Metazoan Ribosome Inactivating Protein encoding genes acquired by Horizontal Gene Transfer

Abstract

Ribosome inactivating proteins (RIPs) are RNA N-glycosidases that depurinate a specific adenine residue in the conserved sarcin/ricin loop of 28S rRNA. These enzymes are widely distributed among plants and their presence has also been confirmed in several bacterial species. Recently, we reported for the first time in silico evidence of RIP encoding genes in metazoans, in two closely related species of insects: Aedes aegypti and Culex quinquefasciatus. Here, we have experimentally confirmed the presence of these genes in mosquitoes and attempted to unveil their evolutionary history. A detailed study was conducted, including evaluation of taxonomic distribution, phylogenetic inferences and microsynteny analyses, indicating that mosquito RIP genes derived from a single Horizontal Gene Transfer (HGT) event, probably from a cyanobacterial donor species. Moreover, evolutionary analyses show that, after the HGT event, these genes evolved under purifying selection, strongly suggesting they play functional roles in these organisms.

Figure 1

Experimental confirmation of the presence and location of RIP gene in C. quinquefasciatus JHB genome. (A) Schematic representation of a fragment of the contig AAWU01015132 depicting the RIP gene (RIPcu) and its closest neighbor gene (XM_001850822). Expected amplicons and relevant EcoRI restriction sites are also presented. The intron of XM_001850822 is represented with a white box. (B) RIP ORF was amplified by PCR and the product was analyzed by gel electrophoresis before (lane 1) and after EcoRI treatment (lane 2). (C) A fragment of 1,882 bp linking the RIP gene with its neighbor-gene was amplified and electrophoresed before (lane 1) and after EcoRI treatment (lane 2).

Figure 2

Phylogenetic tree of RIP proteins family. Bayesian tree topology based on a matrix analysis of 133 proteins sequences with 209 informative sites is presented. Numbers above branches indicate PP support values. Bootstrap values (BS) >50% are shown below branches for nodes where topology of ML analysis was coincident with Bayesian inference. Lineages are indicated by different colors as follows: green (Plantae, including Mo: monocots; Eu: eudicots and Mg: magnoliids), blue (Fungi, including As: Ascomycota and Ba: Basidiomycota), red (Me: Metazoa), orange (Bacteria, including G+: Gram positive, G-: Gram negative, and excluding Cyanobacteria) and turquoise (Cy: Cyanobacteria). The clade of Culicinae RIPs is emphasized with red branches. Information related to the sequences used to infer these trees is available in Supplementary Table 2.

Figure 3

Schematic representation of the shared genomic context between C. quinquefasciatus (DS232037), Aedes aegypti (NW001810221) and Anopheles gambiae (chromosome 3L). Grey shadows link conserved syntenic ORFs. RIP genes are represented with orange arrows. The ORF 4 is equivalent to XM_001850822 and is absent in other scaffolds. Additional information about each gene is available in Supplementary Table 1.

Figure 4

Phylogenetic relationships among Culicinae and bacterial RIPs. Bayesian tree topology based on matrix analysis of 45 proteins sequences with 275 informative sites is presented. Numbers above branches indicate PP support values. Bootstrap values (BS) >50% are shown below branches for nodes where topology of ML analysis was coincident with Bayesian inference. Light blue, green and yellow backgrounds indicate Spiroplasma, Cyanobacteria and mosquito RIPs, respectively. Information related to the sequences used to infer these trees is available in Supplementary Table 2.

Figure 5

Analyses of the synonymous vs. nonsynonymous substitution rates. Codons forming the active site are indicated by colored nucleotides (A and T: orange, G and C: blue). Codons under significant purifying (negative) selection determined by the three tests (SLAC, FEL and REL), or by two out of the three tests, are underlined in red or green color, respectively.