In this issue of Genome Biology, Espagne et al. 2 publish the genome sequence of Podospora anserina, a joint effort between the Podospora research community and Genoscope, the French National Sequencing Center. P. anserina is one of the most recent additions to the constantly growing collection of fungal genomes 2, but it has been around as a fungal genetic model for quite a while, having been introduced in the 1930s by the late French geneticist Georges Rizet. Podospora anserina is a coprophilous fungus inhabiting the dung of various herbivores such as rabbits, goats or horses. In contrast to other popular fungal models such as Aspergillus and Neurospora, it lacks asexual reproduction and it is strictly dependent on the sexual cycle for production of the resistant form, the ascospore. The presence of an appendage on these ascopore spawned the name of the genus: Podospora, spore with a foot. The sexual cycle can be completed in as little as a week and typically produces bunches (rosettes) of four-spored asci (Figure 1). The ascospores are heterokaryotic (that is, they contain nuclei of different genetic constitution) and invariably contain nuclei of opposite mating-types; as a consequence, the mycelium germinating from an ascospore is self-fertile. The early work of Rizet on P. anserina introduced a particular emphasis on nucleocytoplasmic interactions and cytoplasmic inheritance. The discovery of the senescence syndrome, a cytoplasmically transmitted aging process, probably represents the first described example of cytoplasmic inheritance in fungi. Currently, P. anserina is used as a model species in the study of mating type, aging, cell death, genome conflicts, conspecific and heterospecific non-self recognition, and prion biology and structure.

Within the ascomycetes, P. anserina belongs to the sordariomycetes, a group that also includes Neurospora crassa, the rice pathogen Magnaporthe grisea, and the wheat pathogen Fusarium graminearum, all of whose sequences have been published 345. The genome of an even closer relative, Chaetomium globosum, is also publicly available but has not yet been published. The genome sequencing of P. anserina was prompted both by its phylogenetically interesting position as a close relative of N. crassa (the red bread mold) and the existence of a body of fundamental research on its biology spanning seven decades. Podospora never attained the general popularity of Aspergillus or Neurospora, but has been extensively studied as a fungal model, especially in France and Germany. It is also the first coprophilous fungus for which the genome is available, and as such, might serve as a proxy for many other species of this extremely diverse and ecologically important group.

A basic trend revealed by comparative fungal genomics is the high divergence in sequence of species that appear morphologically or even phylogenetically closely related. For instance, the genomes of the three Aspergillus species A. nidulans, A. fumigatus and A. oryzae, revealed only 68% average sequence identity between any species pair 6. Even more strikingly, extensive sequence differences can be found between two isolates of the same species 57. The genome published by Espagne et al. 2 illustrates the trend towards divergence. The authors mainly emphasize the comparison with N. crassa. On average, sequence conservation of orthologous genes between N. crassa and P. anserina is only 60.5%, and roughly a quarter of predicted proteins in P. anserina lack orthologs in either N. crassa, M. grisea or A. nidulans. These numbers serve to remind us that the evolutionary distance separating these seemingly closely related organisms is on the order of the distance between fishes and humans.

Another feature of fungal genome evolution is the important and frequent chromosomal rearrangements that occur through random breakages 1. Analysis of syntenic blocks of orthologous sequences between P. anserina and N. crassa again highlights this tendency. Short blocks of synteny are more frequent than long ones, fitting with the model of random breakage. A surprise, however, is the fact that rearrangements in P. anserina and N. crassa appear mostly intra-chromosomal, as revealed by the high conservation of chromosomal gene content. This is in contrast with previous observations in Aspergillus genomes, where syntenic blocks are spread over all chromosomes between the three compared species 6. As Espagne and co-authors point out, this observation may relate to the heterothallic life-style of N. crassa and P. anserina, which requires the pairing of homologous chromosomes from different nuclei; this would not suit well with interchromosomal rearrangements. Significantly, a single syntenic block encompassing 37 orthologous gene pairs stands out from the random-breakage model. This is the largest syntenic block shared by P. anserina and N. crassa and is centered on the mating-type locus, which controls sexual development. This supports the idea that recombination is restricted around sex-controlling loci, as noted in Aspergillus and N. tetrasperma 68. In the latter species, recombination is actually suppressed over most of the chromosome carrying the mating-type locus. This peculiarity appeared after the split between N. crassa and N. tetrasperma, and it is suggested that in N. tetrasperma we might be witnessing the early steps in the evolution of a proper sex chromosome 8.

Overall, repeated sequences are rare in filamentous fungi compared with plant or animal genomes. This might in part be due to the existence of genetic mechanisms aimed at the inactivation of mobile genetic elements, such as repeat-induced point mutation (RIP), meiotic silencing of unpaired DNA (MSUD) and 'quelling', all originally described in N. crassa. RIP operates during the sexual cycle and heavily mutates and methylates both copies of any repeated sequence as long as it is longer than about 800 bp 9. P. anserina displays the full fungal arsenal against mobile genetic elements, as all the genes required for RIP, MSUD and quelling appear functional. So far, however, only RIP has been demonstrated in P. anserina in laboratory conditions 10. Despite this equipment, the P. anserina genome does have repeated sequences. Gene families generated through duplications have evolved in P. anserina, despite the fact that these duplications could be potential targets for RIP mutagenesis. It would thus appear that the P. anserina genome evolved through a history of transpositions, duplications and gene losses, accompanied by a low level of RIP that preserved it against transposons (as most are inactivated by RIP) while possibly increasing the divergence of copies of duplicated genes 11.

As a coprophilous fungus, P. anserina grows exclusively on herbivore dung. This is an ecologically rich microcosmos where, alongside dozens of fungal species, bacteria, animals and plants are also represented. Coprophilous fungi typically appear in a phylogenetically determined succession during dung degradation. Typically, zygomycetes come first, followed by ascomycetes, which finally leave the last crumbs of the feast to the basidiomycetes. One study on game animal dung from the Kruger National Park in South Africa identified a succession of 106 species belonging to 23 genera over a period of 112 days 12; thus, competition must be fierce for both resources and territory. P. anserina appears to be one of the last of the ascomycetes to reach its ecological peak (the time when it becomes predominant) in this habitat, and by then the simple carbohydrates are depleted. Espagne et al. 2 show that to exploit the limited resources, P. anserina possesses formidable enzymatic tools for degrading complex biopolymers, including enzymes that potentially can degrade cellulose/hemicellulose, xylan and even lignin. The authors report that the ability of P. anserina to grow on media containing different complex carbon sources is in line with the existence of this complex enzymatic tool-box. At the same time, P. anserina has lost the enzymatic potential to degrade 'easier' carbohydrates such as sucrose. This is in sharp contrast to the enzymatic equipment of the ectomycorrhizal basidiomycete Laccaria bicolor, which has lost many enzymes that degrade plant cell walls, presumably to avoid harmful damage to its plant host during symbiotic development 13. The P. anserina enzymatic equipment is unique among the ascomycete genomes sequenced. In certain aspects, P. anserina even rivals basidiomycetes of biotechnological interest, such as the wood-degrading fungus Phanerochaete chrysosporium, which causes white rot 14. It thus appears that P. anserina's life-style on dung promoted the development of a copious assortment of enzymes to degrade complex biopolymers. This rich gene repertoire might potentially turn P. anserina into a viable alternative or complement to the white-rot basidiomycete fungi in biotechnological applications such as bioremediation or industrial biomass processing 15.

For better or worse, we depend for much of our biological knowledge on a handful of model organisms. While the benefits of S. cerevisiae for modern cell biology are undisputable, it is now clear that it is a very peculiar organism and - in some aspects - not that good a model for other fungi or for eukaryotes. It is very fortunate that the field of fungal genetics has entertained a variety of models over the years rather than relying only on one superstar. The value of this diversification now comes to full bloom with the progressive entry of the field into the genomic era. The fundamental impact of comparative genomics is, and will certainly continue to be, considerable. The amusing surprise here with the work by Espagne et al. 2 is that P. anserina, originally selected as a tool for formal genetics, might in the longer run make an unexpected career in biotech.