Portal de Periódicos da CAPES

Plants acquired a major retrotransposon horizontally from fungi during the conquest of land

2021; Wiley; Volume: 232; Issue: 1 Linguagem: Inglês

10.1111/nph.17568

1469-8137

Qin Wang, Yan Wang, Jianhua Wang, Zhen Gong, Guan‐Zhu Han,

Plant Disease Resistance and Genetics

Long-terminal repeat (LTR) retrotransposons are transposable elements that mobilize via RNA intermediates, and comprise a significant portion of land plant genomes. For example, LTR retrotransposons account for more than 4% of the Arabidopsis thaliana genome and more than 74.6% of the maize (Zea mays L.) genome (Peterson-Burch et al., 2004; Schnable et al., 2009). Typically, LTR retrotransposons have been classified into three major types, Ty1/Copia, Ty3/Gypsy and Bel/Pao (Wells & Feschotte, 2020). Essentially reverse-transcribing viruses, Ty1/Copia, Ty3/Gypsy and Bel/Pao retrotransposons have been unified with Caulimoviridae and Retroviridae into a single viral order, Ortervirales (Krupovic et al., 2018). LTR retrotransposons can move horizontally between eukaryotes (including land plants), potentially shaping the evolution of genome complexity in eukaryotes (Novikova et al., 2010; Schaack et al., 2010; Gao et al., 2018). Diverse lineages (or families) of Ty1/Copia and Ty3/Gypsy retrotransposons have been identified within land plant genomes (Feschotte et al., 2002). Among the Ty3/Gypsy retrotransposons, the Athila/Tat retrotransposon lineage comprises a significant component of land plant retrotransposons. For example, the Athila/Tat retrotransposons account for more than 1% of the A. thaliana genome (Pelissier et al., 1995; Peterson-Burch et al., 2004). The Athila/Tat retrotransposons have been known to be widely distributed in the genomes of land plants (Pelissier et al., 1995; Wright & Voytas, 1998, 2002; Vicient et al., 2001; Peterson-Burch et al., 2004). Unique among the Ty3/Gypsy retrotransposons, some Athila/Tat elements encode an env-like gene with predicted transmembrane domains, and the pol gene of some Athila/Tat elements encodes a dual ribonuclease H domain (Peterson-Burch et al., 2000; Vicient et al., 2001; Wright & Voytas, 2002; Ustyantsev et al., 2015). Both features appear to have arisen independently in the Athila/Tat retrotransposons and retroviruses (Peterson-Burch et al., 2000; Ustyantsev et al., 2015). The Athila/Tat retrotransposons have often been referred to as ‘plant retroviruses’ (Peterson-Burch et al., 2000). Archaeplastida originated following primary endosymbiosis between a cyanobacterium and a heterotrophic eukaryote probably during the Paleoproterozoic Era (Furst-Jansen et al., 2020; Strassert et al., 2021). Archaeplastida comprises three major lineages, namely Rhodophyta, Glaucophyta and Chloroplastida (or Viridiplantae; including Prasinodermophyta, Chlorophyta and Streptophyta) (Furst-Jansen et al., 2020; Strassert et al., 2021). Streptophyta consists of Charophyta, a paraphyletic group of freshwater algae, and Embryophyta (land plants). Land plants originated from a charophyte progenitor that colonized terrestrial environments (plant terrestrialization) (Furst-Jansen et al., 2020). In this study, we explored the origin and evolution of the Athila/Tat retrotransposons in plants. First, to explore the distribution of the Athila/Tat retrotransposons within plants, we used a combined similarity search and phylogenetic analysis approach (see Supporting Information Methods S1 for details) to identify the Athila/Tat retrotransposons in a total of 106 representative plant species. These plant species cover the major diversity of plants, including seven angiosperms, four gymnosperms, two ferns, three lycophytes, four hornworts, two liverworts, three mosses, seven charophytes, 63 chlorophytes, one prasinodermophyte, one glaucophyte and nine rhodophytes (Fig. 1). We found the Athila/Tat retrotransposons are present in all the land plants screened in this study (Fig. 1; Table S1). However, we did not identify any Athila/Tat retrotransposon in Archaeplastida outside of land plants (Fig. 1). These results imply that the Athila/Tat elements might have originated after the split of land plants and charophytes when plants colonized terrestrial environments. The copy numbers of Athila/Tat elements vary among different land plant genomes, ranging from 67 (in Pleurozium schreberi) to 103 854 (in Ginkgo biloba) (Table S1). Note that these estimated copy numbers should be taken with caution, because transposable elements represent a challenging problem for genome assembly, and our approach only identified the retrotransposons with reverse transcriptase (RT) proteins. Nevertheless, the Athila/Tat retrotransposons might comprise a significant proportion of land plant genomes. To explore the potential source of land plant Athila/Tat retrotransposons, we used a combined similarity search and phylogenetic analysis approach to identify retrotransposons that are closely related to land plant Athila/Tat retrotransposons in the currently available genomes of eukaryotes outside plants (a total of 2095 genomes, including 53 Excavates, 1868 Amorphea, three Haptista, 168 TSAR (Telonemia, Stramenopiles, Alveolates and Rhizaria), and three Cryptista) (Fig. 1). We identified LTR retrotransposons closely related to the Athila/Tat elements in two oomycete species (Pseudoperonospora cubensis and Phytophthora lateralis) and nine fungus species within the phyla Mucoromycota (two species of Glomeromycotina and two species of Mortierellomycotina) and Zoopagomycota (three species of Zoopagomycotina and two species of Entomophthoromycotina) (Fig. 1; Tables S2, S3). To explore the evolutionary relationships of the Athila/Tat retrotransposons among land plants, oomycetes and fungi, we performed phylogenetic analyses based on the RT protein, a hallmark protein for analyzing the relationship among retrotransposons. Phylogenetic analyses showed that the Athila/Tat-like retrotransposons from oomycetes fall within the diversity of land plant Athila/Tat retrotransposons and form at least six distinct clusters (Fig. 2). All the oomycete Athila/Tat-like retrotransposons are placed at terminal nodes, and the Athila/Tat-like retrotransposons of different oomycetes do not cluster together, suggesting that these retrotransposons are unlikely to be circulating in oomycetes. Together, our phylogenetic analyses indicate that the Athila/Tat-related retrotransposons in oomycetes are likely to represent occasional sequencing contamination or to have arisen through independent recent horizontal transfer events from land plants to oomycetes. Importantly, we identified the presence of Athila/Tat-related retrotransposons in fungus species within the phyla Mucoromycota (early-branching terrestrial fungi) and Zoopagomycota (aquatic fungi). Phylogenetic analyses show that the Athila/Tat retrotransposons from land plants (and oomycetes) form a monophyletic group, and nest within the diversity of the Athila/Tat-related retrotransposons of fungi with strong support (Figs 2, S1). Our phylogenetic analyses suggest that the Athila/Tat retrotransposons in land plants might have originated through horizontal transfer from fungi to land plants. Moreover, phylogenetic analyses also show that land plant Athia/Tat retrotransposons are more closely related to retrotransposons of fungi from the phylum Mucoromycota (two species of Glomeromycotina and two species of Mortierellomycotina). Members of the Glomeromycotina are mutualistically associated with roots of vascular plants, forming arbuscular mycorrhizas (Strullu-Derrien et al., 2018; Berbee et al., 2020; Tedersoo et al., 2020). Our results suggest that plants might have acquired the Athila/Tat retrotransposons horizontally from fungi during the conquest of terrestrial environments. The Athila/Tat retrotransposons are known to exhibit diverse domain architectures and gene structures: some land plant Athila/Tat elements encode a dual RH domain, that is, they acquired cellular-like RH (archeal subtype RH, aRH) multiple times and the original RH domain (gRH) was retained but degenerated (Ustyantsev et al., 2015); and some land plant Athila/Tat retrotransposons acquired an additional env-like gene with transmembrane domain residing between the pol gene and 3′-LTR (Peterson-Burch et al., 2000, 2004; Vicient et al., 2001; Wright & Voytas, 2002; Ustyantsev et al., 2015). To explore the diversity and evolution of the Athila/Tat retrotransposon structures, we analyzed the domain architectures of representative complete or nearly complete Athila/Tat elements from land plants and fungi (Fig. S2). We found the Athila/Tat elements from fungi do not encode a dual RH domain or an env-like gene, representing the original Athila/Tat structure. By contrast, several independent land plant Athila/Tat lineages encode aRH domains, and the positions of aRH vary among different Athila/Tat lineages, indicating that aRHs in land plant Athila/Tat retrotransposons have multiple independent origins through domain shuffling (Ustyantsev et al., 2015). The Athila/Tat retrotransposons closely related to Athila4-1 that was found to encode an env-like gene (Wright & Voytas, 2002) contain a long region between LTR and the pol gene, which might encode an env-like gene. Moreover, we found at least two nonconventional domains in the Athila/Tat retrotransposons, namely Transposase_28 and universal minicircle sequence-binding protein (UMSBP) domains (Fig. S2; Ustyantsev et al., 2015). Our results suggest that the Athila/Tat retrotransposons frequently acquired and lost novel domains and evolved into diverse forms after horizontal transfer from fungi to land plants. In this study, we found that the Athila/Tat retrotransposons are ubiquitously distributed in land plants (Pelissier et al., 1995; Wright & Voytas, 1998, 2002; Vicient et al., 2001; Peterson-Burch et al., 2004; Ustyantsev et al., 2015). However, we did not identify any Athila/Tat retrotransposon in the genomes of Archaeplastida outside of land plants. Through a comprehensive screening of the eukaryote genomes, we identified retrotransposons that are closely related to land plant Athila/Tat retrotransposons in the genomes of fungi within the phyla Mucoromycota (two species of Glomeromycotina and two species of Mortierellomycotina) and Zoopagomycota (three species of Zoopagomycotina and two species of Entomophthoromycotina). Phylogenetic analyses show that land plant Athila/Tat retrotransposons cluster together and nest within the diversity of Athila/Tat retrotransposons from fungi. Therefore, we proposed that plants might have acquired the Athila/Tat retrotransposons horizontally from fungi during the conquest of land. However, there are five possibilities. (1) Fungus Athila/Tat retrotransposons might be generated by contamination of plant retrotransposons. However, we mined all the major lineages of plants, and all the fungus Athila/Tat retrotransposons fall outside the plant Athila/Tat diversity. Many fungus Athila/Tat retrotransposons are embedded within long contigs (Table S4). Moreover, we identified Athila/Tat retrotransposons within nine fungus species that were sequenced independently, and these fungus Athila/Tat retrotransposons do not cluster together but form a paraphyletic group. All these findings make the possibility of contamination highly unlikely. (2) The Athila/Tat retrotransposons were present in the common ancestor of plants and fungi, but lost in many eukaryote lineages. (3) Fungi or land plants acquired Athila/Tat retrotransposons multiple times from unknown sources. For possibilities (2) and (3), many protist lineages have been understudied, and it remains possible that Athila/Tat retrotransposons are present in eukaryotes outside fungi and plants. However, land plant Athila/Tat elements form a monophyletic group, and nest within the fungus Athila/Tat elements, making horizontal transfer from fungi to plants the most likely scenario. (4) Plants might have acquired the Athila/Tat retrotransposons long after they colonized terrestrial environments. If horizontal transfer occurred long after plant terrestrialization, the Athila/Tat retrotransposons are likely to be present in a certain range of plant species, and show a patchy distribution pattern in land plants. However, we found the Athila/Tat retrotransposons are ubiquitously present in land plants, weakening the possibility of horizonal transfer long after plant terrestrialization. (5) Horizontal transfer of the Athila/Tat retrotransposons occurred before plant terrestrialization, namely during the evolutionary course of charophytes. However, we mined seven charophyte genomes and did not identify the presence of the Athila/Tat retrotransposons. Moreover, this scenario requires many independent losses in charophytes, given charophytes are paraphyletic groups. Nevertheless, although the last possibility cannot be formally excluded, we think, based on the currently available evidence, the most likely evolutionary scenario is that plants might have acquired the Athila/Tat retrotransposons horizontally from fungi during (if not upon) the conquest of land. It has been widely speculated that fungi helped land plant progenitors colonize terrestrial environments (Pirozynski & Malloch, 1975; Selosse & Le Tacon, 1998; Delaux et al., 2015; Lutzoni et al., 2018; Strullu-Derrien et al., 2018). Indeed, charophytes, from which land plants originated, appear to encode some genes crucial for forming symbiotic associations with beneficial fungi (Delaux et al., 2015). Pirozynski & Malloch (1975) hypothesized that ‘terrestrial plants are the product of an ancient and continuing symbiosis of a semi-aquatic ancestral green alga and an aquatic fungus’. The discovery of arbuscules in Aglaophyton major, an Early Devonian land plant, provides fossil evidence for plant–fungus interactions more than 400 million years ago (Remy et al., 1994; Humphreys et al., 2010; Strullu-Derrien et al., 2018). Our results provide another layer of evidence for the ancient intimate associations between land plant precursors and fungi, and ancient land plant–fungus interactions might have shaped the evolution of the genome complexity of both partners. Indeed, traveling horizontally from fungi to plants, the Athila/Tat retrotransposons colonized and became a major lineage of retrotransposons in the genomes of land plants, contributing substantially to the genome complexity of land plants. Our study provides novel insights into the question of how diverse retrotransposons originated in land plants. Land plants evolved from a group of fresh water green algae, charophytes, more than 400 million years ago (Delwiche & Cooper, 2015; de Vries & Archibald, 2018; Han, 2019; One Thousand Plant Transcriptomes Initiative, 2019). Two possible scenarios of land plant retrotransposon origins could be conceived: retrotransposons in land plants might have originated from charophytes when plants colonized the terrestrial environment; or plants might have acquired retrotransposons horizontally from intimate associations with other organisms, for example, fungi and arthropods, after terrestrialization. In this study, we found that land plants acquired a major retrotransposon lineage through horizontal transfer from fungi. We speculate that retrotransposons in land plants might have originated multiple times, and the intimate association between land plants and other organisms after colonizing the land might have contributed to the diversity of retrotransposons in land plants. This work was supported by the National Natural Science Foundation of China (31922001). G-ZH conceived and designed the research. QW, YW, JW and ZG performed the research. QW, JW, ZG and G-ZH analyzed the data. G-ZH wrote the manuscript. QW, JW and ZG reviewed and revised the manuscript. All authors read and approved the final manuscript. No new data were generated in support of this research. Fig. S1 The phylogenetic relationships among representative Athila/Tat retrotransposons from plants, fungi and oomycetes, and representative retroelements. Fig. S2 The domain architectures of representative Athila/Tat retrotransposons from plants and fungi. Methods S1 Supplemental materials and methods. Table S1 Information on plant genomes with the presence of Athila/Tat retrotransposons. Table S2 Information on fungus genomes with the presence of Athila/Tat retrotransposons. Table S3 Information on oomycete genomes with the presence of Athila/Tat retrotransposons. Table S4 Information on the representative Athila/Tat retrotransposons identified in this study. Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.