Hybrid Speciation in Experimental Populations of Yeast

See allHide authors and affiliations

Science  29 Nov 2002:
Vol. 298, Issue 5599, pp. 1773-1775
DOI: 10.1126/science.1076374


Most models of speciation require gradual change and geographic or ecological isolation for new species to arise. Homoploid hybrid speciation occurred readily between Saccharomyces cerevisiae and Saccharomyces paradoxus. Hybrids had high self-fertility (about 82%), low fertility when backcrossed to either parental species (about 7.5%), and vigorous growth under different thermal environments that favored one or the other of the parental species. Extensive karyotypic changes (tetrasomy) were observed in the hybrids, although genic incompatibilities accounted for 50% of the variation in self-fertility.

Speciation is thought to arise by gradual evolution of genic incompatibilities (1), ecological specialization (2, 3), or chromosomal differences (4) that prevent mating or cause inviable or infertile hybrid offspring (5). Rapid species formation can potentially occur by hybridization; however, the degree of reproductive isolation between potential new hybrid species and the two parental species is a major limiting factor. Hybrids must be self-fertile and sufficiently reproductively isolated to maintain a distinct lineage, but reproductive barriers between parental species must not preclude the initial hybridization. In postzygotically isolated species, where hybrids are typically inviable or sterile (6), these conflicting requirements can be achieved by a doubling of chromosome complement in the new species to produce an allotetraploid (7). Potentially, these requirements can also be met by maintaining chromosome number (homoploid hybrid speciation) (8, 9), but this mechanism is very uncommon in plants and unknown in animals (10).

Saccharomyces yeast species are postzygotically isolated, because hybrids form readily but are sterile, producing only ∼1% viable gametes (spores) (11–13). However, populations of yeast can be very large (>108), and viable gametes can be easily obtained. Moreover, the ability ofSaccharomyces gametes to divide and switch mating type allows for autofertilization (gametophytic selfing) and, potentially, for instantaneous homoploid hybrid speciation. We investigated this potential with Saccharomyces cerevisiae andSaccharomyces paradoxus and measured the effects of intrinsic incompatibilities (hybrid sterility and infertility) and extrinsic incompatibilities (relative fitness of hybrids under different environmental conditions) (14).

First, we crossed S. cerevisiae and S. paradoxus and isolated 80 independent viable haploid gametes from their F1 hybrid offspring. After allowing for spontaneous hybrid diploid formation by autofertilization (15), we found that 81.25% were capable of sporulation and that fertility (spore viability) was high (median = 90%; mean = 84.40%, with 95% confidence interval of 73.75 to 92.67%) (Fig. 1A) (15). Fertility was slightly reduced from that of the parental species (S. cerevisiae, 99.93%, 99.04 to 99.79%; S. paradoxus, 99.21%, 97.80 to 99.92%) (11), with statistically significant variation among F2 hybrids (F 61,260 = 15.72, P < 0.0001). We tested for reproductive isolation of the fertile F2 hybrids from the parental species (Fig. 1B). The backcross hybrids have fertility that is significantly higher (7.54%, 5.38 to 10.02%) than that of F1 hybrids (0.03%, 0.00 to 0.18%) (11), but they have fertility that is much lower than that of the F2 hybrids (F 1,895 = 817.02, P ≪ 0.0001). Although rare, hybrid F2 diploids are both fertile and isolated from their parental species.

Figure 1

Reproductive isolation of sporulation-proficient F2 hybrids. (A) Hybrids have high fertility when crossed with themselves. (B) Hybrids have low fertility when crossed with either parental species (squares, S. cerevisiae; triangles, S. paradoxus).

Crossing F2 hybrids and assessing fertility of their hybrid offspring demonstrated the existence of multiple different highly fertile F2 hybrids (15). Ten independent F2 genotypes, each having 100% fertility, were randomly paired and used to generate F3 hybrids. All pairs yielded some viable gametes, but the average fertility of F3hybrids (10.64%, 0.93 to 28.97%) was much lower than that of their immediate parents; also, there was genetic variation in fertility among the F3 hybrids caused by interaction between the F2 parental genomes (F 4,94 = 5.65, P < 0.001). Nevertheless, autofertilized F4 hybrid diploids derived from the viable gametes had particularly high fertility (97.33%, 92.10 to 99.81%), greater than that observed in the F2 autofertilized hybrids. As in the F2 hybrids, there was significant variation in fertility among the F4 hybrids (F 13,72 = 5.02, P < 0.0001), but variation in F4fertility was not associated with particular F3 parental genotypes (F 4,13 = 0.56, P> 0.5). Thus, highly fertile hybrid diploids were readily obtained regardless of existing genotype interactions.

There were several possible causes of the reproductive isolation observed between F2 hybrids and their parents, as well as between different F2 hybrid strains. One possibility is that crossing an F2 hybrid with either parental genotype or another F2 hybrid generated incompatible gene combinations in the resulting F3 hybrids, rendering them sterile. A second possibility is chromosomal incompatibility due to the inability of chromosomes from one parental species to pair and cross over with their diverged homeologs from the other parental species (13). A third possible cause is aneuploidy. Meiotic segregation in F1 hybrids is known to be ineffective, and the resulting gametes have a high frequency of disomic chromosomes (13). Such gametes become tetrasomic in F2hybrids and trisomic in backcrosses or in crosses to F2hybrids with different disomic chromosomes (16).

To distinguish between these possible causes for reproductive isolation of the F2 hybrids, we performed two karyotype measurements on a randomly selected set of 38 hybrids. The first measurement detected the presence (or absence) of S. cerevisiae chromosomes through polymerase chain reaction (PCR) assays with S. cerevisiae–specific primers that amplify marker sequences from all S. cerevisiae chromosome ends, if present. This showed that 73% of the chromosomes had not recombined, and crossing over between S. cerevisiae andS. paradoxus chromosomes in the F1 hybrid had been drastically reduced to an average of 8.1% of the S. cerevisiae rate. If reproductive isolation of the F2hybrids was due to the inability of diverged chromosomes to cross over, then strains with more S. cerevisiae chromosomes should be more isolated from S. paradoxus and less isolated fromS. cerevisiae, and vice versa for those with more S. paradoxus chromosomes. This would also be expected if isolation was due to genic incompatibilities, although only if they were frequent enough to be spread across most chromosomes. No correlation betweenS. cerevisiae chromosome number and isolation from eitherS. cerevisiae (r = 0.143, P> 0.2, df = 31) or S. paradoxus (r = –0.156, P > 0.2, df = 31) was detected, nor was there the expected negative correlation between parental F2backcross fertilities (r = –0.093, P> 0.5, df = 63). Nevertheless, the power of these statistical tests is limited by the absence of fertile backcrosses and their relatively small sample sizes.

Instead, extensive tetrasomy was detected. S. cerevisiae ends were present in 797 of the 1216 individual chromosome end assays, which is 31% more frequent than that expected by chance, suggesting the presence of both S. paradoxus andS. cerevisiae homeologs of the same chromosomes in each strain. Tetrasomy was confirmed by the second karyotype assay with pulsed-field gel electrophoresis (15). Three S. paradoxus chromosomes migrate to different positions from theirS. cerevisiae homeologs, allowing a direct and precise measure of tetrasomy. Chromosomes I, II, and VIII were tetrasomic in 26, 20, and 31% of the F2 hybrids, respectively, consistent with the 31% overall tetrasomy estimate from the PCR assays. The abundance of tetrasomy suggests that particular chromosome combinations may have been required for viable hybrid gametes. It also provides a mechanism by which aneuploidy may play a major role in reproductive isolation of hybrids from parental genotypes, although there is no direct proof that aneuploidy is the cause.

However, neither aneuploidy nor chromosomal incompatibility adequately explains the lower self-fertility in F2 hybrids than in the pure parental species. Unlike triploids (16), both nonhybrid and F1 hybrid tetraploids are fertile (11), so we do not expect tetrasomes to have inherent deficiencies in meiotic segregation, although it is possible that unbalanced excess gene dosage from the extra chromosomes might affect fertility. We also can eliminate mitochondrial incompatibility (17) because of the high fertility of hybrid tetraploids. Chromosomal incompatibilities can be excluded because F2 hybrids are fully homozygous, having originated from single autofertilized gametes; thus, all chromosomes can match and pair effectively. The likely explanation is that genic incompatibilities between interacting S. cerevisiae andS. paradoxus genes have a detrimental effect on fertility. The apparent absence of these incompatibilities in full tetraploid F1 hybrids (11), where the complete genomes of both species are present, indicates that they are recessive. We can estimate that they contribute 50% of the variation in self-fertility among the fertile F2 hybrids (F 3,52 = 6.46, P < 0.0001). The nonrandom distribution of chromosomes in the F2 hybrids suggests that similar incompatibilities also cause F1 gamete inviability for certain combinations of chromosomes; however, the current data set is too small to determine specific interactions.

Our results suggest that homoploid hybrid speciation can occur readily and that any intrinsic incompatibilities inSaccharomyces can be overcome relatively easily, but extrinsic barriers, such as fitness under differing environmental conditions, could limit speciation. To address this, we compared the set of 38 F2 hybrids and a common S. cerevisiaecompetitor in nutrient-rich medium at 30° and 10°C, temperatures that favor S. cerevisiae and S. paradoxus, respectively (Fig. 2). There was abundant genetic variation detected for fitness (15) among the hybrids (F 37,37 = 4.49, P < 0.0001), and all hybrids were capable of rapid growth under both environmental conditions. Although hybrids were generally less fit than one of the parental species at each temperature, suggesting mild extrinsic incompatibility, 29% of the hybrids were more fit thanS. paradoxus at 30°C, and 76% of the hybrids were more fit than S. cerevisiae at 10°C. In addition, hybrid fitness in one thermal environment was highly correlated with fitness in the other (r = 0.786, P < 0.001). Therefore, intermediate or fluctuating conditions may provide a mechanism for the selection of hybrids.

Figure 2

Competitive ability of F2 hybrids in two thermal environments. Hybrids have high relative fitness, in comparison with their parental species (SC, S. cerevisiae; SP, S. paradoxus), although their fitness is generally lower than that of either of the parental species at their preferred temperatures.

Recent studies have isolated fertile Saccharomyceshybrids in the laboratory (18, 19) and in nature (20,21). In this study, we showed that homoploid hybrid speciation occurs readily in laboratory populations of Saccharomyces, in contrast to all known animal species and most plant species. In part, this is due to the ability to autofertilize, which produces identical homologs in every chromosome pair (except at the mating-type locus on chromosome III) and thus avoids any incompatibilities that could arise by fusion with other gametes, even from the same parent. Autofertilization is thought to be relatively common in wild yeast (22), and it can also occur in other species with gametophytic selfing (e.g., protists, fungi, algae, ferns). Our results extend the range of known mechanisms that cause reproductive isolation. These act at different levels and in different taxa (23), but all may help produce new species.

Supporting Online Material

Materials and Methods

Table S1

Figs. S1 and S2

References and Notes

  • * To whom correspondence should be addressed. E-mail: mtrav{at}


View Abstract

Navigate This Article