The synthesis of kermesic acid by acetylation-aided tautomerism of 6- chloro-2,5,8-trihydroxynaphtho-1,4-quinone

Article abstract of DOI:10.1039/b002591g

6-Chloro-2,5,8-trihydroxynaphtho-1,4-quinone does not undergo cycloaddition reactions but the 1,2-diacetate, 2-chloro-5,6-diacetoxy-8- hydroxynaphtho-1,4-quinone, formed by acetylation-aided tautomerism added (E)- and (Z)-3-alkoxycarbonyl-2,4-bis(trimethy

Full text of DOI:10.1039/b002591g

The synthesis of kermesic acid by acetylation-aided tautomerism of
6-chloro-2,5,8-trihydroxynaphtho-1,4-quinone
Steve J. Bingham and John H. P. Tyman*
Department of Chemistry, Brunel University, Uxbridge, Middlesex, UK UB8 3PH. E-mail: john.tyman@brunel.ac.uk
Laboratories of European Colour plc, Nathan Way, London, UK SE18
Received (in Cambridge, UK) 14th March 2000, Accepted 20th April 2000
6-Chloro-2,5,8-trihydroxynaphtho-1,4-quinone does not un-
dergo cycloaddition reactions but the 1,2-diacetate,
2-chloro-5,6-diacetoxy-8-hydroxynaphtho-1,4-quinone,
formed by acetylation-aided tautomerism added (E)- and
(Z)-3-alkoxycarbonyl-2,4-bis(trimethylsiyloxy)penta-
1,3-diene to afford kermesic acid after hydrolysis.
Kermesic acid 1a, the red colourant component of kermes, a
dyestuff of great antiquity and probably the earliest of which
Scheme 1 Reagents and conditions: 4 (X = Cl) from 4 (X = H), NCS,
HOAc, Mg(OAc)2 r.t., 24 h: i, DMS, Me2CO, K2CO3, heat, 24 h; ii,
NaOMe, MeOH, (MeO)2CO, hear; iii, (COCl)2, MeNO2, AlCl3, 0 to 80 °C,
3 h; iv, 10% conc. HCl–HOAc, 100 °C, 5 h.
unhalogenated naphthoquinones,8 afforded the two 3-carbo-
nylmethoxy-6-halogeno-2,5,8-trihydroxynaphth-1,4-quinones
7 (X = Br) 48% and 7 (X = Cl) 56% together with some
partially demethylated quinones. Hydrolysis of the trihydrox-
there is a record,1 is also the aglycone of carminic acid2 2, the
colourant principle of cochineal.
Although carminic acid historically, through its brilliant
yquinone esters with either aqueous sodium hydroxide followed
range of red hues, superseded kermesic acid, dyeing with
by acidification, or with hot acetic acid containing hydrochloric
kermes is still active.3 Nevertheless, the natural source of
acid (the preferred method) then gave the parent compounds 3
kermesic acid, for example from the wingless insect Kermes
(X = Br) 69% and 3 (X = Cl) 53%.
illicis which infects the kermes oak, is not abundant, and it was
of interest to devise a synthetic approach that could provide an
intermediate from which, by different strategies, either ker-
mesic or carminic acid could be obtained. It was desirable to
Under a variety of conditions 3 (X = Cl) failed to undergo
Diels–Alder cycloaddition with the dienes (E)- and (Z)-
3-methoxycarbonyl-2,4-bis(trimethylsilyloxy)penta-1,3-diene2
8a. This was attributed primarily to the intramolecular hydrogen
avoid penultimate tricyclic intermediates having an halogeno
bonding9 of the 2-OH group thus inhibiting the normal
group requiring difficult hydrolysis4,5 to the required hydroxy
tautomerism of the naphthazarin system and locking the
group and in another method6 the simultaneous formation of
structure in form 1. Further support for this exclusive structure
isokermesic acid. Therefore, effort was focused on the prepara-
was provided by an X-ray structure determination (Fig. 1)† and
tion of 2-halogeno-5,6,8-trihydroxynaphtho-1,4-quinone 3,
that, under mild conditions, 3 (X = Cl) readily underwent: (a)
Michael addition at the 3-position and (b) selective O-
methylation in 77% yield at the 2-OH with BF3·MeOH which
then removed the impediment to tautomerism and was expected
to assist cycloaddition.
Acetylation of 3 (X = Cl) in dichloromethane with pyridine–
acetic anhydride afforded the triacetoxy-compound in 91%
yield as a 2+1 mixture (confirmed by NMR) of 6-chloro-
2,5,8-triacetoxynaphtho-1,4-quinone and 2-chloro-5,6,8-triace-
toxynaphtho-1,4-quinone. However, upon heating 3 (X = Cl) in
acetic anhydride alone at 100 °C the 2-chloro-5,6-diacetoxy
(form 2) which, as with naphthazarins generally, could be
expected to exist as the 6-halogeno-2,5,8-trihydroxynaphtho-
1,4-quinone tautomer 3 (form 1) where the halogeno atom is
required to aid subsequent cycloaddition.
The starting point of the synthesis (Scheme 1) was 3-bromo-
2-hydroxy-5-methoxyacetophenone7 4 (X
= Br), and its
chloro-analogue 4 (X = Cl), prepared in 80% yield by selective
chlorination of 2-hydroxy-5-methoxyacetophenone with N-
chlorosuccinimide (NCS) in acetic acid containing Mg(OAc)2.
Methylation of 4 with DMS in acetone containing K2CO3
afforded 5 (X = Br) 82% and 5 (X = Cl) 81%. Reaction of each
with dimethyl carbonate in methanolic sodium methoxide gave
the b-ketoesters 6 (X = Br, Cl), both in quantitative yield which
by treatment with oxalyl chloride in nitromethane containing
anhydrous aluminium chloride, a process used to prepare
Fig.
1 X-Ray crystal structure of 6-chloro-2,5,8-trihydroxynaphtho-
1,4-quinone; C–OH: a1, a2, a3 = 1.33, 1.34, 1.35 Å; CNO: b1, b2 = 1.22,
1.24 Å.
DOI: 10.1039/b002591g
Chem. Commun., 2000, 925–926
This journal is © The Royal Society of Chemistry 2000
925

methyl kermesate 1b with DMS in acetone solution containing
potassium carbonate gave methyl 3,5,6,8-tetramethoxy-1-me-
thylanthra-9,10-quinone-2-carboxylate 1c, which was identical
with an authentic sample4 kindly made available by Paul
Brassard. Correct spectroscopic, elemental analysis and mass
spectral data were obtained for all compounds.
Scheme 2 Reagents and conditions: i, ii, Ac2O, 100 °C, 3 h; vac. to
dryness.
compound 9 resulted quantitatively. It is believed (Scheme 2)
that acetylation of the 2-OH occurs first allowing tautomerism
to form 2 to take place and is then followed by acetylation at the
5-position which locks the structure as form 2. It is conjectural
that since hydrogen-bonded OH groups are more difficult to
acetylate, migration of the 6-acetyl group to the 5-position may
possibly occur enabling the more susceptible 6-OH group to
then react. This tentative notion may explain the non-
acetylation of the hydrogen-bonded 8-OH although we have no
distinct evidence to support this pathway. X-Ray crystallog-
raphy provided confirmation of the structure of the 5,6-diace-
toxy compounds (Fig. 2).†
The diacetyl compound 9 readily underwent cycloaddition
with (E)- and (Z)-3-methoxycarbonyl-2,4-bis(trimethylsilyl-
oxy)penta-1,3-diene 8a followed by aromatisation in boiling
toluene to afford 10, methyl 5,6-diacetoxy-3,8-dihydroxy-
1-methylanthra-9,10-quinone-2-carboxylate in 87% yield after
column chromatography, with definitive proof of structure
being provided by an X-ray study (Fig. 3).† The isopropyl
analogue 8b reacted similarly in 64% yield.
Scheme 3 Reagents and conditions: i, 9, toluene, 8, heat, 24 h; col chrom.
SiO2, ii, MeOH, 1% Na2CO3; iii, HOAc–HCl, heat; iv, 1b, DMS, Me2CO,
K2CO3, heat 11 from 3 with BF3·MeOH; toluene, 8, heat; col. chrom.
SiO2.
Finally, it is believed that 3 (X = Cl) also has the potential
utility to facilitate efficient access to carminic acid and work is
continuing in this approach.10
Financial support is acknowledged from the SPUR initiative
(DTI) and Mr G. Marshall (European Colour) is thanked for
critical comments. We thank Professor S. Roberts (Exeter
University) for facilitating and Professor M. Hursthouse
(University of Wales, Cardiff) for effecting X-ray structure
determinations.
2-Chloro-6-methoxy-5,8-dihydroxynaphtho-1,4-quinone 11,
in which normal naphthazarin tautomerism is able to operate,
although the structure is not locked as with the diacetate, also
underwent cycloaddition with the diene 8a in boiling toluene to
afford after work-up, methyl 6-methoxy-3,5,8-trihydroxy-
1-methylanthra-9,10-quinone-2-carboxylate 12 in 42% yield.
Possibly both forms 1 and 2 are present owing to the fact that
there is no locking of the structure in this compound. Hydrolysis
of 10 in 1% methanolic sodium carbonate followed by
acidification gave methyl kermesate 1b quantitatively and
thence by refluxing in acetic acid containing hydrochloric acid,
kermesic acid 1a was obtained (Scheme 3). Methylation of
Notes and references
¯
† Crystal data: 3: C20H10Cl2O10, M 479.16, triclinic, space group P1, a =
6.756(4), b = 9.329(7), c = 11.536(9) Å, V = 676.4(8) Å3, Z = 3, m =
0.425 mm21; 1913 independent reflections (Rint = 0.0484), R indices (all
data): R1 = 0.0894, wR2 = 0.1282.
9: C14H9ClO7, M 324.66, monoclinic, space group C2/c, a = 16.0510(8),
b = 5.5590(8), c = 30.980(8) Å, V = 2743.7(9) Å3, Z = 8, m = 0.313
mm21, 1926 independent reflections: (Rint = 0.0742), R indices (all data):
R1 = 0.1202, wR2 = 0.2532.
10: C21H16O10, M = 428.34, triclinic, space group P1, a = 8.295(4), b
= 8.8970(10), c = 12.799(8) Å, V = 922.0(7) Å3, Z = 2, m = 0.125 mm21
¯
,
=
2459 independent reflections (Rint = 0.0518), R indices (all data): R1
0.0836, wR2 = 0.1125.
CCDC 182/1611.
1 R. H. Thomson, Naturally Occurring Quinones, Academic Press,
London, New York, 1971, p. 458.
2 P. Allevi, M. Anastasia, S. J. Bingham, P. Ciuffreda, A. Fiecchi, G.
Cighetti, M. Muir, A. Scala and J. H. P. Tyman, J. Chem. Soc., Perkin
Trans. 1, 1998, 575; J. H. P. Tyman, Synthetic and Natural Phenols,
Elsevier, Amsterdam, 1996, p. 623.
3 A. Verhecken, J. Soc. Dyers Col., 1989, 105, 389.
4 G. Roberge and P. Brassard, J. Chem. Soc., Perkin Trans. 1, 1978,
1041.
5 D. W. Cameron, D. J. Deutscher, G. I. Feutrill and P. G. Griffiths, Aust.
J. Chem., 1981, 34, 2401.
6 S. J. Bingham and J. H. P. Tyman, J. Chem. Soc., Perkin Trans. 1, 1997,
3637; K. Venkataraman and A. V. Rama Rao, in Some Recent
Developments in the Chemistry of Natural Products, Prentice Hall of
India, New Delhi, 1972.
7 C. T. Chang, M. F. Young and F. C. Chen, Formosan Sci., 1962, 16,
29.
Fig. 2 X-Ray crystal structure of 2-chloro-5,6-diacetoxy-8-hydroxyna-
phtho-1,4-quinone.
8 G. Sartori, F. Bigi, G. Canali, R. Maggi, G. Casnati and X. Tao, J. Org.
Chem., 1993, 58, 840.
9 The 2-, 5- and 8-OH groups in 3 (form 1) are hydrogen bonded, as seen
in the 1H NMR spectrum, with d (DMSO-d6) at 12.75 (br, s), 11.70,
13.35 ppm, respectively, cf. d (CDCl3) (Sadtler 19857) 10.34 for
lawsone (2-hydroxynaphtho-1,4-quinone) and d (Me2CO) (Sadtler
3230) 8.35 for 2-naphthol. No NMR data for the 2-OH are given in ref.
8. In F. Farina, R. Martinez-Utrilla and M. C. Paredes, Synthesis, 1981,
300, the authors do not list the 2-OH group in data for 2-hydroxy-
5,8-dimethoxynaphtho-1,4-quinone).
10 For this 3 (form 1) appears to be an ideal candidate from initial work on
Michael addition and enamine reactions. Chemical approaches to using
kermesic acid itself as an intermediate to carminic acid have not so far
been successful although this or xanthokermesic acid are the likely
biosynthetic precursors.
Fig. 3 X-Ray crystal structure of methyl 5,6-diacetoxy-3,8-dihydroxy-
1-methylanthra-9,10-quinone-2-carboxylate.
926
Chem. Commun., 2000, 925–926