The knowledge of the genome composition and their architecture is essential for understanding the evolutionary processes that occur in species. One of the most abundant and important components of genomes are repetitive elements. These repetitive sequences can be classified in satellite DNA and transposable elements (TEs). Satellite DNA plays a notable role in the evolution of chromosomes, including sex chromosomes, and in the organization and chromosomal speciation (Ruiz-Ruano et al., 2016; Robles et al., 2017; Kretschmer et al., 2022). On the other hand, TEs are sequences that have been present in eukaryotic genomes for a long time, and have had a major influence over millennia (Feschotte and Pritham, 2007; Raskina et al., 2008; Belyayev, 2014; Sotero-Caio et al., 2017; Bourque et al., 2018). These sequences can move through the genome and insert themselves into new chromosomal regions, which contributes significantly to genetic diversity (Makalowski, 2000; Gao et al., 2016; Yuan et al., 2018). Initially, these mobile elements, as long as satellites, did not attract much attention from researchers and were erroneously treated as “junk DNA”. However, recently TEs are recognized as evolutionarily and functionally critical components in genome evolution (Biémont, 2010; Chuong et al., 2017) and are involved in processes of speciation, sex determination, chromosomal rearrangements, creation of new genes, adaptation to the environment, migratory patterns, climate change, etc (Long et al., 2003; Kvikstad and Makova, 2010; Chalopin et al., 2015b; Auvinet et al., 2018; Platt et al., 2018; Carotti et al., 2021; Zhao et al., 2022).

TEs have colonized all sequenced species to date, but with varying success. The abundance of these elements can vary between 4-60% of vertebrate genomes sequenced to date (Sotero-Caio et al., 2017; Chang et al., 2022). Although the mobility of TEs is generally deleterious to the host, the accumulation of TEs in genomes represents a source of raw genetic material that can be used during evolution to benefit a variety of cellular functions, including those related to embryogenesis (Jachowicz et al., 2017; Chang et al., 2022). Among other findings, TEs have been related to the adaptive evolution of warm-blooded fish, such as the opah fish, which contains the highest percentage of LTR elements known in teleosts to date, and where it has been shown that the expansion of these elements in their genome contributed to the opah’s endothermic capacity and adaptation to deep-sea environments (Wang et al., 2022a). In syngnathid fish, TEs have also played a fundamental role in their evolution and adaptation to the environment, through the recent expansion of TEs in the vicinity of existing supernumerary genes in this group of fish (Small et al., 2022). Furthermore, the fundamental role that TEs have played in the adaptive success and invasiveness of tunicates has been demonstrated (Wei et al., 2020). Finally, TEs have been implicated in the diversification of zinc finger genes in animals, where these elements have been linked to the expansion of TEs throughout metazoan evolution (Wells et al., 2023).

Based on their transposition mechanism, TEs are classified into retrotransposons (Class I) and DNA transposons (Class II) (Wells and Feschotte, 2020). Class I elements are characterized by their copy-and-paste transposition mechanism, in which their own RNA is reverse-transcribed into its complementary DNA by an RNA-dependent DNA polymerase (RT) and then reintegrated into the host genome (Carducci et al., 2020; Chang et al., 2022). These Class I elements can be further divided into LTR (Long Terminal Repeats), non-LTR, and Penelope retroelements. In turn, all of these retroelements are divided into multiple superfamilies such as different types of LINEs, SINEs, DIRs, Crypton, among others (Wells and Feschotte, 2020). On the other hand, Class II mobile elements use an intermediate DNA to transpose their copies to a new chromosomal position, and in general their transposition mechanism occurs through a cut-and-paste process, in which both DNA strands are separated (Chen et al., 2014; Chalopin et al., 2015a). These DNA transposons can be further divided into subclasses I and II (Bourque et al., 2018; Goerner-Potvin and Bourque, 2018; Carducci et al., 2020). In those of Subclass I, we can find the superfamilies hAT, Merlin, Tc1-Mariner, among others. The major representatives of the subclass II are Helitrons (Rolling Circles, RC) and Maverick, which, unlike the rest of Class II elements, transpose through a copy-and-paste mechanism (Wicker et al., 2007). Specifically, Helitrons represent a new class of transposable element discovered recently in animals and plants (Kapitonov and Jurka, 2001; Thomas et al., 2010; Grabundzija et al., 2016; Xiong et al., 2016). These elements have two notable characteristics. The first of them is that Helitrons replicate and mobilize through a mechanism known as rolling-circle replication (RCR) (Grabundzija et al., 2016; Xiong et al., 2016; Wang et al., 2022b). This transposition mechanism was first described in phages and plasmids (Khan, 2005; Ruiz-Masó et al., 2015; Zattera and Bruschi, 2022). Later, it has been shown that the transposition of a bat Helitron in a human cell assay system generated covalently closed circular intermediates, as predicted by the RCR model (Grabundzija et al., 2016). On the other hand, Helitrons can capture gene sequences, which makes them an element of notable evolutionary importance (Lai et al., 2005; Xiong et al., 2014; Thomas and Pritham, 2015).

Studies to date confirm that the TE content is highly variable among vertebrates, including fish. Thus, the genomes of species of mammals, reptiles, coelacanths, Xenopus, and fish have been shown to have coverage percentages above 20% (Chalopin et al., 2015a) and yet, some compact genomes, such as those of pufferfishes (fugu and tetraodon) and birds are poor in TEs (<5%). The variation percentages can be enormous, with differences of even 10 times (Chang et al., 2022). The composition also varies between organisms, so in most teleosts, amphioxus, tunicates, and Xenopus, DNA transposons predominate, although mammals, birds, coelacanths, and elephant shark are especially poor in these elements, with retroelements being the main transposable elements. In addition, some actinopterygians and nonbony vertebrates show a higher abundance of LINEs and SINEs. Finally, tunicates present mainly LTR retrotransposons.

In recent years, some studies have been conducted on the abundance and diversity of repetitive elements in fish, but the absence of complete genomic maps for a large number of species has made it difficult to study them systematically (Chalopin et al., 2015a, Chalopin et al., 2015b; Gao et al., 2016). The arrival of high-throughput technologies and bioinformatics has provided a wealth of genomic data on fish (Carducci et al., 2020). The data published to date suggest that, compared to other vertebrate genomes, DNA transposons are the most abundant in most fish genomes (Shao et al., 2019). Another general characteristic of Actinopterigii is the presence of more recent copies of TEs than those found in other vertebrate lineages, and in many cases two rapid amplification of TEs are observed (Carducci et al., 2020).

However, many fish families and orders are still in an early stage of genome sequencing, and therefore of their knowledge, with a significant deficit of information about them. This is the case of flatfish in which the current information is scarce and the information on abundance, divergence, and chromosomal distribution of repetitive sequences is almost non-existent, where frequently only general abundance data obtained during the semi-automatic process of genome sequencing and annotation are available (Chalopin et al., 2015a; Gao et al., 2016; Chalopin and Volff, 2017; Shao et al., 2019; Lü et al., 2021).

The Senegalese sole (Solea senegalensis) is among the most important flatfish, with a wide distribution along the eastern coast of the Atlantic Ocean and in the Mediterranean Sea and a high economic value (Imsland et al., 2004; Díaz-Ferguson et al., 2007, Díaz-Ferguson et al., 2012). This commercial interest has promoted the increase in genomic resources in the last decade (Robledo et al., 2017; García-Angulo et al., 2018; Cross et al., 2020; Merlo et al., 2021; Rodríguez et al., 2021; de la Herrán et al., 2023), including an initial version of its genome (Guerrero-Cózar et al., 2021) and a recent improved version (de la Herrán et al., 2023). Due to the absence, until 2023, of a quality sequenced genome in this species, repetitive sequence studies on Senegalese sole had been limited to the sequence analysis of some BAC clones mapped on the chromosomes of the species (García et al., 2019; Rodríguez et al., 2019; Cross et al., 2020; Rodríguez et al., 2021; Ramírez et al., 2022) and the study of TEs in the Hox gene clusters of three flatfish species, including S. senegalensis (Mendizábal-Castillero et al., 2022).

The present study has allowed us to study the abundance of repetitive elements in 8 species of the order Pleuronectiformes, and two species outside this order (Carangiforme and Spariforme) from recent sequenced genomes available in databases. We have studied in more depth the repetitive elements of the S. senegalensis and described the abundance, divergence and distribution of TEs by chromosomes of this species.

In the present study, a general analysis of repetitive element abundance in the genome of eight flatfish species was carried out. The comparative analysis showed differences in the abundance of this fraction of the genome among pleuronectiform species, as well as between these species and the other species analyzed from Carangiformes and Spariformes orders. The species analyzed in the present study clearly show the common situation of teleosts, with DNA transposons predominating (Chalopin et al., 2015a; Sotero-Caio et al., 2017). In recent genome sequencing and annotation studies of flatfish, general data on the abundance of DNA transposons, LINEs, SINEs, and LTRs in 10 flatfish species showed that in four of the analyzed species, belonging to the Achiridae, Paralichthydae, Cynoglossidae, and Soleidae families, the overall coverage sum of the analyzed retrotransposon categories (LTRs, LINEs, and SINEs) exceeded that of DNA transposons (Lü et al., 2021), although the absence of other more precisely covered data such as Helitrons or another families included in the general categories, prevents a deeper analysis of the observed differences in these species.

In S. senegalensis, there are two references to the composition of repetitive elements based on whole-genome data. In 2021, in the first version of the species’ genome (Guerrero-Cózar et al., 2021), during the genome annotation process, a brief reference is made to the global content of repetitive sequences, giving a value of 23.41% for a female linkage map and 23.55% for the male one, without any additional contribution or evaluation in relation to these sequences and their classes, types, superfamilies, or families. Subsequently, in a new, more complete and improved version of the genome (de la Herrán et al., 2023), more up-to-date general data on repetitive elements were obtained. To do this, the authors created a library of repetitive elements directly from non-assambled sequencing reads (Illumina), using an experimental design based on the de novo analysis performed by the RepeatExplorer platform (Novák et al., 2020). This tool is suitable for obtaining libraries of new repetitive sequences, which are highly represented in genomic reads, such as those from repetitive regions. Some of these reads can be eliminated in assembly processes and therefore is an advantage of the technique. However, the information is never exhaustive, the annotation of the contigs obtained from the program is deficient, because it uses libraries that are not specific, at least for fish, and the quantification that RepeatMasker performs using these RepeatExplorer libraries significantly underestimates the quantification of these repetitive elements. The data obtained in that work showed that, using the library obtained with RepeatExplorer, repetitive sequences made up 8.2% of the genome of S. senegalensis, a percentage much lower than that obtained by Guerrero-Cózar et al. (2021), and much lower than that obtained using the de novo library constructed in the present work (28.68%) or the results also obtained by combining libraries of repetitive elements from teleosts and de novo library (39.54%). The contents of repetitive sequences obtained from different studies are highly dependent on the methodology used. In general, different combinations of annotation of repetitive elements are used, such as homology analyses combined with de novo analyses, as has been done in this work, to obtain a more complete view of the content of repeated sequences in genomes. It is also important to point out that due to this, it is appropriate to study the group of genomes that are being analyzed with the same methodology, as is the case of the study that has been carried out in flatfish in this work. In this way, with this approach, the comparison allows for more robust conclusions to be drawn in relation to the relative differences between the species analyzed. Additionally, the use of RepeatMasker on a common teleost database for all species analyzed avoids biases with respect to the use of species-specific databases, where the different quality of these libraries could produce distortions in the results.

On the other hand, the superfamily analysis carried out in this work in flatfish species has shown them to be poor in SINEs. This is consistent with previous work, where it seems that this absence of SINEs is a common feature in most fish studied (Sotero-Caio et al., 2017; Shao et al., 2019). In flatfish, this same situation has been described in other species belonging to families Achiridae, Bothidae, Rhombosoleidae, or Toxotidae, among others, where the values were close to 0% coverage (Lü et al., 2021). The analysis of other superfamilies has also shown, in this work, a high value of the hobo-Activator DNA transposon, followed by retroelements, LINEs, L2/CR1/rex, LTR elements, and Gypsy/DIRS1 ( Figure 1 , Supplementary Figure 1 ). In most fish genomes studied to date, it has also been seen that hAT, L1, L2, and Gypsy are the most widely distributed and are the most predominant (Shao et al., 2019). In flatfish, however, despite recent sequencing and annotation of new genomes (Lü et al., 2021), the absence of detailed analysis of TE families and superfamilies prevents comparisons between species of this group. On the other hand, in studies carried out with other TE families in other fish species, their abundance has been shown to be more specific to species. For other families, however, abundance may be more specific to species. This is the case of the CR1 superfamily, which in fish species that have diverged more recently, has very low values. Among these species are those that have not undergone the specific genome duplication event of teleosts (Chalopin et al., 2015a; Sotero-Caio et al., 2017). In addition, in other fish species, the levels of each TE superfamily seem to be very specific and dependent on the species itself. This occurs for example for the L2 and RTE elements in Nothobranchius furzeri, for Gypsy elements in Boleophthalmus pectinirostris, or Tc/mariner in Astyoanax mexicanus (Shao et al., 2019). It has also been described that in opah fish approximately 50% of their genome is composed of repetitive sequences (Wang et al., 2022a). The data presented for the first time in this study in flatfish species, show high values of abundance of Class I and Class II TE elements in the species S. senegalensis in relation to the rest of the species analyzed, showing even more similarity of abundance with a species as evolutionarily distant as S. aurata (family Sparidae) than with the rest of the species of Pleuronectiformes or Carangiformes. On the other hand, the low abundance of Helitrons in C. semilaevis compared to the rest of the species is also notable. However, the sequencing technology used to assemble this species (short reads), may in part influence the detection of its low abundance, although future genome reassemblies with new hybrid assemblies (long and short reads) pipelines will improve the analysis of the C. semilaevis genome. It has been described that in general, TEs in fish are regularly distributed and that the relationships between species with similar TE distribution are consistent with phylogenetic relationships. However, the results observed in this study confirm that although there is similarity in abundance at the global level among the genomes of the eight flatfish species, this relationship is not met for all families analyzed. Since genome protection processes (e.g., Piwi-interacting small RNAs, DNA methylation) regulate TEs, the loss and gain of the same must be associated with the host genome itself (Levin and Moran, 2011). Harmful TEs that compete with the host genome are more likely to be eliminated, while more beneficial TEs are likely to be conserved in genomes. In this way, the most specific and abundant superfamilies in some fish species could play a key role in the evolution of their genomes and may even be related to the biological characteristics of the species themselves (Venner et al., 2009). Our results indicate that TE levels in species belonging to the same group can have large differences and be more specific to each species than to a phylogenetic group ( Supplementary Figure 1 ). It is worth noting that there are currently no specific and comparative data on the abundance of repeated sequences in flatfish, so the results shown here provide a very important additional value for the study of the evolution of the genomes of fish in general and pleuronectiformes in particular.

In this work, we deepened the analysis of the distribution of mobile elements in S. senegalensis, since it is the species that has shown the greatest differences compared to the rest of the genomes of the families of Pleuronectiformes analyzed. The interaction of TEs with their host genomes has been compared to the interaction of species with ecosystems. Thus, TEs proliferate and use genome resources while interacting with other mobile elements (Leonardo and Nuzhdin, 2002). In that sense, we carried out the study of abundance by chromosomes, considering that they could have a seemingly more local behavior, like an ecosystem, and reflect their evolution. In previous studies, partial analyses of the distribution of repetitive elements in S. senegalensis have been carried out by analyzing the content of these elements in BAC clones located in cytogenetic maps (Rodríguez et al., 2019, Rodríguez et al., 2021; Ramírez et al., 2022). In these studies, clones (between 4 and 8) were analyzed, spaced out in some chromosomes. Despite the scarce amount of genome analyzed, limited by the small portion of each chromosome contained in these BACs and of genome studied, it was possible to observe certain differences in the abundance of elements in BACs belonging to different chromosomes and different intrachromosomal location (García et al., 2019; Rodríguez et al., 2019, Rodríguez et al., 2021; Ramírez et al., 2022). However, it was only after the publication of the recent complete whole-genome sequence of S. senegalensis (de la Herrán et al., 2023), we have been able to completely analyze the abundance of repetitive elements in each of the 21 pairs of chromosomes of the species. Although the content of the different types of elements in general seemed to show a similar behavior, with homogenized values in the chromosomes, the study by superfamilies did show differences in coverage between chromosomes, very notable in the case of satellites and Helitrons. These elements, although not the only ones, are the ones that show the greatest heterogeneity throughout the chromosomes, mainly the Helitrons, with peaks of abundance located in centromeric regions of some chromosomes, mainly in biarmed chromosomes, although also in acrocentric chromosomes ( Figures 3 – 5 ). Karyotype of S. senegalensis (2n=42) is divided in three pairs of metacentric chromosomes, two submetacentric, four telocentric and twelve acrocentric ones (Vega et al., 2002). Previous analyses demonstrated that the centromeres of 19 out of 21 pairs of S. senegalensis were occupied by PvuII satellite DNA (Robles et al., 2017). Furthermore, our in silico mapping of centromeres is coherent with the morphology in all cases with four exceptions: (1, 2) chromosomes 14 and 18, for which no centromeric sequences were detected. Chromosomes 14 and 18 are telocentric. No telomeric motifs were found at neither of the distal regions. This might indicate that this portion, including the centromeric region, was not assembled, (3) chromosome 19 exhibited two signals of centromeric PvuII in distal regions of both arms and (4) chromosome 5, telocentric according to our characterization but submetacentric according to morphology. The p arm of chromosomal pair 5 bears the ribosomal 45S in this species (Cross et al., 2006). The genome assembly used in our analyses for mapping missed this 45S unit (de la Herrán et al., 2023), which was found to be massively present in unanchored scaffolds (unpublished data). This might be the reason of the discrepancy in morphology we found in our analyses on centromeric position of this chromosome pair. Thus, we were able to unambiguously characterize the centromeric region of 18 out of the 21 chromosome complement.

Furthermore, it is known that the chromosomal evolution in flatfish has presented Robertsonian fusion and intrachromosomal duplication processes (Merlo et al., 2021; Rodríguez et al., 2021; Ramírez et al., 2022). In S. senegalensis it has been described that centromeric or Robertsonian fusions have occurred in 3 of the 9 biarmed chromosomes of the species. These chromosomes were 1, 2 and 4 where, based on BAC clone synteny studies, other pericentric rearrangements (inversions), that have occurred during the evolution of these chromosomes were described (Rodríguez et al., 2019, Rodríguez et al., 2021; Ramírez et al., 2022). There is evidence to support that inversions, as well as other chromosomal rearrangements, are involved in the adaptation of species to the environment and that polymorphisms associated with these inversions are related to geographic distributions (Wellenreuther and Bernatchez, 2018; Amorim et al., 2021). TEs are considered key elements in this chromosomal rearrangement process (Feschotte and Pritham, 2007). TEs and associated machinery play an important role in the evolution of the structure of the centromeres and their function (Wong and Choo, 2004). Regardless of their origin, centromeric sequences in higher eukaryotes contain extensive and homogeneous tandem repeat sequences of satellites and TEs (Wong and Choo, 2004; Klein et al., 2018). Centromeres can be considered functionally defined regions in eukaryotic chromosomes that show strong evidence of recurrent evolutionary novelties facilitated by TE activity. The impact of TEs on centromeres includes both the structure of the centromeric ecosystem itself and the proteins involved in centromeric identity and function (Klein et al., 2018). Studies of human populations have revealed that active insertions of TEs into centromeres have occurred during the evolution of modern humans and can facilitate rare events of centromeric recombination (Contreras-Galindo et al., 2013; Zahn et al., 2015). In Arabidopsis, TEs make up approximately 11% of the genome and are enriched mainly in the pericentromeric heterochromatin regions (Kapitonov and Jurka, 2001; Wong and Choo, 2004). Numerous evidences have shown the implication of TEs and transposases in the evolution of centromeric DNA, among which the presence of TEs (retrotransposons) specific to centromeres of different plants such as maize and grass (Langdon et al., 2000; Nagaki et al., 2003; Jin et al., 2004). In certain maize species and humans, a process of displacement of TEs towards pericentric regions has been observed, with a reduction of TEs within the centromeres themselves. This extraction process provides a mechanism for protection against the potential harmful effects of any newly emerging TE in the developing or established centromeric chromatin. This process offers an explanation for the accumulation and high prevalence of TEs found in the pericentromeric domains of many centromeres (Mroczek and Dawe, 2003). In S. senegalensis, tandem repeats of Helitrons are not localized exactly in the centromeric sequences of the satellite PvuII family previously described (Robles et al., 2017) and used to map the centromeres of this species, but in pericentromeric positions, as described previously in humans and plants.

Interestingly, the potential capacity of TEs to contribute to the formation of satellite arrays in centromeres of genomes has been demonstrated through the production of tandem internal repeats via their folding mechanism (Dias et al., 2014). It has also been described that the insertion of TEs in centromeres is due to the fact that they probably represent safe insertion zones, both for the host and for the TEs (Birchler and Presting, 2012; Sultana et al., 2017). Thus, TEs localized in centromeres could not cause insertional mutagenesis in centromeres since the surrounding repeated sequences could act as a “buffer” and the suppression of crossing-over events in centromeres. This location could, therefore, protect recently inserted TEs from the type of recombination events that cause mutations and usually result in the loss of mobility of these TEs (Gent et al., 2017; Klein et al., 2018).

One of the most relevant findings of the present study is the discovery of multiple insertions of Helitron transposons, through tandem arrays, in centromeric-pericentromeric regions of many of the chromosomes of S. senegalensis ( Figure 5 ). To our knowledge, this has only been previously described in mammals, such as primates or bats, and in plants (Xiong et al., 2016; Klein et al., 2018; Wang et al., 2022b). As has been mentioned, Helitrons are a class of eukaryotic transposon with an important role in the shaping of current genomes (Schnable et al., 2009; Yang and Bennetzen, 2009a, Yang and Bennetzen, 2009b). Helitrons are widely distributed in plants and invertebrates, often contributing to a high percentage of the genome (Putnam et al., 2007; Yang and Bennetzen, 2009b; Han et al., 2013; Peñaloza et al., 2021) and more recently in bats (Pritham and Feschotte, 2007; Thomas et al., 2014; Grabundzija et al., 2016; Kosek et al., 2021). Helitrons have also been described in other species but in lower abundance (Poulter et al., 2003). As previously discussed, Helitrons are replicated by a rolling circle mechanism (RCR) (Khan, 2005; Ruiz-Masó et al., 2015). In a large study conducted in 27 plant genomes (Xiong et al., 2016) it was described that Helitrons were found in tandem repeat arrays in all analyzed species, a configuration predicted by the RCR transposition model. This has also been observed again recently in the wheat genome (Wang et al., 2022b). The number of Helitrons in a tandem array varied in these genomes from a few to hundreds of copies in the case of rice genomes. In particular, it has been observed that this tandem array arrangement occurred mainly in the centromeric regions, intercalated between retrotransposons and satellite repeats (Zattera and Bruschi, 2022). This tandem repeat configuration of Helitrons in centromeric positions, described in plants, is exactly what was discovered in the present study in the flatfish S. senegalensis. This suggests that this distribution is favorable in the evolution of centromeres in eukaryotes. The maximum number of repeats observed in plants, within each array, was described in rice, with more than 150 copies (Xiong et al., 2016) and in wheat (Wang et al., 2022b). In the case of the centromere of chromosome 1 of S. senegalensis, this number is much higher. Oddly, this arrangement of long tandem Helitron arrays does not seem to occur in other plants, like maize, even when more than 80% of the maize genome is composed of transposons (Schnable et al., 2009). In addition, unlike what is observed in plants, the Helitrons of S. senegalensis that are found within the same array are not always in the same direction, as has been observed in the centromeric region of chromosome 1. This could indicate that the different arrays observed in chromosome 1 come from insertions produced at different times in the evolution of the centromeric region of this chromosome.

Interestingly, the Helitrons analyzed in plants presented internal repeats, both isolated Helitrons and tandem repeats (Xiong et al., 2016). In fact, of 1616 Helitrons observed in maize, rice, and Arabidopsis, 81.8% of them present internal repeats (Xiong et al., 2014). This characteristic has also been observed in the main families of Helitrons that form part of the long tandem arrays of chromosomes 1 and 2 of S. senegalensis.

In Actinopterygian fishes, different divergence profiles have been observed (Chalopin et al., 2015a; Sotero-Caio et al., 2017; Shao et al., 2019). In general, transposition explosions occur at least once or twice, if not more, throughout the evolutionary history of a fish. In this process, there is a continuous increase in the number of active transposons, before the explosion event, after which there is a decrease in the number of these active transposons. In most fish genomes, the rate at which the number of active transposons increases is lower than the rate at which it declines, so most fish have fewer ancient copies (K-values > 25) than recent copies (K-values < 25) (Chalopin et al., 2015a; Shao et al., 2019). Recent studies have shown different situations depending on the species in flatfish (Lü et al., 2021). Thus, in most of them, the divergence profiles have mainly shown ancient activity periods in almost all the analyzed species, except in Pseudorhombus dupliocellatus, Platichthys stellatus where additional recent transposon activity peaks were observed. In Trinectes maculatus, only recent explosion processes are observed (Lü et al., 2021). In S. senegalensis, the main peak of divergence is between 12-14%, taking into account all TEs, although there are significant differences between different superfamilies studied. Important differences in their TE profile have been described between evolutionarily close species. Thus, in Japanese and European eels, there are many differences in the evolutionary history and explosions in R2 and Helitrons transposons, respectively (Shao et al., 2019). Among African cichlid species, which generally have two events of explosions of all their superfamilies, a recent explosion has been observed in the species Maylandia zebra (Shao et al., 2019).

The life cycle of a TE goes through periods of activity and inactivity. The process begins with the invasion of a TE into a new genome (through horizontal transfer events) or by the evolution through mutation of a new lineage from an existing one (Shao et al., 2019). The insertion of TEs into genomes generates a series of host responses to prevent their expansion through the genome. However, if the insertion favors the host in some way, the TE will be conserved and a process of coevolution of the element with the host will occur (Kidwell and Lisch, 2001; Hua-Van et al., 2005). One of the most relevant characteristics of the ray-finned fish mobilome is the presence of more recent TE copies than those observed in other vertebrates, in particular fugu, cod, and stickleback present very recent copies. Between closely related species such as medaka and platyfish, differences in TE activity have been identified (Carducci et al., 2020). In S. senegalensis, two points of recent activity are also observed (K-value < 25), although a much more recent explosion of TEs is observed, not very abundant, with values close to 5% divergence. This has been observed in species such as spotted gar, Tetraodon, or Tilapia (Chalopin et al., 2015a), as well as in Haplochromis burtoni, Neolamprologus brichardi, Oreochromis niloticus, or Pundamilia nyererei (Shao et al., 2019). However, in other fish species, different patterns are observed, with a higher abundance of recent TEs (K-value <10) in species such as zebrafish, cod, stickleback, medaka, or Fugu (Chalopin et al., 2015a). Especially important is the recent TE activity event experienced by Maylandia zebra, with a peak of explosion with K-values of 1-2% (Shao et al., 2019). On the other hand, it has also been observed in fish the absence of recent activity as in platyfish, European eel, Latimeria chalumnae, or Callorhinchus milli (Chalopin et al., 2015a; Shao et al., 2019).

In the study of divergence by TE types, we have described in the present work important differences between them in the genome of S. senegalensis. DNA transposons, present greater divergence than the rest of the elements, with no recent activity signals. However, the rest of the TEs show different recent explosion activities, being the majority in the case of LINEs and LTRs, and similar abundance of ancient and recent elements both in Helitrons and in SINEs ( Figure 7 ). However, in previous studies of TE abundance and divergence in other flatfish species, no differences in divergence profile were observed between the studied TE categories. Therefore, with the general coverage and divergence data of TEs in both S. senegalensis described in the present work and other flatfish species described previously (Lü et al., 2021), it can be affirmed that repetitive sequences constitute a considerable portion of the genomes of this group of fish, and that the variety of genome sizes among flatfish can possibly be attributed to the expansion of these repetitive sequences in the genomes after the divergence of these species (Lü et al., 2021). Regarding other fish groups, with the exception of gar fish, most teleosts have modeled their genomes with DNA transposons. This occurs especially in the zebrafish, which shows the highest amplification of DNA-transposons among vertebrates. LINEs have contributed significantly to the genome of species such as fugu, tilapia, and medaka, while a middle-aged explosion of LTR elements has been detected in Tetraodon. In pufferfish, zebrafish, stickleback, and tilapia, a high number of recent copies have been described. In lamprey, many recent copies of DNA transposons can be identified (Chalopin et al., 2015a). In the case of Anguilla japonica and Anguilla anguilla, important differences in the activity of the Helitrons have been observed, with much greater divergence and abundance of these elements in A. anguilla (Shao et al., 2019). DNA transposons in species such as H. burtoni, N. brichardi, O. niloticus, or P. nyererei, present recent activity events, but not ancient copies, as in the rest of the TEs of these species, so it seems that there is a purging process of these elements in these genomes (Shao et al., 2019).

The analysis of the landscape in S. senegalensis by chromosomes showed that there are two important moments of transposon activity in the genome of some types of TEs, where insertions of different ages have been observed in many chromosomes but not in all. This occurs again, mainly in Helitrons ( Supplementary Figure 7 ). Of the two TEs explosion events, one of them has occurred recently from an evolutionary point of view, because the divergence values are around 5-7% (Kimura distance). The study of TE activity by chromosomes has not been carried out to date in any species, so there is no data to be able to make a comparison. However, it can be deduced that TEs burst events generally affect the entire genome, and in the case of Helitrons, in certain positions such as centromeres, the replication of the insertions and the maintenance of them are preferred because they are evolutionarily favored (Birchler and Presting, 2012; Sultana et al., 2017). It is worth noting that in chromosomes such as the 15, there are no recent copies, so they only contain ancestral copies with a higher degree of divergence, from ancient explosion events.

The analysis of Helitron families carried out reflects two large groups of well-differentiated families ( Figure 8 ). One of these clusters contains very few families, all of which have a small genetic distance. In this cluster, two subgroups can be observed, each with very little differentiation between them. One of these groups includes the family located as long tandem series in the centromeric region of chromosome 1. This cluster could be related to those Helitron families that contain more tandem copies, where the divergence is small because it could be part of a recent burst event of this transposon. The presence of another cluster with families with greater divergence between them could be reflecting isolated copies throughout the genome, or with few rounds of tandem replication. These copies have had more time to evolve, and therefore reflect longer branches, or, on the other hand, not have a selective pressure because they are not so involved in the function of the centromere. When the analysis of chromosomal insertions has been carried out, labeling the copies by color according to their origin, it can be observed that different chromosomes actually present several families with different numbers of insertions, but that not all families are in all chromosomes. This is well seen in chromosomes such as 2, 3, and 9 ( Figure 9 ), among others, where there are clearly copies belonging to certain clusters that are not present in them. Again, this indicates that there is a clear divergence between Helitron copies in the genome of S. senegalensis, and that their distribution is not random. The phylogenetic study of Helitrons also addressed the evolution of their insertions in different genomes. The analyses presented interesting results that reflect the existence of clusters of Helitron copies that are evolutionarily close (short branches in the same cluster) belonging to species as distant as S. aurata (Sparidae) and S. maximus (flatfish) ( Figure 10 ). There are also observed events of burst of these Helitrons in S. senegalensis, shared with sequences of Helitrons from S. aurata or S. arureovittata, or clusters with sequences that are almost exclusive to S. senegalensis (Family Soleidae) and H. hippoglossus (Family Hippoglossidae). These data could be explaining two possible evolutionary processes. One of them is the selection of certain families and their copies throughout the evolution not only of flatfish, but also of intermediate species to this order, such as S. aureovittata, or as far away as S. aurata. However, there are no studies on the distribution of these elements along their chromosomes. If these copies with little divergence between species so distant were located in the centromeric regions, as occurs in S. senegalensis, it is probable that these sequences would be maintained by evolution, within the protection that these genomes would be carrying out on these centromeric Helitrons. Another explanation that is not to be ruled out is the existence of horizontal transfer (HT) events between the families of Helitrons that present high homology between these distant species. Although there are various ways to study possible transposition events, the most general one predicts that in TE families where HT events occur, there should be major inconsistencies between the TE family phylogeny and that of its hosts (Hartl et al., 1997; Schaack et al., 2010). Although we have not performed their specific analysis in the present study, these inconsistencies are clearly observable from the genomic study carried out.

In contrast to the knowledge that exists around horizontal HT in the evolution of prokaryotes, the evolutionary importance of HT remains more obscure (Frost et al., 2005; Schaack et al., 2010; Merlo et al., 2012). These differences can be attributed in part to the disproportionate attention that has been given to the transfer of genes, as opposed to non-coding DNA. TEs are not only the most abundant elements in eukaryotic genomes, but they are also one of their most dynamic components. Today, HT of transposable elements (HTT) is considered a relevant mechanism in the modeling of eukaryotic genomes. Both DNA transposons and retrotransposons can be horizontally transferred, and HTT can involve a wide variety of eukaryotic lineages that can transfer TEs between closely or distantly related lineages (Dotto et al., 2015). Additionally, a significant number of TEs are known to have induced important phenotypic changes to their host that have been acquired through HT, thus establishing HTT as a source of variation that feeds adaptive changes (Gilbert and Feschotte, 2018).

Recently, the study of more than 300 vertebrate genomes has shown a minimum of 975 independent events of HTT between lineages that diverge more than 120 million years (Zhang et al., 2020). Of these events, more than 90% (93.7%) have occurred in ray-finned fishes, and less than 3% in mammals and birds. These HTT events occur not only between fish but also between fish and amphibians or birds. The majority of the events recorded in ray-finned fishes involve DNA transposons, specifically the Tc1/Mariner superfamily (Zhang et al., 2020), although other superfamilies belonging to the retrotransposons such as BovB and L1, have also shown significant HTT events in marine eukaryotes (Ivancevic et al., 2018). On the other hand, the Pacific oyster (Crassostrea gigas), the cactus worm (Prriapulus catus) and the marine worm (Saccoclossus kowalevskii) have been described as potential vector species in HTT cross-Phylum events involving marine eukaryotes (Ivancevic et al., 2018). In the case of Helitrons, it has been shown that this transposon has been frequently transferred horizontally in insect genomes (Thomas et al., 2010). However, additional cases of HTT of Helitrons have been identified in vertebrates such as lizards, jellyfish, or jawless fish (Thomas et al., 2010). It has been suggested that ray-finned fishes could be part of an environment that includes both organisms and environments that are particularly susceptible to exchanging TEs, such as viruses and other parasites (Loreto et al., 2008; Gilbert and Cordaux, 2017). Interestingly, the species S. aurata, in which we have described in the present work high similarity between its copies of Helitrons and those of S. maximus, had already shown a possible HT event of the 5S rDNA gene with the toadfish Halobatrachus didactylus (Merlo et al., 2012). This would support the fact that HT processes between marine organisms, in this case between fish, are more frequent than previously studied and described.

The current work introduces novel genetic resources that have broadened our understanding of the abundance, distribution, and evolutionary patterns of repetitive sequences in flatfish species. We have discerned variations in the content of distinct subclasses of transposable elements across eight flatfish species. Specifically, we have deepended into the study of repetitive elements in the S. senegalensis sole genome and their evolution, by examining the divergence of the predominant types of TEs, identifying two burst events of the majority of these elements. We have unveiled an intriguing new discovery of a genomic structure involving tandem repeat insertion arrangement of Helitron families in the pericentromeric regions of the S. senegalensis genome. This discovery, previously identified solely in mammals and plants, significantly augments our knowledge of genome architecture and transposon-mediated evolutionary processes in flatfish. Furthermore, our phylogenomic analysis of Helitron insertions in flatfish and other external species has yielded intriguing results, reflecting the existence of evolutionarily proximate Helitron copy clusters belonging to species as distant as S. maximus (flatfish) and S. aurata (Sparidae). All the findings bear significant implications for our understanding of the chromosomal evolution in S. senegalensis and other studied flatfish species. This taxonomic group holds paramount importance due to its global economic relevance and its remarkable adaptation to benthic life. Of particular interest is the role that transposable elements (TEs) have played in shaping the current chromosomal architecture within this group of fishes.

The original contributions presented in the study are included in the article/ Supplementary Material . Additional TEs annotation dataset is deposited in the Figshare repository and available in https://figshare.com/search?q=10.6084%2Fm9.figshare.25239952. Further inquiries can be directed to the corresponding author.

Ethical approval was not required for the study involving animals in accordance with the local legislation and institutional requirements because This is bioinformatic work from public data bases.

IC: Conceptualization, Writing – original draft, Resources, Investigation, Methodology. MR: Investigation, Resources, Writing – review & editing. SP-B: Investigation, Writing – review & editing. MM: Investigation, Writing – review & editing. AG-S: Investigation, Writing – review & editing. RN-P: Writing – review & editing, Methodology, Resources, Supervision, Writing – original draft. LR: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – review & editing.