Human Fetal Cartilage-Derived Progenitor Cells Exhibit Anti-Inflammatory Effect on IL-1β-Mediated Osteoarthritis Phenotypes In Vitro

Abstract

Background:

In this study, we have investigated whether human fetal cartilage progenitor cells (hFCPCs) have anti-inflammatory activity and can alleviate osteoarthritis (OA) phenotypes in vitro.

Methods:

hFCPCs were stimulated with various cytokines and their combinations and expression of paracrine factors was examined to find an optimal priming factor. Human chondrocytes or SW982 synoviocytes were treated with interleukin-1β (IL-1β) to produce OA phenotype, and co-cultured with polyinosinic-polycytidylic acid (poly(I-C))-primed hFCPCs to address their anti-inflammatory effect by measuring the expression of OA-related genes. The effect of poly(I-C) on the surface marker expression and differentiation of hFCPCs into 3 mesodermal lineages was also examined.

Results:

Among the priming factors tested, poly(I-C) (1 µg/mL) most significantly induced the expression of paracrine factors such as indoleamine 2,3-dioxygenase, histocompatibility antigen, class I, G, tumor necrosis factor- stimulated gene-6, leukemia inhibitory factor, transforming growth factor-β1 and hepatocyte growth factor from hFCPCs. In the OA model in vitro, co-treatment of poly(I-C)-primed hFCPCs significantly alleviated IL-1β-induced expression of inflammatory factors such as IL-6, monocyte chemoattractant protein-1 and IL-1β, and matrix metalloproteinases in SW982, while it increased the expression of cartilage extracellular matrix such as aggrecan and collagen type II in human chondrocytes. We also found that treatment of poly(I-C) did not cause significant changes in the surface marker profile of hFCPCs, while showed some changes in the 3 lineages differentiation.

Conclusion:

These results suggest that poly(I-C)-primed hFCPCs have an ability to modulate inflammatory response and OA phenotypes in vitro and encourage further studies to apply them in animal models of OA in the future.

Introduction

Stem or progenitor cells from fetal tissues can be obtained from various human tissues such as bone marrow, liver, blood, lung, brain, spinal cord, kidney, cartilage, heart, pancreas, placenta, umbilical cord blood, and Wharton’s jelly [1–3]. They are known to have better abilities than adult stem cells such as mesenchymal stem cells (MSCs) in cell proliferation and differentiation into dedicated cell types. They also have an advantage like MSCs of immune-privilege and immune-modulatory activities as well, therefore regarded as a promising source for a transplant [4, 5]. MSCs are well known for their immune-modulatory and anti-inflammatory effects and are under extensive investigation for their therapeutic potential to treat a variety of immune- and inflammation-related disorders [6, 7]. Therapeutic potential of MSCs is generally thought to depend on their paracrine effects of secreted growth factors, cytokines, and exosomes, etc. [6, 7]. Resting MSCs need to be ‘primed’ by certain cytokines or inflammatory factors to become active MSCs and have a high therapeutic activity [8, 9]. However, it is very challenging to find an optimal priming factor because the priming effect is highly variable depending on the characteristics of MSCs and target diseases.

We have previously shown that fetal cartilage-derived progenitor cells (hFCPCs) are a highly promising cell source for cartilage regeneration thanks to their high yield, superior proliferation, and differentiation abilities than adult MSCs and chondrocytes [10, 11]. We have also reported that hFCPCs have an immune-privilege and immune-modulatory activity like MSCs [12]. In the study, hFCPCs did not express human leukocyte antigen (HLA) class II (HLA-DR and HLA-DM) and co-stimulatory molecules (CD40, CD80 and CD86) and expressed anti-inflammatory factors such as transforming growth factor-β1 (TGF-β1) and tumor necrosis factor (TNF) inducible gene-6 (TSG-6). In addition, hFCPCs showed immune-privileged and immune-modulatory properties on allogeneic peripheral blood lymphocytes (PBLs) in the mixed lymphocyte reaction (MLR) and cytokines expression such as TNF-α, interferon-γ (IFN-γ) and interleukin-10 (IL-10) upon concanavalin A stimulation. These results are sufficient to suggest therapeutic potential of hFCPCs as a cell therapy to treat immune-related and inflammatory disorders. However, this report did not investigate any therapeutic potential or priming effect of hFCPCs and remained it to be addressed in the future studies.

Osteoarthritis (OA) is the most common form of degenerative joint disorders, mainly characterized by the degradation of articular cartilage tissue and associated with subchondral bone lesions [13]. Accumulating evidence indicates that a low-grade inflammation contributes to symptoms and progression of OA [13, 14]. It is known that resident synoviocytes, chondrocytes and immune cells increase the production of cytokines, chemokines, growth factors, adipokines, prostaglandins, leukotrienes, nitric oxide, and neuropeptides in response to local physiochemical changes in the OA joint. The failure in the proper control of inflammation can lead to severe inflammation, progressive cartilage loss, pain, and joint dysfunction. Current therapies for OA mainly focus on relieving symptoms or pain using analgesics and non-steroidal anti-inflammatory drugs (NSAIDs) [15, 16]. However, they suffer they have a critical limitation of causing high comorbidity due to their side effects [17]. Hyaluronic acids (HA) is also commonly used in many clinics but it is not officially recommended for OA treatment due to the lack of therapeutic evidence [18]. Recently cell therapy particularly using anti-inflammatory effect of MSCs is emerging as a novel therapeutic modality to treat OA [19–21]. There are many clinical trials as summarized well in several meta-analyses and one product [22, 23] has been approved for commercial use in Korea using MSCs from umbilical cord blood (Cartistem by Medipost) [24, 25]. However, cell therapy to treat OA is still at its early stage and needs further investigation to improve the technology in many aspects and optimize the therapeutic protocol [26].

In this study, we have screened for several priming factors and their combinations to enhance paracrine effects of hFCPCs, namely IL-1β, TNF-α, IFN-γ and poly(I-C), and found that poly(I-C) significantly enhanced expression of anti-inflammatory factors in hFCPCs. Poly(I-C) is a synthetic analog of double-stranded RNA (dsRNA) and activates toll-like receptor-3 (TLR-3) signal pathway, one of pattern recognition receptors of the innate immune system [8, 9]. We have also shown the therapeutic potential of poly(I-C)-primed hFCPCs on the IL-1β-induced OA phenotypes in human articular chondrocytes and SW-982 human synoviocyte cells. This study is a first report using non-MSCs cells to treat OA and can contribute to better understanding about cell therapies against OA.

Materials and methods

Cell culture

hFCPCs were obtained from fetal cartilage tissues of gestation age (GA) 12–14 with the approval of institutional review board (IRB) of Ajou University Medical Center (AJIRB-CRO-07-139) as previously described [10]. After thawing of a frozen stock, hFCPCs were cultured at 8 × 10³ cells/cm² in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco BRL, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT, USA), 100 U/mL penicillin G, and 100 μg/mL streptomycin (Pen-Strep; HyClone). Human chondrocytes were obtained from the knee cartilage tissue of a surgery patient (25 years old) with traumatic injury at Ajou University Hospital (AJIRB-MED-SMP-10–266) and cultured with a seeding density of 1 × 10⁴ cells/cm² in DMEM with 10% FBS as above. SW982, a human synovial sarcoma cell line, was purchased from ATCC and maintained in DMEM with 10% FBS.

Treatment of cells

For immunophenotype analysis, hFCPCs at passage 4 were untreated or stimulated with 100 ng/mL lipopolysaccharide (LPS; Sigma-Aldrich, St. Louis, MO, USA), or 200 U/mL human recombinant IFN-γ (Sigma-Aldrich) for 1 or 3 days. To prime the paracrine activity of hFCPCs, cells at passage 4 were treated with 10 ng/mL IL-1β, 500 U/mL TNF-α, 200 U/mL IFN-γ, 1 µg/mL poly(I-C) or their combinations as descried in figure legends (all from Sigma-Aldrich). To observe the effect of hFCPCs on the OA phenotype in vitro, human young chondrocytes or SW982 cells were cultured in the 6-well insert (Corning, Lowell, MA, USA) and treated with 10 ng/mL IL-1β for 3 days in the absence or presence of hFCPCs cultured in the bottom of 6-well Transwell plates and primed with poly(I-C) for 1 day in advance.

Reverse transcriptase-polymerase chain reaction (RT-PCR)

Total RNA was extracted using TRIzol reagent (Invitrogen, Waltham, MA, USA), followed by a cDNA synthesis with random hexamers according to the manufacturer’s instructions. RT-PCR was performed using The First Strand cDNA Synthesis Kit for RT-PCR (Roche, Mannheim, Germany) with specific primers (Table 1). Real-time quantitative PCR was performed using a CFX96 Touch Real-time PCR Detection System (Bio-Rad, Hercules, CA, USA) with SYBR Green Reaction Mix (Roche) with specific primers (Table 2). Data were normalized against the expression level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene.

Table 1.

Primer sequences used for RT-PCR

Gene	Primer sequence	Accession number	Size (bp)	Annealing temp
GAPDH	5′-ACAACTTTGGTATCGTGGAA-3′ 5′-AAATTCGTTGTCATACCAGG-3′	NM_002046.3	458	55
COX2	5′-ATGAGATTGTGGGAAAATTGCT-3′ 5′-GATCATCTCTGCCTGAGTATC-3′	NM_000963.4	299	56
IDO	5′-CGCTGTTGGAAATAGCTTC-3′ 5′-CAGGACGTCAAAGCACTGAA-3′	NM_002164.5	234	53
HGF	5′-ATGCATCCAAGGTCAAGGAG-3′ 5′-TTCCATGTTCTTTTGTCCCACA-3′	NM_001010932	249	55
HLA-G	5′-CCACCACCCTGTCTTTGACT-3′ 5′-TGGCACGTGTATCTCTGCTC-3′	NM_002127.5	210	63
TSG-6	5′-GGTGTGTACCACAGAGAAGCA-3′ 5′-GGGTTGTAGCAATAGGCATCC-3′	NM_007115.3	284	63
TGF-β	5′-ACCGGCCTTTCCTGCTTCTCA-3′ 5′-CGCCCGGGTTATGCTGGTTGT-3′	NM_000660.4	288	63
IL-6	5′-ATGAACTCCTTCTCCACAAGC-3′ 5′-CTACATTTGCCGAAGAGCCCTCAGGCTGGACTG-3′	NM_000600.3	639	55
IL-1β	5′-AAGCTGATGGCCCTAAACAG-3′ 5′-AGGTGCATCGTGCACATAAG-3′	NM_000576.2	281	53
TNF-α	5′-ATGAGCACTGAAAGCATGATC-3′ 5′-TCACAGGGCAATGATCCCAAAGTAGACCTGCCC-3′	NM_000594.3	702	51
HLA-ABC	5′-GTATTTCTTCACATCCGTGTCCCG-3′ 5′-GTCCGCCGCGGTCCAAGAGCGCAG-3′	XM_005275331	346	37
HLA-DR	5′-CTGATGAGCGCTCAGGAATCATTG-3′ 5′-TGCATTGGCCAACATAGCTG-3′	NM_001242758	220	60
HLA-DM	5′-CCAGCCCAATGGAGACTG-3′ 5′-CAGCCCAGGTGTCCAGTC-3′	NM002118.4	136	57
Beta 2 M	5′-GTGGAGCATTCAGACTTGTC-3′ 5′-AACAAGCTTTGAGTGCAAGAG-3′	NM_004048.2	479	57
CD80	5′-ACTCGCATCTACTGGCAAAAGGA-3′ 5′-ATGGGAGCAGGTTATCAGGAAAA-3′	NM_005191.3	553	59
CD86	5′-GTATTTTGGCAGGACCAGGA-3′ 5′-GCCGCTTCTTCTTCTTCCAT-3′	NM_006889. 4	664	57
CD40	5′-AGAAGGCTGGCACTGTACGA-3′ 5′-CAGTGTTGGAGCCAGGAAGA-3′	NM_152854.2	363	59
CD59	5′-ATGGGAATCCAAGGAGGGT-3′ 5′-ATGAAGGCTCCAGGCTGCT-3′	NM_001127223.1	381	59
MCP1	5′-CTCGCTCAGCCAGATGCAATCAAT-3′ 5′-CCCAGGGGTAGAACTGTGGTTCAA-3′	NM_002982.3	480	66
TRL3	5′-GGACTTTGAGGCGGGTGTT-3′ 5′-TGTTGAACTGCATGATGTACCTTGA-3′	NM_003265	141	60
TRL4	5′-TTGAGCAGGTCTAGGGTGATTGAAC-3′ 5′-ATGCGGGACACACACACTTTCAAATA-3′	NM_138554	143	60

Table 2.

Primer sequences used for qRT-PCR

Gene	Primer sequence	Accession number	Size (bp)	Annealing temp
GAPDH	5′-ATGGGGAAGGTGAAGGTCG-3′ 5′-TAAAAGCAGCCCTGGTGACC-3′	NM_002046	70	60
IDO	5′-GCCTGATCTCATAGAGTCTGGC -3′ 5′-TGCATCCCAGAACTAGACGTGC-3′	NM_002164.5	118	60
LIF	5′-CCTGGACAAGCTATGTGG-3′ 5′-GGTTGAGGATCTTCTGGTC-3′	XM_006724240.1	154	60
TGF-β	5′-CAGCAACAATTCCTGGCGATAC-3′ 5′-GCTAAGGCGAAAGCCCTCAAT-3′	NM_000660.4	139	60
TSG-6	5′-CATCTCGCAACTTACAAGC-3′ 5′-AGACGGATTCCATAATCAATAATG-3′	NM_007115.3	160	60
HGF	5′-TACGCTACGAAGTCTGTG-3′ 5′-TCTTGCCTGATTCTGTATGA-3′	XM_006715956.1	104	60
HLA-G	5′-CCACCACCCTGTCTTTGACT-3′ 5′-TGGCACGTGTATCTCTGCTC-3′	NM_002127.5	210	63
MMP1	5′-GGCTGAAAGTGACTGGGAAACC-3′ 5′-TGCTCTTGGCAAATCTGGCGTG-3′	NM_001145938.1	127	60
MMP13	5′-AGGTAGCGCTCTGCAAACTGG-3′ 5′-AGCTGGACTCATTGTCGGGC-3′	NM_002427.3	92	60
MCP1	5′-TCTGTGCCTGCTGCTCATAG-3′ 5′-GTGACTGGGGCATTGATTG-3′	NM_002982	75	60
ADAMTS4	5′-CAAGGTCCCATGTGCAACGT-3′ 5′-CATCTGCCACCACCAGTGTCT-3′	NM_005099.4	70	60
COL2	5′-GGCAATAGCAGGTTCACGTACA-3′ 5′-CGATAACAGTCTTGCCCACTT-3′	XM_011537935.1	40	60
Aggrecan	5′-CGATAACAGTCTTGCCCCACTT-3′ 5′-TCGAGGACAGCGAGGCC-3′	NM_013227	85	60
IL-6	5′-AGACAGCCACTCACCTCTTCAG-3′ 5′-TTCTGCCAGTGCCTCTTTGCTG-3′	NM_000600	132	60
IL-1β	5′-CCACAGACCTTCCAGGAGAATG-3′ 5′-GTGCAGTTCAGTGATCGTACAGG-3′	NM_000576	131	60

Flow cytometry

Flow cytometry analysis was performed to investigate the surface marker profile of hFCPCs. Briefly, after harvesting cells by enzymatic digestion, cells were centrifuged and washed. Aliquots of 5 × 10⁵ cells were then incubated with monoclonal antibodies conjugated to fluorochromes for 30 min on ice. Fluorescein isothiocyanate (FITC) or phycoerythrin (PE)-conjugated mouse monoclonal antibodies against human HLA-ABC, HLA-DR, CD40, CD80, CD86, CD44, CD54, CD73, CD90 and CD166 were used as recommended by the manufacturer (all from BD Biosciences, San Diego, CA, USA). For the detection of TLR-3 and TLR-4, PE-conjugated mouse monoclonal antibodies and their isotype control antibody were used (Invitrogen). The information of antibodies used is provided in Table 3. Data were obtained by analyzing 10,000 events using BD FACScan flow cytometer (Becton Dickinson, San Jose, CA, USA).

Table 3.

Antibodies used for flow cytometry

Gene	Company	Catalog number	Labeling	Dilution factor
HLA-ABC	BD Bioscience	555552	FITC	20:1
HLA-DR		347363	FITC	20:1
CD40		560963	PE	20:1
CD80		560925	PE	20:1
CD86		560957	PE	20:1
CD44		550989	PE	20:1
CD54		560971	PE	20:1
CD73		550257	PE	20:1
CD90		555595	FITC	2:1
CD166		559263	PE	20:1
Isotype IgG1		555748	FITC	20:1
Isotype IgG1		555749	PE	20:1
TLR3 (CD283)	Invitrogen	12-9039-80	PE	40:1
TLR4 (CD284)		12-9917-42	PE	40:1
Isotype IgG1		12-4714-82	PE	40:1
Isotype IgG2		12-4724-82	PE	40:1

Multipotent differentiation

For adipogenesis, hFCPCs at passage 4 were cultured on 12 well plates at 5 × 10⁴ cells/well. One day after plating, cells were cultured in adipogenic differentiation medium containing minimum essential medium with alpha modification (α-MEM; Gibco-BRL), 10% FBS, 0.5 μM dexamethasone (Sigma-Aldrich), 0.5 mM isobutyl-methylxanthine (Sigma-Aldrich), 10 μg/mL insulin and 50 μM indomethacin (Sigma-Aldrich). Cells were cultured for 21 days with the medium changed every 3 or 4 days. After differentiation, cells were stained with Alizarin red S (Sigma-Aldrich) to observe lipid droplets. Adsorbed dyes were eluted with 0.5 mL isopropyl alcohol and the intensity was measured at 500 nm for quantification. For osteogenesis, hFCPCs at passage 4 were cultured on 12 well plates at 2.5 × 10⁴ cells/well. One day after plating, cells were cultured in osteogenic differentiation medium containing α-MEM, 10% FBS, 10 mM β-glycerophosphate (Sigma-Aldrich), 10 nM dexamethasone and 0.1 mM L-ascorbic acid-2-phosphate (Sigma-Aldrich). Cells were cultured for 21 days with the medium changed every 3 or 4 days. After differentiation, cells were stained with Alizarin red S (Sigma-Aldrich) to observe mineralization. Adsorbed dyes were eluted with 0.5 mL mixture of 10% acetic acid and 10% ammonium hydroxide, and the intensity was measured at 405 nm for quantification. For chondrogenesis, hFCPCs of 3 × 10⁵ cells were centrifuged at 500 g for 10 min in a 15 mL conical tubes to form spherical pellets. The pellets were cultured for 21 days in chondrogenic differentiation media containing high-glucose DMEM, 10 ng/mL TGF-β3 (R&D Systems, Minneapolis, MN, USA), 10 µM dexamethasone, 50 μg/mL L-ascorbic acid-2-phosphate, 40 μg/mL proline (Sigma-Aldrich), 100 μg/mL pyruvate (Sigma-Aldrich), and 50 mg/mL insulin-transferrin-selenium (ITS) mixture (Becton Dickinson). For microscopy, the pellets were embedded in paraffin, cut into 4-μm-thick sections, and stained with 0.1% safranin-O (Sigma-Aldrich). The diameter of pellets was measured for quantification.

Statistical analysis

All data are expressed as mean ± standard deviation (SD) from 3 independent experiments except the case indicated specifically. Statistical differences among experimental groups were determined using one-way analysis of variance (ANOVA) with Tukey–Kramer multiple comparisons test. *^/#p < 0.05 and **^/##p < 0.01.

Results

Immuno-phenotype of hFCPCs did not change by IFN-γ or LPS treatment

Immunophenotypic characteristics of hFCPCs were first examined after stimulating cells with IFN-γ or LPS for 3 days. Then, the expression of selected HLA class I and II and complement molecules was examined by RT-PCR and flow cytometry. In flow cytometry, untreated hFCPCs showed high level expression (99.0 ± 3.7%) of HLA-ABC but barely expressed HLA-DR (0.7 ± 0.1%), CD80 (0.5 ± 0.1%) and CD86 (0.6 ± 0.1%) (Fig. 1A). Treatment of IFN-γ or LPS did not change their expression profiles. RT-PCR analysis showed that hFCPCs did not express inflammatory cytokines of TNF-α, IL-1β, IL-6 and macrophage chemoattractant protein-1 (MCP-1) but expressed anti-inflammatory factors of TGF-β and TSG-6 to some extent (Fig. 1B). Treatment of hFCPCs with LPS showed no changes in the expression of all inflammatory cytokines and TGF-β but increased the expression of TSG-6. These results suggest that, like the case of MSCs, hFCPCs are immunologically inert or privileged and do not induce inflammatory pathways upon stimulating signals like IFN-γ and LPS.

Fig. 1.

Immune phenotypes of hFCPCs. A hFCPCs were treated with 100 ng/mL LPS for 1 or 3 days and the expression of pro-inflammatory genes (TNF-α, IL-1β, IL-6 and MCP-1) or anti-inflammatory genes (TGF-β and TSG-6) was examined by RT-PCR. GAPDH was used as an internal control. B hFCPCs were untreated or treated with 200 U/mL IFN-γ or 100 ng/mL LPS for 3 days. The expression of immune-related genes (HLA-ABC, HLA-DR, CD80 and CD86) was examined in hFCPCs by flow cytometry. The percentages of positive cells are indicated by means with SD from 3 independent experiments (n = 3)

Expression of TLR-3 and TLR-4 in hFCPCs

Since hFCPCs seemed not to respond to LPS, we examined the expression of TLR-4 in comparison with that of TLR-3 upon treatment of poly(I-C) and LPS for 24 h. In RT-PCR analysis, both TLR-3 and TLR-4 mRNAs were observed in hFCPCs and their levels appeared to increase slightly by poly(I-C) and LPS treatment, respectively (Fig. 2A). Flow cytometry analysis also showed that TLR-3 and TLR-4 proteins were expressed in hFCPCs and treatment of poly(I-C) and LPS did not increase their expression (Fig. 2B).

Fig. 2.

Expression of TLR-3 and TLR-4 in hFCPCs. hFCPCs were untreated or treated with 1 µg/mL poly(I-C) or 100 ng/mL LPS for 24 h. A mRNA level of TLR-3 or TLR-4 was examined by RT-PCR using specific primers. GAPDH was used as an internal control. B Expression of TLR-3 or TLR-4 proteins was examined by flow cytometry using specific antibodies conjugated with PE. Isotype antibodies were used as a control

Treatment of poly(I-C) increased the expression of paracrine factors in hFCPCs

We tried to find good priming factors that can increase anti-inflammatory activity of hFCPCs. Cells were treated with 4 factors of IL-1β, TNF-α, IFN-γ and poly(I-C) alone or in combinations for 24 h and the expression of 6 selected anti-inflammatory factors were examined by qRT-PCR (Fig. 3). In the result, treatment of IL-1β, TNF-α or IFN-γ showed increase in the expression of 2 ~ 3 genes only among the 6 selected genes, namely indoleamine 2,3-dioxygenase (IDO), TSG-6 and leukemia inhibitory factor (LIF) by IL-1β, LIF and TGF-β by TNF-α, and TSG-6 and LIF by IFN-γ. In contrast, treatment of poly(I-C) showed increase in the expression of all 6 genes and particularly high-level increase was observed with IDO, HLA-G and TSG-6. Combined treatment of IL-1β, TNF-α and IFN-γ (I + T + IF) showed no additive or synergistic effect at all, while combination of I + T + IF and poly(I-C) showed additive effects on the expression of all 6 genes. These results suggest that poly(I-C) is an efficient priming factor to increase anti-inflammatory paracrine effect of hFCPCs.

Fig. 3.

The expression of anti-inflammatory genes in hFCPCs primed with various stimulations. hFCPCs were untreated or treated with 10 ng/mL IL-1β, 500 U/mL TNF-α, 200 U/mL IFN-γ, 1 µg/mL poly(I-C) or their combinations as indicated for 24 h. mRNA levels of IDO, HLA-G, TSG-6, LIF, TGF-β and HGF were determined by qRT-PCR. The relative gene expression levels to that of untreated control groups were presented by fold increase. Data are presented by mean values with SD from 3 independent experiments (n = 3). *p < 0.05 and **p < 0.01. I, IL-1β; T, TNF-α; IF, IFN-γ; and P, poly(I-C)

We have subsequently examined the expression of anti-inflammatory of IDO, hepatocyte growth factor (HGF), HLA-G, TSG-6 and TGF-β, and pro-inflammatory factors of IL-6 and cyclooxygenase (COX-2) in hFCPCs along with time after poly(I-C) treatment by RT-PCR analysis. When cells were treated with poly(I-C) for 3 days, the expression of these paracrine factors was increased significantly from day 1 and maintained at similar levels until day 3 (Fig. 4A). When cells were treated with poly(I-C) for 1 day and cultured in fresh medium without poly(I-C) for 2 additional days, the increased expression of all paracrine factors was also maintained at similar levels without a decrease (Fig. 4B).

Fig. 4.

Characterization of gene expression patterns in hFCPCs stimulated by poly(I-C). A hFCPCs were treated with 1 µg/mL poly(I-C) for 1, 2 and 3 days, and the expression of inflammation related genes was examined by RT-PCR as indicated. B hFCPCs were treated with 1 µg/mL poly(I-C) for 1 day and cultured for 2 more days without poly(I-C). The expression of the indicated genes was examined at each time point by RT-PCR

Poly(I-C)-primed hFCPCs inhibited IL-1β -mediated OA phenotypes of SW-982 cells and human young chondrocytes

We examined the therapeutic effect of poly(I-C)-primed hFCPCs on OA phenotypes in vitro. hFCPCs were treated with poly(I-C) for 24 h and co-cultured with SW-982 synoviocytes or human young chondrocytes in the presence of IL-1β for another 24 h. In SW-982 cells, treatment of IL-1β alone increased the expression of metalloproteinases (MMPs) like MMP-1, MMP-13 and a disintegrin and metalloproteinase with thrombospondin motifs 4 (ADAMTS4) and inflammatory cytokines of MCP-1, IL-6 and IL-1β), which were inhibited significantly by co-culture of poly(I-C)-primed hFCPCs except MMP-13 that showed no statistically significant difference (Fig. 5). In human young chondrocytes, treatment of IL-1β also increased the expression of metalloproteinases (MMP-1 and MMP-13) and inflammatory cytokines (MCP-1 and IL-6), while did not change the expression of cartilage matrix proteins of aggrecan and collagen type II (Fig. 6). Co-culture of poly(I-C)-primed hFCPCs inhibited the increase of MMP-1 and MCP-1, while showed no significant inhibitory effect on that of MMP-13, ADAMTS4 and IL-6. The expression of ADAMTS4 showed no statistical difference in all experimental groups (Fig. 6A). Unexpectedly co-culture of poly(I-C)-primed hFCPCs increased the expression of aggrecan in human young chondrocytes (Fig. 6B).

Fig. 5.

A, B Effect of poly(I-C)-primed hFCPCs on the IL-1β-induced inflammatory response in SW982 cells. SW982 cells were treated with 10 ng/mL IL-1β to induce OA phenotype in vitro. hFCPCs were primed with 1 µg/mL poly(I-C) for 1 day, and co-cultured with IL-1β -treated SW982 cells for 3 days in a 6-well Transwell plate in the co-treatment group (IL-1β  + hFCPCs-p). The expression of matrix metalloproteinases (MMP-1, MMP-13 and ADAMTS4) (A) and inflammatory cytokines (MCP-1, IL-6 and IL-1β) (B) by qRT-PCR. Data are presented by means with SDs from 3 independent experiments (n = 3). #p < 0.05 and ##p < 0.01 between the control and IL-1β groups. *p < 0.05 and **p < 0.01 between the IL-1β group and IL-1β  + hFCPCs-p groups

Fig. 6.

Effect of poly(I-C)-primed hFCPCs on the IL-1β-induced inflammatory response in human chondrocytes. Human chondrocytes were treated with 10 ng/mL IL-1β to induce OA phenotypes in vitro. hFCPCs primed with poly(I-C) for 1 day in advance was co-cultured with IL-1β-treated SW982 cells for 3 days in the co-treatment group (IL-1β + hFCPCs-p). The expression of A matrix metalloproteinases (MMP-1, MMP-13 and ADAMTS4), B cartilage ECM proteins (aggrecan and collagen type II), and C inflammatory cytokines (MCP-1 and IL-6) was examined by qRT-PCR. Data are presented by mean with SD from 3 independent experiments (n = 3). ##p < 0.01 between the control and IL-1β groups. *p < 0.05 and **p < 0.01 between the IL-1β group and IL-1β + hFCPCs-p groups

Treatment of poly (I-C) did not affect surface marker profile of hFCPCs

We then examined if poly(I-C) priming affects immunophenotype and surface markers expression in hFCPCs. In RT-PCR analysis, hFCPCs highly expressed HLA-ABC, beta2M and CD59 genes, while showed no expression of HLA-DM, HLA-DR, CD40, CD80 and CD86 at all regardless of poly(I-C) treatment (Fig. 7A). Flow cytometry confirmed these expression patterns of HLA-ABC, HLA-DR, CD40, CD80 and CD86 except slight increase in the expression of CD86 upon poly(I-C) treatment (Fig. 7B). In flow cytometry, hFCPCs showed high level expression of selected surface markers enriched in MSCs and poly(I-C) treatment did not change their expression levels significantly either (Fig. 7C).

Fig. 7.

Effect of poly(I-C) treatment on the expression of surface markers of hFCPCs. hFCPCs were treated with 1 µg/mL poly(I-C) for 1 day. A, B The expression of selected immune-related surface markers was examined by RT-PCR and flow cytometry, respectively. C The expression of selected MSCs markers was examined by flow cytometry

Treatment of poly(I-C) did not affect surface marker profile of hFCPCs

The effect of poly(I-C) priming was examined on the differentiation ability of hFCPCs into 3 mesodermal lineages. hFCPCs were differentiated into adipogenic, osteogenic and chondrogenic lineages for 3 weeks (Fig. 8). In the Oil Red O staining, poly(I-C) showed no significant effect on the adipogenic differentiation of hFCPCs (Fig. 8A). In contrast, it reduced osteogenic differentiation of hFCPCs but with no statistical significance in the quantitative analysis, while increased the diameter of chondrogenic pellets with a statistical significance (Fig. 8B, C).

Fig. 8.

Multipotent differentiation of hFCPCs and poly(I-C) treated hFCPCs. A Lipid droplets were observed in hFCPCs and poly(I-C)-hFCPCs after culture in adipogenic medium for 21 days by Oil Red O staining. Scale bar = 200 μm. The adsorbed dye was eluted and quantified at 500 nm. B Alizarin red staining identified calcium deposits in hFCPCs and poly(I-C)-hFCPCs after culture in osteogenic induction medium for 21 days. Error bars indicate SEs of triplicate experiments. Scale bar = 500 μm. The adsorbed dye was eluted and quantified at 405 nm. C hFCPCs and poly(I-C)-hFCPCs derived pellets were cultured in chondrogenic induction medium for 21 days and then stained with safranin O. Scale bar = 500 μm. The diameter of pellets was measured for quantitative analysis. Data are presented by mean with SD from 3 independent experiments (n = 3). **p < 0.01

Discussion

hFCPCs are progenitor cells with characteristics of both MSCs and chondrocytes. We have previously shown that hFCPCs are immune-privileged and can block PBLs activation by concanavalin A [12]. hFCPCs are also shown to express anti-inflammatory cytokines well-known in MSCs in the study. However, the study did not address therapeutic potential of hFCPCs on any specific disease models. In this study, we have investigated an optimal priming signal that can enhance paracrine activity of hFCPCs and their anti-inflammatory effect on the IL-1β -mediated OA phenotypes in human chondrocytes and synoviocytes.

Priming of MSCs using TLR ligands or pro-inflammatory cytokines is an attractive strategy for improving the potency and efficacy of MSC-based therapies. It has been reported that priming of MSCs with poly(I-C) or LPS which are an TLR-3 and TLR-4 agonist respectively can amplify MSCs trophic factors and enhances their therapeutic potency [8, 9, 27]. We have first treated hFCPCs with LPS for 3 days but it did not increase the expression of both inflammatory and anti-inflammatory cytokines. LPS is a well-known endotoxin that can stimulate MSCs to secrete more paracrine factors via TLR4 [8, 9]. Our result suggests that hFCPCs did not respond to LPS. The underlying mechanism of this result is not clear but might be due to cell type differences in the expression of TLR subtypes as shown in the case of MSCs. Raicevic et al. have shown that adipose tissue MSCs express both TLR-3 and TLR-4 at high levels in RT-PCR and flow cytometry, while bone marrow MSCs express them at high levels only in flow cytometry and Wharton’s jelly MSCs express TLR-3 only under inflammatory condition in flow cytometry [28]. They also showed that Warton’s jelly MSCs did not express TLR-4 in flow cytometry neither show increased expression of paracrine factors upon LPS treatment [28]. Interestingly, their study and others using different MSCs types have shown diverse patterns not only in TLRs expression and paracrine secretion but also their anti-inflammatory and immune-modulatory activities [8, 9, 28]. In our study, hFCPCs expressed both TLR-3 and TLR-4 in RT-PCR and flow cytometry. Upon treatment of poly(I-C) or LPS, their expression in hFCPCs appeared to increase slightly in RT-PCR but showed no significant difference in flow cytometry. It is not clear why hFCPCs did not respond to LPS since they express TLR-4 on their surface, which needs a future study to elucidate the underlying mechanism.

The expression pattern of HLA subtypes in MSCs is important in their interaction with immune cells and allogeneic immune response [6, 7]. It is well known that MSCs from various sources express HLA type I molecule but do not express HLA type II molecules and complementary molecules of CD80, CD86 and CD40. Activation of TLR signals is generally reported not to alter the expression of HLA subtypes and complementary molecules [29–32]. Only a rare exception has been reported with umbilical cord derived MSCs which showed increased expression of CD86 upon TLR-4 activation by LPS [27]. Therefore, MSCs are known to interact with immune cells and modulate their activities indirectly via secretion of diverse array of paracrine factors or using various cell adhesion molecules such as hyaluronic acid receptors, vascular cell adhesion molecules (VCAMs) and intercellular adhesion molecules (ICAMs) [30]. Similar to the case of MSCs, hFCPCs also expressed HLA-ABC, an HLA type I molecule only but not HLA-II or complementary molecules, and their expression levels did not change by the treatment of IFN-γ, LPS or poly(I-C). We have previously reported consistent result in the expression of these immune-related surface antigens by IFN-γ in hFCPCs [12]. We have also shown that TLR-3 priming of hFCPCs with poly(I-C) did not affect the expression pattern of so-called MSCs’ surface marker of CD44, CD54, CD73, CD90 and CD166 that are importance in cell adhesion. The expression pattern of other cell adhesion molecules in hFCPCs and their regulation of immune cells need future investigation.

The effect of TLRs priming on the differentiation ability of MSCs into osteogenic, adipogenic and chondrogenic lineages is also various depending on MSCs types and TLR signals and even inconsistent among different reports [8, 9]. When it comes to TLR-3 priming used in this study, it is generally reported to have no significant effect on the differentiation of human MSCs into 3 mesodermal lineages [8, 9], but some reports have shown that osteogenic differentiation was increased in adipose tissue MSCs [33, 34]. In our study, activation of TLR-3 signals in hFCPCs showed no significant effect on adipogenic differentiation, while reduced osteogenic differentiation and enhanced chondrogenic differentiation. This result is rather different from the case of MSCs, which might be due to unique chondrogenic character of hFCPCs. The underlying mechanism of this finding is not clear at this point and needs further investigation.

In this study, we have treated several priming factors to activate hFCPCs and found that poly(I-C), a TLR-3 ligand, was most prominent in increasing paracrine expression in hFCPCs. Once activated, the TLR-3 signal recruits MyD88 and activates NF-κB and MAPKs for the induction of inflammatory cytokine [8, 9]. Priming with proinflammatory factors of IL-1β, TNF-α or IFN-γ did not increase expression of anti-inflammatory or immune-modulatory factors in hFCPCs and the combination of them with poly(I-C) did not show any additive effect on their expression. This result was somewhat unexpected and different from the case of MSCs, which suggests that hFCPCs are less responsive to inflammatory signals other than TLR-3. In the subsequent experiment, poly(I-C) induced not only anti-inflammatory factors but also pro-inflammatory factors like IL-6 and COX-2 in hFCPCs. In human MSCs, TLR-3 priming of bone marrow or nasal mucosa MSCs with poly(I-C) was reported to stimulate expression of pro-inflammatory cytokines such as IL-1β, Il-6, IL-8 and C–C chemokine receptor type 5 (CCR5) [35, 36]. These cytokines are also known to act as a chemoattractant that can recruit immune cells in vivo [35]. Similarly, TLR-3 priming is reported to increase the secretion of many C-X-C motif chemokine ligand (CXCL) family chemokines in human tonsil MSCs [31]. Therefore, it is plausible that TLR-3 priming also increases secretion of these chemotactic factors in hFCPCs, though we did not screen for their expression. Interestingly, Waterman et al. have reported that TLR-3 priming of human MSCs with poly(I-C) increased expression of immune-modulators like IDO and prostaglandin E2 (PGE2), while TLR-4 priming with LPS increased expression of pro-inflammatory factors like IL-6 and IL-8 [29]. Similarly, Kota et al. have also shown similar results that TLR-4 priming highly induced pro-inflammatory secretion while TLR-3 priming indued secretion of some chemokines only like IL-6 and macrophage migration inhibitory factor (MIF) in human MSCS [30]. Therefore, it appears that TLR-3 priming of MSCs with poly(I-C) is a better strategy than TLR-4 priming with LPS to enhance anti-inflammatory and immune-modulatory activities of MSCs and their therapeutic effects.

MSCs are expected to exert various therapeutic effects and mechanisms on OA [19, 20]. They can modulate inflammation and pain, reduce degradation of cartilage extracellular matrices (ECMs), protect chondrocytes from apoptosis and possibly induce cartilage regeneration. Many anti-inflammatory molecules and growth factors secreted from MSCs are reported to play importance roles in these events. However, the OA environment in vivo is complicated with diverse interactions between transplanted MSCs and resident cells such as chondrocytes, synoviocytes and various immune cells, which make it difficult to understand the exact mechanism of MSCs and their paracrine factors to alleviate OA. Studies addressed in vitro have also suggested the role of paracrine factors from MSCs in reducing OA-related inflammatory response and apoptosis of chondrocytes [37–39]. Recently, the role of exosomes and miRNAs with anti-inflammatory activity is also highlighted as a key mechanism of MSCs’ therapeutic effect on OA [40, 41]. In our research, TLR-3-primed hFCPCs clearly reduced IL-1β -induced expression of inflammatory cytokines and MMPs in SW982 synoviocytes and human chondrocytes, while inducing expression of cartilage matrix proteins of aggrecan and type II collagen in chondrocytes. These results are very similar to MSCs’ therapeutic effect on OA phenotypes described above and strongly suggest their therapeutic potential as a cell therapy to treat OA. It is highly like that TLR-3-primed hFCPCs have more therapeutic potency that unprimed ones seeing that therapeutic effect of MSCs on OA was shown to depend critically on their expression of paracrine molecules [37]. Seeing that hFCPCs are highly proliferative and have chondrogenic characteristics [10, 11], we speculate that hFCPCs can not only modulate immune response but also enhance chondrocyte protection and cartilage regeneration in OA patients.

Stem or progenitor cells derived from various fetal tissues have emerged as promising sources of cells to be used in the treatment of various inflammatory diseases. It has been reported that soluble factors secreted by fetal membrane-derived mesenchymal stem cell ameliorated inflammatory responses in a rat model of acute and chronic pancreatitis, possibly mediated by the suppression of the activity of macrophages and pancreatic stellate cells and by the suppression of acinar cell injury [42]. Trophic factors from human Wharton’s jelly-derived stem cells were shown to reduce neurodegeneration and inflammation in the stroke brain [43]. Paracrine release of trophic factors by umbilical cord MSCs were also assumed as an important mediator to reduce inflammation in bleomycin-induced mouse model of lung injury [44]. Fetal tissues are generally considered immune privileged due to maternal–fetal tolerance, or in other words pregnancy-associated immune privilege [45, 46] and cells obtained from fetal tissues are regarded as a good cell source that can overcome allogeneic immune rejection. The immune-privileged property of fetal tissue-derived cells attribute not only to their low expression of HLA-II and complementary molecules but also expression of immune-modulatory secretory factors [46–48]. A particular emphasis has been on the function of HLA-G in the uterine-placenta immune privilege by targeting various subpopulations of hematopoietic cells with immunological functions [49]. In this and our previous studies [12], we have also shown that hFCPCs maintain their immune privileged property in vitro. In addition, hFCPCs could be a better cell source than adult MSCs because they are highly proliferative and obtained in large amount enough to establish a stable cell bank.

Taken all together, we speculate that hFCPCs could be an effective and cost-efficient off-the-shelf cell therapy to treat OA. Future studies need to address exact mechanism of TLR-3-primed hFCPCs to alleviate OA phenotypes including possible roles of hFCPCs-derived exosomes and extension of the findings in vitro to animal models of OA.

Declarations

Conflict of interest

The authors declare that there is no conflict of interest.

Ethical statement

The experiments were conducted with the approval of institutional review board (IRB) of Ajou University Medical Center (AJIRB-CRO-07-139) and (AJIRB-MED-SMP-10–266).