Abstract
The dicarbonyl compound methylglyoxal (MGO) is a major precursor in the formation of advanced glycation endproducts (AGEs). MGO and AGEs are increased in subjects with diabetes and are associated with fatal and nonfatal cardiovascular disease. Previously, we have shown that plasma MGO concentrations rapidly increase in the postprandial phase, with a higher increase in individuals with type 2 diabetes. In current study, we investigated whether postprandial MGO formation in plasma and tissues originates from exogenous glucose and whether the increased plasma MGO concentration leads to a fast formation of MGO-derived AGEs.
We performed a stable isotope-labelled oral glucose tolerance test (OGTT) in 12 healthy males with universally labelled D(+)13C glucose. Analysis of plasma-labelled 13C3 MGO and glucose levels at 11 time-points during the OGTT revealed that the newly formed MGO during OGTT is completely derived from exogenous glucose. Moreover, a fast formation of protein-bound MGO-derived AGEs during the OGTT was observed. In accordance, ex-vivo incubation of MGO with plasma or albumin showed a rapid decrease in MGO and a fast increase in MGO-derived AGEs. In an intraperitoneal glucose tolerance test in C57BL/6J mice, we confirmed that the formation of postprandial MGO is derived from exogenous glucose in plasma and also showed in tissues that MGO is increased and this is also from exogenous glucose.
Collectively, increased formation of MGO during a glucose tolerance test arises from exogenous glucose both in plasma and in tissues, and this leads to a fast formation of MGO-derived AGEs.
Introduction
Methylglyoxal (MGO) is a highly reactive dicarbonyl compound and spontaneously modifies amino groups in protein. MGO is a major precursor in the formation of advanced glycation endproducts (AGEs) [1] such as Nδ-(5-hydro-5-methyl-4-imidazolon-2-yl)-ornithine (MG-H1) and Nε-(carboxyethyl)lysine (CEL) [2–4]. MGO and MGO-derived AGEs have detrimental effects on cellular function [5–7], inflammation, and cell death [8–10], are increased in subjects with type 1 and 2 diabetes and are associated with fatal and nonfatal cardiovascular disease [8,11–15]. We have recently shown that plasma MGO concentrations rapidly increased both during an oral glucose tolerance test (OGTT) and a mixed meal test, with a higher increase in individuals with type 2 diabetes [16,17]. Since MGO is rapidly formed postprandially, repeated episodes of elevated MGO plasma concentration may lead to MGO stress, and may contribute to the detrimental effects of postprandial glucose spikes [18].
MGO is mainly formed from glycolysis-derived triose phosphates [19,20], but minor sources are degradation of glycated proteins, oxidation of acetone, and lipid peroxidation [13]. The exact source of the postprandial MGO in plasma is unknown. Therefore, we investigated in-vivo whether postprandial MGO directly originates from exogenous glucose during an OGTT in healthy humans, using universally labelled D(+)13C glucose, and in mice whether MGO increased in tissues after a GTT. Furthermore, we investigated in-vitro whether the MGO leads to a rapid formation of plasma MGO-derived AGEs.
Materials and methods
Human study
A total of 12 healthy males, with an average age of 25 years (range 21–30 years) and average BMI of 22.5 kg/m2 (range: 19.2–24.7 kg/m2) were recruited. No medication use was reported by the subjects. Approval was obtained from the Medical Ethics Committee Brabant (Tilburg, The Netherlands) on 1 December 2015. Each subject provided written informed consent for the study. This trial was registered at controlled-trials.com as ISRCTN42106325.
Experimental design
After an overnight fasting period, study participants underwent a stable isotope-labelled OGTT with 50 g glucose of which 2% was universally labelled D(+)13C glucose, enabling us to track the occurrence of labelled 13C3 MGO, which we refer to as formation of MGO, in the obtained samples. Blood samples were collected at 11 time-points after the intake of glucose: every 15 min for first 2 h and every half an hour for the next 4 h. Concentrations of MGO and glucose in plasma during OGTT (6 h) were measured at 11 time-points with ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS).
Measurements of MGO, free and protein-bound MGO-derived AGEs, and plasma glucose
MGO and MGO-derived AGEs in plasma were measured with UPLC-MS/MS (Waters, Milford, U.S.A.), as described in detail previously [21,22]. Plasma glucose concentrations were measured on a Roche/Hitachi Modular automatic analyser (Roche Diagnostics, Hitachi) by using a glucose hexokinase method. 13C3 MGO and 13C6 glucose data were corrected for a difference in mass spectrometry response factor between 13C3 MGO and 12C3 MGO, and between 13C6 glucose and 12C6 glucose, respectively.
It was assumed that labelled and unlabelled glucose molecules showed identical behaviour. Based on the 2% dose of the total of 50 g glucose, the appearance of total 13C6 glucose and 13C3 MGO was corrected for dose.
Ex-vivo experiments
Bovine serum albumin (BSA, 50 g/l in PBS) and human plasma from healthy donors was incubated at 37°C for 0, 1, 2, 3, 4, 5, 6, 8, and 24 h with 0, 1, 10, and 100 µmol/l MGO. MGO was produced as previously described [23]. Every sample was snap frozen in liquid nitrogen and stored at −80°C until analysis. Protein-bound MG-H1 and CEL levels and MGO concentrations were measured with UPLC-MS/MS (Waters, Milford, U.S.A.) [21,22].
Mouse experiments
Four-week-old male C57BL/6J mice were purchased from Charles River Europe and were maintained in the animal facility until the age of 10–12 weeks, with four mice per cage. Mice were fasted overnight for 16 h and received an intraperitoneal glucose tolerance test (IPGTT) with a solution of universally labelled D(+) 13C glucose (Sigma-Aldrich), 2 g/kg body weight dissolved in saline. These mice were euthanised by means of CO2/O2 mixture at either 30, 60, or 120 min after the glucose injection, respectively. The control group of mice received no treatment and were euthanised directly, at time 0 min of the IPGTT. Blood samples were collected from the heart in EDTA tubes and stored immediately on ice. Plasma samples were obtained after spinning the blood at 2000 g, 4°C for 5 min, and were snap frozen and stored at −80°C until analysis. Pancreas, spleen, liver, kidney, visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), and skeletal muscle were collected, snap frozen and stored at −80°C. Glucose levels were measured in whole blood using a glucometer (Contour, Bayer, Leverkusen, Germany).
The experimental work protocol was approved by the local Animal Experiments Committee of Maastricht University, project license number AVD1070020187086. Experiments were performed at the Animal Laboratory of Internal Medicine, Maastricht University, by licensed people according to institutional guidelines.
Tissue preparations for measurements
Pancreas, spleen, liver, kidney, and skeletal muscle were homogenised by crushing in liquid nitrogen and dissolved in lysate buffer (0.1 M sodium phosphate buffer supplemented with 0.02% Triton-x (Sigma-Aldrich) and protease inhibitor (Roche)). VAT and SAT were homogenised with a Mini-bead beater homogeniser (Biospec), in the same lysate buffer as used for the other tissues. Concentrations of 13C3 MGO and 12C3 MGO were measured in homogenates of all tissue and plasma samples using UPLC-MS/MS, as described previously [21].
Statistical analysis
All data are presented as mean ± SEM. Statistical analysis were performed using GraphPad Prism 8.0.2. Two-way ANOVA with Bonferroni’s multiple comparisons test was used to compare groups over time. One-way ANOVA with Tukey’s multiple comparisons test was used to compare MGO and glucose levels between different time points. All analysis were considered statistically significant with P-values <0.05.
Results
Postprandial MGO formation in plasma during an OGTT is derived from exogenous glucose in humans
To investigate whether postprandial MGO formation directly originates from exogenous glucose, we performed an OGTT with 50 g glucose of which 2% was universally labelled D(+)13C glucose. Plasma 12C6 glucose levels increased rapidly after the glucose load with a calculated peak at 41 min, after which levels started to decline. The plasma 13C6 glucose concentration showed a similar peak value as for 12C6 glucose. However, the decline of 13C6 glucose levels after the peak was significantly slower than for 12C6 glucose (Table 1 and Figure 1A). After 180 min postload, plasma 12C6 glucose, but not 13C6 glucose, slightly increased (Figure 1A). The curves of plasma 12C3 MGO and 13C3 MGO followed the same pattern as observed for 12C6 glucose and 13C6 glucose, respectively. Plasma 12C3 MGO and 13C3 MGO showed a rapid increase with a calculated Cmax at 49 and 56 min, respectively (Table 1 and Figure 1B). After 180 min, 12C3 MGO slightly increased (Figure 1B). Collectively, these data show that the rapid increase in MGO formation is completely derived from exogenous glucose during the 6-h OGTT.
Plasma glucose and MGO concentrations during a stable isotope-labelled OGTT with universally labelled D(+)13C glucose in humans
(A) Plasma glucose levels, and (B) plasma MGO concentrations were analysed by UPLC-MS/MS. 13C6 glucose and 13C3 MGO data were corrected for dose (*50) and for a difference in mass spectrometry response factor between 13C6 glucose and 12C6 glucose, 13C3 MGO and 12C3 MGO, respectively. 12C6 Glucose and 12C3 MGO levels were corrected for baseline. Data are shown as mean ± SEM, n=12. Comparisons of 13C6 glucose and 12C6 glucose, 13C3 MGO and 12C3 MGO over time were tested using two-way ANOVA with Tukey’s multiple comparison. * represents comparisons with glucose or MGO at the same time point. * indicates P<0.05, ** indicates P<0.01, *** indicates P<0.001. Abbreviation: MGO, methylglyoxal.
(A) Plasma glucose levels, and (B) plasma MGO concentrations were analysed by UPLC-MS/MS. 13C6 glucose and 13C3 MGO data were corrected for dose (*50) and for a difference in mass spectrometry response factor between 13C6 glucose and 12C6 glucose, 13C3 MGO and 12C3 MGO, respectively. 12C6 Glucose and 12C3 MGO levels were corrected for baseline. Data are shown as mean ± SEM, n=12. Comparisons of 13C6 glucose and 12C6 glucose, 13C3 MGO and 12C3 MGO over time were tested using two-way ANOVA with Tukey’s multiple comparison. * represents comparisons with glucose or MGO at the same time point. * indicates P<0.05, ** indicates P<0.01, *** indicates P<0.001. Abbreviation: MGO, methylglyoxal.
Parameter . | Cmax* (mean ± SEM, unit) . | Tmax (mean ± SEM, unit) . |
---|---|---|
12C6 glucose | 328114 ± 40611 (peak area MS) | 41 ± 4.1 (min) |
13C6 glucose | 298345 ± 17997 (peak area MS) | 50 ± 4.3 (min) |
12C3 MGO | 110.8 ± 11.8 (nmol/l) | 48 ± 4.9 (min) |
13C3 MGO | 109.7 ± 6.5 (nmol/l) | 56 ± 4.9 (min) |
Parameter . | Cmax* (mean ± SEM, unit) . | Tmax (mean ± SEM, unit) . |
---|---|---|
12C6 glucose | 328114 ± 40611 (peak area MS) | 41 ± 4.1 (min) |
13C6 glucose | 298345 ± 17997 (peak area MS) | 50 ± 4.3 (min) |
12C3 MGO | 110.8 ± 11.8 (nmol/l) | 48 ± 4.9 (min) |
13C3 MGO | 109.7 ± 6.5 (nmol/l) | 56 ± 4.9 (min) |
C6 glucose and 12C3 MGO were corrected for baseline, 13C6 glucose and 13C3 MGO were corrected for dose (*50). Abbreviation: MS, mass spectrometry.
Postprandial protein-bound AGEs formation in plasma increases rapidly during an OGTT in humans
Since MGO can react rapidly with proteins to form AGEs [24], we next investigated the formation of postprandial MGO-derived AGEs in plasma during the 6h OGTT. Plasma protein-bound CEL increased after a glucose load and reached a peak 60 min (+85%; P<0.001). Plasma protein-bound MG-H1 showed the highest levels after 120 min (+8%; P=0.36), but this was not statistically significant. For both MGO-derived AGEs, the concentrations decreased and the lowest concentrations were found at 240 min during the OGTT, after which the concentrations slightly increased again (Figure 2A,B). In contrast with protein-bound AGEs, the concentrations of free plasma MG-H1 and CEL continuously decreased during the entire 6h OGTT (Figure 2C,D).
MGO-derived AGEs concentrations in plasma during an OGTT in humans
(A) Protein-bound plasma MG-H1, (B) protein-bound plasma CEL, (C) free plasma MG-H1, and (D) free plasma CEL were analysed by UPLC-MS/MS. Data are shown as mean ± SEM, n=12. Comparisons of protein-bound and free AGEs levels at different time points were tested using one-way ANOVA with Tukey’s multiple comparison. * represents comparisons with time 0. * indicates P<0.05, ** indicates P<0.01, *** indicates P<0.001. Abbreviations: CEL, Nε-(carboxyethyl)lysine; MG-H1, Nδ-(5-hydro-5-methyl-4-imidazolon-2-yl)-ornithine.
(A) Protein-bound plasma MG-H1, (B) protein-bound plasma CEL, (C) free plasma MG-H1, and (D) free plasma CEL were analysed by UPLC-MS/MS. Data are shown as mean ± SEM, n=12. Comparisons of protein-bound and free AGEs levels at different time points were tested using one-way ANOVA with Tukey’s multiple comparison. * represents comparisons with time 0. * indicates P<0.05, ** indicates P<0.01, *** indicates P<0.001. Abbreviations: CEL, Nε-(carboxyethyl)lysine; MG-H1, Nδ-(5-hydro-5-methyl-4-imidazolon-2-yl)-ornithine.
MGO rapidly induces AGEs formation in plasma ex vivo
We next investigated whether these MGO-derived AGEs can be directly formed by MGO in plasma. The incubation of human plasma with different concentrations of MGO ex vivo for 24 h, induced a time- and dose-dependent increase in protein-bound MG-H1 and CEL, which occurred rapidly during the first 6 h (Figure 3A,B). After 6 h, the levels of protein-bound MG-H1 and CEL continued to increase, albeit at a slower rate.
Protein-bound AGE formation in plasma and BSA after incubating with MGO ex vivo
The concentrations of (A) protein-bound plasma MG-H1, (B) protein-bound plasma CEL, (C) protein-bound BSA MG-H1, (D) protein-bound BSA CEL, and MGO after (E) 10 μM MGO incubation and (F) 100 μM MGO incubation were measured by UPLC-MS/MS. Comparisons of MGO effects on protein-bound MG-H1 or protein-bound CEL formation were tested using two-way ANOVA with Bonferroni’s multiple comparison. * represents comparisons with 10 μM MGO, $ represents comparisons with BSA. * indicates P<0.05, ** indicates P<0.01, and $$$ndicates P<0.001. Abbreviations: BSA, bovine serum albumin; CEL, Nε-(carboxyethyl)lysine; MG-H1, Nδ-(5-hydro-5-methyl-4-imidazolon-2-yl)-ornithine; MGO, methylglyoxal.
The concentrations of (A) protein-bound plasma MG-H1, (B) protein-bound plasma CEL, (C) protein-bound BSA MG-H1, (D) protein-bound BSA CEL, and MGO after (E) 10 μM MGO incubation and (F) 100 μM MGO incubation were measured by UPLC-MS/MS. Comparisons of MGO effects on protein-bound MG-H1 or protein-bound CEL formation were tested using two-way ANOVA with Bonferroni’s multiple comparison. * represents comparisons with 10 μM MGO, $ represents comparisons with BSA. * indicates P<0.05, ** indicates P<0.01, and $$$ndicates P<0.001. Abbreviations: BSA, bovine serum albumin; CEL, Nε-(carboxyethyl)lysine; MG-H1, Nδ-(5-hydro-5-methyl-4-imidazolon-2-yl)-ornithine; MGO, methylglyoxal.
The incubation of BSA with MGO resulted in a similar pattern for protein-bound MG-H1 formation as in plasma, while protein-bound CEL formation was not induced by MGO (Figure 3C,D). The incubation of BSA with 10 and 100 μM MGO showed a fast decrease in MGO levels during the first 6 h (Figure 3E,F), demonstrating the high reactivity of MGO with proteins.
MGO formation in tissues during an intraperitoneal glucose tolerance test is also derived from exogenous glucose in mice
Next, we studied in mice whether exogenous glucose contributes to in vivo MGO formation in different organs after an intraperitoneal glucose bolus of universally labelled D(+)13C glucose. Blood glucose levels increased rapidly after the intraperitoneal glucose injection and reached a peak at 30 min (Figure 4A). Plasma 13C3 MGO levels (Figure 4B) followed the same trend as glucose, which is in accordance with the findings in humans (Figure 1B). In pancreas, spleen, kidney, SAT, and VAT, 13C3 MGO concentrations increased after the glucose bolus, with a peak at 30 min. After this peak, 13C3 MGO levels declined, with some residual 13C3 MGO in the tissues at 120 min (Figure 4C–G). In the liver and muscle, 13C3 MGO formation also increased during the IPGTT with a peak at 60 min (Figure 4H,I). Nonlabelled MGO levels in all the tissues during IPGTT were not affected by the bolus of glucose or slightly decreased at later time points (Supplementary Figure S1). Taken together, these data in mice showed that, similar to plasma, MGO transiently accumulates in tissues during a glucose tolerance test, and that its formation is directly derived from exogenous glucose.
Blood glucose levels and 13C3 MGO levels in tissues during a 2-h IPGTT in mice
Glucose levels were measured using a glucometer in (A) whole blood. 13C3 MGO concentrations were analysed by UPLC-MS/MS, in (B) plasma, (C) pancreas, (D) spleen, (E) kidney, (F) SAT, (G) VAT, (H) liver, and (I) muscle tissue from C57BL/6J male mice, 10–11 weeks old. Mice were grouped for different time points, n=6 per group. Data are shown as mean ± SEM. Comparisons of glucose levels or 13C3 MGO levels at different time points were tested using the ordinary one-way ANOVA with Tukey’s multiple comparison. * represents comparisons with time 0, $ represents comparisons with time 30 min, and # represents comparisons with 60 min. */$/# indicates P<0.05, **/$$/## indicates P<0.01, and ***/$$$/### indicates P<0.001. Abbreviations: SAT, subcutaneous adipose tissue; VAT, visceral adipose tissue.
Glucose levels were measured using a glucometer in (A) whole blood. 13C3 MGO concentrations were analysed by UPLC-MS/MS, in (B) plasma, (C) pancreas, (D) spleen, (E) kidney, (F) SAT, (G) VAT, (H) liver, and (I) muscle tissue from C57BL/6J male mice, 10–11 weeks old. Mice were grouped for different time points, n=6 per group. Data are shown as mean ± SEM. Comparisons of glucose levels or 13C3 MGO levels at different time points were tested using the ordinary one-way ANOVA with Tukey’s multiple comparison. * represents comparisons with time 0, $ represents comparisons with time 30 min, and # represents comparisons with 60 min. */$/# indicates P<0.05, **/$$/## indicates P<0.01, and ***/$$$/### indicates P<0.001. Abbreviations: SAT, subcutaneous adipose tissue; VAT, visceral adipose tissue.
Discussion
Our study demonstrates that postprandial plasma MGO formation during an OGTT originates from exogenous glucose. This formation of MGO during an OGTT is followed by increased formation of MGO-derived AGEs in plasma. Ex-vivo experiments showed that MGO is rapidly decreased when incubated with plasma or BSA, accompanied by a fast formation of MGO-derived AGEs. In mice, we showed that not only the postprandial increase in plasma MGO is derived from exogenous glucose but also MGO formation in pancreas, spleen, kidney, SAT, VAT, liver, and skeletal muscle.
Under physiological conditions, MGO is mainly formed during glycolysis by the nonenzymatic degradation of the glucotrioses glyceraldehyde 3-phosphate and dihydroxyacetone phosphate [13,19], but MGO can also be produced by auto-oxidation of glucose and degradation of glycated proteins [25], catabolism of threonine and acetoacetate [26,27], and lipid peroxidation [28]. In addition to these endogenous sources, MGO can be ingested from food [29]. We previously showed that MGO levels increased rapidly in plasma during an OGTT and a mixed meal test, and that MGO levels followed the trend of glucose curves [16,17], suggesting that increased levels of postprandial glucose are the source of MGO formation in the postprandial phase. Here, we established that the formation of MGO during the OGTT is indeed completely derived from exogenous glucose.
The decrease in 13C6 glucose and 13C3 MGO levels in plasma after the peak was slower as compared with unlabelled glucose and MGO within 180 min of OGTT, which is most likely due to an overcorrection of the 12C curve in the postprandial phase. The baseline level of 12C glucose is approximately 5 mmol/l and reflects the balance between endogenous production and disposal in fasting state. However, the endogenous production declines rapidly in the postprandial state due to the effects of insulin, to ∼35% of baseline by ∼60 min [17]. By subtracting a fixed baseline level of 5 mmol/l from the total curve of 12C6 glucose, this overcorrects the 12C curve, explaining a lower 12C6 glucose compared with 13C6 glucose curve. This effect becomes most apparent after ∼30–60 min. The same effects play a role for 12C3 MGO. The baseline 12C3 MGO levels probably reflect the endogenous production of glucose in the liver and muscles by oxidation of glucose/glycogen, which are processes that are also reduced in the postprandial phase. In support of this concept, we observed lower levels of 12C3 MGO in mice during the postprandial phase in several tissues including the liver and the muscle. Correction for the baseline MGO levels will again overcorrect the 12C3 MGO curve. Although we cannot exclude the possibility that there is also a kinetic isotope effect resulting in lower rates of utilisation of 13C6 glucose for glycolysis [30], such an effect is likely too small to explain the observed difference between 12C and 13C patterns of glucose and MGO. In addition, it was also observed that unlabelled glucose and MGO mildly increased again after 180 min of the OGTT, which could be due to a restoration in endogenous glucose formation after being suppressed in the postprandial state.
In agreement with our current findings, we have previously shown a similar increase in the MGO levels in plasma in healthy individuals during OGTT, with a higher increase in plasma MGO in individuals with type 2 diabetes as compared with healthy subjects [17]. Given that glucose spikes have been established as a key driver in the development of diabetic vascular complications [13,18], daily spikes of MGO, as a consequence of postprandial glucose peaks, may explain the effects of glucose spikes.
MGO can react with proteins, resulting in a fast formation of AGEs [24]. Indeed, in the OGTT in humans, we showed a fast increased formation of protein-bound CEL and MG-H1, although the later to a lesser extent. In the ex-vivo experiments, we confirmed MGO-induced formation of MG-H1 and CEL in plasma was in a concentration and time-dependent fashion. Interestingly, an incubation of MGO with pure BSA produced MG-H1, but failed to induce protein-bound CEL within 6 h. Since we observed MGO-induced CEL formation in plasma, and since CEL was most profoundly increased during the in vivo OGTT in human subjects, the reaction of MGO with lysine in plasma may preferentially be on proteins other than albumin. The low appearance of MG-H1 in plasma in vivo, in comparison with the high increase in the formation of MG-H1 ex vivo, might be due to a rapid degradation of MG-H1 containing proteins in vivo by epithelial cells of renal tubular [31]. In contrast with protein-bound MG-H1 and CEL, free MG-H1 and CEL were decreased over time during the OGTT. We previously found a strong association between dietary AGEs and free AGEs, but not with protein-bound AGEs [32]. Therefore, this decrease in free MG-H1 and CEL during the OGTT most likely reflects the process of the clearance of free AGEs that were derived from food before fasting.
In line with the findings in humans, changes of plasma 13C3 MGO levels followed the same pattern as of glucose during IPGTT in mice. Furthermore, an elevation of 13C3 MGO levels but not 12C3 MGO, after a bolus of glucose was observed in pancreas, spleen, kidney, SAT, VAT, liver, and muscle, indicating that the increase in MGO in tissues was completely derived from exogenous glucose. The changes of 13C3 MGO in the pancreas, spleen, kidney, SAT, and VAT followed the same pattern as for plasma 13C3 MGO. Thus, exogenous glucose rapidly results in MGO formation in various tissues. It has been previously demonstrated that glycolysis was largely increased in SAT in the postprandial phase, with only a small increase in pancreas and kidney, and no change in spleen and VAT [33]. Therefore, it might be that the elevated 13C3 MGO concentration in spleen and VAT is due to uptake of 13C3 MGO from plasma, rather than from glycolysis. The concentrations of 13C3 MGO in liver and muscle also increased during the IPGTT, however, with some delay compared with the other tissues. These tissues are very insulin dependent for the uptake of glucose and, therefore, this delay in the formation of 13C3 MGO in liver and muscle may reflect the kinetics of insulin release by the pancreas [34]. Another explanation for the relatively later increase in 13C3 MGO in the liver and skeletal muscle may be that the intake of glucose is not directly used for glycolysis but first absorbed and used into the glycogen pool, that was likely emptied due to the overnight fasting, before being used for glycolysis [33]. The fact that hepatic 12C3 MGO levels declined during IPGTT may be explained by a lower use of the glycogen contribution to glycolysis in the liver.
A limitation of the current study is that only male participants were enrolled in the human study and because of gender differences in response to an OGTT [35], these data may not be completely extrapolated to women.
Overall, we found that postprandial MGO formation during a glucose tolerance test is directly derived from exogenous glucose both in plasma and in tissues, followed by a fast formation of MGO-derived AGEs. MGO stress in the postprandial phase may contribute to the detrimental effects and long-term complications due to postprandial glucose spikes. Reducing MGO stress in the postprandial phase may be a way to reduce the burden of cardiovascular disease in diabetes.
Clinical perspectives
Background as to why the study was undertaken: plasma MGO levels are linked to cardiometabolic diseases. Increased levels of postprandial MGO in plasma during a glucose tolerance test or a mix meal have been observed, and with a higher increase in individuals with type 2 diabetes. However, the exact source of postprandial MGO is unknown.
A brief summary of the results: we herein showed that postprandial MGO formation during OGTT is completely derived from exogenous glucose in plasma and in tissues, followed by a fast formation of AGEs in plasma.
The potential significance to human health and disease: our current findings provide new insights in the formation of MGO in the postprandial phase and this may be of help in lowering MGO stress.
Data Availability
All supporting data are included within the article.
Funding
X.Z. is supported by the China Scholarship Council. Unilever was funder of the human study and provider of the samples from the human study.
CRediT Author Contribution
Xiaodi Zhang: Investigation, Writing—original draft. Jean L.J.M. Scheijen: Investigation, Methodology. Coen D.A. Stehouwer: Writing—review & editing. Kristiaan Wouters: Supervision, Writing—review & editing. Casper G. Schalkwijk: Supervision, Project administration, Writing—review & editing.
Acknowledgements
The authors acknowledge Unilever R&D Wageningen for their support of the human study. Unilever was in no way involved with the animal data.
Abbreviations
- AGE
advanced glycation endproduct
- ANOVA
analysis of variance
- BSA
bovine serum albumin
- CEL
Nε-(carboxyethyl)lysine
- IPGTT
intraperitoneal glucose tolerance test
- MG-H1
Nδ-(5-hydro-5-methyl-4-imidazolon-2-yl)-ornithine
- MGO
methylglyoxal
- OGTT
oral glucose tolerance test
- SAT
subcutaneous adipose tissue
- SEM
standard error of the mean
- UPLC-MS/MS
ultra-performance liquid chromatography-tandem mass spectrometry
- VAT
visceral adipose tissue