Posterior isolated gills of Neohelice (Chasmagnathus) granulatus were symmetrically perfused with hemolymph-like saline of varying [HCO3−] and pH. Elevating [HCO3−] in the saline from 2.5 to 12.5 mmol/l (pH 7.75 in both cases) induced a significant increase in the transepithelial potential difference (Vte), a measure of ion transport. The elevation in [HCO3−] also induced a switch from acid secretion (−43.7 ± 22.5 μequiv·kg−1·h−1) in controls to base secretion (84.7 ± 14.4 μequiv·kg−1·h−1). The HCO3−-induced Vte increase was inhibited by basolateral acetazolamide (200 μmol/l), amiloride (1 mmol/l), and ouabain (5 mmol/l) but not by bafilomycin (100 nmol/l). The Vte response to HCO3− did not take place in Cl−-free conditions; however, it was unaffected by apical SITS (2 mmol/l) or DIDS (1 mmol/l). A decrease in pH from 7.75 to 7.45 pH units in the perfusate also induced a significant increase in Vte, which was matched by a net increase in acid secretion of 67.8 ± 18.4 μequiv kg−1 h−1. This stimulation was sensitive to basolateral acetazolamide, bafilomycin, DIDS, and Na+-free conditions, but it still took place in Cl−-free saline. Therefore, the cellular response to low pH is different from the HCO3−-stimulated response. We also report V-H+-ATPase- and Na+-K+-ATPase-like immunoreactivity in gill sections for the first time in this crab. Our results suggest that carbonic anhydrase (CA), basolateral Na+/H+ exchangers and Na+-K+-ATPase and apical anion exchangers participate in the HCO3−-stimulated response, while CA, apical V-H+-ATPase and basolateral HCO3−-dependent cotransporters mediate the response to low pH.
- acid/base regulation
- HCO3− secretion
- carbonic anhydrase
- electrogenic Na+/H+ exchanger
in crustaceans, the gill is the main organ involved in acid/base (A/B) exchange with the environment, and thus it is probably the main A/B regulatory organ (47). The overall branchial mechanism for A/B regulation in crabs is similar to other aquatic animals: apical Na+/H+ and Cl−/HCO3− exchanges (3, 6, 14, 15) energized by ATPases (18). However, the only proteins involved in the gill A/B regulatory mechanism that have been identified so far in crabs are carbonic anhydrase (12, 13, 36) and Na+/K+-ATPase (36). Several other ion-transporting proteins have been molecularly and/or pharmacologically identified in crab gills, but to our knowledge, only their role in ion uptake and not A/B regulation has been directly tested (reviewed in Ref. 7).
A major advantage of the gills of certain crustaceans over other A/B regulatory organs (i.e., fish gill and mammalian kidney) is that they can be easily isolated from the animal and perfused with hemolymph-like saline using a peristaltic pump. This technique allows for the study of the ion transport physiology of the whole organ without the interference of hormonal or nervous factors. Isolated and perfused gills allow for accurate control of the composition of the perfusing and bathing salines, including the addition of specific inhibitors of ion-transporting proteins. These manipulations are impossible to perform in whole animal experiments.
The cellular mechanisms for ion uptake across the posterior gills of Neohelice granulata (formerly named Chasmagnathus granulatus) are one of the better studied among aquatic animals. The basal (nonstimulated) ion transport mechanism in gills from crabs acclimated to low-salinity water (2 ‰) includes basolateral Na+-K+-ATPase, and Cl− and K+ channels (24, 25, 30). In this condition, the apical routes of entry for Cl− and Na+ are apical Na+-K+-2Cl− (NKCC) cotransporters in parallel with K+ channels (25, 30). This basal state of ion uptake can be activated by hormonal and nonhormonal factors. For example, dopamine can activate Na+-K+-ATPase via a D1-like receptor-Gs protein-cAMP-PKA pathway (9, 11). A reduction in the osmolarity of the hemolymph-side of the isolated gill also stimulates ion uptake (24, 39), an effect that is partially mediated by cAMP and Na+-K+-ATPase (39). In addition, CA and apical V-H+-ATPases and Cl−/HCO3− exchangers (CBEs) are also recruited during these stimulated conditions (8). Apical V-H+-ATPase and CBEs working in concert could facilitate Cl− uptake from a dilute external medium (8), but the combined effect of both transporters is neutral in terms of net A/B transport. Interestingly, the branchial cellular mechanisms for ion transport in N. granulata from different salinities resemble those from different segments of the mammalian nephron. This is likely related to similar ionic gradients encountered at the apical membrane in each system.
In the current work, we investigated the involvement of the gills of N. granulata in A/B-relevant ion transport. We used the advantages of the isolated gill preparation to test the hypothesis that ion transport can be directly stimulated by [HCO3−] and pH. Our results indicate that ion transport is stimulated by high [HCO3−] and by low pH and that the specific mechanisms are different in each condition.
MATERIALS AND METHODS
N. granulata (Dana 1851) were collected by hand from a muddy beach at San Antonio Oeste (Rio Negro, Argentina) during the Austral summer of 2006–2007 and spring of 2007. The animals were transported to the Laboratory of Aquatic Ecotoxicology Centro Internacional de Educación para el Desarrollo (CIEDE, San Martin de los Andes, Neuquén, Argentina), where the gill perfusion experiments were performed. Crabs were acclimated in plastic containers with aerated seawater of 2‰ salinity for at least 1 wk. Water temperature was kept at 18 ± 2°C. The animals were fed twice a week with commercially available pellets of trout food. All procedures followed the Canadian Council for Animal care procedures. Stage C intermolt adult male crabs (4) were selected for the study.
Transepithelial potential difference (Vte) across isolated gills.
Crabs were killed by destroying the ventral ganglia using a pair of scissors. The carapace was then removed, and the posterior gills no. 6 and 7 were dissected and placed in a petri dish containing saline. Preliminary experiments had demonstrated that gills no. 6 and 7 responded in identical manner to the manipulations described in this paper (also see Ref. 24). The afferent and efferent vessels were connected by polyethylene tubing (0.4 mm in diameter) to a peristaltic pump (afferent) and to a collecting tube (efferent). Tubing was held in place by a sponge-coated acrylic clamp. The preparation was placed into a glass beaker with the appropriate solution, which was constantly aerated. The perfusion rate was kept at 0.1 ml/min. The salt composition and pH of the perfusate and bath were identical at all times to avoid the passive movement of ions by diffusion. Therefore, the only asymmetries between the internal and external media were those created by the gill itself (Vte, pH).
Transepithelial potential difference (Vte) was measured via Ag+/AgCl electrodes connected by agar bridges (3% agar in 3 mol/l KCl) to the bath (external side) and to the collecting tube (internal side). Vte was measured with two chart recorders equipped with mV-meters (Cole Palmer 8373–20, Chicago, IL; and Kipp & Zonen BD 40, Delft, Holland). Vte is given as the difference in electrical potential between the external and internal medium. For the statistical analyses, we considered the Vte values after stabilization for at least 30 min in each treatment and controls.
Apparent acid and base secretion across isolated gills.
Gills were perfused and bathed in the same saline as explained above. Apparent H+ (JH+) and HCO3− (JHCO3−) secretion were calculated as explained by Siebers et al. (36). Perfusate was collected after passing through the gills, and its pH was measured and compared against the pH of the preperfusion saline. To obtain enough volume for the measurement, perfusate was collected for 1 h under control and experimental (low pH, high [HCO3−]) conditions. The buffering capacity of the solutions was calculated by titration of 20 ml of perfusion saline with 0.1 N HCl. Immediately after the experiment, the gill was dried by gently pressing with a paper towel and its fresh weight (fw) was measured. JH+ and JHCO3− were calculated using: J (μmol·g−1·h−1) = [HCl]·VHCl·ΔpHgill·Vcollected. fw−1·Voltitr−1ΔpHtitr−1, where [HCl] is 0.1 N, VHCl is the volume of HCl needed to reach the final pH, ΔpHgill is the change in the perfusate pH after passing through the gills, Vcollected is the volume of perfusate collected after passing through the gill during 1 h, Voltitr is the volume of saline used for titration (20 ml), fw is the fresh weight of the gill, and ΔpHtitr is the change in pH in the medium used for titration. A positive value represents JHCO3−, and a negative value is indicative of JH+.
Solutions and reagents.
The composition of the control saline was (in mmol/l): 465 NaCl, 9.40 KCl, 7.50 MgCl2, 12.40 CaCl2, 2.50 NaHCO3 and 5.00 HEPES; pH to 7.75 with Tris. In addition, the perfusate also contained 2.00 mmol/l glucose. This saline was used because it has been shown to induce a basal state in ion uptake estimated from Vte, Isc, and 22Na+ radiotracer experiments (24, 30, 39). Combined with the available information on cellular ion transport mechanisms in gills from crabs acclimated to low salinity (8, 24, 25, 30, 39), this facilitated the interpretation of the responses to high [HCO3−] and low pH.
The “high [HCO3−]” saline had 10 mmol/l NaCl substituted with NaHCO3, which resulted in a final nominal [HCO3−] of 12.50 mmol/l, but with osmolarity similar to the control. The actual [HCO3−] was calculated from total CO2 readings as described in Ref. 43 and was in the ranges of 2.3–2.5 mmol/l for control and 11.50–13.50 mmol/l for the high [HCO3−]. The extra HCO3− elevated the pH of the saline to 7.81. In some experiments, the pH was adjusted back to 7.75 with HEPES (“high [HCO3−] pH 7.75”). An additional saline contained control [HCO3−], but its pH was increased to 7.81 with Tris-base. The “low pH” saline was identical in composition to the control saline, but pH was adjusted to 7.45 with HEPES. [HCO3−] in this saline was negligible. We changed the pH of the solutions by adding Tris and HEPES instead of using strong acid or base to avoid changing the strong ion concentrations. The strong ion difference has been reported to affect A/B regulation in crabs (Refs. 1 and 21, among others). The nominal Cl− free solutions were prepared by substituting the Cl− salts by the corresponding nitrates, while the nominal Na+-free solution had choline chloride instead of NaCl.
Acetazolamide, amiloride, bafilomycin, ouabain, DIDS, SITS, and phenamil were purchased from Sigma (St. Louis, MO). Stock solutions of all of these inhibitors were prepared in DMSO and were added at a final dilution of 0.1% to the bath (SITS and phenamil) or perfusate (the rest), with the exception of oaubain, which was directly dissolved in the perfusion saline. The effects of the drugs were only tested on Vte because Vte measurements allow for real-time monitoring and therefore for an accurate estimation of the stable values. The addition of 0.1% DMSO alone to the control saline did not result in any significant effect on Vte. The effect of every drug during control [HCO3−] and pH conditions was also tested, and neither of them had any significant effect on the basal Vte (also see 8, 23, 29), with the exception of ouabain, which inhibits the basal ion transport activity.
Posterior gill no. 6 was excised from crabs acclimated to 2‰ salinity and processed for immunohistochemistry as described in Ref. 40. The primary antibodies used included an anti-V-H+-ATPase antibody designed against a synthetic peptide based on the highly conserved and hydrophilic region in the A-subunit (17) and an anti-Na+-K+-ATPase antibody against a synthetic peptide corresponding to a part of a highly conserved region of the α-subunit (16). Secondary antibody incubation and signal development were performed using the Vectastain ABC kit (Vector Laboratories, Burlingame, CA), according to the manufacturer's directions. The specificity of the antibodies in crab tissues was analyzed by Western blotting (7.5% gels), using a donkey anti-rabbit fluorescent secondary antibody (Li-Cor, Lincoln, NE) (39).
All data are given as means ± SE. Differences between groups were tested using one-way repeated measures ANOVA (1-way repeated-measures ANOVA), followed by Dunnet's post test or Tukey's multiple-comparisons test. Certain data sets were analyzed using paired Student's t-test. Statistical significance was set at P < 0.05. All statistical analyses were performed on GraphPad Prism v. 3.0 (GraphPad Software, San Diego CA).
Elevation of Vte and ΔpH by HCO3− in isolated perfused gills.
Switching from control [HCO3−] saline of 2.50 mmol/l to the “high [HCO3−] pH 7.81” (12.50 mmol/l HCO3−) saline caused an immediate and significant elevation in Vte from 3.02 ± 0.47 to 4.83 ± 0.88 mV (n = 6, P < 0.05). This effect was fully reversible, since Vte returned to its basal value after reintroducing the original control saline (3.15 ± 0.49 mV). A subsequent application of a saline with control [HCO3−] but with a higher pH of 7.81 did not have any significant effect on Vte (2.83 ± 0.45 mV, n = 6, P > 0.05). However, the saline with high [HCO3−] and a control pH of 7.75 induced a significant increase on Vte, of a similar magnitude to the “high [HCO3−] pH 7.81” saline (5.17 ± 0.88 mV, n = 6, P < 0.05). These results indicate that an increase in the saline [HCO3−], and not pH, is the stimulus that ultimately causes the elevations on Vte. Consequently, only the “high [HCO3−] pH 7.75” saline was used for the pharmacological characterization. These results are shown in Fig. 1, A and B.
Gills perfused with the control saline secreted an apparent H+ flux (JH+) of −43.72 ± 22.5 μequiv·kg−1·h−1. Perfusion with “high [HCO3−] pH 7.81” saline induced the reversion in secretion to an apparent HCO3− flux (JHCO3−) of 84.7 ± 14.4 μequiv·kg−1·h−1. Therefore, the net change (Jctrol − Jbicarb) in JHCO3− was of 128.5 ± 31.6 μequiv·kg−1·h−1 (P < 0.05; n = 5) (Fig. 1C). This indicates that the elevations in Vte are accompanied by net base efflux across the gill epithelium.
To test whether carbonic anhydrase (CA) is involved in the Vte response to high [HCO3−], we added 200 μmol/l acetazolamide into the high [HCO3−] perfusate. Acetazolamide completely and reversibly abolished the increased Vte (Fig. 2, A and B). Bafilomycin is a specific inhibitor of V-H+-ATPases at nanomolar concentrations (5). Basolateral application of bafilomycin (100 nmol/l) did not exert any significant effect on the HCO3−-stimulated Vte, suggesting that V-H+-ATPase is not important in this process (Fig. 2, A and C). On the other hand, ouabain (5 mmol/l) almost completely blocked Vte both during control (not shown) and high [HCO3−] conditions (Fig. 2D), demonstrating that Na+-K+-ATPase is the major driving force for the transepithelial transport of ions. To identify the basolateral route of exit of H+ from the cells into the hemolymph space, we tested the effect of basolateral amiloride (1 mmol/l). This treatment completely and reversibly blocked the HCO3−-stimulated Vte, suggesting that basolateral Na+/H+ exchangers (NHE) are critical for the overall transepithelial transport mechanism activated by high [HCO3−] (Fig. 3, A and B). Importantly, amiloride did not have any effect on the basal Vte, and it completely but reversibly blocked the HCO3−- induced response when added prior to elevating [HCO3−] (Fig. 3C). This indicates that the putative NHE are specifically recruited for the base secreting mechanism. This finding also shows that the concentration of amiloride used does not significantly affect Na+-K+-ATPase function.
Our last two experimental series were designed to test for the involvement of apical Cl−/HCO3− exchangers. Introduction of Cl−-free conditions produced a significant decrease in Vte, which reflects the importance of Cl− for the basal ion transport mechanism (see 24, 30, 40). An increase of [HCO3−] under Cl−-free conditions did not result in the typical stimulation of Vte, indicating that Cl− ions must be present for the HCO3− stimulation to occur (Fig. 4, A and B). Returning to normal Cl−-containing conditions did not restore the original Vte, probably because the unnatural Cl−-free saline induced a new basal steady state of ion transport. However, the gill epithelium was able to respond with the typical increase in Vte to an elevation in [HCO3−], indicating that it was still healthy and functional (Fig. 4, A and B). Finally, we tested the effect of apical DIDS (1 mmol/l) on the Vte stimulated by HCO3−, but this treatment did not result in any significant changes in Vte (Fig. 4C). Apical SITS (2 mmol/l) was also without any significant effect on Vte (not shown).
Stimulation of Vte by low pH in isolated perfused gills.
A reduction in the pH from 7.75 to 7.45 pH units in the perfusion conditions induced an immediate and significant stimulation of Vte from 3.68 ± 0.58 to 6.32 ± 0.72 mV (n = 13). The original Vte was restored upon reapplication of the saline with pH of 7.75. The low pH-stimulated Vte and subsequent wash-out was a repeatable event (Fig. 5A), and it was accompanied by an increase in net acid secretion from −79.1 ± 18.5 to −146.9 ± 19.2 μequiv·kg−1·h−1 (net increase of −67.8 ± 18.4 μequiv·kg−1·h−1, P < 0.05; n = 5) (Fig. 5B). To test whether CA is involved in the low pH-stimulating mechanism, we added 200 μmol/l acetazolamide into the 7.45 pH perfusate of isolated gills that had already been stimulated by low pH. Acetazolamide completely and reversibly abolished the stimulated Vte (Fig. 6). We also investigated the involvement of V-H+-ATPase. Basolateral bafilomycin (100 nmol/l) inhibited the low pH-stimulated Vte, but, unlike acetazolamide, its effect was only slightly reversible (Fig. 7). However, the original resting Vte was restored upon reintroduction of the control pH 7.75 saline, indicating that the inhibitory effect of bafilomycin specifically acts on the mechanism stimulated by low pH. Next, we tested the involvement of apical Na+ channels and basolateral Na+/HCO3− cotransporters by adding apical phenamil (50 μmmol/l) and basolateral DIDS (1 mmol/l), respectively, under low-pH stimulating conditions. An inhibitory effect of phenamil was seen in a 3 out of 5 preparations, but its magnitude was highly variable, and thus it is not shown or analyzed. On the other hand, basolateral DIDS totally inhibited the elevated Vte, although it first produced an important (∼10 mV) hyperpolarization of Vte in most of the preparations (Fig. 8). The inhibitory effect of DIDS was not reversible. DIDS did not cause any significant changes when applied under control pH conditions (Fig. 8C), indicating that both the initial hyperpolarization and the inhibitory effect observed are related to the low pH-stimulating mechanism.
Lastly, we tested whether the low-pH stimulated Vte depends on the transepithelial movement of Na+ and Cl− by using solutions with reduced concentrations of these ions. As shown in Fig. 9, these conditions abolished and in some cases even reversed Vte. Introduction of Na+-free pH 7.45 saline did not elicit an increase in Vte, demonstrating that the transepithelial movement of Na+ (apical channel and basolateral NBC?) is essential for the low pH activation of Vte (Fig. 9, A and B). Conversely, a reduction in saline pH to 7.45 during Cl−-free conditions still induced a significant increase of Vte of a magnitude comparable to that in normal, Cl−-containing, saline. This effect was only partially reversible (Fig. 9, C and D).
V-H+-ATPase and Na+/K+-ATPase immunolocalization.
The anti-V-H+-ATPase and anti-Na+-K+-ATPase heterologous antibodies cross-reacted with bands of the appropriate size during Western blot analysis (∼75 and 110 kDa, respectively, Fig. 10). No bands were detected when the primary antibodies were omitted (not shown).
The antibodies resulted in slightly different immunolabeling patterns on gill sections. Both V-H+-ATPase and Na+-K+-ATPase immunolabeling were present throughout the pillar and principal cells, although Na+-K+-ATPase labeling was much stronger (Fig. 11, A and B). However, high-magnification micrographs revealed that V-H+-ATPase is also present in the apical region of some cells (Fig. 11C). No signal was found when the primary antibodies were omitted (not shown).
Our results demonstrate that ion transport across posterior gills of N. granulata is stimulated by an increase in [HCO3−] and also by a decrease in pH. This suggests that the posterior gills are important in detecting and correcting A/B disturbances in the hemolymph of the whole animal.
Gills from crabs acclimated to 2‰ salinity maintain a basal ion uptake activity when perfused with the control saline used in the current study, as estimated from Vte, short-circuit current and 22Na+ uptake (24, 30, 39). Increasing [HCO3−] by 10 mmol/l elevated both Vte and base secretion to the apical medium. An equivalent increase in pH alone did not have any effect on Vte. This indicates that an electrogenic transepithelial base secretion mechanism is activated directly by [HCO3−]. The pharmacology profile suggests that the mechanism activated by high [HCO3−] is different from those described in our previous studies (see below). Our Vte measurements are not without certain limitations. In particular, Vte is a reflection of both current (I) and transcellular and paracellular resistance (R). Therefore, it is possible that the changes observed were produced by changes in R. However, our saline manipulations were always performed at constant osmotic pressure, and thus it is unlikely that paracellular R changes in the dramatic fashion that is required to justify the observed changes in Vte. Similarly, transcellular R would have to increase substantially, which could only happen by a reduction in the conductance of the ion-transporting proteins involved (most likely at the apical membrane). However, this hypothesis would make it difficult to explain the pharmacological inhibition of the stimulated Vte by low pH and HCO3−. Therefore, we are confident that the Vte measurements are indeed a reasonable estimation of net transepithelial ion transport (I) under our experimental conditions. In addition, Vte measurements in open-circuit conditions have certain advantages over short-circuit conditions (i.e., inactivation of V-H+-ATPase, see Refs. 7 and 19).
Our results indicate that CA is essential for the gill responses to both high [HCO3−] and low pH. On the basis of current models for branchial A/B regulation in aquatic animals (33), it is likely that both extracellular and intracellular CA are involved in the response to elevated [HCO3−]. In fact, both types of CA are present in the posterior gills of C. granulatus (8). Extracellular CA would dehydrate H+ and HCO3− into CO2, which can diffuse inside the ion-transporting cells. Once inside, CO2 is probably rehydrated into H+ and HCO3− by intracellular CA (Fig. 12), the main intracellular substrates in the mechanisms stimulated by high [HCO3−] and low pH. Unfortunately, the drug used in our study does not allow us to differentiate between extracellular and intracellular CA, and further studies are necessary to confirm the involvement and specific roles of both CA isoforms. On the other hand, saline [HCO3−] was negligible in the low pH experiments. Therefore, the source of CO2 for the acid secretion mechanism is unclear. One possibility is that the HCO3− that is reabsorbed into the hemolymph combines with H+, generating CO2 that diffuses back into the cell. Yet another interesting option is derived from studies on fish intestinal epithelium, where mitochondria-produced CO2 is an important substrate for ion transport (reviewed in Ref. 10). The gill ionocytes of N. granulata possess large numbers of mitochondria (22, 23), which supports this model. Furthermore, low pH did not produce any change in gill Vte when we attempted the same protocol during the winter months. Although preliminary, these results may indicate that gills from winter crabs are metabolically less active than in summer crabs. Whether winter crabs conform to acidotic conditions or use different mechanisms for recovery is a fascinating topic that we are currently investigating.
Responses to high [HCO3−].
A priori, we were expecting that basolateral V-H+-ATPases energized the secretion of HCO3− as suggested for hagfish (42, 44), elasmobranchs (40, 43, 45), and teleost fish (2, 33, 41). However, the [HCO3−]-dependent Vte was insensitive to 100 nmol/l bafilomycin. Importantly, this was in contrast to the low pH-stimulating mechanism, indicating that the dose of bafilomycin used is effective in inhibiting V-H+-ATPase in isolated perfused gills. On the basis of the inhibition of the HCO3−-induced Vte by amiloride, we conclude that the basolateral route of exit for H+ in crab gills during HCO3− stimulation is via a member of the NHE family. Candidate targets for the amiloride sensitivity are some of the NHEs present in the gills of several crustaceans (18, 37, reviewed in 38). Importantly, at least some crustacean NHE isoforms are electrogenic and transport two Na+ for each H+ (18, 35). However, it is not clear whether the electrogenic NHE is located in the basolateral or apical membrane from those studies. Moreover, our study clearly demonstrates that the driving force for HCO3− secretion and H+ reabsorption is Na+-K+-ATPase, as seen in the inhibitory effect of ouabain.
Finally, the lack of HCO3− stimulation in Cl−-free conditions suggests the involvement of apical Cl−/HCO3− exchangers of some sort (Fig. 12). However, apical application of DIDS and SITS did not have any significant effect in the stimulated Vte. It is interesting to note that apical SITS produced a small but significant inhibition of 16% under hypo-osmotic stimulating conditions, and of 45% in similar, but Na+-free, perfusion saline (8). This was interpreted as indicative of the involvement of a Cl−/HCO3− exchanger in Cl−-uptake. Lack of SITS/DIDS inhibition in our experimental conditions indicates that either a different, DIDS/SITS-insensitive, Cl−/HCO3− exchanger participates in the HCO3−-induced Vte or that these drugs did not cross the cuticle in our experiments.
We have only found one similar study in the literature, performed in isolated perfused gills of the shore crab Carcinus maenas (36). This study concluded that the gill could detect an elevated pH of 8.10 in the perfusate (hemolymph space) and reabsorb H+ to restore a normal pH of ∼7.70. However, a saline with high pH but without HCO3− did not increase H+ reabsorption significantly, which suggests that the actual stimulus was [HCO3−] or at least that HCO3− was necessary to the base secretion mechanism. Siebers et al. (36) tested a variety of ion-transporting protein inhibitors, but only ouabain affected Vte and H+ reabsorption simultaneously. It is thus possible that C. maenas, unlike C. granulatus, relies on electroneutral ion transport for gill A/B regulation. Nonetheless, both isolated gill epithelia demonstrated an ability to detect and correct A/B disturbances, a feature that might be common to crustaceans and other aquatic organisms.
Low pH-stimulating mechanism.
The inhibition of the low pH-stimulated Vte by bafilomycin is a good indicator of the importance of V-H+-ATPase in this response. On the basis of the outside positive Vte, we propose that V-H+-ATPase is located in the apical membrane and acts to secrete the excess H+ into the water covering the gills. Although bafilomycin was applied to the basolateral space, it is a membrane-permeable compound (5) and thus it can inhibit V-H+-ATPase located at the apical membrane even if applied at the basolateral space.
Apical V-H+-ATPase has been proposed to energize apical Cl− absorption in some strong hyperregulating freshwater crabs (26, 27, 29, 34, 48), and also in C. granulatus (8, 24). However, the Cl− independence demonstrated in our study suggests that the low-pH stimulating mechanism is different from the Cl− uptake mechanism. On the basis of Na+-uptake models from certain aquatic organisms (for a review, see Ref. 19), we tested for the putative involvement of apical Na+ channels by applying apical phenamil on gills with stimulated Vte. Although phenamil did inhibit the stimulated Vte in certain preparations, the effect was not consistent, probably because of permeability issues at the apical cuticle (28). Our preliminary model presented here includes apical Na+ channels (Fig. 12), although further investigation is required to confirm this component.
We have recently reported the involvement of a basolateral electrogenic Na+/HCO3− cotransporter (NBC) as the way of exit of HCO3− and Na+ in isolated fish gill cells (31). In the current study, basolateral application of DIDS produced an initial further stimulation of the Vte already stimulated by low pH, followed by a rapid, complete, and irreversible inhibition. We tentatively propose that DIDS indeed inhibits basolateral NBCs and that the initial Vte stimulation is due to the pumping of protons by apical V-H+-ATPases, which is not totally compensated due to the reduction in the transcellular flux of Na+ as a result of the inhibition of its basolateral transport. Additionally, a transient increment in Cl− uptake through Cl−/ HCO3− exchange, favored by the activation of the V-H+-ATPases and by the intracellular accumulation of HCO3−, could be an alternative cause of this peak in Vte. In the longer term, the apical V-H+-ATPase probably shuts down because the HCO3− accumulation inside the cells reduces the availability of H+, and then Vte is inhibited. Lastly, DIDS penetrating into the cell and affecting transporters located at the apical membrane cannot be ruled out. In addition, the results from the Na+ and Cl− substitutions indicate that the low pH-induced increase in Vte depends on transepithelial Na+ transport. This leaves only the possibility of an NBC and/or Na+-K+-ATPase on the basolateral side. The NBC would be more likely as the HCO3− produced via CA hydration of CO2 would need to be transported across the basolateral surface to maintain the charge distribution, as the H+ is pumped out apically.
V-H+-ATPase and Na+-K+-ATPase immunolocalization.
Although V-H+-ATPase and Na+/K+-ATPase have been previously detected in C. granulatus by pharmacological, biochemical (8, 11, 24, 39), and molecular biology techniques (25), this is the first immunolocalization report of these transporters in this crab. Na+-K+-ATPase is present in both principal and pillar cells, being restricted to the basolateral area in both cell types. A basolateral localization is consistent with the literature (reviewed in Ref. 20) and with the role of Na+-K+-ATPase in energizing ion uptake in basal and stimulated conditions (8, 9, 24, 25, 30, 39). A basolateral Na+-K+-ATPase is also important for the high [HCO3−] and low pH stimulatory mechanisms reported in this study (Fig. 12).
The apical V-H+-ATPase localization in some pillar and principal cells is consistent with its role in acid secretion derived from our perfusion experiments. The V-H+-ATPase labeling found throughout the majority of cells could be due to V-H+-ATPase stored in vesicles. These vesicles might insert into the apical membrane for enhanced acid or ammonia secretion (reviewed in Ref. 49) or when ion uptake is stimulated by hypo-osmotic shock. Similar V-H+-ATPase immunolabeling patterns have been recently reported in 13 other species of crabs (46).
Na+-K+-ATPase and V-H+-ATPase are present both in pillar and principal cells. Therefore, it is possible that the gill epithelium of N. granulata has only one cell type for A/B regulation, which could alternatively perform acid or base secretion depending on the physiological status of the animal. This would match hagfish gills, in which Na+-K+-ATPase, V-H+-ATPase, and NHE are all located in the same cells (32, 42, 44). However, it is also possible that the cytoplasmic pool of V-H+-ATPase from the principal and pillar cells differentially insert into the apical membrane during acidosis. Alternatively, it is possible that V-H+-ATPase is present in the basolateral membrane of certain gill cells, but the labeling looks cytoplasmic due to the deep infoldings of the basolateral membrane (22, 23) (see Ref. 41 for a similar situation in teleost fish).
Perspectives and Significance
The low-pH and HCO3− stimulation of Vte reported in this study raise the following interesting questions: 1) Do these two mechanisms take place in the same gill cell type or are there specific cell subtypes for each of them? 2) Is ion regulation linked to A/B regulation in the gills of crustaceans? 3) What is the identity of the pH/[HCO3−] sensor(s) that regulates the activation of one mechanism over the other? Finally, given that the key ion-transporting proteins (e.g., Na+-K+-ATPase, V-H+-ATPase, CA) involved in our proposed A/B regulatory mechanisms in crab gills are also present in ion-transporting epithelia from vertebrates, it is possible that they all share a similar direct activation by blood A/B variables. It is exciting to hypothesize that intrinsic cellular signaling pathways could have been exploited for further regulation via hormones during the course of evolution.
This research was funded by the Izaak Walton Killam Memorial Scholarship, The Company of Biologists Travel Fund and The University of Alberta, Faculty of Graduate Studies and Research. Research Abroad Travel Grant (to M. Tresguerres), the Queen Elizabeth II doctoral scholarship and the University of Alberta, Faculty of Graduate Studies and Research. Travel Grant (to S. K. Parks), Consejo Nacional de Investigaciones Científicas y Tecnicas PIP 6244 and material support from CIEDE to C. M. Luquet and an Natural Sciences and Engineering Research Council Discovery Grant to G. G. Goss.
We acknowledge the assistance by the staffs at Asentamiento Universitario San Martín de los Andes and Centro Internacional de Educación para el Desarollo who were extremely helpful with troubleshooting. We are also grateful to Dra. Iara Rochetta, Mirna Ferrada, Gonzalo Fernandez, Ana Schiffrin, Pablo, Facundo, and Ivan Nonini and Mr. Sabatini (Sr.). We would also like to thank the input of three anonymous reviewers who recommended additional experiments that improved this manuscript.
↵* Both authors contributed equally to this work.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society