Optimization of the therapeutic efficacy of human umbilical cord blood–mesenchymal stromal cells in an NSG mouse xenograft model of graft-versus-host disease

Cytotherapy, 2014; 16: 298e308

ORIGINAL PAPERS

Optimization of the therapeutic efficacy of human umbilical cord bloodemesenchymal stromal cells in an NSG mouse xenograft model of graft-versus-host disease

YUN KYUNG JANG1, MIYEON KIM1, YOUNG-HO LEE2, WONIL OH1, YOON SUN YANG1 & SOO JIN CHOI1 1

Biomedical Research Institute, MEDIPOST Co., Ltd., Seoul, Korea, and 2Department of Pediatrics, Hanyang University Medical Center, Seoul, Korea

Abstract Background aims. Although in vitro studies have demonstrated the immunosuppressive capacity of mesenchymal stromal cells (MSCs), most in vivo studies on graft-versus-host disease (GVHD) have focused on prevention, and the therapeutic effect of MSCs is controversial. Moreover, optimal time intervals for infusing MSCs have not been established. Methods. We attempted to evaluate whether human umbilical cord bloodeMSCs (hUCB-MSCs) could either prevent or treat GVHD in an NSG mouse xenograft model by injection of MSCs before or after in vivo clearance. Mice were infused with either a single dose or multiple doses of 5
105 hUCB-MSCs (3- or 7-day intervals) before or after GVHD onset. Results. Before onset, hUCB-MSCs significantly improved the survival rate only when repeatedly injected at 3-day intervals. In contrast, single or repeated injections after GVHD onset significantly increased the survival rate and effectively attenuated tissue damage and inflammation. Furthermore, the levels of prostaglandin E2 and transforming growth factor-b1 increased significantly, whereas the level of interferon-g decreased significantly in all MSC treatment groups. Conclusions. These data establish the optimal time intervals for preventing GVHD and show that hUCB-MSCs effectively attenuated symptoms and improved survival rate when administered after the onset of GVDH. Key Words: graft-versus-host disease, human umbilical cord bloodemesenchymal stromal cells, NSG mice, therapeutic effect

Introduction Although allogeneic hematopoietic stem cell transplantation can cure certain malignant and nonmalignant diseases, graft-versus-host disease (GVHD) represents the most common complication despite the use of transplants from human leukocyte antigen (HLA)-matched siblings (1,2). GVHD can be induced by multiple factors, including conditioning regimens, total-body irradiation, the bone marrow (BM) microenvironment, patient and donor age or sex, stem cell source and graft composition (3,4). However, most cases of GVHD are caused by the reaction of transplanted T cells with histoincompatible antigens of the recipient. The ensuing proliferation or activation of other immune cells leads to a wide variety of injuries to host tissues caused by the release of inflammatory cytokines (5).

Corticosteroids are generally used as primary treatment for acute GVHD and are more effective when combined with immunosuppressive agents such as cyclosporine or methotrexate (4). Primary treatment with steroids improves skin, liver or gastrointestinal tract lesions (4) and increases the probability of survival (1-year survival: approximately 50%) (6,7). Patients with steroid-resistant GVHD are usually offered second-line therapy such as antithymocyte globulin; however, 31% of patients show initial improvement in signs and symptoms, particularly involving the skin, and only 10% have longterm survival (12e60 months) (8). Therefore, new approaches for increased survival rate are required, and recent strategies have introduced the infusion of mesenchymal stromal cells (MSCs) to take advantage of their immunoregulatory properties (9e12).

Correspondence: Soo Jin Choi, MD, PhD, Biomedical Research Institute, MEDIPOST Co., Ltd., 1571e17 Seocho3-dong, Seocho-gu, Seoul 137e874, Korea. E-mail: [email protected] (Received 28 May 2013; accepted 22 October 2013) ISSN 1465-3249 Copyright Ó 2014, International Society for Cellular Therapy. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcyt.2013.10.012

Optimized MSC therapy of GVHD MSCs suppress T-cell proliferation or cytotoxicity induced by cellular or humoral stimuli in a dosedependent manner (13,14) and mediate T-cell suppression by secreted factors, such as prostaglandin E2 (PGE2) (15), transforming growth factor-b1 (TGFb1) and hepatocyte growth factor (16). Moreover, because there is no immunologic restriction on their ability to suppress T cells, MSCs provide an important tool for transplantation biology. Recent studies on MSCs in vivo have been facilitated by a GVHD animal model (17e20). However, despite the immunosuppressive capacity of MSCs, studies on GVHD focus on prevention, and the use of MSCs for treatment is controversial. Moreover, the therapeutic effect of human umbilical cord blood (hUCB)-MSCs in a xenograft model of GVHD (xeno-GVHD) has not been demonstrated (17) and the optimum dosing frequency has not been established. Therefore, the goal of the present study was to determine optimal time intervals for infusing MSCs and to assess the efficacy of hUCBMSCs for preventing and treating GVHD. Methods Human peripheral blood mononuclear cells Human peripheral blood mononuclear cells (hPBMCs) were acquired from Astarte Biologics (Redmond, WA, USA) for research use only after written informed consent was given by the donor. The hPBMCs were stored in their blood bank, in compliance with the Health Insurance Portability and Accountability Act. We purchased the hPBMCs used to induce GVHD from Lifeline Cell Technology (Frederick, MD, USA), which is the distributor for Astarte Biologics. We stored the hPBMCs at 196 C until use. hUCB-MSCs Culture conditions for hUCB-MSCs were the same as those described in our previous study (21). Human UCB was obtained after written informed consent was given by normal full-term pregnant women. The hUCBMNCs were isolated with the use of a Histopaque (1.077 g/mL, Sigma-Aldrich, St Louis, MO, USA) density gradient centrifuge at 400 g for 30 min and were grown to a density of 5
106 cells/cm2 in 175-cm2 flasks (Nunc, Roskilde, Denmark) in a-minimum essential medium (a-MEM; Gibco/Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Gibco/Life Technologies). Cultures were maintained at 37 C in a humidified atmosphere containing 5% CO2, and half of the media was replaced with fresh media twice each week. After spindle-shaped colonies were observed, cells were harvested with the use of 0.25% trypsineethylenediaminetetraacetic

299

acid (EDTA) (Gibco/Life Technologies) and grown to 80% confluence. Because the immunosuppressive activity of hUCB-MSCs differs for each donor, we selected the optimal lot that significantly reduced lymphocyte proliferation (mixed lymphocyte reaction assays) and secretion of interferon-g (IFN-g enzymelinked immunosorbent spot assays kit [BD Biosciences, San Jose, CA, USA]) (Supplementary Figure 1). We confirmed the phenotype, differentiation ability and immunosuppressive activity of the UCB-MSC lots and show the respective results in Supplementary Figures 1 and 2. Additional information regarding hUCB-MSC preparation is provided in the Supplementary information. Cells were cryopreserved in liquid nitrogen until use. Xenogeneic GVHD mouse model NSG (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ, 7 weeks old) male mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA) and allowed to acclimate for 2 weeks before the experiments. All animal experiments and use of hPBMCs and hUCBMSCs were approved in advance by the Institutional Review Board and the Institutional Animal Care and Use Committee, College of Medicine, Hanyang University, Seoul, Republic of Korea, respectively (permit No. HY-IACUC-10e030, 10e046, 11e057, 12e019). Mice (9 weeks old) were irradiated by means of 2.0 Gy with the use of a Gammacell 1000/ 3000 irradiator (Best Theratonics Ltd, Ottawa, Ontario, Canada). The control (group 1) did not receive hPBMCs or irradiation, and group 2 only was irradiated. The viability of the hPBMCs after thawing was >95%, and hPBMCs (group 3, 0.5
106; group 4, 1
106 or group 5, 2.5
106) were transplanted intravenously within 24 h after irradiation to establish acute or chronic xenogeneic GVHD animal models. Survival, weight loss, fur texture, physical activity, skin integrity and hunched back were recorded every 3e4 days. The severity of GVHD was assessed by means of a clinical scoring system described by Cooke et al. (22). A clinical GVHD index was generated by summing the five criterion scores (0e10), and we evaluated acute and chronic GVHD according to overall survival rate, GVHD clinical score and histological analysis until day 100. Five mice per group were used, and experiments were performed three times. Two mice in each group were euthanized for histological analysis on day 24. Lungs, liver, kidneys and the small intestine were removed to observe morphological changes and lymphocyte infiltration. Paraffin embedding, sectioning, hematoxylin and eosin staining and analysis were performed at the Tissue and Cell (T & C) Pathology Centre (Seoul, Korea), and a clinical pathologist determined the score according to

300

Y. K. Jang et al.

lymphocyte infiltration (0, normal; 0.5, focal and rare; 1, focal and mild; 2, diffuse and mild; 3, diffuse and moderate; 4, diffuse and severe). Tissues affected by GVHD were selected by referring to the results for NOG mice (23). Hematoxylin and eosin estained sections were observed at a magnification of
200 with the use of an inverted microscope (Nikon Eclipse TS100; Nikon Co, Tokyo, Japan) and photographed with the use of a Nikon Digital Sight DS-Fi1 camera (Nikon). Images were analyzed with the use of NISElements F software (Nikon). Analysis of inflammation To establish the optimum time for inducing GVHD, mice were infused with 1
106 hPBMCs, and blood samples were taken on days 0, 3, 12, 18, 24, 30 or 36. Three mice were euthanized at each time point. The mice were anesthetized with Rompun (0.002 mL/10 g [5 mg/kg]) (Bayer, Leverkusen, Germany) and Zoletil (0.008 mL/10 g [40 mg/kg]) (Virbac, Carros, France), and circulating whole blood was collected by means of heart puncture and transferred to a K3EDTA Minicollect tube (Greiner Bio-One GmbH, Kremsmünster, Austria) for separating plasma. Whole blood was centrifuged at 3000 rpm for 10 min, and the plasma was analyzed for levels of human IFN-g, human tumor necrosis factor (TNF)-a, and human interleukin (IL)-2 with the use of Quantikine enzymelinked immunoassay (ELISA) kits purchased from R&D Systems (Minneapolis, MN, USA). Samples were analyzed according to the manufacturer’s instructions. Inflammation was determined in two independent experiments, and plasma collected at each time point was analyzed in triplicate. Optical density (OD) was measured at 450 nm with the use of a 540-nm correction (VersaMax Microplate Reader; Molecular Devices, Sunnyvale, CA, USA). Data are expressed as mean  standard deviation values. Efficacy of naive human UCB-MSCs for preventing or treating GVHD Mice were irradiated by means of 2.0 Gy, and 1
106 hPBMCs were transplanted intravenously within 24 h as described above. The hUCB-MSCs were thawed and washed three times in phenol redefree a-MEM (Gibco/Life Technologies) and resuspended to 5
105 cells/0.1 mL in phenol redefree a-MEM (Gibco/Life Technologies) in sterilized 1.5-mL tubes (AXYGEN Scientific Inc, Union City, CA, USA). For studies on preventing GVHD, 5
105 hUCB-MSCs were infused into the tail vein once (day 0, group 2), on days 0, 3 and 6 (group 3) or days 0, 7 and 14 (group 4) (Figure 1). For studies on treatment, hUCB-MSCs were infused into the tail vein either once on day 18

Figure 1. Experimental scheme. After establishing the optimum hPBMC dose and the xeno-GVHD model, the onset time of GVHD for infusing hUCB-MSCs was determined by means of analysis of inflammatory cytokine levels. The hUCB-MSCs were infused with single or repeated doses. Survival rate, weight loss and GVHD clinical scores were observed until day 60, and all surviving mice were euthanized for cytokine and histological analyses.

(group 5); on days 18, 21 and 24 (group 6), or days 18, 25 and 32 (group 7) after the onset of GVHD (Figure 1). Cryopreserved hUCB-MSCs were used within 3 h after thawing for all experiments. The control mice (group 1, hPBMCs only) were injected with 0.1 mL phenol redefree a-MEM instead of hUCBMSCs. Prevention and treatment experiments were performed six and three times, respectively, with the use of 10 mice from each group. Clinical score, anti-inflammatory cytokine and histological analyses The survival rate, weight loss, fur texture, physical activity, skin integrity and hunched back of the control (hPBMCs only) and experimental (hUCBMSC infusion) groups were observed every 3e4 days for 60 days, and the severity of GVHD was assessed as described above. The average values of the data are shown in Table I. On day 60, all surviving mice were euthanized for histological and anti-inflammatory cytokine analyses. Mice were anesthetized and blood was collected, processed as described above, and the plasma samples were analyzed with the use of the following kits: PGE2 parameter kit, human TGF-b1 Quantikine ELISA kit and human IFN-g Quantikine ELISA kit (R&D Systems) according to the manufacturer’s instructions. Assays were performed in triplicate, and OD values were measured and expressed as described above. Histological analyses were performed as described above. Because

Optimized MSC therapy of GVHD

301

Table I. GVHD clinical score after infusion of mice with naive hUCB-MSCs. Prevention effect

Item Weight loss Posture Activity Fur texture Skin integrity Total

Group 1 (PBMC only) 4 2 2 1 1 10

Group 2 (day 0) 3.4 1.0 1.0 1.0 1.0 7.4

Group 3

Treatment effect Group 4

(days 0, 3 and 6) (days 0, 7 and 14) 1.4 0.4 0.4 0.3 0.4 2.9

1.7 0.8 0.7 0.5 0.7 4.4

Group 5 (day 18) 1.0 0.5 0.3 0.1 0.1 2.0

Group 6

Group 7

(days 18, 21 and 24) (days 18, 25 and 32) 1.0 0.5 0.5 0.3 0.2 2.5

1.0 0.7 0.5 0.1 0.2 2.5

The GVHD clinical score was based on weight loss (1w10%, 1; 11w20%, 2; 21w30%, 3; 31w40%, 4), posture, activity, fur texture and skin integrity (0, normal; 1, mildly abnormal; and 2, severely abnormal). For comparisons with group 1, all groups were evaluated every 3 days from days 24e45. The score shown in the table represents the average value of day 45 (n ¼ 5 or 6).

the death rate of the control group (group 1, received PBMC) was 100% at day 60, two of the surviving mice in group 1 at days 42e45 were euthanized for PGE2, TGF-b1, IFN-g and histological analysis. The two mice were also excluded from the overall survival rate in every experiment. Statistical analysis Survival curves were generated by means of the Kaplan Meier method, and two or more survival curves were compared by means of the log-rank test. Analyses were performed with the use of GraphPad Prism version 4.0 (GraphPad Software, San Diego, CA, USA). Values of P < 0.05 were considered statistically significant. Results Mouse models for acute and chronic GVHD established by optimizing the dose of engrafted hPBMCs To establish mouse xenograft models for acute and chronic GVHD, we first determined the optimum dose of hPBMCs by intravenously injecting 0.5
106, 1
106 or 2.5
106 cells into irradiated NSG mice. The untreated control group (group 1) and the group that received radiation only (group 2) survived for more than 100 days (Figure 2A,B) and did not have symptoms of GVHD such as weight loss, alopecia, hunched back and reduced physical activity. Mice in group 3 (injected with 0.5
106 hPBMCs) showed symptoms of GVHD from day 24, and four of five survived longer than 100 days despite weight loss, alopecia, hunched back and reduced physical activity. In contrast, mice of group 4 (1
106 hPBMCs) or group 5 (2.5
106 hPBMCs) died within 1e1.5 months. The onset of GVHD in group 5 mice was earlier than in group 4 mice, and symptoms began to appear between days 15e20. All mice of group 5 died (Figure 2A,B), and the mean survival time was 24.0  10.4 days. Moderate and severe

lymphocyte infiltrations were observed in the lungs, liver, kidneys and small intestine at day 24 compared with groups 1 and 2 (Figure 2C). These experiments established that 0.5
106 and 2.5
106 hPBMCs were optimal for inducing chronic and acute xenoGVHD models. GVHD symptoms were observed between days 21e24 when 1
106 hPBMCs were transplanted into NSG mice (group 4). Mild and moderate lymphocyte infiltration was observed on day 24, whereas the results of groups 1 and 2 were normal (Figure 2C), and the mean survival time was 41.6  6.8 days. The onset time of GVHD for infusing hUCB-MSCs was determined on the basis of the progress of GVHD symptoms by infusion of 1
106 hPBMCs, and the hPBMC dose was used to study inflammation and to assess the ability of hUCB-MSCs to prevent and treat GVHD.

Inflammatory cytokines in a xenogeneic GVHD model To investigate inflammation in vivo after irradiation and transplantation of 1
106 hPBMCs, the levels of human IFN-g, TNF-a and IL-2 in plasma were determined. Secretion of IFN-g was first detected on day 12, increased dramatically after 18 days and high concentrations persisted until day 36 (Figure 3A). TNF-a was first detected on day 24 and was produced until day 36 (Figure 3B). In contrast, the concentration of IL-2 was unchanged (Figure 3C). When we compared the levels of the three inflammatory cytokines simultaneously, the increase in IFN-g and TNF-a levels correlated with weight loss and GVHD symptoms, although the increased level of TNF-a was lower than that of IFN-g (Figure 3D). Naive hUCB-MSCs can be used to prevent and treat GVHD In a prevention study for GVHD, the survival rates of mice that received either a single injection (group 2)

302

Y. K. Jang et al.

Figure 2. Establishment of acute and chronic xenogenic GVHD animal models with the use of NSG (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ) mice. Mice in groups 1 (n ¼ 4) and 2 (n ¼ 5) survived for 100 days (A), and there was no irradiation-induced weight loss or lymphocyte infiltration (B, C). Four of five mice survived until 100 days when injected with 0.5
106 hPBMCs (G3, n ¼ 5). Mice in groups injected with 1
106 hPBMCs (G4, n ¼ 5) or 2.5
106 hPBMCs (G5, n ¼ 5) died quickly (A), lost weight (B) and showed severe lymphocyte infiltration in their organs (C). Lymphocyte infiltration and blunt villus structures were commonly detected in groups 4 and 5 (magnification
200, scale bar ¼ 100 mm).

or multiple injections at 7-day intervals (group 4) were only 12.5% (versus group 1, P ¼ 0.4805) and 33.3% (versus group 1, P ¼ 0.1252; group 2 versus group 4, P ¼ 0.2083), respectively. In contrast, repeated injections at 3-day intervals (group 3) significantly improved the survival rate up to 70% (versus group 1, P < 0.001; versus group 2, P < 0.05; versus group 4, P ¼ 0.3667) (Figure 4A). The difference in weight loss was compared between days 24 and day 45. Significant weight loss was not observed in group 3 in comparison to groups 1 and 2 (P < 0.05). Between days 27 and 45, weight loss was also ameliorated in group 4 in comparison to group 1 (P < 0.05; day 38, P ¼ 0.05) (Figure 4B). Furthermore, group 3 showed remarkably reduced tissue damage, lymphocyte infiltration, and GVHD clinical scores compared with groups 1 (hPBMCs only) and 2 (Figure 4E; Tables I and II). Group 1 showed high infiltration with lymphoctyes in the lungs, liver, kidneys and small intestine (Table II). The alveolar structure of the lung was changed through lymphocyte infiltration and inflammation, and the villus structure of the small intestine was converted into a blunted form (Figure 4E).

Group 2 showed reduced lymphocyte infiltration in the liver, kidneys and small intestine but not in the lungs. However, group 3 showed dramatically improved histological structures and reduced lymphocyte infiltration levels (Figure 4E, Table II). In contrast, group 4 showed alleviated inflammatory damage in the small intestine; however, the lungs, liver and kidneys showed high infiltration with lymphocytes (Figure 4E, Table II). Therefore, multiple infusions of hUCB-MSCs at 3-day intervals were effective in preventing GVHD. To determine whether the effects of infusions at 3-day intervals were related to cell numbers, we infused a 3-fold higher number of hUCB-MSCs (1.5
106) with a single dose at day 0. However, there was no significant increase in survival (data not shown). In a treatment study for GVHD, either a single injection (group 5) or repeated injections at 3-day intervals (group 6) after the onset of GVHD increased the survival rate to 70% (versus group 1, P < 0.001), and infusions at 7-day intervals (group 7) increased the survival rate to 60% (versus group 1, P < 0.001) (Figure 4C). Moreover, the mice consistently maintained their weight compared with

Optimized MSC therapy of GVHD

303

Figure 3. Inflammatory cytokine levels. After transplantation of hPBMCs, IFN-g levels gradually increased for 12 days and markedly increased on day 18 (A). TNF-a levels increased from day 24 (B), but there was no detectable change in IL-2 secretion (C). IFN-g was the main molecule associated with the onset of GVHD symptoms induced by hPBMCs (D). Each time point represents analysis of three mice measured in triplicate.

group 1 irrespective of the number of doses (Figure 4D). Furthermore, after the onset of GVHD, either a single injection or repeated injections of hUCB-MSCs significantly improved the clinical score and reduced lymphocyte infiltration compared with group 1 (Figure 4E; Tables I and II). These data demonstrate that infusions of naive hUCB-MSCs after the onset of GVHD significantly improved the survival rate and attenuated tissue damage regardless of the time between doses and the number of doses.

levels of both PGE2 and TGF-b1 increased significantly (Figure 5D,E, P < 0.001) when compared with group 1. In contrast, the levels of IFN-g decreased significantly in all experimental groups (Figure 5C,F, versus group 1, P < 0.001). These findings show that elevated levels of PGE2 and TGF-b1 induced by the infusion of hUCB-MSCs correlated with alleviation of GVHD symptoms and improved survival.

Discussion Engrafted naive hUCB-MSCs increase the levels of PGE2 and TGF-b1 and decrease those of IFN-g We measured the levels of PGE2 and TGF-b1 in the surviving mice (of all groups) at day 60 to confirm the immunosuppressive effects by hUCB-MSCs. When only hUCB-MSCs were repeatedly infused before the onset of GVHD, PGE2 levels increased significantly (Figure 5A, P < 0.001) in comparison to groups 1 and 2. TGF-b1 levels were also significantly induced by either single or repeated injections of hUCB-MSCs (Figure 5B, group 2 and group 3; P < 0.001, group 4; P < 0.05) when compared with group 1. However, when hUCB-MSCs were infused once or repeatedly after the onset of GVHD, the

In the present study, we report the development of animal models for acute and chronic GVHD, and, to our knowledge, this study is the first to report the optimum dosing frequency of hUCB-MSCs and successful treatment of GVHD by use of either a single injection or multiple injections of hUCBMSCs after GVHD onset. Allogeneic GVHD models generally require engrafting two types of cells, for example, 5
106 to 1
107 from the BM and spleen (18,19,24). In xeno-GVHD models, greater than 1e2
107 hPBMCs, together with irradiation, are used to induce symptoms of GVHD (17,25). Such high cell numbers limit the number of experimental conditions that can be tested.

304

Y. K. Jang et al.

Figure 4. Increase in survival rates in GVHD mice after infusion of naive hUCB-MSCs. Before the onset of GVHD, the survival rate of group 2 was 12.5% (versus group 1, P ¼ 0.4805) and that of group 4 was 33.3% (versus group 1, P ¼ 0.1252). However, the survival rate of group 3 markedly increased up to 70% (versus group 1, P < 0.001) (A and B) (experiments were repeated six times, n ¼ 10 for each group). The infusion of hUCB-MSCs after the onset of GVHD increased the survival rate by 70% for groups 5 and 6 and 60% for group 7 (C) (versus group 1, P < 0.001), and the mice maintained their weight (D) (experiments were repeated 3 times, n ¼ 10 for each group). Before the onset of GVHD, hUCB-MSCs were injected at 3-day intervals (group 3), resulting in decreased lymphocyte infiltration and significant prevention of tissue damage compared with group 1 and group 2. After the onset of GVHD, the infusion of hUCB-MSCs reduced lymphocyte infiltration and tissue damage compared with group 1, independent of the number of doses (E) (magnification
200, scale bar ¼ 100 mm).

Ito et al. (23) recently established a stable xenoGVHD model with the use of NOG (NOD.Cgprkdcscidil2rgtm1Sug/Jic) mice. With the use of as little as 2.5e5
106 hPBMCs, their xeno-GVHD model showed consistent GVHD induction compared with Rag2null Il2rgnull and NOD/SCID models. In the present study, we succeeded in inducing acute and chronic GVHD symptoms with smaller numbers of hPBMCs than those reported by Ito et al. (23). With the use of NSG mice and 1
106 hPBMCs, our results are similar to those obtained by Ito et al. with the use of 2.5
106 hPBMCs (23). Furthermore, this xeno-GVHD model was useful for investigating the efficacy of prevention or treatment by hUCBMSCs because the onset time of GVHD was reproducible in every experiment.

Although the immunosuppressive activity of MSCs has been demonstrated in in vivo skin graft studies (26) and clinical research (27), whether MSCs are effective for treating GVHD after its onset is still debated. For example, Tisato et al. (17) showed that the systemic infusion of hUCB-MSCs before but not after the onset of GVHD markedly reduced human T-cell proliferation and the ensuing damage to tissues and significantly improved the survival rate of xeno-GVHD-model mice. A study that used an allogeneic model of GVHD with mouse BM-MSCs suggested that MSCs could effectively increase survival rates after GVHD onset (18). In contrast, other studies that used allogeneic models have reported that when MSCs were transfused together with immune cells at the start of the

Optimized MSC therapy of GVHD

305

Table II. Histopathological GVHD score after infusion of naive hUCB-MSCs. Prevention effect

Organ Lung Liver Kidney Small intestine Total

Group 1 (PBMC only) 3 2 2 3 10

Group 2 (day 0) 3 1 1 2 7

Group 3

Treatment effect Group 4

Group 5

Group 6

Group 7

(days 0, 3 and 6) (days 0, 7 and 14) (day 18) (days 18, 21 and 24) (days 18, 25 and 32) 0.5 0.5 0 0 1

2 2 1 0 5

1 0.5 0.5 0 2

0.5 0.5 0.5 1 2.5

1 0.5 1 1 3.5

The histopathological scores of groups infused with hUCB-MSCs were determined according to the extent of lymphocyte infiltration in group 1 (lymphocyte infiltration: 0, normal; 0.5, focal and rare; 1, focal and mild; 2, diffuse and mild; 3, diffuse and moderate; 4, diffuse and severe).

experiments, the symptoms of GVHD were attenuated (19,20,24,28). We wished to determine why the therapeutic effects of hMSCs were not observed from the xenoGVHD animal models, even though MSCs have immunosuppressive abilities (13e16). Therefore, we assessed the levels of inflammatory cytokines after irradiation and injections of hPBMCs to confirm the onset of GVHD. We also determined the optimal time intervals for administering hUCB-MSCs that might be required to enhance prevention and treatment. MSCs are trapped, initially, in the lungs after

systemic intravenous infusion and are then redistributed to other organs, although they can influence the regeneration of injured sites or the engraftment of stem cells (29e31). However, when the distribution of MSCs was observed by means of bioluminescence imaging and quantitative polymerase chain reaction between 1e7 days after injection, the fluorescence signals and messenger RNA transcripts rapidly decreased during the first 24 h and gradually disappeared until 7 days (30). In our previous test of the distribution of hUCB-MSCs with the use of NOD/ SCID mice, human DNA was detected in BM,

Figure 5. Naive hUCB-MSC infusion increases the levels of anti-inflammatory molecules. Before GVHD onset, repeated injections of hUCB-MSCs at 3-day intervals (group 3) and at 7-day intervals (group 4) induced a statistically significant increase in PGE2 levels in the xeno-GVHD model compared with group 1 (*P < 0.001); however, a single injection of hUCB-MSCs (group 2) failed to increase PGE2 levels (P ¼ 0.586) (A). In contrast, the TGF-b1 levels increased significantly in groups 2e4 in comparison to group 1 (B) *P < 0.001; **P < 0.05. After the onset of GVHD, the injection of hUCB-MSCs significantly increased the levels of both PGE2 and TGF-b1 compared with those for group 1 (D, E) *P < 0.001. The levels of IFN-g were significantly decreased with the use of hUCB-MSCs regardless of the injection time (C, F) versus group 1, *P < 0.001.

306

Y. K. Jang et al.

lungs, pancreas and spleen in one or two of five mice 3 days after the infusion of hUCB-MSCs. In contrast, human DNA was undetectable 7 or 14 days after the infusion of hUCB-MSCs (Supplementary Figure 3). Therefore, we infused hUCBMSCs at 3- or 7-day intervals (before and after in vivo clearance) and found that the infusion of hUCB-MSCs at 3-day intervals was more effective for preventing GVHD. This result suggests that before hUCB-MSCs encounter inflammatory cytokines, cells should be repeatedly infused within a short time and at 3-day intervals because of rapid clearance. This protocol resulted in preventing the onset of GVDH. In contrast, infusion of hUCBMSCs after the onset of GVHD significantly increased the survival rate for single and repeated injections because the MSCs were exposed to inflammatory cytokines on infusion. However, if applied to a clinical study, the number of doses may be considered according to the severity of the disease and the patient’s condition. Polchert et al. (18) showed that in a BALB/c mice model, mouse BM-MSCs activated by IFN-g could prevent GVHD without the infusion of additional MSCs and improved survival rate up to 100% with a single injection of only 1
105 MSCs. They explained that immunosuppressive molecules such as TGF-b, indole 2,3-dioxygenase or hemoxygenase1, produced from IFN-gestimulated MSCs, may be important contributors of the MSC effect. As another mechanism, Tobin et al. (32) showed that the prolonged survival of xeno-GVHD model after MSC or IFN-g stimulated BM-MSC infusion was caused by the inhibition of T-cell proliferation but not CD4þCD25þFoxP3þ T-cell populations. In our xeno-GVHD model, we detected immunosuppressive molecules such as PGE2 and TGF-b1 after naive hUCB-MSC injection but could not detect CD4þCD25þFoxP3þ T-cell populations (data not shown) in the mouse blood. This was similar to the findings presented from previous studies. However, when we infused IFN-gestimulated MSCs, we observed only prevention effect by multiple infusions (3-day intervals) of activated hUCB-MSCs, and there was no significant effect on treating established GVHD (Supplementary Figure 4). In our future studies, it is therefore necessary to determine why activated hUCB-MSCs prevented GVHD but were not effective in treating GVHD. On the other hand, Bruck et al. (33) have recently reported that neither infusion of naive BM-MSCs nor IFN-geactivated BM-MSCs consistently prevents xenogeneic GVHD, even though 3
106 MSCs were repeatedly infused. Their results were contrary to those of Tisato et al. (17), in which they demonstrated GVHD prevention effects of

UCB-MSCs. Bruck’s group suggested that differences in MSC source could have accounted for such different results. In addition, the discrepancy observed between the results of Bruck et al. (33) and our data is the time point of MSC injection as well as MSC source. On the basis of our results, if MSCs were repeatedly injected before in vivo clearance, different results might have been obtained. In clinical practice, patients are administered 2e8
106/kg of MSCs for GVHD treatment (10,11). If the patient’s weight increases, the number of required MSCs also must increase. However, acquiring sufficient quantities of stem cells from commercial sources is very costly and places a significant financial burden on patients. Therefore, if we could achieve effective treatments with the use of 20e30% of the total number of cells currently being used in clinical studies, it might reduce patient’s burden and improve quality of life. Furthermore, to increase the efficacy of MSCs for preventing or treating GVHD, it is important that MSCs migrate to injured sites in sufficient numbers. The stromal derived factor-1 and its ligand CXCR-4 play an important role in the homing of stem cells, but the levels of CXCR-4 in the cultured hMSCs were low (34). CXCR-4 expression is induced by hypoxia, which activates the hypoxia inducible factor (HIF) pathway in diverse cell types, such as maturing B cells (35), human umbilical vein endothelial cells (36,37), microglia (38), mesenchymal stromal cells (39) and cardiac myocytes (40) or by overexpression of HIF (28). However, hUCB-MSCs that overexpress CXCR-4 only show a preventive effect, although they significantly reduce tissue damage caused by GVHD and prolong survival (28). In our preliminary experiment, low doses of hUCB-MSCs (2
104 and 1
105) showed a survival rate similar to that achieved with the use of a high dose of hUCB-MSCs (5
105) under normal conditions (Supplementary Figure 5). We confirmed that HIF-1a and CXCR-4 expression increased when hUCB-MSCs were cultured under hypoxic conditions (3% O2, 5% CO2 and 92% N2) (data not shown). Therefore, if hUCB-MSCs are cultured under hypoxic conditions, we expect that the increased expression of CXCR-4 may enhance the effect of a low dose of MSCs with respect to preventing or treating GVHD without pretreatment with IFN-g in our ongoing study. Taken together, the present study suggests that hUCB-MSCs should be infused repeatedly within a short time to prevent GVHD, and the optimal time intervals should be established before the clearance of MSCs in vivo. Furthermore, our data demonstrate that naive hUCB-MSCs can be used to treat GVHD symptoms after the onset of GVHD in a

Optimized MSC therapy of GVHD xeno-GVHD NSG mouse model in which single or repeated injections were effective. Acknowledgments This study was supported by a grant from the Korea Healthcare Technology R&D Project, Ministry of Health and Welfare, Republic of Korea (A121968). We thank Dr Sook-Tae Ha for histopathological diagnoses and Seok-Yoon Hong for preparing the slides in the Tissue and Cell (T & C) Pathology Centre as well as all staff who maintained the specific pathogenefree facility in the College of Medicine, Hanyang University. Disclosure of interests: The authors have no commercial, proprietary, or financial interest in the products or companies described in this article.

References 1. Locatelli F, Rocha V, Chastang C, Arcese W, Michel G, Abecasis M, et al. Factors associated with outcome after cord blood transplantation in children with acute leukemia: EurocordCord Blood Transplant Group. Blood. 1999;93:3662e71. 2. Bacigalupo A. Management of acute graft-versus-host disease. Br J Hematol. 2007;137:87e98. 3. Hill GR, Crawford JM, Cooke KR, Brinson YS, Pan L, Ferrara JL. Total body irradiation and acute graft-versus-host disease: the role of gastrointestinal damage and inflammatory cytokines. Blood. 1997;90:3204e13. 4. Goker H, Haznedaroglu IC, Chao NJ. Acute graft-versus-host disease: pathobiology and management. Exp Hematol. 2001; 29:259e77. 5. Iwasaki T. Recent advances in the treatment of graft-versushost disease. Clin Med Res. 2004;2:243e52. 6. Ho VT, Cutler C. Current and novel therapies in acute GVHD. Best Pract Res Clin Haematol. 2008;21:223e37. 7. MacMillan ML, Weisdorf DJ, Wagner JE, DeFor TE, Burns LJ, Ramsay NK, et al. Response of 443 patients to steroids as primary therapy for acute graft-versus-host disease: comparison of grading systems. Biol Blood Marrow Transplant. 2002;8:387e94. 8. Khoury H, Kashyap A, Adkins DR, Brown RA, Miller G, Vij R, et al. Treatment of steroid-resistant acute graft-versushost disease with anti-thymocyte globulin. Bone Marrow Transplant. 2001;27:1059e64. 9. Le Blanc K, Ringden O. Immunobiology of human mesenchymal stem cells and future use in hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2005;11: 321e34. 10. Kebriaei P, Robinson S. Treatment of graft-versus-host-disease with mesenchymal stromal cells. Cytotherapy. 2011;13:262e8. 11. Tolar J, Villeneuve P, Keating A. Mesenchymal stromal cells for graft-versus-host disease. Hum Gene Ther. 2011;22: 257e62. 12. Kuzmina LA, Petinati NA, Parovichnikova EN, Lubimova LS, Gribanova EO, Gaponova TV, et al. Multipotent mesenchymal stromal cells for the prophylaxis of acute graft-versushost disease: a phase II study. Stem Cells Int. 2012;2012: 968213. http://dx.doi.org/10.1155/2012/968213.

307

13. Krampera M, Glennie S, Dyson J, Scott D, Laylor R, Simpson E, et al. Bone marrow mesenchymal stem cells inhibit the response of naive and memory antigen-specific T cells to their cognate peptide. Blood. 2003;101:3722e9. 14. Maitra B, Szekely E, Gjini K, Laughlin MJ, Dennis J, Haynesworth SE, et al. Human mesenchymal stem cells support unrelated donor hematopoietic stem cells and suppress T-cell activation. Bone Marrow Transplant. 2004;33: 597e604. 15. Aggarwal S, Pittenger MF. Human mesenchymal stem cells modulate allogeneic immune cell responses. Blood. 2005;105: 1815e22. 16. Di Nicola M, Carlo-Stella C, Magni M, Milanesi M, Longoni PD, Matteucci P, et al. Human bone marrow stromal cells suppress T-lymphocyte proliferation induced by cellular or nonspecific mitogenic stimuli. Blood. 2002;99: 3838e43. 17. Tisato V, Naresh K, Girdlestone J, Navarrete C, Dazzi F. Mesenchymal stem cells of cord blood origin are effective at preventing but not treating graft-versus-host disease. Leukemia. 2007;21:1992e9. 18. Polchert D, Sobinsky J, Douglas G, Kidd M, Moadsiri A, Reina E, et al. IFN-gamma activation of mesenchymal stem cells for treatment and prevention of graft versus host disease. Eur J Immunol. 2008;38:1745e55. 19. Joo SY, Cho KA, Jung YJ, Kim HS, Park SY, Choi YB, et al. Mesenchymal stromal cells inhibit graft-versus-host disease of mice in a dose-dependent manner. Cytotherapy. 2010;12: 361e70. 20. Highfill SL, Kelly RM, O’Shaughnessy MJ, Zhou Q, Xia L, Panoskaltsis-Mortari A, et al. Multipotent adult progenitor cells can suppress graft-versus-host disease via prostaglandin E2 synthesis and only if localized to sites of allopriming. Blood. 2009;114:693e701. 21. Yang SE, Ha CW, Jung M, Jin HJ, Lee M, Song H, et al. Mesenchymal stem/progenitor cells developed in cultures from UC blood. Cytotherapy. 2004;6:476e86. 22. Cooke KR, Kobzik L, Martin TR, Brewer J Jr, Delmonte J, Crawford JM, et al. An experimental model of idiopathic pneumonia syndrome after bone marrow transplantation, I: the roles of minor H antigens and endotoxin. Blood. 1996; 888:3230e9. 23. Ito R, Katano I, Kawai K, Hirata H, Ogura T, Kamisako T, et al. Highly sensitive model for xenogenic GVHD using severe immunodeficient NOG mice. Transplantation. 2009;87: 1654e8. 24. Min CK, Kim BG, Park G, Cho B, Oh IH. IL-10-transduced bone marrow mesenchymal stem cells can attenuate the severity of acute graft-versus-host disease after experimental allogeneic stem cell transplantation. Bone Marrow Transplant. 2007;39:637e45. 25. Gregoire-Gauthier J, Selleri S, Fontaine F, Dieng MM, Patey N, Despars G, et al. Therapeutic efficacy of cord bloodderived mesenchymal stromal cells for the prevention of acute graft-versus-host disease in a xenogenic mouse model. Stem Cells Dev. 2012;21:1616e26. 26. Bartholomew A, Sturgeon C, Siatskas M, Ferrer K, McIntosh K, Patil S, et al. Mesenchymal stem cells suppress lymphocyte proliferation in vitro and prolong skin graft survival in vivo. Exp Hematol. 2002;30:42e8. 27. Le Blanc K, Rasmusson I, Sundberg B, Götherström C, Hassan M, Uzunel M, et al. Treatment of severe acute graftversus-host disease with third party haploidentical mesenchymal stem cells. Lancet. 2004;363:1439e41. 28. Chen W, Li M, Li Z, Yan Z, Cheng H, Pan B, et al. CXCR4transduced mesenchymal stem cells protect mice against graftversus-host disease. Immunol Lett. 2012;143:161e9.

308

Y. K. Jang et al.

29. Noort WA, Kruisselbrink AB, in’t Anker PS, Kruger M, van Bezooijen RL, de Paus RA, et al. Mesenchymal stem cells promote engraftment of human umbilical cord blood-derived CD34(þ) cells in NOD/SCID mice. Exp Hematol. 2002;30:870e8. 30. Kraitchman DL, Tatsumi M, Gilson WD, Ishimori T, Kedziorek D, Walczak P, et al. Dynamic imaging of allogeneic mesenchymal stem cells trafficking to myocardial infarction. Circulation. 2005;112:1451e61. 31. Yang X, Balakrishnan I, Torok-Storb B, Pillai MM. Marrow stromal cell infusion rescues hematopoiesis in lethally irradiated mice despite rapid clearance after infusion. Adv Hematol. 2012; 2012:142530. http://dx.doi.org/10.1155/2012/142530. 32. Tobin LM, Healy ME, English K, Mahon BP. Human mesenchymal stem cells suppress donor CD4(þ) T cell proliferation and reduce pathology in a humanized mouse model of acute graft-versus-host disease. Clin Exp Immunol. 2013;172:333e48. 33. Bruck F, Belle L, Lechanteur C, de Leval L, Hannon M, Dubois S, et al. Impact of bone marrow-derived mesenchymal stromal cells on experimental xenogeneic graft-versus-host disease. Cytotherapy. 2013;15:267e79. 34. Wynn RF, Hart CA, Corradi-Perini C, O’Neill L, Evans CA, Wraith JE, et al. A small proportion of mesenchymal stem cells strongly expresses functionally active CXCR4 receptor capable of promoting migration to bone marrow. Blood. 2004;104:2643e5. 35. Piovan E, Tosello V, Indraccolo S, Masiero M, Persano L, Esposito G, et al. Differential regulation of hypoxia-induced CXCR4 triggering during B-cell development and lymphomagenesis. Cancer Res. 2007;67:8605e14.

36. Schioppa T, Uranchimeg B, Saccani A, Biswas SK, Doni A, Rapisarda A, et al. Regulation of the chemokine receptor CXCR4 by hypoxia. J Exp Med. 2003;198:1391e402. 37. Ceradini DJ, Kulkarni AR, Callaghan MJ, Tepper OM, Bastidas N, Kleinman ME, et al. Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1. Nat Med. 2004;10:858e64. 38. Wang X, Li C, Chen Y, Hao Y, Zhou W, Chen C, et al. Hypoxia enhances CXCR4 expression favoring microglia migration via HIF-1alpha activation. Biochem Biophys Res Commun. 2008;371:283e8. 39. Hung SC, Pochampally RR, Hsu SC, Sanchez C, Chen SC, Spees J, et al. Short-term exposure of multipotent stromal cells to low oxygen increases their expression of CX3CR1 and CXCR4 and their engraftment in vivo. PLoS One. 2007;2: e416. 40. Hu X, Dai S, Wu WJ, Tan W, Zhu X, Mu J, et al. Stromal cell derived factor-1 alpha confers protection against myocardial ischemia/reperfusion injury: role of the cardiac stromal cell derived factor-1 alpha CXCR4 axis. Circulation. 2007;116: 654e63.

Supplementary data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.jcyt.2013. 10.012.