Toll-like Receptor–Mediated Signaling in Human Adipose-Derived Stem Cells: Implications for Immunogenicity and Immunosuppressive Potential

Abstract

Human adipose-derived stem cells (hASCs) are mesenchymal stem cells with reduced immunogenicity and the capability to modulate immune responses. These properties make hASCs of special interest as therapeutic agents in the settings of chronic inflammatory and autoimmune diseases. Exogenous and endogenous toll-like receptor (TLR) ligands have been linked with the perpetuation of inflammation in a number of chronic inflammatory diseases such as inflammatory bowel disease and rheumatoid arthritis because of the permanent exposure of the immune system to TLR-specific stimuli. Therefore, hASCs employed in therapy are potentially exposed to TLR ligands, which may result in the modulation of hASC activity and therapeutic potency. In this study, we demonstrate that hASCs possess active TLR2, TLR3, and TLR4, because activation with specific ligands resulted in induction of nuclear factor kappa B–dependent genes, such as manganese superoxide dismutase and the release of interleukin (IL)-6 and IL-8. TLR3 and TLR4 ligands increased osteogenic differentiation, but no effect on adipogenic differentiation or proliferation was observed. Moreover, we show that TLR activation does not impair the immunogenic and immunosuppressive properties of hASCs. These results may have important implications with respect to the safety and efficacy of hASC-based cell therapies.

Introduction

Toll-like receptors (TLRs) are broadly expressed in immune and non-immune cells and are important sensors for the detection of foreign pathogens and endogenous danger signals.^1,2 TLRs activate a rapid effector response, which implies innate and adaptive immune responses.³ Ligand recognition results in the recruitment of intracellular Toll/Interleukin1 Receptor (TIR)-domain-containing adaptor proteins, including myeloid-differentiation primary-response protein 88 (MyD88) and Toll/interleukin (IL)-1R domain-containing adaptor-inducing interferon (IFN)-β (TRIF), leading to the activation of MyD88-dependent or MyD88-independent pathways.^4–6 TLR triggering leads to activation of nuclear factor kappa B (NF-kB) and mitogen-activated protein kinase (MAPK) signalling pathways, which results in the induction of pro-inflammatory cytokines, including tumor necrosis factor-alpha (TNFα), IL-6, IL-1β, and type I IFN and the up-regulation of co-stimulatory molecules.

Eleven human TLRs have been identified that recognize distinct microbial products from bacteria, viruses, protozoa, and fungi. Thus, TLR2 recognizes peptidoglycans (PGNs), lipoproteins, and lipoteichoic acids from gram-positive bacteria; TLR3 recognizes double-stranded RNA (dsRNA) and poly-inosinic acid:cytidylic acid poly I:C, a synthetic dsRNA analog; TLR4 recognizes gram-negative lipopolysaccharide (LPS); TLR5 senses bacterial flagellin; and TLR9 detects unmethylated bacterial DNA.^1,4 In addition, the recognition of endogenous ligands by TLRs (e.g., heat shock proteins, cellular DNA) is thought to have an important role in the regulation of inflammation.^7–11

Mesenchymal stem cells (MSCs) are multipotent adult stem cells capable of differentiation to mesenchymal-type cells (adipocytes, osteoblasts, and chondrocytes) and to myocytes, neurons, endothelial cells, astrocytes, and epithelial cells.^12–14 Adipose tissue is a potential source of MSCs referred to as human adipose-derived mesenchymal stem cells (hASCs).¹⁵ hASCs can be isolated from liposuctioned fat tissue and expanded over a long time in culture. In addition to their differentiation potential, hASCs share with other MSCs the unique ability to suppress immune responses. Ex vivo expanded MSCs and ASCs have been reported to inhibit activation, proliferation, and function of immune cells, including T cells, B cells, natural killer cells, and antigen-presenting cells.^16–18 There is evidence that this ability to modulate immune responses relies on cell contact–dependent mechanisms and soluble factors secreted by MSCs and ASCs in response to cytokines released by activated immune cells. Soluble factors such as hepatocyte growth factor, prostaglandin E2, transforming growth factor (TGF)-β1, indoleamine 2,3-dioxygenase (IDO), nitric oxide, and IL-10 have been implicated.^19–26 Several reports have shown that IFNγ plays an active role in the immunosuppression mediated by MSCs and ASCs through its induction of prostaglandins and IDO release.²¹ An additional immunological feature of all MSCs, hASCs among them, is that they are considered to be poorly immunogenic because they express low levels of human leukocyte antigen (HLA)-I but do not express HLA-II, CD40, CD80, or CD86. This phenotype is thought to lead to the activation of T cells mediated via HLA-I but concomitantly to anergy due to the absence of co-stimulatory molecules.²⁷ Recent studies have shown that MSCs express active TLRs, which may modulate stem cell function.^28–31

The two biological abilities (differentiation and immunosuppression) make hASCs an interesting tool for cellular therapy and regeneration. MSCs are being used in several clinical trials, with a focus on their immunomodulatory capacities (http://clinicaltrials.gov/search/term=stem+cells?term=stem+cells). We are currently conducting a phase III clinical trial using hASCs to treat complex peri-anal fistula in patients with and without Crohn's disease after having successfully completed phase I and II trial with high efficacy rates.^32,33 TLR activation has been implicated in the pathology of a number of inflammatory diseases, including inflammatory bowel disease, because they can initiate or perpetuate the chronic inflammation due to the continued exposure to TLR ligands in the gut.^34,35 Therefore, the use of hASCs in cell therapy for the treatment of inflammatory diseases deserved further investigation regarding the potential effects of TLR signaling on hASCs biology and the potential implications for immunogenicity and immunosuppressive capacity, which are of special relevance in terms of therapeutic potency.

Here we report that hASCs express virtually all TLRs, except TLR8 at the messenger RNA (mRNA) level. Activation with specific ligands resulted in induction of the NF-κB-dependent genes manganese superoxide dismutase (MnSOD), cyclooxygenase (COX)-2, IL-6, and IL-8. TLR3 and TLR4 ligands increased differentiation into osteoblasts, but no effect was observed in proliferation, immunogenicity, or immunosuppressive capabilities. These results have important implications for the safety and efficacy of hASC-based cell therapies.

Materials and Methods

Isolation and expansion of cells

hASC isolation

Lipoaspirates obtained from human adipose tissue from healthy adult male and female donors were washed twice with phosphate buffered saline (PBS) to remove contaminating debris and red blood cells and digested at 37°C for 30 min with 0.075% collagenase (Type I, Invitrogen, Carlsbad, CA) in PBS. The digested sample was washed with 10% fetal bovine serum (FBS), treated with 160 mM of ammonium chloride to eliminate the remaining red blood cells, suspended in culture medium (Dulbecco's modified Eagle medium *DMEM) containing 10% FBS) and filtered through a 40-μm nylon mesh. Cells were seeded (2–3 × 10⁴ cells/cm²) onto tissue culture flasks and expanded at 37°C and 5% carbon dioxide. The culture medium was changed every 3 to 4 days. Cells were passed to a new culture flask (1,000 cells/cm²) when cultures reached 90% of confluence. Cells were phenotypically characterized according to their capacity to differentiate into chondro-, osteo-, and adipogenic lineages. In addition, hASCs were verified by staining with specific surface markers. hASCs were positive for HLA-I, CD90, CD105, and CD59 and negative for HLA-II, CD40, CD80, CD86, CD34, CD14, CD18, and CD45. A pool of six different hASC samples from male and female donors (culture passage 4–6) were used in the study.

Human peripheral blood lymphocytes

The National Transfusion Centre of the Comunidad Autónoma of Madrid (CAM) kindly provided Buffy coats. Peripheral blood mononuclear cells (PBL) were isolated using density centrifugation gradient using Ficollplaque Plus (GE Healthcare Biosciences AB, Uppsala, Sweden). After isolation, cells were washed and stored before use.

Immunomagnetic cell separation

Lymphocyte subpopulations were isolated using antibody-coated magnetic beads (Miltenyi Biotech, Madrid, Spain) against CD8 or CD4 surface molecules. After washing, CD8 fraction was collected using the AUTOMACS system (Miltenyi Biotech), and the depleted CD8 negative fraction was stained for CD4 with the coated beads, following the manufacturer's instructions. The CD4-positive fraction was then collected, and both fractions (CD4 and CD8) were stored separately until use. The medium for culture and proliferation assays was RPMI 1640 supplemented with 10% FBS, 2mM L-glutamine, 1% non-essential amino acids, 1% pyruvate, and 1% penicillin and streptomycin.

Reagents and antibodies

The following anti-human monoclonal antibodies (mAbs) were used for flow cytometry: antibodies against HLA-I, CD80, CD86, and HLA-II were purchased from BD Bioscience (San Jose, CA). Anti-TLR4 was from Serotec (Raleigh, NC), and Anti-TLR2 was from eBioscience (San Diego, CA). Antibodies and their respective isotypes (negative controls) used for surface staining were all titrated under the appropriate conditions. Anti-COX-2 and anti-IκB-α were from Santa Cruz Biotechnology; LPS (Salmonella typhimurium) was purchased from List Biological Laboratories; poly I:C, oligodeoxynucleotides (ODNs), and PGNs were from Invivogen (San Diego, CA) and IFNγ was from PeproTech (Rocky Hill, NJ); 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), sulfanilic acid, naphthylenediamine, 5(6)- carboxyfluorescein diacetate N-succinimidyl ester (CFSE), dexamethasone, insulin, indomethacin, isobutyl-methylxanthine, 1,25-dihydroxyvitamin D3, ascorbate-2-phosphate, β-glycerophosphate, oil-red-O, and alizarin red were from Sigma-Aldrich (St. Louis, MO); and dihydroethidium and chloromethyl-2′,7′-dichlorofluorescein diacetate were from Invitrogen.

Cellular staining and flow cytometry

CFSE labelling

Ten to 20 × 10⁶ cells were extensively washed to remove FBS, resuspended in 200 μL of a 10-μM CFSE solution, and incubated under constant shaking at 37°C for 10 min. The reaction was stopped by adding ice-cold RPMI 10% FBS, after which the cells were washed twice with ice-cold PBS. Cells were then cultured overnight, and one aliquot was used to set up and control the FL-1 voltage for CFSE.

Flow cytometry

A total of 10⁶ hASCs were left untreated or treated with IFNγ (30 ng/mL), LPS (1 μg/mL), PGN (10 μg/mL), poly I:C (1 μg/mL) and maintained in culture for 72 h. Cells were harvested and stained with the monoclonal antibodies, and appropriate isotopic controls were included. Intracellular antibody staining was achieved after fixation and permeabilization as indicated by the manufacturer (cytofix/cytoperm buffers, BD Biosciences). Flow cytometry was performed using a Coulter XL flow cytometer equipped with Expo32 software, 10 × 10³ events were acquired for each tube. The acquisition and analysis gates were chosen based on the forward and side scatter properties of cells.

Cell proliferation assays

hASCs proliferation

At day 0, hASCs were seeded at 2 × 10⁴ cells per well in 24-well plates in 1 mL of DMEM 10% FBS overnight. Cells were then incubated under the indicated experimental conditions. At different time points, medium was removed, and 0.5 mL of MTT solution (0.5 mg/mL in RPMI 10% FBS) was added to each well. After 30 min at 37°C, cells were washed once with PBS, and 100 μL of dimethyl sulfoxide was added. Spectrophotometric measurement at 570 nm was performed using a 96-well plate reader (iEMS Reader, ThermoElectron Corp., Waltham, MA).

PBL proliferation assay

After overnight resting, CFSE-labelled cells were activated with the Pan T cell Activation kit (microbeads loaded with anti-CD3, anti-CD2, and anti-CD28, Miltenyi Biotech, Madrid, Spain) following the manufacturer's instructions or left inactivated. Two days before PBL activation, ASCs were plated in a 24-well plate (5–4 × 10⁴ cells/well). Lymphocytes (10⁶ cells/mL) were cultured with or without hASCs and in the presence or absence of 1 μg/mL of LPS, 1 μg/mL of poly I:C, or 10 μg/mL of PGN. At day 5, PBLs were harvested, and cell proliferation was determined according to loss of CFSE in the FL-1 channel. A FACScalibur cytometer (Becton Dickinson, San Diego, CA) was used to measure the cells. Data were analyzed using CellQuest-pro software (Becton Dickinson) over gated lymphocytes (based on forward scatter/side scatter properties). Percentage of proliferating lymphocytes was obtained by gating in the region (M1) of the FL-1 channel corresponding to the last 2 days of culture.

Reverse transcriptase polymerase chain reaction

Total RNA from 80% confluent hASCs was isolated, and 1 μg of the total RNA was reverse transcribed using SuperScript II (Invitrogen). Human TLRs were amplified using specific primers (5′to 3′): TLR1 forward AAACGGTCTCATCCACGTTC, reverse CCAAGTGCTTGAGGTTCACA; TLR2 forward GGCC AGCAAATTACCTGTGT, reverse TTCTCCACCCAGTAGGCATC; TLR3 forward AGCCTTCAACGACTGATGCT, reverse TTTCCAGAG CCGTGCTAAGT; TLR4 forward CAAAATCCCCGACAACCTCC, reverse TGTAGAACCCGCAAGTCTGTGC; TLR5 forward CAGAAACCTGCCCA ACCTTA, reverse TCCCAAATGAAGGATGAAGG; TLR6 forward TTCCAGAGCTGCCAGAAGAT, reverse CCAGGGCAGATCCAAGTAGA; TLR7 forward GGAAATTGCCCTCGTTGTTA, reverse CTGGGGAGA AAATGCAGAAA; TLR8 forward GTTTCCTCGTCTCGAGTTGC, reverse TCAAAGGGGTTTCCGTGTAG; TLR9 forward CAGCAGCTCTG CAGTACGTC, reverse AAGGCCAGGTAATTGTCACG; TLR10 forward GGCCAGAAACTGTGGTCAAT, reverse AACTTCCTGGCAGCTCTGAA (95°C for 20 s, 52°C for 1 min, and 72°C for 1 min, 35 cycles). The complementary DNA of THP-1 cells was used as a positive control for TLR mRNA expression.

Western blot

Cell extracts were obtained using CelLytic M Cell lysis reagent (Sigma-Aldrich) containing protease inhibitors, and protein concentration was determined using bicinchoninic acid protein assay (Pierce, Rockford, IL). Lysates were separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred onto nitrocellulose membrane, and blotted with primary and secondary antibodies, and proteins were visualized using enhanced chemoluminescence. Anti-MnSOD (SOD-110) was obtained from Stressgen (Ann Harbor, MI). Anti-β-actin was from Sigma-Aldrich.

Cytokine detection

hASCs were treated with TLR ligands, and supernatants were collected at different time points. Cytokine secretion was tested using enzyme-linked immunosorbent assay (ELISA) following the manufacturer's instructions (eBioscience). Measurements were obtained at 450 nm using a 96-well plate reader (Bio-Rad, model 680, Hercules, CA). In addition, simultaneous detection in conditioned supernatants of multiple soluble cytokines was performed using the Cytometric Bead Array immunoassay (BD Biosciences, Becton Dickinson, San Diego, CA). Data were acquired in a FACScalibur and analyzed using Cellquest-Pro software.

IDO activity

IDO activity was measured by determining tryptophan and kynurenine concentrations in conditioned supernatants. Two hundred μL of supernatants was added to 50 μL of trichloroacetic acid 2M and vortexed. After centrifugation for 10 min at 13,000 rpm, 100 μL of supernatant was analyzed using high-performance liquid chromatography (HPLC; Waters 717plus Autosampler, Milford, MA).

Nitric oxide detection

The generation of nitrite was measured using the Griess reaction. Briefly, after stimulation with the corresponding stimuli, the nitrite content in the supernatant was measured using incubation with Griess reagent (sulfanilamide and N-1-napthylethylenediamine dihydrochloride). The nitrite concentration was determined spectrometrically at 540 nm (model 680, Bio-Rad) and calculated from a sodium nitrite standard curve.

hASC differentiation

Adipogenic differentiation was achieved after 3 weeks in adipogenic medium (DMEM 10% FBS), containing 1 μM of dexamethasone, 10 μM of insulin, 200 μM of indomethacin, and 0.5 mM of isobutyl-methylxanthine. Adipocytes were stained with oil-red-O. Osteogenic differentiation was achieved after 3 weeks in DMEM 10% FBS containing 0.01 μM of 1,25-dihydroxyvitamin D3, 50 μM of ascorbate-2-phosphate, and 10 mM of β-glycerophosphate. Osteoblasts were stained with alizarin red. Where indicated, LPS (1 μg/mL), poly I:C (1 μg/mL), or PGN (10 μg/mL) was added during differentiation.

Results

hASCs express TLRs

To determine whether hASCs express TLRs, we analyzed levels of TLR mRNA according to reverse transcriptase polymerase chain reaction (PCR) using specific primers for human TLR1 to TLR10. The PCR revealed that hASCs express mRNA of all TLRs except TLR8, similar to THP-1 cells, a monocytic human cell line that was used as a positive control (Fig. 1A). To confirm expression of TLRs in hASCs at the protein level, TLR2 and TLR4 expression was studied using flow cytometry using specific antibodies. As expected, hASCs stained positively for TLR2, TLR4, and HLA-I, which was used as a control for staining (Fig. 1B). To further understand the regulation of TLR expression by proinflammatory cytokines, hASCs were stimulated with IFNγ, TNFα, LPS, or combinations of them, and TLR2 and TLR4 expression was determined using flow cytometry 72 h after stimulation. None of the stimuli employed showed any significant effect on TLR2 and TLR4 expression (data not shown). These results indicate that hASCs express detectable levels of TLRs.

FIG. 1.

Toll-like receptor (TLR) expression by human adipose-derived stem cells (hASCs). (A) Total RNA was isolated from hASCs and tetrahydropyran (THP)-1 cells. TLR expression was determined according to reverse transcriptase polymerase chain reaction (PCR) using specific primers for human TLR1 to TLR10 or in the absence of any primer as a PCR control (–). (B) Representative flow cytometry analysis of TLR2, TLR4, and human leukocyte antigen-I (black line) and corresponding isotype antibodies controls (grey line) in hASCs. Histograms from one representative experiment are shown (n = 3).

TLR ligands activate hASCs

TLR activation leads to the nuclear translocation of NF-κB and activation of a gene expression program that includes induction of the anti-oxidant gene MnSOD, COX-2, IL-6, and IL-8.^36–41 We therefore investigated the expression of functional TLRs by stimulating hASCs with different concentrations of PGN (TLR2 ligand), poly I:C (TLR3 ligand), LPS (TLR4 ligand), and ODN (TLR9 ligand). MnSOD and COX-2 were detected using Western blot at, after 72 h poststimulation. As shown in Figure 2A, PGN induced MnSOD poorly even at the highest concentration, whereas LPS and poly I:C induced a strong expression of MnSOD even at the lowest concentration. However, no induction was observed after stimulation with ODN. COX-2 expression was stimulated by poly I:C only at the highest concentration (Fig. 2A and data not shown). To further demonstrate that TLRs are expressed and can be activated, IL-6, TNFα, and IFNγ production was determined using ELISA. Of the TLR ligands tested, PGN, LPS, and poly I:C induced IL-6 production, although they failed to induce TNFα and IFNγ production. Again, ODNs failed to induce any tested cytokine (Fig.2B and data not shown). Furthermore, simultaneous detection of soluble cytokines induced by TLR activation was determined using cytometric bead arrays (Becton Dickinson) 72 h after stimulation with LPS, poly I:C, and PGN. Whereas unstimulated hASCs showed expression of IL-6 and IL-8, activation by LPS, poly I:C, and PGN led to a significant increase in IL-6 and IL-8 secretion (Fig. 2C). However, TLR activation did not induce production of IL-1β, IL-2, IL-4, IL-5, IL-10, IL-12, IFNγ, or TNFα. Finally, to confirm that stimulation of hASCs by LPS, poly I:C, and PGN leads to the activation of the NF-κB signalling cascade, degradation of IκB-α, which is required for efficient nuclear translocation of NF-κB, was analyzed at different time points. As shown in Figure 2D, stimulation with LPS, poly I:C, or PGN led to a rapid degradation of IκB-α, demonstrating that NF-κB signalling cascade is turned on. Together, our results show that hASCs express functional TLR2, TLR3, and TLR4.

FIG. 2.

Human adipose-derived stem cells (hASCs) express functional Toll-like receptors (TLRs). (A) hASCs were left unstimulated or were stimulated with increasing concentrations of lipopolysaccharide (LPS), poly-inosinic acid:cytidylic acid (Poly I:C), peptidoglycan (PGN) and ODN. After 72 hours of treatment, TLR-mediated induction of manganese superoxide dismutase and cyclooxygenase-2 was determined using Western blot. (B) Supernatants from the same experiments were collected to determine interleukin (IL)-6 concentration. Unstimulated cells (lined bar) were compared with TLR2- (white bars), TLR3- (light grey bars), TLR4- (dark grey bars), and TLR9-stimulated (black bars) cells at three different concentrations. Means and standard deviations of three different experiments are represented. (C) Simultaneous detection of 10 different cytokines induced 72 h after activation with LPS (1 μg/mL), Poly I:C (1 μg/mL), or PGN (10 μg/mL) was carried out using cytometric bead arrays. Cytokines were gated based on their FL3-FL4 position (upper left dot plot). FL2 intensity of the remaining plots indicates concentration of the gated cytokines. Gray line indicates baseline levels of IL-6 and IL-8 by unstimulated hASCs. Representative plots from one experiment are shown (n = 3). (D) hASC were left unstimulated or were stimulated with LPS (1 μg/mL), Poly I:C (1 μg/mL), or PGN (10 μg/mL). At the indicated time points, degradation of IκB-α was determined according to Western blot.

TLR and hASC differentiation

One of the main features of hASCs is their potential to differentiate to adipocytes, osteoblasts, and chondrocytes when cultured in the appropriate media. To investigate whether TLRs can modulate differentiation, hASCs were differentiated into adipocytes or osteoblasts in the presence or absence of PGN, poly I:C, LPS, and ODNs. Whereas adipocyte differentiation was not affected, poly I:C and LPS stimulation significantly increased osteoblast differentiation (Fig. 3A and data not shown). Levels of reactive oxygen species (ROS) seem to play a role during osteogenic differentiation.^42,43 To determine whether increased ROS levels may play a role in the observed effects of LPS and poly I:C on hASC osteogenesis, staining with 7′-dichlorofluorescein diacetate and dihydroethidium were performed after 48 h of stimulation. Neither LPS nor poly I:C showed any effect on ROS levels of hASCs (data not shown).

FIG. 3.

Toll-like receptor (TLR) ligation effects on human adipose-derived stem cells (hASC) phenotype. (A) hASCs were cultured for 3 weeks in osteogenic differentiation medium in the absence (control) or presence of lipopolysaccharide (LPS; 1 μg/mL) or poly-inosinic acid:cytidylic acid (Poly I:C; 1 μg/mL), and osteogenic differentiation was analyzed using alizarin red staining, as indicated in Materials and Methods. Representative pictures from two independent experiments are depicted. (B) Expression by hASC of the indicated surface markers was determined using flow cytometry after 72 h of culture in the absence (control) or presence of interferon gamma (IFNγ; 30 ng/mL), LPS (1 μg/mL), Poly I:C (1 μg/mL), or peptidoglycan (PGN; 10 μg/mL), as described in Materials and Methods. Overlay histograms from one representative experiment are shown (n = 3). Isotype monoclonal antibody (mAb) (…), specific mAb in control hASCs (–), and specific mAb in stimulated hASCs (—). (C) 2 × 10⁴ hASCs were seeded in 24-well plates and cultured for 2 weeks in the absence (control) or presence of LPS (1 μg/mL), Poly I:C (1 μg/mL), or PGN (10 μg/mL), and the effects on proliferation were analyzed according to 3-(4,5-(95%)dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay as described in Materials and Methods. Data shown are representative of three different experiments.

TLR and hASC immunogenic phenotype

TLRs have been shown to modulate expression of co-stimulatory molecules in immune cells.³ To test whether TLR activation might modify the immunogenic profile of hASCs, we analyzed the expression of HLA-I, HLA-II, CD80, and CD86 using flow cytometry 72 h after stimulation with IFNγ, LPS, poly I:C, and PGN. As expected, IFNγ up-regulated the expression of HLA-I, induced expression of HLA-II, and failed to induce the expression of CD80 and CD86. LPS, poly I:C, and PGN did not alter the expression of HLA-II, CD80, and CD86, whereas poly I:C was the only TLR ligand capable of inducing HLA-I similarly to IFNγ (Fig. 3B). Furthermore, costimulation of hASCs with IFNγ in combination with LPS, poly I:C, or PGN did not alter the INFγ-mediated induction of HLA-I and HLA-II (data not shown). Together, these results indicate that TLR activation does not significantly affect the immunogenic properties of hASCs.

TLR effects on hASC proliferation

To determine the role of TLR ligands in hASC proliferation, we compared the proliferation of hASCs in the absence or presence of TLR ligands using the MTT proliferation assay. As shown in Figure 3C, only LPS induced a moderate increase in proliferation, although it was not statistically significant (p > 0.05). Similar results were obtained using cell counting (data not shown). Together, these results indicate that TLR ligands have no significant effect on hASC proliferation.

TLRs and induction of immune modulators

Adult mesenchymal stem cells, hASCs among them, have been shown to inhibit proliferation of PBLs upon mitogenic or allogeneic activation.¹³ This important role is mediated, at least in part, through the induction of immune modulators upon activation. One of the molecules found to inhibit PBL proliferation is nitric oxide, which has been reported to be induced by TLRs in monocytes and macrophages.⁴⁴ We therefore wondered whether TLR triggering could result in the release of nitric oxide by hASCs. To determine nitric oxide production, hASCs were left unstimulated or were stimulated with increasing concentrations of LPS, poly I:C, and PGN, and nitric oxide concentration was determined in conditioned supernatants 72 h after stimulation (Fig. 4A and data not shown). Bone marrow–derived macrophages (BMDMs) stimulated with LPS were used as a positive control for nitric oxide detection. None of the TLR ligands tested in our experimental conditions (not even at the highest concentration of 10 μg/mL) induced the release of nitric oxide (Fig. 4A and data not shown).

FIG. 4.

Toll-like receptor (TLR) effects on human adipose-derived stem cell (hASC)-mediated production of immunomodulators. hASCs were left unstimulated (control) or were stimulated with increasing concentrations of interferon gamma (IFNγ), lipopolysaccharide (LPS), poly-inosinic acid:cytidylic acid (Poly I:C), and peptidoglycan (PGN). (A) Seventy-two h after stimulation, supernatants were collected and tested for nitric oxide release, as described in Materials and Methods. Bone marrow–derived macrophages (BMDM) stimulated with LPS (100 ng/mL) were used as a positive control for nitric oxide detection. One experiment with hASCs stimulated with 30 ng/mL of IFNγ or 10 μg/mL of the corresponding TLR ligands is shown. (B) Tryptophan and kynurenine concentrations were determined in conditioned supernatants after 72 h of stimulation with the indicated stimuli, as described in Materials and Methods. Means and standard deviations of three experiments are represented.

IDO enzyme is essential in the tryptophan catabolism, leading to the conversion of tryptophan into kynurenine. IDO activity plays a key role in mediating suppression of PBL proliferation.¹⁶ Different stimuli, most prominently IFNγ, induce transcription of IDO. In this regard, it should be noted that TLRs have been shown to induce IDO expression in dendritic cells and macrophages.⁴⁵ Therefore, to determine whether TLRs activation results in IDO expression and activity, we stimulated hASCs with increasing concentrations of LPS, poly I:C, and PGN. Tryptophan and kynurenine concentrations were determined using HPLC in supernatants 72 h after stimulation. As expected, IFNγ induced strong IDO activity, whereas LPS and PGN failed to induce it. However, poly I:C was capable of inducing IDO expression and activity at the highest concentration (10 μg/mL) only, although weaker than that observed with IFNγ (Fig. 4B).

TLRs do not inhibit hASC immunosuppressive capacity

The immunosuppressive capacity of hASCs is a key factor in their therapeutic use and potential. Therefore, we tested the role of TLRs in the immunosuppressive capacity of hASCs. To do so, we analyzed proliferation of activated CFSE-labeled PBLs in the absence or presence of increasing amounts of hASCs pre-cultured for 24 h with medium alone or in the presence of LPS, poly I:C, or PGN (Fig. 5A). As expected, hASCs efficiently suppressed PBL proliferation when 2 × 10⁴ or 4 × 10⁴ cells were plated but not at lower concentrations (5 × 10³ or 1 × 10⁴ cells). No significant effect of TLRs on hASC-mediated suppression was observed. Moreover, neither a longer preincubation of hASCs with TLR ligands for 72 h nor incubation with higher concentrations of TLR ligands (10 μg/mL) altered their suppressive potential in our experimental conditions (data not shown). Furthermore, to exclude the possibility that TLR activation may have an effect on hASC-mediated inhibition of proliferation of T cell subsets, CFSE-labeled CD4+ or CD8+ T cells were stimulated and cultured in the absence or presence of 4 × 10⁴ hASCs pre-cultured for 24h with medium alone or in the presence of LPS (1 μg/mL), poly I:C (1 μg/mL), or PGN (10 μg/mL). As shown in Figure 5B, no effect on hASC-mediated suppression was found. As a whole, these results indicate that activation through TLR2, TLR3, and TLR4 do not significantly interfere with the capacity of hASCs to modulate immune responses in vitro.

FIG. 5.

Toll-like receptor (TLR) signaling does not impair the inhibitory effect of human adipose-derived stem cells (hASCs) on lymphocyte proliferation. (A) Peripheral blood lymphocytes (PBLs) were labeled with carboxyfluorescein diacetate N-succinimidyl ester (CFSE) and stimulated with anti-CD3/CD2/CD28 antibodies. PBLs were cultured in the absence (black line) or presence of increasing numbers of hASCs pre-cultured for 24 h with medium alone (green line) or containing lipopolysaccharide (LPS) (1 μg/mL, red line), poly-inosinic acid:cytidylic acid (Poly I:C; 1 μg/mL, orange line), or peptidoglycan (PGN); 10 μg/mL, blue line). PBLs proliferation was determined according to flow cytometry after 5 days in culture, as described in Materials and Methods. M1 region represents proliferating cells after day 4. One representative experiment of three is shown. (B) Highly purified CD4+ or CD8+ CFSE-labeled T cells were stimulated and cultured in the absence or presence of 4 × 10⁴ hASCs pre-cultured for 24 h with medium alone or containing LPS (1 μg/mL), Poly I:C (1 μg/mL), or PGN (10 μg/mL). Results are expressed as the percentage of proliferating lymphocytes gated after day 4 of culture (M1 region). Means and standard deviations of three different experiments are shown.

Discussion

Because TLR signalling has been associated with the perpetuation of chronic inflammatory and autoimmune diseases,^46–49 the importance of better understanding the potential effect of TLR ligands on hASCs function is highlighted because of the therapeutic use of hASCs in the settings of such diseases. We have successfully conducted a phase II clinical trial for treatment with autologous hASCs (Cx401) for complex perianal fistula that included patients suffering from Crohn's disease,^32,33 and a phase III clinical trial is currently ongoing. Because of the nature of the gut, TLR ligands are highly present in the healthy gut and particularly in the gut of patients suffering from Crohn's disease. Therefore, hASCs employed in the treatment of fistula are probably exposed to TLR ligands, which may result in the modulation of hASC activity and therapeutic potency, including their proliferation, immunogenicity,and immunosuppression. While conducting our research, we found that a few studies have reported that MSCs express active TLRs.^28–31 However, they did not investigate the role of TLRs in the immunosuppressive and immunogenic features of hASCs.

In the present study, we report that hASCs express mRNA for TLR1, 2, 3, 4, 5, 6, 7, 9, and 10, similar to THP-1 cells, a monocytic cell line used as a control for TLR expression (Fig. 1A). Only TLR8 mRNA was absent in our studies. These findings are in agreement with the results recently reported showing high mRNA expression of TLR1 to 6 in adipose and bone marrow MSCs.^27–30

We demonstrated that hASCs possess active and functional TLR2, TLR3, and TLR4, because activation with PGN (TLR2), poly I:C (TLR3), and LPS (TLR4) triggered downstream signalling events, leading to IκB-α degradation and the induction of NF-kB–dependent genes and cytokines (MnSOD, COX-2, IL-6, and IL-8). MnSOD is a well-established TLR downstream gene with a key protective role against oxidative stress in the mitochondria. LPS and poly I:C treatment of hASCs led to the induction of MnSOD, whereas PGN treatment triggered a moderate induction of the gene (Fig. 2A). In the settings of an inflammatory response, immune cells release vast amounts of reactive oxygen species, which results in the generation of an oxidative milieu. In this regard, it has been reported that induction of MnSOD protects cells from oxidative stress, leading to greater survival.⁵⁰ Therefore, it is tempting to speculate that greater expression of MnSOD by hASCs in response to TLR ligand exposure would provide them with better engraftment or survival at injured or inflamed sites, leading to better therapeutic effects.

Cytokine release upon TLR triggering has been determined using ELISA and cytometric bead arrays (Fig. 2B, C). Under our experimental conditions, hASCs constitutively released significant amounts of IL-6 and IL-8. TLR2, TLR3, and TLR4 activation led to the strong up-regulation of IL-6 and IL-8 secretion but failed to induce other known TLR-regulated cytokines such as IL-1β, IL-2, IL-4, IL-5, IL-10, IL-12, TNFα, or IFNγ that immune cells induce highly upon TLR ligation. The significance of a constitutive expression of IL-6 and IL-8, which are considered pro-inflammatory cytokines, in the context of the immunosuppressive activity of hASCs is unclear. It has been recently reported that MSCs inhibit the differentiation of dendritic cells, at least in part, through the release of IL-6.⁵¹ This observation links IL-6 production to the immunosuppression mediated by MSCs. Hence, it is tempting to speculate that induction of IL-6 secretion by TLR activation may enhance hASC-mediated impairment of dendritic cell differentiation and maturation.

To determine the effects of TLR activation on hASC phenotype, we tested the ability of PGN, poly I:C, and LPS to modulate hASC proliferation, differentiation, and immunogenicity. We found no effect on adipogenic differentiation but detected that poly I:C and LPS significantly increased osteogenic differentiation in hASCs (Fig. 3A). The effect of TLR activation on hASC and MSC differentiation remains controversial, with different authors recently reporting contradictory results. Liotta et al.³¹ found no effect of TLR activation on adipogenic, osteogenic, or chondrogenic differentiation of human bone marrow–derived MSCs. However, Pevsner-Fischer et al.²⁹ reported that TLR2 activation by Pam3Cys reduced mouse bone marrow–derived MSC differentiation into the three mesodermal lineages. Furthermore, Hwa Cho et al.²⁸ reported greater osteogenic differentiation of hASC by LPS and PGN activation and less adipogenic differentiation when PGN was present. Differences between bone marrow and adipose-derived mesenchymal stem cells and between mouse and human cells might explain these discrepancies.

The potential use of allogeneic hASCs and MSCs relies on the capacity of these cells to escape to the immune recognition because they display low levels of HLA-I and no expression of HLA-II and costimulatory molecules.⁵² In this study, for the first time, we provide evidence that activation of hASCs through TLRs does not increase hASC immunogenic features, indicated by a lack of expression of several costimulatory molecules. Furthermore, only poly I:C activation led to greater expression of HLA-I (Fig. 3B). These results may have relevance in terms of allogeneic hASC-based cell therapy.

The mechanisms underlying the immunosuppression potential of hASCs are not fully understood but seem to require the release of soluble factors in response to immune cells. Soluble factors such as TGF-β1, IDO, nitric oxide, and IL-10 have been shown to play a role.^19–26 Our study shows that hASC activation by LPS, poly I:C, and PGN does not lead to the production of IL-10 (Fig. 2C), TGF-β1 (data not shown), or nitric oxide (Fig. 4A). Moreover, IDO induction was restricted to poly I:C at the highest concentration (10 μg/mL), which yielded a more-moderate kynurenine production than IFNγ (Fig. 4B). Based on these results, we hypothesized that TLR activation would not impair the suppressive effect of hASCs on PBL proliferation. Proliferation assays with PBLs or CD4+ or CD8+ T cells cocultured with hASCs in the presence of TLR ligands, in which no effect on hASC-mediated immunosuppression was observed further confirmed this hypothesis (Fig. 5).

Our observations, although in agreement with those reported by Pevsner-Fisher et al.²⁹ which showed that TLR2 activation by Pam3Cys does not inhibit mouse MSC-mediated immunosuppression, are somewhat different from those by Liotta et al.³¹ showing that TLR3 and TLR4 ligation on human MSCs may regulate their suppressive activity. The fact that bone marrow MSCs and hASCs are different, although they share numerous similar properties, might explain these differences. Differences in the experimental conditions during the assay may also play a role. In fact, Liotta and colleagues employed purified CD4+ T cells stimulated with allogeneic T cell–depleted PBLs and anti-CD3 mAb, whereas we used whole PBL samples or purified CD4+ and CD8+ fractions (in the absence of any additional allogeneic PBLs) stimulated with beads loaded with anti-CD3, anti-CD2, and anti-CD28 mAbs. Further investigations will be needed to clarify the findings from different authors.

Finally, our demonstration that TLR2, TLR3, and TLR4 ligation does not impair the potential of hASCs to suppress immune responses or to escape the recipient immune surveillance has implications for the safety and efficacy of allogeneic hASC-based cell therapies for inflammatory diseases in which hASCs may eventually be exposed to TLR ligands, such as inflammatory bowel disease (in particular, patients with complex perianal fistula).

Footnotes

Acknowledgments

The authors would like to thank Dr. Sonsoles Hortelano (Centro Nacional de Investigaciones Cardiovasculares, Madrid, Spain) for generously providing reagents for nitrite detection. We would also like to thank Eduardo Suarez for the critical reading of the manuscript. This work was supported by the Ministerio de Industria, Turismo y Comercio of Spain, and by the Ramón y Cajal program from the Ministerio de Educación y Ciencia of Spain (to E.L.).

Disclosure Statement

Authors are employees of Cellerix. A potential conflict of interest may exist. Data are original and unbiased.

References

Akira

, Uematsu

, Takeuchi

Pathogen recognition and innate immunity. Cell, 124:783. 2006.

Ozinsky

, Underhill

D.M.

, Fontenot

J.D.

, Hajjar

A.M.

, Smith

K.D.

, Wilson

C.B.

, Schroeder

, Aderem

The repertoire for pattern recognition of pathogens by the innate immune system is defined by cooperation between toll-like receptors. Proc Natl Acad Sci U S A, 97:13766. 2000.

Pasare

, Medzhitov

Toll-like receptors: linking innate and adaptive immunity. Adv Exp Med Biol, 560:11. 2005.

Hoebe

, Jiang

, Georgel

, Tabeta

, Janssen

, Du

, Beutler

TLR signaling pathways: opportunities for activation and blockade in pursuit of therapy. Curr Pharm Des, 12:4123. 2006.

Hawai

, Akira

TLR signaling. Semin Immunol, 19:24. 2007.

O'Neill

L.A.

, Bowie

A.G.

The family of five: TIR-domain-containing adaptors in Toll-like receptor signalling. Nat Rev Immunol, 7:353. 2007.

Ohashi

, Burkart

, Flohé

, Kolb

Cutting edge: heat shock protein 60 is a putative endogenous ligand of the toll-like receptor-4 complex. J Immunol, 164:558. 2000.

Asea

, Kraeft

S.K.

, Kurt-Jones

E.A.

, Stevenson

M.A.

, Chen

L.B.

, Finberg

R.W.

, Koo

G.C.

, Calderwood

S.K.

HSP70 stimulates cytokine production through a CD14-dependant pathway, demonstrating its dual role as a chaperone and cytokine. Nat Med, 6:435. 2000.

Barrat

F.J.

, Meeker

, Gregorio

, Chan

J.H.

, Uematsu

, Akira

, Chang

, Duramad

, Coffman

R.L.

Nucleic acids of mammalian origin can act as endogenous ligands for Toll-like receptors and may promote systemic lupus erythematosus. J Exp Med, 202:1131. 2005.

10.

Termeer

, Benedix

, Sleeman

, Fieber

, Voith

, Ahrens

, Miyake

, Freudenberg

, Galanos

, Simon

J.C.

Oligosaccharides of hyaluronan activate dendritic cells via Toll-like receptor 4. J Exp Med, 95:99. 2002.

11.

Beutler

Neo-ligands for innate immune receptors and the etiology of sterile inflammatory disease. Immunol Rev, 220:113. 2007.

12.

Friedenstein

A.J.

, Gorskaja

J.F.

, Kulagina

N.N.

Fibroblast precursors in normal and irradiated mouse hematopoietic organs. Exp Hematol, 4:267. 1976.

13.

Pittenger

M.F.

, Mackay

A.M.

, Beck

S.C.

, Jaiswal

R.K.

, Douglas

, Mosca

D, Moorman

M.A.

, Simonetti

D.W.

, Craig

, Marshak

D.R.

Multilineage potential of adult human mesenchymal stem cells. Science, 284:143. 1999.

14.

Jiang

, Jahagirdar

B.N.

, Reinhardt

R.L.

, Schwartz

R.E.

, Keene

C.D.

, Ortiz-Gonzalez

X.R.

, Reyes

, Lenvik

, Lund

, Blackstad

, Du

, Aldrich

, Lisberg

, Low

W.C.

, Largaespada

D.A.

, Verfaillie

C.M.

Pluripotency of mesenchymal stem cells derived from adult marrow. Nature, 418:41. 2002.

15.

Zuk

P.A.

, Zhu

, Ashjian

, De Ugarte

D.A.

, Huang

J.I.

, Mizuno

, Alfonso

Z.C.

, Fraser

J.K.

, Benhaim

, Hedrick

M.H.

Human adipose tissue is a source of multipotent stem cells. Mol Biol Cell, 13:4279. 2002.

16.

Stagg

Immune regulation by mesenchymal stem cells: two sides to the coin. Tissue Antigens, 69:1. 2007.

17.

Puissant

, Barreau

, Bourin

, Clavel

, Corre

, Bousquet

, Taureau

, Cousin

, Abbal

, Laharrague

, Penicaud

, Casteilla

, Blancher

Immunomodulatory effect of human adipose tissue-derived adult stem cells: comparison with bone marrow mesenchymal stem cells. Br J Haematol, 129:118. 2005.

18.

Yañez

, Lamana

M.L.

, García-Castro

, Colmenero

, Ramírez

, Bueren

J.A.

Adipose tissue-derived mesenchymal stem cells have in vivo immunosuppressive properties applicable for the control of the graft-versus-host disease. Stem Cells, 24:2582. 2006.

19.

Tse

W.T.

, Pendleton

J.D.

, Beyer

W.M.

, Egalka

M.C.

, Guinan

E.C.

Suppression of allogeneic T-cell proliferation by human marrow stromal cells: implications in transplantation. Transplantation, 75:389. 2003.

20.

Djouad

, Plence

, Bony

, Tropel

, Apparailly

, Sany

, Noël

, Jorgensen

Immunosuppressive effect of mesenchymal stem cells favors tumor growth in allogeneic animals. Blood, 102:3837. 2003.

21.

Aggarwal

, Pittenger

M.F.

Human mesenchymal stem cells modulate allogeneic immune cell responses. Blood, 105:1815. 2005.

22.

Beyth

, Borovsky

, Mevorach

, Liebergall

, Gazit

, Aslan

, Galun

, Rachmilewitz

Human mesenchymal stem cells alter antigen-presenting cell maturation and induce T-cell unresponsiveness. Blood, 105:2214. 2005.

23.

Sato

, Ozaki

, Oh

, Meguro

, Hatanaka

, Nagai

, Muroi

, Ozawa

Nitric oxide plays a critical role in suppression of T-cell proliferation by mesenchymal stem cells. Blood, 109:228. 2007.

24.

Krampera

, Glennie

, Dyson

, Scott

, Laylor

, Simpson

, Dazzi

Bone marrow mesenchymal stem cells inhibit the response of naive and memory antigen-specific T cells to their cognate peptide. Blood, 101:3722. 2003.

25.

Augello

, Tasso

, Negrini

S.M.

, Amateis

, Indiveri

, Cancedda

, Pennesi

Bone marrow mesenchymal progenitor cells inhibit lymphocyte proliferation by activation of the programmed death 1 pathway. Eur J Immunol, 35:1482. 2005.

26.

Cui

, Yin

, Liu

, Li

, Zhang

, Cao

Expanded adipose-derived stem cells suppress mixed lymphocyte reaction by secretion of prostaglandin E2. Tissue Eng, 13:1185. 2007.

27.

Chamberlain

, Fox

, Ashton

, Middleton

Concise review: mesenchymal stem cells: their phenotype, differentiation capacity, immunological features, and potential for homing. Stem Cells, 25:2739. 2007.

28.

Hwa Cho

, Bae

Y.C.

, Jung

J.S.

Role of toll-like receptors on human adipose-derived stromal cells. Stem Cells, 24:2744. 2006.

29.

Pevsner-Fischer

, Morad

, Cohen-Sfady

, Rousso-Noori

, Zanin-Zhorov

, Cohen

I.R.

, Zipori

Toll-like receptors and their ligands control mesenchymal stem cell functions. Blood, 109:1422. 2007.

30.

Tomchuck

S.L.

, Zwezdaryk

K.J.

, Coffelt

S.B.

, Waterman

R.S.

, Danka

E.S.

, Scandurro

A.B.

Toll-like receptors on human mesenchymal stem cells drive their migration and immunomodulating responses. Stem Cells, 26:99. 2008.

31.

Liotta

, Angeli

, Cosmi

, Filì

, Manuelli

, Frosali

, Mazzinghi

, Maggi

, Pasini

, Lisi

, Santarlasci

, Consoloni

, Angelotti

M.L.

, Romagnani

, Parronchi

, Krampera

, Maggi

, Romagnani

, Annunziato

Toll-like receptors 3 and 4 are expressed by human bone marrow-derived mesenchymal stem cells and can inhibit their T-cell modulatory activity by impairing Notch signaling. Stem Cells, 26:279. 2008.

32.

García-Olmo

, García-Arranz

, Herreros

, Pascual

, Peiro

, Rodríguez-Montes

J.A.

A phase I clinical trial of the treatment of Crohn's fistula by adipose mesenchymal stem cell transplantation. Dis Colon Rectum, 48:1416. 2005.

33.

Garcia-Olmo

, Herreros

, Pascual

J. A.

, Del-Valle

, Zorrilla

, De-La-Quintana

, Garcia-Arranz

, Beraza

, Alemany

, Fernandez

, Portero

, Büscher

, Pascual

Expanded Adipose-Derived Stem Cells (Cx401) for the Treatment of Complex Perianal Fistula; a Randomized, Controlled Clinical TrialIn press.

34.

Yamamoto-Furusho

J.K.

, Podolsky

D.K.

Innate immunity in inflammatory bowel disease. World J Gastroenterol, 13:5577. 2007.

35.

Ishihara

, Rumi

M.A.

, Ortega-Cava

C.F.

, Kazumori

, Kadowaki

, Ishimura

, Kinoshita

Therapeutic targeting of toll-like receptors in gastrointestinal inflammation. Curr Pharm Des, 12:4215. 2006.

36.

Tsan

M.F.

, Clark

R.N.

, Goyert

S.M.

, White

J.E.

Induction of TNF-alpha and MnSOD by endotoxin: role of membrane CD14 and Toll-like receptor-4. Am J Physiol Cell Physiol, 280:1422. 2001.

37.

Crofford

L.J.

, Tan

, McCarthy

C.J.

, Hla

Involvement of nuclear factor kappa B in the regulation of cyclooxygenase-2 expression by interleukin-1 in rheumatoid synoviocytes. Arthritis Rheum, 40:226. 1997.

38.

Matsusaka

, Fujikawa

, Nishio

, Mukaida

, Matsushima

, Kishimoto

, Akira

Transcription factors NF-IL6 and NF-kappa B synergistically activate transcription of the inflammatory cytokines, interleukin 6 and interleukin 8. Proc Natl Acad Sci U S A 90, 10193. 1993.

39.

Libermann, T,A., Baltimore

Activation of interleukin-6 gene expression through the NF-kappa B transcription factor. Mol Cell Biol, 10:2327. 1990.

40.

Mukaida

, Mahe

, Matsushima

Cooperative interaction of nuclear factor-kappa B- and cis-regulatory enhancer binding protein-like factor binding elements in activating the interleukin-8 gene by pro-inflammatory cytokines. J Biol Chem, 265:21128. 1990.

41.

Kunsch

, Rosen

C.A.

NF-kappa B subunit-specific regulation of the interleukin-8 promoter. Mol Cell Biol, 10:6137. 1993.

42.

Fehrer

, Brunauer

, Laschober

, Unterluggauer

, Reitinger

, Kloss

, Gülly

, Gassner

, Lepperdinger

Reduced oxygen tension attenuates differentiation capacity of human mesenchymal stem cells and prolongs their lifespan. Aging Cell, 6:745. 2007.

43.

Potier

, Ferreira

, Andriamanalijaona

, Pujol

J.P.

, Oudina

, Logeart-Avramoglou

, Petite

Hypoxia affects mesenchymal stromal cell osteogenic differentiation and angiogenic factor expression. Bone, 40:1078. 2007.

44.

Weinberg

J.B.

, Misukonis

M.A.

, Shami

P.J.

, Mason

S.N.

, Sauls

D.L.

, Dittman

W.A.

, Wood

E.R.

, Smith

G.K.

, McDonald

, Bachus

K.E.

et al. Human mononuclear phagocyte inducible nitric oxide synthase (iNOS): analysis of iNOS mRNA, iNOS protein, biopterin, and nitric oxide production by blood monocytes and peritoneal macrophages. Blood, 86:1184. 1995.

45.

Fujigaki

, Saito

, Sekikawa

, Tone

, Takikawa

, Fujii

, Wada

, Noma

, Seishima

Lipopolysaccharide induction of indoleamine 2,3-dioxygenase is mediated dominantly by an IFN-gamma-independent mechanism. Eur J Immunol, 31:2313. 2001.

46.

Marshak-Rothstein

, Rifkin

I.R.

Immunologically active autoantigens: the role of toll-like receptors in the development of chronic inflammatory disease. Annu Rev Immunol., 25:419. 2007.

47.

Kanai

, Ilyama

, Ishikura

, Uraushihara

, Totsuka

, Yamazaki

, Nakamuma

, Watanabe

Role of the innate immune system in the development of chronic colitis. J Gastroenterol, 37:38. 2002.

48.

Andreakos

, Sacre

, Foxwell

B.M.

, Feldmann

The toll-like receptor-nuclear factor kappaB pathway in rheumatoid arthritis. Front Biosci, 10:2478. 2005.

49.

Baccala

, Hoebe

, Kono

D.H.

, Beutler

, Theofilopoulos

A.N.

TLR-dependent and TLR-independent pathways of type I interferon induction in systemic autoimmunity. Nat Med, 13:543. 2007.

50.

Kops

G.J.

, Dansen

T.B.

, Polderman

P.E.

, Saarloos

, Wirtz

K.W.

, Coffer

P.J.

, Huang

T.T.

, Bos

J.L.

, Medema

R.H.

, Burgering

B.M.

Forkhead transcription factor FOXO3a protects quiescent cells from oxidative stress. Nature, 419:316. 2002.

51.

Djouad

, Charbonnier

L.M.

, Bouffi

, Louis-Plence

, Bony

, Apparailly

, Cantos

, Jorgensen

, Noël

Mesenchymal stem cells inhibit the differentiation of dendritic cells through an interleukin-6-dependent mechanism. Stem Cells, 25:2025. 2007.

52.

McIntosh

, Zvonic

, Garrett

, Mitchell

J.B.

, Floyd

Z.E.

, Hammill

, Kloster

, Di Halvorsen

, Ting

J.P.

, Storms

R.W.

, Goh

, Kilroy

, Wu

, Gimble

J.M.

The immunogenicity of human adipose-derived cells: temporal changes in vitro. Stem Cells, 24:1246. 2006.