Cardiac Stem Cells in Brown Adipose Tissue Express CD133 and Induce Bone Marrow

江边孤钓

作者：Yoshihiro Yamadaa,b,c, Shin-ichiro Yokoyamad, Xiang-Di Wangb, Noboru Fukudad, Nobuyuki Takakuraa,b,c作者单位：aDepartment of Signal Transduction, Research Institute for Microbial Diseases, Osaka University, Osaka, Japan;bDepartment of Stem Cell Biology, Cancer Research Institute of Kanazawa University, Kanazawa, Japan;cPREST, Japan Science and Technology Agency, Saitama, Japan;dSecond Department of Internal
: Z1 x6 _* d( N1 N                  # L2 x0 }8 ?: n. ^! }) p
                  # @/ X3 }; m; p7 K* F
          4 I- s. C1 f7 p2 S
                         0 h. I7 x3 H3 O: }
            6 O8 C% J7 s0 y) x+ I) |" ]
            
0 X% H: P# L; u            
) c% e; L$ D9 b0 I2 z# j            
6 ]8 W8 ?1 j$ H+ `                     
6 ?+ y; P! S! ?' {* w        " C- p8 u, _5 X
        
4 @2 T2 Y: S4 J, t' k$ w        
/ N5 d5 Z4 N1 y3 R( D          【摘要】& D: E6 G$ M8 |- G  q1 t: s) e
      Correspondence: Nobuyuki Takakura, M.D., Ph.D., Department of Signal Transduction, Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita-shi, Osaka 565-0871, Japan. Telephone: 81-6-6879-8316; Fax: 81-6-6879-8314; e-mail: ntakaku@biken.osaka-u.ac.jp; Yoshihiro Yamada, M.D., Ph.D., Department of Signal Transduction, Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita-shi, Osaka 565-0871, Japan. Telephone: 81-6-6879-8316; Fax: 81-6-6879-8314; e-mail: yamaday@biken.osaka-u.ac.jp1 ~( \3 m1 N1 `
9 j0 a9 @$ {1 E7 U; j5 b
Recently, there has been noteworthy progress in the field of cardiac regeneration therapy. We previously reported that brown adipose tissue (BAT) contained cardiac progenitor cells that were relevant to the regeneration of damaged myocardium. In this study, we found that CD133-positive, but not c-Kit- or Sca-1-positive, cells in BAT differentiated into cardiomyocytes (CMs) with a high frequency. Moreover, we found that CD133  brown adipose tissue-derived cells (BATDCs) effectively induced bone marrow cells (BMCs) into CMs. BMCs are considered to have the greatest potential as a source of CMs, and two sorts of stem cell populations, the MSCs and hematopoietic stem cells (HSCs), have been reported to differentiate into CMs; however, it has not been determined which population is a better source of CMs. Here we show that CD133-positive BATDCs induce BMCs into CMs, not through cell fusion but through bivalent cation-mediated cell-to-cell contact when cocultured. Moreover, BMCs induced by BATDCs are able to act as CM repletion in an in vivo infarction model. Finally, we found that CD45¨CCD31¨C CD105  nonhematopoietic cells, when cocultured with BATDCs, generated more than 20 times the number of CMs compared with lin¨Cc-Kit  HSCs. Taken together, these data suggest that CD133-positive BATDCs are a useful tool as CM inducers, as well as a source of CMs, and that the nonhematopoietic fraction in bone marrow is also a major source of CMs.5 ?, M& W) {7 F! M% u+ ^8 R5 z6 U

( O+ M' l( _0 Y- t8 dDisclosure of potential conflicts of interest is found at the end of this article. 4 M% K/ V4 _0 e, h* I) J& ]
          【关键词】 AC Adult bone marrow Myogenesis Adipogenesis9 m' C' `- |1 S8 h& y. W
                  INTRODUCTION
8 u$ F! G  ?. g3 D' N' F
8 i  \: T  M# P3 dCell transplantation and gene transfer are two of the foremost therapies with potential for regenerating damaged cardiomyocytes (CMs) and enabling revascularization. Embryonic stem cells, skeletal myoblasts, and c-kit- or Sca-1-positive cardiac stem cells (CSCs) in heart tissue have all been reported as candidates for the replacement of CMs; however, each is associated with problems, including those involving allergenic, etiologic, and arrhythmic issues.
0 P- U0 v. Z6 K# @* |
& v( C5 e7 }8 V1 H$ |Bone marrow cells (BMCs) are a well-known source of various kinds of stem cells. BMCs can contribute to the regeneration of various tissues, and their clinical application is easier than that of other candidates. Recent studies have suggested that CMs might also originate from BMCs, which are composed of at least two cell populations. Among these are hematopoietic cells (HCs) . Taken together, the results from these various studies suggest that it remains to be resolved whether plasticity of HSCs truly occurs and which population will prove to be the best source for the repair of damaged cardiac tissue. Moreover, specific molecular cues for the differentiation of CMs from BMCs have not been identified. This lack of knowledge means that, at present, it is difficult to effectively induce immature cells into CMs.
9 e2 Z3 _) J( ~6 ?/ ?' n# n- S& [( _, R( d, S
In this study, we show that CD133 is a CSC marker in brown adipose tissue-derived cells (BATDCs), which had been identified previously as a source of CSCs . These results indicated that CD133 might be an antigen common to several stem cells, including CSCs, in brown adipose tissue (BAT). Second, we have shown that BMCs could differentiate with high efficiency into CMs by coculturing with BATDCs. In this induction system, we have demonstrated that bivalent cation-mediated cell-to-cell contact was critical for the differentiation of BMCs into CMs, and we have also investigated cadherin-mediated cell contact. Finally, we have demonstrated that in the bone marrow (BM) population, CD45¨CCD31¨CCD105  non-HCs could effectively differentiate into CMs, in contrast to the lin¨Cc-Kit  HSCs.
5 U8 D4 X: h3 u, [% a1 s$ G3 \
MATERIALS AND METHODS; ], B. X% a( O, s
0 g1 m3 S: D4 q  R% x0 W8 D1 a3 e
Cell Preparation and Flow Cytometry! V4 J& {* l7 q4 k& \& n
8 a. U& Q2 h, {+ |
BAT was dissected from the interscapular region of postnatal day 1 (P1) to P7 neonates of C57BL/6 mice. BAT was dissociated by DispaseII (Roche Diagnostics, Mannheim, Germany, http://www.roche-applied-science.com), drawn through a 23-gauge needle, and prepared as a single-cell suspension, as previously reported . The monoclonal antibodies (mAbs) used in immunofluorescence staining were anti-CD45, -ter119, -CD31, -CD29, -c-kit, -Sca1, -CD105, and -CD133 mAbs (BD Pharmingen, San Diego, http://www.bdbiosciences.com/pharmingen). All mAbs were purified and conjugated with fluorescein isothiocyanate, phycoerythrin (PE), biotin, or allophycocyanin (APC). Biotinylated antibodies were visualized with PE-conjugated streptavidin (BD Pharmingen) or APC-conjugated streptavidin (BD Pharmingen). Cells were incubated for 5 minutes on ice with CD16/32 (FcIII/II Receptor) (1:100) (Fcblock; BD Pharmingen) prior to staining with primary antibody. Cells were incubated in 5% fetal calf serum (FCS)/phosphate-buffered saline (PBS) (washing buffer) with primary antibody for 30 minutes on ice and washed twice with washing buffer. Secondary antibody was added, and the cells were incubated for 30 minutes on ice. After incubation, cells were washed twice with, and suspended in, the washing buffer for fluorescence-activated cell sorting (FACS) analysis. The stained cells were analyzed and sorted with an EPICS flow cytometer (Beckman Coulter, San Jose, CA, http://www.beckmancoulter.com). The sorted cells were added to 24-well dishes (Nunc, Roskilde, Denmark, http://www.nuncbrand.com), precoated with 0.1% gelatin (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com), and cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FCS and 10¨C5 M 2-mercaptoethanol at 37¡ãC in a 5% CO2 incubator.
. {4 y3 @0 z# K! |- r/ ^3 _3 k
6 w! {, F; `0 w' }Immunohistochemistry- z6 H3 e9 A# W( G& {" w

& x' H0 n' V$ o$ sImmunohistochemical analyses of tissue sections and culture dishes and the tissue fixation procedure were performed as previously described , -GATA-4, -MEF2C, and -pan-cadherin (Sigma-Aldrich) antibodies for CM culture dishes were used in this assay. In brief, anti-SA was developed with Alexa Fluor 488-, 546-, or 633-conjugated goat anti-mouse IgM (Molecular Probes Inc., Eugene, OR, http://probes.invitrogen.com); anti-pan-cadherin was developed with 546-conjugated goat anti-mouse IgG (Molecular Probes); and anti-cardiac troponin T, cardiac troponin I, -MEF2C, -GATA-4, -connexin43, and -ANF antibodies were developed with Alexa Fluor 488-, 546-, or 633-conjugated goat anti-rabbit IgG (Molecular Probes). Nuclear staining was performed with 4,6-diamidino-2-phenylindole or TOPRO3 (Molecular Probes). Finally, the sections and dishes were observed using an Olympus IX-70 microscope equipped with UPlanFI 4/0.13 and LCPlanFI 20/0.04 dry objective lenses (Olympus, Tokyo, http://www.olympus-global.com). Images were acquired with a CoolSnap digital camera (Roper Scientific, Tokyo, http://www.roperscientific.com). In all assays, an isotype-matched control Ig was used as a negative control, and it was confirmed that the positive signals were not derived from nonspecific background. Images were processed using Adobe Photoshop 7.0 software (Adobe Systems Inc., San Jose, CA, http://www.abode.com).3 \8 z) x  c6 f% V& V$ B
! m6 W: e' m7 y, j! u" J
Transmission Electron Microscopy7 k7 z/ B1 K! k- Y+ u; m

6 l( ]5 |8 n" M& ICells were washed in phosphate buffer and fixed with 2% glutaraldehyde and 1% paraformaldehyde in PBS. Samples were postfixed with 1% osmium in PBS, rinsed, dehydrated, and embedded in araldite (DAKO, Glostrup, Denmark, http://www.dako.com). Then, samples were cut with a diamond knife and examined under a Jeol 100CX (Tokyo, http://www.jeol.co.jp) electron microscope.  ]0 c3 x- u( g6 R5 h' d

" Z- U. w" V% I# z, QCell Coculture; x! D: q* a  x, \: P/ {

( A% X$ s7 C0 r4 g: p; M- H2 |BATDCs and BMCs were prepared as described above. In situations where CD133  BATDCs and BMCs were cocultured in contact conditions, 2 x 104 CD133  BATDCs were plated per well of a 24-well plate and cultured for 10 days before being fixed with 0.5% paraformaldehyde for 15 minutes at room temperature. Fixed cells were washed extensively, first with PBS and then with DMEM/10% FCS. After washing, 1 x 105 BM mononuclear cells (BMMNCs), 1 x 104 lin¨Cc-Kit  HSCs, or 1 x 104 CD45¨Cter119¨CCD31¨CCD105  MSCs from GFP mice were cultured with fixed BATDCs, as described above, for 10 days and then stained with anti-SA (Sigma-Aldrich), -GATA-4 (Santa Cruz Biotechnology), -MEF-2C (Cell Signaling Technology), and -GFP (MBL International Corp., Nagoya, Japan, http://www.mblintl.com) antibodies. Under separate conditions, CD133  BATDCs from ROSA26 mice were cocultured on 0.4-µm cell culture inserts (Becton, Dickinson and Company, San Jose, CA, http://www.bd.com) with BMMNCs from green mice in 0.1% gelatin-coated dishes (Becton Dickinson) for 14 days and then stained with anti-SA (Sigma-Aldrich) and anti-GATA-4 (Santa Cruz Biotechnology)./ V$ a' c) r, ]7 d
% R7 F. Q/ \& y/ L/ @- S# ]) l# I
To analyze the diverse signaling between BATDCs and BM-derived cells, we used an anti-E-cadherin antibody (ECCD-1; Calbiochem, La Jolla, CA, http://www.emdbiosciences.com), E-cadherin-Fc, N-cadherin-Fc, and R-cadherin-Fc (all purchased from R&D Systems Inc., Minneapolis, http://www.rndsystems.com).
) l9 e7 y- `% K! G5 U7 ~$ `5 m+ _7 L1 F5 O  h. J" B
Reverse Transcription-Polymerase Chain Reaction Analysis
& y% v- [7 R, E2 }) V
7 g1 B+ e4 C$ FThe RNeasy Mini kit (Qiagen, Hilden, Germany, http://www1.qiagen.com) was used for isolation of total RNA from BAT and BMCs. Total RNA was reverse transcribed using the reverse transcription-polymerase chain reaction (RT-PCR) kit (Clontech, Palo Alto, CA, http://www.clontech.com). The cDNA was amplified using Advantage Polymerase Mix (Clontech) in a GeneAmp PCR system, model 9700 (PerkinElmer Life and Analytical Sciences, Norwalk, CT, http://www.perkinelmer.com), by 40¨C50 cycles. The sequences of the gene-specific primers for RT-PCR were as follows: 5'--myosin heavy chain (MHC), TGTCTGCTCTCCACCGGGAAAATCT; 3'--MHC, CATGGCCAATTCTTGACTCCCATGA; 5'-¦Â-MHC, AACCCACCCAAGTTCGACAAG ATCG; 3'-¦Â-MHC, CCAACTTTCCTGTTGCCCCAAAAATG; 5'--skeletal actin, GGAGATTGTGCGCGACATCAAAGAG; 3'--skeletal actin, CTGGTTCCTCCAATGGGATATCTTC; 5'--cardiac actin, TGTGTTACGTCGCCCTGGATTTTGA; 3'--cardiac actin, TTGCTGATCCACATTTGCTGGAAGG; 5'-myosin light chain (MLC)-2a, AGCAGGCACAACGTGGCTCTTCTAA; 3'-MLC-2a, CCTGGGTCATGAGAAGCTGCTTGAA; 5'-MLC-2v, ATGGCACCTTTGTTTGCCAAGAAGC; 3'-MLC-2v, CCCTCGGGATCAAACACCTTAATG; 5'-GATA4, GAGTGAGTCAATTGTGGGGCCATGT; 3'-GATA4, TGCTGCTAGTGGCA TTGCTGGAGTT; 5'-glyceraldehyde-3-phosphate dehydrogenase (G3PDH), TGAAGGTCGGTGTGAACGGATTTGGC; 3'-G3PDH, CATGTAGGCCATGAG GTCCACCAC; 5'-lacZ, GCGTTACCCAACTTAATCG; 3'-lacZ, TGTGAGCGAGTAACAACC; 5'-GFP, TACGGCAAGCTGACCCTGAA; 3'-GFP, TGTGATCGCGCTTCTCGTTG. Each cycle consisted of denaturation at 94¡ãC for 30 seconds and annealing/extension at 70¡ãC for 4 minutes except lacZ and GFP. For lacZ and GFP, each cycle consisted of denaturation at 94¡ãC for 30 second, annealing at 60¡ãC for 1 minute, and extension 72¡ãC for 1 minute.
2 a; y9 E3 ?4 [2 g9 B
8 Y% }- e0 K1 bMyocardial Infarction and Cell Implantation and Echocardiography3 w, x9 A( y# b0 l: l

- y* ?1 l8 ^5 DBMCs derived from GFP transgenic Sprague-Dawley rats . After verifying the MIs, 10 rats were injected with 2 x 105 cells each, in five opposite regions bordering the infarct, and then sacrificed after 30 days. At each time interval, sham-operated rats were injected with saline as controls. Under ketamine (Dainippon Pharmaceutical, Osaka, Japan, http://www.ds-pharma.co.jp) anesthesia, echocardiography was performed at 29 days. From M mode tracings, left ventriculus end-diastolic and systolic diameter and wall thickness were obtained, and then the percentage of fractional shortening was calculated. Echocardiographic acquisition and analysis were performed by an echocardiographer blinded to treatment group. Results represent the mean of five separate experiments. Mortality was lower, but not significantly different, in the treated rats, averaging 35% in all groups. Protocols were approved by the institutional review board.
( }& o  U( L% M4 W& e4 I- m; i$ Z4 G7 l# ]$ k$ R2 \
Fluorescence In Situ Hybridization Staining
- Q: `7 E, o; x" e: X: q: u* r# I) M' t, Y0 W; x  e
After staining of CMs with anti-SA antibody as described above, we performed fluorescence in situ hybridization (FISH) to determine the presence of rat X and Y chromosomes (Cambio, Cambridge, U.K., http://www.cambio.co.uk) according to the manufacturer's instructions. A digestion with proteinase K (DAKO) (20 µg/ml at 37¡ãC, 5 minutes) was added at the beginning of the FISH protocol. Nuclear staining was performed with TOPRO3 (Molecular Probes). Pictures were taken by confocal microscope (LSM510; Carl Zeiss MicroImaging, Inc., Göttingen, Germany, http://www.zeiss.com).! b; z, N6 `4 f% [
& C% Z7 H4 T3 H4 e) `/ U* \) f9 s. Q
RESULTS
+ ^$ U/ f6 d. W8 N. E  E! ^3 U$ _! T- W+ b$ u& H& l
Surface Phenotype of CM Progenitors in BAT
& b/ w: W/ b5 X8 D5 S  ^1 |  @: u" X$ U, ~  \* X8 x3 v, H' n9 G
A putative stem cell population has recently been identified in adipose tissue . Therefore, using flow cytometric analysis, we analyzed several stem cell markers and multiple CD markers on cells from BATDCs, which can differentiate into CMs. First, we checked for ter119 (erythrocyte marker), CD45 (LCA) (panleukocyte marker), and CD31 (endothelial marker). Among CD45¨Cter119¨CCD31¨C cells (nonhematopoietic, nonendothelial cells), c-Kit, Sca-1, and CD133 expression was observed on 1.4%, 16.6%, and 3.5% of total BATDCs, respectively (Fig. 1A). To clarify which populations have a high potential to produce CMs, we cultured a CD45¨Cter119¨CCD31¨C subpopulation fractionated by c-Kit, Sca-1, or CD133 expression. Among them, CD133-positive BATDCs differentiated into CMs with a higher incidence compared with cells in other fractions and CD29-positive (CD45¨Cter119¨CCD31¨C) cells, which we had previously reported as a CM-enriched population (Fig. 1C). Indeed, approximately 50% of adherent cells from cultured CD133-positive cells among CD45¨Cter119¨CCD31¨C cells differentiated into SA GATA-4  (Fig. 1B, a and b, 1C, a and b), cardiac-troponin T GATA-4  (Fig. 1B, c and d), and cardiac-troponin I MEF2C  (Fig. 1B, e and f). Furthermore, electron microscopic analysis indicated that these cells had cellular structures typical of CMs, such as an organized sarcomere with typical cross-strain, developed Z-bands, long mitochondria in the cytoplasm, and centrally positioned nuclei (Fig. 1B, g). To provide additional evidence that CD133  BATDCs have the phenotype of CMs, we performed pharmacological studies (supplemental online data). The calcium antagonist verapamil slowed the beating rate, and the ¦Â-agonist isoproterenol induced a dose-dependent increase of the spontaneous contraction rate; however, the ¦Â-adrenergic antagonist propranolol reversed the isoproterenol-induced acceleration. These results also indicated that CD133  BATDCs have the functional character of CMs.3 g' Y- g0 \2 ~: R/ O

( l3 h* [- t4 D  XFigure 1. Phenotype of cardiomyocyte (CM) stem cells in brown adipose tissue (BAT). (A): Cells derived from BAT were stained with a mixture of anti-CD45, -ter119, and -CD31 antibodies (x-axis) and anti-c-Kit, -Sca-1 or -CD133 antibody (y-axis). (B): Immunocytochemical analysis and transmission electromicrograph of CMs derived from CD133  (CD45¨Cter119¨CCD31¨C) BATDCs. CD133-positive cells generated SA  (red)/GATA-4  (green) cells (a), cardiac troponin T  (red)/GATA-4  (green) cells (c), or cardiac troponin I  (red)/MEF2C  (green) cells (e). Nuclear staining by TOPRO3 in a, c, and e is shown in b, d, and f, respectively. Notably, well-organized sarcomeres, Z-band, and a large number of mitochondria were observed in CD133  BATDCs. Scale bar = 10 µm (f) and 1 µm (g). (C): a, Existence of SA-positive (green) and GATA4-positive (red) CMs in the culture of CD133  BATDCs. b, Nuclear staining with 4,6-diamidino-2-phenylindole in the same field as shown in panel a. Note that this low-power field view clearly showed that approximately 50% of adherent cells are CMs. Scale bar = 20 µm. c, Quantitative evaluation of differentiation potential for CMs from each fraction of BATDCs as indicated. The same number of cells (1 x 104) of each fraction was cultured. Note that c-Kit  (CD45¨Cter119¨CCD31¨C) cells could not differentiate into CMs, and CD133  (CD45¨Cter119¨CCD31¨C) cells were the most effective at differentiating into CMs." S  A3 x: g# n$ d% S

' K5 k' Q2 R: t' p4 W! Q6 d/ n  d7 kBATDCs Effectively Induce CM Production from BMCs2 _& e# @. P! ?+ O6 @2 C
9 ^3 g6 i- g' ~  O$ ]
A previous in vitro study showed that although BMCs were a source of CMs, they could not spontaneously differentiate into CMs; however, differentiation could be induced with 5-azacytidine, a DNA-hypomethylating agent  were cocultured in direct contact with CD133  BATDCs, as shown in Figure 2B. By coculturing BMMNCs with CD133-positive BATDCs, nuclear-located GATA-4-positive (Fig. 2A, a¨Cc), MEF2C-positive (Fig. 2A, d¨Cf), and SA-positive (Fig. 2A, g¨Ci) contracting cells were produced among the GFP-positive BMMNCs. However, the SA-positive and GATA-4-positive cells were not observed when BMMNCs were cultured alone under the same conditions (Fig. 2C).
% Z0 N; y( d, ^  L6 k/ ]# ]: e" n
Figure 2. BMMNCs can differentiate into cardiomyocytes (CMs) upon coculturing with BATDCs. (A): Immunocytochemical analysis of BMMNCs from green mice cocultured for 14 days with CD133-positive BATDCs from wild-type mice. a¨Cc, GFP (green) (a), GATA-4 (red) (b), and nuclear staining with TOPRO3 (blue) merged with that shown in b (c). d¨Cf, GFP (green) (d), MEF2C (red) (e), and nuclear staining with TOPRO3 (blue) merged with that shown in e (f). g¨CI, GFP (green) (g), SA (red) (h), and GATA-4 (blue) merged with that shown in h (i). Scale bar = 5 µm (i). (B): Schematic presentation of coculturing BMMNCs from green mice with CD133-positive BATDCs from wild-type mice. (C): Quantitative evaluation of differentiated SA- and GATA4-positive CMs among adhering GFP-positive BM-derived cells. Data for BMMNC cocultured with CD133  BATDCs and BMMNCs cultured alone are displayed. Results represent the mean of five independent experiments. Abbreviations: BATDC, brown adipose tissue-derived cell; BMC, bone marrow cell; BMMNC, bone marrow mononuclear cell; GFP, green fluorescent protein.
) _) I/ L, ~# y7 p9 N
$ p* {% {0 d7 ~0 D" VBMMNCs with BATDCs Differentiated into CMs Without Fusion Mechanism
5 h# Q: ^) {1 Y  x+ I/ p
6 s. v4 l9 s4 n9 tRecently, it has been suggested that cell fusion is the main mechanism that contributes to the development or maintenance of cardiac muscle and neuron regeneration . To exclude the possibility of cell fusion in our experiment, we cultured BMMNCs from green mice with BATDCs from ROSA26 mice, which express LacZ ubiquitously in their tissues, and separated the cells from these two different origins with a 0.4-µm-pore membrane (Fig. 3A, a). In this separating coculture system, SA GATA-4  CMs were also produced from BMMNCs when cocultured with BATDCs (Fig. 3B). Moreover, with PCR analysis, GFP-positive signals were detected only in the BMMNC layer, and lacZ-positive signals were detected only in the BATDCs layer from the cells before and after culturing (Fig. 3A, b). Therefore, we concluded that the development of CMs from BMMNCs was not dependent on cell fusion with BATDCs.
) q% @7 g9 B5 D" i
' f! }4 S4 d# ~  [! ^- }" FFigure 3. Bone marrow mononuclear cells (BMMNCs) differentiated into cardiomyocytes (CMs) in separated coculture system. (A): a, Schematic representation of the system for the separate coculture of BMMNCs with CD133  BATDCs. b, Reverse transcription-polymerase chain reaction analysis in separate culture conditions. BMMNCs were derived from green mice (lower chamber), and BATDCs were derived from ROSA26 mice (upper chamber). GFP signal was detected only in the lower chamber, and lacZ signal was detected only in the upper chamber. Lane 1, BMMNCs before culture (lower chamber); lane 2, BMMNCs after 14 days of culture (lower chamber); lane 3, BATDCs before culture; lane 4, BATDCs after 14 days of culture. (B): Quantitative evaluation of the differentiation potential for CMs from BMMNCs in separate culture conditions. Upon coculturing with BATDCs (coculture). Results represent the mean of five independent experiments. Abbreviations: BATDC, brown adipose tissue-derived cell; BMC, bone marrow cell; GFP, green fluorescent protein.
' |+ N: D6 w0 b4 L4 i) F, p. a5 ]: y+ [9 ^/ g( |1 U2 V3 R
Cell Contact Mediated by Bivalent Cation Is Critical for the Differentiation of BMMNCs into CMs
! i+ H" @# F9 W! a! V; z$ p
& {. M& z) F. f3 j1 f( \4 Q, dWe found that BMMNCs could differentiate into CMs under conditions in which the BMMNCs were cultured in direct or indirect contact with BATDCs; however, differentiation of BMMNCs into CMs was promoted more effectively by direct cell-to-cell contact between BMMNCs and BATDCs than under separated culture conditions (Figs. 2B, 3B). Indeed, 520 ¡À 40 versus 105 ¡À 18 SA GATA4  CMs were generated from 1 x 105 BMMNCs under direct and indirect cocultures, respectively. To exclude the possibility of cell fusion in the contact coculture system, we used paraformaldehyde-fixed cultured CD133  BATDCs, which cannot fuse with other cells but have an intact cell surface. After coculturing with fixed BATDCs, some GFP  BMMNC-derived cells (Fig. 4A, a) expressed cardiac-specific antigens, such as SA (Fig. 4A, b and c), GATA-4, and cardiac troponinT (data not shown). Moreover, these cells also had the typical features of CMs, confirmed by electron microscopic analysis (Fig. 4A, f).0 c/ e; ?: Y+ s3 c: `5 n2 L

* b( ]* x8 P: U+ SFigure 4. Differentiation of bone marrow mononuclear cells (BMMNCs) into cardiomyocytes (CMs) is mainly induced by bivalent cation-mediated cell contact with brown adipose tissue-derived cells (BATDCs). (A): Immunocytochemical staining of CMs from BMMNCs cocultured with fixed BATDCs for 14 days. Anti-GFP (green) (a and c), anti-SA(red) (d and e), and anti-pan-cadherin (blue) (b, c, and e) Abs were used in this assay. Arrows indicate SA GFP  cells interacting with SA GFP¨C cells (arrowheads) through cadherin expression. f, Transmission electron micrographic analysis showed that BM-derived CMs had well-organized sarcomeres, Z-band, and a large number of mitochondria. Scale bar = 10 µm (e) and 1 µm (f). (B): Quantitative evaluation of differentiated CMs from BMMNCs in the presence or absence of EGTA, EDTA, or 100 µg/ml neutralizing Ab of Ecad, Ecad-Fc, Ncad-Fc, or Rcad-Fc when cocultured with fixed BATDCs. Adherent SA-positive and GATA-4-positive cells were counted in each culture condition. Results represent the mean of five independent experiments. Abbreviations: Ab, antibody; Ecad, E-cadherin; Ncad, N-cadherin; Rcad, R-cadherin." @8 a. r2 ~; ?6 S9 o8 W: P! C  w

* I$ k# z' v' U, q2 OCadherin-mediated calcium-dependent cell-to-cell contact is widely believed to be involved in the regulation of the diverse signaling process for cell differentiation . Therefore, we elucidated a potential role for calcium-dependent cell-to-cell contact in our coculture system. First, BMMNCs were incubated with the calcium chelator EDTA or EGTA on fixed BATDCs. This treatment abolished the adhesion of BMMNCs to fixed, differentiated BATDCs. By the suppression of cell adhesion of BMCs to differentiated BATDCs (fixed), differentiation of BMCs into CMs was almost completely suppressed (Fig. 4B). Next, we examined which, out of E-cadherin (ECad), NCad, or RCad, was affected in this mechanism. We found that CD133  BATDCs abundantly expressed Ecad, and BMCs expressed Ecad and Ncad, as confirmed by PCR analysis (data not shown). To evaluate the effect of cadherins in this culture system, we added neutralizing antibody (Ab) against Ecad (anti-Ecad Ab) or soluble cadherin, such as Ecad-Fc, Ncad-Fc, or Rcad-Fc. Among these materials, anti-Ecad Ab and Ecad-Fc effectively inhibited the differentiation of BMMNCs into CMs. Taken together, these data suggest thatEcad-mediated cell-to-cell contact is important for the differentiation of BMMNCs into CMs.
; m' [, ]8 a+ O, N$ w! `0 L1 _  r& P* @8 u6 C- i. Y$ L: S+ j
Educated BMMNCs Contributed to CM Regeneration* _6 ]/ G6 V# o5 C

- z: o4 \/ [: CThe data above clearly indicate that CD133-positive BATDCs effectively induced CM production from BMCs. Next, we evaluated the length of time required for BMMNCs to become committed to CM lineage when cultured with BATDCs. For this purpose, we attempted to coculture BMMNCs from green rats expressing GFP with BATDCs for 1 to 10 days and harvested GFP-positive cells at intervals from the culture. GFP-positive cells were then sorted and cultured alone for an additional 14 days (Fig. 5A, a). In this experiment, we found that 5 days was enough for commitment of BMMNCs into CM lineage. Sorted GFP-positive cells did not express SA and GATA-4 (data not shown); however, they started to display contractile activity after 7 days of culturing. After 14 days of culturing, we found that approximately 10% of GFP-positive adherent cells differentiated into SA-positive and MEF2C-positive cells (Fig. 5A, b¨Cd). To examine cardiac-specific genes and transcription factors in detail, we performed RT-PCR analysis and revealed that the expression of -MHC, ¦Â-MHC, -skeletal actin, -cardiac actin, MLC-2v, and GATA4 was detected (Fig. 5A, e). Interestingly, this phenotype was similar to that of CMs derived from BATDCs .7 Z! I- K1 C4 \& s+ [; ]9 V; H
8 \- z% N, ]7 L
Figure 5. Bone marrow mononuclear cells (BMMNCs) cocultured with brown adipose tissue-derived cells (BATDCs) can contribute to cardiac regeneration. (A): Phenotype analysis of cardiomyocytes (CMs) derived from BMMNCs. a, After BMMNCs from green rats were cocultured for 5 days with BATDCs from wild-type rats, FACS analysis showed that 80% of cells were GFP-positive. b¨Cd, Isolated GFP-positive fraction as indicated in panel a were cultured again in the absence of BATDCs for 14 days. Among adherent GFP-positive cells (green) (b), approximately 9.7% showed SA staining (red). d, MEF2C staining (blue) merged with that shown in b and c. e, Reverse transcription-polymerase chain reaction analysis of several CM-specific genes in CMs produced from BMMNCs. Lane 1, cells harvested from the conditions described in b¨Cd (educated BMCs ); lane 2, CMs from neonatal mice as a positive control; lane 3, total RNA used in lane 1 without reverse transcription. Scale bar = 5 µm. (B): Cells obtained as described in (A) were injected into infarcted heart, and the tissue distribution of various marker molecules was determined. a¨Cc, GFP (green) (a), SA (blue) (b), and connexin43 (red) merged with that shown in a and b (c). d¨Cf, GFP (green) (d), ANF (red) (e), and nuclear staining with TOPRO3 (blue) merged with that shown in d and e (f). Scale bar = 5 µm (f). (C): Fluorescence in situ hybridization staining in implanted site of CM. a, Section was stained with anti-SA (green) and TOPRO3 (blue). b, Serial section of panel a. Green indicates X chromosome, and red indicates Y chromosome. Nuclear staining was performed with TOPRO3 (blue). Cells expressing X and Y chromosome in the nuclei indicate that these cells were derived from e-BMCs and did not fuse with host CM expressing only X chromosome. Scale bar = 5 µm. Abbreviations: G3PDH, glyceraldehyde-3-phosphate dehydrogenase; GFP, green fluorescent protein.+ X/ m1 w0 ?: {/ G* U1 o
7 V: W' c, X. A0 y$ x2 R
Next, to determine whether BMMNCs exposed to BATDCs could effectively contribute to the regeneration of the heart, we injected the exposed BMMNCs into the hearts of rats after the induction of an acute MI. At first, we cocultured GFP-positive BMMNCs with CD133-positive BATDCs for 5 days, purified the GFP-positive cells by FACS, and injected the cells into the hearts of experimental MI rats at each of five sites at the border of the infarcted tissue. As a control, infarcted hearts were injected either with equal volumes and numbers of nonexposed (naive) BMMNCs or with saline. First, we revealed that donor-derived, exposed GFP-positive and SA- or ANF-positive cells were detected in abundance in the infarct border zone (Fig. 5B, a and d; 16.8% ¡À 2.1% of total cardiomyocytes in one field), but there were 15-fold fewer after injecting nonexposed GFP-positive cells (1.1% ¡À 0.4%). Hearts injected with exposed BMMNCs expressed SA-positive CMs, which also expressed connexin43 (Fig. 5B, a¨Cc), and ANF (Fig. 5B, d¨Cf). This indicated that transplanted BMMNCs formed gap junctions and intercalated disks and secreted CM-specific factor. Moreover, the assessment of cardiac function by echocardiography revealed that the hearts injected with exposed BMMNCs showed improved contractions of movement of the infarcted anterior walls and reduced left ventricular remodeling compared with the hearts injected with naive BMMNC or saline (Table 1). To exclude the possibilities of cell fusion in this model, we performed FISH staining with chromosome X- and Y-specific paint and revealed that educated BMMNCs (e-BMCs) implanted from male rats into female rats formed CM stained only with XY chromosome, not XXXY (Fig. 5C, a and b). This indicated that in vivo cardiac differentiation of BMMNCs was not induced by the fusion mechanism./ L: r" C. h; [0 `+ F% h' f
4 B6 E- e; B, l. X% H9 A. c, Q
Table 1. Effect of transplanted cells on myocardial performance
8 `0 |! c( m+ B( D3 {8 [& S+ ^" J9 H" s2 z
Nonhematopoietic Cells in the BM Are a Major Source of CM
1 w/ v, G6 a* X* t1 |' D+ U3 D! {+ u9 N+ M" e4 l
Two candidates for the ability to regenerate myocardial tissue are HSCs and nonhematopoietic MSCs, as previously reported . To reveal which stem cells were capable of differentiating into CMs, we compared an HSC population and a nonhematopoietic MSC population for their ability to differentiate into CMs when cocultured with BATDCs. We sorted Lin¨Cc-Kit  cells and CD45¨CCD31¨CCD105  cells in BM from green mice, as typical of HSC and nonhematopoietic MSC populations, respectively (Fig. 6A), and then cocultured them with CD133-positive BATDCs, as shown in Figure 2B. After 14 days, cultured cells were stained with cardiac-specific antigen, which revealed that SA /GATA-4  cells were generated from both populations; however, the efficiency for differentiation into CMs from those two populations was quite different. As shown in Figure 6B, the number of SA /GATA-4  CMs from CD45¨CCD31¨CCD105  was over 20 times greater than that from Lin¨Cc-Kit  cells (Fig. 6B). These results suggested that nonhematopoietic MSCs are the major source of CMs in the BM.
# j' B0 u* H# l9 l- K. n. K' F, X. K& w
Figure 6. Nonhematopoietic MSCs in bone marrow (BM) mainly differentiate into cardiomyocytes (CMs). (A): Cells derived from BM were stained with anti-lineage (mixture of ter119, Mac-1, Gr-1, CD4, CD8, and B220 antibodies) marker or anti-CD31 and anti-CD45 antibodies (x-axis) and anti-c-Kit or anti-CD105 antibody (y-axis). The percentage of lin¨Cc-Kit  cells and CD45¨CCD31¨CCD105  cells in BM are indicated in the box. (B): Quantitative evaluation of differentiated SA- and GATA-4-positive CMs from fractionated BMMNCs, as indicated under coculturing with CD133  BATDCs. Among the adhering BM cells, 10.7% (850 ¡À 172) and 0.4% (23 ¡À 3) were SA /GATA-4  cells from lin¨Cc-Kit  and CD45¨CCD31¨CCD105  cells, respectively. BMMNCs alone did not produce CMs. Results represent the mean of five independent experiments. Abbreviations: BATDC, brown adipose tissue-derived cell; BMMNC, bone marrow mononuclear cell.9 q9 L% f5 p1 `* ]* V4 I

* W* E2 W6 E& I5 Y, C1 q. V3 tDISCUSSION
7 Z6 e5 Q: n8 }- d! W; M& L5 a) V* S$ Z
Previously, we showed that BATDCs possessed cardiac progenitor cells, which contributed to the in vivo regeneration of damaged cardiac tissues . The results from these previous studies, taken alongside those from our present study, strongly suggest that CD133 might be the CSC marker for stem cell sources in adipose and other tissues. CD133-expressing cells in BAT from the interscapular region were relatively abundant in the embryonic and neonatal stage but were reduced in the adult stage (3.5% vs. 0.2%; data not shown). Moreover, BAT does not exist in great abundance in the adult human. Therefore, BAT itself might not have a clinical application for myocardial disease in adult patients. However, using the evidence that BATDCs can differentiate into CMs and CD133-positive cells from BAT can induce BMMNCs into CMs, we might be able to investigate how BMCs can differentiate into CMs at a molecular level.
2 k( g# ]* v- Z- G" F, ~" b' x( b/ ]5 o2 W5 U
In this study, we found that when cocultured in vitro with BATDC, BMCs themselves can give rise to CMs effectively without cell fusion. In the coculture system on CD133  BATDCs, 1 x 105 total BMMNCs, which logically contained about 140 CD45¨CCD31¨CCD105  cells, generated 520 SA GATA4  cells. On the other hand, in the coculture of nonhematopoietic cells from BM with CD133  BATDCs, 1 x 104 CD45¨CCD31¨CCD105  cells generated approximately 850 SA GATA4  cells. This low efficiency of induction for CMs from CD45¨CCD31¨CCD105  cells may have been due to the lack of other adherent cells observed in the culture using total BMMNCs. At present, we could not determine which adherent cells from BMMNCs were important; however, macrophage and/or endothelial cells may play a role in the generation of CMs synergistically with CD133  BATDCs.
5 M9 i0 P8 G. `4 D& o3 |% r, {8 @5 }/ H
In the coculture system, when cell-to-cell contact coculture conditions were compared with this filter-separated coculture system, the former effectively induced BMCs into CMs. The culture supernatant of cultured BATDCs, even concentrated, induced BMCs into CMs with an efficiency similar to that observed in the separate coculture conditions. These results suggested that there might be two independent ways of inducing CMs from BMCs. One way is to use the fact that secreted factors determine the fate of BMCs into CMs and the other is that membrane proteins do. With respect to the latter, we suggested that cadherins may mediate this crucial cell-to-cell contact. For the physical interaction of cohering cells, cadherins regulate diverse signaling process, such as differentiation, proliferation, and migration. As previously reported, Ecad-mediated cell-cell interaction, among cells that contain primordial germ cell precursors, is essential to directing such cells to the germ cell fate . Indeed, calcium depletion with EDTA or EGTA, which prevents cadherin-mediated cell-to-cell contact, abolished the adhesion of BMMNCs to CMs. In addition, cadherins are expressed in GFP  BMMNC-derived cells and are localized to the sites of cell-to-cell contact between BMMNCs and BAT CMs (Fig. 4A). Among several type of cadherins, we detected that the suppression of Ecad function was effective in inhibiting both the adhesion of BMMNCs to BATDCs and their differentiation into CMs (Fig. 4B). These data suggest a potential role of Ecad in the differentiation of BMMNCs into CMs. However, additional experiments will be necessary to show the molecular mechanisms behind the cell fate decision that involve cell-to-cell contact. Moreover, regarding the possibility of the existence of a secreted factor from BATDCs for the induction of BMCs into CMs, we found that CD133  cells expressed platelet-derived growth factor (PDGF)-AB (data not shown), and BMCs expressed platelet-derived growth factor receptor  (PDGFR-) (data not shown). To understand the contribution of the PDGF-PDGFR system, we added neutralizing antibody against PDGFR- into the coculture of CD133  BATDCs and BMCs. Some inhibition of the differentiation from BMCs into SA GATA4  CMs was observed after 14 days of culturing (data not shown). Moreover, we found that BATDCs expressed beneficial cytokines, such as vascular endothelial growth factor (VEGF), hepatocyte growth factor (HGF), and angiopoietin-1 (data not shown). These and other, unknown factors induced the proliferation and antiapoptotic effect of e-BMCs and contributed to the effective differentiation into SA GATA4  CMs. With these data, taken together, we confirmed that there might be two processes involved in the effective induction of BMCs into CMs: a secreted protein and cell-to-cell contact mediated by a membrane-binding protein.
7 Z9 V& j7 X$ [! ?3 ?& S% a6 e4 D0 ]% i6 G8 ]$ @1 n$ e
With the in vivo infarction model, e-BMCs effectively differentiated into CMs and improved cardiac function. In regard to this point, we found that VEGF and HGF were highly expressed by e-BMCs compared with non-e-BMCs. Moreover, approximately twice the number of CD31  ECs was observed in the transplanted region of e-BMCs compared with that of non-e-BMCs, and we also found that a small number of e-BMCs incorporated as CD31  ECs or -smooth muscle actin  mural cells (data not shown). These results indicated that the paracrine effect on neighboring cardiac myocytes and angiogenesis also contributed to the beneficial effects of transplantation.
- Q+ r$ x. z2 H0 t. S6 B0 y
" p  v" r( P1 t0 y" p- `  GFurthermore, we found no significant difference in the total number of GFP-positive cells located around the ischemic border zone between BMMNCs exposed to BATDCs (educated BMMNCs) and noneducated BMMNCs; however, SA /GFP  CMs derived from educated BMMNCs were more than 15 times more abundant in number compared with those from noneducated BMMNCs. Recently, some groups have indicated that VEGFR-2 /VE-cadherin¨C primitive cells from embryonic stem cells do not contribute to vascular cells, such as ECs or SMCs. By contrast, differentiated VE-cadherin  EPCs contributed to the vascular formation as ECs . These studies suggested that committed immature cells are better sources for vascular or myocardial regeneration than undirected mesodermal-derived stem cells.
; V1 r6 M; ^( `5 H' Z' I) ^0 H9 z1 B
) f2 i7 a4 x+ f  nIn summary, we suggest that CD133-positive cells in BATDCs form an enriched stem cell population, as well as being effective inducers from MSCs in BM to CMs in vitro and in vivo. CD133-positive cells in BATDCs might prove to have potential for the regeneration of CMs.  A/ W! i  |, ~' y& ?( S. x: b. _! C

+ r9 k" S/ f3 p) GDISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
' `5 n9 z, E& E$ m! I/ W2 _6 F3 G9 g) k3 O2 l3 R" p
The authors indicate no potential conflicts of interest.
5 V) `( r7 Z+ A7 w( y& o. B, w" ^' X- O3 K1 e
ACKNOWLEDGMENTS1 u& N  w; l" |

( y( H: q( s' ~+ q+ TWe thank Dr. M. Okabe (Osaka University, Osaka, Japan) for providing green mice and green rats and M. Sato for technical support. This study was supported in part by a grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan (N.T.).) p- n8 i* O: Y  f
          【参考文献】
- Z5 {& j! P) {& m) f % w  F' r  @' W" {5 a

9 C/ @$ g% s  N3 i+ `) WOrlic D, Kajstura J, Chimenti S et al. Bone marrow cells regenerate infarcted myocardium. Nature 2001;410:701¨C705.
" Q& q2 [) b8 a4 Y
4 E8 o2 L# U& N2 |Jackson KA, Majka SM, Wang H et al. Regeneration of ischemic cardiac muscle and vascular endothelium by adult stem cells. J Clin Invest 2001;107:1395¨C1402.
' d% m( w3 i. S! a& u
$ G: W3 T% j# T9 _9 vToma C, Pittenger MF, Cahill KS et al. Human mesenchymal stem cells differentiate to a cardiomyocyte phenotype in adult murine heart. Circulation 2002;105:93¨C98.; ~# w7 a" O3 X# h, L' o0 q

/ [/ k# u% k* m! o0 nMakino S, Fukuda K, Miyoshi S et al. Cardiomyocytes can be generated from marrow stromal cells in vitro. J Clin Invest 1999;103:697¨C705.) b0 r9 J  G2 |' P: T0 o. b" W0 ?0 J

- d- R& c7 y' F  M- P% kTomita S, Li RK, Weise RD et al. Autologous transplantation of bone marrow cells improves damaged heart function. Circulation 1999;100:247¨C256.8 }; s/ r) B: `$ D5 n

! g( W  V0 `( D& \7 xMurry CE, Soonpaa MH, Reinecke H et al. Haematopoietic stem cells do not transdifferentiate into cardiac myocytes in myocardial infarcts. Nature 2004;428:664¨C668.7 \& |4 U  S& u% m

7 {, c8 m+ g; ~" DBalsam LB, Wagers AJ, Christensen JL et al. Haematopoietic stem cells adopt mature haematopoietic fates in ischaemic myocardium. Nature 2004;428:668¨C673.
- r+ O" N" M/ C1 m/ N7 T" w  j
5 g: r; t; ?8 s! K8 HAlvarez-Dolado M, Pardal R, Garcia-Verdugo JM et al. Fusion of bone-marrow-derived cells with Purkinje neurons, cardiomyocytes and hepatocytes. Nature 2003;425:968¨C973.8 u& y* |2 K, _) m8 C5 {" g+ [( ^% v
* P5 e8 p8 n2 j  A8 o5 A9 O# i' w/ T
Yamada Y, Wang X-D, Yokoyama S-I et al. Cardiac progenitor cells in brown adipose tissue repaired damaged myocardium. Biochem Biophys Res Commun 2006;342:662¨C670.' }" k' C  `3 g7 A
$ Y, X8 e  \' u" |) j. M8 ~7 l0 `, S1 Q# b
Weigmann A, Corbeil D, Hellwing A et al. Prominin, a novel microvilli-specific polytopic membrane protein of the apical surface of epithelial cells, is targeted to plasmalemmal protrusions of non-epithelial cells. Proc Natl Acad Sci U S A 1997;94:12425¨C12430.
' x& {; y  B% V3 J# a0 X
) X4 h8 }' O& ?0 p7 l2 S1 u0 r" uYin AH, Miraglia S, Zanjani E et al. AC133, a novel marker for human hematopoietic stem and progenitor cells. Blood 1997;90:5002¨C5012.
, m" H9 @' f' I8 m
: p- m$ S, t% B) u; t3 b5 b; f  SPeichev M, Naiyer AJ, Pereira D et al. Expression of VEGFR-2 and ACC133 by circulating human CD34( ) cells identifies a population of functional endothelial precursors. Blood 2000;95:952¨C958.6 d% _9 q& e1 J$ E; N

% e! `( J1 n$ S3 P/ f3 iUchida N, Buck DW, He D et al. Direct isolation of human central nervous system stem cells. Proc Natl Acad Sci U S A 2000;97:14720¨C14725.; E) l9 ]% P% F+ I0 W+ h8 w5 F3 ~1 q

8 f. y, U. m! g' `) _, jCorbeil D, Roper K, Hellwig A et al. The human AC133 hematopoietic stem cell antigen is also expressed in epithelial cells and targeted to plasm membrane protrusions. J Biol Chem 2000;275:5512¨C5520.- j" y% C6 C  T4 P3 p
/ D8 n3 G  I/ x4 ]" J& M
Richardson GD, Robson CN, Lang SH et al. CD133, a novel marker for human prostatic epithelial stem cells. J Cell Sci 2004;117:3539¨C3545.. C$ x. \( v2 t+ G1 y! l: x

8 o. X+ N) @/ g+ k9 b3 _# `  X* \Torrente Y, Belicchi M, Sampaolesi M et al. Human circulating AC133  stem cells restore dystrophin expression and ameliorate function in dystrophic skeletal muscle. J Clin Invest 2004;114:182¨C195.5 l9 [& \9 X5 v$ i' C

, f0 p( A& u3 W7 X) I  ZMaw MA, Corbeil D, Koch J et al. A frameshift mutation in prominin(mouse)-loke 1 causes human retinal degeneration. Hum Mol Gen 2000;9:27¨C34.
! p9 g9 n" o3 F7 E5 B! ]# ~9 ]* Q" B4 W% o8 t* T+ u- V
Singh SK, Hawkins C, Clarke ID et al. Identification of human brain tumour initiating cells. Nature 2004;432:396¨C401." q% |5 ~: `2 I; J' |
# Z7 s7 m2 [+ f. M0 T! h
Yamada Y, Takakura N. Physiological pathway of differentiation of hematopoietic stem cell population into mural cells. J Exp Med 2006;203:1055¨C1065.1 _/ B) }$ \6 z  ]

) ~# C" c: Z% u+ ]7 ?& TYamada Y, Takakura N, Yasue H et al. Exogenous clustered neuropilin 1 enhances vasculogenesis and angiogenesis. Blood 2001;97:1671¨C1678.
/ [3 t: T3 f" E" N7 h: B5 x7 p. b6 m3 d* I# a9 k* s
Oh H, Bradfute SB, Gallardo TD et al. Cardiac progenitor cells from adult myocardium: Homing, differentiation, and fusion after infarction. Proc Natl Acad Sci U S A 2003;100:12313¨C12318." w$ V1 u4 h8 w$ w1 t* D

  A4 u' V8 u5 f1 Y; MMoretti A, Caron L, Nakano A et al. Multipotent embryonic isl1  progenitor cells lead to cardiac, smooth muscle, and endothelial cell diversification. Cell 2006;127:1151¨C1165.6 \  D' F6 N1 V$ E) V3 f' w, \
7 E2 s2 ~8 t4 f+ c  R' ?
Wu SM, Fujiwara Y, Cibulsky SM. Developmental origin of a bipotential myocardial and smooth muscle cell precursor in the mammalian heart. Cell 2006;127:1137¨C1150.
6 v+ a" d' w( V8 l$ p- H4 m+ D# @! T) A6 d4 X
Ito T, Suzuki A, Okabe M et al. Application of bone marrow-derived stem cells in experimental nephrology. Exp Nephrol 2001;9:444¨C450.
: h% ?) a) \, e6 [: W: I* S$ K0 u% q' ?+ _1 L. L8 D0 O: b* _
Zuk PA, Zhu M, Ashjian P et al. Human adipose tissue is a source of multipotent stem cells. Mol Biol Cell 2002;13:4279¨C4295.
, _/ u. {) I. [5 D
2 X8 a+ t. v0 T9 jPlanat-B¨¦nard P, Menard C, Andre M et al. Spontaneous cardiomyocyte differentiation from adipose tissue stroma cells. Circ Res 2004;94:223¨C229.8 A3 U+ H$ Z7 L$ ]1 V+ l
+ Y* l1 b' q6 u
Beltrami AP, Barlucchi L, Torella D et al. Adult cardiac stem cells are multipotent and support myocardial regeneration. Cell 2003;114:763¨C776.
5 c3 ^, A0 G( e9 B" D* ]5 p6 [+ J9 g8 D4 p  {: H0 {
Okabe M, Ikawa M, Kominami K et al. ¡®Green mice' as a source of ubiquitous green cells. FEBS Lett 1997;407:313¨C319.
/ M1 F1 u. M3 R" C/ T& S" Q
( l. u% x3 A7 e9 \# tTerada N, Hamazaki T, Oka M et al. Bone marrow cells adopt the phenotype of other cells by spontaneous cell fusion. Nature 2002;416:542¨C545.' v, `2 X2 j: L6 w" U- j
, }9 Q, W) U2 D0 @4 @; @4 ^* W
Okamura D, Kimura T, Nakano T et al. Cadherin-mediated cell interaction regulates germ cell determination in mice. Development 2003;130:6423¨C6430., }5 ]/ V, q- r5 n) P3 G+ `

! c6 w: z. E3 _( ]- h" s* T+ _, b6 DBadorff C, Brandes RP, Popp R et al. Transdifferentiation of blood-derived human adult endothelial progenitor cells into functionally active cardiomyocytes. Circulation 2003;107:1024¨C1032.5 a' `' U, o6 w) K: c# L' }

5 S3 U8 ^. Z/ o# K" INakamura T, Sano M, Songyang Z et al. A Wnt-and beta-catenin-dependent pathway for mammalian cardiac myogenesis. Proc Natl Acad Sci U S A 2003;100:5834¨C5829.
7 D$ [" u% s' D8 A1 }" W9 p0 @: G/ w3 P/ M7 |6 q
Sone M, Itoh H, Yamashita J et al. Different differentiation kinetics of vascular progenitor cells in primate and mouse embryonic stem cell. Circulation 2003;107:2085¨C2088.

橙味绿茶

看看..  

红旗

表观遗传学

科研人

写得好啊  

大小年

厉害！强~~~~没的说了！  

大小年

哈哈,顶你了哦.  

yukun

好困啊  

nauticus

今天的干细胞研究资料更新很多呀

s06806

楼上的稍等啦  

昕昕

小心大家盯上你哦