Abstract
Purpose :
Keratoconus (KC), a bilateral, asymmetric corneal degeneration, leads to high myopia, irregular astigmatism, and corneal scarring. In a 4-generation KC family, we identified a rare variant rs373951075 (G>A) in a PCSK1 intron co-segregating with KC in an autosomal dominant pattern. Here, we evaluated the genetic contribution of PCSK1 to KC pathogenesis.
Methods :
We used two mouse models of human PCSK1 deficiency: ENU-induced Pcsk1N222D & targeted mutation Pcsk1tm1Dfs mice. We examined the thickness and curvature of mouse corneas using anterior segment optical coherence tomography (OCT) with a Bioptigen Envisu R2200 system at 3 & 6 months. We performed histological analysis at 6 months. In addition, we checked the expression of PCSK1 in a RNA-Seq dataset of human corneas. We performed bioinformatics analysis with rs373951075 using the UCSC Human Genome Browser and Ingenuity Pathway Analysis (IPA) to identify its potential function and target genes.
Results :
We evaluated 70 (37 male, 33 female, 32 WT, 33 heterozygous, 5 homozygous) Pcsk1N222D & Pcsk1tm1Dfs mice. The average corneal thickness measured by OCT was 110±11µm for WT mice, which did not differ significantly from the mutant mice. No curvature abnormalities were detected in either mutant group. Histologic analyses did not reveal morphologic alterations related to KC in corneas of either mutants at 6 months. The expression of PCSK1 mRNA was limited to background/barely detectable levels in human corneas with RNA-Seq. Bioinformatics analysis indicated that rs373951075 is located in a consensus sequence (TGATGTCAT) binding to JUND, an AP1 transcription factor subunit. This region is predicted to be a distal enhancer to neighboring genes based on ENCODE data. Using IPA Pathway Building, we found that JUND, FOS and estradiol can regulate expression of only 1 of 10 neighboring genes: CAST, which is associated with KC. CAST is also highly expressed in human corneas. Our network analysis connects JUND/CAST/Estradiol with many KC-associated genes, including LOX, IL1B, TGFB1, MMP9, SMAD3, COL5A1, RXRA, COL6A1, FNDC3B, WNT10A, WNT7B, LUM, FBN1, LTBP1, FOXO1, CSNK1E, THBS2, and DCN.
Conclusions :
Data from the Pcsk1 mice suggest that Pcsk1 may not contribute to development of KC. Further analyses suggest that the KC-segregating intronic variant rs373951075 in the PCSK1 region may serve as an enhancer for CAST. Future functional experiments will test the role of CAST in KC.
This is a 2021 ARVO Annual Meeting abstract.