ABSTRACT
Sprouting of undamaged, adult neurones has been observed in a number of situations. For instance, in the peripheral nervous system of vertebrates, intact motoneurones exhibit terminal sprouting in response to axotomy of contralateral motoneurones (Rotshenker, 1979; Rotshenker & Tal, 1985) and in the central nervous system, neurones have been observed to sprout and occupy synapses vacated by lesioned fibres (Cotman, Nieto-Sampedro & Harris, 1981; Tsukahara, 1981). Furthermore, evidence also exists for dendritic sprouting and regression in the mature, human brain (Buell & Coleman, 1981). The plastic properties of the neurones of the freshwater pulmonate Helisoma have been examined in a number of regeneration and sprouting studies (e.g. Murphy & Kater, 1980; Bulloch & Kater, 1982; reviewed by Bulloch, 1985a). Recently, it was demonstrated that sprouting and retraction of a central neurite occurred from an undamaged adult neurone of Helisoma in response to stress (Bulloch, 1984). Specifically, sprouting occurred in a pair of buccal neurones, L5 and R5, and aestivation and body wall incision were the most effective treatments examined. Body wall incision results in haemolymph loss, and aestivation results in a loss of water and thus an increase in haemolymph osmolarity (Machin, 1975). These observations raise the possibility that the sprouting of neurones L5 and R5 may occur in response to changes in haemolymph osmoregulation. The current study directly tested the hypothesis that Helisoma neurones can sprout in response to a change in blood osmolarity. This was achieved by exposing snails to hyperosmotic and hyposmotic pond water which has known effects on blood osmolarity (Khan & Saleuddin, 1979).
Experimental animals (13–15 mm shell height) were placed in 250 ml beakers which contained either normal pond water (0·25 g Instant Ocean salts l−1), hyperosmotic pond water, i.e. 20% sea water (7·6 g Instant Ocean salts l−1) or hyposmotic (deionized) pond water, all of which were based on carbon-filtered Millipore ‘Super Q’ water. The two experimental conditions were chosen on the basis of previously reported salinity tolerance limits and osmotic regulation data of Helisoma trivolvis : compared to a normal haemolymph osmotic pressure of 115mosmol l−1, Helisoma maintains a value of 88 mosmol l−1 in distilled water, whereas in hyperosmotic pond water (200 mosmol l−1, i.e. 20 % sea water) the animal maintains a value of 215 mosmol l−1 (Khan & Saleuddin, 1979). After periods of up to 7 days the paired buccal ganglia were dissected and two paired neurones were injected with Lucifer Yellow CH, i.e. neurones L5 and R5, L4 and R4, or L19 and R19 were stained in any given preparation. Control animals were taken directly from holding tanks and dissected, and appropriate pairs of neurones were stained. Ganglia were fixed overnight in phosphate-buffered 4 % formaldehyde within 3 h of dissection. Specimens were dehydrated with a graded ethanol series, cleared in methyl salicylate and viewed as whole mounts under fluorescence microscopy. Tests for significance between the percentage of neurones sprouted in control and experimental groups was by the Fisher’s (2 × 2) exact test which examines the association between two variables.
Neurone 5 is characterized morphologically by a dendritic arborization in the neuropile which is dominated by a medial dendrite (Fig. 1A). No processes normally penetrate the contralateral buccal hemiganglion. Sprouts are defined as neuritic processes which enter the contralateral neuropile and/or foreign nerve trunks. In the present study the normal morphology of neurone 5 was reconfirmed (N = 38), but in a single preparation both neurones L5 and R5 exhibited a neurite that traversed the commissure which resembled that reported to be evoked by stress (Bulloch, 1984). This suggests that a baseline of sprouting (5 % in this study) exists in the ‘normal’ animal. However, these are the first control neurones 5 observed to sprout out of many (>100) used in a variety of studies (e.g. see Bulloch, 1984 for references).
Sprouting of neurone 5 was seen following placement of the animal in hyperosmotic pond water (Fig. 1B). The observed sprouts had a characteristic morphology: they projected across the buccal commissure and extended towards and/or into the dendritic arborization of the contralateral neurone 5. The sprout was never observed to leave the buccal ganglia, i.e. to penetrate a nerve trunk. Furthermore, only a single cell was observed to sprout in any given preparation. The sprout observed in response to animal placement in hyperosmotic pond water is identical in morphology to the sprouts observed in response to body wall incision (Fig. 1C) and aestivation (Bulloch, 1984). The morphological similarities of the sprouts observed under these conditions suggest that the sprouting response may be triggered by the same factor in all three treatments.
The sprouting in response to animal immersion in hyperosmotic pond water was transient, being maximal (26%) at 1 day (Fig. 2). Given the distance traversed by sprouts, the maximum rate of neurite elongation is about 0·5 mm day−1. In contrast to these results, maximal sprouting has been observed at 3 days in response to body wall incision (Bulloch, 1984). This discrepancy may be due to the absence of direct injury in the present protocol. Although the time course of sprouting was not identical in the hyperosmotic vs injury response, the maximal observed sprouting was similar : 27 % for hyperosmotic water at 1 day and 25 % for body wall incision at 3 days (Table 1). The significance of the transient nature of the sprouting response by neurone 5 is unknown, and is especially paradoxical in the present study which involves exposure of animals to a constant environment.
To determine if the sprouting response in neurone 5 was due to non-specific stress (e.g. from handling), as opposed to an increase in haemolymph osmolarity, neurone 5 was examined for sprouting in animals immersed in normal pond water in beakers for 1 day. Although examples of sprouts by neurone 5 were observed under this condition, their occurrence was not statistically significant (Table 1). This control experiment reinforces the idea that the sprouting response may be due to an increase in haemolymph osmolarity.
In animals placed in hyposmotic pond water, a new type of sprouting response was observed that was characterized by multiple processes. Furthermore, 50% of the sprouted neurones had processes which extended into peripheral nerve trunks (Fig. ID). In common with the sprouting response in hyperosmotic pond water, the sprouting found in animals immersed in hyposmotic pond water was maximal (11 %) at 1 day and was transient (Fig. 2). A few examples of neurones with the single sprout described previously were observed, but the occurrence of this type of sprout was not above control values. Thus, the type of sprouting observed in response to body wall incision and aestivation (Bulloch, 1984) is found in animals placed in a hyperosmotic environment which increases haemolymph osmolarity (Khan & Saleuddin, 1979). In contrast, a qualitatively and quantitatively different type of sprouting response occurs when animals are placed in a hyposmotic environment which slightly lowers haemolymph osmolarity (Khan & Saleuddin, 1979) and requires an increase in diuresis and salt retention (Van Aardt, 1968).
The neuronal specificity of the sprouting response to hyperosmotic pond water was determined by examining two other pairs of identified buccal neurones: R4 and L4, and R19 and L19. Neurones R4 and L4 are salivary effector neurones and R19 and L19 are protractor motoneurones. The morphology of these two pairs of buccal neurones has been described previously (see Bulloch, 1985a for references). Both pairs of neurones were examined for sprouts 1 day after animal placement in hyperosmotic pond water. No evidence of sprouting in either pair of identified neurones was observed (N = 28 for each neurone type). Thus the sprouting response in animals placed in hyperosmotic pond water is specific to particular identified neurones. In agreement with this conclusion is the evidence that neurones 4 and 19 do not sprout in response to body wall incision (A. G. M. Bulloch, unpublished observations). Since neurone 5 innervates the oesophagus, it would be of interest to determine the functional significance of sprouting under these various conditions. It is as yet unknown if the electrical properties of neurone 5 are affected by sprouting, but, in contrast to the consequences of axotomy (Bulloch & Kater, 1982), a new electrical synapse does not form between the two neurones 5 that sprout as a result of animal injury (Bulloch, 1985b).
Taken together, evidence presented previously (Bulloch, 1984) and in this report suggests that neurone 5 can sprout in response to altered blood osmoregulation. The evidence presented thus far does not, however, indicate if sprouting by neurone 5 under these conditions is due to changes in haemolymph osmolarity per se (or selected ion levels) or a change in the levels of hormones associated with osmoregulation. Two further observations support the possible involvement of hormones. First, the osmolarity of blood is unchanged after body wall incision (S. J. Rosser & A. G. M. Bulloch, unpublished observation), a condition which evokes sprouting by neurone 5 (Bulloch, 1984). Secondly, a new sprouting response has been observed in the current study from neurones of animals in a hyposmotic environment, i.e. under conditions in which the change of blood osmolarity is comparatively small. The fact that sprouting and retraction of neurites of specific identified neurones can be induced under conditions similar to those which could occur in nature (i.e. spring run-off dilutes pond water and late summer droughts increase its salinity) suggests that this morphological plasticity could be environmentally induced.
ACKNOWLEDGEMENT
I thank Don Murphy and Janet Richmond for critical review of the manuscript, Rae Barolet for assistance in its preparation, and Helen Bowes for technical assistance. Supported by MRC (Canada) grant MT-7619 and grants from the Alberta Heritage Foundation for Medical Research. DJM is an AHFMR Student.