Fig 1.
Number of predicted AST-AR and peptide genes identified in arthropods and C. elegans.
Accession numbers are available in S1 Table. The number of AST-A peptides is indicated within brackets and references are provided. The T. urticae, D. plexippus, H. melpomene, S. invicta and A. darlingi AST-A peptides were predicted by comparison with the insect homologues and identification of the C-terminal FGL-amide motif. * indicates species in which a putative AST-AR pseudogene (orthologue of the third Culicidae AST-AR gene) was identified. Data from D. pulex and A. cephalotes obtained from [115, 116].
Fig 2.
AST-A peptide precursor in A. gambiae.
The deduced sequence of AST-A in A. gambiae (Aga, PEST) was obtained from the AGAP003712 gene and confirmed using EST data. The A. aegypti (Aae, AAEL015251,[81]) and D. melanogaster (Dme, FBgn0015591,[48]) orthologues were used for comparisons. The predicted mature peptides are highlighted in bold and the Gly residues processed to the C-terminal amide in mature AST-A’s are indicated in italics.
Fig 3.
Phylogeny of the AST-AR with the KISSR and GALR.
Phylogenetic analysis was performed using the ML method and three subsets of the same phylogenetic tree showing the expansion of the different family members (A, B and C) are represented to facilitate interpretation. Only bootstrap support values above 50% are indicated. In the most important receptor family nodes statistical support was established using the aLRT SH-like test and is indicated (bootstrap method/ aLRT SH-like test). The deduced A. darlingi (Scaffold_325) was not used, as the receptor sequence was very incomplete and only 3 TM domains were predicted. The phylogenetic tree was rooted with the vertebrate GPR151 cluster (12 sequences). Species names and accession numbers of the receptor genes are available in S1 Table.
Table 1.
Sequence identity and similarity of insect AST-ARs with human GALR1 and KISSR1.
Fig 4.
Conserved gene synteny of the A. gambiae, D. melanogaster and C. elegans AST-AR genome regions with the human KISSR1 chromosomes.
Conservation for T. castaneum is also shown. Horizontal lines represent chromosome fragments and block arrows indicate genes and orientation in the genome. Orthologue genes are represented in the same colour and their position (Mb) is indicated. An arrow with red stripes represents the putative AST-AR pseudogene (AGAP001774) localized near GPRALS2. Dotted boxes represent the absent human KISSR genes (that emerged during early vertebrate tetraploidizations) [67,74] and the T. castaneum AST-AR gene. Note that the mosquito 2R and human ch19 have been divided into two parts (pt1 and pt2) to facilitate visualization. Only shared genes are represented. The number of family members that map to the same chromosome is indicated and the closest to AST-AR and KISSR1 is represented. A full description of gene families and names and accession numbers is given in S3 Table.
Fig 5.
Conserved gene synteny of the genome regions containing AST-A gene in A. gambiae, D. melanogaster and C. elegans (npl-5 and npl-6) compared to the human KISS/GAL/SPX chromosomes.
The gene homologues in T. castaneum are also represented. Horizontal lines indicate chromosome fragments and coloured arrow identify genes and their orientation in the genome. Orthologue genes are indicated in the same colour and their positions are indicated below (Mb). Dotted boxes represent the absent human KISS and SPX2 genes (that emerged during early vertebrate tetraploidizations) [42,74] and the T. castaneum AST-A gene. Only shared gene family members are represented. The number of family members that map to the same chromosome is indicated and those closest to AST-A and KISS/GAL/SPX are represented. A full description of gene families and names and accession numbers is given in S4 Table.
Fig 6.
Gene organisation of the AST-A receptors in Anopheles and D. melanogaster.
The structure of the Anopheles receptor genes was deduced from the consensus organisation obtained from several mosquito genomes (S5 and S6 Tables). The D. melanogaster AST-ARs gene organizations were obtained from ENSEMBL. In the A. gambiae PEST genome duplicated exons highly similar in sequence to GPRALS1 (exon 1) and GPRALS2 (exon 2, 3 and 4) are predicted and are not represented. Closed boxes represent exons and dashed lines introns. Mosquito exons are numbered and exons encoding the UTR are represented by pink boxes. Gene structures were built using FancyGene 1.4 software. The figure is not drawn to scale.
Fig 7.
Sequence conservation of the duplicate dipteran AST-ARs with the insect and human orthologues.
The predicted seven transmembrane domains are boxed in red and numbered. Potential sites for N-glycosylation are underlined in the N-terminal region and two conserved motifs D-R-Y/F localized after TM3 and NSxxNPxxY within TM7 are annotated with asterisks [87,88]. Two conserved cysteine residues that may form a disulphide bond were identified are connected by a line; predicted residues involved in protein kinase C phosphorylation are indicated by a blue square and potential protein A phosphorylation sites are annotated by a green diamond; C-terminal cysteine residues for potential palmitoylation after TM7 are denoted in italics and indicated with an orange pentagon. Amino acids important for binding of human galanin to GALR1 are indicated in red. The arginine residue important for the function of human KISSR1 that is proximate to the end of TM7 is indicated in bold and circled. Shading denotes amino acid conservation and dark grey means 80% and black 100% conservation. Shading after TM4 was manually edited and did not considered the incomplete receptor regions * indicate incomplete mosquito receptor sequences. Accession numbers of receptor genes are available in S1 Table.
Fig 8.
Amino acid sequence alignment of the dipteran AST-A mature peptides with the vertebrate KISS, GAL and SPX family members.
The highly conserved FGL motif between AST-A and KISS peptides is indicated in bold and red and conserved N residues in bold and blue. Sequence conservation of GAL and SPX is indicated in italics and totally conserved are in italics and bold. The vertebrate predicted peptide sequences were obtained from [74] and [42] and the Xenopus laevis mature galanin peptide deduced from EU446417.1.
Fig 9.
Tissue distribution and effect of a blood meal on the expression of the paralogue GPRALS and AST-A transcripts in the female A. coluzzii.
Expression was analysed in the head, fat body, midgut and ovaries 3 hours after a glucose (white bars) or blood (grey bars) meals. Receptor expression levels were normalized using the geometric mean of two reference genes (S7 and MC). The results are represented as mean ± SEM of three biological replicates with the exception of ovaries where only a single biological replicate was analysed (~60 ovaries). Prism GraphPad v5 software was used to assess the significance of differences between experimental groups using the Mann-Whitney (two-tailed) test (*p < 0.05).
Fig 10.
Capacity of the insect AST-A peptides to activate the A. coluzzii GPRALS.
The mosquito Ano_AST-A1 and Ano_AST-A2 peptides and the cockroach BLAST-2 peptide were tested at several different concentrations and the response of the receptor monitored by measuring concentrations of intracellular calcium (RFU). The D. melanogaster DAR-2-RA receptor was used as a positive control: A) Response to BLAST-2 peptide (0.5 μM to 0.005 μM); B) Receptor response to the presence of decreasing concentrations of Ano_AST-A1 and Ano_AST-A2 peptides (1 μM to 1nM). A Kruskal-Wallis test with a Dunn’s Multiple Comparison test was performed using Prism GraphPad version 5 software. No significant differences were found.
Fig 11.
Proposed model for the origin and evolution of the AST-AR genes.
Circles with different colours represent the AST-AR (light blue), KISSR (green) and GALR (pink) family members. The tetraploidization events basal to vertebrate radiation (1R, 2R) and the teleost specific genome duplication (3R) are indicated. The circle with a cross indicates gene loss during evolution. Numbers within the circles indicate predicted gene numbers of each family. Gene number from early deuterostome and lophotrochozoa representatives were obtained from [41]. For simplicity, lineage-specific duplications are not indicated and the time line is not drawn to scale. Receptor mapping for the vertebrate ancestral chromosomes (VAC) was obtained from [42].