OUP user menu

Schizosaccharomyces cryophilus sp. nov., a new species of fission yeast

Rachel M. Helston, Jessica A. Box, Wen Tang, Peter Baumann
DOI: http://dx.doi.org/10.1111/j.1567-1364.2010.00657.x 779-786 First published online: 1 September 2010


The genus Schizosaccharomyces is presently comprised of three species, namely Schizosaccharomyces pombe, Schizosaccharomyces octosporus and Schizosaccharomyces japonicus. Here, we describe a hitherto unknown species, Schizosaccharomyces cryophilus, named for its preference for growth at lower temperatures than the other fission yeast species. Although morphologically similar to S. octosporus, analysis of several rapidly evolving sequences, including the D1/D2 divergent domain of the large subunit (LSU) rRNA gene, the RNA subunit of RNAse P and the internal transcribed spacer elements, revealed significant divergence from any previously characterized Schizosaccharomyces strain. Based on phylogenetic analysis of the D1/D2 domain of the LSU rRNA gene, S. octosporus is the closest known relative of S. cryophilus, with the sequences of the two species differing by 25 nucleotide substitutions (>4%). Sequencing of the S. cryophilus genome and phylogenetic analysis of all 1 : 1 protein orthologs confirmed this observation, and together with morphological and physiological characterization, supports the assignment of S. cryophilus as a new species within the genus Schizosaccharomyces. The type strain of the new species is NRRL Y-48691T (=NBRC 106824T=CBS 11777T).

  • fission yeast
  • D1/D2
  • RNAse P
  • ITS
  • Schizosaccharomyces pombe
  • 454 genome sequence


Fission yeasts are unicellular eukaryotes, allocated to the genus Schizosaccharomyces in the phylum Ascomycota. The main unifying feature of this group of yeasts is division by medial fission (mechanism reviewed in Chang & Nurse, 1996). This genus is currently comprised of three species: Schizosaccharomyces pombe (Linder, 1893), containing the varieties S. pombe var. pombe and S. pombe var. malidevorans (Sipiczki, 1982); Schizosaccharomyces octosporus (Beijerinck, 1894); and Schizosaccharomyces japonicus (Yukawa & Maki, 1931), comprised of the variants S. japonicus var. japonicus and S. japonicus var. versatilis (Wickerham & Duprat, 1945). See Vaughan-Martini & Martini (1998) for a review of the genus. Several lines of evidence suggest that the evolutionary distance between S. pombe and S. octosporus is shorter than the distance between either of them and S. japonicus. Firstly, the U6 snRNA intron sequence is most divergent in S. japonicus (Frendewey, 1990; Reich & Wise, 1990), as are the sequences of 18S and 26S rRNA genes (Yamada, 1973). Secondly, while it is possible to create somatic hybrids between S. pombe and S. octosporus (Sipiczki, 1979), no prototrophs were detected when either of these strains was crossed with S. japonicus (Sipiczki, 1982). Furthermore, while both S. pombe and S. octosporus contain cytochrome oxidase (a+a3) and cytochrome c, the levels of these two proteins are undetectable in S. japonicus (Sipiczki, 1982). These observations refuted the idea that S. octosporus and S. japonicus be grouped together in a separate genus, Octosporomyces, as proposed earlier (Kudriawzew, 1960).

Based on the physiological differences between S. japonicus and the other two Schizosaccharomyces sp. Yamada & Banno (1987) proposed that S. japonicus be assigned to a new genus, Hasegawaea. Subsequent genetic examination of rRNA gene sequences (Naehring, 1995; Kurtzman & Robnett, 1998) rejected this notion and S. japonicus has remained in the Schizosaccharomyces genus.

Fission yeasts normally propagate mitotically as haploids, but under conditions of nitrogen starvation, sexual union between two haploid cells of opposite mating types occurs. The first physical step of mating involves cell elongation with the formation of conjugation tubes as a result of extrinsic growth toward pheromones secreted by cells of opposite mating types (Leupold, 1987). Conjugation between two cells of opposite mating types is followed by karyogamy to produce a diploid zygote which undergoes meiosis, forming four haploid nuclei which are then encapsulated by spore walls. In species that produce eight ascospores (S. japonicus and S. octosporus), meiosis II is followed by a mitotic division resulting in the generation of eight spores (Tanaka & Hirata, 1982).

Since the initial discovery of S. pombe (Linder, 1893), several other species have been described in this genus including Schizosaccharomyces liquefaciens (Osterwalder, 1924) and Schizosaccharomyces slooffiae (Kumbhojkar, 1972), but a closer examination of nuclear DNA/nuclear DNA reassociation kinetics (Martini, 1991) revealed them to be conspecific with one of the three aforementioned species. Here, we describe a new species of fission yeast on the basis of phenotypic differences, genome sequencing and phylogenetic analyses. Based on its preference for growth at lower temperatures than other fission yeasts, we named the new species Schizosaccharomyces cryophilus.

Materials and methods

Strains and culture media

The strains used in this study were obtained from Centraalbureau voor Schimmelcultures (CBS), Utrecht, the Netherlands, or the American Type Culture Collection, Washington, DC and are listed in Table 1.

View this table:

Strains utilized in this study

Strain name
S. pombe var. pombe CBS 356T
S. pombe var. pombe 972 h-ATCC 38366
S. octosporus var. octosporus CBS 371T
S. octosporus var. octosporus (formerly S. slooffiae) CBS 6207T
S. octosporus var. octosporus CBS 6206
S. japonicus var. japonicus CBS 354T
S. japonicus var. versatilis CBS 103T
S. cryophilus NRRL Y-48691T, NBRC 106824T, CBS 11777T
  • Sources: CBS, Centraalbureau voor Schimmelcultures, Utrecht, the Netherlands; ATCC, American Type Culture Collection, Washington, DC. Schizosaccharomyces cryophilus has been deposited at NRRL, Agricultural Research Service Culture Collection, National Center for Agricultural Utilization Research, Peoria, IL; CBS and NBRC, NITE Biological Resource Center, National Institute of Technology and Evaluation, Chiba, Japan.

Strains were propagated on YES (0.5% yeast extract, 3% glucose, 0.01% leucine, 0.01% uracil, 0.01% histidine-HCl, 0.01% adenine, 2% agar) or YPD (1% yeast extract, 2% peptone, 2% glucose, 2% agar) plates. For sporulation, strains were plated on G25YA (29% glucose, 0.5% yeast extract, 1.5% agar), ME (3% malt extract, 2% agar), ME+0.4% glucose, PMG (0.3% potassium hydrogen phthalate, 0.415% Na2HPO4·7H2O, 2% dextrose, 1 × salt stock, 1 × vitamin stock, 1 × mineral stock, 2% agar) or SPA (1% potassium diphosphate, 1 × vitamin stock, 3% agar) media. See Alfa (1993) for the composition of the stock solutions.


Liquid cultures were passed through an 18-gauge needle to break up large cell clumps. Samples were imaged using a Zeiss AxioVert microscope with a 63 × 1.4 NA Plan-APOCHROMAT DIC objective. Cell dimensions were measured using axiovision software (Carl Zeiss Jena GmbH).

Pulsed-field gel electrophoresis (PFGE)

PFGE was performed as described previously (Baumann & Cech, 2000), except that the final concentration of Zymolyase (US Biological) was increased to 0.6 mg mL−1.

PCR amplification of genomic fragments

Genomic DNA was isolated as described previously (Bunch, 2005). The D1/D2 domain of the large subunit (LSU) rRNA gene and the internal transcribed spacer (ITS) elements were amplified and sequenced using primers in flanking conserved regions. RNAse P was amplified using primers complementary to conserved regions of flanking ORFs based on alignment of the S. pombe, S. japonicus and S. octosporus sequences. The ITS elements were amplified from all species, examined using the oligonucleotides BLoli1467 (5′-TCCTAGTAAGCGCAAGTCATCAGC-3′) and BLoli1468 (5′-AATTTGAGCTTTTCCCGCTTCACTCG-3′). The D1/D2 domain of the LSU rRNA gene was amplified according to Kurtzman & Robnett (1998). RNAse P was amplified using the primers BLoli1474 (5′-ACCCAAGTCATYTCAATTTCATGC-3′) and BLoli1475 (5′-GYGCTGGAAAGGGYACACAATGC-3′) for S. japonicus and BLoli1477 (5′-ACCGTTGTAWCCCCAATAAACACC-3′) and BLoli1478 (5′-GTGCTGGWAAGGGTACMCAATGC-3′) for S. cryophilus. PCR reactions (50 μL) contained 1 × Pfu reaction buffer (Stratagene), 200 μM dNTPs, 0.5 μM of each primer, 2.5 U PfuTurbo DNA polymerase and 15 ng genomic DNA. In most reactions, denaturation at 94 °C for 2 min was followed by 30–35 cycles of 94 °C for 30 s, 53 °C for 45 s and 72 °C for 3 min. To amplify RNAse P from S. japonicus, an annealing temperature of 50 °C and an extension time of 5 min were used. Reactions were completed by a final extension at 72 °C for 7–10 min. DNA products were subjected to electrophoresis on agarose gels, recovered using the Qiaquick Gel Extraction kit (Qiagen) and sequenced using the primers listed in Table 2.

View this table:

Sequencing primers

GeneSpeciesOligo sequence (5′–3′)

Sequence retrieval and analysis

RNAse P sequences from S. pombe, S. octosporus and S. cerevisiae were obtained from the RNAse P database: http://www.mbio.ncsu.edu/RnaseP/home.html (Brown, 1999). Flanking ORF sequences were obtained from the Broad Institute. RNAse P from S. japonicus and S. cryophilus was cloned based on homology in flanking ORFs. The D1/D2 LSU rRNA gene sequences for S. pombe, S. octosporus and S. japonicus and the ITS sequences for S. pombe and S. japonicus were retrieved from GenBank. The ITS sequences for S. octosporus and S. cryophilus and the D1/D2 region for S. cryophilus were cloned based on homology to the corresponding S. pombe sequence.

Sequence alignments were performed using jalview (Waterhouse, 2009) and phylogenetic trees were constructed using mega4 (Tamura, 2007). Saccharomyces cerevisiae was included as an outgroup representative. The nucleotide sequences determined in this study have been deposited into GenBank with the following accession numbers: S. cryophilus NRRL Y-48691T D1/D2, GU470882; S. cryophilus NRRL Y-48691T RNAse P, GU470883; S. japonicus CBS 354T RNAse P, GU470884; S. cryophilus NRRL Y-48691T ITS regions, GU470885; and S. octosporus CBS 371T ITS regions, GU470886.

Genome sequencing

Genomic DNA from S. cryophilus was submitted to 454 Life Sciences for whole-genome sequencing. A titanium fragment library was generated from 5 μg genomic DNA and sequenced on a single 454 run using titanium chemistry. A 3K long-tag paired-end (LTPE) library was generated in parallel from the same amount of genomic DNA and sequenced on half of an FLX run. Approximately 1.3 million high-quality reads of 400 bp were generated from the titanium run and 700 000 reads were generated from the LTPE FLX run. Quality-filtered sequences from the whole-genome shotgun fragment library and LTPE library were assembled using the 454 newbler assembler.

Phenotypic characterization

To compare the growth rate of fission yeast strains at different temperatures, liquid cultures (YES and YPD) were diluted to 2 × 106 cells. Serial dilutions (1 : 5) were prepared and 10-μL aliquots were spotted onto YES and YPD agar media. Plates were photographed following incubation at 18, 25, 28, 32 and 36 °C for 3 days.

Fermentation of carbohydrates was assessed according to standard methods (Yarrow, 1998). Carbon assimilation tests were examined using the ID 32 C test strip (bioMérieux) according to the manufacturer's instructions. Urea hydrolysis was tested on a solid medium using BBL urea agar prepared slants (BD Biosciences) according to the manufacturer's instructions.

Results and discussion

As part of the characterization of the telomerase RNA subunit (TER1) from fission yeast (Leonardi, 2008; Webb & Zakian, 2008), we amplified and sequenced the TER1 encoding region from genomic DNA samples of over 100 isolates of fission yeast. During this analysis, which will be published elsewhere, we observed a sequence that differed substantially from all the others. A closer examination of the original culture (S. octosporus, CBS 7191) revealed the presence of a heterogeneous colony morphology on the agarose plate, suggesting that the previously characterized S. octosporus isolate was mixed with another organism. To further investigate the origin of the distinct TER1 sequence, we isolated single colonies and prepared genomic DNA for further sequence analysis.

Sequence analysis

To examine sequence differences at specific loci, we sequenced the D1/D2 divergent domain of the LSU rRNA gene, the RNA subunit of RNAse P and the ITS elements, three rapidly evolving sequences commonly used in phylogenetic analyses (reviewed in Cho, 1998; Haas & Brown, 1998; Kurtzman & Robnett, 1998; Iwen, 2002; Nilsson, 2008). The phylogenetic tree constructed from the D1/D2 nucleotide sequence data using S. cerevisiae as an outgroup representative shows that S. cryophilus clusters with S. pombe, S. octosporus and S. japonicus, supporting the fact that it is a member of the Schizosaccharomyces genus (Fig. 1a). It is most closely related to S. octosporus, although the two species differ by 25 nucleotide substitutions (>4%) and three indels. Given that a nucleotide substitution rate >1% is indicative of distinct species (Kurtzman & Robnett, 1998), we propose that S. cryophilus represents a previously uncharacterized species. The sequences of all S. octosporus strains examined were identical and there were only one and five differences between the two S. pombe strains and the two S. japonicus strains, respectively, thereby confirming these as conspecific strains. Conspecific strains were excluded from further analyses.


Neighbor-joining phylogenetic tree based on sequences of (a) the D1/D2 divergent domains of the LSU rRNA gene and (b) the RNA subunit of RNAse P. Saccharomyces cerevisiae was included as an outgroup representative. GenBank accession numbers are indicated. Numbers represent the bootstrap values from 1000 replicates (values <50% are not shown). Scale bar represents the number of substitutions per nucleotide position.

Phylogenetic analysis of the RNA subunit of RNAse P yielded a similar phylogenetic tree (Fig. 1b). Again, S. cryophilus was most closely related to S. octosporus, with the sequences of these two species differing by 15 nucleotide substitutions (>5%) and three indels.

There was a significant degree of disparity in ITS length between the species examined (Table 3), with the sequence of ITS1 in S. japonicus (183 bp) being over 200 bp shorter than the equivalent region in S. pombe (417 bp). The ITS2 length was more uniform, with a maximum difference of only 46 bp. Comparison of the S. cryophilus and S. octosporus ITS1 and ITS2 sequences revealed 95 nucleotide substitutions/66 indels and 84 nucleotide substitutions/51 indels, respectively, further confirming S. cryophilus as a distinct species.

View this table:

ITS length (bp) in fission yeast species

ITS1ITS2Accession no.
S. pombe CBS 356T417292Z19578
S. japonicus CBS 354T183252AB243296
S. octosporus CBS 371T343298GU470886
S. cryophilus NRRL Y-48691T307279GU470885

As an additional step to examine sequence differences between S. cryophilus and other fission yeast species, intact chromosomal DNA was subjected to NotI digestion and subsequent PFGE. The banding pattern observed in S. cryophilus differed notably from the pattern observed in samples from S. pombe and S. octosporus (Fig. 2), indicative of significant sequence divergence across the genome. It is noteworthy that the digestion patterns of all three S. octosporus strains examined were very similar to each other, but distinct from S. cryophilus. Schizosaccharomyces japonicus was excluded from this analysis as the aforementioned sequence analysis had shown it to be a more distant relative than S. octosporus and S. pombe.


NotI-digested genomic DNA was subjected to PFGE, stained with ethidium bromide and visualized on a Typhoon PhosphorImager using a 532 nm laser and a 610 BP 30 filter. Saccharomyces cerevisiae chromosomal DNA was used as a marker (M).

Having established profound sequence divergence between S. cryophilus and all previously described species of fission yeast, we surmised that an S. cryophilus genome sequence would constitute a useful resource for the fission yeast community as well as for evolutionary and computational biologists interested in this genus. Genomic DNA was sequenced using 454 Life Sciences' technology for whole-genome sequencing and the Broad Institute generously agreed to host and annotate a draft sequence on their Schizosaccharomyces group website (http://www.broadinstitute.org/annotation/genome/schizosaccharomyces_group/GenomesIndex.html). Genome annotation and subsequent phylogenetic analysis of 1 : 1 protein orthologs confirmed that S. cryophilus is most closely related to S. octosporus, with the two species sharing 85% amino acid identity. The amino acid identity between 1 : 1 orthologs is limited to 66% with S. pombe and 55% with S. japonicus (B. Haas, Broad Institute, pers. commun.).

Latin diagnosis of S. cryophilus Helston, Box, Tang & Baumann sp. nov.

In media liquido YPD in 25 °C cellulae sunt spheroidae aut ovoidae (5.25±0.20 μm × 4.85±0.10 μm). In agaro YPD post dies quinque at 25 °C cultura butyrosa et cremea. Margine undulatis. Multiplicatio non-sexualis fissione. Species homothallica. In agaro malti aut PMG asci cum 1-8 sporis formatur.

Glucosum, maltosum et sucrosum (exigue) fermentantur at non galactosum. Glucosum, maltosum et methyl α-d-glucopyranosidum assimilantur at non galactosum, sucrosum, trehalosum, raffinosum, lactosum, d-xylosum, l-sorbosum, glucosaminum, d-ribosum, l-arabinosum, l-rhamnosum, d-cellobiosum, d-melibiosum, d-melezitosum, glycerolum, erythritolum, d-mannitolum, inositolum, 2-keto-d-gluconicum, acidum d-gluconicum, acidum d-glucuronicum, acidum lacticum, d-sorbitolum, palatinosum, acidum levulinicum nec N-acetylglucosaminum. Non crescit in 0.01% cycloheximido. Urea finditur.

Typus stirps NRRL Y-48691T (=NBRC 106824T=CBS 11777T).

Description of S. cryophilus Helston, Box, Tang & Baumann sp. nov.

On YPD agar, after growth for 5 days at 25 °C, colonies are cream-colored and butyrous with undulate margins (Fig. 3a), in contrast to the smooth, rounded appearance of S. pombe colonies and the foam-like appearance of S. octosporus colonies. When propagated in liquid YPD medium at 25 °C, cells are rounded and have a high tendency to flocculate (Fig. 3b). The overall morphology is most similar to S. octosporus, but S. cryophilus cells are smaller in both length (5.25±0.20 vs. 7.00±0.34 μm) and width (4.85±0.10 vs. 5.68±0.09 μm). Cells elongate slightly during division, which occurs by fission. When plated on PMG or ME media, after 3 days, fresh cultures produced one to eight spores, with tetrads and hexads being the most common (Fig. 3c). The ability to sporulate was rapidly lost upon subculturing, as is common in a number of other yeasts (Yarrow, 1998). The observation of conjugation tubes (indicated with arrows in Fig. 3c) suggests that the isolate of S. cryophilus is a homothallic strain.


Schizosaccharomyces cryophilus morphology. (a) Fission yeast cultures were grown on YES agar plates at 32°C (Schizosaccharomyces octosporus CBS 371T and Schizosaccharomyces pombe CBS 356T) or 25°C (S. cryophilus NRRL Y-48691T) for 3–5 days. (b) Schizosaccharomyces cryophilus NRRL Y-48691T and S. octosporus CBS 371T cultures were grown to the mid-log phase in YPD and YES media, respectively. Scale bars=10 μm. (c) Schizosaccharomyces cryophilus NRRL Y-48691T was streaked on PMG or ME agar plates and imaged after 3 days. Scale bars=10 μm.

To determine the optimal growth temperature for S. cryophilus, serial dilutions of cells were spotted onto both YES and YPD plates and incubated at various temperatures. While optimal growth of S. japonicus, S. pombe and S. octosporus occurred at 32 °C, these three species were capable of growth across the temperature range from 18 to 36 °C. In contrast, the growth of S. cryophilus was optimal at 25 °C, with little or no growth at higher temperatures (Fig. 4a). The growth of S. cryophilus was slightly better on YPD media (Fig. 4a) than on YES media (Fig. 4b).


Spotting growth assay of fission yeast strains. Comparison of the growth of fission yeast strains Schizosaccharomyces japonicus CBS 354T, Schizosaccharomyces pombe CBS 356T, Schizosaccharomyces octosporus CBS 371T and Schizosaccharomyces cryophilus NRRL Y-48691T. Cultures of each strain were serially diluted, spotted onto (a) YPD or (b) YES media and incubated at the indicated temperatures for 3 days.

The physiological characteristics of S. cryophilus and the three other fission yeast species are presented in Table 4. As observed for S. pombe, S. cryophilus is able to ferment d-glucose, maltose and also sucrose (albeit weakly), but not d-galactose. Use of the ID 32 C test strip revealed that this species is able to assimilate d-glucose, d-maltose and methyl-α-d-glucopyranoside (albeit delayed). Most ascomycetous yeasts are unable to hydrolyze urea, Schizosaccharomyces sp. being an exception. Inoculation of urea agar slants revealed that S. cryophilus is also able to hydrolyze urea, although at a much slower rate than S. pombe. Growth is negative in the presence of 0.01% cycloheximide.

View this table:

Physiological properties of fission yeast

PropertyS. cryophilus NRRL Y- 48691TS. pombe CBS 356TS. octosporus CBS 371TS. japonicus CBS 354T
Fermentation of carbon compounds
Assimilation of carbon compounds
Lactic acid
Levulinic acid
Other tests
  • * Data from CBS.

  • d, delayed; w, weak; ND, not determined.

Origin and species designation

Schizosaccharomyces cryophilus was isolated in our laboratory as a contaminant of the S. octosporus strain CBS 7191, which had been isolated in Taastrup, Denmark, from the feed of Osmia rufa and deposited by J.P. Skou at the Centraalbureau voor Schimmelcultures. While it is unclear whether S. cryophilus cells were present in the original isolate of S. octosporus, we consider this a reasonable possibility for the following reasons: first, to our knowledge, no other isolate of S. cryophilus has been described, making it unlikely that the S. octosporus culture was contaminated with another strain in our laboratory or elsewhere. Second, the tendency of S. octosporus and S. cryophilus cells to clump together and form mixed colonies favors the possibility that S. cryophilus could have been carried along in the S. octosporus culture until we coincidentally isolated it.

The description of S. cryophilus is based on a single isolate of uncertain origin. While this is unsatisfying from a taxonomic viewpoint, we believe that description of this isolate as a species is justified in light of the substantial sequence divergence from all other known species. Most importantly, the description of S. cryophilus makes the organism and the associated sequencing data available to the scientific community for further analysis. It is hoped that identification of additional isolates by others will soon shed more light on the biology of this organism.

The type strain has been deposited at NRRL, Agricultural Research Service Culture Collection, National Center for Agricultural Utilization Research, Peoria, IL, as Y-48691T, the Centraal Bureau voor Schimmelcultures, Utrecht, the Netherlands, as CBS 11777T and at NBRC, NITE Biological Resource Center, National Institute of Technology and Evaluation, Japan, as NBRC 106824T.


We acknowledge the Stowers Institute Microscopy Center for training and the Molecular Biology Facility for sequencing services. We thank Chad Nusbaum and Brian Haas at the Broad Institute for hosting the S. cryophilus sequence on their web server and for sharing information before publication. We also thank Matthias Sipiczki and Nick Rhind for comments on the manuscript and the members of the Baumann lab for helpful discussions. This study was funded by the Stowers Institute. P.B. is an HHMI Early Career Scientist.


  • Editor: Cletus Kurtzman


View Abstract