Indeed, numerous association studies have suggested the involvement of various genes in human MTLE but data have mainly shown negative or conflicting. There is growing evidence for an important role of the complement system in CNS development and functioning and for the involvement of the complement system and particularly of C3 dysregulation, in the epilepsies – including the MTLE. Hence, the C3 gene represented a good candidate for the genetic susceptibility to human MTLE. In the present study, we identified a novel functional, regulatory CA-repeat polymorphism within the promoter region of C3 and aimed at evaluating the possibility of genetic association between human MTLE and GF100472-related haplotypes. Genetic contribution to human MTLE is well recognized but little is known about the genes participating in the underlying pathophysiology. While there had been previous arguments for dysregulation of the complement system in various epileptic models and in the human MTLE particularly, the genetic influence of the complement system on the risk for epilepsy and seizures had not been suggested so far. In the present study, replicated genetic data and functional analysis suggest that a newly-identified and functional dinucleotide polymorphism in the complement component C3 promoter influences genetic susceptibility to human MTLE with a history of FS, and to pure FS. One rare C3 haplotype that encompasses the GF100472 shortest length allele, protected against MTLE-FS+ in two independent populations of ONX-0914 patients vs controls. Another rare C3 haplotype was also found protective in an independent population of pure FS and was also much less frequent in MTLE-FS+ patients than in controls. Moreover, the GF100472 allele 11 contained within HAP5, was also protective by itself in a replication study composed by novel series of FS patients and controls. Noteworthingly, HAP4 did not show any significant difference in an independent population of pure FS compared to matched controls. Several reasons may account for this latter data. The first one is a difference of power between the studies, hence masking some of the statistical effects. Different genetic contributions to FS developed earlier in life by patients with MTLE, and to FS displayed by the general population, might represent another and non exclusive possibility. MTLE-FS+ patients are very often considered as having complex FS while FS in the general population would represent a more balanced mix of simple and complex FS. Indeed, in the present study 19 out of the 57 MTLEFS+ patients had complex FS as compared to 14 out of 97 FS patients. On the one hand, HAP4 could be specific to complex FS of MTLE patients and hence its effect would be diluted in the FS general population. For instance, HAP4 could influence a sequence of events leading from FS to MTLE rather than directly influencing the risk of FS. On the other hand, HAP5 – including 11 – might protect against FS and against MTLE as a shared susceptibility factor.