l’won Res. Vol. 23. No. 4. pp. 363-369. 1983 0042-6989;R3:040363-07SO3 00.0 Printed in Great Brltam Pergamon Press Ltd KINETICS OF SYNAPTIC TRANSMISSION FROM PHOTORECEPTORS TO HORIZONTAL AND BIPOLAR CELLS IN TURTLE RETINA DAVID R. COPENHAGEN’, JONATHANF. ASHMORE~ and JULIE K. SCHNAPF~ ‘Departments of Ophthalmology and Physiology, University of California at San Francisco, San Francisco, CA 94143, U.S.A. 2 School of Biological Sciences, University of Sussex, Falmer, Brighton BNI 9QCi. England and ‘Department of Neurobiology, Stanford University School of Medicine, Stanford, CA 94305. U.S.A. Abstract-The kinetics of synaptic transfer from photoreceptors to second order neurons have been determined in the retina of the snapping turtle. Analysis of the fluctuations resulting from the random release of cone transmitter molecules reveals that the action on postsynaptic channels of the depolarizing bipolar cells is prolonged by approximately IO times compared to the action on the hyperpolarizing bipolar cells. Determination of the cone and rod to horizontal cell impulse response shows that the kinetics of the cone/HlAT synapse is approximately 8-10 times shorter than the rod/H I zyxwvutsrqponmlkjihgfedcba AT synapse. These results indicate that the cone driven, sign preserving synapses exhibit comparable temporal characteristics. Synaptic transmission Kinetics Bipolar cells Horizontal cells Retina INTRODUCTION The retina is a unique, experimentally accessible seg- ment of neural tissue about which it is possible to address questions both on the mechanisms of synaptic interaction between nerve cells and on the mechan- isms by which visual information is processed. Direct transmission of signals from photoreceptors to gang- lion cells requires transfer through at least two chemi- cal synapses. The kinetic properties of these synapses set bounds on the temporal resolution of the infor- mation that proceeds through the retina. In the hopes of understanding the roles and mechanisms of synap- tic transmission, we have examined the kinetic behav- ior of the first synapses in the retina. The properties of cone to bipolar cell synaptic transmission were inves- tigated by analyzing the random potential fluctu- ations in bipolar cells induced by the action of cone transmitter molecules. We find that the synaptic transfer is 5-10 times faster from cones to hyperpolar- izing bipolar cells (HBC) than from cones to depolar- izing bipolar cells (DBC). This kinetic difference is attributed to differences in the duration of the post- synaptic conductance changes elicited by the cone transmitter(s). In addition, the properties of cone and rod synaptic transmission to horizontal cells were studied by comparing the light responses reco:ded simultaneously in photoreceptors and horizontal cells. We find that synaptic transfer from rods to horizontal cells is almost IO times slower than that from cones to the same horizontal cells. METHODS Intracellular microelectrode recordings were made in eyecup preparations of the snapping turtle, Che- hydra serpentina. The light stimulation and recording system is described elsewhere (Copenhagen and Owen, 1976). In brief, the optical system consisted of a two-channel optical stimulator in which the inten- sity, size, position, timing and wavelength of the stimuli in each channel were independently adjust- able. Microelectrodes were fabricated on either a Liv- ingstone puller or a Brown-Flaming air blast puller. Microelectrodes with resistances of 5ON300 MR were used to make bipolar cell recordings; horizontal and photoreceptor recordings were made with 20& 500 MR microelectrodes. Membrane potentials were amplified, filtered and then recorded on magnetic tape for later analysis. Identification of cell types was based upon retinal depth of penetration, response waveforms, receptive field properties and spectral sen- sitivities (Copenhagen and Owen, 1976; Leeper and Copenhagen. 1979; Ashmore and Copenhagen, 1980). Data from magnetic tapes were digitized and stored on floppy discs in the PDP-I l/O3 laboratory mini- computer. The single sided power spectrum was com- puted by a standard 1024 point FFT algorithm on data which had been cosine tapered (Bendat and Pier- sol, 1971). Averaging of flash responses was also per- formed on this computer. RESULTS Light responses in zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONM bipolar cells Both DBC’s and HBC’s were impaled 120-180 m distal to the internal limiting membrane. Identifica- tion of these two cell types was verified by Lucifer dye injections. For both bipolar cell types, the diameter of the center of the receptor field, as mapped by circular stimuli, was IO@-150pm. Surround antagonism was 363