[CANCER RESEARCH 50, 1459-1463, March 1, 1990]
myc Gene Amplification and Expression in Primary Human Neuroblastoma1
Irene Slave, Richard Ellenbogen, Woo-Hee Jung, Gordon F. Vawter, Cynthia Kretschmar, Holcombe Grier, and
Bruce R. Korf
Division of Genetics [I. S., B. R. K.J, and Departments of Neurosurgery [R. E.J, Pathology ¡W.-H.J., G. F. Y.J, and Neurology [B. R. K.J. The Children's Hospital, and
Department of Pediatrie Oncology, Dana-Farber Cancer Institute/C. K., H. G.J, Boston, Massachusetts 02115
ABSTRACT
Although N-myc amplification in neuroblastomas correlates with poor
prognosis, not all neuroblastomas which fail to respond to therapy have
N-myc amplification. To determine whether other modes of myc gene
activation underlie progression of some neuroblastomas, 45 were analyzed
for amplification of N-wiyc, c-myc and L-myc and 26 were studied for
transcription of these oncogenes. N-myc amplification was found in 6 of
45 tumors; no tumor had amplification of c-myc or L-myc. Transcription
of both N-myr and c-myc occurred in 21 of 26 neuroblastomas. No tumor
without \-myc amplification had a level of \-myc expression near that
of a tumor or cell line with amplification. One tumor with \-myc
amplification was the only specimen with N-myc but not c-myc expres
sion. Five samples had c-myc but not \-myc expression; all had histolog-
ical features of ganglioneuroma. DNA index did not correlate with myc
gene amplification or expression. It is concluded that N-ni.vc and c-myc
are commonly expressed in primary untreated neuroblastomas, but in the
absence of N-myc amplification, expression of these genes does not
appear to correlate with disease progression.
INTRODUCTION
Amplification of the protooncogene N-myc occurs in approx
imately one-third of neuroblastomas and correlates with poor
prognosis (1, 2). Yet the absence of N-myc amplification does
not necessarily predict a favorable outcome since a high pro
portion of children over 1 year of age with stage III or IV
neuroblastomas eventually die from their disease (3, 4). Al
though nonaneuploid DNA content (5-7) and deletion of the
short arm of chromosome 1 (8) have been found to correlate
with advanced disease, no molecular genetic markers have been
identified in the advanced neuroblastomas lacking N-myc am
plification which might explain their poor outcome.
The association of N-myc amplification with poor prognosis
suggests two hypotheses to explain the progression of tumors
without amplification. First, N-myc is one member of a "family"
of nuclear oncogenes (9), which also includes c-myc, L-myc, B-
myc etc. In small cell lung carcinomas, amplification of N-myc,
c-myc, or L-myc occurs in various tumors (10, 11), and c-myc
amplification has been found in other neoplasms (12-15). It is
therefore possible that some neuroblastomas might have am
plification of one of these other myc family oncogenes. A second
possibility is that some tumors might achieve high levels of myc
gene expression by means other than amplification. This phe
nomenon has been observed in some Wilms' tumors, for ex
ample, which do not have N-myc amplification but may have
levels of N-wiyc expression approximating those seen in neu
roblastomas with amplification (16).
In this paper we present results of the study of 45 primary
untreated neuroblastomas for amplification of N-myc, c-myc,
and L-myc. In addition, 26 tumors were analyzed for expression
of these myc family oncogenes. The data are compared with
Received 6/28/89; revised 11/9/89; accepted 11/14/89.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1This investigation received support from a Clinical Investigator Development
Award NS01097 from the National Institute of Neurological Disorders and
Stroke (B. R. K.) and the Max Kade Foundation (I. S.).
clinical outcome of the patients, histological features of the
tumors, and DNA histograms obtained by flow cytometry.
MATERIALS AND METHODS
Tumor Specimens. Tumor samples were obtained either in the oper
ating room or the pathology laboratory, immediately frozen in liquid
nitrogen, and stored at —¿ 70°C. The neuroblastoma cell lines LAN-1
and IMR-32 were used as controls for N-myc amplification.
Pathological Examination. Portions of tumors were fixed and stained
by conventional histopathological means. Samples studied for oncogene
expression were examined by a pathologist for areas of differentiation
and classified according to the criteria of Shimada ( 17). The histological
analysis was performed without knowledge of the results of RNA
studies.
Flow Cytometry. Nuclei were isolated and harvested from 50-/im-
thick sections of formalin-fixed paraffin blocks by deparaffinization,
rehydration, digestion in pepsin, disaggregation by vortex, digestion in
RNase, and filtration through SO-^m nylon mesh (18). Papanicolaou-
stained smears were used to assess proportion of tumor nuclei and
adequacy of disaggregation. The nuclei were stained with propidium
iodide and processed through a Becton-Dickinson fluorescence-acti
vated cell sorter flow cytometer. Data on 5000 nuclei were analyzed
with an EPICS multiparameter data acquisition and display system
single parameter computer. The proliferative compartment of the cells
of the tumor samples was estimated as the percentage of cells in S
phase plus G2 plus M phase by cytographic pattern.
DNA and RNA Analysis. DNA and RNA were prepared as described
previously (19). DNA samples were digested with £coRIand quantified
by DAPI (4'-6-diamidino-2-phenylindole) fluorescence. Three-Mg sam
ples were loaded onto a 0.7% agarose gel, subjected to electrophoresis,
and blotted onto Hybond-N (Amersham). Total cellular RNA was
quantified by spectrophotometry and 25-^g samples were electropho-
resed in formaldehyde-agarose gels and then blotted onto Biotrans
(ICN) membranes.
Hybridization was carried out sequentially using N-myc (20), c-myc
(21), and L-myc (10) DNA probes labeled with 32P by the random
primer method (22). RNA blots were also hybridized with a chick a-
tubulin sequence (23) as a control for amount of RNA loaded into the
gel. Hybridizations were carried out in 50% formamide at 42°Cand
washed at high stringency in 0.1 x standard (1 x = 0.15 M NaCl-0.015
M sodium citrate) saline-citrate at 65°C.Autoradiography was done
using X-Omat AR film and an intensifying screen.
Hybridization intensity was measured using an Ultrascan XL laser
densitometer. N-myc copy number was determined by comparison of
the hybridization intensity of the N-myc band between the tumor, a
normal control sample, and DNA from the neuroblastoma cell line
IMR-32, which has 25-fold N-myc amplification (24). A single copy
sequence (D13S28) hybridized to the same blots was used to correct
for amount of DNA loaded into each lane. For RNA analysis, in order
to compare intensity of hybridization between samples on the same
blot, and between samples on different blots, densitometric readings
were normalized to the intensity of the a-tubulin hybridization for each
sample.
Statistical Analysis. Levels of N-myc and c-myc expression were
tested for correlation with age at diagnosis, stage, site of tumor,
Shimada classification, degree of differentiation, and DNA index using
the Mann-Whitney nonparametric method (25). Correlation of levels
of expression with proliferative compartment was tested using the
Spearman rank correlation coefficient (25).
1459
Research.
on December 29, 2021. © 1990 American Association for Cancer cancerres.aacrjournals.org Downloaded from