Effects of reaction conditions on N^ε-(1-Carboxymethyl)-L-lysine formation in reducing sugar-lysine mode reaction systems

Abstract

A This study aimed to employ three reducing sugar (fructose, galactose and glucose)-lysine mode reaction systems (FLMRS, GaLMRS and GLMRS) in examining the influences of reaction conditions, such as initial pH value, reaction temperature, and time on the N^ε-(1-Carboxymethyl)-L-lysine (CML) formation. Results indicated that the initial pH value were positively correlated with the CML content in the system. However, the influence of reaction temperature and time on the CML content in the system displayed an initial increasing trend followed by a decline. A temperature of 100.58/125.24/123.24 ° C and a heating time of 30.00/24.66/25.38 min maximized the formation of CML on FLMRS/GaLMRS/GLMRS and resulted in a CML production of 2.11/1.39/1.34 mmol/mol lysine of response surface methodology. The results of this study would provide particular theoretical support for studying the formation mechanism and control methods of CML in food processing.

Introduction

Via the Maillard reaction, amino acids and reducing sugars in food can produce a group of products with stable qualities, namely advanced glycation end-products (AGES) (Singh et al., 2001). To date, more than 20 types of AGES had been detected of which N^ε-(1-Carboxymethyl)-L-lysine (CML) is the most important, as well as the first to be identified and separated from food products (Buser et al., 1987). CML can accumulate in a variety of tissues and organs after entering the body via food ingestion. After subjecting a mouse to a diet containing CML, Tessier et al. found accumulations of CML in all tissues and organs except fat, including the kidneys, intestines, lungs, heart, muscles, and liver (Tessier et al., 2016). When the accumulation of CML reaches a specific level, the functionality of tissues and organs will be directly affected, and pathological changes will occur in the body. CML is closely related to the manifestation of various diseases and can promote the development of diabetes, heart disease, atherosclerosis and several other diseases, as well as the rapid aging of human organs (Goldin et al., 2006; Baynes, 2001; Zhang et al., 2016; Begieneman et al., 2015). The CML content in food is affected by many factors, including food composition, temperature and method of preparation, as well as the time length of heat treatment. Food items such as meat products, baked foods, dairy products, and fried foods usually contain higher levels of CML (Assar et al., 2009; Chao et al., 2009). Detection results indicated that the CML content in fried foods ranged from 3.9 to 50.3 mg/kg, from 8.6 to 304.7 mg/kg in meat products, from 40.0 to 87.3 mg/kg in infant formula, and from 2.3 to 171.2 mg/kg in baked foods (Zhou, 2015).

The harmful impact of CML and its abundance in many food products are attracting increased attention. Experts and scholars both at home and abroad devote themselves to studying the formation rule of CML during food processing, and are searching for effective ways to reduce the CML content in food. Goldberg et al. found that foods of the fat group showed the highest amount of CML content with a mean of 100 ± 19 kU/g. High values were also observed for the meat and meat-substitute group, 43 ± 7 kU/g (Goldberg et al., 2004). Chen et al. studied the changes of the CML content in several meat products while being subjected to different processing methods, and found that the CML content were elevated when being subjected to high-temperature conditions involving baking and frying and could reach as much as 21.80 µg/g in beef exposed to high cooking temperatures (Chen and Smith, 2015; Trevisan et al., 2016). The formation path of CML in food processing is complicated, and the amount of CML can be affected by many factors, including cooking temperature, length of cooking time, pH value and presence of moisture (Vlassara and Uribarri, 2004; Poulsen et al., 2013). Furthermore, protein, reducing sugar, lipid, and ascorbic acid can be used as substrates to form CML. The major reactions are glycosylation of amino acids, and the oxidative degradation of reducing sugar, oil and ascorbic acid. CML can be formed simultaneously via one or more of these reactions (Fu et al., 1996; Dunn et al., 1990; Koschinsky et al., 1997). Since current studies involving the formation rule of CML in food focus primarily on the actual food, it is necessary to simplify the process conditions of food and establish a mode reaction system.

Reducing sugars in carbohydrates and the lysine in proteins are essential substrates for forming CML. Fructose, galactose and glucose are the most common reducing sugars in food. Since the lysine is a component of protein, it is present in a variety of protein-rich foods, including meat products, dairy products, and bean products. Therefore, in this experiment, reducing sugars (fructose, galactose and glucose) and lysine were selected as the substrates for the mode reaction system. Three mode reaction systems of reducing sugars (fructose, galactose, and glucose) and lysine (FLMRS, GaLMRS and GLMRS) were established to study the formation rule of CML, with reference to the influences of initial pH value, reaction temperature and time. Furthermore, the response surface methodology (RSM) was applied to optimize the CML formation conditions. The conclusions can provide a theoretical basis for further study involving the formation mechanism and control methods of CML in food processing.

Materials and Methods

Chemicals and reagents  CML standard (with a purity of 98.0 %) and CML-D₄ isotope internal standard (with a chemical purity of 98.0 % and an isotope purity of 97.9 %) were purchased from TRC (Toronto Research Chemicals, Canada). Acetonitrile (mass spectrometric purity) and formic acid (mass spectrometric purity) were purchased from Thermo Fisher Science (Waltham, Massachusetts, USA). D-(-)-fructose (purity ≥ 99.0 %), D-(+)-glucose (purity ≥ 99.5 %), D-(+)-galactose (purity ≥ 99.0 %), and L-lysine (purity ≥ 98.0 %) were purchased from Sigma-Aldrich (St. Louis, USA). The ultrapure water used in the experiment was prepared with the Milli-Q Advantage A10 water system (Millipore, Bedford, USA). The remainder were all analytical reagents.

Preparation of CML standard solution and substrate solution in the reaction system  The working standard solutions for linear calibration were prepared by diluting the stock solution (100.0 ng/mL) to a concentration sequence of 1.0, 2.0, 5.0, 10.0, 20.0, and 50 ng/mL spiked with the CML-D₄ isotope internal standard (10.0ng/mL). Phosphate buffer with a pH of 4.0 was prepared with H₃PO₄ solution (0.2 mol/L) and NH₂PO₄ solution (0.2 mol/L). Phosphate buffers with a pH of 6.0, 7.0 and 8.0, respectively, were prepared with NaH₂PO₄ solution (0.2 mol/L) and Na₂HPO₄ solution (0.2 mol/L), while phosphate buffers with a pH of 10.0 and 12.0, respectively were prepared with Na₃PO₄ solution (0.2 mol/L) and Na₂HPO₄ solution (0.2 mol/L). The separate fructose, galactose, glucose and lysine solutions, each with a concentration of 0.6 mol/L were prepared using the phosphate buffer mentioned above at individual pH levels of 4.0, 6.0, 7.0, 8.0, 10.0, and 12.0. All these solutions were stored at 4 °C.

Establishment of the reducing sugar-lysine mode reaction system  The reducing sugar (fructose or galactose or glucose) and lysine solutions at a concentration of 0.6 mol/L prepared with specific pH phosphate buffer (0.2 mol/L) and the 0.25 mL volume of each solution, were accurately transferred to a reaction tube (outer diameter × length = 18mm × 180mm) and mixed thoroughly using a vortex mixer. The sealed reaction tube was placed in a Type 101 Oven (Yongguangming Medical instrument factory, Beijing, China), which was heated at an exact temperature for a specific time. Following completion of the reaction, the test tube was retrieved from the oven and cooled immediately in a refrigerator at 4.0 °C. It was then stored at −20.0 °C after cooling. All trials were carried out in triplicate.

Pretreatment of end products of the reaction  The final reaction product was transferred to a 25.0 mL volumetric flask, and diluted with water. A 40.0 µL volume of this sample solution and 50.0 µL of CML-D₄ isotopic internal standard solution (400.0 ng/mL) were diluted with 5.0% of the aqueous formic acid solution to 2.0 mL. A mixed cation-exchange (MCX) solid phase extraction (SPE) cartridge (3.0 mL, 60.0 mg, 30.0 um, Waters, Milford, MA, USA), was activated with 3.0 mL methanol and balanced with 3.0 mL water in the VacElut SPS 24 vacuum SPE unit (Agilent Technologies, Santa Clara, USA). A 1.0 mL volume of the composite sample solution was passed through the MCX cartridge at a rate of 1–2 drops/s. After the sample solution was discharged completely, the MCX cartridge was leached with 3.0 mL of 5.0 % aqueous formic acid solution and dried with a pumping vacuum. Then, the MCX cartridge was leached with 3.0 mL of methanol, dried with a pumping vacuum, and all effluent solutions were discarded. Finally, the MCX cartridge was eluted with 5.0 mL of 15.0 % ammonia methanol solution and dried with a pumping vacuum. The effluent solution was collected in a test tube and blow-dried with 40°C nitrogen using the N-EVAP-24 nitrogen-blowing instrument (Organomation Associates Inc, Berlin, USA). The residue was dissolved with 1.0 mL water. The final solution was filtered into a 2.0 mL sample vial with a 0.22 µm polyethersulfone needle filter, and the CML content in the reaction system was determined using Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry (UPLC-MS/MS).

UPLC-MS/MS analysis  An Ultra-Performance Liquid Chromatograph (Waters, Milford, MA, USA) was used for liquid chromatography analysis. The column was UPLC BEH Amide Hydrophilic Interaction Chromatography (HILIC) (2.1 mm x 100 mm, 1.7 µm, Waters, Milford, MA, USA). Mobile phase A and B consisted of 0.1 % formic acid solution and acetonitrile, respectively. The gradient elution parameters are shown as follows: 0–2.0 min 10 % A, 2.0–6.0 min 20 % A, 6.0–6.1 min 90 % A, and 6.1–7.0 min 10 % A. The column temperature was 35 °C, and the injection volume was 5 µL, with a running time of 7.0 min.

For this study, the XEVO TQ Triple Quadrupole Tandem Mass Spectrometer(Waters, Milford, MA, USA) was selected with the ion source in electrospray positive ion mode (ESI⁺), and application of multiple reaction monitoring (MRM). The voltage of the capillary was 3.5 kV, the voltage of the cone hole was 20 V, and the temperature of the source was 150 °C. Desolvation gas temperature was 400 °C, the desolvation flow rate was 700 L/h, and the collision energy was 15V. The setting scheme of the MRM mode was as follows: The mass charge ratio (m/z) of CML parents ion was 205.22, the ion with m/z of 84.00 was set as the quantitative ions of CML, the ion with m/z of 130.00 was set as the qualitative ions of CML, and the ions with m/z of 87.70 was set as the qualitative ions of CML-D₄.

Drawing of the standard curve  The concentrations of the CML standard solutions were 1.0, 2.0, 5.0, 10.0, 20.0 and 50.0 ng/mL, respectively, while the concentration of CML-D₄ internal standard solution in each case was 10.0 ng/mL. The CML content in these standard solutions was determined with UPLC-MS/MS. Taking the response value Y (CML peak area/CML-D₄ peak area) and the concentration ratio of CML and CML-D₄ X as the vertical and horizontal axes, respectively, and the regression equation and correlation coefficient were as follows: Y = 2.435 9*X-1.183 1, and R2 = 0.9989. The internal standard curve of CML is shown in Fig.1. The linear range of CML was 1.0–50.0 ng/mL. It was necessary for the sample solution to be diluted appropriately before sample analysis was conducted to ensure that the concentration remained within the linear range of the standard curve in the case of high CML content.

Fig. 1.

The internal standard curve of CML.

Statistical data analysis methods  Three parallel tests were conducted for each sample. The test data were expressed as “mean ± standard deviation”. Dunnett's T3 method of ANOVA in SPSS Statistics 17.0 was used to test the significance of the test data.

Results and Discussion

Effects of the initial pH value on CML formation  The effects of different initial pH values on the CML formation in the reducing sugar-lysine reaction system are shown in Fig.2A. When the substrate concentration is 0.3mol/L, the reaction temperature is 140.0 °C and the reaction time is 20.0 min, and the initial pH value increases from 4.0 to 12.0. From Fig.2A it is apparent that the CML content rises in conjunction with the increase of initial pH values in the FLMRS. The CML content is 1.24 ± 0.02 mmol/mol lysine when the initial pH value is 4.0, and the CML content is 1.91 ± 0.03 mmol/mol lysine when the initial pH value increases to 12.0. Alkaline conditions are more conducive to CML formation than an acidic environment. The variational rules of the CML content in the GaLMRS and GLMRS are similar to those in the FLMRS. The reason for this trend is that the content of open-loop reducing sugars with reactive activity rises with the increase of the pH values (Yaylayan et al., 1993), which can also affect the reaction activity of lysine. Under aerobic conditions, the reducing sugar undergoes an automatic oxidative degradation reaction to form a dicarbonyl compound GO, which as an intermediate reacts with lysine residues to generate CML. This process is called Wolff Pathway (Wolff and Dean,1987). The foresaid oxidative degradation of the reducing sugar and other degradation reactions, such as enolization and dehydration, are all catalyzed by alkali in the Maillard reaction (Ajandouz et al., 2001; Shao et al., 2012; Sun et al., 2015), increasing the activity of the reaction groups, as well as the CML content. Fig.2A indicates the presence of some differences in the CML content of the three reducing sugar reaction systems. Under the same conditions, the order of the CML content in the three systems is FLMRS > GaLMRS > GLMRS. Since fructose molecules contain free ketones, and free aldehydes are present in both glucose and galactose molecules, these reducing sugars are all able to form CML via Maillard reaction involving the lysine in the solution. Sakai et al. pointed out that there was stronger glycosylation ability in fructose than in glucose (Sakai et al., 2002; Hinton and Ames, 2006). Nguyen et al. studied the formation rules of CML in the casein-glucose and casein-lactose reaction systems and found that the CML content in the lactose-casein solution was higher than in the glucose-casein solution at temperatures of 120 °C and 130 °C, respectively (Nguyen et al., 2016). Furthermore, Oh et al. discovered that the CML content in the casein-galactose system exceeded that in the casein-glucose system (Oh et al., 2018). The results of this experiment conform to those of previous studies.

Fig. 2.

Effects of reaction conditions on CML formation in reducing sugar-lysine mode reaction systems. (A) Initial pH value, (B) Reaction temperature, and (C) Reaction time. Fructose-lysine mode reaction system (FLMRS), Galactose-lysine mode reaction system (GaLMRS), and Glucose-lysine mode reaction system (GLMRS).

Effects of the reaction temperature on CML formation  The effects of different reaction temperatures on the CML formation in the reducing sugar-lysine reaction system are shown in Fig.2B. When the substrate concentration is 0.3 mol/L, with an initial pH value of 7.0 and the reaction time is 20.0 min, the reaction temperature increases from 80.0 °C to 180.0 °C. From Fig.2B it is evident that the CML content displays an initial rise and then decreases in conjunction with higher reaction temperature in the FLMRS. The CML content is 0.52 ± 0.02 mmol/mol lysine when the reaction temperature is 80.0 °C, and reaches the maximum value of 1.76 ± 0.03 mmol/mol lysine when the reaction temperature increases to 120.0 °C. The CML content exhibits a downward trend when the reaction temperature increases continuously. The variational rules of the CML content in the GaLMRS and GLMRS are similar to those in the FLMRS, reaching its maximum values at 140.0 °C and 120.0 °C respectively. The effects of the reaction temperature on the CML formation in the reducing sugar-lysine reaction system are more complicated. By studying the CML formation rules in beef, Sun et al. found higher the CML content in beef when the heating temperature increased between 65 °C-100 °C (Sun et al., 2015). When the reaction temperature continuously increases at a high temperature, the CML will degrade, adduct with lysine and transform into melanoidins. Therefore, the CML content will exhibit a decline when the reaction temperature in the FLMRS is higher than 120.0 °C. The results of this experiment conform to those of previous studies.

Effects of the reaction time on CML formation  The effects of different reaction times on the CML formation in the reducing sugar-lysine reaction system are shown in Fig. 2C. When the substrate concentration is 0.3 mol/L, with an initial pH value of 7.0, and the reaction temperature is 140.0 °C, the reaction time increases from 5.0 to 30.0 min. From Fig.2C, it is apparent that the CML content increases with prolonged reaction time in the FLMRS. The CML content is 0.62 ± 0.01 mmol/mol lysine when the reaction time is 5.0 min and increases to a maximum value of 1.66 ± 0.03 mmol/mol lysine, when the reaction time is extended to 15.0 min. The CML content exhibits a downward trend with a continuous extension of the reaction time. The variational rules of CML content in the GaLMRS and GLMRS are similar to those in the FLMRS, both reaching maximum values at 25.0 min. The formation and elimination of CML occur simultaneously in the reducing sugar-lysine mode reaction system. The elimination rate of CML will increase gradually from low to high during the extended heating process of the reaction system at high temperature. When the CML elimination rate exceeds the formation rate, the CML content in the reaction system will display a downward trend.

Optimization of the formation conditions of CML using RSM  Based on the single factor test results of the initial pH value, reaction temperature and time, the formation conditions of CML were optimized by the Central Composite Design (CCD) of the RSM. The two-factor and five-level optimization design was conducted by selecting two factors of reaction temperature (°C) and time (min) and applying the Design-Expert V 8.0.6.1 (Delaware, USA) software. The design scheme of the factor level is shown in Table 1, With the reaction temperature (°C) and reaction time (min) as the response variables, and the CML content as the response value, the regression analysis was conducted on data derived from the two-factor and five-level tests of every reaction system in Table 2. The fitting equations of quadratic polynomial regression models of the three reaction systems are as follows:

  

Where, Y is the CML content in the system, mmol/mol lysine; A is the reaction temperature in the system, °C; and B is the reaction time in the system, min.

Table 1 The design scheme of factors and levels in RSM.

	Factor
Level	A		B
	Reaction temperature (°C)		Reaction time (min)
−2	80.0		10.0
−1	100.0		15.0
0	120.0		20.0
1	140.0		25.0
2	160.0		30.0

Table 2 Design of RSM and CML content in different reducing sugar-lysine simulated reaction systems.

Numerical order	A Reaction temperature (°C)	B Reaction time (min)	Y CML content(mmol/(mol lysine))
			FLMRS	GaLMRS	GLMRS
1	80.0(-2)	20.0(0)	0.52	0.48	0.31
2	100.0(-1)	15.0(-1)	0.73	0.60	0.54
3	100.0	25.0(1)	1.77	1.19	1.14
4	120.0(0)	10.0(-2)	0.64	0.40	0.33
5	120.0	20.0	1.70	1.20	1.14
6	120.0	20.0	1.77	1.28	1.16
7	120.0	20.0	1.69	1.24	1.22
8	120.0	20.0	1.76	1.29	1.28
9	120.0	20.0	1.82	1.26	1.24
10	120.0	30.0(2)	2.00	1.34	1.32
11	140.0(1)	15.0	1.66	1.26	1.07
12	140.0	25.0	1.52	1.33	1.19
13	160.0(2)	20.0	1.35	1.15	1.05

The variance analysis results in Table 3, indicate that the p values of the regression models of the FLMRS, GaLMRS and GLMRS are all less than 0.0001. Three quadratic equation models are all extremely significant models. While the p values of lacks of fit are 0.0864, 0.0629 and 0.3121, respectively, they are not significant. The correction determination factors R² of the three reaction systems are 0.9826, 0.9805 and 0.9806 respectively. The relatively high level of R² indicates higher reliability of the test design method. Therefore, the established regression model can better analyze and predict the formation rules of CML in the reducing sugar-lysine mode reaction system.

Table 3 variance analysis of RSM model variables in different reducing sugar-lysine simulated reaction systems.

Reaction System	Source of variances	Sum of squares	Degree of freedom	Mean square	F Value	P Value
FLMAS	Model	2.91	5	0.58	79.23	< 0.0001**
	A	0.46	1	0.46	62.18	< 0.0001**
	B	1.09	1	1.09	148.82	< 0.0001**
	AB	0.35	1	0.35	47.44	0.0002**
	A2	0.95	1	0.95	129.39	< 0.0001**
	B2	0.26	1	0.26	35.96	0.0005**
	Residual	0.051	7	0.0073
	Lack of fit	0.040	3	0.013	4.63	0.0864
	Pure error	0.011	4	0.0029
	Cor total	2.96	12
GaLMAS	Model	1.36	5	0.27	70.52	< 0.0001**
	A	0.38	1	0.38	98.82	< 0.0001**
	B	0.54	1	0.54	139.22	< 0.0001**
	AB	0.068	1	0.068	17.50	0.0041**
	A2	0.27	1	0.27	70.42	< 0.0001**
	B2	0.21	1	0.21	53.77	0.0002**
	Residual	0.027	7	0.0039
	Lack of fit	0.022	3	0.0073	5.71	0.0629
	Pure error	0.0051	4	0.0013
	Cor total	1.39	12
GLMAS	Model	1.50	5	0.30	70.66	< 0.0001**
	A	0.35	1	0.35	83.18	< 0.0001**
	B	0.61	1	0.61	142.90	< 0.0001**
	AB	0.058	1	0.058	13.55	0.0079**
	A2	0.40	1	0.40	93.80	< 0.0001**
	B2	0.21	1	0.21	49.33	0.0002**
	Residual	0.030	7	0.0043
	Lack of fit	0.016	3	0.0055	1.65	0.3121
	Pure error	0.013	4	0.0033
	Cor total	1.53	12

A (Reaction temperature, °C), B (Reaction time, min), ** (p < 0.01).

Table 3 signifies that the p values of the linear and quadratic terms of the reaction temperature, the linear and quadratic terms of the reaction time, as well as the interaction terms of the reaction temperature and time are all less than 0.01. Therefore, these factors reach a significant level, indicating the absence of a simple linear relationship between the single factor of response variables and the response value (the CML content) in the three reaction systems. Based on variance analysis of the regression model in the reaction system, the response surface and contour analysis of the interaction of response variables (reaction temperature and time) on the response value (the CML content) in the reducing sugar-lysine mode reaction system are shown in Fig.3. The shape of contour reflects the degree of interaction, and the interaction between response variables is not significant when the contour is circular. However, the interaction between the response variables is significant when the contour is elliptical (Zeng et al., 2009).

Fig. 3.

Response surface curve of the CML content in three reducing sugar-lysine reaction systems. (A) FLMRS, (B) GaLMRS and (C) GLMRS.

From Fig. 3, it is apparent that the contours of the interaction between reaction temperature and time in the FLMRS, GaLMRS and GLMRS are all elliptical, indicating that there is a significant interaction between the response variables. As for the FLMRS, it was calculated that the CML content reached its maximum of 2.11 mmol/mol lysine when the reaction temperature was 100.58 °C with a reaction time of 30.00 min. In the GaLMRS, it was calculated that the CML content reached its maximum of 1.39 mmol/mol lysine when the reaction temperature was 125.24 °C with a reaction time of 24.66 min. Similarly, in the GLMRS, it was calculated that the CML content reached its maximum of 1.34 mmol/(mollysine) when the reaction temperature was 123.24 °C with a reaction time of 25.38 min. The calculation values of the reaction temperature and time for the maximum CML content in the FLMRS, GaLMRS and GLMRS were modified to values which could be easily operated. Under these reaction conditions, the reliability of the predicted values of the regression model was verified and the results are shown in Table 4. The relative standard deviations (RSD) between the predicted values and experimental values in three mode reaction systems were all less than 5.0 %, indicating that the predicted values of the regression model were accurate and reliable.

Table 4 The comparison table of predicted values and experiment values in regression model.

Reaction System	Reaction temperature (°C)	Reaction time (min)	Predicted value (mmol/(mol lysine))	Experiment value (mmol/(mol lysine))	RSD (%)
FLMRS	100.0	30.0	2.11	1.98	4.49
GaLMRS	125.0	25.0	1.39	1.35	2.04
GLMRS	125.0	25.0	1.34	1.29	2.73

Conclusion

This study investigated the effects of the initial pH value, reaction temperature and time on the CML formation in the reducing sugar-lysine mode reaction system. The results showed that the initial pH value positively correlated with the CML content in the system, while a complicated relationship existed between the reaction temperature and time with the CML content of the system, displaying an initial increasing trend followed by a decline. By comparing FLMRS, GaLMRS and GLMRS, it was found that the CML content in the FLMRS was highest, followed by GaLMRS and GLMRS. With a substrate concentration of 0.3 mol/L, an initial pH value of 7.0, and applying RSM for optimizing the reaction conditions of the system, the correction decision coefficient R² of the three reaction system regression models were all above 0.98. Therefore, the highest CML content of calculated value in the FLMRS, GaLMRS and GLMRS was obtained under a reaction temperature of 100.58 °C, 125.24 °C and 123.24 °C, respectively. This was accomplished at individual reaction times of 30.00, 24.66 and 25.38 min, while the highest CML content was 2.11, 1.39 and 1.34 mmol/mol lysine, respectively. The results of this experiment provide certain theoretical support for studying the formation mechanism and control methods of CML in food processing.

Conflict of interest  There are no conflicts of interest to declare.

Ethical approval  This article does not contain any studies with human participants or animals performed by any of the authors.

Informed Consent  Not applicable.

References

•  Ajandouz,  E. H.,  Tchiakpe,  L. S.,  Ore,  F. D.,  Benajiba,  A., and  Puigserver,  A. (2001). Effects of pH on caramelization and Maillard reaction kinetics in fructose-lysine model systems. J. Food Sci., 66, 926–931.
•  Assar,  S., H.,  Moloney,  C.,  Lima,  M.,  Magee,  R., and  Ames  JM. (2009). Determination of n ε-(carboxymethyl)lysine in food systems by ultra performance liquid chromatographymass spectrometry. Amino Acids, 36, 317–326.
•  Baynes,  J.W. (2001). The role of AGES in aging: causation or correlation. Exp. Gerontol., 36, 1527–1537.
•  Begieneman,  M. P. V.,  Rijvers,  L.,  Kubat,  B.,  Paulus,  W. J.,  Vonk,  A. B. A., and  Van Rossum,  A.C. (2015). Atrial fibrillation coincides with the advanced glycation end product N^ε-(carboxymethyl)lysine in the atrium. Am. J. Pathol., 185, 2096–2104.
•  Büser,  W.,  Erbersdobler,  H.F. and  Liardon,  R. (1987). Identification and determination of N^ε-carboxymethyllysine by gas-liquid chromatography. J. Chromatogr. A, 387, 515–519.
•  Chao,  P. C.,  Hsu,  C. C. and  Yin,  M. C. (2009). Analysis of glycative products in sauces and sauce-treated foods. Food Chem., 113, 262–266.
•  Chen,  G. and  Smith,  J. S. (2015). Determination of advanced glycation endproducts in cooked meat products. Food Chem., 168, 190–195.
•  Dunn,  J. A.,  Ahmed,  M. U.,  Murtiashaw,  M. H.,  Richardson,  J. M.,  MD Walla, and  Thorpe,  S. R. (1990). Reaction of ascorbate with lysine and protein under autoxidizing conditions: formation of Nepsilon-(carboxymethyl)lysine by reaction between lysine and products of autoxidation of ascorbate. Biochemistry, 29, 10964–10970.
•  Fu,  M.,  Requena,  J.,  Jenkins,  A.,  Lyons,  T.,  Baynes,  J., and  Thorpe  S. (1996). The advanced glycation end product, N^ε-(carboxymethyl)lysine, is a product of both lipid peroxidation and glycoxidation reactions. J Biol. Chem., 271, 9982–9986.
•  Goldberg,  T.,  Cai,  W.,  Peppa,  M.,  Dardaine,  V.,  Baliga,  B. S. and  Uribarri,  J. (2004). Advanced glycoxidation end products in commonly consumed foods. J. Am. Diet. Assoc., 104, 1287–1291.
•  Goldin,  A.,  Beckman,  J. A.,  Schmidt,  A. M. and  Creager,  M. A. (2006). Advanced glycation end products sparking the development of diabetic vascular injury. Circulation. 114, 597–605.
•  Hinton,  D. and  Ames,  J. M. (2006). Site specificity of glycation and carboxymethylation of bovine serum albumin by fructose. Amino Acids, 30, 425–434.
•  Hull,  G.L.J.,  Woodside,  J.V.,  Ames,  J.M. and  Cuskelly,  G.J. (2012). N^ε-(carboxymethyl)lysine content of foods commonly consumed in a western style diet. Food Chem., 131, 170–174.
•  Koschinsky,  T.,  He,  C.J.,  Mitsuhashi,  T.,  Bucala,  R.,  Liu,  C. and  Buenting,  C. (1997). Orally absorbed reactive glycation products (glycotoxins): an environmental risk factor in. P. Natl. Acad. Sci. Usa., 94, 6474–6479.
•  Nguyen,  H. T.,  Van,  D. and  Boekel,  M. V. (2016). Kinetics of N^ε-(carboxymethyl)lysine formation in aqueous model systems of sugars and casein. Food Chem., 192, 125–133.
•  Oh,  M.J.,  Kim,  Y.,  Hoon Lee,  S.,  Lee,  K.W., and  Park,  H. Y. (2018). Prediction of CML contents in the Maillard reaction products for casein-monosaccharides model. Food Chem., 267, 271–276.
•  Poulsen,  M.W.,  Andersen,  J.M.,  Bugel,  S. and  Skibsted,  L.H. (2013). Advanced glycation endproducts in food and their effects on health. Food Chem. Toxicol., 60, 10–37.
•  Sakai,  M.,  Oimomi,  M. and  Kasuga,  M. (2002). Experimental studies on the role of fructose in the development of diabetic complications. Kobe J. Med. Sci., 48, 125–136.
•  Shao,  L.Y.,  Zhou,  J.W. and  Liu,  D.H. (2012). Research status on mechanism and kinetic modelling of Maillard reaction. J. Chin. Insti. Food Sci. and Technol., 12, 103–112.
•  Singh,  R.,  Barden,  A.,  Mori,  T. and  Beilin  L. (2001). Advanced glycation end-products: a review. Diabetologia, 44, 129–146.
•  Sun,  X.H.,  Tang,  J.M.,  Wang,  T.J.,  Rasco,  B.,  Lai,  K.Q. and  Huang,  Y.Q. (2015). Formation of advanced glycation endproducts in ground beef under paste urination conditions. Food Chem., 172, 802–807.
•  Tessier,  F.J.,  Niquet-Leridon,  C.,  Jacolot,  P.,  Jouquand,  C.,  Genin,  M. and  Schmidt,  A.M. (2016). Quantitative assessment of organ distribution of dietary protein-bound ¹³C-labeled N^ε-carboxymethyllysine after a chronic oral exposure in mice. Mol. Nutr. Food Res., 60, 2446–2456.
•  Trevisan,  A.J.B.,  Lima,  D. D. A.,  Sampaio,  G.R.,  Manolio Soares,  R.A. and  Bastos,  D.M. (2016). Influence of home cooking conditions on Maillard reaction products in beef. Food chem., 196, 161–169.
•  Vlassara,  H. and  Uribarri,  J. (2004). Glycoxidation and diabetic complications: modern lessons and a warning?. Rev. Endocr. Metab. Dis., 5, 181–188.
•  Wolff,  S. P. and  Dean,  R. T. (1987). Glucose autoxidation and protein modification. The potential role of autoxidative glycosylation in diabetes[J]. Biochem. J., 1, 243–250.
•  Yaylayan,  V. A.,  Ismail,  A.A. and  Mandeville,  S. (1993). Quantitative determination of the effect of pH and temperature on the keto form of D-fructose by FT-IR spectroscopy. Carbohyd. Res., 248, 355–360.
•  Zeng,  L.,  Song,  X.Y.,  Sheng,  Z.H.,  Wu,  Q.F. and  Zhu,  M. (2009). The microwave extraction process of polysaccharide from ganoderma lucidum was optimized by response surface analysis. Chin. J. Traditi. Chin. Medic., 27, 631–633.
•  Zhang,  J.H.,  Xu,  H.Z.,  Shen,  Q.F.,  Sun,  C.K. and  Sha,  L. (2016). N^ε-(carboxymethyl)-lysine, White Matter, and Cognitive Function in Diabetes Patients. Can. J. Neurol. Sci., 43, 518–522.
•  Zhou,  Y.Q. (2015). Study on the inhibition effects and mechanism of plant polyphenol on advanced glycation end-products in food. Hangzhou: Zhejiang University (Chapter 2).