Vitamin D3: Synthesis of seco C-9,11,21-trisnor-17-methyl-1α, 25-dihydroxyvitamin D3 analogues

Article abstract of DOI:10.1016/S0960-894X(02)00221-4

The synthesis and biological activities of seco C-9,11,21-trisnor-17-methyl-1α,25-dihydroxyvitamin D₃ analogues (D-ring analogues) are described.

Full text of DOI:10.1016/S0960-894X(02)00221-4

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1629–1632
Vitamin D3: Synthesis of seco C-9,11,21-trisnor-17-Methyl-1ꢀ,
25-dihydroxyvitamin D3 Analogues
Yusheng Wu,a Katrien Sabbe,a Pierre De Clercq,a Maurits Vandewalle,a,*
Roger Bouillonb and Annemieke Verstuyfb
aGhent University, Department of Organic Chemistry, Laboratory for Organic Synthesis, Krijgslaan 281 S4, B-9000 Gent, Belgium
bLaboratorium voor Experimentele Geneeskunde en Endocrinologie, K.U. Leuven, Onderwijs en Navorsing Gasthuisberg,
Herestraat 49, B-3030 Leuven, Belgium
Received 14 January 2002; accepted 3 April 2002
Abstract—The synthesis and biological activities of seco C-9,11,21-trisnor-17-methyl-1a,25-dihydroxyvitamin D3 analogues (D-ring
analogues) are described. # 2002 Elsevier Science Ltd. All rights reserved.
The observation that 1a,25-dihydroxy vitamin D3 (1;
calcitriol) is active in the regulation of cell proliferation
and differentiation, next to the classical role in calcium-
bone homeostasis, has led in recent years to the devel-
to see if its absence can be compensated by a 17-methyl
substituent. Also some 19-nor analogues 4 are described,
as this structural feature lowers the toxic calcemic effect
(1 vs 2).7
opment of analogues capable of dissociating cell
1,2
differentiating effects fromcalcemic effects.
Among
(1S,3R)-Camphoric acid 5 is an ideal template for the
central fragment (D-ring). Construction of analogues
will involve (i) chemoselective manipulation of the car-
boxylic functions, (ii) formation of side chains and (iii)
producing the C-8 aldehyde function needed for
coupling with A-ring precursors 6a,8 and 6b (Fig. 1).7a
the three fragments of the vitamin D skeleton especially
structural modifications of the side chain and of the
A-ring have been studied in the past.3
Some years ago, we embarked on an extensive study of
the structure–function relationship with the focus on the
least studied part of the molecule, that is the central
CD-ring regio.4 In this respect we decided stripping the
molecule to its five-carbon backbone (C-8–C-20) and
resubstituting it again in various ways. Presently we
want to describe 21-nor, 17-methyl D-ring analogues
lacking the six-membered C-ring, that is with general
formula 3 and 4. We decided to select a ‘D-ring’ carry-
ing a gem-dimethyl group at C-13 (steroid numbering)
as these substituents mimic respectively the angular C-
18 methyl group and C-12 in the parent steroid 1, which
are known to have an influence on restricting the side
chain orientations.3,5 It is generally assumed that the
relative position in space of the 1a- and 25-hydroxy
groups is important for the biological activity and that
the side chain occupies a very restricted topology at the
binding site of the vitamin D receptor (VDR).3,5,6 This
is also influenced by the 21-methyl group; it is of interest
*Corresponding author. Tel.: +32-9-264-4460; fax: +32-9-264-4998;
e-mail: pierre.declercq@rug.ac.be
Figure 1.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00221-4

1630
Y. Wu et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1629–1632
A highly efficient two-step differentiation of the car-
boxylic functions in 5 was attained via silylation of the
corresponding diol 7 affording 8a (87%) next to the di-
TBDPS ether (5%) (Scheme 1). This method was far
superior to an approach involving the C-8 mono-methyl
ester (97%; MeOH, concd H2SO4, 30 ꢀC) of 5 and
selective reduction of both carboxylic functions. Intro-
duction of the side chain via tosylate 8b was impossible
because all attempted substitutions at the sterically hin-
dered C-20, failed. We therefore turned our attention to
addition reactions on aldehyde precursor 9. Julia coupling9
with the appropriate phenylsulfone (PhSO2CH2R) and
reductive elimination10 led in good yields to the E-dou-
ble bond isomers 10 (for respective side chain moieties
R, see Scheme 2). The 25-hydroxy group in R is pro-
tected as a TES ether. Subsequent deprotection and
oxidation gave the required C-8-formyl precursors 11
(for analogues WY 621, WY 624 and WY 903). For the
synthesis of aldehyde 12 (synthesis of saturated side
chain analogues KS 699, KS 703, WY 619, WY 625,
WY 838 and WY 836) 10 was first hydrogenated.
step sequence (reduction to the corresponding alcohol
and Swern oxidation12). On large scale the direct Dibal-
H reduction was troublesome as it invariably led to
mixtures of 14, 15 and the primary alcohol. Then,
Of special interest are 23-yne vitamin D3 analogues with
a conformationally restricted side chain; they frequently
display enhanced biological properties.3 In the present
context their synthesis was performed via an initial
Horner–Emmons reaction11 on 9. After catalytic
hydrogenation of resulting 13, the ester function in 14
was transformed into an aldehyde function in a two-
Scheme 2. (a) n-BuLi, THF, ꢁ78!ꢁ20 ꢀC, 3 h; (b) TBAF, THF, rt,
12 h.
Scheme 1. (a) LiAlH4, THF–Et2O, rt, 5 h; (b) TBDPSCI, imidazole, DMF, 0 ꢀC!rt, 5 h; (c) TsCl, Et3N, DMAP, CH2Cl2, Á, 12 h; (d) (COCl)2,
DMSO, CH2Cl2, ꢁ78 ꢀC, then Et3N, ꢁ78!0 ꢀC; (e) (i) LDA, THF, ꢁ78 ꢀC, 2 h; (ii) Ac2O, Et3N, DMAP, ꢁ78 ꢀC!rt, 2 h; (iii) Na–Hg, (3.1 or
4.8%), Na2HPO4, MeOH, ꢁ40!rt, 5 h; (f) TBAF, THF, rt, 4 h; (g) (COCl)2, DMSO, CH2Cl2, then Et3N, ꢁ78 ꢀC!ꢁ10 ꢀC, 4 h; (h) H2, 5% Rh/
Al2O3, H2, EtOAc, rt, 2–4 h; (i) (EtO)2P(O)CH2CO2CH3, NaH, THF, 0 ꢀC, 2 h; (j) 4 atm. H2, 10% Pd/C, 3% EtOAc in hexane, rt, 15 h; (k)
DIBALH, CH2Cl2, ꢁ78 ꢀC, 3 h; (l) (MeO)2P(O)CHN2, t-BuOK, THF, ꢁ78!0 ꢀC, 15 h; (m) LDA, R2CO, THF, ꢁ40!10 ꢀC, 3 h.

Y. Wu et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1629–1632
1631
treatment of 15 with dimethyl diazophosphonate affor-
ded alkyne 16.13 The remaining carbon atoms of the
side chain were introduced upon reaction of lithiated 16
with a ketone (Me2CO for WY 722 and WY 906,
(CF3)2CO for WY 718 and CD 578 and Et2CO for WY
727). Deprotection and Swern oxidation then led to C-8
aldehydes 17.
for the more potent fluorinated analogues; especially
WY 718 and CD 578 with a 26,27-hexafluoro-23-yne
side chain. The latter, a 19-nor-analogue, displays high
ratios of cell antiproliferation activities versus calcemic
effect. Further details of the biological activities and
considerations obtained fromcomparison with other
D-ring analogues will be published elsewhere (Table 1).
Finally, construction of the title compounds 3 and 4
involved Lythgoe coupling of aldehydes 11, 12 and 17
with A-ring phosphine oxides 6a8 and 6b7a followed by
deprotection of the hydroxy functions (Scheme 2).
Acknowledgements
We thank the ‘FWO’, the ‘Ministerie voor Wetenschaps-
beleid’ and Theramex S.A. for financial support.
The coupling was performed on D-ring side chain frag-
ments possessing a free 25-hydroxy function therefore
an excess of 6a,b (4–5 equiv) was used; the A-ring
phosphine oxide could be recuperated. The combined
yield of this two-step sequence was higher (65–80%)
than for the alternative approach involving TES
protection (40–50%).
References and Notes
1. Bouillon, R.; Van Baelen, H. Saudi Med. J. 1989, 10, 260.
2. DeLuca, H. F.; Burmester, J.; Darwish, H.; Krisinger, J. In
Comprehensive Medicinal Chemistry; Pergamon: New York,
1990; vol. 3, p 1129.
3. Bouillon, R.; Okamura, W. H.; Norman, A. W. Endocr.
Rev. 1995, 16, 200.
The affinity to the pig intestinal mucosa vitamin D
receptor (VDR) was evaluated as described pre-
viously.14 The relative affinity of the analogues was cal-
culated fromtheir concentration needed to displace
50% of [3H] 1a,25(OH)2D3 fromits receptor compared
with the activity of 1 (assigned a value of 100%). The
affinity for VDR varied between 40 and 80% compared
to 1.
4. (a) Bouillon, R.; De Clercq, P.; Pirson, P.; Vandewalle, M.
Novel structural analogues of vitamin D; patent PCT/EP
93.202037.3; priority date 09–07–1993 (b) Sabbe, K.; D’Halle-
weyn, C.; De Clercq, P.; Vandewalle, M.; Bouillon, R.; Ver-
stuyf, A. Bioorg. Med. Chem. Lett. 1996, 6, 1697. (c) Gui-
Dong, Z.; Yongjun, C.; Xiaoming, Z.; Vandewalle, M.; De
Clercq, P.; Bouillon, R.; Verstuyf, A. Bioorg. Med. Chem.
Lett. 1996, 6, 1703. (d) Yong-Yung, C.; De Clercq, P.; Van-
dewalle, M. Tetrahedron Lett. 1996, 52, 9361. (e) Wu, Y.; Shi,
L.; D’Halleweyn, C.; Van Haver, D.; De Clercq, P.; Vande-
walle, M. Bioorg. Med. Chem. Lett. 1997, 7, 928. (f) Linclau,
B.; De Clercq, P.; Bouillon, R.; Verstuyf, A.; Vandewalle, M.
Bioorg. Med. Chem. Lett. 1997, 11, 1461 and 1465. (g) Sas, B.;
De Clercq, P.; Vandewalle, M. Synlett 1997, 10, 1167. (h)
Verstuyf, A.; Verlinden, L.; Van Baelen, H.; Sabbe, K.;
D’Hallewyn, C.; De Clercq, P.; Vandewalle, M.; Bouillon, R.
J. Bone Miner. Res. 1998, 13, 549. (i) Zhou, X.; Zhu, G.-D.;
Van Haver, D.; Vandewalle, M.; De Clercq, P. J.; Verstuyf,
A.; Bouillon, R. J. Med. Chem. 1999, 42, 3539. (j) Verstuyf,
A.; Verlinden, L.; Van Etten, E.; Shi, L.; Wu, Y.; D’Halle-
weyn, C.; Van Haver, D.; Zhu, G.-D.; Chen, Y.-J.; Zhou, X.;
Haussler, M. R.; De Clercq, P.; Vandewalle, M.; Van Baelen,
H.; Mathieu, C.; Bouillon, R. J. Bone Miner. Res. 2000, 2, 237;
for an analogous approach, see: Kutner, A.; Zhao, H.; Fitak,
H.; Wilson, S. R. Bioorg. Chem. 1995, 23, 22. Hilpert, H.;
Wirz, B. Tetrahedron 2001, 57, 681.
The in vivo calcemic effects were tested in vitamin D-
replete normal NMRI mice by daily intraperitoneal
injections of 1a,25(OH)2D3 1, the analogues or the sol-
vent during 7 consecutive days, using serumcalcium
concentration as parameter. The biological evaluation
was determined in vitro on different cell lines (HL 60,
MCF-7, MG 63, keratinocytes). All results are the mean
of at least 3 experiments and are expressed as percen-
tage activity (at 50% dose response) in comparison with
1 (=100% activity). All analogues were 10 to more than
100-fold less calcemic than 1a,25(OH)2D3. The
antiproliferative activity was comparable to 1, except
Table 1. Biological activities
5. (a) Okamura, W. H.; Midland, M. M.; Hammond, M. W.;
Rahman, N. A.; Dormanen, M. C.; Nemere, I.; Norman,
A. W. J. Steroid Biochem. Mol. Biol. 1995, 53, 603. (b) Van
Haver, D.; De Clercq, P. Bioorg. Med. Chem. Lett. 1998, 8,
1029.
Compd
In vitro studies
In vivo studies
Ca serum
VDR HL60 MG-63 MCF-7 Keratinocytes
1
KS 699
WY 838 40
KS 703
WY 836 75
WY 619 80 150
WY 625 75
WY 621 80
WY 624 40
WY 903 70
WY 722
WY 906 60
100 100
100
60
100
80
100
40
10
90
100
<1
0.25
5
0.25
10
30
40
9
6. Rochel, N.; Wurtz, J. M.; Mitscher, A.; Klaholz, B.;
Moras, D. Mol. Cell 2000, 173.
40
30
80 100
85
200
80
200
50
250
100
1250
90
7. (a) Perlman, K. L.; Sicinski, H. K.; DeLuca, H. F. Tetra-
hedron Lett. 1990, 31, 1823. (b) Bouillon, R.; Sarandeses, L. A.;
Allewaert, K.; Zhao, J.; Mascarenas, J. L.; Mourino, A.;
Vrielynck, S.; De Clercq, P.; Vandewalle, M. Bone Miner. Res.
1993, 8, 1009.
8. Baggiolini, E. G.; Iacobelli, J. A.; Hennessy, B. M.; Batcho,
A. D.; Sereno, J. F.; Uskokovic, M. R. J. Org. Chem. 3098, 51.
9. Julia, M.; Paris, J. M. Tetrahedron Lett. 1973, 49, 4833.
10. Roush, W. R.; Russo-Rodrigues, S. J. Org. Chem. 1985,
50, 5465.
40
750
1350
300
200
60
400
85
3500
4500
90
80
50
20
20
3
13
50
12
375
70
<0.1
0.5
0.1
0.25
1
10
9
100
60
50
350
20
150
70
3500
2000
150
WY 718 80 215
CD 578 100 300
WY 727 70
1
1
90
11. Tanner, D.; Somfai, P. Tetrahedron 1987, 43, 4395.

1632
Y. Wu et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1629–1632
12. Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651.
13. (a) Seyferth, D.; Marmor, R. S.; Hilbert, P. J. Org. Chem.
1971, 36, 1379. (b) Gilbert, J. C.; Weeracoriva, U. J. Org.
Chem. 1979, 44, 4997.
14. Bouillon, R.; Allewaert, K.; Van Leeuwen, J. P. T. M.;
Tan, B. K.; Xiang, D. Z.; De Clercq, P.; Vandewalle, M.; Pols,
H. A. P.; Bos, M. P.; Van Baelen, H.; Birkenhager, J. C. J.
Biol. Chem. 1992, 267, 3044.
¨