Armando Carrillo-L´opez and Elhadi Yahia
Identification of phenolic compounds was done by means of liquid chromatography (HPLC) coupled to mass spectrometry (MS) using the electrospray ionization interface (ESI). Quantification of phenolic compounds was carried out by using HPLC with diode array detector (DAD) in exocarp and mesocarp of tomato fruit at 6 different ripeness stages (mature-green, breakers, turning, pink, light-red, and red). Several phenolic compounds were identified including chlorogenic acid, caffeic acid, p-coumaric acid, ferulic acid, and rutin and some combined phenolic acids were tentatively identified, mainly glycosides, such as caffeoyl hexose I, caffeoyl hexose II, caffeoylquinic acid isomer, dicaffeoylquinic acid, p-coumaroyl hexose I, p-coumaroyl hexose II, feruloyl hexose I, feruloyl hexose II, siringyl hexose, and caffeoyl deoxyhexose hexose. Fruit exocarp had higher quantities of total soluble phenolics (TSP) compared to mesocarp. During ripening, TSP increased in both exocarp and mesocarp, mainly in exocarp. While rutin increased, chlorogenic acid decreased in both tissues: exocarp and mesocarp.
Abstract:
Keywords: caffeic acid, chlorogenic acid, Licopersicon esculentum, rutin, TOF-MS
Since exocarp showed higher TSP content than mesocarp, it is recommended to food processors that tomato exocarp should not be discarded but processed along with tomato mesocarp.
Practical Application:
Introduction It has been suggested that the consumption of fruit and vegetables is able to reduce the risk of some chronic diseases, such as cancer and cardiovascular diseases (Willett 1994; Temple 2000). For example, the consumption of fresh tomatoes and tomato products has been inversely related to the development of some types of cancer (Giovannucci 1999). Phenolic compounds have been suggested to play an important role in human health due to the properties related to their antioxidant capacity, by which they may provide antioxidant protection against oxidative processes and achieve the prevention of some diseases (Le Marchand and others 2000; Manach and others 2004). Phenolic compounds are important components in many fruits and vegetables. They are water-soluble substances that tend to accumulate in the dermal tissues of plant body because of their potential role in protection against ultraviolet radiation (Strack 1997), and have been related to growth, development, and defense activities of the plant against pathogens, parasites, and predators, in addition of imparting color to some fruits and vegetables (Robards and Antolovich 1997). In tomato, Minoggio and others (2003) reported that the main phenolic acids were chlorogenic and caffeic acid, among others such as ferulic acid, rutin, and a chlorogenic acid derivative that were also present. Slimestad and Verheulm (2009) also reported that chlorogenic acids and related compounds are the main phe-
nolic compounds besides flavonoids in tomatoes. Concerning to chlorogenic acid health properties, Farah and Donangelo (2006) reported that this phenolic compound has a number of beneficial health properties related to their potent antioxidant activity as well as hepatoprotective, hypoglycemic, and antiviral activity. Analyses of phenolic compounds usually start with acid or alkali hydrolysis to release phenolic compounds from their glycosides and ester conjugates. However, in plants the free forms of phenolic acids are rarely present, whereas the natural occurrence of phenolic compounds is mainly as conjugated forms (Robbins 2003). The exocarp of several fruit and vegetables has been shown to be richer in nutrients/phytochemicals (including phenolic compounds) compared to the mesocarp (Arima and Rodriguez-Amaya 1988; Chang and others 2000; Tom´as-Barber´an and others 2001; Silva and others 2002). This work investigated the qualitative and quantitative changes of phenolic compounds in mesocarp and exocarp of “Caiman” tomato fruit during ripening, looking for differences between peel and pulp, among fruit ripeness stages and between hydrolyzed and nonhydrolyzed extracts since the reports of phenolic composition are mostly concerned with the free phenolic acids (those derived from the hydrolyzed extracts plus the naturally occurring free phenolic acids) rather than conjugated phenolics (those derived from the nonhydrolyzed extracts as naturally occurring conjugated compounds)
Materials and Methods MS 20130317 Submitted 3/6/2013, Accepted 9/19/2013. Author Armando Carrillo-L´opez is with Maestr´ıa en Ciencia y Tecnolog´ıa de Alimentos, Facultad de Sources of standards and solvents Ciencias Qu´ımico-Biol´ogicas, Universidad Aut´onoma de Sinaloa, Culiac´an, Sinaloa, Standards of p-coumaric acid, chlorogenic acid, caffeic acid, 80000, M´exico. Author Elhadi M. Yahia is with Facultad de Ciencias Naturales, Universidad Aut´onoma de Quer´etaro, Juriquilla, 76230, Quer´etaro, M´exico. Direct and ferulic acid were purchased from Sigma-Aldrich (St. Louis, Mo., U.S.A.). Methanol (HPLC grade), acetonitrile (HPLC grade) inquiries to author Carrillo-L´opez (E-mail: [email protected]).
and formic acid (99% purity) were obtained from JT baker R C 2013 Institute of Food Technologists
doi: 10.1111/1750-3841.12295 Further reproduction without permission is prohibited
Vol. 78, Nr. 12, 2013 r Journal of Food Science C1839
C: Food Chemistry
HPLC–DAD–ESI–MS Analysis of Phenolic Compounds During Ripening in Exocarp and Mesocarp of Tomato Fruit
Phenolics changes during tomato ripening . . .
Figure 1–Total soluble phenolics content in exocarp and mesocarp of “Caiman” tomato at different stages of ripeness. MG, mature green; B, breakers; T, turning; P, pink: LR, light-red; R, red.
40
Total soluble phenolics, mg GAE/100 g fw
C: Food Chemistry
(Baker Mallinckrodt, M´exico). HPLC-grade water was used for Ciocalteu reagent (dilution 1:10) and 120 μL of 7.5% of Na2 CO3 identification analyses. All other chemical reagents used were of were added. The plates were incubated for 2 h in darkness, and absorbance was measured at 630 nm using a Dynex MRX microplate analytical grade. reader (Dynex Technology Chantilly, Va., U.S.A.). Results were expressed as milligrams of gallic acid equivalents (GAE)/100 g Plant Material and Sampling Tomato (Lycopersicon esculentum Cv Caiman) was grown in fresh weight (fw). greenhouses from Ac´ambaro, Guanajuato, M´exico. It was harvested at its mature-green stage of ripeness and transported to the Phytochemicals & Nutrition Laboratory at the Faculty of Natu- HPLC–DAD–ESI–MS analytical conditions ral Sciences of the Autonomous Univ. in Queretaro. Fruits were Samples (50 μL) from the different extracts were injected into ripened in air at 25 ◦ C, 80% to 85% RH, and sampled at 6 dif- a HP 1100 series HPLC–DAD system (Hewlett-Packard GmbH, ferent ripeness stages (mature green, breakers, turning, pink, light Waldbronn, Germany) coupled to a 6210 time of flight (TOF) red, and red) for evaluation. The ripeness assessment was done mass spectrometer (Agilent, Palo Alto, Calif., U.S.A.). The pheusing the color classification requirements in tomatoes from the nolic compounds were separated using a 5-μm X-Terra (4.6 × 250 United States Standards for Grades of Fresh Tomatoes. Exocarp mm) column (Waters Co., Milford, Mass., U.S.A.) operating at 25 (thickness ∼ 1 mm) was separated from mesocarp using a potato ◦ C. The mobile phase was composed of 1% formic acid in water peeler. Exocarp was freeze-dried, and both exocarp and mesocarp (A) and acetonitrile (B) following a linear gradient from 100% A were kept at –80 ◦ C until analysis. Seeds were discarded. and 0% B to 0% A and 100% B in 60 min using a flow rate of 0.5 mL/min. The 1% formic acid in water was to promote molecular Phenolic compounds extraction ionization. Mass spectrometer system was equipped with an elecFive grams of fresh mesocarp or 0.4 g of freeze-dried exocarp trospray interface operating in the negative ionization mode. A were each mixed with 10 mL of extraction solution (80% methanol mass hunter manager software (A.02.01) was used. Nitrogen was in water, containing 1% formic acid) and thoroughly ground in utilized as drying gas at 350 ◦ C, at a flow rate of 11.5 L/min and mortar and pestle. The homogenate was sonicated during 15 min at a pressure of 45 . Fragmentor and capillary conditions were at room temperature and filtered through Whatman 3 filter paper. 100 and 4000 V, respectively. This was named the nonhydrolyzed extract. Two milliliters of this extract were mixed with 2 mL of 2.4 N HCl for hydrolysis in water bath at 80 ◦ C for 2 h and then the test tubes were cooled Statistical analysis with water at room temperature. This latter extract was named the Data are means of 3 determinations. Analysis of variance hydrolyzed extract. Both extracts were filtered through 0.45-μm (ANOVA) was conducted and the significance of difference beMillipore membrane (Millipore Corp., Bedford, Mass., U.S.A.) tween means was determined by Tukey method at a 5% signifibefore injection to the HPLC–DAD–ESI–MS system. cance level. Limit of detection (LOD) and limit of quantitation (LOQ) were obtained from the calibration curves for chlorogenic, Quantification of TSP caffeic, ferulic, and p-coumaric acids, using the slope value (S) Aliquots from exocarp or mesocarp nonhydrolyzed extracts of the respective regression line and the standard deviation of the were diluted 1:10 with HPLC-grade water and 30 μL of diluted y-intercepts (σ ). The used formulas were LOD = 3.3σ /S and sample per hole was placed in 96-hole plates, and 150 μL of Folin– LOQ = 10σ /S.
Exocarp Mesocarp 30
20
10
0 MG
B
T
P
Tomato ripeness stage
C1840 Journal of Food Science r Vol. 78, Nr. 12, 2013
LR
R
Phenolics changes during tomato ripening . . . Table 1–Fragment-ion identification of nonhydrolyzed extracts from exocarp and mesocarp of tomato.
2 (17.1)
3 (17.3)
4 (17.9) 5 (18.2) 6 (18.5)
7 (18.8)
8 (19.2) Chlo (19.5)
10 (20.8) 11 (21.6)
Caf (22.7) 14 (23.4) Rut (24.2) Cou (25.6)
Fragments of negative ions, (m/z) ]−
325 [326-H 163 [326-H-162] − 119 [326-H-162-44] − 341 [342-H] − 179 [341-H-162] − 135 [341-H-162-44] − 515 [516-H ] − 353 [516-H-162] − 355 [356-H] − 193 [356-H-162] − 341 [342-H] − 179 [342-H-162] − 135 [342-H-162-44] − 353 [354-H] − 173 [354-H-180] − 179 [354-H-174] − 191 [354-H-162 ] − 325 [326-H ] − 163 [326-H-162] − 353 [354-H] − 173 [354-H-162-18] − 179 [354-H-174] − 191 [354-H-162] − 355 [356-H] − 193 [356-H-162] − 359 [359-H] − 197 [359-H-162] − 153 [359-H-162-44] − 179 [360-H-179] − 135 [180-H-44] − 487 [488-H] − 179 [488-H-162-146] − 609 [610-H] − 163 [164-H] − 119 [164-H-44] −
Tentative fragment ion identification
Tentative compound
p-Coumaroyl hexose p-Coumaroyl ion Decarboxilated p-coumaroyl Caffeoyl-hexose Caffeoyl-ion Decarboxilated caffeoyl Dicaffeoylquinic acid ion CQ acid Feruloyl hexose Feruloyl ion Caffeoyl hexose Caffeoyl ion Decarboxilated caffeoyl CQ CQ ion minus dehydrated quinic moiety Caffeic acid ion CQ ion minus caffeoyl moiety p-Coumaroyl hexose p-Coumaroyl ion
p-Coumaroyl hexose I
Chlo acid ion minus dehydrated quinic acid Caffeic acid ion Chlo ion minus caffeoyl moiety Feruloyl hexose Feruloyl ion Syringyl hexose Syringic ion Decarboxylated siringic ion Decarboxylated caffeoyl ion Deoxyhexose hexose caffeoyl Caffeoyl ion
Caffeoyl hexose I
Dicaffeoylquinic acid Feruloyl hexose I p-Caffeoyl hexose II
CQ isomer
p-Coumaroyl hexose II
Feruloyl hexose II Siringyl hexose
Caffeoyl deoxyhexose hexose
Coumaroyl ion Decarboxilated coumaroyl
Peaks 1, 9, 12, 13, 16, 17, 18, 19, 20 were not identified. Tr, retention time; Chlo, Chlorogenic acid; Caf, Caffeic acid; Rut, Rutin; Cou, p-Coumaric acid; CQ, caffeoylquinic acid.
Results and Discussion Total soluble phenolics TSP were more than twice higher in the exocarp than in the mesocarp during all stages of ripening (Figure 1). TSP increased significantly (P < 0.05) in both tissues during ripening, more pronouncedly in exocarp. The TSP levels observed in the mesocarp (12.7 to 16.8 mg/100 g fw) were similar to those observed by Minoggio and others (2003) in tomato, but higher than those observed by Kacjan-Marˇsi´c and others (2011) for 11 cultivars from which the cultivar cherry showed the highest content (10.39 mg/100 g fw). The higher TSP content in the exocarp is in agreement with the fact that phenolic compounds tend to accumulate mainly in the dermal tissues of plant body because of their potential role in protection against ultraviolet radiation (Strack 1997), in addition to the defense activities of the plant against pathogens, parasites, and predators (Robards and Antolovich 1997). Identification and quantification of phenolic compounds in the exocarp and mesocarp HPLC–DAD–ESI–MS detection operating in the negative ion mode was used to identify phenolic compounds in exocarp and mesocarp of tomato fruit at 6 different ripeness stages. The compounds were identified as aglycones obtained after acid hydrolysis (hydrolyzed extracts) and as naturally occurring aglycones or combined phenolic compounds obtained in the nonhydrolyzed extracts. Identification of aglycones was based on retention times, and
UV and mass spectra by comparison with commercial standards, and a tentatively identification of naturally occurring combined phenolic compounds was based on their fragmentation patterns of the mass spectra. The quantification of the identified compounds was done by HPLC–DAD comparing the UV-peaks areas with calibration curves produced with commercially obtained standards. A total of 15 phenolic compounds were identified or tentatively identified, taking into account exocarp, mesocarp and if the extracts were hydrolyzed or nonhydrolyzed.
Naturally occurring aglycones From the nonhydrolyzed extracts, 3 naturally occurring aglycones of phenolic acids were identified; chlorogenic acid in both exocarp and mesocarp, caffeic acid only in exocarp, and p-coumaric acid only in mesocarp.
Naturally occurring combined phenolic acids Ten naturally occurring combined phenolic acids were tentatively identified in mesocarp and/or exocarp (Table 1, Figure 2). p-Coumaroyl hexose I, caffeoyl hexose II, p-coumaroyl hexose II, caffeoyl deoxyhexose hexose, and rutin were shown in both, mesocarp and exocarp, whereas caffeoyl hexose I, feruloyl hexose I, feruloyl hexose II, and syringyl hexose were observed only in mesocarp and dicaffeoylquinic acid only in exocarp). Vol. 78, Nr. 12, 2013 r Journal of Food Science C1841
C: Food Chemistry
Peak(Tr, min)
Phenolics changes during tomato ripening . . .
C: Food Chemistry
[354-H-173]− corresponds to the caffeoyl ion derived from the loss of the dehydrated quinic moiety from the deprotonated caffeoylquinic acid, and 191 [354-H-162]− corresponds to the ion derived from the loss of the dehydrated caffeoyl moiety. Similar results have been reported by Sun and others (2007) and by S´anchez-Rabaneda and others (2003). Caffeic acid showed its [MH]− deprotonated molecule (m/z 179) and the [M-H-44]− fragment ion (m/z 135) corresponding to the fragment derived from the loss of the carbon dioxide moiety of caffeic acid (Table 1). The identification of the naturally occurring glycoside-combined phenolic compounds (Figure 2, Table 1) was based on the fragmentation pattern derived from the mass spectra obtained from the different chromatographic peaks observed (Figure 2). Peaks 2 and 8 presented different retention times, 17.1 and 19.2 min, but similar fragmentation patterns whose m/z of 325 ([326-H]− ), 163 [326-H-162]− , and 119 ([164-H-44]− ) correspond to the deprotonated molecule of p-coumaroyl hexose, the p-coumaroyl ion derived from the deglycosilation of the molecule of coumaroyl hexose, and the decarboxilated coumaroyl ion, respectively. Thus, peaks 2 and 8 were tentatively identified as isomers of glycosilated molecule of p-coumaroyl and were assigned as p-coumaroyl hexose I and p-coumaroyl hexose II, respectively. Peaks 3 and 6 also presented different retention times, 17.3 and 18.5 min, but the same fragmentation pattern whose m/z of 341 ([342-H]− ), 179 ([342-H-162]− ), and 135 ([179-H-44]− ) correspond to the deprotonated molecule of caffeoyl hexose, the deglycosilated molecule of caffeoyl hexose previously dehydrated (caffeoyl ion), and the decarboxilated caffeoyl ion, respectively. Thus, peaks 3 and 6 were tentatively identified as glycosilated isomers of caffeic acid and was assigned as caffeoyl hexose I and caffeoyl hexose II, respectively. Peaks 4, 7, and Chlo, presented the retention times 17.9, 18.8, and 19.5 min, respectively, and showed a fragmentation pattern that exhibited in common a molecular ion m/z 353 corresponding to the caffeoylquinic acid ion. Peak 4 showed a fragmentation pattern whose m/z 515 ([516-H]− ) and 353 ([516-H-162]− ) is believed to correspond to the deprotonated molecule of dicaffeoylquinic acid, in agreement with Moco and others (2006). Both, peak 7 and peak chlo presented the same fragmentation pattern that corresponds to the caffeolquinic acid molecule. Thus, peak 7 corresponds to an isomerization form of the chlorogenic acid, whereas peak chlo was identified as chlorogenic acid (3caffeoylquinic acid) based on the retention time and mass spectra from the standard. Peak 10 presented a retention time of 20.8 min, and a fragmentation pattern whose m/z of 355 ([356-H]− ) and 193 ([356-H-162]− ) correspond to the deprotonated molecule of feruloyl hexose and the feruloyl ion (deglycosilated molecule of feruloyl hexose), respectively. Peak 11 presented a retention time of 21.6 min, and a fragmentation pattern whose m/z of 359 ([360-H]− ), 197 ([360-H-162]− ), and 153 ([198-H-44]− ), correspond to the deprotonated molecule of syringyl hexose, the syringyl ion (the deglycosilated molecule of siringyl hexose), and the decarboxilated siringyl ion, respectively. Peak 14 presented a retention time of 23.4 min, and a fragmentation pattern whose m/z of 487 ([488-H]− ) and 179 ([488-H-162–146]− ) correspond to the deprotonated molecule of deoxyhexose hexose caffeic acid and the caffeic ion derived from the loss of the deoxyhexose hexose moiety, respectively. Peak Rut presented a retention time of 24.2 min, and a fragmentation pattern whose m/z of 609 ([610H]− ) correspond to the deprotonated molecule of rutin. Peak Figure 2–HPLC–DAD chromatograms for phenolic compounds at 320 nm Cou with a retention time of 25.6 min presented a fragmentation of nonhydrolyzed extracts from exocarp and mesocarp of “Caiman” tomato − − fruit. A, ripe mesocarp; B, mature-green mesocarp; C, ripe exocarp; D, pattern whose m/z of 163 ([164-H] ) and 119 ([164-H-44] )
Aglycones derived from acid hydrolysis Four aglycones were identified in the hydrolyzed extracts (chlorogenic acid and caffeic acid in both mesocarp and exocarp, ferulic acid only in mesocarp and p-coumaric acid only in exocarp). The naturally occurring aglycone phenolic acids were identified in agreement with the retention times observed for their respective commercial standards being 19.5 min for chlorogenic acid, 22.7 min for caffeic acid, and 25.6 min for p-coumaric acid, respectively. Furthermore, the identities were corroborated with their respective fragmentation patterns derived from their mass spectra (Table 1). Chlorogenic acid showed its profile of m/z 353, 173, 179, and 191, where 353 ([354-H]− ) corresponds to the [MH]− deprotonated molecule of 3-caffeoylquinic acid, 173 [354-H180]− corresponds to the ion derived from the loss of the caffeoyl moiety from the deprotonated caffeoylquinic acid molecule, 179
mature-green exocarp.
C1842 Journal of Food Science r Vol. 78, Nr. 12, 2013
Phenolics changes during tomato ripening . . . Table 2–Phenolic acids content (mg/100 g fw) of nonhydrolyzed and hydrolyzed extracts from mesocarp and exocarp of “Caiman” tomato fruit.
Ripeness stage Mature green Breakers Turning Pink Light red Red LSD Exocarp Ripeness stage Mature green Breakers Turning Pink Light red Red LSD LOD LOQ
Nonhydrolyzed
Hydrolyzed
Chlorogenic
Caffeic
Chlorogenic
Caffeic
Ferulic
0.68 ab 0.90 a 0.67 ab 0.51 b 0.56 ab 0.40 b 0.344
nd nd nd nd nd nd –
0.46 a 0.57 a 0.49 a 0.44 a 0.44 a 0.41 a 0.344
0.02 a 0.07 a 0.10 a 0.08 a 0.08 a 0.16 a 0.207 Hydrolyzed
U-LOQ 0.14 a 0.21 b 0.20 b 0.23 b 0.30 b 0.10
Caffeic U-LOQ U-LOQ 0.06 a 0.10 a 0.12 a 0.25 a 0.207 0.004 0.012
Chlorogenic 1.90 a 1.80 a 1.80 a 1.80 a 1.40 b 0.97 c 0.344 0.017 0.053
Caffeic 0.54 a 0.60 a 1.43 b 1.61 b 1.34 c 1.53 c 0.207 0.004 0.012
p-Coumaric 0.04 a 0.09 a 0.39 b 0.36 b 0.26 b 0.31 b 0.1430 0.04 0.13
Nonhydrolyzed Chlorogenic 4.03 a 4.09 a 3.72 a 3.30 b 2.45 c 1.32 d 0.344 0.017 0.053
Mean values in the same column followed by different lower-case letters are significantly different (P = 0.05) using the Least Significant Difference (LSD) test. LOD, Limit of Detection; LOQ, Limit of Quantitation; U-LOQ, Under LOQ; nd, not detected (under LOD).
correspond to the deprotonated molecule of p-coumaric acid and the decarboxilated p-coumaric ion, respectively. Based on fragmentation patterns of naturally occurring phenolic compounds in tomato, Moco and others (2006) tentatively identified 3 isomers of chlorogenic acid (3-caffeoylquinic acid, 4-caffeoylquinic, and 5-caffeoylquinic acid), 6 glycosylated forms of caffeic acid (caffeic acid hexose I to VI, 2 glycosilated forms of coumaric acid named coumaric acid hexose I and II, respectively), and 2 glycosilated forms of ferulic acid (ferulic acid hexose I and ferulic acid hexose II). In our work there were tentatively identified 2 isomers of chlorogenic acid, the dicaffeoylquinic acid, 3 glycosilated forms of caffeic acid (caffeoyl hexose I, caffeoyl hexose II, and deoxyhexose hexose caffeic acid), 2 glycosilated forms of p-coumaric acid (pcoumaroyl hexose I and p-coumaroyl hexose II), and 2 forms of ferulic acid (feruloyl hexose I and feruloyl hexose II). It seems to be clear that the different isoforms for a given glycosilated phenolic compound affect the retention time in the HPLC system. This can be due to different molecular conformation among isomers and not necessarily due to changes in molecular polarity. The extraction of nonhydrolized extracts from exocarp and mesocarp made possible the estimation of quantitative changes of the naturally occurring combined-phenolic during ripening, based on their mAU × s values given by the HPLC–DAD system (Figure 2). All the combined phenolic acids showed an increasing tendency during ripening; about 2- to 5-fold in both tissues. When the phenolic compound extracts were nonhydrolyzed, chlorogenic acid was the main naturally occurring phenolic acid observed in tomato fruit as aglycone in both exocarp and mesocap tissues. Caffeic acid was observed to appear as aglycone at the “turning” stage of ripeness only in exocarp tissue and tended to increase about 4-fold during ripening (Table 2). In mature-green tomatoes, chlorogenic acid content was about 6-fold higher in exocarp than in mesocarp, and decreased during ripening about 3-fold in exocarp and about 2-fold in mesocarp. When the extracts were hydrolyzed, a remaining quantity of chlorogenic acid (about 0.46 mg/100 g fw) was observed in mesocarp at the different ripeness stages, whereas in exocarp it was estimated to be 1.9 mg/100 g fw
at the mature-green stage and decreased about 50% at the end of tomato fruit ripening. In mesocarp, chlorogenic acid content from nonhydrolyzed extract tended to be higher than the hydrolyzed extract, whereas in exocarp, chlorogenic acid content from nonhydrolyzed extract was about 2-fold higher than in the hydrolyzed extract. This result suggests that chlorogenic acid was partially hydrolyzed when the exocarp was subjected to acidic extraction conditions (time of exposure and acid concentration) of the hydrolyzed extracts. Interestingly, such effect was observed mainly at relatively high chlorogenic acid levels, such as those shown in exocarp at mature-green ripeness stage (about 4.03 mg/100 g fw). This decreased chlorogenic acid content is in agreement with the concomitantly increment in caffeic acid during ripening observed mainly in the hydrolyzed extracts from exocarp. Taga and others (1984) reported that chlorogenic acid is readily hydrolyzed to caffeic and quinic acids by some acid treatments. In hydrolyzed extracts, caffeic acid was observed to increase in exocarp during ripening from 0.54 to 1.53 mg/100 g fw, whereas in mesocarp it tended to increase from 0.02 to 0.16 mg/100 g fw. The caffeic acid levels showed in hydrolyzed extracts in both, mesocarp and exocarp, could have been derived from the breakdown of caffeoyl hexose and from the partial hydrolysis of chlorogenic acid, in addition to that derived from the naturally occurring biosynthesis of caffeic acid observed only in exocarp. Minoggio and others (2003) similarly to our results, reported the presence in tomato of chlorogenic acid as the main phenolic acid, followed by caffeic acid. From our results, ferulic acid in mesocarp and p-coumaric acid in exocarp appeared as free phenolic acids only when the extracts were hydrolyzed. More advanced ripeness stages showed higher quantities of both compounds. The approach of the present work consisting in the comparison of hydrolyzed versus nonhydrolyzed extracts made possible to identify the naturally occurring phenolic acids present as aglycones in the exocarp (chlorogenic acid and caffeic acid) and in mesocarp (chlorogenic acid) of tomato Cv Caiman. Furthermore, we noticed that ferulic acid is present naturally as conjugated forms (tentatively as feruloyl hexose I and feruloyl hexose II) but not as free acid. In the case of p-coumaric
Vol. 78, Nr. 12, 2013 r Journal of Food Science C1843
C: Food Chemistry
Mesocarp
Phenolics changes during tomato ripening . . .
C: Food Chemistry
acid, this was identified as free form (Figure 2), but the level cuted the lab analysis, interpreted results and drafted the was too low to be quantified (LOQ = 0.0022 mg/100 g fw). manuscript. However, p-coumaric acid was quantified from the hydrolyzed extract of exocarp, and it is believed to be derived from the References hydrolysis of p-coumaroyl hexose I. Luthria and others (2006) Arima HK, Rodriguez-Amaya DB. 1988. Carotenoid composition and vitamin A value of commercial Brazilian squashes and pumpkin. J Micronutr Anal 4:177–91. reported that caffeic acid was the predominant phenolic acid in Bowels BL, Miller AJ.1994. Caffeic acid activity against Clostridium botulinum spores. J Food tomatoes “Oregon Spring” and “Red Sun” (higher in Oregon Sci 59(4):905–8. Spring than in Red Sun), and chlorogenic acid was not detected Castellucio C, Paganga G, Melikian N, Bolwell GP, Pridham J, Sampson J, Rice-Evans C. 1995. Antioxidant potential of intermediates in phenylpropanoid metabolism in higher plants. FEBS because of its conversion to caffeic acid during the base hydrolysis lett 368:188–92. extraction. p-Coumaric acid and ferulic acid were also identified Chang S, Tan C, Frankel N, Barret DM. 2000. Low-density lipoprotein activity of phenolic compounds and polyphenol oxidase activity in selected clingstone peach cultivars. J Agric by Luthria and others (2006). Rutin was the sole flavonol idenFood Chem 48:147–51. tified in our work. This compound was observed in mesocarp deSotillo DR, Hadley M, Wolf-Hall C. 1998. Potato peel extract a nonmutagenic antioxidant with potential antimicrobial activity. J Food Sci 63(5):907–10. and exocarp, showing in both tissues increments during ripening Farah A, Donangelo CM. 2006. Phenolic compounds in coffee. Braz J Plant Physiol 18:23–36. and a level in exocarp 18-fold higher than in mesocarp at the red Giovannucci E. 1999. Tomatoes, tomato-based products, lycopene, and cancer: review of the epidemiological literature. J Natl Cancer Inst 91:317–31. ripeness stage. Stewart and others (2000) reported rutin (quercetin Jassim SAA, Naji MA. 2003. Novel antiviral agents: a medicinal plant perspective. J Appl Microbiol 95(3):412–27. 3-rhamnosylglucoside) as the main flavonol in tomato and it was Kacjan-Marˇsi´c N, Gaˇsperlin L, Abram V, Budiˇc M, Vidrih R. 2011. Quality parameters and found mainly in exocarp (98% of flavonols detected were found in total phenolic content in tomato fruits regarding cultivar and microclimatic conditions. Turk the exocarp). Minoggio and others (2003) also reported the presJ Agric For 35:185–94. Marchand L, Murphy SP, Hankin JH, Wilkens LR, Kolonel LN. 2000. Intake of flavonoids ence in tomato fruit of ferulic acid, rutin, naringenin, naringenin Le and lung cancer. J Natl Cancer Inst 92(2):154–60. chalcone, and a chlorogenic acid derivative. In our results, neither Luthria DL, Mukhopadhyay S, Krizek DT. 2006. Content of total phenolic and phenolic acids in tomato (Lycopersicumesculentum Mill.) fruits as influenced by cultivar and solar UV radiation. naringenin nor naringenin chalcone were identified in mesocarp J Food Comp Anal 19:771–7. or in exocarp. This could be due to differences among cultivars. Manach C, Scalbert A, Morand C, R´em´esy C, Jim´enez L. 2004. Polyphenols: food sources and bioavailability. Am J Clin Nutr 79:727–47. It has been claimed that chlorogenic and caffeic acids have anMinoggio M, Bramati L, Simonetti P, Gardana C, lemoli L, Santangelo E, Mauri PL, Spigno P, tiviral (Jassim and Naji 2003), antibacterial (de Sotillo and others Soressi GP, Pietta PG. 2003. Polyphenol pattern and antioxidant activity of different tomato lines and cultivars. Ann Nutr Metab 47:64–9. 1998), and antifungal (Bowels and Miller 1994) effects. FurtherMoco S, Bino RJ, Vorst O, Verhoeven HA, de Groot J, van Beek TA, Vervoort J, Ric de more, caffeic, chlorogenic, ferulic, and p-coumaric acids have been Voss CH. 2006. A liquid chromatography-mass spectrometry-based metabolome database for tomato. Plant Physiol 141:1205–18. reported to be effective in enhancing the resistance of low-density Robards K, Antolovich M. 1997. Analytical chemistry of fruits bioflavonoids: a review. Analyst lipoproteins to oxidation (Castellucio and others 1995). 122(2):11R–34R.
Conclusion Phenolic compounds were qualitatively and quantitatively different during ripening and between exocarp and mesocarp. Fruit exocarp showed more than twice higher quantities of TSP compared to mesocarp. During ripening, TSP increased in both tissues: exocarp and mesocarp, mainly in exocarp. In nonhydrolyzed extracts, whereas rutin increased mainly in exocarp, chlorogenic acid decreased in both tissues and caffeic acid appeared at the “Turning” stage of ripeness only in exocarp.
Acknowledgments The authors thank Dr. Ernesto Aguilar-Palazuelos for statistical assistance.
Author Contributions Elhadi M. Yahia designed the study and helped in the interpretation of the results. Armando Carrillo-L´opez exe-
C1844 Journal of Food Science r Vol. 78, Nr. 12, 2013
Robbins R. 2003. Phenolic acids in foods: an overview of analytical methodology. J Agric Food Chem 51(10):2866–87. S´anchez-Rabaneda F, J´auregui O, Casals I, Lacueva A, Izquierdo-Pulido M. 2003. Liquid chromatographic/electrospray ionization tandem mass spectrometry study of the phenolic composition of cocoa (Theobroma cacao). J Mass Spectrom 38:35–42. Silva BM, Andrade PB, Ferreres F, Dominigues AL, Seabra RM, Ferreira MA. 2002. Phenolic profile of quince fruit (Cycloniaoblomga Miller) (pulp and peel). J Agric Food Chem 50:4615– 8. Slimestad R, Verheul M. 2009. Review of flavonoids and other phenolics from fruits of different tomato (LycopersiconesculentumMill.) cultivars. J Sci Food Agric 89:1255–70. Strack D. 1997. Phenolic metabolism. In: Dey PM & Harborne JB, editors. Plant biochemistry. London, UK: Academic Press. Stewart AJ, Bozonnet S, Mullen W, Jenkins GI, Lean MEJ, Crozier A. 2000. Occurrence of flavonoids in Tomatoes and Tomato-Based Products. J Agric Food Chem 48:2663–9. Sun J, Liang F, Bin Y, Li P, Duan C. 2007. Screening non-colored phenolics in red wines using liquid chromatography/ultraviolet and mass spectrometry/mass spectrometry libraries. Molecules 12:679–93. Taga MS, Miller EE, Pratt DE. 1984. Chia seeds as a source of natural lipid antioxidants. JAOCS 61:928–31. Temple NJ. 2000. Antioxidants and disease: more questions than answers. Nutr Res 20:449–59. Tom´as-Barber´an FA, Gil MI, Cremin P, Waterhouse AL, Hess-Pierce B, Kader AA. 2001. HPLC-DAD-ESI-MS analysis of phenolic compounds in nectarines, peaches and plums. J Agric Food Chem 49:4748–60. Willett WC. 1994. Diet and health: what should we eat? Science 264:532–7.
Identification of phenolic compounds was done by means of liquid chromatography (HPLC) coupled to mass spectrometry (MS) using the electrospray ionization interface (ESI). Quantification of phenolic compounds was carried out by using HPLC with diode array detector (DAD) in exocarp and mesocarp of tomato fruit at 6 different ripeness stages (mature-green, breakers, turning, pink, light-red, and red). Several phenolic compounds were identified including chlorogenic acid, caffeic acid, p-coumaric acid, ferulic acid, and rutin and some combined phenolic acids were tentatively identified, mainly glycosides, such as caffeoyl hexose I, caffeoyl hexose II, caffeoylquinic acid isomer, dicaffeoylquinic acid, p-coumaroyl hexose I, p-coumaroyl hexose II, feruloyl hexose I, feruloyl hexose II, siringyl hexose, and caffeoyl deoxyhexose hexose. Fruit exocarp had higher quantities of total soluble phenolics (TSP) compared to mesocarp. During ripening, TSP increased in both exocarp and mesocarp, mainly in exocarp. While rutin increased, chlorogenic acid decreased in both tissues: exocarp and mesocarp.
Abstract:
Keywords: caffeic acid, chlorogenic acid, Licopersicon esculentum, rutin, TOF-MS
Since exocarp showed higher TSP content than mesocarp, it is recommended to food processors that tomato exocarp should not be discarded but processed along with tomato mesocarp.
Practical Application:
Introduction It has been suggested that the consumption of fruit and vegetables is able to reduce the risk of some chronic diseases, such as cancer and cardiovascular diseases (Willett 1994; Temple 2000). For example, the consumption of fresh tomatoes and tomato products has been inversely related to the development of some types of cancer (Giovannucci 1999). Phenolic compounds have been suggested to play an important role in human health due to the properties related to their antioxidant capacity, by which they may provide antioxidant protection against oxidative processes and achieve the prevention of some diseases (Le Marchand and others 2000; Manach and others 2004). Phenolic compounds are important components in many fruits and vegetables. They are water-soluble substances that tend to accumulate in the dermal tissues of plant body because of their potential role in protection against ultraviolet radiation (Strack 1997), and have been related to growth, development, and defense activities of the plant against pathogens, parasites, and predators, in addition of imparting color to some fruits and vegetables (Robards and Antolovich 1997). In tomato, Minoggio and others (2003) reported that the main phenolic acids were chlorogenic and caffeic acid, among others such as ferulic acid, rutin, and a chlorogenic acid derivative that were also present. Slimestad and Verheulm (2009) also reported that chlorogenic acids and related compounds are the main phe-
nolic compounds besides flavonoids in tomatoes. Concerning to chlorogenic acid health properties, Farah and Donangelo (2006) reported that this phenolic compound has a number of beneficial health properties related to their potent antioxidant activity as well as hepatoprotective, hypoglycemic, and antiviral activity. Analyses of phenolic compounds usually start with acid or alkali hydrolysis to release phenolic compounds from their glycosides and ester conjugates. However, in plants the free forms of phenolic acids are rarely present, whereas the natural occurrence of phenolic compounds is mainly as conjugated forms (Robbins 2003). The exocarp of several fruit and vegetables has been shown to be richer in nutrients/phytochemicals (including phenolic compounds) compared to the mesocarp (Arima and Rodriguez-Amaya 1988; Chang and others 2000; Tom´as-Barber´an and others 2001; Silva and others 2002). This work investigated the qualitative and quantitative changes of phenolic compounds in mesocarp and exocarp of “Caiman” tomato fruit during ripening, looking for differences between peel and pulp, among fruit ripeness stages and between hydrolyzed and nonhydrolyzed extracts since the reports of phenolic composition are mostly concerned with the free phenolic acids (those derived from the hydrolyzed extracts plus the naturally occurring free phenolic acids) rather than conjugated phenolics (those derived from the nonhydrolyzed extracts as naturally occurring conjugated compounds)
Materials and Methods MS 20130317 Submitted 3/6/2013, Accepted 9/19/2013. Author Armando Carrillo-L´opez is with Maestr´ıa en Ciencia y Tecnolog´ıa de Alimentos, Facultad de Sources of standards and solvents Ciencias Qu´ımico-Biol´ogicas, Universidad Aut´onoma de Sinaloa, Culiac´an, Sinaloa, Standards of p-coumaric acid, chlorogenic acid, caffeic acid, 80000, M´exico. Author Elhadi M. Yahia is with Facultad de Ciencias Naturales, Universidad Aut´onoma de Quer´etaro, Juriquilla, 76230, Quer´etaro, M´exico. Direct and ferulic acid were purchased from Sigma-Aldrich (St. Louis, Mo., U.S.A.). Methanol (HPLC grade), acetonitrile (HPLC grade) inquiries to author Carrillo-L´opez (E-mail: [email protected]).
and formic acid (99% purity) were obtained from JT baker R C 2013 Institute of Food Technologists
doi: 10.1111/1750-3841.12295 Further reproduction without permission is prohibited
Vol. 78, Nr. 12, 2013 r Journal of Food Science C1839
C: Food Chemistry
HPLC–DAD–ESI–MS Analysis of Phenolic Compounds During Ripening in Exocarp and Mesocarp of Tomato Fruit
Phenolics changes during tomato ripening . . .
Figure 1–Total soluble phenolics content in exocarp and mesocarp of “Caiman” tomato at different stages of ripeness. MG, mature green; B, breakers; T, turning; P, pink: LR, light-red; R, red.
40
Total soluble phenolics, mg GAE/100 g fw
C: Food Chemistry
(Baker Mallinckrodt, M´exico). HPLC-grade water was used for Ciocalteu reagent (dilution 1:10) and 120 μL of 7.5% of Na2 CO3 identification analyses. All other chemical reagents used were of were added. The plates were incubated for 2 h in darkness, and absorbance was measured at 630 nm using a Dynex MRX microplate analytical grade. reader (Dynex Technology Chantilly, Va., U.S.A.). Results were expressed as milligrams of gallic acid equivalents (GAE)/100 g Plant Material and Sampling Tomato (Lycopersicon esculentum Cv Caiman) was grown in fresh weight (fw). greenhouses from Ac´ambaro, Guanajuato, M´exico. It was harvested at its mature-green stage of ripeness and transported to the Phytochemicals & Nutrition Laboratory at the Faculty of Natu- HPLC–DAD–ESI–MS analytical conditions ral Sciences of the Autonomous Univ. in Queretaro. Fruits were Samples (50 μL) from the different extracts were injected into ripened in air at 25 ◦ C, 80% to 85% RH, and sampled at 6 dif- a HP 1100 series HPLC–DAD system (Hewlett-Packard GmbH, ferent ripeness stages (mature green, breakers, turning, pink, light Waldbronn, Germany) coupled to a 6210 time of flight (TOF) red, and red) for evaluation. The ripeness assessment was done mass spectrometer (Agilent, Palo Alto, Calif., U.S.A.). The pheusing the color classification requirements in tomatoes from the nolic compounds were separated using a 5-μm X-Terra (4.6 × 250 United States Standards for Grades of Fresh Tomatoes. Exocarp mm) column (Waters Co., Milford, Mass., U.S.A.) operating at 25 (thickness ∼ 1 mm) was separated from mesocarp using a potato ◦ C. The mobile phase was composed of 1% formic acid in water peeler. Exocarp was freeze-dried, and both exocarp and mesocarp (A) and acetonitrile (B) following a linear gradient from 100% A were kept at –80 ◦ C until analysis. Seeds were discarded. and 0% B to 0% A and 100% B in 60 min using a flow rate of 0.5 mL/min. The 1% formic acid in water was to promote molecular Phenolic compounds extraction ionization. Mass spectrometer system was equipped with an elecFive grams of fresh mesocarp or 0.4 g of freeze-dried exocarp trospray interface operating in the negative ionization mode. A were each mixed with 10 mL of extraction solution (80% methanol mass hunter manager software (A.02.01) was used. Nitrogen was in water, containing 1% formic acid) and thoroughly ground in utilized as drying gas at 350 ◦ C, at a flow rate of 11.5 L/min and mortar and pestle. The homogenate was sonicated during 15 min at a pressure of 45 . Fragmentor and capillary conditions were at room temperature and filtered through Whatman 3 filter paper. 100 and 4000 V, respectively. This was named the nonhydrolyzed extract. Two milliliters of this extract were mixed with 2 mL of 2.4 N HCl for hydrolysis in water bath at 80 ◦ C for 2 h and then the test tubes were cooled Statistical analysis with water at room temperature. This latter extract was named the Data are means of 3 determinations. Analysis of variance hydrolyzed extract. Both extracts were filtered through 0.45-μm (ANOVA) was conducted and the significance of difference beMillipore membrane (Millipore Corp., Bedford, Mass., U.S.A.) tween means was determined by Tukey method at a 5% signifibefore injection to the HPLC–DAD–ESI–MS system. cance level. Limit of detection (LOD) and limit of quantitation (LOQ) were obtained from the calibration curves for chlorogenic, Quantification of TSP caffeic, ferulic, and p-coumaric acids, using the slope value (S) Aliquots from exocarp or mesocarp nonhydrolyzed extracts of the respective regression line and the standard deviation of the were diluted 1:10 with HPLC-grade water and 30 μL of diluted y-intercepts (σ ). The used formulas were LOD = 3.3σ /S and sample per hole was placed in 96-hole plates, and 150 μL of Folin– LOQ = 10σ /S.
Exocarp Mesocarp 30
20
10
0 MG
B
T
P
Tomato ripeness stage
C1840 Journal of Food Science r Vol. 78, Nr. 12, 2013
LR
R
Phenolics changes during tomato ripening . . . Table 1–Fragment-ion identification of nonhydrolyzed extracts from exocarp and mesocarp of tomato.
2 (17.1)
3 (17.3)
4 (17.9) 5 (18.2) 6 (18.5)
7 (18.8)
8 (19.2) Chlo (19.5)
10 (20.8) 11 (21.6)
Caf (22.7) 14 (23.4) Rut (24.2) Cou (25.6)
Fragments of negative ions, (m/z) ]−
325 [326-H 163 [326-H-162] − 119 [326-H-162-44] − 341 [342-H] − 179 [341-H-162] − 135 [341-H-162-44] − 515 [516-H ] − 353 [516-H-162] − 355 [356-H] − 193 [356-H-162] − 341 [342-H] − 179 [342-H-162] − 135 [342-H-162-44] − 353 [354-H] − 173 [354-H-180] − 179 [354-H-174] − 191 [354-H-162 ] − 325 [326-H ] − 163 [326-H-162] − 353 [354-H] − 173 [354-H-162-18] − 179 [354-H-174] − 191 [354-H-162] − 355 [356-H] − 193 [356-H-162] − 359 [359-H] − 197 [359-H-162] − 153 [359-H-162-44] − 179 [360-H-179] − 135 [180-H-44] − 487 [488-H] − 179 [488-H-162-146] − 609 [610-H] − 163 [164-H] − 119 [164-H-44] −
Tentative fragment ion identification
Tentative compound
p-Coumaroyl hexose p-Coumaroyl ion Decarboxilated p-coumaroyl Caffeoyl-hexose Caffeoyl-ion Decarboxilated caffeoyl Dicaffeoylquinic acid ion CQ acid Feruloyl hexose Feruloyl ion Caffeoyl hexose Caffeoyl ion Decarboxilated caffeoyl CQ CQ ion minus dehydrated quinic moiety Caffeic acid ion CQ ion minus caffeoyl moiety p-Coumaroyl hexose p-Coumaroyl ion
p-Coumaroyl hexose I
Chlo acid ion minus dehydrated quinic acid Caffeic acid ion Chlo ion minus caffeoyl moiety Feruloyl hexose Feruloyl ion Syringyl hexose Syringic ion Decarboxylated siringic ion Decarboxylated caffeoyl ion Deoxyhexose hexose caffeoyl Caffeoyl ion
Caffeoyl hexose I
Dicaffeoylquinic acid Feruloyl hexose I p-Caffeoyl hexose II
CQ isomer
p-Coumaroyl hexose II
Feruloyl hexose II Siringyl hexose
Caffeoyl deoxyhexose hexose
Coumaroyl ion Decarboxilated coumaroyl
Peaks 1, 9, 12, 13, 16, 17, 18, 19, 20 were not identified. Tr, retention time; Chlo, Chlorogenic acid; Caf, Caffeic acid; Rut, Rutin; Cou, p-Coumaric acid; CQ, caffeoylquinic acid.
Results and Discussion Total soluble phenolics TSP were more than twice higher in the exocarp than in the mesocarp during all stages of ripening (Figure 1). TSP increased significantly (P < 0.05) in both tissues during ripening, more pronouncedly in exocarp. The TSP levels observed in the mesocarp (12.7 to 16.8 mg/100 g fw) were similar to those observed by Minoggio and others (2003) in tomato, but higher than those observed by Kacjan-Marˇsi´c and others (2011) for 11 cultivars from which the cultivar cherry showed the highest content (10.39 mg/100 g fw). The higher TSP content in the exocarp is in agreement with the fact that phenolic compounds tend to accumulate mainly in the dermal tissues of plant body because of their potential role in protection against ultraviolet radiation (Strack 1997), in addition to the defense activities of the plant against pathogens, parasites, and predators (Robards and Antolovich 1997). Identification and quantification of phenolic compounds in the exocarp and mesocarp HPLC–DAD–ESI–MS detection operating in the negative ion mode was used to identify phenolic compounds in exocarp and mesocarp of tomato fruit at 6 different ripeness stages. The compounds were identified as aglycones obtained after acid hydrolysis (hydrolyzed extracts) and as naturally occurring aglycones or combined phenolic compounds obtained in the nonhydrolyzed extracts. Identification of aglycones was based on retention times, and
UV and mass spectra by comparison with commercial standards, and a tentatively identification of naturally occurring combined phenolic compounds was based on their fragmentation patterns of the mass spectra. The quantification of the identified compounds was done by HPLC–DAD comparing the UV-peaks areas with calibration curves produced with commercially obtained standards. A total of 15 phenolic compounds were identified or tentatively identified, taking into account exocarp, mesocarp and if the extracts were hydrolyzed or nonhydrolyzed.
Naturally occurring aglycones From the nonhydrolyzed extracts, 3 naturally occurring aglycones of phenolic acids were identified; chlorogenic acid in both exocarp and mesocarp, caffeic acid only in exocarp, and p-coumaric acid only in mesocarp.
Naturally occurring combined phenolic acids Ten naturally occurring combined phenolic acids were tentatively identified in mesocarp and/or exocarp (Table 1, Figure 2). p-Coumaroyl hexose I, caffeoyl hexose II, p-coumaroyl hexose II, caffeoyl deoxyhexose hexose, and rutin were shown in both, mesocarp and exocarp, whereas caffeoyl hexose I, feruloyl hexose I, feruloyl hexose II, and syringyl hexose were observed only in mesocarp and dicaffeoylquinic acid only in exocarp). Vol. 78, Nr. 12, 2013 r Journal of Food Science C1841
C: Food Chemistry
Peak(Tr, min)
Phenolics changes during tomato ripening . . .
C: Food Chemistry
[354-H-173]− corresponds to the caffeoyl ion derived from the loss of the dehydrated quinic moiety from the deprotonated caffeoylquinic acid, and 191 [354-H-162]− corresponds to the ion derived from the loss of the dehydrated caffeoyl moiety. Similar results have been reported by Sun and others (2007) and by S´anchez-Rabaneda and others (2003). Caffeic acid showed its [MH]− deprotonated molecule (m/z 179) and the [M-H-44]− fragment ion (m/z 135) corresponding to the fragment derived from the loss of the carbon dioxide moiety of caffeic acid (Table 1). The identification of the naturally occurring glycoside-combined phenolic compounds (Figure 2, Table 1) was based on the fragmentation pattern derived from the mass spectra obtained from the different chromatographic peaks observed (Figure 2). Peaks 2 and 8 presented different retention times, 17.1 and 19.2 min, but similar fragmentation patterns whose m/z of 325 ([326-H]− ), 163 [326-H-162]− , and 119 ([164-H-44]− ) correspond to the deprotonated molecule of p-coumaroyl hexose, the p-coumaroyl ion derived from the deglycosilation of the molecule of coumaroyl hexose, and the decarboxilated coumaroyl ion, respectively. Thus, peaks 2 and 8 were tentatively identified as isomers of glycosilated molecule of p-coumaroyl and were assigned as p-coumaroyl hexose I and p-coumaroyl hexose II, respectively. Peaks 3 and 6 also presented different retention times, 17.3 and 18.5 min, but the same fragmentation pattern whose m/z of 341 ([342-H]− ), 179 ([342-H-162]− ), and 135 ([179-H-44]− ) correspond to the deprotonated molecule of caffeoyl hexose, the deglycosilated molecule of caffeoyl hexose previously dehydrated (caffeoyl ion), and the decarboxilated caffeoyl ion, respectively. Thus, peaks 3 and 6 were tentatively identified as glycosilated isomers of caffeic acid and was assigned as caffeoyl hexose I and caffeoyl hexose II, respectively. Peaks 4, 7, and Chlo, presented the retention times 17.9, 18.8, and 19.5 min, respectively, and showed a fragmentation pattern that exhibited in common a molecular ion m/z 353 corresponding to the caffeoylquinic acid ion. Peak 4 showed a fragmentation pattern whose m/z 515 ([516-H]− ) and 353 ([516-H-162]− ) is believed to correspond to the deprotonated molecule of dicaffeoylquinic acid, in agreement with Moco and others (2006). Both, peak 7 and peak chlo presented the same fragmentation pattern that corresponds to the caffeolquinic acid molecule. Thus, peak 7 corresponds to an isomerization form of the chlorogenic acid, whereas peak chlo was identified as chlorogenic acid (3caffeoylquinic acid) based on the retention time and mass spectra from the standard. Peak 10 presented a retention time of 20.8 min, and a fragmentation pattern whose m/z of 355 ([356-H]− ) and 193 ([356-H-162]− ) correspond to the deprotonated molecule of feruloyl hexose and the feruloyl ion (deglycosilated molecule of feruloyl hexose), respectively. Peak 11 presented a retention time of 21.6 min, and a fragmentation pattern whose m/z of 359 ([360-H]− ), 197 ([360-H-162]− ), and 153 ([198-H-44]− ), correspond to the deprotonated molecule of syringyl hexose, the syringyl ion (the deglycosilated molecule of siringyl hexose), and the decarboxilated siringyl ion, respectively. Peak 14 presented a retention time of 23.4 min, and a fragmentation pattern whose m/z of 487 ([488-H]− ) and 179 ([488-H-162–146]− ) correspond to the deprotonated molecule of deoxyhexose hexose caffeic acid and the caffeic ion derived from the loss of the deoxyhexose hexose moiety, respectively. Peak Rut presented a retention time of 24.2 min, and a fragmentation pattern whose m/z of 609 ([610H]− ) correspond to the deprotonated molecule of rutin. Peak Figure 2–HPLC–DAD chromatograms for phenolic compounds at 320 nm Cou with a retention time of 25.6 min presented a fragmentation of nonhydrolyzed extracts from exocarp and mesocarp of “Caiman” tomato − − fruit. A, ripe mesocarp; B, mature-green mesocarp; C, ripe exocarp; D, pattern whose m/z of 163 ([164-H] ) and 119 ([164-H-44] )
Aglycones derived from acid hydrolysis Four aglycones were identified in the hydrolyzed extracts (chlorogenic acid and caffeic acid in both mesocarp and exocarp, ferulic acid only in mesocarp and p-coumaric acid only in exocarp). The naturally occurring aglycone phenolic acids were identified in agreement with the retention times observed for their respective commercial standards being 19.5 min for chlorogenic acid, 22.7 min for caffeic acid, and 25.6 min for p-coumaric acid, respectively. Furthermore, the identities were corroborated with their respective fragmentation patterns derived from their mass spectra (Table 1). Chlorogenic acid showed its profile of m/z 353, 173, 179, and 191, where 353 ([354-H]− ) corresponds to the [MH]− deprotonated molecule of 3-caffeoylquinic acid, 173 [354-H180]− corresponds to the ion derived from the loss of the caffeoyl moiety from the deprotonated caffeoylquinic acid molecule, 179
mature-green exocarp.
C1842 Journal of Food Science r Vol. 78, Nr. 12, 2013
Phenolics changes during tomato ripening . . . Table 2–Phenolic acids content (mg/100 g fw) of nonhydrolyzed and hydrolyzed extracts from mesocarp and exocarp of “Caiman” tomato fruit.
Ripeness stage Mature green Breakers Turning Pink Light red Red LSD Exocarp Ripeness stage Mature green Breakers Turning Pink Light red Red LSD LOD LOQ
Nonhydrolyzed
Hydrolyzed
Chlorogenic
Caffeic
Chlorogenic
Caffeic
Ferulic
0.68 ab 0.90 a 0.67 ab 0.51 b 0.56 ab 0.40 b 0.344
nd nd nd nd nd nd –
0.46 a 0.57 a 0.49 a 0.44 a 0.44 a 0.41 a 0.344
0.02 a 0.07 a 0.10 a 0.08 a 0.08 a 0.16 a 0.207 Hydrolyzed
U-LOQ 0.14 a 0.21 b 0.20 b 0.23 b 0.30 b 0.10
Caffeic U-LOQ U-LOQ 0.06 a 0.10 a 0.12 a 0.25 a 0.207 0.004 0.012
Chlorogenic 1.90 a 1.80 a 1.80 a 1.80 a 1.40 b 0.97 c 0.344 0.017 0.053
Caffeic 0.54 a 0.60 a 1.43 b 1.61 b 1.34 c 1.53 c 0.207 0.004 0.012
p-Coumaric 0.04 a 0.09 a 0.39 b 0.36 b 0.26 b 0.31 b 0.1430 0.04 0.13
Nonhydrolyzed Chlorogenic 4.03 a 4.09 a 3.72 a 3.30 b 2.45 c 1.32 d 0.344 0.017 0.053
Mean values in the same column followed by different lower-case letters are significantly different (P = 0.05) using the Least Significant Difference (LSD) test. LOD, Limit of Detection; LOQ, Limit of Quantitation; U-LOQ, Under LOQ; nd, not detected (under LOD).
correspond to the deprotonated molecule of p-coumaric acid and the decarboxilated p-coumaric ion, respectively. Based on fragmentation patterns of naturally occurring phenolic compounds in tomato, Moco and others (2006) tentatively identified 3 isomers of chlorogenic acid (3-caffeoylquinic acid, 4-caffeoylquinic, and 5-caffeoylquinic acid), 6 glycosylated forms of caffeic acid (caffeic acid hexose I to VI, 2 glycosilated forms of coumaric acid named coumaric acid hexose I and II, respectively), and 2 glycosilated forms of ferulic acid (ferulic acid hexose I and ferulic acid hexose II). In our work there were tentatively identified 2 isomers of chlorogenic acid, the dicaffeoylquinic acid, 3 glycosilated forms of caffeic acid (caffeoyl hexose I, caffeoyl hexose II, and deoxyhexose hexose caffeic acid), 2 glycosilated forms of p-coumaric acid (pcoumaroyl hexose I and p-coumaroyl hexose II), and 2 forms of ferulic acid (feruloyl hexose I and feruloyl hexose II). It seems to be clear that the different isoforms for a given glycosilated phenolic compound affect the retention time in the HPLC system. This can be due to different molecular conformation among isomers and not necessarily due to changes in molecular polarity. The extraction of nonhydrolized extracts from exocarp and mesocarp made possible the estimation of quantitative changes of the naturally occurring combined-phenolic during ripening, based on their mAU × s values given by the HPLC–DAD system (Figure 2). All the combined phenolic acids showed an increasing tendency during ripening; about 2- to 5-fold in both tissues. When the phenolic compound extracts were nonhydrolyzed, chlorogenic acid was the main naturally occurring phenolic acid observed in tomato fruit as aglycone in both exocarp and mesocap tissues. Caffeic acid was observed to appear as aglycone at the “turning” stage of ripeness only in exocarp tissue and tended to increase about 4-fold during ripening (Table 2). In mature-green tomatoes, chlorogenic acid content was about 6-fold higher in exocarp than in mesocarp, and decreased during ripening about 3-fold in exocarp and about 2-fold in mesocarp. When the extracts were hydrolyzed, a remaining quantity of chlorogenic acid (about 0.46 mg/100 g fw) was observed in mesocarp at the different ripeness stages, whereas in exocarp it was estimated to be 1.9 mg/100 g fw
at the mature-green stage and decreased about 50% at the end of tomato fruit ripening. In mesocarp, chlorogenic acid content from nonhydrolyzed extract tended to be higher than the hydrolyzed extract, whereas in exocarp, chlorogenic acid content from nonhydrolyzed extract was about 2-fold higher than in the hydrolyzed extract. This result suggests that chlorogenic acid was partially hydrolyzed when the exocarp was subjected to acidic extraction conditions (time of exposure and acid concentration) of the hydrolyzed extracts. Interestingly, such effect was observed mainly at relatively high chlorogenic acid levels, such as those shown in exocarp at mature-green ripeness stage (about 4.03 mg/100 g fw). This decreased chlorogenic acid content is in agreement with the concomitantly increment in caffeic acid during ripening observed mainly in the hydrolyzed extracts from exocarp. Taga and others (1984) reported that chlorogenic acid is readily hydrolyzed to caffeic and quinic acids by some acid treatments. In hydrolyzed extracts, caffeic acid was observed to increase in exocarp during ripening from 0.54 to 1.53 mg/100 g fw, whereas in mesocarp it tended to increase from 0.02 to 0.16 mg/100 g fw. The caffeic acid levels showed in hydrolyzed extracts in both, mesocarp and exocarp, could have been derived from the breakdown of caffeoyl hexose and from the partial hydrolysis of chlorogenic acid, in addition to that derived from the naturally occurring biosynthesis of caffeic acid observed only in exocarp. Minoggio and others (2003) similarly to our results, reported the presence in tomato of chlorogenic acid as the main phenolic acid, followed by caffeic acid. From our results, ferulic acid in mesocarp and p-coumaric acid in exocarp appeared as free phenolic acids only when the extracts were hydrolyzed. More advanced ripeness stages showed higher quantities of both compounds. The approach of the present work consisting in the comparison of hydrolyzed versus nonhydrolyzed extracts made possible to identify the naturally occurring phenolic acids present as aglycones in the exocarp (chlorogenic acid and caffeic acid) and in mesocarp (chlorogenic acid) of tomato Cv Caiman. Furthermore, we noticed that ferulic acid is present naturally as conjugated forms (tentatively as feruloyl hexose I and feruloyl hexose II) but not as free acid. In the case of p-coumaric
Vol. 78, Nr. 12, 2013 r Journal of Food Science C1843
C: Food Chemistry
Mesocarp
Phenolics changes during tomato ripening . . .
C: Food Chemistry
acid, this was identified as free form (Figure 2), but the level cuted the lab analysis, interpreted results and drafted the was too low to be quantified (LOQ = 0.0022 mg/100 g fw). manuscript. However, p-coumaric acid was quantified from the hydrolyzed extract of exocarp, and it is believed to be derived from the References hydrolysis of p-coumaroyl hexose I. Luthria and others (2006) Arima HK, Rodriguez-Amaya DB. 1988. Carotenoid composition and vitamin A value of commercial Brazilian squashes and pumpkin. J Micronutr Anal 4:177–91. reported that caffeic acid was the predominant phenolic acid in Bowels BL, Miller AJ.1994. Caffeic acid activity against Clostridium botulinum spores. J Food tomatoes “Oregon Spring” and “Red Sun” (higher in Oregon Sci 59(4):905–8. Spring than in Red Sun), and chlorogenic acid was not detected Castellucio C, Paganga G, Melikian N, Bolwell GP, Pridham J, Sampson J, Rice-Evans C. 1995. Antioxidant potential of intermediates in phenylpropanoid metabolism in higher plants. FEBS because of its conversion to caffeic acid during the base hydrolysis lett 368:188–92. extraction. p-Coumaric acid and ferulic acid were also identified Chang S, Tan C, Frankel N, Barret DM. 2000. Low-density lipoprotein activity of phenolic compounds and polyphenol oxidase activity in selected clingstone peach cultivars. J Agric by Luthria and others (2006). Rutin was the sole flavonol idenFood Chem 48:147–51. tified in our work. This compound was observed in mesocarp deSotillo DR, Hadley M, Wolf-Hall C. 1998. Potato peel extract a nonmutagenic antioxidant with potential antimicrobial activity. J Food Sci 63(5):907–10. and exocarp, showing in both tissues increments during ripening Farah A, Donangelo CM. 2006. Phenolic compounds in coffee. Braz J Plant Physiol 18:23–36. and a level in exocarp 18-fold higher than in mesocarp at the red Giovannucci E. 1999. Tomatoes, tomato-based products, lycopene, and cancer: review of the epidemiological literature. J Natl Cancer Inst 91:317–31. ripeness stage. Stewart and others (2000) reported rutin (quercetin Jassim SAA, Naji MA. 2003. Novel antiviral agents: a medicinal plant perspective. J Appl Microbiol 95(3):412–27. 3-rhamnosylglucoside) as the main flavonol in tomato and it was Kacjan-Marˇsi´c N, Gaˇsperlin L, Abram V, Budiˇc M, Vidrih R. 2011. Quality parameters and found mainly in exocarp (98% of flavonols detected were found in total phenolic content in tomato fruits regarding cultivar and microclimatic conditions. Turk the exocarp). Minoggio and others (2003) also reported the presJ Agric For 35:185–94. Marchand L, Murphy SP, Hankin JH, Wilkens LR, Kolonel LN. 2000. Intake of flavonoids ence in tomato fruit of ferulic acid, rutin, naringenin, naringenin Le and lung cancer. J Natl Cancer Inst 92(2):154–60. chalcone, and a chlorogenic acid derivative. In our results, neither Luthria DL, Mukhopadhyay S, Krizek DT. 2006. Content of total phenolic and phenolic acids in tomato (Lycopersicumesculentum Mill.) fruits as influenced by cultivar and solar UV radiation. naringenin nor naringenin chalcone were identified in mesocarp J Food Comp Anal 19:771–7. or in exocarp. This could be due to differences among cultivars. Manach C, Scalbert A, Morand C, R´em´esy C, Jim´enez L. 2004. Polyphenols: food sources and bioavailability. Am J Clin Nutr 79:727–47. It has been claimed that chlorogenic and caffeic acids have anMinoggio M, Bramati L, Simonetti P, Gardana C, lemoli L, Santangelo E, Mauri PL, Spigno P, tiviral (Jassim and Naji 2003), antibacterial (de Sotillo and others Soressi GP, Pietta PG. 2003. Polyphenol pattern and antioxidant activity of different tomato lines and cultivars. Ann Nutr Metab 47:64–9. 1998), and antifungal (Bowels and Miller 1994) effects. FurtherMoco S, Bino RJ, Vorst O, Verhoeven HA, de Groot J, van Beek TA, Vervoort J, Ric de more, caffeic, chlorogenic, ferulic, and p-coumaric acids have been Voss CH. 2006. A liquid chromatography-mass spectrometry-based metabolome database for tomato. Plant Physiol 141:1205–18. reported to be effective in enhancing the resistance of low-density Robards K, Antolovich M. 1997. Analytical chemistry of fruits bioflavonoids: a review. Analyst lipoproteins to oxidation (Castellucio and others 1995). 122(2):11R–34R.
Conclusion Phenolic compounds were qualitatively and quantitatively different during ripening and between exocarp and mesocarp. Fruit exocarp showed more than twice higher quantities of TSP compared to mesocarp. During ripening, TSP increased in both tissues: exocarp and mesocarp, mainly in exocarp. In nonhydrolyzed extracts, whereas rutin increased mainly in exocarp, chlorogenic acid decreased in both tissues and caffeic acid appeared at the “Turning” stage of ripeness only in exocarp.
Acknowledgments The authors thank Dr. Ernesto Aguilar-Palazuelos for statistical assistance.
Author Contributions Elhadi M. Yahia designed the study and helped in the interpretation of the results. Armando Carrillo-L´opez exe-
C1844 Journal of Food Science r Vol. 78, Nr. 12, 2013
Robbins R. 2003. Phenolic acids in foods: an overview of analytical methodology. J Agric Food Chem 51(10):2866–87. S´anchez-Rabaneda F, J´auregui O, Casals I, Lacueva A, Izquierdo-Pulido M. 2003. Liquid chromatographic/electrospray ionization tandem mass spectrometry study of the phenolic composition of cocoa (Theobroma cacao). J Mass Spectrom 38:35–42. Silva BM, Andrade PB, Ferreres F, Dominigues AL, Seabra RM, Ferreira MA. 2002. Phenolic profile of quince fruit (Cycloniaoblomga Miller) (pulp and peel). J Agric Food Chem 50:4615– 8. Slimestad R, Verheul M. 2009. Review of flavonoids and other phenolics from fruits of different tomato (LycopersiconesculentumMill.) cultivars. J Sci Food Agric 89:1255–70. Strack D. 1997. Phenolic metabolism. In: Dey PM & Harborne JB, editors. Plant biochemistry. London, UK: Academic Press. Stewart AJ, Bozonnet S, Mullen W, Jenkins GI, Lean MEJ, Crozier A. 2000. Occurrence of flavonoids in Tomatoes and Tomato-Based Products. J Agric Food Chem 48:2663–9. Sun J, Liang F, Bin Y, Li P, Duan C. 2007. Screening non-colored phenolics in red wines using liquid chromatography/ultraviolet and mass spectrometry/mass spectrometry libraries. Molecules 12:679–93. Taga MS, Miller EE, Pratt DE. 1984. Chia seeds as a source of natural lipid antioxidants. JAOCS 61:928–31. Temple NJ. 2000. Antioxidants and disease: more questions than answers. Nutr Res 20:449–59. Tom´as-Barber´an FA, Gil MI, Cremin P, Waterhouse AL, Hess-Pierce B, Kader AA. 2001. HPLC-DAD-ESI-MS analysis of phenolic compounds in nectarines, peaches and plums. J Agric Food Chem 49:4748–60. Willett WC. 1994. Diet and health: what should we eat? Science 264:532–7.
