Oncogene (2007) 26, 5010–5016
& 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00
www.nature.com/onc
SHORT COMMUNICATION
Hypermethylation and transcriptional downregulation of the CITED4 gene
at 1p34.2 in oligodendroglial tumours with allelic losses on 1p and 19q
B Tews1,6, P Roerig2,6, C Hartmann3, M Hahn1, J Felsberg2, B Blaschke2, M Sabel4, A Kunitz3,
G Toedt1, K Neben1, A Benner5, A von Deimling3, G Reifenberger2,6 and P Lichter1,6
1
Division of Molecular Genetics (B060), Deutsches Krebsforschungszentrum, Heidelberg, Germany; 2Department of Neuropathology,
Heinrich-Heine-University, Düsseldorf, Germany; 3Department of Neuropathology, Charité University Medicine, Berlin, Germany;
4
Department of Neurosurgery, Heinrich-Heine-University, Düsseldorf, Germany and 5Central Unit Biostatistics (C060), Deutsches
Krebsforschungszentrum, Heidelberg, Germany
Deletions of chromosomal arms 1p and 19q are frequent
in oligodendroglial tumours and have been associated
with sensitivity to radio- and chemotherapy as well as
favourable prognosis. By using microarray-based expres-
sion profiling, we found that oligodendroglial tumours
with 1p and 19q losses showed significantly lower
expression of the CBP/p300-interacting transactivator
with glutamic acid/aspartic acid-rich carboxyl-terminal
domain 4 gene (CITED4) at 1p34.2 as compared to
tumours without 1p and 19q losses. Mutational analysis
showed no CITED4 mutations in gliomas. However, 1p
and 19q losses as well as low expression of CITED4
transcripts were significantly associated with hypermethy-
lation of the CITED4-associated CpG island. In line with
the latter finding, treatment of CITED4 hypermethylated
glioma cell lines with 5-aza-20 -deoxycytidine and trichos-
tatine A resulted in a marked increase of the CITED4
transcript levels. Furthermore, CITED4 hypermethylation
was significantly associated with longer recurrence-free
and overall survival of patients with oligodendroglial
tumours. Taken together, our results indicate that
CITED4 is epigenetically silenced in the vast majority
of oligodendroglial tumours with 1p and 19q deletions and
suggest CITED4 hypermethylation as a novel prognostic
marker in oligodendroglioma patients.
Oncogene (2007) 26, 5010–5016; doi:10.1038/sj.onc.1210297;
published online 19 February 2007
Keywords: CITED4; gene expression profiling; oligo-
dendroglioma; CpG island; hypermethylation; tumour
suppressor gene
Oligodendroglial tumours are primary neoplasms of
the central nervous system that frequently demons-
trate allelic losses on chromosomal arms 1p and 19q
Correspondence: Professor G Reifenberger, Department of Neuro-
pathology, Heinrich-Heine-University, Moorenstrasse 5, D-40225
Düsseldorf, Germany.
E-mail: reifenberger@med.uni-duesseldorf.de
6
These authors contributed equally to this work.
Received 5 April 2006; revised 14 December 2006; accepted 3 January
2007; published online 19 February 2007
(Reifenberger and Louis, 2003). Recent data indicate
that these combined 1p and 19q losses are mediated by
an unbalanced t(1;19)(q10;p10) translocation (Griffin
et al., 2006; Jenkins et al., 2006). Importantly, 1p/19q
deletion is associated with favourable response to radio-
and chemotherapy as well as prolonged survival
(Cairncross et al., 1998, 2006; van den Bent et al.,
2006). Thus, molecular testing for these losses provides
clinically valuable information beyond the conventional
histologic classification.
To date, several genes located on 1p or 19q have
been reported as putative oligodendroglioma-associated
tumour suppressor genes. For example, the cyclin-
dependent kinase inhibitor 2C (CDKN2C, 1p32) gene
is mutated or homozygously deleted in rare cases of
anaplastic oligodendroglioma (Reifenberger and Louis,
2003). Other genes showed reduced expression in 1p-
deleted gliomas, sometimes associated with promoter
hypermethylation but never with mutation. These
include the calmodulin-binding transcription activator
1 gene (CAMTA1, 1p36), the DNA fragmentation
factor subunit b gene (DFFB, 1p36), the SHREW1 gene
(1p36.32), the tumour protein p73 gene (TP73, 1p36.3)
and the RAD54 gene (1p32) (Dong et al., 2002;
Barbashina et al., 2005; McDonald et al., 2005, 2006).
The p190RhoGAP gene (19q13.3) was reported as a
candidate tumour suppressor that inhibits PDGF-
induced murine oligodendrogliomas (Wolf et al.,
2003). However, its role in human oligodendrogliomas
is unknown. More recently, the myelin-related epithelial
membrane protein gene 3 (EMP3, 19q13.3) was found
to be epigenetically silenced in malignant gliomas and
neuroblastomas (Alaminos et al., 2005).
We applied microarray-based expression profiling to
identify novel genes that are differentially expressed
between gliomas with and without 1p/19q losses (Tews
et al., 2006). In total, 35 gliomas were studied, including
seven oligodendrogliomas WHO grade II, eight diffuse
astrocytomas WHO grade II, 14 anaplastic oligoden-
drogliomas WHO grade III and six anaplastic oligo-
astrocytomas WHO grade III (Tews et al., 2006).
Differentially expressed genes allowing for the discrimi-
nation between gliomas with and without 1p/19q losses
were revealed by significance analysis of microarrays

(SAM) followed by prediction analysis for microarrays
(PAM) (Tusher et al., 2001; Tibshirani et al., 2002; Tews
et al., 2006). One of the genes with significantly lower
expression in 1p/19q-deleted gliomas was the CREB-
binding protein (CBP)/p300-interacting transactivator
with glutamic acid/aspartic acid-rich carboxyl-terminal
domain 4 gene (CITED4) at 1p34.2 (OMIM 606815),
which has been linked to cancer-relevant cellular
processes (Braganca et al., 2002; Fox et al., 2004). Here,
we report on a detailed molecular analysis of CITED4
aberrations in human gliomas and provide evidence for
an association of CITED4 hypermethylation with longer
survival of oligodendroglial tumour patients.
CITED4 hypermethylation in oligodendroglial tumours
B Tews et al
5011
In total, we investigated tumour samples from 78
patients (40 female and 38 male; mean age at operation:
48 years; range: 11–91 years) with gliomas (21 oligoden-
drogliomas WHO grade II, three oligoastrocytomas
WHO grade II, eight diffuse astrocytomas WHO grade
II, 29 anaplastic oligodendrogliomas WHO grade III
and 17 anaplastic oligoastrocytomas WHO grade III).
The tumours were collected at the Departments of
Neuropathology, Heinrich-Heine-University, Düssel-
dorf, and Charité University Hospital, Berlin, and
investigated in an anonymized manner as approved by
the local institutional review boards. All tumours were
classified according to the WHO classification of

Figure 1 (a) Examples of results obtained by quantitative real-time reverse transcription-PCR, which was performed on 41 tumours
using primers 50 -gggaggacagtttggcttca-30 and 50 -gggagaggacacgatccaag-30 for the detection of CITED4 transcripts, SYBR-Green
fluorescent dye incorporation and the ABI PRISM 5700 sequence detection system (Applied Biosystems, Foster City, CA, USA).
CITED4 expression in each tumour was calculated relative to non-neoplastic brain tissue (BD Biosciences, St Jose, CA, USA) after
normalisation to the mRNA levels of ADP-ribosylation factor 1 (ARF1, GenBank accession no. NM_001658, primers: 50 -gac
cacgatcctctacaagc-30 and 50 -tcccacacagtgaagctgatg-30 ) or actin-g (ACTG1, GenBank accession no. NM_001614, primers 50 -cag
ctctcgcactctgttctt-30 and 50 -cgacgatggaaggaaacac-30 ). Note: reduced level of CITED4 mRNA in a 1p/19q-deleted glioma (AO84) as compared
to a glioma without 1p/19q losses (A82) and non-neoplastic brain tissue (NB). Abscissa, cycle number; Ordinate, relative amount
of PCR product (RFU, relative fluorescence units). The curves obtained for A82 and NB are nearly identical for ARF1 and CITED4.
In AO84, the ARF1 curve is shifted to the left and the CITED4 curve to the right relative to the respective curves for A82 and NB
(t, threshold value). The calculated CITED4 mRNA level in AO84 was 20% of the level in NB and A82. (b) Sequencing of the
CITED4-associated CpG island after sodium bisulfite treatment of DNA revealed methylation of CpG sites in AO84 (arrowheads in
upper lane) but not in A82 and NB (middle lanes). DNA methylated in vitro using SssI (CpG) methylase (NewEngland Biolabs,
Beverly, MA, USA) served as a positive control (lower lane). Shown is the reverse sequence from nucleotides g.40997168 to g.40997212
(GenBank accession no. NC_000001). (c) Induction of CITED4 mRNA in three glioma cell lines (A172 and T98G, American Type
Culture Collection, Manassas, VA, USA; U138MG, German National Ressource Center for Biological Material, Braunschweig,
Germany) by 5-aza-20 -deoxycytidine and trichostatin A treatment as determined by real-time reverse transcription–PCR. Cell lines
were grown either under standard conditions (1) or under two different treatment conditions: (2) 500 nM 5-aza-20 -deoxycytidine for
48 h plus 1 mM trichostatin A for 24 h or (3) 1 mM 5-aza-20 -deoxycytidine for 72 h plus 1 mM trichostatin A for 24 h. (d) Results of reverse
transcription–PCR analysis of CITED4 expression in A172 cells before (1) and after treatment with 5-aza-20 -deoxycytidine and
trichostatin A under two different conditions (see Figure 1c). Note marked induction of CITED4 mRNA after treatment (bp ¼ base
pairs).

Oncogene
 CITED4 hypermethylation in oligodendroglial tumours
B Tews et al
5012
tumours of the nervous system (Reifenberger et al.,
2000). A tumour cell content of at least 80% was
histologically assured for each specimen used for
molecular analysis. Extraction of DNA and RNA from
frozen tumour samples as well as DNA from leukocytes
was carried out as reported elsewhere (van den Boom
et al., 2003).
In line with our microarray results (Tews et al., 2006),
quantitative real-time reverse transcription–polymerase
chain reaction (PCR) analysis of 41 gliomas revealed
CITED4 transcript levels reduced by at least 50%
relative to non-neoplastic brain tissue in 17 of 21
gliomas (81%) with 1p losses spanning the CITED4
locus (Figure 1a). Twelve of these 17 tumours showed

Figure 2 Methylation patterns in the 50 -CpG island of CITED4 in 62 gliomas and two samples of non-neoplastic brain tissue (NB1,
NB2). Methylation at each of the 54 investigated CpG sites was determined by PCR amplification of two adjacent fragments of
the CITED4 CpG island (covering 54 CpG sites between nucleotides 40 996 844 and 40 997 411, Genbank accession no. NC_000001)
from sodium bisulfite-modified DNA using primers 50 -tttggttagtttaagttttattttagt-30 and 50 -aatcacaaaaacccgcaacctataa-30 as well as
50 -ttataggttgcgggtttttgtgatt-30 and 50 -acctctacaccaaacgataacc-30 , followed by direct sequencing. The methylation status at each CpG site
was rated according to the following scale: 0, not methylated; 1, weakly methylated (intensity of the methylated signal lower than 1/3
relative to the unmethylated signal); 2, moderately methylated (intensity of the methylated signal between 1/3 and 2/3 relative to the
unmethylated signal); 3, strongly methylated (intensity of the methylated signal higher than 2/3 of the unmethylated signal). This rating
is represented in a grey-scale pattern as indicated below the figure (na, not analysed). Each tumour was assigned to one of two groups:
(1) no CITED4 hypermethylation (methylation score 1, 2 or 3 in less than 50% of the investigated CpG sites) or (2) CITED4
hypermethylation (methylation score 1, 2 or 3 in more than 50% of the investigated CpG sites). CITED4 mRNA expression (Expr.)
levels relative to non-neoplastic brain tissue were subdivided into two categories as shown below the figure. Allelic losses on 1p and 19q
are also indicated for each tumour ( þ , allelic loss; , no allelic loss; * allelic loss restricted to distal 1p not including the CITED4
locus). At least five microsatellite loci located on 1p (D1S200, D1S211, D1S507, D1S489, D1S468) and five microsatellite loci located
on 19q (D19S572, D19S219, D19S1182, D19S596, D19S210) were assessed for allelic loss in each tumour (Felsberg et al., 2004). The
location of exon 1 and the start codon (ATG) are indicated on top of the figure. Note a close association between 1p/19q losses and
CITED4 hypermethylation. Furthermore, 15 of 19 tumours with hypermethylation showed CITED4 mRNA levels of o0.5-fold
relative to the reference brain tissue, as compared to 7 of 22 tumours without hypermethylation.

Oncogene

transcript levels reduced by at least 75% relative to non-
neoplastic brain tissue when ARF1 was used as the
reference gene. When ACTG1 was used as reference, all
17 tumours showed CITED4 transcript levels reduced
by more than 75% relative to non-neoplastic brain
tissue. In contrast, only 5 of 20 gliomas (25%) with
retention of both CITED4 alleles showed reductions in
the CITED4 transcript levels by at least 50% relative to
non-neoplastic brain tissue (Figure 2). Statistical ana-
lyses confirmed that CITED4 mRNA expression was
significantly lower (Student’s t-test, Po0.001) and more
commonly reduced (Fisher’s exact test, Po0.001) in 1p/
19q-deleted gliomas.
Mutation screening of 45 gliomas, including 22
tumours with 1p/19q losses, detected 15 different
CITED4 polymorphisms, mostly single nucleotide ex-
changes (Table 1). Four polymorphisms (c.95T>A:
Leu32Gln; c.170G>C, Arg57Pro; c.173A>G, Gln58
Arg; c.393C>T, Gly131Gly) were linked to each other,
that is, occurred together in nine patients. A 36-base
pair in-frame deletion polymorphism (c.327-362del,
GenBank accession no. NM_133467.2) was found in
two patients, with one patient being constitutionally
homozygous for this polymorphism. However, none of
Table 1 CITED4 sequence polymorphisms identified in 45 glioma
patients
Polymorphism Sequence change
number
1 NC_000001:g.40997143G>A
2 NC_000001:g.40997205C>G
3 NC_000001:g.40997259del1
4 NC_000001:g.40997260del1
5 NM_133467.2:c.-74C>G
6 NM_133467.2:c.27G>C (Glu9Asp)
7 NM_133467.2:c.54G>T (Pro18 Pro)
8 NM_133467.2:c.-72G>T
9 NM_133467.2:c.95T>A (Leu32Gln)
10 NM_133467.2:c.170G>C (Arg57Pro)
11 NM_133467.2:c.173A>G (Gln58Arg)
12 NM_133467.2:c.327-362del (PPGPQPAPSAAA110-
121del)
13 NM_133467.2:c.393C>T (Gly131Gly)
14 NM_133467.2:c.410T>G (Ile137Ser)
15 NM_133467.2:c.519G>T (Leu173Phe)
Most polymorphisms consist in exchanges or deletions of single
nucleotides. Polymorphism 12 is a 36-bp in-frame deletion polymorph-
ism that was found in two patients, one of which was constitutionally
homozygous for this deletion. To detect sequence changes in CITED4,
we screened its entire coding region in five overlapping fragments using
the following primers: 50 -gggccaagacctagatgcag-30 and 50 -aaaccaaaccc
gactggtg-30 (229-bp amplicon), 50 -cgcaaggtgcgcagtagt-30 and 50 -ctctgca
ccaggcggtag-30 (229-bp amplicon), 50 -catggccgaccacctgat-30 and
50 -gaaagggctggaaggaggac-30 (230-bp amplicon), 50 -ctaccgcctggtgcagag-30
and 50 -agtccgagaagcagtcgaac-30 (488-bp amplicon), 50 -catggacgccgaactc
atc-30 and 50 -agtcgggccctttctcctct-30 (212-bp amplicon). The 488-bp
amplicon was directly sequenced while the other four amplicons
were subjected to single-strand conformation polymorphism/
heteroduplex analysis (van den Boom et al., 2003). PCR products
were sequenced in both directions using cycle-sequencing (BigDye
cycle sequencing kit, Applied Biosystems) and an ABI PRISM 377
semi-automated DNA sequencer (Applied Biosystems). To determine
whether sequence changes detected in tumour DNA were of somatic or
germ-line origin, we additionally sequenced the corresponding
leucocyte DNA.
CITED4 hypermethylation in oligodendroglial tumours
B Tews et al
5013
the 45 tumours carried somatic mutations, indicating
that structural alterations of CITED4 are absent or very
rare in gliomas.
We next investigated the role of aberrant methylation
of the CITED4-associated CpG island in 62 gliomas,
including the 41 tumours studied for CITED4 expres-
sion. CITED4 hypermethylation was detected in 31 of
34 gliomas with 1p/19q losses (91%) but only in 2 of 28
gliomas without 1p/19q losses (7%) (Figures 1b and 2)
(Fisher’s exact test, Po0.001). CITED4 hypermethy-
lation was absent in leucocyte DNA from 16 patients
with CITED4 hypermethylated gliomas and in non-neo-
plastic brain samples from two unrelated individuals
(Figure 2).
Correlation of CITED4 expression and methylation
revealed that CITED4 transcript levels were reduced by
at least 50% relative to non-neoplastic brain tissue in 15
of 19 tumours (79%) with CITED4 hypermethylation
but only in 7 of 22 tumours (32%) without CITED4
hypermethylation. Statistical analyses confirmed that
CITED4 mRNA levels were significantly lower (Student’s
t-test, Po0.01) and more commonly reduced (Fisher’s
exact test, Po0.01) in CITED4-hypermethylated
gliomas. In line with these findings, 5-aza-20 -deoxy-
cytidine and trichostatin A treatment of three CITED4-
hypermethylated glioma cell lines (A172, T98G,
U138MG) resulted in markedly increased mRNA
levels (Figure 1c–d).
To assess the relationship between CITED4 hyper-
methylation and patient survival, we investigated 45
patients with oligodendroglial tumours (19 WHO grade
II, 26 WHO grade III) and available follow-up data
(Felsberg et al., 2004). The median follow-up time after
diagnosis was 122 months. Twenty-three patients died of
their tumours while 22 patients were still alive at last
follow-up. Median overall survival after diagnosis was
103 months, median recurrence-free survival was 42
months. All patients were treated by open resection.
Twenty-four of the 26 patients with WHO grade III
tumours received adjuvant therapy, comprising radio-
therapy in 16 patients, alkylating chemotherapy in one
patient, and combined radio- and chemotherapy in
seven patients. Univariate statistical analysis revealed
that CITED4 hypermethylation was significantly asso-
ciated with longer recurrence-free and overall survival in
the entire group of 45 patients and in the 26 patients
with WHO grade III tumours (Figure 3a–d). Similarly,
significant associations were found between the 1p/19q
allelic status and survival (Figure 3e–h). Multivariate
analysis identified CITED4 hypermethylation and 1p/
19q deletions as significant prognostic factors adjusted
for WHO grade and patient age (Table 2).
The CITED4 gene encodes a 184-amino-acid protein
of 24.5 kDa that binds CBP and the tumour suppressor
protein EP300 (E1A-binding protein, 300 kDa), func-
tions as a co-activator of the transcription factor AP-2
(Braganca et al., 2002), blocks binding of hypoxia-
inducible factor 1 alpha (HIF1a) to EP300 and inhibits
HIF1a transactivation as well as hypoxia-mediated
reporter gene activation (Fox et al., 2004). Loss of
nuclear expression or cytoplasmic translocation of
Oncogene
 CITED4 hypermethylation in oligodendroglial tumours
B Tews et al
5014

Figure 3 (a–d) Significant associations (log rank tests) of CITED4 hypermethylation with longer recurrence-free survival (RFS) and
overall survival (OS) in patients with oligodendroglial tumours (meth., CITED4 hypermethylated; no meth., CITED4 not
hypermethylated). Shown are Kaplan–Meier survival curves obtained for 45 patients (a, b), including 19 patients with WHO grade II
and 26 patients with WHO grade III tumours, as well as in the subgroup of 26 patients with WHO grade III tumours (c, d). (e–h)
Kaplan–Meier survival curves stratified according to the 1p/19q allelic status in the same patients. Similar to CITED4
hypermethylation, 1p and 19q losses are significantly associated with longer RFS and OS in all 45 patients (e, f) and in the subgroup
of 26 patients with WHO grade III tumours (g, h). Kaplan–Meier survival curve estimation and log rank tests were performed with the
GraphPad Prism 4 software.

CITED4 has been implicated in breast cancer develop-
ment (Fox et al., 2004). Here, we report that CITED4
mRNA expression in gliomas is significantly lower in
1p/19q-deleted versus non-deleted tumours. Further-
more, we found reduced expression closely linked to
Oncogene
CITED4 hypermethylation. In fact, our data indicate
biallelic CITED4 inactivation by loss of one allele and
hypermethylation of the other allele in the vast majo-
rity of 1p/19q-deleted oligodendroglial tumours. The
absence of CITED4 hypermethylation in DNA extracted
 CITED4 hypermethylation in oligodendroglial tumours
B Tews et al
5015
Table 2 Results of multivariate Cox proportional hazards regression analysis based on 45 patients with oligodendroglial tumours (19 patients with
WHO grade II and 26 patients with WHO grade III tumours)
Factor RFS OS

Hazard ratio (95% CI)a P-value Hazard ratio (95% CI)* P-value

(a)
Age at operation 1.39 (1.07–1.80) 0.01 1.26 (0.93–1.71) 0.13
WHO grade 1.78 (0.87–3.61) 0.11 3.09 (1.23–7.76) 0.02
CITED4 methylation status 0.34 (0.17–0.70) 0.003 0.25 (0.10–0.62) 0.003

(b)
Age at operation 1.36 (1.05–1.76) 0.02 1.24 (0.92–1.67) 0.16
WHO grade 1.79 (0.88–3.62) 0.11 3.33 (1.33–8.29) 0.01
1p/19q allelic status 0.37 (0.18–0.75) 0.006 0.20 (0.08–0.51) o0.001

As CITED4 hypermethylation and 1p/19q losses were highly correlated, two separate models were calculated, either including the CITED4
methylation status (a) or the 1p/19q allelic status (b), each together with the WHO grade and the patient age at operation. Multivariate analyses
were performed using the statistical software package R, version 2.1.1 (R Development Core Team, 2005). aHazard ratios and confidence intervals
(95% CIs) are computed for a 10-year increment in age, for comparing WHO grade III versus WHO grade II, for comparing CITED4 methylation
status positive versus negative, or for comparing LOH 1p/19q versus no LOH 1p/19q. P-values are computed for the Wald test statistics. RFS,
recurrence-free survival; OS, overall survival.

from leucocytes and non-neoplastic brain tissue indi-
cates that CITED4 is not imprinted. However, it
remains unclear whether CITED4 hypermethylation is
already present in the yet unknown cell of origin of
oligodendrogliomas or represents an acquired aberra-
tion selected for during tumorigenesis.
The molecular mechanisms by which CITED4 in-
activation might promote oligodendroglial tumour
growth are as yet unknown. CITED4 downregulation
could interfere with EP300 tumour suppressor activity
(Iyer et al., 2004), which, among other functions, plays a
crucial role in p53 regulation (Grossman, 2001).
Interestingly, 1p/19q-deleted oligodendrogliomas usual-
ly lack TP53 mutations (Reifenberger and Louis, 2003).
However, further studies need to substantiate a tumour-
suppressive function of the CITED4 protein and to
clarify the functional consequences of its inactivation in
oligodendroglial tumours.
We found a similar prognostic significance of
CITED4 hypermethylation in oligodendroglioma pa-
tients as obtained for 1p/19q deletion. Thus, diagnostic
assessment of CITED4 hypermethylation might be an
alternative approach to 1p/19q-testing. However, the
frequency of CITED4 hypermethylation in other glio-
mas, in particular anaplastic astrocytomas and glioblas-
tomas, remains to be determined. Furthermore, future
studies need to address whether CITED4 protein
expression analysis, for example, by immunohistoche-
mistry, might substitute for methylation analysis and
1p/19q-testing. It also needs to be investigated whether
CITED4 inactivation contributes to radio- and chemo-
sensitivity or just represents a surrogate marker for
other alterations in 1p/19q-deleted oligodendroglio-
mas. For example, 1p/19q-deletion has been linked
to hypermethylation of the O6-methylguanine DNA
methyltransferase (MGMT) gene (Möllemann et al.,
2005), which predicts response to alkylating chemo-
therapy in malignant gliomas (Hegi et al., 2005). Taken
together, our results suggest CITED4 as an interesting
novel candidate gene in oligodendroglial tumours and
should stimulate further research into its functional
roles and clinical significance.
Abbreviations
LOH, loss of heterozygosity; OS, overall survival; RFS,
recurrence-free survival; WHO, World Health Organisation.
Acknowledgements
This study was supported by grants from the Deutsche
Krebshilfe to GR and MS (70-3088-Sa1) as well as to GR
and AvD (70-3163-Wi3), and from the German Bundesmi-
nisterium für Bildung und Forschung (BMBF) within
the National Genome Research Network 2 (NGFN-2) to
PL, MH, GR, CH and AvD (01GS0460, 01GS0462 and
01GS0463). BT was a scholar of the Studienstiftung des
Deutschen Volkes.

References

Alaminos M, Davalos V, Ropero S, Setien F, Paz MF,
Herranz M et al. (2005). EMP3, a myelin-related gene
located in the critical 19q13.3 region, is epigenetically
silenced and exhibits features of a candidate tumor
suppressor in glioma and neuroblastoma. Cancer Res 65:
2565–2571.
Barbashina V, Salazar P, Holland EC, Rosenblum MK,
Ladanyi M. (2005). Allelic losses at 1p36 and 19q13 in
gliomas: correlation with histologic classification, definition
of a 150-kb minimal deleted region on 1p36, and evaluation
of CAMTA1 as a candidate tumor suppressor gene. Clin
Cancer Res 11: 1119–1128.
Oncogene
 CITED4 hypermethylation in oligodendroglial tumours
B Tews et al
5016
Braganca J, Swingler T, Marques FI, Jones T, Eloranta JJ,
Hurst HC et al. (2002). Human CREB-binding protein/
p300-interacting transactivator with ED-rich tail (CITED)
4, a new member of the CITED family, functions as a co-
activator for transcription factor AP-2. J Biol Chem 277:
8559–8565.
Cairncross G, Berkey B, Shaw E, Jenkins R, Scheithauer B,
Brachman D et al. (2006). Phase III trial of chemotherapy
plus radiotherapy compared with radiotherapy alone for
pure and mixed anaplastic oligodendroglioma: Intergroup
Radiation Therapy Oncology Group Trial 9402. J Clin
Oncol 24: 2707–2714.
Cairncross JG, Ueki K, Zlatescu MC, Lisle DK, Finkelstein
DM, Hammond RR et al. (1998). Specific genetic predictors
of chemotherapeutic response and survival in patients
with anaplastic oligodendrogliomas. J Natl Cancer Inst 90:
1473–1479.
Dong S, Pang JC, Hu J, Zhou LF, Ng H. (2002). Transcrip-
tional inactivation of TP73 expression in oligodendroglial
tumors. Int J Cancer 98: 370–375.
Felsberg J, Erkwoh A, Sabel MC, Kirsch L, Fimmers R,
Blaschke B et al. (2004). Oligodendroglial tumors: refine-
ment of candidate regions on chromosome arm 1p and
correlation of 1p/19q status with survival. Brain Pathol 14:
121–130.
Fox SB, Braganca J, Turley H, Campo L, Han C, Gatter KC
et al. (2004). CITED4 inhibits hypoxia-activated transcrip-
tion in cancer cells, and its cytoplasmic location in breast
cancer is associated with elevated expression of tumor cell
hypoxia-inducible factor 1alpha. Cancer Res 64: 6075–6081.
Griffin CA, Burger P, Morsberger L, Yonescu R, Swierczynski S,
Weingart JD et al. (2006). Identification of der(1;19)
(q10;p10) in five oligodendrogliomas suggests mechanism
of concurrent 1p and 19q loss. J Neuropathol Exp Neurol
65: 988–994.
Grossman SR. (2001). p300/CBP/p53 interaction and regula-
tion of the p53 response. Eur J Biochem 268: 2773–2778.
Hegi ME, Diserens AC, Gorlia T, Hamou MF, de Tribolet N,
Weller M et al. (2005). MGMT gene silencing and benefit
from temozolomide in glioblastoma. N Engl J Med 352:
997–1003.
Iyer NG, Ozdag H, Caldas C. (2004). p300/CBP and cancer.
Oncogene 23: 4225–4231.
Jenkins RB, Blair H, Ballman KV, Giannini C, Arusell RM,
Law M et al. (2006). A t(1;19)(q10;p10) mediates the
combined deletions of 1p and 19q and predicts a better
prognosis of patients with oligodendroglioma. Cancer Res
66: 9852–9861.
McDonald JM, Dunmire V, Taylor E, Sawaya R, Bruner J,
Fuller GN et al. (2005). Attenuated expression of DFFB is a
hallmark of oligodendrogliomas with 1p-allelic loss. Mol
Cancer 4: 35.
Oncogene
McDonald JM, Dunlap S, Cogdell D, Dunmire V, Wei Q,
Starzinski-Powitz A et al. (2006). The SHREW1 gene,
frequently deleted in oligodendrogliomas, functions to
inhibit cell adhesion and migration. Cancer Biol Ther 5:
300–304.
Möllemann M, Wolter M, Felsberg J, Collins VP, Reifenber-
ger G. (2005). Frequent promoter hypermethylation and low
expression of the MGMT gene in oligodendroglial tumors.
Int J Cancer 113: 379–385.
R Development Core Team (2005). A language and environment
for statistical computing. R Foundation for Statistical
Computing: Vienna, Austria, ISBN 3-900051-07-0, http://
www.R-project.org.
Reifenberger G, Kros JM, Burger PC, Louis DN, Collins VP.
(2000). Oligodendroglial tumours and mixed gliomas. In:
Kleihues P, Cavence WK (eds). Tumours of the Nervous
System – Pathology and Genetics. World Health Organisation
Classification of Tumours. IARC Press: Lyon, pp 55–70.
Reifenberger G, Louis DN. (2003). Oligodendroglioma:
toward molecular definitions in diagnostic neuro-oncology.
J Neuropathol Exp Neurol 62: 111–126.
Tews B, Felsberg J, Hartmann C, Kunitz A, Hahn M, Toedt G
et al. (2006). Identification of novel oligodendroglioma-
associated candidate tumor suppressor genes in 1p36 and
19q13 using microarray-based expression profiling. Int J
Cancer 119: 792–800.
Tibshirani R, Hastie T, Narasimhan B, Chu G. (2002).
Diagnosis of multiple cancer types by shrunken centroids
of gene expression. Proc Natl Acad Sci USA 99: 6567–6572.
Tusher VG, Tibshirani R, Chu G. (2001). Significance analysis
of microarrays applied to the ionizing radiation response.
Proc Natl Acad Sci USA 98: 5116–5121.
van den Bent MJ, Carpentier AF, Brandes AA, Sanson M,
Taphoorn MJ, Bernsen HJ et al. (2006). Adjuvant procar-
bazine, lomustine, and vincristine improves progression-free
survival but not overall survival in newly diagnosed
anaplastic oligodendrogliomas and oligoastrocytomas: a
randomized European Organisation for Research and
Treatment of Cancer phase III trial. J Clin Oncol 20:
2715–2722.
van den Boom J, Wolter M, Kuick R, Misek DE, Youkilis AS,
Wechsler DS et al. (2003). 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 163: 1033–1043.
Wolf RM, Draghi N, Liang X, Dai C, Uhrbom L, Eklof C
et al. (2003). p190RhoGAP can act to inhibit PDGF-
induced gliomas in mice: a putative tumor suppressor
encoded on human chromosome 19q13.3. Genes Dev 17:
476–487.