An active transposable element, Herves, from the African malaria mosquito Anopheles gambiae

Abstract

Transposable elements have proven to be invaluable tools for genetically manipulating a wide variety of plants, animals, and microbes. Some have suggested that they could be used to spread desirable genes, such as refractoriness to Plasmodium infection, through target populations of Anopheles gambiae, thereby disabling the mosquito's ability to transmit malaria. To achieve this, a transposon must remain mobile and intact after the initial introduction into the genome. Endogenous, active class II transposable elements from An. gambiae have not been exploited as gene vectors/drivers because none have been isolated. We report the discovery of an active class II transposable element, Herves, from the mosquito An. gambiae. Herves is a member of a distinct subfamily of hAT elements that includes the hopper-we element from Bactrocera dorsalis and B. cucurbitae. Herves was transpositionally active in mobility assays performed in Drosophila melanogaster S2 cells and developing embryos and was used as a germ-line transformation vector in D. melanogaster. Herves displays an altered target-site preference from the distantly related hAT elements, Hermes and hobo. Herves is also present in An. arabiensis and An. merus with copy numbers similar to that found in An. gambiae. Preliminary data from an East African population are consistent with the element being transpositionally active in mosquitoes.

Figure 1.—

Consensus sequence of three Herves elements from the An. gambiae genome (see text) and amino acid translation of the transposase. IUPAC ambiguity codes indicate sites where the three copies differ; lowercase letters indicate a deletion in at least one copy. Inverted terminal repeats are underlined. A region containing three almost-perfect tandem repeats is shaded.

Figure 2.—

Phylogenetic relationships of transposase amino acid sequences from selected hAT elements (see text). Boxed numbers next to each node represent measures of nodal support. The top number is the quartet puzzling reliability percentage from TREE-PUZZLE using the WAG matrix substitution model and among-site rate variation estimated using a discrete gamma shape parameter. The bottom number in italics is the percentage of bootstrap support from unweighted maximum-parsimony analysis using PAUP*. Tree topology and branch lengths correspond to the tree obtained with TREE-PUZZLE from 100 maximum-likelihood quartets. All transposase sequences except that of Herves were obtained from GenBank (http://www.ncbi.nih.gov/): Tam3 (accession no. CAA38906), Ac (accession no. CAA29005), hobo (accession no. A39652), Hermes (accession no. AAC37217), Homer (accession no. AAD03082), hermit (accession no. AAA64851), Buster3 (accession no. NP_071373), Tip100 (accession no. BAA36225), Tag1 (accession no. T52187), Ac-like (accession no. NP 509255), hopper (accession no. AAL93203), and Tramp (accession no. CAA76545).

Figure 3.—

Evidence of past mobility of Herves in An. gambiae PEST and RSP strains. Sequences are grouped by locus under the heading of the scaffold from the An. gambiae sequencing project. PEST sequences were obtained from GenBank while RSP sequences were from PCR amplifications; the two RSP lines for scaffold AAAB01008978 represent different PCR amplification products. Herves ITRs are underlined, omitted Herves nucleotides are represented by (…), and dashes indicate insertions/deletions.

Figure 4.—

Transposable element display of the left end of Herves in transgenic D. melanogaster lines 1, 2, and 3. The results from two individuals from each line are shown along with the position of the molecular weight makers. Bands were excised, eluted, reamplified, and sequenced. Chromosomal position was determined by BLAST searching the D. melanogaster sequence database. Bands below the main bands (arrows) are identical in sequence to the main bands and are often associated with abundant PCR products <250 bp in length. The sequences adjacent to the right inverted terminal repeat were determined by direct amplification using primers specific for the Herves element and flanking genomic DNA. The 8-bp target-site duplications flanking each of the three independent transpositions are shown to the right and each corresponds to the arrowed band at the corresponding migration distance on the gel. For line 1, only sequence flanking the left end of Herves was obtained. For line 2, the 8-bp target-site duplication was imperfect with the single-base-pair difference (T vs. G) shown in boldface type.

Figure 5.—

Comparison of the Herves content of laboratory lines of An. gambiae s.s. Transposable element display was performed on individuals from the laboratory lines G3, RSP (R), Suakoko obtained from two sources (S1 and S2), KIL obtained from two sources (K1 and K2), and two recently established isofemale lines from individuals collected from Mali (M1 and M2). Molecular weight markers are indicated.

Figure 6.—

Transposable element display of individuals collected from a local population in Mozambique. Three species are represented: An. arabiensis, An. gambiae s.s., and An. merus. Molecular weight markers (in kilobase pairs) are shown.

Figure 7.—

Herves in three species of Anopheles in Mozambique. Data were derived from transposable element displays on the number of individuals indicated (n). “Site” refers to a genomic position occupied by at least one Herves element in the sample and corresponds to a unique PCR product on a transposable element display. Sites are arbitrarily numbered and are not similar between species. “Site occupancy” refers to the total number of elements within a sample divided by the number of sites.