July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Multimodal classification of mouse retinal ganglion cell types.
Author Affiliations & Notes
  • Megan Lee Zipperer
    Anatomical Sciences and Neurobiology, University of Louisville, Louisville, Kentucky, United States
  • Mark Joseph Kravitz
    Anatomical Sciences and Neurobiology, University of Louisville, Louisville, Kentucky, United States
  • Bart Borghuis
    Anatomical Sciences and Neurobiology, University of Louisville, Louisville, Kentucky, United States
  • Footnotes
    Commercial Relationships   Megan Zipperer, None; Mark Kravitz, None; Bart Borghuis, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 3880. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Megan Lee Zipperer, Mark Joseph Kravitz, Bart Borghuis; Multimodal classification of mouse retinal ganglion cell types.. Invest. Ophthalmol. Vis. Sci. 2019;60(9):3880.

      Download citation file:

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

  • Supplements

Purpose : A major development in neuroscience is the emergence of apparently complete classifications of mouse retinal cell types, defined either functionally, morphologically, or genetically. A first current challenge is to identify corresponding types across these classifications, which for the majority of types remains unknown. A second challenge is to explain at the level of pre-synaptic circuits and cell-intrinsic properties the differences in visual encoding between types. We developed an approach to efficiently address both challenges using intra-experiment functional classification of population-level calcium responses to selectively target identified cell types for morphological identification and electrophysiological characterization.

Methods : We used two-photon fluorescence imaging of ganglion populations (5-15 cells per imaged area) in Thy1-GCaMP6f transgenic mouse retinal explants. We developed a graphic user interface in Matlab to extract from image sequences the evoked response during spatio-temporally diverse visual stimulation of all ganglion cells, in all imaged regions. We also developed algorithms to cluster these responses and find corresponding, previously recorded cell types in a library of responses of many ganglion cells, accumulated across many experiments (>1000 neurons, >30 retinas). We then targeted cell types-of-interest for morphological identification, to link our functionally classified ganglion cell types with corresponding, published morphologically classified ganglion types (museum.eyewire.org). Morphological identification was achieved through targeted neurobiotin injection and immunohistochemical staining of the ChAT bands, followed by confocal imaging and quantitative analysis of dendritic stratification depth and morphology.

Results : Fluorescence imaging readily resolved functionally diverse ganglion cell types, as reported previously (Baden et al., Nature, 2016). Demonstrated morphological identification of functionally identified cell types shows that intra-experiment analysis of calcium responses enables cross-linking functional with existing morphological ganglion cell type classifications.

Conclusions : Our approach allows targeted studies of functionally identified ganglion cell types to resolve the cell-intrinsic and circuit-level mechanisms that give rise to established differences in visual encoding across potentially all functionally identified ganglion cell types.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.


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.