Enzymatic recovery of glycopeptides from different industrial grades edible bird’s nest and its by-products: nutrient, probiotic and antioxidant activities,and physicochemical characteristics

Hiyti Symimi Moh Noor, Rfih Moh Ariff,b, L Sin Chng,c, Xin Yi Chi,Hui Yn Tn, Nur’ Alih Du, Abul Slm Bbji, Sng Jo Lim,,*

a Faculty of Science and Technology, Universiti Kebangsaan Malaysia, Bangi 43600, Malaysia

b International Institute for Halal Research and Training (INHART), International Islamic University Malaysia (IIUM), Jalan Gombak 53100, Malaysia

c Faculty of Applied Sciences, UCSI University Kuala Lumpur, Cheras 56000, Malaysia

d Faculty of Agriculture Science and Technology, Universiti Putra Malaysia Bintulu Campus, Bintulu 97008, Malaysia

e Innovative Centre for Confectionery Technology (MANIS), Faculty of Science and Technology, Universiti Kebangsaan Malaysia, Bangi 43600, Malaysia

ABSTRACT

This study was conducted to recover edible bird’s nest (EBN) hydrolysates from different grades of EBN,including the industrial by-products, using enzymatic treatment. The nutrient, physicochemical properties and antioxidant activities of the recovered hydrolysates at different hydrolysis times were evaluated. Results showed that the recovery yield of enzymatic hydrolysis was above 89% for all grades of EBN and the degree of hydrolysis increased over time. Nitrite content (0.321–0.433 mg/L) was below the permissible tolerance level for all samples. Interestingly, the antioxidant activities (DPPH and ABTS scavenging activities and ferric reducing antioxidant powder (FRAP) activity) were significantly higher (P ≤ 0.05) in hydrolysates recovered from EBN by-products (EBNhC and EBNhD) as compared to the high grade EBN hydrolysates (EBNhA and EBNhB). The in-vitro probiotic activity of EBN and its hydrolysates were examined using the probiotic bacterium Lactobacillus plantarum. Evidently, EBN by-products hydrolysate (EBNhD) recorded the highest number of L. plantarum (1.1 × 1011 CFU/mL), indicating that low grade EBN has the potential as prebiotic material that promotes probiotic activity. This study demonstrated the concept of using EBN by-products hydrolysates for various applications, such as functional ingredients with enhanced bioactivities, to improve its economic value.

Keywords:

Edible bird’s nest by-products

Enzymatic hydrolysis

Glycopeptides

Lactobacillus plantarum

Probiotic activity

1. Introduction

Edible birds’ nest (EBN) is a delicacy made from swiftlets’ saliva,in which the saliva was used to build the swiftlet’s nest during the mating season [1], with glycoprotein (20–140.8 kDa) as the main component [2-4]. EBN is a valuable commodity in Malaysia, priced at USD 1 000–10 000/kg, depending on the EBN quality [5]. The high commercial value of EBN is due to the high market demand,popular for its nutritional and therapeutic properties. EBN was proved to improve the immune system [6], enhance development of infant brain functions [7], induce cell proliferation [8], and exhibit antioxidant [3,9] as well as antihypertension properties [10]. Besides,the risky and extensive procedure of nest harvesting, and the laborious cleaning processes are among the reason of high EBN prices in the market. Thus, researchers aim to enhance the yield of EBN, to increase affordability to the general public, enabling them to enjoy its health benefits.

Nest processing involves immersion of EBN in water to allow swelling, manual impurities removal (egg shells, feathers, leaves, soil,etc.) by forceps, followed by sterilisation, moulding into appropriate shape, dehydration, and finally secondary sterilisation prior to packaging [11]. The grading method is depending on the extent of the cleaning steps, in which the cleaned EBN without impurities is of the highest quality (Grade A), while the semi-cleaned bird’s nest (Grade B) may contain some impurities. However, during the cleaning process, large quantities of EBN by-products are disposed of, consisting mostly of bird feathers with the important glycoprotein attached to them. Our previous study showed that EBN hydrolysed using alcalase produced lower molecular weight glycopeptides with enhanced solubility and bioactivity [12,13]. Hence, this research was carried out to recover glycopeptides from the discarded EBN by-products, whereas the commercially cleaned EBN and the semi-cleaned EBN were used as control. Subsequently, probiotic activity, antioxidant activity and physiochemical properties of recovered glycopeptides were examined. Potential of EBN as prebiotic material was determined through in-vitro digestion system and subsequently inoculation of Lactobacillus plantarum, a probiotic bacterium. This study highlights the potential to recover high-value glycopeptides from the industrial waste that could be used as alternative source of EBN, as well as application in secondary products. This improves the sustainability of the EBN industry as it reduces waste and, in turn, makes EBN highly accessible due to lower costs and improved bioactivities. The recovered EBN glycopeptides are suitable to be applied in various industries as a functional ingredient.

2. Materials and methods

2.1 Materials

Various grades of EBN samples (EBNRA= cleaned bird’s nest,EBNRB= semi-cleaned bird’s nest, EBNRC= by-products from bird’s nest cleaning process, EBNRD= uncleaned bird’s nest residue by-product), as shown in Fig. 1, were supplied by Mobile Harvesters (M) Sdn Bhd. Alcalase enzyme (Bacillus licheniformis) for the enzymatic hydrolysis of EBN was purchased from Novozymes,Denmark. α-Amylase, Type IX-A (1 000–3 000 U/mg pro)from human saliva, pepsin of the porcine gastric mucosa(3 200–4 500 U/mg pro) and pancreatin of the porcine pancreas(8 × USP) used for in-vitro fermentation were bought from Sigma-Aldrich (UK). L. plantarum ATCC 8014 was sourced from Next Gene Scientific Sdn. Bhd. (Malaysia). Other analytical grade chemicals in this study were supplied by Sigma-Aldrich (UK),unless otherwise specified.

Fig. 1 Different grades of EBN samples.

2.2 Analysis of raw EBN

2.2.1 Proximate analysis

Macronutrient content was determined on all raw EBN samples(EBNRA, EBNRB, EBNRC, EBNRD) in accordance with the standard procedures by the Association of Official Analytical Chemists(AOAC) [14]. Proximate composition determined were moisture content (AOAC method 925.10), ash content (AOAC method 942.05),crude fat content (AOAC method 920.39) and crude protein content(AOAC method 984.13). Total carbohydrate was the subtraction of the sums of the weights of moisture, ash, crude fat and crude protein from the total weight of the EBN.

2.2.2 Amino acid profiling

Generally, the amino acid profiles of raw EBNs (EBNRA, EBNRB,EBNRC, EBNRD) were determined by sequences of treatments and analysis, starting from hydrolysis of samples, separation of sediments,derivatisation of the hydrolysates, and amino acids detection using high performance liquid chromatography (HPLC) as defined by Ali et al. [15].Three different hydrolysis methods were used, namely acid hydrolysis,performic acid hydrolysis and alkaline hydrolysis. Acid hydrolysis was performed by adding 15 mL 6 mol/L HCl to the test tube containing 0.03 g of EBN sample. Samples were then placed in an incubator at 110 °C for 24 h and cooled down to room temperature (25 ± 1) °C.Meanwhile, performic acid hydrolysis was performed by adding 2 mL of performic acid (> 95%) to the test tubes containing 0.03 g of EBN. The sample was then incubated at 4 °C for 16 h. After that,0.4 mL of hydrogen bromide (HBr) was added, and the incubation was continued for another 30 min. Finally, the process was completed by adding HCl for acid hydrolysis process, as mentioned above. For both methods, the acid hydrolysates were transferred to a volumetric flask, added with 400 μL of 50 μmol aminobutyric acid solution,AABA (internal standard) and deionized water was used to make up to 100 mL. The solution was filtered via a nylon cellulose membrane(0.45 μm) and 10 μL of the sample was pipetted into a centrifugal tube for derivation using AccQ Flour Reagents. Lastly, the derivatised sample was injected into the HPLC system equipped with a fluorescence detector (λ excitation at 250 nm and λ emission at 395 nm)and separated using the Waters AccQ·Tag Amino Acid column(3.9 mm × 150 mm, 4 μm) (Waters Co., USA). Mobile phases used were eluent A (10% AccQ Tag in Milli-Q water) and eluent B(60% acetonitrile), flow through a gradient system at a flow rate of 1.0 mL/min. The gradient used for the analysis started at 100% eluent A and 0% eluent B, 98% A and 2% B at 0.5 min, 91% A and 9% B at 15 min, 87% A and 13% B at 19 min, 65% A and 35% B at 32 min and held for 2 min. Eluent A was reduced to 0% and eluent B was then increased to 100% at 38 min, 100% A and 0% B at 39 min held until 50 min [16].

On the other hand, alkaline hydrolysis was performed by adding 5 mL of 4.3 mol/L lithium hydroxide monohydrate (LiOH·H2O) to the test tube containing 0.03 g of EBN. The test tube was then placed in an incubator at 120 °C for 16 h. Hydrolysate was then transferred to a beaker and 9 mL 6 mol/L HCl was added. The pH was measured and adjusted to pH 4.5 using 0.1 mol/L HCl. The solution was transferred to a 100 mL volumetric flask and added with deionised water up to a final volume of 100 mL. The solution was then filtered using 0.2 μm cellulose acetate membrane. Finally, the alkaline hydrolysate without derivatisation was injected into HPLCfitted with Nova-Pak C18column (3.9 mm × 150 mm, 4 μm) (Waters, Waters Co., USA),eluted with a mixture of 0.0085 mol/L sodium acetate solution(pH 4.0) and methanol with a ratio of 86.7:13.3. Eluting compounds were recorded through fluorescence detector (285 nm excitation and 345 nm emission of λ).

The protein efficiency ratios (PER) were calculated based on the essential amino acid composition. The PER was determined on the basis of equation (1) [17]:

Where ΣAA is the amount of essential amino acids present in samples: arginine (Arg), histidine (His), isoleucine (Ile), leucine (Leu),lysine (Lys), methionine (Met), phenylalanine (Phe), threonine (Thr),tyrosine (Tyr), and valine (Val).

2.3 Enzymatic hydrolysis of raw EBN

Hydrolysis of raw EBN samples (EBNRA, EBNRB, EBNRC, EBNRD)was performed using the method mentioned by Ali et al. [15].A 4 × 4 factorial design was employed, with EBN sample grades and enzymatic hydrolysis duration as the factors. Brie fly, EBN samples were soaked for 16 h, then double-boiled for 1 h, added with 1% alcalase enzyme at pH 8 and incubated at 60 °C for 1, 2, 3 and 4 h, and subsequently deactivated at 90 °C for 5 min,filtered and freeze-dried for 72 h. The concentration of alcalase added was based on our patented parameters (Patent No. WO2017034390A2), in which the ratio of enzyme:EBN protein fractions was 1:100 (V/m). EBN hydrolysates of EBNRA, EBNRB, EBNRCand EBNRDwere labelled as EBNhA, EBNhB, EBNhCand EBNhD, respectively. The recovery yields were estimated based on equation (2).

where, W1is the dry weight of EBN prior to hydrolysis and W2is the dry weight of EBN hydrolysate.

The degree of hydrolysis (DH) was determined according to equation (3) (pH Stat method):

where, B = volume of 0.1 mol/L NaOH to maintain the mixture at pH 8 (mL); Nb= normality of the titrator; α = pK for amino groups at the specified temperature (1.5 and 1.4 for 25 and 30 °C, respectively);Mp= crude protein mass of the samples (g) and Htot= total number of peptide bonds in protein substrate [18].

2.4 Chemical analyses of EBN hydrolysate

2.4.1 Nitrite and nitrate content

Nitrite content was determined by adding 50 μL EBN hydrolysates samples (1 mg/mL) to 50 μL of 1% sulphanilamide solution.The mixtures were homogenised using vortex and allowed to stand in the dark at room temperature for 10 min. Then, 50 μL of 0.1% N-1-naphthylenediamine dihydrochloride (NED) was applied,homogenised, and allowed to stand in the dark for 10 min at room temperature. The sodium nitrite solution of 1, 2, 4, 6, 8 and 10 mg/L were used for the preparation of the standard curve.

For nitrate content, 3 mL EBN hydrolysate (1 mg/mL), pH 7 (adjusted with acetic acid or sodium hydroxide) was mixed with 3 mL of sulphuric acid (diluted with distilled water 4:1, H2SO4:H2O) and added with 0.25 mL Brucine-sulphanilic acid reagent (1 g brucine sulphate, 0.1 g sulphanilic acid, 3 mL HCl in 100 mL distilled water). The solution was heated in a boiling water bath for 25 min and cooled to room temperature. A standard curve was prepared by a serial dilution of KNO3stock solution (100 mg/L), ranging from 1 mg/L to 10 mg/L. The KNO3stock solution was prepared by dissolving 30 mg KNO3in 300 mL distilled water. Sample absorption was measured using the EpochTMMicroplate Spectrophotometer(BioTek, Vermont, USA) at 540 and 410 nm for nitrite and nitrate contents, respectively [19]. The nitrite and nitrate content were determined according to equation (4):

where, R is the absorbance value obtained of the standard curve.

2.4.2 Total content of soluble protein

The total soluble protein content was calculated based on the Bradford protein assay [20]. EBN hydrolysates samples at 1 mg/mL(5 μL) were mixed with 250 μL Coomassie Brilliant Blue G250 dye reagent and required to react for 5 min at room temperature. Bovine serum albumin (BSA) was used as standard (0.01–0.20 mg/mL).The absorption was assessed at 595 nm with the microplate spectrophotometer [21].

2.4.3 Peptide content

Peptide content was calculated using the O-phthaldialdeyde (OPA)method defined by Church et al. [22]. The OPA solution was prepared by incorporating 12.5 mL of 100 mmol/L sodium tetraborate,1.25 mL of 20% sodium dodecyl sulfate (SDS), 0.5 mL of 4% OPA (dissolved with methanol) and 50 μL β-mercaptoetanol. The distilled water was added to achieve afinal volume of 25 mL. EBN hydrolysates samples (6 μL; 1 mg/mL) and 210 μL OPA reagents were mixed and absorption at 340 nm was determined using the microplate spectrophotometer.

2.4.4 Reducing sugar content

EBN hydrolysate samples at 1 mg/mL (1 mL) were mixed with 4 mL 3,5-dinitrosalicylic (DNS) reagent and placed in a boiling water bath for 5 min. The mixture was cooled to room temperature and the microplate spectrophotometer was used to read the absorbance value at 540 nm. The standard curve was prepared by serial dilution of glucose solution, ranging 0.1%–0.5%, and the percentage of reducing sugar in the sample was recorded as a percentage of glucose equivalent (% GE) [23].

2.5 The antioxidant role of EBN hydrolysate

The 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical scavenging activity was analysed by mixing 500 μL of EBN hydrolysates (1 mg/mL) with 2.5 mL of 100 mmol/L methanolic DPPH reagent. The mixtures were kept in dark for 30 min at room temperature [24]. The 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical scavenging activity was assessed by adding 500 μL of EBN hydrolysates (1 mg/mL) to 1 mL of ABTS solution, prior to incubation in the dark at room temperature for 10 min [25]. Meanwhile, ferric reducing antioxidant powder (FRAP) was determined by adding 500 μL of EBN hydrolysates samples (1 mg/mL)to the FRAP reagent (2.5 mL), and allowed to stand away from light at room temperature for 30 min [26,27]. Ascorbic acid (C6H8O6)solutions were used as standard in FRAP assay (100, 250, 500 and 1 000 μmol/L) and FRAP values were expressed as ascorbic acid equivalent antioxidant capacity (AAEAC). DPPH, ABTS and FRAP were evaluated at 517, 734 and 593 nm, respectively, using the microplate spectrophotometer. The percentage of free radical scavenging activity of DPPH and ABTS was calculated based on equation (5), while the FRAP values were determined using the equation (6):

whereA1is absorbance value of blank andA2is absorbance value of sample.

where,Ris the standard curve reading, and DF is the dilution factor.

2.6 Probiotic activity

2.6.1 In-vitro digestion system

Simulated gastrointestinal digestion was conducted based on the procedures defined by Minekus et al. [28], which consisted of simulated oral, gastric and intestinal phases. Fructose-oligosaccharide(FOS) was used as a positive control, while EBNRA, EBNhA,EBNhB, EBNhC, EBNhDwere digested as samples. FOS was used as a positive control because FOS is a well-known commercial prebiotic material in the form of carbohydrate. It is used to feed probiotic in the intestine as a source of soluble fibre. Simulated salivary fluid (SSF), simulated gastric fluid (SGF) and simulated intestinal fluid (SIF) are composed of the corresponding electrolyte stock solutions, enzymes, CaCl2and water as shown in Table 1.Starting with the oral phase, EBN hydrolysates samples were added with 15 mL of water (1:15,m/V) and mixed with 10.5 mL SSF electrolyte stock solution. Then, 1.5 mL ofα-amylase saline solution (1 500 U/mL) was applied, followed by 0.075 mL of 0.3 mol/L CaCl2and 2.925 mL of water. The mixture was thoroughly mixed and allowed to react and transferred to the gastric phase. The oral phase solution (30 mL) was combined with 22.5 mL of SGF electrolyte stock solution, 4.8 mL of pepsin (25 000 U/mL), 0.015 mL of 0.3 mol/L CaCl2, 0.3 mL of 1 mol/L HCl and 2.385 mL of water and incubated at 37 °C for 2 h. The gastric acidic solution (60 mL) was then combined with 33 mL of SIF electrolyte stock solution, 15 mL of pancreatin stock solution (800 U/mL), 7.5 mL of 160 mmol/L fresh bile, 0.12 mL of 0.3 mol/L CaCl2, 0.2 mL of 1 mol/L NaOH and 4.1 mL of water [28]. All the digested samples were then freeze-dried for 48 h until further use.

Table 1Volume of solution (mL) used in preparation of SSF, SGF and SIF.

2.6.2 In-vitro fermentation

All ingredients (casein peptone (10 g), meat extract (10 g),Tween 80 (5 g), yeast extract (5 g), sodium acetate (5 g), ammonium citrate (2 g), dipotassium hydrogen phosphate (2 g), magnesium sulphate (0.2 g), and manganese sulphate (0.05 g) in 1 000 mL of distilled water) were prepared to make a nutrient broth and the medium was autoclaved at 121 °C for 15 min. Allin-vitrodigested samples (1%, 1.4 g) of the FOS, EBNRA, EBNhA, EBNhB, EBNhCand EBNhDhave been added to 140 mL of nutrient broth. Nutrient broth containing samples (20 mL) was placed in 100 mL Erlenmeyer flasks and 1% (0.2 mL) of activatedL. plantarumhas been added and incubated for 24 and 48 h at 37 °C.

2.6.3 Probiotic activity assessment

GrowthofL. plantarumwas monitored at 24 and 48 h by measuring the cultures’ optical density (OD) at 622 nm using the microplate spectrophotometer. The pH of the culture solutions was also measured using a pH metre. Counting ofL. plantarumcolony forming unit (CFU) on the MRS agar plate was performed, where a serial dilution (up to 10-10) was prepared by mixing 1 mL of culture solution with 9 mL of saline solution and then, 1 mL of saline solution containingL. plantarumwas spread on the MRS plate and incubated at 37 °C overnight.

2.7 Analysis of statistics

All analyses were carried out in triplicates. Factorial experimental design was analysed using two-way ANOVA, whereas the statistical difference between the means was determined using Duncan’s multiple range test atP≤ 0.05, both using SPSS software (IBM SPSS Statistics Version 23, USA).

3. Results and discussion

3.1 Proximate analysis

Proximate compositions of different grades of EBN are shown in Table 2. The significantly highest (P≤ 0.05) moisture(16.62 ± 0.12)% and ash (10.70 ± 0.08)% contents were recorded in EBNRBand EBNRD, respectively. The high ash content in EBNRDmay be due to the sample being uncleaned EBN residue by-products,thus containing feathers, soil and other foreign materials. In addition,the higher ash content in EBNRCand EBNRDcompared to EBNRAand EBNRBwas due to the leaching of water-soluble minerals in EBNRAand EBNRBduring the cleaning process, which involved the immersion of EBN in water.

Table 2Proximate composition (%) of raw EBN of different grades (n = 3).

All EBN samples have protein content ranging from (54.29 ± 0.68)% to (60.59 ± 0.14)%, which was in agreement with Norhayati et al. [29]who recorded EBN protein content of among 56.20%–61.50%.The protein content of EBNRAwas significantly higher (P≤ 0.05)at (60.59 ± 0.14)%. This was expected because the EBNRAhas undergone a thorough cleaning process that removed the impurities and thus increased the overall protein composition. However, the fat contents ranged from 0.43%–1.19%, where all samples showed no significant difference (P> 0.05) among one another. Besides,EBNRChas the significantly highest (P≤ 0.05) carbohydrate content of (26.32 ± 1.37)%. Overall, all EBN samples had very low fat and ash content, as they were mainly made up of glycoproteins; therefore,high protein and carbohydrate contents were recorded [3,13].

3.2 Amino acids profile

The amino acid compositions for different grades of EBN are shown in Table 3 (EBNRAchromatogram is shown in Fig. 2). All EBN samples contain 18 amino acids, with a particularly high amount of valine, leucine and threonine essential amino acids. This is consistent with Ali et al. [15], who also quantified all 18 amino acids in the raw EBN sample. Valine was the highest essential amino acid present at (5.15 ± 0.12), (5.12 ± 0.21), (4.43 ± 0.21) and (5.10 ± 0.11) g/100 g for EBNRA, EBNRB, EBNRCand EBNRD, respectively. Meanwhile, the total amino acid contents of all samples ranged 52.15–61.04 g/100 g and did not show any significant difference (P> 0.05). Among the samples, EBNRChad a significantly lower (P≤ 0.05) amount of Asp, Ser, Gly, Tyr, His, Val, Ile, Leu, and Phe compared to the other samples. This finding was consistent with the proximate composition (section 3.1), where EBNRCshowed the lowest protein content due to the presence of other impurities.

Fig. 2 Typical chromatogram of amino acid profile for EBNRA through (a) hydrochloric acid hydrolysis, where Asp = aspartic acid, Ser = serine, Glu = glutamine acid, Gly = glycine, Ala = alanine, Pro = proline; (b) performic acid hydrolysis, where Cya = cysteine; and (c) alkaline hydrolysis, where Trp = tryptophan.

The PER, which analyses the quality of proteins by considering the requirement of essential amino acids, is presented in Table 3. EBNRAhad the significantly highest (P≤ 0.05) PER value(1.63 ± 0.10), while EBNRCexhibited the significantly lowest(P≤ 0.05) PER value of 1.35 ± 0.11. This indicates the cleaned EBN exhibited better protein efficiency with a higher amount of essential amino acids. Interestingly, EBNRDhave a PER value (P> 0.05) close to the EBNRAand EBNRBwith no significant difference (P> 0.05). This finding indicates the valuable characteristic of EBNRDas it had a high PER value.

Table 3Amino acid profile (g/100 g) of raw EBN of different grades (n = 3).

3.3 Recovery yield of EBN hydrolysates

Fig. 3a displays the recovery yield of different EBN samples with different enzymatic hydrolysis time treatments. No significantly difference (P> 0.05) recovery was obtained throughout 1–4 h of enzymatic hydrolysis in EBNhA, ranging from 96% to 99%, with the highest recovery at 2 h of enzymatic hydrolysis. There was no significant difference (P> 0.05) in recovery yield from 1, 3, 4 h of the hydrolysis period for samples, except for EBNhB, in which sample hydrolysed for 2 h was significantly higher (P≤ 0.05) (95.56 ± 1.41)% than sample hydrolysed for 4 h (88.99 ± 4.86)%. The difference may be due to the purity of the semi-cleaned EBN was lower, that contained lower glycoprotein. The overall lower recovery yields of EBNhB(89%–96%), EBNhC(91%–92%) and EBNhD(92%–94%)throughout the hydrolysis as compared to EBNhAmay be due to higher purity of EBNhAafter removal of impurities in this sample.Overall, this showed that all samples had a high recovery of EBN hydrolysate yield of 89%–99%, which is consistent with the report by Lim [30] and Ibrahim et al. [31] which reported that EBN contained 5%–10% of foreign materials, for instance feathers and dirt. The traditional EBN processing method (soaking and picking) recorded a loss of up to 35%–40% of precious EBN glycoproteins [32],rendering the enzymatic approach an interesting alternative to the EBN processing method, which reduces waste with high recovery rates. On the other hand, the two-way ANOVA results (Table 4)showed a significant effect of sample grades and hydrolysis time(P≤ 0.05), but the inter-factor interactions were not significant (P> 0.05)for the recovery of EBN hydrolysates.

Table 4Result of two-way ANOVA analysis.

Fig. 3b demonstrates the degree of hydrolysis (DH%) for all EBN samples undergoing enzymatic hydrolysis for up to 4 h with 15 min sampling interval. Significant differences (P≤ 0.05) in DH% were recorded among all the EBN samples throughout the hydrolysis period. All EBN samples were found to show an increasing trend of DH% as the enzymatic hydrolysis time increased. A similar trend has been reported by Zulkifli et al. [18], where the DH% of the cleaned EBN sample increased proportionally with hydrolysis time(0.5–3 h), with a maximum DH% of 16.2%. The longer hydrolysis duration allowed alcalase to react efficiently with the protein and to break down the peptide bonds, thus releasing more glycopeptides and subsequently increased the DH% [10,18].

Fig. 3 (a) Recovery yield and (b) degree of hydrolysis of various grades of EBN samples under several hydrolysis time (n = 3).

EBNhChas the highest DH% (13.79%) followed by EBNhB(13.48%), EBNhD(13.33%) and EBNhA(10.83%) after 4 h of incubation. EBNhAhas the significantly lowest DH% (P≤ 0.05)in the entire enzymatic hydrolysis process. This may be due to the presence of more insoluble glycoprotein in the cleaned sample,where most soluble glycoproteins have been removed during the cleaning process. Meanwhile, EBNhChad the highest DH% due to higher soluble glycoproteins. It was therefore necessary to have a longer reaction time between the enzyme and the insoluble protein of EBNAfor the production of EBNhAand vice versa for EBNhC.Enzymatic hydrolysis enables EBN glycoproteins to be hydrolysed into glycopeptides, making them water soluble. Interestingly, for EBNhB, EBNhCand EBNhD, enzymatic hydrolysis helps to clean and improve the purity of the samples by breaking the glycoproteins (water insoluble) into glycopeptides (water soluble) and removing water insoluble impurities and feathers by filtration. This method can be utilised as an alternative cleaning procedure for the EBN industry to replace the laborious and costly manual EBN cleaning process.

3.4 Physicochemical properties

3.4.1 Nitrite and nitrate content

Table 5 demonstrates the nitrite content of EBN samples ranging from (0.321 ± 0.02) mg/L to (0.433 ± 0.03) mg/L, which was relatively low. According to the Malaysian Standard MS 2334:2011 [33], the raw EBN obtained from the purification process cannot contain nitrite content of more than 30 mg/L, in which the EBN hydrolysate samples produced in this study have a nitrite content well below the permitted tolerance level. The nitrite content was not significantly affected (P> 0.05) by the EBN grades, the hydrolysis time and the interaction of these two factors (Table 4). After 4 h of hydrolysis, the nitrite content of the EBNhBsample was significantly lowest (P≤ 0.05).Meanwhile, the other samples showed no significant difference (P> 0.05), indicating the grades of the sample and the hydrolysis time showed no effect on the nitrite content (Table 5).Nitrite content is closely related to soil contamination from the harvesting environment. This indicated that the EBN being studied was considered safe to consume because of the low level of nitrite.

Table 5Physicochemical properties of different grades of EBN hydrolysates under different hydrolysis time (n = 3).

However, the significantly highest nitrate content (P≤ 0.05) was found in EBNhAafter 2 h of hydrolysis (26.91 ± 4.92) mg/L and the significantly lowest nitrate content (P≤ 0.05) was found in EBNhCand EBNhDafter 1 and 3 h of hydrolysis (Table 5). Sample grades,hydrolysis time and inter-factor interactions showed significant effects (P≤ 0.05) on nitrate content (Table 4). The amount of nitrate content in this study was found to be higher than nitrite content,which was consistent with Quek et al. [34] who also found that nitrate content (98.2 μg/g for house nests and 36 999.4 μg/g for cave nests)was significantly higher (P≤ 0.05) than nitrite content (5.7 μg/g for house nests and 843.8 μg/g for cave nest). The higher nitrate content compared to that of nitrite content was due to the fact that nitrate has a more stable structure and can be produced via oxidation of nitrite [35].

3.4.2 Total soluble protein content

Table 5 reports the total amount of soluble protein in EBN hydrolysate samples, where the soluble protein content of EBNhCat 1 h hydrolysis (81.12 ± 1.81)% was the significantly highest (P≤ 0.05), while the significantly lowest (P≤ 0.05) was recorded at 4 h of hydrolysis in EBNhD(65.57 ± 1.64)%. Overall,EBNhDhad a significantly lower (P≤ 0.05) content of soluble protein for all hydrolysis duration analysed. It was evident that the increase in hydrolysis time significantly (P≤ 0.05) decreased the soluble protein content of EBN hydrolysates. The study carried out by G?rgü? et al. [36]using a combination of microwave heating and enzymatic hydrolysis (alcalase) in production of sesame bran hydrolysate exhibited that the protein content increased from 52.7% to 61.3% when the hydrolysis time increase from 10 min to 104 min. Two-way ANOVA showed that all the main factors and interactions between the factors were significant (P≤ 0.05) in terms of soluble protein content (Table 4).Protein solubility depends on the balance of hydrophilicity and hydrophobicity of protein molecules and the composition of amino acids exposed at the protein surface [37]. Proteins are broken down into smaller peptides and released amino acids during enzymatic hydrolysis. The different positions of these amino acids exhibited different hydrophobic and hydrophilic properties that influence protein solubility [37]. In addition, the decrease in protein solubility may be due to exposure to heat treatment during long hydrolysis hours. Moreover, the thermal stability study conducted on single-chain bromelain glycoprotein by measurement of enzymatic activity showed an optimum hydrolysis temperature at 40 °C, and decreasing enzyme activity with increasing temperature [38].

3.4.3 Peptide content

Peptide content ranged from (53.21 ± 1.89)% to (74.99 ± 2.93)% as shown in Table 5. Sample grades, hydrolysis time and inter-factor interactions were found to have significant effects (P≤ 0.05)on peptide content. Overall, the significantly highest (P≤ 0.05)peptide content was observed in EBNhD, indicating that EBNhDhad the smallest glycopeptides compared to other EBN hydrolysates. It is interesting to note that the uncleaned EBN residue by-products showed greater hydrolysis performance with a higher peptide content compared to the cleaned EBN. Meanwhile, significantly higher(P≤ 0.05) peptide content was observed when the hydrolysis time increased from 1 h to 2 h in the EBNhCsample and decreased thereafter. Similarly, Syarmila et al. [39] reported that the soluble protein content of EBN hydrolysate obtained from alcalase treatment increased significantly from the initial value of 25.5-86.7 mg/g after 90 min of hydrolysis, but the value decreased to 70.1 mg/g after the continuous treatment up to 4 h. This is associated with more glycoproteins in EBN being broken down into smaller glycopeptides when the contact time of the enzyme with the samples is increased.However, when the enzyme reached saturated point, the concentration of peptide bonds available for hydrolysis may become the limiting factor, entering into a stationary enzyme activity period [40]. Besides,alcalase was reported to catalyse condensation (joining of peptides)reaction in whey protein hydrolysate over 6 h of incubation time,forming of peptide aggregates, led to a lower peptide content [41].

3.4.4 Reducing sugar content

EBN, being a glycoprotein, has its glycans, which could be measured by reducing sugar content. The DNS method measures the reducing ends of sugars and polysaccharides present in the EBN hydrolysate samples, as shown in Table 5. EBNhAand EBNhCat 1 h of hydrolysis time have a significant higher (P≤ 0.05) content of reducing sugar ((8.83 ± 0.34)% GE and (8.78 ± 1.86)% GE,respectively) while EBNhBat 4 h of hydrolysis had the significantly lowest (P≤ 0.05) reducing sugar content ((4.96 ± 1.37)% GE).The two-way ANOVA analysis showed that the sample grade and hydrolysis time had a significant effect (P≤ 0.05) on the reducing sugar content. However, no significant effect (P> 0.05) was found when the two factors interacted (Table 4). Correspondingly, there was no significant increase in reducing sugar concentration of the cleaned EBN sample during 1–3 h of hydrolysis duration [18]. A significant decrease (P≤ 0.05) in reducing sugar was observed in EBNhAduring 1 and 4 h of hydrolysis duration. This could be due to the depletion of sugar during enzymatic hydrolysis when it was exposed to 60 °C for a long period of time, decreasing the reducing sugar content. Li et al. [42]demonstrated that sugar was well preserved at a lower temperature of 30 °C and degraded at 50 °C.

3.5 Antioxidant activities

The antioxidant capacity of EBN hydrolysates showed very interesting results (Table 6), in which DPPH and ABTS scavenging activities of lower grade EBN samples (EBNhCand EBNhD) were significantly higher (P≤ 0.05) compared to that of the higher grade EBN samples (EBNhAand EBNhB) throughout the 4 h hydrolysis process. This may be due to the loss of valuable and high bioactivity soluble glycopeptides during the washing and cleaning of the EBNhAand EBNhBsamples. The DPPH and ABTS scavenging activity of EBNhCat 3 h hydrolysis time were the significantly highest (P≤ 0.05) at (43.65 ± 0.53)% and (92.52 ± 0.40)%, respectively.In contrast, EBNhBshowed the significantly lowest (P≤ 0.05) DPPH (3 h hydrolysis) and ABTS (1 h hydrolysis) activity at (19.67 ± 0.71)% and (58.87 ± 0.36)%, respectively.

Table 6Antioxidant capacity of different grades of EBN hydrolysates under different hydrolysis time (n = 3).

Statistical analysis showed that the main factors and inter-factor interactions had a significant effect (P≤ 0.05) on DPPH and ABTS scavenging activities. In general, increased DPPH scavenging activity with increased hydrolysis time was observed in EBNhBand EBNhDonly, indicating that more free radical scavenging glycopeptides were released during hydrolysis. Similarly, Ghassem et al. [43] reported that higher DPPH scavenging activity was attributed to exposure of more bioactive EBN peptides with increasing hydrolysis in EBN samples. Aside from that, the high scavenging activity of the ABTS cation radical indicates that the antioxidant compounds are mostly hydrophilic amino acids that donating hydrogen to stabilise the ABTS radical [44].

Meanwhile, the FRAP activity of all the EBN hydrolysate samples ranged from (0.47 ± 0.01) μmol/L AAEAC to (0.53 ± 0.04) μmol/L AAEAC,with no significant difference (P> 0.05) for all samples except EBNhBwith 4 h of hydrolysis time, showed the significantly lowest (P≤ 0.05) FRAP activity. Samples grades, hydrolysis time and inter-factor interactions also showed no significant impact(P> 0.05) on FRAP activity for all samples. Samples were found to demonstrate different antioxidation activities with different assays.This proved that upon enzymatic hydrolysis, different bioactive peptides were exposed, and exhibited different antioxidative mechanisms, which were analysed using different antioxidant tests.In addition, additional treatment, such as microwave and vacuumultrasound combined with alcalase, has been shown to result in increased hydrolysate yield and antioxidant properties of sesame bran(a by-product of the sesame industry) as compared to the conventional alkaline treatment [36,45].

3.6 Probiotic activity

The activity ofL. plantarumbacterial population in different grades of EBN hydrolysates is presented in Fig. 4. A preliminary study (results not shown) was conducted on all samples at 1–4 h of hydrolysis time and EBN hydrolysates produced at 2 h of hydrolysis time showed significantly higher (P≤ 0.05)L. plantarumbetween 24 and 48 h of the fermentation period. Hence, the EBN samples at 2 h of hydrolysis time were selected and used as the sample for the determination of EBN hydrolysates as potential prebiotic material.

From Fig. 4, for thefirst 24 h, all EBN hydrolysate samples except EBNhDdemonstrated significantly higher (P≤ 0.05)L. plantarumin comparison with the positive control, FOS. However, after 48 h of fermentation, only EBNhCand EBNhDexhibited significantly higher (P≤ 0.05)L. plantarumthan FOS. These results were very interesting because the lower grade EBN samples had higher probiotic activity as the growth of the bacteria was greater than the positive control. Meanwhile, by comparing the fermentation time, EBNhDhad the significantly highest (P≤ 0.05) increased inL. plantarumfollowed by EBNhC, as compared to all samples and FOS. This is a very impactful finding, as it indicates that low grade EBNs exhibit high probiotic activity, although they were developed from low value EBN, which are essentially by-products. Overall, this study proved that all EBN samples (raw and hydrolysate) of different grades had a high potential as a prebiotic material, with lower grade EBNhCand EBNhDexhibited higher probiotic activity.

Fig. 4 Growth of L. plantarum of various grades of EBN samples for 24–48 h of fermentation time (n = 3).

4. Conclusions

EBN samples from various grades, including by-products, were enzymatically hydrolysed to recover the valuable glycopeptides, with high recovery yields of 89%–99%. The physicochemical properties of the recovered EBN hydrolysates showed comparable results with the high grade processed cleaned EBN. It is interesting to note that the by-products of the EBN cleaning process (EBNhC) showed significantly higher (P ≤ 0.05) antioxidants and probiotic activity. In conclusion, this study confirms that hydrolysates produced from low grade EBNs (EBNhCand EBNhD) have comparable quality to that of the cleaned and semi-cleaned EBN (EBNhAand EBNhB). Thisfinding could be very beneficial to the EBN industry to reduce waste and EBN processing costs, as well as introducing variation of products to the consumers. Ultimately, this study could revolutionise the EBN processing industry.

Con flict of interest

The author declared they have no con flict of interest.

Acknowledgements

This research was funded by the Research Excellence Consortium(Konsortium Kecemerlangan Penyelidikan) (KKP/2020/UKMUKM/5/1) (JPT(BKPI)1000/016/018/25 (21)) and the Fundamental Research Grant Scheme (FRGS/1/2019/WAB01/UKM/02/1), both provided by Ministry of Higher Education, Malaysia. The authors would like to recognise the Innovation Centre for Confectionery Technology (MANIS) and Department of Food Sciences, Faculty of Science and Technology, Universiti Kebangsaan Malaysia, and Mobile Harvesters Malaysia Sdn. Bhd. for providing the necessary facilities and samples for this research.