[CANCER RESEARCH 55, 2053-2055, May 15, 1995]
Advances in Brief
No CDKN2 Mutations in Neuroblastomas1
Christian P. Beltinger, Peter S. White, Erik P. Sulman, John M. Maris, and Garrett M. Brodeur2
Division of Oncology, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania 19104-9786
Abstract
Mutations of CDKN2 have been found recently in melanoma and many
other tumor types. Neuroblastoma shares with melanoma a neuroectoder-
mal origin and a high incidence of deletions of the short arm of chromo
some 1. Therefore, we analyzed 18 primary neuroblastomas and 9 tumor-
derived cell lines for mutations in CDKN2. We used PCR-single-strand
conformation polymorphism to examine exons 1 and 2 of the CDKN2 gene
for mutations, but none were detected. Furthermore, no homozygous
deletions were detected and there was no loss of heterozygosity at the
closely linked II \\ locus. We conclude that disruption of the CDKN2
gene is not required for malignant transformation of human neuroblas
tomas.
Introduction
The CDKN2 (MTSÌ) gene encodes piò, a protein which inhibits the
CDK43/cyclin D complex. Inhibition by pl6 decreases the phospho-
rylation state of the retinoblastoma gene product (pRb), and this cell
cycle checkpoint limits the G,-S transition and ultimately inhibits
cell growth (1-3).
The CDKN2 gene maps to 9p21, a site of frequent deletion in
melanoma and other tumor types. Interestingly, mutations and ho
mozygous deletions of CDKN2 were found initially in cell lines from
many different tumor types (4, 5). However, the overall incidence of
mutations or homozygous deletions is generally less in primary tu
mors compared to cell lines (6). In primary tumors, homozygous
deletions and intragenic mutations are frequently seen in pancreatic
adenocarcinoma, glioblastoma, anaplastic astrocytoma (but not low-
grade astrocytoma), esophageal squamous cell carcinoma, non-small
cell lung cancer, bladder carcinoma, acute lymphocytic leukemia, and
non-Hodgkin's lymphoma (4, 7-12). Intragenic alterations of CDKN2
have been detected rarely, if at all, in primary breast cancer (12),
primary renal cancer (11), and primary sporadic melanoma (13),
although cell lines derived from these tumors often contain homozy
gous deletions (5). Although cell lines of osteosarcoma, ovarian
tumor, and mesothelioma frequently show homozygous deletions of
CDKN2, no data concerning primary tumors have been published yet
for these cancers (4).
Homozygous deletions of CDKN2 occur in 60% of melanoma cell
lines (4, 5), and disease-specific germ line mutations are found in
many 9p21-linked melanoma families (14, 15). Melanomas frequently
contain deletions of the short arm of chromosome 1 (16), a region that
also shows genetic linkage in several familial melanomas (15). Inter
estingly, deletion of the short arm of chromosome 1 is the most
common genetic abnormality in human neuroblastomas (17, 18).
Furthermore, melanocytes and neuroblasts are both derived from the
Received 3/9/95; accepted 4/5/95.
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 work was supported by Grant CA 39771 (G. M. B.).
2 To whom requests for reprints should be addressed, at Division of Oncology, Room
9(KX). Children's Hospital of Philadelphia, 324 South 34th Street. Philadelphia, PA
19104-9786.
3 The abbreviations used are: CDK4, cyclin-dependent kinasc 4; LOH, loss of het
erozygosity; SSCP, single-strand conformation polymorphism.
neural crest. Because of these similarities, as well as the frequent
involvement of CDKN2 mutations in many other cancers, we exam
ined the CDKN2 gene in a series of neuroblastoma cell lines and
primary tumors. However, no homozygous deletions or mutations
were detected in the coding region, and no LOH was found at the
adjacent IFNA locus, suggesting that CDK.N2 is not involved in
the pathogenesis of neuroblastomas.
Materials and Methods
Tumors and Cell Lines. Nine neuroblastoma cell lines were analyzed, six
of which had amplification of the MYCN oncogene as well as Ip deletions or
rearrangements (NGP, NLF, NAB. SMS-KAN, SMS-KCN, and LA-N-5).
Three cell lines had neither MYCN amplification nor abnormalities of Ip
(NBL-S, SK-NSH, and LHN). Eighteen primary tumor samples were selected
to represent the distribution of stage and MYCN copy number typically found
in neuroblastoma. One of six selected stage I-II tumors and 4 of 10 selected
stage 1II-IV tumors had MYCN amplification. Two tumors were stage IV-S
and not amplified. For LOH studies, constitutional DNA corresponding to the
neuroblastoma samples was extracted and processed as described previously
(17). DNA samples from unrelated probands of 31 Centre d'étudedu poly
morphisme humain pedigrees were provided by H. Donis-Keller. Constitu
tional DNA samples of eight patients with germ line CDKN2 substitutions (one
substitution in exon 1 and seven substitutions in exon 2) were a gift from
N. Dracopoli (15).
PCR-SSCP and DNA Sequencing. PCR of CDKN2 was performed as
described previously (15). One primer pair covered exon 1. Three primer pairs
spanning overlapping regions designated as A, B, and C covered exon 2. SSCP
was performed as described (15), with the following modifications. Samples
were electrophoresed at 25°Con a 4.5% polyacrylamide gel in IX TBE (90
mm Tris-borate, 2 IHMEDTA) at 25 W for 4 h. Duplicates were run at 25°Con
a 0.5X MDE gel (AT Biochem, Malvern. PA) with 5% glycerol in 0.6X TBE
at 6 W for 12 h. Variant bands were cut from the gel and used directly as a
template for a second PCR reaction under the same conditions as described
above. The PCR product was then sequenced in both directions with the same
primers using methods described previously (19).
LOH Analysis. A microsatellite repeat within IFNA, which also maps to
9p21 and is adjacent to CDKN2, was PCR amplified from 17 paired normal
and tumor DNA samples using primers derived from the Genome Data Base.
One primer was end labeled with y-'2P using T4 polynucleotide kinase. PCR
was performed in 20-/il volumes containing 2 ¡u10X PCR buffer II (Perkin
Elmer/Cetus, Branchburg, NJ), 0.4 mM each primer, 0.1 mM each dNTP, 0.2
units AmpliTaq DNA polymerase (Perkin Elmer/Cetus), 1.5 mM MgCK, and
20 ng DNA. Reactions were amplified for one cycle at 95°C(3 min); for 16
cycles at 95°C(45 s) with the annealing/extension temperature starting at 70°C
and decreasing by 0.7°Ceach cycle (1 min); 25 cycles at 95°C(45 s), 55°C(30
s), and 72°C(1 min); and for one cycle at 72°C(5 min). LOH analysis was
performed as described (16).
Results
To first confirm the sensitivity of our PCR-SSCP method, we
examined exons 1 and 2 of CDKN2 in the constitutional DNA of eight
melanoma patients with known germ line CDKN2 substitutions. In
each case, a variant band was clearly detectable by PCR-SSCP. Fig.
1 shows the three germ line samples with substitutions in exon 2
(region C) used as controls (Lanes 6-9). While not all mutations were
seen on the polyacrylamide gels without glycerol, glycerol-supple-
mented MDE gels detected all substitutions.
2053
on March 7, 2021. © 1995 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from