Chemical Properties, Fatty-Acid Composition, and Antioxidant Activity of Goji Berry (Lycium barbarum L. and Lycium chinense Mill.) Fruits

In this study, the content composition and antioxidant activity of goji berry fruits from two species (Lycium barbarum and Lycium chinense) were assessed. The total carbohydrate and phenolic contents were evaluated using attenuated total reflection Fourier-transform infrared (ATR-FT-IR) spectroscopy, while the antioxidant activity of fruits was examined with two in vitro methods, which are based on the scavenging activity of the 2,2-diphenyl-1-picrylhydrazyl (DPPH•) and 2,2’-azino-bis(3-ethyl-benzthiazoline-sulfonic acid) (ABTS•+) free radicals. The fatty-acid profile was determined using gas chromatography coupled with mass spectrometry (GC-MS). The results of this study indicate that the fruits of L. barbarum present higher concentrations in carbohydrates and phenolics than L. chinense Mill. fruits. Furthermore, the antioxidant activity based on the half maximal inhibitory concentration (IC50) measurements of DPPH• and ABTS•+ free-radical scavenging was higher in L. barbarum than L. chinense Mill. Also, the GCMS analysis confirms the high levels of linoleic, palmitic, and oleic acids contained in the fruits of both species. Finally, the results of this study clearly show that the concentration of bioactive and antioxidant molecules is higher in L. barbarum than in L. chinense fruits, which was also confirmed by ATR-FT-IR measurements.


Introduction
Plants and their fruits contain a wide variety of biological active secondary metabolites and, consequently, they are used for the development of drugs, dietary supplements, and functional foods, because of their antioxidant, anti-inflammatory, and antimicrobial properties [1,2]. Lycium barbarum is a perennial deciduous shrub with ellipsoid orange-red berries and a sweet-tangy flavor, also known as goji berry, wolfberry, barbary wolfberry, and Chinese boxthorn (or gouqizi in Chinese). The Lycium genus includes up to 70 species that vegetate in regions from the temperate to the subtropical regions of Eurasia, Australia, southern Africa, and North and South America [3]. Lycium species have a long history in China as they were used as medicine and functional food, and they are also referred to in the Traditional Chinese Pharmacopeia (TCP). Among the functional natural components, Goji berry fruits contain L. barbarum polysaccharides (LPB) that are water-soluble glycoconjugates, and they are the most well-researched components [4]. In addition to LBP, Goji berry fruits also contain carotenoids, (FAME) mix was also used. The FAME mix contains the methyl esters of 37 fatty acids (catalog no. 47885-U) in a concentration of about 2% and 4% of each fatty-acid methyl ester. Palmitic acid methyl ester and tridecanoid acid in FAME mix were also purchased from Sigma-Aldrich in 6% and ≥99% concentrations, respectively.

Determination of Total Carbohydrate Content (TCC) of the Extracts
The phenol/sulfuric method according to Dubois et al. [16] was used for the determination of TCC. Briefly, 1 mL of each sample solution was mixed with 0.5 mL of phenol (4%) and 2.5 mL of sulfuric acid (95%); after an incubation of 10 min, the optical density was measured at 490 nm using a Thermo Scientific (Evolution 201) spectrophotometer (Thermo Scientific, Madison, WI, USA). The TCC was calculated on the basis of the calibration curve of D-glucose and expressed as g of carbohydrates/L of the extract. The linearity range of standard D-glucose was determined as 0.01 to 0.1 mg/L (R 2 = 0.9918). The equation was obtained by linear fitting of y = 8.0357x + 0.0111 (data not shown), where y is the absorbance at 490 nm and x is the concentration of D-glucose. The TCC was expressed as mg of D-glucose equivalent per g of dried extract (mg/g dry weight).

Determination of Total Phenol Content (TPC) of the Extracts
Samples were aqueous ultrasound extracted according to the methodology of Skenderidis et al. [17]. The TPC of the extracts was determined using the Folin-Ciocalteu microscale protocol [18]. Briefly, 20 µL of goji berry extract was added to 1.58 mL of deionized water and 100 µL of Folin-Ciocalteu reagent. After the immediate addition of 300 µL of Na 2 CO 3 solution (20% w/v), the mixture was left for 120 min of incubation in the dark. Finally, the optical density was measured at 765 nm using the abovementioned spectrophotometer. TPC was expressed as mg of Gallic Acid Equivalent (GAE) per g of dried fruit weight (mg GAE/g dry weight) after determination using a standard curve of absorbance values derived from standard concentration solutions of GA. The DPPH radical-scavenging activity of the samples was evaluated according to the method described by Skenderidis et al. [17]. Dissolved in distilled water at different concentrations, goji berry samples were mixed with 1 mL of a freshly made methanol solution of DPPH• radical (100 µM). The contents were vigorously mixed and incubated at room temperature in the dark for 20 min and the absorbance was read at 517 nm. Methanol solutions of tested extracts and DPPH• were used as blank and control measurements, respectively. All experiments were carried out three times on two separate occasions. The percentage of radical-scavenging capacity (RSC) of the tested extracts was calculated according to the following equation: % DPPH• radical scavenging activity = ((Abs control − Abs sample )/Abs control ) × 100 where Abs control and Abs sample are the absorbance values of the control and the tested sample, respectively. The half maximal inhibitory concentration (IC 50 ) value was calculated using the graph-plotted RSC percentage against extract concentration in order to compare radical-scavenging efficiency of the extracts. All experiments were repeated three times on two separate occasions. 2.5.2. ABTS• + (2,2'-Azino-bis-(3-ethyl-benzthiazoline-sulfonic Acid) Radical-Scavenging Activity ABTS• + radical-scavenging activity was measured according to the method described by Kerasioti et al. [19]. The reaction was carried out in 1 mL of final volume, containing 400 µL of H 2 O, 500 µL of ABTS (1 mM), 50 µL of H 2 O 2 (30 µM), and 50 µL of 6 µM solution of the enzyme horseradish peroxidase (HRP). Contents immediately after enzyme addition were thoroughly mixed and incubated for 45 min in the dark at room temperature. After 45 min of incubation in the dark, different concentrations of the tested sample solution were added and mixed thoroughly; then, the absorbance was measured at 730 nm. In each experiment, blank (samples without HRP) and control samples (without goji berry extract) were used. All experiments were repeated three times and on at least two separate occasions. Furthermore, the RSC percentage of the ABTS• + radical-scavenging activity and IC 50 values were determined as described above.

Chemical Components and Structure Analysis by Infrared Spectroscopy
The FT-IR measurements were obtained at room temperature using a Nicolet 6700 (Thermo Scientific) spectrophotometer in order to estimate the chemical composition of the samples. The OMNIC 9 software (Thermo Scientific) was used for the configuration of the spectrophotometer equipped with an attenuated total reflectance (ATR) accessory, a deuterated triglycine sulfate (DTGS) detector (Thermo Scientific), and a KBr beam splitter (Thermo Scientific). A diamond crystal sampling plate Smart iTX, (Thermo Scientific) clamped with a pointed tip was used in order to place and analyze the samples. Background (empty sample plate) scans were acquired and rationed against the sample spectrum. Furthermore, the ATR crystal and pointed tip were cleaned in order to remove any interference from the preceding sample. Spectra were collected at 4 cm −1 spectral resolution in the mid-infrared range of 4000-650 cm −1 with 32 successive scans. Furthermore, all spectra were ATRand baseline-corrected, normalized, and transformed to absorbance spectra.

Fatty-Acid Methyl Ester (FAME) Synthesis
The method of FAME synthesis was applied as described by Gerasopoulos et al. [13]. Firstly, 0.5 mL of sample was placed in heat-resistant pyrex glass (16 cm length, 1.6 cm diameter) sealed with a teflon screw cap. As an internal standard, 1 mL of methanolic solution of tridecanoidacid (C13:0) was added at a concentration of 600 µg/mL. Subsequently, 0.4 mL of 10 N KOH (potassium hydroxide) and 2.7 mL of pure methanol were added. The examined solution was contained in tubes placed in a water bath at 55 • C for 90 min, while vigorous stirring every 20 min followed in order to achieve proper hydrolysis of the samples. Tap water was used in order to cool down the tubes for about 15 min. Fatty-acid methyl esters, 0.3 mL of 24 N H 2 SO 4 were added in order to correct the composition. Thereafter, tubes were placed again in a water bath at 55 • C for 90 min followed by vigorous stirring every 20 min. A cooling bath with tap water followed, as previously. Finally, 3 mL of hexane was added as a solvent, and the samples were stirred with a vortex for 3 min. The treated samples were then placed in the centrifuge at 6000× g for 15 min at room temperature, and the supernatant solution (hexane layer containing the FAME) was placed in GC vials of 2 mL. Finally, the supernatant solution was stored at −20 • C until GS/mass spectrometry (MS) analysis was carried out.

GC-MS Analysis
The fatty-acid profile assessment in goji berry samples was carried out in duplicate. A standard solution containing 37 FAMEs was used to identify the individual FA supplied by Supelco TM (Sigma-Aldrich, Munich, Germany) known as the 37 Component FAME Mix Standard. The peak area of the samples that was corrected with the respective correction factors was used for the calculation of the percentage of each FA [20]. The GC-MS Agilent 7890A GC chromatography apparatus (Agilent, Frankfurt, Germany) was used. The GC silica column (J&W 112-88A7: 804.11246 HP-88 250 • C: 100 m × 250 µm × 0.25 µm) (Agilent, Frankfurt, Germany) was located inside a temperature-controlled oven (Agilent, Frankfurt, Germany). The following MS parameters were used: interface temperature, 250 • C; MS ionization mode; helium gas with 45.2 mL/min flow; electron ionization; detector voltage acquisition mass range, 20-850 amu. Data were collected and processed using the GCMS Agilent integrated software (Agilent).

Statistical Analysis
The standard deviation was calculated and the averaged values along with the standard deviations (SD) are documented in the respective tables or figures. Statistical differences among the means, as well as interactions between the variables used in chemical analyses, were detected by one-way Analysis Of Variance (ANOVA) followed by Tukey's test, and the statistical significance was set at p ≤ 0.05. MiniTab ® 17.1.0 software (Minitab LCC, State College, PA, USA) was used as the tool to perform the abovementioned tests.

Results and Discussion
Since L. barbarum fruits of different areas showed different contents of the main active components, their main medicinal and functional use should be consequently different [21].
The current study was conducted to investigate the chemical profiles of two varieties of the genus Lycium cultivated in central Greece by measuring the TCC and TPC, fatty-acid profile, and the antioxidant activity based on the ability to scavenge the free radicals of DPPH• and ABTS• + .

Total Carbohydrate and Phenol Content
The TCC and TPC of the samples are presented in Table 1. According to the results, L. chinensis sample had lower TCC than L. barbarum, with values of 395 ± 4.1, 440 ± 5.2, and 329 ± 2.7, and 452 ± 3.8, 490 ± 6.8, and 370 ± 4.3 (mgGlu/g dry fruit), respectively, for each specie per collection month. The results for the imported fruits from China and Mongolia were 459 ± 3.8 and 434 ± 4.3, respectively. This is in line with the literature confirming that L. barbarum fruits have higher TCC [22].
The comparison of all samples shows that L. barbarum fruits that were collected in August show higher TCC, while the Mongolian sample had the lowest TCC. The highest rate of TCC of the Thessaly fruits confirms a previous study of our laboratory in which a significantly higher percentage of polysaccharides was present in L. barbarum [14].
Concerning TPC, L. barbarum the results were 9.7 ± 0.2, 10.1 ± 0.3 and 6.9 ± 0.3 compared to L. chinense with 8.5 ± 0.4, 8.9 ± 0.7, and 7.4 ± 0.6 mg GAE/g. Furthermore, the values of the imported from China fruits was 9.9 ± 0.6, while, for the Mongolian fruit, the value was 10.9 ± 0.3 mg GAE/g. The values of total polyphenol content obtained in the present study was higher than the values of 2.59-4.14 mg GAE/g presented by Ionică et al. [23], lower than the values of 106.80 ± 0.46 mg GAE/g reported from Kosar et al. [24], but well correlated with the values of 8.95-10.36 mg GAE/g for goji berries that were reported in the study of Medina [25].

Antioxidant Activity Based on DPPH• and ABTS• +
DPPH• and ABTS• + are assays that measure the sample's ability in deactivating radical species through electron transfer reactions [13,26,27]. The free-radical-scavenging activities determined by DPPH• and ABTS• + are expressed as the IC 50 value which is the effective concentration of the samples required to inhibit 50% of the initial free radical. Results are shown in Table 1.
The DPPH• free-radical-scavenging capacities ranged from 784 to 1254 µg/mL, while the ABTS• + values ranged from 192 to 407 µg/mL. As observed from the results, both Lycium species displayed good antioxidant activity. The best IC 50 values of L. chinense and L. barbarum were achieved in August and, more specifically, they were 950 ± 4.7 and 830 ± 5.4 µg/mL for the DPPH•, while, for ABTS• + , the values were 220 ± 6.1 and 195 ± 3.5, respectively. The fruits imported from Mongolia showed a lower statistical IC 50 value for DPPH•, while, for ABTS• + , they also showed a lower but not significant level when compared with Thessally L. barbarum fruits. The reported IC 50 of DPPH radical-scavenging activity values are lower compared to the 42.76 ± 0.25 mg/mL reported by Benchennouf et al. [8] for their water fraction extract.
Data presented in Table 1 showed that the antioxidant capacity of dehydrated L. barbarum fruits was higher compared to that of the L. chinense fruits cultivated in Greece, and the values are well correlated with the results obtained for the total phenolic content. The Mongolian sample showed the highest antioxidant activity of all samples investigated in this work against DPPH scavenging.
The reported values and this correlation are in accordance with previous studies [23,28,29].

Goji Berry Fatty-Acid Profile
The fatty-acid composition of the goji berry samples is presented in Table 2 [30].
The content of monounsaturated fatty acids (MUFA) ranged from 18.61% of LbC 1 to 26.19% of the Chinese origin sample . Oleic acid was found to be the major MUFA with a percentage ranging from 16.71% of LbC 1 to 22.39% of LcC 3 . Similar resultswere also reported by Cossignani et al. [12], who analyzed Italian, Chinese, and Mongolian goji berry fruits, and by Endes et al. [10] for Turkish origin goji berry fruits.
Polyunsaturated fatty acids (PUFA) were the most represented FA ranging from 43.23% of the LcC 3 sample to 51.11% of LbC 3 , and these results are in line with those reported by Blasi et al. [30]. The α-linolenic acid (n−3) PUFA ranged from 5.39% of the Mongolia origin L. barbarum sample to 8.85% of LbC 2 .
The ratios of PUFA/SFA and n−6/n−3 are significant for human health status and, as recommended by the World Health Organization [31], the PUFA/SFA ratio should be above 0.4-0.5. In this work, these ratios for all samples studied were found to be higher than 0.5, since their values ranged from 1.38 of the LcC 3 sample to 1.95 of LbC 3 . It was suggested that the balance between the intake of n-6 and n-3 PUFA is more important than the levels of intake of individual FAs, with regard to many metabolic functions in the human body [32]. In our study, the ratio of n−6/n−3 ranged from 4.48 in LbC2 to 7.82 in the Mongolia origin L. barbarum sample.
The current study is the first attempt to evaluate the fatty-acid composition contained in Greek goji berries. Similar studies were completed by Jabbar et al. [33] in Ningxia goji berry fruits and pollen, and by Blasi et al. [30], who studied 16 commercial samples of fruits in Italy. The results of both studies are consistent with the results presented in this work.   The reported values are mol.% mean values ± SD (n = 3), calculated using peak area values corrected with theoretical response factors.

ATR-FT-IR Spectroscopy
The fact that bioactive compounds present different ATR-FT-IR spectra denotes that infrared spectroscopy as an identification technique is of great importance for qualitative analysis. Functional groups and the structure characterization of all infrared peaks appearing in the spectra are presented in Table 2.
ATR-FT-IR spectra of LbC 2 and LcC 2 samples are presented in Figure 1. It can be said that there are no differences between the two fruits since both spectra are similar. The intensities of the peaks at 1620 and 1732 cm −1 in the carbonyl region are correlated with the antioxidant activity (unpublished data), and this is in agreement with the literature data, where the powerful antioxidant nature of the compounds is explained by the fact that they possess a great number of compounds like organic acids and carbohydrates [34][35][36].

ATR-FT-IR Spectroscopy
The fact that bioactive compounds present different ATR-FT-IR spectra denotes that infrared spectroscopy as an identification technique is of great importance for qualitative analysis. Functional groups and the structure characterization of all infrared peaks appearing in the spectra are presented in Table 2.
ATR-FT-IR spectra of LbC2 and LcC2 samples are presented in Figure 1. It can be said that there are no differences between the two fruits since both spectra are similar. The intensities of the peaks at ~1620 and 1732 cm −1 in the carbonyl region are correlated with the antioxidant activity (unpublished data), and this is in agreement with the literature data, where the powerful antioxidant nature of the compounds is explained by the fact that they possess a great number of compounds like organic acids and carbohydrates [34][35][36].
The identification of the major chemical groups of the examined compounds is usually based on the "fingerprint" region (950-1200 cm −1 ) of the spectra. The position (frequency) and intensity of the bands are characteristic of each compound, as presented in Table 3 [37]. The characteristic infrared band at ~898 cm −1 may be assigned to the β-anomeric configuration, while the bands at ~l028, 867, 818, and 777 cm −1 (Figure 1) can be attributed to the pyranose ring [38]. The higher intensities of the peaks that emerge in the spectrum of L. barbarum are probably related to the higher TCC and TPC content of this variety.   The identification of the major chemical groups of the examined compounds is usually based on the "fingerprint" region (950-1200 cm −1 ) of the spectra. The position (frequency) and intensity of the bands are characteristic of each compound, as presented in Table 3 [37]. The characteristic infrared band at~898 cm −1 may be assigned to the β-anomeric configuration, while the bands at~l028, 867, 818, and 777 cm −1 (Figure 1) can be attributed to the pyranose ring [38]. The higher intensities of the peaks that emerge in the spectrum of L. barbarum are probably related to the higher TCC and TPC content of this variety. Table 3. Peak analysis of the attenuated total reflection Fourier-transform infrared (ATR-FT-IR) spectra of the functional groups of L. barbarum L. and L. chinense Mill. fruits cultivated in central Greece.

Conclusions
In the present study the composition of the bioactive substances of two Lycium species cultivated in Greece in different fruit periods of production were evaluated and compared with two samples which were imported from China and Mongolia. The results show that the Greek L. barbarum fruits from the Thessaly region contain higher amounts TCC and TPC than L. chinense. Among all samples, the Mongolian fruits showed better results against DPPH radical scavenging. Moreover, a significant variation in the fruit composition between the different cultivated periods emerged from the analysis results. The cultivated fruits in August showed higher amounts of the examined bioactive substances, which leads to the conclusion that this is probably due to the heat stress response of the plants because of the high temperatures during this particular month in the Thessaly area. Furthermore, the GC-MS analysis results show high levels of linoleic, palmitic, and oleic acids for the fruits of both species, which is in agreement with the results of other studies that appear in the literature. The analysis of the ATR-FT-IR spectra of the two varieties shows no differences between the two Lycium species. However, the intensity of the infrared peaks, which are related to the content of functional groups in the fruits, verifies the results obtained for the TP and TC contents. Nevertheless, for a better understanding of their phytochemicals and biological composition, as well as their effects, more investigations are needed.