A murine model of hnRNPH2-related neurodevelopmental disorder recapitulates clinical features of human disease and reveals a mechanism for genetic compensation of HNRNPH2

Abstract

Mutations in HNRNPH2 cause an X-linked neurodevelopmental disorder with a phenotypic spectrum that includes developmental delay, intellectual disability, language impairment, motor function deficits, and seizures. More than 90% of patients with this disorder have a missense mutation within or adjacent to the nuclear localization signal (NLS) of hnRNPH2, although the specific pathogenic consequences of these mutations have not been examined. Here we found that hnRNPH2 NLS mutations result in reduced interaction with the nuclear transport receptor Kapβ2 in vitro and in cultured human cells. These mutations also cause modest accumulation of hnRNPH2 in the cytoplasm, suggesting that mislocalization of the protein might contribute to pathogenesis. We generated two knock-in mouse models with human-equivalent mutations in the endogenous mouse gene Hnrnph2, as well as Hnrnph2 knockout (KO) mice, and subjected them to extensive, deep phenotyping. Mutant knock-in mice displayed a spectrum of phenotypes that recapitulated aspects of the human disorder, including reduced survival in males, craniofacial abnormalities, impaired motor functions, and increased susceptibility to audiogenic seizures. Mutant knock-in male mice developed more severe phenotypes than female mice, likely due to differences in X-chromosome gene dosage. In contrast, two independent lines of Hnrnph2 KO mice showed no detectable phenotypes. Notably, KO mice had upregulated expression of Hnrnph1, a close paralog of Hnrnph2, whereas mutant Hnrnph2 knock-in mice failed to upregulate Hnrnph1. Thus, genetic compensation by Hnrnph1 might be sufficient to counteract the loss of hnRNPH2. These findings suggest that the pathogenesis of HNRNPH2-related disorder in humans may be driven by a toxic gain of function or a complex loss of HNRNPH2 function with impaired compensation by HNRNPH1. The carefully phenotyped mutant knock-in mice described here are an important resource for preclinical studies to assess the potential benefit of either gene replacement or therapeutic knockdown of mutant hnRNPH2.