抗生素導致細胞代謝變化的音調拓樸性與耳蝸外聽毛細胞致病性的關聯：常微分方程數學模型的模擬探討

Tonotopy in Acute Aminoglycoside-induced Metabolic Changes and Vulnerability of Cochlear Outer Hair Cells: An ODE-based Mathematical Modeling Analysis

林惟淨(Wei-Ching Lin)

指導教授 ： 陳義裕

國立臺灣大學/理學院/物理學研究所/碩士(2021年)

https://doi.org/10.6342/NTU202100309

摘要

本篇論文使用常微分方程數學模型，模擬胺基醣苷類抗生素對耳蝸外聽毛細胞能量代謝的即時影響，並比較高頻與低頻外聽毛細胞能量代謝變化的程度，探討何以高頻的外聽毛細胞對抗生素具有較高的致病性。本篇論文提出的外聽毛細胞代謝模型修改自Poliquin等人發表的腦神經細胞模型，包含糖解作用、檸檬酸循環、氧化磷酸化、活性氧類的生成與清除等主要代謝反應，並由Jensen-Smith等人發表的FVB小鼠耳蝸聽毛細胞實驗數據校正。此數學模型的模擬結果顯示：投藥後70分鐘，高頻外聽毛細胞中的ATP濃度下降程度(35%)遠高於低頻細胞中的下降程度(12%)。此研究結果與抗生素導致之耳毒性的音調拓樸性呈現一致。

關鍵字

胺基醣苷類抗生素 ； 耳毒性 ； 音調拓樸性 ； 外聽毛細胞 ； 腺嘌呤核苷三磷酸 ； 動態數學模型

並列摘要

Aminoglycosides, though life-saving and efficient antibiotics, are infamous for their ototoxicity by causing irreversible damage to the cochlear sensory cells. This ototoxicity is known for its tonotopically differential vulnerability in the cochlear: high-frequency, basal turn outer hair cells (OHCs) are preferentially affected. Previous experimental findings suggest that different metabolic biases may vary OHCs’ susceptibility, however, the metabolic mechanisms underlying this high-to-low-frequency propensity are still not fully understood. In this study, we performed in silico experiments to simulate metabolic variations induced by gentamicin (the most commonly prescribed aminoglycoside) in basal and apical OHCs separately. We built a kinetic-metabolic model based on a neuronal-model framework including glycolysis, citric acid cycle, oxidative phosphorylation, ROS production and scavenging systems. Original pathways were modified and new pathways were incorporated to depict hair cell metabolism. The proposed model was then calibrated onto existing experimental data in the literature. Our result of the modeling is that after gentamicin administration, basal turn OHCs suffer from a higher degree of ATP depletion, which correlates with their higher vulnerability to gentamicin.

並列關鍵字

aminoglycoside ； tonotopic ototoxicity ； outer hair cell ； ATP ； dynamic mathematical modeling

參考文獻

[1] K. Ahnert and M. Mulansky. Odeint–solving ordinary differential equations in c++. In AIP Conference Proceedings, volume 1389, pages 1586–1589. American Institute of Physics, 2011.

[2] A. Alharazneh, L. Luk, M. Huth, A. Monfared, P. S. Steyger, A. G. Cheng, and A. J.Ricci. Functional hair cell mechanotransducer channels are required for aminoglycoside ototoxicity. PloS one, 6(7):e22347, 2011.

[3] M. Avent, B. Rogers, A. Cheng, and D. Paterson. Current use of aminoglycosides: indications, pharmacokinetics and monitoring for toxicity. Internal medicine journal, 41(6):441–449, 2011.

[4] J. Bergstra, B. Komer, C. Eliasmith, D. Yamins, and D. D. Cox. Hyperopt: a python library for model selection and hyperparameter optimization. Computational Science Discovery, 8(1):014008, 2015.

[5] N. Berndt, O. Kann, and H.G. Holzhütter. Physiologybased kinetic modeling of neuronal energy metabolism unravels the molecular basis of nad(p)h fluorescence transients. Journal of Cerebral Blood Flow Metabolism, 35(9):1494–1506, 2015.