Structural characterization of an anhydrous polymorph of paclitaxel by solid-state NMR

Structural Refinement of Carbimazole by NMR Crystallography

The characterization of the three-dimensional structure of solids is of major importance, especially in the pharmaceutical field. In the present work, NMR crystallography methods are applied with the aim to refine the crystal structure of carbimazole, an active pharmaceutical ingredient used for the treatment of hyperthyroidism and Grave’s disease. Starting from previously reported X-ray diffraction data, two refined structures were obtained by geometry optimization methods. Experimental ¹H and ¹³C isotropic chemical shift measured by the suitable ¹H and ¹³C high-resolution solid state NMR techniques were compared with DFT-GIPAW calculated values, allowing the quality of the obtained structure to be experimentally checked. The refined structure was further validated through the analysis of ¹H-¹H and ¹H-¹³C 2D NMR correlation experiments. The final structure differs from that previously obtained from X-ray diffraction data mostly for the position of hydrogen atoms.

Modeling Small Structural and Environmental Differences in Solids with 15 N NMR Chemical Shift Tensors

The ability to theoretically predict accurate NMR chemical shifts in solids is increasingly important due to the role such shifts play in selecting among proposed model structures. Herein, two theoretical methods are evaluated for their ability to assign ¹⁵ N shifts from guanosine dihydrate to one of the two independent molecules present in the lattice. The NMR data consist of ¹⁵ N shift tensors from 10 resonances. Analysis using periodic boundary or fragment methods consider a benchmark dataset to estimate errors and predict uncertainties of 5.6 and 6.2 ppm, respectively. Despite this high accuracy, only one of the five sites were confidently assigned to a specific molecule of the asymmetric unit. This limitation is not due to negligible differences in experimental data, as most sites exhibit differences of >6.0 ppm between pairs of resonances representing a given position. Instead, the theoretical methods are insufficiently accurate to make assignments at most positions.

NMR crystallography of molecular organics

Developments of NMR methodology to characterise the structures of molecular organic structures are reviewed, concentrating on the previous decade of research in which density functional theory-based calculations of NMR parameters in periodic solids have become widespread. With a focus on demonstrating the new structural insights provided, it is shown how "NMR crystallography" has been used in a spectrum of applications from resolving ambiguities in diffraction-derived structures (such as hydrogen atom positioning) to deriving complete structures in the absence of diffraction data. As well as comprehensively reviewing applications, the different aspects of the experimental and computational techniques used in NMR crystallography are surveyed. NMR crystallography is seen to be a rapidly maturing subject area that is increasingly appreciated by the wider crystallographic community.

14N–1H HMQC solid-state NMR as a powerful tool to study amorphous formulations – an exemplary study of paclitaxel loaded polymer micelles

¹⁴N–¹H HMQC experiments are powerful experiments to characterize amorphous drug–polymer formulations of paclitaxel yielding well-separated signals in the ¹⁴N dimension as well as information on the symmetry of ¹⁴N and ¹⁴N–¹H interactions.

A Bayesian approach to NMR crystal structure determination

Nuclear Magnetic Resonance (NMR) spectroscopy is particularly well suited to determine the structure of molecules and materials in powdered form. Structure determination usually proceeds by finding the best match between experimentally observed NMR chemical shifts and those of candidate structures. Chemical shifts for the candidate configurations have traditionally been computed by electronic-structure methods, and more recently predicted by machine learning. However, the reliability of the determination depends on the errors in the predicted shifts. Here we propose a Bayesian framework for determining the confidence in the identification of the experimental crystal structure, based on knowledge of the typical errors in the electronic structure methods. We demonstrate the approach on the determination of the structures of six organic molecular crystals. We critically assess the reliability of the structure determinations, facilitated by the introduction of a visualization of the similarity between candidate configurations in terms of their chemical shifts and their structures. We also show that the commonly used values for the errors in calculated ¹³C shifts are underestimated, and that more accurate, self-consistently determined uncertainties make it possible to use ¹³C shifts to improve the accuracy of structure determinations. Finally, we extend the recently-developed ShiftML model to render it more efficient, accurate, and, most importantly, to evaluate the uncertainties in its predictions. By quantifying the confidence in structure determinations based on ShiftML predictions we further substantiate that it provides a valid replacement for first-principles calculations in NMR crystallography.

Rapid Structure Determination of Molecular Solids Using Chemical Shifts Directed by Unambiguous Prior Constraints

NMR-based crystallography approaches involving the combination of crystal structure prediction methods, ab initio calculated chemical shifts and solid-state NMR experiments are powerful methods for crystal structure determination of microcrystalline powders. However, currently structural information obtained from solid-state NMR is usually included only after a set of candidate crystal structures has already been independently generated, starting from a set of single-molecule conformations. Here, we show with the case of ampicillin that this can lead to failure of structure determination. We propose a crystal structure determination method that includes experimental constraints during conformer selection. In order to overcome the problem that experimental measurements on the crystalline samples are not obviously translatable to restrict the single-molecule conformational space, we propose constraints based on the analysis of absent cross-peaks in solid-state NMR correlation experiments. We show that these absences provide unambiguous structural constraints on both the crystal structure and the gas-phase conformations, and therefore can be used for unambiguous selection. The approach is parametrized on the crystal structure determination of flutamide, flufenamic acid, and cocaine, where we reduce the computational cost by around 50%. Most importantly, the method is then shown to correctly determine the crystal structure of ampicillin, which would have failed using current methods because it adopts a high-energy conformer in its crystal structure. The average positional RMSE on the NMR powder structure is ⟨r_av⟩ = 0.176 Å, which corresponds to an average equivalent displacement parameter U_eq = 0.0103 Ų.

Chemical shifts in molecular solids by machine learning

Due to their strong dependence on local atonic environments, NMR chemical shifts are among the most powerful tools for strucutre elucidation of powdered solids or amorphous materials. Unfortunately, using them for structure determination depends on the ability to calculate them, which  comes at the cost of high accuracy first-principles calculations. Machine learning has recently emerged as a way to overcome the need for quantum chemical calculations, but for chemical shifts in solids it is hindered by the chemical and combinatorial space spanned by molecular solids, the strong dependency of chemical shifts on their environment, and the lack of an experimental database of shifts. We propose a machine learning method based on local environments to accurately predict chemical shifts of molecular solids and their polymorphs to within DFT accuracy. We also demonstrate that the trained model is able to determine, based on the match between experimentally measured and ML-predicted shifts, the structures of cocaine and the drug 4-[4-(2-adamantylcarbamoyl)-5-tert-butylpyrazol-1-yl]benzoic acid.

Solid-state nuclear magnetic resonance combined with quantum chemical shift predictions is limited by high computational cost. Here, the authors use machine learning based on local atomic environments to predict experimental chemical shifts in molecular solids with accuracy similar to density functional theory.

Insights on conformation in the solid state: a case study – s-cis and/or s-trans crystallization of 5(3)-aryl-3(5)-carboxyethyl-1-tert-butylpyrazoles

The conformation adopted by COOEt group in solid state were influenced by supramolecular environment and intramolecular interaction for 1,3- and 1,5-regioisomers, respectively.

Polymorphs of Theophylline Characterized by DNP Enhanced Solid-State NMR

We show how dynamic nuclear polarization (DNP) enhanced solid-state NMR spectroscopy can be used to characterize polymorphs and solvates of organic solids. We applied DNP to three polymorphs and one hydrated form of the asthma drug molecule theophylline. For some forms of theophylline, sample grinding and impregnation with the radical-containing solution, which are necessary to prepare the samples for DNP, were found to induce polymorphic transitions or desolvation between some forms. We present protocols for sample preparation for solid-state magic-angle spinning (MAS) DNP experiments that avoid the polymorphic phase transitions in theophylline. These protocols include cryogrinding, grinding under inert atmosphere, and the appropriate choice of the impregnating liquid. By applying these procedures, we subsequently demonstrate that two-dimensional correlation experiments, such as (1)H-(13)C and (1)H-(15)N HETCOR or (13)C-(13)C INADEQUATE, can be obtained at natural isotopic abundance in reasonable times, thus enabling more advanced structural characterization of polymorphs.

Towards understanding P-gp resistance: a case study of the antitumour drug cabazitaxel

A structural and Hirshfeld surface study of five forms of the antitumour drug cabazitaxel suggests key factors contributing to its poor affinity to P-gp.

De novo determination of the crystal structure of a large drug molecule by crystal structure prediction-based powder NMR crystallography

The crystal structure of form 4 of the drug 4-[4-(2-adamantylcarbamoyl)-5-tert-butyl-pyrazol-1-yl]benzoic acid is determined using a protocol for NMR powder crystallography at natural isotopic abundance combining solid-state (1)H NMR spectroscopy, crystal structure prediction, and density functional theory chemical shift calculations. This is the first example of NMR crystal structure determination for a molecular compound of previously unknown structure, and at 422 g/mol this is the largest compound to which this method has been applied so far.

Powder crystallography of pharmaceutical materials by combined crystal structure prediction and solid-state 1H NMR spectroscopy

A protocol for the ab initio crystal structure determination of powdered solids at natural isotopic abundance by combining solid-state NMR spectroscopy, crystal structure prediction, and DFT chemical shift calculations was evaluated to determine the crystal structures of four small drug molecules: cocaine, flutamide, flufenamic acid, and theophylline. For cocaine, flutamide and flufenamic acid, we find that the assigned (1)H isotropic chemical shifts provide sufficient discrimination to determine the correct structures from a set of predicted structures using the root-mean-square deviation (rmsd) between experimentally determined and calculated chemical shifts. In most cases unassigned shifts could not be used to determine the structures. This method requires no prior knowledge of the crystal structure, and was used to determine the correct crystal structure to within an atomic rmsd of less than 0.12 Å with respect to the known reference structure. For theophylline, the NMR spectra are too simple to allow for unambiguous structure selection.

Small molecule, big difference: the role of water in the crystallization of paclitaxel

Paclitaxel is an important antineoplastic drug, which is used widely in the treatment of many forms of cancer. The crystal structures of the anhydrous form and the hemihydrate were determined from laboratory X-ray powder diffraction data, whereas the dihydrate was solved from single-crystal synchrotron diffraction data. Intermolecular spaces allow for the inclusion of loosely bound water molecules, which are then lost easily upon heating. All three forms were found to crystallize in the orthorhombic spacegroup P2(1)2(1)2(1), with Z' = 2. The unit cell parameters were found to be a = 9.6530(3) Å, b = 28.1196(8) Å, c = 33.5378(14) Å, and V = 9103.5(5) Å for the anhydrous form (363 K); a = 9.6890(2) Å, b = 28.0760(4) Å, c = 33.6166(8) Å, and V = 9144.7(3) Å(3) for the hemihydrate (333 K); and a = 9.512(6) Å, b = 28.064(16) Å, c = 33.08(2) Å, and V = 8829.0(9) Å(3) for the dihydrate (120 K). Water loss occurs in two steps between 120 K ≤ t ≤ 363 K. The thermal stability of the hydrates and accompanying unit cell changes were observed in situ via temperature-resolved X-ray powder diffraction and thermogravimetric analysis.

Measurement of the principal values of the chemical-shift tensors of overlapping protonated and unprotonated carbons with the 2D-SUPER technique and dipolar dephasing (DD-SUPER)

A modified 2D-SUPER technique is demonstrated to allow independent measurement of the principal values of the chemical-shift tensors of overlapping protonated and unprotonated carbons. The insertion of a dipolar-dephasing period into the sequence causes loss of signal from protonated carbons. The spectrum obtained with this modification allows one to determine the principal values of the unprotonated carbons with high precision. Subsequent fitting of the usual 2D-SUPER spectrum, with the chemical-shift parameters of the unprotonated carbons fixed, gives the parameters of the overlapped resonances of the protonated carbons. As an example, we report the determination of the (13)C chemical-shift parameters of the carbons of form II of piroxicam. The experimental results are compared with those obtained from calculations using the DFT/GIAO method. Potential applications of this method are discussed.

Phenylisoserine in the gas-phase and water: Ab initio studies on neutral and zwitterion conformers

The conformational landscape of phenylisoserine (PhIS) was studied. Trial structures were generated by allowing for all combinations of single-bond rotamers. Based on the B3LYP/aug-cc-pVDZ calculations 54 conformers were found to be stable in the gas phase. The six most stable conformers were further optimized at the B3LYP/aug-cc-pVTZ and MP2/aug-cc-pVDZ levels for which characteristic intramolecular hydrogen bond types were classified. To estimate the influence of water on PhIS conformation, the IEF-PCM/B3LYP/aug-cc-pVDZ calculations were carried out and showed 51 neutral and six zwitterionic conformers to be stable in water solution. According to DFT calculations, the conformer equilibrium in the gas phase is dominated by one conformer, whereas the MP2 calculations suggest three PhIS structures to be significantly populated. Comparison of DFT and MP2 energies of all 57 structures stable in water indicates that, in practice, one zwitterionic and one neutral conformer determine the equilibrium in water. Based on the AIM calculations, we found that for the neutral conformers in vacuum and in water, d(H...B) is linearly correlated with Laplacian at the H-bond critical point.

Figure

Powder crystallography by combined crystal structure prediction and high-resolution 1H solid-state NMR spectroscopy

A fast method for crystal structure determination using crystal structure prediction and solid-state (1)H NMR is presented. This technique does not need any prior knowledge except the chemical formula; resonance assignment is not necessary. Starting from an ensemble of predicted crystal structures for powdered thymol, comparison between experimental and calculated (1)H solid-state isotropic NMR chemical shifts is sufficient to determine which predicted structure corresponds to the powder under study. The same approach using proton-proton spin-diffusion data is successful and can be used for cross-validation.

Complete (1)H resonance assignment of beta-maltose from (1)H-(1)H DQ-SQ CRAMPS and (1)H (DQ-DUMBO)-(13)C SQ refocused INEPT 2D solid-state NMR spectra and first principles GIPAW calculations

A disaccharide is a challenging case for high-resolution (1)H solid-state NMR because of the 24 distinct protons (14 aliphatic and 10 OH) having (1)H chemical shifts that all fall within a narrow range of approximately 3 to 7 ppm. High-resolution (1)H (500 MHz) double-quantum (DQ) combined rotation and multiple pulse sequence (CRAMPS) solid-state NMR spectra of beta-maltose monohydrate are presented. (1)H-(1)H DQ-SQ CRAMPS spectra are presented together with (1)H (DQ)-(13)C correlation spectra obtained with a new pulse sequence that correlates a high-resolution (1)H DQ dimension with a (13)C single quantum (SQ) dimension using the refocused INEPT pulse-sequence element to transfer magnetization via one-bond (13)C-(1)H J couplings. Compared to the observation of only a single broad peak in a (1)H DQ spectrum recorded at 30 kHz magic-angle spinning (MAS), the use of DUMBO (1)H homonuclear decoupling in the (1)H DQ CRAMPS experiment allows the resolution of distinct DQ correlation peaks which, in combination with first-principles chemical shift calculations based on the GIPAW (Gauge Including Projector Augmented Waves) plane-wave pseudopotential approach, enables the assignment of the (1)H resonances to the 24 distinct protons. We believe this to be the first experimental solid-state NMR determination of the hydroxyl OH (1)H chemical shifts for a simple sugar. Variable-temperature (1)H-(1)H DQ CRAMPS spectra reveal small increases in the (1)H chemical shifts of the OH resonances upon decreasing the temperature from 348 K to 248 K.

Powder NMR crystallography of thymol

A protocol for the structure determination of powdered solids at natural abundance by NMR is presented and illustrated for the case of the small drug molecule thymol. The procedure uses proton spin-diffusion data from two-dimensional NMR experiments in combination with periodic DFT refinements incorporating (1)H and (13)C NMR chemical shifts. For thymol, the method yields a crystal structure for the powdered sample, which differs by an atomic root-mean-square-deviation (all atoms except methyl group protons) of only 0.07 A from the single crystal X-ray diffraction structure with DFT-optimized proton positions.

Solid-state phosphorus-31 NMR spectroscopy of a multiple-spin system: an investigation of a rhodium-triphosphine complex

Phosphorus-31 NMR spectra of solid [tris(dimethylphenylphosphine)](2,5-norbornadiene) rhodium(I) hexafluorophosphate have been acquired at several applied magnetic field strengths. The phosphorus nuclei of the three phosphine ligands are spin-spin coupled to each other and to 103Rh, resulting in complex NMR spectra; however, the three phosphorus chemical shift (CS) tensors were determined through the analysis of NMR spectra of slow magic angle spinning and stationary samples. Spectra of spinning samples in rotational resonance and two-dimensional 31P NMR spectra were particularly useful for determining the magnitudes of the indirect spin-spin couplings, and to probe their signs. Despite being in similar environments, the three phosphorus nuclei of the phosphine ligands have distinct CS tensors. In particular, the spans of these tensors, delta11-delta33, range from 80 to 176 ppm. The phosphorus CS tensors have been assigned to specific sites determined by X-ray crystallography, based on a combination of the experimental results and the results of quantum chemical calculations of the phosphorus shielding and 2J(31P,31P) values. The effect of coordination of dimethylphenylphosphine with rhodium has been investigated by comparing calculated phosphorus CS tensors for the uncoordinated ligand with those obtained for the ligands in the complex.