I. Structure, Folding and Function of Lipid Transfer Proteins (LTPs)

LTPs are well known for their ability to bind with variety of lipid molecules and catalyze the transfer of lipid molecules between membranes in vitro. Besides, some biological functions of the LTPs have been proposed, including biosynthesis of cutin, involvement in defense against pathogens, and managing abiotic stress conditions imparted by temperature and drought situation. LTPs are further subdivided into two different isoforms that differ in their molecular mass, 9kDa (LTP1) and 7kDa (LTP2). We have purified several non-specific lipid transfer proteins (LTPs) from rice (Oryza sativa) seeds and mung bean. The three-dimensional structure of nsLTP2 has been solved for the first time by us. It is found to adopt entirely a new fold containing a triangular hollow box like hydrophobic cavity formed by three prominent helices stabilized. We have also applied molecular modeling and physical methods to study its binding with various ligands. The ultimate goal of this project is to solve LTPs 3D structure and find out their biological function in vivo. The oxidative folding pathway of LTPs have also been studied in our laboratory.

II. Structural and functional studies of Drosophila melanogaster dopamine N-acetyltransferase

Dopamine is recently disclosed as a wake-promoting agent. Both in rats and in Drosophila, wakefulness is positively correlated with dopamine levels in the brain. In Drosophila, the mRNA levels of Dopamine N-acetyltransferase (Dat) were increased by 48% after 2 to 3 hours of waking. Flies lacking Dat (Dat^lo/Dat^lo and Dat^lo/Df flies) increased rest after rest deprivation, and the more severely mutant at the Dat locus, the greater the rest rebound. Those results implicate that catabolism of monoamines such as dopamine may regulate sleep and waking in the fly. This conclusion also suggests that Drosophila may serve as a model system for the genetic dissection of sleep. According to our previous structural analysis, we proposed that the entrances of substrate and cofactor are different. For binding, substrate might pass a tunnel-like pathway to arrive the middle of Dat. So we want to further confirm the existence of this substrate-entry tunnel, and to figure out whether the substrate selectivity will change with different tunnel bottleneck size. The electron density maps of both ternary complex reveled that they were consisted of Dat and two products (CoA and acetyl-substrate), although cofactor and substrate were used to form the crystals. This result indicated that the products might still strongly bind to Dat, and stimulated us to further investigate how the products leave Dat and let the next reaction occur. Since catabolism of monoamines is important in sleep/waking (rest/activity) regulation, investigating the mechanism of Dat would contribute to understanding more about the detail in this regulation.

III. Structural Genomics Project of Helicobacter Pylori

The goal of Structural Genomics Project is to discover and analyze the structures as well as functions of all proteins in nature in order to provide a foundation for a fundamental understanding of biology. Eight of research groups in College of Life Sciences have formed a team to target the Structural Genomics Project of Helicobacter Pylori. Our group use nmr as well as bioinformatics methods to solve and analyze the protein structures from H. Pylori.

IV. Bioinformatics approaches to structural biology

• We have set up several on-line bioinformatics services and mirror sites in our Bioinformatics Center. Please check (http://bioinfo.life.nthu.edu.tw/) for details.
• All information needed to create thermo tolerance is encoded in the protein sequences. Due to the advances in the genome sequencing project, the number of sequences deposited in databanks increases exponentially in the last few years. Thus, with a vast number of sequences available, it becomes possible to predict a myriad of protein properties from their amino acid sequences alone. We apply classical statistical method to study the relationship between Tm (melting temperature) values of proteins and their amino sequences. Using the reference thermo indices based on the statistical correlation between the dipeptide constituents and the Tm values of proteins, we are able to obtain prediction accuracy more than 80% for the test data sets. We have also applied this method to the whole genomes of several hyper-thermophilic bacteria, and found that more than 72% of their proteins were of high thermostability (Tm > 65 °C ), which is significantly different from those of mesophile genome. We have built a web-based program to predict the range of Tm of any given protein sequences and users can access the web page at http://tm.life.nthu.edu.tw/
• In collaboration with researchers from National Yang-Ming University and National Chung-Hsing University, we form a structural genomics team to target Xanthomonas campestris genome. Our lab is in charge the bioinformatics part of the project which includes the gene prediction, annotation, target selection and homology modeling. The detail of the project and preliminary results can be seen at http://xcc.life.nthu.edu.tw/.
• In iSARST service, we implement two protein structural similarity search methods, SARST and CPSARST. Besides, three outstanding structural alignment tools, FAST, TM-align and SAMO, are recruited as refinement engines. SE is applied to improve structure-based sequence alignments. We would like to thank these authors for their excellent developments, which greatly move this research field forward. iSARST allows the user to input many structures at once. Its MPI system will do the similarity searches and structural alignments in a batch mode to rapidly retrieve structural homologs of the query proteins. SARST is designed for co-linear structural similarity search while CPSARST specifically finds circular permutants. Circular permutation (CP) is a kind of structural rearrangement such that homologous proteins have different locations of termini. It is more and more widely used in protein engineering.