Perspectives in Biology and Medicine

GENE REGULATION: THE INVOLVEMENT OF STEREOCHEMICAL RECOGNITION IN DNA-SMALL MOLECULE INTERACTIONS LAWRENCE B. HENDRY, FRANCIS H. WITHAM, and ORVlLLE L. CHAPMAN* A broad spectrum of naturally occurring substances and synthetics is known to influence responses in a variety of biological systems. Substances classified as hormones, phytohormones, pheromones, antigens, carcinogens, antibiotics, etc., range widely in structure from exotic steroids to straight chain aliphatics; the functional groups on these small molecules are equally diverse, for example, the presence of esters, amines, alcohols, phosphates, ethers, and others. Despite this chemical diversity, it appears logical to expect some similarity in the mechanisms by which this array of molecular species influences biological processes. The action of some regulatory molecules may involve their initial interactions with proteins which somehow eventually result in gene action. Accordingly, O'Malley and co-workers [1,2] have detailed such a mechanism in the action of steroid hormones with the nonhistone protein components of DNA. Hence, the preferential binding of biologically active molecules in general to chromosomal proteins could result in the exposure and transcription of deoxyribonucleotide sequences with subsequent production of specific receptor proteins. However, the details concerning the origin of a specific protein(s) which recognizes one regulatory molecule from a host ofother substances in a cell or the regulation of transcription and translation of genetic information into proteins involved in a given cellular function are not clear. Indeed, inspection of the chemistry of DNA and RNA indicates that while the multiplicity of nucleotide sequences may be responsible for coding the primary structural differences among proteins, the array of specific recognition sites *The authors are affiliated with: Department of Chemistry, Pennsylvania State University , 152 Davey Laboratory, University Park, Pennsylvania 16802; Department of Biology, Pennsylvania State University; and Department ofChemistry, University ofCalifornia, Los Angeles, respectively. We wish to express our sincerest appreciation to Professor Kenneth Craig, Edinboro State College, for helpful discussions particularly regarding the importance of hydrogen bonding in small molecules. We also thank Dr. Harold Kaufman, J. T. Baker Chemical Co., and ProfessorJohn Reissner, Pembroke College, for their interest and encouragement. 120 I Lawrence B. Hendry et al. ¦ Gene Regulation in an organism cannot be explained on this basis alone. We maintain that specific proteins involved in molecular biosyntheses and/or chemoreception are most likely a reflection of template modifications which result from the stereochemical recognition between regulatory molecules and nucleic acids. As an initial step in support of this hypothesis, we will illustrate with the use of selected space-filling Corey-Pauling-Koltun (CPK) molecular models the importance of chirality1 in the stereochemical intercalation of regulatory molecules into DNA. The structures of a diverse set of biologically active molecules are illustrated in figure 1. These chemicals were selected because they represent a range of classes of substances which exhibit different kinds of biological activity and because they possess a variety of structural features . Interestingly, many of the chemicals are steroids; others have conformations which are closely related to the basic steroidal skeleton. If the steroidal shape is of any consequence in DNA-small molecule interactions , it would be logical to predict that nucleic acids might contain components which can accept, and perhaps conform to (or vice-versa), the shapes of certain biologically active molecules. The structures of the four major bases in the classical Watson and Crick DNA bound in the two common base pairings are illustrated in figure 2. The outline of the thymine-adenine (T-A) base pair (fig. 2C) bears a striking resemblance to the general steroid skeleton. Similarly, the cytosine-guanine (C-G) base pair (fig. 2D) resembles the general steroid skeleton with an additional ring. This similarity in the structure formed by hydrogen bonding ofthe base pairs and the structure ofthe steroid nucleus have been observed by others [3]. It follows from these observations that chemicals possessing structures similar to base pairs could interact with DNA. Precedence for direct binding of biologically active chemicals with DNA already has been established experimentally for several compounds including certain carcinogens [4], the drug ethidium bromide [5] and the antibiotic, actinomycin [6], where it has been shown that they fit or intercalate between base pairs. The intercalation of actinomycin appears to be highly specific in...