Clinical, biochemical and molecular correlations in MPS II patients
All of the data was available for the MPS II patients except for patients P9-P12 for whom only clinical and biochemical data were known because these patients died before their molecular data could be collected. The delay in diagnosis was explained by a lack of awareness among physicians of the specific MPS II clinical features associated with the adverse socioeconomic conditions of those patients.
Based on the clinical, biochemical, and molecular data, 6 patients (P1-P4 and P6-7) were classified in the MPS II group with a severe disease, and only one patient (P5) presented a mild phenotype.
According to the clinical data, the confirmation of diagnosis in all MPS II patients was done at a mean age of 5 years, unlike what is found in the literature 
The urinary GAG concentration ranged from 30.0 to 125 mg of creatinine, according to the age of each patient. The high level of heparan sulfate in the urine was correlated with the severity of the disease as previously described by Tomastsu S et al., who demonstrated a significant correlation between the level of heparan sulfate and the severity of this disease 
The leukocyte IDS activity in patients (P1-P4; P6-P8) with the severe type of the disease had a mean of 0.13 nmol/h/mg of proteins. Based on the high level of urinary GAGs and the deficiency of IDS activity, a relationship seems to exist between these data and the phenotypic expression of Hunter syndrome, contrasting with what is reported in the literature such as in Filipino patients . However, the clinical profiles of the MPS II patients (P1-P7) were in agreement with several studies described in the literature, and the clinical manifestations of the phenotype of Hunter syndrome ranged from moderate to severe Hunter syndrome phenotypes .
The most recurrent symptoms observed in this series ranged in degree of severity, including hepatosplenomegaly, coarse facial features including broad noses, macroglossia, psychomotor and mental retardation, multiple dysostoses including joint stiffness, oval vertebrae, respiratory problems including otitis, nasal obstruction, and enlarged tongue and adenoids.
Patient P8 has a female gender related to Patient P2 who was hemizygous for the p.R88P mutation. She presented GAG excretion of 125 mg/g/creatinine and leukocyte IDS activity of 1.00%. She died before molecular analysis was conducted, but she probably had the same genetic mutation as patient P2 since he presented the same clinical profile as her cousin P8. The existence of intrafamilial phenotypic heterogeneity suggests that the presence of genetic and epigenetic polymorphisms could be important to confirm their effects on the phenotypic expression of the disease. MPS II females have been noted to present very rare clinical descriptions, and most of them present the severe form of the disease . Importantly, the identification of MPS II heterozygous females by measurement of IDS activity and urinary GAG levels is unreliable. Therefore, the definitive diagnosis should be determined using genetic analysis 
Previous studies [15, 16] showed that the phenotypic expression of this disease in MPS II females is uncommon, and most of the cases described in the literature presented the severe phenotype. MPS II heterozygous females are rarely reported except for the presence of double mutant alleles or a coincidental genetic defect, leading to skewed X-inactivation or hemizygosity in heterozygotes 
Patient P12 was diagnosed at the age of three years old when he had an inguinal hernia operation. However, coarse facial features, including macrocrania, macroglossia and small teeth, had been noted at the age of eighteen months. He presented severe hepatosplenomegaly, skeletal disease, and severe mental retardation. The biochemical test showed that the leukocyte IDS activity in this patient was significantly higher (1.5 nmol/h/mg of proteins) than the enzyme activity of other MPS II patients. Patient P12 presented the severe phenotype of the MPS II disease, but he died before the molecular analysis hence the interest of carrier testing.
In this study, cardiovascular involvement, including arrhythmia and congestive heart failure, was identified in all MPS II patients and has been shown to be the cause of morbidity and mortality in most patients, as has been described previously in the literature .
Seven different mutations were found in the 12 MPS II patients. These nucleotides variations reflect the genetic heterogeneity leading to the wide spectrum of clinical phenotypes of MPS II in agreement with several other studies [4,12].
Sequence alterations in the IDS gene included five previously reported mutations and two novel mutations. The severe phenotype was found in patients who had the following mutations: c.240+1G>A, p.R88P, Ex1_7del, and p.Q396*. This in agreement with several previous studies (Table 1).
The missense mutation p.G94D was associated with a milder phenotype. This finding agrees with the data reported in Australian patients [4, 19]. This mutation occurred within a conserved amino acid of human lysosomal sulfatase, which is essential for the common sulfatase activity .
The p.R88P and p.G94D missense mutations, associated with the same polymorphisms were identified in two patients P2 and P5 whos presented two different phenotypes. The genotype heterogeneity in MPS II could be explained by the possible contribution of other genes in the severity of the phenotype in patient P2.
The first novel alteration p.Q204*(c.610C>T) was a nonsense mutation and was identified in a patient who developed a severe form of MPS II. This mutation was due to a cytosine -to- thymine transversion at position 610 of the cDNA resulting in premature glycopolypeptide truncation at the 204th codon in exon 5 of IDS gene. Carrier testing was performed in the mother, who was found conductive.
The second novel frame shift mutation (p.D450Nfs*95) in exon 9 of the IDS gene is caused by a single-base deletion of guanine at genomic DNA position 1565. This mutation in exon 9 changes codon 450 from aspartic acid (GAT) to a chain termination codon (TAG) that leads to the lack of 95 amino acids at the amino terminus of the IDS protein. This novel mutation may lead misfolding of the glycopeptide resulting in a non-functional protein.
Mutations leading to a premature translation codon have frequently been classified as severe mutations; in agreement with this, the novel frame shift (p.D450Nfs*95) mutation was found in a patient (P6) who presented the severe phenotype.
To probe effects of various amino acid substitutions on catalytic activity or stability of IDS protein, we located the mutations (p.Q240* and p.D450Nfs*95) on the IDS model and characterized their structural effects.
The p.D450Nfs*95 mutation results in exon skipping and introducing premature translation termination codon in exon nine with an abnormal IDS protein and have been classified as severe mutation. The premature stop codon causes a deletion of the last 5 amino acids of the heavy chain which contains the catalytic core (451🡪455) and the entire light chain (456🡪550) of IDS protein (Fig.3A). The predicted premature stop codon could affects protein stability www.pymol.org - PyMOL ; PyMol (pdb : 5FQL). In fact, the light chain of the IDS protein had an important role in the stability of the protein. Furthermore, the four antiparallel strands comprising the light chain are considerably longer than those of other sulfatases, and hence a greater contribution to the shape of the substrate-binding cleft comes directly from the light chain . The expected severity of this mutation was variable and consequence range from local destabilization and misfolding to global unfolding, leading to premature degradation. The K479 residue in the exon 9 was important to the substrate binding . The lack of this residue in our patient (P6) with p.D450Nfs*95 mutation result the non-functional IDS protein by the absence subtrate binding. Moreover, three frame shift mutations were described in the exon 9 of IDS gene: p.R443X, p.R443X, p.Y466X and found in the patients who presented severe phenotype [22 ; 23]. However, investigation of mRNA and expression studies will be necessary to prove this conclusively. Correlation between genotype and phenotype was uncertain using genomic DNA. Further investigations such as transcription tests are useful to predict with confidence the disease phenotype.
Several described mutations directly affect the catalytic core of the enzyme, for example by direct substitution of key active-site residues . In this study, the nonsense mutation p.Q204* was located on the light chain of the IDS protein, downstream of the catalytic core which contains 451-455 residus, destroys the enzyme activity and the β chain of the IDS protein (Fig.3B). Consistent with this concept, this mutation was associated with the severe form of MPS II observed in patient P7.
In this study, there was no relationship between the genotype and phenotype in these MPS II patients except for the significant correlation between the high level of urine GAGs and the severity of the disease. Further studies including a large number of cases in Tunisian population of the same age and genotype are needed in the first time, to confirm this correlation in MPS II patients and in the second time to screening of haplotyping data for the recurrent mutation because of a founder effect.