Inhibition of amyloidogenesis by nonsteroidal anti-inflammatory drugs and their hybrid nitrates

Article abstract of DOI:10.1021/jm101450p

Poor blood-brain barrier penetration of nonsteroidal anti-inflammatory drugs (NSAIDs) has been blamed for the failure of the selective amyloid lowering agent (SALA) R-flurbiprofen in phase 3 clinical trials for Alzheimer's disease (AD). NO-donor NSAIDs (NO-NSAIDs) provide an alternative, gastric-sparing approach to NSAID SALAs, which may improve bioavailability. NSAID analogues were studied for anti-inflammatory activity and for SALA activity in N2a neuronal cells transfected with human amyloid precursor protein (APP). Flurbiprofen (1) analogues were obtained with enhanced anti-inflammatory and antiamyloidogenic properties compared to 1, however, esterification led to elevated A _1-42 levels. Hybrid nitrate prodrugs possessed superior anti-inflammatory activity and reduced toxicity relative to the parent NSAIDs, including clinical candidate CHF5074. Although hybrid nitrates elevated A _1-42 at higher concentration, SALA activity was observed at low concentrations (1 μM): both A_1-42 and the ratio of A ₄₂/A_1-40 were lowered. This biphasic SALA activity was attributed to the intact nitrate drug. For several compounds, the selective modulation of amyloidogenesis was tested using an immunoprecipitation MALDI-TOF approach. These data support the development of NO-NSAIDs as an alternative approach toward a clinically useful SALA.

Full text of DOI:10.1021/jm101450p

ARTICLE
pubs.acs.org/jmc
Inhibition of Amyloidogenesis by Nonsteroidal Anti-inflammatory
Drugs and Their Hybrid Nitrates
Isaac T. Schiefer, Samer Abdul-Hay, Huali Wang, Michael Vanni, Zhihui Qin, and Gregory R. J. Thatcher*
Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, University of Illinois at Chicago, MC 781, 833 South
Wood Street, Chicago, Illinois 60612-7231, United States
ABSTRACT: Poor bloodꢀbrain barrier penetration of nonster-
oidal anti-inflammatory drugs (NSAIDs) has been blamed for the
failure of the selective amyloid lowering agent (SALA) R-flurbi-
profen in phase 3 clinical trials for Alzheimer’s disease (AD).
NO-donor NSAIDs (NO-NSAIDs) provide an alternative, gastric-
sparing approach to NSAID SALAs, which may improve bioavail-
ability. NSAID analogues were studied for anti-inflammatory
activity and for SALA activity in N2a neuronal cells transfected
with human amyloid precursor protein (APP). Flurbiprofen (1)
analogues were obtained with enhanced anti-inflammatory and
antiamyloidogenic properties compared to 1, however, esterifica-
tion led to elevated Aβ1ꢀ42 levels. Hybrid nitrate prodrugs
possessed superior anti-inflammatory activity and reduced toxicity
relative to the parent NSAIDs, including clinical candidate CHF5074. Although hybrid nitrates elevated Aβ1ꢀ42 at higher
concentration, SALA activity was observed at low concentrations (e1 μM): both Aβ1ꢀ42 and the ratio of Aβ1ꢀ42/Aβ1ꢀ40 were
lowered. This biphasic SALA activity was attributed to the intact nitrate drug. For several compounds, the selective modulation of
amyloidogenesis was tested using an immunoprecipitation MALDI-TOF approach. These data support the development of
NO-NSAIDs as an alternative approach toward a clinically useful SALA.
’ INTRODUCTION
The effect of NSAID SALAs on Aβ levels was shown previously
to be independent of the cyclooxygenase (COX) inhibitory activity,
a dissociation of actions considered crucial for drug development,
because NSAID gastrotoxicity is associated with COX inhibi-
tion.10,11 Supported by epidemiological studies that reported a
decreased risk of developing AD after chronic treatment with
NSAIDs,19ꢀ21 the COX-inactive R-flurbiprofen was studied in
phase 3 clinical trials as a gastric-sparing therapeutic agent for AD
(Scheme 1).7 In 2008, this trial was reported as a failure due to lack
of efficacy, possibly linked to poor bioavailability.22,23 Nevertheless,
CHF5074 (11d), a new COX-inactive analogue of flurbiprofen (1)
appears to be on course for clinical development.24 An alternative
approach to developing a gastric-sparing NSAID has been to
incorporate a gastroprotective organic nitrate moiety in a so-called
NO-donating NSAID (NO-NSAID) to counteract the NSAID-
induced inhibition of prostaglandin synthesis.25 HCT-1026 (2),26
an NO-NSAID prodrug of 1, has been reported: (i) to reduce Aβ
load in an amyloid transgenic mouse model,27 (ii) to reverse
cognitive deficits in mice under cholinergic blockade,28 and (iii)
Alzheimer’s disease (AD), the major cause of dementia in the
elderly population, is a neurodegenerative disorder characterized
by the abnormal accumulation of amyloid plaques and neurofibrillary
tangles.1,2 According to the amyloid cascade hypothesis, β-amyloid
(Aβ) peptide, the main constituent of amyloid plaques, is the cause
of the observed neurodegeneration in AD patients.3 In 2001,
Weggen et al. discovered that certain NSAIDs were able to selectively
lower the levels of Aβ1ꢀ42 without affecting levels of Aβ1ꢀ40; these
drugs were referred to as SALAs.4 The importance of this discovery
lay in evidence that the 42 amino acid fragment, Aβ1ꢀ42, aggregates
more rapidly and is a much more potent neurotoxin than the more
abundant Aβ1ꢀ40.
3,5 Aβ1ꢀ42 is widely viewed as the causative mole-
cule in AD. The enzymes responsible for the production of Aβ,
namely β- and γ-secretase, remain primary drug targets for the
treatment of AD.6 Changes in the levels of Aβ elicited by NSAID
SALAs in vitro and in vivo have been attributed to modulation of Aβ
production,4,7ꢀ11 via: modulation of γ-secretase activity,4,7 down-
regulation of β-secretase,12 or Rho kinase inhibition.13,14 In addition,
NSAIDs including flurbiprofen (1) have been reported to inhibit
amyloidosis by stabilizing transthyretin.15 The N2a neuronal cell line
stably transfected with human βAPP695 bearing the Swedish muta-
tion (N2a.hAPP) provides a model for amyloidogenesis, producing
Aβ peptides associated with human pathophysiology. Studies on this
cell line demonstrated that NSAID SALAs stimulate metalloprotease
to reduce inflammatory markers, including iNOS, in adult rats
29
injected with Aβ1ꢀ42
.
The purpose of this study was to assay libraries of indometha-
cin (3), sulindac (4), and flurbiprofen analogues, including novel
compounds and preclinical agents (e.g., CHF5022, 11c,30 and
degradation of Aβ1ꢀ40 and Aβ1ꢀ42
.
16 The enhanced proteolysis of
Received: November 10, 2010
Published: March 15, 2011
Aβ is currently seen as a preferred drug discovery strategy.17,18
r
2011 American Chemical Society
2293
dx.doi.org/10.1021/jm101450p J. Med. Chem. 2011, 54, 2293–2306
|

Journal of Medicinal Chemistry
ARTICLE
Scheme 1
11d), as a basis for development of new NSAID hybrid nitrates as
SALAs. Anti-inflammatory activity and toxicity were assayed in
cell culture and for selected compounds inhibition of amyloido-
genesis was assayed in neuronal cell cultures. On the basis of
these data, prototype hybrid nitrate analogues of 1 were synthe-
sized and studied for SALA activity, revealing further insight into
the mechanism of SALA activity associated with NO-NSAIDs.
synthesized by coupling of 1 with 4-aminobutanol, followed by
treatment with tetrabromomethane to give 24b and subsequent
nitration with AgNO3 to afford the nitrate 27 (Scheme 4).
Activity. Three structural approaches to analogues of 1 were
developed: (1) variations at the terminal phenyl ring (R1), (2)
0
modification of the R-carbon (R2 and R2 ), and (3) isosteric
replacement of the carboxylic acid (R3) (Scheme 1). Derivatization
of the R-carbon is known to be a subtle modification resulting in
reduction or ablation of COX activity, whereas replacement of the
carboxylate group is expected to be important in enhancing bloodꢀ
brain barrier penetration. For these modifications to be successful,
enhanced SALA activity was desired without loss of NSAID anti-
inflammatory activity. Mouse macrophage-like RAW 264.7 cells,
treated with lipopolysaccharide (LPS) bacterial endotoxin, undergo
induction of COX-2, inducible NO synthase (iNOS), and other
inflammatory mediators.31 Thus, the RAW 264.7 cell line is routinely
used to examine the ability of agents to inhibit cellular inflammatory
response. iNOS activity is readily assessed by measuring the levels of
inorganic nitrite, the major product of NO oxidative metabolism,
quantified using the Griess assay.31 Flurbiprofen (1) itself was a
modest inhibitor of iNOS induction in RAW cells treated with LPS,
and no evidence of toxicity was seen at the lower concentration
tested (Table 1).28,32,33
’ RESULTS AND DISCUSSION
Synthesis. Synthesis was directed at the generation of a library
of compounds possessing variations at the terminal phenyl ring
0
(R1), R-carbon (R2 and R2 ), and isosteric replacement of the
carboxylic acid (R3) (Scheme 1).
Synthesis began with allylic bromination of the commercially
available 4-bromo-3-fluorotoluene 6, with N-bromosuccinimide
using a catalytic amount of dibenzoyl peroxide to yield 7, which was
then reacted with potassium cyanide to give 8. Exposure to either
iodoethane or 1,2-dibromoethane yielded alkylated products 9a and
9b that were then oxidized to the carboxylic acid followed by Suzuki
coupling to the appropriate boronic acids to give biarylcarboxylic
acids 11aꢀd in moderate yields (60ꢀ70%). Quantitative yield was
attained through direct coupling of the cyano analogues 8 and 9b,
yielding the R-cyano-biaryls 12a and 12b, along with 13a and 13b.
Fluorination of 13a and 13b was accomplished via reaction with
t-BuLi followed by fluorination with N-fluorobenzenesulfonimide
(NFSi) togive14aꢀband 15aꢀb. While base catalyzed oxidation of
14a, 14b, 15a, and 15b gave the corresponding biaryl fluoro-amides
16a, 16b, 17a, and 17b, the desired biaryl-fluorinated car-
boxylic acids were not isolated. Acid catalyzed hydrolysis of 14a
and 14b yielded 18a and 18b as products that were also evaluated for
inhibition of amyloidogenesis and anti-inflammatory activity. Base
and acid catalyzed oxidation of the diethyl derivative 12a was
hindered by solubility issues, therefore, the corresponding acid
was synthesized by first oxidizing 13a to give the acid 19, which
was then reacted with bromoethane in aqueous NaOH and
toluene to give 20, followed by hydrolysis with LiOH to yield the
desired acid 21 (Scheme 2).
To synthesize the biaryl mono carboxylic acids 23a and 23b, a
synthetic pathway was adopted in which 1 or 19 were reacted with
t-butyldimethylsilyl chloride (TBSCl) and lithium bis(trimethyl-
silyl)amide (LiHMDS) at 0 °C to give the desired bis-ketenes 22a
and 22b, which were then reacted with Selectfluor followed by
acidꢀbase workup to give the desired fluorinated acids, 23a and 23b
(Scheme 3). Flurbiprofen derivatives and analogues were synthe-
sized from 1, 11a, 11c, and 11d through alkylation with 1,2-dibro-
moethane for 11a, 11c, and 11d to give intermediates 25aꢀc, or via
alkylation of 1 with 1,4-dibromobutane to yield 24a. The bromides
were then transformed to the corresponding nitrates with AgNO3
to yield 2 and 26aꢀc. In a similar manner, the amide 27 was
In comparison with 1, several analogues gave a significantly
greater anti-inflammatory response, although in the case of 14b
and 18a, this was associated with somewhat increased toxicity.
These data demonstrate that the three structural modifications
can be accommodated while maintaining or enhancing anti-inflam-
matory activity toward iNOS induction, relative to 1. At higher
concentration, anti-inflammatory activity was increased relative to 1
again without toxicity (Table 2). Cyclopropyl analogues 11a and 11c
gave similar responses; however, 11d was highly toxic toward RAW
cells. Comparison with other test compounds showed that toxicity
was not a feature associated with one of the three structural motifs
(R1, R2, R3) but a property of the molecule as a whole.
We have previously reported use of the Griess assay in
measurement of the anti-inflammatory activity of nitrates toward
LPS-induced RAW cells.31 There is a potential pitfall with this
approach, bꢀec3a4use metabolic denitration of nitrate drugs directly
yields NO2
.
However, a complete analysis of denitration in
induced and noninduced RAW cells demonstrated that simple
aliphatic mononitrates, such as isosorbide mononꢀitrate (ISMN)
and 2, release <3% of the theoretical yield of NO2 after 24 h.31
We previously reported that 2was significantly more potent than 1in
attenuating induction of iNOS activity, whereas the classical mono-
nitrate, ISMN, showed no significant inhibition of iNOS activity, and
in combination with 1 was no more efficacious than 1 alone.28 These
observations extend to the hybrid nitrate derivatives, 26aꢀc, which
were observed to be significantly more potent than their parent
flurbiprofen analogue (Table 2). Interestingly, conversion of the
2294
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
Scheme 2a
a Reagents and Conditions: (i) NBS, CCl4; (ii) KCN, EtOH; (iii) 1,2 dibromoethane for 9a or CH3CH2I for 9b, NaOH, tolueneꢀH2O; (iv) NaOH,
MeOH, 100 °C; (v) R1B(OH)2, Pd(OAc)2, Na2CO3, TBAB, H2O, 120 °C; (vi) NFSi, t-BuLi, THF, ꢀ78 °C; (vii) K2CO3, DMSO, H2O2, 0 °C yields
16aꢀb and 17aꢀb, or H2SO4 (49%), 90 °C to give 18aꢀb; (viii) K2CO3, DMSO, H2O2, 0 °C; (ix) CH3CH2Br, NaOH, tolueneꢀH2O; (x) LiOH,
THF/H2O/MeOH, 0 °C.
Scheme 3a
work published by Peretto et al. focused mainly on substituting the
terminal phenyl ring (R1) of the cyclopropyl analogue of 2, leadingto
compound 11d.11 The murine N2a.hAPP cell line has previously
been employed by various investigators to assay SALAs using ELISA
16,28,35
to measure Aβ1ꢀ40 and/or Aβ1ꢀ42
.
In this assay, confluent
N2a/APPsw cells were treated with drug for 24 h, following which
the supernatant was collected and Aβ1ꢀ42 measured by ELISA.
Compound 11d, observed to be toxic in RAW cells at higher concen-
tration (100 μM), was not toxic toward N2a cells at lower concen-
tration (1 μM). However, in contrast to reports of the activity of this
compound tested at higher concentrations (i.e., 50ꢀ500 μM),30,36 at
the low concentration tested herein (1 μM), 11d did not reduce
levels of Aβ1ꢀ42 (Table 3). Carboxylate analogues possessing a
phenyl R1 substituent uniformly lowered Aβ1ꢀ42, with the exception
of the diethyl analogue 21.
A number of derivatives (i.e., 11b, 18a, and 23b) demonstrated
antiamyloidogenic activity, comparable to 1, in the absence of sub-
stantial toxicity. These compounds all contain the NSAID carboxy-
late group but are otherwise structurally diverse. Two compounds
stood out as superior to 1, 11a and 14a, both bearing an unsub-
stituted phenyl R1 substituent. The activity of the nitrile 14a was of
interest because the observation of antiamyloidogenic activity shows
that the carboxylate group is not an absolute requirement. This was
a Reagents and Conditions: (i) TBSCl, LiHMDS, THF, 0 °C, 12 h; (ii)
Selectfluor, CH3CN, <50 °C.
cytotoxic 11d to a hybrid nitrate (26c) completely ablated toxicity.
However, neither the anti-inflammatory activity nor the ablated
toxicity could be ascribed to the nitrate group because the precursor
bromo-derivatives, 25aꢀꢀc, were the most potent anti-inflammatory
agents, reducing NO2 production 4-fold relative to the parent
carboxylic acids in the absence of any cytotoxicity.
Approaches to the design of improved analogues of 1 as
antiamyloidogenic agents have been published previously. Modifica-
tion of the R-carbon through the incorporation of a cyclopropyl
group11 or a dimethyl group10 were reported to lead to increased
potency for lowering Aβ while eliminating COX activity. Previous
2295
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
Scheme 4a
a Reagents and Conditions: (i) NaOCH3, CH3OH; (ii) Br(CH2)4Br for 24a, Br(CH2)2Br for 25aꢀc, or 4-aminobutanol followed by PPh3 with CBr4 for
24b, DMF; (iii) AgNO3, CH3CN, reflux.
Table 1. Anti-inflammatory Effects of Selected 1 Analogues
(10 μM)
full concentrationꢀresponse data for 2 showed SALA activity
more potent than 1 and similar to 11a at low concentrations and,
in contrast, SARA activity at >10 μM (Figure 1B).
It was important to determine whether this biphasic activity
was peculiar to compound 2 or if, indeed, this was common to
related NSAID hybrid nitrates. Therefore, nitrate 26a was studied
over a range of concentrations, revealing a similar response, albeit
with less efficacy as a SALA than 2 (Figure 2A). To determine if the
nitrate group was relevant to the observed biphasic activity of these
NSAID hybrid nitrates, the antiamyloidogenic activity of the nitrate
2 was compared with that of the simple ethyl ester 38 in N2a.hAPP
cell culture incubations (Figure 2B). The ester behaved as a typical
SARA agent increasing Aβ1ꢀ42 2.5-fold at 10 μM, indicating the
importance of the nitrate for SALA activity at low concentrations.
However, this SALA activity of hybrid nitrate NSAIDs toward
Aβ1ꢀ42 modulation is unlikely to be associated with NO bioactivity
delivered by the nitrate functionality since we have previously noted
that the classical nitrate ISMN is without activity in this assay.28 The
alkyl ester linker of the hybrid nitrates, 2 and 26a, is expected to be
labile toward nonspecific esterase activity; therefore, it was impor-
tant to test a hybrid nitrate with a stable linker. The amide linked
nitrate, 27, was synthesized expressly to test if the more labile ester
linker of 2 and 26a was essential for the observed antiamyloidogenic
activity. The antiamyloidogenic activity at low concentration
(0.5 μM) was retained compared to 26a, indicating that the intact
nitrate is required for SALA activity (Figure 2C). This appears to be
only the second example where biological activity can be attributed
to the intact NO-NSAID, rather than to the NO-NSAID acting as a
prodrug releasing NSAID, aliphatic nitrate, and/or NO.37
The NO-flurbiprofens, NCX-2216 and HCT-1026 (2), have
been reported to lower Aβ in mouse models, however, NCX-
2216 was not detected in brain tissue,38 and both plasma and
brain levels of HCT-1026 were reported to be below detection
limits.39 Hydrolysis of HCT-1026 yields 4-hydroxybutyl nitrate,
and it is possible that this metabolite has brain bioavailaibility
because animal studies have shown that the level of inorganic
nitrite in the brain increases after oral administration of HCT-
1026.29,40 It was of interest both to confirm the increased stability
of the amide linked hybrid nitrate 27 relative to the ester linked
homologue, 2, and to determine if unlike 2, the NO-flurbiprofen,
27, was detectable in the brain. Therefore, 27 (10 mg/kg) was
delivered by ip injection to male C75/BL6 mice (25 g) and brain
and plasma levels detected by LC-MS/MS at 30 and 120 min
after injection. The brain/plasma ratio of 27 was 37% and 13% at
30 and 120 min, respectively (Figure 3). Most importantly, 27
was present in detectable amounts both in the brain and plasma 2
h after administration, in contrast to reports on 2.
cell viability
NO2ꢀ as %a
as %b
0
compd
R1
R2, R2
R3
DMSO
1
100 ( 0.8
80.7 ( 3.6
86.4 ( 3.5
91.3 ( 4.6
65.3 ( 9.0
75.7 ( 9.1
74.7 ( 7.2
82.2 ( 3.9
84.1 ( 5.3
59 ( 3.4
100 ( 2.3
102 ( 2.7
102 ( 1.7
107 ( 1.1
84.5 ( 1.2
98.2 ( 3.3
90.6 ( 1.1
100 ( 0.4
98.8 ( 1
89.1 ( 3.5
102 ( 1
99.7 ( 3.6
92.7 ( 4.1
phenyl
H, CH3
COOH
11b
12b
14b
15b
16b
17a
17b
18a
18b
23a
23b
4-F-phenyl cyclopropyl COOH
napthyl
ethyl, ethyl
CN
CN
CN
CONH2
CONH2
CONH2
COOH
COOH
COOH
COOH
4-F-phenyl H, F
4-F-phenyl F, F
4-F-phenyl H, F
phenyl
4-F-phenyl F, F
phenyl
4-F-phenyl H, OH
phenyl
phenyl
F, F
H, OH
76.3 ( 4.9
80.8 ( 3.4
81.5 ( 3.9
F, CH3
F, H
a Griess assay. b MTT assay.
the most interesting of the novel R-fluoro analogues but obviously
was not suitable for further study as a hybrid nitrate.
Given the ability of at least one analogue, lacking a carboxylate
group, to lower Aβ1ꢀ42, and several such examples inhibiting iNOS
activity, it was of interest to assay analogues of 3 and 4, which
contained examples with amides in place of carboxylic acids. The
derivatives of 3 provided the expected anti-inflammatory activity and
the expected antiamyloid activity for 3, itself, was observed at high
concentration (Table 4). Other activity of note was the amyloid
raising activity of the amide derivative, 34, which was of the same
magnitude as seen for so-called selective amyloid raising agents
(SARAs) such as fenofibrate (39). Derivatives of 4 (28ꢀ33)
provided one compound with a good profile of anti-inflammatory
and antiamyloidogenic activity without cytotoxicity, the desmethy-
lated derivative 28. Amide derivatives did not reduce Aβ1ꢀ42 and
were comparable to 4 (Table 5).
For selected agents, the dependence of antiamyloidogenesis
on concentration was tested further, again using N2a.hAPP cell
cultures treated with agents for 24 h, followed by ELISA assay of
the supernatant. The flurbiprofen analogue 11a was compared to
1 itself and shown to be a more potent SALA, lowering Aβ1ꢀ42 by
40% at the highest concentration studied, with no significant
toxicity (Figure 1A). In contrast, the hybrid nitrate derivative 2,
as we have reported previously,28 showed biphasic activity. The
2296
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
Table 2. Anti-inflammatory Effects of 1 and Analogues (100 μM)
NO2ꢀ as %a
cell viability as %b
0
compd
R1
R2, R2
R3
DMSO
1
2
100 ( 0.8
69.1 ( 3.2
49.8 ( 2.5
60.7 ( 1.3
62.4 ( 3.4
100 ( 2.3
109 ( 2.7
108 ( 3.4
101 ( 3.1
102 ( 4.6
19.1 ( 2.8
109 ( 4.0
113 ( 2.1
114 ( 3.7
113 ( 6.5
128 ( 1.8
135 ( 2.9
phenyl
phenyl
phenyl
4-CF3-phenyl
3,4 dichlorophenyl
phenyl
4-CF3-phenyl
3,4 dichlorophenyl
phenyl
4-CF3-phenyl
3,4 dichlorophenyl
H, CH3
H, CH3
COOH
COO(CH2)4ONO2
COOH
COOH
COOH
COO(CH2)2Br
COO(CH2)2Br
COO(CH2)2Br
COO(CH2)2ONO2
COO(CH2)2ONO2
COO(CH2)2ONO2
11a
11c
11d
25a
25b
25c
26a
26b
26c
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
18.9 ( 1.0
14.7 ( 3.2
27.8 ( 2.0
47.3 ( 1.5
42.8 ( 2.1
40 ( 2.4
a Griess assay. b MTT assay.
Table 3. Effects of 1 and Analogues (1 μM) on Aβ1-42 Levels
cell viability as %a
0
compd
R1
R2, R2
R3
Aβ1ꢀ42 as %
DMSO
1
100 ( 3.2
86.6 ( 0.9
69.5 ( 2.0
90.4 ( 1.3
99.3 ( 10.7
125 ( 5.0
125 ( 5.1
105 ( 3.3
65.6 ( 1.5
100 ( 0.9
101 ( 0.9
103 ( 1.9
102 ( 1.3
97.2 ( 3.1
92.9 ( 0.6
97.3 ( 0.7
91.6 ( 1.4
93.1 ( 1.0
69.4 ( 0.3
98.2 ( 0.6
102.8 ( 1.0
69.1 ( 3.0
102 ( 1.3
95.9 ( 0.4
101 ( 0.3
99.7 ( 1.4
95.5 ( 0.3
101 ( 1.8
98.7 ( 1.3
phenyl
phenyl
4-F-phenyl
4-CF3-phenyl
3,4 dichlorophenyl
phenyl
naphthyl
phenyl
4-F-phenyl
4-F-phenyl
phenyl
4-F- phenyl
phenyl
4-F-phenyl
phenyl
4-F-phenyl
phenyl
phenyl
H, CH3
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
ethyl, ethyl
ethyl, ethyl
F, H
F, H
F, F
F, H
F, H
COOH
COOH
COOH
COOH
COOH
CN
CN
CN
CN
CN
CONH2
CONH2
CONH2
CONH2
COOH
COOH
COOH
COOH
COOH
11a
11b
11c
11d
12a
12b
14a
14b
15b
16a
16b
17a
17b
18a
18b
21
125 ( 5
101 ( 0.9
-
115 ( 3.5
109 ( 5.8
82.5 ( 1.6
106 ( 1.3
124 ( 7.1
98.8 ( 2.1
91.8 ( 1.3
F, F
F, F
OH, H
OH, H
ethyl, ethyl
F, CH3
F, H
23a
23b
phenyl
a MTT assay.
The SARA activity of flurbiprofen derivatives 2, 26a, and 38 at
higher concentrations has precedent in SARA agents, including
NSAIDs such as celecoxib, and other drugs, such as 39, that have
Aβ1ꢀ42 levels 3-fold but had modest effects on Aβ1ꢀ37 and
Aβ1ꢀ38 although lowering the level of the latter fragment
(Figure 4). In contrast, fibric acid, the free acid form of 39, was
been reported to elevate Aβ1ꢀ42 levels while lowering the level of
reported to lower the levels of Aβ1ꢀ42.
16 A similar observation on
41ꢀ43
other fragments including Aβ1ꢀ38
.
To study the effect of
Aβ1ꢀ42 modulation was made herein with 1 and its ester, 38.
Whereas, the ester acted as a SARA, the free acid, 1, at 100 μM,
selectively lowered Aβ1ꢀ42 and elevated Aβ1ꢀ37 relative to the
vehicle control (Figure 4). Indeed, the amyloid peptide product
profiles for the three carboxylate esters, 2, 25c, and 39 are very
similar at high concentrations; all act as SARAs, significantly and
selectively increasing Aβ1ꢀ40, Aβ1ꢀ42 and the ratio Aβ1ꢀ42/
Aβ1ꢀ40. In contrast, the hybrid nitrate ester, 2, behaves quite
differently; at low concentrations, 2 lowers total Aβ generation
agents on amyloidogenesis in more detail, we determined the
profile of Aβ fragment production from N2a.hAPP cells for
selected compounds using an immunoprecipitation MALDI-
TOF technique introduced in 2009.16 This technique quantifies
the amyloid fragments Aβ1ꢀ37, Aβ1ꢀ38, Aβ1ꢀ40, and Aβ1ꢀ42
,
using Aβ1ꢀ43 as an internal standard, because this fragment is not
a product of amyloidogenesis. In the N2a.hAPP amyloidogenesis
assay, we observed that the known SARA 39 (200 μM) increased
2297
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
Table 4. Anti-amyloidogenic Inhibitory Activity and Anti-Inflammatory Activity of 3 Analogues
compd
DMSO
R1
R2
Aβ1ꢀ42 as %a
NO2ꢀ as %b
cell viability as %c
100 ( 3
108 ( 6
100 ( 7.4
70.9 ( 7.4
39.8 ( 10.5
68.6 ( 6.3
65.9 ( 0.4
52.5 ( 2.3
100 ( 7.4
100 ( 7.4
98.7 ( 1
3
CH3
CH3
CH3
H
CH3
CH3
COOH
COOH
3 (100 μM)
76.7 ( 2.9
295 ( 27.2
102 ( 5.3
118 ( 9.4
135 ( 8.7
92.0 ( 7.3
92.2 ( 5.3
97.9 ( 1.1
100 ( 1.6
95.6 ( 2.6
34
35
36
37
CONH-4-chlorophenyl
COOH
NHCH2CH2OH
CONHCH2CH2-3-pyridine
a 1 μM. b Griess assay; 10 μM unless otherwise stated. c MTT assay in RAW cells; 10 μM unless otherwise stated.
Table 5. Anti-amyloidogenic Activity and Anti-inflammatory Activity of 4 Analogues
compd
R1
R2
R3
Aβ1ꢀ42 as %a
NO2ꢀ as %b
cell viability as %c
DMSO
4
5
28
29
30
31
32
33
100 ( 3.8
100 ( 0.1
82.6 ( 2.2
80.8 ( 4.3
95 ( 5
94.5 ( 1.8
95.7 ( 1.8
91 ( 1.5
100 ( 11.4
85.7 ( 9.3
100 ( 1.6
104 ( 1.2
SOCH3
SCH3
SCH3
SOCH3
SCH3
SCH3
CF3
CH3
CH3
H
H
H
H
H
H
COOH
COOH
COOH
COOH
CONHCH2Ph
CONHCH2CH3
COOH
47.5 ( 7.2
91.6 ( 11
58.4 ( 12.5
80.4 ( 11.1
29.2 ( 8
100 ( 1.6
104 ( 1.2
100 ( 1.6
104 ( 1.2
100 ( 1.6
104 ( 1.2
SCF3
COOH
99.4 ( 0.4
1.7 ( 6.3
a 1 μM. b Griess assay at 10 μM. c MTT assay in RAW cells at 10 μM.
with similar efficacy to the flurbiprofen analogue, 11a, although
with less selectivity for Aβ1ꢀ42
.
Sulindac (4) is known to act as an NSAID SALA. The expected
selective lowering of Aβ1ꢀ42 and raising of both Aβ1ꢀ37 and
Aβ1ꢀ38 levels was observed (Figure 4).44 Flurbiprofen derivative
11a was a more potent SALA than both 1 and 4 (Figures 1, 4),
moreover, the profile of Aβ modulation by 11a was very different
from that of 4. The concept that NSAID SALAs act as γ-secretase
modulators was initially rationalized by a shift in secretase
cleavage from producing Aβ1ꢀ42 to production of Aβ1ꢀ37
,
however, current opinion is that SALAs do not operate uni-
versally via this synchronized shift from Aβ1ꢀ42 production to
Aβ1ꢀ38 production.43 As we have seen, the well-established
SALA and SARA activity of several drugs is recapitulated in the
N2a.hAPP model system. However, a simple “control experi-
ment” has shown that in this system, the SALA and SARA activity
is not related to γ-secretase modulation. In this experiment, wild-
type N2a cells were treated with drug or vehicle for 24 h, at which
time, conditioned media (CM) was removed from the culture
dish and added to media containing a freshly made solution of
Aβ1ꢀ42 and Aβ1ꢀ40. CM delivered the same modulation of
Aβ1ꢀ42, Aβ1ꢀ40, and Aβ1ꢀ42/Aβ1ꢀ40 observed when drugs were
incubated in N2a.hAPP cell cultures. Further experiments
showed that the modulation of Aβ1ꢀ42 and Aβ1ꢀ40 levels by
SARA and SALA drugs was mediated by expression of
Figure 1. Flurbiprofen and derivatives showed varied SALA and SARA
activity in neuronal cell incubations. Levels of Aβ1ꢀ42 from N2a/APPsw cell
supernatants were measured by ELISA and normalized to DMSO vehicle
treated control after incubation with different concentrations of compounds
1 and 11a (A) or 1 and 2 (B). The average level of Aβ1ꢀ42 in the DMSO
treated groups was ∼500 pg/mL. The data show mean and SEM from at
least four separate experiments performed using separate cell passages.
metalloprotease activity induced by drug treatment of cells.16
Levels of Aβ1ꢀ42 and Aβ1ꢀ40 were measured by the MALDI-TOF
2298
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
Figure 2. Hybrid nitrate flurbiprofen derivatives caused biphasic regulation of Aβ1ꢀ42 in neuronal cell incubations, acting as SALAs at lower
concentration. Levels of Aβ1ꢀ42 from N2a/APPsw cell supernatants were measured by ELISA and normalized to DMSO vehicle treated control
after incubation with different concentrations of compounds 2 and 26a (A), 2 and 38 (B), or 26a (0.5 μM) and 27 (0.5 μM) (C). The average level
of Aβ1ꢀ42 in DMSO treated group was ∼500 pg/mL. The data show mean and sem from at least four separate experiments performed using
separate cell passages.
Figure 4. Quantitative analysis of Aβ1ꢀ37, Aβ1ꢀ38, Aβ1ꢀ40, and Aβ1ꢀ42
in neuronal cell supernatant after drug treatment. For each individual
experiment, peak heights from IP-MALDI-TOF spectra were normal-
ized to the peak height of the Aβ1ꢀ43 standard after incubation of drugs
with N2a/APPsw cells. These normalized values were then expressed
relative to the DMSO vehicle control in each set of experiments. See ref
16 for full details. The data show mean and sem from at least four
separate experiments.
method after treatment of N2a cells with drugs for 24 h and
transfer of the CM to a solution of Aβ1ꢀ42 and Aβ1ꢀ40
(Figure 5). The results of this experiment confirm that 1, 2,
and 28 are able to act as SALAs by increasing degradation of
Aβ1ꢀ42 and Aβ1ꢀ40 and decreasing the ratio of Aβ1ꢀ42/Aβ1ꢀ40
.
Figure 3. Hybrid nitrate 27 has brain bioavailaibility and in vivo
stability after administration to mice (10 mg/kg ip) 30 or 120 min
prior to sacrifice. Quantitation of 27 by LC-MS/MS used MRM
analysis of the m/z transition 361 f 200. (A) Representative MRM
chromatogram of 27 from mouse brain sample 120 min after
administration, showing fragment observed as protonated cation
(m/z = 200) in MRM. (B) Quantitative analysis of brain and plasma
levels at 30 and 120 min using LC-MS/MS detection. Data (N = 4)
show mean and SEM.
’ CONCLUSIONS
The poor bloodꢀbrain barrier penetration common to most
NSAIDs has been blamed for the failure of the SALA R-flurbiprofen
in phase 3 clinical trial for AD. Hybrid nitrate NO-NSAIDs
provide an alternative, gastric-sparing approach to NSAID SALAs of
use in AD. Because these compounds contain an ester linkage in
place of the carboxylate anion, improved CNS bioavailability is
2299
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
supplemented with 200 μg/mL G418 but omitted during all experi-
ments. RAW 264.7 mouse macrophage-like cells, provided by Dr. J.
Cook (University of Illinois at Chicago, Chicago, IL), were maintained
in DMEM, supplemented with 1% penicillinꢀstreptomycin, and 10%
fetal bovine serum. Cells were maintained at 37 °C and 5% CO2.
Griess Assay. RAW cells were plated at a concentration of 25 ꢁ 104
cells/well in a 24-well plate and incubated at 37 °C for 24 h. The medium
was changed, and the cells were drug treated, followed 30 min later by
the addition of lipopolysaccharide (LPS) from Escherichia coli 055:B5
(LPS; Sigma-Aldrich). Twenty-four h later, 100 μL of the supernatant
was removed and incubated with the Griess reagent (100 μL; 0.5%
sulfanilamide, 0.05% (N-1-naphthyl)ethylenediamine dihydrochloride,
2.5% H3PO4, and 97% H2O by weight) for 30 min at rt in the dark. The
absorbance was measured at 530 nm on a Dynex MRX II microplate
spectrophotometer and calibrated using a standard curve constructed
with sodium nitrite to yield nitrite concentration.
Amyloid β Measurement from N2a/APPsw Cells Super-
natant. N2a/APPsw cells were plated 24 h before the experiment at a
density of 25 ꢁ 104 cells/well in a 24-well plate. Cells were washed with
phosphate buffered saline (PBS; 50 mM, pH 7.4) before the addition of
500 μL of DMEM supplemented with 0.2% fetal bovine serum (FBS)
followed by drug treatment. Cells were incubated for a period of 24 h.
CM was collected, followed by the addition of NaN3 (0.01% final
concentration) and a mixture of protease inhibitors (10 mM phenan-
throline and P2714 protease inhibitor cocktail from Sigma containing
4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride, aprotinin,
bestatin, E-64, leupeptin, and ethylenediaminetetraacetate (EDTA)),
and centrifuged for 2 min at a speed of 10 000g. Aβ1ꢀ42 levels were
determined by sandwich ELISA using human Aβ1ꢀ42 ELISA kit and
following the supplied protocol. For immunoprecipitation, cells were
treated as described above but plated in a 6-well plate at a density of
100 ꢁ 104 cells/well in 2 mL media. Twenty-four h after treatment, 1 mL
of the CM is collected, followed by the addition of NaN3 and protease
inhibitors and spiked with Aβ1ꢀ43 at approximately 1 ng/mL final
concentration as an internal standard before performing immunopreci-
pitation. A stock of 200 ng/mL Aβ1ꢀ43 was prepared in sodium
bicarbonate (50 mM, pH 9.0) and stored in aliquots in 1.5 mL low
binding eppendorf tubes at ꢀ80 °C; once defrosted, any remaining
Aβ1ꢀ43 stock was discarded. Following immunoprecipitation the sam-
ples were analyzed in a MALDI-TOF instrument as described below for
the purpose of measuring the relative abundance of Aβ peptides. In all
experiments, Aβ levels were assayed relative to vehicle controls by
treating all the groups with the identical sample of Aβ1ꢀ43 internal
standard.
Aβ Degradation Associated with N2a/WT Cells. N2a/WT
cells were plated 24 h before the experiments at a density of 100 ꢁ 104
cells/well in 6-well plates. Cells were washed with PBS before the
addition of 2 mL of DMEM supplemented with 0.2% FBS followed by
drug treatment. Eighteen h after drug treatment, CM was collected and
treated with NaN3 and centrifuged at 10 000g; the mixture of Aβ1ꢀ40
and Aβ1ꢀ42 was added and further incubated for 8 h at 37 °C. The CM
was then treated with the protease inhibitor mixture and spiked with
Aβ1ꢀ43 as an internal standard followed by immunoprecipitation and
MALDI-TOF analysis.
Immunoprecipitation and MALDI/TOF Analysis. Immuno-
precipitation of Aβ was performed by the addition of a mixture of
0.25 μL of biotinylated 4G8 antibody and 0.25 μL of biotinylated 6E10
antibody, samples were then rotated at room temperature for 4 h
followed by the addition of 4 μL T1 streptavidin coated magnabeads
and rotated for an additional 2 h. Beads were washed three times with
NH4CO3 (10 mM, pH 8.0), and the Aβ peptides were eluted using 4 μL
of a mixture of 75% acetonitrile and 0.5% TFA in H2O, followed by the
addition of 3 μL of a saturated matrix solution (R-cyano-4-hydroxycin-
namic acid in 1:1 acetonitrile and H2O). One μL of this mixture was
Figure 5. SALA activity was observed in mixtures of Aβ1ꢀ40 þ Aβ1ꢀ42
treated with conditioned media (CM) from drug treated N2a/WT cells.
Cells were treated with 1 (100 μM), 2 (1 μM), and 28 (10 μM) for 24 h
before the collection of the CM. CM was added to a mixture of Aβ1ꢀ40
þ Aβ1ꢀ42 and incubated for 8 h. CM from cell free wells (media) was
used as the control. Aβ was quantified by IP-MALDI-TOF and normal-
ized to “media” control. Data show mean and sd from two separate cell
cultures.
expected. In this work, analogues of indomethacin and sulindac did
not show a significantly improved profile of anti-inflammatory activity
nor Aβ1ꢀ42 lowering. However, analogues of flurbiprofen, 1, were
prepared with superior anti-inflammatory and antiamyloidogenic
activity to the parent NSAID. A number of examples were converted
to hybrid nitrate ester prodrugs, including a prodrug of the AD drug
candidate, 11d.24 The hybrid nitrates possessed superior anti-inflam-
matory activity to the parent NSAID. Esterification of 1 led to
elevated Aβ1ꢀ42, however, this SARA activity was observed only at
high concentrations for the hybrid nitrates, whereas at low concen-
trations (<1 μM), Aβ1ꢀ42 and the ratio of Aβ1ꢀ42/Aβ1ꢀ40 were
lowered. Interestingly, the intact hybrid nitrate was found to be
necessary for the SALA activity of these NO-NSAIDs. An amide-
linked NO-NSAID SALA, 27, was shown to have brain bioavailability
and to be detectable in brain and plasma 120 min after administra-
tion. These data support the optimization of hybrid nitrate SALAs,
based upon modified flurbiprofen scaffolds that retain anti-inflam-
matory activity, and provide enhanced brain bioavailability.
’ EXPERIMENTAL SECTION
All chemicals and reagents were purchased from Sigma-Aldrich (St.
Louis, MO) unless stated otherwise. Murine biotinylated antihuman Aβ
4G8 and 6E10 antibodies produced by Signet were purchased from
Covance Research Products, Inc. (Dedham, MA). All cell culture
supplies, human Aβ ELISA kit, and streptavidin Dynabeads T1 were
purchased from Invitrogen (Carlsbad, CA).
Cell Culture and Treatments. The murine neuroblastoma N2a
cells obtained from American Type Culture Collection (Manassas, VA)
were cultured in 1:1 Dulbecco’s Modified Eagle’s Medium (DMEM)
and OPTI-MEM supplemented with 5% fetal bovine serum, 100 U/mL
penicillin, and 100 U/mL streptomycin. N2a cells stably transfected
with the Swedish mutant of human APP (N2a/APPsw were kind gift
of Dr. Gopal Thinakaran, University of Chicago) were additionally
2300
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
spotted onto a MALDI plate and analyzed using a MALDI-TOF instru-
ment (Applied Biosystems). Mass spectra were acquired automatically
in a linear positive mode at 1950 shot per spectrum. For each individual
experiment, peak heights were normalized to the peak height of the
Aβ1ꢀ43 standard; these normalized values were then expressed relative
to the DMSO vehicle control in each set of experiments.
(2 ꢁ 50 mL), dried over Na2SO4, and concentrated to afford red oil (∼14
g), which was purified by column chromatography (hexane/ethyl acetate
[10:1]) to give 7 as a yellow crystalline solid (12.2 g, yield 75.6%). 1H NMR
(CDCl3, 400 MHz): δ 7.55ꢀ7.53 (t, 1H), 7.19ꢀ7.17 (dd, 1H), 7.08ꢀ7.06
(dd, 1H), 4.42 (s, 2H).
4-Bromo-3-fluorophenyl)acetonitrile (8). A solution of 7 (12.0 g, 41
mmol) and KCN (2 g, 0.04 mol) in ethanol (100 mL) was heated to
80 °C for 2 h and then cooled to rt and concentrated under vacuo. The
residue was suspended in H2O (200 mL) and extracted with ethyl
acetate (3 ꢁ 150 mL). Combined organic extracts were washed with
brine (2 ꢁ 50 mL), dried over Na2SO4, and concentrated to afford a
brown oil. Column chromatography (hexane/ethyl acetate [7:1])
MTT Assay. The MTT assay was performed by incubating the cells
in a culture media containing 0.5 mg/mL solution of 3-(4,5-dimethylthia-
zol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) for 4 h, the cells were
then washed with PBS and the formed purple formazan crystals were
solubilized in DMSO. The plate was shaken on a plate rocker for 30 min,
and then the absorbance was measured at 570 nm using 630 nm as a
reference wavelength on a Dynex MRX II microplate spectrophotometer.
Bioavailability. Use of animals was approved by the Institutional
Animal Care and Use Committee at the University of Illinois at Chicago.
Nitrate 27 was dissolved in DMSO/H2O (1:1 v/v) and administrated at
10 mg/kg via ip injection to male C57BL/6 mice (25 g). Animals were
sacrificed at 30 and 120 min, and both blood and whole brain were collected
for analysis. Two equal portions of each brain or blood sample were
processed for analysis. Brain tissue was homogenized in methanol manually
and then centrifuged at 4 °C, 13000 rpm for 10 min. The remaining pellet
was washed with methanol (0.3 mL) and supernatants combined. After
concentration under N2, reconstitution in MeOH/H2O (1:1, v/v), and
centrifugation at 4 °C, 13000 rpm for 15 min, samples were analyzed by LC-
MS/MS. Plasma was extracted in a similar way prior to analysis by LC-MS/
MS, except acetonitrile was used instead of methanol. LC separation was
achieved using a Waters Xbridge C18 3.0 mm ꢁ 100 mm, 3.5 μm column,
and a linear gradient mobile phase from 55% to 95% MeOH in 0.1% formic
acid over 21 min at a flow rate of 350 μL/min. MS/MS analysis was
performed on a Sciex API 3000 triple-quad instrument monitoring the
transition m/z = 361 f m/z = 200 in positive ion mode. Quantitative
extraction from plasma was estimated by spiking untreated samples.
Concentrations were interpolated from a standard curve constructed from
analyte in MeOH/H2O (4ꢀ500 ng/mL).
1
yielded 8 as a light-yellow solid (5.0 g, yield 52%). H NMR (CDCl3,
400 MHz): δ 7.57 (dd, 1H), 7.13 (dd, 1H), 7.02 (dd, 1H), 3.72 (s, 2H).
1-(4-Bromo-3-fluorophenyl)cyclopropanecarbonitrile (9a). To a
solution of 8 (5.0 g, 23 mmol) and 1,2-dibromoethane (3 mL, 35 mmol)
in toluene (20 mL), 50% NaOH (20 mL), and TBAB (1.6 g, 5 mmol) were
added. The mixture was vigorously stirred at rt for 4 h and then diluted with
H2O (20 mL) and extracted with ethyl acetate (3 ꢁ 150 mL). The organic
layer was washed with 1 M HCl (25 mL) and then with brine (2 ꢁ 15 mL),
dried over Na2SO4, and concentrated to afford a brown solid (5.7 g).
Column chromatography (hexane/ethyl acetate [5:1]) afforded 9a as yellow
solid. 1H NMR (CDCl3, 400 MHz): δ 7.54ꢀ7.50 (t, 1H), 7.05ꢀ7.03 (dd,
1H, J = 7.48 Hz), 7.00ꢀ6.98 (dd, 1H, J = 7.48 Hz), 1.80ꢀ1.77 (q, 2H),
1.43ꢀ1.40 (q, 2H). 13C NMR (CDCl3, 100 MHz): 160.42 (d, J = 240 Hz),
138.01 (d, J = 6.9 Hz), 133.94, 122.54 (d, J = 3.5 Hz), 121.49, 114.07 (d, J =
8.4 Hz), 108.23 (d, 20.8 Hz), 18.71, 13.50.
1-(4-Bromo-3-fluorophenyl)diethylcarbonitrile (9b). To a solution
of 8 (235 mg, 1.1 mmol) and iodoethane (240 mg, 2.2 mmol) in
toluene (20 mL), 50% NaOH (20 mL) and TBAB (420 mg, 2.8 mmol)
were added. The mixture was vigorously stirred at rt for 4 h and then diluted
with H2O (20 mL) and extracted with ethyl acetate (3 ꢁ 100 mL). The
organic solution was washed with 1 M HCl and then with brine, dried over
Na2SO4, and concentrated to afford a brown solid (950 mg). Column
chromatography (hexane/ethyl acetate [5:1]) afforded 9b as white solid
(278 mg, yield 93.5%). 1HNMR(CDCl3, 400 MHz): δ7.73ꢀ7.70 (t, 1H),
7.18ꢀ7.15 (dd, 1H, J = 7.48 Hz), 7.12ꢀ7.09 (dd, 1H, J = 7.48 Hz),
2.10ꢀ2.01 (m, 2H), 1.92ꢀ1.83 (m, 2H), 0.95ꢀ0.91 (t, 6H). 13C NMR
(CDCl3, 100 MHz): 160.11 (d, J = 240 Hz), 139.82 (d, J = 6.9 Hz), 133.46,
122.83 (d, J = 3.5 Hz), 120.97, 114.28 (d, J = 2.4 Hz), 107.96 (d, 21.1 Hz),
49.19, 33.33, 9.82.
1-(4-Bromo-3-fluorophenyl)cyclopropanecarboxylic Acid (10). To
a solution of 9a (5.1 g, 21 mmol) in methanol (10 mL), 35% NaOH
(40 mL) was added and the mixture was heated to 100 °C for 8 h. After
cooling to rt, the mixture was acidified (pH 2) with 2 M HCl, and the pre-
cipitate was filtered off, washed with H2O, and redissolved in 5% NaHCO3.
Insoluble materials were filtered off and the solution acidified with 2 M HCl.
The precipitate was filtered, washed with H2O, and dried under reduced
pressure to afford 10 as a white solid (4.0 g, 72.5%). 1H NMR (CDCl3, 400
MHz): δ 11.82 (bs, 1H), 7.51ꢀ7.47 (t, 1H), 7.14ꢀ7.12 (dd, 1H, J = 9.39
Hz), 7.05ꢀ7.03 (dd, 1H, J = 9.39 Hz), 1.73ꢀ1.70 (q, 2H), 1.29ꢀ1.22 (q,
2H). 13C NMR (CDCl3, 100 MHz): 180.50, 159.83 (d, J = 246.1 Hz),
140.25 (d, J = 7.0 Hz), 133.15, 127.28 (d, J = 3.4 Hz), 118.78 (d, J = 22.2
Hz), 107.20 (d, J = 20.7 Hz), 28.30, 17.54.
General Procedure for Biaryl Suzuki Coupling. A suspension of the
appropriate aryl bromide (1 equiv), arylboronic acid (2 equiv), TBAB
(1.1 equiv), and Pd(OAc)2 (catalytic amount) in 2 M Na2CO3
(5ꢀ10 mL) was heated to 120 °C in a sealed tube for a period of
1ꢀ4 h. After cooling to rt, 2 M HCl (3 mL) and ethyl acetate (10 mL)
were added and the mixture filtered through a Celite pad. The organic
layer was separated, washed with brine (2 ꢁ 30 mL), dried over Na2SO4,
and concentrated in vacuo. The crude product was purified by flash
chromatography (hexane/ethyl acetate [1:1]) to afford the correspond-
ing biaryl compounds.
General Methods. Unless stated otherwise, all reactions were
carried out under an atmosphere of dry argon in oven-dried glassware.
Indicated reaction temperatures refer to those of the reaction bath, while
room temperature (rt) is noted as 25 °C. CH2Cl2 was distilled over
CaH2, and THF was distilled over Na(s)/benzophenone. All other
solvents were of anhydrous quality purchased from Aldrich Chemical
Co. and used as received. Commercially available starting materials and
reagents were purchased from Aldrich, TCI, and Fisher Scientific and
were used as received unless specified otherwise. Indomethacin and
sulindac libraries were provided by Prof. Larry Marnett of Vanderbilt
University. Analytical thin layer chromatography (TLC) was performed
with Sorbent Silica glass backed TLC plates (20 cm ꢁ 20 cm, 60 Å,
250 μm) with UV detection at 254 nm. 1H and 13C NMR spectra were
recorded on either a Bruker Avance 400 MHz spectrometer or a Bruker
DPX 400 MHz spectrometer. Chemical shifts are reported in ppm with
the solvent resonance as internal standard ([CDCl3 7.27 ppm, 77.23
ppm] [DMSO-d6 2.5 ppm, 39.51 ppm] and [MeOD-d4 4.78, 49.0] for
1H, 13C, respectively). Data are reported as follows: chemical shift,
number of protons, multiplicity (s = singlet, d = doublet, dd = doublet of
doublet, t = triplet, q = quartet, bs = broad singlet, m = multiplet), and
coupling constants. Compounds submitted for biological testing were
found to be >95% pure via analytical HPLC analysis using Shimadzu
HPLC. High resolution mass spectral data was collected in-house using
Shimadzu LCMS IT-TOF.
1-Bromo-4-bromomethyl-2-fluorobenzene (7).45. N-Bromosuccini-
mide (11.39 g, 640 mmol) was added to a solution of 1-bromo-2-fluoro-4-
methylbenzene 6 (5.36 mL, 8.01 g, 42 mmol) in CCl4 (100 mL). The
mixture was heated to 80 °C, and then dibenzoyl peroxide (0.27 g, 32
mmol) was added. The mixture was stirred at this temperature for 1 h and
then cooled to rt and washed in sequence with H2O (2 ꢁ 80 mL) and brine
2301
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
1-(2-Fluorobiphenyl-4-yl)cyclopropanecarboxylic Acid (11a). The
general procedure using 10 (259 mg, 1 mmol), phenylboronic acid (244
mg, 2 mmol), TBAB (340 mg, 1 mmol), and Pd(OAc)2 (30 mg) afforded
the title compound as a white solid (yield 73.7%). 1H NMR (CDCl3, 400
MHz): δ 7.63ꢀ7.55 (t, 4H), 7.50ꢀ7.36 (m, 2H), 7.24ꢀ7.16 (q, 2H),
1.76ꢀ1.73 (q, 2H), 1.35ꢀ1.33 (q, 2H). 13C NMR (CDCl3, 100 MHz):
178.68, 160.63 (J = 243.2 Hz), 140.60 (d, J = 8.0 Hz), 135.70, 130.56,
129.15, 129.13, 128.61, 128.21, 128.08, 127.84, 126.56 (d, J = 3.0 Hz),
118.45 (d, J = 23.0 Hz), 28.01, 17.90. ESI-HRMS (m/z): [M ꢀ H]þ calcd
for C16H13FO2, 255.0826; observed, 255.0832.
2-(2-Fluorobiphenyl-4-yl)acetonitrile (13a). The general procedure
using 8 (428 mg, 2.0 mmol), phenylboronic acid (490 mg, 4.0 mmol),
TBAB (710 mg, 2.2 mmol), and Pd(OAc)2 (67 mg) afforded the title
compound as a yellow solid (350 mg, yield 82.8%). 1H NMR (CDCl3, 400
MHz): δ 7.55ꢀ7.42 (m, 6H), 7.18ꢀ7.10 (m, 2H), 3.75 (s, 2H). 13C NMR
(CDCl3, 100 MHz): 159.40 (d, J = 249.7 Hz), 134.52, 131.10 (d, J =
3.8 Hz), 130.61 (d, J = 8.0 Hz), 129.26, 128.58, 128.55, 128.21, 127.93,
127.67, 123.56 (d, J = 3.4 Hz), 116.93, 115.53 (d, J = 24.5 Hz), 22.76.
2-(2,40-Difluorobiphenyl-4-yl)acetonitrile (13b). The general proce-
dure using 8 (500 mg, 2.1 mmol), 4-F-phenylboronic acid (560 mg, 4.0
mmol), TBAB (710 mg, 2.2 mmol), and Pd(OAc)2 (45 mg) afforded the
title compound as a yellow solid (480 mg, yield 90%). 1H NMR (CDCl3,
400 MHz): δ 7.51ꢀ7.39 (m, 3H), 7.19ꢀ7.11 (m, 4H), 3.81 (s, 2H). 13C
NMR (CDCl3, 100 MHz): 162.21 (d, J = 247.6 Hz), 158.04, 130.91 (d,
J = 3.6 Hz), 130.70 (d, J = 8.0 Hz), 130.44, 130.29 (d, J = 2.9 Hz), 130.20
(d, J = 2.9 Hz), 127.59 (d, J = 13.5 Hz), 124.80, 123.61 (d, J = 3.5 Hz),
116.84, 115.57 (d, J = 24.6 Hz), 115.18 (d, J = 21.7 Hz), 22.75.
Preparation of Fluoromethylnitriles (14a, 14b, 15a, 15b).46 To a
solution of the appropriate benzylic nitrile in anhydrous THF (approxi-
mately 5ꢀ10 mL of THF/1 mmol of nitrile) at ꢀ78 °C, t-BuLi (2.2
equiv for difluorination, 1.1 equiv for monofluorination) was added over
a period of 2 min. The resulting orange to dark-red solution was stirred
for 1 h at ꢀ78 °C. A solution of NFSi (2.5 equiv for difluorination, 1.3
equiv for monofluorination) in anhydrous THF (approximately
2ꢀ4 mL of THF/mmol of NFSi) was added over 2 min. After this
addition, the solution was stirred for 3 h at ꢀ78 °C. The reaction was
quenched with 0.01 M HCl, and the resulting solution was extracted
with CH2Cl2 (2 ꢁ 100 mL). The combined organic layers were washed
with 5% NaHCO3 and brine, dried over Na2SO4, and concentrated in
vacuo to give a yellow residue. Purification was achieved using silica gel
flash chromatography (hexane/ethyl acetate [5:1]).
1-(2,40-Difluorobiphenyl-4-yl)cyclopropanecarboxylic Acid (11b).
The general procedure using 10 (150 mg, 0.58 mmol), 4-F-phenylboronic
acid(162mg, 1.16 mmol), TBAB(206mg, 0.64 mmol), andPd(OAc)2 (20
mg) afforded the title compound as a white solid (yield 75.5%). 1H NMR
(CDCl3, 400 MHz): δ 7.70ꢀ7.33 (m, 5H), 7.24ꢀ7.17 (q, 2H), 1.76ꢀ1.70
(q, 2H), 1.34ꢀ1.32 (q, 2H). 13CNMR(CDCl3, 100 MHz): 179.96, 140.99,
132.57, 131.95, 131.32, 130.69, 130.08 (d, J = 3.6 Hz), 128.90, 128.21 (d, J =
3.4 Hz) 127.16, 126.54, 127.61, 118.51 (d, J = 22.8 Hz), 28.35, 17.54. ESI-
HRMS (m/z): [M ꢀ H]þ calcd for C16H12F2O2, 273.0733; observed,
273.0746.
1-(2-Fluoro-40-(trifluoromethyl)biphenyl-4-yl)cyclopropanecarboxy-
lic Acid (11c). The general procedure using 10 (259 mg, 1.0 mmol),
4-CF3-phenylboronic acid (228 mg, 1.2 mmol), TBAB (322 mg, 1.0
mmol), and Pd(OAc)2 (40 mg) afforded the title compound as a white
1
solid (yield 67.8%). H NMR (CDCl3, 400 MHz): δ 11.91 (bs, 1H),
7.72ꢀ7.65 (q, 4H), 7.42ꢀ7.38 (t, 1H), 7.26ꢀ7.19 (q, 2H), 1.76 (s, 2H),
1.35 (s, 2H). 13C NMR (CDCl3, 100 MHz): 180.29, 160.06, 157.58,
140.68 (d, J = 3.2 Hz), 138.65, 129.95, 129.55, 129.22, 128.89 (d, J = 1.3
Hz), 126.47 (d, J = 13.3 Hz), 126.21 (d, 2.9 Hz), 125.02 (d, J = 3.8 Hz),
118.13, 117.90, 28.04, 17.20. ESI-HRMS (m/z): [M ꢀ H]þ calcd for
C17H12F4O2, 323.0701; observed, 323.0710.
2-Fluoro-2-(2-fluorobiphenyl-4-yl)acetonitrile (14a). The general
procedure using 13a (192 mg, 0.91 mmol), t-BuLi (1.2 mL, 2.0 mmol),
and NFSi (710 mg, 2.4 mmol), yielding after purification a mixture of
14a (yellow solid, 42 mg, yield 20.1%) and 15a (yellow solid, 111 mg,
yield 49.4%). 14a: 1H NMR (CDCl3, 400 MHz) δ 7.59ꢀ7.38 (m, 8H),
6.11ꢀ6.09 (d, 1H, J = 11.8 Hz). 13C NMR (CDCl3, 100 MHz): 159.73
(d, J = 250.5 Hz), 134.34, 132.15 (d, J = 6.7 Hz), 132.11, 128.98, 128.79,
128.65, 128.49, 127.92, 123.25 (d, J = 3.9 Hz), 115.39 (d, J = 25.2 Hz),
114.77 (d, J = 33.3 Hz), 79.18 (d, J = 182.8 Hz), 60.37. APCI-HRMS
(m/z): [M ꢀ H]þ calcd for C14H9F2N, 228.0630; observed, 228.0625.
1-(30,40-Dichloro-2-fluorobiphenyl-4-yl)cyclopropanecarboxylic Acid
(11d). The general procedure using 10 (259 mg, 1.0 mmol), 3,4-
dichlorophenylboronic acid (190 mg, 1.2 mmol), TBAB (322 mg, 1.0
mmol), and Pd(OAc)2 (25 mg) afforded the title compound as a white
solid (yield 72.9%) 1H NMR (CDCl3, 400 MHz): δ 7.52ꢀ7.49 (t, 2H),
7.37ꢀ7.33 (t, 1H), 7.23ꢀ7.01 (m, 4H), 1.75ꢀ1.72 (q, 2H), 1.35ꢀ1.32
(q, 2H). 13C NMR (CDCl3, 100 MHz): 179.96, 160.10 (d, J = 246.5
Hz), 140.99 (d, 9.2 Hz), 132.57, 130.69, 130.06, 129.55, 129.22, 128.90,
128.19, 127.61, 126.54, 118.45 (d, J = 22.8 Hz), 28.35, 17.54. ESI-HRMS
(m/z): [M ꢀ H]þ calcd for C16H11Cl2FO2, 323.0047; observed,
323.0050.
2-Ethyl-2-(2-fluorobiphenyl-4-yl)butanenitrile (12a). The general
procedure using 9b (150 mg, 0.56 mmol), phenylboronic acid (177.3
mg, 0.83 mmol), TBAB (206 mg, 0.64 mmol), and Pd(OAc)2 (20 mg)
afforded the title compound as a white solid (140 mg, yield 94.3%). 1H
NMR (CDCl3, 400 MHz): δ 7.56ꢀ7.54 (d, 2H), 7.47ꢀ7.43 (m, 3H),
7.40ꢀ7.36 (t, 1H), 7.27ꢀ7.25 (d, 1H), 7.25ꢀ7.19 (d, 1H), 2.12ꢀ2.03
(m, 2H), 1.97ꢀ1.88 (m, 2H), 0.98ꢀ0.90 (t, 6H). 13C NMR (CDCl3,
100 MHz): 160.05 (d, J = 242.8 Hz), 139.36 (d, J = 7.0 Hz), 135.42,
131.44, 129.35 (d, J = 4.0 Hz), 129.59, 128.94, 127.52, 122.50, 121.91,
118.5 (d, J = 22.8 Hz) 113.79, 113.55, 51.17, 33.38, 10.51.
1
2,2-Difluoro-2-(2-fluorobiphenyl-4-yl)acetonitrile (15a). H NMR
(CDCl3, 400 MHz): δ 7.55ꢀ7.32 (m, 8H). 13C NMR (CDCl3, 100
MHz): 159.58 (d, J = 251.0 Hz), 133.79, 133.74 (d, J = 13.6 Hz), 131.93
(d, J = 3.8 Hz), 131.76 (d, J = 7.8 Hz), 129.00 (d, J = 3.0 Hz), 128.78,
128.34, 124.93, 121.33 (q, J = 5.0 Hz), 116.82, 113.58 (d, J = 26.6 Hz),
112.19 (t, J = 47.9 Hz), 107.92 (t, J = 245.0 Hz). APCI-HRMS (m/z):
[M ꢀ H]þ calcd for C14H8F3N, 246.0536; observed, 246.0515.
2-(2,40-Difluorobiphenyl-4-yl)-2-fluoroacetonitrile (14b). Synthe-
sized via the general procedure using 13b (115 mg, 0.5 mmol), t-BuLi
(0.65 mL, 1.1 mmol), and NFSi (394 mg, 1.5 mmol), yielding after
purification a mixture of 14b (yellow solid, 56 mg, yield 42.1%) and 15b
1
(yellow solid, 32 mg, yield 32%). H NMR (CDCl3, 400 MHz): δ
2-Ethyl-2-(3-fluoro-4-(naphthalen-1-yl)phenyl)butanenitrile (12b).
The general procedure using 9b (51 mg, 0.19 mmol), 2-napthylboronic
acid (120 mg, 0.70 mmol), TBAB (125 mg, 0.21 mmol), and Pd(OAc)2
(20 mg) afforded the title compound as a yellow solid (45 mg, yield
75.1%). 1H NMR (CDCl3, 400 MHz): δ 8.03 (s, 1H), 7.93ꢀ7.88 (m,
3H), 7.58ꢀ7.50 (m, 4H), 7.31ꢀ7.24 (q, 2H), 2.15ꢀ2.05 (m, 2H),
2.03ꢀ1.88 (m, 2H), 0.98ꢀ0.89 (t, 6H). 13C NMR (CDCl3, 100 MHz):
162.09 (d, J = 246.2 Hz), 140.29 (d, J = 9.8 Hz), 133.75, 133.17, 133.87,
131.69 (d, J = 5.5 Hz), 128.80, 128.65, 128.50, 128.07, 127.18 (d, J = 4.1
Hz), 126.80 (d, J = 3.7 Hz), 124.22, 122.81 (d, J = 4.7 Hz), 122.21,
121.15, 115.60, 50.01, 34.21, 10.16.
7.54ꢀ7.30 (m, 5H), 7.15 (t, 2H, J = 8.4 Hz), 6.08 (d, 1H, J = 46.8 Hz).
13C NMR (CDCl3, 100 MHz): 162.50 (d, J = 248.6 Hz), 159.25 (d, J =
250.9 Hz), 131.82 (dd, J = 21.0 Hz, 7.7 Hz), 131.29, 130.74 (dd, J = 13.5
Hz, 2.9 Hz), 130.43 (d, J = 2.7 Hz), 130.34 (d, J = 2.9 Hz), 128.84, 122.96
(t, J = 3.8 Hz), 115.45, 115.24, 115.06 (dd, J = 25.4 Hz, 4.4 Hz), 114.41
(d, J = 33.2 Hz), 78.88 (d, J = 183.0 Hz). APCI-HRMS (m/z): [M ꢀ H]þ
calcd for C14H8F3N, 246.0536; observed, 246.0511.
2-(2,40-Difluorobiphenyl-4-yl)-2,2-difluoroacetonitrile (15b). 1H
NMR (CDCl3, 400 MHz): δ 7.52ꢀ7.37 (m, 5H), 7.09 (t, 2H, J = 8.0
Hz). 13C NMR (CDCl3, 100 MHz): 163.04 (d, J = 249.2 Hz), 159.45 (d,
J = 251.7 Hz), 132.68 (d, J = 13.2 Hz), 131.86 (d, J = 8.0 Hz), 131.73 (d,
2302
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
J = 3.6 Hz), 130.83 (d, J = 3.0 Hz), 130.75 (d, J = 3.0 Hz), 129.92, 129.88,
121.40 (q, J = 4.9 Hz), 115.83 (d, J = 21.6 Hz), 113.64 (dt, J = 26.6 Hz,
5.1 Hz), 112.12 (d, J = 48.3 Hz), 107.79 (t, J = 244.3 Hz). APCI-HRMS
(m/z): [M ꢀ H]þ calcd for C14H7F4N, 264.0442; observed, 264.0460.
General Procedure for Fluoromethylnitrile (14a,b and 15a,b) to
Fluoromethylamide (16a,b and 17a,b). To a stirred solution of
fluoronitrile in DMSO (30 mg/1.5 mL) was added K2CO3 and the
mixture was cooled to 0 °C. To this mixture was added dropwise H2O2 at
the same temperature, and the reaction mixture was stirred for 45 min at rt
under N2. The reaction was quenched by adding cold H2O, and the mixture
was extracted three times with ethyl acetate (3 ꢁ 75 mL). The combined
organic layers were washed with brine, dried over Na2SO4, and concen-
trated. The residue was purified by flash column chromatography (CH2Cl2/
MeOH [95:5]) to give corresponding fluoroamides.
2-(2,40-Difluorobiphenyl-4-yl)-2-hydroxyacetic Acid (18b). White
1
solid, yield 41.5%. H NMR (MeOD-d4, 400 MHz): δ 7.56ꢀ7.54 (t,
J = 6.8 Hz, 1H), 7.46ꢀ7.44 (t, 1H, J = 8.0 Hz), 7.39ꢀ7.31 (m, 3H),
7.18ꢀ7.16 (t, 2H, J = 8.8 Hz), 5.21 (s, 1H). 13C NMR (MeOD-d4, 100
MHz): 175.72, 164.05 (d, J = 245.7 Hz), 160.95, 160.18, 142.99 (d, J =
7.8 Hz), 133.13, 132.11 (d, J = 3.2 Hz), 132.04, 131.82 (d, J = 3.3 Hz),
129.01 (d, J = 13.7 Hz), 124.17 (d, J = 3.1 Hz), 116.43 (d, J = 21.8 Hz),
115.46 (d, J = 24.3 Hz), 73.42. ESI-HRMS (m/z): [M ꢀ H]þ calcd for
C14H10F2O3, 263.0525; observed, 263.0575.
2-(2-Fluorobiphenyl-4-yl)acetonitrile (19). The acid catalyzed hy-
drolysis procedure used in the preparation of 18a, starting with 13a (510
mg, 2.41 mmol) in the presence of H2SO4 (100 mL) afforded the
product as a white solid (480 mg, yield 86.3%). 1H NMR (CDCl3, 400
MHz): δ 7.55ꢀ7.37 (m, 6H), 7.21ꢀ7.15 (m, 2H), 3.80 (s, 2H).
Preparation of the Biaryl-Biethyl Ester (20). To a solution of 19 (350
mg, 1.5 mmol) and bromoethane (500 mg, 5 mmol) in toluene (20 mL),
50% NaOH (20 mL) and TBAB (700 mg, 5 mmol) were added. The
mixture was vigorously stirred at rt for 8 h and then diluted with H2O
(25 mL) and extracted with ethyl acetate (3 ꢁ 50 mL). The organic solution
was washed with 1 M HCl (1 ꢁ 50 mL) and then with brine, dried over
Na2SO4, and concentrated to afford a brown solid. Purification was achieved
using silica gel flash chromatography to give the product as a white solid
(300 mg, 70.3%). 1H NMR (CDCl3, 400 MHz): δ 7.50ꢀ7.36 (m, 7H),
7.05ꢀ7.01 (m, 1H), 4.34ꢀ4.32 (q, 2H), 2.12ꢀ2.02 (m, 2H), 1.99ꢀ1.90
(m, 2H), 1.01ꢀ0.98 (t, 6H), 0.71ꢀ0.70 (t, 3H). 13C NMR (CDCl3, 100
MHz): 174.30, 165.40, 140.05, 130.69, 130.28, 129.61, 128.93, 128.40,
127.58, 125.48, 122.69 (d, J = 4.82 Hz), 117.42, 113.56, 61.33, 52.09,
26.67, 14.30, 8.43.
Preparation of Biaryl-Diethyl Acid (21). The ester 20 (200 mg,
0.5 mmol) was dissolved in CH3CN (15 mL) with a few drops of H2O.
TEA (0.2 mL, 1.5 mmol) was then added, followed by LiBr (430 mg,
5 mmol), and the reaction was vigorously stirred at reflux for 6 h. The
reaction was cooled to rt and extracted with diethyl ether (3 ꢁ 50 mL).
The aqueous layer was then acidified to pH ∼ 2 with 1 M HCl and
extracted with ethyl acetate (3 ꢁ 50 mL). Combined organic extracts
were washed with brine (2 ꢁ 30 mL) and H2O (1 ꢁ 25 mL), dried over
Na2SO4, and concentrated to give the product as a white solid (56 mg,
39% conversion). 1H NMR (CDCl3, 400 MHz): δ 7.59ꢀ7.41 (m, 6H)
7.29ꢀ7.19 (m, 2H), 2.14ꢀ2.07 (m, 2H), 1.99ꢀ1.92 (m, 2H), 1.01ꢀ
0.94 (t, 6H). 13C NMR (CDCl3, 100 MHz): 178.25, 160.67 (d, J = 245.6
Hz), 139.39 (d, J = 7.4 Hz), 134.62, 130.65 (d, J = 3.9 Hz), 129.52,
128.55 (d, J = 2.8 Hz), 127.99, 127.54, 121.92 (d, J = 3.3 Hz), 121.42,
117.20, 113.82 (d, J = 24.8 Hz), 49.19, 33.38, 9.35. ESI-HRMS (m/z):
[M ꢀ H]þ calcd for C18H19FO2, 285.1296; observed, 285.1273.
Preparation of Bis-TBS Ketene Acetals (22a and 22b). The appro-
priate biaryl acid (1 equiv) and TBDSCl (2.3 equiv) were dissolved in
THF at rt. LiHMDS (2.2 equiv) was introduced dropwise at 0 °C for 15
min. The resulting reaction mixture was allowed to warm to rt and stirred
overnight before being concentrated under vacuum. The residual oil was
redissolved in hexanes, filtered, and concentrated. After confirmation of
the olefinic proton and TBDS protection by 1H NMR, the product was
used in the next step without further purification.
2-Fluoro-2-(2-fluorobiphenyl-4-yl)acetamide (16a). White solid,
1
yield 75.4%. H NMR (CDCl3, 400 MHz): δ 7.59ꢀ7.38 (m, 8H),
6.11ꢀ6.09 (d, 1H, J = 11.8 Hz). 13C NMR (CDCl3, 100 MHz): 171.10,
159.73 (d, J = 250.5 Hz), 136.52, 134.34, 132.15 (d, J = 6.7 Hz), 132.11,
128.98, 128.79, 128.65, 128.49, 123.25 (d, J = 3.9 Hz), 115.39 (d, J = 25.2
Hz), 114.77 (d, J = 33.3 Hz), 79.18 (d, J = 182.8 Hz). ESI-HRMS (m/z):
[M þ Na]þ calcd for C14H11F2NO, 270.0701; observed, 270.0709.
2-(2,40-Difluorobiphenyl-4-yl)-2-fluoroacetamide (16b). White so-
lid, yield 82.4%. 1H NMR (400 MHz, CDCl3): δ 7.54ꢀ7.30 (m, 5H),
7.15ꢀ7.13 (t, 2H, J = 8.4 Hz), 6.08ꢀ6.02 (d, 1H, J = 46.8 Hz,). 13C
NMR (CDCl3, 100 MHz): 171.52, 162.50 (d, J = 248.6 Hz), 159.25 (d,
J = 250.9 Hz), 131.82 (dd, J = 21.0 Hz, 7.7 Hz), 131.29, 130.74 (dd, J =
13.5 Hz, 2.9 Hz), 130.43 (d, J = 2.7 Hz), 130.34 (d, J = 2.9 Hz), 128.84,
122.96 (t, J = 3.8 Hz), 115.45, 115.06 (dd, J = 25.4 Hz, 4.4 Hz), 114.41
(d, J = 33.2 Hz), 78.88 (d, J = 183.0 Hz). ESI-HRMS (m/z): [M þ Na]þ
calcd for C14H10F3NO, 288.0607; observed, 288.0609.
2,2-Difluoro-2-(2-fluorobiphenyl-4-yl)acetamide (17a). White so-
lid, yield 81.0%. 1H NMR (DMSO-d6, 400 MHz): δ 8.43 (s, 1H), 8.11
(s, 1H), 7.71ꢀ7.69 (t, 1H, J = 8.0 Hz), 7.76ꢀ7.46 (m, 7H). 13C NMR
(DMSO-d6, 400 MHz): 171.60, 160.63 (d, J = 242.5 Hz), 134.58, 131.22
(d, 3.8 Hz), 128.99 (d, J = 2.9 Hz), 128.58, 128.32, 127.90, 127.63,
125.63, 121.55 (d, J = 3.9 Hz), 117.59, 114.09 (d, J = 10.06 Hz), 113.96
(d, J = 19.5 Hz). ESI-HRMS (m/z): [M þ H]þ calcd for C14H10F3NO,
266.0787; observed, 266.0797.
2-(2,40-Difluorobiphenyl-4-yl)-2,2-difluoroacetamide (17b). White
solid, yield 87.9%. 1H NMR (DMSO-d6, 400 MHz):δ 8.42 (s, 1H), 8.10 (s,
1H), 7.71ꢀ7.62 (m, 3H), 7.53ꢀ7.49 (m, 2H), 7.34 (t, 2H, J = 8.8 Hz). 13C
NMR (DMSO-d6, 100 MHz): 164.77 (t, J = 30.7 Hz), 162.20 (d, J = 245.8
Hz), 158.61 (d, = 247.4 Hz), 134.62 (td, J = 25.7 Hz, 8.1 Hz), 131.50 (d, J =
3.2 Hz), 131.12 (d, J = 2.7 Hz), 131.03 (d, J = 3.0 Hz), 130.36, 129.78 (d, J=
13.0 Hz), 127.64, 121.78, 115.71 (d, J = 21.5 Hz), 113.81 (t, J = 254.0 Hz),
113.82 (dt, J = 25.8 Hz, 6.4 Hz). ESI-HRMS (m/z): [M ꢀ H]þ calcd for
C14H9F4NO, 282.0547; observed, 282.0544.
General Procedure for Acid Hydrolysis of Fluoromethylnitriles (16a,
16b) to Fluoromethylacetic Acids (18a, 18b). A vigorously stirred
emulsion of the fluoromethylnitrile in a 49% aqueous solution of H2SO4
(5 mL per 100 mg of substrate) was heated at 80 °C for 3 h. After being
cooled to rt, the aqueous phase was extracted with diethyl ether (3 ꢁ
50 mL). Acids were washed out from the combined organic phases with
satd NaHCO3. The aqueous phase was acidified to pH 2 and extracted
with Et2O (3 ꢁ 50 mL). The combined extracts were dried over Na2SO4
and concentrated. The residue was recrystallized from chloroform:
petroleum ether.
General Procedure for Fluorination of the R-Carbon Using
Selectfluor.47 The bis-TBS ketene acetal (1.0 equiv) in MeCN
(10 mL) was added dropwise to Selectfluor (1.5 equiv) in MeCN for
15 min while maintaining reaction <50 °C. The reaction was allowed to
warm to rt and stirred for an additional 15 min. The reaction was then
diluted with 1 M HCl (25 mL) and extracted with ethyl acetate (3 ꢁ
50 mL). The combined organic layer was extracted with 1 M NaOH (3 ꢁ
50 mL). The aqueous layer was washed with ethyl acetate (3 ꢁ 50 mL),
followed by acidification of the aqueous layer to pH ∼1 with 3 M HCl. The
acidic aqueous layer was extracted with ethyl acetate (3 ꢁ 75 mL), and
combined organic extracts were dried over Na2SO4 and concentrated.
2-(2-Fluorobiphenyl-4-yl)-2-hydroxyacetic Acid (18a). White solid,
1
yield 56.0%. H NMR (MeOD-d4, 400 MHz): δ 7.54ꢀ7.33 (m, 8H),
5.21 (s, 1H). 13C NMR (MeOD-d4, 100 MHz): 175.72, 165.27, 162.83,
159.73, 143.03 (d, J = 7.8 Hz), 142.95, 133.13, 132.12, 131.84 (d, J = 3.3
Hz), 129.08 (d, J = 13.6 Hz), 122.20 (d, J = 3.1 Hz), 116.54 (d, J = 21.8
Hz), 115.58 (d, J = 24.3 Hz), 73.42. ESI-HRMS (m/z): [M ꢀ H]þ calcd
for C14H11FO3, 245.0609; observed, 245.0639.
2303
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
2-Fluoro-2-(2-fluorobiphenyl-4-yl)propanoic Acid (23a). Synthe-
sized using fluorination general procedure using the following values:
22a (232 mg, 0.491 mmol), Selectfluor (226 mg, 0.638 mmol), yielding
23a as a white solid (42 mg, yield 32.6%). 1H NMR (CDCl3, 400 MHz):
δ 10.15 (bs, 1H), 7.44ꢀ7.27 (m, 6H), 7.14ꢀ7.05 (m, 2H), 1.52 (d, 3H,
J = 7.18 Hz). 13C NMR (CDCl3, 100 MHz): 171.90, 161.52 (d, J = 242.2
Hz), 141.98 (d, J = 7.4 Hz), 134.80, 132.21, 129.84, 129.60, 128.97,
128.94, 128.46, 127.98, 123.60, 115.42 (d, J = 24.8 Hz), 81.81, 19.02.
ESI-HRMS (m/z): [M ꢀ H]þ calcd for C15H12F2O2, 261.0733;
observed, 261.0706.
2-bromoethanol (1.0ꢀ1.5 equiv). After stirring overnight, the solution was
concentrated and purified by column chromatography.
2-Bromoethyl-1-(2-fluorobiphenyl-4-yl)cyclopropanecarboxylate (25a).
Synthesized using general procedure with the following values: 11a (130 mg,
0.50 mmol), oxalyl chloride (1.25 mL, 2.5 mmol), 2-bromoethanol (0.1 mL,
0.8 mmol), and TEA (0.5 mL) afforded 25a as colorless oil (160 mg, yield
86.8%). 1H NMR (CDCl3, 400 MHz): δ 7.59ꢀ7.39 (m, 6H), 7.25ꢀ7.21
(qs, 2H), 4.41ꢀ4.38 (t, 2H), 3.50ꢀ3.46 (t, 2H), 1.75ꢀ1.71 (q, 2H),
1.33ꢀ1.29 (q, 2H). 13C NMR (CDCl3, 100 MHz): 173.68, 159.8 (d, J =
246.6 Hz), 140.57 (d, J = 8.0 Hz), 135.70, 130.56, 129.15, 129.13, 128.61,
128.21, 128.08, 127.84, 126.56, 126.53, 118.32 (d, J = 23.1 Hz), 64.39,
28.87, 17.19. ESI-HRMS (m/z): [M þ H]þ calcd for C18H16BrFO2,
363.0390; observed, 363.0396.
2-Fluoro-2-(2-fluorobiphenyl-4-yl)acetic Acid (23b). Synthesized
using fluorination general procedure using the following values: 22b
(58 mg, 0.13 mmol), Selectfluor (40 mg, 0.10 mmol), yielding 23b as a
1
white solid (12 mg, yield 38.2%). H NMR (CDCl3, 400 MHz): δ
2-Bromoethyl-1-(2-fluoro-40-(trifluoromethyl)biphenyl-4-yl)cyclo-
propanecarboxylate (25b). Synthesized using general procedure with
the following values: 11c (270 mg, 0.83 mmol), oxalyl chloride (2.1 mL,
4.15 mmol), 2-bromoethanol (0.3 mL, 2.4 mmol), and TEA (1.0 mL)
afforded 25b as colorless oil (280 mg, yield 78%). 1H NMR (CDCl3, 400
MHz): δ 7.73ꢀ7.67 (q, 4H), 7.43ꢀ7.39 (t, 1H), 7.28ꢀ7.21 (qs, 2H),
4.42ꢀ4.39 (t, 2H), 3.50ꢀ3.47 (t, 2H), 1.80ꢀ1.73 (q, 2H), 1.33ꢀ1.26
(q, 2H). 13C NMR (CDCl3, 100 MHz): 173.31, 160.44 (d, 246.6 Hz),
141.55, 139.11, 130.25, 130.21, 129.86, 129.27, 129.24, 126.62, 126.59,
125.34, 125.30, 122.80, 118.46, 64.26, 28.60, 16.99. ESI-HRMS (m/z):
[M þ H]þ calcd for C19H15BrF4O2, 431.0264; observed, 431.0264.
2-Bromoethyl-1-(30,40-dichloro-2-fluorobiphenyl-4-yl)cyclopropane-
carboxylate (25c). Synthesized using general procedure with the
following values: 11d (260 mg, 0.80 mmol), oxalyl chloride (2.0 mL,
4.0 mmol), 2-bromoethanol (0.1 mL, 0.8 mmol), and TEA (0.5 mL)
afforded 25c as a white solid (300 mg, yield 87.8%). 1H NMR (CDCl3,
400 MHz): δ 7.69 (s, 1H), 7.65ꢀ7.59 (d, 1H), 7.54ꢀ7.18 (m, 4H),
4.40ꢀ4.37 (t, 2H), 4.14ꢀ4.12 (t, 2H), 1.73ꢀ1.70 (q, 2H), 1.29ꢀ1.24
(q, 2H). 13C NMR (CDCl3, 100 MHz): 172.92, 159.92 (d, J = 242.0
Hz), 141.14 (d, J = 8.0 Hz), 135.04, 132.17, 131.50, 130.35, 129.72,
129.61, 129.57, 127.84, 126.25, 125.12, 118.12 (d, J = 23.0 Hz), 63.89,
28.32, 16.66. ESI-HRMS (m/z): [M]þ calcd for C18H14BrCl2FO2,
430.9538; observed, 430.9611.
General Procedure for Nitration of the Ethyl Bromide Esters (25aꢀc).
To a solution of the appropriate bromide (1 equiv) in anhydrous MeCN
was added AgNO3 (2ꢀ5 equiv), and the reaction was stirred at reflux for
2 h. The mixture was filtered through Celite, the filtrate concentrated, and
the resulting crude oil purified by column chromatography (hexane/ethyl
acetate [3:1]) to give the desired nitrates characterized below.
2-(Nitrooxy)ethyl 1-(2-Fluorobiphenyl-4-yl)cyclopropanecarboxylate
(26a). Synthesized using general nitration procedure with the following
values: 25a (120 mg, 0.33 mmol) and AgNO3 (112 mg, 0.66 mmol)
afforded 26a as yellow oil (85 mg, yield 74.5%). 1H NMR (CDCl3, 400
MHz): δ 7.58ꢀ7.38 (m, 6H), 7.27ꢀ7.14 (m, 2H), 4.64ꢀ4.63 (t, 2H),
4.36ꢀ4.35 (t, 2H), 1.70ꢀ1.68 (m, 2H), 1.32ꢀ1.30 (m, 2H). 13C NMR
(CDCl3, 100 MHz): 173.35, 159.90 (d, J = 242.0 Hz), 141.12(d, J = 8.0
Hz), 139.83, 135.11, 130.07, 129.53, 128.59, 128.06, 127.31, 125.93, 117.79,
117.56, 69.67, 60.75, 28.13, 16.76. ESI-HRMS (m/z): [M þ Na]þ calcd for
C18H16FNO5, 368.0905; observed, 368.0911.
7.55ꢀ7.52 (m, 2H), 7.43ꢀ7.32 (m, 4H), 7.14ꢀ7.09 (m, 2H), 5.40ꢀ
4.28 (d, 1H). 13C NMR (CDCl3, 100 MHz): 174.9, 161.0 (d, J = 242.2
Hz), 141.9 (d, J = 7.4 Hz), 135.50, 132.05, 130.9 (d, J = 3.2 Hz), 129.52,
129.03 (d, J = 13.7 Hz), 128.51, 128.0, 127.83, 123.62, 115.40 (d, J =
24.8 Hz), 72.60. ESI-HRMS (m/z): [M ꢀ H]þ calcd for C14H10F2O2,
247.0576; observed, 247.0601.
4-Bromobutyl R-(ꢀ)-2-Fluoro-a-methyl-4-diphenylacetate (24a).
Flurbiprofen (1.0 g, 4.09 mmol) was dissolved in methanol (10 mL),
and then 0.5 M sodium methoxide in methanol (8.18 mL, 4.09 mmol)
was added. The reaction mixture was stirred at rt for 10 min, and then the
solvent was evaporated under reduced pressure, yielding the sodium salt
(1.08 g, 4.09 mmol), which was suspended in dimethylformide (10 mL)
and stirred for 10 min, followed by dropwise addition of 1,4-dibromobutane
(1.5 mL, 2.74 g, 12.6 mmol). The reaction mixture was stirred at rt for an
additional 48 h. The reaction mixture was poured into H2O (100 mL) and
extracted with ethyl acetate (3 ꢁ 50 mL). The organic layers were washed
with H2O (3 ꢁ 30 mL) and brine (3 ꢁ 50 mL), dried over Na2SO4, and
evaporated under reduced pressure. The resulting residue was purified by
column chromatography (hexane/CH2Cl2 [1/1]) to give the product as
yellow oil (1.35 g, 87.5%). 1H NMR (CDCl3, 400 MHz): δ 7.53ꢀ7.50 (m,
2H), 7.45ꢀ7.34 (m, 4H), 7.14ꢀ7.07 (m, 2H), 4.11ꢀ4.09 (t, 2H, J = 6.15
Hz), 3.73ꢀ3.71 (q, 1H, J = 7.15 Hz), 3.34ꢀ3.32 (t, 2H, J = 6.28 Hz),
1.86ꢀ1.73 (m, 4H), 1.52ꢀ1.50 (d, 3H, J = 7.17 Hz).
N-(4-Bromobutyl)-2-(2-fluorobiphenyl-4-yl)propanamide (24b). A
solution of (R)-2-(2-fluorobiphenyl-4-yl)propanoic acid (0.60 g, 2.4
mmol) in SOCl2 (4.4 g, 36.9 mmol) was heated at reflux. After 2 h, the
reaction mixture was allowed to cool to rt and the excess of SOCl2
removed under vacuum. Both a solution of the residual material in
CH2Cl2 (2 mL) and a solution of NaOH (98.3 mg, 2.40 mmol) in H2O
(0.32 mL) were simultaneously added to a cooled (ꢀ20 °C) solution of
4-amino-1-butanol (438.2 mg, 4.92 mmol) in CH2Cl2 (2 mL) under
vigorous stirring. The reaction was kept at ꢀ20 °C for 2 h and allowed to
warm to rt overnight. The organic layer was washed with H2O, NaHCO3,
and brine, and concentrated to afford the intermediate alcohol (50% yield)
as an oil, which was used in the subsequent procedure without further
purification. The intermediate alcohol (390 mg, 1.24 mmol) and PPh3
(356.3 mg, 1.36 mmol) in CH2Cl2 (8 mL) were cooled to 5 °C, and a
solution of tetrabromomethane (493 mg, 1.48 mmol) in CH2Cl2 (4 mL)
was added dropwise over 10 min. The resulting clear solution was stirred at rt
for 4 h, before removal of solvent under vacuum. The concentrate was
purified by flash chromatography (hexane/ethyl acetate [20:80]) to yield
pure 24b (64% yield) as an oil. 1H NMR (CDCl3, 400 MHz): 7.56ꢀ7.53
(m, 2H), 7.47ꢀ7.35 (m, 4H), 7.16ꢀ7.11 (m, 2H), 3.57ꢀ3.54 (q, 1H),
3.39ꢀ3.37 (t, 2H), 3.31ꢀ3.24 (m, 2H), 1.87ꢀ1.78 (m, 2H), 1.67ꢀ1.60
(m, 2H), 1.55ꢀ1.53 (d, 3H, J = 6.9 Hz).
2-(Nitrooxy)ethyl 1-(2-Fluoro-40-(trifluoromethyl)biphenyl-4-yl)cyclo-
propanecarboxylate (26b). Synthesized using general nitration procedure
with the following values: 25b (220 mg, 0.49 mmol) and AgNO3 (415 mg,
2.44 mmol) afforded 26b as colorless oil (190 mg, yield 90.1%). 1H NMR
(CDCl3, 400 MHz): δ 7.73ꢀ7.67 (q, 4H), 7.43ꢀ7.39 (t, 1H), 7.25ꢀ7.18
(qs, 2H), 4.67ꢀ4.64 (t, 2H), 4.38ꢀ4.36 (t, 2H), 1.73ꢀ1.70 (q, 2H),
1.33ꢀ1.31 (q, 2H). 13C NMR (CDCl3, 100 MHz): 173.54, 160.46 (d, J =
240.2 Hz), 141.33(d, J = 8.0 Hz), 139.10, 130.36 (d, J = 3.6 Hz), 129.85,
129.29 (d, J = 3.1 Hz), 128.23, 126.58 (d, J = 3.3 Hz), 125.38 (q, J = 11.2
Hz), 124.10, 122.83, 118.35, 118.12, 70.00, 61.20, 28.54, 17.09. ESI-HRMS
(m/z): [M-NO3]þ calcd for C18H16FNO5,: 351.1008; observed, 351.1029.
General Procedure for Alkylation of Biaryl Cyclopropyl Acids (11a,
11c, 11d). To a solution of the appropriate carboxylic acid (1 equiv) in
CH2Cl2 was added oxalyl chloride (5 eq, 2.0 M in CH2Cl2), andthemixture
was stirred at rt for 5 h. The mixture was concentrated and redissolved in
CH2Cl2 containing TEA (2.5 equiv), followed by the dropwise addition of
2304
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
2-(Nitrooxy)ethyl 1-(30,40-Dichloro-2-fluorobiphenyl-4-yl)cyclopro-
panecarboxylate (26c). Synthesized using general nitration procedure
with the following values: 25c (300 mg, 0.69 mmol) and AgNO3 (589 mg,
3.47 mmol) afforded 26c as colorless oil (210 mg, yield 73.0%). 1H NMR
(CDCl3, 400 MHz): δ 7.65ꢀ7.16 (m, 6H), 4.66ꢀ4.64 (t, 2H), 4.38ꢀ
4.35 (t, 2H), 1.75ꢀ1.68 (q, 2H), 1.35ꢀ1.28 (q, 2H). 13C NMR (CDCl3,
100 MHz): 172.92, 159.92 (d, J = 242.0 Hz), 141.14 (d, J = 8.0 Hz), 135.04,
132.17, 131.50, 130.35 (d, J= 1.8 Hz), 130.01, 129.72, 129.61, 127.88 (d, J=
4.0 Hz), 126.28 (d, J= 3.8 Hz), 125.26, 118.12 (d, J= 23.0 Hz), 69.59, 60.84,
28.16, 16.76. ESI-HRMS (m/z): [M ꢀ NO3]þ calcd for C18H14Cl2FNO5,
351.0349; observed 351.0536.
’ REFERENCES
(1) Blennow, K.; de Leon, M. J.; Zetterberg, H. Alzheimer’s disease.
Lancet 2006, 368, 387–403.
(2) Lee, V. M.; Trojanowski, J. Q. The disordered neuronal cytos-
keleton in Alzheimer’s disease. Curr. Opin. Neurobiol. 1992, 2, 653–656.
(3) Hardy, J.; Selkoe, D. J. The amyloid hypothesis of Alzheimer’s
disease: progress and problems on the road to therapeutics. Science 2002,
297, 353–356.
(4) Weggen, S.; Eriksen, J. L.; Das, P.; Sagi, S. A.; Wang, R.; Pietrzik,
C. U.; Findlay, K. A.; Smith, T. E.; Murphy, M. P.; Bulter, T.; Kang, D. E.;
Marquez-Sterling, N.; Golde, T. E.; Koo, E. H. A subset of NSAIDs lower
amyloidogenic Abeta42 independently of cyclooxygenase activity. Nat-
ure 2001, 414, 212–216.
4-(Nitrooxy)butyl 2-(2-fluorobiphenyl-4-yl)propanoate (2). The bro-
mide 24a (1.34 g, 3.58 mmol) was dissolved in anhydrous MeCN (20 mL),
and then AgNO3 (1.84 g, 10.74 mmol) was added. The suspension was
stirred at rt for 1 h and then at reflux 14 h. The solid was filtered through
Celite, and the solvent was removed under vacuum. Flash column chroma-
tography of the residue (hexane/ethyl acetate [7:3]) afforded the desired
(5) Findeis, M. A. The role of amyloid beta peptide 42 in Alzheimer’s
disease. Pharmacol. Ther. 2007, 116, 266–286.
(6) Reinhard, C.; Hebert, S. S.; De Strooper, B. The amyloid-beta
precursor protein: integrating structure with biological function. EMBO
J. 2005, 24, 3996–4006.
1
product as yellow oil (1.11 g, 85.9%). H NMR (CDCl3, 400 MHz): δ
(7) Eriksen, J. L.; Sagi, S. A.; Smith, T. E.; Weggen, S.; Das, P.;
McLendon, D. C.; Ozols, V. V.; Jessing, K. W.; Zavitz, K. H.; Koo, E. H.;
Golde, T. E. NSAIDs and enantiomers of flurbiprofen target gamma-
secretase and lower Abeta 42 in vivo. J. Clin. Invest. 2003, 112, 440–449.
(8) Narlawar, R.; Perez Revuelta, B. I.; Haass, C.; Steiner, H.;
Schmidt, B.; Baumann, K. Scaffold of the cyclooxygenase-2 (COX-2)
inhibitor carprofen provides Alzheimer gamma-secretase modulators.
J. Med. Chem. 2006, 49, 7588–7591.
(9) Imbimbo, B. P.; Del Giudice, E.; Colavito, D.; D’Arrigo, A.; Dalle
Carbonare, M.; Villetti, G.; Facchinetti, F.; Volta, R.; Pietrini, V.; Baroc,
M. F.; Serneels, L.; De Strooper, B.; Leon, A. 1-(30,40-Dichloro-2-
fluoro[1,10-biphenyl]-4-yl)-cyclopropanecarboxylic acid (CHF5074), a
novel gamma-secretase modulator, reduces brain beta-amyloid pathol-
ogy in a transgenic mouse model of Alzheimer’s disease without causing
peripheral toxicity. J. Pharmacol. Exp. Ther. 2007, 323, 822–830.
(10) Stock, N.; Munoz, B.; Wrigley, J. D.; Shearman, M. S.; Beher,
D.; Peachey, J.; Williamson, T. L.; Bain, G.; Chen, W.; Jiang, X.; St-
Jacques, R.; Prasit, P. The geminal dimethyl analogue of Flurbiprofen as
a novel Abeta42 inhibitor and potential Alzheimer’s disease modifying
agent. Bioorg. Med. Chem. Lett. 2006, 16, 2219–2223.
7.53ꢀ7.50 (m, 2H), 7.43ꢀ7.32 (m, 4H), 7.14ꢀ7.09 (m, 2H), 4.37 (t, 2H,
J = 6.28 Hz), 4.11 (t, 2H, J = 6.15 Hz), 3.73 (q, 1H, J = 7.15 Hz), 1.86ꢀ1.73
(m, 4H), 1.52 (d, 3H, J = 7.17 Hz). 13C NMR (CDCl3, 100 MHz): 173.9,
161.0, 158.5, 141.9, 135.5, 132.18, 130.9, 129.0, 128.5, 127.83, 123.6, 115.4,
115.1, 72.6, 64.1, 45.1, 25.0, 23.6, 18.3. ESI-HRMS (m/z):[Mþ Na]þ calcd
for C19H20FNO5, 384.1218; observed, 384.1227.
4-[[2-(2-Fluorobiphenyl-4-yl)propanoyl]amino]butyl Nitrate (27).
The bromide 24b (0.24 mmol) was dissolved in CH3CN, and AgNO3
(82.39 mg, 0.48 mmol) was added. The suspension was stirred at rt for
14 h and then at reflux for 2 h. The solid was filtered through Celite pad,
and the solvent was removed under vacuum. The resulting residue was
purified by column chromatography (hexane/ethyl acetate [60:40]) to
1
afford the product as yellow oil (63% yield). H NMR (CDCl3, 400
MHz): δ 7.55ꢀ7.52 (m, 2H), 7.47ꢀ7.34 (m, 4H), 7.15ꢀ7.09 (m, 2H),
4.43 (t, 2H), 3.56 (q, 1H), 3.30ꢀ3.24 (m, 2H), 1.77ꢀ1.53 (m, 7H). 13C
NMR (CDCl3, 100 MHz): 173.85, 161.00, 158.53, 124.85, 135.33,
131.02, 128.90 128.50, 127.90, 127.76, 123.55, 115.32, 115.09, 73.98,
47.58, 40.37, 26.05, 24.13, 18.49. ESI-HRMS (m/z): [M þ H]þ calcd
for C19H21FN2O4, 361.1558; observed, 361.1568.
(11) Peretto, I.; Radaelli, S.; Parini, C.; Zandi, M.; Raveglia, L. F.;
Dondio, G.; Fontanella, L.; Misiano, P.; Bigogno, C.; Rizzi, A.; Riccardi,
B.; Biscaioli, M.; Marchetti, S.; Puccini, P.; Catinella, S.; Rondelli, I.;
Cenacchi, V.; Bolzoni, P. T.; Caruso, P.; Villetti, G.; Facchinetti, F.; Del
Giudice, E.; Moretto, N.; Imbimbo, B. P. Synthesis and biological activity
of flurbiprofen analogues as selective inhibitors of beta-amyloid(1)(ꢀ)-
(42) secretion. J. Med. Chem. 2005, 48, 5705–5720.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: 312-355-5282. Fax: 312 996 7107. E-mail: thatcher@
uic.edu.
(12) Sastre, M.; Dewachter, I.; Landreth, G. E.; Willson, T. M.;
Klockgether, T.; van Leuven, F.; Heneka, M. T. Nonsteroidal anti-inflam-
matory drugs and peroxisome proliferator-activated receptor-gamma ago-
nists modulate immunostimulated processing of amyloid precursor protein
through regulation of beta-secretase. J. Neurosci. 2003, 23, 9796–9804.
(13) Sastre, M.; Dewachter, I.; Rossner, S.; Bogdanovic, N.; Rosen,
E.; Borghgraef, P.; Evert, B. O.; Dumitrescu-Ozimek, L.; Thal, D. R.;
Landreth, G.; Walter, J.; Klockgether, T.; van Leuven, F.; Heneka, M. T.
Nonsteroidal anti-inflammatory drugs repress beta-secretase gene pro-
moter activity by the activation of PPARgamma. Proc. Natl. Acad. Sci. U.
S.A 2006, 103, 443–448.
(14) Zhou, Y.; Su, Y.; Li, B.; Liu, F.; Ryder, J. W.; Wu, X.; Gonzalez-
DeWhitt, P. A.; Gelfanova, V.; Hale, J. E.; May, P. C.; Paul, S. M.; Ni, B.
Nonsteroidal anti-inflammatory drugs can lower amyloidogenic Abe-
ta42 by inhibiting Rho. Science 2003, 302, 1215–1217.
(15) Miller, S. R.; Sekijima, Y.; Kelly, J. W. Native state stabilization
by NSAIDs inhibits transthyretin amyloidogenesis from the most
common familial disease variants. Lab. Invest. 2004, 84, 545–552.
(16) Abdul-Hay, S. O.; Edirisinghe, P.; Thatcher, G. R. Selective
modulation of amyloid-beta peptide degradation by flurbiprofen, fenofi-
brate, and related compounds regulates Abeta levels. J. Neurochem. 2009,
111, 683–695.
’ ACKNOWLEDGMENT
Prof. Larry Marnett of Vanderbilt University is thanked for the
indomethacin and sulindac libraries. This work was funded by
NIH grant AG027425.
’ ABBREVIATIONS USED
SALA, selective amyloid lowering agent; AD, Alzheimer’s disease;
APP, amyloid precursor protein; Aβ, β-amyloid peptide; NSAID,
nonsteroidal anti-inflammatory drug; LPS, lipopolysaccharide; iNOS,
inducible NO synthase; ISMN, isosorbide mononitrate; SARA,
selective amyloid raising agent; CM, conditioned media; DMEM,
Dulbecco’S Modified Eagle’S Medium; PBS, phosphate
buffered saline; FBS, fetal bovine serum; EDTA, ethylenediaminete-
traacetate; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide; NFSI, N-fluorobenzenesulfonimide; TBSCl, tert-butyl-
dimethylsilyl chloride; LiHMDS, lithium bis(trimethylsilyl)-
amide.
2305
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306

Journal of Medicinal Chemistry
ARTICLE
(17) De Strooper, B. Proteases and proteolysis in Alzheimer disease:
a multifactorial view on the disease process. Physiol. Rev. 2010, 90,
465–494.
(18) Chow, V. W.; Mattson, M. P.; Wong, P. C.; Gleichmann, M. An
overview of APP processing enzymes and products. Neuromol. Med.
2010, 12, 1–12.
(19) Rich, J. B.; Rasmusson, D. X.; Folstein, M. F.; Carson, K. A.;
Kawas, C.; Brandt, J. Nonsteroidal anti-inflammatory drugs in Alzhei-
mer’s disease. Neurology 1995, 45, 51–55.
(36) Imbimbo, B.; Hutter-Paier, B.; Villetti, G.; Facchinetti, F.;
Cenacchi, V.; Volta, R.; Lanzillotta, A.; Pizzi, M.; Windisch, M.
CHF5074, a novel gamma-secretase modulator, attenuates brain beta-
amyloid pathology and learning deficit in a mouse model of Alzheimer’s
disease. Br. J. Pharmacol. 2009, 156 (6), 982–993.
(37) Idris, A. I.; Ralston, S. H.; van’t Hof, R. J. The nitrosylated
flurbiprofen derivative HCT1026 inhibits cytokine-induced signalling
through a novel mechanism of action. Eur. J. Pharmacol. 2009,
602, 215–222.
(20) Stewart, W. F.; Kawas, C.; Corrada, M.; Metter, E. J. Risk of
Alzheimer’s disease and duration of NSAID use. Neurology 1997,
48, 626–632.
(21) in t’ Veld, B. A.; Ruitenberg, A.; Hofman, A.; Launer, L. J.; van
Duijn, C. M.; Stijnen, T.; Breteler, M. M.; Stricker, B. H. Nonsteroidal
antiinflammatory drugs and the risk of Alzheimer’s disease. N. Engl. J.
Med. 2001, 345, 1515–1521.
(22) Wilcock, G. K.; Black, S. E.; Hendrix, S. B.; Zavitz, K. H.; Swabb,
E. A.; Laughlin, M. A. Efficacy and safety of tarenflurbil in mild to
moderate Alzheimer’s disease: a randomised phase II trial. Lancet Neurol.
2008, 7, 483–493.
(23) Marder, K. Tarenflurbil in patients with mild Alzheimer’s
disease. Curr. Neurol. Neurosci. Rep. 2010, 10, 336–337.
(24) Imbimbo, B. P.; Giardino, L.; Sivilia, S.; Giuliani, A.; Gusciglio, M.;
Pietrini, V.; Del Giudice, E.; D’Arrigo, A.; Leon, A.; Villetti, G.; Calza, L.
CHF5074, a novel gamma-secretase modulator, restores hippocampal
neurogenesis potential and reverses contextual memory deficit in a
transgenic mouse model of Alzheimer’s disease. J. Alzheimer's Dis. 2010,
20, 159–173.
(38) Wallace, J. L.; Muscara, M. N.; de Nucci, G.; Zamuner, S.;
Cirino, G.; del Soldato, P.; Ongini, E. Gastric tolerability and prolonged
prostaglandin inhibition in the brain with a nitric oxide-releasing
flurbiprofen derivative, NCX-2216 [3-[4-(2-fluoro-alpha-methyl-[1,10-
biphenyl]-4-acetyloxy)-3-methoxyphenyl]-2-propenoic acid 4-nitrooxy
butyl ester]. J. Pharmacol. Exp. Ther. 2004, 309, 626–633.
(39) Govoni, M.; Casagrande, S.; Maucci, R.; Chiroli, V.; Tocchetti,
P. In vitro metabolism of (nitrooxy)butyl ester nitric oxide-releasing
compounds: comparison with glyceryl trinitrate. J. Pharmacol. Exp. Ther.
2006, 317, 752–761.
(40) Prosperi, C.; Scali, C.; Pepeu, G.; Casamenti, F. NO-flurbipro-
fen attenuates excitotoxin-induced brain inflammation, and releases
nitric oxide in the brain. Jpn J. Pharmacol. 2001, 86, 230–235.
(41) Kukar, T.; Murphy, M. P.; Eriksen, J. L.; Sagi, S. A.; Weggen, S.;
Smith, T. E.; Ladd, T.; Khan, M. A.; Kache, R.; Beard, J.; Dodson, M.;
Merit, S.; Ozols, V. V.; Anastasiadis, P. Z.; Das, P.; Fauq, A.; Koo, E. H.;
Golde, T. E. Diverse compounds mimic Alzheimer disease-causing
mutations by augmenting Abeta42 production. Nature Med. 2005,
11, 545–550.
(25) MacNaughton, W. K.; Cirino, G.; Wallace, J. L. Endothelium-
derived relaxing factor (nitric oxide) has protective actions in the
stomach. Life Sci. 1989, 45, 1869–1876.
(26) Chiroli, V.; Benedini, F.; Ongini, E.; Del Soldato, P. Nitric
oxide-donating non-steroidal anti-inflammatory drugs: the case of
nitroderivatives of aspirin. Eur. J. Med. Chem. 2003, 38, 441–446.
(27) van Groen, T.; Kadish, I. Transgenic AD model mice, effects of
potential anti-AD treatments on inflammation and pathology. Brain Res.
Brain Res. Rev. 2005, 48, 370–378.
(28) Abdul-Hay, S. O.; Luo, J.; Ashghodom, R. T.; Thatcher, G. R.
NO-flurbiprofen reduces amyloid-beta, is neuroprotective in cell culture,
and enhances cognition in response to cholinergic blockade. J. Neuro-
chem. 2009, 111, 766–776.
(42) Kukar, T. L.; Ladd, T. B.; Bann, M. A.; Fraering, P. C.; Narlawar,
R.; Maharvi, G. M.; Healy, B.; Chapman, R.; Welzel, A. T.; Price, R. W.;
Moore, B.; Rangachari, V.; Cusack, B.; Eriksen, J.; Jansen-West, K.;
Verbeeck, C.; Yager, D.; Eckman, C.; Ye, W.; Sagi, S.; Cottrell, B. A.;
Torpey, J.; Rosenberry, T. L.; Fauq, A.; Wolfe, M. S.; Schmidt, B.; Walsh,
D. M.; Koo, E. H.; Golde, T. E. Substrate-targeting gamma-secretase
modulators. Nature 2008, 453, 925–929.
(43) Page, R. M.; Baumann, K.; Tomioka, M.; Perez-Revuelta, B. I.;
Fukumori, A.; Jacobsen, H.; Flohr, A.; Luebbers, T.; Ozmen, L.; Steiner,
H.; Haass, C. Generation of Abeta38 and Abeta42 is independently and
differentially affected by familial Alzheimer disease-associated presenilin
mutations and gamma-secretase modulation. J. Biol. Chem. 2008,
283, 677–683.
(29) Prosperi, C.; Scali, C.; Barba, M.; Bellucci, A.; Giovannini,
M. G.; Pepeu, G.; Casamenti, F. Comparison between flurbiprofen and
its nitric oxide-releasing derivatives HCT-1026 and NCX-2216 on
Abeta(1ꢀ42)-induced brain inflammation and neuronal damage in
the rat. Int. J. Immunopathol. Pharmacol. 2004, 17, 317–330.
(30) Imbimbo, B. P.; Del Giudice, E.; Cenacchi, V.; Volta, R.; Villetti,
G.; Facchinetti, F.; Riccardi, B.; Puccini, P.; Moretto, N.; Grassi, F.;
Ottonello, S.; Leon, A. In vitro and in vivo profiling of CHF5022 and
CHF5074 Two beta-amyloid1ꢀ42 lowering agents. Pharmacol. Res.
2007, 55, 318–328.
(31) Hagos, G. K.; Abdul-Hay, S. O.; Sohn, J.; Edirisinghe, P. D.;
Chandrasena, R. E.; Wang, Z.; Li, Q.; Thatcher, G. R. J. Anti-inflammatory,
antiproliferative, and cytoprotective activity of NO chimera nitrates of use in
cancer chemoprevention. Mol. Pharmacol. 2008, 74, 1381–1391.
(32) Hinz, B.; Brune, K.; Rau, T.; Pahl, A. Flurbiprofen enantiomers
inhibit inducible nitric oxide synthase expression in RAW 264.7 macro-
phages. Pharm. Res. 2001, 18, 151–156.
(44) Takahashi, Y.; Hayashi, I.; Tominari, Y.; Rikimaru, K.; Mor-
ohashi, Y.; Kan, T.; Natsugari, H.; Fukuyama, T.; Tomita, T.; Iwatsubo,
T. Sulindac sulfide is a noncompetitive gamma-secretase inhibitor that
preferentially reduces Abeta 42 generation. J. Biol. Chem. 2003,
278, 18664–18670.
(45) Peretto, I.; Radaelli, S.; Parini, C.; Zandi, M.; Raveglia, L. F.;
Dondio, G.; Fontanella, L.; Misiano, P.; Bigogno, C.; Rizzi, A.; Riccardi,
B.; Biscaioli, M.; Marchetti, S.; Puccini, P.; Catinella, S.; Rondelli, I.;
Cenacchi, V.; Bolzoni, P. T.; Caruso, P.; Villetti, G.; Facchinetti, F.;
DelGiudice, E.; Moretto, N.; Imbimbo, B. P. Synthesis and biological
activity of flurbiprofen analogues as selective inhibitors of beta-
amyloid1ꢀ42 secretion. J. Med. Chem. 2005, 48, 5705–5720.
(46) Kotoris, C. C.; Chen, M.-J.; Taylor, S. D. Preparation of
^
^
benzylic I(,I(-difluoronitriles, -tetrazoles, and -sulfonates via electro-
philic fluorination. J. Org. Chem. 1998, 63, 8052–8057.
(47) Zhang, F.; Song, J. Z. A novel general method for preparation of
[alpha]-fluoro-[alpha]-arylcarboxylic acid. Direct fluorination of silyl ketene
acetals with Selectfluor. Tetrahedron Lett. 2006, 47, 7641–7644.
(33) Tegeder, I.; Niederberger, E.; Israr, E.; Guhring, H.; Brune, K.;
Euchenhofer, C.; Grosch, S.; Geisslinger, G. Inhibition of NF-
kappaB and AP-1 activation by R- and S-flurbiprofen. FASEB J 2001,
15, 2–4.
(34) Thatcher, G. R. J.; Nicolescu, A. C.; Bennett, B. M.; Toader, V.
Nitrates and NO release: contemporary aspects in biological and
medicinal chemistry. Free Radical Biol. Med. 2004, 37, 1122–1143.
(35) Gasparini, L.; Rusconi, L.; Xu, H.; del Soldato, P.; Ongini, E.
Modulation of beta-amyloid metabolism by non-steroidal anti-inflamma-
tory drugs in neuronal cell cultures. J. Neurochem. 2004, 88, 337–348.
2306
dx.doi.org/10.1021/jm101450p |J. Med. Chem. 2011, 54, 2293–2306