Structural evolutions of salicylaldoximes as selective agonists for estrogen receptor β

Article abstract of DOI:10.1021/jm801458t

The bioisosteric replacement of the phenol ring, a signature functional group of most estrogen receptor (ER) ligands, with a hydrogen-bonded pseudocyclic ring, led to the development of a novel class of nonsteroidal ER-ligands based on a salicylaldoxime t

Full text of DOI:10.1021/jm801458t

858
J. Med. Chem. 2009, 52, 858–867
Structural Evolutions of Salicylaldoximes as Selective Agonists for Estrogen Receptor ꢀ
Filippo Minutolo,*,† Simone Bertini,† Carlotta Granchi,† Teresa Marchitiello,† Giovanni Prota,† Simona Rapposelli,†
Tiziano Tuccinardi,† Adriano Martinelli,† Jillian R. Gunther,‡ Kathryn E. Carlson,‡ John A. Katzenellenbogen,‡ and
Marco Macchia†
Dipartimento di Scienze Farmaceutiche, UniVersita` di Pisa, Via Bonanno 6, 56126 Pisa, Italy, Department of Chemistry, UniVersity of Illinois,
600 South Mathews AVenue, Urbana, Illinois 61801
ReceiVed NoVember 19, 2008
The bioisosteric replacement of the phenol ring, a signature functional group of most estrogen receptor
(ER) ligands, with a hydrogen-bonded pseudocyclic ring, led to the development of a novel class of
nonsteroidal ER-ligands based on a salicylaldoxime template. A series of structural modifications were applied
to selected molecules belonging to the monoaryl-salicylaldoxime chemical class in an attempt to improve
further their ERꢀ-selective receptor affinity and agonist properties. Among several modifications, the best
results were obtained by the simultaneous introduction of a meta-fluorine atom into the para-hydroxyphenyl
substituent present in the 4-position of salicylaldoxime, together with the insertion of a chloro group in the
3-position of the central scaffold. The resulting compound showed the best affinity (Ki ) 7.1 nM) and
selectivity for ERꢀ over ERR. Moreover, in transcription assays, it proved to be a selective and potent
ERꢀ-full agonist with an EC50 of 4.8 nM.
Introduction
analysis of the ligand binding cavities of the two subtypes
reveals only minor differences. In fact, there are only two amino
acid substitutions: Leu384 and Met421 of ERR are respectively
replaced by Met336 and Ile373 in ERꢀ.2 Overall, the ERꢀ
binding pocket has a smaller volume than that of ERR and there
are also slight differences in the shape of these cavities due to
amino acid residues lining the cavity borders. This high level
of similarity between the ERR and ERꢀ binding pockets has
made the development of highly subtype-selective, ERꢀ agonists
particularly challenging. Nevertheless, the number of novel
chemical classes that demonstrate selective affinity for ERꢀ
continues to increase.
The effects of estrogen hormones are mediated by the
estrogen receptor subtypes alpha (ERR) and beta (ERꢀa), which
are ligand-regulated transcription factors.1 Since the discovery
of the ERꢀ subtype,2 studies have pointed out the existence of
up to five different ERꢀ isoforms (ERꢀ1-5) that arise from
alternative splicing of the last exon coding for ERꢀ.3,4 However,
the originally cloned 59 kDa ERꢀ1 isoform is the only fully
functional isoform and, therefore, is the isoform referred to
simply as ERꢀ.3,4
For many years, ERR was considered the only target for
estrogen-related therapies, which were based on either the
agonist properties of certain drugs, such as ones for hormone
replacement therapy (HRT), osteoporosis, and hot flushes, or
on antagonist properties such as those exhibited by antitumor
agents effective against hormone-sensitive breast cancer. In
many cases, the use of selective estrogen receptor modulators
(SERMs) has offered advantages for both types of therapies.5,6
The more recently discovered receptor, ERꢀ, has inspired
research efforts devoted to distinguishing the biological effects
of estrogen action through each of the two ER-subtypes. Some
ERꢀ-selective agonists have become clinical candidates in
several therapeutic areas such as for the treatment of prostate
hyperplasia and cancer,7 bone loss,8 arthritis, and intestinal
inflammation.9 Notably, the beneficial effects resulting from
selective ERꢀ-stimulation are free from the undesired ERR-
mediated proliferative effects on breast and uterus, and this
constitutes a great advantage for the prospective therapeutic use
of such drugs.
In recent years, we have developed several diaryl-substituted
salicylaldoxime10,11 and anthranylaldoxime12,13 derivatives that
are able to bind to the ERs with high affinity. These compounds
contain a six-membered pseudocycle, formed by an intramo-
lecular H-bond, that was designed to isosterically replace the
steroid phenolic A ring, which is required in ER ligands.14 In
spite of the good binding affinity shown by some compounds
in these classes, a satisfactory level of subtype binding selectivity
was never obtained.
Later, an ERꢀ pharmacophoric model,15 based on a structural
analysis of many nonsteroidal ERꢀ-selective ligands, inspired
the development of our initial series of monoaryl-substituted
salicylaldoximes,16 whose simplest member (1) is shown in
Figure 1. Compound 1 showed a good level of ERꢀ selectivity
in the receptor binding assay, although its absolute affinity was
not as high as desired. We then introduced various substituents
into the 3-position of 1, such as a chlorine atom, a methyl, or
a cyano group. Binding assays showed that the 3-chloro-
substituted derivative 2 (Figure 1) possessed the best properties
in terms of ERꢀ-binding affinity and ꢀ-subtype selectivity.16
Transcriptional assays showed that compound 2 is a partial
agonist on ERꢀ, but unfortunately, its functional activity is much
less ERꢀ selective than its binding affinity.16
The amino acid sequence of ERR and ERꢀ show 59% amino
acid sequence identity in the ligand binding domain (LBD), but
* To whom correspondence should be addressed. Phone: +39 050
2219557. Fax: +39 050 2219605. E-mail: minutolo@farm.unipi.it.
†
Dipartimento di Scienze Farmaceutiche, Universita` di Pisa.
Department of Chemistry, University of Illinois.
‡
a Abbreviations: ER, estrogen receptor; ERR, estrogen receptor subtype
alpha; ERꢀ, estrogen receptor subtype ꢀ; HRT, hormone replacement
therapy; SERMs, selective estrogen receptor modulators; LBD, ligand
binding domain; RBA, relative binding affinity; RTP, relative transcriptional
potency; TLC, thin-layer chromatography; CG, conjugated gradient.
For this reason, we started to consider several possible
molecular variations that could afford new monoaryl-subsituted
salicylaldoximes possessing good ERꢀ-binding affinity and
selectivity, like that of compound 2, but would at the same time
10.1021/jm801458t CCC: $40.75
2009 American Chemical Society
Published on Web 01/07/2009

Salicylaldoximes as ERꢀ Agonists
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 859
Figure 1. Compounds 2-12 deriving from structural evolutions (a-d) of monoaryl-substituted salicylaldoxime 1.
Scheme 1. Synthesis of Salicylaldoximes 3 and 4a
have improved selectivity as ERꢀ-agonists in transcriptional
activity assays. To accomplish this, we planned a series of
structural evolutions that we applied to the simplest member of
this class of compounds (1); these can be summarized as follows
(Figure 1): (a) introduction of a chlorine atom into the 3-position
of the central scaffold (already exploited for the development
of 2,16 now also combined with other modifications), (b)
exchange of the relative positions of the phenol OH and oxime
groups, (c) introduction of a meta-fluorine atom into the
peripheral 4-aryl substituent, and (d) introduction of a methyl
group on the oxime carbon atom. The choice of the combina-
tions of the various structural modifications was mainly dictated
by the synthetic accessibility of the resulting target molecules.
These structural evolutions have led to the development of new
hydroxyphenyl-substituted oximes (RdH: 3, 5, 7, 9, and 11)
and their O-methylated counterparts (RdCH3: 4, 6, 8, 10, and
12), which were synthesized and submitted to biological assays
to determine whether any of these modifications would be
beneficial in terms of selective ERꢀ-ligand binding and agonist
properties.
a Key: (a) 3-F-4-MeO-C6H4B(OH)2, Pd(OAc)2, PPh3, aqueous 2 M
Na2CO3, 1:1 toluene/EtOH, 100 °C, 16 h; (b) OsO4, NaIO4, dioxane-H2O,
2 h; (c) BBr3, CH2Cl2, -78 to 0 °C, 1 h; (d) NH2OH·HCl, MeOH-H2O, 50
°C, 5 h.
Results and Discussion
of osmium tetroxide. The resulting salicylaldehyde (15) was
treated with boron tribromide to obtain O-demethylated aldehyde
16, which was then condensed with hydroxylamine hydrochlo-
ride, thus yielding the final oxime 3. Methoxy-substituted
salicylaldoxime 4 was obtained by direct reaction of aldehyde
15 with hydroxylamine hydrochloride.
Salicylaldoximes 5-8, derived by the formal exchange of
the positions of the OH and CHO groups, were synthesized as
shown in Scheme 2. Commercially available 5-bromosalicyla-
ldehyde (17) was submitted to a Pd-catalyzed cross coupling
reaction with either 4-methoxyphenylboronic acid to give biaryl
derivative 18, or with 3-fluoro-4-methoxyphenylboronic acid,
for the preparation of analogue 19. Subsequent O-demethylation
with BBr3 afforded free phenols 20 from 18, and 21 from 19.
Final condensation with hydroxylamine hydrochloride yielded
final oximes 5 and 7. The preparation of O-methylated oximes
6 and 8 was obtained by direct condensation of O-Me-substituted
aldehydes 18 and 19, respectively, with NH2OH·HCl.
Synthetic Chemistry. The synthesis of 3-chloro-substituted
salicylaldoximes bearing a meta-fluorine atom on the 4-aryl
substituent (3 and 4) followed a synthetic path similar to one
we had used for the synthesis of oxime 2 (Scheme 1).16 A 9:1
E/Z-diastereoisomeric mixture of 3-bromo-2-chloro-6-(prop-1-
enyl)phenol (13), prepared in three steps from 3-bromo-2-
chlorophenol as previously reported,16 was submitted to palla-
dium-catalyzed cross coupling with 3-fluoro-4-methoxy-
phenylboronic acid. In this case, the use of classic Suzuki
conditions, namely, in situ formation of Pd(PPh3)4 by reaction
of palladium acetate with a 5-fold excess of triphenylphosphine
with an aqueous base and conventional heating at 100 °C
overnight, guaranteed the chemoselective replacement of the
bromine atom only, with the chloro group remaining intact.
Subsequent oxidative cleavage of the olefin double bond of the
resulting biphenyl derivative 14 was achieved with two equiva-
lents of sodium periodate in the presence of catalytic amounts

860 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Scheme 2. Synthesis of Salicylaldoximes5-8a
Minutolo et al.
Table 1. Relative Binding Affinitiesa of Compounds 1-12 for the
Estrogen Receptors R and ꢀ
ligand
hERR (%)
hERꢀ (%)
ꢀ/R ratio
estradiol
(100)
(100)
1
79
65
62
1.8
41
3.7
46
1.9
10
1b
2b
3
0.007 ( 0.001
0.065 ( 0.016
0.114 ( 0.030
0.006 ( 0.001
0.064 ( 0.016
0.003 ( 0.001
0.021 ( 0.001
0.007 ( 0.001
0.012 ( 0.004
<0.001
0.553 ( 0.110
4.21 ( 0.66
7.01 ( 1.00
4
5
6
7
8
9
10
11
12
0.011 ( 0.001
2.64 ( 0.62
0.011 ( 0.003
0.970 ( 0.110
0.013 ( 0.001
0.123 ( 0.030
0.002 ( 0.001
0.077 ( 0.010
0.002 ( 0.001
0.006 ( 0.001
<0.001
13
a Determined by a competitive radiometric binding assay with [3H]es-
tradiol; preparations of purified, full-length human ERR and ERꢀ (PanVera)
were used; see Experimental Section. Values are reported as the mean (
the range or SD of 2 or more independent experiments; the Kd for estradiol
for ERR is 0.2 nM and for ERꢀ is 0.5 nM. Ki values for the new compounds
can be readily calculated by using the formula: Ki ) (Kd[estradiol]/RBA)
× 100. b See ref 16.
a Key: (a) 4-MeO-C6H4B(OH)2 or 3-F-4-MeO-C6H4B(OH)2, Pd(OAc)2,
PPh3, aqueous 2 M Na2CO3, 1:1 toluene/EtOH, 100 °C, 16 h; (b)
NH2OH·HCl, MeOH-H2O, 50 °C, 5 h; (c) BBr3, CH2Cl2, -78 to 0 °C,
1 h.
the final ketoximes 9 and 11. Treatment of methoxy-substituted
ketones 25 and 26 afforded oximes 10 and 12.
Scheme 3. Synthesis Salicylaldoximes 9-12a
All the aldoximes (3-8) were obtained as single E-diaste-
reoisomers because only these isomers can form the stabilizing
intramolecular hydrogen bonds. This selectivity had already been
verified for other oxime analogues previously reported,10-13,16
1
and it was confirmed by the H NMR chemical shift value (δ)
of the oxime proton, which is always found downfield from 8
ppm (δ g 8.45 ppm, see Experimental Section).18 The E-
configuration of ketoximes 9-12 was correlated with the
chemical shift of the methyl protons, which is in the range of
2.35-2.41 ppm for these four compounds; by contrast, ke-
toximes of acetophenones with Z-configurations show chemical
shifts for the methyl protons that are less than 2.0 ppm.19 Further
evidence comes from the 13C NMR spectra, which showed δ
values of the methyl carbon atoms in the range of 10.78-11.00
ppm (δ ) 11.00 (9), 10.94 (10), 10.78 (11), 10.78 (12) ppm),
consistent with values found for E-diastereoisomers of aromatic
ketoximes; the Z isomers typically have chemical shifts around
21-22 ppm.19
Estrogen Receptor Binding Assays. ERR and ERꢀ binding
affinities of oximes 3-12 were determined by a radiometric
competitive binding assay, using methods that have been
previously described.20,21 The relative binding affinity (RBA)
values for the newly reported compounds, together with those
previously obtained for compounds 1 and 2,16 are summarized
in Table 1. RBA values are reported as percentages (%) of that
of estradiol, which is set at 100%.
a Key: (a) (i) NaOH, nBu4N+HSO4- (cat.), dioxane, (ii) acetyl chloride,
dioxane, RT, 1 h; (b) AlCl3, 130 °C, 4 h neat; (c) 4-MeO-C6H4B(OH)2 or
3-F-4-MeO-C6H4B(OH)2, Pd(OAc)2, PPh3, aqueous 2 M Na2CO3, 1:1
toluene/EtOH, 100 °C, 16 h; (d) NH2OH·HCl, MeOH-H2O, 50 °C, 5 h;
(e) BBr3, CH2Cl2, -78 to 0 °C, 1 h.
As we had already experienced in the past,16 and in
accordance with the pharmacophore model (Figure 1),15 para-
methoxyphenyl-substituted oximes (4, 6, 8, 10, and 12) did not
show appreciable levels of affinity for either receptor subtype.
Among the para-hydroxyphenyl-substituted compounds, the
simplest member of this class (compound 1) had already shown
a remarkable level of ERꢀ-selectivity (ERꢀ/R ) 79), although
its absolute affinity for ERꢀ was low (RBA ) 0.55%).16 The
first structural evolution (type a, Figure 1) applied to compound
1, was the introduction of a chlorine atom into the 3-position
of the central scaffold, leading to compound 2. This caused an
8-fold increase in binding affinity and, at the same time,
maintained a good level of ERꢀ-selectivity (ERꢀ/R ) 65).16
Oxime 5, derived by exchange of the relative positions of the
phenol and oxime groups (structural modification type b),
showed a 5-fold increase in ERꢀ-binding affinity when com-
The synthesis of ketoximes 9-12 starts from 3-bromo-2-
chlorophenol 22, which was prepared following a previously
reported route (Scheme 3).17 Phenol 22 was esterified with acetyl
chloride under phase-transfer conditions in the presence of base
and, upon heating to 130 °C with aluminum trichloride, the
resulting acetate 23 underwent a Fries rearrangement to give
the ortho-acetyl compound 24. Biaryl derivatives 25 and 26
were obtained from o-acetylphenol 24 by a Pd-catalyzed cross-
coupling reaction with 4-methoxyphenylboronic or 3-fluoro-4-
methoxyphenylboronic acid, respectively. BBr3-promoted O-
demethylation afforded compounds 27 and 28, which were
treated with hydroxylamine hydrochloride to give, respectively,

Salicylaldoximes as ERꢀ Agonists
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 861
Figure 2. Dose-response curves for transcriptional activation by compound 3 (A), compound 5 (B), and estradiol (C) through ERꢀ (dashed line)
and ERR (solid line). Human endometrial cancer (HEC-1) cells were transfected with expression vectors for ERR or ERꢀ and an (ERE)2-pS2-luc
reporter gene and were treated with estradiol, compound 3, or compound 5 at the concentrations indicated. Luciferase activity was expressed
relative to ꢀ-galactosidase activity from an internal control plasmid. The maximal activity with 1 nM E2 was set at 100%. Values are the mean of
duplicate determinations. EC50 values give absolute potencies. The ERꢀ/ERR relative transcriptional potency (RTP(ꢀ/R)) ratios are calculated as
explained in the text.
pared to 1, together with the persistence of a notable level of
ERꢀ-selectivity (ERꢀ/R ) 41). Oximes 3 and 7 were obtained
by modification type c, namely, the introduction of a meta-
fluorine atom into the 4-hydroxyphenyl substituent on com-
pounds 2 and 5, respectively. This kind of substituent is also
present in one of the best ERꢀ-selective ligands, ERB-041,22a,b
which is being studied clinically for the treatment of various
kinds of inflammatory pathologies, including rheumatoid ar-
thritis. Moreover, a beneficial effect of the m-F group on ERꢀ-
affinity was also observed on biphenylcarbaldehyde oxime
derivatives previously reported.22c This modification, however,
was not very effective in improving the biological properties
of compound 5 because the resulting compound 7 showed a
drop (almost 3-fold) in ERꢀ-binding affinity. On the other hand,
the same modification on compound 2 leading to oxime 3
resulted in an increase in ERꢀ binding affinity; this reached
the best value in this series of compounds, a RBA value of
7.0% relative to estradiol, which corresponds to an absolute
binding affinity (Ki) of 7.1 nM. Fortunately, a high level of
subtype-selectivity was also preserved (ERꢀ/R ) 62). Finally,
the last kind of modification (type d), which introduces a methyl
group into the oxime carbon atom of compound 2, leading to
product 9, or of compound 3, leading to 11, did not have any
positive effect in terms of ERꢀ-binding affinity and selectivity.
Transcription Assays. Compounds 3 and 5, showing the
highest binding affinity and selectivity for ERꢀ, were submitted
to further biological testing to assess their pharmacological
character. They were assayed for transcriptional activity through
both receptor subtypes, together with estradiol for reference.
Reporter gene transfection assays were conducted in human
endometrial (HEC-1) cells, using expression plasmids for either
full-length human ERR or ERꢀ and an estrogen-responsive
luciferase reporter gene system.23
In our previous studies, compound 2 proved to be an ERꢀ
partial agonist in these assays, having a maximum activity of
60% relative to estradiol and an EC50 value of 11 nM. However,
it also efficiently stimulated ERR, with an EC50 of 26 nM. This
resulted in low ERꢀ-selectivity in terms of transcriptional
potency (ERꢀ/R ) 2.4) compared to the selectivity of its binding
affinity (ERꢀ/R ) 65).16 It should be noted that reduced ERꢀ-
selectivity in terms of transcriptional potency vs binding affinity
may be ascribed to differences in the manner in which the ERR-
and ERꢀ-ligand complexes interact with various cellular co-
regulators, which can act as modulators of ligand potency.
nounced agonist properties in these assays, with a maximum
ERꢀ-activation of g85% when compared to estradiol (Figure
2A); furthermore, 3 proved to be more potent than 2, with an
EC50 of 4.8 nM, a value close to its absolute affinity Ki (7.1
nM) found in the binding assay (section above). This compound
also had a better ꢀ-selective profile than 2, because its activation
of ERR has an EC50 value of 19 nM. Compound 5 was also a
potent ERꢀ-agonist, with an EC50 of 10 nM and a maximum
activation approaching that of estradiol (∼100%), although the
ERꢀ selectivity level for this compound was lower than for 3
(Figure 2B).
To facilitate comparisons of the ER subtype transcriptional
potencies of our compounds with their ER subtype binding
affinities, we converted the EC50 values from the transcription
assays to relative transcriptional potency (RTP) values, which
(ligand)
were calculated as RTP ) EC50(estradiol)/EC50
× 100 (RTP,
estradiol ) 100). The RTPs provide a measure of transcriptional
potency relative to that of estradiol and thus are appropriate to
compare with their binding affinities, which are also measured
relative to estradiol by the competitive radiometric binding
assays. Estradiol has a 2.5-fold preference in favor of ERR in
terms of binding (Kd [ERR] ) 0.2 nM vs [ERꢀ] ) 0.5 nM) and
a 5.4-fold preference in terms of transcriptional potency (EC50
[ERR] ) 0.13 nM vs [ERꢀ] ) 0.70 nM). By these metrics,
compound 3 has an RBA(ꢀ/R) ratio of 62 and an RTP(ꢀ/R)
ratio of 21.3, and compound 5 has an RBA(ꢀ/R) ratio of 41
and an RTP(ꢀ/R) ratio of 9.1. Thus, measured relative to
estradiol, much of the ERꢀ affinity preference of these
compounds is, indeed, preserved in their ERꢀ transcriptional
potency preference.
Molecular Modeling. Docking studies were undertaken to
explore the interaction of the newly synthesized compounds with
the ERR and ERꢀ ligand binding sites. Figure 3A shows the
docking of compound 3 into ERꢀ, and it shows a binding mode
very similar to that already reported for 4′-hydroxybiphenyl-
carbaldehyde oxime derivatives,22c and for compound 2,16
a
compound from which it differs by only a fluorine atom in
position 3′. In this model, the pseudocycle/oxime system forms
H-bonds with H475 and the backbone carbonyl of G472; the
chlorine atom is inserted into the lipophilic pocket delimited
by A302, W335, M336, and L339, while the phenol OH
substituent is involved in a H-bond network that includes E305,
R346, and a water molecule. The presence of the fluorine
substituent results in an increase of ERꢀ affinity, but the docking
studies do not reveal any important interactions for this group.
However, as reported by Malamas and co-workers in their
Newly synthesized compound 3, which differs from 2 by the
presence of an additional F-atom, showed much more pro-

862 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Minutolo et al.
Figure 3. Docking analysis of ligands into ERꢀ and ERR: (A) docking of 3 (magenta) ERꢀ, (B) docking of 4 (cyan) into ERꢀ, (C) docking of 11
(orange) into ERꢀ, (D) docking of 5 (green) into ERꢀ, (E) docking of 3 and 11 into ERR, (F) docking of 5 into ERR.
analysis of ERB-041 binding,22b the repulsion between the
fluorine atom of compound 3 and the carbonyl oxygen of L339
results in a shift of the molecule by about 0.7 Å; this allows a
stronger interaction of the pseudocycle/oxime system with G472
and H475, while the greater distance between the phenol OH
group and the E305-R346-water system is balanced by the
higher polarization of the phenol OH group due to the fluorine
substituent.
The substitution of the phenol OH with a methoxy group
results in about a 600-fold reduction in ERꢀ affinity (compound
4). As shown in Figure 3B, this substitution causes the ligand
to invert its position in the binding pocket with respect to
compound 3, leading to the loss of the interactions with G472
and H475.
The introduction of a methyl group into the oxime carbon
atom of compound 3 results in about a 90-fold reduction in
ERꢀ affinity (compound 11). We have already reported an
analysis of the cavities surrounding reference compound 2 in
the ERꢀ binding site,16 and this analysis indicated the presence
of two empty voids around the central aromatic ring, while no
space was found around the pseudocycle/oxime unit. This was
confirmed by a docking analysis of 11, as shown in Figure 3C.
Because of the lack of any void in this region, addition of the
methyl substituent onto the oxime carbon atom results in an
inversion of ligand orientation with respect to compound 3. In
this new orientation, a H-bond forms between the phenol OH
substituent and H475, but there is a loss of all the interactions
between the ligand and the E305-R346-water system.
Finally, compound 5, in which there has been an exchange
on the relative positions of the phenol and oxime groups in
compound 1, experiences about a 5-fold increase in ERꢀ-binding
affinity. The docking results showed that the pseudocycle/oxime
system was able to interact with the E305-R346-water system,
whereas the shift of the p-phenol ring from C4 to C5 results in
a loss of the H-bond with H475 and G472, but formation of a
new H-bond with T299 (Figure 3D). This different interaction
may explain the good ERꢀ binding affinity of this compound,
and it is also consistent with the lack of beneficial effects by
the introduction of a 3′-fluoro substituent (cf. 5 and 7) that was
found in going from compound 2 to 3. It is of note that the
strong interaction between the 4′-hydroxyl of 5 and T299 is
possible only with ERꢀ because only in this ER subtype is there
enough space for the phenol group in the area close to M336.
Thus, the phenol OH group comes in close proximity to the
alcohol group of T299; the same does not happen in ERR, where
the methionine (M336 in ERꢀ) is replaced by a bulkier leucine
(L384 in ERR), causing a completely different disposition of
compound 5 in this receptor subtype (Figure 3F).
In general, all the compounds tested proved to be weaker
ligands with ERR. As already reported,16 the compounds having
a Cl substituent in the central aromatic ring, such as compound
3, show, in ERR, the phenol OH interacting with H524 and the
pseudocycle/oxime system forming H-bonds with the E353-
R394-water system (see 3 and 11, Figure 3E). As shown in
Figure 3A, the Cl-substituent of 3 in ERꢀ is able to interact
with the lipophilic pocket delimited by A302, W335, M336,
and L339, whereas in ERR the replacement of M336 by L384
restricts the size of this cavity. This steric restriction causes
compound 3 to bind to ERR in a reversed orientation and with
lower affinity.
As for compound 5, characterized by exchange of the relative
positions of the phenol and oxime groups (Figure 3F), its
phenolic OH interacts in the ERR binding site with the
E305-R346-water system, while the pseudocycle/oxime sys-

Salicylaldoximes as ERꢀ Agonists
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 863
Evaporation was performed in vacuo (rotating evaporator). Sodium
sulfate was always used as the drying agent. Microwave assisted
reactions were run in a CEM microwave synthesizer.
tem does not show any important interactions, confirming its
low affinity for this subtype, and thus its high beta-selectivity.
(E/Z)-2-Chloro-3-(3-fluoro-4-methoxyphenyl)-6-(prop-1-enyl)phe-
nol (14). A solution of Pd(OAc)2 (9 mg, 0.04 mmol) and triph-
enylphosphine (54 mg, 0.20 mmol) in ethanol (1.7 mL) and toluene
(1.7 mL) was stirred at RT under nitrogen for 10 min. After that
period, compound 13 (337 mg, 1.36 mmol, 9:1 E/Z diastereomeric
mixture),16 1.7 mL of an aqueous 2 M solution of Na2CO3, and
3-fluoro-4-methoxyphenylboronic acid (309 mg, 1.76 mmol) were
sequentially added. The resulting mixture was heated at 100 °C in
a sealed vial under nitrogen overnight. After being cooled to RT,
the mixture was diluted with water and extracted with EtOAc. The
combined organic phase were dried and concentrated. The crude
product was purified by flash chromatography (n-hexane/ EtOAc
9:1) to yield 14 as a 9:1 E/Z diastereomeric mixture (352 mg, 88%
yield). 1H NMR (CDCl3, 90:10 E/Z mixture, asterisk denotes minor
isomer peaks) δ (ppm): 1.86* (dd, 3H, J ) 7.2, 1.8 Hz), 1.94 (dd,
3H, J ) 6.6, 1.6 Hz), 3.93* (s, 3H), 3.94 (s, 3H), 5.93 (bs, 1H),
6.33 (dq, 1H, J ) 15.9, 6.6 Hz), 6.70 (dq, 1H, J ) 15.7, 1.5 Hz),
6.84 (d, 1H, J ) 8.1 Hz), 7.01 (t, 1H, J ) 8.6 Hz), 7.12-7.23 (m,
2H), 7.32 (d, 1H, J ) 8.1 Hz).
Conclusions
Several structural evolutions were effected within a series of
monoaryl-substituted salicylaldoximes, starting from the simple
member of this class, 3-(4-hydroxyphenyl)salicylaldoxime 1.
The first modification consisted of introduction of a chlorine
atom on C(3) of the central scaffold, producing compound (2),
which previously showed a very high affinity for the ERꢀ-
subtype and a high binding selectivity (ꢀ/R ratio 65), although
the same compound showed comparable levels of activation of
the two receptor subtypes in transcriptional assays. In this work,
we extended the types of structural modifications and found
that exchange of the positions of the -OH and -CHdNOH
groups constituting the hydrogen-bonded salicylaldoxime unit,
produced a compound (5) with better ERꢀ-affinity than parent
compound 1, although some ꢀ potency selectivity was lost in
gene transcription assays. The best combination of the trans-
formations was found with the simultaneous introduction of a
Cl group in position 3 and a fluorine atom in the meta-position
of the 4-aryl-substituent, affording compound 3, which showed
the highest selective ꢀ affinity (Ki ) 7.1 nM) and still preserved
a good level of ERꢀ selectivity in functional assays, where it
proved to be a partial agonist with high intrinsic activity on
ERꢀ, with a EC50 of 4.8 nM. Docking studies highlighted the
characteristics of compounds 3 and 5, which are important for
their high ERꢀ affinity and ERꢀ/ERR selectivity. Compound 5
is of particular interest because seems to be able to participate
to an unprecedented H-bond with a threonine residue of ERꢀ.
This interaction may be further exploited in the design of more
potent ERꢀ-ligands. Finally, although aldoximes are generally
considered to be quite unstable, and to readily undergo hy-
drolysis, especially under acidic conditions or in biological
systems, this is not true for aromatic oximes, whose aqueous
hydrolysis rate in buffer solutions is extremely slow even at
markedly acidic pH values.24 As a matter of fact, the most
important metabolic pathway of this kind of compounds was
shown to be a slow oxidative (and not hydrolytic) cleavage of
the oxime portion in liver microsomes, which liberates NO.24
In our laboratories, we have verified that chemical stability of
our final salicylaldoximes in buffer (phosphate buffered saline,
pH 7.2) is practically unlimited (>4 weeks). Moreover, cellular
assays did not give any indication ascribable to biological
instability of these compounds, although more precise metabolic
studies are unquestionably necessary. Studies are underway to
examine other kinds of structural evolutions, as well as other
possible combinations of the modifications applied in the present
work, to discover compounds showing improvements in their
ERꢀ-selective agonist potency in gene transcription assays.
3-Chloro-4-(3-fluoro-4-methoxyphenyl)salicylaldehyde (15). A
solution of 14 (270 mg, 0.92 mmol) in dioxane (9 mL) was treated
with 4.5 mL of water, 449 g of sodium periodate (2.10 mmol), and
0.3 mL of a 2.5% solution of osmium tetroxide in tert-butanol (0.02
mmol), and the mixture was stirred at RT for 2 h. Most of dioxane
was then removed under vacuum and the mixture was diluted with
water and extracted with chloroform. The organic phase was dried
and evaporated to afford a crude residue that was purified by flash
chromatography (n-hexane/EtOAc 9:1) to yield pure 15 (248 mg,
1
96% yield) as a yellow solid. H NMR (CDCl3) δ (ppm): 3.96 (s,
3H), 7.02 (d, 1H, J ) 8.1 Hz), 7.03-7.09 (m, 1H), 7.19-7.29 (m,
2H), 7.54 (d, 1H, J ) 7.9 Hz), 9.92 (s, 1H), 11.73 (s, 1H).
3-Chloro-4-(3-fluoro-4-hydroxyphenyl)salicylaldehyde (16). A
solution of 15 (170 mg, 0.60 mmol) in anhydrous dichloromethane
(7 mL) was cooled to -78 °C and treated dropwise with a 1 M
solution of BBr3 in dichloromethane (1.9 mL), and the resulting
solution was stirred at the same temperature for 5 min and at 0 °C
for 1 h. The mixture was then diluted with water and extracted
with ethyl acetate. The organic phase was dried and concentrated.
The crude product was purified by flash chromatography (n-hexane/
ethyl acetate 7:3) to yield pure 16 (117 mg, 73% yield) as a yellow
solid. 1H NMR (acetone-d6) δ (ppm): 7.12-7.26 (m, 3H), 7.35 (dd,
1H, J ) 12.1, 1.8 Hz), 7.87 (d, 1H, J ) 8.1 Hz), 9.09 (d, 1H, J )
1.5 Hz, exchangeable), 10.11 (s, 1H), 11.78 (s, 1H, exchangeable).
(E)-3-Chloro-4-(3-fluoro-4-hydroxyphenyl)salicylaldoxime (3). A
solution of 16 (90 mg, 0.34 mmol) in ethanol (6.5 mL) was treated
with a solution of hydroxylamine hydrochloride (47.3 mg, 0.68
mmol) in water (1.5 mL), and the mixture was heated to 50 °C for
1 h. After being cooled to RT, part of the solvent was removed
under vacuum, and the mixture was diluted with water and extracted
with EtOAc. The organic phase was dried and evaporated to afford
a crude residue that was purified by column chromatography (n-
hexane/ethyl acetate 7:3) to yield pure 3 (76.8 mg, 80% yield) as
a white solid. 1H NMR (acetone-d6) δ (ppm): 7.02 (d, 1H, J ) 8.1
Hz), 7.12 (t, 1H, J ) 8.2 Hz), 7.18 (dd, 1H, J ) 8.2, 2.0 Hz), 7.28
(dd, 1H, J ) 11.7, 2.0 Hz), 7.44 (d, 1H, J ) 8.1 Hz), 8.48 (s, 1H),
8.93 (bs, 1H), 10.94 (s, 1H), 11.00 (bs, 1H). 13C NMR (acetone-
d6) δ (ppm): 114.73, 117.83 (d, J ) 19.2 Hz), 118.24 (d, J ) 3.7
Hz), 120.19, 122.52, 126.56 (d, J ) 3.7 Hz), 129.28, 131.84, 142.31,
145.52 (d, J ) 12.8 Hz), 151.78 (d, J ) 258 Hz), 151.87, 154.00.
MS m/z 281 (M+, 100), 263 (M+ sH2O, 55). Anal. (C13H9ClFNO3)
C, H, N.
Experimental Section
Chemistry. Melting points were determined on a Kofler hot-
stage apparatus and are uncorrected. NMR spectra were obtained
with a Varian Gemini 200 MHz spectrometer. Chemical shifts (δ)
are reported in parts per million downfield from tetramethylsilane
and referenced from solvent references. Electron impact (EI, 70
eV) mass spectra were obtained on a HP-5988A mass spectrometer.
Where indicated, the elemental compositions of the compounds
agreed to within (0.4% of the calculated value. Chromatographic
separations were performed on silica gel columns by flash (Kieselgel
40, 0.040-0.063 mm; Merck) or gravity column (Kieselgel 60,
0.063-0.200 mm; Merck) chromatography. Reactions were fol-
lowed by thin-layer chromatography (TLC) on Merck aluminum
silica gel (60 F254) sheets that were visualized under a UV lamp.
(E)-3-Chloro-4-(3-fluoro-4-methoxyphenyl)salicylaldoxime (4).
Compound 15 (100 mg, 0.36 mmol) was submitted to the same
procedure described above for the preparation of 3. The crude
product was purified by flash chromatography (n-hexane/ethyl
acetate 7:3) to produce pure 4 (68 mg, 62% yield) as a white solid.
1H NMR (acetone-d6) δ (ppm): 3.99 (s, 3H), 7.02 (d, 1H, J ) 7.9
Hz), 7.26-7.35 (m, 3H), 7.45 (d, 1H, J ) 8.1 Hz), 8.49 (s, 1H),

864 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Minutolo et al.
10.95 (s, 1H), 11.02 (s, 1H). 13C NMR (acetone-d6) δ (ppm): 56.52,
114.07, 117.69 (d, J ) 19.2 Hz), 118.10 (d, J ) 4.7 Hz), 120.19,
122.48, 126.39 (d, J ) 2.7 Hz), 129.29, 132.61, 142.08, 148.31 (d,
J ) 10.1 Hz), 151.85, 152.30 (d, J ) 244 Hz), 154.33. MS m/z
295 (M+, 100), 277 (M+ -H2O, 20). Anal. (C14H11ClFNO3) H, N.
C: calcd, 56.87; found, 56.37.
(E)-5-(3-Fluoro-4-hydroxyphenyl)salicylaldoxime (7). Com-
pound 21 (190 mg, 0.82 mmol) was submitted to the same
procedure described above for the preparation of 3. The crude
product was purified by flash chromatography (n-hexane/ethyl
acetate 6:4) to produce pure 7 (162 mg, 81% yield) as a white
1
solid. H NMR (acetone-d6) δ (ppm): 6.97 (d, 1H, J ) 8.6 Hz),
5-(4-Methoxyphenyl)salicylaldehyde (18). Commercially avail-
able 5-bromosalicylaldehyde (17) (500 mg, 2.48 mmol) was
submitted to a cross coupling reaction with 4-methoxyphenylboronic
acid (503 mg, 3.29 mmol) following the same procedure described
above for the preparation of 14. The crude product was purified by
flash chromatography (n-hexane/ethyl acetate 7:3) to produce pure
7.00-7.10 (m, 1H), 7.25-7.33 (m, 1H), 7.38 (dt, 1H, J ) 12.5,
2.4 Hz), 7.54 (dd, 1H, J ) 8.6, 2.4 Hz), 7.65 (d, 1H, J ) 2.4 Hz),
8.45 (s, 1H), 8.68 (d, 1H, J ) 1.5 Hz, exchangeable), 10.10 (s,
1H, exchangeable), 10.74 (s, 1H, exchangeable). 13C NMR (acetone-
d6) δ (ppm): 114.57 (d, J ) 19.2 Hz), 117.49, 118.31, 118.94 (d,
J ) 2.7 Hz), 123.12 (d, J ) 3.7 Hz), 129.15, 129.46, 132.06 (d, J
) 1.8 Hz), 133.28 (d, J ) 5.5 Hz), 144.68 (d, J ) 12.8 Hz), 152.50
(d, J ) 239 Hz), 152.56, 157.42. MS m/z 247 (M+, 100), 229 (M+
-H2O, 27). Anal. (C13H10FNO3) C, H, N.
1
18 (372 mg, 66% yield) as a yellow solid. H NMR (CDCl3) δ
(ppm): 3.86 (s, 3H), 6.99 (AA′XX′, 2H, JAX ) 8.8 Hz, JAA′/XX′
)
2.3 Hz), 7.06 (d, 1H, J ) 9.2 Hz), 7.48 (AA′XX′, 2H, JAX ) 8.8
Hz, JAA′/XX′ ) 2.4 Hz), 7.71 - 7.76 (m, 2H), 9.97 (s, 1H), 10.96
(s, 1H, exchangeable).
(E)-5-(3-Fluoro-4-methoxyphenyl)salicylaldoxime (8). Com-
pound 19 (169 mg, 0.69 mmol) was submitted to the same
procedure described above for the preparation of 3. The crude
product was purified by flash chromatography (n-hexane/ethyl
acetate 7:3) to produce pure 8 (100 mg, 56% yield) as a white
5-(3-Fluoro-4-methoxyphenyl)salicylaldehyde (19). Commer-
cially available 5-bromosalicylaldehyde (17) (500 mg, 2.48 mmol)
was submitted to a cross coupling reaction with 3-fluoro-4-
methoxyphenylboronic acid (551 mg, 3.23 mmol) following the
same procedure described above for the preparation of 14. The crude
product was purified by flash chromatography (n-hexane/ethyl
acetate 7:3) to produce pure 19 (469 mg, 77% yield) as a yellow
solid. 1H NMR (CDCl3) δ (ppm): 3.93 (s, 3H), 6.96-7.13 (m, 3H),
7.21-7.33 (m, 1H), 7.68-7.75 (m, 2H), 9.97 (s, 1H), 10.99 (s,
1H, exchangeable).
1
solid. H NMR (acetone-d6) δ (ppm): 3.91 (s, 3H), 6.98 (d, 1H, J
) 8.4 Hz), 7.15-7.24 (m, 1H), 7.37-7.46 (m, 2H), 7.57 (dd, 1H,
J ) 8.6, 2.4 Hz), 7.68 (d, 1H, J ) 2.4 Hz), 8.46 (s, 1H), 10.11 (s,
1H, exchangeable), 10.74 (s, 1H, exchangeable). 13C NMR (acetone-
d6) δ (ppm): 56.53, 114.57 (d, J ) 22.9 Hz), 114.78 (d, J ) 7.3
Hz), 117.55, 118.35, 122.88 (d, J ) 3.7 Hz), 129.24, 129.53, 131.77
(d, J ) 1.8 Hz), 134.02 (d, J ) 6.4 Hz), 147.57 (d, J ) 11.0 Hz),
152.55, 153.24 (d, J ) 243 Hz), 157.57. MS m/z 261 (M+, 100),
243 (M+ -H2O, 22), 228 (M+ -H2O -CH3, 73). Anal.
(C14H12FNO3) C, H, N.
5-(4-Hydroxyphenyl)salicylaldehyde (20). Compound 18 (200
mg, 0.87 mmol) was submitted to the same procedure described
above for the preparation of 16. The crude product was purified by
flash chromatography (n-hexane/ethyl acetate 7:3) to produce pure
3-Bromo-2-chlorophenyl acetate (23). A solution of 22 (1.8 g,
8.7 mmol)17 in dioxane (13 mL) was treated with solid NaOH (0.83
g, 21 mmol) and a catalytic amount of tetrabutylammonium
hydrogen sulfate (10 mg, 0.028 mmol), and the mixture was stirred
under nitrogen at RT for 30 min. A solution of acetyl chloride (0.9
mL, ∼10 mmol) in dioxane (8 mL) was then added with a syringe
through a silicon septum. Once the addition was completed, stirring
was continued at the same temperature for 1 h. The reaction mixture
was then poured into cold (ice) water and extracted with EtOAc.
The organic phase was dried and concentrated. The crude product
was purified by flash chromatography (n-hexane/ethyl acetate 9:1)
1
20 (133 mg, 71% yield) as a yellow solid. H NMR (CDCl3) δ
(ppm): 5.23 (bs, 1H), 6.92 (AA′XX′, 2H, JAX ) 8.4 Hz, JAA′/XX′
)
2.4 Hz), 7.05 (d, 1H, J ) 9.2 Hz), 7.43 (AA′XX′, 2H, JAX ) 8.6
Hz, JAA′/XX′ ) 2.5 Hz), 7.69-7.74 (m, 2H), 9.96 (s, 1H), 10.96 (s,
1H, exchangeable).
5-(3-Fluoro-4-hydroxyphenyl)salicylaldehyde (21). Compound
19 (300 mg, 0.98 mmol) was submitted to the same procedure
described above for the preparation of 16. The crude product was
purified by flash chromatography (n-hexane/ethyl acetate 7:3) to
1
produce pure 21 (209 mg, 87% yield) as a yellow solid. H NMR
(CDCl3) δ (ppm): 5.20 (bs, 1H), 7.04-7.12 (m, 2H), 7.16-7.32
(m, 2H), 7.68-7.72 (m, 2H), 9.97 (s, 1H), 10.99 (s, 1H, exchange-
able).
1
to yield pure 23 (1.73 g, 79% yield) as a colorless oil. H NMR
(CDCl3) δ (ppm): 2.36 (s, 3H), 7.10 (d, 1H, J ) 8.1, 2.2 Hz), 7.16
(t, 1H, J ) 8.2 Hz), 7.53 (dd, 1H, J ) 7.5, 2.2 Hz).
(E)-5-(4-Hydroxyphenyl)salicylaldoxime (5). Compound 20
(110 mg, 0.51 mmol) was submitted to the same procedure
described above for the preparation of 3. The crude product was
purified by flash chromatography (n-hexane/ethyl acetate 6:4) to
6-Acetyl-3-Bromo-2-chlorophenol (24). Compound 23 (400 mg,
1.60 mmol) was treated neat with AlCl3 (278 mg, 2.08 mmol), and
the mixture was heated to 130 °C in a sealed vial for 2 h. After
cooling to RT, the crude mixture was treated with aqueous 1N HCl
and extracted with EtOAc. The organic phase was dried and
concentrated. The crude product was purified by flash chromatog-
raphy (n-hexane/ethyl acetate 9:1) to yield pure 24 (252 mg, 64%
1
produce pure 5 (73 mg, 62% yield) as a white solid. H NMR
(acetone-d6) δ (ppm): 6.91 (AA′XX′, 2H, JAX ) 8.6 Hz, JAA′/XX′
)
2.6 Hz), 6.96 (d, 1H, J ) 8.6 Hz), 7.46 (AA′XX′, 2H, JAX ) 8.4
Hz, JAA′/XX′ ) 2.7 Hz), 7.51 (dd, 1H, J ) 8.6, 2.4 Hz), 7.60 (d, 1H,
J ) 2.4 Hz), 8.38 (s, 1H), 8.45 (s, 1H, exchangeable), 10.03 (s,
1H, exchangeable), 10.70 (s, 1H, exchangeable). 13C NMR (acetone-
d6) δ (ppm): 116.27, 117.13, 118.30, 127.96, 128.22, 129.02,
131.79, 133.19, 151.43, 156.23, 157.35. MS m/z 229 (M+, 100),
211 (M+ -H2O, 40). Anal. (C13H11NO3) H, N. C: calcd, 68.11;
found, 67.49.
1
yield) as an off-white solid. H NMR (CDCl3) δ (ppm): 2.65 (s,
3H), 7.22 (d, 1H, J ) 8.6 Hz), 7.53 (d, 1H, J ) 8.6 Hz).
6-Acetyl-2-chloro-3-(4-methoxyphenyl)phenol (25). Compound
24 (300 mg, 1.40 mmol) was submitted to a cross coupling reaction
with 4-methoxyphenylboronic acid (275 mg, 1.80 mmol) following
the same procedure described above for the preparation of 14. The
crude product was purified by flash chromatography (n-hexane/
ethyl acetate 7:3) to produce pure 25 (203 mg, 52% yield) as an
(E)-5-(4-Methoxyphenyl)salicylaldoxime (6). Compound 18
(150 mg, 0.66 mmol) was submitted to the same procedure
described above for the preparation of 3. The crude product was
purified by flash chromatography (n-hexane/ethyl acetate 7:3) to
1
off-white solid. H NMR (CDCl3) δ (ppm): 2.68 (s, 3H), 3.87 (s,
3H), 6.92 (d, 1H, J ) 8.2 Hz), 6.99 (AA′XX′, 2H, JAX ) 8.8 Hz,
JAA′/XX′ ) 2.5 Hz), 7.43 (AA′XX′, 2H, JAX ) 8.8 Hz, JAA′/XX′
2.6 Hz), 7.69 (d, 1H, J ) 8.2 Hz).
)
1
produce pure 6 (128 mg, 80% yield) as a white solid. H NMR
(acetone-d6) δ (ppm): 3.83 (s, 3H), 6.97 (d, 1H, J ) 8.4 Hz), 7.00
(AA′XX′, 2H, JAX ) 8.8 Hz, JAA′/XX′ ) 2.7 Hz), 7.53 (dd, 1H, J )
8.6, 2.4 Hz), 7.55 (AA′XX′, 2H, JAX ) 8.8 Hz, JAA′/XX′ ) 2.5 Hz),
7.63 (d, 1H, J ) 2.4 Hz), 8.45 (s, 1H), 10.06 (s, 1H, exchangeable),
10.71 (s, 1H, exchangeable). 13C NMR (acetone-d6) δ (ppm): 55.57,
115.02, 117.44, 118.28, 128.16, 129.11, 129.49, 133.04, 133.32,
152.65, 157.17, 159.85. MS m/z 243 (M+, 100), 225 (M+ -H2O,
25), 210 (M+ -H2O -CH3, 45). Anal. (C14H13NO3) C, H, N.
6-Acetyl-2-chloro-3-(3-fluoro-4-methoxyphenyl)phenol (26).
Compound 23 (350 mg, 1.41 mmol) was submitted to a cross
coupling reaction with 3-fluoro-4-methoxyphenylboronic acid (312
mg, 1.82 mmol) following the same procedure described above for
the preparation of 14. The crude product was purified by flash
chromatography (n-hexane/ethyl acetate 8:2) to produce pure 26
1
(208 mg, 50% yield) as an off-white solid. H NMR (CDCl3) δ

Salicylaldoximes as ERꢀ Agonists
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 865
(ppm): 2.69 (s, 3H), 3.95 (s, 3H), 6.90 (d, 1H, J ) 8.2 Hz), 7.04
(t, 1H, J ) 8.6 Hz), 7.18-7.27 (m, 2H), 7.70 (d, 1H, J ) 8.0 Hz).
6-Acetyl-2-chloro-3-(4-hydroxyphenyl)phenol (27). Compound
25 (155 mg, 0.56 mmol) was submitted to the same procedure
described above for the preparation of 16. The crude product was
purified by flash chromatography (n-hexane/ethyl acetate 7:3) to
Biological Methods. Full-length human ERR and ERꢀ were
obtained from PanVera/Invitrogen (Carlsbad, CA). [3H] Estradiol
([3H]E2) ([2,4,6,7-3H]estra-1,3,5(10)-triene-3,17ꢀ-diol) was ob-
tained from GE Healthcare (Piscataway, NJ) with a specific
activity of 70-120 Ci/mmol. Cell culture media were purchased
from Gibco BRL (Grand Island, NY). Calf serum was obtained
from Hyclone Laboratories, Inc. (Logan, UT), and fetal calf
serum was purchased from Atlanta Biologicals (Atlanta, GA).
The expression vectors for human ERR (pCMV5-hERR) and
human ERꢀ (pCMV5-ERꢀ) were as described previously.25,26
The estrogen responsive reporter plasmid, (ERE)2-pS2-Luc, was
constructed by inserting the (ERE)2-pS2 fragment from (ERE)2-
pS2-CAT into the MluI/BglII sites of pGL3-Basic vector
(Promega, Madison, WI). The luciferase assay system was from
Promega (Madison, WI). The plasmid pCMVꢀ-gal (Clontech,
Palo Alto, CA), which contains the ꢀ-galactosidase gene, was
used as an internal control for transfection efficiency.
Estrogen Receptor Binding Assays. Relative binding affinities
were determined by competitive radiometric binding assays with 2
nM [3H]E2 as tracer, as a modification of methods previously
described.20,21 The source of ER was purified full-length human
ERR and ERꢀ purchased from Pan Vera/Invitrogen (Carlsbad, CA).
Incubations were done at 0 °C for 18-24 h, and hydroxyapatite
was used to absorb the purified receptor-ligand complexes (human
ERs).21 The binding affinities are expressed as relative binding
affinity (RBA) values, where the RBA of estradiol is 100%; under
these conditions, the Kd of estradiol for ERR is ca. 0.2 nM, and for
ERꢀ 0.5 nM. The determination of these RBA values is reproducible
in separate experiments with a CV of 0.3, and the values shown
represent the average ( range or SD of 2 or more separate
determinations.
Cell Culture and Transient Transfections. Human endometrial
cancer (HEC-1) cells were maintained in culture as described.23
Transfection of HEC-1 cells in 24-well plates used a mixture of
0.35 mL of serum-free MEM medium and 0.15 mL of HBSS
containing 5 µL of lipofectin (Life Technologies, Rockville, MD),
20 µL of transferrin (Sigma, St. Louis, MO), 0.2 µg of pCMVꢀ-
galactosidase as internal control, 0.5 µg of the reporter gene plasmid,
and 50 ng of ER expression vector. The cells were incubated at 37
°C in a 5% CO2 containing incubator for 4 h. The medium was
then replaced with fresh medium containing 5% charcoal-dextran
treated calf serum and the desired concentrations of ligands.
Reporter gene activity was assayed at 24 h after ligand addition.
Luciferase activity, normalized for the internal control ꢀ-galactosi-
dase activity, was assayed as described.23
1
produce pure 27 (113 mg, 77% yield) as an off-white solid. H
NMR (CDCl3) δ (ppm): 2.68 (s, 3H), 5.22 (bs, 1H), 6.88-6.96
(m, 3H), 7.37 (AA′XX′, 2H, JAX ) 8.8 Hz, JAA′/XX′ ) 2.5 Hz),
7.69 (d, 1H, J ) 8.2 Hz).
6-Acetyl-2-chloro-3-(3-fluoro-4-hydroxyphenyl)phenol (28).
Compound 26 (120 mg, 0.41 mmol) was submitted to the same
procedure described above for the preparation of 16. The crude
product was purified by flash chromatography (n-hexane/ethyl
acetate 8:2) to produce pure 28 (109 mg, 95% yield) as an off-
white solid. 1H NMR (CDCl3) δ (ppm): 2.69 (s, 3H), 5.35 (bs, 1H),
6.90 (d, 1H, J ) 8.2 Hz), 7.08 (t, 1H, J ) 8.1 Hz), 7.13-7.28 (m,
2H), 7.70 (d, 1H, J ) 8.4 Hz).
(E)-1-(3-Chloro-2-hydroxy-4-(4-hydroxyphenyl)phenyl)etha-
none oxime (9). Compound 27 (80 mg, 0.30 mmol) was submitted
to the same procedure described above for the preparation of 3.
The crude product was purified by flash chromatography (CH2Cl2/
MeOH 95:5) to produce pure 9 (70 mg, 84% yield) as a white
1
solid. H NMR (CD3OD) δ (ppm): 2.35 (s, 3H), 6.83 (AA′XX′,
2H, JAX ) 8.8 Hz, JAA′/XX′ ) 2.5 Hz), 6.87 (d, 1H, J ) 8.2 Hz),
7.26 (AA′XX′, 2H, JAX ) 8.8 Hz, JAA′/XX′ ) 2.5 Hz), 7.46 (d, 1H,
J ) 8.2 Hz). 13C NMR (acetone-d6) δ (ppm): 11.00, 115.58, 119.19,
120.63, 121.63, 126.54, 127.42, 131.31, 147.00, 154.80, 157.92,
158.57. MS m/z 277 (M+, 100), 259 (M+ -H2O, 41). Anal.
(C14H12ClNO3) C, H, N.
(E)-1-(3-Chloro-2-hydroxy-4-(4-methoxyphenyl)phenyl)etha-
none oxime (10). Compound 25 (59 mg, 0.21 mmol) was submitted
to the same procedure described above for the preparation of 3.
The crude product was purified by flash chromatography (n-hexane/
ethyl acetate 7:3) to produce pure 10 (41 mg, 67% yield) as a white
1
solid. H NMR (acetone-d6) δ (ppm): 2.41 (s, 3H), 3.85 (s, 3H),
6.93 (d, 1H, J ) 8.3 Hz), 7.02 (AA′XX′, 2H, JAX ) 8.8 Hz, JAA′/
XX′ ) 2.4 Hz), 7.41 (AA′XX′, 2H, JAX ) 8.8 Hz, JAA′/XX′ ) 2.3
Hz), 7.56 (d, 1H, J ) 8.2 Hz), 10.95 (s, 1H, exchangeable), 12.41
(s, 1H, exchangeable). 13C NMR (acetone-d6) δ (ppm): 10.94, 55.57,
114.24, 119.35, 120.40, 121.54, 126.54, 131.26, 132.30, 142.97,
155.08, 158.82, 160.27. MS m/z 291 (M+, 100), 273 (M+ -H2O,
30). Anal. (C15H14ClNO3) C, H, N.
(E)-1-(3-Chloro-2-hydroxy-4-(3-fluoro-4-hydroxyphenyl)phe-
nyl)ethanone oxime (11). Compound 28 (79 mg, 0.28 mmol) was
submitted to the same procedure described above for the preparation
of 3. The crude product was purified by flash chromatography
(CH2Cl2/MeOH 95:5) to produce pure 11 (46 mg, 55% yield) as a
white solid. 1H NMR (CD3OD) δ (ppm): 2.35 (s, 3H), 6.88 (d, 1H,
J ) 8.2 Hz), 6.96 (pseudo-t, 1H, J ) 8.5 Hz), 7.06 (ddd, 1H J )
8.2, 2.0, 0.6 Hz), 7.15 (dd, 1H J ) 8.6, 2.4 Hz), 7.48 (d, 1H, J )
8.4 Hz). 13C NMR (CD3OD) δ (ppm): 10.78, 117.98 (d, J ) 21.1
Hz), 118.21 (d, J ) 5.5 Hz), 120.12, 121.03, 121.83, 126.53, 126.60
(d, J ) 3.7 Hz), 132.30 (d, J ) 6.0 Hz), 142.41, 145.83 (d, J )
12.6 Hz), 152.13 (d, J ) 240 Hz), 155.14, 158.36. MS m/z 295
(M+, 100), 277 (M+ -H2O, 38). Anal. (C14H11ClFNO3) C, H. N:
calcd, 4.74; found, 4.22.
Docking Methods. The crystal structure of ERR (pdb code
2I0J27) and ERꢀ (pdb code 2I0G27) was taken from the Protein
Data Bank.28 After adding hydrogen atoms the two proteins
complexed with their reference inhibitor were minimized using
Amber 9 software29 and parm03 force field at 300 K. In detail, the
complexes were placed in a rectangular parallelepiped water box,
an explicit solvent model for water, TIP3P, was used, and the
complexes were solvated with a 10 Å water cap. Sodium ions were
added as counterions to neutralize the system. Two steps of
minimization were then carried out; in the first stage, we kept the
protein fixed with a position restraint of 500 kcal/mol/Å2 and we
solely minimized the positions of the water molecules. In the second
stage, we minimized the entire system through 5000 steps of
steepest descent followed by conjugate gradient (CG) until a
convergence of 0.05 Kcal/Å ·mol.
(E)-1-(3-Chloro-2-hydroxy-4-(3-fluoro-4-methoxyphenyl)phe-
nyl)ethanone oxime (12). Compound 26 (63 mg, 0.21 mmol) was
submitted to the same procedure described above for the preparation
of 3. The crude product was purified by flash chromatography (n-
hexane/ethyl acetate 7:3) to produce pure 12 (54 mg, 83% yield)
as a white solid. 1H NMR (CD3OD) δ (ppm): 2.35 (s, 3H), 3.92 (s,
3H), 6.89 (d, 1H, J ) 8.4 Hz), 7.10-7.23 (m, 3H), 7.49 (d, 1H, J
) 8.2. Hz). 13C NMR (CD3OD) δ (ppm): 10.78, 56.72, 114.15 (d,
J ) 1.8 Hz), 117.96 (d, J ) 19.2 Hz), 120.12, 121.01, 121.76,
126.48, 126.55, 133.41 (d, J ) 7.3 Hz), 142.06 (d, J ) 1.8 Hz),
148.63 (d, J ) 10.1 Hz), 152.81 (d, J ) 238 Hz), 155.30, 158.30.
MS m/z 309 (M+, 100), 291 (M+ -H2O, 37). Anal. (C15H13ClFNO3)
C, H, N.
The ligands were built using Maestro30 and were minimized by
means of Macromodel31 in a water environment using the CG
method until a convergence value of 0.05 kcal/Å ·mol, using the
MMFFs force field and a distance-dependent dielectric constant of
1.0.
Automated docking was carried out by means of the AU-
TODOCK 4.0 program;32 Autodock Tools33 was used in order to
identify the torsion angles in the ligands, add the solvent model,
and assign the Kollman atomic charges to the protein. The ligand
charge was calculated using the Gasteiger method. To prevent the
loss of the intramolecular H-bond of the pseudocycle/oxime system,

866 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Minutolo et al.
Estrogen Receptors R and ꢀ. Bioorg. Med. Chem. 2003, 11, 1247–
1257.
during the docking, we blocked the torsions involved in this
intramolecular bond.13 The regions of interest used by Autodock
were defined by considering SERBA-1 into both receptors as the
central group; in particular, a grid of 50, 40, and 46 points in the
x, y, and z directions was constructed centered on the center of
the mass of this compound. A grid spacing of 0.375 Å and a
distance-dependent function of the dielectric constant were used
for the energetic map calculations.
Using the Lamarckian Genetic Algorithm, the docked compounds
were subjected to 100 runs of the Autodock search, using 500000
steps of energy evaluation and the default values of the other
parameters. Cluster analysis was performed on the results using an
rms tolerance of 1.0 Å, and the best docked conformation was used
for the analysis.
(12) Minutolo, F.; Antonello, M.; Bertini, S.; Ortore, G.; Placanica, G.;
Rapposelli, S.; Sheng, S.; Carlson, K. E.; Katzenellenbogen, B. S.;
Katzenellenbogen, J. A.; Macchia, M. Novel Estrogen Receptor
Ligands Based on an Anthranylaldoxime Structure: Role of the Phenol-
Type Pseudocycle in the Binding Process. J. Med. Chem. 2003, 46,
4032–4042.
(13) Tuccinardi, T.; Bertini, S.; Martinelli, A.; Minutolo, F.; Ortore, G.;
Placanica, G.; Prota, G.; Rapposelli, S.; Carlson, K. E.; Katzenellen-
bogen, J. A.; Macchia, M. Synthesis of Anthranylaldoxime Derivatives
as Estrogen Receptor Ligands and Computational Prediction of Binding
Modes. J. Med. Chem. 2006, 49, 5001–5012.
(14) There are several examples of successful bioisosteric replacements in
the field of nuclear receptors; for example, three carbon atoms
constituting the o-aminophenone group of farglitazar, a PPARR
agonist, were removed, leading to a vinylogous amide analogue
retaining activity. Xu, H. E.; Lambert, M. H.; Montana, V. G.; Plunket,
K. D.; Moore, L. B.; Collins, J. L.; Oplinger, J. A.; Kliewer, S. A.;
Gampe, R. T., Jr.; McKee, D. D.; Moore, J. T.; Willson, T. M.
Structural determinants of ligand binding selectivity between the
peroxisome proliferators-activated receptors. Proc. Natl. Acad. Sci.
U.S.A. 2001, 98, 13919–13924.
(15) De Angelis, M.; Stossi, F.; Carlson, K. E.; Katzenellenbogen, B. S.;
Katzenellenbogen, J. A. Indazole Estrogens: Highly Selective
Ligands for the Estrogen Receptor ꢀ. J. Med. Chem. 2005, 48, 1132–
1144.
All graphic manipulations and visualizations were performed by
means of Chimera.34
Acknowledgment. Support of this research from the Na-
tionalInstitutesofHealthisgratefullyacknowledged(R37DK15556
and R01CA18119 to J.A.K.). Molecular graphics images were
produced using the UCSF Chimera package from the Resource
for Biocomputing, Visualization, and Informatics at the Uni-
versity of California, San Francisco (supported by NIH P41 RR-
01081). C.G. gratefully acknowledges Italian Ministry for
University and Research (MIUR), for a Ph.D. fellowship
(“grandi programmi strategici”).
(16) Minutolo, F.; Bellini, R.; Bertini, S.; Carboni, I.; Lapucci, A.;
Pistolesi, L.; Prota, G.; Rapposelli, S.; Solati, F.; Tuccinardi, T.;
Martinelli, A.; Stossi, F.; Carlson, K. E.; Katzenellenbogen, B. S.;
Katzenellenbogen, J. A.; Macchia, M. Monoaryl-Substituted Sali-
cylaldoximes as Ligands for Estrogen Receptor ꢀ. J. Med. Chem.
2008, 51, 1344–1351.
(17) Sanz, R.; Castroviejo, M. P.; Ferna´ndez, Y.; Fan˜ana´s, F. J. A New
and Efficient Synthesis of 4-Functionalized Benzo[b]furans from 2,3-
Dihalophenols. J. Org. Chem. 2005, 70, 6548–6551.
Supporting Information Available: Combustion analysis data
of final compounds 3-12. This material is available free of charge
via the Internet at http://pubs.acs.org.
(18) Karabatsos, G. J.; Hsi, N. Structural Studies by Nuclear Magnetic
Resonance. XI: Conformations and Configurations of Oxime O-Methyl
Ethers. Tetrahedron 1967, 23, 1079–1095.
(19) Beauchemin, A. M.; Moran, J.; Lebrun, M.-E.; Se´guin, C.; Dimitrijevic,
E.; Zhang, L.; Gorelsky, S. I. Intermolecular Cope-Type Hydroami-
nation of Alkenes and Alkynes. Angew. Chem., Int. Ed. 2008, 47,
1410–1413.
(20) Katzenellenbogen, J. A.; Johnson, H. J., Jr.; Myers, H. N. Photoaffinity
Labels for Estrogen Binding Proteins of Rat Uterus. Biochemistry 1973,
12, 4085–4092.
(21) Carlson, K. E.; Choi, I.; Gee, A.; Katzenellenbogen, B. S.; Katzenel-
lenbogen, J. A. Altered Ligand Binding Properties and Enhanced
Stability of a Constitutively Active Estrogen Receptor: Evidence That
an Open-Pocket Conformation is Required for Ligand Interaction.
Biochemistry 1997, 36, 14897–14905.
(22) (a) Malamas, M. S.; Manas, E. S.; McDevitt, R. E.; Gunawan, I.;
Xu, Z. B.; Collini, M. D.; Miller, C. P.; Dinh, T.; Henderson, R. A.;
Keith, J.C., Jr.; Harris, H. A. Design and Synthesis of Aryl
Diphenolic Azoles as Potent and Selective Estrogen Receptor-ꢀ
Ligands. J. Med. Chem. 2004, 47, 5021–5040. (b) Manas, E. S.;
Unwalla, R. J.; Xu, Z. B.; Malamas, M. S.; Miller, C. P.; Harris,
H. A.; Hsiao, C.; Akopian, T.; Hum, W.-T.; Malakian, K.; Wolfrom,
S.; Bapat, A.; Bhat, R. A.; Stahl, M. L.; Somers, W. S.; Alvarez,
J. C. Structure-Based Design of Estrogen Receptor-ꢀ Selective
Ligands. J. Am. Chem. Soc. 2004, 126, 15106–15119. (c) Mewshaw,
R. E.; Bowen, S. M.; Harris, H. A.; Xu, Z. B.; Manas, E. S.; Cohn,
S. T. ERꢀ ligands. Part 5: Synthesis and structure-activity
relationships of a series of 4′-hydroxyphenyl-aryl-carbaldehyde
oxime derivatives. Bioorg. Med. Chem. Lett. 2007, 17, 902–906, and
references therein.
(23) Sun, J.; Meyers, M. J.; Fink, B. E.; Rajendran, R.; Katzenellenbogen,
J. A.; Katzenellenbogen, B. S. Novel ligands that function as selective
estrogens or antiestrogens for estrogen receptor-R or estrogen receptor-
ꢀ. Endocrinology 1999, 140, 800–804.
(24) Kumpulainen, H.; Ma¨ho¨nen, N.; Laitinen, M.-L.; Jaurakkaja¨rvi, M.;
Raunio, H.; Juvonen, R. O.; Vepsa¨la¨inen, J.; Ja¨rvinen, T.; Rautio, J.
Evaluation of Hydroxyimine as Cytochrome P450-Selective Prodrug
Structure. J. Med. Chem. 2006, 49, 1207–1211.
References
(1) Dahlman-Wright, K.; Cavailles, V.; Fuqua, S. A.; Jordan, V. C.;
Katzenellenbogen, J. A.; Korach, K. S.; Maggi, A.; Muramatsu,
M.; Parker, M. G.; Gustafsson, J.-Å. International Union of
Pharmacology. LXIV. Estrogen Receptors. Pharmacol. ReV. 2006,
58, 773–781.
(2) Kuiper, G. G. J. M.; Enmark, E.; Pelto-Huikko, M.; Nilsson, S.;
Gustafsson, J.-Å. Cloning of a novel receptor expressed in rat
prostate and ovary. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 5925–
5930.
(3) Weihua, Z.; Andersson, S.; Cheng, G.; Simpson, E. R.; Warner, M.;
Gustafsson, J.-Å. Update on estrogen signaling. FEBS Lett. 2003, 546,
17–24.
(4) Leung, Y.-K.; Mak, P.; Hassan, S.; Ho, S.-M. Estrogen receptor (ER)-ꢀ
isoforms: A key to understanding ER-ꢀ signalling. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 13162–13167.
(5) Cosman, F.; Lindsay, R. Selective estrogen receptor modulators:
clinical spectrum. Endocr. ReV. 1999, 20, 418–434.
(6) Jordan, V. C. Chemoprevention of breast cancer with selective
oestrogen-receptor modulators. Nat. ReV. Cancer 2007, 7, 46–53.
(7) Norman, B. H.; Dodge, J. A.; Richardson, T. I.; Borromeo, P. S.; Lugar,
C. W.; Jones, S. A.; Chen, K.; Wang, Y.; Durst, G. L.; Barr, R. J.;
Montrose-Rafizadeh, C.; Osborne, H. E.; Amos, R. M.; Guo, S.;
Boodhoo, A.; Krishnan, V. Benzopyrans Are Selective Estrogen
Receptor ꢀ agonists with Novel Activity in Models of Benign Prostatic
Hyperplasia. J. Med. Chem. 2006, 49, 6155–6157.
(8) In this case, the effect due to ERꢀ-stimulation appears to be inferior
with respect to ERR-agonists, see Hillisch, A.; Peters, O.; Kosemund,
D.; Mu¨ller, G.; Walter, A.; Schneider, B.; Reddersen, G.; Elger, W.;
Fritzemeier, K.-H. Dissecting physiological roles of estrogen receptor
R and ꢀ with potent selective ligands from structure-based design.
Mol. Endocrinol. 2004, 18, 1599–1609.
(9) Harris, H. A.; Albert, L. M.; Leathurby, Y.; Malamas, M. S.;
Mewshaw, R. E.; Miller, C. P.; Kharode, Y. P.; Marzolf, J.; Komm,
B. S.; Winneker, R. C.; Frail, D. E.; Henderson, R. A.; Zhu, Y.;
Keith, J. C., Jr. Evaluation of an Estrogen Receptor-ꢀ Agonist in
Animal Models of Human Disease. Endocrinology 2003, 144, 4241–
4249.
(25) Wrenn, C. K.; Katzenellenbogen, B. S. Structure-function analysis
of the hormone binding domain of the human estrogen receptor by
region-specific mutagenesis and phenotypic screening in yeast. J. Biol.
Chem. 1993, 268, 24089–24098.
(10) Minutolo, F.; Bertini, S.; Papi, C.; Carlson, K. E.; Katzenellenbogen,
J. A.; Macchia, M. Salicylaldoxime Moiety as a Phenolic “A-Ring”
Substitute in Estrogen Receptor Ligands. J. Med. Chem. 2001, 44,
4288–4291.
(11) Minutolo, F.; Antonello, M.; Bertini, S.; Rapposelli, S.; Rossello,
A.; Sheng, S.; Carlson, K. E.; Katzenellenbogen, J. A.; Macchia,
M. Synthesis, Binding Affinity, and Transcriptional Activity of
Hydroxy- and Methoxy-Substituted 3,4-Diarylsalicylaldoximes on
(26) McInerney, E. M.; Weis, K. E.; Sun, J.; Mosselman, S.; Katzenel-
lenbogen, B. S. Transcription activation by the human estrogen receptor
subtype ꢀ (ERꢀ) studied with ERꢀ and ERR receptor chimeras.
Endocrinology 1998, 139, 4513–4522.

Salicylaldoximes as ERꢀ Agonists
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 867
(27) Norman, B. H.; Dodge, J. A.; Richardson, I.; Borromeo, P. S.; Lugar,
C. W.; Jones, S. A.; Chen, K.; Wang, Y.; Durst, G. L.; Barr, R. J.;
Montrose-Rafizadeh, C.; Osborne, H. E.; Amos, R. M.; Guo, S.;
Boodhoo, A.; Krishnan, V. Benzopyrans are selective estrogen receptor
beta agonists with novel activity in models of benign prostatic
hyperplasia. J. Med. Chem. 2006, 49, 6155–6157.
(30) Maestro, Version 7.5; Schro¨dinger Inc.: Portland, OR, 1999.
(31) Macromodel, Version 8.5; Schro¨dinger Inc.: Portland, OR, 1999.
(32) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.;
Belew, R. K.; Olson, A. J. Automated docking using a Lamarckian
genetic algorithm and an empirical binding free energy function.
J. Comput. Chem. 1998, 19, 1639–1662.
(33) Sanner, M. F. Python: A Programming Language for Software
Integration and Development. J. Mol. Graphics Modell. 1999, 17, 57–
61.
(34) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.;
Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. UCSF Chimera: A
Visualization System for Exploratory Research and Analysis. J. Com-
put. Chem. 2004, 25, 1605–1612.
(28) RSCB Protein Data Bank; http://www.pdb.org.
(29) Case, D. A.; Darden, T. A.; Cheatham, T.E., III; Simmerling, C. L.;
Wang, J.; Duke, R. E.; Luo, R.; Merz, K. M.; Pearlman, D. A.;
Crowley, M.; Walker, R. C.; Zhang, W. ; Wang, B.; Hayik, S.;
Roitberg, A.; Seabra, G.; Wong, K. F.; Paesani, F.; Wu, X. ; Brozell,
S.; Tsui, V.; Gohlke, H.; Yang, L.; Tan, C.; Mongan, J.; Hornak, V.;
Cui, G.; Beroza, P.; Mathews, D. H.; Schafmeister, C.; Ross, W. S.;
Kollman, P. A. AMBER 9; University of California: San Francisco,
2006.
JM801458T