Closing the Gap: Horizontal Transfer of Mariner Transposons between Rhus Gall Aphids and Other Insects

Simple Summary Transposable elements (TEs) are mobile genetic elements that invade and multiply in host genomes. Besides vertical inheritance, it can transfer from one species to another through a phenomenon called horizontal transfer (HT). HT is crucial for TEs survival in genomes but also a significant disadvantage for host genomes, and recurrent HTT events between different hosts could shape and affect their genome architecture. HTT could be harmful to host genomes, but sometimes it can be useful and may play a role in the adaptive evolution of the host genome. HTT is well reported in many eukaryotes; however, there is still a considerable gap of knowledge about HTT in some organisms. This study closed the knowledge gap about HTT regarding Rhus gall aphids and reported multiple events of HTT involving Rhus gall aphids and other insects. Abstract Horizontal transfer of transposons (HTT) is an essential source of genomic evolution in eukaryotes. The HTT dynamics are well characterized in eukaryotes, including insects; however, there is a considerable gap in knowledge about HTT regarding many eukaryotes’ species. In this study, we analyzed the events of the HTT between Rhus gall aphids (Hemiptera) and other insects. We analyzed the Mariner-like transposable elements (MLEs) belonging to Rhus gall aphids for the possible HT events. The MLEs have a patchy distribution and high similarity over the entire element length with insect MLEs from different orders. We selected representative sequences from the Rhus gall MLEs and identified five events of HT between MLEs of Rhus gall aphids and other insects from five different orders. We also found multiple HTT events among the MLEs of insects from the five orders, demonstrating that these Mariner elements have been involved in recurrent HT between Rhus gall aphids and other insects. Our current study closed the knowledge gap surrounding HTT and reported the events between Rhus gall aphids and other insects for the first time. We believe that this study about HTT events will help us understand the evolution and spread of transposable elements in the genomes of Rhus gall aphids.


Introduction
Transposable elements (TEs) are mobile DNA sequences that can translocate in the host genome and replicate their number. This ability of TEs allows them to invade virtually all kinds of organisms, from prokaryotes to higher vertebrates, and including those of humans [1] and plants [2,3]. However, the mobility of TEs can lead to significant adaptive changes by promoting chromosomal rearrangements such as segmental duplications, deletions, and inversions through phenomena such as non-allelic homologous recombination [4,5]. On the other hand, TEs' expansion in the host genomes can also be harmful, leading eukaryotes to evolve various defense and regulatory mechanisms [6]. Due to their In our previous study, we reported the diversity of MLEs in seven species of Rhus gall aphids, i.e., Schlechtendalia chinensis, Schlechtendalia peitans, Nurudea ibofushi, Melaphis rhois, Meitanaphis flavogallis, Floraphis choui, and Kaburagia rhusicola. The seven species of aphids belonged to six major genera of Rhus gall aphids [36]. The Mariner family of DNA transposons is well known to perform recurrent and successful horizontal transfers, as supported by previous studies' conclusions [17,38]. The present study aimed to analyze all MLEs detected in Rhus gall aphids for possible HTTs between the Rhus gall aphids and other insects. This study will present the first report of HTTs between Rhus gall aphids and other insects belonging to different orders of Class Insecta.

Data Availability
We used 121 Mariner transposons sequences as queries in databases, and the detailed information, including accession numbers, is shown in Supplementary File S1. The qualified MLEs for the HT analysis in this study constitute the set detected in seven Rhus gall aphid species and are available on GenBank as of August 2021. We built a species phylogenetic tree based on 20 genes downloaded from GenBank (Supplementary File S2), while Marinerlike elements for the target insect's species were extracted from genome assemblies at NCBI after Blastn search and from TEs databases (Supplementary File S3). Alignment files used to obtain all the trees reported in this study can also be found in Supplementary File S4.

Identification and Annotation of MLEs in Targeted Species
To infer any possible transfer of Rhus gall MLEs with other insects, we followed the widely used two-step approach in reference to the previous studies [38]. Homology-based strategies were followed to find and extract similar nucleotide sequences from the genome of the target species. For this purpose, we used the transposable elements database, i.e., RepBase [39], and the whole genomes database, i.e., NCBI GenBank. RepBase searches for homologous sequences were performed using the "CENSOR" tool [40], implemented in RepBase [39] using the default parameters against the whole database. We retrieved a few sequences from the Repbase database and extracted sequences that followed our designed criteria (Sequence similarity ≥ 85%, and Query coverage ≥ 80%). At the same time, Blastn searches at NCBI were conducted with default parameters using MLEs of Rhus gall aphids in queries. We extracted the resulting Blastn hit produced by the qualified MLEs from the genome of the insect's species and were manually analyzed for their terminal inverted repeats (TIRs). To confirm the placement of the extracted sequences in the Mariner family, we searched the conserved motifs and domain of the sequences, i.e., Helix-turn-Helix HTH DNA binding motif and DDE catalytic domain using CD-search [41,42] and motif search online, last accessed on September 15, 2021. The open reading frame (ORF) was also predicted using the ORF finder (implemented in Geneious prime), and the MLEs were annotated using Geneious prime v11.1 [43].
The nucleotide sequence's identity between the MLEs of Rhus gall aphids and other insects included in the present study was searched for using Blastn. We used MLEs sequences with a query coverage ≥ 80%, and sequences shorter than this were filtered out from the analysis to avoid false-positive results.

Phylogenetic Analysis
We achieved the phylogenetic analysis in four steps: first, we aligned the full-length nucleotide sequences of all of the 121 MLEs of Rhus gall aphids with MAFFT v.7.1.1 [44] with the default parameter. The alignment was manually curated, followed by the construction of the ML phylogenetic tree. We used jModelTest v2.1.10 [45,46] to select the best evolutionary model that fitted adequately and resulted in a good tree. The Rhus gall MLEs were already classified into four subfamilies and sub-lineages in our previous study [34], and the same classification was followed in the present study.
Second, we aligned the qualified sequences from Rhus gall aphids and the extracted sequences from the targeted species, i.e., other insect orders using MAFFT implemented in Geneious. Alignment was trimmed manually, and MLEs phylogenetic tree was constructed with IQTREE using model GTR+I+G, as suggested by jModelTest.
In the third step, we constructed the species phylogenetic tree of seven species of Rhus gall aphid and the targeted insects belonging to four different orders, using fifteen mitochondrial genes and five nuclear genes. We aligned all of the 20 genes with MAFFT implemented in Geneious with default parameters. The aligned sequences were manually curated, and ML phylogenetic tree was constructed with IQTREE using model GTR+I+G, suggested by jModelTest. The tree was visualized and modified using Figtree v1.3.1 software [47], and bootstrap (1000 replicates) was used as statistical support for each branch. All the trees constructed were unrooted, which were manually rooted at the midpoint, during visualization and modification using Figtree v1.3.1.
In the fourth step, we estimated the divergence time between the species to infer the HT between the species, especially the five different clades representing each order. Divergence time was estimated using TimeTree online.

Estimating the Minimal Number of Horizontal Transfer Events
To infer multiple events of HTs, we analyzed and estimated the minimum number of HTT events in our present study. We considered the possibility that a single HT event may be sufficient to explain several cases of shared MLEs through horizontal transfer if they happened in the common ancestor of recently diverged species. We evaluated and compared the species tree and MLEs tree, considering all of the nodes in the tree, and predicted one HTT event if the descending clades sharing the same MLE were connected by a common ancestor.
Strictly speaking, we inferred and concluded that most of the HT events took place in ancestor branches, which passed to the descendant through a vertical transfer with slight divergence under natural selection. Nevertheless, for simplicity, we approximated that the species of our sample (Rhus gall aphids), whose MLEs have the closest similarity with MLEs of the other species, were potentially involved in HTT. We estimated the confirmed minimal HTT events between Rhus gall aphids and other insects following this procedure.
The MLEs involved in the event of HT were present in the middle of genomic contigs and chromosomes of the studied species. Their flanking sequences at both ends have minimal similarity, which rules out any possible contamination and validates the event of HT.

Results
In this study, we analyzed the MLEs that we detected in our previous study for HTT events between Rhus gall aphids and other insects belonging to different orders. To further understand the origin of evolution and inheritance of MLEs in Rhus gall aphids, we performed a detailed comparative phylogenetic analysis of the Rhus gall aphids and other insects' MLEs. Many MLEs of Rhus gall aphids from different lineages showed high pairwise identity with the TEs of phylogenetically distantly related insect species. The unique identities between TEs of Rhus gall aphids and other insects prompted us to have a systematic search for the HTT events involving Rhus gall aphids Mariner transposons. Many studies have successfully documented thousands of HTT events among the species of Class Insecta [19,48]. Nevertheless, due to the lack of TEs data about Rhus gall aphids, there was no evidence of HTT between Rhus gall aphids and other insects. We followed a detailed two-step approach to discriminate the HTT event in the present studies based on homology or nucleotide sequence identity and species phylogenetic tree comparisons with MLEs phylogenetic tree. We found several events of HTT between Rhus gall aphids and species from five different orders of Class Insecta.

Phylogenetic Relationship of the Rhus Gall Aphids MLEs
We determined the phylogenetic relationship of all of the MLEs of Rhus gall aphids analyzed in the present study with already known MLEs from the Tcl/Mariner Superfamily in our previous study [34]. All of the MLEs in this study belonged to four subfamilies of the Mariner family, i.e., Mauritiana, Irritans, Vertumana, and Drosophila. Phylogenetically, we classified MLEs of the Mauritiana subfamily into two sub-lineages based on sequence similarities, i.e., Botmar-like elements reported in Bombyx mori for the first time and Batmarlike elements found in Bactrocera tryoni. Phylogenetic distribution of all MLEs used in this study can be seen in Figure 1 across the seven species of Rhus gall aphids, among which Mauritiana, Vertumana, and Drosophila subfamily are distributed in all of the seven species. In comparison, MLEs from the Irritans subfamily is present in four species, i.e., Schlechtendalia chinensis, Schlechtendalia peitans, Nurudea ibofushi, and Meitanaphis flavogallis. Tips of branches indicate MLE name followed by GenBank accession number. Five-pointed star represents 100% bootstrap value. The tip of branches is labeled with the first two letters representing the acronym of their scientific names, while the next three-letter followed by digits indicates the Mariner family name and no. of MLEs.
In conclusion, the phylogenetic analysis shows the distribution of DNA transposons of the Mariner family in all of the studied Rhus gall aphids species used for HTT analysis in the present study. All of the MLEs reported in seven species of Rhus gall aphids showed patchy distribution and are not congruent with the species phylogenetic tree, i.e., MLEs from Irritans are present in only four species, which might also reflect HTT events within Rhus gall aphid species. To infer the horizontal transfer events, we selected representative sequences from each lineage of Rhus gall aphids and searched for identical sequences in other insect species following homology-based approaches.

Selection of the MLEs Representative Sequence for HTT
Initially, to infer the HTT event between transposons of Rhus gall aphids and other species, specifically insects, we include all of the 121 Rhus gall MLEs as queries (Supplemen-tary File S1). We blast the MLEs against the NCBI standard database and extensively search the highly similar DNA sequences in distantly related genomes. Among 121 sequences, 40 produced an excellent hit against genomes of other species with query coverage > 90% and similarity ≥ 80%. Although it is possible to include all copies of MLEs detected within one species, most sequences from the same lineages result in similar hits. Hence, it is simpler and quicker to use only a few representatives. As one sequence from the same lineage (subfamily) is enough to infer the desired result, we chose one representative from each lineage or more than one where required in each of the seven species. We selected the complete sequence from each lineage from all of the seven species, which could result in good Blastn hits (Query >90%, similarity ≥80%), and truncated sequences were discarded from the analysis. To further simplify our search and choose the best representative among the complete copies from the same lineages in each species, we chose the sequence which resulted in the best Blastn hit (high similarity and coverage with query sequence). For instance, Fcmar1, Fcmar2, and Fcmar3 belong to the same lineage of MLEs in Floraphis choui and resulted in similar hits, so we discarded Fcmar3 as it was incomplete, and Fcmar2 was selected as its results best Blastn hit against the genome (NCBI standard database). The same rule was applied to MLEs of all seven species, and, finally, 16 MLEs (Fcmar2, Kr-mar1, Krmar4, Krmar5, Mfmar1, Mfmar7, Mrmar9, Mrmar11, Mrmar16, Nimar1, Nimar13, Scmar2, Scmar10, Spmar2, Spmar3, and Spmar4) were selected.
Furthermore, to avoid repetitions and false-positive HTT events among Rhus gall aphids and other non-related species, only one orthologous sequence (MLE) was selected as representative from all of the MLEs in seven different species following the rules explained above. For instance, Fcmar2, Krmar1, Mfmar1, Nimar1, and Scmar2 belong to the same lineage (Figure 1), i.e., the Botmar-like elements of Mauritiana subfamily result in similar hits in Blastn, so only one Fcmar2 with the best hit was selected as representative for this lineage. The same rule was applied to all four subfamilies; for example, the three elements from the Drosophila subfamily, i.e., Mrmar11, Scmar10, and Spmar3, produced similar hits, so only one Mrmar11 was selected as representative for this lineage.
In some cases, more than one representative sequence was selected from the same lineage based on the Blastn hit it produced. For instance, Krmar4 and Krmar5 are derived from Kaburagia rhusicola and belongs to Batmar-like lineage of Mauritiana subfamily, but produced different Blastn hits (see Table 1), so both are included in the analysis. Finally, we selected eight MLEs from four different subfamilies involved in the HTT event between Rhus gall aphids and other insects. We followed the above-discussed criteria again to search for similar MLEs sequences in RepBase, which could be involved in HTT with Rhus gall aphids.

Inference of Horizontal Transfer between Rhus Gall TEs with Other Insects
Horizontal transfer event is well documented in many species of insects, but there are no reports of HTT of transposons in Rhus gall aphids. HT of transposons can be conferred either based on DNA sequence similarities or phylogenetic incongruences of TEs compared to neutrally evolving vertically transmitted genes or by combining both methods. Sequences of distantly related species with query coverage and similarity > 90% are considered to be horizontally transferred [48], while sequences sharing terminal inverted repeats (TIRs) of a similarity > 90% could also be a result of HTT event [16]. We also followed the commonly used two-step approach to unveil the phenomena of HT in Rhus gall aphids and other insects.

Inference of HTT Based on Nucleotides Sequence Similarities
The first step to uncovering the possible events of HT in Rhus gall aphids was based on DNA sequence similarities with other non-related species. We search for homologous sequences of the Rhus gall MLEs in Repbase using the "censor" tool implemented in the Repbase database with default parameters, and Blastn explores the NCBI database using default parameters. Deleted and truncated small copies can lead to false-positive results; we extracted copies only with query > 90% and similarity > 80%. We found very few good hits in Repbase as per the designed criteria for the study, but retrieved many homologous sequences from Blastn searched at NCBI. Many sequences belonging to the same lineages within the same species resulted in similar hits, so we selected sequences with the highest DNA similarity throughout the length as a representative sequence from each lineage, as explained above. The top-ranking results from NCBI Blastn, i.e., query coverage > 90%, identity > 85% with lowest E-value, and higher bit score, are shown in Table 1. Likewise, the top-ranking results from RepBAse, i.e., query coverage > 90%, similarity with consensus sequence > 90%, Pos-value, and higher bit score, are shown in Table 2. The detected sequences in other insects display higher nucleotide identity, which exceeds the expected identity values when comparing transposable elements in distantly related species. We found many sequences in the NCBI and RepBase databases to be very similar to Rhus gall aphids transposons (identity ≥ 85%, and query ≥ 95%) belonging to seventeen insects species from five different orders, i.e., (Hymenoptera, Diptera, Coleoptera, Lepidoptera, and Neuroptera). The Rhus gall aphid species belongs to the order Hemiptera of Class Insecta, so the current study represents HTT events among six different orders of Class Insecta (Tables 1 and 2). As expected for TEs, in some cases, the extent of sequence identity was observed throughout the length of elements of TEs, including TIRs, which strongly supports the hypothesis of the HTT event. Some of the examples that support this can be seen in Figures 2 and 3 and Tables 1 and 2, in which the DNA sequence similarity between TEs of Rhus gall aphid species with distantly related species of different orders are more than 96%, while the amino acid sequences similarities are also more than 96%.

Phylogenetic Analysis of HTT among Rhus Gall Aphids and Other Insects
To further confirm the phenomenon of HTT between Rhus gall aphids and five distantly related insect orders, we constructed the species tree (See Figure 4A) of all included species in this study by selecting 20 highly conserved orthologous genes, resulting in a good quality phylogenetic tree. We could not retrieve the mitochondrial and nuclear genes for Drosophila elegans and Herpegnathos saltator due to unavailability in public databases. Drosophila elegans was branched manually in the species tree using its divergence information from Timetree online, while the position of Herpegnathos saltator can be represented by other ant species, i.e., Acromyrmex echinatior. We concatenate and aligned the selected 20 genes of the species; 15 mitochondrial genes, i.e., 12S rRNA, 16S rRNA, ATP6, APT8, COX1, COX2, COX3, Cyt-b, ND1, ND2, ND3, ND4, ND4L, ND5, and ND6, and 5 nuclear genes, i.e., Long-wavelength rhodopsin resistance gene (lwrh), wingless (wnt-1), Elongation factor 1-alpha (EF1-alpha), Histone (H3), and 18S rRNA gene (see Supplementary File S2) to construct species tree. We also constructed the transposons tree ( Figure 4B) of representative sequences of Rhus gall aphids MLEs and other insects MLEs recovered from NCBI GenBank and Repbase. We compared the MLEs tree to the species tree of the Rhus gall aphids and other insects included in this study (Figure 4). These comparisons clearly showed at least five events of HT among Rhus gall aphids and other insects belonging to five different orders and few events of HTT within the insects of the other five orders. Seven MLEs from three subfamilies and four lineages of Rhus gall aphids clustered with the MLEs of insects from distantly related orders provide strong evidence of HTT. For instance, an MLE (Fc-mar2_MW699035) of the gall aphid Floraphis choui belonging to the Botmar-like lineage of Mauritiana subfamily clustered with these of Ocypus olens (Oomar1_OU343056) from Coleoptera order, Myrmica ruginodis (Myrmar2_AY652426), and other insects from Bombus genus of order Hymenoptera. In the present scenario, MLEs of Rhus gall aphid nested with the MLEs of very distantly related orders, i.e., Coleoptera and Hymenoptera, diverged be-tween 323-392 MYA from order Hemiptera, which provides strong evidence of HT between species of these orders. Similarly, MLEs of Kaburagia rhusicola (Krmar4 & Krmar5) nested with MLEs of Drosophila elegans (Mariner_7_DEI), Bactrocera tryoni (Batmar11_KX931004.1) from order Diptera, and Tinea trinotella (Ttmar2_HG992305) from order Lepidoptera, respectively. Other MLEs from Rhus gall aphid Melaphis rhois (Mrmar11) clustered with these of Chrysoperla carnea (Ccmar1_FR997756) from order Neuroptera, Bactrocera tryoni (Batmar6_KX930994) from order Diptera, and Mellicta athalia (Mamar1_HG992328) from order Lepidoptera. At the same time, one MLE (Mrmar16) nested with that of Herpegnathos saltator (Mariner33_Hsal) from the Hymenoptera order ( Figure 4B).
All the species mentioned above diverged from Rhus gall aphids with more than 300 MYA ( Figure 5). The genes and MLEs of these species cannot nest together in a phylogenetic tree in normal circumstances in the absence of HT. In conclusion, of the above phylogenetic analysis, the clustering of MLEs from five distantly related orders of insects with the Rhus gall aphids MLEs provides strong evidence of HT between these groups of insects. Several HTTs events have also been observed within the other insect orders included in this study. For instance, a Botmar-like MLE from the Mauritiana subfamily (Oomar1_OU343056) extracted from Ocypus olens of Coleoptera order nested with the MLEs of Myrmica ruginodis and Acromyrmex echinatior of order Hymenoptera, which reflects multiple events of HTT of this MLE between Myrmica ruginodis, Ocypus olens, and Floraphis choui ( Figure 4B). Another such event of HTT can be observed in a Batmar-like MLE of Mauritiana subfamily in Acromymex echinatior (Mariner-18_Ace) of Hymenoptera order, which nested with Bactrocera tryoni MLE (Batmar-11_KX931004) of order Diptera. It also reflects multiple events of HTTs of these elements between Acromymex echinatior, Bactrocera tryoni, and Kaburagia rhusicola. Furthermore, one MLE of the Drosophila subfamily (Ccmar_FR997756) extracted from Chrysoperla carnea (Order: Neuroptera) was nested with Batmar6_KX930994 of Bactrocera tryoni (Order: Diptera), Mamar1_HG992203 of Mellicta athalia, and Ttmar1_HG992328 of Tinea trinotella from order Lepidoptera, which indicates several events of HTT by this MLE between Chrysoperla carnea, Bactrocera tryoni, Mellicta athalia, Tinea trinotella, and Melaphis rhois ( Figure 4B).
We could not discriminate the HTT events within Rhus gall aphid species. All the aphids in this study are closely related and belong to the same subfamily Eriostominae, with closed divergence time. As recently diverged species share high nucleotide similarities in their sequences, making it challenging to detect HT in them based on nucleotide identities. Although the Rhus gall MLEs showed patchy distribution among the Rhus gall aphids, but did not produce positive HT signals based on the designed criteria for the present study.
In conclusion, the above results show that Rhus gall aphid transposons had striking identities with transposons of distantly related species and clustered together in the MLEs tree. Meanwhile, all the Rhus gall aphids clustered apart from other insects in the species tree based on nuclear and mitochondrial genes. These unexpected sequence similarities of galling aphids MLEs with other insects from different orders and the uneven distribution of MLEs in the phylogenetic tree are assumed to be solid evidence of HTT between Rhus gall aphids and other insects.

Estimation of Divergence Time
All of the insects in the tree shared a common ancestor and belonged to the same Class; we estimated the divergence time of each species and the order of insects included in this study ( Figure 5). Divergence times were estimated to discover the distances between the species, which will help to discriminate the HT events between the sequences. Due to evolution, species with common ancestors and that diverged a long time ago tend to accumulate more mutations and changes in their nucleotide sequences. We estimated the divergence time of all of the species using Timetree online at (https://www.timetree.org, accessed on 1 January 2022) and the divergence time between Rhus gall aphids with all of the other insects in the tree was more than 300-350 Myr, with an average of 325 Myr. Comparatively, the MLEs sequences of Rhus gall aphids and other insects showed contrasting nucleotide similarities, which was impossible for the neutrally evolving genes diverged so long ago. For example, Sclechtendalia chinensis diverged 350-375 Myr from Bactrocera tryoni and nested very far apart in the species tree. Still, their MLEs showed 96% nucleotide sequence similarities between them and nested as sister sequences in the MLEs tree. The very high divergence time between the studied species and closed MLEs sequence similarities further support our claim of HTT events between these species.

Discussion
HTT is a well-known and reported phenomenon documented in many metazoans, including insects. Although the exact mechanism of HTT is poorly understood, the geographic proximity and host-parasite interactions might help in the exchange of genetic material in distantly related species [49,50]. To date (visited 1 January 2022), there are 5689 cases of HTT that have been reported in the Horizontal transfer of transposons database (HTT-DB) [51]. Among all of the HTT events reported in HTT-DB, the Tc1/Mariner Superfamily of DNA transposons contribute to most cases, i.e., 2523 out of 4271 DNA transposon HT events. HTT is well reported in most insects. Recent studies uncovered thousands of HTT events in the Class Insecta [19], while others reported thousands of HTT events in vertebrates [52]. However, there is no report of HTT involving Rhus gall aphids to date due to the unavailability of TEs data in these aphids. We recently reported Mariner transposons in Rhus gall aphids for the first time. We followed the previous findings to uncover the events of HTT among Rhus gall aphids and other insects and closed the knowledge gap about HTT in galling aphids.
Methodologically, it is not easy to infer HTT events. Several tools and methods can be used; for example, the VHICA tool concludes HTT events based on codon usage analyses and compares synonymous and non-synonymous substitutions rates [18]. Unfortunately, there are limitations to these automatic tools. These perform poorly if the divergence between the species in which HTT is inferred increases, leading to substitution saturation, which causes loss of the phylogenetic signal [53]. In this study, the divergence time between Rhus gall aphids and other species for which HTT was inferred is huge (>300 Myr), and the methods based on phylogeny and genetic distance seem to be more suitable, which is also suggested by previous studies [54,55]. We followed a two-step approach explained above and uncovered a few events of HTTs between Rhus gall aphids and 11 other species of insects belonging to 5 orders of Class Insecta. HTT events detected from more than one species of the same order for the same MLEs were considered single HTT events. Among the seven species of Rhus gall aphids, TEs from three species seem to have undergone HTT events.
We have inferred HTT events involving six DNA transposons of the Mariner family in Rhus gall aphids. Our findings suggest that HTT events between Rhus gall aphids and distantly related insects might have occurred several times. Our results also suggest that the MLEs involved in Rhus gall aphids' HTT events have also undergone many HT between other orders of insects. Moreover, the common ancestor of all the insects, Rhus gall aphids (Order: Hemiptera), is quite distantly related to other insects in this study, and the HTT scenario is very clear from the results; however, it is challenging to infer both the direction and the vector of the HTT events described in this study.
Interestingly, all of the putative cases of HTT detected in this study involved 21 insect species from five different orders. In this respect, Order Hymenoptera seems to be the preferred order in exchanging MLEs since seven other bee species belonging to three different genera and five ants species from different genera are putatively involved in HTT events with four Rhus gall aphids Mariner elements (namely Fcmar2, Krmar4, Krmar5, Mrmar16). Insects from the order Hymenoptera have been involved in many HTT events in previous studies [38], as is supported by the present study. In comparison, four of the twenty-one insects involved in HTT events in the study belong to the order Lepidoptera and four from the order Diptera, respectively, with five Rhus gall elements (Krmar4, Krmar5, Spmar2, Scmar11, Mrmar11). A recent study found that the Lepidoptera order has been the hotspot of HTT in insects [20], while fruitflies from the order Diptera were also suggested as good horizontal transfer candidates [56]. At the same time, one species, each from beetles (Order: Coleoptera) and Laecwings (order: Neuroptera), also seemed to be involved in the HTT event with Rhus gall aphid Mariner elements (Fcmar2, Mrmar11). All the Mariner elements involved in HTT events in this study belong to two subfamilies in which elements of the Mauritiana subfamily seem to be dominantly involved in HTT events.
Although there is a patchy distribution of MLEs among the Rhus gall aphids, and they do not follow the species tree, it is difficult to infer the HTT events among closely related species [19]. All the Rhus gall aphids belong to the same subfamily Eriosomatinae and are closely related phylogenetically. To avoid false-positive results, we could not infer HTT events within Rhus gall aphids but have drawn observations from the study that Rhus gall genomes are equally targeted to transposition and HT events of transposons. The presence of potentially active TEs in the genome could also be involved in HT events within Rhus gall aphids yet be undetectable due to a closed evolutionary relationship.

Conclusions
Our study reveals that the evolutionary history of Mariner transposons in Rhus gall aphids has been subjected to many events of HT, involving a total of five other orders of insects at the same time. Moreover, our results show that Mariner elements from the Mauritiana subfamily are involved in more HT events compared to other MLEs. These results contribute to the description of transposons as genomic symbionts that mobilize and move between different host lineages, evolving and shaping their host genomes. Our study represented the HT events involving the Rhus gall aphids for the first time and closed the information gap about HTs events in galling aphids.