Impacts of the ectoparasitic snail Boonea impressa on growth of postset juvenile oysters

Gastropods of the family Pyramidellidae consist of obligate parasites, many of which show some degree of host specificity (Robertson & Mau-Lastovicka, 1979) and can have dramatic impacts on bivalve fisheries and aquaculture (Wilson, Powell & Ray, 1988a; Cumming & Alford, 1994). A number of pyramidellids affect commercial mollusc species, including Turbonilla sp. parasitic on giant clams (Boglio & Lucas, 1997) and Evalea tenuisculpta on abalone (Maguire & Rogers-Bennett, 2013). One species, Boonea impressa, has received particular attention due to its adverse effects on the eastern oyster, Crassostrea virginica (White, Powell & Kitting, 1984; Ward & Langdon, 1986; Wilson, Powell & Ray, 1991). Boonea impressa is common on oyster reefs from Connecticut to the Gulf of Mexico (Hopkins, 1956; Ward & Langdon, 1986; White et al., 1988b). On natural reefs, densities can vary greatly and have been found as high as 100 individuals per oyster (Hopkins, 1956), although more typically they are found at densities of 5–40 individuals per oyster (White et al., 1984; Wilson et al., 1988a). At least a portion of this variation is likely due to their aggregative behaviour, as mean snail densities per oyster in oyster clumps may be relatively low, while patches of B. impressa within those clumps can exhibit very high densities (Wilson et al., 1991; Wilson, White & Powell, 1988b). Known as the ‘oyster mosquito,’ B. impressa positions itself on the edge of the host shell, inserting its proboscis, penetrating mantle tissue with its stylet and removing oyster haemolymph using a buccal pump (Wise, 1993). This perturbation to the oyster causes reductions in growth and condition, with relatively high densities of B. impressa having been demonstrated to reduce oyster growth by between 73% (Ward & Langdon, 1986) and 100% (White, Powell & Ray, 1988a). Reduction in oyster growth occurs in three distinct ways: (1) snail parasites consume already assimilated nutrients and energy, (2) parasitized oysters filter and assimilate less materials and (3) parasitized oysters experience rapid valve movements, caused by rapid contractions of the adductor muscle, and thus use up energy (Ward & Langdon, 1986). While the impacts of B. impressa on larger oysters at high densities are well documented, the impacts on small oyster spat have not been examined. Given the dramatic decreases in physiological rates for larger oysters in the presence of these snails, it is possible that these ectoparasites could be especially damaging to recruiting cohorts of juvenile oysters. In particular, realistic field densities of B. impressa have the potential to restrict significantly the growth of oyster spat. The objectives of this study were to examine the impact of these parasitic snails, when maintained at field densities, on the growth of oyster spat in a laboratory setting. Oyster spat were received from the University of North Carolina Wilmington Shellfish Research Hatchery. Initial sizes varied between 7 and 10 mm left-valve length. Juvenile oysters were haphazardly glued with super glue onto cleaned and dried oyster shells with a mean (±SE) shell area (both sides of the shell) of 89.2 (±1.1) cm2. The resulting spat density was 0.11 spat cm−2 of available shell area. Numbered oyster shells were then placed into individual polyvinyl chloride (PVC) arenas (Ward & Langdon, 1986), created using 11.4 cm diameter PVC couplings, drilled with four 11.3 cm2 windows and covered with a 1-mm mesh window screen on the bottom and sides of the arenas. The tops were left open. Arenas were placed onto inverted aquaculture trays in a 1240-l water tank (330 × 65 × 60 cm), filled with c. 600 l and receiving constantly flowing filtered seawater at the UNCW Shellfish Research Hatchery. The inverted trays allowed the experimental arenas to be kept off the bottom. The tank received filtered seawater at a rate of c. 3 l min−1 and a constant algal drip of 200 ml min−1 with a mixture of Isochrysis galbana affinis Tahiti (T-iso), Chaetoceros muelleri and Tetraselmus sp. Oysters in arenas were allowed to acclimate for 2 weeks before experiments. Boonea impressa was collected from the field in November 2012 at Hewlett's Creek, Wilmington, NC (34°10′35.29″N, 077°50′33.19″W), near the intercoastal waterway and Masonboro Sound. Snails were collected by shaking clumps of oysters in a bucket filled with seawater, which caused the snails to become dislodged and sink to the bottom (White et al., 1984). The water was changed and individuals of B. impressa were maintained in the bucket with flowing seawater and an additional airstone, and fed dissected oysters. After the 2-week acclimation period, all oyster shells with sets of 10 oyster spat were digitally photographed and initial spat shell area was measured using ImageJ (National Institutes of Health) image analysis software (Newell, Kennedy & Shaw, 2007; Kelly et al., 2011). On 6 December 2012, oysters were assigned to 1 of 3 parasite treatments: no snail controls, low (5 B. impressa per arena) and moderate snail abundance (15 B. impressa per arena; White et al., 1988a). The resulting snail densities were 0.06 cm−2 of oyster shell area (=0.5 per spat) for the low snail treatments, and 0.17 cm−2 (=1.5 per spat) in the moderate snail treatments. These densities were derived from literature values (White et al., 1988a), reports for field densities in the literature (Hopkins, 1956; White et al., 1984) and local field densities of up to 23 snails in 0.04-m2 quadrats (Hanke, 2014). Snails exhibit clumped distribution in the field (Wilson et al., 1991), thus the densities used in this experiment are likely to be encountered in the field. The experiment was allowed to run for 3 weeks. Oysters were checked and cleaned every 2 d for the duration of the experiment. After 3 weeks, all oyster shells with spat were again digitally photographed and, afterwards, oysters were checked to determine whether they were alive or dead. Absolute shell growth for all live oysters was calculated by measuring the surface area (in mm2) of oysters at the beginning and the end of the experiment, using ImageJ software (Kelly et al., 2011). Results of the ANCOVA for shell area growth rate (mm2 d−1) in the various Boonea impressa treatments (black circles, no snails; white triangles, 0.5 snails per oyster; grey squares, 1.5 snails per oyster) with initial shell size (mm2) of Crassostrea virginica as the covariate. SPG was significantly affected by snail treatment (ANOVA F2,28 = 12.716, P < 0.001). Post-hoc analysis revealed that SPG was significantly reduced from no snail treatments by 34% in low and 80% in moderate snail treatments (P = 0.039 and P < 0.001, respectively). Additionally, oysters in the moderate snail treatments experienced 70% reduction in SPG relative to those in low snail treatments (P = 0.007, Fig. 2). Specific growth rate (ln increase in shell area mm2 d−1; SPG) for Crassostrea virginica spat in the presence of no Boonea impressa snails, low snail density (5 individuals per arena) and high snail density (15 individuals per arena). Letters above bars denote significant results from post-hoc tests. Error bars are standard error. These results demonstrate that relatively low, but field-relevant, densities of B. impressa can lead to a significant reduction in the shell growth rates of C. virginica spat. Pyramidellid gastropods have received attention due to concerns over their impacts on commercially valuable species, such as giant clams (Cumming & Alford, 1994; Boglio & Lucas, 1997) and C. virginica (White et al., 1984; Ward & Langdon, 1986; Wilson et al., 1988a). However, these studies have failed to investigate impacts on small bivalve spat, which this study suggests may be especially vulnerable. In previous studies, pyramidellid gastropods have been shown to reduce bivalve growth severely and even lead to mortality. In giant clam culture, high densities of ectoparasitic snails reduced clam growth rate by 25% and also produced clam mortality up to 100% when 10–20 snails were present per clam (Boglio & Lucas, 1997). However, parasite density on clams in culture tends to be very high, because the culture technique also protects the snails from predators, and similar densities are unlikely in the field (Cumming & Alford, 1994; Boglio & Lucas, 1997). For oysters, live-weight growth was reduced by 56% with 3 snails per oyster, 73% with 6 snails per oyster (Ward & Langdon, 1986) and 83% with 10 snails per oyster (White et al., 1984). Shell growth of 20–40-mm oysters was also reduced by 8% with 5 snails per oyster and up to 100% with 15 snails per oyster (White et al., 1988b). While methodological differences make direct comparisons between this study and previous studies difficult, oyster spat growth rates measured as SPG were reduced by 34 and 80% in the low and moderate snail treatments relative to control oysters. While our per-oyster snail densities were at least a factor of 2 (and up to a factor of 20) less than those used in previous studies, our results fall within the ranges of those reported in the literature. It is important to point out that the snail densities used in this study, on a per-oyster basis, while much lower than those used in other studies, are realistic field densities. In Hewlett's Creek, North Carolina, B. impressa densities have been found as high as 575 m−2 of bottom area (Hanke, 2014). While oyster densities also approach or exceed those of snails, the snails exhibit an aggregated distribution, typically occurring as ‘patches’ of between 5 and 15 snails per-oyster clump (defined as one or more oysters physically cemented together; Wilson et al., 1988a, 1991). These high-density patches are maintained for days, despite high snail movement between patches (Wilson et al., 1991). Therefore, the snail density integrated over the clump can be much lower than on a single oyster within the clump. A number of factors suggest that oyster spat may be especially vulnerable to attack by B. impressa. For example, B. impressa seems to prefer oysters <2.5 cm in length (White et al., 1984) and they recruit during the summer months (White, Kitting & Powell, 1985) when oysters are also recruiting. Additionally, juvenile B. impressa have difficulty feeding on larger oysters and thus target smaller prey, in which their proboscis is able to reach the haemolymph more easily (Powell et al., 1987), suggesting that small oyster spat might also be targeted by recruiting B. impressa. It is, therefore, likely that oyster spat are highly vulnerable to B. impressa and regularly experience densities of parasites in the field similar to those used during this study. The impacts of reduced growth on oyster spat can have serious cascading implications. By reducing growth rates, B. impressa also increase the window of vulnerability to predators. This might be especially problematic, as current research suggest that small crabs, such as xanthids, are more important in driving patterns in oyster predation, due to their size, distribution on reefs, natural abundance and claw morphology (Rindone & Eggleston, 2011; Hill & Weissburg, 2013; Johnson, Grabowski & Smee, 2014). Given that young B. impressa may select smaller prey (Powell et al., 1987), that aggregations of snails were not selective when given options of adult, juvenile and spat oysters (Wilson et al., 1988b), that snail recruitment overlaps oyster recruitment (White et al., 1985) and that low snail densities can significantly reduce spat growth (this study), oyster spat are particularly vulnerable to negative consequences of parasitism by B. impressa.