Roles And Mechanisms In Regulation Of T Cell Alloreactivity By Natural Killer Cells In Graft-versus-host Disease

In the current study, we investigated the relationship between the restitution of NK cells receptoires and aGVHD after allogeneic hematopoietic stem cell transplantation(Allo-HSCT). Then we unveiled the roles and mechanisms of natural killer cells(NK cells) on regulation of T cell alloreactivity in acute graft-versus-host disease in vitro and in animal models. Finally we explored the modulation of proliferation, mobilization and biological function of NK cells by two immunoregulatory agents-dasatinib and AMD3100. Our research will help us to understand how NK cells regulate the T cell response in aGVHD, to replenish the mechanism of aGVHD regulation, and to establish biological marks for aGVHD prediction. It will also will provide experimental evidences for establishing a novel strategy for GVHD prevention and treatment.Our study includes3parts:part1to unveiled the relationship between the restitution of NK cells receptoires and aGVHD after Allo-HSCT; part2to investigate the roles and mechanisms of inhibition of T cell alloreactivity in aGVHD by NK cells; part3to explore the modulation of NK cells’ mobilization, proliferation and function of GVHD regulation by AMD3100and dasatinib.Part1The relationship between the restitution of NK cells receptoires and aGVHD after Allo-HSCT Aim:We detected the dynamic restitution regularity of NK cell receptoires after Allo-HSCT, analyzed the relationship between the restitution of NK cells receptoires and aGVHD, aimed to explore the role of donor NK cells in regulation of aGVHD.Methods:Patients who underwent allo-HSCT in Bone Marrow Transplantation Center, the First Affiliated Hospital, School of Medicine Zhejiang University from April2012to December2013were selected.Peripheral blood was drawn from63donors and63recipients receiving allogeneic hematopoietic stem cell transplantation(allo-HSCT) every week for3months. We comprehensively evaluated the subgroup, receptors expression of NKG2D/DNAM-1/NKP46/NKG2A and functions of NK cells by flow cytometry. CD25/CD69/MICA/MICB/PVR/Nectin2expression on T cells were masured simaltaneously.Results:63patients were included in study. The Median age was30years (12-50).Among them,22patients received HLA-matched related HSCT,14patients received HLA-matched unrelated HSCT, and27patients received HLA-haploidentical related HSCT.36patients developed aGVHD (57.14%). For those who developed aGVHD, NK cell percent in grafts was8.95±6.38%, NK cell dose was (9.87±14.72)*106/Kg, NK:T ratio was0.14±0.09.There was no significant difference in gaft NK percent and NK dose between aGVHD and no-aGVHD group, Whereas the NK:T ratio in no-aGVHD group was higher than aGVHD gropu(P<0.05).In the early period after HSCT, reconstituted NK cells were mainly CD56bright and NKG2A+subgroup, and percent of CD11b+CD27+subgroup was significantly higher than in health donors.76.00±9.20%NK cells in aGVHD patients were positive for NKG2A, much more higher than that in no-aGVHD patients(59.59±6.10%).NK cells in GVHD patients expressed lower level of NKG2D and CD226when compared with NK cells in no-aGVHD recipients or their corresponding donors. T cells in aGVHD patients expressed higher level of PVR(a ligand for DNAM-1) and MICA/B(a ligand for NKG2D) when compared with T cells in no-aGVHD recipients or their corresponding donors. In aGVHD patients, both degranulation and cytotoxicity of NK cells towards activated auto-T cells were depressed. NK cells in aGVHD up-regulated the expression of DNAM-1and NKG2D when stimulated ex vivo with IL-15, but not IL-2. Both IL-2and IL-15enhanced the degranulation and cytotoxicity of NK cells towards activated T cells.Conclusion:NK cells are the first lymphocyte subset to reconstitute after Allo-HSCT. The restitution of NK cells receptoires and NK cells’function recovery after Allo-HSCT impact the development of aGVHD. In this part of our research, we demonstrated that the NK:T ratio in grafts and expression of NKG2D/DNAM-1/NKG2A receptors on early reconstituted NK cells are correlated with aGVHD. The ability of donor NK cells to inhibit and lyse autologous activated T cells is impaired during human GVHD, possiblely due to altered subgroup and down regulated activated receptors. The restitution of NK cells receptoires and NK cells’function recovery after Allo-HSCT may represent a suitable markers for early predictation for aGVHD. Part2Regulation of T cell alloreactivity by NK cells in aGVHD and the underlying mechanismsAim:The aim of this part is to investigate the immunoregulatory roles of donor NK cells in regulation of alloreactive T cells response in aGVHD in animal models, and unveil the underlying mechanisms in this process by in vitro cell research.Methods:Haploidentical GVHD model was established in F1hybrid mice. Animals were co-injected with different ratio of T cells and NK cells. Survival time, GVHD manifestations, proliferation and apoptosis of T cells in the transplanted mice were evaluated. CD3+T cels and CD56+NK cells were sorting from donor peripheral blood mononuclear cell(PBMCs) by fluorescence activated cell sorting (FACS). We compared the proliferation rate of CFSE labeled donor Tcell stimulated by phytohemagglutination(PHA), allogeneic dendric cells(Allo-DC) and anti-CD3/ anti-CD28and cocultured with donor NK cells in different NK/T ratio. CFSE-7AAD based cytotoxicity assay and CD107a degranulation assay were used to detect the cytotoxicity against activated donor T cells by autologous NK cells. In blocking experiments, NK cells were pretreated with anti-NKG2D, anti-DNAM-1, anti-LFA-1, anti-FAS-L, anti-NKG-2A or anti-TIM-3mAbs. Western blot was used to detect the phosphorylation of ERK in NK cells after interacting with activated donor T cells.Results:1) In mouse haploidentical aGVHD model, NK cells prevented the development of aGVHD, inhibited the proliferation of T cells and promoted the apoptosis of activated T cells in the transplanted mice;2) Donor T cell proliferation in response to PHA, anti-CD3/CD28and Allo-DCs were suppressed by auto-NK cells, and the inhibiting effects were depended on NK/T ratio;3) Donor NK cells killed proliferating T cells(CFSElow) more efficiently than nonproliferating T cells (CFSEhigh) at all E:T ratios tested. Donor NK cells degranulated in response to activated, but not resting T cells. CD56dim degranulated more than CD56bright NKcells, and NKG2A’ degranulated more than NKG2A+NK cells;4) Activated T cells expressed high levels of the NKG2D ligand s(MIC-A, MIC-B and ULBP-1) and DNAM-1ligand PVR;5) Blocking of LFA-1, NKG2D or DNAM-1led to significant reduction of NK cell cytotoxicity, whereas blocking of NKG2A and TIM-3resulted in an increase in NK cytotoxicity;6) Immunoblotting assay revealed that phosphorylation of ERK was induced10to30mins after NK cells cocultured with activated donor T cells.Conclusion:Our findings demonstrate that NK cells negatively regulate T cells response to allogeneic antigen in mouse aGVHD model and human cell research. Activated donor T cells are susceptive to NK cytotoxicity by up-regulation of ligands for activated receptors expressed by NK cells. ERK signal pathway participates the cytotoxicity process of NK cells against activated T cells. This study provides new insight into the role of NK cells in the regulation of aGVHD. Part3The modulation of NK cell function of aGVHD regulation by AMD3100and TKIsAim:Recent studies have showed the immunoregulatory effects of agents administrated during Allo-HSCT. The aim of this part is to explore the modulation of NK cells’mobilization, proliferation and function of GVHD regulation by AMD3100and TKIs(Imatinib, nilotinib and dasatinib).Methods:Six to8week-old CB6F1mouse (a cross between female BALB/c and male C57BL/6) were divided into4groups based on mobilization protocol:Group1, subcutaneous injection (SC) of PBS as control; Group2, AMD-3100,5mg/kg, single dose; Group3, G-CSF,250ug/kg.d for5days; Group4, AMD3100plus G-CSF. Blood samples were taken from tail vein for cell counting and flow cytometric analysis at baseline (before treatment) and at1,2, and4hours after treatment. Splenocytes from CB6F1mouse and their parent mouse C57BL/6were labeled with0.5μM(CFSE low) and10μM CFSE(CFSE hi) respectively and transfered into CB6F1mouse2h after above treatments. We analyzed the in vivo cytotoxcity of NK cells by detecting the percent of CFSElow and CFSEhigh cells by flow cytometry4days after transfusion. Splenic NK cells(CD3e-DX5+) were sorted by FACS from CB6F1mouse4h after mobilization, then incubated with YAC-1cells at indicated E/T ratio for4h to determine the degranulation and cytotoxicity of NK cells in vitro.Human peripheral blood mononuclear cells (PBMCs) were isolated from adult healthy volunteers and cultured with SCGM medium(CellGenix) containing10%human AB serum, IL-2and IL-15for expansion of NK cells. After Imatinib, nilotinib and dasatinib were added to the culture at gradient concentration, cell counting and phenotyping by Flow Cytometry were used to evaluate the effect of TKIs on NK cell amplification. Subsequently, degranulation assay and CFSE/7AAD based cytotoxicity assay were used to detect the effects of dasatinib on NK cytotoxicity against leukemia cells and activated T cells.Results:single-dose AMD3100(5mg/kg) for hematopoietic stem cell mobilization showed notable effect on NK cell mobilization. Both AMD-3100alone and AMD3100plus G-CSF induced the mobilization of CD3e-DX5+NK cells into blood, the percentage and absolute number of NK cells increased after1hour and reached the maximal values2hours after treatment of AMD3100(±G-CSF), but not G-CSF alone.In PBS control mice69.09±2.64%parental target cells were killed by F1NK cells. We found that an increased proportion of parental target cells were killed by F1NK cells after administration of AMD3100alone(80.67±4.87%) or AMD3100pluse G-CSF (76.03±3.39%), whereas an decreased proportion of parental target cells were killed after administration of G-CSF alone(62.14±2.34%). In vitro function assay showed that NK cells purified from AMD3100injected mouse exhibited similar in vitro cytotoxic activity against YAC-1cells at different E/T ratio, but NK cells from G-CSF plus AMD3100or G-CSF alone treated mouse showed impaired cytotoxic activity when E/T=10:1or5:1. Similar results were detected in the degranulation assay. AMD3100alone showed no significant effect in CD107a expression, but G-CSF plus AMD3100or G-CSF alone lead to lower CD107a expression in NK cells.Among Imatinib, nilotinib and dasatinib, only dasatinib could enhance the efficiency of NK cell amplification in vitro. As the concentration of dasatinib increased, the percent of CD3-CD56+NK cells in culture increased gradually. However, high concentration of dasatinib led to apoptosis of NK cells, so the absolute number of NK cells in culture reached the maximal values when20nM dasatinib was used. Dasatinib treated NK cells could inhibit the proliferation of alloreactive T cells more effectively when compared with control. Furthmore, dasatinib improved the NK cytotoxicity against activated T cells and leukemia cells. Dasatinib increased the NKG2A-CD57+NK cells and up-regulated the expression of activating receptor (CD226, NKP46, NKG2D) on NK cells in culture.Conclusion:Single-dose AMD3100for hematopoietic stem cell mobilization increased the number of NK cells in blood without functional impairment of cytotoxicity. Therefore, AMD3100-based mobilization may enhance the NK dose in grafts and promote the GVHD regulatory effect of NK cells, and improve the outcome of allo-HSCT. Appropriate dose of dasatinib increases the amplification efficiency of NK cells, simultaneously upregulates NK activating receptors and increases the NKG2A-CD57+subgrop, which lead to enhanced NK cytotoxicty against activated T cells and leukemia cells. So AMD3100and Dasatinib may optimize the amplification program of NK cells, promote NK cell function of GVL and GVHD regulation after Allo-HSCT.