 |
Structural Biology of Pathways Involved in Cancer
Summary: Nikola Pavletich is interested in the structural biology of pathways that are altered in cancer, with emphasis on the cell division cycle and the DNA-damage response.
The cell cycle coordinates the nuclear events needed for the growth and proliferation of all eukaryotic cells. These events revolve around the replication of DNA during the S phase of the cell cycle and the segregation of sister chromosomes to daughter cells during M phase. The cell cycle machinery ensures that discrete biochemical processes are executed in an orderly, interdependent, and highly accurate fashion. Throughout the cell cycle, the genome is monitored for DNA damage through processes referred to as DNA-damage checkpoints. These checkpoints can arrest the cell cycle until the damage is repaired, thus preventing the accumulation and propagation of genetic errors to daughter cells. The cell cycle also has an important function in controlling cell growth, exerted at the transition from the G1 to the S phases of the cell cycle. The G1-S transition is when DNA synthesis starts, and it represents an irreversible commitment to complete the cell cycle and divide. The G1-S cell cycle machinery receives signals from diverse extracellular and intracellular growth-regulatory pathways, ensuring that the proliferation of the cell happens at the right time and the right place in the organism.
Because of its roles in ensuring the faithful transmission of genetic information to daughter cells and in controlling cell growth, the cell cycle and its associated checkpoints are among the most frequently deregulated processes in cancer. Mutations and other genetic alterations that deregulate the restriction point can allow the developing tumor to grow in the absence of growth factors. On the other hand, mutations or DNA-damaging environmental carcinogens that interfere with the faithful transmission of genetic information to daughter cells can increase the probability that the cell will accumulate mutations in growth-regulatory genes that will eliminate the normal strictures against uncontrolled growth.
Function of BRCA2 in DNA Repair
Much of our understanding of the role DNA repair has in cancer has emerged from studies of inherited cancer predisposition syndromes caused by mutations in genes involved in the sensing, signaling, or repair of DNA damage. The breast cancer susceptibility protein 2 (BRCA2), which is mutated in inherited breast and ovarian cancer predisposition syndromes, is implicated in the repair of DNA double-strand breaks (DSBs). BRCA2-deficient tumor cells exhibit high levels of chromosomal abnormalities such as breaks, fusions, and translocations, which are among the most tumorigenic lesions known.
DSBs are repaired by a process that utilizes the RAD51 homologous recombination (HR) enzyme. In this process, the DSB is resected to produce single-stranded DNA (ssDNA), which is coated by RAD51 to form a ssDNA-RAD51 nucleoprotein filament. The filament binds to double-stranded DNA (dsDNA) on either the sister or homologous chromosome, finds the homologous DNA sequence, and then initiates a repair process that copies the missing genetic information around the DSB from the sister or homologous chromosome. The requirement of BRCA2 for this process was well established by cellular biological studies, and biochemical analyses had shown that BRCA2 contained eight ~30–amino acid repeats (BRC repeat) that bind to RAD51. The role that BRCA2 played in the HR-mediated DSB repair process was unknown, however.
To investigate this question, we first determined the crystal structure of an ~800–amino acid BRCA2 fragment (thereafter BRCA2CTD) that we thought, based on sequence analyses, is important for BRCA2 function. The structure revealed that BRCA2 contains three structural domains similar to the oligonucleotide-binding domain found in proteins that bind to ssDNA, and another domain similar to the helix-turn-helix motif that binds to dsDNA. In follow-up biochemical experiments, we discovered that the BRCA2CTD binds to ssDNA, and we determined the crystal structure of a BRCA2CTD-ssDNA complex. This structure revealed that several of the BRCA2 amino acids that contact the ssDNA are mutated in cancer, indicating that the DNA-binding activity of BRCA2 is important for its tumor-suppressor function. Our findings led to the model that one function of BRCA2 may be to deliver RAD51 to the ssDNA of the resected DSB.
To address this model, we established an in vitro assay to study the formation of RAD51 nucleoprotein filaments on a DNA substrate that mimics the resected DSB. We found that BRCA2 facilitates the initiation (known as nucleation) of RAD51-ssDNA filament formation. Previous studies had shown that RAD51 filament nucleation is a rate-limiting step, much like other filamentous protein assemblies in the cell. Our finding provides an explanation for the observation that BRCA2 is required for RAD51-mediated DSB repair. We also found that BRCA2 nucleates RAD51 filaments specifically at the dsDNA-ssDNA junction of the processed DSB. Our hypothesis is that filament nucleation at the junction increases the fidelity of repair.
Control of the G1-S Transition by the Retinoblastoma Protein
The major processes that control the cell cycle program are phosphorylation, ubiquitin-dependent proteolysis, and transcription. Phosphorylation is mediated primarily by cyclin-dependent protein kinases (Cdks), which coordinate cell cycle transitions and checkpoints by switching between active and inactive states. Cdks are switched on by the cyclin proteins and off by the CKI family of inhibitory proteins, which are in turn regulated by ubiquitination and transcription in response to signals from growth-regulatory pathways or from cell cycle checkpoints. The cell cycle transcription program is responsible for the production of most proteins involved in the core biochemical processes, such as DNA replication and chromosome segregation. It is controlled by the E2F-DP family of heterodimeric transcription factors and the retinoblastoma (Rb) family of proteins. The cell cycle transcription program is inactive in G1, as the Rb protein binds to E2F-DP complexes and converts them to repressors of transcription at E2F-responsive promoters. Growth factors lead to the activation of Cdk4/6 and Cdk2, which hyperphosphorylate Rb and thereby cause the release of active E2F. Deregulation of the Rb pathway is thought to be a requisite event in tumorigenesis and has been found to occur by mutation of either Rb or the upstream Cdk, cyclin, and CKI proteins.
Our laboratory has been interested in how Rb assembles with the heterodimeric E2F-DP complex, and how this complex is dissociated by Cdk-mediated phosphorylation. Previous studies had focused on the ability of a central Rb domain (termed the pocket) to bind to the transactivation domain of the E2F transcription factors. This interaction alone does not, however, recapitulate the well-demonstrated requirement for the C-terminal domain of Rb (RbC) for its function in G1-S control and growth/tumor suppression.
We have now established that RbC binds directly to E2F-DP. This interaction involves the marked box regions of E2F and DP and has an affinity comparable to that of the Rb pocket binding to the transactivation domain of E2F, indicating it has an important role in the assembly of the Rb-E2F-DP complex. Crystal structure analysis showed that the marked boxes of E2F and DP form an intermolecular β-sandwich structure to which RbC binds. Many of the regulatory phosphorylation sites of Rb map to RbC, and using in vitro biochemistry, we found that phosphorylation of RbC by Cdks dissociates it from the E2F-DP complex in a two-step mechanism. Phosphorylation of RbC at two sites known to be early events in vivo (Ser788 and Ser795) destabilizes one set of RbC-E2F-DP interactions directly, reducing but not eliminating the stability of the complex. Subsequent phosphorylation of RbC at two additional sites (Thr821 and Thr827) induces an intramolecular interaction between RbC and the Rb pocket. This intramolecular interaction involves a region of RbC that is also required for E2F-DP binding, and thus results in the complete dissociation of E2F-DP from RbC.
Grants from the National Institutes of Health provided partial support for these projects.
Last updated May 15, 2006
|
|
|
|
