Stretch of mammalian nerve in vitro: effect on compound action potentials

Stretch of nerve has been reported to decrease the amplitude of the compound action potential (CAP) with a complete block appearing in approximately 30 minutes. But for the most part, those experiments were carried out in vivo, and it is generally accepted that the failure of responses was due to a closure of vessels supplying the nerve with a resulting ischemia and anoxia. These studies were undertaken to determine if stretch of nerve has effects that are independent of interference with its vascular supply. In the studies, lengths of rat sciatic and dog peroneal nerves were removed and placed in a chamber supplied with oxygen in which their CAPs were continuously elicited and recorded. This in vitro preparation obviated interference with the nerve's metabolism on stretching. We have previously shown that the form change termed 'beading,' appearing within 10 seconds and reversing as quickly on relaxation, can be elicited with tensions of only several grams. We wished to determine if stretch adequate to produce beading could alter CAPs with the same rapidity. Tensions below 2 g had little effect. On applying tensions of 10-100 g, levels well above those needed to bead the fibers, both increases and decreases of CAP amplitude were seen. The changes occurred within 10 seconds of stretch application, the time at which beading arises with stretch. Although the decreases of CAP amplitudes could be accounted for by beading, the degree of CAP change did not correspond to the amount of tension applied. We hypothesize that the constrictions in the beaded fibers increase axial resistivity and diminish local currents so as to block conduction. The lack of an increasing degree of decreased CAP amplitude with increases in tension is ascribed to the inhibition of elongation offered by the collagen fibrils present in nerve. Collagenase applied to nerves allowed a further increase in length, producing a 'hyperbeading,' showing much longer lengths of beading constrictions on stretch. This would further increase axial resistance and is taken to account for the greater decreases of CAP amplitudes seen following collagenase treatment. To account for those cases where increases of CAP amplitude were seen on stretch, we hypothesize that stretch can also cause an increase in the excitability of the nodes. The outcome of stretch in any given nerve would be the resultant of two opposing actions; beading of the internodes causes a decrease of local currents leading to block of CAPs, while an increased excitability of the nodes acts to augment the responses.