- Research article
- Open Access
Mariner transposons are sailing in the genome of the blood-sucking bug Rhodnius prolixus
© Filée et al. 2015
- Received: 6 May 2015
- Accepted: 10 October 2015
- Published: 15 December 2015
The Triatomine bug Rhodnius prolixus is a vector of Trypanosoma cruzi, which causes the Chagas disease in Latin America. R. prolixus can also transfer transposable elements horizontally across a wide range of species. We have taken advantage of the availability of the 700 Mbp complete genome sequence of R. prolixus to study the dynamics of invasion and persistence of transposable elements in this species.
Using both library-based and de novo methods of transposon detection, we found less than 6 % of transposable elements in the R. prolixus genome, a relatively low percentage compared to other insect genomes with a similar genome size. DNA transposons are surprisingly abundant and elements belonging to the mariner family are by far the most preponderant components of the mobile part of this genome with 11,015 mariner transposons that could be clustered in 89 groups (75 % of the mobilome). Our analysis allowed the detection of a new mariner clade in the R. prolixus genome, that we called nosferatis. We demonstrated that a large diversity of mariner elements invaded the genome and expanded successfully over time via three main processes. (i) several families experienced recent and massive expansion, for example an explosive burst of a single mariner family led to the generation of more than 8000 copies. These recent expansion events explain the unusual prevalence of mariner transposons in the R. prolixus genome. Other families expanded via older bursts of transposition demonstrating the long lasting permissibility of mariner transposons in the R. prolixus genome. (ii) Many non-autonomous families generated by internal deletions were also identified. Interestingly, two non autonomous families were generated by atypical recombinations (5' part replacement with 3' part). (iii) at least 10 cases of horizontal transfers were found, supporting the idea that host/vector relationships played a pivotal role in the transmission and subsequent persistence of transposable elements in this genome.
These data provide a new insight into the evolution of transposons in the genomes of hematophagous insects and bring additional evidences that lateral exchanges of mobile genetics elements occur frequently in the R. prolixus genome.
- Transposable element (TE)
- Miniature inverted repeat transposable element (MITE)
- Horizontal transfer
The Triatominae blood-sucking bugs (Hemiptera, Reduviidae, Triatominae) are vectors of Trypanosoma cruzi (Kinetoplastida, Trypanosomatidae), the ethiologic agent of Chagas disease. Chagas disease is the most important parasitic disease in Latin America with 7 to 8 million affected people and is one of the most neglected diseases in the world (WHO, 2014). To date, about 140 species of Triatominae have been described into three main genera: Rhodnius, Triatoma, and Panstrongylus. Recently, the genome of R. prolixus has been sequenced (www.vectorbase.org). The availability of high throughput sequencing data has refined our understanding of functional genomics and gene expression and also the identification of adaptation mechanisms that may involve structural variations including gene duplication or transposition of mobile elements . In addition R. prolixus are suspected to transmit transposable elements (TE) horizontally across phyla . TEs, which represent an important part of eukaryotic genomes, play important roles in genome size, genome adaptability, and genome structure and functions [3, 4]. At the gene level, they can trigger dramatic gene inactivation or temperate regulation changes. TEs are usually silent but can occasionally reactivate under environmental changes, notably through epigenetic changes affecting TE copies [5–7]. Hence this reactivation may lead to transposition burst, which will increase (through transposition or recombination) adaptability, genetic diversity, and probability to create beneficial/adaptive alleles . However, TEs have to undergo frequent horizontal transfers (HTs) between different species to avoid stochastic losses . A growing number of cases of TEs HT have been reported in the literature but their underlying mechanisms are still unknown . It has been shown that four TE families in the genome of R. prolixus are almost identical to mammalian TEs . These data support the existence of recent HTs of diverse TEs between this species and their mammalians hosts. They may also indicate that this haematophagous bug plays a pivotal role in the transmission of TEs across a wide range of species. Recently, six additional MITEs almost identical between R. prolixus and the silkworm Bombyx mori have been evidenced . Taken together these data suggest that R. prolixus is an interesting model to document the evolutionary dynamics of TEs, notably the role played by the host/parasite interactions in the mechanism of HT events of transposons.
In this paper we explored the complete genome of R. prolixus for transposons and their non-autonomous derivatives using a combination of library-based and de novo methods. We found that TE derived sequences compose 5.8 % of the Rhodnius genome, a relatively modest contribution in comparison to other insect genomes. But DNA transposons are surprisingly abundant and especially a very large diversity of mariner families accounts for two third of these TEs. We demonstrate that the dominance of mariner-like transposons is the result of recent and older burst events in addition to more continuous expansion of other families. The ongoing invasion of mariner elements is also associated with multiple generations of non-autonomous derivatives that have subsequently expanded. Finally, the identification of several HTs sharing with various species suggests the existence of horizontal transfers of TEs which participated to the recurrent invasion of the R. prolixus genome by exogenous mariner transposons.
Data collection and availability
Rhodnius prolixus assembled genomic sequences (RproC1) were downloaded from VectorBase (htps://www.vectorbase.org/organisms/rhodnius-prolixus). We analyzed TEs in the whole genome using RepeatMasker with default parameters (http://www.repeatmasker.org) and a library of Metazoan TEs extracted from Repbase (http://www.girinst.org/repbase/).
Python scripts and raw data including TE sequences, consensus, alignments and phylogenetic trees… are available at: http://echange.legs.cnrs-gif.fr:5000/fbsharing/LUGs8EBq
Library based method for Tc1-mariner Element searches
Reconstitution of copies by associating hit distant of less than 1000 bp, in correct orientation
Filtering out any copies less than 400 bp-long
Extraction of all the sequences with or without 500 bp flanking sequences each side to get full copies
Clustering copies (without flanking sequences) with Usearch (−id 0.8, −rev) 
Filtering out sequences with “N”, assembly-truncated copies, and duplicated copies (resulting from segmental duplication and not from transposition, as determined by the flanking sequences.
Trimming flanking sequences and generating nucleotide consensus (majority rule with keeping the longest elements), then protein consensus
De novo identification of MITEs
We used a suite of python scripts gathered under the name AutoMitaur (Hua-Van, unpublished) and available at http://www.egce.cnrs-gif.fr/wp-content/uploads/2014/04/AutomitAur.v1.0.1.zip.
Briefly, in this suite of script, BLASTN is used to compare a genome against itself for short hits at least 11 bp-long, distant of 750 bp at most, and in inverted orientation (TIRs). The TIRs, the intervening sequence plus 60 bp flanking sequences on each side are then extracted. Sequences are then clustered and copies with similar flanking sequences are removed. Several filters are applied and only groups with at least ten independent sequences that reach a certain level of homogeneity between the sequences and display bona fide TIRs are kept. A consensus sequence is then determined for each cluster. The pipeline also includes a step consisting of searching (BLASTN-SHORT) for putative autonomous partners, by using the defined TIR sequences as queries against the input genome, keeping only sequences larger than 1 kb. The putative longer elements are then searched against the RepBase protein database (31/01/2014 version) using BLASTX, to automatically identify the potential associated super-family. In parallel, a BLASTX search was realized with the MITE consensus sequences as a query, against the database.
Out of a raw output of 107 clusters, we could then select 41 MITE clusters for further analysis.
TE Classification and phylogenetic analyses
We classified clusters of the Tc1-mariner-IS630 super-family to define homogeneous groups. This computation is based on the UPGM-VM method, an ascending hierarchical classification analogous to the classical UPGMA, with two main differences: 1) there is no arithmetical mean, the sequences are aligned two-per-two and the corresponding distances are computed; 2) the metric varies with the ascending classification. At the beginning, an alignment gap is considered as a fifth nucleotide, and its weight is progressively and rapidly set to zero. This variation of the metric allows gathering in the same group a complete sequence and the corresponding truncated or deleted sequences such as MITEs .
R. prolixus elements found in this study were added to a set of 309 complete sequences previously published in GenBank and representatives of the main clades of the Tc1-mariner-IS630 SuperFamily : mariner (Briggsae, Cecropia, Elegans, Irritans, Mellifera, Mauritiana, Vertumnana), maT (mori), Tc1, Tc2, Tc3, Tc4, Tc5, Tc6, Gambol, Pogo, Fot, Lemi, Plant mariner, Impala, IS630, IS870. We added the 36 Drosophila sequences described by Wallau et al.  and the consensus sequences found here in R. prolixus.
For the phylogenetic analysis we used a representative set of mariner transposase from Repbase covering all the known clades or lineages of the super-family [11, 16, 18]. Sequences were aligned using MUSCLE with default parameters and conserved parts of the alignments usable for phylogenetic analyses were chosen using Gblocks [18, 19]. The best-fitting ML model was selected using Protest and the tree was computed using PhyML 3.0 . Branch supports were calculated using a LRT Shimodaira-Hasegawa (SH) procedure.
We compared R. prolixus mariner consensus sequences to Genbank and WGS NCBI databases (ftp://ftp.ncbi.nlm.nih.gov/) using BLASTN searches . Candidate elements for HT were identified as sequences with more than 75 % of nucleotide identity over more than 90 % of the query sequences. To discard potential cases of contamination with foreign DNA, each genomic context of the putative elements was carefully examined: each 50 kbp adjacent segment was inspected with a BLASTN procedure and only elements within a conserved synteny block were conserved. Cases of HTs were then validated using phylogenetic analyses.
TE amplification dynamics
We inferred species-specific amplification dynamics of single lineages using a new method based on the phylogenetic tree node distributions over time. This method relies on the topology of the phylogenetic tree and offers a visualization of the variation in transposition rate per copy over time. More details are available in Le Rouzic et al. .
Tc1-mariner elements dominate the mobilome of R. prolixus
Large diversity of mariner elements in the R. prolixus genome
In order to identify the different Tc1-mariner transposable elements, we used a homology-based approach (TBLASTN), starting with two sets of transposases, one composed of eight mariner transposases representing the major mariner subfamilies [16, 17], the other set comprising fifteen transposases belonging to other Tc1-like families (classified according to the catalytic domain as in ) (Additional file 1: Table S1). The mariner search retrieved a total of 11,015 copies that could be clustered in 89 groups of copies with similarities higher than 80 % and that likely represent functional lineages (i.e., copies within one lineage can cross-mobilize copies from the same lineage, due to high sequence similarity, usually over 80 %). On the opposite, the non-mariner search retrieved only 502 copies, clustered in 52 groups (Additional file 1: Table S1). This revealed that the large domination of the Tc1-mariner-IS630 elements in R. prolixus is mainly due to elements of the mariner family (characterized by a DD(34)D catalytic domain) both at the abundance and the diversity levels and we subsequently focused on this family.
Characteristics of mariner lineages identified in the Rhodnius prolixus genome. Column “Clean Independent Copy Number” reports the number of copies not truncated by “N” and corresponding to true transposition events (different flanking sequences). Column “Potentially Active Copies” indicates if at least one complete ORF (>1000 bp) has been found among copies
Total Copy Number
Clean Independent Copy Number
Potentially Active Copies
Putative Horizontal Transfer
Putative Horizontal Transfer
Putative Horizontal Transfer
Putative Horizontal Transfer
Putative Horizontal Transfer
Putative Horizontal Transfer
Putative Horizontal Transfer
Putative Horizontal Transfer
The initial 11,015 sequences, consisting only of sequences exhibiting homology with transposase sequences, covered about 7 Mb of the genome, mainly due to the 32 lineages. By comparison the 503 non-mariner Tc1-like copies covered only 0.35 Mb. However, when the full nucleotide consensuses derived from the 32 mariner lineages were used as seeds in a RepeatMasker search, 26.4 Mb were masked, slightly more than the initial search using RepBase as the seed library (24.5 Mb). Then, our TBLASTN methodology based on transposases is not fully exhaustive since it did not allow the recovery of all mariner sequences including degenerated or highly divergent copies. The most probable explanation is that a large amount of mariner fragments, lacking ORF sequences, or shorter than 400 bp (our filtering threshold) exist in the R. prolixus genome. For example, the Rpmar63 encompasses 153 identifiable sequences with our pipeline (Table 1) but a BLASTN with the consensus sequence identify 580 additional short and fragmented sequences. Another problem is the level of assembly quality of the genome. Indeed, the 55,000 contigs include a large proportion of small contigs (only 13 % of them are bigger than 10,000 bp). That may prevent the recovery of long-enough copies, and ultimately makes impossible a precise estimation of the amount of repeated sequences (which often corresponds to unmapped small contigs). Nevertheless, and although both methods are homology-based, our TBLASTN-based method appears more efficient than the RepeatMasker/RepBase strategy, that likely underestimates the amount of repeated sequences, probably due to high divergence between the sequences in the library and the elements in the genome.
Besides these methodological limitations, two facts still account for the exceptional situation encountered in the R. prolixus genome regarding the mariner elements. The first is that the huge amount of mariner sequences is mainly due to one single lineage (Rpmar0) comprising more than 8000 copies (73 % of all mariner elements). Furthermore, seven other lineages display more than 100 copies. Mariner is described as a low copy number family, although high copy number lineages have occasionally been described in some species (see for example ). In a recent analysis of 20 Drosophila genomes  the most prolific mariner lineage exhibited about 500 copies in one genome, most of the other consisting of less than 50 copies/per genome and usually less than 10. The R. prolixus genome appears then rather permissive for mariner amplification, for reasons that still remain to be deciphered.
The second peculiarity in this genome is the huge diversity of mariner elements. 89 different clusters (suggesting about the same number of functional lineages) have been identified. Even by considering only those with at least 5 copies, it is still more than 30 different lineages coexisting in the very same genome, just a few less than in the recently analyzed 20 Drosophila genomes, taken as a whole. Indeed, no more of 23 lineages > 5 independent copies have been identified within one single Drosophila genome . The R. prolixus genome then appears so far the most comfortable ecological niche for mariner elements.
All these lineages fully covered the known mariner diversity and possibly formed at least one new subfamily. We first performed a classification of the R. prolixus nucleotide sequences using a clustering method (UPGM-VM) based on the whole nucleic sequences of 309 Tc1-mariner sequences. This classification allows the use of a large dataset in a reasonable calculation time, including distantly related Tc1 and Tc3 sequences found in animals, plants, fungi and bacteria (Additional file 2: Figure S1). The resulting classification revealed the clustering of R. prolixus sequences within known clades/subfamilies with the exception of four lineages that may define a new subfamily called nosferatis (Nos in Additional file 2: Figure S1).
The typical mariner size is between 1280 and 1350 bp, which is supported by the size of most of the consensus sequences reported in Table 1. Among the 32 lineages analyzed, we found only 9 lineages with at least one full-length copy with an uninterrupted ORF that could witness recent potential activity. Furthermore, we could identify ten lineages for which the consensus sequence (constructed in a way to fit the most complete element) is between 800 bp and 1000 bp-long, meaning that these lineages are only made of shorter elements and then obviously represent non-autonomous lineages. It is noteworthy that six of these lineages belong to the subfamily drosophila, already known to easily generate such kind of deleted lineages in the 20 Drosophila genomes ).
Disregarding the fact that these lineages have kept a reasonable size, they could represent lineages on the way of becoming MITEs (Miniature Inverted-repeat Transposable Elements), that amplify using the transposase of other closely related lineages that share almost identical TIR sequences. MITEs are usually present in high copy number, and supposed to derive from full-length lineages by successive shortening of the internal part, combined with elevated sequence degeneracy, and in some cases rearrangement, while keeping the ability to be mobilized .
One example of ongoing “MITEzation” is provided when comparing one of these shorter lineages (Rpmar35), which is actually directly derived from the dominant Rpmar0 lineage by internal deletion. Yet, Rpmar35 is mainly composed of 2 sets of shorter sequences similar to Rpmar0, and having obviously transposed after internal deletion in the transposase sequence of a Rpmar0 copy.
All the Rpmar17 sequences (except two that correspond to near full-length copies) seemed to have experienced the same kind of rearrangement (5’ part replacement with 3’ part), as for Rpmar49. A striking difference is however that few copies exhibit identical recombination breakpoints, as shown from a subset of complete sequences that we could easily align (Fig. 3b). All the breakpoints seem however localized in the same region. This case is at first glance puzzling, but can actually gives insights on possible initial events responsible for this lineage. One hypothesis is that all these different sequences, made of a transposed 3’ part having replaced the 5’ part, could result from an initial head-to-tail mariner dimer or close copies. From it, a shorter element would have arisen by internal deletion, leaving only extremities made of 3’ parts. After (or during) amplification of this progenitor sequence, resulting copies would have suffered new independent deletions all localized around the hypothetical initial breakpoint. This hypothesis suggests then two unrelated process (the first rearrangement/deletion event and the subsequent deletions centered around the breakpoints.
We performed an additional analysis relying only on the position of the breakpoints relative to the non-rearranged full-length copy, avoiding the problematic step of aligning. A similar pattern was observed for 227 independent rearranged copies, including the variable position of the breakpoints (Fig. 3c).
We noticed that the longest copy that could represent the initial deleted progenitor is more than 1700 bp long. Curiously all the other rearranged copies but one are less than 1000 bp long, with the majority between 900–950 bp (Fig. 3d and e). Element size, as well as internal structure, can influence the transposition efficiency [31, 33, 34]. However, in our case, successful transposition is not observed since most copies exhibit different breakpoints: they are probably not derived from each other by transposition. Hence the size homogeneity is not the result of selection for transposition ability and the observed necessity for a certain size range is difficult to understand here. Alternatively, the apparent propensity to obtain 950 bp copies after deletion could result from structural particularities in the breakpoints regions. For example, these regions could be hotspots for double strand breaks repair , or prone to be joined together during abortive gap repair. Indeed, it was already shown that deletions are not totally random in transposable elements and may depends on sequence characteristics .
R. prolixus mariner elements generate a limited set of MITEs smaller than 900 bp
These few examples described before illustrate the fact that mariner transposons can generate shorter lineages that are able to amplify, although no lineages shorter than 900 bp could be identified. Since MITEs are usually shorter and are sometimes related to autonomous elements only by very short sequences corresponding to TIRs with or without subterminal sequences, they can totally lack any similarities with coding sequences (ORF), and then cannot been retrieved with our method . In order to complete the mariner landscape, we then used a de novo approach based on the presence of short inverted repeats less than 750 bp apart; we retrieved 107 clusters of potential MITEs with at least 10 copies. 33 of them were found to be potentially inserted in a TA TSD, among which the six more abundant (more than 100 copies). For each cluster, a search for longer elements bordered by the same TIRs was run and longer copies blasted against the protein repbase to detect homology with transposase. The same was also carried out using representative or consensus sequences of the different lineages. We then selected 41 clusters meeting one of these criteria (TSD with TA, or Tc1-mariner transposase homology), for further analysis and manual inspection. However, very few families could be confirmed to be Tc1-mariner MITEs. Indeed, for some of them, no similarity to Tc1-mariner sequences could be found, in internal part or within the TIRs. For some other elements the TSD was determined to be larger than just the typical TA observed for Tc1-mariner, suggesting these elements could belong to other super-families (CACTA, P, hAT, piggyBac…). For ten MITE lineages, no clear TSD and TIRs could be defined, weak homology often extending outside the putative limits of the elements. Finally, for several MITE lineages, the homology found in longer element sharing the same TIRs was due to the nested insertion of a Tc1-mariner element in a non- Tc1-mariner MITE.
List of MITE clusters that belong to the Tc1-mariner superfamily. Only clusters with at least one sublineage may represent bona fide MITEs. (a) independent copy number (b) minimum and maximum size are given
Copy number (a)
Partner in Rhodnius genome
TSD and TIRs
We also could detect several clusters that are probably related to Tc-like or pogo/tigger–like elements as well as a prokaryotic IS630 elements. The latter element could originate from endosymbiotic bacteria that are abundant and diverse in Rhodnius species . A contamination with foreign bacterial DNA is also possible.
The MITE_109 comprised 3 sub-lineages that share similar ends but have different breakpoints. Homology with Tigger elements were detected in a region common to the two less abundant sub-lineages, but no longer element that could correspond to the progenitor exists in the Rhodnius genome (Fig. 4b).
The MITE_100 bona fide MITE is composed of 68 independent copies that display high homogeneity in size and sequence. The internal part of the element presents homology with Tc1-like elements, although we again could not find any related longer element in the genome (Fig. 4c).
MITE_120 comprised 174 independent sequences presenting homology on the main part of their external sequences. Three sub-lineages can be recognized, but concern only one third of the copies. Although clearly related, the others seemed to result from independent internal deletions from a larger element (Fig. 4d). Like for the autonomous Rpmar17 lineage, and unlike the MITE_9 previously described, it is possible to locate a potential unique breakpoint in this largest element. All deletions in the other copies are centered on this position including the 3 sub-lineages that experienced further transposition. However, this largest element is non-coding and presents no homology with any known proteins, so the autonomous partner responsible for transposition is still unidentified. Nevertheless, the TIRs sequences including a potential TA dinucleotide TSD resembles that of MITE_100 and MITE_109, suggesting that this element lineage belongs to the Tc1-mariner super-family (Table 2).
MITE_51 present a pattern similar to the MITE_120 pattern, i.e. two sub-lineages but in which most copies have suffered independent deletions, as well than a probable breakpoint at the origin of all the copies (Fig. 4e). Like for MITE_120, no homology with any proteins could be detected, the relationship with Tc1-mariner superfamily being only supported by the TSD and TIR sequences (Table 2).
Globally, it seems that Tc1-mariner, and especially mariner lineages are not prone to generate short MITE families. However, the fine analysis of mariner and the few MITE families raise interesting questions. For mariner the search for MITE smaller than about 800 bp was rather unfruitful. If short mariner MITEs exist, they are obviously in very low copy number, so not quite prone to amplification. In contrast, an important proportion of the mariner lineages identified correspond to shortened non-autonomous lineages usually 800–900 bp long. Altogether, this suggests than mariner elements are prone to deletion but the ability to transpose is likely highly constrained, by a minimum size about 800 bp, preventing the amplification or short copies and then the generation of MITE families. Noteworthy, several other mariner non-autonomous lineages have been detected in the drosophila genomes, most of them exhibiting a size of about 950 bp, supporting the hypothesis of a size constraint .
Dynamic of mariner transposons in the R. prolixus genome
The first pattern is a “S-shaped” curve, which reflects the fact that transposition started with a very slow rate, then the rate increased before slowing down progressively. This pattern can be interpreted as a transposition rate that is dependent of copy number at the beginning. At the end of the amplification period, the slowing down may be due to the progressive loss of active copy (inactivation), or the establishment of regulations. In such a dynamics, the median transposition event is roughly located at mid-course of the amplification time-span. This dynamics is observed for older lineages but also in recent ones, such as Rpmar1.
A second type of dynamics is referred as “Exponential”, and is compatible with a model in which the transposition rate per copy is constant, meaning the more copies the more transposition events. This is expected for the beginning of the amplification (before establishment of regulations), or for active lineages (undergoing amplification), for example in Rpmar11. Rpmar12 also display this dynamics, although it is now inactive, which indicates that the transposition suddenly stopped after the initial transposition burst, maybe due by the rapid loss of all active copies. The median is then shifted to the recent time.
The third dynamics is described as “Linear”, because the transposition rate seems to be constant over time, until it falls rapidly to zero. In this case, it is independent of the copy number. Rpmar17 and Rpmar5 follow this dynamics, characterized by a median centered in the middle of the amplification time-span.
Finally the fourth dynamics is called “Concave” and is characterized by a high transposition rate at the beginning, followed by a progressive slowing down. The median is the shifted to ancient time, and several recent or middle-aged lineages present this dynamics.
This comparative analysis revealed that very different dynamics characterize closely related TE lineages that coexist in the same genome at the same time. These differences can be explained by the intrinsic biochemical properties of the element , or the establishment of specific regulations, through epigenetic silencing, or through cell cycle-coupled controls [38–40]. It should be noticed that methodological biases exist since the method relies on a reconstructed phylogeny based on extant copies, and only duplicative transposition are scored. The resulting dynamics can also be modified by variable deletion rates. However, considering that the same genomic deletion rate will apply for coexisting lineages, we suspect that it cannot be responsible for the dynamics differences we observed between lineages.
Globally, it appears that the R. prolixus genome is recurrently and frequently invaded by mariner elements. Mariner elements seem to escape easily transposition controls since huge copy number are observed for several lineages. In particular the three most abundant mariner lineages (8,041, 767 and 488 total copies) are also the most recent ones. The high level of amplification is not compensated by a high turnover, i.e., the rapid disappearance of older lineages, as shown by the large diversity of mariner lineages and the high copy number in old lineages too.
Evidence of multiple HT of Mariner elements
We screened the complete GenBank and WGS databases with the mariner and MITE mariner consensus in order to document the propensity of mariner TEs in R. prolixus to generate HTs.
List of the different HTs of mariner elements found in the R. prolixus genome
First BLAST Hit
Schmidtea mediterranea (flatworm)
Dendroctonus ponderosae (beetle)
Hymenolepis diminuta (tapeworm)
Haemonchus placei (nematode)
Scolia oculata (parasitic wasp)
Bombus terrestris (bumblebee)
Glossina pallipes (tsetse fly)
Artibeus jamaicensis (bat)
Strongyloides stercoralis (nematode)
The nine other putative cases of HT concerned mariner autonomous transposons. They involved five other insects (Dendroctonus ponderosae, Scolia oculata, Bombus terrestris, Glossina pallipes, and Drosophila sp.), one south American bat (Artibeus jamaicensis), two blood sucker nematodes of mammals (Haemonchus placei and Strongyloides stercoralis) and one tapeworm (Hymenolepis diminuta) that parasites various insects and mammals. Phylogenetic analysis of the transposase of each element confirms the close proximity between these elements and the R. prolixus transposons (Fig. 2). Concerning putative TE transfer between R. prolixus and the four other insects, despite high level of sequence conservation between these elements and the R. prolixus mariner transposons (up to 93 %), the long period of divergence since the split between Hemiptera and Diptera/Hymenoptera (>300 Ma ) is incompatible with a vertical inheritance. For horizontal transfers two scenario of transmission could be examined: direct transmission or indirect via intermediate hosts. Interestingly, the implication of parasitoid insects as vector of HT of hAT and Ginger MITEs between R. prolixus and the silkworm B. mori has been proposed . A similar situation has been reported between R. prolixus and the twisted wing parasite Mengenilla moldrzyki that are known to infect a large variety of insects . In our dataset, we have detected a possible HT between the parasitic wasp S. oculata and R. prolixus. Eggs of Triatomine bugs as Rhodnius species are effectively infected by diverse parasitic wasps . Another example of HT of a mariner element between the parasitic wasp Ascogaster reticulatus and its host the moth Adoxyphyes honmai has also been evidenced . Since the implication of insect parasites as intermediate vectors seems to be plausible, this mechanism could be considered for the sharing of very closely related mariner transposons between R. prolixus and Drosophila sp., B. terrestris and G. pallipes. In the case of the parasite tapeworm H. diminuta, both the strong transposon sequence conservation between this tapeworn and R. prolixus and the ecology of this organism that live as a parasite of various insects, are arguments in favor a recent and direct HT within R. prolixus. Moreover, a direct HT between the South American bat A. jamaicensis and R. prolixus is possible, since R. prolixus is known to feed on bats blood . Interestingly, HTs of mariner elements have been evidenced between various insects and mammals . Concerning the two species of blood feeding nematodes of mammals, as R. prolixus infects the same range of hosts, we cannot ruled out the hypothesis of independent HTs of the transposons from a common but unknown mammalian host.
Taken together our data indicate the existence of frequent HTs of mariner transposons between R. prolixus and a large variety of organisms. In addition, keeping in mind the recent data supporting the existence of other transposon HTs with mammals [2, 47] and insects [10, 48], our analyses demonstrate the existence of a diverse horizontal flux of transposons in the genome of R. prolixus. By providing new invading elements, we can hypothesize that this flux balances the inevitable stochastic losses of mariner elements and thus participate to the strong preponderance of this super-families in the R. prolixus genome.
a long-lasting permissibility of the genome for mariner that leads to lineages with huge copy number. Copy number explosion is especially striking for recent still active mariner lineages, but is also observed in very old lineages supposed to progressively loose copies. A recent burst of a single mariner lineage has led to the generation of more than 8000 copies (two third of the total mariner elements present in the genome);
a huge diversity of mariner lineages that was never observed before since between 32 and 89 different lineages were recovered. These lineages are usually well delimited and reflect the diversity of mariner within the whole metazoan clade, since lineages from most mariner subfamilies could be identified.
frequent occurrence of HT of mariner elements within various species including other insects in particular parasitoids, hematophagous nematodes, parasite worms and a South American bat.
Finally, this huge dataset of copies has revealed some aspect of the biology of mariner elements, for example, the generation of shorter lineages that seems to be highly constrained by size, the fact that these shorter lineages are frequent within some subfamilies only, that rearranged lineages can also arise by 5’ replacement with 3’part. We believe that the data and interpretations provided here will offer a basis to future study aiming to understand the role play by transposable elements during evolution and the adaptation to human of Triatomine bugs.
The authors thank Jean-Michel Rossignol and Nicolas Pollet for critical reading of the manuscript and Arnaud Le Rouzic for helpful discussion. This work is supported by the Agence Nationale de la Recherche (Adaptanthrop project ANR-09-PEXT-009) and the University Paris-Sud (IDEEV grants).
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Stapley J, Reger J, Feulner PG, Smadja C, Galindo J, Ekblom R, et al. Adaptation genomics: the next generation. Trends Ecol Evol. 2010;25(12):705–12.View ArticlePubMedGoogle Scholar
- Gilbert C, Schaack S, Pace 2nd JK, Brindley PJ, Feschotte C. A role for host-parasite interactions in the horizontal transfer of transposons across phyla. Nature. 2010;464(7293):1347–50.PubMed CentralView ArticlePubMedGoogle Scholar
- Fedoroff NV. Presidential address. Transposable elements, epigenetics, and genome evolution. Science. 2012;338(6108):758–67.View ArticlePubMedGoogle Scholar
- Hua-Van A, Le Rouzic A, Boutin TS, Filee J, Capy P. The struggle for life of the genome's selfish architects. Biol Direct. 2011;6:19.PubMed CentralView ArticlePubMedGoogle Scholar
- Grandbastien MA, Audeon C, Bonnivard E, Casacuberta JM, Chalhoub B, Costa AP, et al. Stress activation and genomic impact of Tnt1 retrotransposons in Solanaceae. Cytogenet Genome Res. 2005;110(1–4):229–41.View ArticlePubMedGoogle Scholar
- Capy P, Gasperi G, Biemont C, Bazin C. Stress and transposable elements: co-evolution or useful parasites? Heredity (Edinb). 2000;85(Pt 2):101–6.View ArticleGoogle Scholar
- He F, Zhang X, Hu JY, Turck F, Dong X, Goebel U, et al. Widespread interspecific divergence in cis-regulation of transposable elements in the Arabidopsis genus. Mol Biol Evol. 2012;29(3):1081–91.View ArticlePubMedGoogle Scholar
- Zeh DW, Zeh JA, Ishida Y. Transposable elements and an epigenetic basis for punctuated equilibria. Bioessays. 2009;31(7):715–26.View ArticlePubMedGoogle Scholar
- Schaack S, Gilbert C, Feschotte C. Promiscuous DNA: horizontal transfer of transposable elements and why it matters for eukaryotic evolution. Trends Ecol Evol. 2010;25(9):537–46.PubMed CentralView ArticlePubMedGoogle Scholar
- Zhang HH, Xu HE, Shen YH, Han MJ, Zhang Z. The origin and evolution of six miniature inverted-repeat transposable elements in Bombyx mori and Rhodnius prolixus. Genome Biol Evol. 2013;5(11):2020–31.PubMed CentralView ArticlePubMedGoogle Scholar
- Jurka J. Repbase update: a database and an electronic journal of repetitive elements. Trends Genet. 2000;16(9):418–20.View ArticlePubMedGoogle Scholar
- Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215(3):403–10.View ArticlePubMedGoogle Scholar
- Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26(19):2460–1.View ArticlePubMedGoogle Scholar
- Katoh K, Standley DM. MAFFT: iterative refinement and additional methods. Methods Mol Biol. 2014;1079:131–46.View ArticlePubMedGoogle Scholar
- Larsson A. AliView: a fast and lightweight alignment viewer and editor for large datasets. Bioinformatics. 2014;30(22):3276–8.PubMed CentralView ArticlePubMedGoogle Scholar
- Rouault JD, Casse N, Chenais B, Hua-Van A, Filee J, Capy P. Automatic classification within families of transposable elements: application to the mariner Family. Gene. 2009;448(2):227–32.View ArticlePubMedGoogle Scholar
- Wallau GL, Capy P, Loreto E, Hua-Van A. Genomic landscape and evolutionary dynamics of mariner transposable elements within the Drosophila genus. BMC Genomics. 2014;15:727.PubMed CentralView ArticlePubMedGoogle Scholar
- Edgar RC. MUSCLE: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinformatics. 2004;5:113.PubMed CentralView ArticlePubMedGoogle Scholar
- Castresana J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol. 2000;17(4):540–52.View ArticlePubMedGoogle Scholar
- Guindon S, Delsuc F, Dufayard JF, Gascuel O. Estimating maximum likelihood phylogenies with PhyML. Methods Mol Biol. 2009;537:113–37.View ArticlePubMedGoogle Scholar
- Le Rouzic A, Payen T, Hua-Van A. Reconstructing the evolutionary history of transposable elements. Genome Biol Evol. 2013;5(1):77–86.PubMed CentralView ArticlePubMedGoogle Scholar
- Xu HE, Zhang HH, Xia T, Han MJ, Shen YH, Zhang Z. BmTEdb: a collective database of transposable elements in the silkworm genome. Oxford: Database; 2013. 2013:bat055.Google Scholar
- Agren JA, Wright SI. Co-evolution between transposable elements and their hosts: a major factor in genome size evolution? Chromosome Res. 2011;19(6):777–86.View ArticlePubMedGoogle Scholar
- Wang S, Lorenzen MD, Beeman RW, Brown SJ. Analysis of repetitive DNA distribution patterns in the Tribolium castaneum genome. Genome Biol. 2008;9(3):R61.PubMed CentralView ArticlePubMedGoogle Scholar
- Elsik CG, Worley KC, Bennett AK, Beye M, Camara F, Childers CP, et al. Finding the missing honey bee genes: lessons learned from a genome upgrade. BMC Genomics. 2014;15:86.PubMed CentralView ArticlePubMedGoogle Scholar
- Fernandez-Medina RD, Ribeiro JM, Carareto CM, Velasque L, Struchiner CJ. Losing identity: structural diversity of transposable elements belonging to different classes in the genome of Anopheles gambiae. BMC Genomics. 2012;13:272.PubMed CentralView ArticlePubMedGoogle Scholar
- Clark AG, Eisen MB, Smith DR, Bergman CM, Oliver B, Markow TA, et al. Evolution of genes and genomes on the Drosophila phylogeny. Nature. 2007;450(7167):203–18.View ArticlePubMedGoogle Scholar
- Shao H, Tu Z. Expanding the diversity of the IS630-Tc1-mariner superfamily: discovery of a unique DD37E transposon and reclassification of the DD37D and DD39D transposons. Genetics. 2001;159(3):1103–15.PubMed CentralPubMedGoogle Scholar
- Bigot Y, Brillet B, Auge-Gouillou C. Conservation of Palindromic and Mirror Motifs within Inverted Terminal Repeats of mariner-like Elements. J Mol Biol. 2005;351(1):108–16.View ArticlePubMedGoogle Scholar
- Garcia-Fernandez J, Bayascas-Ramirez JR, Marfany G, Munoz-Marmol AM, Casali A, Baguna J, et al. High copy number of highly similar mariner-like transposons in planarian (Platyhelminthe): evidence for a trans-phyla horizontal transfer. Mol Biol Evol. 1995;12(3):421–31.PubMedGoogle Scholar
- Jiang N, Feschotte C, Zhang X, Wessler SR. Using rice to understand the origin and amplification of miniature inverted repeat transposable elements (MITEs). Curr Opin Plant Biol. 2004;7(2):115–9.View ArticlePubMedGoogle Scholar
- Rubin E, Levy AA. Abortive gap repair: underlying mechanism for Ds element formation. Mol Cell Biol. 1997;17(11):6294–302.PubMed CentralView ArticlePubMedGoogle Scholar
- Lohe AR, Hartl DL. Efficient mobilization of mariner in vivo requires multiple internal sequences. Genetics. 2002;160(2):519–26.PubMed CentralPubMedGoogle Scholar
- Lozovsky ER, Nurminsky D, Wimmer EA, Hartl DL. Unexpected stability of mariner transgenes in Drosophila. Genetics. 2002;160(2):527–35.PubMed CentralPubMedGoogle Scholar
- Brunet F, Giraud T, Godin F, Capy P. Do deletions of Mos1-like elements occur randomly in the Drosophilidae family? J Mol Evol. 2002;54(2):227–34.View ArticlePubMedGoogle Scholar
- da Mota FF, Marinho LP, Moreira CJ, Lima MM, Mello CB, Garcia ES, et al. Cultivation-independent methods reveal differences among bacterial gut microbiota in triatomine vectors of Chagas disease. PLoS Negl Trop Dis. 2012;6(5), e1631.PubMed CentralView ArticlePubMedGoogle Scholar
- Bouuaert CC, Tellier M, Chalmers R. One to rule them all: A highly conserved motif in mariner transposase controls multiple steps of transposition. Mob Genet Elements. 2014;4(1), e28807.PubMed CentralView ArticlePubMedGoogle Scholar
- Spradling AC, Bellen HJ, Hoskins RA. Drosophila P elements preferentially transpose to replication origins. Proc Natl Acad Sci U S A. 2011;108(38):15948–53.PubMed CentralView ArticlePubMedGoogle Scholar
- Ton-Hoang B, Pasternak C, Siguier P, Guynet C, Hickman AB, Dyda F, et al. Single-stranded DNA transposition is coupled to host replication. Cell. 2010;142(3):398–408.View ArticlePubMedGoogle Scholar
- Dufourt J, Vaury C. During a short window of Drosophila oogenesis, piRNA biogenesis may be boosted and mobilization of transposable elements allowed. Front Genet. 2014;5:385.PubMed CentralView ArticlePubMedGoogle Scholar
- Hedges SB, Dudley J, Kumar S. TimeTree: a public knowledge-base of divergence times among organisms. Bioinformatics. 2006;22(23):2971–2.View ArticlePubMedGoogle Scholar
- Tang Z, Zhang HH, Huang K, Zhang XG, Han MJ, Zhang Z. Repeated horizontal transfers of four DNA transposons in invertebrates and bats. Mob DNA. 2015;6(1):3.PubMed CentralView ArticlePubMedGoogle Scholar
- Dos Santos CB, Tavares MT, Leite GR, Ferreira AL, Rocha Lde S, Falqueto A. First Report of Aprostocetus asthenogmus (Hymenoptera: Eulophidae) in South America and Parasitizing Eggs of Triatominae Vectors of Chagas Disease. J Parasitol Res. 2014;2014:547439.PubMed CentralPubMedGoogle Scholar
- Yoshiyama M, Tu Z, Kainoh Y, Honda H, Shono T, Kimura K. Possible horizontal transfer of a transposable element from host to parasitoid. Mol Biol Evol. 2001;18(10):1952–8.View ArticlePubMedGoogle Scholar
- Maia Da Silva F, Junqueira AC, Campaner M, Rodrigues AC, Crisante G, Ramirez LE, et al. Comparative phylogeography of Trypanosoma rangeli and Rhodnius (Hemiptera: Reduviidae) supports a long coexistence of parasite lineages and their sympatric vectors. Mol Ecol. 2007;16(16):3361–73.View ArticlePubMedGoogle Scholar
- Oliveira SG, Bao W, Martins C, Jurka J. Horizontal transfers of Mariner transposons between mammals and insects. Mob DNA. 2012;3(1):14.PubMed CentralView ArticlePubMedGoogle Scholar
- Thomas J, Schaack S, Pritham EJ. Pervasive horizontal transfer of rolling-circle transposons among animals. Genome Biol Evol. 2010;2:656–64.PubMed CentralView ArticlePubMedGoogle Scholar
- Zhang HH, Shen YH, Xu HE, Liang HY, Han MJ, Zhang Z. A novel hAT element in Bombyx mori and Rhodnius prolixus: its relationship with miniature inverted repeat transposable elements (MITEs) and horizontal transfer. Insect Mol Biol. 2013;22(5):584–96.View ArticlePubMedGoogle Scholar