Gene expression profiling of primary normal and glaucomatous trabecular meshwork cells by RNAseq

Abstract

Purpose : The trabecular meshwork (TM) is a major component of the increased resistance to outflow in glaucoma patients, but the changes underlying the diseased TM are not well understood. To better understand mechanisms of glaucoma, we sought to identify differences in baseline global gene expression profiles in primary and immortalized TM cells derived from normal and glaucomatous human donors.

Methods : The following cells were included in the study: 25 primary glaucomatous TM cell strains, 17 primary normal TM cell strains, 7 immortalized glaucomatous TM lines, 9 immortalized normal TM lines, 1 primary normal ciliary muscle cell strain, 1 immortalized ciliary muscle line, and 2 primary microvascular endothelial cell strains (blood and lymphatic). Cells were harvested at 90% confluence and total RNA extracted. RNA integrity was verified and sample libraries prepared using a TruSeq RNA Sample Prep Kit. Libraries were clustered and sequenced using HiSeq PE Cluster and SBS Kits on an Illumina HiSeq 2500 instrument. Expression of selected genes was verified by immunofluorescence, Western blotting, or quantitative RT-PCR.

Results : (1) Comparison of global gene expression profile similarity between groups of cells showed that immortalized TM cells were closer in identity to primary TM cells (R² = 0.897) vs. endothelial cells (R² = 0.667 for primary TM vs. endothelial, R² = 0.722 for immortalized TM vs. endothelial); (2) Immortalized TM cells lost expression of some TM markers and glaucoma-related genes, such as MYOC and MGP; (3) No statistically significant differences in gene expression were observed between primary normal and glaucomatous TM cells.

Conclusions : The strong correlation in gene expression profiles between normal and immortalized TM cells suggest that immortalized TM cells could be useful surrogates for high-throughput screening applications, despite downregulation of some TM markers. There are several potential explanations for the lack of gene expression differences between the primary normal and glaucomatous TM strains in this study: (1) Loss of disease phenotype after adaptation to culture conditions; (2) Disease heterogeneity and lack of markers to classify potential disease subtypes; (3) Inability of baseline growth conditions to reveal gene expression profiles associated with disease phenotypes. These issues will be addressed in future studies that will build upon this baseline dataset.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.