Skip Internal Navigation
We perform studies on cell behavior that can be linked to underlying physical mechanisms, for which we develop and apply methodologies based on mathematical and physical principles. Our research also involves biochemical and cell biological techniques. Projects currently include (i) constructing a physical model to explain the stochastic nature of coated vesicles biogenesis during receptor mediated endocytosis, (ii) determining the mechanical properties of clathrin cages and using that knowledge to gain insight into membrane transformations that occur during vesicle trafficking, (iii) exploring how substrate mechanical properties affect the movements of locomoting eukaryotic cells, and (v) understanding how certain small molecules interact with microtubules and thereby act as antimitotic agents. We also develop new experimental modalities to characterize these and related phenomena. We have a particular interest in the ways cellular activities are coordinated in space and time.
Biophysical methods and models
We develop and employ physics-based methodologies to investigate complex biological structures and materials. In particular, we have constructed model systems to mimic the movement of charged macromolecules within "congested" polymer solutions. The macromolecules are fluorescently labeled and their diffusion is monitored by fluorescence correlation spectroscopy (FCS). The unique power of FCS is that, in principle, one can focus on the motions of fluorescent entities while ignoring signals from non-fluorescent surroundings. Using a small probe (fluorescently labeled RNase A) and dextrans that carry differing electronic charges, we showed that transient, charge-mediated binding can retard the movement of proteins to an extent similar to that attributable to molecular crowding. The parameters of this experimental model can be readily tuned, so it can be used to study the properties of a wide range of crowded systems (1). We also used FCS to study targets located within optically dense media. If the sample interrogated by FCS is optically clear, data interpretation is relatively straightforward, but, if the sample demonstrates a high degree of multiple scattering, the dimensions of the illuminated sample volume may be distorted, confounding interpretation of the measurement. By comparing data from various well-defined scattering models, we gained insight into the reliability of parameters determined when FCS is used to probe the movement of molecules in highly complex, biological environments. We anticipate that knowledge acquired from such models may be used in various biomedical inquiries. Applications might include monitoring antibiotics and viruses diffusing in biofilms, studying morphogens moving within embryotic tissues, and assessing the delivery of drugs from polymer implants.
In a related study, undertaken to gain deeper knowledge of the effects of molecular crowders on supramolecular assembly, we employed a biomimetic model consisting of macroscopic rods immersed in a "fluid" of small spheres. In this setup, mechanical shaking plays a role analogous to thermal excitation. The system is purposely kept "small," in the sense that its physical dimensions are of the order of only 10 – 100 times the length of the rods. Depending on the numbers and sizes of the constituents, the rods self-assemble into raft-like structures when confined to quasi two-dimensional spaces. The structures are predicted by equilibrium thermodynamic simulations, suggesting that entropy maximization is the driving force for bundling. This and related work indicate the roles of solution constituents and boundaries in the assembly of oligomeric structures (2).
Additionally, we collaborated with colleagues at the National Institute of Standards and Technology to develop a low-cost instrument for polarization fluorescence microscopy. We first demonstrated its utility by examining the molecular orientations of probes attached to lipid membranes. The instrument provides a calibrated, tunable method to rotate polarization states of light prior to its being coupled into a fluorescence microscope. To demonstrate the application of this device to cells, we subsequently measured a series of full-field fluorescence polarization images from fluorescent analogs incorporated in the lipid membrane of Burkitts lymphoma CA46 cells. A spatially varying contrast in the normalized amplitude was observed on the cell surface when we used molecules whose head groups were labeled with DiI (1,1'-dioctadecyl3,3,3',3'-tetramethylindocarbocyanine) fluorophores, which orientated tangentially to the cell membrane. Internally labeled cellular structures showed zero response to changes in incident light polarization, with zero net linear polarization amplitude. A paper based on this work has been submitted to the Review of Scientific Instruments.
Complex systems biophysics
We employ advanced physical and mathematical methods to understand the biophysics of complex cellular processes. A major focus has been on the biogenesis of coated vesicles involved in receptor-mediated endocytosis and other intracellular transport processes. Receptor-mediated endocytosis involves the formation of plasma membrane–derived vesicles surrounded by closed polyhedral, cage-like structures assembled from a three-legged heteropolymer composed of three clathrin heavy/light chain complexes joined at a common hub (the "clathrin triskelion"). At the plasma membrane of eukaryotic cells, it is the principal pathway for the regulation of receptors and internalization of certain nutrients and signaling molecules. The early stage of receptor-mediated endocytosis involves the formation of transient structures known as clathrin-coated pits (CCPs) which, depending on the detailed energetics of protein binding and associated membrane transformations, either mature into clathrin-coated vesicles (CCVs) or regress and vanish from the cell surface. The former are referred to as "productive" CCPs and the latter as "abortive" CCPs. We posited a simple model for CCP dynamics and carried out Monte Carlo simulations to investigate the time development of CCP size and explain the origin of abortive pits and features of their lifetime distribution. By fitting the results of the simulations to experimental data, we were able to estimate values of the free-energy changes involved in formation of the clathrin-associated protein complexes that constitute the coat, and showed how the binding of cargo might modify the coat parameters and thereby facilitate CCV formation. Clathrin-mediated endocytosis has been shown to regulate signaling between cells in developing tissue and, in certain instances, influence the establishment of morphogen gradients in embryos.
In devising physical theories to understand how such CCVs arise, we developed various quantitative methods of physical analysis. For example, in earlir work we used novel computer-based structural modeling, combined with dynamic light scattering (DLS), static light scattering (SLS), and small angle neutron scattering (SANS), to examine the conformations and molecular mechanics of clathrin triskelia in solution. In an extension of these studies, we recently used quartz crystal microbalance-dissipation (QCM-D) instrumentation to investigate the mechanical properties of clathrin triskelia, CCVs, and clathrin cages assembled with and without AP180 adaptor proteins (APs). We also performed atomic force microscopy (AFM) measurements (3) that complement the QCM-D measurements and facilitate interpretation of the QCM-D data. Using these methods, we find the apparent shear moduli of these structures to be approximately one to two orders of magnitude smaller than the Young's modulus of a triskelion leg. The values of the shear moduli vary strongly with buffer properties and, to a lesser degree, also depend on properties of their substrate support. For example, we find that the shear modulus of CCVs attached to glass surfaces varies from 5 kPa to 12 kPa when buffer pH is changed over the pH range 6.2–7.5. This investigation is a continuation of our earlier work to establish the various mechanical properties of clathrin structures, as such properties are important elements in our physical models. Moreover, pH modulation of the nanomechanical properties of clathrin lattices and related protein structures may be an essential aspect of vesicle transformations involved in cellular function. A paper based on this work currently is being prepared.
A different project, also relating to mechanical aspects of cell response, involves establishing a reliable method to assess the influence of substrate properties on the locomotion of eukaryotic cells. Our emphasis has been on developing collagen-coupled polymer films (e.g., polyacrylamide) whose formation can be controlled by photopolymerization in a quantitative and reproducible fashion. The scheme uses an ultraviolet (UV) source of spatially-defined intensity to set up rigidity gradients in a film. The efficacy of the procedure has been confirmed by AFM. Currently, our work is directed at examining the response of single cells (e.g., durotaxis). However, an extension will focus on understanding the effects of substrate rigidity on the collective movements of mechanically interacting cells. Such work may have applications in studies of wound healing, cancer metastasis, and normal and aberrant development of embryonic tissues.
Tubulin polymers and cytoskeletal organization
We continued our study on the properties of microtubules (MT) and of drugs that alter these properties. The studies aim to find drugs that are highly active against parasite, but not human, MT and drugs that are active against human MT that can be useful chemotherapy agents. We are also concerned with the detailed mode of action of these drugs in patients.
Our earlier work showed that agents with the chemical core of dinitroaniline (derivatives of the herbicide oryzalin) can show selectivity for parasite microtubules versus human ones. As a result of their high selectivity for plant tubulin over mammalian tubulin, oryzalin and other dinitroanilines are effective herbicides. We showed that these compounds, which destabilize MT, also show selectivity for protozoal parasite tubulin over mammalian tubulin. We continued our effort to understand this selectivity by mapping the binding site for oryzalin on alpha tubulin using detailed analysis of mutations in parasite tubulin that confer resistance to this compound (4). We hope to use such detailed knowledge to design compounds that bind better and with improved selectivity to parasite tubulin, thereby yielding clinically useful antiparasite drugs.
We used a similar approach to define the binding site and mode of action of peloruside, a new MT–stabilizing natural product. We had previously shown by mass spectrometry and molecular modeling that this compound binds to a site on beta tubulin distinct from that of taxol, a clinically important MT–stabilizing drug. Selecting and mapping mutations in human tubulin that confer resistance to peloruside have confirmed our mass spectrometry data and improved our understanding of the binding site, how occupancy alters MT stability, and how this differs from taxol action. We hope to use this knowledge to understand the differing mechanisms of peloruside and taxol and to provide a basis for combination of these drugs in clinical usage (Cell Cycle, in press). It is already clear from binding-site mapping and preclinical studies that taxol and peloruside stabilize MT by distinct mechanisms. Molecular modeling of the two binding sites suggests a differing balance of longitudinal and lateral stabilization in the MT polymer, indicating that the mechanical properties of the MT may differ with the two drugs. Unperturbed MT are the most rigid intracellular protein polymers known, and taxol increases their flexibility 10-fold. We are measuring the rigidity of individual fluorescent MT after binding of taxol or peloruside in order to relate differences in binding site structures to differences in MT properties. This understanding could provide an explanation for the synergistic effect observed for combinations of these drugs in preclinical cellular models.
The roles of MT extend throughout the life of the cell, not only in mitosis, but also in the rest of the cell cycle that is not mitosis (over 98% of the cycle). These vital roles include establishing cellular polarity, supporting intracellular transport and signaling, and allowing directionality in cell movements. MT–targeting drugs are active in all cells, not only those in mitosis, and indeed some targets of clinical use of anti–MT drugs are post-mitotic cells. We have argued that, even in clinical settings, where intuition dictates that mitosis is the target, such as in patient tumors, data indicate that MT–targeting drugs are effective by interfering with non-mitotic processes (5). To improve the clinical usefulness of these agents, we plan to combine various experimental approaches to obtain a better understanding of the nonmitotic processes that are targeted by the action of anti–MT drugs.