Ethan Will Taylor

Ethan Will Taylor
Title Senior Research Professor
Expertise Medicinal, Computational, and Bioorganic Chemistry
Education B.Sc. (Chemistry), University of Winnipeg , 1981
Ph.D. (Pharmacology), University of Arizona , 1985
NIH Postdoc (Medicinal Chemistry), University of Arizona, 1986-1987
Office Eberhart Building , Rm 206
Phone 336.334.3114
E-Mail Email Dr. Taylor (UNCG address) | E-mail Dr. Taylor (Alternate Address)

My research group is involved with a variety of studies in the following general areas:

  1. Computational biology and genomic analysis
  2. Prediction of structure and function of macromolecules
  3. Computational chemistry/molecular modeling
  4. Computer-assisted drug design and drug synthesis
  5. Discovery, cloning and characterization of novel viral genes
  6. Antiviral and anticancer effects of selenium
  7. Nutritional (orthomolecular) medicine

Our research typically begins with theoretical genomic analysis, to develop hypotheses about biological structure and mechanisms, which are then validated in the laboratory by molecular biological and biochemical methods. This approach is applied to various problems of biomedical interest, particularly in the area of viral diseases. A key accomplishment is our demonstration that the genomic complexity of some viruses is greater than previously thought. For example, in the case of HIV-1 we have demonstrated the existence of several novel viral proteins encoded in “hidden” genes, and shown that they function as predicted by theoretical analysis.

In regard to theoretical studies, we are particularly concerned with certain problems in structural bioinformatics, such as:

  1. Identifying novel genes in genomic DNA and mRNA sequences
  2. Predicting protein function and modeling protein structure, especially when homology to known macromolecules is low, and
  3. Developing approaches to assessment of the significance of remote sequence similarities.

In regard to experimental studies, an ongoing focus of the lab has been the identification of viral glutathione peroxidase (GPx) genes in HIV-1 and hepatitis C virus, and other novel viral proteins expressed by ribosomal frameshifting (which makes them non-obvious or “hidden” genes). Several of these genes have now been cloned in our lab. Because GPx is a selenoprotein, and selenium is an essential trace antioxidant mineral, these findings indicate a key role for antioxidant nutritional status in the pathogenesis and progression of AIDS and other chronic viral diseases. This has led to an interest in investigating dietary antioxidants and phytochemicals as chemoprotective agents.

Selected examples of ongoing research projects

1. Redox regulation of activation of gene expression by Nuclear Factor- κB ( NF- k B ).

NF- k B is a cellular transcription factor which functions as a regulator of gene expression via its DNA-binding ability. It is of paramount importance in the immune system, because it serves to regulate hundreds of genes that are involved in immune activation, inflammation, and apoptosis (programmed cell death). Significantly, some viruses, including HIV-1, have evolved to incorporate NF- k B recognition sequences in their promoter regions, so that viral gene expression is activated along with other genes when NF- k B is “turned on”. Thus, there is considerable interest in the use of NF- k B inhibitors, a class that includes many dietary antioxidants, as antiviral and anti-inflammatory chemoprotective agents. This role for antioxidants arises because NF- k B is regulated in a complex manner by the redox status of the cell; furthermore, several selenium containing proteins (“selenoproteins”) play critical roles in this process. In the cytosol, various oxidative stimuli are known to induce translocation of NF- k B into the nucleus, where it must be reduced by the small regulatory protein thioredoxin in order to bind to DNA, and thereby initiate transcription and gene expression. Two cellular selenoproteins have been shown to regulate these events: in the cytosol, glutathione peroxidase, via its antioxidant effects, can inhibit NF- k B activation and nuclear translocation; in the nucleus, the selenoprotein thioredoxin reductase is required to maintain thioredoxin in a reduced state, so that it in turn can reduce a specific disulfide in an NF- k B heterodimer, enabling the molecule to bind to DNA.

Our lab has recently published the first explicit model of the oxidized form of NF- k B, which clearly shows that it is unable to bind to DNA prior to reduction, because access to the DNA binding cleft is blocked. This model is also a snapshot of the state of NF- k B that would be the initial target for binding of thioredoxin, enabling the subsequent conversion of NF- k B to the reduced form, in which the DNA binding cleft can open, and the free thiol groups of the cysteine residues are able to engage in sequence-specific recognition of the DNA target.

Taylor - Figure 1

Figure 1. The structure of the oxidized form of transcription factor NF- k B (right) was generated by molecular modeling, starting from the experimental structure of the reduced NF- k B/DNA complex (left). An inter-subunit disulfide bridge in the oxidized form (yellow spheres) prevents the entry, binding and sequence-specific recognition of DNA.

2. An HIV-1 encoded peptide mimics the DNA binding loop of NF -κB and binds thioredoxin.

A conserved HIV-1-encoded peptide sequence with local similarity to the DNA binding loop of NF-κB was previously identified by our lab, in the -1 reading fame overlapping the protease gene (thus named pro-fs, for protease frameshift). We have shown that recombinant pro-fs is a potent activator of the HIV-1 promoter in transfected cells, an effect which is abolished by specific mutations to either pro-fs or to the NF-κB binding sites in DNA. Furthermore, we have demonstrated experimentally that pro-fs can mimic the ability of NF-κB to bind to thioredoxin, a biological reducing agent essential for NF-κB activation. Figure 2 shows a n energy-minimized model of an HIV-1 pro-fs peptide fragment (RYRSRCYSI) bound to thioredoxin, created by mutating in silico the bound NF- k B peptide fragment (RFR YVCEGP).

Figure 2: Homology model of the HIV-1 pro-fs peptide bound to thioredoxin (Trx), based on an NMR structure of the NF-kB/Trx complex , PBD file 1MDI. The modeled Trx-pro-fs complex is shown in protein ribbon (Trx) and CPK colored stick (pro-fs) renditions, with the homologous NF- k B peptide from the Trx-NF- k B complex overlaid in green for comparison. The N and C termini of the peptide ligands are indicated. Notable differences between the bound pro-fs and NF- k B peptides include the substitution in pro-fs of a Tyr for F2 (Phe) of NF- k B, and a Tyr (Y7 of pro-fs) for a Glu of NF- k B. Possibly in compensation for the greater bulk of Y7, pro-fs has a smaller Ser at the position of Y4 of NF- k B. Energy calculations suggest that Trx binding of the pro-fs peptide is at least as favorable as that for binding of the NF- k B peptide.

Taylor - Figure 2


(from a total of about 55 research articles and 10 professional articles)

Zhao, L., Cox, A.G., Ruzicka, J.A., Bhat, A.A., Zhang, W. and Taylor E.W. (2000): Molecular modeling and in vitro activity of an HIV-1 encoded glutathione peroxidase. Proc. Natl. Acad. Sci. USA 97: 6356-6361.

Taylor, E. W., Cox, A. G., Zhao, L., Ruzicka, J., Bhat, A., Zhang, W., Nadimpalli R. G., and Dean R. G. (2000): Nutrition, HIV and drug abuse: the molecular basis of a unique role for selenium. J. AIDS Human Retrovirol. 25: S53-S61.

Su, G., Min, W. and Taylor, E.W. (2005) An HIV-1 encoded peptide mimics the DNA binding loop of NF-κB and binds thioredoxin with high affinity. Mutat. Res. 579: 133-148.

Zhao, L., Olubajo, B. and Taylor, E.W. (2006) Functional studies of an HIV-1 encoded glutathione peroxidase. Biofactors, 27: 93-107.

Peng, D., Zhang, J, Liu, Q., Taylor, E.W. (2007) Size effect of elemental selenium at nano size (Nano-Se) and supra-nutritional levels on selenium accumulation and glutathione S-transferase activity. J. Inorg. Biochem. 101: 1457-1463.

Chandrasekaran, V. and Taylor , E.W. (2007) Molecular modeling of the oxidized form of Nuclear Factor-κB suggests a mechanism for redox regulation of DNA binding and transcriptional activation. J. Mol. Graph. Model. 26: 861-7.

Chandrasekaran, V, Ambati, J., Ambati, B.K. and Taylor, E.W. (2007) Molecular docking and analysis of interactions between vascular endothelial growth factor (VEGF) and SPARC protein. J. Mol. Graph. Model. 26: 775-782.

M.E. Kleinman, K. Yamada, A. Takeda, V. Chandrasekaran, M. Nozaki, J.Z. Baffi, R.J.C. Albuquerque, S. Yamasaki, M. Itaya, Y. Pan, B. Appukuttan, D. Gibbs, Z. Yang, K. Kariko', B.K. Ambati, T.A. Wilgus, L.A. DiPietro, E. Sakurai, K. Zhang, J.R. Smith, E.W. Taylor, J. Ambti (2008)  Sequence-and target-independent suppression of angiogenesis by siRNA via TLR3. Nature, 452-597.