paper (trần cẩm tú) identiﬁcation of human placenta-derived mesenchymal stem cells involved...

ORIGINAL ARTICLE 224J o u r n a l o fJ o u r n a l o f

CellularPhysiologyCellularPhysiology
Identification of Human
Placenta-Derived MesenchymalStem Cells Involved inRe-Endothelialization
TU CAM TRAN,1,2 KENICHI KIMURA,1 MASUMI NAGANO,1 TOSHIHARU YAMASHITA,1
KINUKO OHNEDA,3 HARUHIKO SUGIMORI,4 FUJIO SATO,4 YUZURU SAKAKIBARA,4

HIROMI HAMADA,5 HIROYUKI YOSHIKAWA,5 SON NGHIA HOANG,2

AND OSAMU OHNEDA1*1Department of Regenerative Medicine and Stem Cell Biology, Graduate School of Comprehensive Human Sciences,

University of Tsukuba, Tsukuba, Japan2Institute of Tropical Biology, Vietnam Academy of Science and Technology, Thu Duc Dist., Ho Chi Minh City, Vietnam3Faculty of Pharmacy, Laboratory of Molecular Pathophysiology, Takasaki University of Health and Welfare, Takasaki, Japan4Department of Cardiovascular Surgery, Graduate School of Comprehensive Human Sciences, University of Tsukuba,

Tsukuba, Japan5Department of Obstetrics and Gynecology, Graduate School of Comprehensive Human Sciences, University of Tsukuba,

Tsukuba, Japan

Human placenta is an attractive source of mesenchymal stem cells (MSC) for regenerative medicine. The cell surface markers expressedon MSC have been proposed as useful tools for the isolation of MSC from other cell populations. However, the correlation between theexpression of MSC markers and the ability to support tissue regeneration in vivo has not been well examined. Here, we established severalMSC lines from human placenta and examined the expression of their cell surface markers and their ability to differentiate towardmesenchymal cell lineages. We found that the expression of CD349/frizzled-9, a receptor for Wnt ligands, was positive in placenta-derivedMSC. So, we isolated CD349-negative and -positive fractions from an MSC line and examined how successfully cell engraftment repairedfractured bone and recovered blood flow in ischemic regions using mouse models. CD349-negative and -positive cells displayed a similarexpression pattern of cell surface markers and facilitated the repair of fractured bone in transplantation experiments in mice. Interestingly,CD349-negative, but not CD349-positive cells, showed significant effects on recovering blood flow following vascular occlusion. Wefound that induction of PDGFb and bFGF mRNAs by hypoxia was greater in CD349-negative cells than in CD349-positive cells while theexpression of VEGF was not significantly different in CD349-negative and CD349-positive cells. These findings suggest the possibility thatCD349 could be utilized as a specialized marker for MSC isolation for re-endothelialization.

J. Cell. Physiol. 226: 224–235, 2010. � 2010 Wiley-Liss, Inc.

The authors indicate no potential conflicts of interest.

Tu Cam Tran and Kenichi Kimura contributed equally to this work.

Additional Supporting Information may be found in the onlineversion of this article.

*Correspondence to: Osamu Ohneda, Department ofRegenerative Medicine, University of Tsukuba, 1-1-1 Tennodai,Tsukuba 305-8575, Japan. E-mail: [email protected]

Received 9 September 2009; Accepted 6 July 2010

Published online in Wiley Online Library(wileyonlinelibrary.com.), 23 July 2010.DOI: 10.1002/jcp.22329

The placenta plays an essential role in the development ofthe embryo. The main functions of the placenta are gaseousexchange between the embryo and the mother, the provisionof maternal nutrients to the fetus and the excretion of wasteproducts from the fetus. It has been reported that the placentais enriched with a variety of stem cells originating fromhematopoietic, trophoblastic, and mesenchymal tissues(Fleischman and Mintz, 1979; Faulk et al., 1990; Fukuchi et al.,2004; Yen et al., 2005; Parolini et al., 2008). Mesenchymal stemcells (MSC) derived from placenta are considered an excellentmaterial for regenerative medicine because of their broaddifferentiation potential, wide accessibility and the lack ofethical concerns.

The two types of MSC, amnionic and chorionic, originatefrom a distinct region of the placenta. Amnionic MSC arise fromthe amnion membrane that consists of amnion epithelial cellsand MSC (Alviano et al., 2007; Bilic et al., 2008; Magatti et al.,2008; Miki et al., 2009). Chorionic MSC develop from thechorion membrane that consists of chorionic trophoblast cellsand MSC (Battula et al., 2008). Amnionic and chorionic MSC canbe isolated separately by mechanical agitation and enzymaticdigestion. Both MSC types possess the potential to differentiateinto three mesodermal cell-types: osteogenic, chondrogenic

� 2 0 1 0 W I L E Y - L I S S , I N C .

and adipogenic lineages. Amnionic MSC can also differentiateinto neuron cells derived from the ectodermal germ layer orinto hepatic and pancreatic cells that arise from the endodermalgerm layer (reviewed in Fukuchi et al., 2004; Parolini et al., 2008;Miki et al., 2009).

The isolation of MSC is generally performed by a procedurebased on the adherence of cells to the surface of culture dishes.

I S O L A T I O N O F P L A C E N T A - M S C F O R R E - E N D O T H E L I A L I Z A T I O N 225

Adherent cells are composed of various cell types, of which onlya small population is stem cells. The major difficulty of thisprocedure is that adherent non-MSC often proliferates morerapidly than MSC. One of the most promising ways to purifyMSC is to utilize MSC-specific markers. Previous studiesshowed that MSC are positive for CD13, CD29, CD44, CD73(SH3, 4), CD90, CD105 (SH2) and CD166 and negative forCD14, CD31, CD34 and CD45 (Portmann-Lanz et al., 2006;Delorme and Charbord, 2007). These markers are helpful fordiscriminating between mesenchymal cells and endothelial orhematopoietic cells. However, there is no established protocolfor isolating MSC solely by these markers. Some reports havedemonstrated that STRO-1, CD271 (nerve growth factorreceptor) and ganglioside molecule GD2 are useful for isolatingbone marrow (BM)-MSC (Bensidhoum et al., 2004; Buhringet al., 2007; Martinez et al., 2007). However, it is not known ifthese markers are applicable to the isolation of placental MSC.

Recently, CD349/Frizzled-9 was shown to be a useful markerfor separating MSC from placenta (Battula et al., 2007, 2008).Frizzled (Fzd) is a family of seven transmembrane-spanningproteins that serve as receptors for Wnt proteins. To date,10 family members (Fzd 1 to 10) with conserved structuralfeatures have been found in vertebrates (Koike et al., 1999).CD349 is expressed in pericytes and mesenchymal cellssurrounding the large blood vessels of the placenta. The colony-forming units-fibroblastic (CFU-F) population is 60 timesgreater in CD349þ/CD10þ/CD26þ cells from placenta thanin CD349þ cells, whereas the CFU-F population is absentin CD349�/CD10�/CD26 cells. Therefore, CD349 mightserve as a useful marker for purifying MSC from placenta,although the difference in properties between CD349-positiveand -negative MSC has not been fully elucidated.

In the present study, we established several lines of MSCfrom human placenta and examined their CD349 expressionlevel and differentiation properties. We isolated CD349-negative and -positive fractions from an MSC line and examinedhow successfully cell engraftment repaired fractured bone andrecovered blood flow in ischemic regions using mouse models.Both CD349-negative and -positive cells showed the potentialto differentiate into an osteogenic lineage and facilitated therepair of fractured bone in transplantation experiments in mice.Interestingly, CD349-negative, but not CD349-positive cells,showed significant effects on recovering blood flow followingvascular occlusion. These results suggest that CD349 might beused as a specialized marker of placental MSC for arteriogenesisand angiogenesis.

Materials and MethodsIsolation of MSC and cell culture

Human full-term placentas were collected by caesarean sectionfrom healthy donor mothers. Tissues were obtained after informedconsent and all experiments were approved by the local ethicsauthorities at the University of Tsukuba.

Chorion leave tissue was manually separated and treated with0.1% collagenase (Nitta Gelatin, Osaka, Japan)/20% FBS (Hyclone,South Logan, UT)/PBS solution at 378C for 1 h. Following filtrationthrough a cell strainer (Falcon 3078; pore size 100mm; BDBioscience, San Jose, CA), the cells were cultured in IMDM(Invitrogen, Carlsbad, CA) with 10% FBS (Hyclone, South Logan,UT), 2 mg/ml L-glutamine (Invitrogen, Carlsbad, CA), 5 ng/mlhuman bFGF (Peprotech, London, United Kingdom) and 0.1%(v/v) penicillin–streptomycin (100 U/ml penicillin, 0.1 mg/mlstreptomycin; Invitrogen). The cells were maintained in a 10 cmtissue culture dish (Sumitomo Bakelite, Osaka, Japan) at 378Cin a humidified atmosphere of 5% CO2. The culture medium wasreplaced with fresh medium once a week.

After adherent cells reached subconfluency, they wereharvested with 0.05% trypsin–EDTA (Invitrogen) and purified for

JOURNAL OF CELLULAR PHYSIOLOGY

CD31�/CD45- cells by FACS to remove endothelial andhematopoietic cells. The cells were cultured with the same mediumused for the isolation except for the concentration of bFGF(10 ng/ml). Furthermore, following two or three passages, the cellswere plated in a 10 cm tissue culture dish at a low confluency.Clusters that formed in the dish were cloned using cloningcylinders (Sigma–Aldrich, St. Louis, MO) and expanded. Frozen cellstocks were prepared using Cell Banker (ZENOAQ, Koriyama,Japan) solution and stored in liquid nitrogen for furtherexperiments. All experiments were performed using at least threedistinct sources of placenta.

Human BM samples were collected from sternum withpermission from the local ethics authorities at the University ofTsukuba. BM derived MSC were cultured in the same way asplacental derived MSC.

Antibodies

The antibodies used in this study were as follows: Fluoresceinisothiocyanate (FITC)-labeled HLA-A,B,C (W6/32), phycoerythrin(PE)-labeled anti-CD31 (WM59), anti-HLA-DR (L243),allophycocyanine (APC)-labeled anti-CD45 (HI30), biotin-labeledCD349 (W3C4E11; BioLegend, San Diego, CA), FITC-labeled anti-CD105 (SN6; Serotec, Oxford, UK), anti-CD90 (5E10; Biolegend),PE-labeled anti-CD166 (3A6), anti-CD73 (AD2), anti-CD14(M5E2), anti-CD13 (WM15), anti-CD271 (LNGFR; MiltenyiBiotec, Auburn, CA), APC-labeled anti-CD34 (581), anti-mouseIgG (BD Biosciences) and PE-labeled anti-SSEA4 (MC-813-70;R&D systems, Minneapolis, MN). After staining the cells withfluorochrome-conjugated antibodies, cells were sorted usingFACSVantageSE (BD Biosciences) as previously described(Ohneda et al., 2001).

In vitro differentiation assay of MSC

To examine their capacity to differentiate toward an osteogeniclineage, MSC were cultured in the presence of 50 ng/ml of humanepidermal growth factor (EGF; Wako, Osaka, Japan) and analyzedfor the expression of alkaline phosphatase (ALP) as reportedpreviously (Kratchmarova et al., 2005). Following differentiation,the cells were harvested on day 9, purified for mRNA and examinedby RT-PCR.

To induce osteogenic differentiation, 5� 104 cells were treatedwith osteogenic differentiation medium in 4-well plates (NalgeNunc, Rochester, NY) for 4 weeks. The osteogenic differentiationmedium consisted of IMDM supplemented with 1% FBS, 0.1 mMdexamethasone (Sigma–Aldrich), 10 mM b-glycerol-2-phosphate(Sigma–Aldrich), 0.2 mM ascorbic acid (Sigma–Aldrich) and 50 ng/ml of human EGF (Kogler et al., 2004; Lee et al., 2004). The culturemedium was replaced with fresh medium once or twice a week.Alkaline phosphatase activity was examined histologicallyaccording to the manufacturer’s instructions (Leukocyte AlkalinePhospatase-Kit, Sigma–Aldrich). The mineralized matrix wasevaluated by von Kossa staining and Alizarin red staining asdescribed previously (Lee et al., 2004).

Adipogenic differentiation was induced in 4-well plates for4 weeks by adipogenic differentiation medium consisting of IMDMsupplemented with 10% FBS, 0.1 mM dexamethasone (Sigma–Aldrich), 0.5 mM 3-isobutyl-1-methylxanthine (IBMX; Sigma–Aldrich), 2 mg/ml insulin (Wako) and 0.1 mM indomethacine(Sigma–Aldrich). The culture medium was replaced with freshmedium once or twice a week. Cultured cells in adipogenicdifferentiation medium were fixed with 10% formaldehyde (Wako)and stained with Oil-Red O solution (Muto Pure Chemicals, Tokyo,Japan) for 30 min at 428C. After the staining, cells were dissolvedwith 4% IGEPAL CA630 (Sigma–Aldrich) in isopropanol and theabsorbance was measured at 480 nm.

To promote chondrogenic differentiation, cells were treatedwith chondrogenic differentiation medium in 96-well spheroidplates for 4 weeks. Chondrogenic differentiation medium

226 T R A N E T A L .

consisted of IMDM supplemented with 1% FBS, 0.1 mMdexamethasone (Sigma–Aldrich), 1 mM sodium pyruvate(Invitrogen), 0.25 nM ascorbic acid, 50 mg/ml ITS premix (BDBioscience), 40 mg/ml proline (Sigma–Aldrich), 10 ng/ml TGF-b1(Wako), and 10 ng/ml BMP-2 (Wako). The culture medium wasreplaced with fresh medium once or twice a week. In order toevaluate chondrocyte differentiation, spheroids were fixed with 4%paraformaldehyde and stained with Toluidine blue solution (MutoPure Chemicals).

Analysis of MSC in a bone fracture mouse model

A bone fracture mouse model was generated according to amodified method of the one reported by Taguchi et al. (2005).Adult C57/BL6 mice were anesthetized and closed transversefractures of the femur were produced in the middle part of thethigh bone. The fractured femurs were connected by pins and anincision was made at the connected site by a 27G-gage needle(2 mm in diameter).

MSC (5� 105) were plated on a 2 mm� 2 mm Gelform (Pfizer,New York, NY) and incubated at 378C for 2 h before insertioninto the mice. The transplanted Gelform was then fixed atthe connected site. Immunosuppression was performed byintraperitoneal injection of 20 mg/kg body weight of cyclosporin-A(Wako) 2 days before the assay. Cyclosporin-A injections werecontinued daily for the entire period of the assay. X-rays weretaken 28 days post transplantation and the density at the bone-gaparea was measured with NIH imaging software.

For histological staining, thigh bones were fixed with 4%paraformaldehyde (Wako) for 7 days and decalcified withPlank-Rychlo solution (Muto Pure Chemicals) for 30 days. Frozenor paraffin-embedded samples were sectioned at a thickness of7mm and stained with an antibody or hematoxylin–eosin (HE)solution (Muto Pure Chemicals) according to the modified methodreported by Kawamoto and Shimizu (2000).

Microscopy analysis

Cell samples were viewed with an Olympus IX71 microscopesystem (Olympus, Tokyo, Japan) using UPlanF objective lenses at4�/0.13PhL and 10�/0.30Ph1. Sample slides were viewed with anOlympus BX51 microscope system (Olympus) using UPlanSApoobjective lenses at 4�/0.16PH, 10�/0.40PH and 20�/0.75PH(Olympus) and mounting reagent (Muto Pure Chemicals). Dataacquisition was carried out using a DP70 digital camera attachedto the microscope and DP controller software (Olympus). Imageswere processed using Adobe Photoshop version 8.0 software(Adobe System, San Jose, CA).

RT-PCR and quantitative PCR

Total RNA (1mg) was reverse transcribed using an RT-PCR kit(BD Biosciences) as described previously (Nagano et al., 2007).Resulting cDNAs were amplified by a GeneAmp PCR System 9100(Applied Biosystems, Foster City, CA) for 23–35 cycles of 958C for5 sec and 688C for 30 seconds. b-actin was used as an internalcontrol. The reaction mixtures for quantitative PCR wereprepared using POWER SYBR1 Green PCR master mix (AppliedBiosystems) and analyzed by a 7700 Sequence Detector (AppliedBiosystems). Experiments were performed in triplicate and datawere calculated by the DDCt method.

The primers used for the PCR reactions were as follows: Oct-4(50- AAGCTCCTGAAGCAGAAGAGGATCACC; 30-GGTTAC-AGAACCACACTCGGACCACAT), NANOG (50-CCTCCAT-GGATCTGCTTATTCAGGACA; 30-CCTTCTGCGTCACACC-ATTGCTATTCT), SOX2 (50-GGAAAACCAAGACGCTCATG-AAGAAGG; 30-GTTCATGTAGGTCTGCGAGCTGGTCAT),Rex1 (50-CAACCCATCCTGGAAGAGGACTCACTT;30-GGAGATGCTTTCTCAGGGCAGCTCTAT), Glut-1(50-CCTTGGATGTCCTATCTGAGCATCG; 30-ATCTCATC-GAAGGTTCGGCCTTTGG), VEGF (50-GAACTTTCTGCTGT-


CTTGGGTGCATTG; 30-CTGCATGGTGATGTTGGACTCC-TCAGT), PDGFb (50-GACCTGTCCAGGTGAGAAAGATCG-AGA; 30-AAATAACCCTGCCCACACACTCTCCTG), bFGF(50-AGAGCGACCCTCACATCAAGCTACAAC; 30-ATAGC-TTTCTGCCCAGGTCCTGTTTTG), TGF-b (50-AGAGCTCCG-AGAAGCGGTACCTGAACCC; 30-GTTGATGTCCACTTGCA-GTGTGTTATCC), Angiopoietin-1 (Ang1) (50-CTGACTCAC-ATAGGGTGCAGCAATCAG; 30-AGGCTGGTTCCTATCTCC-AGCATGGTA), b-actin (50-GTGCGTGACATTAAGGAGAA-GCTGTGC; 30-GTACTTGCGCTCAGGAGGAGCAATGAT).

Mouse vascular occlusion model

Young adult (2 months old) male BDF-1 mice underwent unilateralfemoral artery and vein ligation. Arteries and veins from theproximal end of the femoral vessels to the popliteal vessels wereligated with 6-0 silk (Couffinhal et al., 1998). In addition, all sidebranches of the femoral and popliteal vessels were ligated, whereasneurons were carefully kept unchanged.

On the first day after the surgical process, 5� 105 cells wereinjected intramuscularly into four divided sites. A laser Dopplerblood flow meter (FLO-C1, Omegawave, Tokyo, Japan) wasutilized for the measurement of serial blood flow at the inner ankle.Data were represented as the ratio of blood flow in the ischemiclimb site divided by that in the non-ischemic site.

Mice were analyzed for angiogenesis 2 weeks after induction ofhind limb ischemia. The capillary density in the thigh muscle wasassessed by immunofluorescence using Banderiraea simplicifolialectin I-TRITC (0.1 mg/ml; Sigma–Aldrich) as reported previously(Nagano et al., 2007). Frozen sections were mounted and observedunder a microscope equipped with the appropriate filters(Olympus). The number of capillaries was measured in 10 differentrandomized fields of each mouse.

Animal care and experimental procedures complied with the‘‘Principles of Laboratory Animal Care’’ (Guide for the Care andUse of Laboratory Animals, University of Tsukuba) and wereapproved by the Use Committee of the University of Tsukuba.

Statistical analysis

Statistical evaluations of data were conducted using the Student’s t-test for per-comparison analysis. Data are presented asmeans� SD.

ResultsIsolation of placenta-derived MSC

Placenta cells were obtained from women who had a normalpregnancy and full-term delivery. These cells were grown inculture medium containing bFGF (5 ng/ml) and the adherentcells were collected. The CD45-negative and CD31-negativecell fraction was sorted by FACS to eliminate hematopoieticand endothelial cells (Fig. 1A). These cells were expanded withbFGF (10 ng/ml) in standard culture dishes. At that time point,the morphology of the cells appeared to be heterogeneous(data not shown). Spindle shaped cells were chosen for cloningand at least 12 cell lines were obtained from each placenta.Among them, 6 cell lines (PL56, PL57, PL58, PL59, PL73, andPL74) were chosen because they grew faster than the other6 cell lines. As a control, BM-MSC was prepared by the sameprocedure.

We found that the cloned cell lines could be divided into twogroups by their morphologies. PL57 and PL58 were composedof short spindle-shaped cells, whereas PL73 and PL74 werefibroblast-like and resembled BM-MSC in appearance (Fig. 1B).To examine whether these cells can differentiate into anosteoblast lineage, EGF was added to the culture medium.Kratchmarova et al. (2005) reported that ALP mRNA increasedin response to EGF when the MSC differentiated into anosteogenic lineage. We found that ALP mRNA was clearly

Fig. 1. Isolation of mesenchymal stem cells from human placenta. Adherent cells derived from placenta were purified for non-hematopoietic(CD45-negative) and non-endothelial (CD31-negative) cells by FACS. Cells within the border shown were cultured for further experiments.Phase-contrast micrograph of the four placenta-derived (PL) adherent cell lines (PL57, PL58, PL73, and PL74) and BM-derived MSC (BM-MSC).Note that PL73 and PL74 show the fibroblast-like morphology and resemble BM-MSC, whereas PL57 and PL58 show the spindle-shapedmorphology. Bar indicates 50mm. The expression of alkaline phosphatase (ALP) mRNA was examined for PL adherent cells on day 9 afterthe addition of EGF to the culture. The expressions of b-actin and BM-MSC mRNAs were utilized as an internal control and a positive control,respectively. Data obtained on day 0 was normalized to a value of 1 as the standard for each cell. White column: without EGF (day 0); black column:with EGF (day 9). MMP < 0.01. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]


induced in BM-MSC and in PL73 on the 9th day after induction(Fig. 1C). ALP mRNA was also upregulated in PL74 and PL59in response to EGF (Supplementary Fig. 1A). On the contrary,ALP mRNA was not induced in PL57, PL58 and PL56(Supplementary Fig. 1A). These results indicate that PL73, PL74


and PL59, but not PL57, PL58 and PL56, can differentiate into anosteogenic lineage in response to EGF (Supplementary Fig. 1B).On the basis of these findings, we have chosen a fibroblastic cellline (PL73) and a non-fibroblastic cell line (PL57) for furtheranalyses.

228 T R A N E T A L .

The expression of cell surface markers onplacenta-derived MSC

Since PL73 and PL57 showed the distinct feature of osteogenicdifferentiation, a comparison was made between placental andbone marrow MSC relating to their expression of cell surfacemarkers (Fig. 2). Despite the clear difference in osteogenic

Fig. 2. FACSanalysesofcell surfacemarkersonPLadherentcells.PL57(B(CD14,CD31,CD34,CD45,CD13,CD90,CD73,CD105,CD166,HLA-ABCa control. No significant differences in the expressions of the cell surface


differentiation in vitro, PL73 and PL57 demonstrated a similarcell surface marker expression profile (Fig. 2B,C). Both cell lineswere negative for hematopoietic (CD45 and CD14) andendothelial (CD31 and CD34) cell markers. Expression ofthe MSC markers CD13, CD73, and CD105 was positive forboth PL73 and PL57. Compared to the expression profiles inBM-MSC (Fig. 2A), PL73 and PL57 were negative for CD271 and

)andPL73(C)wereanalyzed fortheexpressionofcell surfacemarkers,HLA-DR,SSEA4,andCD271)byFACS.BM-MSC(A)wasexaminedasmarkers were observed between PL57 and PL73.

Fig. 3. Analysis of the ability of PL-derived cells to differentiate.Data obtained from two representative PL adherent cell groups(PL57: middle parts; PL73: parts on the right) are shown. BM-MSCserved as a positive control (parts on the left). A: Osteogenicdifferentiation was examined by alkaline phosphatase stainingwithout (w/o) induction (top parts) or with induction (bottom parts).Bar indicates 100mm. B,C: Osteogenic differentiation was furtherexamined by observing the formation of the mineralized matrixby Von Kossa (B) and Alizarin red (C) stainings without induction(top parts) or with induction (bottom parts). Bar indicates 100mmin (B) and 50mm in (C). D: Adipogenic differentiation was studiedby the detection of lipid vacuoles by Oil Red O staining withoutinduction (top parts) or with induction (bottom parts). Bar indicates50mm. E: Chondrogenic differentiation was inspected by Toluidineblue staining (bottom parts). HE staining was performed toascertain cell morphology (top parts). Bar indicates 200mm.Note that PL57 was negative for each stain used to detect thedifferentiation potential of MSC.


showed the bimodal distribution of CD90 expression(Fig. 2B,C). The expression of SSEA4, a previously reportedmarker for BM-MSC (Gang et al., 2007), was similar in BM-MSC,PL57 and PL73. We also examined the expression of thesemarkers for PL58 and PL74 (Supplementary Fig. 2). Theexpression of these markers was almost similar among the 4 celllines except that the peak of CD90 expression level variedamong the cell lines.

In vitro differentiation of placenta-derived MSC

Although we have screened the capability of osteogenicdifferentiation by EGF-induced ALP mRNA expression onday 9, the full induction of osteogenic differentiation requires4 weeks of culture in the presence of dexamethasone,b-glycerol-2-phosphate and ascorbic acid in addition to EGF.We therefore cultured PL57 and PL73 for 4 weeks in thiscondition and performed cytochemical analysis of the ALPprotein (Fig. 3A), Von Cossa staining (Fig. 3B), and Alizarin redstaining (Fig. 3C). We also examined these cells for adipocyteand chondrocyte differentiation in vitro by Oil Red O (Fig. 3D)and Toluidine Blue (Fig. 3E) staining, respectively.

Consistent with the results of ALP mRNA expression onday 9, the ALP protein was not detectable in PL57 cells, whereasBM-MSC and PL73 did express ALP protein in this culturecondition. Mineralization of osteocytes in BM-MSC and PL73was further confirmed by Von Cossa staining (Fig. 3B), andAlizarin red staining (Fig. 3C). PL57 failed to differentiateinto other mesenchymal cell lineages, such as adipocytes andchondrocytes. In contrast, BM-MSC and PL73 were able todifferentiate into these cell lineages. Furthermore, neither PL57nor PL58 could differentiate into osteocytes or adipocytes,after a longer culture period (45 days; data not shown).

Consequently, these data indicate that PL73 and PL74possess characteristics of MSC, whereas PL57 and PL58 may bederived from a cell lineage other than MSC. In addition, thesedata suggest the possibility that measurement of the ALPmRNA level in response to EGF on day 9 of cultivation might bean efficient first screening for MSC isolation.

The effects of placenta-derived MSC engraftment onbone repair in a mouse model

Although PL57 failed to differentiate into any mesenchymal celllineages in our cell culture experiments, the profile of cellsurface markers of PL57 was consistent with that of MSC andwas indistinguishable from that of PL73. These observations ledus to test the possibility that PL57 might be able to differentiateinto mesenchymal cells in vivo. To assess this, we transplantedPL57, PL73 and BM-MSC into immunosuppressed mice withsurgically fractured femurs and examined the processes of bonerepair (Fig. 4). Twenty-eight days after surgery, new boneformation was evaluated by the degree of calcification revealedby X-ray. Following the X-ray examination, histological analysiswas performed on the region of transplantation and in the jointregion of the fractured femur. As shown in Figure 4A, BM-MSCand PL73 engraftment revealed the infiltration of mononuclearcells, including inflammatory blood cells and osteocytic cells atthe injection site. In contrast, cells were sparsely distributedwith a few osteocytic cells at the site of PL57 transplantation,which was akin to the control PBS injection. These resultssuggest that PL73 and BM-MSC, but not PL57, can proliferaterapidly in situ and differentiate into osteocytic cells at the site oftransplantation. It is also significant that PL73 and BM-MSC canaccumulate host-derived mesenchymal cells and inflammatorycells into this region. X-ray examination revealed that thedegree of calcification at the joint of each fractured femur wassignificantly higher in PL73 and BM-MSC transplanted micecompared to that in PL57 transplanted and PBS injected mice(Fig. 4B). Consistent with these results, histological


Fig. 4. Repair of bone fracture by transplantation of PL adherent cells. BM-MSC, PL57, PL73 or no cells (PBS) were transplanted into miceat the of bone fracture sites using Gelform (collagen-based gelatin sponge) as a carrier vehicle. We looked at the morphology by HE staining (A),the osteocalcification by X-ray (B; left side) and the density (B; right bar graph) of the transplantation sites. HE staining was performed at thejoint of the fractured bones (C). The boxed regions in the upper parts of (A) and (C) were magnified in the lower parts. Bars in the upper and lowerparts indicate 200mm and 100mm, respectively. Note that when PL73 was transplanted, the growing bone substitution process progressedsimilarly as in the BM-MSC transplant. MP < 0.05, MMP < 0.01.

230 T R A N E T A L .

examination at the joint regions of the fractured bones revealedthe formation of lamellar bone on pre-existing hyaline cartilagein mice transplanted with BM-MSC or PL73 (Fig. 4C). Theseobservations suggest that the endochondral ossification andbony substitution were processed at the joint regions of thefractured bones in these mice.


In contrast, woven bone with fibroblastic cells was observedat the joint regions of the PL57 transplanted and PBS injectedmice (Fig. 4C). Although the accumulation of fibroblastic cellsinto the joint region was observed in both PL57 recipient andcontrol (PBS), the central region of the joint was not completelyfilled with nucleated cells as observed in BM-MSC of PL73


recipient (Fig. 4C). These results suggest that osteogenicdifferentiation in the joint region of the PL57 transplanted micewas obviously delayed compared to that in the BM-MSC orPL73 transplants. Taken together, these results suggest thattransplantation of BM-MSC or PL73 facilitates the repairprocess of a fractured femur. On the contrary, PL57 failed toaccelerate the bone repair process. Despite the profile of cellsurface markers on PL57 being indistinguishable from that forPL73, we found no evidence of mesenchymal cell differentiationof PL57 both in vitro and in vivo.

Separation of placenta-derived MSC on the basis ofCD349 expression

Battula et al. (2007) reported that frizzled-9 (CD349), amember of the Wnt receptor family, is expressed in BM-MSCand placenta MSC when cultivated in a serum-free, b-FGF-containing medium. CD349-positive MSC exhibited multi-lineage differentiation into mesodermal, ectodermal, andendodermal cells. So, CD349 might be utilized as a marker forisolating MSC from BM and placenta. This result suggests thatPL73 might be composed of two subpopulations with distinctfeatures. We then isolated CD349-negative and -positive PL73cell fractions by FACS and determined if the expressions of cellsurface markers other than CD349 differs between theCD349-negative and -positive cells (Fig. 5B,C). As observed inparental PL73 cells (Fig. 2C), both cell fractions were negativefor the hematopoietic cell markers CD45 and CD14 and theendothelial markers CD31 and CD34. The expressions oftypical MSC markers were similar in both CD349-naegative and-positive PL73 (Fig. 5B,C). Exceptionally, the peak of CD90expression was lower in the CD349-positive fractioncompared to that in the CD349-negative fraction. We furtherexamined the expression of CD349 for PL74 and isolatedCD349-positive and –negative fractions (Supplementary Fig. 3).Unlike the case of PL73, the CD349 expression level was lowand did not show a bimodal distribution in PL74. There was nosignificant difference between the CD349-negative and -positive fractions in the expression of MSC markers in PL74.

To further assess the status of differentiation of these cells,we studied the expressions of the immature cell markers Oct4,Nanog, Rex1 and Sox2 by RT-PCR (Fig. 5D). While theexpression of these genes was detectable in BM-MSC, Rex-1and Nanog expressions were higher in BM-MSC compared tothat in PL-MSC by real time PCR (data not shown). Importantly,the expression level of these genes was similar in both CD349-negative and -positive cells, indicating that CD349-negative and-positive cell fractions cannot be distinguished by thesemarkers.

We also examined CD349 expression in PL57 and PL58(Supplementary Fig. 4). As shown above, these cells could notbe defined as MSC by their differentiation potential. The peak ofCD349 expression level was low, but a considerable frequencyof CD349 (þ) cells were observed in the two non-MSC lines.The frequency of positive cells was 58% and 20% for PL57 andPL58, respectively. Taken together, these results suggest thatCD349 may not be a suitable marker to discriminate MSC andnon-MSC from placenta.

Both CD349-negative and -positive cells facilitated thebone repair process in mice

We observed that the transplantation of PL73 facilitated thebone repair process in a mouse model. Therefore, we evaluatedthe therapeutic effects that the expression level of CD349might have on bone repair (Fig. 6). As with the parental PL73transplantation, bone calcification was significantly higher inmice transplanted with CD349-negative or -positive cellscompared to control mice injected with PBS. Furthermore, thedegree of calcification at the joint region was higher in mice


transplanted with CD349-positive cells compared to thosetransplanted with CD349-negative cells. Interestingly, theseresults were consistently observed in the other MSC line, PL74(Supplementary Fig. 5). These results suggest that theexpression level of CD349 might alter the effects on bonerepair to some extent. Nonetheless, CD349-negative cells hadconsiderable effects on bone calcification compared to control.

Transplantation of CD349-negative cells demonstratedsignificant arterio/angiogenic effects following vascularocclusion in mice

Previous studies have shown that MSC frequently localize nearvessels in villi and in parenchymatic tissues of the placenta,suggesting physiological roles for MSC in vascular formation inthe placenta (Battula et al., 2007). In order to examine if theexpression level of CD349 facilitates new vessel formation invivo, a vascular occlusion model was prepared by ligating theproximal end of the femoral vessel with the popliteal vessel inmice. CD349-negative or -positive cells were injectedintramuscularly into the ischemic region and blood flow wasmeasured at the unilateral ankle over a period of time (Fig. 7A).The recovery of blood flow was evaluated by comparing valuesfrom the ischemic and non-ischemic sides of the ankle. As apositive control, endothelial precursor cells (EPC) derivedfrom umbilical cord blood were utilized in this study (Naganoet al., 2007). Prior investigations demonstrated that blood flowrecovery in an ischemic region occurs initially through‘‘arteriogenesis,’’ a process defined as the formation offunctional collateral arteries from pre-existing arterio-arteriolar anastomoses (Heil et al., 2006). It has been reportedthat this process commences one week after femoral vesselocclusion (Ito et al., 1997). Following arteriogenesis, blood flowin the ischemic region is further supported by ‘‘angiogenesis,’’which is characterized by the sprouting of new capillaries frompre-existing vessels. We observed that the recovery of bloodflow on day 3 after surgery was significantly greater in miceinjected with EPC compared to those injected with CD349-negative or -positive cells. These results suggest that EPCdirectly contributes to arteriogenesis and the development ofcollateral blood flow.

Surprisingly, thereafter the blood flow in the ischemic regionrapidly increased in mice injected with CD349-negative cells. Atday 7 and 10 after the transplantation, the relative blood flowwas significantly greater in CD349-negative cell recipientscompared to that in CD349-positive cell recipient mice(Fig. 7A, �P< 0.05). Furthermore, the results were consistentlyobserved in the other MSC line, PL74 (Supplementary Fig. 6).On day 14 following the vascular occlusion, blood flow at theischemic site recovered and resembled the blood flow at thenon-ischemic site, even in control mice (Couffinhal et al., 1998;Tang et al., 2005; Yang et al., 2009). These results indicate thatcollateral blood flow arose naturally by this time point.

Interestingly, we observed a significant accumulation oflectin-binding endothelial cells (EC) in the EPC and CD349-negative cell recipients at the femoral region on day 14 (Fig. 7B).In contrast, EC were sparsely distributed in control andCD349-positive cell recipient mice. The EC at the femoralregion are considered to contribute to angiogenesis bysupplying the blood flow to the peripheral region. Overall, thesedata strongly indicate that CD349-negative PL-MSC showeda greater ability of re-endothelialization compared to CD349-positive PL-MSC.

In order to determine how CD349-negative cells supportnew vessel formation, the mRNA expressions of angiogenicfactors were evaluated by quantitative PCR under normoxicand hypoxic conditions (Fig. 7C). We found that the expressionlevels of PDGFb and bFGF were greater in CD349-negativecells than in CD349-positive cells under hypoxic conditions.

Fig. 5. Separation of PL-MSC on the basis of CD349 expression. PL73 was separated into CD349-negative and -positive cells by FACS.B,C: CD349-negative cells (B) and CD349-positive cells (C) were analyzed for the expressions of CD14, CD31, CD34, CD45, CD13, CD90,CD73, CD105, CD166, HLA-ABC, HLA-DR, SSEA4, and CD271 by FACS. Note the comparable expression profiles between the two CD349fractions.D:Theexpressionsof immaturecellmarkers inCD349-negativeand-positivecellswereexaminedbyRT-PCR.BM-MSCwasutilizedasapositive control.


232 T R A N E T A L .

Fig. 6. Analysis of the differentiation potential of PL73 CD349-negative and -positive cells in a mouse bone fracture model. PL73CD349-negative (left) and CD349-positive (right) cells weretransplanted into mice at the sites of bone fracture using Gelformas a carrier vehicle. The transplantation sites were analyzed forosteocalcification by X-rays (A) and density measurements (B).Note that osteocalcification in the transplant of CD349-positive cellswas greater than that of CD349-negative cells. MP < 0.05, MMP < 0.01.


In addition, the hypoxic induction of PDGFb and bFGF mRNAswas significantly greater in CD349-negative cells comparedto that in CD349-positive cells; PDGFb: 2.88� 0.74 folds(CD349-negative cells) versus 1.46� 1.11 folds (CD349-positive cells), and bFGF: 1.95� 0.28 folds (CD349-negativecells) versus 1.84� 0.43 folds (CD349-positive cells; �P< 0.05).On the other hand, the expression of VEGF mRNA was clearlyupregulated by hypoxia in CD349-negative and CD349-positivecells, but there was no significant difference between them.CD349-negative and -positive cells showed a comparableexpression profile of TGF-b, Ang1, the matrixmetalloproteinases MMP-2, and MMP-9 (Fig. 7C and data notshown).

In summary, the expression levels of the two angiogenicfactors PDGFb and bFGF were greater in the CD349-negativecells than in the CD349-positive cells. These angiogenic factorsmight contribute to the significant arterio / angiogenic effects ofthe CD349-negative cells in vivo.

Discussion

In the present study, we established several MSC lines fromhuman placenta and characterized these cells by analyzing the


expression of cell surface markers and the ability todifferentiate both in vitro and in vivo. Engraftment of bothCD349-positive and -negative PL-MSC subfractionssuccessfully facilitated the bone calcification of fracturedfemurs. Interestingly, the CD349-negative fraction showedsignificant effects on new vessel formation in ischemic tissuefollowing vascular occlusion in mice. These results indicate thatthe CD349-negative and -positive MSC possess partiallyoverlapping, but distinct features in their differentiationpotential. On the basis of these findings, we propose thatCD349 might be utilized as a specialized marker for PL-MSC inarterio/angiogenic therapy. However, we should note thatCD349 may not be a suitable marker to discriminate MSC andnon-MSC from placenta, because the distribution of CD349appeared different in PL73 and PL74 (Fig. 5 and SupplementaryFig. 3) and both CD349-positive and -negative fractionswere observed in the two non-MSC line, PL57 and PL58(Supplementary Fig. 4). Given that human placenta is anattractive source of MSC for cell therapy based clinicalapplications, our data provide a novel insight into the isolationof placenta-derived MSC that have advantageous effects onarteriogenesis and angiogenesis.

The properties of MSC are generally defined by theirpotential to differentiate into mesenchymal and non-mesenchymal cells in cell culture experiments. Severalconventional protocols exist for the in vitro differentiation ofMSC for a variety of cell types (Kogler et al., 2004; Lee et al.,2004). Most of them require several weeks to induce theterminal stage of differentiation. In this study, we screened theosteogenic differentiation of MSC lines by an increase in ALPmRNA expression in response to EGF after 9 days ofcultivation. Furthermore, we confirmed the ability todifferentiate into osteogenic, chondrogenic and adipogeniclineages using conventional protocols. In contrast, PL57, whichshowed no induction of ALP mRNA by EGF, failed todifferentiate into osteogenic, adipogenic and chondrogeniclineages. Thus, measurement of the ALP mRNA level in thepresence of EGF might provide an indication of thedifferentiation potential of MSC and would be a valuablescreening procedure for isolating MSC.

Despite the failure of PL57 to differentiate into anymesenchymal cell lineages in vitro, this cell line wasindistinguishable from PL73 by the expression of multiple cellsurface antigens. We also demonstrated that PL57 failed tocontribute to the processes of bone repair in vivo. These datasuggest that it might be difficult to distinguish between MSC andnon-MSC solely by the expression of cell surface markers.Although the origin of PL57 is uncertain, it is conceivable thatseveral MSC-specific surface antigens are expressed on acertain class of placenta-derived mesenchymal cells that lost thepotential to differentiate toward multiple lineages. The specificisolation of MSC by the exclusive use of cell surface markerswould indeed be a speedy and useful method. However,we realized that the combined assessment of differentiationpotential and cell surface marker expression is necessary forisolating placenta-derived MSC for clinical applications.

Although we and others have confirmed that engraftment ofMSC facilitates collateral blood flow in ischemic regions, howMSC contribute to new vessels is still controversial. It has beenreported that MSC can differentiate into vascular endothelialcells (Nagaya et al., 2004; Pittenger and Martin, 2004; Urbichand Dimmeler, 2004; Wu et al., 2005; Jiang et al., 2006).However, direct in vivo evidence for the contribution ofMSC-derived endothelial cells to new vessel formation islacking. Even though MSC might be able to directly differentiateinto EC in vivo, the proliferation and migration of host-derivedendothelial cells in ischemic regions might be necessary for newvessel development. Alternatively, MSC have been reportedto have a paracrine function by secreting angiogenic factors

Fig. 7. Studyof theangiogeniceffectsofCD349-negativeand-positivecells.A,B:Angiogeniceffectsweredeterminedafterthetransplantation ofPL73 CD349-negative or -positive cells into a mouse model of vascular occlusion. The ratio of ischemic blood flow to non-ischemic blood flow wasmeasured at the inner ankle on days 0, 1, 3, 5, 7, 10, and 14 post transplantation and demonstrated that CD349-negative cells (white square withdot line) were more effective at recovering blood flow than CD349-positive cells (black circle with dot line, A). As controls we used PBS alone(white circle with solid line) or endothelial progenitor cells (EPC; triangle with dot line) derived from umbilical cord blood. MP < 0.05. Vesselformations were measured after lectin-TRIC injections on day 14 (B). Bar indicates 50mm. The number of vessels was scored. MP < 0.05, MMP < 0.01.C: The expressions of angiogenic factors were examined by quantitative PCR after cells were cultured under normoxic conditions (N) or hypoxicconditions(H).RelativemRNAexpressionsweremeasuredandthedataobtained fromCD349-negativecells culturedundernormoxicconditionswere normalized to a value of 1 as the standard. Grey columns: CD349-negative cells; black columns: CD349-positive cells. MP < 0.05, MMP < 0.01.


234 T R A N E T A L .


(Kinnaird et al., 2004; O’Neill et al., 2005; Takahashi et al., 2006;Liu et al., 2008). This paracrine function is a likely explanationfor the significant effects of CD349-negative cells on bloodflow recovery in our experiments, since CD349-positive and-negative cells displayed a clear difference in the expression ofangiogenic factors. VEGF mRNA expression was significantlyupregulated under hypoxic conditions in both CD349-negativeand CD349-positive cells. Interestingly, PDGFb and bFGFmRNA levels and those inductions were greater in CD349-negative cells than in CD349-positive cells under hypoxicconditions. It is noteworthy that CD349-positive and -negativecells showed a similar expression profile of the other angiogenicfactors investigated, such as TGF-b, Ang1, MMP2, MMP9 andMMP14 (Fig. 7C and data not shown). Thus, PDGFb and bFGFmight play specific roles in promoting new vessels in placenta-derived MSC. In contrast to our data, it has been reported thatMSC enhance angiogenesis by secreting VEGF and Ang1 duringwound healing in mice (Wu et al., 2007). Further study isrequired to clarify the roles of PDGFb and bFGF in PL-MSC byusing gene-transduced MSC in transplantation experiments.

In this study, we were able to separate a line of MSC intotwo sub fractions on the basis of CD349 expression. Ourtransplantation experiments in mice revealed that CD349-negative MSC have significant effects on new vessel formation inischemic regions, while both the CD349-positive and -negativefractions facilitate new bone calcification in fractured femurs.On the basis of these findings, it is proposed that CD349 mightbe utilized as a specialized MSC marker for arteriogenesis andangiogenesis. Further understanding of the molecular basis ofMSC differentiation in vivo might allow for the development ofeffective cell therapy.

Acknowledgments

We thank Dr. Daisuke Nozawa and Ms. Naomi Kanekofor technical advice and excellent work on theimmunohistochemistry. We also thank Tania O’Connorfor critical reading of the manuscript.

Literature Cited

Alviano F, Fossati V, Marchionni C, Arpinati M, Bonsi L, Franchina M, Lanzoni G, Cantoni S,Cavallini C, Bianchi F, Tazzari PL, Pasquinelli G, Foroni L, Ventura C, Grossi A, Bagnara GP.2007. Term Amniotic membrane is a high throughput source for multipotent MesenchymalStem Cells with the ability to differentiate into endothelial cells in vitro. BMC Dev Biol21:7–11.

Battula VL, Bareiss PM, Treml S, Conrad S, Albert I, Hojak S, Abele H, Schewe B, Just L,Skutella T, Buhring HJ. 2007. Human placenta and bone marrow derived MSC cultured inserum-free, b-FGF-containing medium express cell surface frizzled-9 and SSEA-4 and giverise to multilineage differentiation. Differentiation 75:279–291.

Battula VL, Treml S, Abele H, Buhring HJ. 2008. Prospective isolation and characterization ofmesenchymal stem cells from human placenta using a frizzled-9-specific monoclonalantibody. Differentiation 76:326–336.

Bensidhoum M, Chapel A, Francois S, Demarquay C, Mazurier C, Fouillard L, Bouchet S,Bertho JM, Gourmelon P, Aigueperse J, Charbord P, Gorin NC, Thierry D, Lopez M. 2004.Homing of in vitro expanded Stro-1- or Stro-1þ human mesenchymal stem cells into theNOD/SCID mouse and their role in supporting human CD34 cell engraftment. Blood103:3313–3319.

Bilic G, Zeisberger SM, Mallik AS, Zimmermann R, Zisch AH. 2008. Comparativecharacterization of cultured human term amnion epithelial and mesenchymal stromal cellsfor application in cell therapy. Cell Transplant 17:955–968.

Buhring HJ, Battula VL, Treml S, Schewe B, Kanz L, Vogel W. 2007. Novel markers for theprospective isolation of human MSC. Ann NY Acad Sci 1106:262–271.

Couffinhal T, Silver M, Zheng LP, Kearney M, Witzenbichler B, Isner JM. 1998. Mouse modelof angiogenesis. Am J Pathol 152:1667–1679.

Delorme B, Charbord P. 2007. Culture and characterization of human bone marrowmesenchymal stem cells. Methods Mol Med 140:67–81.

Faulk WP, Labarrere CA, Carson SD. 1990. Tissue factor: Identification and characterizationof cell types in human placentae. Blood 76:86–96.

Fleischman RA, Mintz B. 1979. Prevention of genetic anemias in mice by microinjection ofnormal hematopoietic stem cells into the fetal placenta. Proc Natl Acad Sci USA 76:5736–5740.


Fukuchi Y, Nakajima H, Sugiyama D, Hirose I, Kitamura T, Tsuji K. 2004. Humanplacenta-derived cells have mesenchymal stem/progenitor cell potential. Stem Cells22:649–658.

Gang EJ, Bosnakovski D, Figueiredo CA, Visser JW, Perlingeiro RC. 2007. SSEA-4 identifiesmesenchymal stem cells from bone marrow. Blood 109:1743–1751.

Heil M, Eitenmuller I, Schmitz-Rixen T, Schaper W. 2006. Arteriogenesis versus angiogenesis:Similarities and differences. J Cell Mol Med 10:45–55.

Ito WD, Arras M, Winkler B, Scholz D, Schaper J, Schaper W. 1997. Monocyte chemotacticprotein-1 increases collateral and peripheral conductance after femoral artery occlusion.Circ Res 80:829–837.

Jiang W, Ma A, Wang T, Han K, Liu Y, Zhang Y, Zhao X, Dong A, Du Y, Huang X, Wang J, Lei X,Zheng X. 2006. Intravenous transplantation of mesenchymal stem cells improves cardiacperformance after acute myocardial ischemia in female rats. Transpl Int 19:570–580.

Kawamoto T, Shimizu M. 2000. A method for preparing 2- to 50-mm-thick fresh-frozensections of large samples and undecalcified hard tissues. Histochem Cell Biol 113:331–339.

Kinnaird T, Stabile E, Burnett MS, Lee CW, Barr S, Fuchs S, Epstein SE. 2004. Marrow-derivedstromal cells express genes encoding a broad spectrum of arteriogenic cytokines andpromote in vitro and in vivo arteriogenesis through paracrine mechanisms. Circ Res94:678–685.

Kogler G, Sensken S, Airey JA, Trapp T, Muschen M, Feldhahn N, Liedtke S, Sorg RV, Fischer J,Rosenbaum C, Greschat S, Knipper A, Bender J, Degistirici O, Gao J, Caplan AI, Colletti EJ,Almeida-Porada G, Muller HW, Zanjani E, Wernet P. 2004. A new human somatic stem cellfrom placental cord blood with intrinsic pluripotent differentiation potential. J Exp Med200:123–135.

Koike J, Takagi A, Miwa T, Hirai M, Terada M, Katoh M. 1999. Molecular cloning of Frizzled-10, a novel member of the Frizzled gene family. Biochem Biophys Res Commun 262:39–43.

Kratchmarova I, Blagoev B, Haack-Sorensen M, Kassem M, Mann M. 2005. Mechanism ofdivergent growth factor effects in mesenchymal stem cell differentiation. Science308:1472–1477.

Lee OK, Kuo TK, Chen W-M, Lee KD, Hsieh SL, Chen TH. 2004. Isolation of multipotentmesenchymal stem cells from umbilical cord blood. Blood 103:1669–1675.

Liu K, Chi L, Guo L, Liu X, Luo C, Zhang S, He G. 2008. The interactions between brainmicrovascular endothelial cells and mesenchymal stem cells under hypoxic conditions.Microvasc Res 75:59–67.

Magatti M, De Munari S, Vertua E, Gibelli L, Wengler GS, Parolini O. 2008. Human amnionmesenchyme harbors cells with allogeneic T-cell suppression and stimulation capabilities.Stem Cells 26:182–192.

Martinez C, Hofmann TJ, Marino R, Dominici M, Horwitz EM. 2007. Human bone marrowmesenchymal stromal cells express the neural ganglioside GD2: A novel surface marker forthe identification of MSCs. Blood 109:4245–4248.

Miki T, Marongiu F, Ellis EC, Dorko K, Mitamura K, Ranade A, Gramignoli R, Davila J, StromSC. 2009. Production of hepatocyte-like cells from human amnion. Methods Mol Biol481:155–168.

Nagano M, Yamashita T, Hamada H, Ohneda K, Kimura K, Nakagawa T, Shibuya M,Yoshikawa H, Ohneda O. 2007. Identification of functional endothelial progenitor cellssuitable for the treatment of ischemic tissue using human umbilical cord blood. Blood110:151–160.

Nagaya N, Fujii T, Iwase T, Ohgushi H, Itoh T, Uematsu M, Yamagishi M, Mori H, Kangawa K,Kitamura S. 2004. Intravenous administration of mesenchymal stem cells improves cardiacfunction in rats with acute myocardial infarction through angiogenesis and myogenesis.Am J Physiol Heart Circ Physiol 287:H2670–2676.

Ohneda O, Ohneda K, Arai F, Lee J, Miyamoto T, Fukushima Y, Dowbenko D, Lasky LA, SudaT. 2001. ALCAM (CD166): Its role in hematopoietic and endothelial development. Blood98:2134–2142.

O’Neill TJ VI, Wamhoff BR, Owens GK, Skalak TC. 2005. Mobilization of bone marrow-derived cells enhances the angiogenic response to hypoxia without transdifferentiationinto endothelial cells. Circ Res 97:1027–1035.

Parolini O, Alviano F, Bagnara GP, Bilic G, Buhring HJ, Evangelista M, Hennerbichler S, Liu B,Magatti M, Mao N, Miki T, Marongiu F, Nakajima H, Nikaido T, Portmann-Lanz CB, SankarV, Soncini M, Stadler G, Surbek D, Takahashi TA, Redl H, Sakuragawa N, Wolbank S,Zeisberger S, Zisch A, Strom SC. 2008. Concise review: Isolation and characterization ofcells from human term placenta: Outcome of the First International Workshop on PlacentaDerived Stem Cells. Stem Cells 26:300–311.

Pittenger MF, Martin BJ. 2004. Mesenchymal stem cells and their potential as cardiactherapeutics. Circ Res 95:9–20.

Portmann-Lanz CB, Schoeberlein A, Huber A, Sager R, Malek A, Holzgreve W, Surbek DV.2006. Placental mesenchymal stem cells as potential autologous graft for pre- and perinatalneuroregeneration. Am J Obstet Gynecol 194:664–673.

Taguchi K, Ogawa R, Migita M, Hanawa H, Ito H, Orimo H. 2005. The role of bone marrow-derived cells in bone fracture repair in a green fluorescent protein chimeric mouse model.Biochem Biophys Res Commun 331:31–36.

Takahashi T, Takahashi K, Mernaugh RL, Tsuboi N, Liu H, Daniel TO. 2006. A monoclonalantibody against CD148, a receptor-like tyrosine phosphatase, inhibits endothelial-cellgrowth and angiogenesis. Blood 108:1234–1242.

Tang GL, Chang DS, Sarkar R, Wang R, Messina LM. 2005. The effect of gradual or acutearterial occlusion on skeletal muscle blood flow, arteriogenesis, and inflammation in rathindlimb ischemia. J Vasc Surg 41:312–320.

Urbich C, Dimmeler S. 2004. Endothelial progenitor cells: Characterization and role invascular biology. Circ Res 95:343–353.

Wu X, Huang L, Zhou Q, Song Y, Li A, Jin J, Cui B. 2005. Mesenchymal stem cells participatingin ex vivo endothelium repair and its effect on vascular smooth muscle cells growth. Int JCardiol 105:274–282.

Wu Y, Chen L, Scott PG, Tredget EE. 2007. Mesenchymal stem cells enhance wound healingthrough differentiation and angiogenesis. Stem Cells 25:2648–2659.

Yang Z, von Ballmoos MW, Diehm N, Baumgartner I, Kalka C, Santo SD. 2009. Call for areference model of chronic hind limb ischemia to investigate therapeutic angiogenesis.Vasc Pharmacol 51:268–274.

Yen BL, Huang HI, Chien CC, Jui HY, Ko BS, Yao M, Shun CT, Yen ML, Lee MC, Chen YC.2005. Isolation of multipotent cells from human term placenta. Stem Cells 23:3–9.

paper (trần cẩm tú) identiﬁcation of human placenta-derived mesenchymal stem cells involved...

Documents

amnionic msc

msc types

types of msc

isolation of msc

cd349positive cells

theexpression of msc

cd349negative cells

chorionic msc canbe