Simultaneous mitigation of 4(5)‐methylimidazole, acrylamide, and 5‐hydroxymethylfurfural in ammonia biscuits by supplementing with food hydrocolloids

Abstract In the current work, methodological approach for the incorporation of food hydrocolloids gum Arabic (GA), pectin, and carboxymethylcellulose (CMC) in biscuit dough was firstly investigated to mitigate simultaneously the formation of 4(5)‐methylimidazole (4(5)‐MI), acrylamide (AA), and 5‐hydroxymethylfurfural (5‐HMF) in ammonia biscuits. Results revealed that the percent inhibition of 4(5)‐MI was ranged from 50.5% to 89.9% by increasing the GA amount from 0.01 g to 0.05 g, respectively. Furthermore, the use of 0.05 g GA reduced significantly AA content up to 58.6% compared to the control biscuits. Moreover, the highest inhibition of 5‐HMF with 74% depression was achieved by 0.05 g GA. Reasons could be referred to the formation of GA layer on the surface of biscuits. This was confirmed by scanning electron microscope and water loss analysis. Additionally, browning intensity and sensory analysis showed that the supplementation with GA had preserved the quality and consumers’ overall acceptability of ammonia biscuits.

In the last few years, several works confirmed the formation of 4(5)-MI, AA, and 5-HMF in ammonia biscuits after baking at 180 ᵒC or 200 ᵒC for a period within 10-20 min. During this process, sucrose thermally degrades into free glucose and fructose which subsequently reacts with ammonium bicarbonate producing 4(5)-MI in levels ranged from 70 to 1,260 μg/kg (Jung et al., 2017).
Furthermore, AA forms initially by reacting free asparagine (as a component of wheat flour) with fructose during baking time producing the Schiff base that underwent decarboxylation to form AA in contents ranged from 10 to 200 μg/kg (Mogol & Gökmen, 2016).
However, AA content is very sensitive to free asparagine content in wheat flour which mainly depends on its growing and fertilizing conditions. AA could be produced with levels from 600 to 900 μg/ kg in wheat cultivated under normal sulfate fertilization; however, more levels of AA between 2,600 and 5,200 μg/kg were also obtained in sulfate-deprived wheat after heating at 160ᵒC for 20 min (Muttucumaru et al., 2006). On the same hand, in ammonia biscuits, the enolization of glucose is known as the de Bruijn van Eckenstein rearrangement followed by dehydration or β-elimination of water giving 3-deoxyhexosulose and further its intramolecular cyclization concurrently with its dehydration produced 5-HMF (Román-Leshkov et al., 2006). It was also reported that the decomposition of sucrose during baking of biscuits released a free glucose and fructofuranosyl cation which rapidly dehydrated to form 5-HMF at 550-3530 μg/kg (Locas & Yaylayan, 2008;Mogol & Gökmen, 2016).
As 4(5)-MI, AA, and 5-HMF toxicants formed spontaneously during the baking of ammonia biscuits, some studies were recently conducted to reduce the formation of these compounds by adding compounds such as iron sulfate, magnesium sulfate, zinc sulfate, tryptophan, cysteine, formic acid, and chitosan (Mogol & Gökmen, 2016;Seo, Ka, & Lee, 2014). It was found that iron sulfate awarded the greatest reduction (about 80%) for 4(5)-MI in biscuits.
Otherwise, formic acid decreased significantly the initial rate of AA formation in biscuits to about four times. The 5-HMF content was slightly decreased in the presence of acids due to its acidic-catalyzed degradation. However, these additives in ammonia biscuits could significantly affect their sensorial quality.
Food hydrocolloids gum Arabic (GA), pectin, and sodium carboxymethyl cellulose (CMC) are commonly used as negatively charged water-soluble polymers and used in low amounts as food additives (Mousa, 2018). These food hydrocolloids had distinguished properties coming from high solubility in water, rigid thin film formation or emulsification and biocompatible. Additionally, they could have antioxidant properties attributed to the existence of stable amino acids in their building structures. Therefore, food hydrocolloids have the ability to form tight thin layer with antioxidant merit during the initial stages of baking for ammonia biscuits. Subsequently, its stable surface structure on biscuits could significantly inhibit or interfere with the factors accelerating the formation of toxicants.
However, there is scarce information about the effect of use of food hydrocolloids on the quality of ammonia biscuits. Therefore, the current work aimed to investigate the influence of incorporation of three types of food hydrocolloids GA, pectin, and CMC with variable concentrations on the simultaneous mitigation of carcinogenic compounds 4(5)-MI, AA, and 5-HMF in ammonia biscuits.
Furthermore, the quality and sensorial acceptability of new biscuit formulations was also investigated. The obtained results would be of great benefit to home workers, researchers, and food industry.
(Tustin, Canada). All solid-phase cartridges and columns were supplied by Waters Corporation (Ireland). Millipore water system (Millipore, Billerica, MA) was used to prepare the deionized pure water.

| Preparation of ammonia biscuits
For the preparation of biscuits dough, 20.0 g of palm oil was creamed with 35.0 g powdered sucrose, 1.0 g NaCl and 0.8 g of NaHCO 3 for 3 min at 60 rpm using a dough-mixer Artisan Kitchen Aid 5KSM150 (MI, USA). For each mixing step, the remained mixture was scratched of the bowl sides in order to retain the ingredients as much as possible. Then, 0.4 g of NH 4 HCO 3 and 17.6 ml of water were added and mixed for 3 min at 150 rpm. After that, 80.0 g of refined wheat flour was added slowly to the mixture and was again mixed for 2 min at 80 rpm to obtain homogenous dough (control biscuit). For hydrocolloids containing biscuits, water in the control biscuits was replaced with variant concentrations of GA, pectin, and CMC dissolved in water (nine treated biscuit samples as shown in Table 1), and subsequently, the other steps were followed as described above. Then, the obtained dough in each sample was rolled in 3 mm thickness using Atlas Brand rolling machine, and subsequently, the sheeted dough was cut into round shape using a 50-mm-diameter cutter. Finally, the obtained dough pieces were baked on an aluminum plate in electric oven (Memmert UNE 400, Germany) at two different temperatures 180°C and 200°C for 10 min. Under these conditions, all the proposed biscuit samples had a fully baked crumb without unacceptable too over-baked conditions. All the obtained biscuits were cooled for 30 min, were packed in low-density polyethylene bags, and then were stored under desiccation prior to subsequent chemical analysis.

| Proximate and structural evaluation of ammonia biscuits
Proximate analysis of proposed baked ammonia biscuits was based on the standard method of AOAC (AOAC, 1995) including moisture, crude fat, crude fiber, crude protein, and ash content. Total carbohydrate was calculated from the sum of moisture, crude fat, crude protein, ash, and crude fiber, and lastly subtracting it from 100. The caloric value was calculated using values of 4.0 k.cal/g of protein, 4.0 k.cal/g of carbohydrate, and 9.0 k.cal/g of fat as described elsewhere (Livesy, 1995).
The microstructure of all studied baked biscuits was investigated by scanning electron microscopy (SEM) after mounting on individual metallic stubs and sputtering with a Balzers SCD 030 conductive coating of gold (Balzers Union LTD. Liechtenstein). The images of all studies were obtained by a JEOL6060LV variable pressure SEM instrument (Jeol (UK) Ltd.). Electron micrographs were produced for cross section of each biscuit type at several different magnifications.

| Determination of 4(5)-MI in ammonia biscuits by GC/MS
The analytical method reported in reference (Jung et al., 2017) was followed to analyze 4-MI in biscuits with little modifications. Two grams of biscuits was dissolved in a mixture containing 4 ml of phosphate buffer (pH 6.6) and 8 ml of 0.1 mol/L BEHPA in chloroform.
Then, the sample solution was centrifuged at 8,000 rpm for 7 min followed by filtration of bottom layer (chloroform phase) through Whatman® No. 2 filter paper. Four milliliters of 0.1 mol/L HCl was added to 6 ml of filtrate followed by shaking for 6 min. Centrifugation of the obtained solution was performed at 3,131 × g for 5 min. Then, 1 ml of aqueous phase (upper layer) was transferred to another vial followed by adding 1 ml of a mixture of acetonitrile, isobutyl alcohol, and pyridine (50:30:20, v/v) and 120 μL of IBCF. After that, 500 μL of 1.0 mol/L sodium bicarbonate solution and isooctane was also added to the mixture and was shaken by hand for 5 s. Finally, 2 μL of the top layer was injected into the GC-MS instrument in the splitless mode at a temperature of 275°C (Agilent 7820A gas chromatograph and 5977E mass spectrometer, Palo Alto, Calif., USA). The ion source and quadrupole were set at 235 and 160°C, respectively. An electron impact at 70 eV was used for the mass spectrometer in the selective ion monitoring mode. A HP-5MS column (30 m × 0.25 mm, 0.25 μm) was used as the stationary phase. Helium was selected as a carrier gas at 1.0 ml/min flow rate. The GC oven temperature program was set in steps starting from 80°C (1 min), ramped to 280°C at 30°C/min, and held for 2 min at 280°C. Ions at m/z 81, 82, 109, and 182 were selected for 4(5)-MI detection, and m/z 182 was chosen for its quantification. A stock solution of 4(5)-MI was prepared in 0.1 mol/L HCl to a concentration of 1 mg/ml. Working solutions were prepared by diluting the stock solution to the concentrations between 50 and 3,000 ng/ml. The limit of detection (LOD, 3.3 SD/ slope) and the limit of quantitation (LOQ, 10 SD/slope) of 4(5)-MI were 19.0 ng/ml and 53.0 ng/ml, respectively. All 4(5)-MI measurements were repeated three times.

| Determination of AA in ammonia biscuits by LC/MS/MS
AA was extracted from ammonia biscuit formulations by following the procedure as described elsewhere (Mogol &Gökmen, 2016, andMousa, 2018). An amount of each biscuit formulation (1.0 gram) was dissolved in a solution mixture containing 9.0 ml of 10 mmol/L formic acid, 0.5 ml of Carrez I, and 0.5 ml of Carrez II. The supernatant of each sample solution was collected by centrifugation at 3,131 × g for 7 min. Then, the extractions were repeated two times using 5 ml of 10 mmol/L formic acid as extraction solvent followed by centrifugation to achieve a clear extract. Further cleaning up of the extract was TA B L E 1 The studied ammonia biscuit samples blended with variant concentrations of GA, pectin, and CMC food hydrocolloids Samples Usage levels (g/ 17.6 ml water)

GA
Pectin CMC Control 0.00 0.00 0.00 MOUSA carried out by passing 1.5 ml of the supernatant through a precon-

| Determination of 5-HMF in ammonia biscuits by HPLC
5-HMF was extracted from ammonia biscuit formulations by following the procedure as described elsewhere (Mogol & Gökmen, 2016).

| Water loss (WL) and browning intensity in ammonia biscuits
Water contents (WCs) of all studied biscuit samples were measured on the dried basis before and after baking process (AOAC, 2005). In these experiments, samples were put in metal plates and then were dried in a forced air oven at 180°C. The following equation is used for the calculation of water loss percent as the following: All results were expressed as mean values of the original samples percent (w/w, %) after three replications.
A baking contrast meter (Konica Minolta Colorimeter BC-10) was used for the determination of browning intensity in the studied ammonia biscuits (Nguyen et al., 2017). The contrast meter ranged from 0 for the darkest to 5.25 for the lightest color. Color measurement was repeated three times in the middle of each biscuit.

| Determination of glucose, fructose, and asparagine in wheat flour
In order to determine the amount of glucose and fructose in wheat flour, the procedure described in Nguyen et al. (2017)  In order to determine asparagine, a sample solution was prepared by weighing 1 g of flour in 25 ml of water. The sample solution was shaken for 2 min and incubated for 1 hr at 25°C followed by cooling at −20°C for 20 min and centrifuged at 4°C for 8 min with 8,000 rpm. Next, the separated supernatant was centrifuged again for 10 min followed by passing through a 0.45-µm filter and finally injected into LC/MS/MS system equipped with a negative ion mode.
A stationary phase of Hypercarb column (100 × 3 mm, 5 µm) was used. The mobile phase (A) was 100% water, and mobile phase (B) was 100% methanol. The gradient run started with 100% water for 6 min, and 100% methanol was set for 1 min. After that, the mobile phase was set back to 100% water. The flow rate was fixed at 0.4 ml/min. Working standards of asparagine were daily prepared by %Water loss WL = WC before baking−WC after baking ∕WC before baking x100 diluting the stock solution (1.0 mg/ml) in water to the concentration range between 1.0 and 500.0 ng/ml. The LOD and LOQ values of asparagine were 0.2 ng/ml and 1.0 ng/ml, respectively. The analysis of asparagine in wheat flour was performed once.

| Hedonic evaluation test
The proposed different types of ammonia biscuits were subjected to consumers' acceptability using hedonic test (Birol, Meenakshi, Oparinde, Perez, & Tomlins, 2015). The panelist is composed of students and collaborators that are usually consuming ammonia biscuits (n = 50). They are free from sinus problem or cold flue during evaluation period. Ten panelists involved daily for 4-day evaluation of 15 different samples distributed in random arrangements among evaluators at a time under "daylight" illumination and in isolated bowls. Control samples were initially presented for each evaluator.
Mean value of each sample was recorded after three evaluations.
Samples were served in white background paper plates which identified by random three digits. Panelists were then asked to write their degree of liking on a paper ballot with a 9-point hedonic rating scale (Mousa, 2018). As usual, color, appearance, aroma, texture, taste, and overall acceptability were evaluated within this test.
Evaluators instructed to clean their palate with water among sample evaluations.

| Statistical Analysis
The data analysis for each set was carried out after repeating measurements of each test three times in the presence of one control.
Mean values were expressed as means ± standard deviation. Minitab statistical software (USA) was used to apply the analysis of variance.
A significance level of p < .05 using the Tukey's range test was used for presenting the significance of differences among values.

| RE SULTS AND D ISCUSS I ON
Ammonia biscuit samples were currently used as real food complex matrices to test the effect of food hydrocolloids GA, pectin, and CMC at different low amounts on the spontaneous formation of 4(5)-MI, AA, and 5-HMF coming from Maillard and caramelization reactions during the baking process.

| Reduction of 4(5)-MI in the studied ammonia biscuits
The formation of 4(5)-MI in control and treated ammonia biscuits (Table 1) was investigated. In control ammonia biscuits, the amount of 4(5)-MI was found to be 505 and 1109 μg/kg by baking at 180 and 200°C, respectively, as shown in Figure 1 (A and B). The results showed that 4 (5) of carcinogenic 4(5)-MI. Therefore, the addition of 0.05 g/ 17.6 mL GA solution to the biscuit dough could significantly (p < 0.05) inhibit 4(5)-MI in the range from 89.9% to 93% after biscuit baking at 180 °C and 200 °C for 10 min.

| Reduction of AA in the studied ammonia biscuits
The AA content is very sensitive to the amounts of free asparagine (limiting factor) in wheat flour. The amount of free asparagine was measured in the wheat flour, and it was found to be 3020.6 mg asparagine/kg wheat flour. Furthermore, fructose and glucose were quantified in wheat flour and were found to be 1020.9 and 420.4 mg/ kg, respectively. In the previous study (Surdyk, Rosén, Andersson, & Åman, 2004), it is well known that added fructose increased AA content more than added glucose in heat-treated food products. Prior to the baking of ammonia biscuits, no AA was detected in the biscuit dough. These results also corroborate with the previous study (Nguyen, Fels-Klerx, Peters, & Boekel, 2016) in which the compound was detected after 10 min of baking in the ammonia biscuits. In the current work, results in Figure 2a showed that the amount of AA in control biscuit after baking at 180°C for 10 min was 604.3 μg/kg dw.
Rising of baking temperature up to 200 °C increased AA content up to 1931.4 μg/kg dry biscuits as shown in Figure 2b, which reflected the direct significant of baking temperature on rising the formation of AA. It is obvious that AA amounts formed in ammonia biscuits at all baking temperatures were higher than the recommended EU benchmark level 500 μg/kg in food (EU, 2017). Figure 2a,

| 3919
MOUSA pH value of GA solution (pH = 4.9) could be another factor to facilitate the reduction of AA formation in biscuits. However, the magnitude of GA influence on the mitigation of AA is lower than 4(5)-MI as described above. The reasons might be due to the molecular movements and interactions between the GA and AA precursors suffered from some limitations in the chemical composition of wheat flour.
The use of pectin and CMC at all concentration levels did not significantly affect the AA formation (p > .05) at all temperatures compared to the control biscuits as depicted in Figure 2a,b. Therefore, the addition of 0.05 g GA/ 17.6 mL water to the biscuit dough could significantly (p < 0.05) inhibit AA content lower than the EU-recommended level in food (EU, 2017) after biscuit baking at 180°C for 10 min.

| Water loss and browning intensity measurement in the studied ammonia biscuits
Changes in water loss (WL) and browning intensity were measured in the nine treated biscuits and compared with the control biscuits. The data revealed that the presence of GA resulted in significant changes (p < .05) in comparison with pectin and CMC after baking at 180°C.

| Proximate chemical composition and caloric value of ammonia biscuits
The mean values of proximate chemical composition and caloric values of biscuits supplemented with hydrocolloids after baking at 180°C for 10 min are given in Table S1. The data of moisture content, crude protein, and crude fiber content (g/ 100 g) revealed that the addition of GA to the biscuit dough showed a significant difference in measurements compared to the control biscuits. The addition of GA in G2 sample significantly increased moisture, protein, and fiber more than 1.7-, 1.1-, and fourfold increments compared to the control biscuits, respectively. However, pectin and CMC increased slightly the moisture content and crude fiber values but did not appear any effect on the others.
Moreover, the addition of GA, CMC, or pectin did not influence crude fat and ash content. However, these values were used in the calculation of total carbohydrate and caloric values as depicted in Table S1. The carbohydrate content varies from a high value of 74.1 ± 0.1 g/100 g in a control biscuit to a low value of 68.3 ± 0.2 g/100 g in the G2 sample. Furthermore, the use of GA in ammonia biscuits reduced significantly (p < .05) the caloric value down to 412.3 ± 0.2 kcal/100 g compared with control biscuits.
Therefore, the formulation of ammonia biscuits with GA reduced the energy by 4.5% compared to the control biscuits. The above results proved that the use of GA-treated biscuits improved also the nutritional profile of control biscuits.

| Consumer acceptability
The results of consumers' acceptability are shown in Table 2. It was found that the different studied ammonia biscuits presented the acceptable scores for all sensory attributes appreciably higher than the minimum acceptability score, that is, 5.0. The overall acceptance among the most studied samples was not significant (p > .05), while the control biscuit sample appeared the highest scores in most of sensory attributes and the overall acceptance. Therefore, the overall sensory attributes of treated ammonia biscuits with hydrocolloids were somewhat worse than the control biscuits; however, the results did not indicate any lack of acceptance. The hydrocolloids, particularly GA, had also retained less carcinogenic compounds 4(5)-MI, AA, and 5-HMF inside the baked ammonia biscuits.

| CON CLUS ION
The incorporation of food hydrocolloids in low amounts, particularly GA, into ammonia biscuit formulations where they are processed at 180 °C for 10 min reduced significantly the formation of thermal process toxicants 4(5)-MI, AA, and 5-HMF. The percent inhibition of 4(5)-MI was ranged from 50.5% to 89.9% after baking at 180 °C for 10 min by increasing the GA amount from 0.01 g to 0.05 g added to the biscuit dough. Furthermore, the use of 0.05 g GA decreased AA content with 58.6% inhibition compared to the control biscuits. Moreover, the highest inhibition of 5-HMF was achieved by 0.05 g GA with 74% depression. The reasons could be due to the formation of tight GA layer as a compact barrier to moisture evaporation during baking process. This was confirmed by SEM and WL analysis. As a result, this would inhibit the molecular interactions of Maillard reaction and caramelization inside ammonia biscuits. However, while the modified GA-ammonia biscuits were evaluated in sensor attributes lower than the control biscuits, they were still overall acceptable.

ACK N OWLED G M ENTS
Author would like to thank Dr. Waleed Elsaid, Assiut University, for his technical assistance with chemical instruments.

CO N FLI C T O F I NTE R E S T
The author declares that I do not have any conflict of interest.

E TH I C A L S TATEM ENT
This study does not involve any human or animal testing.