Synthesis of 22-oxaspiro[4.5]decane CD-ring modified analogs of 1α,25-dihydroxyvitamin D3

Article abstract of DOI:10.1016/j.tetlet.2009.05.005

In search of analogs of 1α,25-dihydroxyvitamin D₃ featuring a dissociation of calcemic and other activities, a series of stereoisomeric 19-nor-22-oxa derivatives, characterized by a spiro[4.5]decane cyclic system instead of the classical CD-rin

Full text of DOI:10.1016/j.tetlet.2009.05.005

Tetrahedron Letters 50 (2009) 4174–4177
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of 22-oxaspiro[4.5]decane CD-ring modified analogs
of 1a,25-dihydroxyvitamin D3
Frederik De Buysser a, Lieve Verlinden b, Annemieke Verstuyf b, Pierre J. De Clercq a,
*
a Department of Organic Chemistry, Ghent University, Krijgslaan 281 S4, B-9000 Gent, Belgium
b Laboratory of Experimental Medicine and Endocrinology, Catholic University of Leuven, Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
In search of analogs of 1a,25-dihydroxyvitamin D3 featuring a dissociation of calcemic and other activi-
Received 1 April 2009
Revised 28 April 2009
Accepted 4 May 2009
Available online 8 May 2009
ties, a series of stereoisomeric 19-nor-22-oxa derivatives, characterized by a spiro[4.5]decane cyclic sys-
tem instead of the classical CD-ring system, have been synthesized in an enantioselective way.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Spirocyclic decanones
Calcitriol analogs
Enantioselective synthesis
The biological action of vitamin D3 (1, cholecalciferol) originates
from the dihydroxylated metabolite 1 ,25-dihydroxyvitamin D3
stereoisomeric series a and b, on the development of analogs such
as 6, 7, and 8, in which the side chain has been modified aiming at
an increase of biological activity.
a
(2, calcitriol). Next to its classical role in the regulation of calcium
homeostasis, these actions also involve immunomodulation, cell
differentiation, and antiproliferation.1 The search for analogs with
potential therapeutic applications was initiated by the discovery
that in the 22-oxa derivative 3 (OCT) the calcemic and antiprolifer-
ative-prodifferentiating activities were separated (Fig. 1).2 Next to
the 22-oxa modification a few other variations are relevant to the
present work. In this context KH-1060 (4) is exemplary.3 This ana-
log, which is still among the most active analogs discovered so far,
features, next to the 22-oxa modification, also epimerization at
C20, chain elongation, and 25-diethyl instead of dimethyl substitu-
tion. The deletion of C19, which is involved in the reversible [1,7]-
sigmatropic shift responsible for the well-known vitamin-previta-
min equilibration (Fig. 1) is usually accompanied by a reduction in
calcemic activity.4 Whereas the above modifications are located in
the flexible parts of the molecule, our laboratories have been
essentially focusing on structural changes in the central CD-ring
system.5,6
In this context we wish to report here on the synthesis and bio-
logical activity of side chain modified 19-nor-22-oxaspiro[4.5]dec-
ane CF-ring analogs 5–8 wherein the spiro-ring system can be
considered as the formal result of the deletion of C15 and C16,
and of the connection between C18 and C21.7,8 First we will de-
scribe analogs 5 featuring the classical vitamin D side chain in four
stereoisomeric series a, b, c, and d. Next we will report, within the
Figure 1. Structures of the natural vitamin D derivatives 1 and 2, of the 22-oxa
analogs 3 and 4, of the reported spiro[4.5]decane analogs 5–8, and of the vitamin–
previtamin equilibrium.
* Corresponding author. Fax: +32 9 264 49 98.
E-mail address: pierre.declercq@ugent.be (P.J. De Clercq).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.05.005

F. De Buysser et al. / Tetrahedron Letters 50 (2009) 4174–4177
4175
Central in the synthesis of the analogs in the four series a, b, c,
and d stands the preparation of the four enantiopure stereoiso-
meric alcohols 10a–d (Fig. 2), which are correctly functionalized
for the introduction of the flexible parts of the molecule: the side
chain via Williamson ether alkylation and the seco-B, A-ring moi-
ety via Wittig–Horner appendage (vide infra).9 Since the four alco-
hols will be obtained upon reduction of the corresponding
enantiomeric ketones (+)-9 and (ꢀ)-9, an efficient synthetic strat-
egy requires the generation of the stereogenic C13 quaternary cen-
ter in 9 to proceed via a reliable and preferably enantiospecific
route.
The chosen synthetic route is illustrated for the synthesis of (+)-
9 in Scheme 1. It essentially follows work previously described by
Yao and Wang,10 which involves the construction of the chiral spi-
rocenter via two well-established procedures: (i) an asymmetric
conjugate addition on 2-cyclohexenone, and (ii) a Rh(II)-catalyzed
carbene insertion reaction with full retention of absolute configu-
ration.11 Also in our hands did Shibasaki’s enantioselective version
of the Michael addition involving the enolate of dimethyl malonate
with (S)-AlLibis(binapthoxide) complex ((S)-ALB, from lithium alu-
minum hydride and (S)-BINOL) as heterobimetallic chiral catalyst
lead to essentially enantiopure diester 11 in high yield (91%; ee
>99%).12 Acetal formation (to 12), followed by Krapcho demethoxy-
carbonylation (to 13) and saponification gave acid 14 in 82% com-
bined yield.13 Arndt-Eistert homologation of acid 14 was
performed via thermal silver(I) benzoate (dioxane/water) induced
Wolff rearrangement of diazoketone 15,14 prepared via reaction
of the acid chloride with trimethylsilyl diazomethane.15 The ob-
tained acid 16 was further directly converted to diazoketone 17
(oxalyl chloride and trimethylsilyl diazomethane). The latter was
then subjected to a stereospecific C–H insertion reaction involving
the rhodium(II) octanoate dimer as the catalyst to yield ketone (+)-
9 in 50% yield with expected full retention of absolute configura-
tion.16 Obviously, the same sequence but involving (R)-ALB as the
chiral catalyst led to the enantiomer (ꢀ)-9. The enantiopurity of
(+)-9 and (ꢀ)-9 was determined as shown in Scheme 2. Acetaliza-
tion of rac-18 with (R,R)-2,3-butanediol gave the 1:1 diastereo-
meric mixture of acetals 19 and 20 which are nicely separated
via chiral VPC.17 Similar treatment of diketone (ꢀ)-18, obtained
upon hydrolysis of (+)-9, led to bisacetal 19 with better than 99%
de.
The final sequence to the analogs in the four stereoisomeric ser-
ies proceeds via the alcohols 10 that are obtained upon reduction
of the enantiomeric ketones 9 (Fig. 2). Reduction of (+)-9 with so-
dium borohydride gave the epimeric alcohols 10a and 10b in 53%
and 25% isolated yields, respectively. The same treatment of the
enantiomer (ꢀ)-9 gave 10c and 10d in 51% and 28% isolated yields,
respectively.18 The relative configuration of the alcohols in 10a and
10b (and in its enantiomeric pair 10c/10d) follows from 1H NMR
measurements (vide infra).
The synthesis of the first series of analogs 5a–d starting from
the corresponding alcohols 10a–d involves the attachment of the
side chain (Scheme 3) and the appendage of the A-ring (Scheme
4). Both sequences are described for the compounds in a series.
After alkylation of the alkoxide derived from 10a with 3-methyl-
2-butenyl bromide and tetrabutylammonium iodide as catalysts
(to 21a in 95% yield), subsequent mercury(II) assisted water addi-
tion (to alcohol 22a in 90% yield) and acid hydrolysis gave alcohol
23a (82% yield), which was protected as triethylsilyl ether 24a
(86% yield).9 The same uneventful sequence led, starting from
10b, 10c, and 10d to the corresponding cyclohexanones 24b, 24c,
and 24d, respectively (not shown).
The appendage of the 19-nor A-ring to ketone 24a was achieved
using the reliable Wittig–Horner procedure involving the known
phosphine oxide 31.19 After fluoride-induced silyl ether deprotec-
tion a 3:1 mixture of dienes (E)-5a and (Z)-5a is obtained (87%
yield) which is separated by HPLC affording pure (E)-5a in 65% iso-
Figure 2. General strategy for obtention of the stereoisomeric series a, b, c, and d.
Scheme 2. Determination of the enantiomeric purity of 9.
Scheme 1. Synthesis of (+)-9.
Scheme 3. Synthesis of spiroketones 24.

4176
F. De Buysser et al. / Tetrahedron Letters 50 (2009) 4174–4177
Scheme 5. Synthesis of spiroketones 27, 28, and 30.
way 10b was converted to 27b, 28b, and 30b (not shown). It is
important to note here that the detailed analysis of 1H NMR data
obtained for ketones 27a and 27b allows for their stereochemical
assignment, and hence also for the structural determination of
the intermediate alcohols 10a (and enantiomer 10c) and 10b
(and enantiomer 10d). The relevant NOE measurements are shown
in Figure 3.
Scheme 4. Synthesis of analogs 5a–d, 6a and 6b, 7a and 7b, and 8a and 8b.
The further conversion of ketones 27a and 27b, 28a and 28b,
30a and 30b into the corresponding analogs 6a and 6b, 7a and
7b, 8a and 8b proceeds in the same way as was described for the
conversion of 24a into (E)-5a (Scheme 4). The separation of the
propargylic (E)- and (Z)-dienes 7 and 8 however proved more dif-
ficult so that the (E)-analogs remained contaminated with varying
amounts of the (Z)-isomer: 7a (20%), 7b (9%), 8a (4%), 8b (<1%).
As was the case for analogs 5a–d, the six new analogs did not
show any relevant affinity for the VDR; the best result was ob-
tained for 6b with 3% of the calcitriol affinity. The activity in the
inhibition of the MCF-7 cells on the other hand increased with
increasing chain length and was markedly higher in the a series
than in the b series: 2%/0.9%, 10%/0%, and 80%/10% of the activity
of calcitriol were measured for 6a/6b, 7a/7b, and 8a/8b, respec-
tively. These preliminary results are in line with the observation
that there exists a relationship between the position of the 25-hy-
droxy group as determined by the conformation of the side chain
and the prodifferentiating/antiproliferative activity of an analog.22
The latter aspect and other biological results will be covered in a
full account.
lated yield. The identification of the major isomer as the (E)-5a
derivative rests on 1H NMR COSY and 2D NOESY measurements
(Fig. 3). In the same way cyclohexanones 24b, 24c, and 24d led
to mixtures of (E)- and (Z)-derivatives affording after HPLC purifi-
cation analogs 5b (56%), 5c (59%) and 5d (68%), respectively.
The biological evaluation of the analogs includes the determina-
tion of the binding affinity for the porcine intestinal VDR, and the
antiproliferative activity in vitro on breast cancer MCF-7 cells.20
The analogs 5c and 5d did not show any relevant biological activ-
ity. On the other hand the analogs 5a and 5b displayed a modest
activity in the inhibition of the proliferation of MCF-7 cells when
compared with the activity of calcitriol (10% of the calcitriol activ-
ity). On the basis of this observation the further search for active
analogs was restricted to derivatives in the a and b stereoisomeric
series.
The synthesis of the analogs 6a, 7a, and 8a proceeded via ke-
tones 27a, 28a, and 30a, respectively (Scheme 5). In the case of
27a the synthesis involves alkylation of 10a to 25a (60% yield), fol-
lowed by hydrogenation (to 26a) and acetal deprotection to yield
27a (73%). Ketone 28a was directly obtained via copper(II) chlo-
ride-induced hydrolysis of 25a (82%).21 In the case of 30a the alkyl-
ation was performed with the corresponding ethyl-substituted
propargylic chloride (to 29a) followed by hydrolysis. In the same
Acknowledgments
This work was supported by grants G.0553.06 and G.0587.09
(FWO) and the KULeuven Research Council (EF/05/007 Symbiosys).
S. Marcelis, B. K. Tan, and I. Jans are thanked for technical
assistance.
References and notes
1. Bouillon, D. R.; Norman, A. W.; Pasqualini, J. R. (Eds.), Proceedings of the 13th
Workshop on Vitamin D. J. Steroid Biochem. Mol. Biol. 2007, 103, 201–203.
2. Nishii, Y.; Okano, T. Steroids 2001, 66, 137–146.
3. Binderup, L.; Latini, S.; Binderup, E.; Bretting, C.; Calverley, M.; Hansen, K.
Biochem. Pharmacol. 1991, 42, 1569–1575.
4. Perlman, K. L.; Sicinski, R. R.; Schnoes, H. K.; DeLuca, H. F. Tetrahedron Lett.
1990, 31, 1823–1824.
5. Bouillon, R.; Okamura, W. H.; Norman, H. W. Endocrine Rev. 1995, 16, 200–257.
6. De Clercq, P. J.; De Buysser, F.; Minne, G.; Schepens, W.; Vrielynck, F.; Van
Haver, D.; Vandewalle, M.; Verstuyf, A.; Bouillon, R. J. Steroid Biochem. Mol. Biol.
2007, 103, 206–212.
Figure 3. 1H NMR NOESY measurements of (E)-dienes 5–8 and of spirocyclic
ketones 27a and 27b.

F. De Buysser et al. / Tetrahedron Letters 50 (2009) 4174–4177
4177
7. For bridged calcitriol analogues, see: (a) Varela, C.; Nilsson, K.; Torneiro, M.;
Mouriño, A. Helv. Chim. Acta 2002, 85, 3251–3261; (b) Cornella, I.; Pérez Sestelo,
J.; Mouriño, A.; Sarandeses, L. A. J. Org. Chem. 2002, 67, 4707–4714.
8. For previous work, see: (a) Schepens, W.; Van Haver, D.; Vandewalle, M.; De
Clercq, P. J.; Bouillon, R.; Verstuyf, A. Bioorg. Med. Chem. Lett. 2004, 14, 3889–
3892; (b) Schepens, W.; Van Haver, D.; Vandewalle, M.; Bouillon, R.; Verstuyf,
A.; De Clercq, P. J. Org. Lett. 2006, 8, 4247–4250.
9. (a) Dai, H.; Posner, G. H. Synthesis 1994, 1383–1398; (b) Zhu, G.-D.; Okamura,
W. H. Chem. Rev. 1995, 95, 1877–1952.
10. Yao, W.-G.; Wang, J.-B. Chinese J. Org. Chem. 2003, 23, 546–549.
11. (a) Adams, J.; Spero, D. M. Tetrahedron 1991, 47, 1765–1808; (b) Padwa, A.;
Krumpe, K. E. Tetrahedron 1992, 48, 5385–5453; (c) Merlic, C. A.; Zechman, A. L.
Synthesis 2003, 1137–1156.
12. (a) Arai, T.; Sasai, H.; Aoe, K.-i.; Okamura, K.; Date, T.; Shibasaki, M. Angew.
Chem., Int. Ed. 1996, 35, 104–106; (b) Shibasaki, M.; Yoshikaw, N. Chem. Rev.
2002, 102, 2187–2209; (c) Xu, Y.; Ohori, K.; Oshima, T.; Shibasaki, M.
Tetrahedron 2002, 58, 2585–2588.
13. Krapcho, A. P. ARKIVOC 2007, ii, 1–53.
14. Kirmse, W. Eur. J. Org. Chem. 2002, 2193–2256.
17. Yamada, S.; Karasawa, S.; Takahashi, Y.; Aso, M.; Suemune, H. Tetrahedron
1998, 54, 15555–15566.
18. Relevant spectroscopic data for 10a: ½a D
ꢁ
1.59 (c 0.95; CHCl3); 1H NMR (CDCl3,
500 MHz): 1.42–1.63 (12H, m), 1.65–1.69 (1H, m), 1.85–1.94 (1H, m), 1.97 (1H,
dd, J = 6.9, 13.9 Hz), 3.90 (4H, s), 4.33 (1H, m). 13C NMR (75 MHz, CDCl3): 21.17
(CH2), 34.71 (CH2), 35.12 (CH2), 37.31 (CH2), 38.70 (CH2), 43.15 (C), 46.12 (CH),
48.56 (CH2), 64.41 (CH2), 64.44 (CH2), 74.20 (CH), 109.42 (C). Relevant
spectroscopic data for 10b:
½
a D
ꢁ
ꢀ8.8 (c 1; CHCl3); 1H NMR (CDCl3,
500 MHz): 1.28 (2H, m), 1.50 (1H, m), 1.54–1.61 (6H, m), 1.66 (2H, q,
J = 13.6 Hz), 1.67 (1H, m), 1.76 (1H, dt, J = 8.2, 12.9 Hz), 1.84 (1H, dd, J = 6.6,
13.9 Hz), 1.87–1.93 (1H, m), 3.89–3.94 (4H, m), 4.32 (1H, m). 13C NMR
(75 MHz, CDCl3): 20.91 (CH2), 34.40 (CH2), 35.04 (CH2), 37.04 (CH2), 37.45
(CH2), 42.91 (C), 46.39 (CH), 48.43 (CH2), 64.09 (CH2), 73.72 (CH), 109.40 (C).
19. Perlman, K. L.; Swenson, R. E.; Paaren, H. E.; Schnoes, H. K.; DeLuca, H. F.
Tetrahedron Lett. 1991, 32, 7663–7666.
20. Verstuyf, A.; Verlinden, L.; Van Baelen, H.; Sabbe, K.; D’Halleweyn, C.; De
Clercq, P.; Vandewalle, M.; Bouillon, R. J. Bone Miner. Res. 1998, 13, 549–558.
21. Saravanan, P.; Chandrasekhar, M.; Vijaya Anand, R.; Singh, V. K. Tetrahedron
Lett. 1998, 39, 3091–3092.
15. Stock, H. T.; Kellogg, R. M. J. Org. Chem. 1996, 61, 3093–3105.
16. Taber, D. F.; Petty, E. H.; Raman, K. J. Am. Chem. Soc. 1985, 107, 196–199.
22. Yamada, S.; Yamamoto, K.; Masuno, H.; Ohta, M. J. Med. Chem. 1998, 41, 1467–
1475.