Holistic Boolean model of cell cycle and investigation of related diseases through perturbation studies : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

The cell cycle is the mechanism by which organisms develop and grow by cell division where a mother cell produces two daughter cells with exact copies of DNA. In the past few decades, much progress has been made in the field of systems biology in studying the complexity of the molecular regulation of cell cycle. However, most recent computational models have focused only on fewer aspects of the cell cycle because of its challenging complexity. Specifically, some core regulatory processes involved in DNA replication are modelled in most studies. However, a very limited attempt has been made to model the other crucial aspect – consistent volume growth during cell cycle to accommodate two sets of DNA and produce two daughter cells. Volume is particularly important because a number of debilitating diseases, including cancer, Alzheimer’s, Parkinson’s, and Down’s syndrome have been attributed to the aberrant cell cycle due to volume dysregulation. DNA replication and volume growth are highly regulated concurrent processes in the cell cycle. This study proposes to develop a holistic computational model of G1/S phase of cell cycle, integrating volume and DNA replication processes in a temporal Boolean model to gain insights into the mammalian cell cycle more comprehensively. It contributes towards the first most comprehensive cell cycle model integrating volume. Additionally, it explores the robustness of cell cycle from a perspective of integrated operations with multiple processes involved in cell division. It also explores the robustness of the cell cycle against single mutations. Further, this study probes into the causes of cell cycle diseases (i.e., volume, neurodegenerative, and cancers) and potential avenues for their understanding, elimination and therapeutics. These aspects along with temporal Boolean modelling are major novel contributions in the proposed study. The cell cycle consists of four main phases, G1, S, G2, and M, representing two Growth phases (G1 and G2) and DNA Synthesis (S) and Mitosis (M) or DNA segregation phases. This study focuses on G1 phase where the cell accomplishes the first volume growth and prepares the conditions (Cyclins and proteins) necessary for DNA synthesis in S phase. Our investigation revealed that these two processes are tightly interlinked and concurrently regulated in G1. The proposed model captures these features to accurately represent this phase of the cell cycle. We focus only on the G1 phase primarily because we realised that closer attention to G1 is needed to bring a clearer picture of some of the crucial aspects of this interlinking that we have uncovered. A cell maintains its volume within close bounds in normal operation and doubles its size in the cell cycle. Volume is increased through osmosis and ion channel operations that bring water into the cell from outside due to gradients in ionic concentrations. Therefore, in cell volume growth, a cell adjusts ion gradients through the operation of a large number of ion channels located on its plasma membrane, which also signifies the important role of bio-electricity (membrane polarisation) in cell cycle. In particular, cell continuously adjusts membrane polarisation to activate the required ion channels throughout volume regulation. Studies found that a large number of ion channels involved in volume regulation in G1 are associated in normal and cancerous cell proliferation. This implication of volume involvement is another reason to keep our focus on G1. Further, during large-scale volume changes, a cell concurrently reorganises its cytoskeleton (CS) to accommodate volume growth. This is achieved primarily through elevated Ca+2 which is established early in G1 phase through the K+ mediated hyperpolarisation of membrane (Vmem) that activates Ca+2 channels to bring Ca+2 into the cell. Ca+2 depolymerises the cytoskeleton and further contributes to increasing Vmem required to activate Cl− channels. This changes the ionic gradient causing the efflux of water through Aquaporin (AQ) and Taurine channels leading to shrinkage of the cell. The purpose of cell shrinkage appears to be primarily to relax the cytoskeleton before swelling. Once the shrinkage has stopped, the swelling process starts through the operation of Cl− and Osmolyte channels activated through increased Vmem for water influx and concurrent repolymerisation of CS that cause the cell to swell. Swelling is stopped upon reaching a volume threshold sensed by sensor protein mTorC1. This sensing defines the volume checkpoint in cell cycle. Ca+2 is a crucial player in controlling and linking both volume regulation and preparation of machinery for DNA replication. Specifically, while contributing to regulation of cell volume as described above, Ca+2 also plays a concurrent role in initiating the DNA replication machinery by activating Immediate Early Genes (IEG) that leads to the production of the first cell cycle Cyclin, Cyclin D. The preparation of the machinery for DNA replication mainly refers to the preparation of Cyclins required for DNA synthesis that takes place in S phase. Cyclins are the drivers of DNA replication, which themselves are tightly regulated by the cell itself through production and degradation processes. The production of Cyclins happens in G1 where they control the transition from G1 to S phase. G1 phase is important for Cyclin production, preparing three Cyclins, Cyclin D, E and A. Among them, Cyclin D is needed to partially release E2F transcription factor which is needed for the production of Cyclin E and A. The first cell cycle Cyclin, Cyclin D, hypophosphorylates Retinoblastoma protein (Rb) bound to partially release E2F for CycE production. This process marks the first checkpoint of the system called Rc, in our proposed model. This and the volume checkpoint are two major checkpoints introduced in our model. The Volume and DNA-replication-related processes further coincide during cell’s passage through Rg-volume checkpoint. Passing of the Rg coincides with complete release of E2F by Cyclin D and Cyclin E. The complete release of E2F factor for Cyclin E synthesis in bulk is done by Cyclin D and E together. During this process, the volume sensor protein, mTorC1, after having ensured that the cell passes Rg, then plays a key role in helping to complete the full release of E2F from Rb to aid bulk CycE synthesis and subsequently CycA. The role of Cyclin E is to assemble the DNA replication machinery on the DNA in late G1; therefore, adequate preparation of CycE signifies the transition from G1 to S phase. We introduce subsystems to integrate the volume regulatory processes, for example, membrane polarisation, CS adjustment, Ca+2, and checkpoints with DNA replicationrelated processes in a holistic system of G1 phase and represent the system in a temporal Boolean model. In particular, our investigation of the literature revealed the two G1 checkpoints mentioned above. One to ensure adequate cell growth (Rg) and the other to ensure the readiness for preparation of Cyclin E (Rc). Only the latter checkpoint Rc has been studied in past computational models and we show in our model how these two checkpoints are operated integrally. This constitutes another important contribution of the study. The goal of the study was to develop and study G1/S network as a holistic system with multiple subsystems. For this, we curated a temporal Boolean core regulatory model of G1 phase of cell cycle with 34 nodes and 42 Boolean Eqs., simplified from over 100 elements. The network model contained six subsystems: signal initiation, Calcium establishment, volume regulation, cytoskeletal regulations, Cyclin synthesis, and checkpoints. The model was implemented on MATLAB. An important aspect of our model is that it incorporates realistic times for the activation and operation of proteins extracted from an extensive literature survey. This gives the model temporal sense while avoiding spurious trajectories commonly found in Boolean models with random asynchronous updates of protein states. This marks another important contribution of this research as other existing Boolean cell cycle models lack time stamps while representing only the DNA-related process. We conducted a comprehensive study of the model to answer a number of questions: (i) Does the model resemble reality? We built the model from an extensive and exhaustive literature survey from which we distilled information for rigorous model building. Still, it is important to assess the correctness of model logic and how well the model represents reality in order to gain valuable insights from the model about the holistic operation of cell cycle and to ensure that the model is realistic to study its response to various perturbations. The model simulation reveals the seamless operation of the subsystems to accomplish the G1/S transfer. Specifically, it correctly unfolds how Vmem, CS, and Ca+2 regulate volume and how volume regulatory processes collaborate with the DNA replication-related processes and how these two strands of activities intricately control the two checkpoints. (ii) How robust is cell cycle design and how vulnerable it is to mutations? We conducted a comprehensive robustness study covering a number of investigations: a) Impact of mutation through element perturbations (Knock on/off), to identify elements and subsystems that crucially impact the main processes of G1: volume regulation, CS and Ca+2, membrane polarisation (MP), checkpoints and G1/S transfer. We found that each subsystem works independently and collaboratively towards the achievement of the G1/S goal. Any failure in one subsystem would either halt the cell cycle progression at various points in G1, or would lead towards cell death. We further found that the minimum working requirement for a subsystem is to achieve its local goal, i.e. to activate the neighbouring subsystem. Moreover, each subsystem has more than one element which can cause total system failure. b) What effects mutations have on volume related diseases, i.e, Alzheimer’s (AD), Parkinson’s (PD), Down’s Syndrome (DS) and other related diseases. We found that components of a sub-system of a larger network can be manipulated to effectively study the impact on the overall disease development. We confirmed through mutational studies that there are components available in the subsystem which can potentially be exploited to stop disease progression, or even eradicate if detected during the early onset of disease. More specifically, the Calcium and volume systems play a role in these disease progression. c) What are the most crucial elements responsible for cancer development and cell cycle-related diseases including, AD, PD and DS. The results showed that Cyclins, checkpoints (including volume checkpoint = Rg), and a few individual nodes from swelling, Calcium, and ion channels can contribute towards cancer development and other major cell cycle-related diseases and therefore be potential interests. These add a wealth of new information to the literature. Further to this, other results are in concert with the existing literature. This first comprehensive G1 model of the cell cycle is a major contribution to cell cycle modelling with its holistic coverage. It has made novel contributions by integrating volume and DNA-related processes and probing the model for cell cycle robustness against mutations, temporal sensitivity, and gaining insights into cancers, other diseases, and system failures. Major Contributions: 1. Extensive Literature Review (over 300) including over 90 Computational Studies 2. Multi-level regulatory Network Building, capturing concurrent regulation of volume increase and preparation of the crucial drivers (Cyclins) of DNA replication 3. Inclusion of volume Sensing, and introduction of the volume checkpoint Rg, and its integration with Cyclin Synthesis Checkpoint Rc 4. Flexible Boolean Logic Synthesis for a Comprehensive representation of the Complex and Dynamic G1/S System 5. Inclusion of Temporal data from in-vitro Expression Studies for Transforming the Standard Boolean Model into a Temporal Boolean Model 6. Investigational Studies via Perturbations on Cell Cycle Robustness, Cancer Development and Cell Cycle Related Diseases
Source DOI
Creative Commons Rights
Attribution 4.0 International
Access Rights