Some substitutions modified directly or indirectly, the binding of factor VII to tissular factor. In our patients we identified 3 missense mutations the novel p.F328Y and the two reported p.R304Q and p.M298I. The novel p.F328Y involves the substitution of a non polar amino acid (F) by a neutral amino acid (Y) in the catalytic domain of the factor VII which may be affects its affinity in binding to the tissular factor. Using PolyPhen, this substitution is predicted to be probably damaging with a score of 1.000 (sensitivity: 0.00; specificity: 1.00). Others studies demonstrate that the p.F328S mutation reduces tissue factor binding, impairs activation by factor Xa and abolishes the coagulant activities
[11, 12]. Since our novel mutation p.F328Y occurs in the same position as the reported mutation p.F328S, we hypothesis that the p.F328Y may be plays the same role by the reduction of tissue factor binding. The mutation p.F328S is reported in homozygous state and in compound heterozygous state and it is associated with severe phenotype
. In our patient the novel mutation p.F328Y is identified in heterozygous state which may be explain the moderate phenotype.
The reported mutations p.M298I and p.R304Q described in the literature, which occur also in the catalytic domain disrupt interactions between factor VII and tissular factor
The p.M298I identified in heterozygous state in our patient (patient 2) and also described with the same state
, is responsible for an inefficient activation of the catalytic site. This mutation is reported in asymptomatic patients when it is in the heterozygous state while in our patient it is associated with epistaxis, dental extraction, gingival and menorrhagia. This may be explained by the presence in heterozygous state of the 3 SNPs; M1M2, P0/P10 >and A1/A2 predicted to reduce the FVII activity
The p.R304Q is reported in homozygous and heterozygous state in many countries indicating that it is a frequent and recurrent mutation
[16, 17]. In our patient (patient 3) it is associated with homozygous state. Correlation genotype/phenotype for this mutation is in accordance with reported data since this mutation is associated with asymptomatic or paucisymptomatic patients
, which is the case for our patient who is asymptomatic.
The 2 splicing mutations IVS1a + 5 G > A and the p.G-39G identified in the heterozygous state respectively in patients 4 and 5, are associated with hemarthrosis, gingival, menorrhagia and epistaxis: clinical symptoms associated with heterozygous mutations described above
[19, 20]. For the patient 5 the presence of the SNPs A1/A2 known to decrease 25% of the FVII levels
 may also has an effect on the phenotype of our patient.
Among the 5 identified mutations, 3 are localized in the exon 8 (catatytic region) which represent the hot spot region of F7 gene as reported in literature
. Only the p.R304Q occurred within the CpG dinucleotides.
No mutations were found in 5 patients from 3 unrelated families with reduced FVII activities (2.5% to 5%) and they share all the wild alleles in the homozygous state concerning the M1, P0 and A1 SNPs. Only patient number 10 represent 7 repetitions of polymorphism H in exon 7 which may be responsible for the reduced FVII activity. Others data reported also the absence of molecular defect in some patients in their series
. The cause may be an intragenic rearrangement escaping our procedure, so we investigate the use of other techniques such as semi-quantitative multiplex PCR or MLPA.
A unique study described molecular pathology within FVII deficiency in Tunisian patients concern Tunisian Jewish
. In this study they identified the p.R304Q mutation which is also present in one of our patients despite their different origin, the p.R304Q is a recurrent mutation known in different ethnics groups and populations. The others mutations described in this study are different from our identified mutations which confirm the Arabic origin of our patients.
A few Tunisian studies were interested in the distribution of some polymorphisms (M1/M2 and P0/P10) in patients with FVII deficiency and healthy groups from different Tunisian regions demonstrate that the most frequent alleles are M1/M1 and P0/P0 in all the studied groups and also in groups from the North of Tunisia
[4, 5]. This distribution is confirmed in our patients who are from the North and in whom we identified only one allele M1/M2 in one patient/7 unrelated patients and we found the allele P0/P10 only in 2 patients. All the others patients are homozygous for the frequent alleles cited above.