The DIRS1 Group of Retrotransposons

Abstract

Only three retrotransposons of the DIRS1 group have previously been described: DIRS1 from the slime mold Dictyostelium discoideum, PAT from the nematode Panagrellus redivivus, and Prt1 from the zygomycetous fungus Phycomyces blakesleeanus. Analyses of the reverse transcriptase sequences encoded by these elements suggest that they are related to the long terminal repeat (LTR) retroelements, such as the Ty3/gypsy retrotransposons and the vertebrate retroviruses. The DIRS1-group elements, however, have several unusual structural features which distinguish them from typical LTR elements: (1) they lack the capacity to encode DDE-type integrases or aspartic proteases; (2) they have open reading frames (ORFs) of unknown function; (3) they integrate without creating duplications of their target sites; and (4) although they are bordered by terminal repeats, these sequences differ from typical LTRs in that they are either inverted repeats or "split" direct repeats. Because of the small number of DIRS1-like elements described, and the unusual structures of these elements, little is known about their evolution, distribution, and replication mechanisms. Here, we report the identification of several new DIRS1-like retrotransposons, including elements from nematodes, sea urchins, fish, and amphibia. We also present evidence for the existence of DIRS1-like sequences in the human genome. In addition, we show that the lack of DDE-type integrase genes from elements of the DIRS1 group is explained by the finding that the previously uncharacterized ORFs of these elements encode proteins related to the site-specific recombinase of bacteriophage lambda. The presence of lambda-recombinase-like genes in DIRS1 elements also accounts for the lack of target-site duplications for these elements and may be related to the unusual structures of their terminal repeats.

Cryptons: a group of tyrosine-recombinase-encoding DNA transposons from pathogenic fungi

“…For instance, for many cryptons we found that the extreme termini contain a short sequence (4-6 bp) which is repeated at either end, and this same sequence appears in the related empty sites. A similar situation was found for DIRS1-like elements (Goodwin & Poulter, 2001a) and also occurs with at least some elements from the Ngaro1 group (T. J. D. Goodwin, unpublished data). For the DIRS1 elements, we interpreted these findings as suggesting that the extrachromosomal transposition intermediates of these elements integrate into the host genome by recombination between a short sequence at the circular junction of the element's termini and an identical sequence in the target site.…”

“…Percentages of bootstrap support, from 1000 replicates, are indicated for branches receiving >40 % support. The sources of many sequences are given in a previous report (Goodwin & Poulter, 2001a). Others are listed in Methods.…”

“…6). As has been noted earlier (Esposito & Scocca, 1997;Goodwin & Poulter, 2001a;Jacobs et al, 2003), phylogenetic analyses of tyrosine recombinases are hindered by the high level of sequence diversity within the group. This usually means that the nature of the more distant relationships within the group cannot be resolved with certainty using current techniques.…”

“…Recently, however, we reported that members of the DIRS1 group of LTR retrotransposons lack genes for DDE-type integrases, but instead encode tyrosine (or lambda) recombinases (Goodwin & Poulter, 2001a). These enzymes are likely to be involved in the insertion of DIRS1-like elements into their hosts' genomes.…”

The Unusual Phylogenetic Distribution of Retrotransposons: A Hypothesis

“…The mechanisms of these two types of retoelements are summarized briefly here and in Figure 2. In addition, two smaller retrotransposon families, the DIRS1 (Goodwin and Poulter 2001) and BEL (Malik et al 2000) groups, appear to be distinct, but are much less widely distributed, and thus, will not be discussed further here.…”