Basic Research—Biology

Isolation and Identification of CXCR4-positive Cells fromHuman Dental Pulp CellsLong Jiang, DDS, MDS, Wei-Wei Peng, DDS, MDS, Li-Fen Li, DDS, MDS,Ya Yang, DDS, PhD, and Ya-Qin Zhu, DDS, PhD

Abstract

Introduction: In previous studies, we found expres-sion of stromal cell–derived factor-1a (SDF-1a)/CXCchemokine receptor 4 (CXCR4) in human dental pulpand the SDF-1a–CXCR4 axis might play a role inthe recruitment of CXCR4-positive dental pulp cells(CXCR4+ DPCs) toward the damaged sites. However,the specific function of CXCR4+ DPCs in the injureddental pulp was still unknown. The purpose of thisstudy was to isolate CXCR4+ DPCs from dental pulpcells in vitro to pave the way for further study oftheir characteristics.Methods: CXCR4+ DPCs were iso-lated with magnetic-activated cell sorting (MACS).Freshly isolated CXCR4+ DPCs were identified by immu-nohistochemistry with light microscopy or confocalmicroscopy. Then the phenotypes CXCR4, stromal cellsurface marker-1 (STRO-1), CD146, and CD34 in 3groups (ie, CXCR4+ DPCs, CXCR4� DPCs, or non-sorted DPCs) were analyzed by flow cytometry afterthey were cultured and expanded in vitro. Results:The results indicated the isolated subpopulation ofDPCs was enriched with CXCR4+ DPCs, and the posi-tive rates of STRO-1 and CD146 in CXCR4+ DPCs groupwere higher than CXCR4� DPCs or non-sorted DPCsgroups (P < .05). There was no expression of CD34in each group. Conclusions: We can isolate CXCR4+

DPCs from DPCs with MACS and identify them by immu-nohistochemistry and flow cytometry. (J Endod2012;38:791–795)

Key WordsCXCR4, dental pulp stem cells, magnetic-activated cellsorting

From the Department of General Dentistry, 9th People’s HospiShanghai, PR China.

Supported by grants from the Science and Technology Commissidation for the Returned Overseas Chinese Scholars, State Educationsion (grant no. 09ZZ116), and the selection and cultivation for outs

Address requests for reprints to Dr Ya-Qin Zhu, Department of GeKey Laboratory of Stomatology, Shanghai 200011, PR China. E-mai0099-2399/$ - see front matter

Copyright ª 2012 American Association of Endodontists.doi:10.1016/j.joen.2012.02.024

JOE — Volume 38, Number 6, June 2012

It has been confirmed that there are stem cells in human dental pulp tissue (1–3). Thedental pulp stem cells (DPSCs), which usually remain quiescent, are thought to

respond during injury (4). After damage of the tooth, these DPSCs will be activatedto migrate to the injury sites and differentiate into odontoblast-like cells, which synthe-size and excrete an extracellular matrix to form the reparative dentin (5, 6).Furthermore, the DPSCs are believed to be promising clinical treatments for dentaldisease in tissue engineering because of the potential capability to repair or evenregenerate a new tooth to replace the diseased tooth.

Many attempts have been made to isolate stem/progenitor cells from dental pulpcells (DPCs). The stromal cell surface marker-1 positive (STRO-1+) cells, with highcolony-forming efficiency, have been isolated from dental pulp (7). Furthermore,STRO-1+ cells from rat dental pulp cells by fluorescence-activated cell sorting wereable to differentiate into the odontogenic pathway, whereas the negative fraction inthe separation failed to differentiate (8). It has been reported that a side cell populationwith characteristic features of stem cell, which cannot be stained by the DNA-binding dyeHoechst 33342, were isolated from dental pulp (9). Recently, the cells expressing highlevel of b1 integrin or expressing the embryonic neural crest cell mark, low-affinitynerve growth factor receptor (P75), were selected by magnetic-activated cell sorting(MACS) (10). Those studies showed that the stem/progenitor cells isolated from dentalpulp have high proliferation andmultilineage differentiation potential. In addition, it hasbeen demonstrated that the dentin/pulp-like complex is formed by them in vivo.However, the mechanisms underlying the recruitment of stem/progenitor cells towardthe injury site remain elusive.

In previous studies, we found the expression of stromal cell–derived factor-1a(SDF-1a) and CXC chemokine receptor 4 (CXCR4) in dental pulp. The SDF-1a–CXCR4 axis was believed to play a role in the recruitment of CXCR4-positive dentalpulp cells (CXCR4+ DPCs) toward the damaged sites (11). Generally, CXCR4 wasconsidered as a surface marker expressed in stem cells (12). Besides hematopoieticstem cells, CXCR4 has also been expressed in other tissue-specific stem cells such asskeletal muscle satellite cells, which are responsible for the migration of satellite cellstoward an SDF-1 gradient (13). Consequently, we hypothesized that CXCR4 might beone of the surface markers of DPSCs, and CXCR4+ DPCs might be a homogeneousstem/progenitor cell population. To verify this hypothesis, we must isolate CXCR4+

DPCs and analyze the characteristics of CXCR4+ DPCs, CXCR4� DPCs, and non-sorted DPCs in vitro. In this study, we successfully isolated the CXCR4+ DPCs byMACS and identified them. The findings of this study are conducive with the systematiccharacterization of CXCR4+ DPCs.

tal, Shanghai JiaoTong University, School of Medicine, Shanghai Key Laboratory of Stomatology,

on of Shanghai Municipality (grant no. 08DZ2271100, 08JC1414500), the Scientific Research Foun-Ministry (grant no. 2008-0890-09), Innovation Program of Shanghai Municipal Education Commis-tanding young teachers in scientific special fund of Shanghai university (grant no. JDY09051).neral Dentistry, 9th People’s Hospital, Shanghai JiaoTong University, School of Medicine, Shanghail address: zyq1590@163.com

Isolation of CXCR4-positive Cells from Human DPCs 791

Basic Research—Biology

Materials and MethodsCell Cultures

Dental pulp tissue was removed from extracted third or premolarsfor orthodontic treatments of patients between 18 and 29 years old. Aninformed consent was obtained from each patient before the extrac-tions. The 9th People’s Hospital Ethics Committee approved the proce-dure and the research protocol for tooth extractions. Primary humanDPCs were cultured as described previously (11). Briefly, the pulptissue was cut into small pieces, and then these tissue fragments werecovered by plastic coverslips. The tissue fragments were maintainedin medium in 37�C incubator with a humidified atmosphere of 5%CO2 in air. The culture medium was Dulbecco modified Eagle medium(DMEM) (Gibco, Grand Island, NY) supplemented with 2 mmol/Lglutamine (Sigma, St Louis, MO), 100 IU/mL penicillin, 100 mg/mLstreptomycin (Gibco), 0.25 g/mL amphotericin B (Fungizone; Gibco),and 20% fetal bovine serum (FBS) (Gibco). The culture medium waschanged at 5-day intervals. Confluent cultures were dissociated by tryp-sinization (0.2% trypsin and 0.02% ethylenediaminetetraacetic acid;Gibco) and subcultured with DMEM plus 10% FBS.

MACSThe procedure was a modification of the manufacturer’s manual.

Briefly, the single-cell suspension of the third passage DPCs (1� 107/mL) was incubated with 5 mg of biotin-labeled anti-human CXCR4 anti-body (eBioscience, San Diego, CA) for 45minutes at room temperature.The cells were washed by using 2 mL phosphate-buffered saline (PBS)with 5% FBS and centrifuged at 1000 rpm for 5minutes. Then the super-natant containing unbound primary antibody was completely removed.The 90 mL of single-cell suspension was resuspended in 10 mL of strep-tavidin linkedmicro-Beads (Miltenyi Biotec, Inc, Auburn, CA) and incu-bated for 45 minutes on ice. Then the DPCs were sorted by using a miniMACS magnetic column (Miltenyi Biotec, Inc) in the magnetic field ac-cording to the protocol. CXCR4+ cells were collected and seeded onto6-well plates in DMEM supplemented with 20% FBS. Once cells werecultured to 80% confluency, they were detached by trypsinization andsubcultured at 1:4 dilution for more than 10 passages. CXCR4� cellsand non-sorted DPCs were also collected and cultured under thesame conditions and used as controls for the following experiments.

ImmunohistochemistryThe 3 types of cell populations were seeded immediately after

being sorted onto the coverslips in the 6-well plates respectively.Then they were cultured in DMEM supplemented with 20% FBS. Thecells were fixed with 4% paraformaldahyde for 20 minutes when theywere 60% confluent. The primary antibody was mouse anti-CXCR4 (im-munglobulin [Ig] G, 10 mg/mL; R&D, Minneapolis, MN) or mouse anti-STRO-1 (IgM, 1:25; Santa Cruz Biotechnology, Inc, Santa Cruz, CA). Theincubation was performed overnight at 4�C. The immunostaining wasrevealed by using the EnVision Detection kit (Dako, Copenhagen,Demark) according to the manufacturer’s instructions. PBS replacedprimary antibodies that were used to serve as the negative controls.Five fields under 100-fold magnification were randomly selected, andimmunoreactive cells were counted. The positive rate was the numberof stained cells/the total number of cells � 100%.

CXCR4+ cells grown on the coverslips were fixed and stained withmouse anti-CXCR4 IgG (10 mg/mL) and mouse anti-STRO-1 IgM(1:25), followed by goat anti-mouse IgG-PE (1:200; Santa Cruz Biotech-nology, Inc) and goat anti-mouse IgM-FITC (1:200; Santa Cruz Biotech-nology, Inc). After washing 3 times with PBS, the nuclei were stainedwith 40 ng/mL Hoechst 33342 (Invitrogen, Carlsbad, CA) for 10minutes. Slides were rinsed and mounted by glycerin. Images were ob-

792 Jiang et al.

tained by confocal microscope (TCS SP2; Leica, Mannheim, Germany).In control groups, PBS replaced primary antibodies.

Flow CytometryWhen the sorted cells had been expanded, the single-cell suspen-

sions of the 3 cells were adjusted to 1 � 105 cells in 100 mL andincubated with mouse anti-CXCR4 IgG (10 mg/mL) or 4 mL of mouseanti-STRO-1 IgM for 30 minutes at 4�C. The secondary antibody wasgoat anti-mouse IgG-FITC (1:200; Santa Cruz Biotechnology, Inc) orgoat anti-mouse IgM-FITC (1:200). After incubating according toprotocol, those cells were resuspended in PBS. Analysis of cells was per-formed by using a flow cytometer (FACSCalibur; BD, Providence, RI).Mouse anti-human CD146:FITC (1:10; ABD Serotec, Oxford, UK) andmouse anti-human CD34:FITC (1:20; eBioscience) were also used totest. All DPCs were from 25 patients’ teeth.

Statistical AnalysisData analysis was performed by using SAS6.12 (SAS Institute Inc,

Raleigh, NC). Statistical significance between groups was determined byusing one-way analysis of variance. A P value <.05 was consideredsignificant.

ResultsImmunohistochemistry

All isolated cells had a typical fibroblast-like morphology, withoutapparent difference observed among CXCR4+ DPCS, CXCR4�DPCs, andnon-sorting DPCs by inverted phase-contrast microscopy. A significantlygreater percentage of immunoreactive CXCR4+ was detected in thesorted subpopulation of cells (P < .05). Many more positively stainedcells were present in CXCR4+ DPCs group (Fig. 1A and D). For CXCR4�

cells group, positively stained cells were almost absent (Fig. 1B and E).The average positive rate (n = 5) of CXCR4 and STRO-1 in CXCR4+ DPCsgroup was approximately 83% and 89%, respectively.

The results of immunofluorescence observed by confocal micros-copy showed CXCR4 and STRO-1 were coexpressed by CXCR4+ DPCs(Fig. 1G–I).

Flow CytometryThe flow cytometric analyses of the expanded CXCR4+ DPCs,

CXCR4� DPCs, and non-sorted DPCs showed differences among the3 cell groups (Fig. 2). The average positive rates (n = 25) of CXCR4were 7.32%, 2.52%, and 4.43%, respectively. The expression ofSTRO-1 in CXCR4+ DPCs, CXCR4� DPCs, and non-sorted DPCs wasapproximately 14%, 3%, and 5%, respectively. The expression ofCD146 in CXCR4+ DPCs, CXCR4� DPCs, and non-sorted DPCs groupswas 57%, 34%, and 47%, respectively. The positive rate of CXCR4,STRO-1, or CX146 in CXCR4+ group was higher than in CXCR4� groupor non-sorted group (P < .05). However, the expressions of CD34 inthe 3 cell groups were negative, and there were no differences in the3 groups (Fig. 2J–L).

DiscussionCXCR4 is the receptor of SDF-1 (14). Compelling evidence is accu-

mulating that stem cells of different organs and tissues resemble hema-topoietic stem cells (HSCs) expressing functional CXCR4 on theirsurface (15, 16). In damaged organs, SDF-1 is up-regulated to attractCXCR4+ stem cells that are mobilized from their niches in response tothe stimulation related to tissue/organ damage (17). Our previous workhas implicated CXCR4 expression in some DPCs. Furthermore, theimmunohistochemical results showed that CXCR4+ DPCs were localizedto the perivascular area in physiological conditions, and they could be

JOE — Volume 38, Number 6, June 2012

Figure 1. Immunostaining of cultured CXCR4+ DPCs, CXCR4� DPCs, and non-sorted DPCs. (A) CXCR4+ DPCs were stained with anti-CXCR4 monoclonal antibody(MoAb). (B) CXCR4� DPCs were stained with anti-CXCR4 MoAb. No staining was observed. (C) Non-sorted DPCs were stained with anti-CXCR4 MoAb. A few DPCswere positive. (D) CXCR4+ DPCs were stained with anti-STRO-1 MoAb. (E) CXCR4� DPCs were stained with anti-STRO-1 MoAb, No staining was observed. (F) Non-sorted DPCs were stained with anti-STRO-1 MoAb. A few DPCs were positive. (A–F) Original magnification, 10�. (G) Red fluorescence presented CXCR4 was ex-pressed by CXCR4+ DPCs. (H) Green fluorescence presented STRO-1 was expressed by CXCR4+ DPCs. (I) Was compounded by (G) and (H). (J–L) Negative control.

Basic Research—Biology

recruited toward the damaged site along an SDF-1 gradient (11).Another study also demonstrated inflammation and hypoxia mightrecruit DPSCs to participate in reparative dentinogenesis with regulatingthe SDF-1a–CXCR4 axis (18). However, the exact function of CXCR4+

DPCs in the area of injury remains unknown. We also do not knowwhether CXCR4+ DPCs have properties of DPSCs.

In general, mesenchymal stem cells have the capacity of self-renewal and multilineage differentiation (19). DPSCs have been furthercharacterized by a single colony isolation method, demonstrating self-renewal capability, multipotent differentiation, and differentiation into

JOE — Volume 38, Number 6, June 2012

odontoblasts forming tubular dentin in vivo after transplantationwith hydroxyapatite/tricalcium phosphate into immunocompromisedmice (20). To investigate the characteristics of CXCR4+ cells, thesuccessful isolation of them from DPCs is a critical step.

At present, stem cells can be identified and isolated from a varietyof cells by using 4 methods:

1. Fluorescence-labeled antibody to recognize and a flow cytometer tosort the cells with specific surface antigen markers in a processcalled fluorescent antibody cell sorting (FACS)

Isolation of CXCR4-positive Cells from Human DPCs 793

Figure 2. Immunophenotype analysis of CXCR4+ DPCs, CXCR4� DPCs, and non-sorted DPCs by flow cytometry. CXCR4+ cells were labeled with antibody againstCXCR4 (A), STRO-1 (D), CD146 (G), or CD34 (J), respectively. CXCR4� cells were labeled with antibody against CXCR4 (B), STRO-1 (E), CD146 (H), or CD34 (K),respectively. Non-sorted cells were labeled with antibody against CXCR4 (C), STRO-1 (F), CD146 (I), or CD34 (L), respectively. Blue area represents controlimmunoglobulin.

Basic Research—Biology

794 Jiang et al. JOE — Volume 38, Number 6, June 2012

Basic Research—Biology

2. MACS3. Immunohistochemical staining4. Physiological and histologic criteria, including phenotype (appear-

ance), chemotaxis, proliferation, differentiation, and mineralizingactivity (21)

In this study, we isolated CXCR4+ DPCs by MACS and examined theefficacy of MACS. Three groups of cells obtained were also assessed forCXCR4+ cell content by immunohistochemistry. The results demon-strated significant difference in the 3 types of cells. After culture expand-ing in vitro, the flow cytometric analyses of CXCR4+ DPCs showed thepositive rate of CXCR4 was 7%–8%. Although it was 2%–3% of CXCR4�

DPCs and 4%–5% of non-sorted DPCs, it was obviously discrepant withthe results of immunohistochemistry, which showed approximately80%–90% of cells were CXCR4+ in light microscope. The very lowpercentage of CXCR4+ cells of the flow cytometric analyses could bedue to the long time of culture in normal DMEM after selection was per-formed. The cells applied to immunohistochemistry were cultured foronly 2 or 3 days after being sorted. However, the cells that were analyzedby flow cytometry must be cultured longer than 2 weeks after beingsorted, so that the number of cells can suffice for the requirement ofthe experiment. Further studies are needed for maintaining the stableexpression of CXCR4 in vitro.

The expression of STRO-1 as an important marker of DPSCs andbone marrow stromal cells (22–25) was similar to CXCR4 in the 3 typesof cells. The STRO-1+ fraction in CXCR4+ DPCs was approximately 14%and more than 3% in CXCR4� DPCs or 5% in non-sorted DPCs. Earlierexperiments showed that DPSCs isolated from the pulp of human exfo-liated deciduous teeth by a clonogenic method exhibited 9% positivityfor STRO-1 (2). This implied we could obtain much more positivityfor the STRO-1 cells from CXCR4+ DPCs population by MACS. CD146is also considered as a marker for DPSCs, although the exact functionof CD146 is not known (7). CD34 is expressed in hematopoietic stemcells (26). Most studies demonstrated that DPSCs did not express CD34(1, 22, 27). In our study, the results indicated that CXCR4+ DPCscomprised the most CD146+ DPCs, and CD34 was almost notexpressed in the 3 groups.

In conclusion, the results presented here demonstrated thatCXCR4+ DPCs can be isolated from DPCs by MACS, and STRO-1 is coex-pressed by CXCR4+ DPCs, although the positive rates would bedecreased after some time of culture in vitro. Further studies will focuson the detection of proliferation and multilineage differentiationcapacity of CXCR4+ DPCs.

AcknowledgmentsThe authors thank Dong-xia Ye, Xiu-li Zhang, and Han-bing Fu

for their excellent technical support.The authors deny any conflicts of interest related to this study.

References1. Gronthos S, Mankani M, Brahim J, et al. Postnatal human dental pulp stem cells

(DPSCs) in vitro and in vivo. Proc Natl Acad Sci U S A 2000;97:13625–30.

JOE — Volume 38, Number 6, June 2012

2. Miura M, Gronthos S, Zhao M, et al. SHED: stem cells from human exfoliated decid-uous teeth. Proc Natl Acad Sci U S A 2003;100:5807–12.

3. Pivoriu�unas A, Surovas A, Borutinskaite V, et al. Proteomic analysis of stromal cellsderived from the dental pulp of human exfoliated deciduous teeth. Stem Cells Dev2010;19:1081–93.

4. Scadden DT. The stem-cell niche as an entity of action. Nature 2006;441:1075–9.5. Smith AJ, Lesot H. Induction and regulation of crown dentinogenesis: embryonic

events as a template for dental tissue repair? Crit Rev Oral Biol Med 2001;12:425–37.

6. Goldberg M, Smith AJ. Cells and extracellular matrices of dentin and pulp: a bio-logical basis for repair and tissue engineering. Crit Rev Oral Biol Med 2004;15:13–27.

7. Shi S, Gronthos S. Perivascular niche of postnatal mesenchymal stem cells in humanbone marrow and dental pulp. J Bone Miner Res 2003;18:696–704.

8. Yang X, van den Dolder J, Walboomers XF, et al. The odontogenic potential ofSTRO-1 sorted rat dental pulp stem cells in vitro. J Tissue Eng Regen Med 2007;1:66–73.

9. Iohara K, Zheng L, Ito M, et al. Side population cells isolated from porcine dentalpulp tissue with self-renewal and multipotency for dentinogenesis, chondrogenesis,adipogenesis, and neurogenesis. Stem Cells 2006;24:2493–503.

10. Waddington RJ, Youde SJ, Lee CP, et al. Isolation of distinct progenitor stem cellpopulations from dental pulp. Cells Tissues Organs 2009;189:268–74.

11. Jiang L, Zhu YQ, Du R, et al. The expression and role of stromal cell-derived factor-1alpha-CXCR4 axis in human dental pulp. J Endod 2008;34:939–44.

12. Majka M, Kucia M, Ratajczak MZ. Stem cell biology: a never ending quest for under-standing. Acta Biochim Pol 2005;52:353–8.

13. Ratajczak MZ, Majka M, Kucia M, et al. Expression of functional CXCR4 by musclesatellite cells and secretion of SDF-1 by muscle-derived fibroblasts is associated withthe presence of both muscle progenitors in bone marrow and hematopoietic stem/progenitor cells in muscles. Stem Cells 2003;21:363–71.

14. Nagasawa T, Hirota S, Tachibana K, et al. Defects of B-cell lymphopoiesis and bone-marrow myelopoiesis in mice lacking the CXC chemokine PBSF/SDF-1. Nature 1996;382:635–8.

15. Ratajczak MZ, Kucia M, Reca R, et al. Stem cell plasticity revisited: CXCR4-positivecells expressing mRNA for early muscle, liver and neural cells ’hide out’ in the bonemarrow. Leukemia 2004;18:29–40.

16. Kucia M, Ratajczak J, Ratajczak MZ. Bone marrow as a source of circulating CXCR4+tissue-committed stem cells. Biol Cell 2005;97:133–46.

17. Imitola J, Raddassi K, Park KI, et al. Directed migration of neural stem cells to sitesof CNS injury by the stromal cell-derived factor 1alpha/CXC chemokine receptor 4pathway. Proc Natl Acad Sci U S A 2004;101:18117–22.

18. Gong QM, Quan JJ, Jiang HW, et al. Regulation of the stromal cell-derived factor-1alpha-CXCR4 axis in human dental pulp cells. J Endod 2010;36:1499–503.

19. Bakopoulou A, Leyhausen G, Volk J, et al. Comparative analysis of in vitro osteo/odontogenic differentiation potential of human dental pulp stem cells (DPSCs)and stem cells from the apical papilla (SCAP). Arch Oral Biol 2011;56:709–21.

20. Nakashima M, Iohara K, Sugiyama M. Human dental pulp stem cells with highlyangiogenic and neurogenic potential for possible use in pulp regeneration. CytokineGrowth Factor Rev 2009;20:435–40.

21. Murray PE, Garcia-Godoy F, Hargreaves KM. Regenerative endodontics: a revised ofcurrent statue and a call for action. J Endod 2007;33:377–90.

22. Shi S, Bartold PM, Miura M, et al. The efficacy of mesenchymal stem cells to regen-erate and repair dental structures. Orthod Craniofac Res 2005;8:191–9.

23. Yang X, Zhang W, van den Dolder J, et al. Multilineage potential of STRO-1+ ratdental pulp cells in vitro. J Tissue Eng Regen Med 2007;1:128–35.

24. Yang X, Walboomers XF, van den Beucken JJ, et al. Hard tissue formation of STRO-1-selected rat dental pulp stem cells in vivo. Tissue Eng Part A 2009;15:367–75.

25. Gronthos S, Graves SE, Ohta S, et al. The STRO-1+ fraction of adult human bonemarrow contains the osteogenic precursors. Blood 1994;84:4164–73.

26. Mao M, Fu G, Wu JS, et al. Identification of genes expressed in human CD34(+)hematopoietic stem/progenitor cells by expressed sequence tags and efficientfull-length cDNA cloning. Proc Natl Acad Sci U S A 1998;95:8175–80.

27. Kara€oz E, Do�gan BN, Aksoy A, et al. Isolation and in vitro characterisation of dentalpulp stem cells from natal teeth. Histochem Cell Biol 2010;133:95–112.

Isolation of CXCR4-positive Cells from Human DPCs 795