A search for reverse transcriptase-coding sequences reveals new non-LTR retrotransposons in the genome of Drosophila melanogaster

Abstract

Background: Non-long terminal repeat (non-LTR) retrotransposons are eukaryotic mobile genetic elements that transpose by reverse transcription of an RNA intermediate. We have performed a systematic search for sequences matching the characteristic reverse transcriptase domain of non-LTR retrotransposons in the sequenced regions of the Drosophila melanogaster genome.

Results: In addition to previously characterized BS, Doc, F, G, I and Jockey elements, we have identified new non-LTR retrotransposons: Waldo, You and JuanDm. Waldo elements are related to mosquito RTI elements. You to the Drosophila I factor, and JuanDm to mosquito Juan-A and Juan-C. Interestingly, all JuanDm elements are highly homogeneous in sequence, suggesting that they are recent components of the Drosophila genome.

Conclusions: The genome of D. melanogaster contains at least ten families of non-site-specific non-LTR retrotransposons representing three distinct clades. Many of these families contain potentially active members. Fine evolutionary analyses must await the more accurate sequences that are expected in the next future.

Figure 1

Reverse transcriptase sequences found in the genome of Drosophila melanogaster. Phylogenetic trees of sequences matching the full-length hidden Markov model of reverse transcriptase as defined by HMMER 2.1.1 that were identified in the analysis of (a) Release 1 and (b) BDGP/EDGP sequences. Names on branches represent GenBank accession numbers followed by coordinates in the respective sequence, which confine the HMM match (start coordinate greater than end means reverse complement of a sequence). Distinct families are separated by clades according to [2]. The tree was constructed by the neighbor-joining method as implemented in CLUSTAL W software. Numbers at the nodes represent bootstrap values as percentages out of 500 replicates and are shown only for values greater than 50%. Because of redundancy present in the BDGP/EDGP data set, absolutely identical sequences were substituted by one representative each in building the tree in (b).

Figure 2

Phylogeny of non-LTR elements based on their reverse transcriptase domain. Drosophila melanogaster non-LTR retrotransposons are shown in bold on the phylogenetic tree based on reverse transcriptase sequences [2]. Sequence alignment was produced by CLUSTAL W software by adding Waldo, JuanDm and You sequences to the alignment DS36752 [2]. Numbers at the nodes represent bootstrap values as percentages out of 500 replicates. The unrooted neighbor-joining tree and bootstrap analysis were inferred as implemented in the MEGA package [57].

Figure 3

Structure of JuanDm elements in strain y; cn bw sp. (a) The full-size JuanDm element is represented as a white box, with two overlapping ORFs indicated as arrows. Positions of SpeI and ClaI restriction sites are shown and the black box above the element indicates the PCR fragment used as a probe in Figure 5a. Thick lines below represent the seven completely sequenced JuanDm elements found in Release 1. The internal deletion in element 5 is indicated as a gray region. (b) Sequences of integration sites of JuanDm elements shown in (a). Target site duplications are shown in bold and underlined. The 5'-most and 3'-most nucleotides of each element are shown in italics and separated by dots. For full-length elements, alignments in 5' region have been made to show variation in the 5' junction. The sequences of elements correspond to the following coordinates in release 1: 1, 13235891-13240123 in AE002787; 2, reverse complement of 7828981-7833212 in AE002566; 3, 4591855 4596087 in AE002593; 4, 268307-272538 in AE002629; 5, reverse complement of 7363833-7366823 in AE002708; 6, 874492-877244 in AE002638; 7, 2582603-2583169 in AE002602.

Figure 4

Alignments of complete amino acid sequences of Juan ORF1 and ORF2. (a) ORF1; (b) ORF2. Identical residues are in highlighted in black, conserved residues are shaded in dark gray, and similar residues are shaded in light gray. Alignments and shadings were performed with VectorNTI software. The sequence of Juan-A in Aedes aegypti is from accession number M95171, the sequence of Juan-C in Culex pipiens is from M91082, the sequence of JuanDm in D. melanogaster is from AC005452 (coordinates 3425-7660).

Figure 5

Southern blot analyses of JuanDm and You in various Drosophila species. Each lane contains 3 μg genomic DNA of the Drosophila melanogaster Cha strain (mel), D. simulans (sim), D. mauritiana (mau), D. teissieri (tei), D. yakuba (yak) and D. virilis (vir). (a) The genomic DNA was digested with SpeI and ClaI and hybridized with a JuanDm-specific probe (see Figure 3). (b) The genomic DNA was digested with SmaI and SacI and hybridized with a You-specific probe (see Figure 6).

Figure 6

Structure of You elements in strain y; cn bw sp. (a) The full-size You element is represented as a white box, with the two ORFs indicated as arrows. Positions of SmaI and SacI restriction sites are shown and the black box above the element indicates the PCR fragment used as a probe in Figure 5b. Thick lines below represent full-size (1-3) and 5' truncated (4) You elements found in Release 1. The internal deletion in element 3 is indicated as a gray region. (b) Sequences of integration sites of You elements shown in (a). Target site duplications are shown in bold and underlined. The most 5' and 3' nucleotides of each element are shown in italics and separated by dots. For full-length elements, alignments in 5' region have been made to show variation in the 5' junction. The sequences of elements corespond to following coordinates in release 1: 1, 56440-61816 in AE002620; 2, 2419411-2424785 in AE002647; 3, reverse complement of 1119136-1124111 in AE002566; 4, reverse complement of 5105-6524 in AE002601.

Figure 7

Alignments of complete amino acid sequences from You and I elements ORF1 and ORF2. (a) ORF1; (b) ORF2. Identical residues are highlighted in black, similar residues are in light gray. Alignment and shading was performed with VectorNTI software. The sequence of I is from accession number M14954 corrected according to [42], the sequence of You is from AL03893 (coordinates 1-5255) and AL022018 (coordinates 38180-38397) assembled manually.