ALTERATIONS ON PHENOLIC COMPOUNDS AND ANTIOXIDANT ACTIVITY DURING SOUR GRAPE JUICE CONCENTRATE PROCESSING

The alterations of phenolic compounds and antioxidant capacity of sour grape juice were investigated during the concentration process stages. Phenolics and antioxidant properties of the samples changed more at the vacuum evaporation stage than that of the other stages. After evaporation, the antioxidant capacity of the samples decreased approximately between 14.2 and 17.0 % for DPPH and ABTS methods, respectively. Besides, phenolic contents also decreased approximately as 14.8%. HPLC data on phenolics of sour grape juice during concentrate processing gave 12 polyphenols, including gallic acid, (+)-catechin, (-)-epigallocatechin, vanillic acid, (-)-epigallocatechin gallate, (-)-epicatechin, caftaric acid, caffeic acid and p -coumaric acid, which were determined as 1.05-1.83 mg/100 g, 5.40-7.83 mg


INTRODUCTION
Fruits play important role on the protection of human health due to their contents such as phenolics, minerals, vitamins and antioxidants. Grape (Vitis sp.) is one of the most produced fruits in the world. It is used for wine, juice, raisin, sour grape juice, vinegar and pekmez (grape concentrate, molasses). Sour grape is used for 'verjuice' or sour grape juice, consuming for salad dressing, processed vegetables and drinks as sherbet with sweeteners (Karapinar and Sengun, 2007;Nickfardjam, 2008). The sour grape juice has different names, such as 'verjuice', 'verjus' and 'koruk' juice, according to the producing country. Additionally, it has been traditionally produced for many years.
Grape includes polyphenols that have protective effect on human health (Chira et al., 2008;Xia et al., 2010;Toaldo et al., 2015). The phytochemical polyphenols have anticancer and anti-inflammatory effects in vitro (Castilla et al., 2006;Capanoglu et al., 2013). It was reported that these compounds inhibit cardiovascular diseases, some types of cancer cells, reduce plasma oxidation stress and aging effect (Meyer et al., 1997;Falchi et al., 2006;God et al., 2007;Xia et al., 2010;Tsanga et al., 2015). Many studies were performed related to the grape, wine and grape juice effects on human health although investigations on improving health of sour grape juice were fairly limited. ZibaeeNezhad et al. (2012) expressed that sour grape juice had improving effect on serum levels of HDL-C but no lipid-lowering effect on triglyceride and serum levels of LDL-C. Grape polyphenols show high antioxidant properties. For this reason, many researchers tried to determine the relationship between phenolic compounds and antioxidant activity of the grape (Xia et al., 2010). It was stated that significant correlations were found among phenolic compounds and antioxidant activity of grape, grape juice, grape concentrate and wine (Castilla et al., 2006;Paixao et al., 2007;Stratil et al., 2008;Gollucke et al., 2009;Buyuktuncel et al., 2014;Lima et al., 2014;Toaldo et al., 2015).
No study related to sour grape juice processing and concentration could be found in literature although many studies were conducted on grape juice processing and concentration (Gollucke et al., 2009;Capanoglu et al., 2013;Lima et al., 2014). On the other hand, the studies that were regarding phenolic compounds and antioxidant properties of sour grape juice were fairly limited.
In the current study, the alterations of phenolic compounds and antioxidant capacity of sour grape juice were investigated during the concentration process stages. The relationship between phenolic content and antioxidant capacity was also observed at six processing stages.

Sour grapes
Sour grape samples of Sultani seedless (Vitis vinifera L.) grape variety used for juice production were obtained from Manisa Viticulture Research Institute vineyards. After harvest, samples were immediately transferred to the grape processing unit of the Institute.
Sour grape samples were harvested before veraison period in 2015. No diseases and insect damage were detected in the sour grape samples at harvest; the cluster and berries of sour grapes had good sanitary appearance. The soluble solid value and total acidity of sour grapes were 9.5 °Brix and 30.45 g/L (as tartaric acid equivalent), respectively. The pH value of sour grape was 2.47.

Sour grape concentrate production
Sour grapes were rinsed to remove the dust, soil and other impurities after harvest. Then, stalks were discarded and clusters were passed through a crusher destemmer machine (Türköz Metal Makine, Turkey). The mash was pressed in a hydraulic press (Türköz Metal Makine, Turkey) and clear 'koruk' juice was obtained [A]. The juice was kept at 2-4 °C cold room for 24 h for precipitation and removing rough residue [B]. Pectolytic enzyme application (Shazym Claro Pectolytic Enzyme, 10.500 PGNU/g polygalacturonase, 0.15 g/L) was performed at 50°C for 2 h [C]. Bentonite and gelatin were applied during clarification process; 10 ml/L from 10% bentonite solution and 25 ml/L from 1% gelatin solution were used at 20 ºC and then 'koruk' juice was kept at 4 °C for 24 h [D]. At the same temperature, 5 g/L potassium bitartrate (KC 4 H 5 O 6 ) was added and left for 7 days for detartarization [E]. The final clarified 'koruk' juice was concentrated to 42-45 ºBrix at 50 ºC and 600 mm Hg vacuum [F]. Flow diagram of sour grape concentrate production is presented in Figure1. The concentration processing was made in duplicate and 150 kg of sour grape was used for each replicate.

Determination of total polyphenols
Total phenolic compounds in the samples were determined according to Folin-Ciocalteu colorimetric method (Singleton and Rossi, 1965). Briefly, 100 µL of Folin-Ciocalteu solution was added to each 4 mL diluted samples and then 500 µL of 20% saturated sodium carbonate (Na 2 CO 3 ) was added to final solution after 3 min and all was shaken. Then the samples were incubated at room temperature (24 ± 1 °C) for 30 min. At the end, 350 µL samples were transferred into a 96 well microplate and the absorbance was measured at 760 nm. 5, 10, 20, 30, 40 and 50 mg/L of standard concentrations were used for calibration curve (y=0.0649 x + 0.0324; R 2 =0.0995). Results were expressed as milligrams of gallic acid equivalents (GE) in 100 g dry matter (dry weight, DM).

Radical scavenging activity assay
2,2-Diphenyl-1-picrylhydrazyl (DPPH) assay was performed according to Brand-Williams et al. (1995). The principle of the method is the measurement of the reduction ability of the DPPH• radical in samples. Briefly, 3 mL of the 1 mM DPPH• solution was transferred into 10 mL tubes and 200, 400, 600, 800 and 1000 µL of diluted samples were added and bring up to 4 mL with methanol and incubated at room temperature (24 ± 1 °C) in the dark during 30 min. Methanol was used as blank. The absorbance was measured at 517 nm wavelength in a spectrophotometer (Thermo scientific, Multiskango, Finland). Percent inhibition values were calculated according to blank absorbance as described in the formula: Inhibition% = ((A DPPH -A SAMPLE )/ A DPPH ) x 100. Calculated inhibitions and the sample volumes were subjected to linear regression on the graphic, and slope of each sample and equilibrium of these slopes were obtained. EC 50 values were calculated based on the equation of obtained slope values (necessary volume of equate for elimination the 50% of DPPH•): EC 50 = [(a × sample volume) ± b]/dilution factor. A calibration curve was drawn using a standard solution of Trolox (8, 16, 32, 64 and 128 µM; y=0.2741x + 0.4972; R 2 =0.9987). Antioxidant activity was expressed as µmol Trolox in 100 g DM.

ABTS
[2,2′-azinobis-(3-ethylbenzothiazoline-6sulfonic acid)] method was used as previously described by Re et al. (1999). Firstly, stable ABTS stock solution was performed by reacting 7 mM ABTS with 2.45 mM potassium persulfate (final concentration) allowing the mixture to stand in the dark for 12-16 h at room temperature before use. The ABTS •+ solution was diluted with methanol to the absorbance of 0.70 (± 0.02) at 734 nm and equilibrated at 30°C. 10, 20 and 30 µL of each diluted samples were pipetted and ABTS •+ radical solution were added to final volume of 1000 µL. Then, absorbance was measured during 6 min with one min intervals. The initial (A 0 ) and the end of the 6 th min (A 6 ) absorbance were used for calculation of the absorbance inhibition percentages of each sample or standard volume. 5, 10, 15 and 20 µM standard Trolox concentration was utilized for calibration curve (y=3.2239x -0.0101; R 2 =0.9991). Trolox equivalent antioxidant capacity (TEAC) of the samples was calculated using slope of the sample and standard. Results were expressed as µmol Trolox in 100 g DM.

HPLC analysis for individual phenolic compounds
Phenolic compounds were evaluated by high performance liquid chromatography (HPLC) method. ODS C18 (250 x 4.6 mm, 5µm) column for analytical separation and diode array detector (DAD) was used in the HPLC system (Agilent 1260 infinity). Detection and quantification of phenolic compounds was carried out according to Ozkan and Gokturk Baydar (2006) and Caponio et al. (1999) with slight modification of mobile phase. 2.5, 5, 10, 20 and 40 mg/L standard concentrations were used for calibration curves of gallic acid, vanillic acid, caftaric acid, caffeic acid, p-coumaric acid, ferulic acid, sinapic acid, quercetin and (-)-epicatechin. 5, 10, 20, 40 and 60 mg/L concentrations were used for (+)catechin, (-)-epigallocatechin and (-)-epigallocatechin gallate calibration curves. The sour grape juice samples taken from the previously defined steps were diluted with methanol and filtered through a 0.45 µm of PTFE syringe filter before the HPLC analysis. Then, the samples were injected directly into the HPLC system. Gradient elution was used; solvents consisting of ultra-pure water: formic acid (99.8:0.2 v/v) (A) and methanol (B). Gradient elution program was as follows: the initial elution 0% B, followed 3 min by linear gradient from 0% to 5% B, 15 min linear gradient to 20% B, 2 min isocratic elution step %20 B, 10 min linear gradient elution to 25% B, 10 min elution to 30% B, 10 min elution to 40% B, 5 min elution to 50% B and 10 min linear gradient elution to 100% B. Then, 5 min 100% A elution was performed for returning to initial condition. Column temperature was set at 30°C and detection was made at 280, 320 and 360 nm. Gallic acid, (+)-catechin, (-)epigallocatechin, vanillic acid, (-)-epigallocatechin gallate and (-)-epicatechin were detected at 280 nm wavelength. Caftaric acid, caffeic acid, p-coumaric acid, ferulic acid and sinapic acid were identified at 320 nm. Finally, quercetin was detected at 360 nm wavelength. The elution time was 65 min, the injection volume was 10 µL and the flow rate was 1mL/min. The phenolic compounds of the samples were identified by comparing their retention times and spectra with those of analytical standards. The concentration of phenolic compounds in the samples was calculated through the calibration curves and expressed as mg/100 g DM. Chromatographic analyses were performed in triplicate.

Statistical analysis
A two-way analysis of variance (ANOVA) was applied to the obtained results. Duncan multiple comparison test was performed to determine the differences between the average values (p<0.05 significance level was used for comparisons). Pearson correlation coefficients were calculated to prove the relationships between total phenolics and antioxidant properties.

RESULTS AND DISCUSSION
The results for total phenolic (TP) contents of the samples are presented in Table I, where it can be observed that the values varied from 245.62 and 288.4 mg/100 g DM at processing steps. Significant statistical differences were found between the TP values of the investigated processing steps (p<0.05). The lowest TP content was found in [F] stage (after evaporation), while the highest TP was found in [A] stage. According to Piva et al. (2008), which investigated the physical and functional alterations during grape juice cooking using three different concentrate ratios (max. 35, 60 and 70%), the TP contents in fresh and concentrated grape must samples varied from 27.1 to 1259 mg/L. Öncül and Karabıyıklı (2015) reported that TP content of 'verjuice' samples varied from 233.44 and 672.75 mg/L. Nikfardjam (2008) reported similar TP content ranges in 'verjuice' (200-1330 mg/L). Turkmen et al. (2017) revealed that TP contents of unripe grape juice extracts ranged from 436.36 and 758.52 mg/L. Hayoglu et al. (2009) reported that TP content of 'verjuice' samples ranged from 2374.8 to 4041.5 mg/L. TP results of the sour grape concentrate are similar to those obtained in previous studies. However, the results are lower than the results reported by Hayoglu et al. (2009). The differences could be due to the maturation of grape, applied process steps and evaporation conditions (Gollücke et al., 2009;Hayoglu et al., 2009;Sabir et al., 2010;Öncül and Karabıyıklı, 2015;Tastan and Baysal, 2015).  On the other hand, Capanoglu et al. (2013) investigated the changes of polyphenols during production of grape juice concentrate and stated that TP content of grape concentrate decreased by 84.4% dry weight basis. The ratio is considerably higher than in the current study.
Antioxidant activities of samples were detected by DPPH and ABTS assays and the obtained results were showed in Table I  Alteração de TEAC durante o processamento do sumo de uvas não amadurecidas. Piva et al. (2008) reported that antioxidant activities of fresh and concentrated grape juice samples varied from 925 to 2100 µM/L. They concentrated the grape juice until 35, 60 and 70% ratios. The antioxidant activities in grape must concentrates decreased compared to initial fresh must values in dry weight bases. The reduction was 24.7 and 27.0% for 60 and 70% concentration ratios, respectively (Piva et al., 2008). It was reported that the antioxidant activity of the grape juice concentrate decreased between 83 and 92% during concentration stages compared to the initial stage. The reduction ratios from pasteurization to concentration stages were 3.6 and 10.4% for ABTS and DPPH methods, respectively. Moreover, it was expressed that antioxidant capacity showed some variation through the concentration process of grape juice (Gollucke et al., 2009). The current findings obtained from the sour grape concentrate have slightly differences with those of previous studies. These differences could be due to the maturity of the grapes because sour grapes are unripe grapes and they have different chemical and physical characteristics comparing to mature grapes. Öncül and Karabıyıklı (2015) have also reported the change of antioxidant components in grape depending on maturation. Additionally, grape varieties, heat treatment in processing and other processing conditions may affect these parameters.
Correlations between TP and antioxidant parameters in the sour grape concentration process are shown in Table II. Significant correlations were found among TP, inhibitions of DPPH and ABTS, EC 50 and TEAC ABTS and TEAC DPPH (p<0.01). The highest positive and the most significant correlation was observed between TP and TEAC ABTS (r = 0.999, p<0.01). The negative significant correlation was found between TP and EC 50 (r = -0.919, p<0.01). Additionally, important correlations were obtained between DPPH and ABTS parameters. Lima et al. (2014) found that the correlations between the total phenolic contents and the DPPH and ABTS antioxidant activities in grape juice samples were 0.94 and 0.84, respectively. In other study, Burin et al. (2010) reported the positive correlations found between total phenolic content and antioxidant activity (DPPH methods) of grape juice samples. Öncül and Karabıyıklı (2015) have also showed similar correlation results. The correlations between TP and antioxidant parameters are consistent with the literature. The values of the 12 phenolic compounds studied during sour grape processing stages are presented in Table III. The HPLC chromatogram obtained for the sour grape juice concentrate of the 12 investigated phenolic compounds is shown in Figure 3. Regarding individual phenolic compounds, caftaric acid was the major compound, with values varying from 12.40 to 37.60 mg/100 g in sour grape juice samples during the concentrate processing. In a study that have investigated phenolic compounds in 'verjuice' samples from different countries also mentioned the caftaric acid, whose ranged from 15.6 to 76.3 mg/L, as the main phenolic compound (Nikdardjam, 2008). Additionally, it was reported that caftaric acid was the major phenolic substance in the grape juice and wine (Toaldo et al., 2015;Yamamoto et al., 2015, Padilha et al., 2017Aleixandre-Tudo et al., 2018). The caftaric acid values in these studies varied from 73.4 to 365.5 mg/L in grape juices and 6.6 to 167.4 mg/L in wines. The findings related to caftaric acid in the current study are accordance with these literature's results.

CONCLUSIONS
Changes in the phenolic profile of the sour grape juice were evaluated during the concentrate processing stages. In addition, the antioxidant activities were analyzed with DPPH and ABTS methods. It was found that concentration process affects phenolic compounds, antioxidant capacity and correlation among these parameters. Especially, total phenolic content and antioxidant properties of the samples changed at the evaporation stage more than in the other processing stages. After evaporation under vacuum conditions, antioxidant activity of the samples decreased approximately between 14.2 and 17.0% for DPPH and ABTS methods, respectively. Besides, TP content of the samples also decreased approximately 14.8% according to [A] processing stage.