Nineteen peach [Prunus persica (L.) Batsch] genotypes and 45 plum (Prunus salicina Erhr. and hybrids) genotypes with different flesh and skin color were analyzed for their antioxidant content and activity. Anthocyanin content, phenolic content, and antioxidant activity were higher in red-flesh than in light-colored flesh peaches. Carotenoid content was higher in yellow-flesh peaches than in light-colored ones. Red-flesh plums generally had higher anthocyanin and phenolic contents than the other plums but not necessarily greater antioxidant capacity. The total phenolic content had the most consistent and highest correlation with antioxidant activity, indicating that it is more important in determining the antioxidant activity of peaches and plums than are the anthocyanin or carotenoid contents. In general, the wide range of phytochemical content and antioxidant activity found indicates that the genetic variability present can be used to develop cultivars with enhanced health benefits.
Fruit have long been promoted for their health benefits in preventing various cancers and age-related diseases (Prior and Cao, 2000; Wargovich, 2000). This interest in functional foods has guided plant breeders of crops such as blueberries (Vaccinium L.) (Prior et al., 1998) and potatoes (Solanum tuberosum L.) (Reyes et al., 2004) to select genotypes with higher phenolic content and antioxidant activity. Likewise, the Prunus L. breeding program at Texas A & M University with the collaboration of the U.S. Dept. of Agr. (USDA) stone fruit breeding program in Byron, Ga. and the USDA stone fruit breeding program in Parlier, Calif., are working toward developing peaches and plums with higher levels of compounds potentially beneficial to human health.
The phytochemicals reported in Prunus L. include carotenoids, anthocyanins, and other phenolics (Cevallos-Casals et al., 2005; Gao and Mazza, 1995; Gil et al., 2002; Radi et al., 1997; Senter and Callahan, 1991; Tourjee et al., 1998; Weinert et al., 1990; Werner et al., 1998). Orange-fleshed peaches have the carotenoids β-carotene and β-cryptoxanthin (Tourjee et al., 1998). Several hydroxycinnamates such as chlorogenic acid and neochlorogenic acid, flavan 3-ols such as catechin and epicatechin, and flavonols such as quercetin 3-rutinoside have been identified in peaches and plums (Kim et al., 2003a; Tomás-Barberán et al., 2001).
Anthocyanins and other phenolic compounds are responsible for many health benefits related to cancer prevention and cardiovascular health (Edenharder et al., 2003; Moline et al., 2000; Sun et al., 2002; Wang et al., 1997, 1999; Zhou et al., 2004). The antioxidant activity in both peaches and plums depends on the genotype tested. Researchers have reported that blueberry has the highest antioxidant activity among fruit; however, the levels found in red-fleshed plums overlap the levels found in blueberry (Cevallos-Casals et al., 2005; Prior et al., 1998; Wang et al., 1996). There is a good correlation between total phenolic compounds and antioxidant activity among red-flesh peaches and plums (Cevallos-Casals et al., 2005). Furthermore, the contribution of phenolic compounds and anthocyanins to the antioxidant activity is much more important than the contribution of vitamin C or carotenoids (Chun et al., 2003; Gil et al., 2002; Kim et al., 2003b).
Thus, phenolic compounds are an interesting target for breeding programs. Because previous work dealt with a narrow range of germplasm, this research examines a wide range of germplasm to determine the variability of these bioactive compounds among peach and plum germplasm.
Materials and Methods
Nineteen peach and 45 plum genotypes from the USDA stone fruit breeding program at Byron, Ga., and the USDA stone fruit breeding program at Parlier, Calif., representing the range of flesh color available in these crops were picked at commercial (firm ripe) maturity during the 2003 harvest season. These were either hand-carried or sent by overnight carrier and immediately on arrival in the laboratory at Texas A & M University, the plums and peaches were stored at 2 to 5 °C. Within 5 d of storage, the fruit was visually inspected, the stones removed, and tissue (flesh plus skin) samples were frozen at –80 °C until analyzed.
Six to 12 fruit at the firm ripe stage were chosen from each genotype for measuring total phenolics, carotenoids, anthocyanin content, and antioxidant activity. Three replicates, each using two to four fruit, were used. Mesocarp and exocarp sections were used together to determine the level of active compounds. All results are expressed on a fresh weight basis.
Phenolics were quantified by the Folin-Ciocalteau method (Cevallos-Casals and Cisneros-Zevallos, 2003; Swain and Hillis 1959). Five grams of frozen tissue (flesh plus skin) was homogenized with 25 mL of methanol in a conical screw-cap tube using a vortex mixer. Samples were stored overnight at 4 °C and then centrifuged (model J2–21; Beckman Instruments, Fullerton, Calif.) for 20 min at 29,000 g at 2 °C. A 0.5-mL aliquot sample of the methanol phase was taken and diluted with 8 mL of nanopure water. At the same time, a blank containing 0.5 mL of methanol was equally diluted and analyzed. Each sample and the blank, were combined with 0.5 mL of 0.25 N Folin-Ciocalteau reagent and allowed to react for 3 min before the addition of 1 mL 1 N Na2CO3. The reaction mixture was incubated for 2 h at room temperature and measurements of absorbance at 725 nm were taken. The spectrophotometer (model 8452A; Hewlett Packard Co., Waldbronn, Germany) was set to zero absorbance using the blank. Measurements were taken in a quartz cuvette. Every time the measurements were above 0.6 absorbance unit (AU), the samples were diluted and reanalyzed. The concentration of total phenolics was estimated from a chlorogenic acid (Sigma Chemical Co., St. Louis) standard curve in terms of milligrams of chlorogenic acid equivalents.
Total carotenoid content protocol was adapted from Talcott and Howard (1999). In indirect light, 2 g of frozen tissue was homogenized with 20 mL of ethanol solution containing 200 mg per L BHT into a falcon tube until uniform consistency. After centrifugation for 20 min at 29,000 g at 2 °C, the supernatant was transferred to a 50-mL graduated cylinder and solvent added to a final volume of 50 mL. The solution was transferred to a plastic container with a screw cap. Twenty-five milliliters of hexane was added to the peach and plum samples and the container was shaken vigorously. The solution was left for 30 min to allow separation of the phases before 12.5 mL of nanopure water was added and the solution was shaken vigorously. Again, the phases were allowed to separate and the hexane phase was used. Spectrophotometer was zeroed using hexane as a standard and the measurements were taken in a quartz cuvette at 470 nm. Every time the measurements were above 0.7 AU, the samples were diluted with hexane and reanalyzed. The concentration of total carotenoids was estimated from a β-carotene (Sigma Chemical Co.) standard curve in terms of milligrams of β-carotene equivalent.
Total anthocyanin content analysis was adapted from Fuleki and Francis (1968) by measuring the absorbance of extracts at pH 1 (Cevallos-Casals and Cisneros-Zevallos, 2003). Five grams of frozen tissue (flesh plus skin) was homogenized with 15 mL of 95% aqueous ethanol:1.5 N HCl solution (85:15) in a conical screw-cap tube using a vortex mixer. Samples were stored overnight at 4 °C and then centrifuged for 15 min at 29,000 g at 2 °C. A 2-g aliquot of the clear supernatant was taken from the sample and placed in a graduated cylinder and the volume was adjusted to 50 mL of solvent. The sample was transferred to a plastic container and was shaken vigorously with 25 mL of hexane to eliminate carotenoids. After 30 min to allow phase separation, the hexane phase was discarded and the spectrophotometer was zeroed with the anthocyanin extraction solvent as the blank. Readings were taken in a quartz cuvette at 535 nm and 700 nm. Every time the measurements were above 0.7 AU, the samples were diluted and reanalyzed. Anthocyanins were quantified as mg cyaniding-3-glucoside using a molar extinction coefficient of 25,965/(cm M) and a molecular weight of 494 (Abdal-Aal and Hucl, 1999).
Antioxidant activity was quantified by the 2, 2-diphenyl-1-picrylhydrazyl (DPPH) radical method (Brand-Williams et al., 1995). Five grams of frozen tissue (flesh plus skin) was homogenized with 25 mL of methanol in a conical screw-cap tube using a vortex mixer. Samples were stored overnight at 4 °C and then centrifuged for 20 min at 29,000 g at 2 °C. Before running the reaction, the spectrophotometer was blanked with methanol, and DPPH was diluted with methanol from a stock solution to reach an absorbance of 1.1 AU at 515 nm; 150 μL of sample was combined with 2850 μL of the DPPH solution. Samples and the methanol blank were left to react for 24 h. Absorbance was measured with a quartz cuvette at 515 nm. When the absorbance was below 0.2 AU, samples were diluted with methanol and reanalyzed. Antioxidant activity was estimated as equivalents of 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox; Sigma Chemical Co.) by comparison with a standard curve.
Specific antioxidant capacity.
In addition, specific antioxidant capacity was defined in this study as the ratio of total antioxidant capacity per total soluble phenolics and expressed as micrograms Trolox equivalents per milligram chlorogenic acid. The specific antioxidant capacity provides information of the effectiveness of phenolics to neutralize free radicals. A higher specific antioxidant capacity means phenolic compounds have a higher capacity to stabilize free radicals.
Analyses of variance was performed and means were compared with the Duncan's multiple range test (P < 0.05). Correlation analysis was applied to assess the relationship between the phytochemical levels and antioxidant activity. Statistical analyses were performed using SPSS (version 11.0 for Windows; SPSS, Chicago).
Results and Discussion
Genotypes with red-colored flesh had higher anthocyanin (milligrams of cyanidin 3-glucoside per 100 g of tissue) content in peach (≈45 to 266) and in plum (≈60 to 611) as compared with the light-colored flesh genotypes of peach (2 to 7) (Tables 1 and 2) and plums (≈2 to 376) (Tables 2 and 3). There is an overlap in values in plum (light versus red) (Table 3) but not in peaches (Table 1). Plums, even among the light-colored genotypes, had greater anthocyanin content than peaches (Table 2). Low concentrations of anthocyanins found in light-colored peaches were from the skin or close to the stone, whereas in light-colored plums, this pigment is from the dark red skins and the late red coloration of the flesh in some genotypes. The main anthocyanins reported in peach and plum were cyanidin 3-glucoside and cyanidin 3-rutinoside (Kim et al., 2003a; Tomás-Barberán, et al., 2001; Wu and Prior, 2005). Other anthocyanins found were cyanidin 3-acetyl glucoside, cyanidin 3-galactoside (small amounts) (Tomás-Barberán et al., 2001; Wu and Prior, 2005), peonidin-3-glucoside (1.1 to 1.2 mg/100 g), and peonidin derivatives (1.9 to 11.5 mg/100 g) (Kim et al., 2003a). Anthocyanins have antioxidant activity in vitro and in vivo (Tsuda et al., 1994, 1998; Wang et al., 1997).
Total anthocyanins, total phenolics, total carotenoids, and antioxidant activity of 19 peach genotypes with flesh color ranging from white to red.
Comparison of phytochemical and antioxidant activity among the different flesh color groups for peach and plum fruit.
Total anthocyanins, total phenolics, total carotenoids, and antioxidant activity of 45 plum genotypes with flesh color ranging from light to red color.
As seen with anthocyanins, the phenolic contents (milligrams chlorogenic acid equivalents per 100 g of tissue) were higher in red-flesh peaches (≈228 to 1260) and plums (≈182 to 898) than in light-colored flesh peaches (≈137 to 371) or plums (≈214 to 474) (Tables 1–3). Light-colored flesh plums had higher phenolic content than light-colored flesh peaches; however, the mean phenolic content of red-fleshed peaches and plums was similar (Table 2).
There was an overlap of values between the light- and red-colored material indicating that genotypes not rich in anthocyanins may contain high contents of other phenolic compounds (Tables 1 and 3). Given that this study surveyed a wider range of peach and plum germplasm, it is not surprising that the level of phenolics measured in the peaches and plums was higher than that previously reported by Los et al. (2000) (≈160 to 300 mg of gallic acid per 100 g) using the Folin method, by Gil et al. (2002) (≈14 to 109 mg of chlorogenic acid per 100 g) using the high-performance liquid chromatography method, or by Kim et al. (2003a) (≈125 to 372.6 mg of gallic acid per 100 g) using the Folin method. However, the phenolic content [milligrams of chlorogenic acid (CGA) per 100 g] of the red-flesh genotypes was similar in concentration to red-flesh peach (≈100 to 563) and red-flesh plum (≈300 to 449) values reported by Cevallos-Casals et al. (2005). Additionally, the phenolic content (milligrams of CGA per 100 g) of the plum genotypes was comparable to those of blueberries reported previously by Connor et al. (2002) (≈335 to 595) and by Cevallos-Casals and Cisneros-Zevallos (2004) (≈292 to 672). In general, the phenolic content found in these genotypes was higher than those previously reported for stone fruit. The genotypes with the highest phenolic concentrations (e.g., BY99P4362 peach and BY00M2977 plum) had a distinct bitter taste indicating that too high a level of phenolics may be detrimental. Phenolic and anthocyanin concentration in the skin of peaches as well as plums is higher compared with the flesh; however, although the exocarp is a concentrated source of these compounds, it only represents ≈8% of the fruit weight. Thus, the total distribution of phenolic compounds in skin and flesh per fruit is ≈30% and 70%, respectively (Cevallos-Casals et al., 2005).
Carotenoid content (milligrams β-carotene per 100 g tissue) found in yellow-fleshed peach genotypes (0.8 to 3.7) was higher than in white-fleshed peaches (0.0 to 0.1 mg β-carotene per 100 g tissue). The concentration of total carotenoids (0.1 to 1.9 mg β-carotene per 100 g tissue) in red-fleshed peaches varied with respect to the white or yellow base color in which the red pigments are in because carotenoid and anthocyanin content of fruit are inherited independently (Werner et al., 1998) (Tables 1 and 2). Carotenoid content in plums varied from 0.2 to 1.5 mg β-carotene per 100 g tissue (Table 3) with light-colored flesh genotypes having slightly higher amounts than the red-fleshed genotypes (Table 2). Carotenoid concentrations found in peaches are higher than in plums (Table 2). Previous studies have reported that peaches have β-carotene and β-cryptoxanthin as the main carotenoids with small amounts of α-carotene (Gil et al., 2002).
The antioxidant activity (micrograms Trolox per gram) of red-fleshed (≈2074 to 13,505) genotypes was higher than the light-colored flesh (≈437 to 1128) genotypes in peaches (Tables 1 and 2) but not in plums (Tables 2 and 3). Light-colored plums had a higher antioxidant activity than light-colored peaches and lower antioxidant activity (AOA) than red-fleshed plums (Table 2). Similar to total phenolics, the AOA found is higher than that previously reported in stone fruit (Cevallos-Casals et al., 2005; Gil et al., 2002).
The most consistent and highest correlations were those between total phenolics and antioxidant capacity (Table 4). These results confirm previous reports that phenolics were better correlated with antioxidant capacity than were anthocyanins, vitamin C, or carotenoids in nectarines, peaches, and plums (Cevallos-Casals et al., 2005; Gil et al., 2002).
Correlation coefficients between antioxidant activity and anthocyanins, antioxidant activity and total phenolics, and antioxidant activity and carotenoid content in peach and plum.
Because the AOA of a phenolic compound depends on its specific structural features like the number of available hydroxyl groups (Rice-Evans et al., 1996, 1997; Shahidi and Naczk, 1995), the antioxidant capacity of a solution containing a mixture of phenolic compounds will depend on the specific phenolic profile, which can be qualitative (type of phenolics present) or quantitative (the relative amounts or proportions of phenolics present). Reporting the antioxidant activity on a phenolic basis (specific antioxidant activity) for peaches and plums will give information on the antioxidant activity of the specific phenolic profile present for each crop and genotype. According to the results, plum genotypes showed specific antioxidant activities ranging from ≈3 to ≈14 μg Trolox equivalents per milligram chlorogenic acid, whereas peaches ranged from ≈2.5 to ≈22 μg Trolox equivalents per milligram chlorogenic acid (Figs. 1 and 2). The larger variability seen in peaches is reflective of the wider diversity of germplasm with both commercial and noncommercial genotypes as compared with the plum germplasm examined, which contained only commercial-type germplasm.
The wide range of phytochemicals and antioxidant activity levels found in the germplasm studied indicates that adequate genetic variability is present to potentially develop cultivars with enhanced health benefits. The selection of crops rich in phenolic compounds with enhanced antioxidant activity would be a first step. Of the phytochemicals analyzed, the phenolic content was better correlated to antioxidant activity than either anthocyanin or carotenoid content. Further studies regarding rapid selection procedures, secondary effects of phenolics on fruit quality and postharvest traits, and the bioactive properties of selected peach and plum genotypes are needed.
Abdal-AalE.S.M.HuclP.1999A rapid method for quantifying total anthocyanins in blue aleurone and purple pericarp wheatsCereal Chem.76350354
Cevallos-CasalsB.ByrneD.OkieW.R.Cisneros-ZevallosL.2005Selecting new peach and plum genotypes rich in phenolic compounds and enhanced functional propertiesFood Chem.96273280
Cevallos-CasalsB.A.Cisneros-ZevallosL.2003Stoichiometric and kinetic studies of phenolic antioxidants from Andean purple corn and red-fleshed sweetpotatoJ. Agr. Food Chem.5133133319
Cevallos-CasalsB.A.Cisneros-ZevallosL.2004Stability of anthocyanin-based aqueous extracts of Andean purple corn and red-fleshed sweet potato compared to synthetic and natural colorantsFood Chem.866977
ChunO.K.KimD.MoonH.Y.KangH.G.LeeC.Y.2003Contribution of individual polyphenolics to total oxidant capacity of plumsJ. Agr. Food Chem.5172407245
ConnorA.M.LubyJ.J.HancockJ.F.BerkheimerS.HansonE.J.2002Changes in fruit antioxidant activity among blueberry cultivars during cold-temperature storageJ. Agr. Food Chem.50893898
EdenharderR.KriegH.KottgenV.PlattK.L.2003Inhibition of clastogenicity of benzo[a]pyrene and of its trans-7,8-dihydrodiol in mice in vivo by fruit, vegetables, and flavonoidsMutat. Res.537169181
FulekiT.FrancisF.J.1968Quantitative methods for anthocyanins 1. Extraction and determination of total anthocyanin in cranberriesFood Sci.337277
GaoL.MazzaG.1995Characterization, quantification, and distribution of anthocyanins and colorless phenolics in sweet cherriesJ. Agr. Food Chem.43343346
GilM.Tomas-BarberanF.Hess-PierceB.KaderA.2002Antioxidant capacities, phenolic compounds, carotenoids, and vitamin A contents of nectarine, peach, and plum cultivars from CaliforniaJ. Agr. Food Chem.5049764982
KimD.O.ChunK.KimY.J.MoonH.LeeC.Y.2003aQuantification of polyphenolics and their antioxidant capacity in fresh plumsJ. Agr. Food Chem.5165096515
PriorR.L.CaoG.MartinA.SoficE.McEwenJ.O'BrienC.LischnerN.EhlenfeldtM.KaltW.KrewerG.MainlandC.1998Antioxidant capacity as influenced by total phenolic and anthocyanin content, maturity, and variety of Vaccinium speciesJ. Agr. Food Chem.4626862693
RadiM.MahrouzM.JaouadA.TacchiniM.AubertS.HuguesM.AmiotM.J.1997Phenolic composition, browning susceptibility, and carotenoid content of several apricot cultivars at maturityHortScience3210871091
ReyesL.F.MillerJ.C.Cisneros-ZevallosL.2004Environmental conditions influence the content and yield on anthocyanins and total phenolics in purple and red-flesh potatoes during tuber developmentAmer. J. Potato Res.81187193
Rice-EvansC.A.MillerN.PagangaG.1996Structure–antioxidant activity relationships of flavonoids and phenolic acidsFree Radic. Biol. Med.20933956
SenterS.D.CallahanA.1991Variability in the quantities of condensed tannins and other major phenols in peach fruit during maturationJ. Food Sci.5615851587
ShahidiF.NaczkM.1995Food phenolics: An overview15ShahidiF.NaczkM.Food phenolics: Sources chemistry effects and applicationsTechnomic PublishingLancaster, Pa
SwainT.HillisW.E.1959The phenolic constituents of Prunus domestica I. The quantitative analysis of phenolic constituentsJ. Sci. Food Agr.106368
TalcottT.S.HowardR.L.1999Phenolic autoxidation is responsible for color degradation in processed carrot pureeJ. Agr. Food Chem.4721092115
Tomás-BarberánF.A.GilM.I.CreminP.WaterhouseA.L.Hess-PierceB.KaderA.A.2001HPLC-DAD-ESIMS analysis of phenolic compounds in nectarines, peaches, and plumsJ. Agr. Food Chem.4947484760
TourjeeK.R.BarrettD.M.RomeroM.V.GradzielT.M.1998Measuring flesh color variability among processing clingstone peach genotypes differing in carotenoid compositionJ. Amer. Soc. Hort. Sci.123433437
WangH.NairM.StrasburgG.ChangY.BoorenA.GrayJ.DeWittD.1999Antioxidant and antiinflammatory activities of anthocyanins and their aglycon, cyanidin from tart cherriesJ. Nat. Prod.62294296
WuX.PriorR.L.2005Systematic identification and characterization of anthocyanins by HPLC-ESI-MS/MS in common foods in the United States: Fruits and berriesJ. Agr. Food Chem.5325892599
ZhouJ.-R.YuL.MaiZ.BlackburnG.L.2004Combined inhibition of estrogen-dependent human breast carcinoma by soy and tea bioactive components in miceInt. J. Cancer108814