Stress-induced variation in host susceptibility to parasitic freshwater mussel larvae
- 395 Downloads
An increasing number of studies demonstrate the critical role of the host–parasite relationship for the persistence and distribution of freshwater mussels. Laboratory experiments are a powerful tool for quantifying the physiological compatibility between parasitic mussel larvae and fish hosts and are clearly applicable to species conservation. Recent findings, however, indicate potential need to control for biases caused by infection intensity and host stress responses. We tested glochidia metamorphosis success and host plasma cortisol response using Lampsilis siliquoidea glochidia on Lepomis macrochirus. The main aims were to (1) compare metamorphosis success in response to infection intensity (number of attached glochidia per fish) and (2) compare metamorphosis success between typical and reduced-stress approaches to the handling and infection of host fish. We found no effect of the glochidia bath density used to infect the fish (1000–4000–8000 glochidia l−1) or the resulting infection intensity on metamorphosis success, although host plasma cortisol was correlated with infection intensity at 24 h post infection. Small but statistically significant differences in metamorphosis success were observed between the typical and reduced-stress approaches. Overall, typical host compatibility testing methods appear to be robust to these variables, but more emphasis on standardizing laboratory protocols may provide more repeatable data.
KeywordsCortisol Glochidia Juvenile Metamorphosis Stress Unionidae
The worldwide decline of unionid bivalves has resulted in increasing efforts to understand their biology, promote conservation, and develop methods for captive propagation and population restoration (reviewed by Haag & Williams, 2014). Knowledge of the host relationships of unionoids is essential to achieve these goals (Kat, 1984; Barnhart et al., 2008, Douda et al., 2012; Haag & Stoeckel, 2015). Detailed knowledge of hosts is needed for population comparison, habitat assessment and management, and captive propagation (Geist, 2010; Douda, 2015; Ford & Oliver, 2015).
Unionid host species can sometimes be inferred from natural infections of glochidia on fish (Blažek & Gelnar, 2006; Österling, 2011; Levine et al., 2012). However, low abundance of mussels or hosts can render the collection of infected hosts impractical (Österling, 2011). Moreover, the observation of encysted glochidia provides little evidence of successful parasitism, because glochidia may be shed before metamorphosis is completed. Therefore, the most widely used approach to the determination of physiological hosts is the monitoring of artificially infected hosts under controlled laboratory conditions. These laboratory methods have been used for decades and provide the majority of current data on host relationships of unionid bivalves (e.g., Keller & Ruessler, 1997; Dodd et al., 2005; Fritts et al., 2012; Ford & Oliver, 2015; Haag & Stoeckel, 2015). Culturing protocols in breeding stations are based upon this knowledge and the functional hosts in natural habitats can be identified if the data on physiological host compatibility are properly combined with knowledge of host ecology or direct observations of natural infections (e.g., Schwalb et al., 2011; Fritts et al., 2012; Levine et al., 2012).
Several lines of evidence indicate that host immunological mechanisms determine the success or failure of attached glochidia. Both innate immune responses and also adaptive (antibody-based) responses are implicated in glochidia rejection (Meyers et al., 1980; Bauer, 1987a; Dodd et al., 2005, 2006, Rogers-Lowery et al., 2007). Low temperature can sometimes improve success of glochidia on hosts, perhaps by depressing host immune response (Roberts & Barnhart, 1999). Cortisol, which is released in response to stress, has a capacity to decrease vertebrate immune responses (Pickering & Pottinger, 1989) and artificially increased levels of cortisol may improve the metamorphosis success rate of glochidia (Kirk & Layzer, 1997; Dubansky et al., 2011).
Dubansky et al. (2011) presented evidence that laboratory host tests might be biased by the stress response and cortisol secretion in host fish. Their study recorded a positive relationship between the intensity of glochidia infestation and cortisol levels in fish blood. Because high glochidia loads are commonly used in experimental laboratory infestations, possible stress-induced alteration of metamorphosis success might bias many host evaluation trials. In the worst case, fish species identified as suitable hosts in high-intensity, high stress laboratory infections might be less suitable hosts in lower intensity infections in natural settings.
In the present study, we investigated the effects of infection intensity and stress stimuli in laboratory tests of glochidia metamorphosis on host fish. Our goal was to determine whether factors that commonly differ among studies might compromise the consistency and applicability of results. We tested different glochidia infection intensities and compared two alternative treatments differing in potential stress stimuli that may occur during laboratory infestations (insufficient fish acclimation, visual contact, handling stress). Cortisol level in host fish plasma was quantified and related to the treatments and to glochidia attachment and metamorphosis success.
The test species of mussel and fish were fatmucket (Lampsilis siliquoidea Barnes, 1823) and bluegill (Lepomis macrochirus Rafinesque, 1819). These species were selected because of their previous use in artificial glochidia infestations (Keller & Ruessler, 1997; Rogers & Dimock, 2003) and medium level of glochidia success recorded in previous experiments (Barnhart, unpublished data). Use of a species pair that yields intermediate metamorphosis success allows for either increase or decrease in response to treatments. Gravid females of L. siliquoidea were sampled in the Bourbeuse River (Gasconade County, MO 38°11′20.53″N, 91°33′54.11″W) and transferred to the laboratory at Missouri State University (MO, USA) on March 17, 2014. Glochidia were obtained by flushing the marsupia with water using a syringe, and their viability was verified by quantifying the closing response to sodium chloride in subsamples. The glochidia from six gravid females with a viability exceeding 90% were pooled and used for the inoculation. Different glochidia bath densities for the experiments were prepared by dilution of the initial bath on the basis of glochidia enumeration in ten 200 µl subsamples, taking care to keep the glochidia in suspension during sampling. Lepomis macrochirus were hatchery-raised and had no previous contact with glochidia. All fish were acclimated for at least 1 month under laboratory recirculating conditions (dechlorinated tap water, 12-h light cycle, temperature 20–24°C). Fish were fed daily ad libitum with frozen chironomid larvae prior to and during the experiments. Treatment of host fish was consistent with an approved institutional animal care and use protocol (Missouri State University IACUC 14-030.0).
Infection design and monitoring system
The monitoring system used was that of Dodd et al. (2005). Briefly, a recirculating aquarium system with multiple 1.5 l polycarbonate tanks holding individual fish (AHAB, Pentair Aquatic Habitats, FL, USA) was equipped with unit filters (105-µm nylon screens) that receive the outflowing water from each tank. The tanks were used both for fish infestation and subsequent holding for recovery of glochidia and juveniles. Temperature in the system was recorded with a HOBO data logger (Onset, USA) at 10 min intervals and was 24.3 ± 0.86°C (mean ± SD) over the course of the experiment.
Two different infection treatments (reported hereafter as “high stress” and “low stress”) were used to infest host fish in both experiments. The “high stress” treatment was designed to include the main likely sources of stress stimuli to host fish in laboratory infestation trials. Fish were acclimated only to laboratory conditions but not to the inoculation and monitoring system itself. The fish were initially held in a larger group tank, and were caught a hand net and moved to the AHAB system immediately before inoculation. The AHAB tanks were left transparent, so that the fish had visual contact with the researchers both during inoculation and the entire monitoring. The “low stress” infection treatment was designed to exclude or minimize these sources of external stress. Prior to infection, fish were acclimated to the AHAB tanks for 48 h before infection, and visual contact with researchers was prevented by covering the sides and bottoms of the tanks with an opaque plastic film.
For inoculation with glochidia, the AHAB system recirculation was paused and a measured volume of glochidia suspension was added to each tank to achieve the desired concentration. Glochidia were kept in suspension by an air-stone and periodic smooth pipetting through a small hole in the tank lid. After the inoculation period of 25 min, the tanks were switched to recirculating conditions for 20 min and at least 3 l min−1 per tank to remove unattached glochidia, which were recovered from the filters.
Experiment I: effect of infection intensity on metamorphosis success
Forty-six individuals of L. macrochirus (weight 3.7 ± 0.7 g, mean ± SD) were randomly divided into “high stress” and “low stress” treatment groups as described above. Fish from both groups were further divided into three different glochidia bath densities and simultaneously exposed to concentrations of 8000, 4000, or 1000 glochidia l−1 as described above. Glochidia and juvenile mussels were collected daily from the filter cups and counted under a stereomicroscope (40× magnification). Each individual was classified either as dead (untransformed glochidia or dead juveniles) or as living juveniles. Live juvenile mussels were identified by foot and valve movements. The total attached glochidia were equated with sum of glochidia and juveniles recovered from each fish during the whole monitoring period. Infection intensity was defined as the total attached glochidia divided by the mass of the fish in grams. Metamorphosis success was defined as the proportion of total attached glochidia that were recovered as live juveniles. Duration of parasitism was calculated from the recovery dates of live juveniles.
Experiment II: cortisol quantification and spatial distribution of glochidia on host fish
Thirty-six individuals of L. macrochirus (weight 3.2 ± 0.8 g, mean ± SD) were divided into high stress and low stress treatment groups that were treated as described above. Fish from both groups were further divided into three different glochidia bath densities and simultaneously exposed to concentrations of 8000, 1000, or 0 glochidia l−1 for 25 min. Fish were sampled 24 h after the exposure to glochidia. Each fish was anesthetized with buffered MS222 (Argent Laboratories, WA, USA) and their body weight and length were recorded. Blood was collected from caudal vein using heparinized capillary tubes (Fisher Scientific, 22-260-950, CA) wetted with a few µl of sample buffer (PBS + 1% BSA with 2 TIU Aprotinin ml−1). The tubes were centrifuged for 5 min at 3000 rpm in a microcapillary centrifuge (Damon IEC International Model MB, MA, USA). The plasma from each sample was collected from microtubes and stored at −80°C. Plasma cortisol concentrations were measured using a cortisol ELISA kit (Prod. No. ADI-900-071, Enzo, USA) according to manufacturer’s instructions. ELISA color reaction intensity was measured in a microplate reader (SpectraMax Plus 384, Molecular Devices, San Francisco, CA, USA) with the manufacturer’s software package. After blood sampling, fish were dissected and the number of glochidia attached to gills, fins, and other body parts (mouth, operculum, nostrils) were counted under stereomicroscope (40× magnification). The proportion of glochidia attached to gills was calculated for each fish.
In Experiment I, two-way analysis of variance (ANOVA) was used to examine the effects of glochidia bath density (8000, 4000, 1000 glochidia l−1) and the infestation approach (“high stress” versus “low stress”) separately on infection intensity (the total number of glochidia attached per gram of fish body weight), metamorphosis success (the percent of attached glochidia recovered as live juveniles), and the duration of parasitism (days from attachment to recovery of a live juvenile). In Experiment II, a two-way ANOVA examined the effects of glochidia bath density (8000, 1000, 0 glochidia l−1) and the infestation approach on cortisol levels recorded in host fish plasma. Additional two-way ANOVAs were made to characterize the initial stage of parasitism in experiment II using dataset restricted to the groups of 8000 and 1000 glochidia l−1. We used the same explanatory variables to examine variation of infection intensity (n g−1) and proportion of glochidia attached to gills. The interaction terms of explanatory variables were included to all two-way ANOVAs but the interactions were not statistically significant and are not reported in the results. Prior to statistical analyses, data were assessed for homogeneity of variance and normality using a Levene’s test and a Kolmogorov–Smirnov test, respectively. Datasets that did not meet the underlying assumptions were transformed by means of a natural log transformation and proportion data were arcsine transformed prior to analyses. In all tests, an alpha level of 0.05 was used. Data analysis was performed using R software (R Development Core Team, 2013).
We found no effect of glochidia infection intensity on metamorphosis success of L. siliquoidea glochidia on L. macrochirus. This result does not support the hypothesis that increasing infection intensities promote enhanced survivorship of the larvae (Jansen et al., 2001; Dubansky et al., 2011). The available studies provide inconsistent results. Dubansky et al. (2011) reported that metamorphosis success of North American Utterbackia imbecillis on Lepomis macrochirus was strongly dependent on infection intensity and on host environment during inoculation. In contrast, Douda et al. (2014) found only weak or nonsignificant relationship between the infestation bath densities ranging from 1000 to 8000 glochidia l−1 and metamorphosis success in 30 experimental infestations with central European Unio crassus. A detailed study on the density-dependent success of M. margaritifera glochidia showed no general relationship between infection intensity and metamorphosis success, which varied across species combinations and fish host age classes (Bauer, 1987b). Hence, density dependence of glochidia success on host fish does not appear to be a common result in experimental studies. This does not rule out the possibility, however, that otherwise stressful experimental conditions might obscure such a relationship.
The slightly but significantly shortened duration of successful parasitism observed in fish infected at higher glochidia densities might be related to an adaptive immune response. In some previous studies, glochidia exhibited reduced duration of successful parasitism and metamorphosis success when fish were reinfected (Rogers & Dimock, 2003; Dodd et al. 2005). Antibody production was also observed (Dodd et al., 2005). The reduced duration of parasitism observed in our study at higher infection intensities might likewise be due to an incipient adaptive immune response.
While the infection intensity had no detectable effect on glochidia metamorphosis success, we recorded weak but significant differences between the high and low stress treatments. The differences between stress levels appeared most pronounced at the intermediate glochidia density (Fig. 1b), but the interaction term (stress × density) was not significant in our analysis and further experiments with larger sample size are needed to investigate whether the effect of host’s stress on glochidia metamorphic success can be density dependent. Anyway, higher metamorphosis success in the “high stress” treatment supports the hypothesis that environmental conditions can influence metamorphosis success. Both the innate and adaptive mechanisms of the fish immune system can be suppressed by acute and chronic effects of environmental factors that exceed specific levels (Bly et al., 1997). Despite the intensive research efforts on the environmental effects on fish immune mechanisms against many parasitic and pathogen groups (e.g., Rohlenová et al., 2011), studies of immune effects on glochidia metamorphosis success are rare (Jansen et al., 2001; Dodd et al., 2005, 2006; Rogers-Lowery et al., 2007). Metamorphosis success of glochidia of North American Anodonta suborbiculata Say 1831 increased on fish hosts but not in vitro at low temperature, suggesting immunosuppression of the host (Roberts & Barnhart, 1999). Immunosuppression could be also responsible for the observed effects of host treatment in our study, although we were unable to support this with cortisol measurements.
Cortisol is considered to be an important signaling mechanism resulting in immunosuppression in fish (Bly et al., 1997). Dubansky et al. (2011) reported a significant increase in plasma cortisol of L. macrochirus exposed to >1000 l−1 of glochidia of Utterbackia imbecillis and related this response to glochidia metamorphosis success. Although we found that plasma cortisol was positively correlated with glochidia infection intensity of individual L. macrochirus at 24 h post infection in Experiment II (in the 8000 glochidia l−1 treatment), we failed to show an effect of infection intensity or glochidia bath density on metamorphosis success in Experiment I. Moreover, although the “high stress” infection method apparently increased metamorphosis success in Experiment I, we were unable to demonstrate an effect of stress treatment on plasma cortisol. Possibly our two infection treatments did not present sufficiently different stress levels. Also, cortisol levels in fish plasma can vary within minutes and hours (Barton et al., 1980) and might have returned to normal after 24 h. Further studies are needed to fully understand the relationship between cortisol level dynamics in fish and glochidia metamorphosis success. Detailed investigation of immunocompetence of fish hosts (e.g., blood cell composition, activation of lysozyme, respiratory burst activity) could be helpful for addressing this issue.
The proportion of glochidia attached to gills versus skin can affect metamorphosis success (Jansen et al., 2001). It appears possible that laboratory infections could differ from natural infections (and among different experimental approaches) in the distribution of glochidia on host surfaces, and such differences might alter the apparent suitability of hosts. Glochidia of many Unionidae attach and metamorphose effectively on skin and fins (e.g., Jansen et al., 2001; Blažek & Gelnar, 2006), whereas most Ambleminae and Margaritiferidae utilize primarily the gills of the host (Barnhart et al., 2008). Lampsilis use a mantle lure to attract the host to strike the marsupial gill and inhale glochidia, so the glochidia are more likely to attach to gills than to skin in natural infections. In contrast, infection by swimming in a suspension provides the opportunity to attach to both gills and skin, and the spatial attachment of glochidia reflects fish behavior such as swimming and ventilation rates during the contact with glochidia (Paling, 1968).
Previous studies of host relationships of unionid bivalves vary in experimental settings for infestations (e.g., bath density, inoculation time, length of fish acclimation period) and monitoring of glochidia development on their host fish (e.g., conditions of fish holding units), which may lead to differences in the observed developmental success. The results of the present study indicate that there is no profound systematic bias caused by elevated glochidia densities typically used in laboratory experiments. In this view, typical host compatibility testing methods are relatively robust and can provide consistent results across variable experimental settings. Yet, more emphasis on laboratory conditions, including careful acclimation of fish hosts to experimental conditions, and minimization of fish handling during experiments, can be recommended for the evaluation of host compatibility. More experiments are necessary to determine whether infestation method might bias apparent metamorphosis success by altering the distribution of glochidia on the host skin, fin, and gills. Efforts to minimize stress stimuli may help to further improve infestation protocols for glochidia of freshwater mussels to provide more repeatable data. Better understanding of the factors that influence glochidia success can improve the evaluation of physiological host compatibility for freshwater mussels and thus increase the efficiency of the management of their host resources.
The authors thank Laszlo Kovacs, Evan Clark, and Paul Durham for their valuable comments and help with experimental procedures. The research was funded by the Czech Science Foundation (13-05872S) and Missouri State University.
- Bauer, G., 1987a. The parasitic stage of the freshwater pearl mussel (Margaritifera margaritifera L.) I. Host response to glochidiosis. Archiv für Hydrobiologie 76: 393–402.Google Scholar
- Bauer, G., 1987b. The parasitic stage of the freshwater pearl mussel (Margaritifera margaritifera L.) III. Host relationship. Archiv für Hydrobiologie 76: 413–423.Google Scholar
- Ford, D. F. & A. M. Oliver, 2015. The known and potential hosts of texas mussels: implications for future research and conservation efforts. Freshwater Mollusk Biology and Conservation 18: 1–14.Google Scholar
- Jansen, W., G. Bauer & E. Zahner-Meike, 2001. Glochidial mortality in freshwater mussels. In Bauer, G. & K. Wachtler (eds.), Ecology and Evolution of the Freshwater Mussels Unionoida. Springer, Berlin.Google Scholar
- Kirk, S. G. & J. B. Layzer, 1997. Induced metamorphosis of freshwater mussel glochidia on nonhost fish. Nautilus 110: 102–106.Google Scholar
- Meyers, T. R., R. E. Millemann & C. A. Fustish, 1980. Glochidiosis of salmonid fishes. IV. Humoral and tissue responses of coho and chinook salmon to experimental infection with Margaritifera margaritifera (L) (Pelecypoda, Margaritanidae). Journal of Parasitology 66: 274–281.CrossRefPubMedGoogle Scholar
- R Development Core Team, 2013. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Viena.Google Scholar