MAL promoter hypermethylation as a novel prognostic marker in gastric cancer.

Abstract

T-lymphocyte maturation associated protein, MAL, has been described as a tumour-suppressor gene with diagnostic value in colorectal and oesophageal cancers, and can be inactivated by promoter hypermethylation. The aim of this study was to analyse the prevalence of MAL promoter hypermethylation and the association with mRNA expression in gastric cancers and to correlate methylation status to clinicopathological data. Bisulphite-treated DNA isolated from formalin-fixed and paraffin-embedded samples of 202 gastric adenocarcinomas and 22 normal gastric mucosae was subjected to real-time methylation-specific PCR (Q-MSP). Two regions within the MAL promoter (M1 and M2) were analysed. In addition, 17 frozen gastric carcinomas and two gastric cancer cell lines were analysed both by Q-MSP and real-time RT-PCR. Methylation of M1 and M2 occurred in 71 and 80% of the gastric cancers, respectively, but not in normal gastric mucosa tissue. Hypermethylation of M2, but not M1, correlated with significantly better disease-free survival in a univariate (P=0.03) and multivariate analysis (P=0.03) and with downregulation of expression (P=0.01). These results indicate that MAL has a putative tumour-suppressor gene function in gastric cancer, and detection of promoter hypermethylation may be useful as a prognostic marker.

Full text

Page 1

MAL promoter hypermethylation as a novel prognostic marker in
gastric cancer
TE Buffart1,4, RM Overmeer1,4, RDM Steenbergen1, M Tijssen1, NCT van Grieken1, PJF Snijders1, HI Grabsch2,
CJH van de Velde3, B Carvalho1and GA Meijer*,1
1Department of Pathology, VU University Medical Center, Amsterdam, The Netherlands;2Pathology and tumour biology, Leeds Institute of Molecular
Medicine, University of Leeds, Leeds, UK;3Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands
T-lymphocyte maturation associated protein, MAL, has been described as a tumour-suppressor gene with diagnostic value in
colorectal and oesophageal cancers, and can be inactivated by promoter hypermethylation. The aim of this study was to analyse the
prevalence of MAL promoter hypermethylation and the association with mRNA expression in gastric cancers and to correlate
methylation status to clinicopathological data. Bisulphite-treated DNA isolated from formalin-fixed and paraffin-embedded samples of
202 gastric adenocarcinomas and 22 normal gastric mucosae was subjected to real-time methylation-specific PCR (Q-MSP). Two
regions within the MAL promoter (M1 and M2) were analysed. In addition, 17 frozen gastric carcinomas and two gastric cancer cell
lines were analysed both by Q-MSP and real-time RT–PCR. Methylation of M1 and M2 occurred in 71 and 80% of the gastric
cancers, respectively, but not in normal gastric mucosa tissue. Hypermethylation of M2, but not M1, correlated with significantly
better disease-free survival in a univariate (P¼0.03) and multivariate analysis (P¼0.03) and with downregulation of expression
(P¼0.01). These results indicate that MAL has a putative tumour-suppressor gene function in gastric cancer, and detection of
promoter hypermethylation may be useful as a prognostic marker.
British Journal of Cancer (2008) 99, 1802–1807. doi:10.1038/sj.bjc.6604777
Published online 11 November 2008
& 2008 Cancer Research UK
www.bjcancer.com
Keywords: gastric cancer; promoter hypermethylation; prognostic marker; MAL
????????????????????????????????????????????????
Despite the overall decreasing rates of incidence and mortality, gastric
cancer remains the second most common cause of cancer death
worldwide (Parkin et al, 2005). The only possible curative treatment
is surgery, but clinical outcome largely depends on the stage of the
disease. Early detection of gastric cancer, before the tumour has
metastasised to the lymph nodes, can therefore contribute to reducing
deaths from gastric cancer. However, the knowledge on the molecular
pathogenesis of gastric cancer and availability of possible biomarkers
with clinical value is limited. Further insight in the molecular
pathogenesis of gastric cancer will aid the discovery of new markers
with high clinical relevance in gastric cancer, which are essential for
improving gastric cancer prognosis.
Gastric cancers, like many other solid tumours, are charac-
terised by the presence of genetic instability leading to the
disruption of many genes, either resulting in their activation
(oncogenes) or inactivation (tumour-suppressor genes). One of the
common mechanisms of inactivation of tumour-suppressor genes
is promoter hypermethylation (Baylin et al, 1998). Gene silencing
by promoter hypermethylation has been described in gastric
cancer for multiple genes, including hMLH1, involved in DNA
mismatch repair, CDH1, involved in cell adhesion and the cell cycle
regulator p16 (Fleisher et al, 1999; Esteller et al, 2001; Machado
et al, 2001; Carvalho et al, 2003). In addition, promoter
hypermethylation of Cox2 has been shown to be an independent
prognostic marker in gastric cancer (de Maat et al, 2007).
In other gastrointestinal cancers, that is, colorectal and
oesophageal cancer, the T-lymphocyte maturation associated
protein MAL, involved in glycolipid-enriched membrane-mediated
apical transport, has been described to be inactivated by promoter
hypermethylation (Puertollano and Alonso, 1999; Mimori et al,
2003; Kazemi-Noureini et al, 2004; Mori et al, 2006; Lind et al,
2007). Promoter hypermethylation of MAL was a frequent event in
these two cancer types, but infrequent in normal mucosa. As
promoter hypermethylation of MAL could already be detected in
cancer precursor lesions, it has been suggested as a tumour-
suppressor gene with diagnostic value (Lind et al, 2007; Mimori
et al, 2007). To the best of our knowledge, MAL promoter
hypermethylation has not yet been shown in gastric cancers. Aim
of this study was therefore to analyse promoter hypermethylation
of MAL in gastric cancers, its relation to gene silencing and to
determine its clinical value as a prognostic marker.
MATERIALS AND METHODS
Material
Two hundred and two formalin-fixed and paraffin-embedded
(FFPE) gastric adenocarcinoma samples, randomly selected from
the Leeds University archive and 22 normal gastric biopsy
specimens, randomly selected from the archives of the VU
University Medical Center, were included in this study. In
addition, 17 snap-frozen gastric cancer tissue samples, obtained
from the archives of the department of pathology of the VU
Received 3 September 2008; revised 10 October 2008; accepted 20
October 2008; published online 11 November 2008
*Correspondence: Professor Dr GA Meijer, Department of Pathology,
VU University Medical Center, PO Box 7057, Amsterdam 1007 MB, The
Netherlands; E-mail: ga.meijer@vumc.nl
4These authors have contributed equally to this work.
British Journal of Cancer (2008) 99, 1802–1807
& 2008 Cancer Research UKAll rights reserved 0007– 0920/08$32.00
www.bjcancer.com
Translational Therapeutics

Page 2

University Medical Center (Weiss et al, 2003) were included.
Patients did not receive chemotherapy, nor radiotherapy. More-
over, two gastric cancer cell lines, IPA220 and GP202 (Gartner
et al, 1996), kindly provided by Professor Dr R Seruca (IPATIMUP,
Porto, Portugal), the cervical cancer cell line SiHA, obtained from
the American Type Culture Collection (Manassas, VA, USA), and
primary human keratinocytes were included. The study was
approved by the Institutional Review Board and was in accordance
with medical and ethical guidelines in place in The Netherlands.
Cell culture
Cells were maintained in standard culturing conditions. IPA220
and GP202 were cultured in RPMI (Life Technologies, Breda, The
Netherlands) supplemented with 10% fetal calf serum, 100Uml?1
penicillin, 100mgml?1streptomycin and 2mmoll?1L-glutamine
(Life Technologies) (Gartner et al, 1996). The cervical cancer cell
line SiHa was maintained in DMEM (Life Technologies) supple-
mented with 10% FCS, 100Uml?1penicillin, 100mgml?1strepto-
mycinand 2mmoll?1
L-glutamine
(Steenbergen et al, 2004). Primary keratinocytes were cultured in
serum-free keratinocyte growth medium (Life Technologies)
supplemented with bovine pituitary extract (50mgml?1), epidermal
growth factor (5ngml?1), penicillin (100Uml?1), streptomycin
(100mgml?1) and L-glutamine (2mmoll?1) (Life Technologies)
(Steenbergen et al, 1996).
(Life Technologies)
DNA and RNA isolation procedures
DNA of the primary gastric tumour tissues was isolated as
described before using a commercially available DNA isolation kit
(QIAamp microkit; Qiagen, Hilden, Germany) (Weiss et al, 1999;
Buffart et al, 2007). Briefly, areas of at least 70% of tumour cells
were marked on a 4mm haematoxylin and eosin stained tissue
section. Tumour tissue was macro dissected from adjacent serial
10mm sections, using a needle. After deparaffinisation, an over-
night incubation at 371C with sodium-thiocyanate (1 M) and a
proteinase K treatment, DNA was extracted.
DNA and RNA of the 17 snap-frozen gastric carcinoma tissue
samples and gastric cancer cell lines, SiHa cervical cancer cell line
and primary keratinocytes was isolated using TRIzol reagent
(Invitrogen, Breda, The Netherlands) according to the instructions
of the manufacturer, with some modifications. Details are
described elsewhere (Weiss et al, 1999, 2003) (http://www.english.
vumc.nl/afdelingen/microarrays/). All DNA and RNA concentra-
tions and purities were measured on a Nanodrop ND-1000
spectrophotometer (Isogen, IJsselstein, The Netherlands).
Bisulphite treatment and real-time methylation specific
PCR
Of each DNA sample, 500ng was used for bisulphite treatment
using a commercially available DNA modification kit (EZ DNA
Methylation Kitt; Zymo Research, Orange, CA, USA).
Real-time methylation specific PCR (Q-MSP) was performed
using primer sets targeting two regions within the MAL promoter
(i.e. from ?680 to ?573bp (M1) and ?92 to ?7bp (M2) relative to
the first ATG). Both regions within the MAL promoter were
selected within the CpG island of the MAL promoter. Amplicons
were detected and quantified using Taqman probes. The house-
keeping gene b-actin was chosen as a reference for total DNA input
measurement.
Q-MSP reactions were carried out in a total reaction volume of
12ml containing 50ng bisulphite treated genomic DNA, 417nM of
each primer, 208nM probe and 2? QuantiTect Probe PCR Kit
master mix (Qiagen, Westburg, Leusden, The Netherlands). For
both MAL M1 and M2 regions, the PCR reaction was performed for
45 cycles (15s at 951C and 60s at 591C) with an initial denaturation
of 15min at 951C. For each Q-MSP a standard curve of serial
dilutions of bisulphite-treated DNA (50, 5, 2.5, 0.5 and 0.25ng) of
the SiHa cervical cancer cell line was used. All samples were
analysed in duplicate. As a negative control, multiple water
samples, unmodified genomic DNA obtained from SiHa cells and
unmethylated DNA obtained from primary keratinocytes were
included. To determine the relative quantity of methylation, we
calculated the ratios between MAL M1 and MAL M2 methylated
DNA vs b-actin DNA (average quantity of methylated MAL DNA/
average DNA quantity for b-actin?1000).
Real-time RT–PCR
Of each RNA sample, 1mg was reverse transcribed to cDNA using
oligo(dT)20Primer (Invitrogen) with AMV reverse transcriptase
(Promega, Leiden, The Netherlands). RT–PCR was performed in a
total reaction volume of 25ml, containing 22.5ml master mix and
2.5ml cDNA (25ng). The master mix contained 12.5ml of SYBR
Green PCR master mix (Applied Biosystems, Nieuwerkerk a/d
IJssel, The Netherlands) and 0.5mM of each primer. All samples
were analysed in duplicate in a 7300 Real-time PCR System
(Applied Biosystems). Amplification was performed in 50 cycles of
951C for 15s and an annealing temperature of 601C for 1min, with
an initial denaturation step of 5min at 951C. Relative expression
levels were determined from the obtained Ct values and the 2DDCt
method, using snRNP U1A as reference (Livak and Schmittgen,
2001), and transformed into a ln scale. Primary keratinocytes and
the SiHa cervical cancer cell line were used as positive and negative
controls respectively. Primer sequences are described earlier
(Wisman et al, 2006; de Wilde et al, 2008; Wilting et al, 2008).
Statistical analysis
Receiver operator characteristic (ROC) curves were analysed for
assessing the best cutoff value for methylation, on all FFPE
samples, assuming that normal gastric mucosae are unmethylated
and gastric carcinomas are methylated. Cutoff points were chosen
based on the point on the ROC curve showing 100% specificity.
Positivity for each methylated promoter region was consi-
dered when a specific sample had a ratio of M1/b-actin?1000
or M2/b-actin?1000 above the respective cutoff value. A sign test
was used for testing significance of differences in frequencies of
M1 vs M2 methylation.
Mann–Whitney U-test was used to determine significance of
differences in expression values between methylated and un-
methylated gastric carcinomas. Survival analysis was performed
using the Kaplan–Meier method, using the survival length starting
from the day of surgery of the primary tumour to the date of death
due to gastric cancer (event) or to the last day of clinical follow-up
(censored). Differences in survival length were analysed using the
log-rank test. Multivariate analysis was performed using a Cox’s
proportional hazard regression model in a forward stepwise
method for variable selection. Gender, histological type, tumour
stage (T-stage) and lymph node stage (N-stage) were entered into
the analysis. w2test was used for calculating differences in
methylation status and tissue type, gender of the patient,
histological type of the tumour and tumour stage. t-Test was used
to evaluate age related differences in methylation status (SPSS 14.0
for Windows, Chicago, IL, USA). P-values below 0.05 were
considered to be significant.
RESULTS
Frequent MAL promoter methylation in gastric
carcinomas
The chosen cutoff values of methylation for the M1 and M2
promoter regions, based on the ROC curve analysis, were ratios of
MAL promoter hypermethylation in gastric cancer
TE Buffart et al
1803
British Journal of Cancer (2008) 99(11), 1802–1807
& 2008 Cancer Research UK
Translational Therapeutics

Page 3

relative methylated DNA quantities of M1/b-actin?1000 and
M2/b-actin?1000 above 95 and 22, respectively, yielding a
specificity of 100% for both promoter regions and a sensitivity
of 71 and 80% for M1 and M2 promoter regions, respectively. ROC
curves for both M1 and M2 promoter regions are shown in
Figure 1.
Both gastric cancer cell lines, IPA220 and GP202, showed
methylation of both M1 and M2 promoter regions. Of all 202
gastric adenocarcinomas tested, 143 carcinomas (70.8%) showed
methylation of the M1 promoter region and 162 carcinomas
(80.2%) showed methylation of the M2 promoter region.
Methylation prevalences of M1 and M2 promoter regions were
significantly different (P¼0.004). Dense methylation, that is,
methylation of both M1 and M2 promoter regions, was detected
in 133 (65.8%) of the carcinomas. Thirty carcinomas (14.9%)
were unmethylated. All normal
were unmethylated for both regions. w2test yielded a signi-
ficant difference between methylation status of gastric carci-
nomas and normal gastric mucosa tissues (Po0.001). An overview
of the methylation status for both promoter regions is given in
Table 1.
gastricmucosa samples
Correlation of MAL promoter methylation and survival
Follow-up data was available for 200 out of 202 patients. Only
patients without distant metastasis at the time of surgery (M0)
were included, leaving 179 patients for the survival analysis.
Patients with a gastric carcinoma methylated for the M2 promoter
region had a significantly better survival compared with patients
with tumours unmethylated for the M2 promoter region (P¼0.03)
(Figure 2). No significant correlation was found between M2
promoter methylation and age or gender of the patient,
histological type of the tumour, T-stage and N-stage (Table 2).
In addition, no significant correlation between M1 promoter
methylation status and clinicopathological characteristics, includ-
ing survival, was found.
Multivariate analysis revealed only N-stage, T-stage and MAL
M2 promoter methylation status, in order of significance, to have
independent prognostic value (Table 2).
MAL promoter methylation is associated with reduced
gene expression
Reduced expression of MAL relative to the housekeeping gene
snRNP U1A was observed in both gastric cancer cell lines
compared with primary keratinocytes that did not show MAL
promoter hypermethylation. Of the 17 gastric cancer tissue
samples tested for MAL mRNA expression, 11 (64.7%) were
methylated for M1, 11 (64.7%) for M2 and 9 (52.9%) samples
showed methylation of both regions. Gastric carcinomas with M2
promoter methylation showed significantly lower expression of the
MAL gene compared with M2 unmethylated gastric cancers
(P¼0.01) (Figure 3, Table 3). For the M1 region, no significant
differences in MAL mRNA expression were found between
methylated and unmethylated tumours.
DISCUSSION
Gastric cancer is a common disease with generally a poor
prognosis (Parkin et al, 2005). Biomarkers can be used to predict
prognosis and optimise therapeutic strategies. Hypermethylation
of the MAL promoter has been shown in colorectal and
oesophageal cancers and MAL has been proposed as a putative
tumour suppressor in these cancer types (Mimori et al, 2003;
Kazemi-Noureini et al, 2004; Mori et al, 2006). Moreover, it has
been proposed as a candidate marker for early detection of these
cancers, as methylation of MAL could already be detected in
precursor lesions (Lind et al, 2007, 2008; Mimori et al, 2007). In
this study we show that MAL might have a similar role in gastric
cancers, as methylation of MAL is detected at high frequency in
gastric cancers and not in normal gastric mucosa samples.
In this study, two regions within the MAL promoter, M1 and
M2, were analysed. Q-MSP analysis showed methylation of both
promoter regions in the two gastric cancer cell lines analysed,
indicating that methylation actually occurs in gastric epithelial
1.0 0.8 0.60.4
1–Specificity
0.20.0
1–Specificity
0.0
0.2
0.4
0.6
0.8
1.0
Sensitivity
Sensitivity
M2
1.0 0.8 0.60.4 0.20.0
0.0
0.2
0.4
0.6
0.8
1.0
M1
Figure 1
methylation and (B) M2 promoter methylation in 202 gastric cancers
and 22 normal gastric mucosa samples (FFPE samples only), assuming that
normal gastric mucosae are unmethylated and gastric carcinomas are
methylated. Chosen cutoff levels for methylation were 95 and 22 for M1
and M2 promoter regions, respectively. This yielded a specificity of 100%
for both promoter regions and a sensitivity of 71% for M1 promoter region
and 80% for the M2 promoter region.
Receiver operator characteristics of (A) M1 promoter
Table 1
M2 (?92 to ?7bp) regions within the MAL promoter for 202 gastric
carcinoma tissues and 22 normal gastric mucosa tissues
Overview of methylation status of M1 (?680 to ?573bp) and
Carcinomas
n¼202
30 (14.9%)
143 (70.8%)
162 (80.2%)
133 (65.8%)
Normal mucosa n¼22
22 (100%)
0 (0%)
0 (0%)
0 (0%)
P-value
No methylation
M1 methylation
M2 methylation
dense methylation
o0.001
o0.001
o0.001
o0.001
w2test yielded a significant difference between methylation status of gastric
carcinomas and normal gastric mucosa tissues (Po0.001).
MAL promoter hypermethylation in gastric cancer
TE Buffart et al
1804
British Journal of Cancer (2008) 99(11), 1802–1807
& 2008 Cancer Research UK
Translational Therapeutics

Page 4

Table 2
tumours used in the univariate and multivariate survival analysis
Overview of patient and tumour characteristics of the 179
Univariate analysis
Methylated UnmethylatedP-value
Age (years) 72 (52–96)71 (54–87)NS
Gender
Male
Female
88 (62%)
55 (38%)
20 (56%)
16 (44%)
NS
Histological type
Intestinal
Diffuse
Mixed
100 (70%)
18 (13%)
25 (17%)
23 (64%)
7 (19%)
6 (17%)
NS
T-stage
T1
T2
T3
T4
10 (7%)
50 (35%)
78 (55%)
5 (3%)
2 (6%)
16 (44%)
18 (50%)
NS
—
N-stage
N0
N1
N2
N3
Unknown
41 (29%)
62 (43%)
32 (22%)
7 (5%)
1 (1%)
9 (25%)
15 (42%)
8 (22%)
4 (11%)
NS
—
Multivariate analysis
HR 95% CIP-value
Gender
Histological type
T-stage
N-stage
MAL M2 methylation
0.94
0.48
0.001
0.001
0.03
1.73
1.52
0.59
1.23–2.42
1.20–1.91
0.36–0.96
CI¼confidence interval; HR¼hazard ratio. Absolute number and percentages are
given for gender, histological type, tumour stage (T-stage) and lymph node status
(N-stage). Age is given as mean age and range. None of the clinicopathological
characteristics were significantly correlated with M2 promoter methylation status
(P¼NS). Multivariate analysis showed that T-stage, N-stage and M2 methylation
status are prognostic variables for patient outcome.
Table 3
status for the M1 and M2 regions within the MAL promoter of the two
gastric cancer cell lines and 17 gastric carcinoma tissues
Relative ln transformed expression values (E) and methylation
SampleE M1 M2
IPA220
GP202
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
0.003
0.0003
0.35
0.24
0.06
0.72
0.10
0.05
0.77
0.28
0.19
2.28
0.28
0.03
0.03
0.04
5.56
0.01
0.33
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Unmethylated
Unmethylated
Unmethylated
Unmethylated
Unmethylated
Methylated
Methylated
Unmethylated
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Methylated
Unmethylated
Unmethylated
Unmethylated
Unmethylated
Methylated
Methylated
Methylated
Methylated
Unmethylated
Methylated
Unmethylated
12 10864
Survival (years)
20
0.0
0.2
0.4
0.6
0.8
1.0
Cum survival
P = 0.03
Log rank = 4.96
Methylation
No
Methylation
Censored
Censored
n = 70 (143)
n = 22 (36)
Figure 2
gastric cancers assessed for the methylation status of the M2 region (?92
to ?7bp) within the MAL promoter. Patients with primary gastric cancers
methylated for the M2 promoter region (n¼143) showed a significantly
better survival compared to patients (n¼36) with gastric cancers without
M2 promoter methylation (P¼0.03; log rank¼4.96). The number of
patients who died of gastric cancer (events) is 70 and 22, respectively.
Kaplan–Meier survival analysis of 179 patients with primary
Methylation No methylation
0
1
2
3
4
5
Expression
6
P = 0.01
M2
Methylation No methylation
0
1
2
3
4
5
6
Expression
P = 0.27
M1
Figure 3
carcinoma tissue samples methylated and unmethylated for the M1 (A) and
M2 (B) promoter regions. Gastric carcinomas methylated for the M2 MAL
promoter region show significantly lower expression of the MAL gene
compared with unmethylated gastric carcinomas (P¼0.01).
Box plots of the relative expression values of 17 gastric
MAL promoter hypermethylation in gastric cancer
TE Buffart et al
1805
British Journal of Cancer (2008) 99(11), 1802–1807
& 2008 Cancer Research UK
Translational Therapeutics

Page 5

cells. Earlier studies showed protein expression of MAL by
immunohistochemistry in normal gastric mucosa. Strong expres-
sion of MAL was observed in parietal and chief cells, but not in
muscle and submucosa cells (Marazuela et al, 2003). Gel-based
MSP analysis revealed a small subpopulation of unmethylated cells
for both the M1 and M2 promoter regions in these cell lines (data
not shown). Expression of MAL was hardly detected in both cell
lines indicating that methylation is probably the main mechanism
of downregulation of this gene. In gastric cancers, methylation of
the M2 region was more frequently observed compared with the
M1 region (80.2 vs 70.8%). In consistence with what has been
described by Lind et al (Lind et al, 2008), an unequal distribution
of DNA methylation within the MAL promoter was observed in
gastric cancers, with the highest frequencies of methylation in the
region closest to the transcription start site. M2 region is located
around the transcription start site.
Results of this study show that MAL may serve as a prognostic
marker in gastric cancer, as patients with tumours methylated for
the M2 region show significantly better survival compared with
patients with tumours unmethylated for MAL or methylated only
for the M1 region. This survival benefit was independent of other
clinicopathological data such as age, gender, histological type of
the tumour, tumour stage and lymph node status. Interestingly,
also in Hodgkin’s lymphoma patients a similar association was
found with a significantly worse survival in patients whose
tumours expressed the MAL protein compared with patients with
tumours lacking MAL expression (Hsi et al, 2006), indicating a
prognostic value of MAL also in other cancer models.
The finding that inactivation of MAL by methylation gives a
better prognosis for the patients may seem contradicting with a
putative tumour suppressor function of this gene in gastric cancer.
However, this finding has been observed earlier for the mismatch
repair gene hMLH1, which also has a tumour suppressor function.
Inactivation of this gene leads to microsatellite instability of the
tumour and patients with microsatellite instable tumours have a
better prognosis compared with patients with microsatellite stable
tumours (Ribic et al, 2003; Parc et al, 2004; Beghelli et al, 2006).
Therefore, tumours without MAL methylation might have a
different biology overall, which could relate to poorer clinical
outcome, rather than the outcome being dependent on MAL itself.
However, this study was performed retrospectively on archival
material, and therefore, rather should be considered as hypothesis
generating. Actual clinical implementation of MAL promoter
hypermethylation as a diagnostic marker requires further valida-
tion in a prospective study.
To test the biological relevance of MAL promoter hypermethy-
lation, in a subset of gastric cancers the association between MAL
promoter hypermethylation and mRNA expression was analysed.
Results showed lower expression of MAL in gastric cancers
methylated for the M1 or M2 region. The association between
promoter methylation and reduced mRNA expression strengthens
the biological relevance of MAL methylation and supports a
putative role as tumour-suppressor gene. However, correlation
between reduced expression and methylation of MAL was only
significant for the M2 region, indicating that M2 methylation
would have more biological relevance. Two out of 17 gastric
cancers showed reduced MAL mRNA expression whereas the gene
was unmethylated for both promoter regions. This indicates that
other regulatory mechanisms of MAL silencing also exist in gastric
cancer, which may include DNA copy number loss, other
epigenetic mechanisms, altered expression of transcription factors
regulating MAL or microRNAs targeting the MAL gene.
In summary, this study shows frequent promoter hypermethyla-
tion of MAL in gastric cancers, and not in normal gastric mucosa
samples. Promoter hypermethylation of MAL is associated with
downregulation of its expression, especially when methylation
occurs at the M2 region within the promoter. Methylation of this
region within the MAL promoter correlates with a significantly
better survival of the patients. Altogether, these results pinpoint
MAL as a putative tumour-suppressor gene with a role in gastric
cancer, which may serve as an independent prognostic marker for
clinical outcome of gastric cancer patients.
ACKNOWLEDGEMENTS
This study was supported by Dutch Cancer Society, grants KWF
2004-3051 and KWF 2005-3276.
REFERENCES
Baylin SB, Herman JG, Graff JR, Vertino PM, Issa JP (1998) Alterations in
DNA methylation: a fundamental aspect of neoplasia. Adv Cancer Res 72:
141–196
Beghelli S, de MG, Barbi S, Tomezzoli A, Roviello F, Di GC, Vindigni C,
Bortesi L, Parisi A, Saragoni L, Scarpa A, Moore PS (2006) Microsatellite
instability in gastric cancer is associated with better prognosis in only
stage II cancers. Surgery 139: 347–356
Buffart T, Carvalho B, Hopmans E, Brehm V, Klein-Kranenbarg E, Schaaij-
Visser T, Eijk P, van Grieken N, Ylstra B, van de Velde CJ, Meijer G
(2007) Gastric cancers in young and elderly patients show different
genomic profiles. J Pathol 211: 45–51
Carvalho B, Pinto M, Cirnes L, Oliveira C, Machado JC, Suriano G, Hamelin
R, Carneiro F, Seruca R (2003) Concurrent hypermethylation of gene
promoters is associated with a MSI-H phenotype and diploidy in gastric
carcinomas. Eur J Cancer 39: 1222–1227
de Maat MF, van de Velde CJ, Umetani N, de HP, Putter H,
van Hoesel AQ, Meijer GA, van Grieken NC, Kuppen PJ, Bilchik AJ,
Tollenaar RA, Hoon DS (2007) Epigenetic silencing of cycloo-
xygenase-2 affects clinical outcome in gastric cancer. J Clin Oncol 25:
4887–4894
de Wilde J, De-Castro AJ, Snijders PJ, Meijer CJ, Rosl F, Steenbergen RD
(2008) Alterations in AP-1 and AP-1 regulatory genes during HPV-
induced carcinogenesis. Cell Oncol 30: 77–87
Esteller M, Corn PG, Baylin SB, Herman JG (2001) A gene hypermethylation
profile of human cancer. Cancer Res 61: 3225–3229
Fleisher AS, Esteller M, Wang S, Tamura G, Suzuki H, Yin J, Zou TT,
Abraham JM, Kong D, Smolinski KN, Shi YQ, Rhyu MG, Powell SM,
James SP, Wilson KT, Herman JG, Meltzer SJ (1999) Hypermethylation of
the hMLH1 gene promoter in human gastric cancers with microsatellite
instability. Cancer Res 59: 1090–1095
Gartner F, David L, Seruca R, Machado JC, Sobrinho-Simoes M (1996)
Establishment and characterization of two cell lines derived from human
diffuse gastric carcinomas xenografted in nude mice. Virchows Arch 428:
91–98
Hsi ED, Sup SJ, Alemany C, Tso E, Skacel M, Elson P, Alonso MA, Pohlman
B (2006) MAL is expressed in a subset of Hodgkin lymphoma and
identifies a population of patients with poor prognosis. Am J Clin Pathol
125: 776–782
Kazemi-Noureini S, Colonna-Romano S, Ziaee AA, Malboobi MA,
Yazdanbod M, Setayeshgar P, Maresca B (2004) Differential gene
expression between squamous cell carcinoma of esophageus and its
normal epithelium; altered pattern of mal, akr1c2, and rab11a
expression. World J Gastroenterol 10: 1716–1721
Lind GE, Ahlquist T, Kolberg M, Berg M, Eknaes M, Alonso MA,
Kallioniemi A, Meling GI, Skotheim RI, Rognum TO, Thiis-Evensen E,
Lothe RA (2008) Hypermethylated MAL gene – a silent marker of early
colon tumorigenesis. J Transl Med 6: 13
Lind GE, Ahlquist T, Lothe RA (2007) DNA hypermethylation of MAL: a
promising diagnostic biomarker for colorectal tumors. Gastroenterology
132: 1631–1632
MAL promoter hypermethylation in gastric cancer
TE Buffart et al
1806
British Journal of Cancer (2008) 99(11), 1802–1807
& 2008 Cancer Research UK
Translational Therapeutics

Page 6

Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data
using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method.
Methods 25: 402–408
Machado JC, Oliveira C, Carvalho R, Soares P, Berx G, Caldas C, Seruca R,
Carneiro F, Sobrinho-Simoes M (2001) E-cadherin gene (CDH1)
promoter methylation as the second hit in sporadic diffuse gastric
carcinoma. Oncogene 20: 1525–1528
Marazuela M, Acevedo A, Adrados M, Garcia-Lopez MA, Alonso MA (2003)
Expression of MAL, an integral protein component of the machinery for
raft-mediated pical transport, in human epithelia. J Histochem Cytochem
51: 665–674
Mimori K, Nishida K, Nakamura Y, Ieta K, Yoshikawa Y, Sasaki A, Ishii H,
Alonso MA, Mori M (2007) Loss of MAL expression in precancerous
lesions of the esophagus. Ann Surg Oncol 14: 1670–1677
Mimori K, Shiraishi T, Mashino K, Sonoda H, Yamashita K, Yoshinaga K,
Masuda T, Utsunomiya T, Alonso MA, Inoue H, Mori M (2003) MAL
gene expression in esophageal cancer suppresses motility, invasion and
tumorigenicity and enhances apoptosis through the Fas pathway.
Oncogene 22: 3463–3471
Mori Y, Cai K, Cheng Y, Wang S, Paun B, Hamilton JP, Jin Z, Sato F,
Berki AT, Kan T, Ito T, Mantzur C, Abraham JM, Meltzer SJ (2006) A
genome-wide search identifies epigenetic silencing of somatostatin, tachy-
kinin-1, and 5 other genes in colon cancer. Gastroenterology 131: 797–808
Parc Y, Gueroult S, Mourra N, Serfaty L, Flejou JF, Tiret E, Parc R (2004)
Prognostic significance of microsatellite instability determined by
immunohistochemical staining of MSH2 and MLH1 in sporadic
T3N0M0 colon cancer. Gut 53: 371–375
Parkin DM, Bray F, Ferlay J, Pisani P (2005) Global cancer statistics, 2002.
CA Cancer J Clin 55: 74–108
Puertollano R, Alonso MA (1999) MAL, an integral element of the apical
sorting machinery, is an itinerant protein that cycles between the trans-
Golgi network and the plasma membrane. Mol Biol Cell 10:
3435–3447
Ribic CM, Sargent DJ, Moore MJ, Thibodeau SN, French AJ, Goldberg RM,
Hamilton SR, Laurent-Puig P, Gryfe R, Shepherd LE, Tu D, Redston M,
Gallinger S (2003) Tumor microsatellite-instability status as a predictor
of benefit from fluorouracil-based adjuvant chemotherapy for colon
cancer. N Engl J Med 349: 247–257
Steenbergen RD, Kramer D, Braakhuis BJ, Stern PL, Verheijen RH, Meijer
CJ, Snijders PJ (2004) TSLC1 gene silencing in cervical cancer cell lines
and cervical neoplasia. J Natl Cancer Inst 96: 294–305
Steenbergen RD, Walboomers JM, Meijer CJ, van der Raaij-Helmer EM,
Parker JN, Chow LT, Broker TR, Snijders PJ (1996) Transition of human
papillomavirus type 16 and 18 transfected human foreskin keratinocytes
towards immortality: activation of telomerase and allele losses at 3p, 10p,
11q and/or 18q. Oncogene 13: 1249–1257
Weiss MM, Hermsen MA, Meijer GA, van Grieken NC, Baak JP, Kuipers EJ,
van Diest PJ (1999) Comparative genomic hybridisation. Mol Pathol 52:
243–251
Weiss MM, Kuipers EJ, Postma C, Snijders AM, Siccama I, Pinkel D,
Westerga J, Meuwissen SG, Albertson DG, Meijer GA (2003) Genomic
profiling of gastric cancer predicts lymph node status and survival.
Oncogene 22: 1872–1879
Wilting SM, de WJ, Meijer CJ, Berkhof J, Yi Y, van Wieringen WN,
Braakhuis BJ, Meijer GA, Ylstra B, Snijders PJ, Steenbergen RD (2008)
Integrated genomic and transcriptional profiling identifies chromosomal
loci with altered gene expression in cervical cancer. Genes Chromosomes
Cancer 47(10): 890–905
Wisman GB, Nijhuis ER, Hoque MO, Reesink-Peters N, Koning AJ, Volders
HH, Buikema HJ, Boezen HM, Hollema H, Schuuring E, Sidransky D, van
der Zee AG (2006) Assessment of gene promoter hypermethylation for
detection of cervical neoplasia. Int J Cancer 119: 1908–1914
MAL promoter hypermethylation in gastric cancer
TE Buffart et al
1807
British Journal of Cancer (2008) 99(11), 1802–1807
& 2008 Cancer Research UK
Translational Therapeutics