DNA methylation profiling reveals novel biomarkers
and important roles for DNA methyltransferases
in prostate cancer
Yuya Kobayashi,1Devin M. Absher,2Zulfiqar G. Gulzar,3Sarah R. Young,3
Jesse K. McKenney,4Donna M. Peehl,3James D. Brooks,3Richard M. Myers,2
and Gavin Sherlock1,5
1Department of Genetics, Stanford University, Stanford, California 94305, USA;2HudsonAlpha Institute for Biotechnology, Huntsville,
Alabama 35806, USA;3Department of Urology, Stanford University, Stanford, California 94305, USA;4Department of Pathology,
Stanford University, Stanford, California 94305, USA
Candidate gene-based studies have identified a handful of aberrant CpG DNA methylation events in prostate cancer.
However, DNA methylation profiles have not been compared on a large scale between prostate tumor and normal
prostate, and the mechanisms behind these alterations are unknown. In this study, we quantitatively profiled 95 primary
prostate tumors and 86 benign adjacent prostate tissue samples for their DNA methylation levels at 26,333 CpGs rep-
resenting 14,104 gene promoters by using the Illumina HumanMethylation27 platform. A 2-class Significance Analysis of
this data set revealed 5912 CpG sites with increased DNA methylation and 2151 CpG sites with decreased DNA methylation
in tumors (FDR < 0.8%). Prediction Analysis of this data set identified 87 CpGs that are the most predictive diagnostic
methylation biomarkers of prostate cancer. By integrating available clinical follow-up data, we also identified 69 prog-
nostic DNA methylation alterations that correlate with biochemical recurrence of the tumor. To identify the mechanisms
responsible for these genome-wide DNA methylation alterations, we measured the gene expression levels of several DNA
methyltransferases (DNMTs) and their interacting proteins by TaqMan qPCR and observed increased expression of
DNMT3A2, DNMT3B, and EZH2 in tumors. Subsequent transient transfection assays in cultured primary prostate cells
revealed that DNMT3B1 and DNMT3B2 overexpression resulted in increased methylation of a substantial subset of CpG sites
that showed tumor-specific increased methylation.
[Supplemental material is available for this article. The microarray data from this study have been submitted to the NCBI
Gene Expression Omnibus (GEO) (http:/ /www.ncbi.nlm.nih.gov/geo) under accession no. GSE26126.]
Prostate cancer is the most commonly diagnosed malignancy for
men in the United States, with an estimated 217,730 new cases
projected for 2010 ( Jemal et al. 2010). After more than two decades
of widespread serum prostate-specific antigen (PSA) testing, clinical
prostate cancer has shifted to a predominantly localized disease.
many cancers that are detected are never destined to progress
(Andriole et al. 2009; Schro ¨der et al. 2009). However, prostate
cancer can have an aggressive and lethal course, and an estimated
of the underlying genomic diversity of the tumors (Taylor et al.
2010). Previous studies of prostate tumors reported significant het-
erogeneity in the gene expression profiles and genomic structural
alterations including DNA copy number changes and gene fusions
often involving the ETS family of transcription factors detectable in
approximately half of prostate tumors (Singh et al. 2002; Lapointe
et al. 2004; Tomlins et al. 2005, 2008; King et al. 2009; Sboner et al.
2010; Taylor et al. 2010; Pflueger et al. 2011; Robbins et al. 2011).
However, exon sequencing of known oncogenes and tumor
suppressors has found few somatic mutations, and the calculated
background mutation rate appears to be relatively low (Taylor et al.
2010). This suggests the presence of other forms of genomic aber-
rations that contribute to the observed gene expression variations,
and, in turn, the diversity in tumor behavior.
DNA methylation has long been suspected to play a role in
et al. 1981; Jones 1986; Laird and Jaenisch 1994, 1996; Ehrlich 2002;
Esteller and Herman 2002; Patra et al. 2002; Das and Singal 2004).
Early studies in cancer epigenetics revealed an overall reduction
of 5-methylcytosine in various tumor genomes (Feinberg and
Vogelstein 1983; Gama-Sosa et al. 1983). In contrast, more recent
studies identified many hypermethylation events in CpG islands
near known tumor-suppressor transcriptional start sites, which
correlatedwith reduction intranscriptlevels(Lee etal. 1994; Brooks
et al. 1998). Many of these candidate gene-based approaches have
led to discovery of potentially prognostic DNA methylation events
(Mu ¨ller et al. 2003; Kim et al. 2008). However, recent advances in
microarray and high-throughput massively parallel sequencing
technologies have enabled investigators to study site-specific DNA
methylation events on a much broader scale. Recent studies of the
DNA methylome in colorectal cancer and glioblastomas have re-
vealed valuable new insights into those diseases, including the
discovery of hundreds of affected genes previously not identified
(Cancer Genome Atlas Research Network 2008; Irizarry et al. 2009;
Noushmehr et al. 2010).
Article published online before print. Article, supplemental material, and pub-
lication date are at http://www.genome.org/cgi/doi/10.1101/gr.119487.110.
21:1017–1027 ? 2011 by Cold Spring Harbor Laboratory Press; ISSN 1088-9051/11; www.genome.org
In prostate cancer, hypermethylation of CpG islands within
et al. 1994; Brooks et al. 1998; Jero ´nimo et al. 2004). In addition,
CpG island microarray. However, this study did not determine the
profiles of normal prostate tissues and was thus limited to compar-
isons between the prostate tumors and six cases of age-matched
lymphocytes. While specific sites of methylation level heterogene-
ity amongtumor samples were identified (Kron etal. 2010; Liu etal.
2011), the study design precluded the examination of changes in
methylation between normal prostate and prostate tumors. In ad-
dition to these studies of prostate cancer methylation profiles, a few
studies looking at DNA methyltransferases (DNMTs) and DNMT-
interacting proteins have suggested that misregulation of these
genes in prostate cancer is responsible for the improper DNA
methylation events in primary tumors and cell lines (Hoffmann
et al. 2007; Yaqinuddin et al. 2008; Ley et al. 2010).
Here, we quantitatively profiled 95 primary prostate tumors
and 86 benign adjacent prostate tissues for their DNA methylation
levels at 26,333 CpG sites in 14,104 gene promoters. Based on the
results, we identified the differentially methylated CpGs and ex-
plored subsets of them that accurately distinguished tumor and be-
nign adjacent prostate tissues. We then integrated available clinical
data to discover novel prognostic markers of aggressive tumors. Fi-
nally, we investigated the DNMT protein family, as well as their in-
in prostate cancer.
To explore the prostate DNA methylome, we profiled 95 primary
prostate tumors and 86 benign adjacent prostate tissues, including
70 matched pairs, using the Illumina HumanMethylation27 micro-
arrays. These tissue samples were harvested from men who under-
went radical retropubic prostatectomy for clinically localized pros-
tate cancer. Surgeries were performed between 1998 and 2007, and
detailed clinical data, including follow-up and recurrence status,
were available in 96 patients (88%). Mean patient age, pre-operative
serum PSA levels, clinical stage, and pathological Gleason grade
were compatible with the risk profiles of contemporary patients
undergoing surgery for prostate cancer (Supplemental Table S1;
Brooks et al. 2008).
The Illumina HumanMethylation27 platform assays 27,578
CpG sites, almost all in the proximal promoter regions of 14,495
transcription start sites (Weisenberger et al. 2008; Hernandez et al.
2011). After batch-correcting and quality-filtering the data, we were
able to determine quantitative methylation status (beta scores;
range: 0 to 1) for 26,333 CpG sites in 14,104 promoters. To in-
vestigate the similarities and differences of the DNA methylation
profiles of the benign adjacent samples and tumor samples, as well
as their heterogeneity, we performed unsupervised hierarchical
clustering on the entire data set (Fig. 1). When the data were clus-
tered by sample, we observed two main clusters—one composed
almost entirely of benign adjacent samples (77/88) and the other
the benign adjacent sample cluster were generally shorter than the
branch lengths in the tumor sample cluster, indicating more het-
erogeneity in methylation profiles among the tumor samples.
Twenty-two of the samples did not fall into either of the two main
clusters and formed long off-shooting branches or small clusters.
Eighteen of these were tumor samples, further indicative of the
heterogeneous nature of the tumor DNA methylome. By visual
inspection, the majority of the samples showed relatively little
methylation change between the tumor and benign adjacent
clusters (Fig. 1), and most of these invariable CpG sites showed
low levels of methylation in both benign adjacent and tumor
samples. However, there were distinct CpG clusters with meth-
ylation patterns that distinguished the benign adjacent or tumor
sample clusters, and, strikingly, a large number of CpG sites
showed increased methylation in the tumor cluster compared to
the benign adjacent cluster.
To identify the CpG sites with statistically different DNA
methylation status between benign adjacent prostate tissues and
tumors, we performed a two-class Significance Analysis of Micro-
arrays (SAM) (Tusher et al. 2001). As we had matched benign ad-
jacent tissues for only 70 of the 95 tumors used in this study, we
conducted the SAM analysis as unpaired. The analysis identified
5912 CpG sites hypermethylated in tumors compared to benign
adjacent tissues and 2151 CpG sites hypomethylated at FDR < 0.8%
(Supplemental Fig. S1; Supplemental Table S2). We performed hier-
archical clustering on all samples based on these 8063 differentially
tion. Unsupervised hierarchical clustering of 181 prostate tissues and
26,333 CpGs, by sample and by CpG. (Red branches) Tumor samples;
(blue branches) benign adjacent samples; (red pixels) high DNA meth-
ylation; (green pixels) low DNA methylation.
Hierarchical clustering of prostate tissues by DNA methyla-
1018 Genome Research
Kobayashi et al.
methylated CpG sites (Fig. 2). When the fold-change was examined
as a 141-fold increase in methylation for a CpG near the transcrip-
tional startsite ofZNF296 (averagenormal beta:5.30310?4, average
tumor beta: 0.0756). CpG island hypermethylation of ZNF296 has
been observed and implicated in its transcriptional silencing in oli-
godendroglioma (Hong et al. 2003; Noushmehr et al. 2010). In ad-
dition, ZNF296 has been reported to be drastically overexpressed in
acute myeloid leukemia (Poland et al. 2009). This suggests that the
aberrant gene expression of ZNF296, as a result of DNA methylation
or otherwise, is a common event in tumorigenesis or tumor pro-
gression. All but 609 of the CpGs had a change of 5% or greater.
While these 609 sites had a low level of fold-change, these were
nonetheless identified as statistically significant changes that were
detectable because of the large sample size (Supplemental Fig. S1).
The 8063 differentially methylated sites corresponded to
4227 promoters with at least one hypermethylated CpG and 1795
promoters with at least one hypomethylated CpG. Of the 11,116
gene promoters represented by two or more CpG sites on the
HumanMethylation27 platform, only 223 had opposite methyla-
from transcriptional start sites were compared in these 223 pro-
moters with opposite methylation effects, we saw enrichment for
hypermethylated CpGs in the ?100-bp to +800-bp range, whereas
we sawenrichmentfor the hypomethylated CpGs in the ?700-bp
to ?200-bp range. Thus, overall, nearly one-third (8063/26,333) of
assayed promoter CpGs had a statistically significant change in
DNA methylation, with most of those showing an increase in
assayed had at least one CpG with a tumor-specific methylation
change. We repeated this analysis using two-class paired SAM
on only the 70 matched sample pairs and observed similar results
(Supplemental Text S1).
Diagnostic methylation markers
in tumor versus benign adjacent prostate tissues by SAM, and
shown clustered in Figure 2, were several sites that had been pre-
viously characterized in prostate tumors, most notably several
CpG island overlapping the transcriptional start site of the GSTP1
gene has been associated with transcriptional silencing and is de-
scribedasthemost commonmolecularalterationinprostate cancer
promoter methylation is very common and specific for prostate
cancer, many investigators have proposed using this methylation
Nakayama et al. 2004). The HumanMethylation27 arrays contain
seven CpG sites in the GSTP1 promoter. Five of these sites showed
located in the promoter CpG island that had been previously char-
acterized as a site of hypermethylation in prostate cancer (Brooks
CpG island boundary (red circles in Fig. 3A). The two remaining
CpGs showed either no differential methylation (gray circle in Fig.
3A) or slight but statistically significant hypomethylation (green
circle inFig.3A); bothliefurther upstreamof the transcriptional start
site, outside of the promoter CpG island. Our data not only confirm
the previously described hypermethylation of the GSTP1 promoter
CpG island, but also show that CpG DNA methylation alteration is
highly context-dependent even within a single promoter.
In addition to GSTP1, we also examined our data specifically
for methylation changes in the promoters of APC and RASSF1,
in prostate cancer (Jero ´nimo et al. 2004) and were represented by
multiple probes on the HumanMethylation27 array. With APC, all
six CpG sites represented on the array showed hypermethylation
in tumors, located 122 bp upstream to 488 bp downstream of the
TSS (Supplemental Fig. S3). With RASSF1, three CpGs sites were
probed, located 58 bp upstream to 176 bp downstream of the TSS
and within a CpG island boundary; all three were hypermethylated
(SupplementalFig.S4). However, fiveofthe six probes located more
than 2 kb downstream from the TSS in a second CpG island did not
show differential methylation.
While hierarchical clustering of samples using the most dif-
ferentially methylated CpG sites (the set shown in Fig. 2) was able
to distinguish most tumors from benign adjacent tissues, the
classification was not perfect, as indicated by the inclusion of
supervised hierarchical clustering of 181 prostate tissues based on the
5912 and 2151 CpG sites hypermethylated and hypomethylated in
prostate tumors, respectively, as identified by 2-class SAM. (Red branches)
Tumor samples; (blue branches) benign adjacent samples; (red pixels)
high DNA methylation; (green pixels) low DNA methylation.
Differentially methylated CpGs of prostate tumors. Un-
Methylation in prostate cancer
benign adjacent tissue samples within the tumor cluster and vice
versa. To identify CpG sites that could best predict either the tumor
state or the benign adjacent state, we performed a Prediction
Analysis of Microarrays (PAM) to perform sample classification
(Tibshirani et al. 2002). This analysis generated a list of 87 pre-
dictive CpG sites, most of which had increased methylation in the
tumor samples (83/87), and represented 82 gene promoters total
(Supplemental Fig. S5; Supplemental Table S4). The CYBA, GSTP1,
KLK10, PPT2, and CXCL1 promoters each had two CpGs repre-
sented in this list. Notably, in this ranked list of 87 predictive
methylation alterations, the GSTP1 hypermethylation was ranked
57th (Supplemental Table S4). Thus, we have identified 56 mo-
lecular events, most of which had not been previously character-
validated several of these diagnostic methylation markers by
PyroMark sequencing (Supplemental Text S2).
Prognostic methylation markers
To explore tumor heterogeneity, we compared the methylation pro-
files of the 86 tumors with respect to Gleason grade and time to
biochemical recurrence (defined as serum PSA > 0.07 ng/mL after
surgery) of the donors. Gleason grade is a powerful predictor of
and biochemical recurrence has also been correlated with prostate
cancer-specific mortality (Freedland et al. 2005). We conducted a
multiclass SAM in an effort to identify methylation events that dis-
tinguished tumors of different Gleason grades but were unable to
identify such events. Next, we conducted a SAM survival analysis
a false discovery rate of 26.8%, we identified six CpGs that showed
greater methylation in tumors from men who had shorter time to
recurrence and 63 CpGs that showed lower methylation in patients
with shorter time to recurrence (Supplemental Table S5). This strong
observed a bias for CpG sites with increased methylation in the tu-
mor/benign adjacent comparison. At a false discovery rate of 26.8%,
we expect 18 of those calls to be false. At a lower false discovery rate
cutoff of 1%, we only observed four CpGs that showed higher
methylation in patients with shorter time to recurrence and none
that showed lower methylation (Supplemental Table S5). Strikingly,
we did not observe the differential methylation of the CpG island
described by Liu et al. (2011), and this may be due to the different
the distribution of samples of various Gleason scores. This discrep-
ancy warrants further investigation. While we were only able to
identify a small number of CpGs whose methylation state correlated
with time to recurrence, we noted that several of these CpG sites are
in the proximal promoter regions of known cancer-related genes, in-
cluding three CpGs near MAGE gene family members that encode for
strictly tumor-specific antigens (Chomez et al. 2001) and four CpGs
near WT1, a transcription factor gene associated with Wilm’s tumor.
Correlation of tumor hypermethylation with DNA
With nearly one-third of assayed CpGs showing changes in DNA
methylation between tumor and benign adjacent samples, we hy-
pothesized that one or more of the DNA methyltransferases
(DNMTs), or a protein that interacts with a DNMT, had altered
calculated by the UCSC Genome Browser. Circles are CpG sites assayed by HumanMethylation27. (Red circles) Probes that were identified to be
hypermethylatedinprostatetumorsby2-classSAM;(greencircle)probe thatwashypomethylated;(graycircle)probethatshowed nosignificant change.
The numbers below the circles indicate the relative distance in base pairs from the predicted TSS. (B) Heatmap depicts DNA methylation pattern of the
seven probes near GSTP1. The dendrogram is based on the hierarchical clustering from Figure 2. (Red branches) Tumor samples; (blue branches) benign
adjacent samples. Coordinates are based on the NCBI36/hg18 human genome assembly.
GSTP1 CpG island hypermethylation in prostate tumors. (A) Diagram of the RefSeq annotation of the GSTP1 gene. (Green box) CpG island
Kobayashi et al.
1020 Genome Research
activity, possibly due to changes in transcript abundance, in pros-
DNA methylation changes. To test this hypothesis, we selected
RNA from 10 of the benign adjacent and 36 of the tumor samples,
and measured the transcript abundance of DNMT1, DNMT3A,
DNMT3A2, DNMT3B, DNMT3L, and EZH2 using the TaqMan Gene
Expression assay. These genes comprise the known maintenance
methyltransferase (DNMT1) (Chuang et al. 1997), all known
methyltransferases with de novo capability (DNMT1 [Este `ve et al.
2005], DNMT3A [Okano et al. 1999], DNMT3B [Okano et al. 1999]),
and two interacting proteins thought to target methyltransferases
to specific genomic regions (DNMT3L [El-
Maarri et al. 2009] and EZH2 [Okano et al.
1999]). In addition, we uniquely assayed
DNMT3A and its alternative promoter var-
iant DNMT3A2 by using transcript-specific
primers and probes. While several splice
variants of DNMT3B have been character-
ized, we were unable to design variant-
specific primers and probes for them, so
instead we designed primers and probes to
the common region of all DNMT3B vari-
ants. We did not observe detectable levels
of DNMT3L transcript abundance from ei-
ther tumor or benign adjacent samples
(data not shown). When the transcript
levels of the remaining genes were com-
pared between benign adjacent and tumor
samples with a two-tailed t-test, three
showed significant changes: DNMT3A2
(P = 0.0013), DNMT3B (P = 0.024), and
EZH2 (P = 0.026), while DNMT1 and
DNMT3A did not (Fig. 4F).
We compared the expression values
for these five genes to global DNA meth-
ylation levels. Specifically, we plotted the
mean percent methylation of all 5912
hypermethylated CpG sites against rela-
tive expression of each methyltransferase
or interacting protein, and calculated re-
gression and the goodness-of-fit of the re-
(r2= 0.272, P = 0.0031), DNMT3B (r2=
0.197, P = 0.0056), and EZH2 (r2= 0.211,
P = 0.0037) all showed significant cor-
relation between expression and global
hypermethylation, while DNMT1 and
DNMT3A did not (Fig. 4A–E). The corre-
lation between DNMT3A2, DNMT3B, and
EZH2 expression and global hypermeth-
overexpression of the same genes in tu-
global methylation changes seen in pros-
DNMT overexpression recapitulates
hypermethylation events seen
in prostate tumors
To determine whether the increased tran-
script abundance of DNMT3A2, DNMT3B,
a large number of promoter CpGs, we expressed these genes from
the CMV promoter in transient transfection assays in primary cul-
tures of normal prostatic epithelial cells. We used plasmids express-
ing DNMT3A, DNMT3A2, DNMT3B1, DNMT3B2, and DNMT3B3,
anEZH2-cDNA plasmid,and a no-insertplasmid.We co-transfected
each cDNA plasmid with the no-insert plasmid, and independently
with the EZH2 plasmid, and also included a mock ‘‘no-insert plas-
mid only’’ transfection. We calculated the change in DNA methyl-
ation for each CpG between each cDNA transfection and the mock
transfection after 48 h. We then plotted the ideal cumulative
mors. Comparison of transcript levels of DNMTs and EZH2measuredbyTaqManqPCRwiththeaverage
DNA methylation levels of CpG sites that are hypermethylated in prostate tumors. (Blue circles) Benign
adjacent samples; (red circles) tumor samples. The P-value was calculated by linear regression analysis.
y-axis: average DNA methylation levels (beta score); x-axis: relative gene expression levels [log2(RQ)];
(D) DNMT3B expression. (E) EZH2 expression. (F) Comparison of DNMTs and EZH2 transcript levels
between benign adjacent tissues (blue) and tumors (red). Significant differences are indicated by as-
is relative gene expression levels [log2(RQ)].
Expression of DNMTs and EZH2 correlates with global hypermethylation in prostate tu-
Methylation in prostate cancer
distribution function of the DNA methylation level change at all
26,333 CpG sites along with the empirical cumulative distribution
function of just the changes at the 5912 CpG sites hypermeth-
ylated in tumors (Fig. 5A–K), and tested the difference between the
two distribution functions using the Kolmogorov-Smirnov (K-S)
test. In all 11 experimental transfections, the distribution of the
5912 CpG sites was significantly enriched compared to the null:
DNMT3A (P = 6.0310?45), DNMT3A2 (P = 3.5310?62), DNMT3B1
(P = 1.2310?31), DNMT3B2 (P = 5.2310?39), DNMT3B3 (P =
4.6310?44), EZH2 (P = 1.1310?59), DNMT3A+EZH2 (P =
7.8310?64), DNMT3A2+EZH2 (P = 9.8310?65), DNMT3B1+EZH2
(P = 2.1310?29), DNMT3B2+EZH2 (P = 6.7310?42), and
when the plots of the empirical cumulative distribution functions
were visually inspected, we observed that the low P-value of the
K-S test appeared to be driven more by the CpGs of increased
methylation rather than CpGs of decreased methylation in all 11
To test specifically whether the list of 5912 CpG sites was sta-
tistically enriched for CpGs with substantially increased DNA
methylation, we performed a series of chi-square tests. Based on the
distribution of CpG methylation levels in tumor and benign adja-
cent tissues at these CpG sites, we set a cutoff value of 0.05. In other
the experimental transfection compared to the mock transfection
were considered to have substantially increased DNA methylation.
We calculated expected values based on the distribution of these
CpGswith substantially increased DNAmethylationin the entire set
of 26,333 CpGs. When chi-square tests were performed, all 11 ex-
perimental conditions had very low P-values: DNMT3A (P =
DNMT3B2 (P = 1.8310?157), DNMT3B3 (P = 6.6310?10), EZH2 (P =
9.4310?31), DNMT3A+EZH2 (P = 1.5310?13), DNMT3A2+EZH2 (P =
1.1310?11),DNMT3B1+EZH2 (P =1.9310?185), DNMT3B2+EZH2
(P = 9.4310?107), andDNMT3B3+EZH2 (P = 2.3310?68). DNMT3B1
and DNMT3B2, which are alternative splicing isoforms differing
by the presence of one exon, both in the presence and absence of
EZH2 co-transfection, showed the lowest P-values, all <1310?100.
From these data, we conclude that our list of 5912 CpGs is, indeed,
enriched for CpGs with substantially increased methylation
when DNMTs or EZH2 were overexpressed, with DNMT3B1 and
DNMT3B2 appearing to have the strongest impact on the DNA
methylation levels at these sites.
Based on these data, we further investigated the altered DNA
methylation in the DNMT3B1 and DNMT3B2 overexpression ex-
periments. Because these splice isoforms differ by only one exon
coding for 21 amino acids in a linker region (Sakai et al. 2004), we
suspected that they would share many targets. To identify the
CpGs targetedby DNMT3B1and DNMT3B2 in prostatetumors,we
examined the list of CpGs that were hypermethylated in prostate
tumors and in the overexpression experiments. Specifically, we
looked for overlaps in the list of CpGs with 5% or greater increase
in methylation compared to the mock in the DNMT3B1 (1267
CpGs), DNMT3B1+EZH2 (1322 CpGs), DNMT3B2 (1261 CpGs),
and DNMT3B2+EZH2 (1235 CpGs) overexpression experiments.
Four hundred and thirty-eight CpGs were represented in all four
lists, and an additional 425 CpGs were represented in three of the
four lists. We performed two permutation tests to determine the
four lists of CpGs (1267, 1322, 1261, and 1235 CpGs, respectively)
drawn randomly from the whole list of 26,333 CpGs and counted
the number of incidences where there was an overlap of 438 CpGs
in the 10,000 permutations. In our second permutation test, we
repeated the first permutation test but changed the criteria to ob-
serving at least 863 CpGs overlapping in three of thefour lists. This,
too, was never observed in 10,000 permutations. This provided
further evidence that the differentially methylated CpGs in the
DNMT3B1 and DNMT3B2 overexpression experiments, indeed,
significantly deviated from random sampling and are likely to be
those that are specifically, directly or indirectly, targeted by these
Alterations in DNA methylation have been shown to play a role in
tumorigenesis and cancer progression in many malignancies, in-
cluding prostate cancer. Until recently, technical limitations have
restricted these findings to either characterization of a handful of
candidate loci or of overall abundance of 5-methylcytosine in the
genome. Although a previously published study reported the DNA
methylation profiles of prostate tumors at CpG islands across the
profiles of normal prostate tissue necessary to determine the meth-
ylation changes that occur during or as a result of tumorigenesis.
Here, we present quantitative DNA methylation levels at more than
26,000 loci across 14,000 gene promoters. Because we assayed 95
cancers and 86 benign adjacent prostate tissues in parallel at CpGs
specifically enriched at gene promoters, we were able to show that
thousands of novel changes, including a set of hypermethylated loci
more strongly predictive of prostate cancer than GSTP1. Our data
show that DNA methylation changes in prostate cancer occur on
a broad scale, at many loci throughout the genome.
DNA methylation alteration has been observed in early cancers
and precursor lesions, suggesting that methylation changes drive
malignant initiation rather than tumor progression (Belinsky et al.
1998; Brooks et al. 1998; Baylin et al. 2001; Guerrero-Preston et al.
the acquisition of DNA methylation alterations continues through-
out tumor progression, variation in methylation profiles should be
observed in tumors of different histological grades and clinical out-
comes. Although we detected more heterogeneity among tumors
than among benign adjacent tissues, the vast majority of tumors fell
in a single cluster and we did not observe obvious subclassifications,
although some tumor samples did cluster with benign adjacent
that clustered with benign adjacent tissues against the donors of the
other tumors but did not observe any differences in Gleason grades
tumor heterogeneity that did exist, we identified several dozen
DNA methylation changes that correlated with patients’ time-to-
recurrence. While the sites we identified were different from those
identified by Liu et al. (2011), both our study and theirs were
CpG islands across tumors.
The fact that we observed changes at a very specific subset of
CpG sites across most tumors, rather than a global DNA methyl-
ation deregulation or instability, suggests a common mechanism
of tumorigenesis among prostate cancers. This specificity in target
sites was particularly apparent in gene promoters assayed by
multiple probes and by the PyroMark assay (Supplemental Text
Kobayashi et al.
empirical (red) cumulative distribution functions of change in DNA methylation after DNMT or EZH2 transfection into cultured normal prostate cells. The
empirical distribution functions are based on the 5912 CpGs that were hypermethylated in prostate tumors, while the ideal distribution functions are
DNMT3A and EZH2, (H) DNMT3A2 and EZH2, (I) DNMT3B1 and EZH2, (J) DNMT3B2 and EZH2, and (K) DNMT3B3 and EZH2.
Overexpression of DNMTs and EZH2 results in increased methylation at a subset of prostate tumor hypermethylation sites. Ideal (black) and
Methylation in prostate cancer
S2). The case of GSTP1 illustrates this point well, where the methyl-
ation changes were highly context-dependent: Only the CpG island
on these findings, we suspect that cellular processes involved with
targeted CpGmethylation regulation are themselves misregulated or
altered in early tumor initiation. The most likely candidates are
DNMTs and DNMT-interacting proteins. In support of this hypoth-
esis, we observed significant correlations between the gene ex-
pression levels and levels of global hypermethylation for several of
these candidates. In vitro experiments in normal prostatic epi-
thelial cells confirmed that overexpression of DNMT3B1 and
DNMT3B2 leads to the hypermethylation of a subset of the pros-
tate tumor-specific changes. These data, together with previous
observations, strongly suggests that dysregulation of DNMTs and
possibly DNMT- interacting proteins are among the earliest events
While we did not address the mechanism for the observed
decreased methylation of some CpGs in tumors, there are three
likely possibilities. First, there may be aberrations in the mainte-
nance DNA methyltransferase gene DNMT1. Although we did not
observe a decrease in the DNMT1 transcript level, there may be a
post-transcriptional dysregulation of this gene or mutations that
lead to decreased activity. Decrease in DNMT1 activity may lead to
improper maintenance and gradual loss of methylation with every
DNA replication. However, this would likely lead to a global loss
rather than a targeted loss at particular CpGs and, therefore, is the
least likely scenario. A second possibility is the dysregulation of
a direct or indirect DNA demethylase. While there have been a few
reports of such enzymes in mammalian cells, none has been con-
clusive, and their existence is still speculative (Iyer et al. 2009;
Bhutani et al. 2010; Okada et al. 2010). Finally, the targeted hypo-
methylation may be the result of dysregulation of an interacting
protein of DNMT1 or the hypothetical DNA demethylase. With
more than 20 DNMT1-interacting proteins already identified, it is
conceivable that one or more of them are involved in DNMT1 tar-
geting.However,abetter understandingofthe biologybehindDNA
demethylation is needed to answer this question.
By approaching DNA methylation in cancer from a genomic
perspective, we were able to gain new insights into the underlying
accurate diagnosis of the disease. However, our study was limited
in scale by technology and practicality: With only 26,333 assayed
CpGs, mostly biased toward gene promoters, it is likely that these
results are not representative of the 28 million CpGs found in the
human genome. Even among nearby CpGs in the same promoter,
the PyroMark assay revealed variability in methylation levels (Sup-
plemental Text S2; Supplemental Fig. S7). Thus, using Reduced
Representation Bisulfite Sequencing (RRBS) (Meissner et al. 2005)
and/or the new Illumina HumanMethylation450 array will likely
uncover additional sites of interest, probably including ones with
better diagnostic and prognostic value. In addition, this is the first
study comparing methylation inprostate cancertobenign adjacent
tissue; while our cohort was clinically representative of patients
presenting with the disease, it is paramount that the findings be
now verified in an independent replication set of samples. Fur-
thermore, recent success in integrative analysis of copy-number
in studying prostate cancer from multiple perspectives (Taylor et al.
2010). Expanding such an integrative analysis to include DNA
to lead to a better understanding of prostate cancer biology, and if
robust diagnostic and prognostic markers can be identified, they
will need to be developed into biomarkers for use in a clinical set-
ting. Finally, as methylation profile data of more tumor types be-
come available, researchers will be able to identify common and
type-specific alterations. While we were largely unable to do this
among existing data sets due to drastically different study designs,
investigations of these similarities and distinctions are likely to lead
to a deeper understanding of cancer biology as a whole.
Sample collection and preparation
All prostate samples used for this study were collected at the Stan-
ford University Medical Center between 1999 and 2007 with pa-
tient’s informed consent under an IRB-approved protocol. Multiple
tissue samples were harvested from each prostate, flash-frozen, and
stored at ?80°C. Sections of each prostate tissue sample were eval-
uated by a genitourinary pathologist. The tumor and non-tumor
areas were marked, and contaminating tissues were trimmed away
fromthe blockasdescribedpreviously(Lapointeetal. 2004).Tumor
samples in which at least 90% of the epithelial cells were cancerous
and non-tumor samples having no observable tumor epithelium
were selected for extraction of DNA and RNA. Clinical information
associated with prostate samples included in the analysis is sum-
marized in Supplemental Table S1.
Primary prostate cell culture and transfection assays
A primary culture of human prostatic epithelial cells (E-PZ-231) was
establishedfrom benigntissueoftheperipheralzone ofthe prostate
of a 56-yr-old man who underwent radical prostatectomy to treat
prostate cancer. Using previously described methods (Peehl 2002),
primary cultures were serially passaged. When tertiary passage cells
were ;50% confluent, they were fed Complete PFMR- 4A medium
(Peehl 2002) without gentamycin until they reached ;85%
confluency. Cells in each 60-mm, collagen-coated dish were then
transfected with 10 mg of plasmid DNA using Lipofectamine 2000
(Invitrogen) according to the manufacturer’s instructions. After
48 h, cells from three 60-mm dishes per condition were dissociated
with TrypLE Express (Invitrogen), centrifuged, and snap-frozen in
liquid nitrogen. These cell pellets were then used for DNA isolation.
Nucleic acid isolation
DNA and RNA were isolated from tissue samples or cell cultures
manufacturer’s protocol, with the exception of the RNA from pri-
mary prostate cell cultures. This RNA was isolated with TRIzol Re-
agent (Invitrogen) according to the manufacturer’s instructions.
Sodium bisulfite conversion
Sodium bisulfiteconversionof genomicDNAwas performed using
The conversion was completed using the alternative incubation
protocol for the Illumina Infinium Methylation Assay, as described
by the manufacturer.
Methylation analysis by Illumina Infinium
Five hundred nanograms of sodium bisulfite–converted genomic
HumanMethylaton27, RevB Beadchip Kits (Illumina). The assay was
performed using the protocol as described by the manufacturer.
Kobayashi et al.
1024 Genome Research
Beta score calculations, quality filtering,
and batch normalization
HumanMethylation27 array results were initially extracted and
analyzed using Illumina BeadStudio software with the Methylation
Module v3.2. Beta scores were calculated manually using values
exported from BeadStudio. For each probe intensity value, we sub-
tracted the median negative background control probe value based
on the color channel. The beta score was calculated using the
background subtracted intensity values as: b = IntensityMethylated/
(IntensityMethylated+ IntensityUnmethylated). Any negative beta scores
were converted to a zero. Any beta scores with an associated de-
tection P-value of >0.01 were converted to ‘‘missing values.’’ To cor-
rect for any array-by-array variation, we imputed all missing values
R-package (Johnson et al. 2007). All previously imputed values were
converted back to ‘‘missing values’’ for subsequent analyses.
To remove CpG probes with potentially problematic hybrid-
ization, we performed BLAT on all 27,578 probe sequences against
the GRCh27/hg19 build of the human genome. One thousand and
twenty-eight probes showed questionable mapping and therefore
were removed from analysis. We also identified 217 probes that
included a SNP of >3% minor allele frequency within 15 bp of the
assayed CpG. These probes were also rejected with consideration to
potential variation in probe hybridization due to the common SNP.
Prior to each hierarchical clustering, the beta scores were mean-
centered. Hierarchical clustering of the arrays was done using the
software Cluster 3.0 with Average Linkage. Because these data sets
were too large to cluster the genes by Cluster 3.0, gene clustering
was done using XCluster, available through the Stanford Micro-
array Database, using non-centered Pearson correlation to perform
the hierarchical clustering.
Significance Analysis of Microarray (SAM)
Each SAM was performed as described in the software manual. The
data were analyzed using the latest version of SAM available at the
was implemented using R version 2.10.0.
Prediction Analysis of Microarray (PAM)
Prior to PAM, the CpGs were sorted by standard deviation across all
tumors and benign adjacent tissue. To improve statistical power,
only CpGs that had a standard deviation of 0.04 or greater were
analyzed. PAM was performed as described in the software manual.
The data were analyzed using the latest version of PAM available at
the time of this manuscript preparation, which was version 2.11.
PAM was implemented using R version 2.10.0. Based on visual ex-
amination of the training errors and the cross-validation results, we
set the shrinkage threshold to 10.5.
PyroMark assays were performed at the Stanford Protein and
Nucleic Acid Facility using the manufacturer’s recommended
protocol (QIAGEN). For each target region, three primers were
used: a forward and reverse PCR primer and a sequencing primer.
Primer sequences are listed in Supplemental Table S6.
TaqMan gene expression assay
Expression levels of genes encoding several DNMTand DNMT-inter-
acting proteins, as well as beta-2-microglobulin as an endogenous
control, were measured in 10 benign adjacent and 36 tumor
samples by the TaqMan Gene Expression Assay. We used the fol-
lowing Applied Biosystems inventoried assays with FAM/MGD la-
beled probes (Assay ID in parentheses): DNMT1 (Hs00945900_g1),
DNMT3A (Hs00173377_m1), DNMT3A2 (Hs00601097_m1),
DNMT3B(Hs01003405_m1), DNMT3L (Hs01081364_m1), EZH2
(Hs01016789_m1), and the Human B2M (beta-2-microglobulin)
Endogenous Control. Twenty-five nanograms of cDNA were as-
sayed in triplicate for each target, using the protocol as described by
the manufacturer, on the ABI PRISM 7900HT instrument. The re-
software. Briefly, the average CT and delta-CT were calculated for
B2M CT, we calculated the delta-delta-CT. All sample delta-delta-CT
values were normalized to that of a tumor sample PC625T to gen-
erate an RQ value. To present the RQ value as a positive value, we
added 5 to each RQ value.
The pcDNA3/Myc-EZH2 construct was a generous gift from A.
Chinnaiyan (University of Michigan) (Okano et al. 1999). The
pcDNA3/Myc-DNMT3A, pcDNA3/Myc-DNMT3A2, pcDNA3/Myc-
constructs were a generous gift from A. Riggs (City of Hope) (Chen
et al. 2005).
We thank Kenneth Day, Kevin Roberts, and Krista Stanton for help
with the HumanMethylation27 and TaqMan assays. We thank Preti
Jain for assistance with HumanMethylation27 quality filtering. We
thank the Stanford Protein and Nucleic Acid Facility for generating
the PyroMark data. We thank Kevin Bowling, Marie Cross, Chris
Gunter, Preti Jain, Brittany Lasseigne, Jun Li, Jonathan Pollack, Rob
Tibshirani, Katherine Varley, and Daniela Witten for stimulating
discussions. We also thank Marie Cross, Barbara Dunn, Dan Kvitek,
was funded by NIH (CA111782 to J.D.B.), an Institutional Training
(to Y.K.), and HudsonAlpha funds (to R.M.M. and D.M.A.).
Authors’ contributions: Y.K., D.M.A., Z.G.G., S.R.Y., J.K.M.,
D.M.P.,J.D.B.,R.M.M.,andG.S. agreewiththemanuscript’s results
and conclusions. Y.K., D.M.A., S.R.Y., D.M.P., J.D.B., R.M.M., and
G.S. designed the experiments. Y.K., D.M.A., J.D.B., R.M.M., and
G.S. analyzed the data. Y.K., D.M.A., Z.G.G., and S.R.Y. collected
data and performed experiments for the study. J.K.M. and J.D.B.
prepared tissue samples. Y.K. wrote the first draft of the paper. Y.K.,
D.M.A., D.M.P., J.D.B., R.M.M., and G.S. contributed to the writing
of the paper.
Andriole GL, Grubb RL, Buys SS, Chia D, Church TR, Fouad MN, Gelmann EP,
Kvale PA, Reding DJ, Weissfeld JL, et al. 2009. Mortality results from
Baylin SB, Esteller M, Rountree MR, Bachman KE, Schuebel K, Herman JG.
2001. Aberrant patterns of DNA methylation, chromatin formation and
gene expression in cancer. Hum Mol Genet 10: 687–692.
Belinsky SA, Nikula KJ, Palmisano WA, Michels R, Saccomanno G,
Gabrielson E, Baylin SB, Herman JG. 1998. Aberrant methylation of
p16INK4a is an early event in lung cancer and a potential biomarker for
early diagnosis. Proc Natl Acad Sci 95: 11891–11896.
Bhutani N, Brady JJ, Damian M, Sacco A, Corbel SY, Blau HM. 2010.
Reprogramming towards pluripotency requires AID-dependent DNA
demethylation. Nature 463: 1042–1047.
Methylation in prostate cancer
Brooks JD, Weinstein M, Lin X, Sun Y, Pin SS, Bova GS, Epstein JI, Isaacs WB,
Nelson WG. 1998. CG island methylation changes near the GSTP1 gene
in prostatic intraepithelial neoplasia. Cancer Epidemiol Biomarkers Prev 7:
Brooks JD, Tibshirani R, Ferrari M, Joseph PC, Harcharan G, King CR. 2008.
The impact of tumor volume on outcomes after radical prostatectomy:
Implications for prostate cancer screening. Open Prost Cancer J 1: 1–8.
Cairns P, Esteller M, Herman JG, Schoenberg M, Jeronimo C, Sanchez-
Cespedes M, Chow N, Grasso M, Wu L, Westra WB, et al. 2001.
Molecular detection of prostate cancer in urine by GSTP1
hypermethylation. Clin Cancer Res 7: 2727–2730.
Cancer Genome Atlas Research Network. 2008. Comprehensive genomic
characterization defines human glioblastoma genes and core pathways.
Nature 455: 1061–1068.
Chen Z, Mann JR, Hsieh C, Riggs AD, Che ´din F. 2005. Physical and
functional interactions between the human DNMT3L protein and
members of the de novo methyltransferase family. J Cell Biochem 95:
Chomez P, De Backer O, Bertrand M, De Plaen E, Boon T, Lucas S. 2001. An
overview of the MAGE gene family with the identification of all human
members of the family. Cancer Res 61: 5544–5551.
Chuang LS, Ian H, Koh T, Ng H, Xu G, Li BF.1997. Human DNA-(cytosine-5)
methyltransferase–PCNA complex as a target for p21WAF1. Science 277:
Das PM, Singal R. 2004. DNA methylation and cancer. J Clin Oncol 22:
Ehrlich M. 2002. DNA methylation in cancer: too much, but also too little.
Oncogene 21: 5400–5413.
El-Maarri O, Kareta MS, Mikeska T, Becker T, Diaz-Lacava A, Junen J, Nusgen
N, Behne F, Wienker T, Waha A, et al. 2009. A systematic search for DNA
methyltransferase polymorphisms reveals a rare DNMT3L variant
associated with subtelomeric hypomethylation. Hum Mol Genet 18:
Esteller M, Herman JG. 2002. Cancer as an epigenetic disease: DNA
methylation and chromatin alterations in human tumours. J Pathol
Este `ve P, Chin HG, Pradhan S. 2005. Human maintenance DNA (cytosine-5)-
methyltransferase and p53 modulate expression of p53-repressed
promoters. Proc Natl Acad Sci 102: 1000–1005.
Feinberg AP, Vogelstein B. 1983. Hypomethylation distinguishes genes
of some human cancers from their normal counterparts. Nature 301:
Freedland SJ, Humphreys EB, Mangold LA, Eisenberger M, Dorey FJ,
Walsh PC, Partin AW. 2005. Risk of prostate cancer–specific mortality
following biochemical recurrence after radical prostatectomy. JAMA
Gama-Sosa MA, Slagel VA, Trewyn RW, Oxenhandler R, Kuo KC, Gehrke
CW, Ehrlich M. 1983. The 5-methylcytosine content of DNA from
human tumors. Nucleic Acids Res 11: 6883–6894.
Guerrero-Preston R,Ba ´ezA,BlancoA,BerdascoM,FragaM,EstellerM.2009.
Global DNA methylation: a common early event in oral cancer cases
with exposure to environmental carcinogens or viral agents. P R Health
Sci J 28: 24–29.
Hernandez DG, Nalls MA, Gibbs JR, Arepalli S, van der Brug M, Chong S,
Moore M, Longo DL, Cookson MR, Traynor BJ, et al. 2011. Distinct DNA
methylation changes highly correlated with chronological age in the
human brain. Hum Mol Genet 20: 1164–1172.
Hoffmann MJ, Engers R, Florl AR, Otte AP, Muller M, Schulz WA. 2007.
Expression changes in EZH2, but not in BMI-1, SIRT1, DNMT1 or
DNMT3B are associated with DNA methylation changes in prostate
cancer. Cancer Biol Ther 6: 1403–1412.
Hong C, Bollen AW, Costello JF. 2003. The contribution of genetic and
epigenetic mechanisms togene silencing in oligodendrogliomas. Cancer
Res 63: 7600–7605.
Irizarry RA, Ladd-Acosta C, Wen B, Wu Z, Montano C, Onyango P, Cui H,
Gabo K, Rongione M, Webster M, et al. 2009. The human colon cancer
methylome shows similar hypo- and hypermethylation at conserved
tissue-specific CpG island shores. Nat Genet 41: 178–186.
Iyer LM, Tahiliani M, Rao A, Aravind L. 2009. Prediction of novel families of
enzymes involved in oxidative and other complex modifications of
bases in nucleic acids. Cell Cycle 8: 1698–1710.
Jero ´nimo C, Henrique R, Hoque MO, Mambo E, Ribeiro FR, Varzim G,
Oliveira J, Teixeira MR, Lopes C, Sidransky D. 2004. A quantitative
promoter methylation profile of prostate cancer. Clin Cancer Res 10:
Johnson WE, Rabinovic A, Li C. 2007. Adjusting batch effects in microarray
expression data using empirical Bayes methods. Biostatistics 8: 118–
Jones PA. 1986. DNA methylation and cancer. Cancer Res 46: 461–466.
Kim E, Kim Y, Jeong P, Ha Y, Bae S, Kim W. 2008. Methylation of the RUNX3
promoter as a potential prognostic marker for bladder tumor. J Urol 180:
King JC, Xu J, Wongvipat J, Hieronymus H, Carver BS, Leung DH, Taylor BS,
Sander C, Cardiff RD, Couto SS, et al. 2009. Cooperativity of TMPRSS2-
ERG with PI3-kinase pathway activation in prostate oncogenesis. Nat
Genet 41: 524–526.
Kron K, Pethe V, Briollais L, Sadikovic B, Ozcelik H, Sunderji A,
Venkateswaran V, Pinthus J, Fleshner N, van der Kwast T, et al. 2009.
Discovery of novel hypermethylated genes in prostate cancer using
genomic CpG island microarrays. PLoS ONE 4: e4830. doi: 10.1371/
Kron KJ, Liu L, Pethe VV, Demetrashvili N, Nesbitt ME, Trachtenberg J,
Ozcelik H, Fleshner NE, Briollais L, van der Kwast TH, et al. 2010. DNA
methylation of HOXD3 as a marker of prostate cancer progression. Lab
Invest 90: 1060–1067.
Laird PW, Jaenisch R. 1994. DNA methylation and cancer. Hum Mol Genet 3:
Laird PW, Jaenisch R. 1996. The role of DNA methylation in cancer genetic
and epigenetics. Annu Rev Genet 30: 441–464.
Lapeyre JN, Walker MS, Becker FF. 1981. DNA methylation and methylase
levels in normal and malignant mouse hepatic tissues. Carcinogenesis 2:
Lapointe J, Li C, Higgins JP, van de Rijn M, Bair E, Montgomery K, Ferrari M,
Egevad L, Rayford W, Bergerheim U, et al. 2004. Gene expression
profiling identifies clinically relevant subtypes of prostate cancer. Proc
Natl Acad Sci 101: 811–816.
Lee WH, Morton RA, Epstein JI, Brooks JD, Campbell PA, Bova GS, Hsieh
WS, Isaacs WB, Nelson WG. 1994. Cytidine methylation of
regulatory sequences near the pi-class glutathione S-transferase gene
accompanies human prostatic carcinogenesis. Proc Natl Acad Sci 91:
Ley TJ, Ding L, Walter MJ, McLellan MD, Lamprecht T, Larson DE, Kandoth
C, Payton JE, Baty J, Welch J, et al. 2010. DNMT3A mutations in acute
myeloid leukemia. N Engl J Med 363: 2424–2433.
Rees B, Gage W, et al. 2001. GSTP1 CpG island hypermethylation is
responsible for the absence of GSTP1 expression in human prostate
cancer cells. Am J Pathol 159: 1815–1826.
Liu L, Kron KJ, Pethe VV, Demetrashvili N, Nesbitt ME, Trachtenberg J,
Ozcelik H, Fleshner NE, Briollais L, van der Kwast TH, et al. 2011.
Association of tissue promoter methylation levels of APC, TGFb2,
HOXD3, and RASSF1Awith prostate cancer progression. Int J Cancer doi:
Meissner A, Gnirke A, Bell GW, Ramsahoye B, Lander ES, Jaenisch R. 2005.
Reduced representation bisulfite sequencing for comparative high-
resolution DNA methylation analysis. Nucleic Acids Res 33: 5868–5877.
Mu ¨ller HM, Widschwendter A, Fiegl H, Ivarsson L, Goebel G, Perkmann E,
Marth C, Widschwendter M. 2003. DNA methylation in serum of breast
cancer patients. Cancer Res 63: 7641–7645.
Nakayama M, Gonzalgo ML, Yegnasubramanian S, Lin X, De Marzo AM,
Nelson WG. 2004. GSTP1 CpG island hypermethylation as a molecular
biomarker for prostate cancer. J Cell Biochem 91: 540–552.
Noushmehr H, Weisenberger DJ, Diefes K, Phillips HS, Pujara K, Berman BP,
Pan F, Pelloski CE, Sulman EP, Bhat KP, et al. 2010. Identification of
a CpG island methylator phenotype that defines a distinct subgroup of
glioma. Cancer Cell 17: 510–522.
Okada Y, Yamagata K, Hong K, Wakayama T, Zhang Y. 2010. A role for the
elongator complex in zygotic paternal genome demethylation. Nature
Okano M, Bell DW, Haber DA, Li E. 1999. DNA methyltransferases Dnmt3a
and Dnmt3b are essential for de novo methylation and mammalian
development. Cell 99: 247–257.
Patra SK, Patra A, Zhao H, Dahiya R. 2002. DNA methyltransferase and
demethylase in human prostate cancer. Mol Carcinog 33: 163–171.
2nd ed. (ed. RI Freshney, MG Freshney), Chapter 6. Wiley, New York.
Pflueger D, Terry S, Sboner A, Habegger L, Esgueva R, Lin P, Svensson MA,
Kitabayashi N, Moss BJ, MacDonald TY, et al. 2011. Discovery of non-
ETS gene fusions in human prostate cancer using next-generation RNA
sequencing. Genome Res 21: 56–67.
Poland KS, Shardy DL, Azim M, Naeem R, Krance RA, Dreyer ZE, Neeley ES,
Zhang N, Qiu YH, Kornblau SM, et al. 2009. Overexpression of ZNF342
by juxtaposition with MPO promoter/enhancer in the novel
translocation t(17;19)(q23;q13.32) in pediatric acute myeloid leukemia
and analysis of ZNF342 expression in leukemia. Genes Chromosomes
Cancer 48: 480–489.
Robbins CM,Tembe WA,BakerA,SinariS,MosesTY,Beckstrom-Sternberg S,
Beckstrom-Sternberg J, Barrett M, Long J, Chinnaiyan A, et al. 2011.
Copy number and targeted mutational analysis reveals novel somatic
events in metastatic prostate tumors. Genome Res 21: 47–55.
Kobayashi et al.
1026 Genome Research
Sakai Y, Suetake I, Shinozaki F, Yamashina S, Tajima S. 2004. Co-
expression of de novo DNA methyltransferases Dnmt3a2 and
Dnmt3L in gonocytes of mouse embryos. Gene Expr Patterns 5:
Sboner A, Demichelis F, Calza S, Pawitan Y, Setlur S, Hoshida Y, Perner S,
Adami H, Fall K, Mucci L, et al. 2010. Molecular sampling of prostate
3: 8. doi: 10.1186/1755-8794-3-8.
Schro ¨der FH, Hugosson J, Roobol MJ, Tammela TL, Ciatto S, Nelen V,
Kwiatkowski M, Lujan M, Lilja H, Zappa M, et al. 2009. Screening and
prostate-cancer mortality in a randomized European study. N Engl J Med
Singh D, Febbo P, Ross K, Jackson D, Manola J, Ladd C, Tamayo P, Renshaw
A, D’Amico A, Richie J, et al. 2002. Gene expression correlates of clinical
prostate cancer behavior. Cancer Cell 1: 203–209.
Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver BS, Arora
VK, Kaushik P, Cerami E, Reva B, et al. 2010. Integrative genomic
profiling of human prostate cancer. Cancer Cell 18: 11–22.
Tibshirani R, Hastie T, Narasimhan B, Chu G. 2002. Diagnosis of multiple
Tomlins SA, Rhodes DR, Perner S, Dhanasekaran SM, Mehra R, Sun X,
Varambally S, Cao X, Tchinda J, Kuefer R, et al. 2005. Recurrent fusion of
TMPRSS2 and ETS transcription factor genes in prostate cancer. Science
Tomlins SA, Rhodes DR, Yu J, Varambally S, Mehra R, Perner S, Demichelis F,
rearrangement-negative prostate cancers. Cancer Cell 13: 519–528.
Tusher VG, Tibshirani R, Chu G. 2001. Significance analysis of
microarrays applied to the ionizing radiation response. Proc Natl Acad
Sci 98: 5116–5121.
Weisenberger D, Van Den Berg D, Pan F, Berman B, and Laird PW. 2008.
Comprehensive DNA methylation analysis on the Illumina Infinium
assay platform. http://www.illumina.com/documents/products/
Woodson K, O’Reilly KJ, Hanson JC, Nelson D, Walk EL, Tangrea JA. 2008.
The usefulness of the detection of GSTP1 methylation in urine as
a biomarker in the diagnosis of prostate cancer. J Urol 179: 508–512.
Yaqinuddin A, Qureshi S, Qazi R, Abbas F. 2008. Down-regulation of
DNMT3b in PC3 cells effects locus-specific DNA methylation, and
represses cellular growth and migration. Cancer Cell Int 8: 13. doi:
Received December 16, 2010; accepted in revised form March 7, 2011.
Methylation in prostate cancer