Figure 1.
Flowchart of genome-mining for the discovery of ant peptides used in this study.
Whole genome shotgun data, in this instance from the three ant species Atta cephalotes, Camponotus floridanus and Harpegnathos saltator, and amino acid sequences of precursor proteins from bioactive peptides of interest (e.g. defense and/or regulatory neuropeptides) were used for database analysis. This included similarity analysis of target DNA sequence and query protein sequence using tBLASTn, DNA to protein translation of discovered hit sequences and identification of open reading frames and coding sequence. The obtained automated results were refined and confirmed manually and used for gene structure prediction using the GeneWise algorithm. Database analysis yielded precursor protein and peptide sequences that were further annotated and analyzed by sequence alignments and similarity comparison to identify signal sequences, propeptides and mature peptide chains. Using this genome-mining methodology it was possible to predict the amino acid sequences of bioactive peptides in ants.
Table 1.
Selected defense- and neuropeptides from ant species characterized by genome-mining.
Figure 2.
Sequence and gene structure of novel ant defensins.
(A) Identified ant defensin precursor sequences from Atta cephalotes, Camponotus floridanus and Harpegnathos saltator were used for similarity alignment (ClustalW2) and compared to known defensins from Formica aquilonia (UniProtKB Q5BU36), Lasius niger (B9TXS0), Myrmecia ruginodis (B9TXS6), Drosophila melanogaster (P36192), Apis mellifera (P17722, Q5MQL3) and Acromyrmex echinatior (F4WLL3). The signal peptide cleavage site (identified by similarity) is shown as an arrow. Mature defensin peptides are indicated in the black box and the conserved cysteine residues are indicated with asterisks. Residue positions of positive evolutionary selection [23] are indicated with a green box. The sequence alignment was prepared using Boxshade. (B) Gene structure of novel ant and known insect defensin genes (GenBank D.melanogaster NT033778.3 and A.mellifera NC007085.3, NC007075.3) was predicted using the GeneWise algorithm. Signal sequences are indicated in light grey, pre-regions in dark grey and the mature peptide domain in white. Intron sequences (including their base pair length) are indicated with upside-down arrow heads. (C) The sequences of 22 known [23] and the novel ant defensins have been compared using a sequence logo to highlight their amino acid variation. Conserved cysteines are colored in yellow, the positions of positive evolutionary selection are colored in green and the conserved negatively-charged Asp and positively-charged Arg are colored in red and blue, respectively.
Figure 3.
Structural model of Atta cephalotes defensin.
The lowest energy model of the ant defensin was prepared by homology to the insect defensin phormicin (pdb code 1ICA) and the synthetic defensin Def-BBB (2E3E). (A) The structure cartoon shows the three conserved disulfide bonds (in yellow) and secondary structure elements (loops, α-helix, and two anti-parallel β-sheets) that form together the so-called ‘cysteine-stabilized αβ (CSαβ) motif’. The cysteines are labeled with roman numerals CI–VI. (B) The putative mature peptide sequence contains two negatively-charged (shown in red) and eight positively-charged (shown in blue) residues. The distances between the side chains of the conserved Asp (D4) and surrounding Lys- (K11 and K30) and Arg-residues (R33 and R42) were measured using PyMol and are indicated by dotted lines (in Å units). The measured distances and orientations of the side-chains suggest possible electrostatic interactions (salt-bridge formation) between these residues. (C, D) Surface representation of the ant defensin pointing out its amphipathic character, i.e. charged surface (cations are shown in blue, anions in red) in combination with a hydrophobic patch (shown in white) that contains two tryptophan-residues (W8 and W12). The amino acid sequence including residue numbers of the mature A.cephalotes defensin is shown below the structural models.
Figure 4.
Sequences, gene- and peptide structures of ant inotocins.
(A) Sequences of prepro-inotocin/neurophysin proteins from Atta cephalotes, Camponotus floridanus and Harpegnathos saltator were compared by similarity alignment to known inotocin/oxytocin/vasopressin prepro-proteins from Tribolium castaneum (UniProtKB A3RE83), Nasonia vitripennis (GenBank XP001606547.1), Homo sapiens (P01185 and P01178) and Lymnaea stagnalis (Q00945). The signal peptide cleavage site (identified by similarity) is shown with an arrow. Mature vasopressin/oxytocin/inotocin peptides are indicated in the box, followed by the canonical GRK amidation signal (dotted line above the sequences). The conserved cysteine residues in the neurophysin domain are indicated with asterisks. The sequence alignment was prepared using Boxshade. (B) Gene structure of novel H.saltator inotocin and known vasopressin-family prepro-protein genes (GenBank H.sapiens NC000020.10 and T.castaneum NC007423.2) was predicted using the GeneWise algorithm. Signal sequences are indicated in light grey, the mature peptide hormone chains (INT, inotocin; AVP, vasopressin; OT, oxytocin) in dark grey, pre-regions in black and the neurophysin domains in white. For the AVP prepro-protein the copeptin region is also marked. Intron sequences (including their base pair length) are indicated with upside-down arrow heads. (C) Evolution of the vasotocin nonapeptide family (simplified illustration for clarity, see also [42], [46]) is indicated by solid arrows. Arginine-vasotocin is the presumed ancestral peptide of oxytocin and vasopressin. Mammalian oxytocin evolved via intermediate forms of isotocin (bony fish) and mesotocin (lung fish, amphibians, reptiles and birds). It is yet to be determined whether invertebrate oxytocin/vasopressin-related peptides in insects or snails (e.g. conopressins, not shown) have also evolved from ancestral vasotocin (indicated as dashed line) [42]. The peptide sequences are shown in one-letter amino acid code. The highly conserved cysteine-residues and disulfide bonds are colored in yellow. Residues in the ancestral arginine-vasotocin and those that are identical to vasotocin are colored in dark grey. Residues that have changed during vasopressin evolution are colored in red, residues that have changed during oxytocin evolution are colored in purple and residues that are unique to insect/ant inotocins are colored in green.
Figure 5.
Phylogenetic relationship of selected ant and insect species.
The phylogenetic relationship of the ant species (Formicidae) used for this study (Atta cephalotes, Camponotus floridanus and Harpegnathos saltator; shown as grey box) is indicated as a phylogram tree (adapted and modified from [16], [66]) in comparison to two Hymenoptera species Apis mellifera (honeybee) and Nasonia vitripennis (parasitoid wasp), the Coleoptera species Tribolium castaneum (red flour beetle), the Nematocera species Anopheles gambiae (mosquitoes), the Brachycera species Drosophila melanogaster (fruit fly) and human as representative vertebra species (Homo sapiens). The absence of the oxytocin/vasopressin peptide hormone system in a specific lineage is indicated with an X.