Valorization of the By-products of Poultry Industry (Bones) by Enzymatic Hydrolysis and Glycation to Obtain Antioxidants Compounds

Currently, one of the fastest growing industries in the world is the poultry industry; however, the increase in demand has generated the production of various byproducts, such as bones, and these byproducts have a negative impact on the environment. The aim of the present work was to evaluate the effect of glycation on the increase in antioxidant compounds and the formation of indicators of advanced glycation end products (AGE) in chicken bone hydrolysates; it also aimed to maximize the protein content, degree of hydrolysis and antioxidant content. Through analysis of variance, the content of AGE products (HMF and furfural) formed in the glycation process was analyzed. The chicken bone hydrolysate had a protein content of 1.42 g/l, a degree of hydrolysis of 17.2% and an antioxidant capacity of 8334 and 10,343 μmol ETrolox/l according to ABTS and ORAC evaluations, respectively. The glycation process increased the ORAC by 6.57%. The presence of hydroxymethylfurfural and furfural was determined in the glycated samples and detected at values between 0.05 and 0.22 and 0 and 0.26 ppm, respectively. In conclusion, hydrolysis and glycation are suitable alternatives that enable the use of chicken bones in producing food ingredients with higher added value.


Introduction
Currently, the world consumption of chicken meat grows on average by 2.6% per year [1]. By 2030, the consumption per capita in the world will be 17.2 kg [2]. However, the increase in the consumption of chicken in its different forms and products generates a variety of byproducts, including viscera, legs, heads, bones, feathers and blood, which are approximately 37% of the live weight of the animal [3]. These byproducts have various impacts on the environment when the industry does not dispose of them properly [4], which leads to biological and biosafety problems [5]. The main poultry meat-producing countries in the world are the United States, Brazil, the European Union and China, which generate 20.3, 14.0, 12.8 and 12.3% world production, respectively; the remaining 40.6% of poultry products are produced in other countries [6]. On the American continent, Colombia ranks fifth with a production of 1.5 million tons of chicken meat per year, below the United States, Brazil, Mexico and Argentina [7].
The increase in the consumption of chicken meat worldwide is due to factors such as low cost, a lack of religious limitations, ease of accessibility, ease of transformation into processed foods [8], and its sensory and nutritional properties [9]. Chicken meat is recognized for its various health benefits, such as its low contents of cholesterol, calories and fat [10]; chicken meat also has a high content of proteins with high biological value [11], as well as essential amino acids and unsaturated fatty acids necessary in the human diet [12].
Given the quantity of production described above, several alternatives have been sought for the use and valorization of the byproducts of the poultry industry to mitigate its environmental impact. For example, the use of chicken feathers [13] and chicken hemoglobin powder [14] in animal food preparation has been reported. In addition, it has been determined that chicken feathers are an important source of amino acids for pigs [3]. Other studies have investigated the production of methane [15], biodiesel [16], flavorings and flavor enhancers from chicken blood, meat and bones [17].
However, one of the most interesting applications of the byproducts of the poultry industry is their use in the extraction of biomolecules, such as polyunsaturated fatty acids and proteins, by enzymatic hydrolysis [18]; chicken bones have specifically been used in this application due to their high protein content of approximately 19% [19]. These components have applications in the pharmaceutical, cosmetic, nutritional, food and microbiological industries [18] and have potential for application as functional ingredients [13]. In general, enzymatic hydrolysis has been used to obtain antioxidants from proteins of various sources, such as wheat germ [20], black beans [21], turkey byproducts [22], salmon gelatin [23] and chicken blood [24]. The extraction of antioxidant peptides derived from food proteins has provided natural ingredients for the formulation of functional foods and has improved their quality [25].
In addition, a prior study [19] determined that once antioxidants are obtained from proteins, they can be increased through a process known as glycation, the reaction between a reducing sugar and the amino acids that come from the peptides released during hydrolysis of the sample, from which the compounds derived from the last stages of the Maillard reaction are formed. [26]. Various authors have reported an increase in antioxidant capacity by the glycation of different products, such as fish [27] and whey [28].
In addition, it has been reported that the antioxidant activity of compounds derived from glycation is due to different mechanisms, such as the chelation of metal ions, breaking of radical chains, decomposition of hydrogen peroxide and elimination of reactive oxygen species [29].
Recent studies report enzymatic hydrolysis as an alternative for the valorization of byproducts of the food industry [25,30]; however, few studies have reported on the effect of glycation in these hydrolysates. This work aimed to evaluate the effect of glycation on the antioxidant capacity and the formation of indicators of thermal damage in CBH to obtain a product with potential applications in the food industry and generate an alternative to mitigate the environmental impact of this waste.

Materials
Chicken bones (Carcasa) were purchased from a local market (Supermarkets EURO). The enzyme used was ALCALASE 2.4L® (Novozymes, Denmark), which contains subtilisin from Bacillus licheniformis. Its optimal temperature is between 55 °C and 70 °C, depending on the substrate, and the optimal pH is between 6.5 and 8.5 [31].

Sample treatment
For the treatment of chicken bones, the methodology described in a pervious study [32] was followed, with some modifications. The chicken carcass was conditioned by removing the adhered meat. The bones were dried in a convective dehydrator (Estructuras y Montajes SAS, Colombia) at 60 °C for 24 h. Once the bones were dried, they were crushed in a Nutribullet food processor at maximum power for approximately 2 min (Nutribullet 600 W, USA).

Degreasing and drying of flour
Once the bone meal was obtained, it was mixed with distilled water at a ratio of 1:3; the solution was heated to 69 °C for 30 min with constant stirring; then, it was cooled to a temperature of 60 °C and refrigerated (Lassele, Mixed Refrigerator LRF-1382PC, Korea) until a temperature of 40 °C was achieved. Once this temperature was reached, the mixture was distributed in 250 mL polypropylene beakers, frozen for 24 h, and stored in a freezer at a temperature of -18 °C (Lassele, LRF-1382PC Mixed Refrigerator, Korea). Subsequently, the sample was removed from the beakers, and the layer of suspended fat on top of the sample was cut. Once the sample had been degreased, it was cut by a chopper into blocks with an approximate thickness of 1.5 cm (Braher, Spain, Tajadora USA-250 Mono), and these were placed in a convective dehydrator (Estructuras y Montajes SAS, Colombia) at 60 °C for 24 h.

Chemical characterization of flour
The proximal analysis of chicken bone meal was performed according to AOAC methods [33], and moisture, protein, fat, ash, and carbohydrates were determined by difference. All tests were done in triplicate. See Table 1.

Enzymatic hydrolysis
For the enzymatic hydrolysis process, the methodology provided in a published study [34] was followed with some modifications. Initially, chicken bone meal was diluted in distilled water in a 500 mL glass reactor at a protein concentration of 20 g/L. Hydrolysis was evaluated at 40, 50 and 60 °C and at pH 7.8 and 9. The apparatus was controlled with sensors connected to a Titrando 842 autotitrator (Metrohm, Switzerland) operated by a computer (Tiamo 1.2.1 software) that maintained the constant stirring of the reaction.
Once the pH and temperature conditions of the experimental design were reached (Table 2), Alcalase 2.4 L food grade was added to start the reaction. The hydrolysis process was carried out over 2 h [35]. The base volume spent was recorded, and the degree of hydrolysis was calculated with the pH stat method, according to a prior study [36]; the released amino -NH groups were calculated according to Eq. 1, and the average degree of dissociation was calculated according to Eqs. 2 and 3: where V B is the volume of base consumed in L, N B is the normality of the base (N), Mp is the mass of the protein in kg, h tot is the total number of peptide bonds in the sample (eq/g protein) (for chicken proteins, its value is 7.6.), and α is the average degree of dissociation of the α-NH 2 groups released in the reaction, which depends on the pK, which in turn is a function of temperature, as indicated in Eqs. (2) and (3) [37]. Finally, the solution was heated at 85 °C for 10 min while stirring to inactivate the enzyme and stop the reaction.

Protein content
The protein content of the hydrolysate was determined by the Bradford method [38,39]. The standard curve was made with bovine serum albumin. The absorbance was measured at 20 °C at a wavelength of 595 nm using a Thermo Scientific ™ Varioskan ™ LUX multimode microplate reader.

Glycation Process
The glycation methodology described in an earlier report [40] was followed; depending on the final protein content of the supernatant obtained from the hydrolysate by the Kjeldahl method, powdered lactose was added in a 1:1 protein mass ratio (Protein:Lactose). This solution was heated in a water bath (Lauda; Aqualine AL 5; Germany) at 60 ± 2 °C for 24 h. Once this process was finished, the product was frozen and stored (Lassele, LRF-1382PC Mixed Refrigerator, Korea) until further analysis.

Antioxidant activity
The antioxidant activity of the hydrolysate supernatant and glycated samples was determined by two methodologies: ORAC and ABTS, which are described below. where AUC is the area under the curve for the sample, AUC • is the area under the curve for the control, AUC Trolox is the area under the curve for Trolox, and f is the dilution factor of the extracts.

ABTS
One hundred microliters of the sample or Trolox standard was taken for the blank, mixed with 1 ml of the ABTS solution, and incubated at 30 °C for 30 min. After that, absorbance measurements were performed at 730 nm. Aqueous solutions of Trolox with concentrations between 0 and 500 M were used for the calibration curve. The results are expressed as micromoles of Trolox equivalents per liter (μMol ETrolox/L) according to a prior study [27] with some modifications.

HMF Y furfural
HMF and furfural compounds were determined as described in a previous study [43], with some modifications. Four milliliters of the homogenized sample was taken and diluted in 2 mL of water, clarified with 250 µL of Carrez I and II, and centrifuged for 10 min at 5000 rpm three times. Subsequently, it was filtered, and a volume of 10 mL was obtained. The supernatant was then passed through a filter with a pore size of 0.2 mm, and 50 µL of the filtered sample was analyzed on a C18 column at a rate of 1 mL/min with a mobile phase of water:acetonitrile (95:5) and a wavelength of 284 nm [44]. The analyses were performed in duplicate.

Experimental design
A factorial design of 3 2 , with 3 center points, was carried out. The hydrolysis temperature and pH were the independent variables, and the protein content, degree of hydrolysis (DH), and antioxidant content by ABTS and ORAC were the dependent variables. This design was performed on hydrolyzed chicken bone meal samples. Each factor was studied at three levels (see Table 2). The design values were evaluated by analysis of variance (ANOVA) to determine the possible significance of the independent variables and their effects on the dependent variables. The experimental trials were randomized to reduce the results of unexpected variability in the observed responses. To find the optimal enzymatic hydrolysis conditions and to analyze how the independent variables affected the overall response, regression analyses and response surface plots were generated. Finally, lack of fit test was performed for the responses variables analyzed. The response variables that were found to be significant in the study were optimized. Several studies use the factorial design 3 2 with central points to optimize processes [45][46][47][48].
The effect of glycation on antioxidant content and the formation of thermal damage indicators was evaluated by analysis of variance (ANOVA) as a function of the glycation process; the antioxidant, HMF and furfural contents were considered the response variables. The samples analyzed by this design were those subjected to hydrolysis and subsequent glycation processes. The data were analyzed with Statgraphics Centurion XVI version 16.1.03 software. Table 1 shows the results obtained from the proximal characterization of chicken bone meal. These values are similar to those reported by the Spanish Foundation for the Development of Animal Nutrition [49][50][51][52], which obtained protein content values of 49.3, 55.45, and 43%, respectively, for chicken bone meal. Several authors have also determined the protein contents in blood [53], offal [54], and chicken feet [55], and the values change depending on the part of the chicken used and can range from 20 to 60%. The contents of other components, such as ash, moisture, fat, and carbohydrates, are reported in Table 1. Table 2 shows the experimental design, as well as the results of the variables analyzed, while Table 3 shows the significance levels for each of the factors and their interactions, as well as the coefficients of determination ( R 2 ). Figure 1 shows the response surface graphs for each variable. Table 2 presents the protein content in the CBH; the results varied between 0.84 and 1.65 mg/mL of protein; these results varied because the limited hydrolysis allows for alterations in the molecule size and the structure and strength of intermolecular and intramolecular interactions between proteins; the molecules are separated into peptides and free amino acids, decreasing the protein content and indicating structural changes [56] that confer functional properties, such as antioxidant activity, to the hydrolysates [57].  According to Table 3, temperature and pH had a significant effect (p < 0.05) on the protein content. Temperatures close to 50 °C allowed a higher enzymatic activity for hydrolysis [58,59], which is evidenced by the decrease in protein content with increasing temperature. As shown in Table 3, lack of fit test was obtained the value of p > 0.05, indicate that the values of the response variables fit a second-order model [60].

Protein content
According to the results obtained for the p value with the goodness of fit test, it can be seen that these are greater than 0.05 for the response variables analyzed, which means that they fit a quadratic model. Table 2 shows that the degree of hydrolysis (DH) ranged from 13.59 to 17.98%, and similar values of DH have been reported in a previous study [35] on salmon byproducts. According to Table 3, DH was significantly affected by pH (p < 0.05); a higher DH was obtained with increasing pH because at a pH close to 9, the enzyme has higher performance [61]. Likewise, Table 3 shows that the quadratic effect of temperature on the degree of hydrolysis is also significant (p < 0.05), showing that a point of maximum DH can be obtained for a temperature of 50 °C (see Fig. 1b). These results are in agreement with those described by [62], who, at pH 8.5 and 50 °C, achieved a higher degree of hydrolysis employing Alcalase 2,4L® in Catla visceral waste proteins. Table 2 shows the results obtained for ABTS and ORAC found for chicken bones under the different processing conditions. These values range from 6531.52 to 8703.26 μmol ETrolox/L for the ABTS method and 2779.7 to 11,565.04 μmol ETrolox/L for ORAC. These antioxidant capacity values are higher than those found for mango [63], orange [64], and beer [65].

Antioxidant content: ABTS and ORAC
According to Table 3, none of the factors analyzed significantly affected the ABTS content (p > 0.05). However, the quadratic effect of temperature did significantly affect the ORAC (p < 0.05). It was also observed that at higher temperatures and pH values, the antioxidant content increased according to both the ABTS and ORAC experiments (Fig. 1c, d). This behavior has been reported by other scholars [66,67]. This behavior occurs because through enzymatic hydrolysis, low molecular weight peptides and amino acids are generated; these factors are related to an increase in antioxidant activity, decreasing the chain reactions of free radical formation through the compounds obtained [61][62][63].
Finally, Eqs. 5, 6, 7, and 8 show the models obtained by the experimental design for each of the response variables analyzed. In this sense, the coefficients of determination R2 for protein content and degree of hydrolysis were 87.91 and 83.83%, respectively, while for the antioxidant content by ABTS and ORAC, the R2 values were 52.43 and 60.75, respectively.

Optimization of the enzymatic hydrolysis process.
To optimize the process, a multiple-response optimization methodology was used, maximizing the protein content and degree of hydrolysis; the pH and temperature conditions were 9 and 50 °C, respectively, and these values were similar to those reported by Novozymes [31]. Table 4 shows the results of the experimental validation of the optimum process conditions, which were tested in   Table 5 shows the content of nitrogen solubilized through hydrolysis, which was 62.42% of the initial content in chicken bone meal, for which it has been reported that, through hydrolysis, the protein can be solubilized up to 90% with Alcalase from treated fleshing meat [68]. It has also been reported that, with this enzyme, a greater number of free amino acids and small peptides is produced than those obtained with the enzyme papain [69]. Table 6 shows the results obtained for the antioxidant content by ABTS and ORAC before and after glycation.

Glycation process
In general, and according to Table 6, the antioxidant content determined by ORAC increased by 6.57%, while the antioxidant content determined by ABTS decreased by 3.73%; this behavior can also be seen in Fig. 2a and b. Likewise, in both cases, the glycation process significantly affected the antioxidant content (p < 0.05), as shown in Fig. 2a and b. The content of antioxidants decreased according to the ABTS tests, probably because they are thermosensitive; however, it can be said that after glycation, antioxidants can undergo further hydrogen transfer (HAT), which is the reason that the ORAC is higher [70].
This result is related to the fact that through the glycation of the amino groups of the proteins or peptides present in the matrix and a reducing sugar, in this case, lactose, antioxidant compounds can be generated; these antioxidant compounds have the capacity to eliminate free radicals. Antioxidants can delay oxidative deterioration in different matrices, which is why it has been suggested that compounds obtained through glycation could be used as functional ingredients in the food industry [71].

HMF and Furfural
Through the glycation process, compounds known as melanoidins are formed; these compounds have been reported to reduce peroxyl radicals [72]. In turn, this reaction also detects the formation of HMF and furfural, which are indicators of advanced glycation end products (AGE) of the intermediate stages of the Maillard reaction [45]. Both compounds have been reported to have toxic and carcinogenic effects [73]. The EFSA has already established an ADI (acceptable daily intake) value of 0.5 mg/kg bw/d for furfural [74]. Additionally, the Scientific Panel on Food Additives, Flavorings, Processing Aids and Food Contact Materials estimated a dietary intake of HMF of 1.6 mg/ person per day based on a modified Theoretical Maximum Daily Added Intake (mTAMDI) approach [75]. According to the European Union Regulation of honey, maximum HMF limits have been established, and these limits vary according to the type of honey; in general, a maximum value of 40 mg/kg is accepted, except for honey for industrial use. A maximum of 80 mg/kg is accepted for honey of tropical origins [76]. Table 2 shows the HMF and furfural contents of the glycated samples, which ranged from 0.05 to 0.22 mg HMF/L and 0 to 0.26 mg furfural/L. The HMF values are lower than those reported in previous studies [77] of beer and balsamic vinegar [78]. The measured furfural contents are lower than those reported to be in fruit juices [77] and [79] in Marsala wine. It should be noted that the HMF and furfural contents are low, so glycation of CBH at 60 °C for 24 h does not lead to a significant formation of the analyzed thermal damage indicator compounds because the Maillard reaction is promoted when food systems containing reducing sugars and amino acids are treated at high temperatures (> 120 °C) [80].  The furfural content presented in Table 2 is low because lactose mainly degrades to form HMF [81], because this disaccharide is composed of two hexoses, which generally promotes the formation of HMF; in contrast, furfural is derived from pentoses [82]. Figure 3a shows that the highest HMF content was observed after treatments 1 and 2, which were significantly different from the other treatments (p < 0.05); these results corresponded to temperature conditions of 50 °C and pH 8 and 9, respectively. Significant differences in the furfural content (Fig. 3b) were found mainly for treatment 4, which involved conditions of 50 °C at pH 7.

Conclusion
In general, it can be concluded that chicken bones are a raw material that can be used in enzymatic hydrolysis to obtain bioactive compounds such as antioxidants. In addition, by means of the factorial experimental design, it was possible to establish the best operating conditions of enzymatic hydrolysis to maximize the protein content and the degree of hydrolysis; it was evidenced that glycation has a significant effect on the content of antioxidant compounds and does not lead to the formation of advanced glycation end products (AGE) such as HMF and furfural. Finally, the compounds obtained here could be used in the development of new food matrices, thus serving as an alternative to the disposal of the byproducts generated by the chicken industry.
Acknowledgements The authors would like to thank the BIOALI (Food Biotechnology Research Group), NUTEC (Food and Nutrition Technology Research Group) and GEMCA (Drug, Cosmetic and Food Stability Group), Faculty of Pharmaceutical and Food Sciences, Universidad de Antioquia, for their support in the development of this project.
Author contribution All authors had readen and agree with the published version of the manuscript. Conceptualization, methodology, formal analysis, experimental research, writing: preparation of the original draft, Luisa Londoño, Sara Franco and Sandra Restrepo; accompaniment of the experimental phase and resources, Lina Suárez, Fáver Gómez; writing: proofreading and editing, Óscar Vega, Pedro Valencia; visualization, Helena Núñez, Ricardo Simpson.

Funding
The authors did not receive support from any organization for the submitted work.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Code availability
The data were analyzed with Statgraphics Centurion XVI version 16.1.03 software.