pex-producinghumanneuralstemcellsinhibittumorgrowth ......pex is a naturally occurring fragment of...

PEX-Producing Human Neural Stem Cells Inhibit Tumor Growthin a Mouse Glioma ModelSeung-Ki Kim,1,3 Theresa G. Cargioli,1Marcelle Machluf,4 Wendy Yang,1Yanping Sun,2

Ruqayyah Al-Hashem,2 Seung U. Kim,5 PeterM. Black,1and Rona S. Carroll1

Abstract A unique characteristic of neural stem cells is their capacity to track glioma cells that havemigrated away from the main tumor mass into the normal brain parenchyma. PEX, a naturallyoccurring fragment of human metalloproteinase-2, acts as an inhibitor of glioma and endothelialcell proliferation, migration, and angiogenesis. In the present study, we evaluated the antitumoractivity of PEX-producing human neural stem cells against malignant glioma.The HB1.F3 cell line(immortalized human neural stem cell) was transfected by a pTracer vector with PEX. The reten-tion of the antiproliferative activity and migratory ability of PEX-producing HB1.F3 cells(HB1.F3-PEX) was confirmed in vitro. For the in vivo studies, DiI-labeled HB1.F3-PEX cellswere stereotactically injected into established glioma tumor in nude mice. Tumor size was sub-sequently measured by magnetic resonance imaging and at the termination of the studies byhistologic analysis including tumor volume, microvessel density, proliferation, and apoptosis rate.Histologic analysis showed that DiI-labeled HB1.F3-PEX cells migrate at the tumor boundaryand cause a 90% reduction of tumor volume (P < 0.03). This reduction in tumor volume in ani-mals treated with HB1.F3-PEX was associated with a significant decrease in angiogenesis(44.8%, P < 0.03) and proliferation (23.6%, P < 0.03). These results support the use of neuralstem cells as delivery vehicle for targeting therapeutic genes against human glioma.

The treatment of malignant gliomas remains a challenge in theneurosurgical field. Despite some progress in standard treat-ment, the overall prognosis of this disease remains dismal, with amean survival period of less than a year. Most treatment failuresin malignant gliomas tend to be local, in the region of theoriginal tumor. These tumors rarely form distant metastasis.Therefore, glioma patients with persistent or recurrent diseasemay be ideal candidates for direct gene transfer intervention. Theeffectiveness of this delivery system has been a long-standingissue. For a therapeutic delivery system to be effective, it shouldbe able to target the tumor microsatellites that are the ultimatecause of recurrence (1). Recent studies show that neural stemcells administered intracranially exhibit significant migratorybehavior and extensive tropism for experimental glioma (2–6).

Therefore, neural stem cells have been suggested to be a tumor-targeting delivery system for glioma gene therapy.

PEX is a naturally occurring fragment of human metal-loproteinase-2. It acts as an inhibitor of glioma angiogenesis,cell proliferation, and migration (7, 8). We have previouslyshown that PEX expression correlated with tumor grade andhistologic subtype, being highly expressed in glioblastomas (7).Systemic or local administration of PEX results in a 97% to 99%suppression of intracranial tumor growth with no signs oftoxicity in a nude mice model (7).

In the present study, we transfected the gene for PEX intohuman neural stem cells and subsequently injected them intoan established glioma in nude mice. The PEX-producinghuman neural stem cells decreased angiogenesis and prolifer-ation, leading to a significant inhibition of tumor growth. Ourfindings lend further support to neural stem cell–based genetherapy for malignant gliomas.

Materials and Methods

Cell culture. HB1.F3 is an immortalized human neural stem cellline derived from the human fetal brain (ventricular zone) at 15 weeksof gestation by an amphotropic, replication-incompetent retroviralvector containing v-myc (9–11). HB1.F3 cells express nestin, a celltype–specific marker for neural stem cells (9, 11). They have proven tobe a multipotent, migratory, non–tumor-producing line in vivo model(9–14). HB1.F3 cells were maintained as adherent cultures in DMEMsupplemented with 10% fetal bovine serum (FBS), 2 mmol/LL-glutamine, and 100 units/mL penicillin, 100 Ag/mL streptomycin,and 0.25 Ag/mL fungizone (Invitrogen, Grand Island, NY). Cells weremaintained in T-75 tissue culture flasks in humidified atmospherecontaining 5% CO2 at 37jC.

Cancer Therapy: Preclinical

Authors’ Affiliations: 1Department of Neurosurgery, Brigham and Women’sHospital and Children’s Hospital and 2Department of Radiology, Brigham andWomen’s Hospital, Harvard Medical School, Boston, Massachusetts; 3Departmentof Neurosurgery, Seoul National University Hospital, Seoul, South Korea;4Laboratory for Cancer Drug Delivery and Mammalian Cell Technology,Technion-Israel Institute of Technology, Haifa, Israel; and 5Division of Neurology, UBCHospital, University of British Columbia, Vancouver, British Columbia, CanadaReceived 2/21/05; revised 4/11/05; accepted 4/26/05.Grant support: NIH, Postdoctoral Fellowship Program of Korea Science andEngineering Foundation (S.K. Kim), andWhitaker Foundation grant RG-01-0251(Y. Sun).The costs of publication of this article were defrayed in part by the payment of pagecharges.This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.Requests for reprints: RonaS.Carroll, DepartmentofNeurosurgery,BrighamandWomen’s Hospital, 221Longwood Avenue, Boston, MA 02115. Phone: 671-278-0177; Fax: 617-232-9029; E-mail: [email protected].

F2005 American Association for Cancer Research.doi:10.1158/1078-0432.CCR-05-0371

www.aacrjournals.org Clin Cancer Res 2005;11(16) August15, 20055965

Research. on August 29, 2021. © 2005 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

The human glioblastoma cell line U87MG (American Type Culture

Collection, Manassas, VA) was cultured in MEMa (Invitrogen)supplemented with 10% FBS, 2 mmol/L L-glutamine, 2 mmol/L

nonessential amino acids, 2 mmol/L sodium pyruvate, 100 units/mL

penicillin, 100 Ag/mL streptomycin, and 0.25 Ag/mL fungizone.In vitro transfection of neural stem cells. A 594 bp fragment of

human PEX was cloned into NheI and EcoRI sites of the pTracer-BsdPEXvector (Invitrogen). HB1.F3 cells were transfected with pTracer-BsdPEXusing the SuperFect transfection reagent (Qiagen, Valencia, CA) andselected under 10 Ag/mL blasticidin for 4 weeks.

Reverse transcription-PCR. Total cellular RNA from HB1.F3 and

HB1.F3-PEX cells was prepared using RNeasy Mini Kit protocol (Qiagen).Single-stranded cDNA was prepared from 1 Ag total RNA using oligo-

p(dT)15 primer following the First Strand cDNA Synthesis kit for reversetranscription-PCR (AMV) protocol (Roche, Indianapolis, IN). cDNA

from the reverse transcription reaction was subject to PCR in the presence

of 0.2 Amol/L of each 5Vand 3Vprimers, 2.5 units of Taq polymerase, 1.5mmol/L MgCl2, 0.2 mmol/L deoxynucleotide mix, and 1� PCR buffer.

Touch-down PCR for PEX was carried out for 25 cycles, each consisting ofdenaturation at 94jC for 4 minutes, annealing at 49.2jC in the first cycle

and decreased by 0.5jC for each subsequent cycle, and extension at 72jC

for 5 minutes. The final extension was continued for 10 minutes. Thehousekeeping gene b-actin , a positive control, was amplified for 32 cycles

(95jC for 1 minute, 55jC for 1 minute, 72jC for 1 minute, with a finalextension at 72jC for 10 minutes). The reaction products were analyzed

on a 2.5% agarose gel stained with ethidium bromide. The sense and

antisense primers, respectively, and the predicted sizes of the reversetranscription-PCR reaction products were as follows: PEX (sense: 5V-TAATACGACTCACTATAGGG-3V, T7 promoter primer in pTracer-CMV/

Bsd plasmid, antisense: 5V-GCCTCGTATACCGCATCAAT-3V, 382 bp) andb-actin (sense: 5V-GCCCAGAGCAAGAGAGGCAT-3V, antisense: 5V-GGCCATCTCTTGCTCGAAGT-3V, 513 bp). PEX is a fragment ofMMP-2, which is expressed at high levels in HB1.F3 cells; therefore, for

PEX amplification, we used a 5V primer was from the pTracer vector

and 3Vprimer from PEX.Assessment of cell viability. The cytotoxic effect of HB1.F3-PEX was

analyzed by coculture experiments. U87MG cells (8 � 103 cells/well)were plated in 96-well plates (Corning, Inc., Acton, MA). After 24hours, variable numbers of HB1.F3 or HB1.F3-PEX cells (0, 2 � 103,4 � 103, 8 � 103, 16 � 103, and 32 � 103 cells/well) were added to the

inside of the inserts (Nalge Nunc International, Rochester, NY) with0.2 Am pore size. After placing the inserts over the lower chamber, cellswere incubated for 96 hours. Quantitation of cell viability was donewith colorimetric assays using Cell Counting Kit-8 (Dojindo Molecular

Technologies, Gaithersburg, MD). All experiments were conducted inquadruplicate. Viability determination was based on the bioconversionof the tetrazolium compound, 2-(2-methoxy-4-nitrophenyl)-3-(4-nitro-phenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium (WST-8), into forma-

zan, as determined by absorbance at 450 nm using a multiwellscanning spectrophotometer. Cell viability was expressed as the mean FSE in percentage of the control viability (=100%).

In vitro migration assay. The migratory ability of HB1.F3-PEX onU87MG-derived extracellular matrix was compared with that of HB1.F3using a modified method previously described by Kaczarek et al. (15).Briefly, U87MG cells (3,000 cells/well) were allowed to grow toconfluence on the wells of the Teflon 10-well printed microscope slides(Creative Scientific Methods, Inc., Phoenix, AZ) for 3 days. The cells werelysed from the extracellular matrix using 20 mmol/L of NH4OH andrinsed with PBS. A 1.5 AL cell suspension containing 3,500 cells/AL ofHB1.F3 or HB1.F3-PEX was added to each cylinder of the 10-cylinder cellsedimentation manifold (Creative Scientific Methods) over the slide. Thecell sedimentation manifold was removed after sedimentation of the cellsand the cells were allowed to spread. The area occupied by the attachedcells was photographed using an inverted microscope (Nikon EclipseTE300) connected to a Spot RT Slider digital camera (DiagnosticInstruments, Inc., Houston, TX). The measurements were conductedevery 24 hours for 72 hours after the removal of the cell sedimentation

manifold in triplicate. The migration was calculated as the increase of theradius beyond the initial radius (the moment when the cell sedimen-tation manifold was removed) and expressed as the mean F SE.

In vivo migration assay. To address the relevance of migration

in vivo , HB1.F3-PEX cells were inoculated into the hemisphere

contralateral to the tumor cell inoculation. Briefly, Swiss nude malemice (n = 4, 6-8 weeks old; Charles River, Wilmington, MA) were

anesthetized (100 mg/kg ketamine and 5 mg/kg xylazine) and stereo-tactically inoculated with 120,000 U87MG cells in 3 AL of PBS via a 30-

gauge Hamilton syringe into the left forebrain (2.5 mm lateral and 1 mm

anterior to bregma, at a 2.5 mm depth from the skull surface). One weekafter tumor cell inoculation, HB1.F3-PEX cells (240,000 cells in 3 AL of

PBS) were inoculated into the right forebrain (2.5 mm lateral and 1 mm

anterior to bregma at a 2.5 mm depth from the skull surface). For thisstudy, the stem cell to tumor cell ratio was 2:1. HB1.F3-PEX was labeled

with the lipophilic tracer DiI (D-282; Molecular Probes, Eugene, OR)immediately before injection for 30 minutes according to the protocol of

the manufacturer. All animal studies were carried out in the animal

facility at Brigham and Women’s Hospital in accordance with federal,local, and institutional guidelines.

Inoculation of therapeutic cells into established intracranial gliomas.

After anesthesia, animals received stereotactic inoculation of U87MGcells (120,000 in 3 AL of PBS) into the left forebrain at the followingcoordinates: 2.5 mm lateral and 1 mm anterior to bregma at a 2.5 mmdepth from the skull surface previously described. Three days after tumorcell injection, the animals were randomized into three groups and treatedwith ipsilateral intratumoral inoculations of PBS (8 AL; n = 6), 480,000HB1.F3 cells (n = 6), or 480,000 HB1.F3-PEX cells (n = 6) in 8 AL of PBSat the established tumor site using the same burr hole and stereotacticcoordinates. For this study, stem cell to tumor cell ratio was 4:1. HB1.F3and HB1.F3-PEX were labeled with the lipophilic tracer DiI (MolecularProbes) immediately before injection.

Magnetic resonance imaging experiments. Seven days after ipsilateralintratumoral inoculation of human neural stem cells, magnetic

resonance imaging (MRI) experiments were done on a Bruker 4.7 Tsystem, operating on Paravision (version 3.0.1) software platform

(Bruker, Billerica, MA). Mice (n = 18) from all experimental groups were

anesthetized with 1% isoflurane in an oxygen/air mixture. Respiratoryrate was monitored using a Bruker Physioguard vital sign monitor. The

animals were maintained at 37jC inside the magnet using a

temperature-controlled water jacket. A T2-weighted image with RAREsequence (TR = 3,000 ms, TE = 50 ms, NEX = 4) was done to acquire 21

coronal slices from the whole brain with a slice thickness of 0.75 mm, amatrix size of 128 � 128 and a field of view of 2.56 � 2.56 cm2. After T2-

weighted image was acquired, Gadopentetate dimeglumine (Berlex

Laboratories, Wayne, NJ) was administered i.p. (0.8 mL/kg bodyweight). T1-weighted post-Gd images were obtained 15 minutes after

contrast injection using TR = 1,000 ms, TE = 8.8 ms, NEX = 2 and a slicethickness of 0.75 mm, matrix size of 128 � 128, and a field of view of

2.56 � 2.56 cm2. Tumor volumes were estimated using Gd-enhanced T1-

weighted spin-echo images, from which three-dimensional renderingsof the tumors were generated with in-house three-dimensional software

(16, 17).Determination of intracranial distribution and in vivo tumor

size. Two weeks after therapeutic intratumoral, ipsilateral inoculation ofhuman neural stem cell, animals were perfused with 4% paraformalde-hyde under deep anesthesia. These animals were the same animalspreviously used in the MRI experiments. The brains were removed,placed in sucrose gradient solution, embedded in optimum cuttingtemperature compound (Tissue-Tek, Miles, Elkhart, IN), and stored at�80jC. Brains were sectioned coronally using a cryostat into 10-Am-thick slices that were mounted on slides and then stained with H&E or4V,6-diamidino-2-phenylindole (DAPI) as per standard protocol. Usingfluorescence microscopy, intracranial distribution of DiI-labeled humanneural stem cells was assessed. Tumor volumes were estimated using theformula for ellipsoid and expressed as a mean F SE as previouslydescribed (16).


www.aacrjournals.orgClin Cancer Res 2005;11(16) August15, 2005 5966



Immunohistochemistry. Immunohistochemistry was carried outusing the Vectastain Elite ABC kit (Vector Laboratories, Burlingame,CA). Brain sections were fixed in cold acetone. Primary antibodiesincluded anti-CD31 (1:100; BD Biosciences PharMingen, San Jose, CA)for blood vessel density, anti-Ki67 nuclear antigen (1:100; DAKO,Carpinteria, CA) for proliferating cells, and anti–cleaved caspase-3(1:100; Cell Signaling Technology, Beverly, MA) for the detection ofapoptosis. Sections were counterstained with hematoxylin, and negativecontrol slides were obtained by omitting the primary antibody.Microvessel count was assessed according to the method described byLeon et al. (18) and Weidner et al. (19). The proliferation and apoptoticindices were defined as the percentage of positively stained cells of 100nuclei from five randomly chosen high-power fields.

Statistics. All of the values were calculated as mean F SE or wereexpressed as percentage of control F SE. Significant differences betweenassessment of cell viability, tumor volume, proliferation, apoptosisindex, and microvessel density were determined using the Mann-Whitney U test. Values of P < 0.05 were considered significant.

Results

Effect of HB1.F3-PEX on viability of U87MG cells in vitro.HB1.F3-PEX cells were obtained by transfection of the pTracer-BsdPEX vector with SuperFect followed by blasticidin selection.Expression of the PEX transcript was confirmed by reversetranscription-PCR. Reverse transcription-PCR showed that the

PEX transcript was present only in HB1.F3-PEX cells, but not inHB1.F3 cells (Fig. 1A) To confirm the biological activity of thePEX released from the transfected HB1.F3 cells, cell viabilitystudies were done (Fig. 1B). U87MG cells cultured in thepresence of HB1.F3-PEX cells, but not HB1.F3 cells, showedsignificant growth inhibition from the ratio of HB1.F3-PEX cellsto tumor cells of 1:2. These results confirmed the ability ofHB1.F3-PEX cells to produce and secrete biologically relevantquantities of the PEX protein.

Migratory capacity of HB1.F3-PEX in vitro and in vivo. Theeffects of PEX transfection on the migration of HB1.F3 wereassessed in vitro and in vivo . The migration pattern of the HB1.F3-PEX cells on the U87MG-produced extracellular matrix wasindistinguishable from the migration pattern of HB1.F3 cells.The incremental expansion of the advancing rim of migratingneural stem cells followed a linear relationship over timeregardless of PEX transfection (Fig. 2).

In vivo studies showed that HB1.F3-PEX cells could migratetoward intracranial glioma through normal brain parenchymain a similar pattern of untransfected cells. U87MG glioma cellswere inoculated into the left forebrain, and 1 week later DiI-labeled HB1.F3-PEX cells were inoculated into the contralateralhemisphere. HB1.F3-PEX cells migrated away from the initialinjection site and targeted the tumor (Fig. 3). DiI-labeledHB1.F3-PEX cells, inoculated intracranially, were found mostlyat the border between tumor and normal parenchyma. Thesefindings confirmed that the potential for migration of HB1.F3-PEX was not affected by the PEX transfection.

Effect of HB1.F3-PEX on in vivo tumor growth. To assess thetherapeutic efficacy of HB1.F3-PEX in glioma animal models, wemonitored tumor volume by MRI 7 days after intracranialhuman neural stem cell inoculation. T2-weighted images showedheterogeneous high signal intensity areas in the left hemisphere,indicating the presence of human glioblastoma xenografts. TheU87MG glioma was clearly visible as an enhanced area by T1-weighted imaging in the coronal section of the brain of controlanimals (9.7 F 2.20 mm3 in PBS-treated and 8.7 F 2.07 mm3

in HB1.F3-treated respectively; P > 0.1, PBS versus HB1.F3). In

Fig. 1. The analysis of expression and bioactivity of PEX from HB1.F3-PEX.A, HB1.F3 cells were transfected by SuperFect pTracer-BsdPEX. Reversetranscription-PCR products from HB1.F3 and HB1.F3-PEX cells were analyzedon a 2.5% agarose gel stained with ethidium bromide. Actin controls confirmedequal protein loading. B, U87MG cells (8� 103 cells/well) were plated in 96-wellcell plates. After 24 hours, HB1.F3 orHB1.F3-PEX cells (0, 2�103, 4�103, 8� 103,16� 103, and 32� 103 cells/well) were added to the inside of the inserts with0.2 Ampore size. After 96 hours of incubation, quantitation of cell viability wasdone with colorimetric assays using Cell Counting Kit-8. Columns, mean cellviability as a percentage of the control viability; bars, SE (*P < 0.05,Mann-WhitneyU test).

Fig. 2. The migratory ability of HB1.F3-PEX on U87MG-derived extracellularmatrix. A10-cylinder cell sedimentationmanifold was placed over the slide onU87MG-derived extracellular matrix. A1.5 AL cell suspension (3,500 cells/AL)of HB1.F3 or HB1.F3-PEX was added to each cylinder of the cell sedimentationmanifold. Migration value was quantified as the increase in the radius of the circleoccupied by the cell population beyond that measured on time 0 and every24 hours for 72 hours after the removal of the cell sedimentationmanifold.

PEX-Producing Human Neural Stem Cells




contrast, significantly smaller tumor volumes were measured inthe brains of animals treated with HB1.F3-PEX (Fig. 4; 1.00 F1.53 mm3; P < 0.03, PBS versus HB1.F3-PEX; P < 0.03, HB1.F3versus HB1.F3-PEX).

Tumor volumes were determined from glioma-bearingbrains harvested from mice 14 days after neural stem cell

inoculation by histologic analysis. The average tumor volumein controls, PBS-treated, and HB1.F3-treated animals were 31.6F 20.89 and 21.7 F 11.75 mm3, respectively, compared withonly 3.29 F 3.44 mm3 in HB1.F3-PEX–treated brains. Thisdecrease in tumor size associated with HB1.F3-PEX treatmentwas highly significant as has also been shown by a MRI study

Fig. 3. Intracranial distribution of DiI-labeledHB1.F3-PEX. HB1.F3-PEX were labeled with thelipophilic tracer DiI immediately before injection intocontralateral (A-C) or ipsilateral hemisphere (D-F).Brains were sectioned coronally into10-Am-thickslices and then stained with DAPI. Using fluorescencemicroscopy, intracranial distribution of DiI-labeledHB1.F3-PEX was assessed; arrow, tumor margin;arrowhead, corpus callosum (original magnification,�100).

Fig. 4. Tumor volumes as determined byMRI at 7 days after therapeutic intratumoralinoculation of humanneural stem cells andhistologic analysis at14 days afterinoculation of humanneural stem cells(n = 18). A , representative images fromMRI (Gd-T1^ weighted images). B,representative photographs from histology(H&E staining, original magnification,�1).C, tumor volumes were estimated fromMRI; columns, mean; bars, SE (P < 0.03,PBS versus HB1.F3-PEX; P < 0.03, HB1.F3versusHB1.F3-PEX,Mann-WhitneyU test).D, tumor volumes were estimated fromhistologic analysis; columns, mean; bars,SE (P < 0.03, PBS versus HB1.F3-PEX;P < 0.03, HB1.F3 versus HB1.F3-PEX,Mann-WhitneyU test).





(P < 0.03, PBS versus HB1.F3-PEX; P < 0.03, HB1.F3 versusHB1.F3-PEX; Fig. 4).

We next sought to determine biological action of PEX onglioma. Histologic analysis revealed a significant decrease intumor vascularization by 44.8% and a significant decrease in theproliferative index by 23.6% in those treated with HB1.F3-PEXcompared with the respective control groups (P < 0.03, PBSversus HB1.F3-PEX; P < 0.03, HB1.F3 versus HB1.F3-PEX; Fig. 5).In contrast, the apoptotic indices revealed no significant differ-ences among groups.

Discussion

The discovery of multipotential, self-renewing neural stemcells offers new approaches to brain tumor researchers. Inneuro-oncology, the biology of neural stem cells has beenpursued in two ways: (a) as cancer stem cells to understand theorigin and maintenance of brain tumors (20–24) and (b) as apotential cell-based vehicle for gene therapy (2–6, 25–29).

Regarding the first issue, some groups have described theexistence of a cancer stem cell population in human braintumors using separation techniques with selective neural stemcell markers (20–22, 24). The identification and characteriza-tion of brain tumor stem cells provides a valuable tool tounderstand their biology more clearly and to develop therapiestargeted to the brain tumor stem cells. On the other hand,several studies have focused on tumor-tropic capacities ofneural stem cells. Neural stem cell lines have been shown to beeffective for delivering transgenes to brain tumors based ontheir unique migratory properties within the central nervoussystem (2–5, 28, 29). Migratory genetically modified neural

stem cells have been reported to promote tumor regression andprolong survival of glioma-bearing animals (3–6, 27, 29).Genes encoded for neural stem cell–based therapy includedimmune enhancing genes, such as interleukin-4 (29), interleu-kin-12 (4, 25); apoptosis promoting gene, such as tumornecrosis factor-related apoptosis-inducing ligand (2, 3); pro-drug converting enzymes, such as cytosine deaminase (5, 27)and herpes simplex virus thymidine kinase (30); and herpessimplex viral therapy (6). Moreover, it was found that theunmodified stem cells also had a potential for tumor outgrowthinhibition, indicating that neural stem cells themselves mayproduce some factors with an antitumor activity (31).

For the treatment of gliomas, we combined the extensivetumor tracking capability of neural stem cells with theantiangiogenic and antiproliferative potency of PEX. Our cellularvehicle system has two very unique characteristics. First, theneural stem cells used in our study are of human origin, asopposed to other reported groups with murine–origin neuralstem cells. In the clinical perspective, human-origin neural stemcells could be used more safely with less immune rejection thanmurine-origin neural stem cells, avoiding the potential dangersof xenograft. Second, the neural stem cells we used as atherapeutic vehicle were engineered to have the ability tosignificantly reduce glioma angiogenesis and migration. This isthe first experimental study in which neural stem cells were usedto deliver an antiangiogenic gene into an established humanxenograft. This resulted in the potent reduction of angiogenesisand proliferation, leading to a highly significant inhibition oftumor growth compared with neural stem cell only or PBS-inoculated controls. In this paradigm, the neural stem cells havethe unique ability to engulf the tumor and concentrate in the

Fig. 5. Treatment effects assessed by immunohistochemistry (n = 18). Primary antibodies included anti-CD31for blood vessels, anti-Ki67 nuclear antigen for proliferatingcells, and anti ^ cleaved caspase-3 for the detection of apoptosis. Sections were counterstained with H&E (original magnification, �200). Columns, mean microvesselcount, proliferation index, and apoptotic index as percentages of the control; bars, SE (microvessel count and proliferation index, P < 0.03, PBS versus HB1.F3-PEX;P < 0.03, HB1.F3 versus HB1.F3-PEX, Mann-Whitney U test).

PEX-Producing Human Neural Stem Cells




tumor boundary where angiogenesis is most active. Treatmenteffects of HB1.F3-PEX in our animal model were confirmed bymagnetic resonance imaging. This noninvasive technique canmonitor dynamic changes of tumor growth during treatmentand provide an independent measurement of tumor volume(17). It has a wide range of applications for the study of noveltherapeutics for brain tumors.

We showed a 90% reduction of tumor growth in a xenograftnude mice model by inoculation of HB1.F3-PEX cells, suggest-ing stable expression of PEX in vivo . However, we did notachieve complete tumor regression by this therapeutic ap-proach. One possible explanation of this is that, in our study, noincrease in apoptosis was observed. Our previous study (7) alsoshowed that to achieve the proapoptotic activity of PEX, muchhigher concentrations were needed than those required toinhibit glioma growth or migration (25 versus 1 versus 10 Ag/mL). In our study, HB1.F3-PEX cells may not be able to producesufficient amount of PEX protein to induce apoptosis of gliomacells. However, high concentration of PEX gene could induceapoptosis of neural stem cells themselves and inhibit theirmigratory ability. For PEX-producing neural stem cells to be aneffective therapeutic delivery vehicle, the amount of PEX

produced must be carefully balanced. Although therapeuticeffect was assessed 2 weeks after inoculation of human neuralstem cell, we did not observe any signs of local or systemictoxicity in HB1.F3- and HB1.F3-PEX–treated groups.

The HB1.F3-PEX could migrate to tumor site and inhibitangiogenesis without inducing apoptosis or inhibiting cellmotility. One way to increase therapeutic efficiency ofHB1.F3-PEX would be using a construct that has an induciblepromoter, such as the tetracycline silencer system (32). Thiscould allow a higher level of PEX protein expression afterneural stem cells migrate within and beyond the tumor mass.

In conclusion, human-origin neural stem cells engineered toproduce PEX can migrate toward glioma and have strongantitumor effect. Our finding strengthens the potential strategiesaimed at malignant gliomas using human neural stem cells as atargeting vehicle for therapeutic gene transfer. Neural stem cellsmight be used with standard therapies in parallel to reduce theincidence of recurrence and improve the survival of patients.After tumor removal surgery, PEX-producing human neural stemcells could be grafted into tumor resection cavity to targetresidual invading tumor cells and to inhibit tumor angiogenesisand proliferation.

References1. Dunn IF, Black PM. The neurosurgeon as local on-cologist: cellular and molecular neurosurgery in ma-lignant glioma therapy. Neurosurgery 2003;52:1411^24.

2. Shah K, Bureau E, Kim DE, et al. Glioma therapyand real-time imaging of neural precursor cell migra-tion and tumor regression. Ann Neurol 2005;57:34^41.

3. EhteshamM, Kabos P, Gutierrez MA, et al. Inductionof glioblastoma apoptosis using neural stem cell-mediated delivery of tumor necrosis factor-relatedapoptosis-inducing ligand. Cancer Res 2002;62:7170^4.

4. EhteshamM, Kabos P, Kabosova A, NeumanT, BlackKL,Yu JS. The use of interleukin 12-secreting neuralstem cells for the treatment of intracranial glioma.Cancer Res 2002;62:5657^63.

5. Aboody KS, Brown A, Rainov NG, et al. Neural stemcells display extensive tropism for pathology in adultbrain: evidence from intracranial gliomas. Proc NatlAcad Sci US A 2000;97:12846^51.

6. Herrlinger U,Woiciechowski C, Sena-Esteves M,et al. Neural precursor cells for delivery of replication-conditional HSV-1vectors to intracerebral gliomas.MolTher 2000;1:347^57.

7. Bello L, Lucini V, Carrabba G, et al. Simultaneousinhibition of glioma angiogenesis, cell proliferation,and invasion by a naturally occurring fragment ofhuman metalloproteinase-2. Cancer Res 2001;61:8730^6.

8. Brooks PC, Silletti S, von SchalschaTL, FriedlanderM, Cheresh DA. Disruption of angiogenesis by PEX, anoncatalytic metalloproteinase fragment with integrinbinding activity. Cell1998;92:391^400.

9. ChoT, Bae JH, Choi HB, et al. Human neural stemcells: electrophysiological properties of voltage-gatedion channels. Neuroreport 2002;13:1447^52.

10.OurednikV, OurednikJ, FlaxJD, et al. Segregationofhuman neural stem cells in the developing primateforebrain. Science 2001;293:1820^4.

11. Flax JD, Aurora S,Yang C, et al. Engraftable human

neural stem cells respond to developmental cues,replace neurons, and express foreign genes. NatBiotechnol1998;16:1033^9.

12. Chu K, Kim M, Chae SH, et al. Distribution andin situ proliferation patterns of intravenously injectedimmortalized human neural stem-like cells in ratswith focal cerebral ischemia. Neurosci Res 2004;50:459^65.

13. Chu K, Kim M, Jeong SW, Kim SU, Yoon BW. Hu-manneural stem cells canmigrate, differentiate, and in-tegrate after intravenous transplantation in adult ratswith transient forebrain ischemia. Neurosci Lett 2003;343:129^33.

14. JeongSW,ChuK, JungKH, KimSU, KimM,RohJK.Human neural stem cell transplantation promotesfunctional recovery in ratswith experimental intracere-bral hemorrhage. Stroke 2003;34:2258^63.

15. Kaczarek E, Zapf S, Bouterfa H,Tonn JC,WestphalM, Giese A. Dissecting glioma invasion: interrelationof adhesion, migration and intercellular contacts deter-mine the invasive phenotype. IntJDev Neurosci1999;17:625^41.

16. Schmidt KF, ZiuM, Schmidt NO, et al.Volume recon-struction techniques improve the correlation betweenhistological and in vivo tumor volume measurementsin mouse models of human gliomas. J Neurooncol2004;68:207^15.

17. SunY, Schmidt NO, Schmidt K, et al. Perfusion MRIof U87 brain tumors in a mouse model. Magn ResonMed 2004;51:893^9.

18. Leon SP, Folkerth RD, Black PM.Microvessel densi-ty is a prognostic indicator for patients with astroglialbrain tumors. Cancer1996;77:362^72.

19.Weidner N, Semple JP,WelchWR, FolkmanJ.Tumorangiogenesis and metastasisMcorrelation in invasivebreast carcinoma. NEnglJMed1991;324:1^8.

20. Galli R, Binda E, Orfanelli U, et al. Isolation andcharacterization of tumorigenic, stem-like neural pre-cursors from human glioblastoma. Cancer Res 2004;64:7011^21.

21. Singh SK, Hawkins C, Clarke ID, et al. Identification

of human brain tumour initiating cells. Nature 2004;432:396^401.

22. Yuan X, Curtin J, XiongY, et al. Isolation of cancerstem cells from adult glioblastoma multiforme. Onco-gene 2004;23:9392^400.

23. Recht L, JangT, Savarese T, Litofsky NS. Neuralstem cells and neuro-oncology: quo vadis? J CellBiochem 2003;88:11^9.

24. Singh SK, Clarke ID,Terasaki M, et al. Identificationof a cancer stem cell in human brain tumors. CancerRes 2003;63:5821^8.

25.Yang SY, Liu H, ZhangJN. Gene therapy of ratmalig-nant gliomas using neural stem cells expressing IL-12.DNACell Biol 2004;23:381^9.

26. Yip S, Aboody KS, Burns M, et al. Neural stem cellbiology may be well suited for improving brain tumortherapies. CancerJ 2003;9:189^204.

27. Barresi V,BelluardoN,SipioneS,MudoG,CattaneoE,CondorelliDF.Transplantationofprodrug-convertingneural progenitor cells for brain tumor therapy. CancerGene Ther 2003;10:396^402.

28. Brown AB,YangW, Schmidt NO, et al. Intravasculardelivery of neural stem cell lines to target intracranialand extracranial tumors of neural and non-neuralorigin. Hum GeneTher 2003;14:1777^85.

29. Benedetti S, Pirola B, Pollo B, et al. Gene therapy ofexperimental brain tumors using neural progenitorcells. NatMed 2000;6:447^50.

30. Uhl M,Weiler M,Wick W, Jacobs AH,Weller M,Herrlinger U. Migratory neural stem cells for improvedthymidine kinase-based gene therapy of malignant gli-omas. Biochem Biophys Res Commun 2005;328:125^9.

31. Staflin K, Honeth G, Kalliomaki S, Kjellman C,Edvardsen K, Lindvall M. Neural progenitor cell linesinhibit rat tumor growth in vivo. Cancer Res 2004;64:5347^54.

32. Freundlieb S, Schirra-Muller C, Bujard H. A tetracy-cline controlled activation/repression system withincreased potential for gene transfer into mammaliancells. JGeneMed1999;1:4^12.





2005;11:5965-5970. Clin Cancer Res Seung-Ki Kim, Theresa G. Cargioli, Marcelle Machluf, et al. Growth in a Mouse Glioma ModelPEX-Producing Human Neural Stem Cells Inhibit Tumor

Updated version

http://clincancerres.aacrjournals.org/content/11/16/5965

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/11/16/5965.full#ref-list-1

This article cites 32 articles, 9 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/11/16/5965.full#related-urls

This article has been cited by 10 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://clincancerres.aacrjournals.org/content/11/16/5965To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/content/11/16/5965.full#ref-list-1

http://clincancerres.aacrjournals.org/content/11/16/5965.full#related-urls

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

mailto:[email protected]



pex-producinghumanneuralstemcellsinhibittumorgrowth ......pex is a naturally occurring fragment of...

Documents