July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Hyperspectral autofluorescence (AF) of lipofuscin (LF) and vitelliform granules in a canine bestrophinopathy
Author Affiliations & Notes
  • Yuehong Tong
    Ophthalmology, Icahn School of Medicine at Mount Sinai, New York, United States
  • Jacob Rosenbloom
    Keck School of Medicine of USC, California, United States
  • Taariq Mohammed
    New York University, New York, United States
  • Nayanika Challa
    The Ohio State University College of Medicine, Ohio, United States
  • R Theodore Smith
    Ophthalmology, Icahn School of Medicine at Mount Sinai, New York, United States
  • Footnotes
    Commercial Relationships   Yuehong Tong, None; Jacob Rosenbloom, None; Taariq Mohammed, None; Nayanika Challa, None; R Theodore Smith, None
  • Footnotes
    Support  R01 EY015520 (RTS)
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 6133. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Yuehong Tong, Jacob Rosenbloom, Taariq Mohammed, Nayanika Challa, R Theodore Smith; Hyperspectral autofluorescence (AF) of lipofuscin (LF) and vitelliform granules in a canine bestrophinopathy. Invest. Ophthalmol. Vis. Sci. 2019;60(9):6133.

      Download citation file:

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

  • Supplements

Purpose : Canine BEST1 disease (cBest1) is a new animal model for human Best disease caused by mutations of bestrophin 1 (BEST1) (PMID: 29721958). Our purpose was to identify and compare the RPE specific spectral signatures from a normal and a cBest1 donor.

Methods : We prepared 3 retina cross sections from a normal canine and 5 from a cBest1 donor. We acquired hyperspectral AF images from 16 locations (6 from the normal dog and 10 from cBest1) at 10nm emission intervals from 420 nm to 720 nm with a Nuance hyperspectral camera connected to a Zeiss microscope with 40X objective and at excitations 436 nm, 480 nm and 505 nm. We applied simultaneous non-negative matrix factorization to each hyperspectral dataset to extract 3 spectra and their corresponding tissue abundances.

Results : The 3 spectra recovered in each case were two spectra that were similar to RPE spectra previously reported in human RPE flat mounts, and one background spectrum from the slide surface. The first spectrum similar to human resembled in all tissues that reported as human S1 (PMID: 27226929) with a peak at about 520 nm. The tissue abundance of this “S1 like” spectrum in the canine cross sections included not only the RPE but also showed non-specific localization to collagen in the choroid and structures in the neural retina (Figure 1, normal dog, green spectrum, S1-like, with non-specific abundance). The second spectrum similar to human resembled either human S2 or S3, and showed very specific abundance in the RPE and any vitelliform granules (Figure 1, blue spectrum, S2-like, peak at about 570 nm, very specific abundance in the RPE. Figure 2, red spectrum, S3-like, peak at about 600 nm, very specific abundance in the RPE and vitelliform granules). Background spectra are shown in gray, localization not shown. In the cBest donor, the second human like spectrum was S3-like in all 10 tissues. The first was still always S1 like, with both RPE and non-specific localization (Fig 2, green spectrum, peak at 530 nm).

Conclusions :
Decomposition of the spectra from a normal and a cBest1 specimen reveals that the RPE-specific spectral components differ, resembling human S2-lke and S3-like, respectively, with the S3-like spectrum also specific to granules in the vitelliform space. IMS will characterize the fluorophores molecularly and provide more understanding of the mechanism of disease.

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.