HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 2007-0106v1 26/2/580    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lu, L. Articles by Xiao, Z.-C. Search for Related Content PubMed PubMed Citation Articles by Lu, L. Articles by Xiao, Z.-C.

TRANSLATIONAL AND CLINICAL RESEARCH

Morphological and Functional Characterization of Predifferentiation of Myelinating Glia-Like Cells from Human Bone Marrow Stromal Cells Through Activation of F3/Notch Signaling in Mouse Retina

^aDepartment of Clinical Research, Singapore General Hospital, Singapore;
^bDepartment of Anatomy, Beijing University of Traditional Chinese Medicine, Beijing, China;
^cDepartment of Anatomy, Shanxi Medical University, Shanxi, China;
^dInstitute of Neuroscience, Nantong University, Nantong, China;
Departments of ^eAnatomy and
^hPharmacology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore;
^fDivision of Life Science and Biotechnology, Ocean University of China, Qingdao, China;
^gCentre of Molecular Medicine, Singapore;
ⁱInstitute of Molecular and Cell Biology, Singapore

Key Words. Oligodendrocytes • F3 • Notch • Retina • Stem cells

Correspondence: Correspondence: Zhi-Cheng Xiao, M.D., Ph.D., Institute of Molecular and Cell Biology, 61 Biopolis Drive, Proteos, Singapore 138673 or Department of Clinical Research, Singapore General Hospital, Block A, 7 Hospital Drive, Singapore 169608. Telephone: 65-6326-6195; Fax: 65-6321-3606; e-mail: xiao.zhi.cheng{at}sgh.com.sg; or Gavin S. Dawe, Department of Pharmacology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore 117597. Telephone: 65-6516-8864; Fax: 65-6873-7690; e-mail: gavindawe{at}nus.edu.sg

Received on February 8, 2007; accepted for publication on October 19, 2007.

First published online in STEM CELLS EXPRESS  November 1, 2007.

	ABSTRACT

Top Abstract Introduction Materials and Methods Results Discussion Disclosure of Potential... Acknowledgments References

 
Recently, we have demonstrated that F3/contactin and NB-3 are trans-acting extracellular ligands of Notch that promote differentiation of neural stem cells and oligodendrocyte precursor cells into mature oligodendrocytes (OLs). Here, we demonstrate that human bone marrow stromal cells (hBMSCs) can be induced to differentiate into cells with myelinating glial cell characteristics in mouse retina after predifferentiation in vitro. Isolated CD90(+) hBMSCs treated with β-mercaptoethanol for 1 day and retinoic acid for 3 days in culture changed into myelinating glia-like cells (MGLCs). More cells expressed NG2, an early OL marker, after treatment, but expression of O4, a mature OL marker, was negligible. Subsequently, the population of O4(+) cells was significantly increased after the MGLCs were predifferentiated in culture in the presence of either F3/contactin or multiple factors, including forskolin, basic fibroblast growth factor, platelet-derived growth factor, and heregulin, in vitro for another 3 days. Notably, 2 months after transplantation into mouse retina, the predifferentiated cells changed morphologically into cells resembling mature MGLCs and expressing O4 and myelin basic protein, two mature myelinating glial cell markers. The cells sent out processes to contact and wrap axons, an event that normally occurs during early stages of myelination, in the retina. The results suggest that CD90(+) hBMSCs are capable of morphological and functional differentiation into MGLCs in vivo through predifferentiation by triggering F3/Notch signaling in vitro.

Disclosure of potential conflicts of interest is found at the end of this article.

	INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion Disclosure of Potential... Acknowledgments References

 
Myelination in the vertebrate central nervous system (CNS) is essential for rapid impulse conduction [1]. Stem cells derived from bone marrow have been successfully engrafted into the CNS after i.v. transplantation [2] and so may have potential for the repair of widespread and diffuse CNS lesions, such as in multiple sclerosis, a demyelinating disease. Stem cells derived from bone marrow are readily differentiated into cells expressing neuronal, astrocytic, and Schwann cell markers in culture [2, 3]. However, it is more difficult to differentiate them into oligodendrocytes (OLs) [2]. At present, there are no simple and reliable means of selectively differentiating human bone marrow stromal cells (hBMSCs) to oligodendrocyte precursor cells (OPCs) or OL lineages in both in vitro and in vivo conditions, although it has been reported previously that hBMSCs can differentiate into myelin basic protein (MBP)-positive OLs in vitro [4].

Recently, we demonstrated that the F3/Notch signaling pathway promotes the differentiation of both neural stem cells and OPCs into mature OLs [1, 5–7]. Notch is a type I transmembrane protein mediating cell fate selection via lateral inhibition. Its core signaling mechanism involves regulated intramembrane proteolysis [8]. Upon binding the classic ligands, Delta, Serrate/Jagged, and Lag-2 (collectively called DSL), Notch undergoes proteolytic cleavage that releases its intracellular domain (Notch intracellular domain [NICD]). NICD translocates to the nucleus and interacts with the RBP-J (CSL) transcription factor to activate gene transcription, for example, transcription of Hes genes [9]. Notch signaling via Deltex1 (DTX1) represses JNK signaling, a pathway regulating OL differentiation, and cooperates with Wingless signaling [9, 10]. F3/contactin and NB-3 are glycosyl phosphatidylinositol (GPI)-anchored neural cell adhesion molecules of the immunoglobulin superfamily [11]. They are neuronal receptors for the oligodendrocyte-related extracellular matrix glycoprotein tenascin-R via epidermal growth factor-like repeats [12]. F3 interacts in cis with receptor protein tyrosine phosphatase [13] to transduce extracellular signals to myelination-related Fyn kinase [14]. In addition, F3 colocalizes and interacts in cis with Caspr/Paranodin and in trans with glial neurofascin 155 at the paranode [15], a key site of axoglial contact for myelination. F3-null mice exhibit partially disrupted paranodal structure and die by P18 [16], suggesting that F3 is critical for development. Moreover, the F3/Notch interaction upregulates the expression of myelin-related proteins in a Deltex1-dependent manner [1, 5–7].

To use hBMSCs as a source of stem cells for cell therapies in demyelinating disease or injury, reliable means for selective predifferentiation into OPCs or OL lineages are required. In the present study, we have demonstrated that hBMSCs can be induced to differentiate into cells with myelinating glial cell characteristics in mouse retina after predifferentiation by stimulating the F3/Notch signaling pathway in vitro.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion Disclosure of Potential... Acknowledgments References

 
Animals
Eight-week-old C57BL/6 mice were used. The Institutional Animal Care and Use Committee of Singapore General Hospital approved all experiments involving mice. The ethical review board of Singapore General Hospital approved the use of human cells.

Culture of hBMSCs
hBMSCs (donated with informed consent by a 38-year-old male and supplied by Cambrex, Walkersville, MD, http://www.cambrex.com) were plated in a plastic dish with complete Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 µg/ml streptomycin and then incubated at 37°C, 95% humidity, and 5% CO₂. After 48 hours, the nonadherent cells were removed by washing with phosphate-buffered saline (PBS). The medium was replaced with fresh medium, and the adherent cells were grown to 90% confluence to obtain samples here defined as passage 0 (P0) hBMSCs. hBMSCs were subcultured four times and used in the following experiments.

In Vitro Differentiation
hBMSCs were chemically induced to oligodendrocyte-like cells in vitro according to a previously described method [17]. Briefly, subconfluent cultures were preinduced for 24 hours with DMEM containing 1 mM β-mercaptoethanol (β-ME; Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com), and the medium were removed, washed with PBS, and replaced with new medium consisting of DMEM, 10% FBS, and 35 ng/ml all-trans-retinoic acid (RA; Sigma-Aldrich) for 3 days. Cells were washed with PBS and randomly divided into two groups treated with either multiple factors (MFS) or F3. The MFS-treated group was cultured in DMEM contained 10% FBS, 5 mM forskolin (Sigma-Aldrich), 10 ng/ml recombinant human basic-fibroblast growth factor (bFGF; Sigma-Aldrich), 5 ng/ml recombinant human platelet-derived growth factor (PDGF)-AA (R&D Systems Inc., Minneapolis, http://www.rndsystems.com), and 200 ng/ml recombinant human heregulin-β1 (HRG; Sigma-Aldrich). The F3-treated group was cultured in DMEM containing 10% FBS and 20 nM F3. Cells were then incubated for 3 days after addition of MFS or F3.

Flow Cytometry Analysis
Single-cell suspensions were obtained from 10-cm-diameter culture dishes using 0.25% trypsin/EDTA treatment. Cells were washed three times in PBS, counted, and adjusted to appropriate concentrations. Per sample, 1 x 10⁶ cells were used; cells were stained with monoclonal antibodies against CD34, CD45, and CD90 (1:100; Chemicon, Temecula, CA, http://www.chemicon.com), which were conjugated with fluorescein isothiocyanate (FITC). The cells were analyzed using a flow cytometer (FACSCalibur and CellQuest Pro software; Becton, Dickinson and Company, San Jose, CA, http://www.bd.com).

To analyze the nature of differentiated hBMSCs in vitro, cells were stained with polyclonal rabbit antibody against NG2 (1:200; Chemicon) and monoclonal mouse antibody against O4 (mouse IgM, 1:200; Chemicon). After staining with the primary unconjugated antibodies, the cells were incubated with phycoerythrin-conjugated anti-rabbit IgG and FITC-conjugated anti-mouse IgM (Invitrogen, Carlsbad, CA, http://www.invitrogen.com).

Luciferase Reporter Assay
The Hes1 assay system has been previously described [5]. Briefly, hBMSCs in 24-well dishes were induced to differentiate under different conditions (RA/β-ME, MFS/RA/β-ME, or F3/RA/β-ME) and transiently transfected using the Human Mesenchymal Stem Cell Nucleofector kit (Amaxa Inc., Cologne, Germany, http://www.amaxa.com). Each well was transfected with pGVB/Hes1 luciferase reporter plasmid together with various plasmids, including DTX1 and dominant-negative DTX1. The β-galactosidase expression plasmid pCMV/β-Gal was included as an internal control. Cells were lysed 24-hour post-transfection and assayed using the Steady-Glo Luciferase Assay Kit (Promega, Madison, WI, http://www.promega.com). The raw data from at least four independent experiments were used to determine the relative reporter activity.

Transplantation of Cells into the Mouse Eye
Eight-week-old C57BL/6 mice were deeply anesthetized with 1% isoflurane. A glass micropipette was introduced into the superior juxtaretinal space of the left eye, and approximately 2 µl of vitreous fluid was removed from the eye under visual control, using a surgical microscope (Leica, Wetzlar, Germany, http://www.leica.com). Subsequently, the same volume of predifferentiated cells (suspended in PBS) was injected (approximately 50,000 cells per microliter). PBS without cells was injected as a negative control (sham injection). Mice were injected with (a) the undifferentiated cells, (b) the predifferentiated cells induced by RA/β-ME, (c) the predifferentiated cells induced by MFS/RA/β-ME, and (d) the predifferentiated cells induced by F3/RA/β-ME. All animals received daily injection of 10 mg/kg cyclosporine (Novartis International, Basel, Switzerland, http://www.novartis.com). Animals with transplanted cells or sham-injected animals were analyzed 2 months after injection.

Immunofluorescence
To characterize undifferentiated and differentiated hBMSCs in vitro, cells were cultured on poly-L-lysine-coated coverslips. After plating, the following antibodies were used: mouse monoclonal antibody against vimentin (1:200; Chemicon), rabbit polyclonal antibody against bone morphogenetic protein receptor (BMPR) 1A (1:100; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com), rabbit polyclonal antibody against Sox2 (1:500; Chemicon), mouse monoclonal antibody against human nestin (1:100; Chemicon), rabbit polyclonal antibody against NG2 (1:100; Chemicon), monoclonal mouse antibody against O4 (1:100; Chemicon), goat polyclonal antibody against Deltex-1 (1:200; Santa Cruz Biotechnology), rabbit polyclonal antibody against Notch/NICD (1:100; Cell Signaling Technology, Danvers, MA, http://www.cellsignal.com), rabbit polyclonal antibody against MBP (1:200; Chemicon), rabbit polyclonal antibody against glial fibrillary acidic protein (GFAP) (1:200; Chemicon), and rabbit monoclonal antibody against β-tubulin III (1:300; Covance, Berkeley, CA, http://www.covance.com). Incubations were performed overnight at 4°C. FITC-conjugated goat anti-mouse IgM (1:200; Chemicon), Cy3-conjugated goat anti-rabbit IgG (Chemicon), and Cy2-conjugated goat anti-rabbit or donkey anti-goat (Chemicon) were used as secondary antibodies.

For immunostaining of retinas, animals were deeply anesthetized by an intraperitoneal injection of sodium pentobarbitone and perfusion-fixed with 0.9% saline followed by PBS containing 4% paraformaldehyde for 20 minutes. After the injected eyes were enucleated, the retinas were incubated overnight with β-tubulin III monoclonal antibody (1:300; Chemicon) and O4 monoclonal antibody (1:100; Chemicon) at 4°C. Primary antibodies were visualized with Cy3-conjugated goat anti-mouse IgG or FITC-conjugated goat anti-mouse IgM antibodies (1:200; Chemicon). Sham-injected eyes were used as a negative control. The preparations were visualized under confocal laser scanning microscopy (LSM510; Carl Zeiss, Jena, Germany, http://www.zeiss.com).

Electron Microscopy
For electron microscopic analysis, animals were deeply anesthetized by an intraperitoneal injection of sodium pentobarbitone and transcardially perfused with 2% paraformaldehyde/3% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4). The retina was removed and postfixed in the same fixative solution for 4 hours. After being finely trimmed, the samples were immersion-washed overnight. Tissue blocks were further postfixed in 1% osmium tetroxide for 2 hours and thereafter subjected to dehydration in an ascending series of alcohol and acetone. After undergoing gradual infiltration with Araldite 502 (Electron Microscopy Systems, Hatfield, PA, http://www.emsdiasum.com), the blocks were embedded and polymerized overnight at 60°C. Ultrathin sections (90 nm in thickness) were placed on 150-mesh copper grids and counterstained with uranyl acetate and lead citrate. All samples were examined and photographed under a Jeol 1220 electron microscope (Jeol, Tokyo, http://www.jeol.com).

Statistical Analysis
Values are given as means ± SEM. All experiments were repeated at least three times. Comparison of data was performed with one-way analysis of variance. Differences yielding p < .05 were regarded as significant.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion Disclosure of Potential... Acknowledgments References

 
Isolation of CD90(+) Cells from hBMSCs
Neural progenitor cells (NPCs) strongly express CD90 [18], which has also been considered to be a marker of hematopoietic stem cells [19]. Moreover, CD90(+) cells isolated from primary neurospheres have the ability to generate neurospheres [20]. To differentiate hBMSCs into myelinating glia-like cells (MGLCs), the first step was to isolate CD90(+) cells from hBMSCs. hBMSCs, bought from Cambrex, were cultured for approximately 2 days with DMEM containing 10% FBS. Nonadherent cells were removed by changing the medium. Adherent cells were predominantly round, but a subpopulation extended processes and had highly refractive cell bodies. After 3 days in culture, the adherent cells had multiplied in culture, and the cells were approaching confluence. The confluent cells that formed colonies displayed diverse appearances: some were large, flat, and polygonal, some were spindle-shaped, and others had multiple processes (Fig. 1Aa, 1Ab). The adherent cells were grown to 90% confluence and then subjected to flow cytometry analysis. Results showed that they were predominantly CD90(+) cells in a range from 83.8% to 97.7% and an average level of 94.78% ± 4.52%, although a few CD34(+) (4.7%) and CD45(+) (5.8%) cells were present (Fig. 1Ac, 1Ad, 1Ae; supplemental online Fig. 1A). In addition, adherent cells were passed through CD90 antibody beads and subjected to flow cytometry analysis. The CD90 bead-collected cells consisted of CD90(+) (98.35%) cells, whereas the flow-through cells from the CD90 beads also consisted of CD90(+) (96.26%) cells (supplemental online Fig. 1B). These results demonstrate that the adherent hBMSCs are mainly CD90(+) cells.

View larger version (46K): [in this window] [in a new window]   Figure 1. Isolation and characterization of CD90(+) human bone marrow stromal cells (hBMSCs). (A): Total hBMSCs were plated on dishes and cultured for 2 days. After the nonadherent cells were removed by changing the medium, the adherent cells were grown to 90% confluence and subjected to flow cytometry analysis. (Aa, Ab): Phase-contrast images of the adherent hBMSCs. (Ac–Ae): The adherent hBMSC were labeled by FITC-conjugated antibodies and examined by flow cytometry. Flow cytometry results indicated that the adherent hBMSCs consisted of CD34(+) (4.7%) (Ac), CD45(+) (5.8%) (Ad), and CD90(+) (83.8%) (Ae) cells. Scale bars = 20 µm (Aa) and 10 µm (Ab). (B): The adherent hBMSC were double-stained with 4,6-diamidino-2-phenylindole and antibodies against vimentin (Ba), BMPR (Bb), Sox2 (Bc), and nestin (Bd). The vimentin(+) (96.45% ± 0.97%), BMPR(+) (97.74% ± 0.79%), Sox2(+) (40.88% ± 3%), and nestin(+) (2.38% + 0.5%) cells among the adherent hBMSCs were counted (Be). Scale bar = 15 µm (Bd) (applies to [Ba–Bd]). Abbreviations: BMPR, bone morphogenetic protein receptor 1A; FITC, fluorescein isothiocyanate.

Induction of Immature NG2(+) MGLCs from CD90(+) hBMSCs
CD90(+) cells isolated from primary neurospheres have the ability to differentiate into immature OLs [25]. To investigate whether the CD90(+) hBMSCs have similar ability, we treated the cells with DMEM containing 1 mM β-ME for 1 day and followed with DMEM, 10% FBS, and 35 ng/ml RA for 3 days. Flow cytometry analysis showed that the untreated cells were negative for both NG2, a marker for immature OLs, and O4, a marker for mature OLs (Fig. 2Aa, 2Ad). The number of NG2(+) cells was increased after treatment with β-ME (5.0%; Fig. 2Ab, 2Ad) and more significantly increased after treatment with β-ME plus RA (54.43%; Fig. 2Ac, 2Ad). In contrast, the number of O4(+) cells was negligible after these treatments (Fig. 2A). Similar results were obtained with NG2 and O4 immunofluorescence staining (Fig. 2B). Notably, after β-ME or β-ME plus RA treatment (Fig. 2Ae), the number of vimentin or BMPR or Sox2(+) cells was slightly reduced, whereas the number of nestin(+) cells was significantly increased, compared with untreated adherent hBMSCs (Fig. 1Ae). CD90(+) hBMSCs have been demonstrated to express some neuron and astrocyte markers under certain treatment conditions [26]. Here, our observations indicated that the cells could be also differentiated into immature NG2(+) MGLCs after induction with β-ME or β-ME plus RA in vitro.

View larger version (27K): [in this window] [in a new window]   Figure 2. Differentiation of CD90(+) human bone marrow stromal cells (hBMSCs) into NG2(+) myelinating glia-like cells. (A): Two-color flow cytometric analysis (NG2 and O4) was performed to characterize the nature of the differentiated hBMSCs under different culture conditions: Dulbecco's modified Eagle's medium (DMEM) plus fetal bovine serum (FBS) (control) (Aa), DMEM plus β-ME for 1 day (Ab), and DMEM plus β-ME for 1 day and then with the addition of RA for 3 days (Ac). The percentages of NG2(+) (β-ME, 5.31% ± 3.02%; RA + β-ME, 51.22% ± 1.62%), O4(+) (β-ME, 0.53% ± 0.21%; RA + β-ME, 5.59% ± 0.46%), vimentin(+) (β-ME, 69.17% ± 1.38%; RA + β-ME, 63.13% ± 1.58%), BMPR(+) (β-ME, 83.22% ± 1.84%; RA + β-ME, 74.6% ± 1.69%), Sox2(+) (β-ME, 31.48% ± 2.56%; RA + β-ME, 12.93% ± 1.61%), and nestin(+) (β-ME, 9.39% ± 0.84%; RA + β-ME, 26.73% ± 1.53%) cells were counted (Ad, Ae). *, p < .05; **, p < .01. (B): Double staining using 4,6-diamidino-2-phenylindole (blue) and antibodies against NG2 (green) (Ba–Bc) or O4 (green) (Bd–Bf) was performed to characterize the nature of the differentiated hBMSCs under different culture conditions: DMEM plus FBS (control) (Ba, Bd), DMEM plus β-ME for 1 day (Bb, Be), and DMEM plus β-ME for 1 day and then with the addition of RA for 3 days (RA + β-ME) (Bc, Bf). Scale bar = 10 µm (Bf) (applies to [Ba–Bf]). Abbreviations: BMPR, bone morphogenetic protein receptor 1A; β-ME, β-mercaptoethanol; RA, all-trans-retinoic acid.

View larger version (29K): [in this window] [in a new window]   Figure 3. Expression of Notch and DTX1 in differentiated human bone marrow stromal cells (hBMSCs). (A): Immunocytochemistry was performed by using antibodies against Notch intracellular domain (green) (Aa–Ae) and DTX1 (green) (Af–Aj) to investigate expression in differentiated hBMSCs under different culture conditions: Dulbecco's modified Eagle's medium (DMEM) plus fetal bovine serum (Aa, Af), DMEM plus β-ME for 1 day (Ab, Ag), DMEM plus β-ME for 1 day and then with the addition of RA for 3 days (Ac, Ah), DMEM/β-ME (for 1 day) plus RA for 3 days and then with the addition of MFS (including forskolin, basic fibroblast growth factor, platelet-derived growth factor, and heregulin-β1) for 3 days (Ad, Ai), and DMEM/β-ME (for 1 day) plus RA for 3 days and then with the addition of F3/contactin for 3 days (Ae, Aj). Scale bar = 10 µm (Aj) (applies to [Aa–Aj]). (Ba): After treatment of RA/β-ME, MFS/RA/β-ME, or F3/RA/β-ME, hBMSCs were transiently cotransfected in 24-well culture dishes with pGVB/Hes1 luciferase reporter plasmid and luciferase internal control plasmid. Normalized luciferase activities in whole-cell lysates were determined in triplicate and expressed relative to activity in lysates prepared from hBMSCs treated with RA/β-ME. (Bb): After treatment of RA/β-ME, MFS/RA/β-ME, or F3/RA/β-ME, hBMSCs were transiently cotransfected in 24-well culture dishes with pGVB/Hes1 luciferase reporter plasmid, DTX1, or a dnDTX1, and luciferase internal control plasmid. Normalized luciferase activities in whole-cell lysates were determined in triplicate and expressed relative to activity in lysates prepared from hBMSCs treated with DTX1. *, p < .05. **, p < .01. Error bars represent SEM. Abbreviations: dnDTX1, dominant-negative DTX1; DTX1, Deltex1; β-ME, β-mercaptoethanol; MFS, multiple factors; RA, all-trans-retinoic acid.

View larger version (34K): [in this window] [in a new window]   Figure 4. Induction of human bone marrow stromal cell (hBMSC) differentiation into O4(+) myelinating glia-like cells. The percentages of MBP(+) (MFS, 74.7% ± 2.4%; F3, 71.27% ± 1.71%), GFAP(+) (MFS, 71.67% ± 2.97%; F3, 69.03% ± 2.6%), β-tubulin III(+) (MFS, 29.17% ± 1.62%; F3, 14.75% ± 1.58%), NG2(+) (MFS, 7.9% ± 1.65%; F3, 8.67% ± 1.86%), and O4(+) (MFS, 36.0% ± 4.48%; F3, 34.29% ± 3.63%) cells were counted (Aa, Ad). Two-color flow cytometric analysis (NG2 and O4 [Ab, Ac]) and immunocytochemistry (NG2, green [Ba, Bd]; O4, green [Bb, Be]; MBP, green [Bc, Bf]) were performed to characterize the nature of the differentiated hBMSCs under two different culture conditions: Dulbecco's modified Eagle's medium (DMEM)/β-ME (for 1 day) plus RA for 3 days and then with the addition of MFS for 3 days (Ac, Ba–Bc) and DMEM/β-ME (for 1 day) plus RA for 3 days and then added F3/contactin for 3 days (Ab, Bd–Bf). Scale bars = 20 µm (Bd) (applies to [Ba, Bd]) and 10 µm (Bf) (applies to [Bb, Bc, Be, Bf]). Abbreviations: GFAP, glial fibrillary acidic protein; MBP, myelin basic protein; β-ME, β-mercaptoethanol; MFS, multiple factors; RA, all-trans-retinoic acid.

The Retina Is a Favorable In Vivo Environment for Examination of the Development of Predifferentiated MGLCs
The retina is devoid of cells of the oligodendrocyte cell lineage and is thus free of myelin [29–33]; it therefore offers an ideal site to further explore the ability of predifferentiated MGLCs to differentiate in vivo (Fig. 5Ab). O4 immunoreactivity was not detectable in sham-injected retina. Notably, the MGLCs derived from CD90(+) hBMSCs in vitro lacked any morphological features of OLs, although they express several oligodendrocyte or Schwann cell markers, such as NG2, O4, and MBP, as evidenced by immunofluorescence microscopy analysis. After in vitro inductions as described above, NG2(+) or O4(+) MGLCs displayed diverse morphologies: some were spindle-shaped; some were large, flat, and polygonal; and others had multiple processes (Fig. 5Ba, 5Ca, 5Da). After induction in vitro, the predifferentiated MGLCs were injected into mouse retina. Two months later, differentiation was assessed by immunocytochemistry using antibodies against O4. Significantly, O4(+) MGLCs from all three groups (RA/β-ME, MFS/RA/β-ME, and F3/RA/β-ME) displayed a typical oligodendrocyte morphology characterized by multiple processes in the retina (Fig. 5Bb, 5Bc, 5Cb, 5Cc, 5Db, 5Dc). The number of branches (longer than the diameter of the soma) per NG2(+) or O4(+) MGLCs was quantified in both in vitro and in vivo conditions. For all three groups, the number of branches per cell was significantly increased in the in vivo condition, compared with the in vitro condition (Fig. 5B–5D). These observations demonstrate that the in vivo retina offers a favorable environment to test the development of predifferentiated MGLCs.

View larger version (39K): [in this window] [in a new window]   Figure 5. Characterization of predifferentiated myelinating glia-like cells derived from human bone marrow stromal cells (hBMSCs) after transplantation into mouse retina. Shown are a structural diagram of MGLCs with branches and processes (Aa) and a schematic diagram of the transplantation procedure (Ab). Immunocytochemistry (NG2, [Ba, Bd]; O4, [Bb, Bc, Be, C, D]) was performed to characterize the nature of the predifferentiated myelinating glia-like cells derived from hBMSCs both in vitro (Ba, Bd, Ca, Cd, Da, Dd) and after transplantation into mouse retina for 2 months (Bb, Bc, Be, Cb, Cc, Ce, Db, Dc, De). (B): Dulbecco's modified Eagle's medium (DMEM) plus β-ME for 1 day and then with the addition of RA for 3 days. (C): DMEM/β-ME (for 1 day) plus RA for 3 days and then with the addition of MFS for 3 days. (D): DMEM/β-ME (for 1 day) plus RA for 3 days then with the addition of F3/contactin for 3 days. Scale bars = 15 µm (Db) (applies to [Ba, Bb, Ca, Cb, Da, Db]), 10 µm (Dc) (applies to [Bc, Cc, Dc]), and 5 µm (De) (applies to [Bd, Be, Cd, Ce, Dd, De]). The number of branches per cell (Ea) and the number of processes per branch (Eb) under both in vitro and in vivo conditions were counted. **, p < .01. Error bars represent SEM. Abbreviations: β-ME, β-mercaptoethanol; MFS, multiple factors; RA, all-trans-retinoic acid.

Functional Maturation of the MGLCs Predifferentiated by F3/Contactin in Mouse Retina
To investigate whether F3 could also play a role in promoting the myelination of the predifferentiated MGLCs in vivo, untreated CD90(+) hBMSCs (Fig. 6B), RA (Fig. 6C), MFS (Fig. 6D), and F3 (Fig. 6E)-predifferentiated MGLCs were injected into mouse retina, respectively, and sham injection was used as a control (Fig. 6A). Two months after transplantation, whole mounts of adult mouse retina were prepared for double immunocytochemistry using monoclonal antibodies against O4 (green) and monoclonal antibodies against β-tubulin III (red; Fig. 6A–6E). Quantification showed that the numbers of O4-positive cells per retina were significantly increased in the F3-treated group (n = 13, p = .04), but not the RA-treated (n = 12, p = .65) and MFS-treated (n = 11, p = .08) groups, compared with the control group (hBMSCs, n = 15), and there was no significant difference between F3- and MFS-treated groups (p = .84; Fig. 6F). Two months after transplantation of RA (Fig. 6C), MFS (Fig. 6D), and, in particular, F3 (Fig. 6E, 6G, 6H)-predifferentiated MGLCs, but not CD90(+) hBMSCs (Fig. 6B), O4(+) cells (green) sent out their processes, which landed onto the surface of retinal ganglion cell (RGC) axons (red). This indicates that axoglial interaction, an important event during early stages of myelination [1], occurs in the contacts between RGC axons and the processes of the F3-predifferentiated MGLCs. Moreover, electron microscopy (EM) was performed in cross-sections of retina to examine the ultrastructure of myelin sheaths. In the sham-injected adult mouse retina, there were unmyelinated axons (Fig. 6I). Interestingly, the F3-predifferentiated MGLCs could form single myelin sheaths that wrapped the RGC axons in the retina 2 months after transplantation (Fig. 6J). These observations support the suggestion that F3/Notch signaling is indeed involved in the promotion of functional maturation of the MGLCs.

View larger version (68K): [in this window] [in a new window]   Figure 6. Possible role of predifferentiated myelinating glia-like cells from hBMSCs in axon-glial interaction in mouse retina. Shown are sham injection (A) and transplantation of CD90(+) hBMSCs (B) and RA (C), MFS (D), and F3 (E)-predifferentiated myelinating glia-like cells derived from hBMSCs for 2 months. Whole mounts of adult mouse retina were prepared for double immunocytochemistry, using monoclonal antibody against O4 (green) (A–E, G, H) and monoclonal antibody against β-tubulin III (red) (A–E, G, H). The numbers of O4-positive cells per retina were counted. The numbers of O4-positive cells per retina were as follows: (a) 51.6 ± 13.18, n = 15 for the hBMSC control group; (b) 68.25 ± 11.72, n = 12, p = .65 for the RA-treated group; (c) 95.55 ± 18.66, n = 11, p = .08 for the MFS-treated group; and (d) 99.46 ± 22.41, n = 13, p < .05 for the F3-treated group. No significant difference was found between the F3- and MFS-treated groups (p = .84) (F). The processes were sent out from the F3-predifferentiated myelinating glia-like cells and landed onto the surface of axons (arrows) (G, H). Scale bars = 20 µm (E) (applies to [A–E]) and 10 µm (H) (applies to [G, H]). Electron microscopy studies in cross sections revealed the ultrastructure of myelin sheaths in adult mouse retina of sham-injected (I) versus transplanted mouse retina (J). Single myelin sheaths were formed in the transplanted mouse retina (arrows) (J). Scale bars = 0.5 µm (I) and 0.3 µm (J). *, p < .05. Error bars represent SEM. Abbreviations: hBMSC, human bone marrow stromal cell; MFS, multiple factors; RA, all-trans-retinoic acid.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion Disclosure of Potential... Acknowledgments References

CD90(+) Cells Derived from hBMSCs Have the Potential to Differentiate as MGLCs
Bone marrow is a readily available source of multipotent stem cells that may represent an alternative to neural or embryonic cells for cell replacement therapies for neurodegenerative diseases. Purified CD90(+)/CD44(+) cells from primary neurospheres have been reported to be able to generate neurospheres again and to differentiate into neurons and astrocytes [20]. Neural stem and progenitor cells isolated from several regions of the brain from mice, rats, and humans are CD90(+)/CD164(+)/CD34(–) cells that resemble a nonhematopoietic stem cell population and have the ability to differentiate along an oligodendrocyte lineage [25]. Pluripotent stem cells isolated from adult muscle have recently been reported to express CD90 and to be able to overcome germ lineage restrictions to express the molecular characteristics of OLs [36]. Moreover, CD90(+) cells derived from mesenchymal stem cells (MSCs) have been reported to express neuroglial transcripts even before any treatment [26]. In particular, these cells acquire astrocytic and neuron-like morphologies under specific culture conditions, suggesting that adult bone marrow may have the capacity for neural differentiation [27]. KP-hMSCs, an immortalized line of hBMSCs, express CD90 and proteins of the neuronal, astrocyte, and oligodendrocyte lineages during expansion in culture [37]. In agreement with these the previous observations, we have further confirmed that the CD90(+) cells isolated from hBMSCs are mainly mesenchymal stem cells or progenitor cells that express both vimentin and BMPR-1A in vitro and can be differentiated into cells with myelinating glia-like cell characteristics both in vitro and in vivo.

Possible Role of F3/Notch Signaling in Predifferentiation of MGLCs from hBMSCs
In the CNS, Notch signaling plays both inhibitory and instructive roles in cell fate selection depending upon the temporal and spatial context [38]. Delta-Notch signaling has been shown to specify glial rather than neuronal fate and also to direct a ventral population of precursors to OPCs rather than motoneurons [39]. In the rat optic nerve, the Jagged1-Notch interaction inhibits further differentiation of OPCs [40]. Our recent work has identified F3/contactin and NB-3, two neural cell adhesion molecules, as functional ligands of Notch [5, 6]. F3/Notch signaling increases oligodendrocyte generation from progenitor cells; in addition, F3 and NB-3/Notch cascades promote OPC commitment to young OLs. Remarkably, these events all required Deltex1 as the intermediate element. Here, we have shown that the CD90(+) cells derived from hBMSCs express both Notch1 and Deltex1. Similar to OLN cells and NPCs [5, 6], F3/contactin triggers Notch/Deltex1 signaling in the CD90(+) cells and induces them to differentiate into MGLCs with a high degree of morphological and functional maturation, as evidenced by increment of the number of processes of the MGLCs and the formation of axoglial contacts and single myelin sheaths in retina. It remains to be determined whether the formation of only single myelin sheaths is a consequence of only partial functional differentiation or a consequence of the fact that the retina is an environment that does not normally support myelination [41].

This study focused on the use of progenitor cells from bone marrow to generate cells belonging to the myelinating glial lineage to be used in transplantation experiments for reparative purposes. In this attempt, the effects of F3/contactin were studied, whereas a mixture of multiple factors, including bFGF, forskolin, PDGF, and HRG, was used as a control. bFGF, PDGF, and HRG are involved in the differentiation and proliferation of glial and Schwann cells [17, 42, 43]. Forskolin increases intracellular cAMP levels and mitogenic responses through upregulation of growth factor receptors [17, 44]. However, the mechanism underlining the role of these multiple factors on hBMSC differentiation remains unknown. To develop a simple and reliable means of selectively differentiating hBMSCs to myelinating glial lineages in both in vitro and in vivo conditions, we have taken advantage of our previous observation that F3/contactin may drive OPCs differentiation in in vitro models and that these effects rely upon activation of the Notch/Deltex1 pathway [5, 6]. Based on these data, we have further explored the possibility that F3/contactin drives commitment of hBMSCs along the myelinating glial lineage through activation of Notch/Deltex1 signaling in both in vitro and in vivo conditions. In particular, these data demonstrate that the predifferentiated MGLCs from hBMSCs may be useful for reparative purposes in central nervous tissue. As F3/contactin is a single factor, it may prove more amenable than multiple factors for the development of contaminant-free differentiation protocols for therapeutic application.

The Retina Is a Useful Transplant Model to Test the Function of Predifferentiated MGLCs
In this study, our primary goal was to test whether the predifferentiated hBMSCs induced by F3/contactin could form axoglial junctions and myelin sheaths by using the retinal transplant model. During development, OPCs migrate from the chiasmal end into the optic nerve [30, 45–47]. Progenitor cells stop migrating shortly before the optic nerve enters the retina. OLs are absent from the retina and the most proximal part of the optic nerve, and segments of retinal ganglion cell axons located in these regions are unmyelinated. Thus, the rodent retina offers several advantages for transplantation experiments with myelinating glial cells [35]. Notably, myelination of the intraretinal axon segments occurs under a variety of pathological conditions [48, 49] or after transplantation of OPCs into the rat retina [34] or neural precursor cells into the retina of young postnatal mice [35]. In agreement with these previous observations, our immunofluorescence and EM results demonstrate that these predifferentiated hBMSCs induced by F3/contactin, which we have called MGLCs, have the potential to form axoglial junctions and myelin sheaths (Fig. 6), although no compacted myelin sheaths were observed.

	DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST

Top Abstract Introduction Materials and Methods Results Discussion Disclosure of Potential... Acknowledgments References

 
The authors indicate no potential conflicts of interest.

	ACKNOWLEDGMENTS

Top Abstract Introduction Materials and Methods Results Discussion Disclosure of Potential... Acknowledgments References

 
We thank Lim Che Kang for technical assistance. This work was supported by grants to Z.-C.X. from Singapore Health Services Pte. Ltd., Department of Clinical Research, Singapore General Hospital, Institute of Molecular and Cell Biology, A*STAR, Singapore; and a grant to Z.-C.X., G.S.D., and S.A. from the A*STAR-Juvenile Diabetes Research Foundation joint program, Singapore.

	REFERENCES

Top Abstract Introduction Materials and Methods Results Discussion Disclosure of Potential... Acknowledgments References

 

1. Ma QH, Nie DY, Dawe GS et al. Physiological roles of neurite outgrowth inhibitors at myelinated axons in the central nervous system—Implications for the therapeutic neutralization of neurite outgrowth inhibitors. Curr Pharm Des 2007;13:2529–2537.[CrossRef][Medline]

2. Sanchez-Ramos JR. Neural cells derived from adult bone marrow and umbilical cord blood. J Neurosci Res 2002;69:880–893.[CrossRef][Medline]

3. Akiyama Y, Radtke C, Kocsis JD. Remyelination of the rat spinal cord by transplantation of identified bone marrow stromal cells. J Neurosci 2002;22:6623–6630.[Abstract/Free Full Text]

4. Hermann A, Gastl R, Liebau S et al. Efficient generation of neural stem cell-like cells from adult human bone marrow stromal cells. J Cell Sci 2004;117:4411–4422.[Abstract/Free Full Text]

5. Hu QD, Ang BT, Karsak M et al. F3/contactin acts as a functional ligand for Notch during oligodendrocyte maturation. Cell 2003;115:163–175.[CrossRef][Medline]

6. Cui XY, Hu QD, Tekaya M et al. NB-3/Notch1 pathway via Deltex1 promotes neural progenitor cell differentiation into oligodendrocytes. J Biol Chem 2004;279:25858–25865.[Abstract/Free Full Text]

7. Hu QD, Ma QH, Gennarini G et al. Cross-talk between F3/contactin and Notch at axoglial interface: A role in oligodendrocyte development. Dev Neurosci 2006;28:25–33.[CrossRef][Medline]

8. Ebinu JO, Yankner BA. A RIP tide in neuronal signal transduction. Neuron 2002;34:499–502.[CrossRef][Medline]

9. Martinez Arias A, Zecchini V, Brennan K. CSL-independent Notch signalling: A checkpoint in cell fate decisions during development? Curr Opin Genet Dev 2002;12:524–533.[CrossRef][Medline]

10. Brennan K, Gardner P. Notching up another pathway. Bioessays 2002;24:405–410.[CrossRef][Medline]

11. Revest JM, Faivre-Sarrailh C, Maeda N et al. The interaction between F3 immunoglobulin domains and protein tyrosine phosphatases zeta/beta triggers bidirectional signalling between neurons and glial cells. Eur J Neurosci 1999;11:1134–1147.[CrossRef][Medline]

12. Xiao ZC, Taylor J, Montag D et al. Distinct effects of recombinant tenascin-R domains in neuronal cell functions and identification of the domain interacting with the neuronal recognition molecule F3/11. Eur J Neurosci 1996;8:766–782.[CrossRef][Medline]

13. Zeng L, D'Alessandri L, Kalousek MB et al. Protein tyrosine phosphatase alpha (PTPalpha) and contactin form a novel neuronal receptor complex linked to the intracellular tyrosine kinase fyn. J Cell Biol 1999;147:707–714.[Abstract/Free Full Text]

14. Umemori H, Sato S, Yagi T et al. Initial events of myelination involve Fyn tyrosine kinase signalling. Nature 1994;367:572–576.[CrossRef][Medline]

15. Girault JA, Peles E. Development of nodes of Ranvier. Curr Opin Neurobiol 2002;12:476–485.[CrossRef][Medline]

16. Berglund EO, Murai KK, Fredette B. Ataxia and abnormal cerebellar microorganization in mice with ablated contactin gene expression. Neuron 1999;24:739–750.[CrossRef][Medline]

17. Dezawa M, Takahashi I, Esaki M et al. Sciatic nerve regeneration in rats induced by transplantation of in vitro differentiated bone-marrow stromal cells. Eur J Neurosci 2001;14:1771–1776.[CrossRef][Medline]

18. Vogel W, Grunebach F, Messam CA et al. Heterogeneity among human bone marrow-derived mesenchymal stem cells and neural progenitor cells. Haematologica 2003;88:126–133.[Abstract/Free Full Text]

19. Vodyanik MA, Bork JA, Thomson JA et al. Human embryonic stem cell-derived CD34+ cells: Efficient production in the coculture with OP9 stromal cells and analysis of lymphohematopoietic potential. Blood 2005;105:617–626.[Abstract/Free Full Text]

20. Oishi K, Ito-Dufros Y. Angiogenic potential of CD44+ CD90+ multipotent CNS stem cells in vitro. Biochem Biophys Res Commun 2006;349:1065–1072.[CrossRef][Medline]

21. Chen X, Xu H, Wan C et al. Bioreactor expansion of human adult bone marrow-derived mesenchymal stem cells. STEM CELLS 2006;24:2052–2059.[Abstract/Free Full Text]

22. Keilhoff G, Stang F, Goihl A et al. Transdifferentiated mesenchymal stem cells as alternative therapy in supporting nerve regeneration and myelination. Cell Mol Neurobiol 2006;26:1235–1252.[Medline]

23. Graham V, Khudyakov J, Ellis P et al. SOX2 functions to maintain neural progenitor identity. Neuron 2003;39:749–765.[CrossRef][Medline]

24. Wislet-Gendebien S, Leprince P, Moonen G et al. Regulation of neural markers nestin and GFAP expression by cultivated bone marrow stromal cells. J Cell Sci 2003;116:3295–3302.[Abstract/Free Full Text]

25. Davis SF, Hood J, Thomas A et al. Isolation of adult rhesus neural stem and progenitor cells and differentiation into immature oligodendrocytes. Stem Cells Dev 2006;15:191–199.[CrossRef][Medline]

26. Bossolasco P, Cova L, Calzarossa C et al. Neuro-glial differentiation of human bone marrow stem cells in vitro. Exp Neurol 2005;193:312–325.[CrossRef][Medline]

27. Izon DJ, Aster JC, He Y et al. Deltex1 redirects lymphoid progenitors to the B cell lineage by antagonizing Notch1. Immunity 2002;16:231–243.[CrossRef][Medline]

28. Matsuno K, Ito M, Hori K et al. Involvement of a proline-rich motif and RING-H2 finger of Deltex in the regulation of Notch signalling. Development 2002;129:1049–1059.[Medline]

29. Bartsch U, Kirchhoff F, Schachner M. Immunohistological localization of the adhesion molecules L1, N-CAM, and MAG in the developing and adult optic nerve of mice. J Comp Neurol 1989;284:451–462.[CrossRef][Medline]

30. Bartsch U, Faissner A, Trotter J et al. Tenascin demarcates the boundary between the myelinated and nonmyelinated part of retinal ganglion cell axons in the developing and adult mouse. J Neurosci 1994;14:4756–4768.[Abstract]

31. Ffrench-Constant C, Miller RH, Burne JF et al. Evidence that migratory oligodendrocyte-type-2 astrocyte (O-2A) progenitor cells are kept out of the rat retina by a barrier at the eye-end of the optic nerve. J Neurocytol 1988;17:13–25.[CrossRef][Medline]

32. Hildebrand C, Remahl S, Waxman SG. Axo-glial relations in the retina-optic nerve junction of the adult rat: Electronmicroscopic observations. J Neurocytol 1985;14:597–617.[CrossRef][Medline]

33. Perry VH, Lund RD. Evidence that the lamina cribrosa prevents intraretinal myelination of retinal ganglion cell axons. J Neurocytol 1990;19:265–272.[CrossRef][Medline]

34. Laeng P, Molthagen M, Gui-Xia Yu et al. Transplantation of oligodendrocyte precursor cells into the rat retina: Extensive myelination of retinal ganglion cell axons. Glia 1996;18:200–210.[CrossRef][Medline]

35. Ader M, Meng J, Schachner M et al. Formation of myelin after transplantation of neural precursor cells into the retina of young postnatal mice. Glia 2000;30:301–310.[CrossRef][Medline]

36. Romero-Ramos M, Vourc'h P, Young HE et al. Neuronal differentiation of stem cells isolated from adult muscle. J Neurosci Res 2002;69:894–907.[CrossRef][Medline]

37. Chu MS, Chang CF, Yang CC et al. Signalling pathway in the induction of neurite outgrowth in human mesenchymal stem cells. Cell Signal 2006;18:519–530.[CrossRef][Medline]

38. Gaiano N, Fishell G. The role of notch in promoting glial and neural stem cell fates. Annu Rev Neurosci 2002;25:471–490.[CrossRef][Medline]

39. Park HC, Appel B. Delta-Notch signaling regulates oligodendrocyte specification. Development 2003;130:3747–3755.[Abstract/Free Full Text]

40. Wang S, Sdrulla AD, diSibio G et al. Notch receptor activation inhibits oligodendrocyte differentiation. Neuron 1998;21:63–75.[Medline]

41. Gao L, Macklin W, Gerson J et al. Intrinsic and extrinsic inhibition of oligodendrocyte development by rat retina. Dev Biol 2006;290:277–286.[CrossRef][Medline]

42. Cohen RI, Mckay R, Almazan G. Cyclic AMP regulates PDGF-stimulated signal transduction and differentiation of an immortalized optic-nerve-derived cell line. J Exp Biol 1999;202:461–473.[Abstract]

43. Jessen KR, Mirsky R. Developmental regulation in the Schwann cell lineage. Adv Exp Med Biol 1999;468:3–12.[Medline]

44. Kim HA, Ratner N, Roberts N et al. Schwann cell proliferative responses to cAMP and Nf1 are mediated by cyclin D1. J Neurosci 2001;21:1110–1116.[Abstract/Free Full Text]

45. Fulton BP, Burne JF, Raff MC. Visualization of O-2A progenitor cells in developing and adult rat optic nerve by quisqualate-stimulated cobalt uptake. J Neurosci 1992;12:4816–4833.[Abstract]

46. Mudhar HS, Pollock RA, Wang C et al. PDGF and its receptors in the developing rodent retina and optic nerve. Development 1993;118:539–552.[Abstract]

47. Small RK, Riddle P, Noble M. Evidence for migration of oligodendrocyte-type-2 astrocyte progenitor cells into the developing rat optic nerve. Nature 1987;328:155–157.[CrossRef][Medline]

48. Wyse JPH. Schwann cell myelination in the nerve fibre layer of the BW rat retina. J Neurocytol 1980;9:107–117.[CrossRef][Medline]

49. Perry VH, Hayes L. Lesion-induced myelin formation in the retina. J Neurocytol 1985;14:297–307.[CrossRef][Medline]

This article has been cited by other articles:

				L.-F. Chen, Z. Q. Yin, S. Chen, and Z.-S. Chen Differentiation and Production of Action Potentials by Embryonic Rat Retina Stem Cells In Vitro Invest. Ophthalmol. Vis. Sci., November 1, 2008; 49(11): 5144 - 5150. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 2007-0106v1 26/2/580    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lu, L. Articles by Xiao, Z.-C. Search for Related Content PubMed PubMed Citation Articles by Lu, L. Articles by Xiao, Z.-C.