The genetic landscape of male infertility is highly complex. It is estimated that at least 4000 genes are involved in human spermatogenesis, but only few have so far been extensively studied. In this study, we investigated by whole exome sequencing two cases of idiopathic non-obstructive azoospermia (NOA) due to severe hypospermatogenesis. After variant filtering and prioritizing, we retained for each patient a homozygous loss-of-function (LoF) variant in a testis-specific gene, C1orf185 (c.250C>T; p.Gln84Ter) and CCT6B (c.615-2A>G), respectively. Both variants are rare according to the gnomAD database and absent from our local control cohort (n = 445). To verify the implication of these candidate genes in NOA, we used the CRISPR/Cas9 system to invalidate the mouse orthologs 4930522H14Rik and Cct6b and produced two knockout (KO) mouse lines. Sperm and testis parameters of homozygous KO adult male mice were analyzed and compared with those of wild-type animals. We showed that homozygous KO males were fertile and displayed normal sperm parameters and a functional spermatogenesis. Overall, these results demonstrate that not all genes highly and specifically expressed in the testes are essential for spermatogenesis, and in particular, we conclude that bi-allelic variants of C1orf185 and CCT6B are most likely not to be involved in NOA and male fertility.