OUP user menu

Saccharomyces cerevisiae and Saccharomyces paradoxus coexist in a natural woodland site in North America and display different levels of reproductive isolation from European conspecifics

Paul D. Sniegowski, Peter G. Dombrowski, Ethan Fingerman
DOI: http://dx.doi.org/10.1111/j.1567-1364.2002.tb00048.x 299-306 First published online: 1 January 2002


We report the isolation of multiple strains of Saccharomyces cerevisiae and Saccharomyces paradoxus from a natural woodland site in southeastern Pennsylvania, USA, using enrichment culturing in a medium containing 7.6% (v/v) ethanol. The method was applied to bark and flux material collected from broad-leaved trees (mostly Quercus spp.) and to associated soils. Many candidate wild strains of Saccharomyces were isolated using this method, most of them from soils associated with oaks. Matings to genetically marked tester strains of S. cerevisiae and S. paradoxus identified roughly equal numbers of these two species within this collection. The S. paradoxus isolates showed significant partial reproductive isolation from a conspecific European strain, whereas the S. cerevisiae isolates did not. Variability in both chromosome size and Ty1 element hybridization profiles was observed within both populations at this site. We discuss the relevance of our data to current debates concerning whether S. cerevisiae is a wild species or a domesticated species.

  • Yeast
  • Reproductive isolation
  • Wild Saccharomyces
  • Saccharomyces paradoxus
  • Saccharomyces cerevisiae

1 Introduction

Studies of hybrid ascospore viability suggest that the Saccharomyces sensu stricto complex contains at least six species delimited by postzygotic isolation: Saccharomyces cerevisiae, Saccharomyces paradoxus, Saccharomyces bayanus, Saccharomyces cariocanus, Saccharomyces kudriavzevii and Saccharomyces mikatae[13]. Relatively little is known about the distribution and population structure of these species. To date, the most extensively studied species in the wild is S. paradoxus, which appears to be present on most continents and exhibits genetic and reproductive differentiation over long distances [2,47]. S. paradoxus strains collected from within broad geographical regions show limited allozyme variability [5], but they are clearly variable at the DNA sequence level as judged by DNA fingerprinting methods [8] and hybridization profiles of Ty1 elements to chromosomes [6,8,9]. S. cerevisiae strains isolated from vineyard grapes show considerable variability in metabolic properties, karyotype, and DNA sequence as judged by the results of RAPD analysis [1013], but no such data are available for S. cerevisiae isolated from uncultivated areas. Indeed, S. cerevisiae has seldom been isolated away from the context of human applications [2,6,14].

The most common uncultivated habitat from which Saccharomyces sensu stricto have been isolated is fluxes of oaks and other broad-leaved trees, although Saccharomyces are not always present at such sites [15]. Previously published methods of isolation from fluxes have involved enrichment in malt extract-based medium, phenotypic screening based on colony, cell, and ascus morphologies, and positive species identification by the analysis of matings to genetically marked tester strains (e.g. [6]). The relative rarity of flux sites and the variable success in obtaining positive enrichments led us to develop a more efficient sampling and enrichment procedure, which we describe here. Although our initial goal was to establish a collection of S. paradoxus for studies of population structure, in the course of this work we obtained comparable numbers of S. cerevisiae isolates from the same habitat. Interestingly, our S. cerevisiae isolates showed no reproductive isolation when crossed with a conspecific tester from Europe, whereas our S. paradoxus isolates showed the expected partial reproductive isolation from a European conspecific. We discuss the implications of our findings for population studies in Saccharomyces and for the controversy concerning whether S. cerevisiae is a wild species or a domesticated species.

2 Materials and methods

2.1 Sample collection and enrichment culturing

All collections were made in July 1999 at the John J. Tyler Arboretum in Lima, PA, USA. The arboretum grounds include 182 ha of mature second-growth forest dominated by stands of native tulip poplar (Liriodendron tulipifera), American beech (Fagus grandifolia), maples (Acer spp.), and oaks (Quercus spp.); this forest is contiguous with 1052 ha of similar natural habitat in Ridley Creek State Park. Exudate material, bark, and soil samples were obtained from around the bases of trees using a sterile scalpel or spatula and deposited in sterile 8-ml glass vials in the field. These collection vials were stored for up to 3 days at 4°C before enrichment culturing. Upon return to the laboratory, the vials were filled with a sterile liquid medium consisting of 3 g yeast extract, 3 g malt extract, 5 g peptone, 10 g sucrose, 76 ml EtOH, 1 mg chloramphenicol, and 1 ml of 1-M HCl per liter. Vials were then capped tightly, incubated for approximately 10 days at 30°C without shaking, and inspected for signs of fermentative CO2 production. In many cases, vials effervesced vigorously upon being opened after a few days; in some cases, a visible white sediment on the bottom of the vial indicated proliferation of yeasts or other cells.

A single aliquot of 10 μl from each of these liquid enrichment cultures was next streaked on solid medium containing 15 g of agar, 20 g of methyl-α-d-glucopyranoside, 6.7 g of yeast nitrogen base with amino acids and ammonium sulfate (Difco, USA) and 4 ml of 1-M HCl per liter. Plates were incubated for several days at 30°C and examined for growth of glossy, off-white yeast colonies. Such colonies were picked and restreaked to the same medium to obtain isolated clones, after which a single clone derived from each was picked into liquid yeast extract–peptone–dextrose (YPD) medium [16], grown to stationary phase, and stored in 15% glycerol at −80°C pending species identification.

Pilot experiments showed that the use of sucrose and methyl-α-d-glucopyranoside in the liquid and solid media described above gave the most dependable isolation of Saccharomyces from natural substrates; however, it is important to note that these media do not permit isolation of strains unable to utilize these carbon sources.

2.2 Sporulation and mutagenesis

Putative Saccharomyces strains obtained from the enrichment procedure described above were regrown and sporulated in standard liquid complete presporulation medium and complete sporulation medium [16]. Most isolates sporulated readily, but a few failed to sporulate and were not subsequently analyzed by test matings. Asci for dissection were obtained by digesting sporulated cultures for 10 min at 30°C in 0.5 mg of zymolyase 100T (US Biological, USA) per ml of 1-M sorbitol. Spores for mutagenesis and mating were obtained by digesting asci for 30 min at 30°C, vortexing on high speed for 2 min, centrifuging, and resuspending in 0.015% Igepal (Sigma, USA) to inhibit clumping.

To facilitate mass mating with tester strains, adenine auxotrophs were obtained from the wild isolates. Appropriate concentrations of spores from each wild isolate to be tested were spread on YPD agar plates and exposed for 5 s to ultraviolet light (254 nm wavelength) on a minitransilluminator (Foto/Phoresis Model 1-1430, FotoDyne, USA). Plates were incubated for several days at 30°C and examined daily for the presence of red colonies indicative of adenine auxotrophy. Putative adenine auxotrophs were confirmed by picking to synthetic dextrose (SD) minimal agar [16] and SD agar plus 2 g l−1 of adenine hemisulfate. Confirmed auxotrophs were grown, resporulated and the asci digested to completion to obtain spore suspensions for test matings.

2.3 Genetic identification of Saccharomyces sensu stricto

Adenine auxotrophs derived from the wild isolates were mass-mated to genetically marked tester strains of S. cerevisiae and S. paradoxus on SD agar. The S. cerevisiae tester, DH46, was a Δleu2 ho MATa strain derived from the laboratory strain Y55 and provided to us by Dr. D. Grieg. Strain Y55 was originally isolated by Ø. Winge from wine grapes in France during the 1930s (J.H. McCusker, personal communication). The S. paradoxus tester, N17-13, was an ho MATa lys2 derivative of wild strain N17, which was originally collected from Tartastan (former USSR) by G.I. Naumov [1]. Strain N17-13 was provided to us by Dr. E.J. Louis. Prototrophic maters between adenine auxotrophic derivatives of the wild isolates and these tester strains were sporulated, and approximately 16 asci of each were dissected on YPD agar to determine ascospore viabilities. Viabilities were scored as the proportion of ascospores yielding colonies visible to the unaided eye after 3 days of incubation at 30°C. Where viable ascospores were obtained, segregation and assortment of both adenine auxotrophy and the tester strain auxotrophy confirmed that the strain analyzed was a mater and not a selfed revertant of the wild isolate.

2.4 Ascospore viabilities and phenotypic tests for heterozygosity in the identified strains

Isolates that were identified by genetic analysis as either S. cerevisiae or S. paradoxus were sporulated and scored for ascospore viability as described above. In addition, colonies derived from dissected complete asci of these identified strains were examined phenotypically for heterozygosity in utilization/fermentation of four sugars, prototrophy, and homothally/sporulation. Colonies were replica-plated to indicator agars containing the pH-sensitive dye bromthymol blue and the sugars galactose, sucrose, maltose and melibiose [16], to SD agar, and to minimal sporulation agar. Colonies replicated to the indicator media and SD medium were inspected visually after 2 days at 30°C; colonies replicated to sporulation medium were inspected microscopically after several days for the presence of asci indicative of homothallic selfing and subsequent sporulation.

2.5 Karyotype and Ty1 element hybridization analyses

Karyotypes of monosporic clones derived from the identified strains were analyzed using contour-clamped homogeneous field (CHEF) gel electrophoresis. Two standard laboratory strains of S. cerevisiae, S288C [17] and YPH80 (obtained commercially from New England BioLabs, USA), were included in the analysis for comparison, as well as the type strain of S. paradoxus, CBS 432 [1]. Preparation of chromosomal DNAs followed a published protocol [18]. Chromosomes were separated on a Bio-Rad CHEF Mapper using the manufacturer's supplied algorithm for separating DNA fragments within the 220–2200-kb size range. Depurinated and denatured chromosomal DNAs were transferred to positively charged nylon membranes using a vacuum blotter (Bio-Rad, USA, Model 785) and cross-linked with a UV transilluminator. Membranes were probed with a 974-bp internal fragment from the TyB open reading frame of the Ty1 transposable element. The probe was amplified from the laboratory S. cerevisiae strain S288C with the primer pair 5′-AAAGCTGTGAGTCCAACCGATTC-3′ and 5′-TTGATTGACTTCTTTGTGGTATGCC-3′ and was PCR-labeled with digoxigenin-11-dUTP. Hybridization and detection were carried out using the DIG Nucleic Acid Labeling and Detection Kit (Roche Molecular Biochemicals, USA) according to the manufacturer's protocol; blots were allowed to develop in color substrate solution for 16 h.

3 Results

3.1 Natural distribution of S. cerevisiae and S. paradoxus

We collected a total of 84 samples for enrichment from 40 different individual trees and their associated soils: 17 white oaks (Quercus alba), 11 red oaks (Quercus rubra), five black oaks (Quercus velutina), two chestnut oaks (Quercus prinus), two American beeches (F. grandifolia), one tulip poplar (L. tulipifera), one red maple (Acer rubrum), and one undetermined oak species. Only the samples associated with oaks yielded putative Saccharomyces yeasts as judged by colony and vegetative cell morphology. Of 79 such oak-associated samples processed through enrichment culture, 18 yielded isolates that were highly fertile with either the S. cerevisiae or the S. paradoxus tester strain. Where fertility in hybrids with both tester species was observed it was always high in one hybrid and very low in the other; in no case was high fertility observed in hybrids with both species.

Table 1 lists all of the strains identified as S. cerevisiae or S. paradoxus and gives the substrate from which each was isolated. As shown in the table, these two species were identified in roughly equal numbers and from similar substrates. Table 2 gives the number of dissected asci and the proportion of viable spores in hybrids with both testers for each identified strain. Auxotrophic mutations in the tester strains (leu2 and lys2) and in the wild isolates (ade1 or ade2) segregated and assorted normally in the viable spores (data not shown). The rare viable spores in interspecific hybrids invariably gave tiny and slowly growing colonies.

View this table:
Table 1

Identified isolates of S. paradoxus and S. cerevisiae and their source habitats in a Pennsylvanian woodland

S. cerevisiaeYPS 128Soil beneath Q. alba
YPS 129Flux from Q. alba
YPS 133Soil beneath Q. alba
YPS 134Soil beneath Q. velutina
YPS 139Soil beneath unidentified Quercus spp.
YPS 141Soil beneath Q. velutina
YPS 142Bark of Q. rubra
YPS 143Soil beneath Q. rubra
YPS 154Bark of Q. velutina
YPS 163Soil beneath Q. rubra
S. paradoxusYPS 125Flux of Q. rubra
YPS 138Soil beneath Q. velutina
YPS 145Soil beneath Q. alba
YPS 150Bark of Q. velutina
YPS 151Soil beneath Q. velutina
YPS 152Soil beneath Q. rubra
YPS 155Bark of Q. rubra
YPS 158Soil beneath Q. alba

Strains YPS 142 and YPS 143 were isolated from different samples associated with the same individual of Q. rubra; strains YPS 150 and YPS 151 were isolated from different samples associated with the same individual of Q. velutina. All other strains were isolated from different trees. The strain numbers are those in the yeast culture collection of the first author in the Department of Biology, University of Pennsylvania.

    View this table:
    Table 2

    Spore viabilities in hybrids produced by crossing auxotrophic derivatives of wild Saccharomyces strains isolated from Tyler Arboretum and auxotrophic tester strains of S. paradoxus (N17) and S. cerevisiae (DH46)

    Wild isolateNo. of tetrads dissected in hybrid with S. paradoxus N17-13Ascospore viability (%) in hybrid with S. paradoxus N17-13No. of tetrads dissected in hybrid with S. cerevisiae DH46Ascospore viability (%) in hybrid with S. cerevisiae DH46
    YPS 1281401593
    YPS 1292502485
    YPS 1332521595
    YPS 1342512590
    YPS 1392502582
    YPS 1411501692
    YPS 1423251694
    YPS 1431501692
    YPS 1543013089
    YPS 1631501577
    YPS 1251542120
    YPS 1381658258
    YPS 1451953256
    YPS 1503151169
    YPS 1511644160
    YPS 1521667160
    YPS 1551656160
    YPS 1581548157

    Spore viabilities were recorded after 3 days of incubation at 30°C. The viable spores recorded in crosses 128-163×N17-13 and 125-158×DH46 yielded microcolonies.

      3.2 Geographic reproductive isolation

      The wild isolates of S. cerevisiae and S. paradoxus differed in their degree of reproductive isolation from their conspecific European tester strains: for S. cerevisiae×Y55 fertile hybrids, the average spore viability ±S.E.M. was 89±1.8%, whereas for S. paradoxus×N17-13 fertile hybrids, it was 52±8.1%. This difference is highly significant (two-tailed t-test of arcsin square root transformed viability proportions: t=10.58; P<0.0001; df=16). The reduction of hybrid spore viability observed in S. paradoxus is not a property of the tester strain, as the S. paradoxus tester parent strain N17 itself has a previously observed spore viability of 100%[1]. To address whether hybrid spore viabilities in both species were a property of the wild isolates or a property of the hybrids, we obtained data on spore viabilities in the original diploid wild isolates as shown in Table 3. Average spore viability in the S. cerevisiae isolates was 82±5.5%, which is not significantly different from that observed in hybrids between these isolates and the S. cerevisiae tester, Y55 (paired t-test of arcsin square root transformed viability proportions: t=0.78; two-tailed P=0.456; df=9). Average spore viability in the S. paradoxus isolates was 81±4.1%, which is significantly higher than that observed in hybrids with the S. paradoxus tester, N17-13 (paired t-test of arcsin square root transformed viability proportions: t=4.98; two-tailedP=0.0016; df=7). There was no significant difference in spore viability between the S. paradoxus and S. cerevisiae isolates themselves.

      View this table:
      Table 3
      SpeciesWild isolateNo. of tetrads dissectedAscospore viability (%)
      S. cerevisiaeYPS 1281598
      YPS 1291677
      YPS 1331564
      YPS 1341677
      YPS 1391691
      YPS 1411689
      YPS 1421684
      YPS 1431698
      YPS 1541644
      YPS 1631698
      S. paradoxusYPS 1251695
      YPS 1381692
      YPS 1451669
      YPS 1501681
      YPS 1511667
      YPS 1521672
      YPS 1551695
      YPS 1581677

      Spore viabilities were recorded after 3 days of incubation at 30°C.

        3.3 Spore phenotypes in the wild isolates

        All of the wild isolates of both species were found to be homozygous for all phenotypes scored: all were prototrophic, homothallic and sporulation-proficient, able to ferment or utilize sucrose and galactose (as judged by colony growth and a color change to yellow in the indicator agar beneath the colonies), and unable to ferment or utilize maltose and melibiose. Inability to ferment or utilize maltose and melibiose was confirmed on SD plates containing these sugars as the sole carbon source.

        3.4 Molecular karyotyping and Ty1 element hybridization

        Fig. 1 illustrates electrophoretic karyotypes of monosporic isolates from a subset of the 18 wild strains along with those of standard S. cerevisiae strains YPH80 and S288C and the type strain of S. paradoxus, CBS 432. All isolates displayed chromosome numbers and sizes typical of those previously observed in the Saccharomyces sensu stricto group. As expected [9], no consistent differences in karyotype between S. cerevisiae and S. paradoxus were observed. In contrast to results previously reported for wild Saccharomyces, however [9,19], we observed substantial size variation in chromosome XII in both species. Fig. 2 shows profiles of hybridization of the Ty1 probe to chromosomes from monosporic clones of a subset of the wild isolates. Hybridization signal was strong in both the wild and laboratory standard S. cerevisiae strains, and visible signal was obtained from most chromosomes. The S. paradoxus type strain CBS 432 also exhibited strong hybridization signal on many chromosomes, but the wild S. paradoxus strains isolated in this study gave a much weaker signal that was detectable on only a small number of chromosomes. Within the wild isolates of each species, multiple strains appeared to share chromosomal Ty1 hybridization profiles; however, at least two distinct hybridization profiles were distinguishable for each species.

        Figure 1

        Two gels showing electrophoretic karyotypes of monosporic derivatives from a subset of the S. cerevisiae and S. paradoxus strains isolated from the Tyler Arboretum, along with the karyotypes of the laboratory S. cerevisiae strains S288C and YPH80 and the S. paradoxus type strain CBS 432. Lanes corresponding to the wild strains are indicated along the bottom of the figure; strain numbers in parentheses refer to the original wild isolates listed in Table 1.

        Figure 2

        Hybridization of a Ty1-specific probe to chromosomes of wild and standard laboratory strains of S. cerevisiae and S. paradoxus. The figure is a composite of two separate blots made under identical conditions and shows a subset of the 18 wild strains isolated from the Tyler Arboretum. Lanes corresponding to wild strains are indicated at the bottom of the figure. Results shown are for monosporic derivatives of the wild strains; strain numbers in parentheses refer to the corresponding original wild isolates listed in Table 1. (The cleared area in the lane for YPS 600 is an artifact introduced during the detection procedure.)

        4 Discussion

        We have shown that the Saccharomyces sensu stricto species S. paradoxus and S. cerevisiae can be obtained from uncultivated habitats by enrichment culturing in a medium containing 7.6% (v/v) ethanol. Ethanol enrichment has been used to isolate S. cerevisiae from vineyard grapes [20,21], but to our knowledge this is the first report of its successful application in uncultivated habitat. The presence of S. cerevisiae in our collection from an uncultivated site is not a local anomaly, as previous studies have reported occasionally finding S. cerevisiae in uncultivated habitats in Central Siberia, Japan, Europe [14], North America [6,22], and South America [23,24]. Nonetheless, this is the first time that substantial numbers of isolates of S. cerevisiae and S. paradoxus have been isolated contemporaneously from the same site.

        Somewhat surprisingly, most of our isolates were obtained from oak-associated soils rather than from oak fluxes as usually reported. The numbers of oak soil and oak flux/bark samples that we collected were approximately equal, yet 12 of our 18 identified isolates originated from soil samples, four from oak bark, and only two from fluxes. It is obviously not necessary to sample oak fluxes in order to have a good chance of isolating Saccharomyces from uncultivated habitat. Indeed, both S. cerevisiae and S. paradoxus have previously been isolated from soils, though never in substantial numbers as reported here [19,2530].

        There is an ongoing debate as to whether S. cerevisiae is exclusively a domesticated organism [12,20,21,3133]. Many isolates identified as S. cerevisiae have been obtained from damaged grape berries and fermented grape must at wineries practicing natural (no added yeast) fermentation [10,11,13,20,21], and human pathogenic strains have been isolated and studied [34,35], but the species has seldom been obtained from habitats less closely associated with humans. At present it remains impossible to rule out any of three explanations for the presence of S. cerevisiae in natural habitats: (1) S. cerevisiae is a domesticated organism that occasionally establishes synanthropic natural populations; (2) a wild population of S. cerevisiae has recently expanded its range from a single region of origin [36], perhaps in association with humans; (3) there are diverse, globally distributed wild populations of S. cerevisiae (like those apparent in S. paradoxus) that predate domestication and have existed independently all along. Some combination of these scenarios is also possible; for example, there could be ongoing gene flow between domesticated and wild populations of S. cerevisiae.

        Previous studies have shown that S. paradoxus populations from different geographic regions exhibit partial reproductive isolation [2,47]. As expected, hybrids between our isolates of S. paradoxus and the S. paradoxus tester strain N17-13, which is originally from Tartastan, showed significantly reduced fertility compared with the parent strains. However, hybrids between our S. cerevisiae isolates and the Y55-derived S. cerevisiae tester, which is originally from France, were as fertile as their parent strains. Naumov et al. [22] previously reported similar results for ten S. paradoxus and three S. cerevisiae strains isolated from multiple North American sites: hybrids between their S. paradoxus strains and an S. paradoxus tester from Denmark showed an average spore viability of 31.6±3.9%, whereas hybrids between their S. cerevisiae strains and S. cerevisiae testers from Russia showed an average spore viability of 87.7±1.5%. Perhaps S. cerevisiae populations from diverse regions share a far more recent common ancestor than comparable S. paradoxus populations and hence have had much less time in which to evolve reproductive divergence.

        Published genetic analyses of population structure in Saccharomyces are limited to a single allozyme study documenting significant genetic differentiation between European and Far East Asian populations of S. paradoxus[5]. That study provided little evidence of intrapopulation variation, but a recent study analyzing multilocus sequence data from a British population of S. paradoxus has documented genetic variation and a low but detectable rate of outcrossing (L. Johnson, personal communication). Our karyotype and Ty1 element hybridization data indicate that both S. paradoxus and S. cerevisiae at the Tyler site are also genetically variable. First, in contrast with some previous studies of wild strains collected from numerous regions [6,9], we observed substantial variation in the size of chromosome XII among our wild isolates (Fig. 1). Size variation in chromosome XII is commonly observed between S. cerevisiae wine strains and has been attributed to length variation in the rDNA gene cluster present on this chromosome [37,38]. Second, although several of our wild isolates within both species shared Ty1 chromosomal hybridization profiles, at least two different hybridization profiles were present in each species. (The possibility that strains sharing hybridization profiles have different Ty1 insertion sites at the nucleotide level also cannot be ruled out.) Taken together, our data on chromosome XII size variation and Ty1 hybridization indicate at least five unique genotypes within the isolates of each species from the Tyler site.

        S. cerevisiae and S. paradoxus at the Tyler site differed markedly in Ty1 hybridization profile and signal intensity (Fig. 2). The signal intensity difference may reflect Ty1 sequence divergence or a copy number difference between the two species; at present we cannot say which. Whatever the cause of the differences in Ty1 hybridization, their presence supports the conclusion that these two populations are reproductively isolated from one another.

        All S. cerevisiae and S. paradoxus strains isolated from the Tyler site yielded identical spore phenotypes: all spores from multiple complete asci in each isolate were prototrophic, homothallic and sporulation proficient, able to ferment or utilize sucrose and galactose, and unable to ferment or utilize maltose and melibiose. Analyses of a wider range of such phenotypic characters in vineyard strains indicated high levels of heterozygosity [1013]. Because of differences in the way data were reported, only one statistical comparison between those previous studies and ours is possible: Mortimer et al. ([11], Table 1) found that eight of 28 S. cerevisiae strains isolated from the Emilia Romagna region of Italy were heterozygous for one or more of the characters homothally, sucrose fermentation, maltose fermentation, and galactose fermentation. Although this suggests a higher overall proportion of heterozygous strains in these vineyard isolates, it is not significantly different from our finding of zero out of 10 heterozygous S. cerevisiae strains at the Tyler site for these same four phenotypic characters (two-tailed Fisher exact test, P=0.082). (Mortimer et al. assayed their strains for fermentation capacity using bromthymol blue indicator agars rather than the standard Durham tube method [39]. Nonetheless, because our strains were assayed using the same indicator agar method, the comparison is a legitimate one.) It is noteworthy that most vineyard S. cerevisiae strains are able to ferment or utilize maltose, whereas all of our strains from uncultivated sites were unable to ferment or utilize maltose. Variation in maltose fermentation capacity has been observed previously in wild and cultivated isolates of S. cerevisiae and S. paradoxus[40].

        S. cerevisiae and S. paradoxus show almost indistinguishable profiles for fermentation and assimilation reactions and other characteristics [3,32] and their life cycles are identical as far as is known. Our finding that these species are coexisting in similar habitat thus raises an ecological question: do S. cerevisiae and S. paradoxus coexist stably in the wild, and, if so, how?


        P.D.S. thanks G.I. Naumov and E.S. Naumova for introducing him to the genetic identification of species in Saccharomyces. We are grateful to E.J. Louis and D. Greig for the tester strains, M.-A. Lachance for advice on the enrichment protocol, J. Sweeney for technical assistance, M. Bucan for use of the CHEF Mapper, T. Pugh for advice on CHEF interpretation, and L. Johnson for sharing data before publication. We also thank Richard Colbert, Director of the Tyler Arboretum, for permission to collect on the arboretum grounds. We thank H. Kuehne, M.-A. Lachance, and G.I. Naumov for comments on the manuscript. This research was supported by startup funds to P.D.S. from the University of Pennsylvania and by a grant from the University of Pennsylvania Research Foundation to P.D.S.


        1. [1].
        2. [2].
        3. [3].
        4. [4].
        5. [5].
        6. [6].
        7. [7].
        8. [8].
        9. [9].
        10. [10].
        11. [11].
        12. [12].
        13. [13].
        14. [14].
        15. [15].
        16. [16].
        17. [17].
        18. [18].
        19. [19].
        20. [20].
        21. [21].
        22. [22].
        23. [23].
        24. [24].
        25. [25].
        26. [26].
        27. [27].
        28. [28].
        29. [29].
        30. [30].
        31. [31].
        32. [32].
        33. [33].
        34. [34].
        35. [35].
        36. [36].
        37. [37].
        38. [38].
        39. [39].
        40. [40].
        View Abstract