- Email: [email protected]
Cytokine interactions in mesenchymal stem cells from cord blood
www.elsevier.com/locate/issn/10434666 Cytokine 32 (2005) 270e279
Cytokine interactions in mesenchymal stem cells from cord blood Chi-Hsien Liu a,b,*, Shiaw-Min Hwang b,* b
a Graduate Institute of Biochemical and Biomedical Engineering, Chang Gung University, Taoyuan, Taiwan Bioresource Collection and Research Center, Food Industry Research and Development Institute, Hsinchu, Taiwan
Received 18 February 2005; received in revised form 21 June 2005; accepted 4 November 2005
Abstract We used cytokine protein array to analyze the expression of cytokines from human cord blood-derived mesenchymal stem cells (CB-MSCs). Several cytokines, interleukins (IL), and growth factors, including ENA-78, GM-CSF, GRO, IL-1b, IL-6, IL-8, MCP-1, OSM, VEGF, FGF-4, FGF-7, FGF-9, GCP-2, IGFBP-1, IGFBP-2, IGFBP-3, IGFBP-4, IP-10, LIF, MIF, MIP-3a, osteoprotegerin, PARC, PIGF, TGF-b2, TGF-b3, TIMP-1, as well as TIMP-2, were secreted by CB-MSCs, while IL-4, IL-5, IL-7, IL-13, TGF-b1, TNF-a, and TNF-b were not expressed under normal growth conditions. IL-6, IL-8, TIMP-1, and TIMP-2 were the most abundant interleukins expressed by CB-MSCs. A set of growth factors were selected to evaluate their stimulatory effects on the IL6 secretion for CB-MSCs. IL-1b was the most important factor inducing CB-MSC to secret IL-6. The mechanism by which IL-1b promoted IL-6 expression in CB-MSCs was studied. By using various inhibitors of signal transduction, we found that activation of p38 mitogen-activated protein kinases (MAPK) and MAPK kinase (MEK) is essential in the IL-1b stimulated signaling cascade which leads to the increase in IL-6 synthesis. Additionally, continuous supplement of IL-1b in the CBMSCs culture will facilitate adipogenic maturation of CB-MSCs as evidenced by the presence of oil drops in the CB-MSCs and secretion of leptin, a molecule marker of adipocytes. These results strongly suggest that cytokine induction and signal transduction are important for the differentiation of CB-MSCs. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Mesenchymal stem cell; Cord blood; Cytokine; Interleukin-1b; Interleukin-6
1. Introduction Mesenchymal stem cells (MSCs) are capable of self-renewal and differentiation into lineages of mesenchymal tissues, including bone, cartilage, fat, tendon, muscle, and hematopoieticsupporting stroma [1]. Although the existence of hematopoietic Abbreviations: MCP, monocyte chemotactic protein; IGFBP, insulin-like growth factor binding protein; ENA-78, epithelial neutrophil-activating protein 78; GRO, growth-related oncogene; OSM, oncostatin M; FGF, fibroblast growth factor; GCP, granulocyte chemotactic protein; IP, interferon-inducible protein; MIF, macrophage migration inhibitory factor; MIP, macrophage inflammatory protein; PARC, pulmonary and activation-regulated chemokine; PIGF, phosphatidylinositol glycan complementation class F; TGF, transforming growth factor; TIMP, tissue inhibitors of matrix metalloproteinases. * Corresponding authors. Graduate Institute of Biochemical and Biomedical Engineering, Chang Gung University, 259 Wen-Hwa first Rd, Kwei-Shan, TaoYuan 333, Taiwan. Tel.: þ886 3 211 8800 3146; fax: þ886 3 211 8700. E-mail addresses: [email protected] (C.-H. Liu), [email protected] (S.-M. Hwang). 1043-4666/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.cyto.2005.11.003
stem cells in cord blood (CB) is well documented, research on MSCs in cord blood is still in progress. Several researchers successfully isolated, expanded, and characterized the MSCs from umbilical cord blood and evaluated their potential for differentiation into osteogenic, chondrogenic or adipogenic lineage [2e 6]. The circulation of pluripotent MSCs in the blood of preterm fetus between liver and bone marrow explains why MSCs can be isolated from cord blood [6]. While the most abundant source of MSCs is bone marrow [7], MSCs from umbilical cord blood can serve as an alternative for cell-based therapies for the following reasons including the significant reduction in number of MSCs with age and the high risk of viral contamination in bone marrow. Although MSCs from placental and umbilical cord blood can be successfully isolated and expanded [8,9], the fact that CB-MSCs are present at a very low frequency [10] may influence their autologous transplantation and other clinical applications. This limitation needs to be further solved by well-designed isolation and expansion procedures.
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
MSCs in the bone marrow (BM) are involved in hematopoiesis, bone formation, bone resorption, and the development of cartilage and fat. From some of these events, BM-MSCs exert their influence through secreting soluble proteins. These macromolecules provide molecular signals to neighboring cells and modulate their mitotic, metabolic, and/or development states. Moreover, MSCs respond to components of the environmental milieu, including autocrine and paracrine factors synthesized by different cells, resulting in modulation of their own mitotic, metabolic, and/or developmental activity [11]. However, the cytokine expression profile of CB-MSCs and the influence of cytokine on the differentiation of CB-MSCs remains poorly documented. Utilization of this serum-free medium will facilitate analysis of interactions between growth factors and cytokines on the proliferation and differentiation of CB-MSCs without the complexity of exogenous serum. The serum-free medium for MSCs was reported and the differentiation potential of adipogenic, osteogenic, and chondrogenic lineages of MSCs was maintained under the serum-free medium [12]. In addition, serum-free medium was used to evaluate the effect of cytokines and growth factors on macrophage colonystimulating factor secretion by human bone marrow stromal cells [13]. In this study, we used commercial human cytokine protein array to identify protein expression profiles of CBMSCs under serum-free culture conditions. The study aimed at characterizing the secreted cytokine profile of CB-MSCs, evaluating the effect of cytokines on IL-6 release of CBMSCs, and determining how their differentiation process was altered when CB-MSCs were exposed to exogenous IL-1b.
2. Materials and methods
271
medium (IMDM þ 20% FBS þ 10 ng/ml basic fibroblast growth factor (bFGF). Before performed the experiment under serum-free conditions, confluent CB-MSCs were digested with 0.25% trypsin-EDTA (JRH Biosciences, KS) from 20% FBS supplemented medium and seeded at the density of 105 cells/ml to serum-free medium. After CB-MSCs were confluent in T75 flasks under the serum-free condition, CB-MSCs were detached with 1 ml of NO-ZYME (JRH Biosciences) and seeded in the six-well plate at the density of 105 cells/well for the following experiments, including effects of cytokines on IL-6 expression and multi-lineage differentiation. 2.2. Effects of cytokines on IL-6 expression For screen of different cytokines on IL-6 expression in CBMSCs, the concentration of 10 ng/ml of different cytokines was added in the serum-free culture medium. The initial seeding density was 105 cells per well in 6-well culture dishes, and the cell-free conditioned medium was analyzed by ELISA method after a 3-day culture. In order to assay the kinase inhibitor effects, 105 cells were plated in 35-mm 6-well culture dishes under serum-free condition after 3 days of growth. Medium was then changed to fresh IMDM medium and cells were treated with freshly prepared kinase inhibitors (PD98059 and SB202190) at various doses for 1 h. The range of concentration tested in the evaluation procedure was 0e50 mM of PD98059, 0e50 mM of SB202190 and 10 ng/ml of IL-1b according to the preliminary experiments. After removal of medium, fresh serum-free medium and/or IL-1b were added and cells were cultured for 24 h according the protocol presented by Huang and Zhang [15]. The conditioned medium was collected and stored at 80 C for further cytokine analysis.
2.1. Isolation and culture of CB-MSCs CB-MSCs were isolated and cultured according to modifications of a previous reported protocol [14]. Briefly, the term cord blood of newborns was layered onto FicollePaque solution (1.077 g/ml, Amersham, Uppsala, Sweden) and centrifuged to deplete the residues of red blood cells, platelets and plasma (700 g for 40 min). Mononuclear cells in the interface were isolated, and seeded at the concentration of 106 cells/cm2 in the serum-containing medium (SCM, IMDM þ 20% FBS (Hyclone, Logan, UT) þ 10 ng/ml basic fibroblast growth factor (bFGF)) at 37 C for 2e3 weeks. The well-developed colonies of fibroblast-like cells were collected and preserved in liquid nitrogen for further experiments. Our laboratory routinely isolated MSCs from cord blood. MSCs can be successfully isolated in 108 units of cord blood out of a total of 144 donated samples. The efficacy of CB-MSCs isolation is 75% (n ¼ 144). Because these isolated MSCs have similar morphology, differentiation potentials, growth characteristics, and the same immunophenotypic markers (CD26, CD31, CD34, CD45, HLA-DR, CD29þ, CD44þ, HLA-ABCþ, SH2þ, SH3þ, SH4þ, Fig. 1), one representative stain of CB-MSCs, CBMSC no. 354, was used as a typical example in our experiments. CB-MSCs were routinely maintained in serum-containing
2.3. Evaluation of differentiation To test chronic supplement of IL-1b on the differentiation of CB-MSCs, serum-free medium supplemented with 10 ng/ml of IL-1b was replaced every 3e4 days for a total of 45 days. The conditioned serum-free medium was collected and preserved at a 80 C refrigerator for further IL-6 and leptin assay. At term, the cultured cells were washed twice with PBS, fixed in 10% formaldehyde for histochemical staining. Differentiation of adipocytes, chondrocytes and osteocytes were evaluated by the appearance of Oil-red-O stained lipid vacuole [16], the Safranin-O stained red accumulation of sulfated proteoglycans [16] and von Kossa method stained mineralization of calcium nodules [17], respectively. 2.4. Immunophenotyping of CB-MSCs After CB-MSCs expanded in SFM for two passages, the surface markers were analyzed using a FACSCalibur flow cytometry system (BD Biosciences, Canada). CB-MSCs (106 cells) were fixed with 70% ethanol (10 min at 4 C). Antibodies against human antigens CD26, CD31, CD34, CD45,
272
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
Fig. 1. Immunophenotyping profile of CB-MSCs cultured in serum-containing or serum-free media.
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
HLA-DR, CD29, CD44 and HLA-ABC were purchased from Becton Dickinson (San Diego, http://www.bd.com). Antibodies against human antigens SH-2, SH-3 and SH-4 were purified from the respective hybridoma cell lines acquired from ATCC. The fixed cells were stained with fluorescein isothiocyanate (FITC) or phycoetrythrin (PE)-conjugated antibodies. A replicate sample was stained with FITC-mouse IgG1 and PE-mouse IgG1 as an isotype control to ensure specificity. The control CB-MSCs were maintained in serumcontaining medium. Their surface marker pattern, including CD26, CD31, CD34, CD45, HLA-DR, CD29þ, CD44þ, HLA-ABCþ, SH2þ, SH3þand SH4þ, was the same as CB-MSCs cultured in serum-containing medium (Fig. 1). 2.5. Antibody-based protein array system Cell-free supernatant was taken from conditioned serumfree medium of 4-day cultured CB-MSCs, and then analyzed by RayBio Human Cytokine Array V kit according to the manufacturer’s instructions. Human Cytokine Array V kit was purchased from RayBiotech (catalog no. H0108005A; Norcross, GA). The array membranes can detect 79 different growth factors/cytokines at a time. The layout of the membrane is listed in Fig. 2. Positive controls are located in the upper left-hand corner (four spots) and lower right-hand corner (two spots) of each membrane. The assay was followed precisely as stated in the directions from the manufacturer. In brief, place each membrane into the provided eight-well tray. Add 2 ml blocking buffer and incubate at room temperature for 30 min to block membranes. Decant blocking buffer from each container, and incubate membranes with 1 ml of conditioned medium at room temperature for 1e2 h. Decant the samples from each container, and wash three times with 2 ml of wash buffer I at room temperature with shaking for 5 min. Wash two times with 2 ml of wash buffer II at room temperature with shaking for 5 min. Add 1 ml of 250 fold diluted biotin-conjugated antibodies to each membrane. Incubate at room temperature for 1e2 h. After washing two times, 12000
IL-6 (pg/mL)
10000 8000 6000 4000 2000
273
add 2 ml of 1000 fold diluted HRP-conjugated streptavidin to each membrane. Incubate at room temperature for 30 min and wash two times. Place membrane into the detection buffer and incubate at room temperature for 5 min. Drain off excess detection reagent and wrap the membrane with PE wrap. The membrane was then exposed to Hyperfilm (Amersham Bioscience, catalog no. RPN2103K). Detectable spots were scanned by densitometer, and detection with fresh culture media was subtracted from the test samples. Positive controls, provided by the manufacturer, were normalized to 1-fold, and the unknown samples were calculated from the normalized spots. 2.6. ELISA IL-6 and leptin were quantified by using the Flexia antibody pair system (BioSource, Camarillo, CA.) and DuoSet ELISA Development System (R&D Systems, Minneapolis, MN), respectively. The assay procedures were performed according to the instructions of the manufacturer. A solution of 3,3#,5,5$-tetramethylbenzidine and H2O2 (Kirkeguard and Perry Laboratories Inc.) was added as the peroxidase substrate. Finally, 1 M phosphoric acid was added to terminate the reaction and the plates were measured at a wavelength of 450 nm. Each sample was tested in duplicate. 2.7. Growth factors, culture medium and chemicals The following recombinant human growth factors or cytokines were used, purchased from PeproTech (Rocky Hill, NJ): transforming growth factor-b1 (TGF-b1); TGF-b2; TGF-b3; bFGF; vascular endothelial growth factor (VEGF); leukemia inhibitory factor (LIF); thrombopoietin (TPO); stem cell factor (SCF); stem cell growth factor-a (SCGF-a); hepatocyte growth factor (HGF); Flt-3 ligand (FL); interleukin (IL)-1b; IL-3; granulocyte colony-stimulating factor (G-CSF); granulocyte-macrophage colony-stimulating factor (GM-CSF). IL-6 and IL-6 sR (soluble receptor) were purchased from R&D Systems. Iscove’s modified Dulbecco’s medium (IMDM, catalog no. I7633) was from Sigma. Human albumin (25%) was from Aventis Behring (Kankakee, IL). Hydrocortisone and SITE (containing 0.5 g/ml sodium selenite, 1 mg/ml insulin, 0.55 mg/ml transferrin and 0.2 mg/ml ethanolamine) were purchased from Sigma. Tissue culture grade T25 and T75 cm2 flasks and 6-well plates were purchased from Corning Costar (Cambridge, MA). PD 98059 and SB 230580 were obtained from CalbiochemeNovabiochem Co. and dissolved in DMSO (Sigma) as a stock at a concentration of 50 mM.
Co nt ro SC l G LI F- F al p V ha EG F G HG M F -C SF SC F IL 3 TP G O -C SF IL FL TG 6F SR TG -be F ta1 TG -be F- ta2 be ta bF 3 IL IL- GF -1 1 be bet ta a +H C
0
Fig. 2. Effects of cytokines on IL-6 secretion by CB-MSCs. The concentration of 10 ng/ml of different cytokines was added in the serum-free medium. The initial seeding density was 105 cells per well in 6-well culture dishes and the cell-free conditioned medium was analyzed by ELISA method after 3 days of culture. Results are expressed as means of triplicate samples and the bar indicates standard deviation.
2.8. Detection of PPARg2 and C/EBPb in CB-MSCs by RT-PCR Total RNAs were extracted from cultured human cells using Trizol (Gibco) according to the manufacturer’s protocol. Genomic DNA contamination was removed by incubation with 10 units of DNAse 1 in solution of 10 mM TriseCl, pH 8.3, 50 mM KCl, and 1.5 mM MgCl2 for 30 min at 37 C.
274
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
Ten micrograms of total RNA from CB-MSCs, were reverse transcribed using 20 units of AMV reverse transcriptase. PCR amplification was performed on 2 ml first-strand cDNA with 0.5 mM of primers and 2.5 units Taq DNA Polymerase, 2 mM dNTP, and 1.5 mM MgCl2. Oligonucleotide primers used for PPARg2, C/EBPb, and b-actin were 5#-CCAGAAAATGACAGACCTCAGACA-3# (forward), 5#-GCAGGAGCGGGTGAAGACT-#’ (reverse); 5#-CCGGGCAGCACCACGACTTCCTCT-3# (forward), 5#-CTTCTTGGCCTTGCTCTTGACCTG-3# (reverse); and 5#-AGACGTATCACCTCTGCAC-3# (forward), 5#-GGAAGCAACGTCTGTGAGGT-3# (reverse), respectively. The PCR products were electrophoresed through 1% agarose gel and the corresponding expected sizes of DNA bands were exercised and purified for sequence confirmation. 2.9. Statistical analysis The data was analyzed by paired samples t-test, and a value of P < 0.05 was considered significant. 3. Results 3.1. Cytokine array analysis of conditioned medium of CB-MSCs from cord blood The conditioned medium after a 4-day culture of MSC from cord blood was assayed using Raybiotech cytokine array. CB-MSCs spontaneously secrete twenty-eight cytokines in their culture supernatants as indicated in the cytokine array (Table 1). Twenty-eight cytokines belonging to the chemokine family and those associated with cell proliferation and immunity were detected from the array, which can analyze 79 growth factors/cytokines at a time. These secreted cytokines can be classified into the following groups. Chemokines: GRO, IL-8, MCP-1, MIP-3a, PARC, IP10, ENA-78, GCP-2, osteoprotegerin; Broad-acting cytokines: IL-1b, IL-6; Antiinflammatory: TGF-b2, TGF-b3, MIF, LIF; Growth factor: GM-CSF, VEGF, FGF-4, FGF-7, FGF-9, PIGF; Growth factor receptor: IGFBP-1, IGFBP-2, IGFBP-3, IGFBP-4; Angiogenic factor: oncostatin M; Proteinase inhibitor: TIMP-1, TIMP-2. Among them, IL-6, IL-8, TIMP-1 and TIMP-2 are the most abundant cytokines secreted by CB-MSCs. MCP-1, IGFBP-4, LIF, and VEGF are also plentifully secreted by CB-MSCs. 3.2. Effects of cytokines on IL-6 secretion IL-6 is one of the most abundantly expressed cytokines by CB-MSCs. Accumulating evidence supports an essential role of IL-6 in the development, differentiation, regeneration and degeneration of different stem cells [18]. IL-6 is also involved in tissue remodeling through its effects on connective tissue cells, proteases and protease inhibitors [19]. The up-regulation effect of different cytokines on the expression IL-6 by CBMSCs was evaluated. First, growth factors and cytokines were used to evaluate their effects on the IL-6 secretion. These tested growth factors and cytokines include LIF, VEGF, SCGF-a, HGF, GM-CSF, SCF, IL-3, TPO, G-CSF, FL, IL-6-SR,
Table 1 Cytokine array result with supernatants from CB-MSCs Cytokines
Fold increase
Chemokines GRO,MIP-3 PARC, IP10, ENA78, GCP MCP-1, Osteoprotegerin IL-8
0.33 0.66 1 2
Broad-acting cytokines IL-1b IL-6
0.33 2
Anti-inflammatory cytokines TGF-b3 MIF TGF-b2, LIF
0.33 0.66 1
Growth factors FGF-4, FGF-7, FGF-9 GM-CSF, PIGF VEGF
0.33 0.66 1
Growth factor receptors IGFBP-1, IGFBP-2 IGFBP-3, IGFBP-4
0.66 1
Protease inhibitors TIMP-1, TIMP-2
2
Angiogenic factor OSM
0.66
Cell-free supernatants were taken from confluent CB-MSCs and then analyzed by microarray. Results are the average of three samples. Detectable spots were scanned by densitometer, and detection with fresh culture media was subtracted from the test samples. Positive controls, provided by the manufacturer, were normalized to 1-fold, and the unknowns calculated accordingly. Details of the procedure are described in Section 2.
TGF-b1, TGF-b2, TGF-b3, bFGF, IL-1b, and hydrocortisone. Among these tested compounds, IL-1b, TGF-b2, TGF-b3, LIF, VEGF, and GM-CSF were also constitutively expressed by CB-MSCs as indicated in the previous cytokine array analysis. These compounds were used to check whether the autocrine effects exist in IL-6 secretion of CB-MSCs. In addition, bFGF and hydrocortisone were supplemented in the serumfree medium. The IL-1b was the most important factor that induces the MSC to secret IL-6. TGF b1 and TGFb2, but not TGFb3, can also induce the MSC to secret IL-6 (Fig. 1). Simultaneous addition of 10 ng/ml of IL-1b and 55.3 mM of hydrocortisone significantly reduced the IL-6 expression of CB-MSCs. Hydrocortisone reduced the stimulatory effects of IL-1b on IL-6 production by CB-MSCs. Basic FGF, supplemented in our serum-free medium, did not stimulate CBMSCs to express IL-6. The other tested cytokines, including LIF, VEGF, SCGF-a, HGF, GM-CSF, SCF, IL-3, TPO, G-CSF, FL, IL-6-SR, TGF-b3, and bFGF, have no stimulation effects on IL-6 production in CB-MSCs at the tested concentration (10 ng/ml) and higher concentration (50 ng/ml, data not shown). 3.3. Effects of IL-1b on IL-6 secretion IL-1b increased IL-6 secretion by CB-MSCs in a dosedependent manner. Fig. 3A illustrates the response of
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
(A)
275
14000 14000
*
12000
IL-6 (pg/mL)
IL-6 (pg/mL)
12000 10000 8000 6000
*
*
6000 4000
2000
2000 0 0
5
10
15
20
25
30
35
IL-1 beta (ng/mL)
10000 IL-1b
8000
IL-1 SB 202190 PD 98059
10ng/mL -
10ng/mL 50uM -
10ng/mL 50µM
10ng/mL 50µM 50µM
Fig. 4. Effects of kinase inhibitors (PD98059 and SB202190) on IL-6 secretion by CB-MSCs. The cultured CB-MSCs were pretreated with kinase inhibitor for 1 h and then stimulated by 10 ng/ml IL-1b for 24 h. The data were average of triplicate experiments. *P < 0.05, compared with the value of control. e, no addition.
14000 12000
IL-6 (pg/mL)
8000
4000
0
(B)
10000
control
6000 4000 2000 0 0
10
20
30
40
50
60
inhibited IL-1b-induced IL-6 production (about 14% inhibition). Our results indicate that IL-1b stimulated CB-MSCs to express IL-6 require the activation of p38 MAPK and ERK pathways.
Hour Fig. 3. (A) Effect of IL-1b dosage on IL-6 secretion in CB-MSCs. Different dosages of IL-1b were added in the serum-free medium. The initial seeding density was 105 cells per well in 6-well culture dishes and the IL-6 in cellfree conditioned medium was analyzed by ELISA method after 3 days of culture. Results are expressed as means of triplicate samples and the bar indicates standard deviation. (B) Effects of culture time on IL-1b induced IL-6 secretion in CB-MSCs. Dosage of 10 ng/ml IL-1b was added in the serum-free culture medium. The initial seeding density was 105 cells per well in 6-well culture dishes and the cell-free conditioned medium was collected at different culture times, and then the IL-6 in conditioned medium was analyzed by ELISA method. Results are expressed as means of triplicate samples and the bar indicates standard deviation.
CB-MSCs to the increasing concentrations of IL-1b in culture medium between 0e30 ng/ml. Additional increases in IL-1b (above 10 ng/ml) to the culture medium generate saturate production of 12,000 pg/ml IL-6. These data were used to choose the optimal concentration of IL-1b to add to the culture medium in the subsequent experiments. Furthermore, IL-1b increased the expression of IL-6 secretion by CB-MSCs in a timedependent manner (Fig. 3B). The IL-6 secretion by CBMSCs reaches the saturate concentration after 24 h culture. Signaling of IL-1b was reported to be initiated by IL-1 receptor, which triggered the mitogen-activated protein kinase (MAPK) cascade to transduce the signal into the nucleus. To assess the role of MAPK pathways in IL-1b-mediated activation of IL-6 expression, two inhibitors (SB202190, PD98059) were used in cells treated with cytokine (Fig. 4). Kinase inhibitor SB202190 (a p38 MAPK inhibitor) and PD98059 (a MAPK kinase (MEK) inhibitor) inhibited the expression of IL-6 induced by IL-1b. However, inhibitor SB202190 had strong inhibition effects on IL-1b induced IL-6 secretion than inhibitor PD98059. We found that the expression in IL-6 protein by IL-1b was eliminated 50% by the pretreatment of both cell types with SB202190. Moreover, PD98059 partially
3.4. Effects of chronic IL-1b supplement on CB-MSCs Initial addition of IL-1b in the medium stimulated the expression level of IL-6. However, the expression of IL-6 decreased gradually when IL-1b was continuously added in the medium (Fig. 5A). Furthermore, consecutive supplement of IL-1b can promote adipogenic maturation of CB-MSCs as evidenced by the presence of oil drops in the CB-MSCs after 45 days of culture (Fig. 6A). The control CB-MSCs were free of oil drops under the culture condition without IL-1b supplement (Fig. 6B). The level of leptin in the conditioned medium was continuously increasing after 17 days of culture (Fig. 5B). The evidence of oil drops in the CB-MSCs by Oil-Red-O stain and the secretion of leptin which is the differentiation marker of adipocytes, support that supplement of IL-1b in vitro can facilitate terminal adipogenic maturation of CB-MSCs. No osteocytes and chondrocytes were existed in these CB-MSC samples as verified by the histochemical stain (data not shown). The mechanism of IL-1b on the commitment of adipogenic differentiation of CB-MSCs and other marker for adipogenic differentiation of CB-MSCs were further investigated. The CCAAT/enhancer binding proteins (C/EBP) of transcription factors were stimulated by IL-1b in human enterocytes [20] and in mouse astrocytes [21]. Peroxisome proliferatoractivated receptors (PPARs) are a transcription factor family related to the induction and differentiation from primitive precursor cells to the terminally differentiated adipocytes [22]. Therefore, we analyzed expression of PPARg2 and C/EBPb by RTePCR in different CB-MSC samples after 21 days of culture. However, PPARg2 and C/EBPb are present in control (cultured serum-free and serum-containing media) and IL-1b treated CB-MSCs (Fig. 7). The existence of PPARg2 and C/EBPb in CB-MSCs might be due to the tendency of CB-MSCs would spontaneously differentiate into
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
276
(A)14000 IL-1b
IL-6 (pg/mL)
12000
control
10000 8000 6000 4000 2000 0 0
5
10
15
20
25
30
35
40
45
50
30
35
40
45
50
Day
(B)
8000 IL-1b
Leptin (pg/mL)
7000
control
6000 5000 4000 3000 2000 1000 0
0
5
10
15
20
25
Day Fig. 5. (A) Effects of consecutive treatment of IL-1b on the secretion of IL-6 by CB-MSCs. (B) Effects of consecutive treatment of IL-1b on the secretion of Leptin by CB-MSCs. The initial seeding density was 105 cells per well in 6-well culture dishes and serum-free medium supplemented with 10 ng/ml of IL-1b was replaced every 3e4 days for 45 days. The conditioned serumfree medium was collected and preserved at 80 C for a further IL-6 and leptin assay. Results are expressed as means of triplicate samples and the bar indicates standard deviation.
adipocytes (although in a low percentage) and the sensitivity of RT-PCR method. 4. Discussion Previous results indicate that CB-MSCs display many common features with BM-MSCs, like adhesion to plastic, morphology, expression of similar surface markers, and a potential to differentiate into osteo-, chondro-, and adipogenic
phenotypes. Potian et al. recently analyzed the cytokine profile of BM-MSCs by the protein array technique [23]. IL-6, IL-8, MCP-1, RANTES, GRO-a, IFNg, IL-1a, TGF-b, GM-CSF, angiogenin, and oncostatin M were constitutively expressed and MIP-1b, IL-2, IL-4, IL-10, IL-12, IL-13 were not expressed by the BM-MSCs [23]. The cytokine profile of BM-MSCs is very similar to that of CB-MSCs in this study, except that CB-MSCs express IL-12 and do not express G-CSF under serum-free condition. Haynesworth et al. reported that constitutively expressed cytokines in this growth phase include IL-6, G-CSF, SCF, LIF, M-CSF, and IL-11, while GM-CSF, IL-3, TGF-2, and OSM were not detected in the growth medium of human bone marrow-derived MSCs [24]. Both MSCs from bone marrow and cord blood abundantly produced IL-6, IL-8, and MCP-1. Interestingly, G-CSF expression of CB-MSCs is distinct from that of BM-MSCs. G-CSF can be detected in the conditioned medium of BM-MSCs, but not in that of CB-MSCs. Besides, tissue inhibitors of metalloproteinase (TIMP)-1 and TIMP-2 were found to be plentifully expressed by CB-MSCs in this study. TIMPs are natural inhibitors of matrix metalloproteinases found in most tissues and body fluids. By inhibiting MMPs activities, they participate in tissue remodeling of the extracellular matrix (ECM). The balance between MMPs and TIMPs activities is involved in tissue remodeling, angiogenesis, and hematopoiesis, as reviewed recently [25]. The implication of TIMP-1 and TIMP-2 in physiological and differentiation processes of MSCs merits further investigation. These results showed that constitutive expression of these growth factors could be related to provide inductive and regulatory information, which is consistent with the ability to support hematopoiesis, and also supply autocrine/ paracrine factors that influence hemopoietic functions, angiogenic properties, and immune regulation. Although the real physiological functions of cytokine expression by CB-MSCs remain to be elucidated, the developed serum-free culture system provides a simple and defined environment to study these cytokine interactions without the complexity of serum. IL-6 is another abundant cytokines secreted by CB-MSCs. Accumulating evidence supports an essential role of IL-6 in the development, differentiation, regeneration and degeneration
Fig. 6. Oil-red-O stain of CB-MSCs in 10 ng/ml of IL-1b supplemented serum-free medium after 45-day culture. The initial seeding density was 105 cells per well in 6-well culture dishes and serum-free medium supplemented with 10 ng/ml of IL-1b was replaced every 3e4 days for 45 days. The CB-MSCs were stained to check the existence of lipid vacuoles by Oil red O stain. Details of the procedure are described in Section 2.
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
Fig. 7. Expression of PPARg2 and C/EBPb in CB-MSCs as determined by RT-PCR. Lane 1: CB-MSCs cultured in serum-free medium plus 10 ng/ml after 21 days of culture. Lane 2: CB-MSCs cultured in serum-free medium after 21 days of culture. Lane 3: CB-MSCs cultured in serum containing medium after 21 days of culture. Lane 4: a negative control.
of different stem cells [18]. IL-6 functions are mediated by a specific receptor system composed of a binding site and a signal transducer. IL-1b is a major mediator of microbial invasion, inflammation, immunological reactions and tissue injury. IL-1b can stimulate human myoblasts [26], human synoviocyte or synovial sarcoma [27], and mouse brown adipocytes [28] to increase expression of IL-6. Furthermore, IL-1a can induce BM-MSCs to increase the expression of G-CSF, M-CSF, GM-CSF, LIF, IL-6, and IL-11 [24]. Both IL-1 and IL-6 are multifunctional cytokines that play an important role as regulators of the immune response and homeostasis processes [29]. The biological relevance of IL-1 and IL-6 in BM-MSCs was studied by several authors [11,24]. Their results suggested that MSCs might be able to influence protective reactions locally as well as at a distance from the bone marrow by expression of their cytokines and growth factor, such as IL-1 and IL-6, during inflammation and/or infection. Hydrocortisone was included in a defined medium to support a high-density culture of chick and mouse limb mesenchymal cells [30]. We also found that hydrocortisone can stimulate the proliferation of CB-MSCs under serum-free condition (data not shown). So our serum-free medium contained hydrocortisone. Glucocorticoids, such as hydrocortisone and dexamesathone, were reported to suppress release of IL-6 and stimulate leptin secretion by human adipocytes in a dose-dependent manner [31]. Hydrocortisone significantly reduced the stimulatory effects of IL-1b on IL-6 production by CB-MSCs (Fig. 2). From these studies, the decreased production of IL-6 during the hMSCs adipo-differentiation process was due to the effects of hydrocortisone. Furthermore, IL-6 was reported to inhibit hydrocortisone-induced adipocyte differentiation of murine stromal cells in a dose-dependent manner [32]. The gradually decreased IL-6 expression might reduce the inhibition effect of IL-6 on adipo-differentiation of CB-MSCs. CB-MSCs were similar to BM-MSCs in morphological features, immunophenotypic markers, and differentiating abilities; however genes related to antimicrobial activity, osteogenesis, matrix remodeling, and angiogenesis are differently expressed in CB-and BM-MSCs [33]. Moreover, CB-MSCs have longer telomeres, which may explain the high expansion capacity of CB-MSCs than that of BM-MSCs [9]. Further differences include absent expression of certain adhesion molecules such as ICAM-3, L-selectin, and VCAM, as well
277
as HAS1 in CB-MSCs. Additionally, CB-MSCs have a much broader differentiation potential and differ in their RNA expression profile as previously reported [8]. Delikat et al. found that addition of IL-1b to human long-term bone marrow culture (LTBMC) inhibited adipocyte formation in serumsupplemented conditions [34,35]. In contrast, we demonstrated that continuous supplementation of IL-1b in serum-free culture conditions promotes adipogenic maturation. The difference response of these two kinds of MSCs to IL-1b might be caused by culture condition (defined and serum-free vs. undefined and 25% serum supplemented) and the isolation sources (cord blood vs. bone marrow). Furthermore, previous studies demonstrated that administration of recombinant IL-1b could increase the concentration of serum leptin in rat [36] as well as human [37] and increase levels of leptin mRNA in adipose tissue of fasted hamster [38]. Our in vitro data also demonstrated that supplement of IL-1b in the CB-MSCs culture will induce CB-MSCs to secret leptin. Adipogenesis is a complex process including proliferation of precursor cells, their commitment to the adipogenic lineage, and terminal differentiation [22]. The early events in adipogenic commitment include the activation of cascade of transcription factors as proposed by Rosen [39]. In brief, induction of C/EBPb and d leads to increase expression of C/EBPa and PPARg, and these two transcription factors trigger the expression of genes important for storage of lipid and secretion of leptin, adipsin, and adiponectin. Hence C/EBPs and PPARs were transcription factors related to the induction and differentiation from primitive precursor cells to the terminally differentiated adipocytes [22]. Besides, C/EBPs were involved in tissue-specific metabolic gene transcription, in signal transduction activated by several cytokines, and in terminal differentiation of a variety of cells including adipocytes [40] and myelomonocytes [21]. IL-1b was reported to strongly stimulate the expression of the C/EBP isoforms in human enterocytes [20] and in mouse astrocytes [21]. Although the terminal differentiation of adipocytes is intensively studied and well described in the murine cell lines, key information regarding the commitment of human precursor cells to the adipogenic lineage is still lacking. Since human MSCs can differentiate into multiple mesenchymal lineages, they were reported to provide a unique model to better understand early differentiation events [22]. We analyzed PPARg2 and C/EBPb by RT-PCR in our CB-MSC samples under different culture conditions (Fig. 7). PPARg2 and C/EBPb are present in control (cultured serum-free and serum-containing media) and IL-1b treated CB-MSCs. The presence of PPARg in uncommitted and adipo-differentiation MSCs is consistent with the recent report of Janderova´ et al. [22]. In BM-MSCs, C/EBPb protein was increased transiently within 24 h and greatly reduced at the end of the 21-day adipo-differentiation culture [22]. On contrary, mRNA of C/EBPb was still detected after 21 days of culture in CB-MSCs. The existence of PPARg2 and C/EBPb in CB-MSCs may be due to the tendency of CB-MSCs would spontaneously differentiate into adipocytes (although in a low percentage) and the sensitivity of RT-PCR method. Therefore, our RT-PCR results cannot
278
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
specify which stage IL-1b is involved in the early event of adipogenic commitment. Leptin is an important differentiation marker of preadipocytes and adipocytes as emphasized by many authors [31,40,41]. The accumulation of lipid-vacuole and the secretion of leptin in our study strongly support that IL-1b can facilitate the terminal adipogenic differentiation and adipocyte maturation process in CB-MSCs. However, mechanisms of IL-1b on the commitment of adipogenic differentiation of CB-MSCs remain to be determined. Further experiments must show whether the effect of IL-1b on adipo-lineage differentiation of CB-MSCs is influenced by the existence of serum. Overall, the present study provides an initial characterization of cytokine expression by CBMSCs and highlights the important role played by the cytokine interaction in the differentiation of CB-MSCs. Acknowledgments This work was supported by the Foundation of Research and Development from FIRDI (03A006) and the Ministry of Economic Affairs, Taiwan (93-EC-17-A-17-R7-0525). We thank Drs. Chao-Ling Yao and Yu-Jen Chang for assistance in immunophenotyping and RTePCR experiments. C.H.L. thanks the support from Grant NSC 93-2214-E-182-007. References [1] Minguell JJ, Erices A, Conget P. Mesenchymal stem cells. Exp Biol Med 2001;226:507e20. [2] Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca R, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143e7. [3] Ye ZQ, Burkholder JK, Qui P, Schultz QJC, Shahidi NT, Yang NS. Establishment of an adherent cell feeder layer from human umbilical cord blood for support of long-term hematopoietic progenitor cell growth. Proc Natl Acad Sci USA 1994;91:12140e4. [4] Erices A, Cognet P, Minguell JJ. Mesenchymal progenitor cells in human umbilical cord blood. Br J Haematol 2000;109:235e42. [5] Gutierrez-Rodriguez M, Reyes-Maldonado E, Mayani H. Characterization of the adherent cells developed in Dexter-type long-term cultures from human umbilical cord blood. Stem Cells 2000;18:46e52. [6] Romanov YA, Svintsitskaya VA, Smirnov VN. Searching for alternative sources of postnatal human mesenchymal stem cells: candidate MSC-like cells from umbilical cord. Stem Cells 2003;21:105e10. [7] Campagnoli C, Roberts IAG, Kumar S, Bennett PR, Bellantuono I, Fisk NM. Identification of mesenchymal stem/progenitor cells in human first-trimester fetal blood, liver, and bone marrow. Blood 2001;38: 2396e402. [8] Ko¨gler G, Sensken S, Airey JA, Trapp T, Mu¨schen M, Feldhahn N, et al. A new human somatic stem cell from placental cord blood with intrinsic pluripotent differentiation potential. J Exp Med 2004;200:123e35. [9] Lee MW, Choi J, Yang MS, Moon YJ, Park JS, Kim HC, et al. Mesenchymal stem cells from cryopreserved human umbilical cord blood. Biochem Biophys Res Commun 2004;320:273e8. [10] Wexler SA, Donaldson C, Denning-Kendall P, Rice C, Bradley B, Hows JM. Adult bone marrow is a rich source of human mesenchymal ‘stem’ cells but umbilical cord and mobilized adult blood are not. Br J Haematol 2003;121:368e74. [11] Majumdar MK, Thiede MA, Mosca JD, Moorman M, Gerson SL. Phenotypic and functional comparison of cultures of marrow-derived mesenchymal stem cells and stromal cells. J Cell Physiol 1998;176:57e66. [12] Marshak DR, Holecek JJ. Chemical defined medium for human mesenchymal stem cells. United States Patent 5,908,782; 1999.
[13] Besse A, Trimoreau F, Praloran V, Denizot Y. Effect of cytokines and growth factors on the macrophage colony-stimulating factor secretion by human bone marrow stromal cells. Cytokine 2000;12:522e5. [14] Yen BL, Huang HI, Chien CC, Jui HY, Ko BS, Yao M, et al. Isolation of multipotent cells from human term placenta. Stem Cells 2005;23:3e9. [15] Huang X, Zhang Q. Coal-induced interleukin-6 gene expression is mediated through ERKs and p38 MAPK pathways. Toxicol Appl Pharmacol 2003;191:40e7. [16] Colter DC, Sekiya I, Prockop DJ. Identification of a subpopulation of rapidly self-renewing and multipotential adult stem cells in colonies of human marrow cells. Proc Natl Acad Sci USA 2001;98:7841e5. [17] Dennis JE, Merriam A, Awadallah A, Yoo JU, Johnstone B, Caplan AI. A quadripotential mesenchymal progenitor cell isolated from the marrow of an adult mouse. J Bone Miner Res 1999;14:700e9. [18] Klassen HJ, Imfeld KL, Kirov II, Tai LQ, Gage FH, Young MJ, Berman MA. Expression of cytokines by multipotent neural progenitor cells. Cytokine 2003;22:101e6. [19] Gadient RA, Otten UH. Interleukin-6 (IL-6) A molecule with both beneficial and destructive potentials. Prog Neurobiol 1997;52:379e90. [20] Hungness EA, Pritts TA, Luo GJ, Hershko DD, Rob BW, Hasselgren PO. IL-1b activates C/EBP-b and delta in human enterocytes through a mitogen-activated protein kinase signaling pathway. Int J Biochem Cell Biol 2002;34:382e95. [21] Cardinaux JR, Allaman I, Magistretti PJ. Pro-inflammatory cytokines induce the transcription factors C/EBP-b and C/EBP-d in astrocytes. Glia 2000;29:91e7. [22] Janderova´ L, McNeil M, Murrell AN, Mynatt RL, Smith SR. Human mesenchymal stem cells as an in vitro model for human adipogenesis. Obes Res 2003;11:65e74. [23] Potian JA, Aviv H, Ponzio NM, Harrison JS, Rameshwar P. Veto-like activity of mesenchymal stem cells: functional discrimination between cellular responses to alloantigens and recall antigens. J Immunol 2003; 171:3426e34. [24] Haynesworth SE, Baber MA, Caplan AI. Cytokine expression by human marrow-derived mesenchymal progenitor cells in vitro: Effects of dexamethasone and IL-1. J Cell Physiol 1996;166:585e92. [25] Lambert E, Dasse E, Haye B, Petitfrere E. TIMPs as multifacial proteins. Crit Rev Oncol Hematol 2004;49:187e98. [26] Mazzarelli P, Scuderi F, Mistretta G, Provenzano C, Bartoccioni E. Effect of transforming growth factor-b1 on interleukin-6 secretion in human myoblasts. J Neuroimmunol 1998;87:185e8. [27] Tsuji F, Oki K, Senda T, Horiuchi M, Mita S. Effects of mitogenactivated protein kinase inhibitors or phosphodiesterase inhibitors on interleukin-1-induced cytokines production in synovium-derived cells. Immunol Lett 1999;68:275e9. [28] Mracek T, Cannon B, Housteka J. IL-1 and LPS but not IL-6 inhibit differentiation and downregulate PPAR gamma in brown adipocytes. Cytokine 2004;26:9e15. [29] Chen CC, Manning AM. TGF-b1, IL-10 and IL-4 differentially modulate the cytokine-induced expression of IL-6 and IL-8 in human endothelial cells. Cytokine 1996;8:58e65. [30] Paulsen DF, Solursh M. Microtiter micromass cultures of limb-bud mesenchymal cells. In Vitro Cell Dev Biol 1988;24:38e47. [31] Vicennati V, Vottero A, Friedman C, Papanicolaou DA. Hormonal regulation of interleukin-6 production in human adipocytes. Int J Obes Relat Metab Disord 2002;26:905e11. [32] Gimble JM, Wanker F, Wang CS, Bass H, Wu X, Kelly K, et al. Regulation of bone marrow stromal cell differentiation by cytokines whose receptors share the gp130 protein. J Cell Biochem 1994;54: 122e33. [33] Panepucci RA, Siufi JLC, Silva WA, Proto-Siquiera R, Neder L, Orellana M, et al. Comparison of gene expression of umbilical cord vein and bone marrow-derived mesenchymal stem cells. Stem Cells 2004;22:1263e78. [34] Delikat SE, Galvani DW, Zuzel M. The metabolic effects of interleukin 1 on human bone marrow adipocytes. Cytokine 1995;7:338e43. [35] Delikat S, Harris RJ, Galvani DW. IL-1b inhibits adipocyte formation in human long-term bone marrow culture. Exp Hematol 1993;21: 31e7.
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279 [36] Reichlin S, Chen GJ, Nicolson M. Blood to brain transfer of leptin in normal and interleukin-1b-treated male rats. Endocrinology 2000;141: 1951e4. [37] Janik JE, Curti BD, Considine RV, Rager HC, Powers GC, Alvord G, et al. Interleukin 1 increases serum leptin concentrations in humans. J Clin Endocrinol Metab 1997;82:3084e6. [38] Grunfeld C, Zhao C, Fuller J, Pollock A, Moser A, Friedman J, et al. Endotoxin and cytokines induce expression of leptin, the ob gene product, in hamsters a role for leptin in the anorexia of infection. J Clin Invest 1996;97:2152e7.
279
[39] Rosen ED. The transcriptional basis of adipocyte development. Prostaglandins Leukotrienes Essent Fatty Acids 2005;73:31e4. [40] Qian H, Hausman GJ, Compton MM, Azain MJ, Hartzell DL, Baile CA. Down-regulation of CCAAT/enhancer binding proteins in adipose tissue by intracerebro-ventricular leptin in rats. Biochim Biophys Acta 1998;1442:245e51. [41] Mitchell SE, Rees WD, Hardie LJ, Hoggard N, Tadayyon M, Arch JRS, et al. ob gene expression and secretion of leptin following differentiation of rat preadipocytes to adipocytes in primary culture. Biochem Biophys Res Commun 1997;230:360e4.
Cytokine interactions in mesenchymal stem cells from cord blood Chi-Hsien Liu a,b,*, Shiaw-Min Hwang b,* b
a Graduate Institute of Biochemical and Biomedical Engineering, Chang Gung University, Taoyuan, Taiwan Bioresource Collection and Research Center, Food Industry Research and Development Institute, Hsinchu, Taiwan
Received 18 February 2005; received in revised form 21 June 2005; accepted 4 November 2005
Abstract We used cytokine protein array to analyze the expression of cytokines from human cord blood-derived mesenchymal stem cells (CB-MSCs). Several cytokines, interleukins (IL), and growth factors, including ENA-78, GM-CSF, GRO, IL-1b, IL-6, IL-8, MCP-1, OSM, VEGF, FGF-4, FGF-7, FGF-9, GCP-2, IGFBP-1, IGFBP-2, IGFBP-3, IGFBP-4, IP-10, LIF, MIF, MIP-3a, osteoprotegerin, PARC, PIGF, TGF-b2, TGF-b3, TIMP-1, as well as TIMP-2, were secreted by CB-MSCs, while IL-4, IL-5, IL-7, IL-13, TGF-b1, TNF-a, and TNF-b were not expressed under normal growth conditions. IL-6, IL-8, TIMP-1, and TIMP-2 were the most abundant interleukins expressed by CB-MSCs. A set of growth factors were selected to evaluate their stimulatory effects on the IL6 secretion for CB-MSCs. IL-1b was the most important factor inducing CB-MSC to secret IL-6. The mechanism by which IL-1b promoted IL-6 expression in CB-MSCs was studied. By using various inhibitors of signal transduction, we found that activation of p38 mitogen-activated protein kinases (MAPK) and MAPK kinase (MEK) is essential in the IL-1b stimulated signaling cascade which leads to the increase in IL-6 synthesis. Additionally, continuous supplement of IL-1b in the CBMSCs culture will facilitate adipogenic maturation of CB-MSCs as evidenced by the presence of oil drops in the CB-MSCs and secretion of leptin, a molecule marker of adipocytes. These results strongly suggest that cytokine induction and signal transduction are important for the differentiation of CB-MSCs. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Mesenchymal stem cell; Cord blood; Cytokine; Interleukin-1b; Interleukin-6
1. Introduction Mesenchymal stem cells (MSCs) are capable of self-renewal and differentiation into lineages of mesenchymal tissues, including bone, cartilage, fat, tendon, muscle, and hematopoieticsupporting stroma [1]. Although the existence of hematopoietic Abbreviations: MCP, monocyte chemotactic protein; IGFBP, insulin-like growth factor binding protein; ENA-78, epithelial neutrophil-activating protein 78; GRO, growth-related oncogene; OSM, oncostatin M; FGF, fibroblast growth factor; GCP, granulocyte chemotactic protein; IP, interferon-inducible protein; MIF, macrophage migration inhibitory factor; MIP, macrophage inflammatory protein; PARC, pulmonary and activation-regulated chemokine; PIGF, phosphatidylinositol glycan complementation class F; TGF, transforming growth factor; TIMP, tissue inhibitors of matrix metalloproteinases. * Corresponding authors. Graduate Institute of Biochemical and Biomedical Engineering, Chang Gung University, 259 Wen-Hwa first Rd, Kwei-Shan, TaoYuan 333, Taiwan. Tel.: þ886 3 211 8800 3146; fax: þ886 3 211 8700. E-mail addresses: [email protected] (C.-H. Liu), [email protected] (S.-M. Hwang). 1043-4666/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.cyto.2005.11.003
stem cells in cord blood (CB) is well documented, research on MSCs in cord blood is still in progress. Several researchers successfully isolated, expanded, and characterized the MSCs from umbilical cord blood and evaluated their potential for differentiation into osteogenic, chondrogenic or adipogenic lineage [2e 6]. The circulation of pluripotent MSCs in the blood of preterm fetus between liver and bone marrow explains why MSCs can be isolated from cord blood [6]. While the most abundant source of MSCs is bone marrow [7], MSCs from umbilical cord blood can serve as an alternative for cell-based therapies for the following reasons including the significant reduction in number of MSCs with age and the high risk of viral contamination in bone marrow. Although MSCs from placental and umbilical cord blood can be successfully isolated and expanded [8,9], the fact that CB-MSCs are present at a very low frequency [10] may influence their autologous transplantation and other clinical applications. This limitation needs to be further solved by well-designed isolation and expansion procedures.
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
MSCs in the bone marrow (BM) are involved in hematopoiesis, bone formation, bone resorption, and the development of cartilage and fat. From some of these events, BM-MSCs exert their influence through secreting soluble proteins. These macromolecules provide molecular signals to neighboring cells and modulate their mitotic, metabolic, and/or development states. Moreover, MSCs respond to components of the environmental milieu, including autocrine and paracrine factors synthesized by different cells, resulting in modulation of their own mitotic, metabolic, and/or developmental activity [11]. However, the cytokine expression profile of CB-MSCs and the influence of cytokine on the differentiation of CB-MSCs remains poorly documented. Utilization of this serum-free medium will facilitate analysis of interactions between growth factors and cytokines on the proliferation and differentiation of CB-MSCs without the complexity of exogenous serum. The serum-free medium for MSCs was reported and the differentiation potential of adipogenic, osteogenic, and chondrogenic lineages of MSCs was maintained under the serum-free medium [12]. In addition, serum-free medium was used to evaluate the effect of cytokines and growth factors on macrophage colonystimulating factor secretion by human bone marrow stromal cells [13]. In this study, we used commercial human cytokine protein array to identify protein expression profiles of CBMSCs under serum-free culture conditions. The study aimed at characterizing the secreted cytokine profile of CB-MSCs, evaluating the effect of cytokines on IL-6 release of CBMSCs, and determining how their differentiation process was altered when CB-MSCs were exposed to exogenous IL-1b.
2. Materials and methods
271
medium (IMDM þ 20% FBS þ 10 ng/ml basic fibroblast growth factor (bFGF). Before performed the experiment under serum-free conditions, confluent CB-MSCs were digested with 0.25% trypsin-EDTA (JRH Biosciences, KS) from 20% FBS supplemented medium and seeded at the density of 105 cells/ml to serum-free medium. After CB-MSCs were confluent in T75 flasks under the serum-free condition, CB-MSCs were detached with 1 ml of NO-ZYME (JRH Biosciences) and seeded in the six-well plate at the density of 105 cells/well for the following experiments, including effects of cytokines on IL-6 expression and multi-lineage differentiation. 2.2. Effects of cytokines on IL-6 expression For screen of different cytokines on IL-6 expression in CBMSCs, the concentration of 10 ng/ml of different cytokines was added in the serum-free culture medium. The initial seeding density was 105 cells per well in 6-well culture dishes, and the cell-free conditioned medium was analyzed by ELISA method after a 3-day culture. In order to assay the kinase inhibitor effects, 105 cells were plated in 35-mm 6-well culture dishes under serum-free condition after 3 days of growth. Medium was then changed to fresh IMDM medium and cells were treated with freshly prepared kinase inhibitors (PD98059 and SB202190) at various doses for 1 h. The range of concentration tested in the evaluation procedure was 0e50 mM of PD98059, 0e50 mM of SB202190 and 10 ng/ml of IL-1b according to the preliminary experiments. After removal of medium, fresh serum-free medium and/or IL-1b were added and cells were cultured for 24 h according the protocol presented by Huang and Zhang [15]. The conditioned medium was collected and stored at 80 C for further cytokine analysis.
2.1. Isolation and culture of CB-MSCs CB-MSCs were isolated and cultured according to modifications of a previous reported protocol [14]. Briefly, the term cord blood of newborns was layered onto FicollePaque solution (1.077 g/ml, Amersham, Uppsala, Sweden) and centrifuged to deplete the residues of red blood cells, platelets and plasma (700 g for 40 min). Mononuclear cells in the interface were isolated, and seeded at the concentration of 106 cells/cm2 in the serum-containing medium (SCM, IMDM þ 20% FBS (Hyclone, Logan, UT) þ 10 ng/ml basic fibroblast growth factor (bFGF)) at 37 C for 2e3 weeks. The well-developed colonies of fibroblast-like cells were collected and preserved in liquid nitrogen for further experiments. Our laboratory routinely isolated MSCs from cord blood. MSCs can be successfully isolated in 108 units of cord blood out of a total of 144 donated samples. The efficacy of CB-MSCs isolation is 75% (n ¼ 144). Because these isolated MSCs have similar morphology, differentiation potentials, growth characteristics, and the same immunophenotypic markers (CD26, CD31, CD34, CD45, HLA-DR, CD29þ, CD44þ, HLA-ABCþ, SH2þ, SH3þ, SH4þ, Fig. 1), one representative stain of CB-MSCs, CBMSC no. 354, was used as a typical example in our experiments. CB-MSCs were routinely maintained in serum-containing
2.3. Evaluation of differentiation To test chronic supplement of IL-1b on the differentiation of CB-MSCs, serum-free medium supplemented with 10 ng/ml of IL-1b was replaced every 3e4 days for a total of 45 days. The conditioned serum-free medium was collected and preserved at a 80 C refrigerator for further IL-6 and leptin assay. At term, the cultured cells were washed twice with PBS, fixed in 10% formaldehyde for histochemical staining. Differentiation of adipocytes, chondrocytes and osteocytes were evaluated by the appearance of Oil-red-O stained lipid vacuole [16], the Safranin-O stained red accumulation of sulfated proteoglycans [16] and von Kossa method stained mineralization of calcium nodules [17], respectively. 2.4. Immunophenotyping of CB-MSCs After CB-MSCs expanded in SFM for two passages, the surface markers were analyzed using a FACSCalibur flow cytometry system (BD Biosciences, Canada). CB-MSCs (106 cells) were fixed with 70% ethanol (10 min at 4 C). Antibodies against human antigens CD26, CD31, CD34, CD45,
272
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
Fig. 1. Immunophenotyping profile of CB-MSCs cultured in serum-containing or serum-free media.
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
HLA-DR, CD29, CD44 and HLA-ABC were purchased from Becton Dickinson (San Diego, http://www.bd.com). Antibodies against human antigens SH-2, SH-3 and SH-4 were purified from the respective hybridoma cell lines acquired from ATCC. The fixed cells were stained with fluorescein isothiocyanate (FITC) or phycoetrythrin (PE)-conjugated antibodies. A replicate sample was stained with FITC-mouse IgG1 and PE-mouse IgG1 as an isotype control to ensure specificity. The control CB-MSCs were maintained in serumcontaining medium. Their surface marker pattern, including CD26, CD31, CD34, CD45, HLA-DR, CD29þ, CD44þ, HLA-ABCþ, SH2þ, SH3þand SH4þ, was the same as CB-MSCs cultured in serum-containing medium (Fig. 1). 2.5. Antibody-based protein array system Cell-free supernatant was taken from conditioned serumfree medium of 4-day cultured CB-MSCs, and then analyzed by RayBio Human Cytokine Array V kit according to the manufacturer’s instructions. Human Cytokine Array V kit was purchased from RayBiotech (catalog no. H0108005A; Norcross, GA). The array membranes can detect 79 different growth factors/cytokines at a time. The layout of the membrane is listed in Fig. 2. Positive controls are located in the upper left-hand corner (four spots) and lower right-hand corner (two spots) of each membrane. The assay was followed precisely as stated in the directions from the manufacturer. In brief, place each membrane into the provided eight-well tray. Add 2 ml blocking buffer and incubate at room temperature for 30 min to block membranes. Decant blocking buffer from each container, and incubate membranes with 1 ml of conditioned medium at room temperature for 1e2 h. Decant the samples from each container, and wash three times with 2 ml of wash buffer I at room temperature with shaking for 5 min. Wash two times with 2 ml of wash buffer II at room temperature with shaking for 5 min. Add 1 ml of 250 fold diluted biotin-conjugated antibodies to each membrane. Incubate at room temperature for 1e2 h. After washing two times, 12000
IL-6 (pg/mL)
10000 8000 6000 4000 2000
273
add 2 ml of 1000 fold diluted HRP-conjugated streptavidin to each membrane. Incubate at room temperature for 30 min and wash two times. Place membrane into the detection buffer and incubate at room temperature for 5 min. Drain off excess detection reagent and wrap the membrane with PE wrap. The membrane was then exposed to Hyperfilm (Amersham Bioscience, catalog no. RPN2103K). Detectable spots were scanned by densitometer, and detection with fresh culture media was subtracted from the test samples. Positive controls, provided by the manufacturer, were normalized to 1-fold, and the unknown samples were calculated from the normalized spots. 2.6. ELISA IL-6 and leptin were quantified by using the Flexia antibody pair system (BioSource, Camarillo, CA.) and DuoSet ELISA Development System (R&D Systems, Minneapolis, MN), respectively. The assay procedures were performed according to the instructions of the manufacturer. A solution of 3,3#,5,5$-tetramethylbenzidine and H2O2 (Kirkeguard and Perry Laboratories Inc.) was added as the peroxidase substrate. Finally, 1 M phosphoric acid was added to terminate the reaction and the plates were measured at a wavelength of 450 nm. Each sample was tested in duplicate. 2.7. Growth factors, culture medium and chemicals The following recombinant human growth factors or cytokines were used, purchased from PeproTech (Rocky Hill, NJ): transforming growth factor-b1 (TGF-b1); TGF-b2; TGF-b3; bFGF; vascular endothelial growth factor (VEGF); leukemia inhibitory factor (LIF); thrombopoietin (TPO); stem cell factor (SCF); stem cell growth factor-a (SCGF-a); hepatocyte growth factor (HGF); Flt-3 ligand (FL); interleukin (IL)-1b; IL-3; granulocyte colony-stimulating factor (G-CSF); granulocyte-macrophage colony-stimulating factor (GM-CSF). IL-6 and IL-6 sR (soluble receptor) were purchased from R&D Systems. Iscove’s modified Dulbecco’s medium (IMDM, catalog no. I7633) was from Sigma. Human albumin (25%) was from Aventis Behring (Kankakee, IL). Hydrocortisone and SITE (containing 0.5 g/ml sodium selenite, 1 mg/ml insulin, 0.55 mg/ml transferrin and 0.2 mg/ml ethanolamine) were purchased from Sigma. Tissue culture grade T25 and T75 cm2 flasks and 6-well plates were purchased from Corning Costar (Cambridge, MA). PD 98059 and SB 230580 were obtained from CalbiochemeNovabiochem Co. and dissolved in DMSO (Sigma) as a stock at a concentration of 50 mM.
Co nt ro SC l G LI F- F al p V ha EG F G HG M F -C SF SC F IL 3 TP G O -C SF IL FL TG 6F SR TG -be F ta1 TG -be F- ta2 be ta bF 3 IL IL- GF -1 1 be bet ta a +H C
0
Fig. 2. Effects of cytokines on IL-6 secretion by CB-MSCs. The concentration of 10 ng/ml of different cytokines was added in the serum-free medium. The initial seeding density was 105 cells per well in 6-well culture dishes and the cell-free conditioned medium was analyzed by ELISA method after 3 days of culture. Results are expressed as means of triplicate samples and the bar indicates standard deviation.
2.8. Detection of PPARg2 and C/EBPb in CB-MSCs by RT-PCR Total RNAs were extracted from cultured human cells using Trizol (Gibco) according to the manufacturer’s protocol. Genomic DNA contamination was removed by incubation with 10 units of DNAse 1 in solution of 10 mM TriseCl, pH 8.3, 50 mM KCl, and 1.5 mM MgCl2 for 30 min at 37 C.
274
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
Ten micrograms of total RNA from CB-MSCs, were reverse transcribed using 20 units of AMV reverse transcriptase. PCR amplification was performed on 2 ml first-strand cDNA with 0.5 mM of primers and 2.5 units Taq DNA Polymerase, 2 mM dNTP, and 1.5 mM MgCl2. Oligonucleotide primers used for PPARg2, C/EBPb, and b-actin were 5#-CCAGAAAATGACAGACCTCAGACA-3# (forward), 5#-GCAGGAGCGGGTGAAGACT-#’ (reverse); 5#-CCGGGCAGCACCACGACTTCCTCT-3# (forward), 5#-CTTCTTGGCCTTGCTCTTGACCTG-3# (reverse); and 5#-AGACGTATCACCTCTGCAC-3# (forward), 5#-GGAAGCAACGTCTGTGAGGT-3# (reverse), respectively. The PCR products were electrophoresed through 1% agarose gel and the corresponding expected sizes of DNA bands were exercised and purified for sequence confirmation. 2.9. Statistical analysis The data was analyzed by paired samples t-test, and a value of P < 0.05 was considered significant. 3. Results 3.1. Cytokine array analysis of conditioned medium of CB-MSCs from cord blood The conditioned medium after a 4-day culture of MSC from cord blood was assayed using Raybiotech cytokine array. CB-MSCs spontaneously secrete twenty-eight cytokines in their culture supernatants as indicated in the cytokine array (Table 1). Twenty-eight cytokines belonging to the chemokine family and those associated with cell proliferation and immunity were detected from the array, which can analyze 79 growth factors/cytokines at a time. These secreted cytokines can be classified into the following groups. Chemokines: GRO, IL-8, MCP-1, MIP-3a, PARC, IP10, ENA-78, GCP-2, osteoprotegerin; Broad-acting cytokines: IL-1b, IL-6; Antiinflammatory: TGF-b2, TGF-b3, MIF, LIF; Growth factor: GM-CSF, VEGF, FGF-4, FGF-7, FGF-9, PIGF; Growth factor receptor: IGFBP-1, IGFBP-2, IGFBP-3, IGFBP-4; Angiogenic factor: oncostatin M; Proteinase inhibitor: TIMP-1, TIMP-2. Among them, IL-6, IL-8, TIMP-1 and TIMP-2 are the most abundant cytokines secreted by CB-MSCs. MCP-1, IGFBP-4, LIF, and VEGF are also plentifully secreted by CB-MSCs. 3.2. Effects of cytokines on IL-6 secretion IL-6 is one of the most abundantly expressed cytokines by CB-MSCs. Accumulating evidence supports an essential role of IL-6 in the development, differentiation, regeneration and degeneration of different stem cells [18]. IL-6 is also involved in tissue remodeling through its effects on connective tissue cells, proteases and protease inhibitors [19]. The up-regulation effect of different cytokines on the expression IL-6 by CBMSCs was evaluated. First, growth factors and cytokines were used to evaluate their effects on the IL-6 secretion. These tested growth factors and cytokines include LIF, VEGF, SCGF-a, HGF, GM-CSF, SCF, IL-3, TPO, G-CSF, FL, IL-6-SR,
Table 1 Cytokine array result with supernatants from CB-MSCs Cytokines
Fold increase
Chemokines GRO,MIP-3 PARC, IP10, ENA78, GCP MCP-1, Osteoprotegerin IL-8
0.33 0.66 1 2
Broad-acting cytokines IL-1b IL-6
0.33 2
Anti-inflammatory cytokines TGF-b3 MIF TGF-b2, LIF
0.33 0.66 1
Growth factors FGF-4, FGF-7, FGF-9 GM-CSF, PIGF VEGF
0.33 0.66 1
Growth factor receptors IGFBP-1, IGFBP-2 IGFBP-3, IGFBP-4
0.66 1
Protease inhibitors TIMP-1, TIMP-2
2
Angiogenic factor OSM
0.66
Cell-free supernatants were taken from confluent CB-MSCs and then analyzed by microarray. Results are the average of three samples. Detectable spots were scanned by densitometer, and detection with fresh culture media was subtracted from the test samples. Positive controls, provided by the manufacturer, were normalized to 1-fold, and the unknowns calculated accordingly. Details of the procedure are described in Section 2.
TGF-b1, TGF-b2, TGF-b3, bFGF, IL-1b, and hydrocortisone. Among these tested compounds, IL-1b, TGF-b2, TGF-b3, LIF, VEGF, and GM-CSF were also constitutively expressed by CB-MSCs as indicated in the previous cytokine array analysis. These compounds were used to check whether the autocrine effects exist in IL-6 secretion of CB-MSCs. In addition, bFGF and hydrocortisone were supplemented in the serumfree medium. The IL-1b was the most important factor that induces the MSC to secret IL-6. TGF b1 and TGFb2, but not TGFb3, can also induce the MSC to secret IL-6 (Fig. 1). Simultaneous addition of 10 ng/ml of IL-1b and 55.3 mM of hydrocortisone significantly reduced the IL-6 expression of CB-MSCs. Hydrocortisone reduced the stimulatory effects of IL-1b on IL-6 production by CB-MSCs. Basic FGF, supplemented in our serum-free medium, did not stimulate CBMSCs to express IL-6. The other tested cytokines, including LIF, VEGF, SCGF-a, HGF, GM-CSF, SCF, IL-3, TPO, G-CSF, FL, IL-6-SR, TGF-b3, and bFGF, have no stimulation effects on IL-6 production in CB-MSCs at the tested concentration (10 ng/ml) and higher concentration (50 ng/ml, data not shown). 3.3. Effects of IL-1b on IL-6 secretion IL-1b increased IL-6 secretion by CB-MSCs in a dosedependent manner. Fig. 3A illustrates the response of
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
(A)
275
14000 14000
*
12000
IL-6 (pg/mL)
IL-6 (pg/mL)
12000 10000 8000 6000
*
*
6000 4000
2000
2000 0 0
5
10
15
20
25
30
35
IL-1 beta (ng/mL)
10000 IL-1b
8000
IL-1 SB 202190 PD 98059
10ng/mL -
10ng/mL 50uM -
10ng/mL 50µM
10ng/mL 50µM 50µM
Fig. 4. Effects of kinase inhibitors (PD98059 and SB202190) on IL-6 secretion by CB-MSCs. The cultured CB-MSCs were pretreated with kinase inhibitor for 1 h and then stimulated by 10 ng/ml IL-1b for 24 h. The data were average of triplicate experiments. *P < 0.05, compared with the value of control. e, no addition.
14000 12000
IL-6 (pg/mL)
8000
4000
0
(B)
10000
control
6000 4000 2000 0 0
10
20
30
40
50
60
inhibited IL-1b-induced IL-6 production (about 14% inhibition). Our results indicate that IL-1b stimulated CB-MSCs to express IL-6 require the activation of p38 MAPK and ERK pathways.
Hour Fig. 3. (A) Effect of IL-1b dosage on IL-6 secretion in CB-MSCs. Different dosages of IL-1b were added in the serum-free medium. The initial seeding density was 105 cells per well in 6-well culture dishes and the IL-6 in cellfree conditioned medium was analyzed by ELISA method after 3 days of culture. Results are expressed as means of triplicate samples and the bar indicates standard deviation. (B) Effects of culture time on IL-1b induced IL-6 secretion in CB-MSCs. Dosage of 10 ng/ml IL-1b was added in the serum-free culture medium. The initial seeding density was 105 cells per well in 6-well culture dishes and the cell-free conditioned medium was collected at different culture times, and then the IL-6 in conditioned medium was analyzed by ELISA method. Results are expressed as means of triplicate samples and the bar indicates standard deviation.
CB-MSCs to the increasing concentrations of IL-1b in culture medium between 0e30 ng/ml. Additional increases in IL-1b (above 10 ng/ml) to the culture medium generate saturate production of 12,000 pg/ml IL-6. These data were used to choose the optimal concentration of IL-1b to add to the culture medium in the subsequent experiments. Furthermore, IL-1b increased the expression of IL-6 secretion by CB-MSCs in a timedependent manner (Fig. 3B). The IL-6 secretion by CBMSCs reaches the saturate concentration after 24 h culture. Signaling of IL-1b was reported to be initiated by IL-1 receptor, which triggered the mitogen-activated protein kinase (MAPK) cascade to transduce the signal into the nucleus. To assess the role of MAPK pathways in IL-1b-mediated activation of IL-6 expression, two inhibitors (SB202190, PD98059) were used in cells treated with cytokine (Fig. 4). Kinase inhibitor SB202190 (a p38 MAPK inhibitor) and PD98059 (a MAPK kinase (MEK) inhibitor) inhibited the expression of IL-6 induced by IL-1b. However, inhibitor SB202190 had strong inhibition effects on IL-1b induced IL-6 secretion than inhibitor PD98059. We found that the expression in IL-6 protein by IL-1b was eliminated 50% by the pretreatment of both cell types with SB202190. Moreover, PD98059 partially
3.4. Effects of chronic IL-1b supplement on CB-MSCs Initial addition of IL-1b in the medium stimulated the expression level of IL-6. However, the expression of IL-6 decreased gradually when IL-1b was continuously added in the medium (Fig. 5A). Furthermore, consecutive supplement of IL-1b can promote adipogenic maturation of CB-MSCs as evidenced by the presence of oil drops in the CB-MSCs after 45 days of culture (Fig. 6A). The control CB-MSCs were free of oil drops under the culture condition without IL-1b supplement (Fig. 6B). The level of leptin in the conditioned medium was continuously increasing after 17 days of culture (Fig. 5B). The evidence of oil drops in the CB-MSCs by Oil-Red-O stain and the secretion of leptin which is the differentiation marker of adipocytes, support that supplement of IL-1b in vitro can facilitate terminal adipogenic maturation of CB-MSCs. No osteocytes and chondrocytes were existed in these CB-MSC samples as verified by the histochemical stain (data not shown). The mechanism of IL-1b on the commitment of adipogenic differentiation of CB-MSCs and other marker for adipogenic differentiation of CB-MSCs were further investigated. The CCAAT/enhancer binding proteins (C/EBP) of transcription factors were stimulated by IL-1b in human enterocytes [20] and in mouse astrocytes [21]. Peroxisome proliferatoractivated receptors (PPARs) are a transcription factor family related to the induction and differentiation from primitive precursor cells to the terminally differentiated adipocytes [22]. Therefore, we analyzed expression of PPARg2 and C/EBPb by RTePCR in different CB-MSC samples after 21 days of culture. However, PPARg2 and C/EBPb are present in control (cultured serum-free and serum-containing media) and IL-1b treated CB-MSCs (Fig. 7). The existence of PPARg2 and C/EBPb in CB-MSCs might be due to the tendency of CB-MSCs would spontaneously differentiate into
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
276
(A)14000 IL-1b
IL-6 (pg/mL)
12000
control
10000 8000 6000 4000 2000 0 0
5
10
15
20
25
30
35
40
45
50
30
35
40
45
50
Day
(B)
8000 IL-1b
Leptin (pg/mL)
7000
control
6000 5000 4000 3000 2000 1000 0
0
5
10
15
20
25
Day Fig. 5. (A) Effects of consecutive treatment of IL-1b on the secretion of IL-6 by CB-MSCs. (B) Effects of consecutive treatment of IL-1b on the secretion of Leptin by CB-MSCs. The initial seeding density was 105 cells per well in 6-well culture dishes and serum-free medium supplemented with 10 ng/ml of IL-1b was replaced every 3e4 days for 45 days. The conditioned serumfree medium was collected and preserved at 80 C for a further IL-6 and leptin assay. Results are expressed as means of triplicate samples and the bar indicates standard deviation.
adipocytes (although in a low percentage) and the sensitivity of RT-PCR method. 4. Discussion Previous results indicate that CB-MSCs display many common features with BM-MSCs, like adhesion to plastic, morphology, expression of similar surface markers, and a potential to differentiate into osteo-, chondro-, and adipogenic
phenotypes. Potian et al. recently analyzed the cytokine profile of BM-MSCs by the protein array technique [23]. IL-6, IL-8, MCP-1, RANTES, GRO-a, IFNg, IL-1a, TGF-b, GM-CSF, angiogenin, and oncostatin M were constitutively expressed and MIP-1b, IL-2, IL-4, IL-10, IL-12, IL-13 were not expressed by the BM-MSCs [23]. The cytokine profile of BM-MSCs is very similar to that of CB-MSCs in this study, except that CB-MSCs express IL-12 and do not express G-CSF under serum-free condition. Haynesworth et al. reported that constitutively expressed cytokines in this growth phase include IL-6, G-CSF, SCF, LIF, M-CSF, and IL-11, while GM-CSF, IL-3, TGF-2, and OSM were not detected in the growth medium of human bone marrow-derived MSCs [24]. Both MSCs from bone marrow and cord blood abundantly produced IL-6, IL-8, and MCP-1. Interestingly, G-CSF expression of CB-MSCs is distinct from that of BM-MSCs. G-CSF can be detected in the conditioned medium of BM-MSCs, but not in that of CB-MSCs. Besides, tissue inhibitors of metalloproteinase (TIMP)-1 and TIMP-2 were found to be plentifully expressed by CB-MSCs in this study. TIMPs are natural inhibitors of matrix metalloproteinases found in most tissues and body fluids. By inhibiting MMPs activities, they participate in tissue remodeling of the extracellular matrix (ECM). The balance between MMPs and TIMPs activities is involved in tissue remodeling, angiogenesis, and hematopoiesis, as reviewed recently [25]. The implication of TIMP-1 and TIMP-2 in physiological and differentiation processes of MSCs merits further investigation. These results showed that constitutive expression of these growth factors could be related to provide inductive and regulatory information, which is consistent with the ability to support hematopoiesis, and also supply autocrine/ paracrine factors that influence hemopoietic functions, angiogenic properties, and immune regulation. Although the real physiological functions of cytokine expression by CB-MSCs remain to be elucidated, the developed serum-free culture system provides a simple and defined environment to study these cytokine interactions without the complexity of serum. IL-6 is another abundant cytokines secreted by CB-MSCs. Accumulating evidence supports an essential role of IL-6 in the development, differentiation, regeneration and degeneration
Fig. 6. Oil-red-O stain of CB-MSCs in 10 ng/ml of IL-1b supplemented serum-free medium after 45-day culture. The initial seeding density was 105 cells per well in 6-well culture dishes and serum-free medium supplemented with 10 ng/ml of IL-1b was replaced every 3e4 days for 45 days. The CB-MSCs were stained to check the existence of lipid vacuoles by Oil red O stain. Details of the procedure are described in Section 2.
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
Fig. 7. Expression of PPARg2 and C/EBPb in CB-MSCs as determined by RT-PCR. Lane 1: CB-MSCs cultured in serum-free medium plus 10 ng/ml after 21 days of culture. Lane 2: CB-MSCs cultured in serum-free medium after 21 days of culture. Lane 3: CB-MSCs cultured in serum containing medium after 21 days of culture. Lane 4: a negative control.
of different stem cells [18]. IL-6 functions are mediated by a specific receptor system composed of a binding site and a signal transducer. IL-1b is a major mediator of microbial invasion, inflammation, immunological reactions and tissue injury. IL-1b can stimulate human myoblasts [26], human synoviocyte or synovial sarcoma [27], and mouse brown adipocytes [28] to increase expression of IL-6. Furthermore, IL-1a can induce BM-MSCs to increase the expression of G-CSF, M-CSF, GM-CSF, LIF, IL-6, and IL-11 [24]. Both IL-1 and IL-6 are multifunctional cytokines that play an important role as regulators of the immune response and homeostasis processes [29]. The biological relevance of IL-1 and IL-6 in BM-MSCs was studied by several authors [11,24]. Their results suggested that MSCs might be able to influence protective reactions locally as well as at a distance from the bone marrow by expression of their cytokines and growth factor, such as IL-1 and IL-6, during inflammation and/or infection. Hydrocortisone was included in a defined medium to support a high-density culture of chick and mouse limb mesenchymal cells [30]. We also found that hydrocortisone can stimulate the proliferation of CB-MSCs under serum-free condition (data not shown). So our serum-free medium contained hydrocortisone. Glucocorticoids, such as hydrocortisone and dexamesathone, were reported to suppress release of IL-6 and stimulate leptin secretion by human adipocytes in a dose-dependent manner [31]. Hydrocortisone significantly reduced the stimulatory effects of IL-1b on IL-6 production by CB-MSCs (Fig. 2). From these studies, the decreased production of IL-6 during the hMSCs adipo-differentiation process was due to the effects of hydrocortisone. Furthermore, IL-6 was reported to inhibit hydrocortisone-induced adipocyte differentiation of murine stromal cells in a dose-dependent manner [32]. The gradually decreased IL-6 expression might reduce the inhibition effect of IL-6 on adipo-differentiation of CB-MSCs. CB-MSCs were similar to BM-MSCs in morphological features, immunophenotypic markers, and differentiating abilities; however genes related to antimicrobial activity, osteogenesis, matrix remodeling, and angiogenesis are differently expressed in CB-and BM-MSCs [33]. Moreover, CB-MSCs have longer telomeres, which may explain the high expansion capacity of CB-MSCs than that of BM-MSCs [9]. Further differences include absent expression of certain adhesion molecules such as ICAM-3, L-selectin, and VCAM, as well
277
as HAS1 in CB-MSCs. Additionally, CB-MSCs have a much broader differentiation potential and differ in their RNA expression profile as previously reported [8]. Delikat et al. found that addition of IL-1b to human long-term bone marrow culture (LTBMC) inhibited adipocyte formation in serumsupplemented conditions [34,35]. In contrast, we demonstrated that continuous supplementation of IL-1b in serum-free culture conditions promotes adipogenic maturation. The difference response of these two kinds of MSCs to IL-1b might be caused by culture condition (defined and serum-free vs. undefined and 25% serum supplemented) and the isolation sources (cord blood vs. bone marrow). Furthermore, previous studies demonstrated that administration of recombinant IL-1b could increase the concentration of serum leptin in rat [36] as well as human [37] and increase levels of leptin mRNA in adipose tissue of fasted hamster [38]. Our in vitro data also demonstrated that supplement of IL-1b in the CB-MSCs culture will induce CB-MSCs to secret leptin. Adipogenesis is a complex process including proliferation of precursor cells, their commitment to the adipogenic lineage, and terminal differentiation [22]. The early events in adipogenic commitment include the activation of cascade of transcription factors as proposed by Rosen [39]. In brief, induction of C/EBPb and d leads to increase expression of C/EBPa and PPARg, and these two transcription factors trigger the expression of genes important for storage of lipid and secretion of leptin, adipsin, and adiponectin. Hence C/EBPs and PPARs were transcription factors related to the induction and differentiation from primitive precursor cells to the terminally differentiated adipocytes [22]. Besides, C/EBPs were involved in tissue-specific metabolic gene transcription, in signal transduction activated by several cytokines, and in terminal differentiation of a variety of cells including adipocytes [40] and myelomonocytes [21]. IL-1b was reported to strongly stimulate the expression of the C/EBP isoforms in human enterocytes [20] and in mouse astrocytes [21]. Although the terminal differentiation of adipocytes is intensively studied and well described in the murine cell lines, key information regarding the commitment of human precursor cells to the adipogenic lineage is still lacking. Since human MSCs can differentiate into multiple mesenchymal lineages, they were reported to provide a unique model to better understand early differentiation events [22]. We analyzed PPARg2 and C/EBPb by RT-PCR in our CB-MSC samples under different culture conditions (Fig. 7). PPARg2 and C/EBPb are present in control (cultured serum-free and serum-containing media) and IL-1b treated CB-MSCs. The presence of PPARg in uncommitted and adipo-differentiation MSCs is consistent with the recent report of Janderova´ et al. [22]. In BM-MSCs, C/EBPb protein was increased transiently within 24 h and greatly reduced at the end of the 21-day adipo-differentiation culture [22]. On contrary, mRNA of C/EBPb was still detected after 21 days of culture in CB-MSCs. The existence of PPARg2 and C/EBPb in CB-MSCs may be due to the tendency of CB-MSCs would spontaneously differentiate into adipocytes (although in a low percentage) and the sensitivity of RT-PCR method. Therefore, our RT-PCR results cannot
278
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279
specify which stage IL-1b is involved in the early event of adipogenic commitment. Leptin is an important differentiation marker of preadipocytes and adipocytes as emphasized by many authors [31,40,41]. The accumulation of lipid-vacuole and the secretion of leptin in our study strongly support that IL-1b can facilitate the terminal adipogenic differentiation and adipocyte maturation process in CB-MSCs. However, mechanisms of IL-1b on the commitment of adipogenic differentiation of CB-MSCs remain to be determined. Further experiments must show whether the effect of IL-1b on adipo-lineage differentiation of CB-MSCs is influenced by the existence of serum. Overall, the present study provides an initial characterization of cytokine expression by CBMSCs and highlights the important role played by the cytokine interaction in the differentiation of CB-MSCs. Acknowledgments This work was supported by the Foundation of Research and Development from FIRDI (03A006) and the Ministry of Economic Affairs, Taiwan (93-EC-17-A-17-R7-0525). We thank Drs. Chao-Ling Yao and Yu-Jen Chang for assistance in immunophenotyping and RTePCR experiments. C.H.L. thanks the support from Grant NSC 93-2214-E-182-007. References [1] Minguell JJ, Erices A, Conget P. Mesenchymal stem cells. Exp Biol Med 2001;226:507e20. [2] Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca R, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143e7. [3] Ye ZQ, Burkholder JK, Qui P, Schultz QJC, Shahidi NT, Yang NS. Establishment of an adherent cell feeder layer from human umbilical cord blood for support of long-term hematopoietic progenitor cell growth. Proc Natl Acad Sci USA 1994;91:12140e4. [4] Erices A, Cognet P, Minguell JJ. Mesenchymal progenitor cells in human umbilical cord blood. Br J Haematol 2000;109:235e42. [5] Gutierrez-Rodriguez M, Reyes-Maldonado E, Mayani H. Characterization of the adherent cells developed in Dexter-type long-term cultures from human umbilical cord blood. Stem Cells 2000;18:46e52. [6] Romanov YA, Svintsitskaya VA, Smirnov VN. Searching for alternative sources of postnatal human mesenchymal stem cells: candidate MSC-like cells from umbilical cord. Stem Cells 2003;21:105e10. [7] Campagnoli C, Roberts IAG, Kumar S, Bennett PR, Bellantuono I, Fisk NM. Identification of mesenchymal stem/progenitor cells in human first-trimester fetal blood, liver, and bone marrow. Blood 2001;38: 2396e402. [8] Ko¨gler G, Sensken S, Airey JA, Trapp T, Mu¨schen M, Feldhahn N, et al. A new human somatic stem cell from placental cord blood with intrinsic pluripotent differentiation potential. J Exp Med 2004;200:123e35. [9] Lee MW, Choi J, Yang MS, Moon YJ, Park JS, Kim HC, et al. Mesenchymal stem cells from cryopreserved human umbilical cord blood. Biochem Biophys Res Commun 2004;320:273e8. [10] Wexler SA, Donaldson C, Denning-Kendall P, Rice C, Bradley B, Hows JM. Adult bone marrow is a rich source of human mesenchymal ‘stem’ cells but umbilical cord and mobilized adult blood are not. Br J Haematol 2003;121:368e74. [11] Majumdar MK, Thiede MA, Mosca JD, Moorman M, Gerson SL. Phenotypic and functional comparison of cultures of marrow-derived mesenchymal stem cells and stromal cells. J Cell Physiol 1998;176:57e66. [12] Marshak DR, Holecek JJ. Chemical defined medium for human mesenchymal stem cells. United States Patent 5,908,782; 1999.
[13] Besse A, Trimoreau F, Praloran V, Denizot Y. Effect of cytokines and growth factors on the macrophage colony-stimulating factor secretion by human bone marrow stromal cells. Cytokine 2000;12:522e5. [14] Yen BL, Huang HI, Chien CC, Jui HY, Ko BS, Yao M, et al. Isolation of multipotent cells from human term placenta. Stem Cells 2005;23:3e9. [15] Huang X, Zhang Q. Coal-induced interleukin-6 gene expression is mediated through ERKs and p38 MAPK pathways. Toxicol Appl Pharmacol 2003;191:40e7. [16] Colter DC, Sekiya I, Prockop DJ. Identification of a subpopulation of rapidly self-renewing and multipotential adult stem cells in colonies of human marrow cells. Proc Natl Acad Sci USA 2001;98:7841e5. [17] Dennis JE, Merriam A, Awadallah A, Yoo JU, Johnstone B, Caplan AI. A quadripotential mesenchymal progenitor cell isolated from the marrow of an adult mouse. J Bone Miner Res 1999;14:700e9. [18] Klassen HJ, Imfeld KL, Kirov II, Tai LQ, Gage FH, Young MJ, Berman MA. Expression of cytokines by multipotent neural progenitor cells. Cytokine 2003;22:101e6. [19] Gadient RA, Otten UH. Interleukin-6 (IL-6) A molecule with both beneficial and destructive potentials. Prog Neurobiol 1997;52:379e90. [20] Hungness EA, Pritts TA, Luo GJ, Hershko DD, Rob BW, Hasselgren PO. IL-1b activates C/EBP-b and delta in human enterocytes through a mitogen-activated protein kinase signaling pathway. Int J Biochem Cell Biol 2002;34:382e95. [21] Cardinaux JR, Allaman I, Magistretti PJ. Pro-inflammatory cytokines induce the transcription factors C/EBP-b and C/EBP-d in astrocytes. Glia 2000;29:91e7. [22] Janderova´ L, McNeil M, Murrell AN, Mynatt RL, Smith SR. Human mesenchymal stem cells as an in vitro model for human adipogenesis. Obes Res 2003;11:65e74. [23] Potian JA, Aviv H, Ponzio NM, Harrison JS, Rameshwar P. Veto-like activity of mesenchymal stem cells: functional discrimination between cellular responses to alloantigens and recall antigens. J Immunol 2003; 171:3426e34. [24] Haynesworth SE, Baber MA, Caplan AI. Cytokine expression by human marrow-derived mesenchymal progenitor cells in vitro: Effects of dexamethasone and IL-1. J Cell Physiol 1996;166:585e92. [25] Lambert E, Dasse E, Haye B, Petitfrere E. TIMPs as multifacial proteins. Crit Rev Oncol Hematol 2004;49:187e98. [26] Mazzarelli P, Scuderi F, Mistretta G, Provenzano C, Bartoccioni E. Effect of transforming growth factor-b1 on interleukin-6 secretion in human myoblasts. J Neuroimmunol 1998;87:185e8. [27] Tsuji F, Oki K, Senda T, Horiuchi M, Mita S. Effects of mitogenactivated protein kinase inhibitors or phosphodiesterase inhibitors on interleukin-1-induced cytokines production in synovium-derived cells. Immunol Lett 1999;68:275e9. [28] Mracek T, Cannon B, Housteka J. IL-1 and LPS but not IL-6 inhibit differentiation and downregulate PPAR gamma in brown adipocytes. Cytokine 2004;26:9e15. [29] Chen CC, Manning AM. TGF-b1, IL-10 and IL-4 differentially modulate the cytokine-induced expression of IL-6 and IL-8 in human endothelial cells. Cytokine 1996;8:58e65. [30] Paulsen DF, Solursh M. Microtiter micromass cultures of limb-bud mesenchymal cells. In Vitro Cell Dev Biol 1988;24:38e47. [31] Vicennati V, Vottero A, Friedman C, Papanicolaou DA. Hormonal regulation of interleukin-6 production in human adipocytes. Int J Obes Relat Metab Disord 2002;26:905e11. [32] Gimble JM, Wanker F, Wang CS, Bass H, Wu X, Kelly K, et al. Regulation of bone marrow stromal cell differentiation by cytokines whose receptors share the gp130 protein. J Cell Biochem 1994;54: 122e33. [33] Panepucci RA, Siufi JLC, Silva WA, Proto-Siquiera R, Neder L, Orellana M, et al. Comparison of gene expression of umbilical cord vein and bone marrow-derived mesenchymal stem cells. Stem Cells 2004;22:1263e78. [34] Delikat SE, Galvani DW, Zuzel M. The metabolic effects of interleukin 1 on human bone marrow adipocytes. Cytokine 1995;7:338e43. [35] Delikat S, Harris RJ, Galvani DW. IL-1b inhibits adipocyte formation in human long-term bone marrow culture. Exp Hematol 1993;21: 31e7.
C.-H. Liu, S.-M. Hwang / Cytokine 32 (2005) 270e279 [36] Reichlin S, Chen GJ, Nicolson M. Blood to brain transfer of leptin in normal and interleukin-1b-treated male rats. Endocrinology 2000;141: 1951e4. [37] Janik JE, Curti BD, Considine RV, Rager HC, Powers GC, Alvord G, et al. Interleukin 1 increases serum leptin concentrations in humans. J Clin Endocrinol Metab 1997;82:3084e6. [38] Grunfeld C, Zhao C, Fuller J, Pollock A, Moser A, Friedman J, et al. Endotoxin and cytokines induce expression of leptin, the ob gene product, in hamsters a role for leptin in the anorexia of infection. J Clin Invest 1996;97:2152e7.
279
[39] Rosen ED. The transcriptional basis of adipocyte development. Prostaglandins Leukotrienes Essent Fatty Acids 2005;73:31e4. [40] Qian H, Hausman GJ, Compton MM, Azain MJ, Hartzell DL, Baile CA. Down-regulation of CCAAT/enhancer binding proteins in adipose tissue by intracerebro-ventricular leptin in rats. Biochim Biophys Acta 1998;1442:245e51. [41] Mitchell SE, Rees WD, Hardie LJ, Hoggard N, Tadayyon M, Arch JRS, et al. ob gene expression and secretion of leptin following differentiation of rat preadipocytes to adipocytes in primary culture. Biochem Biophys Res Commun 1997;230:360e4.