Skip to main content
NCBI home page
Food Chemistry: X logoLink to Food Chemistry: X
. 2024 Sep 27;24:101858. doi: 10.1016/j.fochx.2024.101858

Rapid and convenient determination of α-dicarbonyl compounds via streamlined simultaneous derivatization and extraction coupled with liquid chromatography-fluorescence detection

Yanbo Luo a, Yuwei Liu b,c, Xiangyu Li a, Xingyi Jiang a, Yongqiang Pang a,⁎⁎, Di Chen b,c,⁎
PMCID: PMC11471521  PMID: 39403299

Abstract

α-Dicarbonyl compounds (α-DCs), formed during food heating and storage, are crucial for assessing food safety and quality. However, the low concentration, high reactivity, and absence of chromophores of α-DCs make their detection challenging, often requiring complex derivatization and extraction. This study developed a one-pot method for α-DCs that combines derivatization and magnetic solid-phase extraction. By mixing the sample, 2,3-diaminonaphthalene, and Fe3O4/MWCNTs-OH in a vial, simultaneous derivatization and extraction are achieved. Derivatization converts α-DCs into hydrophobic products, facilitating their adsorption and enabling sensitive liquid chromatography-fluorescence detection. The introduction of the magnetic adsorbent allows phase separation to be easily achieved using an external magnet, simplifying and speeding up the process. The detection limits for six α-DCs (glyoxal, methylglyoxal, diacetyl, 2,3-pentanedione, D-glucosone, and 3-deoxyglucosone) were determined to be in the range of 0.4–3.5 nM. This rapid and convenient analytical approach was successfully applied to analyze α-DCs in juices, coffees, and tea beverages.

Keywords: α-Dicarbonyl compounds, One-pot, Chemical derivatization, Magnetic solid-phase extraction, Liquid chromatography-fluorescence detection

Graphical abstract

Unlabelled Image

Open in a new tab

Highlights

  • •

    A rapid and convenient OPD/MSPE method for α-DCs analysis was developed.

  • •

    Chemical derivatization and extraction occurred simultaneously in one vial.

  • •

    Converting α-DCs into hydrophobic products facilitates extraction/FLD detection.

  • •

    Using magnetic adsorbent avoids centrifugation, simplifying the procedure.

  • •

    α-DCs in various samples (teas, coffees, and juices) were detected.

1. Introduction

α-Dicarbonyl compounds (α-DCs) are a series of vicinal dicarbonyl compounds produced from sugars like glucose, fructose, and pentose through oxidation reactions and the Maillard reaction (He et al., 2024; Yan et al., 2023). Common α-DCs include methylglyoxal (MGO), glyoxal (GO), diacetyl (DA), 2,3-pentanedione (PD), D-glucosone (DS), and 3-deoxyglucosone (3-DG). These compounds are closely linked to the pathogenesis of chronic non-communicable diseases such as diabetes and cardiovascular diseases (Giacco et al., 2013; Rabbani & Thornalley, 2015). Detecting α-DCs in food helps assess potential health risks and ensures food safety (Shakoor et al., 2022). Additionally, α-DCs are formed through sugar oxidation and the Maillard reaction, making their detection vital for evaluating chemical reactions during food processing and controlling food quality (Yan et al., 2019). Monitoring α-DCs also provides scientific data for establishing food safety standards and reflects the freshness and storage conditions of food. Therefore, detecting α-DCs in food is crucial for assessing health risks, ensuring quality control, setting safety standards, and evaluating freshness.

When detecting α-DCs in food, two main issues must be addressed. Firstly, due to their high reactivity and instability, as well as their lack of inherent fluorescence and chromophores, chemical derivatization is usually required, with pre-column derivatization using HPLC being the most common method (Brun et al., 2023; Wang XinJie et al., 2018). Secondly, because of the low concentration of α-DCs and the complex and diverse sample matrices, enrichment and purification are often necessary to improve detection accuracy and sensitivity. Consequently, many methods such as liquid-liquid extraction (LLE), solid-phase extraction (SPE), dispersive liquid-liquid extraction (DLLE) (Rodríguez-Cáceres et al., 2017), and dispersive solid-phase extraction (DSPE) (Custodio-Mendoza et al., 2024) have been developed for this purpose.

Typically, to achieve optimal performance, a combination of chemical derivatization and extraction is used. This allows for converting the analyte into detection-sensitive products, enriching the analytes, removing interferences, and thereby improving sensitivity and selectivity. For example, Song et al. developed a method to determine α-DCs in propolis, utilizing both derivatization and extraction (Song et al., 2021). In their method, the propolis extracts were first derivatized with o-phenylenediamine and then cleaned up using an HLB solid-phase extraction column to remove lipids and beeswax before liquid chromatography-mass spectrometry analysis. However, existing sample preparation methods typically operate the derivatization and extraction steps separately. While this approach is straightforward, it results in tedious procedures and prolonged processing times. Additionally, excessive sample transfer steps increase the likelihood of errors, potentially reducing the accuracy of analytical results (Goedert & Guiochon, 1973). Integrating derivatization and extraction into a single process allows both steps to occur simultaneously, greatly simplifying the sample preparation process (Chen et al., 2021; Chen et al., 2023; Shi et al., 2024). However, as far as we know, no such method has been developed for α-DCs analysis, leading to the current methods being effective yet cumbersome and time-consuming. Thus, developing a new sample preparation method that streamlines the integrated derivatization and extraction procedure is necessary.

This study focused on developing a rapid, convenient, and sensitive method for detecting α-DCs. To achieve this, a one-pot method that integrates derivatization and magnetic solid-phase extraction (OPD/MSPE) into a single step was developed, in which the sample solution, derivatization reagent 2,3-diaminonaphthalene (DMN), and magnetic adsorbent Fe3O4/MWCNTs-OH were mixed together in a vial, achieving simultaneous derivatization and extraction. This approach streamlines the sample preparation process, minimizes sample loss, and boosts detection accuracy and sensitivity. The use of magnetic adsorbent eliminates the need for centrifugation or filtration, providing a time-efficient and rapid approach. Coupled with liquid chromatography-fluorescence detection (LC-FLD), the feasibility of the OPD/MSPE method was evaluated by analyzing six α-DCs and comparing the results with those obtained using a derivatization-only step. The main factors influencing the derivatization and extraction efficiency were optimized. The advantages of the method were illustrated by comparing it with existing methods. The feasibility of the method for real samples was demonstrated by determining α-DCs in various samples, including juices, coffees, and tea beverages.

2. Materials and methods

2.1. Chemicals and reagents

Methanol (MeOH), ethanol (EtOH), acetonitrile (ACN), and acetone (AC) (all HPLC grade) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Hydroxylated multi-walled carbon nanotubes (MWCNTs-OH, purity>95 %, inner diameter 5–12 nm, outer diameter 20–50 nm, length 10–20 μm) were obtained from Shanghai Aladdin Biochemical Technology Co., Ltd. The magnetic particles (Fe3O4, ∼50 nm) were provided by Tianjin Zhiyuan Chemical Reagent Co., Ltd. Analytical grade potassium hydrogen phosphate (K2HPO4, ≥99.0 %), sodium hydroxide (NaOH, 98.0 %), and phosphoric acid (H3PO4, ≥85.0 %) were purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). MGO and DMN were also obtained from Shanghai Aladdin Biochemical Technology Co., Ltd., while GO, DA, and PD were sourced from Adamas Reagent Technology Co., Ltd. (Shanghai, China). 3-DG and DS were purchased from Sigma-Aldrich Co., Ltd. (St. Louis, MO, USA). Ultrapure water (H2O) was produced using a Milli-Q water purification system (Millipore, Billerica, MA, USA).

Stock solutions of the six α-DCs were prepared at a concentration of 100 mM in ultrapure water and stored at 4 °C. A mixed standard stock solution of the six α-DCs was then prepared at a concentration of 1 mM, also stored at 4 °C, for subsequent dilution into working solutions. The derivatization reagent, DMN, was prepared in MeOH at a concentration of 20 mM.

2.2. Sample sources and pretreatment

The samples used in this study included juices, coffees, and tea beverages. The freshly squeezed juices, such as lemon juice, orange juice, strawberry juice, and peach juice, were all obtained from a local beverage shop called Mixue Bingcheng. The tea beverages, including black tea, green tea, oolong tea, and jasmine tea (all Master Kong brand), were purchased from a local supermarket in Zhengzhou (Henan, China). The coffee beverages, including Americano, latte, and coconut latte, were obtained from a local coffee shop called Luckin Coffee.

All samples were subjected to high-speed centrifugation at 15,000 g for 10 min or filtered through a 0.45 μm microporous membrane to remove solid particles. For the more viscous samples such as latte, and coconut latte, the samples were diluted with an equal volume of pure water after centrifugation to facilitate subsequent use.

2.3. Preparation of magnetic adsorbent

The Fe3O4/MWCNTs-OH material was synthesized using the “aggregate wrap” method as previously reported (Shi et al., 2024). In brief, Fe3O4 and MWCNTs-OH, in a mass ratio of 4:1, were precisely measured and thoroughly ground together in a mortar until a uniform fine powder was achieved. This resulting magnetic adsorbent, referred to as Fe3O4/MWCNTs-OH, was then used in the subsequent OPD/MSPE procedure.

2.4. The OPD/MSPE procedure

A schematic diagram of the OPD/MSPE method for α-DCs analysis is shown in Fig. 1. Briefly, 1 mL of the sample solution, 5 mg of Fe3O4/MWCNTs-OH, and 50 μL of DMN solution (20 mM, prepared in MeOH) and 100 μL of phosphate buffer solution (50 mM, pH 7.0) are mixed in a 1.5 mL centrifuge tube. The mixture is incubated at 60 °C in a thermostatic shaker for 40 min. Subsequently, an external magnet is used to attract the magnetic adsorbent material to the side of the centrifuge tube, allowing for easy removal of the supernatant without centrifugation. Following this, 0.3 mL of H2O is added and vortexed for 30 s to wash the adsorbent. After discarding the washing liquid, 150 μL of ACN is added and vortexed for 1 min to achieve desorption. Similarly, by applying an external magnet, the desorption solution is conveniently collected into an HPLC vial, ready for the subsequent LC-FLD analysis.

Fig. 1.

Fig. 1

Open in a new tab

Schematic diagram of (A) chemical derivatization scheme and (B) OPD/MSPE procedure.

The optimization of key parameters influencing the OPD/MSPE method was carried out using a single-factor variable approach. The parameters optimized included the pH of the sample solution, the amount of derivatization reagent (DMN), the mass of magnetic adsorbent (Fe3O4/MWCNTs-OH), incubation time, desorption solvent type and volume, and desorption time. For each optimization, a 10 μM mixed standard solution of α-DCs was used, and the mean peak areas (n = 3) of the α-DC derivatives were compared to determine the best conditions.

2.5. LC-FLD conditions

Chromatographic analysis was performed using an Agilent 1100 liquid chromatography system (Agilent Technologies, Palo Alto, CA, USA), equipped with an online degasser, autosampler, quaternary pump, column oven, and fluorescence detector. Separation was achieved on an Agilent SB-C18 column (250 × 4.6 mm, 5 μm) with water as mobile phase A and ACN as mobile phase B. The gradient elution for the separation of the six α-DCs was as follows: 26 % B for the first 10 min, increased to 45 % B at 10.5 min, held at 45 % B until 25 min, then increased to 80 % B at 26 min, maintained for 8 min, and finally returned to the initial proportion and equilibrated for 5 min. The column temperature was set at 30 °C, with a flow rate of 1 mL/min. The injection volume was 10 μL, and fluorescence detection was conducted with an excitation wavelength of 267 nm and an emission wavelength of 500 nm.

2.6. Data analysis

All chromatographic data were acquired using Agilent ChemStation software (version B.04.03). Peak integration was manually reviewed and adjusted to ensure accuracy. For method optimization, three parallel experiments were conducted, and the mean peak areas (n = 3), along with standard deviations, were compared to determine the optimal conditions. For method validation, each concentration point was tested in triplicate. Calibration curves were constructed by plotting the mean peak areas against known concentrations of α-DC standard solutions, with linear regression analysis used to calculate the correlation coefficients (R2). The limit of detection (LOD) was defined as the concentration corresponding to a signal-to-noise ratio of 3, while the limit of quantification (LOQ) was determined with a signal-to-noise ratio of 10. Accuracy and precision were evaluated through intra- and inter-day relative recoveries and relative standard deviations (RSDs) at three concentration levels. Mean relative recoveries were used to assess method accuracy, and the RSDs were calculated to determine precision. Quantification was based on the external standard method, using the calibration curves for each α-DC. All experiments were performed in triplicate, and the results were expressed as mean ± standard deviation (SD).

3. Results and discussion

3.1. Feasibility study of the OPD/MSPE method

To validate the proposed OPD/MSPE method's capability to simultaneously achieve derivatization and extraction, and to assess its efficiency, chromatograms of the standard sample solution (10 μM) subjected to derivatization alone were compared with those treated using the OPD/MSPE method.

As shown in Fig. 2, panel (a) represents the chromatogram of a 10 μM mixed standard solution of α-DCs treated with OPD/MSPE, while panel (b) shows the chromatogram of the same solution treated with derivatization alone. The retention times of the chromatographic peaks in panels (a) and (b) are consistent, indicating that the OPD/MSPE method successfully achieves derivatization. Additionally, the chromatographic peak heights in panel (a) are significantly higher than those in panel (b) owing to the enrichment effect. This indicates that the α-DC derivatives were successfully adsorbed by Fe3O4/MWCNTs-OH and subsequently desorbed by ACN. Comparing the peak areas of the products in panels (a) and (b), the extraction recoveries are calculated to be between 57 % and 80 % without optimization, indicating that the OPD/MSPE method efficiently performs both derivatization and extraction. Meanwhile, panels (c) and (d) represent the blank solution treated with the OPD/MSPE method and derivatization alone, respectively. The retention times of the derivatization reagent peaks in the blank groups are consistent with those in the experimental groups, and no other interfering peaks are present near the retention times of the α-DC derivatives. This indicates that the excess derivatization reagent does not interfere with the analysis. The above results demonstrate that the proposed OPD/MSPE method is effective.

Fig. 2.

Fig. 2

Open in a new tab

Chromatograms of (a) 10 μM mixed standard solution treated with OPD/MSPE procedure, (b) 10 μM mixed standard solution treated with derivatization alone, (c) the blank solution treated with OPD/MSPE procedure and (d) the blank solution treated with derivatization alone.

3.2. Optimization of OPD/MSPE conditions

3.2.1. The pH of the solution

The pH of the sample solution may play a crucial role in the derivatization efficiency, as it can influence the ionization state of both the analytes and the derivatization reagents. Therefore, the effect of pH was investigated using different phosphate buffer solutions (50 mM) with pH values of 3, 5, 7, 8, and 10. The results shown in Fig. 3A indicate that as the pH increased from 3 to 7, the peak areas of α-DC derivatives gradually increased. However, when the pH reached 8, the peak areas decreased. Further increasing the pH to 10 resulted in a continued decrease in the peak areas of GO and MGO, while the peak areas of other products showed no significant changes. This can be possible that excessively high or low pH values may either suppress the derivatization reaction or destabilize the derivatives formed. Therefore, a pH of 7 is most favorable for the reaction and was therefore selected as the optimal condition.

Fig. 3.

Fig. 3

Open in a new tab

Optimization of key conditions for OPD/MSPE method. (A) pH of the solution, (B) volume of DMN solution, (C) mass of adsorbent, (D) incubation temperature, (E) incubation time, (F) Type of desorption solvent, (G) volume of desorption solvent and (H) desorption time.

3.2.2. The amount of derivatization reagent

The amount of derivatization reagent is critical for ensuring complete derivatization of the α-DCs, which directly impacts the method's sensitivity. The effect of the amount of derivatization reagent was investigated by varying the volume of DMN (20 mM, prepared in MeOH) solution: 10, 30, 50, 70, and 100 μL. The results shown in Fig. 3B indicate that the peak areas of α-DC derivatives initially increase with the volume of derivatization reagent solution, followed by a slight decrease. This trend may be due to the fact that while increasing the volume of DMN solution (prepared in MeOH) initially promotes the derivatization reaction by providing more reagent, an excessive proportion of MeOH in the solution can hinder adsorption, leading to a reduction in extraction recovery. Based on these results, 50 μL was determined to be the optimal amount. Therefore, 50 μL of 20 mM DMN was selected for subsequent experiments.

3.2.3. Amount of magnetic adsorbent

The amount of Fe3O4/MWCNTs-OH can significantly affect both adsorption and desorption processes. Insufficient adsorbent may lead to incomplete adsorption, while excessive adsorbent can make desorption difficult. Therefore, the effect of the amount of magnetic adsorbent on the peak areas of α-DC derivatives was evaluated using 2, 3, 4, 5, and 6 mg of Fe3O4/MWCNTs-OH. The results shown in Fig. 3C indicate that the peak area of GO significantly increased with the amount of extraction material up to 5 mg, while the peak areas of other analytes increased more gradually. When the amount of extraction material was further increased to 6 mg, some analytes' peak areas slightly decreased, while others continued to increase. Given the minimal changes at this point, This suggests that 5 mg of Fe3O4/MWCNTs-OH provides optimal surface area for adsorption without hindering desorption, making it the ideal amount for this method.

3.2.4. Incubation temperature

Temperature is a key factor influencing both the rate of the derivatization reaction and the extraction efficiency. This study examined the effects of constant temperature shaking reactions at 25 °C, 40 °C, 50 °C, 60 °C, and 70 °C. As shown in Fig. 3D, the results indicated that the peak areas of all reaction products increased significantly between 40 °C and 50 °C, indicating enhanced reaction kinetics. However, temperatures above 60 °C resulted in a slower rate of increase or even a decrease in peak areas for some derivatives (e.g., GO and MGO), likely due to thermal degradation or volatility losses. Thus, 60 °C was chosen as the optimal temperature, balancing reaction efficiency and analyte stability.

3.2.5. Incubation time

The duration of the derivatization and extraction steps is another critical factor, as insufficient time may result in incomplete reactions, while excessive time could lead to degradation of the analytes. By comparing the peak areas (Fig. 3E) of constant temperature shaking reactions at 10, 20, 30, 40, 60, and 90 min, it was observed that after 40 min, the peak areas of GO and MGO derivatives began to slightly decrease. Additionally, the amounts of other reaction products also started to decline after 60 min, likely due to over-incubation leading to partial degradation of some analytes. Ultimately, the results at 90 min were similar to those at 40 min. Considering these factors, 40 min was chosen as the optimal reaction time for subsequent experiments.

3.2.6. Type and volume of desorption solvent

The desorption process is a crucial step in the entire experiment, with its efficiency primarily depending on the type and the volume of desorption solvent. First, the effect of different types of desorption solvents was examined using MeOH, EtOH, ACN, and AC. The results in Fig. 3F showed that MeOH had the poorest desorption efficiency. EtOH was slightly less effective than ACN and AC, which had comparable efficiency. Given AC's volatility and toxicity, ACN was chosen as the desorption solvent. Next, the effect of the amount of ACN on desorption efficiency was studied using 150, 200, 250, 300, and 400 μL of ACN. As illustrated in Fig. 3G, A volume of 150 μL provided the maximum peak area, indicating complete desorption without diluting the analyte concentration. Thus, 150 μL of ACN was selected for desorption.

3.2.7. Desorption time

The effect of desorption time on desorption efficiency was investigated at 5, 10, 30, 60, and 120 s. The results, depicted in Fig. 3H, showed that the peak area reached its maximum at 30 s, with no further improvement from longer desorption times. Considering the better repeatability and precision of results at 60 s, this was chosen as the desorption time for the experiments. Thus, the final desorption conditions were 150 μL of ACN for 1 min.

3.3. Method validation

A series of mixed standard solutions of six α-DCs with gradient molar concentrations were prepared and subjected to the OPD/MSPE method under optimal experimental conditions. The peak areas of the α-DC derivatives were plotted as the y-axis against the molar concentrations of the analytes on the x-axis, and standard curves were generated using the least squares method (Guo et al., 2024; Yang et al., 2022). The linear regression equations and correlation coefficients were obtained. As shown in Table 1, the six α-DCs exhibited good linearity within the range of 10 nM to 2 mM, with R2 ≥ 0.998. The LOD for the six α-DCs were determined using a signal-to-noise ratio of 3 and were found to be within the range of 0.4–3.5 nM.

Table 1.

Linear ranges, linear regression equations, LODs, and LOQs for the detection of six α-DCs using the OPD/MSPE method.

Analyte Linearity Range (nM) Linearity Regression Equation R2 LODs
(nM)		 LOQs
(nM)
DS 20–2000 y = 0.1224×-1.5761 0.999 3.5 11.8
3-DG 10–2000 y = 0.3092× + 0.8889 0.999 1.4 4.8
GO 10–2000 y = 0.7442×-9.1107 0.998 0.8 2.7
MGO 10–2000 y = 0.7977× + 12.7909 0.999 0.8 2.6
DA 10–2000 y = 1.3454× + 167.2837 0.999 0.4 1.5
PD 10–2000 y = 1.1354× + 46.2752 0.999 0.4 1.4
Open in a new tab

To evaluate the accuracy and precision of the method, three concentrations (20 nM, 200 nM, 1000 nM) were tested three times in one day and over three consecutive days, with each experiment performed in triplicate (n = 3). The intra-day and inter-day relative recovery rates and standard deviations were calculated (Cao et al., 2024; Paranthaman et al., 2024). As shown in Table S1, the intra-day and inter-day relative recovery rates ranged from 86.2 % to 104.2 % and 86.5 % to 105.3 %, respectively, indicating good accuracy of the method. Additionally, the intra-day and inter-day RSDs were less than or equal to 13.7 %, demonstrating good precision of the method.

3.4. Real sample analysis

The established method was applied to detect the content of six α-DCs in various samples, including four tea beverages, three coffee beverages, and four freshly squeezed juices. Each sample underwent three parallel experiments, and the mean peak areas were used to calculate the concentrations from the previously established calibration curves. When the concentration exceeded the linear range, the samples were diluted and reanalyzed. The detection results are presented in Table S2, and the typical chromatograms for each sample are shown in Fig. 4.

Fig. 4.

Fig. 4

Open in a new tab

Chromatogram of six α-DCs detected in actual samples using the OPD/MSPE method: (A) tea beverages, (B) coffee beverages, and (C) freshly squeezed juices.

From the results, DS, DA, 3-DG, GO, and MGO were detected in all samples. PD was either not detected or present at levels below the quantification limit in most samples. Freshly squeezed fruit juices generally had higher α-DCs content, with strawberry juice having the lowest content. In tea drinks, the levels of DA and PD were very low, almost undetectable. For coffee samples, Americano coffee had lower DS, 3-DG, and GO content compared to lattes, but MGO, DA, and PD levels were higher. Coconut latte had the lowest α-DCs content among coffee drinks.

The detection of actual samples indicates that α-DCs substances are widely present in beverages. The established method can be used to detect α-DCs content in various complex matrix samples. Differences in raw materials, processing methods, and storage conditions may lead to variations in α-DCs content. Detecting α-DCs content can provide a reference for public choices and scientific guidance for patient diets.

3.5. Method comparison

The proposed OPD/MSPE method offers a straightforward and convenient approach to sample preparation for detecting α-DCs, with several distinct advantages over commonly used techniques. By integrating derivatization and extraction into a single step, this method significantly simplifies the procedure and reduces sample loss. Table 2 outlines key differences between the proposed method and other reported techniques.

Table 2.

Comparison of the proposed method with previously reported methods for the determination of α-DCs.

Sample Sample preparation method Sample preparation time Detection technique LODs Ref.
Monovarietal wines / >120 min LC-FLD 2.9–3.7 ng/mL (Rodríguez-Cáceres et al., 2015)
Alcoholic beverages DLLME >20 min LC-FLD 1.5 ng/mL (Rodríguez-Cáceres et al., 2017)
Red wine and urine after drinking Decolorization using activated carbon >60 min LC-MS/MS 12.5–50 fmol (Qi et al., 2023)
Sesame oils DLLME >60 min GC–MS 0.22–0.86 ng/mL (Lee et al., 2024)
Urine SALLE and DLLME >25 h GC–MS 0.06–0.12 ng/mL (Pastor-Belda et al., 2017)
Juices, coffees, and tea beverages OPD/MSPE ∼42 min LC-FLD 0.4–3.5 nM This work
Open in a new tab

First, the OPD/MSPE method is easy to operate, as it completes both derivatization and extraction in one step. This eliminates multiple phases, reduces the potential for errors and sample loss, and improves the repeatability and reliability of the results.

Second, the use of magnetic adsorbents removes the need for centrifugation, streamlining the procedure and greatly reducing sample preparation time. With a total preparation time of approximately 42 min, the OPD/MSPE method is considerably faster than other labor-intensive techniques such as solid-phase extraction (∼120 min) or more complex methods like salting-out-assisted liquid-liquid extraction (SALLE) combined with dispersive liquid-liquid microextraction (DLLME), which can take over 25 h. This time efficiency makes the one-pot method especially valuable for high-throughput analyses in food safety and quality control applications.

In terms of sensitivity, the OPD/MSPE method demonstrates LODs ranging from 0.4 to 3.5 nM, which are competitive with, or even lower than, those reported for LC-MS/MS methods. While mass spectrometry-based techniques are known for their precision and sensitivity, they are often expensive and require specialized equipment, limiting their use in routine analyses. The use of LC-FLD in this method provides a cost-effective yet sensitive alternative, suitable for practical applications that demand high sensitivity without the complexity and expense of LC-MS/MS.

However, the limitations of the OPD/MSPE method should also be acknowledged. One potential limitation is its reliance on hydrophobic interactions between the α-DC derivatives and the magnetic adsorbent (Fe3O4/MWCNTs-OH), which makes it most suitable for water-soluble matrices or samples easily processed into aqueous solutions. Its application to more complex food matrices, particularly those with high lipid content, has not been fully explored. In such cases, additional sample preparation steps may be required to ensure compatibility with lipid-rich matrices. Another limitation is that the method does not systematically address matrix effects, which could impact the accuracy and precision of α-DC quantification in complex samples. Further optimization or validation across a wider range of matrices could improve the method's robustness.

In conclusion, while the OPD/MSPE method offers clear advantages in terms of time efficiency, simplicity, and cost-effectiveness, its limitations should be considered when applied to non-aqueous or highly complex food matrices. Future improvements could focus on expanding its applicability by modifying the derivatization reagent or adsorbent to enhance performance across a broader spectrum of sample types.

4. Conclusion

The study successfully developed a rapid, convenient, and sensitive method for the determination of α-DCs using an OPD/MSPE method coupled with LC-FLD. The method integrates derivatization and extraction into a single step, significantly simplifying the sample preparation process and reducing the risk of sample loss and contamination. The OPD/MSPE method demonstrated excellent linearity (R2 ≥ 0.998) and low detection limits (LODs: 0.4–3.5 nM) for six α-DCs. Method validation confirmed the high accuracy (relative recoveries: 86.2 %–105.3 %) and precision (RSDs: ≤ 13.7 %). Application of the OPD/MSPE method to various real samples, including tea beverages, coffee beverages, and freshly squeezed juices, revealed the presence of α-DCs in all tested samples. The method's efficiency in detecting α-DCs in complex matrices highlights its potential for broad applications in food safety and quality control. Overall, the developed OPD/MSPE method offers a significant improvement over traditional techniques by streamlining the derivatization and extraction processes, thus providing a valuable tool for the rapid and efficient determination of α-DCs in various complex samples. However, it is important to note that this method relies on hydrophobic interactions to adsorb the of α-DCs derivatives in aqueous sample solutions, making it primarily suitable for water-soluble matrices or samples that can be easily processed into aqueous solutions. Its applicability to more complex food matrices, such as those with high lipid content, remains underexplored. Future work could focus on modifying the derivatization reagent and the adsorbent to leverage π-π or ion exchange interactions, potentially expanding the method's utility to non-aqueous and more complex food matrices.

CRediT authorship contribution statement

Yanbo Luo: Writing – original draft, Methodology, Funding acquisition. Yuwei Liu: Writing – original draft, Methodology. Xiangyu Li: Methodology. Xingyi Jiang: Methodology. Yongqiang Pang: Writing – review & editing, Project administration. Di Chen: Writing – review & editing, Project administration, Funding acquisition, Conceptualization.

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.

Acknowledgements

The authors thank the Henan Provincial Science and Technology Research Project (No. 242102320268 and 242102311184), the Science Innovation Foundation of China National Tobacco Quality and Supervision & Test Center (No. 512018CA0050), and the National Key Laboratory of Cotton Bio-breeding and Integrated Utilization Open Fund (No. CB2023A24) for their financial support.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2024.101858.

Contributor Information

Yongqiang Pang, Email: pangyq2726@163.com.

Di Chen, Email: dichen@zzu.edu.cn.

Appendix A. Supplementary data

Supplementary material

mmc1.docx (26.8KB, docx)

Data availability

Data will be made available on request.

References

  1. Brun N., González-Sánchez J.M., Demelas C., Clément J.L., Monod A. A fast and efficient method for the analysis of α-dicarbonyl compounds in aqueous solutions: Development and application. Chemosphere. 2023;319 doi: 10.1016/j.chemosphere.2023.137977. [DOI] [PubMed] [Google Scholar]
  2. Cao M., Wang J., Wang M., Yuan X., Zhang X., Ma J.…Li Q. Simultaneous detection of 22 mycotoxins in grape by QuEChERS and ultra-high-performance liquid chromatography coupled with quadrupole/orbitrap high-resolution mass spectrometry. Journal of Future Foods. 2024;4(4):369–375. doi: 10.1016/j.jfutfo.2023.11.009. [DOI] [Google Scholar]
  3. Chen D., Wang B., Xu X.L., Bu X.M., Zhang M.Y., Xu X.…Shi N. One-pot derivatization/extraction coupled with liquid chromatography-tandem mass spectrometry for furfurals determination. Food Chemistry. 2023;428 doi: 10.1016/j.foodchem.2023.136839. [DOI] [PubMed] [Google Scholar]
  4. Chen D., Zhang J.X., Cui W.Q., Zhang J.W., Wu D.Q., Yu X.R.…Xu X. A simultaneous extraction/derivatization strategy coupled with liquid chromatography–tandem mass spectrometry for the determination of free catecholamines in biological fluids. Journal of Chromatography A. 2021;1654 doi: 10.1016/j.chroma.2021.462474. [DOI] [PubMed] [Google Scholar]
  5. Custodio-Mendoza J.A., Blanco A.L., Ares-Fuentes A.M., Díaz A.M.C. Green infant formula analysis: Optimizing headspace solid-phase microextraction of carbonyl compounds associated with lipid peroxidation using GC-MS and pentafluorophenylhydrazine derivatization. Talanta. 2024;273 doi: 10.1016/j.talanta.2024.125816. [DOI] [PubMed] [Google Scholar]
  6. Giacco F., Du X., D’Agati V.D., Milne R., Sui G., Geoffrion M., Brownlee M. Knockdown of glyoxalase 1 mimics diabetic nephropathy in nondiabetic mice. Diabetes. 2013;63(1):291–299. doi: 10.2337/db13-0316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Goedert M., Guiochon G. Sources of error in chromatographic analysis. Effect of sampling parameters on the accuracy of numerical measurements. Chromatographia. 1973;6(2):76–83. doi: 10.1007/BF02270542. [DOI] [Google Scholar]
  8. Guo X., Zheng X., Guo X., Wu J., Jing X. Determination of chiral prothioconazole and its chiral metabolite in water, juice, tea, and vinegar using emulsive liquid–liquid microextraction combined with ultra-high performance liquid chromatography. Food Chemistry. 2024;440 doi: 10.1016/j.foodchem.2023.138314. [DOI] [PubMed] [Google Scholar]
  9. He J., Wang L., Liu H., Sun B. Recent advances in molecularly imprinted polymers (MIPs) for visual recognition and inhibition of α-dicarbonyl compound-mediated Maillard reaction products. Food Chemistry. 2024;446 doi: 10.1016/j.foodchem.2024.138839. [DOI] [PubMed] [Google Scholar]
  10. Lee J., Nam T.G., Choi H.K., Jang H.W. Determination of toxic α-dicarbonyl compounds in sesame oils using dispersive liquid–liquid microextraction coupled with gas chromatography–mass spectrometry. Food Chemistry: X. 2024;22 doi: 10.1016/j.fochx.2024.101302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Paranthaman R., Moses J.A., Vidyalakshmi R. Simultaneous detection of melamine and cyanuric acid using powder X-ray diffraction. Food Safety and Health. 2024;2(2):289–302. doi: 10.1002/fsh3.12039. [DOI] [Google Scholar]
  12. Pastor-Belda M., Fernández-García A.J., Campillo N., Pérez-Cárceles M.D., Motas M., Hernández-Córdoba M., Viñas P. Glyoxal and methylglyoxal as urinary markers of diabetes. Determination using a dispersive liquid–liquid microextraction procedure combined with gas chromatography–mass spectrometry. Journal of Chromatography A. 2017;1509:43–49. doi: 10.1016/j.chroma.2017.06.041. [DOI] [PubMed] [Google Scholar]
  13. Qi C., Jin Y., Cheng S., Di L., Wang X., Zhang M.…Min J.Z. A novel UHPLC-MS/MS method for the determination of four α-dicarbonyl compounds in wine and dynamic monitoring in human urine after drinking. Food Research International. 2023;163 doi: 10.1016/j.foodres.2022.112170. [DOI] [PubMed] [Google Scholar]
  14. Rabbani N., Thornalley P.J. Dicarbonyl stress in cell and tissue dysfunction contributing to ageing and disease. Biochemical and Biophysical Research Communications. 2015;458(2):221–226. doi: 10.1016/j.bbrc.2015.01.140. [DOI] [PubMed] [Google Scholar]
  15. Rodríguez-Cáceres M.I., Palomino-Vasco M., Mora-Diez N., Acedo-Valenzuela M.I. Novel HPLC – Fluorescence methodology for the determination of methylglyoxal and glyoxal. Application to the analysis of monovarietal wines “Ribera del Guadiana”. Food Chemistry. 2015;187:159–165. doi: 10.1016/j.foodchem.2015.04.103. [DOI] [PubMed] [Google Scholar]
  16. Rodríguez-Cáceres M.I., Palomino-Vasco M., Mora-Diez N., Acedo-Valenzuela M.I. Dispersive liquid-liquid microextraction for a rapid determination of glyoxal in alcoholic beverages. Talanta. 2017;168:100–104. doi: 10.1016/j.talanta.2017.03.031. [DOI] [PubMed] [Google Scholar]
  17. Shakoor A., Zhang C., Xie J., Yang X. Maillard reaction chemistry in formation of critical intermediates and flavour compounds and their antioxidant properties. Food Chemistry. 2022;393 doi: 10.1016/j.foodchem.2022.133416. [DOI] [PubMed] [Google Scholar]
  18. Shi N., Liu Y., Li W., Yan S., Ma L., Xu X., Chen D. One-pot derivatization/magnetic solid-phase extraction coupled with liquid chromatography-fluorescence detection for the rapid determination of sulfonamide residues in honey. Food Chemistry: X. 2024;21 doi: 10.1016/j.fochx.2023.101090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Song M., Wang K., Lu H., Yan S., Wu L., Xue X. Composition and distribution of α-dicarbonyl compounds in propolis from different plant origins and extraction processing. Journal of Food Composition and Analysis. 2021;104 doi: 10.1016/j.jfca.2021.104141. [DOI] [Google Scholar]
  20. Wang XinJie W.X., Gao Feng G.F., Li LuChao L.L., Hui Xian H.X., Li Heng L.H., Gao WenYun G.W. Quantitative analyses of α-dicarbonyl compounds in food samples by HPLC using 4-(2,3-dimethyl-6-quinoxalinyl)-1,2-benzenediamine as a derivatizing reagent. Microchemical Journal. 2018;141:64–70. doi: 10.1016/j.microc.2018.05.006. [DOI] [Google Scholar]
  21. Yan S., Sun M., Zhao L., Wang K., Fang X., Wu L., Xue X. Comparison of differences of α-Dicarbonyl compounds between naturally matured and artificially heated Acacia honey: Their application to determine honey quality. Journal of Agricultural and Food Chemistry. 2019;67(46):12885–12894. doi: 10.1021/acs.jafc.9b05484. [DOI] [PubMed] [Google Scholar]
  22. Yan S., Wu L., Xue X. α-Dicarbonyl compounds in food products: Comprehensively understanding their occurrence, analysis, and control. Comprehensive Reviews in Food Science and Food Safety. 2023;22(2):1387–1417. doi: 10.1111/1541-4337.13115. [DOI] [PubMed] [Google Scholar]
  23. Yang L., Zhang W., Deng W., Wang H., Liu H. Simultaneous quantification and evaluate the differences of two skeleton components in raw, salt and wine Angelicae Pubescentis Radix by UPLC-MS/MS in negative/positive modes coupled with chemometrics. Journal of Future Foods. 2022;2(1):82–90. doi: 10.1016/j.jfutfo.2022.03.020. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material

mmc1.docx (26.8KB, docx)

Data Availability Statement

Data will be made available on request.

Articles from Food Chemistry: X are provided here courtesy of Elsevier

RESOURCES