Eastern oysters alter inducible defense mechanism of shell strengthening with age

Abstract

Eastern oysters, Crassostrea virginica, use inducible defenses in the form of strengthened shells to reduce their risk of predation. Inducible defenses often have trade-offs between the costs and benefits associated with the organism's fitness, as developing defenses requires energetic resources. Shell strength is a product of the amount of material laid by the animal (thickness) and the material properties of the shell (e.g. hardness and fracture resistance). Previous studies have suggested that oysters may trade off between shell thickness or shell hardness as a mechanism for increasing shell strength against predation, which are hypothesized to have different energetic requirements. The present study analyzed the shell structural (thickness of composite layers) and micromechanical properties (microhardness and crack propagation tested within individual shell layers) of predator-induced and non-induced juvenile oysters at 4 and 8 weeks post-settlement to determine which shell strengthening mechanism oysters use in response to predator cues. Younger juveniles did not display any significant differences in micromechanical shell properties or shell thickness as a result of induction, though some marginal differences were detected. In contrast, older juveniles use a combination of increased hardness and thickness in shell defense, where induced oysters were 33% thicker overall and 12% harder within their outer prismatic layer. This suggests that oysters alter shell strength using multiple defense mechanisms depending on age, and we postulate that animals switch mechanisms when necessary to balance the advantages and associated costs of defense with an individual's physiological needs for growth.

Keywords: Crassostrea virginica; Blue crabs; Phenotypic plasticity; Predator–prey interactions; Shell formation.

Fig. 1.

Shell size (mm) of predator induced and non-induced oysters of 4 and 8 weeks of age measured from the present study (n=10 per age and induction state). The boxplot displays median and the first and third quartiles of the data and black dots display data points that fall outside of this range. Age strongly affected size (Type II Wald chi-square test on GLMM with age and induction state as fixed factors), where 8-week-old oysters were significantly larger than 4-week-old oysters. Within-group post hoc comparisons were made using estimated marginal means and brackets indicate significant differences between groups (***P<0.001).

Fig. 2.

Thickness of shell layers in induced and non-induced oyster spat-on-shell of two ages. (A) Overall, (B) foliated and (C) prismatic. The boxplot displays median and the first and third quartiles of the data, and black dots display data points that fall outside of this range. Data were collected from 19–31 measurements (depending on size) on 10 replicate shells per age and induction state. Older, 8-week-old oysters grew significantly thicker shells when induced with predator cues compared with non-induced oysters of the same age. Four-week-old oysters did not exhibit any significant differences between induced and non-induced oyster thickness. Statistical analysis (Type II Wald chi-square test on GLMM with age and induction state as fixed factors) revealed shell thickness, overall and each individual layer, was affected by age, induction and their interaction. Within-group post hoc comparisons were made using estimated marginal means and asterisks indicate significant differences between groups (**P<0.01, ***P<0.001).

Fig. 3.

Vickers microhardness value of shell layers of 4- and 8-week-old oyster spat-on-shell. (A) Foliated and (B) prismatic. The boxplot displays median and the first and third quartiles of the data and black dots display data points that fall outside of this range. Data were collected from 7–10 measurements (Table S2) on 10 shells per age and induction state. Within the foliated layer, oyster age, but not induction status, affected shell microhardness (A, Type II Wald chi-square test on GLMM with age and induction state as fixed factors), where 4-week-old oysters were harder than 8-week-old oysters. The prismatic layer was affected by both age and induction, where older, induced oysters were significantly harder compared with non-induced oysters (B, Type II Wald chi-square test on GLMM with age and induction state as fixed factors). Brackets represent significant difference based on overall effect of age (A) or within-group post hoc comparisons made using estimated marginal means (B). Asterisks indicate significant differences between groups (*P<0.05, ***P<0.001).

Fig. 4.

The mean±s.e.m. number of cracks appearing after Vickers microhardness indent test on shell layers. (A) In the foliated layer, there were no differences in the number of cracks. (B) In the prismatic layer, induction state significantly affected the number of cracks (Type II Wald chi-square test on GLMM with age and induction status as fixed effects), where older induced oysters experienced significantly fewer cracks overall. Brackets represent significant difference based on overall effect of induction (B) and asterisks indicate significant differences between groups (**P<0.01). Data were collected from 7–10 measurements (Table S2) on 10 shells per age and induction state. Individual data points are displayed, where the size of the point indicates the number of individual data points represented.

Fig. 5.

The length of cracks appearing after Vickers microhardness indent test on shell layers.  (A) Folidated and (B) prismatic. The boxplot displays median and the first and third quartiles of the data and black dots display data points that fall outside of this range. Data were collected from 7–10 measurements (Table S2) across 10 shells per age and induction state. The length of cracks in the foliated layer (A) was unaffected by age or induction, but crack size in the prismatic layer (B) was significantly smaller for older oysters (Type II Wald chi-square test on GLMM with age and induction status as fixed effects). Asterisks indicate significant differences between groups (**P<0.01).