A pictet-spengler cyclisation methodology for the construction of nonproteinogenic tetrahydroisoquinoline quaternary amino acids

Article abstract of DOI:10.1055/s-0031-1289901

A simple highly stereoselective route to tetrahydroisoquinoline quaternary amino acids from phenylalanine is described. Georg Thieme Verlag Stuttgart · New York.

Full text of DOI:10.1055/s-0031-1289901

LETTER
3005
A Pictet–Spengler Cyclisation Methodology for the Construction of
Nonproteinogenic Tetrahydroisoquinoline Quaternary Amino Acids
N
onproteinoge
h
nic
T
etrahyd
i
roisoq
l
uinolin
i
e
Q
uate
p
rnary Amino Acid
C
s
. Bulman Page,*a Genna A. Parkes,b Benjamin R. Buckley,b J. Steven Wailesc
a
School of Chemistry, University of East Anglia, Norwich Research Park, Norfolk, NR4 7TJ, UK
Fax +44(1603)593008; E-mail: p.page@uea.ac.uk
b
c
Department of Chemistry, Loughborough University, Ashby Road, Loughborough, Leicestershire, LE11 3TU, UK
Syngenta, Jealott’s Hill International Research Centre, Bracknell, Berkshire, RG42 6EY, UK
Received 16 September 2011
Our attempts to install the Evans chiral auxiliary, howev-
er, proved to be unsuccessful. Several methods were at-
tempted; for example, treatment of the acid chloride 4
with oxazolidinone 5 failed (Figure 3). Coupling of the
acid using EDCI was also attempted, but the required
product could not be obtained, perhaps as a result of steric
hindrance.
Abstract: A simple highly stereoselective route to tetrahydroiso-
quinoline quaternary amino acids from phenylalanine is described.
Key words: diastereoselective, amino acid, phenylalanine, tetra-
hydroisoquinoline, oxidation, enantiomerically pure
In recent years, the biological importance of a-alkylated
nonproteinogenic amino acids has become apparent, and
this has stimulated an increase in a number of reports in
this area.1 In connection with a number of projects within
our laboratories we required a highly enantioselective
synthetic route for a series of a-alkylated tetrahydroiso-
quinoline a-amino acids 1 (Figure 1).
O
H
Cl
HN
O
N
Cbz
O
5
4
Figure 3
O
R
OH
NH
Seebach has reported a robust methodology for the stereo-
controlled a-alkylation of proline without loss of enantio-
meric purity and without the use of an external chiral
auxiliary, by cyclocondensation with pivaldehyde, fol-
lowed by stereoselective alkylation at the re face.4 Several
other groups have attempted a-alkylation of amino acids
without the use of chiral auxiliaries, but isomeric mixtures
were usually produced.1 Compound 1 (R = Me) has been
prepared by a related route involving a double benzylation
using 1,2-bisbromomethyl benzene.5
1
Figure 1
The use of chiral auxiliaries to induce enantiocontrol dur-
ing synthetic procedures has been well documented,2 and
we initially believed that such an approach would be ideal
to allow us to manipulate the formation of a new chiral
centre by a-alkylation of 2 via a suitably derivatised ana-
logue. Using a Pictet–Spengler approach to generate the
tetrahydroisoquinoline 2 from phenylalanine,3 we envis-
aged that a simple application of Evans’ chiral auxiliary
would provide the stereocontrol we required, by the for-
mation of a single diastereoisomer of each of a series of
a-alkylated tetrahydroisoquinolines 3 (Figure 2).
We therefore anticipated that condensation of the tetrahy-
droisoquinoline 2 with pivaldehyde would similarly form
a single diastereoisomer of the oxazolidinone 6
(Figure 4). This could then be treated with a base to form
the enolate, allowing the stereocontrolled addition of the
alkyl group to give 7, and subsequent deprotection of the
oxazolidinone to give the required a-alkylated tetrahy-
droisoquinolines 1.
O
Several attempts to form the oxazolidine 6, however, us-
ing TFA as catalyst gave only unreacted starting materi-
als. Many other conditions were investigated; for
O
O
N
R
H
R
OH
O
NH
N
2
3
O
O
R
N
H
N
Figure 2
O
O
SYNLETT 2011, No. 20, pp 3005–3007
x
x
.x
x
.2
0
1
1
Advanced online publication: 23.11.2011
DOI: 10.1055/s-0031-1289901; Art ID: D2931ST
6
7
© Georg Thieme Verlag Stuttgart · New York
Figure 4

3006
P. C. B. Page et al.
LETTER
example, both PTSA and CSA were used without any suc-
cess, only unreacted starting material being recovered.
Lewis acids such as BF3·OEt2 and ScOTf3 were also with-
out effect. These reactions were attempted both under
Dean–Stark conditions and over molecular sieve, to no
avail. Microwave irradiation was utilised, but no reaction
was observed. This lack of reactivity appears surprising,
but is perhaps a result of conformational restrictions in the
substrate.
O
O
i, ii
OH
ONa
t-Bu
NH2
N
8
iii
O
O
H
Me
iv
O
O
An alternative approach would involve the stereocon-
trolled a-alkylation prior to the Pictet–Spengler reaction.
Smith has published a highly enantioselective synthesis of
a-alkylated amino acids including derivatives of leucine,
valine, and phenylalanine (8).6 Following the work of
Smith (Scheme 1), the dried sodium salt of phenylalanine
was treated with pivaldehyde under reflux in toluene. The
colourless imine precipitated and was collected, dried,
and treated with allyl chloroformate to induce cyclisation
to the corresponding cis-oxazolidinone 9 with concomi-
tant N-acylation in excellent yield.6 We found that the ad-
dition of a catalytic amount of BF3·OEt2 reduced the
reaction time of this last step from two weeks to three days
while retaining a similar high yield. Diastereoselective
alkylation of 9 by deprotonation with KHMDS in THF at
–78 °C followed by addition of iodomethane furnished
the a-methylated oxazolidinone 10 in excellent yield with
a >20:1 ratio of stereoisomers.7 Hydrolysis of the oxazo-
lidinone afforded the alloc-protected methylated phenyl-
alanine 11 in excellent yield. Conversion into the less
water-soluble methyl ester 12 was accomplished in 86%
yield by treatment with methyl iodide and potassium car-
bonate in DMF at 0 °C. The alloc protecting group was
then removed using catalytic Pd(PPh3)4 in the presence of
dimedone, providing the required amino ester 13 in 91%
yield.6
N
N
O
O
O
t-Bu
t-Bu
O
10
9
v
O
O
O
Me
Me
vi
OMe
OH
O
HN
O
HN
O
vii
11
12
O
Me
OMe
NH2
13
Scheme 1 Reagents and conditions: (i) NaOH, EtOH; (ii) t-
BuCHO, pentane, Dean–Stark, 48 h; (iii) AllocCl, BF3·OEt2, CH2Cl2,
5 °C to r.t., 3 d, 93%; (iv) KHMDS, THF, –78 °C; MeI, 81%; (v) Na-
OH, MeOH, reflux, 3 h, 90%; (vi) K2CO3, MeI, DMF, 86%; (vii)
Pd(PPh3)4, dimedone, THF, 3 d, 91%.
the basis that this procedure would generate a reactive
N-acyliminium species 14 in situ (Scheme 2).9
This reaction was successful, but yields of the tetrahy-
droisoquinoline 15 were low (around 27%). Furthermore,
the alloc group could not then be removed by treatment
with catalytic Pd(PPh3)4 in the presence of dimedone as
before. Upon removal of the solvent medium by evapora-
tion of the aqueous layer following the Pictet–Spengler re-
action, however, a colourless solid was also isolated. We
were delighted to find that this solid was the free a-meth-
ylated tetrahydroisoquinoline carboxylic acid 1 (R = Me),
We then returned to the Pictet–Spengler reaction to form
the required tetrahydroisoquinoline. This step, however,
now proved to be problematic. Treatment of 13 under
standard Pictet–Spengler reaction conditions of paraform-
aldehyde and 1 M aqueous HCl under reflux were unsuc-
cessful, providing only recovered starting material,
although the corresponding carboxylic acid in its racemic
form has been reported to undergo Pictet–Spengler reac-
tion under similar conditions.8 A number of other alde-
hydes, including benzaldehyde, 4-bromobenzaldehyde, 4-
nitrobenzaldehyde, and several acids, including PTSA,
TFA, TiCl4, BF3·OEt2, Sc(OTf)3, along with increased re-
action times, were investigated, but none led to the forma-
tion of tetrahydroisoquinoline. It appears that the
introduction of the new tertiary centre prevents this reac-
tion from taking place, perhaps for conformational rea-
sons, for example the conformation required for
cyclisation being highly disfavoured.
O
O
Me
Me
N+
i
OMe
O
OMe
O
HN
H2C
O
O
12
14
O
O
Me
N
OMe
O
Reasoning that the use of an acylated nitrogen species
would enhance the reactivity of the electrophile, we sub-
jected the alloc-protected a-methylated phenylalanine
methyl ester 12 to Pictet–Spengler reaction conditions, on
15
Scheme 2 Reagents and conditions: (i) HCHO, HCl, reflux, 4 h,
27%.
Synlett 2011, No. 20, 3005–3007 © Thieme Stuttgart · New York

LETTER
Nonproteinogenic Tetrahydroisoquinoline Quaternary Amino Acids
3007
port of The Royal Society (PCBP: Industry Fellowship). We are
indebted to the EPSRC Mass Spectrometry Unit, Swansea.
which could be obtained in 77% yield by increasing the
Pictet–Spengler reaction time from 4 hours to 16 hours.10
It appears that the alloc group, as well as the methyl ester,
is removed by hydrolysis subsequent to cyclisation under
the continued acidic Pictet–Spengler reaction conditions
(Scheme 3).
References and Notes
(1) (a) Lohmar, R.; Steglich, W. Chem. Ber. 1980, 113, 3706.
(b) Stork, G.; Leong, A. Y. W.; Touzin, A. M. J. Org. Chem.
1976, 41, 3491. (c) Fitt, J. J.; Gschwend, H. W. J. Org.
Chem. 1977, 42, 2639. (d) Vevert, J.-P.; Dorsselaer, V. V.;
Kolb, M. J. Org. Chem. 1979, 44, 2732.
(2) (a) Evans, D. A. Aldrichimica Acta 1982, 15, 23. (b) Gage,
J. R.; Evans, D. A. Org. Synth., Coll. Vol. VIII 1993, 339;
Org. Synth. 1990, 68, 83. (c) Evans, D. A.; Vogel, E.;
Nelson, J. V. J. Am. Chem. Soc. 1979, 101, 6120.
(3) Pictet, A.; Spengler, T. Ber. Dtsch. Chem. Ges. 1911, 44,
2030.
O
O
Me
NH
Me
i
OH
OMe
O
HN
O
12
1, R = Me
Scheme 3 Reagents and conditions: (i) HCHO, HCl, reflux, 16 h,
77%.
(4) Seebach, D.; Boes, M.; Naef, R.; Schweizer, W. B. J. Am.
Chem. Soc. 1983, 105, 5390.
(5) Schöllkopf, U.; Hinrichs, R.; Lonsky, R. Angew. Chem. Int.
We have found this sequence also to be successful for to
the introduction of an ethyl group to give 1 (R = Et), again
in good to excellent yields (Scheme 4); in this case, the
Pictet–Spengler reaction was carried out successfully us-
ing the carboxylic acid 16,8 so reducing the length of the
sequence by one step.11 Attempts to alkylate derivatives
of 9 with other electrophiles, for example, benzyl bro-
mide, proved unsuccessful.
Ed. 1987, 26, 143.
(6) Smith, A. B. III.; Guzman, M. C.; Sprengeler, P. A.; Keenan,
T. P.; Holcomb, R. C.; Wood, J. L.; Carrol, P. J.;
Hirschmann, R. J. Am. Chem. Soc. 1994, 116, 9947.
(7) Determined by 1H NMR spectroscopy at 400 MHz. Relative
stereochemistry assigned by analogy with the work of Smith,
established by single-crystal X-ray analysis.
(8) Skiles, J. W.; Suh, J. T.; Williams, B. E.; Menard, P. R.;
Barton, J. N.; Love, B.; Jones, H.; Neiss, E. S.; Schwab, A.;
Mann, W. S.; Khandwala, A.; Wolf, P. S.; Weinryb, I.
J. Med. Chem. 1986, 29, 784.
O
O
H
N
Et
i
(9) Allin, S. M.; James, S. L.; Elsegood, M. R. J.; Martin, W. P.
O
O
J. Org. Chem. 2002, 67, 9464.
(10) (S)-3-Methyl-1,2,3,4-tetrahydroisoquinoline-3-
carboxylic Acid 1 (R = Me, HCl Salt)
N
O
O
O
t-Bu
t-Bu
O
A mixture of a-methylated alloc-protected phenylalanine
methyl ester 16 (2 g, 7.97 mmol) and aq formaldehyde (6
mL) in concd HCl (13.6 mL) was stirred at 100 °C under
reflux for 24 h. The reaction mixture was allowed to reach
r.t. and washed with EtOAc (3 × 50 mL). The aqueous layer
was concentrated under reduced pressure to give a colourless
foam (1.40 g, 77%). IR (film): nmax = 3399, 2927, 2361,
1731, 1632, 1454, 1392, 1177, 1131, 1079, 750 cm–1. 1H
NMR (400 MHz, MeOH): d = 1.72 (br s, 3 H), 3.26 (d, J =
17.0 Hz, 1 H), 3.45 (d, J = 17.0 Hz, 1 H), 4.45 (d, J = 16.8
9
ii
O
O
Et
Et
iii
OH
O
OH
HN
NH
O
Hz, 1 H), 4.56 (d, J = 16.4 Hz, 1 H), 7.26–7.34 (m, 4 H). 13
C
1, R = Et
16
NMR (100 MHz, MeOH): d = 21.4, 39.0, 43.1, 60.7, 127.6,
Scheme 4 Reagents and conditions: (i) KHMDS, THF, –78 °C; EtI,
80%; (ii) NaOH, MeOH, reflux, 3 h, 97%; (iii) HCHO, HCl, reflux,
24 h, 80%.
128.1, 128.7, 128.8, 129.6, 131.0, 173.2. ESI-MS: m/z calcd
20
for C11H14NO2 [M + H]: 192.1025; found: 192.1029. [a]D
–17.9 (c 1.06, MeOH).
(11) (S)-3-Ethyl-1,2,3,4-tetrahydroisoquinoline-3-carboxylic
Acid 1 (R = Et, HCl Salt)
In conclusion, several approaches having surprisingly
failed, including standard chemistry, a synthetic sequence
has been developed capable of delivering the target struc-
tures in good yield.
A mixture of alloc-protected phenylalanine carboxylic acid
derivative 20 (2.0 g, 7.9 mmol) and aq formaldehyde (6 mL)
in concd HCl (13.6 mL) was stirred at 100 °C under reflux
for 24 h. The reaction mixture was allowed to reach r.t. and
washed with EtOAc (3 × 50 mL). The aqueous layer was
concentrated under reduced pressure to give a colourless
foam (1.40 g, 80%). IR (film): nmax = 3302, 2976, 2343,
1707, 1631, 1454, 1398, 1188, 1121, 1059, 758 cm–1. 1H
NMR (400 MHz, MeOH): d = 1.09 (t, J = 8.0 Hz, 3 H), 2.00–
2.34 (m, 3 H), 3.21 (d, J = 17.0 Hz, 1 H), 3.44 (d, J = 17.0
Hz, 1 H), 4.42–4.49 (m, 2 H), 7.25–7.32 (m, 4 H). 13C NMR
(100 MHz, MeOH): d = 10.6, 31.7, 37.6, 45.9, 67.5, 130.0,
130.7, 131.2, 131.4, 131.9, 132.9, 174.6. ESI-MS: m/z calcd
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synlett.
Acknowledgment
This investigation has enjoyed the support of Loughborough Uni-
versity, the EPSRC, and Syngenta. We also acknowledge the sup-
20
for C12H16NO2 [M + H]: 206.1181; found: 206.1185. [a]D
–10.7 (c 1.08, MeOH).
Synlett 2011, No. 20, 3005–3007 © 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.