- Stereochemical Recognition in Nucleic Acid-Amino Acid Interactions and its Implications in Biological Coding: A Model Approach
- Perspectives in Biology and Medicine
- Johns Hopkins University Press
- Volume 22, Number 3, Spring 1979
- pp. 333-345
- View Citation
- Additional Information
STEREOCHEMICAL RECOGNITION IN NUCLEIC ACID-AMINO ACID INTERACTIONS AND ITS IMPLICATIONS IN BIOLOGICAL CODING: A MODEL APPROACH* LAWRENCE B. HENDRKt and FRANCIS H. WITHAMt Recently, space-filling molecular models were employed to describe potential in vivo interactions between nucleic acids and a broad range of biologically active small molecules [1, 2]. Stereospecific recognition of certain small molecules, including chiral natural products, and DNA via binding at the phosphate backbone and intercalation between base pairs was postulated. As an extension of this study our initial observations concerning stereochemical recognition between molecular models of amino acids and nucleic acids are described herein. We postulate that the recognition process involves, in part, intercalation of amino acid R groups into double-stranded nucleic acids between two consecutive bases of its codon concomitant with complementary pairing with bases of the opposite strand (anticodon). The details of our hypothesis were developed in stages by considering initially single-stranded nucleic acidamino acid interactions. In this regard, a chronological account of our observations with models, followed by a description of our preliminary criteria for amino acid-nucleic acid interactions, is presented. Single-stranded RNA-Amino Acid Complexes Corey-Pauling-Koltun (CPK) space-filling molecular models of the common L-amino acids were individually positioned into each of the 16 possible base cavities of a single-stranded RNA model (the cavities were constructed with the bases in the anti conformation). In all cases, the *We express our most sincere appreciation to Professor Orville L. Chapman, Department of Chemistry, University of California at Los Angeles, for stimulating discussions regarding various concepts contained in this manuscript. tDepartment of Medicine, Medical College of Georgia, Augusta, Georgia 30901 . ^Department of Biology, Pennsylvania State University, University Park, Pennsylvania 16802.© 1979 by The University of Chicago. 0031-5982/79/2203-0073$01.00 Perspectives in Biology and Medicine · Spring 1979 | 333 intercalation was accomplished by initially binding the positively charged a-amino group of the amino acid(s) to the negatively charged phosphate oxygen of RNA as it might occur at physiological pH (fig. IA). The R group of each L-amino acid was extended subsequently into the cavity formed by two consecutive bases with the a-carboxyl group directed toward the 3' base (fig. IB), so that the carbonyl oxygen was in close proximity to the N7 cap of the 3' purine or Cs hydrogen of the 3' pyrimidine, as shown for phenylalanine between UU (proline, the wellknown helix breaker, was an exception). In this orientation, the R group was not considered properly "seated" when it extended above the Ck and Cs hydrogens of pyrimidines, N7 cap and Ce hydrogen of purines, or beyond the base-pairing hydrogen-bonding surfaces of the dinucleotide (fig. IG-/). A rather convenient, although somewhat crude, method for insuring proper positioning ofthe R groups according to this convention is to align the bases with the intercalated amino acid with a straightedge placed across these surfaces of the complex. Although each L-amino acid model could be intercalated with the a-carboxyl group oriented in the 5' direction (fig. IC), the convention with the carboxyl group in the 3' direction was used because, in tfiis orientation: (1) a recurring pattern of stereochemical recognition features was observed in many specific amino acid-dinucleotide complexes; (2) our assignments ofmany amino acids to a certain base cavity(ies) were reproducible; and, most important, (3) the convention was consistent witfi the current knowledge concerning the translation of amino acids from NH3+ to COO" in the 5' to 3' direction (vide infra). Surprisingly, when a given amino acid was positioned into a specific dinucleotide (often the cavity formed by the first two bases of its codon) according to the aforementioned procedures, we observed that in a number of instances there were conformatipns of L-amino acids in which: (1) aromatic R groups aligned (stacked) with aromatic portions of both bases in the cavity, and atoms of the R group overlapped with similar atoms of the bases; (2) heteroatoms and functional groups of hydrophilic R groups are aligned with those on both bases of the cavity; and (3) nonpolar aliphatic R groups filled hydrophobic regions between aromatic portions of the dinucleotide cavities. Thus, the above observations...