The Role And Mechanism Of Cytotoxic T Lymphocyte-mediated Platelet Destruction In ITP

Primary immune thrombocytopenia (ITP) is one of the most common bleeding diseases, characterized by reduced platelet count in peripheral blood. No obvious bleeding symptoms are observed in some patients, while other patients encounter bleeding symptoms of skin or mucous membrane, menorrhagia, even life-threatening viscera and intracranial hemorrhage. Various factors including increased bleeding risk, unpredictability of outcome, fear of low platelet count, long-term treatment, reduced social activities, incapability of work, and so on, lead to the decrease of the life quality of ITP patents seriously. The pathogenesis of ITP consists of increased platelet destruction, impaired platelet production and immune imbalance of T lymphocytes. The first-line therapies for ITP include corticosteroids, intravenous immunoglobulin (IVIG) and anti-D immunoglobulin (anti-D). The second-line therapies include splenectomy, thrombopoietin receptor (TPO-R) agonists, rituximab, and other immune suppressive agents. Some ITP patients respond poorly to both first-line and second-line therapies, and become refractory ITP after long-term treatment. Therefore, to further study the pathogenesis and new therapeutic targets of ITP is of great significance.Increased platelet destruction is mainly mediated by antiplatelet autoantibodies. Patients lose immune tolerance and produce antiplatelet autoantibodies targeting glycoproteins (GPs) on platelets, mainly including GPIb/IX and GPIIb/IIIa. Antiplatelet autoantibodies bond to GP antigen epitopes on platelet surface and mediate the phagocytosis of platelets by reticuloendothelial system in the spleen. However, antiplatelet autoantibodies can be detected in the plasma of only 50-70% of ITP patients. Furthermore, some ITP patients fail to respond to therapies blocking autoantibody-mediated platelet destruction, indicating that there must be other pathways responsible for platelet destruction in ITP.Studies found that cytotoxic T lymphocytes (CTLs) could directly lyse platelets by perform and granzyme B in ITP, leading to increased platelet destruction. Furthermore, CTLs from ITP patients could inhibit the maturation and apoptosis of megakaryocytes, thus resulting in impaired platelet production. These studies demonstrated that CTLs played a critical role in the pathogenesis of ITP.Corticosteroid is the first-line therapy for ITP patients. It was thought that conventional-dose prednisone was the first choice of corticosteroid in the treatment. Recently, several studies revealed that high-dose dexamethasone (HD-DXM) might be a superior choice to conventional-dose prednisone because of shorter time to response and less adverse events. After HD-DXM therapy, the cytotoxity of CD8+T lymphocytes of ITP patients towards platelets significantly reduced, which might be one of the mechanisms through which HD-DXM achieved its therapeutic effect.CD8 is commonly used as a marker of CTLs. However, CD8+T lymphocytes contain different subsets which have different surface markers and functions. CD28 is a co-stimulatory molecule necessary for the initiation of most T cell responses. It is demonstrated that the lack of CD28 molecule is closely associated with immunosuppressive function of CD8+T lymphocytes. CD8+CD28- T lymphocytes are also called suppressor T cells (Ts). In addition, CD8+ T lymphocytes contain regulatory T cells (Tregs). Many studies showed that the defect of CD4+ Tregs contributed to the pathogenesis of ITP. Reduced frequency and impaired suppressive function of CD4+ Tregs were found in peripheral blood, bone marrow and spleen of ITP patients. The frequency and/or function of CD4+ Tregs could be improved as the platelet count increased after the treatments of dexamethasone, rituximab, or thrombopoietin receptor agonists. CD8+ Tregs were also shown to exert an important regulatory role in several autoimmune diseases, including systemic lupus erythematosus and experimental autoimmune encephalitis, despite of their small population. However, the status and role of CD8+ Tregs in ITP remain unclear.Recent studies have highlighted the role of phagocytosis of desialylated platelets by hepatocytes in platelet destruction in addition to antiplatelet autoantibodies and CTLs. Platelet desialylation is the process in which sialic acids on platelet glycoproteins are cleaved by neuraminidases. Then β-galactose residues (βGals) are exposed and recognized by asialoglycoprotein receptors (ASGPRs)-expressing hepatocytes. This process mediates the platelet clearance in the liver, which is an important pathway for platelet destruction. Platelet desialylation is catalyzed by neuraminidases. Increased exogenous neuraminidases caused by bacteria or virus infections, and increased endogenous neuraminidase activity caused by various factors which can lead to neuraminidase translocation to cell surface or release to plasma, can result in platelet desialylation and then induce platelet clearance in the liver. Studies found that infections, platelet refrigeration, free radicals, aging and vessel damage can cause platelet lesion. Platelet lesion induces neuraminidase translocation from platelet lysosomes to platelet surface or even release to extracellular plasma, and then causes increased platelet desialylation. We speculate that platelet desialylation may participate in the pathogenesis of ITP. To date, platelet desialylation level, its association with antiplatelet autoantibodies and CTLs, and its effect on platelet destruction in ITP remain unclear.We will conduct in vivo and in vitro experiments, and investigate the frequencies of CD8+ lymphocyte subsets, including CTLs, Ts and Tregs, the alteration of these subsets after HD-DXM therapy, the platelet desialylation level in ITP, its effect on platelet destruction, its association with antiplatelet autoantibody specificity and CTLs. This study will further reveal the novel mechanisms of platelet destruction in ITP and shed lights on new therapeutic options.Part I:The study on the regulation of cytotoxic T lymphocytes and CD8+regulatory T cells by high-dose dexamethasone in ITPObjective:To detect the frequencies of CD8+lymphocyte subsets, including CTLs、Ts、Tregs, in ITP patients and controls, measure the plasma levels of transforming growth factor-β1 (TGF-β1) and interleukin 10 (IL-10) in ITP patients and controls, determine the alteration of the frequencies of these CD8+lymphocyte subsets and the plasma levels of TGF-β1 and IL-10 after HD-DXM therapy, elucidate the status of CD8+ lymphocyte subsets and explore the mechanisms of HD-DXM in treating ITP patients.Methods:1. Twenty-one ITP patients were enrolled and received HD-DXM (40mg/d) for 4 consecutive days. Blood samples were collected before treatment and 14 days after the initiantion of HD-DXM therapy. Clinical response and adverse events were also evaluated. Sixteen matched healthy volunteers were enrolled as controls.2. Platelet counts were measured. Peripheral blood mononuclear cells (PBMCs) and plasma were separated.3. PBMCs were stained by mouse anti-human APC-conjugated CD8, FITC-conjugated CD28, PE-Cy5-conjugated CD25 and PE-conjugated Foxp3 monoclonal antibodies. The expression levels of CD8, CD28, CD25 and Foxp3 of lymphocytes in ITP patients and controls were analyzed by flow cytometry.4. The plasma concentrations of TGF-β1 and IL-10 in ITP patients and controls were determined by enzyme-linked immuno sorbent assay (ELISA).Results:1. Of these 21 ITP patients receiving HD-DXM therapy, the rates of complete response (CR), response (R) and no response (NR) were 57.14%,28.57% and 14.29%, respectively. No obvious adverse events were observed.2. There were no significant difference of the percentage of CD8+ cells in lymphocytes between ITP patients and controls, nor between pre-treatment and post-treatment ITP patients.3. The frequency of CTLs (CD28+in CD8+population) was significantly higher in ITP patients than that in controls, which could be markedly reduced by HD-DXM therapy. The frequency of Ts (CD28" in CD8+population) definitely showed an opposite trend with CTLs.4. The frequency of CD8+Tregs (CD25+or Foxp3+in CD8+population) was significantly lower in ITP patients than that in controls, which could be restored to some extent by HD-DXM therapy.5. The plasma levels of both TGF-β1 and IL-10 were lower in ITP patients than that in controls. After HD-DXM administration, the plasma TGF-β1 level of ITP patients was increased to the normal level, while the plasma IL-10 level was also increased but still lower than that in controls.Conclusion:1. The increased CTLs, decreased Ts and Tregs in CD8+lymphocytes may contribute to the pathogenesis of ITP.2. HD-DXM can reduce the frequency of CTLs and increase the frequencies of Ts and Tregs in CD8+lymphocytes, which may partially explain how HD-DXM exerts its therapeutic effect in ITP patients.Part II:The study on the mechanism of cytotoxic T lymphocytes causing platelet desialylation in ITPObjective:To detect the antiplatelet autoantibodies, the cytotoxity of CTLs towards platelets and the platelet desialylation level in ITP patients and controls; and analyze the association between antiplatelet autoantibody specificity or the cytotoxity of CTLs towards platelets and the platelet desialylation level. To measure pre-treatment and post-treatment platelet desialylation levels of ITP patients and determine the effect of response to treatment on platelet desialylation. To evaluate the levels of desialylation and neuraminidase 1 (Neul) expression of platelets after incubation with plasma or CTLs of ITP patients and controls. To study the phagocytosis of desialylated platelets in the liver through animal experiments. To validate the status and role of platelet desialylation in the onset of ITP, the effect of antiplatelet autoantibodies and CTLs on platelet desialylation, and the destruction of desialylated platelets in vivo.Methods:1. Blood samples were obtained from active ITP patients before and after treatment and controls. Platelet counts were measured. Plasma, PBMCs and platelets were separated.2. The exposure of βGals on platelets was detected by flow cytometry. Mouse anti-human PE-Cy5-conjugated CD41a monoclonal antibody was used to label platelets and FITC-conjuagted RCA-I was stained at the same time. The mean fluorescence intensity (MFI) of RCA-I represented for the platelet desialylation level.3. Monoclonal antibody-specific immobilization of platelet antigens (MAIPA) was conducted to detect anti-GPIb and anti-GPIIb/IIIa antibodies in the plasma of ITP patients.4. Assay of CTL-mediated cytotoxicity towards platelets was performed to detect the platelet lysis mediated by CTLs. Briefly, CD8+lymphocytes were separated by positive selection with CD8 magnetic microbeads. Platelets were incubated with CD8+lymphocytes for 4 hours. Then mitochondrial membrane potential of platelets was detected by mitochondrial membrane potential assay kit with JC-1.5. We used plasma and CTLs to incubate with platelets. After incubation, the platelet desialylation level and neuraminidase 1 (Neul) expression were analyzed by flow cytometry.6. CD61 knockout mice were immunized against platelets from wild-type (WT) mice. Labeled WT platelets were cultured with CD8+T cells selected from splenocytes of immunized CD61 knockout mice. After 4 hours, platelets were injected into WT mice through caudal vein. Thirty minutes later, mice were anesthetized and drained of blood. Livers were harvested. Imunofluorescence was conducted to detect the platelet phagocytosis in the liver.Results:1. The MFI of RCA-Ⅰ on platelets was significantly higher in ITP patients than that in controls. The rates of CR, R and NR were 19.36%,45.16% and 35.48%, respectively. In CR and R patients, the MFI of RCA-I decreased significantly after treatment. While in NR patients, there was no change.2. Among these ITP patients, there were 17.57% anti-GPIb positive,13.51% anti-GPIIb/IIIa positive,21.62% double positive and 47.30% double negative. There was no difference of RCA-I binding level on platelets among each group. Moreover, no difference was found between anti-GPIb positive and negative, between anti-GPIIb/IIIa positive and negative, and between antiplatelet autoantibodies positive and negative patients. After incubation of platelets with ITP plasma with different antiplatelet autoantibody specificity in vitro, there was no difference of the MFI of RCA-I or Neul of platelets among each group.3. The CTL-induced platelet apoptosis was significantly higher in ITP patients than that of controls. According to the percent CTL-induced platelet apoptosis, ITP patients were divided into cytotoxic and non-cytotoxic group. The percentage of cytotoxic group was 55% and the rest 45% belonged to noncytotoxic group. The MFI of RCA-I of platelets in cytotoxic group was markedly higher. After incubation of platelets with CTLs of ITP patients in vitro, the MFI of RCA-I and Neul of platelets in cytotoxic group was significantly higher than that in noncytotoxic group and controls, which could be dereased by neraminidase inhibitor.4. After cultured with CD8+ T cells selected from splenocytes of immunized CD61 knockout mice and injected back to WT mice, platelets underwent more obvious phagocytosis in the liver than the control group,which could be inhibited by neraminidase inhibitor.Conclusion:1. The platelet desialylation level is significantly higher in ITP patients than that in controls and effective treatment can markedly decrease the platelet desialylation level in ITP patients. These results indicate that increased platelet desialylation level may play an important role in the pathogenesis and development of ITP. Inhibiting platelet desialylation may provide novel insight into the treatment of ITP.2. There is no assiociation between the platelet desialylation level and antiplatelet specificity in ITP.3. CTLs can lead to a secondary platelet clearance in the liver via platelet desialylation in ITP, which may be a novel mechanism of increased platelet destruction in ITP.