Medical Biology

The Bcl‑2 protein family regulates the intrinsic apoptotic pathway through an intricate network of protein:protein and protein:membrane interactions. The pathway culminates in the permeabilisation of the mitochondrial outer membrane by the pro‑apoptotic effector proteins Bak and Bax, an event that irreversibly commits a cell to death. To facilitate membrane permeabilisation, Bak and Bax undergo a series of conformational changes to convert from inert monomers to membrane‑embedded homodimers that nucleate and propagate apoptotic oligomers. While great strides have been made in structurally characterising these conformational changes, questions remain surrounding homodimer interactions with the membrane, oligomerisation, and membrane pore formation. This thesis addresses these questions by providing structures of lipids bound to Bak BH3:groove core homodimers (Chapter 2). These are the first structures of any Bcl‑2 family protein in complex with lipid. They reveal symmetric binding sites for phospholipid headgroups and acyl chains. In one structure, adjacent Bak homodimers are cross‑linked by the acyl chains of single phospholipids, suggesting homodimer oligomerisation could be mediated by lipid. Bak oligomers could be dissociated with phospholipase A2, supporting a role for lipid in oligomer stability. Collectively, the structures presented here indicate that lipids may play a direct role in Bak oligomerisation. Like Bak, Bax homodimerises and oligomerises on the mitochondrial outer membrane. The original Bax BH3:groove core homodimer structure was solved as a GFP fusion at low resolution. Here, a tetrameric structure consisting of two Bax BH3:groove core homodimers alone was solved at high resolution (Chapter 3), providing details for canonical interactions in atomic detail. A crystal structure of Bax BH3:groove core homodimers containing lipid was also solved, although the structure could not be refined due to severe twinning. This result demonstrates that Bax core domains also associate with lipid, and provides a starting point for crystal optimisation. Pro‑survival Bcl‑2 family proteins antagonise the apoptotic function of Bak and Bax by preventing their activation and sequestering their activated forms. Sequestration of activated Bak and Bax in heterodimeric Mode 2 complexes involves binding of the Bak/Bax BH3 domain to a conserved hydrophobic groove. Beyond this, little is known regarding the topology of these complexes. The pro‑survival protein Bcl‑XL can undergo similar conformational changes to Bak and Bax, but whether it forms BH3:groove heterodimers with Bak/Bax was unknown. Using cysteine cross‑linking on mitochondria, I show that Bcl‑XL can form reciprocal BH3:groove heterodimers with Bax, and possibly Bak (Chapter 4). These results challenge a simplistic view of Mode 2 complexes, implicating more extensive interactions beyond the canonical BH3 in groove interface. Bok is a third potential pro‑apoptotic effector protein that shares sequence similarity with Bak and Bax, but its role in apoptosis remains unresolved. To investigate the structure and function of Bok, I developed a recombinant expression system to produce human, rat, and chicken Bok. The first crystal structure of Bok, from the chicken, reveals the canonical Bcl‑2 family fold, with deviations that may explain its proposed constitutive activity (Chapter 5). The structure paves the way for mutagenesis studies that will further our understanding of this enigmatic protein.

Apoptosis is a form of programmed cell death that is critical for embryonic development, tissue homeostasis and regulation of the immune system of multi-cellular organisms. Aberrations in the apoptotic pathway are thought to be the cause or a contributing factor in a broad range of diseases. Inappropriate survival of cells that should die can cause cancer and also autoimmune disease, whereas killing of normally long-lived cells has been implicated in degenerative disorders. Mammals have two distinct, albeit ultimately converging apoptotic pathways. The so-called “death receptor” (also called “extrinsic”) pathway is triggered by ligand mediated activation of members of the tumour necrosis factor receptor (TNF-R) family with an intra-cellular “death domain”, such as Fas or TNF-R1. The so-called ”Bcl-2-regulated” (also denoted as “stress”, “mitochondrial” or “intrinsic”) apoptotic pathway can be initiated by developmental cues (programmed cell death) or a diverse range of stress stimuli, such as cytokine withdrawal, oncogene activation, hypoxia or DNA damage. The two pathways activate distinct “initiator caspases” (caspase-8 for “death receptor” signalling but caspase-9 in the “Bcl-2-regulated” pathway), which then proteolytically activate common effector caspases (caspase-3, 6 and 7). They promote cellular demolition by cleaving hundreds of critical proteins (e.g. essential structural components) and by proteolytic activation of the DNase CAD (caspase activated DNase), which degrades nuclear DNA. The “Bcl-2-regulated” pathway is controlled by three major subgroups of proteins of the Bcl-2 family. The five pro-survival members, Bcl- 2, Bcl-xL, Mcl-1, Bcl-w and A1, are required for cell survival, acting in a cell type specific manner, although there is also evidence for significant functional overlap. There are two pro-apoptotic sub-groups within the Bcl-2 family that differ in structure and function. The BH3-only proteins (Bim, Puma, Noxa, Bid, Bad, Bik, Bmf, Hrk) are critical for initiation of apoptosis whereas the multi-BH (Bcl-2 Homology) domain pro-apoptotic Bax, Bak (and possibly Bok) are essential for activation of the downstream phases of apoptosis. Death stimuli trigger transcriptional induction and/or post-translational activation of BH3- only proteins. These pro-apoptotic proteins can interact with the pro-survival Bcl-2 family members. Some BH3-only proteins, such as Bim, Puma and activated tBid, can bind to all pro-survival members with high affinity, whereas other BH3-only members have more restricted binding specificity and it is believed that this accounts for the reduced killing capacity (compared to Bim, Puma or tBid). Neutralisation of the pro-survival Bcl- 2 family members by BH3-only proteins facilitates activation of Bax and Bak, but direct interaction with BH3-only proteins has also been implicated in Bax/Bak activation. Activated Bax and Bak can form multimers on the outer mitochondrial membrane resulting in mitochondrial outer membrane permeabilisation (MOMP), with consequent release of cytochrome C into the cytosol where it promotes the assembly of the apoptosome that also comprises pro-caspase-9 and its adaptor Apaf-1. This complex facilitates activation of caspase-9, which in turns proteolytically activates the effector caspases (3, 6 and 7),thereby unleashing cellular demolition. The oncogene Myc is a key transcriptional regulator of cell growth, proliferation and differentiation. Moreover, under conditions of stress, such as limited supply of growth factors, deregulated Myc expression can also promote apoptosis. Myc expression is abnormally elevated in a large fraction (estimated 50-70%) of human cancers. In some of these tumours, such as Burkitt’s lymphoma, Myc is over-expressed as a consequence of chromosomal translocations (t8;14 or t8;22/t2;8) that subjugate the myc gene under the control of the immunoglobulin heavy or light chain gene enhancers, respectively. The elevated expression of c-Myc enforces abnormal proliferation of B lymphoid cells. In many cases of Burkitt’s lymphoma the myc/Ig gene translocation is accompanied by a mutation in the tumour suppressor TP53 gene, which causes severe defects in the in- duction of cell cycle arrest and apoptosis. Lymphomas with intact p53 often present with mutations in genes that regulate the p53 pathways, such as Mdm2, an E3 ligase that inhibits p53 function by promoting its ubiquitin/proteasome dependent degradation. Abnormally increased expression of pro-survival Bcl-2 members has also frequently been observed in Myc over-expressing cancers, particularly lymphoma and leukaemia, and this is thought to promote tumorigenesis by counteracting the aforementioned apoptosis inducing actions of Myc. Although Burkitt’s lymphoma generally responds fairly well to chemotherapeutic drugs and radiotherapy and can be cured by bone marrow transplantation, there remains a need for better understanding of this disease and improved treatment strategies. The Eμ-myc transgenic mouse strain, in which Myc is over-expressed under control of the IgH gene enhancer (Eμ), is the most widely used animal model for studying tumour development. In the pre-leukaemic phase, these animals display abnormally increased cycling with consequently elevated numbers of pro-B and pre-B lymphoid cells and, due to the ability of Myc to inhibit differentiation, a decrease in the more mature sIg+ B cells. Acquisition of additional oncogenic lesions that cooperate with Myc over-expression cause emergence of clonal pre-B or B lymphoma in these mice. Previous work has shown that over-expression of Bcl-2 (or its close relatives, Bcl-xL or Mcl-1) can dramatically accelerate lymphoma development in Eμ-myc mice, most likely because this prolongs the survival of pre-leukaemic cells whilst they are acquiring oncogenic mutations, thereby promoting neoplastic transformation. Although sustained Bcl-2 over-expression was reported to be essential for sustained expansion of pre-B/B lymphomas elicited by combined over-expression of Myc and Bcl-2, endogenous Bcl-2 was found to be dispensable for Myc-induced lymphomagenesis. This indicated that expression of other pro-survival Bcl-2 family members, such as Bcl-xL or Mcl-1, at endogenous levels might be critical for Myc-induced lymphoma development. To investigate the requirement for Bcl-xL in lymphomagenesis in the Eμ-myc transgenic mouse model, pre-leukaemic ani- mals were prophylactically treated with the BH3 mimetic ABT-737 (which binds to Bcl-xL, Bcl-2 and Bcl-w). This study demonstrated that Bcl-xL is critical to sustain the survival of Eμ-myc transgenic B lymphoid cells that are undergoing neoplastic transformation and consequently for Myc-induced lymphoma development. In addition, I examined the role of Mcl-1 in Myc-induced lymphomagenesis. Remarkably, Eμ-myc mice lacking only one allele of mcl-1 survived considerably longer free of lym- phoma compared to control Eμ-myc mice. Mice that are deficient for mcl-1 die around embryonic day 4 prior to implantation and therefore gene-targeted mice in which the mcl-1 gene has been flanked with loxP sites to facilitate its deletion by Cre recombinase in a lineage-specific manner were generated to examine the impact of complete loss of Mcl-1 on Myc induced lymphomagenesis. To allow specific deletion of loxP flanked mcl-1 in B lymphoid cells we employed the CD19-cre transgenic mouse strain and generated Eμ-myc/mcl-1fl/+/CD19-cre and Eμ-myc/mcl-1fl/fl/CD19-cre mice. This analysis revealed that selective loss of one or both alleles of mcl-1 in B lymphoid cells could significantly delay lymphoma development in Eμ-myc mice, although this delay was less pronounced compared to that seen in Eμ-myc/mcl-1+/- mice. This was explained by the discovery that all lymphomas arising in the Eμ-myc/mcl-1fl/+/CD19-cre and Eμ-myc/mcl-1fl/fl/CD19-cre animals had been selected against loss of their floxed mcl-1 alleles. Collectively, these data demonstrate that Mcl-1 is essential for Myc-induced lymphoma development. The tumour suppressor p53 is a transcription factor that is activated by diverse cell stresses, including DNA damage, oncogene activation or hypoxia, and regulates a range of target genes that control cell cycle arrest, apoptosis, cell senescence and many other processes. Mutation or loss of p53 have been implicated in the development of ~50% of human cancers. The development of a substantial fraction of sporadic as well as familial human cancers is facilitated (in part) by environmental stresses, such as exposure to mutagenic chemicals or radiation. In familial cancer syndrome patients, mutations that cause increased predisposition to cancer development are passed on to the next gen- eration. For example, Li-Fraumeni Syndrome patients have a high predisposition to can- cer development due to an inherited mutation in one TP53 allele. Gene-targeted mice, in which one or both alleles of p53 have been deleted or mutated represent a mouse model of Li-Fraumeni Syndrome. Evasion of apoptosis, by up-regulating proteins that secure survival of transformed cells, such as pro-survival Bcl-2-like proteins, is a hallmark of cancer. To test the hypothesis that expression of pro-survival Bcl-2 family members under endogenous control is essential for development of lymphoma or other cancers, p53-deficient mice were prophylactically treated with ABT-737. Analysis of these mice revealed that collectively Bcl-2, Bcl-xL and Bcl-w do not appear to be critical to sustain survival of cells undergoing thymic lymphoma or sarcoma development due to mutations in p53. Hence, Mcl-1 or A1 might be more critical mediators in the setting of tumorigenesis driven by defects in the p53 tumour suppressor pathway. The findings reported in this thesis, in context with published data, demonstrate that Mcl-1 and Bcl-xL are both critical to sustain the survival of cells undergoing neoplastic transformation in Myc-induced pre-B/B cell lymphoma development. In contrast, Bcl-2, Bcl-xL and Bcl-w appear to be dispensable for the survival of cells undergoing neoplastic transformation in T cell lymphoma or sarcoma development initiated by defects in p53. Understanding the molecular mechanisms that are critical to sustain the survival of cells undergoing neoplastic transformation or those that sustain the continued expansion of a tumour will assist the development of improved therapies for treatment of cancer and possibly also early prophylactic intervention.

Medical Biology - Theses

Permanent URI for this collection

Filters

Date

Author

Subject

Type

Settings

Sort By

Results per page

Statistics

Citations

Search Results