Exchanges of Sodium Ions in the Central Nervous System of Anodonta Cygnea

Our understanding of the ionic basis of axonal conduction has been largely derived from studies carried out on vertebrate and invertebrate animals which possess conventional body fluids containing relatively high concentrations of sodium and other inorganic ions. Apart from some recent studies on a phytophagous insect (Treherne, 1965 a, b; Treherne & Maddrell, 1967 a, b) we have very little knowledge of the functioning of the nervous system in animals which possess body fluids of specialized chemical composition. One such animal is the freshwater bivalve, Anodonta cygnea, which possesses the most dilute blood so far described in any animal species. The total concentration of the blood of this mollusc averages only 44·0 m-osmoles, with a sodium content of around 15 mM/kg. (Potts, 1954).

In a recent paper from this laboratory (Treherne, Mellon & Carlson, 1969) it was shown that, despite the specialized nature of the blood, the axons were conventional in possessing sodium-dependent action potentials. Furthermore, it was shown that both the small, slowly conducting fibres, and the larger (2·0—6·0 μ) rapidly conducting ones, appeared to be accessible to water-soluble ions and molecules contained in the bathing medium. The rapidly conducting fibres differed from the slowly conducting ones in their response to the treatment of preparations with sodium-free solutions, for whereas the latter showed a rapid decline in activity in these circumstances the former continued to function for extended periods in sodium-free Ringers, in sucrose and in dextran solutions.

The ability of the rapidly conducting fibres to function in sucrose solution presents something of a paradox, for a recent electron microscopic investigation has shown that the extracellular space surrounding the axons in the connective of Anodonta appears to be freely accessible to ions and molecules in the bathing medium. Apart from the neural lamella no visible structures were interposed between the axon surfaces and the bathing medium (Gupta, Mellon & Treherne, 1969). In this respect the results from the previous investigation (Treherne et al. 1969) agree with these electron microscope pictures of the central nervous tissues. The difficulty arises in the explanation of the ability of the rapidly conducting fibres to function in the absence of external sodium ions, for electron microscope investigation revealed that there is a paucity of glial elements in the central nervous connectives. It is, for example, not possible to postulate any local dynamic extra-axonal sodium regulation of the kind suggested in the central nervous system of the stick insect, Carausius morosus (Treherne & Maddrell, 1967b; Treherne, 1967). In this latter species there is a very close glial-axon association in which each axon is surrounded by an appreciable number of glial folds, and it was suggested that an elevated extra-axonal sodium level might be achieved by a local recycling of sodium ions in the narrow extracellular channels formed at the axon surfaces by the closely applied glial cell membranes.

For the above reasons the present investigation was undertaken in order to discover something about the distribution and exchange of inorganic ions in the central nervous tissues of Anodonta cygnea.

The experiments described in this paper were carried out using isolated cerebrovisceral connectives. To determine the sodium and potassium content of the central nervous tissues the excised connectives were carefully blotted, to remove surface fluid, and weighed on a Cahn electrobalance (with an accuracy of ± 0·001 mg.). The connectives were then placed on pieces of platinum foil and ashed in a muffle furnace at 460—480° C. The ash was dissolved in appropriate volumes of distilled water and the concentrations of the two cations were determined using a Unicam S.P. 900 flame photometer. The sodium and potassium concentrations of the blood were also measured, using aliquots of blood diluted 1/100 with distilled water. Tests were carried out to determine the extent of any interference of other elements contained in the tissue samples on the measurements of sodium concentrations. In these tests quantities of sodium roughly equivalent to that measured in the tissue samples were added to the aliquots. The recovery of added sodium, equivalent to 13·1 μM/L., was found to be 13·3 + 0·7μM/L., which suggests that there was no appreciable interference of other elements in the determination of the tissue sodium concentrations.

The physiological solution used in these experiments was that devised by Potts (1954) and had the following composition: 14·0 mM NaCl, 0·5 mM KC1, 5·0 mM CaCl₂, 0-25 mM Na₂ HPO₄ and 1·0 mM glucose. The pH was adjusted to 7-5 by the addition of dilute NaOH. Radioactive solutions were prepared by substituting ²⁴Na or ²²Na for ²³Na in the above solution. The ²⁴Na was assayed using a GM tube (Mifflard MX 123) linked to a scaler unit (Panax 100c); ²²Na was assayed using a Panax scintillation counter with a dioxan liquid phosphor.

In the radioisotope experiments the cut ends of the connectives were ligatured, with fine hairs, immediately after excision. On removal from the radioactive solutions the ligatured connectives were carefully blotted before assay. In efflux experiments the ligatured nerve cords were washed for varying periods in successive 0·2 ml. amounts of non-radioactive physiological solution. The amount of radiosodium remaining in the connectives at varying times was determined from the measured radioactivity of the washings.

To determine the distribution of radioactivity between the nerve sheath and the underlying tissues isolated, ligatured connectives were soaked in “Na-labelled Ringer solution for 1 hr. The ligatured ends of the connectives were then cut away, using a sharp razor blade, and the nervous tissues were squeezed using a small rubber roller as used by Baker, Hodgkin & Shaw (1962). The radioactivity remaining on the sheath and that contained in the extruded tissues were assayed as indicated above.

The sucrose space of cerebro-visceral connectives was estimated by measuring the radioactivity of weighed ligatured preparations to which had been soaked in blood containing 0·16 mM/l. ¹⁴C-labelled sucrose.

The concentration of sodium and potassium in the blood and in freshly dissected connectives are summarized in Table 1.

The values for the sodium and potassium concentrations of the blood are in approximate agreement with previous measurements which have been made with this species (Florkin & Duchâteau, 1950; Potts, 1954). The gross sodium content of the tissues, expressed per unit of tissue water, was lower than that in the blood, whilst that of potassium exceeded the blood concentration of this cation.

If the measured sucrose space (Table 2) is taken as an estimate of the extracellular volume then the values contained in Table 1 can be used to calculate the intracellular concentrations of sodium and potassium ions in the cerebro-visceral connective. This calculation yields mean values of 8-6 mM/1. for sodium and 19-5 mM/1. for potassium ions, if it is assumed that these cations are passively distributed between the blood and extracellular fluid.

The concentration of ¹⁴C-labelled sucrose in connectives was used as an estimate of the volume of the extracellular fluid. For this the isolated ligatured connectives were soaked in blood, containing 0·16 mM/1.¹⁴C-sucrose, for 15 or 30 min. The radioactivity was assayed, after careful blotting and weighing of the connectives, and the results were expressed in terms of the volume of tissue water accessible to sucrose molecules (Table 2). It will be seen that there was no significant increase in the volume of fluid which was accessible to the sucrose molecules in connectives when the period of exposure was increased. This result would seem to indicate that the sucrose molecules were effectively confined to the extracellular fluid, for any appreciable intracellular uptake would have been reflected in a significant increase in uptake on doubling the period of exposure to the radioactive solution.

In this experiment isolated, ligatured connectives were soaked in physiological solution containing ²⁴Na. The connectives were removed from the experimental solution at varying times, carefully blotted and the radioactivity was determined. After each measurement the connectives were returned to the radioactive solution. At the end of the experiment the sodium content of the connectives was determined by flame photometry. The results showed that there was an initial rapid uptake of ²⁴Na followed by a much slower rise to a mean value of 10·9 mM/kg. after 2 hr. (Fig. 1). The individual variation in this experiment makes it difficult to attach much significance to the slower increase in radioactivity, although it is clear that the radiosodium approached a complete exchange with that contained in the central nervous tissues.

Experiments were carried out to determine what proportion of ²²Na was associated with the peripheral connective tissue nerve sheath and how much penetrated into the underlying tissues of the connective. For this the ligatured connectives were soaked in a solution containing ²²Na for a period of 1 hr. The radioactivities of the tissues squeezed from the connectives and of the sheath and associated tissues were then measured (Table 3). The results show that the greater part of the radioactivity was contained in the tissue contents squeezed from the connectives. It can, therefore, be safely assumed that the exchanges of radiosodium described in the previous section were, in fact, largely taking place between the bathing medium and the nervous tissues of the connectives. The proportion of the radioactivity contained within the nerve sheath must be appreciably smaller than that indicated in Table 3, for it is unlikely that the relatively crude technique employed here would have achieved a complete separation of the nerve sheath from the underlying nervous tissues.

Fig. 2. illustrates the rate of loss of radiosodium ions into non-radioactive solution from ligatured cerebro-visceral connectives which had previously been soaked for 3 hr. in Ringer solution containing ²²Na. It will be seen that the emergence of labelled sodium ions appeared to follow a complex course initially and eventually assumed an exponential form after a period of between 800 and 1000 sec.

It was possible to separate a fast component from the curves for the loss of ²²Na from the connectives by subtraction from the initial values lying above the line extrapolated to zero time. The separation of the efflux into fast and slow components is illustrated in Fig. 2. It will be seen that the fast component was complex initially, but assumed after a few seconds a simple exponential form with a half-time (t_0·5) of approximately 150 sec. The half-time for the slow component illustrated in Fig. 2 was 980 sec.

Efflux experiments were also carried out in physiological solution at 0° C. and in isotonic sucrose solutions. The results obtained from these experiments are sumnarized in Table 4. It will be seen that at low temperature there was a significant reduction in the escape of the slowly exchanging fraction and an increase in the efflux of the initial fast one. The efflux of sodium ions in the fast fraction was not, apparently, affected by the substitution of sucrose for physiological saline as the bathing medium. There was, however, a slight but significant reduction in the escape of radiosodium from the slow fraction in preparations bathed in sucrose solution.

The results showed that, despite the relatively low concentration of sodium ions in the blood, the concentration of this cation exceeds that in the central nervous tissues. In these experiments the sodium level in the blood averaged 14·0 mM/1. as compared with a concentration of 9·9 mM/1. of tissue water in the cerebro-visceral connective. When allowance is made for the volume of the extracellular fluid, estimated from the measured sucrose space, it was found that the mean intracellular level of sodium averaged 8·6 mM/1. It is clear, therefore, that, because of the extremely low intracellular level of this cation, there is a concentration gradient of sodium ions between the extracellular and intracellular compartments in the connective. The situation is thus similar to that encountered in the fast and slow adductor muscles of Anodonta in which the sodium concentration amounted to only 45 and 63 % respectively of the blood concentration (Potts, 1958). The intracellular level of this cation in the muscle fibres was, again, exceptionally low, being only 5·3 mM/1. fibre water in the fast adductor and 7·2 mM/kg. fibre water in the slow adductor.

It was shown that there was a relatively rapid uptake of labelled sodium into the tissues of the cerebro-visceral connective and that there was an approach to a fairly complete exchange with that contained within the nervous tissues. The individual variation encountered in these experiments makes it difficult to attach much significance to the apparently secondary slow uptake of radioactivity by the connectives. The form of the exchange is, in fact, better seen in the experiments on the rate of efflux of labelled sodium from the connectives. In these latter experiments it was shown that sodium efflux can be represented as a two-stage process, in which an initial fast component eventually gives way to a slower exponential decline in radioactivity within the nervous tissues. These results are essentially similar to those obtained with insect central nervous tissues (Treherne, 1961, 1962, 1965, 1966). There are at least two possible explanations for such a two-step efflux. It could, for example, be postulated to result from the different exchange rates of different cellular components within the nervous tissues. Alternatively, it could be envisaged that such a two-stage efflux could result from the different exchange rates of extracellular and intracellular ion fractions. In the case of insect central nervous tissues it was shown that the initial rapid efflux of sodium was unaffected by the presence of ouabain or metabolic inhibitors, but that the secondary component was affected by the presence of these substances. On the basis of these results it was suggested that the fast fraction represented the extracellular sodium, the slow fraction being identified as the intracellular ion fraction. The reduced efflux of the second fraction observed at 0° C in the present study suggests that, as in the insect central nervous tissue, this represents the escape of an intracellular ion fraction. Low temperature was, however, found to have an accelerating effect on the efflux of the fast fraction. It is clear, therefore, that it is not possible to postulate any simple separation in the efflux of extracellular and intracellular ion fractions as in insect central nervous tissues. In the case of the lamellibranch central nervous tissues it could be postulated that the above observations result either from a drastic slowing down of the efflux from a cellular fraction, so as to leave a fast fraction largely composed of the extracellular efflux, or to some reduction in the retention of extracellular sodium which is maintained by active processes. Now the amount of sodium contained in the fast fraction was shown to be reduced from an average of 74·7% of the total sodium at normal temperature to 41·0% at o° C. 41·0 % would correspond to 3·9 HIM Na/kg. tissue. Using a figure of 0·23 /d./mg. tissue for the volume of the extracellular fluid (i.e. the measured sucrose space) then it can be calculated that the concentration of the sodium in the fast fraction, at low temperature, would be approximately 15·6 noM/1. This latter figure is roughly equivalent to the concentration of sodium ion in the bathing medium and would, thus, accord with the idea that the fast fraction, obtained in efflux at low temperature, is largely extracellular. This assumption would further imply that there was no gross regulation of the extracellular sodium or any effects such as the Donnan equilibrium which has been postulated to exist between the blood and extracellular fluid in insect central nervous tissues.

The demonstration that the axons in the cerebro-visceral connective appear to be conventional in possessing a low intracellular concentration of sodium ions, relative to the extracellular fluid, accords with previous observations on the ionic basis of action potential production in this preparation (Treherne et al. 1969). In this earlier study it was shown that the small, slowly conducting axons, were sodium-dependent and developed a conduction block in preparations bathed in sodium-free solutions. It was also demonstrated that dilute tetrodotoxin caused a rapid decline in the action potentials associated with both small and large axons in the connectives.

The relatively rapid movement of labelled sodium between the central nervous tissues and the bathing medium is also in essential agreement with the electrophysiological evidence. Thus, the ready access of sodium ions to the axon surfaces can be inferred from the rapidity of the development of the conduction block in the slowly conducting axons in preparations bathed in sodium-free solutions. Similarly, with the larger, rapidly conducting axons, which continued to function in sodium-free solutions, there appeared to be a ready access to their surfaces. This was shown, for example, by the extremely rapid and reversible effect of tetrodotoxin on the latter axons and by the rapid return of action potentials obtained in preparations poisoned with dinitrophenol when the normal solution was substituted for one containing an elevated concentration of sodium ions.

The apparent mobility of the sodium ions in the extracellular spaces also accords with the structural organization of the connectives as revealed by the electron microscope (Gupta et al. 1969). In this study no evidence was obtained for the existence of any visible structure which would be likely to restrict intercellular ion movements between the bathing medium and the axon surfaces.

Despite the satisfactory degree of agreement between the chemical, electrophysiological and structural evidence outlined above an apparently paradoxical situation has arisen in connexion with the interpretation of the remarkable ability of the rapidly conducting axons to continue conducting action potentials in preparation bathed with sucrose or sodium-free solutions (Treherne et al. 1969). Mention has already been made in the preceding paragraphs of the evidence which suggest that there is very little restriction to the intercellular movement of inorganic ions between the bathing medium and the axon surfaces. There is, furthermore, only a very sparse distribution of glial elements in the central nervous tissues of Anodonta, which does not, apparently, allow the postulation of any local dynamic regulation of the extra-axonal sodium concentration, such as has been postulated in some insect species (Treherne & Maddrell, 19676; Treherne, 1967).

A number of possibilities remain to explain the ability of the larger axons to function in sodium-free media given that there is little restriction in the movements of inorganic ions between the axonal surfaces and the bathing medium. It could be postulated, for example, that the larger axons alter their cation specificity in sodium-free conditions so as to utilize some other cation species. Alternatively, it could be suggested that in the active state the axons switch from sodium-selective mechanism to one involving a net efflux of anions from the axoplasm in sodium-free non-electrolyte solutions. These and some other possibilities will be dealt with in a subsequent investigation being carried out in this laboratory.