Scientific Report 2006

Molecular Biology

Organic Synthesis and Selective Drug Delivery

S.C. Sinha, R.A. Lerner, Z. Chen, S. De, S. Das, S. Abraham, F. Guo

Our main research interests are synthesis of biologically important natural and nonnatural molecules, synthetic methods, and antibody catalysis in organic synthesis and selective drug delivery. During the past year, we focused on 3 different classes of compounds: the anticancer adjacent bis-tetrahydrofuran annonaceous acetogenins, the antibacterial macrocyclic lactones sorangiolides, and nonnatural small-molecule drugs that target G protein–coupled receptors. In our work on antibody catalysis, we developed a proadapter approach for production of the chemically programmed aldolase antibody 38C2 and new doxorubicin prodrugs that are catalyzed by antibody 38C2 faster than the previously reported drugs are.

Selective Chemotherapy With Catalytic Aldolase Antibodies

For selective chemotherapy, we intend to develop drug conjugates and prodrugs that will target cell-surface receptors, such as the glycoprotein integrins α_vβ₃ and α_vβ₅. These integrins are directly implicated in tumor angiogenesis; they are overexpressed in the vasculature of angiogenic tumors and in numerous cancer cells but are less expressed on quiescent blood vessels. Using antibody 38C2 and small-molecule antagonists of α_vβ₃ and α_vβ₅ (or a targeting agent), we developed antagonist-38C2 conjugates, also known as chemically programmed 38C2 (Fig. 1).

Fig. 1. Schematic drawings of the diketone and the proadapter strategies used to produce chemically programmed antibody constructs that target cells expressing the integrins αvβ3 and αvβ5. Abbreviations: Ab, antibody; TA, targeting agent.

The conjugation between the targeting agents and the antibody takes place in the binding sites though the diketone or vinyl ketone linkers. Because the vinyl ketones are highly reactive, we used the corresponding acetone adduct as the prolinker, which undergoes 38C2-catalyzed reaction to produce the active linker before the active linker reacts. This strategy has been termed the proadapter approach. The conjugates prepared by both approaches bound efficiently to cells expressing α_vβ₃ and α_v>β₅, including human breast cancer cell lines MDA-MB-435 and MDA-MB-231, and inhibited the growth of both the primary tumors and secondary metastasis in distant organs.

Development of these strategies for the formation of antibody constructs can have a large effect on the treatment of various diseases, including cancer.

In the alternative approach, we are developing prodrugs that can be efficiently activated by the aldolase antibodies 38C2 and 93F3. These antibodies will be targeted to tumor cells or the tumor vasculature by using antagonists of α_vβ₃ and α_vβ₅. In the past year, we developed new doxorubicin prodrugs that are not only more stable than the previously reported analogous prodrugs but also activated faster by using antibody 38C2. We also produced 38C2-antagonist conjugates. The conjugates bound efficiently to MDA-MB-231 cells expressing α_vβ₃ and α_vβ₅. We also found that the modified antibody can activate new doxorubicin prodrugs.

Therefore, we have all the tools to investigate prodrug therapy in animal models. The selective chemotherapy studies are carried out in collaboration with C.F. Barbas, Department of Molecular Biology, and
B. Mueller, La Jolla Institute for Molecular Medicine, San Diego, California.

Synthesis of Natural and Nonnatural Small Molecules

Total synthesis of naturally occurring and biologically important compounds is important not only for confirming their structures but also for producing the compounds and their analogs for comprehensive biological evaluations. To synthesize these compounds, we are developing methods that involve both antibody catalysis and common synthetic routes. In the past year, in addition to synthesizing sorangiolides, which are naturally occurring macrocyclic lactones, and the library of bis-tetrahydrofuran annonaceous acetogenins, we focused on small-molecule nonnatural ligands of G protein–coupled receptors. The studies on the synthesis of these ligands are carried out in collaboration with E. Roberts, Department of Chemistry.

Sorangiolides (Fig. 2) are weakly active antibacterial compounds. Our goal is to synthesize the highly active sorangiolide analogs. Thus, we have developed synthetic routes that can provide the macrocyclic structure of sorangiolides. Using an intermediate, we will synthesize both the natural and nonnatural molecules. For other bis-tetrahydrofuran acetogenins, which are among the most active cancer agents and are toxic to several human cancer cell lines at much lower concentrations than doxorubicin is, we developed methods that can provide all the stereoisomers of asimicin and bullatacin. The new methods involve a bidirectional approach. Now, we are pursuing synthesis of the 64 stereoisomers of asimicin and bullatacin.

Fig. 2. Structure of sorangiolides A and B (top) and a general structure of bis-tetrahydrofuran annonaceous acetogenins (bottom).

Publications

Das, S., Li, L.-S., Abraham, S., Chen, Z., Sinha, S.C. A bidirectional approach to the synthesis of a complete library of adjacent-bis-THF annonaceous acetogenins. J. Org. Chem. 70:5922, 2005.

Li, L.-S., Babendure, J.L., Sinha, S.C., Olefsky, J.M., Lerner, R.A. Synthesis and evaluation of photolabile insulin prodrugs. Bioorg. Med. Chem. Lett. 15:3917, 2005.

Popkov, M., Rader, C., Gonzalez, B., Sinha, S.C., Barbas, C.F. III. Small molecule drug activity in melanoma models may be dramatically enhanced with an antibody effector. Int. J. Cancer 119:1194, 2006.