1-Desoxy analog of 2MD: Synthesis and biological activity of (20S)-25-hydroxy-2-methylene-19-norvitamin D3

Article abstract of DOI:10.1016/j.jsbmb.2010.03.063

During our ongoing structure-activity studies in the vitamin D area, we obtained (20S)-1α,25-dihydroxy-2-methylene-19-norvitamin D₃ (5). This analog, designated 2MD, is characterized by a significantly enhanced calcemic activity and is currentl

Full text of DOI:10.1016/j.jsbmb.2010.03.063

Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 51–55
Contents lists available at ScienceDirect
Journal of Steroid Biochemistry and Molecular Biology
journal homepage: www.elsevier.com/locate/jsbmb
1-Desoxy analog of 2MD: Synthesis and biological activity of
ଝ
(20S)-25-hydroxy-2-methylene-19-norvitamin D3
Izabela Sibilskaa,b, Katarzyna M. Baryckaa, Rafal R. Sicinskia,b, Lori A. Pluma, Hector F. DeLucaa,∗
a Department of Biochemistry, University of Wisconsin-Madison, 433 Babcock Drive, Madison, WI 53706-1544, USA
b Department of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 30 October 2009
Accepted 10 March 2010
During our ongoing structure–activity studies in the vitamin D area, we obtained (20S)-1␣,25-dihydroxy-
2-methylene-19-norvitamin D3 (5). This analog, designated 2MD, is characterized by a significantly
enhanced calcemic activity and is currently evaluated as a potential drug for osteoporosis. Therefore,
it was of interest to synthesize also its 1-desoxy analog and to evaluate its biological action. These stud-
ies were aimed at solving an intriguing problem: can such a vitamin also be hydroxylated in vivo at the
allylic C-1 position despite lack of the exomethylene moiety at C-10? The Wittig-Horner coupling of the
known protected (20S)-25-hydroxy Grundmann ketone 17 and the phosphine oxides 16 and 33, differing
in their hydroxyls protection, provided the target 1-desoxy-2MD (6) after removal of the silyl protecting
groups. Two synthetic paths have been elaborated leading to the desired A-ring synthons and starting
from commercially available compounds: 1,4-cyclohexanedione monoethylene acetal (7) and (−)-quinic
acid (19). The biological activity in vitro of the synthesized 1-desoxy-2MD (6) was evaluated and this
analog was found to have an affinity for the vitamin D receptor (VDR) similar as its parent compound
2MD (5) while being much less active in the transcriptional assay. The results of the biological tests in
vivo are also discussed.
Keywords:
Vitamin D analogs
19-Norvitamin D
2MD
Vitamin D receptor
© 2010 Published by Elsevier Ltd.
1. Introduction
pound, called 2MD (5), is characterized by a significantly enhanced
activity on bone [9] and is currently in clinical development as a
1␣,25-Dihydroxyvitamin D3 [1␣,25-(OH)2D3, calcitriol, 1; Fig. 1]
is the active, hormonal form of vitamin D3, involved in the regula-
tion of the serum calcium and phosphorus homeostasis as well as in
the wide array of other biological functions [1,2]. Formation of cal-
citriol in mammals includes two steps. In the liver, vitamin D3 (2)
undergoes its first metabolic activation in which it is converted to
25-hydroxyvitamin D3 (3), the major circulating metabolite of vita-
min D [3]. Second step occurs in the kidney [4] where compound 3
is hydroxylated at 1␣-position to form the final vitamin D hormone,
1␣,25-(OH)2D3 (1). Compared to its biological precursor 3, calcitriol
has proved to be about three orders of magnitude more active
then 25-OH-D3 in binding to the VDR and in the differentiation of
human promyelocytic leukemia (HL-60) cells, respectively [5,6]. In
our continued investigation of the structure–activity relationship
in the vitamin D area, we obtained an analog of 1␣,25-(OH)2D3 pos-
sessing the inverted (20S)-configuration and the A-ring exocyclic
methylene group transposed from C-10 to C-2 [7,8]. This com-
potential drug for osteoporosis. Taking into account a very high
calcemic activity of 2MD, we synthesized the related compound 4
(25-desoxy-2MD), that can be “activated” in a living organism by
25-hydroxylation process [10]. Such a “prodrug” turned out to be
a highly potent analog, with calcemic activities approaching these
of the corresponding 25-hydroxylated counterpart 5. The above-
mentioned results, suggesting in vivo side chain hydroxylation of
the analog 4 in the liver, encouraged us to undertake the studies
directed towards the synthesis of the respective analog of 2MD
lacking the 1␣-hydroxyl group (6, 1-desoxy-2MD). We expected
that biological testing of such an analog could provide indication
whether the enzymatic 1␣-hydroxylation process was also possible
in the 2-methylene-19-norvitamin D derivatives.
2. Materials and methods
2.1. Preparation of
(20S)-25-hydroxy-2-methylene-19-norvitamin D3 (6)
ଝ
Special issue selected article from the 14th Vitamin D Workshop held at Brugge,
(20S)-25-Hydroxy-2-methylene-19-norvitamin D3 was syn-
thesized at the Department of Biochemistry, University of
Wisconsin-Madison and at the Department of Chemistry, Uni-
Belgium on October 4–8, 2009.
∗
Corresponding author. Tel.: +1 608 262 1620; fax: +1 608 262 7122.
E-mail address: deluca@biochem.wisc.edu (H.F. DeLuca).
0960-0760/$ – see front matter © 2010 Published by Elsevier Ltd.
doi:10.1016/j.jsbmb.2010.03.063

52
I. Sibilska et al. / Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 51–55
Fig. 1. Chemical structure of 1␣,25-dihydroxyvitamin D3 (calcitriol, 1) and its analogs.
using 1␣,25-(OH)2[26,27-3H]D3 as previously described [11].
The experiments were carried out in duplicate, on two or three
different occasions.
versity of Warsaw, according to the synthetic route presented in
Schemes 1 and 2. All prepared compounds exhibited spectroscopic
and analytical data consistent with their structures. Full details of
their synthesis will be reported elsewhere.
2.2.2. Measurement of cellular differentiation
2.2. In vitro studies
Human leukemia HL-60 cells (obtained from ATTC) were plated
at 2 × 105 cells per plate and incubated. Then the test compounds
were added and, after 4 days, superoxide production was mea-
sured by nitro blue tetrazolium (NBT) reduction. The experiment
was repeated in duplicate, two times. This method is described in
detail elsewhere [6].
2.2.1. Measurement of binding to the rat recombinant vitamin D
receptor
The procedure for obtaining the purified rat recombinant
vitamin D receptor used in the binding studies will be reported
in detail elsewhere. Competition binding assays were performed
Scheme 1. (a) PhNO, l-Pro, CHCl3, 44% (e.e. > 97%); (b) TBDPSCl, Im, DMF, 88%; (c) Ph3P+CH3Br−, n-BuLi, THF, 93%; (d) FeCl3 × H2O, CH2Cl2, 99%; (e) TMSCH2COOMe, LDA,
THF, 90%; (f) DIBALH, PhMe, CH2Cl2, chrom. sep., 92% (ratio of 14:15 = 0.74:1); (g) n-BuLi, p-TsCl, THF; (h) n-BuLi, Ph2Ph, THF; (i) H2O2, H2O, CH2Cl2, 79% (over three steps);
(j) n-BuLi, 17, THF, 70%; (k) n-Bu4NF, THF, 85%.

I. Sibilska et al. / Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 51–55
53
Scheme 2. (a) p-TsOH, PhH, DMF, 72%; (b) TDCI, MeCN, 63%; (c) Bu3SnH, AIBN, PhH, THF, 88% (ratio of 22:23 = 0.06:1; (d) PDC, 4 Å mol. sieves, MeCN, 82%; (e) Ph3P+CH3Br−
,
t-BuOK, THF, 80%; (f) MeOH, 4 Å mol. sieves, 59%; (g) TBDMSOTf, 2,6-lutidine, CH2Cl2, 80% (over two steps); (h) DIBALH, CH2Cl2, 94%; (i) NaIO4, MeOH, H2O, 95%; (j)
Ph3P = CHCOOMe, PhH, 89%; (k) DIBALH, CH2Cl2, chrom. sep., 97% (ratio of 31:32 = 0.67:1); (l) n-BuLi, p-TsCl, THF; (m) n-BuLi, Ph2Ph, THF; (n) H2O2, H2O, CH2Cl2, 69% (over
three steps); (o) PhLi, 17, THF, 70%; (p) CSA, MeOH, 14% (over two steps).
2.2.3. Transcriptional assay
[14] and involving the reaction of a ketone with nitrosobenzene
in the presence of a catalytic amount of l-proline. The introduced
secondary hydroxyl group in the product 8 was silylated and the
protected compound 9 was subjected to the Wittig reaction with an
ylide generated from methyltriphenylphosphonium bromide and
n-butyllithium. In the resulting olefinic compound 10, the carbonyl
group was deprotected in the reaction with the Lewis acid and
the formed cyclohexanone 11 was subjected to a Peterson reac-
tion leading to the mixture of ␣,␤-unsaturated esters 12 and 13.
The separation of the geometric isomers, although possible also
at this stage, was more easily achieved (by column chromatogra-
phy) after the reduction step, providing the E- and Z-allylic alcohols
14 and 15, respectively. The E-isomer 15 was then transformed
in a three-step procedure into the corresponding phosphine oxide
16. The Wittig-Horner coupling of the known Grundmann ketone
17 [7] with a lithium phosphinoxy carbanion generated from the
phosphine oxide 16 was subsequently carried out, producing the
protected 19-norvitamin D compound 18 which, after deprotection
of hydroxyl groups with tetrabutylammonium fluoride, provided
the desired (20S)-25-hydroxy-2-methylene-19-nor-vitamin D3 (6,
1-desoxy-2MD).
A different synthetic sequence led to the building block 33 and
to the final vitamin 6. The chiral starting compound was a com-
mercially available (−)-quinic acid 19, which was at first converted
to the known lactone 20 [15]. Treatment of this compound with
a 1,1ꢀ-thiocarbonyldiimidazole resulted in formation of the cyclic
thiocarbonate 21 [16]. The Barton-McCombie deoxygenation of 21
with tributyltin hydride and AIBN provided two isomeric products:
the known compound 22 [17] and the desired diol 23. Oxidation
of the secondary hydroxyl group in the latter isomer yielded the
ketone 24 which was subjected to the Wittig methylenation. The
lactone ring in the formed compound 25 was then opened and the
Transcription activity was measured in ROS 17/2.8 (bone) cells
that were stably transfected with a 24-hydroxylase (24OHase) gene
promoter upstream of a luciferase reporter gene [12]. The cells were
given a range of doses. Sixteen hours after dosing, the cells were
harvested and luciferase activities were measured using a lumi-
nometer. Eachexperimentwas performedin duplicate, two or three
times.
2.3. In vivo studies
Bone calcium mobilization. Male, weanling Sprague–Dawley
rats were purchased from Harlan (Indianapolis, IN). The animals
were group housed and placed on Diet 11 (0.47% Ca) + AEK oil for
one week followed by Diet 11 (0.02% Ca) + AEK oil for 3 weeks. The
rats were then switched to a diet containing 0.47% Ca [13] for 1
week followed by 2 weeks on a diet containing 0.02% Ca. Dose
administration began during the last week on 0.02% Ca diet. Four
consecutive intraperitoneal doses were given approximately 24 h
apart. Twenty-four hours after the last dose, blood was collected
from the severed neck and the concentration of serum calcium
determined as a measure of bone calcium mobilization.
3. Results and discussion
3.1. Chemical synthesis of 6
For the preparation of the A-ring fragments, phosphine oxides
16 and 33, two alternative synthetic routes were established.
The first synthesis started from an achiral, commercially avail-
able acetal-ketone 7 (Scheme 1), that was enantioselectively
␣-hydroxylated, using the method elaborated by Hayashi et al.

54
I. Sibilska et al. / Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 51–55
Fig. 2. Competitive binding of 1␣,25-(OH)2D3 (1) and the synthesized 1-desoxy-
2MD (6) to the rat recombinant vitamin D receptor. This experiment was carried
out in duplicate on two different occasions.
Fig. 3. Differentiation activity of 1␣,25-(OH)2D3 (1) and the synthesized 1-desoxy-
2MD (6). The differentiation state was determined by measuring the percentage
of cells reducing nitro blue tetrazolium (NBT). The experiment was repeated in
duplicate, two times.
secondary hydroxyl group silylated. The methyl ester moiety in the
obtained product 26 was reduced and the diol 27 was subjected
to periodate oxidation. The Wittig reaction of the obtained cyclo-
hexanone 28 with methyl (triphenylphosphoranylidene)acetate
provided a mixture of ␣,␤-unsaturated esters 29 and 30. These were
reduced with DIBALH and the obtained allylic alcohols separated by
column chromatography. The E-isomer 31 was then converted into
the corresponding allylic phosphine oxide 33. Its anion, generated
by phenyllithium, was coupled with the Grundmann ketone 17 and
the final 19-norvitamin 6 was obtained after acidic deprotection of
hydroxyls.
3.2. Biological evaluation of the synthesized analog 6
Interestingly, the synthesized vitamin 6 was almost as active as
1␣,25-(OH)2D3 and its parent 2MD [7], in binding to the full-length
recombinant rat vitamin D receptor (Fig. 2). This is a surprising
result because removal of the 1␣-hydroxyl group from the native
hormone reduces the binding activity by three orders of magni-
tude [5]. The obtained vitamin 6 proved to be slightly less active
in the HL-60 assay (Fig. 3) than the natural hormone but ca. 70
times less potent compared to 2MD [7]. When the ability of the
analog 6 to induce transcription of vitamin D-responsive genes was
examined using the 24-hydroxylase (CYP-24) luciferase reporter
gene system (Fig. 4), the synthesized compound appeared to be
less potent by three orders of magnitude than 2MD; a similar rela-
tionship was observed for 25-OH-D3 and 1␣,25-(OH)2D3 [12]. The
in vivo tests performed in rats (Fig. 5) showed that removal of the
1␣-hydroxyl group from the native hormone (1) does not appear to
affect bone calcium mobilization of its 1-desoxy form 3 (25-OH-D3)
Fig. 4. Transcriptional activity of 1␣,25-(OH)2D3 (1) and the synthesized 1-desoxy-
2MD (6). Transcriptional assay was carried out in rat osteosarcoma cells stably
transfected with a 24-hydroxylase gene reporter plasmid. Each experiment was
performed in duplicate, two or three separate times.
because the 25-OH-D3 is readily 1-hydroxylated in vivo. However,
in the case of 2MD, elimination of its 1␣-hydroxyl group dramat-
ically reduced (by ca. two orders of magnitude) the activity of its
1-desoxy analog 6 in the bone [7]. Thus it is likely that analog 6 in
clear contrast to 25-OH-D3 is not significantly hydroxylated in vivo.
Fig. 5. Bone calcium mobilization of 1␣,25-(OH)2D3 (1), 25-OH-D3 (3), 2MD (5) and the synthesized 1-desoxy-2MD (6).

I. Sibilska et al. / Journal of Steroid Biochemistry & Molecular Biology 121 (2010) 51–55
55
3.3. Conclusion
[4] M.F. Holick, H.K. Schnoes, H.F. DeLuca, T. Suda, R.J. Cousins, Isolation and iden-
tification of 1,25-dihydroxycholecalciferol. A metabolite of vitamin D active in
intestine, Biochemistry 53 (1971) 2799–2804.
It has been commonly accepted that the presence of 1␣-
hydroxyl group in the vitamin D molecules is of crucial importance
for their binding to the VDR and exertion of other biological activ-
ities. It was, therefore, rather surprising that 1-desoxy analog of
2MD, whose synthesis is described above, retained good affin-
ity for the VDR. However, in other in vitro assays, the analog 6
showed the expected much lower activity compared to its parent
compound 2MD. Lack of 1␣-OH had also a detrimental effect in
vivo, proving a significantly decreased calcemic potency of 6. These
results seem to indicate that 2-methylene-19-norvitamins with a
(20S)-configuration are not enzymatically hydroxylated (at least,
not efficiently) at C-1 despite the fact that this position is allylic,
similarly to the case of the “normal” vitamin D compounds with a
preserved C(10) = C(19) moiety and (20R)-configuration.
[5] H.F. DeLuca, H.E. Paaren, H.K. Schnoes, Vitamin D and calcium metabolism, Top.
Curr. Chem. 83 (1979) 1–65.
[6] V.C. Ostrem, W.F. Lau, S.H. Lee, K. Perlman, J. Prahl, H.K. Schnoes, H.F. DeLuca,
N. Ikekawa, Induction of monocytic differentiation of HL-60 cells by 1,25-
dihydroxyvitamin D analogs, J. Biol. Chem. 262 (1987) 14164–14171.
[7] R.R. Sicinski, J.M. Prahl, C.M. Smith, H.F. DeLuca, New 1␣,25-dihydroxy-19-
norvitamin D3 compounds of high biological activity: synthesis and biological
evaluation of 2-hydroxymethyl, 2-methyl and 2-methylene analogues, J. Med.
Chem. 41 (1998) 4662–4674.
[8] R.R. Sicinski, 2-Alkylidene analogs of 19-nor-1␣,25-(OH)2D3: synthesis and
biological activity, Pol. J. Chem. 80 (2006) 573–585.
[9] N.K. Shevde, L.A. Plum, M. Clagett-Dame, H. Yamamoto, J.W. Pike, H.F. DeLuca,
A potent analog of 1␣,25-dihydroxyvitamin D3 selectively induces bone for-
mation, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 13487–13491.
[10] P. Grzywacz, L.A. Plum, W. Sicinska, R.R. Sicinski, J.M. Prahl, H.F. DeLuca,
2-Methylene analogs of 1␣-hydroxy-19-norvitamin D3: synthesis, biological
activities and docking to the ligand binding domain of the rat vitamin D recep-
tor, J. Steroid Biochem. Mol. Biol. 89–90 (2004) 13–17.
[11] M.C. Dame, E.A. Pierce, J.M. Prahl, C.E. Hayes, H.F. DeLuca, Monoclonal antibod-
ies to the porcine intestinal receptor for 1,25-dihydroxyvitamin D3: interaction
with distinct receptor domains, Biochemistry 25 (1986) 4523–4534.
[12] N.C. Arbour, T.K. Ross, C. Zierold, J.M. Prahl, H.F. DeLuca, A highly sensi-
tive method for large-scale measurements of 1,25-dihydroxyvitamin D, Anal.
Biochem. 255 (1998) 148–154.
[13] T. Suda, H.F. DeLuca, Y. Tanaka, Biological activity of 25-hydroxyergocalciferol
in rats, J. Nutr. 100 (1970) 1049–1052.
[14] Y. Hayashi, J. Yamaguchi, T. Sumiya, K. Hibino, M. Shoji, Direct proline-catalyzed
asymmetric ␣-aminoxylation of aldehydes and ketones, J. Org. Chem. 69 (2004)
5966–5973.
Acknowledgements
The work was supported in part by funds from the Wiscon-
sin Alumni Research Foundation. Special thanks are addressed to
Jean Prahl and Jennifer Vaughan for carrying out in vitro studies
and to Heather Neils and Xiaohong Ma for conducting the in vivo
studies.
[15] J.D. Elliott, M. Hetmanski, R.J. Stoodley, M.N. Palfreyman, Studies related to
cyclopentanoid natural products. 1. Preparation of (4RS) and (4R)-4-hydroxy-
2-hydroxymethylcyclopent-2-en-1-one–a versatile synthetic intermediate, J.
Chem. Soc., Perkin Trans. 1 (1981) 1782–1789.
[16] S. Mills, R. Desmond, R.A. Reamer, R.P. Volante, I. Shinkai, Tetrahedron Lett. 29
(1988) 281–284.
References
[1] D. Feldman, J.W. Pike, F.H. Glorieux (Eds.), Vitamin D, second edition, Elsevier
Academic Press, 2005.
[2] G. Jones, S.A. Strugnell, H.F. DeLuca, Current understanding of the molecular
action of vitamin D, Physiol. Rev. 78 (1998) 1193–1231.
[3] J.W. Blunt, H.F. DeLuca, H.K. Schnoes, 25-Hydroxycholecalciferol. A biologically
active metabolite of vitamin D3, Biochemistry 7 (1986) 3317–3322.
[17] C. Gonzales-Bello, J.R. Coggins, A.R. Hawkins, C. Abell, J. Chem. Soc., Perkin Trans.
1 (1999) 849–854.