Broad Overview
Dr. Zhiguo Zhang’s laboratory studies epigenetic inheritance and cancer epigenetics. How epigenetic states are transmitted into daughter cells is a challenging but yet poorly understood question in the chromatin and epigenetic fields. To gain mechanistic insight into epigenetic inheritance, we focus on elucidating how DNA replication-coupled nucleosome assembly is regulated using a variety of systems including yeast, mouse embryonic stem (ES) cells, and human cells. In cancer epigenetics, we are discovering how onco-histone mutations (H3K27M and H3K36M found in diffuse midline gliomas (DMG) and chondroblastomas, respectively) reprogram cancer epigenomes and drive tumorigenesis, while also working on identifying drug targets for DMG. Finally, we are also developing methods to analyze DNA methylation and hemi-methylation in plasma cell-free DNA for cancer detection.
Specific Aims
1. Elucidate pathways underlying histone deposition onto replicating DNA strands and the connection to transposon silencing
Assembling replicating DNA into nucleosomes using both parental and newly synthesized H3-H4 is the first step for the inheritance of high order chromatin structures. While we have been working on how newly synthesized H3-H4 proteins are deposited following DNA replication, we are currently focusing on understanding how parental H3-H4 tetramers, the primary carriers of epigenetic information, are transferred onto replicating DNA strands for nucleosome assembly.
We developed the eSPAN (enrichment and Sequencing of Protein-Associated Nascent strand) method, which can discern whether a protein is enriched at leading or lagging strands of DNA replication forks in budding yeast and mammalian cells. Using this tool, we discovered that on the one hand, Dpb3-Dpb4 (POLE3 and POLE4 in mammalian cells), two subunits of leading strand DNA polymerase, Pol ε, function as a H3-H4 chaperone for the transfer of parental H3-H4 onto leading strand of DNA replication forks; on the other hand, the Mcm2-Ctf4-Polα axis facilitates the transfer of parental H3-H4 onto lagging strands. Furthermore, these two pathways are conserved from yeast to mammalian cells.
We also found that cells with mutations at genes involved in parental histone transfer show defects in silencing transposons including endogenous retroviral elements. We are presently working on identifying novel factors/genes involved in parental histone transfer and silencing of transposons. Furthermore, we are actively exploring whether targeting these factors will lead to an increase in anti-tumor immunity.
Figure 1. Factors that regulate de novo deposition of new H3-H4.
Figure 2. A model depicting two pathways that mediate the transfer of parental H3-H4 to leading and lagging strands of DNA replication forks.
2. Uncover mechanisms of onco-histone mutations
It has been reported that genes encoding histone H3 proteins are mutated in high-grade pediatric brain tumors, giant cell tumors of bone, and chondroblastoma. There are 13 genes encoding H3.1/H3.2 and two genes encoding H3.3. It was largely unknown how these single-allele mutations at one histone H3 gene impact tumorigenesis. We discovered that the H3.3K27M mutation found in high-grade pediatric brain tumors dominantly reprograms H3K27 methylation and gene expression. We have also provided insight into how the H3.3K36M mutation found in chondroblastoma reshapes the cancer epigenome. Currently, we are focusing identifying drug targets for these deadly diseases.
Figure 3. A model depicting two pathways that mediate the transfer of parental H3-H4 to leading and lagging strands of DNA replication forks.
3. Identify drug targets for diffuse midline glioma (DMG)
To identify drug targets for diffuse midline glioma, we are performing CRISPR/Cas9 screens to identify genes that when depleted kill H3K27M DMG cells specifically. One of the targets we found through this effect is SMARCA4, the catalytic subunit of mammalian SWI/SNF chromatin remodeling complex. Currently, we are testing the hypothesis that epigenomic rewiring creates a dependence of H3K27M DMG cells on this SMARCA4-containing and DMG-specific chromatin remodeling complex. Concurrently, we are also characterizing other proteins revealed by the screens. In the long run, we hope to identify drug targets for this deadly disease.
Figure 4. A model for the role of SMARCA4 (Brg1) in H3.3K27M DMG tumor cells.
4. Develop methods for analysis of DNA methylomes of plasma cell free DNA for multi-tumor detection
We have developed a unique method for analysis of the DNA methylomes of plasma cell-free DNA for early tumor detection. Currently, we continue to test the feasibility of the method using plasma samples from patients with brain, liver, and ovarian cancers.