Liquid chromatography-tandem mass spectrometry analysis for identification and quantification of antimicrobial compounds in distillery wastewater

Highlights • Analyze and identify 4 levels of small molecule compounds in distillery wastewater.• Simple method for quantification of five antimicrobial compounds.• Column temperature affected the lactic and succinic acid chromatographs significantly.


Introduction
Distillery wastewater could cause many environment issues due to its high generation amount and high concentration of organics and nutrients [1] . Therefore, it is important to develop methods to analyze the composition of distillery wastewater to support the improvement of resource recovery and treatment process of distillery wastewater. In this study, a high-resolution mass spectrometry (HR-MS) method was developed to analyze and identify small molecules compounds in distillery wastewater and 4 levels of compounds were identified. And an effective and rapid method has been developed for simultaneous determination of lactic acid, succinic acid, acetophenone, cinnamic acid and phenyllactic acid (the five identified major antimicrobial compounds) in the distillery wastewater using a simple one-step sample dilution preparation couple with UPLC-MS/MS.

Materials and reagents
Lactic acid, succinic acid, acetophenone, cinnamic acid and phenyllactic acid were purchased from the Sigma-Aldrich Company Ltd.
HPLC-grade formic acid and MS-grade methanol purchased from Merck (Darmstadt, Germany) were used for HPLC analysis and sample preparation.

Preparation of standard solution and distillery wastewater samples
Concentrated stock solutions of analytes were prepared by dissolving the appropriate amount of the standard samples in 50% methanol at a concentration of 1 mg/mL. And then it was further diluted with acetonitrile to form a series of working solutions used to prepare the calibration curve. All the solutions were stored at -20 °C.
A10 μl of the distillery wastewater sample was added with a 20 mL of 50% methanol solution was added. Then, the mixture was vortexed for 2 min and centrifugation at 13,0 0 0 rpm for 10 min at 4 °C. Subsequently, the supernatant liquor was transferred to centrifugation at 13,0 0 0 rpm for 5 min at 4 °C again, then the supernatant liquor was injected into the HPLC-MS/MS for analysis.

Analytical instrumentation
The LC-MS/MS system used was a Thermo Scientific Ultimate 30 0 0 liquid phase system equipped with Q Exactive Orbitrap and an electrospray ionization source. A volume of 2 μl sample was injected to a Hypersil Gold C18 column (100 × 2.1 mm, 1.9 μm, Thermo Scientific) at 20 °C. The LC flow was set to 250 μl/min using H 2 O (0.1% formic acid) and methanol as eluents. The gradient elution started with 98% H 2 O for 2 min and was changed to 95% methanol over the course of 13 min, maintained for 3 min, then returned to 98% H 2 O within 0.1 min, and equilibrated for 1.9 min prior to the next injection. The heated electrospray ionization source had a capillary temperature of 350 °C. Both positive and negative electrospray ionization were employed to obtain MS signals of analytes with spray voltages of + 3.5 kV and -2.5 kV, respectively. Sheath gas flow rate, aux gas flow rate and sweep gas flow rate were set to 40, 10 and 0 (arbitrary units), respectively. Capillary temperature and aux gas heater temperature were set to 320 °C and 350 °C, respectively. The MS was set at full scan mode and acquire targeted first MS signals in at 70,0 0 0 fwhm and targeted MS/MS scan was set at a resolution of 175,00 fwhm with isolation width of 2.0 m/z. The instrument would automatically switch the positive and negative ion scanning mode and the scan mode was chosen as full MS scandd MS2 and acquire first MS signals at 70,0 0 0 fwhm and targeted MS/MS scan was set at a resolution of 175,00 fwhm with isolation width of 2.0 m/z. Meanwhile, the m/z scan range was 70-700.

Data processing
Peak detection and alignment of the LC −MS data were performed using Compound Discoverer 2.0 (Thermo Scientific) to obtain a peak list with peak areas, molecular weight, and retention time with the following settings: S/N threshold, 3; mass tolerance, 10 ppm; minimum peak intensity, 1 × 10 5 . With the application of the software, a possible molecular formula fitting the exact mass and isotope patterns was calculated. Furthermore, the MS/MS fragments were compared to the mzCloud database. Fig. 1 and S1-S4 (in the supplementary materials) show how compounds were identified. As can be seen, the MS and, MS/MS information and retention time of the unknown compound were highly consistent with the reference substance.
According to Identification confidence levels reported by Schymanski et al. [2] , 4 levels of unknown compound were classified in Table 2 .   Table 3 showed the compounds contained in rice spirit distillery wastewater identified with four different confidence levels by HR-MS. Lactic acid, succinic acid, L-phenylalanine, caffeine, adenosine, D( + )-phenyllactic acid, DL-arginine, acetophenone and cinnamic acid were confirmed using the standard compounds. The MS, MS/MS and retention time compared with reference standards (lactic acid, succinic acid, acetophenone, cinnamic acid and phenyllactic acid) were shown in Fig. 1 and S1-4. Approximate 60 compounds were converged to level 2 in the identified top 100 most abundant compounds (based on peak area). Their MS/MS fragments were compared to the mzCloud database and had a direct matching. In Fig. S5, ϒ-aminobutyric acid, L-glutamic acid, proline and D-( + )pyroglutamic acid were chosen as representatives to show the MS2 spectrum comparison between the sample and mzCloud library. Fig. S6 was the chromatogram and ms2 spectrum of extract mass 132.1019, indicated the existence of leucine or isoleucine. In level 4, a possible molecular formula fitting the exact mass and isotope patterns was calculated.

Quantification of antimicrobial compounds by LC-MS-MS
Among the compounds detected by LC-MS-MS, five of them are reported with antimicrobial activity and had relatively high concentrations in distillery wastewater, which may affect the resource recovery process for distillery wastewater via microorganisms. They are lactic acid [3] , succinic acid [4] , cinnamic acid [5] , phenyllactic acid [6] , acetophenone [7] . Therefore, an effective and rapid quantification method has been developed for these compounds in this study.

Analytical instrumentation
The LC-MS/MS system consisted of a Thermo Scientific Ultimate 30 0 0 liquid phase system and TSQ Endura triple quadrupole mass spectrometer with an electrospray ionization source. Chromatographic separation was achieved at 20 °C on a Hypersil Gold C18 column (100 × 2.1 mm, 1.9 μm, Thermo Scientific) by gradient solution with 0-2 min, 98% mobile phase A;2-4 min, 98% → 80% mobile phase A; 4-7 min, 80% → 10% mobile phase A; 7-9 min, 10% mobile phase A;9.1-12 min, 98% mobile phase A, flowing at 0.25 mL/min. Eluent A was water containing 0.1% formic acid, and B was methanol. The injection volume was 2 μL. Note: * qualitative ion. To achieve better retention and separation of both hydrophilic and polar compounds, two chromatographic columns with different stationary phases (i.e. a HILIC column and a C18 column) were examined with various mobile phases and additives (i.e. formic acid, acetic acid and ammonium acetate). Additionally, gradients, flow rate and column temperatures (20-40 °C) were also explored. It was found that the chromatographs of lactic acid and succinic acid were significantly affected by the column temperatures. Based on the chromatograph of lactic acid and succinic acid under 20 °C and 30 °C (Fig. S7), 20 °C was selected as the column temperature to obtain a good peak shape.
The addition of ammonium acetate into formic acid water or acetic acid water as mobile phase significantly decreased peak responses while did not improve peak shapes simultaneously. Compared with acetic acid in water, formic acid in water as the mobile phase could narrow peak widths. Therefore, 0.1% formic acid in water was selected as one of the mobile phases. Though the two columns had similar performance in resolution, retention time and peak shape, Hypersil Gold C18 as chromatographic separation column was chosen rather than Syncronis Hilic column (for polar components) because the former one was more commonly used.
The mass spectrometer was operated in negative ion mode using SRM to detect the mass transitions. High purity nitrogen served as both nebulizing and drying gas. Compound-dependent parameters of the mass spectrometer were set as follows: spray voltage at 2500 V, capillary temperature at 320 °C, vaporizer temperature at 350 °C, sheath gas at 35 (Arb) and auxiliary gas at 10 (Arb). The parameters of SRM scan mode for each compound are shown in Table 4 . Fig. 2 demonstrated typical chromatograms of the five analytes.

Validation of the method
The developed method was validated based on the recommendations published by FDA (Food and Drug Administration) [8] . The calibration curve consisted of five concentration levels. The linear regression of the areas of the analyte peaks versus the concentration were weighted with weighing factor 1/x 2 (where x = concentration). The concentrations of the analyte were determined by interpolation from the calibration curve. Concentration of the standard sample in solvents with a signal-to-noise ratio (S/N) of 3 times is defined as instrumental detection limit. As shown in Table 5 , all the analytes showed good linearity with regression coefficients (R 2 ) values above 0.9981 (R > 0.9990). Linear ranges and IDL of the analytes were also shown in Table 5 . The calibration curves of the five analytes were shown in Fig. S8.
Three levels (low, medium and high) of organic acids were added to distillery wastewater samples to determine the precision (relative standard deviation, RSD) and extraction recovery (relative error,   Fig. 3 and Table S1. All the recoveries were between 95.89% and 116.39% (RSD% < 9.80) at the three concentration levels of the analytes. These results were with the acceptance criteria and indicated that the method was accurate, reliable, and reproducible. Meanwhile, the wastewater samples were pretreated simply through dilution and centrifugation. These results of recoveries indicate that there was no significant matrix effect.

Application
The established LC-MS/MS method was applied for determining the concentration of the five major antimicrobial compounds in distillery wastewater obtained from the rice spirit distillery located in Foshan city, Guangdong, Southern China. Table 6 was the quantitative analysis results of the five analytes in distillery wastewater.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.