Gill Na⁺/H⁺ and Cl⁻/HCO₃⁻ Exchange Systems Evolved Before the Vertebrates Entered Fresh Water

Hagfish (Myxine glutinosa, 14·6 to 73·1 g) were collected in the Bay of Fundy by the staff of the Huntsman Marine Laboratory, St Andrews N.B. and maintained in running sea water (12–15 °C) at the Mt Desert Island Biological Laboratory, Salsbury Cove, Maine, or flown to Gainesville, Florida and maintained in plastic pails, supplied with corner charcoal filters, in a constant temperature room (12−13 °C). To test the effect of external ionic substitution on net H⁺ and ammonia efflux, individual hagfish were placed in plastic containers in 200 ml of gently bubbled (either air or 5 % CO₂ in air) sea water (maintained at the acclimation temperature). After a 1-to 4-h acclimation period, a 5 ml sample of the seawater bath in each container was taken as a time-zero control. Subsequent samples were removed during 1-2h of continued aeration in sea water. At the end of this control period in normal sea water, individuals were gently transferred to new experimental baths containing either Na⁺-free or Cl⁻-free artificial sea water (Evans & Cooper, 1976; Kormanik & Evans, 1979) for a period of 1 h, and then returned to a second control sea water bath. Five-ml sample? of these solutions were taken at the end of 1 h. The experimental animals were weighed at the end of the experiment, and the samples of the baths were analysed for both ammonia and H⁺ concentrations, and net efflux of ammonia and H⁺ were calculated as described previously (Evans, 1977, 1982). All data are expressed as X̄ S.E. (TV).

Preliminary studies indicated that summer hagfish excreted net amounts of H⁺, while other individuals excreted net amounts of base during the winter months. This apparent seasonality was especially evident in a single population of hagfish which was tested in the winter in Gainesville and during the summer at MDIBL. As before, the winter fish excreted net base, while the summer fish excreted net H⁺. Neither of the net effluxes was affected by hypercapnia produced by bubbling the experimental baths with 5 % CO₂ in air (unpublished results). Experimental animals were not fed in either the summer or winter experiments, so the causes of this seasonality in net base or H⁺ efflux are unknown. The insensitivity to environmental hypercapnia is especially interesting since various studies have shown that increased P_CO2 stimulates a net H⁺ secretion in many species (for a review of relevant literature see Heisler, 1980, 1984; Evans, 1984a) of teleosts and elasmobranchs. One might argue that the sessile and scavenging habitat of the hagfish would select for a relative insensitivity to environmental hypercapnia, but the true causes for this phenomenon are still to be discovered. Transfer of summer M. glutinosa into Na⁺-free solutions resulted (Table 1) in a significant (P < < 0·01), and reversible, reduction in the net efflux of H⁺, but had no effect on ammonia efflux, a situation similar to that described for the skate Raja erinacea (Evans et al. 1979). It should be noted that it is experimentally impossible to distinguish between the net efflux of H⁺ or net influx of either HCO₃⁻ or OH⁻under these circumstances. We have chosen to call a net drop in external pH a net efflux of H⁺, especially since it is sensitive to the external Na⁺ concentration. Recent studies of teleosts and elasmobranchs (Heisler, 1980, 1984) have demonstrated that the renal contribution to net H⁺ or base efflux is quite small. Attempts at maintaining a patent renal cannula in hagfish were unsuccessful due to their ability to displace the cannula by moving a knot via muscle contraction down the length of their body. Since the response to ionic substitutions in the external solutions was relatively rapid it is unlikely that our data are biased by renal effects. Moreover, the urinary output of hagfish is extremely small 4 −40 μl 100g⁻¹ h⁻¹ (Hardisty, 1979), even below that of marine teleosts, so it is unlikely that renal transport mechanisms play a substantial role in net H⁺ or base effluxes. It is also obvious that ionic substitution could alter passive efflux of ionic species via changes in the transepithelial electrical potential (TEP) (e.g. Evans & Cooper, 1976). Measurement of the TEP across hagfish is extremely difficult for the reasons given for cannulation problems. However, a few preliminary experiments (using the techniques described in Evans & Cooper, 1976) indicate that the TEP across anaesthetized hagfish in sea water is of the order of — 1 mV (blood relative to sea water). Transfer to Na⁺-free sea water hyperpolarized the TEP by some 3 mV, transfer to Cl⁻-free sea water depolarized the TEP by some 5 mV. These results indicate an extremely low ionic permeability, which is supported by our preliminary ²²Na and ³⁶C1 efflux experiments, which found Na⁺ and Cl⁻ turnover rates of less than 0-1 % and 10% of the body ion content per hour, respectively (D. H. Evans & K. More, unpublished). Moreover, the reduction in H⁺ efflux in Na-free artificial sea water seen in Table 1 would require a depolarization of something greater than 100 mV (see Evans, Carrier & Bogan, 1974 for a discussion of the relationship between TEP and diffusive efflux) if it were due to changes in the TEP. Thus it appears unlikely that the results of ionic substitutions in the present experiments were secondary to changes in the TEP.

The fact that seven of the twelve hagfish in the present experiments (Table 1) actually excreted a net amount of base (negative H⁺ efflux) when Na⁺ was preferentially removed from the surrounding sea water implies that mechanisms for base efflux were also present, which may have been overshadowed by a dominant H⁺ excretion. To test this proposition, we examined the effect of Cl”-free sea water on the net excretion of base, which we discovered was typical of M. glutinosa during the winter months. Table 2 shows clearly that removal of external Cl⁻ not only reduced the net base efflux to zero, it reversed it to a net efflux of H⁺, which was completely reversible when the fish were replaced into Cl⁻-containing sea water. Thus a Cl⁻ -sensitive base efflux is present, apparently running in parallel with a Na⁺-sensitive H⁺ extrusion system which predominates in the summer but is hidden behind a dominant base extrusion system in the winter. Another experiment corroborates this proposition.

During the late summer of 1982 we found that, in a single experiment, three fish from the population of hagfish which had excreted net base during the winter, and net H’ during the early summer, excreted a small net amount of base, even under hypercapnic conditions. When these fish were transferred to Na⁺-free artificial sea water, net base efflux increased substantially, and declined again after the fish were replaced into Na⁺-containing sea water (Table 3). In this case, it appears obvious that in normal sea water the net base excretion was reduced by a concomitant extrusion of H⁺, and that elimination of the H⁺ extrusion by removal of external Na⁺ allowed the full expression of a substantial net extrusion of base.

These data are certainly consistent with the presence of at least the Na⁺/H⁺ and Cr/HCO₃⁻ exchange systems in the branchial epithelium of the hagfish. Moreover, they demonstrate directly for the first time that the two exchangers are running in parallel and that the net extrusion of H⁺ or base depends on the relative activity of the respective carriers, and possibly the time of year. Most importantly, this study shows that Na⁺/H⁺ and CF/HCO₃⁻ exchange, so important for ionic regulation in fresh water vertebrates (Evans, 1979), actually arose (presumably for acid/base regulation, Heisler, 1980, 1984) in the seawater ancestors of the vertebrates and provided a ‘preadaptation’ for NaCl regulation in fresh water. One might suppose that the same scenario is true for various invertebrate groups, but the presence of these exchange mechanisms has never been demonstrated in stenohaline marine invertebrates, with no history of freshwater ancestory (Kirschner, 1983). It is also important to note that these data indicate that entry into fresh water during vertebrate evolution was not dependent upon the origin of the necessary ionic uptake systems, any more than euryhalinity of modern marine fishes is apparently limited by the presence or absence of Na⁺ and Cl⁻ uptake mechanisms (Evans, 1984b). What, then, is the determining factor allowing ancient and modern entry into brackish/fresh waters? It appears obvious that it is ionic permeability and the kinetics of these already-present Na⁺ and Cl⁻ uptake systems vis-à-vis the kinetics of ionic loss (Evans, 1979, 1984b). Finally, it is intriguing that we found no evidence for the presence of Na⁺/NH4⁺ exchange in the hagfish, corroborating earlier work with the skateR. erinacea (Evans et al. 1979). These findings are to be contrasted with earlier studies which have demonstrated the presence of both Na⁺/H⁺ and Na⁺/NH4⁺ exchange in marine teleosts and elasmobranchs (Evans, 1973, 1975a, 1977, 1982; Evans et al. 1979). It therefore appears that Na⁺/NH4⁺ exchange evolved subsequent to the origin of Na⁺/H⁺ exchange, and its presence is variable in modern marine fish species, possibly because nitrogen excretion has an alternative pathway via non-ionic (Cameron & Heisler, 1983) and ionic (Goldstein, Claiborne & Evans, 1982) diffusion of ammonia.

In summary, these experiments demonstrate for the first time that the ionic exchange systems (at least Na⁺/H⁺ and CD/HCO₃⁻) which characterize branchial NaCl regulation by freshwater vertebrates were actually present in the early marine vertebrates before this subphylum entered the freshwater environment at least 500 million years ago. Whether they were present in protochordates and other invertebrate, deuterostome ancestors remains to be determined.

This research was supported by NSF Grants PCM 77-2670 and PCM 80-03866 to the author, as well as NSF PCM 77-2670 and NIH Bio-Medical Research Suppor Grant S07 RR 05764 to the Mt Desert Island Biological Laboratory. Appreciation is expressed to Dr Bruce Sidell, University of Maine, Orono for kindly supplying me with hagfish, and to Kane More, Craig Hooks and Andrew Evans for fine technical assistance during segments of this study. A portion of these data was published in abstract form (Evans, 1980).