Portal de Periódicos da CAPES

Is aristolochic acid a risk factor for Balkan endemic nephropathy‐associated urothelial cancer?

2002; Wiley; Volume: 101; Issue: 5 Linguagem: Inglês

10.1002/ijc.10602

1097-0215

Volker M. Arlt, Dušan Ferluga, Marie Stiborová, Annie Pfohl‐Leszkowicz, Mato Vukelić, Stjepan Čeović, Heinz H. Schmeiser, Jean‐Pierre Cosyns,

Heavy Metals in Plants

Chinese herbs nephropathy (CHN), associated with the ingestion of herbal remedies containing aristolochic acid (AA), was first reported in young Belgian women who have been on a slimming regimen including the Chinese herb Aristolochia fangchi.1 CHN is a unique rapidly progressive nephropathy characterized by extensive renal interstitial fibrosis, tubular proteinuria, early and severe anemia and a high risk of urothelial cancer.2, 3, 4 AA is a mixture of structurally related nitrophenanthrene carboxylic acids with aristolochic acid I (AAI) being the major component. AA is nephrotoxic in several species, mutagenic in bacteria and carcinogenic in rodents.1 The detection of specific AA-DNA adducts by 32P-postlabeling in kidney and ureter tissue of CHN patients unambiguously demonstrated a causative role of AA in CHN.4, 5, 6, 7 CHN and urothelial tumors have recently been reported in patients who were exposed to Aristolochia species and had no relationship with the Belgian slimming clinic.8, 9 These reports demonstrate that the development of CHN lesions may be ascribed to the toxicity of AA alone without requiring the other drugs prescribed in the slimming regimen. Moreover, the recent demonstration of renal interstitial fibrosis and urothelial malignancy in rabbits treated with AA alone, removed any doubt on the causal role of AA in CHN.10 It has been proposed to designate the interstitial nephropathy in which the unequivocal role of AA has been fully documented as aristolochic acid nephropathy (AAN).1, 9 On both clinical and morphological grounds, AAN is very similar to another fibrosing nephropathy, the Balkan endemic nephropathy (BEN), which is found in certain rural areas of Rumania, Croatia, Bosnia, Serbia and Bulgaria along the Danube river basin.2, 11, 12 Both diseases share normal blood pressure, aseptic leukocyturia, early and severe anemia on clinical ground and morphologically extensive hypocellular interstitial sclerosis, tubular atrophy, global sclerosis of glomeruli, cellular atypia and malignant transformation of the urothelium.2-4, 11, 12 These similarities have led to the hypothesis of a common etiological agent for both diseases. In this context it is noteworthy that food contamination by AA has been suggested as a possible causal factor in BEN in 1970.13 Alternatively evidence has accumulated that BEN is an environmentally induced disease strongly associated with the oral intake of food from plant origin contaminated with the mycotoxin ochratoxin A (OTA).11, 12 Like AA, OTA is nephrotoxic and carcinogenic. Similarly to DNA adduct formation by AA found in AAN, high levels of OTA-related DNA adducts have been found in urinary tract tumors of Bulgarian patients suffering from BEN supporting the hypothesis that OTA is a causal factor in the development of BEN.14 Thus, OTA-related and AA-DNA adducts are suitable biomarkers for elucidating the molecular epidemiology of both diseases. We investigated recently the potential role of OTA exposure in AAN.4, 7 A major role of OTA in AAN was excluded. In this pilot study we want to assess AA exposure in endemic areas for BEN. Therefore, we analyzed kidney tissue for both OTA-related and AA-DNA adducts by 32P-postlabeling from 3 female patients from endemic regions (Table I). A unilateral nephroureterectomy had been carried out for pyeloureteral urothelial malignancy (Cases 1 and 2) and for ureteral stenosis (Case 3). The unavailability of sufficient clinical and renal morphological data did not allow us to classify these patients as clearly suffering from BEN. It is of interest to note, however, that they lived in endemic villages near Slavonski Brod (Croatia) where they had a farming activity, and that 2 of them had developed an upper urinary tract malignancy. Tissues were stored at −80°C until analysis. For the detection of AA-DNA adducts we used the 32P-postlabeling assay as described previously.6 As shown in Figure 1, we detected 1 major adduct spot (Spot A1) in Patients 1 and 3, chromatographically indistinguishable from the adenosine adduct of AAI, dA-AAI [7-(deoxyadenosine-N6-yl)-aristolactam I], the most abundant AA-DNA adduct found in AAN patients (compare Fig. 1a).4, 5, 6, 7, 8, 9 In renal tissue of Patient 2 no AA-DNA adducts were identified (Fig. 1c). As a second independent chromatographic procedure to confirm the identity of the adduct spots we employed reversed-phase HPLC analysis.6 To obtain a reference compound for the identification of the adduct dA-AAI was prepared by in vitro incubation.6 Co-chromatographic analysis on HPLC with the reference compound showed that the adduct spot A1 in Figure 1 were indistinguishable chromatographically from dA-AAI and eluted with a retention time of 23.45 min, identical to the dA-AAI standard. Thus, the spot was assigned as 3′,5′-bisphospho-7-(deoxyadenosine-N6-yl)-aristolactam I. Quantitative analysis showed that adduct levels of the dA-AAI adduct were 17.1 and 5.6 adducts per 109 nucleotides in Patients 1 and 3, respectively (Table I). The dA-AAI adduct is also the major AA-DNA adduct observed in rats treated orally with the plant extract AA.1 Moreover, the dA-AAI adduct showed an apparently life-long persistence in various organs in rats and it was still detectable in renal tissue of Belgian AAN patients more than 7 years after the patients stopped taking the herbal slimming regimen.4, 6 Because the renal tissue samples were collected between 1987–90 (Table I) our results confirm that the dA-AAI adduct is a suitable biomarker for exposure to AA years later. Autoradiographic pattern of DNA adducts found in kidney tissue of a Belgian AAN patient (a) and in kidney tissue of patients living in endemic areas of BEN analyzed for AA-DNA adducts:6 Patient 1 (b), Patient 2 (c) and Patient 3 (d). Spot A1, dA-AAI. The role of AA in the genesis of BEN was under debate, because AA was found in flour obtained from wheat contaminated with seeds of Aristolochia clematitis in endemic regions for BEN.13 This hypothesis, however, has not yet received sufficient support.11, 12 Our findings here provide the first evidence that people living in endemic areas for BEN have been exposed to AA. Because many herbal remedies are used locally in these areas,12 the cases reported here raises major concerns for public health. Whether our patients took herbal medication is not known, however, and we can only speculate about the source of exposure to AA. DNA adducts can be considered both as markers of the biologically effective dose and as markers of cancer risk.15 As pointed out recently AA-DNA adducts may trigger the carcinogenic process in AAN patients.1 Indeed, the dA-AAI adduct is a premutagenic lesion and is associated with mutations in genes involved in carcinogenesis, such as the H-ras protooncogene and the p53 gene.1, 3, 16 Our results may therefore provide a molecular link to the cause of urothelial tumors observed in patients living in BEN areas. Whether AA-DNA adducts play a causal role also in BEN awaits further elucidation. Some clinical and morphological features of the 2 diseases still differ. For instance by the rapidity of evolution from a few months to years in AAN vs. several years in BEN and also the age of the patients that is generally higher in BEN than for AAN.2 As mentioned before, exposure to the mycotoxin OTA was strongly associated with BEN. We therefore also looked for OTA-related DNA adducts in these patients. The detection of OTA-related DNA adducts by 32P-postlabeling requires chromatographic conditions different to those routinely used for lipophilic adducts, like AA-DNA adducts.7 Several adduct spots were present in all 3 renal tissues analyzed (Fig. 2). Patient 1 and 3 showed a similar adduct pattern consisting of 1 major adduct spot (Spot O1) and a few minor adducts (Spots O2–O4) (Fig. 2a,c) similar to the adduct pattern observed in renal tissue of rat and mouse treated with OTA that served as reference. In renal tissue of Patient 2 no OTA-related DNA adducts were found (Fig. 2b). Total OTA-related adduct levels were 4.7 and 3.1 adducts per 109 nucleotides in Patients 1 and 3, respectively (Table I). These findings are in line with the detection of high levels of OTA-related DNA adducts in urinary tract tumors of BEN patients,14 but the low adduct levels prevented co-chromatographic identification. These results clearly show that 2 of the Croatian patients living in high-risk areas for BEN have been exposed to OTA. Thus, our findings confirm a potential role of OTA in the development of upper urinary tract tumors in these areas. Whether OTA plays also a role in the interstitial nephropathy characterizing BEN remains to be determined. Autoradiographic pattern of DNA adducts found in kidney tissue of patients living in endemic areas of BEN analyzed for OTA-related DNA adducts:14 Patient 1 (a), Patient 2 (b) and Patient 3 (c). Spots O1–O4: OTA-related DNA adducts, structures unknown. Interestingly the same 2 patients exposed to OTA have been exposed to AA, too. The detection of both OTA-related and AA-DNA adducts in our patients may suggest interactive toxicologic effects of both toxins in endemic areas in the development of urothelial malignancy. Given the absence of evidence that these patients clearly suffered from BEN, the effects of these toxins in the development of BEN remain to be assessed. Other contaminations may come from citrinin, another mycotoxin, which has also been found in food in endemic areas and showed a synergistic effect with OTA.11 AAN patients have been exposed to high amounts of AA within a few months (≈15–20 months) providing a reason for the rapid development of AAN and AAN-associated urothelial tumors (≈2–6 years).2, 4, 11, 12 In contrast, people in endemic areas for BEN might have been exposed chronically to low amounts of AA over years, which might explain the slow development of BEN and BEN-associated urothelial malignancy. In summary, we found a high prevalence of AA exposure (2/3 cases) in a small number of farmers coming from endemic areas for BEN with tumoral or stenotic urinary tract obstruction. This provides evidence that in addition to OTA, AA may be a potential risk factor in the development of urothelial cancer in endemic areas. Whether AA plays a role also in BEN awaits further investigation in patients with unequivocal diagnosis of BEN. The respective role of AA and OTA in this entity should be evaluated by the detection of both OTA-related and AA-DNA adducts in urinary tract tissue from patients with unequivocal diagnosis of BEN and patients with other nephropathies but living in endemic areas for BEN. Yours sincerely, Volker M. ARLT, Dusan FERLUGA, Marie STIBOROVA, Annie PFOHL-LESZKOWICZ, Mato VUKELIC, Stjepan CEOVIC, Heinz H. SCHMEISER and Jean-Pierre COSYNS Volker M. Arlt*, Dusan Ferluga , Marie Stiborova , Annie Pfohl-Leszkowicz§, Mato Vukelic¶, Stjepan Ceovic¶, Heinz H. Schmeiser**, Jean-Pierre Cosyns , * Section of Molecular Carcinogenesis, Institute of Cancer Research, Sutton, United Kingdom, Institute of Pathology, University of Ljubljana, Ljubljana, Slovenia, Department of Biochemistry, Charles University, Prague, The Czech Republic, § Ecole Nationale Supérieure Agronomique de Toulouse, Laboratoire de Toxicologie et Sécurité Alimentaire, Auzeville Tolosane, France, ¶ Department of Pathology, General Hospital, Slavonski Brod, Croatia, ** Division of Molecular Toxicology, German Cancer Research Center, Heidelberg, Germany, Department of Pathology, Université Catholique de Louvain, Medical School, Brussels, Belgium.