SUMMARY1. Responses to light were recorded from bipolar cells in the retina of the dogfish, Scyliorhinus canicula, under dark-adapted conditions. The identity of the cells was confirmed by Procion Yellow staining.2. More than 95 % of the bipolar cells sampled were of the type which depolarized to a spot of light. These are termed depolarizing bipolar cells. In most cells, illumination of the surround had little effect on the responses elicited from the central receptive field.3. The mean flash sensitivity of the depolarizing bipolar cells was 270 mV/Rh** (where Rh** signifies rhodopsin photoisomerization per rod for full field illumination).4. The mean flash sensitivity of horizontal cells under the same conditions was 8 mV/Rh**. In a limited sample of hyperpolarizing bipolar cells the highest flash sensitivity was 42 mV/Rh**.5. The high flash sensitivity of the depolarizing bipolar cells indicates a large voltage gain at its synapse with rods. On the assumption of a rod flash sensitivity of 2 mV/Rh** the mean gain at the synapse was 135, but for some cells the gain was in excess of 500.6. Responses of depolarizing bipolar cells to dim flashes could be approximated by the impulse response of a 12-16 stage low-pass filter, whereas horizontal cell responses could be fitted by a low-pass filter of six sections. The implied filter at the rod-bipolar cell synapse is tuned to the higher frequency components of rod signals, thereby improving temporal resolution in the rod pathway.7. Depolarizing bipolar cell responses to test flashes are reduced by weak background illumination (less than 0-1 Rh**/sec). This desensitization, which would not be expected to affect rod responses, could be explained by a shift in the operating point to a less sensitive region of the intensity-response curve as a result of the large depolarization elicited by the background.8. The results of current injection into the cell in darkness and during the response to light are consistent with the release by rod terminals of a transmitter which closes ionic channels in a conductance path having a reversal potential of -8 mV, transmitter release being suppressed by light.

No abstract

The linear electrical properties of muscle fibres have been examined using intracellular electrodes for a. c. measurements and analyzing observations on the basis of cable theory. The measurements have covered the frequency range 1 c/s to 10 kc/s. Comparison of the theory for the circular cylindrical fibre with that for the ideal, one-dimensional cable indicates that, under the conditions of the experiments, no serious error would be introduced in the analysis by the geometrical idealization. The impedance locus for frog sartorius and crayfish limb muscle fibres deviates over a wide range of frequencies from that expected for a simple model in which the current path between the inside and the outside of the fibre consists only of a resistance and a capacitance in parallel. A good fit of the experimental results on frog fibres is obtained if the inside-outside admittance is considered to contain, in addition to the parallel elements R m = 3100 Ωcm 2 and C m = 2.6 μF/cm 2 , another path composed of a resistance R e = 330 Ωcm 2 in series with a capacitance C e = 4.1 μF/cm 2 , all referred to unit area of fibre surface. The impedance behaviour of crayfish fibres can be described by a similar model, the corresponding values being R m = 680 Ωcm 2 , C m = 3.9 μF/cm 2 , R e = 35 Ωcm 2 , C e = 17 μF/cm 2 . The response of frog fibres to a step-function current (with the points of voltage recording and current application close together) has been analyzed in terms of the above two-time constant model, and it is shown that neglecting the series resistance would have an appreciable effect on the agreement between theory and experiment only at times less than the halftime of rise of the response. The elements R m and C m are presumed to represent properties of the surface membrane of the fibre. R e and C e are thought to arise not at the surface, but to be indicative of a separate current path from the myoplasm through an intracellular system of channels to the exterior. In the case of crayfish fibres, it is possible that R e (when referred to unit volume) would be a measure of the resistivity of the interior of the channels, and C e the capacitance across the walls of the channels. In the case of frog fibres, it is suggested that the elements R e , C e arise from the properties of adjacent membranes of the triads in the sarcoplasmic reticulum . The possibility is considered that the potential difference across the capacitance C e may control the initiation of contraction.

Whole-cell patch-clamp recordings were obtained from light-responsive on-bipolar cells in retinal slices of the dogfish. Inclusion of the G-protein activator, GTP gamma S, in the intracellular patch solution mimicked the action of glutamate, inducing an increase in net outward current (interpreted as a decrease in inward current), a decrease in membrane conductance and block of light responses. Cyclic GMP (cGMP) in the patch pipette increased inward current and membrane conductance, and blocked light responses. Cyclic AMP had no effect. IBMX, a phosphodiesterase inhibitor, produced the same effect as cGMP, suggesting the presence of a cGMP phosphodiesterase in rod bipolar cells. These results indicate that the glutamate receptors of on-bipolar cells are coupled via a G-protein to regulate intracellular cGMP, which, in turn, results in the opening of sub-synaptic membrane channels. The similarity to phototransduction is striking, and the proposed scheme would account for the high gain in transmission of rod signals to on-bipolar cells.

No abstract