Figure 1.
Dinoponera quadriceps in the field and its dissected venom apparatus.
Part A - A single specimen of D. quadriceps protecting the nest's entrance. Part B - Dissected D. quadriceps venom apparatus (×40). Abbreviations: Dg - Doffur's gland; cv - convoluted gland (not observable); st - secretory tubule; vs- venom sac; sg – sting.
Figure 2.
Overall classification of ESTs from the D. quadriceps venom gland cDNA (ESTs) library based on function.
EST sequences were annotated by comparing contigs and singlets (E-values≤1.0E-5) with the non-redundant Genbank translated protein database using BLASTX. Search stringency was eventually adjusted (E-values lower than 1.0E-5) and database restricted to hymenopterans with the aim of finding relevant familial- and structural-related sequences of hymenopterans toxins. Part A – relative number of polypeptides which were counted as individual random sequenced clones. The group of <cellular process> refers to enzymes and polypeptides involved in the venom gland metabolism. Part B – relative numbers of venom-related sequences found in the overrepresented small toxin dataset. Under the term <toxin> is included dinoponeratoxins, venom allergens, and venom cysteine rich peptides (part B).
Figure 3.
General overview of the transcripts in the D. quadriceps venom gland identified by deep RNA sequencing.
Annotation of contigs from the RNA-Seq assembly of the giant ant venom following the conventions of figure 2; i.e., the E-value cut-off was at least 1.0E-5 for BLASTX functional comparison.
Figure 4.
Distribution of the Hymenoptera species as determined by best protein hits.
The percentages of homologs from distinct hymenopterans species with which the 6,429 contigs of the D. quadriceps venom gland shared the high sequence similarities.
Table 1.
Summary of assembled contigs derived from RNA sequencing and comparison with the ESTs of the D. quadriceps venom gland transcriptome.
Figure 5.
Proportions of transcript toxin and toxin-related components expressed in the venom gland of D. quadriceps.
Toxins and venom-related protein precursors represented less than 1% of all transcripts (18,524) in the giant ant venome. As shown, the major toxin core was composed of four groups of polypeptides.
Figure 6.
Multiple alignments of the deduced amino acid sequences of venom allergen I (Sol i 1) from D. quadriceps with known sequences from different ant and wasp species.
The identical and conserved amino acid residues of diverse ant and wasp hymenopterans are highlighted in black and gray, respectively. Dots represent gaps.
Figure 7.
Multiple alignments of the deduced amino acid sequences of the venom allergen antigen 5 (Sol i 3) of D. quadriceps with known sequences from different Hymenoptera species.
The conserved regions of the Sol i 3/allergen antigen 5 (Ag 5) polypeptide sequences of D. quadriceps and other hymenopterans are presented in black boxes. Conserved amino acid residue mutations are highlight in light gray boxes. Dots represent gaps. The unique amino acid sequences of the D. quadriceps allergen homologous to Sol i 3 protein are displayed in open boxes.
Figure 8.
Comparison of Sol i 3/Allergen Antigen 5 polypeptides from D. quadriceps with their hymenopteran orthologs.
Phylogenetic tree based on neighbor-joining analyses of a concatenated alignment of allergen 5/Sol i 3 and the orthology relationships of multiple insects. The scale bar indicates 0.05 substitutions per site. Vespula maculifrons, Vespula vulgaris, Vespula germanica, Vespula squamosal, Vespa crabro, Dolichovespula maculate, Polybia paulista, Polistes dominulus, Polistes annularis, and Solenopsis invicta.
Figure 9.
Multiple alignments of deduced amino acid sequences of different identified dinoponeratoxins from the D. quadriceps venom gland transcriptome and D. australis.
Deduced D. quadriceps dinoponeratoxin cDNA precursor sequences (contig_1 and contig_9). RNA deep sequencing contigs (contig_2 and contig_145) were compared to mature peptide sequences from D. australis (Da-3177 and Da-3105) (part A) and from another species of Dinoponera (Da-2501) (part B). ClustalW was used to multi-align the sequences. Identical amino acid residues are marked with asterisks. Stretches of deduced amino acid sequences supported by EST sequences are boxed. Signal peptides (pre-peptides) are doubled underlined, and pro-regions of pro-peptides are shown with a single line under the sequence. Contigs 1 and 9 were first identified in the EST library, and contigs 2 and 145 came primarily from RNA-Seq.
Figure 10.
The Inhibitor cystine knot- (ICK-) toxin from D. quadriceps venom.
A) Dinoponera ICK-like tridimensional homology model from contig 05 (Table S1) mature peptide (yellow – β-sheet; green – cysteine; ted – 3/10-helix. N-terminus (N); C-terminus (C). The numbers indicate the cysteine positions in the Dinoponera sequence. The presence of a ‘disulfide through disulfide knot’ structurally defines this peptide as a knottin. B) Schematic representation of a knottin obtained when one disulfide bridge crosses the macrocycle formed by two other disulfides, and the interconnecting backbone (disulfide III–VI) goes through disulfides I–IV and II–V. C) Alignments of the Dinoponera ICK-like mature peptide with the tarantula and Conus snail sequences of the templates used in the homology model showing the highly conserved cysteine framework [C-C-CC-C-C] that is characteristic of the omega-toxin-like family. TXFK1_PSACA: psalmopeotoxin I (PcFK1) from the venom of the tarantula Psalmopoeus cambridgei (NMR structure 1X5V; UNIPROT Accession P0C201). CO16B_CONMR: Mu-Conotoxin MrVIB from conus snail Conus marmoreus (NMR structure 1RMK; UNIPROT Accession Q26443).
Table 2.
Overview of major venom polypeptide components of D. quadriceps and other ant species.