(CANCER RESEARCH 56, 2025-2028,
May I. I996|
Advances in Brief
Detection of DNA Adducts Formed by Aristolochic Acid in Renal Tissue from
Patients with Chinese Herbs Nephropathy
Heinz H. Schmeiser,1 Christian A. Bieter, Manfred Wiessler, Charles van Ypersele de Strihou, and
Jean-Pierre Cosyns
Division of Molecular Toxicology. German Cancer Research Center. Im Neuenheimer Feld 280. 69120 Heidelberg. Germany IH.H.S.. C. A. B., M. W.J, and Departments of
Pathology [J. P. C.¡and Renal Diseases (C. v. Y. d. S.), University of Louvain Medical School, Cliniques Universitaires Si-Luc, 10 Avenue Hippocrate, 1200 Bruxelles. Belgium
Abstract
A unique type of rapidly progressive renal fibrosis, designated Chinese
herbs nephropathy (CHN), has been described in young Belgian women
who had followed a slimming regimen including recently introduced
Chinese herbs (Stephania tetrandra and Magnolia officinalis). Aristolochic
acid (AA), a known nephrotoxin and carcinogen, was suspected as its
causal factor. To substantiate this hypothesis, renal tissue from five pa
tients with CHN and six patients with other renal diseases was analyzed
for the presence of AA-derived DNA adducts, a described biomarker of
AA exposure associated with its carcinogenic and mutagenic activity.
Using the J2P-postlabeling method, a major distinct DNA adduct spot was
found in all five cases of CHN and identified by cochromatographic
analyses with authentic markers as the deoxyadenosine adduct of AA-I
[7-(deoxyadenosin-/V6-yl)-aristolactam
I], the major component of the
plant extract AA. This DNA adduct was absent in the six control cases.
The 7-(deoxyadenosin-/V*-yl)-aristolactam
I adduct levels in CHN ranged
from 0.7 to 5.3/107 nucleotides. Our data demonstrate that AA is impli
cated in CHN. They suggest a mechanism for the urothelial atypia and
cancers observed in this disease and raise the possibility that a DNA
mutation is responsible for the kidney-destructive Hbrotic process.
Materials and Methods
Nephroureterectomy Specimens. Seven native kidneys were removed at
the time of transplantation or subsequently in five patients with documented
CHN. Samples were analyzed by light and immunofluorescent microscopy as
described previously (9). In addition, corticomedullary (cases 1 and 2) and/or
cortical tissue samples (cases 1-5) were frozen at -20°C for DNA extraction.
Similarly, corticomedullary
samples were obtained from six end-stage kidneys
removed from six patients at the time of transplantation. Our diagnosis in this
control group included lupus nephritis, membranoproliferative
glomerulonephritis type 1, chronic interstitial nephritis, reflux nephropathy, cystic nephrop
athy, and nephrosclerosis (one case each).
Identification of DNA Adducts. The method has been described else
where (12). Briefly, DNA was isolated from kidney tissue by the phenol
extraction method. DNA samples (12.5 /¿g)were digested enzymatically to
nucleoside 3'-phosphates and enriched for adducts by nuclease PI digestion
(13). The digests were then analyzed by 12P-postlabeling. Nuclease Pi-resist
ant adducts were 5' labeled with [<2P]orthophosphate, separated by PEIcellulose TLC, and detected by autoradiography; autoradiography was per
formed at 23°Cfor 6 h. Adducts were measured by assay of their 32P content
Introduction
CHN2 nephropathy
indicate that AA is implicated in CHN and provide an explanation for
the urothelial atypias and transitional cell carcinomas reported in
patients with CHN (9, 10, 11).
is a recently described,
rapidly progressive
interstitial nephritis associated with the intake of Chinese herbs during
a slimming cure ( 1). The nature of the postulated toxic agent, how
ever, remained in doubt. The observation that one of the prescribed
Chinese herbs (Stephania tetrandra) had probably been replaced by
another one (Aristolochia fangchi', Ref. 2) led to the hypothesis that
AA might be the causative agent, despite the fact that it could not be
identified in the original pills ( 1). The active principle of herbal drugs
derived from Aristolochia species is AA. AA is a mixture of struc
turally related nitrophenanthrene carboxylic acids, with AA-1 [8-methoxy-6-nitrophenanthro-(3,4-d)-l
,3-dioxolo-5-carboxylic acid] being
the major component.
In 1964, AA was shown to be nephrotoxic in humans (3), and in
1982, it was shown to be a strong carcinogen in rodents (4). Recently,
we demonstrated that its carcinogenic and mutagenic effects were
associated with the formation of AA DNA adducts (5, 6). We have
characterized the major adducts and used the 32P-postlabeling method
(12). Their tissue levels were determined in at least three separate experiments
and expressed as relative adduct labeling values, as described previously (13.
14).
The comparison of different Chromatographie mobilities on PEI-cellulose or
on reversed-phase HPLC of single 12P-postlabeled adducts obtained in vivo
with those of in wiro-synthesized reference compounds was used for identi
fication purposes. Individual spots detected by the 32P-postlabeling assay were
excised from thin-layer plates and extracted as described (12).
Results
Patients. The characteristics of patients with CHN are described in
Table 1. All were female, aged 27-42 years. Medical histories prior to
attending the slimming clinic were unremarkable. They had taken the
Chinese herbs-containing pills (formula 2 regimen as described in the
original publication) for 13-23 months. Renal transplantation was per
formed after an average of 5 (range, 2-10) months of hemodialysis in
four patients (three cadavers and one living donor grafts) and prior to
hemodialysis in one patient (living donor graft). The delay between the
to detect and identify them (7, 8). In the present study, we have taken
end of the slimming cure and renal transplantation ranged from 9 to 27
advantage of this method and demonstrate the presence of AA DNA
months. Pathological features have been reported extensively (9) for the
adducts in the kidney tissue of five patients with CHN. These results
first three patients. The kidneys of the two last patients exhibited the same
characteristics, including extensive hypocellular interstitial fibrosis, tubu
Received 2/28/96; accepted 3/25/96.
lar atrophy, global glomerular sclerosis with a corticomedullary gradient,
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance with
and fibrosis of mainly interlobular arteries. Extensive atypia and atypical
18 U.S.C. Section 1734 solely to indicate this fact.
hyperplasia of the epithelial lining of the collecting ducts, calyces, pelves,
1 To whom requests for reprints should be addressed, at Division of Molecular
and ureter were also present. Additional findings included: in case 4,
Toxicology. German Cancer Research Center. Im Neuenheimer Feld 280. 69120 Heidel
berg. Germany.
diffuse, intense granular mesangial deposits of IgA; and in case 5,
2 The abbreviations used are: CHN, Chinese herbs nephropathy: AA, aristolochic acid:
numerous corticomedullary cysts ranging from 1 to 15 mm in diameter,
PEI, polyelhylenimine:
HPLC, high-performance
liquid chromatography; dA-AA-1.
7-(deoxyadenosin-/v*-yl)-arislolactam
I.
typical of early polycystic disease.
2025
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1996 American Association for Cancer Research.
DNA ADDUCTS
OF ARISTOLOCHIC
ACID IN HUMAN KIDNEYS
Table I Relevant clinical data and relative adduci labeling fRAL) values of dA-AA-l in kidnev tissue of patients with CHN
in renal tissue
(mean ±SD/IO7
nucleolides/'kidney
Duration
offormili;)
"^Case12345Age
tnRAL
(yrVregimenSex
(months)32/F28/F27/F42/F42/F"
HI)"(months)041022s
removal(months)RightLeft,Right.Left,Right.Right.Right,9172015271811Cortical1.72.35.34.13.02.50.91.23.22.71.82.10.7
2.42.6
±
2.03.4
±
±2.1n.d.<n.d.n.d.n.d.VariaBladder
TCC*Bilateral
type¡nations
pretransplantation'
PreTP HI),
duration.Intervalfrom»nil
detenu1
Mean of at least three
experiments.transitional
in separate
n.d..•'TCCnot
determined.1913202123hemodialysPreTP
0.5Cortico-mcdullary3.5
TCCIgA ureter.
nephropathyPolycystic
kidney disease, adult
cell carcinoma.
This conclusion hinges on the reliability of the method used to
identify the DNA adducts. The 32P-postlabeling method has proven
Table 2 Cimimi subjects
HD"(months)51
DiagnosisLupus
extremely sensitive and specific to detect lipophilic DNA adducts
(13). The nuclease PI modification of the 12P-postlabeling assay is
(yr)/Sex14/F
nephritis
PCIN'
M PON
12/F27/F
07
PN''Cystic
41/F10/F
117
nephropathy
NephrosclerosisAge
38/MPreTP
" PreTP
TP HD.
HI), pretransplantation hemodialysis duration.
"GN I.
I. mcmnranopronierauve
memhranoproliferative giomeruio
glomerulonephritis type I.
MPGN
' CIN, chronic idiopathic interstitial nephritis
'' PN. pyelonephritis due to lithiasis.
12
The characteristics of the control patients are given in Table 2. They
include four females and one male aged 10-41 years. Renal transplan
tation was performed after an average of 18 (range, 7-51) months of
hemodialysis in five patients and prior to hemodialysis in one patient.
DNA Adduci Detection and Identification. Using the nuclease
Pi-enhanced variation of the '"P-postlaheling assay, one major (spot
highly sensitive for the recognition of DNA adducts formed in vivo by
AA-I and -II (12). The level of these specific DNA adducts in tissues
or body fluids provides an internal dosimeter of individual AA expo
sure. All CHN samples showed a distinctive adduci spot characteristic
of the deoxyadenosine adduci of AAI (dA-AA-I). Cochromatography
in different independent systems with authentic markers confirmed
the identity of spot 1. Because no such spots were found in control
kidneys removed at the time of transplantation from six patients with
renal failure of other causes, cross-contamination or a nonspecific
effect per se can be excluded.
1) and several minor spots of DNA adducts were observed in all renal
samples obtained from the five patients with CHN (Fig. 1). To obtain
reference compounds for the identification of AA DNA adducts by
cochromatography.
the deoxyadenosine
(dA-AA-I) and deoxyguanosine adducts |7-(deoxyguanosin-/V~-yl)-aristolactam
11of AA-I
were prepared by /';; vitro incubation, as described previously ( 12). By
cochromatographic analysis on PEI-cellulose TLC plates (data not
shown) and reversed-phase HPLC (Fig. 2) with those reference com
pounds, we found that the spots in Fig. 1 were chromatographically
indistinguishable from dA-AA-I and thus assigned as 3'.5'-bisphospho-7-(deoxyadenosin-/V*'-yl)-aristolactam
I.
It is clear from Fig. 2 that spot 1 from Fig. 1«
eluted with a retention
time of 24 min (Fig. 2h), identical to the dA-AA-I standard (Fig. 2a).
The identity of the two spots was also confirmed by cochromato
graphic HPLC analysis, shown in Fig. 2c. Equal amounts of radioac
tivity obtained from spot 1 (Fig. la) and the dA-AA-I standard were
mixed prior to analysis and found to elute as a single peak.
The levels of this adduci in DNA samples ranged from 0.7 to
5.3/107 nucleotides and were similar in cortical and corticomedullary
f
samples (Table 1).
By contrast, DNA isolated from the control kidneys was virtually
free of adducts in the area where AA-derived spots were located
(Fig. I/).
Discussion
These results conclusively demonstrate for the first time that CHN
is associated with the intake of AA. Furthermore, they provide a
physiopathological clue as to the cause of the urothelial atypias
observed in several patients and of the transitional cell carcinoma of
the bladder observed in patient 3.
Fig. 1. Autoradiographs of two-dimensional ehromatograms of '-P-laheled DNA
adducts from human kidneys. («-<•).
patients with CHN: a. case 1: b, case 2; c, case 3; d.
case 4; e. case 5./ control patient with lupus nephritis.
2026
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1996 American Association for Cancer Research.
DNA ADDUCTS
OF ARISTOLOCHIC
ACID IN HUMAN
KIDNEYS
probably explains why dA-AA-I was still detectable in the kidneys of
patients with CHN 9-27 months after the interruption of formula 2
ijcp.
pills. The persistence of such specific adducts suggests nonreparable
lesions in a variety of tissues exposed to carcinogens (17).
The specific adduci levels of 0.7-5.3/107 nucleotides found in this
study are approximately 10 limes higher than the levels of bulky DNA
adducts in humans reported by others (18, 19). They are similar to
those reported in DNA from forestomachs of rals irealed wiih AA at
a dose of 10 mg/kg body weight twice weekly for 2 weeks (12). Both
interindividual variabilily in Ihe melabolism of AA and balch-to-batch
5
IB
1C
1C
IE
it
2S
25
25
3B
3e
35
46
differences of AA conlenls in Ihe capsules are major faclors influenc
ing adduci levels. Hence, a correlalion belween adduci levels and the
duration of capsule intake is unlikely and was indeed noi found.
Specific DNA adducls within the genomic DNA have an important
biological significance as premutagenic lesions. This has been verified
by us for dA-AA-I adducts; UÃ -ras proto-oncogenes are activaled
with high frequency by a A —¿Â»
T transversion mulalion in codon 61
from CAA to CTA in the DNA of AAI-induced foreslomach carci
«Inni
nomas in rals (5).
The role of Ihe dA-AA-I adducls in AA-induced mulagenesis and
carcinogenesis appears relevanl and might be responsible for Ihe
alypia observed in Ihe urothelia of palienls with CHN (9) as well as
for the multifocal iransilional cell carcinomas observed in iwo of them
(10, 11).
Finally, the possibility lhat Ihe AAI DNA adduci also initiated the
renal fibrolic process lhal destroyed Ihe kidney remains an inlriguing
possibility. The faci lhal the fibrotic process was noi confined lo the
kidney but also extended to ihe pelvis and ureier in one palienl (9) and
conlribuled perhaps lo ihe occurrence of aortic valvular lesions in
40% of ihe patients with CHN3 supports a generalized effect of the
35
C
5
1C
19
2»
29
3>
39
«
«!:«.-:
minutes
Fig. 2. HPLC chromatograms showing comigration of radioactive peaks of '2P-labeled
adduci«excised from thin-layer plates and extracted as described (10). a, dA-AA-1
obtained from AAI activated by xanthine oxidase in the presence of 3'-phosphodeoxyadenosine; b, spot 1 from Fig. \a, obtained from DNA digest of cortical tissue of case 1;
c. equal radioactive amounts of a and h mixed prior to HPLC analysis. Chromatography
was performed on a phenyl-modified. reversed-phase HPLC column, as described previ
ously (10).
ingested toxic compound. The amount of ingesied AA calculaled on
Ihe basis of Ihe batch analyses (2) is minimal when compared with
thai given experimentally lo rals (12) and in ihe range prescribed
wilhoul untoward effecls in iradilional Chinese medicine (20). Still,
the faci lhal such quanlilies are capable of allering DNA lo a significanl extent raises the possibility that AA has a toxic effecl synergislic
with some of the olher compounds included in ihe slimming cure (15).
Altogelher, we demonslrale ihe presence of AA melabolites as
DNA adducts in patienls with CHN, thus establishing thai CHN is
indeed associaled wiih the presence of AA. This observation further
suggesls lhal dA-AA-I DNA adducls are responsible for ihe urothelial
cancers observed in CHN. Therefore, patients with CHN should
undergo regular follow-ups for urothelial cancer.
References
The role of AA in the genesis of CHN is still under debate (15).
Although the evidence that the prescribed herb, S. tetrandra, was
replaced by another is compelling, because the Stephania alkaloid
tetrandrine was not present in the pills used (1), AA was not initially
identified in the Chinese herbs contained in the pills. Subsequent
analysis of the batches of Chinese herbs imported in Belgium between
July 1990 and August 1992 under the name of 5. tetrandra disclosed
variable amounts of AA. The failure to detect AA in the pills was then
ascribed to an inadequate extraction procedure (2). The discovery of
AA adducts in the kidneys of five patients with CHN conclusively
demonstrates that AA had indeed been ingested in amounts sufficient
to alter cellular DNA.
The natural extract of aristolochia plants consists of structurally
related nitrophenanthrene carboxylic acids, AA-I and -II being the
major components. In rats given AA-I or the natural extract AA, the
predominant adduci found in all organs analyzed is dA-AA-I (12). It
persists apparently throughout the animals' lives (16), a feature that
1. Vanherweghem. J. L.. Depierreux. M.. Tielemans, C. Abramowicz. D.. Dratwa. M..
Jadoul. M.. Richard. C.. Vanderwelde. D.. Verbeelen. D., Vanhaelen-Fastre. R.. and
Vanhaelen. M. Rapidly progressive interstitial renal fibrosis in young women: asso
ciation with a slimming regimen including Chinese herbs. Lancet, 341: 387-391,
1993.
2. Vanhaelen. M.. Vanhaelen-Fastre. R.. But. P.. and Vanherweghem. J. L. Identification
of aristolochic acid in Chinese herbs. Lancet, 343: 174, 1994.
3. Jackson. L., Kofman. S.. Weiss. A., and Brodovsky. H. Aristolochic acid (NSC50413): Phase 1 clinical study. Cancer Chemother. Rep., 42: 35-37, 1964.
4. Mengs, U.. Lang. W., and Poch, J. A. The carcinogenic action of aristolochic acid in
rats. Arch. Toxicol., 61: 107-119. 1982.
5. Schmeiser. H. H.. Janssen, J. W. G., Lyons, J.. Scherf. H. R.. Pfau. W., Buchmann,
A.. Bartram. C. R., and Wiessler. M. Aristolochic acid activates ras genes in rat
tumors at deoxyadenosine residues. Cancer Res.. 50: 5464-5469. 1990.
6. Broschard. T. H., Wiessler, M., von der Lieth, C. W., and Schmeiser. H. H. Translesional synthesis on DNA templates containing site-specifically placed deoxyade
nosine and deoxyguanosine adducts formed by the plant carcinogen aristolochic acid.
Carcinogenesis (Lond.), /5: 2331-2340. 1994.
7. Schmeiser. H. H.. Schoepe, K. B., and Wiessler. M. DNA adduci formation of
aristolochic acid I and II in vitra and in vim. Carcinogenesis (Lond.), 9: 297-303,
1988.
3 Unpublished observations.
2027
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1996 American Association for Cancer Research.
DNA ADDL'CTS OF ARISTOLOCHIC
8. Pfau, W., Schmeiser, H. H., and Wiessler, M. Arislolochic acid binds covalently to
Cheexocyclic amino group of purine nucleolides in DNA. Carcinogenesis (Lond.), II:
313-319, 1990.
9. Cosyns, J. P., Jadoul. M. Squifflet. J. P.. De Plaen. J. F., and van Ypersele de Strihou.
C. Chinese herbs nephropathy: a clue to Balkan endemic nephropathy? Kidney Int.,
45: 1680-1688. 1994.
10. Cosyns, J. P.. Jadoul. M.. Squifflel. J. P., and van Ypersele de Strihou, C. Urothelial
malignancy in nephropathy due to Chinese herbs. Lancet. 344: 188. 1994.
11. Vanherweghem. J. L.. Tielemans. C.. Simon. J., and Depierreux, M. Chinese herbs
nephropathy and renal pelvic carcinoma. Nephrol. Dial. Transplant.. 10: 270-273. 1995.
12. Stiborova, M., Fernando, R. C., Schmeiser, H. H., Frei, E.. Pfau, W., and Wiessler,
M. Characterization of DNA adducts formed by arislolochic acids in the target organ
(forestomach) of rats by 32P-postlabelling analysis using different Chromatographie
procedures. Carcinogenesis (Lond.), 15: 1187-1192. 1994.
13. Reddy. M. V.. and Randerath. K. Nuclease PI-mediated enhancement of sensitivity
of 32P-postlabelling test for structurally diverse DNA adducts. Carcinogenesis
(Lond.). 7: 1543-1551, 1986.
14. Randerath, K., Reddy, M. V., and Gupta. R. C. "P-labelling test for DNA damage.
Proc. Nati. Acad. Sci. USA, 78: 6126-6129, 1981.
ACID IN HUMAN KIDNEYS
15. van Ypersele
Chinese herbs
16. Fernando, R.
persistence of
17.
18.
19.
20.
de Strihou, C., and Vanherweghem. J. L. The tragic paradigm of
nephropathy. Nephrol. Dial. Transplant.. 10: 157-160. 1995.
C., Schmeiser. H. H.. Schert'. H., and Wiessler. M. Formation and
specific purine DNA adducts by 12P-postlabelling in target and non-
target organs of rats treated with aristolochic acid I. In: D. H. Phillips, M. Castegnaro.
and H. Bartsch (eds.l. Postlabeling Methods for Detection of DNA Adducts, pp
167-171. 1ARC Scientific Pubi. No. 124. Lyon. France: IARC. 1993.
Moorthy. B.. Chen. S., Li, D., and Randerath. K. 3-Methylcholanthrene-inducible
liver cytochrome(s) P450 in female Sprague-Dawley rats: possible link between P450
turnover and formation of DNA adducts and I-compounds. Carcinogenesis (Lond.),
14: 879-886, 1993.
Phillips, D. H.. Hewer, A.. Malcolm. A. D. B., Ward, P.. and Coleman. D. V.
Smoking and DNA damage in cervical cells. Lancet, 335: 417. 1990.
Talaska. G., al .1ninni. A. Z.. and Kadlubar. F. F. Smoking related carcinogen-DNA
adducts in biopsy samples of the urinary bladder: identification of W-ideoxyguanosin8-yl)-4-aminobiphenyl as a major product. Proc. Nati. Acad. Sci. USA, 88: 53505354, 1991.
Kee Chang Huang. The Pharmacology of Chinese Herbs. Boca Raton. FL: CRC
Press. Inc.. 1993.
2028
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1996 American Association for Cancer Research.
Detection of DNA Adducts Formed by Aristolochic Acid in Renal
Tissue from Patients with Chinese Herbs Nephropathy
Heinz H. Schmeiser, Christian A. Bieler, Manfred Wiessler, et al.
Cancer Res 1996;56:2025-2028.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/56/9/2025
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1996 American Association for Cancer Research.