ROBERT W. ARMSTRONG, Associate Professor

(b. 1956) B.S. 1979, University of California, San Diego; Ph.D., 1984, Colorado State University; NIH Postdoctoral Fellow, 1984-86, Harvard University.

The goals of our research program involve the synthesis of natural products (and designed molecules) of potent and relevant biological activity and elucidation at the molecular level of the chemical events responsible for these activities. These bioactivities include inactivation processes and site-specific recognition of macromolecules (DNA, RNA, proteins), factors important in the understanting and design of antitumor drugs, antiviral agents, antihypertensive compounds, immune response regulators, cell-surface recognition factors, and general enzyme inhibitors. This program provides an opportunity to combine synthetic skills with biological evaluation and molecular modeling in order to understand the mechanism of action of the target molecules. Emphasis is strongly directed towards utilizing a broad spectrum of synthetic techniques and procedures in achieving the goals of each thesis. Multiple isotope and 2D and 3D NMR experiments provide an opportunity to elucidate structural information of drug-biopolymer interactions. Use of biochemical tools in order to solve particular aspects of each project provides a well-rounded background in synthetic and bioorganic chemistry. Several major areas of interest are listed below.

Natural Products- Total synthesis, methodology, analog design

We have targeted the synthesis of novel antitumor antibiotics in order to establish a stereospecific protocol for elucidating their mechanism of action and to establish the molecular basis of their specificity. Carzinophilin and ficellomycin are two such agents whose potential therapeutic capacity appears to be a result of interference with DNA synthesis due to the potential alkylating ability of the [3.1.0] bicyclic aziridines found in each compound. Synthetic schemes which provide stereospecific modular approaches are central to the total synthesis. A combinatorial array used to synthesize large libraries of structural analogs of these antitumor agents has been developed. Application to the SAR profile of other targeted drugs is the goal of this work.

A second series of targets include the calyculin antibiotics, potent phosphotase inhibitors.

The renin inhibitor cyclothiazomycin contains several unnatural amino acids and conjugated thiazoles which are implicated as potential DNA-binding motifs. The nucleoside FR-900848 contains a unique and unprecedented polycyclopropane system with interesting structural implications. Cyclindrospermopsin and polycavernoside are highly potent marine toxins.

Carbohydrate Chemistry- methodology, conformational analysis, 3D design

Because of the ubiquitous role carbohydrates play in biology, carbohydrate analogs are valuable tools for the study of biochemical systems. Since the chemistry of sugars is dominated by the reactivity of the glycosidic bond, a great deal of effort has gone into the synthesis and study of C-glycosides. We are interested in developing synthesis of C-oligosaccharides in which the "floppy" glycosidic linkage has been re-engineered to produce unnatural linkages with predictable conformations. We are building large and rigid (carbohydrate-derived) water-soluble molecules which may extend between two distant sites (30-120) of a biopolymer. For example, targets such as the acetylene-substituted oligomer shown below provide a base structure which may be modified at either termini or along the oligomer. Synthesis of more rigid fused bicyclic linear dimers has also been achieved. Further modification of these substrates should provide a synthesis of highly complex C-oligosaccharides, including the pentasaccharide analog of ganglisoide G2.

Implementation of an iterative strategy involving the fused oligosaccharides provides an entry to the synthesis of natural products containing this moiety. Maitotoxin is the most complex target in this series.

Applications involving polydentate binding requirements include the generation of dendridic structures containing multiple ligands with predictable three-dimensional structure. Such substrates may provide a systematic study of, for instance, viral adhesion properties. Targets under development include the adamantyl based dendromer shown below.

The synthesis of conformationally fixed oligosaccharides may be achieved by the formation of polycyclic C-oligosaccharide derivatives. For example, the sialyl Lewis A termini (A) can be structurally mimicked by a C-glycoside (B).

DNA/RNA Modifications-Asymmetric synthesis of modified oligos, methodology, biopolymer synthesis

A combination of synthetic modifications of target antitumor agents along with screening of their in vitro activity can potentially identify locations and modifications induced by these compounds. For example, the use of gel electrophoresis in combination with chemical degradation of radiolabeled oligonucleotides allows us to elucidate unambiguously the formation of interstrand crosslinks induced by carzinophilin. The molecular basis for this recognition/activation can also be addressed via the synthesis and incorporation of modified oligonucleotides.

Modeling studies of chemical intermediates in the reaction of antitumor drugs with DNA provides some understanding of the molecular basis for specificity of these drugs for particular bases or sequences. For example, in the carzinophilin system the reaction takes place via the initial monoalkylation of one strand followed by interstrand delivery of a second less reactive functional group. Identification of these types of intermediates in the reaction process provides unique information in understanding the mechanism of action of such drugs.

Publications

1. Armstrong, R.W.; Teegarden, B.R. "Synthesis of a-methyl-1'-2'-Dideoxycellobiose: A Novel C-Disaccharide," J. Org. Chem., 1992, 56, 915.

2. Armstrong, R.W.; Moran, E.J. "Stereoselective Synthesis of a 1-Aza-Bicyclo-[3.1.0]-Hex-2-Ylidine Dehydroamino Acid Derivative Related to the Azinomycin Antitumor Antibiotics," J. Am. Chem. Soc. 1992, 114, 371-373.

3. Armstrong, R.W.; Salvati, M.E.; Nguyen, M., "Novel Interstrand Cross-Links Induced by the Antitumor Antibiotic Carzinophilin," J. Am. Chem. Soc. 1992, 114, 3144.

4. Armstrong, R.W.; Tellew, J.E.; Moran, E.J. "Stereoselective Synthesis of E and Z 1-Aza-Bicyclo-[3.1.0]-Hex-2-Ylidine Dehydroamino Acid Derivatives," J. Org. Chem. 1992, 57, 2208.

5. Salvati, M.E.; Armstrong, R.W., "Simplified Method for the Isolation of Thermally Labile Drug-DNA Adducts: Characterization of Chlorambucil and Azinomycin B Alkylation Products," Tetrahedron Lett. 1992, 33, 3711.

6. Moran, E.J.; Tellew, J.E.; Zhao, Z.; Armstrong, R.W. "Synthesis and Intramolecular Reactions of Dehydroamino Acid Derivatives Derived from D-Arabinose and L-Serine: Application to the Synthesis of the Azinomycin Antitumor Antibiotics," J. Org. Chem. (in press.)

7. Sutherlin, D.P.; Armstrong, R.W. "Stereoselective Synthesis of Dipyranyl C-Disaccharides, " Tetrahedron Lett. ( in press)