Distribution of acetylcholine receptors at frog neuromuscular junctions with a discussion of some physiological implications.

1. The distribution of acetylcholine receptors (AChR) at frog cutaneous pectoris neuromuscular junctions was studied quantitatively using [1125]alpha‐bungarotoxin (alpha‐BTX) labelling and EM autoradiography. 2. We found that, as in mouse end‐plates, the AChR is localized uniformly along the thickened post‐junctional membrane. In the frog muscle this specialized membrane constitutes approximately the top 50% of the junctional folds. 3. The receptor site density is approximately 26,000 +/‐ 6000 sites/micrometer2 on the thickened post‐junctional membrane and falls sharply to approximately 50 sites/micrometer2 within 15 micrometer from the axon terminal. 4. alpha‐BTX site density on the presynaptic axonal membrane was directly determined to be at most 5% of the value on the thickened post‐junctional membrane. 5. The high post junctional AChR site density leads us to conclude that: (a) each quantum of ACh needs to spread only over a very small post‐junctional area (to be called the 'critical area') before it encounters as many AChR (plus AchE) sites as there are ACh molecules in the quantum (for a packet of 10(4) ACh molecules this critical area is approximately 0.3 micrometer2), (b) the average concentration of ACh prevailing in the cleft over this critical area during a quantal response will be approximately 10(‐3)M (independent of the size of the quantal packet), and (c) since 10(‐3)M‐ACh is large compared to any estimates of the dissociation constant Kd for ACh binding to the AChR, the ACh will essentially saturate the AChR within the critical area (provided the ACh binding rate is sufficiently faster than the ACh spreading rate). 6. The total receptive surface for a frog end‐plate is calculated to be approximately 1500 micrometer2, and therefore an end‐plate potential resulting from 300 quanta will be due to the activation of less than 10% of the total receptive area. 7. Free diffusion would allow each small post‐junctional critical area to be reached in less than 15 musec. Therefore, either the recorded rise time of the miniature end‐plate is not predominantly a function of ACh diffusion time, or, as suggested by Gage & McBurney (1975), the net rate of movement of ACh in the cleft is much slower than indicated by the free diffusion constant.