Structures of myocilin olfactomedin domain variants provide new clues into its misfolding and elusive function

Abstract

Purpose : Mutations in the myocilin olfactomedin domain (myoc-OLF) are a strong genetic link to open angle glaucoma. Myoc-OLF is exquisitely sensitive to mutations: the vast majority of documented variants from patients have been shown to be destabilizing and lead to facile aggregation. Myoc-OLF is significantly less stable than the phylogenetically more primitive gliomedin OLF domain (glio-OLF), which also lacks a calcium binding site. The purpose of this study is to probe the structure-stability relationship in myoc-OLF by generating variant(s) that increase myoc-OLF stability, and then characterize resultant biophysical and structural differences.

Methods : Phylogenetic and other computational analysis, site directed mutagenesis, recombinant expression and purification, circular dichroism (CD), differential scanning fluorimetry, crystallization, X-ray diffraction data collection, model building, and refinement.

Results : After in silico analysis of available OLF structures, a series of single and double point mutants of myoc-OLF, including the glaucoma-associated D380A, were generated. Proteins were characterized for intrinsic stability, enhancement of stability in the presence of Ca²⁺, and structural changes. Although the previous trend of destabilization holds true for myoc-OLF variants, we identified several mutant myoc-OLFs that are considerably more stable than wild-type, do not bind Ca²⁺, and propagate large changes in the myoc-OLF structure.

Conclusions : Nature utilizes the metastable, aggregation-prone myoc-OLF domain in the eye even though alternative sequences are available that fold into a similar architecture with higher stability. Relevant to myoc-OLF glaucoma-associated misfolding, our results demonstrate that certain substitutions yield a myoc-OLF nearly as stable as glio-OLF. Newly revealed structural conformations of myoc-OLF are likely pertinent to its still-elusive biological function.

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