I. INTRODUCTION
Eltrombopag is a medication that has been developed to treat conditions that lead to thrombocytopenia (abnormally low platelet counts). It is a small molecule agonist of the c-mpl (TpoR) receptor, which is the physiological target of the hormone thrombopoietin (PubChem; Kim et al., Reference Kim, Chen, Cheng, Gindulyte, He, He, Li, Shoemaker, Thiessen, Yu, Zaslavsky, Zhang and Bolton2021). Eltrombopag was discovered as a result of research collaboration between GlaxoSmithKline and Ligand Pharmaceuticals. Designated an orphan drug in the USA and EU, it is being manufactured and marketed by Novartis under the trade name Promacta® in the USA and is marketed as Revolade® in the EU. Eltrombopag olamine is the orally active ethanolammonium salt of eltrombopag, a small-molecule, nonpeptide thrombopoietin receptor agonist with megakaryopoiesis-stimulating activity. Eltrombopag binds to and stimulates the transmembrane domain of the platelet thrombopoietin receptor (TPO-R or CD110), a member of the hematopoietin receptor superfamily. The activation of TPO-R leads to the proliferation and differentiation of cells in the megakaryocytic lineage and an increase in platelet production. The systematic name (CAS Registry Number 496775-62-3) is bis(ethanolammonium) 3-[3-[[2-(3,4-dimethylphenyl)-5-methyl-3-oxo-1H-pyrazol-4-yl]diazenyl]-2-hydroxyphenyl]benzoate. The molecular structure of bis(ethanolammonium) eltrombopag is illustrated in Figure 1.
The molecular structure of bis(ethanolammonium) eltrombopag.
The bis(ethanolamine) salt of eltrombopag was disclosed in US Patent Application 2006/0178518 A1 (Moore, Reference Moore2006; SmithKline Beecham; equivalent to WO 2003/098992), but no powder diffraction data were provided. X-ray powder patterns of Forms I, II, and III of the bis(ethanolamine) salt of eltrombopag are reported in US Patent Application 2011/0196008A1 (Leksic et al., Reference Leksic, Ceric, Mundorfer, Antonac and Samardzic2011; Pliva Hrvatska).
This work was carried out as part of a project (Kaduk et al., Reference Kaduk, Crowder, Zhong, Fawcett and Suchomel2014) to determine the crystal structures of large-volume commercial pharmaceuticals and include high-quality powder diffraction data for these pharmaceuticals in the Powder Diffraction File.
II. EXPERIMENTAL
The sample was a commercial reagent, purchased from Sigma (Lot #B29271), and was used as received. The white powder was packed into a 1.5-mm diameter Kapton capillary and rotated during the measurement at ~50 Hz. The powder pattern was measured at 295 K at beamline 11-BM (Lee et al., Reference Lee, Shu, Ramanathan, Preissner, Wang, Beno, Von Dreele, Ribaud, Kurtz, Antao, Jiao and Toby2008; Wang et al., Reference Wang, Toby, Lee, Ribaud, Antao, Kurtz, Ramanathan, Von Dreele and Beno2008) of the Advanced Photon Source at Argonne National Laboratory using a wavelength of 0.457899 Å from 0.5 to 50° 2θ with a step size of 0.001° and a counting time of 0.1 s step−1.
The pattern was indexed on a primitive monoclinic unit cell with a = 17.66301, b = 7.55911, c = 22.05342 Å, β = 105.944°, V = 2831.2 Å3, and Z = 4 using N-TREOR (Altomare et al., Reference Altomare, Cuocci, Giacovazzo, Moliterni, Rizzi, Corriero and Falcicchio2013). Analysis of the systematic absences using EXPO2014 (Altomare et al., Reference Altomare, Cuocci, Giacovazzo, Moliterni, Rizzi, Corriero and Falcicchio2013) suggested the space group P21/n, which was confirmed by successful solution and the refinement of the structure. A reduced cell search in the Cambridge Structural Database (Groom et al., Reference Groom, Bruno, Lightfoot and Ward2016) yielded four hits but no structures for eltrombopag derivatives. A connectivity search yielded no hits.
An eltrombopag molecule was downloaded from PubChem as ConformerD_CID_135448332.sdf. It is worth noting that the molecular structure in this PubChem entry is incorrect and had to be corrected based on the molecular connectivity in the patents. The protons were removed from the carboxylic acid and hydroxyl groups of the neutral molecule, and it was converted into a .mol2 file using Materials Studio (Dassault Systèmes, 2020) and a Fenske-Hall Z-matrix using Open Babel (O'Boyle et al., Reference O'Boyle, Banck, James, Morley, Vandermeersch and Hutchison2011). An ethanolammonium cation was built using Spartan'18 (Wavefunction, 2020), saved as a .mol2 file, and converted into a Fenske-Hall Z-matrix using the same tools. The structure was solved by Monte Carlo simulated annealing (parallel tempering) techniques as implemented in FOX (Favre-Nicolin and Černý, Reference Favre-Nicolin and Černý2002). One anion and two cations were used as fragments, along with a bump penalty and 010 preferred orientation. Of the 34 runs, one had a cost factor (Figure 2; FOX's measure of the quality of the solution; lower is better) much lower than the others. The similarity of the two ends of the cations (N vs. O) suggested the possibility of false minima in the structure solution, as the ends might be difficult to distinguish. Some of the hydrogen positions in the best solution were recalculated using Materials Studio, and the N and O in one cation were swapped to make a better hydrogen-bonding pattern. (We would expect the positive N end to be nearer the negatively charged portions of the anion.)
Distribution of FOX cost factors among the multiple structure solutions.
Rietveld refinement was carried out using GSAS-II (Toby and Von Dreele, Reference Toby and Von Dreele2013). Only the 1.0–26.0° portion of the pattern was included in the refinement (d min = 1.018 Å). All non-H-bond distances and angles were subjected to restraints based on a Mercury/Mogul Geometry Check (Bruno et al., Reference Bruno, Cole, Kessler, Luo, Motherwell, Purkis, Smith, Taylor, Cooper, Harris and Orpen2004; Sykes et al., Reference Sykes, McCabe, Allen, Battle, Bruno and Wood2011) of the molecule. The Mogul average and standard deviation for each quantity were used as the restraint parameters. The restraints contributed 5.5% to the final χ 2. The hydrogen atoms were included in calculated positions, which were recalculated during the refinement using Materials Studio (Dassault Systèmes, 2020). The U iso were grouped by chemical similarity. The U iso of the hydrogen atoms were constrained to be 1.3× that of the heavy atoms to which they are attached. The background was modeled using a four-term shifted Chebyshev polynomial, along with one peak at 6.55° to model the scattering from the Kapton capillary and any amorphous component of the sample. The peak profiles were described using the generalized microstrain model, and a second-order spherical harmonic preferred orientation model was included.
After initial refinement, a DFT geometry optimization was carried out, and the refinement was restarted from this model. Although the R wp decreased to ~12%, this model proved to be unsatisfactory. When the displacement coefficients were refined, the common U iso for the dimethylphenyl ring C1–C8 refined to a value much larger (>0.10 Å2) than that of the other portions of the molecule. Refining the occupancies of the carbon atoms of this ring led to values of 0 for C8, 1.6 for C1, and ~1.0 for the remaining atoms. This pattern suggested that the substituted ring needed to be rotated by 180°. The N10–N9–C6–C2 torsion angle was manually adjusted using Materials Studio, and the refinement continued. The R wp decreased significantly, and the U iso refined to a value consistent with the rest of the molecule. A new DFT calculation was carried out, which resulted in a structure lower in energy by 2.5 kcal mol−1 cell−1.
The final refinement of 156 variables using 25 040 observations and 98 restraints yielded the residuals of R wp = 0.0866 and GOF = 1.60. The largest peak (1.51 Å from N40) and hole (1.67 Å from C27) in the difference Fourier map were 0.31 and −0.28(7) eÅ−3, respectively. The largest errors in the fit (Figure 3) are in the shapes of some of the low-angle peaks.
The Rietveld plot for the refinement of bis(ethanolammonium) eltrombopag. The blue crosses represent the observed data points, and the green line is the calculated pattern. The cyan curve is the normalized error plot. The vertical scale is multiplied by a factor of 10× for 2θ > 10.0°.
The density functional geometry optimizations were carried out using CRYSTAL14 (Dovesi et al., Reference Dovesi, Orlando, Erba, Zicovich-Wilson, Civalleri, Casassa, Maschio, Ferrabone, De La Pierre, D-Arco, Noël, Causà and Kirtman2014). The basis sets for the H, C, N, and O atoms were those of Gatti et al. (Reference Gatti, Saunders and Roetti1994). The calculation was run on eight 2.1 GHz Xeon cores (each with 6 GB RAM) of a 304-core Dell Linux cluster at IIT, using eight k-points and the B3LYP functional, and took ~134 h.
III. RESULTS AND DISCUSSION
This synchrotron pattern of bis(ethanolammonium) eltrombopag of this study is similar enough to that reported in figure 28 of Leksic et al. (Reference Leksic, Ceric, Mundorfer, Antonac and Samardzic2011) to conclude that both samples represent Form I (Figure 4). The refined atom coordinates and the coordinates from the DFT optimization are reported in the CIFs deposited with the ICDD.
Comparison of the synchrotron pattern of bis(ethanolammonium) eltrombopag to the experimental pattern reported by Leksic et al. (Reference Leksic, Ceric, Mundorfer, Antonac and Samardzic2011). The published pattern was digitized using UN-SCAN-IT (Silk Scientific, 2013) and scaled to the synchrotron wavelength of 0.457899 Å using MDI JADE Pro (MDI, 2020).
The root-mean-square (rms) Cartesian displacement of the non-hydrogen atoms in the Rietveld-refined and DFT-optimized cation structures of the eltrombopag anion is 0.091 Å (Figure 5). The rms displacements for the ethanolammonium cations are even smaller at 0.017 and 0.042 Å. The excellent agreement of the two structures is strong evidence that the experimental structure is correct (van de Streek and Neumann, Reference van de Streek and Neumann2014). This discussion concentrates on the DFT-optimized structure. The asymmetric unit (with the atom numbering) is illustrated in Figure 6, and the crystal structure is presented in Figure 7. The crystal structure is dominated by columns of hydrogen-bonded cations and anions along the short b-axis. van der Waals interactions bind the columns together.
Comparison of the Rietveld-refined (red) and single-crystal (blue) structures of the eltrombopag anion in bis(ethanolammonium) eltrombopag.
The asymmetric unit of bis(ethanolammonium) eltrombopag, with the atom numbering. The atoms are represented by 50% probability spheroids.
The crystal structure of bis(ethanolammonium) eltrombopag, viewed down the b-axis.
Almost all of the bond distances, angles, and torsion angles fall within the normal ranges indicated by a Mercury Mogul Geometry check (Macrae et al., Reference Macrae, Sovago, Cottrell, Galek, McCabe, Pidcock, Platings, Shields, Stevens, Towler and Wood2020). Only the C17–N16–N15 angle of 125.2° (average = 119.3(14)°; Z-score = 4.36) is flagged as unusual. The standard uncertainty on the average is small, thus leading to a high Z-score. This amine nitrogen bridges two rings, and participates in an intramolecular hydrogen bond, so perhaps “unusual” nature of this angle is not surprising.
Quantum chemical geometry optimization of the eltrombopag anion (DFT/B3LYP/6-31G*/water) using Spartan'18 (Wavefunction, 2020) indicates that the observed solid-state conformation is very similar to the local minimum conformation of the isolated anion. In the minimum-energy conformation, the orientation of the ring containing the carboxyl group differs, but the rest of the molecule is similar. Intermolecular interactions are important in determining the solid-state conformation.
Analysis of the contributions to the total crystal energy using the Forcite module of Materials Studio (Dassault Systèmes, 2020) suggests that angle deformation terms are the dominant contributions to the intramolecular deformation energy, and that torsion terms are also significant. The intermolecular energy is dominated by van der Waals attractions and electrostatic repulsions, which, in this force-field-based analysis, include hydrogen bonds. The hydrogen bonds are better analyzed using the results of the DFT calculation.
As expected, hydrogen bonds are important in the crystal structure (Table I). Two H atoms of each ammonium group in the ethanolammonium cations participate in strong H-bonds, and the third H forms weaker bifurcated H-bonds (Figure 8). The energies of the N–H⋯O hydrogen bonds were calculated using the correlation of Wheatley and Kaduk (Reference Wheatley and Kaduk2019). One hydroxyl group of an ethanolammonium cation forms an O–H⋯O hydrogen bond to a carbonyl group, the energy of which was calculated using the correlation of Rammohan and Kaduk (Reference Rammohan and Kaduk2018). The hydroxyl group of the other cation forms an O–H⋯C hydrogen bond to the carbon atom of the carboxylate group. The methylene groups of one cation participate in C–H⋯O and C–H⋯N hydrogen bonds, while the methyl groups of the other cation are inactive. The amine group N16–H43 forms an intramolecular hydrogen bond to the phenoxyl O23. Several C–H⋯O and C–H⋯N hydrogen bonds also contribute to the crystal energy.
The principal hydrogen bonds in bis(ethanolammonium) eltrombopag.
Hydrogen bonds (CRYSTAL14) in bis(ethanolammonium) eltrombopag.
a Intramolecular.
The volume enclosed by the Hirshfeld surface (Figure 9; Hirshfeld, Reference Hirshfeld1977; Turner et al., Reference Turner, McKinnon, Wolff, Grimwood, Spackman, Jayatilaka and Spackman2017) is 697.61 Å3, 98.66% of one-fourth of the unit cell volume. The molecules are thus not tightly packed. All of the significant close contacts (red in Figure 9) involve the hydrogen bonds. The volume/non-hydrogen atom is 17.2 Å3.
The Hirshfeld surface of bis(ethanolammonium) eltrombopag. Intermolecular contacts longer than the sums of the van der Waals radii are colored blue, and contacts shorter than the sums of the radii are colored red. Contacts equal to the sums of radii are white.
The Bravais–Friedel–Donnay–Harker (Bravais, Reference Bravais1866; Friedel, Reference Friedel1907; Donnay and Harker, Reference Donnay and Harker1937) morphology suggests that we might expect elongated morphology for bis(ethanolammonium) eltrombopag, with 〈010〉 as the principal axes, or platy morphology, with {10-1} as the major faces. A second-order spherical harmonic model for preferred orientation was incorporated into the refinement. The texture index was 1.000(0), indicating that preferred orientation was not present in this rotated capillary specimen. The powder pattern of bis(ethanolammonium) eltrombopag from this synchrotron data set has been submitted to the ICDD for inclusion in the Powder Diffraction File.
IV. DEPOSITED DATA
The Crystallographic Information Framework (CIF) files containing the results of the Rietveld refinement (including the raw data) and the DFT geometry optimization were deposited with the ICDD. The data can be requested at info@icdd.com.
ACKNOWLEDGEMENTS
The use of the Advanced Photon Source at Argonne National Laboratory was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-06CH11357. This work was partially supported by the International Centre for Diffraction Data. We thank Lynn Ribaud and Saul Lapidus for their assistance in the data collection, and Andrey Rogachev for the use of computing resources at IIT.
CONFLICTS OF INTEREST
The authors have no conflicts of interest to declare.
