Are the stereochemistry and mechanism of action of thyroid hormones predicted by the structure of DNA?

Abstract

ARE THE STEREOCHEMISTRY AND MECHANISM OF ACTION OF THYROID HORMONES PREDICTED BY THE STRUCTURE OF DNA? TERRYf. SMITH* LAWRENCE B. HENDRY,t and EDWIN D. BRANSOME, fR.t Although the existence of rigorous limitations on the structures of biologically active molecules is well known, whether there is a rationale for diese constraints remains a mystery. We have observed previously that the structures ofbiologically active natural products are reflected in the unique physicochemical and topographic properties of DNA and/or RNA [I]. Since the information governing the structure as well as the function and metabolism of biologically active molecules is known to reside ultimately in the genetic template, we have reasoned diat die complementarity of these molecules with the structure of DNA may define a comprehensible stereochemical blueprint for this information. This blueprint, which we have termed a "stereochemical logic" [2], has led to the discovery ofa rationale for the genetic code [3] and has proved useful in the formulation of a dieoretical framework for the mechanism of action of steroid hormones [4]. In this paper we raise the question of whether this stereochemical logic is applicable to thyroid hormones. The naturally occurring iodothyronine hormones, L-3,5,3',5'-tetraiodothyronine (L-thyroxine, T4) and L-3,5,3'-triiodothyronine (T3) (fig. 1), regulate many aspects of mammalian cellular metabolism, including growth, respiration, and protein synthesis [5-8], While T4 is by far the most abundant product released from die diyroid gland, die biological potency ofT3 gready exceeds diat OfT4 [9]. Triiodothyronine arises primarily from 5' monodeiodination of T4 in peripheral target tissues. Monodehalogenation of T4 at the 5 position also occurs in vivo, resulting in the formation of 3,3',5'-triiodothyronine (reverse-Ts), a bioThe authors express their appreciation to Ms. Connie Natoli for secretarial assistance. ?Department of Medicine, Medical College of Georgia, Augusta, Georgia, 30912; and Veterans Administration Medical Center, Augusta, Georgia 30910. tDepartments of Medicine and Endocrinology, Medical College of Georgia, Augusta, Georgia 30912.© 1984 by The University of Chicago. All rights reserved. 0031-5982/84/2703-0395$01.00 408 I TerryJ. Smith, Lawrence B. Hendry, andEdwin D. Bransome,Jr. · ThyroidAction HO^------^ I ^^ F??, L-3,5,3· - triiodothyronine (T1) I^ ^\ .. ^ I coo© COMPLEMENTARY FIT INTO DNA HYDROGEN BONDING BIOLOGICAL H ? CAVITYTODNAACTIVITY ^r: «Si HO^^f I >^ F??,_ + T, L-3'-lsopropyl - 3,5-dllodothyronlne (lsopropyl TJ HO i r# H ? C00fc T??, L-3,3',5'-trllodothyronlne (Reverse T1) Fig. 1.—Correlation ofbiological activity oftriiodothyronine and related molecules with fit into cavities in DNA. logically inactive analog [9]. One synthetic analog of thyroid hormone, L-3'-isopropyl-3,5-diiodothyronine (isopropyl-T2), possesses greater biological activity than any ofthe naturally occurring iodothyronines [9]. Fit of T= and Related Molecules into DNA A Corey-Pauling-Koltun (CPK) space-filling model of the original Watson and Crick right-handed double-helical B form of DNA was constructed using known coordinates from the National Institutes ofHealth X-ray computer graphics system [1O]. Models were also constructed of T3, T4, reverse-T3, and isopropyl-T2 (figs. 2, 3) and then tested for fit (intercalation) between DNA base pairs. To fit the analogs into the double helix, it was necessary to uncoil partially the DNA model resulting in a cavity approximately 6 angstroms wide and 14 angstroms in length between neighboring base pairs (fig. 2). Two criteria had to be met: (1) the candidate molecule had to be accommodated widiin and fill the cavity between neighboring base pairs in the double helix without obviously extending beyond die surfaces of the base pairs or distorting DNA components; (2) heteroatoms capable of forming hydrogen bonds in the candidate molecule had to form Perspectives in Biology andMedicine, 27, 3 · Spring 1984 | 409 5' 3' ? 3' 5' 3' 3NE S«14A° G ñ a t«6 A0 5' 5'r HO 3' ÌQ 13' NH3** ? Fig. 2.—Block diagram depicting intercalation of T3 into DNA: sequence TA/GC in coiled DNA (left); partially uncoiled DNA showing cavity between TA/GC base pairs (middle); T3 inserted into cavity in TA/GC with asterisks denoting positions of hydrogen bonding and die saltbridge to die phosphate-deoxyribose backbone on adjacent strands of DNA (see text; cf. fig...