Expression of a functional, hybrid murine αIIb-human β3 integrin complex. (A) Schematic diagram of lentivirus vector β3-WPTS. The enhancer/promoter of the viral 3′-long terminal repeat (LTR) was removed to allow the vector to self-inactivate (SIN), and the human ITGA2B gene promoter (nucleotides –889 to +35) was used to direct megakaryocyte-specific synthesis of human β3 in mice. The promoter binds GATA and Ets for high-level gene transcription in megakaryocytes, and there is also a repressor region that inhibits gene expression in other lineages. The woodchuck hepatitis virus postregulatory element (WPRE) and the central polypurine tract (cPPT) were used to enhance the efficiency of transgene expression. (B) β3-WPTS–transduced bone marrow was transplanted into β3–/– mice (as described in “Materials and methods”). Flow cytometric histograms of murine platelets isolated from circulating whole blood of transplant recipients A to F showed that 5 mice exhibited significant levels of β3 on their platelet surface (shaded peak) compared with MFI levels for platelets from a β3–/– control (overlay histogram) using a PE-conjugated antibody to human β3. (C) Flow cytometric analysis detected integrin αIIb on the surface of platelets from recipients A to F with an FITC-conjugated antibody to murine αIIb. Histograms showed that mice B to F expressed αIIb on platelets (shaded peak) at intermediate MFI levels compared with the levels (in parentheses) on platelets from β3–/–, β3+/–, and β3+/+ controls. (D, left) Flow cytometric analysis revealed that Alexa 488–conjugated antibody (7E3) specific for human β3 in complex with αIIb or αv reacted positively with platelets from a β3–/– mouse expressing human β3 in complex with murine αIIb (shaded peak) in comparison with murine β3+/+ platelets serving as a negative control and human β3+/+ platelets used as a positive control. (Middle) Platelets from the mouse in the left panel were used to show that a fibrinogen mimetic peptide containing Arg-Gly-Asp (+RGD) could induce murine platelets expressing human β3 (shaded peak) to bind a monoclonal antibody (D3) (plus PE-F(ab′)2 goat anti–murine IgG Fc secondary antibody) that recognizes a ligand induced binding site (LIBS) exposed only on the high-affinity conformation of human β3. The platelets failed to bind D3 in the absence of RGD peptide (–RGD). (Right) Histogram demonstrating that an antibody (PE-Jon/A) specific for the high-affinity conformation of murine αIIbβ3 reacted positively with platelets expressing human β3 from the mouse described in the left and middle panels (shaded peak) following treatment with a cocktail of physiologic agonist of platelet activation (ADP, epinephrine, PAR4). Quiescent platelets were not recognized by PE-Jon/A in the absence of agonist (–Agonist). Results shown were observed in at least 2 separate experiments analyzing platelets from 3 separate mice that expressed human β3 at similar MFI levels. (E) Platelets from the mouse described in panel D were fixed and permeabilized to perform quantitative analysis with rabbit polyclonal antibodies to detect the intracellular storage of major ligands for αIIbβ3, fibrinogen, and VWF. (Left) Histogram shows that a nonreactive Alexa 647 rabbit polyclonal antibody did not react with murine platelets expressing human β3 (shaded peak) nor did it stain platelets from β3–/–, β3+/–, and β3+/+ controls. A nonreactive FITC-Ig showed identical results (not shown). (Middle, right) Histograms reveal that an FITC-antibody to fibrinogen (middle) and an Alexa 647–antibody to VWF (right) recognized platelets from the mouse expressing human β3 (shaded peak) at intermediate MFI levels compared with the level (in parentheses) in platelets from β3–/–, β3+/–, and β3+/+ controls. Results shown were observed in at least 2 separate experiments analyzing platelets from 3 separate mice that expressed human β3 at similar MFI levels.
