Challenges in the development of mGluR5 positive allosteric modulators: The discovery of CPPHA

Article abstract of DOI:10.1016/j.bmcl.2006.11.081

This Letter describes, for the first time, the synthesis and SAR, developed through an iterative analog library approach, that led to the discovery of the positive allosteric modulator (PAM) of the metabotropic glutamate receptor mGluR5 CPPHA. Binding to

Full text of DOI:10.1016/j.bmcl.2006.11.081

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1386–1391
Challenges in the development of mGluR5 positive allosteric
modulators: The discovery of CPPHA
Zhijian Zhao,a,* David D. Wisnoski,a Julie A. O’Brien,b Wei Lemaire,b
David L. Williams, Jr.,b Marlene A. Jacobson,b Marion Wittman,b
Sookhee N. Ha,b Herve Schaffhauser,b Cyrille Sur,b Doug J. Pettibone,b
Mark E. Duggan,a P. Jeffrey Conn,b,c George D. Hartmana
and Craig W. Lindsleya,c
aDepartment of Medicinal Chemistry, Merck & Co., Inc., PO Box 4, West Point, PA 19486, USA
bDepartment of Neuroscience, Merck & Co., Inc., PO Box 4, West Point, PA 19486, USA
cVICB Program in Drug Discovery, Department of Pharmacology, Vanderbilt Medical Center, Nashville, TN 37232, USA
Received 4 October 2006; revised 21 November 2006; accepted 30 November 2006
Available online 3 December 2006
Abstract—This Letter describes, for the first time, the synthesis and SAR, developed through an iterative analog library approach,
that led to the discovery of the positive allosteric modulator (PAM) of the metabotropic glutamate receptor mGluR5 CPPHA. Bind-
ing to a unique allosteric binding site distinct from other mGluR5 PAMs, CPPHA has been the focus of numerous pharmacology
studies by several laboratories.
ꢀ 2006 Elsevier Ltd. All rights reserved.
N
F
Our laboratory recently disclosed the structures and
pharmacology of two series of mGluR5 positive alloste-
ric modulators (PAMs), represented by DFB (1) and
CPPHA (2), and one account of the synthesis, SAR,
and in vivo efficacy of the first centrally active mGluR5
PAM, CDPPB (3) (Fig. 1).1–4 Data acquired with all
three series of mGluR5 PAMs, by indirect activation
of NMDA receptors, strongly support the NMDA
hypofunction hypothesis of schizophrenia.5 Since these
initial reports, CPPHA has proven to be an important
pharmacological tool for both our laboratory and a
number of other pharmaceutical and academic laborato-
ries.2,6,7 Of particular note is the fact that DFB and
CDPPB bind at the same allosteric binding site as the
mGluR5 negative allosteric modulator, MPEP; in con-
trast, CPPHA does not bind at the MPEP site and
therefore interacts with an as yet unknown and distinct
allosteric binding site.1–4 The novel allosteric binding
site of CPPHA has engendered unique pharmacology
to CPPHA (vide infra) as well as significant challenges
in the lead optimization program that led to the discov-
F
N
1, DFB
Cl
O
H
O
N
N
N
CN
N
HN
OH
O
O
2, CPPHA
3, CDPPB
Figure 1. mGluR5 positive allosteric modulators (PAMS).
ery of CPPHA. This Letter will detail the screening
leads and subsequent synthesis and SAR that ultimately
led to the discovery of CPPHA, as well as an overview
of the pharmacological profile of CPPHA.
We screened a portion of our sample collection searching
for compounds that selectively potentiated the mGluR5
response to 300 nM glutamate (an EC20 concentration
Keywords: MGluR5; Allosteric modulator; Potentiation.
*
Corresponding author. E-mail: zhijian_zhao@merck.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.081

Z. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1386–1391
1387
R1
R1
R1
Br
Cl
a
b
O
O
O
O
O
NH2
HN
Br
HN
N
HN
HN
R2
R2
N
6
7
8
O
O
4
5
R1
R1
Figure 2. mGluR5 PAM screening leads.
R3
O
O
N
c
d
O
R4
O
8
N
HN
HN
of glutamate).1–4 Hits from this screen were then titrated
in a FLIPR assay at a single, low concentration of ago-
nist to evaluate potency (EC50) and fold potentiation.8
This led to the identification of 4 (human mGluR5
EC50 = 750 nM, 7.6-fold potentiation) and 5 (human
mGluR5 EC50 = 2.7 lM, 5-fold potentiation) as selec-
tive mGluR5 PAMs (Fig. 2). Moreover, 4 alone (up to
100 lM) caused no response by human mGluR5 in this
assay and both 4 and 5 were selective of mGluR5 (selec-
tive vs mGluR1–4, 7, and 8).2 Note that fold-potentia-
tion is established at a single agonist concentration to
provide a convenient and rapid metric to compare com-
pounds, and all values for fold potentiation presented in
this Letter were determined in this way. More detailed
studies were performed on selected compounds of inter-
est. In this case, agonist concentration–response curves
were determined in the presence of several fixed
concentrations of test compound. Figure 3 shows the
concentration dependent potentiation of the response
of CHO cells expressing mGluR5 to glutamate, which
is manifested as increased sensitivity to agonist.
R2
R3
R2
O
10
9
Scheme 1. Synthesis of PAMs 9 and 10. Reagents and conditions: (a)
i—R2COCl, PS–DIEA, DCM, rt, 90–95%, ii—PS–trisamine, 70–95%;
(b) NBS, Bz2O2, CCl4, 90 ꢁC, 45–69%; (c) functionalized phthalimide/
succinimide, K2CO3, DMF, 50 ꢁC, 70–85%; (d) i—R3NH2, PS–DIEA,
DCM, rt, 80–90%, ii—R4COCl, PS–DIEA, DCM, rt, 90–95%. All
compounds purified by mass-directed HPLC.9
good, with all final products being purified by mass-di-
rected preparative HPLC.9 We first explored substitu-
tions on the central phenyl ring (Table 1).
Replacement of the bromine atom with hydrogen, as
in 11a, led to an ꢀ19-fold loss in potency; in contrast,
replacement with a chlorine atom, as in 11c, afforded a
250 nM mGluR5 PAM with 6.8-fold potentiation. After
surveying 48 diverse substituents, we found that only
4-position halogen and OMe were tolerated; all other
substituents were inactive as well as substitutions at all
other positions on the central phenyl ring. Interestingly,
11e, a 6-F analog, was devoid of activity at mGluR5,
but displayed PAM of mGluR7 (EC50 = 7 lM).
Since 4 was modular, an iterative analog library ap-
proach was employed to rapidly develop SAR following
the synthetic route shown in Scheme 1. In the event,
commercially available functionalized 2-methyl anilines
6 were acylated with acid chlorides to deliver amides
7. Benzylic bromination provided 8 which could be
treated with various phthalimides or succinimides to
provide analogs represented by general structure 9, or
8 could be treated with a secondary amine followed by
acylation to provide analogs 10. Yields were generally
Having identified the optimal substituents for the central
phenyl ring, the next rounds of library synthesis focused
on incorporating functionalized succinimide and phthal-
imide moieties, as the phthalimide group engendered
poor physical properties and aqueous solubility to the
series.
Table 1. EC50 and fold potentiation of analogs 11
140
R
human mGluR5
120
O
100
O
EC50, nM
80
N
HN
10 μM 4 87
N
60
1 μM 4 220
O
40
0.1 μM 4 514
11
mGluR5 EC50a(lM) Fold potentiation
Vehicle 667
20
0
Compound
R
4
4-Br
H
0.75
13.9
1.5
7.6
3.4
2.6
6.8
3.5
0.9
-9
-8
-7
-6
-5
-4
-3
11a
11b
11c
11d
11e
log [glutamate] (M)
4-F
4-CI
0.25
Figure 3. Compound 4 potentiation of response to glutamate is
manifested as increased mGluR5 sensitivity. The glutamate EC50 value
is shifted from 667 nM to 87 nM with the addition of 4. Similar data
acquired with quisqualate and 3,5-DHPG.2
4-OMe 1.4
6-F >100
a Human mGluR5.

1388
Z. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1386–1391
Table 3. EC50 and fold potentiation of analogs 14
Cl
Cl, Br
O
O
O
R = alkly, Ph, OH, Het
O
R
R
N
HN
N
HN
N
O
O
12
14
Compound
R
mGluR5
EC50a(lM)
Fold
potentiation
Figure 4. Succinimide analogs 12.
4
0.25
35
6.8
1.8
1.3
0.6
0.6
3.5
1.6
0.6
7.1
11.3
2.9
4.1
6.2
9.2
All attempts to replace the phthalimide moiety with
functionalized succinimides (Fig. 4) were unproductive,
leading to inactive molecules (mGluR5 EC50 > 100 lM).
Thus far, SAR for this series was ‘flat’, with slight struc-
tural changes leading to a complete loss of activity.
However, substitution was tolerated on the phthalimide
moiety (Table 2), with a general trend toward the 4-Cl
analogs being more efficacious. Despite affording PAMs
of comparable activity to 11c, none of the analogs 13
displayed improvements in physical properties.
N
14a
14b
14c
14d
14e
14f
14g
2
N
N
2.1
N
N
>100
>100
5.7
N
Based on the data generated thus far, subsequent efforts
held the unsubstituted phthalimide and the 4-Cl func-
tionality constant, and explored a diverse array of amide
analogs 14 (Table 3). From several libraries, totaling
over 500 analogs, only ꢀ10% retained mGluR5 PAM
activity. Again, on this PAM program and other alloste-
ric modulator programs, SAR tends to be very flat
which makes series development difficult.3,4,10 However,
this effort did produce several compounds worthy of fur-
ther study; notably, 4, 2 (also known as CPPHA), and
14h. None of these analogs potentiated rat mGluR5
with EC50s of <1 lM; therefore, they were not viable
for in vivo proof of concept experiments in rat behavior-
al models, but could serve as tools for in vitro
experiments.2
CF3
MeO
1.5
>100
0.25
0.26
5.7
OMe
OH
N
14h
14i
14j
14k
OH
Table 2. EC50 and fold potentiation of analogs 13
X
NO2
O
O
1.5
R
N
HN
N
O
O
2.3
13
Compound
R
X
mGluR5
EC50a(lM)
Fold
potentiation
S
N
14l
1.9
13a
13b
13c
13d
13e
13f
5-F
5-F
Br
Cl
Br
Cl
Br
Cl
Br
Cl
Br
Cl
Br
Cl
0.93
0.33
0.17
0.61
0.72
0.28
>100
>100
>100
2.2
11.1
4.6
4.5
4.8
10.7
4.5
0.6
0.6
0.6
3.7
7.2
8.9
a Human mGluR5.
3-F
3-F
While 4, 2, and 14h were being screened in ancillary as-
says, synthetic efforts focused on identifying an alterna-
tive for the phthalimide group. According to Scheme 1,
benzyl bromide 8 was treated with various primary
amines and subsequently acylated to deliver tertiary
amide analogs 10 (Fig. 5). This effort produced 150 ana-
logs; however, the majority of these analogs were inac-
tive or had EC50s > 5 lM with <5-fold potentiation
for human mGluR5 and EC50s > 1 lM for rat mGluR5.
4-OEt
4-OEt
4,5-diOMe
4,5-diOMe
4,5-diMe
4,5-diMe
3,5-diCI
3,5-diCI
2.1
0.89
a Human mGluR5.

Z. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1386–1391
1389
Cl
Cl
Cl
Cl
Cl
Cl
Ph
O
N
O
N
R4
O
O
O
O
O
I
HN
HN
HN
HN
R3
N
20
Ph
N
N
N
O
Pd(OAc)2
150oC, uw
10
Figure 5. Tertiary amide analogs 10 and 15.
15
21
19
Scheme 3. Synthesis of unsaturated analog 21. Yields range from 79%
to 88%; Compounds purified by mass-directed HPLC.9
The most potent compound from this effort was 15
(EC50 = 470 nM, 5.9-fold potentiation for human
mGluR5); unfortunately, 15 offered no advantage over
phthalimide analogs in terms of physical properties or
efficacy for rat mGluR5.
dard Heck conditions with unsaturated ketone 20; sur-
prisingly, the desired saturated product 21 resulted
instead of the expected unsaturated Heck product 17.
Unfortunately, 21 was only a weak mGluR5 PAM with
an EC50 of 4.5 lM (Scheme 3).
Concurrently, synthetic effort was applied to the other
screening lead 5, with the aim of merging the two series
into a hybrid molecule, recognizing that the unsaturated
phenyl ketone moiety could be a phthalimide equivalent.
A number of analogs were prepared from intermediate 8
by conversion to the phosphonium salt, and a subse-
quent microwave-assisted Wittig reaction to deliver ana-
logs 16 (Scheme 2). Of these analogs, the unsubstituted
phenyl analog 17 proved to be the most active with a hu-
man mGluR5 EC50 of 1.7 lM and 8-fold potentiation
(Fig. 6). Unexpectedly, as DMSO stock solutions of
17, and related analogs of general structure 16, were
stored at room temperature and then assayed again,
the EC50 diminished to >13 lM. NMR analysis con-
firmed that 17 quantitatively underwent a hetero-inverse
electron demand Diels–Alder reaction to generate the
highly functionalized pyran 18. To avoid this unwanted
reaction, the unsaturated analog of 17 was prepared and
evaluated. In the event, iodide 19 was subjected to stan-
After evaluation in CYP assays, a panel of 120 enzymes,
ion channels, and transporters, and physiochemical prop-
erty measures, 2 (CPPHA) proved to be the best mGluR5
PAM in the series for further proof of concept studies. As
we reported previously, CPPHA was selective for
mGluR5 (no PAM of mGluRs 1, 2, 3, 4, 7 or 8) and the
only detected ancillary pharmacology was weak negative
allosteric modulation of mGluR4 and 8.2 In radioligand
binding assays, CPPHA did not affect binding of the
orthosteric ligand 3[H]quisqualate, nor did CPPHA affect
3
negative allosteric ligand [H]PEPy.2 This was in sharp
contrast to DFB and CDPPB which compete for
3[H]PEPy binding, and therefore share the same allosteric
binding site as MPEP.1–4 It is interesting to note that
CDPPB displayed more tractable SAR than CPPHA
which could be the result of the novel allosteric binding
site utilized by CPPHA. In total, 985 analogs of 4 and 5
were prepared and evaluated; only 45 analogs (4.5%) pos-
sessed mGluR5 EC50s below 5 lM and no tractable SAR
could ever be established in these series, highlighting the
power and utility of library synthesis to identify CPPHA.
Cl
1) PPh3, CH3CN, rt
O
8
Despite the medicinal chemistry challenges, CPPHA
proved to be an important small molecule tool for in vi-
tro experiments. In electrophysiological studies of brain
slice preparations, CPPHA (10 lM) potentiated
NMDA currents in hippocampal slices induced by sub-
threshold levels of DHPG and also potentiated
mGluR5-mediated, DHPG-induced depolarization of
subthalmic nucleous neurons. In the absence of DHPG,
CPPHA had no effect.2
HN
O
2)
R
N
R
H
O
O
150oC, uw
16
Scheme 2. Synthesis of analogs 16. Yields range from 22% to 86%; All
compounds purified by mass-directed HPLC.9.
Cl
After our initial publication, Conn and Liu disclosed the
effects of CPPHA on signaling pathways in brain slices.6,7
Importantly, Conn’s study demonstrated that CPPHA
has qualitatively different effects on mGluR5-mediated
calcium responses and ERK1/2 phosphorylation in rat
cortical astrocytes, whereas, DFB, an mGluR5 PAM that
binds at the MPEP allosteric site, had the same effects on
both signaling pathways.6 Liu reported that CPPHA
potentiated (R,S)-3,5-dihydroxyphenylglycol (DHPG)-
induced ERK and CREB phosphorylation in rat
cortical and hippocampal slices, as well as potentiated
N
Cl
Ph
O
HN
O
HN
O
NH
N
O
O
N
O
Ph
Cl
17
18
Figure 6. Structure of 17 and cycloaddition product 18.

1390
Z. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1386–1391
Cl
O
Cl
Cl
O
O
O
HN
O
NBS/cat. Bz2O2
23
CCl4, 90 oC, 4 hrs
PS-DIEA/DCM
rt, 1 h
NH2
O
22
24
O
Cl
O
Cl
NH
26
O
O
O
O
N
H
N
HN
HO
DMF/K2CO3/ca. KI
50 oC overnight
Br
O
O
then LCMS purification
2
25
Scheme 4. Synthesis of CPPHA 2.11
3. Lindsley, C. W.; Wisnoski, D. D.; Duggan, M. E.; Leister,
W. H.; Huff, J. R.; Williams, D. L., Jr.; O’Brien, J. A.;
Lemaire, W.; Sur, C.; Pettibone, D.; Kinney, G.; Burno,
M. J. Med. Chem. 2004, 47, 5825.
4. Kinney, G. G.; O’Brien, J. A.; Lemaire, W.; Burno, M.;
Bickel, D. J.; Clements, M. K.; Wisnoski, D. D.; Lindsley,
C. W.; Tiller, P. R.; Smith, S.; Jacobson, M. A.; Sur, C.;
Duggan, M. E.; Pettibone, D. J.; Williams, D. W., Jr.
J. Pharmacol. Exp. Therapeut. 2005, 313, 199.
5. (a) Williams, D. L., Jr.; Lindsley, C. W. Curr. Topics Med.
Chem. 2005, 5, 825; (b) Lindsley, C. W.; Shipe, W. D.;
Wolkenberg, S. E.; Theberge, C. R.; Williams, D. L., Jr.;
Sur, C.; Kinney, G. G. Curr. Topics Med. Chem. 2006, 8,
771, and references therein; (c) Lindsley, C. W.; Wolken-
berg, S. E.; Shipe, W.; Williams, D. L., Jr. Curr. Opin.
Drug Disc. Dev 2005, 8, 449.
DHPG-induced phosphorylation of the NR1 (Ser 897)
subunit of NMDA receptors.7
In summary, we have disclosed the synthesis and SAR
of the mGluR5 positive allosteric modulator CPPHA,
2, with a unique allosteric binding site, distinct from that
of MPEP, DFB, and CDPPB. SAR was flat and devel-
oping SAR was virtually impossible through traditional
medicinal chemistry strategies. Even with an iterative
analog library synthesis approach, only 4.5% of 985 ana-
logs based on screening leads 4 and 5 possessed mGluR5
PAM activity below 5 lM. In general PAMs have
proved far more challenging to develop robust, tractable
SAR for standard GPCR agonists/antagonists. Despite
these challenges, CPPHA has become an important tool
to probe the pharmacology of mGluR5, and demon-
strated that different PAMs can differentially modulate
coupling of a single receptor to different signaling
pathways.
6. Zhang, Y.; Rodriguez, A. L.; Conn, P. J. J. Pharmacol.
Exp. Therapeut. 2005, 315, 1212.
7. Liu, F.; Zhang, G.; Hornby, G.; Vasylyev, D.; Bowlby,
M.; Park, K.; Gilbert, A.; Marquis, K.; Andree, T. H. Eur.
J. Pharmacol. 2006, 536, 262.
8. EC50 of potentiation and fold potentiation were deter-
mined from concentration–response curves of test com-
pound at a single, low (EC20) concentration of agonist.
Fluo-4 (calcium sensitive dye) was loaded into CHO cells
expressing human or rat mGluR5. These cells were then
pretreated with a range of concentrations of test com-
pound, and finally challenged with an approximate EC20
concentration of agonist (in this case, 300 nM glutamate).
The inflection point of the concentration–response curve
of the test compound was taken as the EC50 for poten-
tiation, and the ratio of the maximum agonist response (at
the agonist EC20) in the presence of test compound versus
the response to the same concentration of agonist in the
absence of test compound is taken as the fold potentiation.
Ref. 2 provides additional experimental details and a curve
(Fig. 2) showing how EC50 and fold potentiation are
determined. The referenced figure provides an illustration
of the variability for each data point. Variability of EC50
for potentiation (ranging from 9.4% to 32.8%), is detailed
in Table of Ref. 2. Note that Figure 3 in the present
paper represents the effect of several fixed concentrations
of test compound on the agonist concentration–response
curve.
Acknowledgments
The authors wish to thank Dr. Sandor Varga for exten-
sive NMR work to confirm the structural assignments of
the analogs represented by general structure 9, 11, 13,
and 14, and Dr. Charles W. Ross III for obtaining
HRMS data.
References and notes
1. O’Brien, J. A.; Lemaire, W.; Chen, T.-B.; Chang, R. L. S.;
Jacobson, M. A.; Ha, S.; Lindsley, C. W.; Sur, C.;
Pettibone, D. J.; Conn, P. J.; Williams, D. L., Jr. Mol.
Pharmacol. 2003, 64, 731.
2. O’Brien, J. A.; Lemaire, W.; Wittmann, M.; Jacobson, M.
A.; Ha, S. N.; Wisnoski, D. D.; Lindsley, C. W.;
Schaffhauser, H. J.; Sur, C.; Duggan, M. E.; Pettibone,
D. J.; Conn, J.; Williams, D. L. J. Pharmacol. Exp.
Therapeut. 2004, 309, 568.

Z. Zhao et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1386–1391
1391
9. Leister, W. H.; Strauss, K. A.; Wisnoski, D. D.; Zhao, Z.;
Lindsley, C. W. J. Comb. Chem. 2003, 5, 322.
10. Lindsley, C. W.; Zhao, Z.; Leister, W. H.; Robinson, R.
G.; Barnett, S. F.; Defeo-Jones, D.; Jones, R. E.;
Hartman, G. D.; Huff, J. R.; Huber, H. E.; Duggan, M.
E. Bioorg. Med. Chem. Lett. 2005, 15, 761.
along with a light source, until completion (4 h) by LCMS.
After cooling down to room temperature, the mixture
solution was filtered through a small plug of silica gel,
yielding 25 as a yellow solid (2.29 g, 74%). Analytical LC/
MS: (CH3CN/H2O/0.1% TFA, 4 min gradient), 89% pure,
m/e 384 (M+1).
11. The synthesis of general structure represented by 9, 11, 13,
and 14 is exemplified by the synthesis of CPPHA (2) as
shown in Scheme 4.
CPPHA (2): a mixture of 25 (120 mg, 0.31 mmol), phthal-
imide 26 (67 mg, 0.45 mmol), K2CO3 (207 mg, 1.5 mmol),
and KI (17 mg, 0.01 mmol) in 1.2 mL DMF was allowed to
stir at 50 ꢁC overnight.14 Next morning, the reaction
mixture was filtered and the collected DMF solution of
crude product was purified by LC/MS9 to afford the pure
product CPPHA (2) as a slightly yellow solid (66.5 mg,
53%). 1H NMR (600 MHz, CDCl3): d (ppm) 4.83, (s, 2 H),
7.02–7.07 (m, 2H), 7.36 (dd, J = 2.5 Hz, 8.8 Hz, 1H), 7.50
(ddd, J = 1.4 Hz, 8.0 Hz, 8.0 Hz, 1H), 7.59 (d, J = 2.5 Hz,
1H), 7.73 (d, J = 8.8 Hz, 1H), 7.76–7.79 (m, 2H), 7.90–7.93
(m, 2H), 8.18 (dd, J = 1.4 Hz, 8.0 Hz, 1H), 10.17 (s, 1H),
12.26 (s, 1H); Analytical LC/MS: (CH3CN/H2O/0.1%
TFA, 4 min gradient), single peak (214 nm), m/e 407
2-{[(4-chloro-2-methylphenyl)amino]carbonyl}phenyl ace-
tate (24): to a mixture solution of N,N-(diisopropyl)ami-
nomethylpolystyrene (PS-DIEA, loading 3.6 mmol/g)
(7.1 g, 26 mmol)12 and 4-chloro-2-methylaniline 22
(1.42 g, 10 mmol) was slowly added acetylsalicyloyl chlo-
ride 23 (1.99 g, 10 mmol) with shaking. The resulting
mixture was shaken until the reaction completion (1 h) by
LCMS. The resin was filtered and washed with DCM (2·).
The collected DCM solution was dried under vacuum to
afford 24 as a slightly yellow solid (2.64 g, 87%). Analytical
LC/MS: (CH3CN/H2O/0.1% TFA, 4 min gradient), 93%
pure, m/e 304 (M+1).
2-({[(2-Bromomethyl)-4-chlorophenyl]amino}carbonyl)
phenyl acetate (25): compound 24 (2.64 g, 8.7 mmol)
obtained from the fist step was immediately taken up in a
solution of CCl4 with fresh-recrystallized NBS (1.62 g,
9.1 mmol) and benzoyl peroxide (Bz2O2)13 (218 mg,
0.9 mmol). The reaction mixture was heated at 90 ꢁC,
(M+1)
at
3.633 min;
HRMS:
calculated
for
[C22H15N2O4Cl + H]+, 407.0799; found407.0793.
12. Booth, R. J.; Hodges, J. C. J. Am. Chem. Soc. 1997, 119,
4882.
13. Sorrell, T. N.; Pigge, F. C. J. Org. Chem. 1993, 58, 784.
14. Bookser, B. C.; Bruice, T. C. J. Am. Chem. Soc. 1991, 113,
4208.