Horizontal transmission of an R4 clade non-long terminal repeat retrotransposon between the divergent Aedes and Anopheles mosquito genera

Abstract

AaegR4_1 and AgamR4_1 are the sole R4 clade non-long terminal repeat (non-LTR) retrotransposons in Aedes aegypti and Anopheles gambiae, two species that diverged approximately 145-200 million years ago. Twelve full-length copies were found in Ae. aegypti and have less than 1% nucleotide (nt) divergence, suggesting recent activity on an evolutionary time scale. Five of these copies have intact open reading frames and the 3.6 kb open reading frame of AaegR4_1.1 has 78% nt identity to AgamR4_1.1. No intact copies were found in An. gambiae. Searches of 25 genomic databases for 22 mosquito species from three genera revealed R4 clade representatives in Aedes and Anopheles genera but not in Culex. Interestingly, these elements are present in all six species of the An. gambiae species complex that were searched but not in 13 other anopheline species. These results combined with divergence vs. age analysis suggest that horizontal transfer is the most likely explanation for the low divergence between R4 clade retrotransposon sequences of the divergent mosquito species from the Aedes and Anopheles genera. This is the first report of the horizontal transfer of an R4 clade non-LTR retrotransposon and the first report of the horizontal transfer of a non-LTR retrotransposon in mosquitoes.

Keywords: evolution; horizontal transfer; mosquito; non-long terminal repeat retrotransposon; transposable elements.

Figure 1. Schematic of AaegR4_1 structural organization

AaegR4_1.1 is approximately 4.0 kb and contains a single 3.6 kb ORF encoding reverse transcriptase (RT) and a restriction enzyme-like endonuclease (REL). Full-length copies are flanked by TAA tandem repeats. The location of a CCHC nucleic acid-binding motif is indicated.

Figure 2. Host species phylogeny and mosquito R4 clade sequence distribution

Mosquito phylogeny including 7 species of 3 genera based on the white gene. Shaded boxes show species containing R4 clade sequences. Ae. aeg. (Ae. aegypti); Ae. alb. (Ae. albopictus); C. qui. (C. quinquefasciatus); A. dar. (An. darlingi); An. sin. (A. sinensis); A. ste. (An. stephensi); A. gam. (An. gambiae). Divergence time between aedine and anopheline lineages of 145 to 200 Million Years Ago (MYA) is from Krzywinski et al., 2006. Maximum likelihood phylogeny was constructed using Phylogeny.fr online at http://www.phylogeny.fr/version2_cgi/simple_phylogeny.cgi. Scale bar shows substitutions per site. Branch support values are indicated above branches as percentages. Sequences used for phylogeny can be found in S2.

Figure 3. RT divergence vs. age of host divergence

Non-LTR retrotransposon RT sequences were aligned (Thompson et al., 1997) and distance values were determined by PAUP (Swofford, 2002). Points 1–10 represent arthropod comparisons taken from Malik, Burke, and Eickbush, 1999. Points a and b represent B. mori Dong compared to AgamR4_1.1 (a) and AaegR4_1.1 (b). Points c and d represent A. lumbricoides R4 compared to AgamR4_1.1 (c) and AaegR4_1.1 (d). Dipteran (mosquito)/Lepidopteran divergence is estimated at 340 MY (Gaunt and Miles, 2002), and 600 MY is used for A. lumbricoides/mosquito divergence. Divergence times are taken from Malik et al., 1999. Thirteen mosquito non-LTR retrotranspon comparisons representing all 10 non-LTR retrotransposon clades in An. gambiae (e–q) are shown in the rectangle where 145 MYA was used as a conservative estimate for Ae. aegypti and An. gambiae divergence. A logarithmic trendline is shown and statistics were calculated using log host age vs. amino acid divergence. Pearson’s correlation coefficient (r) is given with p-value. RStudent, Studentized deleted residual. An asterisk indicates a value based on the inclusion of only one of the 13 Ae. aegypti/An. gambiae non-LTR retrotranspon comparisons, the mosquito R4 clade elements (RStudent = 5.7). This removes the influence of other mosquito comparisons that may include paralogous sequences or sequences that have unusually high conservation. All sequences and distance values are available in S3.