Absorptivity Is an Important Determinant in the Toxicity Difference between Aristolochic Acid I and Aristolochic Acid II

Abstract

Inadvertent exposure to aristolochic acids (AAs) is causing chronic renal disease worldwide, with aristolochic acid I (AA-I) identified as the primary toxic agent. This study employed chemical methods to investigate the mechanisms underlying the nephrotoxicity and carcinogenicity of AA-I. Aristolochic acid II (AA-II), which has a structure similar to that of AA-I, was investigated with the same methods for comparison. Despite their structural similarities, findings from cultured human cells and gut sac experiments showed that AA-I is absorbed more effectively than AA-II (∼3 times greater for AA-I than for AA-II; p < 0.001). This increased absorption, along with the previously observed higher activity of reductive activation enzymes for AA-I, results in greater DNA damage and oxidative stress, both of which are key factors in AA-related toxicity. The similar patterns of cell mortality (34.4 ± 2.3% vs 9.7 ± 0.1% for AA-I and AA-II at 80 μM; p < 0.0001), DNA adduct formation (∼3 times greater for AA-I than for AA-II; p < 0.001), and oxidative stress levels in relation to the concentrations of AA-I and AA-II indicate that the higher absorption rate of AA-I is a significant contributor to its greater toxicity. The toxicity of AA-I was also found to be further enhanced by its (natural) coexistence with AA-II. Since AA-I and AA-II differ only by a methoxy group, future research on reducing risks associated with AA exposure should focus on strategies to lower the absorption of these compounds.

Keywords: Balkan endemic nephropathy; DNA adduct; aristolochic acids; dietary exposure; food contamination.

Scheme 1. Metabolic Activation and DNA Adduct Formation of Aristolochic Acid I (R = OCH₃) and Aristolochic Acid II (R = H)

Scheme 2. Metabolism of AA-I Forming AA-Ia and Aristolactam Ia, which Undergoes Glucuronide Conjugation, Lowering the Toxicity of AA-I

Figure 1

Cell mortality rates (A,B) and DNA adduct formation (C,D) in cultured kidney (A,C) and liver (B,D) cells that were exposed to different concentrations of AA-I or AA-II for 48 h. No AA-DNA adduct was detected in cells that were treated with the dosing vehicle (DMSO). Significant higher mortality rates and AA-DNA adduct formation were observed in AA-I than AA-II-exposed kidney and liver cells, with higher rates in kidney cells compared to liver cells. Student’s t-test where ns, p > 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. The data represent means ± SD for three independent experiments.

Figure 2

Exposure-time-dependent formation of the AA-DNA adduct (ALI-dA for AA-I; ALII-dA for AA-II) in cultured kidney (A) and liver (B) cells exposed to 30 μM of AA-I and AA-II for varying duration, along with the kinetics of AA-DNA adduct elimination from the AA-exposed kidney (C) and liver (D) cells. Faster AA-DNA adduct formation was observed for AA-I than AA-II in both kidney (A) and liver (B) cells. AA-DNA adducts formed by AA-I were eliminated at a slower rate than those formed by AA-II (C,D). Student’s t-test where ns, p > 0.05; *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. The data represent means ± SD for three independent experiments.

Figure 3

8-Oxo-dG (A,B) and GSH/GSSG ratio (C,D) in cultured kidney (A,C) and liver (B,D) cells exposed to different concentrations of AA-I or AA-II for 48 h. Higher 8-oxo-dG frequencies and GSH depletion were observed in cells exposed to AA-I compared to AA-II. Higher 8-oxo-dG frequencies and GSH depletion were also observed in kidney cells compared to liver cells. Student’s t-test where ns, p > 0.05; *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. The data represent means ± SD for three independent experiments.

Figure 4

Concentrations of AA-I and AA-II in intracellular fluid of cultured kidney (A) and liver (B) cells exposed to different concentrations of AA-I or AA-II for 48 h, along with the concentrations of AL-I or AL-II in the cell culture medium (C,D). Higher cellular absorptivity of AA-I compared to AA-II was observed in the exposed cells. A higher reductive metabolism rate was observed for AA-I compared to AA-II. Student’s t-test where **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. Fitting the data by linear regression yielded lines with the following equations. (A) AA-I: y = 0.25x – 2.7 (r² = 0.99), AA-II: y = 0.050x – 0.54 (r² = 0.94); (B) AA-I: y = 0.28x – 4.2 (r² = 0.97), AA-II: y = 0.065x – 0.87 (r² = 0.99); (C) AL-I: y = 1.5x – 38.7 (r² = 0.87), AL-II: y = 0.13x – 2.2 (r² = 0.99); and (D) AL-I: y = 0.56x – 10.5 (r² = 0.99), AL-II: y = 0.15x – 2.7 (r² = 0.99). The data represent means ± SD for three independent experiments.

Figure 5

Correlation of the concentrations of DNA adducts with ALs in the culture medium of (A) kidney and (B) liver cells exposed to different concentrations of AAs and (C) in organs of mice that were treated with 10 mg/kg of AA-I or AA-II. Fitting the data by linear regression yielded lines with the following equations. (A) AA-I: y = 0.73x + 12.4 (r² = 0.90), AA-II: y = 2.3x + 0.29 (r² = 0.96); (B) AA-I: y = 1.1x + 0.22 (r² = 0.99), AA-II: y = 0.73x + 0.14 (r² = 0.99); and (C) AA-I: y = 3.8x + 46.7 (r² = 0.90), AA-II: y = 2.9x + 14.4 (r² = 0.95). The data represent means ± SD for three and five independent cell and mice experiments, respectively.

Figure 6

DNA adduct formation in cultured kidney (A,C) and liver (B,D) cells exposed to mixtures of AA-I (ALI-dA; A,B) and AA-II (ALII-dA; C,D) at various concentrations for 48 h. Significantly higher levels of AA-DNA adducts were observed in cells coexposed to the AA-I and AA-II mixture compared to those exposed to the same concentrations of AA-I or AA-II separately. Student’s t-test where ns, p > 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. Fitting the data by linear regression yielded lines with the following equations. (A) ALI-dA: y = 2.0x – 37.7 (r² = 0.96) vs y = 0.92x – 16.0 (r² = 0.95); (B) ALI-dA: y = 0.77x – 9.7 (r² = 0.99) vs y = 0.39x – 4.5 (r² = 0.99); (C) ALII-dA: y = 0.29x – 5.1 (r² = 0.99) vs y = 0.28x – 4.7 (r² = 0.99); and (D) ALII-dA: y = 0.17x – 2.9 (r² = 0.98) vs y = 0.099x – 1.7 (r² = 0.98). The data represent means ± SD from three independent experiments.

Figure 7

Concentrations of (A) ALI-dA and ALII-dA and (B) AL-I and AL-II in different organs of AA-I- or AA-II-exposed mice, respectively. Student’s t-test where ****, p < 0.0001. The data represent means ± SD for five independent experiments. No AA-DNA adducts or ALs were detected in internal organs of control mice that were dosed with the dosing vehicle.

Figure 8

Time-course cumulative absorption of AA-I and AA-II in Tyrode’s solution outside the gut sacs, which contain 1 μM of AA-I and AA-II, respectively. Student’s t-test where **, p < 0.01. The data represent means ± SD for five independent experiments.