A one-pot synthesis of tetrahydroisoquinolin-4-ols via a novel acid-catalyzed rearrangement of 5-aryloxazolidines

Article abstract of DOI:10.1016/j.tetlet.2013.02.087

Benzaldehydes with electron-donating substituents react smoothly with a nonstabilized azomethine ylide derived from sarcosine and formaldehyde to form 5-aryloxazolidines as intermediates, which undergo rearrangement into 2-methyl-1,2,3,4-tetrahydroisoquinolin-4-ols in high yields by simple heating with hydrochloric acid. This one-pot synthesis of tetrahydroisoquinolines can be considered as a formal [3+3] cycloaddition of the azomethine ylide to the aromatic aldehyde.

Full text of DOI:10.1016/j.tetlet.2013.02.087

Accepted Manuscript
A one-pot synthesis of tetrahydroisoquinolin-4-ols via a novel acid-catalyzed
rearrangement of 5-aryloxazolidines
Vladimir S. Moshkin, Vyacheslav Ya. Sosnovskikh
PII:
S0040-4039(13)00348-1
DOI:
http://dx.doi.org/10.1016/j.tetlet.2013.02.087
TETL 42596
Reference:
To appear in:
Tetrahedron Letters
Received Date:
Revised Date:
Accepted Date:
3 January 2013
11 February 2013
27 February 2013
Please cite this article as: Moshkin, V.S., Sosnovskikh, V.Y., A one-pot synthesis of tetrahydroisoquinolin-4-ols
via a novel acid-catalyzed rearrangement of 5-aryloxazolidines, Tetrahedron Letters (2013), doi: http://dx.doi.org/
10.1016/j.tetlet.2013.02.087
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical abstract
________________________________________________________________________________
A one-pot synthesis of tetrahydroisoquinolin-4-ols via a novel acid-catalyzed rearrangement of 5-aryloxazolidines
Vladimir S. Moshkin *, Vyacheslav Ya. Sosnovskikh
________________________________________________________________________________________________________________________
∗ Corresponding author. Fax: +7 343 261 59 78;
e-mail: mvslc@mail.ru

A one-pot synthesis of tetrahydroisoquinolin-4-ols via a novel acid-catalyzed
rearrangement of 5-aryloxazolidines
Vladimir S. Moshkin *, Vyacheslav Ya. Sosnovskikh
Department of Chemistry, Ural Federal University, prosp. Lenina 51, 620000 Ekaterinburg, Russia
A B S T R A C T
Benzaldehydes with electron-donating substituents react smoothly with a nonstabilized azomethine
ylide derived from sarcosine and formaldehyde to form 5-aryloxazolidines as intermediates, which
undergo rearrangement into 2-methyl-1,2,3,4-tetrahydroisoquinolin-4-ols in high yields by simple
heating with hydrochloric acid. This one-pot synthesis of tetrahydroisoquinolines can be considered
as a formal [3+3] cycloaddition of the azomethine ylide to the aromatic aldehyde.
Keywords: 1,2,3,4-Tetrahydroisoquinolin-4-ols; 5-Aryloxazolidines; Nonstabilized azomethine
ylide; [3+2] Cycloaddition; Rearrangement; Pictet–Spengler reaction
The chemistry of 4-substituted-l,2,3,4-tetrahydroisoquinolines has attracted considerable
attention from the synthetic community due to their distribution in Nature and their wide spectrum
of biological activity.1 Examples of such tetrahydroisoquinolin-4-ols include longimammamine (1)
and 4,6-dihydroxy-2-methyl-1,2,3,4-tetrahydroisoquinoline (2) isolated from Dolichothele
longimamma and Theobroma cacao, respectively (Figure 1).2
Figure 1. Examples of natural alkaloids containing a tetrahydroisoquinolin-4-ol framework
Synthetic analogs of tetrahydroisoquinolinols 1 and 2 have been proposed as potential
antihypertensive drugs.3a,b Recently, a Lewis acid catalyzed arylation of tetrahydroisoquinolin-4-ols
2

with thioanisole was reported.3c This reaction opens a new approach to the synthesis of 4-
aryltetrahydroisoquinolines, which can function as serotonin–norepinephrine–dopamine reuptake
inhibitors.3d–f Taking into account other methods of functionalization, the tetrahydroisoquinolin-4-ol
core is a valuable building block for the construction of a wide range of isoquinoline derivatives.4a–i
Due to the important applications of this class of compounds, their synthesis has been
extensively studied.4 There are two classical methods for the preparation of tetrahydroisoquinolin-4-
ols. The first is the Pictet–Spengler cyclization of hydroxyphenethylamines 3, which are not easily
accessible,4a–d and the second is the Pomeranz–Fritsch–Bobbitt acid-catalyzed cyclization of N-
benzyl-N-methylaminoacetaldehyde diethyl acetals 4.4d–l Both reactions involve multistep
procedures and require electron-donating substituents on the aromatic ring of the substrate. The
second method is often more efficient due to the fewer number of steps involved from the starting
benzaldehydes (Scheme 1).
Scheme 1. The classical syntheses of tetrahydroisoquinolin-4-ols
On the other hand, it is well-known that nonstabilized azomethine ylides generated from
sarcosine (5) and formaldehyde,5a,b or silyl precursor 6 in the presence of trifluoroacetic acid5c,d add
3

to the carbonyl group of benzaldehydes and benzophenone5e to form 5-aryloxazolidines 7. Their
methylene group, the former azomethine ylide cationic center, still has electrophilic character after
the cycloaddition step due to the presence of two geminal acceptor atoms and the easy cleavage of
the C–O acetal bond. It would be reasonable to assume the possibility of the Pictet-Spengler
intramolecular recyclization with the formation of the tetrahydroisoquinoline core 8 (Scheme 2). In
general, these two successive reactions can be considered as a formal [3+3] cycloaddition of the
nonstabilized azomethine ylide with the aromatic aldehyde or ketone, followed by proton migration.
In this context, and in connection with our interest in the development of azomethine ylide
chemistry,6 a method based on a 1,3-dipolar cycloaddition followed by an acid-catalyzed
rearrangement of the initial 5-aryloxazolidine 7 into tetrahydroisoquinoline system 8, in a one-pot
procedure, attracted our attention. To the best of our knowledge, no such transformation has been
reported.
Bn
CO2H
NH
N
SiMe3
OMe
Me
6
5
H+
CH2O
O
O
N
R'
[3+2]
R
N
R
+
R'
7
?
formal [3+3]
OH
R
N
R'
8
R = H, Ph; R' = Me, Bn
Scheme 2. Cycloaddition of aromatic aldehydes and ketones with azomethine ylides
To test the feasibility of the envisioned idea, the reaction of m-anisaldehyde with the
nonstabilized azomethine ylide derived from sarcosine and formaldehyde was first investigated. We
found that refluxing m-anisaldehyde, sarcosine and paraformaldehyde in benzene for six hours with
azeotropic removal of water resulted in the formation of previously unknown 5-(3-methoxyphenyl)-
4

3-methyloxazolidine (7а) (yield 97%). The crude product was sufficiently pure according to the 1H
NMR spectrum and only a trace (4%) of the starting aldehyde was observed. To our delight, simple
heating of oxazolidine 7а in 6 M HCl at 60 °C resulted in full recyclization into the desired product,
6-methoxy-2-methyl-1,2,3,4-tetrahydroisoquinolin-4-ol (8a).7,8 Subsequent basification of the
reaction mixture and extraction with ether gave the rearranged product 8a as a white solid in 89%
isolated yield, based on the starting m-anisaldehyde (Scheme 3). Apparently, this intramolecular
reaction was mediated by HCl, where protonation of the oxygen atom facilitates breakdown of the
oxazolidine ring to give cationic intermediate A, which acts as an internal electrophile.
Scheme 3. One-pot synthesis of the tetrahydroisoquinolin-4-ol 8a
With optimal conditions for the one-pot synthesis of tetrahydroisoquinolinol 8a in hand,
reactions with other aromatic aldehydes bearing electron-donating substituents were investigated.
As can be seen from Table 1, the reactions of veratraldehyde, 3,4,5-trimethoxybenzaldehyde,9 O-
benzyl vanillin, and 3,4-methylendioxybenzaldehyde proceeded smoothly to afford the desired
products 8b–e in 62–87% yields. Tetrahydroisoquinoline 8b has been reported earlier.4e–g Using this
approach we were also able to obtain 5-methyl-4,5,6,7-tetrahydrothieno[3,2-c]pyridin-7-ol (8f)
from thiophene-2-carbaldehyde in 63% yield under the same conditions, which may be of interest
as a therapeutically valuable compound.10 However, all our attempts to rearrange the oxazolidines
obtained from benzaldehyde and anisaldehyde failed (the former at a higher temperature led to
partial hydrolysis of the oxazolidine ring with extrusion of formaldehyde, while the latter gave an
insoluble dark gum). The lack of reactivity in these cases was apparently due to the low
5

nucleophilicity of the benzene ring. These results clearly show that at least one electron-donating
meta-substituent must be present on the aromatic aldehyde for the successful electrophilic
aryloxazolidine–tetrahydroisoquinoline rearrangement. The structures of products 8a–f were
characterized by IR, 1H, and 13C NMR spectral data as well as by elemental analysis.
Table 1. Yields and melting points of the products 8a–f
Entry
Aromatic aldehyde
Tetrahydroisoquinolinol 8
Yield (%)
Mp (°C)
1
89
84–86a
2
3
62
87
124–126b
107–109c
4
5
6
71
83
63
127–130
87–89d
79–81e
O
O
O
Me
N
S
OH
8f
a This compound was mentioned in ref. 4c without any data. b Mp 127–128 °C (ref. 4e,f), mp 124–126 °C (ref. 4g). c Mp
of the hydrochloride 189–191 °C (ref. 4b). d This compound was reported in ref. 4k without any data, only the mp 106–
108 °C was described. e This compound was reported in ref. 10a without any data, only the mp 145 °C of the maleate
was given.
6

Compared with the Pictet-Spengler cyclization of the corresponding hydroxyphenethylamines,
this reaction proceeds with high regioselectivity. No 8-methoxy regioisomer was observed.2a Also it
should be noted that the preparation of N-benzyl-N-methylaminoacetaldehyde diethyl acetal from
m-anisaldehyde followed by Pomeranz–Fritsch–Bobbitt cyclization leads to the 7-
methoxytetrahydroisoquinoline, while the reaction described here gives the 6-methoxy isomer and
is a good addition to the classical method. Moreover, unlike both previously known approaches for
the synthesis of tetrahydroisoquinolines from aromatic aldehydes, this process does not require any
reducing agent or chromatographic purification of the intermediate liquid oxazolidines, and thereby
greatly facilitates the preparation of the target tetrahydroisoquinolinols.
In conclusion, we have developed a simple and convenient one-pot method for the synthesis of
2-methyl-1,2,3,4-tetrahydroisoquinolin-4-ols from aromatic aldehydes bearing electron-donating
substituents and an azomethine ylide derived from sarcosine and formaldehyde, via a novel
aryloxazolidine–tetrahydroisoquinoline rearrangement. Compared with the previously known
procedures, our method shows several advantages, the main of which are simplicity, efficiency and
easy availability of the starting materials. Further application of this reaction for the construction of
substituted tetrahydroisoquinolines and their heteroanalogs is underway in our laboratory and will
be reported in due course.
Acknowledgment
This work was supported financially by the RFBR (Grant 12-03-31036).
References and notes
1. (a) Chemistry of Heterocyclic Compounds: Isoquinolines, ed. Taylor, E. C.; Weissberger, A.;
Grethe, G., John Wiley & Sons, New York, 1981, Vol. 38; (b) Rozwadowska, M. D.
Heterocycles 1994, 39, 903–931; (c) Charifson, P. S. Drugs Fut. 1989, 14, 1179–1185.
2. (a) Ranieri, R. L.; McLaughlin, J. L. J. Org. Chem. 1976, 41, 319–323; (b) Ranieri, R. L.;
McLaughlin, J. L. Lloydia 1977, 40, 173–177; (c) Kenyhercz, T. M.; Kissinger, P. T. Lloydia
1978, 41, 130–139.
7

3. (a) Grethe, G.; Lee, H. L.; Uskokovic, M. R. US Patent 3772304, 1973; Chem. Abstr. 1974, 80,
59880; (b) Grethe, G.; Lee, H. L.; Uskokovic, M. R. US Patent 3629265, 1971; Chem. Abstr.
1972, 76, 113088; (c) Deng, X.; Liang, J. T.; Liu, J.; McAllister, H.; Schubert, C.; Mani, N. S.
Org. Process Res. Dev. 2007, 11, 1043–1050; (d) Rheiner, A. DE Patent 2062001, 1971; Chem.
Abstr. 1971, 75, 129683; (e) Molino, B. F.; Berkowitz, B.; Cohen, M. US Patent 20060111394,
2006; Chem. Abstr. 2006, 144, 488538; (f) Molino, B. F.; Liu, S.; Berkowitz, B. A.; Guzzo, P.;
Beck, J. P.; Cohen, M. WO Patent 2006020049; Chem. Abstr. 2006, 144, 254149.
4. (a) Kondo, T.; Tanaka, S. J. Pharm. Soc. Jpn. 1930, 119–121; (b) Friedman, O. M.;
Parameswaran, K. N.; Burstein, S. J. Med. Chem. 1963, 6, 227–229; (c) Simig, G; Schlosser, M.
Tetrahedron Lett. 1994, 35, 3081–3082; (d) Quessy, S. N.; Williams, L. R. Aust. J. Chem. 1979,
32, 1317–1327; (e) Lesèche, B.; Gilbert, J.; Viel, C. J. Heterocycl. Chem. 1981, 18, 143–153; (f)
Umezawa, B.; Hoshino, O.; Terayama, Y. Chem. Pharm. Bull. 1968, 16, 180–182; (g) Moreno,
I.; SanMartin, R.; Tellitu, I.; Domínguez, E. Heterocycles 1999, 51, 2311–2320; (h) Dyke, S. F.;
Bather, P. A.; Garry, A. B.; Wiggins, D. W. Tetrahedron 1973, 29, 3881–3888; (i) Morita, Y.;
Imaki, N.; Takayanagi, H.; Shirasaka, T.; Shimpuku, T.; Takuma, Y.; Oishi, M. EP Patent
214905, 1987; Chem. Abstr. 1987, 106, 213780; (j) Morita, Y.; Imaki, N.; Takayanagi, H.;
Shirasaka, T.; Shimpuku, T.; Takuma, Y.; Oishi, M. EP Patent 214051, 1987; Chem. Abstr.
1987, 106, 196434; (k) Schlosser, M.; Simig, G.; Geneste, H. Tetrahedron 1998, 54, 9023–
9032; (l) Simig, G.; Schlosser, M. Synlett 1990, 50–51; (m) Anakabe, E.; Vicario, J. L.; Badía,
D.; Carrillo, L.; Yoldi, V. Eur. J. Org. Chem. 2001, 4343–4352; (n) Vicario, J. L.; Badía, D.;
Carrillo, L.; Anakabe, E. Tetrahedron: Asymmetry 2003, 14, 347–353.
5. (a) Nyerges, M.; Fejes, I.; Virányi, A.; Groundwater, P. W.; Töke, L. Synthesis 2001, 1479–
1482; (b) Tóth, J.; Blaskó, G.; Dancsó, A.; Tőke, L.; Nyerges, M. Synth. Commun. 2006, 36,
3581–3589; (c) Ryan, J. H.; Spiccia, N.; Wong, L. S.-M.; Holmes, A. B. Aust. J. Chem. 2007,
60, 898–904; (d) D’Souza, A. M.; Spiccia, N.; Basutto, J.; Jokisz, P.; Wong, L. S.-M.; Meyer,
8

A. G.; Holmes, A. B.; White, J. M.; Ryan, J. H. Org. Lett. 2011, 13, 486–489; (e) Padwa, A.;
Dent, W. J. Org. Chem. 1987, 52, 235–244.
6. (a) Moshkin, V. S.; Sosnovskikh, V. Y.; Slepukhin, P. A., Röschenthaler, G.-V. Mendeleev
Commun. 2012, 22, 29–31; (b) Moshkin, V. S.; Sosnovskikh, V. Y.; Röschenthaler, G.-V.
Tetrahedron Lett. 2012, 53, 3568–3572.
7. General procedure. A stirred mixture of the substituted benzaldehyde (1.0 mmol), finely
ground sarcosine (0.27 g, 2.0 mmol) and paraformaldehyde (0.10 g, 4.0 mmol) was refluxed in
dry benzene (3.3 mL) with removal of the water formed by means of a Dean-Stark trap.
Refluxing was continued for 5–8 h. The resulting solution was evaporated in vacuo to give the
intermediate 5-aryl-3-methyloxazolidine as an oily residue. This was dissolved in 6 M HCl with
heating at 60 °C and then left overnight at room temperature. The mixture was extracted with
Et2O (2 × 3 mL) followed by basification with an excess of a cold concentrated solution of
NaOH. The crude tetrahydroisoquinolin-4-ol 8 was extracted into Et2O or CH2Cl2, dried over
Na2SO4, and evaporated to give the pure solid product.
8. 6-Methoxy-2-methyl-1,2,3,4-tetrahydroisoquinolin-4-ol (8a). White powder, yield 89%, mp 84–
86 °C (8a was previously reported in ref. 4c without any data). IR: 3300–3000, 2944, 2836,
2796, 1610, 1586, 1500, 1439, 1371, 1284, 1169, 1135, 1054, 1030, 888, 811, 735, 488 cm–1. 1H
NMR (400 MHz, CDCl3) δ 2.35 (s, 3H, MeN), 2.48 (dd, J = 11.8, 2.8 Hz, 1H, 3-CHH), 2.92
(ddd, J = 11.8, 2.8, 1.0 Hz, 1H, 3-CHH), 3.10 (d, J = 14.5 Hz, 1H, 1-CHH), 3.33 (d, J = 14.5
Hz, 1H, 1-CHH), 3.80 (s, 3H, MeO), 3.8–4.1 (br s, 1H, OH), 4.54 (t, J = 2.8 Hz, 1H, H-4), 6.80
1
(dd, J = 8.5, 2.6 Hz, 1H, H-7), 6.87 (d, J = 8.5 Hz, 1H, H-8), 6.94 (d, J = 2.6 Hz, 1H, H-5). H
NMR (400 MHz, DMSO-d6) δ 2.26 (dd, J = 10.9, 7.8 Hz, 1H, 3-CHH), 2.31 (s, 3H, MeN), 2.84
(ddd, J = 10.9, 5.4, 0.8 Hz, 1H, 3-CHH), 3.25 (d, J = 14.4 Hz, 1H, 1-CHH), 3.50 (d, J = 14.4
Hz, 1H, 1-CHH), 3.72 (s, 3H, MeO), 4.59 (q, J = 6.5 Hz, 1H, H-4), 5.25 (d, J = 6.9 Hz, 1H,
OH), 6.74 (dd, J = 8.4, 2.6 Hz, 1H, H-7), 6.95 (d, J = 8.4 Hz, 1H, H-8), 6.99 (d, J = 2.6 Hz, 1H,
H-5). 13C NMR (126 MHz, CDCl3) δ 45.9, 55.2, 56.5, 60.4, 66.7, 112.7, 114.5, 126.5, 127.3,
9

137.1, 158.1. Anal. Calcd for C11H15NO2: C, 68.37; H, 7.82; N, 7.25. Found: C, 68.03; H, 7.82;
N, 7.10.
9. 6,7,8-Trimethoxy-2-methyl-1,2,3,4-tetrahydroisoquinolin-4-ol (8с). Light yellow powder, yield
87%, mp 107–109 °C (8c was previously reported in ref. 4b as the hydrochloride). IR: 3300–
3020, 2938, 2834, 2794, 1602, 1451, 1342, 1117, 1077, 1056, 1038, 1022, 919, 900, 859, 785,
1
580 cm–1. H NMR (400 MHz, CDCl3) δ 2.47 (s, 3H, MeN), 2.53 (dd, J = 11.6, 2.6 Hz, 1H, 3-
CHH), 2.8–3.0 (br s, 1H, OH), 2.90 (ddd, J = 11.6, 3.2, 1.0 Hz, 1H, 3-CHH), 3.10 (d, J = 15.4
Hz, 1H, 1-CHH), 3.80 (d, J = 15.4 Hz, 1H, 1-CHH), 3.85, 3.86, 3.87 (all s, 3H, MeO), 4.51 (t, J
= 2.7 Hz, 1H, H-4), 6.73 (s, 1H, H-5). 13C NMR (126 MHz, CDCl3) δ 46.0, 52.8, 55.9, 60.1,
60.4, 60.7, 67.1, 107.4, 121.0, 131.9, 141.2, 149.3, 152.3. Anal. Calcd for C13H19NO4: C, 61.64;
H, 7.56; N, 5.53. Found: C, 61.24; H, 7.60; N, 5.49.
10. (a) Maffrand, J.-P. DE Patent 2623567, 1976; Chem. Abstr. 1977, 86, 89789; (b) Matsuoka, M.;
Oyama, T. WO Patent 2008120761; Chem. Abstr. 2008, 149, 425917.
10