Final Report Summary - NEURAL CIRCUITS (Functional analysis of genetically identified retinal interneurons)
Retinal processing in known to change during different ambient light conditions. In particular, the different photoreceptor dominate the transformation of light energy into neural signals. Rods operate at low light levels, while cones dominate during daylight conditions. Aside from the use of different photoreceptors the receptive field properties of retinal ganglion cells, the output of the retina, are known to change with the ambient light level. In particular, the cells selectivity for different spatial frequencies.
During this project we have uncovered a neural circuit capable of switching on the inhibitory surround of a neuron in a input state dependent manner. Specifically, we have been able to describe the circuitry that mediates the break down of the centre-surround receptive field organisation of retinal ganglion cells in scotopic conditions. Using a combination of two-photon microscopy and a mouse expressing a florescent protein (EYFP) in a subset of its ganglion cells we performed targeted recordings from a large ON ganglion cell type, termed PV1.
The spatial receptive field properties were explored using spots and annuli of different sizes presented at background light intensities that stimulated only rods, both rods and cones or only cones. Similar to studies by Kuffler and Barlow in the 1950s, we found that the PV1 cell has no surround for stimuli that activated only rods, while for light levels that stimulated cones the PV1 cell exhibits clear centre-surround antagonism. These results in the specific suppression of the cells response to the presentation of large spots while the response of smaller spots remain unaffected. Surprisingly, the appearance of the surround was not graded but abrupt and switch-like, appearing with full effectiveness in less then 15 s.
Having identified this robust phenomenon we turned our attention to the responsible circuitry. As a first step we recorded from different circuit components in the retina that mediated either scotopic or photopic vision, i.e. the rod and cone pathways respectively. We performed recording from both rod and cone bipolar cells in slice preparation of the retina. We found that the threshold for ON cone bipolar cells was identical to the light level at which we begin to see the affect of inhibition in the PV1 cell. Illustrating that the light level at which inhibition appears corresponds with the threshold of cones activation.
In order to elucidate how the inhibitory surround is selectively activated by the cone pathway we used a combination of pharmacological manipulation and electrophysiological recordings. Voltage clamp experiments suggest that the inhibitory surround is mediated by postsynaptic inhibition from wide field spiking GABAergic amacrine cells. The application of the GABA antagonist picrotoxin and the sodium channel blocker tetrodotoxin, but not glycinergic antagonist strychnine strongly reduced the inhibitory currents while leaving the excitatory unaffected.
Additionally, GABAergic inhibition is only present at light levels bright enough to activate cones. It is mediated by ON cone bipolar cells as demonstrated by its sensitivity to the mGluR6 agonist, APB. Our work suggests a neural circuit switch that turns on a widefield GABAergic amacrine cell via electrical coupling with ON cone bipolar cells and is toggled by the activation of cones.
During this project we have uncovered a neural circuit capable of switching on the inhibitory surround of a neuron in a input state dependent manner. Specifically, we have been able to describe the circuitry that mediates the break down of the centre-surround receptive field organisation of retinal ganglion cells in scotopic conditions. Using a combination of two-photon microscopy and a mouse expressing a florescent protein (EYFP) in a subset of its ganglion cells we performed targeted recordings from a large ON ganglion cell type, termed PV1.
The spatial receptive field properties were explored using spots and annuli of different sizes presented at background light intensities that stimulated only rods, both rods and cones or only cones. Similar to studies by Kuffler and Barlow in the 1950s, we found that the PV1 cell has no surround for stimuli that activated only rods, while for light levels that stimulated cones the PV1 cell exhibits clear centre-surround antagonism. These results in the specific suppression of the cells response to the presentation of large spots while the response of smaller spots remain unaffected. Surprisingly, the appearance of the surround was not graded but abrupt and switch-like, appearing with full effectiveness in less then 15 s.
Having identified this robust phenomenon we turned our attention to the responsible circuitry. As a first step we recorded from different circuit components in the retina that mediated either scotopic or photopic vision, i.e. the rod and cone pathways respectively. We performed recording from both rod and cone bipolar cells in slice preparation of the retina. We found that the threshold for ON cone bipolar cells was identical to the light level at which we begin to see the affect of inhibition in the PV1 cell. Illustrating that the light level at which inhibition appears corresponds with the threshold of cones activation.
In order to elucidate how the inhibitory surround is selectively activated by the cone pathway we used a combination of pharmacological manipulation and electrophysiological recordings. Voltage clamp experiments suggest that the inhibitory surround is mediated by postsynaptic inhibition from wide field spiking GABAergic amacrine cells. The application of the GABA antagonist picrotoxin and the sodium channel blocker tetrodotoxin, but not glycinergic antagonist strychnine strongly reduced the inhibitory currents while leaving the excitatory unaffected.
Additionally, GABAergic inhibition is only present at light levels bright enough to activate cones. It is mediated by ON cone bipolar cells as demonstrated by its sensitivity to the mGluR6 agonist, APB. Our work suggests a neural circuit switch that turns on a widefield GABAergic amacrine cell via electrical coupling with ON cone bipolar cells and is toggled by the activation of cones.
