Expression of Neurogenic Basic Helix-Loop-Helix Genes in Primitive Neuroectodermal Tumors Robert C. Rostomily, Olivia Bermingham-McDonogh, Mitchel S. Berger, et al. Cancer Res 1997;57:3526-3531. Published online August 1, 1997.
Updated Version
Citing Articles
E-mail alerts Reprints and Subscriptions Permissions
Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/57/16/3526
This article has been cited by 14 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/57/16/3526#related-urls
Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at
[email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at
[email protected].
Downloaded from cancerres.aacrjournals.org on July 16, 2011 Copyright © 1997 American Association for Cancer Research
[CANCER RESEARCH 57. 3526—3531.August 15, 19971
Expression of Neurogenic Basic Helix-Loop-Helix
Genes in Primitive
Neuroectodermal Tumors1 Robert C. Rostomily,2 Olivia Bermingham-McDonogh, Mitchel S. Berger, Stephen J. Tapscott, Thomas A. Reh, and James
M. Olson3
Departments of Neurological Surgery (R. C. R., M. S. B.J, Biological Structure fR. C. R., 0. B-M., T. A. RI, and Neurology [S. .1. Ti, The University of Washington School of Medicine, Seattle, Washington 98195; Division of Molecular Medicine, The Fred Hutchinson Cancer Research Center, Seattle, Washington 98109 (S. J. T., J. M. 0.1; and Divisions ofNeurological Surgery (R. C. R.] and Pediatric Hematology-Oncology fJ. M. 0.], The Children ‘s Hospital and Medical Center, Seattle, Washington 98105
ABSTRACT The basic helix-loop-helix (bHLH) class of transcription factors plays a pivotal
role in tissue-specific
determination
and differentiation.
Moreover,
dysregulated expression or loss of function of these factors contributes to leukemogenesis and solid tumor development. Neurogenesis is regulated
by genes of the NEUROD/atonaland ACHAETESCUTE families. We analyzed expression of human NEURODJ, NEUROD2, NEUROD3, and ACHAETE SCUTE 1 (HASHJ) in cerebellar and cerebral primitive neu
commit progenitors to a neural fate and induce terminal neuronal differentiation. The NeuroD family, bike the myogenic MyoD family, consists of members of class I bHLH transcription factors that regulate tissue specific differentiation (3—8). MyoD family members orchestrate myogenic differentiation and are universally expressed in rhab domyosarcomas, yet rhabdomyosarcomas evade terminal differen
tiationat beastin part by inhibitingthe activityof MyoD(9—12).
variety of neuroectodermal tumors by Northern analysis and in situ hybridization. NEURODJ was expressed in each of the 12 medulloblas toma specimens, whereas NEUROD2 and NEUROD3/neurogenin were
On the basis of similarities between NeuroD and MyoD, we hypothesized that NeuroD family members may be expressed in medubbobbastomas as markers of a neurogenic lineage. Because distinct populations of neurons express different neurogenic bHLH
expressed
transcription
roectodermal
tumors
(PNETs),
in partly
gliomas,
overlapping
and
cell lines
derived
subsets of medulloblastomas.
from
a
All of the
tumors that presented with distant metastases expressed NEUROD3. The only other NEUROD3-positive tumor progressed early in treatment. Hu man ACHAETESCUTE homologue (HASHJ) was not expressed in medal loblastomas (infratentorial PNETs) but was expressed in three of five supratentorial
PNETs.
Neuroectodermal
tumor
cell lines derived
from
other sites (e.g., neuroblastoma and retinoblastoma) expressed NeuroD and ACHAETE SCUTE family members. No NEUROD message was detected in glial tumors or cell lines. Neurogenic bHLH transcription factor expression patterns suggest that specific family members may contribute to or reflect biological differences that arise during malignant transformation.
INTRODUCTION Molecular events associated with medulboblastoma and other child hood brain tumor development are very poorly understood. Currently, no
biological
markers
facilitate
stratification
of
treatment
groups,
assist with prognosis, or serve as targets for therapeutic intervention (1).
Medullobbastomas express neuronab intermediate filaments, synap tic vesicle
proteins,
growth
factor receptors,
and adhesion
molecules,
suggesting that they arise from neurobbasts that escape terminal dif ferentiation (2). The mechanisms by which cells escape terminal differentiation can now be addressed because the NeuroD family of bHLH4 proteins was recently identified. NeuroD and the rebated achaete scute transcription factors regulate activation of genes that Received1/14/97; accepted6/10/97. The costs of publication of this article were defrayed in part by the payment of page
factors
during
that bHLH patterns would different neuronal lineages. NEUROD
and ACHAETE
development,
we further
vary between SCUTE
PNETs
genes are expressed
hypothesized
arising
from
in overlap
ping but distinct populations of neuroblasts during neuronab develop ment. A population of immature proneurab cells that reside in the ventricular zone first express NEUROD3/neurogenin at mouse em bryonic day 9.5 (E9.5; Refs. 4 and 6). Ventricular zone cells migrate to the periventricubar zone, where NEURODJ is first expressed (3, 6). NeuroD2 expression commences on Ell.5 (4). By E16, NEUROD3 can no longer be detected by Northern analysis of RNA derived from brain tissue (4). In contrast, NEURODJ and NEUROD2 are expressed throughout cerebeblar development and in adult cerebellum, primarily in the granule cell layer (3, 4). The importance of these genes in neuronal development was established by their ability to induce ectopic neurogenesis and premature neuronal differentiation when injected into Xenopus embryos (3, 4, 6). ACHAETE SCUTE homo bogues make up a distinct family of neurogenic bHLH transcription factors that share homology with the NEUROD family. Achaete scute null mutant mice failed to develop sympathoadrenal neural cells and olfactory neurons (13). Mammalian achaete scute (mashi) expression was reported in cerebellum, but it remains unclear whether it is expressed in the same neural precursors as neuroD family members (14). MATERIALS
AND METHODS
Surgical specimens from 13 medulbobbastomas, 5 supratentorial PNEIs, and 6 pediatric gliomas and cell lines derived from 14 neuroectodermal tumors and
charges. This article must therefore be hereby marked advertisement in accordance with
3 gliomas
18 U.S.C. Section 1734 solely to indicate this fact.
HASHJ by Northern analysis. Because surgicaltissue quantity was limited, not
t This work
was supported
by NINDS
NS 09458 (to R. C. R), NIH Grants ROl
28308 and NS 30304 (to T. A. R.), and the Seattle Children's
NS
Hospital, “Jesse'sPerfect
Peach― Neurooncology Research Fund (to J. M. 0). R. C. R. was supported by the Charles A. Elsberg Fellowship in Neurological Surgery of the New York Academy of Medicine
and by Neurosurgery Training Program Grant N07l44. S. J. T. was supported by an American Cancer Society grant. J. M. 0. was supported by the Emily Dorfman Founda tion through the American Brain Tumor Association. 2 Present
address:
59th
Medical
WingIMKFN,
Department
of
Neurosurgery,
whom
requests
for
reprints
should
be addressed,
at the Fred
Hutchinson
Cancer
Research Center, Mailstop C3-l68, 1100 Fairview Avenue, Seattle, WA 98109. 4 The abbreviations used are: bHLH, basic helix-loop-helix; ectodermal tumor.
evaluated
for expression
of NeuroD
family
members
and
all bHLH members were tested on each specimen. In situ analysis was used to confirm Northern data and determine whether expression was ubiquitous in
tumor cells. Studies were done in accordance with the Children's Hospital Human Subjects Institutional Review Board guidelines. Human tumor tissue samples
were
obtained
at craniotomy.
Samples
were
either
snap
frozen
in
liquid nitrogen and stored at — 80°C,or they were placed in tissue culture
Wilford
Hall Medical Center, 2200 Bergquist Drive, STE1, Lackland AFB, San Antonio, TX 78246-5300. 3 To
were
media and then mechanically dissociated and filtered, and the centrifuged cell
pellet was snap frozen and stored at — 80°C. Cell Culture. The cell lines listed in Table2 were obtainedfromAmerican Type Culture Collection with the following exceptions: NLF, NGP, and NMB
PNET, primitive neuro
(G. Brodeur, University of Pennsylvania, Philadelphia, PA); UW228, I98G, 3526
Downloaded from cancerres.aacrjournals.org on July 16, 2011 Copyright © 1997 American Association for Cancer Research
NEURODAND HASHI IN NEUROECI'ODERMALTUMORS
and SNB 19 (J. Silber and M. Berger, University ofWashington, Seattle, WA). Y79, WERI, H209, and H82 were grown in RPM! 1640 (Life Technologies, Inc.) with 10% FBS (Hyclone). The remainder were grown in Eagle's MEM with Earle's balanced salt solution, non-essential amino acids (Life Technol ogies, Inc.), 1 mist pyruvate (Life Technologies,
MMMMM 1 2345 A
Inc.), and 10% fetal bovine
serum or bovine calf serum. Northern Analysis. Total RNA was extracted from confluent cell lines and tissue specimens or cell pellets using Trizol reagent (Life Technolo gies, Inc.). Northern
analysis
was performed
according
to published
-2.6
pro
cedures with minor modifications (15, 16). Due to limited sample quantity, NEUROD2 and NEUROD3 expression was determined for some specimens on stripped membranes that had previously been used for NEURODI and HASHJ expression,
respectively.
Stripped filters exposed to film for 4 days
were found to be negative for NEURODJ bands prior to reprobing. Human NEURODJ message was detected by either a full-length 1.6-kb cDNA probe or an 800-bp divergent probe (carboxyl to bHLH domain) isolated with Kpn digestion (3). The human NEUROD2 probe spanned 500 bp in the 3' divergent
region between
Pst and Sac! excision
sites (4). The human
NEUROD3/neurogenin probe spanned 800 bp, from the Sma to the Pst restriction sites of the clone SK2OA1 (4). HASHJ probe spanned the bHLH domain using the PCR primers GTCACAAGTCAGCGCCCAAG and CGACGAGTAGGATGAGACCG.HASHJfindings were confirmed with a 0.9-kb fragment isolated from a human fetal cDNA library (Stratagene) that was homologous to the published sequence in the 3' untranslated region (17). None of the divergent
probes cross-reacted
1.1-IL -2.6
C
-1.9
with RNA from other
family members under the conditions used. In Situ Hybridization. Paraformaldehyde-fixed frozen tumor specimens
S
D
were sectioned at l2-g.@mintervals and then pretreated with 4% proteinase K, acetic anhydride, and 0.1% Triton X-lOO (NEURODI) or 4% proteinase K and
acetic anhydride (NEUROD2 and NEUROD3). In situ analysis was performed according to a published procedure except that the concentration of NBT was reduced
10-fold
for NEUROD2
and NEUROD3
assays
-2.6
(18).
E$j)S@
RESULTS Neurogenic bHLH Genes in Human Tumor Specimens. Neu roDi was expressed in each of the 12 medulbobbastoma surgical specimens analyzed (Fig. 1; Table 1). Specimens 2—4,6, and 10 also expressed NEUROD2, and specimens 1, 3, 5, 8 and 9 ex pressed NEUROD3/neurogenin. Four of the patients whose tumors expressed NEUROD3/neurogenin presented with disseminated dis ease, and the fifth developed clinical and radiographic progression in the 3-week interval between tumor resection and initiation of radiation therapy. Although the association between NEUROD3/ neurogenin expression and medulboblastoma dissemination is sig nificant using the Fisher exact test, a larger study will be necessary to conclusively evaluate the prognostic implications of NEUROD3/ neurogenin expression. In situ hybridization using antisense probes for each NEUROD gene revealed robust staining in the characteristically scant cytoplasm, whereas background hybridization with sense probes was minimal (Fig. 2 and data not shown). In an expressing tumor, nearly all of the cells were positive for the particular NeuroD mRNA, consistent with the homogeneous nature of medulboblastomas. Esthesioneuroblas
tomasamplesassayedbyinsituanalysisdemonstrated patchyexpres sion of NEURODJ and NEUROD3/neurogenin, suggesting that cx pression may be more heterogeneous in some tumors (data not shown). Medulbobbastoma (infratentoriab PNET) samples probed with sequence from a human clone of HASH! were uniformly negative (Fig. 1; Table 1). In contrast, three of five supratentorial PNETs expressed HASh (Fig. 3; Table 1). Conversely, all medulboblas toma samples expressed NEURODJ, but only one of four supra tentorial PNETs expressed NEURODI . These results suggest that although medulboblastoma and supratentoriab PNET are similar
F@ftØ Fig. 1. Representative
Northern analyses of NEURODI
(A), NEUROD2 (B), NEU.
ROD3/neurogenin (C), and HASH! (D) expression in medulloblastoma surgical speci mens. E and F, loading variability of 20 @g of RNA as assayed by a probe to the human glyceraldehyde-3-phosphate deshydrogenase gene (GAPDH). Lanes (left to right), pa tients Mt—MS.These and other patient data are summarized in Table I . Molecular weight
markers are indicated on the right.
histologically, they probably derive from distinct progenitor cell populations. Six pediatric gliomas, including cerebelbar pilocytic astrocytoma, cerebral pilocytic astrocytoma, ependymoma, and malignant astrocy toma specimens, and three human glioma cell lines failed to express any of the neurogenic bHLH messages (Table 2 and data not shown). Neurogenic bHLH Expression In Neuroectodermal Tumor Cell Lines. We next evaluated cell lines derived from neuroectodermal tumors to determine the patterns of neurogenic bHLH expression. Northern analysis was performed on thirteen cell lines derived from tumors of presumed neuroectodermal origin. Consistent with patient samples and developmental data discussed above, D283 and D34l metastatic medulloblastoma cell lines expressed NEURODI and NEUROD3/neurogenin (Table 2). Although the medulloblastoma cell lines did not express HASHI, it was notable that the neuroblastoma cell lines SKN-SH, NGP, NMB, and 1MR32 expressed HASHJ, a gene that is critical for sympathoadrenal lineage development (13). SKN-SH, NGP, and 1MR32 coexpressed NEURODI, but NEUROD2
3527
Downloaded from cancerres.aacrjournals.org on July 16, 2011 Copyright © 1997 American Association for Cancer Research
NEURODAND HASH! IN NEUROECTODERMALTUMORS
medulloblastomaNorthern
Table I NEUROD genes and HASH! in
analysis was performed on 20 @sg of RNA from 12 patients with cerebellar PNETs and 5 patients with supratentorial PNETs using cDNA probes to NEUROD!, NEUROD2, NEUROD3. or HASH!. Data from patientsMI—MSare shown in Fig. I, and data from patientsMb, Ml3, and Sl—S5are shown in Fig. 3. Sampleswere scoredpositive if the
appropriate —80°C.DistantPatient size band was present after ovemight hybridization under high-stringency conditions followed by 24—72h of film exposure at HASH!Medulloblastomas Age PNETs)Ml
Sex
metastasis
NEURODI
NEUROD2
NEUROD3
(infrasentorial
—M2
S
M―
Yes
+
—
+
—M3 -M4
7 9
M M
No Yes
+ +
+ +
— +
10
M
No
+
+
—
to
M
No―
+
—
+
—
M? —M6
—M7
I
F
No
+
+
-M8 —M9 -MlO
7 13 7
M M F
No Yes Yes
+ + ND
ND ND
+ +
9
M
No
+
+
ND
8 6 3
M M M
No No No
+ + +
ND
ND ND ND
SI +52
17
M
ND
-
ND
ND
+S3' -54 +S5
2 2 3 7
M M M F
ND ND ND ND
ND + -
ND ND ND ND
ND ND ND ND
-Mll -Ml2 -Ml3 -Supratentorial
—
PNETs
aM.male: F,female, ND, not done. I, Tumor
progressed
( Recurrent
tumor
early in treatment. specimen
from
patient
52.
and NEUROD3 were absent in neuroblastoma cell lines. None of the NEUROD
-
mRNAs
were detectable
by Northern
analysis
other than those tested (e.g. , atonal homobogues) or that expression is below the limit of detection by Northern analysis. Retinoblastoma cell lines expressed NEUROD genes but did not express HASh. Y79, a cell line derived from familial retinobbastoma,
in DAOY
and UW228 medubboblastoma cells or NLF and SKN-MC neuroblas toma cells. It is possible that these cell lines express bHLH proteins
;A
Fig. 2. In situ analysis of NEUROD!, NEUROD2, and NEUROD3 in medulloblastoma samples. Top row, patient samples Ml0 (A), M4 (B), and M3 (C) stained with H&E. Bottom row, in situ analysis using antisense digoxin-labeled riboprobes NEUROD! on sample MlO (D), NEUROD2 on sample M4 (E), and NEUROD3 on sample M3 (F). Ten-gsm markers are shown in bottom left comer of each specimen in the top row. 3528
Downloaded from cancerres.aacrjournals.org on July 16, 2011 Copyright © 1997 American Association for Cancer Research
NEUROD AND HASH! IN NEUROECFODERMAL TUMORS
MSMSSSS
lines;
(c) all of the medubbobbastomas
1
312345
tisera
C
patients
who presented
reactive
to
neurod2
demonstrated
that
protein
expression
correlated with in situ and Northern analysis data.6 Every tumor sample derived from neuroectodermal tissue cx pressed one or more neurogenic bHLH genes, whereas no glial tumors expressed NeuroD family members. Medulboblastomas expressed
-2 6
11
NEURODJ, NEUROD2, and NEUROD3/neurogenin, allofwhich are expressed in developing cerebellum or cerebellar precursors (3, 4, 6). Likewise, neuroblastomas expressed HASH] and NEURODI, both of which
•0Sa
D,...@,..
Fig. 3. Northem analysis ofNEUROD! and HASH expression in supratentorial PNETs. A, NEUROD! expression in medulloblastoma samples MlO and Ml3 and in supratentorial
PNET samples S1-S5; B, HASH! expression in corresponding samples; C and D, GAPDH expression on lanes represented in rows A and B, respectively.
expressed NEURODJ and NEUROD3/neurogenin. WERI, a sponta neous retinoblastoma cell line, expressed NEURODJ but no others. NEURODJ is expressed in developing and mature retina, and retinal expression of NEUROD2 and NEUROD3/neurogenin is not yet known.5 Achaete scute homobogues are expressed in late but not early retinal progenitors (19, 20). The lineage from which small cell lung cancers arise remains controversial. The H82 cell line expressed NEURODJ and NEU ROD3/neurogenin, and H209 cells expressed HASHJ and NEUROD3 (Table 2). Additional experiments are needed to determine whether neurogenic bHLH factors are expressed because these tumors arise from neuroendocrine cells or whether NEUROD/ACHAETE SCUTE genes are deregulated in these tumors conferring a neuronal pheno type to cells derived from a nonneuronab lineage. NeuroD2 was not expressed in any neuroectodermal tumor line tested. It is possible that none of the cell lines tested were derived from NEUROD2-expressing tumors. Alternatively, NeuroD2 expres sion might confer a growth disadvantage to cells so that they are selected against in culture. DISCUSSION
(a) NEUROD] was expressed in all medulboblastomatumorsand two medublobbastoma cell lines; (b) NEUROD2 and NEUROD3/neuroge nm were expressed in subsets of medulboblastoma tumor specimens, and NEUROD3/neurogenin was expressed in medulboblastoma cell and T. A. Reh, unpublished
data.
are expressed
in the developing
autonomic
nervous
system
(3,
13, 14). In this study, HASH] expression distinguished some suprat entorial PNETs from infratentonal PNETs (medubloblastoma). Cur rently, PNETs from both sites are classified together and treated similarly, although they respond differently to therapy (1, 2 1—23). Thus, along with other genes expressed in restricted regions of the developing nervous system, such as pax, homeobox, and zinc finger genes, bHLH genes may provide clues to the lineage of tumor origin (24—26). Transcription factors contribute to malignant transformation by either dysregulation of factors that enhance growth and loss of cell cycle control (oncogenes) or by inactivation of transcription factors that normally regulate cell cycle or terminal differentiation (tumor suppressors or anti-oncogenes). bHLH family members have been described in each of these roles, although they have not traditionally been called proto-oncogenes or tumor suppressors. For example, chromosomal translocations of the bHLH genes E2A and Ta! to the DNA-binding domain of a homeobox gene and to the enhancer region of a T-cell receptor, respectively, result in leukemic transformation, presumably by inducing transcription of inappropriate target genes (27—29).In contrast, MyoD normally maintains a tumor suppressor robe by orchestrating myogenic cell cycle control and terminal differ entiation, but its function is diminished in rhabdomyosarcomas by expression of MyoD repressors and by limited availability of neces sary cofactors
(1 1, 12).
Within the limits of Northern analysis sensitivity, NEUROD3/ neurogenin has been described in two populations: the immature proneural cells that reside in the mitotically active ventricular zone and medulloblastoma samples from patients who presented with dis tant metastasis or who rapidly progressed (4, 7). This raises the question of whether NEUROD3/neurogenin serves a different role in normal development and malignancy than the other neurogenic bHLH factors tested. We previously localized NEURODJ, NEUROD2, and NEUROD3/neurogenin to chromosome regions 2q32, 17ql2, and 5q23—31, respectively (30, 3 1). Comparative genomic hybridization analysis
In this study, we examined the expression of neurogenic bHLH transcription factors in neuroectodermal tumors and cell lines by Northern analysis and in situ hybridization. We found the following:
5 0. Bermingham-McDonogh
from
with early signs of biologically aggressive disease expressed NEU ROD3/neurogenin; (d) NEUROD family members were also cx pressed in other neuroectodermal cell lines, including neuroblastomas, retinoblastomas, and small cell lung cancer; and (e) HASHJ was not expressed in medulloblastoma tumor samples or cell lines but was expressed in supratentorial PNETs and cell lines derived from periph eral nervous system neuroectodermal tumors, including neuroblas toma and small cell lung cancer. Preliminary experiments using an
of medulboblastoma
samples
demonstrated
gain of distal
5q
in 3 of 18 cases, consistent with the possibility that NeuroD3 is amplified or translocated in a subset of medulboblastomas (32). The more downstream neuronal differentiation factors, NEURODJ and NEUROD2, are more reminiscent of MyoD in rhabdomyosar coma. Whether these transcription factors regulate terminal differen tiation and induce cell cycle control in mammalian cells remains to be determined.
6 J@ M.
Olson,
If so, then the paradoxical
unpublished
observations.
3529
Downloaded from cancerres.aacrjournals.org on July 16, 2011 Copyright © 1997 American Association for Cancer Research
expression
of these genes
in
NEUROD
linesRNA
AND HASH)
IN NEUROECTODERMAL
4. McCormick,
Table 2 NEUROD genes and HASH! in cell
activation potentials within the neuroD gene family. Mol. Cell. Biol., !6: 5792—5800, 1996. 5. Kume, H., Maruyama, K., Tomita, T., Iwatsubo, T., Saido, T. C., and Obata, K. Molecular cloning of a novel basic helix-loop-helix protein from the rat brain. Biochem. Biophys. Rca. Commun., 2!9: 526—530, 1996.
byNorthem cancer, and glioma cell lines near confluence. Twenty @sg of RNA were analyzed
analysis and scored following ovemight hybridization at high stringency fol lowed by 24—72h film—80°C.NEURODI exposure at NEUROD2
M. B., Tamimi, R. M., Snider, L., Asakura, A., Bergstrom, D., and
Tapscott. S. J. NeuroD2 and neuroD3: distinct expression patterns and transcriptional
celllung was harvested from medulloblastoma, neuroblastoma, retinoblastoma, small
HASH!MedulloblastomaD283
TUMORS
NEUROD3
6. Ma, Q., Kintner, C., and Anderson, D. J. Identification of neurogenin, a vertebrate —D34l + —UW228 + -DAOY -NeuroblastomaSKN-SH -
— — -
+ + -
+NLF -NGP
+ -
-
-
+NMB
+
-
-
+1MR32 +SKN-MC
+
—
—
-
-
-
-RetinoblastomaY79
+
-
+
—Small
—WERI
+
—
—
cancerH82 cell lung —H209 +GliomaSNBI9
+ -
— -
+ +
—T98G -SF767 —a
— —
ND,
— NDa —
neuronal determination gene. Cell, 87: 43—52,1996. 7. Weintraub, H., Davis, R., Tapscott, S., Thayer, M., Krause, M., Benezra, R.,
Blackwell, T. K., Tumer, D., Rupp, R., Hollenberg, S., Zhuang, Y., and Lassar, A. The myoD gene family: nodal point during specification of the muscle cell lineage. Science (Washington DC), 25!: 761—766,1991. 8. Tapscott, S. J., Davis, R. L., Thayer, M. J., Cheng, P-F., Weintraub, H., and Lassar,
A. B. MyoDl : a nuclear phosphoprotein requiring a Myc homology region to convert fibroblasts to myoblasts. Science (Washington DC), 242: 405—411, 1988.
9. Clark, J., Rocques, P. J., Braun, T., Bober, E., Arnold, H. H., Fisher, C., Fletcher, C., Brown, K., Gusterson, B. A., Carter, R. L., and Cooper, C. S. Expression of members of the myfgene family in human rhabdomyosarcomas. Br. J. Cancer, 64: 1039—1042, 1988. 10. Tonin, P. N., Scrable, H., Shimada, H., and Cavenee, W. K. Muscle-specific gene expression in rhabdomyosarcomas and stages of human fetal skeletal muscle devel
opment. Cancer Res., 5!: 5100—5106,1991. I I . Tapscott, S. J., Thayer, M. J., and Weintraub, H. Deficiency in rhabdomyosarcomas of a factor required for MyoD activity and myogenesis. Science (Washington DC). 259: 1450—1453,1993. 12. Fiddler, T. A., Smith, L., Tapscott, S. J., and Thayer, M. J. Amplification of MDM2 inhibits MyoD-mediated myogenesis. Mol. Cell. Biol., 16: 5048—5057, 1996. 13. Guillemot, F., La L-C., Johnson, J. E., Auerback, A., Anderson, D. J., and Joyner,
— ND —
A. L. Mammalian achaete-scute homolog-l is required for the early development of
not done.
14.
neuroectodermal tumors suggests that these cancer cells may gain a growth advantage by limiting the function of bHLH transcription factors. Prior to the isolation of NEURODJ, a study of MyoD function in neuroblastoma and medulboblastoma cell lines suggested that rhab domyosarcomas and neuroectodermal tumors may share mechanisms of evading bHLH transcription control (33). In addition, NEUROD2 expression or function might be affected by rearrangements associated with isochromosome 17q, the most common cytogenetic abnormality in medubboblastomas (32, 34). The study of mammalian neurogenic bHLH transcription factors promises to be rather complex. Multiple proneurab bHLH genes work combinatorialby with each other and with other devebopmen tal transcription activators to orchestrate development of the rela tively simple neural structures in Drosophila (35—37). From the more intricate vertebrate nervous system, a number of neurogenic bHLH proteins have been described (3—6, 19, 38—44). Basic information about these proteins, such as dimerization partners, DNA binding preferences, and biological targets, remains to be elucidated. As that work progresses, the current study shows that NeuroD family members may be useful diagnostically and pro vides a rationale
for a study addressing
of NEUROD3/neurogenin
the prognostic
implications
to this
work;
17.
Cumaraswamy, A., Borges, M., and Nelkin, B. D. Identification of a human achaete
sautehomologhighlyexpressedin neuroendocrinetumors.Proc.Natl.Acad.Sci. USA, 90: 5648—5652,1993. I 8. Schaeren-Wiemers, N., and Gerfin-Moser, A. A single protocol to detect transcripts of various types and expression levels in neural tissue and cultured cells: in situ hybridization using digoxigenin-labelled cRNA probes. Histochemistry. !00: 431— 440, 1993. 19. Jasoni, C. L., Walker, M. B., Morris, M. D., and Reh, T. A. A chicken achaete-scute
homolog (CASH-l) is expressed in a temporally and spatially discrete manner in the developing nervous system. Development (Camb.), 120: 769—783, 1994. 20. Jasoni, C. L., and Reh, T. A. Temporal and spatial pattem of MASH-I expression in the developing rat retina demonstrates progenitor cell heterogeneity. J. Comp. Neu
rol., 319—327,1996. 21. Rorke, L. B., Gilles, F. H., Davis, R. L., and Becker, L. E. Revision of World Health Organization classification of brain tumors for childhood brain tumors. Cancer (Phila.), 56: 1869—1886, 1996.
22. Packer, R. J., Sutton, L. N., Elterman, R., Lange, B., Goldwein, J., Nicholson, H. S., Mulne, L., Boyets, J., Dangio, G., Wechslerjentzsch, K., Reaman, G., Cohen, B. H., Bruce, D. A., Rorke, L. B., Molloy, P., Ryan, J., Lafond, D., Evans, A. E., and Schut,
L. Outcome for children with medulloblastoma treated with radiation and cisplatin, CCNU, and vincristine chemotherapy. J. Neurosurg., 8!: 690—698,1994. S. B., Wisoff, J., Stevens, K. R., and Albright, A. L. Prognostic factors and treatment results for supratentorial primitive neuroectodermal tumors in children using radiation
We thank Jackie Lee, Anthony Gerber, Mary Beth McCormick, and Rulla for contributions
16.
23. Cohen, B. H., Zeltzer, P. M., Boyett, J. M., Geyer, J. R., Allen, J. C., Finlay, J. L., McGuire-Cullen, P., Milstein, J. M., Rorke, L. B., Stanley, P., Stehbens, J. A., Shurin,
expression.
ACKNOWLEDGMENTS
Tamimi
15.
olfactory and autonomic neurons. Cell, 75: 463—476, 1993. Guillemot, F., and Joyner, A. L. Dynamic expression of the murine Achaete-Scute homologue Mash-I in the developing nervous system. Mech. Dcv., 42: 171—185, 1993. Bennahmias, S. An altemative to the water bath/plastic bag method for hybridization of Southem and Northem blots. Am. Biotechol. Lab., 10—12,1989. Chen, M. S., Bermingham-McDonogh, 0., Danehy, F. T., Nolan, C., Scherer, S. S., Lucas, J., Gwynne, D., and Marchionni, M. A. Expression of multiple neuregulin transcripts in postnatal rat brains. J. Comp. Neurol., 349: 389—400, 1994. Ball, D. W., Azzoli, C. G., Baylin, S. B., Chi, D., Dou, S., Donis-Keller, H.,
Kathleen
Patterson
for pathology
assistance; Wendy Leisenring for statistical assistance; and J. Russel Geyer for critical evaluation of the manuscript.
and chemotherapy: a children's cancer group randomized trial. J. Clin. Oncol., 13: 1687—1696, 1995. 24. Kozmik, Z., Sure, U., Ruedi, D., Busslinger, M., and Aguzzi, A. Deregulated expression of PAX5 in medulloblastoma. Proc. Nail. Acad. Sci. USA, 92: 5709— 5713, 1996. 25. Yokota, N., Aruga, J., Takai, S., Yamada, K., Hamazaki, M., Iwase, T., Sugimura, H., and Mikoshiba, K. Predominant expression of human zic in cerebellar granule cell lineage and medulloblastoma. Cancer Res., 56: 377—383,1996. 26. Aruga, J., Yokota, N., Hashimoto, M., Furuichi, T., Fukuda, M., and Mikoshiba, K.
A novel zinc finger protein, zic, is involved in neurogenesis, especially in the cell
REFERENCES 1. Heideman, R. L., Packer, R. J., Albright, L. A., Freeman, C. R., and Rorke, L. B.
Tumors of the central nervous system. In: P. A. Pizzo and D. G. Poplack (eds.), Principles and Practice of Pediatric Oncology. ed. 2, pp. 633—682.Philadelphia: J. B. Lippincott Co., 1993. 2. Trojanowski, J. Q., Tohyama, T., and Lee, V. M-Y. Medulloblastomas and related
primitiveneuroectodennalbraintumorsofchildhood recapitulatemolecularmilestonesin the maturationof neuroblasts.Mol. Chem. Neuropathol.,!7: 121—135, 1992. 3. Lee, J. E., Hollenberg, S. M., Snider, L., Tumer, D. L., Lipnick, N., and Weintraub,
H. NeuroD, a new basic helix-loop-helix protein, can convert Xenopus ectoderm into neurons. Science (Washington DC), 268: 836—844, 1995.
lineage of cerebellar granule cells. J. Neurochem., 63: 1880—1890, 1994. 27. Nourse, J., Mellentin, J. D., Galili, N., Wilkinson, J., Stanbridge, E., Smith, S. D., and
Cleary,M.,L.Chromosomaltranslocationt(l;19)resultsin synthesisof a homeobox fusion mRNA that codes for a potential chimeric transcription factor. Cell, 60: 535—545, 1990. 28. Chen, Q., Yang, C., Tsan, J. T., Xia, Y., Ragab, A. H., Peiper, S. C., Carroll, A., and Baer, R. Coding sequences of the tal-l gene are disrupted by chromosomal translo cation in human T-cell leukemia. J. Exp. Med., 172: 1403—1408,1990. 29. Aplan, P. D., Begley, C. G., Bertness, V., Nussmeier, M., Ezquerra, A., Coligan, J.,
and Kirsch, I. R. The SCL gene is formed from a transcriptionally complex locus. Mol. Cell. Biol., !0: 6426—6435,1990.
3530
Downloaded from cancerres.aacrjournals.org on July 16, 2011 Copyright © 1997 American Association for Cancer Research
NEUROD AND HASH) IN NEUROECTODERMAL
30. Tamimi, R., Steingrimsson,
E., Copeland, N. G., Dyer-Montgomery,
K., Lee, J. E.,
Hemandez,R.,Jenkins,N. A.,andTapscou,S. J. TheneuroDgenemapsto human chromosome 2q32 and mouse chromosome 2. Genomics, 34: 418—421,1996. 31. Tamimi, R. M., Steingrimsson, E., Montgomery-Dyer, K., Copeland, N. G., Jenkins, N. A., and Tapscott, S. J. NEUROD2 and NEUROD3 genes map to human chromo
somes llql2 and 5q23—q31 and mousechromosomes11 and 13, respectively. Genomics, 40: 355—357,1997. 32. Schutz, B. R., Scheurlen, W., Krauss, J., du Manoir, S., Joos, S., Betas, M., and
Lichter, P. Mapping of chromosomal gains and losses in primitive neuroectodermal tumors by comparative genomic hybridization. Genes Chromosomes & Cancer, !6: 196—203, 1996. 33. Gerber, A. N., and Tapscott, S. J. Tumor cell complementation groups based on myogenic potential: evidence for inactivation of loci required for basic helix-loop
helix protein activity. Mol. Cell. Biol., 16: 3901—3908,1996. 34. Biegel, J. A., Rorke, L. B., Jansa, A. J., Sutton, L. N., and Parmiter, A. H. Isochro mosome 17q demonstrated by interphase fluorescence in situ hybridization in prim
A., and Dambly-Chaudiere,
C. From DNA to form: the achaete-scute
proneural gene atonal is a positive transcriptional regulator expressed in the devel oping nervous system. J. Biol. Chem., 270: 8730—8738, 1995. 42. Shimizu, C., Akazawa, C., Nakanishi, S., and Kageyama, R. Math-2, a mammalian helix-loop-helix factor structurally related to the product of Drosophila proneural 239—248, 199S. 43. Naya, F. J., Stellrecht, C. M. M., and Tsai, M-J. Tissue-specific regulation of the insulin gene by a novel basic helix-loop-helix transcription factor. Genes Dcv., 9:
1009—1019,1995.
complex. Genes Dcv., 2: 495—501,1988. 36. Ghysen. A., Dambly-Chaudiere,
38. Johnson, J. E., Birren, S. J., and Anderson, D. J. Two rat homologues of Drosophila achaete-scute specifically expressed in neuronal precursors. Nature (Land.), 346: 858—861. 1990. 39. Sasai, Y., Kageyama, R., Tagawa, Y., Shigemoto, R.. and Nakanishi, S. Two mammalian helix-loop-helix factors structurally related to Drosophila hairs' and enhancer ofsplit. Genes Dcv., 6: 2620—2634, 1992. 40. Bartholoma, A., and Nave, K-A. Nex-l : a novel brain-specific helix-loop-helix protein with autoregulation and sustained expression in mature cortical neurons. Mech. Dcv.,48:217—228, 1994. 41. Akazawa, C., Ishibashi, M.. Shimizu, C., Nakanishi, S., and Kageyama, R. A mammalian helix-loop-helix factor structurally related to the product of Drosophila
gene atonal, is specifically expressed in the nervous system. Eur. J. Biochem., 229:
itive neuroectodermal tumors of the central nervous system. Genes Chromosomes & Cancer, 14: 85—96,1995. 35. Ghysen,
TUMORS
C., Jan, L. Y., and Jan, Y. N. Cell interactions and
44. Yasunami., M., Suzuki, K., Maruyama. H., Kawakami, H., Nagai. Y.. Hagiwara, M.,
gene interactions in peripheral neurogenesis. Genes Dev., 7: 723—733,1993.
and Ohkubo, H. Molecular cloning and characterization of a cDNA encoding a novel
37. Jan, Y. N., and Jan, L. Y. HLH proteins, fly neurogenesis, and vertebrate myogenesis. Cell, 75: 827—830, 1993.
basic helix-loop-helix protein structurally related to NeuroDIBHFl . Biochem. Bio phys. Res. Commun., 220: 754—758, 1996.
3531
Downloaded from cancerres.aacrjournals.org on July 16, 2011 Copyright © 1997 American Association for Cancer Research
