Development of a three-dimensional unit cell to model the micromechanical response of a collagen-based extracellular matrix.
Development of a three-dimensional unit cell to model the micromechanical response of a collagen-based extracellular matrix.
复制标题
开发三维单位细胞来模拟基于胶原蛋白的细胞外基质的微机械响应。
DOI：
10.1016/j.actbio.2009.11.014
复制
发表时间：
2010
影响因子：
9.7
通讯作者：
Nauman,EricA
中科院分区：
文献类型：
--
作者：
Susilo,MonicaE;Roeder,BlayneA;Voytik-Harbin,SherryL;Kokini,Klod;Nauman,EricA
细胞外基质（ECM）内胶原原纤维的三维微观结构和力学特性目前被认为是调节细胞增殖和分化的主要因素。因此，对细胞与其ECM相互作用的机械方面的理解对于工程组织的发展是必要的。最终，使用这些相互作用来设计组织等效物需要具有三维结构的数学模型。在本研究中，利用细胞固体建立了经受单轴拉伸应力的胶原原纤维基质的三维模型。选择由细支柱组成的结构来代表工程ECM内胶原原纤维的排列。为了解释组织的大变形，胶原原纤维被建模为超弹性neo-Hookean或Mooney-Rivlin材料。细胞状固体的使用使得纤维的性质以封闭的形式与ECM的性质相关，这反过来又允许使用ECM实验数据来估计纤维的性质。一组先前获得的实验数据包括纤维微观结构和力学测试的同时测量，用于评估模型在给定组织尺度数据时估计胶原纤维力学性能的能力，以及在给定估计纤维刚度时预测组织尺度力学性能的能力。当纤维被建模为neo-Hookean和Mooney-Rivlin材料时，纤维的切模量分别为1.26±0.70和1.62±0.88MPa。各组间纤维切模量的估计差异无统计学意义。灵敏度分析表明，纤维力学性能和体积分数是两个需要精确值的输入参数。虽然从凝胶的初始图像中很容易获得体积分数，但纤维的力学性能却不易获得。因此，在留一交叉验证（LOOCV）分析中估计纤维的力学性能。LOOCV分析表明，该模型能够预测ECM应力-拉伸曲线，平均均方根误差为9.71kPa2。三维结构扩展了以前的连续体模型和二维表示，为研究ECM微观结构对细胞功能的分层效应提供了一个有用的模型。该模型可以作为设计工具来设计细胞功能的最佳微观结构。
The three-dimensional microstructure and mechanical properties of the collagen fibrils within the extracellular matrix (ECM) is now being recognized as a primary factor in regulating cell proliferation and differentiation. Therefore, an appreciation of the mechanical aspects by which a cell interacts with its ECM is required for the development of engineered tissues. Ultimately, using these interactions to design tissue equivalents requires mathematical models with three-dimensional architecture. In this study, a three-dimensional model of a collagen fibril matrix undergoing uniaxial tensile stress was developed by making use of cellular solids. A structure consisting of thin struts was chosen to represent the arrangement of collagen fibrils within an engineered ECM. To account for the large deformation of tissues, the collagen fibrils were modeled as hyperelastic neo-Hookean or Mooney–Rivlin materials. The use of cellular solids allowed the fibril properties to be related to the ECM properties in closed form, which, in turn, allowed the estimation of fibril properties using ECM experimental data. A set of previously obtained experimental data consisting of simultaneous measures of the fibril microstructure and mechanical tests was used to evaluate the model’s capability to estimate collagen fibril mechanical property when given tissue-scale data and to predict the tissue-scale mechanical properties when given estimated fibril stiffness. The fibril tangent modulus was found to be 1.26±0.70 and 1.62±0.88MPa when the fibril was modeled as neo-Hookean and Mooney–Rivlin material, respectively. There was no statistical significance of the estimated fibril tangent modulus among the different groups. Sensitivity analysis showed that the fibril mechanical properties and volume fraction were the two input parameters which required accurate values. While the volume fraction was easily obtained from the initial image of the gel, the fibril mechanical properties were not readily available. Therefore the fibril mechanical properties were estimated in the leave-one-out cross-validation (LOOCV) analysis. The LOOCV analysis showed that the model was able to predict the ECM stress–stretch curve with an average mean squared error of 9.71kPa2. The three-dimensional architecture expands on previous continuum models and two-dimensional representations to provide a useful model for studying the hierarchical effects of ECM microstructure on cell function. This model can be used as a design tool to engineer the optimum microstructure for cells to function.