The presence of lipid-containing extracellular deposits in human retina, including drusen, subretinal drusenoid deposits, and basal laminar deposits, is a major risk factor for vision loss in advanced age-related macular degeneration (AMD). Recent advances in imaging mass spectrometry (IMS) have allowed determination of spatially-specific molecular signatures in a variety of tissues, including retina. We have applied IMS to determine the spatial distribution of lipids in retina cell layers in aged human eyes and eyes with age-related macular degeneration (AMD).
Fixed retina tissue from deceased donors (80-85 years) was sectioned to 12-14 µm thickness and placed on ITO slides for IMS experiments. MALDI matrix was applied by sublimation and images were acquired at 10-15 µm spatial resolution using a Bruker Solarix 9.4T FTICR instrument and a Bruker timsTOF Pro instrument. Data were analyzed with FlexImaging and SCiLS software. Registration between optical images and MALDI images was aided by autofluorescence microscopy of sections and custom registration software. To image neutral lipids including unesterified and esterified cholesterol, tissue sections were mounted on gold coated slides or silicon nanopost arrays and matrix-free laser ablation was used to desorb and ionize the neutral lipids. Lipid identifications were carried out by LC-MS/MS analysis of tissue extracts.
Registration of images using autofluorescence images greatly facilitated assignment of IMS signals to specific retina cell layers and extracellular deposits. In AMD eyes, basal laminar deposits and drusen, showed distinct molecular profiles including high levels of unesterified and esterified cholesterol, e.g. CE 18:2, as well as specific sphingomyelin lipids, e.g. SM(d18:2_24:1), compared to surrounding cells. The matrix-free laser ablation method provided high neutral lipid intensities in the sclera and choroid regions as well as in the retinal pigment epithelium (RPE) and Bruch’s membrane. These spatially-resolved molecular images provide clues to the underlying pathobiology of deposit formation that often leads to AMD.
Molecular features of retina deposits were determined using IMS and a custom registration workflow. Sphingomyelins and cholesterol (esterified and unesterified) were prevalent in the observed retina deposits. Because these deposits are direct precursors of the vision-robbing end-stages of neovascularization and atrophy, molecular characterization will help identify new therapeutic approaches.