Long-Term Adaptations of Sabella Giant Axons Tohyposmotic Stress

The preceding paper (Carlson, Pichon & Treherne, 1978) showed that the polychaete worm, Sabella penicillus L., is a euryhaline osmoconformer in which the nervous system can be exposed to variations in osmotic and ionic concentration of the blood sufficient to cause irreversible damage to the axons of a stenohaline osmoconformer (Pichon & Treherne, 1976). The ability of the giant axon of Sabella to with-stand, in vitro, abrupt hyposmotic stress was shown to result from an increased restriction to intercellular ion movements between the axon surface and the external medium (Carlson et al. 1978). This restriction appears to be a homeostatic device to protect the axon from short-term fluctuations in osmotic and ionic concentration of the body fluids. Continued exposure to diluted saline in the extreme hyposmotic range experienced by the blood, nevertheless, caused a reversible axonal conduction block. This suggests that in vivo the axon must be capable of adapting to changes in the osmotic and ionic concentration of the blood resulting from alterations in external salinity.

In this paper we describe the results of experiments designed to elucidate the physiological mechanisms involved in the hyposmotic adaptation of the Sabella giant axon.

Worms were collected at low tide near Roscoff or trawled in the Tamar estuary near Plymouth, U.K. and maintained in aquaria in fresh, flowing, sea water until used. Hyposmotic adaptation was achieved by exposing freshly collected individuals to progressive dilution by mixing flowing and well-aerated sea water with Roscoff tap water. To adapt to 50 % sea water the worms were successively exposed, each day, to 70, 60, 55 and, finally, 50% dilutions of sea water. Increased salinity was achieved by appropriate addition of sodium chloride to freshly collected sea water. The osmotic concentration of the blood, collected from the tentacular crown, was measured using the method of Ramsay & Brown (1955).

The electrophysiological techniques used were essentially similar to those used previously (Carlson et al. 1978). The dissection of hyposmotically adapted worms was carried out in appropriately diluted artificial sea water (ASW). In vitro hyposmotic adaptation of the axons was achieved by progressive dilution of the ASW in the experimental chamber. This dilution was made by vigorously syringing the diluted saline onto the preparation, in the absence of constant flow through the chamber, to avoid fluid stratification in the immediate vicinity of the axon surfaces (Treherne, Benson & Skaer, 1977).

The experimental saline used in these experiments was an artificial sea water and had the following composition: Na⁺, 470 mM; K+, 10 mM; Ca²⁺, 11 mM; Mg²⁺, 55 mM; Cl, 609·7 mM, and HCO₃⁻, 2·3 mM.

Freshly caught individuals were rapidly killed by transfer from normal to 50% sea water (Fig. 1). Progressive daily hyposmotic dilution enabled the animals to survive in 50% sea water, although subsequent transfer to 40% caused rapid mortality (Fig. 1).

A number of attempts to adapt worms to below 30% dilution were unsuccessful, despite very gradual rates of dilution, which suggests that one half dilution of normal sea water is probably near the physiological limit for this species. Increase in salinity of above about 150% was also found to be lethal for sea-water-adapted individuals.

The osmotic concentrations of the blood declined rapidly following exposure to diluted sea water (Fig. 2). As can be seen from Fig. 2 the half-time for dilution of the blood was 1 h for transfer of sea-water-adapted animals to 70% or 50% sea water.

In contrast to the axons of sea-water-adapted individuals (Carlson et al. 1978) those of individuals adapted to 50 % sea water were able to conduct action potentials during prolonged exposure to 50% ASW (Fig. 3).

The resting potentials measured in the axons of hyposmotically adapted worms in 50% ASW averaged 63·9 ± 5·8 mV (n = 7) and were not significantly different from those of sea-water-adapted individuals in 100% ASW (66·0 ± 2·2 mV, n = 15) or in 50% isosmotic (sucrose-substituted), ASW (66·6 ± 3·7, n = 5).

The overshoot of the action potentials recorded in the axons of hyposmotically adapted individuals in 50% ASW averaged 42·3± 2·2 mV (n = 6) and were significantly smaller than those recorded in the axons of sea-water-adapted individuals in 100% ASW (52·7 ±2·6, n = 32). They were, on the other hand, significantly larger than those of sea-water-adapted worms in 50% isosmotic ASW (35·7 ±4·6 mV, n = 12).

As previously shown (Carlson et al. 1978) abrupt exposure to hyposmotic saline (60 % ASW) eventually led to reversible conduction block and axonal depolarization (Fig. 4 A). In contrast, progressive hyposmotic dilution of the bathing medium resulted in sustained axonal function, with overshooting action potentials persisting at 60% hyposmotic dilution of the bathing medium (Fig. 4B). The maximal rate of rise of action potentials declines, in the experiment illustrated in Fig. 4A, from 706·8 V s⁻¹, in the 100 % ASW, to after hyposmotic adaptation in 60% ASW. The resting potential showed no appreciable overall change during in vitro hyposmotic adaptation of the giant axon. Attempts to adapt the giant axon, in vitro, to < 60 % were unsuccessful.

Elevated external potassium concentrations resulted in relatively rapid depolarization and subsequent repolarization on return to the normal (6 mM) external potassium level in axons adapted, in vitro, to 60 % hyposmotic ASW (Fig. 5). In this respect the hyposmotically adapted axons were similar to those of sea-water-adapted individuals in 100% ASW (Carlson et al. 1978). They differ, however, from normal axons exposed to abrupt hyposmotic dilution in which a supposed reduced intercellular access resulted in relatively slow potassium depolarization of the axonal membrane (Carlson et al. 1978).

Fig. 6 illustrates the relation between membrane potential and external potassium concentration in axons from sea-water-adapted individuals, in 100% ASW, and in axons which were adapted in vitro (Fig. 4B) to 60% hyposmotic, ASW. The slope relating membrane potentials to [K⁺]₀ for the hyposmotically adapted axons showed 36·8 mV alteration (for decade change in external potassium) which was not significantly different from that of 41·8 mV for the axons from sea-water animals (Fig. 6). Extrapolation of the exponential relations shown in Fig. 6 to zero potential indicates that the estimated intracellular potassium concentration had decreased from 490 to 340 mM during in vitro hyposmotic adaptation.

The effects of variation in external sodium concentration on the intracellularly recorded action potentials was measured in the axons of individuals taken from 100% sea water and from individuals which were adapted to 50 % sea water (Fig. 7). The slope of the calculated regression line for the axons from hyposmotically adapted animals shown in Fig. 7 was 59·8 ± 6·4 mV for decade change in [Na⁺]₀ as compared with that of 48·8 ± 2·6 mV for the axons of sea-water-adapted worms. The significant difference between the Nernst slopes in Fig. 7 (t = 3·595, P < 0·01; for 8 degrees of freedom) indicates that the relative sodium permeability of the active nerve membrane had increased during hyposmotic adaptation. The intracellular sodium concentrations (estimated by extrapolation of the exponentials to zero potential) were, however, similar in the axons from sea water and hyposmotically adapted individuals (Fig. 7). In both cases [Na⁺]₀ can be estimated to be ca. 45 mM.

Transfer of hyposmotically adapted axons (taken from individuals adapted to 50% sea water) from 50 % to normal ASW caused axonal depolarization and, subsequently, incomplete repolarization (Fig. 8). The initial depolarization was accompanied by a rapid decline in the overshoot and was followed by partial recovery. No significant repolarization or increase in overshoot occurred on return to 50% ASW. The overshoot, in fact, showed a progressive decline in 50% ASW presumably as a result of irreversible damage caused by the abrupt exposure of the hyposmotically adapted axon to the osmotic concentration of normal ASW (i.e. from 520 to 1040 m-osmol).

The giant axons of this polychaete osmoconformer are able to adapt to progressive dilution of the bathing medium (both in vivo and in vitro) and to function at hyposmotic concentrations (down to 50%) sufficient to induce conduction block in unadapted axons.

Unlike the extreme osmoconformer, Mercierella, hyposmotic adaptation of the Sabella giant axon is not accompanied by appreciable hyperpolarization of the axonal membrane. In Mercierella this axonal hyperpolarization results from dilution of [K⁺]₀ and from the fact that the intracellular potassium is not proportionally diluted during hyposmotic adaptation (i.e. one quarter dilution of the external medium producing only a halving of [K⁺]_i) (Benson & Treherne, 1978b). In the Sabella giant axon, on the other hand, the intracellular potassium shows a proportionally greater dilution (i.e. 69% dilution of [K⁺]_i for 60% dilution of external medium) which compensates for the reduction in [K⁺]₀.

The decrease in intracellular potassium concentration during hyposmotic adaptation of the giant axon cannot be attributed merely to dilution of this cation due to axonal swelling, for there was no equivalent reduction in the estimated intracellular sodium concentration. Furthermore, it has been shown by direct microscopical measurements that there is no appreciable increase in axonal diameter during progressive dilution of the bathing medium (Carlson et al. 1978). It appears, therefore, that the observed decrease in [K⁺]_i during hyposmotic adaptation must have resulted from a net loss of this major intracellular cation from within the giant axon : a strategy which is employed by a variety of cells to reduce their internal osmotic concentrations during hyposmotic stress (cf. Hoffman, 1977).

The average overshoot of the action potential recorded in hyposmotically adapted axons in 50% hyposmotic ASW (42·3 ± 2·2 mV) was significantly larger than that for the action potentials recorded in unadapted axons bathed with 50 %, isosmotic, ASW (35·3 ± 4·6 mV). This effect could not result from an increased sodium gradient across the membrane of hyposmotically adapted axons, since in both sets of experiments [Na⁺]₀ was the same and similar values for [Na⁺], (ca. 45 mM) were estimated (from the relation between [Na+]₀ and the peak of the overshoot) in adapted and unadapted axons. The larger overshoot measured in hyposmotically adapted axons, in 50% ASW, appears, however, to result from an increase in the relative sodium permeability of the active membrane. This was shown by the 59·8 mV alteration for decade change in [Na⁺]₀ (which approximates to the relation predicted by the Nernst equation for an ideal sodium electrode) measured in hyposmotically adapted axons as compared with the 48·8 mV slope obtained with unadapted axons (Fig. 7).

It appears then that hyposmotic adaptation of the Sabella giant axon involves a decrease in intracellular potassium concentration which tends to maintain a relatively constant resting potential, despite dilution of external potassium ions, during hyposmotic adaptation. This is accompanied by an increase in the relative sodium permeability of the active membrane in hyposmotically adapted axons so as to increase the extent of the overshoot without appreciable change in the effective sodium gradient across the membrane.

The responses to progressive hyposmotic dilution of the bathing medium differ in a number of ways from those observed in the axons of the extreme euryhaline osmoconformer Mercierella enigmatica (Treherne, Benson & Skaer, 1977). As has already been emphasized, the absence of appreciable axonal hyperpolarization in the Sabella giant axon is an obvious difference from the physiological adaptations of the Mercierella axon to extreme hyposmotic stress (Benson & Treherne, 1978 a, b). In the Mercierella axon the hyperpolarization tends to compensate for the reduction in overshoot resulting from the decrease in E_Na during dilution of the external ions, so as to maintain the amplitude of the action potentials. This hyperpolarization also tends to reduce the degree of inactivation of the sodium channels. These effects are reinforced in the Mercierella axon by a progressive decrease in intracellular sodium concentration during hyposmotic adaptation so as to maintain the relative sodium gradient across the axon membrane. There is, however, no change in the apparent relative sodium permeability of the active axonal membrane which still approximates to an ideal sodium electrode during hyposmotic adaptation. In contrast the Sabella giant axon shows no change in intracellular sodium concentration but does exhibit an increase in the relative sodium permeability of the axon membrane so as to increase the extent of the overshoot despite the similarity of the sodium gradients across the axon membranes of unadapted and adapted axons.

The physiological responses of the Sabella giant axon can be regarded as effective adaptations to the limited, osmotic stress, experienced by this axon as compared with that of Mercierella, the most extreme osmoconformer known (Skaer, 1974). First, the Sabella giant axon appears to receive short-term protection by a reduction in intercellular accessibility during exposure to abrupt hyposmotic dilution of the bathing medium (Carlson et al. 1978). Despite an earlier belief to the contrary (Treherne, Carlson & Skaer, 1977) the Mercierella axon does not appear to receive appreciable short-term protection from the immediate effects of abrupt hyposmotic stress (Treherne, Benson & Skaer, 1977). The Mercierella axon can, however, adapt more rapidly, and to a much lower degree of hyposmicity (down to 7·5 %) (Benson & Treherne, 19786), than the Sabella axon. Secondly, the long-term adaptation of the Sabella axon is associated with a proportionally larger reduction of intracellular potassium, which is present at the high concentration of 490 mM in the axons of sea-water-adapted individuals (Carlson et al. 1978). The reduction in the concentration of this major internal cation during hyposmotic adaptation, which is presumably accompanied by an equivalent loss of anions, must therefore contribute substantially to achieving osmotic equilibrium with the external medium. In the Mercierella axon there is a proportionally smaller reduction in intracellular potassium during hyposmotic adaptation (Benson & Treherne, 19786). This enables hyperpolarization of the axonal membrane to occur, despite reduction in [K⁺]₀, and to reduce apparent sodium inactivation during hyposmotic adaptation (Benson & Treherne, 1978a). However, the membrane of the Mercierella axon possesses structural specializations (in the form of hemidesmosome-like structures linked to a system of intracellular filaments) which could enable it to withstand excess internal hydrostatic pressure resulting from osmotic imbalance with the bathing medium (Skaer et al. 1978).

The present results show that abrupt exposure of hyposmotically adapted axons to solutions of similar concentration to the blood of sea-water-adapted individuals eventually causes irreversible axonal damage. It appears, therefore, that long-term physiological adjustments are necessary to enable the giant axon to adapt to changes in blood ionic and osmotic concentration associated with increases in external salinity from dilute to normal sea water.

It should be emphasized that Sabella penicillus can experience in natural conditions changes in salinity sufficient to produce alterations in the ionic and osmotic concentration of the blood of equivalent magnitude to those employed in this investigation. This is shown, for example, in the occurrence of S. penicillus in Stonehouse Pool, Plymouth, U.K. in 1967 (P. E. Gibbs, personal communication). During that year the salinity of the water varied between approximately 35 and 17%₀ (Gibbs, 1971). The physiological adaptations described in this paper are, thus, likely to be employed in natural conditions and to be of adaptive value for this modest euryhaline osmoconformer.

This research was supported in part by grants from the U.S. European Research Office and the Royal Society.