Mitochondrial genome invaders: an unselfish role as molecular markers

‘Until now, attempts to generate reliable strain-specific genetic markers have generally failed, but Formey and colleagues have found a way around this problem.’ The cellular biology and genetics of AMF are atypical compared with those of most eukaryotes. Indeed, these supposedly asexual organisms contain coenocytic cells (i.e. there is no cellular separation within their hyphae) that are multinucleated throughout their entire life cycle, and their nuclei are often found to harbor highly polymorphic genes (Corradi & Bonfante, 2012). The origin of this intracellular polymorphism has been a source of considerable debate, especially among evolutionary biologists (i.e. result from extensive gene duplications vs from heterokaryosis), but this unusual feature has also sometimes hampered our understanding of genetic processes that occur within natural populations of these plant symbionts. Specifically, the presence of sequence variability for a number of AMF nuclear genes (regardless of its origin), coupled with difficulties in amplifying nuclear-encoded genes from single spores, has resulted in frustrating attempts to identify genetic nuclear markers suitable for population-based studies of AMF. This has prompted the exploration of mitochondrial genes as a potential alternative to nuclear markers because of their intra-strain sequence homogeneity and high copy number per spore (i.e. each AMF spore typically contains hundreds of mitochondria). Consequently, mitochondrial genes, such as the large subunit of ribosomal RNA (LSU rRNA), became an attractive tool for studies of AMF diversity in the field, and have served as a sensitive marker for distinguishing Rhizophagus irregularis isolates (Borstler et al., 2008) and different Rhizophagus species (Thiery et al., 2010), mostly because of the ‘optional’ nature of introns and their associated homing endonucleases (HEOs). A new study has now further explored the nature of ‘variability-generating-elements’ in the mitochondrial genomes of AMF, unraveling their potential utility for population genetics purposes and the development of strain-specific markers. These elements have shaped the mitochondrial genomes of Rhizophagus irregularis strains, which range from just over 70 to c. 88 kb. Such inter-strain variation had already been reported in other fungi (e.g. Schizosaccharomyces pombe; Schafer, 2003), but it was unclear to what extent it existed in AMF. So what are these mobile genetic elements, and where do they come from? The mitochondrial genomes of Rhizophagus irregularis strains have numerous molecular fossils left by elements which have invaded the genomes during evolution and left genetic signatures that are well-suited for genotyping arbuscular mycorrhizal fungal strains. A common type of mobile element inhabiting fungal mitochondrial genomes is the group I intron family which in their fully-competent form encode HEOs (Edgell et al., 2011). They are tolerated within genes because their ribozyme structure allows precise excision at the RNA level. These strains of Rhizophagus irregularis possess varying numbers of introns and HEOs, especially in their cox1 and LSU rRNA genes. At the extreme, in strain 43204 these two genes occupy over 15 kb, which is almost as long as the whole human mitochondrial genome. The cox1 gene is well-known to possess strong homing sites, leading to the horizontal movement of introns even across eukaryotic kingdoms. One of these has received particular attention because in flowering plants it is the sole mitochondrial group I intron identified (although they have many group II introns that spread by ‘retrohoming’), and it is present in only certain disparate lineages of plants (Sanchez-Puerta et al., 2008). Given their close physical association with plants, AMF are prime candidates as donors for this lateral transfer, and the intronic HEO at this cox1 site in strain 43204 shares strong sequence similarity with the plant mitochondrial one, although not quite to the level seen for the zygomycete Rhizopus oryzae (Seif et al., 2005). The LSU rRNA gene in fungal mitochondria also has very hospitable integration sites, and in addition to group I introns, in strain 43204 it has a group II-type intron related in sequence to ones located at the same site in algae such as Pedinomonas minor (Turmel et al., 1999), again illustrating the extensive spread of these group I and group II intron mobile elements. A second category of invasive elements in Rhizophagus irregularis is represented by the DNA polymerase domain-containing open reading frame (ORF) (DPDCO) family. When such ORFs are present in fungal or plant mitochondria, they are typically found on plasmids (often in conjunction with other ORFs) and are believed to be of viral origin (Griffiths, 1995). If integrated into mitochondrial genomes, such elements degenerate because they are not under functional constraint, and a cluster of such relics is found in the spacer between the cox3 and LSU rRNA genes in these strains. This region is prime territory for PCR-based length polymorphism analysis. Notably, the longest DNA polymerase-related ORF in R. irregularis strains shows appreciable similarity to a plasmid-located DNA polymerase gene in Gigaspora rosea (Nadimi et al., 2012). Finally, these mitochondrial genomes also contain dispersed palindromic repeats that generate length polymorphism useful for genotyping. Short inverted repeats (SIRs) are common features of fungal mitochondrial genomes, primarily located within spacers but also occasionally within introns, as is the case for Rhizophagus irregularis. SIRs have been implicated in recombination events that contribute to the plasticity of fungal mitochondrial genomes, and interestingly, in the mitochondria of laboratory strains of Saccharomyces cerevisiae, 1-2 copies of GC-rich hairpins are located within the var1 (ribosomal protein) gene (Hudspeth et al., 1984) and so can even contribute directly to the evolution of core mitochondrial genes. Altogether, the fact that fungal mitochondrial genomes have been invaded by ‘selfish’ mobile elements provides the means to track the evolutionary history of their hosts. This is especially true for these closely-related strains of Rhizophagus irregularis because they have a completely-conserved order and orientation of all the core mitochondrial genes. This contrasts with more distantly-related AMF such as Gigaspora spp., where two genes have even had their coding segments dispersed and are expressed through trans-splicing (Nadimi et al., 2012; Pelin et al., 2012). It should be noted that because HEOs can also be independently acquired by a self-splicing intron core, their utility in phylogenetic analysis requires caution and sequence similarity must be carefully assessed to determine whether or not it reflects a long common history or horizontal acquisition/loss at orthologous sites. The presence of AMF in the soil is known to have direct and indirect beneficial effects on many terrestrial ecosystems, most notably by improving the access and absorption of many nutrients (e.g. phosphate) for many plants and by representing a habitat and a source of food for multiple microorganisms. Within this context, the overall biodiversity of AMF is believed to be crucial – that is, it is generally assumed that the more AMF species are present in one ecosystem, the more diverse the ecosystem will be. This correlation is of obvious relevance, especially for applications of AMF in different environmental practices, but it has long been based on the simplistic assumption that all members of one species affect ecosystems in similar ways. However, it has now been demonstrated that this is not the case, as strains from one AMF species can differently alter plant growth (Koch et al., 2004, 2006). In other words, some AMF individuals are better than others at improving the fitness of plants, and this is likely to result in variable outcomes for terrestrial ecosystems as a whole. Until now, attempts to generate reliable strain-specific genetic markers have generally failed, but Formey et al. have found a way around this problem. In particular, these authors have been successful in designing primers for PCR that are able to distinguish between five laboratory strains of Rhizophagus irregularis, all of which had been previously problematic to genetically ‘barcode’. This important accomplishment was made possible thanks to the presence of distinctive mobile genetic elements, whose variation in frequency, position and degree of sequence degeneration generates useful length polymorphisms among strains; a bonanza for this type of research. It seems likely that the approach used by Formey et al. will be applicable to other AMF species, especially if their mitochondrial genomes are not too recombinogenic, and could provide a wealth of molecular markers for future studies. Besides their utility for strain identification and studies of AMF diversity, the presence of numerous ‘variability-generating-elements’ in their mitochondria also raises interesting questions about the evolution of AMF genetic systems in a broader sense. In particular, it will now be interesting to see how abundant mobile elements are in their nuclear genomes, and determine if and how they have contributed to their pronounced polymorphism. Interestingly, unlike those of land plants, fungal nuclear genomes are not typically known for harboring many transposable elements. A notable exception, however, is represented by the genomes of ectomycorrhizal fungi in the genus Laccaria or Tuber (Martin et al., 2008, 2010); notorious plant symbionts. Whether the unusual wealth of transposable elements in ectomycorrhizal fungal genomes is related to their long-term association with land plants is currently unknown, but it will be interesting to see if this intriguing correlation still holds when the first AMF genome sequence becomes available. Finally, the identification of abundant transposable elements in AMF genomes could help answer a number of important questions, especially those related to their supposed long-term evolution in the absence of sexual reproduction. In particular, these may be involved in producing the remarkable intra-individual polymorphism, and the genetic variation they generate could mirror that typically produced by sexually-related processes (i.e. meiosis). This speculative idea may partially explain how these ancient asexuals can shuffle their alleles in order to reduce the overall mutational load carried by their nuclei.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.