A practical synthesis of 4-amino-2-(trifluoromethyl)nicotinic acid

Article abstract of DOI:10.1055/s-0030-1258481

A practical synthesis of 4-amino-2-(trifluoromethyl)-nicotinic acid is described. 2-(Trifluoromethyl)pyridine was lithiated using lithium 2,2,6,6-tetramethylpiperidide (LTMP) in the presence of 1,3-dimethyl-2- imidazolidinone (DMI) and followed by CO₂ quench to give the C-3 carboxylation product. Subsequent directed C-4 lithiation of carboxylation product afforded 4-iodo-2-(trifluoromethyl)nicotinic acid, which was coupled with tert-butyl carbamate under Pd-catalyzed conditions and followed by Boc deprotection to yield the title product in four steps and 50% overall yield. Georg Thieme Verlag Stuttgart New York.

Full text of DOI:10.1055/s-0030-1258481

LETTER
2133
A Practical Synthesis of 4-Amino-2-(Trifluoromethyl)nicotinic Acid
S
ynthesis of 4-A
m
r
ino-2-(
y
T
rifluoromet
a
hyl)nicotin
n
ic
A
cid Li,* F. Christopher Bi,1 Richard A. Buzon, Ming Kang, Robert M. Oliver, John F. Sagal, Lacey Samp,
Daniel P. Walker, Zhijun Zhang
Pfizer Global Research and Development, Groton Laboratories, Groton, CT 06340, USA
E-mail: bryan.li@pfizer.com
Received 3 February 2010
rination step was not considered an environmentally be-
nign reagent. On the other hand, a known literature
synthesis of the ethyl ester of 14 involved the highly toxic
gaseous and low boiling (bp –64 °C5) trifluoroacetonitrile
Abstract: A practical synthesis of 4-amino-2-(trifluoromethyl)-
nicotinic acid is described. 2-(Trifluoromethyl)pyridine was lithiat-
ed using lithium 2,2,6,6-tetramethylpiperidide (LTMP) in the pres-
ence of 1,3-dimethyl-2-imidazolidinone (DMI) and followed by
CO2 quench to give the C-3 carboxylation product. Subsequent di- as the starting material.
rected C-4 lithiation of carboxylation product afforded 4-iodo-2-
Attracted by the metalation reaction of 2-(trifluorometh-
(trifluoromethyl)nicotinic acid, which was coupled with tert-butyl
carbamate under Pd-catalyzed conditions and followed by Boc
deprotection to yield the title product in four steps and 50% overall
yield.
yl)pyridine (2) at the C-3 described by Schlosser et al.6
and the availability of 2 in bulk from a number of suppli-
ers,7 we decided to enable a scalable synthesis starting
from 2. Following the reported procedures (LTMP, –78 °C,
6 h; CO2), the C-3 carboxylated product 3 was obtained in
60–72% yields. In this reaction, the lithium anion solution
was required to equilibrate for six hours at –78 °C before
it was quenched with CO2. The long equilibration time
was necessary to allow the C-4 lithiopyridine anions
formed initially under kinetic control to equilibrate to the
thermodynamically more favorable C-3 lithiopyridine.
Since it is costly to run cryogenic reactions on large scales
for long hours, we sought to shorten the lithium anion
equilibration time, and found the introduction of 1.0
equivalent of 1,3-dimethyl-2-imidazolidinone (DMI)8
dramatically accelerated the anion equilibrium process.
Thus, quenching the anion solution with CO2 within one
hour gave 3 in 63–77% yields, and the carboxylation
product resulted from C-4 lithiation was not detected un-
der the reaction conditions. With 3 in hand, the C-4
lithiation9 and halogenation were carried out uneventfully
under similar conditions shown in Scheme 1. We opted to
Key words: pyridine, directed lithiation, trifluoromethyl, iodina-
tion, Boc-amide, amination
4-Amino nicotinic acids are useful building blocks for
syntheses of heterocyclic molecules of medicinal rele-
vance, such as naphthyridines, azaquinazolines, aza-
quinazolinones, and pyridyldiazepinones.2 Recently, to
support the preparation of 5-(trifluoromethyl)pyrido[4,3-
d]pyrimidine-4(3H)-one derivatives as calcium-sensing
receptor antagonists for treatment of osteoporosis,3 we
were tasked to devise a practical process-scale synthesis
of the critical intermediate 4-amino-2-(trifluorometh-
yl)nicotinic acid (1, Scheme 1).
NH2
N
R1
R2
CO2H
CF3
N
N
N
1
CF3
O
Br
Br
I
Br
Scheme 1 Retrosynthetic analysis of azaquiazolinones
c
a
b
N
Br
N
N
CF3
Although the original eight-step linear sequence starting
from 2,3-dibromopyridine (Scheme 2) served well for
discovery analogue syntheses, it suffered several draw-
backs limiting its potential as a feasible scale-up option.
Specifically, in this synthesis, the installation of the CF3
group at C-2 position with Ruppert’s reagent was de-
manding, which required flame-heating a mixture of
spray-dried potassium iodide and copper(I) iodide at a re-
duced pressure of 1.33·10–3 bar. The synthesis was also
complicated with thermal hazards associated with the in-
volvement of sodium azide, diazomethane, and an azide
intermediate. Furthermore, hexachloroethane in the chlo-
Cl
Cl
CO2H
CF3
CO2Me
CF3
CO2H
CF3
f
d
e
N
N
N
NH2•HCl
CO2H
N3
NH2
CO2Me
CF3
CO2Me
CF3
g
h
N
CF3
N
N
1
Scheme 2 Original synthesis of 4-amino-2-trifluoromethyl nicoti-
nic acid. Reagents and conditions: (a) TMSCl, NaI, EtCN, reflux,
85%; (b) TMSCF3, KF (spray-dried), CuI, NMP, r.t. to 35 °C, 62%;
(c) Mg turnings, THF, reflux; CO2 (g), 74%; (d) LTMP, THF, –70 °C,
–85 °C to –75 °C, then C2Cl6, –75 °C. (e) CH2N2, EtOAc, 99%; (f)
NaN3, DMF, 45–50 °C, 98%; (g) concd HI, 0 °C, 96%; (h) LiOH,
THF–dioxane–H2O (3:2:1), 85 °C, 95%.
SYNLETT 2010, No. 14, pp 2133–2135
x
x
.x
x
.2
0
1
0
Advanced online publication: 09.07.2010
DOI: 10.1055/s-0030-1258481; Art ID: S00410ST
© Georg Thieme Verlag Stuttgart · New York

2134
B. Li et al.
LETTER
use iodine instead of hexachloroethane as halogenation removed by recrystallization after converting the crude
reagent, and obtained 47e in 65–77% yields. Both lithia- product to the hydrochloride salt, and the desired product
tion steps were scaled up successfully to multikilogram was isolated in 74% yield without chromatography.
batches.
With the robustness issues encountered in the benzyl-
To streamline the synthesis, we planned to avoid the ester amine displacement and the subsequent debenzylation
intermediate and the subsequent step of ester hydrolysis. steps in mind, we turned our attention to Pd-catalyzed aryl
Although amination of halogen-substituted nicotinic acids amination. As the free-base form of 1 is zwitterionic, and
using ammonia is not reported,10 2-halo benzoic acids its isolation from aqueous phase is cumbersome, we
were known to undergo direct substitution using ammonia quickly settled on the use of tert-butyl carbamate as the
(ethanolic or aqueous) to give the corresponding 2-ami- choice of the nitrogen nucleophile for the convenience of
nobenzoic acids.11 The electron-withdrawing trifluoro- Boc removal and formation of the hydrochloride salt in
methyl group at the C-2 was expected to facilitate the one operation. Uneventfully, reaction of 4 with tert-butyl
SNAr reaction. Thus, we attempted the use of ammonia (7 carbamate in the presence of Pd2(dba)3 and Xantphos pro-
M in MeOH or concd NH4OH; with or without the pres- ceeded to give 5b in 84% yield (Scheme 3).12 The removal
ence of copper salts as catalysts) under sealed-tube condi- of the Boc group was readily effected with HCl (g) to give
tions (170 °C bath temperature, 4 h). To our surprise, we 113 as a hydrochloride salt in quantitative yield. This
did not observe any desired product from the SNAr reac- method was proved superior to the benzylamine substitu-
tion, rather, 4-iodo-2-(trifluoromethyl)pyridine, resulted tion approach and allowed us to provide substantial quan-
from the decarboxylation, was detected at ca. 25% (by tities of 1 to support preclinical needs.
HPLC) under these conditions. Nevertheless, the employ-
In summary, we have developed a practical synthesis of
ment of large excess of benzylamine (5.8 equiv) at 120 °C
4-amino-2-(trifluoromethyl)nicotinic acid from commer-
gave the desired product 5a as the major peak by HPLC
cially available 2-(trifluoromethyl)pyridine in four steps
analysis. Again 5–8% of 4-iodo-2-(trifluoromethyl)pyri-
and 50% overall yield. The C-3 carboxylation step was
dine was observed in the reaction. Removal of the excess
made more feasible on production scale with the addition
benzylamine was found tedious. Further decarboxylation
of DMI as a co-solvent that significantly shortened the an-
occurred when the reaction mixture was subjected to high
ion equilibration time in a cryogenic reactor. The Pd-cat-
alyzed amination of 4 in its free carboxylic acid form
further streamlined the synthesis. It is noteworthy that
vacuum distillation. It was subsequently found convenient
to derivatize the excess benzylamine to the corresponding
acetamide (Ac2O–Et3N) that was readily removed by
analogous amination reactions precedented in the
acid–base extractive workup. Compound 5a was isolated
literature14 had been limited to esters of nicotinic acids.
in 73% yield after crystallization using this method.
References and Notes
I
CO2H
CF3
CO2H
CF3
(1) Current Address: Pfizer Global Research & Development,
10770 Science Center Drive, San Diego, CA 92121, USA.
(2) (a) Hawes, E. M.; Gorecki, D. K. J. J. Heterocycl. Chem.
1972, 9, 703. (b) Armarego, W. L. F. J. Chem. Soc. 1962,
4094. (c) Ismail, A. G.; Wibberley Denman, G. J. Chem.
Soc. 1967, 24, 2613. (d) Stukenbrock, H.; Mussmann, R.;
Geese, M.; Ferandin, Y.; Lozach, O.; Lemcke, T.; Kegel, S.;
Lomow, A.; Burk, U.; Dohrmann, C.; Meijer, L.; Austen,
M.; Kunick, C. J. Med. Chem. 2008, 51, 2196. (e) Wall,
M. J.; Chen, J.; Meegalla, S.; Ballentine, S. K.; Wilson,
K. J.; DesJarlais, R. L.; Schubert, C.; Chaikin, M. A.;
Crysler, C.; Petrounia, I. P.; Donatelli, R. R.; Yurkow, E. J.;
Boczon, L.; Mazzulla, M.; Player, M. R.; Patch, R. J.;
Manthey, C. L.; Molloy, C.; Tomczuk, B.; Illig, C. R.
Bioorg. Med. Chem. Lett. 2008, 18, 2097. (f) Boerjesson,
L.; Connolly, S.; Johansson, H.; Kristoffersson, A.;
Linnanen, T.; Shamovsky, I.; Skrinjar, M. WO 2006-
SE1012 20060904, 2007; Chem. Abstr. 2007, 146, 337739.
(g) Nagase, T.; Sato, N.; Kanatani, A.; Tokita, S. US
2005182045, 2005; Chem. Abstr. 2005, 143, 211923.
(3) Didiuk, M. T.; Griffith, D. A.; Benbow, J. W.; Liu, K. C.;
Walker, D. P.; Bi, F. C.; Morris, J.; Guzman-Perez, A.; Gao,
H.; Bechle, B. M.; Kelley, R. M.; Yang, X.; Dirico, K.;
Ahmed, S.; Hungerford, W.; DiBrinno, J.; Zawistoski,
M. P.; Bagley, S. W.; Li, J.; Zeng, Y.; Santucci, S.; Oliver,
R.; Corbett, M.; Olson, T.; Chen, C.; Li, M.; Paralkar, V. M.;
Riccardi, K. A.; Healy, D. R.; Kalgutkar, A. S.; Maurer,
T. S.; Nguyen, H. T.; Frederick, K. S. Bioorg. Med. Chem.
Lett. 2009, 19, 4555.
a
b
N
CF3
N
N
2
3
4
R
NH2•HCl
CO2H
CO2H
CF3
c or d
e or f
N
N
CF3
5a R = NHBn
5b R = NHBoc
1
Scheme 3 Synthesis of 4-amino-2-(trifluoromethyl)nicotinic acid.
Reagents and conditions: (a) LTMP, –70 °C, then CO2 (g) or dry ice,
77%; (b) LTMP, –70 °C; then I2; (c) BnNH2, 120 °C, then Ac2O,
Et3N; (d) BocNH2, Pd2(dba)3, Xantphos, Cs2CO3, tert-amyl alcohol,
84%; (e) H2 (100 psi), Pd(OH)2/C, MeOH, AcOH, 50 °C, 73%; (f)
HCl (g), MeOH, 100%.
The debenzylation also turned out to be troublesome due
to the complication of the pyridine reduction giving rise to
the corresponding piperidine impurities. After screening
the hydrogenolysis conditions, the reaction was optimized
to run in methanol (10 mL/g) and acetic acid (2 mL/g) un-
der 6.89 bar H2 in the presence of Pd(OH)2/C (10% wt/wt)
at 50 °C. Under these conditions, the pyridine reduction
was minimized to ca. 10%. The piperidine impurities were
Synlett 2010, No. 14, 2133–2135 © Thieme Stuttgart · New York

LETTER
Synthesis of 4-Amino-2-(Trifluoromethyl)nicotinic Acid
2135
(4) (a) Vasil’ev, L. S.; Surzhikov, F. E.; Dorokhov, V. A. Russ.
Chem. Bull. Int. Ed. 2004, 53, 2319. (b) An alternative and
attractive synthesis of ethyl ester of 2-trifluoromethyl
nicotinic acid via construction of the pyridine ring was
reported after this work was completed: Kiss, L. E.; Ferreira,
H. S.; Learmonth, D. A. Org. Lett. 2008, 10, 1835.
(5) Gilman, H.; Jones, R. G. J. Am. Chem. Soc. 1943, 53, 2319.
(6) (a) Schlosser, M.; Marull, M. Eur. J. Org. Chem. 2003,
1569. (b) Schlosser, M. Angew Chem. Int. Ed. 2006, 45,
5432. (c) Schlosser, M. Synlett 2007, 3096.
(7) For methods developed specifically for CF3-group
installation: (a) Jarvie, J. M. S.; Fitzgerald, W. E.; Janz, G. J.
J. Am. Chem. Soc. 1956, 78, 978. (b) Clark, J. H.;
McClinton, M. A.; Blade, R. J. J. Chem. Soc., Chem.
Commun. 1988, 638. (c) Paratian, J. M.; Sibille, S.;
Perichon, J. J. Chem. Soc., Chem. Commun. 1992, 53.
(d) Cottet, F.; Schlosser, M. Eur. J. Org. Chem. 2002, 327.
(e) Cottet, F.; Marull, M.; Mongin, F.; Espinosa, D.;
Schlosser, M. Synthesis 2004, 1619.
(8) Li, B.; Buzon, R. A.; Castaldi, M. J. Org. Process Res. Dev.
2001, 5, 609.
(9) Mongin, F.; Trécourt, F.; Quéguiner, G. Tetrahedron Lett.
1999, 40, 5483.
(10) An N-oxide intermediate was used for the preparation of 4-
dimethylamino nicotinic acid: (a) Tono-Oka, S. Bull. Chem.
Soc. Jpn. 1982, 55, 1531. (b) Yamada, S.; Misono, T.; Iwai,
Y.; Masumizu, A.; Akiyama, Y. J. Org. Chem. 2006, 71,
6872.
completion by HPLC analysis, the reaction was cooled to
r.t., and the solids (mostly Cs2CO3) were removed by
filtration. The filter cake was rinsed with EtOAc. The filtrate
was concentrated under reduced pressured to give a dark
orange oil. To this was added CH2Cl2 (250 mL), the
resulting mixture was stirred for 10 min, and a slurry was
obtained. The solids were collected by filtration, rinsed with
CH2Cl2 (25 mL), and dried to give 5b (40.6 g, 84%) as a
white solid: mp 180–182 °C (EtOAc). 1H NMR (400 MHz,
MeOH-d4): d = 8.36 (d, 1 H, J = 4.0 Hz), 8.31 (d, 1 H, J = 4.0
Hz), 1.51 (s, 9 H). 13C NMR (100 MHz, MeOH-d4): d =
173.1, 170.4, 153.4, 149.0, 145.8, 144.6 (q, J = 33 Hz),
115.96, 82.7, 28.8. MS (ESI+): m/z = 307.0 [M + 1]+, 251
[M + 1 – t-Bu]+, 206.9 [M + 1 – Boc]+. Anal. Calcd for
C12H13F3N2O4·H2O: C, 44.45; H, 4.66; F, 17.58; N, 8.64.
Found: C, 44.16; H, 4.77; F, 17.28; N, 8.56.
(13) Compound 5b was dissolved in MeOH (150 mL), and HCl
(gaseous, 5.75 g, 158 mmol) was bubbled into the solution.
A white solid began to precipitate in 5 min. MeOH was
removed under reduced pressure to give 1 as a white solid:
mp 214–215 °C (MeOH, dec. observed). 1H NMR (400
MHz, MeOH-d4): d = 7.85 (d, 1 H, J = 7.0 Hz), 6.90 (d, 1 H,
J = 7.0 Hz). 13C NMR (100 MHz, MeOH-d4): d = 166.1,
157.8, 139.3, 136.2 (q, J = 37.5 Hz), 119.2 (q, J = 275.8 Hz),
116.8 (q, J = 2.0 Hz), 112.2. MS (ESI+): m/z = 207.2 [M +
1]+, 189.0 [M + 1 – H2O]+. Anal. Calcd for C7H6ClF3N2O2:
C, 34.66; H, 2.49; Cl, 14.61; F, 23.50; N, 11.55. Found: C,
34.52; H, 2.55; Cl, 14.48; F, 23.36; N, 11.37.
(11) (a) Xu, H.; Wolf, C. Chem. Commun. 2009, 3035.
(b) Roehrscheid, F.; Rapp, J.; Papenfuhs, T. EP 647614,
1995; Chem. Abstr 1995, 122, 290454.
(14) For similar reactions with halo nicotinic acid esters:
(a) Spicer, J. A.; Rewcastle, G. W.; Kaufman, M. D.; Black,
S. L.; Plummer, M. S.; Denny, W. A.; Quin, J. III;
(12) 4-Iodo-2-(trifluoromethyl)nicotinic acid (50 g, 158 mmol),
Boc-amide (22.2 g, 189 mmol), and Cs2CO3 (103 g, 315
mmol) were combined in 2-methyl-2-butanol (500 mL) that
was previously bubbled with dry nitrogen. The reaction flask
was purged four times with nitrogen by applying vacuum to
the flask then flushed with dry nitrogen. Xantphos (2.74 g,
4.73 mmol) and Pd2(dba)3 (2.89 g, 3.15 mmol) were added.
The nitrogen purging sequence was repeated four times. The
reaction was then heated to reflux (104–107 °C internal
temperature) for 2 h. Upon confirmation of reaction
Shahripour, A. B.; Barrett, S. D.; Whitehead, C. E.; Milbank,
J. B. J.; Ohren, J. F.; Gowan, R. C.; Omer, C.; Camp, H. S.;
Esmaeil, N.; Moore, K.; Sebolt-Leopold, J. S.;
Pryzbranowski, S.; Merriman, R. L.; Ortwine, D. F.;
Warmus, J. S.; Flamme, C. M.; Pavlovsky, A. G.; Tecle, H.
J. Med. Chem. 2007, 50, 5090. (b) Collins, I.; Reader, J. C.;
Matthews, T. P.; Cheung, K. M.; Proisy, N.; Williams, D.;
Hugh, K.; Sukhbinder, S.; Scanlon, J. E.; Piton, N.; Addison,
G. J.; Cherry, M. WO 2009044162, 2009; Chem. Abstr.
2009, 150, 398578.
Synlett 2010, No. 14, 2133–2135 © Thieme Stuttgart · New York