In silico analysis of structural modifications in and around the integrin αIIb genu caused by ITGA2B variants in human platelets with emphasis on Glanzmann thrombasthenia

Abstract

Background: Studies on the inherited bleeding disorder, Glanzmann thrombasthenia (GT), have helped define the role of the αIIbβ3 integrin in platelet aggregation. Stable bent αIIbβ3 undergoes conformation changes on activation allowing fibrinogen binding and its taking an extended form. The αIIb genu assures the fulcrum of the bent state. Our goal was to determine how structural changes induced by missense mutations in the αIIb genu define GT phenotype.

Methods: Sanger sequencing of ITGA2B and ITGB3 in the index case followed by in silico modeling of all known GT-causing missense mutations extending from the lower part of the β-propeller, and through the thigh and upper calf-1 domains.

Results: A homozygous c.1772A>C transversion in exon 18 of ITGA2B coding for a p.Asp591Ala substitution in an interconnecting loop of the lower thigh domain of αIIb in a patient with platelets lacking αIIbβ3 led us to extend our in silico modeling to all 16 published disease-causing missense variants potentially affecting the αIIb genu. Modifications of structuring H-bonding were the major cause in the thigh domain although one mutation gave mRNA decay. In contrast, short-range changes induced in calf-1 appeared minor suggesting long-range effects. All result in severe to total loss of αIIbβ3 in platelets. The absence of mutations within a key Ca2+-binding loop in the genu led us to scan public databases; three potential single allele variants giving major structural changes were identiffied suggesting that this key region is not protected from genetic variation.

Conclusions: It appears that the αIIb genu is the object of stringent quality control to prevent platelets from circulating with activated and extended integrin.

Keywords: ITGA2B; Glanzmann thrombasthenia; genetic variants; in silico analysis; αIIb genu.

Figure 1

Ribbon diagrams highlighting selected missense mutations in and around the genu region of αIIb. The lower left panel shows a cartoon representation of the entire extracellular domain of αIIb with the genu in dark blue. Missense mutations most closely affecting this region are in red and numbered from 1 to 7. Framed are groups of missense variants within (a) connecting loops from the base of the thigh domain; (b) the unstructured linker ribbon between the thigh and the calf‐1 domains that contains a Ca2+ loop; and (c) the lower bend of the genu which contains a H‐bond clasp. In small windows are enlarged views of mutations illustrated as graphical sticks with superimposed the natural amino acid represented as lines. Graphical “bumps” (red discs) reveal steric encumbrance caused by the amino acid substitution. Amino acids engaged in H‐bonds (dotted lines) are represented as sticks with C atoms in white, N atoms in blue, O atoms in red and S atoms in orange. In (a) the black arrow indicates the position of Ile596; for Ile596Thr numbers in red are interatom distances in Angstroms, while a red dotted line represents a potential H‐bond introduced by the presence of Thr596. In (b) the Ca2+ is represented as a gray sphere and the dotted lines coordination links to nearby amino acids. Models were obtained using the PyMol Molecular Graphics System, version 1.3 and the 3fcs pdb file

Figure 2

Cartoon representations of selected missense mutations within the αIIb calf‐1 domain. On the left hand side is a ribbon diagram depicting the extracellular domains of αIIb with the genu region colored in dark blue. Framed is the calf‐1 domain with missense mutations colored in red and numbered from 1 to 5. On the right hand side are selected representations of the calf‐1 domain with selected mutations as red sticks. The calf‐1 domain is structured as a β‐barrel composed of multiple anti‐parallel β‐strands. Amino acids whose side chain point to the core of the barrel are colored in gray and represented as spheres in the far right representations. In the small windows are enlarged views with each mutation illustrated as graphical sticks with superimposed, the natural amino acid represented as lines. C atoms are colored in white, N atoms in blue, O atoms in red and S atoms in orange. Graphical “bumps” (red discs) reveal steric encumbrance caused by the amino acid substitution. Models were obtained using the PyMol Molecular Graphics System, version 1.3 and the 3fcs and 1uv9 pdb files