July 10, 2007

Formation and Reduction of Acrylamide in Maillard Reaction: A Review Based on the Current State of Knowledge

By Zhang, Yu Zhang, Ying

The recent report of elevated acrylamide levels in heat processing foods evoked an international health alarm. Acrylamide, an acknowledged potential genetic and reproductive toxin with mutagenic and carcinogenic properties in experimental mammalians, has been found in various heat processing foods. Many original contributions reported their findings on the formation mechanism and possible reduction methods of acrylamide. The aim of this review article is to summarize the state-of-the-art about the formation and reduction of acrylamide in the Maillard reaction. This research progress includes mechanistic studies on the correlation between the Maillard reaction and acrylamide, the formation mechanism of acrylamide, the main pathways of formation and impact factors on formation including cultivars, storage temperature, storage time, heat temperature, heat time, environmental pH, concentration of precursors, effects of food matrixes, type of oil, etc. Meanwhile, primary mechanisms on the reduction of acrylamide as well as reduction pathways including material and processing related ways and use of exogenous chemical additives are systematically reviewed. The mitigation studies on acrylamide are also summarized by the Confederation of the Food and Drink Industries of the EU (CIAA) "Toolbox" approach. Keywords acrylamide, formation, reduction, Maillard reaction, heat processing foods


Acrylamide (2-propenamide, CAS No. 79-06-1), a colorless and odorless crystalline solid with a melting point of 84.5[degrees]C, is formed from the hydration of acrylonitrile. Such a small compound is soluble in water, acetone, and ethanol, has a high mobility in soil and groundwater, and is biodegradable (Smith et al., 1996, 1997). Polyacrylamide polymers and copolymers are used in the paper and textile industries, as flocculants in the treatment of wastewater, as soil conditioners, in ore processing, and cosmetics. Studies by Smith et al. (1996) showed that heat, light, and outdoor environmental conditions, but not pH, promoted depolymerization of polyacrylamide to acrylamide.

Subsequently, such a contaminant drew wide attention because of an occasional finding. The observation that acrylamide used as a sealing adjuvant in tunnel construction in Sweden was responsible for adverse health effects in exposed workers finally led researchers to an association of acrylamide with foods. The results from animal tests showed that rats fed a fried chow diet had significantly higher levels of the hemoglobin (Hb) adduct of acrylamide, measured as N-(2-carbamoylethyl) valine, than those fed a control diet (Tareke et al., 2000). Nevertheless, an analysis of the heat-treated feed revealed the presence of acrylamide in amounts that paralleled those of the Hb adducts. In early 2002, the Swedish National Food Administration and the University of Stockholm together announced that carbohydraterich foods that are processed at relatively high temperature may contain considerable levels of acrylamide (SNFA, 2002).

Such findings have attracted considerable interest and wide attention and were very quickly validated by several governments (Ahn et al., 2002; UK FSA, 2002), and subsequently all available data on acrylamide were reviewed at an international level by the World Health Organization (WHO), the Food and Agriculture Organization (FAO) of the United Nations (UN) (FAO & WHO, 2002), and the Joint Institute for Food Safety and Applied Nutrition/National Center for Food Safety and Technology (JIFSAN/NCFST) Workshop (JIFSAN/NCFST, 2002) by expert working groups. Meanwhile, European researchers published their research about the acrylamide from the Maillard reaction products for the first time (Stadler et al., 2002; Mottram et al., 2002). Under such a situation, researchers from all over the world published their latest original contributions on various studies, such as the origin of acrylamide in foods, the health risk to humans, and intensive investigations in a global scope were launched into acrylamide.

Acrylamide has been classified as "probably carcinogenic to humans" (class 2A) by the International Agency for Research on Cancer (IARC, 1994) and exposure at high levels causes damage to the nervous system (Tilson, 1981; LoPachin and Lehning, 1994). Acrylamide is also considered as a reproductive toxin (Dearfield et al., 1988; Costa et al., 1992) with mutagenic and carcinogenic properties, as shown by in vitro and in vivo mammalian studies (Dearfield et al., 1995). Recently in 2005, WHO and FAO together announced that certain foods processed or cooked at high temperature especially Western-style snacks contain considerable levels of acrylamide and may harm human health to a certain extent again (INFOSAN, 2005). Therefore, research on acrylamide in different food matrixes has once again become a hotspot. Especially a monograph named "Chemistry and safety of acrylamide in food" and edited by Friedman and Mottram has recently been published (Friedman and Mottram, 2005). This book comprehensively introduced the discovery, analytical method, occurrence, toxicology, metabolism, and the risk assessment of acrylamide in heat-treated foods. On the other hand, this monograph summarized the state-of-the-art about all aspects of dietary acrylamide, which can be regarded as a fundamental research and still should be further investigated.

During these years, many review articles focused on various aspects of acrylamide research have been published in peerreview journals and most of them referred to comprehensive research progress to a macroscopical extent (Friedman, 2003; Stadler and Scholz, 2004; Taeymans et al., 2004; Blank, 2005). Meanwhile, some special reviews on acrylamide focused on coordination chemistry (Girma et al., 2005), analytical methods (Wenzl et al., 2003; Zhang et al., 2005a), neurotoxicology (LoPachin and Lehning, 1994; LoPachin et al., 2002), human exposure and internal dose assessments (Dybing et al., 2005), or carcinogenicity (Rice, 2005) have also been attracted. Recently, a special edition of the Journal of AOAC International was dedicated to a survey on 2 years of research activities (Anklam and Wenzl, 2005). In this issue, several aspects dealing with the acrylamide including risk management (Slayne and Lineback, 2005), exposure reduction (Grob, 2005), stability in food during storage (Hoenicke and Gatermann, 2005), perspective in mechanism formation (Yaylayan and Stadler, 2005), analytical methodologies (Castle and Eriksson, 2005), monitoring data bases (Lineback et al., 2005), activities conducted by the food industry (Taeymans et al., 2005) and assessment of results coming from interlaboratory proficiency tests (Owen et al., 2005) were reviewed. Another special issue of Mutation Research-Genetic Toxicology and Environmental Mutagenesis was published for the demonstration on carcinogenicity (Rice, 2005), DNA adduction, and mutagenic properties (Besaratinia and Pfeifer, 2005), the influence of glutathione transferases and epoxide hydrolase on detoxification (Paulsson et al., 2005), DNA strand breaking capacity (Puppel et al., 2005), and dietary exposure (Boon et al., 2005) of acrylamide in mammalian studies.

The chemical mechanism leading to the formation of acrylamide in foodstuffs derives from the Maillard reaction that occurs between reducing sugars and proteins/amino acids (Stadler et al., 2002). One of the main contributing compounds seems to be asparagine, an amino acid frequently occurring in foods (Yaylayan et al., 2003). The cooking process, in particular frying or roasting at high temperature, then induces the formation of acrylamide. More and more studies reported the details in many aspects of acrylamide formation. It is necessary to summarize their research and generalize the formation pathway of acrylamide in Maillard reaction. On the other hand, scientists have made their efforts for finding the chemical or physical way to reduce the amount of acrylamide in Maillard reaction during these years. Their original contributions are also worthy of review and summarization. With the development of further research on the formation of acrylamide and the findings of reduction pathways, corresponding important contributions are needed to emphasize.

This review article is an approbatory medium to list and summarize the above aspects for the formation and reduction of acrylamide by citing corresponding references. This review will address the formation and reduction aspects of the acrylamide issue and summarizes the progress made to date in these key areas. Since acrylamide formation is closely linked to endogenous and exogenous conditions of Maillard reaction, factors such as the presence of sugars and the availability of free amino acids are also considered. Many new findings that contribute towards a better understanding of the presence and control of acrylamide in Maillard reaction are presented. This review can also be considered as a basis for further research.


Maillard Reaction

The Maillard reaction has played an important role in improving the appearance and taste of various foods. The first coherent scheme for the general pathway of Maillard reaction was demonstrated by Hodge (1953) (see the detailed scheme in Figure 1). Actually, it proclaims that in an early stage, a reducing sugar, such as glucose, condenses with a specific compound possessing a free amino group (of an amino acid or in proteins mainly the e-amino group of lysine, but also the a-amino groups of terminal amino acids) to give a condensation product N-substituted glycosilamine, which rearranges to form the Amadori rearrangement product (ARP). The subsequent degradation of the Amadori product is dependent on the pH of the system. At pH 7 or below, it undergoes 1,2-enolization mainly with the formation of furfural (when pentoses are involved) or hydroxymethylfurfural (HMF) (when hexoses are involved). At pH > 7, the degradation of the Amadori compound is thought to involve mainly 2,3-enolization, where reductones, such as 4hydroxy-5-methyl-2,3- dihydrofuran-3-one (HMF^sub one^). and a variety of fission products, including acetol, pyruvaldehyde, and diacetyl are formed. All these compounds are highly reactive and take part in further reactions. Carbonyl groups can condense with free amino groups, which results in the incorporation of nitrogen into the reaction products. Dicarbonyl compounds will react with amino acids with the formation of aldehydes and alpha-aminoketones. This reaction is known as the Strecker degradation. Subsequently, in an advanced stage, a range of reactions takes place, including cyclizations, dehydrations, retroaldolizations, and mechanism using ^sup 13^C- labeled sugars. It involves different reaction pathways, in which the key intermediates are the 1-, 3- and 4-deoxyhexosuloses. Along with the enolization reactions, the Amadori product and rearrangements, isomerizations, and further condensations, which ultimately, in a final stage, lead to the formation of brown nitrogenous polymers and co-polymers, known as melanoidins (Martins et al., 2001). Figure 1 General pathway of Maillard reaction according to Hodge (1953) and Martins et al. (2001) with some modifications.

In spite of all the work that has been done and the great progress that has been achieved, the mechanism of the Maillard reaction is still regarded as a controversial issue. Such a controversial topic focuses on the formation of advanced glycation end products (AGEs) and melanoidins (Zheng and Xu, 2005). The chemical structures and corresponding formation mechanism of AGEs have attracted wide attention. Hitherto, several important monomers of AGEs, i.e. the crosslinks between adjacent proteins or other amino groups, have been found and isolated. These important compounds are demonstrated as pentosidine (Dyer et al., 1991; Grandhee and Monnier, 1991; Sell and Monnier, 1989), N^sup epsilon^- (carboxymethyl)lysine (CML) (Ahmed et al., 1986; Ahmed et al., 1997; Machado et al., 2006), LM-1 (versperlysine A) (Nagaraj and Monnier, 1992; Tessier et al., 1999), crossline (Aoki et al., 2000; Obayashi et al., 1996), and pyrraline (Nagaraj et al., 1995; Mendez and Leal, 2004; Miyata and Monnier, 1992). In general, they are considered as the products deriving from the reaction between dicarbonyl intermediates and amino groups of proteins (Thornalley, 2005). On the other hand, melanoidins are known as the brown nitrogenous polymers and co-polymers formed in the final stage of the Maillard reaction. So far, only partial structures of melanoidins have been elucidated. The origin and actual chemical species responsible for it remain largely undefined. Studies on the formation of melanoidins have been summarized in different review articles (Friedman, 1996; Rizzi, 1997), suggesting many important findings reported.

The two contributions published in Nature clarified the correspondence between the Maillard reaction and acrylamide. Original findings from Mottram et al. (2002) indicated that the Maillard reaction involving asparagine could produce acrylamide and might elucidate the elevated concentrations of acrylamide in certain plant-derived foods after cooking. They also concluded proposed pathways for the formation of acrylamide after Strecker degradation of the amino acids asparagine and methionine in the presence of dicarbonyl products from the Maillard reaction. Furthermore, a new insight from Stadler et al. (2002) showed that N-glycoside formation could be favored in foodprocessing matrixes that combined conditions of high temperature and water loss; when such condensation occurs between reducing sugars and certain amino acids, a direct pathway is opened up to the potential origin of acrylamide. Based on these two pioneer studies, more and more details and further mechanisms have been found. Many instrumental methods, such as proton transfer reaction mass spectrometry (PTR-MS), pyrolysis gas chromatography mass spectrometry (Py-GC/MS), and Fourier transform infrared (FT- IR) analysis, have been applied to on-line monitoring of the formation and control of intermediates and acrylamide (Pollien et al., 2003; Yaylayan et al., 2003). A detailed review will be addressed in the following section.

Fundamental Mechanistic Studies

Shortly after the proclamation of the discovery of acrylamide in heat-treated foods (SNFA, 2002; Tareke et al., 2002), numerous research groups in academic schools, industry, and official laboratories commenced studies into the possible sources and corresponding mechanisms. Several hypotheses on formation pathways were discussed at the very early stages of investigations, focusing initially on vegetable oils or lipids, since the problem mainly encompassed carbohydrate-rich foods that were fried or baked. In general, some critical and direct precursors contributing to the formation of acrylamide were demonstrated as 3-aminopropionamide, decarboxylated Schiff base (Zyzak et al., 2003), decarboxylated Amadori product (Yaylayan et al., 2003), acrylic acid (Becalski et al., 2003; Stadier, 2003a; Stadler et al., 2003b), and acrolein (Yasuhara et al., 2003). At the early stage of mechanistic study, researchers focused on parameters affecting the formation of acrylamide, such as heating time, heating temperature, pH, the ratio of amino acid and reducing sugar, etc. Heating equimolar amounts of asparagine and glucose at 180[degrees]C for 30 min resulted in the formation of 368 [mu] of acrylamide per mol of asparagine (Stadler et al., 2002). Adding water to the reaction mixture resulted in an increase of acrylamide up to 960 /imol/mol. A temperature-dependent study suggested that acrylamide formation also increases with temperature from about 120 to 17O0C and then decreases. Under similar conditions, methionine formed about one-sixth the amount of acrylamide. Similar temperature dependence was observed by Tareke et al. (2002) in laboratory heated foods. Moderate amounts (5-50 Mg/ kg) were detected in heated protein-rich foods and higher levels (150-4000 /zg/kg) in carbohydrate-rich foods such as beetroot and potatoes. No acrylamide was found in unheated or boiled foods. However, Ezeji et al. (2003) found that acrylamide is formed during the boiling or autoclaving of starch. Other amino acids producing low amounts of acrylamide include alanine, arginine, aspartic acid, cysteine, glutamine, methionine, threonine, and valine.

Several researchers have established that the main approach of formation of acrylamide in foods is related to the Maillard reaction and especially the amino acid asparagine via water or food matrix model systems (Mottram et al., 2002; Stadier et al., 2002). The link of acrylamide to asparagine, which directly provides the backbone chain of acrylamide molecule, was established by labeling experiments. Mass spectral studies showed that the three carbon atoms and the nitrogen atom of acrylamide are all derived from asparagine. The mechanism of acrylamide formation from a decarboxylated Amadori product of asparagine was shown in Figure 2. On the other hand, lipid oxidation has been suggested as a minor pathway, with acrylic acid as a direct precursor formed via acrolein by oxidative degradation of lipids (Gertz and Klostermann, 2002).

As the pathway shown in Figure 2, the first critical step is the amino-carbonyl reaction between asparagine and a carbonyl substance, preferably a-hydroxycarbonyls such as reducing sugars, resulting in the corresponding W-glycosyl conjugation and forming the Schiff base as a key intermediate after dehydration under elevated temperatures. Under low moisture conditions, both the W-glycosyl conjugation and the Schiff base are relatively stable (Robert et al., 2004). In aqueous systems, however, the Schiff base may hydrolyze to the precursors or rearrange to the Amadori compound (Fig. 2), which is not an efficient precursor in acrylamide formation (Yaylayan et al., 2003). Even under low moisture conditions, this reaction is most likely the main pathway initiating the early Maillard reaction stage that leads to 1- and 3-deoxyosones, which further decompose, generating color and flavor. This is in accordance with the relatively low transformation yield of asparagine to acrylamide (Blank, 2005). Alternatively, the Schiff base may be apt to make decarboxylation either directly via the Schiff betaine or via the intermediary oxazolidine-5-one to generate the azomethine ylide I, which furnishes the decarboxylated Amadori product after tautomerization (Yaylayan et al., 2003). In summary, acrylamide may be released via the following pathways: (a) directly from azomethine ylide I (Zyzak et al., 2003); (b) (^-elimination reaction from the Maillard intermediate, i.e. decarboxylated Amadori product (Yaylayan et al., 2003); and (c) loss of ammonia from 3-aminopropionamide deriving from the azomethine ylide II. Such reaction has been shown to preferentially proceed under aqueous conditions in the absence of sugars (Granvogl et al., 2004). Besides the main reaction precursors (reducing sugars and amino acids) and key intermediates, it is suggested that fat or oils can also play an important role in the acrylamide formation pathway. Another factor under consideration is the general background of food systems such as carbohydrate-rich or protein-rich matrix (Claeys et al., 2005a). During these years, many of the acrylamide-correlative studies focused on the investigation of impact factors on the formation of acrylamide. These factors include heating time, heating temperature, pH, the concentration of precursors and their compositions, heat processing methods, etc. Potato materials and corresponding products such as French fries, potato chips, and crisps are regarded as the most important food matrixes for the research on acrylamide formation. Recently, besides the abovementioned factors, other multiple factors such as potato cultivar, farming systems, field site, fertilization, pesticide/ herbicide application, time of harvest, storage time, and temperature, may all impact the final levels of the precursors to acrylamide to some degree. The importance of cultivar selection in controlling the acrylamide content has been emphasized in several studies. On the other hand, sugar control is an important mechanism of reacting to environmental change and stress, and leveraging the carbohydrate level is therefore a potential means of controlling acrylamide in the cooked product. Meanwhile, the fundamental formation studies based on asparagine and carbohydrate model systems were further conducted. The details on the formation study of acrylamide and main contribution from various laboratory colleagues all over the world were summarized in Table 1.

Figure 2 Mechanism of acrylamide formation from a decarboxylated Amadori product or dicarbonyls of asparagine.

Research groups from Mottram et al. (2002) and Stadier et al. (2002) have supplied evidence for the important contribution of the Schiff base deriving from asparagine, which relates to the dehydrated N-glucosyl compound. Nevertheless, the key intermediates were only partially clarified or not at all, and therefore the chemical interactions leading to acrylamide remained largely hypothetical. To fill in the gaps under currently proposed routes, Stadier et al. (2004) further demonstrated the mechanism of acrylamide formation in food by comparing the two major hypothetical pathways, i.e. via (a) the Strecker aldehyde route and (b) glycoconjugates of asparagine. Consequently, they have synthesized appropriate model intermediates and employed these in model systems to obtain a deeper insight into the key precursors and the salient reaction steps. The synthesis of Amadori compounds and N-glycosides of amino acids was further demonstrated. At first, Maillard intermediates related to the synthesis of Amadori compounds were prepared following the general procedure described by Lopez and Gruenwedel (1991) and some key intermediates were identified by NMR (Beksan et al., 2003; Stadier et al., 2004), which included:

2,3:4,50 -di0 -00 -isopropylidene-beta-D-fructopyranose 1

2,3:4,5-(di-O-isopropylidene-1-O-trifluoromethanesulfonyl)-beta- D-fructopyranose 2

tert-butyl N-(2,3:4,5-di- 0-isopropylidene-1 -deoxy-D-fructosl- yl)-L-asparaginate 3

sodium N-(2,3:4,5-di-0 -isopropylidene-1 -deoxy-D-fructos-1-yl)- L-asparaginate 4

N-(1-deoxy-D-fructos-1 -yl)-L-asparagine 5

Table 1 Summary of studies on the formation of acrylamide and major contributions from various laboratory colleagues

Table 1 Summary of studies on the formation of acrylamide and major contributions from various laboratory colleagues (

Table 1 Summary of studies on the formation of acrylaniide and major contributions from various laboratory colleagues

Table 1 Summary of studies on the formation of acrylamide and major contributions from various laboratory colleagues

N -(2,3:4,5-di- O -isopropy lidene-1 -deoxy-D-fructos-l-yl)- benzylamine 6

N -(2,3:4,5-di- O -isopropy lidene-1 -deoxy-D-fructos-1 -yl)-2- phenylethylamine 7

N-(1-deoxy-D- fructos-l-yl)-benzylamine 8

Second, N-glycosides of amino acids were prepared by adapting the general procedure for obtaining N-glycosides by condensation of reducing sugars and amino acids in anhydrous methanol under basic pH conditions, first described by Weitzel et al. (1957). The main N- glucosides included:

potassium N-(D-Glucos-l-yl)-L-asparaginate 9

potassium N-(D-Fructos-2-yl)-L-asparaginate 10.

The above compounds 3-10 are suitable compounds to study the reaction mechanisms leading to acrylamide under food processing conditions. The experimental results are consistent with the reaction mechanism based on (a) a Strecker type degradation of the Schiff base leading to azomethine ylides, followed by (b) a beta- elimination reaction of the decarboxylated Amadori compound to afford acrylamide. Meanwhile, alpha-hydroxycarbonyls are much more efficient than alpha-dicarbonyls in converting asparagine into acrylamide (Stadler et al., 2004).

Besides the Strecker degradation and N-glucosides steps of Maillard reaction suggested up to now as the key pathways in the formation of acrylamide, it might be proposed that decarboxylases present in the raw materials might generate the biogenic amine 3- aminopropionamide (3-APA) from asparagine, which is then thermally deaminated into acrylamide. This process would run without involving reducing carbohydrates. The formation of biogenic amines from amino acids via pyridoxal phosphate as cofactor is a reaction depending on the enzyme present (Karlson, 1988).

Recent publications (Zyzak et al., 2003) pointed to 3-APA as a transient intermediate in acrylamide formation during the thermal degradation of asparagine. As compared to an asparagine/glucose mixture, the amounts of acrylamide formed from 3-APA were reported to be five times higher. Further experiments indicated that 3-APA is formed during storage of intact potatoes (20 or 35[degrees]C) or after crushing of the cells. The heating of 3-APA under aqueous or low moisture conditions at temperatures between 100 and 180[degrees]C in model systems always generated more considerable acrylamide than in the same reaction of asparagine, therefore pointing to 3-APA as a very effective precursor of acrylamide. While the highest yields measured were about 28% (mol percent) in the presence of carbohydrates (170[degrees]C; aqueous buffer), in the absence of carbohydrates, 3-APA was even converted by about 63% (mol percent) into acrylamide upon heating at 170[degrees]C under aqueous conditions. Propanoic acid amides bearing an amino or hydroxyl group in the alpha-position, such as 2-hydroxypropionamide and L- alaninamide, were ineffective in acrylamide generation, indicating that elimination occurs only from the beta-position (Granvogl et al., 2004).

Importance of Asparagine

The first concrete evidence of the origin of acrylamide and the critical participation of asparagine in the reaction was presented in 2002 by several different research groups working independently all over the world (Stadler and Scholz, 2004). Using ^sup 15^N- labeled asparagine and mass spectral studies, it was unambiguously demonstrated that the three-carbon backbone of acrylamide and the amide nitrogen originated from corresponding positions in the asparagine molecule (Becalski et al., 2003). In principle, structural considerations dictate that asparagine alone may be converted thermally into acrylamide through decarboxylation and deamination reactions. However, the main product of the thermal decomposition of asparagine was maleimide, mainly due to the fast intramolecular cyclization reaction that prevents the formation of acrylamide. On the other hand, asparagine, in the presence of reducing sugars, was able to generate acrylamide in addition to maleimide (Yaylayan et al., 2003). Furthermore, subjecting asparagine alone to high temperatures does not lead to the formation of meaningful amounts of acrylamide, and the early studies cited above demonstrated the necessity of reducing sugars for the reaction to proceed, typically at temperatures above 100[degrees]C (Liu and Jiang, 2004; Stadler and Scholz, 2004).

Because asparagine is approved to be a major precursor for the heat-induced formation of acrylamide, control of the biosynthesis of free asparagine could turn out to be a useful approach to reduce acrylamide formation. Many researchers investigated the asparagine level in various plants. Martin and Ames (2001) found that asparagine is the free amino acid present in the highest amount in potatoes (939 mg/kg). Rozan et al. (2001) found that asparagine was quantitatively by far the most important amino acid present in five varieties of lentil seeds, ranging from 18000 to 62000 mg/kg of dry weight. Especially, a very high level of asparagine was detected in asparagus, i.e. 11000-94000 mg/kg (Hurst et al., 1998). Changes in asparagine content appear to be a good indicator of the changes in nitrogen metabolism of plants induced by pesticides and environmental factors (Friedman, 2003). Possible consequences of suppressing genes that govern the formation of enzymes involved in asparagine biosynthesis are still not known.

Importance of Sugars

As for the formation study of acrylamide, reactions in asparagine model systems containing either glucose, fructose, sucrose, sorbitol, glyceraldehyde, glycolaldehyde, or 2,3-pentanedione were investigated. All model systems generated acrylamide at 250[degrees]C; however, the intensity of the acrylamide increased at higher temperatures. Interestingly, at 250[degrees]C, the asparagine/ glycolaldehyde model system was more efficient than the asparagine/ glucose system in generating acrylamide; however, at 350[degrees]C asparagine/sucrose was more efficient than the asparagine/glucose model system. The least efficient system in generating acrylamide was that of asparagine/2,3-pentanedione, which generated acrylamide in trace amounts at both temperatures (Yaylayan et al., 2003). Biedermann et al. (2002b) also reported fructose is more efficient than glucose in forming acrylamide in a potato model. Similarly, Pollien et al. (2003) found fructose is more efficient in generating acrylamide from asparagine, compared to glucose. Comparisons between various sugars were also reported by Stadler et al. (2002). They found fructose, galactose, lactose, and sucrose to release acrylamide with comparable yields. Many researchers found that the reaction of acrylamide formation involves the reaction of a specific amino acid with a reducing sugar in the presence of heat (Mottram et al., 2002; Stadler et al., 2002; Becalski et al., 2003). However, the reaction with sucrose resulted in acrylamide levels similar to those recorded for fructose and glucose. A possible explanation to acrylamide formation from sucrose is that the sucrose is hydrolyzed upon thermal treatment to the individual monosaccharide. It considered that one sucrose molecule could then, in theory, provide two reducing hexoses, i.e., a molar ratio of sugar to amino acid of 2:1 (Taeymans et al., 2004). On the other hand, although glucose is a representative source for the formation of acrylamide, 2- deoxyglucose is a weak source because 2-deoxyglucose does not have a hydroxyl group adjacent to the carbonyl and it can only form a Schiff base adduct and cannot undergo the Amadori rearrangement, which leads to the formation of dicarbonyl compounds, e.g. 3- deoxyglucosone (Zyzak et al., 2003). Another possible route involves the reaction of asparagine with octanal to form acrylamide. These findings suggest the necessity of carbonyls in the formation of acrylamide, but that dicarbonyls are not essential (Becalski et al., 2003; Zyzak et al., 2003). In a word, acrylamide is formed in different amounts with several mono- ordisaccharides. Even non- reducing sugars, such as sucrose, are efficient reactants leading to the release of reducing sugars that are then available to react with the alpha-NH^sub 2^ group of asparagine via the Maillard pathway after thermally-induced hydrolysis.

Model Studies in Foods

During these years, the formation mechanism of acrylamide in asparagine-carbohydrate model systems has been nearly clarified. But actually, the formation of acrylamide is also affected by food matrixes and some important effect parameters on generating acrylamide should be investigated based on the asparagine- carbohydrate model studies. Many researchers involve such food matrix studies to find the factors affecting acrylamide formed in certain matrixes including potato, cereal, bread, almond, coffee, etc.


Potato tubers contain considerable amounts of the acrylamide precursors, i.e. free asparagine, glucose, and fructose, which may explain the high concentrations of acrylamide in certain potato products (Amrein et al., 2003). Potatoes of the cultivar Erntestolz stored at 4[degrees]C for 15 days showed an increase in reducing sugars from 80 to 2250 mg/kg (referring to fresh weight). As a consequence, the potential of acrylamide formation at 120[degrees]C rose by a factor of 28 (Biedermann et al., 2002b). Currently served French fries, a kind of favorite processing food made of potatoes, are the primary source of acrylamide and it is correspondingly important to reduce concentrations as far as reasonably possible. As we all know, any modifications performed on the raw material constituents will inevitably impact the Maillard reaction and its products and, concomitantly, the organoleptic properties including taste and color of the fried food. However, even though small scale and laboratory trials have shown that products such as French fries can be prepared with acrylamide amounts below 100 [mu]g/kg (Grob et al., 2003). Furthermore, Fiselier and Grob (2005) found that an average concentration of 50 [mu]g/kg acrylamide in French fries could be targeted by limiting the reducing sugars in the prefabricates to 0.7 g/kg and the frying temperature to 170[degrees]C. Even considerable over-frying in terms of duration can be tolerated then.


Cereal-based foods become another key source for the formation of acrylamide after heat treatment. Wheat and rye are two important cereal matrixes for daily staple foods in some Western countries. As for the contents of two acrylamide precursors, free asparagine and reducing sugars, Noti et al. (2003) reported levels of 0.15-0.4 g/ kg of asparagine in 10 samples of wheat flour. Surdyk et al. (2004) measured asparagine levels of 0.17 g/kg in white wheat flour. However, not enough data are available on the free amino acid compositions of rye. Springer et al. (2003) showed that levels of free asparagine varied across the rye grain, with the lowest level in the endosperm and the highest level in the bran. Acrylamide levels in wafers made from rye flours reflected their free asparagine content, with levels of acrylamide in bran being greater than in cooked wholemeal rye, which were greater than in cooked rye flour without bran. Elmore et al. (2005) studied the relationship between acrylamide and its precursors in wheat and rye. Between 5 and 20 min, major losses of asparagine, water, and total reducing sugars were accompanied by large increases in acrylamide, which maximized between 25 and 30 min, followed by a slow linear reduction when the cereal matrixes were cooked at 180[degrees]C. Acrylamide formation did not occur to a large degree until the moisture contents of the cakes fell below 5%. Possibly, a combined treatment of pH (citric acid) and low levels of added amino acids (soy protein hydrolysate) could result in significant reductions in acrylamide, but the effects of such treatments on sensory properties such as color and flavor remain to be evaluated (Cook and Taylor, 2005).


Besides the presence of two precursors, the considerable amount of acrylamide in bread was due to comfortable bakery time and temperature. Brathen and Knutsen (2005) studied the dry starch systems, containing freeze-dried rye-based flat bread doughs, flat bread and bread, which were baked at varying temperatures and time. The amount of acrylamide formation went through a maximum at approximately 200[degrees]C, depending on the system and the baking time. The amount of acrylamide was reduced at long baking time. However, in bread crust, the amount of acrylamide increased with both baking time and temperature in the interval test. Another possible source for high amounts of acrylamide, gingerbread, is demonstrated in some European countries. Gingerbread may contain up to 1000 [mu]g of acrylamide/kg of fresh weight. Konings et al. (2003) measured acrylamide in Dutch gingerbread products ranging from 260 to 1410 [mu]g/kg (average = 890 [mu]g/kg; median = 1070 [mu]g/kg). Because gingerbread is consumed frequently and all through the year in the Netherlands, these products were estimated to contribute 16% of the total acrylamide exposure of the Dutch population. As for the control of acrylamide in gingerbread, ammonium hydrogencarbonate strongly enhanced acrylamide formation, but its N atom was not incorporated into acrylamide, nor did acrylic acid form acrylamide in gingerbread (Amrein et al., 2004a). It is concluded that a significant reduction of acrylamide in gingerbread can be achieved by using hydrogencarbonate as baking agent, minimizing free asparagine, and avoiding prolonged baking.


Almonds also contain both acrylamide precursors in appreciable levels. The content of free asparagine is reported in the range of 2000-3000 mg/kg (Seron et al., 1998) while glucose and fructose levels were determined to be 500-1300 mg/kg, and sucrose contents ranged from 2500 to 5300 mg/kg (Ruggeri et al., 1998). As a result, the detection of acrylamide in roasted almonds in concentrations from 260 to 1530 [mu]g/kg (Becalski et al., 2003) was not surprising. As for the area difference on acrylamide levels in roasted almonds, almonds of European origin contained significantly less free asparagine and formed accordingly less acrylamide during roasting compared to the almonds from the U.S. (Amrein et al., 2005b). Interestingly, acrylamide was found to decrease in roasted almonds during storage at room temperature (Amrein et al., 2005a). It can be assumed that reactive compounds formed during the roasting process may be responsible for the decrease of acrylamide in these products during storage.


Coffee is generally roasted at temperatures in the range of 220[degrees]C to 250[degrees]C, and the time and speed of roasting play an important role on the sensorial properties, such as color, aroma, and taste. These are carefully fine-tuned to a characteristic profile leading to the clear identity of the coffee product. Coffee beans are thus subjected to relatively higher temperatures than other foods, and one can expect more than one pathway leading to acrylamide (Stadler and Scholz, 2004). The free asparagine concentration in green coffee beans lies within a very narrow range, typically from 30 to 90 mg/100g. Two research groups have found that acrylamide is not stable in commercial coffee stored in its original container (Andrzejewski et al., 2004; Delatour et al., 2004). Losses of 40% to 60% have been recorded in ground coffees stored at room temperature. On the other hand, experiments have shown that acrylamide is degraded or eliminated during roasting, and the profile of acrylamide formation during the roasting of coffee reflects this effect very clearly (Taeymans et al., 2004). In coffee, acrylamide is formed at the very beginning of the roasting step, reaching more than 7 mg/kg, and then declining sharply toward the end of the roasting cycle due to higher rates of elimination. Within the commercial roasting range, the acrylamide amount was reduced by a factor of approximately 10 compared with the deepest level recorded during the complete roasting cycle (Stadler and Scholz, 2004). Nevertheless, deeper roasting as a potential choice to reduce acrylamide could generate other undesirable compounds and negatively impact the sensory properties of the product. Consequently, no applicable solutions are today in practice that would reduce acrylamide levels and concomitantly retain the quality characteristics of coffee, because the roasting step cannot be thoroughly changed. Kinetic Studies

The kinetic study of acrylamide in the Maillard reaction includes the formation and the elimination processes (Claeys et al., 2005b). According to the reaction in the model system, the formation of acrylamide from asparagine and reducing sugar, such as glucose, can be modeled by a second-order reaction because these two reactants are present at an equimolar level in the model system. Therefore, the concentration of asparagine (C^sub Asn^) and reducing sugar (C^sub Sugar^) can be expressed as the concentration of reactant (C^sub R^). On the other hand, the elimination of acrylamide can be regarded as a first-order kinetic process. Biedermann et al. (2002a), monitoring the elimination of acrylamide by D^sub 3^- acrylamide, proposed first-order kinetics for describing the elimination of acrylamide. However, whereas both asparagine and sugar undergo the Maillard reaction, the sugar is consumed by caramelization reactions as well. Moreover, in the Maillard reaction, the loss of sugar has been reported to be faster than the loss of amino acid. On the basis of these arguments, asparagine can be considered in excess as compared to the sugar. In other words, the kinetics of the formation process lies on the concentration of sugar (C^sub Sugar^). Based on all of the above consequences, the acrylamide yield (C^sub AA^) can be described by a first-order formation/first-order elimination kinetic model and corresponding kinetic equation can be written as follows, with k^sub F^, k^sub E^, and t the formation rate constant, the elimination rate constant and the treat time.

Equation (1) reflecting the kinetics of acrylamide is conveniently integrated when kinetics is analyzed under isothermal conditions (constant temperature) and k values can be assumed to be constant. However, because heating of the samples involved nonisothermal conditions (variable temperature), the integrated effect of temperature on the reaction rate constant has to be taken into account (Claeys et al., 2005b; Granda and Moreira, 2005). Traditionally, a well-known theorem named the Arrhenius equation clearly demonstrated the effect of temperature on the reaction rate constant k, in which the temperature dependence of k is quantified by the activation energy E^sub a^ (J/mol).

As for the convention of variables in the equation (2), R is the universal gas constant [8.314 J/(mol.K)], T is the absolute temperature (kelvins, abbr. K), k^sub 0^ is the the reaction rate constant, and T^sub 0^ is the reference temperature (Claeys et al., 2005b). During the modeling phase, the kinetic parameters describing acrylamide formation and elimination, i.e. k^sub F0^, k^sub E0^, E^sub aF^, and E^sub aE^, could be estimated by nonlinear regression according to Gauss-Newton algorithm.

A primary kinetic study demonstrated the formation and elimination of acrylamide in the Maillard reaction. However, the loss of precursors, the dynamic balance between formation and elimination of some key intermediates, and the formation of other compounds such as melanoidins cannot be monitored based on such kinetic study. Therefore, the target of further kinetic study is to fill in the above-mentioned gaps. The reaction network can be simplified and shown in Figure 3 (Stadler et al., 2004; Knol et al., 2005). As for the kinetic rate constants, k^sub 1^, k^sub 2^, k^sub 3^, k^sub 4^, k^sub 5^, and k^sub 6^ are expressed as loss of asparagine and glucose, formation of fructose, loss of asparagine and fructose, formation of acrylamide, formation of melanoidins, and elimination of acrylamide, respectively. For each reaction step, a differential equation was set up by the use of the law of mass action, and the obtained differential equations were solved by numerical integration. The temperature dependence, which plays a major role in the Maillard reaction, was also taken into account by including an Arrhenius relationship (Equation 2) among the rate constants (k) of the different reactions (Knol et al., 2005).


As for the control of acrylamide in foods, the basic knowledge opens the way to focus studies on kinetic modeling (formation over temperature and time, the effect of water activity, elimination of acrylamide, and the formation of final products), and identifying the rate-limiting steps under actual food matrix systems (Blank, 2005). Measures can then be optimized to attempt to reduce acrylamide in commercial food products. Hitherto, most of the publications have been carried out on fried potatoes to understand the critical factors that may control or reduce acrylamide (Grob et al., 2003; Taubert et al., 2004). Possible pathways on the reduction of acrylamide in potato products include a combination of measures, such as controlling the storage temperature of the raw materials, variety, and optimized processing conditions (time and temperature).

Figure 3 Reaction kinetic model for the formation of acrylamide from asparagine and glucose/fructose.k^sub 1^, loss of asparagine and glucose; k^sub 2^, formation of fructose; k^sub 3^, loss of asparagine and fructose; k^sub 4^, formation of aerylamide; k^sub 5^, formation of melanoidins; k^sub 6^, elimination of acrylamide.

Mechanistic Study on Reduction of Acrylamide

Many fundamental studies demonstrated the formation mechanism of acrylamide in either asparagine and carbohydrate model systems or various food matrixes (Stadler et al., 2004; Yaylayan et al., 2003; Zyzak et al., 2003). Based on the current knowledge, acrylamide may be reduced under the following situations: (i) Some key intermediates may be eliminated under the change of reaction conditions; (ii) Other vinylogous compounds other than acrylamide are formed; (iii) Some key pathways such as the formation of Schiff base, Strecker type degradation, N-glucoside pathway, and beta- elimination reaction of the decarboxylated Amadori compounds are blocked.

Yaylayan et al. (2003) reported the possible formation pathways of two substitutes, i.e. maleimide and niacinamide instead of acrylamide generation. First, the reduction of acrylamide can be achieved by preventing the decarboxylation of asparagine itself, which can further generate acrylamide via deamination. Obviously, intramolecular cyclization to form an imide is compared much faster to the decarboxylation reaction due to its entropy value. It is approved that the tendency of chemical reactions is in favor of the formation of maleimide. Therefore, this reaction should theoretically reduce the latent toxicity of asparagine in food by preventing its conversion into acrylamide. The intermediate 3- aminosuccinimide was detected in the pyrogram of asparagine model systems along with maleimide. Second, the reduction of acrylamide can be achieved by preventing the Schiff base of N- glycosylasparagine from stabilizing itself through intramolecular cyclization initiated by the carboxylate anion and formation of oxazolidin-5-one (Manini et al., 2001). There are two main stabilization forms of the Schiff base of N-glycosylasparagine, i.e. Amadori rearrangement and intramolecular cyclization (Yaylayan et al., 2003). The Amadori product eliminates the carboxylic acid free via intramolecular cyclization at elevated temperatures and generates an Amadori product with N-substituted succinimide. The main residue generated from the amino acid moiety identified in the pyrolysates of asparagine model systems was succinimide. This pathway, similar to that occurring in asparagine alone, prevents the formation of acrylamide. According to the mechanism of competitive inhibition, such formation pathway should be encouraged because it allows the Amadori product to decarboxylate and inhibits the intramolecular cyclization of the Schiff base.

Raw Materials and Processing

With the development of acrylamide formation studies, more and more researchers are involved in the reduction of acrylamide contaminant. Especially since 2004, studies on ways to reduce acrylamide in foods have become a hotspot. Excitingly, various effective ways for the reduction of acrylamide have been found during these years such as the change of precursors in food materials, change of heat processing methods, optimization of suitable cultivar and the storage temperature of food materials, fermentation, etc. However, most of the reduction mechanisms of acrylamide via the above-mentioned pathways especially via effects of exogenous chemical additives are unclear. Meanwhile, different methods for the reduction of acrylamide related to raw materials and processing will be briefly demonstrated as follows:

Decrease of Reducing Sugars in Food Matrixes

Fiselier and Grob (2005) found that a suitable limit for the reducing sugars in the prefabricates for French fries is a simple and efficient measure to reduce the exposure to acrylamide from the predominant source for many consumers. Meanwhile, many researches (Biedermann-Brem et al., 2003; Chuda et al., 2003; Grob et al., 2003) demonstrated that the acrylamide level in potato chips made from tubers stored at low temperature (10[degrees]C). Besides this, another important factor which affects the acrylamide level is the potato cultivar (Biedermann-Brem et al., 2003; Grob et al., 2003). Therefore, this method can be considered as the control of precursors and achieved based on some additional conditions: (i) choice of plant cultivars low in reducing sugar; (ii) avoidance of storage at temperatures below 10[degrees]C; (iii) optimization of the blanching process to extract the maximum amount of sugar and asparagine; (iv) avoidance of adding reducing sugars in further treatments. Modification of Processing Conditions

The control of important processing parameters could be regarded as the most direct way to reduce acrylamide. These parameters include heating temperature, heating time, oil type, etc. The temperature dependence of acrylamide formation from asparagine indicates that this is favored above 100[degrees]C and that very high temperatures are not necessary (Mottram et al., 2002). That means there is not a simple linear increase between the acrylamide level and heating temperature. However, the acrylamide concentrations were noticed to increase with heating time. Furthermore, the formation of acrylamide in food was much higher when using palm olein or frying oils containing silicone (Gertz and Klostermann, 2002). Therefore, acrylamide reduction may be achieved by low temperature heating such as low temperature vacuum frying (Granda et al., 2004), short time heating and avoiding the use of palm olein as for modification of processing.

Change of Heat Processing Methods

Blanching instead of frying or soaking before frying could significantly reduce the acrylamide level (Haase et al., 2003; Pedreschi et al., 2004). However, such an idea seems not very practical because the negative effects on the connatural and sensory characteristics of processing foods may inevitably occur.


Acrylamide formed in the Maillard reaction may also be reduced via the addition of exogenous chemical additives, which should comply with the following conditions: (i) The addition level should be properly controlled according to corresponding criteria of food or chemical additives; (ii) the selected additives should be regarded as no toxicity demonstrated by toxicity test from previous publications; (iii) the additives applied to the food systems cannot affect the connatural and sensory characteristics of processing foods. During these years, many additives have been found to have the inhibitory effect of acrylamide formation in the Maillard reaction.

Use of Citric Acid to Adjustify pH

Several studies (Cook and Taylor, 2005; Jung et al., 2003; Pedreschi et al., 2006; Rydberg et al., 2003; Stadler et al., 2003) have demonstrated that pH modification are potential ways in which to reduce acrylamide formation in food and model systems. Lowering the pH of the food system to reduce acrylamide generation may attribute to protonating the alpha-amino group of asparagine, which subsequently cannot engage in nucleophilic addition reactions with carbonyl sources (Jung et al., 2003). Cook and Taylor (2005) reported that the effect of citric acid addition alone was a 23.5% reduction in acrylamide at pH 4.48 (product pH lowered by 1.05 units) and 47% reduction at pH 3.93 (product pH lowered by 1.6 units). Such reduction rates were less in percentage terms than those reported by Jung et al. (2003) when modifying the pH of corn grits (also with citric acid) prior to frying. The correlation between pH decrease and acrylamide reduction may vary among products, due to multiple factors or different starting pH values of the products.

Use of Proteins and their Hydrolysates

As for in vitro studies, Schabacker et al. (2004) studied the reduction of acrylamide uptake by dietary proteins in Caco-2 gut model. They found that although acrylamide monomers easily diffuse through Caco-2 monolayers, acrylamide uptake from food in the human intestine may differ from these experimental conditions. Acrylamide molecular contains an active double bond, which may interact with food ingredients, mainly proteins, DNA and RNA. In vitro incubation of acrylamide together with glutathione (GSH) at pH 8 statistically significantly reduces the amount of acrylamide monomers to 81%. Meanwhile, a higher availability of cysteines (molar ratio 1:10; acrylamide/GSH) led to 48% reduction of acrylamide because acrylamide is most likely bound covalently to glutathione via Michael addition of cysteine residues to the terminal double bond (Schbacker et al., 2004). The addition of glycine or glutamine during blanching of potato crisps reduced the amount of acrylamide by ~30% compared to no addition (Claeys et al., 2005e). Similar results have been demonstrated via in vitro scavenging reactions of acrylamide with GSH by Cui et al. (2005). The degradation mechanism of acylamide in vitro by GSH is mainly via the decomposition of glycine fragment of GSH. The glycine moiety can be eliminated more readily by the cleavage of peptide bonding in the presence of acrylamide. The glycine in the solution is degraded by the Streck degradation to aldehyde, ammonia, and carbon dioxide. This degradation product, such as formaldehyde, reacts with acrylamide, which leads to the decomposition of acrylamide to small molecular fragments. Then, acrylamide can be transformed to acrylate which is susceptible to consecutive decarboxylation and further total oxidation by the catalytic processes of degraded products of glycine (Cui et al., 2005). This scavenging reaction is similar to the catalytic total oxidation of acrylamide in the presence of water via the formation of acrylate (Hawrylak and Szymanska, 2004). As for plant-derived proteins, soy protein hydrolysate can also be used to reduce acrylamide due to the fact that soy protein hydrolysate is believed to reduce acrylamide by introducing additional amino acids to compete with asparagine for key reaction intermediates (Cook and Taylor, 2005; Wedzicha et al., 2005).

Use of Hydrogencarbonates

The effect on the acrylamide reduction by hydrogencarbonates is complex (Amrein et al., 2004a; Biedermann and Grob, 2003). Studies on elimination and net formation of acrylamide caused by sodium hydrogencarbonate (NaHCO^sub 3^) showed that three different addition levels had a similar effect on elimination. Net acrylamide increased somewhat as bicarbonate dropped to near 1%, but remained significantly suppressed compared with the control sample. Potassium hydrogencarbonate (KHCO^sub 3^) had an inhibitory effect similar to the sodium compound except that the net formation appeared to be slightly higher. Conversely, ammonium hydrogencarbonate (NH^sub 4^HCO^sub 3^) with the same addition levels as NaHCO^sub 3^ and KHCO^sub 3^ could dramatically enhance the acrylamide concentration (Levine and Smith, 2005). The interpretation of great differences among three bicarbonates is also complicated by the change in the temperature curve known to accompany bicarbonate addition. Nevertheless, such phenomena are in agreement with the others in finding that samples baked with NH^sub 4^HCO^sub 3^ have more acrylamide than samples baked with NaHCO^sub 3^ or KHCO^sub 3^ (Levine and Smith, 2005). Another variable is the relatively high volatility of NH^sub 4^HCO^sub 3^ upon heating compared with the sodium or potassium salt.

Use of Antioxidants

The relationship between the antioxidant effects and acrylamide reduction deserves to be discussed. During these years, many correlative tests have been performed and positive or negative effects on acrylamide reduction have been demonstrated by using different kinds of antioxidants. Tareke (2003) found that the addition of antioxidants (BHT, sesamol, and Vitamin E) to meat prior to heating enhanced the formation of acrylamide, probably by protection of acrylamide against free radical initiated reactions. Meanwhile, decreased acrylamide formation when adding rosemary extracts to the oil used for frying potato slices has been found (Taeymans et al., 2004). Zhang et al. (2005b) demonstrated that the addition of antioxidant of bamboo leaves (AOB) could effectively reduce acrylamide in various heat-treated foods and also summarized the relationship between the inhibitory rates and the addition level of AOB. AOB, a new flavone-rich extract, has been recently certificated as a novel kind of natural antioxidant for the application of food additives by the Ministry of Health, P.R. China. Shortly after such findings, the results from similar tests showed that AOB combined with other plant-derived extracts such as ginkgo biloba extracts, tea extracts, grape seed extracts, etc., could also reduce acrylamide to a different extent. The selected plant-derived extracts also included some antioxidants such as rosemary extracts and liquorice extracts (Zhang et al., 2005e). Furthermore, relatively lower amounts of acrylamide after the addition of a flavonoids' mix have also been reported by Ferna;ndez et al. (2003). A liquid flavonoids' mix was added to potato slices before frying, and a powder mix was also added to the potato slices after frying. After a 4-day incubation time, the acrylamide levels were detected to be reduced by up to 50% in the flavonoids' mix treatment (Kurppa, 2003). Biedermann et al. (2002b) found a relatively weak reduction of the acrylamide formation by the addition of ascorbic acid to a potato model. Moreover, similar results were obtained by Levine and Smith (2005) when using ascorbate as the additive. Hitherto, it is still difficult to obtain affirmatory conclusions on either positive or negative relationship between the addition of antioxidants and the reduction of acrylamide. Summa et al. (2006) demonstrated a direct correlation that was found between the concentration of acrylamide and the antioxidant activity. However, as for the addition of exogenous antioxidants, the mechanism on positive or negative effects has not been found so far.

Final Preparation

The Maillard reaction, which leads to the generation of acrylamide, also produces the colors and flavors which give baked cereal products their essential characteristics. The color, texture, flavor and shelf-life of products also affect the acrylamide level. In general, acrylamide could theoretically be reduced in lighter colored and less baked products (CIAA, 2004; Taeymans et al., 2004). However, in some cases a darker color may be associated with less acrylamide, e.g. digestive biscuits and some breakfast cereals (Taeymans et al., 2004). In bread, the endpoint color does in most cases reflect the acrylamide content (Surdyk et al., 2004). As for the texture and flavor of products especially with the addition of exogenous chemical additives to prepare the heat processing foods, it is unfortunate that the Maillard reaction also develops flavor and color. In some products (e.g. gingerbread) reducing sugars, such as glucose or fructose, are deliberately added so as to achieve particular flavors (and color). Such products also tend to be higher in acrylamide. Nowadays, many researches (Jung et al., 2003; Pedreschi et al., 2004) have found effective ways to reduce acrylamide during heat processing, some of which were also performed via immersion, but their sensory evaluation was not reported or not very reasonable, even not acceptable. For instance, the largest decrease of the acrylamide content (90%) in crisps was obtained when potato slices were soaked in acetic acid solution for 60 min at 20[degrees]C, and a large decrease of acrylamide content (74%) was also observed after soaking of the potato slices in 1% NaOH solution. However, a sour and acerbic taste from both of the treatments greatly influenced the appearance as well as the taste and flavor of crisps, which were not sensorially acceptable (Kita et al., 2004). Baked cereal products have a shelf-life ranging from a few days for bread to nine months or more for some biscuits and breakfast cereals. Some baked cereal products have a shelf life of several months, typically at least nine months. Since it is known that baking at a high temperature and/or to lower moisture content generally increases formation of acrylamide, it follows that higher moisture and therefore shorter shelf life products will tend to be lower in acrylamide. They will also be softer textured and much less crisp and more prone to mould growth. The acrylamide concentration in cereal-based products does not change during shelf-life (Delatour et al., 2004; Hoenicke and Gatermann, 2005). Besides the above- mentioned reducing methods, the elimination of acrylamide could also be achieved via lactic acid fermentation (Baardseth et al., 2006), use of asparaginase (Hendriksen et al., 2005), coating with egg/ breadcrumbs (Fiselieret al., 2004), addition of trehalose (Oku et al., 2005), etc.

Mitigation Studies by the CIAA "Toolbox" Approach

The CIAA "Toolbox" reflects the results of more than three years of industry cooperation to understand acrylamide formation and potential mitigation steps. Its aim is to provide brief descriptions of the mitigation steps evaluated and, in many cases, already implemented by food manufacturers and other partners in the food chain (CIAA, 2005). This approach allows individual manufacturers, including also small and medium size enterprises with limited research and development resources, to assess and evaluate which of the mitigation steps identified so far may be helpful to reduce the acrylamide formation in their specific manufacturing processes and products. It is important that they assess the suitability of proposed mitigation steps in light of the actual composition of their products, their manufacturing equipment, and their need to continue to provide consumers with quality products consistent with their brand image and consumer expectations. It is anticipated that some of the tools and parameters will also be helpful within the context of domestic food preparation and in catering establishments, where stringent control of cooking conditions may be more difficult.

Table 2 Summary of studies on the reduction of acrylamide and major contributions from various laboratory colleagues by the CIAA "Toolbox" approach

Table 2 Summary of studies on the reduction of acrylamide and major contributions from various laboratory colleagues by the CIAA "Toolbox" approach

Table 2 Summary of studies on the reduction of acrylamide and major contributions from various laboratory colleagues by the CIAA "Toolbox" approach

Essentially, the summaries describing the acrylamide reduction tools developed by industry are intended to be generic. This is necessary to take account of the differences between product recipes, designs of processes and equipment, and brand-related product characteristics even within a single product category. As for the definition of CIAA "Toolbox" parameters, a total of 13 additional parameters grouped within the four major "Toolbox" compartments have now been identified, which include the agronomical (sugars and asparagine), recipe (ammonium bicarbonate, pH, minor ingredients, dilution and rework), processing (fermentation, thermal input and pretreatment) and final preparation (color endpoint, texture/flavor and storage/shelf-life/consumer preparation). These parameters can be selectively applied by each food producer in line with their particular needs and product/process criteria. In addition, the stage at which the different studies have been conducted, i.e. laboratory, pilot, or in a factory setting (industrial), are aligned to the potential mitigation measures. This approach ensures that all pertinent tests and studies are captured independent of their (immediate) applicability.

The expanded "Toolbox" is not meant as a prescriptive manual nor formal guidance. It should be considered as a "living document" with a catalog of tested concepts at different trial stages that will be updated as new findings are communicated. Furthermore, it can provide useful leads in neighboring sectors such as catering, retail, restaurants, and domestic cooking. The final goal is to find appropriate and practical solutions to reduce the overall dietary exposure to acrylamide. During these years, especially in the past two years, many mitigation studies on acrylamide have been reported. To review the research progress on acrylamide reduction and identify the key parameters across the food chain that may impact the formation of acrylamide, the CIAA "Toolbox" app