Direct determination of 3-chloro-1,2-propanediol esters in beef flavoring products by ultra-performance liquid chromatography tandem quadrupole mass spectrometry

Abstract

A sensitive ultra-performance liquid chromatography tandem quadrupole mass spectrometry (UPLC-TQ-MS) method coupled with amino solid-phase extraction was developed for the direct determination of 3-chloro-1,2-propanediol (3-MCPD) esters which were firstly detected in the natural beef flavoring products. The analyzed 3-MCPD esters in this paper contain three monoesters and six diesters. The validation data indicated that the proposed method provided good linearity, repeatability and sensitivity. The method showed a good linearity within the range of 40–200 μg kg⁻¹ for monoesters and 20–100 μg kg⁻¹ for diesters with the determination coefficients (R²) ranging from 0.9912 to 0.9993. The limits of detection (LODs) for monoesters and diesters of 3-MCPD were in the range of 0.04–5.0 μg kg⁻¹ and 0.13–16.0 μg kg⁻¹, respectively. The method accuracy was confirmed by higher sample recovery which ranged from 80.5 to 113.7%. The repeatability expressed as intra-day precision ranged from 0.8 to 9.9% and inter-day precision varied from 2.8 to 13.8%. The validated method was successfully applied for determination of 3-MCPD esters in beef flavoring products.

---

1. Introduction

Free 3-chloro-1,2-propanediol (3-MCPD) is known as a food toxicant which exhibits adverse effects on the kidney, immune, and central nervous system functions, and testicular organogenesis.^1–4 It was mostly found in acid hydrolyzed vegetable protein (HVP) and its derived products.^5–7 Recently, 3-MCPD was discovered in many foods not only in free form but mainly as 3-MCPD esters.⁸ 3-MCPD esters existed in some types of foodstuffs, such as refined oils and products containing significant amounts of refined oils (e.g. French fries, potato crisps, roasted coffee and infant formulae).^8,9 Refined edible oils had a higher amount of 3-MCPD esters than other foodstuffs. And 3-MCPD esters are mainly formed during the deodorization step of the refining process due to the effect of high temperature.^10–12 3-MCPD esters can also be formed by intrinsic thermal formation at high temperatures during the manufacture of refined edible oils and some products with the presence of 3-MCPD ester precursors, such as nuts and biscuits.¹¹

Compared with the simple thermal reaction, it shows more favorable for the formation of 3-MCPD esters in the process of producing the beef flavorings and the earlier work of this study has proved their formation. However, less work was focused on the relationship between 3-MCPD esters and the savory flavorings. In order to determine the level of 3-MCPD esters in foodstuffs, different determination methods have also been published, and the methods can be divided into two categories: direct and indirect. Currently, most analytical methods for determining the accurate concentration of 3-MCPD esters in food are indirect methods based on the measurement of free 3-MCPD after transesterification.^13–15 This method typically includes the following steps: the release of free 3-MCPD from esterified form based on either acidic¹⁶ or alkaline,¹⁷ purification by liquid–liquid extraction, then derivatization of the released 3-MCPD by PBA,^18,19 HFBI²⁰ and quantification by gas chromatography-mass spectrometry (GC-MS).²¹

However, the major drawback of the above-mentioned method based on alkaline hydrolysis is the possible formation of additional 3-MCPD during the analysis, leading to positive biased results, which was attributed to the fact that glycidol and its fatty acid esters from alkaline catalyzed transesterification process can be converted into 3-MCPD.²² Moreover, there are considerable impacts of the time and other factors (e.g. pH value) of the transesterification, salting out and derivatization processing on the recovery, specificity and trueness of indirect methods for the determination of 3-MCPD esters.²³

The indirect methods provided information about the total content of 3-MCPD, but cannot confirm the detailed information concerning that 3-MCPD esters exist in monoesters or diesters and the types of fatty acids. Though toxicological data on 3-MCPD esters were rarely limited, the experiment with a simple intestinal model confirmed that 3-MCPD esters could be hydrolyzed by mammalian intestinal lipases to release the free 3-MCPD which was toxicologically well characterized.²⁴ However, the 3-MCPD esters structure would affect their absorption rate, metabolism pathways and toxicological properties.^24–26 Therefore, it is deeply necessary to develop a highly efficient separation and analysis method for 3-MCPD esters (monoesters and diesters). With the development of analytical techniques, the present direct methods mainly include the use of ultra-high performance liquid chromatography,²⁷ time-of-flight MS²⁸ or LC-MS/MS.^29–31

Compared to refined oils, the savory flavoring systems are much more complicated, including fats, sodium chloride, amino acids, amino acid salts, and spice. The aim of this study was to develop a (UPLC-TQ-MS) method, which was suitable to analyze 3-MCPD esters in the natural beef flavoring products. The most important part of this work is to develop a sample preparation procedure for the natural beef flavoring samples. The method was validated for linearity, precision, accuracy, LODs and LOQs.

2. Materials and methods

2.1 Chemicals and reagents

The 3-MCPD ester standards were purchased from Toronto Research Chemicals Inc. (North York, Canada), including 1-stearoyl-3-chloro-1,2-propanediol (1-St), 1-oleoyl-3-chloro-1,2-propanediol (1-OL), 1-palmitoyl-3-chloro-1,2-propanediol (1-Pa), 1,2-dioleoyl-3-chloropropanediol (OL-OL), 1,2-dipalmitoyl-3-chloropropanediol (Pa-Pa), 1,2-distearoyl-3-chloropropanediol (St-St), 1-palmitoyl-2-stearoyl-3-chloropropanediol (Pa-St), 1-oleoyl-2-stearoyl-3-chloropropanediol (OL-St) and 1-palmitoyl-2-oleoyl-3-chloropropanediol (Pa-OL). The purities are 98% except for OL-OL and 1-OL (95%). Methanol, hexane, 2-propanol (IPA), ammonium acetate and dichloromethane (DCM) are all of HPLC grade. Ethyl acetate, hexane and methyl tert-butyl ether (MTBE) are analytical grade. Sep-Pak® Vac silica cartridges (1000 mg) and Sep-Pak® C₁₈ Vac. Cartridges (1000 mg) were obtained from Waters (Ireland). CNWBOND NH₂ SPE cartridges (1000 mg) were purchased from ANPEL Laboratory Technologies Inc (shanghai, China).

2.2 Sample preparation

2.2.1 Preparation of enzymatic hydrolyzed tallow. Crude tallow in phosphate buffered solution was placed in the enzyme reactor with mechanical stirring. Lipase was added to the reactor with enzyme/substrate (E/S) ratio of 9.5 × 10⁻³ (g lipase/g tallow) when the mixture was adjusted to be isothermal to the water bath (45 °C). The sample was heated to 95 °C for 10 min to deactivate the enzyme after reacting for 4–8 h and stored at −18 °C for further analysis.

2.2.2 Preparation of enzymatic hydrolyzed bovine bone. Bone cement (water content, 65.38%; protein content, 9.75%) and deionized water were placed in enzyme reactor with magnetic stirring for thermal denaturation at 50 °C for 5 h. Protease was added to the reactor with enzyme/substrate (E/S) ratio of 1.5 × 10⁻² (g lipase/g protein) when the mixture was adjusted to suitable temperature and pH. After reacting for 5 h, the sample was heated to 95 °C for 10 min to deactivate the enzyme and then stored at −18 °C for further analysis.

2.2.3 Preparation of beef flavorings (BFs). A mixture of HVP, yeast extract, DL-methionine, D-xylose, glucose, L-glutamic acid, L-cysteine, taurine, spices and enzymatic hydrolyzed tallow, was dissolved in solution of the bovine bone hydrolysate. The solution was transferred into 50 mL screw-sealed tubes and then heated in a thermostatic oil bath with mechanical stirring (150 rpm) at 105 °C for 70 min. After reaction, the tubes were immediately cooled in ice-water and six Maillard reaction products named BF1-6 were stored at −18 °C for further analysis. The ultimate goal of this study is to ensure the safety production of beef flavor, and these six samples were randomly selected from samples which were prepared during the study of control strategies.

2.3 Purification of samples

Beef flavoring samples (1 g ± 0.02 g) were accurately weighed and dissolved in a mixture of MTBE and ethyl acetate (4 : 1 v/v, 5 mL), vortexed for 5 min and centrifuged to obtain a clear organic phase layer at 4000 rpm for 5 min (TDL-5-A, shanghai, China). Two milliliters of the organic phase layer (supernatant) were transferred to a 10 mL centrifuge tube and the solvent was removed under nitrogen stream at room temperature. The residue was dissolved in 2 mL mixture of IPA : DCM (1 : 1, v/v). Subsequently, 0.5 mL of the mixture was removed and loaded into a SPE cartridge (silica, C₁₈ or NH₂) pre-conditioned with hexane (10 mL). The target compounds were eluted by 6 mL IPA : DCM (1 : 1, v/v) and the solvents were removed under nitrogen stream at room temperature. The residue was dissolved in 1 mL methanol : IPA (7 : 3, v/v) for further quantification of 3-MCPD esters on UPLC-TQ-MS.

2.4 UPLC-TQ-MS analysis

The qualitative analysis and quantitative determination of 3-MCPD esters was carried out by UPLC-TQ-MS (Waters, USA). UPLC separation was performed using a reversed phase ACQUITY UPLC® BEH C₈ column (1.7 μm particle, 2.1 × 50 mm, Waters). The chromatographic conditions were as follows: flow rate 0.3 mL min⁻¹, the column oven was set at 40 °C, sample injection volume of 1 μL and mobile phase A (100% methanol) and mobile phase B (3 mM ammonium acetate solution). A gradient programme was used as follows: 70% A and 30% B, 0–0.5 min, linearly changed to 100% A within 12.5 min and remained for 3 min, back to 70% A and 30% B in 0.5 min, 10 min of reconditioning before the next injection.

Mass spectrometry was performed with a Waters TQD system for data collection. Compounds were analyzed under the positive ion (PI) mode with electrospray ionization (ESI) source. The parameters of the mass spectrometer were shown as follows: the ion source temperature 130 °C; cone gas flow 50 L h⁻¹; the argon collision gas flow 0.1 mL min⁻¹; the capillary voltage 3.47–3.50 kV, cone voltage 20.00–32.48 V, second cone voltage 3.00–3.17 V, hexapole lens voltage 0.30 V; the desolvation gas temperature 400 °C at a flow rate of 600 L h⁻¹. The mass range was set between m/z 100 and 1000.

2.5 Method validation

The method was validated according to the guideline of the European Commission Decision 2002/657/EC in terms of linearity, matrix effects (ME), limit of detection (LOD) and limit of quantification (LOQ), accuracy and precision expressed by the results of recovery experiments.³²

Linearity was verified by standard solutions in beef flavorings at the concentration of 40–200 μg kg⁻¹ for monoesters and 20–100 μg kg⁻¹ for diesters. In this study, the 3-MCPD ester standards were added to blank beef flavorings samples to evaluate the matrix interference. The detailed procedures were that appropriate volumes of 3-MCPD esters standard solution in blank beef flavorings samples in order to achieve these concentration levels, and the following procedures were accordance with sample preparations. Matrix effects (ME) were determined by constructing calibration curves in blank beef flavorings samples and in the pure solvent and were calculated as slope of spiked extract/slope of pure solvent. The accuracy was evaluated by recovery test, and the precision was expressed as intra-day precision (RSD_r) and inter-day precision (RSD_R). The recovery was calculated as the ratio of the quantified levels of 3-MCPD esters in the spiked beef flavorings to the known amount of 3-MCPD esters used for spiking in samples. Intra-day precision was evaluated by the blank spiked samples, and analyzed in the same day at four concentration levels (Table 3). For inter-day precision (RSD_R), the four concentrations were analyzed in three different days. LOD and LOQ were estimated for a signal-to-noise (S/N) ratio of 3 and 10, respectively and were determined experimentally by analyzing spiked blank beef flavorings samples.

3. Results and discussion

3.1 Selection of standards

For direct determination of 3-MCPD esters in beef flavoring, it is crucial to select the categories of fatty acid ester standards. The fatty acids in beef flavoring are provided by the added oxidized tallow and the fatty acids of the used tallow in this work mainly contained palmitic acid (19.5%), oleic acid (38.4%) and stearic acid (29.3%). In this work, therefore, the detected 3-MCPD monoesters and diesters with the most abundant fatty acids (palmitic, oleic, and stearic acids) were selected to characterize the 3-MCPD esters in beef flavoring. To date, only a few of the MCPD esters were commercially available. According to Dubois,³³ the most important MCPD diesters consisting of at least 95% total MCPD diester content are 1-oleoyl-2-dipalmitate-3-chloropropanediol (OL-Pa), 1,2-dipalmitoyl-3-chloropropanediol (Pa-Pa), 1,2-dioleoyl-3-chloropropanediol (OL-OL), 1-linoleoyl-2-palmitoyl-3-chloropropanediol (Li-Pa), 1-linoleoyl-2-oleoyl-3-chloropropanediol (Li-OL), 1-stearoyl-2-palmitoyl-3-chloropropanediol (St-Pa) 1-oleoyl-2-stearoyl-3-chloropropanediol (OL-St), 1,2-linoleoyl-3-chloropropanediol (Li-Li), 1-linolenoyl-2-palmitoyl-3-chloropropanediol (Ln-Pa), 1-linoleoyl-2-stearoyl-3-chloropropanediol (Li-St), 1-linolenoyl-2-oleoyl-4-chloropropanediol (Ln-OL) and 1,2-distearoyl-3-chloropropanediol (St-St) in most of the oils. Considering the fatty acids composition of tallow, nine 3-MCPD esters detailed in Section 2.1 have been detected in this study.

3.2 Detection of 3-MCPD esters by UPLC-TQ-MS

The ESI source can operate under both positive and negative ionization modes, but no ions of analytes could be observed in the negative ionization mode. Due to the isotope clusters characteristic for chlorine containing compounds, three kinds of adduct ions were induced including [M + H]⁺, [M + NH₄]⁺ and [M + Na]⁺. It was very difficult to detect the protonated precursor ion of 3-MCPD esters. The [M + Na]⁺ adduct ions were induced by adding sodium acetate³⁴ or sodium formate³⁵ into the mobile phase. However, the used sodium salts would inevitably contaminate the MS system. On the other hand, the intensities of [M + NH₄]⁺ adduct ions were superior to those of the [M + Na]⁺ adduct ions. In order to maintain the stability and sensitivity of the detection, the [M + NH₄]⁺ adduct ions were selected for further studies. The skeletal structures of the target compounds are shown in Fig. 1.

Fig. 1 Structure of 3-MCPD esters. R₁ or R₂ means either palmitoyl, oleoyl or stearoyl group.

The ammonium adduct was first detected in the positive ion mode and used as a precursor ion to form the product ions in the collision cell. It was assumed that product ions were [M − (CH₂Cl–CH(OH)–CH₂–O)]⁺ ion for 3-MCPD monoesters and [M − (R₁C(O)O) + 2H]⁺ ion for 3-MCPD diesters. The multi-reaction monitoring (MRM) mode was constructed for its high sensitivity and selectivity. Retention times for analyte were detected by analyzing a mixed standard under the conditions described above and the UPLC gradient program resulted in a good fractionation (Fig. 2). Cone voltage, collision energy, monitoring ions (m/z) and retention times (min) are shown in Table 1.

Fig. 2 Chromatograms of nine 3-MCPD esters.

Table 1 Cone voltage (V), collision energy (eV), monitoring ions (m/z) and retention times (min) for 3-MCPD esters

Compound	MW^a	Cone voltage (V)	Collision energy (eV)	Precursor ion [M + NH4]^b (m/z)	Product ion^c (m/z)	Retention times (min)
a Molecular weight, calculated using 35 Da as the mass of chlorine M. The M was 3-MCPD ester.b The precursor ion was the ammonium adduct.c Product ions were [R1 + H]^+ ion for monoesters and [(CH2Cl–CH–O–R2–CH3) + H]^+ ion for diesters.
Monoesters
1-Pa	348.95	26	8	366.28	239.23	7.70
1-OL	374.99	28	10	394.24	267.24	8.03
1-St	377.00	26	8	392.29	265.26	8.90
Homo-diesters
Pa-Pa	587.36	28	18	604.5	331.22	13.06
St-St	643.36	34	20	660.57	359.23	13.85
OL-OL	639.43	34	20	656.54	357.21	13.26
Hetero-diesters
Pa-OL	612.39	30	20	630.48	357.25	13.16
Pa-St	614.41	28	16	632.48	359.26	13.46
OL-St	640.44	28	24	658.54	359.26	13.33

---

3.3 Pretreatment of samples

Similar to oils matrix, the beef flavoring systems contained abundant TAG, DAG and MAG. Based on pretreatment of oils,^24,29–31 three different fillers SPE columns (C₁₈, silica gel, amino) were chosen in this study and were evaluated by the spiked recoveries experiments using the blank matrix containing lower 3-MCPD esters concentration. The two-step solid-phase extraction (SPE) procedure similar to oils pretreatment (C₁₈–Si column for 3-MCPD monoester and Si–C₁₈ for diesters) was firstly used.^29–31 For 3-MCPD diesters, Si–C₁₈ columns were used and the results showed lower recoveries (only 56.9–79.8%). However, the results of 3-MCPD monoesters (C₁₈–Si separation procedure) showed much higher recoveries (115.3–177.8%) compared to the acceptable range of recoveries (80–110%).³⁶ The higher recoveries might be attributed to the co-eluted TAGs, MAGs and DAGs, which could also induce [R₁ + H]⁺ (R₁ see Fig. 1) product ions under LC-MS/MS detection condition. In addition, sample pretreated with only C₁₈ or Si SPE column were also evaluated. The 3-MCPD monoesters and diesters recoveries were (132.7–151.3%) and (22.9–62.9%) on C₁₈ SPE column respectively. Though Si SPE column was suitable for the separation of diesters with higher recovery (107.6–116.2%), it showed poor recovery for 3-MCPD monoesters (0–44.7%). Therefore, C₁₈ and Si SPE column were not suitable for the separation of 3-MCPD mono-/di-ester.

NH₂ SPE column were used with 6 mL DCM and DCM : IPA (9 : 1; 7 : 3; 1 : 1, v/v) for the reason that 3-MCPD esters are highly soluble in DCM and IPA. Four solvents were effectively used to elute 3-MCPD esters and DCM : IPA solvent (1 : 1, v/v) showed the higher optimal elution effect (with 87.9–103.0% and 92.9–113.0% recovery for monoester and diesters respectively). The mixture of DCM : IPA (1 : 1, v/v) was chosen as elution for further studies.

Further determination of the eluent solvent volume, the DCM–IPA (1 : 1 v/v) solvent was used to elute 3-MCPD monoester and diester. Two milliliters of eluent were separately collected and the solvent was removed under nitrogen stream at room temperature, then the residues were dissolved in 5 mL methanol : IPA (7 : 3, v/v) for UPLC-TOQ-MS quantification. Results showed that both 3-MCPD monoester and diester could be recovered with 6.0 mL of DCM–IPA (1 : 1 v/v) elution solvent (94.6–99.3%). There was slight change in recovery with increasing the elution solvent volume to 8 mL. And there might be slight changes in experiments, such as the tightness of the filter, the elution flow rate difference and relatively higher content of 3-MCPD mono-/di-ester in some samples. Therefore, eight milliliters of DCM–IPA (1 : 1 v/v) elution solvent volume were selected over 6 mL for further studies.

3.4 Method validation

3.4.1 Linearity and matrix effect. The calibration curves of the 9 target 3-MCPD esters were linear in the range from 40 to 200 μg kg⁻¹ for monoesters and from 20 to 100 μg kg⁻¹ for diesters. As shown in Table 2, where x was the concentration of 3-MCPD esters, and y was the peak area. The determination coefficients (R²) ranged from 99.12 to 99.93%, indicting a good linearity and satisfactory determination coefficients for nine analytical 3-MCPD esters.

Table 2 Linearity equation, linear range, R², LOD, LOQ and matrix effect of 3-MCPD esters in beef flavorings

Compound	Linear range (μg kg^−1)	Linearity equation	R^2	LOD (μg kg^−1)	LOQ (μg kg^−1)	Matrix effect
1-Pa	40–200	y = 18.57x − 0.0903	0.9993	5.0	16	1.01
1-St	40–200	y = 15.541x + 10.63	0.9946	1.0	4	0.99
1-OL	40–200	y = 9.3163x + 4.3642	0.9912	2.0	6	1.18
Pa-Pa	20–100	y = 329.01x + 29.145	0.9953	0.5	1.8	1.02
St-St	20–100	y = 470.46x − 2.7796	0.9945	1.0	4.0	1.07
OL-OL	20–100	y = 6231.1x − 25.014	0.9973	0.04	0.13	1.14
Pa-OL	20–100	y = 1178x − 15.411	0.9989	0.08	0.30	1.09
Pa-St	20–100	y = 407.62x − 43	0.9925	0.2	0.8	0.93
OL-St	20–100	y = 1483.2x − 34.846	0.9969	0.3	1.0	1.15

---

Matrix effect is an analytical phenomenon, which matrix components may enhance or suppress the response signal when ESI is used.³⁷ The beef flavoring is a complicated system including amino acids, sugars and fats, which might influence the detection of 3-MCPD esters. In this study, matrix effect was calculated as slope of spiked blank beef flavorings samples/slope of pure solvent and it was considered tolerable if the value was between 0.8 and 1.2.³⁸ As shown in Table 2, the values were ranged from 0.93 to 1.18. Therefore, calibration standard curves can be used to quantify 3-MCPD esters in beef flavoring samples.

3.4.2 Limit of detection (LOD) and limit of quantification (LOQ). As shown in Table 2, the LODs for 3-MCPD monoesters and diesters were ranged from 1.0 to 5.0 μg kg⁻¹ and from 0.04 to 1.0 μg kg⁻¹ in beef flavorings, and the LOQs were ranged from 4 to 16 μg kg⁻¹ and from 0.13 to 4.0 μg kg⁻¹, respectively.

The content of 3-MCPD esters in sample was calculated using the following formula:

C_S = 5 × C_C × V/m

Where C_S is the content of 3-MCPD esters in sample (μg kg⁻¹); 5 is the ratio of extraction solution/solution added to NH₂ SPE column; C_C is the calculated concentration of 3-MCPD esters by calibration curve (μg kg⁻¹); V is the volume of solvent DCM–IPA (1 : 1 v/v) (mL); m is the weight of sample (g).

3.4.3 Accuracy and precision. The recovery and repeatability were evaluated by analyzing the blank beef flavoring sample spiked with 9 analytes at four different concentration levels (Table 3). The recovery values were ranged from 83.6 to 105.5% for 3-MCPD monoesters and from 80.5 to 113.7% for 3-MCPD diesters, respectively. The results indicated the acceptable recovery. The intra-day precision ranged from 0.8 to 9.9%, and the inter-day precision ranged from 2.8 to 13.8% within the acceptable levels. The recovery and RSD results indicated the suitability of this method for 3-MCPD esters analysis.

Table 3 The recovery and RSD results of 3-MCPD esters

Compound	Spiked level (μg kg^−1)	Recovery (%)	RSDr^a (%)	RSDR^b (%)
a Relative standard deviation of intra-day (n = 6).b Relative standard deviation of inter-day (n = 6).
1-Pa	20	83.6	7.4	10.7
	50	92.9	8.9	9.6
	250	104.1	3.6	5.4
	1000	105.5	2.7	3.1
1-OL	20	101.8	9.3	9.8
	50	88.7	6.6	7.8
	250	96.6	8.5	11.2
	1000	100.7	2.8	3.4
1-St	20	88.7	7.9	10.4
	50	95.8	9.9	11.4
	250	92.5	8.6	13.8
	1000	104.5	1.5	2.6
Pa-Pa	3	105.3	8.2	12.8
	50	85.6	6.5	7.3
	250	97.9	0.8	8.6
	1000	88.9	1.6	2.4
St-St	6	82.4	8.9	10.9
	50	91.6	6.9	10.8
	250	90.2	0.4	4.9
	1000	109.7	2.2	2.8
OL-OL	3	109.2	8.2	11.3
	50	108.6	2.5	11.3
	250	90.6	3.2	8.4
	1000	110.2	1.4	8.2
Pa-OL	3	113.7	6.9	11.5
	50	81.2	4.8	10.6
	250	103.7	3.1	10.5
	1000	87.6	3.5	9.9
Pa-St	3	110.3	4.6	6.3
	50	80.5	3.8	12.4
	250	105.8	2.8	9.2
	1000	94.3	2.4	10.8
OL-St	3	105.5	5.6	11.6
	50	94.8	4.6	9.3
	250	82.2	3.4	9.6
	1000	109.2	2.8	10.8

---

3.4.4 Application to actual samples. The developed method with UPLC-TQ-MS to analyse 3-MCPD esters in natural beef flavoring samples was validated using six beef flavoring products. The contents of 3-MCPD esters were calculated as 3-MCPD equivalent. As shown in Table 4, the 3-MCPD esters contents are 30.6–501.7 μg kg⁻¹. There are no related reports on 3-MCPD esters in the savory flavoring production up to day; however, 3-MCPD contents of beef flavoring with the Maillard reaction were detected (2.9167–53.8714 μg kg⁻¹).³⁹ The 3-MCPD esters contents are much more than 3-MCPD contents and it can explain 3-MCPD mainly exists in the form of fatty acid ester in beef flavoring products.

Table 4 Contents of 3-MCPD esters in actual beef flavoring products^a

Sample	3-MCPD monoesters (μg kg^−1)	3-MCPD diesters (μg kg^−1)	3-MCPD esters (μg kg^−1)
a The contents of 3-MCPD esters were all calculated as 3-MCPD.b ND: not detected.
BF1	ND^b	85.1	85.1
BF2	50.3	56.3	106.6
BF3	49.5	365.3	414.8
BF4	ND^b	399.2	399.2
BF5	244.7	257.0	501.7
BF6	ND^b	30.6	30.6

---

4. Conclusion

Although the issue of 3-MCPD esters has aroused worldwide attention, there are no relevant reports on 3-MCPD esters from beef flavoring products. A direct analytical method for quantification of 3-MCPD esters in beef flavoring products using LC-MS/MS without ester cleavage was developed and validated. This method was established according to the characteristics of the beef flavoring products based on detection methods of 3-MCPD esters in edible oils and fats. However, the required time and organic reagent for the cleanup of one sample were far less than those of the previous the two-step SPE method. Also, the results showed that this method could be a promising tool for quantitative determination of 3-MCPD esters in the beef flavoring products due to its good linearity, repeatability and low limit of quantification.

Acknowledgements

The authors greatly appreciate the joint support for the study by the program of ‘‘Collaborative innovation center of food safety and quality control in Jiangsu Province’’ and National Program of China (2013AA102204).

References

1. J. K. Lee, J. A. Byun, S. H. Park, H. S. Kim, J. H. Park, J. H. Eom and H. Y. Oh, Toxicology, 2004, 204, 1–11.
2. K. Kim, C. Song, Y. Park, S. Koh, J. Kim, S. Kim, Y. Kim, S. U. Kim and H. Jung, Neurotoxicology, 2004, 25, 377–385.
3. K. Kim, Food Chem. Toxicol., 2008, 46, 955–960.
4. R. El Ramy, M. Ould Elhkim, M. Poul, M. G. Forest, P. Leduque and B. Le Magueresse-Battistoni, Reprod. Toxicol., 2006, 22, 485–492.
5. K. O. Wong, Y. H. Cheong and H. L. Seah, Food Control, 2006, 17, 408–413.
6. R. Macarthur, C. Crews, A. Davies, P. Brereton, P. Hough and D. Harvey, Food Addit. Contam., 2000, 17, 903–906.
7. W. S. Fu, Y. Zhao, G. Zhang, L. Zhang, J. G. Li, C. D. Tang, H. Miao, J. B. Ma, Q. Zhang and Y. N. Wu, Food Addit. Contam., 2007, 24, 812–819.
8. B. Svejkovska, O. Novotny, V. Divinova, Z. Reblova, M. Dolezal and J. Velisek, Czech J. Food Sci., 2004, 22, 190–196.
9. R. Weißhaar, Eur. J. Lipid Sci. Technol., 2008, 110, 671–672.
10. Z. Zelinková, B. Svejkovská, J. Velíšek and M. Doležal, Food Addit. Contam., 2006, 23, 1290–1298.
11. C. G. Hamlet, L. Asuncion, J. Velíšek, M. Doležal, Z. Zelinková and C. Crews, Eur. J. Lipid Sci. Technol., 2011, 113, 279–303.
12. K. Franke, U. Strijowski, G. Fleck and F. Pudel, LWT--Food Sci. Technol., 2009, 42, 1751–1754.
13. J. Kuhlmann, Eur. J. Lipid Sci. Technol., 2011, 113, 335–344.
14. K. Hrncirik, Z. Zelinkova and A. Ermacora, Eur. J. Lipid Sci. Technol., 2011, 113, 361–367.
15. V. Divinova, B. Svejkovska, M. Dolezal and J. Velisek, Czech J. Food Sci., 2004, 22, 182–189.
16. V. Divinova, B. Svejkovska, M. Dolezal and J. Velisek, Czech J. Food Sci., 2004, 22, 182–189.
17. R. Weißhaar, Eur. J. Lipid Sci. Technol., 2008, 110, 183–186.
18. M. Kusters, U. Bimber, S. Reeser, R. Gallitzendorfer and M. Gerhartz, J. Agric. Food Chem., 2011, 59, 6263–6270.
19. J. Kuhlmann, Eur. J. Lipid Sci. Technol., 2011, 113, 335–344.
20. C. G. Hamlet and L. Asuncion, Eur. J. Lipid Sci. Technol., 2011, 113, 345–355.
21. R. A. A. Razak, A. Kuntom, W. L. Siew, N. A. Ibrahim, M. R. Ramli, R. Hussein and K. Nesaretnam, Food Control, 2012, 25, 355–360.
22. L. Karasek, T. Wenzl and F. Ulberth, Eur. J. Lipid Sci. Technol., 2011, 113, 1433–1442.
23. K. Hrncirik, Z. Zelinkova and A. Ermacora, Eur. J. Lipid Sci. Technol., 2011, 113, 361–367.
24. W. Seefelder, N. Varga, A. Studer, G. Williamson, F. P. Scanlan and R. H. Stadler, Food Addit. Contam., Part A: Chem., Anal., Control, Exposure Risk Assess., 2008, 25, 391–400.
25. M. Liu, B. Y. Gao, F. Qin, P. P. Wu, H. M. Shi, W. Luo, A. N. Ma, Y. R. Jiang, X. B. Xu and L. L. Yu, Food Chem. Toxicol., 2012, 50, 3785–3791.
26. T. Buhrke, R. Weisshaar and A. Lampen, Arch. Toxicol., 2011, 85, 1201–1208.
27. H. Li, D. Chen, H. Miao, Y. Zhao, J. Shen and Y. Wu, J. Chromatogr. A, 2015, 1410, 99–109.
28. E. Moravcova, L. Vaclavik, O. Lacina, V. Hrbek, K. Riddellova and J. Hajslova, Anal. Bioanal. Chem., 2012, 402, 2871–2883.
29. K. Yamazaki, M. Ogiso, S. Isagawa, T. Urushiyama, T. Ukena and N. Kibune, Food Addit. Contam., Part A: Chem., Anal., Control, Exposure Risk Assess., 2013, 30, 52–68.
30. S. MacMahon, E. Mazzola, T. H. Begley and G. W. Diachenko, J. Agric. Food Chem., 2013, 61, 4737–4747.
31. S. MacMahon, T. H. Begley and G. W. Diachenko, J. Agric. Food Chem., 2013, 61, 4748–4757.
32. European Commission Decision, 2002/657/EC, Implementing Council Directive96/23/EC, Off. J. Eur. Communities: Legis., 2002, 211, 8–36.
33. M. Dubois, A. Tarres, T. Goldmann, A. M. Empl, A. Donaubauer and W. Seefelder, J. Chromatogr. A, 2012, 1236, 189–201.
34. T. D. Haines, K. J. Adlaf, R. M. Pierceall, I. Lee, P. Venkitasubramanian and M. W. Collison, J. Am. Oil Chem. Soc., 2011, 88, 1–14.
35. K. Hori, N. Koriyama, H. Omori, M. Kuriyama, T. Arishima and K. Tsumura, LWT--Food Sci. Technol., 2012, 48, 204–208.
36. S. A. Rodrigues, S. S. Caldas, M. H. Scherer Kurz, L. da Costa Cabrera, F. A. Duarte, R. Zanella and E. G. Primel, Anal. Methods, 2012, 4, 1820–1824.
37. Y. Shen, C. Han, B. Liu, Z. Lin, X. Zhou, C. Wang and Z. Zhu, J. Dairy Sci., 2014, 97, 679–686.
38. A. G. Frenich, R. Romero-Gonzalez, M. L. Gomez-Perez and J. L. Vidal, J. Chromatogr. A, 2011, 1218, 4349–4356.
39. M. Huang, Study on chlopropandiol formation in thermo-reaction meat flavours and development of new meat flavours [D], 2012.

---