Having a famous relative can be a mixed blessing. This is certainly the case for the yeast Schizosaccharomyces pombe, commonly referred to as 'fission yeast' both because it divides by binary fission and to distinguish it from its distantly related cousin, Saccharomyces cerevisiae or 'budding yeast'. Saccharomyces cerevisiae is considered by many to be the single-celled model for research into eukaryotes1 (those organisms, including humans, whose cells have a defined nucleus) and is also of major industrial importance. So, although those studying S. pombe have benefited from discoveries about S. cerevisiae, the fission yeast is often viewed as 'the other yeast', taking a back seat in research and funding.
Well, get ready everyone, because a fight for glory is brewing in the yeast family. First, a few months ago, Paul Nurse was announced as co-winner of the 2001 Nobel Prize in Physiology or Medicine, being recognized largely for his work on the cell cycle in S. pombe2, 3. Now, on page 871 of this issue, Nurse and colleagues4 report on the sequencing and analysis of the complete S. pombe genome, officially bringing the other yeast into the post-genomics era.
Schizosaccharomyces pombe is the sixth free-living eukaryotic species whose genome has been reported as completely sequenced5, 6, 7, 8, 9, 10. (Some, such as the human genome, have been announced as 'completed' even though they are not; the S. pombe sequence is actually nearer to completion than many of the others.) The analyses presented in the new paper, the sequence itself and the many bits of extra information available on websites devoted to S. pombe (see, for example, ref. 11) together represent a landmark achievement. The analyses should also satisfy those who have asked: "Why another yeast genome?".
Wood et al.4 use the S. pombe genome sequence to reveal new features of S. pombe biology, and to uncover further evidence of how different the fission and budding yeasts are. For example, S. pombe has hundreds of genes that are apparently absent in S. cerevisiae, and vice versa. The genetic differences are not as great in some areas as S. pombe researchers may have hoped; for example, there are only three disease-linked human genes that have counterparts in S. pombe but not in S. cerevisiae. But overall the differences are quite significant, and show why S. cerevisiae may not always be the preferred model eukaryote.
For instance, Wood et al. find that, compared with S. cerevisiae, S. pombe has significantly more 'intron' sequences (roughly 4,700 compared with 275), which interrupt the coding regions of genes, and very few transposable — mobile — genetic elements. S. pombe also has more proteins that appear to be involved in transporting sugars or other molecules; larger centromeres (chromosome regions needed for the accurate partitioning of chromosomes after cell division); and an apparent lack of recent whole-genome duplication. These differences could make S. pombe a better model than S. cerevisiae for understanding some eukaryotic processes.
It does not particularly surprise me that budding and fission yeast differ so much at the genomic level, as they are not very closely related12, and many genetic and physical differences had been known before the genomes were sequenced (see, for example, ref. 13). But the fact that many further differences have been uncovered by genomic comparisons4 suggests that it could prove valuable to sequence the genomes of other biologically diverse yeast species, and, more broadly, other fungi.
Wood et al. also attempt to identify genes that might be specific to eukaryotes (and so probably evolved on the branch of the evolutionary tree that separates these species from prokaryotes; Fig. 1). The authors use a very conservative approach, identifying only those genes that are highly conserved in eukaryotes and have no apparent matches in any prokaryote, so they may have missed many eukaryotic-specific genes. Nevertheless, many of those identified are predicted to function in processes specific to, or highly developed in, eukaryotes, such as the cell cycle, RNA 'splicing', construction of the cytoskeleton, protein degradation and signal transduction. So these genes may be fundamental to understanding the origin and evolution of eukaryotes. In a separate analysis, Wood et al. identified protein 'domains' — structurally defined portions of proteins — that are more abundant in eukaryotes than in prokaryotes; these may also be important in understanding eukaryotic biology.
Figure 1: The tree of life, with the branches labelled according to Wood et al.'s analysis4 of genes that might be specific to eukaryotes versus prokaryotes, and to multicellular versus single-celled organisms. Bacteria and archaea are prokaryotes (they do not have nuclei).
The eukaryotic part of the tree is based on ref. 18. Only representative lineages are shown.
High resolution image and legend (66K)The S. pombe genome is the second of a free-living, single-celled eukaryote to be completely sequenced (the first being that of S. cerevisiae). Wood et al. take advantage of this to try to identify genomic properties that are related to multicellularity. They used another conservative approach to compare the genomes of the two yeast species with the available genome sequences of multicellular eukaryotes (a plant and three animals), and found only three genes that were specific to all the multicellular species.
This may seem surprising, but it probably should not. First, the comparison did not take into account that multicellularity probably evolved separately in plants and animals14 (Fig. 1), so different multicellularity-related genes may have evolved in these two evolutionary lineages. Second, the time interval during which these genes could have evolved is much shorter than for the eukaryotic versus prokaryotic comparison — in other words, there has been less time to 'invent' new genes. Perhaps more usefully here, the authors found that, even after correcting for differences in genome size, some protein domains are more common in the multicellular than in the unicellular organisms, probably reflecting the expansion of certain protein families. This implies that such expansions may have occurred in parallel during the evolution of multicellular animals and plants, but the same genes were rarely if ever invented by both groups.
So what next? Clearly, a better comparison of eukaryotes and prokaryotes requires complete genome sequences from a more diverse sampling, not just those eukaryotes from the 'top' of the tree (Fig. 1). Lumped together as 'protists', these other eukaryotes show remarkable diversity and include many parasitic species, such as the malaria-causing Plasmodium; species such as Giardia that lack the cellular powerhouses, mitochondria; and organisms with several nuclei and unusual genome-rearrangement processes, such as Tetrahymena. It will also be interesting to use the S. pombe and other fungal genomes to do a more thorough comparison with genomes from the Microsporidia, such as the parasitic Encephalitozoon15. These organisms were once classified with the protists but are now thought to be related to fungi16.
In all these comparisons, it will be important to go beyond simply identifying the similarities and differences between species, and to analyse the origin of the differences, for example the gain, loss and possible transfer of genes over time17. But today we should do a little dance for the other yeast, and hope that in the future, when someone says 'yeast', scientists will give equal thought to the species that was first isolated from a traditional African beer known as Pombe.
