Structural and Mechanistic Studies of the SWI/SNF Family of Chromatin Remodelling Complexes

DNA, the blueprint of life, is found within 46 chromosomes in every human cell; if stretched out end to end, it would measure two metres in length. In order for these 46 chromosomes to fit into the "control-room" of every cell, known as the nucleus, the DNA must be very tightly packaged. This is achieved by wrapping the DNA around the surface of special proteins, called histones, that are spaced regularly along the DNA like beads on a string. Each DNA-histone bead is known as a nucleosome and each nucleosome is able to pack very closely against its neighbours to form highly compact fibres called chromatin. Although chromatin fibers are very good at compacting DNA into small spaces, they are poor at allowing other proteins access to the DNA.

Many normal processes in the cell involve proteins binding to DNA, such as when genes are decoded to make protein, when chromosomes are replicated prior to cell division, and when special repair proteins are called upon to fix sites of DNA damage. Consequently, complicated organisms like humans, with highly packaged DNA, have had to develop specialized machinery for opening chromatin at very specific regions of the chromosome. This machinery includes a large number of specialized proteins. One group of proteins, known as chromatin remodelling complexes (CRCs), help to expose DNA by using energy to slide or remove nucleosomes within chromatin. CRCs are protein machines, much larger than a single nucleosome, and made up of proteins responsible for recruitment to the correct chromosome region, binding to nucleosomes, or helping to generate the force required for nucleosome sliding or removal. Multiple different CRCs exist in human cells, each made up of similar proteins that can interact with chromatin in subtly different ways or target the complex to different regions of the chromosome.

The recruitment of a CRC to the start of a gene is an important early step in the process of turning on, or activating, that gene. If the cell makes a mistake, failing to recruit CRCs to their correct sites or instead recruiting those complexes to the wrong set of genes, then a dangerous cascade can result in the loss of control of normal cell events such as cell division and death. This type of gene dysregulation takes place at an early stage in the development of every human cancer. Over the past decade, advances in DNA sequencing technology have allowed scientists to identify mistakes in the DNA code, known as DNA mutations, that are found in many types of human cancer. A surprising finding from these studies was that DNA mutations leading to the loss of protein components from one single CRC, known as the BAF complex, are present in as many as 20% of all human cancers. Further investigations showed that DNA mutations altering the BAF complex cause many normal target genes to be switched off whereas new and inappropriate genes often become active.

Although the BAF complex is very often the target of mutations in cancer, surprisingly little is currently known about this important machine, with many questions still unaddressed. For example, how are the different proteins organized in the BAF complex? What roles do the different proteins play in the recruitment of the complex to the correct target genes? How does the BAF complex interact with nucleosomes? How does BAF use energy to bring about nucleosome sliding or eviction? The overarching goal of my future research will be to address these questions using a repertoire of cutting-edge structural biology techniques such as cryo-electron microscopy, protein cross-linking, X-ray crystallography and computational modelling, in order to provide a detailed description of the organization, recruitment and remodelling activity of this important human complex. Such findings will provide a framework for understanding the molecular basis of BAF complex dysregulation, with broad implications for the future treatment of many human cancers.

Chromatin remodelling complexes (CRCs) overcome the barrier presented by the packaging of DNA into nucleosomal and supra-nucleosomal structures and allow access to the factors responsible for transcription, DNA repair, replication and recombination. Genome sequencing of cancer cells has identified the 2MDa human SWI/SNF (BAF) remodelling complex as the most frequently mutated chromatin regulator in human cancer - mutations affecting BAF complex subunits were observed in 20% of all tumors. Although the BAF complex is an important global tumour suppressor, little is known about its architecture and the molecular mechanisms underpinning its recruitment to specific genomic loci and chromatin remodelling activities. My research team will strive to achieve a detailed molecular understanding of the structure, mechanism and promoter recruitment of the eukaryotic SWI/SNF family of chromatin remodelling complexes. Structural work will initially target global architectural details using a proven integrative structural biology pipeline comprising cryo-EM, cross-linking mass spectrometry (CXMS) and computational modelling. High-resolution studies of the complete BAF complex will utilize cutting-edge cryo-EM instrumentation with complementary X-ray crystallography studies targeting sub-complex assemblies. To dissect the molecular determinants of BAF complex recruitment to target gene promoters, we will reconstitute a yeast SWI/SNF-nucleosome-activator protein complex from purified components in vitro. This system will enable us to systematically probe the numerous interactions between CRC, nucleosome and transcriptional activator protein. The structural basis of these CRC-nucleosome interactions will be determined by solving the cryo-EM structure of a BAF-nucleosome complex. Subsequent mechanistic studies will explore the nucleoprotein perturbations accompanying remodelling by trapping "stalled" remodelling intermediates harbouring engineered site-specific DNA-histone crosslinks.

The origin and design of the proposed research program were motivated by a strong commitment to applying the tools of structural and molecular biology to characterize, in detail, a human transcriptional complex found frequently to drive tumorigenesis in a large number of human cancers. Aside from this core focus on impacting the development of future cancer therapies, the research will also have a positive impact on secondary beneficiaries in the UK private sector and junior scientists receiving scientific training. The list of beneficiaries from the proposed research is therefore the following:

Core beneficiaries:
-National health: facilitating the future development of novel chemotherapeutics targeting mutant BAF remodelling complexes in a variety of human tumours

Secondary beneficiaries:
-UK private sector: pharmaceutical companies developing small-molecule therapeutics using structure-guided design principles
-Future UK research leaders: junior scientists in the research program receiving training in cutting-edge structural biology techniques

The analysis of whole genome sequences from human cancer cells has demonstrated that roughly 20% of all people with primary tumours will be harbouring mutations that alter at least one subunit of the BAF chromatin remodelling complex. This number increases to 39% people suffering from Melanoma, 55% cases of Colorectal cancer and 75% diagnosed with Ovarian clear-cell tumours. Follow-up studies have shown that the dysregulated activity of mutant BAF complexes is critical to cancer cell proliferation and survival, suggesting that the mutant BAF complex represents a very promising target for the development of small molecular inhibitors as chemotherapeutic agents. Although such therapeutic development goes beyond the remit of the proposed research program, our structural characterization of the BAF complex will ultimately have an impact in this area by laying the foundations for future structure-guided drug design pipelines within UK pharmaceutical research and development programs. Specifically, we propose to define the subunit topology of the complex for the first time. This would identify and direct the isolation of stable BAF sub-complexes that we would then target for atomic resolution structural studies using X-ray crystallography. Therefore, within the term of the fellowship, my lab will strive to contribute both BAF complex architectural maps and atomic structures of isolated BAF subassemblies, as a solid foundation for future structure-guided drug design efforts. Beyond the term of the fellowship, longer-term future plans include contributing to the translation of these basic research findings into small molecule BAF inhibitors through collaborations with a UK pharmaceutical partner. In this way, my lab's research would continue to have an impact both on UK health and economic prosperity. Opportunities for such commercial research and development partnership are available at the host institute through the Apollo Therapeutics consortium or independently through programs such as the GSK discovery partnership with academia (see Pathways to Impact for more information).

The proposed research program will also have an impact on the training of future leaders of UK research. Junior researchers in my lab will be trained in cutting-edge integrative structural biology involving a collaborative effort with world leaders in mass spectrometry method development, at the University of California, San Francisco, and in integrative computational modelling at the Pasteur institute, Paris. At the end of the research program contributing staff will have mastered many elements of integrative structural biology including single-particle cryo-EM, X-ray crystallography, cross-linking mass spectrometry, computational modelling and biophysical binding measurements. The research program will therefore have an impact on the foundation of future UK research excellence.