Structural Studies of Bcl-xL/ligand Complexes using 19F NMR

Fluorine atoms are often incorporated into drug molecules as part of the lead optimization process in order to improve affinity or modify undesirable metabolic and pharmacokinetic profiles. From an NMR perspective, the abundance of fluorinated drug leads provides an exploitable niche for structural studies using ¹⁹F NMR in the drug discovery process. As ¹⁹F has no interfering background signal from biological sources, ¹⁹F NMR studies of fluorinated drugs bound to their protein receptors can yield easily interpretable and unambiguous structural constraints. ¹⁹F can also be selectively incorporated into proteins to obtain additional constraints for structural studies. Despite these advantages, ¹⁹F NMR has rarely been exploited for structural studies due to its broad lines in macromolecules and their ligand complexes, leading to weak signals in ¹H/¹⁹F heteronuclear NOE experiments. Here we demonstrate several different experimental strategies that use ¹⁹F NMR to obtain ligand–protein structural constraints for ligands bound to the anti-apoptotic protein Bcl-xL, a drug target for anti-cancer therapy. These examples indicate the applicability of these methods to typical structural problems encountered in the drug development process.     

31P and19F NMR studies of glycophorin-reconstituted membranes: Preferential interaction of glycophorin with phosphatidylserine

Summary Glycophorin A, a major glycoprotein of the erythrocyte membrane, has been incorporated into small unilamellar vesicles composed of a variety of pure and mixed phospholipids. Nuclear spin labels including³¹P and¹⁹F have been used at natural abundance or have been synthetically incorporated in lipids to act as probes of lipid-protein interaction. Interactions produce broadening of resonances in several cases and it can be used to demonstrate preferential interaction of certain lipids with glycophorin.³¹P and¹⁹F probes show a strong preferential interaction of glycophorin with phosphatidylserine over phosphatidylcholine. There is some evidence that interactions are more pronounced at the inner surface of the bilayer and these results are rationalized in terms of the asymmetric distribution of protein and lipid.     

Difluorophosphate as a 19F NMR probe of erythrocyte membrane potential

Erythrocyte membrane potential can be estimated by measuring the transmembrane concentration (activity) distribution of a membrane-permeable ion. We present here the study of difluorophosphate (DFP) as a ¹⁹F NMR probe of membrane potential. This bicarbonate and phosphate analogue has a pK_a of 3.7±0.2 (SD, n = 4) and therefore exists almost entirely as a monovalent anion at physiological pH. When it is incorporated into red cell suspensions, it gives two well resolved resonances that arise from the intra- and extracellular populations; the intracellular resonance is shifted 130 Hz to higher frequency from that of the extracellular resonance. Hence the transmembrane distribution of DFP is readily assessed from a single ¹⁹F NMR spectrum and the membrane potential can be calculated using the Nernst equation. The membrane potential was independent of, DFP concentration in the range 4 to 59 mM, and haematocrit of the cell suspensions of 31.0 to 61.4%. The membrane potential determined by using DFP was 0.94±0.26 of that estimated from the transmembrane pH difference. The distribution ratios of intracellular/extracellular DFP were similar to those of the membrane potential probes, hypophosphite and trifluoroacetate. DFP was found to be transported across the membranes predominantly via the electrically-silent pathway mediated by capnophorin. Using magnetization transfer techniques, the membrane influx permeability-coefficient of cells suspended in physiological medium was determined to be 7.2±2.5 × 10^–6 cm s^–1 (SD, n=4). Offprint requests to: P. W Kuchel     

Fluoride transmembrane exchange in human erythrocytes measured with 19F NMR magnetization transfer

¹⁹F NMR spectra of sodium fluoride in suspensions of human erythrocytes were seen to yield separate resonances for the F^- populations inside and outside the cells. Selective saturation of the magnetization of the intracellular population gave rise to transfer of that saturation to the extracellular population. The extent of magnetization transfer was high and it was blocked by the capnophorin (band 3) anion exchange inhibitor 4,4-dini-trostilbene-2,2-disulfonic acid (DNDS). A series of magnetization-inversion transfer experiments was carried out for the range of intracellular fluoride concentrations of 11 mM to 136 mM and analysed using one-dimensional overdetermined exchange analysis. This yielded an estimate of the equilibrium exchange Michaelis constant and maximal velocity of 27 ± 3 mM and 180 ± 5 × 10^-16 mol cell^-1 s^-1, respectively. There was no alteration of exchange flux of fluoride at an intracellular concentration of 49 mM in the presence of 50 mM glucose; thus suggesting no interaction between glucose and anions in capnophorin-mediated exchange of solutes.     

Measurement of chemical exchange between RNA conformers by F NMR

Many noncoding RNA molecules adopt alternative secondary and tertiary conformations that are critical for their roles in gene expression. Although many of these rearrangements are mediated by other biomolecular components, it is important to evaluate the equilibrium relationship of the conformers. To measure the spontaneous interconversion in a bi-stable RNA stem loop sequence into which a single ¹⁹F-uridine label was incorporated, a ¹⁹F–¹⁹F EXSY experiment was employed. The kinetic exchange rate measured from EXSY experiments for this system was 37.3 ± 2.8 s⁻¹. The advantage of this approach is that exchange kinetics can be monitored in any RNA sequence into which a single ¹⁹F nucleotide is incorporated by commercial synthesis. This method is therefore suitable for application to biologically significant systems in which dynamic conformational rearrangement is important for function and may therefore facilitate studies of RNA structure–function relationships.     

Lysine side-chain dynamics derived from 13C-multiplet NMR relaxation studies on di- and tripeptides

Summary ¹³C NMR relaxation data have been used to determine dipolar auto- and cross-correlation times for the di- and tripeptides GK, KG and GKG, primarily to analyze lysine side-chain motional dynamics. In general, correlation times are largest for backbone positions and decrease on going through the lysine side chain, consistent with the idea of increased mobility at C and C methylenes. Correlation times, however, vary with the peptide ionization state. In the zwitterionic state of GK, for example, both auto-and cross-correlation times are at their maximum values, indicating reduced internal motions probably resulting from intramolecular electrostatic interactions. Modifying the charge state increases motional fluctuations. Activation energies determined from the temperature dependence of CH rotational autocorrelation times at neutral pH are approximately equal for glycine and lysine C and lysine C and C positions (4.1±0.2 to 4.5±0.2 kcal/mol) and tend to decrease slightly for lysine C and C (3.8±0.2 to 4.3±0.2 kcal/mol). The sign of lysine side-chain cross-correlations could not be explained by using any available rotational model, including one parameterized for multiple internally restricted rotations and anisotropic overall tumbling. Molecular and stochastic dynamics calculations were performed to obtain insight into correlated internal rotations and coupled overall tumbling and internal motions. Relatively strong correlations were found for i,i+1 backbone and lysine side-chain internal bond rotations. Stochastic dynamics calculations were more successful at explaining experimentally observed correlation times. In the fully charged state, a preferred conformation was detected with an all-trans lysine side chain.Abbreviations rf radio frequency - GK dipeptide glycine-lysine - KG dipeptide lysine-glycine - GKG tripeptide glycine-lysine-glycine     

19F NMR measurements of NO production in hypertensive ISIAH and OXYS rats

Recently we demonstrated the principal possibility of application of 19F NMR spin-trapping technique for in vivo *NO detection [Free Radic. Biol. Med. 36 (2004) 248]. In the present study, we employed this method to elucidate the significance of *NO availability in animal models of hypertension. In vivo *NO-induced conversion of the hydroxylamine of the fluorinated nitronyl nitroxide (HNN) to the hydroxylamine of the iminonitroxide (HIN) in hypertensive ISIAH and OXYS rat strains and normotensive Wistar rat strain was measured. Significantly lower HIN/HNN ratios were measured in the blood of the hypertensive rats. The NMR data were found to positively correlate with the levels of nitrite/nitrate evaluated by Griess method and negatively correlate with the blood pressure. In comparison with other traditionally used methods 19F NMR spectroscopy allows in vivo evaluation of *NO production and provides the basis for in vivo *NO imaging.     

Clinically Relevant Concentration Determination of Inhaled Anesthetics (Halothane,Isoflurane, Sevoflurane,and Desflurane) by <Superscript>19</Superscript>F NMR

Biophysical studies of protein–anesthetic interactions using nuclear magnetic resonance (NMR) spectroscopy are often conducted by the addition of micro amounts of neat inhaled anesthetic which yields much higher than clinically relevant (0.2–0.5 mM) anesthetic concentrations. We report a ¹⁹F NMR technique to measure clinically relevant inhaled anesthetic concentrations from saturated aqueous solutions of these anesthetics (halothane, isoflurane, sevoflurane, and desflurane). We use a setup with a 3-mm NMR tube (containing trifluoroacetic acid as standard), coaxially inserted in a 5-mm NMR tube containing anesthetic solution under investigation. All experiments are conducted in a 5-mm NMR probe. We also have provided standard curves for four inhaled anesthetics using NMR technique. The standard curve for each of these anesthetics is helpful in determining the prerequisite amount of aqueous anesthetic solution required to prepare clinically relevant concentrations for protein–anesthetic interaction studies. Parts of the results to be presented at Society for Neuroscience meeting, 2008.     

Construction of a 19F-lectin biosensor for glycoprotein imaging by using affinity-guided DMAP chemistry

In this study, assisted by affinity-guided DMAP strategy, we developed a novel ¹⁹F-modified lectin as a biosensor for specific detection and imaging of glycoproteins. Exploited the large chemical shift anisotropy property of ¹⁹F nuclei, glycoproteins detected by our ¹⁹F-biosensor are signatured by broadened peaks in ¹⁹F NMR, hence enabled the distinction between glycoproteins and small molecule saccharides. Such signal on/off switching was also applied to glycoprotein imaging by ¹⁹F MRI.     

13C and31P NMR spectroscopic studies on glucose metabolism inTribolium confusum

Nuclear magnetic resonance (NMR) technology was applied to study the glucose metabolism inTribolium confusum (Coleoptera).¹³C signals of D-(1-¹³C)glucose eaten by beetles were clearly detected in such metabolites of the glucose metabolism as glycogen, trehalose, triacylglycerol, alanine and proline by¹³C-NMR. After glucose feeding the³¹P-NMR spectra ofT. confusum showed the signal intensity increases in arginine-phosphate, sugar-phosphate and uridine diphosphoglucose. The results demonstrated the potential of NMR analysis for the study of glucose metabolism inT. confusum.     

5-Fluorotryptophan as dual probe for ground-state heterogeneity and excited-state dynamics in apoflavodoxin

The apoflavodoxin protein from Azotobacter vinelandii harboring three tryptophan (Trp) residues, was biosynthetically labeled with 5-fluorotryptophan (5-FTrp). 5-FTrp has the advantage that chemical differences in its microenvironment can be sensitively visualized via ¹⁹F NMR. Moreover, it shows simpler fluorescence decay kinetics. The occurrence of FRET was earlier observed via the fluorescence anisotropy decay of WT apoflavodoxin and the anisotropy decay parameters are in excellent agreement with distances between and relative orientations of all Trp residues. The anisotropy decay in 5-FTrp apoflavodoxin demonstrates that the distances and orientations are identical for this protein. This work demonstrates the added value of replacing Trp by 5-FTrp to study structural features of proteins via ¹⁹F NMR and fluorescence spectroscopy.     

19F NMR study on the biodegradation of fluorophenols by various Rhodococcus species

Of all NMR observable isotopes 19F is the one perhaps most convenient for studies on biodegradation of environmental pollutants. The reasons underlying this potential of 19F NMR are discussed and illustrated on the basis of a study on the biodegradation of fluorophenols by four Rhodococcus strains. The results indicate marked differences between the biodegradation pathways of fluorophenols among the various Rhodococcus species. This holds not only for the level and nature of the fluorinated biodegradation pathway intermediates that accumulate, but also for the regioselectivity of the initial hydroxylation step. Several of the Rhodococcus species contain a phenol hydroxylase that catalyses the oxidative defluorination of ortho-fluorinated di- and trifluorophenols. Furthermore, it is illustrated how the 19F NMR technique can be used as a tool in the process of identification of an accumulated unknown metabolite, in this case most likely 5-fluoromaleylacetate. Altogether, the 19F NMR technique proved valid to obtain detailed information on the microbial biodegradation pathways of fluorinated organics, but also to provide information on the specificity of enzymes generally considered unstable and, for this reason, not much studied so far.     

Internal motions of apo-neocarzinostatin as studied by 13C NMR methine relaxation at natural abundance

Summary Dynamics of the backbone and some side chains of apo-neocarzinostatin, a 10.7 kDa carrier protein, have been studied from ¹³C relaxation rates R1, R2 and steady-state ¹³C-{¹H} NOEs, measured at natural abundance. Relaxation data were obtained for 79 nonoverlapping C resonances and for 11 threonine C single resonances. Except for three C relaxation rates, all data were analysed from a simple two-parameter spectral density function using the model-free approach of Lipari and Szabo. The corresponding C–H fragments exhibit fast (_e < 40 ps) restricted libration motions (S²=0.73 to 0.95). Global examination of the microdynamical parameters S² and _e along the amino acid sequence gives no immediate correlation with structural elements. However, different trends for the three loops involved in the binding site are revealed. The -ribbon comprising residues 37 to 47 is spatially restricted, with relatively large _e values in its hairpin region. The other -ribbon (residues 72 to 87) and the large disordered loop ranging between residues 97–107 experience small-amplitude motions on a much faster (picosecond) time scale. The two N-terminal residues, Ala¹ and Ala², and the C-terminal residue Asn¹¹³, exhibit an additional slow motion on a subnanosecond time scale (400–500 ps). Similarly, the relaxation data for eight threonine side-chain C must be interpreted in terms of a three-parameter spectral density function. They exhibit slower motions, on the nanosecond time scale (500–3000 ps). Three threonine (Thr⁶⁵, Thr⁶⁸, Thr⁸¹) side chains do not display a slow component, but an exchange contribution to the observed transverse relaxation rate R2 could not be excluded at these sites. The microdynamical parameters (S², _e and R2_ex) or (S _infslow ^sup2 , S _inffast ^sup2 and _slow) were obtained from a straightforward solution of the equations describing the relaxation data. They were calculated assuming an overall isotropic rotational correlation time _e for the protein of 5.7 ns, determined using standard procedures from R2/R1 ratios. However, it is shown that the product (1–S²)× _e is nearly independent of _e for residues not exhibiting slow motions on the nanosecond time scale. In addition, this parameter very closely follows the heteronuclear NOEs, which therefore could be good indices for local fast motions on the picosecond time scale.     

A novel view of domain flexibility in E. coli adenylate kinase based on structural mode-coupling (15)N NMR relaxation.

Adenylate kinase from Escherichia coli (AKeco), consisting of a single 23.6 kDa polypeptide chain folded into domains CORE, AMPbd and LID, catalyzes the reaction AMP+ATP-->2ADP. In the ligand-free enzyme the domains AMPbd and LID execute large-amplitude movements controlling substrate binding and product release during catalysis. Domain flexibility is investigated herein with the slowly relaxing local structure (SRLS) model for (15)N relaxation. SRLS accounts rigorously for coupling between the global and local N-H motions through a local ordering potential exerted by the protein structure at the N-H bond. The latter reorients with respect to its protein surroundings, which reorient on the slower time scale associated with the global protein tumbling. AKeco diffuses globally with correlation time tau(m)=15.1 ns, while locally two different dynamic cases prevail. The domain CORE features ordering about the equilibrium N-H bond orientation with order parameters, S(2), of 0.8-0.9 and local motional correlation times, tau, mainly between 5-130 ps. This represents a conventional rigid protein structure with rapid small-amplitude N-H fluctuations. The domains AMPbd and LID feature small parallel (Z(M)) ordering of S(2)=0.2-0.5 which can be reinterpreted as high perpendicular (Y(M)) ordering. M denotes the local ordering/local diffusion frame. Local motion about Z(M) is given by tau( parallel) approximately 5 ps and local motion of the effective Z(M) axis about Y(M) by tau( perpendicular)=6-11 ns. Z(M) is tilted at approximately 20 degrees from the N-H bond. The orientation of the Y(M) axis may be considered parallel to the C(alpha)(i-1)-C(alpha)(i) axis. The tau( perpendicular) mode reflects collective nanosecond peptide-plane motions, interpretable as domain motion. A powerful new model of protein flexibility/domain motion has been established. Conformational exchange (R(ex)) processes accompany the tau( perpendicular) mode. The SRLS analysis is compared with the conventional model-free analysis.     

Histidine side-chain dynamics and protonation monitored by 13C CPMG NMR relaxation dispersion

The use of ¹³C NMR relaxation dispersion experiments to monitor micro-millisecond fluctuations in the protonation states of histidine residues in proteins is investigated. To illustrate the approach, measurements on three specifically ¹³C labeled histidine residues in plastocyanin (PCu) from Anabaena variabilis (A.v.) are presented. Significant Carr-Purcell-Meiboom-Gill (CPMG) relaxation dispersion is observed for ¹³C^ε1 nuclei in the histidine imidazole rings of A.v. PCu. The chemical shift changes obtained from the CPMG dispersion data are in good agreement with those obtained from the chemical shift titration experiments, and the CPMG derived exchange rates agree with those obtained previously from ¹⁵N backbone relaxation measurements. Compared to measurements of backbone nuclei, ¹³C^ε1 dispersion provides a more direct method to monitor interchanging protonation states or other kinds of conformational changes of histidine side chains or their environment. Advantages and shortcomings of using the ¹³C^ε1 dispersion experiments in combination with chemical shift titration experiments to obtain information on exchange dynamics of the histidine side chains are discussed.     

Quantitation of protein expression in a cell-free system: Efficient detection of yields and <Superscript>19</Superscript>F NMR to identify folded protein

We have developed an efficient and novel filter assay method, involving radioactive labelling and imaging, to quantify the expression of soluble proteins from a cell-free translation system. Here this method is combined with the conformational sensitivity of ¹⁹F NMR to monitor the folded state of the expressed protein. This report describes the optimisation of 6-fluorotryptophan incorporation in a His-tagged human serum retinol-binding protein (RBP), a disulphide bonded -barrel protein. Appropriate reagent concentrations for producing fluorine labelled RBP in a cell-free translation system are described. It is shown that ¹⁹F NMR is a suitable method for monitoring the production of correctly folded protein from a high-throughput expression system.     

Rat brain slices oxidize glucose at high rates: a (13)C NMR study

Since glucose is the main cerebral substrate, we have characterized the metabolism of various ¹³C glucose isotopomers in rat brain slices. For this, we have used our cellular metabolomic approach that combines enzymatic and carbon 13 NMR techniques with mathematical models of metabolic pathways. We identified the fate and the pathways of the conversion of glucose carbons into various products (pyruvate, lactate, alanine, aspartate, glutamate, GABA, glutamine and CO₂) and determined absolute fluxes through pathways of glucose metabolism. After 60 min of incubation, lactate and CO₂ were the main end-products of the metabolism of glucose which was avidly metabolized by the slices. Lactate was also used at high rates by the slices and mainly converted into CO₂. High values of flux through pyruvate carboxylase, which were similar with glucose and lactate as substrate, were observed. The addition of glutamine, but not of acetate, stimulated pyruvate carboxylation, the conversion of glutamate into succinate and fluxes through succinate dehydrogenase, malic enzyme, glutamine synthetase and aspartate aminotransferase. It is concluded that, unlike brain cells in culture, and consistent with high fluxes through PDH and enzymes of the tricarboxylic acid cycle, rat brain slices oxidized both glucose and lactate at high rates.     

13C NMR relaxation studies of RNA base and ribose nuclei reveal a complex pattern of motions in the RNA binding site for human U1A protein

The widespread importance of induced fit and order-disorder transition in RNA recognition by proteins and small molecules makes it imperative that RNA motional properties are characterized quantitatively. Until now, however, very few studies have been dedicated to the systematic characterization of RNA motion and to their changes upon protein or small-molecule binding. The U1A protein-RNA complexes provide some of the best-studied examples of the role of RNA motional changes upon protein binding. Here, we report (13)C NMR relaxation studies of base and ribose dynamics for the RNA internal loop target of human U1A protein located within the 3'-untranslated region (3'-UTR) of the mRNA coding for U1A itself. We also report the semi-quantitative analysis of both fast (nano- to picosecond) and intermediate (micro- to millisecond) motions for this paradigmatic RNA system. We measure (13)C T(1), T(1rho) and heteronuclear nuclear Overhauser effects (NOEs) for sugar and base nuclei, as well as the power dependence of T(1rho) at 500 MHz and 750 MHz, and analyze these results using the model-free formalism. The results provide a much clearer picture of the type of motions experienced by this RNA in the absence of the protein than was provided by the analysis of the structure based solely on NOEs and scalar couplings. They define a model where the RNA internal loop region "breathes" on a micro- to millisecond timescale with respect to the double-helical regions. Superimposed on this slower motion, the residues at the very tip of the loop undergo faster (nano- to picosecond) motions. We hypothesize that these motions allow the RNA to sample multiple conformations so that the protein can select a structure within the ensemble that optimizes intermolecular contacts.     

Resonance assignment and structural analysis of acid denatured E. coli [U-15N]-glutaredoxin 3

Virtually complete sequence specific ¹H and ¹⁵N resonance assignments are presented for acid denatured reduced E. coli glutaredoxin 3. The sequential resonance assignments of the backbone rely on the combined use of 3D F₁-decoupled ROESY-¹⁵N-HSQC and 3D ¹⁵N-HSQC-(TOCSY-NOESY)-¹⁵N-HSQC using a single uniformly ¹⁵N labelled protein sample. The sidechain resonances were assigned from a 3D TOCSY-¹⁵N-HSQC and a homonouclear TOCSY spectrum. The presented assignment strategy works in the absence of chemical exchange peaks with signals from the native conformation and without ¹³C/¹⁵N double labelling. Chemical shifts, ³J(H, NH) coupling constants and NOEs indicate extensive conformational averaging of both backbone and side chains in agreement with a random coil conformation. The only secondary structure element persisting at pH 3.5 appears to be a short helical segment comprising residues 37 to 40.Abbreviations HSQC heteronuclear single quantum coherence - NMR nuclear magnetic resonance - NOE nuclear Overhauser effect - NOESY two-dimensional NOE spectroscopy - ROE nuclear Overhauser effect in the rotating frame - ROESY two-dimensional ROE spectroscopy - TOCSY total correlation spectroscopy - TPPI time proportional phase incrementation Correspondence to: G. Otting     

F nuclear magnetic resonance screening of glucokinase activators

Human hexokinase enzyme IV (EC 2.7.1.1) catalyzes the phosphorylation of glucose and regulates the level of glucose. This enzyme exhibits strong positive cooperativity due to an allosteric transition between an inactive form and a closed active form. This form can be stabilized by activators and, thus, can increase its turnover by a kinetic memory effect characterized by a slow decay to the inactive state. The structural details of this kinetic allostery are known. Several synthetic activators have been reported. We present a preliminary nuclear magnetic resonance (NMR) screening of a chemical library in search of molecules with some affinity for glucokinase (GK). The library, composed of eight molecules with known activity as well as molecules that display no interaction, has been tested using the FAXS (fluorine chemical shift anisotropy and exchange for screening) method, based on monitoring the R₂ relaxation of the ¹⁹F spin. To ensure a valid interaction measurement, the enzyme was placed in the presence of glucose and magnesium. The binding signal of one known fluorinated ligand was measured by determining the displacement of the known ligand. This simple measure of the ¹⁹F signal intensity after an 80-ms spin echo correlates nicely with the EC₅₀, opening a route for NMR screening of GK activators.