CXC chemokine receptor, CXCR2

Cellular Immunology 279 (2012) 1–11

Contents lists available at SciVerse ScienceDirect

Cellular Immunology journal homepage: www.elsevier.com/locate/ycimm

Mesenchymal stem cells regulate the proliferation of T cells via the growth-related oncogene/CXC chemokine receptor, CXCR2 Yong-Soo Lee a,1, Kyung-Jong Won b,1, Sung-Won Park a,1, Hyeon-Woo Lee c, Bokyung Kim b, Jin-Hoi Kim d,⇑, Dong-Ku Kim d,⇑ a

Graduate School of Life Science and Biotechnology, CHA University, Seoul 135-081, Republic of Korea Department of Medical Science, School of Medicine, Konkuk University, Chungju 380-701, Republic of Korea Department of Pharmacology, Institute of Oral Biology, School of Dentistry, Kyung Hee University, Seoul 131-701, Republic of Korea d Department of Animal Biotechnology, Konkuk University, Seoul 143-701, Republic of Korea b c

a r t i c l e

i n f o

Article history: Received 9 December 2010 Accepted 24 August 2012 Available online 5 September 2012 Keywords: Mesenchymal stem cells T cells GRO CXCR2 receptor Immunosuppression Tumor animal model

a b s t r a c t Mesenchymal stem cells (MSCs) have known to induce immunosuppressive properties by preventing T cell proliferation. However, it is remains unclear how MSCs inhibit T cell proliferation. To identify the factor that inhibits T cell proliferation, we conducted a cytokine array analysis of culture medium from a coculture of MSCs and T cells and found that the chemokines, CXCL1, 2 and 3, were induced in T cells. MSCs also induced the expression of the CXCR2 receptor on T cell surface. Particularly, CXCL3 inhibited proliferation and increased apoptosis in T cells, which were reversed by CXCR2 inhibitor treatment. Moreover, CXCL3 decreased JAK2, STAT3, and AKT phosphorylation and these responses were also abolished by CXCR2 inhibitor treatment. MSCs suppressed the proliferation of T cells into tumor tissue. Collectively, these data demonstrate that MSCs directly regulate T cell proliferation by induction of CXCL3 chemokine and its receptor, CXCR2 on the surface in T cells. Ó 2012 Elsevier Inc. All rights reserved.

1. Introduction Mesenchymal stem cells (MSCs) have multipotential differentiation ability into chondrocytes, tenocytes, skeletal myocytes, and neuronal cells, and are also able to improve the engraftment, survival, and differentiation of human hematopoietic stem cells [1– 4]. MSCs is known to have immunomodulatory activity; they suppress the T cell-mediated response by inducing apoptosis of activated T cells and contribute to the successful control of severe graft-versus-host disease [5,6]. The main suppressive reactions by MSCs were involved in the enhancement of the regulatory T cells, such as CD4+CD25hi, CD4+, and CTLA4+, and eventually the modulation of the cytokine or allogeneic T cell response [7,8]. MSCs also attenuated CD8+ T cell-mediated lysis in mixed lymphocyte reactions [9]. Moreover, the MSCs-induced inhibition of T cellmediated immune responses were also caused by several factors, such as interferon(IFN)-c, interleukin(IL)-10, tumor necrosis facAbbreviations: MSC, mesenchymal stem cells; GRO, growth-related oncogene; NOD/SCID, non-obese diabetic severe combine immunodeficient disease; CB-MNCs, cord blood-derived mononuclear cells. ⇑ Corresponding authors. Fax: +82 2 20307889 (D.-K. Kim), fax: +82 2 4551044 (J.-H. Kim). E-mail addresses: [email protected] (J.-H. Kim), [email protected] (D.-K. Kim). 1 These three authors have contributed equally to this study. 0008-8749/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cellimm.2012.08.002

tor(TNF)-a, IL-2, inoleamin 2,3-dioxygenase(IDO), and prostaglandin E [10–12]. It is demonstrated that MSCs can inhibit almost all immune responses in immune cells including B lymphocytes, natural killer cells, dendritic cells, and monocytes [7,13,14]. Therefore, it is assumed that MSCs may act as an important regulator in immune responses induced by T cells. Chemokines are cytokine-like proteins that selectively regulate the recruitment and trafficking of leukocyte subsets into inflammatory site by chemoattraction [15]. CXC chemokine family has been found to be associated with tumorigenesis, angiogenesis, and metastasis [16–18]. Moreover, growth-related oncogene (GRO) is known as a member of the CXC chemokine subfamily and plays a major role in inflammation and wound healing [18]. GRO chemokine is composed of CXCL1, 2, and 3 (called as GROa, b, and c, respectively) and binds to its common receptor CXCR2. The ligands of this receptor are also reported to be several chemokines such as IL-8, CXCL5(ENA-78), CXCL6(GCP2), and CXCL7(NAP2) [19]. Signaling via the CXCR2 chemokine receptor is mediated by various signaling pathways, including extracellular signal-regulated kinase [18–21]. Moreover, GRO chemokines inhibited the ability of downstream signals of the CXCR2 chemokine receptor to suppress monocyte arrest and also played a pivotal role in metastasis of several tumor cell lines and in engraftment of hematopoietic stem cells [22–24]. Previous reports demonstrated that MSCs downregulated the expression of chemokine receptor

2

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

CXCR4, CXCR5, and CCR7B, and reduced chemotaxis into CXCL12, a CXCR4 ligand, CXCL13, and the CXCR5 ligand [25–29], suggesting that MSCs may affect chemotactic properties via chemokine receptors in immune cells. Although MSCs is known to exert inhibitory effect on T cell proliferation induced by various stimuli [30,31], it remains unclear how MSCs inhibit T cell proliferation. In the present study, to define the immunosuppressive effect of MSCs, we explored the factors that modulate the ability of MSCs involved in T cell proliferation, using both an in vivo proliferative animal model and an in vitro cytokine microarray technique. We here showed that MSCs induced the expression of GRO, CXCL1, CXCL2, and CXCL3 and the expression of the CXCR2 receptor on T cells when co-cultured with T cells and that MSCs regulated immunosuppressive activity through CXCL3 signaling by reducing the activation of AKT, Janus kinase (JAK) 2, and signal transducer and activator of transcription (STAT) 3 signaling via the CXCR2 receptor in T cells. 2. Materials and methods 2.1. Primary cells and cell lines Cord blood (CB) samples from human were obtained from umbilical and placental tissues according to the institutional guidelines of CHA General Hospital (Seoul, Korea). Mononuclear cells (MNCs) from CB were isolated by density gradient centrifugation in Ficoll-PaqueTM Plus (Amersham Biosciences, Uppsala, Sweden). Purified cells were washed and suspended in phosphate buffered saline (PBS; Gibco, Gaithersburg, MD, USA) containing 2% fetal bovine serum (FBS) (Gibco). After washing twice, CB-derived MNCs (CB-MNCs) were placed on ice until transplantation into 1–3 day old neonate non-obese diabetic-severe combined immunodeficient (NOD/SCID) mice (KKIBB, Seoul, Korea). human cervical tumor (HeLa) cell lines were purchased from the Korean Cell Line Bank (KCLB, Seoul, Korea) and cultured in Dulbecco’s modified eagle’s medium (Gibco) medium supplemented with Eagle’s minimal essential medium (Gibco) and 10% heat-inactivated FBS. Human MSCs were purchased from Chambrex Bioscience (Walkersville, MD, USA). MSCs were cultured in minimum essential a-medium (a-MEM; Gibco), 10% FBS, and 2 mM L-glutamine (Gibco). Media were changed every three days, and cells were sub-cultured at the 70–80% confluent stage. MSCs were used at passage 4–8 for all experiments. 2.2. Animal model mice NOD/SCID mice were purchased from the Animal Laboratory of KKIBB and maintained in an animal facility at the CHA Stem Cell Institute at CHA University. Six- to ten-week-old mice were used for transplantation of CB-MNCs and the tumor formation experiment. To reconstitute NOD–SCID mice-T cells, we injected 1  107 CB-MNCs into 1–3 days old NOD/SCID mice. Treatment with 1  106 MSCs was initiated simultaneously with injection of CB-MNCs into NOD/SCID mice. To investigate the effects of the MSCs on human cervical tumorgenesis, 2  106 HeLa cells were subcutaneously injected into the right back-pad without MSCs and into the left back-pad with MSCs (4  105) of NOD/SCID mice-T cells that had been pre-injected with CB-MNCs for 4 weeks according to the above-mentioned method. The volume and weight of tumors were determined when the animals were sacrificed five weeks post injection with tumor cells with or without MSCs to determine tumor growth and MSC-mediated inhibition of T cells in vivo. The tumor volume was measured before after excision (volume = length  with  height). All animal experi-

ments in this study were approved by the CHA University Ethical Committee for Animal Experiment Regulation. 2.3. Flow cytometry analysis To determine the chimerism of human hematopoietic cells in peripheral blood (PB), engrafted human hematopoietic cells were investigated at four, six, and eight weeks after injection of CBMNCs. Mice were sacrificed and the bone marrow, spleen, and lymph nodes were collected to confirm the presence of human hematopoietic cells in organs. The tissues were teased apart and passed through a nylon filter to remove debris. Samples were prepared as single cell suspensions in staining medium with PBS and 2% FBS. Cells were stained with the following labeled antibodies: Fluorescein isothiocyanate (FITC)-conjugated anti-human CD45 (HI30), CD4 (RPA-T4), CD45RO (UCHL1), CD8 (HIT8a), CD45RA (HI100), and CD182 (CXCR2); PE-conjugated anti-human CD34 (581), CD33 (WIM53), CD19 (HIB19), CD3 (UCHT1), CD4 (RPAT4), CD8 (HIT8a), and CD182 (CXCR2); and APC-conjugated antihuman CD56 (B159) and CD3 (UCHT1). Activated and isotypic control antibodies were purchased from BD Pharmingen, (San Diego, CA, USA). Stained cells were analyzed with a fluorescence-activated cell sorter (FACS) VantageSE flow cytometer (BD Biosciences, CA, USA). Data were live-gated by forward and side scatter and by lack of propidium iodide (PI) uptake. The frequencies in quadrant corners are given as percentages of gated cells. Collected data were analyzed with CELLQUEST software (BD Biosciences). 2.4. Human T cell isolation To isolate purified T cells, human CD3+ T cells were selectively isolated from splenocytes of NOD/SCID mice injected with CBMNCs for four to six weeks prior to sacrifice or from human PB. Splenocytes were resuspended in PBS supplement with 2% FBS and stained with anti-human FITC-conjugated anti-CD3 (UCHT1) antibody (BD Bioscience), and the EasySep FITC Selection Kit (StemCell Technologies, Seattle, WA, USA) was used to isolate CD3-positive T cells. Enrichment of purified cells was confirmed by flow cytometric analysis, which showed a >90% positive cell population, and these purified cells were used for further experiments. 2.5. Proliferation assays To determine the anti-proliferative effects of MSCs, 2  104 MSCs/well were seeded in 96-well dishes containing a-MEM with 10% FBS. After a 24 h culture period, the medium was removed and the MSCs were washed with PBS and then further cultured in aMEM with 10% FBS and human CD3+ T cells (1  105/well) purified from CB-MNCs, human PB leukocytes, and splenocytes from NOD/ SCID mice pre-established by intraperitoneal injection with CBMNCs. To induce T cell proliferation, cells were stimulated with 10 ng/ml IL-2 (BD Bioscience) for two days and cell proliferative ability was measured using a CCK Solution Kit (Dojindo, kumamoto, Japan). To investigate the effect of chemokines on T cells, MSCs were cultured for 24 h and then washed twice with PBS. Purified CD3+ T cells obtained from human PB leukocytes or NOD–SCID mice pre-established with CB-MNCs for five to seven weeks were co-cultured with MSCs and anti-CD3/CD28 antibody. After two days of culture, T cells were isolated and transferred into 96-well plates in serum-free medium containing 100 ng/ml each of CXCL1, CXCL2, and CXCL3 chemokine (R&D, Minneapolis, MN, USA) for two days. The inhibition test of CXCR2 signaling was conducted using 200 nM of CXCR2-specific antagonist, SB225002 (Calbiochem, San Diego, CA, USA) [25,32], before a 30 min treatment with CXCL3 in serum-free culture medium. After two days of treatment,

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

3

cell proliferation was measured with a CCK Cell Counting Kit (Dojindo) using a microplate immunoreader (NJ-2001; Nihon InterMed, Tokyo, Japan) at a wavelength of 450 nm.

USA) to visualize cell nuclei. After incubation with the antibodies, sections were washed twice and analyzed by immunofluorescence microscopy using an Apotome (Carl Zeiss, Oberkochen, Germany).

2.6. Cytokine antibody array

2.9. Cell death assay

MSCs (4  105/well) were seeded in 60-mm dishes containing

a-MEM with 10% FBS. After 24 h of culture, MSCs were washed with PBS and further cultured in serum-free a-MEM with purified CD3+ T cells (2  106/well) from NOD-SCID mice injected with CBMNCs. After one day, culture media from MSCs alone, T cells alone, and an MSC and T cell co-culture were harvested for cytokine tests using ChemiArray™ Human Antibody Array I (Chemicon, Germany), according to the manufacturer’s instructions. Briefly, the cytokine array membranes were blocked with blocking buffer at room temperature and then incubated overnight at 4 °C. After incubation, the membranes were washed three times with 2 ml Wash Buffer I, followed by two washes with 2 ml Wash Buffer II at room temperature, with shaking. The membranes were then incubated with 2 ml biotin-conjugated antibodies (1:500 dilution) for 2 h at room temperature and washed as described above; this was followed by incubation with 1 ml streptavidin-conjugated peroxidase (1:1000 dilution) for 1 h at room temperature. After thorough washing, the membranes were exposed to a peroxidase substrate (R&D) for 5 min in the dark prior to imaging. The membranes were exposed to an X-ray film within 10 min of exposure to the substrate. 2.7. ELISA assay MSCs (2  105/ml) were seeded in 24-well plates containing aMEM with 10% FBS medium. After 25 h, cells were washed with PBS and further cultured in serum-free a-MEM with CD3+ T cells purified from human PB or splenocytes of NOD/SCID mice that had been injected with CB-MNCs five to eight weeks earlier. After culture for 48 h, culture media were prepared from T cells alone, MSCs alone, or a T cell and MSC co-culture to analyze the concentration of chemokines, CXCL1, CXCL 2, or CXCL 3. Chemokines, CXCL1 and CXCL2 were measured using the commercial human CXCL1 ELISA Kit (R&D) and the human CXCL2 ELISA Kit (IBL, Tokyo, Japan), respectively, according to the manufacturer’s instructions. To determined CXCL3 level, we coated 96-well plates with 100 ll/well of 5 lg/ml goat anti-human CXCL3 capture antibody in coating buffer (eBioscience, San Diego, CA, USA). After an overnight incubation at 4 °C, each well was washed thoroughly and blocking was conducted with 200 ll/well blocking buffer (R&D) at room temperature for 1 h. After three washes, cultured medium was added into CXCL3 specific capture antibody coated plate and incubated overnight at 4 °C. After thorough washing, 100 ll/well rabbit anti-human CXCL3 capture antibody (0.5 lg/ml) was added to the 96-well plates. After incubation with horseradish peroxidase-conjugated donkey anti-rabbit IgG antibody, 100 ll/well of substrate solution was added to each well. The absolute absorbance was calculated with a microplate immunoreader (NJ-2001; Nihon InterMed) at 450 nm. 2.8. Immunofluorescence of tissues To visualize human cells in tissues, frozen sections were fixed with 100% ethanol, the tissues were blocked by incubation with 1% bovine serum albumin buffer at 4 °C for 30 min, and directly stained for 1 h at room temperature with either FITC- or PE-conjugated anti-human-CD45 (HI30), CD3 (UCHT1), CD4 (RPA-T4), CD8 (HIT8a), CD33 (WIM53), CD19 (HIB19), CD56 (B159), CD90 (5E10), and CD105 (439–9B) antibodies. All sections were counterstained with 1 g/ml Hoechst 33342 (Sigma–Aldrich, St. Louis, MO,

To confirm the apoptotic effects of CXCL3, 1 x 106 purified CD3 + T cells were co-cultured with MSCs and anti-CD3/CD28 antibody for two days. T cells were isolated and transferred into 96well plates and treated with 10 lg/ml CXCL3 with or without 200 nM CXCR2 antagonist SB225002. After two days of culture, cells were stained with Annexin V (eBioscience) and PI, and then apoptosis was analyzed using flow cytometry (FACSvantageSE; BD Pharmingen). 2.10. Intracellular staining To characterize the CXCL3 signaling pathway through CXCR2 receptor, MSCs (2  105/well) were seeded in 24-well plates, to which 2  106 CD3+ T cells and 2 lg/ml anti-CD3/CD28 antibody were added for two days. T cells were harvested and stimulated or not with 100 ng/ml CXCL3 for 30 min. Cells were pre-treated with CXCR2 antagonist, SB205002, 30 min prior to CXCL3 treatment. After fixation with the BD Cytofix/Cytoperm Kit (BD Pharmingen), cells were subjected to intracellular staining with the following signaling protein antibodies: anti-phospho-JAK2 (Tyr 1007/Tyr 1008), anti-phospho-STAT3 (ser727), and anti-phoshoAKT (ser473) (Cell Signaling Technology, Beverly, MA, USA), according to the manufacturer’s instructions. The flow cytometry analysis was conducted using FACSvantageSE (BD Pharmingen). 2.11. Statistics The data are expressed as means ± SD. Statistical significance of the differences between experimental groups was assessed by Student’s t test. Statistical significance was determined at the level of P < 0.05. 3. Results 3.1. In vivo suppressive effect of MSCs on engraftment of cord blood hematopoietic cells Human MSCs promoted the engraftment and differentiation of hematopoietic stem cells [1,2]. To confirm these properties of human MSCs using CB hematopoietic cells in vivo, we injected peritoneally CB-MNCs into newborn NOD/SCID mice with or without MSCs. We determined the engraftment efficiency of CB hematopoietic cells in PB at four, six, and eight weeks post-injection by flow cytometry. NOD/SCID mice untreated with MSCs had a high percentage of CD45+ human hematopoietic cells from four- to eightweeks after injection. In contrast, these responses were not observed in NOD/SCID mice treated with MSCs during all periods tested. Moreover, we established that CD45+ human hematopoietic cells consisted of the majority of CD3+ T cells and a few cells of CD19+ B cells, but did not associated with CD33+ myeloid cells and CD56+ NK cells in NOD/SCID mice injected without MSCs (Fig. 1). We next analyzed the bone marrow, spleen, and lymph nodes of NOD/SCID mice injected with or without MSCs 10 weeks after injection with CB-MNCs. NOD/SCID mice untreated with MSCs showed higher levels of human CD45+ cells in the bone marrow, spleen, lymph nodes, and bone marrow than NOD/SCID mice treated with MSCs (Fig. 2A and B). The percentages of human CD45+ cells in bone marrow, spleen, and lymph nodes were 3.2 ± 2.1%, 21.7 ± 21.4%, and 38.1 ± 34.9%, respectively, in MSCs-

4

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

Fig. 1. Engraftment of human hematopoietic cells in PB of NOD/SCID mice injected with human CB mononuclear cells with or without MSCs. Cells were harvested on the indicated day from neonate NOD-SCID mice transplanted with 10  106 CB-MNC cells with or without 1  106 MSCs. (A) Human cells in NOD/SCID mice injected or not with MSCs were detected by staining with the human leukocyte marker, CD45, and the CD3, CD19, CD33, and CD56 antibodies. Analysis was performed using flow cytometry at 4 weeks mice. (B) The percentage of human CD45+ cells was determined using flow cytometry in PB of NOD/SCID mice treated with MSCs (j) or without MSCs (s) at four, six, and eight weeks after injection. Data are representative from five mice in each group.

Fig. 2. Flow cytometry analysis of bone marrow, spleen, and lymph nodes of NOD/SCID mice injected with human CB mononuclear cells with or without MSCs. CB-MNC cells (10  106/mouse) were transplanted into neonate NOD/SCID with or without 1  106 MSCs. Cells was harvested from the bone marrow, spleen, and lymph nodes of mice 10 weeks after injection. Cells were stained with anti-human CD45 and CD3 antibody and analyzed by flow cytometry. (A) Representative flow cytometry data from bone marrow (BM), spleen, and lymph nodes (LN) of NOD/SCID mice treated with or without MSCs. (B) The percentage of human CD45+ hematopoietic cells in BM, spleen, and lymph nodes was determined by flow cytometry. (C) Immunofluorescence analysis was performed on spleen from NOD/SCID mice treated with or without MSCs using antihuman CD4-FITC and anti-human CD8-PE antibody. Hoechst dye staining was used as counter stain. Data are representative from five mice in each group.

untreated mice and 0.1 ± 0.1%, 0.2 ± 0.1%, and 0.5 ± 0.2%, respectively, in MSCs-treated mice (Table 1). In these organs, the majority

of engrafted human CD45+ cells were CD3+ T cells, including CD4+ T cells and CD8+ T cells, similar to the phenotype of cells in the PB

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

5

Table 1 Engraftment of human hematopoietic lineage cells in organ of NOD/SCID mice injected with cord blood-MNC with or without MSCs. Cells

Human Antibodies

Bone marrow

Spleen

Lymph node

Control

CD45 CD3 CD19 CD33 CD56 CD45 CD3 CD19 CD33 CD56

3.2 ± 2.1 3.1 ± 1.9 0 0.1 ± 0.2 0.1 ± 0.1 0.1 ± 0.1 0.1 ± 0.1 0 0 0

7 ± 21.4 10.6 ± 10.6 1.9 ± 2.2 0.3 ± 0.6 0.5 ± 0.6 0.2 ± 0.1 0.1 ± 0.1 0 0 0

38.1 ± 34.9 24.9 ± 31.4 0.2 ± 0.2 0.5 ± 0.3 0.7 ± 0.9 0.5 ± 0.2 0.1 ± 0.1 0 0 0

MSC

Data are represented as mean percentage of positive cells ± SEM (n = 8).

from NOD/SCID mice. In the histological analysis, the spleens from mice untreated with MSCs showed a high engraftment of human CD4+ T cells and CD8+ T cells, but no engraftment was detected in those from MSCs-treated mice (Fig. 2C). Furthermore, we confirmed the profile of CD45RA and CD45RO expression in T cells and tested whether engrafted CD3+ T cells in NOD/SCID mice might be differentiated into the mature phenotype from immature T cells in CB, because the majority of T cells from CB were positive for CD45RA, a marker of naïve T cells. From the flow cytometric analysis, we determined that the majority of CD3+ T cells were of the mature phenotype (CD45RO+ T cells), and that there was a reduction in the number of naïve CD45RA+ T cells by co-culture with MSCs (Suppl. Fig. S1). 3.2. In vitro suppressive effect of MSCs on T cell proliferation We next tested the immunosuppressive effects of MSCs on T cell proliferation in vitro. CD3+ T cells were purified from CBMNCs, human PB, and splenocytes of NOD/SCID mice injected with CB-MNCs and co-cultured with or without MSCs in presence of IL-2 to stimulate T cell proliferation. As shown in Fig. 3, IL-2 increased the proliferations of T cells from human PB and NOD/SCID mice and these responses were significantly inhibited by co-cultured with MSCs. However, IL-2-stimulated proliferation of in T cells from CB was not inhibited by co-culture with MSCs. Based on the proliferation responses of T cells demonstrated above, it is assumed that MSCs can affect the level of cytokine IL2 induced in T cells. Moreover, it is known that one of inhibitory mechanism by MSCs in T cells results from the inhibition of IFNc secretion and the blocking of IFN-c by MSCs played important for the T cell inhibition [9]. Thus, to examine correlation between

Fig. 4. Regulation of IL-2 and IFN-c expression by MSC co-culture. CD3+ T cells were purified using EasySep selection kit from splenocytes of NOD–SCID mice injected with CB-MNCs. Cells were co-cultured with or without MSCs in the presence of anti-CD3/CD28 antibody stimulation. Expression of IL-2 or IFN-c was determined by intracellular staining of CD4+ T cells (A) or CD8+ T cells (B) using flow cytometry. The number in the quadrant box indicates the positive percentage of target cells. Results are representative of three experiments.

MSCs and the cytokine production of T cells, we determined the expression level of intracellular cytokines, especially IFN-c and IL-2, in T cells co-cultured with MSCs or in splenocytes of NOD/ SCID mice pre-treated with CB-MNCs using flow cytometry. When

Fig. 3. In vitro immunosuppressive effects of MSCs on T cells from CB, human PB, or NOD/SCID mice injected with CB-MNCs. T cells were purified by MACS from cord blood (A), PB (B), or NOD/SCID mice (C) treated with cord blood-MNCs. CD3+ T cells (1  105) were stimulated with IL-2 (10 ng/ml) cytokine in the presence or absence of 2  104 MSCs in 96 well plate for two days, and CCK analyzed. Cultures were performed in triplicate, and the mean of four independent experiments is presented. The histogram shows the mean of cpm + SEM. means statistically significant (P < 0.05), as compared with control cells.

6

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

Fig. 5. ELISA assay of chemokines, CXCL1, CXCL 2, or CXCL 3. CD3+ T cells prepared from human PB or splenocytes of NOD-SCID mice-T cells were co-cultured with MSCs. (A) CD3+ T cells derived from human PB were co-cultured with or without MSCs, and then the volume of indicated chemokines was calculated in an ELISA assay using ELISA kit. Results are representative of three experiments. (B) CD3+ T cells were purified from splenocytes of NOD/SCID mice injected with CB-MNCs and cultured with or without MSCs. The expression of the indicated chemokines, CXCL1, 2, or 3, was determined in the culture medium by an ELISA assay using ELISA kit. The data indicate fold-increase of expression level (in volume) of chemokine compared to that in media containing MSCs or T cells alone. Results are representative of three experiments.

cells were stimulated with anti-CD3 or CD28 antibodies, there was a decrease in the proportion of IFN-c+ cells, but did not show effect on IL-2+ cells, in CD4+ T cells (Fig. 4A) or CD8+ T cells (Fig. 4B) in the presence of MSCs compared to the absence of MSCs. 3.3. Cytokine array From in vivo and in vitro data, we hypothesized that the soluble factors produced by MSCs might be involved in the inhibition of engraftment and in the immunosuppressive effects against T cells. To identify the suppressive effect factor produced by MSCs, we performed a cytokine array analysis using culture medium from a coculture of MSCs and T cells from splenocytes of NOD/SCID mice engrafted with CB T cells. We observed that, compared to cells cultured in the media with only MSCs or T cells, co-culture of MSCs and T cells significantly induced the expression of the GRO family

of cytokines (Suppl. Fig. S2). In the ELISA assay to measure the expression level of each chemokine, CXCL1, 2, and 3, in co-culture medium of MSCs and T cells, the expression of CXCL1, 2, and 3 in co-culture of MSCs and T cells from human PB and NOD/SCID mice engrafted with CB-MNCs showed a dramatic up-regulation compared to the culture media containing only MSCs or CD3+ T cells. Indeed, CXCL1 expression was about 30-fold higher than PB T cells alone, and that of CXCL2 or CXCL3 was approximately 4- to 8-fold higher in the co-culture medium of PB-T cells and MSCs (Fig. 5A). Moreover, the expression of CXCL1, CXCL2, and CXCL3 was about 4- to 7-fold higher in the co-culture medium of NOD/SCID-T cells and MSCs than in the other culture media without co-culture (Fig. 5B). We next tested whether MSCs could induce the expression of the CXCL1, CXCL2, and CXCL3 common receptor, CXCR2, on the surface of T cells by co-culture with MSCs. CXCR2 was expressed

Fig. 6. Induction of CXCR2 expression by co-culture of MSCs with T cells. CD3+ T cells purified from splenocytes of NOD/SCID mice injected with CB-MNCs were co-cultured with or without MSCs in the presence of anti-CD3/CD28 stimulation for two days. The expression of the CXCR2 receptor was determined by flow cytometry and staining with CD4 (A) or CD8 (B) antibody. The number in the histogram represents the percentage of positive cells in CD4- or CD8-positive gated cells. (C) Immunohistochemical analysis was conducted to survey the expression of the CXCR2 receptor on the cell surface of T cells with or without MSCs.

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

7

Fig. 7. Anti-proliferative effects of CXCL3 in T cells. (A) Purified CD3+ T cells were obtained from splenocytes of NOD/SCID mice treated with CB-MNCS. T cells were cocultured with MSCs in presence of anti-CD3/CD28 antibody to induce the expression of CXCR2 receptor on T cells. After stimulation, T cells were harvested and treated with CXCR1, 2, or 3 for two days, and the effects on proliferation were determined. (B) After T cells were cultured with anti-CD3/CD28 antibody and MSCs, cells were harvested and treated with CXCL3 and SB225002, an inhibitor of CXCR2, and cell growth was quantified. (C) The effect on cell death was determined by measuring Annexin V and PI staining following treatment with CXCL3 and/or SB225002 inhibitor. The culture was conducted in triplicate. Significant differences compared with a group of CXCL3-treated cells are marked All of data are representative of three experiments (⁄P < 0.05).

on myeloid cells and, to a lesser extent, on T cells of PB, but was not significantly expressed in T cells from CB and NOD/SCID mice engrafted with CB-MNCs (Suppl. Fig. S3). When T cells from NOD– SCID mice engrafted with CB-MNCs were co-cultured with MSCs and anti-CD3/CD28 antibody, there was an up-regulation in the expression of CXCR2 in CD4+ T and CD8+ T cells. The proportion of CXCR2+ cells was 4.5-fold higher in CD4+ T cells (Fig. 6A) and 2.3-fold higher in CD8+ T cells (Fig. 6B) in the co-culture with MSCs than in cultures treated only with anti-CD3/CD28. Moreover, in immunofluorescence analysis, the expression of CXCR2 in T cells from the co-culture with MSCs showed a significantly greater proportion of CXCR2-positive cells than that in the control cells (Fig. 6C).

To identify how CXCL3 inhibits the proliferation of T cells, we first examined whether T cells stimulated with CXCL3 undergo cell death, using flow cytometry analysis of Annexin V and PI. After coculture with MSCs and anti-CD3/CD28 antibodies, T cells were isolated and treated with CXCL3 in the absence or presence of CXCR2 inhibitor, SB225002. Flow cytometry analysis revealed that CXCL3treated cells showed a remarkable increase in Annnexin V+ apoptotic cells compared to untreated control cells. However, the cell death was restored in cells treated with the specific CXCR2 inhibitor, SB225002, in response to CXCL3 signaling, to levels observed in control cells (Fig. 7C). 3.5. CXCL3 suppresses T cell proliferation by inhibiting AKT, JAK2, and STAT3 pathways

3.4. CXCL3 inhibits T cell proliferation To define the biological activity of CXCL1, 2, and 3 chemokines, T cells from NOD/SCID mice pre-engrafted with CB-MNCs were cocultured with MSC in presence of anti- CD3/CD28 antibodies to induce CXCR2 expression and proliferation of T cells. Isolated cells were treated with CXCL1, 2, and 3 and the proliferative effects were measured two days after stimulation. The proliferations of T cells treated with CXCL1 or CXCL2 did not differ from those of control cells, but CXCL3-treated T cells showed a significant reduction in cell proliferation compared with control cells (Fig. 7A). To obtain further evidence that these suppressive effects of CXCL3 were derived from direct signaling via the CXCR2 receptor, we treated cells with a specific CXCR2 inhibitor, SB225002, under the same culture conditions. Pre-treatment of T cells with SB225002 abolished CXCL3-induced inhibition of T cell proliferation evoked by anti-CD3/CD28 antibody stimulation (Fig. 7B).

To define how CXCL3 suppresses the proliferation of T cells, we investigated the activity of three signaling molecules, AKT, JAK2, and STAT3, involved in T cell survival and growth. T cells were cultured in the presence of MSCs, exposed to anti-CD3/CD28 antibody stimulation, and then treated with CXCL3. Phosphorylation of AKT, JAK2, and STAT3 were assessed by flow cytometry analysis with specific antibody staining. As shown in Fig. 8A, CXCL3 treatment reduced JAK2, STAT3, and AKT phosphorylation compared to that in untreated control cells. The phosphorylation percentages of JAK2, STAT3, and AKT were 37.3 ± 2.1%, 73.3 ± 1.5%, 32.0 ± 1.9%, respectively and 0.1 ± 0.1%, 0.2 ± 0.1%, and 0.5 ± 0.2%, respectively, in untreated control cells and 30.8 ± 1.8%, 62.7 ± 2.0%, 20.9 ± 1.2% in CXCL3-treated cells, respectively. Cells co-treated with CXCL3 and CXCR2 inhibitor, SB225002 did not show CXCL3-induced reduction of phosphorylation of AKT, STAT3, and JAK2 compared with control T cells (Fig. 8B).

8

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

Fig. 8. CXCL3 involves the survival or proliferative signaling pathway of T cells. CD3+ T cells prepared from splenocytes of NOD/SCID mice injected with CB-MNCs and cocultured with MSCs in presence of anti-CD3/CD28 antibodies. After two days, T cells were harvested and stimulated with CXCL3 alone (A) or with CXCL3 and SB 225002, a CXCR2 receptor-specific inhibitor (B). Flow cytometry revealed the percentage of phospholated-JAK2, -STAT3, and -AKT positive cells in comparison with those in the isotype control (unfilled histogram). In all cases, data are representative of three independent experiments.

3.6. T cell-mediated anti-tumor activity is suppressed by MSCs We previously showed that MSCs directly regulate the immunosuppressive activity against T cells in vivo and in vitro. Thus, the capacity of MSCs to regulate anti-tumor activity directly by suppressing the T cell response to tumors was tested in a tumor animal model that consisted of a mixture of T cells, tumor, and MSCs. To establish this animal model, we transplanted CB-MNCs into NOD/SCID mice, following the previously described procedure, for four to six weeks and confirmed the engraftment of human T cells in the PB of NOD/SCID mice. Next, HeLa cells were subcutaneously co-injected with MSCs at the left flank site and without MSCs at the right flank site into pre-established NOD/SCID mice. Five weeks after tumor injection, the mice were sacrificed and the tumor was measured and weighted to determine the anti-tumor effect of MSCs. Interestingly, the tumors in mice co-injected with MSCs were larger than those of mice not injected with MSCs (Fig. 9A–C). In the immunohistochemical analysis, tumor tissue had a high level in infiltration by CD8+ T cells in mice injected without MSCs, but that a few CD8+ T cells were present in the tumor tissue of mice co-injected with MSCs (Fig. 9D). Moreover, the presence of MSCs, which showed CD90+ and CD105+ double positive response, were observed in tumor tissue (Fig. 9E). 4. Discussion MSCs possess immunomodulatory activity on a variety of immune cell lineages, including T cells, by acting on several different factors or by arresting cell division [6,20,33]. T cells are a key component of the adaptive immune system and play a major role in the immune response in vivo. However,

the mechanism underlying the anti-proliferative role of MSCs against T cells is largely unknown. In the present study, we tested the immunosuppressive effect of MSCs to identify the factors that modulate the ability of MSCs involved in T cell proliferation, using both an in vivo proliferative animal model and an in vitro cytokine microarray technique. We found that MSCs showed the dramatic anti-proliferative effects on human hematopoietic cells in vivo by transplanting CB-MNCs into neonate NOD/SCID mice. Moreover, in vitro test on immunosuppressive effects of MSCs on T cell proliferation, T cells from human PB- and NOD/SCID mice showed the diminishment of IL-2-incrrased proliferation by co-culture with MSCs. However, T cells from CB were not inhibited by co-culture with MSCs. This unresponsiveness in CB T cells may be attributed to the property of CB T cells, which are based on our data that CB T cells consisted of the majority of CD4+ CD45RA+ naïve cells and a report that CB CD4+ CD45RA+ T cells have lower magnitude of activation and have lower expression level of CD25 (IL-2 receptor) when compare with adult counterparts [34]. Therefore, these results imply that MSCs can exert an inhibitory effect on proliferation of T cells in vivo and in vitro. It is reported that MSCs reduced the secretion of proinflammatory cytokines, IL-2 and IFN-c, a phenomenon that could be involved in T cell proliferation [35,36]. In addition, the immunosuppressive activity of MSCs on T cell responses was modulated even when the T cells were stimulated by mitogen, CD3/ CD28 antibody or alloantigen [38]. In the present study, we confirmed that T cells co-cultured with MSCs showed severe inhibition in the expression level of intracellular cytokine IFN-c, but did not affect IL-2 expression. These data indicate that MSCsinhibited T cell proliferation may be associated with secretion of pro-inflammatory cytokine IFN-c.

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

9

Fig. 9. Suppressive effect of MSCs on the inhibition of tumor growth by T cells. Before establishing a tumor model, all NOD/SCID mice were injected with cord blood mononuclear cells and engraftment of the cord blood T cells was confirmed by flow cytometry. Mice were injected subcutaneously into the posterior right flank with 2  106 human cervical tumor (HeLa) cells alone or into the posterior left flank with conjunction of HeLa and 2  106 MSCs. (A) Representative tumors in untreated (control) or MSCtreated mice eight weeks after tumor injection. Tumor size (B) and weight (C) were measured eight weeks post injection. Histogram data show the mean of cpm + SEM  means statistically significant (P < 0.05) as compared with control cells. The tumor volume was measured after excision (volume = length  width  height). (D) Immunofluorescence analysis revealed if CD4+ T or CD8+ T cells were present or absent in cervical tumor tissue. (E) The presence of MSCs was confirmed by staining with MSC-specific markers, anti-human CD90 and anti-human CD105 antibody with tumor tissues. Arrow indicates CD90 and CD105 double positive cells. All of data are representative of three repetitive experiments with at three mice per group.

To determine a soluble factor that mediates the immunosuppressive properties in T cells, the present study characterized it using a cytokine array system with a co-culture of MSCs and T cells. We found that the expressions of all three GRO chemokines, CXCL1, CXCL2, and CXCL3, were significantly enhanced in both human PB-derived T cells and NOD/SCID mice-derived human ones. This result implies that MSCs can induce GRO chemokines in T cells. We also confirmed that an MSC co-culture enhanced the expression of CXCR2, which is known as a specific receptor of GRO chemokines CXCL1, CXCL2, or CXCL3 [19], on CD4+ T and CD8+ T cells in flow cytometry and immunohistochemical analysis, indicating that MSCs also induce a specific receptor CXCR2 of GRO chemokines on T cell surface. Previous investigations demonstrated that CXCR2 chemokine receptor was associated with various signaling pathways [18–21]. Moreover, GRO chemokines regulated downstream signals of the CXCR2 chemokine receptor in various cells [22–24]. Therefore, it is possible that MSCs may regulate cellular events involved in signaling via GRO chemokines and their receptor in T cells. Moreover, a functional activity assay of T cells in the present study showed that the treatment of CXCL3 significantly suppressed T cell proliferation compared with that of CXCL1 or CXCL2 in T cell. This inhibitory effect by CXCL3 treatment was abolished by the CXCR2 receptor inhibitor, SB225002, demonstrating that signaling through CXCR2 was responsible for CXCL3-mediated inhibition of T cell proliferation. In cell death and cell cycle analysis, we also found that CXCL3 induced apoptotic cell death and this was sup-

pressed by CXCR2 inhibition, when co-cultured with MSCs or with both MSCs and anti-CD3/CD28. However, CXCL3 did not affect cell cycle arrest (data not shown). These results imply that the CXCL3 produced by MSCs may mainly exert the suppressive activity of proliferation in T cells by regulating apoptotic cell death, but not by cell division. A research group has reported that inoleamin 2,3dioxygenase(IDO) induced MSCs-mediated apoptosis in proliferating T cells, but not in resting T cells [31], indicating that signaling from CXCL3 might be involved in the inhibition of T cell proliferation through the activation of the IDO enzyme. Collectively, these observations indicated that the MSCs-induced anti-proliferative activity in T cells may regulated by signaling pathway mediated by binding of CXCL3 to its receptor, CXCR2, on the T cell surface. Signal transduction cascades associated with the activation of signaling molecules, including Raf-1, AKT, STATs, JAKs, are known to play important roles in cellular events including proliferation in T cells [37–39]. AKT, JAK2, and STAT3 participated in T cell survival and proliferation [37,38,40]. In the present study, we observed that CXCL3 inhibited proliferation and survival signals, while reducing the phosphorylation activity of JAK2, STAT3, and AKT kinases in T cells. Moreover, these CXC3-induced events in T cells were recovered by the pretreatment with CXCR2 inhibitor, SB225002. These results suggest that the anti-proliferative activity of CXCL3 could be attributed to signals involved in CXCR2 receptor in T cells. To further define the anti-proliferative activity of MSCs against T cells in vivo, we used human tumor animal model, in which CB-

10

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11

MNCs were pre-engrafted into neonate NOD/SCID mice. This model had a high level of engraftment with human T cells, which showed anti-tumor effects by tumor-infiltrated T cells from CB via cytotoxic activity by T cells in cervical tumor tissue. In mice injected with tumor cells and MSCs, we observed that the proliferative activity of CD4+ or CD8+ T cells was greatly inhibited, and the anti-tumor activity was blocked, which leads to the high tumor growth phenotype compared to control MSCs-untreated tumor. Moreover, in immunohistochemical analysis, we also confirmed that MSCs were present in tumor tissue, implying that this might inhibit the proliferation of tumor-specific T cells. MSCs suppressed T cell proliferation induced by various stimuli [30,31]. These data therefore indicate that MSCs may suppress anti-tumor immune response by regulating proliferation or infiltration of T cell into tumor tissue.

[9]

[10]

[11]

[12]

[13]

[14]

5. Conclusions [15]

In the present study, we demonstrated that MSCs induced the expression of GRO chemokines CXCL1, CXCL2, or CXCL3 and their specific receptor, CXCR2 on T cell surface. Moreover, CXCL3 inhibited proliferation and increased apoptosis in T cells, which were reversed by treatment of CXCR2 inhibitor. CXCL3 also decreased JAK2, STAT-3, and AKT phosphorylation and this was abolished by treatment of CXCR2 inhibitor. MSCs suppressed the proliferation of T cells into tumor tissue. Therefore, our work suggests that human MSCs suppress T cell proliferation and survival through CXC chemokine signaling, CXCL3, bound with CXCR2 on T cells, and this MSC-mediated T cells suppression effects may influence anti-tumor activity in vivo. Acknowledgment This work was supported by a grant from the BioGreen 21 Program (PJ0071822011) and from Woo Jang-Choon (PJ007849) projects, Rural Development Administration and by the Korea Republic of Korea.

[16]

[17]

[18]

[19] [20]

[21]

[22] [23]

[24]

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cellimm.2012. 08.002. References [1] G. Almeida-Porada, A.W. Flake, H.A. Glimp, E.D. Zanjani, Cotransplantation of stroma results in enhancement of engraftment and early expression of donor hematopoietic stem cells in utero, Exp. Hematol. 27 (1999) 1569–1575. [2] M. Angelopoulou, E. Novelli, J.E. Grove, H.M. Rinder, C. Civin, L. Cheng, D.S. Krause, Cotransplantation of human mesenchymal stem cells enhances human myelopoiesis and megakaryocytopoiesis in NOD/SCID mice, Exp. Hematol. 31 (2003) 413–420. [3] P.S. in ‘t Anker, W.A. Noort, A.B. Kruisselbrink, S.A. Scherjon, W. Beekhuizen, R. Willemze, H.H. Kanhai, W.E. Fibbe, Nonexpanded primary lung and bone marrow-derived mesenchymal cells promote the engraftment of umbilical cord blood-derived CD34(+) cells in NOD/SCID mice, Exp. Hematol. 31 (2003) 881–889. [4] Z.J. Liu, Y. Zhuge, O.C. Velazquez, Trafficking and differentiation of mesenchymal stem cells, J. Cell. Biochem. 106 (2009) 984–991. [5] B. Maitra, E. Szekely, K. Gjini, M.J. Laughlin, J. Dennis, S.E. Haynesworth, O.N. Koc, Human mesenchymal stem cells support unrelated donor hematopoietic stem cells and suppress T-cell activation, Bone Marrow Transplant. 33 (2004) 597–604. [6] A. Uccelli, V. Pistoia, L. Moretta, Mesenchymal stem cells: a new strategy for immunosuppression?, Trends Immunol 28 (2007) 219–226. [7] S. Aggarwal, M.F. Pittenger, Human mesenchymal stem cells modulate allogeneic immune cell responses, Blood 105 (2005) 1815–1822. [8] R. Maccario, M. Podesta, A. Moretta, A. Cometa, P. Comoli, D. Montagna, L. Daudt, A. Ibatici, G. Piaggio, S. Pozzi, F. Frassoni, F. Locatelli, Interaction of human mesenchymal stem cells with cells involved in alloantigen-specific

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

immune response favors the differentiation of CD4+ T-cell subsets expressing a regulatory/suppressive phenotype, Haematologica 90 (2005) 516–525. I. Rasmusson, O. Ringden, B. Sundberg, K. Le Blanc, Mesenchymal stem cells inhibit the formation of cytotoxic T lymphocytes, but not activated cytotoxic T lymphocytes or natural killer cells, Transplantation 76 (2003) 1208–1213. E. Klyushnenkova, J.D. Mosca, V. Zernetkina, M.K. Majumdar, K.J. Beggs, D.W. Simonetti, R.J. Deans, K.R. McIntosh, T cell responses to allogeneic human mesenchymal stem cells: immunogenicity, tolerance, and suppression, J. Biomed. Sci. 12 (2005) 47–57. J. Liu, X.F. Lu, L. Wan, Y.P. Li, S.F. Li, L.Y. Zeng, Y.Z. Zeng, L.H. Cheng, Y.R. Lu, J.Q. Cheng, Suppression of human peripheral blood lymphocyte proliferation by immortalized mesenchymal stem cells derived from bone marrow of Banna Minipig inbred-line, Transplant. Proc. 36 (2004) 3272–3275. R. Meisel, A. Zibert, M. Laryea, U. Gobel, W. Daubener, D. Dilloo, Human bone marrow stromal cells inhibit allogeneic T-cell responses by indoleamine 2,3dioxygenase-mediated tryptophan degradation, Blood 103 (2004) 4619– 4621. A. Corcione, F. Benvenuto, E. Ferretti, D. Giunti, V. Cappiello, F. Cazzanti, M. Risso, F. Gualandi, G.L. Mancardi, V. Pistoia, A. Uccelli, Human mesenchymal stem cells modulate B-cell functions, Blood 107 (2006) 367–372. X.X. Jiang, Y. Zhang, B. Liu, S.X. Zhang, Y. Wu, X.D. Yu, N. Mao, Human mesenchymal stem cells inhibit differentiation and function of monocytederived dendritic cells, Blood 105 (2005) 4120–4126. K. Ebnet, D. Vestweber, Molecular mechanisms that control leukocyte extravasation: the selectins and the chemokines, Histochem. Cell Biol. 112 (1999) 1–23. J. Luan, R. Shattuck-Brandt, H. Haghnegahdar, J.D. Owen, R. Strieter, M. Burdick, C. Nirodi, D. Beauchamp, K.N. Johnson, A. Richmond, Mechanism and biological significance of constitutive expression of MGSA/GRO chemokines in malignant melanoma tumor progression, J. Leukoc. Biol. 62 (1997) 588–597. D.R. Smith, P.J. Polverini, S.L. Kunkel, M.B. Orringer, R.I. Whyte, M.D. Burdick, C.A. Wilke, R.M. Strieter, Inhibition of interleukin 8 attenuates angiogenesis in bronchogenic carcinoma, J. Exp. Med. 179 (1994) 1409–1415. B. Wang, D.T. Hendricks, F. Wamunyokoli, M.I. Parker, A growth-related oncogene/CXC chemokine receptor 2 autocrine loop contributes to cellular proliferation in esophageal cancer, Cancer Res. 66 (2006) 3071–3077. J. Vandercappellen, J. Van Damme, S. Struyf, The role of CXC chemokines and their receptors in cancer, Cancer Lett. 267 (2008) 226–244. J.C. Acosta, A. O’Loghlen, A. Banito, M.V. Guijarro, A. Augert, S. Raguz, M. Fumagalli, M. Da Costa, C. Brown, N. Popov, Y. Takatsu, J. Melamed, F. d’Adda di Fagagna, D. Bernard, E. Hernando, J. Gil, Chemokine signaling via the CXCR2 receptor reinforces senescence, Cell 133 (2008) 1006–1018. V. Shyamala, H. Khoja, Interleukin-8 receptors R1 and R2 activate mitogenactivated protein kinases and induce c-fos, independent of Ras and Raf-1 in Chinese hamster ovary cells, Biochemistry 37 (1998) 15918–15924. P. Dhawan, A. Richmond, Role of CXCL1 in tumorigenesis of melanoma, J. Leukoc. Biol. 72 (2002) 9–18. L.M. Pelus, S. Fukuda, Peripheral blood stem cell mobilization: the CXCR2 ligand GRObeta rapidly mobilizes hematopoietic stem cells with enhanced engraftment properties, Exp. Hematol. 34 (2006) 1010–1020. D.F. Smith, E. Galkina, K. Ley, Y. Huo, GRO family chemokines are specialized for monocyte arrest from flow, Am. J. Physiol. Heart Circ. Physiol. 289 (2005) H1976–H1984. J. Catusse, A. Liotard, B. Loillier, D. Pruneau, J.L. Paquet, Characterization of the molecular interactions of interleukin-8 (CXCL8), growth related oncogen alpha (CXCL1) and a non-peptide antagonist (SB 225002) with the human CXCR2, Biochem. Pharmacol. 65 (2003) 813–1821. G. Chamberlain, K. Wright, A. Rot, B. Ashton, J. Middleton, Murine mesenchymal stem cells exhibit a restricted repertoire of functional chemokine receptors: comparison with human, PLoS One 3 (2008) e2934. J. Croitoru-Lamoury, F.M. Lamoury, J.J. Zaunders, L.A. Veas, B.J. Brew, Human mesenchymal stem cells constitutively express chemokines and chemokine receptors that can be upregulated by cytokines, IFN-beta, and Copaxone, J. Interferon Cytokine Res. 27 (2007) 53–64. A.L. Ponte, E. Marais, N. Gallay, A. Langonne, B. Delorme, O. Herault, P. Charbord, J. Domenech, The in vitro migration capacity of human bone marrow mesenchymal stem cells: comparison of chemokine and growth factor chemotactic activities, Stem Cells 25 (2007) 1737–1745. J. Ringe, S. Strassburg, K. Neumann, M. Endres, M. Notter, G.R. Burmester, C. Kaps, M. Sittinger, Towards in situ tissue repair: human mesenchymal stem cells express chemokine receptors CXCR1, CXCR2 and CCR2, and migrate upon stimulation with CXCL8 but not CCL2, J. Cell. Biochem. 101 (2007) 135– 146. I. Rasmusson, O. Ringden, B. Sundberg, K. Le Blanc, Mesenchymal stem cells inhibit lymphocyte proliferation by mitogens and alloantigens by different mechanisms, Exp. Cell Res. 305 (2005) 33–41. J. Plumas, L. Chaperot, M.J. Richard, J.P. Molens, J.C. Bensa, M.C. Favrot, Mesenchymal stem cells induce apoptosis of activated T cells, Leukemia 19 (2005) 1597–1604. J.R. White, J.M. Lee, P.R. Young, R.P. Hertzberg, A.J. Jurewicz, M.A. Chaikin, K. Widdowson, J.J. Foley, L.D. Martin, D.E. Griswold, H.M. Sarau, Identification of a potent, selective non-peptide CXCR2 antagonist that inhibits interleukin-8induced neutrophil migration, J. Biol. Chem. 273 (1998) 10095–10098. S. Glennie, I. Soeiro, P.J. Dyson, E.W. Lam, F. Dazzi, Bone marrow mesenchymal stem cells induce division arrest anergy of activated T cells, Blood 105 (2005) 2821–2827.

Y.-S. Lee et al. / Cellular Immunology 279 (2012) 1–11 [34] J. Hassan, D.J. Reen, Cord blood CD4+ CD45RA+ T cells achieve a lower magnitude of activation when compare with their adult counterparts, Immunology 90 (1997) 397–401. [35] S. Beyth, Z. Borovsky, D. Mevorach, M. Liebergall, Z. Gazit, H. Aslan, E. Galun, J. Rachmilewitz, Human mesenchymal stem cells alter antigen-presenting cell maturation and induce T-cell unresponsiveness, Blood 105 (2005) 2214–2219. [36] G.M. Spaggiari, A. Capobianco, S. Becchetti, M.C. Mingari, L. Moretta, Mesenchymal stem cell-natural killer cell interactions: evidence that activated NK cells are capable of killing MSCs, whereas MSCs can inhibit IL2-induced NK-cell proliferation, Blood 107 (2006) 1484–1490. [37] P. Brennan, J.W. Babbage, B.M. Burgering, B. Groner, K. Reif, D.A. Cantrell, Phosphatidylinositol 3-kinase couples the interleukin-2 receptor to the cell cycle regulator E2F, Immunity 7 (1997) 679–689.

11

[38] S. Matikainen, T. Sareneva, T. Ronni, A. Lehtonen, P.J. Koskinen, I. Julkunen, Interferon-alpha activates multiple STAT proteins and upregulates proliferation-associated IL-2Ralpha, c-myc, and pim-1 genes in human T cells, Blood 93 (1999) 1980–1991. [39] A.E. Nel, T-cell activation through the antigen receptor. Part 1: signaling components, signaling pathways, and signal integration at the T-cell antigen receptor synapse. J Allergy, Clin. Immunol. 109 (2002) 758–770. [40] W.M. Yu, S. Wang, A.D. Keegan, M.S. Williams, C.K. Qu, Abnormal Th1 cell differentiation and IFN-gamma production in T lymphocytes from motheaten viable mice mutant for Src homology 2 domain-containing protein tyrosine phosphatase-1, J. Immunol. 174 (2005) 1013–1019.