Ultrasound aided debittering of bitter variety of citrus fruit juice: Effect on chemical, volatile profile and antioxidative potential.

In the present study, sonication assisted debittering of pomelo fruit juice was carried out and the effect of sonication along with resin/enzyme on the chemical, phytochemical and volatile composition of juice was also investigated. The optimum conditions for sonication coupled debittering using resin were 50 kHz, 2 min, and 45 ℃ while 50 kHz, 60 min, and 60 ℃ were obtained for enzyme hydrolysis. Sonication treatment not only reduced the debittering time but also enhanced the adsorption and hydrolysis of naringin by 17% and 20% in resin and enzyme respectively. In addition, enzymatic activity was also improved and weakened C-O bonds in naringin. At the same time, sonication significantly affected the bioactive compounds of juice, chemical composition, and volatile compounds of juice. Flavor compounds including octanal, linalool, citral, and ethyl butyrate were enhanced by sonication-assisted enzymatic treated juice.

Introduction

Citrus bitterness is a major problem observed in a certain variety of citrus fruits including grapefruit, Kinnow mandarin, Valencia orange, pomelo, etc. Despite the appreciable nutritional properties of these citrus fruits, still face challenges in the acceptance and commercialization. Also, in comparison to the other climacteric fruits, products from these fruits are very limited. It has been reported that bitterness may create problems during the storage of extracted juice due to the presence of naringin, nomilin, hesperidin, nobiletin, tangeretin, and limonin. However, naringin is responsible for prime bitterness, whereas limonin is responsible for delayed bitterness [40], [43], [61]. Previous studies have shown that the presence of these bitter chemicals causes decreased acceptance, economic loss, and quality degradation [28]. Bitterness levels are also affected by the age of the fruits, extraction procedures, pH, climate, and geographical factors [25], [22]. Number of debittering methods including physical, chemical, and microbiological methods have been reported and their merits and demerits been discussed by previous authors [40].

Resin-based physical adsorption and enzyme- or bacteria-based bioconversion ([43], [58], [62]) were employed for debittering of citrus juice. Physical adsorption considerably reduced the bitterness and bioactive compounds in orange juice [46]. A. niger has potential to hydrolyze the naringin into non-bitter compound [37], [38]. Beside these reports, few reports have also emphasized the enzymatic modification of naringin [37]), where the isolation, purification and characterization from Bacillus strain was detailed and studied the hydrolysing effects on naringin and other aglycones flavonoids [65].

Ultrasound (US) is currently gaining popularity as a new technique for reducing processing stages, improving quality, and ensuring the safety of food items. [3]. Furthermore, sonication is widely recognised as a green, innovative, cost-effective, and rapidly evolving technique that is scalable [53], implying its potential for industrial production. Ultrasound is commonly employed in freezing, extraction, separation, emulsification, tenderization, crystallisation, desiccation, filtering, quality control, meat processing, and functionality modification, among other applications [6], [9], [14], [23], [30]. Several recent studies have suggested that the mechanical, chemical, and cavitation effects of ultrasound can rectify the structures of enzymes and their substrates, facilitating interaction reactions and lowering activation energy, as well as increasing enzymatic hydrolysis of substrate and enzymatic reaction velocity [11], [23], [52].

Bhat et al. [4] applied sonication treatment to kasturi lime for 60 min and observed the significant retention of bioactive compounds and reduced the microbial load. Gao et al. [17] also reported the influence of sonication along with Apergillus niger koji extract on Ougan juice where hydrolysis efficiency enzyme was improved, however; debittering time was reduced significantly. This study reported the significant impact on volatile compounds of juice during sonication. Also, impact on phytochemical compounds and antioxidant activity of juice is still need to explore. Many studies have suggested that sonication as a strategy for preserving the quality of fruit juices by preserving beneficial bioactive components while decreasing pathogenic bacteria. Despite numerous investigations, however, there is little scientific literature on the effect of ultrasound on bioactive components in bitter citrus juice with regard to bitter compounds removal and improving the quality attributes of juice. So, keeping in view the nutritional value of citrus juice and increasing trend toward commercialization and acceptability of nutritious juice, this study was designed to debitter pomelo juice with minimal processing.

Thus, the present study is undertaken to investigate the (a) hydrolysis efficiency of naringinase in the presence of sonication; (b) study the impacts of ultrasound-assisted debittering on the possible mechanisms to enhance the degrees of hydrolysis of bitter compounds in juice; finally, (c) assessing the effect of sonication on the chemical composition, volatile compounds, and phytochemical properties of pomelo juice.

Materials and methods

Raw material and chemicals

Mature pomelo fruits (Citrus grandis) were obtained from the department of horticultural of Tezpur University. Aspergillus niger culture (ATCC 16888TM) was procured from KWIKSTIK Plus. Resin (amberlite IRA-400Cl-), analytical grade naringin and other necessary chemicals used in the present study were procured from Merck-Sigma Aldrich, India. Also, sugars and organic acids were of HPLC grade and purchased from Sigma-Aldrich.

Juice extraction and debittering of pomelo juice

The fruit of optimum maturity was collected from healthy pomelo trees and peeled, with the albedo, flavedo, membrane, seed, and juice sacs being meticulously separated using a scalpel and tweezers. A laboratory-scale juicer (Philips HR1832/00) was used to extract the juice from the separated juice sacs. To remove the remnants, the juice was passed through a 3-layered cheesecloth, then centrifuged for 10 min at 4° C and 5000 rcf to obtain a clear solution. For further examination, the clear fluid was stored in screw-capped odour-free vials at −20 °C.

Optimization of ultrasound coupled debittering parameters using response surface methodology (RSM)

With three independent variables and four layers of central composite rotatable design (CCRD), 20 tentative runs were constructed, including six imitates at the centre point. The independent factors were frequency (A), time (B) (on-time 10 s and off-time 5 s), and temperature (oC), which ranged from 10 to 50 kHz, 0–120 min (for resin: 0–10 min), and 20–70 °C, respectively. The degree of naringin hydrolysis was chosen as the preferred dependent variable and a quality criterion for debittered juice during sonication. A significance test with a 95% confidence interval was performed using the total error criteria. Analysis of variance (ANOVA) was used to determine the significance of each result.

Experimental condition for resin debittering

The ultrasonic bath was utilised for ultrasonication treatment (amplitude: 100%, power: 100 W, Genaxy Scientific Pvt. Ltd., India). 100 mL of juice containing the optimal amount of activated resin juice was sonicated with occasional shaking in a 250 mL beaker at the optimal frequency (30–50 kHz), time (0–10 min), and temperature (20–70 °C). After the juice had been ultrasonically treated, the resin was centrifuged off and the juice was filtered. The juice that had not been sonicated was used as a control for the resin debittered juice. For further investigation, the treated juice was kept at −20 °C.

Enzymatic debittering

The enzyme was purified, characterised, and kept at 4 °C for the debittering test as described in our earlier study [62]. In a conical flask, freshly extracted juice (100 mL) was combined with 2 mL enzyme (52.8 U/mL activity) and 2 mL phosphate buffer solution (pH 7.20) and treated at optimal frequency (10–50 kHz), time (0–120 min), and temperature (20–70 °C). The juice that had not been sonicated was used as a control for the enzyme debittered juice. For further examination, the debittered juice was stored at −20 °C. To determine the rate of naringin hydrolysis, the reducing sugar of treated juice was calculated using the Dinitrosalicylic (DNS) acid method [24] where 1 mL of juice was mixed with 80% ethanol. After mixing in 3 mL of DNS, the solution was heated for 15 min at 90 °C. After allowing the solution to cool to room temperature, 1 mL sodium potassium tartrate tetrahydrate solution was added. Using a UV–Vis spectrophotometer, the solution's intensity was then measured at 575 nm (Cary 60, Agilent UNICO Products and Instruments Inc., Shanghai, China).

Effect of sonication on enzymatic activity

The impact of sonication on α-L-rhamnosidase activity of the juice was examined as per Li et al. [31]. The reaction was carried out by incubating the enzymatic hydrolysed mixture consisting of enzyme (50 μL), naringin solution (1 mL of 300 μg/mL) and citrate buffer (0.95 mL of 20 mM, pH 4.0) at 60 °C for 15 min followed by heating at 100 °C for 30 min to deactivate the enzyme. Deactivated mixture was filtered through an inorganic membrane (0.22 μm) and the naringin concentration of juice was estimated according to Mishra and Kar [35], where 0.1 mL of juice was mixed with 0.1 mL 4 N NaOH followed by the addition of diethylene glycol (9.8 mL). The solution was incubated at 40 ℃ for 30 min and then the intensity of the solution was measured at 420 nm using a UV–Vis spectrophotometer against diethylene glycol as a blank. Control experiment was performed using deactivated enzyme extracts instead of active enzyme. One unit (U) of activity was defined as the quantity of β-glucosidase required to hydrolyze naringin and release 1 μM of glucose per min at 60 °C. Inactivated β-glucosidase was treated as the blank control [66].

Phytochemical and antioxidant properties of juice

Naringin content of juice was determined using the method suggested by Mishra and Kar [35]. The total phenolic content (TPC) was determined using the Folin-Ciocalteu reagent (FCR) method as per Mishra and Kalita [34] with slight modification. In brief, extract (1 mL) was added in 0.25 mL of FCR (1:2 diluted) and then incubated at 40 ℃ for 30 min followed by the addition of 1 mL of sodium bicarbonate solution (20%). The mixture was vortexed and incubated for 60 min at 40 ℃, and then the intensity of developed color was analysed using a UV–Vis spectrophotometer at 750 nm. The result was expressed in terms of gallic acid equivalent (GAE)/mL. The antioxidant property of juice extract was analyzed in terms of DPPH* scavenging activity and determined according to Gupta et al. [19]. Briefly, extract (0.5 mL) was added in DPPH solution (75 mmol/L) prepared in methanol and incubated in the dark at 37 ℃ for 30 min. The change in the color of the solution was observed at 517 nm against methanol using a UV–Vis spectrophotometer. Total flavonoid content (TFC) was calculated using the method of Zhishen et al. [57] with slight modification, where extract (2.5 mL) was mixed with 5% sodium nitrate solution (150 µL) followed by the addition of 10% aluminum chloride (150 µL). The mixture was vortexed and 1 mL of NaOH was added after 5 min. Then, the absorbance of the solution was taken at 510 nm against distilled water using a UV–Vis spectrophotometer. The result was expressed in terms of quercetin equivalent (QE) mL−1.

For the superoxide radical scavenging activity of juice, method of Cavia-Saiz et al. (2010) was adopted where the activity was determined by monitoring the reduction of nitroblue tetrazolium (NBT). The reaction mixture contained 20 µL of the juice sample, 78 µM NADH, 50 µM NBT and 10 µM PMS in Tris-HCl buffer (16 µM, pH 8.0) in a total volume of 2.5 mL. The samples were incubated at room temperature for 5 min and the absorbance was read at 560 nm. The OH– radical scavenging activity of juice samples was determined according to Halliwell et al. [21] where 10 mM deoxyribose, 50 mM phosphate buffer, pH 7.4, 1 mM ascorbic acid, 100 mM ferric chloride, 1 mM hydrogen peroxide, 10 mM EDTA and 20 µL of the juice samples was taken in a test tube and incubated for 60 min at 37 °C. Afterwards, 1 mL of the incubated sample was mixed with 0.5 mL of trichloroacetic acid (2.8%, w/v) and 0.5 mL of thiobarbituric acid reagent (1% w/v, in 0.05 M NaOH), followed by heating at 100 °C for 15 min and subsequent cooling to room temperature. The intensity of the solution was measured using a UV–vis spectrophotometer at 750 nm.

Fourier transform infrared (FTIR)

To study the effect of sonication on naringin, 20 ppm of naringin solution was prepared in deionized water and the pH of the solutions was adjusted to 3.5. The sonicated naringin solution under the optimized condition (40 kHz and 60 min) and the non-sonicated naringin solution was designated as samples and controls, respectively. The FTIR spectra of naringin and debittered pomelo juice was obtained through scanning from 4000 to 400 cm−1 using Impact 410 spectrometer (Nicolet, USA, Software: Omnic ESP50) [17].

Effect of debittering on chemical compounds of juice

Juice sample (5 mL) was mixed with ethanol solution (5.0 mL of 80%) at 40 °C followed by centrifugation at 7000 rpm at 4 °C for 15 min. The supernatant after the centrifugation was collected using a pipette and volume was made up to 25 mL with 80% ethanol solution. Extract (1 mL) was kept at 45 °C to evaporate ethanol followed by dissolving the residues in distilled water (0.5 mL). The solution was filtered through a Whatman syringe filter (0.2 µm) and further used for the analysis of soluble sugar and organic acids according to Cheong et al. [10]. HPLC (Waters Corporation, USA) equipped with two pumps, NH2 column (250 mm × 4.6 mm, GL Sciences, Inc., Japan), and RI detector was used for the quantification of soluble sugars present in the sample. The operating conditions for the quantification were set where acetonitrile: water (80:20, v/v) was used as the mobile phase at 1.0 mL/min. The standard sugar solution in the range of 0.1–4.0 mg/mL was prepared and was plotted to determine the concentrations of sugars in the juice. The same system was used for the detection of organic acids using a C-18 column (4.6 mm × 250 mm, Beckman, USA), and 0.1% sulphuric acid as the mobile phase with an isocratic flow rate of 0.4 mL/min. The organic acids were quantified at 210–260 nm while to avoid interference with other acids, ascorbic acids were dtected at 260 nm. For calibration, standard organic acid solution was prepared in the serial concentration of 0.3–3.00 mg/mL for ascorbic acid, succinic acid, and malic acid while 2.4–24.00 mg/mL was prepared for citric acid.

Analysis of volatile components of juice

Volatile components of pomelo juice were determined according to Cheong et al. [10] where 10 mL of juice was mixed with standard 5-methyl-2-hexanone (100 µL of 10,000 ppm) and were extracted twice with dichloromethane (8 mL). The extracted sample was vortexed for 30 min followed by centrifugation at 7000 rpm for 5 min at 4 ℃. The gel was broken up using a glass rod and collected the organic phase which was further dried with anhydrous sodium sulphate. The solution was filtered using a Whatman syringe filter (0.2 µm) and concentrated to a final volume of 0.10 mL using nitrogen stream. All the prepared extracts were kept at −20 ℃ before determination. Volatile compounds of pomelo juices were analysed using gas chromatography (GC) equipped (Clarus 600, PerkinElmer, USA) with FID and a 5975 inert MSD with software (Turbomass-Ver) according to Cheong et al. [10]. In brief, extract (1 µL) was injected directly into the GC injector under spitless mode. The concentration of juice's volatiles and aromatic components was measured in parts per million (ppm), with each compound’s relative peak area matched to the internal standard and the response factor. NIST 8.0 MS library was explored to identify the volatiles by comparing the mass spectra (National Institute of Standards and Technology, Gaithersburg, MD, USA), and volatile components were identified by matching the retention times with standard volatile and aromatic compounds.

Statistical analysis

IBM SPSS Statistics Version 20.0, Armonk, NY: IBM Corporation software was utilized for statistical analysis of the obtained data. The Duncan’s multiple range test (p < 0.05) was applied to separate the means. All the data were presented in the present study in the form of mean with the standard deviation. The present study used Design-Expert software version 10.0.7.0 (Statease Inc., Minneapolis, USA) to optimize the debittering parameters.

Results and discussion

Optimization parameters for debittering of pomelo juice

The debittering parameters i.e. frequency, time, and temperature were optimized by RSM-CCRD and the optimized condition was selected for further debittering of juice via US coupled with resin and US coupled with enzyme respectively. The fitted CCRD design was acceptable for significant regression with nonsignificant lack of fit and satisfactory R2 value for the naringin response, according to ANOVA data. The R2 value for naringin content using the US coupled resin debittering was found to be 0.96, while and 0.92 for the US coupled enzymatic debittering. The 3D graphical illustration for naringin response was established as a function of three independent factors (not reported here). The optimized process variables for the US coupled resin debittering such as time, frequency, and temperature were 2 min, 50 kHz, and 45 °C respectively, while 50 kHz, 60 °C, and 60 min were observed for enzymatic debittering. The predicted value for naringin content was 35 µg/mL (US + resin) and 415 µg/mL (US + enzyme) at desirability level of 0.98 and 0.96 respectively. Also, the predicted value was closed to the experimental value of 40.35 µg/mL and 430 µg/mL for resin and enzymatic debittering respectively. As a result, the chosen model may be utilised to successfully optimise the debittering parameters for removing naringin from pomelo fruit juice.

Debittering of juice

Naringin is a major bittering agent in the pomelo juice than limonin [39], thus hydrolysis degree of naringin was opted as a preferential dependent variable during optimization process for debittering using ultrasound.

Resin

The juice used in the present study had 1308 ± 10.52 µg/mL of naringin content. US combined resin debittering result in the naringin content of 40 ± 1.15 µg/mL while resin assisted debittered juice had 278 ± 3.25 µg/mL of naringin content. In the case of US, a significant reduction in the naringin content could easily be observed than only resin assisted, hence; the juice debittered using ultrasound and resin has a profound effect on the naringin content of juice. Such a considerable variation in naringin might be traced to the US, which altered the physiochemical properties of adsorbates and enhanced the availability of distinct binding sites of pomelo juice on the surface of adsorbents. The affinity between resin and juice has improved due to US treatment [13]. When the ultrasound was used, naringin adsorption was improved. This behaviour can be explained by the cavitation bubbles, which could improve naringin transit [16]. The findings showed that the amount of naringin transferred to the resin relies on the resin type, chemical matrix, surface area, and pore size. Surprisingly, naringin adsorption on polar resins may occur both on the surface and within the pores of the resins. The rationale is that water molecules containing phenolic compounds are easily transported into the pores due to the hydrophilic character of polar resins. Conversely, the matrices inside the non-polar resins contain non-polar covalent or polar covalent bonds, in both of which electrons are shared between the atoms involved, which makes it difficult for the electrons to be drawn away. As a result, the molecules are more difficult to break apart, and the flavonoids are unable to be adsorbed inside the pores. This suggests that physical interactions may be involved in the adsorption of flavonoids onto the surfaces of polar resins, aiding desorption, whereas chemical interactions may be involved in the adsorption of flavonoids inside the pore [32].

The adsorption of naringin in US coupled resin juice correspond to 96%; however, adsorption of naringin in only resin debittered juice was 78.73% under optimized condition as obtained from RSM-CCRD. From the result, it could be assumed that, ultrasound has the synergistic effect on the rate of naringin adsorption. Even, the sonication treatment found to reduce the treatment time (at least 2 min) which was beneficial to minimizing the unnecessary loss of vital nutrients of juice.

Enzyme

Juice debittered via US coupled enzyme and only enzyme had the naringin content of 430 ± 7.62 µg/mL and 684 ± 4.78 µg/mL respectively, which correspond to 67.14% and 47.67% respectively under optimized condition. In addition, sonication reduced the hydrolysis time by 33% resulting minimal loss of quality attributes and vital nutrients. Previous authors have reported the debittering mechanism of naringin [12], [55], [56]. Recently, Gao et al. [17] documented that, US assisted debittering of Ougan juice using Aspergillus niger koji at 40 kHz, 80 W/L and 90 min reduced naringin content significantly by 73.01%. Ultrasound can enhance the interfacial area between the phases of a heterogeneous liquid/liquid system through mechanical and vibrational effects. Furthermore, sonication, which enhances the breaking of naringin's C-O bonds, may increase the activity of naringinase [17]. Previous studies have found that sonication improves not just the hydrolysis of bittering compounds, but also the mass transfer and structure of bitter compounds [38,71].

Phytochemical potential

Total phenolic content (TPC)

In combined US and resin assisted debittering, TPC was significantly decreased from 1834 ± 20.74 µg GAE/ mL to 1542 ± 14.35 µg GAE/ mL, while only resin debittering reduced TPC from 1834 ± 20.74 µg GAE/ mL to 1632.80 ± 18.49 µg GAE/ mL (Fig. 1a). Breitbach et al. [5] revealed that ultrasound technique enhances the mass transfer process from juice to adsorbents and break the affinity between adsorbents and adsorbates by ultrasonic cavitation. Generally, reduction in TPC might be due to interaction between ion-exchange resins and phenolic compound. The binding of phenolic compounds to ion exchange resins is affected by interactions between functional groups, hydrophobic interactions, and hydrogen bonding. As the resin having hydrophobic composition of resin resulting higher binding capability than other matrices [26] and interaction of phenolic compounds and amberlite resin during debittering leads to a reduction in the phenolic content. At different pH levels, hydrophobic interactions and hydrogen bonding with anion exchanger resins revealed different mechanism. Under alkaline condition, adsorption and ion exchange were the most important processes. Because the phenols are undissociated and dispersion interactions prevail at acidic pH, the absorption of phenolic compounds by various adsorbents is improved [45].

Fig. 1

Impacts of ultrasound debittering on (a) Total phenol content (b) Total flavonoid content (c) Naringin content (d) DPPH* scavenging activity (e) Superoxide anion radical scavenging activity (f) Hydroxyl radical scavenging activity.

Enzyme debittered juice had a higher content of TPC than fresh juice. Juice treated with only enzyme result in enhanced phenolic content whereas 9.59% of increment was seen in US assisted enzymatic hydrolysis. An increased bioactivity of the juice upon combined treatment could be due to acoustic cavitation produce by US and controlled enzymatic cell wall degradation which facilitated the release of more soluble antioxidant polyphenols into the juice [33]. The result of present study was supported by Vila-Real et al. [50], who documented the similar observation and stated that the by-product of naringin i.e., naringenin, exhibit antioxidative and other bioactive properties. In case of US assisted debittering under optimized frequency and temperature, accelerated the contact between naringin and naringinase resulting enhanced biological activity and more production of by product [23]. Increased TPC could be due to the generation of OH– radicals to the aromatic rings of phenol compounds in the presence of sonication. Bhat et al. [4] observed the increased phenolic compounds in sonicated Kasturi lime juice which might be due to the disruption of cell walls as a result release of bound phenolics from the cell walls.

Total flavonoid content (TFC)

The traditional adsorption technique takes a long time to attain equilibrium, and flavonoids are removed at a rate that is less than satisfactory [54]. As a result, the traditional resin adsorption technique has failed to satisfy the demands of current industrial production, making it necessary to modify the method for the adsorption of flavonoids from target material.

In the present study, the flavonoid content of pomelo juice was 529 ± 17.75 µg QE/ mL and juice debittered by combined resin and US, result in a loss of 50% of flavonoid content (Fig. 1b). In general, the majority of phenols and flavonoids tend to form hydrogen bonding with the resin, which is likely to play a role in the decrease in the bioactive compounds [8]. The type of the adsorbent, solvent, and adsorbate are all thought to have a role in flavonoid adsorption [41]. The mechanism of flavonoid adsorption and removal from the target matrix has been documented in previous investigations. Flavonoids have benzene rings and many hydrogen groups connected to their aglycone, which helps the resin to adsorb these compounds if it has the right average pore diameter, polarity, and surface area. Some of the findings suggest that the first stage of flavonoid adsorption by resins is shown by the initial intra-particle transport of flavonoids, which is governed by a surface diffusion mechanism, whereas the latter stage is governed by pore diffusion. The purpose of the US in this study is to impart mechanical and cavitation effects to the pomelo juice, resulting in a considerable increase in flavonoids adsorption. Shen et al. [42] recently used macroporous resin to elucidate the likely mechanism of flavonoid adsorption from jujube peel. They observed that the US improved adsorption capacity mostly by strengthening hydrogen bond formation and increasing resin surface roughness. Pomelo juice debittered in combination of enzyme and US had a slightly higher value of flavonoid content (547.25 ± 13.25 µg QE/mL) than fresh juice (529 ± 17.75 µg QE/mL); which is concordant to the result of TPC. The elevation in TFC could be due to number of OH– groups present in naringenin than naringin [2]. US treatment might have contributed the more by-products of naringin resulting more OH groups and higher flavonoid content in debittered juice [2], [4].

Naringin content

Ultrasound also has a mechanical impact, allowing more solvent to penetrate the sample matrix and increasing the surface area of contact between the solid and liquid phases. This, along with increased mass transfer and severe cell damage via cavitation bubble collapse, results in increased intracellular product release into the bulk medium [48]. As indicated in Fig. 1c, naringin content was significantly influenced by the sonication treatment. It can be assumed that, sonication synergistically enhanced the catalytic activity of enzyme resulting the lower level of bitterness in pomelo juice. Several recent studies have suggested that the mechanical, chemical, and cavitation effects of ultrasound can correct the structures of enzymes and their substrates, then facilitate their interaction and lower their activation energy, and finally improve the enzymatic hydrolysis degree of substrate and enzymatic reaction velocity [23], [52]. According to Dabbour et al. [11], sonication increased the binding between alcalase and sunflower meal protein while lowering their reaction activation energy, resulting in a significant improvement in sunflower meal protein hydrolysis efficiency. According to the findings, sonication has the potential to improve the enzymatic hydrolysis of naringin in pomelo juice.

On the other hand, sonication synergistically enhanced the adsorption capacity of resin and thereof decreased level of naringin content in pomelo juice. It is because of the limited solubility of naringin in water and hydrophobic character. The adsorption of naringin requires a huge amount of surface area associated with resin and have the maximum tendency to get pushed from solution to the adsorbent surface [41]. This explains the maximum removal of naringin from the pomelo juice.

Antioxidant properties

DPPH* scavenging activity

The highest DPPH* scavenging activity was observed in US assisted enzyme treated juice followed by the only enzyme treated juice (Fig. 1d). Synergistic or additive effects are the most possible reason for increased DPPH* scavenging activity in US aided enzyme treated juice. Compounds like gallic acid, catechin, naringin, citric acid, vitamin C, and others, are well known for their antioxidative properties; thus, the interaction of these antioxidant constituents can result in synergistic or additive effects on scavenging free radicals [1], [47], resulting in increased antioxidant activity. However, DPPH* scavenging decreased significantly in resin treated juice. The findings of TPC of juice can be correlated with the findings of DPPH* scavenging activity. In addition, loss of ascorbic acid could be one of the probable reason for reduced DPPH* scavenging activity of debittered juice. Sonication treatment reduced the ascorbic acid in Kasturi lime juice [4] and tomato juice and these studies well supported the observation of present study.

Superoxide anion radical scavenging activity

This superoxide anion radical scavenging activity is performed to evaluate the effect of sonication on scavenging efficiency of O2 (Fig. 1e). In the case of only enzymatic treatment, slight enhancement (8%) of superoxide anion radical scavenging activity was noticed, however; combined US and enzymatic debittering, result in an increase of 42.2% of total activity. Sizable loss in superoxide anion radical scavenging activity was observed in resin treated juice where resin coupled with US reduced the activity by 54.78% and 41.14% respectively. These results are concordant to DPPH* scavenging activity where similar variation was seen. Cavia-Saiz et al. [7] also stated the similar observation for enzymatic and resin treated citrus juice. Končić et al. [27] also found a correlation between the content of phenolic compounds and the antioxidative activity of B. vulgaris and B. croatica roots, twigs and leaves.

OH– radical scavenging activity

Figure 1f illustrates hydroxyl scavenging activity of debittered juice; where activity was found to decrease due to sonication while non-sonicated juice presented the increased activity by 50%. The highest inhibition of OH– radical result in deoxyribose degradation and was observed in the juice obtained from enzyme treated juice. Resin debittered juice has shown decreased activity and the similar results were presented by Cavia-Saiz et al. [7]. Differences in scavenging activity might be due to alteration of one or more OH or oxidation during US debittering. It might also be attributed to loss of both 7-and 5-OH groups in the aglycone resulting decreased scavenging reaction and quenching efficiency of juice, thus decreased activity [7].

Effect of sonication on α-L-rhamnosidase and β-galactosidase activities

Impact of sonication on α-L-rhamnosidase and β-galactosidase activity of enzyme was determined (Fig. 2). Initially, the α-L-rhamnosidase and β-galactosidase activity was observed as 48 U/ml and 43 U/ml respectively. In the case of α-L-rhamnosidase, decrease in activity due to sonication can be seen with an increase of time of incubation. The degradation inactivity was significantly higher in US treated juice than non-sonicated counter parts (p < 0.05). Decrease in rhamnosidase activity was ranged from 0 to 48% in non-sonicated sample while 0–65% observed in case of sonicated samples. It has been reported that the sonication improves the hydrolysis of substrate resulting rapid formation of by-products [17].

Fig. 2

Influence of ultrasound debittering on enzymatic activity (a) α-L-rhamnosidase activity (b) β-glucosidase activity.

Alternatively, β-galactosidase activity was comparatively less affected than the activity of α-L-rhamnosidase. Enzyme assisted debittering retained the maximum activity with an increase of reaction time. The maximum activity was observed after 30 min of incubation and there after decreased. According to Huang et al. [23], low-intensity sonication at a suitable frequency can cause cavitation, magnetostrictive, and mechanical oscillation effects, which can enhance enzyme and substrate conformations, resulting in increased enzyme activity and reaction rate. Gao et al. [18] reported that frequency and intensity of sonication increased the activity during fermentation of soy sauce and the similar observation was made during the debittering of Ougan juice assisted by US [17]. From this activity, it can be assumed that, juice debittered along with sonication has substantially manage to retain the activity along with retention of vital nutrition in 60 min of treatment.

Impact of sonication on naringin content and structure of naringin

FTIR spectra was collected to study the effect of sonication on bond stretching of naringin (Fig. 3). IR study reveals the prominent effect on bond stretching of naringin and alteration of naringin structure when compared with non-sonicated naringin (Fig. 3a). A broad peak was observed at 3424 cm−1 which corresponds to OH stretching while peaks at 2974 and 2929 cm−1 designated to CH stretching. In addition, peaks at 1715, 1448, 915, and 518 cm−1 correspond to C = O stretching and C–H “oop” bend in the aromatic group respectively [17].

Fig. 3

FTIR spectra (a) Naringin (control and ultrasound treated), (b) Resin treated juice, and (c) Enzyme treated.

On contrary, profound changes in the bands and stretching were noticed due to US treatment along with enzymatic hydrolysis (Fig. 3b). However, fingerprint regions were not significantly affected by US treatment. The bond appeared such as OH stretch, C = C stretch, C = O at 3436, 2086, and 1639 cm−1 respectively. Peaks at 1408 represent O-C–H bending and deformation of carbohydrate while -C = O acid stretching was designated to 1234 cm−1 that also belong to the C-O stretch related to the liberations of rhamnose and glucose (Del Fresno et al., 2018). Peaks in the fingerprint regions, notably 1060, 994, and 729 correspond to -C = O acid stretching, trans = C–H, cis = C–H, C–H stretching, and vibration of sugars. Thus, enzymic activity leads to utilization of sugar and formation of alcohols and esters, these results are in agreement with previous studies [63], [64], [59]. The increased C-O stretch would obviously result in the fracture of C-O bonds in sonicated naringin. It would also make it easier for the naringinase to target the C-O bonds, resulting in maximum hydrolysis of sonicated naringin [17].

Effect of debittering on chemical composition

Non-volatile compounds including sugars and organic acids were estimated and are presented in Table 1. Fresh juice had a higher portion of sucrose (50 ± 2.53 g/l) followed by fructose (12 ± 0.61 g/l) and glucose (11 ± 0.57 g/l). Organic acids including citric acid (12 ± 0.62 g/l) was found as predominant followed by malic acids (1.5 ± 0.07 g/l), succinic (0.22 ± 0.01 g/l), tartaric acids (0.13 ± 0.01 g/l) and ascorbic acid (0.32 ± 0.01 g/l).

Table 1

Impact of ultrasound debittering on non-volatile composition of juice.

Non volatiles (g/l)	Fresh juice	Resin	Ultrasound + Resin	Enzyme	Ultrasound + Enzyme
Fructose	12 ± 0.61a	9 ± 0.45d	7 ± 0.35e	11 ± 0.55bc	11 ± 0.58b
Glucose	11 ± 0.57a	9 ± 0.46d	8 ± 0.40e	10 ± 0.50c	11 ± 0.55ab
Sucrose	50 ± 2.53a	34 ± 1.70c	27 ± 1.37d	48 ± 2.43b	49 ± 2.45ab
Citric acid	12 ± 0.62a	7 ± 0.38d	5 ± 0.25e	10 ± 0.52c	11 ± 0.55b
Malic acid	1.5 ± 0.07a	0.75 ± 0.03d	0.41 ± 0.02e	1 ± 0.052c	1 ± 0.07ab
Tartaric acid	0.13 ± 0.01a	0.07 ± 0.00d	0.05 ± 0.00e	0.09 ± 0.00c	0.11 ± 0.00b
Succinic acid	0.22 ± 0.01b	0.10 ± 0.01d	0.08 ± 0.00a	0.18 ± 0.01c	0.21 ± 0.01b
Ascorbic acid	0.32 ± 0.01a	0.21 ± 0.01d	0.12 ± 0.00e	0.30 ± 0.01b	0.28 ± 0.01c

Values are mean ± standard deviation (n = 15). Values followed by different superscript letter in a row are significantly different (p ≤ 0.05).

Sonication assisted debittering has a substantial impact on non-volatile compounds of juice. Resin assisted debittering result in loss of more than 30–50% of non-volatile compounds. Juice debittered with combined resin and US, had only 57.48% fructose, 69.63% glucose and 54.36% of sucrose. Sonication also reduced the organic acids where 58.66% of citric acid, 62.5% of ascorbic acid and 73.71% of malic acids in 2 min of juice treatment. Sizable loss in the sugars and acids was observed due to enhanced adsorption of the solutes because of sonication [46].

On the other hand, enzymatic debittering retained the maximum vital components of juice. The enzyme debittered pomelo juice had fructose (11 ± 0.55 g/l), glucose (10 ± 0.50 g/l), and sucrose (48 ± 2.43 g/l) which is higher than resin treated juice. In the case of, US combined enzymatic debittering found effective to retain sugars and organic acids and no significant difference was observed when compared to fresh juice. Decreased levels of ascorbic acid in the treated juice may be a result of oxygen exposure during processing. As it has been reported that, ascorbic acid is prone to oxidation, temperature, and light [36]. Similar observation was reported by Gao et al. [17] where no significant difference in the composition of US assisted debittered juice.

GC–MS analysis of volatile compounds

Citrus juice is well known for its volatiles including citrus-based flavors and aroma active compounds such as intense green, waxy odors, woody, and bitter taste (Cheong et al., 2011). Volatiles were extracted using dichloromethane, and major active compounds responsible for characteristics of citrus aroma and fruity odor were identified (Table 2). Like monoterpenes constitutes a major portion (approx. 90%) in which limonene and α-pinene are topmost compounds followed by octanal, linalool, and citral [60]. The significant compounds have a significant role in the flavor contribution of the pomelo juice. Terpineol is also identified and has been reported for off-flavor formed from limonene or linalool. In general, debittering of pomelo juice has drastically reduced the aromatic and volatile compounds by 60% which was further increased by 10% when sonication was applied.

Table 2

Volatile composition of fresh and debittered pomelo juice.

Volatiles (ppm)	Fresh juice	Resin	Ultrasound + Resin	Enzyme	Ultrasound + Enzyme
(R)-limonene	1668 ± 83.44a	792 ± 39.63c	662 ± 33.10d	1052 ± 52.6b	1031 ± 51.55b
Octanal	13 ± 0.65a	5 ± 0.26d	3 ± 0.19e	9 ± 0.48c	10 ± 0.50b
Linalool	21 ± 1.05a	15 ± 0.76d	12 ± 0.64e	19 ± 0.95c	19 ± 0.98b
Ethyl Butanoate	107 ± 5.36a	84 ± 4.20c	64 ± 3.23d	97 ± 4.88b	95 ± 4.78b
Terpineol	13 ± 0.68a	8 ± 0.43d	7 ± 0.37e	11 ± 0.56b	10 ± 0.56c
Citral	16 ± 0.81a	8 ± 0.44c	6 ± 0.31d	15 ± 0.76b	15 ± 0.78b
α-pinene	21 ± 1.06 a	12 ± 0.62c	9 ± 0.48d	19 ± 0.97b	19 ± 0.97b
Ethyl Butyrate	1020 ± 51.00a	596 ± 29.81c	542 ± 27.10 cd	780 ± 39.33b	790 ± 39.50b
2-phenylethanol	1736 ± 86.80a	1684 ± 84.21b	1544 ± 77.20c	1086 ± 54.30d	1006 ± 50.34de

Values are mean ± standard deviation (n = 15). Values followed by different superscript letter in a row are significantly different (p ≤ 0.05).

Juice obtained after the resin treatment had a sizable loss of volatiles where the reduction was ranged from 2.98 to 59.23%. During debittering, octanal, limonene, citral, α-pinene, and ethyl butyrate were decreased significantly (p < 0.05) by 41–59.23%. However, losses were increased in sonicated samples which were ranged from 11.05 to 70.46%. Percentages of reductions were from 16% to 61%. These findings are consistent with those of Kranz et al. [29], who found that debittering process had a significant influence on volatile taste components in grapefruit juice. The present findings revealed that limonene has a higher adsorption affinity for the adsorption resin and decreases in a higher proportion (50.04%). But Fernández-Vázquez et al. [60] documented that 61% reduction of octanal decreases during debittering of orange juice.

On the contrary, volatile compounds in the enzyme-treated juice were retained comparatively higher than resin treated. In the enzyme-assisted debittered juice, the reduction was observed from 9.18 to 37.44%. There was a slight effect of sonication was noticed where 2-phenylethanol reduced significantly followed by limonene and octanal. The results are comparable to the previous study done by Wahia et al. [51], who observed the impact of sonication on acetaldehyde and ethyl acetate in citrus fruit juice, while the opposite effect on d-carvone, ethyl butyrate, ethyl 3-hydroxyhexanoate, and (R)-limonene was noticed during sonication of juice.

Conclusion

From the present study, it can be concluded that sonication treatment at optimized conditions along with debittering agents results in a substantial effect on the naringin content of pomelo juice. The sonication treatment improves the adsorption efficiency of resin and markedly improved the hydrolysis efficiency of the enzyme. It also decreased the processing time by 30 min. The hydrolytic activity of rhamnosidase and glucosidase was also driven by sonication treatment and therefore, improved activity and lower level of bitterness in pomelo juice. The bioactive compounds of juice were also affected by the treatment where resin treatment results in the maximum loss of phytochemicals while enzymatic treatment improved certain activities. Also, enhancement of C-O stretch results in higher fracture possibility of the C-O bonds in the sonicated naringin thus driving the naringinase to attack the C-O bonds. Sizable loss in sugars and organic acids was noticed in ultrasound-assisted resin debittering. The volatile compounds in the enzyme-treated juice were retained comparatively higher than resin treated. The findings of the present study will be helpful to reduce the debittering time of citrus juice.

Industrial relevance

Improving the debittering efficiency and reducing the processing time of resin and enzyme through ultrasonication could make the process more economical. The undesired bitter taste due to naringin can be significantly reduced in a shorter time in ultrasound assisted debittering and thus the overall quality and consumer acceptability of juice can be increased.

CRediT authorship contribution statement

Arun Kumar Gupta: Data curation, Investigation, Methodology, Software, Validation, Visualization, Writing – original draft. Partha Pratim Sahu: Supervision, Writing – review & editing. Poonam Mishra: Conceptualization, Supervision, Validation, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.