Immune thrombocytopenia (ITP) is an acquired autoimmune bleeding disorder caused by increased platelet destruction and impaired platelet production, induced by IgG antiplatelet autoantibodies [1]. Recently, we found that CD4+CD25+ regulatory T-cell (Treg) depletion in mice causes the spontaneous development of sustained thrombocytopenia with increases in the platelet-associated IgG and the proportion of reticulated platelets [2]. Platelet eluates and the supernatants of splenocyte cultures prepared from these thrombocytopenic mice contained IgG antibodies capable of binding to intact platelets derived from normal mice, indicating that the primary mechanism was autoantibody-mediated, analogous to the pathophysiology of ITP. To identify molecular targets recognized by antiplatelet autoantibodies in Treg-deficient ITP mice, we performed antiplatelet antibody detection assays using platelets from mice deficient in functional platelet membrane glycoprotein (GP) IIb/IIIa or the GPIb/IX complex, which are the major targets of antiplatelet autoantibodies in ITP patients [3].

Treg-deficient mice were generated by transferring Treg-depleted CD4+ T cells from BALB/c mice into syngeneic nude mice [2]. Approximately one-third of the mice spontaneously developed thrombocytopenia 3 weeks after the transfer. As IgG bound to platelets from wild-type (WT) BALB/c mice were detected in platelet eluates and supernatants of splenocyte cultures, but not in plasma samples, from thrombocytopenic mice [2], splenocyte culture supernatants were used for evaluation of antigenic specificity of antiplatelet autoantibodies. Specifically, splenocytes were obtained from 10 Treg-deficient mice with thrombocytopenia (ITP mice; #60, #77, #105, #106, #108, #135, #139, #182, #199 and #200) 8 weeks after the transplantation and from 10 mice without thrombocytopenia (control mice), and cultured in RPMI1640 plus 10% fetal bovine serum for 4 days without antigenic or mitogenic stimulation. Platelets from WT mice were incubated with culture supernatants, and subsequently with AlexaFluor® 488-labeled anti-mouse IgG polyclonal antibodies (Life Technologies, Carlsbad, CA, USA). Platelets derived from GPIb-knockout (GPIb−/−) or GPIIIa-knockout (GPIIIa−/−) mice [4], which lack a functional GPIb/IX or GPIIb/IIIa complex on platelet surfaces, were also used instead of WT platelets. The IgG bound to platelets was quantitatively analyzed by a FACS® Calibur flow cytometer and CellQuest™ software (Becton Dickinson, Franklin Lakes, NJ, USA). The amount of bound IgG antibody was calculated as the ratio of the mean fluorescence intensity (MFI) of platelets treated with culture supernatant and fluorophore-conjugated anti-mouse IgG to the MFI of platelets treated with fluorophore-conjugated anti-mouse IgG alone. Relative IgG binding to the GPIb−/− or GPIIIa−/− platelets was obtained by comparison with the binding to WT platelets. All experimental protocols were approved by the Keio University Ethics Committee for Animal Experiments.

Flow cytometric analysis showed IgG antibodies capable of binding to WT platelets in the culture supernatant from representative ITP mouse #60, but not from control mouse #101 (Fig. 1A). Consistent results were obtained from 10 ITP and 10 control mice. When GPIb−/− platelets were used instead of WT platelets in the assay, IgG from a representative ITP mouse, #106, failed to bind (Fig. 1B). Concordant results were obtained from all ITP mice, although IgG from three ITP mice (#77, #105 and #199) bound to GPIb−/− platelets to a small degree. Binding of IgG from another representative ITP mouse, #77, to GPIIIa−/− platelets was reduced, compared with binding to WT platelets (0Fig. 1C), but binding to GPIIIa−/− platelets and WT platelets was nearly concordant in IgG from the remaining nine ITP mice. Relative binding of IgG from 10 ITP mice to GPIb−/− or GPIIIa−/− platelets compared with their binding to WT platelets is summarized in Fig. 1(D). IgG binding was diminished by the use of GPIb−/− platelets in all ITP culture supernatants, and was completely lost in half the samples. IgG binding to GPIIIa−/− platelets was reduced in six (60%) samples, but the reduction in binding was much less than with GPIb−/− platelets. IgG from ITP mouse #105 bound both GPIb−/− and GPIIIa−/− platelets, suggesting that in this mouse, antiplatelet antibodies also recognized other platelet surface molecules.


Figure 1. Detection of IgG autoantibodies to GPIb/IX and GPIIb/IIIa in splenocyte culture supernatants from Treg-deficient ITP mice. IgG anti-GPIb/IX and anti-GPIIb/IIIa antibodies were analyzed by flow cytometry using WT, GPIb−/− or GPIIIa−/− platelets as a source of platelet antigens. (A) IgG bound to WT platelets incubated with culture supernatants from ITP mouse #60 (left) or control mouse #101 (right). Open histograms represent staining with culture supernatants and the secondary antibody; shaded histograms represent control staining with secondary antibody alone. (B) IgGs from ITP mouse #106 bound to GPIb−/− or WT platelets (solid and dotted lines, respectively). Shaded histogram represents staining of WT platelets with secondary antibody alone. (C) IgGs from ITP mouse #77 bound to GPIIIa−/− or WT platelets (solid and dotted lines, respectively). Shaded histograms represent staining of WT platelets with secondary antibody. (D) Relative binding of IgGs from 10 ITP mice to GPIb−/− (black bars) or GPIIIa−/− (white bars) platelets, compared with binding to WT platelets (defined as 100%). (E) Supernatants containing IgGs from ITP mouse #106 were pre-incubated with GPIb−/− or WT platelets (solid and dotted lines, respectively) and IgG binding to WT platelets was assayed. Shaded histogram represents staining with mock pre-incubated culture supernatant. (F) IgGs from ITP mouse #77 were treated as described in E except that the supertantants were pre-incubated with GPIIIa−/− or WT platelets (solid and dotted lines, respectively). Shaded histogram represents staining with mock pretreated culture supernatant.

Download figure to PowerPoint

To confirm the autoantibody binding of IgG from ITP mice to GPIb/IX and GPIIb/IIIa, we conducted antibody depletion assays, in which splenocyte culture supernatants were pre-incubated with platelets derived from WT, GPIb−/− or GPIIIa−/− mice and then subjected to the antiplatelet antibody assay using WT platelets as the source of antigen. Pre-incubation of the supernatant from representative ITP mouse #106 with WT platelets completely abolished IgG binding to platelets, but pre-incubation with GPIb−/− platelets had no effect (Fig. 1E). A concordant result was obtained from additional ITP mice (#135 and #200). In contrast, IgG binding of supernatant from ITP mouse #77 to platelets was partially suppressed by pre-incubation with GPIIIa−/− platelets (Fig. 1F).

Our results clearly indicate that both GPIb/IX complex and GPIIb/IIIa complex are the major autoantigens recognized by IgG antiplatelet autoantibodies in an ITP mouse model generated by Treg deficiency. This heterogeneous autoantigen recognition profile is analogous to the antiplatelet autoantibody response seen in ITP patients [3]. Antiplatelet autoantibodies in our ITP mouse model preferentially targeted GPIb/IX over GPIIb/IIIa. In contrast, the most common autoantigen in ITP patients is GPIIb/IIIa, which is recognized by approximately 30–93% of them, depending on the assay [5–11], and anti-GPIb/IX is the second-most prevalent antibody, found in 12–83% of the patients [5–10]. However, some reports have shown that anti-GPIb/IX antibodies are more frequent than anti-GPIIb/IIIa antibodies in ITP patients [9,10]. Interestingly, ITP patients with anti-GPIb/IX antibodies are refractory to treatment with intravenous IgG (IVIG), and those without these antibodies are responsive [12]. In addition, IVIG treatment is effective in a mouse model of experimental thrombocytopenia induced by an anti-GPIIb/IIIa monoclonal antibody, but not in a model induced by an anti-GPIbα monoclonal antibody [13]. We have yet to examine the efficacy of IVIG treatment in Treg-deficient ITP mice.

The antiplatelet autoantibody response in individual mice was heterogeneous, including recognition of GPIb/IX alone and of GPIb/IX and GPIIb/IIIa together, even though all the Treg-deficient mice had an identical BALB/c genetic background and were grown in the same colony. This heterogeneous nature of the autoantibody response is in sharp contrast to another ITP mouse model, which is characterized by induction of antibody- and CD8+ T cell-mediated responses specific to GPIIIa [14]. It has been recently reported that peripheral Treg deficiency is observed in association with thymic retention of Tregs in this ITP model [15], but the immune target of this model is restricted to GPIIIa. This is not surprising because this model is generated by immunization of a GPIIIa−/− mouse with wild-type platelets, and subsequent transfer of splenocytes from the immunized mouse into a severe combined immunodeficiency mouse. Alternatively, this finding strongly suggests that peripheral Treg deficiency alone is not sufficient to induce autoreactive responses to multiple GPs on platelets.

In contrast, in our ITP model, a small number of Treg-depleted CD4+ T cells derived from WT mice are transferred into syngeneic nude mice that lack functional T cells. The transferred CD4+ T cells proliferate in the lymphopenic condition by a process termed ‘homeostatic proliferation’ [16]. The clonal expansion of an autoreactive CD4+ T-cell repertoire during this process is reported to be responsible for the development of autoimmunity [17]. In our ITP model, the absence of Tregs during homeostatic proliferation is likely to promote expansion of a variety of autoreactive T-cell reportoires, resulting in elicitation of autoantibody responses to multiple platelet GPs. The mechanisms that control the expansion of autoreactive T cells have not been fully elucidated, but contributing factors may include precursor frequency in the transferred T cells and the presentation of a variety of agonistic peptides required for the survival and expansion of autoreactive donor T cells in the recipient nude mice [17]. Differences in these factors among individual Treg-deficient ITP mice may contribute to their heterogeneous antiplatelet autoimmune responses.

In conclusion, IgG autoantibody responses to multiple platelet surface GPs, including GPIb/IX and GPIIb/IIIa, observed in Treg-deficient ITP mice are analogous to the antiplatelet autoantibody response in ITP patients. Our new mouse ITP model should be useful for analyzing the mechanisms behind the antiplatelet autoantibody response and for developing novel therapeutic strategies to suppress harmful autoimmune responses in ITP patients.


T. Nishimoto performed research, collected data, analyzed and interpreted data and wrote the manuscript; T. Satoh and E. K. Simpson performed research and collected data; H. Ni interpreted data and supervised the study; M. Kuwana designed the research, analyzed and interpreted data, supervised and organized the study and wrote the manuscript.


This work was supported by a research grant on intractable diseases from the Japanese Ministry of Health, Labour, and Welfare, and Canadian Institutes of Health Research (MOP-97918). E.K. Simpson was a recipient of the studentship award of LKSKI of St. Michael's Hospital and the Heart & Stroke Foundation of Canada/Ontario Graduate Scholarship in Science & Technology.

Disclosure of Conflict of Interests

The authors state that they have no conflict of interest.


  1. Top of page
  2. References
  • 1
    Cines DB, Blanchette VS. Immune thrombocytopenic purpura. N Engl J Med 2002; 346: 9951008.
  • 2
    Nishimoto T, Satoh T, Takeuchi T, Ikeda Y, Kuwana M. Critical role of CD4(+)CD25(+) regulatory T cells in preventing murine autoantibody-mediated thrombocytopenia. Exp Hematol 2012; 40: 27989.
  • 3
    McMillan R. Autoantibodies and autoantigens in chronic immune thrombocytopenic purpura. Semin Hematol 2000; 37: 23948.
  • 4
    Li C, Piran S, Chen P, Lang S, Zarpellon A, Jin JW, Zhu G, Reheman A, van der Wal DE, Simpson EK, Ni R, Gross PL, Ware J, Ruggeri ZM, Freedman J, Ni H. The maternal immune response to fetal platelet GPIbalpha causes frequent miscarriage in mice that can be prevented by intravenous IgG and anti-FcRn therapies. J Clin Invest 2011; 121: 453747.
  • 5
    Warner MN, Moore JC, Warkentin TE, Santos AV, Kelton JG. A prospective study of protein-specific assays used to investigate idiopathic thrombocytopenic purpura. Br J Haematol 1999; 104: 4427.
  • 6
    Berchtold P, Müller D, Beardsley D, Fujisawa K, Kaplan C, Kekomäki R, Lipp E, Morell-Kopp MC, Kiefel V, McMillan R, von dem Borne AE, Imbach P. International study to compare antigen-specific methods used for the measurement of antiplatelet autoantibodies. Br J Haematol 1997; 96: 47783.
  • 7
    Brighton TA, Evans S, Castaldi PA, Chesterman CN, Chong BH. Prospective evaluation of the clinical usefulness of an antigen-specific assay (MAIPA) in idiopathic thrombocytopenic purpura and other immune thrombocytopenias. Blood 1996; 88: 194201.
  • 8
    McMillan R, Wang L, Tani P. Prospective evaluation of the immunobead assay for the diagnosis of adult chronic immune thrombocytopenic purpura (ITP). J Thromb Haemost 2003; 1: 48591.
  • 9
    Kiefel V, Santoso S, Kaufmann E, Mueller-Eckhardt C. Autoantibodies against platelet glycoprotein Ib/IX: a frequent finding in autoimmune thrombocytopenic purpura. Br J Haematol 1991; 79: 25662.
  • 10
    He R, Reid DM, Jones CE, Shulman NR. Spectrum of Ig classes, specificities, and titers of serum antiglycoproteins in chronic idiopathic thrombocytopenic purpura. Blood 1994; 83: 102432.
  • 11
    Kuwana M, Okazaki Y, Satoh T, Asahi A, Kajihara M, Ikeda Y. Initial laboratory findings useful for predicting the diagnosis of idiopathic thrombocytopenic purpura. Am J Med 2005; 118: 102633.
  • 12
    Go RS, Johnston KL, Bruden KC. The association between platelet autoantibody specificity and response to intravenous immunoglobulin G in the treatment of patients with immune thrombocytopenia. Haematologica 2007; 92: 2834.
  • 13
    Webster ML, Sayeh E, Crow M, Chen P, Nieswandt B, Freedman J, Ni H. Relative efficacy of intravenous immunoglobulin G in ameliorating thrombocytopenia induced by antiplatelet GPIIbIIIa versus GPIbalpha antibodies. Blood 2006; 108: 9436.
  • 14
    Chow L, Aslam R, Speck ER, Kim M, Cridland N, Webster ML, Chen P, Sahib K, Ni H, Lazarus AH, Garvey MB, Freedman J, Semple JW. A murine model of severe immune thrombocytopenia is induced by antibody- and CD8+ T cell-mediated responses that are differentially sensitive to therapy. Blood 2010; 115: 124753.
  • 15
    Aslam R, Hu Y, Gebremeskel S, Segel GB, Speck ER, Guo L, Kim M, Ni H, Freedman J, Semple JW. Thymic retention of CD4+CD25+FoxP3+ T regulatory cells is associated with their peripheral deficiency and thrombocytopenia in a murine model of immune thrombocytopenia. Blood 2012; 120: 212732.
  • 16
    Bell EB, Sparshott SM. The peripheral T-cell pool: regulation by non-antigen induced proliferation? Semin Immunol 1997; 9: 34753.
  • 17
    Le Campion A, Gagnerault MC, Auffray C, Bécourt C, Poitrasson-Rivière M, Lallemand E, Bienvenu B, Martin B, Lepault F, Lucas B. Lymphopenia-induced spontaneous T-cell proliferation as a cofactor for autoimmune disease development. Blood 2009; 114: 178493.