Col4a1 mutation causes endoplasmic reticulum stress and genetically modifiable ocular dysgenesis

Douglas B. Gould 3 5 Jeffrey K. Marchant 0 Olga V. Savinova 2 3 5 Richard S. Smith 2 3 Simon W.M. John 2 3 4 0 Department of Anatomy and Cell Biology 1 Present address: Department of Chemistry and Biochemistry , UCSD, La Jolla, CA, USA 2 The Howard Hughes Medical Institute , Bar Harbor, ME, USA 3 The Jackson Laboratory 4 Department of Ophthalmology, Tufts University School of Medicine , Boston, MA, USA 5 Present address: Department of Ophthalmology and Department of Anatomy, Institute for Human Genetics, UCSF School of Medicine , San Francisco, CA, USA Ocular anterior segment dysgenesis (ASD) is a complex and poorly understood group of conditions. A large proportion of individuals with ASD develop glaucoma, a leading cause of blindness resulting from retinal ganglion cell death. Optic nerve hypoplasia is thought to have distinct causes and is a leading cause of blindness in children. Here, we show that a mutation in the type IV collagen alpha 1 (Col4a1) gene can cause both ASD and optic nerve hypoplasia. COL4A1 is a major component of almost all basement membranes. The mutation results in non-secretion of the mutant COL4A1 proteins, which instead accumulate within cells. Basement membrane abnormalities may, therefore, contribute to the phenotype. The mutation also induces endoplasmic reticulum stress and so intracellular stress may contribute to pathogenesis. The overall consequence of the Col4a1 mutation depends on genetic context. In one genetic context, the mutation causes severe ASD with intraocular pressure abnormalities and optic nerve hypoplasia. In a different genetic context, both the ASD and optic nerve hypoplasia are rescued, and we have identified a single dominant locus that confers the phenotypic modification. - INTRODUCTION Glaucoma describes a heterogeneous group of neurodegenerative diseases where death of retinal ganglion cells (RGCs) leads to vision loss (1). One of the strongest known risk factors for glaucoma is an elevated intraocular pressure (IOP). However, some patients have normal-tension glaucoma where death of RGCs occurs in the absence of detected IOP elevation (2). Lowering IOP can often slow disease progression, even in normal tension glaucoma patients (3). This suggests that normal tension glaucoma patients have RGCs that are susceptible to pressure-related death even at normal IOPs. Together, this illustrates that glaucoma is a complex disease where factors that influence IOP regulation and factors that determine ganglion cell viability interact to influence the final course of the disease. IOP is a balance of aqueous humor production, by the ciliary body (CB), and aqueous humor drainage. Aqueous humor drains through the trabecular meshwork and Schlemms canal in the iridocorneal angle and through the uveoscleral drainage pathway. Dysgenesis of the ocular anterior segment can impede aqueous humor outflow and lead to IOP elevation. Consequently, patients with anterior segment dysgenesis (ASD) are at an elevated risk for developing glaucoma. A number of genes have been identified in which mutations lead to ASD in human patients and in mice, but the precise pathogenic mechanisms remain largely unknown (4). Pathogenic alleles of developmental genes often cause a spectrum of ocular phenotypes that vary in severity (5). It is possible that some of these same genes contribute to age-related, open angle glaucoma, where the ocular drainage structures have abnormalities that are not clinically visible but that cause dysfunction with age. Similarly, genes influencing survival of RGCs during development (where severe mutations might lead to optic nerve aplasia or hypoplasia) may modulate RGC susceptibility to glaucoma. Therefore, continued characterization of factors influencing ocular development and dysgenesis may identify new pathways and processes important for age-related ocular diseases. Integration of this information will be important for understanding specific disease processes leading to glaucoma susceptibility. Phenotype-driven approaches represent an unbiased mechanism to identify new genetic factors and biological pathways underlying disease processes. In a mutagenesis screen performed to identify mice with abnormalities in IOP regulation, we discovered a semi-dominant mutation in type IV collagen alpha 1 (Col4a1) (6). The mutation is within the triple helical domain of the COL4A1 protein and leads to an inhibition of secretion of both COL4A1 and its binding partner COL4A2. COL4A1 is the most abundant and ubiquitous basement membrane protein, and the mutation has pleiotropic effects (7). In the eye, COL4A1 is present in the basal lamina of the conjunctiva, corneal epithelium, corneal endothelium, trabecular meshwork, Schlemms canal, lens, CB, retinal inner limiting membrane (ILM), Bruchs membrane and vascular basement membranes (8 11). Here, we show that on the C57BL6/J genetic background Col4a1Dex40 mice have severe ocular dysgenesis including ASD and optic nerve hypoplasia. IOP in the mutant mice is variable with mice exhibiting both high and low pressures. To begin to understand the pathogenic mechanism(s) and to identify genes that might interact with Col4a1 in ocular development, we tested other genetic backgrounds to determine if they modify the phenotypes. We found that they do and we mapped a dominant CAST/Ei derived modifier locus to Chromosome 1. Mutation of Col4a1 causes ASD We performed a random embryonic stem cell mutagenesis screen to identify genes that contribute to glaucoma pathology (12). We identified a mutant lineage with ocular ASD and buphthalmos (enlargement of the eye), and we determined that the causative mutation was in a splice acceptor site of Col4a1 resulting in the absence of exon 40 from the mature transcript (Col4a1Dex40) (6). Col4a1Dex40/Dex40 mice are not viable, and Col4a1/Dex40 mice have decreased viability. On the C57BL/6J genetic background, all surviving Col4a1/ Dex40 mice have clinically obvious ASD. The phenotype is variable and includes all tissues of the anterior segment. Some combination of buphthalmos, corneal opacification, pigment dispersion, iridocorneal synechiae (attachments of the iris to the cornea), cataracts, persistence of tunica vasculosa lentis and abnormal iris vasculature is present in all mutant mice (Fig. 1). To understand how Col4a1Dex40 caused ASD, we determined the age-at-onset of observed dysgenesis. Histologic analysis of eyes from control and mutant embryos at embryonic day (E) 16.5 revealed anterior hyphema (hemorrhage in the anterior chamber) in five of six eyes from Col4a1/Dex40 mice but not in control mice (Fig. 2). At E18.5, remnants of anterior hyphema were still present. Despite the hyphema, the overall anterior segment morphology, including the iridocorneal angle, was indistinguishable between control and mutant mice until birth. The iridocorneal angle in mice continues to develop postnatally (4,13). Histologic analysis of anterior Figure 1. Col4a1 muta (...truncated)