Stereoselective synthesis of vitamin D3 analogues with cyclic side chains

Article abstract of DOI:10.1016/S0040-4039(00)01220-X

The synthesis of a new class of vitamin D analogues having a cyclic side chain is described. Readily available starting material, a short sequence with high yielding steps are key features of the synthesis. Alcohols 12c and 12d were readily separable by flash chromatography and the stereochemistry at C22 was established unambiguously by X-ray diffraction analysis of 12c. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01220-X

Tetrahedron Letters 41 (2000) 7323±7326
Stereoselective synthesis of vitamin D3 analogues with cyclic
side chains
Yagamare Fall,a,* Carlos Fernandez,a Cristian Vitaleb and Antonio Mourinob
a
Â
Departamento de Quõmica Fõsica y Quõmica Organica, Facultade de Ciencias, Universidad de Vigo,
Â
Â
Â
36200 Vigo, Spain
b
Â
Departamento de Quõmica Organica y Unidad Asociada al CSIC, Universidad de Santiago de Compostela,
Â
15706 Santiago de Compostela, Spain
Received 31 May 2000; accepted 18 July 2000
Abstract
The synthesis of a new class of vitamin D analogues having a cyclic side chain is described. Readily
available starting material, a short sequence with high yielding steps are key features of the synthesis.
Alcohols 12c and 12d were readily separable by ¯ash chromatography and the stereochemistry at C22 was
established unambiguously by X-ray di€raction analysis of 12c. # 2000 Elsevier Science Ltd. All rights
reserved.
The importance of 1a,25-dihydroxyvitamin D3 [Calcitriol, 1a,25-(OH)2-D3, 1] the hormonally
active metabolite of vitamin D3 is presently well recognized.1 It exhibits control over the expres-
sion of various genes which are involved in calcium and phosphorus metabolism, cellular di€er-
entiation and regulation of the immune system.2 1a,25-(OH)2-D3 is a multifunctional hormone3
which is believed to exert its activities by a mechanism mediated by the nuclear vitamin D
receptor (VDR).4 In the expression of vitamin D function, two proteins play important roles: the
speci®c nuclear receptor protein (vitamin D receptor, VDR) and the transport protein (vitamin D
binding protein, DBP). It is of crucial importance to ®nd out the conformation of 1a,25-(OH)2-
D3 responsible for the binding to those proteins. For that purpose, Yamada et al.5,6 designed the
synthesis of four new vitamin D analogues (2a,2b,3a,3b) with restricted side chain conformation.
* Corresponding author. Fax: 34 986 81 23 82; e-mail: yagamare@uvigo.es
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01220-X

7324
The structure and function studies of these active vitamin D analogues showed that they are
indeed of extreme importance not only as potentially useful therapeutic drugs, but also as a tool
for studying the molecular mechanism of vitamin D mediated gene expression. This prompted us
to release our preliminary results on the synthesis of vitamin D analogues such as 4a and 4b with
restricted side chain conformation, based on the retrosynthetic analysis depicted in Scheme 1.
Scheme 1.
The synthesis of key precursors 8a and 8b bearing a cyclic side chain is detailed in Scheme 2.
The nitrile 6,7 readily obtained from the Inho€en±Lythgoe diol 9 was deprotonated with 2
equivalents of LDA in THF at ^78ꢀC. A solution of the bromide 58 in THF was added and the
mixture allowed to reach room temperature overnight, a€ording the cyanoalcohol 109 (89%) as a
mixture of unseparable diastereoisomers. Reaction of the nitrile 10 with DIBAH in dichloro-
methane at ^10ꢀC and subsequent acid work-up a€orded the corresponding aldehyde which was
taken up in methanol and reacted with excess of sodium borohydride10 to a€ord a 2:1.5 mixture
of alcohols 11c and 11d which were cleanly separated by ¯ash chromatography (15% EtOAc±
hexanes) as colorless oils (40 and 24%, respectively). At this stage of the synthesis the stereo-
chemistry of both compounds at C22 was unknown. Deprotection of the TES group gave the
corresponding triols 12c and 12d in 92 and 90% yield, respectively, as white solids. Recrystallization
of 12c11 from dichloromethane±methanol a€orded crystals which were subjected to X-ray
crystallographic analysis, thus establishing the structure to be that shown in Fig. 1.
Selective tosylation of the primary alcohol of 12c and 12d gave the corresponding tosylates 13
and 14 in 80 and 84% yield, respectively. The tosylates 13 and 14 were individually treated with

7325
Scheme 2. Reagents and conditions: (i) LDA,THF, ^78ꢀC; 5 (89%); (ii) DIBAH,CH2Cl2, ^10ꢀC; HCl; NaBH4, MeOH
(64% global yield 11c+11d, two steps); (iii) nBu4NF, THF, rt; (iv) TsCl, Pyr, 0ꢀC, 12 h; (v) NaH, DMF, rt (61%); (vi)
PDC, CH2Cl2, rt 5 h; (vii) (a) 7; nBuLi, THF, ^78ꢀC; (b) nBu4NF, THF, rt
Figure 1. Molecular structure of 12c
sodium hydride, in DMF at rt to a€ord the corresponding alcohols 15b and 15a bearing a cyclic
side chain. Pyridinium dichromate oxidation of the alcohols 15a and 15b a€orded the ketones 8a
and 8b in 90 and 95% yield, respectively. With these key precursors in hand the stage was set for
the Wittig±Horner reaction using phosphine oxyde 7.12 Coupling reaction of 8a and 8b with 7,

7326
and subsequent removal of the silyl protecting groups, a€orded the corresponding targets 4a13
and 4b14 in 75 and 78% yield, respectively. In conclusion, we have developed a convergent route
to a new class of vitamin D analogues having a cyclic side chain. This approach is also useful for
the preparation of other vitamin D analogues modi®ed at the side chain including Yamada's type
compounds.5 Work is in progress for the synthesis of those analogues.
Acknowledgements
We are grateful to the DGES (Spain, project PM97-0166) for ®nancial support and Solvay
Pharmaceuticals (Weesp, The Netherlands) for the gift of starting materials. Manuel Marcos
(CACTI, Vigo) is gratefully acknowledged for the X-ray structure determination of 12c.
References
1. Bouillon, R.; Okamura, W. H.; Norman, A. W. Endocr. Rev. 1995, 16, 200±257.
2. Feldman, D.; Glorieux, F. H.; Pike, J. W., Vitamin D. Eds.; Academic Press, 1997.
3. Reichel, H.; Koe‚er, H. P.; Norman, A. W. New. Engl. J. Med. 1989, 320, 980±991.
4. (a) Minghetti, P. P.; Norman, A. W. FASEB J. 1988, 2, 3043±3053. (b) Deluca, H. F.; Krisinger, J.; Darwish, H.
Kidney Int. (Suppl.) 1990, 29S, 2±8.
5. Yamamoto, K.; Takahashi, J.; Hamano, K.; Yamada, S.; Yamaguchi, K.; Deluca, H. F. J. Org. Chem. 1993, 58,
2530±2537.
6. Yamamoto, K.; Yan Sun, W.; Ohta, M.; Hamada, K.; Deluca, H. F.; Yamada, S. J. Med. Chem. 1996, 39, 2727±
2737.
7. (a) Leyes, G. A.; Okamura, W. H. J. Am. Chem. Soc. 1982, 104, 6099±6105. (b) Sardina, F. J.; Mourino, A.;
Castedo, L. J. Org. Chem. 1986, 51, 1264±1269. (c) Fall, Y.; Torneiro, M.; Castedo, L.; Mourino, A. Tetrahedron
1997, 53, 4703±4714.
8. Andrews, D. R.; Barton, D. H. R.; Hesse, R. H.; Pechet, M. M. J. Org. Chem. 1986, 51, 4819±4828.
9. All new compounds exhibited satisfactory 1H and 13C NMR, analytical, and/or high resolution mass spectral data.
10. Wovkulich, P. M.; Barcelos, F.; Batcho, A. D.; Sereno, J. F.; Baggiolini, E. G.; Hennessy, B. M.; Uskokovic,
M. R. Tetrahedron 1984, 40, 2283±2296.
11. The X-ray data for 12c will be deposited at the Cambridge Data Centre, Cambridge, UK.
12. Mourino, A.; Torneiro, M.; Vitale, C.; Fernandez, S.; Perez-Sestelo, J.; Anne, S.; Gregorio, C. Tetrahedron Lett.
1997, 38, 4713±4716.
1
13. Compound 4a: H NMR (300 MHz, CD2Cl2), ꢀ: 6.34 and 5.99 (2H, AB, J=11.2, H-6 and 7), 5.27 (1H, br s, H-
19), 4.94 (1H, br s, H-19), 4.36 (1H, m), 4.15 (1H, m), 3.51 (1H, dd, J=11.0, 2.3, H-28), 3.41 (1H, t, J=11.0, H-
28), 2.83 (1H, dd, J=12.0, 3.6), 2.53 (1H, dd, J=12.3, 3.5), 2.24 (1H, dd, J=13.4, 6.7), 1.12 (3H, s, CH3-26), 1.11
(3H, s, CH3-27), 0.87 (3H, d, J=6.5, CH3-21), 0.51 (3H, s, CH3-18); 13C NMR (CD2Cl2), ꢀ: 148.80 (Cˆ), 143.57
(Cˆ), 134.29 (Cˆ), 125.29 (CHˆ), 117.93 (CHˆ), 112.25 (ˆCH2), 71.61 (C-25), 71.56, 67.49, 62.46 (CH2), 57.01,
54.82, 46.63 (C-13), 46.13 (CH2), 43.72 (CH2), 41.32 (CH2), 40.18, 39.40, 37.94 (CH2), 31.99, 29.78 (CH2), 28.15
(CH2), 25.77 (CH2), 24.36 (CH2), 22.94 (CH2), 21.99, 15.12, 12.05. HRMS calcd for C28H44O3: 428.3290; found:
428.3318.
1
14. Compound 4b: H NMR (300 MHz, CD2Cl2), ꢀ: 6.34 and 5.99 (2H, AB, J=11.2, H-6 and 7), 5.27 (1H, br s, H-
19), 4.94 (1H, br s, H-19), 4.35 (1H, m), 4.15 (1H, m), 3.53 (1H, t, J=11.2, H-28), 3.32 (1H, dd, J=11.2, 4.0, H-
28), 2.82 (1H, dd, J=12, 3.3), 2.52 (1H, dd, J=13.3, 3.0), 2.26 (1H, dd, J=13.3, 6.6), 1.13 (3H, s, CH3-26), 1.11
(3H, s, CH3-27), 0.84 (3H, d, J=6.6, CH3-21), 0.53 (3H, s, CH3-18); 13C NMR (CD2Cl2), d: 148.85 (Cˆ), 143.51
(Cˆ), 134.38 (Cˆ), 125.23 (CHˆ), 117.94 (CHˆ), 112.18 (ˆCH2), 71.87 (C-25), 71.47, 67.49, 66.51 (CH2), 57.09,
54.48, 46.49 (C-13), 46.09 (CH2), 43.72 (CH2), 41.40 (CH2), 39.28, 37.58, 37.50 (CH2), 31.98, 29.79 (CH2), 28.17
(CH2), 24.38 (CH2), 22.91 (CH2), 22.02, 19.71 (CH2), 15.44, 12.24. HRMS calcd for C28H44O3: 428.3290; found:
428.3278.