Search

If the guinea‐pig superior cervical ganglion is partially denervated in a way which spares some preganglionic axons arising from each of the spinal cord segments which normally innervate this ganglion (C8‐T7), sprouting of the intact preganglionic axons occurs without a loss of selective end‐organ responses to stimulation of individual ventral roots (Maehlen & Njå, 1981). In the present work we examine the selective innervation of individual ganglion cells by ventral roots after sprouting and compare it to that in normal ganglia. After sprouting the pattern of ganglion cell innervation by the intact preganglionic axons shows some of the features typical of normal ganglia. Thus each ganglion cell is strongly innervated by one or two neighbouring spinal cord segments, with the adjacent segments contributing a synaptic influence which diminishes with distance from the dominant one. Acutely after the partial denervation there is a tendency for the rostral segments to innervate ganglion cells more strongly than do caudal segments, compared to the situation in normal ganglia. The same is true following sprouting. The pattern of ganglion cell innervation observed after sprouting can be explained if the affinities of ganglion cells for innervation from particular spinal levels are assumed to be unchanged and the shift in the relative availability of different preganglionic axons, caused by the partial denervation, is taken into account. This view was substantiated by statistical analysis based on a model of selective synapse formation in the guinea‐pig superior cervical ganglion. We conclude that the ability of individual ganglion cells to distinguish preganglionic axons arising from different spinal levels is maintained during sprouting. Moreover, the way in which selective recognition and the availability of different preganglionic axons combine to produce particular patterns of innervation appears to be similar in normal development and after partial denervation in maturity.

1. The synaptic connexions established by sprouting of intact preganglionic sympathetic axons were examined by intracellular recording in vitro and by observing the sympathetic end organ responses to ventral root stimulation in vivo. 2. The superior cervical ganglion of the guinea‐pig was partially denervated (70‐85%) by crushing the cervical sympathetic trunk at the level of the subclavian artery, leaving the ansa subclavia intact. The intact nerve carried some preganglionic axons arising from each of the eight spinal cord segments (C8‐T7) contributing innervation to the ganglion. 3. During the first 4 weeks after the operation, there was a two‐to threefold increase in the number of steps in the synaptic response elicited in individual ganglion cells by graded stimulation of the ansa subclavia. There was also an increase in the amplitude of the synaptic potential elicited by each preganglionic axon. 4. This increase in the synaptic contribution of the intact nerve to neurones in the superior cervical ganglion after partial denervation was attributed to sprouting of residual preganglionic axons. A major contribution from collateral connexions between ganglion cells was ruled out by intracellular recording form neurones during antidromic stimulation of their axons in the inferior post‐ganglionic nerve. 5. After sprouting, the specificity of the sympathetic end organ responses elicited by stimulation of the ventral roots of spinal segments T1 and T4 in vivo was indistinguishable from normal, although the strength of these responses increased from just perceptible acutely after partial denervation to near normal 3‐6 weeks after the operation, when sprouting was largely complete. 6. These results show that intact preganglionic axons arising from different spinal levels established selective connexions with different classes of ganglion cells during sprouting.

1. The effects of electrical stimulation of presynaptic and post‐synaptic cells on sprouting after partial denervation were examined in the guinea‐pig superior cervical ganglion with intracellular recording. The partial denervation reduced the mean number of preganglionic axons innervating each ganglion cell from about eleven to about two (Maehlen & Njå, 1981). 2. One week after partial denervation alone, the number of residual preganglionic axons innervating each superior cervical ganglion cell was increased by about 30%. This means that the number of ganglion cells contacted by each residual preganglionic axon was increased by the same amount, which represents an early stage of sprouting. 3. Preganglionic stimulation for 1 hr immediately after the partial denervation (with 100 pulses at 20 Hz every 25 sec) increased the rate of sprouting. Thus 1 week after partial denervation and preganglionic stimulation the number of residual preganglionic axons innervating each ganglion cell was increased by about 70%. Preganglionic stimulation had no similar effect on the innervation of normal ganglia. 4. The acceleration of sprouting caused by preganglionic stimulation was abolished by blocking ganglionic transmission with hexamethonium (30‐‐60 mg kg‐1 hr‐1 I.V.) during the stimulation. 5. Furthermore, the rate of sprouting of residual preganglionic axons was increased by electrical stimulation of the ganglion cells alone. 6. These results show that after partial denervation of the superior cervical ganglion, a period of impulse activity in ganglion cells enhances their subsequent ability to receive innervation from sprouts arising from residual preganglionic axons.

Intact preganglionic axons can sprout and form new functional synaptic connexions with neurones in the partially denervated superior cervical ganglion of the guinea‐pig (Maehlen & Njå, 1981). In the present work we have examined to what extent the degree of sprouting, as measured by intracellular recording, is paralleled by changes in ultrastructural and neurochemical parameters. The mean number of preganglionic axons innervating each neurone, as estimated by intracellular recording from ganglion cells during stimulation of the individual ventral roots, was reduced from about eleven to about two immediately after the partial denervation. However, 5‐7 weeks after the operation, when sprouting was complete, each neurone was innervated on average by about seven axons. The estimated mean amplitude of the synaptic potential elicited by each innervating axon (about 5 mV) was only slightly increased after sprouting. Counts of the number of synaptic profiles per unit area of electron microscopical sections of the ganglion were reduced to about 15% of normal 3‐4 days after the operation, and increased to about 60% of normal after sprouting (5‐7 weeks). The activity of the enzyme choline acetyltransferase was about 20% of the normal value in ganglia examined 3‐4 days after partial denervation. After sprouting (5‐7 weeks), this value was increased to about 35% of that in normal ganglia. These results show that intact preganglionic sympathetic axons have a substantial growth potential. However, the increase in the choline acetyltransferase activity was smaller than the increase in electrophysiological and ultrastructural measurements. Therefore, compensatory axonal branching and synapse formation may cause a reduction in the supply of choline acetyltransferase to each presynaptic terminal from the parent soma.

The rearrangement of synaptic connexions maintained by the intact preganglionic axons was examined with intracellular recording after partial denervation of the guinea‐pig superior cervical ganglion. After sprouting, the number of superior cervical ganglion cells innervated by each residual preganglionic axon was increased by a factor of about three. These new synaptic connexions were stable in the absence of reinnervation of the ganglion by the injured axons. The intact preganglionic axons that sprouted in the superior cervical ganglion abandoned most of the synaptic connexions they normally maintain with neurones in the stellate ganglion. Several months after the partial denervation some of the injured preganglionic axons had usually re‐established synaptic connexions with neurones in the superior cervical ganglion. However, both the rate and final degree of reinnervation were poor compared to reinnervation after complete denervation. During reinnervation of the superior cervical ganglion by the injured axons, there was a reduction in the degree of innervation of this ganglion by the sprouting axons. These results support the idea of an ongoing competition between preganglionic axons, and indicate some of the factors that probably determine the synaptic balance under normal and experimental conditions.

1. The synaptic connexions of the guinea‐pig superior cervical ganglion were examined after collateral sprouting provoked by cutting the preganglionic nerve supply from spinal segments T3‐T7. The selective properties of the connexions made by the remaining segments C8, T1 and T2 were explored both with respect to the segmental origin of the preganglionic axons (Njå & Purves, 1977 a) and with respect to their conduction velocity (Wigston, 1983). 2. After sprouting, the ventral root T1 elicited strong sympathetic effects on both the eye and the ear, while the normal response is largely confined to the region of the eye. This effect on the expressed selectivity was confirmed by intracellular recording in vitro, which showed that virtually all the ganglion cells were now innervated by T1. 3. However, neurones that received dominant innervation from T1 were more frequently innervated by sprouted axons from C8 than were neurones whose dominant innervation derived from T2. This indicates that basic mechanisms promoting segmental selectivity were still functioning. 4. The selective innervation of ganglion cells with respect to the conduction velocity of the preganglionic axons was well maintained after sprouting. 5. These results show that during sprouting from a restricted set of preganglionic axons the synaptic partners are still matched according to both types of selectivity. This suggests that the recognition mechanisms which guide neural development are preserved in adult life, but that the differences in the available sets of preganglionic axons lead to different competitive interactions and different resulting patterns of innervation.