A highly selective monoalkylation of tert-butyl 3,4-dihydroxy-4- phenylpiperidine-1-carboxylate under phase transfer conditions

Article abstract of DOI:10.1016/j.tetlet.2011.01.029

Liquid-liquid phase transfer conditions have been found to provide highly selective monoalkylation of tert-butyl 3,4-dihydroxy-4-phenylpiperidine-1- carboxylate, a transformation for which many other common alkylation protocols proved inadequate. The high

Full text of DOI:10.1016/j.tetlet.2011.01.029

Tetrahedron Letters 52 (2011) 1199–1201
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
A highly selective monoalkylation of tert-butyl 3,4-dihydroxy-4-
phenylpiperidine-1-carboxylate under phase transfer conditions
⇑
Mark I. Lansdell , David Fradet
Pfizer Global Research and Development, Ramsgate Road, Sandwich, Kent CT13 9NJ, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Liquid–liquid phase transfer conditions have been found to provide highly selective monoalkylation of
tert-butyl 3,4-dihydroxy-4-phenylpiperidine-1-carboxylate, a transformation for which many other com-
mon alkylation protocols proved inadequate. The high preference for 3-O-alkylation in the phase transfer
alkylation is emphasised by the absence of diether formation even in the presence of a large excess of
reagents.
Received 1 September 2010
Revised 18 December 2010
Accepted 7 January 2011
Available online 14 January 2011
Ó 2011 Elsevier Ltd. All rights reserved.
F
We have recently described the discovery of PF-00446687
(Fig. 1), a potent and selective small-molecule agonist of the mel-
anocortin-4 (MC4) receptor, which showed efficacy in a pilot clin-
ical study of male erectile dysfunction, thereby providing the first
evidence that selective MC4 receptor activation may be sufficient
to elicit pro-erectile effects in humans.1
O
HO
N
F
N
A key structure–activity relationship (SAR) observed during the
optimization of the chemical series was the profound potency
enhancement resulting from the addition of methyl substituents
at C3 and C5 of the piperidine ring. To investigate more fully the
SAR in this region, we wished to prepare analogues containing alk-
oxy substituents at C3/C5. We identified the piperidine diol 2/ent-2
as a key potential intermediate, attracted by its straightforward
preparation via Sharpless asymmetric dihydroxylation of a pro-
tected tetrahydropyridine 1 (Scheme 1).2
Despite the simple accessibility of such enantiopure piperidine
3,4-diols, there have been remarkably few reports of their selective
mono-O-alkylation. The few successful examples have been lim-
ited to benzylations,3,4 and the only previous monomethylation
of which we are aware proceeded in very low yield with significant
quantities of starting material recovered.5 In our own hands, a sim-
ilar protocol based on deprotonation with sodium hydride pro-
vided a quantitative yield of the diether 3 when excess reagents
were employed (Table 1, entry 1),6 but attempts to achieve mono-
alkylation by limiting either the base or iodomethane to stoichiom-
etric quantities were unsuccessful: the crude reaction mixtures
were found to contain a mixture of unreacted diol and all three
possible alkylation products (diether 3 and each of the mono-
ethers). Separation of the components on a preparative scale
proved difficult, and variations in the quantities of reagents used
or in the reaction procedure provided inadequate alteration of
Figure 1. Structure of PF-00446687.
OH
OH
α
AD-mix-
N
O
O
O
2
t-BuOH/H2O
0-15 °C, 16 h, 90%
N
OH
O
O
OH
−β
AD-mix
1
N
O
ent-2
Scheme 1.
the product ratios. Consequently, we sought an alternative method
for achieving the desired monoalkylation.
In parallel with our attempts to identify a direct monoalkylation
protocol, we also explored the potential of employing a protecting
group strategy. Whilst ultimately unsuccessful in terms of provid-
ing access to the desired targets, these efforts yielded a useful
insight into the relative reactivity of the two hydroxyl groups.
For example, it was found that good selectivity can be achieved
⇑
Corresponding author. Tel.: +44 1304 645037; fax: +44 1304 651819.
E-mail address: mark.i.lansdell@googlemail.com (M.I. Lansdell).
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.01.029

1200
M. I. Lansdell, D. Fradet / Tetrahedron Letters 52 (2011) 1199–1201
Table 1
Reagents and conditions for attempted selective derivatisation of diol 2
2
OR
OR
OH
OH
1
N
N
O
O
O
O
2
3-11
Entry
Reagent
Conditionsa
R1
R2
Product
Yieldb
1
2
MeI (3 equiv)
Ac2O (2.5 equiv)
NaH (3 equiv), THF, 4 h
Et3N, CH2Cl2, 16 h
Me
Ac
Me
H
3
4
100%
89%
Ac
Ac
H
TMS
Me
H
Me
Me
H
Me
H
Me
H
Me
H
Me
H
H
5
6
7
3
8
9
3
8
3
8
3
8
3
8
9
10
11
5%
100%
87%
10%
80%
Trace
3
4
5
TMSCl (4 equiv)
TMSCl (4 equiv)
Me3OBF4 (4 equiv)
Et3N, CH2Cl2, 16 h
Imidazole, DMF, 16 h
Proton sponge (4 equiv), CH2Cl2, 72 h
TMS
TMS
Me
Me
H
Me
Me
Me
Me
Me
Me
Me
Me
H
6
7
8
9
MeI (10 equiv)
MeI (10 equiv)
MeI (20 equiv)
MeI (20 equiv)
KOH(s) (10 equiv), DMSO, 5 min
KOH(s) (10 equiv), DMSO, 15 min
K2CO3(s) (10 equiv), MeCN, reflux, 72 h
Tracec
Major productc
Major productc
Not detectedc
Not detected
10%
Trace
96%
1.5%
92%
NaOH (20 equiv), Bu4NHSO4 (1 equiv), H2O/PhMe (1:1), 16 h
10
11
EtI (20 equiv)
BnBr (2 equiv)
NaOH (20 equiv), Bu4NHSO4 (1 equiv), H2O/PhMe (1:1), 16 h
NaOH (10 equiv), Bu4NHSO4 (0.5 equiv), H2O/PhMe (1:1), 16 h
Et
Bn
93%
a
b
c
All reactions carried out at 20 °C unless otherwise stated.
Isolated yields after chromatography unless otherwise stated.
Judged by TLC.
by direct reaction of the diol with acetic anhydride in the absence
of prior alkoxide generation, producing only 5% of diacetate 5 even
in the presence of excess reagent over an extended reaction time
(Table 1, entry 2).
on silica). As a consequence, examination of alternative alkylation
conditions continued, the focus now moving to those under which
alkylation occurs via the intermediacy of an alkoxide, but without
complete deprotonation prior to exposure to the electrophile. Use
of powdered potassium hydroxide in DMSO9 provided in principle
the necessary selectivity (Table 1, entry 6), but subsequent pro-
gression to the dialkylated product occurred too rapidly to be of
practical use (Table 1, entry 7; no attempt was made to limit the
quantities of reagents or reduce the temperature, owing to the
more favourable results described below). The selectivity similarly
offered by use of potassium carbonate in acetonitrile at reflux10
was, in contrast, rendered impractical by the very slow progression
of the reaction (Table 1, entry 8).
The desired balance of reaction rate versus selectivity was final-
ly obtained with the use of liquid–liquid phase transfer conditions.
Such conditions are well known for alkylation of alcohols with
benzylic and allylic halides, but they have been relatively seldom
used for simple alkyl halides. In the present case, reaction of the
diol with excess iodomethane at room temperature in a rapidly
stirred two-phase water/toluene system containing sodium
hydroxide and a stoichiometric quantity of a phase transfer cata-
lyst provided an excellent product profile within a reasonable reac-
tion time (Table 1, entry 9).
Most significantly, only a trace of dialkylated product 3 was
generated, which enabled vastly simpler purification of the desired
mono-3-O-methylated product 8 (the presence of mono-4-O-
methylated by-product 9 is less problematic for purification owing
to a larger polarity difference). The high selectivity for monoalkyl-
ation is particularly notable given the large excess of electrophile
used in this reaction.
Reaction with iodoethane (Table 1, entry 10) occurs equally
smoothly under these conditions. In the case of benzyl bromide
(Table 1, entry 11), reduced quantities of reagents may be em-
ployed without detriment to the yield or rate of reaction.
However, it also became clear that the extent of selectivity was in
some cases highly dependent on the choice of reaction conditions.
For example, reaction with excess chlorotrimethylsilane in the pres-
ence of triethylamine in dichloromethane provided monosilylether
6 with complete selectivity even after an extended reaction time
(Table 1, entry 3), whereas exposure to the same excess of reagent
in the presence of imidazole in DMF produced smooth formation
of the disilylether 7 in the same time period (Table 1, entry 4).
These results encouraged an examination of the methods for
achieving monoalkylation not dependent on complete formal alk-
oxide formation. Numerous methods based on the activation by
silver(I) salts7 were found to be unsuccessful owing to impracti-
cally slow or incomplete conversion or, under more forcing condi-
tions, substrate decomposition. Our attention then turned to
metal-free conditions promoted by Evans8 as useful alternatives
to standard alkoxide procedures, relying on particularly active
alkylating agents. Disappointingly, no reaction was obtained with
methyl triflate in the presence of 2,6-di-tert-butylpyridine at room
temperature, and at elevated temperatures only decomposition of
the diol was observed. Much more encouraging was the reaction
with trimethyloxonium tetrafluoroborate in the presence of proton
sponge (Table 1, entry 5). This delivered by far the cleanest mono-
alkylation seen by us at that point, providing the desired monoe-
ther 8 in 80% yield.
Despite this significant step forward, further improvements
remained desirable. Not only was the reaction very slow (the mass
balance in this reaction at three days comprised ꢀ10% unreacted
2), but also there was still sufficient formation of dialkylated prod-
uct 3 to complicate purification of the targeted 3-O-monoalkylated
product 8 (the two compounds having very similar retention times

M. I. Lansdell, D. Fradet / Tetrahedron Letters 52 (2011) 1199–1201
1201
References and notes
OH
OBn
OR
OBn
OR
1. Lansdell, M. I.; Hepworth, D.; Calabrese, A.; Brown, A. D.; Blagg, J.; Burring, D. J.;
Wilson, P.; Fradet, D.; Brown, T. B.; Quinton, F.; Mistry, N.; Tang, K.; Mount, N.;
Stacey, P.; Edmunds, N.; Adams, C.; Gaboardi, S.; Neal-Morgan, S.; Wayman, C.;
Cole, S.; Phipps, J.; Lewis, M.; Verrier, H.; Gillon, V.; Feeder, N.; Heatherington,
A.; Sultana, S.; Haughie, S.; Martin, S. W.; Sudworth, M.; Tweedy, S. J. Med.
Chem. 2010, 53, 3183–3197.
2. Bursavich, M. G.; West, C. W.; Rich, D. H. Org. Lett. 2001, 3, 2317–2320.
3. Chang, M.-Y.; Pai, C.-L.; Kung, Y.-H. Tetrahedron Lett. 2005, 46, 8463–8465.
4. Kim, K.; Liu, E. A.; Mischke, S. G. US Patent Appl. 20040180929; Chem. Abstr.
2004, 141, 277500.
5. Davies, D. T.; Jones, G. E.; Markwell, R. E.; Pearson, N. D. International PCT
patent application WO2003010138; Chem. Abstr. 2003, 138, 153541.
6. Andrews, M. D.; Brown, A. D.; Fradet, D. S.; Lansdell, M. I. International PCT
patent application WO2007015162; Chem. Abstr. 2007, 146, 229186.
7. Burk, R. M.; Gac, T. S.; Roof, M. B. Tetrahedron Lett. 1994, 35, 8111–8112.
8. Evans, D. A.; Ratz, A. M.; Huff, B. E.; Sheppard, G. S. Tetrahedron Lett. 1994, 34,
7171–7172.
OH
b
a
N
N
N
O
O
O
O
O
O
11
12 R = Me
13 R = Et
14 R = Me
15 R = Et
Scheme 2. Reagents and conditions: (a) RI (2 equiv), KOH(s) (4 equiv), DMSO, 20 °C,
72 h, 98% for 12, 66% for 13; (b) 15 wt % of 20 wt % Pd(OH)2/C, 1-methylcyclohexa-
1,4-diene, EtOH, reflux, 1.5 h, 100% for 14, 100% for 15.
Access to mono-4-O-alkylated compounds was provided by
derivatisation of benzyl ether 11 with haloalkanes in DMSO in
the presence of powdered potassium hydroxide,9 followed by
benzyl-deprotection of the resulting diethers 12 and 13 under
transfer hydrogenation conditions (Scheme 2).
The O-alkylated piperidines 3, 8–11, 14 and 15 shown above
were all cleanly Boc-deprotected (4 M hydrogen chloride in 1,4-
dioxane (10 equiv), dichloromethane, 20 °C, 30–60 min) to give
quantitative yields of the corresponding amines as their hydrogen
chloride salts.
In conclusion, we have identified that a convenient phase trans-
fer alkylation protocol6,11 provides high selectivity in the derivati-
sation of diol 2, a challenging setting in which many other common
conditions for hydroxyl alkylation have proven inadequate. The
MC4 agonist activity of compounds analogous to PF-00446687 de-
rived from such 3- and 4-O-alkylated piperidines will be described
in detail elsewhere.
9. Johnstone, R. A. W.; Rose, M. E. Tetrahedron 1979, 35, 2169–2173.
10. Davis, R.; Muchowski, J. M. Synthesis 1982, 987–988.
11. Representative experimental procedure. A solution of NaOH (544 mg, 13.6 mmol)
in H2O (3.4 mL) was added to a solution of tert-butyl (3S,4S)-3,4-dihydroxy-4-
phenylpiperidine-1-carboxylate (2) (200 mg, 0.68 mmol) in PhMe (3.4 mL)
followed by MeI (0.85 mL, 13.6 mmol) and Bu4NHSO4 (231 mg, 0.68 mmol).
The mixture was stirred vigorously at 20 °C for 16 h, then diluted with H2O
(20 mL) and extracted with CH2Cl2 (3 Â 20 mL). The combined organic layers
were dried (MgSO4) and filtered, and the filtrate was concentrated in vacuo.
The residual oil was purified by column chromatography on silica gel (pentane/
EtOAc 1:0–7:3 gradient elution) to give methyl ether 8 as a colourless oil
(200 mg, 96%). 1H NMR (CD3OD, 400 MHz) d 1.50 (s, 9H), 1.68 (dt, J = 2.4,
14.2 Hz, 1H), 1.93 (td, J = 4.7, 13.3, 13.5 Hz, 1H), 3.03 (br s, 1H), 3.11 (s, 3H),
3.17 (br s, 1H), 3.57 (dd, J = 4.9, 10.5 Hz, 1H), 3.85–3.90 (m, 1H), 4.19 (br s, 1H),
7.23 (t, J = 7.4 Hz, 1H), 7.34 (t, J = 7.7 Hz, 2H), 7.51 (d, J = 7.7 Hz, 2H); 13C NMR
(CD3OD, 500 MHz) d 28.7, 39.5, 54.8, 58.5, 75.0, 81.3, 81.8, 126.2, 127.8, 129.1,
147.3, 156.5; LRMS (APCI) m/z 208 [MH À Boc]+. HRMS (ESI+) found m/z
330.1687 (calculated for C17H25N1Na1O4 = 330.167579).
Acknowledgement
We thank Denise Burring for assistance in preparing this
manuscript.