(395b) Transient Elasticity Gradients for Studying Cell Mobility

Introduction Investigations of mechanotransduction and cell mobility often utilize gels with mechanical property gradients. Poly(acrylamide) is generally the polymer of choice for such studies and is covalently crosslinked using photopolymerization. Crosslink density gradients are achieved using photomasks. The resulting gel has irreversible crosslinks and, therefore, a fixed elasticity gradient. However, toxicity of these materials may be a concern due to the presence of unreacted monomers. Furthermore, the gradient magnitude is limited by the nature of the photomask. Since mechanotransduction and cell mobility are obviously dynamic in nature, detailed study may require transient substrates with larger elastic gradients. We have developed a novel method for producing hydrogels with transient mechanical property gradients using a biopolymer called gellan. Gellan is a naturally derived polysaccharide which is capable of gelling at very low concentrations (<0.1 wt %) due to conformational changes caused by temperature or cation concentration effects. Gellan exists as a random coil at low temperatures in water, and changes to a triple helix conformation upon heating and/or the addition of cations. In this helical conformation, the gellan molecules form ionic intermolecular crosslinks. Cell behaviors are generally of interest only at physiological conditions (i.e. 37°C). As the coil-to-helix transformation occurs in the presence of cations, we utilized mass transport of cations to manipulate the crosslink density of these gels; however, thermal control may be used in a similar fashion. Methods A diffusion chamber was constructed in a cylindrical shape. One end of the cylinder was sealed, while the other was covered with a dialysis membrane (molecular weight cutoff 3500 Da). This prevented the gellan (MW 500 kDa) from leaving the chamber, while presenting little resistance to the diffusion of small cations such as Ca2+ and Na+. The chamber was filled with solutions of low-acyl gellan at various concentrations. Cation solutions of various concentrations were pumped across the membrane, then allowed to passively diffuse into the chamber based on the concentration gradient, for 24 hours. The resulting gel was then removed from the mold. A microindentation device was used to measure the elastic modulus of the gel along its length. The concentrations of gellan and cation solutions were determined using a statistical experimental design. Elastic modulus as a function of position, as well as the effective diffusivity, were taken as responses. This allowed rapid correlation between the independent concentrations and the dependent responses with a minimum number of experiments. A second diffusion chamber with slab geometry was constructed with dialysis membranes at both ends. One side of the slab was removable to allow determination of the mechanical property gradient in situ. Solutions of cations or deionized water were pumped across either end to control the local concentrations of cations within the chamber. Gels were again created at various gellan concentrations. Indentation was performed at 24 hours. After indentation, boundary concentrations were altered and indentation was performed after 24 hours. Thus, reversibility of crosslinking and the transient nature of the mechanical property gradient could be assessed. A 1-D diffusion model was developed to relate the boundary concentrations, initial gellan concentration, and the resulting elastic modulus profile. The model was trained using the results of the experimental design in the cylindrical diffusion chamber. It was then used to predict transient effects in the slab diffusion chamber. Results The effective diffusivity of cations decreased as gellan concentration increased (Fig. 1). This was a result of gellan-cation binding, which effectively inhibited diffusion. The elastic modulus increased as a power law function of the gellan and cation concentrations. Gels had elastic moduli from 500-15000 Pa, though stiffer gels could be formulated simply by increasing the concentration of gellan. The diffusion model predicted the transient behavior of the gel elastic modulus profile under all conditions. Conclusions This system is able to reproduce many diverse elastic gradients which may be of interest to those studying mechano-transduction and cell mobility. The diffusion model allows prediction and control of the elasticity profile of the substrate at all times, thus allowing the detailed study of the dynamics of such systems.

>