An improved synthesis of (3R)-2-(tert-Butoxycarbonyl)-7-hydroxy-1,2,3,4- tetrahydroisoquinoline-3-carboxylic acid

Article abstract of DOI:10.1055/s-2008-1032211

An improved synthesis of (3R)-2-(tert-butoxycarbonyl)-7-hydroxy-1,2,3,4- tetrahydroisoquinoline-3-carboxylic acid is described wherein a modified Pictet-Spengler reaction was employed to provide 95% yield of the product with 7% racemization or less. The enantiomeric excess of the final product was improved to 99.4% via recrystallization. The overall yield of this four-step synthesis provides the title compound in 43% starting from D-tyrosine. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2008-1032211

856
SHORT PAPER
An Improved Synthesis of (3R)-2-(tert-Butoxycarbonyl)-7-hydroxy-1,2,3,4-
tetrahydroisoquinoline-3-carboxylic Acid
(
3
R
)-2-(tert-Buto
o
xycarbonyl)-
n
7-hydroxy-1
g
,2,3,4-tetrahydrois
L
i
rboxyli
u
c
Acid , James B. Thomas, Larry Brieaddy, Bertold Berrang, F. Ivy Carroll*
Organic and Medicinal Chemistry, Research Triangle Institute, Research Triangle Park, NC 27709, USA
Fax +1(919)5418868; E-mail: fic@rti.org
Received 30 October 2007; revised 10 December 2007
The synthesis of 5a using the diiodo intermediate 4a was
Abstract: An improved synthesis of (3R)-2-(tert-butoxycarbonyl)-
7-hydroxy-1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid is de-
scribed wherein a modified Pictet–Spengler reaction was employed
to provide 95% yield of the product with 7% racemization or less.
described to be both higher yielding (55%) and less prone
to racemization (<5%) than the sequence starting from di-
bromo intermediate 4b (30% yield). In our hands, how-
The enantiomeric excess of the final product was improved to ever, the syntheses of 5a from 4b was quite cumbersome,
99.4% via recrystallization. The overall yield of this four-step syn-
forming tar-like intermediates that were very difficult to
thesis provides the title compound in 43% starting from D-tyrosine.
handle as the process was scaled up. The conversion of di-
Key words: Pictet–Spengler, 1,2,3,4-tetrahydroisoquinoline-3-car-
boxylic acid, JDTic, synthesis
bromo compounds 4b into 5b on the other hand produced
easily manipulated solids. This feature prompted us to fo-
cus attention on improving this synthetic route to 2. Bro-
mination of D-tyrosine (3) in acetic acid provided 3¢,5¢-
We recently reported the discovery of (3R)-7-hydroxy-N- dibromo-D-tyrosine hydrobromide (4b·HBr) suitable for
[(1S)-1-{[(3R,4R)-4-(3-hydroxyphenyl)-3,4-dimethylpip- use in the Pictet–Spengler reaction.8 After some experi-
eridin-1-yl]methyl}-2-methylpropyl]-1,2,3,4-tetrahydro-
3-isoquinolinecarboxamide (JDTic, 1) as the first orally 5b·HBr could be improved to 95% versus the 30% report-
active selective opioid receptor antagonist ed earlier by using hydrogen bromide/acetic acid and tri-
mentation, it was found that the conversion of 4b·HBr into
k
(Scheme 1).1–6 The synthesis of JDTic is convergent and fluoroacetic acid as the catalyst/solvent combination. Our
utilizes (3R)-2-(tert-butoxycarbonyl)-7-hydroxy-1,2,3,4- initial trials of the cyclization of 4b·HBr to 5b·HBr at 80
tetrahydroisoquinoline-3-carboxylic acid (2) as one of the °C for 18 hours showed significant racemization (~30%),
key starting materials. In preparing large quantities of 1 but we found that this could be suppressed to 7% by con-
(>200 g) for biological testing we viewed the high cost of ducting the reaction at 55 °C for 72 hours. Attempts to re-
this key starting material 2 as an incentive to improve its move the 7% of S-isomer of 5b directly by
preparation. The results of this investigation are described recrystallization of diastereomeric mixtures of 5b·HBr
herein.
with camphorsulfonic acids were not successful.9 Howev-
er, we found that compound 6 could be purified to 99.4%
ee by catalytic dehydrobromination of 5b·HBr using pal-
ladium-on-carbon and hydrogen to give tetrahydroiso-
Compound 2 has been previously synthesized from D-
tyrosine (3) using the Pictet–Spengler reaction when the
ortho-hydrogen atoms were replaced by bromine or io-
dine to give 4a,b to prevent phenol-formaldehyde poly-
merization (Scheme 2).7 Vershceuren et al. described the
cyclization of both the diiodo and dibromo intermediates
4a,b to the corresponding tetrahydroisoquinolines 5a,b.7
quinolin-3-carboxylic acid
6
followed by two
recrystallizations from water. Protection of 6 with a tert-
butoxycarbonyl group gave 2 in 43% overall yield from D-
tyrosine.
OH
OH
Me
Me
Me
HO
HO
Me
N
+
N
OH
NH
N
N
H
O
Boc
O
NH2
2
JDTic (1)
Scheme 1
SYNTHESIS 2008, No. 6, pp 0856–0858
x
x
.
x
x
.2
0
0
8
Advanced online publication: 28.02.2008
DOI: 10.1055/s-2008-1032211; Art ID: M08107SS
© Georg Thieme Verlag Stuttgart · New York

SHORT PAPER
(3R)-2-(tert-Butoxycarbonyl)-7-hydroxy-1,2,3,4-tetrahydroisoquinoline-3-carboxylic Acid
857
X
X
HO
X
HO
X
HO
a
b
OH
OH
95%
98%
OH
H2N
4a: X = I
N
H2N
H
O
O
O
3
5a: X = I
4b: X = Br
5b: X = Br
HO
HO
c
d
OH
87%
53%
OH
N
H
N
Boc
6
O
2
O
Scheme 2 Reaction conditions: (a) Br2, AcOH; (b) 33% HBr in AcOH, TFA, 55 °C; (c) 10% Pd/C, H2, EtOH; (d) Boc2O, Et3N, DMF.
Determination of the Ratio of Enantiomers; Typical Procedure
The ratio of enantiomers was determined in the following way: Et3N
(0.4 mL) and Boc2O (164 mg) were added sequentially to a soln of
5b·HBr (43 mg) in DMF–H2O (1 mL/0.5 mL) at r.t. The mixture
was stirred for 1 h and diluted with EtOAc. The organic phase was
washed with 20% aq NaHSO4 (1 ×), H2O (2 ×), and brine (1 ×),
dried (Na2SO4), and concentrated. The residue was passed through
silica gel (pipette column) (hexanes–EtOAc, 1:1). Fractions were
collected and concentrated. The residue was diluted with hexanes–
i-PrOH (95:5) to yield a soln (1–2 mg/mL) for HPLC: tR = 76.4 (R-
isomer), 81.8 min (S-isomer).
In conclusion, we have made significant improvements to
the reported synthesis of (3R)-2-(tert-butoxycarbonyl)-7-
hydroxy-1,2,3,4-tetrahydroisoquinoline (2) via modifica-
tion of the Pictet–Spengler conditions together with crys-
tallizations of the key intermediate (3R)-7-hydroxy-
1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid (6). This
provided the title compound in 43% overall yield and
99.4% ee starting from D-tyrosine.
NMR spectra were obtained on a Bruker spectrometer operating at
1
300 MHz for H NMR and 75 MHz for 13C NMR. Mass spectral
(3R)-7-Hydroxy-1,2,3,4-tetrahydroisoquinoline-3-carboxylic
Acid (6)
data were obtained using a Finnegan LCQ electrospray mass spec-
trometer in positive ion mode at atmospheric pressure. Enantiomer
ratios were obtained on a Dynamax HPLC system equipped with a
Chiralcel OD column. Each HPLC sample was eluted with hex-
anes–i-PrOH–TFA (95:5:0.1) at 0.2 mL/min and the absorption was
detected at 220 nm. Optical rotations were obtained on an Autopol
III automatic polarimeter.
EtOH (50 mL), H2O (50 mL), Et3N (12.4 mL, 89.2 mmol), and 10%
Pd/C (0.97 g) were added sequentially to a soln of 5b·HBr (9.7 g,
0.0224 mol) in MeOH (50 mL) at r.t. The resulting mixture was
placed on a Parr hydrogenator at 3.1 bar for 3 h. The mixture was
filtered and the Pd cake was washed with H2O. The pH of the filtrate
was adjusted to 6 by 1 M HCl to give a white precipitate. After stor-
age at 5 °C overnight, the solids were collected by filtration and
washed with cold H2O (2 ×) to yield 6 (3.0 g, 69%) as a white crys-
talline powder; 97.8% ee. The enantiomeric ratio of 6 was measured
as described for the typical procedure for 5b. HPLC: tR = 86.5 (S-
isomer), 90.1 min (R-isomer). Compound 6 (3.0 g) was further re-
crystallized (H2O, 90 mL) to yield 6 (2.3 g, 77%) as a crystalline
solid; 99.4% ee.
3¢,5¢-Dibromo-D-tyrosine Hydrobromide (4b·HBr)
Br2 (173.0 g, 1.08 mol) in AcOH (30 mL) was added dropwise, at
r.t., to a heterogeneous soln of D-tyrosine (3, 100.0 g, 0.55 mol) in
AcOH (300 mL). The mixture was stirred for 4 h, and then concen-
trated on a rotary evaporator. The residue was further dried under
high vacuum to give 4b·HBr (227.1 g, 98%) as an off-white solid.
1H NMR (CD3OD): d = 7.45 (s, 2 H), 4.27 (t, J = 6.6 Hz, 1 H), 3.21
(dd, J = 5.7, 14.2 Hz, 1 H), 3.11 (dd, J = 6.6, 14.2 Hz, 1 H).
[a]D25 +157.5 (c 0.78, AcOH) {Lit.7 [a]D –169.09 (c 1.0, AcOH)}.
1H NMR (DMSO-d6): d = 9.44 (br s, 2 H), 6.96 (d, J = 8.3 Hz, 1 H),
6.66 (dd, J = 2.3, 8.3 Hz, 1 H), 6.59 (d, J = 2.3 Hz, 1 H), 3.52 (dd,
J = 4.9, 10.6 Hz, 1 H), 3.03 (dd, J = 4.5, 16.6 Hz, 1 H), 2.87 (dd,
J = 10.6, 16.6 Hz, 1 H).
(3R)-6,8-Dibromo-7-hydroxy-1,2,3,4-tetrahydroisoquinoline-3-
carboxylic Acid Hydrobromide (5b·HBr)
Compound 4b·HBr (10.0 g, 23.1 mmol) was finely ground with a
mortar and pestle and suspended in TFA (30 mL). 33% HBr in
AcOH (4.4 mL) was added dropwise to the suspension. The vigor-
ous gaseous evolution was allowed to pass through aq NaOH soln
before vented in the hood. Upon the complete addition of HBr,
paraformaldehyde (1.44 g) was added. The mixture was stirred at 55
°C for 72 h, cooled to r.t., and stored in the refrigerator overnight.
The solids were collected by filtration and washed (EtOAc) to give
5b·HBr (9.7 g, 95%) as an off-white solid; 86% ee.
(3R)-2-(tert-Butoxycarbonyl)-7-hydroxy-1,2,3,4-tetrahydroiso-
quinoline-3-carboxylic Acid (2)
Et3N (6.6 mL, 47.6 mmol) and Boc2O (3.9 g, 0.018 mol) were added
sequentially to a soln of 6 (2.3 g, 0.012 mol) in DMF–H2O (4:1, 50
mL) at r.t. After stirring for 4 h, the mixture was concentrated to a
residue that was dissolved in EtOAc (150 mL). The organic phase
was washed with 20% NaHSO4 (1 ×), H2O (2 ×), and brine (1 ×),
dried (Na2SO4), and concentrated. The residue was treated with
hexanes (200 mL) whereupon white precipitates formed. These sol-
ids were collected by filtration and washed (hexanes) to give 2 (3.1
g, 87%) as a white solid; 99.4% ee. The enantiomeric ratio for 2 was
determined as described for the typical procedure for 5b. HPLC:
tR = 86.5 (S-isomer), 90.1 min (R-isomer).
1H NMR (CD3OD): d = 7.53 (s, 1 H), 4.47 (d, J = 16.5 Hz, 1 H),
4.38 (dd, J = 5.1, 10.8 Hz, 1 H), 4.25 (d, J = 16.5 Hz, 1 H), 3.41 (dd,
J = 8.1, 16.8 Hz, 1 H), 3.18 (dd, J = 11.1, 16.8 Hz, 1 H).
Synthesis 2008, No. 6, 856–858 © Thieme Stuttgart · New York

858
C. Liu et al.
SHORT PAPER
25
[a]D –15.0 (c 0.16, MeOH) {Lit.7 S-isomer [a]D +8.2 (c 0.16,
Y.; Cantrell, B. E.; Zimmerman, D. M.; Carroll, F. I. J. Med.
Chem. 2001, 44, 2687.
MeOH)}.
(2) Beardsley, P. M.; Howard, J. L.; Shelton, K. L.; Carroll, F. I.
Psychopharmacology (Berlin) 2005, 183, 118.
(3) Carroll, F. I.; Harris, L. S.; Aceto, M. D. Eur. J. Pharmacol.
2005, 524, 89.
(4) Thomas, J. B.; Fix, S. E.; Rothman, R. B.; Mascarella, S. W.;
Dersch, C. M.; Cantrell, B. E.; Zimmerman, D. M.; Carroll,
F. I. J. Med. Chem. 2004, 47, 1070.
(5) Carroll, F. I.; Thomas, J. B.; Dykstra, L. A.; Granger, A. L.;
Allen, R. M.; Howard, J. L.; Pollard, G. T.; Aceto, M. D.;
Harris, L. S. Eur. J. Pharmacol. 2004, 501, 111.
(6) Thomas, J. B.; Atkinson, R. N.; Vinson, N. A.; Catanzaro, J.
L.; Perretta, C. L.; Fix, S. E.; Mascarella, S. W.; Rothman,
R. B.; Xu, H.; Dersch, C. M.; Cantrell, B. E.; Zimmerman,
D. M.; Carroll, F. I. J. Med. Chem. 2003, 46, 3127.
(7) Verschueren, K.; Toth, G.; Tourwe, D.; Lebl, M.; Van Binst,
G.; Hruby, V. Synthesis 1992, 458.
1H NMR (CD3OD): d = 6.98 (t, J = 5.1 Hz, 1 H), 6.65–6.57 (m, 2
H), 4.96–4.70 (m, 1 H), 4.56–4.43 (m, 2 H), 3.15–3.07 (m, 2 H),
1.53 (s, 4 H), 1.47 (s, 5 H).
13C NMR (CD3OD): d = 175.8, 175.3, 164.1, 157.8, 157.7, 157.3,
136.3, 135.6, 130.8, 130.3, 124.9, 124.4, 115.6, 114.1, 113.9, 82.4,
82.3, 56.4, 54.8, 46.4, 45.7, 32.2, 31.9, 29.2, 29.0, 28.3.
MS (APCI): m/z = 294.3 [M + H]+.
Acknowledgment
This research was supported by the National Institute on Drug
Abuse under Grant DA09045.
References
(8) Vert, M. Eur. Polym. J. 1972, 8, 513.
(9) Shiraiwa, T.; Furukawa, T.; Tsuchida, T.; Sakata, S.;
Sunami, M.; Kurokawa, H. Bull. Chem. Soc. Jpn. 1991, 64,
3729.
(1) Thomas, J. B.; Atkinson, R. N.; Rothman, R. B.; Fix, S. E.;
Mascarella, S. W.; Vinson, N. A.; Xu, H.; Dersch, C. M.; Lu,
Synthesis 2008, No. 6, 856–858 © Thieme Stuttgart · New York