Acetylcholine receptor bars and transmitter release in frog neuromuscular junctions

After labelling with rhodaminated alpha-bungarotoxin, acetylcholine receptors in cutaneous pectoris muscles of normal adult frogs (Rana temporaria) appear as brightly fluorescent straight bars, usually extending over the whole gutter. Here we investigated first whether receptor bars can undergo changes and secondly whether they would provide a structural correlate for the strength of a junction. Bars of low fluorescence intensity, as well as short or discontinuous receptor bars consisting of two or three segments, suggest plasticity at the receptor/active zone level. In order to elucidate this notion, receptor bars were studied at different seasons which have previously been shown to be associated with structural changes. In two groups of frogs kept under laboratory conditions simulating wintertime and summertime, respectively, the length and number of receptor bars and the amount of discontinuous bars were investigated. Synaptic contact length, which is the summed length of labelled synaptic branches, and the number and total length of receptor bars did not differ significantly. A clear difference between Group I ("winter" frogs) and Group II ("summer" frogs) was found in the number of discontinuous bars, which was almost twice as high in Group I compared with Group II (6.4 +/- 3.3% S.D. vs 3.4 +/- 1.3% S.D., n = 8 and 7 muscles, respectively, P < 0.05). In addition, the average length of individual bars was slightly longer in Group I frogs (2.16 +/- 0.7 micron S.D. vs 2.07 +/- 0.12 microns S.D., 0.1 < P < 0.05). Transmitter release has been shown to be different in these two groups--as determined from endplate potential measurements in tubocurarine-containing bathing solutions--although it was equal when measured in low Ca2+/high Mg2+ [Dorlöchter M. et al. (1991) Pflügers Arch. 418, Suppl. 1, R31]. We also investigated whether receptor bars would be a reasonable structural correlate of synaptic function by comparing different measures of transmitters release with different structural parameters in 19 identified junctions. The mean quantal content (m) of a junction was positively correlated with the number and total length of receptor bars, but not with synaptic contact area or length. Amplitudes of the first, maximum, and plateau endplate potentials (corrected for a common resting potential and apparent input resistance) at tetanic nerve stimulation (40 Hz for 2 s) in tubocurarine block were strongly correlated with both synaptic contact length and total receptor bar length (r = 0.90 for maximum endplate potential); correlations between m and any structural measure were significantly worse.(ABSTRACT TRUNCATED AT 400 WORDS)