Research

Cell adhesion molecules, e.g., integrin, intercellular adhesion molecule, selectin, mediate cell-cell and cell-matrix interaction, and are important therapeutic targets in autoimmune-related diseases and cancer. Integrins are a family of heterodimers of alpha and beta subunits, and each of subunits consists of multiple modular domains, among which one or two domains serve as ligand-binding sites. Integrins exhibit sophisticated allosteric conformational change of bent to extended conformation induced either by signals from inside (inside-out signaling) or by ligand binding to integrins (outside-in signaling)(Figure 1). Previously we have studied experimentally and computationally an allosteric activation mechanism of ligand binding domain of some members of integrins, known as Inserted or I domain. Impressively, using various protein engineering tools, we haveengineered an I domain with an increase of 200,000-fold in the affinity to Intercellular adhesion molecule-1 (ICAM-1). When the high affinity I domain mutant was tested for a potential therapeutic use in blocking the migration of leukocyte through endothelium whose aberrant activation is associated with autoimmune-related diseases, it was found to be as effective as anti-LFA-1 antibody or Raptiva that was approved for treating psoriasis.

ICAM-1 is a ligand for some integrins and subverted as a receptor for human rhinovirus, which is the major causative agent of common colds. A binding site for rhinovirus in ICAM-1 is limited to the first N-terminal domain (D1), which by itself is not stable. D1 belongs to the immunoglobulin-like fold or immunoglobulin superfamily (IgSF), and we identified mutations that enable the production of D1 from bacterial expression system. ICAM-1 was shown to be effective in preventing rhinovirus infection in clinical studies, but the high cost of producing ICAM-1 from mammalian system prevented a further development. Therefore, the production of ICAM-1 D1, which was stabilized by mutations and optimized for a low-cost and large-scale production from bacteria, may open a new approach for developing ICAM-1 D1 as rhinovirus therapeutics. Additionally we plan to use this D1 for obtaining a high-resolution structure of rhinovirus bound with D1, which can facilitate developing small molecule drug that inhibits rhinovirus infection.

With the advent of modern technologies of producing antibodies, e.g., transgenic animal producing human antibodies, phage display of human antibodies, etc., antibody-based therapeutics are entering market with great anticipation. In the process of antibody production, a target antigen has to be purified and it is often desired that the antigen to be locked into an active conformation for isolation of activation-dependent antibodies. We aim to develop a novel and efficient system for engineering conformation-specific antibodies based on yeast display system. We will create yeast display vector that can express two proteins of interest by Gal1/10 promoter. Among the two proteins, a target antigen is displayed on the cell surface and a library of single-chain antibody (scFv) fused to eGFP will be expressed as a soluble protein. By isolating cells that exhibit higher level of fluorescence, the scFv that is reactive with a target antigen can be isolated.

Targeted delivery of nanoparticles containing drug molecules, siRNA, and fluorophore is becoming the technology of choice for potential cancer therapy, gene knock out, as well as for diagnostics and basic research. One of the key issues in the success of the targeted delivery of nanoparticles, e.g., liposomes, quantum dots, magnetic particle, etc., is how to ensure their specificity to targets. We tackle this problem through a surface modification of nanoparticles by proteins that will specifically bind to target molecules. Antibodies and physiological ligands to cell surface receptors, e.g., integrins, EGFR, etc., will be used for the decoration of nanoparticles, where these molecules can be engineered for enhanced affinity and stability, directly related to the Project I & II.s Currently, no diagnostic system is available for detecting early onset of brain disease. One of the difficulties in developing brain diagnostics arises from the blood brain barrier (BBB) in cerebral vasculature that prevents large particles emigrating from blood into brain. We propose to develop as MR contrast agent iron oxide nanoparticles coated with antibody against transferrin receptor, which will elicit transferrin receptor-mediated transcytosis of the particles to the brain. The particles will be coated with second set of antibody against NMDA receptors expressed highly at hippocampus of the brain, and their distribution will be analyzed by MRI at sub-millimeter resolution.

Adeno-associated virus (AAV) mediated gene delivery targeting brain and connective tissue. AAV is a non-pathogenic virus, and yields long-term gene transfer and minimal toxicity as a gene delivery system. We are interested in modifying viral capsid protein to create new tropism for the virus for efficient gene delivery into different tissues and for direct transport through the BBB into the brain.