- Email: [email protected]
Determination of aflatoxin and zearalenone analogs in edible and medicinal herbs using a group-specific immunoaffinity column coupled to ultra-high-performance liquid chromatography with tandem mass spectrometry
Journal of Chromatography B 1092 (2018) 228–236
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/jchromb
Determination of aflatoxin and zearalenone analogs in edible and medicinal herbs using a group-specific immunoaffinity column coupled to ultra-highperformance liquid chromatography with tandem mass spectrometry
T
Shujuan Suna,1, Kai Yaoa,1, Sijun Zhaob, Pimiao Zhenga, Sihan Wanga, Yuyang Zenga, ⁎ Demei Lianga, Yuebin Kec, Haiyang Jianga, a
Beijing Advanced Innovation Center for Food Nutrition and Human Health, College of Veterinary Medicine, China Agricultural University, Beijing Key Laboratory of Detection Technology for Animal-Derived Food Safety, Beijing 100193, People's Republic of China China Animal Health and Epidemiology Center, Qingdao 266032, People's Republic of China c Shenzhen Center for Disease Control and Prevention, Shenzhen 518055, People's Republic of China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Aflatoxins Zearalenone analogs Edible and medicinal herbs Group-specific Immunoaffinity column UPLC–MS/MS
Six aflatoxins (AFs; AF B1, B2, G1, G2, M1 and M2) and six zearalenone (ZEN) analogs (ZEN, zearalanone, αzeralanol, β-zeralanol, α-zearalenol, and β-zearalenol) were simultaneously extracted from edible and medicinal herbs using a group-specific immunoaffinity column (IAC) and then identified by ultra-high-performance liquid chromatography tandem mass spectrometry (UPLC–MS/MS). The IAC was prepared by coupling N-hydroxysuccinimide-activated Sepharose 4B Fast Flow gel with two group-specific monoclonal antibodies. The column capacities to six AFs and six ZEN analogs ranged from 100.2 ng to 167.1 ng and from 59.5 ng to 244.4 ng, respectively. The IAC–UPLC–MS/MS method was developed and validated with three different matrices (Chinese yam [Dioscorea polystachya], Platycodon grandiflorum and coix seed [Semen Coicis]). Recoveries of twelve analytes from edible and medicinal herbs were in the range of 64.7%–112.1%, with relative standard deviations below 13.7%. The limits of quantification were in the range from 0.08 μg kg−1 to 0.2 μg kg−1. The method was proven to be sensitive and accurate, and suitable for the determination of real samples.
1. Introduction Herbal medicines have been used for thousands of years in China, and some herbal medicines have been considered as daily food (known as edible and medicinal herbs) because of their health-promoting functions and disease treatment properties [1]. With the improvement of people's living standards, individuals increasingly care more about their health. To remain healthy, individuals have used many kinds of edible and medicinal herbs, such as Chinese yam, Platycodon grandiflorum, and coix seed, as alternative medicines. A survey conducted by World Health Organization showed that about 70%–80% of the world populations rely on non-conventional medicines comprised mainly of herbal sources in their primary healthcare [2]. Although many health benefits are present in edible and medicinal herbs, the safe consumption of these products has gained much concern because of contaminations in raw materials by aflatoxins (AFs) and zearalenone (ZEN) structural analogs during cultivating, harvesting, processing and storage [3].
⁎
1
Corresponding author. E-mail address: [email protected] (H. Jiang). These authors contributed equally to this work.
https://doi.org/10.1016/j.jchromb.2018.06.012 Received 20 April 2018; Received in revised form 4 June 2018; Accepted 6 June 2018 Available online 07 June 2018 1570-0232/ © 2018 Elsevier B.V. All rights reserved.
AFs are produced by fungi belonging to Aspergillus flavus and Aspergilllus parasiticus under warm and moist conditions, which are designated as a group 1 carcinogenic compound by the International Agency for Research on Cancer [3]. AF B1, B2, G1, G2, M1 and M2 (Fig. 1; AFB1, AFB2, AFG1, AFG2, AFM1 and AFM2, respectively) are the most ubiquitous members of AF family and have received increasing attention because of their great harm to the liver [4]. Recently, AFs have been widely detected in herbal medicines [5–8]. Han et al. reported the mean levels (incidence) of AFB1, B2, G1 and G2 in herbal medicines samples were 1.40 (68.8%), 1.27 (50.0%), 0.50 (43.8%) and 0.94 (43.8%) μg kg−1; AFM1 was also detected with maximum concentrations of 0.70 μg kg−1 [6]. ZEN has also been found in herbal medicines [9, 10], but has not been studied as extensively as AFs. Zhang et al. detected ZEN in coix seed, with levels ranging from 18.7 μg kg−1 to 211.4 μg kg−1. ZEN is a naturally occurring nonsteroidal estrogenic mycotoxin produced by genus Fusarium [11], and its derivatives (Fig. 1; zearalanone [ZAN], α-zeralanol [α-ZAL], β-zeralanol [β-ZAL], α-zearalenol [α-ZOL], β-zearalenol [β-ZOL]) have also been characterized
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Fig. 1. Chemical structures of the AFs and ZEN structural analogs.
against ZEN analogs (named 3D4) [19], 5H3 recognizes six AFs (AFB1, AFB2, AFG1, AFG2, AFM1 and AFM2), and whereas 3D4 recognizes six ZEN analogs (ZEN, ZAN, α-ZAL, β-ZAL, α-ZOL and β-ZOL), respectively. As far as we know, this study is the first report on a multiple mycotoxins IAC for the simultaneous determination of AFs and ZEN analogs in three kinds of edible and medicinal herbs, including Chinese yam, Platycodon grandiflorum and coix seed. Finally, the developed IACLC-MS/MS method proved to be sufficiently sensitive and accurate and hence successfully used in real sample detection.
with estrogenic effects [12]. Therefore, to ensure consumers' health and minimize economic losses from contaminated herbal medicines, sensitive and accurate analytical methods are urgently needed for the detection of AF and ZEN analogs concentrations in herbal medicines. To date, the most common analytical methods include liquid chromatography electrospray ionization tandem mass spectrometry (LC–ESI-MS/MS) [6, 7, 9, 13, 14] or LC coupled with fluorescence detection (FLD) [5, 8, 10, 15]. However, LC–ESI-MS/MS is more preferred because of its excellent selectivity and sensitivity. The complexities of herbal medicine matrices considerably challenge the attainment of accurate data under detection by MS/MS. Therefore, the clean-up performance of analytical methods is of great importance during the analytical process. The current purifying technologies for determination of AFs or ZEN are mainly based on solid phase extraction (SPE) [6], quick, easy, cheap, effective, rugged, and safe (QuEChERS) method [14] and immunoaffinty column (IAC) [5, 7, 8, 10]. SPE is easy to perform and solvent conserving, but retainment of analytes is based on nonspecific binding, which can lead to the coextraction of analytes and matrix impurities. QuEChERS offers a less expensive and time-saving approach to determine target analytes. However, its strong matrix effect is a central issue that cannot be ignored. IAC is a powerful purification technique that relies on the specific recognition between the antibody and complementary analytes [16]. Compared with SPE and QuEChERS, IAC is most frequently used because of its high specificity and efficiency. However, most studies have only detected AFs or ZEN with commercial IACs [7, 10, 17], the IACs may not be suitable for all the herbal medicines, and none has been used to determine two classes of analytes. The application of the screening and determination of two different types of mycotoxin has been largely restricted by the high cost of commercial IACs. Thus, a cost-saving IAC that can purify two classes of mycotoxins is needed for research. An IAC preparation that uses two group-specific monoclonal antibodies (Mabs) has not been reported. Such lack may be due to the difficulty in obtaining ZEN group-specific Mabs and the requirement for steady preparative techniques for further developing a multi-mycotoxin IAC. In this paper, a novel IAC was prepared using a recently obtained Mab against AFs (named 5H3) [18] and a previously obtained Mab
2. Materials and methods 2.1. Reagents and materials The AFs Mab (5H3) [18] and the ZENs Mab (3D4) [19] were previously produced by our laboratory. AFB1, AFB2, AFG1, AFG2, AFM1, AFM2, ZEN, ZAN, α-ZAL, β-ZAL, α-ZOL and β-ZOL standards were purchased from Sigma–Aldrich (St. Louis, MO). N-hydroxysuccinimide (NHS)-activated Sepharose 4B Fast Flow (FF) and polypropylene columns (3 mL) with polytetrafluoroethylene (PTFE) frits were purchased from Biocomma (Shenzhen, China). LC-MS grade acetonitrile (ACN) and methanol (MeOH) and were supplied by Merck (Darmstadt, Germany). Ultrapure water was obtained from a Milli-Q water purification system (Millipore, Billerica, MA, USA). All other chemicals and solvents were of analytical grade orhigher. Stock mixed standard solutions (10 mg mL−1) were prepared by dissolving 10 mg of each standard substance in MeOH (10 mL) in amber glass vials. Working mixed standard solutions (10 μg mL−1) were prepared by diluting the stock solutions with MeOH. The working solutions were stored at 4 °C. Phosphate buffered saline (PBS) was prepared by dissolving 8.8 g NaCl, 0.02 g KCl, 2.9 g NaH2PO4·2H2O and 0.59 g Na2HPO4·12H2O in 1 L of water, and the pH was adjusted to 7.4 with 1 M NaOH aqueous solution. 2.2. Samples Edible and medicinal herb samples including Chinese yam, 229
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
the gels coupled with the AFs Mab and the ZEN analogs Mab were mixed together into a SPE column and washed with 30 mL PBS. Finally, the prepared multiple IAC column was stored at 4 °C in PBS containing 0.02% (w/v) Proclin 300.
Platycodon grandiflorum, and coix seed were randomly purchased from supermarkets in Beijing. All samples were stored at 4 °C. 2.3. LC–MS/MS analysis
2.5. Sample extraction and IAC clean-up
The LC-MS/MS analysis was performed with a Xevo TQ-S triplequadrupole tandem mass spectrometer (Waters, Milford, MA, USA). Chromatographic separation of AFs and ZEN analogs was carried out using an Acquity HSS T3 column (2.1 mm × 100 mm; 1.8 μm; Waters) and an Acquity BEH C18 column (2.1 mm × 100 mm; 1.7 μm; Waters), respectively. The mobile phase for positive-ion electrospray ionization (ESI+) consisted of 0.1% formic acid aqueous solution (mobile phase A1) and ACN:MeOH mixture (50:50, v/v; mobile phase B1). The mobile phase for negative-ion ESI (ESI−) consisted of ultrapure water (mobile phase A2) and ACN (mobile phase B2). The gradient elution program for ESI+ was as follows: 0–5.0 min, 30%–100% B1; 5.0–6.0 min, 100% B1; 6.0–6.5 min, 100%–30% B1; 6.5–8.0 min, 30% B1. The gradient elution program for ESI− was as follows: 0–3.0 min, 15%–85% B2; 3.0–3.5 min, 85%–90% B2; 3.5–4.0 min, 4.0–4.3 min, 90%–15% B2; 4.3–5.5 min, 15% B2. The column temperature and sampler temperature were set at 40 °C and 10 °C, respectively. The flow rate and injection volume were 0.3 mL min−1 and 10 μL, respectively. The MS/MS acquisition was operated in ESI+ and ESI− with multiple reaction monitoring (MRM). The capillary voltage was 2.0 kV for ESI+ and 3.2 kV for ESI−. The source temperature and desolvation temperature were set at 80 °C and 450 °C for two ESI modes, respectively. The nitrogen flow rate was 800 L h−1 and 550 L h−1 for ESI+ and ESI−, respectively. The ultra-pure argon flow rate, used as the collision gas, was 0.13 mL min−1 for both modes. The MS/MS parameters for AFs and ZEN analogs are presented in Table 1.
Samples (Chinese yam and Platycodon grandiflorum) were cut into small pieces (coix seed did not need to be cut), lyophilized for 24 h, and ground to fine powder using a TissueLyser II (Qiagen, Germany), and then the sample powders were passed through 40 mesh screen to obtained the homogenized samples. Homogenized samples (1.00 ( ± 0.01) g) were weighed in 10 mL polypropylene centrifuge tubes. After addition of 4 mL MeOH–water (60:40, v/v), the mixture was mixed for 10 min at 600 rpm with a VXR shaker (IKA, Germany). After centrifugation at 8000 rpm for 10 min, the supernatant was transferred to a 50 mL polypropylene centrifuge tube and diluted to 24 mL with PBS. Subsequently, the diluted extraction solvent was passed through an IAC at a flow of one drop per second, the column was washed with 5 mL PBS and 5 mL water. Finally, the target analytes were eluted with 3 mL MeOH and evaporated to dryness by a gentle stream of N2 at 40 °C. The residue was reconstituted in 1 mL ACN–water (30:70, v/v) and the solution was filtered with 0.22 μm nylon filter membrane before injection into UPLC-MS/MS. 2.6. IAC working conditions optimization Loading conditions were evaluated by using the following MeOHwater /PBS ratios (v/v, 1:7, 1:5, 1:4 and 1:3, which correspond to MeOH percentages of 7.5%, 10%, 12% and 15%), and the optimal MeOH volume (1, 2, 3, 4, and 5 mL) were both determined to achieve satisfactory recoveries. Extraction solvent (4 mL MeOH–water) was diluted with PBS (28, 20, 16 and 12 mL) and spiked with 50 μL 40 ng mL−1 mixed standard solution. The resulting mixture was loaded onto the IAC. After washing with 5 mL PBS and 5 mL water, the two classes of compounds were eluted with various volume of MeOH and evaporated to dryness by a gentle stream of N2 at 40 °C.
2.4. IAC preparation The immunosorbent was prepared in accordance with our previous study [20]. NHS-activated Sepharose 4B FF gel (1 mL) was transferred to an empty polypropylene SPE tube (3 mL) with a PTFE filter. The gel was washed with 10 mL 1 mM hydrochloric acid and then with 10 mL 0.2 M NaHCO3 (pH 8.3) solution. After this procedure, the pre-equilibrated gel was mixed with 0.5 mg AFs Mab and 0.5 mg ZEN analogs Mab solution. The mixture was gently shaken end over end at room temperature for 2 h. The eluted solution was then collected to detect the amount of Mabs by a Nanodrop 2000 UV–Vis spectrometry (Thermo Fisher, USA) and the coupling efficiency was determined. The mixture was blocked by reacting with the blocking buffer (0.2 M NaHCO3, pH 8.3, containing 50 mM ethanolamine) for 2 h at room temperature to block the unreacted active sites on the sorbent. Subsequently, the blocked gel was consecutively washed twice with 40 mL Tris–HCl buffer (0.1 M, pH 7.0) and 40 mL HOAc-NaOAc buffer (0.1 M, pH 4.0). Finally,
2.7. Column capacity and reproducibility Column capacity was determined by over-loading solutions containing 1000 ng of each AFs and ZEN analogs that was mixed with loading solution. The subsequent steps were the same as those described in Section 2.5. The reproducibility was tested by calculating the relative standard deviation (RSD) of column capacities of five IACs in one batch and three IACs in three consecutive days. The whole column capacity process was performed as described in the above-mentioned paragraph.
Table 1 MS/MS parameters and retention time of 12 analytes. Analytes AFB1 AFB2 AFG1 AFG2 AFM1 AFM2 ZEN ZAN α-ZAL, β-ZAL, α-ZOL, β-ZOL ⁎
Retention time (min) 2.38 2.20 2.19 2.01 1.77 1.61 3.09 3.11 2.74 2.57 2.79 2.60
Mode +
ESI ESI+ ESI+ ESI+ ESI+ ESI+ ESI− ESI− ESI− ESI− ESI− ESI−
Precursor ion (m/z)
Product ion (m/z)
Cone voltage (V)
Collision energy (eV)
313.1 315.1 329.1 331.2 329.1 331.2 317.0 319.0 321.0 321.0 319.0 319.0
285.1⁎/240.9 287.1⁎/259.1 243.1⁎/128.0 245.1⁎/217.0 273.1⁎/259.1 273.1⁎/285.0 175.0⁎/273.0 275.3⁎/205.0 277.0⁎/303.0 277.0⁎/303.0 275.2⁎/301.0 275.2⁎/301.0
45 45 10 30 20 20 25 25 25 25 25 25
20/35 26/30 26/45 24/26 22/20 20/20 24/21 22/21 22/22 22/22 22/22 22/21
Ion used for quantification. 230
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
2.8. Recovery
3.2. IAC conditions optimization
The recoveries of the method were evaluated by using blank samples (1.00 g) spiked with standard analytes at four concentration levels (LOQ, 5-fold LOQ, 10-fold LOQ and 50-fold LOQ for each analyte), each concentration was repeated five times.
The evaluation of different IAC working conditions is essential for a satisfying recovery. Organic solvents or a mixture of organic solvent and water are often used to extract AF and ZEN analogs. However, the sample extract containing a high percentage of MeOH was found unfit for directly loading the IAC. To obtain an optimal composition of the loading solution, various loading conditions were evaluated. Recoveries were the highest when the MeOH–water/PBS ratio was increased to 1:5 (Fig. 2A) and decreased when the MeOH–water/PBS ratio exceeded 1:5. This pattern may be explained by the fact that a high ratio means a high concentration of MeOH in the loading buffer, and MeOH greatly influences the analyte–antibody binding. A low percentage of organic solvent can achieve satisfactory recoveries and reduce non-specific adsorption [23]. Therefore, MeOH–water/PBS ratio of 1:5 was used in the following experiment. After the washing process, the analytes retained on the IAC were eluted with 1–5 mL of pure MeOH. Previous studies reported that analytes can be released completely from the analyte-antibody complex when using pure MeOH or a MeOH-water mixture as elution solvent [22, 24, 25]. However, a small quantity of water in the elution solvent may increase the elution volume and render the evaporation difficult. No recovery increase was found when the elution volume reached 3 mL (Fig. 2B).
3. Results and discussion 3.1. IAC preparation Polyclonal antibody (Pab) and Mab are the most important reagents for preparing IAC, and there is a direct correlation between the antibody performance and the potential use of IAC. In order to obtain a consistent clean-up and extraction performance of IAC, the long-term preparation of massive and reproducible antibodies must be guaranteed. However, Pabs are usually generated from just one animal, and the supply of reproducible Pabs is commonly limited [21]. Hence, Mabs are generally used instead of Pabs to obtain IACs with high repeatability. To develop an IAC which is simultaneously group-selective to both AFs and ZEN analogs, the process requires group-specific Mabs that can recognize several analogs. Mabs with uniform affinity to all analytes in the same class are highly necessary to avoid the competition between the structurally related analogs for limited antibody binding sites during the IAC loading process [22]. The competition in the IAC loading process may lead to an unequal adsorption of analytes onto IAC when the analytes are in high concentration in samples. Thus, two group-specific Mabs against both AF (5H3) and ZEN analogs (3D4) with uniform cross-reactivities were utilized in this study. According to previous research, the Mab 3D4 had cross-reactivities of 57.5%–108.5%. The cross-reactivities of our recently prepared Mab (5H3) were determined by enzyme-linked immunosorbent assay and shown to lie in the range of 64.4%–140.6%. The details on the IC50 (half maximal inhibitory concentration) and cross-reactivities are listed in Table 2. The two Mabs were coupled with NHS-activated Sepharose 4B FF to prepare an IAC that can concomitantly extract AF and ZEN analogs (12 analytes in total) by using only one column. The IAC was prepared by packing the Sepharose 4B FF that was coupled with two Mabs into an empty column. NHS-activated Sepharose 4B FF was selected not only because of its chemically and biologically inertness but also because of the benefit of fast flow. Compared with the most commonly used solid support, namely, CNBractivated Sepharose 4B, the NHS-activated Sepharose 4B FF exhibited a significant flow rate improvement, which can conserve much time during the IAC loading procedure. 3D4 Mab (0.5 mg) and 5H3 Mab (0.5 mg) were immobilized on 1 mL NHS-activated Sepharose 4B FF gel. The coupling efficiency was determined in accordance with the ratio of coupled antibodies to added antibodies. In this study, coupling efficiency was shown to be 98.1%.
3.3. Column capacity and reproducibility of IAC Column capacities for AF and ZEN analogs were determined and summarized in Table 3. In this study, two group-specific Mabs against AF and ZEN analogs were combined for the preparation of IAC. High column capacities were in the range of 100.2–167.1 ng for AFs and 59.5–244.4 ng for the ZEN analogs, respectively. Interestingly, the lower cross-reactivities of the Mab 3D4 for β-ZOL (57.5%) and ZAN (66.4%) than for other ZEN analogs appeared to slightly influence the column capacity (59.5 ng for β-ZOL and 244.4 ng for ZAN). This phenomenon also occurred in the Mab 5H3 for AFM2. In summary, the column capacities of the IAC column for each analyte were mainly dependent on the experimental data and not the cross-reactivities [25]. The prepared IAC in this research showed wider recognition abilities than those in previous works for the six AFs (AFB1, AFB2, AFG1, AFG2, AFM1 and AFM2) and six ZEN analogs (ZEN, ZAN, α-ZAL, β-ZAL, α-ZOL and β-ZOL) under study. Zhang et al. [17] applied commercial IAC products to analyze AFB1, AFB2, AFG1 and AFG2, but the parameters of IAC and the information on the antibody were not given. Some researchers reported IACs that only recognized limited numbers of AFs [5, 6, 8] or ZEN [10, 26]. Hu et al. [24] developed a multiple IAC that can capture AFB1, AFB2, AFG1, AFG2 and ZEN, but the remaining two AFs and five ZEN analogs were not included. Mabs (40 mg) were also used for the generation of an IAC, and the column capacities were 194.2, 185.4, 159.0, 150.5, and 135.5 ng for AFB1, AFB2, AFG1, AFG2 and ZEN, respectively. In this study, 1 mg Mab was utilized for achieving comparable column capacities because of the extremely low IC50 values of two group-specific Mabs. A greater number of IACs can be produced at a relatively low cost when a fixed amount of Mabs are used. We conducted intra-assay and inter-assay column capacity determinations to evaluate the reproducibility of IAC (Table 3). The intraand inter-day RSD values of IAC column capacities were 2.3%–10.1% and 2.8%–13.7%, respectively, which reveal the satisfactory reproducibility of IAC.
Table 2 Cross-reactivities and IC50 values of AF and ZEN analogs. Analytes
IC50 (pmol mL−1)
Cross-reactivities (%)a
AFB1 AFB2 AFG1 AFG2 AFM1 AFM2 ZEN ZAN α-ZAL β-ZAL α-ZOL β-ZOL
0.2972 0.3686 0.2114 0.3443 0.3001 0.4612 0.1289 0.1941 0.1230 0.1724 0.1188 0.2241
100.0% 80.6% 140.6% 86.3% 99.0% 64.4% 100.0% 66.4% 104.8% 74.8% 108.5% 57.5%
a
3.4. Method validation The developed method was validated in terms of matrix effect, linearity, specificity, limit of detection (LOD), limit of quantification (LOQ), and recovery, and precision of the method was analyzed by spiking mixed standards into edible and medicinal herb samples.
Cross-reactivities = (IC50 of AFB1 or ZEN /IC50 of analogs) × 100%. 231
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Fig. 2. The optimized working conditions of IAC determined by UPLC–MS/MS (n = 3). (A) MeOH–water /PBS ratios; (B) elution volume. Table 3 IAC capacities and RSDs for each analyte. AFs
Column capacity (ng) Intra-day RSD (%) Inter-day RSD (%)
ZEN analogs
AFB1
AFB2
AFG1
AFG2
AFM1
AFM2
ZAN
ZEN
α-ZOL
β-ZOL
α-ZAL
β-ZAL
130.0 4.1 2.9
152.4 5.5 8.1
126.6 2.3 4.8
160.8 8.3 7.9
100.2 2.5 4.1
167.1 3.2 2.8
244.4 7.1 6.8
209.2 2.6 3.7
183.8 3.9 5.2
59.5 7.5 9.9
239.1 5.4 13.7
175.7 10.1 5.6
Table 4 Matrix effect, LOQ and LOQ of the method. Analytes
AFB1
AFB2
AFG1
AFG2
AFM1
AFM2
ZEN
ZAN
α-ZAL
β-ZAL
α-ZOL
β-ZOL
Matrices
Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed
LOD (μg kg−1)
0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.05 0.05 0.05 0.07 0.07 0.07 0.03 0.03 0.03 0.05 0.05 0.05 0.06 0.06 0.06 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.02 0.02 0.02
LOQ (μg kg−1)
Slopes of two standard curves
0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.15 0.15 0.15 0.20 0.20 0.20 0.08 0.08 0.08 0.15 0.15 0.15 0.20 0.20 0.20 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10
232
Linearity range (μg kg−1)
Matrix-matched
Solvent
SSE (%)
0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.15–20.00 0.15–20.00 0.15–20.00 0.20–20.00 0.20–20.00 0.20–20.00 0.08–20.00 0.08–20.00 0.08–20.00 0.15–20.00 0.15–20.00 0.15–20.00 0.20–20.00 0.20–20.00 0.20–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00
22,925 16,736 22,745 17,990 14,837 16,634 17,026 13,666 19,206 2756 2418 3289 5272 4899 5953 24,678 22,658 26,116 22,767 24,815 24,522 10,067 10,711 9610 14,685 15,018 15,613 14,260 15,042 15,476 14,561 17,156 19,193 17,419 17,134 24,387
20,198
113.5 82.9 112.6 112.5 92.8 104.0 102.6 82.4 115.8 96.9 85.0 115.6 89.7 83.4 101.3 106.2 97.5 112.4 85.6 93.3 92.2 97.0 103.2 92.6 83.9 85.8 89.2 82.1 86.6 89.1 80.8 95.2 106.5 84.1 85.5 80.3
15,994
16,588
2844
5877
23,236
26,597
10,379
17,504
17,370
18,022
20,374
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Fig. 3. Typical MRM chromatograms of Platycodon grandiflorum matrix-matched standard solutions of 5 ng mL−1 for each analyte: (A) six AFs; (B) six ZEN analogs.
medicinal herb samples. The limits were based on signal-to-noise ratios of 3:1 and 10:1 for AF and ZEN analogs, respectively. The LODs ranged from 0.02 μg kg−1 to 0.07 μg kg−1, and the LOQs were in the range from 0.08 μg kg−1 to 0.2 μg kg−1. The recovery and precision of five blank samples fortified at three fortification levels of 12 mixed standards were determined within the same day. The mean recoveries lay between 64.7% and 112.1%, with RSDs lower than 13.7% (n = 5) in the intra-day experiment (Table 5). The results fulfilled the current demand for simultaneous detection of AF and ZEN analogs. The methods reported previously were compared with the present study (Table 6). The recoveries of the present work were slightly lower than those of others, but the LODs in this study were improved because of the good purifying performance of the multiple IAC. Moreover, six AF and six ZEN analogs can be simultaneously extracted and purified by a single IAC.
For the development of a reliable IAC–UPLC–MS/MS method, the matrix effect in case of inaccurate results must be carefully considered because ion suppression or enhancement induced by sample matrices and interferences during the ESI process may occur. The matrix effect is commonly evaluated as a signal suppression/enhancement effect (SSE) [27], which is quantitatively calculated in accordance with the following equation: SSE (%) = (the slope of matrix-matched standard curves / the slope of solvent standard curves) × 100%. SSE < 100% is defined as ion suppression, SSE > 100% is defined as ion enhancement, and an SSE of 80%–120% is considered to be negligible [28]. As shown in Table 4, the SSE values of all the herbal medicines matrices are in the negligible range, indicating that all the analytes in this study can be determined directly by using solvent standard curves [29]. The correlation coefficients (R2) of the matrix-fortified calibration curves were all > 0.995 and showed very good linearities. The specificity of the method was evaluated by analyzing blank samples spiked with the standard solvent. No interfering peaks were observed at the retention time of each analyte and demonstrated a high specificity of the established method (Fig. 3). The LODs and LOQs of the method were determined by confirming the minimum detectable concentration of 12 analytes in edible and
3.5. Application to real samples The validated results demonstrate that our IAC-UPLC-MS/MS method is applicable for determining AF and ZEN analogs in herbal 233
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Table 5 Recoveries and RSD (n = 5) of 12 analytes in three edible and medicinal herbs. Analytes
AFB1
AFB2
AFG1
AFG2
AFM1
AFM2
ZEN
ZAN
α-ZAL
β-ZAL
α-ZOL
β-ZOL
Spiked levels (μg kg−1)
0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00 0.15 0.75 1.50 7.50 0.20 1.00 2.00 10.00 0.08 0.40 0.80 4.00 0.15 0.75 1.50 7.50 0.20 1.00 2.00 10.00 0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00
Chinese yam
Platycodon grandiflorum
coix seed
Recovery (%)
RSD (%)
Recovery (%)
RSD (%)
Recovery (%)
RSD (%)
89.4 95.2 90.7 93.1 77.3 84.2 86.7 88.2 81.7 92.3 85.9 82.5 88.9 95.8 90.5 93.2 87.2 80.3 92.3 90.6 79.1 84.6 81.5 78.3 73.2 78.5 85.3 80.5 64.7 71.3 74.6 70.4 80.7 69.1 77.3 81.4 73.6 85.4 79.9 83.1 80.6 76.1 88.7 85.0 74.4 83.2 81.9 85.4
4.6 9.1 7.3 5.3 4.1 3.7 8.3 4.3 6.6 3.9 12.1 7.2 8.5 12.4 7.1 4.5 3.8 2.4 9.2 5.3 4.3 2.8 7.1 9.9 8.5 3.6 11.3 7.3 2.9 4.7 6.9 5.8 11.5 6.4 8.8 3.7 4.3 8.1 2.4 5.5 5.3 3.8 10.2 4.2 3.7 1.9 7.5 4.6
93.4 98.1 85.2 89.5 88.4 81.4 91.2 90.1 73.0 78.1 80.2 85.8 97.2 92.0 87.3 89.1 83.7 90.8 88.9 87.3 81.2 87.3 75.9 84.5 91.6 86.3 82.4 85.2 83.3 76.4 85.9 82.1 91.2 84.3 82.1 85.2 89.1 94.4 97.5 91.3 79.9 88.2 74.7 81.2 84.7 90.1 93.4 92.2
11.3 8.2 5.2 4.9 10.2 5.4 12.3 2.2 5.8 4.7 7.5 3.1 10.8 8.3 2.4 7.7 8.5 13.1 9.2 8.5 10.2 4.6 5.3 4.1 11.7 5.8 2.1 3.9 7.4 8.2 12.1 9.2 4.3 9.1 6.3 5.8 7.9 12.2 10.9 3.4 5.1 10.3 7.4 3.5 2.8 9.1 7.8 7.4
80.3 89.3 93.6 86.6 81.1 90.6 85.1 87.7 79.5 75.6 87.1 80.2 90.9 99.1 92.5 85.8 112.1 103.2 100.4 104.1 108.1 91.3 95.8 100.2 87.3 80.1 92.8 90.1 92.7 94.8 90.2 88.7 78.0 84.9 74.1 77.9 80.0 85.6 81.3 80.4 67.7 70.3 74.2 69.3 81.5 84.3 89.1 82.4
6.3 4.9 10.8 7.5 4.0 6.1 10.3 7.2 9.2 7.3 11.7 6.9 3.8 8.4 4.2 9.2 4.1 13.7 6.0 4.4 5.9 3.2 10.8 3.6 5.0 6.9 4.4 3.0 3.9 4.5 7.7 6.9 2.0 11.2 5.3 4.7 9.6 7.4 6.3 8.8 9.1 2.4 3.3 5.9 8.1 5.2 3.6 6.7
Table 6 Comparison of IAC-UPLC-MS/MS with other reported commercially IAC (Vicam, Watertown, MA, USA) clean-up methods for the determination of AFs and ZEN analogs. Matrices
Analyte
Detection
Recovery
LOD (μg kg−1)
Ref.
Chinese yam Platycodon grandiflorum Coix seed 107 Chinese medical herbs Schisandra chinensis fructus, Crataegus fructus, Citri reticulata percarpium, Polygoni Multiflori Radix Fructus aurantii immaturus, Gynostemma pentaphyllum, folia Eriobotrya, dark plum Ginger
6 AFs and 6 ZEN anglogs
UPLC–MS/MS
AFs: 73.0%–112.1% ZEN analogs: 64.7%–97.5%
This study
ZEN AFB1, AFB2, AFG1, AFG AFB1, AFB2, AFG1, AFG2 AFB1, AFB2, AFG1, AFG2
HPLC–FLD HPLC–FLD
80.8%–98.3% 63.3%–109.8%
AFs: 0.03–0.07 ZEN analogs: 0.02–0.06 9.50 0.04–0.02
HPLC–FLD
62.3%–115.0%
0.02–0.10
[31]
ultra-fast (UF) LC–MS/ MS
82.0%–100.2%
0.04–0.30
[32]
[10] [17]
ranging from 0.15 μg kg−1 to 0.54 μg kg−1 and from 0.12 μg kg−1 to 10.5 μg kg−1, respectively. The results indicated that herbal medicines were often contaminated with AF and ZEN analogs because of inappropriate drying, processing procedures and storage [30]).
medicines. To test whether the validated method is feasible for real samples, we analyzed 15 real samples purchased from supermarkets. As shown in Table 7, three samples were found positive for AFs, and 5 samples were found positive for ZEN analogs, with concentration 234
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Table 7 IAC–UPLC–MS/MS was applied in 15 real samples. Numbera
S01 S02 S03 S04 S05 S06 S07 S08 S09 S10 S11 S12 S13 S14 S15 a b
AFs (μg kg−1)
ZEN class (μg kg−1)
AFB1
AFB2
AFG1
AFG2
AFM1
AFM2
ZEN
ZAN
α-ZOL
β-ZOL
α-ZAL
β-ZAL
– < LOQ – – – – – – – – – 0.15 – – –
– – – – – – – – – – – – 0.54 – –
– – – – – – – – – – – – – – –
– – – – – – – – – – – – – – –
– – – – – – – – – – – – – – –
– – – – – – – – – – – – – – –
– – 2.78 – – – 1.59 – – – – – – 0.12 –
– – – – – – – – – – – – – – 10.5
– – – – – – – – – – – – – – –
– – – – – 0.20 – – – – – – – – –
– – – – – – – – – – – – – – –
– – – – – – – – – – – – – 0.19 –
b
S01-S05, Chinese yam; S06-S10, Platycodon grandiflorum; S11-S15, coix seed. –, not detected (below LOD).
4. Conclusions
Anal. Technol. Biomed. Life Sci. 1022 (2016) 118–125. [10] X. Zhang, W. Liu, A.F. Logrieco, M.H. Yang, Z. Ou-yang, X. Wang, Determination of zearalenone in traditional Chinese medicinal plants and related products by HPLCFLD, Food Addit. Contam., Part A 28 (2011) 885–893. [11] A. Logrieco, G. Mule, A. Moretti, A. Bottalico, Toxigenic Fusarium species and mycotoxins associated with maize earrot in Europe, Eur. J. Plant Pathol. 108 (2002) 597–609. [12] F. Minervini, A. Giannoccaro, A. Cavallini, A. Visconti, Investigations on cellular proliferation induced by zearalenone and its derivatives in relation to the estrogenic parameters, Toxicol. Lett. 159 (2005) 272–283. [13] A.V. Sartori, M.H.P.D. Moraes, R.P.D. Santos, Y.P. Souza, A.W.D. Nóbrega, Determination of mycotoxins in cereal-based porridge destined for infant consumption by ultra-high-performance liquid chromatography-tandem mass spectrometry, Food Anal. Methods 10 (2017) 4049–4061. [14] E. Miróabella, P. Herrero, N. Canela, L. Arola, F. Borrull, R. Ras, N. Fontanals, Determination of mycotoxins in plant-based beverages using QuEChERS and liquid chromatography-tandem mass spectrometry, Food Chem. 229 (2017) 366–372. [15] M.N. Irakli, A. Skendi, M.D. Papageorgiou, HPLC-DAD-FLD method for simultaneous determination of mycotoxins in wheat bran, J. Chromatogr. Sci. 55 (2017) 1–7. [16] J. Fitzgerald, P. Leonard, E. Darcy, S. Sharma, R. O'Kennedy, Immunoaffinity chromatography: concepts and applications, Methods, Mol. Biol. 1485 (2017) 27–51. [17] L. Zhang, X. Dou, W. Kong, M. Yang, Assessment of critical points and development of a practical strategy to extend the applicable scope of immunoaffinity column cleanup for aflatoxin detection in medicinal herbs, J. Chromatogr. A 1483 (2017) 56–63. [18] X. Zhang, Immunoassay for the detection of four mycotoxins in maize, PhD thesis, Beijing, China Agricultural University, (2017), pp. 20–39. [19] X. Zhang, S.A. Eremin, K. Wen, X. Yu, C. Li, Y. Ke, Y. Jiang, J. Shen, Z. Wang, Fluorescence polarization immunoassay based on a new monoclonal antibody for the detection of the zearalenone class of mycotoxins in maize, J. Agric. Food Chem. 65 (2017) 2240–2247. [20] K. Yao, K. Wen, W. Shan, S. Xie, T. Peng, J. Wang, H. Jiang, B. Shao, Development of an immunoaffinity column for the highly sensitive analysis of bisphenol A in 14 kinds of foodstuffs using ultra-high-performance liquid chromatography tandem mass spectrometry, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 1080 (2018) 50–58. [21] N.G. Di, E. Cimpoies, A. Dossi, L. Paolini, A. Radeghieri, L. Caimi, D. Ricotta, Polyclonal versus monoclonal immunoglobulin free light chains quantification, Ann. Clin. Biochem. 52 (2015) 327–336. [22] C. Li, Z. Wang, X. Cao, R.C. Beier, S. Zhang, S. Ding, X. Li, J. Shen, Development of an immunoaffinity column method using broad-spectrum specificity monoclonal antibodies for simultaneous extraction and cleanup of quinolone and sulfonamide antibiotics in animal muscle tissues, J. Chromatogr. A 1209 (2008) 1–9. [23] N. Delaunay, V. Pichon, M.C. Hennion, Immunoaffinity solid-phase extraction for the trace-analysis of low-molecular-mass analytes in complex sample matrices, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 745 (2002) 15–37. [24] X. Hu, R. Hu, Z. Zhang, M. Wang, Development of a multiple immunoaffinity column for simultaneous determination of multiple mycotoxins in feeds using UPLC–MS/MS, Anal. Bioanal. Chem. 408 (2016) 1–10. [25] L. Sun, L. Mei, H. Yang, K. Zhang, J. Li, D. Jiang, M. Li, A. Deng, Development and application of immunoaffinity column for the simultaneous determination of norfloxacin, pefloxacin, lomefloxacin, and enrofloxacin in swine and chicken meat samples, Food Anal. Methods 9 (2016) 1–11. [26] S.J. Macdonald, S. Anderson, P. Brereton, R. Wood, A. Damant, Determination of zearalenone in barley, maize and wheat flour, polenta, and maize-based baby food
Two group-specific Mabs with uniform affinity were initially applied for the preparation of an IAC that can simultaneously capture AF and ZEN analogs in edible and medicinal herb (Chinese yam, Platycodon grandiflorum and coix seed) samples. The validated IAC–UPLC–MS/MS method has the advantages of simple pretreatment, excellent clean-up performance and high sensitivity. Compared with conventional IAC, which uses large amounts of one or more Mabs with low cross-reactivities to a group of analogs, the present IAC showed group-specific and a great ability to coextract two groups of analogs in complex edible and medicinal herbs. Notably, the IAC clean-up procedure is not only a time saving but also a cost saving method, and more than one groupspecific Mab may be used for multiple IAC applications for determining various groups of analytes in edible and medicinal herbs. Acknowledgements This work was supported by the Ministry of Science and Technology of People's Republic of China (2017YFF0211003) and Sanming Project of Medicine in Shenzhen (SZSM201611068). References [1] D.-X. Kong, X.-J. Li, H.-Y. Zhang, Where is the hope for drug discovery? Let history tell the future, Drug Discov. Today 14 (2009) 115–119. [2] O. Akerele, Nature's medicinal bounty: don't throw it away, World Health Forum 14 (1993) 390–395. [3] International Agency for Research on Cancer, IARC Monographs on the Evaluation of Carcinogenic Risks to Humans: Some Traditional Herbal Medicines, Some Mycotoxins, Naphthalene and Styrene, IARC Press, Lyon, 2002, pp. 171–179. [4] T.W. Kensler, B.D. Roebuck, G.N. Wogan, J.D. Groopman, Aflatoxin: a 50-year odyssey of mechanistic and translational toxicology, Toxicol. Sci. 120 (Suppl. 1) (2011) S28–S48. [5] M.-H. Yang, J.-M. Chen, X.-H. Zhang, Immunoaffinity column clean-up and liquid chromatography with post-column derivatization for analysis of aflatoxins in traditional Chinese medicine, Chromatographia 62 (2005) 499–504. [6] Z. Han, Y. Zheng, L. Luan, Z. Cai, Y. Ren, Y. Wu, An ultra-high-performance liquid chromatography-tandem mass spectrometry method for simultaneous determination of aflatoxins B1, B2, G1, G2, M1 and M2 in traditional Chinese medicines, Anal. Chim. Acta 664 (2010) 165–171. [7] S. Liu, F. Qiu, W. Kong, J. Wei, X. Xiao, M. Yang, Development and validation of an accurate and rapid LC-ESI-MS/MS method for the simultaneous quantification of aflatoxin B1, B2, G1, and G2, in lotus seeds, Food Control 29 (2013) 156–161. [8] I. Arranz, E. Sizoo, E.H. Van, K. Kroeger, T.M. Legarda, P. Burdaspal, K. Reif, J. Stroka, Determination of aflatoxin B1 in medical herbs: interlaboratory study, J. AOAC Int. 89 (2006) 595–605. [9] M. Li, W. Kong, Y. Li, Mei. Yange, High-throughput determination of multi-mycotoxins in Chinese yam and related products by ultra fast liquid chromatography coupled with tandem mass spectrometry after one-step extraction, J. Chromatogr. B
235
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
medicines by LC-MS-MS, Anal. Bioanal. Chem. 406 (2014) 3031–3303. [30] J. Xue, D. Liu, S. Chen, Y. Liao, Z. Zhou, Overview on external contamination sources in traditional Chinese medicines, World Sci. Technol. 10 (2008) 91–96. [31] G. Yang, X. Liu, J. Zou, W. Fu, J. Xiao, Q. Fang, Determination of aflatoxins in some traditional Chinese medicine rich in pigments by high performance liquid chromatography-fluorescence, J. Food Safety Quality 5 (2014) 1125–1132. [32] Y. Yang, J. Wen, W. Kong, Q. Liu, H. Luo, J. Wang, M. Yang, Simultaneous determination of four aflatoxins and ochratoxin A in ginger after inoculation with fungi by ultra-fast liquid chromatography-tandem mass spectrometry, J. Sci. Food Agric. 96 (2016) 4160–4167.
by immunoaffinity column cleanup with liquid chromatography: interlaboratory study, J. AOAC Int. 88 (2005) 1733–1740. [27] C. Ferrer, A. Lozano, A. Agüera, J. Girón, A.R. Fernández-Alba, Overcoming matrix effects using the dilution approach in multiresidue methods for fruits and vegetables, J. Chromatogr. A 1218 (2011) 7634–7639. [28] C. Ferrer, M.J. Martínezbueno, A. Lozano, A.R. Fernándezalba, Pesticide residue analysis of fruit juices by lc-ms/ms direct injection. One year pilot survey, Talanta 83 (2011) 1552–1561. [29] R. Zheng, H. Xu, W. Wang, R. Zhan, W. Chen, Simultaneous determination of aflatoxin B1, B2, G1, G2, ochratoxin A, and sterigmatocystin in traditional Chinese
236
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/jchromb
Determination of aflatoxin and zearalenone analogs in edible and medicinal herbs using a group-specific immunoaffinity column coupled to ultra-highperformance liquid chromatography with tandem mass spectrometry
T
Shujuan Suna,1, Kai Yaoa,1, Sijun Zhaob, Pimiao Zhenga, Sihan Wanga, Yuyang Zenga, ⁎ Demei Lianga, Yuebin Kec, Haiyang Jianga, a
Beijing Advanced Innovation Center for Food Nutrition and Human Health, College of Veterinary Medicine, China Agricultural University, Beijing Key Laboratory of Detection Technology for Animal-Derived Food Safety, Beijing 100193, People's Republic of China China Animal Health and Epidemiology Center, Qingdao 266032, People's Republic of China c Shenzhen Center for Disease Control and Prevention, Shenzhen 518055, People's Republic of China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Aflatoxins Zearalenone analogs Edible and medicinal herbs Group-specific Immunoaffinity column UPLC–MS/MS
Six aflatoxins (AFs; AF B1, B2, G1, G2, M1 and M2) and six zearalenone (ZEN) analogs (ZEN, zearalanone, αzeralanol, β-zeralanol, α-zearalenol, and β-zearalenol) were simultaneously extracted from edible and medicinal herbs using a group-specific immunoaffinity column (IAC) and then identified by ultra-high-performance liquid chromatography tandem mass spectrometry (UPLC–MS/MS). The IAC was prepared by coupling N-hydroxysuccinimide-activated Sepharose 4B Fast Flow gel with two group-specific monoclonal antibodies. The column capacities to six AFs and six ZEN analogs ranged from 100.2 ng to 167.1 ng and from 59.5 ng to 244.4 ng, respectively. The IAC–UPLC–MS/MS method was developed and validated with three different matrices (Chinese yam [Dioscorea polystachya], Platycodon grandiflorum and coix seed [Semen Coicis]). Recoveries of twelve analytes from edible and medicinal herbs were in the range of 64.7%–112.1%, with relative standard deviations below 13.7%. The limits of quantification were in the range from 0.08 μg kg−1 to 0.2 μg kg−1. The method was proven to be sensitive and accurate, and suitable for the determination of real samples.
1. Introduction Herbal medicines have been used for thousands of years in China, and some herbal medicines have been considered as daily food (known as edible and medicinal herbs) because of their health-promoting functions and disease treatment properties [1]. With the improvement of people's living standards, individuals increasingly care more about their health. To remain healthy, individuals have used many kinds of edible and medicinal herbs, such as Chinese yam, Platycodon grandiflorum, and coix seed, as alternative medicines. A survey conducted by World Health Organization showed that about 70%–80% of the world populations rely on non-conventional medicines comprised mainly of herbal sources in their primary healthcare [2]. Although many health benefits are present in edible and medicinal herbs, the safe consumption of these products has gained much concern because of contaminations in raw materials by aflatoxins (AFs) and zearalenone (ZEN) structural analogs during cultivating, harvesting, processing and storage [3].
⁎
1
Corresponding author. E-mail address: [email protected] (H. Jiang). These authors contributed equally to this work.
https://doi.org/10.1016/j.jchromb.2018.06.012 Received 20 April 2018; Received in revised form 4 June 2018; Accepted 6 June 2018 Available online 07 June 2018 1570-0232/ © 2018 Elsevier B.V. All rights reserved.
AFs are produced by fungi belonging to Aspergillus flavus and Aspergilllus parasiticus under warm and moist conditions, which are designated as a group 1 carcinogenic compound by the International Agency for Research on Cancer [3]. AF B1, B2, G1, G2, M1 and M2 (Fig. 1; AFB1, AFB2, AFG1, AFG2, AFM1 and AFM2, respectively) are the most ubiquitous members of AF family and have received increasing attention because of their great harm to the liver [4]. Recently, AFs have been widely detected in herbal medicines [5–8]. Han et al. reported the mean levels (incidence) of AFB1, B2, G1 and G2 in herbal medicines samples were 1.40 (68.8%), 1.27 (50.0%), 0.50 (43.8%) and 0.94 (43.8%) μg kg−1; AFM1 was also detected with maximum concentrations of 0.70 μg kg−1 [6]. ZEN has also been found in herbal medicines [9, 10], but has not been studied as extensively as AFs. Zhang et al. detected ZEN in coix seed, with levels ranging from 18.7 μg kg−1 to 211.4 μg kg−1. ZEN is a naturally occurring nonsteroidal estrogenic mycotoxin produced by genus Fusarium [11], and its derivatives (Fig. 1; zearalanone [ZAN], α-zeralanol [α-ZAL], β-zeralanol [β-ZAL], α-zearalenol [α-ZOL], β-zearalenol [β-ZOL]) have also been characterized
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Fig. 1. Chemical structures of the AFs and ZEN structural analogs.
against ZEN analogs (named 3D4) [19], 5H3 recognizes six AFs (AFB1, AFB2, AFG1, AFG2, AFM1 and AFM2), and whereas 3D4 recognizes six ZEN analogs (ZEN, ZAN, α-ZAL, β-ZAL, α-ZOL and β-ZOL), respectively. As far as we know, this study is the first report on a multiple mycotoxins IAC for the simultaneous determination of AFs and ZEN analogs in three kinds of edible and medicinal herbs, including Chinese yam, Platycodon grandiflorum and coix seed. Finally, the developed IACLC-MS/MS method proved to be sufficiently sensitive and accurate and hence successfully used in real sample detection.
with estrogenic effects [12]. Therefore, to ensure consumers' health and minimize economic losses from contaminated herbal medicines, sensitive and accurate analytical methods are urgently needed for the detection of AF and ZEN analogs concentrations in herbal medicines. To date, the most common analytical methods include liquid chromatography electrospray ionization tandem mass spectrometry (LC–ESI-MS/MS) [6, 7, 9, 13, 14] or LC coupled with fluorescence detection (FLD) [5, 8, 10, 15]. However, LC–ESI-MS/MS is more preferred because of its excellent selectivity and sensitivity. The complexities of herbal medicine matrices considerably challenge the attainment of accurate data under detection by MS/MS. Therefore, the clean-up performance of analytical methods is of great importance during the analytical process. The current purifying technologies for determination of AFs or ZEN are mainly based on solid phase extraction (SPE) [6], quick, easy, cheap, effective, rugged, and safe (QuEChERS) method [14] and immunoaffinty column (IAC) [5, 7, 8, 10]. SPE is easy to perform and solvent conserving, but retainment of analytes is based on nonspecific binding, which can lead to the coextraction of analytes and matrix impurities. QuEChERS offers a less expensive and time-saving approach to determine target analytes. However, its strong matrix effect is a central issue that cannot be ignored. IAC is a powerful purification technique that relies on the specific recognition between the antibody and complementary analytes [16]. Compared with SPE and QuEChERS, IAC is most frequently used because of its high specificity and efficiency. However, most studies have only detected AFs or ZEN with commercial IACs [7, 10, 17], the IACs may not be suitable for all the herbal medicines, and none has been used to determine two classes of analytes. The application of the screening and determination of two different types of mycotoxin has been largely restricted by the high cost of commercial IACs. Thus, a cost-saving IAC that can purify two classes of mycotoxins is needed for research. An IAC preparation that uses two group-specific monoclonal antibodies (Mabs) has not been reported. Such lack may be due to the difficulty in obtaining ZEN group-specific Mabs and the requirement for steady preparative techniques for further developing a multi-mycotoxin IAC. In this paper, a novel IAC was prepared using a recently obtained Mab against AFs (named 5H3) [18] and a previously obtained Mab
2. Materials and methods 2.1. Reagents and materials The AFs Mab (5H3) [18] and the ZENs Mab (3D4) [19] were previously produced by our laboratory. AFB1, AFB2, AFG1, AFG2, AFM1, AFM2, ZEN, ZAN, α-ZAL, β-ZAL, α-ZOL and β-ZOL standards were purchased from Sigma–Aldrich (St. Louis, MO). N-hydroxysuccinimide (NHS)-activated Sepharose 4B Fast Flow (FF) and polypropylene columns (3 mL) with polytetrafluoroethylene (PTFE) frits were purchased from Biocomma (Shenzhen, China). LC-MS grade acetonitrile (ACN) and methanol (MeOH) and were supplied by Merck (Darmstadt, Germany). Ultrapure water was obtained from a Milli-Q water purification system (Millipore, Billerica, MA, USA). All other chemicals and solvents were of analytical grade orhigher. Stock mixed standard solutions (10 mg mL−1) were prepared by dissolving 10 mg of each standard substance in MeOH (10 mL) in amber glass vials. Working mixed standard solutions (10 μg mL−1) were prepared by diluting the stock solutions with MeOH. The working solutions were stored at 4 °C. Phosphate buffered saline (PBS) was prepared by dissolving 8.8 g NaCl, 0.02 g KCl, 2.9 g NaH2PO4·2H2O and 0.59 g Na2HPO4·12H2O in 1 L of water, and the pH was adjusted to 7.4 with 1 M NaOH aqueous solution. 2.2. Samples Edible and medicinal herb samples including Chinese yam, 229
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
the gels coupled with the AFs Mab and the ZEN analogs Mab were mixed together into a SPE column and washed with 30 mL PBS. Finally, the prepared multiple IAC column was stored at 4 °C in PBS containing 0.02% (w/v) Proclin 300.
Platycodon grandiflorum, and coix seed were randomly purchased from supermarkets in Beijing. All samples were stored at 4 °C. 2.3. LC–MS/MS analysis
2.5. Sample extraction and IAC clean-up
The LC-MS/MS analysis was performed with a Xevo TQ-S triplequadrupole tandem mass spectrometer (Waters, Milford, MA, USA). Chromatographic separation of AFs and ZEN analogs was carried out using an Acquity HSS T3 column (2.1 mm × 100 mm; 1.8 μm; Waters) and an Acquity BEH C18 column (2.1 mm × 100 mm; 1.7 μm; Waters), respectively. The mobile phase for positive-ion electrospray ionization (ESI+) consisted of 0.1% formic acid aqueous solution (mobile phase A1) and ACN:MeOH mixture (50:50, v/v; mobile phase B1). The mobile phase for negative-ion ESI (ESI−) consisted of ultrapure water (mobile phase A2) and ACN (mobile phase B2). The gradient elution program for ESI+ was as follows: 0–5.0 min, 30%–100% B1; 5.0–6.0 min, 100% B1; 6.0–6.5 min, 100%–30% B1; 6.5–8.0 min, 30% B1. The gradient elution program for ESI− was as follows: 0–3.0 min, 15%–85% B2; 3.0–3.5 min, 85%–90% B2; 3.5–4.0 min, 4.0–4.3 min, 90%–15% B2; 4.3–5.5 min, 15% B2. The column temperature and sampler temperature were set at 40 °C and 10 °C, respectively. The flow rate and injection volume were 0.3 mL min−1 and 10 μL, respectively. The MS/MS acquisition was operated in ESI+ and ESI− with multiple reaction monitoring (MRM). The capillary voltage was 2.0 kV for ESI+ and 3.2 kV for ESI−. The source temperature and desolvation temperature were set at 80 °C and 450 °C for two ESI modes, respectively. The nitrogen flow rate was 800 L h−1 and 550 L h−1 for ESI+ and ESI−, respectively. The ultra-pure argon flow rate, used as the collision gas, was 0.13 mL min−1 for both modes. The MS/MS parameters for AFs and ZEN analogs are presented in Table 1.
Samples (Chinese yam and Platycodon grandiflorum) were cut into small pieces (coix seed did not need to be cut), lyophilized for 24 h, and ground to fine powder using a TissueLyser II (Qiagen, Germany), and then the sample powders were passed through 40 mesh screen to obtained the homogenized samples. Homogenized samples (1.00 ( ± 0.01) g) were weighed in 10 mL polypropylene centrifuge tubes. After addition of 4 mL MeOH–water (60:40, v/v), the mixture was mixed for 10 min at 600 rpm with a VXR shaker (IKA, Germany). After centrifugation at 8000 rpm for 10 min, the supernatant was transferred to a 50 mL polypropylene centrifuge tube and diluted to 24 mL with PBS. Subsequently, the diluted extraction solvent was passed through an IAC at a flow of one drop per second, the column was washed with 5 mL PBS and 5 mL water. Finally, the target analytes were eluted with 3 mL MeOH and evaporated to dryness by a gentle stream of N2 at 40 °C. The residue was reconstituted in 1 mL ACN–water (30:70, v/v) and the solution was filtered with 0.22 μm nylon filter membrane before injection into UPLC-MS/MS. 2.6. IAC working conditions optimization Loading conditions were evaluated by using the following MeOHwater /PBS ratios (v/v, 1:7, 1:5, 1:4 and 1:3, which correspond to MeOH percentages of 7.5%, 10%, 12% and 15%), and the optimal MeOH volume (1, 2, 3, 4, and 5 mL) were both determined to achieve satisfactory recoveries. Extraction solvent (4 mL MeOH–water) was diluted with PBS (28, 20, 16 and 12 mL) and spiked with 50 μL 40 ng mL−1 mixed standard solution. The resulting mixture was loaded onto the IAC. After washing with 5 mL PBS and 5 mL water, the two classes of compounds were eluted with various volume of MeOH and evaporated to dryness by a gentle stream of N2 at 40 °C.
2.4. IAC preparation The immunosorbent was prepared in accordance with our previous study [20]. NHS-activated Sepharose 4B FF gel (1 mL) was transferred to an empty polypropylene SPE tube (3 mL) with a PTFE filter. The gel was washed with 10 mL 1 mM hydrochloric acid and then with 10 mL 0.2 M NaHCO3 (pH 8.3) solution. After this procedure, the pre-equilibrated gel was mixed with 0.5 mg AFs Mab and 0.5 mg ZEN analogs Mab solution. The mixture was gently shaken end over end at room temperature for 2 h. The eluted solution was then collected to detect the amount of Mabs by a Nanodrop 2000 UV–Vis spectrometry (Thermo Fisher, USA) and the coupling efficiency was determined. The mixture was blocked by reacting with the blocking buffer (0.2 M NaHCO3, pH 8.3, containing 50 mM ethanolamine) for 2 h at room temperature to block the unreacted active sites on the sorbent. Subsequently, the blocked gel was consecutively washed twice with 40 mL Tris–HCl buffer (0.1 M, pH 7.0) and 40 mL HOAc-NaOAc buffer (0.1 M, pH 4.0). Finally,
2.7. Column capacity and reproducibility Column capacity was determined by over-loading solutions containing 1000 ng of each AFs and ZEN analogs that was mixed with loading solution. The subsequent steps were the same as those described in Section 2.5. The reproducibility was tested by calculating the relative standard deviation (RSD) of column capacities of five IACs in one batch and three IACs in three consecutive days. The whole column capacity process was performed as described in the above-mentioned paragraph.
Table 1 MS/MS parameters and retention time of 12 analytes. Analytes AFB1 AFB2 AFG1 AFG2 AFM1 AFM2 ZEN ZAN α-ZAL, β-ZAL, α-ZOL, β-ZOL ⁎
Retention time (min) 2.38 2.20 2.19 2.01 1.77 1.61 3.09 3.11 2.74 2.57 2.79 2.60
Mode +
ESI ESI+ ESI+ ESI+ ESI+ ESI+ ESI− ESI− ESI− ESI− ESI− ESI−
Precursor ion (m/z)
Product ion (m/z)
Cone voltage (V)
Collision energy (eV)
313.1 315.1 329.1 331.2 329.1 331.2 317.0 319.0 321.0 321.0 319.0 319.0
285.1⁎/240.9 287.1⁎/259.1 243.1⁎/128.0 245.1⁎/217.0 273.1⁎/259.1 273.1⁎/285.0 175.0⁎/273.0 275.3⁎/205.0 277.0⁎/303.0 277.0⁎/303.0 275.2⁎/301.0 275.2⁎/301.0
45 45 10 30 20 20 25 25 25 25 25 25
20/35 26/30 26/45 24/26 22/20 20/20 24/21 22/21 22/22 22/22 22/22 22/21
Ion used for quantification. 230
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
2.8. Recovery
3.2. IAC conditions optimization
The recoveries of the method were evaluated by using blank samples (1.00 g) spiked with standard analytes at four concentration levels (LOQ, 5-fold LOQ, 10-fold LOQ and 50-fold LOQ for each analyte), each concentration was repeated five times.
The evaluation of different IAC working conditions is essential for a satisfying recovery. Organic solvents or a mixture of organic solvent and water are often used to extract AF and ZEN analogs. However, the sample extract containing a high percentage of MeOH was found unfit for directly loading the IAC. To obtain an optimal composition of the loading solution, various loading conditions were evaluated. Recoveries were the highest when the MeOH–water/PBS ratio was increased to 1:5 (Fig. 2A) and decreased when the MeOH–water/PBS ratio exceeded 1:5. This pattern may be explained by the fact that a high ratio means a high concentration of MeOH in the loading buffer, and MeOH greatly influences the analyte–antibody binding. A low percentage of organic solvent can achieve satisfactory recoveries and reduce non-specific adsorption [23]. Therefore, MeOH–water/PBS ratio of 1:5 was used in the following experiment. After the washing process, the analytes retained on the IAC were eluted with 1–5 mL of pure MeOH. Previous studies reported that analytes can be released completely from the analyte-antibody complex when using pure MeOH or a MeOH-water mixture as elution solvent [22, 24, 25]. However, a small quantity of water in the elution solvent may increase the elution volume and render the evaporation difficult. No recovery increase was found when the elution volume reached 3 mL (Fig. 2B).
3. Results and discussion 3.1. IAC preparation Polyclonal antibody (Pab) and Mab are the most important reagents for preparing IAC, and there is a direct correlation between the antibody performance and the potential use of IAC. In order to obtain a consistent clean-up and extraction performance of IAC, the long-term preparation of massive and reproducible antibodies must be guaranteed. However, Pabs are usually generated from just one animal, and the supply of reproducible Pabs is commonly limited [21]. Hence, Mabs are generally used instead of Pabs to obtain IACs with high repeatability. To develop an IAC which is simultaneously group-selective to both AFs and ZEN analogs, the process requires group-specific Mabs that can recognize several analogs. Mabs with uniform affinity to all analytes in the same class are highly necessary to avoid the competition between the structurally related analogs for limited antibody binding sites during the IAC loading process [22]. The competition in the IAC loading process may lead to an unequal adsorption of analytes onto IAC when the analytes are in high concentration in samples. Thus, two group-specific Mabs against both AF (5H3) and ZEN analogs (3D4) with uniform cross-reactivities were utilized in this study. According to previous research, the Mab 3D4 had cross-reactivities of 57.5%–108.5%. The cross-reactivities of our recently prepared Mab (5H3) were determined by enzyme-linked immunosorbent assay and shown to lie in the range of 64.4%–140.6%. The details on the IC50 (half maximal inhibitory concentration) and cross-reactivities are listed in Table 2. The two Mabs were coupled with NHS-activated Sepharose 4B FF to prepare an IAC that can concomitantly extract AF and ZEN analogs (12 analytes in total) by using only one column. The IAC was prepared by packing the Sepharose 4B FF that was coupled with two Mabs into an empty column. NHS-activated Sepharose 4B FF was selected not only because of its chemically and biologically inertness but also because of the benefit of fast flow. Compared with the most commonly used solid support, namely, CNBractivated Sepharose 4B, the NHS-activated Sepharose 4B FF exhibited a significant flow rate improvement, which can conserve much time during the IAC loading procedure. 3D4 Mab (0.5 mg) and 5H3 Mab (0.5 mg) were immobilized on 1 mL NHS-activated Sepharose 4B FF gel. The coupling efficiency was determined in accordance with the ratio of coupled antibodies to added antibodies. In this study, coupling efficiency was shown to be 98.1%.
3.3. Column capacity and reproducibility of IAC Column capacities for AF and ZEN analogs were determined and summarized in Table 3. In this study, two group-specific Mabs against AF and ZEN analogs were combined for the preparation of IAC. High column capacities were in the range of 100.2–167.1 ng for AFs and 59.5–244.4 ng for the ZEN analogs, respectively. Interestingly, the lower cross-reactivities of the Mab 3D4 for β-ZOL (57.5%) and ZAN (66.4%) than for other ZEN analogs appeared to slightly influence the column capacity (59.5 ng for β-ZOL and 244.4 ng for ZAN). This phenomenon also occurred in the Mab 5H3 for AFM2. In summary, the column capacities of the IAC column for each analyte were mainly dependent on the experimental data and not the cross-reactivities [25]. The prepared IAC in this research showed wider recognition abilities than those in previous works for the six AFs (AFB1, AFB2, AFG1, AFG2, AFM1 and AFM2) and six ZEN analogs (ZEN, ZAN, α-ZAL, β-ZAL, α-ZOL and β-ZOL) under study. Zhang et al. [17] applied commercial IAC products to analyze AFB1, AFB2, AFG1 and AFG2, but the parameters of IAC and the information on the antibody were not given. Some researchers reported IACs that only recognized limited numbers of AFs [5, 6, 8] or ZEN [10, 26]. Hu et al. [24] developed a multiple IAC that can capture AFB1, AFB2, AFG1, AFG2 and ZEN, but the remaining two AFs and five ZEN analogs were not included. Mabs (40 mg) were also used for the generation of an IAC, and the column capacities were 194.2, 185.4, 159.0, 150.5, and 135.5 ng for AFB1, AFB2, AFG1, AFG2 and ZEN, respectively. In this study, 1 mg Mab was utilized for achieving comparable column capacities because of the extremely low IC50 values of two group-specific Mabs. A greater number of IACs can be produced at a relatively low cost when a fixed amount of Mabs are used. We conducted intra-assay and inter-assay column capacity determinations to evaluate the reproducibility of IAC (Table 3). The intraand inter-day RSD values of IAC column capacities were 2.3%–10.1% and 2.8%–13.7%, respectively, which reveal the satisfactory reproducibility of IAC.
Table 2 Cross-reactivities and IC50 values of AF and ZEN analogs. Analytes
IC50 (pmol mL−1)
Cross-reactivities (%)a
AFB1 AFB2 AFG1 AFG2 AFM1 AFM2 ZEN ZAN α-ZAL β-ZAL α-ZOL β-ZOL
0.2972 0.3686 0.2114 0.3443 0.3001 0.4612 0.1289 0.1941 0.1230 0.1724 0.1188 0.2241
100.0% 80.6% 140.6% 86.3% 99.0% 64.4% 100.0% 66.4% 104.8% 74.8% 108.5% 57.5%
a
3.4. Method validation The developed method was validated in terms of matrix effect, linearity, specificity, limit of detection (LOD), limit of quantification (LOQ), and recovery, and precision of the method was analyzed by spiking mixed standards into edible and medicinal herb samples.
Cross-reactivities = (IC50 of AFB1 or ZEN /IC50 of analogs) × 100%. 231
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Fig. 2. The optimized working conditions of IAC determined by UPLC–MS/MS (n = 3). (A) MeOH–water /PBS ratios; (B) elution volume. Table 3 IAC capacities and RSDs for each analyte. AFs
Column capacity (ng) Intra-day RSD (%) Inter-day RSD (%)
ZEN analogs
AFB1
AFB2
AFG1
AFG2
AFM1
AFM2
ZAN
ZEN
α-ZOL
β-ZOL
α-ZAL
β-ZAL
130.0 4.1 2.9
152.4 5.5 8.1
126.6 2.3 4.8
160.8 8.3 7.9
100.2 2.5 4.1
167.1 3.2 2.8
244.4 7.1 6.8
209.2 2.6 3.7
183.8 3.9 5.2
59.5 7.5 9.9
239.1 5.4 13.7
175.7 10.1 5.6
Table 4 Matrix effect, LOQ and LOQ of the method. Analytes
AFB1
AFB2
AFG1
AFG2
AFM1
AFM2
ZEN
ZAN
α-ZAL
β-ZAL
α-ZOL
β-ZOL
Matrices
Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed Chinese yam Platycodon grandiflorum Coix seed
LOD (μg kg−1)
0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.05 0.05 0.05 0.07 0.07 0.07 0.03 0.03 0.03 0.05 0.05 0.05 0.06 0.06 0.06 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.02 0.02 0.02
LOQ (μg kg−1)
Slopes of two standard curves
0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.15 0.15 0.15 0.20 0.20 0.20 0.08 0.08 0.08 0.15 0.15 0.15 0.20 0.20 0.20 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10
232
Linearity range (μg kg−1)
Matrix-matched
Solvent
SSE (%)
0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.15–20.00 0.15–20.00 0.15–20.00 0.20–20.00 0.20–20.00 0.20–20.00 0.08–20.00 0.08–20.00 0.08–20.00 0.15–20.00 0.15–20.00 0.15–20.00 0.20–20.00 0.20–20.00 0.20–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00 0.10–20.00
22,925 16,736 22,745 17,990 14,837 16,634 17,026 13,666 19,206 2756 2418 3289 5272 4899 5953 24,678 22,658 26,116 22,767 24,815 24,522 10,067 10,711 9610 14,685 15,018 15,613 14,260 15,042 15,476 14,561 17,156 19,193 17,419 17,134 24,387
20,198
113.5 82.9 112.6 112.5 92.8 104.0 102.6 82.4 115.8 96.9 85.0 115.6 89.7 83.4 101.3 106.2 97.5 112.4 85.6 93.3 92.2 97.0 103.2 92.6 83.9 85.8 89.2 82.1 86.6 89.1 80.8 95.2 106.5 84.1 85.5 80.3
15,994
16,588
2844
5877
23,236
26,597
10,379
17,504
17,370
18,022
20,374
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Fig. 3. Typical MRM chromatograms of Platycodon grandiflorum matrix-matched standard solutions of 5 ng mL−1 for each analyte: (A) six AFs; (B) six ZEN analogs.
medicinal herb samples. The limits were based on signal-to-noise ratios of 3:1 and 10:1 for AF and ZEN analogs, respectively. The LODs ranged from 0.02 μg kg−1 to 0.07 μg kg−1, and the LOQs were in the range from 0.08 μg kg−1 to 0.2 μg kg−1. The recovery and precision of five blank samples fortified at three fortification levels of 12 mixed standards were determined within the same day. The mean recoveries lay between 64.7% and 112.1%, with RSDs lower than 13.7% (n = 5) in the intra-day experiment (Table 5). The results fulfilled the current demand for simultaneous detection of AF and ZEN analogs. The methods reported previously were compared with the present study (Table 6). The recoveries of the present work were slightly lower than those of others, but the LODs in this study were improved because of the good purifying performance of the multiple IAC. Moreover, six AF and six ZEN analogs can be simultaneously extracted and purified by a single IAC.
For the development of a reliable IAC–UPLC–MS/MS method, the matrix effect in case of inaccurate results must be carefully considered because ion suppression or enhancement induced by sample matrices and interferences during the ESI process may occur. The matrix effect is commonly evaluated as a signal suppression/enhancement effect (SSE) [27], which is quantitatively calculated in accordance with the following equation: SSE (%) = (the slope of matrix-matched standard curves / the slope of solvent standard curves) × 100%. SSE < 100% is defined as ion suppression, SSE > 100% is defined as ion enhancement, and an SSE of 80%–120% is considered to be negligible [28]. As shown in Table 4, the SSE values of all the herbal medicines matrices are in the negligible range, indicating that all the analytes in this study can be determined directly by using solvent standard curves [29]. The correlation coefficients (R2) of the matrix-fortified calibration curves were all > 0.995 and showed very good linearities. The specificity of the method was evaluated by analyzing blank samples spiked with the standard solvent. No interfering peaks were observed at the retention time of each analyte and demonstrated a high specificity of the established method (Fig. 3). The LODs and LOQs of the method were determined by confirming the minimum detectable concentration of 12 analytes in edible and
3.5. Application to real samples The validated results demonstrate that our IAC-UPLC-MS/MS method is applicable for determining AF and ZEN analogs in herbal 233
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Table 5 Recoveries and RSD (n = 5) of 12 analytes in three edible and medicinal herbs. Analytes
AFB1
AFB2
AFG1
AFG2
AFM1
AFM2
ZEN
ZAN
α-ZAL
β-ZAL
α-ZOL
β-ZOL
Spiked levels (μg kg−1)
0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00 0.15 0.75 1.50 7.50 0.20 1.00 2.00 10.00 0.08 0.40 0.80 4.00 0.15 0.75 1.50 7.50 0.20 1.00 2.00 10.00 0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00 0.10 0.50 1.00 5.00
Chinese yam
Platycodon grandiflorum
coix seed
Recovery (%)
RSD (%)
Recovery (%)
RSD (%)
Recovery (%)
RSD (%)
89.4 95.2 90.7 93.1 77.3 84.2 86.7 88.2 81.7 92.3 85.9 82.5 88.9 95.8 90.5 93.2 87.2 80.3 92.3 90.6 79.1 84.6 81.5 78.3 73.2 78.5 85.3 80.5 64.7 71.3 74.6 70.4 80.7 69.1 77.3 81.4 73.6 85.4 79.9 83.1 80.6 76.1 88.7 85.0 74.4 83.2 81.9 85.4
4.6 9.1 7.3 5.3 4.1 3.7 8.3 4.3 6.6 3.9 12.1 7.2 8.5 12.4 7.1 4.5 3.8 2.4 9.2 5.3 4.3 2.8 7.1 9.9 8.5 3.6 11.3 7.3 2.9 4.7 6.9 5.8 11.5 6.4 8.8 3.7 4.3 8.1 2.4 5.5 5.3 3.8 10.2 4.2 3.7 1.9 7.5 4.6
93.4 98.1 85.2 89.5 88.4 81.4 91.2 90.1 73.0 78.1 80.2 85.8 97.2 92.0 87.3 89.1 83.7 90.8 88.9 87.3 81.2 87.3 75.9 84.5 91.6 86.3 82.4 85.2 83.3 76.4 85.9 82.1 91.2 84.3 82.1 85.2 89.1 94.4 97.5 91.3 79.9 88.2 74.7 81.2 84.7 90.1 93.4 92.2
11.3 8.2 5.2 4.9 10.2 5.4 12.3 2.2 5.8 4.7 7.5 3.1 10.8 8.3 2.4 7.7 8.5 13.1 9.2 8.5 10.2 4.6 5.3 4.1 11.7 5.8 2.1 3.9 7.4 8.2 12.1 9.2 4.3 9.1 6.3 5.8 7.9 12.2 10.9 3.4 5.1 10.3 7.4 3.5 2.8 9.1 7.8 7.4
80.3 89.3 93.6 86.6 81.1 90.6 85.1 87.7 79.5 75.6 87.1 80.2 90.9 99.1 92.5 85.8 112.1 103.2 100.4 104.1 108.1 91.3 95.8 100.2 87.3 80.1 92.8 90.1 92.7 94.8 90.2 88.7 78.0 84.9 74.1 77.9 80.0 85.6 81.3 80.4 67.7 70.3 74.2 69.3 81.5 84.3 89.1 82.4
6.3 4.9 10.8 7.5 4.0 6.1 10.3 7.2 9.2 7.3 11.7 6.9 3.8 8.4 4.2 9.2 4.1 13.7 6.0 4.4 5.9 3.2 10.8 3.6 5.0 6.9 4.4 3.0 3.9 4.5 7.7 6.9 2.0 11.2 5.3 4.7 9.6 7.4 6.3 8.8 9.1 2.4 3.3 5.9 8.1 5.2 3.6 6.7
Table 6 Comparison of IAC-UPLC-MS/MS with other reported commercially IAC (Vicam, Watertown, MA, USA) clean-up methods for the determination of AFs and ZEN analogs. Matrices
Analyte
Detection
Recovery
LOD (μg kg−1)
Ref.
Chinese yam Platycodon grandiflorum Coix seed 107 Chinese medical herbs Schisandra chinensis fructus, Crataegus fructus, Citri reticulata percarpium, Polygoni Multiflori Radix Fructus aurantii immaturus, Gynostemma pentaphyllum, folia Eriobotrya, dark plum Ginger
6 AFs and 6 ZEN anglogs
UPLC–MS/MS
AFs: 73.0%–112.1% ZEN analogs: 64.7%–97.5%
This study
ZEN AFB1, AFB2, AFG1, AFG AFB1, AFB2, AFG1, AFG2 AFB1, AFB2, AFG1, AFG2
HPLC–FLD HPLC–FLD
80.8%–98.3% 63.3%–109.8%
AFs: 0.03–0.07 ZEN analogs: 0.02–0.06 9.50 0.04–0.02
HPLC–FLD
62.3%–115.0%
0.02–0.10
[31]
ultra-fast (UF) LC–MS/ MS
82.0%–100.2%
0.04–0.30
[32]
[10] [17]
ranging from 0.15 μg kg−1 to 0.54 μg kg−1 and from 0.12 μg kg−1 to 10.5 μg kg−1, respectively. The results indicated that herbal medicines were often contaminated with AF and ZEN analogs because of inappropriate drying, processing procedures and storage [30]).
medicines. To test whether the validated method is feasible for real samples, we analyzed 15 real samples purchased from supermarkets. As shown in Table 7, three samples were found positive for AFs, and 5 samples were found positive for ZEN analogs, with concentration 234
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
Table 7 IAC–UPLC–MS/MS was applied in 15 real samples. Numbera
S01 S02 S03 S04 S05 S06 S07 S08 S09 S10 S11 S12 S13 S14 S15 a b
AFs (μg kg−1)
ZEN class (μg kg−1)
AFB1
AFB2
AFG1
AFG2
AFM1
AFM2
ZEN
ZAN
α-ZOL
β-ZOL
α-ZAL
β-ZAL
– < LOQ – – – – – – – – – 0.15 – – –
– – – – – – – – – – – – 0.54 – –
– – – – – – – – – – – – – – –
– – – – – – – – – – – – – – –
– – – – – – – – – – – – – – –
– – – – – – – – – – – – – – –
– – 2.78 – – – 1.59 – – – – – – 0.12 –
– – – – – – – – – – – – – – 10.5
– – – – – – – – – – – – – – –
– – – – – 0.20 – – – – – – – – –
– – – – – – – – – – – – – – –
– – – – – – – – – – – – – 0.19 –
b
S01-S05, Chinese yam; S06-S10, Platycodon grandiflorum; S11-S15, coix seed. –, not detected (below LOD).
4. Conclusions
Anal. Technol. Biomed. Life Sci. 1022 (2016) 118–125. [10] X. Zhang, W. Liu, A.F. Logrieco, M.H. Yang, Z. Ou-yang, X. Wang, Determination of zearalenone in traditional Chinese medicinal plants and related products by HPLCFLD, Food Addit. Contam., Part A 28 (2011) 885–893. [11] A. Logrieco, G. Mule, A. Moretti, A. Bottalico, Toxigenic Fusarium species and mycotoxins associated with maize earrot in Europe, Eur. J. Plant Pathol. 108 (2002) 597–609. [12] F. Minervini, A. Giannoccaro, A. Cavallini, A. Visconti, Investigations on cellular proliferation induced by zearalenone and its derivatives in relation to the estrogenic parameters, Toxicol. Lett. 159 (2005) 272–283. [13] A.V. Sartori, M.H.P.D. Moraes, R.P.D. Santos, Y.P. Souza, A.W.D. Nóbrega, Determination of mycotoxins in cereal-based porridge destined for infant consumption by ultra-high-performance liquid chromatography-tandem mass spectrometry, Food Anal. Methods 10 (2017) 4049–4061. [14] E. Miróabella, P. Herrero, N. Canela, L. Arola, F. Borrull, R. Ras, N. Fontanals, Determination of mycotoxins in plant-based beverages using QuEChERS and liquid chromatography-tandem mass spectrometry, Food Chem. 229 (2017) 366–372. [15] M.N. Irakli, A. Skendi, M.D. Papageorgiou, HPLC-DAD-FLD method for simultaneous determination of mycotoxins in wheat bran, J. Chromatogr. Sci. 55 (2017) 1–7. [16] J. Fitzgerald, P. Leonard, E. Darcy, S. Sharma, R. O'Kennedy, Immunoaffinity chromatography: concepts and applications, Methods, Mol. Biol. 1485 (2017) 27–51. [17] L. Zhang, X. Dou, W. Kong, M. Yang, Assessment of critical points and development of a practical strategy to extend the applicable scope of immunoaffinity column cleanup for aflatoxin detection in medicinal herbs, J. Chromatogr. A 1483 (2017) 56–63. [18] X. Zhang, Immunoassay for the detection of four mycotoxins in maize, PhD thesis, Beijing, China Agricultural University, (2017), pp. 20–39. [19] X. Zhang, S.A. Eremin, K. Wen, X. Yu, C. Li, Y. Ke, Y. Jiang, J. Shen, Z. Wang, Fluorescence polarization immunoassay based on a new monoclonal antibody for the detection of the zearalenone class of mycotoxins in maize, J. Agric. Food Chem. 65 (2017) 2240–2247. [20] K. Yao, K. Wen, W. Shan, S. Xie, T. Peng, J. Wang, H. Jiang, B. Shao, Development of an immunoaffinity column for the highly sensitive analysis of bisphenol A in 14 kinds of foodstuffs using ultra-high-performance liquid chromatography tandem mass spectrometry, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 1080 (2018) 50–58. [21] N.G. Di, E. Cimpoies, A. Dossi, L. Paolini, A. Radeghieri, L. Caimi, D. Ricotta, Polyclonal versus monoclonal immunoglobulin free light chains quantification, Ann. Clin. Biochem. 52 (2015) 327–336. [22] C. Li, Z. Wang, X. Cao, R.C. Beier, S. Zhang, S. Ding, X. Li, J. Shen, Development of an immunoaffinity column method using broad-spectrum specificity monoclonal antibodies for simultaneous extraction and cleanup of quinolone and sulfonamide antibiotics in animal muscle tissues, J. Chromatogr. A 1209 (2008) 1–9. [23] N. Delaunay, V. Pichon, M.C. Hennion, Immunoaffinity solid-phase extraction for the trace-analysis of low-molecular-mass analytes in complex sample matrices, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 745 (2002) 15–37. [24] X. Hu, R. Hu, Z. Zhang, M. Wang, Development of a multiple immunoaffinity column for simultaneous determination of multiple mycotoxins in feeds using UPLC–MS/MS, Anal. Bioanal. Chem. 408 (2016) 1–10. [25] L. Sun, L. Mei, H. Yang, K. Zhang, J. Li, D. Jiang, M. Li, A. Deng, Development and application of immunoaffinity column for the simultaneous determination of norfloxacin, pefloxacin, lomefloxacin, and enrofloxacin in swine and chicken meat samples, Food Anal. Methods 9 (2016) 1–11. [26] S.J. Macdonald, S. Anderson, P. Brereton, R. Wood, A. Damant, Determination of zearalenone in barley, maize and wheat flour, polenta, and maize-based baby food
Two group-specific Mabs with uniform affinity were initially applied for the preparation of an IAC that can simultaneously capture AF and ZEN analogs in edible and medicinal herb (Chinese yam, Platycodon grandiflorum and coix seed) samples. The validated IAC–UPLC–MS/MS method has the advantages of simple pretreatment, excellent clean-up performance and high sensitivity. Compared with conventional IAC, which uses large amounts of one or more Mabs with low cross-reactivities to a group of analogs, the present IAC showed group-specific and a great ability to coextract two groups of analogs in complex edible and medicinal herbs. Notably, the IAC clean-up procedure is not only a time saving but also a cost saving method, and more than one groupspecific Mab may be used for multiple IAC applications for determining various groups of analytes in edible and medicinal herbs. Acknowledgements This work was supported by the Ministry of Science and Technology of People's Republic of China (2017YFF0211003) and Sanming Project of Medicine in Shenzhen (SZSM201611068). References [1] D.-X. Kong, X.-J. Li, H.-Y. Zhang, Where is the hope for drug discovery? Let history tell the future, Drug Discov. Today 14 (2009) 115–119. [2] O. Akerele, Nature's medicinal bounty: don't throw it away, World Health Forum 14 (1993) 390–395. [3] International Agency for Research on Cancer, IARC Monographs on the Evaluation of Carcinogenic Risks to Humans: Some Traditional Herbal Medicines, Some Mycotoxins, Naphthalene and Styrene, IARC Press, Lyon, 2002, pp. 171–179. [4] T.W. Kensler, B.D. Roebuck, G.N. Wogan, J.D. Groopman, Aflatoxin: a 50-year odyssey of mechanistic and translational toxicology, Toxicol. Sci. 120 (Suppl. 1) (2011) S28–S48. [5] M.-H. Yang, J.-M. Chen, X.-H. Zhang, Immunoaffinity column clean-up and liquid chromatography with post-column derivatization for analysis of aflatoxins in traditional Chinese medicine, Chromatographia 62 (2005) 499–504. [6] Z. Han, Y. Zheng, L. Luan, Z. Cai, Y. Ren, Y. Wu, An ultra-high-performance liquid chromatography-tandem mass spectrometry method for simultaneous determination of aflatoxins B1, B2, G1, G2, M1 and M2 in traditional Chinese medicines, Anal. Chim. Acta 664 (2010) 165–171. [7] S. Liu, F. Qiu, W. Kong, J. Wei, X. Xiao, M. Yang, Development and validation of an accurate and rapid LC-ESI-MS/MS method for the simultaneous quantification of aflatoxin B1, B2, G1, and G2, in lotus seeds, Food Control 29 (2013) 156–161. [8] I. Arranz, E. Sizoo, E.H. Van, K. Kroeger, T.M. Legarda, P. Burdaspal, K. Reif, J. Stroka, Determination of aflatoxin B1 in medical herbs: interlaboratory study, J. AOAC Int. 89 (2006) 595–605. [9] M. Li, W. Kong, Y. Li, Mei. Yange, High-throughput determination of multi-mycotoxins in Chinese yam and related products by ultra fast liquid chromatography coupled with tandem mass spectrometry after one-step extraction, J. Chromatogr. B
235
Journal of Chromatography B 1092 (2018) 228–236
S. Sun et al.
medicines by LC-MS-MS, Anal. Bioanal. Chem. 406 (2014) 3031–3303. [30] J. Xue, D. Liu, S. Chen, Y. Liao, Z. Zhou, Overview on external contamination sources in traditional Chinese medicines, World Sci. Technol. 10 (2008) 91–96. [31] G. Yang, X. Liu, J. Zou, W. Fu, J. Xiao, Q. Fang, Determination of aflatoxins in some traditional Chinese medicine rich in pigments by high performance liquid chromatography-fluorescence, J. Food Safety Quality 5 (2014) 1125–1132. [32] Y. Yang, J. Wen, W. Kong, Q. Liu, H. Luo, J. Wang, M. Yang, Simultaneous determination of four aflatoxins and ochratoxin A in ginger after inoculation with fungi by ultra-fast liquid chromatography-tandem mass spectrometry, J. Sci. Food Agric. 96 (2016) 4160–4167.
by immunoaffinity column cleanup with liquid chromatography: interlaboratory study, J. AOAC Int. 88 (2005) 1733–1740. [27] C. Ferrer, A. Lozano, A. Agüera, J. Girón, A.R. Fernández-Alba, Overcoming matrix effects using the dilution approach in multiresidue methods for fruits and vegetables, J. Chromatogr. A 1218 (2011) 7634–7639. [28] C. Ferrer, M.J. Martínezbueno, A. Lozano, A.R. Fernándezalba, Pesticide residue analysis of fruit juices by lc-ms/ms direct injection. One year pilot survey, Talanta 83 (2011) 1552–1561. [29] R. Zheng, H. Xu, W. Wang, R. Zhan, W. Chen, Simultaneous determination of aflatoxin B1, B2, G1, G2, ochratoxin A, and sterigmatocystin in traditional Chinese
236