Abstract Study question What is the functional relevance of GLUD2, FAM47A, and TAF7L in spermatogenesis and male infertility? Summary answer Knockdown of orthologues leads to infertility in male flies substantiating GLUD2, FAM47A, and TAF7L as candidate genes for human male infertility. What is known already Non-obstructive azoospermia (NOA) is clinically the most severe form of spermatogenic failure and male infertility. Despite advances in the genetic diagnostics, the precise cause and molecular diagnosis remains elusive for the majority of affected men. GLUD2, FAM47A, and TAF7L are X-linked and testis-enriched genes, with increasing evidence of their relevance in male infertility. However, so far, no functional study or animal model supporting their contribution to the disease exist. Study design, size, duration We queried exome/genome data of more than 2, 300 infertile men with severe quantitative spermatogenic failure from the Male Reproductive Genomics (MERGE) cohort for variants in GLUD2, FAM47A, and TAF7L. To assess the impact of these three genes on male infertility we performed functional studies using Drosophila melanogaster as a model organism. Participants/materials, setting, methods Exome/genome data was filtered for rare (MAF <0.001, gnomAD), hemizygous loss-of-function (LoF) and potentially pathogenic missense variants (CADD score >20) in GLUD2 (NM_012084.4), FAM47A (NM_203408.3), and TAF7L (NM_024885.4). Depending on sample availability, the X-linked inheritance pattern was confirmed by segregation analysis. To investigate the relevance of these genes in male infertility, we generated and functionally characterised testis-specific knockdown (KD) of orthologues in Drosophila melanogaster using GAL4-UAS system. Main results and the role of chance In GLUD2, we identified four missense and one LoF variant in five azoospermic men. All variant carriers revealed impaired spermatogenesis with predominant meiotic arrest (MeiA). In FAM47A, we identified three missense and one LoF variant in five azoospermic men, four of whom consistently exhibited a testicular phenotype of Sertoli cell-only (SCO). In TAF7L, we identified four missense variants in three azoospermic men. One of them was a carrier of two variants. Notably, all the identified variants were clustered in the mutation hotspot region within exons 9, 10, and 13. Testis-specific KD of GLUD2, FAM47A, and TAF7Lorthologues consistently resulted in complete sterility in male flies. Compared to controls, Glud2-KD flies had normal testicular morphology but revealed small seminal vesicles without mature sperm. Testicular squash analysis showed germ cell maturation arrest at the elongating spermatid stage, leading to individualisation complex ablation. Immunofluorescence staining revealed a lack of cystic bulges and waste bags, indicating disrupted spermatogenesis. In contrast, Fam47a-KD and Taf7l-KD flies were infertile due to total germ cell loss in the testis, resulting in an extremely small testis size compared with controls, consistent with the SCO testicular phenotype observed in infertile men with FAM47A and TAF7L variants. Limitations, reasons for caution The main limitation is that the study does not analyse a gene-specific genotype-phenotype correlation of the missense variants identified. Furthermore, extensive research is required to elucidate the underlying molecular mechanisms by which GLUD2, FAM47A, and TAF7L regulate male fertility in humans. Wider implications of the findings This study highlights the utility of Drosophila for studying the role of human male fertility genes, underlining the importance of validating genetic findings. Additionally, it contributes significantly to advancing our understanding of the genetic basis of azoospermia, offering potential targets for research aimed at addressing male infertility. Trial registration number not applicable