Final Report Summary - METHDRE (Distal regulatory elements in cancer progression and treatment: focus on DNA methylation and hydroxymethylation)
Despite great advances in research and treatment strategies cancer remains a key health and economic problem within the EU. Aberrant epigenetic regulation is one of the primary characteristics of many cancers including Head and Neck Squamous Cell Carcinoma (HNSCC). HNSCC is the 6th most common cancer worldwide, with relatively low survival rates which have not improved in recent decades, therefore highlighting a pressing need for novel therapies. In addition to tobacco and alcohol exposure, infection by the human papillary virus (HPV) is an important aetiological factor linked to approximately 40-80% of oropharyngeal cases. There is an increasing knowledge of the role of regulatory pathways involved in HPV-positive and HPV-negative epithelial tumours. However, little is known about the role of cis-regulatory elements and their epigenetic regulation in HNSCC.
The goal of the methDRE was to establish a group with expertise in HNSCC epigenetic regulation and which would explore the potential for epigenetic therapies. The scientific objective was to understand the mechanisms by which chromatin organization and DNA methylation regulate gene expression in healthy and tumour cells of the oral mucosa. The investigations focused on the role of transcription factors, DNA methylation (5mC) and hydroxymethylation (5hmC) in the functioning of gene regulatory elements, especially distal regulatory elements (DREs). We also studied how gene regulation is affected by treatment with DNA demethylating agent (5-aza-2’-deoxycytidine, DAC) and whether DAC can be used in combination with other therapeutic agents. The main findings include:
We have established the genome-wide localization of regulatory elements (DNase I hypersensitive sites, DHSs) in a panel of five HNSCC cell lines, both HPV-positive and HPV-negative; around 114,000 regulatory elements have been identified and analyzed. We show for the first time that AP-1, TP63 and TEADs TFs form the main regulatory network in HNSCC and the interaction between the three transcription factors was further confirmed by ChIP-seq experiments. A major overlap was identified between them, suggesting co-regulatory mechanisms involving TGFβ, WNT/β-catenin, Notch, Hippo and EGFR pathways. Gene ontology analysis confirmed previously identified regulatory pathways of therapeutic value in HNSCC, such as EGFR and WNT, but also pointed towards the significant involvement of the Hippo pathway and its main TF effectors, the TEAD TF family in both HPV-positive and HPV-negative HNSCC cells. This is a novel observation potentially creating new therapeutic options. Indeed, the Hippo pathway inhibitor, verteporfin reduces HNSCC cell viability in therapeutically relevant doses and affects the regulation of genes involved in EGFR and TGFbeta and WNT regulatory pathways as supported by RNA-seq analyses. Therefore, we propose the Hippo pathway as a potential therapeutic target in HNSCC irrespective of aetiology (HPV-positive and HPV-negative) and postulate its role in affecting other HNSCC-relevant pathways (manuscript in preparation).
Understanding methylation processes at regulatory elements is very important since these elements are rich in transcription factor binding sites and play key roles in epigenetic gene control. The response to DAC was established in five HNSCC cell lines and the extensive MeDIP-seq and hMeDIP-seq experiments were carried out in one of the sensitive cell lines to identify methylated and hydroxymethylated regions, respectively, in a genome wide manner. We have identified the genome wide distribution of methylated and hydroxymethylated regions before and after treatment with low (100nM) and high (1µM) doses of DAC and compared them to DHS regions under the same conditions. Around 50% of DHSs, as well as methylated and hydroxymethylated regions change upon DAC treatment and these regions have certain unique characteristics. Although, the project was focused on regulatory elements the results highlighted a unique effect of DAC on epigenetic change at SINE/Alu repetitive sequences, namely decrease in 5mC accompanied by an increase in 5hmC. In particular, the DAC-induced Alu modifications result in epigenetic landscape shift towards the one observed in normal keratinocytes (manuscript in preparation). This opens new directions for follow up studies including the stimulation of interferon response to dsRNAs created by elements activated by DAC as well as activation of new Alu-derived regulatory elements.
Significantly, the DHS, ChIP, MeDIP and hMeDIP genome-wide data sets form basis for more focused investigations where epigenetic status of regulatory elements can be studied in base pair resolution. The need for single base analysis led to a CIG diversion towards chemical capture of DNA epigenetic modifications linked with MinION sequencing technology. In addition, genetic DNA variation was identified within our preliminary DHS-seq data sets, by adapting variant calling techniques typically used in genome sequencing projects. The ability to identify the variation within enhancer regions provides the possibility to discover previously missed disease-causing mutations. The diagnostic potential of the epigenetic-genetic variation of HNSCC-specific enhancers is currently one of the main directions in the lab which resulted in multiple interdisciplinary collaborations and a grant application.
The potential clinical suitability of low concentration DAC was additionally explored by screening a library of 100 commonly used drugs for DAC-sensitizing properties. This resulted in three candidates from which paracetamol was further investigated. DAC and paracetamol were established to work in synergy (using the Chou-Talalay method), allowing DAC to be used at therapeutically relevant doses (below 500nM). We investigated the mechanisms underlying the DAC-paracetamol synergy which appear to be multifactorial and encompass both effects of DAC on paracetamol action (alterations in the cyclooxygensase (COX) pathway and mimicry of paracetamol overdose) as well as decreased DNA methylation by paracetamol. Therefore, we propose DAC to be a potential therapeutic at least in a subset of HNSCC patients with its efficacy significantly increased by use of the common analgesic paracetamol (manuscript submitted). Such interaction is shown for the first time and may carry important clinical implications for treatments and clinical trials using DAC.
Epithelial barrier function highly depends on its three-dimensional structure and complex cell-cell interactions between epithelial cells (keratinocytes). Therefore we have successfully introduced in the lab the 3D culture models (epithelial rafts) that allow mimicking the environment of differentiated epithelium. So far, they have been used to investigate pro-differentiation and anti-migratory properties of retinoic acid (ATRA) as well as the effects of retinoic acid receptor beta deprivation. The results reveal mechanisms significantly different to those observed in a monolayer cell culture (manuscript in preparation).
The methDRE has been a success. The flexibility of the CIG funding delivered significant support for Dr Wiench’s research independence and smooth career integration after she moved to the UK from the United States. It allowed her to establish her research group and a new programme within the UoB with a focus on molecular basis of HNSCC and development of targeted therapies. It directly led to multiple collaborations, four grant applications (one received), it supported three PhD projects as well as laid basis for subsequent PhD and undergraduate projects. The project so far resulted in numerous conference presentations, two papers published in collaboration, two submitted manuscripts and three manuscripts in preparation. Career integration, facilitated by the project, is progressing well and Dr Wiench is now fully engaged in under- and post-graduate activities. The current Dr Wiench’s position offers a career security and good institutional support as she has moved to permanent position as a lecturer in cancer biology and will soon be applying for a position of senior lecturer. Throughout the project Dr Wiench engaged in a number of dissemination, training and supervisory roles proving an effective transfer of knowledge and expertise to the University of Birmingham. These include mentoring research students, the delivery of undergraduate lectures within three modules, training of researchers in epigenetic technologies and public engagement.
The goal of the methDRE was to establish a group with expertise in HNSCC epigenetic regulation and which would explore the potential for epigenetic therapies. The scientific objective was to understand the mechanisms by which chromatin organization and DNA methylation regulate gene expression in healthy and tumour cells of the oral mucosa. The investigations focused on the role of transcription factors, DNA methylation (5mC) and hydroxymethylation (5hmC) in the functioning of gene regulatory elements, especially distal regulatory elements (DREs). We also studied how gene regulation is affected by treatment with DNA demethylating agent (5-aza-2’-deoxycytidine, DAC) and whether DAC can be used in combination with other therapeutic agents. The main findings include:
We have established the genome-wide localization of regulatory elements (DNase I hypersensitive sites, DHSs) in a panel of five HNSCC cell lines, both HPV-positive and HPV-negative; around 114,000 regulatory elements have been identified and analyzed. We show for the first time that AP-1, TP63 and TEADs TFs form the main regulatory network in HNSCC and the interaction between the three transcription factors was further confirmed by ChIP-seq experiments. A major overlap was identified between them, suggesting co-regulatory mechanisms involving TGFβ, WNT/β-catenin, Notch, Hippo and EGFR pathways. Gene ontology analysis confirmed previously identified regulatory pathways of therapeutic value in HNSCC, such as EGFR and WNT, but also pointed towards the significant involvement of the Hippo pathway and its main TF effectors, the TEAD TF family in both HPV-positive and HPV-negative HNSCC cells. This is a novel observation potentially creating new therapeutic options. Indeed, the Hippo pathway inhibitor, verteporfin reduces HNSCC cell viability in therapeutically relevant doses and affects the regulation of genes involved in EGFR and TGFbeta and WNT regulatory pathways as supported by RNA-seq analyses. Therefore, we propose the Hippo pathway as a potential therapeutic target in HNSCC irrespective of aetiology (HPV-positive and HPV-negative) and postulate its role in affecting other HNSCC-relevant pathways (manuscript in preparation).
Understanding methylation processes at regulatory elements is very important since these elements are rich in transcription factor binding sites and play key roles in epigenetic gene control. The response to DAC was established in five HNSCC cell lines and the extensive MeDIP-seq and hMeDIP-seq experiments were carried out in one of the sensitive cell lines to identify methylated and hydroxymethylated regions, respectively, in a genome wide manner. We have identified the genome wide distribution of methylated and hydroxymethylated regions before and after treatment with low (100nM) and high (1µM) doses of DAC and compared them to DHS regions under the same conditions. Around 50% of DHSs, as well as methylated and hydroxymethylated regions change upon DAC treatment and these regions have certain unique characteristics. Although, the project was focused on regulatory elements the results highlighted a unique effect of DAC on epigenetic change at SINE/Alu repetitive sequences, namely decrease in 5mC accompanied by an increase in 5hmC. In particular, the DAC-induced Alu modifications result in epigenetic landscape shift towards the one observed in normal keratinocytes (manuscript in preparation). This opens new directions for follow up studies including the stimulation of interferon response to dsRNAs created by elements activated by DAC as well as activation of new Alu-derived regulatory elements.
Significantly, the DHS, ChIP, MeDIP and hMeDIP genome-wide data sets form basis for more focused investigations where epigenetic status of regulatory elements can be studied in base pair resolution. The need for single base analysis led to a CIG diversion towards chemical capture of DNA epigenetic modifications linked with MinION sequencing technology. In addition, genetic DNA variation was identified within our preliminary DHS-seq data sets, by adapting variant calling techniques typically used in genome sequencing projects. The ability to identify the variation within enhancer regions provides the possibility to discover previously missed disease-causing mutations. The diagnostic potential of the epigenetic-genetic variation of HNSCC-specific enhancers is currently one of the main directions in the lab which resulted in multiple interdisciplinary collaborations and a grant application.
The potential clinical suitability of low concentration DAC was additionally explored by screening a library of 100 commonly used drugs for DAC-sensitizing properties. This resulted in three candidates from which paracetamol was further investigated. DAC and paracetamol were established to work in synergy (using the Chou-Talalay method), allowing DAC to be used at therapeutically relevant doses (below 500nM). We investigated the mechanisms underlying the DAC-paracetamol synergy which appear to be multifactorial and encompass both effects of DAC on paracetamol action (alterations in the cyclooxygensase (COX) pathway and mimicry of paracetamol overdose) as well as decreased DNA methylation by paracetamol. Therefore, we propose DAC to be a potential therapeutic at least in a subset of HNSCC patients with its efficacy significantly increased by use of the common analgesic paracetamol (manuscript submitted). Such interaction is shown for the first time and may carry important clinical implications for treatments and clinical trials using DAC.
Epithelial barrier function highly depends on its three-dimensional structure and complex cell-cell interactions between epithelial cells (keratinocytes). Therefore we have successfully introduced in the lab the 3D culture models (epithelial rafts) that allow mimicking the environment of differentiated epithelium. So far, they have been used to investigate pro-differentiation and anti-migratory properties of retinoic acid (ATRA) as well as the effects of retinoic acid receptor beta deprivation. The results reveal mechanisms significantly different to those observed in a monolayer cell culture (manuscript in preparation).
The methDRE has been a success. The flexibility of the CIG funding delivered significant support for Dr Wiench’s research independence and smooth career integration after she moved to the UK from the United States. It allowed her to establish her research group and a new programme within the UoB with a focus on molecular basis of HNSCC and development of targeted therapies. It directly led to multiple collaborations, four grant applications (one received), it supported three PhD projects as well as laid basis for subsequent PhD and undergraduate projects. The project so far resulted in numerous conference presentations, two papers published in collaboration, two submitted manuscripts and three manuscripts in preparation. Career integration, facilitated by the project, is progressing well and Dr Wiench is now fully engaged in under- and post-graduate activities. The current Dr Wiench’s position offers a career security and good institutional support as she has moved to permanent position as a lecturer in cancer biology and will soon be applying for a position of senior lecturer. Throughout the project Dr Wiench engaged in a number of dissemination, training and supervisory roles proving an effective transfer of knowledge and expertise to the University of Birmingham. These include mentoring research students, the delivery of undergraduate lectures within three modules, training of researchers in epigenetic technologies and public engagement.