Effect of pulsed light treatment on the phytochemical, volatile, and sensorial attributes of lactic-acid-fermented mulberry juice

ABSTRACT Lactic-acid-fermented mulberry juice (LFMJ) was subjected to pulsed light (PL) treatment at exposure time of 2, 4, and 8 s at high insensitive pulses of 14.0 J/cm2. The effect of PL treatment on the microbial inactivation, physicochemical, phytochemical, volatile, and sensory characteristics of LFMJ was evaluated. It was found that the PL was able to reduce the microbial load to acceptable levels (1.02 ± 0.04 log10 cfu/mL) with no significant impact on the physicochemical properties of LFMJ. It was also observed that the PL treatment caused a slight decrease in anthocyanin concentration at 8 s exposure time. The color difference (∆E) of the juice treated for 2 and 4 s fell below the slightly noticeable range 0.5<ΔE<1.5 while ∆E values for the 8 s (0.55 ± 0.02) and the thermal (0.50 ± 0.02) treated samples were slightly noticeable. The volatile profile and odor activity values were positively affected by increasing the exposure time. The results depict that, under the present experimental conditions, the application of the PL resulted in a fermented juice with superior quality attributes as compared to the thermal treated juice.


Introduction
Mulberry is a monoecious plant cultivated in Africa, Asia, Europe, South America, and North America. [1] It belongs to the genus Morus from the Moraceae family. There are 24 species of Morus and one subspecies, with at least 100 known varieties. [2] The fruit has been extensively used in traditional medicine in treating ailments such as malaria, diabetes, hepatitis, dermatitis, atherosclerosis, asthma, and rheumatism. [3] Besides, it is regarded as a nutritious food that varies according to variety, environmental factors, and geographical location. [4] The fruit is very perishable [5] hence the need to preserve it from postharvest loss. To address this challenge, the fruit is often processed into fruit jellies, jam, wines, and nonalcoholic beverage, among others. [6,7] To date, the application of conventional thermal processing of fruit juices remains the most common technique for preservation and shelf-life extension of fruit juice because of its ability to inactivate microorganisms and spoilage enzymes. However, heat processing particularly under certain conditions have been reported to induce detrimental effect on the nutritional, phytochemical, and sensory qualities of food [8][9][10] as well as reducing the content or bioavailability of some bioactive compounds. [9,11] Consumer's demand for nutritious, natural, and healthier foods, which are minimally and naturally processed makes it imperative for food manufacturers to explore various technologies that impact less on the quality of processed foods. This has necessitated the need for exploration into novel nonthermal pasteurization technologies such as high hydrostatic pressure, pulsed electric fields, pulsed magnetic field, pulsed light, and ultrasonication, among others, to complement or replace the conventional heat

Microbiological enumeration
The microbial assay for the fermented samples was performed using the plate count method described by Donsì et al. [13] with slight modifications. Briefly, ten-fold dilution series of samples were prepared using sterile distilled water and 0.1 mL aliquots of appropriate dilutions pour plated using plate count agar (PCA). Plates were incubated at 37°C for 36 h, and colony forming units (CFU)/mL were estimated.

Pulse light treatment
Pulsed light (PL) treatment was carried out using RS-3000B Steripulse-XL system (Xenon Corporation, Wilmington, MA, USA) as described by Palgan et al. [17] with slight modifications. The sample was dispensed into petri dishes (100 mm diameter) to ensure that the entire dish surface was covered with the sample to a depth of 1 mm. To minimize temperature increases of the samples, petri dishes containing the samples were cooled to 3°C and positioned on a stainless-steel shelf located at 2.5 cm of vertical distance from the quartz window of the Xenon flash lamp (Model No. RC 747, Xenon Corporation MA, USA). To reduce differences in radiation dose absorption, the samples were placed within a uniform area of the radiation field. Light pulses were delivered at a pulse width of 360 µs, frequency of 3 Hz, and delivering radiant energy of 1.17 J/cm 2 /pulse. The samples were exposed to a high intensity light pulse of 14 J/cm 2 for 2, 4, and 8 s.

Thermal treatment
The thermal pasteurization was carried out as described by Caminiti et al. [24] with modifications. Briefly, the tubular heat exchanger (Model No. FT74T, Armfield, Ringwood, UK) was thoroughly cleaned according to the manufacturer's instructions. Thereafter, the juice was pasteurized at a flow rate of 94 mL/min with the temperature of the holding tube set at 72°C, a residence time of 26 s, and a come-up time of 3.5 min. All other important parameters were monitored using the equipment's logging system.
pH titratable acidity and total soluble solids Total soluble solids (TSS) were determined using a digital refractometer (Beijing Yaxingtai Electrical Equipment, Beijing, China). The pH of the samples was measured using LIDA Instrument PHS-3C Precision pH/mV meter (LIDA Instrument, Shanghai, China). Titratable acidity (TA) was determined using the method described by Bhat et al. [25] with some modifications. Briefly, 3.0 g of the sample was allotted in a 200.0 mL beaker after which 30.0 mL distilled water was added. Using phenolphthalein as an indicator, the mixture was titrated against a standardized 0.1 N NaOH. The total acidity in gram of lactic acid per liter of the juice was calculated using the following equation (Eq. 1): where v is the titer value of NaOH, z is milliequivalent weight of lactic acid (0.09), and m is the mass of the sample.

Total phenolics concentration
Total phenolic concentration (TPC) was estimated using the Folin-Ciocalteu method described by Aydın et al. [26] with some modifications. A 2 mL freshly prepared Folin-Ciocalteu reagent (1:1 v/v) and 2 mL of sodium carbonate (75 g·L −1 ) were added to 0.2 mL of the sample in a test tube, and vortexed for 60 s. The mixture was incubated at 25°C for 30 min, and the absorbance was read at 760 nm using UV spectrophotometer (UV-1600, Beijing Rayleigh analytical instrument, Beijing, China). The TPC was expressed in terms of milligrams of gallic acid equivalent per milliliter of juice.

Total anthocyanin concentration
The total anthocyanin concentration (TAC) was determined by the pH differential method described by Kwaw et al. [7] Two buffer solutions: CH 3 COONa (0.4 mol·L −1 ) at pH 4.5 and KCl (0.25 mol·L −1 ) at pH 1 were prepared. A 0.1 mL each of the sample was dispensed into two sets of tubes. One set of the sample was adjusted to 10 mL with the KCl buffer while the other set was adjusted with the CH 3 COONa. The tubes were vortex, and their absorbances were read at 510 nm and 700 nm using UV spectrophotometer (UV-1600) against a blank (water and reagents). The TAC was calculated using Eq.
(2) and expressed as milligrams equivalent of cyanidin 3-glucoside per milliliter of the juice.
where A 1 is the absorbance at 510 nm at pH 1.0, A 2 is the absorbance at 700 nm at pH 1.0, A 3 is the absorbance at 510 nm at pH 4.5, A 4 is the absorbance at 700 nm at pH 4.5, and MW is the molecular weight of cyanidin-3-glucoside (449.2 g·mol −1 ); DF is the dilution factor (100); L is the path length (1 cm); ɛ is the molar extinction coefficient for cyanidin-3-glucoside (26,900 L·mol −1 ·cm)

Total flavonoid concentration
The aluminum chloride colorimetric method as described by Tchabo et al. [27] with slight modifications was employed in the determination of the total flavonoid concentration (TFC) of the juice. Briefly, 0.3 mL NaNO 2 (50 g/L) and 4 mL distilled water were added to the sample (1 mL) and vortexed for 60 s. The mixture was allowed to stand for 5 min after which 1 mL of AlCl 3 (100 g/L) was added, vortexed, and allowed to stand. After 5 min, 2 mL of NaOH (1 mol/L) was added and the final volume was adjusted to 10 mL with 2.4 mL distilled water. The mixture was incubated at 25°C for 10 min with 2 min periodic vortexing (15 s). Thereafter, the absorbance was read at 510 nm using UV spectrophotometer (UV-1600).
The TFC was expressed as milligrams of rutin equivalents per milliliter of the juice.

Volatile compounds and odor activity analysis
The headspace gas chromatography-mass spectrometry method was used in the analysis of the volatile compounds. The odor activity value (OAV) for each volatile component was valued by dividing its concentration by the odor threshold. [28] Solid-phase microextraction. The headspace solid-phase microextraction (HS-SPME) method described by Qin et al. [29] with slight modification was used in the extraction of the volatile compounds. Briefly, 1.5 g NaCl was added to 5 mL of sample in a 15.0 mL glass vial. Thereafter, 2-octanol (10 µL, 800 µg/L) was added to the mixture, which was vial sealed with silicone septum and the mixture equilibrated for 20 min at 40°C. The DVB/CAR/PDMS fiber (50/30 μm) (Supelco, Pennsylvania, USA) was exposed to the headspace of the vial with continuous stirring at 750 rpm. After 30 min, the fiber was removed and desorption was carried out at 250°C for 5 min by introducing fiber into the injection port of the Agilent 6890N-5973B gas chromatograph (GC) coupled with a mass spectrometer detector (MSD).
Identification and semi-quantification of volatile compounds. The identification of the volatile components was carried out by comparing their GC retention times with those of pure standards carried out under the same chromatographic conditions and their mass spectra with those in library databases (Wiley Spectral Library and NIST Library 2005 v 2.0). Series of n-alkanes (C5-C25) (Sigma-Aldrich) were run under the same conditions and used in calculating the linear retention indices (RI) of the compounds. [31] Color assessment Color measurement of CIE parameters (L*, a*, and b*) as described by Fazaeli et al. [32] using Hunter colorimeter model color Quest XE (HunterLab, Reston, USA) was carried out on the samples. The total color difference (ΔE) was calculated as the chroma (C) as (a *2 + b *2 ) 1/2 , and the hue angle (H°) as tan −1 (b*/a*). L Ã o ; a Ã o ; andb Ã o are the untreated mulberry juice chromatic values obtained for brightness, red/green, and yellow/blue; whereas, L*, a*, and b* are the respective chromatic values for the treated samples.

Sensory evaluation
Sensory attributes (aroma, color, taste, flavor, mouthfeel, and overall acceptability) of the samples (FMJ, TFMJ, PFMJ-1, PFMJ-2, and PFMJ-3) were assessed using 45 semi-trained panelists (20 males and 25 females) made up of staff of the School of Food and Biological Engineering, Jiangsu University. The samples were randomly presented to the panelist in clear plastic cups (15.0 mL) with a 3-digit code number. Each panel evaluated the samples (10 mL) on a 9-point hedonic scale (1 = extremely dislike; 2 = very much dislike; 3 = dislike; 4 = slightly dislike; 5 = neither like nor dislike; 6 = slightly like; 7 = like; 8 = like very much; 9 = like extremely). [33] Statistical analysis The results from the study were presented as mean ± standard deviation of three independent treatments and assays. The analysis of variance (ANOVA) was carried out using OriginPro version 2015 (OriginLab, Northampton, USA). Differences were considered to be significant at p < 0.05 (95% confidence level) using the Tukey test.

Results and discussions
Physicochemical and microbial analysis The study (Table 1) showed no significant changes (p < 0.05) in the pH, brix, and titratable acidity of the treated samples regardless of the treatment applied compared to the LFMJ. Similar trends have been reported in the literature on juice subjected to PL treatments. [17,18,34] The counts of the estimated microorganism after the 36 h fermentation, thermal, and pulsed light treatments are presented in Table 1. The fermented mulberry juice (LFMJ) had 7.62 log CFU/mL, but the thermal treatment (TT) and pulsed light (PL) treatment significantly (p < 0.05) reduced the microbial load of the juice. However, the application of TT was less effective than the PL. Similar trends in microbial inactivation using PL have been reported in orange and apple juice. [17] The foremost outcome on microbial inactivation produced by PL has been ascribed to the photochemical action of the UV radiation that modifies the structure of the DNA, impaired replication, and gene transcription. [16] Besides, the release of energy from the numerous penetrating flashes per second that upsurges the rapid energy intensity also contributes to microbial inactivation. [35] This was attested in the study with increasing exposure time resulting in a significant (p < 0.05) log reduction in microbial load of the juice ( Table 1). This could be due to a progressive increase in the lethal effect that might have caused microbial cells inactivation. Other factors such as the food adsorption physiognomies or hypnotic power might have contributed to the level of inactivation with regard to the PL. [36] Phytochemical and color analysis The results on the influence of PL on the phytochemical properties of the LFMJ are presented in Table 2. The results depicted a slight increase in the TPC and TFC of the LFMJ [37] but a significant (p < 0.05) decrease in TAC of PLFMJ-3. The significant change in TAC of PLFMJ-3 could be due to adverse photochemical reactions that might have taken place with the exposure of the juice to the PL. [38] The results on the color attributes (L*, a*, and b*) and the total color differences (ΔE) as influenced by the PL are shown in Table 2. The lightness (L*) increased slightly as the exposure time increased (Table 2) but was only significant in PLFMJ-3 (8 s of treatment). This could be attributed to a decrease in monomeric anthocyanin concentration of the juice during the treatment. [38,39] This confirms the significant negative correlation observed in Table 3 between TAC and L* (r = −0.936) that is in line with the findings of Wang et al. [40] The study showed a slight decrease in b* values but the extent of the decrease was not significant (P < 0.05). It was observed that increasing the exposure time resulted in a slight decrease in a* and an increase in b* but were only significant (p < 0.05) in PLFMJ-3. The heat treatment (TLFMJ) on the other hand caused an increase in L* and b* that could be the result of the negative impact of thermal pasteurization on the phytochemical components. [41] The ΔE values increased with increasing PL exposure time (Table 2) but the ΔE values for PLFMJ-1 and PLFMJ-2 fell below the slightly noticeable range 0.5<ΔE<1.5 according to the classification reported by Cserhalmi, et al. [42] However, the ΔE for TLFMJ (0.50 ± 0.02) and PLFMJ-3 (0.55 ± 0.02) were slightly noticeable ( Table 2). These noticeable color differences could be as a result of the heat in the case of TLFMJ and photochemical reactions (PLFMJ-3) that resulted in the degradation of TAC of those juices. Anthocyanins are mainly responsible for the color of mulberry, [43] and hence TAC are associate with changes in L*, a*, and b*. [44] The decrease in TAC ( Table 2) led to a decrease in a* thus making the sample dark thereby an increase in L* ( Table 2). This presupposes that the slight degradation of anthocyanins with increasing exposure time might have resulted in the formation of new polymeric complex hence, the upsurge in b* and a decline in a*. [45,46] This confirms the significant positive correlation (r = 0.932) between TAC and a* and negative correlation (r = −0.936) between TAC and L* (Table 3). Besides, H°( Table 3) was found to be significantly negative correlated to TAC (r = −0.881), TFC (R = −0.967), and a* (r = −0.985) but positively correlated with L* (r = 0.975) and b* (r = 0.987). Meanwhile, C* was noted to be positively correlated with TAC (r = 0.946) and a* (r = 0.998) but negatively correlated with L* (r = −0.999), b* (r = −0.921), and H°(r = −0.972). These findings substantiate the notion that change in phytochemical components relates to CIE color components (a and b) as emphasized by H°and C*, which are quantitative and qualitative colorimetric information associated with color recognition by humans. [44,46]  Changes in volatile compounds of pulsed light treated lactic-acid-fermented mulberry juice The volatile compounds found in the mulberry juice samples are enumerated in Table 4. A total of 34, 55, and 36 volatile compounds comprising aldehydes, ketones, esters, alcohols phenols, furans, lactones, and acids with varying concentrations were determined in the RMJ, LFMJ, and TLFMJ samples respectively; whereas, 51 volatile compounds each were found in the PL treated samples ( Table 4). The decrease in the number of volatile compounds after the PL occurred mostly in the ketones that could be due to photochemical reactions during the treatment. There were, however, increments in the concentrations of the volatile compounds with increasing exposure time (PLFMJ-1 (102.70 ± 1.56 mg/L), PLFMJ-2 (106.26 ± 1.70 mg/L), and PLFMJ-3 (107.34 ± 1.79 mg/L)) compared to the fermented juice not subject to PL (LFMJ (101.08 ± 1.43 mg/L)). These results are in line with those of Matak et al. [47] who demonstrated that increasing exposure time of light treatment increased the concentration of volatile compounds in goat milk.

Odor activity values assessment
The core aromatic notes of the samples as illustrated by their odor activity value (OAV) are presented in Table 5. The PL treatments had positive impact on the OAV of the fermented sample with the highest OAV of 541.56 after exposure time of 8 s. The TLFMJ on the other hand recorded the least OAV of 259.08 (Table 5). This could be due to the effect of heat on the key odorants responsible for the aroma of the juice. [48] A total of 32 (48.48%) aromatic compounds out of the 66 volatile compounds identified (Table 4) were noted to be accountable for the aromatic notes of the juice (Table 5). Based on the classifications of odorants compounds (odorant with OAV > 1 are significant and those <1 but > 0.2 may have synergistic impact on product's aroma), [49,50] odorants that had significant impact (OAV > 1) on the aromatic tone of the PL treated samples were 2-phenylethyl acetate, (E)-β-damascenone, acetaldehyde, 1-nonanal, 3-ethoxypropan-1-ol, heptyl alcohol, ethyl acetate, isopentyl acetate, ethyl hexanoate, octanal, and hexyl acetate. In addition, ethyl butyrate, benzaldehyde, 2-phenylethanal, isobutyl acetate, butyl acetate, 1-hexanol, ethyl octanoate, and ethyl benzoate were found to have possible synergistic impact on the aromatic tone of the juice (Table 5).

Sensory assessment
The mean sensory scores of the juices are presented in Fig. 2. The fermented samples were scored higher in all the assessable areas compared to the unfermented juice. [51] Among the fermented samples, the Pl treated samples had higher mean scores in terms of color, aroma, taste, flavor, and mouthfeel whereas the thermal treated sample had lower scores. These confirm the detrimental effect of thermal treatment on the juice color (Table 2) and the OAV (Table 5) and the positive influence of PL on the volatile and phytochemicals characteristics of the juice as discussed earlier. The flavor of the PL treated juices as perceived by the panelist followed a similar trend as the OAV of the juices ( Table 5). The overall acceptability scores depict that PLFMJ-2 was the most preferred (Fig. 2).

Conclusion
This study demonstrated that pulsed light is effective in reducing microbial levels in fermented mulberry juice without changing their physicochemical properties. Pulsed light also enhances the total phenolics, volatile composition, and the aromatic tones of lactic-acid-fermented mulberry juice with increasing pulsation time. The results also suggested that total anthocyanin concentration of lactic-acid-fermented mulberry juice is unlikely to be damaged under short PL exposure  Table 4. RMJ: raw mulberry juice; LFMJ: lactic-acid-fermented mulberry juice; TLFMJ: thermal pasteurized fermented mulberry juice; PLFMJ: pulsed light pasteurized fermented mulberry juice (exposure time of 1 -2 s; 2 -4 s; 3 -4 s) PCprincipal component time. The sensory analysis revealed that the nonthermal treated fermented juices were more preferred than the pasteurized fermented juice. This clearly shows that the integration of PL into juice processing lines could be considered a modest and cost-effective substitute to improve the safety and quality of juices.