*Article* **Identification of Azoxystrobin Glutathione Conjugate Metabolites in Maize Roots by LC-MS**

#### **Giuseppe Dionisio 1,\* ,**† **, Maheswor Gautam 2,**†**,**‡ **and Inge Sindbjerg Fomsgaard <sup>2</sup>**


Academic Editor: Thomas Letzel

Received: 9 June 2019; Accepted: 3 July 2019; Published: 5 July 2019

**Abstract:** Xenobiotic detoxification in plant as well as in animals has mostly been studied in relationship to the deactivation of the toxic residues of the compound that, surely for azoxystrobin, is represented by its β-methoxyacrylate portion. In maize roots treated for 96 h with azoxystrobin, the fungicide accumulated over time and detoxification compounds or conjugates appeared timewise. The main detoxified compound was the methyl ester hydrolysis product (azoxystrobin free acid, 390.14 *m*/*z*) thought to be inactive followed by the glutathione conjugated compounds identified as glutathione conjugate (711.21 *m*/*z*) and its derivative lacking the glycine residue from the GSH (654.19 *m*/*z*). The glycosylated form of azoxystrobin was also found (552.19 *m*/*z*) in a minor amount. The identification of these analytes was done by differential untargeted metabolomics analysis using Progenesis QI for label free spectral counting quantification and MS/MS confirmation of the compounds was carried out by either Data Independent Acquisition (DIA) and Data Dependent Acquisition (DDA) using high resolution LC-MS methods. Neutral loss scanning and comparison with MS/MS spectra of azoxystrobin by DDA and MSe confirmed the structures of these new azoxystrobin GSH conjugates.

**Keywords:** azoxystrobin; glutathione; glutathione conjugate; untargeted metabolomics

#### **1. Introduction**

Azoxystrobin is one of the widely used fungicides in plants. The metabolism of azoxystrobin in plants is well described [1] and recently our laboratory has reported mass spectrometric methods to identify and quantify azoxystrobin metabolites in plants [2–4]. However, glutathione conjugation of azoxystrobin in plants has not been reported yet [1,5]. The only known report on GSH conjugation of azoxystrobin comes from an animal metabolism study where azoxystrobin was reported to conjugate with glutathione in the cyanophenyl aromatic ring with the electrophilic carbon in the –ortho position to the cyano group [6]. However, a later in vitro experiment undertaken to conjugate azoxystrobin with glutathione by plant-derived glutathione transferase was unsuccessful [5]. Glutathione (GSH) conjugation is a crucial step in the detoxification process of xenobiotics in living organisms [5–7]. GSH conjugation occurs in the cytoplasm of plant cells when electrophilic centers in xenobiotics are attacked by the nucleophilic thiol (–SH) group of glutathione [5]. This process increases the solubility of xenobiotics, which can thus be transported from cytoplasm to vacuoles. As increasing concentration of GSH-xenobiotic could be potentially harmful for plant cells, the conjugates are sequentially degraded in vacuoles by enzymatic actions thereby enabling plants to tackle harmful

effects of xenobiotics [7]. During the degradation of GSH-xenobiotic conjugate in vacuoles, glycine residue from GSH is cleaved off enzymatically, which results in glutamic acid and cysteine residue of GSH attached to the xenobiotic [5,7].

GSH detoxification has been shown to occur in xenobiotics compound by conjugation, at a structural level, to aliphatic, aromatic, benzylic, disulfide, and thioester groups [8]. However, no other conjugation than aromatic conjugation of azoxystrobin with GSH has been reported or studied so far. We hypothesized that the carbon in the –CN group of azoxystrobin is a favorable candidate for the nucleophilic attack by the –SH group in GSH. The synthesis of analytical standards of such conjugates of GSH with azoxystrobin was out of the scope of the present study. Thus, we aimed to verify the presence of GSH-azoxystrobin conjugates at the –CN group by suspect screening of such metabolites and others possible ones by using high-resolution mass spectrometry. Our initial candidate structure for GSH conjugates were aromatic and cyano conjugation with azoxystrobin. However, the algorithm used here to detect GSH conjugates of azoxystrobin, a priori of its structure determination, is to detect at MS/MS level the simultaneous presence of GSH fragments neutral loss and the main azoxystrobin MS/MS fragments. Untargeted LC-MS/MS analysis was automated by using Progenesis QI for metabolomics where the candidate MS scans were compared also over time for increased appearance due to the plant detoxification of azoxystrobin. Once we had identified by MS/MS spectral matching we did confirm its reproducibility in detection by setting up a method to use in future for targeted detection by quadrupole-linear ion trap mass spectrometry to develop sensitive triple-quadrupole mass spectrometry-based methods for quantification of GSH-azoxystrobin conjugates.

The conjugation of GSH to xenobiotics is operated enzymatically by plant Glutathione Transferases (GSTs), which is a multi-functional superfamily of detoxification enzymes [9]. They possess a hydrophobic binding domain (H-site) and the ability to bind glutathione (G-site). The greatest similarity between members of the GST family is observed in the GSH binding domain containing a motif of four highly conserved amino acids [10]. Here, a catalytically essential tyrosine or serine activates glutathione by lowering the pKa of the thiol group from around pH 9 to approximately pH 6, thus enhancing the rate of nucleophilic attack by the resulting thiolate anion toward electrophilic substrates at physiological pH [11–13]. The H site is more variable, accounting for the broad range of endogenous and xenobiotic substrates utilized by this family of GSH dependent enzymes [13].

Despite the fact that reduced glutathione (GSH) is capable of undergoing spontaneous conjugation with compounds containing electrophilic centers for the detoxification of several xenobiotics compounds, different classes of GSTs are, in fact, involved because they can lower by several fold the Km related to the nucleophilic thiolate anion attack of GSH to the electrophilic substrates, with concomitant displacement of a leaving group (i.e., a proton) which more rarely takes part in addition reactions [14].

In our former study, we established a maize root culture model system to study the biotransformation of azoxystrobin in plants [3]. In the current study, we report on more results from the use of maize root culture as a model system to study not only the demethylated metabolite, azoxystrobin free acid (AzFA), which represents the main plant metabolite but also, in lower extent of abundancy, the GSH-azoxystrobin conjugates.

#### **2. Results**

The first observation to note in treated maize roots samples as compared to control is the accumulation of azoxystrobin (monoisotopic mass 403.1168), as expected, judged by looking at the appearance of the peak at 404.1197 *m*/*z* representing MH+ of azoxystrobin in the MS survey profile (Figure 1), that is accumulating after 96h as compared to the control. Further evidence that the peak of 404.1197 *m*/*z* eluting at about 28.5 min represents for real the MH+ of azoxystrobin is given by its fragmentation pattern when observing its MS/MS profile (Figure 2). Furthermore, the major methyl ester hydrolysis product of azoxystrobin, azoxystrobin free acid, was differentially present in the treated samples as compared to the controls (Supplementary Figure S1).

*Molecules* **2019**, *24*, x FOR PEER REVIEW 3 of 14

**Figure 1.** Q-TOF survey profile of the total metabolites extracted from maize root in control (panel **A**) and azoxystrobin treated ones (Panel **B**). Untargeted MS of roots methanol extracts at 96h has revealed the presence of a differential peak occurring at about 28.5 min with 404.1197 as *m/z* which is correspond to MH+ of azoxystrobin (exact mass, 403.1168; IUPAC Name, methyl (*E*)-2-[2-[6-(2 cyanophenoxy)pyrimidin-4-yl]oxyphenyl]-3-methoxy-prop-2-enoate). In the ordinate is shown the signal intensity in percentage; in the abscissa the retention time (RT) in minutes. **Figure 1.** Q-TOF survey profile of the total metabolites extracted from maize root in control (panel **A**) and azoxystrobin treated ones (Panel **B**). Untargeted MS of roots methanol extracts at 96h has revealed the presence of a differential peak occurring at about 28.5 min with 404.1197 as *m*/*z* which is correspond to MH+ of azoxystrobin (exact mass, 403.1168; IUPAC Name, methyl (*E*)-2-[2-[6-(2-cyanophenoxy)pyrimidin-4-yl]oxyphenyl]-3-methoxy-prop-2-enoate). In the ordinate is shown the signal intensity in percentage; in the abscissa the retention time (RT) in minutes. **Figure 1.** Q-TOF survey profile of the total metabolites extracted from maize root in control (panel **A**) and azoxystrobin treated ones (Panel **B**). Untargeted MS of roots methanol extracts at 96h has revealed the presence of a differential peak occurring at about 28.5 min with 404.1197 as *m/z* which is correspond to MH+ of azoxystrobin (exact mass, 403.1168; IUPAC Name, methyl (*E*)-2-[2-[6-(2 cyanophenoxy)pyrimidin-4-yl]oxyphenyl]-3-methoxy-prop-2-enoate). In the ordinate is shown the signal intensity in percentage; in the abscissa the retention time (RT) in minutes.

**Figure 2.** Q-TOF MS/MS DDA profile of the root metabolite extracted from maize root at 28.667 min as retention time. The precursor specie isolated by the quadrupole at 404.09 *m/z* was subjected to high energy ramping from 10 to 40 eV and its MS/MS profile shown here clearly indicates the presence of fragments of azoxystrobin as assessed by comparing the peak with similar fragmentation pattern **Figure 2.** Q-TOF MS/MS DDA profile of the root metabolite extracted from maize root at 28.667 min as retention time. The precursor specie isolated by the quadrupole at 404.09 *m/z* was subjected to high energy ramping from 10 to 40 eV and its MS/MS profile shown here clearly indicates the presence of fragments of azoxystrobin as assessed by comparing the peak with similar fragmentation pattern **Figure 2.** Q-TOF MS/MS DDA profile of the root metabolite extracted from maize root at 28.667 min as retention time. The precursor specie isolated by the quadrupole at 404.09 *m*/*z* was subjected to high energy ramping from 10 to 40 eV and its MS/MS profile shown here clearly indicates the presence of fragments of azoxystrobin as assessed by comparing the peak with similar fragmentation pattern from massbank (http://www.massbank.jp/RecordDisplay.jsp?id=AU324504&dsn=Athens\_Univ). The peak annotation has extracted by Progenesis QI Metascope built-in fragmentation and annotation system.

The conjugated azoxystrobin metabolite analysis relied on the fact that the MS/MS spectra of the main peaks of azoxystrobin-conjugates should contain, at least, one of the most abundant MS/MS peak as shown in Figure 2. Using Masslynx function for extracting selected peaks from a chromatogram, it is possible to create one extracted chromatogram profile at survey as well as at MS/MS level. By scanning the MS/MS fragment in order to filter out peaks with 372 *m*/*z* from the MS/MS spectra of the treated versus untreated controls it was possible to visualize the presence of four main peaks at MS/MS level that contains the species 372 *m*/*z* (Figure 3). Four potential azoxystrobin metabolites containing at MS/MS level the 372 *m*/*z* species were found at retention time (RT) respectively: (a) 18.36 min; (b) 18.68 min; (c) 19.72 min; and (d) 30.5 min, while RT 28.79 min was related to the azoxystrobin itself. The conjugated azoxystrobin metabolite analysis relied on the fact that the MS/MS spectra of the main peaks of azoxystrobin-conjugates should contain, at least, one of the most abundant MS/MS peak as shown in Figure 2. Using Masslynx function for extracting selected peaks from a chromatogram, it is possible to create one extracted chromatogram profile at survey as well as at MS/MS level. By scanning the MS/MS fragment in order to filter out peaks with 372 m/z from the MS/MS spectra of the treated versus untreated controls it was possible to visualize the presence of four main peaks at MS/MS level that contains the species 372 m/z (Figure 3). Four potential azoxystrobin metabolites containing at MS/MS level the 372 m/z species were found at retention time (RT) respectively: a) 18.36 min; b) 18.68 min; c) 19.72 min; and d) 30.5 min, while RT 28.79 min was related to the azoxystrobin itself.

*Molecules* **2019**, *24*, x FOR PEER REVIEW 4 of 14

from massbank (http://www.massbank.jp/RecordDisplay.jsp?id=AU324504&dsn=Athens\_Univ). The

**Figure 3.** Extracted MS chromatograms at MS/MS (MSe) level of the precursor specie 372 *m/z*. A comparative extracted chromatogram for control untreated (**top** MS/MS trace) and treated with azoxystrobin (**lower** MS/MS trace) extracted and analyzed after 96 h. Four potential azoxystrobin metabolites are differentially displayed containing the 372 *m/z* species as visualized by the related extracted chromatogram at *m/z* 372 with different retention times: a) RT 18.36 min; b) RT 18.68 min; c) RT 19.72 min; and d) RT 30.5 min, being RT 28.79 min related to the azoxystrobin itself. **Figure 3.** Extracted MS chromatograms at MS/MS (MSe) level of the precursor specie 372 *m*/*z*. A comparative extracted chromatogram for control untreated (**top** MS/MS trace) and treated with azoxystrobin (**lower** MS/MS trace) extracted and analyzed after 96 h. Four potential azoxystrobin metabolites are differentially displayed containing the 372 *m*/*z* species as visualized by the related extracted chromatogram at *m*/*z* 372 with different retention times: (a) RT 18.36 min; (b) RT 18.68 min; (c) RT 19.72 min; and (d) RT 30.5 min, being RT 28.79 min related to the azoxystrobin itself.

Matches were found in relation to the metabolites proposed to be azoxystrobin-glutathione conjugates at the cyano group due to the bonding of sulphur with the electrophilic carbon with monoisotopic mass 710.200 (M + H 711.30 *m/z*) and after glycil neutral loss 653.1792 (M + H 654.19 *m/z)* (Figure 4). The peak eluting at retention time (RT) 18.36 min is therefore compatible with precursor ion 711.30 *m/z* (Figure 5) and the peak eluting at RT 18.67 min could be represented by the same azoxystrobin-glutathione conjugate with glycil neutral loss, represented by the precursor ion 654.19 *m/z* (Supplementary Figure 2). The correct identification of potential azoxystrobin conjugate at RT 19.72 min was indicated by the neutral loss of 162 Da (glucosyl) and corresponds to azoxystrobin free acid glucoside with a monoisotopic mass of 551.1540 Da (M + H 552.19 *m/z*) (Supplementary Figure 3). An analysis with QTRAP IDA method revealed similarity in MS/MS spectra for 711.3 *m/z* including the evidence for a neutral loss of 129 corresponding to the loss of Matches were found in relation to the metabolites proposed to be azoxystrobin-glutathione conjugates at the cyano group due to the bonding of sulphur with the electrophilic carbon with monoisotopic mass 710.200 (MH+ 711.30 *m*/*z*) and after glycil neutral loss 653.1792 (MH+ 654.19 *m*/*z*) (Figure 4). The peak eluting at retention time (RT) 18.36 min is therefore compatible with precursor ion 711.30 *m*/*z* (Figure 5) and the peak eluting at RT 18.67 min could be represented by the same azoxystrobin-glutathione conjugate with glycil neutral loss, represented by the precursor ion 654.19 *m*/*z* (Supplementary Figure S2). The correct identification of potential azoxystrobin conjugate at RT 19.72 min was indicated by the neutral loss of 162 Da (glucosyl) and corresponds to azoxystrobin free acid glucoside with a monoisotopic mass of 551.1540 Da (MH+ 552.19 *m*/*z*) (Supplementary Figure S3). An analysis with QTRAP IDA method revealed similarity in MS/MS spectra for 711.3 *m*/*z* including the evidence for a neutral loss of 129 corresponding to the loss of glutamic acid residue as shown in Supplementary Figure S4. Finally, the correct identification of potential azoxystrobin conjugate at RT 30.55 min was revealed by the neutral loss discovery of demethylated azoxystrobin-GSH conjugate of the precursor specie 697.48 *m*/*z* (Supplementary Figure S5).

glutamic acid residue as shown in Supplementary Figure 4. Finally, the correct identification of potential azoxystrobin conjugate at RT 30.55 min was revealed by the neutral loss discovery of demethylated azoxystrobin-GSH conjugate of the precursor specie 697.48 *m/z* (Supplementary Figure

**Figure 4.** Neutral loss discovery of azoxystrobin-GSH conjugate at MS survey conditions of the precursor specie 711 *m/z*. Possible identification of potential azoxystrobin conjugate at RT 18.10–18.36 min. The analysis of MS survey profile for untreated (panel **A**) and treated sample (panel **B**) reveals that both 711.25 and 654.22 *m/z* species are present only in the treated sample. The glycil neutral loss (654.22 *m/z*) is already occurring at +5eV low energy level (MS1 or MS survey). The azoxystrobinglutathione conjugate (panel **C**) at the cyano group has the MS/MS signature of azoxystrobin alongside the evidence of neutral loss of the glycil residue (neutral loss 57). **Figure 4.** Neutral loss discovery of azoxystrobin-GSH conjugate at MS survey conditions of the precursor specie 711 *m*/*z*. Possible identification of potential azoxystrobin conjugate at RT 18.10–18.36 min. The analysis of MS survey profile for untreated (panel **A**) and treated sample (panel **B**) reveals that both 711.25 and 654.22 *m*/*z* species are present only in the treated sample. The glycil neutral loss (654.22 *m*/*z*) is already occurring at +5eV low energy level (MS1 or MS survey). The azoxystrobin-glutathione conjugate (panel **C**) at the cyano group has the MS/MS signature of azoxystrobin alongside the evidence of neutral loss of the glycil residue (neutral loss 57).

**Figure 5.** Analysis of the precursor specie 711 *m/z*. The identification of potential azoxystrobin conjugate at RT 18.36 min was possible by the analysis of MS survey profile (panel **A**) of azoxystrobin treated roots and its DDA MS/MS profile (panel **B**). The isolated MS/MS fragmentation profile reveals that the precursor of this azoxystrobin-conjugate/metabolite fits well with the azoxystrobin glutathione conjugate (panel **C**) at the cyano group compatible with the compound 711.30 *m/z* as confirmed by the presence of 372 *m/z* in MS/MS of the isolated peak at 711.30 *m/z* (panel B). Further confirmation of this compound has been done by studying the neutral losses. **Figure 5.** Analysis of the precursor specie 711 *m*/*z*. The identification of potential azoxystrobin conjugate at RT 18.36 min was possible by the analysis of MS survey profile (panel **A**) of azoxystrobin treated roots and its DDA MS/MS profile (panel **B**). The isolated MS/MS fragmentation profile reveals that the precursor of this azoxystrobin-conjugate/metabolite fits well with the azoxystrobin glutathione conjugate (panel **C**) at the cyano group compatible with the compound 711.30 *m*/*z* as confirmed by the presence of 372 *m*/*z* in MS/MS of the isolated peak at 711.30 *m*/*z* (panel B). Further confirmation of this compound has been done by studying the neutral losses.

The presence of these metabolites of azoxystrobin in maize roots was also investigated by differential untargeted metabolomics using Progenesis QI ver. 2.0. A database of SDF structures to search such metabolites was custom made from known and possible metabolites and validation of such metabolites was performed by the internal Metascope theoretical fragmentation engine using DDA MS/MS fragmentation profile both at MS1 and the most abundant MS/MS fragments. The presence of these metabolites of azoxystrobin in maize roots was also investigated by differential untargeted metabolomics using Progenesis QI ver. 2.0. A database of SDF structures to search such metabolites was custom made from known and possible metabolites and validation of such metabolites was performed by the internal Metascope theoretical fragmentation engine using DDA MS/MS fragmentation profile both at MS1 and the most abundant MS/MS fragments.

The top relative maximum abundance is shown in Table 1 and it was recorded respectively for compounds at *m/z* 404.1068, 372.0798, 549.1369, 549.1715, 654.1935, 711.2189, 557.1727, and 552.1740. The validation of each peak was performed by looking at the control versus the treated root MS/MS profile for each *m/z* species above which a signature of azoxystrobin MS/MS spectra could be found. The top relative maximum abundance is shown in Table 1 and it was recorded respectively for compounds at *m*/*z* 404.1068, 372.0798, 549.1369, 549.1715, 654.1935, 711.2189, 557.1727, and 552.1740. The validation of each peak was performed by looking at the control versus the treated root MS/MS profile for each *m*/*z* species above which a signature of azoxystrobin MS/MS spectra could be found.


**Table 1.** Possible azoxystrobin containing peaks differentially present in the treated maize root as compared untreated roots over 96 h. The searched azoxystrobin SDF structures and related metabolites present here were not all validated since most of them were also present in the control roots. Their relative max abundance as compared to the control and spectral counting is shown in the last column, where clearly the relative max abundance was recorded respectively for compounds at *m*/*z* 404.1068, 372.0798, 549.1369, 549.1715, 654.1935, 711.2189, 557.1727, and 552.1740. In bold are reported the compounds which have been confirmed being azoxystrobin metabolites.


**Scan** 

Not all the metabolites of azoxystrobin were found by the Progenesis Metascope engine but at least the main azoxystrobin conjugates were detected in order of significance (ANOVA) and abundance. The peaks were presented by precursor ions 654.1935 *m*/*z*, 711.2189 *m*/*z*, and 552.1740 *m*/*z*, corresponding respectively to azoxystrobin-GSH conjugate with glycil loss, azoxystrobin-GSH conjugate, and azoxystrobin free acid glucoside conjugate (Figure 6). Time course accumulation of selected metabolites have been chosen for visualization in Progenesis QI where the untreated control (CTRL) and the azoxystrobin treated samples (AZO) have been analyzed at 3 h, 24 h, 48 h, and 96 h (Figure 6). The typical representation of the trend of *m*/*z* over time and its isotope distribution can be visualized in 2D or 3D and the related area quantified by relative units which are normalized by abundance related to a standard (i.e., CTRL) in Figure 6. Not all the metabolites of azoxystrobin were found by the Progenesis Metascope engine but at least the main azoxystrobin conjugates were detected in order of significance (ANOVA) and abundance. The peaks were presented by precursor ions 654.1935 *m/z*, 711.2189 *m/z*, and 552.1740 *m/z*, corresponding respectively to azoxystrobin-GSH conjugate with glycil loss, azoxystrobin-GSH conjugate, and azoxystrobin free acid glucoside conjugate (Figure 6). Time course accumulation of selected metabolites have been chosen for visualization in Progenesis QI where the untreated control (CTRL) and the azoxystrobin treated samples (AZO) have been analyzed at 3 h, 24 h, 48 h, and 96 h (Figure 6). The typical representation of the trend of *m/z* over time and its isotope distribution can be visualized in 2D or 3D and the related area quantified by relative units which are normalized by abundance related to a standard (i.e., CTRL) in Figure 6.

*Molecules* **2019**, *24*, x FOR PEER REVIEW 8 of 14

**Figure 6.** Relative Ion abundance distribution of azoxystrobin-GSH (glutathione) conjugate at MS survey conditions as quantified by the Progenesis QI software for metabolomics. The label free relative spectral counting quantification analysis (here depicted as 3D isotope abundance) has revealed the differential presence of the *m/z* 711.2189, accumulating over time (max at 96 h) in azoxystrobin treated roots versus the untreated control. The same trend has been detected for the peaks 654.1935 *m/z* and 552.1740 *m/z*. **Figure 6.** Relative Ion abundance distribution of azoxystrobin-GSH (glutathione) conjugate at MS survey conditions as quantified by the Progenesis QI software for metabolomics. The label free relative spectral counting quantification analysis (here depicted as 3D isotope abundance) has revealed the differential presence of the *m*/*z* 711.2189, accumulating over time (max at 96 h) in azoxystrobin treated roots versus the untreated control. The same trend has been detected for the peaks 654.1935 *m*/*z* and 552.1740 *m*/*z*.

All the rest of Progenesis identified peaks at MS1 containing at MS/MS level the 372 *m/z* species, shown in Table 1, were not validated because they were also present in the control roots extract and lacked full similarity to all MS/MS spectra of azoxystrobin. All the rest of Progenesis identified peaks at MS1 containing at MS/MS level the 372 *m*/*z* species, shown in Table 1, were not validated because they were also present in the control roots extract and lacked full similarity to all MS/MS spectra of azoxystrobin.

The QTRAP MS was hence used to setup a quantification method to be further implemented in future screening of the 711.3 *m/z* metabolite in plant once it would be commercially available to be used as standards. The MS conditions used for the future QQQ detection of 711.3 *m/z* metabolite are summarized in Table 2. The QTRAP MS was hence used to setup a quantification method to be further implemented in future screening of the 711.3 *m*/*z* metabolite in plant once it would be commercially available to be used as standards. The MS conditions used for the future QQQ detection of 711.3 *m*/*z* metabolite are summarized in Table 2.

**Table 2.** Mass spectrometric parameters used for the Multiple Reaction Monitoring-Information Dependent Acquisition (MRM-IDA) spectra acquisition of glutathione conjugate of azoxystrobin.

**(psi) CAD IS (V) TEM** 

**(°C)** 

**GS 1 (psi)** 

**GS 2 (psi)** 

**type Q1 Q3 DP (V) EP (V) CE (V) CXP (V) CUR** 

MRM 711 404 60 10 30 13 35 High 4300 450 90 50


**Table 2.** Mass spectrometric parameters used for the Multiple Reaction Monitoring-Information Dependent Acquisition (MRM-IDA) spectra acquisition of glutathione conjugate of azoxystrobin.

#### **3. Discussion**

Azoxystrobin is a strobilurin fungicide that was developed by Zeneca UK Ltd., in 1992. Its mechanism of action in fungi and bacteria rely on inhibiting the electron transport system by binding the Qo site of mitochondrial cytochrome bcl complex to inhibit microbial respiration [15]. The toxophore (chemical group that produces the toxic effect) portion of the molecule is actually not the cyano group, but the β-methoxyacrylate portion that is double methylated and that, hence, should represent the immediate sensitive region of the molecules for enzymatic deactivation by conjugation or demethylation. One of the azoxystrobin environmental degradation products, azoxystrobin free acid, was present in the methanol extracts of maize roots for the treated samples as compared to the controls (Supplementary Figure S1) indicating an active demethylation mechanism of detoxification occurring *in vivo*. This initial demethylation opens the molecule to a definitive inactivation of its toxophore by glycosylation as was shown by the presence of the peak with 552.1740 *m*/*z* (Supplementary Figure S3).

The conjugation of azoxystrobin with glutathione by enzymatic action of glutathione S-transferase enzyme has not been described yet in plants, although a glutathione (GSH) conjugate of azoxystrobin has been identified in rats where sulphur is conjugated to the phenolic ring in the ortho position relative to the cyano group where the conjugate species have shown *m*/*z* values of 708.18 and 694.16 respectively for the fully methylated and demethylated GSH azoxystrobin conjugate [6] not found here in our experiments. Here, we present mass spectrometric evidences that GSH was conjugated with azoxystrobin through the –CN group thereby potentially allowing detoxification of azoxystrobin. We have also proposed chemical structures of the corresponding precursors of *m*/*z* 711.2189 (Figures 4 and 5) and 654.1935 (Supplementary Figure S2) as being full azoxystrobin-GSH and azoxystrobin-cysteine-γglutamic acid conjugates. During the MS survey at low energy, the glycine neutral loss is already present in the MS/MS spectra of *m*/*z* 711.2189 (Figure 4), but the elution of the peak with *m*/*z* 654.1935 is at different time than that with *m*/*z* 711.2189 meaning that the loss could be happening *in vivo*. In barley, Glutathione S-conjugates accumulates in the plant vacuole and that the first step of its degradation, the formation of the respective T-glutamylcysteinyl-S-conjugate, is catalyzed by a vacuolar carboxypeptidase [16]. Following the action of a vacuolar carboxypeptidase that degrades GS-conjugates by cleaving the glycine, it is also reported a removal of the amino terminal glutamic acid residue, cleaved also successively enzymatically [17]. Besides, a membrane associated γ-glutamyl transpeptidase (γ-GTase) has been identified [18], as part of the transport from the vacuole to the cytosol, that then removes the terminal glutamic acid leaving the S-cysteinyl derivative [7]. The loss of the glycine residue from glutathione conjugates of a xenobiotic is well documented [5,19] and is known to occur during transport of xenobiotic glutathione conjugates from cytosol to vacuoles (for review see [5]). Alongside the glycine neutral loss (∆57) (Figure 4), γ-Glutamic acid neutral loss (∆129) has been detected (Supplementary Figure S4) confirming the nature of the analyte with precursor ion *m*/*z* 711.2189. The neutral loss of 129 Da corresponding to the loss of glutamic acid from GSH conjugates is previously described [20] and validated by the method for simultaneous screening of GSH and CN adducts using precursor ion (PI) and neutral loss (NL) scans dependent product ion spectral acquisition and datamining tools on QTRAP mass spectrometry [21]. Here, hence, we have demonstrated that the 711.3 *m*/*z* metabolite possesses a 129 Da neutral loss (Supplementary Figure S4) and that can be quantified by QQQ MS methods (Table 2) once the standard is available as pure compound.

Naturally occurring isothiocyanate compounds present in cruciferous vegetables, e.g., sulforaphane, are powerful inducers of glutathione S-transferase activities in laboratory animals, meaning that the GSH conjugation at the cyano group is operated enzymatically and the reactive group is represented by the carbonyl portion of the isothiocyanate [22]. Similar results have been shown for phenyl isocyanate detoxification by glutathione conjugation [23]. Here we have proposed and have shown evidence that glutathione is conjugated to the electrophilic carbon of the cyano group (–CN) of azoxystrobin. Further experiments are currently underway in order to identify possible enzymes, i.e., glutathione S-transferases, that might mediate in vivo GSH conjugation and detoxification of azoxystrobin in maize roots.

#### **4. Materials and Methods**

#### *4.1. Chemicals*

Analytical grade azoxystrobin (purity 99.4%) and sucrose (≥99.5%) were purchased from Sigma-Aldrich (Copenhagen, Denmark). Azoxystrobin free acid ((*E*)-2-(2-((6-(2-cyanophenoxy) pyrimidin-4-yl)oxy)phenyl)-3-methoxyacrylic acid) (10 ng/µL in acetonitrile) was procured from Dr. Ehrenstorfer GmbH (Augsburg, Germany). Acetonitrile and methanol (TOF/MS grade) were obtained from Fisher Chemical (Roskilde, Denmark). Milli-Q water was generated from a Milli-Q system (Millipore, MA, USA). Murashige and Skoog (M&S) media was purchased from Duchefa Biochemie (Harlem, The Netherlands).

#### *4.2. Maize Root Culture Growth*

Maize seeds (*cv*. Kaspian) were purchased from KWS Scandinavia (Vejle, Denmark) and maize roots were cultured as explained in our earlier study [3]. Briefly, maize seeds were surface sterilized with ethanol and sodium hypochlorite and rinsed with sterilized Milli-Q water. The seeds were transferred into petri plates containing sucrose and full-strength M&S media that had been solidified with phytagel. The pH of the media was maintained at 5.8. The plates were incubated under light/dark condition for 16/8 h at 23 ◦C for seven days. The roots from germinated seeds were detached and transferred into Erlenmeyer flasks containing half-strength M&S and sucrose where the pH was maintained at 5.8. The flasks were covered with aluminum foil to maintain a dark condition within the flasks. The flasks were shaken at 100 rpm at 25 ◦C for seven days and the maize root culture was ready for azoxystrobin uptake and metabolism study.

#### *4.3. Azoxystrobin Uptake*

Azoxystrobin was dissolved in methanol and added to sterile half-strength liquid M&S growth medium (pH 5.8) contained in a beaker to yield a final azoxystrobin concentration 100 µM in 500 mL medium as explained in Gautam et al. (2018). Maize roots were transferred into the beaker and swirled gently to mix azoxystrobin thoroughly. A beaker with similar content except azoxystrobin was used as control. Azoxystrobin treatment continued for three hours after which a portion of roots (5–10 g) was cut-off with sterile scissors, washed thoroughly with sterile water, wiped with tissue paper, wrapped in aluminum foil, and flash frozen in liquid nitrogen. Control maize roots was sampled similarly after three hours. The samples were labelled as 3 h sample.

The treatment solution was decanted off, maize roots were first washed with sterilized Milli-Q water and then with M&S medium, and finally transferred to sterile liquid M&S medium devoid of azoxystrobin. The control was also washed and transferred similarly. The flasks were wrapped with aluminum foil to prevent exposure to light and allowed to stand in a sterile bench throughout the uptake experiment period. Samples were collected after 3, 24, 48, 72, and 96 h of initial exposure with azoxystrobin.

#### *4.4. Sample Preparation and Extraction*

Frozen maize roots were freeze dried on a DryWinner 6-85 freeze drier (Holm & Halby, Brøndby, Denmark) while still wrapped in aluminum foil. The freeze-dried roots were ground in GenoGrinder 100 (Spex, Metuchen, NJ, USA). 20.0 mg of homogenized root tissue was weighed in 2-mL Eppendorf tube and suspended in 1 mL of 50% methanol (*v*/*v*) containing 0.1% formic acid. The tube was shaken vigorously, ultrasonicated for 15 min, and centrifuged at 16,000× *g* at 20 ◦C for 15 min in a Sigma 1-14K microcentrifuge (Buch and Holm, Herlev, Denmark). The supernatant was filtered with 0.22 µm KX Syringe PTFE filter (Mikrolab, Aarhus, Denmark) and analyzed immediately with or stored at −20 ◦C before analysis.

#### *4.5. LC-MS*/*MS Analysis*

#### 4.5.1. Q-TOF Analysis

Methanol content in the extracts were adjusted to 25% before injecting. The liquid chromatography system was composed of a nano Acquity UPLC (Waters, Milford, CT, USA) equipped with a nano Trap column (Waters XSelect™ HHS T3, 5 µm beads diameter, 180 µm × 20 mm) for accumulating the injected samples and an analytical capillary column with the same chemistry (Waters nano RP-C18 HHS T3 type, 1.8µm, 150µm × 100 mm) for sample separation. The nano capillary column was attached with a nano ESI source to a Waters Q-TOF premiere (Waters, Milford, USA). The Q-TOF operational conditions were the following: 3.2 kV for capillary cone voltage and extraction, 30 kV as sampling cone voltage, and 3.4 kV as ion guide voltage, and the source temperature was set to 120 ◦C. The instrument was calibrated in V positive mode using the MS/MS profile of M2H + 785.8426 *m*/*z* GluFib-B ([Glu1]-Fibrinopeptide B human, F3261, Sigma Aldrich, Merck, Germany) as Z-lock mass (reference calibration) at 23 eV as collision energy obtaining a calibration range from 72 to 1286 Da, and with a ppm error of +/− 2.

The entire length of the LC run was 42 min. Mobile phase A was 0.1% formic acid (FA, Sigma Aldrich, Germany) and phase B was acetonitrile (ACN, Thermo-Fischer Scientific, Hvidovre, Denmark) with 0.1% FA. The gradient conditions were from 15% phase B to 70% phase B in 25 min and from 25 min to 27 min from 70% phase B to 85% phase B; from 27 to 30 min, the percentage of mobile phase B raised from 85% to 95%. After 30 min the percentage of phase B was constant set to 95%. Data acquisition was performed in Masslynx version 4.1 (Waters, Milford, USA) either as data-independent acquisition (DIA) mode by all ion fragmentation (MSe) with MS1 and MS/MS (MS2) range from 50 to 2000 *m*/*z*, or data dependent (DDA). For the DDA mode the MS1 survey was acquired from 50 to 1000 *m*/*z* and MS/MS was acquired from 500 Da above after peak deisotoping: scan time 1.5 s with variable collision energy from 5 to 40 V.

Differential untargeted azoxystrobin metabolite, by the means of spectral counting comparison, was performed by Progenesis QI, version 2.0 (Nonlinear Dynamics, a Waters company, Newcastle upon Tyne, UK). The import of raw MSe data acquired by Masslynx was transformed into centroid data by M2H + 785.8426 *m*/*z* GluFib-B lock mass data calibration and refinement including dead time correction and charge deconvolution. Data runs, in triplicate, were hence layered into 2D graphs obtained by the retention time versus the *m*/*z* indexing of the MS1 and MS/MS peaks. After the multirun 2D alignment the peptide3D Apex3D algorithm was used to peak picking and quantification [24]. Label-free ion/adducts quantitation was performed by general all ion compound normalization using built in LOWESS (locally weighted scatterplot smoothing) algorithm and weighted spectral counting alongside the ion/adduct isotopic peaks [25]. The search algorithm was the built-in Progenesis Metascope using a custom Structure Data File (SDF) databases created by joining together many single chemical 2D structures of azoxystrobin precursors as SDF files by Progenesis SDF studio. Chemical 2D structures of azoxystrobin precursors were downloaded from ChemSpider website or manually drawn either by ChemDraw Pro (CambridgeSoft, PerkinElmer, Waltham, MA, USA) or BIOVIA Draw 2018 (BIOVIA, Accelrys, San Diego, CA, USA). The Progenesis Metascope search of the custom SDF database was carried out assuming 50 ppm error for precursor ions and 30 ppm for theoretical fragments.

#### 4.5.2. QTRAP Analysis

The LC-MS/MS analysis was done with a QTRAP 4500 mass spectrometer (AB Sciex, Framingham, MA, USA) coupled to an Agilent 1260 Infinity HPLC system (Santa Clara, CA, USA). Reversed phase HPLC was done with a BDS Hypersil C18 column (250 × 2.1; 5 µm) (Thermo Scientific, Waltham, MA, USA) using a gradient with eluent A 0.1% formic acid in Milli-Q water and B 0.1% formic acid in acetonitrile. Column compartment was maintained at 30 ◦C. The column was equilibrated with 20% B for 10 min in the beginning of each run. The linear gradient was increased from 20% to 95% B (0–15 min) and maintained at 95% B (15–25 min). The mobile phase flow rate was maintained at 300 µL/min with 5 µL injection volume. Samples were filtered through a 0.22 µm KX Syringe PTFE filters (Mikrolab, Aarhus, Denmark) before LC-MS/MS analysis.

The QTRAP MS was used in Information Dependent Acquisition (IDA) mode. After the suspect screening of glutathione conjugates with Q-TOF analysis, MS/MS spectra of the selected hits were acquired using IDA.

The MS/MS spectra from one to two of the most intense peaks matching the survey MRM transitions outlined in Table 1 were acquired if the peak intensity was above IDA threshold of 3000 counts per second (cps). Dynamic exclusion was applied to avoid acquiring MS/MS spectra from the same transition for the next 15 s after 3 occurrences of a transition, thus allowing for the detection of co-eluting peaks. Two EPI experiments were done for the two most intense peaks of each transition meeting the IDA criteria. EPI spectra were acquired between *m*/*z* 80 to 720 in positive ionization mode with a scan rate at 20,000 amu per second. Two EPI spectra from each transition was summed to give a final spectrum which resulted in a dwell time of 1.6 s for a complete MRM-EPI scan. During the acquisition of EPI spectra, DP was ramped between 50 to 70 V. Collision energy of 35 V and a spread of 15 V was used thus acquiring EPI at CE 20, 35, and 50 V.

**Supplementary Materials:** The supplementary materials are available online.

**Author Contributions:** Conceptualization, G.D., I.S.F., M.G.; methodology, G.D. and M.G.; software, (Q-TOF) G.D., (QTRAP) M.G.; data analysis, G.D. I.S.F., M.G.; data curation, G.D. I.S.F., M.G.; writing—original draft preparation, G.D. and M.G.; writing—review and editing, I.S.F., G.D. and M.G.; supervision, I.S.F.

**Funding:** The study was funded by the PhD project (Project no. 17638) of Maheswor Gautam at the Graduate School of Science and Technology, Aarhus University (GSST, AU).

**Acknowledgments:** The GSST, AU is thanked for the funding for the PhD project. Bente Laursen, Kirsten Heinrichson and Elena Claudia Jensen from the Team Natural Products Chemistry and Environmental Chemistry at AU are thanked for their assistance in the laboratory. Agnes Corbin and Juliet Evans from Nonlinear Dynamics, a Waters company (Newcastle upon Tyng, United Kingdom), are thanked a lot for their assistance during the data processing in Progenesis QI for our experiments.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**


#### **References**


**Sample Availability:** Samples of the compounds are not available from the authors.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Metabolomics Approach on Non-Targeted Screening of 50 PPCPs in Lettuce and Maize**

**Weifeng Xue \*, Chunguang Yang, Mengyao Liu, Xiaomei Lin, Mei Wang and Xiaowen Wang**

Technical Center of Dalian Customs, Dalian 116000, China; 2004ycg51@163.com (C.Y.); laoyao1024@163.com (M.L.); lynn9857@163.com (X.L.); monkeywangcn@sina.com (M.W.); wxw0652@sina.com (X.W.)

**\*** Correspondence: xwf526@163.com

**Abstract:** The metabolomics approach has proved to be promising in achieving non-targeted screening for those unknown and unexpected (U&U) contaminants in foods, but data analysis is often the bottleneck of the approach. In this study, a novel metabolomics analytical method via seeking marker compounds in 50 pharmaceutical and personal care products (PPCPs) as U&U contaminants spiked into lettuce and maize matrices was developed, based on ultrahigh-performance liquid chromatography-tandem mass spectrometer (UHPLC-MS/MS) output results. Three concentration groups (20, 50 and 100 ng mL−<sup>1</sup> ) to simulate the control and experimental groups applied in the traditional metabolomics analysis were designed to discover marker compounds, for which multivariate and univariate analysis were adopted. In multivariate analysis, each concentration group showed obvious separation from other two groups in principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) plots, providing the possibility to discern marker compounds among groups. Parameters including S-plot, permutation test and variable importance in projection (VIP) in OPLS-DA were used for screening and identification of marker compounds, which further underwent pairwise *t*-test and fold change judgement for univariate analysis. The results indicate that marker compounds on behalf of 50 PPCPs were all discovered in two plant matrices, proving the excellent practicability of the metabolomics approach on non-targeted screening of various U&U PPCPs in plant-derived foods. The limits of detection (LODs) for 50 PPCPs were calculated to be 0.4~2.0 µg kg−<sup>1</sup> and 0.3~2.1 µg kg−<sup>1</sup> in lettuce and maize matrices, respectively.

**Keywords:** metabolomics; marker compounds; non-targeted screening; pharmaceutical and personal care products; plant-derived food

#### **1. Introduction**

Pharmaceutical and personal care product (PPCP) contamination in animal-derived foods has attracted worldwide attention, and a series of formal regulatory documents on the maximum residue limits (MRLs) of PPCPs from different countries and organizations has been issued [1–4]. However, PPCPs-induced contamination in plant-derived foods has not been fully addressed [5]. Previous studies [6–14] indicate that some plant-derived foods (e.g., corn, barley, pea, wheat, carrot, potato, cucumber and lettuce) can easily absorb PPCPs from soil with animal manure used as a fertilizer, which contains several kinds of commonly used antibiotics, e.g., tetracyclines, quinolones, sulfonamides and β-lactam, with their total concentration from the µg kg−<sup>1</sup> to the mg kg−<sup>1</sup> level in the plants [9,15–18]. Due to the lack of evaluation standards of PPCPs in plant-derived foods, it is hard to directly judge whether the residue concentrations of PPCPs can induce adverse effects on human health. Referring to the regulatory files on MRLs of PPCPs in animal-derived foods [2,4], which proposed a concentration of 10 µg kg−<sup>1</sup> as the threshold of safety for most PPCPs, it can be inferred that if the concentrations of PPCPs in plant-derived foods exceed 10 µg kg−<sup>1</sup> , it triggers a food safety risk. Therefore, the top priority is to develop reliable analytical methods for the investigation of PPCP residues in plant-derived foods.

**Citation:** Xue, W.; Yang, C.; Liu, M.; Lin, X.; Wang, M.; Wang, X. Metabolomics Approach on Non-Targeted Screening of 50 PPCPs in Lettuce and Maize. *Molecules* **2022**, *27*, 4711. https://doi.org/10.3390/ molecules27154711

Academic Editor: Thomas Letzel

Received: 23 June 2022 Accepted: 18 July 2022 Published: 23 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Previous studies [6,9,10,19,20] have proposed some analytical methods based on high-performance liquid chromatography-tandem mass spectrometer (HPLC-MS/MS) for PPCPs detection in plant matrices. These studies mainly focused on the contamination of antibiotics, especially for tetracyclines, quinolones and sulfonamides. Most methods are customized and showcase the excellent detection performance for specific PPCPs, but are invalid for detecting other PPCPs not in the customized database. With the rapid development of the modern chemical industry's ability to synthesize new compounds, there is reason to believe that more and more PPCPs will be produced and applied in animal husbandry; as a result, continuous uptake of PPCPs by plant-derived foods will probably lead to more complicated, serious and underlying food safety risk. The United States, China and Japan, as the world's top three economies, plus the European Union, have issued regulatory documents on MRLs of only 95, 128, 180 and 139 PPCPs in animal-derived foods, and some listed PPCPs are of repeated emergence [1–4]. The sticking point is that the existing technologies cannot meet the detection requirements for increasing unknown and unexpected (U&U) PPCPs, for which the most effective method is to develop non-targeted screening methods, as proposed by the NORMAN network (www.norman-network.net, accessed on 20 November 2021) founded in 2005 by the European Commission [21,22].

Non-targeted screening can be defined from the narrow and broad senses. The former is reliant on the established screening database to discern contaminants [23]. The contaminants in the database are known, but those existing in the matrix are obscure, thus the screening practicability depends on the database size. The latter sense is to employ omicsrelated approaches to complete U&U contaminant screening [24,25], which can be realized by high resolution mass spectrometry (HRMS) technology [26,27]. To date, LC-MS/MSbased metabolomics analytical methods have showed good practicability on non-targeted screening of some pesticides in food matrices, e.g., orange juice [28], milk [24] and tea [29], with the screening ratio of pesticides depending on their contents. These studies have obtained desirable outcomes, but the methods they proposed are so sophisticated that they are not favorable for wide application. Nowadays, the development of non-targeted metabolomics analysis still encounters many great obstacles, especially for data analysis, which is the bottleneck to be urgently solved through the advancement of data processing tools and improvement of HRMS data quality.

In view of this, we developed a novel metabolomics-based analytical method via seeking marker compounds on behalf of 50 PPCPs as U&U contaminants spiked in lettuce and maize matrices to achieve non-targeted screening. Ultrahigh-performance liquid chromatography-tandem mass spectrometer (UHPLC-MS/MS) was used to obtain output results for metabolomics analysis. Herein, 14 sulfonamides, 12 quinolones, 10 nitroimidazoles, 7 agonists, 4 steroids and 3 tetracyclines were selected as target PPCPs, in which quinolones, sulfonamides and tetracyclines are of relatively high detection frequency in plant-derived foods [9,15–18]. Lettuce and maize are consumed in high quantities worldwide, and have proved to easily absorb PPCPs from the soil [6,19]. Lack of formal documents to regulate the MRLs of PPCPs in plant-derived foods makes it difficult to directly evaluate whether the contents of PPCPs in the foods are in the safety range. According to the guidelines of GB 31650-2019 [4] and Commission Regulation (EU) No 37/2010 [2], the MRLs of most PPCPs in animal-derived foods are no lower than 10 µg kg−<sup>1</sup> , which was used as the test concentration of 50 PPCPs in our study to perform screening analysis. The goal of this study is to develop an applicable analytical method on the basis of metabolomics, which can accurately, rapidly and comprehensively achieve the screening and identification of potential non-targeted contaminants in plant-derived foods.

#### **2. Materials and Methods**

#### *2.1. Chemicals and Materials*

The lettuce was bought from a local market in Dalian City. Ethylenediamine tetraacetic acid disodium salt (Na2EDTA), citric acid, sodium hydrogen phosphate (Na2HPO4), anhydrous sodium sulfate (Na2SO4), sodium chloride (NaCl), sodium hydroxide (NaOH), hydrochloric acid (HCl) and C18 powder (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China); methanol and acetonitrile (HPLC grade, Merck, Darmstadt, Germany); formic acid (HPLC grade, Shanghai ANPEL Laboratory Technologies Inc., Shanghai, China); filter membrane (0.22 µm, Agilent Technologies, Singapore, MI, USA); ultrapure water (Milli-Q ultrapure water system, Merck, Darmstadt, Germany); ciprofloxacin-d8 hydrochloride solution (100 µg mL−<sup>1</sup> in methanol, First Standard, Ridgewood, NY, USA). Analytical standard compounds for 50 PPCPs (purity > 98.3%) were obtained from First Standard (Ridgewood, NY, USA), Sigma (Alexandria, VA, USA), TRC (Toronto, ON, Canada) and Dr. Ehrenstorfer (Augsburg, Germany). More details on the 50 PPCPs are shown in Table 1.


**Table 1.** Basic information on the 50 PPCPs.

#### *2.2. Solution Preparation*

A total of 50 PPCPs were separately prepared with methanol at 100 µg mL−<sup>1</sup> , 1 mL of which was withdrawn, mixed together and further diluted with methanol to obtain a 1 µg mL−<sup>1</sup> solution. Then, 100 ng mL−<sup>1</sup> ciprofloxacin-d8 methanol solution was prepared by diluting its 100 µg mL−<sup>1</sup> solution. A 0.1 mol L−<sup>1</sup> Na2EDTA-Mcllvaine buffer solution was prepared with Na2HPO<sup>4</sup> (5.5 g), citric acid (12.9 g) and Na2EDTA (37.2 g) dissolved in 1 L pure water, which was further adjusted to pH 4.0 with 0.1 mol L−<sup>1</sup> HCl or NaOH solution.

#### *2.3. Sample Preparation and Pretreatment Process*

(a) Lettuce sample was cut into small pieces, then ground into batter by tissue homogenizer; (b) 2.0, 5.0 and 10.0 g lettuce batters, together with one-to-one corresponding 20, 50 and 100 µL of 50 PPCPs mixed solutions (1 µg mL−<sup>1</sup> ) were poured into 50 mL polypropylene centrifuge tubes. To calibrate the recovery during the sample pretreatment process, ciprofloxacin-d8 methanol solution (0.5 mL, 100 ng mL−<sup>1</sup> ) as recovery internal standard was further added, as adopted in previous studies [30–32]; (c) 5 mL Na2EDTA-Mcllvaine buffer solution (0.1 mol L−<sup>1</sup> ) was dumped into the tube, vortexed for 1 min, then 20 mL 1% (*V*/*V*) formic acid/acetonitrile solution was added further, stirring for 1 min. An extraction salt package (10.0 g Na2SO<sup>4</sup> + 2.0 g NaCl) was added for stratification under salting out after the solution standing for 10 min, centrifuging at 4500 r min−<sup>1</sup> for 5 min; (d) then, after transferring all the supernatant into new 50 mL polypropylene centrifuge tubes, adding 100 mg C18 powder, vortexing for 1 min, centrifuging at 4500 r min−<sup>1</sup> for 3 min, the solution was extracted to another 50 mL centrifuge tube, dried with N<sup>2</sup> blowing by nitrogen blowing apparatus (N-EVAP-112, Organomation, Berlin, MA, USA), and redissolved in 1 mL 40% (*V*/*V*) methanol 0.1% formic acid/water solution, vortexed for 1 min; (e) then, filtered with a 0.22 µm filter membrane, the sample solutions of 50 PPCPs at the theoretical concentrations of 20, 50 and 100 ng mL−<sup>1</sup> were prepared. Each concentration experiment was repeated nine times.

#### *2.4. Sample Grouping and Naming*

Samples of 20 ng mL−1–1~20 ng mL−1–9, 50 ng mL−1–1~50 ng mL−1–9 and 100 ng mL−1–1~100 ng mL−1–9 were employed to label samples from three concentration groups. Each sample provided a 30 µL solution as a quality control (QC) sample [29,33,34], which experienced 3 injections before and after each concentration group. As a result, 12 samples marked as QC-1, QC-2, and QC-12 were obtained to evaluate the stability of LC-MS/MS.

#### *2.5. Analytical Method*

The 50 PPCPs and ciprofloxacin-d8 were analyzed on a quadrupole/electrostatic field orbitrap LC-MS/MS system (Q Exactive Plus, Thermo Fisher Scientific Inc., Waltham, MA, USA) under the positive mode of electrospray ion (ESI) source. Components in the sample solution underwent separation within an Accucore RP-MS column (100 × 2.1 mm, 2.6 µm particle diameter, Thermo Fisher Scientific Inc., Waltham, MA, USA), with injection volume of 10 µL. Next, 0.1% (*V*/*V*) formic acid/water and 0.1% (*V*/*V*) formic acid/methanol solutions were prepared as the mobile phase A and B, respectively, with flow rate of 0.3 mL min−<sup>1</sup> . In consideration of the matrix complexity of lettuce and maize, there may be some impurities not eluted from the LC-MS/MS system in a relatively short time (738 s for the last eluted target PPCP in this study) designed only for 50 PPCPs, leading to the potential disruption for the elution and analysis of the next sample. Therefore, a longer elution program was designed as follows: gradient started from 5% B, kept for 2 min, then increased to 30% B in 1 min, at a duration of 7 min, further increased to 90% B in 1 min, holding on 25 min, finally decreased to 5% B in 1 min, equilibrating for 16 min. The oven temperature was set at 40 ◦C. Other parameter settings were as follows: heating and capillary temperature 320 ◦C; lens and spray voltage 50 and 3200 V, respectively; auxiliary and sheath gas N2, with flow rate at 10 and 40 arb, respectively; scan mode: full-scan/datadependent two-stage scanning; MS parameters: full-scan resolution 70,000, maximum dwell time 100 ms, AGC target 1 <sup>×</sup> <sup>10</sup><sup>6</sup> , *m*/*z* scan range 100~1000; MS/MS parameters: resolution 17,500, maximum dwell time 50 ms, AGC target 2 <sup>×</sup> <sup>10</sup><sup>5</sup> .

LC-MS/MS output results of 50 PPCPs and ciprofloxacin-d8 were analyzed by Trace Finder 3.3 software, with screening conditions as follows: (a) for primary parent ion, signal to noise ratio 5.0, response intensity threshold 10,000, and mass error 5 ppm; (b) for secondary fragment ions, minimum matching number of ion 1, response intensity threshold 10,000, and mass error 5 ppm. On the basis of the peak area of the primary parent ion, ciprofloxacin-d8 was quantified with standard curve for recovery calculation.

#### *2.6. Metabolomics Data Processing*

LC-MS/MS was operated in full scan mode with RAW-formatted files as the direct output, which underwent conversion to corresponding mzXML-formatted files via the ProteoWizard software [35]. These new files are adaptable to the upload to the Workflow4Metabolomics (W4M) platform (https://workflow4metabolomics.usegalaxy.fr/, accessed on 20 November 2021) for metabolomics analysis [36]. After peak detection, alignment and retention time calibration, plus data normalization, centralization, scaling and transformation performed on the W4M platform, the data matrix was obtained in the format of variable and sample named as abscissa and ordinate, respectively [36,37]. Variable contains a series of information, e.g., molecular weight and retention time, with every marker compound corresponding to its unique variable, that is to say, the process to pursue marker compounds is actually a process to pursue eligible variables. Multivariate statistical analysis including principal component analysis (PCA) [38–40] and orthogonal partial least squares discriminant analysis (OPLS-DA) [41,42] was performed in SIMCA 14.1 software [43] after importing the data matrix. A permutation test with 200 iterations was employed for over-fitting judgement of the OPLS-DA model [43,44]. Other parameters to screen marker compound candidates include the absolute value of variable confidence in the S-plot plot [45] and variable importance in projection (VIP) [43,44,46], with the threshold above 0.9 and 1, respectively. After this, eligible marker compound candidates from 20 and 100 ng mL−<sup>1</sup> groups can both be obtained, and only overlapped candidates in two groups, representing their significantly low and high concentration in the corresponding 20 and 100 ng mL−<sup>1</sup> groups, were further investigated by pairwise *t*-test [47–49] in SPSS Statistics V17.0 software and fold change judgement for the univariate analysis. Univariate analysis is simple, intuitive and easy to be understood. It was used to quickly investigate the differences of marker compound candidates in different groups. To more rapidly verify the identity of marker compounds on behalf of 50 PPCPs, we directly compared the precise molecular weight (<5 ppm in absolute value of error), retention time and the adduct structure of marker compounds with that of the authentic 50 PPCPs (Table 1).

#### **3. Results and Discussion**

#### *3.1. Data Preprocessing*

As indicated in Figure 1, although only part of the total ion chromatograms at the retention time of 0~900 s is shown, during which all 50 PPCPs were eluted, obvious differences in peak intensity have already been observed in three concentration groups, implying the possibility to seek marker compounds among groups. The principle for relative standard deviation of peak intensity above 30% was employed to filter out invalid variables in QC and three concentration groups [50], with a final 6512 × 39 data matrix obtained for further analysis.

#### *3.2. Multivariate Analysis*

#### 3.2.1. PCA Analysis

As Taguchi [51] pointed out, PCA can make a natural classification for sample groups and eliminate the extreme data without knowing their categories, thus PCA can be used in metabolomics to assess the data quality and to identify outliers [38–40]. As indicated in

Figure 2, no extreme data and outliers were observed. Samples at the same concentration gathered together, indicating the good classification of groups. Obvious separation among three concentration groups indicates the existence of major discrepancies, further paving the way to seek marker compounds from different groups. ferences in peak intensity have already been observed in three concentration groups, implying the possibility to seek marker compounds among groups. The principle for relative standard deviation of peak intensity above 30% was employed to filter out invalid variables in QC and three concentration groups [50], with a final 6512 × 39 data matrix obtained for further analysis.

As indicated in Figure 1, although only part of the total ion chromatograms at the retention time of 0 ~ 900 s is shown, during which all 50 PPCPs were eluted, obvious dif-

corresponding 20 and 100 ng mL−1 groups, were further investigated by pairwise t-test [47–49] in SPSS Statistics V17.0 software and fold change judgement for the univariate analysis. Univariate analysis is simple, intuitive and easy to be understood. It was used to quickly investigate the differences of marker compound candidates in different groups. To more rapidly verify the identity of marker compounds on behalf of 50 PPCPs, we directly compared the precise molecular weight (<5 ppm in absolute value of error), retention time and the adduct structure of marker compounds with that of the authentic 50

*Molecules* **2022**, *27*, x FOR PEER REVIEW 6 of 15

PPCPs (Table 1).

**3. Results and Discussion** *3.1. Data Preprocessing*

Retention Time (s) **Figure 1.** Total ion chromatograms (0~900 s) of spiked lettuce sample groups on the W4M platform.

**Figure 2.** PCA score plot of spiked lettuce sample groups. **Figure 2.** PCA score plot of spiked lettuce sample groups.

#### 3.2.2. OPLS-DA Analysis

3.2.2. OPLS-DA Analysis Theoretically speaking, the peak intensities of variables ought to increase with their rising concentrations, i.e., 20 and 100 ng mL−1 groups should present the minimum and maximum peak intensities, respectively. However, the reality may be different, due to the discrepancies in sample recoveries. Previous studies [30–32] proposed deuterated antibiotics as recovery internal standards to correct losses of PPCPs during sample preparation. In consideration of this, ciprofloxacin-d8 (parent ion *m/z* 340.19132; fragment ions *m/z* 296.20156, 253.15933 and 239.14367; retention time 6.73 min) was employed here to eliminate the peak intensity errors of variables induced by disparate recoveries of PPCPs during the pretreatment process. As shown in Table S1 (Supplementary Materials), the recoveries of ciprofloxacin-d8 were calculated to be 80.1 ~ 85.9%, 80.3 ~ 86.2% and 81.6 ~ 87.7% Theoretically speaking, the peak intensities of variables ought to increase with their rising concentrations, i.e., 20 and 100 ng mL−<sup>1</sup> groups should present the minimum and maximum peak intensities, respectively. However, the reality may be different, due to the discrepancies in sample recoveries. Previous studies [30–32] proposed deuterated antibiotics as recovery internal standards to correct losses of PPCPs during sample preparation. In consideration of this, ciprofloxacin-d8 (parent ion *m*/*z* 340.19132; fragment ions *m*/*z* 296.20156, 253.15933 and 239.14367; retention time 6.73 min) was employed here to eliminate the peak intensity errors of variables induced by disparate recoveries of PPCPs during the pretreatment process. As shown in Table S1 (Supplementary Materials), the recoveries of ciprofloxacin-d8 were calculated to be 80.1~85.9%, 80.3~86.2% and 81.6~87.7% in the 20, 50 and 100 ng mL−<sup>1</sup> groups, respectively, based on the ciprofloxacin-d8 standard curve solutions (100, 50, 25, 10 and 5 ng mL−<sup>1</sup> ) prepared in blank lettuce extract solution. After

in the 20, 50 and 100 ng mL−1 groups, respectively, based on the ciprofloxacin-d8 standard curve solutions (100, 50, 25, 10 and 5 ng mL−1) prepared in blank lettuce extract solution. After this, the recoveries of ciprofloxacin-d8 were all calibrated to 100% by multiplying a

As shown in Figure 3, we can observe the separation of two camps on the first principal component axis. One camp represents the specific concentration group (green part), and the other camp is on behalf of the remaining two groups (blue part), indicating the existence of variables with significant differences between the two camps. Each point in the S-plot plots (Figure 4) represents a variable, which keeps away from the origin along *X*- and *Y*-axis, implying more contribution and higher confidence level of the variable to the difference. Therefore, the points at the two ends of 'S' can be deemed the most differentiating components. In the S-plot analysis, absolute value of confidence > 0.9 has been proposed to screen variables as marker compound candidates [45], which at the significantly low and high concentration should be searched at the right and left ends of S-plot

plots in Figure 4a,b, respectively.

were also calibrated, together with peak intensities for all the variables.

this, the recoveries of ciprofloxacin-d8 were all calibrated to 100% by multiplying a corresponding calibration coefficient, with which the peak intensities of ciprofloxacin-d8 were also calibrated, together with peak intensities for all the variables.

As shown in Figure 3, we can observe the separation of two camps on the first principal component axis. One camp represents the specific concentration group (green part), and the other camp is on behalf of the remaining two groups (blue part), indicating the existence of variables with significant differences between the two camps. Each point in the S-plot plots (Figure 4) represents a variable, which keeps away from the origin along *X*- and *Y*-axis, implying more contribution and higher confidence level of the variable to the difference. Therefore, the points at the two ends of 'S' can be deemed the most differentiating components. In the S-plot analysis, absolute value of confidence > 0.9 has been proposed to screen variables as marker compound candidates [45], which at the significantly low and high concentration should be searched at the right and left ends of S-plot plots in Figure 4a,b, respectively. *Molecules* **2022**, *27*, x FOR PEER REVIEW 8 of 15 *Molecules* **2022**, *27*, x FOR PEER REVIEW 8 of 15

**Figure 3.** OPLS-DA score plots of spiked lettuce sample groups. **Figure 3.** OPLS-DA score plots of spiked lettuce sample groups. **Figure 3.** OPLS-DA score plots of spiked lettuce sample groups.

SIMCA 14.1 software performed permutation tests with 200 iterations to investigate whether the OPLS-DA models underwent data over-fitting, for which R2Y and Q<sup>2</sup> are two SIMCA 14.1 software performed permutation tests with 200 iterations to investigate SIMCA 14.1 software performed permutation tests with 200 iterations to investigate whether the OPLS-DA models underwent data over-fitting, for which R2Y and Q<sup>2</sup> are two

from Figure 5, R2Y and Q<sup>2</sup> values were no less than 0.991, indicating the good reliability, predictability and no over-fitting for all OPLS-DA models. VIP > 1 principle continues to screen marker compounds. Eventually, marker compounds on behalf of 50 PPCPs were all screened out as shown in Table 2. Negligible concentrations (<0.1 ng mL−1) of 50 PPCPs in the blank lettuce extract solution were obtained by the metabolomics analysis, which eliminates the interference of inherent (rather than spiked) 50 PPCPs residues in lettuce

(or equal) to 1, the OPLS-DA models are not susceptible to over-fitting. As can be seen from Figure 5, R2Y and Q<sup>2</sup> values were no less than 0.991, indicating the good reliability, predictability and no over-fitting for all OPLS-DA models. VIP > 1 principle continues to screen marker compounds. Eventually, marker compounds on behalf of 50 PPCPs were all screened out as shown in Table 2. Negligible concentrations (<0.1 ng mL−1) of 50 PPCPs in the blank lettuce extract solution were obtained by the metabolomics analysis, which eliminates the interference of inherent (rather than spiked) 50 PPCPs residues in lettuce

matrix to seek marker compounds.

matrix to seek marker compounds.

common parameters to describe the interpretation level of the model in the *Y*-axis direc-

whether the OPLS-DA models underwent data over-fitting, for which R2Y and Q<sup>2</sup> are two common parameters to describe the interpretation level of the model in the *Y*-axis direc-

**Var ID (Primary)**

**Marker Compounds**

M162T266 5-Chloro-1-methyl-4-

**VIP Pred** *<sup>a</sup>*

M226T148 Terbutaline 2.297/2.467 (−0.049, −0.969)/(0.049, 0.945) −1.751 2.0 M228T491 Tolobuterol 4.045/3.311 (−0.098, −0.922)/(0.063, 0.929) −0.294 0.6 M234T261 Cimbuterol 3.226/4.071 (−0.075, −0.934)/(0.082, 0.965) −2.061 1.1 M260T571 Propranolol 3.710/4.505 (−0.089, −0.927)/(0.099, 0.945) 0.617 0.5 M273T138 Sotalol 1.513/1.453 (−0.025, −0.923)/(0.025, 0.916) 0.563 0.7 M310T388 Nadolol 4.165/2.141 (−0.101, −0.981)/(0.038, 0.933) −2.351 0.7

M170T467 Ipronidazole 4.537/4.102 (−0.111, −0.921)/(0.087, 0.927) 2.920 0.4 M172T152 Metronidazole 2.838/2.675 (−0.064, −0.928)/(0.052, 0.967) 2.916 1.7

nitroimidazole 1.854/1.675 (−0.038, −0.971)/(0.032, 0.948) <sup>−</sup>0.394 0.6

common parameters to describe the interpretation level of the model in the *Y*-axis direction and the prediction level of the model [52,53], respectively. If R2Y and Q<sup>2</sup> are both close (or equal) to 1, the OPLS-DA models are not susceptible to over-fitting. As can be seen from Figure 5, R2Y and Q<sup>2</sup> values were no less than 0.991, indicating the good reliability, predictability and no over-fitting for all OPLS-DA models. VIP > 1 principle continues to screen marker compounds. Eventually, marker compounds on behalf of 50 PPCPs were all screened out as shown in Table 2. Negligible concentrations (<0.1 ng mL−<sup>1</sup> ) of 50 PPCPs in the blank lettuce extract solution were obtained by the metabolomics analysis, which eliminates the interference of inherent (rather than spiked) 50 PPCPs residues in lettuce matrix to seek marker compounds. *Molecules* **2022**, *27*, x FOR PEER REVIEW 9 of 15

**Figure 5.** Permutation test plots of spiked lettuce sample groups. **Figure 5.** Permutation test plots of spiked lettuce sample groups.

*3.3. Univariate Analysis* **Table 2.** Marker compounds screened in lettuce sample groups.


**Coordinate** 

**Mass Error (ppm)** *<sup>c</sup>*

**LOD (µg kg−1)**


**Table 2.** *Cont.*

Note: *<sup>a</sup>* two VIP values from 100 and 20 ng mL−<sup>1</sup> groups, respectively; *<sup>b</sup>* two-group coordinate values from 100 and 20 ng mL−<sup>1</sup> groups, respectively; *<sup>c</sup>* Mass error (ppm) = (extracted molecular weight from W4M platform extracted molecular weight from LC-MS/MS) <sup>×</sup> <sup>10</sup>6/extracted molecular weight from LC-MS/MS.

#### *3.3. Univariate Analysis*

After multivariate analysis, a pairwise *t*-test [47–49] was firstly employed to examine whether marker compounds from a specific concentration group presented significant differences in peak intensity with those from other two groups. Pairwise *t*-test, as a reliable statistical test method, was performed to calculate *p* values between the two concentration groups and the *p* < 0.05 observed in this study indeed showed the existence of significant differences among groups. Previous studies [29,54] also adopted fold change of concentration > 2 to discern variables with high contrast among groups as marker compounds. Herein, marker compounds on behalf of 50 PPCPs all presented fold change values above 2, supporting the validity of marker compounds obtained with our analytical strategy.

The limits of detection (LODs) for 50 PPCPs were also considered here. Firstly, a 2.0 g blank lettuce sample was used to prepare an extract solution (1 mL) after the same pretreatment mentioned above. Then, a 20 ng mL−<sup>1</sup> PPCPs solution was obtained by diluting their mixed methanol solution (20 µL, 1 µg mL−<sup>1</sup> ) with 1 mL blank lettuce extract solution. The experiments were repeated in septuplicate to obtain seven samples, which underwent the same metabolomics analysis to obtain the peak intensities of 50 PPCPs. For each PPCP, a 20 ng mL−<sup>1</sup> concentration level was deemed to correspond to average values of seven samples in peak intensity; therefore, the concentration (unit: ng mL−<sup>1</sup> ) of each PPCP in a sample was calculated by its own peak intensity × 20/average peak intensity for the standard deviation measurement of the seven samples. According to the method proposed by US Environmental Protection Agency [55], the LOD values for 50 PPCPs were calculated to be 0.4~2.0 µg kg−<sup>1</sup> , as shown in Table 2.

#### *3.4. Method Applicability in Maize Matrix*

Maize as the primary food crop in China has proved to easily absorb PPCPs from the soil [19]; therefore, it was selected as another plant matrix different from vegetables to investigate the applicability of the developed metabolomics-based screening method. Maize sample was purchased from the local market and turned into a powder by a grinder. Then, it underwent the same above-mentioned pretreatment process after 50 PPCPs spiked at 10 µg kg−<sup>1</sup> as well. Ciprofloxacin-d8 methanol solution (0.5 mL, 100 ng mL−<sup>1</sup> ) was added for recovery calibration, with the results shown in Table S2. The same metabolomics analysis was performed as indicated in Figures S1–S5 (Supplementary Materials). Marker compounds to represent 50 PPCPs were also discovered (Table S3), proving the good applicability of the metabolomics analytical method to non-targeted screening of various PPCPs residues in different plant matrices. As can be seen from Table S3, the LOD values for 50 PPCPs in maize matrix were calculated to be 0.3~2.1 µg kg−<sup>1</sup> .

#### *3.5. Real Sample Test*

We collected lettuce and maize samples from six administrative districts including Zhongshan, Xigang, Shahekou, Gaoxin, Ganjingzi and Jinpu affiliated to Dalian City, each district with two sampling points. A total of 12 fresh lettuce samples were purchased from the local farmer's market and immediately delivered to the laboratory for testing. The above process was also applied to the maize samples. After pretreatment experiments and metabolomics analysis, only one lettuce sample from Jinpu District was found to contain enrofloxacin and its content was 17.4 µg kg−<sup>1</sup> . Other samples had no detection of PPCPs. Although the detection rate of PPCPs in all the samples is only 1/24, and seemingly only one district is vulnerable to PPCPs contamination, the results are enough to show that our proposed method is competent for the screening of PPCPs in plant-derived foods. These spot check results alert us to the fact that PPCP-induced safety risk of plant-derived foods is on the horizon.

Previous studies have successfully applied non-targeted screening methods on the basis of metabolomics to pesticide residues in plant matrices, e.g., orange juice [28] and tea [29], providing the feasibility to screen PPCPs residues in plant-derived foods. In light of the otherness of analytes, the reported methods may not be completely applied to our study. Herein, we firstly considered spiked contaminants to be marker compounds and then implemented a marker compound-seeking analytical strategy of metabolomics to finish the non-targeted screening of contaminants in plant-derived foods, which is the biggest difference from previous studies [24,28,29]. Despite only 50 PPCPs and two plant matrices considered here, the developed method still has wide applicability due to the representation of these PPCPs and universal consumption of lettuce and maize.

Extensive use of PPCPs in livestock farming raises the risk that these compounds end up in soil where animal waste is used as fertilizer [9,56], which leads to the uptake of PPCPs by plant-derived foods from the soil [57–64]. Compared with other plants, leafy vegetables generally show higher detection ratio and concentrations of PPCPs [60,64] and therefore deserve more attention in their food safety risk. Although there are no official documents to explicitly clarify the MRLs of PPCPs in plant-derived foods, we can still deduce their safety thresholds from their corresponding MRLs in animal-derived foods [1–4]. Relative to the colossal number of analytical methods for PPCPs in animal-derived foods [65–69], the methods for PPCPs detection in plant matrices are in short supply. To better cope with the complicated PPCPs contamination in plants, the top priority is to develop a high-throughput screening method that can accurately, rapidly and comprehensively determine which PPCPs exist in the foods. With this consideration, we developed this novel metabolomics-based analytical method to achieve non-targeted screening of PPCPs in plant-derived foods.

#### **4. Conclusions**

The newly developed metabolomics analytical method was successfully applicable to non-targeted screening of 50 PPCPs residues in lettuce and maize matrices. We intentionally designed three concentration groups of PPCPs (20, 50 and 100 ng mL−<sup>1</sup> ) to simulate the experimental and control groups adopted in the traditional metabolomics analytical procedures to search for marker compounds on behalf of 50 PPCPs. The process to perform metabolomics analysis has less artificial interference, a more concise workflow and higher screening efficiency. It is worth mentioning that this is the first implemented analytical strategy of metabolomics for non-targeted screening of PPCPs in plant-derived foods through seeking marker compounds. Due to the lack of binding legal documents on MRLs of PPCPs in plant matrices, together with constant development and application of new PPCPs in animal husbandry, it is urgent to compile legal rules to control MRLs of PPCPs in plant-derived foods, otherwise it may evolve as a serious food safety issue. To date, plant uptake from PPCP-contaminated soil is a known source of PPCP residues in plant-derived foods. It is not yet clear whether other ways can also induce the accumulation of PPCPs in the foods, potentially increasing the complexity of PPCPs contamination. Even worse, this increases the exposure risk of PPCPs to human health via the food chain. Therefore, we advocate that early attention to this issue would help defuse the potential crisis.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/molecules27154711/s1, Figure S1: Total ion chromatograms (0~900 s) of spiked maize sample groups on the W4M platform; Figure S2: PCA score plot of spiked maize sample groups; Figure S3: OPLS-DA score plots of spiked maize sample groups; Figure S4: S-plot plots of spiked maize sample groups; Figure S5: Permutation test plots of spiked maize sample groups; Table S1: Recovery (%) of spiked ciprofloxacin-d8 in lettuce sample groups (*n* = 9); Table S2: Recovery (%) of spiked ciprofloxacin-d8 in maize sample groups (*n* = 9); Table S3: Marker compounds screened in maize sample groups.

**Author Contributions:** Conceptualization, W.X.; Data curation, W.X.; Investigation, C.Y., M.L., X.L., M.W. and X.W.; Methodology, W.X.; Writing—original draft, W.X.; Writing—review and editing, W.X. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was funded by the National Natural Science Foundation of China (No. 21777014), Natural Science Foundation of Liaoning Province of China (No. 2019-BS-008), Special Foundation for Basic Research Program of Dalian Customs (No. 2021DK11).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available in this article and Supplementary Materials.

**Conflicts of Interest:** The authors declare no conflict of interest.

**Sample Availability:** Samples of the compounds are available from the authors.

#### **References**

