Dietary glycation compounds - implications for human health.

The term "glycation compounds" comprises a wide range of structurally diverse compounds that are formed endogenously and in food via the Maillard reaction, a chemical reaction between reducing sugars and amino acids. Glycation compounds produced endogenously are considered to contribute to a range of diseases. This has led to the hypothesis that glycation compounds present in food may also cause adverse effects and thus pose a nutritional risk to human health. In this work, the Senate Commission on Food Safety (SKLM) of the German Research Foundation (DFG) summarized data on formation, occurrence, exposure and toxicity of glycation compounds (Part A) and systematically assessed potential associations between dietary intake of defined glycation compounds and disease, including allergy, diabetes, cardiovascular and renal disease, gut/gastrotoxicity, brain/cognitive impairment and cancer (Part B). A systematic search in Pubmed (Medline), Scopus and Web of Science using a combination of keywords defining individual glycation compounds and relevant disease patterns linked to the subject area of food, nutrition and diet retrieved 253 original publications relevant to the research question. Of these, only 192 were found to comply with previously defined quality criteria and were thus considered suitable to assess potential health risks of dietary glycation compounds. For each adverse health effect considered in this assessment, however, only limited numbers of human, animal and in vitro studies were identified. While studies in humans were often limited due to small cohort size, short study duration, and confounders, experimental studies in animals that allow for controlled exposure to individual glycation compounds provided some evidence for impaired glucose tolerance, insulin resistance, cardiovascular effects and renal injury in response to oral exposure to dicarbonyl compounds, albeit at dose levels by far exceeding estimated human exposures. The overall database was generally inconsistent or inconclusive. Based on this systematic review, the SKLM concludes that there is at present no convincing evidence for a causal association between dietary intake of glycation compounds and adverse health effects.

Considering the implication of endogenous glycation compounds in aging and disease, dietary exposure via consumption of an "AGE (advanced glycation end product) rich diet" is increasingly suggested to pose a potential health risk. However, studies attempting to assess an association between dietary glycation compounds and adverse health effects frequently suffer from insufficient chemical analysis of glycation compounds, including inadequate structural characterization and limited quantitative data. The Senate Commission on Food Safety (SKLM) of the German Research Foundation (DFG) previously defined quality criteria for studies designed to assess the effects of dietary glycation compounds on human health. The aim of the present work is to summarize data on formation, occurrence, exposure and toxicity of glycation compounds (Part A) and to systematically evaluate if the currently available scientific database allows for a conclusive assessment of potential health effects of defined glycation compounds (Part B).The term "glycation compounds" comprises a wide range of structurally diverse compounds that derive from the Maillard reaction, a chemical reaction between reducing carbohydrates and amino compounds that occurs during food processing. In the first stage of the Maillard reaction, reducing sugars such as glucose and fructose react for instance with the ε-amino group of lysine, which is most abundant in food ("glycation" of lysine). Subsequently, these primary reaction products undergo Amadori rearrangement to yield products (ARP) such as fructosyllysine (FL) from glucose and also Heyns rearrangement products (HRPs) such as glucosyl- and mannosyllysine from fructose. While ARPs are rapidly formed during food processing, they are not stable and undergo degradation reactions, predominantly to 1,2-dicarbonyl compounds such as glyoxal (GO), methylglyoxal (MGO) and 3-deoxyglucosone (3-DG), which are highly reactive. The last stage of the Maillard reaction is characterized predominantly by the reaction of these dicarbonyl compounds with nucleophilic groups of proteins. The side-chains of lysine and arginine residues as well as the N-termini of proteins are important reaction sites. Carboxyalkylated amino acids such as N-ε-carboxymethyllysine (CML) and N-ε-carboxyethyllysine (CEL) result from reaction of the ε-amino group of lysine with the dicarbonyl compounds GO and MGO. Dicarbonyl compounds with C5 or C6 chains can form cyclic pyrrole derivatives at the ε-amino group of lysine. The most important example for this reaction is pyrraline, which is formed from reaction of 3-DG and lysine. The reaction of dicarbonyl compounds with the guanidino group of arginine mainly leads to hydroimidazolones, of which the MGO-derived hydroimidazolone 1 (MG-H1) is best described in food systems.ARPs are the most abundant glycation products found in food. Up to 55% of the lysine residues in food may be modified to ARPs at the side-chain. Food items particularly rich in ARPs include bread, rusk, biscuits, chocolate, and powdered infant formulas. Exposure estimates range between 0.6­1.6 mg/kg body weight (bw), although exposure may be as high as 14.3 mg/kg bw in individuals consuming foods with extreme ARP concentrations. Foods particularly rich in dicarbonyl compounds include heat-treated or long-term stored items rich in reducing sugars such as jams, alternative sweeteners, soft drinks, honey, candies, cookies, and vinegars, especially balsamico-type vinegars. The main contributors to the daily intake of MGO, GO, and 3-DG are coffee and bread. Dietary exposure to dicarbonyl compounds has been estimated to range between 0.02­0.29 mg/kg bw/d for MGO, 0.04­0.16 mg/kg bw/d for GO, 0.14­2.3 mg/kg bw/d for 3-DG, and 0.08­0.13 mg/kg bw/d for 3-deoxygalactosone (3-DGal). Dietary intake of 5-hydroxymethylfurfural (HMF), which can be formed from 3-DG, is estimated to range between 0.0001­0.9 mg/kg bw/d. Exposure estimates for individual glycated amino acids range from 0.03­0.35 mg/kg bw/d for CML, 0.02­0.04 mg/kg bw/d for CEL and 0.19­0.41 mg/kg bw/d for MG-H1. From a model diet consisting of 1 L milk, 500 g bakery products and 400 mL coffee, an intake of pyrraline corresponding to 0.36 mg/kg bw/d for a 70 kg person was estimated.Quantitative analysis of individual glycation compounds or their metabolites in tissues or body fluids as well as their reaction products with amino acids, proteins or DNA may serve to monitor exposure to glycation compounds. However, since glycation compounds are also formed endogenously, these biomarkers reflect the totality of the exposure, making it inherently difficult to define the body burden due to dietary intake against the background of endogenous formation.Information on the toxicokinetics and toxicity of glycation compounds is scarce and mostly limited to the reactive dicarbonyl compounds GO, MGO, 3-DG, HMF, and individual glycated amino acids such as CML and CEL. Acute toxicity of dicarbonyl compounds is low to moderate. There are some data to suggest that rapid detoxification of dicarbonyls in the gastrointestinal tract and liver may limit their oral bioavailability. Biotransformation of GO and MGO occurs predominantly via the glutathione (GSH)-dependent glyoxalase system, and to a lesser extent via glutathione-independent aldo-keto-reductases, which are also responsible for biotransformation of 3-DG. GO, MGO and 3-DG readily react with DNA bases in vitro, giving rise to DNA adducts. There is clear evidence for genotoxicity of GO, MGO and 3-DG. Repeated dose toxicity studies on GO consistently reported reduced body weight gain concomitant with reduced food and water consumption but did not identify compound related changes in clinical chemistry and hematology or histopathological lesions. There is also no evidence for systemic carcinogenicity of GO and MGO based on the available studies. However, initiation/promotion studies indicate that oral exposure to GO may exhibit genotoxic and tumor promoting activity locally in the gastrointestinal tract. From a 2-year chronic toxicity and carcinogenicity study in rats, a NOAEL for systemic toxicity of GO administered via drinking water of 25 mg/kg bw was reported based on reduced body weight and erosions/ulcer in the glandular stomach. Other non-neoplastic and neoplastic lesions were not observed. Acute toxicity of HMF is also low. From a 90-day repeated dose toxicity study in mice, a NOAEL of 94 mg/kg bw was derived based on cytoplasmic alterations of proximal tubule epithelial cells of the kidney. HMF was mostly negative in in vitro genotoxicity tests, although positive findings for mutagenicity were obtained under conditions that promote formation of the chemically reactive sulfuric acid ester 5-sulfoxymethylfurfural. There is some evidence of carcinogenic activity of HMF in female B6C3F1 mice based on increased incidences of hepatocellular adenoma, but not in male mice and rats of both sexes. Although data on oral bioavailability of glycated amino acids are mostly limited to CML, it appears that glycated amino acids may be absorbed from the gastrointestinal tract after oral exposure to their free and protein bound form. Glycated amino acids that are not absorbed in the intestine may be subject to metabolism by the gut microbiome. Glycated amino acids present in the systemic circulation are rapidly eliminated via the urine. Acute oral toxicity of CML is low. Studies in mice and rats reported changes in clinical chemistry parameters indicative of impaired renal and hepatic function. However, these changes were not dose-related and not supported by histopathological evaluation.Previous risk assessments of individual glycation compounds did not identify a health concern at estimated human exposures (GO, HMF) but also noted the lack of data to draw firm conclusions on health risks associated with exposure to MGO.To identify potential associations between dietary intake of defined glycation compounds and disease a systematic review was carried out according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) model, applying the quality criteria previously defined by the SKLM. Using a combination of keywords defining individual glycation compounds and relevant disease patterns linked to the subject area of food, nutrition and diet, a systematic search in Pubmed (Medline), Scopus and Web of Science was performed. Although the present systematic review identified numerous studies that investigated an association between an "AGE-rich diet" and adverse health effects, only a subset of studies was found to comply with the quality criteria defined by the SKLM and was thus considered suitable to assess potential health risks of dietary glycation compounds.For each adverse health effect considered in this assessment, only limited numbers of human studies were identified. Although studies in humans offer the advantage of investigating effects at relevant human exposures, these studies did not provide compelling evidence for adverse effects of dietary glycation compounds. Animal studies identified in this systematic review provide some evidence for induction of impaired glucose tolerance, insulin resistance, cardiovascular effects and renal injury in response to oral exposure to GO and MGO as representatives of dicarbonyl compounds. Only limited evidence points to a link between high intake of glycated amino acids and metabolic disorders. However, these effects were typically reported to occur at dose levels that exceed human dietary exposure, often by several orders of magnitude. Unfortunately, most studies employed only one dose level, precluding characterization of dose-response and derivation of a point of departure for riskassessment. While in vitro studies provide some evidence for a potential mechanistic link between individual glycation compounds and presumed adverse health effects, the clinical and toxicological relevance of the in vitro findings is often limited by the use of high concentrations of glycation compounds that by far exceed human dietary exposure and by insufficient evidence for corresponding adverse effects in vivo. A key question that has not been adequately considered in most studies investigating systemic effects of glycation compounds is the extent of oral bioavailability of dietary glycation compounds, including the form in which MRPs may be taken up (e.g. free vs. peptide bound glycated amino acids). Understanding how much dietary glycation compounds really add to the significant endogenous background is critical to appraise the relevance of dietary MRPs for human health.While it appears mechanistically plausible that glycation of dietary allergens may affect their allergenic potential, the currently available data do not support the hypothesis that dietary glycation compounds may increase the risk for diet-induced allergies. There are no human studies addressing the immunological effects of dietary AGEs. Accordingly, there are no data on whether dietary AGEs promote the development of allergies, nor whether existing allergies are enhanced or attenuated. In numerous in vitro studies, the IgG/E binding ability of antigens and therefore their allergenic potential has been predominantly reported to be reduced by glycation. However, some in vitro studies showed that glycated proteins bind to receptors of immunological cells, and thus may have promoting effects on immune response and inflammation.Although experimental data from animal studies provide some evidence that high doses of individual glycation compounds such as MGO and protein-bound CML may produce certain adverse health effects, including diabetogenic, cardiovascular, metabolic and renal effects, the doses required to achieve these effects by far exceed human dietary exposures. Of note, in the only long-term study identified, a high dose of MGO administered via drinking water to mice for 18 months had no adverse effects on the kidneys, cardiovascular system, or development of diabetes.Experimental data from animal studies provide evidence that high doses of defined glycation compounds such as MGO or protein-bound CML may affect glucose homeostasis. However, the doses required to produce these effects markedly exceed human dietary exposure. Results from human studies are inconclusive: Three short-term intervention studies suggested that diets rich in AGEs may impair glucose homeostasis, whereas one recent intervention study and two observational studies failed to show such an effect.For the cardiovascular system, there is some evidence from in vitro and in vivo studies that high concentrations of MRPs, well above the dietary exposure of humans, may enhance inflammation in the cardiovascular system, induce endothelial damage, increase blood pressure and increase the risk of thrombosis. Only a limited number of human intervention studies investigated potential effects of short-term exposure and longer-term effects of glycation compounds on the cardiovascular system, and yielded inconsistent results. The few observational studies available either found no association between dietary MRP intake and cardiovascular function or even reported beneficial effects. Therefore, currently no definitive conclusion on potential acute and chronic effects of dietary MRPs on inflammation and cardiovascular function can be drawn. However, there is currently also no convincing evidence that potential adverse effects on the cardiovascular system are triggered by dietary MRP intake.Furthermore, human studies did not provide evidence for an adverse effect of dietary MRPs on kidney function. In animal studies with high levels of oral intake, MGO was reported to cause structural and functional effects in the kidney. Several studies show that the concentration of modified proteins and amino acids, such as CML, increases significantly in kidney tissue after oral intake. One study showed a negative effect of a high-temperature-treated diet containing increased CML concentrations on kidney structure integrity and impaired glomerular filtration. The causative relationship of accumulation of dietary MRPs and a functional decline of the kidneys, however, needs further confirmation.With regard to gut health, there is some evidence for alterations in gut microflora composition and the production of individual short-chain fatty acids (SCFAs) upon dietary exposure to glycation compounds. However, this has not been linked to adverse health effects in humans and may rather reflect adaptation of the gut microbiota to changing nutrients. In particular, a human observational study and several animal studies did not find a correlation between the intake of glycation compounds and increased intestinal inflammation. In animal studies, positive effects of glycation compounds on gut tissue damage and dysbiosis during colitis were described.Considering clear evidence for DNA reactivity and genotoxicity of the dicarbonyl compounds GO, MGO and 3-DG, it is plausible to suspect that dicarbonyl compounds may induce mutations and cancer. Although there is some evidence for tumor promoting activity of GO locally in the gastrointestinal tract, the only guideline-compatible chronic rodent bioassays reported erosions and ulcer in the glandular stomach but no treatment-related neoplastic lesions. A recent multinational cohort study with focus on CEL, CML, and MG-H1 found no evidence to support the hypothesis that dietary AGEs are linked to cancer risk.Evidence for an association between human exposure to dietary glycation compounds and detrimental effects on the brain and on cognitive performance is far from being compelling. No human studies fully complying with the defined quality criteria were identified. A few experimental studies reported neuroinflammation and cognitive impairment following dietary MRP exposure, but these can be considered indicative at best and do not support firm conclusions for human health. In addition to utilizing exceedingly high dosages of individual agents like CML, harsh processing conditions causing a multitude of major process-related changes do not allow to convincingly reconcile effects observed with measured/supposed contents of free and protein-bound CML alone.Overall, although dietary glycation compounds have been claimed to contribute to a wide range of adverse health effects, the present critical evaluation of the literature allows the conclusion that the available data are insufficient, inadequate or inconclusive and do not compellingly support the hypothesis of human health risks being related to the presence of glycation compounds in food. The study limitations detailed above, together with the fact that a large number of studies did not comply with the defined quality criteria and therefore had to be excluded highlight the importance of performing adequately designed human or animal studies to inform scientifically reliable health risk assessment.To achieve this, high quality, dependable scientific cooperation within various disciplines is pivotal.

Controversy on health-based guidance values for bisphenol A-the need of criteria for studies that serve as a basis for risk assessment.

Since 2006, the responsible regulatory bodies have proposed five health-based guidance values (HBGV) for bisphenol A (BPA) that differ by a factor of 250,000. This range of HBGVs covers a considerable part of the range from highly toxic to relatively non-toxic substances. As such heterogeneity of regulatory opinions is a challenge not only for scientific risk assessment but also for all stakeholders, the Senate Commission on Food Safety (SKLM) of the German Research Foundation (DFG) analyzed the reasons for the current discrepancy and used this example to suggest improvements for the process of HBGV recommendations. A key aspect for deriving a HBGV is the selection of appropriate studies that allow the identification of a point of departure (PoD) for risk assessment. In the case of BPA, the HBGV derived in the 2023 EFSA assessment was based on a study that reported an increase of Th17 cells in mice with a benchmark dose lower bound (BMDL40) of 0.53 µg/kg bw/day. However, this study does not comply with several criteria that are important for scientific risk assessment: (1) the selected end-point, Th17 cell frequency in the spleen of mice, is insufficiently understood with respect to health outcomes. (2) It is unclear, by which mechanism BPA may cause an increase in Th17 cell frequency. (3) It is unknown, if an increase of Th17 cell frequency in rodents is comparably observed in humans. (4) Toxicokinetics were not addressed. (5) Neither the raw data nor the experimental protocols are available. A further particularly important criterion (6) is independent data confirmation which is not available in the present case. Previous studies using other readouts did not observe immune-related adverse effects such as inflammation, even at doses orders of magnitude higher than in the Th17 cell-based study. The SKLM not only provides here key criteria for the use of such studies, but also suggests that the use of such a "checklist" requires a careful and comprehensive scientific judgement of each item. It is concluded that the Th17 cell-based study data do not represent an adequate basis for risk assessment of BPA.

Commentary of the SKLM to the EFSA opinion on risk assessment of N-nitrosamines in food.

Dietary exposure to N-nitrosamines has recently been assessed by the European Food Safety Authority (EFSA) to result in margins of exposure that are conceived to indicate concern with respect to human health risk. However, evidence from more than half a century of international research shows that N-nitroso compounds (NOC) can also be formed endogenously. In this commentary of the Senate Commission on Food Safety (SKLM) of the German Research Foundation (DFG), the complex metabolic and physiological biokinetics network of nitrate, nitrite and reactive nitrogen species is discussed with emphasis on its influence on endogenous NOC formation. Pioneering approaches to monitor endogenous NOC have been based on steady-state levels of N-nitrosodimethylamine (NDMA) in human blood and on DNA adduct levels in blood cells. Further NOC have not been considered yet to a comparable extent, although their generation from endogenous or exogenous precursors is to be expected. The evidence available to date indicates that endogenous NDMA exposure could exceed dietary exposure by about 2-3 orders of magnitude. These findings require consolidation by refined toxicokinetics and DNA adduct monitoring data to achieve a credible and comprehensive human health risk assessment.