains had the same karyotype pattern as the recipient strain.

Plasmid transfer from C. glabrata to S. bayanus.

Top

Plasmid transfer from c....

C. glabrata strain Y760 was mutagenized with ethyl methanesulfonateand selected with 5-FOA. Double mutant Y765 was transformedwith URA3 gene-containing plasmid P157 (Y784) or P159 (Y781and Y782) and was used as a C. glabrata donor of genetic material(Table 2). A food-borne yeast species belonging to the Saccharomycessensu stricto group, S. bayanus (Y244), was chosen as the potentialplasmid acceptor. Plasmid transfer experiments were carriedout as described above, and prototrophic strains were isolated.Plasmid transfer frequencies were calculated; for plasmid P157the transfer frequency was 2.2 x 10^–4, and for plasmidP159 the transfer frequency was 1.6 x 10^–4. For severalstrains chromosomes were separated by pulsed-field gel electrophoresisusing the program described above, and all of the strains testedexhibited the S. bayanus chromosomal pattern (Fig. 2B). Thus,S. bayanus cells acquired the plasmid from the C. glabrata donorstrain. The presence of the plasmid in S. bayanus plasmid recipientstrain Y244 was also confirmed by PCR when plasmids were firstisolated from the strains and rescued in Escherichia coli strainXL-1 Blue (laboratory designation, P311; Stratagene) (24), andthen the plasmid tetracycline resistance-encoding gene was amplifiedwith gene-specific primers Tet^r11.27 (5'-AGT GCC ACC TGA CGTCTA AGA-3') and Tet^r11.28 (5'-GTT TGC GCA TTC ACA GTT CTC C-3').The following conditions were used for the PCR: 94°C ofinitial denaturation for 4 min and then 35 cycles of 40 s at94°C, 45 s at 54°C, and 2.5 min at 72°C, followedby 72°C for 5 min. In conclusion, S. bayanus also acquiredthe plasmid when it was grown in the presence of C. glabratacells.

Concluding remarks.

Top

Concluding remarks.

Although sexual activity (zygote, G₁ arrest) between C. glabrataand S. cerevisiae cells has not been observed previously, weobserved a relatively high frequency of transfer of geneticmaterial between these yeast species. The mechanism(s) for thistransfer is not clear. Previously, DNA transfer between bacteria(E. coli) and yeasts (S. cerevisiae and S. kluyveri) was reportedas transkingdom conjugation (10, 11). In our experiments recipientspecies could have picked up plasmid molecules released by thedonor strains through cell lysis, or the plasmid transfer couldhave resulted from cytoduction. In addition, despite the "asexual"nature of C. glabrata, a potential cryptic sexual life cycle(27) could promote rare interspecies mating. The transfer observedcould allow the spread of virulence factors and resistance tomedical drugs even between distantly related yeast species andcould probably help in the "transformation" of harmless saprophytesinto potential causative agents of human infections.

ACKNOWLEDGMENTS

We thank Jørgen Stenderup for his interest in this work.

FOOTNOTES

* Corresponding author. Mailing address: Department of Cell and Organism Biology, Lund University, Sölvegatan 35, SE-223 62 Lund, Sweden. Phone: 46 46 222 83 73. Fax: 46 46 222 41 13. E-mail: jure.piskur{at}cob.lu.se.

Present address: Department of Biochemistry, Faculty of NaturalSciences, Comenius University, Mlynská dolina CH-1, 84215 Bratislava, Slovakia.

Present address: Department of Chromosome Biology, Faculty ofLife Sciences, University of Vienna, Dr. Bohr Gasse 1, A-1030Vienna, Austria.

REFERENCES

Top