|
|
Prof. Seth Fraden
Dept. of Physics, MS 057
Brandeis University
Waltham, MA USA 02454
Phone (781) 736-2888
Fax (781) 736-2915
My group studies macromolecules in suspension and we explore the relation between the macromolecules' interparticle interactions and the macroscopic properties of the suspensions. In our research we utilize a variety of methods and approaches covering experiment, theory, and simulation. Areas of interest are described below:
- Electro-rheological fluids are composed of dielectric spheres suspended in a solvent to which an electric field is applied. Under certain conditions the suspension rapidly solidifies. These materials are being developed for applications in the automotive industry. The microscopic structure is being studied using digital video optical microscopy, confocal laser scanning microscopy and light scattering.
- We study entropy driven phase transitions in colloidal liquid crystals of rod-like virus particles. The virus particles are very interesting because Nature has engineered them to be monodisperse -- which has not been achieved in the laboratory. Because of the simplicity of interparticle interactions there is hope that a microscopic theory of the phase behavior and bulk properties can be undertaken. These viral colloidal systems offer the possibility of serving as model systems for understanding all liquids composed of anisotropic particles. We have studied phase behavior of nematic, cholesteric, smectic, and colloidal crystals of virus particles. Currently we are modifying the viruses using a combination of genetic engineering to make viruses of different lengths, and chemistry to modify the diameter of the particles by either coating them with metal or bonding water soluble polymers, such as PEO to the surface of the virus.
- Macromolecular Crowding is a project of direct biological relevance. Macromolecules occupy 30% of the volume of the cell, strongly influencing inter-molecular interactions. Even in the absence of any direct interactions between particles, such as electrostatic, hydrophobic, or van der Waals forces, the macromolecules feel each others presence simply because two molecules cannot occupy the same place at the same time. This crowding of molecules causes like species to phase separate into different regions of the cell, leading to macromolecular compartmentalization without the need for any intracellular membranes. Strong partitioning and bundling is observed for rodlike biopolymers such as actin or microtubules when mixed with globular proteins or polymers.
The biochemistry of the cell has evolved in this crowded, thermodynamically non-ideal environment, and it may be that the cell has exploited this fact. Prof. Herzfeld [Accounts of Chemical Research 29, 31-37 (1996)] has proposed that that bundling of filaments is driven by macromolecular crowding and that the role of specific bundling proteins is to fine tune certain aspects of the bundling, like the relative alignment of the biopolymers in a bundle. There are a number of questions we are seeking to answer. How crowded do suspensions have to be in order to partition? How strong is the degree of partitioning of the species? What is the organization of the macromolecules in the partitioned phases? How different do the species have to be from each other for this to occur? How does the interparticle potential affect the phenomena? To what extent is all this relevant to cellular biology?
We are studying the partitioning of mixtures of globular and filamentous proteins in vitro using suspensions of the biopolymer fd bacteriophage and Tobacco Mosaic Virus mixed with polymers such as polyethylene glycol and dextran or globular proteins like BSA. Genetic engineering methods are used to systematically alter the length of the biopolymers, an important thermodynamic variable. The viruses are labeled with fluorescent dyes and equilibrated samples are observed in the light microscope. Electron microscopy, light and x-ray scattering, and optical microscopy are used to determine the structure of the macromolecular suspension. We compare our experimental measurements of the phase behavior of these mixtures with several theoretical statistical mechanical models developed by ourselves, by Prof. J. Herzfeld of the Brandeis University Chemistry department and others, as well as with Monte Carlo computer simulations. Our recent results are available online.
- Protein Crystallization. The goal of the human genome project is to sequence the genes that code the approximately 30,000 proteins that comprise the human body. While much valuable information is contained in a protein's genetic code, it is often desirable to obtain the three-dimensional structure of the protein in order to elucidate its function. The first step in structure determination is obtaining purified protein, either directly through biochemical separation processes, or if the DNA sequence is known, from expression of the protein using genetically modified organisms. The next step is crystallizing the protein, and the final step is solving the protein structure from x-ray diffraction measurements. However, the structures of only 4,000 of the 30,000 human proteins have been solved, and these represent fewer than 300 of the estimated 1,000 unique protein folds. More problematic is the lack of human membrane protein structures. Only 1 of 10,000 have been solved!
The rate-limiting step is the crystallization process. Crystallographers follow a set of recipes, which are the result of years of experience, and although intuition gleaned from colloidal chemistry aids in improvisation the crystallization process is poorly understood. A typical crystallography lab contains cold rooms stacked floor-to-ceiling with racks upon racks of suspensions of proteins in a range of solvent conditions. A theoretical framework for protein crystallization is needed in order to restrict the parameter space and eliminate crystallization as a bottleneck in structural determination of proteins. Such a theory has for the first time been formulated in the last months of 1997 by tenWolde and Frenkel.
In September 1997, ten Wolde and Frenkel (Science) approached the protein crystallization problem from the perspective of classical nucleation theory applied to the phase behavior of colloidal sphere and polymer mixtures. In this remarkable paper, a complete theory for protein crystallization was succinctly elucidated. Their theory suggests a prescription for how to prepare protein suspensions most conducive to crystallization. We are exploring the applicability of this idealized theory of protein crystallization to real proteins, and are working in collaboration with the Rosenstiel x-ray crystallography group here at Brandeis. The ambitious goal of this research is to transform the art of protein crystallization into a science.
Microfluidics and the Phase Chip. Protein crystals are necessary in order to determine protein structure using x-ray diffraction. Typically the number of crystallization trials are limited by the availability of protein, hence the drive to minimize sample volume. To address this problem a high-throughput, low volume microfluidic device denoted the Phase Chip is being developed. On this device different microfluidic components have been designed, fabricated, and interconnected in order to precisely meter, mix, and store sub-nanoliter amounts of sample, solvent, and other reagents. The Phase Chip can store thousands of sub-nanoliter drops of protein solution in individual wells and a total of 103 crystallization trials can be accomplished with 1 - 10 microgram of protein thereby enabling high-throughput crystallization of mammalian proteins expressed in tissue culture. Additionally each sample well is in contact with a reservoir through a dialysis membrane through which only water and other low molecular weight organic solvents can pass. Thus the concentration of all solutes in an aqueous solution can be reversibly, rapidly, and precisely varied in contrast to current microfluidic crystallization methods, which are irreversible. Rapid reversible dialysis solves a major problem in protein crystallization, the decoupling of nucleation from growth. Using the phase chip we will screen crystallization conditions using proteins that are not available in sufficient quantities for current techniques. The protein targets are bacterially-expressed recombinant channel proteins, G protein-coupled receptors heterologously expressed in a mammalian cell culture system, and enzymes which produce crystals too small for diffraction.
Publications list
To get to the Electrorheological Fluids page
Protein Crystallization page
Colloidal Suspensions of virus page
Complex Fluids Group
Physics Department
Brandeis University
415 South Street
Waltham, MA 02454, USA