Research Interests
My research focuses on understanding and controlling phase transitions and transport properties of macromolecular solutions relevant to biotechnology and medicine. These studies involve techniques such as interferometry, static and dynamic light scattering, microscopy and chromatography. Below I summarize some relevant aspects.
Aggregation of Macromolecules in Solution. Macromolecules (proteins, polymers) in solution are subject to crystallization, liquid-liquid phase transition, self-assembly, aggregation and gelation. These processes are condensations of macromolecules because they lead to the formation of a new phase rich in macromolecules. Understanding and controlling condensation is important for protein-crystal growth, protein-aggregation diseases and the processing and use of biomaterials. One aspect of our research is to investigate the effect of isothermal protein oligomerization on protein condensation in aqueous solutions. We find that additives such as salts or polymers can be used to rationally design the morphology of the protein aggregate into microspheres or amorphous aggregates under non-denaturing conditions: Wang et al., Langmuir, 24, 2799 (2008).
We also investigate condensation in other systems such as dendrimer solutions. Dendrimers are a class of globular branched macromolecules, which can be used as uni-molecular micelles, catalytic devices, drug delivery systems, protein mimics and building blocks for the formation of larger supra-molecular and super-molecular structures. Specific types and concentration of additives bring about reversible dendrimer condensation in aqueous solutions. Finally, we also use additives to induce and control the formation of nano/micro gels from aqueous solutions using biocompatible linear polymers such as poly(vinyl alcohol).
DENDRIMER
Macromolecular Crowding. Macromolecular crowding is a property of multicomponent solutions that favors phase separation by excluded-volume interactions between two macromolecules with different size.
For protein systems, this phenomenon is important because 1) it affects protein self-assembly in crowded environments such as cell cytoplasm; 2) it is implicated in protein crystallization; 3) it can be used for the rationale design of protein-based materials for enzymology and drug delivery (see above). Polymers are usually added to induce phase separation in protein solutions by macromolecular crowding. However protein-polymer interactions are not well understood and simple excluded-volume models have been only qualitatively successful in describing protein-polymer interactions. We experimentally investigate protein-polymer interactions by inducing and characterizing liquid-liquid phase separation (LLPS) in protein-polymer aqueous mixtures. The corresponding phase boundary is represented by a surface in the temperature-concentration phase diagram.
PHASE DIAGRAM
Accurate microscopic models must be consistent with two thermodynamically independent effects of polymer on the LLPS phase boundary: 1) the effect of polymer concentration on the LLPS temperature (red line in the figure), 2) protein/polymer partitioning in the two liquid coexisting phases (green lines in the figure). Thus models can be built to verify both observables: Wang et al., J. Phys. Chem. B, 111, 1222 (2007). Using the obtained experimental results, we design accurate models for macromolecular crowding in protein-polymer systems. One important aspect of our microscopic models is to take into account the chemical structure of polymer coils. This includes 1) bond length and angles and 2) segment-segment conformational interaction in order to fit the experimentally known relation between the polymer radius of gyration and its molecular weight. Our primary theoretical tools are thermodynamic perturbation theory and Monte Carlo simulations.
Macromolecule-Additive Interactions by Ternary Diffusion. Measurements of diffusion coefficients are typically used to predict and model the kinetics of several in vivo, laboratory, medical, pharmaceutical and manufacturing applications. Diffusion in multicomponent mixtures is usually approximated by defining one single diffusion coefficient for each solute, where the diffusion coefficient relates the concentration gradient of the solute to its flux (Fick’s law). This is however only an incomplete description. In fact, if a solution contains more than two components, the fluxes of the components are also coupled with each other. For instance, in the case of a ternary solution (two solutes and one solvent) a matrix of four diffusion coefficients is required: two main diffusion coefficients which describe the flux of the two solutes due to their own concentration gradients and two cross diffusion coefficients which describe the flux of each solute due to the gradient of the other solute. The determination of the diffusion coefficient matrix requires the use of a highly precise technique such as optical interferometry. Next figure shows the Rayleigh interferometric pattern created by the diffusion mixing of two aqueous solutions with different solute concentrations. At TCU, we are lucky to have the most precise optical diffusiometer in the world, the Gosting diffusiometer.
Rayleigh Intereference
Pattern
We have developed a novel method that uses ternary diffusion coefficients to determine the effect of an additive concentration and type on the chemical potential of a macromolecule. This in turn provides information on the charge and solvation of macromolecules. In this way, we investigate the effect of salts and osmolytes on proteins and polymers in solutions: Tan et al., J. Phys. Chem. B, 112, 4967 (2008).
Protein Solubility and Crystal Thermodynamics. Understanding protein solubility is important for a rational design of the conditions of protein crystallization. We focus on the experimental investigation of the effect of salts on protein solubility. Our solubility results are directly compared to thermodynamic parameters of the liquid phase obtained by ternary diffusion (see above). Preferential hydration and common-ion effects (Donnan effect) are the main factors influencing protein-salt interactions in the liquid phase and the dependence of protein solubility as a function of salt interactions. Comparison between thermodynamic parameters of the liquid phase and protein solubility demonstrate that the solubility dependence on salt concentration is substantially affected by the corresponding change of protein chemical potential in the crystalline phase. We introduce models for the crystalline phase based on salt partitioning between solution and the hydrated protein crystal. This allows us to derive novel solubility equation that quantitatively explain the observed experimental dependence of protein solubility on salt concentration: Annunziata et al., J. Am. Chem. Soc., 130, 13347 (2008).
Modulation of Drug Transport by Multicomponent Diffusion. We are interested in using the concept of coupled diffusion to design novel kinetic profiles of drug release. In several systems, the release of the drug occurs by diffusion of the drug through a microporous media. In these cases the flux of the drug at the surface of the delivery media and its time dependence are the factors to be controlled. Kinetic models for diffusion-mediated release of drug from delivery devices are derived from Fick's diffusion laws. An optimal drug release behavior requires a controllable release rate in many cases. The introduction of another diffusible component in the drug delivery device can act on the flux of the drug by coupled diffusion and thus regulate drug release profiles.
We have focused in the characterization of ternary diffusion in drug-surfactant-water systems. In this way we learn on how the concentration and the concentration gradient of micelles affect drug diffusion: Zhang et al., Langmuir, 24, 10680 (2008).