STUDY QUESTION: Can we identify new sequence variants in the aurora kinase C gene (AURKC) of patients with macrozoospermia and establish a genotype-phenotype correlation? SUMMARY ANSWER: We identified a new non-sense mutation, p.Y248*, that represents 13% of all mutant alleles. There was no difference in the phenotype of individuals carrying this new mutation versus the initially described and main mutation c.144delC. WHAT IS KNOWN ALREADY: The absence of a functional AURKC gene causes primary infertility in men by blocking the first meiotic division and leading to the production of tetraploid large-headed spermatozoa. We previously demonstrated that most affected men were of North African origin and carried a homozygous truncating mutation (c.144delC). STUDY DESIGN, SIZE, DURATION: This is a retrospective study carried out on patients consulting for infertility and described as having >5% large-headed spermatozoa. A total of 87 patients are presented here, 43 patients were published previously and 44 are new patients recruited between January 2008 and December 2011. PARTICIPANTS/MATERIALS, SETTING, METHODS: All patients consulted for primary infertility in fertility clinics in France (n = 44), Tunisia (n = 30), Morocco (n = 9) or Algeria (n = 4). Sperm analysis was carried out in the recruiting fertility clinics and all molecular analyses were performed at Grenoble teaching hospital. DNA was extracted from blood or saliva and the seven AURKC exons were sequenced. RT-PCR was carried out on transcripts extracted from leukocytes from one patient homozygous for p.Y248*. Microsatellite analysis was performed on all p.Y248* patients to evaluate the age of this new mutation. MAIN RESULTS AND THE ROLE OF CHANCE: We identified a new non-sense mutation, p.Y248*, in 10 unrelated individuals of European (n = 4) and North African origin (n = 6). We show that this new variant represents 13% of all mutant alleles and that the initially described c.144delC variant accounts for almost all of the remaining mutated alleles (85.5%). No mutated transcripts could be detected by RT-PCR suggesting a specific degradation of the mutant transcripts by non-sense mediated mRNA decay. A rare variant located in the 3' untranslated region was found to strictly co-segregate with p.Y248*, demonstrating a founding effect. Microsatellite analysis confirmed this linkage and allowed us to estimate a mutational age of between 925 and 1325 years, predating the c.144delC variant predicted by the same method to have arisen 250-650 years ago. Patients with no identified AURKC mutation (n = 15) have significantly improved parameters in terms of vitality and concentration of normal spermatozoa, and a decreased rate of spermatozoa with a large head and multiple flagella (P < 0.001). LIMITATIONS, REASONS FOR CAUTION: Despite adherence to the World Health Organization guidelines, large variations in most characteristic sperm parameters were observed, even for patients with the same homozygous mutation. We believe that is mainly related to inter-laboratory variability in sperm parameter scoring. This prevented us from establishing clear-cut values to indicate a need for molecular analysis of patients with macrozoospermia. WIDER IMPLICATIONS OF THE FINDINGS: This study confirms yet again the importance of AURKC mutations in the aetiology of macrozoospermia. Although a large majority of patients are of North African origin, we have now identified European patients carrying a new non-sense mutation indicating that a diagnosis of absence of a functional AURKC gene should not be ruled out for non-Magrebian individuals. Indirect evidence indicates that AURKC might be playing a role in the meiotic spindle assembly checkpoint (SAC) during meiosis. We postulate that heterozygous men might have a more relaxed SAC leading to a more abundant sperm production and a reproductive advantage. This could be the reason for the rapid accumulation of the two AURKC mutations we observe in North African individuals. STUDY FUNDING/COMPETING INTEREST(S): None of the authors have any competing interest. This work is part of the project 'Identification and Characterization of Genes Involved in Infertility (ICG2I)' funded by the programme GENOPAT 2009 from the French Research Agency (ANR).