Bioutilization of Chicken Feather Wastes by Newly Isolated Keratinolytic Bacteria Into Protein Hydrolysates With Improved Functionalities

13 In this study, a novel feather-degrading bacteria B. amyloliquefaciens KB1 was isolated from 14 chicken farm bed (CFB), identified by morphological, physico-biochemical tests followed by 15 16s rDNA analysis. Among observed isolates, bacterial isolate (KB1) showed the highest 16 degree of feather degradation (74.78 ± 2.94 %) and total soluble protein (205 ± 0.03 mg/ g). 17 Using the same species of bacteria, the optimum fermentation condition was found at 40 o C, 18 pH 9, and 1 % (w/v) feather concentration that produced 260 mg/ g of soluble protein and 19 86.16 % feather degradation using response surface methodology in a Box-Behnken design 20 space. The obtained hydrolysates exhibited bioactive properties. The amino acid profile 21 showed the increase in concentration of essential amino acid compared with feather meal 22 broth. The selection of safe screening source of this new bacteria in CFB produced 23 hydrolysates with enhanced bioactivity applicable for food, feed, and cosmetic applications 24 along with environmental remediation. 25


27
Intense growth and development of food processing industries have led to a huge amount of 28 waste as a by-product that is mostly discharged into the environment. Chicken feather 29 remains one of the significant by-products from the poultry industry, mainly due to keratin 30 protein that is hard to degrade [1]. In general, each bird has up to 125 g of feathers and taking 31 into account the daily processing of chicken at 400 million/ day worldwide; this waste 32 reaches five million tons of dry feathers per day [2]. However, chicken feathers are excellent 33 reservoirs of biomolecules with more than 82% crude protein, out of which 91% is keratin, 34 predominantly β-keratin [3]. The higher amount of protein in keratinous waste presents great 35 potential as a source of protein and amino acids for feed, food, and cosmetic applications. 36 Keratin is generally characterized by its ability to resist common proteolytic enzymes and 37 mechanical stability to chemical, hydrothermal and thermo-chemical treatments under high 38 steam. Currently, the industrial process for feather meal involves high temperature and thus 39 the process is costly and energy-intensive. It also results in denaturation and significant loss 40 of essential amino acids producing low-quality protein products [4]. Alkali pretreatments 41 using KOH, NaOH, Ca (OH)2 increase the extraction and yield but possess threats in dealing 42 with toxic effluents [5]. Land dumping and incineration are other methods that are likely to 43 result in environmental vandalism. The generation of toxic air emissions from burning 44 feathers is higher than that generated from coal combustion plants [2]. 45 Biotechnological methods have been employed recently to biologically degrade feather 46

Identification and molecular phylogenetic studies 116
The identification of the Genomic DNA of feather degrading bacteria was based on 5' 16S 117 rDNA gene sequence comparison. This DNA was amplified with universal 16S rDNA 118 primers under following PCR (T100 TM Thermal Cycler, Bio-Rad Laboratories, Inc., 119 Thailand) conditions: 25 cycles of denaturation at 94 °C for 1 min, annealing at 50 °C for 1 120 min and elongation at 72 °C for 3 min. PCR product was amplified using forward primer; 121 20F (5'-GAG TTT GAT CCT GGC TCA G-3') and reverse primer; 1500R (5'-GTT ACC 122 TTG TTA CGA CTT-3'). The nucleotide sequences obtained from all primers were 123 assembled using the BioEdit program (http://www.mbio.ncsu.edu/BioEdit/bioedit.html), 124 followed by deposition of this sequence into the NCBI GenBank 125 (https://www.ncbi.nlm.nih.gov/). The identification of closest phylogenetic neighbors was 126 performed using the BLASTN program against the 16S rDNA sequence from previous 127 prokaryotes' database collection. The pairwise sequence similarity with the highest value was 128 calculated using the Global alignment algorithm. 129

Production of feather protein hydrolysate 130
Keratinolytic bacteria with the highest feather degradation was used as an inoculum for 131 fermenting the raw chicken feather. The bacteria was cultured in nutrient broth for 24 h at 132 37 o C. Then, an inoculum 1% (v/v) containing 10 7 CFU/ml was added to a 250 mL flask 133 containing whole feather (1 g) and MGM (100 mL) as a basal medium for 7 days with 134 shaking incubator at 150 rpm (M2019, Velp Scientifica, Europe). After, every 24 h, sample 135 (5 mL) was harvested, filtered (Whatman filter paper No. 1, GE Healthcare UK), and 136 centrifuged (Centrikon T-324, Germany) (6000 × for 15 min). The supernatant was used 137 to detect the total soluble protein (TSP) and pH. The degree of feather degradation (DFD) 138 was determined from the residual feather on the seventh day. The broth was passed through filter paper (Whatman filter paper No. 1, GE Healthcare UK), followed by washing to 140 remove the cell debris and finally dried in a hot-air oven (SLW115TOP,Gibthai,Thailand)  Aldrich, 3000 IU/g), and Pancreatin (EC 232-468-9, from porcine pancreas, Sigma Aldrich, 183 1400 IU/g) were used for the digestibility studies. Freeze-dried protein hydrolysate (1 g) 184 resuspended in Milli-Q water (1:1) (1 mL) and feather (1 g) were taken in a glass beaker, and it was dissolved with 2 mg/ mL of pepsin prepared with 2 M HCl and incubated for 2 h at 186 37 o C. By the end of the incubation period, the pH was changed to 8 with 2 M NaHCO3. 187 Then, pancreatin (2 mg/ mL) prepared with 2 M HCl was added, and incubation was carried 188 for further 16 h. After completion of digestion, the mixtures were centrifuged (Centrikon T-189 324, Germany). The solubilized protein content in the supernatant was determined by the 190 Kjeldahl method, and % protein digestion was calculated. 191

Oil holding capacity (OHC) 200
The oil holding capacity of the protein hydrolysates was determined according to Jain and 201 Anal (22). Keratin hydrolysate (100 mg) was dissolved in soybean oil (10 mL) and vortexed 202 for 1 min. They were then centrifuged (Centrikon T-324, Germany) at 2500 × for 30 min. 203 Free oil was removed, and the adsorbed oil weighed. OHC was calculated as the weight of oil 204 adsorbed per gram of sample. 205

Water holding capacity (WHC) 206
The water holding capacity of the protein hydrolysates was determined according to 207 hydrolysate (400 mg) was loaded in the centrifugal tube in which distilled water (10 mL) was 209 added, stirred (5 min), and then centrifuged (Centrikon T-324, Germany) at 5000 × for 30 210 min. The unabsorbed water was removed by decantation after centrifugation, and the final 211 weight of the tube was recorded. Finally, the amount of water absorbed (g) was calculated per 212 gram of protein hydrolysates. 213

Chemical fingerprinting by FTIR spectra 214
Using FTIR spectrophotometer (Bruker Vertex 70, Billerica, MA, USA), the structural and 215 functional groups present on the keratin hydrolysate were evaluated, and all spectra were 216 collectively attenuated in the frequency range of 4000-400 cm -1 using 16 scans and 2 cm -1 217 resolution [26]. The lyophilized sample (2 mg) was pressed into the carver hydraulic press 218 after mixing with KBr (100 mg). The spectra were analyzed for the structural characteristics 219 of the protein hydrolysates. 220

Amino acid profile 221
The amino acid composition of the feather protein hydrolysates was analyzed according to 222 Cambridge, UK) using ninhydrin as a color reactant and on a single ion-exchange resin 228 column. The amino acid composition was converted into mg amino acid per 100 g of protein 229 in feather protein hydrolysates and compared with the raw chicken feather meal based on 230 previous studies. 231

Antioxidant activity (DPPH assay) 234
Feather protein hydrolysates were assessed to analyze its ability to reduce the DPPH radical 235 (2,2-diphenyl-1-picrylhydrazyl) (D9132, Sigma-Aldrich, USA) by measuring its absorbance 236 decrease at 517 nm. DPPH solution was made by using DPPH powder (0.004 g in 100 mL 237 95% ethanol), according to Garrido et al. (28). Then stock sample solution of 3 mg/ mL was 238 prepared and diluted to different concentrations (0.125, 0.250, 0.50, 1 and 2 mg/ mL) with 239 distilled water. DPPH solution was then mixed with a sample solution (1:1) in an opaque 240 glass test tube. A blank solution was prepared with DPPH solution (1 mL) and distilled water 241 (1 mL). The samples were incubated (30 min) in the dark, and the absorbance was read at 517 242 nm. The DPPH inhibition activity was determined by using equation (4). 243 Where, and are the absorbance of blank and keratin, respectively. The IC50 value, 245 which is the half-maximal concentration of feather keratin hydrolysate to inhibit a substance, 246 was determined using the Graph Pad Prism 7. 247

Statistical analysis 248
All the experimental tests were carried out in triplicates. The results were expressed as the 249 mean of the replicas with the standard deviation. Similarly, IBM SPSS statistics 21 was used 250 to analyze the Analysis of Variance (ANOVA). Tukey's method was used as a post-hoc to 251 analyze the significant difference among the samples at 95% confidence level.

Isolation of keratinolytic bacteria from chicken farm bed (CFB) 254
CFB was selected as the source for the isolation of keratinolytic bacteria. The bed soil sample 255 (1 g) was serially diluted to 10 -8 . The highest dilution showed 3×10 10 CFU/ mL bacterial 256 population after spreading (100 µL) sample on the skim milk agar (SMA) (incubated at 37 o C 257 for 24 h). Each distinct colony was streaked on the SMA plates to get pure culture. A single 258 colony from each of the thirty plates were then tested to observe their ability to degrade the 259 feather in a test tube containing 10 mL of MGM broth and a single feather piece as a sole 260 source of carbon and nitrogen. The initial pH was maintained 7.5 and incubated for 7 days at 261 37 o C. Out of thirty test isolates, only eight were found to show the feather degradability after 262 7 days of hydrolysis, and thus the isolate was named KB1, KB2, KB3, KB4, KB5, KB6, 263 KB7, and KB8. 264

Measurement of the degree of feather degradation and total soluble proteins of isolates 265
The isolates were grown in an Erlenmeyer flask (250 mL) with MGM broth (100 mL) and 266 chicken feathers (1 g). Table 1 illustrates the degree of feather degradation and total soluble 267 proteins released by these bacterial isolates during feather degradation. The maximum (74.78 268 ± 2.94 %) and minimum (11.1 ± 1.23 %) degradation of the whole feather in the broth were 269 shown by the isolates KB1 and KB2, respectively. The same isolates produced maximum 270 soluble protein KB1 (205 ± 0.03 mg/ g of the dry feather) and KB2 (39 ± 0.06 mg/ g dry 271 feather). Based on these findings, keratinolytic bacterial isolate (KB1) was chosen as 272 appropriate. Fig. 1 illustrates the degradation of a feather (a) control (in a feather meal) (b) 273 solubilized due to isolated bacteria KB1. 274

Optimal growth conditions 284
The growth of the bacterial isolate KB1 was studied in nutrient broth for 72 h, which showed 285 the initial lag phase of 2 h and onset of log-phase till 42 h of growth (Supplementary file 286 figure S1). Then, the optical density started to fall, exhibiting the decline phase. The optical 287 density (OD) measured as absorbance at 600 nm drops gradually with no visible stationary 288 phase. This is because the identified isolate was endospore former, which means due to the 289 depletion of nutrient and accumulation of toxic substances, vegetative cells of endospore 290 former start to undergo spore formations which are smaller in size than vegetative cells [29]. 291

Identification of the bacteria by 16s rDNA 292
The bacterial isolate KB1 was identified using single-strand 16s rDNA sequencing for 293 species characterization phylogenetically. BLAST search engine showed that the species had 294 the highest similarity with the Bacillus siamensis and Bacillus velezensis 99.78%. The 295 phylogenetic analysis of the bacteria that was observed to be located in the same cluster with 296 Bacillus siamensis (Fig. 2)

Optimization of fermentation condition using response surface methodology 303
The identified bacterial isolate KB1 was used for the fermentation of raw chicken feathers in 304 MGM broth. The fermentation process was optimized by using RSM with independent 305 factors; initial feather concentration (1%, 3%, and 5 % w/v): Initial pH (6, 7.5, and 9): 306 fermentation temperature (30, 40 and 50 o C) against the response variables; total soluble 307 protein (mg/ g) and the degree of feather degradation (%) using Box-Behnken design. 308 Where Y1 and Y2 are total soluble protein and degree of feather degradation, respectively. 318 Similarly, X1, X2 and X3 are initial feather concentration % (w/v), pH and fermentation 319 temperature ( o C) respectively. 320 The effects of each independent variable on the response were determined with the help of F-321 test (ANOVA), where initial feather concentration (% w/v) and pH had a significant effect on the production of total soluble protein (mg/ g) and degree of feather degradation (%) (p < 323 0.001). Lack of fit test helps to measure the model's failure to represent predicted and 324 observed data in the experimental design space. The model had a non-significant lack of fit 325 value (p values) of 0.0789 and 0.4136, respectively for total soluble protein and degree of 326 feather degradation, meaning the variation of data fits the actual response variable with the 327 model able to predict values of total soluble protein (mg/ g) and degree of feather degradation 328 (%). The R 2 value for soluble protein and the degree of feather degradation was 0.9866 and 329 0.9781, respectively, which showed a good fit of the empirical model with the experimental 330 data. 331 Table 2 here 332 Fig. 3 (a) and 3 (b) illustrate the response surface (3-D) plots with interactive effects of two 333 independent variables on a single response variable. The soluble proteins were found to be 334 increasing at increasing alkaline condition and decreasing feather condition. Similar findings 335 were observed in the degree of feather degradation. The maximum production of soluble 336 protein (260 mg/ g) was observed at initial pH 9 and 1% (w/v) feather concentration, with 337 86.16 % feather degradation. At higher initial feather concentration, it was observed that 338 soluble protein release and feather degradation was minimum representing 56.20 mg/ g and 339 34.19 %, respectively. At higher substrate concentration, the enzyme excretion is lower and 340 hence is the lower degradation of feather and soluble protein production [31]. However, the 341 influence of temperature was not significant for both the production of soluble protein and 342 feather degradation (p > 0.05). The Design Expert Software determined the optimum 343 fermentation conditions based on the desirability function (Design Expert). The optimum 344 condition (desirability = 0.976) includes initial feather concentration of 1% (w/v), pH of 9, 345 and fermentation temperature of 40 o C at which the maximum protein concentration and Fig. 3 (a) and 3 (b) here 348 of protein (78.45 ± 0.38 %). The reduction in the protein content is due to utilization by 357 bacterial culture to increase biomass; hence the output protein is slightly less than that of a 358 raw feather. The low moisture content (3.54 ± 0.04 %) of the hydrolysate is due to the freeze-359 drying, which helps extend the product's shelf-life. 360 Table 3 here 361

Physico-chemical composition of raw feathers and Feather protein hydrolysates (FPH) 350
In vitro protein digestibility plays a significant role in the formulation of food and feed 362 products. The in vitro protein digestibility of raw feather and feather protein hydrolysates 363 were observed in vitro using pepsin and pancreatin and it was observed as 1.75 ± 0.5 % and

Color of protein hydrolysates 371
The color parameters of sample hydrolysates were measured with Hunter Colorimeter. The 372 parameters were expressed as L* for darkness to lightness, a* for greenness to redness, and 373 b* for blueness to yellowness. L* a*and b* values were found to be 76.49 ± 0.08, 3.19 ± 374 0.22, and 23.27 ± 1.63, respectively. The whiteness index of produced protein hydrolysates 375 was found to be 66.77 ± 1.12. 376

Oil holding capacity (OHC) and water holding capacity (WHC) feather protein 377 hydrolysates 378
Feather protein hydrolysate exhibited excellent OHC (5.46 g/ g) and WHC (3.35 g/ g) of 379 protein hydrolysate, respectively. The increased concentration of polar groups such as COOH 380 and NH2 that is caused by enzymatic hydrolysis has a substantial effect on the amount of 381 adsorbed oil and water [22]. Oil and water-holding properties of protein hydrolysates are 382 crucial in food and feed formulation. These properties directly affect the texture, color, 383 appearance, and the shelf-life of the final product. The higher the water holding capacity, the 384 more the energy to reduce the moisture content and even reduce shelf life. However, it can 385 help to solubilize the water-soluble component in the food matrix (Jain & Anal, 2017). 386 Similarly, higher the oil holding capacity, the product's palatability will be increased with a 387 soft texture and higher fat-soluble nutrients. However, the products will have a lesser shelf-388 life as the product may face rancidity. The water-holding capacity is due to the hydrophilic 389 moiety of proteins. The oil holding capacity results from a lipophilic and non-polar moiety of 390 protein [32]. 391

Fourier-transform infrared spectroscopy (FTIR) of feather protein hydrolysates 392
The FTIR spectra of feather protein hydrolysates exhibiting different peaks of wave numbers 393 representing the presence of Amide A, I, II, and III bands is shown in Fig. 4. The hydrolysate 394 exhibited a peak at 3404.39 cm -1 , which showed the presence of amide A with a wave 395 number close to 3500-3200 cm -1 . N-H stretching vibration is associated with the absorption 396 characteristic of amide A [33]. Amide I exhibit the wavenumber of 1700-1600 cm -1 , which is 397 due to the stretching vibration of C=O bonds, whereas Amide II exhibits the wave number

Amino acid composition 413
The study of amino acid composition is significant as the biological and functional activities 414 of protein hydrolysates depend on the type and composition of amino acids within the protein 415 sequence. The amino acid composition of feather protein hydrolysates was determined, as 416 shown in table 4. The amino acid -glutamic acid, leucine, proline, valine, and aspartic acid 417 were found to be present in the highest amount. Similarly, the protein hydrolysates showed 418 high levels of hydrophobic amino acids (alanine, cystine, isoleucine, leucine, methionine, 419 phenylalanine, proline, tryptophan, tyrosine, and valine) contributing to 55.01% and good 420 quantities of aromatic amino acids (tryptophan, phenylalanine, and tyrosine) contributing to 421 14.29 % of the total amino acids respectively. These amino acids are known to possess 422 antioxidant activities, which help to justify the high free radical scavenging abilities obtained 423 from the fermented protein hydrolysates. The protein hydrolysates also demonstrated good 424 amounts of essential amino acids (histidine, isoleucine, leucine, lysine, methionine, 425 phenylalanine, threonine, tryptophan, and valine) in various food and cosmetic applications. 426 Zhao et al. (36) evaluated the composition of essential amino acids in chicken feather as 427 histidine (0.5 mg/ 100 g protein), isoleucine (3.51 mg/ 100 g protein), leucine (6.16 mg/ 100 428 g protein), lysine (1.12 mg/ 100 g protein), methionine (0.41 mg/ 100 g protein), 429 phenylalanine (3.18 mg/ 100 g protein), Threonine (4.04 mg/ 100 g protein) and cysteine 430 (5.07 mg/ 100 g protein). In our study, as illustrated in table 4, all essential amino acids 431 increased with the most significant increment seen in lysine (3.78 mg/ 100 g protein) and 432 Methionine (0.99 (mg/ 100 g protein). Improved amino acid profile can potentially enhance 433 the growth and meat weight of the chicken applicable to chicken feed industry. 434 Table 4 here 435

In vitro antioxidant properties of feather protein hydrolysates 436
As illustrated in Fig. 5, the antioxidant abilities of feather protein hydrolysates were studied 437 with DPPH radical scavenging activity, which showed that the inhibition activity increases 438 with the hydrolysate concentration. DPPH being a free radical, when protonated, is 439 scavenged, which reduces the absorbance at 517 nm, which is the measure of radical 440 scavenging activity. The IC50 value of the feather keratin hydrolysate was found to be 0.7 441 mg/ mL. The IC50 DPPH radical scavenging activity of this hydrolysate was found to be lower (meaning higher antioxidant abilities) than the two chemically extracted keratin 443 hydrolysate A and hydrolysate C by Alahyaribeik and Ullah (37)  Chryseobacterium sediminis RCM-SSR-7 isolated and identified from feather dumping sites 447 in India exhibited 0.102 mg/ mL radical scavenging activity in its hydrolysates [38]. 448 The free radical scavenging abilities of the feather keratin hydrolysate can be positively 449 correlated with its amino acid composition. As can be observed from the amino acid analysis, 450 the protein hydrolysates showed high levels of hydrophobic amino acids (alanine, cystine, 451 isoleucine, leucine, methionine, phenylalanine, proline, tryptophan, tyrosine, and valine). 452 Cysteine is produced as a product during the breakdown of the disulfide bond present in the 453 feather by microbial keratinase, which acts as a potent antioxidant. Also, Cysteine-SH present 454 in feather peptide is a strong hydrogen donor to free radicals. Furthermore, sulfenic acid (-455 SOH) is produced when the chicken feather is reduced under alkaline conditions. This acid is 456 yet another prime antioxidant in keratin hydrolysate [12]. This confirms the presence of 457 electron-donating hydrolysates and peptides in feather keratin hydrolysate, which could be 458 used as primary antioxidants that are applicable to many food, pharmaceutical, and cosmetic 459 industrial products. 460

Conclusion 462
Chicken feather degrading bacteria Bacillus amyloliquefaciens KB1 was isolated and 463 identified from chicken farm bed and the same isolated bacteria with maximum total soluble 464 protein (250.33 mg/ mL) and the highest feather degradation (86.17 %) obtained after 465 fermentation was utilized successfully to degrade chicken feather. The fermentation process 466 of a chicken feather by isolated bacteria KB1 was optimized using feather concentration, 467 initial pH, and incubation temperature. The feather protein hydrolysates were characterized 468 using FTIR spectroscopy and amino acid analysis. Similarly, these hydrolysates enhanced 469 functional properties like antioxidant abilities and in-vitro digestibility, which can be 470 associated with the breakdown of protein (keratin) during the fermentation process. Thus, the 471 application of green technology-based fermentation by newly isolated and identified bacteria 472 was highly effective in valorizing feather waste in producing feather keratin hydrolysate. 473 Such hydrolysates of chicken feathers from chicken feather waste hold tremendous potential 474 for various feed, pharmaceutical, and cosmetic industries. 475

Funding 476
The authors did not receive any funding for this research. 477

Conflict of interest 478
There are no conflict of interest among the author (s).    (a-f) used in the same column represent significant difference (p < 0.05) 708 Table 2. Box-Behnken experimental design with experimental and predicted values for 709     In vitro % inhibition of DPPH antioxidant activity of feather protein hydrolysates

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.