ABSTRACT
In vitro experiments were conducted to determine the factors contributing to the unusual distribution of CO2 in the blood of the sea lamprey. When rainbow trout red blood cells (RBCs) were equilibrated with a 3% CO2:nitrogen mixture in either normal saline or sodium-free saline, the extracellular total carbon dioxide content was highly dependent upon the fraction of RBCs in the suspension. In contrast, when lamprey RBCs were equilibrated in normal saline, the decreased with increasing hematocrit. In the absence of extracellular sodium, however, the in the lamprey RBC suspension also became positively correlated with hematocrit. These results suggest that the membrane of sea lamprey RBCs may be somewhat permeable to bicarbonate, but that transmembrane bicarbonate movements may only be detectable in vitro when Na+/H+ exchange is inactivated. Also in contrast to the results for rainbow trout, the changes in that occurred in lamprey RBC suspensions following a step increase in were not associated with any change in RBC chloride concentration and were not markedly affected by the chloride/bicarbonate exchange inhibitor 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid (DIDS). When lamprey RBCs in sodium-free saline were treated with an ionophore for anions, tributyl tin chloride (TBTC), however, the distribution of anions across the RBC membrane came to resemble that of the trout. Furthermore, the relationship between and hematocrit in suspensions of TBTC-treated lamprey RBCs also resembled that of trout in normal saline. Thus, these results demonstrate that both the presence of Na+/H+ exchange and the absence of significant anion exchange contribute to the unique CO2 transport properties of sea lamprey blood.
INTRODUCTION
It is generally accepted that CO2 transport in blood adheres to the same basic pattern in most vertebrates (Roughton, 1964; Obaid et al. 1979; Cameron, 1979; Swenson, 1990). Briefly, CO2 diffuses from the tissues down its concentration gradient into the erythrocyte. Within the erythrocyte, CO2 is immediately hydrated to bicarbonate ions (HCO3−) and protons (H+) by carbonic anhydrase and much of the resulting bicarbonate is rapidly exchanged for plasma chloride. In the respiratory organ, the reverse process occurs; plasma bicarbonate is rapidly exchanged for erythrocyte chloride and combines with a proton to form CO2. The CO2 then diffuses down its concentration gradient into the ambient medium, air or water. In most vertebrates, the majority of CO2 transported by the blood is therefore carried in the plasma as HCO3−. Furthermore, CO2 transport is vitally dependent upon the hydration/dehydration of CO2/HCO3− within the red blood cells (RBCs), catalyzed by the soluble protein carbonic anhydrase, and chloride/bicarbonate (Cl−/HCO3−) exchange across the RBC membrane by the integral membrane protein band 3.
Recent studies have shown that the process of CO2 transport in agnathans, such as the sea lamprey, differs markedly from that of other vertebrates (Tufts and Boutilier, 1989, 1990; Ferguson et al. 1992; Nikinmaa and Matsoff, 1992; Tufts et al. 1992). Unlike that in most vertebrates, CO2 transport in lampreys is largely dependent on CO2 carriage by the erythrocyte. At present, however, the reasons for these unique CO2 transport properties in lampreys are not entirely clear. Earlier studies suggested that Cl−/HCO3− exchange may be absent or very limited in agnathan RBCs (Ohnishi and Asai, 1985; Ellory et al. 1987; Nikinmaa and Railo, 1987). On the basis of these findings, Tufts and Boutilier (1989) proposed that the unusual CO2 transport properties of lamprey blood might be due to the fact that their erythrocytes are impermeable to anions such as bicarbonate. However, Tufts (1992) has more recently shown that sodium-dependent proton movements across the RBC membrane also influence the CO2 transport properties in sea lampreys. Moreover, in the absence of Na+, the RBC membrane of sea lampreys appears to be permeable to HCO3− (Tufts, 1992). Brill et al. (1992) have also recently provided further evidence that the RBC membrane of sea lampreys is permeable to anions. In apparent contrast to earlier studies (Ohnishi and Asai, 1985; Nikinmaa and Railo, 1987), Brill et al. (1992) also found that the lamprey RBC membrane contains significant quantities of a protein that binds irreversibly with DIDS (4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid), presumably a Cl−/HCO3− exchanger. In view of these somewhat conflicting results, it is therefore unclear whether the unique CO2 transport characteristics in lamprey blood are actually due to reduced permeability to HCO3− caused by the physical absence of a Cl−/HCO3− exchange protein or whether they are largely a result of simultaneous movements of HCO3− and H+ across the RBC membrane. In previous in vitro studies (Nikinmaa and Railo, 1987; Tufts and Boutilier, 1989, 1990), simultaneous movements of HCO3− and H+ across the RBC membrane would have resulted in the formation of CO2, which could evolve in an open system, giving the impression that the RBC membrane of lampreys is totally impermeable to HCO3−.
The purpose of the present study was therefore to investigate further the factors contributing to the unique CO2 transport properties of blood in the sea lamprey. Specifically, we examined whether the transmembrane movements of bicarbonate that occur when Na+/H+ exchange is inactivated are due to a Cl−/HCO3− exchange mechanism and/or whether apparent anion exchange limitations persist under these conditions. As a positive control, several experiments also involved a comparison with trout RBCs, which are known to possess large quantities of band 3 (Romano and Passow, 1984).
MATERIALS AND METHODS
Animals
Adult lampreys, Petromyzon marinus L. (320±20.0 g, N=75), were collected during their spawning migration from the Department of Fisheries and Oceans fish ladder at the estuary of the Shelter Valley River and Lake Ontario. Adult rainbow trout, Oncorhynchus mykiss (Walbaum) (0.5–1.0 kg, N=50), were obtained from Linwood Acres Trout Farm (Campbellcroft, Ontario). The animals were transported to the Biology Department of Queen’s University, where they were held in dechlorinated fresh water at 6–9°C for a minimum of 2 weeks before the experiments.
Blood sampling
Animals were anesthetized in an aerated and pH-balanced solution of tricane methane sulfonate (66.7 mg l−1 MS-222 and 133.3 mg l−1 NaHCO3). Using a heparinized syringe, approximately 12 ml of blood was collected from two lampreys via the cardinal vein and pooled in a chilled round-bottomed flask pre-rinsed with EDTA (20 mmol l−1) saline. A similar volume of blood was collected from the caudal vein of one or two rainbow trout and pooled in a second chilled flask.
CO2 distribution experiments
Erythrocytes were washed three times with physiological saline (10.0 mmol l−1 NaHCO3, 124 mmol l−1 NaCl, 5.0 mmol l−1 KCl, 0.5 mmol l−1 MgCl2, 1.1 mmol l−1 CaCl2 and 5.5 mmol l−1 of glucose) containing 5.0 mmol l−1 EDTA. For erythrocytes that were to be equilibrated in sodium-free saline, the three washing procedures were repeated with identical saline except that choline replaced sodium.
The purpose of the first series of experiments was to investigate the distribution of CO2 across the red blood cell membrane in the presence, or absence, of sodium. After the appropriate washing procedure, 2 ml samples of lamprey and trout blood were adjusted to hematocrits ranging from 2.5 to 78% and equilibrated for 60 min at 10°C with humidified 3% CO2 in nitrogen from a Wösthoff gas-mixing pump. It should be noted that nitrogen was used instead of air as an equilibration gas throughout this study in order to maximize the amount of bicarbonate formed within the red blood cell at any given (Ferguson et al. 1992). Following equilibration, approximately 200 μl of blood was dispensed into three hematocrit tubes which were spun at 20 000 g for 3 min in a microhematocrit centrifuge. Hematocrit measurements were then recorded for each tube and 100 μl of the supernatant was drawn into a gas-tight Hamilton syringe for analysis of extracellular total CO2.
The second series of experiments examined the effects of the Cl−/HCO3− inhibitor DIDS on the distribution of CO2 across the membranes of lamprey and trout erythrocytes. Red blood cells were washed and suspended in sodium-free saline and equilibrated for 30 min in paired intermittently rotating glass tonometers (6 ml each) at 10°C with a humidified gas mixture of 0.2% CO2 in air. At this point, the blood in one tonometer was transferred to a flask rinsed with saline containing EDTA (5.0 mmol l−1) and placed on ice. The blood in the other tonometer was centrifuged at 3000 g for 3 min at 10°C and the supernatant was removed by aspiration. The erythrocytes were then resuspended at approximately 30% hematocrit in sodium-free saline plus DIDS (final concentration, 10−4 mol l−1). The equilibration period then continued for another 30 min with 0.2% CO2 in nitrogen. Next, a 1.2 ml blood sample was removed from the tonometer using two Hamilton gas-tight syringes. A 100 μl fraction of this sample was immediately analyzed for whole-blood total CO2 content. Another 400 μl sample of blood was dispensed into a pre-weighed 0.5 ml Eppendorf tube and was centrifuged at 14 000 g for 4 min. The supernatant was then removed from the tube and the RBC pellet saved for chloride determination. The remaining blood sample was used for duplicate hematocrit determination. The supernatant from the hematocrit tubes was drawn into a 100 μl Hamilton syringe for analysis of extracellular total CO2 content. At this point, the gas mixture was changed to 3.0% CO2 in nitrogen and the sampling procedure was repeated after 5, 15 and 30 min. After the DIDS experiment had been completed, the control blood sample was transferred to a tonometer and the protocol was repeated. In subsequent experiments, the order of the control and DIDS experiments was often reversed.
The third series of experiments examined the effects of an ionophore for anions, tributyl tin chloride (TBTC, Pfaltz & Bauer, Waterbury, CT, USA; final concentration 5.0 μ,mol l−1) dissolved in 10 μl of dimethylsulfoxide (DMSO, Sigma) on the total CO2 distribution across the membrane of erythrocytes equilibrated in sodium-free saline. The protocol was identical to that in the previous experiment except that TBTC was used instead of DIDS, and an equal amount of the vehicle (10 μl of DMSO) was added to the control. In this experimental series, 25 μl aliquots of the supernatant from each sample were also used to determine extracellular [Cl−]. An additional set of experiments further examined the relationship between hematocrit and the distribution of CO2 across the erythrocyte plasma membrane in the presence of the ionophore TBTC. The protocol was identical to that of the sodium-free experiments in the first series except that blood was also equilibrated in the presence of TBTC.
Analyses
where Hct is the hematocrit. For the determination of [Cl−]RBC, the red cell pellet was first dissolved overnight in 8% perchloric acid. The [Cl−] of the saline or the extract from the RBCs was then measured with a CMT 10 chloride titrator (Radiometer, Copenhagen).
Statistics
Significance of results for the DIDS and TBTC experiments was assessed using a Student’s paired t-test. Significant differences between the slopes and intercepts of the relationships between and hematocrit were assessed using a Student’s t-test (Zar, 1984). All tests considered P⩽0.05 as significant.
RESULTS
In trout blood, extracellular total carbon dioxide content in both Na+ and Na+-free treatments was highly dependent on the fraction of erythrocytes in the incubation (hematocrit, Fig. 1A). In addition, extracellular Na+ had no significant impact on in trout blood since the slopes and elevations of the relationship between and hematocrit were not signficantly different between Na+ and Na+-free treatments (Fig. 1A). In contrast, the presence of Na+ in the lamprey experiments had a large impact on the relationship between and hematocrit and the slopes of the Na+ and Na+-free regression lines were significantly different. Fig. 1B shows that, in the absence of extracellular Na+, increased with respect to hematocrit, but the slope and elevation of the lamprey Na+-free regression line were still significantly different from those of the trout (Fig. 1A). Fig. 1B also shows that there is a negative relationship between and hematocrit in lamprey when Na+ is present.
In trout blood, treatment with the Cl−/HCO3− exchange inhibitor DIDS resulted in a significantly lower than that of the control following a step increase from 0.2% to 3% CO2:nitrogen (Fig. 2A). In trout, DIDS treatment also resulted in a significantly higher than was observed in the control suspension (Fig. 2B). After DIDS treatment, the chloride concentration of trout RBCs ([Cl−]RBC) was also significantly lower than that of the control following the step increase in CO2 (Fig. 2C). In contrast, the effect of DIDS on the measured variables in the lamprey suspension was minimal. In the lamprey, the was only significantly lower than that of the control 30 min after the step increase to 3% CO2:nitrogen (Fig. 3A). In addition, DIDS had no significant effect on the (Fig. 3B) or [Cl−]RBC (Fig. 3C) at any sampling time in lamprey.
After equilibration with 0.2% CO2:nitrogen in the presence of the anionic ionophore TBTC, the of lamprey RBC suspensions was significantly greater than that of the control (Fig. 4A). Following treatment with TBTC, the was also significantly reduced when compared with the control (Fig. 4B). In addition, TBTC affected the movement of chloride across the lamprey red blood cell membrane. The [Cl−]ext in the TBTC-treated suspension was significantly lower than that in the control (Fig. 4C) and the [Cl−]RBC was significantly higher at all sampling times (Fig. 4D). TBTC also significantly affected the slope and the elevation of the relationship between and hematocrit in lamprey. In the presence of TBTC, the slope of this relationship became significantly elevated from that of the Na+-free lamprey suspension (Fig. 5). Fig. 5 also shows that, after equilibration with this ionophore, the slope of the regression between Hct and was not significantly different from that of trout in normal saline.
DISCUSSION
In the rainbow trout, the in both the Na+ and Na+-free suspensions is clearly dependent upon the fraction of red blood cells (hematocrit, Fig. 1A). The increase in that occurs as hematocrit is elevated can be explained by the increase in proton buffering, and therefore bicarbonate formation, at higher blood hemoglobin concentrations and to the transmembrane movement of bicarbonate to the extracellular compartment through the Cl−/HCO3− exchange protein band 3 (Heisler, 1986). In contrast to the rainbow trout, an elevation in hematocrit only resulted in an increase in in suspensions of sea lamprey red blood cells when extracellular sodium was absent (Fig. 1B). Even in Na+-free saline, however, the slope of the relationship between and hematocrit in the lamprey was significantly less than that of the trout red blood cell suspension. At a hematocrit of 30%, the amount of bicarbonate added to the extracellular compartment by the red blood cells appeared to be about six times greater in the trout than in the lamprey. Nevertheless, these results support the findings of Tufts (1992) and indicate that the red blood cell membrane of the sea lamprey Petromyzon marinus may be somewhat permeable to bicarbonate. One can speculate that these results may also partially explain the significant arteriovenous differences in plasma total CO2 concentration in the lamprey (in vivo) documented by Tufts et al. (1992). Although changes in RBC total CO2 concentration accounted for the majority of the CO2 excretion across the gills, changes in plasma total CO2 excretion were still responsible for almost 40% of the arteriovenous difference in total CO2 in the sea lamprey (Tufts et al. 1992). If the erythrocyte of the sea lamprey were totally impermeable to bicarbonate and, as in other vertebrates, the CO2 reactions in the plasma were uncatalysed, the magnitude of these arteriovenous changes in plasma total CO2 concentration would be difficult to reconcile. The present results suggest, however, that the apparent contribution of plasma bicarbonate to CO2 excretion in vivo (Tufts et al. 1992) may reflect the fact that the red blood cells of the sea lamprey are actually somewhat permeable to bicarbonate, although probably less so than those of other vertebrates. These results also suggest that further studies to quantify the relative permeability of sea lamprey red blood cells in comparison with that of other vertebrates are probably warranted.
In the presence of extracellular Na+, the in the lamprey suspension actually decreased as hematocrit was elevated (Fig. 1B). Further experiments in the presence of ion transport inhibitors, such as amiloride, would be required to determine fully the nature of the sodium-dependent mechanism affecting the in the present experiments. However, there is now substantial evidence that the RBC pH of lampreys is regulated by a Na+/H+ exchange mechanism (Nikinmaa, 1986; Nikinmaa et al. 1986, 1993; Tufts, 1992; Ferguson et al. 1992; Vrikki and Nikinmaa, 1994). Moreover, Ferguson et al. (1992) found that both increased levels and deoxygenation activated Na+/H+ exchange across the red blood cell membrane of the sea lamprey. Thus, in the present study, an increase in hematocrit would probably have been associated with a relative increase in the number of protons being added to the extracellular fluid when extracellular Na+ was available. The extruded protons could then combine with the extracellular bicarbonate, forming CO2, which would be dissipated in the open in vitro system used in these experiments. As observed, this titration of extracellular bicarbonate would then reduce in proportion to the number of RBCs present (Fig. 1B). These results therefore provide further evidence that a sodium-dependent mechanism, possibly Na+/H+ exchange, has an important impact on the characteristics of CO2 transport in sea lamprey blood. Indeed, it is likely that activation of this mechanism in vivo would result in an increase in blood levels and a disequilibrium state for CO2 reactions in the blood, but this has not yet been investigated.
The anion exchange inhibitor DIDS had a significant impact on CO2 distribution across the red blood cell membrane of trout (Fig. 2A,B). Since much of the increase in in the control experiment is due to extrusion of HCO3− from the erythrocyte via Cl−/HCO3− exchange, these results indicate that DIDS was effective in inhibiting anion exchange in trout. Moreover, because Cl−/HCO3− exchange is carried out in an obligate 1:1 process (Obaid et al. 1979), the lower [Cl−]RBC in DIDS-treated blood following the step increase in CO2 is further evidence that the anion exchanger was inhibited in trout (Fig. 2C). In lamprey, however, there was no consistent evidence that DIDS inhibited the movement of anions across the RBC membrane. Although significant, the in DIDS-treated lamprey blood, was only 0.7 mmol l−1 lower than in the control 30 min after the step increase in CO2 (Fig. 3A). In comparison, the in the DIDS-treated trout suspension was 7.6 mmol l−1 lower than that of the control at the same time (Fig. 3A). Furthermore, there were no significant differences in [CCO2]RBC and [Cl−]RBC between control and DIDS-equilibrated lamprey blood (Fig. 3B,C). Thus, although there appears to be limited, but significant, HCO3− movement across the plasma membrane of the sea lamprey when extracellular sodium is removed, these movements do not appear to be the result of the typical DIDS-sensitive Cl−/HCO3− exchange found in most vertebrates. In view of these results, an alternative explanation for the apparent bicarbonate movements should also be considered. In choline medium, the putative Na+/H+ exchanger may actually transport these ions in the opposite direction (i.e. Na+ out, H+ in). This process would also lead to an increased alkalization of the saline and thus to an increase in extracellular bicarbonate concentration by hydration of CO2. Clearly, further experiments will be required to determine fully the mechanism for these apparent bicarbonate movements across the RBC membrane of the sea lamprey.
The distribution of anions in suspensions of lamprey erythrocytes equilibrated in Na+-free saline was significantly different when the ionophore for anions tributyl tin chloride was present (Fig. 4). This ionophore mediates the translocation of OH− and Cl− across the RBC membrane (Motais et al. 1977). As explained by Tufts and Boutilier (1990), since OH− reacts with water to form HCO3−, the presence of this ionophore will also affect the transmembrane distribution of HCO3−. In the red blood cells of most vertebrates, the Donnan equilibrium across the red blood cell membrane is largely determined by the apparent high permeability of the membrane for anions due to the presence of the anion exchange mechanism. The present results suggest, however, that an anion-dependent Donnan equilibrium is only established across the RBC membrane of the sea lamprey when the anion permeability of the membrane is increased by the ionophore TBTC. In this regard, it is noteworthy that the anion distributions in lamprey blood after TBTC treatment (Fig. 4) became very similar to those of trout blood under control conditions (Fig. 2). Indeed, in the presence of TBTC, the slope of the relationship between and hematocrit in lamprey blood (Fig. 5) was no longer significantly different from that in trout (Fig. 5). Thus, anion exchange limitations appear to influence the CO2 transport characteristics of sea lamprey blood even in the absence of Na+/H+ exchange.
In conclusion, the present results confirm that sodium/proton exchange across the RBC membrane contributes to the unique CO2 transport properties of lamprey blood. When extracellular sodium is removed, there appears to be limited movement of bicarbonate across the membrane of lamprey RBCs, but these anion movements apparently do not occur via the typical DIDS-sensitive anion exchange mechanism found in most vertebrate RBCs. Moreover, even in the absence of extracellular sodium, anion exchange limitations continue to influence the distribution of Cl− and HCO3− across the RBC membrane in sea lampreys.
Acknowledgements
This study was supported by an NSERC Operating Grant to B.L.T. The authors would like to thank Mr Rod McDonald and the staff of the Lamprey Control Centre (Department of Fisheries and Oceans) in Sault Sainte Marie, Ontario, for their assistance in obtaining the lampreys.