Cellular and Genetic Characterization of Human Adult Bone ... · [CANCER RESEARCH 63, 8877–8889, December 15, 2003] Cellular and Genetic Characterization of Human Adult Bone Marrow-Derived

[CANCER RESEARCH 63, 8877–8889, December 15, 2003]

Cellular and Genetic Characterization of Human Adult Bone Marrow-DerivedNeural Stem-Like Cells: A Potential Antiglioma Cellular Vector

Jeongwu Lee,1 Abdel G. Elkahloun,2 Steven A. Messina,1 Nicolay Ferrari,1 Dan Xi,1 Carolyn L. Smith,3

Ronald Cooper, Jr.,2 Paul S. Albert,4 and Howard A. Fine1

1Neuro-Oncology Branch, National Cancer Institute, National Institute of Neurological Disorders and Stroke, NIH; 2Cancer Genetics Branch, National Human Genome ResearchInstitute-NIH, Microarray Unit, National Institute of Neurological Disorders and Stroke, NIH; 3National Institute of Neurological Disorders and Stroke Light Imaging Facility,National Institute of Neurological Disorders and Stroke; and 4Biometric Research Branch, DCTD, National Cancer Institute, Bethesda, Maryland

ABSTRACT

We describe the in vitro isolation and expansion of cells capable offorming neurosphere-like aggregates from human adult bone marrow.Cells within these passaged spheroids can differentiate into astrocytes,specific neuronal subtypes, and oligodendrocytes and have gene expres-sion profiles similar to human fetal brain-derived neural stem cells. Ge-netically modified neural-competent bone marrow-derived cells efficientlymigrate toward distant sites of brain injury and tumor in vivo, where theydifferentiate and express therapeutic transgenes when transplanted intothe brains of mice. These studies suggest that adult bone marrow mayserve as a large reservoir for autologous neural stem-like cells for futuretherapeutic strategies.

INTRODUCTION

Malignant gliomas represent an important cause of cancer-relatedmortality for which standard treatments are suboptimal (1). Althoughgliomas do not generally metastasize, their propensity to deeplyinfiltrate adjacent cerebral cortex precludes definitive surgical resec-tion or other local therapies. Intrinsic resistance of glioma cells toradiation and chemotherapy limits traditional cancer therapies,whereas newer approaches such as gene therapy are confounded bythe inefficiencies of viral vectors to transduce the deeply infiltratingglioma cells. An autologous cellular vector that could migrate throughbrain parenchyma and deliver a therapeutic transgene to the site ofinfiltrating tumor would theoretically represent an optimal gene de-livery strategy for brain tumors. Neural stem cell (NSC) lines havebeen demonstrated to be effective for delivering transgenes to braintumors based on their unique migratory properties within the centralnervous system [CNS (2–4)]; however, the ability to safely obtainsufficient numbers of autologous NSCs represents a significant andpotentially prohibitive technical challenge. The demonstration thatfreshly isolated bone marrow cells injected into the brains of animalscan express neuronal and glial markers in mice suggests that tissue-specific stem cells possess broad developmental potential and raisesthe possibility of a readily available source of autologous NSC-likecells (5–15). Here we demonstrate the in vitro isolation, expansion,and therapeutic potential of a neural-competent population of cellsfrom adult human bone marrow.

MATERIALS AND METHODS

Cell Culture. RBCs from human adult (donors, 20–40 years old) bonemarrow cells purchased from AllCells (Berkely, CA) were depleted by densitygradient, and the remaining cells were plated in uncoated tissue culture plasticdishes. After 10 days of culture in 15% ES-qualified serum (Life Technologies,Inc., Gaithersburg, MD) containing media with basic fibroblast growth factor

[bFGF (50 ng/ml)]/fibroblast growth factor (FGF) 8 (100 ng/ml)/sonic hedge-hog (500 ng/ml)/leukemia inhibitory factor (10 ng/ml; from R&D Systems,Minneapolis, MN), nestin-positive cell aggregates were formed. Cells werethen trypsinized and replated in serum-free N2/B27 neurobasal media (LifeTechnologies, Inc.) with bFGF/FGF8/sonic hedgehog. Floating aggregateswere separated and cultured for 10 days. Brain-derived neural stem cells(BDNSCs) were obtained from Clonetics (Walkersville, MD). Conditionedmedia from BDNSC culture were filtered through 0.22 �m Stericup (Milli-pore, Bedford, MA) and used in the culture of BDNSCs and marrow-derivedneural-competent cells [MDNCCs (25% conditioned media �75% fresh me-dia)]. To induce differentiation in vitro, cell aggregates were plated on fi-bronectin-coated plates in B27 media containing 10% fetal bovine serum. Invivo conditioning consisted of the injection of cell aggregate-dissociated,nestin-positive cells into the lateral ventricle of neonatal severe combinedimmunodeficient mouse. Transplanted human cells were examined by confocalmicroscopy (Zeiss LSM 510; Zeiss). For ex vivo characterization, mice trans-planted with human cells were killed; brains were dissociated by digestion withtrypsin, collagenase, and papain; and then the harvested cells were cultured inthe presence of bFGF and differentiated by the removal of bFGF (16).

Immunohistochemistry. Immunohistochemistry was carried out usingstandard protocols. The following primary antibodies were used at the follow-ing dilutions: (a) TuJ1 rabbit polyclonal antibody, 1:2000; green fluorescenceprotein (GFP) monoclonal antibody, 1:200 (both from Babco, Richmond, CA);(b) TuJ1 mouse monoclonal antibody, 1:1000; MAP2 mouse monoclonalantibody, 1:2000; glial fibrillary acidic protein (GFAP) mouse monoclonalantibody, 1:400; 2�,3�-cyclic nucleotide 3�-phosphodiesterase monoclonal an-tibody, 1: 1000; and �-aminobutyrate rabbit polyclonal antibody, 1:1000 (allfrom Sigma, St. Louis, MO); (c) A2B5 monoclonal antibody, 1:100(BioTrend); (d) nestin polyclonal antibody, 1:1000; NeuN mouse monoclonalantibody, 1:100; GFAP rabbit polyclonal antibody, 1:400; human ribonucleo-protein monoclonal antibody, 1:50; HuMac mouse monoclonal antibody, 1:50;P1H12 monoclonal antibody, 1:100; and neurofilament M rabbit polyclonalantibody, 1:200 (all from Chemicon); (e) Ki67 rabbit polyclonal antibody(Novocastra); (f) GFP rabbit polyclonal antibody, 1:200 (Molecular Probes,Eugene, OR); (g) F4/80 rat monoclonal antibody, 1:50; S100� rabbit poly-clonal antibody, 1:100 (Abcam); (h) Glut1 rabbit polyclonal antibody, 1:1000(Calbiochem); (i) CD45 mouse monoclonal antibody, 1:100 (PharMingen), (j)O4 monoclonal antibody, 1:40 (Developmental Studies Hybridoma Bank); (k)aquaporin4 rabbit polyclonal antibody, 1:200 (Santa Cruz Biotechnology,Santa Cruz, CA); (l) tyrosine hydroxylase rabbit monoclonal antibody, 1:400(Pel-Freez, Rogers, AK); and (m) bromodeoxyuridine rat monoclonal anti-body, 1:100 (Accurate, Westbury, NY). For detection of primary antibodies,fluorescence-labeled secondary antibodies (Molecular Probes) were used ac-cording to the manufacturer’s recommendations.

Telomeric Repeat Amplification Protocol Assay. Telomeric repeat am-plification protocol assay was carried out using Telomerase PCR kit (Roche)following the manufacturer’s recommendations.

Microarray. Human cDNA microarray chips consisting of 13,826 se-quence-verified cDNA clones were printed onto glass slides as describedonline.5 Gene names are according to build 138 of the Unigene humansequence collection.6 Total RNA from all samples was extracted using TrizolReagent (Life Technologies, Inc.), and further purified using the RNeasy kit(Qiagen). One ug of total RNA was subjected to linear RNA amplificationusing the modified Eberwine protocol.7 Microarrays were hybridized and

Received 3/27/03; revised 8/27/03; accepted 10/8/03.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Howard A. Fine, Room 225, The Bloch Building (Building82), 9030 Old Georgetown Road, Bethesda, Maryland 20892. Phone: (301) 402-6298;Fax: (301) 480-2246; E-mail: [email protected].

5 http://www.nhgri.nih.gov/UACORE/.6 http://www.ncbi.nlm.nih.gov/UniGene/build.html.7 http://cmgm.stanford.edu/pbrown/protocols/ampprotocol_2.txt.

8877

Research. on June 9, 2017. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

scanned, and image analysis was performed as described online.1 Briefly,fluorescence-labeled cDNA was synthesized from 1 �g of sample or referencemRNA by random hexamers and oligo(dT)-primed reverse transcription in thepresence of either Cy3 or Cy5 fluor-dUTP (Amersham Pharmacia Biotech,Piscataway, NJ). All samples were compared with the same reference RNA toallow for normalization of each clone’s expression relative to the reference foreach sample. The reference consisted of RNA extracted from adult whole bonemarrow depleted of RBCs. Image analyses were performed using DeArraysoftware.8 The two fluorescent images (red and green channel) obtainedconstitute the raw data from which differential gene expression ratio valueswere calculated. The ratio normalization was performed based on all datapoints. All data from all experiments were then entered into a relationaldatabase using the Filemaker Pro software.

Microarray Analysis. All analyses were done on median centered log(base 2) expression ratios. Hierarchical clustering was performed based onPearson correlation and average linkage. We used multidimensional scaling(MDS) to graphically assess the similarity between groups. MDS was per-formed using Euclidean distance (root sum of squared differences) as thedistance metric.

Commonly overexpressed/underexpressed genes between the differentiatedand undifferentiated BDNSC groups were selected based on all samples eitherbeing overexpressed (over 1.8) or underexpressed (0.56); 133 genes wereidentified. We compared the median correlation between BDNSC andMDNCC groups with the median correlation between BDNSCs and othergroups as a measure of the closeness of BDNSCs and MDNCCs. A permuta-tion test (17) was used to assess the statistical significance of the differencebetween these two median correlations. The permutation P was obtained usinga null distribution estimated by scrambling individual labels for the control,differentiated MDNCC, and undifferentiated MDNCC groups, and recomput-ing the difference in the two median correlations (10,000 permutations weregenerated).

Genes that were differentially expressed between differentiated and undif-ferentiated BDNSCs were selected based on (a) all of the differentiated(undifferentiated) BDNSCs being overexpressed (over 1.8), and none of theundifferentiated (differentiated) BDNSCs being overexpressed (under 1.8),and (b) all of the differentiated (under-differentiated) BDNSCs being under-expressed (under 0.56), and none of the undifferentiated (differentiated)BDNSCs being underexpressed (over 0.56). There were 78 genes that met thiscriterion. We assessed the closeness of the expression patterns betweenMDNCC and BDNSC groups by averaging the median correlation betweenundifferentiated MDNCC and BDNSC samples and the median correlationbetween differentiated MDNCC and BDNSC samples. A permutation test wasused to test whether MDNCC and BDNSC expression patterns of the samedifferentiation status were more highly correlated than MDNCC and BDNCpatterns of a different differentiation status. The permutation P was obtainedusing a null distribution estimated by scrambling labels for the differentiatedand undifferentiated MDNCC samples and recomputing the average correla-tion measure described above.

In Vivo Models. The freezing insult was induced by stabbing the brain ofanesthetized animals with a Hamilton syringe chilled in liquid nitrogen understereotactic guidance (2 mm lateral and 1 mm anterior to bregma). CM-DiI(Molecular Probes)-labeled MDNCCs (2 � 105), suspended in 5 �l of HBSS,were then stereotactically implanted into a distant site within the ipsilateral orcontralateral hemisphere (2 mm lateral and 2.5 mm posterior to bregma; depthof 2.5 mm from dural surface; 5 mice/group). On the specified days, animalswere sacrificed, and brains were processed for immunohistochemical stainingand microscopy.

Experimental gliomas were produced by injecting U87-MG human gliomacells (5 � 105), suspended in 5 �l of HBSS, into the forebrain (2 mm lateraland 1 mm anterior to bregma; depth of 2.5 mm from dura) of female adultnu/nu mice over a 3–5-min period. Three days later, animals received a secondstereotactic injection (2 mm lateral and 2.5 mm posterior to bregma; depth of2.5 mm from dura) of genetically transduced MDNCCs.

In Vitro Migration Assays. In vitro migration assays were performedusing a 96-well microchemotaxis chamber with polyvinylpyrrolidone-freepolycarbonate filters with a pore size of 8 �m (Neuroprobe, Gaithersburg,

MD). Various concentrations of the cytokines (R&D Systems), U87-MG cells(2.5 � 104), or supernatant from homogenized tissues were placed in B27media (serum free, 0.1% BSA), and 29 �l of the solution were placed in thelower chamber. After placing the filter over the lower chamber, 2.5 � 104

MDNCCs, BDNSCs, or human bone marrow cells were resuspended andpipetted onto the surface of the filter. The apparatus was then incubated for 5 hat 37°C in a humidified chamber with 5% CO2 to allow cell migration. Afterthe incubation period, the filter was removed, and the upper side of the filterwas washed and scraped with a rubber policeman. The filters were fixed withmethanol and stained with Giemsa. We quantified migration by counting cellsin four random high-power fields (�400) in each well. Experiments wereconducted in quadruplicate.

RESULTS

Isolation of Neural Competent Cells in Vitro from Human BoneMarrow. We proceeded under the assumption that the bone marrowcontains a population of pluripotent stem cells that could be driventoward a cell type with NSC-like properties, given that neural fate isone of the principle default differentiation pathways of uncommittedmultipotent cells in vitro (15, 18, 19). To select for such an undiffer-entiated population of cells in vitro, we chose to culture adult bonemarrow under conditions that are conducive to the expansion ofpluripotent embryonic stem cells [Fig. 1A (18–20)]. Small clusters ofcells began to appear after 10 days in embryonic stem cell cultureconditions, at which point we switched the cells to serum-free growthconditions similar to those used for culturing NSCs [Fig. 1B (20)].The media contained basic FGF2, sonic hedgehog, and FGF8, cyto-kines known to be important both in proliferation and fate determi-nation of NSCs in early development (21, 22). The majority of theseclumped cells began to grow as nestin-positive spheroids with mor-phologies identical to typical BDNSC-derived neurospheres within 10days in NSC culture conditions [Fig. 1B, b�e (23)]. These floatingspheroids were first seen as 10–20-cell aggregates but then quicklygrew to clusters consisting of hundreds of floating cells, facilitatingtheir separation from cells that remained attached to the plates. Ap-proximately 2000 spheroids were typically generated from 107 humanadult bone marrow cells after 20 days in culture. Our cultures gener-ally yielded between 5 � 105 and 5 � 106 cells after 1 month inculture starting from a population of 10 million whole bone marrowcells. Cells from dissociated marrow-derived neurosphere-like aggre-gates continued to proliferate, reformed spheroids, and could bepassaged for more than 2 months.

We next sought to determine whether these bone marrow-derivedcellular aggregates could be induced to differentiate into cells ofneural lineage, given their morphological similarity to NSC-derivedneurospheres. After the addition of serum and the removal of growthfactors, spheroid-derived cells attached to the polyornithine/fibronec-tin-coated plates and spread out (Fig. 1B). Some nestin-positive cellsalso became positive for A2B5, a marker for glial progenitor cells,during the first few days in the differentiation culture conditions (Fig.1B, g). Immunohistochemical staining of cells after 7 days in differ-entiation conditions demonstrated that 10–15% of cells were positivefor TuJ1 (neuron-specific � III tubulin), whereas 50–80% of cellswere positive for GFAP, which are markers for neuronal and glialdifferentiation, respectively. (Fig. 1B, h�l). GFAP cells were alsopositive for S100� staining, another marker of astrocytic differentia-tion (Fig. 1B, i). All cells were negative for HuMac (marker for humanmacrophages) and CD45 (marker for differentiated hematopoieticcells), respectively (Fig. 1B, f and h). Despite the abundance of cellswith astrocytic and neuronal morphology and immunophenotypes, wesaw only a few O4-positive cells with morphology consistent witholigodendrocyte differentiation (data not shown).

We performed single cell limiting dilution assays to evaluate the8 http://www.nhgri.nih.gov/DIR/microarray.

8878

MARROW-DERIVED NEURAL STEM-LIKE CELLS AS THERAPEUTIC VECTORS



clonogenicity of individual cells derived from our cellular aggregates.Growth of individual cells derived from dissociated spheroids metwith limited success; however, clumps of a minimum of 10 cells in a96-well microtiter plate were sufficient to generate cellular aggregatescharateristic of neurospheres in 23 � 3.2% of the wells (89 of 384wells). To evaluate the growth and differentiation potential of singlecells, we performed mixing experiments whereby single cells express-ing GFP from a dissociated GFP-transduced spheroid were mixedwith 9 cells from non-GFP-expressing spheroids. We observed GFP-positive spheroids forming in 2.1 � 0.85% of the microtiter wells (6of 288 wells). Both GFP-expressing astrocytes and neurons wereformed after exposure of these GFP-positive spheroids to differenti-ation conditions. We therefore estimate that approximately 2% of thecells within our bone marrow-derived spheroids are clonogenic andcapable of differentiating toward both glial and neuronal lineage.

Further evidence for the stem cell-like nature of the cells within thebone marrow-derived neurosphere-like aggregates was their expres-sion of telomerase, as demonstrated in various other stem cell popu-lations (16, 24). By contrast, the differentiated progeny of thesespheroids were negative for telomerase expression (Fig. 1C). Based

on these in vitro data, we named these cells MDNCCs. All MDNCCsand their differentiated progeny demonstrated a normal 2N humankaryotype (data not shown).

In Vivo Differentiation of MDNCCs. We transduced MDNCCswith GFP and injected them (50,000 GFP-expressing cells) into theventricles of neonatal (P1 to P3) immunodeficient severe combinedimmunodeficient mice (25, 26) to evaluate the effects of in vivoconditioning on the differentiation of MDNCCs. We subsequentlyharvested the brains and performed immunohistochemistry on brain-derived single cell suspensions (Fig. 2A) or on whole brain mounts(Fig. 2B) at various time points after injection (5 days, 10 days, 1month, and 2 months). Examination of the single cell suspensionsdemonstrated the presence of GFP-expressing TuJ1/neurofilamentM-, GFAP-, and O4-positive cells with morphology consistent withneuronal, astrocytic, and oligodendroglial differentiation, respectively(Fig. 2A). We next evaluated the whole brain sections of the 1-month-old MDNCC intracerebrally injected mice by exclusively using con-focal microscopy to unambiguously identify labeled cells in situ.Approximately 32.9% of the GFP-positive cells coexpressed GFAP(342 of 1038 cells counted), whereas 6.5% of the GFP-expressing

Fig. 1. Marrow-derived neural-competent cells (MDNCCs) are phenotypically similar to the neural stem cells (NSCs) in vitro. A, protocol schema for generating MDNCCs in vitro.B, light photomicrographs of cellular aggregates after in vitro bone marrow cultures as diagrammed in A: a, day 1; b, day 18; c, day 35; nestin staining (red) of fresh bone marrowcells (d) and cell aggregates [e; blue 4�,6-diamidino-2-phenylindole (DAPI) staining is shown in the inset]. Differentiation of cellular aggregates plated on fibronectin: f, nestin (red)and HuMac (green); g, A2B5 (red), nestin (green), and DAPI (blue); h, glial fibrillary acidic protein (GFAP; red), CD45 (green), and DAPI (blue); i, S100� (green) and GFAP (red);j, TuJ1 (red) and DAPI (blue); k, TuJ1 (green) and GFAP (red; �200); and l, TuJ1 (green), GFAP (red), and DAPI (blue); �400. C, telomeric repeat amplification protocol analysisfor telomerase activation. Cell lysates extracted from 10,000 cells were used in each PCR reaction. Arrow shows the internal control for PCR. Lane 1, undifferentiated Brain-derivedneural stem cells (BDNSCs); Lane 2, differentiated BDNSCs; Lane 3, fresh bone marrow cells; Lane 4, bone marrow cells cultured for 3 weeks in 10% fetal bovine serum; Lane 5,undifferentiated MDNCCs; Lane 6, differentiated MDNCCs. Typical ladder patterns are only shown in undifferentiated BDNSCs and MDNCCs.

8879




cells coexpressed the neuronal marker NeuN (49 of 757 cells). Im-munostained whole brain mounts revealed GFP-expressing cells withimmunophenotypes and morphologies consistent with glial and neu-ronal subtypes including GFAP-positive astrocytes, CNPase-positiveoligodendrocytes, and neurons positive for TuJ1, NeuN, GABA, andtyrosine hydroxylase expression (Fig. 2B). In addition, we foundnumerous GFP-positive cells located within the ventricular liningzone. We did not detect GFP-positive glial and neuron-like cells whenwe injected either 50,000 or 500,000 fresh bone marrow cells (datanot shown).

Global Gene Expression Profiling of MDNCCs. To first confirmthe neural lineage of the spheroids containing MDNCCs and examinethe similarity between BDNSC-derived neurospheres and MDNCC-

derived spheroids during the process of differentiation, we character-ized the global gene expression profiles of these two populations ofcells using cDNA microarrays. Total RNA was isolated from differentpopulations of cells, and the gene expression pattern of 13,826 geneswas analyzed (27, 28). MDS analysis and hierarchical clustering wereperformed to compare gene expression profiles of undifferentiatedand differentiated BDNSCs with MDNCCs as well as with other bonemarrow-derived and other unrelated cell populations [Fig. 3 (29, 30)].For the first analyses, we sought to identify a group of genes withexpression profiles representative of the neural lineage to confirm theneural characteristics of MDNCCs. We reasoned that the genes in theundifferentiated and differentiated BDNSC groups demonstrating themost extreme concordant gene regulation compared with a control

Fig. 2. In vivo differentiation of marrow-derived neural-competent cells (MDNCCs). In vitro generated, green fluorescence protein (GFP)-transduced cellular aggregates weretransplanted into the intracerebral ventricles of P1-P3 mice; and brains were harvested. A, single cell isolates from brain demonstrating (a) neuronal differentiation; (b) astrocyticdifferentiation; and (c) oligodendroglial differentiation. White arrows represent GFP-expressing human cells positive for the lineage selective marker. B, confocal microscopy fromwhole brain mounts of differentiated neuronal subtypes and astrocytes in situ. a, nestin (red)-positive MDNCCs in ventricular zone; b�d, neuronal differentiation of MDNCCs in cortex;e, tyrosine hydroxylase (TH)-positive MDNCCs in striatum; f, GFAP-positive MDNCCs along the falx in the frontal cortex; g, CNPase-positive MDNCCs in corpus callosum.

8880




nonneural control cell would define such a neural-selective gene set.We identified 133 such genes and compared them across all groups.We found a high degree of similarity in gene expression patternsbetween the BDNSC and the MDNCC groups as compared with theother cell populations [Fig. 3B (median correlation between BDNSCsand MDNCCs � 0.872; median correlation between BDNSCs andother groups � 0.628, P � 0.001)].

The pattern of gene expression necessary to differentiate a stem cellto a specific terminally differentiated cell must involve a unique andregulated sequence of events. Thus, for our second analyses, we askedwhether the gene expression patterns seen during the differentiation ofBDNSC-derived neurospheres were similar to those of MDNCC-derived spheroids. We identified 78 genes that were highly up- ordown-regulated during the process of BDNSC differentiation, based

Fig. 2. Continued.

8881




on the assumption that a subgroup of these genes may be coregulatedin cells with NSC-like potential. When we performed hierarchicalclustering of our different population of cells using these 78 genes, wefound that the undifferentiated MDNCC spheroids were closely cor-related with undifferentiated BDNSC neurospheres, and the differen-tiated MDNCCs showed strong correlation with the differentiatedBDNSC populations (Fig. 3C). Clustering analyses graphically dem-onstrate the relatedness of these groups compared with each other andall other control cells (Fig. 3C). As expected, analyses of the expres-sion data using MDS revealed little similarity between undifferenti-ated and differentiated BDNSC-derived neurospheres because the 78genes used for this analysis were selected based on their differentialexpression between these two populations of cells. By contrast, theMDS demonstrated a strong similarity between the undifferentiatedMDNCC spheroids and undifferentiated BDNSC-derived neuro-spheres [Fig. 3C (median correlation between samples � 0.84)] andbetween the differentiated MDNCC spheroids and the differentiatedBDNSC neurospheres (median correlation between samples � 0.83).

The average of these two median correlations is 0.836, which issubstantially higher than a similar correlation measure relatingMDNCC spheroids and BDNSC neurospheres of different differenti-ation status (average correlation of 0.579), suggesting a high degree ofsimilarity between these two cell populations relative to their differ-entiation status (P � 0.01 by a permutation test).

In both analyses, we obtained similar results when we selectedgenes based on t-values and average log expression values with thelargest magnitude, although the weighted genes were different (datanot shown). As strong as these correlations are, however, some genesdid have different expression patterns between the MDNCC andBDNSC groups. It remains to be determined whether these differencessimply reflect differences in the conditions used for in vitro acquisi-tion and propagation of cells or contamination of a minor cell popu-lation (i.e., hematopoietic and/or stromal progenitors) or rather rep-resent intrinsic differences between NSC-like populations. Takentogether, these microarray data suggest that MDNCC-derived sphe-

Fig. 3. Gene expression analyses of bone mar-row-derived and neural stem cells (NSCs). A, de-scription of cells used. B, table of correlation coef-ficients for analyses of 133 weighted neurallineage-selective genes. The degree of similaritybetween groups is represented as a number between�1 and 1, with 1 being identical. Blue box repre-sents coefficients � 0.85. Connected bars, as de-termined by hierarchical clustering analysis, repre-sent the similarity between groups of cells. Note thehigh degree of similarity in gene expression pat-terns between all of the brain-derived neural stemcell (BDNSC) and the marrow-derived neural-com-petent cell (MDNCC) groups, regardless of differ-entiation status, as compared with the other cellpopulations. C, analyses of 78 weighted genes thatdemonstrated significant change during BDNSCdifferentiation. a, hierarchical clustering analysisusing 78 weighted genes and their respective cloneIDs that demonstrated significant change duringBDNSC differentiation (see text). Expression levelsof specific genes compared with reference RNA arevisualized as red (up-regulation) or green (down-regulation). b, table of correlation coefficients formultidimensional scaling (MDS) analysis. The de-gree of similarity between groups is represented asa number between �1 and 1, with 1 being identical.Blue box represents coefficients � 0.85. Note thatthere is little similarity between undifferentiated(red) and differentiated (yellow) BDNSCs. By con-trast, the MDS demonstrated a strong similaritybetween the undifferentiated BDNSCs (red) andundifferentiated MDNCCs (pink) and between thedifferentiated BDNSC spheroids (yellow) and thedifferentiated MDNCCs (light yellow). c, visualiza-tion of MDS.

8882




roids have genetic profiles and neural differentiation potential verysimilar to those of BDNSC-derived neurospheres.

In Vivo Migration of MDNCCs in Freeze Injury Model. Wewere interested in determining whether MDNCCs have migratoryproperties similar to NSCs, given the known ability of such cells tomigrate within the brain in response to physiological and pathologicalstimuli [i.e., the rostral migratory stream and trauma, respectively (31,32)]. We therefore investigated the migratory properties of theMDNCCs using an in vitro migration assay by using lysates fromnormal and freeze-injured brain tissue. Similar to BDNSCs, MDNCCsmigrated toward freeze-injured brain tissue lysates to a far greaterextent than to normal brain lysates (�7-fold increase; Fig. 4A). Bycontrast, our starting population of whole bone marrow cells demon-strated significantly less migration (�2-fold increase; P � 0.001, ttest). The effect of time from injury on the migration of cells withineach group (fresh bone marrow, BDNSCs, and MDNCCs) was ana-lyzed using a two-way ANOVA. The effect of time on migrationvaried significantly between the different groups when all groupswere considered [F � 17.41; degree(s) of freedom (df) � 8,45;P � 0.001]. However, there was no significant variation in the effectof time on cell migration when only the BDNSC (F � 1.62; df � 4,30;

P � 0.2) and MDNCC (F � 2.23; df � 1,34; P � 0.14) groups wereanalyzed, suggesting no strong differences in temporal migrationpatterns between BDNSCs and MDNCCs.

To evaluate the response of MDNCCs to brain injury in vivo, westereotactically injected CM-DiI (fluorescent red)-labeled MDNCCsat a distance from the site of freeze-injury. Seven days after injury,fluorescence-labeled cells from both the ipsilateral and contralateralMDNCC injection sites were found within the area of the injuredbrain (Fig. 4B). In all of the animals we examined (n � 5 MDNCCsand 5 controls), the migration patterns of the MDNCCs were largelythrough white matter tracts, including the corpus callosum and exter-nal capsule. MDNCCs migrated in a chain-like formation, character-istic of the pattern of NSC migration seen in the rostral migratorystream [Fig. 4C (31)]. Human ribonucleoprotein-specific antibodystaining colocalized with the DiI-labeled cells, confirming that thelabeled cells were of human origin. The transplanted MDNCCs werefound adjacent to reactive astrocytes, which were identified by intenseGFAP and nestin immunopositivity (33). Immunostaining of GFP-expressing MDNCCs at the trauma site also revealed dual nestin- andGFAP-positive cells, demonstrating the ability of the MDNCCs todifferentiate along a glial lineage in response to brain injury (gliosis).

Fig. 3. Continued.

8883




This pattern of glial differentiation and cellular migration was mir-rored identically by the BDNSCs, in contrast to the starting populationof freshly isolated bone marrow cells that demonstrated little migra-tion or glial differentiation (data not shown).

Migration and Differentiation of MDNCCs in Tumor Model.To further confirm the migratory properties of the MDNCCs, weinvestigated their ability to migrate toward intracranial gliomas, as hasbeen reported previously for a murine NSC line (2, 3). Adenovirus-transduced MDNCCs expressing GFP were injected distally (3.5 mm)into the ipsilateral cortical hemisphere of mouse brains harboring anintracerebral established U87-MG human glioblastoma. Nearly all ofthe GFP-positive cells migrated toward the established U87 gliomawithin 12–20 days of injection (Fig. 5). More than 90% of the labeledMDNCCs were found at the interface of the tumor mass and “normal”brain (an area known to contain infiltrative tumor cells) rather thanwithin the center of the tumor (Fig. 5). A similar pattern of MDNCCmigration was observed in the brains of animals harboring rat glio-blastoma CNS-1 tumors (data not shown). We performed in vitro

assays to better understand this pattern of migration. Normal brainlysates or lysates from U87-MG gliomas implanted intracranially ors.c. induced only minimal migration of MDNCCs and BDNSCs invitro. Likewise, cultured U87-MG cells, vascular endothelial growthfactor, and bFGF did not stimulate significant cellular migration. Bycontrast, lysates of brain tissue at the edge of the invading tumorborder induced migration of both MDNCCs and BDNSCs in vitro byan order of magnitude greater than any other stimuli (Fig. 5A). Usinga two-way ANOVA, we found a highly significant difference in thecell migration pattern between the different populations of cells(BDNSCs, MDNCCs, and fresh bone marrow) when all three groupswere included (F � 12.02; df � 10,54; P � 0.001). There were nosignificant differences, however, when only BDNSCs (F � 1.21;df � 5,36; P � 0.33) and MDNCCs (F � 2.11; df � 1,41; P � 0.15)were analyzed, suggesting the similarity between BDNSCs andMDNCCs to chemotaxic stimuli. These data further demonstrate thesimilar migratory properties between MDNCCs and BDNSCs. Inaddition, the in vitro and in vivo data suggest that brain tissue adjacent

Fig. 3C. Continued.

8884




to or infiltrated by tumor results in the stimulus for migration ofMDNCCs and BDNSCs rather than the tumor itself. The mechanismfor this migratory stimulus remains unknown, although we suspectthat the destructive and/or inflammatory environment at the edge ofthe advancing tumor might be the stimulus for migration, consistentwith our demonstration that brain injury is a potent stimulus forMDNCC and BDNSC migration.

We performed immunohistochemical studies of the MDNCCs thathad migrated toward the intracerebral glioma to reveal their differen-tiation fate (Fig. 5B). Whereas numerous tumor cells stained posi-tively for the proliferation marker Ki67, GFP-expressing MDNCCswere Ki67 negative, suggesting that these cells are largely nonprolif-erative in vivo. MDNCCs were also negative for Glut1 and HuMac,markers for endothelial cells and human macrophage cells, respec-tively. Within the border between tumor and cerebral cortex, however,were large numbers of GFP-expressing MDNCCs that stained positivefor GFAP. Consistent with these findings, the tumor-bearing hemi-spheres showed increased GFAP immunoreactivity characteristic ofreactive gliosis. These cells constituted approximately 72% of allGFP-expressing cells. Some cells coexpressing GFP and GFAP werealso positive for nestin, aquaporin 4, and S100�, consistent with areactive astrocyte phenotype (33). These cells coexpressing GFAPand nestin constituted approximately 30–40% of all of the MDNCCs.

Approximately 2.5% of the MDNCCs stained positive for NeuN.Finally, approximately 20% of MDNCCs expressed no identifiableneural, migroglial, or hematopoietic markers (data not shown). Thereactive astrocyte phenotype observed around the tumor was notdetected in the brains of either normal adult mice or neonatal mice thatreceived injection with MDNCCs. Taken together, these data suggestthat MDNCCs not only migrate toward intracerebral gliomas but alsorespond to environmental cues.

MDNCCs as a Therapeutic Cellular Vector. We chose to ex-plore the potential therapeutic implications of the MDNCC migratoryproperties by evaluating the ability of these cells to carry therapeuticgenes to sites of intracerebral gliomas. The decision to geneticallymodify the MDNCCs to express antitumor transgenes that are non-toxic to normal cells was based on the migratory pattern of MDNSCthrough the CNS, as well as the infiltrative nature of malignantgliomas into normal brain tissue. We have demonstrated previouslythat a secreted form of platelet factor 4 (PF4), an antiangiogenicprotein, could be safely expressed within normal cerebral tissue,thereby inhibiting tumor angiogenesis and tumor growth (34). Thus,MDNCCs transduced by an adenoviral vector encoding the sPF4transgene were stereotactically injected into an intracerebral site dis-tant from the pre-established U87-MG glioblastomas. Animals treatedwith sPF4-transduced MDNCCs survived significantly longer than

Fig. 4. In vitro and in vivo migratory properties ofmarrow-derived neural-competent cells (MDNCCs).A, in vitro migration assay. Lysates of in vivo freeze-injured brain tissue taken at various time points afterinjury were used as chemoattractants. B, a, sagittalview of CM-DiI dye-labeled MDNCC (red) migra-tion toward injury in vivo after injection into brain ata distance from injury site. b, coronal view of injurysite as seen by up-regulation of glial fibrillary acidicprotein (GFAP; left); DiI-labeled MDNCCs areshown at injury site at day 7 after injection in wholemount brain section (middle) and in higher-powersection (�400; right). c, CM-DiI-labeled MDNCCsmigrating to injury site along the corpus callosumand striatum. HP, hippocampus; CC, corpus callo-sum, ST, striatum.

8885




mice treated with the MDNCCs transduced by a GFP-expressingcontrol adenoviral vector or mice with tumor only (P � 0.001,log-rank test; Fig. 6A, a). The reduced microvascular density in thePF4-transduced MDNCC-treated animals is consistent with an anti-angiogenic mechanism of tumor growth inhibition (Fig. 6A, b and c).

To further evaluate the potential utility of MDNCCs as a tumor-targeting vector, we examined the efficacy of transducing MDNCCswith a transgene encoding tumor necrosis factor-related apoptosis-inducing ligand (TRAIL). TRAIL induces apoptotic cell death inmany tumor cell lines but generally spares normal cells (35, 36).Transduction of normal cells, including MDNCCs, by an adenoviralvector expressing a human TRAIL cDNA results in potent tumor cellkilling through normal cell-mediated presentation of membrane-bound TRAIL to the tumor cell (36). Injection of either an excess orequal number TRAIL-transduced MDNCCs directly into the site oftumor cell implantation at the time of initial tumor cell implantationresulted in a very large and significant survival advantage comparedwith control vector-transduced MDNCCs (data not shown). Suchexperiments, however, do not reflect the reality of the clinical situa-tion, where there generally exists a pre-established tumor with largenumbers of distantly infiltrating tumor cells that far exceed the num-ber of “therapeutic” or “targeting cells” that could be practicallyadministered. We therefore stereotactically injected a log fewerTRAIL-transduced MDNCCs than tumor cells into the brain at adistance from a pre-established and growing intracerebral gliomas toexperimentally mirror the clinical scenario more precisely. TRAIL-transduced MDNCCs mediated a significant prolongation of survivalcompared with control-transduced MDNCCs, even under these mostdisadvantageous yet clinically relevant conditions (log-rank test:P � 0.017 for a and P � 0.001 for b; Fig. 6B). We observed increasedtumor cell apoptosis in the MDNCC/TRAIL-treated tumors, consist-ent with a TRAIL-mediated antitumor effect (Fig. 6B, c and d).Furthermore, increasing the ratio of TRAIL-transduced MDNCCs totumor cells resulted in an even longer prolongation of survival (Fig.6B).

Finally, immunodeficient mice (n � 3) transplanted with MDNCCsremained tumor free and were clinically and histologically normalwhen sacrificed more than a year after transplantation, demonstratingthe overall safety and lack of tumorigenic potential of the MDNCCs.

DISCUSSION

A number of groups have recently demonstrated that bone marrowcells could be differentiated into neurons and glial cells in vitro and invivo based on immunohistochemical studies (5, 6, 37, 38). We havefurthered those observations by demonstrating that adult human bonemarrow-derived neural progenitor-like cells can be propagated invitro, can differentiate into different subtypes of neuronal and glialcells, and have gene expression profiles as well as migratory functionssimilar to those of endogenous brain-derived NSCs. Furthermore, wehave demonstrated that these MDNCCs can be effectively used ascellular vectors to deliver therapeutic genes to intracranial tumors inan animal model of glioblastoma.

We believe that our data have several interesting implications forautologous stem cell-based therapeutics, potentially avoiding many ofthe difficult issues surrounding the use of adult brain, fetal, or em-bryonic stem cell-derived NSCs (39–41). For purposes of exploringone such potential therapeutic application, we have focused our initialexperiments on the use of MDNCCs as a cellular targeting vector.Cell-based gene transfer offers the potential advantage of controlled invitro cellular transduction, thereby increasing the possibility for highcellular transduction efficiency while lessening the potential toxicityassociated with direct vector administration to patients. MDNCCsappear particularly well suited as a cellular vector for CNS genedelivery given (a) their ability to be expanded in vitro, (b) their abilityto respond to environmental cues based on three different in vivomodels (the developing, the injured, and the tumor-containing brain),(c) their apparent lack of neurotoxicity or tumorigenicity, (d) the easewith which they are transduced by genetic vector in vitro, and (e) theready availability of a nearly limitless supply of autologous MDNCCsfor every patient. In particular, MDNCC-based gene transfer may beparticular promising for therapeutic strategies aimed at malignantgliomas, given their ability to migrate to sites of tumor infiltrationwithin the brain, thereby potentially overcoming the current limitationof most direct gene transfer approaches. Furthermore, the ability ofMDNCCs to migrate toward sites of brain injury around tumor or sitesof trauma and differentiate in vivo into both glial and neuronalsubtypes raises the intriguing prospect of using MDNCCs for CNSrepair, although many more studies remain to be done before such a

Fig. 5. In vivo migration and differentiation ofmarrow-derived neural-competent cells (MDNCCs)in tumor-bearing brain. A, migration of MDNCCsto intracerebral glioma. a, schematic depiction ofMDNCC site of injection relative to tumor. Left,whole mount brain sections (light photographed in1, costained for MAP2 in 2) demonstrating anatom-ical location of green fluorescence protein (GFP)-expressing MDNCCs to tumor periphery. Right,higher power cross-section (�100) demonstratingGFP-transduced MDNCCs surrounding the intrac-erebral glioma 12 days after injection into a distantsite within the brain. GFP-expressing cells werevisualized by anti-GFP antibody detected by 3,3�-diaminobenzidine. Arrows demonstrate tumor bor-der. Tu, tumor. b, in vitro migration assay. Basicfibroblast growth factor (highly overexpressed bygliomas) and lysates from either glioma cells, s.c.glioma, normal brain, or brain/tumor border wereused as chemoattractants. B, differentiation ofMDNCCs at intracerebral tumor site. Tumor andarea of brain surrounding tumor were visualized forGFP (MDNCC), 6-diamidino-2-phenylindole (cel-lular nuclei, pseudocolored), and either Ki67 (cel-lular proliferation; a), Glut1 (endothelial marker;b), HuMac (human microglia/macrophage marker;c), F4/80 (mouse microglia/macrophage marker;inset in c), Nestin (d), AQP4 (astrocytic; e), S100�(astrocytic; f), or glial fibrillary acidic protein (as-trocytic; g) coexpression. Bar, 100 �m.

8886




Fig. 5. Continued.

8887




prospect can be realistically entertained. These studies will need toaddress important questions such as whether the neurons generated bythe MDNCCs form appropriate and functional synaptic connections invivo, whether MDNCC-generated oligodendrocytes contribute to there-myelination of de-myelinated neurons, and whether MDNCC-gen-erated astrocytes are capable of constructing an in vivo niche condu-cive to neuronal regeneration and repair of a functional blood-brainbarrier (42–44).

A number of groups have demonstrated that cells from a specificdifferentiated tissue type can be induced to express phenotypic mark-ers and/or have functional properties consistent with differentiatedcells from a different lineage (5–15). One potential interpretation ofthese observations is that tissue-specific stem cells possess the abilityto “transdifferentiate” toward other fates. Several recently publishedreports, however, have failed to duplicate these observations, raisingthe possibility that the apparent plasticity of the tissue-specific stemcells may be a function of a contaminating, yet to be characterized raremultipotent cell population (45–47). We were unable to convincinglydemonstrate the presence of NSC-like cells from fresh, unculturedhuman bone marrow. By contrast, we do demonstrate that adulthuman bone marrow-derived neural progenitor-like cells can be prop-agated in vitro under selective growth conditions. These data areconsistent with the presence of and in vitro selection for an otherwise

rare bone-marrow derived cell with neural lineage competent proper-ties. Formal testing of this hypothesis, however, awaits the prospec-tive isolation and identification of the clonal cell of origin.

Experimental approaches addressing issues regarding stem cell fatedetermination have been largely based on morphological and immu-nohistochemical studies to date (42–47). We believe the results of ourgene expression profiling, although preliminary, demonstrate thepower of a genetic approach for determining similarities between stemcell populations and lineages and may be useful as a starting point forunderstanding the genetic program responsible for the induction ofNSC lineage commitment. The potential for these genetic approacheshas been demostrated in recent studies describing transcriptionalprofiling of embryonic, neural, and hematopoietic stem cells (48–50).These reports demonstrate that NSCs and hematopoietic stem cellsshare some of the same “stem cell” genes, yet not unexpectedly,express many distinct genes. Our data showing that the gene expres-sion profiles of MDNCCs are much closer to those of BDNSCs thanthose of hematopoietic stem cells (CD34-positive cells) and mesen-chymal stem cells (stroma cells) are of interest and further attest to theneural lineage-like properties of the MDNCCs.

In conclusion, we have demonstrated that human neural progenitor-like cells, generated from adult bone marrow cells, can be partiallypurified and propagated in vitro and used as cellular targeting vectors

Fig. 6. Marrow-derived neural-competent cells(MDNCCs) as tumor-targeting cellular vectors.Animals with pre-established intracerebral gliomasunderwent stereotactic injection into a distant sitewithin the brain with MDNCCs expressing thera-peutic genes. A, a, survival of animals injected withMDNCCs transduced with Ad-green fluorescenceprotein [GFP (control)] or Ad-platelet factor 4 [PF4(n � 7)]. b, diminished intratumor microvasculardensity (endothelial cells visualized with P1H12antibody) in animals given MDNCCs expressingPF4 compared with GFP. c, quantitation of vesselcount. Vessel counts were assessed by countingP1H12-positive cells within tumor of brain sections(n � 3/group; P � 0.001). B, survival of animalsinjected with Ad-GFP (control) or Ad-tumor necrosisfactor-related apoptosis-inducing ligand (TRAIL)-transduced MDNCCs (n � 8, except n for MDNCC/GFP in b � 5). The ratio between the number ofinitially injected tumor cells and effector cells is 10:1(a) and 1:1 (b). c, TUNEL staining of brain sections inanimals given MDNCCs expressing either GFP orTRAIL. d, quantitation of TUNEL-positive cells ofbrain sections in MNDCC/GFP or TRAIL (n � 3/group; P � 0.001).

8888




within the CNS. The data presented here suggest that this nearlyunlimited reserve of autologous cells might one day prove useful forthe treatment of malignant brain tumors and possibly even otherneurological diseases, should the MDNCC-derived cells ultimately bedemonstrated to have physiological functions comparable with thoseof cells derived from NSCs of the CNS.

REFERENCES

1. Maher, E. A., Furnari, F. B., Bachoo, R. M., Rowitch, D. H., Louis, D. N., Cavenee,W. K., and DePinho, R. A. Malignant glioma: genetics and biology of a grave matter.Genes Dev., 15: 1311–1333, 2001.

2. Benedetti, S., Pirola, B., Pollo, B., Magrassi, L., Bruzzone, M. G., Rigamonti, D. etal. Gene therapy of experimental brain tumors using neural progenitor cells. Nat.Med., 6: 447–450, 2000.

3. Aboody, K. S., Brown, A., Rainov, N, G., Bower, K. A., Liu, S., Yang, W. et al.Neural stem cells display extensive tropism for pathology in adult brain: evidencefrom intracranial gliomas. Proc. Natl. Acad. Sci. USA, 97: 12846–12851, 2000.

4. Ehtesham, M., Kabos, P., Gutierrez, M. A. R., Chung, N. H. C., Griffith, T. S., Black,K. L., and Yu, J. S. Induction of glioblastoma apoptosis using neural stem cell-mediated delivery of tumor necrosis factor-related apoptosis-inducing ligand. CancerRes., 62: 7170–7174, 2002.

5. Mezey, E., Chandross, K. J., Harta, G., Maki, R. A., and McKercher, S. R. Turningblood into brain: cells bearing neuronal antigens generated in vivo from bone marrow.Science (Wash. DC), 290: 1779–1782, 2000.

6. Brazelton, T. R., Rossi, F. M., Keshet, G. I., and Blau, H. M. From marrow to brain:expression of neuronal phenotypes in adult mice. Science (Wash. DC), 290: 1775–1779, 2000.

7. Krause, D. S., Theise, N. D., Collector, M. I., Henegariu, O., Hwang, S., Gardner, R.,Neutzel, S., and Sharkis, S. J. Multi-organ, multi-lineage engraftment by a single bonemarrow-derived stem cell. Cell, 105: 369–377, 2001.

8. Orlic, D., Kajstura, J., Chimenti, S., Jakoniuk, I., Anderson, S. M., Li, B. et al. Bonemarrow cells regenerate infarcted myocardium. Nature (Lond.), 410: 701–705, 2001.

9. Lagasse, E., Connors, H., Al-Dhalimy, M., Reitsma, M., Dohse, M., Osborne, L. et al.Purified hematopoietic stem cells can differentiate into hepatocytes in vivo. Nat.Med., 6: 1229–1234, 2000.

10. Pittenger, M. F., Mackay, A. M., Beck, S. C., Jaiswal, R. K., Douglas, R., Mosca,J. D., Moorman, M. A., Simonetti, D. W., Craig, S., and Marshak, D. R. Multilineagepotential of adult human mesenchymal stem cells. Science (Wash. DC), 284: 143–1466, 1999.

11. Kondo, T., and Raff, M. Oligodendrocyte precursor cells reprogrammed to becomemultipotential CNS stem cells. Science (Wash. DC), 289: 1754–1757, 2000.

12. Clarke, D. L., Johansson, C. B., Wilbertz, J., Veress, B., Nilsson, E., Karlstrom, H.,Lendahl, U., and Frisen, J. Generalized potential of adult neural stem cells. Science(Wash. DC), 288: 1660–1663, 2000.

13. Toma, J. G., Akhavan, M., Fernandes, K. J., Barnabe-Heider, F., Sadikot, A., Kaplan,D. R., and Miller, F. D. Isolation of multipotent adult stem cells from the dermis ofmammalian skin. Nat. Cell Biol., 3: 778–784, 2001.

14. Reyes, M., Lund, T., Lenvik, T., Aguiar, D., Koodie, L., and Verfaillie, C. M.Purification and ex vivo expansion of postnatal human marrow mesodermal progen-itor cells. Blood, 164: 2615–2625, 2001.

15. Jiang, Y., Jahagirdar, B. N., Reinhardt, R. L., Schwartz, R. E., Keene, C. D.,Ortiz-Gonzalez, X. R., et al. Pluripotency of mesenchymal stem cells derived fromadult marrow. Nature (Lond.), 418: 41–49, 2002.

16. Ostenfeld, T., Caldwell, M. A., Prowse, K. R., Linskens, M. H., Jauniaux, E., andSvendsen, C. N. Human neural precursor cells express low levels of telomerase invitro and show diminishing cell proliferation with extensive axonal outgrowth fol-lowing transplantation. Exp. Neurol., 164: 215–226, 2000.

17. Manly, B. F.J. Randomization and Monte-Carlo Methods in Biology. London: Chap-man and Hall, 1991.

18. Tropepe, V., Hitoshi, S., Sirard, C., Mak, T. W., Rossant, J., and van der Kooy, D.Direct neural fate specification from embryonic stem cells: a primitive mammalianneural stem cell stage acquired through a default mechanism. Neuron, 30: 65–78,2001.

19. Lee, S. H., Lumelsky, N., Studer, L., Auerbach, J. M., and McKay, R. D. Efficientgeneration of midbrain and hindbrain neurons from mouse embryonic stem cells. Nat.Biotechnol., 18: 675–679, 2000.

20. Carpenter, M. K., Cui, X., Hu, Z. Y., Jackson, J., Sherman, S., Seiger, A., andWahlberg, L. U. In vitro expansion of a multipotent population of human neuralprogenitor cells. Exp. Neurol., 158: 265–278, 1999.

21. Lai, K., Kaspar, B. K., Gage, F. H., and Schaffer, D. V. Sonic hedgehog regulatesadult neural progenitor proliferation in vitro and in vivo. Nat. Neurosci., 6: 21–27,2003.

22. Bhardwaj, G., Murdoch, B., Wu, D., Baker, D. P., Williams, K. P., Chadwick, K.,Ling, L. E., Karanu, F. N., and Bhatia, M. Sonic hedgehog induces the proliferationof primitive human hematopoietic cells via BMP regulation. Nat. Immunol., 2:172–180, 2001.

23. Lendahl, U., Zimmerman, L. B., and McKay, R. D. CNS stem cells express a newclass of intermediate filament protein. Cell, 60: 585–595, 1990.

24. Meyerson, M., Counter, C. M., Eaton, E. N., Ellisen, L. W., Steiner, P., Caddle, S. D.et al. hEST2, the putative human telomerase catalytic subunit gene, is up-regulated intumor cells and during immortalization. Cell, 90: 785–795, 1997.

25. Flax, J. D., Aurora, S., Yang, C., Simonin, C., Wills, A. M., Billinghurst, L. L. et al.Engraftable human neural stem cells respond to developmental cues, replace neurons,and express foreign genes. Nat. Biotechnol., 16: 1033–1039, 1998.

26. Zhang, S. C., Wernig, M., Duncan, I. D., Brustle, O., and Thomson, J. A. In vitrodifferentiation of transplantable neural precursors from human embryonic stem cells.Nat. Biotechnol., 19: 1129–1133, 1998.

27. Khan, J., Simon, R., Bittner, M., Chen, Y., Leighton, S. B., Pohida, T. et al. Geneexpression profling of alveolar rhabdomyosarcoma with cDNA microarrays. CancerRes., 58: 5009–5013, 1998.

28. Bittner, M., Meltzer, P., Chen, Y., Jiang, Y., Seftor, E., Hendrix, M. et al. Molecularclassification of cutaneous malignant melanoma by gene expression profiling. Nature(Lond.), 406: 536–540, 2000.

29. Golub, T. R., Slonim, D. K., Tamayo, P., Huard, C., Gaasenbeek, M., Mesirov, J. P.et al. Molecular classification of cancer: class discovery and class prediction by geneexpression monitoring. Science (Wash. DC), 286: 531–537, 1999.

30. Bittner, M., Meltzer, P., and Trent, J. Data analysis and integration: of steps andarrows. Nat. Genet., 22: 213–215, 1999.

31. Lois, C., Garcia-Verdugo, J. M., and Alvarez-Buylla, A. Chain migration of neuronalprecursors. Science (Wash. DC), 271: 978–981, 1996.

32. Johansson, C. B., Momma, S., Clarke, D. L., Risling, M., Lendahl, U., and Frisen, J.Identification of a neural stem cell in the adult mammalian central nervous system.Cell, 96: 25–34, 1999.

33. Ridet, J. L., Malhorta, S. K., Privat, A., and Gage, F. H. Reactive astrocytes: cellularand molecular cues to biological function. Trends Neurosci., 20: 570–577, 1997.

34. Tanaka, T., Manome, Y., Wen, P., Kufe, D. W., and Fine, H. A. Viral vector-mediatedtransduction of a modified platelet factor 4 cDNA inhibits angiogenesis and tumorgrowth. Nat. Med., 3: 437–442, 1997.

35. Pollack, I. T., Erff, M., and Ashkenazi, A. Direct stimulation of apoptotic signalingby soluble Apo2l/tumor necrosis factor-related apoptosis-inducing ligand leads toselective killing of glioma cells. Clin. Cancer Res., 7: 1362–1369, 2001.

36. Lee, J., Hampl, M., Alberts, P., and Fine, H. A. Anti-tumor activity and prolongedexpression from a TRAIL-expressing adenoviral vector. Neoplasia, 4: 312–323, 2002.

37. Woodbury, D., Schwarz, E. J., Prockop, D. J., and Black, I. B. Adult rat and humanbone marrow stromal cells differentiate into neurons. J. Neurosci. Res., 61: 364–370,2000.

38. Sanchez-Ramos, J., Song, S., Cardozo-Pelaez, F., Hazzi, C., Stedeford, T., Willing,A. et al. Adult bone marrow stromal cells differentiate into neural cells in vitro. Exp.Neurol., 164: 247–256, 2000.

39. McKay, R. D. Stem cells in the central nervous system. Science (Wash. DC), 276:66–71, 1997.

40. Donovan, P. J., and Gearhart, J. The end of the beginning for pluripotent stem cells.Nature (Lond.), 414: 92–97, 2001.

41. Weissman, I. L. Translating stem and progenitor cell biology to the clinic: barriersand opportunities. Science (Wash. DC), 287: 1442–1446, 2000.

42. Svendsen, C. N., Bhattachayya, A., and Tai, Y-T. Neurons from stem cells: prevent-ing an identity crisis. Nat. Rev. Neurosci., 2: 831–834, 2001.

43. Temple, S. The development of neural stem cells. Nature (Lond.), 414: 112–117,2001.

44. Anderson, D. J., Gage, F. H., and Weissman, I. L. Can stem cells cross lineageboundaries? Nat. Med., 7: 393–398, 2001.

45. Orkin, S. H., and Zon, L. I. Hematopoiesis and stem cells:plasticity versus develop-mental heterogeneity. Nat. Immunol., 3: 323–328, 2002.

46. Castro, R. F., Jackson, K. A., Goodell, M. A., Robertson, C. S., Liu, H., and ShineH. D. Failure of bone marrow cells to transdifferentiate into neural cells in vivo.Science (Wash. DC), 297: 1299, 2002.

47. Wagers, A. J., Sherwood, R. I., Christensen, J. L., and Weissman, I. L. Little evidencefor developmental plasticity of adult hematopoietic stem cells. Science (Wash. DC),297: 2256–2259, 2002.

48. Terskikh, A. V., Easterday, M. C., Li, L., Hood, L., Kornblum, H. I., Geschwind,D. H., and Weissman, I. L. From hematopoiesis to neuropoiesis: evidence of over-lapping genetic programs. Proc. Natl. Acad. Sci. USA, 98: 7934–7939, 2001.

49. Ivanova, N. B., Dimos, J. T., Schaniel, C., Hackney, J. A., Moore, C. A., andLemischka, I. R. A stem cell molecular signature. Science (Wash. DC), 298: 601–604, 2002.

50. Ramalho-Santos, M., Yoon, S., Matsuzaki, Y., Mulligan, R. C., and Melton, D. A.“Stemness”: transcriptional profiling of embryonic and adult stem cells. Science(Wash. DC), 298: 597–600, 2002.

8889




2003;63:8877-8889. Cancer Res Jeongwu Lee, Abdel G. Elkahloun, Steven A. Messina, et al. Antiglioma Cellular VectorMarrow-Derived Neural Stem-Like Cells: A Potential Cellular and Genetic Characterization of Human Adult Bone

Updated version

http://cancerres.aacrjournals.org/content/63/24/8877

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/63/24/8877.full.html#ref-list-1

This article cites 47 articles, 19 of which you can access for free at:

Citing articles

/content/63/24/8877.full.html#related-urls

This article has been cited by 13 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

SubscriptionsReprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

[email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications


http://cancerres.aacrjournals.org/content/63/24/8877

http://cancerres.aacrjournals.org/content/63/24/8877.full.html#ref-list-1

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

mailto:[email protected]


Documents

Cellular and Genetic Characterization of Human Adult Bone ... · [CANCER RESEARCH 63, 8877–8889, December 15, 2003] Cellular and Genetic Characterization of Human Adult Bone Marrow-Derived