Transcript Presentation Sabine Zöchbauer-Müller

Life Sciences Day 2013
Mapping of CpG island methylation
and its prognostic relevance in lung
cancer patients
Sabine ZÖCHBAUER-MÜLLER, MD
Medical University of Vienna
Department of Medicine I
Clinical Division of Oncology
OVERVIEW
• Background of DNA methylation
• Mapping of CpG island (CGI) methylation and its prognostic
relevance in lung cancer patients
• Genome-wide microRNA (miRNA) expression profiling identifies
targets for DNA methylation in non-small cell lung cancers
• Human ressources development
• Publications
• Research grants
• Outlook
• Acknowledgement
EPIGENETIC MECHANISMS CONTRIBUTE TO
TRANSCRIPTIONAL GENE SILENCING
DNA methylation …..
• epigenetic change occurring at
CG dinucleotides within CGI
located in 5` region of many
cancer-related genes
• affects together with other
epigenetic changes binding
of transcription factors to DNA
leading to gene silencing
• frequently occuring change in neoplastic cells
• reversibel by demethylating drugs
DNA METHYLATION IN LUNG CANCER
• Lung cancer leading cause of cancer deaths worldwide with
patient`s 5-year overall survival rates of ~ 14%
• Many protein-encoding genes identified to be frequently
methylated in non-small cell lung cancers (NSCLC)
• So far mainly analyses of single genes or small numbers of genes
• Methylation of certain protein-encoding genes of potential
clinical relevance in NSCLC patients
• Mechanism for deregulation of microRNA expression in NSCLC?
MAPPING OF CGI METHYLATION AND ITS PROGNOSTIC
RELEVANCE IN LUNG CANCER PATIENTS
Aims …
• Genome-wide search for methylated CGIs in tumor (TU) and
corresponding non-malignant lung tissue samples (NL) from
a large number of NSCLC patients
• Identification of tumor-specifically methylated genes
• Confirmation of methylation by gene-specific analyses
• Comparison of methylation with expression patterns
• Investigation of effects of epigenetically active drugs on
gene expression
• Investigation of potential tumor suppressor gene function
• Determination of potential clinical relevance of methylated
genes in NSCLC patients
Heller et al, Carcinogenesis 2013
METHODS
• Methylated DNA immunoprecipitation and microarray analysis using
NimbleGen`s 385K Human CGI plus Promoter array (MeDIP-chip)
including statistical analyses of primary TU and corresponding NL tissue
samples of 101 stage I-IIIA NSCLC patients
• Functional characterization using Ontologizer
• Gene-specific methylation analyses using methylation-sensitive high
resolution melting (MS-HRM) approach
• Bisulfite genomic sequencing (BGS)
• Microarray expression analyses of untreated/drug treated NSCLC cells
• Immunohistochemistry (IHC)
• Overexpression of genes in NSCLC cells followed by cell viability/proliferation
assays
• Comparison of methylation results with clinico-pathological characteristics
of patients
Heller et al, Carcinogenesis 2013
CHROMOSOMAL DISTRIBUTION OF PROBES
FOUND TO BE DIFFERENTIALLY METHYLATED
• Identification of 2.414 probes
differentially methylated between
TU and corresponding NL samples
• 97% of probes tumor-specifically
methylated (dots above upper red
line)
• Tumor-specifically methylated probes
at all chromosomes except
chromosome 22 and Y chromosome
• 3% of probes methylated at higher
extent in NL samples (dots below
lower red line)
X-axis, probe position; y-axis, t-statistics; upper and lower
red lines indicate significance levels (adjusted p-values for
step-down multiple testing < 0.05)
Heller et al, Carcinogenesis 2013
FROM PROBES TO GENES
• Annotation and dereplication of differentially methylated probes
• Identification of 477 unique tumor-specifically methylated genes
• 406 characterized protein encoding genes, 38 predicted genes,
33 non-coding RNA encoding genes
• Association of 95% of protein encoding genes with a 5`CGI
• Several tumor-specifically methylated genes located in gene clusters:
HOXA, HOXB, HOXD, PCDHA, PCDHGA, PCDHGB
• 149 tumor-specifically methylated genes involved in transcriptional
gene expression and 61 in cell adhesion
• From majority of genes methylation in NSCLCs unknown
Heller et al, Carcinogenesis 2013
COMPARISON OF PERCENT OF METHYLATION
IN TU AND NL SAMPLES BY MS-HRM ANALYSES
y-axis: % of methylation; each line represents an individual patient; * P ≤ 0.05; ** P ≤ 0.01; ***P ≤ 0.001
Heller et al, Carcinogenesis 2013
TUMOR-SPECIFIC METHYLATION BY BGS
3 clones of TU and corresponding NL sequenced;
relatively homogeneous methylation in TU; no or low methylation in corresponding NL;
Heller et al, Carcinogenesis 2013
EFFECT OF EPIGENETICALLY ACTIVE DRUGS ON
EXPRESSION OF TUMOR-SPECIFICALLY METHYLATED GENES
•
Microarray expression analyses of untreated and with Aza-dC or Aza-dC/TSA treated NSCLC cells
•
Upregulation of expression of 31% of tumor-specifically methylated genes* after drug treatment in
NSCLC cell lines (fold change ≥ 1.5, FDR ≤ 0.1);
*identified by MeDIP- chip analyses of NSCLC cell lines A549, NCI-H1993 and NCI-H2073
Heller et al, Carcinogenesis 2013
IHC STAINING OF TU AND NL SAMPLES
Lack of expression in
cancer cells
Lack of expression in
cancer cells
Lack of expression in
cancer cells
Weak expression in
cancer cells
Nuclear expression in
normal bronchiolar
epithelial cells
Cytoplasmic expression
in normal bronchiolar
epithelial cells
Cytoplasmic
expression in alveolar
epithelial cells and
macrophages
Stronger cytoplasmic
expression in normal
bronchiolar epithelial
cells and leucocytes
TU samples methylated for individual gene;
Heller et al, Carcinogenesis 2013
REDUCED CELL VIABILITY/CELL PROLIFERATION AFTER L1TD1
OVEREXPRESSION IN L1TD1 METHYLATED NSCLC CELLS
Tumor-specific L1TD1
methylation determined
by MS-HRM analysis
Up-regulated L1TD1 expression
after Aza-dC treatment
Cell viability normalized to GFP control
Cell proliferation determined in real-time
Heller et al, in preparation
REDUCED CELL VIABILITY/CELL PROLIFERATION AFTER ZNF677
OVEREXPRESSION IN ZNF677 METHYLATED NSCLC CELLS
Empty
ZNF677
NCSLC cells transfected
with empty control vector
or ZNF677 expression
vector
Cell viability normalized
to GFP control
Cell proliferation determined
in real-time
Heller et al, in preparation
STATISTICALLY SIGNIFICANT SURVIVAL DIFFERENCES*
REGARDING METHYLATION
Cumulative survival
Methylation of
Shorter DFS
Shorter OS
HOXA2§
HOXA2
HOXA10§
HOXA2 and/or HOXA10§
HOXA2 and/or HOXA10
ZNF677§
* Univariate analysis;
§ Independent prognostic parameters by multivariate analysis;
HOXA2, HOXA10, HOXA2/HOXA10 : SCC patients;
ZNF677: NSCLC patients;
P = 0.013
Months
Kaplan-Meier plot of OS according
to ZNF677 methylation.
Heller et al, Carcinogenesis 2013
Heller et al, in preparation
CONCLUSION
• Identification of large number of tumor-specifically methylated genes
in NSCLC patients
• From many of them methylation in NSCLCs unknown so far
• Involvement of about half of genes in regulation of gene expression or
cell adhesion
• Association of methylation of many genes with transcriptional regulation
• Putative tumor suppressor gene function of L1TD1 and ZNF677
• Identification of HOXA2, HOXA10 and ZNF677 as potential prognostic
markers
• Findings emphasize impact of methylation on pathogenesis of NSCLCs
Heller et al, Carcinogenesis 2013
Heller et al, in preparation
Heller et al, in preparation
GENOME-WIDE miRNA EXPRESSION PROFILING IDENTIFIES TARGETS
FOR DNA METHYLATION IN NON-SMALL CELL LUNG CANCER
Aims …
• Investigation of potential role of DNA methylation on miRNA
silencing in NSCLCs
• Investigation of miRNA methylation in TU and corresponding
NL samples of NSCLC patients
• Determination of potential clinical relevance of miRNA
methylation in NSCLC patients
Heller et al, Clin Cancer Res 2012
METHODS
• Determination of genome-wide miRNA expression of untreated
and with epigentically activs drugs treated A549 cells by
microarray analyses
• miRNA target prediction and functional annotation
• Gene-specific methylation analyses using MS-HRM approach
in NSCLC cell lines, primary TU and corresponding NL tissue
samples from 101 stage I-IIIA NSCLC patients
• BGS
• Comparison of methylation with gene expression patterns
• Comparison of methylation results with clinico-pathological
characteristics of patients
Heller et al, Clin Cancer Res 2012
MICROARRAY EXPRESSION ANALYSIS OF 856 miRNAs
IN A549 CELLS
•Upregulation of expression of 66 miRNAs
after drug treatment
(fold change ≥
1.5; P < 0.1)
•Association of 33 miRNAs with CGI e.g.
miR-7, miR-9-3, miR-29c, miR-34a, miR125a, miR-193a, miR-200c, miR-375
GO analysis of predicted targets of 33 miRNAs
Heller et al, Clin Cancer Res 2012
TUMOR-SPECIFIC METHYLATION OF miR-9-3 AND
miR-193a BY MS-HRM IN NSCLC PATIENTS
Each circle represents individual sample. Heatmaps show comparison of percentage of methylation
between TU and NL samples of each patient.
Heller et al, Clin Cancer Res 2012
miR-9-3 METHYLATION AND SURVIVAL OF LUNG
SQUAMOUS CELL CARCINOMA PATIENTS
Kaplan-Meier plots of disease-free and overall survival according to mi-9-3 methylation.
Heller et al, Clin Cancer Res 2012
CONCLUSION
• Identification of 66 miRNAs with upregulated expression after
drug treatment
• Association of 33 miRNAs with a CGI
• Identification of tumor-specific miR-9-3 and miR-193a methylation
in NSCLC patients
• Association of miR-9-3 methylation with shorter DFS and OS
of SCC patients
• Identification of methylation as mechanism for miRNA silencing
in NSCLCs
• Findings of potential clinical relevance
Heller et al, Clin Cancer Res 2012
HUMAN RESSOURCES DEVELOPMENT
• Gerwin Heller*, PhD, post doc: permanent employment contract
at the MUV, ready for “Venia docendi“
• Valerie Babinsky*, diploma student: finished thesis
• Marlene Weinzierl, diploma student: finished thesis
• Christian Noll*, diploma student: finished thesis
• Corinna Altenberger*, diploma student: finished thesis, now PhD student
• Bianca Schmid, diploma student: thesis in progress
• Barbara Ziegler, technician: Nomination as lab manager
*Financed by WWTF
MAJOR PUBLICATIONS
•
•
•
•
•
•
•
•
•
•
•
•
Heller G et al, Anticancer Res 2009
Heller G et al, Cancer Metastasis Rev 2010
Kollmann K et al*, Blood, 2011
Kollmann K et al*, Oncotarget 2011
Heller G et al, Clin Cancer Res 2012
Ghanim V et al*, Blood 2012
Brodowicz T et al*, Ann Oncol 2012
Heller G et al, Carcinogenesis 2013
Kollmann K et al*, Cancer Cell 2013
Preusser M et al*, Lung Cancer 2013
Höbaus J et al*, Int J Cancer 2013
Berghoff AS et al*, APMIS 2013
Peer reviewed journals; * Co-Autorship; IF 2012
IF 1.713
IF 7.787
IF 9.060
IF 6.636
IF 7.837
IF 9.060
IF 4.120
IF 5.635
IF 24.755
IF 3.392
IF 6.198
IF 2.068
IF 88.261
RESEARCH GRANTS
• Research grant from the FWF to Zöchbauer-Müller S:
DNA methylation mediated microRNA gene silencing in non-small cell
lung cancer patients, duration 3 years
• Special research program (SFB) grant from the FWF to the SFB consortium
(Valent P, Kralovics R, Lion T, Sexl V, Moriggl R, Zöchbauer-Müller S, Zuber
J, Superti-Furga G, Mannhalter C, Nijman S):
Myeloproliferative neoplasms - pathogenesis and development of new
therapeutic strategies; subtheme: Epigenetic mechanisms involved in
disease manifestation and progression in MPN, duration 4 years
• Research grant from the Austrian Society of Hematology and Oncology to
Heller G: Identification of novel epigenetically silenced tumor suppressor
genes in NSCLCs, duration 1 year
OUTLOOK
• Characterization of additional tumor-specifically methylated
genes regarding function and potential prognostic relevance
in NSCLC patients
• Identification of patients with “bad prognosis“ regarding
pattern of tumor-specifically methylated genes
• Identification of exclusively in NSCLCs tumor-specifically
methylated genes
• Investigation of methylation from panel of markers in
serum samples to diagnose NSCLC
• Relevance for more individualized treatment of patients
and early detection of NSCLC ?
THANK YOU FOR THE WWTF GRANT!
Martin Posch, PhD, MUV
Leonhard Müllauer, MD, MUV
Walter Klepetko, MD, MUV
Kwun Fong, MD, Prince Charles Hospital, Brisbane
Balasz Döme, MD, PhD, Budapest
Balasz Hegedüs, PhD, MUV
Veronika Sexl, MD, Vetmeduni
Peter Valent, MD, MUV
Christoph Bock, PhD, CeMM
Martin Bilban, PhD, MUV
Christine Mannhalter, MD, MUV