Ion Transport and Acid–Base Balance in Freshwater Bivalves

Although freshwater bivalves are a polyphyletic group, they display striking convergence in some of their physiological adaptations to a dilute medium. In particular, the osmolality of their body fluids is remarkably low, ranging from 35 to 50 mosmol kg⁻¹ in Dreissena polymorpha (Dreissenidae) to 50–65 mosmol kg⁻¹ in Corbicula fluminea (Corbiculidae), and with values for members of the Unionidae averaging 40–50 mosmol kg⁻¹ (Potts, 1954; Dietz, 1979; Byrne et al. 1989; Horohov et al. 1992; Dietz et al. 1996). In general, Na⁺ is the predominant cation, with HCO₃⁻ and Cl⁻ being the major anions. There are significant differences between freshwater bivalve groups: Dreissena and Corbicula have Cl⁻ as their predominant anion, whereas the unionids typically employ Cl⁻ and HCO₃⁻ equally as major anions in their blood (Dietz, 1979; Horohov et al. 1992).

In order to maintain even these low levels of solute, freshwater bivalves must actively transport the major ions from the dilute medium. The gills are the site of both Na⁺ (Dietz and Findley, 1980) and Cl⁻ uptake (Dietz and Hagar, 1990). For the unionids and corbiculids, a Na⁺/H⁺ exchanger and a Cl⁻/HCO₃⁻ exchanger are the main transporters present, as shown by transport kinetic studies and the consequences of application of specific inhibitors (Dietz, 1978; Dietz and Branton, 1979).

Na⁺ transport is consistently stimulated in unionids and corbiculids by exogenous application of the neurotransmitter serotonin (5-hydroxytryptamine, 5-HT) in the bathing medium (Dietz et al. 1982; Zheng, 1996). Serotonin also stimulates Cl⁻ transport in corbiculids (Zheng, 1996) and in D. polymorpha (Horohov et al. 1992).

The blood (more accurately the extracellular fluid) of freshwater and marine bivalve groups is very low in protein (Byrne and McMahon, 1991; Booth et al. 1984). Thus, the non-bicarbonate buffering capacity of the blood, which is usually directly related to the protein content, is also extremely low. Therefore, changes in the controlling components of acid–base balance, such as ⁠, and changes in the strong ion difference (SID) (Stewart, 1978) result in relatively large changes in blood pH (Byrne et al. 1991, 1995; Byrne and McMahon, 1991, 1994).

The interaction between ion transport and acid–base balance has been well studied in aquatic vertebrates, and particular attention has been given to mechanisms involving the gills of freshwater fish (Maetz et al. 1976; Perry et al. 1982; for a review, see McDonald et al. 1989). A combination of Na⁺/H⁺; Na⁺/NH₄⁺ and Cl⁻/HCO₃⁻ exchange on the apical (bath side) surface of gill epithelia is largely responsible for gill-mediated ion exchange. This combination of cation and anion transport exchanges a strong ion for an acid or base equivalent. Thus, measurement of net movements of Na⁺ and Cl⁻ yields information on excretion or uptake of acid equivalents (McDonald et al. 1989).

In the present study, we examined the following questions. Can measurements of Na⁺ and Cl⁻ transport aid in the understanding of acid–base state in freshwater bivalves? Can neuromodulators that alter ion transport variables exert predictable effects on acid–base state? Are there differences in response or mechanism between phyletically distinct groups of freshwater bivalve, and does this information aid in the understanding of their physiology?

Experiments were designed to examine Na⁺ and Cl⁻ fluxes and blood acid–base variables of clams exposed to a pondwater medium and to compare these responses with those of animals incubated in medium deficient in either Na⁺ or Cl⁻. Animals were also stimulated with the neuromodulator serotonin so that the acid–base effects of changes in epithelial ion transport would be more apparent.

Specimens of the corbiculid Corbicula fluminea (Müller) were collected, under permit, either from Anococo Creek, 10 km northwest of Leesville, Louisiana, or from the Tangipahoa River at Percy Quin State Park, Mississippi. We collected the unionid Carunculina texasensis (Lea) from local ponds in the Baton Rouge area. Animals were returned to the laboratory and maintained unfed in artificial pondwater (APW, in mmol l⁻¹: NaCl, 0.5; NaHCO₃, 0.2; KCl, 0.05; CaCl₂, 0.4; Dietz and Branton, 1975) at ambient temperature (22–24 °C) for at least a week prior to use.

We measured blood acid–base variables and ion composition in bivalves following incubation in a series of ion-restricted media, with or without the addition of serotonin. In addition, we measured the components of ion flux (net flux, influx and efflux) in these bathing media and in the presence or absence of 5-HT.

Specimens of Corbicula fluminea and Carunculina texasensis (N=5–10 per treatment category) were incubated for 3 h in 50 ml of one of the following eight media: artificial pondwater containing 6×10³–14×10³ cts min⁻¹ ml^{−1 22}Na or ³⁶Cl ± 0.1 mmol l⁻¹ serotonin; 1 mmol l⁻¹ choline chloride (Na⁺-free medium) + ³⁶Cl ± 0.1 mmol l⁻¹ serotonin; 0.5 mmol l⁻¹ Na₂SO₄ (Cl⁻-free medium) + ²²Na ± 0.1 mmol l⁻¹ serotonin. Bath samples (5.0 ml) were taken at time zero and after the 3 h incubation. Animals were removed, and the soft tissues were dried (90 °C; 24 h) and weighed. The radioactivity of bath samples was measured on 0.5 ml subsamples using liquid scintillation methods (Weigman et al. 1975). Bathing medium [Na⁺] was measured by flame photometry and [Cl⁻] by electrometric titration. Net ion flux was determined from the change in [Na⁺] or [Cl⁻] in the bathing medium and was expressed as μmol g⁻¹ dry mass h⁻¹. Ion influx (J_i) was determined by the loss of isotope from the bathing medium using standard methods (Graves and Dietz, 1982), and ion efflux (J_o) was calculated as the algebraic difference between influx and net flux (J_net).

All data are expressed as means ± S.E.M.; significance was accepted at P<0.05. Overall differences between means were determined by single-factor analysis of variance (ANOVA) followed by Fisher’s protected least significant difference (PLSD) to determine differences between specific means. Data from individual species were analyzed separately. Analyses were performed following Statview (Abacus Systems) procedures.

The acid–base responses of both species to serotonin treatment and treatment with ion-restrictive media are summarized in a pH–HCO₃⁻ (Davenport) diagram (Fig. 1). The blood of pondwater-acclimated Corbicula fluminea is characterized as having a relatively low pH of approximately 7.9 and a low HCO₃⁻ concentration, whereas the blood of pondwater-acclimated Carunculina texasensis, in common with that of most other unionid species, has a higher pH and a higher bicarbonate content (Fig. 1).

Application of exogenous serotonin to pondwater-acclimated Corbicula fluminea had no effect on blood acid–base variables or ion composition (Table 1; Fig. 1). However, animals incubated in Cl⁻-free medium developed a significant base excess demonstrated by an increase in blood [HCO₃⁻] with no significant change in blood pH or (Table 1; Fig. 1). Application of serotonin to these animals resulted in a further increase in base excess, this time with a concomitant increase in ⁠, but with no change in blood pH (Table 1; Fig. 1). Serotonin stimulation of clams bathed in Cl⁻-free medium did not affect blood [Na⁺], but blood Cl⁻ levels did decline significantly compared with values for animals bathed in Na⁺-free medium (Table 1). When specimens of C. fluminea were incubated in a Na⁺-free medium, a base deficit occurred relative to pondwater-acclimated controls, as HCO₃⁻ content fell, but with non-significant changes in or in blood pH (Table 1; Fig. 1).

Treatment of these animals with serotonin resulted in a significant decrease in blood pH with no further change in [HCO₃⁻], but with a 25 % increase in mean blood that was not statistically significant (Table 1; Fig. 1). Blood Cl⁻ levels increased to 30 mmol l⁻¹, whereas blood Na⁺ levels were lower than those of pondwater-bathed animals.

The effects of serotonin and the compounding effects of the ion-restricted medium are more apparent on blood acid–base variables and ion composition in the unionid Carunculina texasensis (Table 2; Fig. 1). Blood pH and HCO₃⁻ levels for pondwater-acclimated C. texasensis are elevated significantly over those of C. fluminea (Fig. 1). Blood pH was significantly elevated in pondwater-acclimated C. texasensis when stimulated with serotonin (Table 2; Fig. 1). There was a significant serotonin-induced base excess, with blood HCO₃⁻ content increasing and a significant reduction in blood ⁠, resulting in a combined respiratory and non-respiratory alkalosis (Table 2; Fig. 1). Although blood [Na⁺] was unaltered, blood Cl⁻ levels were significantly reduced following serotonin treatment, and there was an equivalent increase in [HCO₃⁻] (Table 2). Incubation in Cl⁻-free medium resulted in an elevation in base excess compared with that of pondwater-incubated controls similar to that noted with C. fluminea. There was no change in blood pH, [Na⁺] or [Cl⁻], but increased (Table 2; Fig. 1). When stimulated with serotonin, there was a significant alkalosis in these animals due entirely to a reduction in blood with no change in [HCO₃⁻] (Table 2; Fig. 1). Blood [Na⁺] remained at pondwater-acclimated levels but blood [Cl⁻] declined significantly (Table 2). In the absence of external Na⁺, blood pH was not significantly altered compared with that of pondwater-incubated controls; none of the other variables measured was altered (Table 2). Application of serotonin to animals incubated in a Na⁺-free medium resulted in no significant change in blood pH compared with Na⁺-free controls, but blood pH was significantly elevated over that of pondwater-acclimated animals (Table 2; Fig. 1).

Serotonin stimulated Na⁺ influx in both C. fluminea and C. texasensis (Table 3). In pondwater, both species were essentially in steady state with regard to Na⁺, having net fluxes of approximately zero. There were no differences between fluxes (J_net, J_i and J_o) in control animals in pondwater and in Cl⁻-free medium. Serotonin stimulated Na⁺ influx by more than twofold for C. fluminea and by approximately fivefold for C. texasensis when bathed in pondwater (Table 3). Efflux was not affected by either bathing medium or serotonin stimulation in C. texasensis. Pondwater-bathed C. fluminea displayed no alteration in efflux in response to serotonin stimulation. However, when C. fluminea were stimulated with serotonin in a Cl⁻-free medium, Na⁺ efflux and Na⁺ influx increased. Examination of fluxes of individual animals demonstrated that those with high influxes also had the highest effluxes (data not shown), suggesting high turnover rates in these individuals.

Specimens of C. fluminea utilized as controls in this study were experiencing net Cl⁻ losses both in pondwater and in Na⁺-free bathing medium, whereas C. texasensis were in approximate steady state (Table 4). In both species, serotonin stimulation resulted in an increase in Cl⁻ influx in both bathing media. In C. fluminea, there was an approximate doubling of the influx, resulting in positive net Cl⁻ fluxes, while efflux was not significantly altered. There was no effect of Na⁺ absence on any of the fluxes in C. fluminea (Table 4). In C. texasensis, serotonin stimulated Cl⁻ influx to a similar extent in both pondwater and Na⁺-free medium (Table 4). Unlike C. fluminea, incubation of C. texasensis in Na⁺-free medium resulted in a reduction in Cl⁻ efflux to approximately half the rate observed in pondwater-incubated controls. The effluxes were not further affected by serotonin application.

Summary diagrams for both C. fluminea (Fig. 2) and C. texasensis (Fig. 3) highlight changes in components of acid flux and blood acid–base balance. The net acid flux can be expressed as the algebraic difference between the net flux of Cl⁻ and the net flux of Na⁺ (McDonald et al. 1989). Pondwater-acclimated C. fluminea were in net acid loss because Cl⁻ net flux was lower than Na⁺ net flux (Fig. 2). Serotonin stimulation did not alter blood pH or the net loss of acid equivalents, despite net increases in Na⁺ and Cl⁻ flux (Fig. 2). In pondwater, Na⁺ influx is approximately twice Cl⁻ influx in C. fluminea, and serotonin stimulation doubles the influx of both Na⁺ and Cl⁻, thereby maintaining the relative stoichiometry (Tables 3, 4). The elevated ion fluxes could result in an increase in acid secretion with consequent changes in blood variables. However, the relative difference between the net Na⁺ and Cl⁻ fluxes was maintained upon serotonin stimulation, resulting in no net change in acid secretion (Fig. 2).

The responses of these species to ion-deficient media accentuate the significance of the differences both in ion-transporting properties and in the ionic composition of the blood. In the absence of external Cl⁻, there were offsetting increases in base excess and in C. fluminea (Fig. 2). Stimulation by serotonin resulted in a large increase in Na⁺ net flux and caused an equally large increase in net acid equivalent loss. Blood pH was not altered, despite further increases in base excess, because of a compensatory increase in ⁠. In Na⁺-free controls, there was no net movement of acid equivalents because Na⁺ and Cl⁻ net fluxes were similar (Fig. 2). Base excess was negative (=base deficit), but was lower than in pondwater-acclimated animals, resulting in no significant change in blood pH. Serotonin stimulated Cl⁻ transport and, as net Na⁺ flux was unaltered in the absence of exogenous Na⁺, net acid flux was positive (the animals retained protons). The base deficit was not changed, but increased, causing a significant decline in blood pH (Fig. 2).

Carunculina texasensis was in a steady state in pondwater with acid flux at around zero (Fig. 3). There was a significant net loss of acid equivalents when these animals were stimulated with serotonin, with a concurrent increase in base excess and a decline in blood combining to cause a rise in blood pH (Fig. 3). In a Cl⁻-free medium, there was a slightly positive net acid flux (proton retention), but a large base excess was offset by an increase in blood ⁠. Serotonin stimulation significantly increased net Na⁺ flux in this medium, resulting in a loss of acid equivalents, base excess was not altered significantly and blood declined, the end result being an increase in blood pH. Stimulation of Cl⁻ flux in C. texasensis in a Na⁺-free medium did not result in a significant change in acid uptake (proton retention); base excess was not affected, nor was blood ⁠, resulting in no significant change in blood pH (Fig. 3). Thus, in C. texasensis, serotonin affects Na⁺ influx to a significantly greater extent than it affects Cl⁻ influx. Consequently, the differences in net Na⁺ and Cl⁻ flux have a significantly greater effect on acid secretion than that observed in C. fluminea.

The interrelationships between Na⁺/Cl⁻ transport and acid–base balance in these two species of freshwater bivalve are dependent on the relative activities of the transporters and the overall blood ion composition of each species. Corbicula fluminea, in common with other freshwater bivalve species with relatively short histories in fresh water (Keen and Casey, 1969), has a blood composition relying on Cl⁻ as the major anion (Dietz, 1979). Consequently, rates of Cl⁻ transport in C. fluminea are high and are apparently under neural control as shown by serotonin stimulation of Cl⁻ transport in this species (this study; Zheng, 1996). In contrast, Carunculina texasensis conforms to the unionid model of having approximately equal amounts of Cl⁻ and HCO₃⁻ in the blood (Dietz, 1979). In the present study, we report the first instance of effective stimulation of Cl⁻ influx by exogenously applied serotonin in a unionid species, but the level of stimulation is significantly lower than that of Na⁺ stimulation and lower than that observed in C. fluminea.

The disturbances in blood acid–base state in these species both in ion-deficient media and in response to serotonin stimulation are consistent with the actions of largely independent Na⁺/H⁺ and Cl⁻/HCO₃⁻ exchangers in the gill epithelia. Blood pH in C. texasensis was alterable to a greater extent than that of C. fluminea. The change in pH was primarily due to the greater activity of the Na⁺/H⁺ exchanger relative to the Cl⁻/HCO₃⁻ exchanger in C. texasensis so that, under conditions of serotonin stimulation, acid secretion increased. Also, a higher blood pH is subject to greater perturbations by addition/deletion of acid equivalents than a lower blood pH, given similar buffering environments. This is a consequence of measuring pH on a logarithmic scale. Although the HCO3− buffering capacity increases with increasing HCO₃⁻ content, changes in have a greater influence at higher pH. In both species, incubation in a Cl⁻-free medium resulted in retention of HCO₃⁻ in the blood as a result of a reduced ability to exchange HCO₃⁻ with the environment and continued Na⁺/H⁺ exchange, with H⁺ excretion being equivalent to HCO₃⁻ retention.

For the unionid, maintenance of ionic homeostasis requires greater reliance on active Na⁺ uptake than on uptake of Cl⁻. The anionic component of blood consists of a combination of Cl⁻ and HCO₃⁻, with substitution of one for the other over a substantial concentration range (Scheide and Dietz, 1982). As HCO₃⁻ can be mobilized or removed relatively easily by metabolic and respiratory means, the requirement for a tightly regulated Cl⁻ transport system, controlled by hormonal and/or neural means, is much less important. In contrast, C. fluminea must regulate Cl⁻ uptake in order to maintain blood Cl⁻ at a high and relatively constant level. Although blood HCO₃⁻ levels increase and decrease in response to treatment, C. fluminea has a much less well-developed capacity to substitute Cl⁻ and HCO₃⁻ in the blood.

It is uncertain whether bivalve molluscs actively regulate blood pH. Acid–base responses to changes in environmental gases result in predictable and large changes in blood pH in the unionids Elliptio complanata (Byrne et al. 1995) and Anodonta grandis simpsoniana (Byrne and McMahon, 1990). Although some compensatory mechanisms to alleviate perturbations in blood pH have been described for these species, the acid–base responses are generally regulated only within wide limits. Booth et al. (1984) concluded that the marine bivalve Mytilus edulis had a limited ability to regulate blood pH and that maintenance of blood pH within narrow limits may be unnecessary. This view was supported by a related study (Walsh et al. 1984) in which it was demonstrated that intracellular pH was not strictly dependent on extracellular pH in this species. The gas-carrying characteristics of respiratory pigments are functionally dependent on blood pH; thus, the absence of such pigments from the blood of most marine and all freshwater bivalve species would remove a major reason for maintenance of close extracellular pH regulation (Booth et al. 1984).

However, shell-forming animals, such as bivalve molluscs, must maintain an overall acid–base balance because the deposition of shell (CaCO₃) is accompanied by the release of acid equivalents (Wilbur, 1983). Therefore, if a bivalve is growing a shell, it must excrete protons in order to remove the accumulated proton load due to carbonate deposition. The time course of this pH regulation may encompass longer periods than are generally utilized in laboratory experiments. Indeed, shell deposition normally occurs in a distinctly discontinuous manner as shown by the presence of growth lines. Zheng (1996) demonstrated that C. fluminea maintains a constant ratio of Na⁺ and Cl⁻ net flux, the result of which is a constant rate of acid secretion over a wide range of Na⁺ and Cl⁻ net fluxes, suggesting that, over time, acid excretion is maintained as a function of the regulation of specific ions.

Thus, control of blood acid–base balance in the long term may have more to do with regulating shell deposition/reabsorption and of maintaining Ca²⁺ and HCO3− levels than with maintaining short-term extracellular homeostasis. Shell formation is influenced by environmental factors such as hypoxia (Wilbur, 1983) and pH (Machado et al. 1988). Calcium stores in unionid bivalve species are defended during environmental hypoxia by sequestration of blood Ca²⁺ released from the shell into calcium phosphate concretions located extracellularly in the gills (Silverman et al. 1983). These studies suggest that perturbations in blood acid–base balance and the associated changes in calcium dynamics affect the process of shell formation occurring in the separate extrapallial compartment. Coimbra et al. (1993) postulate that the dynamics of shell deposition are ultimately under the control of [Ca²⁺] in the blood and of the transepithelial electrical potential difference across the outer mantle epithelium. These factors exert a controlling effect on the direction of Ca²⁺ movement into or out of the extrapallial compartment (Coimbra et al. 1993). This is well illustrated when bivalves are exposed to air, which results in a respiratory acidosis because the pathway for metabolic CO₂ exchange with the environment is severely compromised. When C. fluminea is exposed to air, Ca²⁺ and HCO₃⁻ are mobilized from the shell and appear in the blood (Byrne et al. 1991). The respiratory acidosis causes alterations in the Ca²⁺ and HCO₃⁻ dynamics between the blood/extrapallial compartments, resulting in shell reabsorption. The effect on blood acid–base balance is to partially alleviate the acidosis by providing a large base excess of HCO₃⁻ (Byrne et al. 1991). In addition, blood could rise to high levels without resulting in severe acidosis; this provides a sufficient gradient for gas loss to the environment, thereby achieving a steady-state acid–base balance (Byrne et al. 1991; Byrne and McMahon, 1994). The blood acid–base responses to emersion were similar in the unionid Anodonta grandis simpsoniana (Byrne and McMahon, 1991), suggesting that broadly similar processes were occurring in both species and that the importance of the regulation of blood acid–base balance is to provide a suitable environment to control the deposition and reabsorption of shell calcium.

The two species of freshwater bivalve utilized in this study represent lineages with sharply different histories in fresh water. Unionids have inhabited fresh water since the Cambrian, whereas corbiculids invaded fresh water in the relatively recent geological past (Triassic) (Keen and Casey, 1969). Other bivalve groups with a recent history in fresh water are some members of the Dreissenidae, represented in North America by Dreissena polymorpha. D. polymorpha conforms to the model that recent invaders of fresh water retain Cl⁻ as the major anion and have a relatively lower level of HCO₃⁻ in the blood (Horohov et al. 1992). This species also possesses Na⁺ and Cl⁻ transporters capable of high rates of transport; both these transporters are stimulated by serotonin (Horohov et al. 1992). Pondwater-acclimated bivalves having a greater reliance on Cl⁻ as the major anion have a comparatively lower blood pH. Blood with a low level of HCO₃⁻ may be the more archetypal condition for bivalves coming into fresh water, a position supported by the low levels of HCO₃⁻ found in the blood of marine bivalve species and in D. polymorpha (Booth et al. 1984; Horohov et al. 1992). The significance to those unionids that shift to a blood with higher [HCO₃⁻] is open to speculation. It is possible that, because all ions tend to be at low concentration in fresh water, reducing the dependence on one of these ions (Cl⁻) and replacing it with an ion that can be endogenously produced from metabolism (HCO₃⁻) would be an adaptive strategy. The relationship between blood acid–base balance and the dynamics of shell formation seems to be one of shifting equilibria so that a change in blood pH brought on by retaining HCO₃⁻ would need only to require a compensatory shift in the ionic components of the extrapallial fluid compartment in order to achieve a new equilibrium. The maintenance of freshwater bivalve blood pH and acid–base balance may be important to this interrelationship.

This research was supported, in part, by an NSF Research Opportunity Award supplementing NSF grant DCB90-17461.