Fragment-based lead discovery of a novel class of small molecule antagonists of neuropeptide B/W receptor subtype 1 (GPR7)

Article abstract of DOI:10.1016/j.bmcl.2020.127510

Here, we report the discovery of a new class of NPBWR1 antagonists identified from a fragment-based screen. Compound 1 (cAMP IC₅₀ = 250 μM; LE = 0.29) emerged as an initial hit. Further optimization of 1 by SAR-by-catalogue and chemical modification produced 21a (cAMP IC₅₀ = 30 nM; LE = 0.39) with a 6700-fold increase in potency from fragment 1. Somewhat surprisingly, Schild analysis of compound 21a suggested that in vitro inhibition of NPW-mediated effects on upon cAMP accumulation were saturable, and that compound 21a dose-dependently increased [125I]-hNPW23 dissociation rate constants from NPBWR1 in kinetic binding studies. Collectively, these data are inconsistent with a classic surmountable, orthosteric mechanism of inhibition. The benzimidazole inhibitors reported herein may therefore represent a mechanistically differentiated class of compounds with which to form a better appreciation of the pharmacology and physiological roles of this central neuropeptide system.

Full text of DOI:10.1016/j.bmcl.2020.127510

Bioorganic&MedicinalChemistryLetters30(2020)127510
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Fragment-based lead discovery of a novel class of small molecule antagonists
of neuropeptide B/W receptor subtype 1 (GPR7)
⁎
Remond Moningkaa, , F. Anthony Romeroa, Nicholas B. Hastingsb, Zhiqiang Guoa, Ming Wanga,
Jerry Di Salvoc, Ying Lic, Dorina Truscac, Qiaoling Dengd, Vincent Tonge, Jenna L. Terebetskif,
Richard G. Ballg, Feroze Ujjainwallaa
a Merck & Co., Inc., Department of Medicinal Chemistry, PO Box 2000, Rahway, NJ 07065, USA
b Merck & Co., Inc., Department of Neuroscience and Ophthalmology, Kenilworth, NJ 07033, USA
c Merck & Co., Inc., Department of In Vitro Pharmacology, PO Box 2000, Rahway, NJ 07065, USA
d Merck & Co., Inc., Department of Chemistry Modeling and Informatics, PO Box 2000, Rahway, NJ 07065, USA
e Merck & Co., Inc., Department of Pharmacokinetics, Pharmacodynamics, and Drug Metabolism, PO Box 2000, Rahway, NJ 07065, USA
f Merck & Co., Inc., Department of Basic Pharmaceutical Sciences, PO Box 2000, Rahway, NJ 07065, USA
g Merck & Co., Inc., Department of Analytical Chemistry, PO Box 2000, Rahway, NJ 07065, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
Here, we report the discovery of a new class of NPBWR1 antagonists identified from a fragment-based screen.
Compound 1 (cAMP IC50 = 250 µM; LE = 0.29) emerged as an initial hit. Further optimization of 1 by SAR-by-
catalogue and chemical modification produced 21a (cAMP IC50 = 30 nM; LE = 0.39) with a 6700-fold increase
in potency from fragment 1. Somewhat surprisingly, Schild analysis of compound 21a suggested that in vitro
inhibition of NPW-mediated effects on upon cAMP accumulation were saturable, and that compound 21a dose-
dependently increased [125I]-hNPW23 dissociation rate constants from NPBWR1 in kinetic binding studies.
Collectively, these data are inconsistent with a classic surmountable, orthosteric mechanism of inhibition. The
benzimidazole inhibitors reported herein may therefore represent a mechanistically differentiated class of
compounds with which to form a better appreciation of the pharmacology and physiological roles of this central
neuropeptide system.
NPBWR1
GPR7
Anti-obesity
G protein-coupled receptor
Schild plot analysis
Introduction
acute hyperphagia in male rats, respectively.1b,d Furthermore, NPBWR1
The G protein-coupled receptor 7 (GPR7) was deorphanized with
the identification of endogeneous ligands neuropeptide B (NPB) and
neuropeptide W (NPW)1 and was reclassified as the neuropeptide B/W
receptor-1 (NPBWR1). NPBWR1 is present in peripheral tissues, but is
primarily expressed within the central nervous system (CNS) and
widely distributed in the hippocampus, subcortical limbic tele-
ncephalon, hypothalamus, olfactory cortex, midbrain periaqueductal
gray and tegmental area.2
knockout mice developed an adult-onset obese phenotype that pro-
gressively worsened with age and was significantly exacerbated when
animals were fed a high-fat diet.4 Encouraged by these studies we
sought to probe NPBWR1 as a potential novel anti-obesity target.
We have previously reported the development of a picomolar small
molecule antagonist of NPBWR1 from a high-throughput screening hit
in order to develop tools to study the physiological function of
NPBWR1.5 A distinct chemical series has also been developed from a
group at The Scripps Research Institute culminating in CYM50769.6
While these compounds represent potential tool compounds to inter-
rogate NPBWR1, we had continued interest in developing novel che-
mical matter. Here we describe the fragment-based design and struc-
Several studies have suggested that NPBWR1 has a role in regulating
feeding behavior, energy homeostatis, neuroendocrine function, and
modulating inflammatory pain.3 Intracerebroventricular (i.c.v.) injec-
tion of NPW and NPB was shown to increase food intake and cause
⁎ Corresponding author.
E-mail address: remond_moningka@merck.com (R. Moningka).
https://doi.org/10.1016/j.bmcl.2020.127510
Received 2 June 2020; Received in revised form 14 August 2020; Accepted 17 August 2020
Availableonline06September2020
0960-894X/©2020ElsevierLtd.Allrightsreserved.

R. Moningka, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127510
Table 2
Exploration of the para position at the right-hand aryl group.
Fig. 1. In vitro potency of fragment hit 1, SAR-by-catalogue hit 2 and analog 3
at NPBWR1.
Scheme 1. Reagents and conditions: (a) o, m, p-substituted benzaldehyde,
ammonium acetate, 75 °C.
Table 1
Effects of ortho, meta and para substituents on potency and LE.
30K member hybrid fragment library, the informer library, which
consisted of molecules that were > 200 MW and < 350 MW. By
biasing the deck with larger fragments we hoped to increase of the
possibility of a hit with < 250 μM potency in a high concentration
screen.8 Since G-protein coupled receptors still remain a challenge for
biophysical fragment screening methods, we proposed that our hybrid-
fragment library might prove productive in identifying novel chemical
matter for GPCR targets.
Our work initiated with screening our hybrid fragment library at
100 µM in a cAMP assay, which led to the identification of 500 hits with
inhibition of > 30% inhibition. A counter-screen of these 500 hits
against a non-related GPCR (GPR74) resulted in 20 unique compounds.
The compounds were purified and titrated in quadruplicate against
NPBWR1 giving three compounds that showed dose-dependent inhibi-
tion. One of these hits was methylsulfone benzimidazole 1 (Fig. 1).
Further follow-up of 1 by use of SAR-by-catalogue resulted in com-
pound 2 with a 25-fold improvement in potency. Initial SAR of 2
quickly identified compound 3c with a further 7-fold increase in po-
tency. For the purpose of this article ligand efficiency (LE) index9 and
potency were used as guidelines for SAR optimization. The potency
ture–activity relationship (SAR) studies of a negative allosteric mod-
ulator of NPBWR1.
Fragment-based drug discovery is now a routine technology to find
novel chemical matter for drug discovery projects.7 While an x-ray
structure of the fragment bound to the protein often selects for fragment
follow-up and provides a path for hypothesis-driven chemistry to de-
velop the fragment hit, it is not necessarily required for fragment-based
lead discovery. Biophysical methods also exist to drive fragment follow-
up, such as surface plasmon resonance (SPR) and NMR. We developed a
2

R. Moningka, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127510
Table 3
Table 4
Exploration of the disubstituted and trisubstituted right-hand aryl ring.
Exploration of heteroaromatic groups at the para position.
described will be at the human receptor unless otherwise indicated.
Compound 3c was chosen as a starting point to further optimize as it
retained excellent LE (0.42). The systematic initial SAR focused on
exploring the effect of a monosubstituted ortho, meta, and para right-
hand aryl moiety by condensation of nitro phenylenediamine 7 with the
appropriate benzaldehydes to afford corresponding benzimidazoles
(Scheme 1, Table 1). Regardless of the electronic effect of the selected
functional groups and without exception, para substituents (3c, 4c, 5c,
6c) were the most potent.
Based on these data, we further investigated the SAR at the mono-
substituted para position to improve potency of 3c (Table 2). Extension
of the alkyl chain from a methyl group of compound 3c to a longer alkyl
chain (8a-b) increased the potency by two-fold. Propyl compound 8b
was the most potent. The potency of 8b increased 2-fold over 3c,
however this was at the cost of a decrease in LE. Substitution to bulkier
t-butyl (8c) and phenyl (8d) groups further decreased the functional
activity. Replacement of the methyl group of compound 3c with a tri-
fluoromethyl (8e) also proved detrimental to potency. Next, we
Fig. 2. In vitro potency of analog 10 at human NPBWR1.
Scheme 2. Reagents and conditions: (a) 1,4-benzo-
quinone, p-bromobenzaldehyde, 75 0C, 2 h; (b) Pd
(PPh3)4, 1.0 M K2CO3, R-B(OH)2, 110 0C, 15 min.
3

R. Moningka, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127510
Scheme 3. Reagents and conditions: (a) SnCl2·H2O,
reflux, 2 h; (b) NaNO2, KI, TsOH, 10 0C to rt, 1 h; (c)
NBS, benzoyl peroxide, reflux, overnight; (d) i. KOAc,
rt, overnight, ii. NaOH, reflux, 1 h; (e) Dess-Martin
Periodinane, rt, 2 h; (f) 4-trifluorobenzene-1,2-dia-
mine, 1,4 benzoquinone, 75 0C, overnight; (g) Pd
(PPh3)4, R-B(OH)2, 1.0 M K2CO3, 110 0C, 15 min.
the ortho position (9e, 9g) provided an additional small enhancement
(1.5-fold) to the potency of 4c. However, the substitution at the meta
position (9f, 9h) proved detrimental to potency, suggesting that 2,4-
disubstitution on the aryl group was superior compared to its coun-
terpart. Interestingly, introduction of another substituent at the 6 po-
sition, trisubstituted 9i, resulted in a significant improvement in both
potency (~13-fold) and LE relative to 4c (0.44 vs 0.40).
Table 5
Trisubstituted aryl group with narrowly focused p-heteroaromatics for an ad-
ditive effect in potency.
Concurrent with exploring the SAR on the nitro containing benzi-
midazole series, we aimed to replace the nitro moiety due to its po-
tential inherent metabolic liabilities in vivo. Early in our work we
discovered that trifluromethyl containing benzimidazole 10 had a two-
fold increase in potency over 3c (Fig. 2).
We further studied the scope of the SAR by examining heteroaro-
matic groups at the para substitution of the phenyl group of compound
10. As illustrated in Scheme 2, trifluoromethyl phenylenediamine 11
was converted to trifluoromethyl benzimidazole 12 by condensation
with p-bromobenzaldehyde. Compound 12 served as an ideal key in-
termediate to quickly access analogs via Suzuki cross-coupling.
Table 4 shows various heteroaromatics that were examined. Pyr-
azole containing compound 13a was 3-fold more potent than the parent
compound 10. Additionally, other heteroaromatic containing com-
pounds (13b-e) were well tolerated. However, triazole analog 13f was
4-fold less potent than 10. We then further functionalized the pyrazole
moiety (13g-h). Both compounds resulted in potency loss.
investigated substitution with polar substituents baring hydrogen bond
donors and acceptors (5c, 8f-i). All resulted in a reduction in potency
and LE.
To expand the SAR further, our efforts were directed towards ex-
ploring 2,4- (ortho, para), 3,4- (meta, para) disubstitution and subse-
quently 2,4,6-trisubstitution10 on the distal aryl ring while maintaining
methyl and chloro groups at the para position. Table 3 summarizes
some of these combinations. Analogs with small substitution such as
methyl (9a-d) were comparable to 3c and 4c, and there was no dis-
tinction in potency difference whether the methyl moiety was at the
ortho or meta position. The presence of a chlorine and bromine atom at
With the discovery of productive SAR on the right-hand side, we
investigated whether the potency gains from trisubstituted and para
heteroaromatic analogs would be additive. Aldehyde 19 was quickly
synthesized in a linear fashion (Scheme 3).11 Aryl nitro 14 was reduced
with tin (II) chloride to give aniline 15, which underwent Sandmeyer
chemistry to furnish aryl iodide 16. Bromination of the methyl group
4

R. Moningka, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127510
with NBS and benzoyl peroxide afforded benzyl bromide 17, which was
then followed by displacement of the bromide with potassium acetate
and a subsequent treatment with sodium hydroxide to give benzyl al-
cohol 18. Oxidation of 18 with Dess-Martin led to the formation of
benzaldehyde 19, which was then condensed with trifluoromethyl
phenylenediamine 11 to yield benzimidazole 20. Suzuki coupling of
iodide with tetrakis(triphenylphosphine)palladium furnished final
compound 21.
A
As shown in Table 5, the hybrid compounds (21a-c) had a sig-
nificant improvement in potency, with IC50 values in the nanomolar
range. Pyrazole containing compound 21a resulted in the most potent
analog. Indeed, by comparing 13a and 21a, analog 21a exhibited both
potency (~10 fold) and LE (0.37 vs. 0.39) gains.
To obtain a more rigorous estimate of antagonist potency, we at-
tempted to characterize the pKB of benzimidazole representative com-
pound 21a by Schild analysis. As demonstrated in Fig. 3A, and con-
sistent with inhibition of agonist activity in our cAMP assay, titration of
compound 21a resulted in a parallel rightward shift of equipotent
agonist doses. Unexpectedly however, the dextral displacement of
agonist dose response curves elicited by compound 21a appeared to be
saturable and, indeed, the Schild replot of the data (Fig. 3B) is curvi-
linear. Moreover, compound 21a appeared to dose-dependently inhibit
forskolin-stimulated cAMP accumulation in cells expressing GPR7 in its
own right, both in the context of cross-titration and when in-
dependently tested in the agonist mode of the assay (red trace, Fig. 3A).
That is, compound 21a appears to act as a partial agonist of GPR7. To
further investigate the mechanism of inhibition of compound 21a, we
turned to a homogeneous, real-time SPA binding assay to examine the
off-rate kinetics of 125I-hNPW23 both in the presence and absence of
compound 21a (Fig. 3C). Toward this end, membranes were prepared
from CHO cells recombinantly overexpressing GPR7, and pre-equili-
brated with a saturating concentration of 125I-hNPW23. Dissociation of
the pre-bound radioiodinated agonist was then initiated by dilution into
a large volume of binding buffer containing 0–30 μM compound 21a,
and revealed that the koff of 125I-hNPW23 is dose dependently ac-
celerated by 21a. Taken collectively, these data are inconsistent with
neutral, strictly competitive antagonism of NPW23 at GPR7 and, rather,
suggest that compound 21a may act as an antagonist of GRP7, at least
in part, by allosterically reducing the affinity of hNPW23 for its cognate
receptor. The Schild replot for such data is linear.
B
In summary, a fragment-based screen was utilized to find a novel
benzimidazole series as GPR7 antagonists. Potency and LE index were
used as guidelines to optimize the series. As indicated in Scheme 4, we
have demonstrated the ability to advance a weak binder and low LE of
methylsulfone benzimidazole 1. The first success came from the dis-
covery of nitro benzimidazole 2 and eventually trifluoromethyl benzi-
midazole 22, which enhanced the potency to a low micromolar range.
Subsequent optimization at the para position with pyrazole followed by
an additive effect at the 2,6-position produced notable compound 21a
with a 6,700-fold increase in potency and a significant gain in LE.
Additionally to have a deeper understanding of the molecular interac-
tion between the benzimidazole class and the GPR7 receptor, we con-
ducted Schild plot analysis and off-rate kinetics experiments using
compound 21a to represent the benzimidazole series. Based on Schild
analysis, the graph diverged from linearity. As the concentration of the
compound increased, the graph would eventually plateau, indicating
that the series did not exhibit competitive antagonism. Furthermore by
examining the off-rate kinetics study, it showed that the series ac-
celerated the off-rate between the indigenous ligand (NPW) and the
GPR7 receptor as the concentration of 21a increased.
C
Fig. 3. Using compound 21a as a representative in the benzimidazole class, A.
Agonist titration curves ([A] = concentration of NPW, [B] = concentration of
21a), B. Schild plot analysis, C. Off-rate kinetics experiment.
5

R. Moningka, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127510
Scheme 4. From hit 1 to lead 21a.
Declaration of Competing Interest
+H]+. To a solution of tosic acid (3.66 g, 19.26 mmol) in acetonitrile (25.7 mL) was
added 3,5-dichloro-4-methylaniline 15 (1.13 g, 6.42 mmol). The resulting suspension
of amine salt was cooled to 10-15 °C, and to this was added a solution of sodium
nitrite (0.886 g, 12.84 mmol) and potassium iodide (2.66 g, 16.05 mmol) in water (4.
0 mL). The reaction mixture was stirred for 10 min, then allowed to come to rt and
stirred for 1 h. Water was then added followed by 1.0 M aqueous NaHCO3 and
Na2S2O3. The mixture was extracted with DCM (3x), and the combined organic
layers were dried over Na2SO4, filtered, and concentrated under vacuum. The crude
residue was purified by silica gel chromatography, eluting with 0-10% EtOAc/hex-
anes to give 2,6-dichloro-4-iodotoluene 16 (1.23 g, 67%) as a pale brown solid. 1H
NMR (d6-DMSO, 500 MHz): δ 7.77 (s, 2H), 2.21 (s, 3H). To a solution of 1,3-di-
chloro-4-iodotoluene 16 (1.23 g, 4.29 mmol) in CCl4 (14.29 mL) was added NBS (1.
907 g, 10.72 mmol) followed by benzoyl peroxide (0.052 g, 0.214 mmol). The re-
action mixture was heated at reflux overnight, then cooled to rt, and filtered through
a pad of celite. The solvent was removed under vacuum, and the residue was purified
by silica gel chromatogarphy, eluting with 0-5% EtOAc/hexanes to afford 5-iodo-2-
(bromomethyl)-1,3-dichlorobenzene 17 (1.54 g, 98%) as an orange crystal. 1H NMR
(CDCl3, 500 MHz): δ 7.67 (s, 2H), 4.67 (s, 2H). To a solution of 5-iodo-2-(bromo-
methyl)-1,3-dichlorobenzene 17 (700 mg, 1.914 mmol) in acetonitrile (9.50 mL) was
added potassium acetate (751 mg, 7.65 mmol). The reaction was stirred at rt over-
night. The solid was then filtered, and the filtrate was evaporated under vacuum. The
resulting residue was dissolved in EtOH (10 mL), and treated with 5.0 M aqueous
sodium hydroxide (765 µL, 3.83 mmol). The reaction mixture was heated at reflux
for 1 h, then cooled to rt, and extracted with EtOAc (3x). The combined extracts were
dried over Na2SO4, filtered, and concentrated under vacuum to yield crude (2,6-
dichloro-4-iodophenyl)methanol 18 (547 mg, 94%) as an orange solid. (2,6-Dichloro-
4-iodophenyl)methanol 18 (547 mg, 1.806 mmol) was dissolved in dichloromethane
(18.0 mL), and Dess-Martin Periodinane (1149 mg, 2.71 mmol) was added in one
portion. The reaction mixture was stirred at rt for 2 h before the solvent was eva-
porated under vacuum. The crude residue was purified by silica gel chromatography,
eluting with 0-15% EtOAc/hexanes to afford 2,6-dichloro-4-iodobenzaldehyde 19
(443 mg, 82%) as a pale-orange solid. 1H NMR (CDCl3, 500 MHz): δ 10.4 (s, 1H), 7.
76 (s, 2H). Trifluoromethyl phenylenediamine 11 (235 mg, 1.334 mmol), 1,4-ben-
zoquinone (159 mg, 1.468 mmol), and 2,6-dichloro-4-iodobenzaldehyde 19 (442 mg,
1.468 mmol) were dissolved in Ethanol (8894 µL), and the reaction mixture was
heated to 75 °C overnight. The reaction mixture was then allowed to cool to rt, and
the solvent was evaporated under vacuum. The crude mixture was purified by silica
gel chromatography, eluting with 0-25% EtOAc/hexanes to afford 2-(2,6-dichloro-4-
iodophenyl)-5-(trifluoromethyl)-1H-benzimidazole 20 (449 mg, 74%) as a brownish
gray solid. LCMS: m/z 457 [M+H]+. 2-(2,6-Dichloro-4-iodophenyl)-5-(tri-
fluoromethyl)-1H-benzimidazole 20 (30 mg, 0.066 mmol), palladium tetrakis (7.59
mg, 6.56 µmol), 1.0 M aqueous potassium carbonate (197 µL, 0.197 mmol), and 4-
pyrazoleboronic acid pinacol ester (5.88 mg, 0.131 mmol) were mixed in dioxane
(656 µL). The resulting mixture was heated to 110 °C for 15 min under microwave
irradiation. The solvent was then evaporated under vacuum, and the crude mixture
was purified by RP HPLC to give 2-[2,6-dichloro-4-(1H-pyrazol-4-yl)phenyl]-5-(tri-
fluoromethyl)-1H-benzimidazole 21a (9.10 mg, 34%). 1H NMR (500 MHz, Methanol-
d4) δ 8.22 (s, 1H), 8.06 (s, 1H), 7.93 (s, 1H), 7.89 (d, J = 8.5 Hz, 1H), 7.71 (dd, J =
8.6, 1.4 Hz, 1H), 7.70 – 7.64 (m, 1H), 7.61 – 7.56 (m, 1H). LCMS: m/z 397 [M
+H]+.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2020.127510.
References
1. (a) Fujii R, Yoshida H, Fukusumi S, et al. J. Biol. Chem. 2002;277:34010
(b) Shimomura Y, Harada M, Goto M, et al. J. Biol. Chem. 2002;277:35826
(c) Brezillon S, Lannoy V, Franssen JD, et al. J. Biol. Chem. 2003;278:776
(d) Tanaka H, Yoshida T, Miyamoto N, et al. Proc. Natl. Acad. Sci. U.S.A.
2003;100:6251.
2. Jackson VR, Lin SH, Wang Z, Nothacker HP, Civelli O. J. Comp. Neurol.
2006;497:367.
3. Hondo M, Ishii M, Sakurai T. Results Probl. Cell. Differ. 2008;46:239.
4. Ishii M, Fei H, Friedman JM. Proc. Natl. Acad. Sci. U.S.A. 2003;100:10540.
5. Romero FA, Hastings NB, Moningka R, et al. Bioorg. Med. Chem. Lett. 2012;22:1014.
6. (a) Urbano M, Guerrero M, Zhao J, et al. Bioorg. Med. Chem. Lett. 2012;22:7135
(b) Guerrero M, Urbano M, Schaeffer M-T, Brown S, Rosen H, Roberts E. Bioorg. Med.
Chem. Lett. 2013;23:614.
7. Rees DC, Congreve M, Murray CW, Carr R. Nat. Rev. Drug Discov. 2004;3:660.
8. Lan P, Romero FA, Wodka D, et al. J. Med. Chem. 2017;60:9040.
9. For ligand efficiency index concept and calculation: (a) Kuntz, I. D.; Chen, K.; Sharp,
K. A.; Kollman, P. A. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 9997; (b) Hopkins, A. L.;
Groom, C. R.; Alex, A. Drug Discovery Today. 2004, 9, 430; (c) Abadzapatero, C.;
Metz, J. Drug Discovery Today. 2005, 10, 464; (d) Leeson, P. D.; Springthorpe, B.
Nat. Rev. Drug Discov. 2007, 6, 881; (e) Keseru, G. M.; Makara, G. M. Nat. Rev. Drug
Discov. 2009, 8, 203.
10. Numbering system on the right-hand aryl group was depicted in figure 1 com-
pound 1.
11. Representative experimental procedure (compound 21a): A mixture of 1,3-dichloro-
2-methyl-5-nitrobenzene (3.00 g, 14.56 mmol) 14 and tin(II) chloride dihydrate (13.
14 g, 58.2 mmol) in ethanol (48.5 mL) was heated under reflux for 2 h. The reaction
mixture was allowed to cool to rt, and quenched carefully with water and sat. aqu-
eous solution of NaHCO3 until slightly basic. The thick slurry mixture was filtered
through a pad of celite, and the filtrate was extracted with EtOAc (3x). The combined
extracts were dried over Na2SO4, filtered, and concentrated under vacuum. The
crude residue was purified by silica gel chromatography, eluting with 0-20% EtOAc/
hexanes to give 3,5-dichloro-4-methylaniline 15 (1.13 g, 44%). LCMS: m/z 176 [M
6