Preview

PDF, English - main document Download (6MB) | Terms of use

Abstract

Glycosylation is one of the most abundant forms of ubiquitous co- and posttranslational modifications affecting protein stability, transportation, and function. The most common type of glycosylation is N-glycosylation, which refers to the addition of glycans to the Nitrogen of Asparagine. Mutations of enzymes involved in this pathway give rise to Congenital Disorders of Glycosylation (CDG) by generating hypomorphic alleles rendering glycosylation enzymes with partial activity. Since animal models of CDGs mostly rely on mutations leading to complete loss of function, the human situation is not recapitulated. Here, we present two distinct hypo-N-glycosylation fish models to address the effects of reduced glycosylation on animal physiology and at the molecular level. The alg2hypo model was generated by introducing a well-defined mutation of alpha-1,3/1,6- mannosyltransferase 2, Alg2, from an ALG2-CDG patient allele into the orthologous region of medaka (Oryzias latipes) via CRISPR/Cas9 targeted genome editing. Under homozygosity, this model displayed multisystemic phenotypes closely resembling patient symptoms, including reduction in white matter, motor development abnormalities and craniofacial dismorphism. N-glycome analysis revealed a total reduction in N-glycan occupancy both in whole medaka and patient fibroblast samples, which in turn led to changes in protein abundance. Accordingly, mass spectrometry of both whole hatchling and dissected eye samples showed reduction in proteins responsible for photoreceptor signaling, as well as upregulation of proteins responsible for nucleotide-sugar metabolism, N-glycosylation machinery, vesicle trafficking and protein folding. Immunofluorescence analysis revealed rod cell death leading to a condition called retinitis pigmentosa or night blindness seen among patients with other CDG types. Finally, mRNA injections into alg2hypo line synthesized from both medaka and human full-length Alg2 coding sequences rescued the multisystemic phenotypes and overcame early-juvenile lethality. In a second model, phosphomannomutase 2 (Pmm2) enzyme was depleted by an engineered degron, deGradFP, to create a proxy for PMM2-CDG model by reducing the enzyme amount to different levels via auxin induction. To this end, Pmm2 was tagged endogenously in its C-terminal end with GFP and was successfully depleted when F-box protein, TIR1, was ectopically expressed together with an auxin-induced degron (AID) fused to a GFP nanobody binding to GFP tagged proteins. All in all, this study presents two different medaka models of hypo-N-glycosylation to study the mechanism behind multisystemic disease phenotypes seen among CDG patients. Both models serve as promising platforms for pre-clinical studies, such as drug-screening and gene therapy approaches to develop novel therapeutic avenues for treating symptoms of CDGs.