Effect of Electrospun Fiber Mat Thickness and Support Method on Cell Morphology.

Electrospun fiber mats (EFMs) are highly versatile biomaterials used in a myriad of biomedical applications. Whereas some facets of EFMs are well studied and can be highly tuned (e.g., pore size, fiber diameter, etc.), other features are under characterized. For example, although substrate mechanics have been explored by several groups, most studies rely on Young's modulus alone as a characterization variable. The influence of fiber mat thickness and the effect of supports are variables that are often not considered when evaluating cell-mechanical response. To assay the role of these features in EFM scaffold design and to improve understanding of scaffold mechanical properties, we designed EFM scaffolds with varying thickness (50-200 µm) and supporting methodologies. EFM scaffolds were comprised of polycaprolactone and were either electrospun directly onto a support, suspended across an annulus (3 or 10 mm inner diameter), or "tension-released" and then suspended across an annulus. Then, single cell spreading (i.e., Feret diameter) was measured in the presence of these different features. Cells were sensitive to EFM thickness and suspended gap diameter. Overall, cell spreading was greatest for 50 µm thick EFMs suspended over a 3 mm gap, which was the smallest thickness and gap investigated. These results are counterintuitive to conventional understanding in mechanobiology, which suggests that stiffer materials, such as thicker, supported EFMs, should elicit greater cell polarization. Additional experiments with 50 µm thick EFMs on polystyrene and polydimethylsiloxane (PDMS) supports demonstrated that cells can "feel" the support underlying the EFM if it is rigid, similar to previous results in hydrogels. These results also suggest that EFM curvature may play a role in cell response, separate from Young's modulus, possibly because of internal tension generated. These parameters are not often considered in EFM design and could improve scaffold performance and ultimately patient outcomes.

Beyond Linear Elastic Modulus: Viscoelastic Models for Brain and Brain Mimetic Hydrogels.

With their high degree of specificity and investigator control, in vitro disease models provide a natural complement to in vivo models. Especially in organs such as the brain, where anatomical limitations make in vivo experiments challenging, in vitro models have been increasingly used to mimic disease pathology. However, brain mimetic models may not fully replicate the mechanical environment in vivo, which has been shown to influence a variety of cell behaviors. Specifically, many disease models consider only the linear elastic modulus of brain, which describes the stiffness of a material with the assumption that mechanical behavior is independent of loading rate. Here, we characterized porcine brain tissue using a modified stress relaxation test, and across a panel of viscoelastic models, showed that stiffness depends on loading rate. As such, the linear elastic modulus does not accurately reflect the viscoelastic properties of native brain. Among viscoelastic models, the Maxwell model was selected for further analysis because of its simplicity and excellent curve fit (R2 = 0.99 ± 0.0006). Thus, mechanical response of native brain and hydrogel mimetic models was analyzed using the Maxwell model and the linear elastic model to evaluate the effects of strain rate, time post mortem, region, tissue type (i.e., bulk brain vs white matter), and in brain mimetic models, hydrogel composition, on observed mechanical properties. In comparing the Maxwell and linear elastic models, linear elastic modulus is consistently lower than the Maxwell elastic modulus across all brain regions. Additionally, the Maxwell model is sensitive to changes in viscosity and small changes in elasticity, demonstrating improved fidelity. These findings demonstrate the insufficiency of linear elastic modulus as a primary mechanical characterization for brain mimetic materials and provide quantitative information toward the future design of materials that more closely mimic mechanical features of brain.

Exploring the mechanical behavior of degrading swine neural tissue at low strain rates via the fractional Zener constitutive model.

The ability of the fractional Zener constitutive model to predict the behavior of postmortem swine brain tissue was examined in this work. Understanding tissue behavior attributed to degradation is invaluable in many fields such as the forensic sciences or cases where only cadaveric tissue is available. To understand how material properties change with postmortem age, the fractional Zener model was considered as it includes parameters to describe brain stiffness and also the parameter α, which quantifies the viscoelasticity of a material. The relationship between the viscoelasticity described by α and tissue degradation was examined by fitting the model to data collected in a previous study (Bentil, 2013). This previous study subjected swine neural tissue to in vitro unconfined compression tests using four postmortem age groups (<6h, 24h, 3 days, and 1 week). All samples were compressed to a strain level of 10% using two compressive rates: 1mm/min and 5mm/min. Statistical analysis was used as a tool to study the influence of the fractional Zener constants on factors such as tissue degradation and compressive rate. Application of the fractional Zener constitutive model to the experimental data showed that swine neural tissue becomes less stiff with increased postmortem age. The fractional Zener model was also able to capture the nonlinear viscoelastic features of the brain tissue at low strain rates. The results showed that the parameter α was better correlated with compressive rate than with postmortem age.

Constitutive modeling of rate-dependent stress-strain behavior of human liver in blunt impact loading.

An understanding of the mechanical deformation behavior of the liver under high strain rate loading conditions could aid in the development of vehicle safety measures to reduce the occurrence of blunt liver injury. The purpose of this study was to develop a constitutive model of the stress-strain behavior of the human liver in blunt impact loading. Experimental stress and strain data was obtained from impact tests of 12 unembalmed human livers using a drop tower technique. A constitutive model previously developed for finite strain behavior of amorphous polymers was adapted to model the observed liver behavior. The elements of the model include a nonlinear spring in parallel with a linear spring and nonlinear dashpot. The model captures three features of liver stress-strain behavior in impact loading: (1) relatively stiff initial modulus, (2) rate-dependent yield or rollover to viscous "flow" behavior, and (3) strain hardening at large strains. Six material properties were used to define the constitutive model. This study represents a novel application of polymer mechanics concepts to understand the rate-dependent large strain behavior of human liver tissue under high strain rate loading. Applications of this research include finite element simulations of injury-producing liver or abdominal impact events.

Mechanical characterization of electrospun polycaprolactone (PCL): a potential scaffold for tissue engineering.

This paper investigates the mechanical behavior of electrospun polycaprolactone (PCL) under tensile loading. PCL in bulk form degrades slowly and is biocompatible, two properties that make it a viable option for tissue engineering applications in biomedicine. Of particular interest is the use of electrospun PCL tubes as scaffolds for tissue engineered blood vessel implants. Stress relaxation and tensile tests have been conducted with specimens at room temperature (21 degrees C) and 37 degrees C. Additionally, to probe the effects of moisture on mechanical behavior, specimens were tested either dry (in air) or submerged in water. In general, the electrospun PCL was found to exhibit rate dependence, as well as some dependence on the test temperature and on whether the sample was wet or dry. Two different models were investigated to describe the experimentally observed material behavior. The models used were Fung's theory of quasilinear viscoelasticity (QLV) and the eight-chain model developed for rubber elastomers by Arruda and Boyce (1993, "A Three-Dimensional Constitutive Model for the Large Stretch Behavior of Rubber Elastic Materials," J. Mech. Phys. Solids, 41(2), pp. 389-412). The implementation and fitting results, as well as the advantages and disadvantages of each model, are presented. In general, it was found that the QLV theory provided a better fit.

Using pressure to predict liver injury risk from blunt impact.

Liver trauma research suggests that rapidly increasing internal pressure plays a role in causing blunt liver injury. Knowledge of the relationship between pressure and the likelihood of liver injury could be used to enhance the design of crash test dummies. The objectives of this study were (1) to characterize the relationship between impact-induced pressures and blunt liver injury in an experimental model to impacts of ex vivo organs; and (2) to compare human liver vascular pressure and tissue pressure in the parenchyma with other biomechanical variables as predictors of liver injury risk. Test specimens were 14 ex vivo human livers. Specimens were perfused with normal saline solution at physiological pressures, and a drop tower applied blunt impact at varying energies. Impact-induced pressures were measured by transducers inserted into the hepatic veins and the parenchyma (caudate lobe) of ex vivo specimens. Experimentally induced liver injuries were consistent with those documented in the Crash Injury Research and Engineering Network (CIREN) database. Binary logistic regression analysis demonstrated that injury predictors associated with tissue pressure measured in the parenchyma were the best indicators of serious liver injury risk. The best injury predictor overall was the product of the peak rate of tissue pressure increase and the peak tissue pressure, P T max * P T max (pseudo-R2 = .82, p = .001). A burst injury mechanism directly related to hydrostatic pressure is postulated for the ex vivo liver loaded dynamically in a drop test experiment.