Issue contents
Download PDF of article
Download CIF
3D view
Navigation
Highlighting
Dictionary of Crystallography reference
IUPAC Gold Book reference
more ...
Download citation
FormatBIBTeX
EndNote
RefMan
Refer
Medline
CIF
SGML
Plain Text
share
Previous article
Next article
Previous article
Issue contents
Download PDF of article
Download CIF
3D view
Navigation
Highlighting
Dictionary of Crystallography reference
IUPAC Gold Book reference
more ...
Download citation
FormatBIBTeX
EndNote
RefMan
Refer
Medline
CIF
SGML
Plain Text
share
Next article

research papers\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoSTRUCTURAL
CHEMISTRY
ISSN: 2053-2296
Volume 72| Part 1| January 2016| Pages 14-20
doi:10.1107/S2053229615022305

Structural corroboration of two important building blocks of the anti­cancer drug eribulin mesylate through two-dimensional NMR and single-crystal X-ray diffraction studies

Srinivas Venkata Pullela,a Vinod Acharya,a* Nagarjuna Reddy,a Jayvant Harlikar,b Amar Kulkarni,b Rakesh Chavan,b Ajay Yadav,c Shuvendu Mannac and Angshuman Ghoshc

aAPI R&D, Cipla Ltd, MIDC Patalganga, Rasayani, Maharashtra 410 220, India, bAPI–ADL, Cipla Ltd, MIDC Patalganga, Rasayani, Maharashtra 410 220, India, and cBlock BN, Plot 7, TCG Life Sciences, Salt Lake, Sector V, Kolkata, West Bengal 700 091, India
*Correspondence e-mail: vinod.acharya@cipla.com

Edited by P. Fanwick, Purdue University, USA (Received 14 October 2015; accepted 20 November 2015; online 1 January 2016)

Eribulin mesylate, one of the most synthetically challenging drugs to date, pos­sesses 19 stereocentres in its structure and ascertaining the absolute stereochemistry at every stage of the 64-stage synthesis is crucial. In our quest to synthesize eribulin, we identified two critical building blocks of this mol­ecule, namely 3,4:6,7-di-O-cyclo­hexyl­idene-D-glycero-α-L-talo-hepto­pyran­ose methanol monosolvate, C19H30O7·CH3OH, and (2R,3R,4R,5S)-5-allyl-2-[(S)-2,3-di­hy­droxy­prop­yl]-4-[(phenyl­sulfon­yl)meth­yl]­tetra­hydrofuran-3-ol, C17H24O6S, for which two-dimensional NMR (2D-NMR) data were not sufficient to prove the absolute configuration. To ensure structural integrity, single-crystal X-ray diffraction data were obtained to confirm the structures. This information provides useful insights into the structural framework of the large eribulin mesylate mol­ecule.

CCDC references: 1438133; 1438132

Similar articles

1. Introduction

Eribulin mesylate is a structurally truncated synthetic ana­logue of Halichondrin B (Hirata & Uemura, 1986[Hirata, Y. & Uemura, D. (1986). Pure Appl. Chem. 58, 701-710.]; Kishi et al., 1994[Kishi, Y., Fung, F. G., Forsynth, C. J., Scola, P. M. & Yoon, S. K. (1994). US Patent 5 338 865.]), the most bioactive natural product isolated from the marine sponge Halichondria okadai commonly found off the coasts of Japan and New Zealand. Eribulin mesylate (Halaven®) inter­feres with microtubule dynamics (Smith et al., 2010[Smith, J. A., Wilson, L., Azarenko, O., Zhu, X., Lewis, B. M., Littlefield, B. A. & Jordan, M. A. (2010). Biochemistry, 49, 1331-1337.]) and was approved in November 2010 by the United States Food and Drug Administration (USFDA) for the treatment of metastatic breast cancer (Towle et al., 2001[Towle, M. J., et al. (2001). Cancer Res. 61, 1013-1021.]). The structure of eribulin resembles a macrocyclic ketone showcasing a fully synthetic drug available on the market today. 19 of the 36 C atoms that constitute the skeleton of the mol­ecule are stereogenic in nature.

The structural framework of eribulin is built up by assembling three key fragments: (i) the C1–C13 aldehyde fragment (I), (ii) the C14–C26 vinyl triflate fragment (II) and (iii) the C27–C35 phenyl sulfone fragment (III) (Fig. 1[link]).

[Figure 1]
Figure 1
Retero-synthetic scheme for the coupling of fragments I and IV, and the construction of eribulin mol­ecule (V).

Several synthetic routes have been used for the preparation of eribulin, (V), each of which utilize the same strategy described by Kishi and co-workers, known for their pioneering work on the total synthesis of Halichondrins, and Halichondrin B in particular (Aicher et al., 1992[Aicher, T. D., Buszek, K. R., Fang, F. G., Forsyth, C. J., Jung, S. H., Kishi, Y., Matelich, M. C., Scola, P. M., Spero, D. M. & Yoon, S. K. (1992). J. Am. Chem. Soc. 114, 3162-3164.]). Over the years, synthetic routes have evolved continuously, with scale-up and route refinement as the key areas of improvement (Yu et al., 2013[Yu, M. J., Zheng, W. & Seletsky, B. M. (2013). Nat. Prod. Rep. 30, 1158-1164.]; Austad et al., 2013[Austad, B. C., et al. (2013). Synlett, 24, 6327-6332.]).

The three key fragments (I)–(III) (Fig. 1[link]) are synthesized in 15–20 stages each and have been reported in detail. In the later stages, (II) and (III) are subjected to Nozaki–Hiyama–Kishi (NHK) coupling (Hiyama et al., 1981[Hiyama, T., Kimura, K. & Nozaki, H. (1981). Tetrahedron Lett. 22, 1037-1040.]; Jin et al., 1986[Jin, H., Uenishi, J., Christ, W. & Kishi, Y. (1986). J. Am. Chem. Soc. 108, 5644-5646.]) to yield fragment (IV) (the C14–C35 fragment) (Fig. 1[link]), which is then coupled with (I) under basic conditions. The final stages (Fig. 2[link]) involve a macrocyclization achieved again through NHK coupling of (VIII), the formation of cyclic ketal (IX) and the conversion of terminal diol (X) to an in situ epoxide on which the primary amine is added to finally realize eribulin mesylate, (V).

[Scheme 1]
[Figure 2]
Figure 2
The final stages in the synthesis of eribulin mesylate, (V).

The goal of our synthesis was to conduct feasibility studies and develop an optimized and scalable process for the preparation of (V). Owing to the presence of 19 stereocentres in this mol­ecule, it was required to continuously ensure the correct stereoisomer at every stage of this extensive synthesis. The important check-point was to characterize newly formed stereocentres in selected mol­ecules through two-dimensional NMR (2D-NMR) data. There were, however, instances where 2D-NMR data were not conclusive enough to prove the absolute stereochemistry or there was a need to prove the mol­ecular geometry for subsequent reactions. Hence, single-crystal X-ray diffraction analysis was employed to identified inter­mediates of compounds (I)–(III) in order to conclusively establish the structures (Fig. 1[link]).

The single-crystal X-ray structure of (IV) has been reported previously (Austad et al., 2013[Austad, B. C., et al. (2013). Synlett, 24, 6327-6332.]). Also, a 3,5-di­nitro­phenyl ester derivative of the alcohol variant of (III) (Fig. 1[link]) has been reported, denoted (IIIa) (Yang et al., 2009[Yang, Y. R., Kim, D. S. & Kishi, Y. (2009). Org. Lett. 11, 4516-4519.]) (Scheme 1)[link].

We report here the single-crystal X-ray data for two relatively important inter­mediates of (I) and (III) which have an important bearing on the structural orchestration of the eribulin mol­ecule.

The selected inter­mediate of (I) is 3,4:6,7-di-O-cyclo­hexyl­idene-D-glycero-α-L-talo-hepto­pyran­ose, (XII) (Scheme 2[link]), whose stereochemistry has important implications for the C-allyl­ation product (XIV) two stages later (Fig. 3[link]). The synthesis of (I) starts with D-gulono-γ-lactone (Fig. 3[link]) having two sets of vicinal diol groups, the stereochemistries of which are fixed. The D-gulono-γ-lactone is converted into its lactol which reacts with MeOCH2PPh3Cl under Wittig conditions to pro­duce hy­droxy­alkene (XI). The Sharpless di­hydroxy­lation protocol through which (XII) is finally obtained involves a catalytic osmylation (Kolb et al., 1994[Kolb, H. C., VanNieuwenhze, M. S. & Sharpless, K. B. (1994). Chem. Rev. 94, 2483-2547.]) of (XI) using the double stereo-differentiation strategy (Masamme et al., 1985[Masamme, S., Choy, W., Peterson, J. S. & Sita, L. R. (1985). Angew. Chem. 24, 1-76.]). It provides a 4:1 mixture of diastereoisomers, which after purification delivers (XII) as a single diastereoisomer in the form of a white crystalline solid. This mechanism should force the H atom on C1 (i.e. H1) to lie on the opposite side of the tetrahydropyran ring plane in the desired diastereoisomer, compared with the other H atoms on atoms C2–C5 (Fig. 4[link]). However, the nuclear Overhauser effect spectroscopy (NOESY) (Fig. 5[link]) of this mol­ecule unusually showed that atom H1 has an inter­action with H2, which normally means that they are on the same side of the tetrahydropyran ring plane. Further insight into this mol­ecule could be provided by using single-crystal X-ray diffraction to align the mechanistic rationale and determine the absolute configuration. Com­pound (XII) is then subjected to acetyl­ation yielding tetra­ace­tate (XIII). C-Allyl­ation of (XIII) using methyl 3-(tri­methyl­silyl)pent-4-enoate (also known as the Kishi alkene) retains the configuration on C1 after displacement of the O-acetyl group to yield allylic ester (XIV) which exists as a 1:1 mixture of conformers (Stamos & Kishi, 1996[Stamos, D. P. & Kishi, Y. (1996). Tetrahedron Lett. 37, 8643-8646.]).

[Scheme 2]
[Figure 3]
Figure 3
An overview of the synthetic scheme for the preparation of (I).
[Figure 4]
Figure 4
The mol­ecular structure of (XII), showing atom H1 in a different plane to the other H atoms of the cyclo­hexyl­idene ring. Displacement ellipsoids are drawn at the 50% probability level.
[Figure 5]
Figure 5
The 2D-NOESY spectrum of (XII).

The second inter­mediate, (2R,3R,4R,5S)-5-allyl-2-[(S)-2,3-di­hydroxy­prop­yl]-4-[(phenyl­sulfon­yl)meth­yl]­tetra­hydrofuran-3-ol, (XV) (Scheme 2[link]), from the synthesis of (III) is obtained through a sequence of 15 chemical transformations (Fig. 6[link]) starting from D-glucorono-6,3-lactone (Yang et al., 2009[Yang, Y. R., Kim, D. S. & Kishi, Y. (2009). Org. Lett. 11, 4516-4519.]). The correlation NMR spectra (COSY) of the tetra­hydro­furan framework present in (V) and its inter­mediates show that the vicinal H atoms on C2 and C3 do not correlate. It became vital to understand the spatial orientation of these two H atoms along with the structural framework so as to deduce the relative configuration of this mol­ecule. Further, the C6—C7 diol functionality is introduced through asymmetric di­hydroxy­lation in a 3:1 diastereoisomeric ratio in (XX). The undesired isomer after two stages at (XXI) is removed through recrystallization. Our attempts to prepare single crystals of (XXI) were unsuccessful. We then went five stages further to triol (XV) (Fig. 4[link]) which exists as a white crystalline solid and was successful crystallized.

[Figure 6]
Figure 6
An overview of the synthetic scheme for the preparation of (V).

2. Experimental

2.1. General considerations

Compounds (XII) and (XV) were prepared according to existing synthesis routes (Charles et al., 2013[Charles, E. C., Francis, G. F., Bryan, M. L., Gordon, D. W., Matthew, J. S. & Xiaojie, Z. (2013). Synlett, 24, 323-326.]; Brian et al., 2013[Brian, C. A., et al. (2013). Synlett, 24, 327-332.]). NMR spectra were recorded at room temperature using an Agilent 500 MHz spectrometer. 1H chemical shifts are reported in p.p.m. referenced to tetra­methyl­silane (TMS, 0.0 p.p.m.). 13C chemical shifts are reported in p.p.m. referenced to the solvent resonance of 39.5 p.p.m. for DMSO-d6.

2.2. NMR data

2.2.1. 1H and 13C NMR data for (XII)

1H NMR (500 MHz, DMSO-d6): δ 1.30–1.35 (m, 4H), 1.42–1.60 (m, 16H), 3.43–3.44 (dd, J = 3 Hz, 1.5 Hz, 1H), 3.56–3.58 (dd, J = 2, 1.5 Hz, 1H), 3.66–3.69 (t, 1H), 3.94–3.97 (q, 1H), 4.00–4.05 (q, 1H), 4.14–4.16 (dd, J = 2.5, 1.5 Hz, 1H), 4.29–4.32 (dd, J = 2.5, 3 Hz, 1H), 4.79–4.80 (d, J = 6 Hz, 1H), 5.23–5.24 (d, 1H), 6.34–6.35 (d, 1H). 13C NMR (125 MHz, DMSO-d6): δ 23.38 (–CH2), 23.44 (–CH2), 23.50 (–CH2), 23.62 (–CH2), 24.63 (–CH2), 24.66 (–CH2), 34.60 (–CH2), 34.98 (–CH2), 35.47 (–CH2), 35.86 (–CH2), 64.78 (–CH2), 68.87 (–CH), 70.93 (–CH), 74.20 (–CH), 74.85 (–CH), 75.65 (–CH), 93.59 (–CH), 108.84 (quaternary C), 109.48 (quaternary C). The NMR spectra are available in the Supporting information .

2.2.2. 1H and 13C NMR data for (XV)

1H NMR (500 MHz, DMSO-d6): δ 1.51–1.53 (m, 1H), 1.69–1.71 (m, 1H), 2.00–2.01 (t, 1H), 2.22–2.54 (m, 2H), 3.25–3.28 (m, 2H), 3.31–3.32 (d, 1H), 3.43–3.47 (m, 2H), 3.53–3.54 (q, 1H), 3.72–3.73 (q, 1H), 3.95–3.97 (t, 1H), 4.39–4.42 (m, 2H), 4.80–4.81 (d, J = 5 Hz, 1H), 4.87–4.90 (t, 2H), 5.61–5.69 (m, 1H), 7.64–7.67 (t, 2H), 7.73–7.76 (t, 1H), 7.92-7.93 (d, J = 8 Hz, 2H). 13C NMR (125 MHz, DMSO-d6): δ 32.53 (–CH2), 38.75 (–CH2), 47.49 (–CH), 56.52 (–CH2), 65.80 (–CH2), 68.89 (–CH), 75.75 (–CH), 78.64 (–CH), 81.20 (–CH), 116.67 (–CH2), 127.78 (2-aryl-CH), 129.38 (2-aryl-CH), 133.83 (–CH), 135.06 (–CH), 139.31 (quaternary C). The NMR spectra are available in the Supporting information .

2.3. Crystallization

Compound (XII) was crystallized by slow evaporation from a solution in methanol, yielding crystals suitable for single-crystal X-ray diffraction analysis. Compound (XV) was recrystallized by slow evaporation from a solution in a mixture of toluene and n-butanol (7:1 v/v), yielding crystals in the form of colourless prisms ideal for single-crystal X-ray diffraction analysis.

2.4. Refinement

Crystal data, data collection and structure refinement de­tails for (XII) and (XV) are summarized in Table 1[link]. H atoms were placed in geometrically optimized positions and con­strained to ride on their parent atoms, with C—H = 0.96 (methyl), 0.97 (methylene) or 0.98 Å (methine) and O—H = 0.82 Å for (XII), and C—H = 0.95 (aryl), 0.99 (methylene) or 1.00 Å (methine) and O—H = 0.84 Å for (XV). Displacement parameters for all H atoms were assigned as Uiso(H) = 1.2Ueq(C,O). The orientation of the solvent methyl group in (XII) was allowed to rotate about the C—O bond to find the best fit to the ob­served electron-density peaks and the hydroxy groups in (XV) were similarly allowed to rotate. For (XII), the hydroxy groups at O3 and O9 were also treated this way, but the hydroxy group at O2 was oriented such that the H atom was in a staggered position with respect to the substituents on the parent C atom.

Table 1
Experimental details

  (XII) (XV)
Crystal data
Chemical formula C19H30O7·CH4O C17H24O6S
Mr 402.48 356.43
Crystal system, space group Orthorhombic, P212121 Orthorhombic, P212121
Temperature (K) 293 200
a, b, c (Å) 7.2240 (18), 12.935 (3), 22.0691 (10) 7.593 (6), 8.542 (7), 26.70 (2)
V (Å3) 2062.2 (7) 1732 (2)
Z 4 4
Radiation type Mo Kα Mo Kα
μ (mm−1) 0.10 0.22
Crystal size (mm) 0.20 × 0.20 × 0.20 0.40 × 0.32 × 0.20
 
Data collection
Diffractometer Rigaku XtaLAB mini diffractometer Rigaku XtaLAB mini diffractometer
Absorption correction Numerical (NUMABS; Rigaku, 1999[Rigaku (1999). NUMABS. Rigaku Corporation, Tokyo, Japan.]) Numerical (NUMABS; Rigaku, 1999[Rigaku (1999). NUMABS. Rigaku Corporation, Tokyo, Japan.])
Tmin, Tmax 0.967, 0.980 0.887, 0.958
No. of measured, independent and observed [F2 > 2.0σ(F2)] reflections 20377, 4735, 4568 14279, 3942, 3733
Rint 0.027 0.034
(sin θ/λ)max (Å−1) 0.649 0.650
 
Refinement
R[F2 > 2σ(F2)], wR(F2), S 0.037, 0.100, 1.08 0.033, 0.082, 1.05
No. of reflections 4735 3942
No. of parameters 256 220
H-atom treatment H-atom parameters constrained H-atom parameters constrained
Δρmax, Δρmin (e Å−3) 0.29, −0.29 0.23, −0.24
Absolute structure Flack x determined using 1853 quotients [(I+) − (I−)]/[(I+) + (I−)] (Parsons et al., 2013[Parsons, S., Flack, H. D. & Wagner, T. (2013). Acta Cryst. B69, 249-259.]) Flack x determined using 1453 quotients [(I+) − (I−)]/[(I+) + (I−)] (Parsons et al., 2013[Parsons, S., Flack, H. D. & Wagner, T. (2013). Acta Cryst. B69, 249-259.])
Absolute structure parameter 0.0 (3) 0.00 (3)
Computer programs: CrystalClear-SM Expert (Rigaku, 2014[Rigaku (2014). CrystalClear-SM Expert. Rigaku Corporation, Tokyo, Japan.]), SIR92 (Altomare et al., 1994[Altomare, A., Cascarano, G., Giacovazzo, C., Guagliardi, A., Burla, M. C., Polidori, G. & Camalli, M. (1994). J. Appl. Cryst. 27, 435.]), SHELXL2014 (Sheldrick, 2015[Sheldrick, G. M. (2015). Acta Cryst. C71, 3-8.]) and CrystalStructure (Rigaku, 2015[Rigaku (2015). CrystalStructure. Rigaku Corporation, Tokyo, Japan.]).

For compound (XII), the chosen absolute configuration of the model was based on known stereochemistry as discussed below, whereas for (XV), the absolute stereochemistry was deduced independently from Parson's z parameter [0.00 (3); Parsons et al., 2013[Parsons, S., Flack, H. D. & Wagner, T. (2013). Acta Cryst. B69, 249-259.]].

3. Results and discussion

The single-crystal X-ray structure of (XII) (Fig. 4[link]) shows that the compound crystallizes in a 1:1 ratio with methanol solvent mol­ecules. The cyclo­hexyl­idene rings assume chair conformations, whereas the five-membered rings have envelope conformations. The mol­ecules pack together in the solid state with a few inter­molecular hydrogen bonds. The O2 and O3 hy­droxy groups form hydrogen-bonding inter­actions with methanol atom O8i and with the O6ii atom of another mol­ecule, respectively (Table 2[link]). These interactions link the constituents into extended chains which run parallel to the [100] direction.

Table 2
Hydrogen-bond geometry (Å, °) for (XII)[link]

D—H⋯A D—H H⋯A D⋯A D—H⋯A
O2—H21⋯O8i 0.82 1.98 2.653 (3) 139
O3—H22⋯O6ii 0.82 2.02 2.829 (2) 171
O8—H23⋯O3i 0.82 1.92 2.727 (3) 169
Symmetry codes: (i) [x-{\script{1\over 2}}, -y+{\script{3\over 2}}, -z+1]; (ii) [x+{\script{1\over 2}}, -y+{\script{3\over 2}}, -z+1].

The NOESY spectrum of (XII) (Fig. 5[link]) shows that the H atom at C1 (H1, 4.8 p.p.m.) has an inter­action with H2, which was unexpected from mechanistic considerations. The single-crystal X-ray diffraction data, however, confirm the mechanistic rationale wherein atom H1 lies on the opposite side of the ring plane from the other H atoms (those on on C2, C3, C4 and C5) of the tetra­hydro­pyran ring. Furthermore, the H1⋯H2 distance of 2.83 Å supports a NOESY inter­action between these atoms. The inter­atomic distance between atoms C1 and C2 is 1.523 (3) Å. Also, atoms H1 and H5 are on opposite sides of the C1—O1—C5 unit, with an H1⋯H5 inter­atomic distance of 3.55 Å. The inter­atomic distance between atoms C1 and H5 is 2.68 Å, and that between atoms C5 and H1 is 3.12 Å.

We further focused our attention on the starting material, i.e. D-gulono-γ-lactone, which has a specific optical rotation (SOR) of −55°. The stereochemistry at atoms C3 and C4 in this mol­ecule are fixed and this is carried through onto atoms C3 and C4 in (XII). Because the known stereochemistry at atoms C3 and C4 remains fixed during the synthesis, the crystal structure allows the configuration of all stereocentres in (XII) to be established relative to those. We were thus able to establish the absolute configuration of (XII) in this manner with the aid of both the NOESY and the single-crystal X-ray data.

The single-crystal X-ray analysis of (XV) shows that the tetra­hydro­furan ring has an envelope conformation (Fig. 7[link]). Again the mol­ecules pack together in the solid state with few inter- or intra­molecular hydrogen bonds. The O2 and O3 hy­droxy groups form an intra­molecular hydrogen-bonding inter­action, and also form inter­molecular hydrogen-bonding inter­actions with atoms O1i and O4ii of other mol­ecules (Table 3[link]). Also, atom O4 forms an inter­molecular hydrogen-bonding inter­action with atom O2iii of another mol­ecule. These interactions link the molecules into two-dimensional networks, which lie parallel to the (001) plane.

Table 3
Hydrogen-bond geometry (Å, °) for (XV)[link]

D—H⋯A D—H H⋯A D⋯A D—H⋯A
O2—H18⋯O1i 0.84 2.05 2.859 (3) 161
O2—H18⋯O3 0.84 2.46 2.817 (3) 107
O3—H19⋯O4ii 0.84 1.98 2.813 (3) 172
O4—H20⋯O2iii 0.84 1.89 2.719 (3) 171
Symmetry codes: (i) [-x, y-{\script{1\over 2}}, -z+{\script{3\over 2}}]; (ii) [-x+1, y-{\script{1\over 2}}, -z+{\script{3\over 2}}]; (iii) x+1, y, z.
[Figure 7]
Figure 7
Single-crystal X-ray view of (XV), showing atoms H2 and H3 in a perpendicular orientation to each other. Displacement ellipsoids are drawn at the 50% probability level.

We were particularly inter­ested to see the orientation of the vicinal H atoms on atoms C2 and C3 of the tetra­hydro­furan ring, as these two H atoms show an unusually weak inter­action in the COSY spectrum (Fig. 8[link]; H2 at 4 p.p.m. and H3 at 2 p.p.m.). This can happen when the C—H bonds form a 90° dihedral angle. The Karplus relationship is based on the observation that vicinal H–H couplings will be at a maximum for H atoms with dihedral angles of 180 and 0° (an anti or eclipsed relationship results in optimal orbital overlap) and that coupling (3JHH) will be minimal (near 0°) for H atoms that are 90° from each other. Atoms H2 and H3 exhibit minimal J coupling, suggesting that their C—H bonds are nearly perpendicular to one another. The tetrahydrofuran ring with its envelope conformation facilitates this orientation of the vicinal H atoms. This conclusion is consistent with the molecular structure elucidated from the single-crystal X-ray diffraction data, which shows that the H2—C2—C3—H3 torsion angle is −110°. The H2⋯H3 interatomic distance is 2.71 Å.

[Figure 8]
Figure 8
The 2D-COSY spectrum of (XV).

In conclusion, we have synthesized two compounds, (XII) and (XV), which have important structural implications for the Eribulin framework. We deduced the relative molecular framework of both structures through 2D-NMR experiments and conclusively proved the absolute structure of both these structures using single-crystal X-ray diffraction techniques.

Supporting information

CCDC references: 1438133; 1438132

Crystal structure: contains datablocks global, Comp_XII, Comp_XV. DOI: 10.1107/S2053229615022305/fn3209sup1.cif

Supporting information file. DOI: 10.1107/S2053229615022305/fn3209sup4.pdf

Structure factors: contains datablock Comp_XII. DOI: 10.1107/S2053229615022305/fn3209Comp_XIIsup5.hkl

Structure factors: contains datablock Comp_XV. DOI: 10.1107/S2053229615022305/fn3209Comp_XVsup6.hkl


Introduction top

Eribulin mesylate is a structurally truncated synthetic analogue of Halichondrin B (Hirata & Uemura, 1986; Kishi et al., 1994), the most bioactive natural product isolated from the marine sponge Halichondria okadai commonly found off the coasts of Japan and New Zealand. Eribulin mesylate (Halaven Reg) inter­feres with microtubule dynamics (Smith et al., 2010) and was approved in November 2010 by the United States Food and Drug Administration (USFDA) for the treatment of metastatic breast cancer (Towle et al., 2001). The structure of eribulin resembles a macrocyclic ketone showcasing a fully synthetic drug available on the market today. 19 of the 36 C atoms that constitute the skeleton of the molecule are stereogenic in nature.

The structural framework of eribulin is built up by assembling three key fragments: (i) the C1–C13 aldehyde fragment I, denoted (III), (ii) the C14–C26 vinyl triflate fragment II, denoted (IV), and (iii) the C27–C35 phenyl sulfone fragment III, denoted (V) (Fig. 1).

Several synthetic routes have been used for the preparation of eribulin, (I), each of which utilize the same strategy described by Kishi and co-workers known for their pioneering work on the total synthesis of Halichondrins, and Halichondrin B in particular (Aicher et al., 1992). Over the years, synthetic routes have evolved continuously, with scale-up and route refinement as the key areas of improvement (Yu et al., 2013; Austad et al., 2013).

The three key fragments (III)–(V) (Fig. 1) are synthesized in 15–20 stages each and have been reported in detail. In the later stages, (IV) and (V) are subjected to Nozaki–Hiyama–Kishi (NHK) coupling (Hiyama et al., 1981; Jin et al., 1986) (Fig. 1) to yield fragment IV [denoted (II), the C14–C35 fragment], which is then coupled with (III) under basic conditions. The final stages (Fig. 2) involve a macrocyclization achieved again through NHK coupling of (VIII), the formation of cyclic ketal (IX) and the conversion of terminal diol (X) to an in situ epoxide on which the primary amine is added to finally realize eribulin mesylate, (I).

The goal of our synthesis was to conduct feasibility studies and develop an optimized and scalable process for the preparation of (I). Owing to the presence of 19 stereocentres in this molecule, it was required to continuously ensure the correct stereoisomer at every stage of this extensive synthesis. The important check-point was to characterize newly formed stereocentres in selected molecules through two-dimensional NMR (2D-NMR) data. There were, however, instances where 2D-NMR data were not conclusive enough to prove the absolute stereochemistry or there was a need to prove the molecular geometry for subsequent reactions. Hence, single-crystal X-ray diffraction analysis was employed to identified inter­mediates of compounds (III)–(V) in order to conclusively establish the structures (Fig. 1).

The single-crystal X-ray structure of (II) has been reported previously (Austad et al., 2013). Also a 3,5-di­nitro­phenyl ester derivative of the alcohol variant of (V) (Fig. 1) has been reported, denoted (Va) (Yang et al., 2009) (Scheme 1).

We report here the single-crystal X-ray data for two relatively important inter­mediates of (III) and (V) which have an important bearing on the structural orchestration of the eribulin molecule.

The selected inter­mediate of (III) is 3,4:6,7-di-O-cyclo­hexyl­idene-D-glycero-α-L-talo-hepto­pyran­ose, (XII), whose stereochemistry has important implications for the C-allyl­ation product (XIV) two stages later (Fig. 3). The synthesis of (III) starts with D-gulono-γ-lactone (Fig. 3) having two sets of vicinal diol groups, the stereochemistries of which are fixed. The gulono lactone is converted into its lactol which reacts with MeOCH2PPh3Cl under Wittig conditions to produce hy­droxy­alkene (XI). The Sharpless di­hydroxy­lation protocol through which (XII) is finally obtained involves a catalytic osmylation (Kolb et al., 1994) of (XI) using the double stereo-differentiation strategy (Masamme et al., 1985). It provides a 4:1 mixture of diastereoisomers, which after purification delivers (XII) as a single diastereoisomer in the form of a white crystalline solid. This mechanism should force the H atom on C1 (i.e. H1) to assume a different orientation on the desired diastereoisomer, as compared to the other H atoms on atoms C2–C5 (Fig. 4). However, the nuclear Overhauser effect spectroscopy (NOESY) of this molecule unusually showed that atom H1 has an inter­action with H2, which means that they are in the same plane (Fig. 5). Further insight into this molecule could be provided by single-crystal X-ray diffraction to align the mechanistic rationale and determine the absolute configuration. Compound (XII) is then subjected to acetyl­ation yielding tetra­acetate (XIII). C-Allyl­ation of (XIII) using methyl 3-tri­methyl­silylpent-4-enoate (also known as the Kishi alkene) retains the configuration on C1 after displacement of the O-acetyl group to yield allylic ester (XIV) which exists as 1:1 conformers (Stamos & Kishi, 1996).

The second inter­mediate, (2R,3R,4R,5S)-5-allyl­tetra­hydro-2-[(S)-2,3-di­hydroxy­propyl]-4-[(phenyl­sulfonyl)­methyl]­furan-3-ol, (XV), from the synthesis of (V) is obtained through a sequence of 15 chemical transformations (Fig. 6) starting from D-glucorono-6,3-lactone (Yang et al., 2009) [see (Va) in Scheme 1]. The correlation NMR spectra (COSY) of the tetra­hydro­furan framework present in (V) and its inter­mediates show that the vicinal H atoms on C2 and C3 do not correlate. It became vital to understand the spatial orientation of these two H atoms along with the structural framework so as to deduce the relative configuration of this molecule. Further, the C6—C7 diol functionality is introduced through asymmetric di­hydroxy­lation in a 3:1 diastereoisomeric ratio in (XX). The undesired isomer after two stages at (XXI) is removed through recrystallization. Our attempts to prepare single crystal of (XXI) were unsuccessful. We then went five stages further to triol (XV) (Fig. 4) which exists as a white crystalline solid and was successful crystallized.

Experimental top

General considerations top

Compounds (XII) and (XV) were prepared according to existing synthesis routes. NMR spectra were recorded at room temperature using an Agilent 500 MHz spectrometer. 1H chemical shifts are reported in p.p.m. values referenced to tetra­methyl­silane (TMS, 0.0 p.p.m.). 13C chemical shifts are reported in p.p.m. referenced to the solvent resonance of 39.5 p.p.m. for DMSO-d6.

NMR data top

1H NMR and 13C NMR data for (XII) top

1H NMR (500 MHz, DMSO-d6): δ 1.30–1.35 (m, 4H), 1.42–1.60 (m, 16H), 3.43–3.44 (dd, J = 3 Hz, 1.5 Hz, 1H), 3.56–3.58 (dd, J = 2, 1.5 Hz, 1H), 3.66–3.69 (t, 1H), 3.94–3.97 (q, 1H), 4.00–4.05 (q, 1H), 4.14–4.16 (dd, J = 2.5, 1.5 Hz, 1H), 4.29–4.32 (dd, J = 2.5, 3 Hz, 1H), 4.79–4.80 (d, J = 6 Hz, 1H), 5.23–5.24 (d, 1 H), 6.34–6.35 (d, 1H).

13C NMR (125 MHz, DMSO-d6): δ 23.38 (–CH2), 23.44 (–CH2), 23.50 (–CH2), 23.62 (–CH2), 24.63 (–CH2), 24.66 (–CH2), 34.60 (–CH2), 34.98 (–CH2), 35.47 (–CH2), 35.86 (–CH2), 64.78 (–CH2), 68.87 (–CH), 70.93 (–CH), 74.20 (–CH), 74.85 (–CH), 75.65 (–CH), 93.59 (–CH), 108.84 (quaternary C), 109.48 (quaternary C).

1H NMR and 13C NMR data for (XV) top

1H NMR (500 MHz, DMSO-d6): δ 1.51–1.53 (m, 1H), 1.69–1.71 (m, 1H), 2.00–2.01 (t, 1H), 2.22–2.54 (m, 2H), 3.25–3.28 (m, 2H), 3.31–3.32 (d, 1H), 3.43–3.47 (m, 2H), 3.53–3.54 (q, 1H), 3.72–3.73 (q, 1H), 3.95–3.97 (t, 1H), 4.39–4.42 (m, 2H), 4.80–4.81 (d, J = 5 Hz, 1H), 4.87–4.90 (t, 2H), 5.61–5.69 (m, 1H), 7.64–7.67 (t, 2H), 7.73–7.76 (t, 1H), 7.92-7.93 (d, J = 8 Hz, 2H).

13C NMR (125 MHz, DMSO-d6): δ 32.53 (–CH2), 38.75 (–CH2), 47.49 (–CH), 56.52 (–CH2), 65.80 (–CH2), 68.89 (–CH), 75.75 (–CH), 78.64 (–CH), 81.20 (–CH), 116.67 (–CH2), 127.78 (2-aryl-CH), 129.38 (2-aryl-CH), 133.83 (–CH), 135.06 (–CH), 139.31 (quaternary C).

Synthesis and crystallization top

Compound (XII) was crystallized by slow evaporation from a solution in methanol, yielding crystals suitable for single-crystal X-ray diffraction analysis. Compound (XV) is recrystallized by slow evaporation from a solution in a mixture of toluene and n-butanol (7:1 v/v), yielding crystals in the form of colourless prisms ideal for single-crystal X-ray diffraction analysis.

Refinement top

Crystal data, data collection and structure refinement details for (XII) and (XV) are summarized in Table 1. H atoms were placed in geometrically optimized positions and constrained to ride on their parent atoms with C—H = 0.96 (methyl), 0.97 (methyl­ene) and 0.98 Å (methine), and O—H = 0.82 Å for (XII) and C—H = 0.95 (aryl), 0.99 (methyl­ene) and 1.00 Å (methine), and O—H = 0.84 Å for (XV). Displacement parameters for all H atoms were assigned as Uiso(H) = 1.2Ueq(C,O). The orientation of the solvent methyl group in (XII) was allowed to rotate about the C—O bond to find the best fit to the observed electron density peaks and the hy­droxy groups in (XV) were similarly allowed to rotate. For (XII), the hy­droxy groups at O3 and O9 were also treated this way, but the hy­droxy group at O2 was oriented such that the H atom was in a staggered position with respect to the substituents on the parent C atom.

For compound (XII), the relative structure was deduced based on known stereochemistry discussed below, whereas for (XV), the structure was deduced based on the Flack parameter [0.00 (3); Parsons et al., 2013].

Results and discussion top

The single-crystal X-ray structure of (XII) (Fig. 4) shows that the compound crystallizes with a methanol solvent molecule. The cyclo­hexyl­idene rings assume chair conformations, whereas the five-membered rings have envelope conformations. The molecules pack together in the solid state with a few inter­molecular hydrogen bonds. The O2 and O3 hy­droxy groups form hydrogen-bonding inter­actions with the methanol atom O9 and with the O6 atom of another molecule, respectively (Table 2).

The NOESY spectrum of (XII) (Fig. 5) shows that the H atom at C1 (H1A, 4.8 p.p.m.) shows an inter­action with H2B, which was unusual as per mechanistic understanding (Table 3). The single-crystal X-ray diffraction data, however, confirm the mechanistic rationale wherein atom H1A is in a different plane from the other H atoms (those on on C2, C3, C4 and C5) of the tetra­hydro­pyran ring. Furthermore, the H1A···H2B distance of 2.83 Å supports a NOESY inter­action between these atoms. The inter­atomic distance between atoms C1 and C2 is 1.523 (3) Å. Also, atoms H1A and H5A are on opposite sides of the C1—O1—C5 unit, with an H1A···H5A inter­atomic distance of 3.55 Å. The inter­atomic distance between atoms C1 and H5A is 2.68 Å, and that between atoms C5 and H1A is 3.12 Å.

We further focused our attention on the starting material, i.e. D-gulono-lactone, which has a specific optical rotation (SOR) of -55°. The stereochemistry at atoms C3 and C4 in this molecule are fixed and this is carried through onto atoms C3 and C4 in (XII). The NOESY spectrum of (XII) showed strong contours for the inter­action of atoms H2B and H3B, suggesting them to be in the same spatial orientation. Furthermore, though atom H1A showed a weak inter­action with H2B, as suggested by NOESY, this was ruled out as the X-ray structure suggested that H1A is in an opposite plane to H2B. The NOESY inter­action for H1A and H2B must stem from the atoms coming into close proximity with each other (2.8 Å) as discussed previously. We were thus able to establish the absolute configuration of (XII) in this manner with the aid of both the NOESY and the single-crystal X-ray data.

The single-crystal X-ray analysis of (XV) shows that the tetra­hydro­furan ring has an envelope conformation (Fig. 7). Again the molecules pack together in the solid state with few inter- or intra­molecular hydrogen bonds. The O2 and O3 hy­droxy groups form an intra­molecular hydrogen-bonding inter­action and also an inter­molecular hydrogen-bonding inter­action with atoms O1 and O4 of another molecule (Table 4). Also, atom O4 forms an inter­molecular hydrogen-bonding inter­action with atom O2 of another molecule, and atom O3 forms an inter­molecular hydrogen-bonding inter­action with atom O4 of another molecule.

We were particularly inter­ested to see the orientation of the vicinal H atoms on atoms C2 and C3 of the tetra­hydro­furan ring, as these two H atoms show an unusually weak inter­action in the COSY spectrum (Fig. 8; H2A at 4 p.p.m. and H3A at 2 p.p.m.). This can happen when the H atoms form a 90° dihedral angle. The Karplus relationship is based on this observation that vicinal H–H couplings will be at a maximum for H atoms with dihedral angles of 180 and 0° (an anti or eclipsed relationship results in optimal orbital overlap) and that coupling (3JHH) will be minimal (near 0°) for H atoms that are 90° from each other. Atoms H2A and H3A, being in near perpendicular positions with respect to each other, possess minimal J coupling. To add to this, the tetra­hydro­furan ring assumes an envelope confirmation giving rise to the feature observed for the vicinal H atoms. This observation was further substanti­ated with the aid of the single-crystal X-ray diffraction data. The bond angles C2—C3—H3A of 110° and C3—C2—H2A of 111° support the COSY spectrum. The H2A···H3A inter­atomic distance is 2.71 Å (Table 5).

In conclusion, we have synthesized and generated single-crystal X-ray diffraction data for two compounds, (XII) and (XV), having important structural implications for the Eribulin framework. Compound (XII) is the starting point of structural integrity for fragment I, (III) (Fig. 1), and thus needed to be thoroughly characterized before proceeding forward with the synthesis. The 2D-NOESY spectrum deduced that atom H1A inter­acted with atom H2B, which was contrary to the mechanistic observations. The single-crystal X-ray analysis, however, confirmed that this was the case and atom H1A in a different plane to the other H atoms on the tetra­hydro­pyran ring. This strategy delivers the correct framework for onward C-allyl­ation and subsequent cyclization.

Compound (XV) gave a typical observation wherein the vicinal H atoms at C2 and C3 on the tetra­hydro­furan ring showed a weak inter­action in the 2D-COSY spectrum. This observation can result from H atoms exhibiting a near perpendicular dihedral angle with respect to each other. Although 2D-NMR techniques helped in deciphering the relative stereochemistry, there remained ambiguity with respect to atoms H2A and H3A since there was no correlation found among them in the COSY analysis. A single crystal of this compound was thus prepared and its X-ray diffraction data showed that the H atoms at C2 and C3 were in a perpendicular orientation with respect to each other. The absolute configuration of the molecule was also determined. The present study thus proves unambiguously that the two H atoms are positioned on the tetra­hydro­furan ring as depicted in Fig. 7. This observation was helpful in deducing the structural framework of fragment III [see (V) in Fig. 1].

Structure description top

Eribulin mesylate is a structurally truncated synthetic analogue of Halichondrin B (Hirata & Uemura, 1986; Kishi et al., 1994), the most bioactive natural product isolated from the marine sponge Halichondria okadai commonly found off the coasts of Japan and New Zealand. Eribulin mesylate (Halaven Reg) inter­feres with microtubule dynamics (Smith et al., 2010) and was approved in November 2010 by the United States Food and Drug Administration (USFDA) for the treatment of metastatic breast cancer (Towle et al., 2001). The structure of eribulin resembles a macrocyclic ketone showcasing a fully synthetic drug available on the market today. 19 of the 36 C atoms that constitute the skeleton of the molecule are stereogenic in nature.

The structural framework of eribulin is built up by assembling three key fragments: (i) the C1–C13 aldehyde fragment I, denoted (III), (ii) the C14–C26 vinyl triflate fragment II, denoted (IV), and (iii) the C27–C35 phenyl sulfone fragment III, denoted (V) (Fig. 1).

Several synthetic routes have been used for the preparation of eribulin, (I), each of which utilize the same strategy described by Kishi and co-workers known for their pioneering work on the total synthesis of Halichondrins, and Halichondrin B in particular (Aicher et al., 1992). Over the years, synthetic routes have evolved continuously, with scale-up and route refinement as the key areas of improvement (Yu et al., 2013; Austad et al., 2013).

The three key fragments (III)–(V) (Fig. 1) are synthesized in 15–20 stages each and have been reported in detail. In the later stages, (IV) and (V) are subjected to Nozaki–Hiyama–Kishi (NHK) coupling (Hiyama et al., 1981; Jin et al., 1986) (Fig. 1) to yield fragment IV [denoted (II), the C14–C35 fragment], which is then coupled with (III) under basic conditions. The final stages (Fig. 2) involve a macrocyclization achieved again through NHK coupling of (VIII), the formation of cyclic ketal (IX) and the conversion of terminal diol (X) to an in situ epoxide on which the primary amine is added to finally realize eribulin mesylate, (I).

The goal of our synthesis was to conduct feasibility studies and develop an optimized and scalable process for the preparation of (I). Owing to the presence of 19 stereocentres in this molecule, it was required to continuously ensure the correct stereoisomer at every stage of this extensive synthesis. The important check-point was to characterize newly formed stereocentres in selected molecules through two-dimensional NMR (2D-NMR) data. There were, however, instances where 2D-NMR data were not conclusive enough to prove the absolute stereochemistry or there was a need to prove the molecular geometry for subsequent reactions. Hence, single-crystal X-ray diffraction analysis was employed to identified inter­mediates of compounds (III)–(V) in order to conclusively establish the structures (Fig. 1).

The single-crystal X-ray structure of (II) has been reported previously (Austad et al., 2013). Also a 3,5-di­nitro­phenyl ester derivative of the alcohol variant of (V) (Fig. 1) has been reported, denoted (Va) (Yang et al., 2009) (Scheme 1).

We report here the single-crystal X-ray data for two relatively important inter­mediates of (III) and (V) which have an important bearing on the structural orchestration of the eribulin molecule.

The selected inter­mediate of (III) is 3,4:6,7-di-O-cyclo­hexyl­idene-D-glycero-α-L-talo-hepto­pyran­ose, (XII), whose stereochemistry has important implications for the C-allyl­ation product (XIV) two stages later (Fig. 3). The synthesis of (III) starts with D-gulono-γ-lactone (Fig. 3) having two sets of vicinal diol groups, the stereochemistries of which are fixed. The gulono lactone is converted into its lactol which reacts with MeOCH2PPh3Cl under Wittig conditions to produce hy­droxy­alkene (XI). The Sharpless di­hydroxy­lation protocol through which (XII) is finally obtained involves a catalytic osmylation (Kolb et al., 1994) of (XI) using the double stereo-differentiation strategy (Masamme et al., 1985). It provides a 4:1 mixture of diastereoisomers, which after purification delivers (XII) as a single diastereoisomer in the form of a white crystalline solid. This mechanism should force the H atom on C1 (i.e. H1) to assume a different orientation on the desired diastereoisomer, as compared to the other H atoms on atoms C2–C5 (Fig. 4). However, the nuclear Overhauser effect spectroscopy (NOESY) of this molecule unusually showed that atom H1 has an inter­action with H2, which means that they are in the same plane (Fig. 5). Further insight into this molecule could be provided by single-crystal X-ray diffraction to align the mechanistic rationale and determine the absolute configuration. Compound (XII) is then subjected to acetyl­ation yielding tetra­acetate (XIII). C-Allyl­ation of (XIII) using methyl 3-tri­methyl­silylpent-4-enoate (also known as the Kishi alkene) retains the configuration on C1 after displacement of the O-acetyl group to yield allylic ester (XIV) which exists as 1:1 conformers (Stamos & Kishi, 1996).

The second inter­mediate, (2R,3R,4R,5S)-5-allyl­tetra­hydro-2-[(S)-2,3-di­hydroxy­propyl]-4-[(phenyl­sulfonyl)­methyl]­furan-3-ol, (XV), from the synthesis of (V) is obtained through a sequence of 15 chemical transformations (Fig. 6) starting from D-glucorono-6,3-lactone (Yang et al., 2009) [see (Va) in Scheme 1]. The correlation NMR spectra (COSY) of the tetra­hydro­furan framework present in (V) and its inter­mediates show that the vicinal H atoms on C2 and C3 do not correlate. It became vital to understand the spatial orientation of these two H atoms along with the structural framework so as to deduce the relative configuration of this molecule. Further, the C6—C7 diol functionality is introduced through asymmetric di­hydroxy­lation in a 3:1 diastereoisomeric ratio in (XX). The undesired isomer after two stages at (XXI) is removed through recrystallization. Our attempts to prepare single crystal of (XXI) were unsuccessful. We then went five stages further to triol (XV) (Fig. 4) which exists as a white crystalline solid and was successful crystallized.

Compounds (XII) and (XV) were prepared according to existing synthesis routes. NMR spectra were recorded at room temperature using an Agilent 500 MHz spectrometer. 1H chemical shifts are reported in p.p.m. values referenced to tetra­methyl­silane (TMS, 0.0 p.p.m.). 13C chemical shifts are reported in p.p.m. referenced to the solvent resonance of 39.5 p.p.m. for DMSO-d6.

1H NMR (500 MHz, DMSO-d6): δ 1.30–1.35 (m, 4H), 1.42–1.60 (m, 16H), 3.43–3.44 (dd, J = 3 Hz, 1.5 Hz, 1H), 3.56–3.58 (dd, J = 2, 1.5 Hz, 1H), 3.66–3.69 (t, 1H), 3.94–3.97 (q, 1H), 4.00–4.05 (q, 1H), 4.14–4.16 (dd, J = 2.5, 1.5 Hz, 1H), 4.29–4.32 (dd, J = 2.5, 3 Hz, 1H), 4.79–4.80 (d, J = 6 Hz, 1H), 5.23–5.24 (d, 1 H), 6.34–6.35 (d, 1H).

13C NMR (125 MHz, DMSO-d6): δ 23.38 (–CH2), 23.44 (–CH2), 23.50 (–CH2), 23.62 (–CH2), 24.63 (–CH2), 24.66 (–CH2), 34.60 (–CH2), 34.98 (–CH2), 35.47 (–CH2), 35.86 (–CH2), 64.78 (–CH2), 68.87 (–CH), 70.93 (–CH), 74.20 (–CH), 74.85 (–CH), 75.65 (–CH), 93.59 (–CH), 108.84 (quaternary C), 109.48 (quaternary C).

1H NMR (500 MHz, DMSO-d6): δ 1.51–1.53 (m, 1H), 1.69–1.71 (m, 1H), 2.00–2.01 (t, 1H), 2.22–2.54 (m, 2H), 3.25–3.28 (m, 2H), 3.31–3.32 (d, 1H), 3.43–3.47 (m, 2H), 3.53–3.54 (q, 1H), 3.72–3.73 (q, 1H), 3.95–3.97 (t, 1H), 4.39–4.42 (m, 2H), 4.80–4.81 (d, J = 5 Hz, 1H), 4.87–4.90 (t, 2H), 5.61–5.69 (m, 1H), 7.64–7.67 (t, 2H), 7.73–7.76 (t, 1H), 7.92-7.93 (d, J = 8 Hz, 2H).

13C NMR (125 MHz, DMSO-d6): δ 32.53 (–CH2), 38.75 (–CH2), 47.49 (–CH), 56.52 (–CH2), 65.80 (–CH2), 68.89 (–CH), 75.75 (–CH), 78.64 (–CH), 81.20 (–CH), 116.67 (–CH2), 127.78 (2-aryl-CH), 129.38 (2-aryl-CH), 133.83 (–CH), 135.06 (–CH), 139.31 (quaternary C).

The single-crystal X-ray structure of (XII) (Fig. 4) shows that the compound crystallizes with a methanol solvent molecule. The cyclo­hexyl­idene rings assume chair conformations, whereas the five-membered rings have envelope conformations. The molecules pack together in the solid state with a few inter­molecular hydrogen bonds. The O2 and O3 hy­droxy groups form hydrogen-bonding inter­actions with the methanol atom O9 and with the O6 atom of another molecule, respectively (Table 2).

The NOESY spectrum of (XII) (Fig. 5) shows that the H atom at C1 (H1A, 4.8 p.p.m.) shows an inter­action with H2B, which was unusual as per mechanistic understanding (Table 3). The single-crystal X-ray diffraction data, however, confirm the mechanistic rationale wherein atom H1A is in a different plane from the other H atoms (those on on C2, C3, C4 and C5) of the tetra­hydro­pyran ring. Furthermore, the H1A···H2B distance of 2.83 Å supports a NOESY inter­action between these atoms. The inter­atomic distance between atoms C1 and C2 is 1.523 (3) Å. Also, atoms H1A and H5A are on opposite sides of the C1—O1—C5 unit, with an H1A···H5A inter­atomic distance of 3.55 Å. The inter­atomic distance between atoms C1 and H5A is 2.68 Å, and that between atoms C5 and H1A is 3.12 Å.

We further focused our attention on the starting material, i.e. D-gulono-lactone, which has a specific optical rotation (SOR) of -55°. The stereochemistry at atoms C3 and C4 in this molecule are fixed and this is carried through onto atoms C3 and C4 in (XII). The NOESY spectrum of (XII) showed strong contours for the inter­action of atoms H2B and H3B, suggesting them to be in the same spatial orientation. Furthermore, though atom H1A showed a weak inter­action with H2B, as suggested by NOESY, this was ruled out as the X-ray structure suggested that H1A is in an opposite plane to H2B. The NOESY inter­action for H1A and H2B must stem from the atoms coming into close proximity with each other (2.8 Å) as discussed previously. We were thus able to establish the absolute configuration of (XII) in this manner with the aid of both the NOESY and the single-crystal X-ray data.

The single-crystal X-ray analysis of (XV) shows that the tetra­hydro­furan ring has an envelope conformation (Fig. 7). Again the molecules pack together in the solid state with few inter- or intra­molecular hydrogen bonds. The O2 and O3 hy­droxy groups form an intra­molecular hydrogen-bonding inter­action and also an inter­molecular hydrogen-bonding inter­action with atoms O1 and O4 of another molecule (Table 4). Also, atom O4 forms an inter­molecular hydrogen-bonding inter­action with atom O2 of another molecule, and atom O3 forms an inter­molecular hydrogen-bonding inter­action with atom O4 of another molecule.

We were particularly inter­ested to see the orientation of the vicinal H atoms on atoms C2 and C3 of the tetra­hydro­furan ring, as these two H atoms show an unusually weak inter­action in the COSY spectrum (Fig. 8; H2A at 4 p.p.m. and H3A at 2 p.p.m.). This can happen when the H atoms form a 90° dihedral angle. The Karplus relationship is based on this observation that vicinal H–H couplings will be at a maximum for H atoms with dihedral angles of 180 and 0° (an anti or eclipsed relationship results in optimal orbital overlap) and that coupling (3JHH) will be minimal (near 0°) for H atoms that are 90° from each other. Atoms H2A and H3A, being in near perpendicular positions with respect to each other, possess minimal J coupling. To add to this, the tetra­hydro­furan ring assumes an envelope confirmation giving rise to the feature observed for the vicinal H atoms. This observation was further substanti­ated with the aid of the single-crystal X-ray diffraction data. The bond angles C2—C3—H3A of 110° and C3—C2—H2A of 111° support the COSY spectrum. The H2A···H3A inter­atomic distance is 2.71 Å (Table 5).

In conclusion, we have synthesized and generated single-crystal X-ray diffraction data for two compounds, (XII) and (XV), having important structural implications for the Eribulin framework. Compound (XII) is the starting point of structural integrity for fragment I, (III) (Fig. 1), and thus needed to be thoroughly characterized before proceeding forward with the synthesis. The 2D-NOESY spectrum deduced that atom H1A inter­acted with atom H2B, which was contrary to the mechanistic observations. The single-crystal X-ray analysis, however, confirmed that this was the case and atom H1A in a different plane to the other H atoms on the tetra­hydro­pyran ring. This strategy delivers the correct framework for onward C-allyl­ation and subsequent cyclization.

Compound (XV) gave a typical observation wherein the vicinal H atoms at C2 and C3 on the tetra­hydro­furan ring showed a weak inter­action in the 2D-COSY spectrum. This observation can result from H atoms exhibiting a near perpendicular dihedral angle with respect to each other. Although 2D-NMR techniques helped in deciphering the relative stereochemistry, there remained ambiguity with respect to atoms H2A and H3A since there was no correlation found among them in the COSY analysis. A single crystal of this compound was thus prepared and its X-ray diffraction data showed that the H atoms at C2 and C3 were in a perpendicular orientation with respect to each other. The absolute configuration of the molecule was also determined. The present study thus proves unambiguously that the two H atoms are positioned on the tetra­hydro­furan ring as depicted in Fig. 7. This observation was helpful in deducing the structural framework of fragment III [see (V) in Fig. 1].

Synthesis and crystallization top

Compound (XII) was crystallized by slow evaporation from a solution in methanol, yielding crystals suitable for single-crystal X-ray diffraction analysis. Compound (XV) is recrystallized by slow evaporation from a solution in a mixture of toluene and n-butanol (7:1 v/v), yielding crystals in the form of colourless prisms ideal for single-crystal X-ray diffraction analysis.

Refinement details top

Crystal data, data collection and structure refinement details for (XII) and (XV) are summarized in Table 1. H atoms were placed in geometrically optimized positions and constrained to ride on their parent atoms with C—H = 0.96 (methyl), 0.97 (methyl­ene) and 0.98 Å (methine), and O—H = 0.82 Å for (XII) and C—H = 0.95 (aryl), 0.99 (methyl­ene) and 1.00 Å (methine), and O—H = 0.84 Å for (XV). Displacement parameters for all H atoms were assigned as Uiso(H) = 1.2Ueq(C,O). The orientation of the solvent methyl group in (XII) was allowed to rotate about the C—O bond to find the best fit to the observed electron density peaks and the hy­droxy groups in (XV) were similarly allowed to rotate. For (XII), the hy­droxy groups at O3 and O9 were also treated this way, but the hy­droxy group at O2 was oriented such that the H atom was in a staggered position with respect to the substituents on the parent C atom.

For compound (XII), the relative structure was deduced based on known stereochemistry discussed below, whereas for (XV), the structure was deduced based on the Flack parameter [0.00 (3); Parsons et al., 2013].

Computing details top

For both compounds, data collection: CrystalClear-SM Expert (Rigaku, 2014); cell refinement: CrystalClear-SM Expert (Rigaku, 2014); data reduction: CrystalClear-SM Expert (Rigaku, 2014); program(s) used to solve structure: SIR92 (Altomare et al., 1994); program(s) used to refine structure: SHELXL2014 (Sheldrick, 2015); molecular graphics: CrystalStructure (Rigaku, 2015); software used to prepare material for publication: CrystalStructure (Rigaku, 2015).

Figures top
[Figure 1] Fig. 1. Synthetic scheme for the coupling of fragments III and IV, and the construction of the eribulin molecule.
[Figure 2] Fig. 2. The final stages in the synthesis of eribulin mesylate, (I).
[Figure 3] Fig. 3. An overview of the synthetic scheme for the preparation of (III), the precursor [which compound is this specifically?] and the product (XII).
[Figure 4] Fig. 4. The molecular structure of (XII), showing atom H1A in a different plane to the other H atoms of the cyclohexylidene ring. [Please provide a new version with a white background and no overlapping atoms labels. Also, the plot should be an ellipsoid plot]
[Figure 5] Fig. 5. The 2D-NOESY spectrum of (XII).
[Figure 6] Fig. 6. An overview of the synthetic scheme for the preparation of (V).
[Figure 7] Fig. 7. Single-crystal X-ray view of (XV), showing atoms H2 and H3 in a perpendicular orientation to each other. [Please provide a new version with a white background and no overlapping atoms labels. Also, the plot should be an ellipsoid plot]
[Figure 8] Fig. 8. The 2D-COSY spectrum of (XV).
(Comp_XII) 3,4:6,7-Di-O-cyclohexylidene-D-glycero-α-L-talo-heptopyranose methanol monosolvate top
Crystal data top
C19H30O7·CH4ODx = 1.296 Mg m−3
Mr = 402.48Mo Kα radiation, λ = 0.71075 Å
Orthorhombic, P212121Cell parameters from 6335 reflections
a = 7.2240 (18) Åθ = 3.2–27.5°
b = 12.935 (3) ŵ = 0.10 mm−1
c = 22.0691 (10) ÅT = 293 K
V = 2062.2 (7) Å3Prism, colorless
Z = 40.20 × 0.20 × 0.20 mm
F(000) = 872.00
Data collection top
Rigaku XtaLAB mini
diffractometer		4568 reflections with F2 > 2.0σ(F2)
Detector resolution: 13.653 pixels mm-1Rint = 0.027
ω scansθmax = 27.5°, θmin = 3.2°
Absorption correction: numerical
(NUMABS; Rigaku, 1999)		h = −9→9
Tmin = 0.967, Tmax = 0.980k = −16→16
20377 measured reflectionsl = −28→28
4735 independent reflections
Refinement top
Refinement on F2Hydrogen site location: inferred from neighbouring sites
R[F2 > 2σ(F2)] = 0.037H-atom parameters constrained
wR(F2) = 0.100 w = 1/[σ2(Fo2) + (0.0514P)2 + 0.2988P]
where P = (Fo2 + 2Fc2)/3
S = 1.08(Δ/σ)max < 0.001
4735 reflectionsΔρmax = 0.29 e Å−3
256 parametersΔρmin = −0.29 e Å−3
0 restraintsAbsolute structure: Flack x determined using 1853 quotients [(I+)-(I-)]/[(I+)+(I-)] (Parsons et al., 2013)
Primary atom site location: structure-invariant direct methodsAbsolute structure parameter: 0.0 (3)
Secondary atom site location: difference Fourier map
Crystal data top
C19H30O7·CH4OV = 2062.2 (7) Å3
Mr = 402.48Z = 4
Orthorhombic, P212121Mo Kα radiation
a = 7.2240 (18) ŵ = 0.10 mm−1
b = 12.935 (3) ÅT = 293 K
c = 22.0691 (10) Å0.20 × 0.20 × 0.20 mm
Data collection top
Rigaku XtaLAB mini
diffractometer		4735 independent reflections
Absorption correction: numerical
(NUMABS; Rigaku, 1999)		4568 reflections with F2 > 2.0σ(F2)
Tmin = 0.967, Tmax = 0.980Rint = 0.027
20377 measured reflections
Refinement top
R[F2 > 2σ(F2)] = 0.037H-atom parameters constrained
wR(F2) = 0.100Δρmax = 0.29 e Å−3
S = 1.08Δρmin = −0.29 e Å−3
4735 reflectionsAbsolute structure: Flack x determined using 1853 quotients [(I+)-(I-)]/[(I+)+(I-)] (Parsons et al., 2013)
256 parametersAbsolute structure parameter: 0.0 (3)
0 restraints
Special details top

Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.

Refinement. Refinement was performed using all reflections. The weighted R-factor (wR) and goodness of fit (S) are based on F2. R-factor (gt) are based on F. The threshold expression of F2 > 2.0 sigma(F2) is used only for calculating R-factor (gt).

Neutral-atom scattering factors were taken from international tables for crystallography, Vol. C, Table 6.1.1.4. Anomalous dispersion effects were included in Fcalc (Ibers & Hamilton, 1964); the values for ?f' and ?f" were extracted from international tables of crystallography (Creagh & McAuley, 1992). The values for the mass attenuation coefficients are those of Creagh and Hubbell.

Creagh, D. C. & Hubbell, J.H..; "International Tables for Crystallography", Vol C, (A.J.C. Wilson, ed.), Kluwer Academic Publishers, Boston, Table 4.2.4.3, pages 200-206 (1992).

Creagh, D. C. & McAuley, W. J. (1992). International Tables for Crystallography, (A. J. C. Wilson, ed.), Kluwer Academic Publishers, Boston, Vol C, Table 4.2.6.8, 219-222.

Ibers, J. A. & Hamilton, W. C. (1964). Acta Cryst. 17, 781–782.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/Ueq
O10.35295 (19)0.58282 (12)0.48548 (6)0.0371 (3)
O20.3077 (2)0.72969 (12)0.42698 (7)0.0421 (3)
H210.20180.71080.41940.050*
O30.6894 (2)0.72168 (11)0.39171 (6)0.0384 (3)
H220.67890.78480.39110.046*
O40.75860 (19)0.51822 (10)0.42770 (6)0.0311 (3)
O50.6597 (2)0.43865 (10)0.51248 (7)0.0382 (3)
O60.19472 (19)0.55968 (11)0.60599 (6)0.0336 (3)
O70.3195 (2)0.46408 (14)0.68318 (7)0.0471 (4)
O80.5386 (3)0.8094 (4)0.65156 (10)0.1275 (16)
H230.42960.80740.64120.153*
C10.4200 (3)0.64349 (14)0.43614 (8)0.0294 (4)
H10.41910.60110.39930.035*
C20.6160 (3)0.68268 (13)0.44707 (8)0.0283 (4)
H20.61110.73920.47660.034*
C30.7420 (2)0.59962 (13)0.47104 (8)0.0273 (4)
H30.86400.62840.48060.033*
C40.6584 (3)0.54600 (13)0.52731 (8)0.0280 (4)
H40.73650.55870.56290.034*
C50.4611 (3)0.58198 (14)0.53966 (8)0.0279 (4)
H50.46470.65200.55650.033*
C60.3620 (2)0.51148 (14)0.58383 (8)0.0287 (4)
H60.33160.44590.56390.034*
C70.4637 (3)0.49056 (17)0.64252 (9)0.0351 (4)
H7A0.55080.43400.63790.042*
H7B0.52970.55150.65620.042*
C80.1523 (3)0.51299 (15)0.66350 (9)0.0331 (4)
C90.0938 (4)0.5960 (2)0.70757 (11)0.0524 (6)
H9A−0.01060.63370.69090.063*
H9B0.19480.64440.71330.063*
C100.0394 (5)0.5495 (2)0.76878 (12)0.0625 (7)
H10A0.14740.51820.78740.075*
H10B−0.00390.60420.79530.075*
C11−0.1116 (4)0.4685 (3)0.76198 (12)0.0617 (7)
H11A−0.13740.43770.80110.074*
H11B−0.22410.50120.74760.074*
C12−0.0527 (4)0.3850 (2)0.71778 (12)0.0530 (6)
H12A−0.15450.33720.71170.064*
H12B0.05020.34660.73480.064*
C130.0050 (3)0.43058 (18)0.65660 (10)0.0418 (5)
H13A0.05190.37570.63090.050*
H13B−0.10260.46030.63690.050*
C140.7801 (3)0.42518 (14)0.46154 (8)0.0305 (4)
C150.7147 (3)0.33586 (16)0.42330 (11)0.0423 (5)
H15A0.59170.35070.40790.051*
H15B0.70710.27420.44820.051*
C160.8454 (4)0.31621 (18)0.37032 (11)0.0475 (5)
H16A0.80460.25540.34840.057*
H16B0.84080.37460.34280.057*
C171.0433 (4)0.3003 (2)0.39150 (13)0.0526 (6)
H17A1.12350.29170.35660.063*
H17B1.05070.23790.41580.063*
C181.1088 (3)0.3921 (2)0.42881 (12)0.0478 (5)
H18A1.11270.45320.40340.057*
H18B1.23300.37890.44360.057*
C190.9804 (3)0.41131 (17)0.48222 (10)0.0384 (4)
H19A1.02040.47280.50370.046*
H19B0.98770.35340.51000.046*
C200.5497 (5)0.8009 (5)0.70985 (16)0.1046 (17)
H20A0.55700.86840.72770.126*
H20B0.65830.76200.72030.126*
H20C0.44190.76580.72490.126*
Atomic displacement parameters (Å2) top
U11U22U33U12U13U23
O10.0301 (6)0.0475 (8)0.0337 (7)−0.0081 (6)−0.0046 (6)0.0125 (6)
O20.0391 (8)0.0433 (8)0.0439 (8)0.0115 (7)−0.0042 (7)0.0044 (6)
O30.0463 (8)0.0315 (6)0.0373 (7)0.0007 (6)0.0088 (7)0.0103 (6)
O40.0377 (7)0.0272 (6)0.0283 (6)0.0048 (5)0.0037 (5)−0.0007 (5)
O50.0444 (8)0.0265 (6)0.0436 (8)0.0037 (6)0.0168 (7)0.0042 (5)
O60.0306 (6)0.0381 (7)0.0321 (7)0.0063 (6)0.0068 (6)0.0102 (5)
O70.0331 (7)0.0700 (11)0.0382 (8)0.0060 (7)0.0031 (6)0.0235 (8)
O80.0373 (10)0.296 (5)0.0495 (12)0.015 (2)0.0035 (10)0.006 (2)
C10.0320 (9)0.0305 (8)0.0257 (8)0.0019 (7)−0.0004 (7)0.0011 (7)
C20.0324 (9)0.0241 (8)0.0284 (8)−0.0005 (7)0.0029 (7)0.0015 (6)
C30.0261 (8)0.0267 (8)0.0291 (8)−0.0006 (7)0.0005 (7)−0.0013 (7)
C40.0295 (8)0.0280 (8)0.0266 (8)0.0014 (7)0.0000 (7)0.0005 (6)
C50.0283 (8)0.0279 (8)0.0273 (8)0.0002 (7)0.0017 (7)0.0012 (7)
C60.0271 (8)0.0297 (8)0.0292 (8)−0.0003 (7)0.0036 (7)0.0022 (7)
C70.0291 (9)0.0412 (10)0.0352 (9)0.0028 (8)0.0016 (8)0.0062 (8)
C80.0313 (9)0.0379 (9)0.0302 (9)0.0032 (8)0.0048 (7)0.0088 (7)
C90.0696 (17)0.0458 (12)0.0418 (12)−0.0032 (12)0.0181 (12)−0.0015 (10)
C100.0811 (19)0.0695 (17)0.0369 (12)−0.0062 (15)0.0208 (13)−0.0041 (11)
C110.0542 (14)0.088 (2)0.0426 (12)−0.0018 (14)0.0173 (12)0.0176 (13)
C120.0490 (13)0.0579 (15)0.0521 (13)−0.0116 (11)0.0022 (11)0.0207 (11)
C130.0395 (11)0.0465 (11)0.0396 (11)−0.0043 (9)0.0002 (9)0.0080 (9)
C140.0331 (9)0.0267 (8)0.0319 (9)0.0023 (7)0.0051 (7)0.0025 (7)
C150.0408 (11)0.0328 (10)0.0534 (12)−0.0055 (8)0.0008 (10)−0.0082 (9)
C160.0569 (14)0.0408 (11)0.0450 (12)−0.0035 (11)0.0005 (11)−0.0151 (9)
C170.0536 (14)0.0463 (13)0.0581 (15)0.0107 (11)0.0114 (12)−0.0125 (11)
C180.0335 (10)0.0513 (13)0.0585 (14)0.0060 (9)0.0055 (10)−0.0108 (11)
C190.0373 (10)0.0378 (10)0.0401 (10)0.0094 (8)−0.0041 (9)−0.0040 (8)
C200.0575 (19)0.192 (5)0.064 (2)−0.003 (3)−0.0071 (17)−0.044 (3)
Geometric parameters (Å, º) top
O1—C11.427 (2)C9—H9A0.9700
O1—C51.428 (2)C9—H9B0.9700
O2—C11.394 (2)C10—C111.520 (4)
O2—H210.8200C10—H10A0.9700
O3—C21.424 (2)C10—H10B0.9700
O3—H220.8200C11—C121.517 (4)
O4—C141.425 (2)C11—H11A0.9700
O4—C31.428 (2)C11—H11B0.9700
O5—C41.427 (2)C12—C131.531 (3)
O5—C141.432 (2)C12—H12A0.9700
O6—C81.439 (2)C12—H12B0.9700
O6—C61.445 (2)C13—H13A0.9700
O7—C71.417 (2)C13—H13B0.9700
O7—C81.431 (2)C14—C151.507 (3)
O8—C201.294 (4)C14—C191.528 (3)
O8—H230.8200C15—C161.524 (3)
C1—C21.523 (3)C15—H15A0.9700
C1—H10.9800C15—H15B0.9700
C2—C31.504 (2)C16—C171.518 (4)
C2—H20.9800C16—H16A0.9700
C3—C41.545 (2)C16—H16B0.9700
C3—H30.9800C17—C181.520 (3)
C4—C51.524 (3)C17—H17A0.9700
C4—H40.9800C17—H17B0.9700
C5—C61.515 (2)C18—C191.520 (3)
C5—H50.9800C18—H18A0.9700
C6—C71.513 (3)C18—H18B0.9700
C6—H60.9800C19—H19A0.9700
C7—H7A0.9700C19—H19B0.9700
C7—H7B0.9700C20—H20A0.9600
C8—C91.509 (3)C20—H20B0.9600
C8—C131.514 (3)C20—H20C0.9600
C9—C101.530 (3)
C1—O1—C5117.22 (14)C9—C10—H10A109.3
C1—O2—H21109.5C11—C10—H10B109.3
C2—O3—H22109.5C9—C10—H10B109.3
C14—O4—C3106.31 (13)H10A—C10—H10B108.0
C4—O5—C14107.60 (14)C12—C11—C10110.7 (2)
C8—O6—C6107.20 (13)C12—C11—H11A109.5
C7—O7—C8108.75 (14)C10—C11—H11A109.5
C20—O8—H23109.5C12—C11—H11B109.5
O2—C1—O1110.68 (15)C10—C11—H11B109.5
O2—C1—C2107.32 (16)H11A—C11—H11B108.1
O1—C1—C2112.19 (15)C11—C12—C13111.6 (2)
O2—C1—H1108.9C11—C12—H12A109.3
O1—C1—H1108.9C13—C12—H12A109.3
C2—C1—H1108.9C11—C12—H12B109.3
O3—C2—C3109.24 (15)C13—C12—H12B109.3
O3—C2—C1109.16 (15)H12A—C12—H12B108.0
C3—C2—C1112.36 (15)C8—C13—C12111.97 (19)
O3—C2—H2108.7C8—C13—H13A109.2
C3—C2—H2108.7C12—C13—H13A109.2
C1—C2—H2108.7C8—C13—H13B109.2
O4—C3—C2110.01 (14)C12—C13—H13B109.2
O4—C3—C4103.90 (13)H13A—C13—H13B107.9
C2—C3—C4111.54 (15)O4—C14—O5104.05 (14)
O4—C3—H3110.4O4—C14—C15108.65 (16)
C2—C3—H3110.4O5—C14—C15110.03 (17)
C4—C3—H3110.4O4—C14—C19111.03 (16)
O5—C4—C5110.13 (15)O5—C14—C19110.79 (16)
O5—C4—C3104.48 (14)C15—C14—C19111.98 (17)
C5—C4—C3111.83 (14)C14—C15—C16111.30 (18)
O5—C4—H4110.1C14—C15—H15A109.4
C5—C4—H4110.1C16—C15—H15A109.4
C3—C4—H4110.1C14—C15—H15B109.4
O1—C5—C6106.55 (15)C16—C15—H15B109.4
O1—C5—C4111.36 (14)H15A—C15—H15B108.0
C6—C5—C4111.91 (15)C17—C16—C15111.7 (2)
O1—C5—H5109.0C17—C16—H16A109.3
C6—C5—H5109.0C15—C16—H16A109.3
C4—C5—H5109.0C17—C16—H16B109.3
O6—C6—C7101.16 (14)C15—C16—H16B109.3
O6—C6—C5110.69 (14)H16A—C16—H16B107.9
C7—C6—C5115.41 (16)C16—C17—C18110.74 (19)
O6—C6—H6109.8C16—C17—H17A109.5
C7—C6—H6109.8C18—C17—H17A109.5
C5—C6—H6109.8C16—C17—H17B109.5
O7—C7—C6103.20 (15)C18—C17—H17B109.5
O7—C7—H7A111.1H17A—C17—H17B108.1
C6—C7—H7A111.1C17—C18—C19111.0 (2)
O7—C7—H7B111.1C17—C18—H18A109.4
C6—C7—H7B111.1C19—C18—H18A109.4
H7A—C7—H7B109.1C17—C18—H18B109.4
O7—C8—O6105.87 (14)C19—C18—H18B109.4
O7—C8—C9110.8 (2)H18A—C18—H18B108.0
O6—C8—C9109.25 (17)C18—C19—C14111.43 (18)
O7—C8—C13108.20 (17)C18—C19—H19A109.3
O6—C8—C13110.89 (16)C14—C19—H19A109.3
C9—C8—C13111.64 (18)C18—C19—H19B109.3
C8—C9—C10111.2 (2)C14—C19—H19B109.3
C8—C9—H9A109.4H19A—C19—H19B108.0
C10—C9—H9A109.4O8—C20—H20A109.5
C8—C9—H9B109.4O8—C20—H20B109.5
C10—C9—H9B109.4H20A—C20—H20B109.5
H9A—C9—H9B108.0O8—C20—H20C109.5
C11—C10—C9111.6 (2)H20A—C20—H20C109.5
C11—C10—H10A109.3H20B—C20—H20C109.5
C5—O1—C1—O2−109.98 (18)C7—O7—C8—O66.6 (2)
C5—O1—C1—C29.9 (2)C7—O7—C8—C9−111.7 (2)
O2—C1—C2—O3−71.31 (18)C7—O7—C8—C13125.54 (17)
O1—C1—C2—O3166.90 (15)C6—O6—C8—O717.8 (2)
O2—C1—C2—C3167.35 (14)C6—O6—C8—C9137.22 (19)
O1—C1—C2—C345.6 (2)C6—O6—C8—C13−99.31 (18)
C14—O4—C3—C2−147.13 (15)O7—C8—C9—C10−66.1 (3)
C14—O4—C3—C4−27.62 (17)O6—C8—C9—C10177.7 (2)
O3—C2—C3—O4−58.95 (19)C13—C8—C9—C1054.6 (3)
C1—C2—C3—O462.34 (19)C8—C9—C10—C11−55.9 (3)
O3—C2—C3—C4−173.70 (14)C9—C10—C11—C1255.7 (3)
C1—C2—C3—C4−52.4 (2)C10—C11—C12—C13−54.5 (3)
C14—O5—C4—C5135.78 (16)O7—C8—C13—C1268.4 (2)
C14—O5—C4—C315.54 (19)O6—C8—C13—C12−175.94 (19)
O4—C3—C4—O57.32 (18)C9—C8—C13—C12−53.9 (3)
C2—C3—C4—O5125.79 (16)C11—C12—C13—C853.9 (3)
O4—C3—C4—C5−111.78 (16)C3—O4—C14—O537.86 (18)
C2—C3—C4—C56.7 (2)C3—O4—C14—C15155.06 (16)
C1—O1—C5—C6−179.18 (15)C3—O4—C14—C19−81.37 (18)
C1—O1—C5—C4−56.9 (2)C4—O5—C14—O4−32.95 (19)
O5—C4—C5—O1−69.48 (18)C4—O5—C14—C15−149.19 (17)
C3—C4—C5—O146.22 (19)C4—O5—C14—C1986.43 (18)
O5—C4—C5—C649.67 (19)O4—C14—C15—C1669.4 (2)
C3—C4—C5—C6165.37 (15)O5—C14—C15—C16−177.24 (18)
C8—O6—C6—C7−33.24 (18)C19—C14—C15—C16−53.6 (2)
C8—O6—C6—C5−156.06 (16)C14—C15—C16—C1754.8 (3)
O1—C5—C6—O6−71.94 (18)C15—C16—C17—C18−56.1 (3)
C4—C5—C6—O6166.11 (14)C16—C17—C18—C1956.3 (3)
O1—C5—C6—C7173.96 (15)C17—C18—C19—C14−55.2 (3)
C4—C5—C6—C752.0 (2)O4—C14—C19—C18−67.5 (2)
C8—O7—C7—C6−26.9 (2)O5—C14—C19—C18177.44 (17)
O6—C6—C7—O736.38 (19)C15—C14—C19—C1854.2 (2)
C5—C6—C7—O7155.87 (16)
Hydrogen-bond geometry (Å, º) top
D—H···AD—HH···AD···AD—H···A
O2—H21···O8i0.821.982.653 (3)139
O3—H22···O6ii0.822.022.829 (2)171
O8—H23···O3i0.821.922.727 (3)169
Symmetry codes: (i) x−1/2, −y+3/2, −z+1; (ii) x+1/2, −y+3/2, −z+1.
(Comp_XV) (2R,3R,4R,5S)-5-Allyl-2-[(S)-2,3-dihydroxypropyl]-4-[(phenylsulfonyl)methyl]tetrahydrofuran-3-ol top
Crystal data top
C17H24O6SDx = 1.367 Mg m−3
Mr = 356.43Mo Kα radiation, λ = 0.71075 Å
Orthorhombic, P212121Cell parameters from 4380 reflections
a = 7.593 (6) Åθ = 3.1–27.5°
b = 8.542 (7) ŵ = 0.22 mm−1
c = 26.70 (2) ÅT = 200 K
V = 1732 (2) Å3Prism, colorless
Z = 40.40 × 0.32 × 0.20 mm
F(000) = 760.00
Data collection top
Rigaku XtaLAB mini
diffractometer		3733 reflections with F2 > 2.0σ(F2)
Detector resolution: 13.653 pixels mm-1Rint = 0.034
ω scansθmax = 27.5°, θmin = 3.1°
Absorption correction: numerical
(NUMABS; Rigaku, 1999)		h = −9→9
Tmin = 0.887, Tmax = 0.958k = −11→11
14279 measured reflectionsl = −34→34
3942 independent reflections
Refinement top
Refinement on F2Hydrogen site location: inferred from neighbouring sites
R[F2 > 2σ(F2)] = 0.033H-atom parameters constrained
wR(F2) = 0.082 w = 1/[σ2(Fo2) + (0.0405P)2 + 0.1659P]
where P = (Fo2 + 2Fc2)/3
S = 1.05(Δ/σ)max = 0.002
3942 reflectionsΔρmax = 0.23 e Å−3
220 parametersΔρmin = −0.24 e Å−3
0 restraintsAbsolute structure: Flack x determined using 1454 quotients [(I+)-(I-)]/[(I+)+(I-)] (Parsons et al., 2013)
Primary atom site location: structure-invariant direct methodsAbsolute structure parameter: 0.00 (3)
Secondary atom site location: difference Fourier map
Crystal data top
C17H24O6SV = 1732 (2) Å3
Mr = 356.43Z = 4
Orthorhombic, P212121Mo Kα radiation
a = 7.593 (6) ŵ = 0.22 mm−1
b = 8.542 (7) ÅT = 200 K
c = 26.70 (2) Å0.40 × 0.32 × 0.20 mm
Data collection top
Rigaku XtaLAB mini
diffractometer		3942 independent reflections
Absorption correction: numerical
(NUMABS; Rigaku, 1999)		3733 reflections with F2 > 2.0σ(F2)
Tmin = 0.887, Tmax = 0.958Rint = 0.034
14279 measured reflections
Refinement top
R[F2 > 2σ(F2)] = 0.033H-atom parameters constrained
wR(F2) = 0.082Δρmax = 0.23 e Å−3
S = 1.05Δρmin = −0.24 e Å−3
3942 reflectionsAbsolute structure: Flack x determined using 1454 quotients [(I+)-(I-)]/[(I+)+(I-)] (Parsons et al., 2013)
220 parametersAbsolute structure parameter: 0.00 (3)
0 restraints
Special details top

Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.

Refinement. Refinement was performed using all reflections. The weighted R-factor (wR) and goodness of fit (S) are based on F2. R-factor (gt) are based on F. The threshold expression of F2 > 2.0 sigma(F2) is used only for calculating R-factor (gt).

Neutral-atom scattering factors were taken from international tables for crystallography, Vol. C, Table 6.1.1.4. Anomalous dispersion effects were included in Fcalc (Ibers & Hamilton, 1964); the values for ?f' and ?f" were extracted from international tables of crystallography (Creagh & McAuley, 1992). The values for the mass attenuation coefficients are those of Creagh and Hubbell.

Creagh, D. C. & Hubbell, J.H..; "International Tables for Crystallography", Vol C, (A.J.C. Wilson, ed.), Kluwer Academic Publishers, Boston, Table 4.2.4.3, pages 200-206 (1992).

Creagh, D. C. & McAuley, W. J. (1992). International Tables for Crystallography, (A. J. C. Wilson, ed.), Kluwer Academic Publishers, Boston, Vol C, Table 4.2.6.8, 219-222.

Ibers, J. A. & Hamilton, W. C. (1964). Acta Cryst. 17, 781–782.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/Ueq
S10.75191 (7)0.69616 (6)0.55350 (2)0.03182 (14)
O10.26736 (18)0.69593 (16)0.68646 (5)0.0270 (3)
O2−0.16817 (19)0.34917 (19)0.72260 (6)0.0327 (4)
H18−0.17260.30740.75110.039*
O30.1105 (2)0.3635 (2)0.79208 (5)0.0328 (4)
H190.17520.28370.79230.039*
O40.64460 (18)0.61630 (18)0.70728 (5)0.0284 (3)
H200.70530.53760.71490.034*
O50.8699 (2)0.6311 (2)0.58990 (7)0.0467 (5)
O60.7510 (3)0.6297 (2)0.50392 (6)0.0502 (5)
C10.3456 (3)0.5447 (2)0.68089 (7)0.0241 (4)
H10.28540.48840.65280.029*
C20.5358 (2)0.5797 (2)0.66544 (7)0.0221 (4)
H20.58640.49060.64580.027*
C30.5122 (2)0.7255 (2)0.63198 (7)0.0228 (4)
H30.59960.80750.64180.027*
C40.3228 (3)0.7839 (2)0.64351 (7)0.0251 (4)
H40.24410.75900.61450.030*
C50.3185 (3)0.4530 (3)0.72883 (8)0.0277 (4)
H5A0.38620.50390.75600.033*
H5B0.36640.34620.72430.033*
C60.1256 (3)0.4405 (2)0.74488 (7)0.0247 (4)
H60.07940.54940.74910.030*
C70.0107 (3)0.3587 (3)0.70659 (8)0.0299 (4)
H7A0.01640.41630.67450.036*
H7B0.05650.25170.70080.036*
C80.5323 (3)0.6877 (3)0.57634 (7)0.0280 (4)
H8A0.45940.76210.55690.034*
H8B0.48540.58130.57020.034*
C90.3069 (3)0.9562 (2)0.65609 (8)0.0292 (4)
H9A0.18410.97890.66620.035*
H9B0.38420.98010.68500.035*
C100.3560 (3)1.0602 (3)0.61316 (9)0.0350 (5)
H100.29761.04490.58210.042*
C110.4748 (4)1.1710 (3)0.61602 (10)0.0475 (6)
H11A0.53531.18890.64660.057*
H11B0.50041.23340.58750.057*
C120.7918 (3)0.8995 (3)0.54793 (8)0.0309 (5)
C130.8810 (3)0.9767 (3)0.58547 (10)0.0416 (6)
H130.92280.92170.61400.050*
C140.9085 (4)1.1358 (4)0.58069 (13)0.0564 (7)
H140.96901.19110.60630.068*
C150.8489 (4)1.2146 (4)0.53927 (14)0.0605 (9)
H150.86981.32380.53620.073*
C160.7584 (4)1.1360 (3)0.50181 (12)0.0529 (7)
H160.71671.19150.47330.064*
C170.7289 (3)0.9771 (3)0.50593 (9)0.0386 (5)
H170.66690.92200.48050.046*
Atomic displacement parameters (Å2) top
U11U22U33U12U13U23
S10.0310 (3)0.0329 (3)0.0315 (3)0.0026 (2)0.0104 (2)0.0010 (2)
O10.0278 (7)0.0240 (7)0.0293 (6)0.0051 (7)0.0081 (6)0.0045 (5)
O20.0242 (7)0.0388 (9)0.0351 (8)−0.0017 (6)0.0005 (6)0.0113 (6)
O30.0306 (8)0.0414 (9)0.0263 (7)0.0036 (7)0.0059 (6)0.0081 (6)
O40.0253 (7)0.0307 (8)0.0292 (7)0.0008 (6)−0.0051 (6)0.0030 (6)
O50.0337 (9)0.0484 (10)0.0582 (11)0.0123 (8)0.0058 (8)0.0165 (9)
O60.0644 (11)0.0472 (10)0.0389 (9)−0.0066 (10)0.0241 (10)−0.0119 (7)
C10.0240 (9)0.0223 (10)0.0259 (9)0.0003 (8)0.0020 (7)0.0009 (7)
C20.0217 (9)0.0218 (10)0.0228 (9)0.0013 (7)0.0018 (7)−0.0001 (7)
C30.0211 (8)0.0236 (10)0.0238 (9)−0.0001 (8)0.0012 (7)0.0011 (7)
C40.0225 (9)0.0284 (11)0.0244 (9)0.0016 (8)0.0015 (7)0.0031 (8)
C50.0228 (9)0.0301 (11)0.0301 (10)0.0014 (8)0.0027 (8)0.0074 (8)
C60.0255 (9)0.0225 (10)0.0261 (9)0.0010 (8)0.0037 (8)0.0046 (8)
C70.0261 (9)0.0348 (12)0.0289 (10)−0.0030 (9)0.0041 (8)0.0015 (8)
C80.0250 (9)0.0356 (11)0.0233 (9)−0.0028 (9)0.0017 (7)0.0019 (9)
C90.0298 (10)0.0264 (11)0.0315 (10)0.0042 (8)0.0017 (8)0.0019 (8)
C100.0398 (12)0.0289 (12)0.0364 (11)0.0082 (10)0.0049 (10)0.0042 (9)
C110.0582 (16)0.0312 (13)0.0531 (15)−0.0015 (12)0.0139 (13)0.0030 (11)
C120.0246 (9)0.0372 (12)0.0311 (10)0.0003 (9)0.0066 (8)0.0015 (9)
C130.0338 (12)0.0479 (15)0.0432 (13)−0.0022 (11)−0.0006 (10)−0.0030 (11)
C140.0419 (14)0.0501 (17)0.077 (2)−0.0097 (14)0.0021 (14)−0.0164 (15)
C150.0414 (14)0.0361 (15)0.104 (3)−0.0017 (13)0.0187 (16)0.0052 (16)
C160.0389 (13)0.0508 (16)0.0691 (17)0.0089 (13)0.0140 (14)0.0251 (14)
C170.0317 (12)0.0471 (14)0.0370 (11)0.0027 (11)0.0055 (10)0.0073 (10)
Geometric parameters (Å, º) top
S1—O51.4339 (19)C6—C71.515 (3)
S1—O61.4403 (19)C6—H61.0000
S1—C121.769 (3)C7—H7A0.9900
S1—C81.777 (2)C7—H7B0.9900
O1—C11.430 (3)C8—H8A0.9900
O1—C41.434 (2)C8—H8B0.9900
O2—C71.426 (3)C9—C101.497 (3)
O2—H180.8400C9—H9A0.9900
O3—C61.426 (2)C9—H9B0.9900
O3—H190.8400C10—C111.310 (4)
O4—C21.424 (2)C10—H100.9500
O4—H200.8400C11—H11A0.9500
C1—C51.514 (3)C11—H11B0.9500
C1—C21.531 (3)C12—C131.378 (3)
C1—H11.0000C12—C171.388 (3)
C2—C31.543 (3)C13—C141.381 (4)
C2—H21.0000C13—H130.9500
C3—C81.528 (3)C14—C151.371 (5)
C3—C41.554 (3)C14—H140.9500
C3—H31.0000C15—C161.387 (5)
C4—C91.514 (3)C15—H150.9500
C4—H41.0000C16—C171.380 (4)
C5—C61.530 (3)C16—H160.9500
C5—H5A0.9900C17—H170.9500
C5—H5B0.9900
O5—S1—O6118.24 (13)C7—C6—H6107.6
O5—S1—C12109.31 (11)C5—C6—H6107.6
O6—S1—C12108.07 (11)O2—C7—C6111.86 (17)
O5—S1—C8109.73 (11)O2—C7—H7A109.2
O6—S1—C8107.17 (12)C6—C7—H7A109.2
C12—S1—C8103.27 (10)O2—C7—H7B109.2
C1—O1—C4105.59 (14)C6—C7—H7B109.2
C7—O2—H18109.5H7A—C7—H7B107.9
C6—O3—H19109.5C3—C8—S1114.77 (14)
C2—O4—H20109.5C3—C8—H8A108.6
O1—C1—C5108.84 (16)S1—C8—H8A108.6
O1—C1—C2104.07 (16)C3—C8—H8B108.6
C5—C1—C2117.17 (16)S1—C8—H8B108.6
O1—C1—H1108.8H8A—C8—H8B107.5
C5—C1—H1108.8C10—C9—C4112.76 (18)
C2—C1—H1108.8C10—C9—H9A109.0
O4—C2—C1112.26 (16)C4—C9—H9A109.0
O4—C2—C3110.11 (16)C10—C9—H9B109.0
C1—C2—C3101.79 (15)C4—C9—H9B109.0
O4—C2—H2110.8H9A—C9—H9B107.8
C1—C2—H2110.8C11—C10—C9123.7 (2)
C3—C2—H2110.8C11—C10—H10118.1
C8—C3—C2112.39 (17)C9—C10—H10118.1
C8—C3—C4110.67 (16)C10—C11—H11A120.0
C2—C3—C4104.60 (15)C10—C11—H11B120.0
C8—C3—H3109.7H11A—C11—H11B120.0
C2—C3—H3109.7C13—C12—C17121.9 (2)
C4—C3—H3109.7C13—C12—S1119.54 (19)
O1—C4—C9107.97 (16)C17—C12—S1118.57 (18)
O1—C4—C3105.16 (15)C12—C13—C14118.6 (3)
C9—C4—C3115.47 (17)C12—C13—H13120.7
O1—C4—H4109.3C14—C13—H13120.7
C9—C4—H4109.3C15—C14—C13120.5 (3)
C3—C4—H4109.3C15—C14—H14119.7
C1—C5—C6113.76 (16)C13—C14—H14119.7
C1—C5—H5A108.8C14—C15—C16120.5 (3)
C6—C5—H5A108.8C14—C15—H15119.7
C1—C5—H5B108.8C16—C15—H15119.7
C6—C5—H5B108.8C17—C16—C15119.9 (3)
H5A—C5—H5B107.7C17—C16—H16120.0
O3—C6—C7109.69 (17)C15—C16—H16120.0
O3—C6—C5110.90 (15)C16—C17—C12118.6 (2)
C7—C6—C5113.23 (17)C16—C17—H17120.7
O3—C6—H6107.6C12—C17—H17120.7
C4—O1—C1—C5170.74 (15)C2—C3—C8—S1−87.49 (19)
C4—O1—C1—C245.09 (18)C4—C3—C8—S1155.97 (15)
O1—C1—C2—O481.59 (18)O5—S1—C8—C339.8 (2)
C5—C1—C2—O4−38.6 (2)O6—S1—C8—C3169.40 (16)
O1—C1—C2—C3−36.12 (18)C12—S1—C8—C3−76.62 (18)
C5—C1—C2—C3−156.31 (17)O1—C4—C9—C10−178.13 (17)
O4—C2—C3—C8135.57 (17)C3—C4—C9—C1064.6 (2)
C1—C2—C3—C8−105.19 (17)C4—C9—C10—C11−125.5 (3)
O4—C2—C3—C4−104.31 (17)O5—S1—C12—C13−18.8 (2)
C1—C2—C3—C414.93 (19)O6—S1—C12—C13−148.67 (19)
C1—O1—C4—C9−158.33 (16)C8—S1—C12—C1398.00 (19)
C1—O1—C4—C3−34.53 (19)O5—S1—C12—C17162.38 (17)
C8—C3—C4—O1131.96 (18)O6—S1—C12—C1732.5 (2)
C2—C3—C4—O110.70 (19)C8—S1—C12—C17−80.87 (19)
C8—C3—C4—C9−109.15 (19)C17—C12—C13—C14−0.2 (4)
C2—C3—C4—C9129.58 (18)S1—C12—C13—C14−179.1 (2)
O1—C1—C5—C655.3 (2)C12—C13—C14—C15−0.4 (4)
C2—C1—C5—C6172.89 (17)C13—C14—C15—C160.8 (4)
C1—C5—C6—O3−175.56 (17)C14—C15—C16—C17−0.5 (4)
C1—C5—C6—C760.6 (2)C15—C16—C17—C12−0.1 (4)
O3—C6—C7—O255.6 (2)C13—C12—C17—C160.5 (3)
C5—C6—C7—O2−179.89 (16)S1—C12—C17—C16179.33 (18)
Hydrogen-bond geometry (Å, º) top
D—H···AD—HH···AD···AD—H···A
O2—H18···O1i0.842.052.859 (3)161
O2—H18···O30.842.462.817 (3)107
O3—H19···O4ii0.841.982.813 (3)172
O4—H20···O2iii0.841.892.719 (3)171
Symmetry codes: (i) −x, y−1/2, −z+3/2; (ii) −x+1, y−1/2, −z+3/2; (iii) x+1, y, z.

Experimental details

(Comp_XII)(Comp_XV)
Crystal data
Chemical formulaC19H30O7·CH4OC17H24O6S
Mr402.48356.43
Crystal system, space groupOrthorhombic, P212121Orthorhombic, P212121
Temperature (K)293200
a, b, c (Å)7.2240 (18), 12.935 (3), 22.0691 (10)7.593 (6), 8.542 (7), 26.70 (2)
V (Å3)2062.2 (7)1732 (2)
Z44
Radiation typeMo KαMo Kα
µ (mm−1)0.100.22
Crystal size (mm)0.20 × 0.20 × 0.200.40 × 0.32 × 0.20
Data collection
DiffractometerRigaku XtaLAB miniRigaku XtaLAB mini
Absorption correctionNumerical
(NUMABS; Rigaku, 1999)		Numerical
(NUMABS; Rigaku, 1999)
Tmin, Tmax0.967, 0.9800.887, 0.958
No. of measured, independent and
observed [F2 > 2.0σ(F2)] reflections		20377, 4735, 4568 14279, 3942, 3733
Rint0.0270.034
(sin θ/λ)max (Å−1)0.6490.650
Refinement
R[F2 > 2σ(F2)], wR(F2), S 0.037, 0.100, 1.08 0.033, 0.082, 1.05
No. of reflections47353942
No. of parameters256220
H-atom treatmentH-atom parameters constrainedH-atom parameters constrained
Δρmax, Δρmin (e Å−3)0.29, −0.290.23, −0.24
Absolute structureFlack x determined using 1853 quotients [(I+)-(I-)]/[(I+)+(I-)] (Parsons et al., 2013)Flack x determined using 1454 quotients [(I+)-(I-)]/[(I+)+(I-)] (Parsons et al., 2013)
Absolute structure parameter0.0 (3)0.00 (3)

Computer programs: CrystalClear-SM Expert (Rigaku, 2014), SIR92 (Altomare et al., 1994), SHELXL2014 (Sheldrick, 2015), CrystalStructure (Rigaku, 2015).

 

Acknowledgements

We wish to thank our integrated product development (IPD) leadership team comprising of Ms Geena Malhotra, Mr Dharmaraj Rao and Dr Manish Gangrade for their continual support and guidance to this project. We also extend our heartfelt thanks to Professor Vedavati Puranik from the National Chemical Laboratory (NCL), Pune, India, for providing important input during the drafting of this paper.

References

First citationAicher, T. D., Buszek, K. R., Fang, F. G., Forsyth, C. J., Jung, S. H., Kishi, Y., Matelich, M. C., Scola, P. M., Spero, D. M. & Yoon, S. K. (1992). J. Am. Chem. Soc. 114, 3162–3164.  CrossRef CAS Web of Science Google Scholar
 First citationAltomare, A., Cascarano, G., Giacovazzo, C., Guagliardi, A., Burla, M. C., Polidori, G. & Camalli, M. (1994). J. Appl. Cryst. 27, 435.  CrossRef Web of Science IUCr Journals Google Scholar
 First citationAustad, B. C., et al. (2013). Synlett, 24, 6327–6332.  Google Scholar
 First citationBrian, C. A., et al. (2013). Synlett, 24, 327–332.  Google Scholar
 First citationCharles, E. C., Francis, G. F., Bryan, M. L., Gordon, D. W., Matthew, J. S. & Xiaojie, Z. (2013). Synlett, 24, 323–326.  Google Scholar
 First citationHirata, Y. & Uemura, D. (1986). Pure Appl. Chem. 58, 701–710.  CrossRef CAS Web of Science Google Scholar
 First citationHiyama, T., Kimura, K. & Nozaki, H. (1981). Tetrahedron Lett. 22, 1037–1040.  CrossRef CAS Web of Science Google Scholar
 First citationJin, H., Uenishi, J., Christ, W. & Kishi, Y. (1986). J. Am. Chem. Soc. 108, 5644–5646.  CrossRef CAS Web of Science Google Scholar
 First citationKishi, Y., Fung, F. G., Forsynth, C. J., Scola, P. M. & Yoon, S. K. (1994). US Patent 5 338 865.  Google Scholar
 First citationKolb, H. C., VanNieuwenhze, M. S. & Sharpless, K. B. (1994). Chem. Rev. 94, 2483–2547.  CrossRef CAS Web of Science Google Scholar
 First citationMasamme, S., Choy, W., Peterson, J. S. & Sita, L. R. (1985). Angew. Chem. 24, 1–76.  Google Scholar
 First citationParsons, S., Flack, H. D. & Wagner, T. (2013). Acta Cryst. B69, 249–259.  Web of Science CrossRef CAS IUCr Journals Google Scholar
 First citationRigaku (1999). NUMABS. Rigaku Corporation, Tokyo, Japan.  Google Scholar
 First citationRigaku (2014). CrystalClear-SM Expert. Rigaku Corporation, Tokyo, Japan.  Google Scholar
 First citationRigaku (2015). CrystalStructure. Rigaku Corporation, Tokyo, Japan.  Google Scholar
 First citationSheldrick, G. M. (2015). Acta Cryst. C71, 3–8.  Web of Science CrossRef IUCr Journals Google Scholar
 First citationSmith, J. A., Wilson, L., Azarenko, O., Zhu, X., Lewis, B. M., Littlefield, B. A. & Jordan, M. A. (2010). Biochemistry, 49, 1331–1337.  Web of Science CrossRef CAS PubMed Google Scholar
 First citationStamos, D. P. & Kishi, Y. (1996). Tetrahedron Lett. 37, 8643–8646.  CrossRef CAS Web of Science Google Scholar
 First citationTowle, M. J., et al. (2001). Cancer Res. 61, 1013–1021.  Web of Science PubMed CAS Google Scholar
 First citationYang, Y. R., Kim, D. S. & Kishi, Y. (2009). Org. Lett. 11, 4516–4519.  Web of Science CSD CrossRef PubMed CAS Google Scholar
 First citationYu, M. J., Zheng, W. & Seletsky, B. M. (2013). Nat. Prod. Rep. 30, 1158–1164.  Web of Science CrossRef CAS PubMed Google Scholar

© International Union of Crystallography. Prior permission is not required to reproduce short quotations, tables and figures from this article, provided the original authors and source are cited. For more information, click here.

Journal logoSTRUCTURAL
CHEMISTRY
ISSN: 2053-2296
Volume 72| Part 1| January 2016| Pages 14-20
doi:10.1107/S2053229615022305
Follow Acta Cryst. C
Sign up for e-alerts
E-alerts
Follow Acta Cryst. on Twitter
Twitter
Follow us on facebook
Facebook
Sign up for RSS feeds
RSS