Macrocycles via Doebner reaction followed by macrolactonization

Article abstract of DOI:10.1055/s-2006-939718

The synthesis of 14 different macrocycles is reported. Upon reacting different formylphenoxyacetic acid derivatives with 5-aminopentanol and a diketo compound, eight different Doebner product intermediates were obtained. These intermediates were then subj

Full text of DOI:10.1055/s-2006-939718

LETTER
1315
Macrocycles via Doebner Reaction Followed by Macrolactonization
Macrocycles via
e
oebner
R
r
Morphochem AG, Gmunder Strasse 37-37A, 81379 München, Germany
Fax +49(89)78005555; E-mail: bernd.henkel@morphochem.de
Received 1 March 2006
Firstly, eight different intermediates were synthesized
using the Doebner reaction.13a–f This multicomponent
reaction uses an amino component and an aldehyde to
form an imine which itself reacts with the enol form of a
Abstract: The synthesis of 14 different macrocycles is reported.
Upon reacting different formylphenoxyacetic acid derivatives with
5-aminopentanol and a diketo compound, eight different Doebner
product intermediates were obtained. These intermediates were
then subjected to cyclization by using polymer-bound carbodiimide pyruvic acid derivative to give a compound which
to give macrolactones. Both the monomers and the dimers could be
cyclizes to tetrahydro-2,3-pyrrolidione upon liberation of
isolated and characterized. In some cases trimers could be seen from
the HPLC-MS. The ratios and distribution of the different-sized
rings are discussed.
an alcohol (Scheme 1).
The reaction was performed as follows:14 The amino com-
ponent and the aldehyde were precondensed followed by
Key words: macrocycles, multicomponent reactions, lactones,
heterocycles, tandem reactions
the addition of the pyruvic acid derivative. After stirring
for 16 hours at room temperature the crude product ob-
tained could be used without purification in some cases.
Details are given in Table 1.
Macrolactones are known to be medically relevant com-
pounds. The antibacterial erythromycin1 and analogues
thereof,2 the immuno suppressant agents Tacrolimus3 and
Sirolimus,4 the antifungal agent amphotericin5 and
nystatin6 and the anticancer drug epothilone7 presently in
the clinical phases are examples.
Table 1 Synthesis Details for the Doebner Products
OH
R
O
O
OH
HO
O
OH
HO
R
O
N
N
R
HO
N
Another interesting system is the tetrahydro-2,4-pyrroli-
diones backbone (tetramic acid) which can be found in
numerous natural compounds like the antifungal
cryptocin,8 the antineoplastic agent tenuazonic acid9 or
ikarugamycin10a–c (Figure 1) which shows a strong
antiprotozoal activity and activity against some Gram-
positive bacteria.11
O
HO
O
O
OH
O
O
HO
Structure C
Structure A
Structure
A
Structure B
No.
1
R
Purification Yield (%)
Yes
71
O
O
H
H
O
2
3
A
A
Yes
62
HO
H
H
Cl
O
N
H
Yes
58
O
N
H
O
Figure 1 Ikarugamycin
4
5
A
A
Yes
Yes
85
51
O
O
Here, we report about the combination of an isoform
(tetrahydro-2,3-pyrrolidione) of the tetrahydro-2,4-pyr-
rolidione structure with a macrolactone thus leading to
potentially medically interesting compounds. This two-
step synthesis concept also combines a multicomponent
reaction with a cyclization method. Some examples are
already given in the literature.12a–c
S
6
A
Yes
49
O
O
7
8
B
C
No
No
82
80
O
O
SYNLETT 2006, No. 9, pp 1315–1318
0
1.
0
6.
2
0
0
6
Advanced online publication: 22.05.2006
DOI: 10.1055/s-2006-939718; Art ID: G08006ST
© Georg Thieme Verlag Stuttgart · New York

1316
B. Henkel et al.
LETTER
O
N
+
HO
NH2
O
OH
O
O
OH
O
HO
O
O
O
OH
O
O
H
OH
O
N
N
O
O
O
OH
O
HO
O
O
– EtOH
O
OH
O
O
O
O
O
N
N
HO
HO
O
O
OH
O
OH
Scheme 1 Mechanism of the Doebner reaction
Compounds 1–8 were cyclized using polymer-bound car- products. Even trimers could be found. The respective
bodiimide.15a,b This Steglich method15c was selected be- yields and distribution of the monomers, dimers and tri-
cause it can be easily performed in parallel fashion and mers are depicted in Table 2.
gives access to the desired compounds in viable yields for
screening. The focus was not on obtaining every single
cyclic compound in the best yield. The polymer-bound
carbodiimide replaced N,N-dicyclohexylcarbodiimide or
N,N-diisopropylcarbodiimide and thus considerably sim-
plified the work-up, for it is difficult to separate off the
The comparison of the rings derived from 4, 7 and 8
shows that the cyclization works best when using the
ortho-substituted precursor (structure A in Table 1) re-
sulting in a 13-membered ring system for the monomer
(major product) and a 26-membered ring system for
the dimer. For compounds 7 and 8 it is different: Due to
steric strain the yields of monomer and dimer decreased
dramatically, the ratio of compounds 15 and 23 being
evolving urea derivatives completely from the reaction
mixture.16 It could be seen from the HPLC-MS that mono-
cycles were formed as well as head-to-tail cyclized
Table 2 Results from the Cyclization
Starting material
No., monomer,a yieldb
9, 60 mg, 25%
10, 49 mg, 22%
11, 53 mg, 26%
12, 44 mg, 26%
13, 21 mg, 9,8%
14, 24 mg, 12%
15, 5 mg, 2.8%
16, n.o.c
No., dimer, yield
17, 28 mg, 5.8%
18, 28 mg, 6.3%
19, 22 mg, 5.4%
20, 22 mg, 6.4%
21, 4 mg, 0.9%
22, 12 mg, 3%
23, 9 mg, 2.5%
24, 15 mg, 4.2%
No., trimer, yield
25, 6 mg, 0.8%
26, 3 mg, 0.4%
27, 7 mg, 1.1%
28, 4 mg, 0.8%
29, n.o.
1
2
3
4
5
6
30, n.o.
7
31, Traces
8
32, Traces
a Isolated quantities in mg.
b Percentage yields (higher molecular weights result in lower yields).
c n.o. = not obtained.
Synlett 2006, No. 9, 1315–1318 © Thieme Stuttgart · New York

LETTER
Macrocycles via Doebner Reaction
1317
O
O
O
O
N
O
O
3
3
N
O
O
O
O
O
O
8
7
16
O
O
N
O
O
O
O
Scheme 2 Monomer and dimer
(7) Höfle, G.; Bedorf, N.; Steinmetz, H.; Schomburg, D.; Gerth,
K.; Reichenbach, H. Angew. Chem., Int. Ed. Engl. 1996, 35,
1567.
(8) Li, J. Y.; Strobel, G.; Harper, J.; Lobkovsky, E.; Clardy, J.
Org. Lett. 2000, 2, 767.
about 1:1. Looking at the cyclization of compound 8 only
the dimer was obtained. The steric strain is caused by the
substitution pattern (ortho, meta, para) of the aldehydes
used for the Doebner reaction. The usual rules for the
formation of medium-sized rings (monomer) therefore
cannot be necessarily applied.17
(9) Stickings, C. E. Biochem. J. 1959, 72, 332.
(10) (a) Ito, S.; Hirata, Y. Tetrahedron Lett. 1972, 13, 1181.
(b) Ito, S.; Hirata, Y. Tetrahedron Lett. 1972, 13, 1185.
(c) Ito, S.; Hirata, Y. Tetrahedron Lett. 1972, 13, 2557.
(11) Jomon, K.; Ajisaka, M.; Sakai, H. J. Antibiot. 1972, 25, 271.
(12) (a) Bughin, C.; Zhao, G.; Bienaymè, H.; Zhu, J. Chem. Eur.
J. 2006, 12, 1174. (b) Cristau, P.; Vors, J.-P.; Zhu, J. Org.
Lett. 2001, 3, 4079. (c) Beck, B.; Larbig, G.; Mejat, B.;
Magnin-Lachaux, M.; Picard, A.; Herdtweck, E.; Dömling,
A. Org. Lett. 2003, 5, 1047.
1
The H NMR signals could be assigned to the respective
protons via NOE experiments. The ortho-monomer ring
systems 9–14 are of a very rigid character. This could be
seen from the fact that the two protons in the CH2 groups
3, 7 and 8 are not equivalent, rotation is hardly possible
(Scheme 2).
In most of the cases, coupling constants could be deter-
mined. For the dimers, splitting-up of signals could only
be clearly seen for signals of the protons of the CH2
groups 3 and 16. No coupling constants could be deter-
mined. In all other cases multiplets were encountered. The
1NMR data of the compounds 15 and 23 are similar to the
(13) (a) Döbner, O. Ber. Dtsch. Chem. Ges. 1887, 20, 277.
(b) Gein, V. L.; Popov, A. V.; Andeichikov, Y. S. Zh. Org.
Khim. 1992, 28, 2774. (c) Andeichikov, Y. S.; Gein, V. L.;
Anikina, I. N. Zh. Org. Khim. 1988, 24, 875.
(d) Andeichikov, Y. S.; Gein, V. L.; Ivanenko, O. I.;
Maslivets, A. N. Zh. Org. Khim. 1986, 22, 2208. (e) Schiff,
R.; Bertini, C. Ber. Dtsch. Chem. Ges. 1897, 30, 601.
(f) Dohrn, M.; Thiele, A. Ber. Dtsch. Chem. Ges. B 1931, 64,
2863.
1
monomers 9–14 and the dimers 17–22. The H NMR
spectrum of compound 24 is different: It shows two
signals for the CH3 groups giving rise to the assumption
that this compound is not of symmetrical structure. This
is supported by the fact that partially a double set of
13C NMR signals exists. The same was found in the
13C NMR spectra for 19 and 22. The gap between the two
adjacent signals was bigger in compound 24 than it was in
19 and 22.
(14) Amine (2 mmol) and aldehyde (2 mmol) were dissolved
in 3 mL of a 1:1 mixture of CH2Cl2 and EtOH. After 30 min
stirring at r.t. pyruvic acid derivative (2 mmol) was added.
The mixture was stirred for 16 h. The solvent was removed
and the remainder taken up in Et2O for solidifying.
Compounds 4, 5 and 6 could be used for the next step
without purification. All other compounds were purified via
preparative HPLC (column Grom-Sil 120 ODS-5, 10 × 4
cm, 10 mm, flow 50 mL/min, gradient 20–100% A in 15 min,
solvent A = MeOH + 0.5% AcOH, solvent B = H2O + 0.5%
AcOH). Yields of 1–8 are given in Table 1.
(15) (a) Buckman, B. O.; Morrissey, M. M.; Mohan, R.
Tetrahedron Lett. 1998, 39, 1487. (b) Keck, G. E.; Sanchez,
C.; Wagner, C. A. Tetrahedron Lett. 2000, 41, 8673.
(c) Neisses, B.; Steglich, W. Angew. Chem., Int. Ed. Engl.
1978, 17, 522.
In conclusion, a simple two-step synthesis strategy for
complex ring systems is described leading to potential
medically interesting compounds of different ring sizes.
While cyclic compounds of the ortho-substituted inter-
mediates can be obtained easily, it is difficult to gain these
compounds from the meta- or para-substituted inter-
mediates in good yields due to steric strain.
(16) Compound 1 (250.8 mg, 0.6 mmol) was dissolved in 5 mL
of anhydrous CH2Cl2 together with N,N-dimethylamino-
pyridine (15.9 mg, 0.13 mmol). Then, N-cyclohexyl-N¢-
methyl polystyrene (Novabiochem, 750 mg, 1.2 mmol) was
added and the mixture agitated overnight at r.t. Resin was
filtered off and thoroughly washed with CH2Cl2. The
combined solvents were evaporated and the remainder was
subjected to preparative HPLC (column Waters SunFire
Prep C18 OBD, 19 × 50 mm, 5 mm, flow 30 mL/min,
gradient 20–90% A in 15 min, solvent A = MeOH + 0.03%
formic acid, solvent B = H2O + 0.03% formic acid). Yield of
9: 60 mg (25%), 17: 28 mg (5.8%), 25: 6 mg (0.8%).
Compound 9: 1H NMR (400 MHz, CDCl3): d = 1.05 (9 H, s,
CH3), 1.10–1.80 (6 H, m, CH2CH2CH2), 2.79–2.88 (1 H, t,
CH of CH2, J = 13.3 Hz), 3.51–3.61 (1 H, t, CH¢ of CH2,
References and Notes
(1) McGuire, J. M.; Bunch, R. L.; Anderson, R. C.; Boaz, H. E.;
Flynn, E. H.; Powell, H. M.; Smith, J. W. Antibiot.
Chemother. 1952, 2, 281.
(2) Katz, L.; Ashley, G. W. Chem. Rev. 2005, 105, 499.
(3) Kino, T.; Hatanaka, H.; Hashimoto, M.; Nishiyama, M.;
Goto, T.; Okuhara, M.; Kohsaka, M.; Aoki, H.; Imanaka, H.
J. Antibiot. (Tokyo) 1987, 40, 1249.
(4) Sehgal, S. N.; Baker, H.; Vezina, C. J. Antibiot. (Tokyo)
1975, 28, 727.
(5) Gold, W.; Stout, H. A.; Pagano, J. F.; Donovick, R. Antibiot.
Ann. 1955-1956, 579.
(6) Hazen, E. L.; Brown, R. Science 1950, 112, 423.
Synlett 2006, No. 9, 1315–1318 © Thieme Stuttgart · New York

1318
B. Henkel et al.
LETTER
J = 13.3 Hz), 4.02–4.14 (2 H, m, CH2), 4.76–4.82 (1 H, d,
CH of CH2, J = 17.2 Hz), 4.92–4.98 (1 H, d, CH¢ of CH2,
J = 17.2 Hz), 6.06 (1 H, s, CH), 6.76–6.88 (3 H, m, Harom),
7.16–7.22 (1 H, d, J = 8.6 Hz, Harom). 13C NMR (100 MHz,
CDCl3): d = 24.3, 25.4, 25.8, 26.1, 40.5, 43.0, 54.2, 64.4,
67.7, 110.2, 118.1, 122.0, 124.8, 127.0, 129.8, 154.5, 155.5,
164.5, 168.9, 202.5.
4.93 (4 H, m, 2 × CH2), 5.83 (2 H, s, 2 × CH), 6.70–6.78 (2
H, m, Harom), 6.85–6.92 (2 H, m, Harom), 6.95–7.03 (2 H, m,
H
arom), 7.17–7.26 (2 H, m, Harom). 13C NMR (100 MHz,
DMSO-d6): d = 22.4, 24.6, 25.3, 27.0, 40.1, 42.5, 54.1, 64.3,
67.3, 110.0, 118.5, 122.1, 125.2, 127.4, 129.9, 156.0, 156.3,
165.0, 168.5, 202.7.
Compound 25: No NMR spectra could be recorded due to
lack of substance.
Compound 17: 1H NMR (400 MHz, DMSO-d6): d = 1.10 (18
H, s, CH3), 1.14–1.68 (12 H, m, 2 × CH2CH2CH2), 3.30–
3.54 (4 H, m, 2 × CH2), 4.06–4.18 (4 H, m, 2 × CH2), 4.83–
(17) Granik, V. G. Russ. Chem. Rev. 1982, 51, 119.
Synlett 2006, No. 9, 1315–1318 © Thieme Stuttgart · New York