Oligopotent Mesenchymal Stem Cell-Like Clone Becomes Multinucleated Following Phorbol Ester TPA Stimulation.код для вставкиСкачать
THE ANATOMICAL RECORD 120:1256–1267 (2007) Oligopotent Mesenchymal Stem Cell-Like Clone Becomes Multinucleated Following Phorbol Ester, TPA Stimulation KEIICHIRO YOSHIDA,* MICHIO ONO, TATSUO MAEJIMA, MICHIYO ESAKI, AND HAJIME SAWADA Department of Histology and Cell Biology, Yokohama City University School of Medicine, Kanazawa-ku, Yokohama, Japan ABSTRACT We established a mesenchymal stem cell clone, 5F9A, from rat bone marrow substrate adherent cells by repeated limiting dilutions. The cells have a ﬁbroblastic shape and form intimate contacts with adjacent cells with interdigitations and junctions similar to adherence and tight junctions in a semi-conﬂuent culture. Analysis of the phenotypes of these cells by RT-PCR and FACS demonstrated that they resembled mesenchymal stem cells, and the cells could differentiate into adiopocytes and osteoblasts under appropriate conditions in vitro showing their oligopotency. Furthermore, the cells were induced to become multinuclear cells by TPA (12-o-tetradecanoylphorbol 13-acetate) stimulation. Anat Rec, 290:1256– 1267, 2007. Ó 2007 Wiley-Liss, Inc. Key words: mesenchymal stem cell; junction; phorbol ester; TPA; multinucleation Mesenchymal stem cells (MSCs) reside in the bone marrow, like hematopoietic stem cells, and can differentiate into various types of cells in appropriate conditions, both in vitro and in vivo (Pittenger et al., 1999; Jiang et al., 2002). MSCs can differentiate into tendon, endothelial, and nerve cells, as well as osteoblasts, adipocytes, and chondrocytes (Pittenger et al., 1999; Dezawa et al., 2001, 2004; Jiang et al., 2002). MSCs can also become stromal cells, supporting the differentiation of hematopoietic stem cells into mature blood cells (Dennis and Charbord, 2002). MSCs are ﬁbroblastic in shape, adhere to culture substrates, and do not phagocytose particles (Castro-Malaspina et al., 1980). By ﬂuorescent-activated cell sorting (FACS) analysis, MSCs were classiﬁed into small and agranular cells (Colter et al., 2000). As for the cell surface markers, MSCs express CD105 (SH-2; Barry et al., 1999), CD73 (SH-3 and SH-4; Barry et al., 2001), CD90, STRO-1, and CD106, but not the hematopoietic markers CD45 and CD34 (Tocci and Forte, 2003). However, it was also reported that MSCs freshly prepared from bone marrow aspirates are phenotypically different from those cultured in vitro for some time (Gronthos et al., 2003). Jiang et al. (2002) suggested that a single MSC Ó 2007 WILEY-LISS, INC. could differentiate into almost all types of cells in the body. Many cells develop junctional structures when they make contact with each other. These structures include tight junctions, adherence junctions, gap junctions, desmosomes, and interdigitations, and cells communicate through these junctions. Although these junctional structures are not prominent for cells of mesenchymal origin, human mesenchymal stem cells are reported to build gap junctions among themselves or with cardiomyocytes in some in vitro culture systems (Valiunas et al., Grant sponsor: The Ministry of Education, Culture, Sports, Science and Technology, Japan Grant-in-Aid for Scientiﬁc Research; Grant numbers: 14370008 and 16659048. *Correspondence to: Keiichiro Yoshida, Department of Histology and Cell Biology, Yokohama City University School of Medicine, Fukuura 3–9, Kanazawa-ku, Yokohama, Japan 236-0004. Fax: 81-45-787-2568. E-mail: email@example.com Received 29 November 2006; Accepted 12 July 2007 DOI 10.1002/ar.20590 Published online in Wiley InterScience (www.interscience.wiley. com). MULTINUCLEATION OF MESENCHYMAL STEM CELL 2004; Beeres et al., 2005). These cells interact electrically and can repair experimentally induced conduction blocks. TPA is a potent tumor promoter and known to be a protein kinase C (PKC) activator. TPA has been reported to induce formation of multinuclear cells, either by cell– cell fusion (syncytium; Hassan et al., 1989; David et al., 1990) or by nuclear division without cell division (plasmodium; Menaya and Clemens, 1991). Syncytia occur in striated muscle cells and macrophage-related cells, and their formation results in the generation of multinuclear cells (Anderson, 2000; Taylor, 2003). Plasmodia were reported in some tumors stimulated with TPA (Menaya and Clemens, 1991), and it was reported that this effect was dependent on PKCa and PKCd activation (Watanabe et al., 1992; Yamaguchi et al., 1995). TPA has also been reported to induce endomitosis (Murate et al., 1991; Bermejo et al., 2002). An endomitotic cell has multiple sets of its genome, for example, 4n, 8n, 16n, or more in one nucleus (Vitrat et al., 1998). Endomitosis occurs in leukemic cells (e.g., K562, HEL MEG-01) stimulated by TPA, and these cells change into megakaryocytes (Murate et al., 1991; Bermejo et al., 2002). In this report, we established an oligopotent (Smith, 2006) mesenchymal stem cell clone that could differentiate into adipocytes and osteoblasts, and we investigated the ﬁne structure of these cells. Upon stimulation with TPA, these cells converted into multinucleated cells. MATERIALS AND METHODS Cell Culture Normal rat bone marrow (BM) was obtained from the femurs of DA/Slc rats (Japan SLC. Inc., Hamamatsu, Japan). Eagle’s minimum essential medium with alpha modiﬁcation (a-MEM; Sigma, St. Louis, MO) including 10% fetal calf serum (Lot.S04301S1820, BioWest, Miami, FL), 300 mg/ml L-glutamine (WAKO, Osaka, Japan), and 60 mg/ml kanamycin sulfate (WAKO) was used for culturing cells. Culture was performed at 378C in a humidiﬁed atmosphere containing 5% CO2 and 95% air. The medium was changed every third day. A few MSC clones were obtained through successive limiting dilutions, and clone 5F9A was used in this study. Transmission Electron Microscopy Cells were ﬁxed with 1% glutaraldehyde and 2% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) overnight at 48C, post-ﬁxed with 1% OsO4 for 1 hr on ice, and further stained with 0.5% uranyl acetate for 30 min at room temperature. Fixed and stained cells were dehydrated with ascending concentrations of ethanol, and embedded in Epon 812. Ultrathin sections were obtained and stained with 2% uranyl acetate in 70% ethanol and 0.4% lead citrate, and observed using a H-7500 transmission electron microscope (Hitachi, Tokyo, Japan) operated at 80 kV. Immunohistochemistry Cells on coverslips were ﬁxed with PLP solution (0.01 M NaIO4, 0.075 M lysine, 0.0375 M phosphate 1257 buffer, and 2% paraformaldehyde, pH 6.2), blocked with 0.4% bovine serum albumin in phosphate buffered saline (BSA-PBS), and permeabilized with 0.1% Triton X-100 in BSA-PBS. Thereafter, the cells were labeled with the primary antibodies and secondary antibodies (Alexa Fluor1488 goat anti-mouse IgG and anti-rabbit IgG; Invitrogen, Carlsbad, CA). The stained samples were observed by confocal laser scanning microscopy (LSM510, Carl Zeiss Ltd. Oberkochen, Germany). For the primary antibodies, rabbit anti–ZO-1 antibody (Invitrogen), mouse anti–b-catenin monoclonal antibody (BD Transduction Lab., Lexington, KY), mouse anti-desmoglein antibody (BD Transduction Lab.), and mouse anti–E-cadherin monoclonal antibody (BD Transduction Lab.) were used in this study. Reverse-Transcription Polymerase Chain Reaction Total RNA was extracted from 5F9A cells using TRI REAGENTTM RNA, DNA, and protein isolation solution (Sigma) according to the manufacturer’s instructions. cDNA was synthesized by reverse transcription with M-MLV reverse transcriptase (Invitrogen). Polymerase chain reaction (PCR) was performed in a PCR thermal cycler (Takara, Tokyo, Japan), and the products were subjected to electrophoresis on 2% agarose gels after 35 cycles of ampliﬁcation and stained with 100 ng/ml ethidium bromide. The sizes of the PCR products were estimated by comparison with Marker 4 (WAKO). Oligonucleotide primers used in PCR are shown in Table 1. FACS For FACS analysis, we used biotin–anti-CD29 antibody, ﬂuorescein isothiocyanate (FITC) –anti-CD45 antibody (BD Biosciences Pharmingen, San Diego, CA), antiCD11b antibody, FITC–anti-CD44 antibody (Chemicon, Tamecula, CA), and FITC–anti-CD90 antibody (eBioscience, San Diego, CA). As a secondary antibody, we used FITC-Streptavidin (Vector, Burlingame, CA) and FITC goat anti-mouse IgG (Fab0 ) antibody (PROTOS IMMUNORESEARCH, San Francisco, CA). The expression of the antigens was analyzed using a MoFlo ﬂow cytometer (Dako Cytomation, Fort Collins, CO). Phagocytosis of Latex Beads 5F9A cells were cultured in medium containing latex beads (1.07-mm diameter, Sigma) for 3 hr. These cells were washed gently with PBS, ﬁxed with methanol, and stained with Giemsa solution (MERCK, Dermstadt, Germany). The cells were observed using a light microscope. Induction of Differentiation For 5F9A cells to differentiate into osteoblasts, they were cultured in a-MEM containing 50 mM ascorbate-2phosphate (Asc-2-P, Sigma), 10 mM b-glycerophosphate (Sigma), and 0.1 mM dexamethasone (WAKO; Zuk et al., 2001). The medium was changed every third day. After 5 weeks, total RNA was extracted from the cells and applied to reverse transcription (RT) PCR, and the cells were stained by von Kossa staining for the detection of calcium. Concerning the differentiation into adipocytes, the cells were cultured in a-MEM containing 4 mg/ml 1258 YOSHIDA ET AL. TABLE 1. Primers used for reverse-transcription polymerase chain reaction Oligonucleotide sequence Rat CD14 Rat CD29 Rat CD90 Rat CD104 Rat CD106 Rat CD126 Rat c-kit Rat CD45 Rat CD44 Rat CD34 Mouse CD73 Rat CD105 Rat E-Cadherin Rat N-Cadherin Rat Runx2 Rat PTHR Rat BSP Rat Osteonectin Rat GPIIb Rat GPV Rat MyoD Rat Myf-5 Mouse Myogenin Rat albumin Rat a-Fetoprotein Rat Integrin aV Rat Integrin b3 Rat Cathepsin K Rat c-fms Rat TRAP Rat RANK Rat CTR Rat b-actin 0 0 5 -AGTCGGAGGCGTATAACTCTG-3 50 -GTAGAGCGCTTGCCTAGTGA-30 50 -TAAGTGCACAGATCCCAAGTT-30 50 -TGATGTCGGGACCAGTAG-30 50 -CCGTCATGAGAATAACACCA-30 50 -CCCAACCAGTCACAGAGAAAT-30 50 -TCCCTATTGCGACTACGAAAT-30 50 -TGGCCAGGTTGATGATAGC-30 50 -ACGTTGCTCCGAAAGAAC-30 50 -AGATGGGAAGACTGTAAGCTG-30 50 -GTTCAGTTCGCACAGGTGTAT-30 50 -GTTTCCTTTCTCCCAAGACA-30 50 -ACAACGGCACGGTGGAGT-30 50 -GGGCACTTGGTTTGAGCATCT-30 50 -GATCCACCCAGTGACCCGTCT-30 50 -ACCCTGGGCAGCAATCATCAT-30 50 -GGGACTACTTTGCCTCTTACA-30 50 -ATGCGAGGAGGATATACACTC-30 50 -GCGTGAGAATGCCGGTCCA-30 50 -TCAGCAAACACTCGGGCCTAA-30 50 -CTGTGCTGCCCTTTGGAG-30 50 -CGGCTGGGTAAACTACTTTCA-30 50 -AGACAGCTGCCATCTGGATTT-30 50 -GGGCACGTGTGTGAGAATAGA-30 50 -GTGTGGGTCTGGAGATCGCATCTTAC-30 50 -AGAGATTCTTGTTCTGCGTCATTGTACTCA-30 50 -AGTGCAGAGAACTCGGAACAT-30 50 -TGCTAGTGCTGCCAGTACCA-30 50 -CCAGATGGGACTGTGGTTACC-30 50 -ACTTGGTGCAGAGTTCAGGG-30 50 -CCGGGTGCTTGCCACTAA-30 50 -GCTTTGCGCTTGAAGTCCAAC-30 50 -ATCGTAAATCAGGGCTGTTGA-30 50 -GGCGTCTCACACCGTGTAT-30 50 -CGCATGCGTGACTGGC-30 50 -CAAAGAAGCGAGTGGTGCAAT-30 50 -CCTGGAAGAAGCTGGAGATT-30 50 -CGAGTACGGTAGGGAGGAGA-30 50 -GCTCTTCCGTAACCTCAGCAG-30 50 -TTTCCCATTGCCTCAAACAAG-30 50 -TGGCGCCGCTGCCTTCTACG-30 50 -ACACGGCCGCACTCTTCCCTG-30 50 -CAGCCAAGAGTAGCAGCCTTCG-30 50 -GTTCTTTCGGGACCAGACAGGG-30 50 -TGGAGCTGTATGAGACATCCC-30 50 -TGGACAATGCTCAGGGGTCCC-30 50 -CAGGCCACTATCTCCAGCAAA-30 50 -GTGCCCACTCTTCCCAGGTT-30 50 -TGGCCGACATTTACATTG-30 50 -AAGAGGCCAGAGAAATCAGCA-30 50 -TGGGCCGATGAACTGTACTGC-30 50 -GGCGCTCTTCCCTCTATCCAG-30 50 -AAAATCCGTTCTAAAGTAGA-30 50 -ATCTCCCCTTTGTAGCGGA-30 50 -ATGTATAACGCCACGGCAAAG-30 50 -CAGATGGGCTGGCTGGCTTGA-30 50 -GGCGTAGATACCTATGTGGAG-30 50 -GGAGGATGCCGTAGGAC-30 50 -CAAGAACTTGAGGCCATTGTT-30 50 -CAGAGTACCAGGGCAGAGAAA-30 50 -TTAAGCCAGTGCTTCACGGG-30 50 -ACGTAGACCACGATGATGTCGC-30 50 -ACACCCTGACAGCAACCGAACCT-30 50 -GAACCCCCAGCCAAGTAAATAGTA-30 50 -AACTGGGACGATATGGAGAA-30 50 -GTAACCCTCATAGATGGGCA-30 Product (base pair) Annealing (8C) 390bp 558C 335bp 568C 382bp 568C 392bp 568C 308bp 568C 443bp 558C 435bp 608C 306bp 608C 308bp 558C 744bp 608C 387bp 558C 374bp 558C 451bp 558C 353bp 558C 380bp 558C 315bp 588C 560bp 558C 316bp 588C 682bp 548C 549bp 548C 230bp 608C 440bp 608C 184bp 608C 503bp 588C 310bp 568C 220bp 558C 556bp 558C 382bp 558C 379bp 558C 438bp 618C 497bp 558C 447bp 558C 276bp 608C MULTINUCLEATION OF MESENCHYMAL STEM CELL 1259 Fig. 2. Junctional structures between 5F9A cells. A–C: An adherence junction-like structure (A), a tight junction-like structure (B), and an interdigitation-like structure (C) are indicated by arrows. D: Fluorescence micrograph of anti–ZO-1 antibody staining of 5F9A cells. The borders of the cells were stained by the antibody (arrow). Scale bars 5 600 nm in A, 100 nm in B, 200 nm in C, 100 mm in D. insulin (Sigma), 500 mM 3-isobutyl-1-methylxanthine (IBM, WAKO), 0.4 mM dexamethasone (WAKO), and 60 mM indomethacin (WAKO) for 3 weeks, changing the medium twice a week, and stained with Sudan III for the detection of lipids. In both cases, the culture dishes were coated with Growth Factor Reduced MATRIGEL1Matrix (40 mg/ml; BD Biosciences, Discovery Labware, Bedford, MA), and the initial cell density was 2 3 104 cells per 2 ml of medium in a 35 mm dish. Induction of Multinuclear Cells For the induction of multinuclear cells (MNCs), cells were cultured in a 35-mm dish in the presence of 20 ng/ ml TPA (Sigma) for 3 days. For light microscopy, the cells were ﬁxed with absolute methanol for 5 min and stained with Giemsa solution (MERCK). Additionally, TPA-stimulated cells were ﬁxed with 10% formalin, permeabilized with 0.5% Triton X-100 in PBS, and stained with SYBR Green I (SYBR-I; Molecular Probes, Inc., Eugene, OR). Twenty optical sections of multinuclei along the z-axis (0.9 mm in thickness and 0.45 mm in focus step) were analyzed using confocal laser scanning microscopy (LSM510, Carl Zeiss Ltd.). Fig. 1. The 5F9A cells. A: Photograph of 5F9A cells. The 5F9A was established from adherent cells in DA/Slc rat bone marrow by repeated limiting dilutions. The cells were ﬁxed with absolute methanol and stained by Giemsa solution. B: Electron micrographs of 5F9A cells. The arrow indicates heterochromatin, the arrowhead indicates Golgi apparatus, the asterisk indicates polysomes, and the white arrow indicates rough endoplasmic reticulum. C: Actin bundle is indicated beneath the plasma membrane at the basal surface (arrow). Scale bars 5 25 mm in A, 1 mm in B,C. 1260 YOSHIDA ET AL. Fig. 3. Phenotype of 5F9A cells. A: Reverse transcription-polymerase chain reaction was performed using total RNA extracted from nontreated and TPA-treated 5F9A cells with speciﬁc primers, which are shown in Table 1. The left column shows the results with total RNA from 5F9A cells without TPA stimulation. The column in the cen- ter shows the results from the TPA-stimulated cells. Total RNA prepared from normal rat spleen cells was used as a control (right column). B: Phenotype analyses using ﬂuorescent-activated cell sorting. 5F9A cells stained with ﬂuorescence were analyzed using a MoFlo ﬂow cytometer. MULTINUCLEATION OF MESENCHYMAL STEM CELL 1261 RESULTS 5F9A cells The 5F9A cell clone was obtained through repeated limiting dilutions of bone marrow adherent cells. After establishment, this clone was maintained continually by dilution with fresh medium because it needed neither feeder cells nor special supplements. The cells spread on plastic surfaces and exhibited a ﬁbroblast-like shape (Fig. 1A). The population doubling time was approximately 20 hr. Disperse heterochromatin could be observed in the nuclei by electron microscopy. Golgi apparati, polysomes, and rough endoplasmic reticulum developed in the cytoplasm, which suggested the presence of active protein synthesis in 5F9A cells (Fig. 1B). Thick actin bundles could be observed on the substratum at the basal surface beneath the plasma membrane (Fig. 1C). Junctional Structures Between 5F9A Cells Conspicuously, the cells formed numerous junctional structures among themselves. This observation is unique considering their probable mesenchymal origin. Some junctions resembled adherence junctions (Fig. 2A), some resembled tight junctions (Fig. 2B), and some seemed to be interdigitations formed by protrusions of one cell ensheathed by the plasma membrane of the adjacent cell (Fig. 2C). Actin bundles were also seen in the vicinity of the adherence junction-like structures. The protrusions were longer than 1 mm and the longest one we observed was approximately 3 mm long. The width appeared to be 100–200 nm. The interdigitations contained actin ﬁlaments extending along their long axis. However, no thickening or undercoating of plasma membranes reminiscent of junctional specialization could be observed between these protrusions and their sheaths (Fig. 2C). Next, we stained 5F9A cells with antibodies that recognize the molecules forming the junctions. This cell clone did not react with anti–E-cadherin antibody (data not shown). Neither could we detect any signal of b-catenin in the cells (data not shown). These ﬁndings suggest that typical adherence junctions are not constructed between 5F9A cells. However, TEM analysis suggested the presence of adherence junction-like structures (Fig. 2A), which may indicate the presence of an atypical member of the adherence junction family. An antibody against the ZO-1 molecule, which is one of the components of tight junctions, stained borders between 5F9A cells (Fig. 2D), and, furthermore, a junction-like structure similar to a tight junction was observed during TEM analysis. Fused outer leaﬂets of the plasma membrane could also be detected (Fig. 2B). In addition, we could not stain 5F9A cells with an anti-desmogrein antibody that reacts with desmosomes (data not shown), which suggests that desmosomes did not exist between the cells. Therefore, the junctional structures between 5F9A cells likely consist of tight junction- and adherence junction-like structures, and protrusions with no communication with the sheath. Phenotype of 5F9A Cells To examine the phenotype of the 5F9A clone, we performed RT-PCR using speciﬁc primers for various markers (Table 1; Fig. 3A). The cells showed positive Fig. 4. The 5F9A cells did not ingest latex beads. The 5F9A cells were cultured with latex beads (1.07 mm diameter). Rat peritoneal adherent cells were used as a positive control, and they could internalize beads (arrows). Scale bar 5 25 mm. reactions for CD90 (Thy-1), CD29 (b1-integrin), CD73 (Ecto-50 -nucleotidase), CD105 (Endoglin), CD44, CD14, and N-Cadherin, but did not express CD126 (IL-6 receptor), CD104 (b4-integrin), CD106 (VCAM-1), E-Cadherin, CD45, CD34, or c-kit. The results of FACS were consistent with the results by RT-PCR. The expression of CD29, CD44, and CD90 but not CD11b and CD45 could be observed both in the presence and absence of TPA (Fig. 3B). Additionally, each histogram shows a single peak of staining except CD44 with TPA, which suggests that the population of cells was homogenous. However, TPA seemed to induce some changes of CD44 expression in the cells. Because the 5F9A clone did not express CD45, CD34, or c-kit (Fig. 3A,B), this clone did not likely originate from hematopoietic lineage cells. Although this clone was positive for CD14 (Fig. 3A), which is a macrophage marker, it did not ingest latex beads, indicating that it was not a phagocyte (Fig. 4). The negative expression of CD11b (Fig. 3B) conﬁrms these results. 5F9A Cells Are Oligopotent 5F9A cells were able to differentiate to contain lipid droplets positive for Sudan III staining in the cytoplasm when cultured in a-MEM medium containing insulin, IBM, dexamethasone, and indomethacin for 3 weeks (Fig. 5A,B). This ﬁnding indicates that 5F9A cells can differentiate into adipocytes, as previously reported for mesenchymal stem cells. After reaching conﬂuence, the cell sheets often rolled up to form aggregates of cells. In these aggregates, adipocytes were very often observed (Fig. 5C). When the cells were cultured for 5 weeks in the presence of Asc-2-P, b-glycerophosphate, and dexamethasone (Zuk et al., 2001), brown deposits were detected on the cells by von Kossa staining (Fig. 6A,B). These deposits were regarded to be calcium, and the cells, therefore, seemed to differentiate into osteoblasts. To conﬁrm osteoblastic differentiation, analysis by RT-PCR was performed. Cells cultured for 5 weeks in osteoblast differentiation medium were shown to express Runx-2 and bone sialoprotein, which are markers of osteoblasts (Laino et al., 2006; Friedman et al., 2006; Fig. 6C), whereas control cells did not express these markers. These results indicated that the 5F9A clone maintained the capacity to differentiate into both adipocytes and osteoblasts under suitable culture conditions, suggesting its oligopotent nature. 1262 YOSHIDA ET AL. Fig. 5. Differentiation into adipocytes. The 5F9A cells were cultured in adipogenic differentiation medium, as described in the Materials and Methods section. A: The cells were ﬁxed and stained by Sudan III staining for adipocytes. Lipid droplets were stained (arrows). B: Sudan III staining of cells cultured without supplements. C: Photograph of transmission electron microscopy of an aggregate of cultured 5F9A cells. There are many lipid droplets in the cells (arrows). Scale bars 5 25 mm in A,B, 5 mm in C. MULTINUCLEATION OF MESENCHYMAL STEM CELL 1263 Fig. 6. Differentiation into osteoblasts. The 5F9A cells were cultured in osteogenic differentiation medium, as described in Materials and Methods. A: The cells were ﬁxed and stained by von Kossa staining for osteoblasts. Calcium deposits could be observed (arrows). B: Shown is the von Kossa staining of the cells cultured without supplements. C: Reverse transcription-polymerase chain reaction analysis of the expression of the osteoblast speciﬁc genes. Total RNA was extracted from cells cultured for 5 weeks in the osteogenic differentiation medium (1) and the control medium (without supplements) (2), and used as templates with the primers listed in Table 1. Asterisks indicate speciﬁc bands of Runx-2 and bone sialoprotein. Scale bars 5 25 mm in A,B. 5F9A Cells Can Be Induced to Develop Into Multinuclear Cells optimal conditions, 1020% of the cells were multinucleated. Although some 5F9A cells spontaneously became multinuclear, the frequency was less than 3%, and the difference between nontreated and TPA-treated 5F9A cells was easily recognizable by microscopic examination. When 5F9A cells were seeded in a 35-mm culture vessel at cell densities of 1 3 104/ml (2 3 104/dish), 2.0 3 1044.0 3 104 cells/dish were multinuclear by day 3 when the cells reached conﬂuence (approximately When 5F9A cells were cultured in the presence of TPA, they developed multiple nuclei, some of which were large and irregular in shape (Fig. 7). This phenomenon was conﬁrmed by confocal laser scanning microscopy, which showed that more than one distinct nucleus existed in one cell following TPA stimulation (Fig. 8). In 1264 YOSHIDA ET AL. Fig. 7. TPA-induced multinucleation. The 5F9A cells were cultured in the presence or absence of 20 ng/ml of TPA for 3 days. Cells were stained with Giemsa solution. Multinucleated cells were induced (arrows). Scale bar 5 25 mm. 2.0 3 105/dish). Upon TPA stimulation, multinuclear cells spread (Fig. 7), and highly distinct stress ﬁbers could be observed. It also appears that cells grew very slowly after becoming multinucleated (data not shown). TPA stimulation did not change the phenotype of 5F9A cells as assessed by RT-PCR (Fig. 3A), and this may reﬂect that only approximately 1020% cells became multinucleated. There are some naturally occurring multinuclear cells, including megakaryocytes (although strictly speaking, they are not multinucleated), striated muscle cells, hepatocytes, and osteoclasts. In RT-PCR analyses, neither nonstimulated nor TPA-stimulated cells expressed GPIIb or GPV (megakaryocyte markers; Lepage et al., 2000), MyoD, Myf-5, or myogenin (striated muscle cell markers; Berkes and Tapscott, 2005), or albumin or a-fetoprotein (hepatocyte markers; Hay et al., 2007; Fig. 9). However, some osteoclastic markers were expressed in 5F9A cells. These cells were positive for integrin aV, integrin b3 (Ptaff and Jurdic, 2001), cathepsin K, M-CSF (macrophage-colony stimulating factor) receptor (c-fms), and TRAP (tartrate-resistant acid phosphatase; Fujisaki et al., 2007), but negative for RANK (receptor activator of NF-jB) and CTR (calcitonin receptor; Myers et al., 1999) both in the presence and absence of TPA stimulation (Fig. 9). Because 5F9A cells are oligopotent, they would likely be stem cells that retain some osteoclast markers. We, therefore, could not acquire any evidence that 5F9A cells differentiated into megakaryocytes, striated muscle, hepatocytes, or osteoclasts after TPA stimulation. Although we tried to induce multinuclear cells with various substances other than TPA during short-term culturing (3 days), a signiﬁcant number of multinuclear cells could not be induced. The substances included dibutyryl cyclic-AMP (250 mg/ml), staurosporine (0.2 mM), activin (10 ng/ml), vitamin A (50 pg/ml), retinoic acid (50 nM), insulin (5 mg/ml), concanavalin A (2 mg/ml), lipopolysaccharide (10 mg/ml), and lysophosphatidic acid (300 ng/ml; data of all materials listed above are not shown). Additionally, rat normal skin ﬁbroblasts did not become multinucleated following TPA stimulation (data not shown), which may suggest that multinucleation is a speciﬁc phenomenon characteristic of 5F9A cells. DISCUSSION 5F9A Cells Are Phenotypically Similar to Mesenchymal Stem Cells RT-PCR and FACS (Fig. 3) revealed that 5F9A cells express CD14, CD29, CD44, CD90, CD73, CD105, and N-Cadherin. They do not express E-Cadherin, CD104, CD106, CD126, CD34, c-kit, CD45, or CD11b (Fig. 3). Deans and Moseley (2000) reported that the phenotypes of MSC were CD90, CD29, and CD106 positive, and CD14, CD34, and CD45 negative. On the contrary, Reyes et al. (2001) reported they were CD44 and CD90 positive, and CD34, CD106, c-Kit, and CD45 negative. The phenotypes of 5F9A cells in the present study were consistent with the results of Reyes et al. and those of Deans and Moseley except CD14 and CD106. 5F9A cells were negative for CD45, CD34, and c-Kit, which suggests that 5F9A cells are not hematopoietic. In human mesenchymal stem cells, CD73 is recognized by SH-3 and SH-4 antibodies (Barry et.al., 2001) and CD105 is recognized by SH-2 antibodies (Barry et al., 1999); Therefore, these antibodies may be applied to the separation and characterization of rat mesenchymal stem cells by FACS or other methods. 5F9A cells are CD44 positive. Deans and Moseley (2000) also reported that CD34-positive hematopoietic stem cells require CD44-mediated signals expressed by mesenchymal stem cells for hematopoiesis. 5F9A cells may, therefore, have the ability to maintain hematopoiesis. 5F9A cells are also positive for CD14, which is one of the components of the LPS (lipopolysaccharide) receptor (Miller et al., 2005) and is regarded as a speciﬁc marker for macrophages and granulocytes (Fig. 3A). This ﬁnding may suggest that 5F9A cells are of macrophage/granulocyte lineage. However, 5F9A cells were CD11b negative and could not internalize latex beads (Figs. 3B, 4). Together with their lack of many hematopoietic markers, we consider that 5F9A cells are not of macrophage/granulocyte lineage. In RT-PCR analyses, 5F9A cells expressed many but not all osteoclast markers. They were positive for integrin aV, integrin b3, cathepsin K, c-fms, and TRAP, but negative for RANK and CTR (Fig. 9). Although osteoclasts were reported to be positive for CD45 and negative for CD14 (Athanasou and Quinn, 1990), 5F9A cells showed the opposite pattern of results (Fig. 3). Although these results suggest that 5F9A cells might be related to osteoclasts, we could not deﬁnitively deﬁne 5F9A cells as osteoclasts because of some phenotypic differences between 5F9A cells and osteoclasts. When taken into consideration that 5F9A cells had oligopotency (Figs. 5, 6), it is possible that they may instead be stem cells expressing some osteoclast markers. 5F9A Cells Are Oligopotent 5F9A cells could differentiate into adipocytes and osteoblasts in the presence of adipogenic and osteogenic differentiation cocktails, as determined by Sudan III staining and electron microscopy for adipocytes, and von Kossa staining and RT-PCR analysis for osteoblasts (Figs. 5, 6). Although osteonectin and PTHR (parathy- Fig. 8. Three-dimensional analysis of multinuclei in the cell using confocal microscopy. A,B: SYBR-I stained nuclei in one TPA-treated cell were observed using confocal microscopy (0.9 mm in thickness and 0.45 mm in focus step along the z-axis from the upper left to the lower right). Twenty optical sections were analyzed in each cell, and more than one distinct nucleus could be observed in one cell. Scale bar 5 10 mm. MULTINUCLEATION OF MESENCHYMAL STEM CELL Figure 8. 1265 1266 YOSHIDA ET AL. We detected N-cadherin on the surface of 5F9A cells (Fig. 3), which was previously reported to be associated with the osteoblastic niche for hematopoietic stem cells (Zhang et al., 2003). Although at present, we do not know if these molecules affect the function of 5F9A cells, a more intensive study of the junctional structures of 5F9A cells would reveal their important roles in the interaction of the cells during events such as cell–cell fusion. Multinucleation by TPA Fig. 9. Analyses of the expression of differentiation markers by TPA-induced multinucleated cells. Total RNA of 5F9A cells with or without TPA treatment was analyzed by the reverse transcription-polymerase chain reaction. Total RNA of rat fetal limb was used as a positive control for MyoD, Myf-5, and myogenin (striated muscle markers). Total RNA of the rat liver was used as a positive control for albumin and a-fetoprotein (hepatocytes markers). Positive controls of GPIIb and GPV (megakaryocyte markers) and other groups (osteoclast markers) were also assessed. roid hormone receptor) were expressed not only in osteogenically differentiated (Runx-2 and bone sialoprotein positive), but also in undifferentiated cells (Fig. 6C), previous ﬁndings showed that osteonectin was expressed in bone marrow-derived (undifferentiated) MSCs (Silva et al., 2003), and MSCs in fetal circulating blood spontaneously became PTHR-positive cells (Naruse et al., 2004). Therefore, our ﬁndings do not necessarily deny the ability of 5F9A cells to undergo osteoblastic differentiation. Rather, our data suggest that 5F9A cells have oligopotency. In brief, 5F9A cells would give us a valuable tool for studying rat MSCs because of their species importance. Cell Communication Between Mesenchymal Stem Cells Although adherence junctions, tight junctions, or desmosomes, that could be observed mainly in epithelial cells have not been reported in MSCs, 5F9A cells were shown to form junction-like structures by electron microscopy (Fig. 2). One type of junction (an interdigitation-like structure) had a bundle of microﬁlaments in the interdigitations along their long axis. Actin cables in the cells were connected to adherence junction-like structures. However, we could not detect E-cadherin or b-catenin in 5F9A cells either by immunoﬂuorescence techniques or RT-PCR analysis (Fig. 3). This ﬁnding may suggest that typical epithelial-type adherence junctions are not constructed between these cells. We could not detect desmogrein either, indicating that few, if any, desmosomes were formed (data not shown). The tumor promoter TPA could cause 5F9A cells to become cells with more than one nucleus (multinuclear; TPA-response; Figs. 7, 8) and it induced 1020% of the total cell population into multinucleated cells, whereas nontreated control cells contained less than 3% multinucleated cells. In general, it has been thought that genome ampliﬁcation in a cell is carried out by one of the mechanisms of syncytium, plasmodium, and endomitosis. All these processes could be induced by TPA stimulation, possibly depending on the cell types (Hassan et al., 1989; David et al., 1990; Murate et al., 1991; Menaya and Clemens, 1991; Bermejo et al., 2002). This is the ﬁrst report indicating that MSCs could be multinucleated by TPA stimulation. Further studies should be done to clarify the relationship between stem cells and multinucleation. ACKNOWLEDGMENTS We thank Dr. Y. Ikeda, Dr. K. Ohbo, and Dr. H. Tanaka for helpful discussions and encouragement during the course of this work. LITERATURE CITED Anderson JM. 2000. Multinucleated giant cells. Curr Opin Hematol 7:40–47. Athanasou NA, Quinn J. 1990. Immunophenotypic differences between osteoclasts and macrophage polykaryons: immunohistological distinction and implications for osteoclast ontogeny and function. J Clin Pathol 43:997–1003. Barry FP, Boyton RE, Haynesworth S, Murphy M, Zaia J. 1999. The monoclonal antibody SH-2 raised against human mesenchymal stem cells recognizes an epitope on Endoglin (CD105). Biochem Biophys Res Commun 265:134–139. Barry F, Boyton R, Murphy M, Zaia J. 2001. The SH-3 and SH-4 antibodies recognize distinct epitopes on CD73 from human mesenchymal stem cells. Biochem Biophys Res Commun 289:519–524. Beeres SLMA, Atsma DE, van der Laarse A, Pijnappels DA, van Tuyn J, Fibbe WE, de Vries AAF, Ypey DL, van der Wal EE, Schalij MJ. 2005. Human adult bone marrow mesenchymal stem cells repair experimental conduction block in rat cardiomyocyte cultures. J Am Coll Cardiol 46:1943–1952. Berkes CA, Tapscott SJ. 2005. MyoD and transcriptional control of myogenesis. Semin Cell Dev Biol 16:585–595. Bermejo R, Vilaboa N, Cales C. 2002. Regulation of CDC6, Geminin, and CDT1 in human cells that undergo polyploidization. Mol Biol Cell 13:3989–4000. Castro-Malaspina H, Gay RE, Resnick G, Kapoor N, Meyers P, Chiarieri D, McKenzie S, Broxmeyer HE, Moore MAS. 1980. Characterization of human bone marrow ﬁbroblast colony-forming cells (CFU-F) and their progeny. Blood 56:289–301. Colter DC, Class R, DiGirolamo CM, Prockop DJ. 2000. Rapid expansion of recycling stem cells in cultures of plastic-adherent cells from bone marrow. Proc Natl Acad Sci U S A 97:3213–3218. David JD, Faser CR, Perrot GP. 1990. Role of protein kinase C in chick embryo skeletal myoblast fusion. Dev Biol 139:89–99. MULTINUCLEATION OF MESENCHYMAL STEM CELL Deans RJ, Moseley AB. 2000. Mesenchymal stem cells: biology and potential clinical uses. Exp. Hematol. 28:875–884. Dennis JE, Charbord P. 2002. Origin and differentiation of human and murine stroma. Stem Cells 20:205–214. Dezawa M, Takahashi I, Esaki M, Takano M, Sawada H. 2001. Sciatic nerve regeneration in rats induced by transplantation of in vitro differentiated bone-marrow stromal cells. Eur J Neurosci 14:1771–1776. Dezawa M, Kanno H, Hoshino M, Cho H, Matsumoto N, Itokazu Y, Tajima N, Yamada H, Sawada H, Ishikawa H, Mimura T, Kitada M, Suzuki Y, Ide C. 2004. Speciﬁc induction of neuronal cells from bone marrow stromal cells and application for autologous transplantation. J Clin Invest 113:1701–1710. Friedman MS, Long MW, Hankenson KD. 2006. Osteogenic differentiation of human mesenchymal stem cells is regulated by bone morphogenetic protein-6. J Cell Biochem 98:538–554. Fujisaki K, Tanabe N, Suzuki N, Kawato T, Takeichi O, Tsuzukibashi O, Makimura M, Ito K, Maeno M. 2007. Receptor activator of NF-jB ligand induces the expression of carbonic anhydrase II, cathepsin K and matrix metalloproteinase-9 in osteoclast precursor RAW264.7 cells. Life Sci 80:1311–1318. Gronthos S, Zannettino ACW, Hay SJ, Shi S, Graves SE, Kortesidis A, Simmons PJ. 2003. Molecular and cellular characterization of highly puriﬁed stromal stem cells derived from human bone marrow. J Cell Sci 116:1827–1835. Hassan NF, Kamani N, Meszaros MM, Douglas SD. 1989. Induction of multinucleated giant cell formation from human blood-derived monocytes by phorbol myristate acetate in in vitro culture. J Immunol 143:2179–2184. Hay DC, Zhao D, Ross A, Mandalam R, Lebkowski J, Cui W. 2007. Direct differentiation of human embryonic stem cells to hepatocyte-like cells exhibiting functional activities. Cloning Stem Cells 9:51–62. Jiang Y, Jahagirdar BN, Reinhardt RL, Schwartz RE, Keene CD, Ortiz-Gonzalez XR, Reyes M, Lenvik T, Lund T, Blackstad M, Du J, Aldrich S, Lisberg A, Low WC, Largaespada DA, Verfaillie CM. 2002. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418:41–49. Laino G, Carinci F, Graziano A, d’Aquino R, Lanza V, De Rosa A, Gombos F, Caruso F, Guida L, Rullo R, Menditti D, Papaccio G. 2006. In vitro bone production using stem cells derived from human dental pulp. J Graniofac Surg 17:511–515. Lepage A, Leboeuf M, Cazenave J, Salle C, Lanza F, Uzan G. 2000. The aIIbb3 integrin and GPIb-V-IX complex identify distinct stages in the maturation of CD341 cord blood cells to megakaryocytes. Blood 96:4169–4177. Menaya J, Clemens MJ. 1991. Phorbol ester-induced inhibition of proliferation of Daudi Burkitt’s Lymphoma cells by impairment of cytokinesis. Exp Cell Res 194:260–266. Miller SI, Ernst RK, Bader MW. 2005. LPS, TLR4 and infectious disease diversity. Nat Rev Microbiol 3:36–46. Murate T, Hotta T, Tsushita K, Suzuki M, Yoshida T, Saga S, Saito H, Yoshida S. 1991. Aphidicolin, an inhibitor of DNA replication, 1267 blocks the TPA-induced differentiation of a human megakaryoblastic cell line, MEG-O1. Blood 78:3168–3177. Myers DE, Collier FM, Minkin C, Wang H, Holloway WR, Malakellis M, Nicholson GC. 1999. Expression of functional RANK on mature rat and human osteoclasts. FEBS Lett 463:295– 300. Naruse K, Urabe K, Mukaida T, Ueno T, Migishima F, Oikawa A, Mikumi-Takagaki Y, Itoman M. 2004. Spontaneous differentiation of mesenchymal stem cells obtained from fetal rat circulation. Bone 35:850–858. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, Moorman MA, Simonetti DW, Craig S. Marshak DR. 1999. Multilineage potential of adult human mesenchymal stem cells. Science 284:143–147. Ptaff M, Jurdic P. 2001. Podosomes in osteoclast-like cells: structural analysis and cooperative roles of paxillin, proline-rich tyrosine kinase 2 (Pyk2) and integrin aVb3. J Cell Sci 114:2775– 2786. Reyes M, Lund T, Lenvik T, Aguiar D, Koodie L, Verfaillie CM. 2001. Puriﬁcation and ex vivo expansion of postnatal human marrow mesodermal progenitor cells. Blood 98:2615–2625. Silva WA Jr, Covas DT, Panepucci RA, Proto-Siqueira R, Siuﬁ JLC, Zanette DL, Santos ARD, Zago MA. 2003. The proﬁle of gene expression of human marrow mesenchymal stem cells. Stem Cells 21:661–669. Smith A. 2006. A glossary for stem-cell biology. Nature 441:1060. Taylor MV. 2003. Muscle differentiation: signaling cell fusion. Curr Biol 13:R964–R966. Tocci AL, Forte L. 2003. Mesenchymal stem cell: use and perspectives. Hematol J 4:92–96. Valiunas V, Doronin S, Valiuniene L, Potapova I, Zuckerman J, Walcott B, Robinson RB, Rosen MR, Brink PR, Cohen IS. 2004. Human mesenchymal stem cells make cardiac connexins and form functional gap junction. J Physiol 555:617–626. Vitrat N, Cohen-Solal K, Pique C, Pierre Le Couedic J, Norol F, Larsen AK, Katz A, Vainchenker W, Debili N. 1998. Endomitosis of human megakaryocytes are due to abortive mitosis. Blood 91:3711–3723. Watanabe T, Ono Y, Taniyama Y, Hazama K, Igarashi K, Ogita K, Kikkawa U, Nishizuka Y. 1992. Cell division arrest induced by phorbol ester in CHO cells overexpressing protein kinase C-d subspecies. Proc Natl Acad Sci U S A 89:10159–10163. Yamaguchi K, Ogita K, Nakamura S, Nishizuka Y. 1995. The protein kinase C isoforms leading to MAP-kinase activation in CHO cells. Biochem Biophys Res Commun 210:639–647. Zhang J, Niu C, Huang H, He X, Tong W, Ross J, Haug J, Johnson T, Feng JQ, Harris S, Wiedemann LM, Mishina Y, Li L. 2003. Identiﬁcation of the haematopoietic stem cell niche and control of the niche size. Nature 425:836–841. Zuk PA, Zhu M, Mizuno H, Huang J, Futrell JW, Katz AJ, Benhaim P, Lorenz HP, Hedrick MH. 2001. Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng 7:211–228.