A Comparative Investigation on Phenolic Composition, Characterization and Antioxidant Potentials of Five Different Australian Grown Pear Varieties

Pear (Pyrus communis L.) is widely spread throughout the temperate regions of the world, such as China, America and Australia. This fruit is popular among consumers due to its excellent taste and perceived health benefits. Various bioactive compounds, which contribute to these health benefits, have been detected in the pear fruits, including a range of phenolic compounds. Five Australian grown pear varieties, which include Packham’s Triumph, Josephine de Malines, Beurre Bosc, Winter Nelis and Rico were selected for this study to examine the phenolic compounds in pears. Beurre Bosc exhibited the highest total polyphenol content (TPC) (3.14 ± 0.02 mg GAE/g), total tannin content (TTC) (1.43 ± 0.04 mg CE/g) and 2,2′-diphenyl-1-picrylhydrazyl (DPPH) (5.72 ± 0.11 mg AAE/g), while the Josephine de Malines variety was high in total flavonoid content (TFC) (1.53 ± 0.09 mg QE/g), ferric reducing antioxidant power (FRAP) (4.37 ± 0.04 mg AAE/g), 2,2′-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) (4.44 ± 0.01 mg AAE/g) and total antioxidant capacity (TAC) (5.29 ± 0.09 mg AAE/g). The liquid chromatography coupled with electrospray-ionization quadrupole time-of-flight mass spectrometry (LC-ESI-QTOF-MS/MS) data indicate that a total of 73 phenolic compounds were detected in Beurre Bosc (37 compounds), Josephine de Malines (34), Rico (22), Packham’s Triumph (15) and Winter Nelis (9), respectively. From HPLC-PDA quantification, the Beurre Bosc pear variety showed significantly higher in phenolic acids (chlorogenic acid; 17.58 ± 0.88 mg/g) and while flavonoids were significantly higher in Josephine de Malines (catechin; 17.45 ± 1.39 mg/g), as compared to other pear varieties. The analyses suggest that the Australian grown pears might contain an ideal source of phenolic compounds which benefit human health. The information provided by the present work can serve as practical supporting data for the use of pears in the nutraceutical, pharmaceutical and food industries.


Introduction
Pears (Pyrus communis L.) are one of the most common fruits in people's daily life. Pears are not only delicious and cheap, but they are also rich in phytochemicals [1]. Pears are grown in the temperate regions of the world which includes more than 50 countries. In 2019, the global production of pears reached 23.1 million tons [2] with around 10% of pears production being processed into various products, such as canned pears, concentrated pear juice and fresh cut pears [3]. Pears are rich sources of bioactive compounds such as phytochemicals, soluble sugars, amino acids, vitamins and minerals [4]. Bioactive compounds extracted from different pears are reported as being protective against various human disorders including ageing, cancer, cardiovascular disease, nerve dysfunction,

Sample Preparation
The fresh fruits of five Australian grown pear varieties, Rico, Packham's Triumph, Beurre Bosc, Winter Nelis and Josephine de Malines ( Figure 1) grown in different region of Victoria were purchased from local markets in Melbourne, Victoria, Australia. A fully mature pear samples were harvested, stored at room temperature for 24-48 h for optimum ripening followed by transportation and distribution to the local retailers within 2-3 days using refrigerated trucks. Two to three kilogram samples of each variety were cleaned and peeled; the seeds were removed and pulp was blended using the 1.5 L blender (Russell Hobbs Classic, model DZ-1613, Melbourne, VIC, Australia). The pear pulps were kept at −20 • C for 48 h and lyophilized at −45 • C/50 MPa by Dynavac engineering FD3 Freeze Drier (W.A., Australia) and Edwards RV12 oil sealed rotary vane pump (Bolton, UK) and the freeze dried powders were stored at −20 • C.

Extraction of Phenolic Compounds
The extracts were made by modifying the protocol of Peng et al. [10], 5 g of pear powder was mixed with 20 mL 70% ethanol and homogenized with the IKA Ultra-Turrax ® T25 homogenizer (Rawang, Selangor, Malaysia) and subjected to shaking incubator (ZWYR-240, Labwit, Ashwood, VIC, Australia) at 120 rpm for 12 h (4 • C). After incubation, the pear extracts were centrifuged with Hettich Refrigerated Centrifuge (ROTINA380R, Tuttlingen, Baden-Württemberg, Germany) at 1000 rpm for 15 min. The supernatants were collected and stored at −20 • C before further analysis.

Estimation of Phenolic Compounds and Antioxidant Assay
The TPC, TFC and TTC assays were conducted to estimate the polyphenols in the samples. For antioxidant activities, three different antioxidant assays, including DPPH, ABTS and FRAP assay were used, which are based on the method and parameter from Tang et al. [11]. The spectrophotometric data were collected on the Multiskan ® Go microplate photometer (Thermo Fisher Scientific, Waltham, MA, USA).

Determination of Total Phenolic Content (TPC)
The TPC of pear samples was determined by the spectrophotometric method of Samsonowicz et al. [12] with some modification. The sample extracts (25 µL), Folin-Ciocalteu reagent solution (25 µL, 1:3 diluted with water) and Milli-Q water (200 µL) were added in a 96-well plate (Costar, Corning, NY, USA). After incubation (25 • C, 5 min), 25 µL 10% (w:w) sodium carbonate was added and followed by incubation in the dark for 60 min. The absorbance was determined at 764 nm in a microplate reader (Thermo Fisher Scientific, Waltham, MA, USA). The quantification of each sample was based on the standard curve that was generated with 0-200 µg/mL gallic acid in ethanol. The result was expressed as mg of gallic acid equivalents per gram dry weight of sample (mg GAE/g of DW).

Determination of Total Flavonoid Content (TFC)
The TFC of the pear sample was measured by modifying the aluminum chloride method of Stavrou et al. [13]. The sample extracts (80 µL), 2% aluminum chloride (80 µL, w/v, diluted with ethanol) and sodium acetate solution (120 µL, 50 g/L) were mixed in a 96-well plate and then incubated in the dark at room temperature for 60 min. The absorbance was measured at 440 nm in a microplate reader. The calculation of TFC of each sample was based on the standard curve of quercetin (0 to 50 µg/mL) and the result was expressed as mg of quercetin equivalent per g (mg QE/g DW) of dry weight.

Determination of Total Tannin Content (TTC)
The TTC of the pear samples was estimated by the method of Stavrou et al. [13] with some modification. The sample extracts (25 µL, 1:50 diluted with methanol), 4% vanillin solution (150 µL, diluted with methanol) and 32% sulfuric acid (25 µL) were mixed in a 96-well plate and followed by incubation at 25 • C for 15 min. The absorbance was measured at 500 nm against a blank in a microplate reader. The calculation of results was based on the standard curve of catechin solution (concentration from 0-1000 µg/mL). The result was expressed as mg of catechin equivalent per g (mg CE/g DW) of dry weight.

2,2 -Diphenyl-1-picrylhydrazyl (DPPH) Assay
The radical scavenging activity of pears was determined by DPPH assay method with some modification of Sogi et al. [14]. Sample extract (40 µL) was mixed with 0.1 M DPPH radical methanol solution (260 µL) in a 96-well plate, then incubated at 25 • C for 30 min. The absorbance was measured at 517 nm in a microplate reader. The standard curve was generated with different concentration of ascorbic acid (0 to 50 µg/mL). The result was expressed as mg ascorbic acid equivalents per g of dry weights (mg AAE/g DW).

Ferric Reducing Antioxidant Power (FRAP) Assay
The reducing capacity of the pear samples was determined based on the method of Sogi et al. [14] with some modification. This method involved the determination of the ability to reduce Fe 3+ in the Fe 3+ -TPTZ complex (ferric-2,4,6-tripyridyl-s-Triazine) into Fe 2+ -TPTZ. Sodium acetate solution (300 mM), TPTZ solution (10 mM) and ferric chloride (20 mM) was mixed in a ratio of 10:1:1 (v/v/v) to freshly made FRAP reagent. Then, sample extract (20 µL) and prepared FRAP dye solution (280 µL) were added to a 96-well plate and followed by incubation at 37 • C for 10 min. The absorbance was measured at 593 nm in microplate reader. The standard curve was generated with ascorbic acids range from 0 to 50 µg/mL. The results were expressed as mg of ascorbic acid equivalents per g (mg AAE/g DW) of dry weight.

2,2 -Azino-bis-3-ethylbenzothiazoline-6-sulfonic Acid (ABTS) Assay
The free radical scavenging capacity of the pear samples was determined by ABTS + radical cation decolorization assay with some modification of Sogi et al. [14]. 7 mmol/L of ABTS solution (5 mL) and 140 mM potassium persulfate decolorization solution were mixed to make ABTS + stock solutions. The mixture was incubated in the dark for 16 h (room temperature) and then ABTS + stock solution was diluted with ethanol to obtain an initial absorbance of 0.700 at 734 nm. Afterwards, the pear sample extract (10 µL) was mixed with the prepared diluent in 96-well plate and followed by incubation in the dark at room temperature for 6 min. The absorbance was measured at 734 nm in microplate reader. The standard curve was generated with ascorbic acid with concentration ranging from 0 to 150 µg/mL. The ABTS value was expressed as mg of ascorbic acid equivalents per g (mg AAE/g DW) of dry weight.

Total Antioxidant Capacity (TAC)
The TAC was determined by following the modified protocol of Subbiah et al. [15]. First, 0.6 M H 2 SO 4 , 0.028 M sodium phosphate and 0.004 M ammonium molybdate were mixed to make the phosphomolybdate reagent. Next, the pear sample extract (40 µL) and phosphomolybdate reagent (260 µL) were mixed in 96-well plate and followed by incubation at 95 • C for 10 min. The absorbance was determined at 695 nm. The standard curve was generated with ascorbic acid with concentration ranging from 0-200 µg/mL. The TAC values were expressed as mg of ascorbic acid equivalents per g (mg AAE/g DW) of dry weight.

Characterization of Phenolic Compounds by LC-ESI-QTOF-MS/MS Analysis
Characterization of phenolic compounds in pears was carried out using the method of Suleria et al. [16]. An Agilent 1200 series HPLC (Agilent Technologies, Santa Clara, CA, USA) equipped with Agilent 6520 I Accurate-Mass Q-TOF LC-MS/MS (Agilent Technologies, Santa Clara, CA, USA) was used in the characterization and identification of phenolic compounds in pears. The separation of different phenolic compounds was conducted in Synergi Hydro-RP 80Å LC reverse phase with 250 mm × 4.6 mm internal diameter and 4 µm particle size (Phenomenex, Torrance, CA, USA). The mobile phase A, which consists of water/acetic acid solution (98:2, v/v), was combined with mobile phase B, which consists of acetonitrile/water/acetic acid solution (50:49.5:0.5, v/v/v) to form the binary solvent system at the flow rate of 0.8 mL/min with a sample injection volume 6 µL. The gradient elution lasted for 85 min with the following conditions: 0 min, 90% A and 10% B; 20 min, 75% A and 25% B; 30 min, 65% A and 35% B; 40 min, 60% A and 40% B; 70 min, 45% A and 55% B; 75 min, 20% A and 80% B; 77-79 min, 0% A and 100% B; 82-85 min, 90% A and 10% B. Both the positive and negative ion mode were used in peak identification with capillary at 3.5 LV and nozzle voltage at 500 V. The parameters of mass spectrometry were set as follows: for nitrogen gas, the pressure was set at 45 psi with the flow rate of 5 L/min at 300 • C; for sheath gas, the flow rate and temperature were set at 11 L/min and 250 • C. The range of mass spectra was set at m/z 50-1300. Further, MS/MS analyses were carried out in automatic mode with collision energy (10, 15 and 30 eV) for fragmentation. Data collection and processing was conducted using MassHunter (Qualitative Analysis, version B.03.01, Agilent).

HPLC-PDA Analysis
The quantification of targeted phenolic compounds in pears was measured by the method of Zhong et al. [17] and performed by Agilent 1200 series HPLC (Agilent Technologies, Santa Clara, CA, USA) equipped with a photodiode array (PDA) detector. The column and conditions in HPLC-PDA analysis were similar to LC-MS/MS analysis, while the sample injection volume was changed to 20 µL. The phenolic compounds were determined at three different wavelengths, including 280 nm, 320 nm and 370 nm. The quantification of the concentration of individual polyphenols was based on the calibration standard curve and the result was expressed as mg/g of sample. Data collection and processing was performed using Agilent LC-ESI-QTOF-MS/MS MassHunter (Qualitative Analysis, version B.03.01, Agilent).

Statistical Analysis
Three parallel experiments were conducted in each test and the data are expressed as mean ± standard deviation. Statistical analysis was performed by using Minitab ® 18 Statistical software (Minitab Inc., State College, PA, USA). One-way analysis of variance (ANOVA) followed by Tukey's honestly significant differences (HSD) multiple rank test was carried out to test whether there was a significant difference between the antioxidant activities and polyphenol content of each sample at p < 0.05.

Phenolic Compound Estimation (TPC, TFC and TTC)
Previously, several studies have estimated the phenolic content of different pear varieties by determining different parameters. In this study, the phenolic content in pears were determined by the TPC, TFC and TTC and the results were expressed as mg of gallic acid equivalents per gram dry weight of sample (mg GAE/g of DW) mentioned in Table 1. The TPC of Beurre Bosc pulp (3.14 ± 0.02 mg GAE/g) was higher than Josephine de Malines, Packham's Triumph, Winter Nelis and Rico. Our results were slightly higher than Turkish grown pear varieties including Egirsah, Gugum, Deveci, Kizil and Banda, also extracted in methanol with different concentration (1.75 ± 0.13 mg GAE/g) [18]. The difference in the sample variety, growing region, sample extraction techniques, type of solvent, solute to solvent ratio may contribute to the difference in results. However, Manzoor et al. [19] reported the total phenolic content of Nakh and Naspati pear were in range of our study.
Regarding the TFC, the Josephine de Malines (1.53 ± 0.09 mg QE/g DW) had higher flavonoid content than Beurre Bosc, Packham's Triumph, Winter Nelis and Rico. The results were similar to the TFC values (ranging from 0.3 to 6 mg QE/g) reported by Azzini et al. [20] and Li et al. [21]. Rawat et al. [22] also determined the higher total flavonoid content in one of the southern Asia grown pear varieties (Pyrus pashia) by aluminum chloride colorimetric assay and extracted with different solvent and solvent to solute ratio. Previously, Patricia et al. [23] also confirmed that the total flavonoid content of pears can be varied significantly using different extraction solvents including n-hexane, ethyl acetate, ethanol and methanol.
The Beurre Bosc pear has the highest TTC value (1.43 ± 0.04 mg CE/g DW) as compared to other pear varieties (Table 1). Previously, only a few studies have focused on the total tannin content of pears. Ma et al. [24] have detected slightly higher total tannin content in one of the apple-shaped pear varieties (Pyrus pyrifolia) using another method of n-BuOH-HCl-Fe-III. In addition, Velmurugan and Bhargava [25] had also reported difference in total tannin content of European grown pear varieties extracted with different solvents (chloroform, aqueous, ethyl acetate and ethanol) and using another method of Folin-Denis. However, due to the difference in the detection methods and calibrators, it is difficult to be compare these findings with the TTC value in our experiment. This is why we have expanded our suite of antioxidant assays to better characterize the antioxidant potential of Australian grown pear samples.

Antioxidant Activity (DPPH, FRAP, ABTS and TAC)
The antioxidant activities of pears were determined by DPPH, FRAP, ABTS and TAC assays. In current research, different types of antioxidant assays were performed, involved different mechanisms, to understand overall and true antioxidant potential of pear fruit. The results of DPPH test ranged from 3.25 ± 0.03 to 5.72 ± 0.1 mg AAE/g, with statistically significant differences between pear varieties except for Josephine de Malines and Rico ( Table 1). The Beurre Bosc had the highest DPPH free radical scavenging capacity followed by Josephine de Malines, Rico, Packham's Triumph and Winter Nelis. In the previous study, Nomura et al. [26] measured the DPPH radical scavenging activity of 29 varieties of Japanese and European grown pears, some of their pear varieties had the similar DPPH free radical scavenging activity with our Australian grown pear varieties. In addition, Galvis Sánchez et al. [27] determined the DPPH free radical scavenging capacity of six Chile grown European pear varieties while their DPPH values were slightly lower than our study. The difference in pear varieties, extraction solvent, growing region, condition and harvesting season may contribute to the difference of results.
For the FRAP assay, all the varieties show significant differences from each other. Josephine de Malines had the highest FRAP activity which is 4.37 ± 0.04 mg AAE/g, followed by Beurre Bosc, Packham's Triumph, Rico and Winter Nelis. Previously, the total antioxidant activity of two Greece grown pears varieties (Naoussa and Vergina) measured by FRAP method ranged from 1.41 mg AAE/g to 1.93 mg AAE/g which is slightly lower than our FRAP values [28,29]. Gu et al. [30] also reported the slightly lower antioxidant capacity of European pears extracted with different solvent to solute ratio. However, Jamuna et al. [31] detected the antioxidant activity of an Indian grown pear variety (Pyrus communis) by FRAP assay and the result (3.00 mg AAE/g) was consistent with our Australian grown pear varieties.
Based on the ABTS assay, Josephine de Malines and Beurre Bosc were not different, but they were significantly higher than the other varieties. Thus, Josephine de Malines and Beurre Bosc have the highest ABTS value (4.44 ± 0.01 and 4.41 ± 0.07 mg AAE/g, respectively) followed by Packham's Triumph, Rico and Winter Nelis (Table 1). In the previous study, Batista et al. [32] reported that the ABTS values of different Portugal grown pear varieties including S. Bartolomeu and Amêndoa pears and the results were slightly lower than our investigation. Erbil et al. [18] also measured the ABTS free radical scavenging activity of five different Turkish grown pear varieties (Egirsah, Gugum, Deveci, Kizil and Banda) and found that there was a significant difference of the ABTS free radical scavenging capacity between different pear varieties grown in the same region.
When antioxidant activity was measured using the TAC assay there were significant differences between all varieties except Beurre Bosc and Winter Nelis. Josephine de Malines had the higher TAC value (5.29 ± 0.09 mg AAE/g), followed by Packham's Triumph, Winter Nelis, Beurre Bosc and Rico. Previously, Jamuna et al. [33] reported the TAC capacity (1.45 mg AAE/g) of Indian grown pear variety (Pyrus communis) extracted using chloroform-water which was lower than our values. The difference in pear varieties, growing region, extraction solvent, solute to solvent ratio and harvesting season may contribute to the difference of results.
In the present study, Beurre Bosc had the highest antioxidant capacity when measured using TPC, TTC, DPPH, FRAP and ABTS assays while Winter Nelis was consistently low TAC for all assays. Thus, the antioxidant capacity of pears can be associated with their polyphenol content. Different varieties, genotypes and agronomy of pears influence the polyphenol content and thereby influence their antioxidant activity.

Correlation among Different Antioxidant Variables
The correlations between TPC, TFC, TTC and antioxidant assays (DPPH, FRAP, ABTS and TAC) were performed by Pearson's correlation test by Graphpad Prism 8 ( Table 2). There was a significant positive correlation between TPC and TTC (r = 0.93, p ≤ 0.05) and a highly significant positive correlation between TPC and DPPH value (r = 0.97, p ≤ 0.01).
This result is consistent with the research reported by Kolniak-Ostek [1]. The TPC estimate the phenolic content in samples and DPPH scavenging capacity measures the antioxidant activities, therefore the phenolic content in pears mainly contribute to the antioxidant activities of sample. Furthermore, TFC is strongly correlated with FRAP value with Pearson's correlation coefficient (r = 0.93, p ≤ 0.01). Previously, the strong positive correlation between TFC and FRAP value of pear was also observed by Azzini et al. [20] with r = 0.919. FRAP measures the ability to reduce the Fe 3+ -TPTZ complex into Fe 2+ -TPTZ, while TFC only determined the flavonoid content in samples. Thus, this correlation indicates that the antioxidant capacity is positively related to the presence of flavonoid compounds. Furthermore there was also a significant positive correlation between TFC and ABTS (r = 0.87, p ≤ 0.05) which was similar to the findings of Floegel et al. [34]. However, Li et al. [21] also observed a strong correlation between DPPH and TFC which was not significant in this study. This may due to the flavonoid compounds have no obvious effect on DPPH free radical scavenging ability.
Apart from above, TTC displays a highly significant positive correlation with DPPH assay with (r = 0.96, p ≤ 0.01) which was similar to the findings of Li et al. [21]. This strong correlation indicates that the tannin in pears were related to the DPPH free radical scavenging capacity. This may due to the molecular structure of tannin in pears can provide H to DPPH free radicals to form DPPH-H [21]. Additionally, the FRAP assay is highly correlated with ABTS assay in the present study (r = 0.94, p ≤ 0.01) which was consistent with the findings of Du et al. [35] who also observed a strong positive correlation (r = 0.92) between TAC measured using the FRAP and ABTS methods. The strong correlations between TAC measured using ABTS and FRAP is expected because both methods are relying on the single electron transfer principle. For further elucidation of the sources of the TAC activity, LC-ESI-QTOF-MS/MS and HPLC-PDA were used for identification, characterization and quantification of phenolic compounds in different pear samples.

LC-MS Analysis
In the present work, the quantitative analysis of the phenolic compounds in five  Figures S1 and S2). The data analysis was carried out by using Agilent LC/MS Mass Hunter Qualitative software and Personal Compound Database and Library (PCDL). Only the compounds with a PCDL score higher than 80 and the mass error less than ±5 ppm were selected for characterization and verification purposes. A total of 73 phenolic compounds were identified in different pear varieties, which include phenolic acids (30), flavonoids (29), stilbenes (1) and other polyphenols (13) mentioned in Table 3.

Phenolic Acids
In this study, 4 classes of phenolic acids were detected in five pear samples (Table 3) resulting in two dominant subclasses of phenolic acids in pear samples being hydroxybenzoic acids (eight phenolic acids) and hydroxycinnamic acids (17), respectively. In addition, two hydroxyphenylacetic acids and three hydroxyphenylpropanoic acids were also identified in pear samples.

Hydroxybenzoic Acids
Hydroxybenzoic acids are widely present in different fruits and they are not only capable of inhibit α-amylase and α-glucosidase activity, but also reduce the enzymes responsible for breaking down complex carbohydrates, which help to control glycemia in humans [36]. According to the LC-MS results, three out of the eight different hydroxybenzoic acids were identified in only one pear sample (Compounds 1, 3, 7).  [38,39].
In a previous study, the appearance of gallic acid in the peel and pulp of different pear varieties have been reported by Yim and Nam [4] and Li et al. [21]. Additionally, some hydroxybenzoic acids were characterized by several studies on different fruits or vegetables. For example, Rajauria et al. [38] identified the gallic acid 4-O-glucoside in Himanthalia elongata Irish seaweed by LC-DAD-ESI-MS/MS and using 60% methanol as the extracting agent. Catarino et al. [40] identified the protocatechuic-acid 4-O-glucoside in the stems of Eriocephalus africanus by LC-DAD-ESI/MS n method. 2-Hydroxybenzoic acid, which is also known as salicylic acid, was identified not only in pears, but also in other fruits, including kiwifruit and grapes [41]. In the previous study, vanillic acid 4-sulfate and 2, 3-dihydroxybenzoic acid were identified by Zhong et al. [17] and Pj et al. [42] in seaweed and Flacourtia inermis fruit, respectively.
Hydroxycinnamic Acids Only 6 out of 17 different hydroxycinnamic acids (Compounds 9, 10, 12, 15, 19, 20) were detected in more than one pear sample. Among them, compounds 10, 12, 19, 20 were identified in both positive and negative modes. Compound 22 was tentatively identified as ferulic acid according to the precursor ion [M − H] − at m/z 193.0499. In the MS 2 experiment of compound 22, the product ion at m/z 178, m/z 149 and m/z 134 were due to the loss of CH 3 , CO 2 and CH 3 with CO 2 from the precursor, respectively [43]. Caffeic acid (Compound 12) was observed in with [M − H] − m/z at 179.0346 in both the negative and positive ionization mode. The identification of caffeic acid was achieved by the MS 2 experiment which displayed the product ions at m/z 143 and m/z 133 result from the loss of 2 water (36 Da) and HCOOH (46 Da), respectively [44]. Compound 10, which was present in all pear varieties except Josephine de Malines, was tentatively detected as cinnamic acid based on the precursor ion [M − H] − m/z at 147.0461 in both the negative and positive ionization modes and further confirmed by the MS 2 experiment which shows a loss of CO 2 (44 Da) at m/z 103 [45].
Previously, the appearance of cinnamic acid in 10 different pear varieties had been reported by Sun et al. [46]. Furthermore, the presence of caffeic and m-coumaric acid in the flesh and peel of pear varieties were also determined by Öztürk et al. [47]. Simirgiotis et al. [48] identified caffeoyl glucose in the peel of the small Easter Pear. In an earlier study, ferulic acid was been identified in five different pear varieties [49,50]. Regarding other plants, Suleria et al. [16] characterized the feruloyl tartaric acid in pear peels and other fruit peels. In addition, Lin and Harnly [51] confirmed the appearance of the 3-caffeoylquinic acids in the pear skin. According to Ludwig et al. [52], the ferulic acid 4-O-glucuronide has been detected in the red raspberry by UHPLC-MS analysis. Piovesana and Noreña [53] isolated 3-p-coumaroylquinic acid from Hibiscus sabdariffa L. by HPLC-QTOF-MS method. In addition, coffee and stone fruit was also reported as a good source of 3-feruloylquinic acid by the previous study [54]. Regarding the rosmarinic acid, several studies have reported the presence of this compound. Hossain et al. [55] [57]. In a previous study, Cuadrado-Silva et al. [58] quantified only trace amounts of 3,4-dihydroxyphenylacetic acid in P. friedrichsthalianum fruit by the UHPLC-ESI/QqQ method. Trautvetter et al. [59] also identified this compound in five varieties of honey at m/z 167.0344 by the UPLC-MS method. However, to our best knowledge the present study is the first to report the presence of 3-hydroxy-3-(3-hydroxyphenyl) propionic acid in pears.

Flavonoids
Flavonoids are the second largest group of phenolic compounds in the five pear varieties. The subclass of flavonoids which were detected in pears include anthocyanins, dihydrochalcones, dihydroflavonols, flavanols, flavanones, flavones, flavonol and isoflavonoids. The flavanols and flavonols are the dominant subclasses among them with 7 of each group of compounds being detected.

Anthocyanins, Dihydrochalcones and Dihydroflavonols
Only 2 anthocyanins, 2 dihydrochalcones and 2 dihydroflavonols were detected in five pear varieties and both compounds 31 and 32 were detected in the positive ionization mode (Table 3) [61].
According to Raja et al. [62], dihydroquercetin is one of the highly specific polyphenols. They have identified this compound in two pear varieties, which are conference and Alejandrina. Previously, CHUNG [63] have identified petunidin 3-O-(6"-acetyl-glucoside) in Highbush blueberry fruit with the MS value m/z at 521 and MS 2 value m/z at 317, this result is same to our research. 3-Hydroxyphloretin 2'-O-glucoside in the peel of apples was also quantified in the previous study by HPLC analysis [64,65]. In addition, another dihydrochalcone, 3-hydroxyphloretin 2'-O-xylosyl-glucoside, was also identified in apple pomace by Alvarez Arraibi [66].

Flavanols
Only two (compounds 37 and 41) out of seven flavanols were detected in more than a single pear variety. Compound 41 was tentatively characterized as (+)-catechin in the negative ESImode at m/z 289.0712 and further confirmed by the MS 2 experiment which displayed a characteristic loss of CO 2 (44 Da), flavonoid a ring (84 Da) and flavonoid a ring (110 Da) at m/z 245, m/z 205 and m/z 179, respectively [37]. While compound 39 only presenting in Josephine de Malines in the negative ESImode was proposed as procyanidin dimer B1 based on the [M − H] − m/z at 577.1318. In the MS 2 spectra, procyanidin dimer B1 showed the fragments at m/z 451 due to a loss of phloroglucinol (126 Da) [67].
Previously, Brahem et al. [7] detected the phenolic compounds in 19 pear varieties and characterized (+)-catechin in the flesh of four pear varieties and the peel of 15 pear varieties. In addition, the appearance of procyanidin dimer B1 and procyanidin trimer C1 in pear varieties were identified by de Pascual-Teresa et al. [68]. In the previous study, Yuzuak et al. [69] have confirmed the presence of (+)-gallocatechin in berries of Two Muscadine grape hybrids by applying HPLC-QTOF-MS/MS method. But Arts et al. [70] indicate that there was no (+)-gallocatechin in the peel and pulp of pears. The difference may result from the different pear varieties and the variance of the extraction method. In addition, (+)-catechin 3-O-gallate and 3'-O-methylcatechin were the first time identified in European pears to our best knowledge.
Previously, Dairpoosh [65] have characterized myricetin 3-O-rhamnoside in pear peel. In addition, myricetin 3-O-galactoside were characterized from grape wine by LC-MS method [74]. Barbosa et al. [75] reported that myricetin 3-O-rutinoside in Chrysobalanus icaco L. was detected in the negative mode with the quasi-molecular ion [M − H] − at m/z 625 and the fragment ion at m/z 317 corresponding to the loss of rhamnose unit. They showed slightly different results from our work, which might be due to the difference in mass spectrophotometric methods and techniques applied.

Flavanones, Flavones and Isoflavonoids
In total, 2 flavanones, 4 flavones and 3 isoflavonoids were detected in pear samples with five out of nine being found in only one pear variety (Table 3). There have been very few studies focusing on the isoflavonoids in pears, although the isoflavonoid content of other fruits have been reported. Peng et al. [10] characterized 3'-hydroxydaidzein with [M + H] + at m/z 271 in mango peel while 2-dehydro-Odesmethylangolensin has been identified in palm fruits and reported by Ma et al. [77]. Although no research has confirmed the presence of rhoifolin in pears, the presence of these compound in Pummelo juice and Mandarin juice has been reported by Nogata et al. [78]. Additionally, 6-hydroxyluteolin 7-O-rhamnoside can be found in Mexican oregano [79].

Other Phenolic Compounds
Thirteen other polyphenols were also detected in the pear samples, which include 1 curcuminoids, 1 furanocoumarin, 2 hydroxybenzaldehydes, 2 hydroxybenzoketones, 1 hydroxycoumarins, 1 phenolic terpene, 2 tyrosols and 3 other phenolic compounds (Table 3). Only two of these compounds (Compounds 63 and 66), were observed in both negative and positive ionization modes. Compound 64 and was found in in all pear varieties except Winter Nelis and was identified as 4-hydroxybenzaldehyde having the precursor ion [M − H] − at m/z 121.0298 in the negative ionization mode and further confirmed by the MS 2 experiment which displayed a characteristic loss of CO 2 (44Da) [80]. In the MS 2 spectra of carnosic acid, the product ion at m/z 287 and m/z 269 were due to the loss CO 2 and further loss of H 2 O from the parent ion [82].
Arbutin is an important phenolic compound with antibiotic properties and has been confirmed as the dominant phenolic compounds in 8 varieties of pears by Öztürk et al. [47] and Li et al. [83]. Some of the other compounds were reported in other plants but not pears. Pistelli et al. [84] have identified the presence of demethoxycurcumin in Curcuma longa by LC-DAD-ESI-MS method. While the presence of isopimpinellin have also been confirmed by Dehghan et al. [85] in Heracleum persicum. In addition, Welke et al. [86] used comprehensive two-dimensional gas chromatography to identify p-anisaldehyde in the Chardonnay wine. In addition, both lithospermic acid salvianolic acid B were identified in Chinese Wild Salvia miltiorrhiza based on UPLC-QqQ-MS method by Zhang et al. [87]. In addition, although carnosic acid was not identified in pears to our best knowledge, but it has been reported widely distributed in the plant species of the Lamiaceae family [88].

HPLC Analysis
In the present work, the quantification of phenolic compounds was based on comparing retention time with HPLC grade reference standards. In total, 10 phenolic compounds were quantified through HPLC-PDA, including five targeted phenolic acids (gallic acid, protocatechuic acid, p-hydroxybenzoic acid, chlorogenic acid and caffeic acid) and five flavonoids (catechin, epicatechin, epicatechin gallate, quercetin and kaempferol) ( Table 4).
Beurre Bosc has the highest phenolic acid contents while Winter Nelis and Rico have lower phenolic acids contents (Table 4) which is consistent with the other estimations of TPC in the present study. Among the five targeted phenolic acids, protocatechuic acid was only detected in Beurre Bosc, Josephine de Malines and Packham's Triumph. In addition, a small amount of p-hydroxybenzoic acid was detected in five pear varieties, which ranged from 0.95 ± 0.05 mg/g to 3.14 ± 0.15 mg/g. They were abundant in Packham's Triumph and less abundant in Josephine de Malines and Rico, respectively. Chlorogenic acid is the dominant phenolic acids among the five selected phenolic acids. It is abundant in all five pear varieties except Rico. Previously, both Öztürk et al. [47] and Liaudanskas et al. [89] have reported that chlorogenic acid is the dominant phenolic compounds in several Lithuania and Sinop grown pear varieties, including Conference, Concordia, Grabova and Patten. Further, protocatechuic acid was also previously quantified by Truong et al. [90] through HPLC analysis in the Asian grown pear varieties (Pyrus spp.). In addition, Tanrıöven and Ekşi [91] have quantitated chlorogenic acid and caffeic acid in pear juice using HPLC method from several pear varieties, for instance, Williams, Santa Maria and Starkrimson. In the previous study, both Li et al. [21] and Li et al. [92] have quantitated the gallic acid in different pear varieties growing in China, which includes P. bretschneideri, P. pyrifolia, P. pyrifolia Nakai and Pyrus sp. nr. Communis. Regarding flavonoids, catechin is the major flavonoids in the five pear varieties except for Packham's Triumph. It is also the dominant flavonoids among the five selected flavonoids, this is consistent with the conclusion from Öztürk et al. [89]. In a small amount, epicatechin gallate was only detected in three pear varieties, which are Beurre Bosc, Josephine de Malines and Rico. Epicatechin is high in Beurre Bosc but less abundant in Winter Nelis. In addition, quercetin is abundant in Josephine de Malines and low in Packham's Triumph and Rico. In the previous study, Brahem et al. [7] also quantified the epicatechin in the flesh and peel of 16 European grown pear varieties (such as Rochas, William rouge and William vert) and found that the concentration is higher in the peel than flesh. However, Arts et al. [70] quantified that there is no epicatechin gallate detected in 2 Netherlands grown pear varieties, including Conference and Doyenne du Comice. This may be due to the difference in pear varieties, growing regions and extraction solvent.

Conclusions
Based on these data, Beurre Bosc has the highest TPC, TTC and DPPH concentrations while Josephine de Malines was high in TFC, ABTS, FRAP and TAC values. Winter Nelis was consistently low across the antioxidant assays except for TAC. The present work successfully employed LC-ESI-QTOF-MS/MS analysis to sperate and identify the phenolic profile in the pulp of five Australian grown pear varieties. A total of 73 phenolic compounds were detected in the five Australian grown pear varieties studied. The present work also used HPLC to quantify the phenolic compounds in pears and a total of five phenolic acids and five flavonoids were quantified. Chlorogenic acid and catechin are the dominant phenolic compounds in pears. This project provides comprehensive qualitative and quantitative information about the phenolic compounds present in Australian grown pears. These data may be useful for developing strategies and products in the nutraceutical, pharmaceutical and food industries.