19F NMR Chemical Shifts Research Articles

Evaluation of machine learning models for the accelerated prediction of density functional theory calculated 19F chemical shifts based on local atomic environments

The introduction of fluorine in compounds plays a crucial role in drug development as it greatly influences their final pharmacokinetic and dynamic properties. Due to the prevalence of fluorine in FDA-approved drugs in recent years, identifying the mechanisms driving their chemical transformations has become crucial in the drug discovery landscape. 19F NMR spectroscopy is a powerful analytical technique that allows for the examination of fluorine-containing compounds, offering valuable information about their structure, dynamics, and reactivity. NMR spectra can be interpreted through the leveraging of Density Functional Theory (DFT). However, the screening of compounds and discovery of feasible drug candidates is limited due to its computational cost. Here, we present a machine learning approach to accelerate the prediction of DFT-calculated 19F NMR chemical shifts. The fluorine atoms’ features in the models were derived from their local three-dimensional environments, representing their neighboring atoms within a radius of n Å away from the given fluorine atom in the compound. A comparative analysis of thirteen regression models was conducted using features extracted from 501 fluorinated compounds in our laboratory’s chemical inventory. Among the models, Gradient Boosting Regression (GBR) exhibited the highest performance, achieving a mean absolute error of 3.31 ppm with a local environment radius of 3 Å. This demonstrates a comparable accuracy to DFT calculations while reducing computational time from several hundred seconds to milliseconds. 3 Å was also found to be the most optimal radius across all models when encoding features for local atomic environments.

Exploring Molecular Heteroencoders with Latent Space Arithmetic: Atomic Descriptors and Molecular Operators.

A variational heteroencoder based on recurrent neural networks, trained with SMILES linear notations of molecular structures, was used to derive the following atomic descriptors: delta latent space vectors (DLSVs) obtained from the original SMILES of the whole molecule and the SMILES of the same molecule with the target atom replaced. Different replacements were explored, namely, changing the atomic element, replacement with a character of the model vocabulary not used in the training set, or the removal of the target atom from the SMILES. Unsupervised mapping of the DLSV descriptors with t-distributed stochastic neighbor embedding (t-SNE) revealed a remarkable clustering according to the atomic element, hybridization, atomic type, and aromaticity. Atomic DLSV descriptors were used to train machine learning (ML) models to predict 19F NMR chemical shifts. An R2 of up to 0.89 and mean absolute errors of up to 5.5 ppm were obtained for an independent test set of 1046 molecules with random forests or a gradient-boosting regressor. Intermediate representations from a Transformer model yielded comparable results. Furthermore, DLSVs were applied as molecular operators in the latent space: the DLSV of a halogenation (H→F substitution) was summed to the LSVs of 4135 new molecules with no fluorine atom and decoded into SMILES, yielding 99% of valid SMILES, with 75% of the SMILES incorporating fluorine and 56% of the structures incorporating fluorine with no other structural change.

Next-Generation Tags for Fluorine Nuclear Magnetic Resonance: Designing Amplification of Chemical Shift Sensitivity.

Fluorine NMR is a highly sensitive technique for delineating the conformational states of biomolecules and has shown great utility in drug screening and in understanding protein function. Current fluorinated protein tags leverage the intrinsic chemical shift sensitivity of the 19F nucleus to detect subtle changes in protein conformation and topology. This chemical shift sensitivity can be amplified by embedding the fluorine or trifluoromethyl reporter within a pyridone. Due to their polarizability and rapid tautomerization, pyridones exhibit a greater range of electron delocalization and correspondingly greater 19F NMR chemical shift dispersion. To assess the chemical shift sensitivity of these tautomeric probes to the local environment, 19F NMR spectra of all possible monofluorinated and trifluoromethyl-tagged versions of 2-pyridone were recorded in methanol/water mixtures ranging from 100% methanol to 100% water. 4-Fluoro-2-pyridone and 6-(trifluoromethyl)-2-pyridone (6-TFP) displayed the greatest sensitivity of the monofluorinated and trifluoromethylated pyridones, exceeding that of known conventional CF3 reporters. To evaluate the utility of tautomeric pyridone tags for 19F NMR of biomolecules, the alpha subunit of the stimulatory G protein (Gsα) and human serum albumin (HSA) were each labeled with a thiol-reactive derivative of 6-TFP and the spectra were recorded as a function of various adjuvants and drugs. The tautomeric tag outperformed the conventional tag, 2-bromo-N-(4-(trifluoromethyl)phenyl)acetamide through the improved resolution of several functional states.

Prediction of 19F NMR chemical shifts for organic compounds with ORCA

Accurate assignment of 19F NMR has long been a challenge, and quantum chemical methods are possible solutions. Herein we reported a scaling method for the prediction of 19F NMR chemical shift with freely available ORCA program package. Performance of 31 DFT functionals coupled with 11 basis sets were evaluated and influence of geometry optimization was also studied with five functionals coupled with three basis sets. The significance of geometry was further examined through the execution of relaxed surface scans of seven flexible compounds, and averaged shieldings of obtained conformers yielded notable improvement of the correlation between calculated isotropic shielidings and experimental chemical shifts. Utilization of the best scaling factor obtained successfully assigned of fluorine atoms in multifluorinated molecules with different conformations. The method reported here was computationally inexpensive, easily available with acceptable accuracy.

On the relativistic effects on 19F nuclear magnetic resonance chemical shifts in the presence of iodine atoms

In this paper, the influence of the relativistic effects on 19F NMR chemical shifts has been thoroughly investigated in diverse fluorine compounds containing one or more iodine atoms. It was found that the magnitude of the relativistic correction to fluorine shielding constant strongly depends on the arrangement of iodine atom(s) in molecule relative to the fluorine atom under consideration. Thus, specific cases were found where the relativistic corrections to fluorine shielding constants reached at as much as 6–8 ppm, causing noticeable changes in the corresponding fluorine chemical shifts. In general, the relativistic corrections to fluorine shielding constants were found to vary in the range of 0–3 % of their nonrelativistic values.

Assessment of the Commercially Available Chemical Space for Using in the 19F NMR FAXS Method: a Enamine Ltd. Case

Aim. To analyze commercially available fluorine containing compounds for the possibility of their use in the 19F NMR FAXS method.Materials and methods. The selection of fluorine-containing fragments for the study was performed using 3.9 million instock screening compounds and 248,000 in-stock building-blocks from Enamine Ltd library. The selection and classification of the compounds was carried out using the DataWarrior and KNIME software. The Fluorinated Fragments library of Enamine Ltd. containing 6377 compounds, was also analyzed. To analyze the abovementioned sets of substances, the multistep workflows specially designed were used.Results and discussion. As a result of applying the workflow developed to the compound sets (both screening compounds and building blocks), 13 800 compounds were selected and further classified according to the presence of one out of 12 fluorine-containing groups. The Fluorinated Fragments library was also subjected to a similar workflow. For the latter, 8 out of 12 fluorine-containing groups were identified. Additionally, experimental 19F NMR chemical shift values for Fluorinated Fragments library compounds spectra were analyzed. It has been found that some structural classes have areas of chemical shifts intersection. On the other hand, the ranges from –40 to –60 ppm and beyond –160 ppm are free from any group of compounds from the library analyzed.Conclusions. The analysis has shown that commercially available fluorine-containing fragments do not satisfy the needs of the 19F NMR FAXS method, and further expansion of the chemical space of fluorine-containing compounds by increasing their diversity is required.

Where Are Sodium Ions in AOT Reverse Micelles? Fluoride Anion Probes Nanoconfined Ions by 19F Nuclear Magnetic Resonance Spectroscopy.

Confining water to nanosized spaces creates a unique environment that can change water's structural and dynamic properties. When ions are present in these nanoscopic spaces, the limited number of water molecules and short screening length can dramatically affect how ions are distributed compared to the homogeneous distribution assumed in bulk aqueous solution. Here, we demonstrate that the chemical shift observed in 19F NMR spectroscopy of fluoride anion, F-, probes the location of sodium ions, Na+, confined in reverse micelles prepared from AOT (sodium dioctyl sulfosuccinate) surfactants. Our measurements show that the nanoconfined environment of reverse micelles can lead to extremely high apparent ion concentrations and ionic strength, beyond the limit in bulk aqueous solutions. Most notably, the 19F NMR chemical shift trends we observe for F- in the reverse micelles indicate that the AOT sodium counterions remain at or near the interior interface between surfactant and water, thus providing the first experimental support for this hypothesis.

Solid-State 19F NMR Chemical Shift in Square-Planar Nickel-Fluoride Complexes Linked by Halogen Bonds.

The halogen bond (XB) is a highly directional class of noncovalent interactions widely explored by experimental and computational studies. However, the NMR signature of the XB has attracted limited attention. The prediction and analysis of the solid-state NMR (SSNMR) chemical shift tensor provide useful strategies to better understand XB interactions. In this work, we employ a computational protocol for modeling and analyzing the 19F SSNMR chemical shifts previously measured in a family of square-planar trans NiII-L2-iodoaryl-fluoride (L = PEt3) complexes capable of forming self-complementary networks held by a NiF···I(C) halogen bond [Thangavadivale, V.; Chem. Sci. 2018, 9, 3767-3781]. To understand how the 19F NMR resonances of the nickel-bonded fluoride are affected by the XB, we investigate the origin of the shielding in trans-[NiF(2,3,5,6-C6F4I)(PEt3)2], trans-[NiF(2,3,4,5-C6F4I)(PEt3)2], and trans-[NiF(C6F5)(PEt3)2] in the solid state, where a XB is present in the two former systems but not in the last. We perform the 19F NMR chemical shift calculations both in periodic and molecular models. The results show that the crystal packing has little influence on the NMR signatures of the XB, and the NMR can be modeled successfully with a pair of molecules interacting via the XB. Thus, the observed difference in chemical shift between solid-state and solution NMR can be essentially attributed to the XB interaction. The very high shielding of the fluoride and its driving contributor, the most shielded component of the chemical shift tensor, are well reproduced at the 2c-ZORA level. Analysis of the factors controlling the shielding shows how the highest occupied Ni/F orbitals shield the fluoride in the directions perpendicular to the Ni-F bond and specifically perpendicular to the coordination plane. This shielding arises from the magnetic coupling of the Ni(3d)/F(2p lone pair) orbitals with the vacant σNi-F* orbital, thereby rationalizing the very highly upfield (shielded) resonance of the component (δ33) along this direction. We show that these features are characteristic of square-planar nickel-fluoride complexes. The deshielding of the fluoride in the halogen-bonded systems is attributed to an increase in the energy gap between the occupied and vacant orbitals that are mostly responsible for the paramagnetic terms, notably along the most shielded direction.

19F NMR chemical shifts are sensitive to remote functional group variations

Covalently identical ortho-F nuclei display dual 19F NMR signals in di-ortho-F substituted N-phenyl γ-lactams. We found that the deshielding 19F signal (δdown) was significantly affected by C9-substituent, especially when C9- substituents are steric bulky. Notably, the deshielded 19F chemical shift is sensitive to chirality variations at C9-alkyls, while the shielded 19F NMR signal δup manifests negligible changes among various C9-alkyl derivatives. Calculated molecular electrostatic surface potentials (ESPs) were used to interpret the experimental observations. 19F NMR signal assignments in an exemplary rotameric molecule are presented based on our understanding on structure-19F shielding relationship. Our work described here provides basics to deciphering 19F shielding in biologically important molecules and drugs, and to use of 19F chemical shifts in monitoring molecular dynamics and conformations.

Extreme NMR shielding in fluoro-nitrogen cations.

The structure and NMR shielding of a set of N-F containing cations is reported to near-quantitative accuracy from extensive ab initio calculations. Currently, the shortest experimentally confirmed N-F bond is 1.2461(10) Å in NNF+, however CCSD(T)-F12b/cc-pVQZ-F12 optimised geometries suggest that even shorter N-F bonds are possible for both monocations (1.236 Å, HNF+) and dications (1.098 Å, NF2+). NMR shielding constants have been calculated in a composite manner with individual components from coupled-cluster expansions up to CCSDTQP and basis sets up to aug-cc-pCV8Z, together with vibrational and relativistic corrections. 15N and 19F NMR chemical shifts correlate well with available experimental data. Extreme 19F chemical shifts are predicted for HNF+ (1628.9 ppm) and NH2F2+ (1298.0 ppm), which are by far the largest 19F chemical shifts ever reported and well outside the known range of +865 ppm (F2O2) to -448 ppm (ClF). The 15N chemical shift of -1283.07 ppm in HNF+ is similarly extreme, being well outside the known range of 15N chemical shifts of -730 to 260 ppm (CH3NO2 reference). This work highlights the application of state-of-the-art theoretical techniques, and provides accurate NMR properties of both isolated and yet unknown N-F cations, which can serve to guide and supplement NMR experimentation.

Structural Parameters of the Interaction between Ciprofloxacin and Human Topoisomerase-II β Enzyme: Toward New 19F NMR Chemical Shift Probes

New tools for cancer diagnosis are being studied since early diagnosis can be crucial for a successful treatment. In this context, the use of NMR probes constitutes an efficient method of diagnosis. In this study, we investigated the use of ciprofloxacin to indirectly label the overexpression of topoisomerase-II enzymes by changes in 19F NMR chemical shifts of ciprofloxacin. Increased topoisomerase-II expression has been associated with cancer occurrence, mainly with aggressive forms of breast cancer, thus constituting a promising molecular target for new tumor cell identifiers. Using DFT calculations, we performed a spectroscopy analysis of ciprofloxacin in different chemical environments and evaluated the solvent and enzymatic effects. Our results show that ciprofloxacin forms a stable complex with the enzyme, and the main intermolecular interactions between ciprofloxacin and human topoisomerase-IIβ are hydrogen bonds, followed by π-π stacking and electrostatic interactions. Additionally, a shift of 6.04 ppm occurs in the 19F NMR signal when ciprofloxacin interacts with the human topoisomerase-IIβ enzyme, and this parameter may be an indirect marker indicating the overexpression of these enzymes in the body.

A Chiral 19F NMR Reporter of Foldamer Conformation in Bilayers.

Understanding and controlling peptide foldamer conformation in phospholipid bilayers is a key step toward their use as molecular information relays in membranes. To this end, a new 19F "reporter" tag has been developed and attached to dynamic peptide foldamers. The (R)-1-(trifluoromethyl)ethylamido ((R)-TFEA) reporter was attached to the C-terminus of α-amino-iso-butyric acid (Aib) foldamers. Crystallography confirmed that the foldamers adopted 310 helical conformations. Variable temperature (VT) NMR spectroscopy in organic solvents showed that the (R)-TFEA reporter had an intrinsic preference for P helicity, but the overall screw-sense was dominated by a chiral "controller" at the N-terminus. The 19F NMR chemical shift of the CF3 resonance was correlated with the ability of different N-terminal groups to induce either an M or a P helix in solution. In bilayers, a similar correlation was found. Solution 19F NMR spectroscopy on small unilamellar vesicle (SUV) suspensions containing the same family of (R)-TFEA-labeled foldamers showed broadened but resolvable 19F resonances, with each chemical shift mirroring their relative positions in organic solvents. These studies showed that foldamer conformational preferences are the same in phospholipid bilayers as in organic solvents and also revealed that phospholipid chirality has little influence on conformation.

1H and 19F NMR chemical shifts for hydrogen bond strength determination: Correlations between experimental and computed values

1H and 19F NMR methods based on chemical shift measurements of different hydrogen bond donors used for quantifying hydrogen bond strength were analyzed and compared. The extracted values from these different methods are shown to be highly correlated with each other and with several experimental and ab initio computed quantities characterizing hydrogen bond formation. The titration method based on 19F NMR spectroscopy was performed for detecting and quantifying the formation of very weak hydrogen bond complexes such as those involving fluorine atoms as hydrogen bond acceptors. This approach represents a powerful and reliable method for studying and characterizing these complexes that, although weak, are very relevant.

Finding Fluorine: Photoproduct Formation during the Photolysis of Fluorinated Pesticides.

The photolysis of pesticides with different fluorine motifs was evaluated to quantify the formation of fluorinated products in buffered aqueous systems, advanced oxidation (AOP) and reduction processes (ARP), and river water. Simulated sunlight quantum yields at pH 7 were 0.0033, 0.0025, 0.0015, and 0.00012 for penoxsulam, florasulam, sulfoxaflor, and fluroxypyr, respectively. The bimolecular rate constants with hydroxyl radicals were 2 to 5.7 × 1010 M-1 s-1 and, with sulfate radicals, 1.6 to 2.6 × 108 M-1 s-1 for penoxsulam, florasulam, and fluroxypyr, respectively. The rate constants of sulfoxaflor were 100-fold lower. Using quantitative 19F-NMR, complete fluorine mass balances were obtained. The maximum fluoride formation was 53.4 and 87.4% for penoxsulam and florasulam under ARP conditions, and 6.1 and 100% for sulfoxaflor and fluroxypyr under AOP conditions. Heteroaromatic CF3 and aliphatic CF2 groups were retained in multiple fluorinated photoproducts. Aryl F and heteroaromatic F groups were readily defluorinated to fluoride. CF3 and CF2 groups formed trifluoroacetate and difluoroacetate, and yields increased under oxidizing conditions. 19F-NMR chemical shifts and coupling analysis provided information on hydrogen loss on adjacent bonds or changes in chirality. Mass spectrometry results were consistent with the observed 19F-NMR products. These results will assist in selecting treatment processes for specific fluorine motifs and in the design of agrochemicals to reduce byproduct formation.

19F-NMR Unveils the Ligand-Induced Conformation of a Catalytically Inactive Twisted Homodimer of tRNA-Guanine Transglycosylase.

Understanding the structural arrangements of protein oligomers can support the design of ligands that interfere with their function in order to develop new therapeutic concepts for disease treatment. Recent crystallographic studies have elucidated a novel twisted and functionally inactive form of the homodimeric enzyme tRNA-guanine transglycosylase (TGT), a putative target in the fight against shigellosis. Active-site ligands have been identified that stimulate the rearrangement of one monomeric subunit by 130° against the other one to form an inactive twisted homodimer state. To assess whether the crystallographic observations also reflect the conformation in solution and rule out effects from crystal packing, we performed 19F-NMR spectroscopy with the introduction of 5-fluorotryptophans at four sites in TGT. The inhibitor-induced conformation of TGT in solution was assessed based on 19F-NMR chemical shift perturbations. We investigated the effect of C(4) substituted lin-benzoguanine ligands and identified a correlation between dynamic protein rearrangements and ligand-binding features in the corresponding crystal structures. These involve the destabilization of a helix next to the active site and the integrity of a flexible loop-helix motif. Ligands that either completely lack an attached C(4) substituent or use it to stabilize the geometry of the functionally competent dimer state do not indicate the presence of the twisted dimer form in the NMR spectra. The perturbation of crucial structural motifs in the inhibitors correlates with an increasing formation of the inactive twisted dimer state, suggesting these ligands are able to shift a conformational equilibrium from active C2-symmetric to inactive twisted dimer conformations. These findings suggest a novel concept for the design of drug candidates for further development.

Pore-Directed Solid-State Selective Photoinduced E – Z Isomerization and Dimerization within Metal–Organic Frameworks

Pore-Directed Solid-State Selective Photoinduced <i>E</i> – <i>Z</i> Isomerization and Dimerization within Metal–Organic Frameworks

Electrostatic effects on 19F NMR chemical shifts in N-phenyl γ-lactam derivatives

Covalently identical ortho-F nuclei display dual 19F NMR signals in di-ortho-F substituted N-phenyl γ-lactams. Structural variations at C9-substituent on the γ-lactam lead to larger chemical shift changes at downfield 19F signals (δdown) than those at upfield 19F signals (δup). A “reverse” correlation between downfield 19F chemical shifts and electron-withdrawing abilities of C9-phenylamides is observed. Electrostatic effects are the major forces between C9-substituents and ortho-F nuclei, and Hammett parameters can be used to describe such through-space interactions. Structural factors influencing ortho-19F shielding among diverse C9-substituted N-phenyl γ-lactams are discussed.

A computational tool to accurately and quickly predict 19F NMR chemical shifts of molecules with fluorine-carbon and fluorine-boron bonds.

We report the evaluation of density-functional-theory (DFT) based procedures for predicting 19F NMR chemical shifts at modest computational cost for a range of molecules with fluorine bonds, to be used as a tool for assisting the characterisation of reaction intermediates and products and as an aid to identifying mechanistic pathways. The results for a balanced learning set of molecules were then checked using two further testing sets, resulting in the recommendation of the ωB97XD/aug-cc-pvdz DFT method and basis set as having the best combination of accuracy and computational time, with a RMS error of 3.57 ppm. Cationic molecules calculated without counter-anion showed normal errors, whilst anionic molecules showed somewhat larger errors. The method was applied to the prediction of the conformationally averaged 19F chemical shifts of 2,2,3,3,4,4,5,5-octafluoropentan-1-ol, in which gauche stereoelectronic effects involving fluorine dominate and to determining the position of coordination equilibria of fluorinated boranes as an aid to verifying the relative energies of intermediate species involved in catalytic amidation reactions involving boron catalysts.

The relative stability of SCM-14 germanosilicate with different distributions of germanium ions in the absence and presence of structure-directing agents

According to periodic DFT calculations suggest that: Ge atoms in SCM-14 germanosilicate are clustered forming D4Rs populated entirely by Ge; structures with more Ge–O–Ge bridges are more stable; simulated19F NMR chemical shifts do not correlated with Ge content of the D4R.

Synthesis, Reactivity, and Bonding of Gold(I) Fluorido-Phosphine Complexes.

Gold(I) fluorido complexes with phosphine ligands have been synthesized from their respective iodido precursors. The bonding situation in comparison between complexes bearing phosphines and N-heterocyclic carbenes (NHCs) was explored quantum-chemically, obtaining similar results for both. Calculations of the 19F NMR chemical shifts match the experimental values well, including the approximately 40 ppm low-field shifts for the phosphine complexes compared to the NHC complexes, in spite of similar negative charges on fluorine. The reactivity of the highly water-sensitive gold(I) fluorido complexes was studied, resulting in substitution at the metal using trimethylsilyl reagents. The compounds studied were characterized using NMR as well as X-ray diffraction methods.