A Mathematical Model of Bivalent Binding Suggests Physical Trapping of Thrombin within Fibrin Fibers
A Mathematical Model of Bivalent Binding Suggests Physical Trapping of Thrombin within Fibrin Fibers
复制标题
DOI：
10.1016/j.bpj.2019.09.003
复制
发表时间：
2019-10-15
影响因子：
3.4
通讯作者：
Leiderman, Karin
中科院分区：
文献类型：
--
作者：
Kelley, Michael;Leiderman, Karin
凝血酶是一种在血液凝固过程中发挥许多重要作用的酶;它激活血小板，切割反馈回路内的凝固蛋白，并将纤维蛋白原切割成纤维蛋白，纤维蛋白聚合成纤维，在生长的凝块中和周围形成稳定的凝胶基质。凝血酶也与形成的纤维蛋白基质结合，但这种相互作用还不清楚。凝血酶-纤维蛋白结合通常被描述为两个独立的单步结合事件，一个高亲和力和一个低亲和力。然而，动力学方案描述这些单步结合事件不解释实验观察到的纤维蛋白结合凝血酶的驻留时间。在这项工作中，我们研究了一个二价，顺序步骤结合方案作为替代的高亲和力的事件，除了低亲和力的。我们开发的数学模型的单步和连续的步骤计划组成的反应扩散方程相互比较和实验数据。然后，我们使用贝叶斯推理，在马尔可夫链蒙特卡罗的形式，学习模型参数分布从以前公布的实验数据。为了使模型最好地拟合数据，我们做了一个额外的假设，即凝血酶是不可逆的隔离;我们假设这可能是由于凝血酶在纤维蛋白纤维形成时被物理捕获在纤维蛋白纤维内。我们进一步估计，在我们比较模型输出的实验中，类似于30%的凝血酶被物理捕获。物理捕获的凝血酶的概念可能会提供新的见解，关于纤维蛋白溶解的速度相互矛盾的意见。最后，我们表明，我们的新模型可用于进一步探测处理凝血酶变构的情况。
Thrombin is an enzyme that plays many important roles in the blood clotting process; it activates platelets, cleaves coagulation proteins within feedback loops, and cleaves fibrinogen into fibrin, which polymerizes into fibers to form a stabilizing gel matrix in and around growing clots. Thrombin also binds to the formed fibrin matrix, but this interaction is not well understood. Thrombin-fibrin binding is often described as two independent, single-step binding events, one high-affinity and one low-affinity. However, kinetic schemes describing these single-step binding events do not explain experimentally-observed residency times of fibrin-bound thrombin. In this work, we study a bivalent, sequential-step binding scheme as an alternative to the high-affinity event and, in addition to the low-affinity one. We developed mathematical models for the single- and sequential-step schemes consisting of reaction-diffusion equations to compare to each other and to experimental data. We then used Bayesian inference, in the form of Markov chain Monte Carlo, to learn model parameter distributions from previously published experimental data. For the model to best fit the data, we made an additional assumption that thrombin was irreversibly sequestered; we hypothesized that this could be due to thrombin becoming physically trapped within fibrin fibers as they formed. We further estimated that similar to 30% of thrombin in the experiments to which we compare our model output became physically trapped. The notion of physically trapped thrombin may provide new insights into conflicting observations regarding the speed of fibrinolysis. Finally, we show that our new model can be used to further probe scenarios dealing with thrombin allostery.