Structural characterization of mini-PCDH15 proteins for Usher syndrome type 1F therapy.

Abstract

Hair cells, the sensory cells of the inner ear, carry actin-filled microvillus-like projections called stereocilia, which are interconnected by long protein filaments called “tip links”. Upon sound stimulation, tip links enable hair-cell mechanotransduction, the critical conversion of mechanical deflection into an electrical signal our brain can understand. Tip links are formed by two pairs of large Ca2+-binding proteins, cadherin-23 (CDH23) and protocadherin-15 (PCDH15), interacting tip-to-tip in a Ca2+-dependent manner. The ectodomain of PCDH15 is made of 11 extracellular cadherin (EC) repeats, with EC linkers stabilized by Ca2+ ions bound to acidic residues. Mutations in PCDH15 cause Usher Syndrome Type 1F, characterized by congenital hearing loss, balance deficit, and progressive blindness. Adeno-associated virus (AAV) vectors have been found to be efficient and effective for gene therapy in the inner ear. However, the PCDH15 coding sequence is too large to fit in AAV. We have developed novel “mini-PCDH15” constructs that retain critical domains but lack 3-5 EC repeats, and that consequently fit in AAV vectors. Some of the mini-PCDH15s are functional and rescue hearing and balance in Pcdh15-knockout mice despite having engineered Ca2+-binding linkers that are not found in nature. Here, we present structural and computational studies of mini-PCDH15s, including an X-ray crystal structure of an artificial EC3-EC7 bent linker of a mini-PCDH15 that partially rescues hearing. In addition, we report on biochemical assays, low-resolution cryo-EM, and molecular dynamics simulations used to compare the flexibility, rigidity, and Ca2+-binding ability of wild-type and mini-PCDH15 ectodomains. Our mini-PCDH15s show different stability and behave as dimers in solution, showing diverse conformational arrangements that might modulate binding to CDH23. Our results provide an atomistic view of engineered mini-PCDH15s and promote the use of rational, iterative, structure-based mini-gene approach to develop gene therapies for large proteins.