Novel clades of chromodomain-containing Gypsy LTR retrotransposons from mosses (Bryophyta)

The Plant Journal (2008) 56, 562–574

doi: 10.1111/j.1365-313X.2008.03621.x

Novel clades of chromodomain-containing Gypsy LTR retrotransposons from mosses (Bryophyta) Olga Novikova1,*, Vladimir Mayorov2, Georgiy Smyshlyaev3, Michail Fursov4, Linda Adkison2, Olga Pisarenko5 and Alexander Blinov1 1 Laboratory of Molecular Evolution, Institute of Cytology and Genetics SB RAS, Novosibirsk, Russia, 2 Department of Basic Medical Sciences, Mercer University School of Medicine, Macon, GA, USA, 3 Novosibirsk State University, Novosibirsk, Russia, 4 Novosibirsk Center of Information Technologies ‘UniPro’, Novosibirsk, Russia, and 5 Central Siberian Botanical Garden, Novosibirsk, Russia Received 20 June 2008; accepted 25 June 2008; published online 11 August 2008. * For correspondence (fax +7 383 3331278; e-mail [email protected]).

Summary Retrotransposons are the major component of plant genomes. Chromodomain-containing Gypsy long terminal repeat (LTR) retrotransposons are widely distributed in eukaryotes. Four distinct clades of chromodomain-containing Gypsy retroelements are known from the vascular plants: Reina, CRM, Galadriel and Tekay. At the same time, almost nothing is known about the repertoire of LTR retrotransposons in bryophyte genomes. We have combined a search of chromodomain-containing Gypsy retroelements in Physcomitrella genomic sequences and an experimental investigation of diverse moss species. The computerbased mining of the chromodomain-containing LTR retrotransposons allowed us to describe four different elements from Physcomitrella. Four novel clades were identified that are evolutionarily distinct from the chromodomain-containing Gypsy LTR retrotransposons of other plants. Keywords: LTR retrotransposons, chromodomain-containing Gypsy, Physcomitrella patens, mosses, phylogeny.

Introduction Transposable elements compose a significant portion of many eukaryotic genomes (Langdon et al., 2000). Retrotransposons are ubiquitous in plant genomes, and are instrumental in the evolution of genes and genomes because of the broad spectrum of mutations produced (Kidwell and Lisch, 2000). In plant genomes, all major types of retroelements exist, but the quantitative and qualitative distribution of these differ, even between closely related species (Feschotte et al., 2002). Plant retrotransposons are structurally and functionally similar to retrotransposons of other eukaryotes. However, there are essential differences in the genomic organization of plant transposable elements in comparison with elements found in other kingdoms. These differences include an increased number, highly heterogeneous populations and chromosome distribution patterns of retrotransposons (Kumar and Bennetzen, 1999). Long terminal repeat (LTR) retrotransposons are flanked by LTRs in direct orientation. The mechanism of transposi562

tion is similar to that of retroviruses (Finnegan, 1992; Wicker et al., 2007). Full-sized autonomous LTR retrotransposons are 4–10-kb long and contain at least two genes, gag and pol (Wicker et al., 2007). The gag gene encodes a protein similar to the retrovirus capsid protein; the pol gene encodes a protein that has protease (PR), reverse transcriptase (RT), ribonuclease H (RNaseH) and integrase (Int) activities (Wicker et al., 2007). Phylogenetic analyses based on comparison of RT domains demonstrate that most LTR retrotransposons can be classified into several superfamilies: Copia (Pseudoviridae), Gypsy (Metaviridae), retroviruses (Retroviridae), Bel and DIRS (Goodwin and Poulter, 2001; Havecker et al., 2004; Peterson-Burch and Voytas, 2002). Gypsy or Metaviridae elements have structural similarity to retroviruses, whereas Copia or Pseudoviridae elements differ from retroviruses. In particular, retrovirus and Gypsy elements have a pol gene organized as PR-RT-RNaseH-Int, whereas the pol gene of ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd

Gypsy elements from mosses 563 Copia elements is organized as PR-Int-RT-RNaseH (Feschotte et al., 2002; Kim et al., 1998; Miller et al., 1998; Wicker et al., 2007). Three genera are known in Gypsy: Errantivirus, Metavirus and Chromovirus (Kordis, 2005; Marı´n and Llore´ns, 2000). Errantivirus and Metavirus differ by the presence (Errantivirus) or absence (Metavirus) of an envelope gene. Elements from the genus Chromovirus have a chromodomain, which targets integration to heterochromatin by recognizing histone modifications (Gao et al., 2008; Gorinsek et al., 2004; Malik and Eickbush, 1999; Marı´n and Llore´ns, 2000). Chromodomain-containing LTR retrotransposons are the most widespread clade of Gypsy, and are present in the genomes of plants, fungi and vertebrates (Gorinsek et al., 2004; Marı´n and Llore´ns, 2000). They seem to be widespread among plants (Gorinsek et al., 2004). However, the majority of these are described from angiosperms (Vitte and Panaud, 2005). Based on investigations of numerous angiosperms, it was shown that the genus Chromovirus in plants is represented by four distinct clades: Reina, CRM, Galadriel and Tekay (Gorinsek et al., 2004; Marı´n and Llore´ns, 2000). A number of comprehensive studies of coniferous species were also conducted (Friesen et al., 2001). Nevertheless, the study of the evolution of chromodomain-containing Gypsy could not be complete without investigations of other groups from the plant kingdom. Bryophytes are one of the oldest groups of plants on Earth, with their origin dating back to the Devonian. They are small land plants, and are quite different from other higher (vascular) plants. The haploid gametophyte is the dominant generation in the life cycle in all bryophytes. In this respect, the bryophytes are different from vascular plants, in which the polyploid sporophyte dominates the life cycle. The moss Physcomitrella patens (Hedw.) Bruch et al. is an emerging plant model system because of its high rate of homologous recombination, haploidy, simple body plan and physiological properties, as well as phylogenetic position. The differences in gene content between Physcomitrella and Arabidopsis was recently shown by Rensing et al. (2005). It was proposed that the potential ‘retained genes’ might have been lost during seed plant evolution (Rensing et al., 2005; Reski and Frank, 2005; Stenøien, 2007). Functional annotation of these genes revealed an unequal distribution among taxonomic groups, as well as intriguing functions such as cytotoxicity and nucleic acid repair. A high proportion of the sequences from unicellular eukaryotes were retained by moss homologs, apparently since the last common ancestor. Retrotransposons can reorganize genomes, and thus change the structure and regulation of genes through insertions, deletions, chromosomal reorganizations and ectopic recombinations (Bowen and Jordan, 2002; Fedoroff, 1999). The high rate of recombination in mosses and their

haploidy make retrotransposons even more powerful agents in the evolution of this plant group. Moreover, some retrotransposons that were lost by higher vascular plants could be retained in mosses from their common ancestor with plants. The present study was initiated to investigate the diversity of chromodomain-containing Gypsy LTR retrotransposons in mosses. The analysis of the entire P. patens genome detected new PpatiensLTR retrotransposons, which belong to the novel moss-specific clade of chromodomain-containing Gypsy retrotransposons. This clade is distantly related to other known clades of chromodomain-containing Gypsy retrotransposons from plants. The data suggest that this clade appeared before the divergence of plants and Fungi/ Metazoa groups, and that it was lost by seed plants (gymnosperms and angiosperms). PCR screening and dotblot hybridization of 34 bryophyte taxa showed a wide distribution of the newly identified clade. Moreover, three additional bryophyte-specific clades were identified, which were also classified as the ‘retained clades’ of retrotransposons, as none had plant-wide distribution. Results Initial identification of chromodomain-containing Gypsy elements from P. patens The conservation of coding domain sequences, such as RT, within retrotransposons allowed primers to be designed for degenerate PCR (Friesen et al., 2001; Hirochika et al., 1992). Initially, the presence of the chromodomain-containing Gypsy retrotransposons in P. patens was tested by amplifying genomic DNA with degenerate oligonucleotides based on conserved amino acid sequence domains of chromodomain-containing Gypsy-like RT. A 330-bp PCR product was cloned, and seven sequences were determined. All of the sequences showed clear similarity to the RT domain of chromodomain-containing Gypsy LTR retrotransposons. One of these was a fragment of the degenerate RT, which was interrupted by several stop codons (PhyPatTy3-7). The clone, named PhyPatTy3-5, represented a truncated RT, whereas the remaining clones contained the intact RT. Comparative nucleotide sequence analyses revealed a high similarity among six clones. They shared more than 96% of similarity, and it was concluded that all fragments were amplified from the copies of the same retrotransposon, which we named PpatensLTR1. The seventh clone (PhyPatTy3-6) significantly differed from the rest: only 65% of its nucleotide sequence was shared in pairwise comparisons with other clones. This RT fragment obviously belongs to another retrotransposon, PpatensLTR2. Pairwise comparisons of the amino acid sequences showed an average similarity of 93% among copies of PpatensLTR1, and 68% between different retrotransposons. Preliminary BLAST

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

564 Olga Novikova et al. analysis showed the RT sequences of both PpatensLTR1 and PpatensLTR2 elements were more closely related to the fungal RTs than to known retroelements from higher plants. An alignment of the derived amino acid sequences of clones is presented in Figure 1a, along with chromodomain-containing Gypsy retrotransposons from fungi and angiosperm plants. The phylogenetic tree resulting from neighbor-joining (NJ) analysis of newly identified chromodomain-containing Gypsy retrotransposons, and known elements from plants, fungi and vertebrates, which are available in GenBank, showed that PpatensLTRs do not belong to any known clades of chromodomain-containing Gypsy retrotransposons from plants (Figure 1b). Moreover, retrotransposons from Physcomitrella formed a common cluster with fungal chromodomain-containing Gypsy elements. PpatensLTR1 reconstruction based on genomic sequences It was proposed earlier that no full-length chromodomaincontaining Gypsy elements could be obtained from the genomic sequences of P. patens (Gorinsek et al., 2004). Indeed, the initial analysis of genomic sequences showed that PpatensLTR1 was represented by degenerate elements, and that no full-length copies were found. Moreover, the sequencing of the Physcomitrella genome is still in progress, and is represented in databases as a draft assembly (Rensing et al., 2008). The scaffold sequences have multiple interruptions by polyN tracks. Nevertheless, we performed additional surveys of LTR retrotransposons from Physcomitrella whole genomic sequences (WGS). Using computer-based sequence similarity searches, we mined PpatensLTR1 from the draft genomic sequence of Physcomitrella (http://genome.jgi-psf.org). Our preliminary search revealed several sequences among WGS with more than 95% amino acid sequence similarity with PhyPatTy3 clones. Next, we used the UniPro GenomeBrowser (http:// genome.unipro.ru) to run the whole element search (see Experimental procedures). The majority of the RT fragments of the elements detected were flanked by long polyN tracks. Nevertheless, several copies were found that represented almost intact retroelements, or had short truncations at the 5¢ or 3¢ ends. Moreover, some of the copies showed the presence of putative intact open reading frames (ORFs). All detected copies shared a high similarity of nucleotide acid sequences, and a complete retroelement could easily be reconstructed from these data. We used multiple alignments of more than 50 copies for PpatensLTR1 reconstruction (see Supplementary material for further details). A single copy was used as the basis for reconstruction of PpatensLTR1. It was located in scaffold_1 of the draft WGS (from position 4 281 237 to 4 286 966). This copy of PpatensLTR1 carries both LTRs, but has no recognizable ORFs. Additionally, it displayed the presence of two short

deletions (100 and 99 bp in length) in comparison with other copies. The internal part of the PpatensLTR1 retrotransposon was restored based on the copies that had putative intact ORFs. One of these was located in scaffold_67, between position 17 128 and 20 328. This copy is somewhat truncated at the 3¢ end, and carries a single ORF that was referred to as gag ORF1. The second copy of PpatensLTR1 was detected in scaffold_103, between 1 063 841 and 1 068 274. This copy is also truncated at the 3¢ end, but has two long coding regions. Based on these regions, and additional element copies, the pol-like ORF2 was restored. The complete reconstructed PpatensLTR1 retrotransposon is 6157 bp in length (Figure 2; Table 1), has two LTRs of 488 bp in length and terminates in a short inverted terminal repeat (ITR, TG…CA), which is common for the LTRs of LTR retroelements and retroviruses (Dej et al., 1998). Direct flanking repeats or terminal site duplicated (TSD) sequences are generated at the final step of the LTR retrotransposon transposition in the processes of reparation of the cleaved DNA (Deka et al., 1988). However, no target site duplications were identified at the end of the reconstructed PpatensLTR1. Long terminal repeats of retrotransposons and retroviruses carry regulatory sequences important for the processes of transcription and transposition as a whole, and contain several structural features. They bear an internal promoter and sites for the recognition of transcription factors. The LTRs of PpatensLTR1 were analyzed for DNA regulatory sequences. Retroviral TATA and CAAT boxes were detected in both repeats. A primer-binding site (PBS) and a polypurine tract (PPT) were identified outside of the coding region of PpatensLTR1. These sites are required for replication of LTR retroelements. The PBS is necessary for the initiation of reverse transcription and for the synthesis of the first-strand complementary 5¢LTR sequence (Telesnitsky and Goff, 1997). The PBS of PpatensLTR1 is located downstream of the 5¢LTR, and is complementary to the 3¢end of tRNAMet (Figure 2a). The PPT sequence is involved in second-strand DNA synthesis, and is represented as the stretch of 11 purines located adjacent to the 3¢LTR of PpatensLTR1. The putative first ORF (ORF1) of PaptensLTR1 is 1002 bp (334 aa) in length. A second reconstructed ORF (ORF2) encodes a polyprotein that bears resemblance to the characteristic retroviral aspartyl PR, RT, RNase H and Int domains of retrotransposon proteins, in the order indicated (Figure 2a). The chromo (chromatin organization modifier) domain has also been identified in the protein product of ORF2. Additionally, the cysteine or the Zn-finger motif with His(2)-Cys(2) composition, typical for DNA-binding proteins, was found (Vallee and Auld, 1993). It appears that PpatensLTR1 is a long-standing part of the Physcomitrella genome. It is represented by at least 10 000 copies per 480 Mb of draft scaffold sequence (Rensing et al., 2008). However, all investigated copies are degenerate, and it remains unclear whether a complete PpatensLTR1 exists.

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

Gypsy elements from mosses 565 Figure 1. Multiple alignment of newly isolated chromodomain-containing Gypsy long terminal repeat (LTR) retrotransposons PhyPatTy3 from Physcomitrella patens, along with chromodomain-containing Gypsy retrotransposons from fungi and angiosperm plants (a), and the neighbor-joining (NJ) phylogenetic tree based on reverse transcriptase (RT) nucleotide sequences of LTR retrotransposons, including PhyPatTy3 clones (b). The statistical support for the tree was evaluated by bootstrapping (1000 replications); nodes with bootstrap values over 50% are shown. The genus Chromovirus and the plant-specific clades: Reina, CRM, Galadriel and Tekay, are indicated. The name of the host species and accession number is indicated for LTR elements taken from GenBank. RT_Cn, reverse transcriptase from Cryptococcus neoformans (Fungi; GenBank Acc. No XM_571377); RT_Cc, reverse transcriptase Coprinopsis cinerea (Fungi; GenBank Acc. No XM_001834840).

(a)

(b)

Nevertheless, the high copy number and the closely relatedness of the copies of PpatensLTR1 suggest that this element was recently active. The limits of the current unfinished Physcomitrella WGS, and incorrect assembling as a result of the high copy numbers of repeated sequences, could explain the absence of intact PpatensLTR1 in our survey.

PpatensLTR2 and other chromodomain-containing Gypsy retrotransposons from P. patens The clone PhyPatTy3-7 was used as a query for a BLAST search of the PpatensLTR2 element in unmasked WGS of Physcomitrella. We were able to detect the intact

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

566 Olga Novikova et al. Figure 2. The structural organization of PpatensLTR retroelements from Physcomitrella (a), and amino acid sequence alignment of retrotransposable chromodomains (b). PR, aspartyl protease; RT, reverse transcriptase; Rnase H, ribonuclease H; Int, integrase; chromo, chromodomains; H2C2, Zn-finger motif; 5¢ and 3¢LTRs, 5¢ and 3¢ long terminal repeats; PBS, primer binding site; PPT, polypurine tract. The conserved aromatic residues typical for group I of the chromodomains are marked by arrows. The GenBank accession numbers for the retrotransposon sequences are as follows: Skippy, L34658; Cft1, AF051915; Pyret, AB062507; MGRL3, AF314096; MarY1, AB028236; MAGGY, L35053; Osr35, AC068924; rn377–208, AK068625; Reina, U69258; RIRE3, AC119148; Tekay, AF050455; RetroSor2, AF061282; Tma, AF147263.

(a)

(b)

Table 1 Characteristics of the chromodomain-containing Gypsy long terminal repeat (LTR) retrotransposons from Physcomitrella patens

Element name

LTRs Element Intact size size (bp) copies (bp)

LTRs identity (%) TSD ITR

PpatensLTR1 PpatensLTR2 PpatensLTR3 PpatensLTR4

6157 6476 6081 5726

91.6 92 97.2 99.3

ND + ND ND

486 474/452 394 434

ND + + +

ta…ta tgt…aca tg…ca tg…ca

ND, not detected; TSD, terminal site duplications; ITR, dinucleotide inverted repeat.

PpatensLTR2 in spite of the great number of degenerate copies. PpatensLTR2 carries two LTRs that are 474 bp (5¢LTR) and 452 bp (3¢LTR) in length, and have conserved features, including the dinucleotide end sequences (TG…CA). Target site duplications were also detected (ATCTC…ATCTC). The LTRs showed the presence of a viral TATA box. An 11-bp PBS-like sequence, complementary to the 3¢-end region of the transfer RNA (tRNA), was present downstream of the 5¢LTR, and a PPT was detected immediately upstream of the 3¢LTR (Figure 2a; Table 1).

The ORF1 of PpatensLTR2 is 1047 bp in length, and encodes a protein (349 aa) that shows a high similarity with the Gag protein from retroviruses. The pol-like ORF2 starts 8 bp downstream of the gag-like ORF1. A search of characteristic motifs within the protein sequence identified all typical domains, i.e. PR, RT, Rnase H, Int and chromo. The BLAST analysis of Physcomitrella WGS revealed more than 1000 copies of PpatensLTR2 per haploid genome. A subsequent analysis showed that these copies share a very high similarity at the DNA level (95% on average in the RT region), and some copies showed the presence of either intact ORF1 or ORF2. Additionally, we identified a number of copies with pseudo ORFs, which are interrupted by several stop codons, but without frameshifts. All these facts may suggest a relatively recent activity of the PpatensLTR2 element in Physcomitrella. Our survey and analysis of the PpatensLTR2-like elements from Physcomitrella WGS yielded two more chromodomain-containing Gypsy retroelements, PpatensLTR3 and PpatensLTR4 (Figure 2a; Table 1), which are represented by a lower copy number per haploid genome (less than 500 copies per genome) than the PpatensLTR1 retroelement. Both elements are degenerate; no potential intact copies were identified. PpatensLTR3 was identified in scaffold_3

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

Gypsy elements from mosses 567 (between position 2 549 334 and 2 543 187). This element is 6081 bp in length and carries 394-bp-long terminal repeats with 97.2% of similarity (Figure 2a). It is flanked by target site duplications CTTGT. The LTRs contain the terminal sequences (TG…CA) and regulatory elements. The 11-bp PPT was found, which is located upstream of the 3¢LTR. The potential tRNA primer binding site was also identified downstream of the 5¢LTR. The intact ORF1 (885 bp) and a putative pseudo-ORF2 occurred within the element. The putative pseudo-ORF2 is 3462-bp long, and is interrupted by three stop codons. Analyses of several PpatensLTR3 copies showed that all of them had pseudo-ORFs, but that the stop codons were located in different positions. Thus, substitutions that led to coding sequence variations occurred independently in PpatensLTR3 copies after their insertion into the genome. The high similarity of LTRs, presence of the recognizable target site duplications and existence of the relatively integral copies indicate a recent transpositional activity of the PpatensLTR3 elements. PpatensLTR4 was also found in scaffold_3 (between position 4 194 293 and 4 199 821), and showed the presence of pseudo-ORF in the internal part of the body. The element is flanked by 434-bp LTRs. Target site duplications have an AAATA sequence. PpatensLTR4 possesses a long pol-like ORF2 coding region (Figure 2a). However, instead of a gag-like ORF1, which should be expected upstream of the ORF2, two short ORFs (546 and 405 bp) were identified. Further comparative analyses of different copies showed that this particular copy of PpatensLTR4 has a 49-bp deletion resulting in a frameshift and a disrupted coding region. The reconstructed PpatensLTR4 has a 999-bp, gag-like ORF1. There are two distinct groups of chromodomains. Group I is characterized by the presence of the three conserved aromatic residues (Gao et al., 2008; Figure 2b). This group of chromodomains was found in diverse eukaryotic LTR retrotransposons, including fungal Gypsy elements. Representatives of the group II of chromodomains lack the first conserved aromatic residue, and usually also lack the third. Group II was identified only in plant retrotransposons (Gao et al., 2008). The comparative analysis of chromodomains from PpatensLTRs clearly demonstrated that newly identified LTR retrotransposons from Physcomitrella have group-I chromodomains, but not group-II, like other chromodomaincontaining Gypsy retrotransposons from plants (Figure 2b). It indicates that PpatensLTRs are more closely related to the fungal retrotransposons than to the clades Galadriel, Tekay, Reina or CRM, which were described for plants. Chromodomain-containing Gypsy LTR retrotransposons in diverse mosses and other plants The presence of chromodomain-containing Gypsy retrotransposons among various mosses was tested by ampli-

fying genomic DNA with degenerate oligonucleotide primers and dot-blotting. The distribution of chromodomain-containing Gypsy LTR retrotransposons has been analyzed in 34 moss species belonging to 27 families (Table 2). We also included an analysis of several other green plants, listed in Table 3. Following the PCR analysis, in which the expected 330-bp product was obtained for 17 mosses and 12 species from other green plants, amplicons were cloned and sequenced. In total, 174 independent clones were isolated. Of these, 162 clones, including 98 clones from mosses, showed sequence similarity with Gypsy RT sequences. Corresponding amino acid sequences were analyzed for the presence of stop codons and frame shifts. We found that 24 clones had stop codons, whereas 17 clones possessed frame shifts either with or without stop codons, and five clones were truncated. Finally, 151 clones, including 96 clones from mosses, were selected for further phylogenetic analysis. The phylogenetic relationships among the obtained clones and the elements extracted from databases were reconstructed using NJ analysis based on a multiple alignment of nucleotide sequences (Figure 3; see Supplementary material for further details). The Gypsy LTR retroelements from Drosophila melanogaster were used as an outgroup. The phylogenetic analysis showed a high level of sequence heterogeneity among moss-derived clones, which could be grouped into at least four distinct groups (clades A–D). The majority of moss retrotransposons obtained, including PpatensLTR elements, belonged to clade A (Figure 3). The NJ phylogeny demonstrated 95% bootstrap support for the monophyletic origin of this clade. The relationships among a large number of well-supported small families inside clade A remain unresolved. It is clear, however, that clade A is more closely related to the Fungi/ Metazoa group of chromodomain-containing Gypsy retroelements than to the known plant-specific clades. The phylogenetic analysis clearly demonstrated that all investigated moss clones could be subdivided into two groups. The first group includes transposable elements from those species that demonstrated a high homogeneity of the nucleotide sequences, and all RT fragments from such species formed a single clade on the phylogenetic tree (Figure 3). For example, 13 RT fragments isolated from Pleurozium schreberi (PleSchrebTy3) shared more than 83% similarity; nine clones from Dicranum polysetum (DicPolyTy3) showed an average similarity of 92.6%. Putative chromodomain-containing Gypsy elements isolated from Sphagnum obtusum, Racomitrium canescens, Splachnum ampullaceum, Pylaisia polyantha and Plagiomnium cuspidatum also showed very high intraspecific similarity. Species of the second group contain several lineages of retrotransposons in their genomes. The PpatensLTR1

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

568 Olga Novikova et al. Class

Family

Sphagnopsida

Sphagnaceae

Species

PCR/dot

Sphagnum fuscum (Schimp.) H.Klinggr. Sphagnum girgensohnii Russow Sphagnum obtusum Warnst. Andreaeopsida Andreaeaceae Andreaea rupestris Hedw. Tetraphidopsida Tetraphidaceae Tetraphis pellucida Hedw. Polytrichopsida Polytrichaceae Polytrichum commune Hedw. Bryopsida Timmiaceae Timmia megapolitana Hedw. Funariaceae Physcomitrella patens (Hedw.) Bruch et al. Funaria hygrometrica Hedw. Encalyptaceae Encalypta rhaptocarpa Schwa¨gr. Grimmiaceae Niphotrichum canescens (Hedw.) Bednarek-Ochyra & Ochyra = Racomitrium canescens (Hedw.) Brid. Leucobryaceae Leucobryum glaucum (Hedw.) Angstr. Dicranaceae Dicranum polysetum Sw. Orthotrichaceae Orthotrichum speciosum Nees Splachnaceae Splachnum ampullaceum Hedw. Hedwigiaceae Hedwigia ciliata (Hedw.) P.Beauv. Bryaceae Bryum argenteum Hedw. Bryum capillare Hedw. Orthodontiaceae Orthodontopsis bardunovii Ignatov & B.C.Tan Mniaceae Plagiomnium cuspidatum (Hedw.) T.J.Kop. Bartramiaceae Bartramia pomiformis Hedw. Aulacomniaceae Aulacomnium palustre (Hedw.) Schwa¨gr. Fontinalaceae Dichelyma falcatum (Hedw.) Myrin Plagiotheciaceae Plagiothecium laetum Bruch et al. Hypnaceae Hypnum cupressiforme (Hedw.) P.Beauv. Vesicularia dubyana (Mu¨ll. Hal.) Broth. Leskeaceae Lescuraea saxicola (Bruch et al.) Molendo Neckeraceae Neckera pennata Hedw. Hylocomiaceae Hylocomium splendens (Hedw.) Bruch et al. Pleurozium schreberi (Brid.) Mitt. Brachytheciaceae Brachythecium salebrosum (F.Weber & D.Mohr) Bruch et al. Scorpidiaceae Scorpidium scorpidioides (Hedw.) Limpr. Pylaisiaceae Pylaisia polyantha (Hedw.) Bruch et al. Stereodon pallescens (Hedw.) Mitt.

)/+ )/+ +/) +/+ +/) )/+ )/+ +/+ +/) )/+ +/+

Table 2 List of the moss species used in this study. The taxonomy follows that of Ignatov et al. (2006)

+/) +/) )/+ +/) +/+ )/+ )/+ )/+ +/) )/+ )/+ )/+ +/) )/+ +/) +/) )/+ )/+ +/) )/+ +/) +/) )/+

Table 3 List of the other plant species used in present study Class

Order

Family

Species

Lycopodiopsida Polypodiopsida Equisetopsida Ginkgoopsida Cycadopsida Gnetopsida Pinopsida = Coniferae

Lycopodiales Polypodiales Equisetales Ginkgoales Cycadales Ephedrales Coniferales

Magnoliopsida

Nymphaeales

Lycopodiaceae Hypolepidaceae Equisetaceae Ginkgoaceae Cycadaceae Ephedraceae Taxaceae Pinaceae Taxodiaceae Nymphaeaceae Cabombaceae Magnoliaceae Piperaceae

Lycopodium lagopus (Laest.) Zinserl. ex Kuzen. Pteridium aquilinum Gled. ex Scop. Equisetum arvense L. Ginkgo biloba L. Cycas revoluta Thunb. Ephedra distachya L. Taxus baccata L.* Pinus radiata D. Don* Sciadopitys verticillata (Thunb.) Siebold & Zucc.* Nymphaea tetragona Georgi Brasenia schreberi J.F. Gmelin Magnolia grandiflora L. Peperomia caperata Yunck.

Magnoliales Piperales

*DNA was provided by the Royal Botanic Gardens, Kew, London, UK (http://www.kew.org).

element is evolutionarily distant from other PpatensLTR retroelements. The moss Funaria hygrometrica contains at least two lineages of retroelements inside clade A (lineages

FunHygTy3.1 and FunHygTy3.2 in Figure 3). Moreover, one of these formed common clusters with PpatensLTRs, and appears to be a recent common ancestor of the PpatensLTR3

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

Gypsy elements from mosses 569

Figure 3. Neighbor-joining (NJ) phylogenetic tree based on reverse transcriptase (RT) nucleotide sequences of long terminal repeat (LTR) retrotransposons, including newly described elements from mosses. Statistical support was evaluated by bootstrapping (1000 replications); nodes with bootstrap values over 50% are indicated. The name of the host species and accession number is indicated for LTR elements taken from GenBank, Oryza sativa retroelements are available from the RetrOryza database. Four novel clades (A–D) described for mosses are shown.

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

570 Olga Novikova et al. element. The LTR retrotransposons from Tetraphis pellucida are represented by two clusters of RT fragments that belonged to clade A (TetPellTy3.1 and TetPellTy3.2 lineages). Elements TetPellTy3-5 and TetPellTy3-8 showed 98.7% similarity (lineage TetPellTy3.1). Three clones TetPellTy3-6, TetPellTy3-9 and TetPellTy3-10 had more than 97% identity (lineage TetPellTy3.2). Pairwise comparisons between elements from these groups showed an average similarity of only 71%. Mosses Hedwigia ciliate and Scorpidium scorpidioides possessed at least three different elements in their genomes (HedCilTy3.1–HedCilTy3.3 and ScoScorpTy3.1–ScoScorpTy3.3 lineages). Three more additional clades were found. Two clades were represented by elements from single species. Clade B was found only in Plagiothecium laetum, whereas clade C comprised retrotransposons solely from F. hygrometrica (Figure 3). Finally, eight retrotransposons from T. pellucida and Vesicularia dubyana formed their own clade D. Although the bootstrap supports for the clades B, C and D were very strong, the phylogenetic positions of these clades remained unclear, as the statistical supports were very weak for clustering these with elements from other plants. It is interesting that two clones isolated from the bracken Pteridium aquilinum (PteAquTy3-1 and PteAquTy3-2) formed a sister clade to clade C. However, the bootstrap support showed that such branching is unreliable (only 33% bootstrap; data not shown). All other clones from non-moss plants were clearly divided into the well-known chromodomain-containing clades: Galadriel, Tekay, Reina and CRM (Figure 3). As we were not able to detect the elements closely related to the retrotransposons from moss-specific clades A–D in other plants, it remains unclear whether vascular plants contain elements from the novel branches or not. Nevertheless, based on the published reports concerning the whole genome sequence analysis of a number of vascular plant

species, we propose that the described elements were specific for mosses only (Gorinsek et al., 2004). Total DNA from 21 bryophyte species were dot-hybridized with the PlaCuspTy3-4 retroelement from P. cuspidatum (Figure 4). The strongest hybridization signals were obtained for the mosses belonging to the classes Andreaeopsida (Andreaea rupestris) and Tetraphidopsida (Polytrichum commune), as well as for Timmia megapolitana and P. patens from the class Bryopsida. The remaining species tested showed weaker hybridization signals, thereby supporting the diversity in the copy number of the elements in the different species. Elements from novel clades are transcriptionally inactive in mosses BLASTN was used to search the expressed sequence tag (EST) database to determine whether the PpatensLTR elements are transcribed in the Physcomitrella genome. Whole sequences of the PpatensLTR elements were used as the queries. Only two fragments from the EST database revealed a significant similarity. A fragment named pph17l15 (615 bp) showed a 95% similarity at the DNA level with the sequence of the PpatensLTR1. The 175-bp contig13819 was 97% identical to PpatensLTR1. No EST sequences were identified with significant similarities to other PpatensLTR elements. To test whether chromodomain-containing Gypsy retrotransposons were transcribed in other mosses, RT-PCR with degenerate ty3-S and ty3-A primers was performed. Total RNAs were isolated from four species: T. pellucida, V. dubyana, D. polysetum, and P. cuspidatum. No products were detected in all four RNA samples (Figure 5). At the same time, bands of the expected size were produced from PCR with template genomic DNA used as a positive control. The positive control for RNA quality was also

Figure 4. Dot-blot hybridization of 21 moss species with the PlaCuspTy3-4 retroelement.

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

Gypsy elements from mosses 571

Figure 5. RT-PCR analysis for the transcription activity of the chromodomaincontaining Gypsy retrotransposons using degenerate ty3 primers and the resistance (R) gene using Pp15 primers, described by Akita and Valkonen (2002), in four mosses.

included in the experiment, as we had a negative reaction in RT-PCR with all RNA templates. We chose the plant resistance (R) gene, and used the primers Pp15 described by Akita and Valkonen (2002). All four RNA templates gave a positive reaction in amplification with Pp15 primers. These data suggest that almost all chromodomain-containing Gypsy elements are transcriptionally inactive in Physcomitrella and other mosses, or that the level of their activity is undetectably low. Discussion Gypsy LTR retrotransposons, including chromodomaincontaining retrotransposons, are one of the major components of the plant genome, in which multiple families are present (Vitte and Panaud, 2005). Furthermore, the differences in genome size observed across the plant kingdom are accompanied by variation in LTR retrotransposon content, suggesting that LTR retrotransposons can be a significant factor in the evolution of plant genome size (Liu et al., 2007; Vitte and Panaud, 2005). Comprehensive studies of the LTR retrotransposons in plant genomes are published regularly (Gorinsek et al., 2004; Marı´n and Llore´ns, 2000; Terol et al., 2001; Wicker et al., 2001; Wright and Voytas, 1998). However, our understanding of the mobile elements in plants is still incomplete. The majority of the known families of LTR retrotransposons have been found and fully characterized in angiosperms, whereas other large groups have not been investigated so widely (Hirochika et al., 1992; Kumekawa et al., 1999; Vitte and Panaud, 2005). A detailed structural analysis of the whole genome provides the most robust data about the repertoire of retrotransposable elements, but the survey of the elements

from individual genomes cannot reflect the diversity of mobile elements in the whole taxonomic group. We have combined a search of chromodomain-containing Gypsy retrotransposons in Physcomitrella genomic sequences with an experimental investigation of diverse moss species. The computer-based mining of the chromodomain-containing elements allowed us to describe four different elements from Physcomitrella. Investigation of a number of bryophyte species showed that the described elements from Physcomitrella belonged to a widely distributed clade found in all investigated mosses, named clade A. The diversity of chromodomain-containing Gypsy retrotransposons in bryophytes is not limited to a single phylogenetic group. Three other distinct, moss-specific clades were found (B, C and D; see Figure 3). Our analyses show that mosses contain retrotransposons that are evolutionarily distinct from the chromodomaincontaining Gypsy retroelements of other plants. Phylogenetic investigation clearly demonstrates that the newly described clade A is more closely related to the Fungi/ Metazoa group of chromodomain-containing retrotransposons than to the clades Galadriel, Tekay, Reina or CRM, which were described for other plants (mostly for angiosperms). One could propose horizontal transmission as an explanation. Horizontal transfer is well described for gypsy LTR retrotransposons in Drosophila (Jordan et al., 1999; Terzian et al., 2000), and has also been suggested or discussed for plants (Friesen et al., 2001; Hirochika and Hirochika, 1993; Marı´n and Llore´ns, 2000). We adhere to the opinion that only vertical evolution led to the appearance of such a repertoire of retrotransposable elements in mosses. Evolutionary dynamics of transposable elements could sufficiently differ, even in closely related taxa. As a result, phylogenetic studies of mobile elements often show incongruence with host species phylogenies. Moreover, it was demonstrated previously that the EST data of Physcomitrella contained a fraction of transcripts derived from ‘retained genes’, which are not present in seed plants but can be found in other kingdoms (Rensing et al., 2005; Reski and Frank, 2005; Stenøien, 2007). It is highly possible that mosses also kept the retrotransposons that are closely related to the elements in the Fungi/Metazoa group from their last common ancestor. There is limited information about the diversity and distribution of chromodomaincontaining Gypsy retrotransposons in non-seed plants, such as other bryophytes (liverworts and hornworts), lycophytes and ferns (Kumekawa et al., 1999; Suoniemi et al., 1998). The elements from the bracken Pteridium (Filicopsida) described in the present study did not belong to any known clades from plants. They formed their own branch on a phylogenetic tree, as did clades B, C and D from mosses. Comprehensive studies of the elements from non-seed plants should provide additional insights into the diversity and evolution of mobile elements in the plant kingdom.

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

572 Olga Novikova et al. Analysis of Physcomitrella genomic sequences clearly showed that the majority of chromodomain-containing Gypsy elements are represented as degenerate copies, and that only a few putatively intact copies were found. Moreover, analysis of the Physcomitrella EST database and RT-PCR analyses of four additional species showed that the elements are transcriptionally inactive in mosses, except for PpatensLTR1 elements, for which one EST match was found. It was shown previously that chromodomaincontaining retrotransposons are under-represented in the plant, fungal and vertebrate EST databases, as only a few families from the particular species were found to be transcribed (Gorinsek et al., 2004). Nevertheless, EST matches to the plant chromodomain-containing Gypsy elements have been found in all of the available seed plants, indicating that transcriptionally active chromodomaincontaining retrotransposons are present in plant genomes (Kordis, 2005). Understanding the evolutionary history of bryophyte genomes is compelling because they represent an exceptional group of plants. They are the most ancient living land plants, which evolved numerous important adaptations, including the domination of the haploid (gametophytic) generation in the life cycle, specialized gametangia and desiccation-resistant spore walls (Renzaglia et al., 2002). Although the gene composition of the moss Physcomitrella resembles that of Arabidopsis, there are significant differences, including genes with intriguing functions, such as signal transduction and cytotoxicity (Rensing et al., 2005). The addition of information about biodiversity, distribution and organization of the LTR retrotransposons, as one of the most abundant genomic components, is very important for the further understanding of genome structure and evolution.

material). Such models are constructed with position-specific scores for amino acids and position-specific penalties for opening and extending an insertion or deletion, and represent a statistical description of a certain multiple alignment. Profile HMMs can be used for searching for additional remote homologous sequences. The nucleotide sequences of the elements of interest were also extracted with the assistance of the UniPro GenomeBrowser. After the localization of amino acid sequences obtained during the HMMER search in the initial genome in its nucleotide representation, the sequences were expanded to LTRs using the ‘Repeat find’ options in the UniPro GenomeBrowser. The BLAST analysis was essentially performed using sequence databases accessible from the National Center for Biotechnology Information (NCBI) server (http://www.ncbi.nlm.nih.gov/BLAST). Other websites used were: PHYSCObase (http://moss.nibb.ac.jp/ dnadb.html); NCBI conserved domain database and search service (http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml); the gypsy database of mobile genetic elements (http://gydb.uv.es) and RetrOryza (the Rice LTR Retrotransposon database, http://www. retroryza.org; Chaparro et al., 2007). The Physcomitrella EST database is available at PHYSCObase (http://moss.nibb.ac.jp/dnadb. html) and contains 102 553 putative transcripts and 22 885 clusters (Nishiyama et al., 2003). The BLASTN service, which is available at http://moss.nibb.ac.jp/blast/blast.html, was used to search the EST database. Multiple DNA alignments were performed by CLUSTALW (Thompson et al., 1994), and were edited manually. Phylogenetic analyses were performed using the NJ method in the MEGA 3.0 program (Kumar et al., 2004). The statistical support for the tree was evaluated by bootstrapping (Felsenstein, 1985).

Species collection and total DNA isolation

Experimental procedures

Table 2 lists moss species, and Table 3 lists the other plant species used in this study. The taxonomy of mosses follows that of Ignatov et al. (2006). Species were collected in the wild and were then dried. The detailed labeling data are available from the authors. The genomic DNA of P. patens subsp. patens was kindly provided by Dr Yasuhiko Sekine (College of Science, Rikkyo University, Tokyo, Japan). Genomic DNA was isolated from the gametophyte (moss) and leaves. Extraction was performed using the CTAB method (Murray and Thompson, 1980).

Genomic sequence screening, sequence and phylogenetic analysis

Gypsy PCR amplification and sequencing

The data for P. patens genomic sequences were directly obtained from the DOE Joint Genome Institute (JGI, http://genome. jgi-psf.org; July 2007). We used the UniPro GenomeBrowser (http://genome.unipro.ru) for the identification of chromodomaincontaining elements. The investigated genome was translated over six possible reading frames to the protein form, for which the searching of homologous regions was performed using the ‘HMMER search’ options in the UniPro GenomeBrowser. The algorithm of the HMMER search is based on profile hidden Markov models (HMM), which can perform amino acid sequence searches using an appropriate profile (Eddy, 1998; McClure et al., 1996). For the analysis, we used a multiple alignment consensus sequence, which contains information about RT and Int domains. The profile HMM, based on this consensus sequence, was also built using UniPro GenomeBrowser software (see Supplementary

Degenerate PCR primers for chromodomain-containing Gypsy LTR retrotransposons were designed based on the conserved amino acid sequences in the reverse transcriptase (RT) domains of different published LTR retroelements. Primer sequences were as follows: ty3-S, 5¢-AATTCTGGCACTTTTCGACTNTGYRTNGAYTA-3¢ and ty3-A, 5¢-AATTCGCTGCCGCTAAAGATNARNADRTCRTC-3¢, where Y = C + T, R = A + G, D = A + G + T and N = A + G + C + T. The expected length of PCR products was about 330 bp. PCR amplification with degenerate primers was performed using 0.1 lg of genomic DNA in a 10-ll volume of 10 mM Tris–HCl (pH 8.9), 1 mM (NH4)2SO4, 4 mM MgCl2, 200 lM each of four dNTPs, 0.5 lM primers and 2.5 U of Taq polymerase. After an initial denaturation step for 3 min at 94C, the PCR reactions were subjected to 30 cycles of amplification, consisting of 30-s denaturation at 94C, 42-s annealing at 52C and a 1-min extension at 72C. PCR products were separated by agarose gel electrophoresis. The resulting PCR products

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

Gypsy elements from mosses 573 were directly ligated into a pGEM vector using a pGEM-T-Easy cloning kit (Promega, http://www.promega.com) for sequence determination. Clones were amplified by PCR with M13 primers, and 40 ng of the product was used in a 10-ll cycle sequencing reaction with the ABI BigDye Terminator Kit on an ABI 310 Genetic Analyser (Applied Biosystems, http://www.appliedbiosystems.com), or sequencing reactions were performed with the Dye Terminator Cycle Sequencing Kit (Beckman Coulter, http://www.beckmancoulter.com), and were analyzed on a CEQ 8000 Genetic Analysis System (Beckman Coulter). Sequences were deposited in GenBank under accession numbers AY959213–AY959270, DQ054442, DQ054443, DQ054454– DQ054458, DQ206457, DQ206458 and EU408372–EU408471.

Table S1. List of scaffolds used in the reconstruction of the PpatensLTR1 element, and locations of PpatensLTR1 fragments. Appendix S1. Nucleotide sequence of the reconstructed PpatensLTR1 element from Physcomitrella patens, with annotations. Appendix S2. Nucleotide sequences of PpatensLTR elements from Physcomitrella patens, with annotations. Appendix S3. The profile HMM used for the isolation of retroelements from the genomic sequence of Physcomitrella patens. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

References Dot-blot hybridization Dot-blots were prepared by applying genomic DNA (500 and 100 ng) to Hybond-N+ (Amersham, http://www.amersham.com) nylon membrane using the alkali blotting protocol as described in the user’s manual, and were crosslinked by UV light. The blots were hybridized for 1 h at 52C in a solution containing QuikHyb Hybridization Solution (Stratagene, http://www.stratagene.com), 100 lg ml)1 fragmented and denatured herring sperm DNA and 32 P-labeled DNA probe labeled by random priming using DNA Polymerase I Large (Klenow) Fragment Mini Kit (Promega). The membrane was washed twice in a solution containing 0.1 · SSC and 0.1% SDS for 45 min at 52C, and was then exposed to X-ray film.

Total RNA isolation and RT-PCR Total RNA from the gametophyte was extracted using the SV Total RNA Isolation System (Promega). RT-PCR was carried out with 1 lg of RNA template using the SuperScript III One-Step RT-PCR System with Platinum Taq DNA Polymerase (Invitrogen, http:// www.invitrogen.com). All PCR conditions were as described in the user’s manual. The degenerate primers ty3-S and ty3-A were used to amplify RT domains (annealing temperature 52C), and primer pair Pp15-S and Pp15-A, described by Akita and Valkonen (2002), was used to amplify a fragment of the R gene (annealing temperature 55C).

Acknowledgements The P. patens sequence data were produced by the US Department of Energy Joint Genome Institute (http://www.jgi.doe.gov). The genomic DNA of P. patens subsp. patens was kindly provided by Dr Yasuhiko Sekine (College of Science, Rikkyo University, Tokyo, Japan). This work was supported in part by state contract 10002251/P-25/155-270/200404-082 and the Siberian Branch of the Russian Academy of Sciences (project No. 10.4). We are grateful to Prof. Victor Fet (Marshall University, http://www.marshall.edu) for his critical reading of the manuscript.

Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Detailed neighbor-joining (NJ) phylogenetic tree based on reverse transcriptase (RT) nucleotide sequences of long terminal repeat (LTR) retrotransposons, including newly described elements from the mosses represented in Figure 3.

Akita, M. and Valkonen, J.P. (2002) A novel gene family in moss (Physcomitrella patens) shows sequence homology and a phylogenetic relationship with the TIR-NBS class of plant disease resistance genes. J. Mol. Evol. 55, 595–605. Bowen, N.J. and Jordan, I.K. (2002) Transposable elements and the evolution of eukaryotic complexity. Curr. Issues Mol. Biol. 4, 65– 76. Chaparro, C., Guyot, R., Zuccolo, A., Pie´gu, B. and Panaud, O. (2007) RetrOryza: a database of the rice LTR-retrotransposons. Nucleic Acids Res. 35(Database issue), D66–D70. Dej, K.J., Gerasimova, T., Corces, V.G. and Boeke, J.D. (1998) A hotspot for the Drosophila gypsy retroelement in the ovo locus. Nucleic Acids Res. 26, 4019–4025. Deka, N., Willard, C.R., Wong, E. and Schmid, C.W. (1988) Human transposon-like elements insert at a preferred target site: evidence for a retrovirally mediated process. Nucleic Acids Res. 16, 1143–1151. Eddy, S.R. (1998) Profile hidden Markov models. Bioinformatics, 14, 755–763. Fedoroff, N.V. (1999) Transposable elements as a molecular evolutionary force. Ann. N Y Acad. Sci. 870, 251–264. Felsenstein, J. (1985) Confidence limits on phylogenies: an approach using the bootstrap. Evolution, 39, 783–791. Feschotte, C., Jiang, N. and Wessler, S.R. (2002) Plant transposable elements: where genetics meets genomics. Genetics, 3, 329–341. Finnegan, D.J. (1992) Transposable elements. Curr. Opin. Genet. Dev. 2, 861–867. Friesen, N., Brandes, A. and Heslop-Harrison, J.S. (2001) Diversity, origin, and distribution of retrotransposons (gypsy and copia) in conifers. Mol. Biol. Evol. 18, 1176–1188. Gao, X., Hou, Y., Ebina, H., Levin, H.L. and Voytas, D.F. (2008) Chromodomains direct integration of retrotransposons to heterochromatin. Genome Res. 18, 359–369. Goodwin, T.J. and Poulter, R.T. (2001) The DIRS1 group of retrotransposons. Mol. Biol. Evol. 18, 2067–2082. Gorinsek, B., Gubensek, F. and Kordis, D. (2004) Evolutionary genomics of chromoviruses in eukaryotes. Mol. Biol. Evol. 21, 781–798. Havecker, E.R., Gao, X. and Voytas, D.F. (2004) The diversity of LTR retrotransposons. Genome Biol. 5, 225. Hirochika, H. and Hirochika, R. (1993) Ty1-copia group retrotransposons as ubiquitous components of plant genomes. Jpn. J. Genet. 68, 35–46. Hirochika, H., Fukuchi, A. and Kikuchi, F. (1992) Retrotransposon families in rice. Mol. Gen. Genet. 233, 209–216. Ignatov, M.S., Afonina, O.M., Ignatova, E.A. et al. (2006) Checklist of mosses of East Europe and North Asia. Arctoa, 15, 1–130.

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574

574 Olga Novikova et al. Jordan, I.K., Matyunina, L.V. and McDonald, J.F. (1999) Evidence for the recent horizontal transfer of long terminal repeat retrotransposon. Proc. Natl Acad. Sci. USA, 96, 12621–12625. Kidwell, M.G. and Lisch, D.R. (2000) Transposable elements and host genome evolution. Trends Ecol. Evol. 15, 95–99. Kim, J.M., Vanguri, S., Boeke, J.D., Gabriel, A. and Voytas, D.F. (1998) Transposable elements and genome organization: a comprehensive survey of retrotransposons revealed by the complete Saccharomyces cerevisiae genome sequence. Genome Res. 8, 464–478. Kordis, D. (2005) A genomic perspective on the chromodomaincontaining retrotransposons: chromoviruses. Gene, 347, 161–173. Kumar, A. and Bennetzen, J.L. (1999) Plant retrotransposons. Annu. Rev. Genet. 33, 479–532. Kumar, S., Tamura, K. and Nei, M. (2004) MEGA3: Integrated software for Molecular Evolutionary Genetics Analysis and sequence alignment. Brief Bioinform. 5, 150–163. Kumekawa, N., Ohtsubo, E. and Ohtsubo, H. (1999) Identification and phylogenetic analysis of gypsy-type retrotransposons in the plant kingdom. Genes Genet. Syst. 74, 299–307. Langdon, T., Seago, C., Mende, M., Legett, M., Thomas, H., Forster, J.W., Thomas, H., Jones, R.N. and Jenkins, G. (2000) Retrotransposon evolution in diverse plant genomes. Genetics, 156, 313–325. Liu, R., Vitte, C., Ma, J., Mahama, A.A., Dhliwayo, T., Lee, M. and Bennetzen, J.L. (2007) A GeneTrek analysis of the maize genome. Proc. Natl Acad. Sci. USA, 104, 11844–11849. Malik, H.S. and Eickbush, T.H. (1999) Modular evolution of the integrase domain in the Ty3/gypsy class of LTR retrotransposons. J. Virol. 73, 5186–5190. Marı´n, I. and Llore´ns, C. (2000) Ty3/Gypsy retrotransposons: description of new Arabidopsis thaliana elements and evolutionary perspectives derived from comparative genomic data. Mol. Biol. Evol. 17, 1040–1049. McClure, M.A., Smith, C. and Elton, P. (1996) Parameterization studies for the SAM and HMMER methods of hidden Markov model generation. Proc. Int. Conf. Intell. Syst. Mol. Biol. 4, 155–164. Miller, J.T., Dong, F., Jackson, S.A., Song, J. and Jiang, J. (1998) Retrotransposon-related DNA sequences in the centromeres of grass chromosomes. Genetics, 150, 1615–1623. Murray, M.G. and Thompson, W.F. (1980) Rapid isolation of high molecular weight plant DNA. Nucleic Acids Res. 8, 4321–4325. Nishiyama, T., Fujita, T., Shin-I, T. et al. (2003) Comparative genomics of Physcomitrella patens gametophytic transcriptome and Arabidopsis thaliana: implication for land plant evolution. Proc. Natl Acad. Sci. USA, 100, 8007–8012. Peterson-Burch, B.D. and Voytas, D.F. (2002) Genes of the Pseudoviridae (Ty1/copia retrotransposons). Mol. Biol. Evol. 19, 1832– 1845. Rensing, S.A., Fritzowsky, D., Lang, D. and Reski, R. (2005) Protein encoding genes in an ancient plant: analysis of codon usage,

retained genes and splice sites in a moss, Physcomitrella patens. BMC Genomics, 6, 43. Rensing, S.A., Lang, D., Zimmer, A.D. et al. (2008) The Physcomitrella genome reveals evolutionary insights into the conquest of land by plants. Science, 319, 64–69. Renzaglia, K.S., Duff, R.J.T., Nickrent, D.L. and Garbary, D.J. (2002) Vegetative and reproductive innovations of early land plants: implications for a unified phylogeny. Philos. Trans. R. Soc. Lond. B Biol. Sci. 355, 769–793. Reski, R. and Frank, W. (2005) Moss (Physcomitrella patens) functional genomics – Gene discovery and tool development, with implications for crop plants and human health. Brief Funct. Genomic Proteomic, 4, 48–57. Stenøien, H.K. (2007) Compact genes are highly expressed in the moss Physcomitrella patens. J. Evol. Biol. 20, 1223–1229. Suoniemi, A., Taskanen, J. and Schulman, A.H. (1998) Gypsy-like retrotransposons are widespread in the plant kingdom. Plant J. 13, 699–705. Telesnitsky, A. and Goff, S.P. (1997) Reverse transcriptase and the generation of retroviral DNA. In Retroviruses (Coffin, J.M., Hughes, S.H. and Varmus, H.E., eds). Cold Spring Harbor: Cold Spring Laboratory Press, pp. 121–160. Terol, J., Castillo, M.C., Bargues, M., Pe´rez-Alonso, M. and de Frutos, R. (2001) Structural and evolutionary analysis of the copialike elements in the Arabidopsis thaliana genome. Mol. Biol. Evol. 18, 882–892. Terzian, C., Ferraz, C., Demaille, J. and Bucheton, A. (2000) Evolution of the Gypsy endogenous retrovirus in the Drosophila melanogaster subgroup. Mol. Biol. Evol. 17, 908–914. Thompson, J.D., Higgins, D.G. and Gibson, T.J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673– 4680. Vallee, B.L. and Auld, D.S. (1993) Cocatalytic zinc motifs in enzyme catalysis. Proc. Natl Acad. Sci. USA, 90, 2715–2718. Vitte, C. and Panaud, O. (2005) LTR retrotransposons and flowering plant genome size: emergence of the increase/decrease model. Cytogenet. Genome Res. 110, 91–107. Wicker, T., Stein, N., Albar, L., Feuillet, C., Schlagenhauf, E. and Keller, B. (2001) Analysis of a contiguous 211 kb sequence in diploid wheat (Triticum monococcum L.) reveals multiple mechanisms of genome evolution. Plant J. 26, 307–316. Wicker, T., Sabot, F., Hua-Van, A. et al. (2007) A unified classification system for eukaryotic transposable elements. Nat. Rev. Genet. 8, 973–982. Wright, D.A. and Voytas, D.F. (1998) Potential retroviruses in plants: Tat1 is related to a group of Arabidopsis thaliana Ty3/gypsy retrotransposons that encode envelope-like proteins. Genetics, 149, 703–715.

Accession numbers: AY959213–AY959270; DQ054442; DQ054443; DQ054454–DQ054458; DQ206457; DQ206458; EU408372– EU408471.

ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 56, 562–574