About Genomics and Infection

The Keck Center integrates four different areas of current technology:

• Functional Genomics
• Structural Genomics
• Chemical Genomics
• Biology of Infection

Functional Genomics

Functional genomics and proteomics is used to determine which proteins are encoded by the genomes of pathogens, what these proteins do, and how they interact with each other.

Click image. Close new window when done.

The complete sequence of the prototype PA01 strain of Pseudomonas aeruginosa (determined by Maynard Olson and his colleagues) was used to construct a P. aeruginosa DNA microarray containing ~5600 predicted open reading frames, which represents 98% of the genes of PAO1. Sam Miller’s laboratory used this array to evaluate gene expression profiles of one isolate from an infant with Cystic fibrosis without clinical lung disease and another from a patient with chronic bronchiectasis, a non-CF lung infection.

The expression of 213 genes, termed cystic fibrosis activated (CFA) genes, was increased in the CF strain (including transporters and genes involved in the remodeling of surface structures and in nutrient transport), and the expression of 120 genes, termed cystic fibrosis repressed (CFR) genes (including genes involved in O antigen synthesis), was decreased compared to the transcriptional profile of the PAO1 strain. By contrast, the transcriptional profile of the bronchiectasis strain was substantially different. In addition to these studies, ongoing work by Maynard Olson’s group indicates that strains from CF patients contain additional genes not found in laboratory strains, which may play a role in adaptation to the CF lung environment. These novel genes, and the CFA genes noted above, are candidates for further study and potential targets for development of new antimicrobial drugs.

As a basis for the design and development of new drugs for the treatment of the most virulent species of malaria, Plasmodium falciparum, Pradip Rathod’s laboratory is exploiting the newly developed tools of functional genomics in three specific areas:

1. They have found that P. falciparum from Southeast Asia rapidly develops resistance to new drugs due to a novel mutator-like phenotype, designated Accelerated Resistance to Multiple Drugs (ARMD). They are using P. falciparum microarrays as one approach to define the basis for this trait, with the goal of evading or defeating this response.
   
   
2. Using the known malarial drug target dihydrofolate reductase-thymidylate synthase (DHFR-TS) as a model, they are exploiting parasite-specific protein-protein interactions and metabolic circuitry to aid in drug design and RNA molecule (“aptamers”) to rapidly define new molecules that are essential for parasite survival; once identified these molecules become targets for rationale drug design in the chemical biology part of the Center.

Proteomics

A critical component of any functional genomics effort is proteomics, i.e., the ability to determine what proteins are being expressed in any given cell, and how this protein complement changes in response to growth conditions, environmental cues, infection, and candidate drug treatment.

Click image. Close new window when done.

Mass spectrometric analyses, incorporating the powerful ICAT (Isotope-Coded Affinity Tags) approach to comparative protein quantification developed by Center investigator Michael Gelb in collaboration with Rudi Aebersold, is being applied to the study of Pseudomonas by Miller’s group. While Stan Fields’ group is using the yeast two-hybrid system developed in his laboratory to screen for protein-protein interactions in protozoans and between microbial components and host Toll-like receptors, and, as part of the chemical biology aspect of the Center, for novel compounds that disrupt their interactions.

Structural Genomics

Structural genomics builds on functional genomics by determining the three-dimensional structure of the encoded proteins, and thus the physical basis for their functions and interactions.

Chemical Genomics

Chemical biology will exploit the information obtained by functional and structural genomics in order to develop new drugs and approaches for modulating or blocking the activity of these proteins.

As an example of the synergy of these approaches, Fan, Gelb, and Verlinde (see lab directory) have used the structures of glycolytic enzymes from Trypansomes, determined by Hol and Merritt, together with the available structures of the human functional homologs, to design parasite-specific, tight-binding inhibitors. The glycolytic enzyme glyceraldehyde phosphate dehydrogenase (GAPDH) uses NAD+ as a co-substrate, and the structure of the enzyme-NAD+ complex has been solved. Since NAD+ contains an adenosine moeity, we are designing adenosine analogs as GAPDH inhibitors (adenosine is a very weak inhibitor with a KI of 50 mM). Molecular modeling suggests that substituents can be added to the N6 of the adenine ring and to the 2'-OH of ribose. To date our best GAPDH inhibitor is N6-naphthylenemethyl-2'-(m-methoxybenzamido)-adenosine. This compound is a potent inhibitor of the parasite enzyme (KI = 200 nM), binding more than 5 orders of magnitude more tightly than adenosine, our original lead compound.

Click image. Close new window when done.

Figure shows N6-benzyl-2'-(m-methoxybenzamido)-adenosine bound in the active site of L. mexicana GAPDH (the active sites of T. brucei and T. cruzi GAPDHs are virtually identical). The 2'-substituent is predicted to sterically clash with residues of human GAPDH. Indeed, this compound does not inhibit human GAPDH at concentrations up to its solubility limit of ~40 mM. Further work to improve the potency of this class of inhibitors is exploiting combinatorial chemistry to prepare libraries of adenosine analogs containing substituents that are rationally chosen based on the X-ray structure of GAPDH. Similar studies are being carried out with trypanosomal phosphoglycerate kinase (PGK) and glycerol-3-phosphate dehydrogenase (GPDH). Baker is using protein structure prediction tools, including the ROSETTA program created by his group, to model new interactions and thereby facilitate the throughput of drug development. Samudrala is using a related approach to model molecular structures of interest in Pseudomonas.

Biology of Infection

The biology of infection component tests the utility of these new drugs and therapeutic approaches in model systems. These include in vitro systems and murine models of infection.

Burns, Miller and Wilson (see lab directory) are using cell culture systems to screen components of P. aeruginosa for their capacity to induce inflammation through human and mouse Toll-like receptors. This has led to the identification of molecular adaptations in the LPS component of the bacterial envelope that stimulate intense inflammatory mediator production through human but not murine Toll-like receptor 4. This information is being used to design and generate mice with human receptor systems, to provide an animal model through which to test the importance of these (and other) CF-specific bacterial adaptations in lung injury and to define approaches to ameliorate this injury. Similarly, Buckner and van Voorhis use cell culture and whole animal approaches to test candidate compounds for the treatment of protozoan infections, as noted, such as the GAPDH inhibitors and DHFR-TS inhibitors described above.