Heterozygous ITGA2B Phe1024 Deletion Associated with Abnormal αIIbβ3 Function in a Patient with Congenital Thrombocytopenia

Integrin αIIbβ3 (previously called glycoprotein [GP] IIb/IIIa) is a transmembrane GP receptor, which plays a crucial role in blood coagulation by mediating platelet aggregation.[1] [2] After activation, this receptor initiates platelet aggregation rapidly through changing its conformation and introducing the interaction of platelets with a wide array of ligands such as fibrinogen, fibronectin, and von Willebrand factor (VWF). Germline mutations in genes encoding the subunits of the complex, integrin subunit α 2b (ITGA2B) and integrin subunit β 3 (ITGB3), have been reported mainly in the platelet-type bleeding disorder Glanzmann thrombasthenia (GT).[3] GT is a congenital disease that follows an autosomal recessive inheritance pattern and exhibits failure or deficiency of platelet aggregation, although with normal platelet counts and morphology.[4] However, several αIIbβ3 variants have also been identified in patients with congenital macrothrombocytopenia, a genetically heterogeneous group of rare disorders characterized by abnormal giant platelets, reduction of platelets, and bleeding tendency with variable severity.[5]

Our case is a 13-year-old boy who was referred to our hospital because of hemostatic difficulty after a first episode of epistaxis. His platelet count was 79 × 10⁹/L (normal range: 150–407 × 10⁹/L); the mean platelet volume (MPV) and platelet distribution width (PDW) were close to the upper limit of the reference intervals (MPV: 13.7fL, normal range: 9.0–13.8 fL, PDW: 21.5 fL, normal range: 9.8–21.9 fL). Six months later, the patient was referred to our hospital again due to sporadic petechia of skin. At that time, routine blood test of the patient showed mild reduction in platelet count (102 × 10⁹/L), with no abnormalities of MPV (12.9 fL) and PDW (18.6 fL). His blood cell smear identified a few giant platelets (4 in a standard automated smear in Mindray blood analyzer MC-80), while without obvious reduction of cytoplasmic granules ([Fig. 1A] and [B]). Except for a slight rise near the upper boundary of activated partial thromboplastin time (37.0 seconds; normal range: 25.1–36.5 seconds), other coagulation indicators, including prothrombin time (PT), prothrombin activity, fibrinogen, thrombin time (TT), and international normalized ratio, were all within normal ranges. VWF level and activity were also within normal limits. Platelet aggregation induced by adenosine diphosphate (ADP), collagen, adrenalin, or arachidonate were each substantially reduced ([Fig. 2A]), to varying degrees (ADP: 9.1%, collagen: 17.7%, adrenalin: 28.6%, arachidonate: 1.0%), whereas aggregation induced by ristomycin (ristocetin) was close to normal level (52.2%, [Fig. 2B]), only slightly lower than the bottom of normal range (55.0–100.0%). In addition, tests for autoimmune antibodies were negative for common anti-platelet antibodies, including GPIX (CD42a), GPIB (CD42b), GPIIB (CD41), GPIIIA (CD61), and GMP140 (CD61p). Detailed laboratory results are summarized in [Table 1].

Fig. 1 The morphology of platelets and blood cells of the proband. (A) Blood smear of the proband under a microscope at 3 times high-power field. (B) Blood smear of the proband under a microscope at 1,000 times oil mirror field showed rare giant platelets.

Fig. 2 Platelet aggregation curves induced by different stimulators. (A) Platelet aggregation induced by ADP, collagen, adrenalin, and arachidonate. (B) Platelet aggregation induced by ristomycin (ristocetin).

Table 1 Laboratory test results of the proband

Assays	Proband's results	Reference interval
Blood cell counts[a]
Platelet count, ×109/L	102.0	150.0–407.0
MPV, fL	12.9	9.0–13.8
PDW, fL	18.6	9.8–21.9
Thrombosis/hemostasis tests[b]
PT, s	12.3	9.4–12.5
Prothrombin activity, %	97	80–130
INR	1.04	0.85–1.15
Fibrinogen, g/L	2.52	2.38–4.98
aPTT, s	37.0	25.1–36.5
TT, s	15.1	10.3–16.6
VWF assays[c]
VWF:Ag (activity), %	111	50–200
VWF:Ag (level), %	105	50–200
VWF:Ag (activity)/VWF:Ag (level) ratio	1.06
Platelet aggregation[d]
ADP (6 μM), %	9.1	50.0–100.0
Collagen (5 mg/L), %	17.7	55.0–100.0
Adrenalin (180 μM), %	28.6	55.0–100.0
Arachidonate (0.3 g/L), %	1.0	55.0–100.0
Ristomycin (1.5 g/L), %	52.2	55.0–100.0

Abbreviations: ADP, adenosine diphosphate; aPTT, activated partial thromboplastin time; INR, international normalized ratio; MPV, mean platelet volume; PDW, platelet distribution width; PT, prothrombin time; TT, thrombin time; VWF, von Willebrand factor.

^a The analysis of blood routine test was accomplished on Mindray CAL8000 hematology system.

^b The coagulation indicators PT, aPTT, TT, and fibrinogen were determined by coagulation assays using corresponding STA reagents (Diagnostica Stago, France) on a STA MAX automatic coagulation analyzer (Diagnostica Stago, France) according to the manufacturer's instructions. The indicators prothrombin activity and INR were calculated based on the result of PT.

^c The VWF antigen activity and level were detected by immunoturbidimetric assay using HemosIL VWF activity or antigen kits on ACL TOP 700 automatic coagulation analyzer (Instrumentation Laboratory, United States).

^d The platelet aggregation tests were conducted by optical turbidimetric assay using Helena platelet aggregation kit on an Aggregation Remote Analyzer Module (Helena Laboratories, United States).

Investigation of family history of the propositus suggested potential familial history of mild bleeding tendency and thrombocytopenia ([Fig. 3A]). The mother of the proband had bleeding diathesis with a lifelong easy bruising and a history of recurrent epistaxis before 10 years old. A mild bleeding tendency was also reported for the maternal grandmother, maternal uncle, and maternal cousin of the propositus. Considering this familial history, we conducted genetic screening on the proband by sequencing a panel of genes related to hemorrhagic and clotting disorders caused by platelet dysfunction. Based on the image of platelet and results of platelet aggregation, we first excluded the possibility of storage pool disease and Bernard–Soulier syndrome. The normal VWF content and activity essentially ruled out von Willebrand disease. Finally, we included 17 candidate genes in our panel, including ITGA2B, ITGB3, MYH9 (myosin heavy chain 9), TBXA2R (thromboxane A2 receptor), TBXAS1 (thromboxane A synthase 1), MPL (MPL proto-oncogene, thrombopoietin receptor), MASTL (microtubule-associated serine/threonine kinase like), ACBD5 (acyl-CoA binding domain containing 5), ACTN1 (actinin α 1), ANKRD26 (ankyrin repeat domain containing 26), ANO6 (anoctamin 6), P2RY1 (purinergic receptor P2Y1), P2RY12 (purinergic receptor P2Y12), PLA2G4A (phospholipase A2 group IVA), RASGRP2 (RAS guanyl releasing protein 2), TUBB1 (tubulin β 1 class VI), and CYCS (cytochrome c, somatic). Potential germline mutations in the coding exons and intron–exon boundaries (±5 bp) of the target genes were initially discovered by next-generation sequencing using Ion Torrent Personal Genome Machine platform (Life Technologies). The clinical significance of the captured genetic variants were evaluated according to standards and guidelines of the American College of Medical Genetics and Genomics,[6] by integrating population frequency (recorded on databases like ExAC, 1000G, and gnomAD), clinical databases such as Human Gene Mutation Database and ClinVar database, and functional annotation resources including SIFT, polyphen-2, MutationTaster, and PROVEAN. Genetic variants with potential clinical significance (pathogenic, likely pathogenic, splicing variants, or novel variants) were further verified by Sanger sequencing.

Fig. 3 The mutations found in the proband and his family members. (A) Pedigrees of the proband and his family members. Affected individuals are indicated by filled symbols. The arrow indicates the proband. The triangle indicates the individual whose DNA was subjected to Sanger sequencing. (B) Sanger sequencing peaks of the identified mutation site (ITGA2B: c.3070–3072delTTC, p.Phe1024del). A total of six family members received Sanger sequencing validation. Among these, four members with mild bleeding tendency, including the proband, his mother, maternal grandmother, and maternal cousin, all harbored the same heterozygous small deletion, whereas this variant was absent in unaffected family members, the father and sister of the proband.

As a result, we identified a heterozygous small deletion of gene ITGA2B c.3070–3072delTTC, leading to a phenylalanine (Phe) deletion at amino acid 1024 (p.Phe1024del) of the encoded protein. This deletion was predicted to be deleterious and has been reported previously in a Japanese patient with macrothrombocytopenia (referred to as p.Phe993del in that report),[7] supporting its identification as the causative mutation in this family. Further gene detection of this mutation in some of the family members revealed that the relatives with bleeding tendencies, including the mother, grandmother, and maternal cousin of the proband, all harbored the same heterozygous small deletion. However, this variant was absent in unaffected family members, including the father and sister of the proband ([Fig. 3B]).

The p.Phe1024del in ITGA2B resides in a highly conserved juxtamembrane Gly–Phe–Phe–Lys–Arg sequence in a membrane proximal region of αIIbβ3. In addition to this variant, three germline mutations in this sequence, including p.Gly1022Cys (referred to as p.Gly991Cys in previous reports), p.Arg1026Gln (referred to as p.Arg995Gln in some reports), and p.Arg1026Trp (referred to as p.Arg995Trp in some reports), have been identified in patients with autosomal dominant congenital macrothrombocytopenia.[7] [8] [9] [10] By using SWISS-model, we predicted that the Phe1024 deletion would give rise to a weaker bridging of αIIb subunit to transmembrane compared with the wild type conformation ([Fig. 4A]), which could be the structural foundation of the dysfunction of αIIbβ3.[11] Impairment of surface αIIbβ3 expression has been consistently observed in variants of ITGA2B and ITGB3, especially in homozygous or compound heterozygous mutations. We thus further determined αIIbβ3 expression on surface of platelets through labeling the αIIb and β3 subunits for CD41a (Pharmingen PE mouse anti-human CD41a, BD Biosciences, United States) and CD61 (BD FITC mouse anti-human CD61, BD Biosciences, United States), respectively. Platelet-rich plasma samples were collected from family members with wild type and mutated ITGA2B and then examined with flow cytometry (FACSCanto flow cytometer, BD Biosciences, United States). As illustrated in [Fig. 4B], in a subject with wild type ITGA2B, we observed normal signals of both positive CD41a and CD61, which represent the externalized receptor αIIb and β3, respectively. However, in case of ITGA2B p.Phe1024del, an abnormal clustering of αIIbβ3 (∼7.7%) was observed ([Fig. 4B]). This subgroup of platelets showed positive CD61 but negative CD41a signals, indicating potential deficiency of αIIb subunit. According to previous studies, variants on membrane proximal regions of αIIbβ3 complex permit residual or even total αIIbβ3 expression but give rise to conformational changes that propagate through the integrin, leading to αIIbβ3 gain of function.[12] However, a description about a series of patients in 10 families revealed that constitutive αIIbβ3 activation only occurred in two out of nine patients.[13] Intriguingly, in a pedigree with macrothrombocytopenia, Miyashita et al reported that a heterozygous small deletion in membrane proximal region of β3 (ITGB3 p.T720del) induced spontaneous activation of integrin αIIbβ3, but with downregulated platelet expression of αIIbβ3 and impaired aggregation.[14] In line with this study, we observed a subset of defective αIIb, which would contribute to aggregation dysfunction. However, whether the residual αIIbβ3 exerted more active or gain of function needs further exploration. We speculated that the aberrant clustering of αIIbβ3 may interfere with proper proplatelet formation and cause thrombocytopenia and GT-like syndrome of the patient. Nevertheless, the underlying mechanism remains to be demonstrated.

Fig. 4 Function prediction of the mutation and platelet αIIbβ3 expression detection. (A) Structural in silico modeling of the wild type ITGA2B (top) and p.Phe1024del mutated ITGA2B (bottom) by SWISS-MODEL. (B) Expression of platelet membrane glycoprotein GPIIB (CD41a) and GPIIIA (CD61) in individual with wild type ITGA2B (top) or p.Phe1024del mutated ITGA2B (bottom) by flow cytometry.

Homozygous or compound heterozygous germline mutations in ITGA2B and ITGB3 have been well recognized as cause of GT, a bleeding disorder characterized by normal platelet count but abnormal platelet function. In addition to the normal platelet count, the other typical laboratory phenotype of GT is the decrease or absence of platelet aggression with all agonists except ristocetin.[15] The platelet count of our case was decreased and although platelet aggregation with all agonists except ristocetin was reduced (10–30%), it was not reduced to the degree usually observed in GT patients (<10% with all physiological agonists, together with a normal agglutination response to ristocetin).[15] Unexpectedly, several variants in membrane proximal regions of αIIbβ3 have also been identified in congenital macrothrombocytopenia, mostly in Japanese individuals.[7] [9] [13] [14] Kunishima and other researchers have demonstrated that gain-of-function mutations around juxtamembrane region of αIIbβ3 led to a highly activated conformation of αIIbβ3 and further induced activation of outside-in signaling like Focal adhesion kinase.[9] [10] The variant ITGA2B p.Phe1024del, which is located in this region, was previously reported in a 9-year-old Japanese boy and was thought to result in activation of the αIIbβ3 receptor.[7] The blood routine tests of the patient in that report indicated macrothrombocytopenia (thrombocytopenia with large platelet size), while there was no episode of bleeding tendency. In contrast to that patient, the family in our study is of Chinese ancestry and exhibits thrombocytopenia with normal platelet size. More importantly, the variant carriers in this family manifested mild bleeding tendency. Considering the platelet morphology, results of aggregation assays, platelet GP expression, and clinical features, we are inclined to believe that the Phe1024 deletion in ITGA2B gives rise to αIIbβ3 dysfunction and causes thrombocytopenia with GT-like phenotype. To our knowledge, our study is the first report about a family of autosomal dominant thrombocytopenia with normal platelet size caused from a small deletion in ITGA2B in Chinese. Coincidentally, Khoriaty et al reported a similar thrombocytopenia family of European ancestry without giant platelets, resulting from p.Arg1026Trp substitution in ITGA2B.[16] Interestingly, this substitution has been reported in several Japanese families/patients with macrothrombocytopenia and all of these patients had increased platelet size. Taken together, we think the lack of macrothrombocytopenia in our investigated family might be attributed to the modification effects of genetic variant in different populations with disparate genetic background.

In summary, we found a germline mutation ITGA2B p.Phe1024del in a Chinese family with thrombocytopenia. Our report supports recent findings that heterozygous variants in membrane proximal regions of integrin αIIb or β3 represent an etiology of a subset of macrothrombocytopenia and additionally extends perspectives regarding diverse laboratory and clinical phenotypes associated with the same variant in different genetic background.

Ethical Approval

The study protocol was approved by the Tongji Hospital Ethics Committee for Research in Health. Informed consent was obtained from individuals included in this study.

Authors' Contributions

J.L. and N.T. designed the research study, B.W. and J.L. performed the research and wrote the paper, H.H. contributed essential reagents and analysis of flow cytometry, J.C. and X.W. analyzed the data. All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.

Publication History

Article published online:
11 April 2024

• References

• 1 Coller BS. αIIbβ3: structure and function. J Thromb Haemost 2015; 13 (01) 17-25
• 2 Huang J, Li X, Shi X. et al. Platelet integrin αIIbβ3: signal transduction, regulation, and its therapeutic targeting. J Hematol Oncol 2019; 12 (01) 26
• 3 Nurden AT, Pillois X, Wilcox DA. Glanzmann thrombasthenia: state of the art and future directions. Semin Thromb Hemost 2013; 39 (06) 642-655
• 4 Nurden AT, Fiore M, Nurden P. et al Glanzmann thrombasthenia: a review of ITGA2B and ITGB3 defects with emphasis on variants, phenotypic variability, and mouse models. Blood 2011; 118 (23) 5996-6005
• 5 Kunishima S, Saito H. Congenital macrothrombocytopenias. Blood Rev 2006; 20 (02) 111-121
• 6 Richards S, Aziz N, Bale S. et al; ACMG Laboratory Quality Assurance Committee. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med 2015; 17 (05) 405-424
• 7 Kashiwagi H, Kunishima S, Kiyomizu K. et al. Demonstration of novel gain-of-function mutations of αIIbβ3: association with macrothrombocytopenia and Glanzmann thrombasthenia-like phenotype. Mol Genet Genomic Med 2013; 1 (02) 77-86
• 8 Peyruchaud O, Nurden AT, Milet S. et al. R to Q amino acid substitution in the GFFKR sequence of the cytoplasmic domain of the integrin IIb subunit in a patient with a Glanzmann's thrombasthenia-like syndrome. Blood 1998; 92 (11) 4178-4187
• 9 Kunishima S, Kashiwagi H, Otsu M. et al. Heterozygous ITGA2B R995W mutation inducing constitutive activation of the αIIbβ3 receptor affects proplatelet formation and causes congenital macrothrombocytopenia. Blood 2011; 117 (20) 5479-5484
• 10 Favier M, Bordet JC, Favier R. et al. Mutations of the integrin αIIb/β3 intracytoplasmic salt bridge cause macrothrombocytopenia and enlarged platelet α-granules. Am J Hematol 2018; 93 (02) 195-204
• 11 Waterhouse A, Bertoni M, Bienert S. et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res 2018; 46 (W1): W296-W303
• 12 Kim C, Lau TL, Ulmer TS. et al Interactions of platelet integrin alphaIIb and beta3 transmembrane domains in mammalian cell membranes and their role in integrin activation. Blood 2009; 113 (19) 4747-4753
• 13 Morais S, Oliveira J, Lau C. et al. αIIbβ3 variants in ten families with autosomal dominant macrothrombocytopenia: expanding the mutational and clinical spectrum. PLoS One 2020; 15 (12) e0235136
• 14 Miyashita N, Onozawa M, Hayasaka K. et al. A novel heterozygous ITGB3 p.T720del inducing spontaneous activation of integrin αIIbβ3 in autosomal dominant macrothrombocytopenia with aggregation dysfunction. Ann Hematol 2018; 97 (04) 629-640
• 15 Botero JP, Lee K, Branchford BR. et al; ClinGen Platelet Disorder Variant Curation Expert Panel. Glanzmann thrombasthenia: genetic basis and clinical correlates. Haematologica 2020; 105 (04) 888-894
• 16 Khoriaty R, Ozel AB, Ramdas S. et al. Genome-wide linkage analysis and whole-exome sequencing identifies an ITGA2B mutation in a family with thrombocytopenia. Br J Haematol 2019; 186 (04) 574-579