Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out
Get this document free

College of Biological Science and Technology

VIEWS: 0 PAGES: 31

									                 College of Biological Science and Technology

Title : Rigid Monomeric and Polymeric Macrocyclic Polyaminocarboxylate
      Lanthanide Coordination Chemistry          and    Development     of   Artificial
      (Ribo)Nucleases
Principal Investigator:C. ALLEN CHANG
Sponsor:National Science Council
Keywords:Rigid    Monomeric,       Dimeric     and    Polymeric     Macrocyclic
          Polyaminocarboxylate Ligands, Lanthanides Complexes, Stability
          Constants, Selectivity, Formation and Dissociation Kinetics, Structure,
            Luminescence, NMR, Molecular Mechanics, DNA/RNA Cleavage


     The primary objective of this multi-year proposed research is to develop
fundamental understanding of the major thermodynamic, kinetic, and structural
factors that influence the desired physico-chemical properties of lanthanide complexes
of macrocyclic polyaminocarboxylate ligands (e.g. complex formation stability and
selectivity, reaction kinetics, NMR relaxation, luminescence, and structure) for
applications in magnetic resonance imaging (MRI), solvent extraction, photodynamic
therapy, luminescence labeling for biomolecules and catalysis for DNA and RNA
phosphate diester bond cleavage. Recently, we have found that the rate constants
measured for the Eu(DO2A)+ reaction with BNPP (a model compound with
phosphate diester bond) had a titration-curve-like dependence with pH. Our initial
kinetic studies revealed that at high pH, Eu(DO2A)+ could form more reactive
hydroxo-bridged [EuL(μ-OH)]2 dimers, Eu(NO2A)+ has similar results. However,
both Eu(DO2A)+ and Eu(NO2A)+ at high pH tend to form hydroxo-bridged
polynuclear species and precipitates, e.g. [EuL(μ-OH)]4. To verify this point, and to
develop more reactive and stable complexes as artificial nucleases, the research
targets of this present project are primarily on the rigid monomeric and dimeric
lanthanide macrocyclic complexes. The specific aims include the following:
(1) Synthesize and characterize (NMR, IR, X-ray structural determination and
      potentiometry) new monomeric and dimeric macrocyclic polyaminocarboxylate
      ligands (e.g. DO2A and NO2A derivatives) with variable cumulative ring strains
      and pendent arms (e.g. carboxymethyl, amidemethyl and hydroxyethyl groups).
(2) Determine the thermodynamic and conditional complex formation constants of
    these macrocyclic ligands with various metal ions including all trivalent
    lanthanide ions, alkaline earth metal ions, selected first and second row divalent
    transition metal ions (e.g. Ni2+, Cu2+, Zn2+ and Cd2+) and post transition metal
    ions (e.g. Pb2+) in aqueous solution. Evaluate ligand selectivity toward lanthanide
    complex formation.
(3) Determine the macrocyclic lanthanide complex formation, dissociation, and
      metal/ligand exchange reaction rates in aqueous solution at various conditions
     (i.e. changing pH, metal ion/ligand concentration, temperature and ionic strength)
     and evaluate possible reaction mechanisms.
(4) Determine if the thermodynamic and kinetic parameters, i.e. stability and
    selectivity constants, formation and dissociation reaction rates, activation enthalpy
    and entropy, of lanthanide DO2A/NO2A monomeric derivatives and dimeric
    complexes correlate with the ligand conformation, whether preorganized or not.
    And verify it with molecular modeling studies.
(5) Synthesize and characterize (while possible, by single-crystal X-ray analyses,
    solution NMR, laser-excitation luminescence spectroscopy, NMR relaxation and
    molecular mechanics calculations) the macrocyclic complexes of La3+, Eu3+,
    Gd3+, and Yb3+. Correlate structural features (e.g. number of inner-sphere
    coordinated water molecules, number of ligand coordinated donor atoms, NMR
    chemical shifts, ligand cavity size, ligand cumulative ring strain, and metal ionic
    radii) with previously found thermodynamic and kinetic properties.
(6) Use the designed cationic macrocyclic lanthanide complexes and dimeric
   structural analogues to promote the cleavage of phosphate ester bonds of model
   compounds (e.g. disodium 4-nitrophenyl phosphate and diphenyl 4-nitrophenyl
   phosphate) and DNA and RNA. Determine reaction rates and possible
  mechanisms. Examine the effects of structure, lanthanide ionic radius and charge
  of the complexes.
NSC95-2113-M-009-025 (95R232)
------------------------------------------------------------------------------------------------------

Title:Protein Folding and Aggregation(2/3)
Principal Investigator:chia-ching chang
Sponsor:National Science Council
Keywords:Protein, Folding, Aggregation, First-order State Transition, Diffusion
                 Limited Aggregation (DLA), Auto-correlation Function


     The first order-like state transition model is the proper model to describe the
global reaction of protein folding. In this project, the enthalpy change of protein
folding under each different folding path will be measured. By comparing the
enthalpy changes between each folding state, the state transition boundary of protein
folding can be obtained. Meanwhile the reaction heats of protein folding path and
unfolding path can be compared and these results can be used to verify the
assumption that the mechanisms of protein folding and unfolding are different. In
addition to the protein folding, the relationship between protein aggregation and
folding time will be revealed by direct folding process. Therefore, the refolding time
of protein folding can be derived from aggregation fraction of auto-correlation
function. Meanwhile protein folding reactions will be simulated by three-dimensional
random walk. The hypothesis of the protein folding process is an antagonistic
reaction between protein spontaneously folding and diffusion limited aggregation
(DLA) will be verified in this project, too. By systematic change the conformation
related amino acids or solvent environment, such as temperature, pH and ionic
strength, the change of critical time of autocorrelation function will be obtained.
The two state co-existing phenomena of first-order-like state transition will be
verified too.
NSC95-2112-M-009-019(95R078)
---------------------------------------------------------------------


Title:Molecular Mechanistic Study of Natural Product Methyltransferases and Their
       Applications in Drug Discovery
Principal Investigator:Hsien-Tai Chiu
Sponsor:National Science Council
Keywords:Methyltransferases, Biosynthesis, Secondary Metabolites, Polyketide
         Glycosides, Drug Discovery, Enzymology, Enzyme Kinetics


     Derived from diverse biosynthetic origins in nature, secondary metabolite
glycosides are small organic molecules composed of core chemical structures
decorated with a variety of carbohydrate moieties. Their extremely structural diversity
has made these glycosides an inexhaustible source of new antimicrobials, antivirals,
antitumor drugs, agricultural and pharmacological agents. The great structural
complexity is attributed to the enzymes involved in the biosynthetic pathways leading
to their formation.One important class of these enzymes is methyltransferase.
Methyltransferases play important functional roles in biological systems in nature,
including DNA methylation, tRNA maturation and natural products biosynthesis. This
project aims to study three O-methyltransferases (OMT) involved in the biosynthesis
of spinosyns, and to utilize the enzymes to generate a library of analogues of
secondary metabolites. Specific goals to conduct research are summarized: 1. The
three OMT will be cloned, overexpressed, and purified to near homogeneity in large
scale. 2. The enzymes will be functionally characterized with regards to their
substrates, structures, biochemical features and cofactors. 3. Substrate specificity of
OMT will be examined towards a library of substrate analogues and their kinetics will
be systematically analyzed.4. Molecular recognition of OMT will also be probed with
a group of inert substrate analogues by inhibition kinetics. 5. Site-directed
mutagenesis of OMT will be conducted to reveal functionally critical amino acids and
their molecular interactions of substrates and cofactors with protein structures. 6. A
library of mono-, di- or tri-methylated glycosides will be generated and their
biological activity and functions will be evaluated, e.g. antibacterial and antifungal
activities. 7. Ability of OMT or its mutated forms to methylate other secondary
metabolite glycosides will be evaluated in enzyme level. 8. OMT will be intensively
purified for structural determination by X-ray crystallography. O-methyltransferases
are responsible for maturation and diversity of many secondary metabolites of
biological importance. This research aims to reveal the factors controlling or dictating
molecular recognition of OMT, by cross-examinations of the kinetics experiments
towards a library of their substrate analogues that are made by in-house combinatorial
biosynthesis in our laboratory. In conjugation with enzyme mutagenesis, the OMT
will be studied extensively in molecular and enzymatic level for the first time.
Moreover, the study will substantially result in a group of brand new methylated
glycosides that possess great potential to exhibit novel or improved biological
activities. Specifically, the study can not only shed new light on understanding the
intriguing family of methyltransferases, especially those in secondary metabolite
biosynthesis, but also their applications in combinatorial biosynthesis can greatly
expand chemical diversity of secondary metabolites for drug discovery and
development.
NSC95-2113-M-009-022 (95R230)
--------------------------------------------------------------------------------------------------------

Title:Developing A Parallel Intelligent Optimization System Based on Evolutionary
       Algorithm for Genetic Network Modeling
Principal Investigator:Shinn-Ying Ho
Sponsor:National Science Council
Keywords:Evolutionary Algorithm, Genetic Network, Parallel Intelligent
         Optimization System, Capability Maturity Model Integration, S-system
         Model, Microarray Analysis


     This project uses an intelligent evolutionary algorithm to efficiently build celluar
dynamic regulation networks to discover more delicate and substantial functions in
molecular biology, biochemistry, bioengineering, and pharmaceutics. In order to build
an optimized genetic network, we adopt S-system model to formulate the genetic
network. Recent literatures showed that S-system model is suitable to characterize
biochemical network systems and capable to analyze the regulatory system dynamics.
However, inference of the S-system model of large-scale genetic networks is in a set
of non-linear differential equations to be optimized. Looking after all series data in the
meantime is time-consuming but necessary for real data with noise. The time
complexity is increased in exponential time as expected but the inferred S-system
model would be more reliable in this case. Therefore, a more powerful algorithm for
global-view of optimization is required. In this project, we proposed an evolutionary
optimization approach for the design of large S-system model, which can be archived
through the following stages. First of all, we use proposed algorithm to optimize the
model equations for a set of time series data without noise at a time based on a novel
intelligent genetic algorithm (IGA). In the early experimental results, extremely
accurate S-system models can be obtained by the proposed method. Second, we
analyze and modified the algorithm for parallel computation. A parallel evolutionary
algorithm based on message passing interface (MPI) is developed to optimize all
time-series data with noise for inferring optimized S-system model. In order to
develop a robust and qualified software system, we also conduct Capability Maturity
Model® Integration (CMMI) into our system development process. According to
process improvement, CMMI helps integrate traditionally separate organizational
functions, set process improvement goals and priorities, provide guidance for quality
processes, and provide a point of reference for appraising current processes. It is
expected the final system to be useful for related researches.
NSC95-2221-E-009-116(95R308)
--------------------------------------------------------------------------------------------------------

Title:Bio-inspired Optimization Methods for Protein Structures Analysis And
       Prediction: From Genomic Sequences to Protein Structures and Functions
Principal Investigator:Shinn-Ying Ho
Sponsor:National Science Council
Keywords:Protein, Structure Prediction, Scoring Function, Parameter Optimization,
                Evolutionary Computation


      This is a three-year subproject: bio-inspired optimization methods for protein
structures analysis and prediction: from genomic sequences to protein structures and
functions. The objectives are to develop various high-performance optimization
algorithms for large-scale parameter optimization problems of bioinformatics,
formulate the protein structure prediction and analysis as optimization problems, and
then provide the methodologies, solutions, and experience in solving bioinformatics
problems.
      Both protein structure prediction and molecular docking involve three important
phases: (a) design of accurate energy functions, i.e., scoring functions, (b) design of
optimization problems by identify system parameters to be optimized from combining
bio-knowledge and computing techniques, and (c) design of powerful optimization
algorithms for obtaining near-optimal solutions. After analyzing the obtained
solutions, we can further realize the structure and function of protein. The individual
projects for each year are fully cooperated, described below.
      Year 1: A good scoring function should truly reflect the native structure and can
distinguish it from its decoys. The performance of a scoring function can be further
improved by adjusting the interior energetic parameters and exterior weights.
However, the scoring function usually has hundreds of energetic parameters.
Evaluation of the scoring function is also very expensive. Thus the optimization of a
scoring function is extremely difficult. In this project, the scoring function based on
AMBER force field is optimized by using protein decoys and Z-score objective
function, as well as a newly-developed orthogonal simulated annealing algorithm. The
achievement can be used to improve the accuracy of protein structure prediction.
      Year 2: Based on the achievement of the first year, this project will design an
optimized scoring function for flexible protein-ligand docking and an associated
high-performance optimization algorithm for predicting the best ligand conformation.
Due to the strong interactions among conformation parameters, we will develop a
Particle Swarm Optimization based algorithm instead of existing evolutionary
algorithms for efficiently obtaining a near-optimal solution. Both prediction accuracy
and computing efficiency are expected to be superior to the existing genetic algorithm
based docking approaches.
      Year 3: The ab initio protein folding and protein-protein docking problems are
investigated. Since both problems are large-scale parameter optimization problems
and rely on accurate scoring functions of protein structure, we will apply the
achievements of the first two years, use the Intelligent Evolutionary Algorithms
proposed by the principal investigator of this project, and integrate bio-knowledge to
provide efficient designs of optimization problems and their problem-solving
methods.
      Some parts of achievements of this project such as 1) the obtained optimized
scoring functions for protein structure analysis including protein-ligand docking,
protein-protein docking, and protein folding; and 2) powerful evolutionary
optimization methods for the design of problem-dependent scoring functions and
prediction of the best conformation, can be directly applied to other subprojects and
further improve the performance of protein structures analysis and prediction. The
bio-inspired optimization methods can be implemented as software tools for
researchers to advance the research of interdisciplinary.
NSC95-2627-B-009-002(95R633)
--------------------------------------------------------------------------------------------------------

Title:Targeting RNA Motifs And Regulatory Elements: Hunting for New Drug Target
       Using Bioinformatics Approach (2/3)
Principal Investigator:Hsien-Da Huang
Sponsor:National Science Council
Keywords:RNA Structural Motif, Database, MicroRNA


   Current drug therapy with registered medicines targets only a few hundred
endogenous targets, mainly receptors and enzymes. Genomics have revealed many
novel candidates drug targets. The unique shapes in various target RNAs create
potential binding sites for small molecules. The specific aim in this investigation is to
develop a “RNA drug-targets hunting system” for the discovery of RNA functional
and regulatory elements. These elements are potential targets sites for the drug design
to induce or inhibit the RNA functions.
    The proposed system hunts for RNA drug targets by computationally detecting
RNA structural motifs within a set of aligned or unaligned RNA sequences. The
proposed integrated system comprises two biological databases and two proposed
software, which are briefly described are as follows: RNAWare (RNA data
warehousing system), RegRNA (Regulatory RNA motifs and element database),
RNAMST (An efficient and flexible search tool for RNA structural homologs), and
miRNAMap (Genomic maps for microRNA genes and their targets).
      In the first year of the project, we have collected the RNA-related data from the
biological databases in public domain and constructed the RNAWare. The microRNAs
and their targets in mammalian genomes were collected in a database, namely
miRNAMap (http://miRNAMap.mbc.nctu.edu.tw/), which has been published in
Nucleic Acids Research (Vol. 34, D135-D139). We also survey the literatures to
extract the RNA structural motifs and their functions to construct the RegRNA
database (http://RegRNA.mbc.nctu.edu.tw). The RegRNA database is now submitted
to Nucleic Acids Research.
NSC95-3112-B-009-002(95R042)
------------------------------------------------------------------------------------------------------

Title:The Computational Prediction, Experimental Verification And Functional
       Analyses of Human And Mouse MicroRNAs(1/3)
Principal Investigator:Hsien-Da Huang
Sponsor:National Science Council
Keywords: Significance And Motivation, The Specific Aims, The Proposed Methods

       Significance and Motivation: Recent work has demonstrated that microRNAs
(miRNAs) areinvolved in critical biological processes by suppressing the translation
of coding genes. The specific functions of the miRNAs are unknown in most
eukaryotic genomes. Therefore, computational methods for efficiently identifying the
miRNA precursors, mature miRNAs and their targets in a genome are crucial. The
comprehensive mapping between all the known/putative microRNA genes and their
target mRNAs in genomes will also be very effective for further investigation of gene
regulation, post-transcriptional control, alternative splicing pathway and other
mechanisms in biological systems.
      The Specific Aims: (1) To develop an integrated database, miRNAMap, to store
the known miRNA genes, the putative miRNA genes, the known miRNA targets and
the putative miRNA targets in human and mice. (2) To experimentally confirm the
predicted miRNAs, which are both conserved in human and mice. (3) To investigate
the regulatory relationships between miRNAs and their targets by using array CGH
data and microarray expression profiles. To systematically identify
disease/cancer-related miRNAs.
      The Proposed Methods: (1) The known miRNA genes in four mammalian
genomes such as human, mouse, rat and dog will be obtained from miRBase, and
experimentally validated miRNA targets will be extracted in a survey of the literature.
Putative miRNA precursors will be computationally identified using a non-coding
RNA prediction tool based on comparative sequence analysis. The mature miRNA of
the putative miRNA genes will be determined using a machine learning approach.
miRanda will be applied to predict the miRNA targets within the conserved regions in
3’-UTR of the genes in the human and mouse. (2) The miRNA candidates conserved
in human and mouse will be identified and experimentally verified by Northern
Blotting. Several human cell-lines and mouse tissues are selected for the experiments
for monitoring the tissue-specificity of the miRNAs. (3) Based on the constructed
miRNAMap database and experimental validated results, the isease/cancer-related
miRNAs can be systematically analyzed by referring to array-CGH data and
microarray gene expression data.
       The Anticipated Results: (1) The miRNAMap database will be constructed. It
will provide comprehensive miRNA information, the expression profiles of the known
miRNAs, cross-species comparisons, gene annotations and cross-links to other
biological databases. Both textual and graphical web interface will be provided to
facilitate the retrieval of data from the miRNAMap. (2) We hope more than thirty
miRNA candidates can be confirmed experimentally and at least 15 novel miRNAs in
human or mouse can be identified. (3) We expect that a systematic approach can be
developed to investigate the regulatory relationships among miRNAs, miRNA target
genes and to identify disease/cancer-related miRNAs.
      The main contributions of this project are addressed as follows. (1) To establish
a microRNA databases to curate known miRNAs, novel miRNAs and predicted
miRNAs. (2) A variety of novel miRNAs are expected to be identified in this
investigation. (3) Disease/cancer-related miRNAs can be determined to effectively
enhance the investigation of the disease-cause mechanisms involved by icroRNAs.
NSC95-2311-B-009-004-MY3 (95R490-1)
---------------------------------------------------------------------


Title:Bioinformatics Core for Genomic Medicine And Biotechnology Development
Principal Investigator:Jenn-Kang Hwang
Sponsor:National Science Council
Keywords:Structural Bioinfomratics
     It usually happens that the structure provides the only missing link between
sequences and function. In the study of protein-protein interactions, the
structure-based protein-protein docking algorithm often provides a more detailed and
accurate picture about residues at the interfaces between proteins than those methods
based on sequence alone. Therefore, the development of accurate, high-throughput
approach to predict three-dimensional structures from sequences becomes even more
important nowadays. In general, there are two types of theoretical approaches to
predict protein structure. One is the physics-based method based on the general
physico-chemical principles. However, at present, this so-called ab initio method are
not yet practical for real system. The second type of approach relies on the empirical
knowledge of proteins structures or sequences to assign the query sequences to the
proper folds by either comparative modeling, threading techniques (or reverse folding)
or taxonometric approach. However, the knowledge-based methods depend critically
on the training dataset, and it is usually less reliable in either the twilight or the
midnight zone where sequence homology is low. In this proposal, we will develop a
hybrid physics-based and knowledge-based approach that can be efficiently applied to
a genome-scale protein structure prediction. Hence, the specific aims are:
1. Generation of a decoy set of plausible structures of query sequences.
2. Develop a mixed physics - and knowledge based energy or scoring function to
distinguish native conformations from other decoy structures. We believe that this
approach may provide a useful and practical tool for predicting protein.
NSC95-3112-B-009-004(95R047)
-------------------------------------------------------------------------------------------------------

Title:Protein Structure Prediction Using The Hybrid Physics/Profile-based Approach(1/3)
Principal Investigator:Jenn-Kang Hwang
Sponsor:National Science Council
Keywords: Protein Profiles, Protein Structures, Molecular Mechanics

     It usually happens that the structure provides the only missing link between
sequences and function. In the study of protein-protein interactions, the
structure-based protein-protein docking algorithm often provides a more detailed and
accurate picture about residues at the interfaces between proteins than those ethods
based on sequence alone. Therefore, the development of accurate, high-throughput
approach to predict three-dimensional structures from sequences becomes even more
important nowadays. In general, there are two types of theoretical approaches to
predict protein structure. One is the physics-based method based on the general
physico-chemical principles. However, at present, this so-called ab initio method are
not yet practical for real system. The second type of approach relies on the empirical
knowledge of proteins structures or sequences to assign the query sequences to the
proper folds by either comparative modeling, threading techniques (or reverse folding)
or taxonometric approach. However, the knowledge-based methods depend critically
on the training dataset, and it is usually less reliable in either the twilight or the
midnight zone where sequence homology is low. In this proposal, we will develop a
hybrid physics-based and knowledge-based approach that can be efficiently applied to
a genome-scale protein structure prediction. Hence, the specific aims are:
1. Generation of a decoy set of plausible structures of query sequences.
2. Develop a mixed physics - and knowledge based energy or scoring function to
   distinguish native conformations from other decoy structures. We believe that this
   approach may provide a useful and practical tool for predicting protein.
NSC95-2113-M-009-030-MY3 (95R431-1)
-------------------------------------------------------------------------------------------------------

Title:Study of The Novel Gene(s) of H. Pylori Involving in Suppression of Host
       Dendritic Cells And Macrophages
Principal Investigator:KW Liao
Sponsor:National Science Council
Keywords: H. Pylori, Host Immune, Dendritic Cells and Macrophages

     The chronic infection of H. pylori is the main factor of severe digestive ulcer. It
has been indicated that H. pylori can inhibit a host’s immunity, which, when
suppressed, will be unable to clear the pathogens. According to our experimental data,
H. pylori was able to induce the host to produce a high amount of TGF-beta by a still
unknown gene product. Then, the host’s immunity will be suppressed completely. The
functional test equipment of TGF-beta inducing substrate (TIS) has been built in our
lab. No matter they are wild-type bacteria (ACTT43504) or clinical bacteria (HC28);
co-culturing H. pylori with AGS cells for 16 hours were both able to induce AGS
cells to produce TGF-beta while others could not. Moreover, the amount of induced
TGF-beta increased at higher infection quantity of H. pylori. Futhermore, co-culturing
AGS cells with the sonicate of bacteria whose cag A, VacA, babA, or sabA genes
were deleted could also induce AGS cells to produce a great amount of TGF-beta,
indicating that it was a unknown protein of H. pylori. Our preliminary data showed
that proteins between 40 to 100 kDa could do so whereas proteins lying out of this
range could not. Therefore, proteins in this range were collected by gel filtration.
Different molecular weights were separated by LC column. Twenty-five μl of the
separated proteins were taken out from each tube, mixed together every five tubes and
co-cultured with AGS cells for 16 hours. It was discovered that among the collected
proteins, some could induce AGS cells to produce TGF-beta. In the analysis of
SDS-PAGE, there were two sharp bands around 70 kDa, both of which were then
recollected by in gel digestion and analyzed by MALDI-TOF. Later, proteins were
blasted with the database on the Internet. Two TGF-beta inducing substances (TIS)
were candidates, which were UreB and Hsp60. In this year’s project, we expect to
produce a high amount of TIS by cloning techniques and search the mechanism by
which a TIS affects a host’s natural immune cells. To conclude, our main goals of this
year are: (1) to complete TIS cloning and TIS protein expression, and (2) to
nderstand the suppressive ability of their genes on dendritic cells and macrophages.
NSC95-2320-B-009-002 (95R383)
--------------------------------------------------------------------------------------------------------
Title:High Efficiency DNA Vaccine Against Toxoplasmosis
Principal Investigator:KW Liao
Sponsor:COUNCIL OF AGRICULTURE, EXECUTIVE YUAN
Keywords: Toxoplasmosis, SAG-1 Antigen, Adeno-associated Virus (AAV) Vector,
          Muscle-specific Promoter, DNA Vaccine


       Toxoplasma gondii is the etiological agent of toxoplasmosis and is the most
frequent and best known of the parasitic diseases. Many mammals serve as
intermediate hosts, and members of the cat family (Felidae) are the definitive hosts.
Infection is contracted by ingesting either oocysts or meat containing live organisms.
Toxoplasma infection is often asymptomatic, but pregnant individuals may suffer
from stillbirth, abortion, and fetus congenital malformation. Encephalities is also an
important and severe manifestation of toxoplasmosis in immunosuppressed patients
and may cause to death. As the most well known parasitic disease in the world, there
is no effective vaccine to protect mammals from toxoplasmosis attack. Our project is
to build up an anti-SAG-1 (Toxoplasma gondii surface antigen-1) AAV
(adeno-associated virus)-DNA vaccine to prevent cats from toxoplasmosis affection,
for cats are the major source of infection. This vaccine will be produced in a
well-established, and severely-guaranteed procedure to make the titer 5 ×10 12/ml.
For the numerous advantages including stringent safety, high titer, and long-term gene
expression, AAV is used as the vehicle of our DNA vaccine. Also we replace the
Cytomegalovirus (CMV) promoter by muscle-specific promoter to drive the SAG-1
gene, therefore it avoids the host immune response due to virus promoter and
prolongs the lifetime of the vaccine. The safety and protective efficiency of
anti-SAG-1 AAV-DNA vaccine will be tested in the experimental mice and cats model.
Then we will incorporate with animal hospitals to further investigate the vaccine
efficacy of the cats clinical trials. We sincerely hope this anti-SAG-1 AAV-DNA
vaccine can effectively prevent toxoplasmosis infection and lower the huge impaction
of the public health and agricultural economics.
I95811(95.01.01-95.12.31)
------------------------------------------------------------------------------------------------------
Title:Interplay of transcriptional regulation and activation of matrix etalloproteinases
     (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) in the pathogenesis
     of left ventricular remodeling
Principal Investigator:Chich Sheng Lin
Sponsor:National Science Council
Keywords: Pecies-Specific Sequences of DNA

     Left ventricular (LV) remodeling after myocardial infarction (MI) contributes
significantly to heart failure (HF) and death. Over last 2 decades, the researchers
support that dysregulating extracellular matrix (ECM) plays a major role in the
progress of LV remodeling.Matrix metalloproteinases (MMPs) are a family of
proteolytic enzymes responsible for ECM degradation in tissue during various
pathophysiological conditions. In addition, MMP activity is tightly controlled by
endogenous inhibitors, named tissue inhibitors of metalloproteinases (TIMPs).
Alteration in the balanced between MMP and TIMP in the levels of transcriptional
induction and enzymatic activation, resulting in enhanced MMP activity, has been
performed to occur in pathological process, such as MI, HF, and dilated
cardiomyopathy (DCM). However, interplay and dynamic regulation of MMPs,
TIMPs, and their effectors in cardiac ECM remodeling during MI or HF progresses
are largely unclear and remain to be explored. In the present project, we propose three
research strategies addressing on the above subjects. In Strategy 1, the dual
MMP/TIMP promoter activity vectors, MMP-2/TIMP-2 and MMP-9/TIMP-4, will be
constructed. In both vectors, green fluorescent protein (GFP) and red fluorescent
protein (dsRed) will be constructed downstream of the gene promoter (approximate
1.5 - 2 kb) of MMP and TIMP as reporters, respectively. Two vectors will be
transferred into cultured myofibroblast H9c2 cells or rat primary myofibroblasts by
DNA-liposome method to explore the interplay of transcriptional regulation of MMP

and TIMP under TGF-β, TNF-α, angiotensin peptides, or hypoxia treatments, all of

the factors reported as a trigger in signaling pathways of ECM metabolism in
myocardium. In Strategy 2, a CardioMP gene chip will be fabricated for studying the
global and specific portfolio of ECM, MMPs, TIMPs, and their relative genes that are

regulated within MI or failing myocardium (in the Strategy 3), and TGF-β, TNF-α,

angiotensin peptides as well as hypoxia treated mrofibroblasts (in the Strategy 1). In
the CardioMP Gene Chip, about 100 oligonucleotide probes with 65-70 mer will be
designed according to the gene sequences encoding ECM proteins, MMPs, TIMPs
and their regulated proteins, synthesized, and spotted onto a glass chip. A competitive
hybridization with Cy3 and Cy5 labeled sequences as targets will be performed in this
chip system. In Strategy 3, rat MI and HF models by coronary artery ligation (CAL)
and adriamycin induction will be established and applied to study the dynamic role
and network among ECM, MMPs, TIMPs, and their relative factors during LV
remodeling. The profiles of plasma and cardiac MMPs and TIMPs activities (by
zymography), and the transcriptional patterns of LV tissues (by CardioMP Gene Chip)
sampled from rats will be assayed over a time-course following the CAL procedure
and adriamycin injection. Metabolism of ECM is a tightly and dynamically regulated
process occurring in and among cells. Dysregulation of ECM metabolism by the
MMP/TIMP imbalance plays an important role in cardiovascular pathogenesis. Our
studies will be valuable in understanding this modulation and will provide new insight
into the pathogenesis of cardiovascular diseases, particular HF, and further illustrate
new potential targets or methods for MMPs inhibition.
NSC95-2313-B-009-002-MY3(95R491-1)
 -------------------------------------------------------------------------------------------------------

Title:A Hydrophobic Surface Substitution on The Stability of Thioredoxin
Principal Investigator:Tiao-Yin Lin
Sponsor:National Science Council
Keywords: Thioredoxin, SEHC

       Stabilization or destabilization of a protein by a hydrophobic surface
substitution is a complicated phenomenon, and can not be well explained or predicted
in every nown case. It was proposed that the effect is context dependent. Surface
hydrophobic stitution on proteins often led to destabilization of proteins or without
effects. Stabilization of proteins by a hydrophobic amino acid substitution occurred
less often, but was found in some cases. Small exterior hydrophobic cluster (SEHC)
has been proposed to explain the stabilization effect of a hydrophobic surface
substitution.
       However, experimental designs to examine the creation of SEHC were
practically lacking in the literatures. Replacement of the surface Gly33 with Ile
stabilizes the oxidized thioredoxin. In this proposal, we will study whether in the
context of breaking a disulfide bond, the degree of stabilization by a hydrophobic
surface substitution will be altered. Experiments are designed to investigate the
structural and other physicochemical features in order to explore the possibility of
forming SEHC for the oxidized and the reduced forms of the mutant. The difference
of these properties between the mutant and wild-type proteins can be compared, and
related to the degree of stabilization effect. Combining with the structural modeling
and further structural determination, we will gain insight of the stabilization effect
elicited by a hydrophobic surface substitution. Understanding of the factors
contributing to the stabilization effects of the surface hydrophobic substitution will
facilitate the engineering of more stable proteins for various applications.
NSC95-2311-B-009-003 (95R381)
-----------------------------------------------------------------------------------------------------

Title:The Algorithmic Study of Detecting Programmed Ribosomal Frameshifting
       And Its Software Tool Development
Principal Investigator:Chin Lung Lu
Sponsor:National Science Council
Keywords: Bioinforamtics, Computational Biology, Recoding; Programmed
          Ribosomal Frameshifting, RNA Pseudoknot


      Programmed ribosomal frameshifting (PRF) is a recoding mechanism that causes
the translational ribosome to switch from one reading frame to another reading frame
at a specific position such that the translation continues in the new frame and as a
result, a fusion protein encoded by these two reading frames is produced. Currently,
PRF has been found to occur in a variety of pathogenic viruses (including SARS-CoV)
and transposable elements as well as a few of bacteria and eukaryotes. In all viruses
examined so far, even small alternations in the efficiencies of PRF inhibit virus
propagation, which implies that PRF presents a potential target for antiviral
therapeutics. Hence, it is very important to design an efficient and effective
bioinformatics algorithms and tools for quickly and accurately identifying the PRF
sites in an RNA sequence. The most common PRF is the so-called programmed -1
ribosomal frameshifting (-1 PRF for short) in which the ribosome slips a single
nucleotide in the upstream direction. Currently, several computational approaches,
such as pattern recognition, statistical analysis and machine learning, have been
proposed for the identification of -1 PRFs in a given RNA sequence. However, most
of them usually predict too many candidates of false positive and even fail to identify
the candidates of true positive in some RNA sequences. Particularly, one common
weakness for these programs is worth mentioning that the methods by which they
predicted the stimulatory RNA structure (should be H-type pseudoknots typically) for
-1 PRFs are very simple such that the obtained RNA structures (that are usually
simple stem-loops) are not stable enough to function efficiently as a stimulator of -1
PRF. Recently, we have successfully developed a heuristic approach that is capable of
efficiently and accurately predicting H-type pseudoknots in a given RNA sequence
[Huang, C.-H., Lu, C. L. and Chiu, H.-T. Bioinformatics, Vol. 21, 2005, pp.
3501-3508]. We believe that this approach is helpful for identifying the stimulatory
H-type pseudoknots of -1 PRFs so that we have a great chance of significantly
improve the sensitivity and specificity of the currently existing programs. Hence, in
this proposal, we plan to further modify accordingly our developed approach of
locating H-type pseudoknots so that it can well serve as the kernel of our designed
algorithms for efficiently and accurately detecting -1 PRF sites in an RNA sequence.
In addition, we are going to implement our algorithms as a web server that is open to
the public for the online analysis of -1 PRFs. To demonstrate the applicability and
effectiveness of our system, we plan to test it on several RNA sequences (including
SARS-CoV sequences) that are known to possess -1 PRF sites in advance, and also to
compare our experimental results with those obtained by the currently existing
programs.
NSC95-2221-E-009-231 (95R358)
------------------------------------------------------------------------------------------------------

Title: Physiological Role of Milk β-lactoglobulin(2/2)
Principal Investigator: Simon J.T Mao
Sponsor: National Science Council
Keywords: Fatty Acids, Steroids

     β -Lactoglobulin (LG) is one of the major milk whey proteins, containing about
10% of the total by weight. The molecular mass of LG is 18.5 kDa belonging to the
lipocalin family. In secondary structure, it consists of nine β strands and one
alpha-helix. The central hydrophobic pocket (calyx) possesses the property in binding
to vitamin D, vitamin A (retinol and retinoic acids), fatty acids, and some steroids.
Unlike to α-lactalbumin, LG is quite sensitive to thermal denaturation, the secondary
structure is altered upon the heating with a transition temperature at 70-80 ℃. In the
last three and half years, we have constructed a detailed thermal denaturation curve
for LG with its time and temperature, the data provide dairy industry a valuable
reference (Chen et al, 2005, JDS) (1). We have also mapped out a specific amino acid
sequence region that is responsible for the thermal change above 80 ℃. Such changes
also results in losing its ligand binding (retinol and palmitic acid) (Song et al, 2004,
JBC) (2). In addition, we have reported that a monoclonal antibody can only
recognize the dry milk, but not the fresh raw milk (Chen et al, 2004, JDS) (3).
Reversely, it would be important to prepare a monoclonal antibody that only
recognize the native form of LG, so that the un-denatured LG content in the processed
milk can be determined.
     Nevertheless, the physiological role of LG, by thus far, remains unclear. One
recent study points out that LG possesses a hypocholesterolemic effect, but whether
the heat treatment of LG may affect its efficacy is not yet known. More recently, we
have shown that LG dramatically stimulates the proliferation of hybridoma B cells.
The LG may therefore used as a modulator for cell growth in general and may even
used in the medical and nutrtional industry for wound healing.
     Therefore, the specific aims in the next 2 years will be to:
1. Use monoclonal antibody as a probe to analyze the thermal sensitive region of LG
   and to relate it in functional role (Year 1).
2. Produce monoclonal and polyclonal antibodies that specifically recognize native
   LG and to employ it for determining the native LG levels in processed milk (Year
   1).
3. Study the hypocholesterolemic effect of LG with and without heat treatment (years
   1 and 2).
4. Define the mechanism by which LG stimulates the hybridoma B cell proliferation
   and to isolate the LG receptors for LG binding domain (Years 1 and 2).
NSC95-2313-B-009-001(95R382)
--------------------------------------------------------------------------------------------------------

Title:Physiologic Role of Antiinflammatory Molecule Haptoglobin(1/2)
Principal Investigator:Simon J.T. MAO
Sponsor:National Science Council
Keywords:β-lactoglobulin


      The pathogenesis of mammary gland mastitis and atherosclerosis is associated
withinflammation, which induces acute phase proteins. Haptoglobin (Hp) is one of the
acute phase proteins, possessing a high binding affinity with hemoglobin (Hb) in
preventing the oxidative damage from releasing free hemoglobin. Human Hp is
classified as three phenotypes: Hp 1-1, 2-1, and 2-2. This phenotype plays essential
roles in determining the clinical outcomes and the prognosis of some inflammatory
diseases. Hp is also a potent antioxidant that modulates immunoresponse. However,
the physiologic roles of Hp remain unclear. There is only one phenotype in cattle and
other animals, and the structure of bovine Hp is polymeric similar to that of human
Hp 2-2. A recent study shows that H p increases in infected mammary gland resulting
in high level of Hp in bovine milk. Therefore, Hp is a potential diagnostic marker for
bovine mastitis. The role involved in the increased Hp triggered by inflammation
needs further investigation. Our preliminary studies showed that Hp is markedly
expressed in atherosclerosis lesion of cholesterol-fed rabbits, especially with the foam
cells. We also demonstrated that oxidized LDL stimulates Hp expression by
macrophage and smooth muscle cells, followed by inhibiting the proliferation of
smooth muscle cells. We postulate that Hp may play important roles in pathogenesis
of atherosclerosis. We attempt to explore the regulation of Hp during inflammation,
and investigate the influence of Hp on the pathogenesis using mastitis and
atherosclerosis as a model system. Therefore, the specific aims in the next two years

will be to

1) Investigate the relationship between Hp and inflammation. (Years 1 and 2)
2) Study the relationship between Hp and bovine mastitis. (Years 1 and 2)
3) Utilize Hp monoclonal antibody to develop a rapid kit for bovine mastitis. (Years 1
  and 2)
4) Determine the roles of Hp in the pathogenesis of atherosclerosis by using miniature
   pig as an animal model. (Years 1 and 2)
NSC95-2313-B-009-003-MY2 (95R492-1)
-------------------------------------------------------------------------------------------------------

Title:Investigation The PPO in Litchi Pericarp And Its Application in Browning
       Protection
Principal Investigator:Simon J.T. MAO
Sponsor:COUNCIL OF AGRICULTURE, EXECUTIVE YUAN
Keywords: PPO in Litchi, Pericarp

      Polyphenol oxidase (PPO) is a copper-containing enzyme that catalyzes the
chain-oxidation from monophenol or polyphenols to o-diphenols and subsequent
o-quinones. The enzyme reflects the browning reaction in fruits. In the present study,
we investigated the oxidation activity of PPO in litchi pericarp and the mechanism by
which PPO instantly makes pericarp browning. PPO of litchi pericarp was initially
extracted and isolated through gel filtration chromatography and then eluted directly
from SDS-PAGE. Two molecular forms of litchi PPO were identified as 86 and 66
kDa, and thereafter named as PPO-86 and PPO-66, respectively. The Km and Vmax
were determined as 66 mM and 382 μM/min for PPO-86 and 102 mM and 290
μM/min for PPO-66, respectively. Most importantly, PPO-86 could trigger the
3,3’-diaminobenzidine and 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)
oxidation, which was not found in other plant PPO. Thus, the unique PPO activity of
litchi might account for its superior rapid-browning reaction. In addition, we show the
presence of a potent volatile inhibitor(s) for PPO in litchi pericarp. The surface of
post-harvesting litchi pericarp revealed an opening ultra structure under the scanning
electron macroscopic examination, therefore allowing an instant evaporation of PPO
inhibitor. As such, the PPO oxidation was proceeded. The novel finding clarifies the
mechanism involved in the rapid browning phenomenon of post-harvesting litchi
pericarp. The finding of evaporation of potent PPO inhibitor may be potentially used
as a strategy in developing a novel method for preventing litchi browing.
I95515(95.01.01-95.12.31)
-------------------------------------------------------------------------------------------------------

Title:Genome wide analysis of expression of the fimbrial operons in Klebsiella
       pneumoniae
Principal Investigator:Peng Hwei-Ling
Sponsor:National Science Council
Key words: Klebsiella pneumoniae, fimbriae, ahesins, adherence specificities
     During bacterial infection, fimbriae (also called pili) play a key role for bacteria
to establish persistence on the host cells. It has long been noted that many bacteria
express multiple fimbrial adhesins, each of which may contribute unique or
overlapping specificity in host cell recognition. Using bioinformatics tools to analyze
the K. pneumoniae NTUH K-2044 genome, we have found up to nine distinct fimbrial
operons including type 1 and type 3, and seven novel types fimbrial gene clusters,.
These fimbrial operons, either functionally synergistic or regulated differentially,
allow the genomic study to unravel the molecular complexity of the adhesion step in
the bacterial infection. Therefore, we propose here a 3-year study plan respectively:
1. To investigate the prevalence of the nine fimbrial gene clusters among K
   pneumoniae clinical isolates from various types of disease, and to determine the
   possibility and correlation between the fimbriae and disease types.
2. To analyze differential expressions of the nine fimbriae under various
   environmental factors as well as during biofilm formation, and to identify possible
   regulators for the fimbrial expression in the bacteria. Moreover, we anticipate
   building up a regulatory network of differential expression of fimbriae in K.
   pneumoniae.
3. To characterize the adherence specificities and identify the corresponding cell
    receptors for each of the fimbrial adhesins, and to analyze the correlation between
    the nine fimbrial adhesions and tissue tropism. In addition, putative anti-adhesion
    drugs will be assayed through rational drug design and competition experiment.
    The pathogenic mechanism of K. pneumoniae is largely unknown. After the above
    analyses, we anticipate to further understand the fimbrial adhesion displayed by K.
    pneumoniae. Moreover, increasing knowledge of the mechanism of host
    cell-pathogen interaction holds promise to identify intervening targets for
    antimicrobial agents.
NSC95-3112-B-009-001(95R041)
--------------------------------------------------------------------------------------------------------

Title:Study Environmental Factors Regulation of RNase E Activity And Rne Gene
       Expression in Escherichia Coli
Principal Investigator:Ching-Ping Tseng
Sponsor:National Science Council
Keywords: Escherichia coli, Starvation, Growth rate, Oxygen, RNase E

     RNase E plays as a key role for mRNA degradation in E. coli. Previous studies
have shown that there were two- to five-fold changes of rne-lacZ xpression and rne
mRNA stability under different carbon sources, growth rates, starvation and oxygen
availability. In additions, the 5’-UTR region of rne-lacZ reporter fragment affected
the stability of rne gene. These results suggested that these environmental factors
regulated rne gene expression at transcriptional and post-transcriptional levels. In
order to understand the regulation of carbon source, growth rate, starvation and
oxygen on rne gene expression, we will examine the effects of above factors on rne
gene, including rne transcription, mRNA stabilities, protein translation and RNase E
enzyme activity. Also, the possible regulators in response to environmental
stimulations, such as ppGpp, ArcA and Fnr, Hfq and Lon protease will be identified at
different extracellular conditions. RNase E is a part of degradosome that contains
PNPase, PPK, RhlB and enolase. Earlier reports have shown that the intracellular
RNase E concentration and its activity may change in response to modification of
environmental conditions. Since PPK, RhlB and enolase activities are affected by the
presence of ATP, it will be also interested to know how RNase E gene expression is
regulated by the energy state and ATP level. Thus, in this study we will further
analyze the energy state and RNase E activity at different growth conditions. The
results of this study could provide useful information to understand the role of energy
state on gene expression and physiological change of microorganisms.
NSC95-2311-B-009-002(95R380)
 -------------------------------------------------------------------------------------------------------
Title:Removal of Organic Waste Air Trichloroethane by A Biofilter
Principal Investigator:Ching-Ping Tseng
Sponsor:Tsing Hua Testing & Analysts Co.,Ltd.
Keywords: Opto-electronics, Biofiltration

     Recently we know that in some process of manufacturing product, some
chlorinated volatile organic compounds could be commonly used for solvent in the
semiconductor and opto-electronics field, most being used of them is trichloroethane.
Trichloroethane is commonly used for cleaning the oil stain or oil on metal in
industrial applications. 1,1,1-trichloroethane is an artificial substance, it doesn't exist
in nature. Most people could smell trichloroethane when the concentration is about
120-500 ppm in air. We will screen some useful bacteria that can utilizing
trichloroethane and fix them in a biofilter to remove organic waste air. We did a paper
research and found there were not much research related to biofiltration for
trichloroethane removal in Taiwan. There are some technical reports about physical
and chemical treatment for trichloroethane removal, the most disadvantage of physical
and chemical treatment is to use much energy, maybe the process could cause waste
water. We believe that if we could develop an efficient biofilter successfully, it could
be helpful in the market for organic waste air removal.
NSC95-2622-B-009-001-CC3(95R978)
-------------------------------------------------------------------------------------------------------
Title:Biological Treatment to Remove The Waste Gases from The Composting Field
       And Analysis Microbial Community Changes During Deodorous Process (3/3)
Principal Investigator:Ching-Ping Tseng
Sponsor: Tsing Hua Testing & Analysts Co.,Ltd.
Keywords: Biofilter, Organic Sulfur, Organic Amine, Fatty acid, Compost Field,
          Environmental Biotechnology


     Waste gases emitted from composting field have brought many environmental
problems. These emitted odors include ammonia, hydrogen sulfide, organic sulfur and
organic nitrogen compounds. All of them can cause the damage on human body.
Except organic sulfur and nitrogen compounds, the short-chain fatty acids (C2~C6)
are also commonly odors because of their low threshold values. The high removal
performance of volatile malodorous is controlled by engineering methods and bacteria
inoculated into the biofilter.
     In previous studies, we have successfully removed the ammonia and hydrogen
sulfide at the field test. Because traditional method is required large space, much
electric power and results in low treatment efficiency, in this study we will remove the
waste gases emitted from composting field by combining biological and engineering
techniques. The modified bioreator has been proven to be a feasible method, which
showed high removal efficiency during the past one year experiment. The
experimental results during past four months operation indicated that the removal
efficiency for 5-40 ppm of NH3 and 15-120 ppm of organic nitrogen compounds
achieved 94% and 99%, respectively. Organic sulfur, short-chain fatty acids and H2S
were simultaneously 100% removed. In addition, the changes of pH, pressure drop,
moisture content and bacterial communities showed that the system was applicable for
the further commercial application. The biological technique for waste gases
treatment also possesses the advantages including high removal efficiency, stable
operation, cheap and no secondary pollution. Based on these results, we will use these
technologies to treat the odor emitted from compost field. This technique
development can elevate the safety of staff and decrease the environmental impact
derived from odor or toxic gases. In addition, the platform of scale up and technique
transfer will be established. The achievement of this study will develop the
environmental biotechnology and expand the market of biological waste gases
treatment.
C95006(94.12.01-95.11.30)
------------------------------------------------------------------------------------------------

Title:Removal of Organic Waste Air Trichloroethane by A Biofilter
Principal Investigator:Ching-Ping Tseng
Sponsor:National Science Council
Keywords: Opto-electronics, Biofiltration

      Recently we know that in some process of manufacturing product, some
chlorinated volatile organic compounds could be commonly used for solvent in the
semiconductor and opto-electronics field, most being used of them is trichloroethane.
Trichloroethane is commonly used for cleaning the oil stain or oil on metal in
industrial applications. 1,1,1-trichloroethane is an artificial substance, it doesn't exist
in nature. Most people could smell trichloroethane when the concentration is about
120-500 ppm in air. We will screen some useful bacteria that can utilizing
trichloroethane and fix them in a biofilter to remove organic waste air. We did a paper
research and found there were not much research related to biofiltration for
trichloroethane removal in Taiwan. There are some technical reports about physical
and chemical treatment for trichloroethane removal, the most disadvantage of physical
and chemical treatment is to use much energy, maybe the process could cause waste
water. We believe that if we could develop an efficient biofilter successfully, it could
be helpful in the market for organic waste air removal.
NSC95-2622-B-009-001-CC3(95R978)
--------------------------------------------------------------------------------------------------------

Title: The Research on Oxidosqualene Cyclase-Binding Peptide Aptamers Interaction
       and Its Peptidomimetics(2/3)
Principal Investigator: Tung-Kung Wu
Sponsor: National Science Council
Keywords: Peptide Aptamer, Phage Display, Combinatorial Peptide Library,
            Antifungal, Hypocholesteremic Drug, Peptide Mimetics, Chemical
            Biology, Chemoproteomics


     Oxidosqualene cyclases (OSCs) catalyze the cyclization/rearrangement of
(3S)-2,3-oxidosqualene to a diverse array of sterols and triterpenoids, including
lanosterol, cycloartenol, and -amyrin. Yeast and other fungi metabolize lanosterol
to ergosterol, which function as primary fungal membrane sterol and is essential for
yeast growth. Mammals utilize lanosterol to synthesize cholesterol and varieties of
steroids. Clinical research results have demonstrated that the relationships between
increase of coronary heart disease and elevated plasma LDL cholesterol and the
unique position of OSC in cholesterol biosynthesis, which made oxidosqualene
cyclase a unique target for antimicrobial, antifungal, and hypocholesterolemic agent
development.
      In order to advance the abovementioned goals, we will apply chemoproteomic
strategies to screen and identify putative oxidosqualene cyclases-binding ligand.
Two different approaches, the phage display and nanoparticle-conjugated peptide
libraries, will be applied to generate combinatorial libraries for in vitro peptide
aptamers screening. Following the identification of the OSC-peptide aptamers, the
detailed molecular kinetics will be studied using Biacore techniques. In parallel, the
high resolution biophysical tools such as nuclear magnetic resonance and
femtosecond fluorescence/absorption laser spectrometries, will be applied to
investigate the structure and mechanistic bases of the molecular interactions.
Besides, modified peptide or nonpeptide strategies will also be applied to generate
improved peptidomimetics with better binding affinity, specificity, or stability for
better drug lead design and synthesis.
NSC95-2113-M-009-005(95R223)
-----------------------------------------------------------------------------------------------------

Title:The Research on Antiviral Drug Screening of Thymidine Kinase from White
       Spot Syndrome Virus
Principal Investigator:Wu Tung-Kung
Sponsor:National Science Council
Keywords:White Spot Syndrome Virus, Thymidine Kinase, Molecular Docking,
               Drug Screening


     The crustacean-infected white spot syndrome virus (WSSV), widely spread in
most Asia countries as well as in Gulf of Mexico and South America, caused high
shrimp mortalities and severe economical damage to shrimp culture industry. The
WSSV is a DNA virus. A unique life cycle between host and the virus provides a
unique opportunity for developing antiviral agents. Thymidine kinase is a key
enzyme involved in the pyrimidine salvage pathway and the enzyme catalyzes the
phosphorylation of deoxythymidine to thymidine monophosphate in the presence of
Mg2+ and ATP. In normal cell proliferation, the enzymatic activity of TK is strictly
regulated. On the other hand, rapid dividing cells required high level of TK activity
for DNA synthesis. Taking the knowledge of this difference, development of TK
specific inhibitor represents a specific merit for the development of chemotherapeutic
agents for the treatment of virus or virus-directed cancer. We have successfully cloned
and sequenced the thymidine kinase gene from WSSV as well as functional expressed
the corresponding enzyme. A modified UV-Vis spectrophotometric assay, based on
coupled-enzyme assay method, was also developed for rapid enzymatic activity assay.
Preliminary inhibition studies showed ambiguous effects on commercially available
nucleoside and analogues tested. In addition, structural homology simulation and
docking experiments have been performed to elucidate the substrate-binding
mechanism and virtual screening of putative inhibitors. In the period of this project,
we will evaluate the inhibitory effect of these putative inhibitors. Moreover, the
functional role of putative active site residues and the possible inhibitory mechanism
will be verified by site-directed mutagenesis and subsequent inhibition assay. With
the execution of the project, the possible binding mode of both substrate and putative
inhibitors will be obtained. The obtained information may help to rationally design
more and potent drug or prodrug for antiviral therapy. On the other hand, the
molecular biology techniques and drug screening technologies will be transferred
from the research laboratory to the biotech company in order to help the biotech
company to establish the research capability for future development and growth.
NSC95-2113-M-009-019-CC3(95R961)
--------------------------------------------------------------------------------------------------------

Title:The Research on Antiviral Drug Screening of Thymidine Kinase from White Spot
         Syndrome Virus
Principal Investigator:Wu Tung-Kung
Sponsor:Navigator Biotechnology Co., Ltd.
Keywords: White Spot Syndrome Virus, Thymidine Kinase, Molecular Docking,
          Drug Screening


    The crustacean-infected white spot syndrome virus (WSSV), widely spread in
most Asia countries as well as in Gulf of Mexico and South America, caused high
shrimp mortalities and severe economical damage to shrimp culture industry. The
WSSV is a DNA virus. A unique life cycle between host and the virus provides a
unique opportunity for developing antiviral agents. Thymidine kinase is a key
enzyme involved in the pyrimidine salvage pathway and the enzyme catalyzes the
phosphorylation of deoxythymidine to thymidine monophosphate in the presence of
Mg2+ and ATP. In normal cell proliferation, the enzymatic activity of TK is strictly
regulated. On the other hand, rapid dividing cells required high level of TK activity
for DNA synthesis. Taking the knowledge of this difference, development of TK
specific inhibitor represents a specific merit for the development of chemotherapeutic
agents for the treatment of virus or virus-directed cancer. We have successfully cloned
and sequenced the thymidine kinase gene from WSSV as well as functional expressed
the corresponding enzyme. A modified UV-Vis spectrophotometric assay, based on
coupled-enzyme assay method, was also developed for rapid enzymatic activity assay.
Preliminary inhibition studies showed ambiguous effects on commercially available
nucleoside and analogues tested. In addition, structural homology simulation and
docking experiments have been performed to elucidate the substrate-binding
mechanism and virtual screening of putative inhibitors. In the period of this project,
we will evaluate the inhibitory effect of these putative inhibitors. Moreover, the
functional role of putative active site residues and the possible inhibitory mechanism
will be verified by site-directed mutagenesis and subsequent inhibition assay. With
the execution of the project, the possible binding mode of both substrate and putative
inhibitors will be obtained. The obtained information may help to rationally design
more and potent drug or prodrug for antiviral therapy. On the other hand, the
molecular biology techniques and drug screening technologies will be transferred
from the research laboratory to the biotech company in order to help the biotech
company to establish the research capability for future development and growth.
C95065(95.05.01-96.04.30)
--------------------------------------------------------------------------------------------------------

Title:A Study of Quantitative Structure-activity Relationships And Virtual Screening
Principal Investigator:Jinn-Moon Yang
Sponsor:National Science Council
Keywords:QSAR, Molecular Docking, GEMDOCK

       Molecular docking and quantitative structure-activity relationship(QSAR)are
 core technologies in computer-aided rational drug design. Previous studies have
 revealed that protein-ligand docking conformation is close to naturally crystal
 structure, but the scoring function of docking still obtains few or no relationship
                                                      (
 between predicted energy and truly biological activity e.g., binding affinity or IC50).
 3D QSAR analysis such as CoMFA and COMBINE also suffers several challenges,
 such as superposition of steric structures or selection of molecular descriptors. In
 recent five years, we have focused on computer-aided drug design and published
 seven journal papers1-7. We have successfully developed a molecular docking and
 virtual drug screening tool, GEMDOCK1,2,7, and a QSAR prediction tool,
 GEMQSAR8,9. Although our GEMDOCK performs very well predictive power, its
 scoring function still could not precisely correspond to real biological activity of
 ligands. And GEMQSAR, like other 3D-QSAR methods, also encounters the
 problems of superposition of 3D molecular structures.
      In this project we will improve the prediction performance of GEMDOCK and
 GEMQSAR by combining molecular docking tool and QSAR analysis. Original
 GEMQSAR uses general evolution strategy for feature selection and partial least
 square analysis for building QSAR model. Here, we will combine our method with
 kNN(k-nearest neighbor)and PLS. Then, according to their molecular similarities,
 we will choose kNN or PLS to enhance predictive power of GEMQSAR. The
 precision of scoring function in GEMDOCK will be improved through
 pharmacophore profile derived form GEMQSAR analysis. This scoring function will
 make the docked results and energies corresponding to real binding modes.
      The core steps in this project are described as follows: We first obtain multiple
 docked conformations against specific target by GEMDOCK. With docking process
 and protein-ligand interactions, superimposition of molecules will be naturally made.
 We can extract information of interactions, structures, and physico-chemical
 properties from superimposed molecules, and use GEMQSAR to build QSAR model.
 The pharmacophore profile of GEMQSAR will be the searching guide and basis of
 enhancing pharmacophore scoring function of GEMDOCK. With this profile we
 could modify the parameters in docking process, improve the precision of scoring
 function, and make docked conformations and energies reflecting real binding.
 Finally, we will apply this profile to virtual screening of GEMDOCK revealing the
 discrimination between binding conformations and improving accuracy of virtual
 screening. This system will complement defects of docking method and QSAR
 analysis, accelerate drug discovery process and save cost. For validating and testing
 our method, we will adopt several well-known and broadly applied cases to study,
 such as HIV-1 protease inhibitors or AChE inhibitors, and apply to practical targets,
 for example, inhibitor screening of AGHO.
       We will develop a virtual high-throughput screening system combining QSAR
and molecular docking. This system will establish specific pharmacophore profile for
interesting target protein. These profiles will be able to apply to real protein
inhibitors/substrates searching and predict biological activities of lead compounds.
We believe this research would be valuable for protein research and drug discovery.
NSC95-2113-M-009-007(95R224)
------------------------------------------------------------------------------- -----------------

Title:An Integrated Analysis of Molecular Structures And Interactions in Biological
       Systems
Principal Investigator:Jinn-Moon Yang
Sponsor:National Science Council
Keywords:Protein-protein Interaction, Docking, Domain-domain Interaction,
              Co-evolutionary Analysis Pathway


     Protein-protein interactions play a central role in numerous processes in a cell and
are one of the main issues of functional proteomics. Prediction of protein-protein
interactions and binding sites is crucial to provide insight into the nature of
macromolecular recognition and biochemical mechanisms. The numerous
physiological and pathological cell processes depend on protein-protein interactions,
which can be influenced by small molecules offers possibilities for the treatment of
human diseased states. Recently, a number of protein-protein interactions have been
considered to be pharmaceutical targets and some methods have been employed for
the successful identification of small molecule modulators of these protein-protein
interactions. We believe that protein-protein interactions and protein recognitions
should be the key roles of understanding how the expressions of proteins within a cell
translate into biological functions.
     In the first year, this project is aimed to develop a novel computational tool that is
able to predict domain-domain interaction. As physical interactions between protein
domains are fundamental to the workings of a cell, domain-domain interactions can be
useful for validating, annotating, and even predicting protein interactions. We will
describe a novel approach to computationally derive domain interactions in S.
cerevisiae from multiple data sources, including protein-protein complexes (e.g., PDB
and SCOP), protein-protein interaction databases (e.g., DIP and MIPS), and protein
domain sequences (e.g., Pfam). In addition, we will analyze protein binding-site
characteristics and develop a new scoring function for protein-protein docking by
combining empirical-based and knowledge-based scoring functions.
    In the second year, we further prove the usefulness of such an integrative
approach by applying the derived domain interactions (derived in the first) to predict
and validate protein–protein interactions. A novel domain-domain interaction
probability model will be also developed. Based on the new scoring function and
binding-site analysis achieved in the first year, we will use our well-developed generic
evolutionary method to develop a new protein-protein recognition tool. In addition,
we will explore small molecules, which may influence protein-protein interaction, by
mining from crystal complexes in PDB and by enhancing our well-developed
protein-ligand docking tool (GEMDOCK1-6) through the structure-based screening of
chemical libraries.
     In the third year, we finally will construct a new domain-domain network and a
protein-protein interaction network to analyze biochemical pathways based on the
achievements of the first two years. Moreover, we will evolve the networks of other
species, including Drosophila, C. elegans, and Homo sapiens. We believe that the
networks combining the small molecules and biochemical pathways play a key role to
understand physiological and pathological cell processes and to offer possibilities for
the treatment of human diseased states. Finally, these pathways will be evaluated on
several practical applications and a web-base service will be built to help researchers
to proceed with further studies.
NSC95-2627-B-009-001(95R632)
------------------------------------------------------------------------------------------------

Title:A Computational Model for Predicting Antigenic Evolution of Influenza A
       Viruses
Principal Investigator:Jinn-Moon Yang
Sponsor:National Science Council
Keywords:Avian    Influenza,    Human     Influenza, Antigenic   Evolution,
         Antigen-antibody Interaction, Vaccine, Feature, Machine Learning,
         Decision Tree, Genetic Algorithm


     Avian and novel human influenza probably cause disastrous damage to human
society and economics.We will combine computer-aided methods with viral genome
and protein information to build up a model for predicting antigenic evolution of
influenza A viruses. The model can predict antigenicity of prospective influenza
viruses on basis of global hemagglutinin protein sequences (from 1968-2005, 974
isolates). In our preliminary studies, we have acquired higher accuracy (91.7%) than
previous studies. The preliminary model, which could predict antigenic distance, was
built up by decision tree and genetic algorithm. Our previous studies, including 1)
global comparison and evolutionary trend of dengue envelope glycoprotein, and 2) a
better regression model for predict antigenic variation of influenza A/H3N2 viruses by
using support vector machine, were announced in international conferences.
In this project, we have five objectives:
1. To construct databases of viral genes, antigen sequences, and protein structures. We
   will supply web services
2. We will apply this model to vaccine selection and development
3. This model can recognize and predict important sites which are directly related to
   antigenic drift
4. By machine learning and statistical methods, to construct a computational model
   for predicting antigenic evolution of influenza A viruses
5. We will apply this model to antigen-antibody interactions, and other viruses, such
 as dengue virus We have had studied in machine learning methods for many years.
 Furthermore, we have cooperated with Dr. Chwan-Chuen King (Institute of
 Epidemiology, National Taiwan University) to study influenza from June, 2004.
 We had applied our model to the studies of Taiwanese influenza virus isolates. In
 the future, this predictive model will be used to predict the antigenic variant of
 influenza A virus, possible epidemic strain in next outbreak, and vaccine selection.
NSC95-2745-B-009-001(95R666)
---------------------------------------------------------------------

Title:An Artificial –Bio Hybrid Nano-System Capable of Sensing And
       Regulation(2/3)
Principal Investigator:Yuh-Shyong Yang
Sponsor:National Science Council
Keywords:Silicon, Nanowire, Biosensor, Bioregulator, Bionic System


     We plan to design and fabricate a hybrid nano-system for the sensing and
regulation of biochemical reaction on a semiconductor chip. Owing to the large
surface to volume ratio, the conductance of a semiconductor nanowire (or a nanotube)
is significantly affected by the surface charge. Many research groups have exploited
in this property to develop highly sensitive biosensors. However, essentially all
nanowire technologies reported in literature so far are still having difficulty to
efficiently reproduce the biosensors’performance with nanowire FET devices. This
issue limits its expanded purpose in research, not to mention its potential application,
especially for detecting a complex system at bioreaction level. The three principal
goals of this project are therefore, 1) to develop a highly sensitive and selective
nano-biosensor, of which the scheme must be highly reproducible and compatible
with low-cost Si process, 2) to develop a novel bioactivity regulator, which uses a
locally-intensified electric field to induce the conformational change of enzymes
immobilized on a Si nanowire, and therefore regulates their activities, 3) to construct a
bionic system at bioreaction level by integrating the above-mentioned nano-biosensor
and bioregulator developed in this project. The strategies to meet our goals are a) to
develop a nanowire biosensor with novel and reliable processes, b) to integrate
enzyme engineering and SAMs technology for biosensing and bioregulation, c) to
construct a bionic system with a cascade of bioreactions on the nano-electronic
devices, and d) to integrate and assemble the biosensing, bioregulating, and NEMS
technologies into a multi-functioned bionic platform.
NSC95-2120-M-009-003(95R251)
---------------------------------------------------------------------


Title:The Mechanism of Action And Biological Function of Sulfotransferase
Principal Investigator:Yuh-Shyong Yang
Sponsor:National Science Council
Keywords:Dehydroepiandrosterone Sulfotransferase, Neurosteroids, Drosophila
              Melanogaster, Neuropharmacology


       Sulfonation of biomolecules has been well-known to take place in lots of
organisms, from prokaryotes to vertebrate species, and responsible for variety of
critical biological functions including detoxification, hormone regulation and even the
dynamic balance of catecholamine hormones and neurotransmitters. Since the insect
system, Drosophila melanogaster, has been perpetually utilized as models for
pathological and neuropharmacological research, at least in part, this tiny living being
is able to mimic the intricate neuroregulation that are effective for most of the
neuronal network in human beings. However, no evidence had been revealed that
whichever cytosolic sulfotransferase is in attendance on this important animal model.
The aim of this study is to determine for the first time the protein expression of
cytosolic sulfotransferase, dehydroepiandrosterone sulfotransferase (DHEA-ST), has
subsisted in Drosophila neuronal system. Result from immunoblot analysis
demonstrated the protein expression of DHEA-ST was present in brain homogenates
of Drosophila. Drosophila DHEA-ST enzymatic activity was also detected by using a
sensitive fluorometric assay in fly brain lysates. Moreover, six populations of
DHEA-ST immunoreactive neurons were observed in sections of Drosophila brain
and DPM2, and several bundles of positive neuron fibers were visualized as well.
These data firstly demonstrated the presence of DHEA-ST in Drosophila brains and
provided a foundation for the future investigation of cytosolic sulfotransferase in
invertebrate species.
NSC95-2311-B-009-001(95R379)
--------------------------------------------------------------------------------------------------------

Title:
Principal Investigator:Yang Yuh-Shyong
Sponsor:National Science Council
Keywords:Biological Electronics, Course Evaluation, New Technological
               Development


      This project (subordinate project three) is to assist the first category
development course. The research tasks are: course development assistance, outer
course evaluation and training of potential teachers. The issues of course development
assistance are instructing the establishment and operation of equipment in biological
electronics laboratory, instructing the experiment and product manufacture and
offering services about development of instruction materials. The content of outer
course evaluation include evaluation the inner thought of the course, evaluation of the
instruction materials, in-field observation of the teaching experiment , and
investigation of course satisfaction. The training of potential teachers has two
categories: research and study of fusion, and that of project-making.
     The research items which will be achieved include: (1) to establish the basic
equipment in biological electronics laboratory; (2) to train the project members in
using newly set-up equipments; (3) to assist writing and editing basic instruction
materials of biological inspection in electronics techniques course; (4) The program of
training the potential teachers in summer vacation.
NSC95-2514-S-009-001-GJ (95R664)
---------------------------------------------------------------------


Title:Understanding The Molecular Mechanism of Enzyme Families from The
       Analysis Their Sequences, Structures and Functions
Principal Investigator:Yang Yuh-Shyong
Sponsor:National Science Council
Keywords:Amidohydrolase Superfamily, Imidase, Sulfotransferase, Nucleotide
               Kinase, FAVA (Feature Amplified Voting Algorithm), GemDock
               (Generic Evolutionary Method for Molecular DOCKing)


      Amidohydrolase superfamily includes imidase, allantoinase, dihydroorotase,
hydantoinase, and other related enzymes that were proposed based on the rigidly
conserved structural domains in identical positions. Although the overall sequence
homology within the enzyme superfamily can be lower than 30%, these conserved
regions, compose of four histidines, one aspartate, and one modified lysine which is
bridging one or two metal ions are involved in the catalysis. In addition,
sulfotransferases and nucleotide kinases were also found to be similar in structure but
different in sequence identity (<20%). The conservation of structural features may be
correlated to specific functions such as substrate/ligand binding. The correlation
between sulfotransferase and nucleotide kinase is very worth study. The goal of this
proposal is to understand the mechanism and functional properties of these enzyme
families by using combined techniques of bioinformatics, biochemistry and molecular
biology. A newly developed bioinformatics tools, FAVA (Feature Amplified Voting
Algorithm) has been used to identify critical amino acids for metal binding in
amidohydrolase superfamily. In order to knowing how imidase select specific
substrate and the role of metals, an upgraded FAVA will be developed to couple with
a well-developed tool -- Gem Dock (Generic Evolutionary Method for molecular
DOCKing). FAVA and Gem Dock will be used to identify critical amino acids
through available sequences and structures, respectively, in the amidohydrolase
superfamily. Biochemical and molecular biological studies will follow to confirm the
prediction and to further study the mechanism of imidase. In particular, we would like
to distinguish the need for either a binuclear or mononuclear metal center in imide
hydrolysis. In the future, we wish our understanding in this superfamily will provide
us tools to protein engineering imidase for novel enzymatic properties. The tools
developed in this research will also be used to study sulfotransferase family in order
to clarify the functional and evolutionary correlation in sulfortransferase and ucleotide
kinase families. We plan to study the functional features of sulfotransferase and
nucleotide kinase families by mapping their sequence in a topological concept with
their secondary and tertiary structures.
NSC95-2627-B-009-004(95R635)
-------------------------------------------------------------------------------------------------------

Title:Elucidation of The Functions of REP1 in Drug Resistance in Candida Albicans II
Principal Investigator:Yun-Liang Yang
Sponsor:National Science Council
Keywords:Candida Albicans, Drug Resistance,Trans-regulatory
           Factors,REP1,Functional Study


     In the past decade, yeast infections have played an important role in nosocomial
infections. In the United States the yeast infection ranks as the fourth most common
cause of nosocomial bloodstream infection. Among the fungal pathogens, Candida
albicans is the most frequently isolated fungal pathogen in humans and has caused
morbidity in seriously debilitated and immunocompromised hosts. In Taiwan the
prevalence of nosocomial candidemia increased 16-fold from 1981 through 2000 at
one teaching hospital in Taipei. In addition to side effects, drug resistance is the major
concern for currently available antifungal agents. Overexpression of CDR1, an efflux
pump, is a major mechanism contributing to drug resistance in C. albicans. To
dentify the trans-regulatory factors of CDR1, we have constructed CDR1
promoter-lacZ (CDR1p-lacZ) fusion gene and transformed this construct into
Saccharomyces cerevisiae strain such that the lacZ gene can be used as the reporter to
monitor the activity of CDR1 promoter when a Candida library was introduced into
the cells. One of the genes obtained were named REP1 (Regulator of Efflux Pump 1).
       Overexpression of REP1 increased the expression of CDR1p-lacZ in S.
cerevisiae.The aim of this proposal is to elucidate the functions of REP1 in C.
albicans. reliminary data have showed that overexpression of REP1 in S. cerevisiae
altered the susceptibility to antifungal agents. Furthermore, REP1 and CaNDT80, a
previously isolated regulatory factor of CDR1, share similar structure in their DNA
binding domain. To study the function of REP1 in C. albicans, especially to
determine if REP1 is also involved in drug resistance in C. albicans, we will construct
a rep1/rep1 C. albicans mutant to assess the effect. We have successfully replaced
one copy of the wild-type REP1 gene by the selective marker ARG4 to generate the
REP1/rep1 heterozygous mutant. We are in the process to construct the rep1/rep1
homozygous mutant. In addition, to investigate the functional relationship between
REP1 and CaNDT80, we will also construct a rep1/rep1 Candt80/Candt80 double
mutant in C. albicans to determine their combinational effect. Furthermore, to reveal
different functional domains of the REP1, it will also be worthy while to perform
domain swapping between REP1 and CaNDT80.
NSC95-2320-B-009-001-MY2 (95R493-1)
--------------------------------------------------------------------------------------------------------

								
To top