Linking human Dead end 1 (DND1) variants to male infertility employing zebrafish embryos.

Westerich, Kim Joana; Reinecke, Solveig; Emich, Jana; Wyrwoll, Margot Julia; Stallmeyer, Birgit; Meyer, Matthias; Oud, Manon S; Fietz, Daniela; Pilatz, Adrian; Kliesch, Sabine; Reichman-Fried, Michal; Tarbashevich, Katsiaryna; Limon, Tamara; Stehling, Martin; Friedrich, Corinna; Tüttelmann, Frank; Raz, Erez

Abstract

Study question: Is the vertebrate protein Dead end (DND1) a causative factor for human infertility and can novel in vivo assays in zebrafish help in evaluating this? Summary answer: Combining patient genetic data with functional in vivo assays in zebrafish reveals a possible role for DND1 in human male fertility. What is known already: About 7% of the male population is affected by infertility but linking specific gene variants to the disease is challenging. The function of the DND1 protein was shown to be critical for germ cell development in several model organisms but a reliable and cost-effective method for evaluating the activity of the protein in the context of human male infertility is still missing. Study design, size, duration: Exome data from 1305 men included in the Male Reproductive Genomics cohort were examined in this study. A total of 1114 of the patients showed severely impaired spermatogenesis but were otherwise healthy. Eighty-five men with intact spermatogenesis were included in the study as controls. Participants/materials, setting, methods: We screened the human exome data for rare, stop-gain, frameshift, splice site, as well as missense variants in DND1. The results were validated by Sanger sequencing. Immunohistochemical techniques and, when possible, segregation analyses were performed for patients with identified DND1 variants. The amino acid exchange in the human variant was mimicked at the corresponding site of the zebrafish protein. Using different aspects of germline development in live zebrafish embryos as biological assays, we examined the activity level of these DND1 protein variants. Main results and the role of chance: In human exome sequencing data, we identified four heterozygous variants in DND1 (three missense and one frameshift variant) in five unrelated patients. The function of all of the variants was examined in the zebrafish and one of those was studied in more depth in this model. We demonstrate the use of zebrafish assays as a rapid and effective biological readout for evaluating the possible impact of multiple gene variants on male fertility. This in vivo approach allowed us to assess the direct impact of the variants on germ cell function in the context of the native germline. Focusing on the DND1 gene, we find that zebrafish germ cells, expressing orthologs of DND1 variants identified in infertile men, failed to arrive correctly at the position where the gonad develops and exhibited defects in cell fate maintenance. Importantly, our analysis facilitated the evaluation of single nucleotide variants, whose impact on protein function is difficult to predict, and allowed us to distinguish variants that do not affect the protein's activity from those that strongly reduce it and could thus potentially be the primary cause for the pathological condition. These aberrations in germline development resemble the testicular phenotype of azoospermic patients. Limitations, reasons for caution: The pipeline we present requires access to zebrafish embryos and to basic imaging equipment. The notion that the activity of the protein in the zebrafish-based assays is relevant for the human homolog is well supported by previous knowledge. Nevertheless, the human protein may differ in some respects from its homologue in zebrafish. Thus, the assay should be considered only one of the parameters used in defining DND1 variants as causative or non-causative for infertility. Wider implications of the findings: Using DND1 as an example, we have shown that the approach described in this study, relying on bridging between clinical findings and fundamental cell biology, can help to establish links between novel human disease candidate genes and fertility. In particular, the power of the approach we developed is manifested by the fact that it allows the identification of DND1 variants that arose de novo. The strategy presented here can be applied to different genes in other disease contexts. Study funding/competing interest(s): This study was funded by the German Research Foundation, Clinical Research Unit, CRU326 'Male Germ Cells'. There are no competing interests. Trial registration number: N/A.