Published online before print July 3, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Division of Clinical Immunology and
Microbiology and Tumor Biology Center, and
Hematology Center and Division of Clinical Immunology, Karolinska Institutet, Stockholm, Sweden
2 Correspondence: Division of Clinical Immunology, F79, Karolinska University Hospital, SE-141 86 Stockholm, Sweden. E-mail: Ida.Rasmusson{at}ki.se
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and TNF-
production by the CTLs. Furthermore, peptide-pulsed MSCs were inefficient in stimulating tyrosine phosphorylation in specific CTLs. Our results demonstrate that MSCs induce only an abortive activation program in fully differentiated, effector CTLs, which does not involve activation of major CTL effector functions. These data may have important implications for the development of therapeutic strategies based on administration of in vitro-expanded MSCs.
Key Words: marrow stromal cells • effector CD8 T cells • activation • EBV • alloantigen
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We previously found that MSCs are capable of inhibiting the induction of CTL responses in vitro [5 ], and these findings were further confirmed by others [8 ]. It is still not clear, however, whether MSCs also exert an inhibitory effect on activated effector CTLs, which represent a cellular subset mediating graft-versus-host disease (GVHD) in transplant recipients and are used for adoptive cancer immunotherapy.
Specific activation of CTLs leads to phosphorylation of a number of signaling molecules, increased production of cytokines, and up-regulation of the IL-2R -chain (CD25). Activation of CTLs also induces release of cytolytic granules, the process termed degranulation, and up-regulates expression of ligands for death receptors, such as FasL and TRAIL (reviewed in ref. [9
]). In the present study, we demonstrate that MSCs used as antigen-presenting cells (APCs) fail to induce IFN- and TNF- production in human CD8+ CTL clones and trigger weaker up-regulation of CD25 and down-regulation of the TCR complex as compared with the effects exerted on these molecules upon CTL stimulation with a lymphoblastoid cell line (LCL). This correlated with reduced, overall phosphorylation of cellular proteins detected in CTLs after MSC-induced stimulation. The potential biological implications of this phenomenon are discussed.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Generation and maintenance of MSCs
For MSC isolation, bone marrow aspirates were obtained from the iliac crest of healthy adult volunteers, and mononuclear cells were separated by Percoll gradient centrifugation as described previously [7
]. Briefly, bone marrow mononuclear cells were plated at 160,000 cells/cm2, and suspension cells were washed away after 48 h. When adherent colonies had formed, the cells were trypsinizied and replated at 4000 cells/cm2. The cells were classified as MSCs based on their ability to differentiate into bone, fat, and cartilage [10
] and by flow cytometric analysis (positive for CD29, -44, -73, -105, and -166 and HLA class I but negative for CD14, -34, and -45). MSCs of Passages 3–8 have been used in this study.
Cell lines, CTL cultures, and clones
The HLA A11-transfected subline of the HLA class I-negative mutant cell line C1R [11
] was generated by transfection with a pHEBO vector-based HLA A11 expression vector [12
] and maintained in RPMI-1640 medium supplemented with 100 µg/ml streptomycin, 100 UI/ml penicillin, and 10% FCS (complete medium) with 200 µg/ml hygromycin B.
EBV-specific CTL cultures and clones specific to the EBV nuclear antigen 4-derived epitope IVTDFSVIK (IVT) were obtained from HLA A11-positive, healthy, Caucasian EBV carriers by stimulation of their PBMCs with a B95.8 EBV strain-transformed, autologous LCL, as described previously [13 ], to induce activation and proliferation of EBV-specific T cells. After three consecutive restimulations, the cultures were expanded in RPMI with 10% FCS supplemented with 10 U/ml recombinant IL-2 and 30% (vol/vol) culture supernatant from the gibbon lymphoma line MLA144 [14 ]. Single-cell cloning was done by limiting dilution in 96-well plates with the above medium and 100,000 irradiated (3000 rad), allogeneic, PHA-activated PBLs as feeder [15 ]. Growing cultures were transferred into 48-well plates and fed twice a week by replacing half of the medium. The EBV specificity and HLA class I restriction of the CTL preparations were investigated by testing their cytotoxic activity against a panel of EBV-positive and -negative targets including autologous LCLs, allogeneic LCLs matched through single HLA class I alleles, at least two cell lines for each allele, PHA-activated blasts, HLA-mismatched LCLs, and the prototype NK-sensitive target K562. Alloreactive CTL clones were expanded in a similar manner. HLA A2-restricted CTLs specific to the BMLF1-derived peptide epitope GLCTLVAML (GLC) [16 ] were derived from the peripheral blood of a healthy EBV carrier using FACS sorting of lymphocytes, which were stained specifically with GLC-containing HLA A2 tetramer purchased from ProImmune Ltd. (Oxford, UK). Sorted lymphocytes were subcloned and expanded as described above.
Synthetic peptides
Synthetic peptides were synthesized by the Merrifield solid-phase method at Alta Bioscience (University of Birmingham, School of Biochemistry, Birmingham, UK) and purified by HPLC on SuperPac Pep-S 5 mm reverse-phase columns (Pharmacia, Uppsala, Sweden). The purified peptides were dried using SpeedVac (Pharmacia) and diluted in DMSO at a concentration of 1 x 10–2 M, as determined by Biuret assays.
T cell activation assays
Target cells were pulsed with the indicated peptide for 1 h at 37°C, washed, and mixed with CTLs at an E:T ratio of 3:1 and incubated for 20 h in 5 ml tubes (1 million T cells/ml). The cells were washed in PBS and stained with PE-labeled CD3-, CD8-, or CD25-specific antibodies and analyzed on a FACS analyzer using CellQuest software (Becton Dickinson Labware, Franklin Lakes, NJ, USA). IFN- and TNF- were measured by intracellular fluorescent staining after activation with peptide-pulsed targets. CTLs were treated with golgistop (Becton Dickinson) for 1 h to sequester newly synthesized proteins in the Golgi complex. After incubation with targets, the cells were fixed using cytofix/cytoperm solution and stained with PE- or allophycocyanin-conjugated antibodies for 30 min on ice.
Cytotoxicity assay
Standard, 4 h 51Cr-release assays were performed as described previously [15
]. Target cells were preincubated with the relevant peptide at a range of peptide concentrations and labeled with 0.1 µCi/106 cells Na51CrO4 at 37°C for 1 h (Amersham Pharmacia Biotech UK Ltd., Buckinghamshire, UK). After extensive washing, target cells were incubated with HLA-A11 or A2-restricted, EBV-specific CTLs at an E:T ratio of 10:1 or 3:1 in triplicates for 4 h at 37°C. 51Cr release in the supernatants was measured by a -counter (Wallac Sverige AB, Stockholm, Sweden).
Analysis of protein tyrosine phosphorylation by immunoblotting
CTLs incubated with the target cells, unpulsed or pulsed with the indicated synthetic peptides for 1, 10, and 20 min in 5 ml tubes (1 million T cells/ml), were lysed in electrophoresis sample buffer (2x105 cells in 10 µl) consisting of 50 mM Tris, pH 6.8, 1 mM EDTA, 150 mM NaCl, 10% glycerol, 5% ME, 1% bromophenol blue, and 1% Triton X-100. Cell lysates were separated on a SDS-PAGE. The gels were blotted onto nitrocellulose filters, which were then incubated with the phosphotyrosine-specific antibody diluted in PBS containing 5% skimmed milk. After incubation with anti-mouse or -rabbit HRP-conjugated antibodies, the blots were visualized by SuperECL (Amersham Pharmacia Biotech UK Ltd.). Images were acquired on a FujiFilm phosphoimager using Las-100Pro ImageReader Version 2.1 software and analyzed using ScienceLab'98 Image Gauge Version 3.4X image analysis software.
Statistical analysis
The Student's t-test was used for statistical calculations. P values below 0.05 were considered statistically significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Lysis by HLA A11-specific, allogeneic CTL clones was assayed against LCLs and MSCs. As shown in Figure 1c
, lysis of A11-positive MSCs was comparable with the A11-negative control.
 View larger version (19K): [in a new window] |
Figure 1. CTL-mediated lysis of peptide-pulsed target cells. CTL lysis of C1R/A2 or C1R/A11 cells and A2 or A11 expressing MSCs were assessed using A2 (a)- or A11 (b)-restricted CTLs specific for the GLC or IVT peptide, respectively. Lysis was monitored by 4 h standard chromium release assay and is presented as percentage of specific lysis. The target cells were pulsed with the relevant synthetic peptide at the indicated concentrations. The graphs show the means and SEM for three experiments. (c) Lysis of the indicated targets by a HLA A11-specific, allogeneic CTL clone. MSCs were tested mock-treated or preincubated in the presence of IFN-.
|
increased MHC class I expression only slightly (data not shown). These results rule out the possibility that resistance of MSCs to CTL lysis is a result of low levels of MHC class I expression at the cell surface. Moreover, the variations in HLA expression observed between different MSCs could not be correlated to differences in lysis. We also analyzed the surface expression of nonclassical MHC class I molecules HLA-E and -G, which are known to inhibit activation of T cells [17
, 18
] and NK cells [19
, 20
] and therefore, could potentially suppress activation of CTLs. However, expression of these molecules at the surface of MSCs was not detected (data not shown).
 View larger version (30K): [in a new window] |
Figure 2. Expression of total HLA class I and HLA A alleles on target cells. Expression of total MHC class I or HLA A2 molecules was determined by immunostaining and FACS analysis as described in Materials and Methods. The filled graphs show total HLA class I expression on C1R/A2 (a) and MSCs (b). The dark lines show A2 allele expression, and the dashed lines are the negative control, which represents staining with a CD3-specific antibody. The figure shows one representative experiment out of three.
|
or TNF- production in CTLs and TNF- in CTLs activated with C1R/A2 or A2-positive MSCs, pulsed or unpulsed with the GLC peptide. Triggering with peptide-pulsed C1R/A2 cells increased the percentage of IFN-- and TNF--producing CTLs significantly, as was determined by intracellular staining with lymphokine-specific antibodies (Fig. 3
). In contrast, incubation with peptide-pulsed MSCs did not lead to any detectable induction of IFN- and TNF- secretion by specific CTLs.
 View larger version (32K): [in a new window] |
Figure 3. MSCs fail to induce cytokine production in triggered CTLs. GLC-specific CTLs were preincubated with control or GLC peptide-pulsed C1R/A2 cells or A2-positive MSCs. Production of IFN- and TNF- was revealed by intracellular staining with the relevant, specific antibodies and FACS analysis. The dashed lines indicate staining with an isotype control antibody (Ctrl), filled graphs show staining obtained after preincubation of CTLs with nonpulsed target cells (Unstim), and the dark line histograms show staining of CTLs stimulated by the peptide-pulsed targets (Stim). The upper and lower panels show results obtained with C1R/A2 cells or MSCs as target cells. The figure shows one representative experiment out of three.
|
-chain expression by MSC-mediated, CTL triggering-chain) enables T cells to respond to IL-2, which promotes cell proliferation. The efficiency of TCR engagement on activated CTLs is also reflected in the level of TCR and CD8 down-regulation at the cell surface [21
]. CTLs were activated by peptide-pulsed C1R/A2 or A2-positive MSCs, and levels of surface expression of CD25, CD3, and CD8 were determined 20 h after triggering (Fig. 4
). Activation with peptide-pulsed C1R/A2 cells resulted in more than tenfold up-regulation of CD25 expression and considerable CD3 and CD8 down-regulation at the surface of CTLs. In contrast, MSC-mediated triggering induced low CD25 up-regulation (approximately threefold, P<0.05, compared with C1R) but failed to induce CD3 or CD8 down-regulation in specific CTLs (not significant compared with C1R).
 View larger version (14K): [in a new window] |
Figure 4. Alterations in surface expression of CD3, CD8, and CD25. (a) IVT-specific CTLs were stimulated with peptide-pulsed C1R/A11 cells or A11-positive MSCs. Expression of CD3 and CD8 on the surface of activated CTLs was monitored by immunostaining and FACS analysis as described in Materials and Methods. The data are shown as percentage of CD3 or CD8 down-regulation on activated CTLs as compared with cells preincubated with control APCs without the peptide. (b) The experiments were performed essentially as described in a. Expression of CD25 is presented as percentage of its up-regulation on the surface of CTLs, preincubated with C1R/A2, C1R/A11, or MSCs expressing the relevant class I alleles. The means and SEM of data obtained in three independent experiments performed with different APCs and CTLs are shown in the figure.
|
 View larger version (28K): [in a new window] |
Figure 5. Tyrosine phosphorylation in peptide-specific CTLs after C1R/A11 or MSC-mediated triggering. Tyrosine phosphorylation was analyzed in IVT-specific CTLs after incubation with peptide-pulsed or unpulsed targets (C1R/A11 or A11-positive MSCs) at an E:T ratio of 3:1. The figure shows three separate time-points of incubation: 1, 10, and 20 min. Position 1, CIR/A11, peptide-pulsed; 2, C1R/A11, unpulsed; 3, A11-positive MSCs, peptide-pulsed; 4, A11-positive, unpulsed MSCs. The figure shows one representative blot out of three performed experiments.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and INF- secretion, which are readily released by CTLs on triggering with other types of APCs.
In our experiments, MSCs were lysed by specific CTLs inefficiently, despite intermediate-to-high expression of HLA class I, corroborating previous findings that PBMCs do not proliferate in response to MSCs, even if the latter are exposed to IFN-, which induces high levels of HLA classes I and II expression [24
]. In addition, MSCs did not induce production of the proinflammatory cytokines IFN- and TNF-. This is consistent with studies reported previously, indicating that MSCs do not induce production of IFN- or increased expression of lymphocyte activation markers CD25, CD38, and CD69, whereas fibroblasts do trigger these changes in T cells [1
, 22
, 23
, 26
]. The effect could reflect an inefficient activation or active suppression of effector CTLs by MSCs or both.
When MSCs are present in MLC, T, B, and dentritic cell proliferation is suppressed through MHC-independent mechanisms [1 , 2 , 4 , 8 , 27 ]. The factor mediating suppression by MSCs appears to be soluble, as proliferation is still reduced if MSCs and lymphocytes are separated in a transwell system [2 3 4 , 7 , 28 ]. MSCs suppress the formation of effector CTLs if present during the stimulation phase but do not affect lysis of third-party targets by effectors formed already [5 ]. This, however, does not exclude the possibility that MSCs actively suppress other T cell activation events, e.g., lymphokine production and proliferation. Triggering of cytotoxic activity usually requires a low level of activating signal, and lytic activity remains intact in CTLs, which are nonresponsive to the specific stimulus in terms of other parameters of activation [21 , 29 , 30 ]. In accordance with this possibility, the analysis of protein phosphorylation performed in this study suggests that MSCs affect the amplitude and probably the quality of signal transduction events induced by specific TCR triggering in CTLs. Our data are generally consistent with an inability of MSCs to trigger CTL activation. However, the delayed kinetics of tyrosine phosphorylation in CTLs suggests that MSCs may modulate signal transduction actively. Resistance of MSCs to CTL-mediated lysis may be important for their capacity to suppress immune activation across major and minor histocompatibility barriers.
Several animal studies have shown that allogeneic MSCs persist when transplanted, without immunosuppression, across MHC barriers [31 32 33 34 ]. Low levels of engraftment were also observed after infusion of haploidentical MSCs to a patient with severe aplastic anemia and in patients treated with third-party-derived, HLA-mismatched MSCs for steroid-resistant GVHD after allogeneic stem cell transplantation [35 , 36 ]. We reported recently about fully mismatched, allogeneic MSCs transplanted into an immunocompetent fetus with osteogenesis imperfecta [37 ]. Persistence of transplanted cells 9 months post-transplant and the lack of immunoreactivity when patient lymphocytes were re-exposed to the MSC graft in vitro indicate that MSCs can be tolerated when transplanted across MHC barriers in humans.
Studies in experimental animals and patients suffering from severe GVHD suggest that MSCs may be immunosuppressive in vivo [36 , 38 39 40 41 ]. In contrast, Sudres et al. [42 ] reported recently that although MSCs suppress alloantigen-induced proliferation in vitro, the cells failed to prevent GVHD in a murine model. Previous reports indicate that MSCs derived from C57BL/6 mice, as used by the authors, evoke an immune response with subsequent graft rejection in BALB/c recipients, although no immunoreactivity was detected in MSC lymphocyte cocultures prior to transplant [43 ]. Immunoreactivity, not initially detectable in vitro, was also reported between other strains of mice and in xenogenic systems [44 , 45 ]. Clearly, species-specific differences exist as suppression by mouse MSCs is cell–cell contact-dependent and induces a cell-division arrest not observed in human lymphocytes [1 , 22 , 46 47 48 ].
Autologous and allogeneic MSCs are evaluated currently in clinical trials as suppressors of pathologic immune responses and treatment of organ damage. Passive or active resistance of MSC to CTLs may be important for transplantation across MHC barriers as well as for the capacity of MSCs to inhibit ongoing GVHD and autoimmune reactions.
Received March 8, 2007; revised May 10, 2007; accepted May 21, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
protects short-term ovarian carcinoma cell lines from CTL lysis via a CD94/NKG2A-dependent mechanism J. Clin. Invest. 110,1515-1523[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  H. Karlsson, S. Samarasinghe, L. M. Ball, B. Sundberg, A. C. Lankester, F. Dazzi, M. Uzunel, K. Rao, P. Veys, K. Le Blanc, et al. Mesenchymal stem cells exert differential effects on alloantigen and virus-specific T-cell responses Blood, August 1, 2008; 112(3): 532 - 541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Morandi, L. Raffaghello, G. Bianchi, F. Meloni, A. Salis, E. Millo, S. Ferrone, V. Barnaba, and V. Pistoia Immunogenicity of Human Mesenchymal Stem Cells in HLA-Class I-Restricted T-Cell Responses Against Viral or Tumor-Associated Antigens Stem Cells, May 1, 2008; 26(5): 1275 - 1287. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
