Experimental and computational assessment of F-actin influence in regulating cellular stiffness and relaxation behaviour of fibroblasts.

In biomechanics, a complete understanding of the structures and mechanisms that regulate cellular stiffness at a molecular level remain elusive. In this paper, we have elucidated the role of filamentous actin (F-actin) in regulating elastic and viscous properties of the cytoplasm and the nucleus. Specifically, we performed colloidal-probe atomic force microscopy (AFM) on BjhTERT fibroblast cells incubated with Latrunculin B (LatB), which results in depolymerisation of F-actin, or DMSO control. We found that the treatment with LatB not only reduced cellular stiffness, but also greatly increased the relaxation rate for the cytoplasm in the peripheral region and in the vicinity of the nucleus. We thus conclude that F-actin is a major determinant in not only providing elastic stiffness to the cell, but also in regulating its viscous behaviour. To further investigate the interdependence of different cytoskeletal networks and cell shape, we provided a computational model in a finite element framework. The computational model is based on a split strain energy function of separate cellular constituents, here assumed to be cytoskeletal components, for which a composite strain energy function was defined. We found a significant influence of cell geometry on the predicted mechanical response. Importantly, the relaxation behaviour of the cell can be characterised by a material model with two time constants that have previously been found to predict mechanical behaviour of actin and intermediate filament networks. By merely tuning two effective stiffness parameters, the model predicts experimental results in cells with a partly depolymerised actin cytoskeleton as well as in untreated control. This indicates that actin and intermediate filament networks are instrumental in providing elastic stiffness in response to applied forces, as well as governing the relaxation behaviour over shorter and longer time-scales, respectively.

Influence of dispersion in myosin filament orientation and anisotropic filament contractions in smooth muscle.

A new constitutive model for the biomechanical behaviour of smooth muscle tissue is proposed. The active muscle contraction is accomplished by the relative sliding between actin and myosin filaments, comprising contractile units in the smooth muscle cells. The orientation of the myosin filaments, and thereby the contractile units, are taken to exhibit a statistical dispersion around a preferred direction. The number of activated cross-bridges between the actin and myosin filaments governs the contractile force generated by the muscle and also the contraction speed. A strain-energy function is used to describe the mechanical behaviour of the smooth muscle tissue. Besides the active contractile apparatus, the mechanical model also incorporates a passive elastic part. The constitutive model was compared to histological and isometric tensile test results for smooth muscle tissue from swine carotid artery. In order to be able to predict the active stress at different muscle lengths, a filament dispersion significantly larger than the one observed experimentally was required. Furthermore, a comparison of the predicted active stress for a case of uniaxially oriented myosin filaments and a case of filaments with a dispersion based on the experimental histological data shows that the difference in generated stress is noticeable but limited. Thus, the results suggest that myosin filament dispersion alone cannot explain the increase in active muscle stress with increasing muscle stretch.

Modeling of fibroblast-controlled strengthening and remodeling of uniaxially constrained collagen gels.

A theoretical model for the remodeling of collagen gels is proposed. The collagen fabric is modeled as a network of collagen fibers, which in turn are composed of collagen fibrils. In the model, the strengthening of collagen fabric is accomplished by fibroblasts, which continuously recruit and attach more collagen fibrils to existing collagen fibers. The fibroblasts also accomplish a reorientation of collagen fibers. Fibroblasts are assumed to reorient collagen fibers toward the direction of maximum material stiffness. The proposed model is applied to experiments in which fibroblasts were inserted into a collagen gel. The model is able to predict the force-strain curves for the experimental collagen gels, and the final distribution of collagen fibers also agrees qualitatively with the experiments.

A calcium-driven mechanochemical model for prediction of force generation in smooth muscle.

A new model for the mechanochemical response of smooth muscle is presented. The focus is on the response of the actin-myosin complex and on the related generation of force (or stress). The chemical (kinetic) model describes the cross-bridge interactions with the thin filament in which the calcium-dependent myosin phosphorylation is the only regulatory mechanism. The new mechanical model is based on Hill's three-component model and it includes one internal state variable that describes the contraction/relaxation of the contractile units. It is characterized by a strain-energy function and an evolution law incorporating only a few material parameters with clear physical meaning. The proposed model satisfies the second law of thermodynamics. The results of the combined coupled model are broadly consistent with isometric and isotonic experiments on smooth muscle tissue. The simulations suggest that the matrix in which the actin-myosin complex is embedded does have a viscous property. It is straightforward for implementation into a finite element program in order to solve more complex boundary-value problems such as the control of short-term changes in lumen diameter of arteries due to mechanochemical signals.

On the correlation between continuum mechanics entities and cell activity in biological soft tissues: assessment of three possible criteria for cell-controlled fibre reorientation in collagen gels and collagenous tissues.

The biomechanical behaviour of biological cells is of great importance in many physiological processes. One such process is the maintenance of fibrous networks, such as collagenous tissues. The activity of the fibre-producing cells in this type of tissue is very important, and a comprehensive material description needs to incorporate the activity of the cells. In biomechanics, continuum mechanics is often employed to describe deforming solids, and modelling can be much simplified if continuum mechanics entities, such as stress and strain, can be correlated with cell activity. To investigate this, a continuum mechanics framework is employed in which remodelling of a collagen gel is modelled. The remodelling is accomplished by fibroblasts, and the activity of the fibroblasts is linked to the continuum mechanics theory. The constitutive model for the collagen fabric is formulated in terms of a strain energy function, which includes a density function describing the distribution of the collagen fibre orientation. This density function evolves according to an evolution law, where fibroblasts reorient fibres towards the direction of increasing Cauchy stress, elastic deformation, or stiffness. The theoretical framework is applied to experimental results from collagen gels, where gels have undergone remodelling under both biaxial and uniaxial constraint. The analyses indicated that criteria 1 and 2 (Cauchy stress and elastic deformations) are able to predict the collagen fibre distribution after remodelling, whereas criterion 3 (current stiffness) is not. This conclusion is, however, tentative and pertains, strictly speaking, only to fibre remodelling processes, and may not be valid for other types of cell activities.

A constitutive model for smooth muscle including active tone and passive viscoelastic behaviour.

A new constitutive model for the biomechanical behaviour of smooth muscle tissue is proposed. The active muscle contraction is accomplished by the relative sliding between actin and myosin filaments, comprising contractile units in the smooth muscle cells. The model includes a chemical part, governing the cross-bridge (myosin head) cycling, that is responsible for the filament sliding. The number of activated cross-bridges govern the contractile force generated and also the contraction speed. A strain-energy function is used to describe the mechanical behaviour of the smooth muscle tissue. Besides the active contractile apparatus, the mechanical model also incorporates a passive viscoelastic part. The constitutive model was calibrated with respect to experiments on smooth muscle tissue from swine carotid artery and guinea pig taenia coli, in terms of isometric and isotonic tensile test results. The model was fully able to reproduce the experimental results.

Influence of medial collagen organization and axial in situ stretch on saccular cerebral aneurysm growth.

A model for saccular cerebral aneurysm growth, proposed by Kroon and Holzapfel (2007, "A Model for Saccular Cerebral Aneurysm Growth in a Human Middle Cerebral Artery," J. Theor. Biol., 247, pp. 775-787; 2008, "Modeling of Saccular Aneurysm Growth in a Human Middle Cerebral Artery," ASME J. Biomech. Eng., 130, p. 051012), is further investigated. A human middle cerebral artery is modeled as a two-layer cylinder where the layers correspond to the media and the adventitia. The immediate loss of media in the location of the aneurysm is taken to be responsible for the initiation of the aneurysm growth. The aneurysm is regarded as a development of the adventitia, which is composed of several distinct layers of collagen fibers perfectly aligned in specified directions. The collagen fibers are the only load-bearing constituent in the aneurysm wall; their production and degradation depend on the stretch of the wall and are responsible for the aneurysm growth. The anisotropy of the surrounding media was modeled using the strain-energy function proposed by Holzapfel et al. (2000, "A New Constitutive Framework for Arterial Wall Mechanics and a Comparative Study of Material Models," J. Elast., 61, pp. 1-48), which is valid for an elastic material with two families of fibers. It was shown that the inclusion of fibers in the media reduced the maximum principal Cauchy stress and the maximum shear stress in the aneurysm wall. The thickness increase in the aneurysm wall due to material growth was also decreased. Varying the fiber angle in the media from a circumferential direction to a deviation of 10 deg from the circumferential direction did, however, only show a little effect. Altering the axial in situ stretch of the artery had a much larger effect in terms of the steady-state shape of the aneurysm and the resulting stresses in the aneurysm wall. The peak values of the maximum principal stress and the thickness increase both became significantly higher for larger axial stretches.

A theoretical model for fibroblast-controlled growth of saccular cerebral aneurysms.

A new theoretical model for the growth of saccular cerebral aneurysms is proposed by extending the recent constitutive framework of Kroon and Holzapfel [2007a. A model for saccular cerebral aneurysm growth by collagen fibre remodelling. J. Theor. Biol. 247, 775-787]. The continuous turnover of collagen is taken to be the driving mechanism in aneurysmal growth. The collagen production rate depends on the magnitude of the cyclic deformation of fibroblasts, caused by the pulsating blood pressure during the cardiac cycle. The volume density of fibroblasts in the aneurysmal tissue is taken to be constant throughout the growth process. The growth model is assessed by considering the inflation of an axisymmetric membranous piece of aneurysmal tissue, with material characteristics representative of a cerebral aneurysm. The diastolic and systolic states of the aneurysm are computed, together with its load-free state. It turns out that the value of collagen pre-stretch, that determines growth speed and stability of the aneurysm, is of pivotal importance. The model is able to predict aneurysms with typical berry-like shapes observed clinically, and the predicted wall stresses correlate well with the experimentally obtained ultimate stresses of this type of tissue. The model predicts that aneurysms should fail when reaching a size of about 1.2-3.6mm, which is smaller than what has been clinically observed. With some refinements, the model may, however, be used to predict future growth of diagnosed aneurysms.

Modeling of saccular aneurysm growth in a human middle cerebral artery.

Saccular aneurysm growth in a human middle cerebral artery is modeled. The aneurysm growth model was presented in a companion paper by Kroon and Holzapfel ("A Model for Saccular Cerebral Aneurysm Growth by Collagen Fibre Remodelling," J. Theor. Biol., in press) and was assessed there for axisymmetric growth. The aneurysm growth model is now evaluated for a more realistic setting. The middle cerebral artery is modeled as a two-layered cylinder, where the layers correspond to the media and the adventitia. An instant loss of the media in a region of the artery wall initiates the growth of the saccular aneurysm. The aneurysm wall is assumed to be a development of the adventitia of the original healthy artery, and collagen is assumed to be the only load-bearing constituent in the adventitia and in the aneurysm wall. The collagen is organized in a number of distinct layers where fibers in a specific layer are perfectly aligned in a certain fiber direction. The production of new collagen is taken to depend on the stretching of the aneurysm wall, and the continuous remodeling of the collagen fibers is responsible for the aneurysm growth. The general behavior of the growth model is investigated and also the influence of the structural organization of the collagen fabric. The analysis underlines the fact that the material behavior of aneurysmal tissue cannot be expected to be isotropic. The model predictions agree well with clinical and experimental results, for example, in terms of aneurysm size and shape, wall stress levels, and wall thickness.

A new constitutive model for multi-layered collagenous tissues.

Collagenous tissues such as the aneurysmal wall or the aorta are multi-layered structures with the mean fibre alignments distinguishing one layer from another. A constitutive representation of the multiple collagen layers is not yet developed, and hence the aim of the present study. The proposed model is based on the constitutive theory of finite elasticity and is characterized by an anisotropic strain-energy function which takes the material structure into account. The passive tissue behaviour is modelled and the related mechanical response is assumed to be dominated by elastin and collagen. While elastin is modelled by the neo-Hookean material the constitutive response of collagen is assumed to be transversely isotropic for each individual layer and based on an exponential function. The proposed constitutive function is polyconvex which ensures material stability. The model has five independent material parameters, each of which has a clear physical interpretation: the initial stiffnesses of the collagen fabric in the two principal directions, the shear modulus pertaining to the non-collagenous matrix material, a parameter describing the level of nonlinearity of the collagen fabric, and the angle between the principal directions of the collagen fabric and the reference coordinate system. An extension-inflation test of the adventitia of a human femoral artery is simulated by means of the finite element method and an error function is minimized by adjusting the material parameters yielding a good agreement between the model and the experimental data.

A model for saccular cerebral aneurysm growth by collagen fibre remodelling.

The first structural model for saccular cerebral aneurysm growth is proposed. It is assumed that the development of the aneurysm is accompanied by a loss of the media, and that only collagen fibres provide load-bearing capacity to the aneurysm wall. The aneurysm is modelled as an axisymmetric multi-layered membrane, exposed to an inflation pressure. Each layer is characterized by an orientation angle, which changes between different layers. The collagen fibres and fibroblasts within a specific layer are perfectly aligned. The growth and the morphological changes of the aneurysm are accomplished by the turnover of collagen. Fibroblasts are responsible for collagen production, and the related deformations are assumed to govern the collagen production rate. There are four key parameters in the model: a normalized pressure, the number of layers in the wall, an exponent in the collagen mass production rate law, and the pre-stretch under which the collagen is deposited. The influence of the model parameters on the aneurysmal response is investigated, and a stability analysis is performed. The model is able to predict clinical observations and mechanical test results, for example, in terms of predicted aneurysm size, shape, wall stress and wall thickness.