Conversion of aristolochic acid i into aristolic acid by reaction with cysteine and glutathione: Biological implications

Article abstract of DOI:10.1021/np300822b

Aristolochic acid I (AA-I), naturally occurring in Aristolochia plants, is a potent nephrotoxin and carcinogen. Here we report that AA-I suffers hydrogenolysis with loss of the nitro group by reaction with cysteine or glutathione to give aristolic acid. Since the reaction can proceed in aqueous solutions at pH 7.0 and 37 C, it is inferred that it may also occur in biological systems and contribute to the nephrotoxic effects induced by AA-I.

Full text of DOI:10.1021/np300822b

Note
pubs.acs.org/jnp
Conversion of Aristolochic Acid I into Aristolic Acid by Reaction with
Cysteine and Glutathione: Biological Implications
Horacio A. Priestap*,† and Manuel A. Barbieri†,‡
†Department of Biological Sciences, Florida International University, 11200 Southwest 8th Street, Miami, Florida 33199, United
States
‡Fairchild Tropical Botanic Garden, 10901 Old Cutler Road, Coral Gables, Florida 33156, United States
S
* Supporting Information
ABSTRACT: Aristolochic acid I (AA-I), naturally occurring
in Aristolochia plants, is a potent nephrotoxin and carcinogen.
Here we report that AA-I suffers hydrogenolysis with loss of
the nitro group by reaction with cysteine or glutathione to give
aristolic acid. Since the reaction can proceed in aqueous
solutions at pH 7.0 and 37 °C, it is inferred that it may also
occur in biological systems and contribute to the nephrotoxic
effects induced by AA-I.
ristolochic acids (AAs) are nitro compounds found in
Aristolochia spp. These compounds are nephrotoxic and
Cys and GSH participate in a number of reactions. For
example, the free thiol group of Cys and GSH can enter into
thiol−disulfide exchange reactions with S−S bridges or can
react with α,β-unsaturated enones to give Michael adducts.4,5 It
is also known that thiol-bearing compounds (propanedithiol,
dithiothreitol, mercaptoethanol, glutathione) reduce aliphatic
and aromatic azides to the corresponding amines.6,7 A new type
of reaction of Cys and GSH is now described.
A
also display mutagenic and carcinogenic properties in both
humans and mammals.1 The chemical reactivity of AAs toward
model biological nucleophiles is being investigated, and we now
describe the reaction of AA-I (1a) with the amino acid cysteine
(Cys) and the tripeptide glutathione (GSH).
The nitro group of AAs can be easily reduced. Thus, AA-I
(1a) can be converted into the corresponding aristolactam (2)
by a variety of reducing agents.2 However, under certain
reduction conditions the nitro group of AA-I (1a) can also
suffer hydrogenolysis, a reaction in which the nitro group is
replaced by hydrogen, to produce the corresponding denitro
compound, aristolic acid (3a). Hydrogenolysis of AA-I (1a) to
aristolic acid (3a) with sodium borohydride or ammonium
sulfide has been reported.2,3 Here we report that the nitro
group of AA-I (1a) can be removed to give aristolic acid (3a)
by treatment with Cys or GSH, under conditions in which the
aristolactam (2) is not formed.
Preliminary experiments showed that aqueous yellow
solutions of AA-I (1a) containing Cys or GSH tend to fade
upon standing at room temperature. This behavior is due to
conversion of AA-I (1a) into aristolic acid (3a), a compound
that does not absorb above 400 nm. HPLC analysis of reaction
mixtures containing AA-I (1a) and Cys or GSH at different
times showed that the concentrations of AA-I progressively
decrease, whereas those of aristolic acid (3a) increase
correspondingly (Figure 1). Side products were not observed
(Figure 1). Complete conversion of AA-I (1a) into aristolic
acid (3a) can be achieved at room temperature, or higher
temperatures, in an aqueous solution containing a sufficient
excess of Cys or GSH at pH 7.0. A nitrogen atmosphere was
used to repress the spontaneous oxidation of Cys to cystine. A
typical experiment illustrating the conversion of AA-I (1a) into
aristolic acid (3a) with Cys is described in the Experimental
Section, and HPLC profiles are shown in Figure 1. GSH is
equivalent to Cys in chemical reactivity toward AA-I. However,
the reaction is slightly slower than that with Cys. The reactions
of Cys and GSH with AA-I are slow as compared with other
reactions reported for Cys, such as those with iodoacetate or α-
methylene γ-lactones.4 In this work aristolic acid (3a) was
Received: November 23, 2012
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
dx.doi.org/10.1021/np300822b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Note
Figure 1. Reaction of AA-I (4.8 mM) and Cys (37.5 mM) in aqueous solution at pH 7.0 and 50 °C. Typical HPLC profiles showing decay of AA-I
(1a) concentrations and increase of aristolic acid (3a) concentrations as a function of the reaction time. About 50% of AA-I was converted into
aristolic acid in 300 min. Under the same conditions but incubating the reaction mixture at 37 °C about 10% of AA-I was converted into aristolic acid
in 300 min.
isolated and characterized by spectrometric methods. This
compound (3a) naturally occurs in Aristolochia plants.8 In
studies performed to evaluate the nephrotoxic potential of AA
analogues with LLC-PK1 cells, aristolic acid (3a) was nontoxic,
indicating that the nitro group is important for the activity and
that this acid would not be involved in the AA-induced
apoptosis of renal tubule cells.9
In order to get information about the mechanism, the AA-I−
Cys reaction was performed in D2O. It was found that a
deuterium atom is introduced at C-10 of aristolic acid, as
1
1
evidenced by H NMR analysis. The H NMR spectrum of
aristolic acid (3a) shows H-9 and H-10 as two doublets (J = 7.6
Hz) centered at δ 8.02 and 8.79, respectively (the relative
positions of H-9 and H-10 resonances were established by
NOE experiments in related denitroAAs).13 When the reaction
is carried out in D2O, the 1H NMR spectrum of compound 3b
was found to be similar to 3a, except for the signals of H-9 and
H-10. The signal of H-10 (doublet, J = 7.6 Hz) was drastically
reduced, whereas the H-9 signal was transformed into a singlet
(accompanied by two small satellite signals; a doublet
corresponding to the 9,10-diH species 3a; Supporting
Information, Figure S1), consistent with the introduction of a
deuterium atom at C-10 of aristolic acid (3b). Signal intensities
showed that the reaction product in the presence of D2O
consisted of 8% of the 9,10-diH species (3a) and 92% of the 9-
H-10-D species (3b). The above findings suggest that this
hydrogenolysis reaction involves direct transfer of two electrons
and a proton, or H−, from the thiol group (SH) of Cys to the
C-10−NO2 position of AA-I, hence resulting in the elimination
of the nitro group to give aristolic acid (3a). If the AA-I−Cys
reaction proceeds through single electron transfer, it may be
assumed that the transfer of the first electron creates a radical-
ion pair, within a solvent cage, that is immediately stabilized by
the second electron (and proton) transfer, so that DTBN
cannot interact with the transient radical species formed during
The reaction of AA-I with Cys or GSH generates aristolic
acid and presumably nitrous acid and an oxidized form of Cys
or GSH. Cystine is not a product from the reaction of AA-I
(1a) with Cys, so disulfides may not be the oxidation products
of Cys and GSH. Cys occurs in various ionic species as a
function of the pH.10 At pH 7 the ionic form H3N+CH-
(CH2SH)COO− is largely predominant, and it could be the
species that reacts with AA-I. Reductive replacement of the
nitro group by hydrogen in aliphatic and aromatic compounds
has been reported.11,12 A mechanism involving electron
transfers and radical species was postulated by Kornblum et
al. (1979)11 for the hydrogenolysis of the nitro group of tertiary
nitroparaffins with methyl mercaptan (CH3SH). A proof in
favor of this free radical chain mechanism is that in the presence
of 10 mol % of di-tert-butylnitroxide (DTBN; free radical
scavenger) the replacement of the nitro group by hydrogen is
completely inhibited.11 However, the AA-I−Cys reaction
proceeds by a different pathway since 10% mol of DTBN
(with respect to AA-I) in the medium does not affect the
formation of aristolic acid (3a).
B
dx.doi.org/10.1021/np300822b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Note
the reaction. Reactions with a mixture of AAs showed that AA-I
(1a) and AA-II (1b) are equally reactive toward Cys and GSH.
In contrast, AA-III (1c) and AA-IV (1d) remain unaffected.
Thus, the reaction seems to be restricted to AA-I (1a) and AA-
II (1b). This result indicates that introduction of a methoxy
group at C-6 of AAs significantly alters the electron distribution
and electrophilic properties of the molecule. There is a
correlation between chemical structure and toxicity since in
studies with cultured renal epithelial cells AA-I (1a) was found
to be the most toxic, followed by AA-II (1b), whereas AA-III
(1c), with a 6-OMe group, was nontoxic.9
The facile replacement of the nitro group of AA-I by
hydrogen by interaction with thiols seems to be restricted to
AAs since related compounds such as 1-nitronaphthalene and
8-nitro-1-naphthoic acid proved to be unreactive toward Cys,
even forcing the reaction at higher temperatures. Consequently,
it is concluded that this AA-I−thiol reaction is also closely
related to the known high reactivity of the C-9−C-10 bond in
phenanthrene.
modifying key proteins or enzymes essential for cell function.
The reaction of AA-I with thiol-containing molecules is slow,
the biological impact of which remains to be determined.
However, the above evidence suggests that it may have some
implication in vivo and that thiol depletion induced by AA-I
could be one of the primary events in the ethiology of AAN.
In summary, a new type of reaction, which occurs under
physiological conditions, is described for Cys and GSH, namely,
the hydrogenolysis of the nitro group in an aromatic molecule.
The AA-I−thiol reaction may serve as a plausible explanation
for the appearance of aristolic acid (3a) and its demethylated
derivative (3c) in urine and feces of AA-I-administered rats and
probably also contributes to the observed depletion of GSH in
AA-I-treated cells. If this AA-I−thiol reaction occurs in vivo and
induces adverse effects, a diet supplemented with thiol-bearing
agents such as N-acetylcysteine could display protective effects
against AAN and other ailments caused by AAs.
EXPERIMENTAL SECTION
■
The nephropathy caused by AAs has been designated
aristolochic acid nephropathy (AAN). Although a number of
mechanisms have been postulated to explain this disease, the
General Experimental Procedures. Aristolochic acids are from
plant origin. Pure AA-I was obtained from a mixture of AAs by
preparative HPLC. Cysteine (C-7755), glutathione (G-4251), and di-
tert-butylnitroxide (300721-1G) were purchased from Sigma. pH
measurements were carried out with a Jenco model 60 portable digital
pH meter provided with a PHR-146 Micro Combination pH electrode
(Lazar Research Laboratories, Los Angeles, CA, USA). NMR spectra
were recorded on a Bruker 400 MHz FT-NMR spectrometer in
DMSO with TMS as internal standard. ESI mass spectra were acquired
on a Thermo Scientific LCQ Deca XP MAX instrument. HPLC
analysis was carried out with a Thermo-Finnigan chromatograph
(Thermo Electron Corporation, San Jose, CA, USA). The chromato-
graph consisted of a SpectraSystem SMC1000 solvent delivery system,
vacuum membrane degasser, P4000 gradient pumps, and AS3000
autosampler. Column effluent was monitored at 226 or 254 nm with a
SpectraSystem UV6000LP variable-wavelength PDA detector and
ChromQuest 4.1 software. Analytical separations were performed
using a C18 RP Hypersil GOLD column (RP5, 250 × 4.6 mm, pore
size 5 μm, Thermo Electron Corporation). The following eluting
system was used: A, MeCN; B, 0.1% TFA in H2O, linear gradient 30%
to 45% A in 60 min (AA-I, tR 22.06 min; aristolic acid, tR 26.88 min),
flow rate 1.0 mL/min at room temperature. Reaction mixtures (20 μL)
were injected after 1:100 dilution in DMSO.
true pathogenic mechanism leading to AAN is still unclear.14
A
number of natural products and synthetic drugs, or their
metabolites, are known to induce cytotoxic lesions through
depletion of intracellular thiols that result in a variety of toxic
effects including nephropathy.15−20 Intracellular thiol depletion
produces cell injury and apoptosis due to generation of reactive
oxygen species (ROS), DNA damage, mitochondrial dysfunc-
tion, and other processes15−20 that have also been observed to
be induced by AA-I,13,21−25 thereby suggesting a connection
between the mechanisms of toxicity produced by thiol-
depleting agents and AA-I. It remains to be determined
whether the AA-I−thiol reaction described in this study takes
place in cells. However, evidence supporting the occurrence of
this reaction in vivo is provided by the fact that aristolic acid
(3a) and its demethylation product 3,4-methylenedioxy-8-
hydroxy-1-phenanthrenecarboxylic acid (3c) were identified in
urine and feces of AA-I-treated rats.26,27 In these experiments
aristolic acid (3a) and its demethylation product (3c) may
account for up to 8% of the dose in the urine and feces
collected over 72 h after administration of AA-I.26 These results
suggest that the metabolites 3a and 3c may arise from the slow
interaction of AA-I with intracellular thiol residues. In vitro
studies using mice and human cells as models also showed that
thiol depletion can be triggered by AA-I.14,25 Thus, it was found
that AA-I depletes intracellular GSH in human HL-60 and mice
C3H/He cells.14,25 GSH is an intracellular reducing agent
whose primary function may be to maintain the SH groups of
proteins and also plays a role in antioxidant defense, protecting
the cell against reactive electrophilic xenobiotics.28,29 Depletion
of cellular GSH can impair cellular defenses against ROS or
toxic compounds and lead to cellular injury followed by
apoptosis and necrosis.30 Decreased GSH levels in human HL-
60 and mice C3H/He cells exposed to AA-I are believed to be
responsible for the generation of ROS and apoptosis of renal
tubular cells.14,25 The reaction of AA-I with GSH described
herein may contribute to the total GSH depletion induced by
AA-I. The high concentration of GSH in animal cells (0.5−10
mM)30 together with the observed accumulation of AA-I in
kidney31,32 may favor the AA-I−GSH interaction so that it may
become significant. In addition to the possible depletion of
GSH, AA-I can also interact with Cys residues of proteins, thus
3,4-Methylenedioxy-8-methoxy-1-phenanthrenecarboxylic
Acid, Aristolic Acid (3a). AA-I (ca. 6 mg, 18 μmol) was suspended in
H2O (1 mL). The suspension was treated with NaHCO3 (ca. 20 mg)
and heated to dissolve the acid, and the solution cooled at room
temperature. Some turbidity was removed by centrifugation. The exact
amount of AA-I in the solution was determined by HPLC analysis
versus a calibration curve. Cysteine (31 mg, ca. 250 μmol) was added
to the solution, and the pH was adjusted to 7.0 with 8.5% phosphoric
acid. The reaction mixture was purged with N2 to remove oxygen. The
yellow solution was heated at 37 °C overnight with magnetic stirring.
Analysis of the reaction mixture by HPLC showed a predominant peak
of aristolic acid (3a) and traces of unreacted AA-I (1). The reaction
mixture was centrifuged to remove some insoluble material. The
supernatant was diluted (×2) with H2O and loaded to a C18 cartridge
(SepPack Vac C18 cartridge, 500 mg, WAT036905, Waters;
conditioned with MeOH and H2O). The cartridge was washed with
H2O (4 mL) and 1% HOAc in CH3CN/H2O 4:6 (4 mL). Aristolic
acid was eluted with 1% HOAc in CH3CN/H2O (7:3, 4 mL). The
eluate containing aristolic acid was extracted with EtOAc. The EtOAc
phase was evaporated to dryness to give aristolic acid, 4 mg: UV λmax
255, 296, 319 sh, 327, 355, 375 nm; 1H NMR δ ppm (DMSO-d6) 4.00
(3H, s, 8-OCH3), 6.39 (2H, s, −OCH2O−), 7.22 (1H, d, J7,6 = 8.0 Hz,
7-H), 7.61 (1H, t, J6,5 = J6,7 = 8.2 Hz, 6-H), 7.80 (1H, s, 2-H), 8.02
(1H, d, J9,10 = 7.6 Hz, 9-H), 8.64 (1H, d, J5,6 = 8.4 Hz, 5-H), and 8.79
(1H, d, J10,9 = 7.6 Hz, 10-H) (coherent with Mukhopadhyay et al.,
1983);2 13C NMR δ ppm (DMSO-d6) 168.4 (CO), 154.7 (C-8),
C
dx.doi.org/10.1021/np300822b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Note
145.6 (C-4), 144.3 (C-3), 128.4 (C-4b), 128.0 (C-10a), 127.1 (C-6),
123.9 (C-10), 121.8 (C-8a), 119.5 (C-1), 119.1 (C-9), 118.7 (C-5),
115.5 (C-4a), 111.7 (C-2), 107.3 (C-7), 102.5 (OCH2O), 55.7 (OMe)
(assignments were based on ref 33); ESIMS (MeOH) positive mode
297 [M + H]+; 279 [M + H − H2O]+.
(19) Wullner, U.; Seyfried, J.; Groscurth, P.; Beinroth, S.; Winter, S.;
Gleichmann, N.; Heneka, M.; Loschmann, P.-A.; Schultz, J. B.; Weller,
M.; Klockgether, T. Brain Res. 1999, 826, 53−62.
(20) Tukov, F. F.; Anand, S.; Gadepalli, R. S. V. S.; Gunatilaka, A. A.
L.; Matthews, J. C.; Rimoldi, J. M. Chem. Res. Toxicol. 2004, 17, 1170−
1176.
(21) Qi, X.; Cai, Y.; Gong, L.; Liu, L.; Chen, F.; Xiao, Y.; Wu, F.; Li,
Y.; Xue, X.; Ren, J. Toxicol. Appl. Pharmacol. 2007, 222, 105−110.
(22) Wang, W.; Zhang, J. Eur. J. Pharmacol. 2008, 588, 135−140.
(23) Liu, Q.; Wang, Q.; Yang, X.; Shen, X.; Zhang, B. Toxicol. Lett.
2009, 184, 5−12.
ASSOCIATED CONTENT
■
S
* Supporting Information
1H NMR spectrum of deuterated aristolic acid (1b). This
information is available free of charge via the Internet at http://
pubs.acs.org.
(24) Yang, H.; Dou, Y.; Zheng, X.; Tan, Y.; Cheng, J.; Li, L.; Du, Y.;
Zhu, D.; Lou, Y. Toxicology 2011, 287, 38−45.
(25) Yu, F.-Y; Wu, T.-S.; Chen, T.-W.; Liu, B.-H. Toxicol. in Vitro
2011, 25, 810−816.
AUTHOR INFORMATION
■
(26) Krumbiegel, G.; Hallensleben, J.; Mennicke, W. H.; Rittmann,
N.; Roth, H. J. Xenobiotica 1987, 17, 981−991.
Corresponding Author
*Phone: 305-348-0375. Fax: 305-348-1986. E-mail: priestap@
fiu.edu.
(27) Chan, W.; Cui, L.; Xu, G.; Cai, Z. Rapid Commun. Mass
Spectrom. 2006, 20, 1755−1760.
(28) Sen, C. K. J. Nutr. Biochem. 1997, 8, 660−672.
(29) Lu, S. C. FASEB J. 1999, 13, 1169−1183.
Notes
The authors declare no competing financial interest.
(30) Reed, J. D.; Fariss, M. W. Pharmacol. Rev. 1984, 36, 25S−33S.
(31) Xiao, I.; Ge, M.; Xue, X.; Wang, C.; Wang, H.; Wu, X.; Li, L.;
Liu, L.; Qi, X.; Zhang, Y.; Li, Y.; Lou, H.; Xie, T.; Gu, J.; Ren, G.
Kidney Int. 2008, 73, 1231−1239.
ACKNOWLEDGMENTS
■
We are grateful to Y. L. Hsu and Y. Song (Florida International
University) for recording NMR and ESIMS spectra. We
specially thank Dr. S. Rose for her support (School of
Integrated Science and Humanity, Florida International
University).
(32) Priestap, H. A.; Torres, M. C.; Rieger, R. A.; Dickman, K. G.;
Freshwater, T.; Taft, D. R.; Barbieri, M. A.; Iden, C. R. Chem. Res.
Toxicol. 2011, 25, 130−139.
(33) Achari, B.; Bandyopadhyay, S.; Chakravarty, A. K.; Pakrashi, S.
Org. Magn. Reson. 1984, 22, 741−746.
REFERENCES
■
(1) Arlt, V. M.; Stiborova, M.; Schmeiser, H. H. Mutagenesis 2002,
17, 265−277.
(2) Mukhopadhyay, S.; Funayama, S.; Cordell, G. A.; Fong, H. H. J.
Nat. Prod. 1983, 46, 507−509.
(3) Krumbiegel, G.; Roth, H. J. Arch. Pharm. 1987, 320, 264−270.
(4) Kupchan, M.; Fessler, D. C.; Eakin, A.; Giacobbe, T. J. Science
1970, 168, 376−378.
(5) Baker, L. M. S.; Baker, P. R. S.; Golin-Bisello, F.; Schopfer, F. J.;
Fink, M.; Woodcock, S. R.; Branchaud, B. P.; Radi, R.; Freeman, B. A.
J. Biol. Chem. 2007, 282, 31085−31093.
(6) Bayley, H.; Standring, D. N.; Knowles, J. R. Tetrahedron Lett.
1978, 3633−3634.
(7) Handlon, A. L.; Oppenheimer, N. J. Pharm. Res. 1988, 5, 297−
299.
(8) Pakrashi, S. C.; Ghosh-Dastidar, S. B.; Basu, S.; Achari, B.
Phytochemistry 1977, 16, 1103−1104.
(9) Balachandran, P.; Wei, F. R.; Lin, R.; Khan, I. A.; Pasco, D. S.
Kidney Int. 2005, 67, 1797−1805.
(10) Dixon, H. B. F.; Tipton, K. F. Biochem. J. 1973, 133, 837−842.
(11) Kornblum, N.; Carlson, S. C.; Smith, R. G. J. Am. Chem. Soc.
1979, 101, 647−657.
(12) Severin, T.; Schmitz, R.; Temme, H.-L. Chem. Ber. 1963, 96,
2499−2503.
(13) Wu, T.-S.; Leu, Y.-L.; Chan, Y.-Y. Chem. Pharm. Bull. 1998, 46,
1301−1302.
(14) Li, Y.-C.; Tsai, S.-H.; Chen, S.-M.; Chang, Y.-M.; Huang, T.-C.;
Huang, Y.-P.; Chang, C.-T.; Lee, J.-A. Biophys. Biochem. Res. Commun.
2012, 423, 832−837.
(15) Benesic, A.; Schwerdt, G.; Freudinger, R.; Mildenberger, S.;
Groezinger, F.; Wollny, B.; Kirchhoff, A.; Gekle, M. Kidney Blood
Pressure Res. 2006, 29, 280−293.
(16) Yao, X.; Panichpisal, K.; Kurtzman, N.; Nugent, K. Am. J. Med.
Sci. 2007, 334, 115−124.
(17) Yang, C. F.; Shen, H. M.; Ong, C. M. Arch. Biochem. Biophys.
2000, 380, 319−330.
(18) Seyfried, J.; Soldner, F.; Schultz, J. B.; Klockgether, T.; Kovar, K.
A.; Wullner, U. Neurosci. Lett. 1999, 264, 1−4.
D
dx.doi.org/10.1021/np300822b | J. Nat. Prod. XXXX, XXX, XXX−XXX