Dynamic Flux Balance Analysis of Hepatic Lipid Metabolism: Cell and Organ Studies

许多代谢疾病和紊乱的治疗发展受到对详细机制和系统目标的不完整理解的阻碍。由于肝脏的内在复杂性及其在维持整个身体的血液营养水平方面的作用，肝脏的代谢性疾病尤其具有挑战性。特别是肝性脂肪变性，或脂肪肝，估计影响美国人口的15%至20%。其主要特征是肝细胞内脂肪滴的积累，随后是肝细胞功能的退化。人们认识到肝细胞中的脂肪积累往往是更严重的肝脏问题的前兆，从而对脂肪肝的因果机制产生了兴趣。此外，出现更严重问题的风险与脂肪堆积程度成正比。不幸的是，由于肝脏代谢网络的复杂性和各种可能导致脂肪变性的代谢紊乱，对因果机制的识别受到阻碍。因此，定量了解肝脏脂质代谢网络对于确定脂肪变性的原因和开发治疗方法显然是必要的。本研究的长期目标是建立肝脏代谢的数学模型，并将其用于分析和计算优化治疗脂肪变性的干预措施。该项目的目的是获得实验数据，以训练和测试肝细胞脂肪代谢的系统尺度数学模型。该模型有望合理准确地预测肝细胞对各种干扰的反应。该提议的中心假设是：(A)肝脏脂肪代谢的细胞控制表现为最优反馈控制器，(B)基于代谢最优控制的数学模型可以预测培养的肝细胞对刺激的反应。所提出的模型既可以用于识别导致脂肪堆积的病理机制，也可以用于测试关于操纵特定代谢靶点以逆转脂肪变性和重建体内平衡的方法的替代假设。具体目的是：(1)收集暴露于脂质干扰刺激下的培养肝细胞的代谢物数据；(2)模拟体外培养肝细胞和离体肝脏脂质代谢控制的机制。这项工作的结果将是一个适合分析和预测脂肪变性结果的肝脏代谢模型，并对其准确性和能力进行评估。此外，所建立的建模框架可能适用于涉及多途径的其他组织和代谢疾病。更广泛的影响提出的工作旨在扩展优化为基础的数学建模技术在哺乳动物生理学的应用。因此，它有望刺激这些复杂的建模原理应用于其他复杂的组织和疾病。从长远来看，这些预测模型的成功部署将为使用系统生物学方法的个性化医疗领域提供强大的工具。研究结果将被纳入代谢模型和细胞培养生物反应器的讲座，目前由研究人员教授的研究生课程。具体地说,?组织工程?然后呢?高等数学吗?将受益于研究衍生，学习模块和项目活动。一门新的高级/研究生课程名为？生物医学干预的系统生物学？也是计划的，并且将涉及至少一个研究生来教授建模方法。每年将对两名研究生和两名本科生（包括一名代表性不足的少数民族学生）进行研究方法培训。在第三年，一名研究生将由我们在哈佛医学院的合作者进行动物外科和器官灌注方法的培训。额外的影响将通过参与两个代表性不足的少数民族高中学生在暑期研究实现。研究成果将通过在诸如由美国生物医学工程学会、英国生物医学工程学会和美国化学学会主办的年度会议上发表报告以及在高影响力期刊上发表手稿的方式向科学界传播。

Therapy development for many metabolic diseases and disturbances in hampered by an incomplete understanding of the detailed mechanisms and system objectives. Metabolic diseases of the liver are particularly challenging due to the intrinsic complexity of this organ and its role in maintaining blood levels of nutrients for the entire body. In particular, hepatic steatosis, or fatty liver, is a condition estimated to affect 15 to 20% of the US population. Its key feature is an accumulation of fat droplets within hepatocytes, followed by a degradation of hepatocyte function. Interest in the causal mechanisms of fatty liver has been driven by the realization that fat accumulation in hepatocytes is often a precursor to more serious liver problems. In addition, the risk of more serious problems is proportional to the level of fat accumulation. Unfortunately, identification of the causal mechanisms is hampered by the complexity of the liver metabolic network, and the variety of metabolic disturbances that may lead to steatosis. Thus, a quantitative understanding of the lipid metabolism network in liver is clearly necessary for identification of the causes and developing therapeutic approaches for steatosis.The long-term goals of this research are to develop a mathematical model of liver metabolism and to employ it for analysis and computational optimization of interventions for treating steatosis. The objective of this project is to obtain experimental data for training and testing a system-scale mathematical model of hepatocyte fat metabolism. This model is expected to enable reasonably accurate prediction of the responses of hepatocytes cells to a variety of disturbances. The central hypotheses of the proposal are: (A) the cellular control of hepatic fat metabolism behaves as an optimal feedback- controller and, (B) a mathematical model based on the optimal control of metabolism can predict the responses of cultured hepatocytes to stimuli. The proposed model can be used either for identification of pathological mechanisms leading to fat accumulation, or for testing of alternate hypotheses regarding methods for manipulating specific metabolic targets to reverse steatosis and re-establish homeostasis. The Specific Aims are: (1) To collect metabolite data for cultured hepatocytes exposed to lipid-disturbing stimuli; and (2) To model the mechanisms of lipid metabolic control in cultured hepatocyte and ex vivo livers. The output of the proposed work will be a liver metabolism model suitable for analysis and prediction of steatosis outcomes, with its accuracy and capabilities assessed. In addition, the modeling framework developed is likely to be applicable to other tissues and metabolic diseases involving multiple pathways.Broader ImpactsThe proposed work seeks to extend optimization-based mathematical modeling techniques to applications in mammalian physiology. As such, it is expected to stimulate application of these sophisticated modeling principles to other complex tissues and diseases. In the longer term, successful deployment of these predictive models will contribute powerful tools to the field of personalized medicine using a systems biology approach. The results of the research will be incorporated into lectures on metabolic modeling and cell culture bioreactors, in graduate courses currently taught by the Investigators. Specifically, ?Tissue Engineering? and ?Advanced Mathematics? will benefit from research-derived, learning modules and project activities. A new senior/graduate course entitled ?Systems Biology for Biomedical Interventions? is also planned, and will involve at least one graduate student in teaching the modeling approaches. Two graduate students and two undergraduate students (including one underrepresented minority student per year will be trained in the research methods. In the third year, one graduate student will be trained in animal surgery and organ perfusion methods by our collaborators at Harvard Medical School. Additional impact will be achieved through the involvement of two underrepresented minority high school students in summer research. The research results will be disseminated to the scientific community by presentations at annual conferences such as those sponsored by AICHE, BMES and ACS as well as publication of manuscripts in high impact journals.