Molecular genetic diagnosis of Glanzmann syndrome in Iranian population; reporting novel and recurrent mutations

Glanzmann thrombasthenia (GT, OMIM: 273800) is a rare inherited disorder of platelet function, caused by quantitative or qualitative defects of the platelet membrane glycoprotein (GP) IIb-IIIa (αIIbβ3) complex [1]. Patients have lifelong mucosal bleeding and may require platelet transfusions for severe bleeding episodes. Bleeding complications are also frequent after dental extraction, surgery, or delivery [2]. GT is a rare disorder worldwide but is relatively frequent in populations such as Iraqi, Jews, Jordanian Arabs, Iran, Palestine, South Indians and French Gypsies with a high frequency of consanguineous marriages [3, 4].

The molecular basis of GT is linked to quantitative and/or qualitative abnormalities of αIIbβ3 integrin that mediates binding of the adhesive proteins to attach aggregating platelets and ensure thrombus formation at sites of injury in blood vessels. GT is classified into 3 subtypes, depending on the level of αIIbβ3. Affected individuals show mucocutaneous bleeding with the absent of platelet aggregation in response to all physiologic stimuli. They show normal platelet count and morphology. In addition, platelet αIIbβ3 deficiency or dysfunction should always be confirmed.

Laboratory tests in severe GT show no platelet aggregation in response to all physiologic agonists and show reduced or absent clot reaction [5]. Patients with type I or classical GT have a virtual absence of αIIbβ3 complex (< 5% of normal). In type II GT, patients have up to 25% of the normal level of the complex. In the variant type, αIIbβ3 level is near normal but functionally impaired, leading to defective binding of fibrinogen [6, 7].

The αIIbβ3 complex integrin is encoded by ITGA2B and ITGB3 genes (GPIIb and GPIIIa respectively). In brief, integrin synthesis occurs in the megakaryocytes with the αIIbβ3 complex formation in the endoplasmic reticulum (ER). The pre-GPIIb-IIIa complex is then transported to the Golgi apparatus, where sugar modification and cleavage of GPIIb into a heavy (M, 125,000) and light chain (M, 23,000) takes place [5]. GPIIb heavy and light chains remain covalently linked by a single disulfide bond. Following these processing events within the Golgi, the mature GPIIb-IIIa complex is then transported to the cell surface, where it exists in a resting state awaiting cellular activation. Noncomplexed or incorrectly folded gene products fail to undergo processing in the Golgi apparatus and are rapidly degraded intra-cellularly.

A continually updated mutation database is available on the sinaicentral.mssm.edu/intranet/research/glanzmann website [3, 4]. The molecular and functional characterization of the mutations has provided important insights to the better understanding of the biosynthesis and structure-function relationships of the αIIbβ3complex with the disease. It also adds important knowledge about the biology of other molecules of the integrin family [8, 9].

Genetic testing in GT patients is very important in the quality of life of the affected individuals and it facilitates prenatal diagnosis (PND) or pre-implantation genetic diagnosis (PGD) for at-risk families. The identified mutations will span the current database of the mutations of GT.

The study included sixteen unrelated families with the diagnosis of GT. As it was expected most of the patients were the result of consanguineous marriages (14/16 families, 87%). Sequencing analysis revealed totally 15 different mutations in the ITGB3 and the ITGA2B genes (Table 1). All identified mutations were homozygous except one in the ITGB3, which was in the form of compound heterozygous. Mutations in the ITGB3 gene consist 7 different mutations including 4 missense mutations [c.2T > C, c.155 G > T, c. 538 G > A, c.1990 G > T], a nonsense mutation [c.1303 G > T], one small 3 nt deletion [c.1656_1658delCTC] and a deletion of a single nt [c.401delA]. Interestingly all identified mutations in this gene were novel (Table 1).

The identified variants were heterozygote in parents and were not present as homozygote in their healthy members of the family as expected in cases where mutations are deleterious. To further confirm the pathogenicity of the missense variants, we checked the mutation in 100 random people and they were normal which supports the pathogenicity of the novel variants.

In a family, while sequencing exons 2 and 3, we could not get any band in the patients suggesting that theses exons may be deleted. In order to characterize the deletion breakpoints, primer walking and sequencing were performed. While the size of DNA fragment in normal cases was nearly 1500 bp as expected but the size of DNA product from the patient was around 1200 bp. Subsequent sequencing of this fragment confirmed deletion of 367 bp covering 319 nt from intron 1 and 48 nt in the second exon (Fig. 1).

GT is a rare congenital bleeding syndrome, which results in deficient platelet aggregation due to the absence or dysfunction of the fibrinogen receptor αIIbβ3 [12]. However, GT is rare worldwide but is relatively frequent in inbred populations such as Iraqi-Jews, Indians, and Arabs in Jordan, Israel, and Saudi Arabia. The exact number of GT patients in Iran has not been clearly reported. But it seems that the disease incidence will be higher from the estimated 1 in 200,000 in the US population. Because Iran is a country with nearly 38.4% rate of consanguineous marriage and at the other hand autosomal recessive mode of inheritance of the disease will add to the number of patients [13]. To the best of our knowledge, this study is the first comprehensive study on the molecular aspects of GT in Iranian population. Needless to say there is another similar study carried out by Kazemi et al. [14] on Iranian population but only three deleterious mutations have been reported out of 20 families affected with GT. However, none of the mutations reported by them is repeated in our study.

In Table 1 we have provided types of mutations and their effects at the amino acid level. Further in silico analysis was performed (Table 2) to find how these changes can be pathogenic. Obvious ones will not be discussed.

The first patient (Table 1) with c.2T > C point mutation which is located at the very beginning of the first exon of the ITGB3 gene (β3 subunit,) will change the methionine to threonine. Initiator methionine is very important for the initiation of protein synthesis.

Regarding the third patient with c.401delA mutation in the ITGB3, and based on mutation taster prediction [15] it is disease causing change. The fourth patient with c.1303G > T [p. Glu435Ter] mutation, causes change of glutamic acid to a termination codon leading to premature stop codon with only 434 residues which most probable will not function properly. The other patient had c.155 G > T mutation in the ITGB3 gene, changing cysteine into phenylalanine. The wild-type residue is involved in a cysteine bridge formation, which is important for the stability of the protein. Only cysteine can make these types of bonds, so the mutation causes loss of this interaction and will have a severe effect on the 3D-structure of the protein. Together with the loss of the cysteine bond, the differences between old and new residue can cause destabilization of the structure [16], This prediction is also supported by SIFT and Polyphen results. The c. 538 G > A [p. Gly180Arg] mutation in the ITGB3 changes glycine into arginine and introduces an amino acid with different properties that can disturb this domain and abolish its function. Because the wild-type residue, glycine, is the most flexible of all residues. Mutation of this glycine can abolish this function. This flexibility might be necessary for the protein’s function and by this mutation the torsion angles for this residue are unusual [16]. This prediction is also compatible with the SIFT and Polyphen prediction (Table 2).

The deletion of T nucleotide in exon 19 of ITGA2B gene (i.e. c.1899 del T [p. Cys633TrpfsTer17]) causes a frameshift in protein structure. HSF (Human splice finder) predicted that this frameshift created an exonic ESS site that has a potential to alter the splicing [17]. Regarding c.409–3 C > G and c.1392 A > G mutations in the ITGA2B which occurred near the potential splice sites of the protein, this can alter the wild type acceptor and donor site respectively. The c.189-319_236del mutation at ITGA2B gene leads to the elimination later part of the intron one and the beginning part of second exon. This deletion, therefore, deletes the splice acceptor site. This in turn should cause deletion of the entire exon 2 because there is no alternative splice site within the remaining part of exon 2 and therefore the remaining part of this exon will be eliminated via splicing of intron 1 (donor site), joining of exon1 and exon3. Therefore it is expected that through this splicing the exon 1 will be joined with exon 3. Figure 1 shows that if this would be the case then amino acids 63rd to 103rd will be lost and a frameshift would be resulted. The frameshift mutation leads to a truncated protein which loss of function would be resulted. Since we could not obtain fresh blood from the patient to perform RNA analysis, the above conclusion can be provided. However, whatever the consequence at the molecular level, we can claim that resulting effect at the protein level is having damaging effect on the patient.

It should be added that the identified variants were not present in the single nucleotide polymorphism database (Exome Variant Server, http://​evs.​gs.​washington.​edu/​EVS/​).

Regarding phenotype-genotype correlation, more studies with larger sample size is recommended.

It is interesting to see so many novel mutations in a rather small number of families (14/16) indicating several independent founder mutations. If we combine other mutations reported by others (Kazemi et al.) with ours it might become even more interesting. This may mean that either these genes are highly mutable (which may not be true since the disease is not so prevalent), or Iranian population is so diverse. This diversity has many historical proofs, which are beyond the scope of this paper but it worth mentioning.

In conclusion, our investigation of 16 patients, and in some cases including their families, revealed a number of different mutations within the αIIb and β3 subunits associated with causing GT. This vast number of different mutations in different provinces of Iran indicates the genetic heterogeneity in the studied group as well as in GT in general. Further investigation on the prevalence of each mutation from each ethnicity may reveal different picture.