Physical Principles Of Biomolecular Recognition

Child Health And Human Development

Investigators

Linked publications & trials

Abstract

Section on Physical Biochemistry, OD/NICHD conducts experimental and theoretical studies of structure and function of biomolecules with emphasis on molecular mechanisms of pathology in connective tissue disorders. Through collaboration with clinical researchers, we strive to gain better knowledge and develop novel techniques for diagnostics, characterization and treatment of osteogenesis imperfecta (OI) and other diseases. Over the years we reported first direct measurements and established physical nature of forces between collagen molecules. We discovered that collagen triple-helix is intrinsically unstable at physiological conditions and that micro-unfolding of most thermally labile helical regions is necessary for proper molecular recognition and fiber formation. In fibers, collagen helices are protected from complete unfolding but they constantly undergo transient local unfolding and refolding giving the fibers their unique combination of elasticity and strength. Currently, our research focus is gradually shifting from fundamental studies of these processes to understanding how different OI mutations affect them. In particular, during the last year we found that procollagen monomers are also thermally unstable at body temperature. Therefore, cells have to use molecular chaperones to fold procollagen within Endoplasmic Reticulum (ER). Furthermore, the temperature and kinetics of procollagen and collagen denturation are similar, except for OI collagens with mutations in the N-terminal 90 amino acids of the triple helix. In collaboration with HDB/NICHD scientists, we established that these first 90 amino acids form an important folding domain of collagen triple helix. OI mutations within this region disrupt the structure of the whole domain resulting in abnormal secondary structure, interaction and cleavage of N-propeptides. Incorporation of uncleaved molecules into fibers produces smaller diameter fibers with reduced strength, causing early onset scoliosis, hyper extensibility and joint laxity which are less common in other OI patients. From systematic analysis of physical and chemical properties of collagen with these and other OI mutations, we also demonstrated the following: Substitutions of obligate Gly residues (most common OI cause) may reduce the melting temperature of collagen triple helix by 5 C or have virtually no effect on triple helix stability depending on the position of the mutation within certain domains, while the identity of the substituting residue or its immediate local environment appear to be less important. Reduction of collagen stability by about 5 C or more results in such rapid denaturation of mutant molecules at physiological conditions that these molecules cannot be incorporated into fibers. At least in the case of G349C substitution (the only mutation for which a mouse model is available), OI phenotype is related to abnormal interactions of mutant collagen helices with other matrix molecules or abnormal function of osteoblasts rather than to abnormal structure, physical properties or interactions between mutant collagen helices. Our preliminary data indicate that abnormal osteoblast function in G349C mice may be caused by ER stress resulting from failure to clear collagen helices with one mutant chain through normal secretion pathway. Another important direction of our research is closely related recognition and assembly reactions involving DNA. In particular, we uncovered several common physical principles, which govern formation, structure and physical properties of collagen and DNA aggregates. We suggested mechanisms for counter-ion specificity in DNA condensation, DNA overwinding from 10.5 base pairs per helical turn in solution to 10.0 bp/turn in aggregates, sequence homology recognition in pairing of duplex DNA and several other phenomena. The present focus of these studies is measurement of sequence effects in formation, structure and properties of DNA aggregates. During the last year, we extended the theory to describe formation of aggregates from non-ideal helices with sequence-dependent twist between adjacent base pairs. These calculations showed that homologous fragments should exhibit aggregation at slightly lower concentration of condensing counterions and that the structure of such aggregates should be distinguishable from the structure of aggregates formed by nonhomologous DNA fragments. Our, predictions rationalized several previous experimental observations, e.g., torsional deformation of DNA in aggregates. Most importantly, we formulated several predictions for measurable parameters of DNA aggregation to be tested in in vitro experiments which are presently under way.