Human Neutralising Antibodies Cloned from Convalescent Patient B Cells: A New Route to Anti-Viral Therapy

A number of important viral pathogens such as Dengue, Hand Foot and Mouth disease, SARS and Avian flu threaten human health in Asia. While vaccines offer the best long term solution to protection against these viral pathogens new developments in antibody isolation offer exciting prospects for passive antibody based immunotherapies. Such therapies may offer protection for health and relief workers and the military entering contaminated areas, may alleviate and reduce apparent infections and my protect the very young , the very old and the immunosuppressed. Such passive immunotherapy has reached clinical approval for the prevention of RSV infection. In Singapore Dengue virus infections remain a source of concern. Working as a consortium of scientists from ETC A*Star, NUS and the local hospitals we have isolated and immortalised B cells clones from individuals recovering from Dengue infection. The mRNA from clones secreting Dengue neutralising antibodies was converted to cDNA using appropriate primers and heavy and light chain encoding clones expressed in a variety of vectors. Human neutralising antibodies were produced and are now being characterised in detail.

Quorum Sensing as a Target for Novel Antibacterial Agents

Although unicellular, bacteria are capable of complex patterns of co-operative behaviour that result from the coordination of the activities of individual cells. This is primarily achieved through the deployment of small diffusible signal molecules which facilitate the regulation of gene expression as a function of cell population density. This phenomenon is termed “quorum sensing” (QS). As the bacterial population density increases, so does the synthesis of QS signal molecules and consequently, their concentration in the external environment rises. Once a threshold concentration has been attained, activation of a signal transduction cascade leads to the induction or repression of QS target genes often incorporating those required for QS signal molecule synthesis so providing an auto-regulatory mechanism for amplifying signal molecule production. QS signal molecules are chemically diverse and many bacteria possess several interacting QS gene regulatory ‘modules’ employing multiple signal molecules from the same or different chemical classes. QS systems generally facilitate the co-ordination of population behaviour to enhance access to nutrients or specific environmental niches, collective defence against other competitor organisms or community escape where survival of the population is threatened. Since the QS systems of bacterial pathogens such as Staphylococcus aureus and Pseudomonas aeruginosa regulate virulence, they represent highly attractive targets for the development of novel therapeutic agents which attenuate infection through the blockade of bacterial intercellular communication. Natural products and synthetic signal molecule analogues have been identified as QS inhibitors and these offer considerable promise for the treatment of infections caused by multi-antibiotic resistant bacteria.

Mutation-resistant Antiviral

A common viral entry mechanism requires fusion of the viral and cellular membranes mediated by envelope glycoprotein which has a homotrimeric quaternary structure and forms a hairpin-like assembly of six-helix bundle (6HB) during the fusion event. The 6HB-mediated membrane fusion is common in all Class-1 envelope viruses including HIV-1. The 6HB employs a 3-on-3 locking mechanism in which 3 HRC (heptad repeating-carboxyl region) chains crosslink 3 HRN (heptad repeating-aminal region) chains to enable membrane fusion. This locking mechanism confers avidity due to multi-chain interactions and a high genetic barrier to mutations in the viral entry strategy despite the high mutation rate of their envelope protein. This mechanism also provides inspiration to our approach in designing mutation-resistant entry inhibitors of HIV-1.

T20/enfuvirtide, a synthetic HRC-peptide monomer designed to interrupt membrane fusion, is the first and only FDA-approved entry inhibitor against HIV-1. However, T20 becomes ineffective in 20% of AIDS patients due to the loss of efficacy by generating T-20 resistant HIV-1 mutants during the course of treatment. Our inhibitor design is based on a novel quaternary protein mimetic approach. Key elements include a covalent-link 3-parallel-chain construct mimicking the quaternary structure of gp41 in its pre-fusion state and an inhibition mechanism mimicking the multimeric interactions found in the highly conserved 6HB formation. Our hypothesis is that covalent-linked, 3-chain quaternary mimetics (called 3a mimetics) of HRC peptides can confer multi-chain binding to the HRN region in a multi-chain locking mechanism that may lead to mutation-resistant HIV fusion inhibitors. We also extend our design to 2-chain HRC mimetics (called 2a mimetics) which may bind to the HRN as a 2-on-3 locking mechanism.

In contrast, T20 and other single-chain peptide inhibitors with a 1-on-3 binding mechanism lacks the advantages offered by the quaternary mimetics and is susceptible to gp41 mutations. This report will describe all three types of inhibitors as a model to further our understanding of gp41 mutations and viral fusion mechanism in developing mutation-resistant entry inhibitors.

Sialidases as Drug Targets for Respiratory Disease

Sialdiases, or neuraminidases, catalyse the removal of sialic acid from the surface of cells. Sialic acid is a negatively-charged carbohydrate that decorates various glycoconjugates, and its terminal location has been exploited by a number of pathogens for attachment during initial infection. Sialic acid can also be removed by pathogens to reveal other cryptic carbohydrate receptors for cell attachment, and the released sugar can be used by certain bacteria as an energy source. The influenza virus neuraminidase is the best studied member of the sialidase superfamiliy, the structure of which has been used for the development of Tamiflu and Relenza, drugs used for the treatment of influenza. My own laboratory has been studying the structural biology of sialidases from paramyxoviruses and a number of bacteria. Studies on the paramyxovirus hemagglutinin-neuraminidase enabled the development of an inhibitor that protects mice from a lethal paramyxovirus, and has promise in the treatment of human parainfluenza. The Streptococcus pneumoniae genome encodes three sialdiases, and we have recently determined the structures of two of these, NanA and NanB. Gene knockout experiments have shown that these sialidases are crucial for bacterial colonisation in the respiratory tract and in blood, and hence may be good drug targets. Pseudomonas aeruginosa, another respiratory pathogen, is a particular problem for cystic fibrosis patients where it can form biofilm in the lungs. Mouse experiments have shown that knocking out the sialidase of P. aeruginosa severely reduces biofilm formation. We have recently determined the structure of this sialidase that shows an unusual trimeric fold with alterations in the active site that suggest that sialic acid is not the substrate, but rather pseudaminic acid, another carbohydrate that decorates the flagella and pili of certain bacteria. We have embarked on the structure-based development of inhibitors of these bacterial sialidases.

Preventing Infection after Cataract Surgery with Antibiotics: Current Evidence and Controversy

Infection of the intraocular tissues of the eye or infectious endophthalmitis, is a devastating and potentially sight-threatening complication occurring after cataract surgery. Although relatively rare with a reported incidence ranging from 0.03 - 0.3 %, endophthalmitis can rapidly lead to destruction of intraocular tissues, resulting in a blind and painful eye, if appropriate treatment is delayed or not instituted. Prevention of this dreadful complication is therefore of paramount importance. However, there is presently a relative lack of definitive, well-controlled studies on prevention methods for endophthalmitis. Such prospective clinical studies would require a large number of subjects and would be extremely costly to undertake. The evidence supporting a range of prevention methods such as topical povidone-iodine antisepsis, topical antibiotics, subconjunctival antibiotics and intraocular (intracameral) antibiotics, would be critically examined, including evidence from recent multicentre clinical trials. The merits and shortfalls of administering antibiotic prophylaxis via the different routes, namely topical, subconjunctival, intracameral and infusion fluid would be reviewed in the light of this current evidence.

Targeting Hbv Infected Cells Using a Monoclonal Antibody with T Cell Receptor-Like Specificity

Hepatitis B chronic infection still affects 350 million people, 75% of which live in Asia. Current therapeutic options suppress HBV replication but elimination of the virus is rarely achieved. Specifically delivering cytokines or anti-viral drugs to HBV infected hepatocytes could increase the efficacy of anti- HBV therapy. One approach to achieve this is the use of monoclonal antibodies that specifically target HBV infected cells. To this end, we have generated antibodies that recognize an HBV peptide/MHC-class I complex (HBV c18-27/A201) expressed at the surface of HBV infected cells. With this approach, antibodies bind epitopes classically recognized by CD8 T lymphocytes through their T cell receptor (TCR) and do not bind free HBV virions or antigens in the circulation. In addition to their potential in delivering cytokines and/or antiviral drugs, these antibodies could be linked with a traceable dye, opening the possibility to obtain precise diagnostic information about the in vivo distribution of HBV infected hepatocytes in the liver of patients with chronic hepatitis B.

Preventing Pili’s Growth in Bacteria: Insights from Simulation and Rational Analysis of Experimental Data

Pili are adhesive multi-subunit fibres assembled on the surface of many pathogenic bacteria. In several systems such as E. Coli and Salmonella enterica this occurs via the chaperone-usher pathway: in the periplasm, a chaperone donates a beta-strand to a pilus subunit to complement its incomplete immunoglobulin-like fold; this is replaced at the outer membrane with a beta-strand formed from the N-terminal extension (Nte) of an incoming pilus subunit. Molecular simulations demonstrated that this reaction occurs through an intermediate; and that this intermediate occurs because a binding pocket is present (or forms dynamically) to accept the Nte while the chaperon has not been yet released. I will show further evidence of the importance of dynamics in tuning the affinities of the various subunits in the case of Pap and Fim pili from uropathogenic E. Coli. Such affinities are also responsible for the precise ordering of subunits in such pili and the growth termination which in turn are important for their function. These results altogether suggest that pilus growth could be hindered by developing antibacterial agents which target the dynamical correlations in the subunit-chaperon complexes.

Developing Antimicrobial Peptides with Increased killing Ability and Decreased Cytotoxicity

The international surge in infectious disease with attendant human suffering and morbidity has significantly raised research activity in the area of anti-infectives. Defensins, a major family of natural peptides found in mammals and known for their broad spectrum antimicrobial activity has earned them the name of “natural antibiotics”. As a component of the innate immune system which protects mucosal surfaces such as the front of the eye, mouth and lungs as a first line of defense against infection. In humans, there are six characterized alpha-defensins, human neutrophil peptides (HNP1 to HNP6) and four beta-defensins (HBD1 to HBD4). Based on the spatial distribution of the cysteine linkages, the mammalian defensins can de divided into two major groups termed alpha- and beta-defensins which are produced in neutrophils and mucosal surface epithelial cells of humans. However, defensins have some limitations such as toxicity at higher concentrations, expensive to synthesize, properties that are not optimized for human application. At the Singapore Eye Research Institute we have developed a multi-disciplinary approach to engineering new anti-microbial peptides. Short-branched and linear molecules with moderate levels of hydrophobicity and net charge have been found to have little or no cytotoxicity to host mucosal epithelial cells, while exhibiting low MIC values. In addition, these have been shown to be effective toward fungus, as well as clinical strains of Pseudomonas some of which are gentamicin resistant and a MRSA. Recent studies suggest that defensins may have much broader functions, acting as chemo-attractant agents for monocytes and dendritic cells. Therefore we have examined the immunomodulatory properties of the native peptides and defensins analogues. A possible advantage of defensins over conventional antibiotics is their positive participation in wound healing.

Defensins stimulate the proliferation of fibroblasts and epithelial cells, increase the expression of genes involved in wound healing responses, further suggesting that the function of defensins in the host system may be well beyond direct killing of microorganisms.

Antimicrobial Surfaces and Materials: An Engineering Approach

As the use of man-made materials and devices such as catheters, cardiac pacemakers, and prosthetic implants in the human body continues to escalate, the risk of biofilm-associated infection will concomitantly increase. The increasing use of antibiotics to combat infections is recognized as the main cause for the emergence of antimicrobial resistance which has become a major public-health problem worldwide. Biofilms-related problems are also present in food, marine and other industries. Our research is targeted at the development of novel materials and surfaces which can inhibit microbial adherence and growth, thus providing protection against biofilm formation. The strategies we employed aim to achieve effectiveness against a spectrum of microbial species, provide long lasting effects and do not adversely affect the material or the system of interest other than the targeted species. These strategies include (i) the grafting of quaternary ammonium groups on different materials such as polymer films, microbeads and membranes, fabric, cellulose and activated carbon, (ii) nanoparticulate antimicrobial polymer for bone cement and endodontic applications, and (iii) the functionalisation of metal surfaces with anti-adhesive and bactericidal biocompatible polymers.

Natural Products as a Source of New Antibiotics

Historically, natural products have been a valuable source of new antibiotics. Despite a slowing in the rate of new discoveries it is still the case that many of the antibacterial drugs entering the market or in clinical trials are natural products or are derived from natural product templates. However, doing natural product discovery is expensive and resource intensive and the number of companies who retain capabilities in this area continue to fall. It is important that we develop ways to make the discovery process more efficient and improve success rates. MerLion Pharma has extensive and biologically diverse microbial and plant collections and has developed a chemical fingerprinting system which allows the characterisation of tens of thousands of natural extracts. This is providing unique insights into the extent and breadth of natural product chemistry and is opening up opportunities to make natural product discovery better and accessible to all.

Starting with Fragments

The talk will describe recent progress in the application of fragment-based methods to develop biologically active small molecules. The approach involves using a range of biophysical and structural techniques to identify small molecular fragments that bind to a target protein with low affinity but high ligand efficiency. There then follows iterative cycles of chemical synthesis and structure-based rational design as the fragments are elaborated to increase affinity and specificity. There will be a particular focus on finding fragments and using them as starting points to develop enzyme inhibitors against Mycobacterium tuberculosis. This work fits as part of joint international efforts to develop new lead compounds to aid TB drug discovery.

Structure and Folding Dynamics of Novel Antimicrobial Peptides

Jagadeesh. Mavinahalli^1,2, SP Liu², L. Zhou^2,3, J. Li^2,3 , C. Tang⁴, Roger W Beuerman^2,3, Chandra Verma1 ¹Bioinformatics Institute, Singapore; ²Singapore Eye Research Institute, Singapore; ³Dept of Opththalmology, National University of Singapore; ⁴Pharmaceutical Microbiology Laboratory, Singapore General Hospital.

We need antimicrobials which are robot and resist pathogen resistance. Nature has already designed such molecules like defensins. Can we learn and exploit the knowledge embedded in these fascinating molecules? A novel peptide dimers which are designed using a ten residue C-terminus fragment of human defensin-3 will be discussed. These molecules are found to have increased broad spectrum antimicrobial activity and decreased cytotoxicity on epithelial cells. They have high potential to be optimized for therapeutic use. Focus of the discussion will be on optimization strategy, in particular exploiting simulations techniques.

We find that high temperature MD simulation are advantageous which spans larger conformational space of these molecules giving clues about possible secondary structures which might adapt under various conditions. We have observed that one of our most active compounds is unstructured in water, which is in consistent with CD data, and might be structured upon reaching its target, that is lipid bi layer. We propose that an unstructured solutions structure with a propensity for secondary structure, which can be observed by high temperature MD simulation, is very important. Our evaluation effort to access factors which might be important for optimal activity will be shared.

Structure Based Approaches for the Discovery of Anti-Infective Ligands

Data Base Mining tools have been developed to aid the search for small molecule compounds that will bind and inhibit target proteins. A range of different protein targets are being investigated including allosteric proteins of the glycolytic pathway as well as kinases and other effector and inhibitor proteins that control cell proliferation. In most cases the 3D structure of the target is known and a druggable pocket can be identified for docking studies. We have developed a searchable database of 4 million commercially available compounds (EDULISS) which is used in a variety of types of searches including our docking program LIDAEUS that is implemented on BlueGene. Potential hits are screened using various biophysical methods.

Networking the Bacterial Virulence Regulons

To gain an uphand in pathogen-host interactions, bacterial pathogen produce a wide array of virulence factors. These virulence factors are known to play critical roles at the different stages of bacterial infection. For example, some virulence factors are essential for acute infection whereas the others contribute to chronic infection. The genes encoding these virulence factors of different functions can be functionally grouped into various virulence regulons. Typically, these virulence regulons are under tight genetic regulation and modulated by various environmental cues, including bacterial quorum sensing signals, host signals, and other environmental conditions. Therefore, understanding microbial virulence regulation demands systemic approaches to study complex interactions among pathogens, host and environment. This talk will summarize our latest findings on the signalling mechanisms with which bacterial pathogens modulate production of different sets of virulence factors. Understanding the signalling networks that control the expression of key virulence regulons seems to hold a bright promise for developing new strategies to prevent and control bacterial infections.

A Better Vancomycin – Application of Cytotopic Membrane-Localising Technology to an Antibiotic

Membrane-associated proteins are of central importance in the regulation of signal transduction, vesicular transport, receptor recycling and many other essential processes in biology. Certain of these proteins (e.g Ras, Src, Rab and MARCKS), possess C or N-terminal lipophilic groups (such as myristoyl or palmitoyl) that insert into the hydrophobic core of a lipid bilayer. Elsewhere in these proteins, clusters of basic amino acids bind to negatively charged components of the cell membrane. The combination is known as the “myristoyl-electrostatic switch”. It has been mimicked with synthetic peptides which, when attached to therapeutic proteins, enable their localisation on mammalian cell membranes and thereby the therapeutic manipulation of cell surface phenotype. This process is known as cytotopic modification and has a wide variety of applications such as protection of the vasculature of solid organs prior to transplantation and in cell therapy. These will be outlined briefly. Bacteria have a distinct membrane composition with components not found in mammalian cells. We have developed a strategy to target the glycopeptide antibiotic vancomycin to bacterial cell membranes. We sought firstly to increase the local concentration at the site of action (cell wall synthesis), secondly to increase the affinity for the target enzyme and thirdly to access and inhibit other membrane-bound bacterial targets.

Biased combinatorial libraries of peptides based on the effector sequences of myristoyl-electrostatic switch proteins were ligated to libraries of membrane-insertive hydrophobic elements. This approach used an easily prepared common C-terminal vancomycin derivative which permitted facile coupling of these amphipathic peptides. The conjugates were highly active against a broad range of bacteria, including vancomycin-resistant enterococci (VRE), methicillin-resistant S. aureus (EMRSA) and glycopeptide intermediate-resistance S. aureus (GISA). The strategy allowed for rapid optimization of anti-bacterial activity with enhanced affinities for bacterial membranes compared to eukaryotic ones and led to potent broad spectrum compounds that were only weakly haemolytic against mammalian erythrocytes but lytic to bacteria.

Mechanistic studies showed that the compounds were bacteriocidal rather than simply bacteriostatic and that enzymes other than bacterial transpeptidase appeared to be targeted resulting in an inhibition profile more similar to ramoplanin than vancomycin (the compounds were shown also to inhibit the intracellular production of the peptidoglycan precursor Lipid II). Membrane localisation was a key determinant of potency and toxicity. Lead compounds have been identified and further development of these agents is planned.

Optimization and Clinical Validation of a Pathogen Detection Microarray
Wong CW, Heng CL, Wan Yee L, Soh SW, Kartasasmita CB, Simoes EA, Hibberd ML, Sung WK, Miller LD

DNA microarrays used as 'genomic sensors' have great potential in clinical diagnostics. Biases inherent in random PCR-amplification, cross-hybridization effects, and inadequate microarray analysis, however, limit detection sensitivity and specificity. Here, we have studied the relationships between viral amplification efficiency, hybridization signal, and target-probe annealing specificity using a customized microarray platform. Novel features of this platform include the development of a robust algorithm that accurately predicts PCR bias during DNA amplification and can be used to improve PCR primer design, as well as a powerful statistical concept for inferring pathogen identity from probe recognition signatures. Compared to real-time PCR, the microarray platform identified pathogens with 94% accuracy (76% sensitivity and 100% specificity) in a panel of 36 patient specimens. Our findings show that microarrays can be used for the robust and accurate diagnosis of pathogens, and further substantiate the use of microarray technology in clinical diagnostics.

Encapsulation of Living Cells and Virus Surface Modification: Two Technologies for Treating Pathogenic Microbes and Other Serious Diseases

John A. Dangerfield^1,2,3, Brian Salmons¹, Christoph Metzner^3,4, Eva Maria Brandtner¹, Mary M.L. Ng⁵ and Walter H. Günzburg^2,3,4,5
1Austrianova Singapore Pte Ltd; ²Christian Doppler Laboratory Foreign Module for Virology; ³Institute of Virology, Department of Pathobiology, University of Veterinary Medicine, Vienna, Austria; ⁴Christian Doppler Laboratory for Gene Therapeutic Vector Development, Vienna, Austria; ⁵Department of Microbiology, National University of Singapore, Singapore

Encapsulation of living cells into polymers of sodium cellulose sulphate allows potentially any cell type to be implanted into human or animal patients in an immunoprivileged setting. The capsules localise the cells to the area of implantation and allow the therapeutic, antimicrobial or cytotoxic biomolecules (or combinations thereof) to be produced and targeted to that specific site. Over nearly two decades we have developed this technology from proof-of-concept to a multifaceted, GMP standardised process and undertaken clinical trials for cancer. With respect to antimicrobial applications, we have achieved immunotherapy of a neurotoxic viral disease by systemic, long term delivery of neutralising monoclonal antibodies in immunocompetent animals and data will be presented. More recently we have initiated a project to encapsulate hybridoma cells producing dengue and West Nile virus neutralising antibodies and to test their release and efficacy in animal models. A second, novel approach for the treatment of infectious diseases utilises a unique method that we have recently developed to modify the outer surface of enveloped viruses. Recombinant glycosylphposphatidylinositol (GPI) anchored proteins can be engineered to contain bio-tags, expressed in eukaryotic cells, purified and then exogenously added to the surface of potentially any enveloped virus, and we have coined this process “viral painting”. So far we have shown this to be feasible for the modification of the envelopes of retrovirus, lentivirus and herpes virus. Viral painting can be achieved without any genetic manipulation of the virus or virus producing cells, the infectivity of the virions is not significantly affected by the process and the recombinant GPI proteins retain their biological function in all cases tested so far. Virus painting has many potential applications in biomedicine and bionanotechnology such as the fluorescence labelling of viral particles, modulation of virus-host interactions e.g. for gene therapy and vaccination approaches and manipulation of viral vectors using magnetic nanoparticles and these applications will be discussed.

Chemical Biology of Fungal Pathogenesis: New Strategies and Antifungal Metabolites

Loss of Abc3, an MDR efflux pump essential for virulence of the rice-blast fungus Magnaporthe grisea, leads to reduced viability and non-pathogenicity due to accumulation of a cytotoxic metabolite(s) in the infection structures. Using a novel strategy, we purified this endogenous metabolite (ATS; Abc3 Transporter Substrate) and showed that it is indeed the efflux substrate of Abc3. ATS shows a strong antifungal activity against yeast and filamentous fungi (including Magnaporthe). More importantly, yeast strains expressing M. grisea Abc3p did not show the cytotoxic effect of the ATS. Detailed characterisation of its physiological function during fungus-host interaction and its potential application in controlling fungal disease(s) would be presented. We’d also highlight the novel strategies that we have developed to identify new antifungal agents.

Challenges in TB Drug Discovery: Understanding the Mechanism of Action of New Chemical Entities

Tuberculosis (TB) is caused by a gram positive bacterium Mycobacterium tuberculosis. TB is an important public health problem worldwide due to multi-drug resistance, AIDS epidemic and mycobacterial persistency. TB is responsible for nearly 2 millions deaths each year. Novartis Institute for Tropical Diseases, Singapore is involved in discovery and early development of novel treatment for Tuberculosis. Through the chemical genetics approach with whole cell high through-put screens, we identify many New Chemical entities (NCE) having cell based activities. Understanding the mechanism of action of these new scaffolds and elucidating the molecular target/s is very critical for further medicinal chemistry efforts. In this presentation our recent findings on understanding the mechanism of action of bicyclic nitroimidazoles will be presented and discussed.