вход по аккаунту


Occurrence of the Fungal Toxin Orellanine as a Diglucoside and Investigation of Its Biosynthesis.

код для вставкиСкачать
Biosynthesis of Orellanine
Occurrence of the Fungal Toxin Orellanine as a
Diglucoside and Investigation of Its
Peter Spiteller, Michael Spiteller, and Wolfgang Steglich*
Dedicated to Professor Meinhart H. Zenk
on the occasion of his 70th birthday
The deadly poisonous toadstool Cortinarius orellanus (Fr.)
Fr.[1] causes an irreversible loss of kidney function within one
to two weeks after ingestion.[2] The toxicity of this fungus was
recognized nearly 50 years ago after a tragic mass poisoning in
Poland, which caused several deaths.[3] The responsible toxin,
orellanine,[4] was assigned the structure 2,2’-bipyridine3,3’,4,4’-tetrol-1,1’-dioxide (1),[5] which has been confirmed
by several syntheses[6] and a crystal structure analysis.[7] In this
communication we report initial investigations on the biosynthesis of orellanine and the discovery that the toxin occurs
mainly as the water-soluble 4,4’-diglucopyranoside 5.
[*] Prof. Dr. W. Steglich, Dr. P. Spiteller
Department Chemie
Ludwig-Maximilians-Universit#t M$nchen
Butenandtstrasse 5–13, Haus F, 81377 M$nchen (Germany)
Fax: (+ 49) 89-2180-77756
Prof. Dr. M. Spiteller
Institut f$r Umweltforschung
Universit#t Dortmund
Otto-Hahn-Strasse 6, 44227 Dortmund (Germany)
[**] We thank Dr. N. Arnold for his assistance and valuable mycological
hints, H. Avak and J. Oeßelmann (Themo Finnigan) for the IR-MS
measurements, and Dr. F. Hampson for linguistic help. This work
was supported by the Deutsche Forschungsgemeinschaft
(SFB 369).
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200351066
Angew. Chem. Int. Ed. 2003, 42, 2864 – 2867
It is tempting to speculate from the symmetric structure of
orellanine (1) that it is generated from two molecules of 3hydroxy-1H-pyridine-4-one (3). Oxidative 2,2’-dimerization
would yield orelline (2), which is then converted into 1 by two
consecutive N-oxidations. This assumption is supported by
the detection of 3 in the acidified methanolic extract of
C. orellanus together with 3,4-dihydroxypyridine-2-carboxylic
acid (4), a further potential precursor of orellanine. Both
compounds have been identified by a GC-MS comparison
with synthetic samples. Compound 3 is already known to be a
degradation product of mimosine and is present in dried
leaves of Leucaena leucocephala.[8]
Several routes must be considered for the biosynthesis of
simple pyridine derivatives. As in mimosine, the pyridine ring
could be formed from lysine.[9] Alternatively, 3-hydroxy-1Hpyridine-4-one might be generated from ammonia and 3hydroxy-4H-pyran-4-one, which in turn could, like maltol,[10]
be derived from carbohydrates. In addition, anthranilic acid
and tryptophan are also possible precursors. Both are transformed via 3-hydroxyanthranilic acid into nicotinic acid in
animals and yeasts,[11] whereas in bacteria and higher plants
nicotinic acid is built up from aspartic acid and dihydroxyacetone phosphate.[12]
Feeding experiments were carried out with toadstools in
the natural environment in order to clarify the biosynthesis of
orellanine. C. orellanus is very rare in Germany. For this
reason C. rubellus Cooke (C. speciosissimus K7hn. Romagn.),
which is more common and produces the same toxin, was used
instead.[13] After injection of an aqueous solution of the
labeled precursor into 5–20 fruit bodies of C. rubellus, the
toadstools were left in the forest for 5 to 12 days. Then they
were collected and the orellanine was isolated. Feeding
experiments with [U-13C]glucose or a mixture of U-13Clabeled amino acids showed, according to the NMR analysis,
no significant 13C enrichment, which would rule out a direct
biosynthetic origin from glucose or lysine. A feeding experiment with [U-13C]glycerol was similarly negative.
In contrast, when [15N]anthranilic acid was applied to the
toadstool, 15N NMR measurements indicated a minor incorporation in orellanine. However, the result was not conclusive
on account of the very low rate of incorporation, which
was approximately 0.1–0.2 %. For this reason, the incorporation was determined by using highly sensitive 15N/14N
isotope ratio mass spectrometry (IR-MS),[14] a method that
was adapted from 13C/12C mass ratio measurements only a
couple of years ago.[15] The 15N enrichments obtained after
feeding experiments with [15N]anthranilic acid are given in
Table 1.
The 15N enrichment of approximately 30 % unambiguously confirms the incorporation of [15N]anthranilic acid into
orellanine. By application of this method the detection limit
for an enrichment is below 0.1 %. When [ring-15N]tryptophan
was applied, the enrichment was only approximately 7 %.
This result suggests that anthranilic acid is converted directly
into 3-hydroxyanthranilic acid, while the kynurenine pathway
commencing from tryptophan obviously plays only a minor
role. In order to exclude the possibility that the incorporation
is due to degradation of anthranilic acid and an unspecific
dissemination of 15N in the fungi, 15NH415NO3 was fed in high
Angew. Chem. Int. Ed. 2003, 42, 2864 – 2867
Table 1: 15N enrichment experiments with 15N-labeled compounds fed to
C. rubellus.
Substance fed
ratio [%][a]
N enrichment [%][a]
rate [%]
water (blank)
[15N]anthranilic acid
(sample 1)
[15N]anthranilic acid
(sample 2)
[a] The 15N enrichments refer to the natural 15N/14N ratio of 0.36630 % in
nitrogen from air (international standard).
doses (50 mg per fruit body) to C. rubellus, which led to a
N enrichment of only 13 %.
The yield of orellanine was always considerably higher
when the methanolic crude extract was acidified with some
drops of 2 n HCl during the workup procedure from
C. orellanus and C. rubellus. It turned out that the neutral
methanolic extract also contains two very polar compounds,
which are converted to orellanine by acidification. The liquid
chromatography mass spectra (LC-MS) suggest that these
compounds are a mono- and a diglycoside of orellanine. The
major component could be separated from accompanying
carbohydrates by carefully performing HPLC on an amino
phase. It was identified by one- and two-dimensional NMR
spectroscopy to be orellanine-4,4’-di-b-d-glucopyranoside
(5).[16] This conclusion is consistent with the fact that feeding
C. rubellus with [U-13C]glucose leads to high incorporation
rates of up to 10 % in the sugar moieties of 5. The
incorporation can be detected in the LC-MS by a mass shift
of 6 or of 12 units, respectively, and the fragmentation pattern
in the LC-MSMS (loss of m/z 168 (13C6H4O5)). The position of
the glucose units has been proven by the HMBC correlations
shown in Figure 1. The occurrence of orellanine in the form of
a diglucoside that is soluble in water explains the contradictory reports in the literature on the solubility of this
Based on the results outlined above we propose the
following hypothetical pathway depicted in Scheme 1. However, the order of some steps remains arbitrary. The biosynthesis starts with anthranilic acid (6), which is converted via 3hydroxyanthranilic acid (7) into 3,6-dihydroxyanthranilic acid
Figure 1. Orellanine-4,4’-diglucopyranoside (5) with selected HMBC
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
in 1 mL MeOH prepurified by passage through an
RP 18 cartridge. Compound 1 was isolated by
preparative HPLC, tR = 23.5 min (column: Nucleosil 100 C-18, 7 mm, 20 E 250 mm (Knauer); gradient:
100 % H2O linear to 100 % MeOH in 40 min; flow
rate: 6 mL min1; detection: UV at l = 280 nm).
Yield: 12 mg (0.12 % based on the fresh weight).
N NMR (61 MHz, [D6]DMSO, reference: DMF
d = 77.7 ppm): d = 207.8 ppm. LC/ESI-MS: tR =
16.4 min (column: Nucleosil 100 C-18, 5 mm, 250 E
2 mm (Macherey-Nagel); gradient: 5 min isocratic
with 100 % H2O, then linear from 100 % H2O to
100 % MeCN in 30 min, flow rate: 0.25 mL min1;
detection: UV at l = 300 nm); m/z: 253 [MþH]+.
Isolation of 5: The crude extract was prepared as
described above for the isolation of 1, except that the
addition of acid was omitted and 50 g of frozen
toadstool material was used. Compound 5 was
Scheme 1. Hypothetical biosynthesis of orellanine (1) from anthranilic acid. Intermediates in
isolated by HPLC from the extract. tR = 7–12 min
brackets have not yet been identified in the toadstool.
(column: Nucleosil 100 NH2, 5 mm, 25 cm E 4.6 mm
(Macherey-Nagel); solvent: MeCN/H2O (1:1), isocratic; flow rate: 2 mL min1; detection: UV at l =
(8). Recently, 8 has been recognized by us as a building block
330 nm). Yield: 12 mg (0.024 % based on the fresh weight),[16]
colorless solid. UV/Vis (H2O): lmax (lg e) = 237 (3.68), 262 (sh, 3.14),
of the toadstool pigment blennione.[18] After oxidative ring
318 nm (3.15). CD (H2O): lmax (De) = 220 (1.5), 237 (16.5), 257 (sh,
opening and a new ring closure, 4-hydroxypyridine-2,35.1), 284 (0.1), 314 (5.6), 344 (0.9), 360 (0.1), 373 nm (0.7).
dicarboxylic acid (9) could be generated from 8 and then
H NMR (600 MHz, D2O): d = 3.43 (dd, 3J = 9.0, 8.4 Hz, 2 H, 4’’/4’’’converted by oxidative decarboxylation into 3,4-dihydroxyH), 3.53 (dd, 3J = 9.4, 9.0 Hz, 2 H, 3’’/3’’’-H), 3.58 (dd, 3J = 9.4, 7.5 Hz,
pyridine-2-carboxylic acid (4), the precursor of 3-hydroxy2 H, 2’’/2’’’-H), 3.63 (dd, 3J = 8.4, 5.5 Hz, 2 H, 5’’/5’’’-H), 3.68 (dd, 2J =
1H-pyridine-4-one (3). At the latest, at this stage formation of
12.6, 3J = 5.5 Hz, 2 H, 6’’/6’’’-H), 3.84 (dd, 2J = 12.6, 3J = 1.8 Hz, 2 H, 6’’/
the 4-O-glucoside 10 should take place, which enables the
6’’’-H), 5.07 (dd, 3J = 7.5, 4J = 1.6 Hz, 2 H, 1’’/1’’’-H), 7.01 (d, 3J =
Hz, 2 H, 5/5’-H), 7.53 ppm (d, 3J = 6.8 Hz, 2 H, 6/6’-H). 13C NMR
regioselective coupling to orelline-4,4’-di-b-d-glucopyrano(151 MHz, D2O): d = 61.1 (C-6’’/6’’’), 69.9 (C-4’’/4’’’), 73.2 (C-2’’/2’’’),
side (11) by generation of a phenoxy radical in position 2.[19]
75.7 (C-3’’/3’’’), 76.9 (C-5’’/5’’’), 100.9 (C-1’’/1’’’), 111.0 (C-5/5’), 126.3
Twofold N-oxidation could finally lead to orellanine-4,4’-di-b(C-6/6’), 132.7 (C-2/2’), 151.5 (C-4/4’), 154.5 ppm (C-3/3’). LC/ESId-glucopyranoside (5). Orellanine (1) itself is generated by
MS: tR = 13.1 min (same program and column as described for the
hydrolysis of the diglucoside 5 via the monoglucoside.
isolation of 1), m/z: 577 [MþH]+. LC/ESI-MS/MS (parent ion: m/z =
The result of a 18O/16O IR-MS experiment is in agreement
577, 25 eV) m/z (%): 577 (24) [MþH]+, 415 (65) [MþHC6H10O5]+,
253 (100) [[MþH(2 E C6H10O5)]+.
with the proposed route.[20] It indicates that all oxygen atoms
Identification of 3 and 4 by GC-MS: Frozen fruit bodies (2 g)
of orellanine originate from atmospheric oxygen, since a
C. rubellus and from C. orellanus were extracted with MeOH
value of 11% was measured for D O. This value is estimated
(2 E 20 mL). The combined extracts were filtered, 2 n HCl (2 mL)
to be approximately 30% for oxygen originating from
added, and the solvents removed in vacuo at 50 8C. The residue was
water,[21] excluding precursors from carbohydrate metabodissolved in MeOH (2 mL), and carbohydrates were removed by
preparative HPLC (RP-18 phase). The fractions containing the
Orellanine is another example for the involvement of
pyridine compounds were dried and treated with N-methyl-(trimethylsilyl)trifluoroacetamide (100 mL) for 2 h at 40 8C. GC-MS comanthranilic acid in the biosynthesis of the secondary metabparison was made by coinjection with corresponding authentic
olites of toadstools. In addition, 2-aminobenzaldehyde, the
derivatives, which had been synthesized by literature methods[6e].
fragrance from Hebeloma sacchariolens, the chromoalka3,4-Bis(trimethylsilyloxy)pyridine: GC-MS (column: DB5 ms,
loid necatorone
from Lactarius necator, and the amino30
E 0.25 mm, 0.25 mm, program: 2 min isotherm at 50 8C, then
benzoquinone pigments blennione[18] and lilacinone[24] from
with 10 K min1 linear to 300 8C): Ri = 1430; EI-MS: m/z (%): 255
some Lactarius species are also derived from it.
(76), 240 (100), 182 (1), 168 (17), 147 (21), 152 (7), 75 (23), 73 (77), 45
3,4-Bis(trimethylsilyloxy)pyridine-2-carboxylic acid trimethylsilyl ester: GC-MS (measurement conditions as above): Ri = 1820;
Experimental Section
EI-MS: m/z (%): 371 (4), 356 (100), 298 (9), 254 (4), 194 (3), 147 (2),
Feeding experiments with C. rubellus were performed from July to
73 (15).
September in the years 1998 to 2001 in moist coniferous forests near
Penzberg and Leutstetten in southern Bavaria (Germany). Typically,
Received: January 30, 2003 [Z51066]
5–20 fruit bodies were used for a single experiment. Each fruit body
received 20 mg of the potential precursor dissolved in 20 mL of water,
Keywords: anthranilic acid · biosynthesis · mass spectrometry ·
which was applied with a syringe. The treated fruit bodies were
natural products · orellanine
collected after 5–10 days and frozen at 20 8C.
Isolation of 1: The frozen fruit bodies (10 g) were extracted with
MeOH (3 E 100 mL) in a beaker for 30 min at 25 8C. The combined
[1] H. Besl, A. Bresinsky, Colour Atlas of Poisonous Fungi, CRC,
extracts were filtered and treated with 2 n HCl (10 mL). After
Boca Raton, FL, 1990.
evaporation of the solvent at 60 8C in vacuo, the residue was dissolved
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2003, 42, 2864 – 2867
182; b) T. P. Begley, C. Kinsland, R. A. Mehl, A. Osterman, P.
Dorrestein, Vitam. Horm. 2001, 61, 103 – 119.
Review: R. F. Dawson, D. R. Christman, R. U. Bjerrum, Methods Enzymol. 1971, 18B, 90 – 113.
G. E. B. Koller, K. Høiland, K. Janak, F. C. Stormer, Mycologia
2002, 94, 752 – 756. We also could detect the simple pyridine
derivatives 3 and 4 in C. rubellus.
W. A. Brand, J. Mass Spectrom. 1996, 31, 125 – 235.
R. A. Werner, B. A. Bruch, W. A. Brand, Rapid Commun. Mass
Spectrom. 1999, 13, 1237 – 1241.
The yield of the diglucoside varies and depends on the age and
the condition of the toadstools. In addition, major losses occur
due to high sensitivity of 5 to hydrolysis and strong adsorption to
the amino phase.
H. K7rnsteiner, M. Moser, Mycopathologia 1981, 74, 65 – 72.
P. Spiteller, W. Steglich, J. Nat. Prod. 2002, 65, 725 – 727.
It cannot be excluded that the carboxylic acid 4 is coupled
directly to the orelline diglucoside 11, after its glucosylation. We
are grateful to one of the reviewers for pointing this out.
H.-L. Schmidt, R. A. Werner, A. Roßmann, Phytochemistry
2001, 58, 9 – 32.
J. Koziet, J. Mass Spectrom. 1997, 32, 103 – 108.
F. v. Nussbaum, W. Spahl, W. Steglich, Phytochemistry 1997, 46,
261 – 264.
a) J. D. Klamann, B. Fugmann, W. Steglich, Phytochemistry 1989,
28, 3519 – 3522; b) “Bioactive Compounds from Plants”: W.
Steglich, B. Steffan, T. EizenhSfer, B. Fugmann, R. Herrmann, J.D. Klamann, Ciba Found. Symp. 1990, 154, 56 – 65.
P. Spiteller, N. Arnold, W. Steglich, J. Nat. Prod., submitted.
2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[2] Examples: a) J. F. Marichal, F. Trilby, J. L. Wiederkehr, R.
Carbiener, Nouv. Presse Med. 1977, 6, 2973 – 2975; b) R.
Flammer, Schweiz. Med. Wochenschr. 1982, 112, 1181 – 1184;
c) T. Schumacher, K. Høiland, Arch. Toxicol. 1983, 53, 87 – 106;
d) J.-M. Richard, J. Louis, D. Cantin, Arch. Toxicol. 1988, 62,
242 – 245; e) H. Prast, E. R. Werner, W. Pfaller, M. Moser, Arch.
Toxicol. 1988, 62, 81 – 88; f) N. Delpech, S. Rapior, P. Donnadieu,
A. P. Cozette, J. P. Ortiz, G. Huchard, Nephrologie 1991, 12, 63 –
[3] S. Grzymala, Z. Pilzkd. 1957, 23, 139 – 142.
[4] S. Grzymala, Bull. Trimest. Soc. Mycol. Fr. 1962, 78, 394 – 404.
[5] W. Z. Antkowiak, W. P. Gessner, Tetrahedron Lett. 1979, 1931 –
[6] a) E. V. Dehmlow, H.-J. Schulz, Tetrahedron Lett. 1985, 26,
4903 – 4906; b) M. Tiecco, M. Tingoli, L. Testaferri, D. Chianelli,
E. Wenkert, Tetrahedron 1986, 42, 1475 – 1485; c) E. V. Dehmlow, H.-J. Schulz, Liebigs Ann. Chem. 1987, 857 – 861; d) H. A.
Hasseberg, H. Gerlach, Helv. Chim. Acta 1988, 71, 957 – 963;
e) F. TrLcourt, M. Mallet, O. Mongin, B. Gervais, G. QuLguiner,
Tetrahedron 1993, 49, 8373 – 8380.
[7] M. Kubicki, T. Borowiak, W. Z. Antkowiak, J. Crystallogr.
Spectrosc. Res. 1991, 21, 401 – 405.
[8] T. Acamovic, J. P. F. D'Mello, K. W. Fraser, J. Chromatogr. 1982,
236, 169 – 179, and references therein.
[9] a) J. W. Hylin, Phytochemistry 1964, 3, 161 – 164; b) H. P. Tiwari,
W. R. Penrose, I. D. Spenser, Phytochemistry 1967, 6, 1245 –
[10] H. Chen, G. Agnihotri, Z. Guo, N. L. S. Que, X. H. Chen, H.-W.
Liu, J. Am. Chem. Soc. 1999, 121, 8124 – 8125.
[11] Reviews: a) G. Magni, A. Amici, M. Emanuelli, N. Raffaelli, S.
Ruggieri, Adv. Enzymol. Relat. Areas Mol. Biol. 1999, 73, 135 –
Angew. Chem. Int. Ed. 2003, 42, 2864 – 2867
Без категории
Размер файла
114 Кб
investigation, orellanine, occurrence, diglucoside, toxic, fungal, biosynthesis
Пожаловаться на содержимое документа