Adiponectin Assists Thrombopoietic Agents in ITP Treatment by Enhancing Myosin-9/Rab6A-Mediated Trafficking of c-Mpl in MKs

Abstract

Immune thrombocytopenia (ITP) is an autoimmune disorder characterized by accelerated platelet destruction and defective platelet production. The thrombopoietin (TPO)-activated c-Mpl signaling pathway has been proven to promote megakaryocyte differentiation and platelet production and thus has significant value in the clinical treatment of ITP. However, individual differences and unsustainable responses limit the clinical application of c-Mpl agonists. The cell membrane distribution of c-Mpl is crucial for the cell response to c-Mpl agonists. In the present study, this is observed that the distribution of c-Mpl on the megakaryocyte membrane is significantly reduced in ITP patients. The reduction is more severe in refractory patients. Then, this is verified that the membrane trafficking of c-Mpl is mediated by the Myosin-9/Rab6A complex, and demonstrates that the stability of this complex depends on Rab6A-GTPase. Furthermore, this is found that adiponectin promotes the membrane localization of c-Mpl by increasing the expression of Rab6A GEFs. Finally, this is revealed that adiponectin can assist thrombopoietic agents in the treatment of ITP mice. The results clarify the c-Mpl distribution defects in the cell membrane and the corresponding c-Mpl membrane transport mechanisms in megakaryocytes of ITP, providing new insights to improve the clinical efficacy of thrombopoietic agents for ITP patients.

Keywords: ITP; Myosin‐9/Rab6A complex; adiponectin; c‐Mpl; membrane trafficking.

Scheme 1

This schematic illustrates the mechanism by which adiponectin enhances the therapeutic efficacy of thrombopoietic agents in ITP. In megakaryocytes (MKs), adiponectin promotes the expression of Rab6A GEFs, which in turn stabilizes the Myosin‐9/Rab6A trafficking complex. This complex facilitates the membrane localization of c‐Mpl, thereby restoring TPO responsiveness and enhancing megakaryocytic maturation.

Figure 1

Cell‐surface c‐Mpl levels are reduced in ITP MKs A) Schematic for the acquisition and staining of bone marrow mononuclear cells (BMMCs). B,C) Immunofluorescence (IF) microscopy images (B) and the quantitative summary (C) of c‐Mpl expression on the surface of MKs from treatment‐naïve primary ITP patients (n = 38), rhTPO/TPO‐RAs resistant primary ITP patients (n = 19) and healthy control subjects (n = 14). At least 15 cells were quantified per donor. CD41a, red; c‐Mpl, green. Scale bars, 10 µm. ***P < 0.001; ****P < 0.0001. D) Flow cytometry strategy for the selection of the MK population. E) Representative histograms and statistical analysis for c‐Mpl expression on the surface of MKs from BM samples by flow cytometry. ISO Control represents staining with no primary antibody. The geometric mean fluorescence intensity is shown. **P < 0.01; ***P < 0.001. F) Scheme of the design and procedures of the passive ITP murine model in which was administered CD41a antibody through intraperitoneal injection into C57BL/J mice (6–8 weeks). (G) Platelet counts of peripheral blood from NC and ITP mice were monitored at different time points. (n = 5). ***P < 0.001; ****P < 0.0001. H) Imaging of megakaryocytes stained with Wright–Giemsa stain from the NC and ITP mouse bone marrow samples. I) IF imaging and quantitative summary of the expression of c‐Mpl on the MK surface from the bone marrow of NC and ITP mice. At least 50 cells were quantified per mouse bone marrow (n = 4). CD41a, red; c‐Mpl, green. Scale bars, 10 µm. *P < 0.05. The means ± SDs are shown for the statistical analysis.

Figure 2

The Myosin‐9/Rab6A trafficking complex transports c‐Mpl to the cell membrane A) An illustration of the structures of c‐Mpl and the c‐Mpl‐C‐terminal domain are presented as cartoons. TMD, transmembrane domain. ECD, extracellular domain. ICD, intracellular domain. B) A silver‐stained SDS gel shows proteins immunoprecipitated to the c‐Mpl‐C‐terminal domain from HEK‐293T cells using an anti‐Flag antibody. C) The top four trafficking‐associated proteins detected by mass spectrometry from the IP gel products. D,E) Immunoblotting shows that Flag‐c‐Mpl does coimmunoprecipitates with Myc‐Myosin‐9 (D) and GFP‐Rab6A (E) in HEK‐293T cells. F,G) The protein expression (F) and relative mRNA expression (G) of Myosin‐9 in shMyosin‐9 UT‐7 cells (n = 3). ****P < 0.0001. H,I) The protein expression (H) and relative mRNA expression (I) of Rab6A in shRab6A UT‐7 cells (n = 3). ****P < 0.0001. J,L) IF images and normalized quantitative analysis of the expression of c‐Mpl on the cell surface of shMyosin‐9 (J) and shRab6A (L) UT‐7 cells. At least 30 cells were quantified per individual experiment (n = 3). c‐Mpl, green. Scale bars, 10 µm. *P < 0.05; **P < 0.01; ***P < 0.001. K,M) Representative histograms and statistical analysis of c‐Mpl expression on the surface of shMyosin‐9 K) and shRab6A (M) UT‐7 cells determined via flow cytometry. The normalized geometric mean fluorescence intensity is shown (n = 3). *P < 0.05; **P < 0.01. All of the UT‐7 cells were transduced with Flag‐c‐Mpl. The means ± SDs are shown.

Figure 3

Myosin‐9/Rab6A trafficking complex enhances MK maturation via the TPO/c‐Mpl signaling pathway A,B) Photomicrograph showing morphological differences (cell size) in NC and shMyosin‐9 DAMI cells after 72 h of PMA treatment. Scale bars, 20 µm. Quantitative analysis of the percentage of mature DAMI cells (Feret's diameter ≥20 µm) is shown. At least 30 cells were quantified per individual experiment (n = 3). **P < 0.01; ***P < 0.001. C) Imaging of PMA‐treated NC and shMyosin‐9 DAMI cells stained with Wright–Giemsa. The black arrows show morphological mature DAMI cells with multiple nuclei. Scale bars, 20 µm. D,E) Detection of DNA content (≥4 N) in PMA‐treated NC and shMyosin‐9 DAMI cells (n = 3). ****P < 0.0001. F) Representative histograms and statistical analysis of CD41a expression in PMA‐treated NC and shMyosin‐9 DAMI cells determined by flow cytometry. The normalized geometric mean fluorescence intensity is shown (n = 3). *P < 0.05; **P < 0.01; ***P < 0.001. G,H) Western blot assays were conducted to analyze the expression of the Stat3/Erk pathway from TPO/c‐Mpl signaling in PMA‐treated NC and shMyosin‐9 DAMI cells. Quantitative analysis of p‐Stat3/p‐Erk is shown. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. I‐P) Same as in Figure 3A–H but with shRab6A instead of shMyosin‐9 DAMI cells. The data are shown as the mean ± SD of three independent experiments.

Figure 4

The formation of the Myosin‐9/Rab6A trafficking complex is dependent on Rab6A GTPase A) Co‐IP immunoblotting shows that the coimmunoprecipitation of Myc‐Myosin‐9 and Flag‐c‐Mpl was destroyed by Rab6A interference in HEK‐293T cells. B) Co‐IP of Myc‐Myosin‐9 and Flag‐c‐Mpl immunoblotting shows that Rab6A‐Q72L promotes the coprecipitation of c‐Mpl with Myosin‐9, while Rab‐T27N does opposite. C,D) Immunofluorescence staining (C) and normalized quantitative analysis (D) of c‐Mpl expression on the surface of the indicated UT‐7 cells. At least 30 cells were quantified per individual experiment (n = 3). c‐Mpl, red; GFP, green. Scale bars, 10 µm. *P < 0.05; **P < 0.01; ***P < 0.001. E‐G) Western blot assays were conducted to analyze the expression of the Stat3/Erk pathway in the indicated HEK‐293T cells with rhTPO added. Quantitative analysis of p‐Stat3/p‐Erk is shown. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. The means ± SDs are shown for the statistical analysis.

Figure 5

Adiponectin promotes the distribution of c‐Mpl on cell membranes via enhancing the expression of Rab6A GEFs A) Co‐IP immunoblotting of GFP‐Rab6A and Flag‐c‐Mpl shows that the coimmunoprecipitation of c‐Mpl and Rab6A is promoted by adiponectin (Adi). B) Representative histograms and statistical analysis of c‐Mpl expression on surface of NC and adiponectin treated UT‐7 cells determined via flow cytometry. The geometric mean fluorescence intensity is shown(n = 3). ***P < 0.001. C,D) The relative mRNA expression (C) and protein expression (D) of adiponectin receptor 2 (AdipoR2) from the mice bone marrow. (n = 3). **P < 0.01. E) The relative mRNA expression of AdipoR2 in the DAMI cells after the adiponectin treatment. ****P < 0.0001. F) IF images and the normalized quantitative analysis of the expression of AdipoR2 on cell surface from the adiponectin treated DAMI cells. At least 30 cells were quantified per individual experiment (n = 3). AdipoR2, green. Scale bars, 10 µm. **P < 0.01. G) Co‐IP immunoblotting of Flag‐c‐Mpl with GFP‐Rab6A shows that the coimmunoprecipitation of c‐Mpl and Rab6A was down‐regulated by siAPPL1. H,I) Co‐IP immunoblotting of HA‐Ric1(H) and Myc‐Rgp1(I) with GFP‐Rab6A and GFP‐Rab6A mutants shows that the Ric1 and Rgp1 bind to Rab6A‐T27N. J) The mRNA relative expression of Ric1 and Rgp1 in the mice bone marrow. (n = 3). **P < 0.01; ****P < 0.0001. K) The mRNA relative expression of Ric1 and Rgp1 in the DAMI cells after the adiponectin treatment. (n = 3). *P < 0.05; **P < 0.01. L) Co‐IP immunoblotting of Flag‐c‐Mpl with GFP‐Rab6A shows that overexpression (OE) of GEF promoted the binding of c‐Mpl to Rab6A, whereas knockdown (KD) of GEF decreases the binding. Overexpression or knockdown of GEF represents simultaneous overexpression or knockdown of Ric1 and Rgp1. M) Representative histograms and statistical analysis for CD41a expression of indicated DAMI cells by flow cytometry. Normalized geometric mean fluorescence intensity is shown(n = 3). *P < 0.05; ***P < 0.001. N) Western blot assays were conducted to analyze the expression of Stat3/Erk pathway from TPO/c‐Mpl signaling in PBS or adiponectin treated DAMI cells. Quantitative analysis of p‐Stat3/p‐Erk was shown. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data are shown as mean ± SD of three independent experiments.

Figure 6

Adiponectin combined with rhTPO increases therapeutic efficacy in ITP mice. A) Schematic outline of the experimental strategy of ITP mice is shown. B) Peripheral blood platelet counts of mice subjected to different treatments were monitored at different time points (n = 5). One‐way ANOVA. **P < 0.01; ***P < 0.001; ****P < 0.0001. C,D) Platelet counts at day 8 (C) and day 10 (D). **P < 0.01; ****P < 0.0001. ##P < 0.01, ####P < 0.0001, versus the PBS group treated with rhTPO and adiponectin; &, interaction effect; two‐way ANOVA. E) IF confocal microscopy images show the expression of c‐Mpl on the MK surface in the bone marrow of mice subjected to different treatments. CD41a, red; c‐Mpl, green. Scale bars, 10 µm. F) Statistical analysis of c‐Mpl expression on the surface of mouse MKs (n = 4). **P < 0.01; ***P < 0.001; ****P < 0.0001. G) Imaging of megakaryocytes stained with Wright–Giemsa from mouse bone marrow. H) Quantitative analysis of the mice MK diameter. At least 30 cells were quantified per mouse bone marrow (n = 4). ****P < 0.0001. The means ± SDs are shown for the statistical analysis.