adult human dental pulp stem cells differentiate toward ... › images › pdf ›...

Adult Human Dental Pulp Stem Cells Differentiate TowardFunctionally Active Neurons Under AppropriateEnvironmental Cues

AGNES ARTHUR,a,b,c GRIGORI RYCHKOV,d SONGTAO SHI,e SIMON ANDREA KOBLAR,a,b STAN GRONTHOSc

aThe Australian Research Council, Centre for the Molecular Genetics of Development, University of Adelaide,Adelaide, Australia; bSchool of Molecular and Biomedical Science (Genetics), University of Adelaide, Adelaide,Australia; cMesenchymal Stem Cell Group, Division of Haematology, Institute of Medical and Veterinary Science,Hanson Institute/Adelaide University, Adelaide, Australia; dSchool of Molecular and Biomedical Science(Physiology), University of Adelaide, Adelaide, Australia; eSchool of Dentistry, University of Southern California,Los Angeles, California, USA

Key Words. Dental pulp stem cells • Neuronal differentiation

ABSTRACT

Human adult dental pulp stem cells (DPSCs) reside withinthe perivascular niche of dental pulp and are thought tooriginate from migrating cranial neural crest (CNC) cells.During embryonic development, CNC cells differentiate intoa wide variety of cell types, including neurons of the periph-eral nervous system. Previously, we have demonstrated thatDPSCs derived from adult human third molar teeth differ-entiate into cell types reminiscent of CNC embryonic ontol-ogy. We hypothesized that DPSCs exposed to the appropri-ate environmental cues would differentiate into functionallyactive neurons. The data demonstrated that ex vivo-ex-panded human adult DPSCs responded to neuronal induc-tive conditions both in vitro and in vivo. Human adultDPSCs, but not human foreskin fibroblasts (HFFs), ac-quired a neuronal morphology, and expressed neuronal-

specific markers at both the gene and protein levels. Cul-ture-expanded DPSCs also exhibited the capacity to producea sodium current consistent with functional neuronal cellswhen exposed to neuronal inductive media. Furthermore,the response of human DPSCs and HFFs to endogenousneuronal environmental cues was determined in vivo usingan avian xenotransplantation assay. DPSCs expressed neu-ronal markers and acquired a neuronal morphology follow-ing transplantation into the mesencephalon of embryonicday-2 chicken embryo, whereas HFFs maintained a thinspindle fibroblastic morphology. We propose that adult hu-man DPSCs provide a readily accessible source of exogenousstem/precursor cells that have the potential for use in cell-therapeutic paradigms to treat neurological disease. STEMCELLS 2008;26:1787–1795

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

INTRODUCTION

Regeneration of the disrupted human nervous system fromdisease or trauma is a challenge for stem cell-based therapeuticparadigms (reviewed by Lindvall and Kokaia [1]). Thus, thegoal of neuroregeneration may entail a complex process ofneuroprotection, immunomodulation of inflammation or glio-genesis, neuroplasticity, and neurogenesis. Understanding theunderlying cellular and molecular mechanisms that may facili-tate this complex process is paramount to achieve an improve-ment in neurological function.

In model organisms, both endogenous and exogenous neuralstem cells (NSCs) have been investigated for their capacity toregenerate a damaged nervous system. Due to the low incidenceof human adult NSCs and problems with accessibility, the use of

exogenous sources of stem cells with neural potential has beensuggested as a plausible approach to stem cell therapy. Althoughembryonic stem cells [2] (reviewed by Lindvall et al [3] andLindvall and Kokaia [4]) and bone marrow stem cells have beenassessed as potential candidates for neuronal therapy, adult stemcell populations derived from cranial neural crest (CNC) cellsmay possess a greater propensity for neuronal differentiationand repair.

Dental pulp tissue is thought to be derived from migratingneural crest cells during development [5, 6], and has been shownto harbor various populations of multipotential stem/progenitorcells [7–10]. In vitro neural differentiation studies of rat andhuman adult dental pulp cells (DPCs) and stem cells fromhuman exfoliated deciduous teeth (SHEDs) demonstrated thatthese stem/precursor cell populations were able to differentiateinto neurons based on cellular morphology and the expression of

Author contributions: A.A.: conception and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing;G.R.: collection and/or assembly of data, data analysis and interpretation, manuscript writing; S.S.: conception and design, provision of studymaterial or patients; S.K. and S.G.: conception and design, data analysis and interpretation, manuscript writing, financial support, finalapproval of manuscript.

Correspondence: Prof. Stan Gronthos, Ph.D., Bone and Cancer Laboratories, Division of Hematology, Institute of Medical and VeterinaryScience/Hanson Institute, Frome Road, Adelaide 5000, SA, Australia. Telephone: �61-8-8222-3460; Fax: �61-8-8222-3139; e-mail:[email protected] Received November 26, 2007; accepted for publication May 9, 2008; first published online in STEM CELLSEXPRESS May 22, 2008. ©AlphaMed Press 1066–5099/2008/$30.00/0 doi: 10.1634/stemcells.2007-0979

TISSUE-SPECIFIC STEM CELLS

STEM CELLS 2008;26:1787–1795 www.StemCells.com

early neuronal markers [8, 9]. Furthermore, in vivo studiesdemonstrated that rat DPCs and SHEDs when transplanted intothe adult rodent brain survived and expressed neuronal markers[8, 9].

We have previously reported that multipotential dental pulpstem cells (DPSCs) [7] retain their neural crest properties fol-lowing ex vivo expansion [11–15]. Preliminary investigationsinto the neural potential of human adult DPCs and DPSCs haveshown that under non-neuronal inductive conditions, these cellsexpressed the neural progenitor marker, Nestin, and glialmarker, glial fibrillary acidic protein (GFAP), at both the geneand protein levels [12, 16]. Subsequent examinations docu-mented that DPSCs expressed the postmitotic neuron-specificmarker, neuronal nuclei (NeuN), when cultured under neuralinductive conditions [11]. We hypothesized that DPSCs ex-posed to the appropriate environmental cues would differentiateinto functionally active neurons and that neural crest-derivedadult DPSCs may provide an alternative stem cell source fortherapy-based treatments of neuronal disorders and injury.

The present study examined the neuronal differentiationpotential of adult human DPSCs both in vitro and in vivo, usingfactors known to initiate neuronal differentiation [17–20]. Thefindings of the present study demonstrated that adult humanDPSCs have a predisposition to a neuronal fate that is furtherenhanced when exposed to neuronal differentiation and envi-ronmental factors.

MATERIALS AND METHODS

Isolating DPSCs and Human Foreskin FibroblastsDPSCs and SHEDs were isolated as described by [7, 8, 13]. Briefly,discarded normal human impacted third molars were collected fromadults (19–35 years of age) or exfoliated teeth (7–8 years of age)with informed consent of patients undergoing routine extractions atthe Dental Clinic of the University of Adelaide, under approvedguidelines set by the University of Adelaide and Institute of Medicaland Veterinary Science Human Subjects Research Committees (H-73–2003). Tooth surfaces were cleaned and cracked open using avice to reveal the pulp chamber. The pulp tissue was gently sepa-rated from the crown and root and then digested in a solution of 3mg/ml collagenase type I and 4 mg/ml dispase for 1 hour at 37°C.Single-cell suspensions were obtained by passing the cells througha 70-�M strainer. Cultures were established by seeding single-cellsuspensions (1–2 � 105) of dental pulp into T-25 flasks in growthmedia, (�-modification of Eagle’s medium supplemented with 20%fetal calf serum, 100 �M L-ascorbic acid 2-phosphate, 2 mML-glutamine, 100 U/ml penicillin, and 100 �g/ml streptomycin),then incubated at 37°C in 5% CO2.

Human foreskin fibroblasts (HFFs) were isolated by collage-nase/dispase digestion from explants of foreskin biopsies fromneonatal male donors, with informed parental consent.

Neural Induction AssayTissue culture flasks or wells were coated with polyornithine (10�g/ml final concentration) overnight at room temperature. Flasksand wells were washed twice with water and then coated withlaminin (5 �g/ml final concentration) overnight at 37°C in a humidincubator. Flasks and wells were washed with phosphate-bufferedsaline (PBS), and then with growth media before cells were added.Cells were thawed and grown in T25-coated flasks until 80%confluent. Cells were reseeded at a density of 1.5 � 105/well in 2ml/well growth media, for 3 days in 6-well coated plates, or 2 � 103

cells in 8-well coated chamber slides. Cells were then cultured ineither media A or media B for 3 weeks. Media A consists ofNeurobasal A Media (10888-022; Invitrogen, Carlsbad, CA, http://www.invitrogen.com), 100 U/ml penicillin, 1� B27 supplement,100 �g/ml streptomycin, 20 ng/ml epidermal growth factor (EGF,100-15; PeproTech, Rocky Hill, NJ, http://www.peprotech.com),

and 40 ng/ml basic fibroblast growth factor (FGF, no. 104FGFB01;Prospec Tany Techno Gene, Rehovot, Israel). Media B consists ofthree separate media conditions for the duration of the 3-weekperiod: the first incubation was media A for 7 days, followed by achange in media consisting of 50:50 ratio of Dulbecco’s modifiedEagle’s medium (DMEM, 12100-046; Invitrogen) and F12 media(21700-075; Invitrogen), and insulin-transferrin-sodium-selenitesupplement (ITTS, 11074547001; Roche Diagnostics, Basel, Swit-zerland, http://www.roche-applied-science.com), 100 U/ml penicil-lin, 100 �g/ml streptomycin, and 40 ng/ml FGF for 7 days. The final7 days of incubation consisted of media containing 50:50 ratio ofDMEM and F12 media, ITTS, 100 U/ml penicillin and 100 �g/mlstreptomycin, 40 ng/ml FGF, and 0.5 �M retinoic acid (RA, R2625;Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com). Controlsamples were maintained in growth media for the duration of theneuronal inductive assay. The media for all conditions were re-placed twice weekly. Following the final incubation, the 8-wellcoated chamber slides were fixed with 4% paraformaldehyde(PFA), and cells in 6-well plates were lysed with TRIzol (Invitro-gen) and stored for RNA isolation and real time-polymerase chainreaction (RT-PCR) analysis or replated for electrophysiologicalanalysis.

2-(4-Iodophenyl)-3-(4-Nitrophenyl)-5-(2,4-Disulfophenyl)-2H-Tetrazolium, Monosodium SaltAssayCell proliferation of human adult DPSCs undergoing neuronal dif-ferentiation was performed as follows. Human adult DPSCs fromeach donor were seeded at 8 � 103 cells/well in 96-well plates inquadruplicate for each of the media conditions and for each mediachange time points and were incubated at 37°C in a 5% CO2-containing humidified atmosphere. Cells were cultured in controlmedia for 2 days and then changed to the corresponding mediaconditions as per the neuronal differentiation method. Cell numberand viability were quantitated at each media change using thecolorimetric reagent 2-(4-Iodophenyl)-3-(4-nitrophenyl)-5-(2,4-di-sulfophenyl)-2H-tetrazolium, monosodium salt (WST)-1 (RocheMolecular Biochemicals, Indianapolis). The absorbance was mea-sured directly with a plate reader (BIO-RAD Model 3550 micro-plate reader; Bio-Rad, Hercules, CA, http://www.bio-rad.com) us-ing the test wavelength of 450 nm. Results of representativeexperiments are presented as the mean � SEM.

Real-Time Polymerase Chain ReactionRNA was extracted from DPSC neuronal differentiation assays.Briefly, total RNA was extracted using the TRIzol method. RNAsamples were quantified by spectrophotometer (Eppendorf, Ham-burg, Germany, http://www.eppendorf.com), and RNA integritywas checked on 1% agarose gels using a deionized formamide-based loading buffer. Reverse-transcription reactions were per-formed using Superscript III reverse transcriptase (Invitrogen).cDNA samples were diluted to a uniform concentration of 50 ng/�l.Real-time PCR reactions were performed using TaqMan master mixon an ABI SDS 7000 light cycler driven by ABI prism SDS v1.1(Applied BioSystems, Foster City, CA, http://www.appliedbiosystems.com). TaqMan primers were designed usingPrimer Express v2.0 (Applied Biosystems) and synthesized locally(GeneWorks, South Australia, Australia, http://www.geneworks.com.au/). Primers TBP (TATA box-binding protein) forward (CTGGAA AAG TTG TAT TAA CAG GTG CT), reverse (CCA TCACGC CAC AGT TTC C); Nestin (NM 006617) forward (5� GTC-CATCCTCAGTGGGTCAGA 3�) reverse (5� CCGATTGAGCTC-CCACATCT 3�); �-III tubulin (accession NM 006086) forward(5� GGGCCAAGTTCTGGGAAGTC 3�) reverse (5� ATCCGCTC-CAGCTGCAAGT 3�); neurofilament-medium chain (NF-M; acces-sion NM 005382) forward (5� GACGGCGCTGAAGGAAATC3�) reverse (5� CTCTTCGCCCTGGTGCATAT 3�); and neurofila-ment-heavy chain (NF-H; accession NM 021076) forward (5�CAGCCAAGGTGAACACAGAC 3�) reverse (5� GCTGCT-GAATGGCTTCCT 3�) were used at a final concentration of 300nM, and reactions for each sample were performed in triplicate.

1788 Neuron Differentiation of Human DPSCs

ElectrophysiologyAt the completion of the 3-week neuronal differentiation assay, cellswere liberated with trypsin and replated onto glass coverslipstreated with hydrochloric acid at a concentration of 2 � 104 cells/500 �L in neuronal differentiation media or growth media andincubated for 24 hours. Whole-cell patch clamping was performedat room temperature using a computer-based patch clamp amplifier(EPC-9; HEKA Electronics, Lambrecht/Pfalz, Germany, http://www.heka.com/) and PULSE software (HEKA Electronics). Thebath solution contained (in mM): NaCl, 140; KCl, 4; CaCl2, 2;MgCl2, 2; glucose, 10; and HEPES, 10; adjusted to pH 7.4 withNaOH. To block K� channels and separate Na� currents, K gluta-mate and KCl in the internal solution were replaced with equimolaramounts of Cs glutamate and CsCl. Patch pipettes were pulled fromborosilicate glass and fire polished; pipette resistance ranged be-tween 2–4 M�. All voltages shown were corrected for liquidjunction potential (�14 mV and �18 mV for K�- and Cs�-basedinternal solutions, correspondingly), estimated by JPCalc (providedby Dr. P.H. Barry, University of New South Wales, Sydney, Aus-tralia, 1994). The holding potential was �90 mV throughout.

Avian Embryo InjectionsEthical approval was obtained from the University of Adelaide(approval number S-59–2003). Chicken eggs (white leghorn;HICHICK Breeding Company, Bethal, Australia) were placed long-ways in egg trays and incubated in a humidified incubator at 37°Cfor approximately 40 hours to reach stages 10–12 [21] prior toinjection. Green fluorescent protein (GFP)-transduced adult humanDPSCs or HFFs (5 � 103 cells/�L) were injected into the devel-oping avian embryo. Fast green dye (2 �L) was added to the cellsto visualize them during the injection procedure. Embryos wereprepared by wiping the egg with 70% ethanol, albumin was ex-tracted from the egg using a needle, a window was cut into the topof the egg, and the embryo was visualized by injecting Indian ink(Winsor and Newton, Middlesex, England, http://www.winsornewton.com/) (1:10 prepared in Ringer’s solution [7.2 g NaCl2, 0.17 gCaCl2, 0.37 g KCl, 0.115 g Na2HPO4 in 1 L ddH2O, adjusted to pH7.4, filter sterilized]) underneath the embryo. Ringer’s solution wasplaced on top of the embryo to maintain moisture. The vitellinemembrane was removed from around the head of the embryo usingtungsten wire. The cells were placed in a glass capillary needle(GC100TF-10; SDR Clinical Technology, Sydney, Australia, http://www.sdr.com.au/), attached to a micromanipulator and pressureinjector (Narishage, Tokyo), set at 25 volts. The micromanipulatorwas used to guide the needle into the region directly adjacent torhombomere 2 in the developing hindbrain. Cells were injected intothe embryo using a foot pump attached to the pressure injector. Theglass needle was removed and a few drops of Ringer’s solution wereplaced on top of the embryo. The egg was sealed with sticky tapeand placed back in the humidified incubator. Following the incu-bation period, embryos were cut out of the egg and placed in coldPBS; GFP-injected cells were visualized using a fluorescent dis-secting microscope. Embryos were then dissected; the head was cutas an open book, that is, cut from the nose toward the hindbraindown the length of the head, along the ventral side. Dissectedembryos were fixed in 4% PFA for either 2 hours at room temper-ature or overnight at 4°C, and then washed twice with PBS prior toimmunofluorescent staining.

Immunocytochemistry of Cultured Cell PreparationsChamber slides were fixed with 4% PFA for 30 minutes at roomtemperature and then washed with PBS and 0.1% Tween 20 (Sigma-Aldrich) (PBS-T). Cultures were blocked (10% horse serum inPBS-T) for 30 minutes at room temperature and then incubated withprimary antibody (1:100 Nestin (611658; BD Biosciences, SanDiego, http://www.bdbiosciences.com); 1:500 polysialic acid-neu-ral cell adhesion molecule (PSA-NCAM) (gift from Dr. T. Seki),1:500 �-III tubulin clone, TUJ1 (MMS-435P, 1 mg/ml; Covance,Princeton, NJ, http://www.covance.com/); 1:200 NF-M (13–0700,0.5 mg/ml; Zymed, South San Francisco, CA); 1:500 NF-H(AB1991; Chemicon, Temecula, CA, http://www.chemicon.com) inblocking solution overnight at 4°C. Mouse, goat, and rabbit (Ig)

controls were treated under the same conditions. After washing, thesecondary antibodies (1:200 goat anti-mouse Alexa 488 [A11029,1.5 mg/ml; Jackson Immunoresearch Laboratories, West Grove,PA, http://www.jacksonimmuno.com]; 1:200 goat anti-rabbit Alexa488 [A11034, 2 mg/ml [Molecular Probes Inc., Eugene, OR, http://probes.invitrogen.com]) were added in blocking solution for 2 hoursat room temperature in the dark. Finally, the slides were washed andcoverslipped with Prolong gold antifade with 4�,6-diamidino-2-phenylindole dihydrochloride (DAPI, P36931; Invitrogen).

Chicken Embryo StainingFollowing 4% PFA fixation and open book dissection, injectedavian embryos were washed five times, 5 minutes per wash, withPBS and 0.3% Triton-X (Sigma-Aldrich) (PBS-TX). Embryos werethen incubated with constant movement in blocking solution (10%horse serum in PBS containing 1% Triton-X) for 5 hours at roomtemperature. The solution was then replaced with either 1:100dilution of mouse anti-human integrin-�1 (Chemicon, MAB2000, 1mg/ml) or 1:250 dilution of anti-goat GFP antibody (600–101-215,1 mg/ml; Rockland Immunochemicals, Inc., Gilbertsville, PA,http://www.rockland-inc.com) and either anti-mouse �-III tubulinclone, TUJ1 (1:250), or NF-M (1:100) in PBS containing 10% horseserum and 0.1% Triton-X; samples were incubated with constantmovement for 24 hours at room temperature. Embryos were washedwith PBS-TX and incubated with corresponding secondary 1:200dilution of donkey anti-goat Alexa 488 (A11055, 2 mg/ml; Molec-ular Probes) and donkey anti-mouse cyanin 3 (Cy3, 715–165-150,1.5 mg/ml; Jackson Immunoresearch Laboratories) overnight at 4°Cwith constant movement in the dark. Samples were washed withPBS-TX, and then transferred to 80% glycerol; once equilibrated inglycerol, the embryos were placed as open books on a slide andcoverslipped with Prolong gold antifade with DAPI.

Imaging and Image ProcessingThe bright-field images of the neuronal differentiation cultures wereacquired using Olympus CKX41 fitted with an Olympus DP11camera (Olympus, Tokyo, http://www.olympus-global.com). The invitro neuronal differentiation culture-stained images were acquiredusing the Olympus AX70 microscope fitted with a cooled CCDcamera (Olympus), and V�� v.4 software (Nikon Australia, Lid-combe NSW, Australia, http://www.nikon.com.au). Whole-mountchicken embryo images were taken using confocal microscopy(Bio-Rad Radiance 2100). Confocal images and z-images wereprocessed and analyzed using Confocal Assistant v.4.02 (Bio-Rad). All digital images were processed using Adobe Photoshop6 (Adobe Systems, San Jose, CA, http://www.adobe.com) andonly brightness and/or contrast were altered.

RESULTS

Neural Induction of DPSCs In VitroIn the present study, we examined the effect of different factorsand media formulations to facilitate neural differentiation ofhuman adult DPSCs. We assessed two different neuronal mediaconditions previously described to promote the neural differen-tiation of postnatal SHEDs and embryonic stem cells [8, 22, 23].Following 3 weeks of induction, DPSCs exposed to eitherneuronal inductive media A or B acquired a bipolar and stellatemorphology consistent with that of sensory and motor neurons,respectively (Figs. 1A, 2A). Assessment of the cell proliferationstatus of DPSCs over the same time period showed that therewas a significant (media A �, � p �. 002; media B #, � p �.00000007; Student t test) 2-fold and 2.6-fold decrease in theproliferation rate of DPSCs cultured in neuronal inductive me-dia A and media B conditions, respectively, compared withcontrol cultures, as demonstrated using the WST-1 assay (Fig.1B). Furthermore, the reduction in cell proliferation was not aresult of cell death, as determined by the counting of trypanblue-positive cells relative to living cells in the supernatant

1789Arthur, Rychkov, Shi et al.

www.StemCells.com

following each media change (supplemental online Fig. 1);whereas dead or dying cells uptake trypan blue, living or healthycells remain clear.

The protein expression patterns of early (Nestin, PSA-NCAM), intermediate (�-III tubulin), and late (NF-M or NF-H)neuronal-associated markers expressed by DPSCs cultured inthe different neuronal inductive conditions were determined byimmunocytochemical analysis. The data demonstrated a pro-gressive increase in the early/intermediate (Nestin, PSA-NCAM, and �-III tubulin) and mature (NF-M and NF-H) neu-ronal markers (Fig. 1A, 1C). In contrast to the proteinexpression of the Nestin, the proportions of PSA-NCAM-, �-IIItubulin-, NF-M-, and NF-H-positive cells were significantlylower (media A: p � .0005, p � .000002, p � .0004, and p �.003, respectively; media B: p � .0005, p � .000005, p � .005,and p � .003, respectively; Student t test; n � 3 DPSC donors)in control media compared with DPSCs cultured under bothneuronal inductive conditions (Fig. 1C).

Supportive real-time PCR analysis indicated that the geneexpression profile of DPSCs was more consistent with that of

mature neuronal cells, following 3 weeks of neuronal induc-tion. When DPSCs derived from three different donors werecultured in neuronal inductive media, the transcript levels ofthe neuronal precursor marker, Nestin, appeared to be un-changed or slightly downregulated (Fig. 1D). Similarly, therewas a minor decrease observed in the expression of theearly/intermediate neuronal gene, �-III tubulin, over thesame time period. Conversely, it was found that when DPSCswere cultured with either media A or media B, there wassignificant increase (p � .015 and p � .0004, respectively,Student t test), in the transcript levels for NF-M comparedwith cells maintained in control media (Fig. 1D). Further-more, a significant increase in the gene expression of themature neuronal marker, NF-H, was also observed whenDPSCs were cultured under both media A and B (p � .02 andp � .00001, respectively, Student t test), in comparison withcontrol cultures (Fig. 1D). Collectively, these data suggestthat in response to the neuronal inductive stimuli, a greaterproportion of DPSCs had stopped proliferating and had ac-quired a phenotype resembling mature neurons.

A

Nes

tinβ-

III tu

bulin

NF-

MN

F-H

C ontro l M ed ia N eura l Induction

PSA-

NC

AM0

1

2

3

4

5

M E M A M E M A BS et U p F inal

Abso

rban

ce 4

50nm

B

* #B

* *# #^ +± ±C

0

20

40

60

80

100

N estin P S A β-III tubu lin N F-M N F-H% o

f pos

itive

cel

ls /

field

of v

iew

M E M M edia A M ed ia B

D* # ^ +

-4-202468

10

N estin β-III tubu lin N F-M N F-H

Fold

Cha

nge

M edia A M ed ia B

Nes

tinβ-

III tu

bulin

NF-

MN

F-H


PSA-

NC

AMN

estin

β-III

tubu

linN

F-M

NF-

H


PSA-

NC

AM0

1

2

3

4

5


Abso

rban

ce 4

50nm

B

* #

0

1

2

3

4

5


Abso

rban

ce 4

50nm

B

* #

* *# #^ +± ±

0

20

40

60

80

100


f pos

itive

cel

ls /

field

of v

iew

M E M M edia A M ed ia B* *# #^ +± ±

0

20

40

60

80

100


f pos

itive

cel

ls /

field

of v

iew

M E M M edia A M ed ia B

* # ^ +

-4-202468

10


Fold

Cha

nge

M edia A M ed ia B* # ^ +

-4-202468

10


Fold

Cha

nge

M edia A M ed ia B

Figure 1. Dental pulp stem cells (DPSCs) cultured in neural inductive media expressed neural markers. (A): Representative images of DPSCsstained with Nestin, �-III tubulin, polysialic acid-neural cell adhesion molecule (PSA-NCAM), neurofilament-medium chain (NF-M), and matureneuronal marker neurofilament-heavy chain (NF-H). Scale bar � 20 �m. (B): The proliferation potential of DPSCs following the neuronaldifferentiation assay is represented as absorbance readings following a 2-(4-Iodophenyl)-3-(4-nitrophenyl)-5-(2-4-disulfophenyl)-2H-tetrazolium,monosodium salt (WST-1) assay of cells cultured in the corresponding media. When DPSCs were cultured in media A or media B for 3 weeks, therewas a 2-fold and 2.6-fold decrease, respectively, in cell proliferation, compared with the control media (MEM) (media A �, p � .002; media B #,p � .00000007; Student’s t test). (C): The number of DPSCs positively stained for the respective neuronal-specific antibodies represented as apercentage of field of view demonstrate a 2-fold, 3-fold, 3.5-fold, and 6.5-fold increase in PSA-NCAM, �-III tubulin, NF-M, and NF-H staining,respectively, of DPSCs cultured in neuronal inductive conditions compared with control (�, p � .0005; ∧, p � .00002; �, p � .000005; �, p � .0004;# p � .005; Student’s t test). (D): Real time-polymerase chain reaction analysis of total RNA isolated from three independent DPSC donors exposedto neuronal and control media conditions, expressed as a fold change relative to control media. Samples were normalized to control gene, TBP. DPSCsin both media A and media B downregulated the expression of neuronal precursor, Nestin, and early neuronal marker, �-III tubulin gene, whiledemonstrating a fourfold and sevenfold up-regulated gene expression of mature neuronal marker, NF-M, compared with control media, respectively;n � 3 replications (�, p � .015; #, p � .0004, respectively; Student’s t test). Furthermore, gene expression for the more mature neuronal markerindicated a sixfold and fivefold increase in NF-H expression when DPSCs were cultured in media A and B, respectively; n � 3 replications (∧, p �.03; �, p � .00002, respectively; Student’s t test) compared with control media.


Functional Analysis of the Neuronal Properties ofDPSCs In VitroTo confirm that adult human DPSCs were differentiating intofunctionally active neurons, electrophysiology was performedon cells grown under normal and neuronal inductive conditions(Fig. 2A). The presence of the ionic currents was first investi-gated by applying voltage ramps between �120 and 120 mV tothe cells patch clamped using K�-based pipette solution. DPSCscultured in standard growth media (Fig. 2C, nondifferentiation[ND]) displayed a small inward current, whereas DPSCs cul-tured in neuronal differentiation media produced a current thatincreased approximately 8.5-fold (Fig. 2C, 2E, differentiation[D]). To separate inward currents, we used Cs�-based pipettesolution. Moreover, differentiated DPSC (D) voltage steps re-vealed large, fast, inactivating inward current with a threshold atapproximately �55 mV and a maximum at approximately �5mV (Fig. 2D, 2E). This current was approximately 65% blockedby 30 nM tetrodotoxin (TTX), although 300 nM TTX blocked itcompletely (Fig. 2E). This confirmed that differentiated DPSCsexpress voltage-gated Na� channels. To investigate further theionic composition of the inward current, Na� in the externalsolution was replaced with Ba2�. As Ba2� permeates all volt-age-gated Ca2� channels, but not Na� channels, this wouldreveal the presence of voltage-gated Ca2� channels. When 140mM NaCl in the external solution was replaced with 100 mM

BaCl2, the inward current completely disappeared (not shown),which confirmed that this current was carried by Na� and notCa2� channels, thus validating the presence of voltage-gatedNa� channels in DPSCs. In contrast, HFFs (Fig. 2A) showed nosignificant inward currents and high-threshold outward currentsin either ND or D media (Fig. 2B). Furthermore, no differencein current amplitude was observed by HFFs cultured in controlor neuronal induction media (Fig. 2B). In addition, although K�

was the main cation in the pipette solution, neither HFFs norDPSCs showed appreciable outward K� currents in the phys-iologically relevant range of membrane potentials (Fig. 2B,2C). The size of the outward current was not affected bydifferentiation. Our in vitro data suggested that DPSCs havea predisposition to differentiate into functionally active neu-rons, which is augmented when cultured in neuronal induc-tive media.

Neural Induction of DPSCs In VivoA developmental avian embryo model was adapted to investi-gate the neuronal differentiation capacity of human DPSCsduring a time of active neurogenesis in vivo [24]. Stably trans-duced GFP-expressing DPSCs (n � 7 donors) and control HFFs(n � 1 cell line) were injected into regions adjacent to thedeveloping neural tube at stages 10–12 [21]. At this stage ofdevelopment, CNC cells coalesce, migrate to their target tissue,

Figure 2. Dental pulp stem cells (DPSCs) produce a sodium current that is accentuated in neuronal inductive conditions. (A): Representativebright-field images of DPSCs and human foreskin fibroblasts (HFFs), cultured in nondifferentiation (ND) and differentiation (D) media that underwentelectrophysiological analysis. Scale bar � 25 �m. (B): Representative I-V plots obtained in differentiated (D) and nondifferentiated (ND) HFFs inresponse to 100 ms voltage ramps. (C): Representative I-V plots obtained in differentiated (D) and nondifferentiated (ND) DPSCs in response to 100ms voltage ramps. (D): Average I-V plots of peak Na� currents obtained in differentiated and nondifferentiated DPSCs in response to voltage steps.(E): Representative traces of Na� currents recorded in differentiated DPSCs under normal conditions (control) and in the presence of Na�-blockingagent tetrodotoxin (TTX) at 30 nM and 300 nM, demonstrating the block of Na� currents in differentiated DPSCs by TTX.


www.StemCells.com

and differentiate into neural derivatives [25]. In the presentstudy, the neuronal differentiation of transplanted DPSCs andHFFs was investigated at 24, 48, and 72 hours after injection.

The detection of injected human DPSCs was predominantlyassessed using an antibody to GFP; however, to further confirmthat the injected cells were of human origin, injected embryoswere also stained with human-specific integrin-�1 antibody at48 hours after injection (supplemental online Fig. 2). At 24hours after injection, DPSCs exhibited extended processes andlocalized to regions normally followed by migrating endoge-nous CNC cells, but did not colocalize with neuronal-specificmarkers (data not shown). However, the response of DPSCs(n � 53 embryos) to the endogenous neuronal environment wasmore pronounced at 48 hours after injection. DPSCs displayedthe morphology of bipolar cells (Fig. 3B, asterisk) and neuronswith multiple neurites (Fig. 3B, 3D, arrows; supplemental onlineFig. 3A, 3B), which appeared to have migrated into the centralnervous system (CNS) based on merged z-series confocal imageanalysis. Importantly, the DPSCs that displayed the multipleneurite morphology also colocalized with both early and matureneural markers, �-III tubulin (Fig. 3A, 3B; supplemental onlineFig. 3A, 3B) and NF-M (Fig. 3C, 3D, arrows; supplementalonline Fig. 3C, 3D), respectively. NF-M was lowly expressed(Fig. 3D, arrows) by the differentiated DPSCs in comparisonwith GFP staining. Growth cone-like structures were also ob-served with NF-M staining, suggesting that these DPSCs haddifferentiated into neuron-like cells (Fig. 3C, 3D, asterisks),which appeared similar to endogenous avian axonal processes.SHEDs also responded in a similar manner to DPSCs (n � 15embryos), which migrated closely to the axons of the trigeminalganglion (TG) (data not shown).

To validate that the response of adult human DPSCs andSHEDs to the in vivo neural environment was specific to thesecell types or whether all ectodermal-derived cells were able torespond to the same embryonic environmental cues, transducedGFP-expressing HFFs (n � 17 embryos) were injected into theavian embryo. In these studies, the HFFs were observed tomigrate toward the epidermal layer of the embryo rather thanwithin the CNS or peripheral nervous system (PNS) or alongaxonal processes of the TG. Although the morphology of HFFswas thin and spindle-like, similar to bipolar neural cells, GFP-labeled HFFs lacked expression of �-III tubulin and NF-M (Fig.3E–3H; supplemental online Fig. 3E, 3F) at 24, 48, and 72 hoursPI.

Seventy-two hours after injection, adult human DPSCs sur-vived and displayed bipolar processes when in close proximityto endogenous sensory axons (Fig. 4A, 4B, asterisks; supple-mental online Fig. 4A), whereas the DPSCs exhibited multineu-rite processes when localized to the CNS (Fig. 4C, 4D, arrow-heads; supplemental online Fig. 4B). Furthermore, 7 days afterinjection, GFP could no longer be detected, however, the in-jected human DPSCs were still viable, as demonstrated withintegrin-�1 staining of frozen sections (supplemental online Fig.5). These observations suggested that human DPSCs were ca-pable of responding to endogenous migratory and neuronaldifferentiation cues and followed CNC pathways.

SUMMARY

In the present study, we investigated the neuronal potential ofhuman adult DPSCs that have previously been shown to retainproperties characteristic of neural crest cells following ex vivoexpansion [15, 26, 27]. Ex vivo-expanded DPSCs were culturedin different media formulations previously reported to induceneuronal differentiation of embryonic stem cells [22] and post-natal stem cell populations [8, 28] when cultured in the presenceof growth factors such as EGF [17], FGF [18], and/or RA [19].Similar media conditions have also been reported to be capableof maintaining neuronal cell types in culture [29–31]. Our datashowed that DPSCs were receptive to neural differentiationusing both media conditions over a 3-week duration. Theseconditions are in accord with other investigations that examinedthe neural differentiation potential of SHEDs and embryonicstem cells [8, 22, 23]. However, although these conditions dovary, both media were able to induce both a morphological anda gene expression response, suggesting that factors present inboth media conditions (EGF and FGF) were predominatelyresponsible for the neuronal induction.

Following neural induction in vitro, DPSCs were shown toconstitutively express the immature markers Nestin and PSA-NCAM, which are associated with neuronal precursors andimmature neuronal-committed progenitors through to neuro-blasts, respectively. This confirms our original findings that exvivo-expanded postnatal dental pulp-derived stem cells consti-tutively express Nestin and the GFAP [8, 12], where these twomarkers are also expressed within fibrous dental pulp tissue insitu [32, 33]. Whereas Nestin protein expression continued to bedetected by the majority of DPSCs following neuronal induc-

Figure 3. Dental pulp stem cells (DPSCs)injected into the avian embryo displayedneural morphology and localized with neuralmarkers 48 hours after injection. DPSCs(green) migrated within the developingavian embryo into facial structures and intothe central nervous system. DPSCs dis-played neural morphology, both bipolar (as-terisk) and with multiple processes (arrow),localizing with (A–B) �-III tubulin (red, n �22) and (C–D) neurofilament-medium chain(NF-M; arrows) (red, n � 31). (C–D): Com-pressed image of confocal z-series, demon-strating (C) green fluorescent protein-la-beled DPSCs (arrows) colocalizing with (D)NF-M staining. (E–H): Control cell line hu-man foreskin fibroblasts (HFFs) injected inthe same way as DPSCs did not display aneuronal morphology, nor did they colocal-ize with early or mature neuronal markers(E–F) �-III tubulin or (G–H) NF-M, respec-tively. Abbreviations: HB, hindbrain; TG,trigeminal ganglion. Scale bar � 100 �m.


tion, gene transcription was found to be downregulated at 3weeks after induction, correlating to neuronal maturation. Earlyneuronal differentiation was further investigated by assessing�-III tubulin expression. This molecule is the only phosphory-lated tubulin, considered to be a neuronal-specific marker that isexpressed during early brain development and is down-regu-lated as the brain develops into adulthood [34]. We observedthat although the number of �-III tubulin positive cells increasedover the 3-week induction period, the mRNA levels were foundto be downregulated, which also correlated to the observed reduc-tion in Nestin gene expression. Furthermore, mature neuronaldifferentiation was evaluated by assessing the expression ofboth NF-M and NF-H, where NF-M is known to be expressedprior to NF-H during neuronal differentiation. NF-M is one ofthree intermediate filaments, and is an important structural com-ponent of the axon cytoskeleton [35, 36], whereas it is expressedby large projection or myelinated neurons in vivo [37–39]. Thenumber of DPSCs expressing either NF-M or NF-H gene orprotein were found to be significantly higher following neuronalinduction compared with control cultures. Overall, the gene andprotein expression profiles suggested that when DPSCs areexposed to different neuronal inductive conditions, these cellscease to proliferate and begin to undergo differentiation with agene and protein expression profile consistent with mature neu-rons. Although mostly circumstantial, our in vitro studies sup-port the notion that at least a proportion of stem cells derivedfrom postnatal dental pulp tissues may be capable of differen-tiating into neuronal cells when cultured under appropriateinductive conditions.

More in-depth functional studies were undertaken to furthersubstantiate this concept by performing electrophysiology.Functional neurons express different types of voltage-gated ionchannels, which are required for the generation and propagation

of action potentials. These include voltage gated K�, Na�, andCa2� [40]. In our study, only functional voltage-gated Na�

channels and not Ca2� were present in adult DPSCs whenexposed to neuronal inductive conditions. Importantly, the pres-ence of functional Na� channels, and the size and the propertiesof the voltage-gated Na� currents in differentiated DPSCs wereconsistent with those reported for neuronal cell types [41]. Theexpression of neuronal voltage-gated Na� channels is a criticalstage for the generation of action potentials [41]. More recently,voltage-dependent Na� currents have been used as an electro-physiological marker for assessing neuronal maturation of dif-ferentiating embryonic stem cell-derived NSCs toward the gen-eration of fully functional neurons [42]. Biella et al were able todemonstrate that there was a correlation between cell excitabil-ity and the Na� channel system during neuronal differentiation.Cells cultured in neuronal inductive media up to 21 days re-sulted in a shift in the steady-state inactivation, whereby agreater percentage of Na� current activation was available withgreater duration in neuronal inductive media. Furthermore, thesechanges in Na� channel system were associated with the abilityof differentiating NSCs to generate action potentials [42]. Thepresent study observed a correlation between the gene andprotein expression of mature neuronal markers, NF-M andNF-H, with a 8.5-fold increase in inward current following 3weeks of neuronal induction conditions, consistent with thefindings of Biella et al. [42]. However, undifferentiated DPSCswere also able to produce a minor inward current, implying thatDPSCs or a proportion of the total cell population may have apredisposition to differentiate into neuronal cell types.

Previous studies have demonstrated that cultured rat andhuman DPCs survived and expressed some neural-associatedmarkers when transplanted into the rodent brain [8, 9]. In thepresent study, the in vivo neural differentiation potential ofhuman adult DPSCs was assessed using an avian embryo modelas a host environment due to the absence of a functional immunesystem and because of the presence of active sites of earlyneuronal development. The chicken embryo has previously beenused to investigate cellular migration and differentiation usingquail-chicken [43] and mouse-chicken chimera [44] experi-ments and following transplantation of rat mesenchymal stemcells [45] and mouse embryonic endothelial progenitor cells intothe chicken embryo [46]. In the present study, GFP-labeledDPSCs were found to coexpress neuronal-associated markerswithin 48 hours after injection into 4-day-old chicken embryos,indicating a rapid response by DPSCs to endogenous environ-mental cues. It must be noted that the possibility of the neuronalmultiple neurite morphology being attributed to multiple GFP-expressing DPSCs was highly unlikely, as the figures representa composite of merged z-series confocal images, and individualimages from the z-series, with XZ and YZ axis (supplementalonline Figs. 3 and 4). Furthermore, the robustness and reliabilityof the avian xenogeneic assay was demonstrated with more than200 embryos injected with either adult human DPSCs, SHEDs,or HFFs, which survived for the duration of the experiments at24, 48, and 72 hours.

Importantly, the location of DPSCs within the embryo ap-peared to be essential for their specific neural morphology anddifferentiation capacity. DPSCs and SHEDs located near sen-sory TG neurons displayed a bipolar morphology characteristicof sensory neurons, whereas in the CNS, DPSCs exhibitedmultidendritic processes associated with motor neurons. Thesignificance of this observation suggests that transplanted hu-man adult DPSCs have a propensity to differentiate into CNS-type neurons when exposed to the correct environmental con-ditions. Furthermore, these DPSCs were not manipulated priorto transplantation, and therefore demonstrate their ability torespond directly to their surrounding environment. Although it

Figure 4. Dental pulp stem cells (DPSCs) continued to display aneuronal morphology consistent with the surrounding environment 72hours after injection in the avian embryo. (A–B): DPSCs survive andlocalize to regions of axonal processes near the hindbrain, displayingbipolar (asterisk) processes consistent with the surrounding neuronalcell. (C–D): DPSCs also localized the regions within the hindbrain;here, the DPSCs displayed a dendritic (arrowhead) morphology, con-sistent with the central nervous system. Abbreviations: E, eye; HB,hindbrain; Op, ophthalmic process; TG, trigeminal ganglion. Scale bar� 100 �m.


www.StemCells.com

is not common for CNC cells to exhibit CNS neural morphol-ogy, Ruffins et al have observed that neural crest cells exhibitthe capacity to differentiate and express markers specific forneurons of the CNS, when transplanted into the ventral neuraltube [47]. It has been proposed that neural crest and neural tubederivatives originate from the same precursor within the dorsalneural tube, which contributes to both CNS and PNS derivatives[48]. Since DPSCs retain neural crest properties in vitro [15], itis not surprising then, that DPSCs displayed characteristics ofboth PNS and CNS neurons in response to the environmentalcues when transplanted in vivo. Our study showed that DPSCswere able to survive and differentiate into neuronal derivativesand potentially integrate into neuronal networks within thechicken embryo up to 7 days after injection. In contrast, thecontrol cell type, HFFs, derived from the same germ layer asthe pulp tissue did not respond to the in vitro neuronal inductiveconditions or the in vivo microenvironment. Overall, thesestudies suggest that DPSCs are responsive to the surroundingmicroenvironment, surviving, migrating, and differentiating ac-cordingly into the appropriate cell types within the avian hostneural tissue. These observations confirm the specificity of theneuronal differentiation response by adult human DPSCs as

previously reported for SHEDs. The data suggest that DPSCsand SHEDs are appropriate candidates for further evaluation asstem cell therapy-based treatments using animal models repre-sentative of neurological diseases and injury.

ACKNOWLEDGMENTS

This study was supported in part by the Hanson Institute Re-search Bridging Grant, the National Health and Medical Re-search Council (Canberra, Australia) Project Grant no. 453497,and the Australian Research Council (Canberra, Australia) Spe-cial Research Centre for the Molecular Genetics of Develop-ment, grant number: S00001531. S.A.K. and S.G. contributedequally to this work.

DISCLOSURE OF POTENTIAL CONFLICTS

OF INTEREST

The authors indicate no potential conflicts of interest.

REFERENCES

1 Lindvall O, Kokaia Z. Recovery and rehabilitation in stroke: Stem cells.Stroke 2004;35(suppl 1):2691–2694.

2 Keirstead HS, Nistor G, Bernal G et al. Human embryonic stem cell-derivedoligodendrocyte progenitor cell transplants remyelinate and restore locomo-tion after spinal cord injury. J Neurosci 2005;25:4694–4705.

3 Lindvall O, Kokaia Z, Martinez-Serrano A. Stem cell therapy for humanneurodegenerative disorders—How to make it work. Nat Med 2004;10(suppl):S42–S50.

4 Lindvall O, Kokaia Z. Stem cells for the treatment of neurologicaldisorders. Nature 2006;441:1094–1096.

5 Thesleff I, Aberg T. Molecular regulation of tooth development. Bone1999;25:123–125.

6 Peters H, Balling R. Teeth. Where and how to make them. Trends Genet1999;15:59–65.

7 Gronthos S, Mankani M, Brahim J et al. Postnatal human dental pulpstem cells (DPSCs) in vitro and in vivo. Proc Natl Acad Sci U S A2000;97:13625–13630.

8 Miura M, Gronthos S, Zhao M et al. SHED: Stem cells from humanexfoliated deciduous teeth. Proc Natl Acad Sci U S A 2003;100:5807–5812.

9 Nosrat IV, Smith CA, Mullally P et al. Dental pulp cells provideneurotrophic support for dopaminergic neurons and differentiate intoneurons in vitro; Implications for tissue engineering and repair in thenervous system. Eur J Neurosci 2004;19:2388–2398.

10 Nosrat IV, Widenfalk J, Olson L et al. Dental pulp cells produceneurotrophic factors, interact with trigeminal neurons in vitro, and rescuemotoneurons after spinal cord injury. Dev Biol 2001;238:120–132.

11 Gronthos S, Cherman N, Robey PG et al. Human Dental Pulp StemCells: Characterization and Developmental Potential. Totowa, NJ: Hu-mana Press Inc., 2004.

12 Gronthos S, Brahim J, Li W et al. Stem cell properties of human dentalpulp stem cells. J Dent Res 2002;81:531–535.

13 Shi S, Gronthos S. Perivascular niche of postnatal mesenchymal stemcells in human bone marrow and dental pulp. J Bone Miner Res 2003;18:696–704.

14 Shi S, Robey PG, Gronthos S. Comparison of human dental pulp andbone marrow stromal stem cells by cDNA microarray analysis. Bone Dec2001;29:532–539.

15 Stokowski A, Shi S, Sun T et al. EphB/ephrin-B interaction mediatesadult stem cell attachment, spreading, and migration: implications fordental tissue repair. Stem Cells 2007;25:156–164.

16 Davidson RM. Neural form of voltage-dependent sodium current inhuman cultured dental pulp cells. Arch Oral Biol 1994;39:613–620.

17 Reynolds BA, Weiss S. Clonal and population analyses demonstrate thatan EGF-responsive mammalian embryonic CNS precursor is a stem cell.Dev Biol 1996;175:1–13.

18 Palmer TD, Markakis EA, Willhoite AR et al. Fibroblast growth factor-2activates a latent neurogenic program in neural stem cells from diverseregions of the adult CNS. J Neurosci 1999;19:8487–8497.

19 Itoh F, Nakane T, Chiba S. Gene expression of MASH-1, MATH-1,

Neurod and NSCL-2, basic helix-loop-helix proteins, during neural dif-ferentiation in P19 embryonal carcinoma cells. Tohoku J Exp Med1997;182:327–336.

20 Neuhoff S, Moers J, Rieks M et al. Proliferation, differentiation, andcytokine secretion of human umbilical cord blood-derived mononuclearcells in vitro. Exp Hematol 2007;35:1119–1131.

21 Hamburger V, Hamilton HL. A series of normal stages in the develop-ment of the chick embryo. J Embryol Exp Morphol 1951;88:49–92.

22 Nakayama T, Momoki-Soga T, Yamaguchi K et al. Efficient productionof neural stem cells and neurons from embryonic stem cells. Neuroreport2004;15:487–491.

23 Gallo R, Zazzeroni F, Alesse E et al. Ren: A novel, developmentallyregulated gene that promotes neural cell differentiation. J Cell Biol2002;158:731–740.

24 LeDouarin NM, Kalcheim C. From neural crest to the ganglia of theperipheral nervous system: the sensory ganglia. In: The Neural Crest. 2nded. Cambridge, U.K.: Cambridge University Press, 1999:153–196.

25 LeDouarin NM. The avian embryo as a model to study the developmentof the neural crest: a long and still ongoing story. Mech Dev 2004;121:1089–1102.

26 Thesleff I, Vaahtokari A. The role of growth factors in determination anddifferentiation of the odontoblastic cell lineage. Proc Finn Dent Soc1992;88(suppl 1):357–368.

27 Miletich I, Sharpe PT. Neural crest contribution to mammalian toothformation. Birth Defects Res Part C Embryo Today 2004;72:200–212.

28 Scintu F, Reali C, Pillai R et al. Differentiation of human bone marrowstem cells into cells with a neural phenotype: Diverse effects of twospecific treatments. Available at http://www.biomedcentral.com/1471-2202/7/14. Bmc Neurosci 2006;7:14.

29 Ciccolini F, Svendsen CN. Fibroblast growth factor 2 (FGF-2) promotesacquisition of epidermal growth factor (EGF) responsiveness in mousestriatal precursor cells: Identification of neural precursors responding toboth EGF and FGF-2. J Neurosci 1998;18:7869–7880.

30 Weiss S, Dunne C, Hewson J et al. Multipotent CNS stem cells arepresent in the adult mammalian spinal cord and ventricular neuroaxis.J Neurosci 1996;16:7599–7609.

31 Gritti A, Cova L, Parati EA et al. Basic fibroblast growth factor supportsthe proliferation of epidermal growth factor-generated neuronal precur-sor cells of the adult mouse CNS. Neurosci Lett 1995;185:151–154.

32 Hainfellner JA, Voigtlander T, Strobel T et al. Fibroblasts can expressglial fibrillary acidic protein (GFAP) in vivo. J Neuropathol Exp Neurol2001;60:449–461.

33 About I, Laurent-Maquin D, Lendahl U et al. Nestin expression inembryonic and adult human teeth under normal and pathological condi-tions. Am J Pathol 2000;157:287–295.

34 Jiang YQ, Oblinger MM. Differential regulation of beta III and othertubulin genes during peripheral and central neuron development. J CellSci 1992;103(pt 3):643–651.

35 Lopez-Picon FR, Uusi-Oukari M, Holopainen IE. Differential expressionand localization of the phosphorylated and nonphosphorylated neurofila-ments during the early postnatal development of rat hippocampus. Hip-pocampus 2003;13:767–779.


36 Lariviere RC, Julien JP. Functions of intermediate filaments in neuronaldevelopment and disease. J Neurobiol 2004;58:131–148.

37 Szebenyi G, Smith GM, Li P et al. Overexpression of neurofilament Hdisrupts normal cell structure and function. J Neurosci Res 2002;68:185–198.

38 Giasson BI, Mushynski WE. Study of proline-directed protein kinasesinvolved in phosphorylation of the heavy neurofilament subunit. J Neu-rosci 1997;17:9466–9472.

39 Shea TB, Dahl DC, Nixon RA et al. Triton-soluble phosphovariants ofthe heavy neurofilament subunit in developing and mature mouse centralnervous system. J Neurosci Res 1997;48:515–523.

40 Hille B. Ionic Channels of Excitable Membranes. 3rd ed. Sunderlend,MA: Sinauer Associates, 2001.

41 Kandel ER, Schwartz TM, Jessell TM. Principles of Neural Sciences. 4thed. New York: McGraw-Hill, 2000.

42 Biella G, Di Febo F, Goffredo D et al. Differentiating embryonic stem-derived neural stem cells show a maturation-dependent pattern of volt-age-gated sodium current expression and graded action potentials. Neu-roscience 2007;149:38–52.

43 LeDouarin NM, Kalcheim C. Methods for Identifying Neural Crest Cellsand Their Derivatives. In: The Neural Crest. 2nd ed. Cambridge, U.K.:Cambridge University Press; 1999:1–22.

44 Mitsiadis TA, Cheraud Y, Sharpe P et al. Development of teeth in chickembryos after mouse neural crest transplantations. Proc Natl Acad SciU S A 2003;100:6541–6545.

45 Pochampally RR, Neville BT, Schwarz EJ et al. Rat adult stem cells(marrow stromal cells) engraft and differentiate in chick embryos with-out evidence of cell fusion. Proc Natl Acad Sci U S A 2004;101:9282–9285.

46 Hatzopoulos AK, Folkman J, Vasile E et al. Isolation and characteriza-tion of endothelial progenitor cells from mouse embryos. Development1998;125:1457–1468.

47 Ruffins S, Artinger KB, Bronner-Fraser M. Early migrating neural crestcells can form ventral neural tube derivatives when challenged by trans-plantation. Dev Biol 1998;203:295–304.

48 Bronner-Fraser M. Molecular analysis of neural crest formation.J Physiol Paris 2002;96:3–8.

See www.StemCells.com for supplemental material available online.


www.StemCells.com

adult human dental pulp stem cells differentiate toward ... › images › pdf ›...

Documents