J. Evan Ward

Marine aggregates facilitate ingestion of nanoparticles by suspension-feeding bivalves.

As the application of nanomaterials to science and technology grows, the need to understand any ecotoxicological effects becomes increasingly important. Recent studies on a few species of fishes and invertebrates have provided data which suggest that harmful effects are possible. The way in which nanoparticles are taken up by aquatic organisms, however, has been little studied. We examined uptake of nanoparticles by two species of suspension-feeding bivalves (mussels, Mytilus edulis; oysters, Crassostrea virginica), which capture individual particles < 1 microm with a retention efficiency of <15%. Given this limitation, it would appear that nanoparticles could not be ingested in large numbers. During certain times of the year, however, >70% of suspended particles are incorporated within aggregates that are > 100 microm in size. Therefore, we delivered bivalves fluorescently labeled, 100-nm polystyrene beads that were either (1) dispersed or (2) embedded within aggregates generated in the laboratory. Results indicate that aggregates significantly enhance the uptake of 100-nm particles. Nanoparticles had a longer gut retention time than 10-microm polystyrene beads suggesting that nanoparticles were transported to the digestive gland. Our data suggest a mechanism for significant nanoparticle ingestion, and have implications for toxicological effects and transfer of nanomaterials to higher trophic levels.

Site of particle selection in a bivalve mollusc

Bivalve molluscs form dense populations that exert profound effects on the particle loads and phytoplankton composition of coastal waters. It has long been known that bivalves can select among different particle types, including selecting against those of poor nutritional value, but because of difficulties in observing particle transport processes in the pallial cavity in vivo, the mechanism of selection was not known. We now use a combination of video endoscopy and flow cytometry to show that oysters can select living particles from non-living detritus on the gills. Our methods could aid the study of suspension feeding in many animal groups.

Biodynamics of suspension-feeding in adult bivalve molluscs : particle capture, processing. and fate

The technique of endoscopic examination and video-image analysis has allowed for in vivo observations of adult, bivalve molluscs and has led to a more accurate understanding of suspension-feeding processes in these animals. Several mechanisms, however, needed to be examined further. For example, the process by which particles are captured by the ctenidia remained unclear, and current models of this process did not adequately explain in vivo observations of capture events. In addition, the kinematics of particle processing by the labial palps and mode of particle ingestion have remained undefined. In this study I use results from endoscopic observations of nine species of bivalves to provide an integrated description of suspension feeding. These observations reveal that particle feeding is accomplished by both mucociliary and hydrodynamic mechanisms that often act simultaneously on the pallial organs. I introduce a model of particle capture that focuses on the ctenidial filament as the capture unit, unlike previous work that emphasized the laterofrontal cilia or cirri. The model is consistent with accepted fluid dynamic principles. In vivo observations in bivalves indicate that the suspension-feeding complex as a whole is, functionally, more than merely the sum of its parts. Additional key words: transport, ingestion, kinematics Suspension-feeding bivalves are a large group of ecologically and economically important aquatic animals. In many habitats they can be dominant infaunal or epibenthic organisms, affecting the surrounding community through processes such as benthic-pelagic coupling, particle depletion, and nutrient cycling (Peterson & Black 1987; Asmus & Asmus 1993; Smaal & Prins 1993). Therefore, many aspects of bivalve biology and ecology have been studied and described (Bayne et al. 1988; Bayne & Hawkins 1992; Bayne 1993; Dame 1993; Grant et al. 1993; Newell & Shumway 1993), including anatomical and physiological aspects of suspension-feeding (J0rgensen 1966, 1990; Owen 1974; Morton 1983). Most of these descriptions of suspension-feeding have been largely based on surgically altered specimens or on feeding structures isolated from the pallial cavity. While valid under certain circumstances, results from most of these studies should be interpreted with caution because surgery and isolation can: (1) cause feeding structures to function abnormally, (2) alter hydrodynamics of flow around the structures, and (3) destroy subtle interactions between adjacent feeding structures (e.g., ctenidia and labial palps). In addition, extrapolation of in vitro results to in vivo mechanisms implicitly assumes that the filaments or structures function as discrete units and that the feeding complex functions merely as the sum of its constituent parts (e.g., Nielsen et al. 1993). This may not be a valid assumption. In fact, previous researchers have suggested that mechanisms of feeding cannot be understood from studies on exposed ctenidia or ctenidial filaments (e.g., J0rgensen 1976). During the past several years, in vivo studies in intact bivalves by means of video endoscopy have provided a better understanding of suspension-feeding processes (Beninger et al. 1992; Ward et al. 1993; Ward et al. 1994). However, the details of many feeding processes, especially as they occur in intact animals, remain unclear. For example, the process by which particles are captured by the ctenidia has been a subject of debate for many years. Some workers have maintained that particles are mechanically trapped by rows of laterofrontal cilia or cirri (Tammes & Dral 1955; Dral 1967; Moore 1971), while others have suggested that particles are retained by hydrodynamic forces that operate at the level of interfilamentar spaces (Owen & McCrae 1976; J0rgensen 1981, 1990). AsThis content downloaded from 207.46.13.129 on Wed, 29 Jun 2016 04:33:02 UTC All use subject to http://about.jstor.org/terms Suspension-feeding biodynamics in bivalves pects of both theories, however, violate some of the physical and biological constraints imposed by the intact feeding system, and neither adequately explains in vivo observations of capture. The goal of this study was to integrate the observations and results obtained by video endoscopy, including previously unpublished data, with those obtained by more invasive techniques, in order describe a more holistic model of suspension-feeding biodynamics. Specifically, data on particle capture and transport by the ctenidia, transfer of particulate matter to the labial palps, processing of material by the palps, and ingestion of food material will be presented. I do not attempt a detailed, exhaustive explanation of feeding processes; rather, I focus on general mechanisms that are common to most suspension-feeding bivalves and that serve to unify this group. Where appropriate, my results are placed in the context of previous descriptions of feeding mechanisms in bivalves.

Lectins Associated With the Feeding Organs of the Oyster Crassostrea virginica Can Mediate Particle Selection

Despite advances in the study of particle selection in suspension-feeding bivalves, the mechanisms upon which bivalves rely to discriminate among particles have not been elucidated. We hypothesized that particle sorting in suspension-feeding bivalves could be based, in part, on a biochemical recognition mechanism mediated by lectins within the mucus that covers the feeding organs. Using Crassostrea virginica, the Eastern oyster, our investigations demonstrated that lectins from oyster mucus can specifically bind several microalgal species as well as different types of red blood cells (RBC), triggering their agglutination. Agglutination of microalgal species and RBC varied with the source of mucus (gills vs. labial palps). Hemagglutination and hemagglutination inhibition assays emphasized that mucus contains several lectins. In feeding experiments, Nitzschia closterium and Tetraselmis maculata were separately incubated with mucus before being fed to oysters. Results showed that pre-treating these microalgae with mucus significantly alters the ability of oysters to sort particles. In another experiment, oysters were fed a mixture of microspheres coated with either bovine serum albumin (BSA) or glucosamide-BSA. Results show that oysters preferentially ingest microspheres with bound carbohydrates, highlighting probable interactions between lectins and carbohydrates in the mechanisms of microalgae recognition. This study confirms the presence of lectins in mucus that covers the feeding organs of oysters and suggests a new concept with regard to particle processing by suspension-feeding bivalves: specific interactions between carbohydrates on the surface of particles and lectins within the mucus mediate the selection and rejection processes.

Influence of diet on pre-ingestive particle processing in bivalves: I: Transport velocities on the ctenidium

Suspension-feeding bivalve molluscs can assume a large ecological role by linking benthic and pelagic ecosystems. Therefore, a knowledge of the factors that influence feeding rates and processes at the level of the individual is important in understanding bivalve-dominated environments. We examined the roles of diet quality and concentration on particle processing by the ctenidia of four species of bivalves: the mussels Mytilus edulis L. and Mytilus trossulus Gould, and oysters Crassostrea virginica (Gmelin) and Crassostrea gigas (Thunberg). Bivalves were delivered diets of three different qualities at three different concentrations (1–2×103, 104, 105 particles ml−1). The high-quality diet consisted of the cryptophyte Rhodomonas lens Pascher et Ruttner; the low-quality diet consisted of particles prepared from ground Spartina sp. detritus; the medium-quality diet consisted of a 50:50 mixture (by particle number) of both particle types. Particle transport velocities on the ventral groove and dorsal tracts were then measured by means of video endoscopy. Ventral-groove transport velocities of M. edulis were the most responsive, demonstrating a significant increase with increasing diet quality, and a significant decrease with increasing diet concentration. Transport velocities in the ventral groove and dorsal tracts of C. virginica were not significantly affected by changes in diet quality, but significantly increased with increasing diet concentration. Transport velocities of M. trossulus and C. gigas demonstrated little change with diet quality or concentration, indicating that responses are species specific. Our data suggest that differential control of particle transport on the bivalve ctenidium is one of the underlying mechanisms that contribute to compensatory feeding responses exhibited by the entire organism.

Feeding activity, absorption efficiency and suspension feeding processes in the ascidian, Halocynthia pyriformis (stolidobranchia: Ascidiacea): responses to variations in diet quantity and quality

The benthic suspension feeding ascidian, Halocynthia pyriformis (Rathke, 1806), is often exposed to high concentrations of resuspended sediment in the Bay of Fundy. Resuspended sediment can change diet quantity and quality that may alter the ascidians ability to feed and gain energy. The feeding activity of H. pyriformis exposed to bottom sediment was examined using standard physiological techniques and video endoscopy. Ascidians were exposed to natural seston plus additions of bottom sediment ranging in concentration from 0 to 46 mg l(-1). For each sediment concentration, clearance rate, ingestion rate, and retention efficiency of the ascidians was estimated using flow-through feeding chambers. Samples of suspended particles and feces were collected to estimate absorption efficiency and absorption rate. Results indicate that with increasing sediment concentration, ingestion rate increased to a constant level, absorption rate increased linearly despite a logarithmic decrease in absorption efficiency, and the retention of small particles (2-5 &mgr;m) increased while retention of larger particles (5-15 &mgr;m) decreased. As sediment concentration increased, squirting frequency increased and diameter of the siphon was reduced. Endoscopic observation of feeding structures and processes and the measurement of particle velocity was performed on ascidians exposed to 0 and 10 mg l(-1) of bottom sediment. An increase in squirting frequency at the high concentration facilitated the rejection of unwanted material and altered the structure and transport velocity of mucus. Mucus velocity was five times slower at 10 mg l(-1) than at 0 mg l(-1), however, the overall distance of mucus travel and the probability of clogging was reduced at 10 mg l(-1). H. pyriformis appears to compensate for episodic changes in the quantity and quality of available food particles by altering siphon-opening diameter, squirting frequency, structure and transport of mucus, and retention efficiency to maintain constant clearance rates.

Microalgal Cell Surface Carbohydrates as Recognition Sites for Particle Sorting in Suspension-Feeding Bivalves

Cell surface carbohydrates play important roles in cell recognition mechanisms. Recently, we provided evidence that particle selection by suspension-feeding bivalves can be mediated by interactions between carbohydrates associated with the particle surface and lectins present in mucus covering bivalve feeding organs. In this study, we used lectins tagged with fluorescein isothiocyanate (FITC) to characterize carbohydrate moieties on the surface of microalgal species and evaluate the effect of oyster mucus on lectin binding. These analyses revealed that concanavalin A (Con A), one of six lectins tested, bound to Isochrysis sp., while Nitzschia closterium reacted with Pisum sativum agglutinin (PNA) and peanut agglutinin (PEA). The cell surface of Rhodomonas salina bound with PNA and Con A, and Tetraselmis maculata cell surface was characterized by binding with PNA, PEA, and Con A. Pre-incubation of microalgae with oyster pallial mucus significantly decreased the binding of FITC-labeled lectins, revealing that lectins present in mucus competitively blocked binding sites. This decrease was reversed by washing mucus-coated microalgae with specific carbohydrates. These results were used to design a feeding experiment to evaluate the effect of lectins on sorting of microalgae by oysters. Crassostrea virginica fed with an equal ratio of Con A-labeled Isochrysis sp. and unlabeled Isochrysis sp. produced pseudofeces that were significantly enriched in Con A-labeled Isochrysis sp. and depleted in unlabeled microalgae. Selection occurred even though two physical-chemical surface characteristics of the cells in each treatment did not differ significantly. This work confirms the involvement of carbohydrate-lectin interaction in the particle sorting mechanism in oysters, and provides insights into the carbohydrate specificity of lectins implicated in the selection of microalgal species.

Characteristics of Marine Aggregates in Shallow-water Ecosystems: Implications for Disease Ecology

Marine aggregates were evaluated for their potential role in the ecology of aquatic pathogens using underwater video surveys coupled with direct collection of aggregates in modified settling cones. Six locations, two each in New York, Connecticut, and Massachusetts, were surveyed over 8 months to explore differences in the characteristics of aggregates found in habitats populated by clams (Mercenaria mercenaria) and oysters (Crassostrea virginica). Microaggregate (<500 μm) concentrations were always greater than macroaggregate (>500 μm) concentrations, but peak concentrations of macroaggregates and microaggregates, mean size of particles, and volume fraction of aggregated material varied among the six shallow-water habitats. Concentrations (colony-forming units per ml) of total heterotrophic bacteria (THB) and total mesophilic pathogenic bacteria (MPB) from samples of aggregates were significantly different among the four locations bordering Long Island Sound (LIS). The highest concentrations and enrichment factors in aggregates were observed in August for THB and in June for MPB. Significant correlations were detected for salinity and the concentrations and enrichment factors of THB in aggregates and for the concentrations and percentages of MPB in seawater samples. Significant correlations were also detected for temperature and the concentrations of MPB in aggregates and the enrichment factors for THB and MPB (marginal significance). Bacterial species identified in association with aggregates included: Vibrio cholerea, V. parahaemolyticus, V. vulnificus, V. alginolyticus, Aeromonas hydrophila, Pseudomonas aeruginosa, Escherichia coli, and Mycobacteria sp. These results have important implications for the way in which aquatic pathogens are collected, quantified, and monitored for risk-based surveillance in shallow-water ecosystems.

Influence of diet on pre-ingestive particle processing in bivalves: II. Residence time in the pallial cavity and handling time on the labial palps

Abstract Previous studies have demonstrated that bivalve molluscs respond to changing food conditions through processes such as preferential selection and ingestion of particulate matter. Little is known, however, about the underlying mechanisms accountable for these responses. To further explain feeding processes at the level of the pallial organs, we determined pallial cavity residence times, or the amount of time it took particles to travel from the inhalant aperture to the stomach, in two species of bivalves, Crassostrea virginica and Mytilus edulis, under conditions of differing particle quality, particle concentration, and temperature. From these residence times, particle-handling times on the labial palps were determined. Diets of three different qualities were tested, including Rhodomonas lens cells, particles prepared from ground Spartina sp. detritus, and a 50/50 mixture of both. Bivalves were delivered one of the three diets along with 10-μm fluorescent polystyrene beads (tracer), removed from feeding chambers at intervals from 30 s up to 20 min, and placed in liquid nitrogen to halt particle transport. Digestive systems of bivalves were then dissected and examined for the presence of tracer beads. Particle-residence times in the pallial cavity and handling times on the labial palps of C. virginica were significantly affected by changes in diet type. Particle-handling times on the palps decreased with increasing diet quality and ranged from 2.2 min (100% R. lens) to 22.8 min (100% ground Spartina sp.), accounting for 88% and 99%, respectively, of the total time particles spent in the pallial cavity. In contrast, diet quality had little effect on particle-residence times in the pallial cavity of M. edulis. However, residence times were affected by temperature and diet concentration. Temperature significantly affected residence times at particle concentrations of both 20 and 100 particles μl−1, whereas particle concentration affected residence times at 20 °C, but not at 5 °C. Particle-handling times on the labial palps ranged from less than 1 to 5.5 min, depending on temperature and concentration, accounting for 50% to 82%, respectively, of the total time particles spent in the pallial cavity. We suggest that (1) observed interspecific differences in particle handling on the labial palps may be due to differences in palp morphology and function, and (2) particle sorting and selection on the labial palps is a rate-limiting step of pre-ingestive feeding processes in by bivalves.

Biodynamics of Particle Processing in Bivalve Molluscs: Models, Data, and Future Directions

Previous models of particle feeding have focused on optimal solutions for particle acquisition or absorption. We propose two conceptual approaches to treat particle feeders as an integrated system of compartments, in hopes of understanding critical limiting factors that might be overlooked by focusing on only one part. The compartment model treats a particle feeder as a series of structures that process particles, with characteristic residence times within compartments and transfer points between them. These might change with overall particle food value and proportion of poor particles. As a non-exclusive alternative, the pathway model considers particle transfer as being analogous to enzyme control systems, with feedback loops that may involve interactions such as negative feedback between compartments that engage in no direct transfer. We examine these models in the light of some studies of particle handling by the deposit-feeding bivalves Yoldia limatula, Macoma secta, and M. nasuta, and the suspension-feeding oyster Crassostrea gigas. In Y. limatula, palp overloading results in feedback that shuts down the particle-collecting palp proboscis. In Macoma, nearly all particles are rejected, suggesting that rejection is necessary because digestion and gut residence time are limiting factors. We suggest that a whole-system approach is important in understanding particle processing by deposit feeders and suspension feeders. Additional key words: feeding, deposit feeding, suspension feeding, symmorphosis The purpose of this paper is to provide a conceptual framework for analyzing the processing of food particles by bivalve molluscs. We model the component organ systems as a series of compartments with transfer points between them, but with potential feedback loops among all potential compartments. Because the various compartments-ctenidium, palp, gut, and their components-have differing structures and functions, mechanisms are necessary for transfers of particles between them, e.g., the mucus string that transports particles between the ctenidium and palp in suspension feeders. Owing to disparate evolutionary origins and inherent structural difficulties, the supply of particles from one compartment to the next might cause overloading or rates of particle supply from one compartment that are below the capacity of another. In other words, the particle-processing system found in bivalve molluscs and other organisms might not function optimally with fixed transfer rates. Alternatively, the system might consist of a series of components whose individual processing rates are fixed and perfectly adapted to each other, much as has been suggested in arguments for other complex systems (Taylor & Weibel 1981; Weibel et al. 1991). We will first present a general compartment model for examining particle feeders such as bivalve molluscs. We describe compartments specifically for a selected group of bivalve species and show how problems ranging from phylogeny to the study of particle transfer between compartment structures can be studied. We will also use another conceptual model, derived from enzyme pathway analysis, to point out how interactions among the compartments can be considered. Finally, we will present some data on two species of deposit-feeding bivalves and some preliminary data on a suspension feeder to demonstrate how this approach allows a sharpening of our understanding of particle feeding and processing. This content downloaded from 207.46.13.52 on Mon, 24 Oct 2016 04:13:08 UTC All use subject to http://about.jstor.org/terms Models of particle processing Bivalve particle feeding Benthic particle feeders are typically divided into deposit feeders and suspension feeders. These represent a continuum rather than two discrete modes, although specific mechanisms of collection and processing may differ. Within the superfamily Tellinacea, for example, we encounter a range of morphologies, from those best suited for suspension feeding to those more efficient at deposit feeding (Yonge 1949; Pohlo 1966, 1967). Many particle-feeding species, including bivalves, switch from deposit to suspension feeding, depending upon substratum and hydrodynamic conditions (Taghon et al. 1980; Dauer 1983; Miller 1984; Olafsson 1986; Miller et al. 1992). The food source for both feeding types, however, is often quite similar, as it may consist of material resupplied by storm and other resuspension events (Miller et al. 1984; Frechette et al. 1989; Judge et al. 1993). In both bivalve feeding types, particles are collected actively, followed by one or more particle-processing stages, which usually involve particle rejection by the ctenidium and palp before particles enter the gut. Some food particles then pass directly through a tubular gut while others may be shunted to a diverticular structure where complete intracellular digestion occurs. Benthic particle feeders can be examined with regard to two principal problems. How do they deal with (1) variation of particle load (volume of particles supplied per unit time), and (2) varying proportions of particles of differing quality? Both deposit feeders and suspension feeders are confronted with widely varying particle loads and ranges of quality. Deposit feeders deal with high loads of particles that are mostly indigestible and that may require rejection before appropriate particles enter the gut (Lopez & Levinton 1987). They deal with this problem partly by non-random search and particle-collection strategies (Taghon et al. 1978; Levinton 1980; Whitlatch & Weinberg 1982), by variation of ingestion rate when local patches of high quality particles are encountered (Taghon 1981; Taghon & Jumars 1984), and by rejection of low quality particles (Hylleberg & Gallucci 1975; Taghon 1982). The bivalve Macoma nasuta typically rejects over 95% of the material that enters its incurrent siphon (Hylleberg & Gallucci 1975). By contrast, some sea cucumbers feed nearly indiscriminately, and digestive processing must be a major consideration in maximizing energy uptake (Powell 1977; Sloan & von Bodungen 1980). Suspension feeders also deal with a wide variety of particle loads and with particles of varying quality. At extreme loading, suspension feeders may adjust pumping rates, but bivalve molluscs also collect particles on the ctenidium and then reject those that are non-nutritive in the form of pseudofeces (Ki0rboe et al. 1980). There is growing evidence for qualitative selection (Newell & Jordan 1983; Ward & Targett 1989; MacDonald & Ward 1994) even of similar-sized organic particles (Shumway et al. 1985). Ingestion rates and other feeding processes appear to respond to changing content of non-nutritive particles (Bayne et al. 1989; Iglesias et al. 1992). Depending upon the degree of non-nutritive particle loading, cockles may respond by altering rates of particle uptake and the degree of rejection of particles as pseudofeces (Iglesias et al. 1992). In sum, particle feeders appear to respond in manifold and complex ways to the two cardinal problems outlined above. Previous conceptual approaches Three distinct but certainly not contradictory approaches have been applied to the modeling of particle processing, as well as other functional problems. The first approach involves models that place constraints upon feeding. Certain features of bivalve feeding have been argued to have characteristic hydrodynamic properties, and several authors have devised models that limit bivalves to certain physical mechanisms of particle processing, such as mechanical transport (Whitlatch & Weinberg 1982; J0rgensen 1989). More general models of particle feeding predict that certain particle sizes are collected with less efficiency than others, owing to hydrodynamic constraints (Rubenstein & Koehl 1977; Shimeta 1993). Processing rates also may be fixed, which could lead to characteristic transfer rates and loadings in a particle-processing chain. The second approach involves development of compensation models of feeding. Compensation is a change in behavior (e.g., rate of water transport through the siphon) following a change in feeding conditions (e.g., change in particle concentration in the water) that increases the rate of food uptake by the animal over the rate that would be found if no compensation occurred. For example, some studies report changes in rate of inflow that compensate for changes in phytoplankton concentration (Bayne et al. 1988). These models implicitly assume an array of optimal solutions corresponding to changing conditions and that bivalves can adjust at least in the direction of these optimal states. Finally, optimal models are more explicit renditions of compensation models. Such models take a series of boundary conditions and predict an optimal solution to foraging or digestion. For example, Penry & Jumars (1987) argued that during deposit feeding, particles 233 This content downloaded from 207.46.13.52 on Mon, 24 Oct 2016 04:13:08 UTC All use subject to http://about.jstor.org/terms Levinton, Ward, & Thompson Fig. 1. Conceptualization of a bivalve mollusc as a system of compartments with transfer between