The cranial gland system of Nasonia spp.: a link between chemical ecology, evo-devo, and descriptive taxonomy (Hymenoptera: Chalcidoidea)

Abstract

Nasonia is an emerging model system for investigating the evolution of complex species-specific behavioral and morphological phenotypes. For example, the male head shape differs considerably between Nasonia Ashmead (Hymenoptera: Chalcidoidea) species. In addition, differences in courtship behaviors, and possibly influences of a male-specific aphrodisiac pheromone, contribute to interspecific prezygotic isolation. However, the possible relationships between courtship, pheromone signaling, and male head shape are unknown. Using multimodal imaging techniques, we conducted a comprehensive examination of the skeletomuscular and exocrine gland systems of the lower head region of all 4 Nasonia species and their sister genus Trichomalopsis Crawford (Hymenoptera: Chalcidoidea). This analysis reveals the presence of 3 undescribed exocrine glands in the lower head region and a unique mandibular modification, the basal mandibular carina, that might be involved in pheromone spread. We performed morphometric and volumetric analyses using 3D datasets from synchrotron X-ray microtomography and found that the size of the genomandibular gland and the corresponding basal mandibular carina correlates with both interspecific courtship length and head shape differences, indicating that this gland is a likely source of the oral aphrodisiac pheromone. These differences correlate with the prevalence of within-host mating rather than phylogenetic relatedness in Nasonia species, with increased within-host mating occurring in species with larger genomandibular glands. Our findings create an opportunity to better understand the complex gene regulatory networks underlying superficially unrelated traits and serve as a link between behavior, chemical ecology, evo-devo, and descriptive taxonomy.

Keywords: SBF-SEM; craniofacial; mandibular rods; parasitoid; speciation.

Fig. 1.

Phenotypic traits related to the aphrodisiac pheromone in Nasonia species. A, Head nodding during courtship of Nasonia. Males touch the female antennae preceding copulation, increasing the sexual desire of the female. B, CLSM micrograph of a bisected head of N. giraulti showing the enlarged mandibular gland (mdg). C–E, Frontal view of male heads of Nasonia vitripennis (C), N. longicornis (D), and N. giraulti (E) showing interspecific differences in gena convexity (mdg = mandibular gland, md = mandible, M1a = mandibular adductor anterior, M1b = mandibular adductor posterior, br = brain, max = maxilla).

Fig. 2.

Volume rendered micrographs (Drishti) showing cuticular anatomical structures of the cranium and mandible and the basimandibular gland of Nasonia giraulti. A–D, cranium, male, E–H, cranium, female. A, E—posterior view, B, F—ventral (oral) view, C, G—anterior view, D, H—lateral view. I–L, mandible, male, M-P, mandible, female. I, M—ventral view, 0 transparency, J, N—ventral view 50% transparency showing basimandibular gland (img), K, O—posterior (internal) view, 0 transparency, L, P—posterior (internal) view, 50% transparency showing basimandibular gland (bmg, aam = anterior angle of the mandible, bmc = basal mandibular carina, lab = labium, max = maxilla, pam = posterior angle of the mandible, plm = maxillary palp).

Fig. 3.

Species level variability of the cranium and mandible in male Pteromalidae. A, B, Nasonia vitripennis. C, D, N. longicornis. E, F, N. oneida. G, H, N. giraulti. I, J, Trichomalopsis sarcophagae. K, BMCH (basimandibular carina heigth) by species, Kruskal-Wallis test: The differences between the rank sums of 425 (N. vitripennis), 198 (N. longicornis), 45 (N. giraulti), 135 (N. oneida), and 325 (T. sarcophagae) were significant, H(4) = 42.4283, P < 0.0001. Post hoc pairwise comparisons using Dunn’s test indicated that BMCH in N. vitripennis were observed to be significantly different from those of N. longicornis (P = 0.0023), N. giraulti (P < 0.001), and N. oneida (P = 0.0009). This metric in T. sarcophagae was observed to be significantly different from that of N. giraulti (P = 0.0001). No other differences were statistically significant. L, GH (gena height) by species, Kruskal-Wallis test: The differences between the rank sums of 425 (N. vitripennis), 321 (N. longicornis), 46 (N. giraulti), 142 (N. oneida), and 194 (T. sarcophagae) were significant, H(4) = 41.5628, P < 0.0001. Post hoc pairwise comparisons using Dunn’s test indicated that normalized gena heights in N. vitripennis were observed to be significantly different from those of N. giraulti (p < 0.001), N. oneida (P = 0.0015), and T. sarcophagae (P = 0.0018). This metric in N. giraulti was also observed to be significantly different from that of N. longicornis (P = 0.0002). No other differences were statistically significant. M, HW/BMCH vs HW/GH, Spearman’s correlation: There was a significant positive monotonic relationship between the two metrics, rs(45) = 0.7454, P = < 0.0001.

Fig. 4.

Volume rendered micrographs showing the mandibles of Nasonia specimens. A–H, The anterior and posterior angles of the mandibles (aam, pam) do not articulate with the cranium at rigid articulations but can slide on the pleurostoma (ple) and move in multiple axes (A–C, E–G, N. vitripennis, D, H, N. giraulti). I–P, The size of the basal mandibular carina (bmc) relative to the length of the mandible is sexually dimorphic and exhibits species-level variation: it is largest in Nasonia giraulti, medium size in N. longicornis and N. oneida, and smallest in N. vitripennis. In females of all species, it is reduced to a low ridge. The anterior and posterior angles of the mandible are neither sexually dimorphic nor species-specific.

Fig. 5.

Volume rendered micrographs (Drishti) showing the head and lower cranial regions of Nasonia giraulti. A, C, D, G, H, male, B, E, F, I, J, female. A, B, oral view showing the mouthparts, C–J, Sagittal sections of the head showing the epithelial cell layer of the genomandibular gland, the subcuticular space and the adjacent surfaces of the basal mandibular carina and epistoma (aam = anterior angle of the mandible, bmc = basal mandibular carina, ec = epithelium of the genomandibular gland, ple = pleurostoma, gmg=genomandibular gland, lab = labium, M1a, M1b = mandibular adductor muscles, M2a, M2b = mandibular abductor muscles).

Fig. 6.

Volume rendered micrographs (Drishti) showing the mandibular muscles and the genomandibular gland of Nasonia giraulti. A–D, external view, E–L, internal view. A, E, I, posterior, B, F, J ventral (oral) view, C, G, K anterior view, D, H, L lateral view. M–P, Sagittal sections of the head of N. giraulti showing the epithelial cell layer of the genomandibular gland, the subcuticular space and the adjacent surfaces of the basal mandibular carina and epistoma. (aam = anterior angle of the mandible, bmc = basal mandibular carina, ec = epithelium of the genomandibular gland, epi = epistoma, gmg = genomandibular gland, lab = labium, M1a, M1b = mandibular adductor muscles, M2a, M2b = mandibular abductor muscles, max = maxilla, pam = posterior angle of the mandible, pll = labial palp, plm = maxillary palp, scs = subcuticular space of the genomandibular gland).

Fig. 7.

micro-CT micrographs showing the skeletomuscular and exocrine gland systems of the lower head regions of Pteromalidae. A, D, F, G, I, Nasonia giraulti male, B, E, H, N.giraulti female, C, Trichomalopsis sarcophagae male. Animated GIF version of D is available from 10.6084/m9.figshare.28300592, and animated GIF version of F is available from 10.6084/m9.figshare.28300607. The mandibular abductor muscles (M2a, b) inserts at the base of the basal mandibular carina (bmc). The mandibular rods (mar) are non-glandular and solid invaginations of the mandibular sclerite. The epithelial cell layer of the class I basimandibular gland (bmg) corresponds to a thinner cuticular region of the mandible (tcr). In some cases, it is complicated to separate the cuboidal fat body cells (fb) from the adjacent columnar epithelial cells of the genomandibular gland (ec) in males as the border between them is not always clear (A, D, G). The epithelium of the genomandibular gland is much thinner in females (ec: E, H). The mandible is connected to the pleurostoma (ple) by the basal mandibular conjunctiva (bmm). The subcuticular space of the genomandibular gland develops between the epithelial gland cells (ec) and the cuticle of the pleurostoma (ple), basal mandibular conjunctiva (bmm) and the basal mandibular carina (bmc). Emptying of the gland might be governed by the contraction of the mandibular abductor muscles (M2a, M2b). The epithelial cells of the class III genal gland (ec: F, I) are adjacent to the genal cuticle just ventral to the compound eye and open, via separated cuticular canals (cc), on the pleurostoma (ple).

Fig. 8.

SBFSEM micrographs of the exocrine gland systems of the lower head region of Nasonia giraulti male. Animated GIF version of A is available from 10.6084/m9.figshare.28300616, B, C from 10.6084/m9.figshare.28300670 and D–G from 10.6084/m9.figshare.28300748. The epithelial cells of the class I genomandibular gland (gmg) are characterized by the electron-dense cytoplasm and the microvilli rich distal cell membrane (mic) and the lack of end apparatuses and cuticular canals (A). Cells of the genal gland (gg) are less electron dense and possess microvilli-embedded end apparati (ea) that continues into cuticular canals (cc) exiting at the pleurostoma (B, C). The mandibular rods (mar) are solid cuticular invaginations that are not associated with any glandular tissue but are connected to cellular structures resembling scolopale cells (sco?). (bmc = basal mandibular carina, bmg = basimandibular gland, bmm = basal mandibular membrane, cc = cuticular ducts of genal gland, ea = end apparatus, ec = epithelial cell of the genomandibular gland, epi = epistoma, fb = fat body, gmg = genomandibular gland, gg = genal gland, man = mandible, mar = mandibular rod, mic = microvilli, scs = subcuticular space, pam = posterior mandibular margin, neu = neuron, nuc = nucleus).

Fig. 9.

Variation of the genomandibular gland in Nasonia species. A–C. Volume rendered micrographs (Drishti) showing the variation in mandibular gland size in Nasonia species, N. vitripennis (A), N. longicornis (B), N. giraulti (C). The gland size in N. oneida (not illustrated) is not different from that of N. longicornis. D. Kruskal-Wallis: The differences between the rank sums of 58 (N. vitripennis), 294 (N. longicornis), 387 (N. giraulti), 237 (N. oneida), and 152 (T. sarcophagae) were significant, H(4) = 41.9177, P = < 0.0001. Post hoc pairwise comparisons using Dunn’s test indicated that normalized subcuticular gland volume in N. vitripennis was observed to be significantly different from those of N. longicornis (P = 0.0013), N. giraulti (P < 0.001), and N. oneida (P = 0.0027). This metric in N. giraulti was also observed to be significantly different from that of T. sarcophagae (P = 0.0001). No other differences were statistically significant.