Macroevolution of venom apparatus innovations in auger snails (Gastropoda; Conoidea; Terebridae)

Abstract

The Terebridae are a diverse family of tropical and subtropical marine gastropods that use a complex and modular venom apparatus to produce toxins that capture polychaete and enteropneust preys. The complexity of the terebrid venom apparatus suggests that venom apparatus development in the Terebridae could be linked to the diversification of the group and can be analyzed within a molecular phylogenetic scaffold to better understand terebrid evolution. Presented here is a molecular phylogeny of 89 terebrid species belonging to 12 of the 15 currently accepted genera, based on Bayesian inference and Maximum Likelihood analyses of amplicons of 3 mitochondrial (COI, 16S and 12S) and one nuclear (28S) genes. The evolution of the anatomy of the terebrid venom apparatus was assessed by mapping traits of six related characters: proboscis, venom gland, odontophore, accessory proboscis structure, radula, and salivary glands. A novel result concerning terebrid phylogeny was the discovery of a previously unrecognized lineage, which includes species of Euterebra and Duplicaria. The non-monophyly of most terebrid genera analyzed indicates that the current genus-level classification of the group is plagued with homoplasy and requires further taxonomic investigations. Foregut anatomy in the family Terebridae reveals an inordinate diversity of features that covers the range of variability within the entire superfamily Conoidea, and that hypodermic radulae have likely evolved independently on at least three occasions. These findings illustrate that terebrid venom apparatus evolution is not perfunctory, and involves independent and numerous changes of central features in the foregut anatomy. The multiple emergence of hypodermic marginal radular teeth in terebrids are presumably associated with variable functionalities, suggesting that terebrids have adapted to dietary changes that may have resulted from predator-prey relationships. The anatomical and phylogenetic results presented serve as a starting point to advance investigations about the role of predator-prey interactions in the diversification of the Terebridae and the impact on their peptide toxins, which are promising bioactive compounds for biomedical research and therapeutic drug development.

Figure 1

Map showing localities sampled for Terebridae study. AU, Queensland, Australia; CH, Coral Sea; FI, Fiji; SMA, South Madagascar; MO, Mozambique; NMA, North Madagascar; NNC, North New Caledonia; PA, Pacific Panama; PH, Philippines; SNC, South New Caledonia; SO, Solomon Islands; TA, Tahiti; VA, Vanuatu.

Figure 2

Flat (A-E) and solid recurved (F-G) teeth of Terebridae. A – Pellifronia jungi (IM_2007_30591), ventral view of radular membrane, only half shown; B – Clathroterebra poppei (IM_2007_30546), ventral view of radular membrane; C – Terebra succincta (IM_2007_30582), separate marginal tooth; D – Terebra trismacaria (IM_2007_30579), ventral vies of radular membrane; E – Myurella lineaperlata (IM_2007_30635), group of teeth attached to the subradular membrane; F – Euterebra fuscolutea (IM_2009_10133), ventral view of radular membrane, only half shown; G – Duplicaria sp. 2 (IM_2009_10164), ventral view of radular membrane, only half shown. Scale bars – 10 μm.

Figure 3

Hypodermic (A-O) and semienrolled (Q) teeth in Terebridae. Clade C (A-G): A, B – Terebra cingulifera (IM_2007_30382); C – Triplostephanus fenestratus (IM_2007_30418); D-E – Triplostephanus triseriatus (IM_2007_30404); F-G – Terebra guttata (IM_2007_30376);. Clade E5 (H-I) – Myurella kilburni (IM_2007_30461); Clade D (J-P): J- K – Hastula hectica, Philippines, Panglao Island; L – Hastula lanceata (IM_2007_30535); M-N – Hastula penicillata (IM_2007_30540), N – central part of the radular membrane; O-P – Hastula strigilata (IM_2007_30607); Q – Hastula stylata (IM_2009_10106). Scale bars: 50 μm (except E, G, P – 10 μm).

Figure 4

Likelihood phylogenetic tree obtained with 410 specimen sequences for the COI, 12S and 16S genes. Boostraps and Posterior Probabilities are indicated for each node (when > B = 70% and > PP = 0.90 respectively). The ten collapsed clades of Terebridae (A, B, C, D, E1, E2, E3, E4, E5 and F) are detailed on Figures 2-5.

Figure 5

Likelihood phylogenetic tree for clades A, B, C, D, F. Boostraps and Posterior Probabilities are indicated for each node (when > 70 and > 0.90 respectively). For clarity purposes, intraspecific support values are not shown.

Figure 6

Likelihood phylogenetic tree for the clades E1-E5. Boostraps and Posterior Probabilities are indicated for each node (when B > 70% and PP > 0.90 respectively). For clarity purposes, intraspecific support values are not shown.

Figure 7

Illustration of some specimens in each clade. From left to right: Clade A: Pellifronia jungi IM_2007_30539; Clade B: Oxymeris maculata IM_2007_30370, Oxymeris crenulata IM_2007_30377, Oxymeris dimidiata IM_2007_30379; Clade C: Terebra argus IM_2007_30383, Terebra guttata IM_2007_30387, Terebra funiculata IM_2007_30394, Triplostephanus fujitai IM_2007_30482, Terebra cingulifera IM_2007_30485, Terebra tricolor IM_2007_30493; Clade D: Hastula strigilata IM_2007_30416, Hastula hectica IM_2007_30426, Hastula albula IM_2007_30437; Clade E1: Terenolla pygmaea IM_2009_10121, Hastulopsis pertusa IM_2007_30388, Clathroterebra fortunei IM_2007_30391, Myurella affinis IM_2007_30415; Clade E2: Terebra fijiensis IM_2007_30423, Terebra succincta IM_2007_30433, Terebra textilis IM_2007_30451, Myurella lineaperlata IM_2007_30471, Duplicaria sp. 3 IM_2009_10151; Clade E3: Terebra succincta IM_2007_16731, Myurella orientalis IM_2007_30515; Clade E4: Terebra elata IM_2007_42111, Terebra larvaeformis IM_2007_42113, Terebra puncturosa IM_2007_42116, Terebra berryi IM_2007_42144; Clade E5: Myurella undulata IM_2007_30384, Myurella paucistriata IM_2007_30453, Terebra sp. 5 IM_2007_30946; Clade F: Euterebra fuscolutea IM_2009_10112, Duplicaria albofuscata IM_2009_10162

Figure 8

Likelihood phylogenetic tree obtained with 63 specimens sequences for the COI, 12S, 16S and 28S genes.

Figure 9

Character mapping of the six characters presented in the Table 2. Bootstraps are shown for each node.