Determination of 5-Hydroxymethylfurfural, N (cid:0) -(Carboxymethyl)Lysine and N (cid:0) -(Carboxyethyl)Lysine in 16 Traditional Chinese Medicine Injections Based on HPLC and UPLC-MS/MS

Background: Traditional Chinese medicine injections (TCMIs) are widely applied to treat many chronic diseases. However, product quality problems occur occasionally due to unknown constituents in TCMIs. 5-hydroxymethylfurfural (5-HMF), N (cid:0) -(carboxymethyl)lysine (CML) and N (cid:0) -(carboxyethyl)lysine (CEL) are three compounds generated during food and Chinese medicinal herb processing and may be harmful to human health. Methods: In this study, the contents of 5-HMF, CML and CEL were determined by high-performance liquid chromatography (HPLC) and ultra-performance liquid chromatography (UPLC). For 5-HMF, the separation was performed on a Hypersil ODS2 column (250 mm×4.6 mm, 5 µm), and the column temperature was set at 30 ℃ . The mobile phase was composed of water-methanol (95:5) at a ow rate of 1.0 mL/min. For CML and CEL, separation was performed on a CORTECS HILIC UPLC column (2.1 mm×50 mm, 1.6 µm), and the column temperature was set at 40 ℃ . The mobile phase was composed of acetonitrile-water (3:7) at a ow rate of 0.3 mL/min. Multiple-reaction monitoring mode was employed for analyte determination with positive ionization. Results: The contents of 5-HMF in 16 TCMIs varied from 0.19 to 74.98 µg/mL, with a larger variation than the contents of CML and CEL. The Ciwujia injection had the highest content of 5-HMF, and the Qingkailing injection had the lowest 5-HMF content. The contents of CML and CEL among these TCMIs were 0.51-7.32 ng·mL -1 and 0.38-5.49 ng·mL -1 , respectively. The contents of CML and CEL in the Shuxuetong injection were much higher than in the others. Conclusions: The methods established in this study were simple, rapid and accurate and could provide a theoretical basis for the quality evaluation of TCMIs. traditional Chinese medicine injections; ADRs: adverse drug reactions; 5-HMF: 5-Hydroxymethylfurfural; AGEs: Advanced glycation end-products; CML: N (cid:0) -(carboxymethyl)lysine; CEL: N (cid:0) -(carboxyethyl)lysine; MGO: methylglyoxal; GO: glyoxal; HPLC: high-performance liquid chromatography; RP-HPLC: reversed-phase high-performance liquid chromatography; LC-MS: liquid chromatography-mass spectrometry; HPLC-MS/MS: high-performance liquid chromatography coupled to tandem mass spectrometry; UPLC-MS: ultra-performance liquid chromatography-mass spectrometry; ELISA: enzyme linked immunosorbent assays; ESI: electrospray ionization; MRM: multiple reaction monitoring; LODs: limits of detection; LOQs: limits quantication.


Background
Traditional Chinese medicine (TCM) is a diagnostic system of ancient medical practice that has evolved over thousands of years to help prevent and cure diseases [1]. As the most effective and popular administration form of TCM [2], traditional Chinese medicine injections (TCMIs) are widely applied to treat acute upper respiratory tract infections [3], angina pectoris [4], heart failure [5] and other chronic diseases. Owing to the uncertainty of the active constituents and the complexity of TCM prescriptions, adverse drug reactions (ADRs) induced by the toxicity of TCMIs have occurred occasionally [6]. In recent years, the potential toxicity of TCMIs has seriously affected its clinical application [7]. 5-Hydroxymethylfurfural (5-HMF, Fig. 1A) is a furanic compound that is produced by the Maillard reaction and caramelization during food processing and Chinese medicinal herb preparation [8]. 5-HMF formation in TCMs is mainly derived from the heating decomposition of sugars under acidic conditions [9]. The intake of 5-HMF is harmful to human health because it stimulates the upper respiratory tract, skin and mucous membranes [10][11][12]. In Chinese Pharmacopeia, the content of 5-HMF in glucose injections has been regulated since 1985 [13].
To analyse the contents of 5-HMF, CML and CEL in food matrices, many techniques, including highperformance liquid chromatography (HPLC) [25], reversed-phase high-performance liquid chromatography (RP-HPLC) [26], liquid chromatography-mass spectrometry (LC-MS) [27], highperformance liquid chromatography coupled to tandem mass spectrometry (HPLC-MS/MS) [11], ultraperformance liquid chromatography-mass spectrometry (UPLC-MS) [28] and enzyme linked immunosorbent assays (ELISA) [29], have been employed to develop analytical methods. Because of the lack of speci c antibodies and matrix effects, the ELISA method has not been widely applied as a detection method. With the advantage of repeatability and stability, HPLC, HPLC-MS/MS and UPLC-MS have been commonly used to identify and determine the contents of AGEs in various matrices [25,28]. To date, 5-HMF has been analysed in shengmaiyin [30], Zuo Gui Wan [31] and other TCMIs [32]. However, studies on the determination of CML and CEL in TCMIs are lacking.
In the present study, we developed HPLC and UPLC-MS/MS techniques to determine the levels of 5-HMF, CML and CEL in TCMIs. Validated methods were used to measure the contents of 5-HMF, CML and CEL in different TCMIs, as they may contribute to AGE formation. Preliminary analysis and evaluation were performed according to the results, which may provide a theoretical basis for standard quality evaluation of TCMIs.

Determination of 5-HMF
A stock solution of 5-HMF was prepared at a concentration of 0.15 mg/mL, which was further diluted with 10% methanol to a series of calibration standards with concentrations of 0.15, 2.34, 9.38, 37.5, and 75 µg/mL. Quality control (QC) samples were also diluted to concentrations of 0. 15, 9.38, and 75 µg/mL. A total of 200 µL of each Chinese medicine injection was diluted 10-fold with 10% methanol prior to HPLC analysis.
HPLC was conducted to determine the content of 5-HMF in different Chinese medicine injections. The separation was performed on a Hypersil ODS2 column (250 mm×4.6 mm, 5 µm, Waters Corp., Milford, MA, USA), and the column temperature was set at 30°C The mobile phase was composed of water (solvent A) and methanol (solvent B). Gradient elution began with 5% solvent B for 5 min, followed by a linear gradient of 5-15% solvent B from 5 min to 15 min, and then returning to a linear gradient of 15%-5% solvent B from 15 min to 20 min. The ow rate was 1.0 mL/min, and the injection volume was 20 µL.
Method validation was conducted according to the US Food and Drug Administration guidelines [33], and method validation included selectivity, linearity, accuracy, precision, recovery and stability determination.
The selectivity of the method was determined by evaluating 6 blank reagent injections (10% methanol). The linearity was determined by using calibration standard samples at concentrations of 0.15, 2.34, 9.38, 37.5, and 75 µg/mL, and each level was independently injected 3 times. Calibration curves were constructed by plotting the average peak areas (y) of each sample against the calibration concentrations (x). The limits of detection (LODs) and quanti cation (LOQs) were evaluated by calculating the concentrations corresponding to signal-to-noise (S/N) ratios of 3 and 10, respectively [34]. The intra-and inter-day precision and accuracy were determined by measuring the concentrations of QC samples and were evaluated by the relative standard deviation (RSD). The stability of 5-HMF was determined by evaluating QC samples under three conditions, including three freeze-thaw cycles (-80°C), storage at room temperature for 4 h, storage at 4°C for 1 d, and storage at -20°C for 21 d. The recovery was assessed by calculating the ratios of concentrations between analytes in the QC samples and those in samples with known concentrations.

Determination of CML and CEL
A stock solution of CML and CEL was prepared with each at a concentration of 1 mg/mL, which was diluted to working solutions at a series of concentrations of 0.078, 0.313, 1.25, 5, 20, and 640 ng/mL. Two hundred microlitres of a Chinese medicine injection was measured and diluted 10-fold with ultrapure water, and 2 µL was injected into the UPLC system for analysis. The method validation conducted as outlined in the description of "Determination of 5-HMF". In brief, 6 blank reagent samples (ultrapure water) were used to measure the selectivity. The intra-/inter-day precision and accuracy were calculated using calibration standards at 0.078, 1.25 and 20 ng/mL. The precision was determined by the RSD, and the accuracy was expressed as RE. Recovery was determined by comparing the peak area ratios of samples spiked with calibration standards and the original amount.
The stability of the working solution for CML and CEL was evaluated under three freeze-thaw cycles (-80°C), at room temperature for 4 h, at 4°C for 1 d, and at -20°C for 21 d.

Results And Discussion
Quanti cation of 5-HMF

Chromatogram of 5-HMF
The chromatograms of 5-HMF in the reference solution and that in the Honghua injection are presented in Fig. 2. The retention time of 5-HMF was 9 min, and no interference peaks were observed.

Precision, accuracy and recovery
The intra-day precision and inter-day precision are listed in Table 1, and the inter-day precision (RSD) and accuracy (RE) ranged from 2.15 to 5.35% and 2.27 to 4.79%, respectively; the intra-day precision and accuracy ranged from 1.53 to 3.97% and 1.70 to 3.88%, respectively. The recoveries of the three tested concentrations were more than 80.2%.

Stability
The stability of 5-HMF was veri ed under various storage conditions ( Table 2). The levels of 5-HMF did not change signi cantly, with an RE ranging from -0.31 to 6.16. The results showed that the analyte exhibited excellent stability under the four storage conditions. Quanti cation of 5-HMF in TCMIs by HPLC The contents of 5-HMF in 16 TCMIs are listed in Table 3, and the results showed that the contents of 5-HMF were notably high in the Ciwujia (74.98 µg/mL) and Shengmai (65.18 µg/mL) injections and were low in the Qingkailing, Kudiezi and Tianmasu injections, which were even under 1 µg/mL. 5-HMF could not be detected in the Qukeluding, Xingnaojing, and Yinxing leaf extracts or in the Shuxuening, Salvianolate and Xueshuantong injections. According to the literature, the contents of 5-HMF in different TCMIs varied from 0.009 to 4100 mg/L [35], and the results in our research were also within that range. Although the content of 5-HMF in glucose injections is limited (11.8 µg/mL) in the Chinese Pharmacopeia [13], we found that the contents of 5-HMF varied with different injection types, manufacturers and batches; thus, we suggested formulating uni ed content standards for 5-HMF in different TCMIs.

Mass spectra and chromatography
The UPLC-MS chromatograms of CML and CEL standards and in the Ciwujia injection are shown in Fig.   3. The chromatographic separation was optimized, and the retention times of CML and CEL were 0.6 min. The precursor ions for CML and CEL were observed at m/z 130.14 and 130.15, respectively. CML and CEL were quanti ed by the major ion at m/z 84.10 (Fig. 4).

Method validation
As shown in Table 4, within the range of 0.078-640 ng/mL, the regression equations of CML and CEL presented great linearity with correlation coe cients (R 2 ) >0.99. The LODs and LOQs were 0.15 and 0.48 for CML and 0.13 and 0.41 for CEL, respectively.
In this study, the inter-/intra-day precision of CML was less than 3.52% and 2.96%, while the inter-/intraday accuracy of CML was below 2.68% and 3.48%, respectively. For CEL, the inter-/intra-day precision ranged from 0.89% to 3.31% and 1.90% to 3.77%, respectively, and the inter-/intra-day accuracy ranged from 0.94% to 3.20% and -0.29% to 4.15%, respectively. The mean recoveries of CML and CEL were 81.4% and 79.6%, respectively. All results were investigated and are listed in Table 4 and Table 5. These results indicated that the method was feasible and e cient for the detection and quanti cation of CML and CEL.

Stability
The stability of CML and CEL was investigated under four storage conditions ( Table 6). The results indicated that the analytes exhibited great stability, as the concentrations were not signi cantly changed under the different conditions.

Quanti cation of CML and CEL in TCMIs by UPLC-MS/MS
The contents of CML and CEL in 16 TCMIs are listed in Table 7, and the results indicated that CML and CEL were detected in 5 and 9 TCMIs, respectively. The contents of CML and CEL in these injections were 0.51-7.32 ng·mL-1 and 0.38-5.49 ng·mL-1, respectively. The contents of CML (7.32 ng/mL) and CEL (5.49 ng/mL) in the Shuxuetong injection were much higher than in the other injections. The CEL level in the Shuanghuanglian powder injection was the lowest (0.38 ng/mL), while the CML level in the Ciwujia injection was the lowest (0.51 ng/mL). We deduced that high protein and fat levels raised the CML and CEL contents in the Shuxuetong injection, as its components were Hirudo and Lumbricus; likewise, high levels of CML and CEL were frequently generated during milk processing [24]. Moreover, the Ciwujia injection is rich in avones, anthraquinones and other secondary metabolites that could stabilize MGO and GO to limit the generation of CML and CEL during the Maillard reaction [36].

Conclusions
A validated, accurate analytical method using HPLC was developed for the determination of 5-HMF in TCMIs. Simultaneously, we developed and validated a rapid and sensitive method using UPLC-MS/MS for the determination of CML and CEL in TCMIs for the rst time. Based on the methods, the contents of 5-HMF, CML and CEL in 16 TCMIs were successfully quanti ed. The methods exhibited excellent recoveries, inter-/intra-day precision and low LODs and LOQs. Among the results, the Ciwujia injection had the highest content of 5-HMF, and the Qingkailing injection had the lowest 5-HMF content. On the other hand, the contents of CML and CEL in the Shuxuetong injection were much higher than in the others. The methods exhibited simple, rapid and high accuracy that could provide a theoretical basis for the quality evaluation of TCMIs.     Typical MRM chromatograms of CML and CEL in a standard solution (A) and in the Ciwujia injection