Clinical Graft-versus-host Disease Research Articles

Redundant Mechanisms for Dendritic Cell Activation in GVHD Induction: Signalings Via TLRs, TNF-α, IL-1 and CD40 Are Not Required

Graft-versus-host disease (GVHD) is initiated by antigen-presenting cells (APCs) that prime alloreactive donor T cells. Amongst APCs, dendritic cells (DCs) are the most potent at priming naïve T cells. A paradigm of DC function holds that instructive maturation signals transform “immature” DCs into “mature” DCs which are optimized for T cell activation. Signaling via Toll-like receptors (TLRs) has emerged as a central mechanism for inducing this maturation in anti-pathogen responses. However, in allogeneic stem cell transplantation, there is no specific pathogen and the signals that induce DC maturation in GVH responses have not been defined. To study potential maturation stimuli we performed GVHD experiments in model systems wherein host APCs are essential for GVHD induction (C3H.SW (H-2b)→B6 and B6bm12→B6). We then compared GVHD in wild type (wt) recipients to that in hosts genetically unable to respond to various DC maturation signals. We first used as hosts mice deficient in TLR4, CD40, IL1-R or TNFR1/R2 and found that in C3H.SW→B6 model, GVHD in each of these was similar. Likewise, GVHD was similar in wt and mice deficient in MyD88, essential for signaling via all TLRs except TLR3 and TLR4. To focus our studies on APCs, we used as hosts radiation bone marrow (BM) chimeras in which wt B6 (CD45.1) mice were reconstituted with BM from CD45.2 wt, MyD88−/−, TRIF−/− (required for signaling via TLR3 and, in-part, TLR4), and MyD88/TRIF−/− mice, in which signaling by all TLRs, IL-1 and IL-18 is completely blocked. When retransplanted with C3H.SW BM and CD8 cells, all chimeras developed similar clinical and histologic GVHD. To exclude that T cells were being primed by residual wt host APCs in BM chimeric recipients, we used as hosts B6bm12 mice that were reconstituted with BM from B6 wt, MyD88−/−, TRIF−/− or MyD88/TRIF−/− mice. These chimeras were then retransplanted with B6bm12 BM and CD4 cells. Because residual host APCs in these BM chimeric hosts were B6bm12 and therefore syngeneic with the donor cells, T cell priming could only occur on B6 background APCs derived from the first transplant. As in the prior experiments, all chimeras developed similar clinical GVHD; pathology scoring is underway. Therefore in contrast to nearly all adaptive T cell responses, GVHD induction in both MHC-matched and MHC-mismatched systems did not require APC maturation signals mediated by TLRs. Absent any phenotype when APC maturation signals were blocked, we used the same approach in the C3H.SW→B6 model to study the roles of IL12 and type I IFNs, which are produced as a consequence of DC maturation. B6 p35−/− and B6 wt mice developed similar GVHD, excluding an essential role for host IL12. Type I IFN receptor-deficient (IFNIR−/−) mice developed similar GVHD as did wt controls, except that IFNIR−/− CD8 recipients lost more weight but had less histologic liver GVHD (P=0.0024). To focus on the activity of Type I IFNs on host hematopoietic cells, we used as recipients IFNIR−/− → B6 chimeras. Both clinical and histologic GVHD were similar in IFNIR−/−→B6 and wt→B6 control chimeras, suggesting that the differences in weight loss and hepatic GVHD observed in IFNIR−/− hosts, were due to parenchymal expression of IFNIR−/−. In sum, these data indicate that DC maturation signals in GVHD are at a minimum redundant, and independent of TLR ligands such as LPS, which are certainly available to the host immune system. This may be a consequence of the ubiquitous expression of alloantigen, such that any maturation signal will create APCs primed to activate donor T cells. If so, strategies for preventing GVHD by blocking APC maturation may need to target final common APC maturation pathways. Alternatively, instructive APC maturation may not be required for alloreactive T cell activation in GVHD.

Recipient Langerhans Cells Are Neither Required Nor Sufficient for GVHD Induction in MHC-Matched Allogeneic BMT, but a Langerin+ Cell Is a Pivotal Regulator of Langerhans Cell Turnover Post Transplantation

Graft-versus-host disease (GVHD) is initiated when alloreactive donor T cells are primed by professional antigen-presenting cells (APCs) to undergo clonal expansion and maturation. Host APCs that survive pretransplant conditioning play an essential role in this T cell activation, and are an attractive target for GVHD prevention and treatment. However, APCs are diverse in phenotype, location and function and an understanding of the roles of distinct subsets is an important first step in developing APC-targeted therapies. Skin is the most frequently affected organ in GVHD. Langerhans cells (LCs), characterized by expression of Langerin, are a major APC in the epidermis, and thus it was logical to hypothesize that host LCs would have a role in GVHD induction. Indeed, in an MHC-mismatched model, Merad et al. showed that host LCs persist after T cell-depleted (TCD) but not T cell-replete bone marrow transplant (BMT), and that these host LCs in donor→host chimeras are sufficient to induce skin GVHD after a second allogeneic bone marrow transplant (alloBMT). However, this work did not examine the role of recipient LCs when all other APCs are intact, the scenario at the time of transplant in all patients. To address this question, we created a transgenic mouse that constitutively lacks epidermal LCs. We did so by expressing diphtheria toxin A chain (DTA) driven by the human Langerin gene (Kaplan, et al 2005) in a bacterial artificial chromosome (BAC). We used Langerin-DTA BAC transgene positive (Tg+) mice or Tg-littermates as recipients in the C3H.SW (H-2b)→B6 (H-2b) strain paring, in which recipient APCs are necessary and sufficient for GVHD induction. Tg+ and Tg− CD8 recipients developed similar GVHD as measured by weight loss and clinical skin disease. Tg+ and Tg− CD8 recipients also had comparable pathologic GVHD of the skin, ear, liver and colon. To generalize these findings, we used B6bm12 →B6 strain pairing, an MHCII-mismatched CD4-dependent GVHD model, in which recipient APCs are also required (Teshima et al, 2002). Tg+ and Tg− CD4 recipients developed similar weight loss and pathologic changes in the tongue and liver, primary sites of GVHD in this model. Thus, in both MHC-matched and MHC-mismatched models in which recipient APCs are absolutely required, the specific absence of recipient epidermal LCs did not affect clinical or histological GVHD. We also analyzed LC turnover in these alloBMT recipients. As previously reported, LCs remained host-derived in B6 Tg− recipients of TCD C3H.SW bone marrow. Given our prior result that C3H.SW → B6 chimeras are resistant to GVHD induction by a second alloBMT from C3H. SW donors (Shlomchik, et al 1999), unlike in the MHC-mismatched model employed by Merad, residual host LCs are insufficient to initiate GVHD in this MHC-matched system. In B6 Tg− recipients of TCD C3H.SW bone marrow plus GVHD-inducing CD8 cells, LC turnover varied by mouse and ranged from all host or donor to a mix of donor and host LCs. This variability could relate to the extent of skin GVHD, as we previously found that epidermal MHCII+ cells in skin GVHD lesions in this model are donor-derived (Matte et all, 2004). Strikingly, in contrast to Tg− recipients, donor-derived LCs developed in Tg+ recipients of TCD C3H.SW bone marrow. Donor LCs also engrafted in Tg+ recipients of TCD bone marrow from Tg− but otherwise syngeneic littermates or B6 RAG1−/− T cell-deficient donors. Thus, in contrast to LC-replete mice, neither allogeneic donor T cells nor UV-induced inflammation was required for donor LC engraftment in LC-deficient hosts. These data indicate that a Langerin+ cell, absent in Langerin-DTA Tg+ mice, regulates LC turnover in the absence of inflammation. Work is underway to identify this key cell.

SELDI-TOF-Based Proteomic Analysis of Plasma Samples as a Diagnostic Tool to Predict Acute Graft-Versus-Host Disease after Allogeneic Stem Cell Transplantation

Allogeneic hematopoietic stem cell transplantation (SCT) offers a curative treatment option for high-risk hematological tumors. Graft-versus-host disease (GvHD) is a severe and potentially life-threatening complication of allogeneic SCT. GvHD is caused by the activation of transplanted alloreactive T-cells and results in severe inflammation of host tissues in skin, liver, and gastrointestinal tract. Manifest GvHD can be diagnosed by clinical criteria and histology only. So far, no diagnostic blood test is able to detect imminent GvHD. GvHD is prevented and treated by systemic immunosuppression which is associated with opportunistic infections and higher relapse rates. Therefore, reliable detection of incipient GvHD would allow to investigate preemptive immunosuppressive therapy tailored to individual patients at appropriate time points. Proteomic profiling is based on the premise that alterations related to pathological states (e.g., GvHD) are accompanied by quantitative and qualitative alterations in plasma protein profiles. The surface-enhanced laser desorption/ionization (SELDI) technology relies on time-of-flight mass spectrometry for accurate measurement of the mass-to-charge ratio (m/z) of proteins that have been preselected on appropriate functional surfaces. We conducted a prospective case-control study based on >3000 serial plasma samples of 236 consecutive patients undergoing allogeneic SCT. 43 patients developed acute GvHD grade I-IV during their sampling period and were matched to control patients remaining GvHD-free. Matching criteria were prior antithymocyte globulin treatment, donor type (related versus unrelated), conditioning protocol, remission status at SCT, diagnosis, age, and sex. These factors were ranked in the listed order. Plasma protein patterns were obtained by SELDI-TOF on day -2 prior to the onset of clinical GvHD and compared to 43 plasma samples obtained on the corresponding day after SCT from the matched GvHD-free patients. In order to enhance the number of analyzable proteins, plasma samples were fractionated to yield 5 different sets of SELDI spectra. Protein peak positions and calibrated peak heights were determined according to Yasui et al (J Biomed Biotechnol, 2003). To identify a limited number of predictive peaks, a sparse logistic regression model was fitted by a boosting approach (Tutz & Binder, Computational Statistics & Data Analysis, 2007), resulting in 4 peaks at 11963, 22363, 22479, and 30977 Da. Prediction performance of a signature derived from this approach on new data was evaluated using bootstrap samples, resulting in an estimated misclassification rate of 38.5%, an estimated sensitivity of 61.0%, and specificity of 60.5% In addition, the identified predictive peaks could be recovered in most bootstrap samples, indicating sufficient stability of the signature. While the performance of this assay strategy is insufficient for use in clinical practice at present, our data nevertheless demonstrate the potential of proteomic plasma analysis in combination with a tailored statistical analysis to identify GvHD prior to its clinically recognizable onset. In contrast to the clinical routine of immunosuppressive GvHD prophylaxis for all patients after allogeneic SCT with treatment escalation in the case of clinically relevant GvHD, this diagnostic tool therefore offers the perspective for preemptive GvHD therapy administered only to patients with an imminent GvH reaction. Such a tailored GvHD strategy may represent a cornerstone to reduce the risk of infectious complications and tumor relapse while maintaining effective GvHD control in patients at risk. Current work addresses the identity of the index peaks, the temporal evolution of the signature, and refinement of the proteomic analysis.

Depletion of Vascular Endothelial Progenitor Cells Inhibits Inflammation

Neo-vascularization has been implicated in a number of inflammatory diseases as well as tumor growth. Both angiogenesis, the sprouting of resident tissue endothelial cells (ECs), and vasculogenesis, the recruitment of bone marrow (BM)-derived circulating endothelial progenitor cells (EPCs), are thought to participate in neo-vascularization. EPCs have been implicated in tumor growth, however, the biologic significance of EPCs during inflammation is unclear.We studied neo-vascularization and the role of EPCs during inflammation in well-characterized murine models of graft-versus-host disease (GVHD). We found a significantly increased number of donor BM-derived EPCs in peripheral blood and BM in allogeneic bone marrow transplantation (allo-BMT) recipients with GVHD at different time points after BMT. We next quantified neo-vascularization during inflammation in GVHD target organs by immunofluorescence microscopy and by flow cytometry. We found significantly increased numbers of donor-derived ECs in the liver as well as a significantly higher vessel density in the liver, illeum and colon. We adoptively transferred selected GFP+ EPCs and observed incorporation into the neo-vasculature of the inflamed intestines (Fig. 1A) and liver during GVHD. Taken together, these data suggest that neovascularization during GVHD is due to vasculogenesis from donor EPCs.Next we used an antibody (E4G10) against the vascular endothelial adhesion molecule VE-cadherin, which recognizes a terminal epitope that is exposed on circulating EPCs, but is masked in the established vasculature, and found a significant reduction of EPCs in the peripheral blood and BM. We observed that depletion of EPCs was associated with a significant inhibition of donor BM-derived neo-vascularization in the liver, illeum and colon during GVHD. E4G10 treated recipients had significantly better survival and lower clinical GVHD scores in all tested models (B6BALB/c [1×106 T], B6B6D2F1 [1×106 T], B6B6D2F1 [2×106 T], B6B6D2F1 [3×106 T]). We found significantly reduced numbers of allo-reactive donor T cells in secondary lymphoid organs during GVHD, but no changes in the expression of activation markers and homing molecules, as a consequence of E4G10 administration. In blinded histopathological analyses we found significantly less GVHD and reduced numbers of tissue-infiltrating CD3+ T cells in the liver, illeum and colon in E4G10-treated allo-BMT recipients.To better emulate the clinical setting, we first assessed the role of EPCs in tumor growth in allo-BMT recipients. We transferred sorted GFP+ EPCs as well as renal carcinoma (RENCA) cells to BALB/c recipients and found that GFP+ EPCs were recruited to the neo-vasculature of lung metastases. We detected a significant inhibitory effect of E4G10 administration on tumor growth, as determined by in vivo bioluminescence imaging, in both tumors tested (RENCA, A20 lymphoma) as well as a significant survival prolongation in tumor-bearing mice that were treated with E4G10 in the RENCA and C1498 (AML) model.Finally, we performed experiments in which tumor-bearing allo-BMT recipients received allogeneic T cells, which mediate the favorable graft-versus-tumor (GVT) effect but also cause inflammation in GVHD target organs. We found that administration of E4G10 led to a significantly higher rate of tumor-free survival in all models (B6'BALB/c [1×105 B6 T and 2×105 RENCA], B10BR'B6 [1×105 B10BR T and 2×105 C1498], B6'BALB/c [2×105 B6 T and 5×105 A20]), which was due to both attenuation of GVHD as well as inhibition of tumor growth (Fig. 1B).We conclude that depletion of EPCs is a strategy to simultaneously ameliorate inflammatory disease and tumors, providing a new approach to improve therapeutic outcome of allogeneic hematopoietic stem cell transplantation. This study demonstrates the biological significance of EPCs for neo-vascularization during inflammation and identifies the specific targeting of EPCs to disrupt neo-vascularization as a novel therapeutic concept to decrease inflammation. [Display omitted]

Prevention of Acute GvHD during Haplo-BMT: Evaluating the Efficacy of T-Cell Costimulation Blockade Using a Novel Rhesus Macaque Transplant Model

Introduction: While bone marrow transplantation (BMT) offers curative therapy for many malignant and non-malignant hematologic diseases, it remains plagued by high morbidity and mortality, most prominently through the complications of graft-versus-host disease (GvHD). The field of GvHD research has been limited by a small number of clinically relevant model systems, which, until now, have focused on murine and canine models. However, many new human therapeutics, including those directed at T cell costimulation pathways, do not cross-react with targets in these model systems, limiting their utility to fully evaluate the new generation biologic anti-GvHD therapies.Methods: We have developed a fully pedigreed and MHC-typed rhesus macaque bone marrow transplant system, with which to model GvHD and its prevention. This model relies upon microsatellite typing of both autosomal chromosomes to create genetically-verified family trees and microsatellite-and allele-specific PCR-based MHC typing, with which to identify MHC haplotypes within each family group. By this methodology, full and half-sibling pairs can be identified with either full MHC matching, one MHC haplotype matching or full MHC disparity. This represents a significant advance compared to previous primate haplo-studies, which relied upon parent-offspring transplants, as it optimizes animal availability and colony stability as well as permiting the rigorous evaluation of the degree of MHC disparity on the non-shared chromosome 6. For the current study, we have concentrated on MHC haplo-identical BMT. Our preparative regimen consists of TBI (8.00 Gy with lung shielding to 6.00 Gy), which is performed on the day of transplant with hematopoietic stem cells. After transplant, donor engraftment is measured by microsatellite analysis, from both whole blood and from flow cytometrically sorted T and B cell populations. GvHD is graded using standard clinical grading scales. The immune phenotype after transplant is also measured by cell- and serum-based flow cytometric analyses.Results: Three animals have so far been studied. The first animal served as a control for TBI-based preparation. This animal exhibited both neutropenia (nadir = 0.015 × 103/uL at 12 days) and thrombocytopenia (nadir = 2 × 103/uL at 12 days) and prolonged anemia (hgb=5.8 g/dL on the day of sacrifice). This animal was sacrificed at day 29 due to prolonged pancytopenia. The second animal was transplanted with haploidentical hematopoietic stem cells (4.47 × 108 total nucleated cells/kg and 1.10 × 108 CD3+ T cells/kg) and was treated with only rapamycin for immunosuppression. He exhibited profuse diarrhea and necrotic skin changes within 8 days of transplant, coincident with early engraftment and was sacrificed at day 14. A full histologic analysis to determine the cause of these lesions (and the histopathologic extent of GvHD) is currently underway. The third animal was transplanted with haploidentical donor bone marrow (1.1 × 109 total nucleated cells/kg and 6.9 × 107 CD3+ T cells/kg), but was given T cell costimulation blockade with abatacept (targeting the CD28/B7 pathway) and an anti-CD40 antibody (targeting the CD40/CD154 pathway) in addition to rapamycin. Abatacept is currently clinically approved for treatment of rheumatoid arthritis, but has not been added to clinical GvHD regimens, largely due to the paucity of translational efficacy data. This animal is currently >43 days post-transplant, with full donor chimerism (including 100% T and B cell chimerism), without signs of GvHD: he has no rash, no diarrhea and is stable clinically. Additional animals are now being added to the study, to confirm these initial results.Conclusions:We have established a robust non-human primate model of GvHD using pedigreed and MHC-typed rhesus macaques.We have used this model to begin to test the efficacy of a novel agent combination, capable of preventing the onset and complications of acute GvHD.Preliminary results suggest that CD28- and CD40-directed costimulation blockade may be active agents for the prevention of GvHD. A large scale analysis of their efficacy and immune consequences is currently underway.

Expanded donor natural killer cell and IL‐2, IL‐15 treatment efficacy in allogeneic hematopoietic stem cell transplantation

Graft-versus-host disease (GVHD), leukemia relapse, and immune deficiency remain the major limitations of allogeneic hematopoietic stem cell transplantation (allo-HSCT). Donor natural killer cells (NK) and cytokines have good potential in GVHD prevention and immune reconstitution enhancement. Improved survival after allo-HSCT therefore requires effective prophylaxis to reduce GVHD and strategy to mediate graft-versus-leukemia (GVL) effect. We studied the administration of expanded donor NK cell infusion and interleukin-2 (IL-2) and IL-15 mixture treatment in a murine allo-HSCT model for its effects on GVHD, immune reconstitution and leukemia relapse. In the GVHD model, recipient mice were reconstituted with bone marrow (BM) cells and splenocytes via vein. In the leukemia model, recipient mice were inoculated with EL9611 leukemia cells via vein 8 d prior to transplant. NK cell infusion mice group had lower clinical GVHD scores and suffered less severe GVHD-associated weight loss than control mice group. 90% of control mice died of leukemia relapse within 52 d post-transplant. NK cell infusion and IL-2 and IL-15 treatment recipient mice had improved survival compared with control mice. NK cell infusion and IL-2 and IL-15 treatment recipient mice had also accelerated lymphoid immune reconstitution compared with control mice group. Expanded donor NK cell infusion and IL-2 and IL-15 treatment could promote lymphoid immune reconstitution, mitigate GVHD, and reduce leukemia relapse in allo-HSCT recipients. The findings may have important significance for complication prevention in clinical allo-HSCT.

A Preliminary Gene Expression Profile of Acute Graft-Versus-Host Disease

Allogeneic hematopoietic stem cell transplantation (HSCT) is the treatment of choice for high-risk hematological malignancies, yet a major complication associated with this therapy is acute graft-versus-host disease (GVHD). Despite a well-defined pathophysiological mechanism, there are no definitive markers for predicting acute GVHD development or progression to advanced stages. In the current study, we enrolled four acute GVHD and four acute GVHD-free recipients of allogeneic HSCT and collected peripheral blood just prior to onset of clinical acute GVHD for analysis on Affymetrix GeneChip Human Genome U133 Plus 2.0 microarrays. We noted significant differences in expression of 1,658 genes between control and acute GVHD patients, based on an analysis of covariance (ANCOVA) by type of transplant, a pooled error estimate, and a false discovery rate (FDR) of 10%. In conclusion, we offer the first report of a preliminary molecular signature of acute GVHD in allogeneic HSCT patients.

Combined Th2 cytokine deficiency in donor T cells aggravates experimental acute graft-vs-host disease

The role of T helper (Th) 1 and Th2 polarization in acute graft-vs-host-disease (GVHD) is unclear. We investigated the role of Th2 cytokine secretion by utilizing donor T cells that cannot make interleukin (IL)-4, IL-5, IL-9, and IL-13 from quadruple cytokine-deficient (Quad-KO) animals in a well-characterized BALB/c-->C57BL/6 model of allogeneic bone marrow transplantation. B6 recipients of BALB/c Quad-KO T cells demonstrated greater clinical severity, target organ damage, and mortality from GVHD than recipients of BALB/c wild-type (WT) T cells. When compared with donor T cells that are deficient in signal transducers and activators of transcription 6 signaling or the signature Th2 cytokine, IL-4, Quad-KO T cells demonstrated greater GVHD mortality. Mechanistic studies demonstrated that Quad-KO T cells demonstrated enhanced T-cell proliferation than WT T cells when stimulated with either allogeneic antigen-presenting cells or with nonspecific stimuli, such as anti-CD3 monoclonal antibody. Quad-KO T cells also secreted greater amounts of Th1 cytokines and IL-17 compared to WT T cells. Deficiency of Th2 cytokines, however, did not alter the allospecific cytotoxic responses, the numbers of immunoregulatory CD4(+)CD25(+) Foxp3(+) T cells or their suppressive functions. Our data thus unequivocally demonstrate that deficiency of the four classical Th2 cytokine enhances T-cell proliferative responses and aggravates GVHD.

CD134-Allodepletion Allows Selective Elimination of Alloreactive Human T Cells without Loss of Virus-Specific and Leukemia-Specific Effectors

Graft-versus-host disease (GVHD) remains a frequent and severe complication of allogeneic stem cell transplantation (SCT). One approach to reducing alloreactivity is to deplete the graft of alloreactive T cells. Global T cell depletion results in poor immune reconstitution with high mortality from viral infections and disease relapse. Therefore, an approach to selectively deplete alloreactive T cells without compromising other responses would be highly beneficial. We undertook studies to identify an inducible activation marker expressed on alloreactive effector T cells following culture with HLA-mismatched allostimulators. Compared to other markers, CD134 was superior because of its negative baseline expression and rapid upregulation after activation. Depletion of CD134 + cells from responder populations dramatically reduced specific alloreactivity as determined by reduction of helper T cell precursor frequencies below the threshold predicting development of clinical GVHD while retaining responses to third-party alloantigens. CD134-allodepleted populations retained effectors specific for the Wilms' tumor (WT1) leukemia antigen as determined by WT1 specific pentamers, and CMV-specific effectors as determined by CMV-specific pentamers and CMV-specific ELISpot. Thus, use of CD134-allodepleted grafts may improve allogeneic SCT by reducing GVHD without loss of pathogen-specific and leukemia-specific immunity.

Murine bone marrow derived stromal progenitor cells fail to prevent or treat acute graft‐versus‐host disease

Human and murine stromal progenitor cells (SPCs) can suppress alloresponse in vitro, suggesting that SPCs may have clinical application toward prevention or treatment of graft-versus-host disease (GVHD). However, th eresults of in vivo studies have been conflicting. This study utilized an established murine model of acute GVHD to assess the ability of bone marrow derived murine SPCs (mSPCs) to prevent or treat GVHD. GVHD was established by transplantation of B6 bone marrow and spleen cells into lethally irradiated (900 cGy) B6 x BALB/c F(1) recipients. mSPCs were administered using various dose and timing protocols designed to either prevent or treat GVHD. After transplantation, mice were monitored daily for weight and survival. Differences in symptom severity were compared using a clinical GVHD scoring system. All GVHD control mice died of lethal GVHD. All groups treated with mSPCs for the prevention of GVHD went on to develop clinical GVHD with no alteration of the disease course or severity compared to controls. Administration of mSPC after the development of GVHD failed to improve the disease course. We conclude that in this model, the ability of SPCs to suppress alloresponse in vitro does not correlate with in vivo prevention or treatment of acute GVHD.

Predominant donor CD103+CD8+ T cell infiltration into the gut epithelium during acute GvHD: a role of gut lymph nodes

The existence of donor effector cell subsets responsible for either gut or skin graft-versus-host disease (GvHD) is still undetermined. We examined the trafficking and role of donor CD8(+) intra-epithelial lymphocytes (IELs) in the gut and skin epithelia concerning alpha E beta 7 integrin (CD103) expression, using a rat acute lethal GvHD model. Most CD103(+) donor cells were CD8(+) and showed a proliferative activity in the target epithelia. On the other hand, activated donor T cells in the host lymphoid tissues did not express CD103, indicating the presence of CD8(+) IEL precursors in the lymphoid tissues that may up-regulate CD103 only after migrating to the target organs. At the late stage of GvHD, while >80% of the donor CD8(+) IELs were CD103(+) in the gut epithelium, both CD4(+) and CD103(+)CD8(+) T cells evenly accumulated in the skin epidermis. The CD103 expression by donor CD8(+) IELs especially in the gut was also correlated with the clinical GvHD manifestations. Furthermore, the selective removal of gut lymph nodes (LNs) but not skin LNs suppressed the infiltration of CD103(+) donor IELs in the gut and alleviated intestinal GvHD. In conclusion, CD103(+)CD8(+) donor T cells predominantly infiltrate into the gut epithelium and are responsible for the manifestations of intestinal GvHD. This pathology is at least partly dependent on the gut LNs.

459: Mapping Graft Versus Host Disease after Allogeneic Stem Cell Transplant-a Prospective Approach to Recognize and Capture Signs and Symptoms of Graft Versus Host Disease (GVHD) and Evaluate Treatment Effectiveness

Initial assessment begins pretransplant by recording information on a timed assessment calendar. It includes patient name, age, diagnosis, performance status, HLA matching status, source of stem cells (related, unrelated, hablo, cord) peripheral blood versus bone marrow, CD34/kg of graft, CD 3/kg of graft, departmental specific immunological reconstitution post transplant research protocol, banking protocol for donor/recipient pairs (DNA). The post transplant portion of the assessment calendar uses a time line for fixed assessment dates: day zero, fourteen, thirty, sixty, one hundred, one hundred eighty, two hundred seventy and one year. A monthly or more frequent assessment is warranted if active GVHD is being treated. The validated assessment tool addresses all mayor systems and documents the onset and degree f the present organ involvement. GVHD treatment methods are documented with specifics such as type of treatment, current dosage, onset, effectiveness- ineffectiveness and discontinuation. This also includes Standard of Care prophylaxis including: antibiotic, antiviral, anti fungal and anti PCP. Clinical GVHD findings are validated whenever possible through tissue and blood sampling and other testing (such as skin, mucosal biopsies, blood markers, Schiller test). A multidisciplinary team is formed to participate in the treatment of GVHD. The core team members are from departments of Dermatology, Ophthalmology, Pulmonology, GastroEnterology, Gynecology as well as Bone and Mineral, Physical therapy and Psychology. Regular discussion meeting dates are set to present current complex GVHD patient issues. The team also explores new treatment options. Whenever possible the patient will be enrolled into institutional or national GVHD clinical trials. Literature searches for pertinent GVHD information and collegial information exchanges are encouraged. Consideration is given to non-traditional approaches for example acupressure-puncture. A future goal is establishing a clinic for patients with prolonged severe clinical GVHD symptoms and those enrolled in GVHD clinical trials.

Incidence and Clinical Outcome of Intestinal Thrombotic MicroAngiopathy (TMA) in Patients Suspected to Clinical Graft Versus Host Disease (GVHD).

Intestinal TMA, a upcoming and serious problem after HSCT, usually manifests diarrhea, vomiting, GI bleeding and abdominal pain as initial clinical presentations similar to those of intestinal GVHD. But intestinal TMA should be distinguished from GVHD, because they needed to be treated in absolutely different way. Here we investigated the incidence of intestinal TMA and evaluated their clinical progressions in HSCT patients. We retrospectively reviewed 243 gastrointestinal mucosal biopsy slides obtained from HSCT patients having suffered from GI problems in the Catholic University Hospital, from 2000 to 2004. The collected cases were reviewed by two pathologists blindly and classified into TMA; showing microthrombi formation in the lumina of microvasculatures, GVHD; showing epithelial apoptosis and/or loss of glands, TMA+GVHD and non-specific inflammation. And next, their incidence and clinical outcomes were analyzed. On histopathological examination, 15 out of 243 cases (6.2%) showed luminal thrombi of microvasculature compatible with TMA, 58 cases (23.9%) showed findings compatible with GVHD, 3 cases (1.2%) showed both TMA and GVHD, and 173 cases (71.2%) showed nonspecific gastroenteritis. On peripheral blood smear examination, all 15 TMA cases showed less than 1% of fragmented RBC at the time of biopsy. But 2 out of 15 cases (13.3%) showed increase to 8% and 10% of fragmented RBC in follow-up, and they had been aggravated clinically and improved only after treatments including discontinuing immunosuppressant and plasma exchange. Among 58 GVHDs, 34 patients (75.9%) were improved with steroid pulse therapy, but 14 patients (24.1%) failed. These results show that intestinal TMA is not infrequent in HSCT patients and can be combined with intestinal GVHD. In conclusion, intestinal TMA should be recognized in pathological diagnosis of HSCT patients. It is important to differentiate intestinal TMA from GVHD in patients suffering from severe and refractory diarrhea after HSCT.

Lymphopenia Promotes the Development of Graft-Versus-Host Disease upon Donor Leukocyte Infusion.

In murine models of hematopoeitic cell transplantation (HCT), a graft-versus-leukemia (GVL) effect can be effectively separated from graft-versus-host disease (GVHD) by giving delayed donor-leukocyte infusions (DLI) to established mixed hematopoietic chimeras (MC) (Mapara, Transplantation 2003). In these models, the absence of conditioning-induced inflammation at the time of DLI prevents trafficking to GVHD target tissues, limiting the GVH response to the lymphohematopoeitic system (Chakraverty, J Exp Med 2006). While clinical trials have shown that the same outcome is possible in humans, some patients develop severe GVHD following delayed DLI (Dey, Blood Biol Marrow Transplant 2003). In contrast to mice, who recover T cell numbers quickly after HCT through de novo thymopoiesis, humans remain lymphopenic for many months. The current study investigates the role of recipient lymphopenia in the development of GVHD following delayed DLI to MC. Allogeneic (B10.A) or syngeneic (B6) lymphocytes (30 × 106 whole splenocytes) and bone marrow cells (10 × 106) were administered to unconditioned RAG–1 knockout (KO) B6 recipients or to established MC, and animals were followed for development of GVHD. By day 7 after injection, the RAG KO recipients of allogeneic lymphocytes (allogeneic group) had lost significantly more weight (p<0.001) and had significantly higher clinical GVHD scores (p<0.0001) than RAG KO recipients of syngeneic lymphocytes (syngeneic group) or MC recipients of allogeneic lymphocytes (MC group). Histologic evaluation on day 7 showed significantly increased GVHD in liver and colon of the allogeneic group. Chimerism analysis on day 7 showed the presence of an ongoing lymphohematopoietic GVH response (LGVHR) in both the allogeneic and MC groups, though the response was slower in the MC group. When followed over time, clinical GVHD in the allogeneic group resolved by day 56, weight differences disappeared by day 65, but mild histologic GVHD persisted as late as day 239. Importantly, the same dose of allogeneic T cells given to freshly conditioned animals results in rapid, lethal GVHD, whereas these unconditioned, lymphopenic animals developed only a relatively mild, largely self-limited GVHD. These results demonstrate that in the absence of conditioning-induced inflammation, lymphopenic recipients of allogeneic T cells do develop mild GVHD, consistent with the possibility that recipient lymphopenia contributes to the GVHD seen in human MC recipients of DLI. The mechanisms by which lymphopenia promotes GVHD in non-conditioned MC recipients of DLI require further investigation, but may include the absence of regulatory T cells, the potential for lymphopenia-induced proliferation of T cells, and the presence of infection-induced inflammation.

Host Plasmacytoid or Conventional Dendritic Cells Alone Are Sufficient To Initiate Graft-Versus-Host Disease.

Graft-versus-host disease (GVHD) is a major complication after allogeneic hematopoietic stem cell transplantation. Alloantigen expression on host dendritic cells (DCs) is critical to initiate GVHD. DCs can be divided into two main subpopulations; conventional DCs (cDCs) and plasmacytoid DCs (pDCs), however, the contribution of each DC subset to elicit GVHD remains unclear. We examined the ability of cDCs and pDCs to initiate GVHD. pDCs, cDCs and B cells were isolated from C57BL/6 (B6: H–2b) mice treated with Flt3 ligand in order to expand DCs. pDCs were enriched from bone marrow by depleting CD3+, CD19+, CD11b+, and CD49b+ cells, followed by a FACS sorting of CD11cint B220+ cells. cDCs and B cells were sorted from splenocytes as CD11chi B220− cells and CD11c− B220+ cells, respectively. Isolated pDCs showed plasmacytoid morphology, produced IFN-α in response to CpG oligonucleotide. Although pDCs stimulated allogeneic T cells far less potently than cDCs, stimulation with CpG enhanced their allostimulatory capacity as potent as cDCs. We compared the ability of each DC subset to initiate GVHD by an add-back study of MHC class II-expressing DCs into MHC class II-deficient (II−/−) mice that were resistant to CD4-dependent GVHD. Lethally irradiated II−/− B6 mice were injected with 2 × 106 pDCs, cDCs or B cells from wild-type (II+/+) B6 mice on day -1 and injected with 2 × 106 CD4+ T cell from BALB/c (H–2d) mice on day 0. A flow cytometric analysis of the mesenteric lymph nodes on day +5 demonstrated significantly greater expansion of donor CD4+ T cells in recipients of pDCs or cDCs than those of B cells (Table). While injection of B cells did not cause any sign of GVHD, injection of pDCs or cDCs alone was sufficient to produce clinical and pathological GVHD (Table), thus breaking GVHD resistance of II−/− mice. We next examined the ability of pDCs to induce CD8-dependent GVHD in MHC-matched transplant using mice deficient in functional MHC class I expression (β2m−/−). Again, injection of pDCs or cDCs alone was sufficient to cause expansion of donor CD8+ T cells (p<0.05). We next asked whether signaling through Toll-like receptors (TLRs) could be required for pDCs to initiate GVHD. However, injection of pDCs isolated from MyD88/TRIF-double deficient mice was able to initiate GVHD as potent as wild-type pDCs, thus demonstrating that pDCs initiate GVHD in a TLR signaling independent manner. These results provide important information for developing strategies aimed at inactivating host DCs to prevent GVHD.Impact of each APC subpopulation on GVHDAPCDonor CD4 expansion (×103±SE)Clinical GVHD score (mean±SE)Pathological GVHD score (mean±SE)B cell0.1 ± 0.02.1 ± 0.22.1 ± 0.2pDC5.3 ± 2.4*4.3 ± 0.3*7.4 ± 0.5*cDC9.7 ± 3.8 *3.8 ± 0.5 *7.2 ± 0.7**p<0.05 compared with B cells

A Shift in the Intestinal Microflora towards Pro-Inflammatory Bacteria and Signaling Via Toll-Like-Receptor 2 Triggers Early Onset of Graft-Versus-Host Disease.

Pattern recognition receptors such as Toll-like-receptors (TLR) and NOD proteins bind specifically to microbial components and activate innate immune responses. Reducing the bacterial load by antibiotic treatment has been shown to ameliorate the severity of graft-versus-host-disease (GvHD) after allogenic stem cell transplantation. The impact of bacteria and TLRs on the induction of T-cell-alloreactivity and GvHD was investigated in a murine transplantation model. C57B/10 wild-type (wt) mice served as recipients of MHC-mismatched Balb/c grafts after conditioning with treosulfan and cyclophosphamide. Analyses of the gut microbiota in fecal samples showed a pronounced increase of luminal load of both gram-positive and gram-negative bacteria within 11 days after stem cell transplantation. The increase in the Escherichia coli load was significantly correlated with the severity of graft-versus-host-disease and with decreased survival rates. Analysis of the kinetics of fecal E. coli load and the histologial and clinical GvHD scores revealed that the increase of the fecal E. coli load is detected simultanously to the onset of clinical overt GvHD. Because TLR 2 and 4 represent the major receptors for gram-positive and gram-negative bacteria, respectively, the incidence and severity of GvHD was investigated in TLR2, TLR4 and TLR2/4 deficient mice on a C57B/10 background. The onset of GvHD was delayed in recipient mice lacking TLR2 and both TLR2 and 4 (p<0.05) whereas a single TLR4 deficiency had no influence. Interestingly, there was no correlation of E. coli load and GvHD in TLR2-deficient mice.Taken together, these results support potential roles of TLR2-dependent signaling and gut bacterial components in the induction and early onset of intestinal GvHD. In expansion of these findings, our data elucidate the role of the innate immune system and its receptors for the induction of alloreactive T-cell responses, providing evidence that modulation of the intestinal gut flora might prevent or reduce GvHD in humans.

Donor Lymphocyte Infusion for Mixed Chimerism or Residual Disease after Reduced-Intensity T Cell Depleted Stem Cell Transplantation Results in Conversion to Full Donor Chimerism Combined with Graft Versus Tumor Responses and Limited GVHD.

We perform reduced intensity allogeneic stem cell transplantation (SCT) with HLA identical sibling and matched unrelated donors using a conditioning regimen consisting of fludarabine, busulphan and ATG combined with alemtuzumab added to the stem cell graft for extensive donor T cell depletion. With this regimen hardly any acute and chronic graft versus host disease (GVHD) are observed after SCT and mixed chimerism is induced in most patients (Barge et al, Exp Hematol 2003). In patients transplanted for malignant disease, donor lymphocytes (DLI) are administered in escalating doses from 6 to 9 months to convert mixed into complete donor chimerism and to induce graft versus tumor (GVT) responses. We report results of chimerism analysis, clinical antitumor response and GVHD after DLI. 41 patients (median age 54 years, range 37–65) transplanted with stem cells from sibling donors (29) or matched unrelated donors (12) for various malignancies received DLI with a median dose of 5 × 106 CD3+ cells/kg (range 1–5) at a median time point of 7 months after SCT (range 5.1–15). After DLI administration 35 of 38 evaluable patients (92%) developed an immune response as defined as a major decrease in patient chimerism. In 26 patients full donor chimerism was achieved, in 9 patients a low patient signal remained present (1–2%). Patient chimerism decreased from median 12% (range 1 to 86%) to 0% (range 0–2%). In most patients conversion occurred after the first DLI, in 4 patients after the second and in 1 patient after the third DLI. In 31 of 38 patients measurable disease was present before DLI. In 22 of these patients conversion from mixed to full donor chimerism coincided with clinical GVT responses. Seven patients with active myeloid disease all achieved complete remission (4 AML/MDS patients with 8–65% blasts, 2 patients with progressive CML and 1 with active CMMOL). All these patients are still in CR after a median time of 22 months. In 8 myeloma patients with measurable disease 5 complete and 1 partial response were observed. However, most responding myeloma patients subsequently developed bone or extramedullary relapses without evidence of bone marrow involvement. In 13 patients with lymphoid disease 9 responses were observed. One patient with an immunocytoma and one with T-PLL achieved CR. In 7 CLL patients 3 CR and 1 PR were observed, notably two CR occurred in patients with massive bone marrow involvement. Two of four patients with aggressive NHL showed a CR to DLI in combination with rituximab and one patient a PR to DLI. Although no regression of tumor was observed in three patients with renal cell carcinoma after DLI, disease progression was halted for several years. Grade 3–4 acute GVHD developed in 22% of patients, grade 1–2 in 39%. Limited or extensive chronic GVHD was observed in 30 and 23% of patients, respectively. GVHD responded to therapy in most patients, only in 9% of patients chronic GVHD did not resolve. Mortality due to acute and chronic GVHD was 10%. In conclusion, after T cell depleted reduced intensity SCT a state of mixed chimerism is induced in most patients, which can be converted into full donor chimerism with DLI in more than 90% of patients. Conversion to full donor chimerism is accompanied with clinical GVT responses in a high percentage of patients with acceptable GVHD.

Benzodiazepine-423, an Inhibitor of Mitochondrial Respiration, Causes Selective Apoptosis of Activated Lymphocytes and Reverses Experimental GVHD While Preserving GVL Effects.

Benzodiazepine (Bz)-423 is a benzodiazepine derivative that targets activated lymphocytes through the mitochondrial F1F0-ATPase, causing loss of mitochondrial membrane potential and apoptosis without affecting resting lymphocytes. We tested Bz-423 in a non-irradiated B6-Ly5.2 → B6D2F1 model of graft-versus-host disease (GVHD) where donor cells were labeled with CFSE to discriminate 3 days after injection between activated (CFSElo) and unactivated (CFSEhi) Ly5.2+ donor or host Ly5.1+ lymphocytes. Compared to controls, Bz-423 caused loss of mitochondrial membrane polarization within 6h of delivery as measured by 3,3'-dihexyloxacarbocyanine iodide (DiOC6(3)) staining in both activated donor CD4+ (12.6% vs 3.3%, p=0.002) and activated CD8+ (12.9% vs 3.0%, p<0.001) T cells but not in unactivated donor or host cells. Loss of mitochondrial membrane potential was followed by selective apoptosis (Annexin-V+) of donor CD4+ (34.9% vs 16.5%, p=0.04) and CD8+ (29.8% vs 12.2%, p=0.03) T cells. Intraperitoneal injection of 60mg/kg Bz-423 3 times weekly beginning 7d after GVHD induction significantly reduced mortality (50% vs 100%, p<0.02). We next used Bz-423 in a miHA-disparate, CD8+ T cell-mediated model of GVHD (C3H.SW → B6) in which B6 hosts received 9 Gy of TBI followed by injection of 5x106 C3H.SW BM cells and 4x106 T cells. We initiated Bz-423 injections 7d after BMT, when GVHD was already established. The drug significantly reduced GVHD clinical scores and improved survival compared to controls (74% vs 29%, p≤0.02). Bz-423 also significantly reduced quantitative GVHD histologic damage indices in the liver (3.6 vs 11.2, p<0.03) and the GI tract (7.0 vs 15.8, p<0.02). Complete donor engraftment was observed in all animals. Bz-423 reduced IFN-γ, a known mediator of GVHD, in the serum (8.4 vs 21.7 pg/ml, p<0.03) and decreased IFN-γ+CD8+ effector spleen T cells (0.64x105 vs 2.2x105, p=0.008), but did not impair the lysis of tumor targets by CD8+ T cells ex vivo. We tested Bz-423 next in a graft-versus-leukemia (GVL) model where EL-4 lymphoma cells (4x103) that are syngeneic to B6 recipients were injected on the day of BMT. No recipients of syngeneic BMT survived EL-4 challenge (0/12) and no untreated allogeneic BMT survived GVHD (0/12) but 9/14 (64%) of Bz-423 treated allogeneic recipients were alive on day 50 without evidence of lymphoma (p=0.003, Fig 1a). We confirmed the effectiveness of Bz-423 in a third model of GVHD to MHC differences (Balb/c → B6) where again Bz-423 significantly reduced all clinical, biochemical and histologic GVHD parameters and improved day 60 survival (58% vs 8%, p<0.002). Bz-423 also preserved GVL effects in this model where 50% of recipients survived without evidence of EL-4 lymphoma compared to 0% of controls (p<0.04, Fig 1b). We conclude that Bz-423, a first-in-class compound that selectively inhibits mitochondrial respiration and causes apoptosis of activated lymphocytes, can reverse experimental GVHD while preserving beneficial GVL effects. [Display omitted]

Targeting CD134 Results in Selective Depletion of Alloreactive Human T Cells without Loss of Virus-Specific and Leukemia-Specific Effectors.

Graft versus host disease (GvHD) remains a frequent and severe complication of allogeneic stem cell transplantation (SCT) and is mediated by recognition of recipient alloantigens by donor T cells. Numerous interventions have been employed against GvHD which target the number and function of T cells including pharmacological non-specific immunosuppression, depletion of donor T cells or positive selection of donor CD34+ stem cells. While these strategies reduce the risk of GvHD, the benefits are often offset by prolonged immunodeficiency and increased risk of infection and disease relapse. Thus, research efforts have focused on developing a method to selectively deplete alloreactive T cells without compromising other immune responses. One approach to reducing alloreactivity is to remove donor T cells that are activated after ex-vivo co-culture with recipient antigen-presenting cells (APC). Clinical studies have been performed using anti-CD25-conjugated immunotoxin to deplete alloreactive T cells after incubation with haplo-identical PBMC ex-vivo. Although this approach is promising, post-transplant infections, disease relapse and residual GvHD remain a problem. These observations mandate further experimentation to improve our strategies for allodepletion. We undertook studies to identify an inducible surface activation marker expressed selectively on alloreactive effector T cells following culture with HLA-mismatched allostimulators. Costimulatory molecules play a major role in activation of the effector T cells involved in GvHD. In contrast to CD28, which is constitutively expressed, TNF receptor (TNFR) superfamily members are induced and therefore their expression can also be used as an indicator of T cell activation during GvHD, in experimental animal models and allogeneic human transplantation. Here, using peripheral blood mononuclear cells as responders and HLA-mismatched dendritic cells as allostimulators, we observed that compared to other activation markers the TNFR superfamily member, CD134, was superior because of its completely negative base-line expression and rapid upregulation upon activation. Depletion of CD134+ cells from responder populations dramatically reduced specific alloreactivity as determined by reduction of helper T cell precursor (HTLp) frequencies below the threshold predicting development of clinical GvHD and reduction of cytolytic T cell precursor (CTLp) frequencies by 92% against original allostimulators, while retaining responses to third party alloantigens. However, effector CD8+ T cells expressing perforin and granzyme B were preserved in the CD134 allo-depleted populations. Detailed analysis revealed that, by this approach, functional CMV-specific effectors were retained as determined by their identification with CMV-specific pentamers and assessment of their functional responses by CMV-specific ELISpot. Effectors specific for WT1 leukemia antigen were also retained. In addition, Foxp3+ Treg cells were preserved in the allodepleted populations. These observations suggest that CD134-based selective allodepletion may allow allogeneic SCT without requirement for global T cell depletion or non-specific immunosuppression and without loss of pathogen-specific and leukemia-specific immunity.

Therapeutic Effect of CXCR3-Expressing Regulatory T Cells on Liver and Intestinal Damages in a Murine Acute GVHD Model.

[Purpose] Adoptive transfer of CD4+CD25+ regulatory T cells has been shown to have therapeutic effects in experimental graft-versus-host disease (GVHD) models. Chemokines play an important role in the recruitment of alloreactive donor T cells into target organs during GVHD. In this study, we investigated the effectiveness of targeted delivery of CD4+CD25+ regulatory T cells via a transfected chemokine receptor on reduction of organ damage during acute GVHD.[Materials & Methods] Haploidentical parent-to-F1 transplants were performed, in which lethally irradiated B6D2F1 mice were given T cell-depleted BM cells supplemented with splenic T cells from C57BL/6. Expression of Th1-associated chemokines (CXCL9, CXCL10, and CXCL11) and their receptor CXCR3 in target organs during acute GVHD was examined with real-time PCR and ELISA assay. CD4+CD25+Foxp3+ T cells (Treg cells) were reprogrammed by infection of recombinant retrovirus carrying Foxp3 gene into cultured CD4+CD25+ T cells of donor C57BL/6 mice. CD4+CD25+Foxp3+ CXCR3-transfected T cells (CXCR3-Treg cells) were isolated by FACS sorting after infection with recombinant retrovirus carrying CXCR3 into Treg cells. Treg and CXCR3-Treg cells were transferred 7 days after BMT and histopathological analysis was preformed 35 days after BMT. The infiltration of Treg and CXCR3-Treg cells into target organs was analyzed with Foxp3 expression and fluorescence.[Results] High levels of expression of Th1-associated chemokines and their receptor CXCR3 were observed in the liver and intestines of GVHD-induced recipient mice from 1 to 3 weeks post-BMT. There was no significant difference in suppressive properties between Treg and CXCR3-Treg cells. However, in the CXCL10-driven chemotactic assay, the chemotactic activity of CXCR3-Treg cells was much higher (>100-fold) than that of Treg cells. Either 4 × 105 or 1 × 106 cells were injected into the mice 1 week after BMT, and the mice were then monitored in terms of clinical GVHD score, before being sacrified four weeks after the injection of regulatory T cells for evaluation of the pathological changes. Recipient mice that had undergone transfer of 1 × 106 CXCR3-Treg cells demonstrated a significantly reduced (P<0.01) clinical GVHD score and weight loss compared with Treg cell-transferred recipients (at 4 weeks after injection: clinical GVHD score, 0.25 ± 0.5 vs 5.25 ± 0.5; and weight change, 104.4% ± 1.9% vs 87.2% ± 4.7%, respectively). The recipients that had undergone transfer of 1 × 106 CXCR3-Treg cells showed a significantly reduced (P<0.05) histopathological GVHD score for the liver and intestines compared with Treg cell-transferred recipients (liver, 1.36 ± 0.38 vs 7.71 ± 1.07; and intestines, 1.00 vs 5.00 ± 1.35, respectively). The localization of CXCR3-Treg cells to the liver and intestines was increased significantly at 24 h and 7 days after transfer compared with that of Treg cells, and the regional localization continued to increase until 28 days post-BMT. These findings indicate that, in comparison with Treg cells, more CXCR3-Treg cells migrate into CXCL10-expressing target organs, and remain localized there for a longer time, resulting in stronger suppressive activity.[Conclusion] We succeeded in preparing chemokine receptor-expressing Treg cells and demonstrated their ability to ameliorate disease progression upon accumulation in target organs. This method may provide a new therapeutic approach for organ damage in acute GVHD.