Unseen sex in ancient virgin fungi

Abstract

‘The number of MATA-HMGs identified in the AMF is strikingly high … why was this family of genes that is typically involved in sexual reproduction expanded in an ancient, putatively asexual lineage?’ Sex is nearly universal among eukaryotes and is present in all eukaryotic lineages. However, sexual reproduction through fusion of two compatible gametes produced by meiotic recombination is a double-edged sword. Compared with mitotic propagation, sex poses several significant short-term costs to the individuals, such as the two-fold cost of producing offspring (two parents produce one offspring and only 50% of a parent's genes are transferred to an individual progeny), the time spent to find a compatible mating partner, the possibility of introducing genetic conflicts, as well as the recombinational load (the breaking up of beneficial epistatic interactions accumulated through selection). On the other hand, sex also provides many long-term benefits, which include: facilitating the removal of deleterious mutations; fixation of beneficial mutations; and generation of genetic diversity that enables organisms to better adjust to changing environments and cope with threats from pathogens (Heitman et al., 2012). Because of the near universal presence of sex in most eukaryotic species, it is commonly held that although asexual reproduction may confer short-term benefits, the ability to undergo sexual reproduction is critical for a lineage to be successful over long-term evolutionary time frames. In contrast to animals, for the vast majority of fungal species, sex is not obligatory. Only a very few fungi may be obligately sexual, and most can reproduce either sexually or asexually. Additionally, there are fungal species currently classified as asexual for which no extant sexual reproduction has been observed in the laboratory. One example of such asexual fungi is the important plant symbionts: the AMF. Arbuscular mycorrhizal fungi belong to the phylum Glomeromycota, and colonize roots of most land plants and are thought to have enabled the exodus of aquatic plants from the ocean to become terrestrialized across the continents of the planet. In exchange for plant-assimilated carbon, AMF assist plants in their uptake of mineral nutrients. Of the > 200 species that have been identified in the Glomeromycota, none has been observed to undergo sexual reproduction in the laboratory. Because of this, it has been widely thought that AMF are ancient asexuals and have propagated solely through mitosis during the c. 500 million yr history of their evolution. However, this long held view is being challenged by recent studies. Specifically, several population genetics analyses provide evidence of recombination in some natural AMF populations, suggesting sexual reproduction might be occurring in these fungal species, albeit at a low frequency and only within some subpopulations (Vandenkoornhuyse et al., 2001; Croll & Sanders, 2009; den Bakker et al., 2010). Additionally, a recent study identified key genes that belong to the meiotic machinery in the genomes of four AMF species (Halary et al., 2011). Furthermore, homologues of the fungal mating type high mobility group (MATA-HMG) gene family have been identified in Glomus intraradices (Tisserant et al., 2012), also an AMF. In many fungal lineages, genes from the MATA-HMG family are normally found to be either located within the mating type loci acting as sex determinants, or in other genomic locations and acting as regulators involved in sexual development (Fig. 1; Lee et al., 2010). Thus, the presence of MATA-HMG homologues in AMF further supports the argument that an extant cryptic sexual cycle might be present in this group of fungi. However, it was not clear how diverse the MATA-HMG genes are in AMF, as well as whether they are located within the mating type loci and indeed are playing important roles in regulating a cryptic sexual cycle in AMF. The study by Riley et al. sheds light on the answers to these questions and adds further intrigue to the possibility of sexual reproduction in AMF. By surveying the transcriptomes of three isolates that belong to two AMF species: Rhizophagus irregularis and R. diaphanous, Riley et al. found that AMF not only have MATA-HMGs as shown by a previous study, they actually have a staggering number of them! Specifically, the authors identified at least 76 homologues of MATA-HMGs in the two AMF species. Some of these genes are similar to the sexM and sexP genes found in the mating type loci of Mucormycotina fungi (Idnurm et al., 2008; Lee et al., 2008), whereas others are related to diverse mating-type genes of filamentous Ascomycetes and Schizosaccharomyces pombe. The presence of an expanded and diverse MATA-HMG gene family further supports the argument for a cryptic sexual life cycle in AMF, as these genes have been shown to be key sex determinants in several fungal lineages (Fig. 1). Riley et al. also showed that at least some of these identified MATA-HMG genes are expressed in conditions that might be expected to favor sexual reproduction. Furthermore, by analyzing the MATA-HMG homologues, Riley et al. identified 11 recombination events, although they were not able to unambiguously determine their mitotic or meiotic origins. All of these findings are consistent with the hypothesis that sexual reproduction may indeed occur in AMF. The mating type locus of AMF has yet to be identified. It is possible that some of these identified MATA-HMG homologues are orthologs of the SexM and SexP fungal sex determinants, and reside within the mating type locus. But it should be noted that not all sex related HMGs are encoded by mating type loci; for example in Basidiomycetes, MAT loci do not encode HMGs. However, it has been shown that a number of HMGs are key regulators of sexual development, such as Rop1 and Prf1 in Ustilago maydis, Pcc1 in Coprinus cinereus, and Mat2 in Cryptococcus neoformans (Murata et al., 1998; Brefort et al., 2005; Lin et al., 2010). Additionally, in Podospora anserina, of the 11 HMG genes that are involved in regulating the sexual cycle, only two are encoded by the mating type locus (Ait Benkhali et al., 2013). Thus, even though most of the MATA-HMGs identified in the AMF cannot be key sex determinants encoded by the mating type locus, they could still play important roles in a possible sexual cycle. The number of MATA-HMGs identified in the AMF is strikingly high, which raises an intriguing question: why was this family of genes that is typically involved in sexual reproduction expanded in an ancient, putatively asexual lineage? As Riley et al. pointed out in their paper, these genes might have been conserved and expanded because they were recruited for other functions, which is supported by several studies indicating that mating-type transcription factors can control genes not directly related to sexual reproduction (see references in Lee & Elion (1999) and Riley et al.). Another possible explanation for the surprisingly high number of MATA-HMGs could be due to the multi-nuclear nature of the AMF. Glomeromycota reproduce via spores containing hundreds of nuclei. It has been shown that the nuclei within the spores are not from the divisions of a single founder nucleus. Rather, these spores are populated by an influx of nuclei from surrounding mycelium (Jany & Pawlowska, 2010). Thus, genetic polymorphisms may exist among these nuclei. Indeed, the heterogeneous, multi-nuclear nature of the AMF, coupled with inter-nuclear selection, has been proposed as a way through which AMF cope with an absence of sexual reproduction. If so, this might suggest that inter-nuclear selection might favor diversification, or pose less stringent purifying selection pressure, on the MATA-HMGs. Mating of AMF has yet to be observed in the laboratory. It is possible that laboratory conditions required for mating remain to be defined, or compatible mating partners have yet to be isolated from natural environments. It is also possible that the AMF undergo sexual reproduction only in certain locations and/or at certain times. Or maybe their sexual cycle is so unique that it has yet to be recognized? What if the MATA-HMGs are involved in some type of genetic information exchange among the nuclei within the individual, which would in turn provide the selection pressure to maintain the highly diversified pool of MATA-HMGs? Many questions still remain regarding the posited sexual reproduction of AMF. Nevertheless, accumulating evidence suggests that an extant cryptic sexual cycle may indeed be present in these ‘ancient asexuals’. Preparation of this commentary was supported by NIH/NIAID R37 award AI39115-15 and R01 award AI50113-10 to J.H. The authors apologize to scientists whose work was not cited here due to space limitations.