The structure of poly(dA).poly(dT) in aqueous solution has been studied by using 1H two-dimensional nuclear Overhauser effect (2D NOE) spectroscopy and relaxation rate measurements on the imino and nonexchangeable protons. The assignments of the 1H resonances are determined from the observed cross-relaxation patterns in the 2D NOE experiments. The cross-peak intensities together with the measured relaxation rates show that the purine and pyrimidine strands in poly(dA).poly(dT) are equivalent in aqueous solution. The results are consistent with a right-handed B-form helix where the sugars on both strands are in the C2'-endo/anti configuration. These observations are inconsistent with a proposed heteronomous structure for poly(dA).poly(dT) [Arnott, S., Chandrasekaran, R., Hall, I. H., & Puigjaner, L. C. (1983) Nucleic Acids Res. 11, 4141-4155]. The measured relaxation rates also show that poly(dA).poly(dT) has fast, large-amplitude local internal motions (+/- 20-25 degrees) in solution and that the amplitudes of the base and sugar motions are similar. The motion of the bases in poly(dA).poly(dT) is also similar to that previously reported for poly(dA-dT).poly(dA-dT) and poly(dG-dC).poly(dG-dC) [Assa-Munt, N., Granot, J., Behling, R. W., & Kearns, D. R. (1984) Biochemistry 23, 944-955; Mirau, P. A., Behling, R. W., & Kearns, D. R. (1985) Biochemistry 24, 6200-6211].

We report here the biologically active conformation ofacetylcholine when bound to the high-affinity state of the receptor from Torpedo cakfornica. The acetylcholine conformation was determined in the free and bound states by proton NMR two-dimensional nuclear Overhauser effects. In agreement with x-ray crystallographic data, acetylcholine in solution has an extended conformation with an average distance between the acetyl methyl and choline methyl protons of =5 A. When bound to the acetylcholine receptor, acetylcholine adopts a conformation where the acetyl methyl group is dose (3.3 A) to the methyl groups of the choline moiety. This bent conformation places the oxygens adjacent to one another and allows the methyl groups to form an uninterrupted hydrophobic surface over the rest of the acetylcholine molecule. The significant difference between the fre-and bound-state conformations implies that structure-activity studies based solely on molecular modeling strategies must be approached with caution. To understand the interactions of biologically interesting molecules, it is necessary to know the shape ofthe interaction sites as well as the conformations of the interacting molecules. We report here the biologically active conformation of acetylcholine (AcCho) when bound to the high-affinity (desensitized) state ofthe acetylcholine receptor (AcChoR) from Torpedo californica. The exact location ofthe AcCho binding site on the a-subunits of the AcChoR is still unknown, although it has recently been localized to residues 158-216 by several groups using a variety oftechniques (1-4). In contrast to the uncertainty in the AcCho binding site, crystal structures are available for AcCho and other agonists that also bind to the AcChoR (5-7). However, we show here that the conformation of AcCho in its receptor-bound state is distinctly different from its conformation in solution and in the crystalline solid state. MATERIALS AND METHODSAll experiments used purified AcChoR in asolectin vesicles. The receptor was purified from the electroplax organs of T. californica as described (8). We have shown that under the conditions used in this study, AcCho binds to the AcChoR and that selective proton relaxation measurements are very sensitive to this binding. [Under these conditions the ratio of Kd (acetylcholine)/Kd (nicotine) is 0.14.] Conditions (see text) have been determined to distinguish between the contributions from site-specific binding ofAcCho to the AcChoR and contributions from nonspecific interactions of AcCho both with the lipid and with other areas of the receptor (8).All proton NMR spectra were acquired on a 360-MHz Bruker AM series spectrometer with an Aspect 3000 computer. The two-dimensional nuclear Overhauser effect (NOE) data sets were obtained in the phase-sensitive mode (9) and were processed with Dennis Hare's Fourier transform NMR program. The temperature in all experiments was held at 250C to within PC with the Bruker variable temperature unit by passing heated nitrogen gas through the probe.The conf...

No abstract

Measurements on the thymine imino proton relaxation rates have been used to study various structural and dynamic properties of 53 +/- 15 base pair long poly(dA-dT). Below 10 degrees C, the relaxation is dominated by dipolar magnetic interactions. At 1 degrees C the relaxation of the transverse magnetization is exponential (R2 = 124 s-1), but the relaxation of longitudinal magnetization is highly nonexponential due to spin-diffusion effects (initial decay rate constant of 28 s-1 and a slower rate of approximately 2.5 s-1 after equilibration of spin polarization). Neither a rigid-rod model nor simple wormlike motions can account for the observed low-temperature relaxation behavior. However, when localized internal motions of the base pairs (three-state jump model) are allowed for, a good fit of the experimental data is obtained by using a correlation time for internal motion of 7 X 10(-10) s and an angular displacement of the bases of +/- 32 degrees relative to the helix axis. The observed R2/R1 ratio for the thymine imino proton yields a value of 1.14 +/- 0.08 A for the imino proton nitrogen distance. Nuclear Overhauser effect (NOE) measurements establish that the base pairing in poly(dA-dT) is Watson-Crick in solution and not Hoogsteen. Exchange of the T-imino protons with H2O dominates the longitudinal relaxation above 28 degrees C (activation energy of 17 +/- 2 kcal and an exchange rate of 5 +/- 2 s-1 at 300 K). Similar values have been reported for the A X T base pairs in DNA restriction fragments and for A X U base pairs in poly(A) X poly(U). These observations can be explained by a model in which exchange of T-imino protons occurs as a result of a single base pair opening, with a rate that is approximately independent of nearest-neighbor sequences and DNA length. Our observations appear to be inconsistent with a soliton model of proton exchange.

A method is presented that uses selective proton Nuclear Magnetic Resonance (NMR) relaxation measurements of nicotine in the presence of the acetylcholine receptor to obtain relative binding constants for acetylcholine, carbamylcholine, and muscarine. For receptors from Torpedo californica the results show that (a) the binding constants are in the order acetylcholine greater than nicotine greater than carbamylcholine greater than muscarine; (b) selective NMR measurements provide a rapid and direct method for monitoring both the specific and nonspecific binding of agonists to these receptors and to the lipid; (c) alpha-bungarotoxin can be used to distinguish between specific and nonspecific binding to the receptor; (d) the receptor--substrate interaction causes a large change in the selective relaxation time of the agonists even at concentrations 100x greater than that of the receptor. This last observation means that these measurements provide a rapid method to monitor drug binding when only small amounts of receptor are available. Furthermore, the binding strategies presented here may be useful for the NMR determination of the conformation of the ligand in its bound state.