An efficient regioselective synthesis of endocrocin and structural related natural anthraquinones starting from emodin

Article abstract of DOI:10.1016/j.tetlet.2005.02.061

Endocrocin and related naturally occurring anthraquinone pigments like cinnalutein could be synthesized regioselectively via a Marschalk type reaction, starting from the natural hydroxy anthraquinone emodin. Furthermore, the new tri-O-methyl protected emodin-2-carbaldehyde may serve as a promising synthon for new bathochromically shifted, higher generation photodynamically active hypericin derivatives.

Full text of DOI:10.1016/j.tetlet.2005.02.061

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 2377–2380
An efficient regioselective synthesis of endocrocin and
structural related natural anthraquinones starting from emodin
Mario Waser, Bernd Lackner, Joachim Zuschrader, Norbert M u¨ ller and Heinz Falk*
Institute of Organic Chemistry, Johannes Kepler University Linz, Altenbergerstr. 69, 4040 Linz, Austria
Received 28 December 2004; revised 7 February 2005; accepted 14 February 2005
Abstract—Endocrocin and related naturally occurring anthraquinone pigments like cinnalutein could be synthesized regioselectively
via a Marschalk type reaction, starting from the natural hydroxy anthraquinone emodin. Furthermore, the new tri-O-methyl
protected emodin-2-carbaldehyde may serve as a promising synthon for new bathochromically shifted, higher generation photo-
dynamically active hypericin derivatives.
Ó 2005 Elsevier Ltd. All rights reserved.
1
. Introduction
selectively substitute the emodin moiety in position 2,
affording not only the carboxylic acid 2 in a six step syn-
thesis, but also other structurally related compounds
like cinnalutein (1,8-dihydroxy-3-methyl-6-methoxy-9,10-
anthraquinone-2-carboxylicacid) (3) and 1,6,8-trimethoxy-
2-formyl-3-methyl-9,10-anthraquinone (4). The latter
is of particular interest in our search for potential
precursors for a new generation of photodynamically
The hydroxylated anthraquinones emodin (1,6,8-tri-
1
hydroxy-3-methyl-9,10-anthraquinone) (1) and endo-
crocin (1,6,8-trihydroxy-3-methyl-9,10-anthraquinone-
2-carboxylic-acid) (2) are natural pigments mainly
2
occurring in fungi and lichens and are biosynthesized
on a polyketide pathway. Contrary to previous
4
assumptions, 2 is not the biosynthetic precursor of 1.
3
5
9
active hypericin derivatives. It might serve as a promis-
It has also been found, that decarboxylation of similar
anthraquinone carboxylic acids by conventional means
is hardly possible on the stage of the anthraquinone
ing synthon for new 9,12-substituted hypericin deriva-
tives with a bathochromically shifted long wavelength
absorption overcoming dimerization problems which
occurred in the case of 3-stilbenoid-substituted emodin
6
but on the corresponding anthrone (see Fig. 1).
1
0
derivatives.
Synthesis pathways for 2 include total synthesis of the
7
anthraquinone skeleton via Friedel–Crafts acylations
as well as rather tedious modifications of 2,3-dimethyl-
anthraquinones. We have now found a way to regio-
2. Results and discussion
8
The main challenge in this work was to regioselectively
substitute the emodin skeleton in position 2in a simple
1
1
and reliable way. We have recently shown, that substi-
tution of this position via an intramolecular Friedel–
Crafts acylation is possible in principle, but with an
unsatisfying regioselectivity. Thus, conventional electro-
philic substitutions like Friedel–Crafts and Gattermann
type reactions seem to be not the methods of choice, not
only because of the missing regioselectivity but also be-
cause of the severe deactivation of the anthraquinone
system with respect to an electrophilic attack.
Figure 1. Structures of emodin (1) and endocrocin (2).
Keywords: Marschalk reaction; Emodin; Endocrin; Cinnalutein;
Regioselectivity; Anthraquinonoes; Natural compounds.
*
Corresponding author. Tel.: +43 732 24 68 8748; fax: +43 732 24 68
747; e-mail addresses: heinz.falk@jku.at; mario.waser@jku.at
Our approach was to introduce a hydroxymethyl group
1
via a Marschalk type reaction, followed by subsequent
2
8
0
040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.02.061

2378
M. Waser et al. / Tetrahedron Letters 46 (2005) 2377–2380
oxidation to the aldehyde and eventually to the carbox-
ylic acid. Marschalk type reactions are quite common in
methyl-3-methyl-9,10-anthraquinone (8) in 90% yield.
Subsequent oxidation of 8 with PCC gave the aldehyde
4 in 87% yield. Accordingly, the promising synthon 4
could be obtained from 1 in a five step synthesis with
an overall yield of 45%.
1
3
the syntheses of anthracyclinone type antibiotics, as
the electrophilic attack always occurs ortho to a phenolic
group. Accordingly, it was necessary to selectively pro-
tect the hydroxyl groups in position 6 and 8. We there-
fore converted 1 to tri-O-methyl emodin (5) by
microwave assisted synthesis (cf. Ref. 14; 98%), followed
by a selective ether cleavage, using an improved proce-
Tri-O-methyl endocrocin (9) could be obtained via a
Jones oxidation of 8 in 78% yield. Due to the fact, that
not only the non-methylated 2, but also the monomethyl
ether 3 and the dimethyl ether 10 (1-hydroxy-6,8-di-
methoxy-3-methyl-9,10-anthraquinone-2-carboxylicacid)
1
5
dure of HassallÕs deprotection (starting at À5 °C
instead of À70 °C increases the yield) to obtain the 1-
hydroxy-6,8-dimethoxy-3-methyl-9,10-anthraquinone (6)
in 93% yield. Due to the high reactivity of the interme-
2
are interesting naturally occurring pigments, we also
investigated the selective deprotection of 9 to 3 and 10.
According to the procedure used for the synthesis of 6,
a BBr3 mediated deprotection of 9 gave the mono-
deprotected 10 in 72% yield, whereas a HBr/AcOH med-
iated deprotection selectively yielded the dideprotected
compound cinnalutein (3) in 85% yield (35% overall
yield).
1
3
diate ortho-quinone methide hydroxymethylation of
required a rather short reaction time (40 min) and
6
low temperature (0 °C). After purification and separa-
tion of by-products by column chromatography 6,8-
dimethoxy-1-hydroxy-2-hydroxymethyl-3-methyl-9,10-
anthraquinone (7) was obtained in 63% yield. Benzylic
oxidation of 7 by means of pyridiniumchlorochromate
7
,15,16
(
PCC) or CrO to the corresponding mono-deprotected
3
The few published procedures,
describing the di-
aldehyde and carboxylic acid proceeded with insufficient
selectivity and rather low yields, especially in the case of
the carboxylic acid, which might be due to the unpro-
tected phenolic group. Furthermore, with respect to
the aldehyde as a suitable starting point for extensions
of the chromophoric system, a protection of the hydro-
xyl group seems to be necessary anyway.
rect total deprotection of methyl ether protected 1,6,8-
hydroxylated anthraquinones, give either rather low
yields, or lead to a decarboxylation of carboxylic acid
derivatives. This is also in accordance with our experi-
ence. Thus we always observed either a destruction of
the compound under rather harsh conditions (e.g., HI
1
6
in boiling AcOH, AlCl /NaCl melt), or only a didepro-
3
tection, leaving the 6-O-methyl ether untouched, under
7
Thus, a selective etherification of the phenolic hydroxyl
group leaving the primary alcohol untouched was car-
ried out affording the 1,6,8-trimethoxy-2-hydroxy-
ÔsofterÕ conditions (e.g., HBr). Steglich and Reininger
described the total O-demethylation of 10 by means of
BBr . In our case, treatment of 9 with an excess of
3
Scheme 1. Reagents and conditions: (a) cf. Ref. 14: Me
À5 to 2 5°C, 1.5 h, 93%; (c) Na /CH O (37%)/MeOH/NaOH (1 N), 0 °C, 40 min, subsequent H
reflux, 16 h, 90%; (e) PCC/CH , 2h, 87%; (f) CrO /H SO
/acetone, 0–10 °C, 1.5 h, 78%; (g) HBr/AcOH, reflux, 30 min, 85%; (h) BBr
to 5 °C, 0.5 h, 72%; (i) BBr /CH Cl , reflux, 10 h, 78%.
2
SO
4
/K
2
CO
3
/tetrabutylammonium bromide, 600 W (75 °C), 20 min, 98%; (b) BBr
oxidation, 63%; (d) Me SO /K CO
/CH
3
/CH
/acetone,
Cl
, À5
2
Cl
2
,
2
S
2
O
4
2
O
2 2
2
4
2
3
2
Cl
2
3
2
4
3
2
2
3
2
2

M. Waser et al. / Tetrahedron Letters 46 (2005) 2377–2380
2379
BBr in boiling CH Cl for 10 h was the only possibility
3
17. Steglich, W.; Reininger, W. Chem. Ber. 1972, 105, 2922–
927.
8. Selected properties of compound 2: Mp: decomp P
2
2
2
found to obtain 2 in a reasonable yield of 78%. Accord-
ingly, endocrocin (2) could be obtained from emodin (1)
in a six step synthesis with an overall yield of 32% (see
Scheme 1).
1
1
3
10 °C. H NMR (500 MHz, d, DMSO-d
s, 3H, ar-CH ), 6.64 (d, 1H, J = 2.14 Hz, ar-H7), 7.17
d, 1H, J = 2.14 Hz, ar-H5), 7.60 (s, 1H, ar-H4), 11.44 (s,
6
, 30 °C): 2.38
(
(
3
1
1
d
H, 3-OH), 11.92(br s, 1H, COOH), 11.95 (s, 1H, 8-OH),
2.37 (s, 1H, 1-OH) ppm. C NMR (125 MHz, d, DMSO-
In conclusion, we found an efficient way to regioselec-
tively substitute emodin in position 2, yielding endocro-
cin-like naturally occurring pigments as well as the
promising hypericin precursor 4 in satisfying overall
yields, with a Marschalk type reaction as the key step.
All compounds were fully characterized on basis of their
IR, UV/vis, MS, and NMR spectra, particularly by 2D
NMR measurements including HSQC, HMBC, and
13
6
, 30 °C): 19.4 (ar-CH
3
), 107.9 (C7), 108.9 (C5), 109.1
(C8a), 114.9 (C9a), 120.5 (C4), 129.1 (C2), 130.5 (C4a),
144.0 (C3), 157.9 (C1), 164.5 (C8), 165.7 (C6), 181.8 (C10)
ppm, C9, C10a, and –COOH not observed due to
insufficient solubility. ESI-MS (neg. ion mode): m/z = 313
(
À
[MÀH] ). IR (KBr): 3445, 2954, 2924, 2854, 1740, 1713,
À1
1
463, 1377, 1262, 1208, 1071, 801, 722 cm . UV–vis
1
8
(CHCl ): kmax = 242 (100), 284 (69), 443 (20) nm (rel. int.);
3
NOESY experiments. Compound 5 displayed spectro-
scopic data in accordance to Ref. 14. Melting points of
1
compound 3: Mp: 275–278 °C. H NMR (500 MHz, d,
DMSO-d , 30 °C): 2.42 (s, 3H, ar-CH ), 3.94 (s, 3H, 6-
6
3
7
,15,17
2
, 3, 6, and 10 were according to literature.
3
OCH ), 6.89 (d, 1H, J = 2.20 Hz, ar-H7), 7.21 (d, 1H,
J = 2.20 Hz, ar-H5), 7.57 (s, 1H, ar-H4), 12.14 (s,1H, –
OH), 12.29 (br s, 2H, –OH and –COOH) ppm. C NMR
1
3
Acknowledgements
(
125 MHz, d, DMSO-d , 30 °C): 19.7 (ar-CH ), 56.4 (6-
6
3
3
OCH ), 106.6 (C7), 107.7 (C5), 110.0 (C8a), 114.0 (C9a),
This work was supported by the Austrian Science Fund
FWF), project P16969. The cryogenic 500 MHz probe
1
1
20.5 (C4), 130.4 (C2 or C3), 132.5 (C4a), 134.8 (C10a),
43.9 (C2or C3), 164.4 (C1 and C8), 166.2(C6), 167.0
(
used was purchased from FWF project P15380 (project
leader: Professor Dr. Norbert M u¨ ller). We are grateful
to Professor Dr. Christian Klampfl and DI Werner Hu-
ber for recording of mass spectra.
(–COOH), 180.9 (C10), 189.6 (C9) ppm. ESI-MS (neg. ion
mode): m/z = 327 ([MÀH] ). IR (KBr): 2924, 2854, 1709,
À
1675, 1601, 1465, 1394, 1285, 1244, 1216, 1174, 1100, 958,
7
À1
66 cm . UV–vis (CHCl ): k
= 268 (93), 290 (100), 442
max
3
1
(
(
40) nm (rel. int.); compound 4: Mp: 246–248 °C. H NMR
500 MHz, d, DMSO-d , 30 °C): 2.67 (s, 3H, ar-CH ), 3.98
6
3
References and notes
(s, 3H, 6-OCH ), 4.00 (s, 3H, 8-OCH ), 4.08 (s, 3H, 1-
3 3
OCH ), 6.81 (d, 1H, J = 2.14 Hz, ar-H7), 7.35 (d, 1H,
3
1
2
3
. This numbering, although arbitrary, was chosen for a
better illustration of the regioselectivity in this work.
. Thomson, R. Naturally Occuring Quinones III; Chapman
and Hall Ltd.: London, 1987, pp 403–457.
. For reviews see: (a) Han, Y.-S.; Van der Heijden, R.;
Verpoorte, R. Plant Cell Tiss. Org. Cult. 2001, 67, 201–
J = 2.14 Hz, ar-H5), 7.87 (s, 1H, ar-H4), 10.65 (s, 1H,
1
3
–CHO) ppm. C NMR (125 MHz, d, DMSO-d , 30 °C):
6
21.9 (ar-CH ), 56.3 (6-OCH ), 57.0 (8-OCH ), 65.2(1-
OCH ), 102.8 (C5), 105.9 (C7), 118.3 (C8a), 125.9 (C4),
3
126.4 (C9a), 134.2 (C2), 136.7 (C10a), 136.8 (C4a), 146.8
3
3
3
(C3), 162.4 (C8), 164.7 (C6), 164.9 (C1), 180.9 (C9), 183.6
(C10), 192.9 (–CHO) ppm. ESI-MS (pos. ion mode):
m/z = 341 ([M+H] ). IR (KBr): 2940, 2849, 1686, 1676,
2
20; (b) Gill, M.; Steglich, W. In Progress in the Chemistry
of Organic Natural Products; Springer, Wien, 1987; Vol.
1, pp 147–149.
. (a) Birch, A. J.; Donovan, F. W. Aust. J. Chem. 1953, 6,
60–368; (b) Robinson, R. Structural Relations of Natural
+
À1
5
1599, 1456, 1427, 1377, 1324, 1257, 1209, 985 cm . UV–
4
vis (CHCl ): k
max
= 282 (100), 346 (14) nm, 405 (14) (rel.
3
1
3
int.); compound 6: Mp: 204–205 °C. H NMR (500 MHz,
Products; Clarendon: Oxford, 1955, pp 10.
d, CDCl , 2 5°C): 2.43 (s, ar-CH ), 3.99 (s, 8-OCH ), 4.03
3
3
3
5
6
7
8
9
. Steglich, W.; Arnold, R.; L o¨ sel, W.; Reininger, W. J.
Chem. Soc., Chem. Commun. 1972, 2, 102–103.
. Yamaguchi, M.; Hasebe, K.; Higashi, H.; Uchida, M.;
Irie, A.; Minami, T. J. Org. Chem. 1990, 55, 1611–1623.
. Steglich, W.; Reininger, W. J. Chem. Soc., Chem. Com-
mun. 1970, 3, 178.
. Joshi, B. S.; Ramanathan, S.; Venkataraman, K. Tetra-
hedron Lett. 1962, 21, 951–955.
. (a) Falk, H. Angew. Chem., Int. Ed. 1999, 38, 3136–3316;
(s, 6-OCH ), 6.79 (d, J = 2.3 Hz, ar-H7), 7.08 (s, ar-H2),
7.46 (d, J = 2.3 Hz, ar-H5), 7.57 (s, ar-H4), 13.09 (s, 1-OH)
3
1
3
ppm. C NMR (125 MHz, d, CDCl , 2 5°C): 22.19 (ar-
3
CH ), 56.27 (8-OCH ), 56.84 (6-OCH ), 104.2(C5), 104.9
3
3
3
(C7), 115.0 (C9a), 115.5 (C8a), 120.2 (C4), 125.0 (C2),
132.5 (C4a), 137.9 (C10a), 147.1 (C3), 162.9 (C1), 163.3
(C6), 165.5 (C8), 183.2(C10), 187.7 (C9) ppm. ESI-MS
+
(pos. ion mode): m/z = 299 ([M+H] ). IR (KBr): 3083,
2945, 2846, 1670, 1630, 1593, 1555, 1493, 1460, 1364, 1326,
1263, 1230, 1202, 1163, 1135, 1060, 1012, 946, 885, 838,
(
0. Oberm u¨ ller, R. A.; Hohenthanner, K.; Falk, H. Photo-
b) Falk, H. Angew. Chem. 1999, 111, 3306–3326.
À1
1
1
1
1
1
1
1
757, 611 cm . UV–vis (CHCl ): k
max
= 272 (100), 280
3
chem. Photobiol. 2001, 74, 211–215.
1. Waser, M.; Falk, H. Monatsh. Chem. 2005, DOI:10.1007/
S00706-004-0263-X.
2. Marschalk, C.; Koenig, F.; Ouroussoff, N. Bull. Soc.
Chim. Fr. 1936, 3, 1545–1575.
3. For a review see: Krohn, K. Tetrahedron 1990, 46, 291–
(96), 426 (36) nm (rel. int.); compound 7: Mp: 237–240 °C.
1
H NMR (500 MHz, d, DMSO-d , 30 °C): 2.49 (s, 3H, ar-
6
CH ), 3.97 (s, 3H, 8-OCH ), 3.98 (s, 3H, 6-OCH ), 4.60 (d,
2H, J = 5.19 Hz, ar-CH –), 4.92(t, 1H, J = 5.19 Hz, –OH),
3
3
3
2
7.02(d, 1H, J = 2.14 Hz, ar-H7), 7.29 (d, 1H, J = 2.14 Hz,
ar-H5), 7.46 (s, 1H, ar-H4), 13.68 (s, 1H, 1-OH) ppm.
1
3
C
3
18.
NMR (125 MHz, d, DMSO-d , 30 °C): 19.4 (ar-CH ), 53.1
6
3
4. Lackner, B.; Popova, Y.; Etzlstorfer, C.; Klampfl, C.;
Smelcerovic, A.; Falk, H. Monatsh. Chem., in press.
5. Hassall, C. H.; Morgan, B. A. J. Chem. Soc., Perkin
Trans. 1 1973, 2853–2861.
6. Kelly, T. R.; Xu, W.; Ma, Z.; Bushan, V. J. Am. Chem.
Soc. 1993, 115, 5843–5844.
2 3 3
(ar-CH –), 56.1 (8-OCH ), 56.6 (6-OCH ), 104.4 (C7),
104.5 (C5), 113.9 (C8a), 114.1 (C9a), 119.7 (C4), 130.5
(C4a), 134.8 (C2), 136.6 (C10a), 146.5 (C3), 160.0 (C1),
163.1 (C8), 165.1 (C6), 181.8 (C10), 186.7 (C9) ppm. ESI-
+
MS (pos. ion mode): m/z = 329 ([M+H] ). IR (KBr): 3472,
3335, 2920, 2850, 1699, 1618, 1489, 1457 1376, 1321, 1268,

2380
M. Waser et al. / Tetrahedron Letters 46 (2005) 2377–2380
À1
1
220, 1160, 1058, 1017 cm . UV–vis (CHCl
3
): kmax = 2 73
3
62.7 (1-OCH ), 102.7 (C5), 105.0 (C7), 116.9 (C8a), 123.5
(
100), 425 (32) nm (rel. int.); compound 8: Mp: 249–252 °C.
(C4), 125.2 (C9a), 131.8 (C4a), 135.7 (C10a), 138.6 (C2 or
C3), 140.0 (C3 or C2), 155.7 (C1), 161.6 (C8), 163.8 (C6),
167.6 (–COOH), 179.5 (C9), 182.4 (C10) ppm. ESI-MS
1
H NMR (500 MHz, d, DMSO-d , 30 °C): 2.52 (s, 3H, ar-
6
CH
3
), 3.84 (s, 3H, 1-OCH
H, 6-OCH ), 4.60 (d, 2H, J = 5.19 Hz, ar-CH
3
), 3.93 (s, 3H, 8-OCH
3
), 3.95 (s,
À
3
1
7
3
2
–), 5.03 (t,
(neg. ion mode): m/z = 355 ([MÀH] ). IR (KBr): 3502,
H, J = 5.19 Hz, –OH), 7.00 (d, 1H, J = 2.14 Hz, ar-H7),
.20 (d, 1H,J = 2.14 Hz, ar-H5), 7.73 (s, 1H, ar-H4) ppm.
2923, 2855, 1687, 1673, 1598, 1458, 1394, 1353, 1255, 1213,
1163, 1151, 1084, 1042, 984, 750 cm . UV–vis (CHCl3):
À1
1
3
C NMR (125 MHz, d, DMSO-d
6
, 30 °C): 19.4 (ar-CH
), 62.9 (1-
3
),
kmax = 276 (100), 344 (19), 396 (20) nm (rel. int.);
1
54.2(ar-CH –), 55.9 (6-OCH ), 56.5 (8-OCH
OCH ), 102.5 (C5), 105.0 (C7), 117.2 (C8a), 123.4 (C4),
1
1
(
(
1
3
2
3
3
7
2
3
3
compound 10: Mp: 273–275 °C. H NMR (500 MHz, d,
DMSO-d , 30 °C): 2.39 (s, 3H, ar-CH ), 3.97 (s, 3H, 8-
3
6
3
25.5 (C9a), 132.6 (C4a), 135.7 (C10a), 141.4 (C2 or C3),
45.0 (C3 or C2), 158.0 (C1), 161.4 (C8), 163.6 (C6), 180.2
OCH
3
), 4.00 (s, 3H, 6-OCH
3
), 7.05 (d, 1H, J = 2.14 Hz, ar-
H7), 7.32(d, 1H, J = 2.14 Hz, ar-H5), 7.53 (s, 1H, ar-H4),
1
3
C9), 182.8 (C10) ppm. ESI-MS (pos. ion mode): m/z = 343
[M+H] ). IR (KBr): 3481, 2925, 2853, 1669, 1593, 1458,
13.45 (s, 1H, 1-OH), 13.55 (br s, 1H, –COOH) ppm.
C
+
NMR (125 MHz, d, DMSO-d , 30 °C): 19.4 (ar-CH ), 56.2
6
3
À1
328, 1263, 1162 cm . UV–vis (CHCl
49 (15) nm, 393 (16) (rel. int.); compound 9: Mp: 228–
3
): kmax = 278 (100),
3 3
(6-OCH ), 56.7 (8-OCH ), 104.5 (C5), 104.8 (C7), 113.9
(C8a), 114.7 (C9a), 119.4 (C4), 128.6 (C10a), 131.5 (C3),
131.6 (C10a), 142.2 (C3), 158.0 (C1), 163.2 (C8), 165.3
(C6), 167.2(–COOH), 181.6 (C10), 186.3 (C9) ppm. ESI-
1
32 °C. H NMR (500 MHz, d, DMSO-d , 30 °C): 2.38 (s,
H, ar-CH ), 3.83 (s, 3H, 1-OCH ), 3.94 (s, 3H, 8-OCH ),
6
3
3
3
À
.96 (s, 3H, 6-OCH
.22 (d, 1H, J = 2.14 Hz, ar-H5), 7.80 (s, 1H, ar-H4), 13.65
3
), 7.02(d, 1H, J = 2.14 Hz, ar-H7),
MS (neg. ion mode): m/z = 341 ([MÀH] ). IR (KBr): 3421,
2924, 2854, 1735, 1685, 1627, 1559, 1508, 1363, 1254, 972,
748 cm . UV–vis (CHCl ): k
(rel. int.).
1
3
À1
(
br s, 1H, –COOH) ppm. C NMR (125 MHz, d, DMSO-
= 284(100), 432 (36) nm
3
max
d , 30 °C): 18.9 (ar-CH ), 55.9 (6-OCH ), 56.5 (8-OCH ),
6
3
3
3