Final Report Summary - EPIX (Genome-wide study of human epigenetic regulators)
Our research focuses on DNA methylation, a paradigmatic epigenetic mark. Methylation of genomic DNA is a vital phenomenon in mammals and a fundamental mechanism of gene expression control. It is also central in a number of human diseases, including neurodevelopmental disorders and cancer (PMID: 22641018).
This has motivated a search for the positions in the genome that are methylated and for the proteins that can recognize methylated DNA and mediate the effect of DNA methylation. The EPIX project focuses on this latter endeavor. Our goal is to study the binding sites of methyl-CpG binding proteins (MBP) in human cells at the genome wide level. This will allow us to draw hypotheses regarding their targets, modes of action and biological functions that we will test experimentally in relevant model systems.
Three groups of proteins that bind methylated DNA in mammals were predominantly investigated in our work. First: the proteins containing the MBD (methyl-CpG binding domain) motif. These are MBD1, MBD2, MBD4 and MeCP2. Second: the proteins containing an SRA (Set and ring associated domain): UHRF1 and UHRF2. Third: the zinc finger proteins ZBTB33 (also called KAISO), ZBTB4 and ZBTB38.
Since the start of the EPIX project, we have made significant progress in understanding the targets, mode of action and biological function of these methyl-CpG binding proteins (MBPs).
Function(s) of MBPs in gene expression regulation.
We identified a number of functional genomic targets for MBPs. We demonstrated that ZBTB38 regulates the expression of DNA replication factor MCM10, and hence plays a critical role in the regulation of DNA replication (PMID: 24726359). We demonstrated that MBD5 and MBD6 targets are also bound by PR-DUB, a multiprotein complex important for histone ubiquitylation (PMID: 24634419). We demonstrated that MBD4, in cooperation with DNA methyltransferase DNMT1 silenced the meiotic gene MSH4 (PMID: 24434851). Finally, we discovered that ZBTB4 regulates the expression of several genes important for the control of mitosis (manuscript in preparation) and cell transformation (PMID: 21765466).
We also characterized the protein partners of MBPs in the nucleus by a proteomic strategy. This approach (not planned in the EPIX proposal) enabled us to better characterize the mode of action of MBD4, MBD5, MBD6, ZBTB38 and ZBTB4 in gene expression regulation, and other nuclear processes (PMID: 24726359 ; PMID: 24634419 ; PMID: 24434851 ; PMID: 21765466).
Function(s) of MBPs in DNA replication.
We identified and characterized a new function of ZBTB38 in DNA replication. We demonstrated that the accumulation of ZBTB38 protein in the nucleus, caused by the inactivation of E3 ligase RBBP6, leads to massive DNA replication defects and eventually chromosomal instability (PMID: 24726359). We also better characterized the known function of UHRF1 in DNA replication. We were able to demonstrate that not only UHRF1 recognizes hemi-methylated DNA at the replication forks; it actually interacts with replication factors present at the fork (data still unpublished).
Function(s) of MBPs in DNA repair.
We demonstrated that many MBPs are involved in DNA repair mechanisms, an observation clearly unpredictable when we started the EPIX project. We identified and characterized the role of ZBTB38, UHRF1, MBD6 and MBD4 in DNA repair processes. We published that MBD6 cooperate with PR-DUB in the repair of UV-induced damages (PMID: 24634419). We published that MBD4 is essential for cell survival upon oxidative stress and it is recruited at sites of damage (PMID: 24434851). We also observed that UHRF1 and ZBTB38 are recruited at sites of UV- and oxidative-induced damages respectively. We are currently investigating the molecular function of these factors in the DNA repair process, and how this function may impact their role in DNA replication and gene expression regulation.
Function(s) of MBPs during development.
We have generated knockout model mice for ZBTB4 and ZBTB38.
In the case of ZBTB4, we were able to show that the inactivation does not cause premature death or sterility. Nevertheless, adult mutant mice present an array of phenotypes ranging from neurodevelopmental disorders (such as depression and heighten anxiety) to developmental phenotypes (smaller size, smaller organs, abnormal blood values). The causes of these phenotypes remain elusive and are still under investigation.
In the case of ZBTB38 we only obtained the first pups recently and the consequences of ZBTB38 inactivation are still under preliminary evaluation.
Our plan for the future is to generate the triple knockout mice ZBTB4/ZBTB38/ZBTB33, and be able to investigate the redundancy between the different factors of the KAISO family. The KAISO knockout mice were kindly provided by Prokhortchouk Egor (Russian Academy of Sciences), a long-term collaborator of Pierre-Antoine Defossez.
Function(s) of MBPs in cancer.
Many research labs have noticed the mys-regulation of MBPs expression in cancer over the last years. We demonstrated that the expression of ZBTB4 and ZBTB38 mRNAs is reduced in tumors from breast and prostate cancer respectively compared to healthy tissue (PMID: 21765466 and manuscript under preparation). Using cellular models, that mimic the disease, we demonstrated that ZBTB4 regulates the expression of several genes important for cellular transformation and invasion (PMID: 21765466). In the case of ZBTB38 we observed that the expression level of ZBTB38 in prostate tumors correlate with the grade of the tumor (Gleason score), the resurgence of cancer following surgery and the clinical response to chemotherapy.
Overall, the EPIX project successfully achieved its goals to better characterize the function of MBPs in human cells. We identified their implication in cellular processes that were clearly unexpected when we started the project such as DNA repair. Further, our work in mice and human cohorts led to clear advances in our understanding of the molecular bases of diseases such as cancer. In the future, our basic knowledge on ZBTB4 and ZBTB38 may be used to diagnose these diseases or develop new therapeutic strategies.
This has motivated a search for the positions in the genome that are methylated and for the proteins that can recognize methylated DNA and mediate the effect of DNA methylation. The EPIX project focuses on this latter endeavor. Our goal is to study the binding sites of methyl-CpG binding proteins (MBP) in human cells at the genome wide level. This will allow us to draw hypotheses regarding their targets, modes of action and biological functions that we will test experimentally in relevant model systems.
Three groups of proteins that bind methylated DNA in mammals were predominantly investigated in our work. First: the proteins containing the MBD (methyl-CpG binding domain) motif. These are MBD1, MBD2, MBD4 and MeCP2. Second: the proteins containing an SRA (Set and ring associated domain): UHRF1 and UHRF2. Third: the zinc finger proteins ZBTB33 (also called KAISO), ZBTB4 and ZBTB38.
Since the start of the EPIX project, we have made significant progress in understanding the targets, mode of action and biological function of these methyl-CpG binding proteins (MBPs).
Function(s) of MBPs in gene expression regulation.
We identified a number of functional genomic targets for MBPs. We demonstrated that ZBTB38 regulates the expression of DNA replication factor MCM10, and hence plays a critical role in the regulation of DNA replication (PMID: 24726359). We demonstrated that MBD5 and MBD6 targets are also bound by PR-DUB, a multiprotein complex important for histone ubiquitylation (PMID: 24634419). We demonstrated that MBD4, in cooperation with DNA methyltransferase DNMT1 silenced the meiotic gene MSH4 (PMID: 24434851). Finally, we discovered that ZBTB4 regulates the expression of several genes important for the control of mitosis (manuscript in preparation) and cell transformation (PMID: 21765466).
We also characterized the protein partners of MBPs in the nucleus by a proteomic strategy. This approach (not planned in the EPIX proposal) enabled us to better characterize the mode of action of MBD4, MBD5, MBD6, ZBTB38 and ZBTB4 in gene expression regulation, and other nuclear processes (PMID: 24726359 ; PMID: 24634419 ; PMID: 24434851 ; PMID: 21765466).
Function(s) of MBPs in DNA replication.
We identified and characterized a new function of ZBTB38 in DNA replication. We demonstrated that the accumulation of ZBTB38 protein in the nucleus, caused by the inactivation of E3 ligase RBBP6, leads to massive DNA replication defects and eventually chromosomal instability (PMID: 24726359). We also better characterized the known function of UHRF1 in DNA replication. We were able to demonstrate that not only UHRF1 recognizes hemi-methylated DNA at the replication forks; it actually interacts with replication factors present at the fork (data still unpublished).
Function(s) of MBPs in DNA repair.
We demonstrated that many MBPs are involved in DNA repair mechanisms, an observation clearly unpredictable when we started the EPIX project. We identified and characterized the role of ZBTB38, UHRF1, MBD6 and MBD4 in DNA repair processes. We published that MBD6 cooperate with PR-DUB in the repair of UV-induced damages (PMID: 24634419). We published that MBD4 is essential for cell survival upon oxidative stress and it is recruited at sites of damage (PMID: 24434851). We also observed that UHRF1 and ZBTB38 are recruited at sites of UV- and oxidative-induced damages respectively. We are currently investigating the molecular function of these factors in the DNA repair process, and how this function may impact their role in DNA replication and gene expression regulation.
Function(s) of MBPs during development.
We have generated knockout model mice for ZBTB4 and ZBTB38.
In the case of ZBTB4, we were able to show that the inactivation does not cause premature death or sterility. Nevertheless, adult mutant mice present an array of phenotypes ranging from neurodevelopmental disorders (such as depression and heighten anxiety) to developmental phenotypes (smaller size, smaller organs, abnormal blood values). The causes of these phenotypes remain elusive and are still under investigation.
In the case of ZBTB38 we only obtained the first pups recently and the consequences of ZBTB38 inactivation are still under preliminary evaluation.
Our plan for the future is to generate the triple knockout mice ZBTB4/ZBTB38/ZBTB33, and be able to investigate the redundancy between the different factors of the KAISO family. The KAISO knockout mice were kindly provided by Prokhortchouk Egor (Russian Academy of Sciences), a long-term collaborator of Pierre-Antoine Defossez.
Function(s) of MBPs in cancer.
Many research labs have noticed the mys-regulation of MBPs expression in cancer over the last years. We demonstrated that the expression of ZBTB4 and ZBTB38 mRNAs is reduced in tumors from breast and prostate cancer respectively compared to healthy tissue (PMID: 21765466 and manuscript under preparation). Using cellular models, that mimic the disease, we demonstrated that ZBTB4 regulates the expression of several genes important for cellular transformation and invasion (PMID: 21765466). In the case of ZBTB38 we observed that the expression level of ZBTB38 in prostate tumors correlate with the grade of the tumor (Gleason score), the resurgence of cancer following surgery and the clinical response to chemotherapy.
Overall, the EPIX project successfully achieved its goals to better characterize the function of MBPs in human cells. We identified their implication in cellular processes that were clearly unexpected when we started the project such as DNA repair. Further, our work in mice and human cohorts led to clear advances in our understanding of the molecular bases of diseases such as cancer. In the future, our basic knowledge on ZBTB4 and ZBTB38 may be used to diagnose these diseases or develop new therapeutic strategies.