May 2007
Volume 48, Issue 13
ARVO Annual Meeting Abstract  |   May 2007
Dendritic Spikes Amplify the Synaptic Signal in a Simulation of the Direction-Selective Ganglion Cell
Author Affiliations & Notes
  • R. G. Smith
    Dept of Neurosci, Univ of PA, Philadelphia, Pennsylvania
  • M. J. Schachter
    Dept of Neurosci, Univ of PA, Philadelphia, Pennsylvania
  • N. Oesch
    Neurol Sci Inst, OHSU, Beaverton, Oregon
  • W. R. Taylor
    Neurol Sci Inst, OHSU, Beaverton, Oregon
  • Footnotes
    Commercial Relationships R.G. Smith, None; M.J. Schachter, None; N. Oesch, None; W.R. Taylor, None.
  • Footnotes
    Support NIH Grants EY016607 and EY014888
Investigative Ophthalmology & Visual Science May 2007, Vol.48, 5967. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      R. G. Smith, M. J. Schachter, N. Oesch, W. R. Taylor; Dendritic Spikes Amplify the Synaptic Signal in a Simulation of the Direction-Selective Ganglion Cell. Invest. Ophthalmol. Vis. Sci. 2007;48(13):5967.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Purpose:: The direction-selective ganglion cell (DSGC) generates spikes in its dendritic tree in response to preferred-direction stimuli, but responds weakly to flashed spots of light traditionally used to excite ganglion cells (Oesch et. al, 2005). However, the somatic PSP is weakly tuned to the DS signal, while dendritic spikelets are strongly tuned. Therefore, the dendrites are thought to be the primary site of spike initiation. Since the presynaptic input to the cell is known to be weakly DS, this implies that dendritic spikes serve to amplify the direction-selective signal. Because dendritic spikelets propagate to the soma with high probability, one question is whether this DS processing is performed by blocking dendritic spike initiation or blocking spike propagation.

Methods:: To investigate how the morphology of the dendritic tree affects its ability to initiate and propagate spikes, we constructed multi-compartment models of 6 digitized On-Off DSGCs, and gave the axon, dendrites, and soma a set of membrane properties and voltage-gated channels calibrated to give passive and spiking properties similar to the real DSGC. Semi-random arrays of bipolar and amacrine cells provided excitation and inhibition. To determine the extent of local processing, we mapped the dendritic tree's passive and active properties by stimulating one synapse at a time.

Results:: We found that distal dendritic regions were electrotonically isolated from the soma. Sub-threshold distal synaptic input was heavily attenuated on its path to the soma. Therefore, the amplitude and tuning of the somatic PSP was not indicative of the distal synaptic input. However, we found distal dendrites highly excitable, and initiated and propagated dendritic spikes robustly to the soma. Spike propagation was affected strongly by axial resistance, but less so by on-path inhibition. However, inhibition within a local distal region prevented the initiation of dendritic spikes.

Conclusions:: These results suggest that nonlinear processing within local dendritic regions of the DSGC amplifies is directional index by selectively initiating spikes for preferred-direction stimuli. Inhibition appears to tune dendritic spike initiation, but not propagation.

Keywords: ganglion cells • computational modeling • ion channels 

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.