Fatty acid amide hydrolase inhibitors. 2. Novel synthesis of sterically hindered azabenzhydryl ethers and an improved synthesis of VER-156084

Article abstract of DOI:10.1016/j.tetlet.2010.07.136

We report an improved synthesis of the fatty acid amide hydrolase (FAAH) inhibitor VER-156084. The key step is a novel, environmentally benign etherification to form an unusual, highly hindered azabenzhydryl ether. The method is applied to a variety of pr

Full text of DOI:10.1016/j.tetlet.2010.07.136

Tetrahedron Letters 51 (2010) 5191–5194
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Fatty acid amide hydrolase inhibitors. 2. Novel synthesis of sterically hindered
azabenzhydryl ethers and an improved synthesis of VER-156084
*
Stephen D. Roughley , Terance Hart
Vernalis (R&D) Ltd, Granta Park, Cambridge CB21 6GB, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
We report an improved synthesis of the fatty acid amide hydrolase (FAAH) inhibitor VER-156084. The key
step is a novel, environmentally benign etherification to form an unusual, highly hindered azabenzhydryl
ether. The method is applied to a variety of primary and secondary alcohols.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 2 June 2010
Revised 14 July 2010
Accepted 23 July 2010
Available online 30 July 2010
Keywords:
Etherification
Fatty acid amide hydrolase (FAAH) inhibitor
Azabenzhydryl ether
Azetidine ether
Hindered ether
Green chemistry
We have recently disclosed the discovery of VER-156084 (7), a
potent and selective inhibitor of fatty acid amide hydrolase (FAAH).1
Our initial synthesis of the compound was according to that previ-
ously described for our CB1 antagonist programme (Scheme 1),2
however, the incorporation of the benzhydryl ether portion of the
molecule resulted in poor yields in the ether formation step. As part
ofourongoingstudiesintoFAAHinhibitorswerequiredan improved
route to azabenzhydryl ethers to provide analogues of VER-156084
and sufficient quantities of VER-156084 to support in vivo studies.
In this Letter, we describe the results of our investigations in this
area, leading to a new method for the formation of hindered aza-
benzhydryl ethers, which we have demonstrated to be of broad syn-
thetic utility.
Our original route to VER-156084 involved a seven-step synthe-
sis, in 5% overall yield (Scheme 1).1 Whilst the three-step route to the
azabenzhydrol intermediate 4 worked reasonably well (74% over
three steps), a shorter route would offer operational advantages.
Consequently, we investigated the addition of 4-chlorophenylmag-
nesium bromide to the commercially available 2-chloronicotinonit-
rile (8).3,4 In our hands, none of the expected ketone 3 was obtained
in this reaction (Scheme 2). Instead, we considered the direct forma-
tion of 4 by reaction of 2-chloronicotinaldehyde (9)4 with the Grig-
nard reagent. Surprisingly, we were unable to find any literature
reports of Grignard reagent additions to 9, although the analogous
reaction of 2-chloroquinoline-3-carboxaldehyde (THF, rt 10 min,
85–95%) had been reported.5 Reaction of 9 with 4-chlorophenyl-
magnesium bromide under these conditions gave 4 as a crystalline
solid in 87% yield. This route proved readily amenable to the synthe-
sis of 30 g quantities of 4.
N
Cl
Cl
N
Cl
N
Cl
R
O
b
c
OH
O
1 R=Cl
2 R=N(OMe)Me
3
4
a
Cl
d
N
Cl
N
Cl
O
O
f, g
N
N
N
R
O
Cl
Cl
VER-156084
5 R=CHPh2
6 R=H•HCl
e
7
Scheme 1. Synthesis of VER-156084.1 Reagents and conditions: (a) MeNHOMeÁHCl,
83%; (b) 4-Cl-C6H4MgBr, 100%; (c) NaBH4, 89%; (d) 1-Ph2CH-azetidin-3-ol, PTSA,
PhMe 35%; (e) 1-chloroethyl chloroformate then MeOH, 85%; (f) chiral HPLC; (g)
triphosgene, piperidine, 50%.
* Corresponding author. Tel.: +44 (0) 1223 895555; fax: +44 (0) 1223 895556.
E-mail address: s.roughley@vernalis.com (S.D. Roughley).
Present address: Geistlich Pharma AG, Bahnhofstrasse 40, CH-6110 Wolhusen,
Switzerland.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.07.136

5192
S. D. Roughley, T. Hart / Tetrahedron Letters 51 (2010) 5191–5194
a
X
R
b
Cl
Cl
Cl
N
N
Cl
Cl
N
3
4
b
a
N
Cl
OMs
OMe
(from 8)
(from 9)
OH
8 R = CN
9 R = CHO
Quant. by
LC-MS
11
4
10
Scheme 2. Reagents and conditions: (a) 4-Cl-C6H4MgBr, THF, rt, 4 d, 0%; (b) 4-Cl-
C6H4MgBr, THF, rt, 10 min; 87%.
Cl
Cl
d
N
97%
N
e
With improved access to 4, we turned our attention to the trou-
blesome ether formation step. Whilst the synthesis of benzhydryl
ethers is well documented,6 there are very few reports on the syn-
thesis of azabenzhydryl ethers with hindered alcohols.7,8 We
therefore carried out a screen of conditions for the synthesis of
the azabenzhydryl ether 13, using N-benzylpiperidin-4-ol 15 as
the nucleophilic alcohol component, as it is readily available com-
mercially,4 and the resulting 1H NMR spectrum of the product
would have fewer overlapping aromatic resonances than the benz-
hydryl-protected azetidine.
Our initial investigations centred on introduction of a leaving
group to one component and generation of the alkoxide of the
other component (Scheme 3). Thus, mesylation of 4 (MsCl, Et3N,
CH2Cl2 or THF) proceeded cleanly to give the highly activated mes-
ylate 10, which was used immediately in the following reactions.
Addition of mesylate 10 to the pre-formed alkoxide of 15 failed
to give any of the desired ether (Table 1). This result is surprising,
as 10 reacts rapidly with DMSO at ambient temperature in a
Swern-like oxidation9 process to give the ketone 14, and with
MeOH under similar conditions to give the methyl ether 11. In-
stead, heating the alkoxide of 15 with mesylate 10 in THF–DMF re-
sulted in mesylate transfer to give 16 with t1/2 ꢀ3 h. Reversal of the
coupling partners by synthesis of the mesylate 16 of the piperidi-
nol component and reaction with the alkoxide of 4 also failed to re-
sult in any ether formation, in this case 16 underwent elimination
to 17 on prolonged heating, even on addition of catalytic NaI.
We next turned our attention to chloride as a leaving group.
Thus, chlorination of 4 (SOCl2, CH2Cl2, cat. DMF) proceeded cleanly,
the resulting chloride 12 being considerably more stable than the
corresponding mesylate 10, however, reaction with the alkoxide
of 15 again failed to give any of the desired product under a variety
of conditions (Table 1).
c
X
O
Cl
Cl
N
14
Quant.
Cl
O
c
Cl
N
Ph
by LC-MS
c
t
13
12
1/2 ~3 h by LC-MS
Cl
N
Cl
c
X
HO
a
MsO
c
Ph
N
17
Ph
89%
N
16
Ph
Quant. by
LC-MS
15
Scheme 3. Reagents and conditions: (a) MsCl, Et3N, CH2Cl2 or THF; (b) MeOH, rt; (c)
See Table 1; (d) DMSO, rt; (e) SOCl2, CH2Cl2, cat. DMF.
Table 1
Nucleophilic displacement conditionsa
Electrophile Base
Solvent
CH2Cl2
T (°C) Commentb
10
10
10
10
10
10
16
16
16c
12
12c
12c
—
rt
rt
NR
K2CO3 CH2Cl2
Trace product by LC–MS
NaH
NaH
—
THF/DMFd rt
THF/DMFd 75
NR
Mesylate transfer t1/2 ꢀ3 h
DMSOe
MeOHe
DMF
rt
rt
rt
75
75
125
125
125
Oxidation to 14
—
Conversion to methyl ether 11
NaH
NaH
NaH
—
NR
DMF
DMF
PhMe
PhMe
Elimination of OMs to give 17
Elimination of OMs to give 17
NR
NR
NR
—
K2CO3 PhMe
Following the failure of our attempts to form the ether by intro-
duction of a leaving group on one of the coupling partners, we re-
turned to the investigation of acid- or Lewis acid-catalysed
conditions, in order to see if we could improve the poor-yielding
conditions previously reported (Scheme 1; PTSA, PhMe, Dean–
Stark, 36%).1,2 We believe that the poor yields in this reaction com-
pared to the corresponding benzhydryl system2,8 to be most likely
due to the competing protonation of the pyridine group under
acidic conditions, disfavouring generation of the intermediate car-
bocation (Fig. 1).
The results of our condition screen are shown in Table 2. Lewis
acids in CH2Cl2 gave no reaction.10 Three equivalents of PTSA were
required, but in CH2Cl2 the reaction was slow. Higher temperatures
gave an increased conversion as determined by LC–MS, but a num-
ber of unidentified by-products began to form. Addition of 4 Å
molecular sieves hindered the reaction. Changing the solvent to
PhMe gave a slight improvement, but the reaction mixture under-
went significant decomposition at 200 °C. MeCN, 1,4-dioxane or
water were detrimental to the reaction. Changing the catalyst to
the milder PPTS in PhMe gave no reaction at 120 °C, but surpris-
ingly gave 60% conversion into the highly hindered symmetrical
ether 18 at 200 °C.11 TFA or AcOH gave no reaction or acylation
of 4 at higher temperatures. However, reaction at 125 °C in neat
PTSA gave clean conversion into the required ether product 13 in
90% isolated yield after 3 h.12 Using cH2SO4 resulted in extremely
rapid and exothermic decomposition at rt.
a
All reactions carried out with 1.25 equiv of nucleophilic partner with LC–MS
monitoring for 3 d.
b
NR—no reaction.
Catalytic NaI added.
2.5 equiv of nucleophile.
Solvent is the nucleophile—reaction complete within 5 min.
c
d
e
With the optimal conditions identified, we next investigated the
scope of the reaction (Table 3).12,13 Nitrogen protection was not re-
quired, and Boc groups underwent very rapid cleavage under the
reaction conditions. The reaction appeared to work well with a
range of primary and secondary alcohols, although the reaction
with N-benzhydryl azetidin-3-ol was poor. The reaction of 4 with
commercially available 3-azetidinol hydrochloride in 81% yield al-
H
H
N+ Cl
N+ Cl
N
Cl
+
H+
H+
H+
OH
+
4
Disfavoured
Cl
Cl
Cl
Figure 1. Competing pyridine protonation in the azabenzhydryl ether synthesis
under acid catalysis disfavouring cation formation.

S. D. Roughley, T. Hart / Tetrahedron Letters 51 (2010) 5191–5194
5193
Table 2
Table 3
Acid- and Lewis acid-catalysed conditionsa
Scope of the ether formationa
N
Cl
Ar
N
Cl
Cl
ROH
Ar
N
Cl
Cl
Ar
N
PTSA•H2O
15
O
O
OH
O
120 ºC
R
4
Cl
N
N
18
13
Bn
Ar = 4-chlorophenyl
Cl
Solvent
Catalyst
(equiv)
Tb
(°C)
Commentc
ROH
Time (h)
3
Yield (%)
90
15
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
FeCl3 (0.1)
FeCl3 (0.1)
20
120
NR
NR
NR
NR
NR
NR
HO
5.75
5
100% LC–MS conversion
72 (Boc group lost)b
78 (Boc group lost)b
68b
Sc(OTf)3 (0.1) 20
Sc(OTf)3 (0.1) 120
NH
PTSA (0.1)
PTSA (1.1)
PTSA (3.0)
PTSA (3.0)
120
120
120
140
HO
HO
N
Boc
10% conversion into 13
40% conversion into 13;
by-products forming
NR
10% conversion into 13
20% conversion into 13
5
N
Boc
CH2Cl2
CH2Cl2
PhMe
PhMe
PTSA (3.0)d
PTSA (3.0)d
PTSA (3.0)
PTSA (3.0)
120
140
120
200
120
200
5.75
5.75
NH•HCl
NH•HCl
HO
All 4 consumed, many by-products
79b
1,4-dioxane PTSA (3.0)
1,4-dioxane PTSA (3.0)
Trace 13 formed
HO
HO
40% conversion into 13;
by-products forming
5
5
73b
72b
NH
NH
Ph
N
MeCN
MeCN
H2O
PTSA (3.0)
PTSA (3.0)
PTSA (3.0)
PTSA (3.0)
PPTS (3.0)
PPTS (3.0)
AcOH (3.0)
AcOH (3.0)
—
120
200
120
200
120
200
120
200
120
200
120
140
Trace 13 formed
Decomposition
HO
HO
HO
NR
H2O
Decomposition
PhMe
PhMe
PhMe
PhMe
AcOH
AcOH
CH2Cl2
CH2Cl2
NR
1.75
3.25
Decomp., trace product
81
60% conversion into 18, no 13 formed
Ph
NR
NR
NR
NH
•HCl
a
All reactions performed with 1:1 4: ROH with 3 equiv of PTSA at 125 °C in an
open tube.
—
4-Ac and des-Cl-4
NR
TFA (3.0)
TFA
b
Product isolated as 1:1 mixture of diastereomers.
10% conversion into the
trifluoroacetate of 4
60% conversion into 13
e
—
—
PTSA (3.0)
PTSA (3.0)
—
125
125f 90% isolated yield of 13
e
H2SO4
rt
Decomposition
Acknowledgement
a
All reactions carried out with a 1:1 mixture of 4 and 15 for 20 min unless
otherwise indicated.
The authors thank Heather Simmonite for NMR spectroscopy
(compound 13).
b
Reactions heated to indicated temperature using
a Biotage Synthesizer
microwave.
c
Reactions monitored by LC–MS—NR = no reaction, % conversion based on LC–
References and notes
MS.
d
4 Å molecular sieves added.
Reaction performed as a melt in an open tube.
Reaction run for 3 h.
e
f
1. Hart, T.; Macias, A. T.; Benwell, K.; Brooks, T.; D’Alessandro, J.; Dokurno, P.;
Francis, G.; Gibbons, B.; Haymes, T.; Kennett, G.; Lightowler, S.; Mansell, H.;
Matassova, N.; Misra, A.; Padfield, A.; Parsons, R.; Pratt, R.; Robertson, A.; Walls,
S.; Wong, M.; Roughley, S. Bioorg. Med. Chem. Lett. 2009, 19, 4241.
2. Davidson, J. E. P.; Bentley, J. M.; Dawson, C. E.; Harrison, K.; Mansell, H. L.;
Pither, A. L.; Pratt, R. M.; Roffey, J. R. A.; Ruston, V. J.; WO2004/096794, 2004;
Chem. Abstr. 2004, 141, 410806.
3. (a) Rynbrandt, R. H.; Tiffany, B. D.; Balgoyen, D. P.; Nishizawa, E. E.; Mendoza, A.
R. J. Med. Chem. 1979, 22, 525; (b) Dunn, A. D.; Norrie, R. Z. Chem. 1990, 30, 245.
4. All reagents were purchased from the Sigma–Aldrich Chemical Company
except for (R)- and (S)-1-Boc-3-hydroxymethylpiperidine (CNH Technologies
Inc., Woburn, MA), 3-hydroxyazetidine hydrochloride (Atlantic SciTech Group
Inc., Linden, NJ) and 1-benzhydrylazetidin-3-ol (Fluorochem Ltd, Old Glossop,
Derbyshire, UK). All commercial reagents were used as supplied without
further purification.
lowed an improved synthesis of VER-156084 (7).13 Completion of
the synthesis required urea formation and resolution as described
in our earlier report,1 giving the racemic material in three steps,
with an overall yield of 35%. Scale-up of the ether formation was
best achieved by carrying out the reaction as a thin layer melt in
a Petri dish immersed in a sandbath, under which conditions, the
reaction could be run on at least 7.8 mmol scale.
In conclusion, we have described the development of a novel
ether synthesis, using molten PTSA as the catalyst. We believe this
methodology to be generally applicable to the synthesis of hin-
dered ethers resistant to conventional conditions and have demon-
strated its applicability to a range of hindered nucleophilic alcohols
bearing protected and free amine functionalities. Furthermore, the
solvent-free, environmentally benign non-toxic nature of the re-
agents has significant benefits over alternative methodologies.
We have demonstrated the utility of the methodology in an im-
proved synthesis of the FAAH inhibitor VER-156084 and will dis-
cuss its wider application in the investigation of FAAH inhibitors
elsewhere.
5. (a) Belmont, P.; Andrez, J.-C.; Allan, C. S. M. Tetrahedron Lett. 2004, 45, 2783; (b)
Bhat, B.; Bhaduri, A. P. Synthesis 1984, 673.
6. See: e.g. Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic Synthesis, 3rd
ed.; Wiley-Interscience: New York, 1999.
7. For a direct etherification, see: e.g. (a) De Martino, G.; La Regina, G.; Di Pasquali,
A.; Ragno, R.; Bergamini, A.; Ciaprini, C.; Sinistro, A.; Maga, G.; Crespan, E.;
Artico, M.; Silvestri, R. J. Med. Chem. 2005, 48, 4378; (b) Many alternative
examples are prepared by indirect substrate-specific methods, see: e.g. Boyd,
M.; Gagnon, M.; Lau, C.; Mellon, C.; Scheigetz, J. WO2004/022526, 2003; Chem.
Abstr. 2003, 140, 270877.
8. Zhang, M.-Q.; Wada, Y.; Sato, F.; Timmerman, H. J. Med. Chem. 1995, 38, 2472.
9. (a) Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651; (b) Huang, S. L.; Omura, K.;
Swern, D. Synthesis 1978, 297; (c) Mancuso, A. J.; Huang, S.-L.; Swern, D. J. Org.
Chem. 1978, 43, 2480.

5194
S. D. Roughley, T. Hart / Tetrahedron Letters 51 (2010) 5191–5194
10. Sharma, G. V. M.; Prasad, T. R.; Mahalingam, A. K. Tetrahedron Lett. 2001, 42,
compound 13: dH (400 MHz; MeOH-d4) 8.28 (1H, dd, J = 4.5 and 1.8 Hz), 8.07
(1H, dd, J = 7.8 and 1.8 Hz), 7.40 (1H, dd, J = 7.8 and 4.5 Hz), 7.35 (2H, d,
J = 8.8 Hz), 7.32 (2H, d, J = 8.8 Hz), 7.34–7.22 (5H, m), 5.86 (1H, s), 3.47 (2H, s),
3.46–3.40 (1H, m), 2.74–2.66 (2H, m), 2.21–2.12 (2H, m), 1.92–1.80 (2H, m)
and 1.72–1.62 (2H, m); dC (100 MHz; MeOH-d4) 150.3, 149.7, 140.4, 139.1,
138.4, 138.3, 134.8, 130.8, 130.2, 129.6, 129.3, 128.4, 124.7, 76.7, 74.4, 63.9,
51.7, 32.0 and 31.9; LC–MS >95%; m/z 427 ([M+H]+; 100%); HRMS found
[M+H]+ 427.1330, calcd for C24H25Cl2N2O: 427.1344.
759.
11. Compound 18 had been observed as a trace by-product under the original
Dean–Stark conditions.2
12. Typical experimental procedure:
A
test-tube containing
4
(1.0 equiv), 15
(1.0 equiv) and PTSAÁH2O (3.0 equiv) was immersed in
a
sand-bath pre-
heated to 125 °C. After 3 h, the mixture was cooled, diluted with CH2Cl2
(25 mL) and washed with NaOH (2.0 M in H2O; 25 mL). The CH2Cl2 layer was
dried by passing through a phase separator frit and evaporated to give the
13. In order to investigate further the difference in reactivity of the N-benzhydryl
azetidin-3-ol and unprotected 3-hydroxyazetidine, we attempted the reaction
of 4 with 3-hydroxyazetidine hydrochloride under our original Dean–Stark
conditions, but no product formation was observed.
ether product as
a pale yellow gum. All compounds showed satisfactory
analytical data (LC–MS purity >95%, molecular ion and 1H NMR) and were
identical to previously published data where available. Analytical data for