Chemistry and Safety of Acrylamide in Food

The unexpected finding that humans are regularly exposed to relatively high doses of acrylamide (AA) through normal consumption of cooked food was a result of systematic research and relevant developments in methodology over decades, as well as a chain of certain coincidences. The present paper describes the scientific approach, investigations and events leading to the discovery of the formation of AA during cooking of foods. In addition, related issues concerning assessment, communication and management of cancer risks and associated ethical questions raised by the finding of the presence of AA in foods will be discussed.

Acrylamide (AA) monomer is used in numerous chemical industries and is a contaminant in potato- and grain-based foods prepared at high temperatures. Although experimental animal studies have implicated carcinogenicity and reproductive toxicity as possible consequences of exposure, neurotoxicity is the only outcome identified by epidemiological studies of occupationally exposed human populations. Neurotoxicity in both humans and laboratory animals is characterized by ataxia and distal skeletal muscle weakness. Early neuropathological studies suggested that AA neurotoxicity was mediated by distal axon degeneration. However, more recent electrophysiological and quantitative morphometric analyses have identified nerve terminals as primary sites of AA action. A resulting defect in neurotransmitter release appears to be the pathophysiological basis of the developing neurotoxicity. Corresponding mechanistic research suggests that AA impairs release by adducting cysteine residues on functionally important presynaptic proteins. In this publication we provide an overview of recent advances in AA research. This includes a discussion of the cumulative nature of AA neurotoxicity and the putative sites and molecular mechanisms of action.

Since April 2002, when the Swedish National Food Administration first reported its finding of elevated levels of the substance acrylamide in commonly consumed foods (Swedish National Food Administration, 2002), there has been considerable debate about the health effects of dietary exposure to acrylamide. In particular, researchers have speculated on whether the amount of acrylamide consumed through the typical diet could increase the risk of cancer in humans. In this paper, we review the epidemiological data to date examining dietary acrylamide in relation to cancer risk. We highlight the strengths and limitations of using epidemiology to address this public health question. Finally, we provide an overview of future directions of epidemiological research on the health effects of dietary acrylamide.

Acrylamide is a monomer of polyacrylamide, used in biochemistry, in paper manufacture, in water treatment, and as a soil stabilizer. The monomer can cause several toxic effects and has the potential for human exposure either through the environment or from occupational exposure. Recently, additional concern for the potential toxicity of acrylamide in humans has arisen with the finding of acrylamide formation in some processed foods. It has been established that following chronic exposure, rats exhibited an increase in the incidence of adrenal pheochromocytomas, testicular mesotheliomas, thyroid adenomas and mammary neoplasms in F344 rats. This has raised increased concerns regarding the carcinogenic risk to humans from acrylamide exposure. Studies examining the DNA reactivity of acrylamide have been performed and have had differing results. The tissue and organ pattern of neoplastic development seen in the rat following acrylamide exposure is not consistent with that seen with other strictly DNA reactive carcinogens. Based on the pattern of neoplastic development, it appears that acrylamide is targeting endocrine sensitive tissues. In the current monograph, studies on the effect of acrylamide on DNA reactivity and on altered cell growth in the target tissues in the rat are reported. DNA synthesis was examined in F344 rats treated with acrylamide (0, 2, or 15 mg/kg/day) for 7, 14, or 28 days. Acrylamide increased DNA synthesis in the target tissues (thyroid, testicular mesothelium, adrenal medulla) at all doses and time points examined. In contrast, in a non-target tissue (liver), no increase in DNA synthesis was seen. Examination of DNA damage using single cell gel electrophoresis (the Comet assay) showed an increase in DNA damage in the target tissues, but not in non-target tissue (liver). In addition, a cellular transformation model, (the Syrian Hamster Embryo (SHE) cell morphological transformation model), was used to examine potential mechanisms for the observed carcinogenicity of acrylamide. SHE cell studies showed that glutathione (GSH) modulation by acrylamide was important in the cell transformation process. Treatment with a sulfhydryl donor compound (NAC) reduced acrylamide transformation while depletion of GSH (BSO) resulted in an enhancement of transformation. In summary, acrylamide caused both an increase in DNA synthesis and DNA damage in mammalian tissues and cells suggesting that DNA reactivity and cell proliferation, in concert, may contribute to the observed acrylamide-induced carcinogenicity in the rat and has implication on the possible risk for human neoplasm development.

This paper attempts to assess possible risks that may result from human exposure to dietary intake of acrylamide.

Acrylamide (AA) is a carcinogen as demonstrated in animal experiments, but the relevance for the human situation is still unclear. AA and its metabolite glycidamide (GA) react with nucleophilic regions in biomolecules. However, whereas AA and GA react with proteins, DNA adducts are exclusively formed by GA under conditions simulating in vivo situations. For risk assessment it is of particular interest to elucidate whether AA or GA within the plasma concentration range resulting from food intake are “quenched” by preferential reaction with non-critical blood constituents or whether DNA in lymphocytes is damaged concomitantly under such conditions. To address this question dose- and time-dependent induction of hemoglobin (Hb) adducts as well as genotoxic and mutagenic effects by AA or GA were studied in human blood as a model system.

Food is assumed to be one major source of acrylamide exposure in the general population. Acrylamide exposure is usually assessed by measuring hemoglobin adducts of acrylamide and its primary metabolite glycidamide as biomarkers. Little is known about the impact of acrylamide in food on biomarkers of acrylamide exposure. Therefore, CDC is conducting a feeding study to investigate the effect of consumption of endogenous acrylamide in food on biomarkers of acrylamide exposure. As part of this study, we performed a pilot study to obtain further information on the magnitude of the changes in biomarker levels after consumption of high amounts of potato chips (21 ounces) over a short period of time (1 week) in non-smokers. After 1 week, biomarkers levels increased up to 46% for acrylamide adducts and 79% for glycidamide adducts. The results indicate that changes in biomarker levels due to consumption of potato chips can be detected. However, because of the design of this pilot study, the observed magnitude of change cannot be generalized and needs to be confirmed in the main study.

Hemoglobin adducts of acrylamide and its primary metabolite, glycidamide are used as biomarkers of acrylamide exposure. Several methods for analyzing these biomarkers in blood have been described previously. These methods were developed to analyze small numbers of samples, not the high sample throughput that is needed in population screening. Obtaining data on exposure of the US population to acrylamide through food and other sources is important to initiate appropriate public health activities. As part of the Centers for Disease Control and Prevention biomonitoring activities, we developed a high throughput liquid chromatography tandem mass spectrometry (LC/MS/MS) method for hemoglobin adducts of acrylamide. The LC/MS/MS method consists of using the Edman reaction and isolating the reaction products by protein precipitation and solid-phase extraction (SPE). Quantitation is achieved by using stable-isotope labeled peptides as internal standards. The method is performed on an automated liquid handling and SPE system. It provides good sensitivity in the low-exposure range as assessed in pooled samples and enables differentiation between smokers and non smokers.

Acrylamide is metabolized by direct conjugation with glutathione or oxidation to glycidamide, which undergo further metabolism and are excreted in urine. In rats administered 3 mg/kg 1,2,3-

13

C

3

acrylamide, 59 % of the metabolites excreted in urine was from acrylamide-glutathione conjugation, whereas 25% and 16% were from two glycidamide-derived mercapturic acids. Glycidamide and dihydroxypropionamide were not detected at this dose level. The metabolism of acrylamide in humans was investigated in a controlled study with IRB approval, in which sterile male volunteers were administered 3 mg/kg 1,2,3-

13

C

3

acrylamide orally. Urine was collected for 24 h after administration, and metabolites were analyzed by

13

C NMR spectroscopy. At 24 h, urine contained 34 % of the administered dose, and 75 % of the metabolites were derived from direct conjugation of acrylamide with glautathione. Gycidamide, dihydroxypropionamide and one unidentified metabolite were also detected in urine. This study indicated differences in the metabolism of acrylamide between humans and rodents.

A pharmacokinetic (PBPK) model has been developed for acrylamide (AMD) and its oxidative metabolite, glycidamide (GLY), in the rat based on available information. Despite gaps and limitations to the database, model parameters have been estimated to provide a relatively consistent description of the kinetics of acrylamide and glycidamide using a single set of values (with minor adjustments in some cases). Future kinetic and mechanistic studies will need to focus on the collection of key data for refining certain model parameters and for model validation, as well as for conducting studies that elucidate the mechanism of action. Development of a validated human AMD/GLY PBPK model capable of predicting target tissue doses at relevant dietary AMD exposures, in combination with expanding data on modes of action, should allow for a substantive improvement in the risk assessment of acrylamide in food.

Several enzymes involved in the metabolism of xenobiotic substances are polymorphic in humans. Inter-individual differences in response to certain chemicals, such as acrylamide, as a result of such genetic polymorphisms might affect health-risk assessments. Detoxification by, for example, conjugation with glutathione (GSH) will decrease the concentration. The dose of the compound and enzymes that enhance the conjugation with GSH will increase the detoxification rate. The dose of acrylamide or glycidamide has been measured in blood samples from individuals with defined genotypes for the glutathione transferases GSTT1 and GSTM1 after

in vitro

incubation with these compounds. The results indicate that these enzymes have no significant effect on the blood dose, measured as Hb adducts over time, after exposure to acrylamide or glycidamide.

The heat-induced reaction of amino groups of amino acids, peptides, and proteins with carbonyl groups of reducing sugars such as glucose results in the concurrent formation of so-called Maillard browning products and acrylamide. For this reason, reported studies of adverse biological effects of pure acrylamide may not always be directly relevant to acrylamide in processed food, which may contain Maillard and other biologically active products. These may either antagonize or potentiate the toxicity of acrylamide. To stimulate progress, this paper presents an overview of selected reported studies on the antiallergenic/allergenic, antibiotic, anticarcinogenic/carcinogenic antimutagenic/mutagenic, antioxidative/oxidative, clastogenic (chromosome-damaging), and cytotoxic activities of Maillard products, which may adversely or beneficially impact the toxicity of acrylamide. The evaluation of biological activities of Maillard products and of other biologically active food ingredients suggests that they could both enhance and/or ameliorate acrylamide toxicity, especially carcinogenicity, but less so neurological or reproductive manifestations. Future studies should be directed to differentiate the individual and combined toxicological relationships among acrylamide and the Maillard products, define individual and combined potencies, and develop means to prevent the formation of both acrylamide and the most toxic Maillard products. Such studies should lead to safer foods.

This paper summarizes the progress made to date on acrylamide research pertaining to analytical methods, mechanisms of formation, and mitigation research in the major food categories. Initial difficulties with the establishment of reliable analytical methods have today in most cases been overcome, but challenges still remain in terms of the needs to develop simple and rapid test methods. Several researchers have identified that the main pathway of formation of acrylamide in foods is linked to the Maillard reaction and in particular the amino acid asparagine. Decarboxylation of the resulting Schiff base is a key step, and the reaction product may either furnish acrylamide directly or via 3-aminopropionamide. An alternative proposal is that the corresponding decarboxylated Amadori compound may release acrylamide by a beta-elimination reaction. Many experimental trials have been conducted in different foods, and a number of possible measures identified to relatively lower the amounts of acrylamide in food. The validity of laboratory trials must, however, be assessed under actual food processing conditions. Some progress in relatively lowering acrylamide in certain food categories has been achieved, but can at this stage be considered marginal. However, any options that are chosen to reduce acrylamide must be technologically feasible and also not negatively impact the quality and safety of the final product.

The formation of acrylamide (AA) from L-asparagine was studied in Maillard model systems under pyrolysis conditions. While the early Maillard intermediate

N

-glucosylasparagine generated ∼2.4 mmol/mol AA, the Amadori compound was a less efficient precursor (0.1 mmol/mol). Reaction with α-dicarbonyls resulted in relatively low AA amounts (0.2–0.5 mmol/mol), suggesting that the Strecker aldehyde pathway is of limited relevance. Similarly, the Strecker alcohol 3-hydroxypropanamide generated low amounts of AA (0.2 mmol/mol). On the other hand, hydroxyacetone afforded more than 4 mmol/mol AA, indicating that α-hydroxycarbonyls are more efficient than α-dicarbonyls in transforming asparagine into AA. The experimental results are consistent with the reaction mechanism proposed,

i.e.

(i) Streckertype degradation of the Schiff base leading to azomethine ylides, followed by (ii) β-elimination of the decarboxylated Amadori compound to release AA, The functional group in β-position on both sides of the nitrogen atom is crucial. Rearrangement of the azomethine ylide to the decarboxylated Amadori compound is the key step, which is favored if the carbonyl moiety contains a hydroxyl group in β-position to the N-atom. The β-elimination step in the amino acid moiety was demonstrated by reacting under pyrolysis conditions decarboxylated model Amadori compounds obtained by synthesis.

Studies on model systems of amino acids and sugars have indicated that acrylamide can be generated from asparagine or from amino acids that can produce acrylic acid either directly such as β-alanine, aspartic acid and carnosine or indirectly such as cysteine and serine. The main pathway specifically involves asparagine and produces acrylamide directly after a sugar-assisted decarboxylation and 1,2-elimination steps and the second nonspecific pathway involves the initial formation of acrylic acid from different sources and its subsequent interaction with ammonia to produce acrylamide. Aspartic acid, β-alanine and carnosine were found to follow acrylic acid pathway. Labeling studies with [

13

C-4]aspartic acid have confirmed the occurrence in aspartic acid model system, of a previously proposed sugarassisted decarboxylation mechanism identified in asparagine model systems. In addition, creatine was found to be a good source of methylamine and was responsible for the formation of N-methylacrylamide in model systems through acrylic acid pathway. Furthermore, certain amino acids such as serine and cysteine were found to generate pyruvic acid that can be converted into acrylic acid and generate acrylamide when reacted with ammonia.

The effectiveness of different compounds in the generation of acrylamide (AA) from asparagine, was determined by reacting asparagine with mono-, diand polysaccharides, as well as four different oxo-compounds known to be involved in carbohydrate metabolism/degradation. Quantitation of AA formed either under aqueous conditions or in low water model systems revealed glucose and 2-oxopropionic acid as the most effective compounds in AA generation, when reacted in model systems with a low water content (about 1 mol-% yield). Interestingly, heating of asparagine in the presence of 2-oxopropionic acid generated quite high amounts of 3-aminopropionamide (3-APA), which itself effectively generated AA upon heating in aqueous solution, as well as in low water systems. Because this is the first report on amounts of 3-APA generated by Maillard-type reactions, the general role of 3-APA as key intermediate in AA formation is discussed in detail. In addition, first results on the development and application of an HPLC/fluorescence method for AA quantitation are presented.

Heating amino acids with dietary oils or animal fats at elevated temperatures produced various amounts of acrylamide. The amount of acrylamide formation corresponded to the degree of unsaturation of the oils and animal fats. The decreasing order of acrylamide formation from dietary oils or animal fats with asparagine was sardine oil (642 μg/g asparagine) > cod liver oil (435.4 μg/g) > soybean oil (135.8 μg/g) > corn oil (80.7 μg/g) > olive oil (73.6 μg/g) > canola oil (70.7 μg/g) > corn oil (62.1 μg/g) > beef fat (59.3 μg/g) > lard (36.0 μg/g). Three-carbon unit compounds such as acrylic acid and acrolein, which are formed from lipids by oxidation also produced acrylamide by heat treatment with amino acids, in particular with asparagine. The results of the present study suggest that acrylamide forms in asparagine-rich foods during deep fat frying in the absence reducing sugars.

A kinetic model for the formation of acrylamide in potato, rye and wheat products has been derived, and kinetic parameters calculated for potato by multi-response modeling of reducing sugar (glucose and fructose), amino acid, asparagine and acrylamide concentrations with time. The kinetic mechanism shares, with Maillard browning, a rate limiting (probably dicarbonylic) intermediate, and includes reaction steps of this intermediate which are competitive with respect to acrylamide formation. A pathway representing physical and/or chemical losses of acrylamide accounts for the measured reduction of acrylamide yield at long reaction times. A mechanistic hypothesis regarding the competing reactions of Strecker aldehyde formation and tautomerization followed by beta-elimination to give acrylamide, features in the kinetic model and can be used to determine the factors which steer the reaction towards acrylamide. A predictive application of this model is for ‘what-if’ experiments to explore the conditions which lead to reduced acrylamide yields.

The relationship between acrylamide and its precursors, namely free asparagine and reducing sugars, was studied in simple cakes made from potato flake, wholemeal wheat and wholemeal rye, cooked at 180°C, from 5 to 60 min. Between 5 and 20 min, large losses of asparagine, water and total reducing sugars were accompanied by large increases in acrylamide, which maximized in all three products between 25 and 30 min, followed by a slow linear reduction. Acrylamide formation did not occur to any extent until the moisture contents of the cakes fell below 5%. A comparison of each type of cake with a commercial product, made from the same food material, showed that acrylamide levels in all three commercial products were well below the maximum levels in the cooked cakes.

Recent concerns surrounding the presence of acrylamide in many types of thermally processed food have brought about the need for the development of analytical methods suitable for determination of acrylamide in diverse matrices with the goals of improving overall confidence in analytical results and better understanding of method capabilities. Consequently, the results are presented of acrylamide testing in commercially available food products — potato fries, potato chips, crispbread, instant coffee, coffee beans, cocoa, chocolate and peanut butter, obtained by using the same sample extract. The results obtained by using LC-MS/MS, GC/MS (EI), GC/HRMS (EI) — with or without derivatization — and the use of different analytical columns, are discussed and compared with respect to matrix borne interferences, detection limits and method complexities.

Acrylamide in food is normally measured as “free water-soluble acrylamide”. However, it is shown that certain extraction techniques, like extraction as for dietary fibre or at high pH can affect the result. This has to be accounted for, particularly in exposure assessment and in studies of bioavailability and, in the long run, the health risk assessment.

Since the first discovery of the presence of acrylamide in a variety of food products in April 2002, numerous methods have been developed to determine the acrylamide monomer in heat-treated carbohydrate-rich food. These detection methods are mainly MS-based, coupled with a chromatographic step using LC or GC. The Food Chemistry Institute (LCI) of the Association of the German Confectionery Industry (BDSI) therefore established a detection method by means of aqueous extraction plus a cleaning step and LC-MS/MS detection, making great efforts to ensure internal and external validation. Citing potato crisps as an example, we will in the following show how the German manufacturing companies have gone to great pains to reduce acrylamide levels in their products.

A system to monitor the formation of acrylamide in model systems and from real food products under controlled conditions of temperature, time and moisture content has been developed. By humidifying the gas that flows through the sample, some control over moisture content can be affected. Results are presented to show the validity and reproducibility of the technique and its ability to deliver quantitative data. The effects of different processing conditions on acrylamide formation and on the development of color, due to the Maillard reaction, are evaluated.

Our finding that acrylamide is formed during heating of food initiated a range of studies on the formation of acrylamide. The present paper summarizes our follow-up studies on the characterization of parameters that influence the formation and degradation of acrylamide in heated foods. The system designed and used for studies of the influence of added factors was primarily homogenized potato heated in an oven. The net content of acrylamide after heating was examined with regard to the following parameters: heating temperature, duration of heating, pH and concentrations of various components. Higher temperature (200°C) combined with prolonged heating led to reduced levels of acrylamide, due to elimination/degradation processes. At certain concentrations, the presence of asparagine or monosaccharides (in particular fructose, glucose and glyceraldehyde) was found to increase the net content of acrylamide. Addition of other free amino acids or a protein-rich food component strongly reduced the acrylamide content, probably by promoting competing reactions and/or covalently binding of formed acrylamide. The pH-dependence of acrylamide formation exhibited a maximum around pH 8; lower pH enhanced elimination and decelerated formation of acrylamide. In contrast, the effects of additions of antioxidants or peroxides on acrylamide content were not significant. The acrylamide content of heated foods is the net result of complex reactions leading to both the formation and elimination/degradation of this molecule.

Simulated food pieces constructed from fiberglass pads (models for French fries and chips) were used as carriers for defined aqueous solutions, dispersions of test substances and ingredients to evaluate acrylamide formation. The pads were loaded with a solution containing asparagine and glucose (10 mM each) plus selected reaction modulators before deep fat frying and analysis for acrylamide. Data from fiberglass models along with companion sliced potato samples were used in developing hypotheses for the mechanisms involved in the suppression of acrylamide formation by polyvalent cations, polyanionic compounds, pH, and altered food polymer states in fried potato products.

We previously reported that in potato chip and French fry models, the formation of acrylamide can be reduced by controlling pH during processing steps, either by organic (acidulants) or inorganic acids. Use of phytate, a naturally occurring chelator, with or without Ca

++

(or divalent ions), can reduce acrylamide formation in both models. However, since phytate itself is acidic, the question remains as to whether the effect of phytate is due to pH alone or to additional effects. In the French fry model, the effects on acrylamide formation of pH, phytate, and/or Ca

++

in various combinations were tested in either blanching or soaking (after blanching) steps. All treatments significantly reduced acrylamide levels compared to control. Among variables tested, pH may be the single most important factor for reducing acrylamide levels, while there were independent effects of phytate and/or Ca

++

in this French fry model. We also developed a mathematical formula to estimate the final concentration of acrylamide in a potato chip model, using variables that can affect acrylamide formation: glucose and asparagine concentrations, cut potato surface area and shape, cooking temperature and time, and other processing conditions.

An integrated approach is described with respect to acrylamide minimization in heated foodstuffs. All relevant variables have to be considered and the main focus is on maintaining the expected product quality. The role of the processes at the interface between product and heating medium during processing is characterized for the case of frying operations. Examples of parameters influencing these processes with respect to minimizing acrylamide and maintaining product quality (e.g. brown color) are described. First, the local distribution of acrylamide in a French fries type model food was investigated. Lowering water activity at the surface of French fries before frying contributes to a reduction of acrylamide without lowering product quality. Both pre-drying of the potato sticks before frying and an increasing of salt concentration at the product surface by coating with a salt solution showed positive effects. Additionally, it was demonstrated by simulation that combined effects of these measurements may enable a reduction of up to 80% in the acrylamide content.

The discovery of acrylamide in processed potato products has brought increased interest in the controlling Maillard reaction precursors (reducing sugars and amino acids) in potato tubers. Because of their effects on nonenzymatic browning of fried potato products, reducing sugars and amino acids have been the focus of many potato research and breeding programs. This study focused on changes in sugars and amino acids in diploid potatoes selected for their storage qualities and their effect on acrylamide formation in the fried product. In addition, a second study was performed using cultivated lines that evaluated the effect of nitrogen fertilization on amino acid levels in tubers. Glucose, fructose, sucrose, and asparagine concentrations in tubers increased upon storage at 2°C. Glucose and fructose concentrations in the tubers were significantly and positively correlated with subsequent acrylamide formation in the products. Tuber sucrose and asparagine concentrations did not have an effect on acrylamide levels. Acrylamide levels in the products were significantly reduced if tubers were preconditioned before being placed in storage at 2°C. Higher rates of nitrogen fertilization resulted in increased amino acid concentrations in the tubers.

The discovery of the formation of acrylamide in fried and baked foods containing high levels of starch and the amino acid asparagine, prompted widespread concern. Both processed and home cooked foods are affected and this has led to the increased study of variations in cooking and processing conditions to minimize formation. While changes in cooking protocols have been in part successful, particularly when lower frying and baking temperatures are used, pretreatments to reduce levels of acrylamide by prevention of formation or acceleration of destruction have been investigated. In this study, a range of pretreatments of grilled potato were investigated and compared with surface washing to remove asparagine and reducing sugars. Synergies were observed between different treatments, and reductions of up to 40% were achieved in a non-optimized system.

Prototype processes were developed for the substantial suppression of acrylamide formation (40–95% compared to untreated controls) in cut surface fried potato products using potato chips (crisps) as the primary model. The most efficacious procedures employed sequentially both surface preparation and subsequent acrylamide precursor complexation and/or competitive inhibition processing steps. Surface preparation processing involved either various low-temperature (50–75°C) aqueous (5–30 min) or

ca.

80% ethanol blanch solutions for various times (1–5 min) combined with aqueous leaching steps (1–10 min) to reduce concentration of acrylamide precursors in the critical frying zone of cut potato surfaces. Acrylamide precursor complexation and/or competitive inhibition processing strategies included immersion exposure of prepared cut potato surfaces to solutions or dispersions of various combinations of either calcium chloride, phytic acid, chitosan, sodium acid pyrophosphate, or N-acetylcysteine.

Acrylamide concentrations in processed foods sold in Japanese markets were analyzed by LC-MS/MS and GC-MS methods. Most potato chips and whole potato-based fried snacks showed acrylamide concentration higher than 1000 µg/kg. The concentrations in non-whole potato based Japanese snacks, including rice crackers and candied sweet potatoes, were less than 350 µg/kg. Those in instant precooked noodles were less than 100 µg/kg with only one exception. The effect of storage condition of potato tubers on acrylamide concentration in potato chips after frying was also investigated. Sugar content in the tubers increased during cold storage, and the acrylamide concentration increased accordingly. The concentrations of asparagine and other amino acids, however, did not change during the cold storage. High correlations were observed between the acrylamide content in the chips and glucose and fructose contents in the tubers. This fact indicated that the limiting factor for acrylamide formation in potato chips is reducing sugar, not asparagine content in the tubers. Effects of roasting time and temperature on acrylamide concentration in roasted green tea are also described.

Many bakery products sold in the UK such as crumpets, batch bread and Naan might be expected to show high levels of acrylamide because they have strong Maillard colours and flavours. However, analysis of commercial products has shown that the highest levels of acrylamide are seen in dry biscuit type products. With the exception of spiced products such as ginger cake, moist high sugar products (e.g. cakes and fruit loaves) show relatively low levels of acrylamide, even in darkly browned crusts. This is in contrast to bread where acrylamide levels in excess of 100 µg/kg are common in the crust region, but are diluted by low levels in the crumb. Acrylamide levels in bread are significantly raised by domestic toasting, but other products such as crumpets and Naan bread have been found to be less sensitive. A mathematical model has been developed (and validated against tests on model dough) which shows that once obvious recipe differences are allowed for, the key factor limiting acrylamide levels is crust moisture. Chemical decay of acrylamide and depletion of amino acids are also limiting factors at higher temperatures.

The influence of ingredients, additives, and process conditions on the acrylamide formation in gingerbread was investigated. The sources for reducing sugars and free asparagine were identified and the effect of different baking agents on the acrylamide formation was evaluated. Ammonium hydrogencarbonate strongly enhanced the acrylamide formation, but its N-atom was not incorporated into acrylamide, nor did acrylic acid form acrylamide in gingerbread. Acrylamide concentration and browning intensity increased both with baking time and correlated with each other. The use of sodium hydrogencarbonate as baking agent reduced the acrylamide concentration by more than 60%. Free asparagine was a limiting factor for acrylamide formation, but the acrylamide content could also be lowered by replacing reducing sugars with sucrose or by adding moderate amounts of organic acids. A significant reduction of the acrylamide content in gingerbread can be achieved by using sodium hydrogencarbonate as baking agent, minimizing free asparagine, and avoiding prolonged baking.

Acrylamide is formed in high-carbohydrate foods during high temperature processes such as frying, baking, roasting and extrusion. Although acrylamide is known to form during industrial processing of food, high levels of the chemical have been found in home-cooked foods, mainly potato- and grain-based products. This chapter will focus on the effects of cooking conditions (e.g. time/temperature) on acrylamide formation in consumer-prepared foods, the use of surface color (browning) as an indicator of acrylamide levels in some foods, and methods for reducing acrylamide levels in home-prepared foods. As with commercially processed foods, acrylamide levels in home-prepared foods tend to increase with cooking time and temperature. In experiments conducted at the NCFST, we found that acrylamide levels in cooked food depended greatly on the cooking conditions and the degree of “doneness”, as measured by the level of surface browning. For example, French fries fried at 150–190°C for up to 10 min had acrylamide levels of 55 to 2130 µg/kg (wet weight), with the highest levels in the most processed (highest frying times/temperatures) and the most highly browned fries. Similarly, more acrylamide was formed in “dark” toasted bread slices (43.7–610.7 µg/kg wet weight), than “light” (8.27–217.5 µg/kg) or “medium” 10.9–213.7 µg/kg) toasted slices. Analysis of the surface color by colorimetry indicated that some components of surface color (“a” and “L” values) correlated highly with acrylamide levels. This indicates that the degree of surface browning could be used as an indicator of acrylamide formation during cooking. Soaking raw potato slices in water before frying was effective at reducing acrylamide levels in French fries. Additional studies are needed to develop practical methods for reducing acrylamide formation in home-prepared foods without changing the acceptability of these foods.