Analysis of Aristolochic Acids in Herbal
Medicines by HPLC/UV and LC/MS
Research & Development Center
Sun Ten Pharmaceutical Co. Ltd., Taipei, Taiwan
The cause of rapidly progressive renal fibrosis found in young Belgian women was reported to have been associated with aristolochic acids, a group of compounds known to be nephrotoxic and carcinogenic. Similar cases were also reported in Japan and Taiwan. It has been investigated that aristolochic acids can be found in Aristolochia species (spp.), Bragantia spp. or Asarum spp. The Aristolochia spp. are usually found incorporated in herbal formulas. Determination of aristolochic acid I was ever addressed from FDA in the U.S.. We describe here the development of a more accurate method for analyzing aristolochic acid I and II by solid phase extraction (SPE) combined with HPLC/UV or LC/MS in herb simples or herbal formulas.
2. Results and Discussion
2.1. Identification of Aristolochic Acids in Analytical Samples
2.1.1. Identification by HPLC/UV and LC/MS Method
The aristolochic acid I and II in the samples were identified by comparing the retention times and UV spectra of authentic standards with those obtained from the samples.
The presence or absence of aristolochic acids in the samples was confirmed by LC/MS. Positive-ion electrospray ionization MS [(+)-ESI-MS] was applied to aristolochic acids in the samples. The spectrum for aristolochic acid I exhibited the characteristic molecular adduct ions at m/z 359 ([M+NH4]+), m/z 324 ([M+H-H2O]+), m/z 298 ([M+H-CO2]+), and m/z 296 ([M+H-NO2]+), and the spectrum for aristolochic acid II exhibited the characteristic molecular adduct ions at m/z 329 ([M+NH4]+), m/z 294 ([M+H-H2O]+), and m/z 268 ([M+H-CO2]+). Selected ion chromatograms at m/z 359, m/z 324, m/z 298, m/z 296 were shown in Fig. 1.
1. Any sample found to be negative for aristolochic acid I would be re-analyzed with the addition of 0.5 ppm spike of aristolochic acid I.
2. In the re-analysis of spiked sample, the signal-to-noise ratio for the ions m/z 359, 324, 298 and 296 must be not less than 3, otherwise the sample must be rejected.
2.2. Determination of Aristolochic acids by HPLC/UV Method
The reproducibility (relative standard deviation) of the proposed method, on the basis of peak area, for six replicated injections is between 0.10~1.89%. The variation of retention time of each peak was less than 1.78% for six replicated injections.
Calibration graphs (peak area y vs. injection amount x ) were constructed by injecting different volumes with the autosampler to be within the ranges: 0.01859~1.4868 for aristolochic acid I and 0.0118~0.236 for aristolochic acid II. The regression equations of these curves and their coefficients were calculated as follows: aristolochic acid I, y=1048590x+1829 (r2=1.0000); aristolochic acid II, y=559238x+1138 (r2=0.9999).
2.2.3. Determination of Aristolochic Acids in Herbal Materials
When the test solutions of single herbal material were analyzed by HPLC under the selected conditions, the chromatograms were obtained without any interference from nearby aristolochic acids. By substituting the peak areas of individual peaks for y in the regression equations, the contents of the individual aristolochic acids in samples are obtained.
Determination of aristolochic acids in single herbal material samples was directly done by HPLC/UV analyses. Separate determinations of herbal formulas would depend on the resolution and purity of aristolochic acids. Determination of aristolochic acid I and II with poor resolution in herbal formulas would be analyzed by SPE-HPLC/UV. The test solutions with SPE treatment were analyzed by HPLC under the selected conditions, the chromatograms would be obtained without any interference from nearby aristolochic acids. By substituting the peak areas of individual peaks for y in the regression equations, the contents of the individual aristolochic acids in samples are obtained.
2.2.4. Accuracy in SPE-HPLC/UV Method
Suitable amounts of aristolochic acid I and II were added sample of known aristolochic acid contents and the mixture was extracted and treated by SPE, and then analyzed using the proposed procedure. The range of recoveries of aristolochic acid I and II were between 99.08~111.58% and 98.49~99.35% respectively.
2.3 Detection Limit of Aristolochic Acid I
The detection limit of aristolochic acid I is 0.19 /g by HPLC/UV method (as shown in Fig. 2A), and 0.45 /g by LC/MS method (as shown in Fig. 2B).
In recent years, SPE has been broadly applied to the analyses of minor constituents in herbal medicines. The main purpose of the present study was to establish a method to confirm the existence of aristolochic acids, and to further determine their contents in herbal medicines. It was achieved successfully by applying SPE in combination with HPLC/UV or LC/MS to analyze aristolochic acids.
Share this Post: