Dietary Acrylamide and the Risks of Developing Cancer: Facts to Ponder

Acrylamide (AA) is a water soluble white crystalline solid commonly used in industries. It was listed as an industrial chemical with potential carcinogenic properties. However to date, AA was used to produce polyacrylamide polymer, which was widely used as a coagulant in water treatment; additives during papermaking; grouting material for dams, tunnels, and other underground building constructions. AA in food could be formed during high-temperature cooking via several mechanisms, i.e., formation via acrylic acid which may be derived from the degradation of lipid, carbohydrates, or free amino acids; formation via the dehydration/decarboxylation of organic acids (malic acid, lactic acid, and citric acid); and direct formation from amino acids. The big debate is whether this compound is toxic to human beings or not. In the present review, we discuss the formation of AA in food products, its consumption, and possible link to the development of any cancers. We discuss the body enzymatic influence on AA and mechanism of action of AA on hormone, calcium signaling pathways, and cytoskeletal filaments. We also highlight the deleterious effects of AA on nervous system, reproductive system, immune system, and the liver. The present and future mitigation strategies are also discussed. The present review on AA may be beneficial for researchers, food industry, and also medical personnel.


FORMATiON OF AA iN FOOD PRODUCTS
Later in the year 2002, AA was detected in heated foods where its formation was temperature dependent (6). Using liquid chromatography-mass spectrometry, moderate levels of AA (5-50 µg/kg) were detected in heated protein-rich foods, but higher contents of AA (150-4,000 µg/kg) were detected in carbohydrate-rich foods (6). Importantly, AA could not be detected in unheated or boiled foods (6). AA in food could be formed during high-temperature cooking via several mechanisms, i.e., formation via acrylic acid which may be derived from the degradation of lipid, carbohydrates, or free amino acids; formation via the dehydration/decarboxylation of organic acids (malic acid, lactic acid, and citric acid); and direct formation from amino acids (7).
Studies have shown that AA is formed mainly from free amino acid asparagine and reducing sugars such as glucose and fructose, during high-temperature cooking through Maillard reactions, a series of non-enzymatic reactions between free amino acids and reducing sugars which is responsible for the flavor and color generated during baking (8,9). This would explain the formation of AA in cooked food rich in asparagine, e.g., in cereals and potatoes (8). In fact, the concentration of reducing sugars in food is the primary determinant of AA formation, compared to asparagine content (10).
The mean AA contents in potato crisps prepared from different UK-grown potatoes ranged from 131 to 5,360 µg/kg (11). Rosti, a popular Swiss dish, made of grated and fried potatoes, contains an average of 702 µg/kg AA (12). The AA concentration in different commonly consumed breads ranges from <limit of quantification (LOQ) to 695 µg/kg, where the highest AA concentration was detected in wheat bran and whole wheat breads (13). Similarly, AA was also detected in different bakery products: biscuits (LOQ to 2,405.0 µg/kg), sandwich biscuits with cream (112.6-570.4 µg/kg), gingerbread (349.5-955.5 µg/kg), and crackers (347.8-366.1 µg/kg) (14). In addition to potato-and cereal-based food, coffee beverages acquired from coffee vending machines also contain AA concentration which varies from 7.7 to 40.0 µg/L (15).
Estimated AA intakes were measured using food frequency questionnaire, total diet study, hemoglobin adducts of AA, and glycidamide (GA, primary metabolite of AA) (16,17). For adults, the estimated average AA intakes range from 0.3 to 0.6 µg/kg of body weight per day, where the intake is higher in children (18).
Infant cereal-based foods are important nutrient source for children around the world, and the AA concentration in these cereal-based baby food (ready-to-eat and instant baby foods, candy bars, and cakes) varied between 10 and 60 µg/kg (19). In another study, the mean AA level in cereal-based baby foods ranged from 36 to 604 µg/kg, and it was estimated the mean AA exposures of toddlers from the cereal-based baby food was 1.43 µg/kg of body weight per day (20).

AA CONSUMPTiON AND CANCeR
High-income countries have the highest incidence rates for lung, colorectal, breast, and prostate cancers, while low-and middleincome countries have the highest rates of stomach, liver, and esophageal and cervical cancer (21). Global Burden of Disease Cancer Consortium et al. (22) has reported that there were a total of 17.5 million cancer cases globally, 8.7 million deaths, and 208.3 million disability-adjusted life-years, in the year 2015. Between the year 2005 and 2015, incident cancer cases were found to be increased by 33% (22).
Few studies reported possible association between dietary AA intake and cancer risk, e.g., dietary AA may increase the risk for cutaneous malignant melanoma in men (HR: 1.13, 95% CI: 1.01-1.26) per 10 µg increment (33). Similarly, dietary AA may also increase the risk of lymphatic malignancies (i.e., multiple myeloma and follicular lymphoma) in men (34). The adjusted risk of all esophageal tumors combined among Sweden esophageal cancer patients was in the highest quartile of AA intake compared to the lowest, particularly among overweight or obese individuals (35). However, in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, an association between dietary AA intake and an increased risk of developing esophageal cancer was observed in the middle quartiles, not in the highest quartile (36). A recent study revealed dietary AA intake was associated with increased overall cancer mortality in elderly Chinese (37). When compared to the lowest (<9.9 μg/day) and the highest (>17.1 μg/ day) quartile of AA intake, the multivariable hazard ratios were 1.9, 1.9, and 2.0 for the overall, digestive, and respiratory system cancer mortality, respectively (37).

BODY eNZYMATiC iNFLUeNCe ON AA
In human and animals, AA is primarily metabolized to an epoxide metabolite, GA, by an enzyme known as CYP2E1 (38,39). GA may subsequently conjugated with reduced glutathione (GSH) to GSH conjugates by GSH transferases, or GA also can be further metabolized to glyceramide by epoxide hydrolase.
GA PRODUCT OF CYP2e1: ReLevANCe iN POTeNTiAL CARCiNOGeNiC eFFeCT OF AA Using CYP2E1-null and wild-type mice, Sumner et al. (40) showed that the absence of CYP2E1 enzyme in the mutant mice led to non-excretion of GA or its metabolites in the urine of 13 C-AA-treated mice. Subsequent studies by Ghanayem et al. (39) showed that CYP2E1-mediated oxidation of AA to GA is required for the induction of male germ cell mutagenicity in mice exposed to AA. In their subsequent study, the researchers

GSH, GST, AND AA
GSH is an important antioxidant that prevents cellular damage caused by reactive oxygen species such as peroxides, free radicals, heavy metals, and lipid peroxides (42). GST is a group of enzymes that mainly involved in detoxifying xenobiotics, cell death, and cell proliferation (43). GSH is employed by GST in numerous cellular reactions. Thus, the enzymatic activities of GST depend upon the availability of GSH. Srivastava et al. (44) investigated the effect of single and repeated administration of AA on GSH content and GST activity in rat brain. The researchers reported single administration of AA to deplete GSH content without affecting GST activity. However, repeated administration of AA (50 mg/ kg × 10 days) depleted GSH content as well GST activity in the rat brain. Altered GST activity may affect the redox status of the cell. This may subsequently affect various redox-dependent gene expression, cell transformation or proliferation, or even apoptosis may occur (45)(46)(47)(48).

HORMONAL MeCHANiSM OF ACTiON OF AA iN THYROiD TUMOR iN MALe RATS
Exposure of male F344 rats to AA (50 mg/kg) in drinking water for 14 days resulted in a significant decrease in serum T4 level (49). In contrary, exposure to lower doses of AA (2.5 or 15 mg/kg) had no significant effect on serum T4 or thyroid-stimulating hormone (TSH) level in male F344 rats (50). This led to the suggestion that hormonal mechanism is likely to attribute to thyroid tumorigenesis, at least at high doses of AA (51). Hormonal mechanism is thought to be influenced by decreased circulating levels of T3 or T4 levels, which would subsequently stimulate the release of TSH by the pituitary gland via hypothalamus-pituitary-thyroid axis. Continuous stimulation of the thyroid gland by TSH results in increased proliferation of cells and so the formation of tumor (52). A schematic diagram showing the mechanism of action of AA in thyroid tumor was highlighted in Figure 1.

CALCiUM SiGNALiNG-CYTOSKeLeTAL MODe OF MeCHANiSM
A recent study correlated calcium signaling and cytoskeletal filaments to genotoxic effects of AA in testes using F344 rats (53).
Similar to this finding, existing literature implicates calcium signaling with several types of cancers (54). Binding of AA toward redox-sensitive Cys residues on key regulatory proteins initiate disruption of calcium signaling genes which could lead to reduced expression of calcium signaling genes and disruption of calcium homeostasis. Reduced expression of calcium signaling genes may destabilize microtubules and microfilaments which carry out endo-and excytotic functions (Figure 2). It is hypothesized that disruption in calcium signaling also could lead to cancer development via hormonal actions. However, lack of empirical proof warrants his hypothesis to be further investigated (55).

CYTOSKeLeTAL-AA-GeNOTOXiC eFFeCTS
The idea of cytoskeletal proteins mediating the genotoxic effects of AA were implicitly corroborated using various study models (55,56). Cytoskeletal proteins are tasked with multitude of functions that integral for cell survival such as movement of vesicles and cell compartments, regulation of cell division (57), and cell mechanics (58). Few studies reported AA to bind directly to cytoskeletal proteins (59)(60)(61). Given the direct functional relation between the cytoskeletal proteins and cell divisions (mitosis/meiosis), AA's disruptive effects pose various adverse effects, including delays in cell cycle (62), heritable translocation of chromosomes (63), chromosomal aberrations (64), blocks in mitotic and meiotic process (59,65), and aneuploidy (66). The primary targets of AA are kinesin motors which are proteins critically involved in cell divisions (67). Kinesins are motor proteins that move along microtubules, fueled by hydrolysis of ATP. Movement of kinesins aids various cellular functions including the transport of cellular cargo from central toward cell periphery. As tailored to their functions, the most common feature of kinesin is presence of motor domain for microtubule binding and ATPase activities. Kinesins are categorized as N (N-terminal), C (C-terminal), and M or I-type (middle or also known as non-motor) based on the location of motor domain within the macromolecule (68). AA was shown to inhibit kinesin in neurons (60) and testes (61). In 2007, Sickles and colleagues investigated the effects of AA, glycidamide (AA's toxic metabolite), and propionamide (non-neurotoxic metabolite) on function of specific kinesins, KIFC5A (member of kinesin-14 subfamily with motor domain located at the carboxy terminus of the polypeptide), and KRP2 (member of kinesin-12 subfamily with motor domain located more toward the center of the polypeptide). Generally, KIFC5A stabilizes spindle pole by crosslinking the adjacent microtubules (69,70) and KRP2 regulates microtubule dynamics and chromosome segregation during anaphase (71). The researchers reported AA and glycinamide to specifically inhibit both KIFC5A and KRP2 kinesins (61). Their findings were similar to their previous findings where AA and glycinamide inhibited neuronal N-type kinesins (60). Despite their findings which were convincing at in vitro level, the authors concluded that the carcinogenicity of AA and glycinamide were inconclusive and this warrants future studies to further elucidate the conundrum.  The neurotoxic effects of AA is the only toxic effect proven to be harmful on humans via occupational exposure (72,73). The clinical signs of AA neurotoxicity were peripheral neuropathies such as upper and lower limb numbness and tingling (74). Chronic administration of AA to laboratory animals resulted in skeletal muscle weakness and ataxia (72,75,76). Numerous excellent reviews have covered in-depth literature on the neurotoxicity effects of AA (73,77). The mechanism underlying the neurotoxicity of AA is inconclusive. However, several mechanisms were proposed to mediate the neurotoxic effects of AA. Interaction of AA with kinesin motor protein in neurons impaired fast anterograde transport of nerve growth factors from cell body to periphery, resulting in nerve death (60). AA was also shown to inhibit uptake of neurotransmitter into striatal synaptic vesicle through possible interaction with sulfhydryl groups on specific proteins and thereby disrupting presynaptic release of neurotransmitters (78).

Reproductive Toxicity of AA
To date, reproductive toxicity of AA in humans is unheard of. In laboratory animals, exposure to high levels of AA proven to cause reproductive toxicity (79,80). Various concentration range of AA and duration of AA exposure have been experimented on the reproductive toxic effects of AA in laboratory animals which resulted in decreased fertility rates, decreased litter sizes, abnormal sperm, decreased sperm counts, reduced copulatory activity, and decreased weight of testes (81)(82)(83). Further corroborating the reproductive toxicity of AA, Yilmaz et al. (84) investigated the effects of AA and GA exposure on mouse Leydig and Sertoli cells. The researchers showed exposure to AA and GA to reduce cell viability and increase oxidative stress and apoptosis in both Leydig and Sertoli cells (84). Several theories were proposed to delineate the mechanistic effects of AA on reproductive toxicity. One of the theories relate the neurotoxic effects of AA to mating behaviors. AA-induced peripheral neuropathies such as reduced hind-limb function could impair copulatory behavior, mounting responses, and intromission (83), subsequently affecting proper deposition of sperm in the vagina and uterus and also the resulting hormonal changes FiGURe 2 | A schematic depiction correlating calcium signaling and cytoskeletal filaments to genotoxic effects of acrylamide (AA) in rats' testes. AA binds toward redox-sensitive Cys residues on key regulatory proteins such as transcription factors, which disrupts the protein's function. Loss of the transcription regulator's function causes reduced expression of calcium signaling genes and disruption of calcium homeostasis. Reduced expression of calcium signaling genes also destabilizes microtubules and microfilaments which carry out endo-and excytotic functions. (85). AA-induced inhibition of kinesin motor proteins, which are also found in flagella of sperm in addition to nervous system, could affect sperm motility and fertilization events (80,86,87). Regarding hormonal mechanism of action, AA reduced serum testosterone and prolactin levels (88) which could lead to testicular atrophy, affects sperm development and motility (81,89).

immunotoxicity of AA
The literature on the immunotoxic potential of AA is limited. In rats, exposure to AA significantly decreased the weight of spleen, thymus, and mesenteric lymph nodes (90). Orally administered AA reduced alpha-naphthyl acetate esterase-positive lymphocytes counts and also caused lesions in ileal Peyer's patches in a dose-dependent manner (91). In other words, AA was detrimental toward gut-associated lymphoid tissue and peripheral blood lymphocytes in rats (91). Fang et al. (92) reported decreased terminal body weight, thymus and spleen weights, and lymphocyte counts following exposure to AA in female BALB/c mice. The researchers also reported significant reduction in the percentage of natural killer cells, interleukin-6, and concanavalin A-induced splenocyte proliferation (92). Kim et al. (93) demonstrated AA-induced senescence in C57BL/6 male mice macrophages via ATF3-mediated pathway. The researchers also associated the AA-induced senescence to ROS production, activation of p38 and JNK signaling pathways, and also increased expression of p53 ( Table 1) (93).

Hepatotoxicity of AA
Similar to the reproductive toxicity, there were no reports of AA hepatotoxicity in humans, although AA metabolism takes place in the liver. However, numerous studies have reported the harmful effect of dietary AA in the liver of experimental animals due to oxidative stress (95,96). A high dose of 25 mg/kg AA administrated for 21 days resulted in significant decrease of liver GSH level and total antioxidant status in experimental adult rats (95). AA administration also lead to decrease of liver enzymes (e.g., aspartate aminotransferase, alanine aminotransferase, and alkaline phosphatase), superoxide dismutase, catalase activities, while total oxidant status and malondialdehyde levels increased (95). Moreover, the total cholesterol, triglyceride, and low-density lipoprotein cholesterol levels were increased, while high-density lipoprotein cholesterol was decreased following AA administration (96). Waterborne exposure of zebrafish in water containing AA (300 ppm) resulted in acute death within 26 h along with elevation of blood glucose, triglyceride, hepatic inflammation, and fat accumulation in the liver (94).

AA POiSONiNG: PReSeNT AND FUTURe MiTiGATiON STRATeGieS
In the year 2005, European Food Safety Authority and the Joint Food and Agriculture Organization of the United Nations     (FAO)/World Health Organization committee recommended to employ "margin of exposure (MOE)" approach to evaluate the risk assessment of AA (7,97). The numerical value of MOE represents the ratio between the dose-response curve results in tumors in experimental animals and human dietary intake. Based on the modified risk assessment, the MOE value of AA showed the chemical to impose a severe human health concern compared to other chemicals, even the likes of benzopyrene (98). Numerous strategies were shown to be able to inhibit or mitigate AA formation, both in model systems and actual food systems. These strategies include pH modification, decreasing the heating temperature and time, choosing raw materials with fewer precursors, adding the different exogenous additives, and process technology interventions (e.g., acids, amino acids, hydrogen carbonates, proteins, or antioxidant) (99,100).
The AA formation in bread are significantly influenced by the yeast amount, fermentation time, fermentation temperature, and yeast types used in bread baking (101). Treatment with asparaginase produced from Cladosporium sp. (to convert the precursor asparagine to aspartic acid) to the wheat-based dough were able to reduce 97 and 73% of AA formation in the crust and crumb regions of the bread (102). Similarly, the use of asparaginase was able to cause 90% reduction in AA content after using a combination of asparaginase treatment and blanching of potato slices (103). In a separate study, asparaginase was also experimented on chilled French fries which reduced the AA content in the fries up to 90% without affecting final taste of the products (104).
Numerous studies used antioxidants as additives to inhibit AA formation. The inhibition of AA formation was in correlation with antioxidant activity of the additives (105). The addition of flavonoids significantly inhibited AA formation during microwave heat processing, where the inhibition strength were shown to be related to the number of phenolic hydroxyls of flavonoids (105). In addition, phenolic antioxidants also demonstrated an inhibiting effect in AA formation, with the most significant reductions (≈60%) observed for caffeoylquinic acids (100). Alternatively, addition of natural herbal extract also demonstrated inhibition of AA formation. Addition of antioxidants of bamboo leaves (AOB) and acylated AOB were able to reduce AA formation in fried potato crisps ranging from 30.7 to 46.9% (106). Similarly, the addition of green tea extract into burgers and nuggets coating were able to reduce AA formation (107).

SUMMARY
The research studies conducted on AA described the potential health risk of fried and baked carbohydrate-rich foods. There is clear evidence to show that AA is carcinogenic in rodents. However, there is paucity of facts depicting the health risk in humans as evident from numerous epidermiological and toxicological studies. There are lacunae in the past studies, especially with regard to lack of findings in estimating dietary AA intake through questionnaires and AA content database in foods, lack of repeated exposure estimations, and lack of statistical power to detect small increases in risk. Nevertheless, it is important that efforts should continue to reduce AA levels in food products. Further studies are suggested in human beings to show the detrimental effect of AA.

AUTHOR CONTRiBUTiONS
Conceptual framework and design: SD. Searched references: ST and JK. Drafted manuscript: ST, SD, and JK. Critically revised the manuscript: ST, SD, and JK.