A practical synthesis of differentially protected 4,4′-dipiperidinyl ethers: Novel ligands of pharmaceutical interest

Article abstract of DOI:10.1055/s-0028-1088154

An efficient four-step synthesis (requiring no purification) of Boc-protected 4,4′-dipiperidinyl ethers is described. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0028-1088154

LETTER
1051
A Practical Synthesis of Differentially Protected 4,4¢-Dipiperidinyl Ethers:
Novel Ligands of Pharmaceutical Interest
D
ifferentia
a
lly Protec
t
m
ed 4,4
¢
-Dipiperidiny
e
l
E
ther
s
s M. Bailey,* Gordon Bruton, Anthony Huxley, Vicky Johnstone, Peter H. Milner, Barry S. Orlek,
Geoffrey Stemp
GlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2NY, UK
Fax +44(1438)768232; E-mail: James.M.Bailey@gsk.com
Received 8 December 2008
Our synthesis of the previously unknown ether 9 began by
coupling hydroxy piperidine 5 with 4-chloropyridine, fol-
lowed by formation of the quaternary alkylammonium
salt 7, under standard conditions (Scheme 1).5 A simple
one-pot reduction of the pyridine (step c) then gave dipi-
peridine ether 8. Subsequent Boc deprotection afforded 9
which allowed further elaboration of the template by func-
tionalisation of the secondary amine.
Abstract: An efficient four-step synthesis (requiring no purifica-
tion) of Boc-protected 4,4¢-dipiperidinyl ethers is described.
Key words: ethers, hydrogenations, piperidines, pyridines, hista-
mine
The design of new ligands for clinical and pharmacologi-
cal profiling of key targets of biological interest, in partic-
ular for disease areas of unmet medical need, is an
important goal for the pharmaceutical industry. As part of
a program directed towards finding novel histamine-sub-
type-selective ligands, we identified the novel 4,4¢-dipi-
peridinyl ether template 1 which showed unexpected
potency at the histamine H3 receptor.1
OH
(a)
BocN
O
BocN
(b)
O
I–
BocN
N+
6
5
N
7
(c)
Surprisingly, this simple ether template was relatively un-
known at the commencement of our studies.2 For exam-
ple, we found no reference to the parent diamine, and
scarce reports of saturated 4,4¢-dipiperidinyl ethers of any
substitution in the last 100 years. First reported in 1909,2a
the 2,2¢,6,6¢-tetraphenyl decorated compound 2 was a side
product from condensation chemistry of acetophenones
(Figure 1). The bistropyl ether 3 is known,2b which con-
tains the 4,4¢- dipiperidinyl ether as a fragment of the
structure. In addition, N-methyldipiperidinyl ether is
found as a spacer group in a series of bicyclic pyrimidine
anti-inflammatory agents 4.2c,3 A recent paper by Chao et
al. highlighted this deficit and reported the synthesis of a
series of piperidine containing bis-N-heterocyclic amine
ethers by Mitsunobu reaction followed by acid-catalysed
pyridyl reduction.4a Our findings are reported herein.
O
O
(d)
⋅2HCl
HN
N
BocN
N
9
8
Scheme 1 Reagents and conditions: (a) 4-chloropyridine·HCl, NaH
(60% dispersion), DMSO, 70 °C (79%); (b) i-PrI, CH2Cl2, r.t.
(quant.); (c) LiBH4, ammonium formate, 10% Pd/C, MeOH, reflux;
(d) TFA, r.t. then 1 M HCl (93% over 2 steps). Alternative conditions:
(c) PtO2, H2 (3.45 bar, r.t.), EtOH, 6 d (54% 8 + approximately 20%
5 due to ether cleavage); or (c) i. NaBH4, MeOH; ii. ammonium for-
mate, 10% Pd/C, MeOH, reflux (66%).
This initial strategy was encouraging but capricious yields
in the reduction step led us to explore alternative ap-
proaches. Stepwise reduction of 7 using NaBH4 in MeOH
followed by transfer hydrogenation afforded 8 in reason-
able yield. More concisely, 8 could be obtained by direct
hydrogenation of 7, although yields were modest and vari-
able, and significant ether cleavage was observed.
O
CO2H
CO2H
O
N
N
Ar
N
N
1
Me
3
Me
NMe
The approach shown in Scheme 1 allowed us to access de-
rivatives containing an N-isopropyl substituent. In order
to investigate the effect of varying the substituents at both
terminal amino groups we required a more versatile late
stage intermediate such as 11 (Scheme 2). Direct reduc-
tion of 6 would be an attractive solution to this problem
but hydrogenation required forcing conditions which gave
an undesirable mixture containing starting material, the
desired product 11, and again significant cleavage of the
ether to afford the hydroxypiperidine 5 (Table 1, entry
1).4b
Ph
O
Ph
RHN
HN
NH
N
N
O
O
2
Ph
Ph
N
N
R2
4
Figure 1 Examples of dipiperidinyl ethers
SYNLETT 2009, No. 7, pp 1051–1054
x
x.
x
x.
2
0
0
9
Advanced online publication: 20.03.2009
DOI: 10.1055/s-0028-1088154; Art ID: D40308ST
© Georg Thieme Verlag Stuttgart · New York

1052
J. M. Bailey et al.
LETTER
Table 1 Reduction Conditions and Yields
Entry
Substrate
Conditions
Product
11 + 5
11
Yield (%)a
1
2
3
4
5
6
10
10
10
10
PtO2, H2 (3.45 bar), AcOH, EtOH, 3 d
n.d.
19
+
–
LiBH4, MeOH, NH4 HCO2 , 10% Pd/C, reflux
PtO2, H2 (1.013 bar), EtOH, 2.5 h
–
n.r.
60b
69
PtO2, H2 (3.45 bar), EtOH, 24 h
11
+
–
NaBH4, MeOH, NH4 HCO2 , 10% Pd/C, MeOH, reflux
11
a Yield of desired piperidine product; n.d. = not determined, n.r. = no reaction.
b Based on 1H NMR analysis of crude product mixture. Approximately 30% of N-(cyclohexyl)methylpyridinium species was also present.
Br–
BocN
O
BocN
O
BocN
O
BocN
O
(b)
(a)
(b)
(a)
10
6
N+
12
11
10
11
N
NH
NH
Bn
Bn
Scheme 3 Reagents and conditions: (a) NaBH4, MeOH, r.t., then
direct reduction possible?
10% Pd/C, H2 (1.013 bar, r.t.), acetone (84%); (b) see text
Scheme 2 Reagents and conditions: (a) BnBr, CH2Cl2, r.t. (99%);
(b) see Table 1
the transfer method employed above (see Table 1 entry 7).
For all conditions, the untreated material failed to react at
all. Both batches of purified 12 behaved similarly: Cata-
lytic hydrogenolysis failed to give any appreciable
amount of 11 (at 1.013 bar and 3.45 bar there was maxi-
mum 30% conversion as measured by HPLC), however,
in comparison the hydrogen-transfer reduction gave a
good yield of 11 (74%) from the purified benzylamine.
These observations led us to investigate the reduction of a
more suitable pyridine nucleus. Hence, intermediate 6
was converted to the N-benzyl quaternary ammonium salt
10 (Scheme 2). Reduction using the ‘one-pot’ LiBH4, fol-
lowed by transfer-hydrogenation conditions gave a poor
recovery of 11 (Table 1, entry 2). Catalytic hydrogenation
of 10 gave either unreacted starting material or was com-
plicated by competing reduction of the benzyl aromatic
group to give a mixture of products (entries 3 and 4). The
reduction was best accomplished by a two-stage process,
using NaBH4 followed by transfer hydrogenation, giving
the desired dipiperidine ether in an acceptable yield (entry
5). This four step synthesis of 11 from Boc-piperidinol 5
proceeded without the need for chromatography.
Two findings were evident from these studies. Firstly, it is
clear that the benzyl group in 10 and 12 is recalcitrant to-
wards removal by catalytic hydrogenolysis. Secondly, it
appears that some component of the crude reaction mix-
ture was inhibiting the debenzylation step; transfer hydro-
genation of purified 12 routinely gave good yields of 11.
We wanted to avoid chromatography on a large scale dur-
ing this work, and the quantities of activated charcoal re-
quired to remove impurities working on >50 gram scale
was inconvenient.
With a viable route to 11 in hand, we scaled up the synthe-
sis to provide bulk material for our medicinal chemistry
program. However, a reliable robust synthesis appeared
elusive. In one experiment the hydrogenation step re-
quired several charges of fresh catalyst before the reaction
went to completion. In other cases less than 10% of the de-
sired product was formed even after fresh catalyst and ex-
tended reaction times. Analysis of the crude mixtures
indicated that removal of the N-benzyl group required ex-
tended reaction times resulting in competing reactions.6
For these reasons we re-examined our large-scale synthe-
sis in an effort to eliminate any impurity at source. While
exploring different workup procedures for the pyridine
ether 6, it was found that the use of nonchlorinated extrac-
tion solvents such as ethyl acetate or diethyl ether ensured
that the subsequent reduction steps proceeded cleanly. Us-
ing CH2Cl2 in the workup gave us slightly higher crude
yields of 6, but with the capricious results discussed earli-
er. From these findings we inferred that a trace contami-
nant from step 1 (a, Scheme 1) appeared to poison the
catalyst during the production of amine intermediate 11.
We reasoned that a sulfur residue from the DMSO used in
the etherification step (a) was carried through during ex-
traction of the product into CH2Cl2. Simply switching to a
less polar extraction solvent ensured the preparation of 11
proceeded cleanly providing hundreds of grams of this
novel template in yields of up to 76% from pyridyl ether
6.9
In order to examine the debenzylation step in more detail,
borohydride reduction of 10 gave the tetrahydropyridine
ether, which was further reduced in situ to N-benzylpipe-
ridine 12 (Scheme 3). This was purified either by filtration
through silica, or by simply stirring with activated char-
coal.7 For comparison, an additional third batch of un-
treated 12 was used as a control sample.
These three batches were then exposed to the same reduc-
tion conditions, either hydrogenolysis over 10%Pd/C,8 or
Synlett 2009, No. 7, 1051–1054 © Thieme Stuttgart · New York

LETTER
Differentially Protected 4,4¢-Dipiperidinyl Ethers
1053
O
(b)
N
NBoc
(a)
(c)
Ar
13
O
O
HN
NBoc
N
N
Ar
R
11
15
O
(d)
BocN
N
R
14
Scheme 4 Reagents and conditions: (a) aryl halide, K2CO3, DMSO, 60–120 °C (56–100%); (b) i. 4 M HCl in dioxane or TFA, CH2Cl2 (95–
100%); ii. R2C=O, Et3N, NaBH(OAc)3 or RX, heat (34–77%); (c) R2C=O, Et3N, NaBH(OAc)3 or RX, heat (47–93%); (d) i. 4 M HCl in dioxane
(79–93%); ii. aryl halide, K2CO3, DMSO, 60–250 °C (thermal or MW heating, 9–100%), or Buchwald–Hartwig coupling (6–74%).
With supplies of intermediate 11 in hand, we were now racemic azepine alcohol 22. It is noteworthy that none of
able to explore variation at both the R and Ar positions these steps were optimised, yet they still provided good
(Scheme 4), a key focus of our medicinal chemistry pro- yields of these amine templates.
gram. Amine 11 was reacted with appropriate aryl halides
via direct displacement to afford excellent yields of Boc-
In conclusion, we have developed an efficient synthesis of
the 4,4¢-dipiperidinyl ether template 11, a novel scaffold
protected amines of formula 13. These were deprotected
and alkylated to give compounds of formula 15. This ini-
tial route via 13 provided a method to obtain and screen a
range of different alkyl groups (R), but a modification to
the synthetic route (Scheme 4, path c and d) was required
to allow late stage variation of the aryl group. To this end,
precursor 11 was first alkylated to give 14, followed by
deprotection and derivatisation with the appropriate aryl
halides either by direct displacement (e.g., using K2CO3,
DMSO, at elevated temperature in the microwave10 or
Buchwald displacement11). Unoptimised yields ranged
from no reaction in the case of secondary pyridinecarbox-
amides,12 to quantitative for activated 2-chloropyridines
or 4-fluoroacetophenone.
for the preparation of histamine H3 ligands.1 The current
route provides access to this template on multigram scale,
requires no chromatography, and solves several synthetic
problems encountered in the preparation. Importantly,
this key dipiperidine intermediate is monoprotected and is
suitable for further elaboration. The template can be deri-
vatised as desired through a variety of synthetic methods.
The chemistry is amenable to ring sizes from four to sev-
en, and proceeds with retention of stereochemistry in
chiral alcohols. Further details of our medicinal chemistry
program will be discussed elsewhere.
Acknowledgment
The authors would like to thank Dr Mervyn Thompson for his re-
view of the manuscript.
This chemistry could also be applied to the synthesis of
templates with alternative ring sizes (Scheme 5). The aze-
tidine-containing scaffold 18 was readily prepared from
commercially available azetidinol 16. Switching to the
Boc-protected amine 19 and generating the benzyl salt of
20 (analogous to dipiperidine intermediate 10) resulted in
successful reduction to give pyrrolidinyl ether 21. Simi-
larly, ent-21 could also be prepared from (3S)-19 (not
shown), and comparable results were observed with the
References and Notes
(1) Bailey, J. M.; Bruton, G.; Huxley, A.; Milner, P. H.; Orlek,
B. S. WO 2005014571, 2005.
(2) (a) Fromm, E. Ber. Dtsch. Chem. Ges. 1909, 41, 3644.
(b) Gadamer, J. Arch. Pharm. 1921, 259, 234. (c) Arai, H.;
Matsumura, T.; Ishida, H.; Yamaura, Y.; Aratake, S.;
OH
O
O
(a)
(b)
HN
Ph
N
N
N
N
Ph
Ph
Ph
16
OH
17
18
O
O
(a)
(c)
N
NH
N
N
N
Boc
Boc
19
20
Boc
21
O
OH
O
(a)
(c)
BocN
BocN
BocN
NH
N
23
24
22
Scheme 5 Reagents and conditions: (a) 4-chloropyridine·HCl, NaH (60% dispersion), DMSO, 70 °C (54–92%); (b) i-PrI, CH2Cl2, r.t., then
+
–
+
–
NaBH4, MeOH, NH4 HCO2 , 10% Pd/C, MeOH, reflux (quant.); (c) BnBr, CH2Cl2, r.t., then NaBH4, MeOH, NH4 HCO2 , 10% Pd/C, MeOH,
reflux (33–73%).
Synlett 2009, No. 7, 1051–1054 © Thieme Stuttgart · New York

1054
J. M. Bailey et al.
LETTER
Ohshima, E.; Yanagawa, K.; Miyama, M.; Suzuki, K.;
Kawabe, A.; Nakanishi, S.; Kobayashi, K.; Sato, T.; Miki, I.;
Ueno, K.; Fujii, S.; Iwase, M. WO 2003104230, 2003; JP
2002-166504, 2002.
stirred at r.t. for 4 h. The solvent was removed in vacuo, and
the crude residue was redissolved in a minimum quantity of
CH2Cl2. Diethyl ether was added to the stirred CH2Cl2
solution until the product precipitated. The pale pink solid
was isolated by filtration and dried at 50 °C under high
vacuum overnight to give 10 (60.0 g) as a pink solid. MS
(ES+): m/z = 369 [M+]. 1H NMR (400 MHz, CDCl3): d =
9.20 (2 H, d, J = 7.5 Hz), 7.58 (2 H, m), 7.50 (2 H, d, J = 7.5
Hz), 7.41 (3 H, m), 6.04 (2 H, s), 5.00 (1 H, m), 3.71 (2 H,
m), 3.38 (2 H, m), 2.03 (2 H, m), 1.74 (2 H, m), 1.47 (9 H, s).
Compound 10 (55.2 g) was stirred in MeOH (300 mL) under
argon at 0 °C and NaBH4 (pellets, 9.3 g) was added
portionwise over 1 h. The reaction was allowed to warm to
r.t. for a further 90 min, and then acetone (50 mL) was added.
The reaction was stirred for 1 h. The solvent was evaporated
and residue partitioned between sat. aq NaHCO3 solution
and EtOAc (200 mL of each). The aqueous phase was
separated and re-extracted with EtOAc (3 × 100 mL). The
combined organics were dried over MgSO4 with activated
charcoal added, filtered, and evaporated to give the enol
ether as a yellow-to-pink oil (40.6 g).
The crude oil (59.2 g) was dissolved in MeOH (900 mL) and
ammonium formate (100.2 g) was added followed by 10%
Pd/C (paste, 30 g). The reaction was heated to 60 °C (bath
temperature, when internal temperature achieved 30 °C
effervescence was observed), maintained at 55 °C for 1.5 h.
The reaction was filtered and concentrated. The residue was
re-dissolved in EtOAc (1 L) and washed with sat. K2CO3
solution (3 × 400 mL), dried (MgSO4), and evaporated to
give an oil which crystallized on standing to give 11 as a
white solid (39.0 g). MS (ES+): m/z = 285 [MH+]. 1H NMR
(400 MHz, CDCl3): d = 3.79 (2 H, m), 3.65–3.38 (2 H, m),
3.06 (4 H, m), 2.60 (2 H, m), 1.91–1.67 (4 H, m), 1.59–1.31
(13 H, m).
(3) (a) Ye, X. M.; Garofalo, A. W.; Lawler, R. D.; Fukuda, J. Y.;
Konradi, A. W.; Holcomb, R.; Rossiter, K. I.; Wone, D. W.
G.; Wu, J. WO 2006113140, 2006. (b) Botez, I.; David-
Basei, C.; Gourlaoueen, N.; Nicolaie, E.; Balavoine, F.;
Valette, G.; Serradeil-Le Gal, C. WO 2006108965, 2006.
(4) (a) Chao, J.; Israiel, M.; Zheng, J.; Aki, C. Tetrahedron Lett.
2007, 48, 791. (b) The authors have reported similar
findings in their two-step synthesis of 10, which proceeds in
37–40% yield and requires chromatography after both steps.
(5) All products gave satisfactory MS and 1H NMR (400 MHz)
spectra.
(6) For example, on a 50 gram scale, extended reaction times
resulted in the deprotected amine reacting with excess
ammonium formate in the reaction mixture to give N-formyl
piperidines.
(7) This latter procedure was chosen to mimic the contact with
the Pd/C catalyst experienced during hydrogenolysis
experiments, which in some cases would drive the reduction
over time.
(8) More vigorous conditions were avoided due to our previous
observation that the phenyl methyl group in 10 was labile to
reduction (see Table 1, footnote b).
(9) Synthesis of Compounds 6, 10, and 11
Sodium hydride (20.88 g) was suspended in DMSO (600
mL) under argon and 4-chloropyridine hydrochloride (31.0
g), suspended in DMSO (150 mL), was added slowly over
45 min. The reaction was then stirred for 10 min, and 5 (35
g), dissolved in DMSO (150 mL), was added over 15 min.
The reaction was stirred at r.t. overnight. Saturated NaHCO3
solution (150 mL) was then added slowly and the reaction
stirred for 20 min. The mixture was evaporated to a
minimum, redissolved in EtOAc (600 mL), and washed with
sat. NaHCO3 (150 mL) and H2O (150 mL), followed by H2O
(5 × 250 mL). The organic layer was then dried (MgSO4).
The solution was filtered and evaporated to give a yellow
solid, which was triturated with hexane and then dried at
50 °C overnight to give 6 as a pale yellow solid (38.0 g).
MS (ES+): m/z = 279 [MH+]. 1H NMR (400 MHz, CDCl3):
d = 8.43 (2 H, d, J = 4.8 Hz), 6.87 (2 H, d, J = 4.8 Hz), 4.57
(1 H, m), 3.68 (2 H, m), 3.37 (2 H, m), 1.90 (2 H, m), 1.78 (2
H, m), 1.47 (9 H, s).
(10) (a) Kappe, O. C. Chimia 2006, 60, 308. (b) Kappe, O. C.;
Dallinger, D. Nat. Rev. Drug Discovery 2006, 5, 51.
(11) (a) Jiang, L.; Buchwald, S. L. In Metal-Catalysed Cross-
Coupling Reactions, 2nd ed., Vol. 2; de Meijere, A.;
Diederich, F., Eds.; Wiley-VCH: Weinheim, 2004, Chap.
13, 699. (b) Buchwald, S. L.; Muci, A. R. Top. Curr. Chem.
2002, 219, 131.
(12) Recently advances in catalysis have shown that palladium-
catalyzed C–N arylation is possible in the presence of
heteroaromatic halides and NH carboxamides. See, for
example: Anderson, K. W.; Tundel, R. E.; Ikawa, T.;
Altman, R. A.; Buchwald, S. L. Angew. Chem. Int. Ed. 2006,
45, 652.
Compound 6 (37.5g) was dissolved in CH2Cl2 (400 mL).
Benzyl bromide (32.26 mL) was added, and the reaction was
Synlett 2009, No. 7, 1051–1054 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.