Bongkrekic Acid and Its Novel Isomers: Separation, Identification, and Determination in Food Matrices

Abstract

The toxicity associated with bongkrekic acid (BKA) is severe due to its chemical structure, which also facilitates high mortality rates; however, its isomer, isobongkrekic acid (iBKA), with only minor structural variance, demonstrates marked differences in toxicity. This discrepancy in structural properties and toxicity highlights that risks have been potentially underestimated within current detection standards for BKAs. In this study, a novel BKA trans isomer at the C8 and C9 double carbon bonds (E-configuration), termed iBKA-neo, was successfully separated and identified. Subsequently, the multiple reaction monitoring parameters and chromatographic conditions for three BKA isomers were optimized, enabling effective separation within 15 min via UHPLC-MS/MS, among which the ammonium positive adduct ions yielded significantly higher response intensities for all BKA isomers than traditional deprotonated molecules. Additionally, distinct differences in the ion ratios between iBKA-neo and BKA were utilized for preliminary screening. On this basis, the extraction and enrichment strategies for BKAs were optimized in food matrices and validated comprehensively with good linearity (0.25-500 μg/kg), a superior limit of quantification (0.25 μg/kg), acceptable recoveries (82.32-114.84%), and stable intraday and interday precision (an RSD less than 12.67%). These findings significantly contribute to ecotoxicology and the formulation of safety standards concerning BKAs.

Keywords: Burkholderia gladioli; UHPLC-MS/MS; biotoxin; bongkrekic acid; cis–trans isomer; food matrix.

Figure 1

The structural identification of iBKA-neo. (A) The structural formulas of BKA, iBKA, and iBKA-neo (the double carbon bonds marked in red represent the cis–trans isomeric position). (B) ¹H NMR of iBKA-neo (400 MHz, CD₃OD). (C) ¹³C NMR of iBKA-neo (101 MHz, CD₃OD). (D) ¹H-¹H DQF-COSY of iBKA-neo (F1, 3424.7 Hz; F2, 3424.7 Hz, CD₃OD). (E) ¹H-¹H NOESY of iBKA-neo (F1, 3424.7 Hz; F2, 3424.7 Hz, CD₃OD). (F) ¹H-¹³C HSQC of iBKA-neo (¹H 3424.7 Hz, ¹³C 16,602.4 Hz). (G) ¹H-¹³C HMBC of iBKA-neo (¹H 3268.0 Hz, ¹³C 22,137.0 Hz). In detail, ¹H NMR (400 MHz, CD₃OD): δ 7.56 (d, J = 12.0 Hz, 1H), 7.48 (d, J = 16.1 Hz, 1H), 6.42 (d, J = 12.2 Hz, 1H), 5.95 (dd, J = 14.6, 10.3 Hz, 1H), 6.11 (dd, J = 16.1, 7.5 Hz, 1H), 6.03 (dd, J = 14.7, 10.3 Hz, 1H), 5.72 (s, 1H), 5.56 (dt, J = 14.6, 6.4 Hz, 1H), 5.42–5.33 (m, 1H), 5.42–5.33 (m, 1H), and 5.46 (dd, J =14.7, 7.3 Hz, 1H). 3.33 (s, 2H), 3.21 (s, 3H), 4.34 (t, J = 7.0 Hz, 1H), 2.46 (dt, J = 13.5, 7.0 Hz, 1H), 2.21–2.15 (m, 1H), 2.37–2.27 (m, 1H), 2.14–2.01 (m, 2H), 2.14–2.01 (m, 2H), 2.14–2.01 (m, 2H), 1.91 (s, 3H), 1.83 (s, 3H), 1.03 (d, J = 6.7 Hz, 3H). ¹³C NMR (101 MHz, CD₃OD): δ 132.36, 124.75, 124.55, 130.36, 143.91, 132.88, 118.64, 132.23, 128.06, 131.46, 126.45, 39.80, 55.28, 78.98, 36.70, 37.59, 39.52, 32.32, 32.11, 10.92, 17.50, 18.45, 172.92, 168.00.

Figure 2

Mass spectrometry fragmentation patterns and detection mode characteristics of BKA isomers. (A) Fragment ion characteristics of BKA in ESI negative mode [M-H]⁻, high-resolution mass spectrometry in SIM-ddMS2 detection mode, and HCD (%) steps of 5, 10, 15, and 20 for normalization. (B) Fragment ion characteristics of BKA in ESI positive mode [M+NH₄]⁺, high-resolution mass spectrometry in SIM-ddMS2 detection mode, with HCD steps of 5, 10, 15, and 20 for normalization. (C) Extraction ion chromatograms (XIC) of fragment ions at m/z 419 and m/z 437 of BKA and iBKA-neo in ESI positive mode. (D) The ion ratios of fragments at m/z 437 to m/z 419 for BKA and iBKA-neo in ESI positive mode. (E) MS response total ion chromatogram (TIC) of BKA isomers with the same concentration detected in ESI positive and negative mode. (F) Linear range of the MS response detected by BKA in ESI positive and negative mode.

Figure 3

Optimization of the chromatographic separation method for BKA isomers (BEH C18 column). (A) TICs of MEOH and ACN as organic phases with initial mobile phases of 20%, 40%, and 60% and ACN segmented gradient changes of 43%, 55%, and 80% for the separation efficiency of BKA isomers. (B) TICs of the separation efficiency of different proportions of FA (0.05%, 0.1%, 0.5%) on BKA isomers in the mobile phase. (C) TICs of the separation efficiency of different concentrations of AF (1 mM, 2 mM, and 5 mM) on BKA isomers in the mobile phase. (D) TICs of the separation efficiency of BKA isomers when different ratios of ACN (10%, 30%, 50%, 70%, and 90%) were used as dilution solvents. (E) TIC of the separation efficiency of BKA isomers optimized by chromatographic conditions. W_1/2 represents the full width at half maximum; Rs represents the separation degree.

Figure 4

Optimization of extraction and enrichment methods for BKA isomers. (A) The influence of MEOH and ACN as extraction reagents on the extraction recovery rate and matrix effect of BKA isomers. (B) Effects of different ratios of HAc (0.5%, 1%, and 2%) in the extraction reagent on the extraction recovery rates and matrix effects of BKA isomers. (C) Effects of the compatibility ratio (1:5, 2:5, or 5:5) of the matrix sample (g) and extraction reagent (mL) on the extraction recovery rates and matrix effects of BKA isomers. (D) The influence of different types of SPE columns (WAX, MAX, HLB) on the extraction recovery rates and matrix effects of BKA isomers. (E) The optimized extraction and enrichment method was used to determine the recovery rates of BKA isomers in different food matrices. The blue area represents the acceptable range of the recovery rate of 70–120%, the red area represents the signal enhancement of the matrix effect, and the green area represents the signal suppression of the matrix effect.

Figure 5

Methodological validation (Tremella fuciformis matrix). In the Tremella fuciformis matrix, the signal-to-noise ratio, LOQ, LOD (left), linear range, and correlation coefficient (right) of BKA (A), iBKA-neo (B), and iBKA (C) are shown. The blue area represents the data source of the noise when confirming the signal-to-noise ratio. (D) The recovery rates of BKA isomers at the LOQ, 5 LOQ, and 10 LOQ in the Tremella fuciformis matrix. The blue area represents the acceptable range of recovery rates from 70% to 120%. (E) The intraday and interday variations in the LOQ, 5 LOQ, and 10 LOQ concentrations of BKA isomers in food matrices. (F) The matrix effect in determination of BKA, iBKA, and iBKA-neo in Tremella fuciformis matrix.

Figure 6

Application of the UHPLC‒MS/MS method in real-world samples. (A) Tremella fuciformis, (B) rice flour, (C) sweet soup dumpling flour, and (D) sour noodles (n = 3, respectively) were tested as fermentation media, and Burkholderia gladioli of the same source was inoculated. After 14 days of fermentation, according to the extraction and enrichment methods provided in this study, the accurate content of each BKA isomer in the sample was detected, and the TICs of BKA isomers detected in different food matrices were generated.