An o -aminoanilide analogue of 1α,25-dihydroxyvitamin D3 functions as a strong vitamin D receptor antagonist

Article abstract of DOI:10.1021/jm1007159

Vitamin D receptor (VDR) antagonists have therapeutic potential in treatment of allergic conditions and hypercalcemia driven by granulomatous diseases. We have identified an o-aminoanilide analogue of the hormonal form of vitamin D with a dienyl side chai

Full text of DOI:10.1021/jm1007159

J. Med. Chem. 2010, 53, 7461–7465 7461
DOI: 10.1021/jm1007159
An o-Aminoanilide Analogue of 1r,25-Dihydroxyvitamin D3 Functions as a Strong
Vitamin D Receptor Antagonist
Marc Lamblin,† Russell Spingarn,‡ Tian-Tian Wang,‡ Melanie C. Burger,† Basel Dabbas,‡ Nicolas Moitessier,†
John H. White,*,‡,§ and James L. Gleason*,†
†
Departments of Chemistry, ‡Physiology, and §Medicine, McGill University, Montreal, Quebec, H3A 2K6, Canada
Received June 16, 2010
Vitamin D receptor (VDR) antagonists have therapeutic potential in treatment of allergic conditions
and hypercalcemia driven by granulomatous diseases. We have identified an o-aminoanilide analogue of
the hormonal form of vitamin D with a dienyl side chain that functions as a strong VDR antagonist.
Modeling studies indicate that antagonism arises from side chain rigidity, when compared to a more
flexible saturated analogue, which interferes with H12 folding/alignment.
Chart 1. 1,25D, TSA, and 1,25D Analogues
Introduction
The hormonal form of vitamin D3, 1R,25-dihydroxyvita-
min D3 (1,25D,a 1) (Chart 1) has long been known for its cen-
tral role in controlling calcium homeostasis. More recently,
data from several sources have provided evidence that endog-
enous 1,25D has cancer chemopreventive properties and that
it functions as a key regulator of innate and adaptive immune
responses.1,2 As the therapeutic activity of exogenously ad-
ministered 1,25D is limited by its capacity to induce hyper-
calcemia, numerous laboratories have developed a wide range
of 1,25D analogues that retain VDR agonism but minimize its
calcemic actions.3 In contrast, only a small number of VDR
antagonists have been reported, and of these most are only
partial antagonists.4 VDR antagonists have therapeutic po-
tential in modulation of allergic responses5 and in suppression
of hypercalcemia associated with excessive macrophage ac-
tion in granulomatous diseases such as sarcoidosis.6
We have been interested in developing bifunctional 1,25D
analogues that combine VDR agonism with histone deacety-
lase (HDAC) inhibitory activity,7,8 as both 1,25D and HDACs
are modulators of gene transcription and have potential
in cancer therapy. HDAC inhibitors (HDACi) such as tri-
chostain A (TSA, 2) (Chart 1) contain terminal zinc binding
groups such as hydroxamic acids. In the bifunctional ana-
logue triciferol (3), the side chain of the 19-nor analogue of19 is
replaced by the dienylhydroxamic acid of 2. Detailed bio-
chemical studies showed that 3 is a VDR agonist and HDACi.7
We have since developed a range of analogues possessing
alternative terminal zinc binding groups including o-amino-
anilides, thioglycloates, and sulfonamides.8 Notable among
these second-generation hybrid molecules was o-aminoanilide
ML 2-250 (4),8 which retained significant VDR agonism even
with a sterically demanding o-aminoanilide group at the side
chain terminus. Here we show that a minor structural change,
reverting tothe dienyl sidechain found in TSA andtriciferolin
place of the saturated side chain, transforms agonist 4 into
strong VDR antagonist ML 3-452 (5).
*To whom correspondence should be addressed. For J.H.W.: phone,
514-398-8498; fax, 514-398-7452; e-mail, john.white@mcgill.ca. For
J.L.G.: phone, 514-398-5596; fax, 514-398-3797; e-mail, jim.gleason@
mcgill.ca.
a Abbreviations: VDR, vitamin D receptor; VDR-LBD, vitamin D
receptor ligand binding domain; HDAC, histone deacetylase; HDACi,
histone deacetylase inhibitor; 1,25D, 1,25-dihydroxyvitamin D3; TSA,
trichostatin A; TSLP, thymic stromal lymphoprotein; PCR, polymerase
chain reaction; qPCR, quantitative polymerase chain reaction; MD,
molecular dynamics; rmsd, root mean square deviation.
Results and Discussion
VDR antagonist 5 was synthesized following the procedure
described for 3, with o-aminoanilide formation replacing
hydroxamic acid formation (see Supporting Information for
synthetic scheme). The agonist activity of 1, 4, and 5 was
compared in a 1,25D-sensitive squamous carcinoma cell line
SCC2510 by screening for induction of the CYP24 gene which
r
2010 American Chemical Society
Published on Web 09/30/2010
pubs.acs.org/jmc

7462 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 20
Lamblin et al.
encodes the enzyme that initiates metabolism of 1. Hybrid
5 (1 μM, Figure 1a) induced substantially less CYP24 expres-
sion than a suboptimal concentration of 1 (10 nM) or 4
(1 μM). Similarly, 5 (1 μM) failed to induce transcription of
the gene encoding thymic stromal lymphoprotein (TSLP),
whereas TSLP expression was strongly stimulated by 1
(100 nM, Figure 1b). Direct binding of 5 to the VDR was
tested by a fluorescence polarization competition assay7 using
a fluorescent tracer bound to the purified VDR ligand binding
domain. These studies (Figure 1c) confirmed that 5 bound
directly to the VDR ligand binding domain and displaced the
bound tracer with an estimated IC50 of 107 nM, or about
8-fold less potently than 1 in this assay. For comparison,
agonist 4 binds to the VDR with an IC50 of 248 nM, approxi-
mately 2.5 times weaker than 5, yet shows clear dose-
dependent induction of CYP24 expression in SCC25 cells.8
The lack of significant CYP24 and TSLP expression
coupled with the affinity of 5 for the VDR raised the possi-
bility that it could act as an antagonist for the receptor.
Returning to gene expression, analysis by semiquantitative
and real-time PCR showed that 5 blocked CYP24 induction
by 1 (10 nM), with substantial inhibition observed at concen-
trations as low as 100 nM (Figure 2). Again, no substan-
tial CYP24 expression was observed in the presence of 5
(1-10 μM) as measured by semiquantitative or qPCR. We
further compared antagonist 5 with established antagonist
ZK159222 (6).3 6 appeared to be a more potent antagonist
than 5 (Figure 2b). However, in contrast to 5, 6 induced
measurable CYP24 expression (Figure 2).
In the presence of 1, the DNA-bound VDR activates
transcription in part by recruiting coactivator proteins to pro-
moter regions of target genes.1 Unlike 1, 5 failed to induce
association of the VDR coactivator AIB1 with the vitamin
D-responsive region of the CYP24 promoter and blocked
AIB1 recruitment induced by 1, as assessed by chromatin
immunoprecipitation (ChIP) assay (Figure 3).7 Moreover, in
a similar assay, 5 partially induced association of corepressor
NCoR with the CYP24 promoter, whereas 1 suppressed
NCoR binding (Figure 3), further substantiating function of 5
as an antagonist.
Intriguingly, in addition to its different action on the VDR,
hybrid 5 also is devoid of the HDACi activity of 4. Treatment
of SCC4 cells with5 (1 μM) did not induce hyperacetylation of
histone H4 or tubulin as judged by Western blotting. This is in
contrast to 4 which induces histone H4 and tubulin hyper-
acetylation at equivalent concentrations.8 Moreover, no inhibi-
tion of purified recombinant human HDAC6 by 5 was observed
up to 1 mM using a standard fluorescence assay.11 Again, by
contrast, hybrid 4 inhibits HDAC6 with an IC50 of 81 μM.8
The VDR antagonist activity of 5 is striking given the
agonist action of 4 and 3. Antagonism of the VDR is usually
suggested to arise by destabilization of the required position-
ing of H12 that forms part of a crucial recognition element for
coactivator recruitment.4b,g,12 To gain insight into the differ-
ing effects of 4 and 5 on the VDR, we modeled their interac-
tions using virtual docking and molecular dynamics simu-
lations. We employed FITTED 2.6, a suite of programs for
virtual screening.13 Visual inspection of the docked poses to
the full-length VDR X-ray structure14 revealed that 4 may
Figure 1. Expression of (a) CYP24 and (b) TSLP in SCC25 cells, as
monitored by CYP24, indicates no VDR agonism by o-aminoanilide
5. GAPDH is used as a control. (c) A fluorescence polarization assay
of hybrid 5 shows binding to the VDR with an IC50 of 107 nM.
Figure 2. Semiquantitative (a) or qPCR (b) analysis of CYP24 regulation in SCC25 cells. Clear dose-dependent inhibition of expression
induced by 1,25D (1) and no substantial agonism of o-aminoanilide (5) were observed over 0.01-10 μM. Analysis of the partial agonist and
antagonist activity of 6 is provided for comparison.

Brief Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 20 7463
stabilize the agonist conformation through a hydrogen
bonding interaction between the o-aminoanilide and histi-
dine 395 on helix 11 (H11). The repositioning of H11 so that
it is contiguous with H10 is required for folding over of H12
to generate an agonized conformation of the VDR.15 Con-
versely 5 was unable to orient the o-aminoanilide to interact
with His395 in an agonized conformation of the VDR
(Figure 4b).16
The only X-ray crystal data available for VDR antagonists
are structures that have been cocrystallized with coactivator
peptides, which presumably leads to an active conformation
of the receptor.4g,h To mimic an inactive conformation of the
receptor, we truncated the 1,25D-liganded VDR14 by remov-
ing H12. When docked to this structure, the side chains of 4
and 5 extended into the volume previously occupied by H12
(Figure 4). However, when energies of binding are compared
to those of the full and truncated VDR, an energy benefit of
-20 kcal/mol was observed for 5, which was considerably
stronger than that observed for 4 (-6 kcal/mol). The larger
energy difference is partially accounted for by loss of syn-
pentane and A-1,3 strain in the extended conformation of 5
and suggests that it is much less tolerant of compaction into
the ligand binding pocket and may significantly hinder proper
positioning of H12.17
To further examine the potential influence of 5 on H12, we
compared effects of 1, 4, and 5 bound to VDR in molecular
dynamics (MD) simulations using Amber 10 (Figure 5).12a,18
MDsimulationof VDR complexesof 1 and 4 over a 40nstime
frame showed minor variations in the position of H12 resi-
dues. In the case of 4, H12 closed very slightly over the binding
pocket during the time frame we examined, while with 1, H12
fluctuated insignificantly about the average agonist confor-
mation. In the case of antagonist 5, H12 was displaced slightly
away from the binding pocket during the first portion of
the simulation. More strikingly, during the second half of the
simulation, H12 began to partially unwind. This change is
most readily apparent by observing the movement of the side
chains of Glu420, Val421, and Phe422 (Figure 5c), the last of
˚
˚
which is displaced with an rmsd of 5.26 A (vs 1.43 A for 1 and
˚
2.26 A for 4). This is considerable movement over a short time
frame and suggests the potential to completely displace H12
over the time frame normally expected for movement of larger
subunits (on the order of seconds). This movement also
Figure 5. Molecular dynamics simulations showing initial posi-
tions (in gray) of ligands (right) and H12 (left) and final positions
of ligands and H12 (in color) after 40 ns for (a) 1,25D (1), (b) agonist
4, and (c) antagonist 5 showing significant movement and partial
unwinding of H12 induced by 5.
Figure 3. ChIP analysis of the effects of 1 (10 nM) or 5 (1 μM) alone
or in combination on the association of VDR coactivator AIB1
(top) or corepressor NCoR (bottom) with the VDR binding region
of the proximal promoter of the CYP24 gene.
Figure 4. Docking of (a) agonist 4 and (b) antagonist 5 to VDR-LBD showing hydrogen bonding in the former. Docking of (c) agonist 4 and
(d) antagonist 5 to VDR-LBD lacking H12 indicates a preference for both ligands for an extended conformation. Extracted conformation of
agonist 5 bound to (e) VDR-LBD and (f ) VDR-LBD lacking H12 shows relief of syn-pentane and A-1,3 interactions in H12 deleted structure.

7464 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 20
Lamblin et al.
significantly affected the distance between Lys248 and Glu420
that form a charge clamp for coactivator binding. For 1 and 4,
˚
this distance varied between 18 and 19.6 A during the course
of the simulation, while for 5, the distance increased substan-
˚
tially to 21.5 A at the end of the simulation. Taken together,
the MD and docking studies suggest that the relative rigidity
and propensity to avoid internal A-1,3 strain of LBD-bound 5
make it more difficult for H12 to be properly positioned
within the VDR to allow efficient coactivator binding, thus
leading to an inactive holo-VDR and explaning the antago-
nistic effect of 5 on 1 when present in concentrations high
enough to compete effectively for binding.
(2H, brs), 7.06 (1H, d, J = 7.6 Hz), 6.16 (1H, d, J = 11.2 Hz),
5.92 (1H, d, J = 14.8 Hz), 5.80 (1H, d, J = 11.2 Hz), 5.72 (1H, d,
J = 10.0 Hz), 4.10-4.02 (2H, m), 3.94 (1H, brs), 2.81 (1H, d,
J = 12.0 Hz), 2.55 (1H, d, J = 6.8 Hz), 2.40-2.31 (2H, m), 2.26
(1H, d, J = 12.8 Hz), 2.10 (1H, dd, J = 12.8, 8.0 Hz), 2.05-1.95
(2H, m), 1.87-1.30 (16H, m), 1.13 (1H, brs), 1.02 (3H, d, J = 6.4
Hz), 0.87 (9H, s), 0.57 (9H, s), 0.58 (3H, s), 0.05 (12H, s); 13C
NMR (75 MHz, CDCl3) δ 165.3, 148.5, 141.1, 140.6, 134.1,
129.6, 128.5, 127.2, 125.3, 124.8, 121.9, 119.8, 118.5, 117.6,
116.5, 68.3, 68.2, 56.6, 56.3, 46.2, 45.9, 43.9, 40.7, 37.1, 36.3,
28.9, 27.4, 26.11 (3C), 26.09 (3C), 23.6, 22.5, 20.1, 18.40, 18.37,
13.0, 12.7, -4.4, -4.5, -4.6, -4.7; IR (film) ν 3248 (br), 2952,
2856, 1613, 1530, 1456, 1253, 1086, 836, 775 cm-1. HRMS (ESI)
m/z calcd for [(M þ H)þ], 733.5160; found, 733.5152.
(R,2E,4E)-N-(2-Aminophenyl)-6-((1R,3aS,7aR,E)-4-(2-((3R,5R)-
3,5-dihydroxycyclohexylidene)ethylidene)-7a-methyloctahydro-
1H-inden-1-yl)-4-methylhepta-2,4-dienamide (5). TBAF (256 μL
of a 1 M solution in THF, 0.064 mmol, 4 equiv) and Et3N (7 μL,
0.048 mmol, 3 equiv) were added to a stirred solution of 8
(12 mg, 0.016 mmol, 1 equiv) in THF (1 mL) under argon, and
the mixture was stirred for 48 h. The solution was concentrated
in vacuo and loaded directly on to silica gel. Purification by silica
gel chromatography (20-60% acetone in hexanes) provided
5 as a clear oil in 62% yield (5 mg, 0.010 mmol). Further puri-
fication by reverse phase HPLC (C18, 70/30 MeCN/H2O)
yielded an analytically pure (>97%) sample which was used
for biological assays. Rf = 0.5 (60% acetone in hexanes);
1H NMR (400 MHz, CDCl3) δ 7.35 (1H, d, J = 15.2 Hz),
7.27-7.17 (1H, m), 7.15-6.97 (2H, m), 6.80 (1H, d, J = 8.0 Hz),
6.77 (1H, brs), 6.31 (1H, d, J = 11.2 Hz), 5.94 (1H, d, J = 14.8
Hz), 5.84 (1H, d, J = 11.2 Hz), 5.73 (1H, d, J = 10.6 Hz),
4.17-4.08 (1H, m), 4.08-3.99 (1H, m), 3.90 (2H, brs), 2.82 (1H,
d, J = 16.0 Hz), 2.72 (1H, dd, J = 12.8, 3.6 Hz), 2.62-2.51 (2H,
m), 2.48 (1H, d, J = 13.6 Hz), 2.27-2.14 (4H, m), 1.82 (3H, s),
1.92-1.31 (10H, m), 1.06 (3H, d, J = 6.4 Hz), 0.59 (3H, s); 13C
NMR (75 MHz, CDCl3) δ 165.2, 148.3, 142.9, 141.0, 131.6,
129.7, 127.3, 125.3, 124.0, 119.8, 118.5, 117.7, 115.7, 67.6, 67.4,
56.6, 56.3, 46.1, 44.9, 42.4, 40.6, 37.4, 36.2, 29.1, 27.3, 23.7, 22.5,
20.1, 13.0, 12.7; IR (film) ν 3351 (br), 2936, 2871, 1654, 1612,
1528, 1454, 1045, 908, 732 cm-1. HRMS (ESI) m/z calcd for
[(M þ H)þ], 505.3430; found, 505.3424.
Conclusion
We have identified an o-aminoanilide analogue of 1,25D
that is a potent antagonist of the VDR. The antagonistic
activity appears to arise because of a more rigid side chain,
which inhibits the alignment of H12 necessary for coactivator
binding to the VDR.
Experimental Section
((R,2E,4E)-6-((1R,3aS,7aR,E)-4-(2-((3R,5R)-3,5-Bis(tert-
butyldimethylsilyloxy)cyclohexylidene)ethylidene)-7a-methyl-
octahydro-1H-inden-1-yl)-4-methylhepta-2,4-dienoic Acid (7).
LiOH H2O (4.8 mg, 0.114 mmol, 3 equiv) was added to a
3
stirring solution of methyl-(2E,4E,6R)-6-((1R,3R,7E,17β)-1,3-
bis[tert-butyl(dimethyl)silyloxy]-9,10-secoestra-5,7-dien-17-yl)-
4-methylhepta-2,4-dienoate7 (25.5 mg, 0.038 mmol) in THF
(0.5 mL), MeOH (200 μL), and H2O (200 μL). The reaction
vessel was fitted with a reflux condenser, and the mixture
brought to reflux for 3 h via a heating mantle. The mixture
was cooled to room temperature and diluted with EtOAc
(5 mL), then quenched with a 1.0 M solution of HCl (5 mL).
The layers were separated, and the aqueous layer was further
extracted with EtOAc (2 ꢀ 5 mL). The organic layers were
combined and extracted with distilled H2O (5 mL) and brine
(5 mL), then dried (MgSO4), and concentrated in vacuo to give
the crude product. The acid was carried forward without further
purification. Quantitative yield was 24.4 mg, 0.038 mmol. Rf =
VDR Binding and Agonism Assays. VDR binding using a
flourescence polarization assay and VDR agonism/antagonism
assessed by PCR on CYP24 followed our published proce-
dures.7 For TSLP, the following primers were used: 50, -caacttg-
tagggct; 30, -gtcgattgaagcga..
1
0.30 (20% EtOAc in hexanes); H NMR (300 MHz, CDCl3)
δ 10.00-9.30 (1H, br s), 7.39 (1H, d, J = 15.5 Hz), 6.16 (1H, d,
J = 11.0 Hz), 5.82 (2H, m), 5.78 (1H, d, J = 15.5 Hz), 4.15-4.00
(2H, m), 2.87-2.78 (1H, m), 2.63-2.53 (1H, m), 2.44-2.24 (3H,
m), 2.16-1.96 (3H, m), 1.83 (3H, s), 1.82-1.36 (11H, m), 1.04
(3H, d, J = 6.5 Hz), 0.89 (9H, s), 0.88 (9H, s), 0.60 (3H, s), 0.06
(12H, m); 13C NMR (75 MHz, CDCl3) δ 173.1, 152.7, 149.8,
140.3, 134.0, 129.9, 121.7, 116.5, 114.8, 68.3, 68.1, 56.6, 56.3,
46.2, 46.0, 43.9, 40.7, 37.1, 36.4, 28.9, 27.5, 26.1 (6C), 23.6, 22.5,
20.1, 18.4 (2C), 12.9, 12.7, -4.3, -4.4, -4.5, -4.6; IR (film) ν
3000 (br), 2956, 1686, 1618, 1417, 1254, 1207, 1088, 1026, 908,
834, 801 cm-1. HRMS (ESI) m/z calcd for [(M þ H)þ], 643.4578;
found, 643.4570.
(R,2E,4E)-N-(2-Aminophenyl)-6-((1R,3aS,7aR,E)-4-(2-((3R,5R)-
3,5-bis(tert-butyldimethylsilyloxy)cyclohexylidene)ethylidene)-
7a-methyloctahydro-1H-inden-1-yl)-4-methylhepta-2,4-dien-
amide (8). HBTU (7.9 mg, 0.021 mmol, 1.1 equiv) was added to a
solution of acid 7 (12.2 mg, 0.019 mmol, 1 equiv), 1,2-phenyl-
enediamine (2.0 mg, 0.019 mmol, 1 equiv), HOBt (7.7 mg, 0.095
mmol, 5 equiv), and DIPEA (9.9 μL, 0.057 mmol, 3 equiv) in
DMF (190 μL). The mixture was stirred at room temperature for
1 h, then diluted with EtOAc (10 mL). The organic solution was
washed with saturated NaHCO3 (5 mL), water (5 mL), brine
(5 mL) and dried (Na2SO4). The solution was filtered, concen-
trated, and the oil was purified by silica gel chromatography
(10-30% EtOAc in hexanes) to provide 8 as a clear oil in 86%
yield (12 mg, 0.016 mmol). Rf = 0.6 (40% EtOAc in hexanes);
1H NMR (400 MHz, CDCl3) δ 7.35 (1H, d, J = 15.2 Hz), 7.23
HDAC Inhibition. HDAC inibition was measured using a
standard fluorometric assay.8,11
Chromatin Immunoprecipitation Assay. ChIP assays assessing
association of VDR coactivator AIB1 and corepressor NCoR
with the promoter of the CYP24 gene were performed eesen-
tially as described.19 AIB1 and NCoR were immunoprecipitated
with ChIP grade antibodies ab2831 and ab24552, respectively,
from Abcam (Cambridge, MA).
Molecular Modeling and Dynamics. Initial coordinates of
VDR ligand binding domain (LBD) were obtained from the
X-ray crystal structure of the VDR-LBD-1R,25(OH)2D3 com-
˚
plex (Protein Data Bank code 1DB1) determined at 1.80 A
resolution.14 Hydrogen atoms were added, and water molecules
were removed. The N(ε)-H tautomer of H397 was selected in
accordance with surrounding hydrogen bond network.14 Next,
hydrogen positions were optimized through energy minimiza-
tion of the VDR-LBD using OPLS_2005 in MacroModel
€
(Maestro 9.0, Schrodinger, Inc.). Ligands 4 and 5 were prepared
in Maestro 9.0 (Schrodinger, Inc.) and docked to the full-length
€
and H12-truncated (residues 417-423) VDR-LBD protein. Ten
docking runs were performed on the two ligands using the
FITTED docking program in the rigid protein mode and SAR
option.13 The structure of 1 cocrystallized with VDR-LBD, and
the top-ranking docked poses of 4 and 5 were selected for MD

Brief Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 20 7465
simulations. These simulations were carried out using the
Amber 10 molecular dynamics package following a modified
version of the protocol described by Kieltyka et al.20 Briefly,
the atomic charges of the ligands were calculated within the
SMART module of FITTED and assigned GAFF21 atom types.
Protein atoms were assigned AMBER atom types and used with
the parm99 force field, while ligands were described with GAFF
in combination with ad hoc parameters generated for 5 with
parmchk. LEaP (Amber 10) was used to combine the initial
structure of the VDR-LBD with 1, 4, and 5; to neutralize the
system with the addition of sodium cations; and to solvate the
whole in a truncated octahedron of TIP3P water molecules with
Nakabayshi, M.; Ikura, T.; Tamamura, H.; Ito, N.; Shimuzu, M.;
Yamamoto, K. A new class of vitamin D analogues that induce
structural rearrangement of the ligand-binding pocket of the receptor.
J. Med. Chem. 2009, 52, 1438–1449.
(5) Mora, J. R.; Iwata, M.; von Andrian, U. H. Vitamin effects on the
immune system: vitamins A and D take centre stage. Nat. Rev.
Immunol. 2008, 8, 685–698.
(6) Iannuzzi, M. C.; Rybicki, B. B.; Tierstein, A. S. Sarcoidosis.
N. Engl. J. Med. 2007, 357, 2153–2165.
(7) Tavera-Mendoza, L. E.; Quach, T.; Dabbas, B.; Hudon, J.; Liao,
X.; Palijan, A.; Gleason, J. L.; White, J. H. Incorporation of
histone deacetylase inhibition into the structure of a nuclear
receptor agonist. Proc. Nat. Acad. Sci. U.S.A. 2008, 105, 8250–
8255.
(8) Lamblin, M.; Dabbas, B.; Spingarn, R.; Mendoza-Sanchez, R.;
Wang, T.-T.; An, B.-S.; Huang, D. C.; Kremer, K.; White, J. H.;
Gleason, J. L. Vitamin D receptor agonist/histone deacetylase
inhibitor molecular hybrids. Bioorg. Med. Chem. 2010, 18, 4119–
4137.
(9) Perlman, K. L.; Sicinski, R. R.; Schnoes, H. K.; DeLuca, H. F.
1R,25-Dihydroxy-19-norvitamin D3, a novel vitamin D-related
compound with potential therapeutic activity. Tetrahedron Lett.
1990, 31, 1823–1824.
(10) Lin, R.; Nagai, Y.; Sladek, R.; Bastien, Y.; Ho, J.; Petrecca, K.;
Sotiropoulou, G.; Diamandis, E. P.; Hudson, T.; White, J. H.
Expression profiling in squamous carcinoma cells reveals pleiotro-
pic effects of vitamin D3 signaling on cell proliferation, differentia-
tion and immune system regulation. Mol. Endocrinol. 2002, 16,
1243–1256.
˚
a 10 A cutoff. The system was energy minimized stepwise, first
relaxing the water molecules and ions while keeping all other
atoms fixed and then minimizing all atoms. A periodic bound-
ary, constant volume simulation was run over 20 ps with a 1.6 fs
time step, keeping the protein and ligands restrained with a
˚
10 kcal mol-1
A
-1 harmonic constant, using SHAKE to omit
3
bond stretching involving H atoms and allowing the system to
heat up from 0 to 300 K using Langevin dynamics with a
collision frequency of 1.0 ps-1. Further 100 ps of simulation at
300K but at constant pressure (1 bar) was performed to relax the
system. From then on, 40 ns of unconstrained simulation at
otherwise identical conditions as the relaxation was performed,
with snapshots saved every 10 ps.
(11) (a) Wegener, D.; Wirsching, F.; Riester, D.; Schwienhorst, A. A fluo-
rogenic histone deacetylase assay well suited for high-throughput
activity screening. Chem. Biol. 2003, 10, 61–68. (b) Wegener, D.;
Hildmann, C.; Riester, D.; Schwienhorst, A. Improved fluorogenic histone
deacetylase assay for high-throughput-screening applications. Anal. Bio-
chem. 2003, 321, 202–208.
(12) (a) Perakyla, M.; Molnar, F.; Carlberg, C. A structural basis for the
species-specific antagonism of 26,23-lactones on vitamin D signal-
ing. Chem. Biol. 2004, 11, 1147–1156. (b) Hashimoto, Y.; Miyachi, H.
Nuclear receptor antagonists designed based on the helix-folding
inhibition hypothesis. Bioorg. Med. Chem. 2005, 13, 5080–5093.
(c) Yoshimoto, N.; Inaba, Y.; Yamada, S.; Makishima, M.; Shimizu,
M.; Yamamoto, K. 2-Methylene 19-nor-25-dehydro-1R-hydroxyvita-
min D3 26,23-lactones: synthesis, biological activities and molec-
ular basis of passive antagonism. Bioorg. Med. Chem. 2008, 16,
457–473.
Acknowledgment. This work was supported by Proof of
Principle and operating grants to J.H.W. and J.L.G. from the
Canadian Institutes of Health Research. J.H.W. is a Chercheur-
ꢀ
Boursier National of the Fonds de Recherche en Sante du
Quebec. M.C.B. thanks the Natural Science and Engineering
ꢀ
€
€
ꢀ
ꢀ
Research Council of Canada for a predoctoral fellowship.
Supporting Information Available: Scheme for the synthesis
of 5, energies for docked ligands, and additional molecular
dynamics structures. This material is available free of charge
via the Internet at http://pubs.acs.org.
(13) (a) Corbeil, C. R.; Englebienne, P.; Moitessier, N. Docking ligands
into flexible and solvated macromolecules. 1. Development and
validation of FITTED 1.0. J. Chem. Inf. Model. 2007, 47, 435–449.
(b) Corbeil, C. R.; Englebienne, P.; Moitessier, N. Docking ligands into
flexible and solvated macromolecules. 2. Development and application
of FITTED 1.5 to the virtual screening of potential HCV polymerase
inhibitors. J. Chem. Inf. Model. 2008, 48, 902–909. (c) Corbeil, C. R.;
Moitessier, N. Docking ligands into flexible and solvated macromole-
cules. 3. Impact of input ligand conformation, protein flexibility, and
water molecules on the accuracy of docking programs. J. Chem. Inf.
Model. 2009, 49, 997–1009.
(14) Rochel, N.; Wurtz, J. M.; Mitschler, A.; Klaholz, B.; Moras, D.
The crystal structure of the nuclear receptor for vitamin D bound to
its natural ligand. Mol. Cell 2000, 5, 173–179.
(15) Bourguet, W.; Germain, P.; Gronemeyer, H. Nuclear receptor
ligand-binding domains: three-dimensional structures, molecular
interactions and pharmacological implications. Trends Pharmacol.
Sci. 2000, 21, 381–388.
References
(1) Lin, R.; White, J. H. The pleiotropic actions of vitamin D.
BioEssays 2004, 26, 21–28.
(2) White, J. H. Vitamin D signaling, infectious diseases and regulation
of innate immunity. Infect. Immun. 2008, 76, 3837–3843.
(3) Eeln, G.; Gysemans, C.; Verlinden, L.; Vanoibeek, E.; De Clercq,
P.; Van Haver, D.; Mathieu, C.; Bouillon, R.; Verstuyf, A. Mech-
anisms and potential of the growth inhibitory actions of vitamin D
and analogs. Curr. Med. Chem. 2007, 14, 1893–1910.
(4) Selected examples: (a) Miura, D.; Manabe, K.; Ozono, K.; Saito,
M.; Gao, Q.; Norman, A. W.; Ishizuka, S. Antagonistic action of
novel 1R,25-dihydroxyvitamin D3-26,23-lactone analogs on differ-
entiation of human leukemia cells (HL-60) induced by 1R,25-
dihydroxyvitamin D3. J. Biol. Chem. 1999, 274, 16392–16399.
(b) Herdick, M.; Steinmeyer, A.; Carlberg, C. Antagonistic action of
a 25-carboxylic ester analogue of 1R,25-dihydroxyvitamin D3 is medi-
ated by a lack of ligand-induced vitamin D receptor interaction with
coactivators. J. Biol. Chem. 2000, 275, 16506–16512. (c) Herdick, M.;
Steinmeyer, A.; Carlberg, C. Carboxylic ester antagonists of 1R,25-
dihydroxyvitamin D3 show cell-specific actions. Chem. Biol. 2000, 7,
885–894. (d) Kato, Y.; Nakano, Y.; Sano, H.; Tanatani, A.; Kobayashi,
H.; Shimazawa, R.; Koshino, H.; Hashimoto, Y.; Nagasawa, K. Synthe-
sis of 1R,25-dihydroxyvitamin D3-26,23-lactams (DLAMs), a novel
series of 1R,25-dihydroxyvitamin D3 antagonist. Bioorg. Med. Chem.
Lett. 2004, 14, 2579–2583. (e) Saito, N.; Kittaka, A. Highly potent
vitamin D receptor antagonists: design, synthesis, and biological
evaluation. ChemBioChem 2006, 7, 1478–1490. (f) Cho, K.; Uneuchi,
F.; Kato-Nakamura, Y.; Namekawa, J.-I.; Ishizuka, S.; Takenouchi, K.;
Nagasawa, K. Structure-activity relationship studies on vitamin D
lactam derivatives as vitamin D receptor antagonist. Biorg. Med.
Chem. Lett. 2008, 18, 4287–4290. (g) Nakabayashi, M.; Yamada, S.;
Yoshimoto, N.; Tanaka, T.; Igarashi, M.; Ikura, T.; Ito, N.; Makishima,
M.; Tokiwa, H.; Deluca, H. F.; Shimizu, M. Crystal structures of rat
vitamin D receptor bound to adamantyl vitamin D analogs: structural
basis for vitamin D receptor antagonism and partial agonism. J. Med.
Chem. 2008, 51, 5320–5329. (h) Inaba, Y.; Yoshimoto, N.; Sakamaki, Y.;
(16) Graphics prepared in the PyMOL Molecular Graphics System,
€
version 1.2r2pre, from Schrodinger, LLC.
(17) Feng, W.; Ribeiro, R. C. J.; Wagner, R. L.; Nguyen, H.; Apriletti,
J. W.; Fletterick, R. J.; Baxter, J. D.; Kushner, P. J.; West, B. L.
Hormone-dependent coactivator binding to a hydrophobic cleft on
nuclear receptors. Science 1998, 280, 1747–1749.
€
€
(18) Perakyla, M. Lect. Notes Comput. Sci. (including subseries Lect.
Notes Artif. Intell. and Lect. Notes Bioinf.) 2007, 4699, 82-89.
(19) Wang, T.-T.; Nestel, F. P.; Bourdeau, V.; Nagai, Y.; Wang, Q.;
Liao, J.; Tavera-Mendoza, L.; Lin, R.; Hanrahan, J. H.; Mader, S.;
White, J. H. 1,25-Dihydroxyvitamin D3 directly stimulates anti-
microbial peptide gene transcription. J. Immunol. 2004, 173, 2909–
2912.
(20) Kieltyka, R.; Englebienne, P.; Moitessier, N.; Sleiman, H. Quanti-
fying interactions between G-quadruplex DNA and transition
metal complexes. Methods Mol. Biol. 2010, 608, 223–255.
(21) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A.
Development and testing of a general amber force field. J. Comput.
Chem. 2004, 25, 1157–1174.