Molecular and cell biology of brain tumor stem cells: lessons from neural progenitor/stem cells

ALIGNANT brain tumors are among the most dead-ly cancers. Current treatment strategies, includingsurgery, radiotherapy, and chemotherapy, only

modestly improve patient survival. The limited success ofthese strategies is largely due to a high incidence of tumorrecurrence following treatment. The results of studies con-ducted in the past several years suggest that a rare popula-tion of stem cell–like cells, known as brain tumor stem cells(BTSCs), may be responsible for the recurrence of braintumors. Based on the expression of the plasma membraneprotein CD133 by BTSCs and the ability of BTSCs to self-renew in neurosphere assays, several groups have been ableto isolate these cells from human brain tumor sam-ples.31,35,38,70,86 When orthotopically transplanted intoimmunodeficient mice, purified BTSCs gave rise to braintumors that histologically resembled original tumors.31,35,

70,86 Thus, targeting BTSCs may be an effective strategy toprevent brain tumor recurrence.

The Relationship Between BTSCs and NPSCs

The rational design of BTSC-based brain tumor thera-pies requires a thorough understanding of how BTSCsacquire and maintain their unusual ability to self-renew andinfiltrate. Our knowledge in this field has been greatly en-hanced by studies of NPSCs, precursor cells that can self-renew and generate neurons and glia. Different types ofNPSCs exist in the mammalian brain during different de-velopmental stages.20,33,51 The earliest NPSCs are neuroep-ithelial cells, which appear prior to the onset of neurogene-sis during embryonic development. Neuroepithelial cellsinitially expand via symmetric cell division and then devel-op into more lineage-restricted NPSCs known as radialglia. Radial glia may give rise to neurons via asymmetriccell division or by generating intermediate neural pro-genitors, which then produce 2 daughter neurons via sym-metric cell division. During postnatal stages, radial gliatransform into slowly dividing neurogenic astrocytes in 2specific brain areas: the subventricular zone of the lateralventricle and the subgranular zone of the hippocampal den-tate gyrus. Neurogenic astrocytes undergo asymmetricdivision to self-renew and generate transient amplifyingcells, which rapidly divide to produce neuroblasts.

The ability of NPSCs to self-renew and the persistenceof neurogenesis in the adult brain have led to the hypothe-sis that NPSCs may give rise to BTSCs through genetic

Neurosurg. Focus / Volume 24 / March/April 2008

Neurosurg Focus 24 (3&4):E24, 2008

Molecular and cell biology of brain tumor stem cells:lessons from neural progenitor/stem cells

ZHIGANG XIE, PH.D.,1,2 AND LAWRENCE S. CHIN, M.D.1

Departments of 1Neurosurgery and 2Pharmacology and Experimental Therapeutics, Boston UniversitySchool of Medicine, Boston, Massachusetts

PThe results of studies conducted in the past several years have suggested that malignant brain tumors may harbor asmall fraction of tumor-initiating cells that are likely to cause tumor recurrence. These cells are known as brain tumorstem cells (BTSCs) because of their multilineage potential and their ability to self-renew in vitro and to recapitulateoriginal tumors in vivo. The understanding of BTSCs has been greatly advanced by knowledge of neural progeni-tor/stem cells (NPSCs), which are multipotent and self-renewing precursor cells for neurons and glia. In this article,the authors summarize evidence that genetic mutations that deregulate asymmetric cell division by affecting cell polar-ity, spindle orientation, or cell fate determinants may result in the conversion of NPSCs to BTSCs. In addition, theyreview evidence that BTSCs and normal NPSCs may reside in similar vascularized microenvironments, where simi-lar evolutionarily conserved signaling pathways control their proliferation. Finally, they discuss preliminary evidencethat mechanisms of BTSC-associated infiltrativeness may be similar to those underlying the migration of NPSCs andneurons. (DOI: 10.3171/FOC/2008/24/3-4/E24)

KEY WORDS • infiltration • proliferation • renewal • tumor

M

1

Abbreviations used in this paper: BMP = bone morphogeneticprotein; BTSC = brain tumor stem cell; CNS = central nervous sys-tem; EGF = epidermal growth factor; EGFR = EGF receptor; FGF =fibroblast growth factor; MCPH = autosomal recessive primarymicrocephaly; NICD = Notch intracellular domain; NPSC = neuralprogenitor/stem cells; Shh = Sonic hedgehog; VEGF = vascularendothelial growth factor.

mutations.65,76 Several lines of evidence support this hy-pothesis. First, NPSCs and BTSCs share molecular mark-ers such as CD133 and nestin, and gene profiles of variousbrain tumors recapitulate that of NPSCs at different devel-opmental stages.57,75 Next, the proliferation of NPSCs andBTSCs is controlled by similar molecular pathways (dis-cussed later in this paper). Third, NPSCs and BTSCsbehave similarly in culture, forming neurospheres undergrowth factor–supplemented non-adherent conditions andgiving rise to a mixture of neurons, astrocytes, and oligo-dendrocytes when induced to differentiate.31,35,38,70,86 Lastly,analyses of mouse models have revealed that malignantbrain tumors may arise in regions containing NPSCs andthat genetic alterations in NPSCs can induce malignantbrain tumors.24,36,62,71,87 Together, the results of these studiessuggest that NPSCs are likely, although not necessarilyexclusive, sources of BTSCs.

Mechanisms of BTSC Self-Renewal

The investigation of the mechanisms that control BTSCproliferation has been guided by our knowledge of the mol-ecular and cellular basis of NPSC self-renewal. Neural pro-genitor/stem cells exhibit highly polarized cell morphologyand undergo asymmetric or symmetric divisions in specif-ic microenvironments, where a number of signaling path-ways maintain the balance between self-renewal and dif-ferentiation of the NPSCs.

Asymmetric Division of NPSCs

All primary NPSCs, including neuroepithelial cells, radi-al glia, and neurogenic astrocytes, exhibit highly polarizedcell morphology. Both neuroepithelial cells and radial gliaspan the entire cerebral wall by extending a thin basalprocess that contacts the basement membrane and a rela-tively thick apical process that contacts the ventricular sur-face. The basal process serves as a support for the migrationof newly generated neurons away from the proliferativezone, whereas the apical process forms intercellular adhe-sion structures at the tip to connect neighboring NPSCs.33

The plasma membrane of the tip of the apical process maycontain cell fate determinants. Whether these cell fate de-terminants are segregated into daughter cells symmetrical-ly or asymmetrically depends on the orientation of themitotic spindle during mitosis.12,33 Neurogenic astrocytes inthe adult brain also exhibit polarized morphology witheither an apical process or a basal process, although they donot span the entire cerebral wall.20

The establishment of cell polarity is necessary for un-equal distribution of cell fate determinants and, therefore,proper asymmetric cell division and the specification ofdaughter cell fates. For instance, conditional ablation of aRho GTPase cdc42 in the embryonic mouse telencephalonresulted in the retraction of the apical process from the ven-tricular surface and the loss of apical Par complex andadherens junctions in primary NPSCs.14 This disruption ofapical polarity in turn caused increased production of inter-mediate neural progenitors from primary NPSCs.14

In addition to cell polarity, the orientation of the mitoticspindle also controls the distribution of cell fate determi-nants by governing the cleavage plane.12,33,85 The criticalrole of spindle orientation in cell fate specification during

neurogenesis was initially discovered in the Drosophilanervous system.85 Recent studies suggest that spindle ori-entation also plays a key role in mammalian neurogenesis.For instance, loss of Aspm,28 Nde1,27 or Gbg signaling64 inprimary NPSCs of the developing neocortex results in al-tered spindle orientation, which may underlie a prematuredepletion of primary NPSCs.

Brain Tumor Stem Cells: Deregulation of AsymmetricDivision?

A tight control of asymmetric division ensures that thebalance between self-renewal and differentiation is main-tained in NPSCs. Since BTSCs are characterized by uncon-trolled proliferation or self-renewal, one possibility is thatNPSCs harboring mutations that impair asymmetric divi-sion may become BTSCs and thus promote tumorigenesis.In support of this possibility, genetic studies of DrosophilaCNS neurogenesis have revealed that mutations that affectcell polarity, spindle orientation, or cell fate determinantsmay induce NPSC overproliferation and brain tumors. Forinstance, overexpression of a membrane-targeted atypicalprotein kinase C, a highly conserved protein that controlsthe polarity of many different types of cells, induced neo-plastic overgrowth of neuroblasts,45 which are NPSCs inthe Drosophila CNS. Correspondingly, loss of atypical pro-tein kinase C caused reduced neuroblast proliferation.63

Loss of lethal giant larvae (lgl)45 or polo,78 which areimportant for polarized distribution of cell fate determi-nants,54,56,78 also led to neoplastic overgrowth of neuroblasts.In addition to deregulation of cell polarity, the misposi-tioning of a mitotic spindle resulting from loss of a NuMAhomolog, mud;11,59 a centrosomal protein, cnn;44 or a kinase,aurora,44,79 has also been linked to excess neuroblast self-renewal and brain tumor formation. Finally, mutations ingenes encoding cell fate determinants such as brain tumor(brat) and prospero (pros) caused increased self-renewal ofneuroblasts and brain tumor formation.6,8,15,46

Much less is known about how asymmetric division maycontribute to brain tumorigenesis in mammals. Nonethe-less, several lines of evidence suggest that the underlyingmechanisms may be at least partially conserved betweenDrosophila and mammals. First, in mice lacking the polar-ity gene Lgl1, the homolog of Drosophila lgl, cortical neu-roepithelial cells exhibited increased self-renewal andformed rosette-like structures similar to those found inhuman primitive neuroectodermal tumors.40 Second, mam-malian polo-like kinase 1 (Plk1), a homolog of Drosophilapolo, has been implicated in various human cancers includ-ing gliomas,74 the most common type of primary brain tu-mors. Third, mammalian Aurora A kinase, a homolog ofDrosophila aurora, is also a key regulator of the mitoticspindle.4 While the function of Aurora A kinase in NPSCself-renewal and brain tumorigenesis remains to be deter-mined, the expression of Aurora A kinase has been foundto be elevated in several types of human cancers.4 Fourth,one of the causative genes for human MCPH, CDK5RAP2,is a homolog of Drosophila cnn and also encodes a centro-somal protein.10 Since MCPH is a brain development dis-order characterized by reduced size of both the brain andthe NPSC pool,22 it is likely that, similar to the productof cnn, the protein product of CDK5RAP2 also regulatesNPSC self-renewal via mitotic spindle–dependent mecha-

Z. Xie and L. S. Chin

2 Neurosurg. Focus / Volume 24 / March/April 2008

nisms. Lastly, studies of another causative gene for MCPH,ASPM,9 have provided a more direct link between asym-metric division and brain tumorigenesis in mammals. In thedeveloping mouse brain, the Aspm protein has been shownto promote NPSC self-renewal by maintaining the mitoticspindle in an orientation that favors symmetric division.28

In a gene profiling analysis of glioblastomas, the most com-mon and malignant human primary brain tumors, ASPMwas identified as an overexpressed gene.37 Furthermore,RNA interference–mediated silencing of ASPM inhibitedthe proliferation of cultured glioblastoma cells.37 Thus, theAspm protein may support the expansion of both NPSCsand brain tumor cells by inhibiting asymmetric division.

While these findings are consistent with the possibilitythat NPSCs with dysregulated asymmetric division maybecome BTSCs, solid evidence is lacking. To further testthis possibility, it will be necessary to analyze asymmetricdivision directly on BTSCs and to determine whether ge-netic manipulations that target asymmetric division will besufficient to convert mammalian NPSCs to BTSCs.

A Vascular Niche for NPSCs and BTSCs

The microenvironment plays a critical role in howNPSCs divide. In the adult mammalian brain, NPSCs (thatis, neurogenic astrocytes) reside in niches enriched withcapillaries.20 In the embryonic mammalian brain, NPSCsare also closely associated with the vasculature.7,77 Neuralprogenitor/stem cells and blood vessels may communicatebidirectionally. On one hand, NPSCs may induce endothe-lial cells to form vascular tubes by secreting VEGF andbrain-derived neurotrophic factor.48 On the other hand, en-dothelial cells may support the self-renewal of NPSCs bysecreting trophic factors such as VEGF and pigment ep-ithelium-derived factor.43,61,67

A recent study using 3D reconstruction of immuno-stained brain tumors revealed that, similar to NPSCs,BTSCs also establish close relationships with blood ves-sels.13 Both in culture and when transplanted into the mousebrain, the self-renewal of BTSCs was enhanced by thepresence of endothelial cells.13 Furthermore, depletion ofthe tumor vasculature eradicated BTSCs and inhibited thegrowth of tumor xenografts in the mouse brain.13 Thesedata suggest that BTSCs reside in a vascular niche whereendothelium-derived trophic factors support BTSC self-renewal. Within this vascular niche, BTSCs may also pro-mote angiogenesis by secreting VEGF.2,55 Therefore, theexcessive vascularization typical of malignant brain tumorsmay result from positive feedback between BTSCs andblood vessels.

Control of BTSC Proliferation by DevelopmentalSignaling Pathways

In addition to being governed by vasculature-related tro-phic factors, the proliferation of BTSCs is also governed byother signaling pathways within the niche, such as BMPs,Notch, and Shh.18,72 These pathways are known to play keyroles in several aspects of neural development, includingthe self-renewal and differentiation of NPSCs.

Bone morphogenetic proteins belong to the transforminggrowth factor–b superfamily of secreted ligands.66 Bybinding to cell-surface receptors (BMP receptors), BMPsactivate transcription factors (SMADs), which then translo-

cate into the nucleus to regulate gene expression. Orig-inally identified as bone-inducing factors, BMPs were laterfound to also have important functions in neurogenesis.Interestingly, the role of BMPs in neurogenesis is compli-cated and may even appear paradoxical.16 Under certainconditions BMPs may specify the dorsal identity of theneural tube and promote the proliferation of NPSCs, where-as under other conditions BMPs may induce apoptosis,mitotic arrest, and differentiation of NPSCs. The results ofa recent study using mice transplanted with human gli-oblastoma cells suggest that the activation of BMP signal-ing may inhibit the proliferation and promote the differen-tiation of BTSCs, thereby reducing the ability of BTSCs toinitiate tumors.58 Analysis of a medulloblastoma xenograftmodel suggests that BMPs may also inhibit brain tumori-genesis by inducing apoptosis;34 however, it is unclearwhether BTSCs undergo apoptosis in this model.

Notch signaling is an evolutionarily conserved pathwaythat mediates the interaction between neighboring cells.49

The core components of this pathway include the ligandsdelta and serrate; the receptor Notch; a g-secretase complexthat cleaves the receptor Notch to produce NICD; NICD-associated nuclear effectors Su(H) and MAM; and Notchtarget genes, such as HES, HES-related, and BLBP. Notchsignaling has pleiotropic functions in neural development,such as maintaining NPSC proliferation, inducing glialfate, and promoting neurite elaboration.49 The activation ofthe Notch pathway has been linked to many human can-cers,69 including malignant brain tumors.23,39,57,60 Two recentstudies suggest that the role of Notch signaling in brain tu-morigenesis may involve BTSCs. In the first study, Shihand Holland68 showed that Notch signaling enhanced theexpression of nestin, a marker for BTSCs, in glioblastomatissues. Furthermore, when combined with K-ras, activa-tion of Notch in nestin-expressing cells induced progenitor-like brain lesions. In the second study, Fan et al.26 blockedNotch signaling in medulloblastoma cells by using a spe-cific inhibitor of g-secretase, which is required for thecleavage of Notch receptor to the activated form NICD.This inhibitor effectively ablated the BTSC population,possibly by inducing apoptosis, and markedly impairedmedulloblastoma growth both in vitro and in a xenograftmouse model. Together, these 2 studies suggest that Notchsignaling may promote brain tumorigenesis by stimulatingthe proliferation and blocking the apoptosis of BTSCs.

Sonic hedgehog is a secreted protein that belongs to thehedgehog family, members of which are best known asmorphogens important for the development of Drosophilaand vertebrates.30 Sonic hedgehog signaling is initiatedby the binding of Shh to a 12-pass membrane receptor“patched” (PTC1). This binding relieves the repression of a7-pass G-protein–coupled receptor “Smoothened” (SMO)by PTC1. The activation of SMO leads to a series of sig-naling cascades that result in nuclear translocation of GLItranscription factors, which then regulate the expression oftarget genes. Sonic hedgehog signaling plays multifacetedroles during CNS development, including ventral pattern-ing, oligodendrocyte specification, the proliferation andsurvival of NPSCs, and axon guidance.30 Genetic analysesof human brain tumor tissues and mouse models haverevealed that increased activity of the Shh pathway is in-volved in the development of both medulloblastoma andglioblastoma.18 The contribution of Shh to brain tumori-


Molecular and cell biology of brain tumor stem cells

3

genesis may be at least partially through BTSCs. In humanglioma samples, BTSCs may exhibit elevated Shh activi-ty.3,19,25 Importantly, pharmacological inhibition or RNAinterference–mediated silencing of Shh signaling decreasedthe self-renewal of BTSCs derived from these tumor sam-ples and impaired the ability of these BTSCs to initiatetumors in a xenograft mouse model.3,19 Thus, Shh signalingmay promote brain tumorigenesis by maintaining the ca-pacity of BTSCs to self-renew.

The proliferation of BTSCs is likely to also be controlledby other signaling pathways, such as those involving Wnt,EGF, and FGF. A conserved pathway important for thedevelopment of various systems, Wnt signaling has beenshown to promote NPSC proliferation and has been linkedto multiple cancers, including brain tumors.17,18 IncreasedEGF signaling resulting from gain-of-function mutationsof its receptor EGFR or loss-of-function mutations of itsdownstream regulator PTEN is typical of primary glioblas-



FIG. 1. A tentative model for the origin of BTSCs from NPSCs. The NPSCs exhibit polarized distribution of cell fatedeterminants, which are segregated unequally into the 2 daughter cells during asymmetric division. A tight control of thepolarized distribution of the cell fate determinants and the orientation of the mitotic spindle ensures that a balancebetween asymmetric division and symmetric division is maintained. This balance is shifted toward symmetric divisionin NPSCs harboring genetic mutations that disrupt cell polarity, alter spindle orientation, or inactivate cell fate determi-nants important for non–stem-cell specification. The increased symmetric division results in uncontrolled self-renewal, ahallmark of BTSCs.

tomas.53 The FGFs promote NPSC proliferation50 and havealso been implicated in brain tumorigenesis.29,82,83 BothEGF and FGF2 are key supplements used for maintainingBTSCs in neurosphere culture assays.47 Nevertheless, thereis no direct evidence that Wnt, EGF, or FGF signaling reg-ulates BTSC self-renewal in vivo.

Brain Tumor Invasion and BTSCs

Malignant tumors are characterized by both uncontrolledproliferation and the ability to infiltrate. While much hasbeen learned about the molecular and cellular basis ofBTSC proliferation since the discovery of BTSCs severalyears ago, mechanisms underlying BTSC-associated in-filtrativeness remain poorly understood. Previous studiesusing non-BTSC brain tumor cells have identified manycell migration pathways that regulate cell–cell adhesion,cell–ECM adhesion, or cell motility.52 While these path-ways are probably important to brain tumor invasion, it isalso necessary to perform studies using purified BTSCs,which may give rise to tumors with enhanced ability to in-filtrate.31,35,47,70,86

In xenograft mouse models, brain tumor cells derivedfrom purified BTSCs migrate from the ventral forebraintoward the olfactory bulb and along axon tracts to the dor-sal cortex,31,47 recapitulating the paths of neuroblasts de-rived from neurogenic astrocytes in the adult SVZ and in-terneurons derived from NPSCs in the embryonic ventralforebrain.1 This observation suggests that similar mecha-nisms may underlie the migration of BTSCs (or their prog-eny) and the migration of NPSCs and neurons. Consistentwith this, preliminary evidence suggests that genes impor-tant for neuronal migration, such as Slit2 and Lis1, mayalso be involved in the migration of medulloblastoma orglioma cells.73,80 In addition, transforming acid coiled-coilproteins (TACCs), which promote nuclear migration inembryonic NPSCs,81 have been implicated in many humancancers, including glioma.21,32,41,42 Finally, the ability oftransplanted NPSCs to track down brain tumor cells thathave migrated away from the tumor bulk in the rodentbrain84 suggests that NPSCs and BTSCs (or their progeny)may migrate through the same extracellular environment.

Conclusions and Future Directions

Since the identification of BTSCs several years ago,these cells have become promising targets for treating brain

tumors. The study of BTSCs has benefited tremendouslyfrom our knowledge of NPSCs, which may give rise toBTSCs through genetic mutations that deregulate asym-metric cell division (Fig. 1). Brain tumor stem cells andNPSCs reside in similar microenvironments enriched incapillaries, respond to similar signaling pathways, and giverise to daughter cells that migrate along similar paths(Table 1).

While great progress has been made in the field of mo-lecular and cell biology of BTSCs, many questions remainunanswered. First, although the findings of a number ofstudies are consistent with the possibility that deregulationof asymmetric division may convert NPSCs to BTSCs,convincing evidence is lacking. Next, how can we distin-guish BTSCs from normal NPSCs in BTSC-based braintumor therapies? Since similar molecular pathways maygovern the proliferation of BTSCs and normal NPSCs, tar-geting these pathways may not only inhibit brain tumori-genesis, but also impair endogenous neurogenesis, whichplays important physiological functions. Third, molecularmechanisms underlying the migration of BTSCs and theirprogeny remain largely unknown. Lastly, in most of thestudies of BTSCs, investigators have assumed that thesecells express a cell surface marker CD133 and grow nonad-herently in vitro. A recent study suggests, however, that asubset of BTSCs derived from glioblastomas grow ad-herently and may not express CD133.5 The significance ofthis subtype of BTSCs in brain tumorigenesis needs to befurther clarified.

References

1. Ayala R, Shu T, Tsai LH: Trekking across the brain: the journeyof neuronal migration. Cell 128:29–43, 2007

2. Bao S, Wu Q, Sathornsumetee S, Hao Y, Li Z, Hjelmeland AB,et al: Stem cell-like glioma cells promote tumor angiogenesisthrough vascular endothelial growth factor. Cancer Res 66:7843–7848, 2006

3. Bar EE, Chaudhry A, Lin A, Fan X, Schreck K, Matsui W, et al:Cyclopamine-mediated hedgehog pathway inhibition depletesstem like cancer cells in glioblastoma. Stem Cells 25:2524–2533,2007

4. Barr AR, Gergely F: Aurora-A: the maker and breaker of spindlepoles. J Cell Sci 120:2987–2996, 2007

5. Beier D, Hau P, Proescholdt M, Lohmeier A, Wischhusen J, Oef-ner PJ, et al: CD133(+) and CD133(-) glioblastoma-derived can-cer stem cells show differential growth characteristics and molec-ular profiles. Cancer Res 67:4010–4015, 2007



5

TABLE 1Comparison of NPSCs and BTSCs

Feature NPSCs BTSCs

molecular marker nestin, CD133 nestin, CD133gene expression profile NPSCs similar to NPSCsability to form neuro- yes yes

spheresability to self-renew yes yeslineage potential neurons, astrocytes, oligodendrocytes neurons, astrocytes, oligodendrocytesproliferation pathways BMPs, Notch, Shh, Wnt, growth factors BMPs, Notch, Shh, Wnt? growth factors?microenvironment associated w/ vasculature, communicate w/ endothelial associated w/ vasculature, communicate w/ endothelial cells

cellsmodel of division tight control of asymmetric vs symmetric division increased self-renewing symmetric division?migration daughter cells generated at ventral forebrain migrate to daughter cells generated at ventral forebrain migrate to

olfactory bulb & along axon tracts to dorsal cortex olfactory bulb & along axon tracts to dorsal cortex

6. Bello B, Reichert H, Hirth F: The brain tumor gene negatively reg-ulates neural progenitor cell proliferation in the larval central brainof Drosophila. Development 133:2639–2648, 2006

7. Bertossi M, Virgintino D, Errede M, Roncali L: Immunohisto-chemical and ultrastructural characterization of cortical plate mi-crovasculature in the human fetus telencephalon. Microvasc Res58:49–61, 1999

8. Betschinger J, Mechtler K, Knoblich JA: Asymmetric segregationof the tumor suppressor brat regulates self-renewal in Drosophilaneural stem cells. Cell 124:1241–1253, 2006

9. Bond J, Roberts E, Mochida GH, Hampshire DJ, Scott S, AskhamJM, et al: ASPM is a major determinant of cerebral cortical size.Nat Genet 32:316–320, 2002

10. Bond J, Roberts E, Springell K, Lizarraga SB, Scott S, HigginsJ, et al: A centrosomal mechanism involving CDK5RAP2 andCENPJ controls brain size. Nat Genet 37:353–355, 2005

11. Bowman SK, Neumüller RA, Novatchkova M, Du Q, KnoblichJA: The Drosophila NuMA Homolog Mud regulates spindle ori-entation in asymmetric cell division. Dev Cell 10:731–742, 2006

12. Buchman JJ, Tsai LH: Spindle regulation in neural precursors offlies and mammals. Nat Rev Neurosci 8:89–100, 2007

13. Calabrese C, Poppleton H, Kocak M, Hogg TL, Fuller C, HamnerB, et al: A perivascular niche for brain tumor stem cells. CancerCell 11:69–82, 2007

14. Cappello S, Attardo A, Wu X, Iwasato T, Itohara S, Wilsch-Brau-ninger M, et al: The Rho-GTPase cdc42 regulates neural progeni-tor fate at the apical surface. Nat Neurosci 9:1099–1107, 2006

15. Caussinus E, Gonzalez C: Induction of tumor growth by alteredstem-cell asymmetric division in Drosophila melanogaster. NatGenet 37:1125–1129, 2005

16. Chen HL, Panchision DM: Concise review: bone morphogeneticprotein pleiotropism in neural stem cells and their derivatives—alternative pathways, convergent signals. Stem Cells 25:63–68,2007

17. Ciani L, Salinas PC: WNTs in the vertebrate nervous system: frompatterning to neuronal connectivity. Nat Rev Neurosci 6:351–362, 2005

18. Clark PA, Treisman DM, Ebben J, Kuo JS: Developmental sig-naling pathways in brain tumor-derived stem-like cells. Dev Dyn236:3297–3308, 2007

19. Clement V, Sanchez P, de Tribolet N, Radovanovic I, Ruiz i Al-taba A: HEDGEHOG-GLI1 signaling regulates human gliomagrowth, cancer stem cell self-renewal, and tumorigenicity. CurrBiol 17:165–172, 2007

20. Conover JC, Notti RQ: The neural stem cell niche. Cell TissueRes 331:211–224, 2008

21. Conte N, Delaval B, Ginestier C, Ferrand A, Isnardon D, LarroqueC, et al: TACC1-chTOG-Aurora A protein complex in breast can-cer. Oncogene 22:8102–8116, 2003

22. Cox J, Jackson AP, Bond J, Woods CG: What primary micro-cephaly can tell us about brain growth. Trends Mol Med 12:358–366, 2006

23. Dakubo GD, Mazerolle CJ, Wallace VA: Expression of Notch andWnt pathway components and activation of Notch signaling inmedulloblastomas from heterozygous patched mice. J Neuro-oncol 79:221–227, 2006

24. Dai C, Celestino JC, Okada Y, Louis DN, Fuller GN, Holland EC:PDGF autocrine stimulation dedifferentiates cultured astrocytesand induces oligodendrogliomas and oligoastrocytomas fromneural progenitors and astrocytes in vivo. Genes Dev 15:1913–1925, 2001

25. Ehtesham M, Sarangi A, Valadez JG, Chanthaphaychith S, BecherMW, Abel TW, et al: Ligand-dependent activation of the hedge-hog pathway in glioma progenitor cells. Oncogene 26:5752–5761, 2007

26. Fan X, Matsui W, Khaki L, Stearns D, Chun J, Li YM, et al: Notchpathway inhibition depletes stem-like cells and blocks engraft-ment in embryonal brain tumors. Cancer Res 66:7445–7452,2006

27. Feng Y, Walsh CA: Mitotic spindle regulation by Nde1 controlscerebral cortical size. Neuron 44:279–293, 2004

28. Fish JL, Kosodo Y, Enard W, Pääbo S, Huttner WB: Aspm specif-ically maintains symmetric proliferative divisions of neuroepithe-lial cells. Proc Natl Acad Sci U S A 103:10438–10443, 2006

29. Fogarty MP, Emmenegger BA, Grasfeder LL, Oliver TG,Wechsler-Reya RJ: Fibroblast growth factor blocks Sonic hedge-hog signaling in neuronal precursors and tumor cells. Proc NatlAcad Sci U S A 104:2973–2978, 2007

30. Fuccillo M, Joyner AL, Fishell G: Morphogen to mitogen: themultiple roles of hedgehog signaling in vertebrate neural develop-ment. Nat Rev Neurosci 7:772–783, 2006

31. Galli R, Binda E, Orfanelli U, Cipelletti B, Gritti A, De Vitis S, etal: Isolation and characterization of tumorigenic, stem-like neu-ral precursors from human glioblastoma. Cancer Res 64:7011–7021, 2004

32. Gergely F: Centrosomal TACCtics. Bioessays 24:915–925, 200233. Götz M, Huttner WB: The cell biology of neurogenesis. Nat Rev

Mol Cell Biol 6:777–788, 200534. Hallahan AR, Pritchard JI, Chandraratna RA, Ellenbogen RG,

Geyer JR, Overland RP, et al: BMP-2 mediates retinoid-inducedapoptosis in medulloblastoma cells through a paracrine effect. NatMed 9:1033–1038, 2004

35. Hemmati HD, Nakano I, Lazareff JA, Masterman-Smith M,Geschwind DH, Bronner-Fraser M, et al: Cancerous stem cellscan arise from pediatric brain tumors. Proc Natl Acad Sci U S A100:15178–15183, 2003

36. Holland EC, Celestino J, Dai C, Schaefer L, Sawaya RE, FullerGN: Combined activation of Ras and Akt in neural progenitorsinduces glioblastoma formation in mice. Nat Genet 25:55–57,2000

37. Horvath S, Zhang B, Carlson M, Lu KV, Zhu S, Felciano RM, etal: Analysis of oncogenic signaling networks in glioblastomaidentifies ASPM as a molecular target. Proc Natl Acad SciU S A 103:17402–17407, 2006

38. Ignatova TN, Kukekov VG, Laywell ED, Suslov ON, Vrionis FD,Steindler DA: Human cortical glial tumors contain neural stem-like cells expressing astroglial and neuronal markers in vitro. Glia39:193–206, 2002

39. Kanamori M, Kawaguchi T, Nigro JM, Feuerstein BG, BergerMS, Miele L, et al: Contribution of Notch signaling activationto human glioblastoma multiforme. J Neurosurg 106:417–427,2007

40. Klezovitch O, Fernandez TE, Tapscott SJ, Vasioukhin V: Loss ofcell polarity causes severe brain dysplasia in Lgl1 knockout mice.Genes Dev 18:559–571, 2004

41. Lauffart B, Howell SJ, Tasch JE, Cowell JK, Still IH: Interactionof the transforming acidic coiled-coil 1 (TACC1) protein with ch-TOG and GAS41/NuBI1 suggests multiple TACC1-containingprotein complexes in human cells. Biochem J 363:195–200, 2002

42. Lauffart B, Vaughan MM, Eddy R, Chervinsky D, DiCioccio RA,Black JD, et al: Aberrations of TACC1 and TACC3 are associat-ed with ovarian cancer. BMC Womens Health 5:8, 2005

43. Le Bras B, Barallobre MJ, Homman-Ludiye J, Ny A, Wyns S,Tammela T, et al: VEGF-C is a trophic factor for neural progeni-tors in the vertebrate embryonic brain. Nat Neurosci 9:340–348,2006

44. Lee CY, Andersen RO, Cabernard C, Manning L, Tran KD,Lanskey MJ, et al: Drosophila Aurora-A kinase inhibits neuro-blast self-renewal by regulating aPKC/Numb cortical polarity andspindle orientation. Genes Dev 20:3464–3474, 2006

45. Lee CY, Robinson KJ, Doe CQ: Lgl, Pins and aPKC regulate neu-roblast self-renewal versus differentiation. Nature 439:594–598,2006

46. Lee CY, Wilkinson BD, Siegrist SE, Wharton RP, Doe CQ: Bratis a Miranda cargo protein that promotes neuronal differentiationand inhibits neuroblast self-renewal. Dev Cell 10:441–449, 2006

47. Lee J, Kotliarova S, Kotliarov Y, Li A, Su Q, Donin NM, et al: Tu-mor stem cells derived from glioblastomas cultured in bFGF and



EGF more closely mirror the phenotype and genotype of primarytumors than do serum-cultured cell lines. Cancer Cell 9:391–403,2006

48. Li Q, Ford MC, Lavid EB, Madri JA: Modeling the neurovascularniche: VEGF- and BDNF-mediated cross-talk between neuralstem cells and endothelial cells: an in vitro study. J Neurosci Res84:1656–1668, 2006

49. Louvi A, Artavanis-Tsakonas S: Notch signaling in vertebrateneural development. Nat Rev Neurosci 7:93–102, 2006

50. Mason I: Initiation to end point: the multiple roles of fibroblastgrowth factors in neural development. Nat Rev Neurosci 8:583–596, 2007

51. Merkle FT, Alvarez-Buylla A: Neural stem cells in mammaliandevelopment. Curr Opin Cell Biol 18:704–709, 2006

52. Nakada M, Nakada S, Demuth T, Tran NL, Hoelzinger DB, Be-rens ME: Molecular targets of glioma invasion. Cell Mol Life Sci64:458–478, 2007

53. Ohgaki H, Kleihues P: Genetic pathways to primary and secon-dary glioblastoma. Am J Pathol 170:1445–1453, 2007

54. Ohshiro T, Yagami T, Zhang C, Matsuzaki F: Role of cortical tu-mor-suppressor proteins in asymmetric division of Drosophilaneuroblast. Nature 408:593–596, 2000

55. Oka N, Soeda A, Inagaki A, Onodera M, Maruyama H, Hara A, etal: VEGF promotes tumorigenesis and angiogenesis of humanglioblastoma stem cells. Biochem Biophys Res Commun 360:553–559, 2007

56. Peng CY, Manning L, Albertson R, Doe CQ: The tumor-suppres-sor genes lgl and dlg regulate basal protein targeting in Drosophilaneuroblasts. Nature 408:596–600, 2000

57. Phillips HS, Kharbanda S, Chen R, Forrest WF, Soriano RH, WuTD, et al: Molecular subclasses of high-grade glioma predictprognosis, delineate a pattern of disease progression, and resem-ble stages in neurogenesis. Cancer Cell 9:157–173, 2006

58. Piccirillo SG, Reynolds BA, Zanetti N, Lamorte G, Binda E,Broggi G, et al: Bone morphogenetic proteins inhibit the tumori-genic potential of human brain tumor-initiating cells. Nature 444:761–765, 2006

59. Prokop A, Technau GM: Normal function of the mushroom bodydefect gene of Drosophila is required for the regulation of thenumber and proliferation of neuroblasts. Dev Biol 161:321–337,1994

60. Purow BW, Haque RM, Noel MW, Su Q, Burdick MJ, Lee J, etal: Expression of Notch-1 and its ligands, Delta-like-1 and Jag-ged-1, is critical for glioma cell survival and proliferation. CancerRes 65:2353–2363, 2005

61. Ramírez-Castillejo C, Sánchez-Sánchez F, Andreu-Agulló C,Ferrón SR, Aroca-Aguilar JD, Sánchez P, et al: Pigment epitheli-um-derived factor is a niche signal for neural stem cell renewal.Nat Neurosci 9:331–339, 2006

62. Rao G, Pedone CA, Valle LD, Reiss K, Holland EC, Fults DW:Sonic hedgehog and insulin-like growth factor signaling synergizeto induce medulloblastoma formation from nestin-expressing neu-ral progenitors in mice. Oncogene 23:6156–6162, 2004

63. Rolls MM, Albertson R, Shih HP, Lee CY, Doe CQ: DrosophilaaPKC regulates cell polarity and cell proliferation in neuroblastsand epithelia. J Cell Biol 163:1089–1098, 2003

64. Sanada K, Tsai LH: G protein betagamma subunits and AGS3control spindle orientation and asymmetric cell fate of cerebralcortical progenitors. Cell 122:119–131, 2005

65. Sanai N, Alvarez-Buylla A, Berger MS: Neural stem cells and theorigin of gliomas. N Engl J Med 353:811–822, 2005

66. Schmierer B, Hill CS: TGFbeta-SMAD signal transduction: mo-lecular specificity and functional flexibility. Nat Rev Mol CellBiol 8:970–982, 2007

67. Shen Q, Goderie SK, Jin L, Karanth N, Sun Y, Abramova N, et al:Endothelial cells stimulate self-renewal and expand neurogenesisof neural stem cells. Science 304:1338–1340, 2004

68. Shih AH, Holland EC: Notch signaling enhances nestin expressionin gliomas. Neoplasia 8:1072–1082, 2006

69. Shih I, Wang TL: Notch signaling, gamma-secretase inhibitors,and cancer therapy. Cancer Res 67:1879–1882, 2007

70. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, etal: Identification of human brain tumor initiating cells. Nature432:396–401, 2004

71. Su X, Gopalakrishnan V, Stearns D, Aldape K, Lang FF, Fuller G,et al: Abnormal expression of REST/NRSF and Myc in neuralstem/progenitor cells causes cerebellar tumors by blocking neu-ronal differentiation. Mol Cell Biol 26:1666–1678, 2006

72. Sutter R, Yadirgi G, Marino S: Neural stem cells, tumor stem cellsand brain tumors: dangerous relationships? Biochim BiophysActa 1776:125–137, 2007

73. Suzuki SO, McKenney RJ, Mawatari SY, Mizuguchi M, MikamiA, Iwaki T, et al: Expression patterns of LIS1, dynein and theirinteraction partners dynactin, NudE, NudEL and NudC in humangliomas suggest roles in invasion and proliferation. Acta Neu-ropathol 113:591–599, 2007

74. Takai N, Hamanaka R, Yoshimatsu J, Miyakawa I: Polo-like kin-ases (Plks) and cancer. Oncogene 24:287–291, 2005

75. Taylor MD, Poppleton H, Fuller C, Su X, Liu Y, Jensen P, et al:Radial glia cells are candidate stem cells of ependymoma. CancerCell 8:323–335, 2005

76. Vescovi AL, Galli R, Reynolds BA: Brain tumor stem cells. NatRev Cancer 6:425–436, 2006

77. Virgintino D, Maiorano E, Errede M, Vimercati A, Greco P, Sel-vaggi L, et al: Astroglia-microvessel relationship in the develop-ing human telencephalon. Int J Dev Biol 42:1165–1168, 1998

78. Wang H, Ouyang Y, Somers WG, Chia W, Lu B: Polo inhibitsprogenitor self-renewal and regulates Numb asymmetry by phos-phorylating Pon. Nature 449:96–100, 2007

79. Wang H, Somers GW, Bashirullah A, Heberlein U, Yu F, Chia W:Aurora-A acts as a tumor suppressor and regulates self-renewal ofDrosophila neuroblasts. Genes Dev 20:3453–3463, 2006

80. Werbowetski-Ogilvie TE, Seyed Sadr M, Jabado N, Angers-Loustau A, Agar NY, Wu J, et al: Inhibition of medulloblastomacell invasion by Slit. Oncogene 25:5103–5112, 2006

81. Xie Z, Moy LY, Sanada K, Zhou Y, Buchman JJ, Tsai LH:Cep120 and TACCs control interkinetic nuclear migration and theneural progenitor pool. Neuron 56:79–93, 2007

82. Yamada SM, Yamada S, Hayashi Y, Takahashi H, Teramoto A,Matsumoto K: Fibroblast growth factor receptor (FGFR) 4 corre-lated with the malignancy of human astrocytomas. Neurol Res24:244–248, 2002

83. Yamada SM, Yamaguchi F, Brown R, Berger MS, Morrison RS:Suppression of glioblastoma cell growth following antisense oli-gonucleotide-mediated inhibition of fibroblast growth factorreceptor expression. Glia 28:66–76, 1999

84. Yip S, Sabetrasekh R, Sidman RL, Snyder EY: Neural stem cellsas novel cancer therapeutic vehicles. Eur J Cancer 42:1298–1308, 2006

85. Yu F, Kuo CT, Jan YN: Drosophila neuroblast asymmetric celldivision: recent advances and implications for stem cell biology.Neuron 51:13–20, 2006

86. Yuan X, Curtin J, Xiong Y, Liu G, Waschsmann-Hogiu S, FarkasDL, et al: Isolation of cancer stem cells from adult glioblastomamultiforme. Oncogene 23:9392–9400, 2004

87. Zhu Y, Guignard F, Zhao D, Liu L, Burns DK, Mason RP, et al:Early inactivation of p53 tumor suppressor gene cooperating withNF1 loss induces malignant astrocytoma. Cancer Cell 8:119–130, 2005

Manuscript submitted December 15, 2007.Accepted January 3, 2008.Address correspondence to: Zhigang Xie, Ph.D., Department of

Neurosurgery, Boston University School of Medicine, 720 HarrisonAvenue, Suite 7600, Boston, Massachusetts 02118. email:[email protected].



7

Documents

Molecular and cell biology of brain tumor stem cells: lessons from neural progenitor/stem cells