Transposon diversity is higher in amphioxus than in vertebrates: functional and evolutionary inferences Cristian Can estro and Ricard Albalat

Transcription

1 BRIEFINGS IN FUNCTIONAL GENOMICS. VOL 11. NO ^141 doi: /bfgp/els010 Transposon diversity is higher in amphioxus than in vertebrates: functional and evolutionary inferences Cristian Can estro and Ricard Albalat Advance Access publication date 2 March 2012 Abstract Transposable elements (TEs) are main components of eukaryote genomesçup to 50% in some vertebratesçwhich can replicate and jump to new locations. TEs contribute to shape genome evolution, actively by creating new genes (or exons) or altering gene expression as consequence of transposition, and passively by serving as illegitimate recombinational hotspots. Analysis of amphioxus TEs can help to shed light on the ancestral status of chordate TEs and to understand genome evolution in cephalochordates and early vertebrates. The Branchiostoma floridae genome project has revealed that TE content constitutes 28% of the amphioxus genome. Amphioxus TEs belong to more than 30 superfamilies, which represent a higher diversity than in vertebrates. Amphioxus TE families are also highly heterogeneous as generally none of their members are drastically more abundant than others, and none of the TEs seems to have suffered any massive expansion. Such diversity and heterogeneity make the amphioxus genome not to be particularly prone to major evolutionary changes mediated by TEs, and therefore favoring genomic evolutionary stasis. Comparison of TE diversity and content between amphioxus and vertebrates allows us to discuss whether or not a burst of TEs happened after the two rounds of whole-genome duplication that occurred during early vertebrate evolution. Keywords: Amphioxus cephalochordates; Transposable elements; Chordate and vertebrate evolution; Two rounds of whole-genome duplication; DNA methylation; Genome stasis INTRODUCTION Transposable elements (TEs) are DNA segments able to jump into new genomic locations, process known as transposition, sometimes replicating during the mobilization and reaching very high copy numbers that can represent a major fraction of the genome (up to half of the genome such in humans or platypus [1, 2]). According to the mechanisms of transposition, TEs are grouped in two classes: retrotransposons and DNA transposons [3, 4]. Retrotransposons or Class I elements transpose via an RNA intermediate that is reverse transcribed into the genome. Retrotransposons are classified in different groups, including the long-terminal repeat (LTR) retrotransposons, the tyrosine-recombinase retrotransposons (YR or DIRS-like elements), the non-ltr retrotransposons (or LINE-like elements) and the Penelope-like elements (PLE) [4, 5]. On the Corresponding author. Ricard Albalat, Departament de Genètica, Facultat de Biologia and Institut de Recerca de la Biodiversitat (IRBIO), Universitat de Barcelona, Edifici Prevosti, Av. Diagonal, 643, E Barcelona, Spain. Tel. þ ; Fax. þ ; ralbalat@ub.edu Cristian Can estro is an Assistant Professor at the Department of Genetics, in the University of Barcelona, Spain, since He made his postdoctoral research in the Institute of Neuroscience at the University of Oregon ( ). He received his PhD from the Department of Genetics, University of Barcelona, Spain (2001). He is interested in Genomics, Evo-Devo and Epigenetics, and his research focused on the impact of gene losses on developmental programs. His projects aim to understand the evolution of chordates and the origin of vertebrate innovations. Amphioxus, ascidians, larvaceans and zebrafish are the main model organisms in which he studies his favorite signaling molecule, the retinoic acid, and its epigenetic role on embryo development. Ricard Albalat, PhD (Universitat de Barcelona) and Postdoc (Skirball Institute; New York University), is an Associate Professor of Genetics at the Universitat de Barcelona. His current research is focused on the evolution of the retinoid and steroid signaling pathways during chordate evolution. He is also interested in the impact of transposable elements and the changes of epigenetics mechanisms in the evolution of chordate genomes. ß The Author Published by Oxford University Press. All rights reserved. For permissions, please journals.permissions@oup.com

2 132 Can estro and Albalat other hand, DNA transposons or Class II elements move through a double-strand DNA intermediate and comprise the original cut and paste TEs discovered by Barbara McClintock over 60 years ago [6], the rolling-circle DNA transposons (Helitrons) and the self-synthesizing DNA transposons (Polintons-Maverik) [4, 7]. Both classes of TEs can be categorized in autonomous or nonautonomous elements. A transposable element is defined as autonomous if it belongs to a group of TEs that encode all the domains that are typically necessary for its mobilization, without implying that the element is either functional or active [4]. Autonomous TE families can include defective elements due to a limited number of mutations or indels, but still retain sufficiently high DNA identity with other functional elements. In contrast, nonautonomous elements lack some (or all) domains necessary for transposition, although in some cases can be still active by borrowing the transposition machinery from autonomous TEs [4]. Short interspersed nuclear elements (SINEs) and miniature inverted repeat transposable elements (MITEs) are among the nonautonomous classes I and II TEs, respectively, that can reach very high copy numbers in eukaryote genomes [8, 9]. Theoretical considerations and empirical approaches had led to suggest that TEs could be conceived as mere molecular selfish parasites that fill up genomes, owing their survival to their capability to replicate faster than the accumulation of mutations that would erase them from the genome. The amplification of TE copy number is not unlimited, however, because selective negative forces restrict their expansion when this becomes detrimental for the proper function of the host genome. Nowadays, TEs are not perceived anymore as selfish DNA entities, but as the most dynamic structural components of genomes that act as one of the main players facilitating and driving the evolution of each species genome (for recent reviews [10 13]). TEs can be substrates for unequal and illegitimate recombination and can be thereby responsible for several genome reorganizations, including chromosomal deletions, inversions, translocations and duplications. It has been described that polyploidization can trigger genomic stress associated with drastic genomic rearrangements, in many cases mediated by bursts of mobilizations of TEs [14, 15]. Lineage-specific genome rearrangements mediated by TEs might facilitate rapid evolution, reproductive isolation of different populations and, consequently, species diversification [16]. Hence, to fully understand the evolution of a given group of organisms, it is necessary to know the functional and structural complexity of their genomes in terms of TE content, diversity and relative abundance of TE families, TE distribution throughout the genome, and the genomic traces that might reflect the dynamic of past or recent TE mobilization. In this review, we focus on the TEs of the amphioxus Branchiostomafloridae, which belongs to the cephalochordate subphylum, the most basally divergent group of our own phylum, the chordates. Recently, the genome of amphioxus has been fully sequenced [17, 18]. The analysis of this genome has been crucial for understanding the evolution of chordates, and the origin and successful diversification of vertebrate features, probably facilitated by two rounds of whole-genome duplication (2R-WGD) that took place in stem vertebrates after their divergence from non-vertebrate chordates (i.e. urochordates and cephalochordates) [17, 18]. We review the current knowledge on the TEs in amphioxus, analyzing their diversity, abundance, mobility and other specific features that characterize the TE families of amphioxus, and we evaluate how these features might have shaped the evolution of the amphioxus genome. Comparison of the TE content and diversity between amphioxus and vertebrates allows us to discuss possible trace evidence of a burst of TE mobilization that could have happened after the 2R-WGD, and the evolutionary dynamics of TEs during early vertebrate evolution. TRANSPOSABLE ELEMENTS IN AMPHIOXUS Transposable elements have been found in most eukaryotic species investigated so far [19], and amphioxus is not an exception. More than 700 families of TEs constitute about 28% of the 520 Mb of the amphioxus genome, from which 9% is occupied by class I elements, 15% by class II and 4% by still unclassified TEs [17] (Table 1). Class I transposable elements The first TE identified in amphioxus was a non-ltr retrotranposon named Bf-CR1 (B. floridae Chicken Repetitive 1 element) [20, 21]. Analysis of the sequence of the Bf-CR1 element showed the presence of an ORF that coded for an apurinic/apyrimidinic endonuclease domain and a reverse transcriptase

3 Amphioxus Transposable Elements 133 Table 1: Transposable elements in the amphioxus genome TE copy number (total/in the most abundant family) Number of families (relative abundance of the biggest family) References Class I retroelements: autonomous LTR retrotransposons BEL ND ^ [17] Copia 137/80 3 [17, 38] ERV ND ^ [17] Gypsy/Troyka 19/16 (Gypsy) 2 [17, 39] 131/31 ( Troyka) 5 YR retrotransposons DIRS ND ^ [17] Non-LTR retrotransposons (LINE-like) CR1/L / (35%) [17, 21, 40^ 42] Crack / (7%) [42, 43] I/Vingi/Outcast 125/125 (I) 1 [17, 40, 42, 44^ 46] 709/318 (Vingi) 3 734/547 (Outcast) 2 Jockey ND ^ [17, 31] LOA ND ^ [17, 40, 44] L1/Tx1 2800/ (13%) [17, 40, 47] NeSL/Hero 403/155 3 [17, 40, 42, 48] Proto2 302/302 1 [42, 49] REX1 312/312 1 [50] RTE/RTEX 8357/1236(RTE) 19 (15%) [17, 40, 42, 51, 52] /3604 (RTEX) 16 (12%) Penelope-like Penelope ND ^ [17] Non-autonomous SINEs Amphi^Alu ND ^ [25] BflSINE / [26] Class II DNA transposons: autonomous cut and paste TEs Academ 21594/ [27] CMC (CACTA-Mirage-Chapaev) 419/346 Chapaev 2 [17, 27, 53, 54] 6372/1903 EnSpm 8 Ginger 4029/ [55] hat / (16%) [17, 27, 56, 57] IS4eu 757/757 1 [58] Kolobok 290/205 2 [17, 27, 59] Merlin ND ^ [27] MULE (MuDR/Rehavkus) 5008/ [17, 27, 60] P ND ^ [17, 27] PIF/Harbinger 61480/ (15%) [17, 27, 31, 61] PiggyBac 11262/ (27%) [17, 27] Sola-2 ND ^ [27, 62] Sola / [27, 62] Tc/mariner 4438/ [17, 27] Zator 575/575 1 [62] Rolling-circle transposons Helitrons ND ^ [17] Self-synthesizing transposons Polintons ND ^ [17] Non-autonomous MITEs ATE-1 ND ^ [28] LanceleTN ^ [29] LanceleTN ^ [29] LanceleTN-3a, -3b ^ [29] LanceleTN ^ [29] LanceleTN-5 ND ^ [31] LanceleTN-6 ND ^ [31] BelcheriTn-1 ND ^ [30]

4 134 Can estro and Albalat domain, characteristic of non-ltr retrotransposons [20]. Phylogenetic analysis using the reverse transcriptase domain revealed that this amphioxus TE belonged to CR1 retrotransposon superfamily [20], which is present in most animals so far analyzed [22]. Recently, sequence analysis of the whole genome of amphioxus has shown that indeed CR1 non-ltr retrotransposons form the most diverse superfamily of amphioxus TEs, including more than 100 families that occupy 2.5% of the genome [17]. Elements show high similarities within families, as observed by a 98% identity between Bf-CR1 and CR1-11_BF elements, but in general, TEs show significant sequence diversity between families, <75% identity [17]. Evidence of TE sequence diversity has been provided by experiments of Southern blot, in which the Bf-CR1 element from the presinilin locus hybridized with some Bf-CR1 copies (approximately 15 copies, most likely of the CR1-11_BF family), but did not cross-hybridize with the thousands copies of other CR1 families identified by in silico analysis of the amphioxus genome [17, 21]. Evidence supporting Bf-CR1 mobilization has been inferred from the variability of Southern blot patterns among different animals, as well as by the cloning and sequencing of various presenilin alleles that differ by the presence or absence of the Bf-CR1 element [21]. TE mobilization, together with instability of other repetitive DNA sequences such minisatellites, might account for a significant fraction of the high polymorphism observed in the amphioxus population [23]. Besides CR1 elements, the sequencing of the amphioxus genome has identified additional class I TEs, including several non-ltr retrotransposons: RTE/RTEX (0.4% of the genome), L1 (0.3%), I / LOA (0.1%), Jockey (1.3%), NeSL/Hero (<0.01%), REX1 (<0.01%), Crack (<1%) and Proto2 (<0.05%); some LTR retrotransposons: Gypsy (0.3%), BEL (0.1%), Copia (0.1%) and ERV (0.1%); few YR retrotransposons: DIRS (0.1%); and some Penelope-like elements (1.9%) (Table 1 and Figure 1; information about amphioxus TE content was provided by Jerzy Jurka, after running CENSOR program against an updated inventory of amphioxus TEs in Repbase Update: [24]). Within class I, non-ltr retrotransposons are, therefore, the most abundant retrotransposons, contributing to 4.6% of the amphioxus genome, whereas LTR retrotransposons represent only 0.6%. The amphioxus genome also has families of nonautonomous retrotransposons that rely on the transposition machinery from autonomous TEs for mobilization. Among nonautonomous retrotransposons, SINEs are short DNA sequences (less than 500 bases) that do not encode a functional reverse transcriptase protein. The first amphioxus SINE reported in amphioxus, named Amphi Alu, was identified in the regulatory region of the FoxD gene and interestingly it contained AluI restriction sites such those described in Alu SINEs of primates [25]. The lack of sequence similarity between Amphi Alu and primate Alu elements, however, suggested that both Alu elements evolved independently [25]. The structure of the Amphi Alu is characterized by a 375-bp sequence related to trnas with a putative RNA polymerase III promoter and a polya tail of variable length [25]. The recent availability of the sequence of the whole genome has revealed high similarity (97%) between Amphi Alu and the 3 0 region of RTEX non-ltr retrotransposons, opening the possibility that Amphi Alu could be a truncated RTEX element, or alternatively that an Amphi Alu could have been captured by an RTEX element (L. Holland, personal communication). Evidence of mobilization of Amphi Alu has been inferred from the fact that different allelic variants of the FoxD locus differed by the presence or absence of the Amphi Alu element in their regulatory regions [25]. In addition to Amphi Alu, a second SINE has been described in amphioxus [26]. This SINE, named BflSINE1, is 383 bp long, it is also a trna-derived element, and it might represent 1 2% of the genome. BflSINE1 belongs to the evolutionary conserved DeuSINE superfamily with members in several deuterostomes, including mammals, chicken, coelacanth, zebrafish, catfish, salmon, dogfish shark, hagfish and sea urchin [26]. Class II transposable elements In silico analysis of amphioxus genome has revealed that 15% of the genome is occupied by class II TEs. Class II TEs are characterized by the presence of nt terminal inverted repeats (TIRs) flanking a central region that encodes for a transposase. The amphioxus genome has a great variety of class II TE, including the three main types, namely, cut and paste (13.5% of the genome), rolling-circle (1.5%) and self-synthesizing (0.1%) DNA transposons (Table 1 and Figure 1) [17]. Among the cut and paste superfamilies amphioxus has members

5 Amphioxus Transposable Elements 135 Figure 1: Amphioxus TE content is highly diverse. Phylogenetic distribution of TEs in different animal lineages, in which gray and white boxes indicate presence and absence of TEs superfamilies, respectively. Information has been gathered from [27] and Repbase Update: [24].TE diversity is represented by the number of different superfamilies present in a given lineage. Amphioxus, with 33 superfamilies, shows a higher TE diversity than any vertebrate species analyzed so far (10 ^28 superfamilies). representing 14 of the 17 known families, which reveals that the amphioxus genome has maintained a high TE diversity since its split from the last common chordate ancestor [17, 27]. Thus, amphioxus contains cut and paste transposons Academ (2.4%), CMC (CACTA-Mirage-Chapaev) (0.3%), Ginger (0.3%), hat (1.2%), Kolobok (0.1%), Merlin (undetermined), MULE (MuDR- Rehavkus) (0.3%), P (0.3%), PIF/Harbinger (3.0%), PiggyBac (1.0%), Sola-2 (undetermined), Sola-3 (0.3%), Tc/mariner (0.4%) and Zator (0.1%), rolling-circle Helitrons (1.5%) and selfsynthesizing Polintons (0.1%) (Table 1 and Repbase Update: index.html [24]). Before the identification in silico of all the autonomous class II TEs in the genome of B. floridae, there had been already reported a nonautonomous class II TE named Amphioxus Transposable Element 1 (ATE-1) [28]. ATE-1 was found in introns of the Adh3 locus in two amphioxus species, B. floridae (BfATE-1) and Branchiostoma lanceolatum (BlATE-1) (Figure 2). The ATE-1 structure consisted of a DNA segment flanked by two 16 nt TIRs and target site duplications (TSDs) of 7 nt, with adjacent direct repeat (DRs) of about 100 nt, and a central region of variable size, but lacking any ORF that could code for a transposase [28]. These features of the ATE-1 led to classify it as nonautonomous class II composite MITE (Miniature inverted repeat TE). Simple MITEs are small sequences (usually less than 600 bp) with TIRs and TSDs, forming stable secondary structures but without coding potential. ATE-1 was composed of two MITEs, namely DRa of 109 bp and DRb of 93 bp, encompassing a central region in which later a novel MITE named LanceleTn-2 was identified (see below) [29]. ATE-1 elements were also found in different loci (e.g. EF-1a and Tip20 genes besides Adh3 gene), and copy number estimations by Southern or slot blot suggested less than 10 conserved copies per haploid genome [28]. After ATE-1 was described, many other MITEs were identified in amphioxus [29 31]. Five MITEs (LanceleTN-1, -2, -3, -3b and 4) were identified around NK homeobox genes and the ParaHox gene cluster [29]. LanceleTN-1 has 21 bp TIRs, 8 bp TSDs, and it was tentatively classified into the hat (hobo, Ac, Tam3) superfamily. LanceleTN-2 has TIRs up to 59-bp long, TSDs made of the dinucleotide TA and apparently is not related to any known autonomous element. LanceleTN-3a and 3b have 21 and 19 bp TIRs, respectively, 9 bp TSDs, and were classified into the Mutator-like superfamily. LanceleTN-4 has TIRs of up to 53 bp long, TSDs of the dinucleotide TA and remains unclassified [29]. Two additional MITEs, LanceleTN-5 and LanceleTN-6, were later identified [31]. LanceleTN-5 has 164 bp TIRs, 8 9 bp TSDs, and was related to Merlin superfamily, while LanceleTN-6 has 33 bp TIRs, 9 bp TSDs, and was related to Mutator-like superfamily. The abundance of MITEs has been estimated around 1.7% of the genome of amphioxus B. floridae and, in average, there is 7.3 MITEs per 100 kb [31], suggesting that

6 136 Can estro and Albalat Figure 2: Convergent insertion of TEs in the same locus occurred independently in two amphioxus species, Branchiostoma lanceolatum and Branchiostoma floridae. BlATE-1 element, which is a composite-mite that contains two MITES (DRa and DRb) and a lanceletn-2 element, was inserted within a minisatellite in intron 9 of the Adh3. BfATE-1, which is a mini-ate1 that does not contain the lanceletn-2 element, was inserted within a mirage-minisatellite in intron 8 of the Adh3 [28, 29]. The fact that the insertion occurred within a minisatellite structure in both species suggests that regions rich in tandem repeats could be hotspots for TE insertions. MITEs are abundant TEs in the B. floridae genome. Interestingly, MITEs have been found not only in B. floridae but also in other amphioxus species, e.g. B. lanceolatum (BlATE-1) [28] and B. belcheri (BelcheriTn-1) [30]. Their presence in three Branchiostoma species suggests that MITEs are widespread in the genus. Comparison of BlATE-1 and BfATE-1 MITEs, both inserted in the Adh3 locus of B. lanceolatum and B. floridae, respectively, reveals an interesting case of recent independent insertions in the same locus [32] (Figure 2). Both TEs transposed into tandem repeat minisatellites and mirage minisatellites located in different introns at the 3 0 -end of the Adh3 gene (Figure 2). The independent transposition of the same type of TE at the same locus could be a fortuitous event or, alternatively, it could suggest that there may be hotspots for TE insertions, and minisatellites appear to be potential candidates for increased TE insertion rate. In addition, the fact that the same gene has gained independently the same type of TE raises the possibility of beneficial functional effects of the TE insertion that could have been positively selected in both amphioxus species. AMPHIOXUS TES:FUNCTIONAL AND EVOLUTIONARY INFERENCES Amphioxus TEs and their impact on gene evolution Transposable elements can generate genetic novelty actively by adding new coding regions via alternative splicing in a process termed exonization [33], or by incorporating cis-regulatory elements to the genes neighboring the insertion sites [34 37]. Transposition, therefore, has a clear mutagenic activity, sometimes beneficial by promoting functional innovations due to an increase of protein versatility or to changes in the expression of genes, and sometimes negative by causing detrimental effects due to the alteration of the normal function of genes [34 36]. Although the beneficial or detrimental impact of TEs on gene evolution in amphioxus is mostly unknown, analyses of some genomics regions have provided some hints. The Amphi Alu, for instance, contains binding sites of transcription factors that are essential for notochord expression and, interestingly, Amphi Alu has been found in the promoter of FoxD and in the nearby of other genes also expressed in the notochord such Tbx15/18/21 and Pax1/9 [25]. Although the role of Amphi Alu in the regulation of these genes is unclear, it is possible that Amphi Alu may affect gene expression by altering cis-regulatory regions [25]. A second example in amphioxus in which a gene acquires a new functional domain comes from the analysis of the Dkk3 gene, which seems to have recruited a new domain from a TGF-b protein that provides new inhibiting functions to Dkk3 (L. Holland, personal communication). Further analyses are needed to understand the origin of this newly acquired domain, but it is possible that the recruitment of this domain resulted from a transposition event as a Jockey-5 TE is found in the nearby (at a 135-bp distance from the 3 0 -end)

7 Amphioxus Transposable Elements 137 of the Dkk3 according to the current genome assembly v2. In contrast to the beneficial effects of TEs shaping gene expression or co-opting new protein domains, some genomic regions appear to be refractory to TE insertions. Analysis of the genomic region of the Hoxgene cluster has shown that the content of TEs in the cluster was smaller than expected [31]. On the other hand, the ParaHox-gene cluster was indeed a hotspot for TE insertion [29, 31], suggesting a significant difference between Hox- and ParaHox-gene clusters in amphioxus, and clearly indicating that not all the genomic regions are equally suitable for TE invasion. Diversity and heterogeneity of amphioxus TE families The genome of B. floridae is characterized by its high diversity and heterogeneity of TEs. The diversity of amphioxus TEs, comprising 33 superfamilies [17], is higher than in any vertebrate species so far analyzed (Figure 1). In fact, several class I (i.e. Crack, Jockey, LOA and Proto2) and class II (i.e. Sola3 and Ginger) TE families are present in amphioxus, as well as in other deuterostomes and protostomes, but absent so far in vertebrates (Figure 1 and [27]). Amphioxus TE families are heterogeneous in the sense that overall none of the families within each superfamily are exceptionally abundant in comparison with others. From the approximately TE sequences in the amphioxus genome, 95% of them belong to families with less than 5000 copies, while only three autonomous TE families (i.e. Academ-2, CR1-1 and Sola3-3) have more than copies (Figure 3A and Table 1). These three families appear probably as the only cases in which a moderate expansion of a particular set of TEs has occurred, reaching copies for Bf_Academ-2, copies for Bf_CR1-1 and copies for Bf_Sola3-3, while the rest of Academ, CR1 and Sola3 families have an average of , and copies, respectively. Within the Harbinger superfamily, which is the most abundant TE superfamily (approximately copies), none of the 27 Harbinger families accounts for >15% of the total number of Harbinger elements (e.g. the most abundant family, Harbinger-N1B, has approximately 9000 copies) (Figure 3B), and the same situation occurs in other highly heterogeneous superfamilies, including Crack (32 families), hat (20 families), RTE (19 families), RTEX (16 families) and Tx1 (17 families) (Figure 3B and Table 1). How the diversity and heterogeneity of the TE content can have affected the evolution of the amphioxus genome? The amphioxus genome appears as the opposite situation of many vertebrates, in which recent bursts of different TE families have led to homogenous populations of highly abundant TEs (e.g. chicken: CR1 copies representing 3.1% of the genome; bat: HeliBat copies and nhat copies representing 3 and 1.3% of the genome, respectively; human: more than L1 copies representing 17% of the genome; platypus: 1.5 million LINE2 copies and 2.1 million Mon1 copies representing 19 and 21% of the genome, respectively) [1, 2, 19, 63, 64]. In vertebrates, it has been suggested that the low diversity of abundant homogenous populations of TEs provided homologous sequences that might have facilitated illegitimate recombination events, leading to chromosomal rearrangements, quick genetic differentiation, reproductive isolation of different populations and consequently, species diversification [65]. In the case of amphioxus, however, the high diversity of heterogeneous TE families and the apparent absence of any massive expansion of any particular TE (Table 1, Figures 1 and 3) probably make the amphioxus genome not to be prone to major evolutionary changes mediated by TEs, favoring genomic evolutionary stasis. Although the amphioxus genome has changed throughout evolution [66, 67], the idea of genomic evolutionary stasis is consistent with the idea of morphological stasis in the evolution of the cephalochordate lineage, in which different amphioxus species (e.g. B. floridae, B. lanceolatum and B. belcheri) are morphologically very similar despite they diverged Mya [32, 68]. This morphological conservation has been linked at the molecular level to stasis in the expression patterns of key developmental genes [69], supporting the view of general conservation of the amphioxus lineage. Future genome sequencing of additional cephalochordate species from different genus will allow testing whether TE diversity and heterogeneity correlate with genomic stasis in the entire cephalochordate subphylum. Transposable elements, 2R-WGD and vertebrate genome evolution In principle, a recently duplicated genome can be relatively tolerant of transposition because it contains many redundant genes, which buffer against insertional mutagenesis, and contains substantial

8 138 Can estro and Albalat Figure 3: Amphioxus TE families are remarkably heterogeneous. (A) Frequency distribution of amphioxus TE families according to their number of TE copies. About 95% of TEs belongs to families with less than 5000 copies, and only one nonautonomous (BflSINE1: 18737) and three autonomous TE families have more than copies (Bf_Academ-2: copies; Bf_CR1-1: copies and Bf_Sola3-3: copies). From left to right, columns represent number of families with <100 copies, from 100 to 1000, from 1000 to 5000, from 5000 to 10000, and > (B) Copy number (left axis) and relative abundance (right axis) for each TE family within superfamilies with more than 10 TE families or with at least copies (Table 1): Academ (4 families), CR1 (56 families), Crack (32 families), EnSpm (8 families), Harbinger (27 families), hat (20 families), LanceleTn (5 families), PiggyBack (11 families), RTE (19 families), RTEX (16 families), Tx1 (17 families) and Sola3 (3 families). The BflSINE1 superfamily has not been included in the graph because it only has one family (Table 1). (A). Information about amphioxus TE content was kindly provided by Jerzy Jurka, after running CENSOR program against the Repbase Update (RU) database ( org/repbase) [24]. repetitive DNA, which could serve as a sink for TEs [70]. According to this expectation, bursts of TE mobilization have been reported after polyploidization in different organisms [14, 15, 71 73]. The occurrence of 2R-WGD during early vertebrate evolution (the 2R-WGD hypothesis [74]) [17, 75] raises the question of whether there was a burst of TE mobilization after the 2R-WGD during early vertebrate evolution. Comparison of the diversity and content of TEs between amphioxus and vertebrates might help to answer this question, although the long time since the 2R-WGD after the amphioxus and vertebrate split hinders a definitive response. Expansion of the TE content within a genome cannot grow to infinity, and theoretical models suggest that the total number of TEs might reach equilibrium between TE mobilization activity, defense mechanisms of the genome (e.g. RNA interference, DNA methylation), and natural selection constraints preserving the functionality of the genome [76]. A model of TE competition predicts that the expansion of a particular type of TE might cause the reduction of other types of TEs, consequently reducing the TE diversity, until a new equilibrium is reached [76]. According to this competition model, if a burst of TE mobilization occurred after the 2R-WGD, we expect that the diversity of TEs

9 Amphioxus Transposable Elements 139 shared among vertebrates should be smaller than in cephalochordates. Consistent with this prediction, we observe that the diversity of TEs shared among vertebrates (15 superfamilies in mammals, 10 in birds, 14 in reptiles, 20 in amphibians, 28 in ray finned fishes and 14 in lampreys) is lower than in amphioxus (33 superfamilies) (Figure 1). Although this observation is not conclusive, the reduction of TE diversity in vertebrates is compatible with a burst of TEs after the 2R-WGD during early vertebrate evolution. We cannot discard, however, that the reduction of TE diversity in vertebrates might have been caused by independent extinctions of TEs in several vertebrate lineages, nonrelated, therefore, to a burst of TEs after the 2R-WGD. The diversity of retrotransposons in tetrapods, for instance, was significantly reduced before the split of mammals and birds [77]. The increased gene redundancy after a genome duplication event relaxes the selective constrains against insertional mutagenesis, and thereby allows the increase of TE content. If the 2R-WGD allowed a burst of TEs in the vertebrate ancestor, we expect that the content of TEs in vertebrates should be higher than in amphioxus. Genomic analyses, however, show that the amount of genome covered by TEs in amphioxus (28%) is similar or even higher than that in some vertebrate genomes (e.g. 30% in lizard, 10% in chicken, 26% in zebrafish or 7% medaka). Although the similar content of TEs in amphioxus and some vertebrates does not support an ancient TE burst after the 2R-WGD, we cannot discard the possibility that secondary reductions of the content of TEs in different vertebrate lineages might have erased the trace evidence of an ancient increase in the TE content. There have been described cases in which genome duplication in allopolyploid plants was not followed by a burst of TEs, and interestingly, it was accompanied by the recruitment of DNA methylation as a genome defense mechanism against TE activity [78, 79]. Remarkably, while TEs are heavily methylated and DNA methylation plays a defense role against TE mobilization in vertebrates [80], current data do not provide evidence supporting that TEs are major targets of methylation in non-vertebrate chordates [80 83]. It is, therefore, tempting to extrapolate this idea from the plant world to vertebrate evolution, and to speculate that the recruitment of the DNA methylation for the control of TE mobilization in vertebrates might have been crucial for restraining the mutational impact of TE proliferation. Probably, regulation of TE mobility was the first step before the molecular domestication or exaptation of TEs facilitated the emergence of gene novelties during vertebrate diversification. Overall, it is clear that amphioxus has a significant amount of TEs, and interestingly the diversity of TEs is higher in amphioxus than in vertebrates. Further analysis, however, needs to be performed to understand the regulation, mobility, and rates of expansion and extinction of TEs in cephalochordates, and how TE diversity and heterogeneity might have led to the evolutionary genomic stasis of this lineage. On the other hand, vertebrates seem to have recruited DNA methylation as a defense mechanism that controls TE mobility. It remains uncertain, however, whether a burst of TE actually happened after the 2R-WGD, because secondary bursts and extinctions of TEs occurred in different vertebrate lineages might have obscured such ancient event. Key Points Amphioxus TE content is considerably big (28%), even bigger than in some vertebrates. AmphioxusTE catalog, which includes members from more than 30 superfamilies, is more diverse than in vertebrates. Amphioxus TE families are highly heterogeneous, and no evidence of a massive expansion of any particularte has been found. The diversity and heterogeneity of amphioxus TEs might have favored genomic stasis during the evolution of amphioxus. Although it remains unclear whether there was a burst of TEs after the 2R-WGD during early vertebrate evolution, the reduced TE diversity shared among vertebrates compared to amphioxus is compatible with such burst of TEs. Acknowledgements The authors thank Jerzy Jurka for generously sharing valuable TE information on the amphioxus genome. The authors also thank Linda Holland for kindly sharing with us unpublished data on amphioxus Dkk gene, and for discussing with us interesting insights on the evolution of amphioxus MITES. FUNDING This work has been supported by grant 2009SGR336 from Generalitat de Catalunya and grant BFU from the Ministerio de Ciencia e Innovación (Spain). References 1. Lander ES, Linton LM, Birren B, et al. Initial sequencing and analysis of the human genome. Nature 2001;409: Warren WC, Hillier LW, Marshall Graves JA, etal. Genome analysis of the platypus reveals unique signatures of evolution. Nature 2008;453:

10 140 Can estro and Albalat 3. Finnegan DJ. Eukaryotic transposable elements and genome evolution. Trends Genet 1989;5: Wicker T, Sabot F, Hua-Van A, etal. A unified classification system for eukaryotic transposable elements. Nat Rev Genet 2007;8: Eickbush TH, Jamburuthugoda VK. The diversity of retrotransposons and the properties of their reverse transcriptases. Virus Res 2008;134: McClintock B. The origin and behavior of mutable loci in maize. Proc Natl Acad Sci USA 1950;36: Kaphingst K. A universal classification of eukaryotic transposable elements implemented in Repbase. Nat Rev Genet 2008;9:411 12, author reply Feschotte C, Zhang X, Wessler SR. Miniature invertedrepeat transposable elements (MITEs) and their relationship with established DNA transposons. In: Craig N.L, Craigie R, Gellert M, et al (eds). Mobile DNA II. Washington, DC: ASM Press, 2002: Cordaux R, Batzer MA. The impact of retrotransposons on human genome evolution. Nat Rev Genet 2009;10: Feschotte C, Pritham EJ. DNA transposons and the evolution of eukaryotic genomes. Annu Rev Genet 2007;41: Oliver KR, Greene WK. Transposable elements: powerful facilitators of evolution. Bioessays 2009;31: Zeh DW, Zeh JA, Ishida Y. Transposable elements and an epigenetic basis for punctuated equilibria. Bioessays 2009;31: Biemont C. A brief history of the status of transposable elements: from junk DNA to major players in evolution. Genetics 2010;186: Matzke MA, Matzke AJ. Polyploidy and transposons. Trends Ecol Evol 1998;13: Comai L. Genetic and epigenetic interactions in allopolyploid plants. Plant Mol Biol 2000;43: Rebollo R, Horard B, Hubert B, et al. Jumping genes and epigenetics: towards new species. Gene 2010;454: Putnam NH, Butts T, Ferrier DE, et al. The amphioxus genome and the evolution of the chordate karyotype. Nature 2008;453: Holland LZ, Albalat R, Azumi K, et al. The amphioxus genome illuminates vertebrate origins and cephalochordate biology. Genome Res 2008;18: Pritham EJ, Feschotte C. Massive amplification of rolling-circle transposons in the lineage of the bat Myotis lucifugus. Proc Natl Acad Sci USA 2007;104: Martinez-Mir A, Cañestro C, Gonzalez-Duarte R, et al. Characterization of the amphioxus presenilin gene in a high gene-density genomic region illustrates duplication during the vertebrate lineage. Gene 2001;279: Albalat R, Permanyer J, Cañestro C, et al. The first non-ltr retrotransposon characterised in the cephalochordate amphioxus, BfCR1, shows similarities to CR1-like elements. Cell Mol Life Sci 2003;60: Malik HS, Burke WD, Eickbush TH. The age and evolution of non-ltr retrotransposable elements. MolBiolEvol 1999;16: Cañestro C, Gonzàlez-Duarte R, Albalat R. Minisatellite instability at the Adh locus reveals somatic polymorphism in amphioxus. Nucleic Acids Res 2002;30: Jurka J, Kapitonov VV, Pavlicek A, et al. Repbase Update, a database of eukaryotic repetitive elements. Cytogenet Genome Res 2005;110: Holland LZ. A SINE in the genome of the cephalochordate amphioxus is an Alu element. IntJ Biol Sci 2006;2: Nishihara H, Smit AF, Okada N. Functional noncoding sequences derived from SINEs in the mammalian genome. Genome Res 2006;16: Yuan YW, Wessler SR. The catalytic domain of all eukaryotic cut-and-paste transposase superfamilies. Proc Natl Acad Sci USA 2011;108: Cañestro C, Albalat R, Gonzalez-Duarte R. Isolation and characterization of the first non-autonomous transposable element in amphioxus, ATE-1. Gene 2003;318: Osborne P, Luke G, Holland P, et al. Identification and Characterisation of five novel Miniature Inverted-repeat Transposable Elements (MITEs) in amphioxus (Branchiostoma floridae). IntJ Biol Sci 2006;2: Xing F, Tan X, Zhang PJ, et al. Characterization of amphioxus GDF8/11 gene, an archetype of vertebrate MSTN and GDF11. Dev Genes Evol 2007;217: Osborne PW, Ferrier DE. Chordate Hox and ParaHox gene clusters differ dramatically in their repetitive element content. MolBiolEvol 2010;27: Cañestro C, Albalat R, Hjelmqvist L, et al. Ascidian and amphioxus Adh genes correlate functional and molecular features of the ADH family expansion during vertebrate evolution. JMolEvol 2002;54: Sorek R. The birth of new exons: mechanisms and evolutionary consequences. RNA 2007;13: Thornburg BG, Gotea V, Makalowski W. Transposable elements as a significant source of transcription regulating signals. Gene 2006;365: Polak P, Domany E. Alu elements contain many binding sites for transcription factors and may play a role in regulation of developmental processes. BMCGenomics 2006;7: Bourque G, Leong B, Vega VB, et al. Evolution of the mammalian transcription factor binding repertoire via transposable elements. Genome Res 2008;18: Fablet M, Rebollo R, Biemont C, et al. The evolution of retrotransposon regulatory regions and its consequences on the Drosophila melanogaster and Homo sapiens host genomes. Gene 2007;390: Kapitonov VV, Jurka J. LTR retrotransposons from Branchiostoma floridae. Repbase Rep 2009;9: Kapitonov VV, Jurka J. Troyka a distinctive group of gypsy-like LTR retrotransposons inducing 3-bp target-site duplications. Repbase Rep 2008;8: Permanyer J, Albalat R, Gonzalez-Duarte R. Getting closer to a pre-vertebrate genome: the non-ltr retrotransposons of Branchiostoma floridae. IntJ Biol Sci 2006;2: Kapitonov VV, Jurka J. Young families of CR1 non-ltr 2009;9: Kapitonov VV, Tempel S, Jurka J. Simple and fast classification of non-ltr retrotransposons based on phylogeny of their RT domain protein sequences. Gene 2009;448: Kapitonov VV, Jurka J. Young families of Crack non-ltr 2009;9:

11 Amphioxus Transposable Elements Kojima KK, Kapitonov VV, Jurka J. Recent expansion of a new Ingi-related clade of Vingi non-ltr retrotransposons in hedgehogs. MolBiolEvol 2011;28: Kapitonov VV, Jurka J. Young families of I non-ltr retrotransposons from the amphioxus genome. Repbase Rep 2009;9: Kapitonov VV, Jurka J. New families of I non-ltr retrotransposons from animals. Repbase Rep 2009;9: Kapitonov VV, Jurka J. Young families of Tx1 non-ltr 2009;9: Kapitonov VV, Jurka J. Young families of HERO non-ltr retrotransposons from the amphioxus and sea urchin genomes. Repbase Rep 2009;9: Kapitonov VV, Jurka J. Proto2, a novel clade of metazoan non-ltr retrotransposons. Repbase Rep 2009;9: Kapitonov VV, Jurka J. Young families of REX1 non-ltr 2009;9: Kapitonov VV, Jurka J. Young families of RTE non-ltr 2009;9: Kapitonov VV, Jurka J. Young families of RTEX non-ltr 2009;9: Kapitonov VV, Jurka J. Chapaev a novel superfamily of DNA transposons. Repbase Rep 2007;7: Kapitonov VV, Jurka J. EnSpm-BF a family of autonomous DNA transposons from the amphioxus genome. Repbase Rep 2008;8: Bao W, Kapitonov VV, Jurka J. Ginger DNA transposons in eukaryotes and their evolutionary relationships with long terminal repeat retrotransposons. Mob DNA 2010;1: Kapitonov VV, Jurka J. hat transposons in the amphioxus genome. Repbase Rep 2008;8: Bao W, Jurka J. hatw-type DNA transposons from Branchiostoma floridae. Repbase Rep 2009;9: Kapitonov VV, Jurka J. IS4EU, a novel superfamily of eukaryotic DNA transposons. Repbase Rep 2007;7: Kapitonov VV, Jurka J. Kolobok, a novel superfamily of eukaryotic DNA transposons. Repbase Rep 2007;7: Bao W, Jurka J. MuDr-type DNA transposons from Branchiostoma floridae. Repbase Rep 2009;9: Kapitonov VV, Jurka J. Harbinger-BF a family of autonomous DNA transposons from the amphioxus genome. Repbase Rep 2008;8: Bao W, Jurka MG, Kapitonov VV, et al. New superfamilies of eukaryotic DNA transposons and their internal divisions. MolBiolEvol 2009;26:983 93, Ray DA, Pagan HJ, Thompson ML, et al. Bats with hats: evidence for recent DNA transposon activity in genus Myotis. MolBiolEvol 2007;24: Wicker T, Robertson JS, Schulze SR, et al. The repetitive landscape of the chicken genome. Genome Res 2005;15: Ayala FJ, Coluzzi M. Chromosome speciation: humans, Drosophila, and mosquitoes. Proc Natl Acad Sci USA 2005; 102(Suppl 1): Hufton AL, Groth D, Vingron M, et al. Early vertebrate whole genome duplications were predated by a period of intense genome rearrangement. Genome Res 2008;18: Garcia-Fernàndez J, Ferrier DE, Minguillón C, et al. The amphioxus genome in Evo-Devo: archetype or cul de sac. IntJ Dev Biol 2001;45(S1):S Nohara M, Nishida M, Manthacitra V, et al. Ancient phylogenetic separation between Pacific and Atlantic cephalochordates as revealed by mitochondrial genome analysis. Zoolog Sci 2004;21: Somorjai I, Bertrand S, Camasses A, et al. Evidence for stasis and not genetic piracy in developmental expression patterns of Branchiostoma lanceolatum and Branchiostoma floridae, two amphioxus species that have evolved independently over the course of 200 Myr. Dev Genes Evol 2008;218: Matzke MA, Mette MF, Matzke AJ. Transgene silencing by the host genome defense: implications for the evolution of epigenetic control mechanisms in plants and vertebrates. Plant Mol Biol 2000;43: SanMiguel P, Gaut BS, Tikhonov A, etal. The paleontology of intergene retrotransposons of maize. Nat Genet 1998;20: Wessler SR. Transposable elements associated with normal plant genes. Physiol Plant 1998;103: SanMiguel P, Tikhonov A, Jin YK, et al. Nested retrotransposons in the intergenic regions of the maize genome. Science 1996;274: Ohno S. Evolution by gene duplication. New York: Springer-Verlag, Dehal P, Boore JL. Two rounds of whole genome duplication in the ancestral vertebrate. PLoS Biol 2005;3:e Abrusán G, Krambeck H-J. Competition may determine the diversity of transposable elements. Theor Popul Biol 2006;70: Böhne A, Brunet F, Galiana-Arnoux D, et al. Transposable elements as drivers of genomic and biological diversity in vertebrates. Chromosome Res 2008;16: Parisod C, Salmon A, Zerjal T, et al. Rapid structural and epigenetic reorganization near transposable elements in hybrid and allopolyploid genomes in Spartina. New Phytol 2009;184: Parisod C, Alix K, Just J, et al. Impact of transposable elements on the organization and function of allopolyploid genomes. New Phytol 2010;186: Zemach A, McDaniel IE, Silva P, et al. Genome-wide evolutionary analysis of eukaryotic DNA methylation. Science 2010;328: Suzuki MM, Kerr AR, De Sousa D, et al. CpG methylation is targeted to transcription units in an invertebrate genome. Genome Res Permanyer J, Gonzalez-Duarte R, Albalat R. The non-ltr retrotransposons in Ciona intestinalis: new insights into the evolution of chordate genomes. Genome Biol 2003;4:R Albalat R, Martí-Solans J, Cañestro C. DNA methylation in amphioxus: from ancestral functions to new roles in vertebrates. Brief Funct Genomics. (in press).