| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
Department of Pathology, Division of Molecular Histopathology, University of Cambridge, Addenbrooke’s Hospital, Cambridge CB2 2QQ, United Kingdom [K. I., E. E. S., V. P. C], and Ludwig Institute for Cancer Research and Unit of Tumorpathology, Department of Oncology and Pathology, Karolinska Hospital, 171 76 Stockholm, Sweden [K. I., M. B. B., H. M. G., E. E. S., A. M., V. P. C.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There are several lines of evidence to indicate that deregulation of G1-S transition control alone may be detrimental for cell survival: (a) transfection of rodent cells with the adenoviral E1A and E1B genes resulted in transformation due to the viral proteins binding to and inactivating pRB and p53. Transfection with E1A alone induced p53 dependent apoptosis (10) ; (b) when E2F1 was ectopically expressed in rodent embryo fibroblasts, although the cell indeed entered S phase (11) , apoptosis was induced in a p53-dependent manner (12) ; (c) loss of pRB function leads to inappropriate progression through S phase and induces apoptosis in developing mouse lens fiber cells, which can be overcome by simultaneous loss of p53 (13) or E2F1 (14) . This evidence suggests that, at least in some cell types, p53 prevents cells with deregulated G1-S control from abnormal proliferation by inducing apoptosis.
p53 is a key regulator of cell cycle checkpoints. p53 binds to DNA in a sequence-specific manner and functions as a transcription factor. It induces either G1 arrest or apoptosis in response to various forms of cell stresses (reviewed in Ref. 15 ). Expression of the p53 protein is mainly regulated posttranscriptionally and maintained at a very low level in normal cells. MDM2 is an important regulator of p53. MDM2 binds to p53 and inhibits its function by concealing the activation domain of p53 (16 , 17) and by promoting degradation of p53, most likely through the ubiquitin-proteasome pathway (18 , 19) . In response to DNA damage, the MDM2 binding site of p53 is phosphorylated, the p53-MDM2 interaction is attenuated and p53 accumulates rapidly, relieved from MDM2-mediated suppression (reviewed in Ref. 20 ). MDM2 is also one of the transcriptional targets of p53 (21) .
Recent investigations have identified another important regulator in
this pathway, p14ARF (human homologue of mouse
p19ARF). The p14ARF protein
is encoded by the CDKN2A/INK4A locus but is distinct from
the p16INK4A protein (22)
.
p14ARF is encoded by the unique exon 1ß (E1ß)
and exon 2 and 3 of p16INK4A, using an
alternative reading frame. E1ß is located between exon 1
of
CDKN2A (E1) and exon 2 of CDKN2B on 9p21
(23
, 24)
. It has been shown that the
p14ARF protein binds to the p53/MDM2 complex and
inhibits MDM2-mediated degradation of p53, which indicates that
p14ARF is an upstream regulator of p53 via MDM2
(25, 26, 27, 28)
. There is also evidence suggesting that p53
down-regulates expression of p14ARF, which would
establish an autoregulatory feedback loop between p53, MDM2, and
p14ARF (29)
. The most striking
finding is that E2F1 transcriptionally up-regulates expression of
p14ARF (29
, 30)
. It has been shown
that in p19ARF null mouse embryonic fibroblasts,
accumulation of p53 and induction of apoptosis after introduction of
E1A was attenuated (31)
. Thus,
p14ARF seems to be a critical component in the
scrutiny of proliferation signals by the p53 system (32)
.
These findings prompted us to determine the status of genes involved in G1-S transition control and the p14ARF/MDM2/p53 pathway in a large series of astrocytic gliomas. We found that almost all astrocytic gliomas with altered G1-S transition control genes also had abnormalities of the p14ARF/MDM2/p53 pathway genes. Our findings indicated that disruption of the p53 pathway is virtually obligatory for these tumors with G1-S transition control gene abnormality.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A total of 190 astrocytic gliomas were included in the present investigation. All of the tumors had been used in previous studies (1, 2, 3 , 5 , 33 , 34) . The histopathological diagnosis was carried out strictly according to the WHO classification (8) . All of the cases collected before publication of the second edition of WHO classification (8) were reviewed in detail and reclassified. A new tumor number with the appropriate prefix representing the diagnosis (GB, AA, or A) was assigned to each reclassified case. All of the diagnoses are based on the histopathology of the case, not the tumor piece, although the pieces chosen are representative for each case.
Allelic Assessment.
Allelic assessment of RB1, CDK4,
CDKN2A, CDKN2B, and MDM2 was done by
densitometry on Southern hybridization using a PhosphorImager and
ImageQuant software (Molecular Dynamics, Sunnyvale, CA). DNA/RNA
extraction, probes, Southern blotting, hybridization and PhosphorImager
analysis have been described previously (1, 2, 3
, 5
, 35)
. A
probe for E1ß of p14ARF was generated by
PCR using primer pair PC978 (CGAGTGAGGGTTTTCGTGGTTC, forward) and PC977
(CGTTGTAACCCGAATGG-GAAGC, reverse), which amplify a genomic segment
containing a 3' portion (138 bp) of E1ß and part of intron 1ß (145
bp). The primer sequences were chosen based on the published genomic
sequence around E1ß (23)
. STS markers
WI-3306 or WI-7427 were used as control probes
for allelic assessment of p14ARF.
WI-3306 is located on 2q close to D2S121,
D2S112 and D2S44 and WI-7427 is
located on 21q between D21S1257 and D21S270.
[Chromosome 2 workshop Consensus Map, GDB: 4225469; An STS-Based Map
of the Human Genome, Data Files Release 12 (36)
]. For
each tumor, the allelic status on 2q and 21q was determined by RFLP
analysis with D2S44 or microsatellite analysis with
D2S121, D2S112, D21S1257, and
D21S270 (data not shown). A control locus from a chromosomal
region that did not have allelic imbalance was chosen for each case. To
assess allelic numbers at E1ß of p14ARF,
densitometry on TaqI digested Southern blotting was carried
out. The signal intensity of the bands corresponding to
p14ARF and the control locus in tumor and
control DNA was measured using the ImageQuant, version 3.3, software
(Molecular Dynamics, Sunnyvale, CA). The number of alleles at
p14ARF was determined by the following
formula:
 |
Mutation Analysis of CDKN2B and
p14ARF.
Mutation of exon 1 and exon 2 of CDKN2B was examined by
direct sequencing of PCR amplified products as described
(3)
. Mutation of E1ß of
p14ARF was analyzed by direct sequencing of
the PCR products that covered the entire coding region (PC1340:
TGCAGTTAAGGGGGCAGGAG, forward; and PC1343: TTATCTCCTCCTCCTCCTAG-CCTG,
reverse; putative start codon was according to Refs. 25
and 30
). The PCR products were directly sequenced
using the same set of primers using the ABI PRISM BigDye Terminator
Cycle Sequencing Kit (PE Biosystems, Foster City, CA) on an ABI PRISM
373aXL and a 377 DNA Sequencer (PE Biosystems). Mutation in exon 2 of
p14ARF was assessed by reevaluating the
previously analyzed sequences of exon 2 of CDKN2A in the
same tumor series (3
, 5)
. Any nucleotide changes in exon 2
of CDKN2A that can cause amino acid changes in the
p14ARF protein were documented.
DGGE Analysis of the TP53 Gene.
Mutation of the TP53 gene (the gene encoding the p53
protein) was analyzed using DGGE. The entire coding region of
TP53 as well as intron sequences adjacent to exons (minimum
of six bases) were covered by the analysis with the primer pairs listed
in Table 1
. A stretch of a GC-rich sequence (GC-clamp) was attached to one
of the primers for each pair as indicated (Table 1)
. Primers for exons
5–8 are modified from Hamelin et al. (37)
. The
details of DGGE conditions and the strategy of DGGE analysis will be
presented
elsewhere.5
Paired DNA from tumors and patients’ blood were aligned in 96-well
microtiter plates at a concentration of 10 ng/µl. Thirty to 50 ng of
DNA were used as PCR templates. A heteroduplex of tumor and blood PCR
products was prepared for each sample to increase the sensitivity of
analysis. The heteroduplex was made by mixing an equal volume of PCR
products from tumor and blood, denaturing at 94°C for 15 min,
incubating at 65°C for 1 h and then at room temperature for
3 h. Thirty-four µl of the heteroduplex was mixed with 10 µl
of loading buffer (0.25% bromo-phenol-blue, 0.25% xylene cyanol, and
30% glycerol). The D GENE system (Bio-Rad, Hercules, CA)
was used to perform DGGE. An appropriate amount of urea and formamide
was mixed with 6.5% acrylamide and cast according to the
manufacturer’s recommendation. Gels were run at 60°C at 160V for an
optimal length of time for each primer pair. The gels were then stained
with SYBR green I (FMC, Rockland, ME) and digitally documented by an
EagleEye II system (Stratagene, La Jolla, CA). When an abnormal shift
was detected, the corresponding exon was amplified from both tumor
DNAand the patient’s blood DNA and directly sequenced to
confirm the presence of a somatic mutation. DNA sequencing was
performed using the ABI PRISM Ready Reaction DyeDeoxy Terminator Cycle
Sequencing Kit (PE Applied Biosystems) on an ABI PRISM 373 DNA
Sequencer (PE Applied Biosystems) as described previously
(3)
.
|
2 test was used to statistically evaluate an
association between various genetic abnormalities and to compare the
incidences of abnormalities among different malignancy grades. When the
expected frequency was less than five in any category, a Fisher’s
exact test was applied instead. |
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The cumulated data showed the following results (see Table 2
). Three GBs had homozygous deletion of RB1 and an additional
15 had loss of one allele and mutation of the other allele of
RB1. In addition, one GB (GB185) had an aberrant transcript
lacking exons 14–17 of RB1. This case had loss of one
allele at RB1 without a detectable mutation in the other
allele.6
In all, 19 GBs (14%) had total loss of wild-type RB1. No
homozygous deletion or mutation of RB1 was found in the AAs
or As. Eighteen GBs (13%) and three AAs (8%) had CDK4
amplification. Fifty GBs and four AAs had homozygous deletion of
CDKN2A, and an additional four GBs and one AA had loss of
one allele and mutation of the other allele of CDKN2A. In
all, 54 GBs (40%) and 5 AAs (13%) had total loss of wild type of
CDKN2A. Fifty GBs (37%) and three AAs (8%) had homozygous
deletion of CDKN2B. No mutation of CDKN2B was
identified.
|
. Eighteen of 19 GBs with RB1 mutation/homozygous deletion
had neither CDK4 amplification nor CDKN2A/CDKN2B
homozygous deletion. One GB (GB47) had mutation of both RB1
and CDKN2A. Eighteen GBs and three AAs with CDK4
amplification retained at least one wild-type allele of RB1,
CDKN2A, and CDKN2B, whereas one GB (GB28) had
both amplification of CDK4 and homozygous deletion of
CDKN2A (Table 3)
. Forty-eight GBs and three AAs had
homozygous deletion or mutation of both CDKN2A and
CDKN2B but no RB1 mutation or CDK4
amplification. Six GBs and two AAs had homozygous deletion of only
CDKN2A but not CDKN2B. Two GBs (GB199 and GB147)
had homozygous deletion of CDKN2B but not CDKN2A.
When all of these abnormalities were combined, 91 GBs (66%), 8 AAs
(21%), and no As had either total loss of wild type of RB1,
CDKN2A, or CDKN2B, or amplification of
CDK4 (Tables 2
and 3)
.
|
. In all, 100 mutations were identified in 85 tumors (Table 2)
. There were 72 missense mutations, 3 inframe deletions, 10
frame-shift deletions, 3 insertions, and 9 nonsense mutations. Three
point mutations were found in intronic consensus splice junction
sequences. Thirteen tumors had two mutations (9 GBs, 3 AAs, and 1 A)
and one GB (GB27) had three mutations. The details of the
TP53 mutations as well as the allelic status of the
TP53 locus and expression of the p53 protein as assessed by
immunohistochemistry will be presented elsewhere.5
In summary, 49 GBs (36%), 26 AAs (67%), and 10 As (67%) had
TP53 mutations (Table 2)
.
|
; Fig. 1B
). No
amplification was found among AAs or As.
The allelic status of E1ß of p14ARF was
assessed in the following manner. E1ß of
p14ARF is located approximately 20 kb
centromeric to E1 of CDKN2A/p16INK4A,
and telomeric and in the close vicinity to exon 2 of CDKN2B
(23)
. When homozygous deletion of both CDKN2A
and CDKN2B was identified E1ß of
p14ARF was considered as being
homozygously codeleted. This cohomozygous deletion was demonstrated in
an analysis of 13 glioma cell lines in which
p14ARF E1ß, CDKN2A and
CDKN2B were specifically examined (data not shown). On the
basis of this principle, 48 GBs and 3 AAs were considered as
having homozygous deletion of E1ß of
p14ARF (Table 3)
.
The allelic status of E1ß of p14ARF was
specifically examined in 15 cases using a genomic fragment containing
part of E1ß and the adjacent intron as a probe for Southern
hybridization. These include the cases with homozygous
deletion/mutation of only CDKN2A (GB14, GB184, GB36, GB236,
GB177, AA87, and AA86) or only CDKN2B (GB199 and GB147) and
the cases with abnormalities of RB1 or CDK4
without TP53 mutation or MDM2 amplification
[GB47, GB100, GB25, GB157, GB11, and AA7 (Table 3)
]. As a result, 5
tumors (GB14, GB184, GB199, GB147, and AA87) were determined to have
homozygous deletion of E1ß of p14ARF
(Fig. 1C
). For the remaining 10 cases, the coding region of
E1ß was sequenced to determine mutations. No nucleotide changes in
E1ß were identified. For mutation in exon 2 of
p14ARF,
CDKN2A/p16INK4A sequences were reevaluated
in the context of the alternative reading frame for
p14ARF [see above and (3
, 5)
].
Among five CDKN2A/p16INK4A mutations
identified, four were located in the common exon 2. These four
CDKN2A/p16INK4A mutations also altered the
amino acid sequence of p14ARF (Table 3)
.
Profiles of p14ARF/MDM2/p53 pathway gene
abnormalities in individual tumors were then compared. Eleven tumors (9
GBs and 2 AAs) had both TP53 mutation and homozygous
deletion of p14ARF. Three tumors (two GBs
and one AA) had mutation in TP53 and exon 2 of
p14ARF. No tumor with MDM2
amplification had either TP53 mutation or
p14ARF deletion/mutation. When
abnormalities of all of the three genes were combined, 103 GBs (76%),
28 AAs (72%), and 10 As (67%) had either mutation of TP53,
amplification of MDM2, or homozygous deletion or mutation of
p14ARF (Table 2)
.
Correlation of G1-S Transition Control Gene and p53
Pathway Gene Abnormalities.
On the basis of the above findings, profiles of
G1-S control gene abnormalities and p53 pathway
gene abnormalities in each individual tumor were compared. Among the 19
GBs with RB1 homozygous deletion or mutation, 14 tumors had
TP53 mutations and 1 tumor (GB140) had MDM2
amplification (Table 3
; Fig. 1B
). One GB (GB47) with both
RB1 and CDKN2A mutation had a mutation of exon 2
of p14ARF (Table 3)
. Three GBs (GB100,
GB25, and GB157) had no TP53 mutation, MDM2
amplification, nor p14ARF homozygous
deletion/mutation.
Among the 21 cases with CDK4 amplification (18 GBs, 3 AAs),
7 GBs and 2 AAs had TP53 mutation (Table 3
; see GB154 in
Fig. 1, A and B
). Among them, one GB with both
CDK4 amplification and homozygous deletion of
CDKN2A/CDKN2B (GB28) had both TP53 mutation and
homozygous deletion of p14ARF (Table 3)
.
An additional 10 GBs had amplification of MDM2 (see GB90 in
Fig. 1B
). Only two tumors (GB11 and AA7) had no
TP53 mutation, MDM2 amplification, nor
p14ARF homozygous deletion/mutation (Table 3)
.
Among the 48 GBs and 3 AAs in which E1ß of
p14ARF was considered as being
homozygously codeleted with
CDKN2A/p16INK4A and
CDKN2B/p15INK4B, 9 GBs and 1 AA had
TP53 mutation (Table 3)
. Among two GBs and one AA with
homozygous deletion of only CDKN2A but not
CDKN2B, two (GB14 and AA87) had both homozygous deletion of
E1ß of p14ARF and TP53
mutation (see GB14 in Fig. 1C
), and one (GB184) had
homozygous deletion of E1ß of p14ARF but
not TP53 mutation. Among the five tumors with
CDKN2A/p16INK4A mutation (GB47, GB36,
GB236, GB177, and AA86), GB36, GB236, and AA86 had mutation of both
TP53 and exon 2 of p14ARF (see
GB36 in Fig. 1, A and C
, and Ref.
3
), GB177 had only TP53 mutation, and GB47 had
only exon 2 mutation of p14ARF (described
above). Both of the cases with homozygous deletion of only
CDKN2B but not CDKN2A (GB199 and GB147) had
homozygous deletion of E1ß of p14ARF
(see GB147 in Fig. 1C
).
When all of the genetic abnormalities described above were combined and
compared, 87 GBs and 7 AAs had abnormalities of both
G1-S transition control genes and p53 pathway
genes (Tables 2
and 3)
. Among the tumors with
G1-S transition control gene abnormalities, this
corresponds to 96% of GBs (87 of 91) and 88% of AAs (7 of 8). Among
tumors with p53 pathway gene abnormalities, 84% of GBs (87 of 103) and
25% of AAs (7 of 28) had deregulated G1-S
transition control (Table 3)
. The association between abnormalities of
G1-S control genes and p53 pathway genes was
highly significant among GBs and among all of the tumors considered
together (2 test, P < 0.0001).
Four GBs and one AA with either RB1 mutation (GB100, GB25, and GB157) or CDK4 amplification (GB11 and AA7) had no TP53 mutation, MDM2 amplification, nor p14ARF homozygous deletion/mutation. All of the tumors with homozygous deletion of CDKN2A and/or CDKN2B or CDKN2A mutation had either TP53 mutation or p14ARF homozygous deletion, or both. Sixteen GBs, 21 AAs, and 10 A grade II had TP53 mutation but none of the G1-S control gene abnormalities. MDM2 amplification was always associated with either RB1 mutation or CDK4 amplification. Twenty-nine GBs, 10 AAs, and 5 As had neither homozygous deletion/mutation of RB1, CDKN2A, CDKN2B, or TP53, nor amplification of CDK4 or MDM2.
In two patients, tumors from the first and second surgeries were
studied. In one of them, the initial tumor (AA100) had two mutations in
exon 8 of TP53 (R273C and T304ins). When the tumor recurred
and progressed (GB177), it acquired a mutation in E1 of
CDKN2A (Y44X) while retaining the same TP53
mutations as AA100.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our findings demonstrate that in primary human gliomas with altered G1-S transition control, the p53 pathway is almost always inactivated. Normally, p53 is involved in preventing cells from uncontrolled proliferation and tumor formation by inducing either G1 arrest or apoptosis when the cell enters the cell cycle in a nonphysiological manner. Cell biological experiments have suggested that abnormalities of G1-S transition control genes may give a growth advantage to the cell; however, simultaneous inactivation of the p53 pathway is a prerequisite for cell survival when such abnormalities are present (10 , 12 , 13) . It has been shown that introduction of wild-type p53 into the glioma cell lines U251 and A172, which had mutation of TP53 and homozygous deletion of CDKN2A, led to apoptosis (42, 43, 44) . Our results substantiate the hypothesis proposed as a result of such experiments using cultured cells.
The two systems can be targeted by a single genetic event. CDKN2A, CDKN2B, and p14ARF reside in a small region on 9p21, and they are frequently homozygously codeleted in diverse types of human tumors including astrocytic gliomas (2 , 45) . CDK4 and MDM2 are located in the same chromosomal region, 12q13–15, albeit with a 4-Mb gap between them (46) . Amplification of this region may be initiated as a single event encompassing both MDM2 and CDK4. Later the amplicon may be rearranged to include only the genes necessary to target, thus, explaining the finding that all of the loci between MDM2 and CDK4 are not amplified in all of the tumors when studied in detail (47) . Rearrangements of amplicons in tumor cells have been well documented. One example is the rearrangement of the amplified EGFR gene in some GBs producing an aberrant, constitutively activated, receptor (48) .
However, in a large proportion of the cases, mutations of TP53 (located on 17p13) and abnormalities of the G1-S transition control genes examined here (RB1, on 13q14; CDK4, on 12q13–15; CDKN2A and CDKN2B, on 9p21) must have occurred as independent and nonsynchronized events, because no single genetic event that would involve these genes simultaneously can be conceived. The association of RB1 mutations/deletions and TP53 mutations alone is statistically significant (P = 0.00081). Therefore, the association between the deregulated p53 pathway and G1-S transition control identified in this study is unlikely to be caused merely by the genetic proximity of some of these genes but is rather a consequence of selection for specific combinations of abnormalities.
The coinactivation of the two systems was found exclusively in GBs and
some AAs, which indicates that the combination of these abnormalities
contributes to a more aggressive biological tumor phenotype.
Inactivation of the p53 pathway alone, exclusively by mutation of
TP53, was observed in all of the malignancy grades. In fact,
the frequency of TP53 mutation is higher in As and AAs than
in the GBs (2 test, P = 0.0224 and 0.0007, respectively; see Table 2
). It has been shown
that As with mutation of TP53 are more likely to recur and
progress as compared with those without TP53 mutation
(49)
. On the basis of these findings, an attractive model
of astrocytic glioma progression considers mutation of TP53
an early step and the acquisition of G1-S
transition control gene abnormalities a later step. A good example is
given by one AA (AA100) in our series with TP53 mutation
that progressed to GB (GB177) 3 years after the first operation. It had
then acquired a mutation in E1 of CDKN2A in addition to
the TP53 mutation documented in the AA.
It is clinically well documented that GB may arise as a de novo tumor or may progress from astrocytic gliomas of lower malignancy. Judging from clinical data the majority of GBs are de novo tumors. The arrangement of the genes in the genome provides the basis for single events that can inactivate the two cellular control systems simultaneously, providing a credible explanation for the prevalence of de novo GBs.
Abnormalities of genes involved in G1-S
transition and the p53 pathway are frequently found in different human
tumors. In our series, an individual tumor usually has an alteration of
only one component from each regulatory system, which supports the idea
that abnormalities of each component may have approximately equivalent
effects. For example, RB1 mutation and CDK4
amplification are completely mutually exclusive in our tumors. Not a
single case had both a TP53 mutation and MDM2
amplification (see Table 3
). However, in some tumors there were
apparently redundant genetic abnormalities. One explanation could be
that certain genetic events involving more than one gene may also
include a neighboring gene, i.e., homozygous deletion or
amplification. In cases that have both TP53 mutation and
p14ARF homozygous deletion,
p14ARF may simply have been codeleted with
CDKN2A or CDKN2B without giving an additional
selective advantage. This is supported by our finding that such
redundant abnormalities were found when CDKN2A,
CDKN2B, or p14ARF were
involved. Another possibility is that abnormalities of the different
components may not have an equivalent effect. Even different mutations
of the same gene may have varying impact on the function of its protein
product. Additionally, it cannot be excluded that multiple
abnormalities in the same complex pathway could provide an additional
growth advantage.
p14ARF (p19ARF in mouse) has an unusual feature. Using an alternative reading frame, it is translated partially from common exons with CDKN2A/p16INK4A, yet results in a totally different protein (22, 23, 24) . p14ARF has been found frequently deleted in a variety of human tumors, most often codeleted with CDKN2A but sometimes specifically targeted (50) . We demonstrated that in cases that have homozygous deletion of only CDKN2A or CDKN2B, E1ß of p14ARF was always involved in the deletion, which indicated that E1ß of p14ARF was the most frequently deleted region on 9p21. p14ARF knockout mice that specifically lack E1ß developed normally but are predisposed to various types of tumors including glioma (51 , 52) . These findings support p14ARF as a tumor suppressor gene. The growth suppressive ability of p14ARF in p14ARF null human cells (or p19ARF in p19ARF null mouse cells) further supports this (28 , 53 , 54) .
Thus far no germ-line or somatic mutation in E1ß has been identified in human tumors (see Ref. 52 and references therein). Mutations in exon 2 of CDKN2A/p16INK4A often alter the amino acid sequence of p14ARF simultaneously, and four such mutations were identified in this study. It has been reported that the NH2-terminal domain of p14ARF, exclusively coded by E1ß, is necessary and sufficient for binding to MDM2 and inducing G1 arrest (27 , 28 , 54) . The growth suppressive activity of p14ARF was not abrogated by a number of common mutations in exon 2 that alter the amino acid sequences of both p16INK4A and p14ARF (53 , 54) . These findings question the significance of exon 2 mutations in p14ARF. Three of our four cases with mutations in exon 2 of p14ARF also had a TP53 mutation.
The role of CDKN2A/p16INK4A as a tumor suppressor gene has been well established by both cell biological experiments and genetic analysis of primary human tumors in sporadic cases and in familial tumor syndromes (55) . Nevertheless, the mutation frequency of CDKN2A is relatively low compared with the frequency of homozygous deletion. It could simply reflect that simultaneous inactivation of both CDKN2A and p14ARF by homozygous deletion is a more efficient way to simultaneously abrogate the p53 pathway and the G1-S transition control.
It is intriguing that five cases had G1-S transition control gene abnormalities without TP53 mutation, MDM2 amplification or p14ARF homozygous deletion/mutation. The most likely explanation is that another component of the p53 pathway, either a regulating protein, for example, p300 (56) , or downstream effector in the apoptosis pathway, for example, BAX (57) , may be altered in these cases. It is also possible that some TP53 mutations have escaped detection, although we think that this is unlikely because primers and conditions for DGGE were optimized to cover the entire coding region and splice junction sequences. However, mutations in regulatory sequences located outside of the screened region would naturally be missed.
Approximately one-third of GBs as well as the majority of AAs and all of the As in our series lack clear evidence of a deregulated G1-S control system. Many tumors in this category have loss of one allele at RB1 or CDKN2A, without detectable mutation in the other alleles (3 , 5) . Methylation of the promoter region of CDKN2A has been suggested to be an alternative mechanism to inactivate the gene (58) . However, our previous analysis of a limited number of tumors showed only a very low incidence of methylation, and this did not correlate to expression (3) . A preliminary mutation analysis of the promoter region of CDKN2A failed to identify somatic mutations.7 Mutation of codon 24 of CDK4 has been reported in familial melanoma kindreds and sporadic melanomas, and the mutated protein has been demonstrated to function as kinase, yet it is unable to bind to p16INK4A (59 , 60) . Mutation analysis of CDK4 exon 2, which contains codon 24, failed to identify any mutation in this tumor series.7 Other components in the same pathway, for example, CDK6 or members of Cyclin Ds, could possibly be involved (42 , 61) and remain to be studied.
The molecular mechanisms of growth control in cells is undoubtedly more complex than our current understanding. The identification of the molecular basis for the tumors without genetic abnormalities of the genes studied here will give further insight into tumorigenesis of astrocytic brain tumors.
ACKNOWLEDGMENTS
We thank Lotta Asplund and Susanne Öhlin for excellent
technical assistance and Alice Johnson-Marshall, Lu Liu, and Shohreh
Varmeh-Ziaie for critical reading of the manuscript. We also thank the
Departments of Neurosurgery at Sahlgrenska Hospital, Göteborg,
and the Karolinska Hospital, Stockholm, Sweden for their cooperation in
the collection of clinical material.
|
FOOTNOTES |
|---|
1 Supported by grants from the Swedish Cancer
Society, Stockholm’s Cancer Society, King Gustaf V. Jubilee Fund, Axel
and Margaret Ax:son Johnsons Funds, Lars Hierta Foundation, the Funds
of the Karolinska Institute, and CAMPOD. 
2 Present address: Department of Cell and
Molecular Biology, Medical Nobel Institute, Karolinska Institute, 171
77 Stockholm, Sweden.
3 To whom requests for reprints should be
addressed, at Department of Pathology, Division of Molecular
Histopathology, University of Cambridge, Addenbrooke’s Hospital, Box
235, Cambridge CB2 2QQ, United Kingdom. Phone: 44-1223-336072; Fax:
44-1223-216980; E-mail: vpc20{at}cam.ac.uk
4 The abbreviations used are: GB, glioblastoma; A,
astrocytoma; AA, anaplastic A; DGGE, denaturing gradient gel
electrophoresis; STS, sequence-tagged site.
5 K. Ichimura, manuscript in preparation.
6 H. M. Goike. Clonal selection for genetic
abnormalities for the G1/S transition control and p53
pathways in human glioblastoma xenografts, manuscript in
preparation.
7 K. Ichimura and V. P. Collins, unpublished
data.
Received 7/14/99. Accepted 11/12/99.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
/
gene cluster and the D9S171 locus. Cancer Res., 54: 3127-3130, 1994.This article has been cited by other articles:
|
|
 |
|
  Y. Ruano, T. Ribalta, A. R. de Lope, Y. Campos-Martin, C. Fiano, E. Perez-Magan, J.-L. Hernandez-Moneo, M. Mollejo, and B. Melendez Worse Outcome in Primary Glioblastoma Multiforme With Concurrent Epidermal Growth Factor Receptor and p53 Alteration Am J Clin Pathol, February 1, 2009; 131(2): 257 - 263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Suri, P. Das, A. Jain, M. C. Sharma, S. A. Borkar, A. Suri, D. Gupta, and C. Sarkar Pediatric glioblastomas: A histopathological and molecular genetic study Neuro-oncol, January 1, 2009; 11(3): 274 - 280. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ichimura, D. M. Pearson, S. Kocialkowski, L. M. Backlund, R. Chan, D. T.W. Jones, and V. P. Collins IDH1 mutations are present in the majority of common adult gliomas but rare in primary glioblastomas Neuro-oncol, January 1, 2009; 11(4): 341 - 347. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Hawes, J. D. Nerva, and K. M. Reilly Novel Dual-Reporter Preclinical Screen for Antiastrocytoma Agents Identifies Cytostatic and Cytotoxic Compounds J Biomol Screen, September 1, 2008; 13(8): 795 - 803. [Abstract] [PDF]
|
|
|
|
 |
|
 L. P. Petalidis, A. Oulas, M. Backlund, M. T. Wayland, L. Liu, K. Plant, L. Happerfield, T. C. Freeman, P. Poirazi, and V. P. Collins Improved grading and survival prediction of human astrocytic brain tumors by artificial neural network analysis of gene expression microarray data Mol. Cancer Ther., May 1, 2008; 7(5): 1013 - 1024. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Huang, A. Wallqvist, and D. G. Covell Targeting changes in cancer: assessing pathway stability by comparing pathway gene expression coherence levels in tumor and normal tissues. Mol. Cancer Ther., September 1, 2006; 5(9): 2417 - 2427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. F. Pollack, R. L. Hamilton, R. W. Sobol, J. Burnham, A. J. Yates, E. J. Holmes, T. Zhou, and J. L. Finlay O6-Methylguanine-DNA Methyltransferase Expression Strongly Correlates With Outcome in Childhood Malignant Gliomas: Results From the CCG-945 Cohort J. Clin. Oncol., July 20, 2006; 24(21): 3431 - 3437. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Reardon, J. N. Rich, H. S. Friedman, and D. D. Bigner Recent Advances in the Treatment of Malignant Astrocytoma J. Clin. Oncol., March 10, 2006; 24(8): 1253 - 1265. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ghosh and G. J. Duigou Decreased Replication Ability of E1-Deleted Adenoviruses Correlates with Increased Brain Tumor Malignancy Cancer Res., October 1, 2005; 65(19): 8936 - 8943. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. K. Wiencke, K. Aldape, A. McMillan, J. Wiemels, M. Moghadassi, R. Miike, K. T. Kelsey, J. Patoka, J. Long, and M. Wrensch Molecular Features of Adult Glioma Associated with Patient Race/Ethnicity, Age, and a Polymorphism in O6-Methylguanine-DNA-Methyltransferase Cancer Epidemiol. Biomarkers Prev., July 1, 2005; 14(7): 1774 - 1783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Gomez-Manzano, W.K. A. Yung, R. Alemany, and J. Fueyo Genetically modified adenoviruses against gliomas: From bench to bedside Neurology, August 10, 2004; 63(3): 418 - 426. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V P Collins Brain tumours: classification and genes J. Neurol. Neurosurg. Psychiatry, June 1, 2004; 75(suppl_2): ii2 - ii11. [Full Text] [PDF]
|
|
|
|
|
|
S. Godard, G. Getz, M. Delorenzi, P. Farmer, H. Kobayashi, I. Desbaillets, M. Nozaki, A.-C. Diserens, M.-F. Hamou, P.-Y. Dietrich, et al. Classification of Human Astrocytic Gliomas on the Basis of Gene Expression: A Correlated Group of Genes with Angiogenic Activity Emerges As a Strong Predictor of Subtypes Cancer Res., October 15, 2003; 63(20): 6613 - 6625. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Watanabe, Y. Katayama, A. Yoshino, C. Komine, and T. Yokoyama Deregulation of the TP53/p14ARF Tumor Suppressor Pathway in Low-Grade Diffuse Astrocytomas and Its Influence on Clinical Course Clin. Cancer Res., October 15, 2003; 9(13): 4884 - 4890. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Backlund, B. R. Nilsson, H. M. Goike, E. E. Schmidt, L. Liu, K. Ichimura, and V. P. Collins Short Postoperative Survival for Glioblastoma Patients with a Dysfunctional Rb1 Pathway in Combination with No Wild-type PTEN Clin. Cancer Res., September 15, 2003; 9(11): 4151 - 4158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. van den Boom, M. Wolter, R. Kuick, D. E. Misek, A. S. Youkilis, D. S. Wechsler, C. Sommer, G. Reifenberger, and S. M. Hanash Characterization of Gene Expression Profiles Associated with Glioma Progression Using Oligonucleotide-Based Microarray Analysis and Real-Time Reverse Transcription-Polymerase Chain Reaction Am. J. Pathol., September 1, 2003; 163(3): 1033 - 1043. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Gilbertson, D. A. Hill, R. Hernan, M. Kocak, R. Geyer, J. Olson, A. Gajjar, L. Rush, R. L. Hamilton, S. D. Finkelstein, et al. ERBB1 Is Amplified and Overexpressed in High-grade Diffusely Infiltrative Pediatric Brain Stem Glioma Clin. Cancer Res., September 1, 2003; 9(10): 3620 - 3624. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Berggren, R. Kumar, S. Sakano, L. Hemminki, T. Wada, G. Steineck, J. Adolfsson, P. Larsson, U. Norming, H. Wijkstrom, et al. Detecting Homozygous Deletions in the CDKN2A(p16INK4a)/ARF(p14ARF) Gene in Urinary Bladder Cancer Using Real-Time Quantitative PCR Clin. Cancer Res., January 1, 2003; 9(1): 235 - 242. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Uhrbom, C. Dai, J. C. Celestino, M. K. Rosenblum, G. N. Fuller, and E. C. Holland Ink4a-Arf Loss Cooperates with KRas Activation in Astrocytes and Neural Progenitors to Generate Glioblastomas of Various Morphologies Depending on Activated Akt Cancer Res., October 1, 2002; 62(19): 5551 - 5558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Wadhwa, T. Sugihara, Md. K. Hasan, K. Taira, R. R. Reddel, and S. C. Kaul A Major Functional Difference between the Mouse and Human ARF Tumor Suppressor Proteins J. Biol. Chem., September 20, 2002; 277(39): 36665 - 36670. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sanchez-Aguilera, M. Sanchez-Beato, J. F. Garcia, I. Prieto, M. Pollan, and M. A. Piris p14ARF nuclear overexpression in aggressive B-cell lymphomas is a sensor of malfunction of the common tumor suppressor pathways Blood, February 15, 2002; 99(4): 1411 - 1418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. F. Pollack, S. D. Finkelstein, J. Woods, J. Burnham, E. J. Holmes, R. L. Hamilton, A. J. Yates, J. M. Boyett, J. L. Finlay, R. Sposto, et al. Expression of p53 and Prognosis in Children with Malignant Gliomas N. Engl. J. Med., February 7, 2002; 346(6): 420 - 427. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Garcia, R. Villuendas, M. Sanchez-Beato, A. Sanchez-Aguilera, L. Sanchez, I. Prieto, and M. A. Piris Nucleolar p14ARF Overexpression in Reed-Sternberg Cells in Hodgkin's Lymphoma : Absence of p14ARF/Hdm2 Complexes Is Associated with Expression of Alternatively Spliced Hdm2 Transcripts Am. J. Pathol., February 1, 2002; 160(2): 569 - 578. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. F. Pollack, S. D. Finkelstein, J. Burnham, E. J. Holmes, R. L. Hamilton, A. J. Yates, J. L. Finlay, and R. Sposto Age and TP53 Mutation Frequency in Childhood Malignant Gliomas: Results in a Multi-institutional Cohort Cancer Res., October 1, 2001; 61(20): 7404 - 7407. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Sonoda, T. Ozawa, K. D. Aldape, D. F. Deen, M. S. Berger, and R. O. Pieper Akt Pathway Activation Converts Anaplastic Astrocytoma to Glioblastoma Multiforme in a Human Astrocyte Model of Glioma Cancer Res., September 1, 2001; 61(18): 6674 - 6678. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Calogero, A. Arcella, G. De Gregorio, A. Porcellini, D. Mercola, C. Liu, V. Lombari, M. Zani, G. Giannini, F. M. Gagliardi, et al. The Early Growth Response Gene EGR-1 Behaves as a Suppressor Gene That Is Down-Regulated Independent of ARF/Mdm2 but not p53 Alterations in Fresh Human Gliomas Clin. Cancer Res., September 1, 2001; 7(9): 2788 - 2796. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Sonoda, T. Ozawa, Y. Hirose, K. D. Aldape, M. McMahon, M. S. Berger, and R. O. Pieper Formation of Intracranial Tumors by Genetically Modified Human Astrocytes Defines Four Pathways Critical in the Development of Human Anaplastic Astrocytoma Cancer Res., July 1, 2001; 61(13): 4956 - 4960. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Nicholson, N. T. Okby, M. A. Khan, J. A. Welsh, M. G. McMenamin, W. D. Travis, J. R. Jett, H. D. Tazelaar, V. Trastek, P. C. Pairolero, et al. Alterations of p14ARF, p53, and p73 Genes Involved in the E2F-1-mediated Apoptotic Pathways in Non-Small Cell Lung Carcinoma Cancer Res., July 1, 2001; 61(14): 5636 - 5643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Chen, K. Aldape, J. K. Wiencke, K. T. Kelsey, R. Miike, R. L. Davis, J. Liu, A. Kesler-Diaz, M. Takahashi, and M. Wrensch Ethnicity Delineates Different Genetic Pathways in Malignant Glioma Cancer Res., May 1, 2001; 61(10): 3949 - 3954. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. Di Bacco, K. Keeshan, S. L. McKenna, and T. G. Cotter Molecular Abnormalities in Chronic Myeloid Leukemia: Deregulation of Cell Growth and Apoptosis Oncologist, October 1, 2000; 5(5): 405 - 415. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
exon 18; Ref.