Fifty-two strains of the genus Succhuromyces, including 50 yeasts of the Succharomyces sensu stricto complex (sensu Vaughan Martini 1989 [A. Vaughan Martini, Syst. Appl. Microbiol. 12119-122, 1989]), were analyzed for nine isozyme loci with starch gel electrophoresis. The genetic analysis of isozyme data provided further evidence that the Succhuromyces sensu stricto complex is a genetically heterogeneous group. Twenty-two different electrophoretic types were identified and could be grouped into four clusters. However, no correlation between these clusters and the currently defined species Succhuromyces cerevisiue, Succhuromyces buyunus, Succhuromyces pustoriunus, and Succharomyces paradom was detected. Several strains could not be assigned to their putative cluster.The criteria used in traditional yeast taxonomy are not always satisfactory for differentiation of species, in part because many separations are based on the presence or absence of assimilative abilities, which are often controlled by single mutable genes (15). Vaughan Martini (45) differentiated several species within the Saccharomyces sensu stricto group, while Yarrow (52) considered these as a single species (S. cerevisiae). This is a good example of this problem of strain variability in standard carbon assimilation and fermentation tests (3, 52).
Other traditional methods also lead to ambiguous results (5).Recently there have been several reports on the use of molecular techniques to distinguish between strains of the Saccharomyces sensu stricto group. These criteria include levels of rRNA and DNA relatedness (16, 36,46), electrophoretic whole-cell protein patterns (44), restriction endonuclease patterns (17, 22, 28, 30), and electrophoretic karyotyping (23, 31,49). Using DNA-DNA reassociation, Vaughan Martini et al. (46,48) found that at least four species exist in the Saccharomyces sensu stricto complex: S. cerevisiae, S. bayanus, S. paradoxus, and S. pastotianus. This technique has been widely used to define species limits and relationships. However, it lacks the precision required for the analysis of genetic relationships among closely related strains (37). Analysis of whole-cell protein patterns and restriction enzyme analysis of genomic DNA or genes coding for rRNA were largely limited to differentiation of strains (12). Restriction enzyme analysis of mitochondrial DNA has given good results in the past (6, 22, 34) but requires a large number of screenings in order to define a suitable restriction enzyme. Recently, Guillamon et al. (9) have differentiated among the four species with four restriction enzymes. However, their results were not statistically checked.