The Comparison of Interleukin 6–Associated Immunosuppressive Effects of Human ESCs, Fetal-Type MSCs, and Adult-Type MSCs

The Comparison of Interleukin 6–AssociatedImmunosuppressive Effects of Human ESCs, Fetal-Type

MSCs, and Adult-Type MSCs

Chin-Kan Chan,1,2 Kang-Hsi Wu,3,4 Yun-Shen Lee,5,6 Shiaw-Min Hwang,7 Maw-Sheng Lee,8,9

Shuen-Kuei Liao,1,10 En-Hui Cheng,11 Lai-Chu See,12,13 Chi-Neu Tsai,1 Ming-Ling Kuo,14,16

and Jing-Long Huang15,16

Background. Although human embryonic stem cells (ESCs) and mesenchymal stem cells (MSCs) from various sourcesdisplay immunomodulatory effects, direct comparisons among these classes of stem cells have not been reported.Methods. Peripheral blood mononuclear cell suppression assays and carboxyfluorescein diacetate succinimidyl esterassays were used to assess the immunosuppressive effects of stem cells. Gene expression was measured using DNAmicroarrays. Enzyme-linked immunosorbent assays were used to determine interleukin (IL)-6 levels.Results. We found that fetal-type MSCs proliferated significantly faster than adult-type MSCs. Compared with ESCsgrown on feeder cells, ESCs grown in feeder cellYfree conditions exhibited decreased immunosuppressive effects. Thesuppressive effects of ESCs were significantly stronger than those of MSCs, and the suppressive effects of fetal-typeMSCs were significantly stronger than those of adult-type MSCs at each tested dose level. Analysis of gene expressionby microarray and MetaCore pathway mapping revealed changes in eight different immune response pathways; weobserved that IL-6 gene expression was highly significantly involved in all eight pathways. Significantly higher IL-6elevation ratios (IL-6after:IL-6before) were found in ESCs compared with fetal-type MSCs, and these were also foundin fetal-type MSCs compared with adult-type MSCs. Furthermore, IL-6 levels were found to correlate with cell dosagesof MSCs and the suppressive effects.Conclusions. The ease of obtaining fetal-type MSCs and their rapid proliferation make these cells ideal candidates forcell-based therapies, especially for diseases associated with immune responses, given the immunosuppressive effects ofthese cells. IL-6 might play an important role in the immunosuppressive effects of various stem cells.

Keywords: Embryonic stem cells, Mesenchymal stem cells, Immunosuppressive effects, Interleukin 6.

(Transplantation 2012;94: 132Y138)

This work was supported by the Taoyuan General Hospital Project(PTH9702 and PTH9809) and grants from China Medical UniversityHospital (DMR-100-055) and the National Science Council of Taiwan(Grant NSC 100-2314-B-039-014-). The study was conducted under theauspices of the Department of Health (North 98017) and the Ministry ofEducation, Taiwan (EMRPD170101).

The authors declare no conflicts of interest.1 Graduate Institute of Clinical Medical Sciences, Chang Gung University,

Taoyuan, Taiwan.2 Department of Pediatrics, Taoyuan General Hospital, Taoyuan, Taiwan.3 School of Chinese Medicine, China Medical University, Taichung, Taiwan.4 Department of Pediatrics, China Medical University Hospital, Taichung, Taiwan.5 Department of Biotechnology, Ming Chuan University, Taoyuan, Taiwan.6 Genomic Medicine Research Core Laboratory, Taoyuan, Taiwan.7 Bioresource Collection and Research Center, Food Industry Research and

Development Institute, Hsinchu, Taiwan.8 Institute of Medicine, Chung Shan Medical University, Taichung, Taiwan.9 Department of Obstetrics and Gynecology, Chung Shan Medical University

Hospital, Taichung, Taiwan.10 Cancer Immunotherapy Center, Taipei Medical University Hospital,

Taipei, Taiwan.11 Department of Biochemistry, School of Medicine, Chung Shan Medical

University, Taichung, Taiwan.12 Biostatistics Consulting Center, Department of Public Health, Chang

Gung University, Taoyuan, Taiwan.13 Biostatistics Core Laboratory, Molecular Medical Research Center, Chang

Gung University, Taoyuan, Taiwan.

14 Department of Microbiology and Immunology, Graduate Institute ofBiomedical Sciences, Chang Gung University, Taoyuan, Taiwan.

15 Division of Allergy, Asthma and Rheumatology, Department of Pediatrics,Chang Gung Memorial Hospital, Taoyuan, Taiwan.

16 Address correspondence to: Jing-Long Huang, M.D., Division of Allergy,Asthma and Rheumatology, Department of Pediatrics, Chang GungMemorial Hospital, No. 5, Fu-Hsin St., Kweishan, Taoyuan, Taiwan; orMing-Ling Kuo, Ph.D., Department of Microbiology and Immunology,Graduate Institute of Biomedical Sciences, Chang Gung University, No.259, Wen-Hwa 1st Rd., Kweishan, Taoyuan, Taiwan.

E-mail: [email protected]; or [email protected]. and K.-H.W. made the research design and wrote the article. Y.-S.L.

and C.-N.T. performed the microarray analysis. S.-M.H., M.-S.L., andE.-H.C. provided study materials. S.-K.L. participated in the study de-sign. L.-C.S. performed the statistical analyses. M.-L.K. provided dataanalysis and interpretation. J.-L.H. participated as leader of the studydesign. K.-H.W. and Y.-S.L. contributed equally to this work.

Received 27 February 2012. Revision requested 16 March 2012.Accepted 5 April 2012.Supplemental digital content (SDC) is available for this article. Direct URL

citations appear in the printed text, and links to the digital files areprovided in the HTML text of this article on the journal’s Web site(www.transplantjournal.com).

Copyright * 2012 by Lippincott Williams & WilkinsISSN: 0041-1337/12/9402-132DOI: 10.1097/TP.0b013e31825940a4

BASIC AND EXPERIMENTAL RESEARCH

132 www.transplantjournal.com Transplantation & Volume 94, Number 2, July 27, 2012

Copyright © 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

The clinical appeal of human embryonic stem cells (ESCs)follows from the immunosuppressive effects displayed by

these cells (1, 2). Conventionally, ESCs have been culturedwith mouse embryonic fibroblast (MEF) (3, 4), leading toconcerns about xenogenic stromal line contamination inclinical applications. Ideally, clinical-grade ESCs should bederived from and maintained in xeno-free culture condi-tions. However, there have been no reports (to our knowl-edge) directly comparing the immunosuppressive activity ofcells grown under these two conditions.

Mesenchymal stem cells (MSCs) with similar surfacephenotype expression patterns can be isolated from varioussources. MSCs derived from adipose tissues (ADMSCs) (5, 6)and bone marrow (BMMSCs) (7Y9) are considered adult-type MSCs, whereas MSCs derived from amniotic fluid(AFMSCs) (10), amniotic membrane (AMMSCs) (11), cordblood (CBMSCs) (12), or umbilical cord (UCMSCs) (13, 14)are considered fetal-type MSCs. Clinical interest has beenraised by the observation that MSCs are immune privilegedand, more importantly, exhibit immunomodulatory capac-ities (5Y15). Based on the immunomodulatory properties ofMSCs, MSCs have been used not only for graft-versus-hostdisease (7Y9, 15) and organ rejection after transplantation(16) but also for other immune disorders, such as enceph-alomyelitis, collagen-induced arthritis, interstitial fibrosisof chronic kidney disease, and glomerulonephritis (17Y20).Therefore, MSCs are ideal candidates for the treatment ofdiseases associated with aberrant immune responses.

In most previous reports, MSCs dosed in humans werederived from bone marrow (7Y9). However, the MSCs countin bone marrow decreases significantly with the age of thedonor (7Y9). In addition, acquiring BMMSCs or ADMSCsrequires invasive procedures, whereas obtaining fetal-typeMSCs is easy and safe for donors. Thus, the derivation ofMSCs from alternative tissues is appealing for future clinicalapplications. In our previous studies, we found that UCMSCsshowed more immunosuppressive effects than BMMSCs,and it was effective in treating severe steroid-resistant acutegraft-versus-host disease without any severe adverse effects(14). In addition, we found that UCMSCs could enhanceengraftment after hematopoietic stem-cell transplantation(21). Therefore, fetal-type MSCs may be to substitute adult-type MSCs in human clinical application.

It has been demonstrated that MSCs are capable ofexerting immunomodulatory effects on virtually all cells ofthe immune system (14, 15); however, the mechanism behindthese effects remains unclear. Interleukin (IL)-6 cytokine canactivate target genes involved in differentiation, survival,apoptosis, and proliferation. IL-6 possesses proinflammatoryand anti-inflammatory properties (22). MSCs have beenfound to secrete high levels of IL-6 on stimulation (23, 24);however, the association of IL-6 on the immunomodulatoryeffects of MSCs from various sources has not been reported.

The results of these studies have been encouraging, butthe immunomodulatory effects previously reported for ESCsand MSCs from various sources were obtained in a range ofdifferent laboratories. To our knowledge, this is the firststudy to directly compare the immunosuppressive propertiesof ESCs and MSCs from various sources. To better definepotential mechanisms of immunomodulatory effects of stemcells, we used microarrays, functional network analysis, and

MetaCore pathway mapping (GeneGo, Saint Joseph, MI)(26) to analyze and compare the gene expression among thesestem cells.

RESULTS

Characteristics of Fetal- and Adult-Type MSCsThe AFMSCs, AMMSCs, CBMSCs, UCMSCs, ADMSCs,

and BMMSCs showed identical uniform spindle-shapedmorphologies. All of these lines were positive for markersCD29, CD44, CD73, CD90, and CD105, as well as humanleukocyte antigen (HLA)-A, HLA-B, and HLA-C, but all ofthese lines tested negative for markers CD31, CD34, CD45,CD117, CD184, and HLA-DR. No significant differences werenoted in the expression levels of any single surface markersbetween fetal- and adult-type MSCs (Table 1; see Figure 1,SDC, http://links.lww.com/TP/A677). Under the respectiveinduction conditions, all of the MSCs were capable of achiev-ing osteogenic, adipogenic, and chondrogenic differentiation.The fetal-type MSCs showed a significantly stronger osteo-genic potential, but a lower capacity for adipogenic differen-tiation, than adult-type MSCs.

Proliferative Potential of Fetal- and Adult-TypeMSCs

No differences were found in proliferative potentialamong fetal-type MSCs (AFMSCs, AMMSCs, CBMSCs, andUCMSCs). Similarly, no differences were noted in prolifer-ative potential between ADMSCs and BMMSCs. The fetal-type MSCs exhibited a faster expansion rate than adult-typeMSCs, measured as average population doubling (PD) ofeach passage (passages 3Y6) (all P values G0.05) (Fig. 1A). Asignificant increase in the cumulative PD from passages 3 to6 was found when comparing fetal-type MSCs with adult-type MSCs (medianTSEM, 11.08T2.09 vs. 4.29T0.84; PG0.05)(Fig. 1B). These results indicated that fetal-type MSCs pro-liferated faster than adult-type MSCs.

Suppressive Effects of ESCs, Fetal-Type MSCs,and Adult-Type MSCs on Peripheral BloodMononuclear Cells

The ESCs cultured under feeder cellYfree conditions(on Matrigel-coated dishes) (BD Biosciences, San Jose, CA)(PG0.0001) or with MEF feeder cells (PG0.0001) exhibited sig-nificant suppressive effects on peripheral blood mononuclearcells (PBMCs) (see Figure 2A, SDC, http://links.lww.com/TP/A677).Whereas ESCs cultured under either condition suppressedPBMC proliferation,direct comparison revealed that sup-pression was significantly greater with ESCs co-culturedwith MEF feeder cells (PG0.0001) (see Figure 2A, SDC,http://links.lww.com/TP/A677). However, direct compari-son among the four classes of ESCs (using the Kruskal-Wallis test) revealed that differences among the four classeswere not significant at each dose level (see Figure 2B, SDC,http://links.lww.com/TP/A677). Similarly, suppression did notsignificantly differ between ADMSCs and BMMSCs at each doselevel (see Figure 2C, SDC, http://links.lww.com/TP/A677).All six classes of MSCs inhibited PBMC proliferation ina dose-dependent manner (see Figure 2B and 2C, SDC,http://links.lww.com/TP/A677).

In a comparison among ESCs, fetal-type MSCs, andadult-type MSCs, we found that suppressive effects onPBMCs of ESCs cultured in Matrigel-coated dishes were

* 2012 Lippincott Williams & Wilkins Chan et al. 133


significantly stronger than those of fetal-type MSCs at eachdose level (all P values G0.001) (Fig. 2A). The suppressiveeffects of fetal-type MSCs were significantly stronger thanthose of adult-type MSCs at each dose level (all P valuesG0.001) (Fig. 2A). Post hoc comparisons were also calcu-lated. The suppressive effects of these stem cells on PBMCswere confirmed by the results of the carboxyfluoresceindiacetate succinimidyl ester (CFSE) assays (Fig. 2B,C).

Gene Expression Profiles of ESCs, Fetal-TypeMSCs, and Adult-Type MSCs

Of 31,099 microarray genes analyzed, 260 up-regulationgenes (0.84%) (more than threefold change in expression)and 699 down-regulation genes (2.25%) (more than three-fold change in expression) were found in ESCs, fetal-typeMSCs, and adult-type MSCs (Fig. 3A; see Table 1, SDC,http://links.lww.com/TP/A677). After functional networkanalysis of the highly expressed genes and MetaCore pathwaymapping, we observed significant changes in eight differentimmune response pathways, including T helper 17 cell differ-entiation, macrophage migration inhibitory factorYmediatedglucocorticoid regulation, histamine H1 receptor signaling inimmune response, triggering receptor expressed on myeloidcells 1 signaling pathway, IL-17 signaling pathways, CD40signaling, histamine signaling in dendritic cells, and macro-phage migration inhibitory factor in innate immunity response.Notably, IL-6 is involved in all eight of these pathways. Theseeight immune response pathways associated with IL-6 aresummarized in Figure 3(B).

Correlation Between IL-6 Secretion Levels andSuppressive Effects on PBMCs

We detected induction of IL-6 production by MSCseven in the absence of stimulation. Specifically, IL-6 levels incontrol groups (MSCs only) increased when cell doses ofMSCs increased. In addition (and as expected), after PBMCswere stimulated by phytohemagglutinin (PHA) and co-cultured with stem cells, we detected significant increases inIL-6 secretion in response to all three groups, includingESCs (P=0.001), fetal-type MSCs (PG0.001), and adult-typeMSCs (P=0.005), when compared with the level of IL-6secretion seen under basal conditions (i.e., without PHAstimulation). We found that IL-6 elevated ratios from ESCswere significantly higher than those from fetal-type MSCs

TABLE 1. The phenotypic profile of fetal- and adult-type mesenchymal stem cells

Cell sources

Markers

CD29 CD31 CD34 CD44 CD45 CD73 CD90 CD105 CD117 CD184 HLA-A, HLA-B, and HLA-C HLA-DR

Amniotic fluid + j j + j + + + j j + j

Amnioticmembrane

+ j j + j + +/j + +/j + or j + j

Cord blood + j j + j + + + j j + j

Umbilical cord + j j + j + + + j j + j

Adipose tissue + j j + j + + + j j + j

Bone marrow + j j + j + + + j j + j

+/j indicates the dim results of our samples.+ or j indicates that some samples showed positive results but some showed negative results.HLA, human leukocyte antigen.

FIGURE 1. Comparison of proliferative potential betweenfetal- and adult-type mesenchymal stem cells (MSCs). A,Growth curves of fetal- and adult-type MSCs are shown.Significant differences of mean population doubling (PD)from passages 3 to 6 were observed between the two groupsat each passage (all P values G0.05). B, Higher proliferationpotential of fetal-typeMSCs alsowas demonstrated by largercumulative PD from passages 3 to 6 than that of adult-typeMSCs. * PG0.05. These results indicated that fetal-type MSCsproliferated faster than adult-type MSCs.



and that the ratios from fetal-type MSCs were significantlyhigher than those from adult-type MSCs (all P valuesG0.001) (Fig. 4A). After PHA stimulation, the level ofsecreted IL-6 was significantly higher in cells co-culturedwith fetal-type MSCs, and this effect was dose depen-dent (PG0.001) (Fig. 4B). Analysis also revealed that theincrease in IL-6 levels on co-culture with fetal-type MSCswas significantly associated with the suppression of PBMCproliferation (R2=0.9105, PG0.01) (Fig. 4C).

DISCUSSIONIt has been demonstrated that MSCs are capable of

exerting immunomodulatory effects on virtually all cells

including T lymphocytes, dendritic cells, natural killer cells,and B lymphocytes of the immune system (15, 26). In thepresent study, PBMCs were used as responder cells becausePBMCs are known to include all of the cells involved incomplicated immune interactions. We found that the im-munosuppressive effects of ESCs were stronger than those ofMSCs, whereas the immunosuppressive effects of fetal-typeMSCs were stronger than those of adult-type MSCs. To ourknowledge, this study is the first to directly compare theimmunosuppressive properties of ESCs and MSCs fromvarious sources using a single experimental setting.

Conventionally, ESCs required co-culturing with feed-er cells (typically MEFs), raising concerns regarding xeno-contamination. In the present study, we found that ESCscultured with MEFs showed stronger immunosuppressiveeffects than those cultured under feeder-free conditions. Wealso noted that immunosuppressive effects by ESCs grownunder feeder-free conditions remained higher than those seenwith MSCs. Hence, even in the absence of feeder cells, ESCsstill possessed strong immunosuppressive effects. To ourknowledge, this study represents the first survey of the im-munosuppressive properties of ESCs maintained under MEF-and feeder-free culture conditions. Although ESCs, even underfeeder-free conditions, were found to have strong immuno-suppressive effects, when ESCs are applied clinically, concernsabout ethics and tumor formation (27) will still need to betaken into consideration.

Acquiring BMMSCs or ADMSCs requires an invasiveprocedure. Our study found that the immunosuppressiveeffects of fetal-type MSCs are significantly stronger thanthose of adult-type MSCs. We also demonstrated that fetal-type MSCs have greater proliferative potential than adult-type MSCs. Therefore, we expect that doses of cells adequatefor clinical applications can be derived more quickly fromfetal-type MSCs than from adult-type MSCs. Thus, fetal-typeMSCs not only are easier to obtain than adult-type MSCs butalso proliferate faster, making the fetal-type MSCs idealcandidates for clinical cell-based therapies. Because of theirimmunosuppressive effects, fetal-type MSCs are especiallyappealing for the treatment of diseases associated with ab-errant immune responses.

FIGURE 2. Comparisons among embryonic stem cells(ESCs), fetal-type mesenchymal stem cells (MSCs), andadult-type MSCs in suppressive effects on peripheral bloodmononuclear cells (PBMCs). A, The reduction rate of PBMCproliferation activated by phytohemagglutinin after co-culture with ESCs, fetal-type MSCs, or adult-type MSCs atvarious cell dosages was compared. PBMC proliferationwas assessed by pulsing with WST-1. Stem cell (ESCs orMSCs):PBMC ratios are indicated on the x-axis. Data arepresented as a box plot (median and range) calculated fromexperiments performed in triplicate. *** PG0.001. B, Thereduction rate of PBMC proliferation activated by anti-CD3/CD28 antibodies after co-culture with ESCs, fetal-typeMSCs, or adult-type MSCs at various cell dosages wascompared. PBMCs labeled with carboxyfluorescein diace-tate succinimidyl ester (CFSE)were analyzed by flow cyto-metry. *** PG0.001. C, After stimulation with anti-CD3/CD28antibodies, cell divisionwas analyzed byCFSE labeling andflow cytometry before and after co-culturing with MSCs.UCMSCs, MSCs derived from umbilical cord.



Half of the fetal genome is derived from the father; as aresult, the fetus is considered a foreign body by the maternalimmune system (28). Nonetheless, this ‘‘natural’’ allograftis not normally rejected. In the present study, it was foundthat the more ‘‘primitive’’ the stem cells, the stronger theimmunosuppressive effects. For example, the immunosup-pressive effects of ESCs were found to be significantly strongerthan those of fetal-type MSCs, and the immunosuppressiveeffects of fetal-type MSCs were found to be significantlystronger than those of adult-type MSCs. We therefore spec-ulate that the suppressive effects of ESCs and MSCs in the fetusmight protect the fetus from rejection by the maternal im-mune systems in utero. This hypothesis is consistent with thetheory of Medawar regarding fetomaternal tolerance (29, 30).

Although we found that ESCs and MSCs have im-munosuppressive effects, the mechanism behind these effectsremains unclear. It has been found that IL-6 induces intra-cellular signaling cascades that give rise to inflammatorycytokine production (22). Dysregulation of IL-6Ytype cyto-kine signaling contributes to the onset of rheumatoid ar-thritis and systemic lupus erythematosus (31). MSCs havebeen found to secrete higher levels of IL-6 on stimulation,and IL-6 may be important for MSC-mediated regulation oflocal inflammatory responses through modulation and in-hibition of T-cell and dendritic cell proliferation (23Y25).However, the association of IL-6 with the immunosuppres-sive effects of stem cells has not been reported. After PHAstimulation, we found that IL-6 elevation ratios (IL-6after:IL-6before) of ESCs were significantly higher than those of fetal-type MSCs whereas the ratios of fetal-type MSCs weresignificantly higher than those of adult-type MSCs. This el-evation of IL-6 secretion was dose dependent and correlatedwith suppressive effects on PBMC proliferation. To ourknowledge, this is the first study to directly compare theimmunosuppressive properties of these classes of stem cellsand the first demonstration of a role for IL-6 in this stemcellYspecific immunomodulatory process. We speculate thatIL-6 plays an important role in the immunosuppressive effectof various stem cells.

MATERIALS AND METHODS

Culture of Human ESCsThe institutional review board of Taoyuan General Hospital

(Taoyuan, Taiwan) approved this study. For ESCs, TW1, TW2, TW3, TW4,

and TW5 cell lines were generously provided by LMS and CEH, who pre-

viously have published studies on these cell lines (3). Briefly, before use

in the PBMC proliferation assay, the ESCs were cultured either with MEF

feeder cells or under feeder-free culture conditions (using Matrigel-coated

dishes) (3). The basic culture medium for ESCs was Dulbecco Modified

Eagle Medium: Nutrient Mixture F-12 basal medium supplemented with 15%

knockout serum replacement, 1-mM L-glutamine, 0.1-mM A-mercaptoethanol,

0.1-mM MEM nonessential amino acids (Gibco; Invitrogen, Carlsbad, CA)

and 4-ng/mL recombinant human basic fibroblast growth factor (Gibco;

Invitrogen, Carlsbad, CA).

Culture and Identification of MSCsSix different classes of MSCs, including fetal-type MSCs (AFMSCs,

AMMSCs, CBMSCs, and UCMSCs) and adult-type MSCs (ADMSCs and

BMMSCs), were used. All of these MSCs were generously provided by

the Bioresource Collection and Research Center (Hsinchu, Taiwan) (25, 32).

Briefly, MSCs were cultured as previously reported (11, 25, 31) in >-MEM

(HyClone; Gibco, Invitrogen, Carlsbad, CA). supplemented with 20% fetal

bovine serum (HyClone) and 4-ng/mL basic fibroblast growth factor (R&D

Systems, Minneapolis, MN). The MSCs were immunolabeled with mouse

antihuman antibodies against one of the following antigens: CD34, CD45,

CD29, CD31, CD44, and CD90, as well as HLA-A, HLA-B, HLA-C, and

HLA-DR (BD Biosciences, San Jose, CA); CD105 (AbD Serotec, Oxford,

UK); and CD73, CD117, and CD184 (BD Pharmingen, San Diego, CA).

The cells were then incubated with a secondary antibody, antimouse IgGY

fluorescein isothiocyanate or IgG-phycoerythrin, and analyzed using flow

cytometry (BD Biosciences). The specific conditions used to induce osteo-

genic, adipogenic, and chondrogenic differentiation of the MSCs were as pre-

viously described (11, 33).

FIGURE 3. Analysis of gene expression profiles of em-bryonic stem cells, fetal-type mesenchymal stem cells(MSCs), and adult-type MSCs. A, Volcano analysis of RNAexpression profiles. Of 31,099 microarray genes analyzed,260 up-regulation genes (more than threefold change) and699 down-regulation genes (more than threefold change)were found among embryonic stem cells, fetal-type MSCs,and adult-type MSCs. B, After functional network analysisof the highly expressed genes and MetaCore pathwaymapping, eight kinds of immune response pathways wereimplicated; we observed that interleukin (IL)-6 gene ex-pression is highly significantly involved in all eight of thesepathways. MIF, macrophage migration inhibitory factor; Th,T helper; TREM1, mean triggering receptor expressed onmyeloid cells 1.



Cell Proliferation Assays for MSCsTo prevent hematopoietic cell contamination, which might be

present in earlier passages, or the presence of senescent or differentiating

MSCs in later passages, we used cells from passages 3 to 6 for the study of

growth kinetics. The yield of cells at each passage was enumerated using

trypan blue (Gibco, Invitrogen, Carlsbad, CA) to exclude dead cells. The PD

of the cultured MSCs was calculated using the equation: PD=log2 of the ratio

of the number of viable cells at harvest to the number of seeded cells (33).

PBMC Proliferation Assay and CFSE AssayAll of the PBMC proliferation assay procedures were performed

as previously reported (11). Briefly, for use as responder cells, PBMCs

were cultured in triplicate as 200 HL at 5�105 cells/mL per well in 96-well

U-bottom microtiter cell plates (Costar; Bio-Rad, Hercules, CA), and each

well was stimulated using either 5-Hg/mL PHA (Sigma-Aldrich, St Louis,

MO) or 1-Hg/mL anti-CD3/CD28 antibodies (Dynal Biotech, Oslo, Norway).

Given their high proliferation kinetics, the MSCs and ESCs used were all

F irradiated (25 Gy) before use in this assay. A series of MSC:PBMC or

ESC:PBMC ratios were used, including 1:1000, 1:100, 1:10, 1:5, 1:2, 1:1,

and 2:1. After PBMCs were incubated with or without MSCs or ESCs for

3 days, then 10 HL of cell proliferation reagent WST-1 (Roche Diagnostics

GmbH, Mannheim, Germany) was added to each well. After 1- to 4-hr in-

cubation at 37-C, absorbance at 450 Hm was measured with a microplate

reader (Molecular Devices Corporation, Sunnydale, CA) using a reference

wavelength of 600 nm.

In addition, a CFSE assay was used to confirm the results. The CFSE-

labeled PBMCs were collected, resuspended at 5�105 cells/mL in RPMI-1640

with anti-CD3/CD28 antibodies, and transferred into 24-well plates at 500 HL/

well. MSCs or ESCs were added to the individual wells, and cultures were

incubated at 37-C. After 72 hr, the cells were harvested and washed twice with

phosphate-buffered saline. Analysis of cell division was performed by flow

cytometry (11, 34).

RNA Preparation and Microarray AnalysisGene expression was assessed using hybridization of cellular RNA

with the human U133A GeneChip (Affymetrix). Expression was tested using

five lines of ESCs, four classes of fetal-type MSCs, and two classes of adult-

type MSCs. RNA purification and testing were performed as previously

described (25). Briefly, cells were grown to 90% confluence, rinsed with ice-

cold phosphate-buffered saline, and lysed with TRIzol reagent (Invitrogen).

RNA was isolated from the lysates using RNeasy purification kits (Qiagen).

RNA quality and quantity were assessed using the Bioanalyzer 2100 (Agilent

Technologies). Hybridization with the microarray was performed per the

manufacturer’s protocol.

Enzyme-Linked Immunosorbent AssayIL-6 protein levels were determined by enzyme-linked immuno-

sorbent assay (ELISA) using a commercially available ELISA kit (BD

Biosciences). ELISAs were performed according to the manufacturer’s

instructions; all samples and standards were tested in duplicate (23).

Statistical AnalysisThe ESCs, fetal-type MSCs, and adult-type MSCs were tested under

identical experimental settings, thereby allowing direct comparisons. All

experiments were performed in triplicate. We defined the controls as PBMCs

FIGURE 4. The association of interleukin (IL)-6 levels andsuppressive effects among embryonic stem cells (ESCs),fetal-type mesenchymal stem cells (MSCs), and adult-typeMSCs on peripheral blood mononuclear cells (PBMCs). A,IL-6 elevated ratios (IL-6after/IL-6before) were defined as theIL-6 level after phytohemagglutinin (PHA) stimulation di-vided by the IL-6 level before PHA stimulation (stemcells only). We found that elevation ratios of IL-6 secretedby ESCs were significantly higher than those of fetal-typeMSCs, and elevation ratios of IL-6 secreted by fetal-typeMSCs were significantly higher than those of adult-type MSCs.Stem cell (ESC or MSC):PBMC ratios are indicated on thex-axis. Data are presented as a box plot (median andrange)calculated from experiments performed in tripli-cate. *** PG0.001. B, After PBMCs were stimulated byPHA, IL-6 levels significantly increased as cell doses ofco-cultured fetal-typeMSCs increased. *** PG0.001. C, IL-6levels in cells stimulated with PHA and co-cultured withfetal-type MSCs were significantly associated with the re-duction rate of suppressive effects on PBMCs (R2=0.9105,PG0.01).



stimulated with PHA or anti-CD3/CD28 antibodies without stem cells. The

‘‘reduction rate’’ of diminished PBMC proliferation was defined as the

optical density (OD) of the control group (PBMCs plus PHA or anti-CD3/

CD28 antibodies) minus the OD of the experimental group (PBMCs plus

PHA or anti-CD3CD28 antibodies plus stem cells) then divided by the OD of

the control group. We compared the reduction rates among ESCs, fetal-type

MSCs, and adult-type MSCs. Comparisons of reduction rates were performed

as nonparametric analyses, either by Wilcoxon rank sum tests (for comparisons

between two groups or two different culture conditions) or by Kruskal-Wallis

tests (for comparisons among three or more groups). Analyses of reduction

rates after co-culture with MSCs or ESCs at varying dosages were performed by

Kruskal-Wallis with a Bonferroni adjustment.

We defined that IL-6 levels in the supernatants of the wells with stem

cells (ESCs or MSCs) only as control groups. After PBMCs were stimulated

by PHA, IL-6 levels in the supernatants of the wells with stem cells co-cul-

tured with PBMCs were assayed as the experimental groups. To quantify any

stimulatory effect, we calculated (for each culture) an ‘‘IL-6 elevation ratio,’’

defined as the IL-6 level after PHA stimulation divided by the IL-6 level

before PHA stimulation (IL-6after/IL-6before). We compared IL-6 elevation

ratios among ESCs, fetal-type MSCs, and adult-type MSCs. The comparisons

of IL-6 levels and IL-6 elevation ratios among ESCs, fetal-type MSCs, and

adult-type MSCs were analyzed by the Kruskal-Wallis test. Linear correlation

and linear regression were used to analyze the correlation between the di-

minished PBMC proliferation and IL-6 levels.

Statistical analyses were performed using the SPSS statistical

package version 16.0 for Windows (SPSS Inc., Chicago, IL). P values of

less than 0.05 were considered statistically significant.

ACKNOWLEDGMENTSThe authors thank the Genomic Medicine Research Core

Laboratory for microarray analysis. The authors thank Pei-YzuLee, Chen-Hsu Chen, and Hsiu-Li Chou for their excellenttechnical assistance.

REFERENCES1. Koch CA, Geraldes P, Platt JL. Immunosuppression by embryonic

stem cells. Stem Cells 2008; 26: 89.2. Yen BL, Chang CJ, Liu KJ, et al. Human embryonic stem cellYderived

mesenchymal progenitors possess strong immunosuppressive effectstoward natural killer cells as well as T lymphocytes. Stem Cells 2009;27: 451.

3. Cheng EH, Chen W, Chang SY, et al. Blastocoel volume is related tosuccessful establishment of human embryonic stem cell lines. ReprodBiomed Online 2008; 17: 436.

4. Thomson JA, Itskovitz-Eldor J, Shapiro SS, et al. Embryonic stem celllines derived from human blastocysts. Science 1998; 282: 1145.

5. McIntosh K, Zvonic S, Garrett S, et al. The immunogenicity ofhuman adipose derived cells: temporal changes in vitro. Stem Cells2006; 24: 1246.

6. Puissant B, Barreau C, Bourin P, et al. Immunomodulatory effect ofhuman adipose tissueYderived adult stem cells: comparison with bonemarrow mesenchymal stem cells. Br J Haematol 2005; 129: 118.

7. Sessarego N, Parodi A, Podesta MF, et al. Multi-potent mesenchymalstromal cells from amniotic fluid: solid perspectives for clinical ap-plication. Haematologica 2008; 93: 339.

8. Wolbank S, Peterbauer A, Fahrner MS, et al. Dose-dependent immu-nomodulatory effect of human stem cells from amniotic membrane: acomparison with human mesenchymal stem cells from adipose tissue.Tissue Eng 2007; 13: 1173.

9. Oh W, Kim DS, Yang YS, et al. Immunological properties of umbilicalcord blood derived mesenchymal stromal cells. Cell Immunol 2008;251: 116.

10. Romanov YA, Svintsitskaya VA, Smirnov VN. Searching for alter-native sources of postnatal human mesenchymal stem cells: candidateMSC-like cells from umbilical cord. Stem Cells 2003; 21: 105.

11. Wu KH, Chan CK, Tsai C, et al. Effective treatment of severe steroid-resistant acute graft-versus-host disease with umbilical cordYderivedmesenchymal stem cells. Transplantation 2011; 91: 1412.

12. Nauta AJ, Fibbe WE. Immunomodulatory properties of mesenchymalstromal cells. Blood 2007; 110: 3499.

13. Ball LM, Bernardo ME, Roelofs HA, et al. Co-transplantation of ex vivoexpanded mesenchymal stem cells accelerates lymphocyte recovery andmay reduce the risk of graft failure in haploidentical hematopoieticstem cell transplantation. Blood 2007; 110: 2764.

14. Le Blanc K, Rasmusson I, Sundberg BC, et al. Treatment of severe acutegraft-versus-host disease with third-party haploidentical mesenchymalstem cells. Lancet 2004; 363: 1439.

15. Macmillan ML, Blazar BR, DeFor TE, et al. Transplantation of ex-vivoculture-expanded parental haploidentical mesenchymal stem cells topromote engraftment in pediatric recipients of unrelated donor um-bilical cord blood: results of a phase IYII clinical trial. Bone MarrowTransplant 2009; 43: 447.

16. Inoue S, Popp FC, Koehi GE, et al. Immunomodulatory effects ofmesenchymal stem cells in arat organ transplant model. Transplan-tation 2006; 81: 1589.

17. Zappia E, Casazza S, Pedemonte E, et al. Mesenchymal stem cellsameliorate experimental autoimmune encephalomyelitis inducingT-cell anergy. Blood 2005; 106: 1755.

18. Djouad F, Fritz V, Apparailly F, et al. Reversal of the immunosup-pressive properties of mesenchymal stem cells by tumor necrosis factoralpha in collagen-induced arthritis. Arthritis Rheum 2005; 52: 1595.

19. Ninichuk V, Gross O, Segerer S, et al. Multipotent mesenchymal stemcells reduce interstitial fibrosis but do not delay progression of chronickidney disease in collagen4A3-deficient mice. Kidney Int 2006; 70: 121.

20. Kunter U, Rong S, Djuric Z, et al. Transplanted mesenchymal stemcells accelerate glomerular healing in experimental glomerulonephritis.J Am Soc Nephrol 2006; 17: 2202.

21. Chao YH, Tsai C, Peng CT, et al. Cotransplantation of umbilical cordMSCs to enhance engraftment of hematopoietic stem cells in patientswith severe aplastic anemia. Bone Marrow Transplant 2011; 46: 1391.

22. Heinrich PC, Behrmann I, Haan S, et al. Principles of interleukin (IL)-6Ytype cytokine signalling and its regulation. Biochem J 2003; 374: 1.

23. Djouad F, Charbonnier LM, Bouffi C, et al. Mesenchymal stem cellsinhibit the differentiation of dendritic cells through an interleukin-6dependent mechanism. Stem Cells 2007; 25: 2025.

24. Liu CH, Hwang SM. Cytokine interaction in mesenchymal stem cellsfrom cord blood. Cytokine 2005; 32: 270.

25. Tsai MS, Hwang SM, Chen KD, et al. Functional network analysis ofthe transcriptomes of mesenchymal stem cells derived from amnioticfluid, amniotic membrane, cord blood, and bone marrow. Stem Cells2007; 25: 1511.

26. Ramasamy R, Fazekasova H, Lam EW et al. Mesenchymal stem cellsinhibit dendritic cell differentiation and function by preventing entryinto the cell cycle. Transplantation 2007; 83: 71.

27. Amariglio N, Hirshberg A, Scheithauer BW, et al. Donor-derived braintumor following neural stem cell transplantation in an ataxia telan-giectasia patient. PLoS Med 2006; 6: e1000029.

28. Guleria I, Sayegh MH. Maternal acceptance of the fetus: true humantolerance. J Immunol 2007; 178: 3345.

29. Billingham RE, Brent L, Medawar PB. ‘Actively acquired tolerance’ offoreign cells. Nature 1953; 172: 603.

30. Medawar PB. Some immunological and endocrinological problemsraised by the evolution of viviparity in vertebrates: prevention of al-logeneic fetal rejection by tryptophan catabolism. Symp Soc Exp Biol1953; 7: 320.

31. Sun L, Akiyama K, Zhang H, et al. Mesenchymal stem cell trans-plantation reverses multiorgan dysfunction in systemic lupus erythe-matosus mice and humans. Stem Cells 2009; 27: 1421.

32. Chang YJ, Shih DT, Tseng CP, et al. Disparate mesenchyme-lineagetendencies in mesenchymal stem cells from human bone marrow andumbilical cord blood. Stem Cells 2006; 24: 679.

33. Chao YH, Peng CT, Harn HJ, et al. Poor potential of proliferation anddifferentiation in bone marrow mesenchymal stem cells derived fromchildren with severe aplastic anemia. Ann Hematol 2010; 89: 715.

34. Olson JA, Zeiser R, Beilhack A, et al. Tissue-specific homing and ex-pansion of donor NK cells in allogeneic bone marrow transplantation.J Immunol 2009; 183: 3219.



The Comparison of Interleukin 6–Associated Immunosuppressive Effects of Human ESCs, Fetal-Type MSCs, and Adult-Type MSCs

Documents