Wheldone Revisited: Structure Revision Via DFT-GIAO Chemical Shift Calculations, 1,1-HD-ADEQUATE NMR Spectroscopy, and X-ray Crystallography Studies

Wheldone is a fungal metabolite isolated from the coculture of Aspergillus fischeri and Xylaria flabelliformis, displaying cytotoxic activity against breast, melanoma, and ovarian cancer cell lines. Initially, its structure was characterized as an unusual 5-methyl-bicyclo[5.4.0]undeca-3,5-diene scaffold with a 2-hydroxy-1-propanone side chain and a 3-(2-(1-hydroxyethyl)-2-methyl-2,5-dihydrofuran-3-yl)acrylic acid moiety. Upon further examination, minor inconsistencies in the data suggested the need for the structure to be revisited. Thus, the structure of wheldone has been revised using an orthogonal experimental-computational approach, which combines 1,1-HD-ADEQUATE NMR experiments, DFT-GIAO chemical shift calculations, and single-crystal X-ray diffraction (SCXRD) analysis of a semisynthetic p-bromobenzylamide derivative, formed via a Steglich-type reaction. The summation of these data now permits the unequivocal assignment of both the structure and absolute configuration of the natural product.

F ungi are a proven source of bioactive compounds, some of which are problematic for our food supply (i.e., mycotoxins), 1−3 while others have been harnessed for the benefit of humanity. 4,5It is now possible to analyze the genetic architecture of fungi, 6 where the annotation of biosynthetic gene clusters 6,7 leads to predictions of their ability to biosynthesize far more secondary metabolites than are observed under standard laboratory conditions. 8To explain this disconnect, a common hypothesis is that fungi must survive hostile environments in Nature, typically competing for resources with other microorganisms and causing them to generate defensive chemicals. 9−15 Recently, coculture studies of Aspergillus f ischeri (strain NRRL181) and Xylaria f labelliformis (strain G536) 16 were shown to stimulate the biosynthesis of a suite of metabolites, 17 including one cytotoxic compound (i.e., wheldone), which combined an interesting molecular structure with notable activity against breast, melanoma, and ovarian cancer cell lines. 18Wheldone was characterized as consisting of an unusual 5-methyl-bicyclo [5.4.0]undeca-3,5-diene scaffold with a 2hydroxy-1-propanone side chain and a 3-(2-(1-hydroxyethyl)-2-methyl-2,5-dihydrofuran-3-yl)acrylic acid moiety, which were established based on 1D-and 2D-NMR spectroscopy and mass spectrometry data.At that time, the proposed structure was not verified via orthogonal studies such as single-crystal X-ray diffraction (SCXRD).However, upon further optimization of the coculture conditions, it was possible to generate a larger stock of wheldone, which was needed for additional pharmacological studies that will be reported in the future.During these experiments, re-examination of the spectral data highlighted several anomalies in the original structure assignment, and therefore, a more in-depth analysis of the structure elucidation was embarked upon.Thus, the goal of the current study was to re-evaluate the structure assignment of wheldone by an orthogonal experimental-computational approach, through (a) examination of the chemical shift assignments by DFT-calculations, (b) performance of additional NMR studies, particularly 1,1-HD-ADEQUATE, 19 which provides visualization of carbon−carbon connectivity through the structural backbone, and (c) generation of crystals suitable for SCXRD studies. 20Analysis of the data from this combination of molecular characterization techniques led to the structural revision of wheldone.Two of the authors (NHO and HAR) of this manuscript were coauthors of the originally published structure of wheldone and apologize to the scientific community for this error, which has been rectified, vide infra.

■ RESULTS AND DISCUSSION
Our re-examination of the wheldone structural assignment began with a comparison of the experimental NMR spectra with those calculated.Initially, an exhaustive conformational search was conducted by MMFF94 force field in Spartan'10 (Wave function).All conformers with relative energies within 5 kcal/ mol were then optimized in Gaussian 16 at the M062X/6-31+G(d,p) level of theory using the integral equation formalism polarizable continuum model (IEFPCM) in CH 3 OH.Next, the optimized conformers were used to calculate both the 1 H and 13 C NMR chemical shifts at the B3LYP/6-311+G(2d,p) level of theory with the IEFPCM, also in CH 3 OH.At first glance, the calculated chemical shifts suggested that the published structure of wheldone indeed displayed structural assignment issues, with mean absolute errors for proton ( 1 H-MAE) of 0.24 and 5.9 ppm for carbon ( 13 C-MAE) (Table S9).Discrepancies between the calculated and experimental NMR chemical shifts were most evident in the seven-membered ring and the 3-(2,5-dihydrofuran-3-yl)acrylic acid moiety (Figure 1).Specifically, significant errors between calculated and experimental chemical shift values of approximately 20 ppm were noted for both C-18 and C-19, as well as deviations of about 10 ppm for the chemical shifts of C-10 and C-15.Likewise, the chemical shifts from C-2 to C-5 showed deviations that averaged around 5.0 ppm.A similar pattern was observed in the 1 H NMR data, where major differences (ranging between 0.09 to 1.04 ppm) were observed in both H-3 and H-5, suggesting issues with original assignment of the 3-(2,5-dihydrofuran-3-yl)acrylic acid moiety, and in H-9, H-10, H-11, H-15, and H-18, indicating that revisions might be needed in the bicyclo [5.4.0]undeca-3,5-diene framework.
In reviewing the previous structure elucidation data, 18 key concerns were also raised with the unusual four-bond HMBC ( 4 J CH ) couplings proposed from H 3 -25 to C-19 and from H 3 -23 to C-5 (Figure 1).It is well-established that the intensity of the HMBC cross-peaks depends on the magnitude of 2 J CH , 3 J CH , and 4 J CH coupling constants as well as the geometry of the coupled atoms.The conventional settings of an HMBC experiment are optimized in the range of 6 to10 Hz, which typically correspond to the magnitude of 2 J CH and 3 J CH correlations and, to a lesser extent, the much smaller 4 J CH couplings. 21For example, a weak 4 J CH coupling was observed from H-18 to C-8, as expected. 18owever, the 4 J CH couplings observed from H 3 -25 to C-19 and from H 3 -23 to C-5 appear to be of the same intensity as the other 2 J CH and 3 J CH couplings observed from H 3 -25 to C-17/C-16 and H 3 -23 to C-7/C-8, respectively.Despite the fact that 4 J CH HMBC correlations can be observed in conjugated systems (as noted with termicalcicolanone B) 22 and in some rigid ring systems (such as yardenones A and B 23 and alternarilactone A), 24 they are typically less intense than 2 J CH and 3 J CH couplings present in those molecules.Thus, we hypothesized that some of the 4 J CH HMBC correlations proposed for wheldone 18 likely corresponded to 3 J CH correlations.Further, reanalysis of the HMBC correlations of the 2,5-dihydrofuran ring suggested that the acrylic acid moiety was more likely to be at position C-5 than at position C-4.In a similar fashion, δ C 157.3 was most likely to be in position C-18 rather than C-19, leading to potential improved assignment of 3 J CH HMBC correlations from H 3 -23 to δ C 142.8 and from H 3 -25 to δ C 157.3, respectively, instead of the 4 J CH HMBC correlations described previously.In summary, the evidence obtained from DFT calculations, coupled with the improbable 4 J CH HMBC correlations (Figure 1), suggested that the previously published structure was incorrect, pointing toward a different fused ring system and an alternate substitution pattern around the 2,5-dihydrofuran ring.
To elucidate the backbone of wheldone, particularly the fused ring system, a new set of 1D-and 2D-NMR experiments, including the 1,1-HD-ADEQUATE experiment, 19 were carried out.Note, for the following discussion, the positions in the structure of wheldone have been renumbered for clarity (see Figure 2, Table S7).Analysis of 1,1-HD-ADEQUATE and δ C 157.3 (C-18) suggested a bicyclo[4.4.0]dec-4-ene ring arrangement with an exocyclic 4-hydroxypent-1-en-3-one moiety instead of the bicyclo[5.4.0]undeca-3,5-diene proposed previously (Figure 2).This decalin-like ring system was supported by the observed 3 J CH HMBC correlations from δ H 1.91 (H 3 -25) to δ C 143.0 (C-16) and δ C 157.3 (C-18) and the NOESY correlation observed between δ H 1.91 (H 3 -25) and δ H 6.57 (H-19), which indicates an E-configuration of the exocyclic double bond.A long-range 1,n-HD-ADEQUATE 19 correlation was observed between δ C 122.0 (C-19) and δ C 74.2 (C-21), which provided additional evidence for the 4-hydroxypent-1-en-3-one moiety (Figure 2).Moreover, 1,1-HD-ADEQUATE connectivities showed attachments of δ C 142.8 (C-5) to δ C 95.2 (C-6) and of δ C 74.8 (C-7) to δ C 137.1 (C-4); these findings  Note, the original proposed structure of wheldone was redrawn (left), so that some of the positions would align with the newly proposed structure (right).In addition, it was necessary to renumber some of the positions (noted in red) in the revised structure to fit the IUPAC rules.
From the previous study, 18 the configuration of position 21 was determined as S via Mosher's ester analysis (confirmed via SCXRD analysis, vide infra).Using that information as a reference point, we could derive additional support for the revised structure of wheldone from the NOESY data.Of particular importance, the NOESY cross peaks observed between H-10 and H-15 indicated a cis-decalin-like fused ring.Also, NOESY correlations observed between H-9/H-10/H 3 -24 supported the cis-decalin conformation, and the equatorial orientation of H 3 -24 suggested either a 9S,10R,11S,15S or a 9R,10S,11R,15R configuration (Figure S12).Thus, the NOESY correlations observed between H-8 and H-5/H 2 -7/H-10/H-15/ H 3 -23 and between H 3 -23 and H-5/H-8/H-9/H-10 were possible regardless of the configuration of positions 6 and 8, due to the free rotation of the C-6/C-8/C-9 bonds.Thus, analysis of the NOESY correlations, while inconclusive, permitted a reduction of the assignment of the absolute configuration of wheldone to eight possibilities (Table S15).
To explore these structural hypotheses, exhaustive conformer search was performed by combining the results from five different conformer generators: Schrodinger Macro-Model Monte Carlo Multiple Minima (version 2021-1), Schrodinger Confgenx (version 2021-1), OpenEye Omega Classic (version 3.1.2.2), OpenEye Omega Macrocycle (version 3.1.2.2), and MOE Low Mode MD (version 2020.09).Each conformer was minimized using the OPLS4 25 force field implemented in MacroModel, and conformers less than a 10.0 kcal/mol cutoff were eliminated as were any redundant conformers based on atomic RMSD.This resulted in final ensembles of between 100 to 500 conformers for each test structure derived from the 1,1-HD-ADEQUATE and NOESY correlations.Conformers were optimized by using DFT calculations at the B3LYP/6-31+G(d,p) level of theory in the gas phase.Based on DFT methodologies reported by Pierens, 26 1 H and 13 C NMR chemical shifts were calculated at the WP04/ aug-cc-pVZD and mPW1PW91/6-311+G(2d,p) levels of theory, respectively, with the IEFPCM in CH 3 OH.Comparisons between the calculated and experimental chemical shifts showed that the newly proposed decalin-like part of the structure of wheldone was a much better fit than prior structure proposals.Specifically, the experimental and calculated 13 C NMR chemical shift values were in closer agreement in the decalin-like part of the structure compared to the originally proposed structure (see Table S12).Among the eight structure possibilities, DFT NMR chemical shift calculations for configuration 6R,8S,9S,10R,11S,15S,21S most closely matched the experimental data with a MAE of 0.09 and 2.4 ppm for the 1 H and 13 C NMR data, respectively (Table S15).In short, these data provided a possible, but not unequivocal, assignment of the configuration of the asymmetric centers in wheldone.
To unambiguously determine the structure and configuration of wheldone, we aimed to generate crystals suitable for SCXRD analysis.Initially, high-throughput crystallization techniques using ENaCt 20,27 (encapsulated nanodroplet crystallization) were attempted, whereby 480 nanoscale parallel crystallization experiments were carried out using approximately 7 mg of isolated natural product.Unfortunately, only microcrystalline materials were observed and from only a limited number of ENaCt experiments (see Supporting Information).−31 Thus, taking advantage of the carboxylic acid moiety in wheldone, a Steglich-type 32−34 reaction was carried out using p-bromobenzylamine in dichloromethane with DCC and DMAP, leading to the formation of the wheldone p-  bromobenzylamide derivative. 35,36HRESIMS analysis helped to establish the molecular formula of this derivative as C 32 H 40 O 5 NBr (m/z 598.2150, [M + H] + ), which corresponded to an index of hydrogen deficiency of 13 and was consistent with the addition of a phenyl.Two doublets at δ H 7.22 (J = 8.10 Hz, 2H) and δ H 7.47 (J = 8.10 Hz, 2H), characteristic of a 1,4disubstituted benzene, a singlet signal of a methylene at δ H /δ C 4.41/43.6ppm, and an amide carbonyl at δ C 168.1, as well as HMBC correlations from δ H 7.33/5.87/4.41 to δ C 168.1 ppm, confirmed the position of the p-bromobenzylamide moiety (Table S8, Figures S18−S23).This compound was subjected to a series of classical crystallization experiments utilizing slow evaporation from nine different solvents.Microcrystalline solids were observed from CHCl 3 , THF, 2-MeTHF, EtOAc, and nitromethane but were unsuitable for SCXRD analysis.Based on those initial results, follow-up crystallizations by layered diffusion 37 were carried out, with a 2-MeTHF/cycloheptane solvent system ultimately providing suitable single crystals for SCXRD analysis.
Wheldone p-bromobenzylamide crystallized in the P1 space group as a 2-methyl tetrahydrofuran and cycloheptane solvate.Further analysis of the single-crystal structure solution (Figure 4; Table S5) illustrated agreement with the backbone proposed from the 1,1-HD-ADEQUATE experiment, showing unambiguously a cis-decalin-like ring system and the E-orientation of the exocyclic double bond in the 4-hydroxypent-1-en-3-one moiety.Moreover, based on the Flack parameter (−0.02(4),Parsons' method), the absolute configuration was assigned as 6R,8S,9S,10R,11S,15S,21S; importantly, this was in agreement with the configuration of C-21 that was assigned previously by Mosher's ester analysis 18 and with the observed NOESY correlations (Figure 3).Additionally, this configuration agreed with the structure that yielded the highest accuracy for the 1 H and 13 C NMR chemical shift predictions.The structure also exhibited an S(8) intramolecular hydrogen bond between the hydroxy at C-8 and the ketone at C-20 in the solid state.An additional C(4) hydrogen bonding network was also observed between N-1 and O-1. 38otably, the diffraction pattern of wheldone p-bromobenzylamide revealed a second set of much weaker reflections, indicative of incommensurate modulation within the structure; the modulation vector was able to be refined to q = (0,− 0.004(8),0.2414(8)).Analysis of the modulation using the program Superflip 39,40 revealed that the modulation is localized mainly in the solvent region and did not affect the main molecule.It would be very difficult to model the modulation of the solvent, and the impact on the structure description of the molecule would be minimal.Therefore, we proceeded with the structure analysis using only the main reflections and, consequently, the average structure, as detailed in the Supporting Information.
In summary, the structure of wheldone, a product of the competition of two fungal strains in coculture, has been revised, including the unequivocal assignment of the absolute configuration as 6R,8S,9S,10R,11S,15S,21S.To do so, a combination of orthogonal techniques was used, including working through computational data, NMR spectroscopy using the 1,1-HD-ADEQUATE experiment, appending a p-bromobenzyl moiety via semisynthesis to enhance crystallization properties, and finally, analyzing X-ray crystallographic data.
■ EXPERIMENTAL SECTION General Experimental Procedures.Optical rotation data were obtained using a Rudolph Research Autopol III polarimeter, and UV spectra were measured with a Varian Cary 100 Bio UV−vis spectrophotometer.NMR data were collected using an Agilent 700 MHz NMR spectrometer equipped with a cryoprobe, operating at 700 MHz for 1 H and 175 MHz for 13 C, a Bruker AVANCE III 600 NMR spectrometer with a BCU-05 cooling unit with liquid N 2 dewar, operating at 600 MHz for 1 H and 150 MHz for 13 C, and a Bruker Neo NMR spectrometer equipped with a H/F C/N TCI 5 mm Prodigy CryoProbe operating at a 1 H observation frequency of 500 MHz for 1 H and 125 MHz for 13 C.In all cases, the NMR data were referenced to the residual solvent peaks for methanol-d 4 , specifically δ H /δ C 3.31/49.0.HRMS data were acquired using a Thermo LTQ Orbitrap XL linear ion trap mass spectrometer equipped with a heated electrospray ionization source coupled to a Waters Acquity UPLC system by using an ACQUITY UPLC BEH C 18 column (1.7 μm; 50 × 2.1 mm) set to 40 °C and a flow rate of 0.3 mL/min.The elution method consisted of a linear gradient of CH 3 CN-H 2 O (both acidified with 0.1% formic acid), starting at 15% CH 3 CN and holding it for 1 min, and then increasing linearly to 100% CH 3 CN over 8 min with a 1.5 min hold before returning to the starting conditions.Analytical and preparative HPLC analyses were performed with a Varian Prostar HPLC system equipped with two Prostar 210 pumps, a Prostar 701 fraction collector, a Prostar 335 photodiode array detector (PDA; Varian Inc.), and a SEDEX75 evaporative light scattering detector (ELSD; SEDERE Inc.) using either Phenomenex Synergi Max-RP C 12 80 Å analytical (4 μm; 250 × 4.6 mm) and preparative (4 μm; 250 × 21.2 mm) or Phenomenex Gemini−NX C 18 110 Å analytical (5 μm; 250 × 4.6 mm) and preparative (5 μm; 250 × 21.2 mm) columns.Data collection and analysis were carried out using Galaxie Chromatography Workstation software (version 1.9.3.2,Varian Inc.).Flash chromatography was performed on a Teledyne ISCO CombiFlash Rf 200 using various sizes of RediSep Rf GOLD silica gel columns and monitored by both ELSD and PDA detectors.Single-crystal X-ray diffraction analysis was performed using a Rigaku XtaLAB Synergy diffractometer equipped with a microfocus sealed Cu Kα X-ray tube (λ = 1.54184Å) and a HyPix Arc-100 detector.
Fungal Strains, Fermentation, and Isolation Procedures.Aspergillus f ischeri (strain NRRL 181) was obtained from the ARS Culture Collection (NRRL), as noted previously. 41Xylaria f labelliformis (strain G536) 17,18 was isolated as an endophyte from surface sterilized twigs of Asimina triloba and identified using molecular methods, as detailed previously. 16These cultures were grown, first individually and then in coculture on Quaker breakfast oatmeal, essentially as described previously. 18To generate enough wheldone for the NMR spectroscopy and X-ray crystallography studies, the cocultures were grown in batches of 25 flasks each (i.e., each 250 mL Erlenmeyer flask containing 10 g of autoclaved oatmeal).The isolation and purification procedures were modified from the reported procedure, 18 as will be detailed in a forthcoming manuscript, so as to yield ∼1 mg of wheldone per flask.
Crystallization Experiments.Crystallization of wheldone was attempted using the encapsulated nanodroplet crystallization (ENaCt) methods. 20This proceeded through a series of experiments, as detailed in the Supporting Information.Additionally, a classical slow evaporation method was also attempted (see Supporting Information).Eventually, it was necessary to generate an analogue, wheldone, pbromobenzylamide, as detailed below.The crystallization of this derivative was attempted by both classical slow evaporation and layered diffusion methods (see Supporting Information), the latter of which was successful.See "X-ray Crystal Structure Analysis of the Wheldone p-bromobenzylamide Derivative" in the Supporting Information for further details.
Preparation of the Wheldone p-Bromobenzylamide Derivative.To a solution of wheldone (12.51  (750 μL).The mixture was stirred using a magnetic bar and kept under a nitrogen atmosphere at 0 °C in a water-ice bath for 8 h and then kept at room temperature for another 17 h.Next, the reaction mixture was dried under nitrogen, reconstituted in dioxane-CH 3 OH (1:1), and immediately fractionated using a Synergi Max-RP C 12 column using a solvent system that started at 40:60 CH 3 CN-H 2 O (0.1% formic acid in both solvents) over 3 min, then 40:60 to 60:40 over 27 min, then 60:40 to 100:0 over 25 min, and finally 100:0 over 7 min at a flow rate of 21.2 mL/min and a Phenomenex Gemini−NX C 18 column with a gradient system of 60:80 to 80:20 CH 3 OH-H 2 O (0.1% formic acid) over 30 min, then 80:20 to 100:0 over 10 min, and finally 100:0 over 7 min at a flow rate of 17.0 mL/min.
Wheldone (1) Computational Details.3D models of the originally proposed wheldone structure were built using Spartan'10 software.Conformational analysis was performed using the MMFF94 molecular mechanics force field with the Monte Carlo search protocol.The resulting conformers under the 5 kcal/mol cutoff were checked for duplicates, and then geometry and frequency were optimized using the densityfunctional theory (DFT) method at the M06-2X/6-31+G(d,p) theory level in the gas phase.For the isotropic shielding tensor calculations, the gauge invariant atomic orbital (GIAO) method was used at B3LYP/6-311+G(2d,p) level of theory with the integral equation formalismpolarizable continuum model (IEFPCM) model in CH 3 OH.Upon acquisition of 1,1-HD-ADEQUATE and NOESY NMR data, conformer searches were carried out for possible structures of wheldone using the following programs: Schrodinger MacroModel Monte Carlo Multiple Minima (version 2021-1), Schrodinger Confgenx (version 2021-1), OpenEye Omega Classic (version 3.1.2.2), OpenEye Omega Macrocycle (version 3.1.2.2), and MOE Low Mode MD (version 2020.09).Conformers were minimized using the OPLS4 25 force field in MacroModel, and higher energy conformers (>10.0 kcal/mol) were eliminated.Redundant conformers sharing equivalent energies were also removed.Conformer optimizations were carried out at the B3LYP/6-31+G(d,p) level of theory in the gas phase. 1 H and 13 C NMR chemical shifts were calculated at the WP04/aug-cc-pVZD and mPW1PW91/6-311+G(2d,p) levels of theory, respectively, with the IEFPCM in CH 3 OH, following methods reported by Pierens. 26The final 1 H and 13 C chemical shifts were calculated for each structure in accordance with the Boltzmann distribution and their relative energies.All calculations were performed employing the Gaussian'16 program software package. 42

■ ASSOCIATED CONTENT Data Availability Statement
The NMR data for wheldone have been deposited in the Natural Products Magnetic Resonance Database (NP-MRD; www.np_mrd.org)and can be found at NP0021048.Also, CCDC 2330603 contains the supplementary crystallographic data for this paper, and these data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_request@ccdc.cam.ac.uk, or by contacting The Cambridge Crystallographic Data Centre,12 Union Road, Cambridge CB2 1EZ, UK; fax:+441223336033.
Experimental procedures, characterization data, spectroscopic and crystallographic data (PDF) Illustrated crystallographic information (CIF) Illustrated crystallographic information (CIF)

Figure 1 .
Figure 1.Identification of NMR spectral discrepancies in the originally proposed structure of wheldone.(a) Key HMBC correlations noted in the original publication with 4 J CH in red.(b) Selected Δδ ppm 13 C NMR (red) and Δδ ppm 1 H NMR (blue) of the experimental vs predicted chemical shifts.

Figure 2 .
Figure 2. 1,1-HD-ADEQUATE connectivities and key HMBC and NOESY correlations that supported the revised structure of wheldone.Note, the original proposed structure of wheldone was redrawn (left), so that some of the positions would align with the newly proposed structure (right).In addition, it was necessary to renumber some of the positions (noted in red) in the revised structure to fit the IUPAC rules.

Figure 3 .
Figure 3. (a) Key COSY and additional NOESY correlations that supported the structure revision of wheldone.(b) Most probable configuration of wheldone based on DFT-GIAO calculations.

Figure 4 .
Figure 4. Structure of wheldone p-bromobenzylamide with the chiral centers labeled and assigned.The crystallographically independent, nondisordered molecule is shown, and anisotropic displacement parameters are displayed at 50%.Key: brown−bromine, red−oxygen, blue−nitrogen, gray−carbon, green−hydrogen.An additional perspective of the molecule is displayed in Figure S3.

■ AUTHOR INFORMATION Corresponding Author Nicholas
H. Oberlies − Department of Chemistry and Biochemistry, University of North Carolina at Greensboro, Greensboro, North Carolina 27402, United States; orcid.org/0000-0002-0354-8464;Email: Nicholas_Oberlies@uncg.edu Advisory Board of Clue Genetics, Inc. NHO is a member of the Scientific Advisory Boards of Mycosynthetix, Inc. and Ionic Pharmaceuticals, LLC.MJH and MRP are directors of and shareholders in Indicatrix Crystallography Ltd.