Identification and Quantification of Metallo–Chlorophyll Complexes in Bright Green Table Olives by High-Performance Liquid Chromatrography–Mass Spectrometry Quadrupole/Time-of-Flight

Abstract

Five different samples of table olives, two regular Spanish table olives and three “bright green table olives”, have been analyzed by HPLC–MS/MS to determine their pigment profile. Typical pigment profiles of almost all table olives show primarily chlorophyll derivatives lacking metals (e.g., pheophytin a/b and 15²-Me-phytol-chlorin e₆). Bright green table olives have a unique profile including metallo–chlorophyll complexes (Cu-15²-Me-phytol-chlorin e₆ with 26–48% and Cu-pheophytin a with 3–18%) as their major pigments. New tentative structures have been identified by MS such as 15²-Me-phytol-rhodin g₇, 15²-Me-phytol-chlorin e₆, 15²-Me-phytol-isochlorin e₄, Cu-15²-Me-phytol-rhodin g₇, Cu-15²-Me-phytol-chlorin e₆, and Cu-15²-Me-phytol-isochlorin e₄, and new MS/MS fragmentation patterns are reported for Cu-15²-Me-phytol-rhodin g₇, Cu-15²-Me-phytol-chlorin e₆, Cu-pheophytin b, Cu-pheophytin a, Cu-pyropheophytin b, and Cu-pyropheophytin a. The presence of metallo–chlorophyll derivatives is responsible for the intense color of bright green table olives, but these metallo–chlorophyll complexes may be regarded as a “green staining” defect that is unacceptable to consumers.

INTRODUCTION

The color of vegetable products is one of the most important sensory attributes evaluated by consumers worldwide, and, in green fresh vegetables and fruits, the green color is due to the presence of chlorophylls a and b. For table olives, color is a very critical attribute.

The most common color changes of fresh vegetables during cooking, freeze preservation, storage, or heat processing are due to the conversion of chlorophylls (a and b), responsible for bright green color, into their respective pheophytins (a and b) that are responsible for yellow-brown olive color. This dramatic change in color involves the release of the magnesium from the porphyrin ring. However, the insertion of metals within the porphyrin ring, such as zinc (Zn²⁺) or copper (Cu²⁺), produces a regreening effect. These metallo–chlorophyll complexes are more stable, due to higher metal–porphyrin bond energies, and more resistant to acid and heat than are naturally occurring magnesium (Mg²⁺) chlorophyll complexes.¹ In addition, the formation of copper complexes is much more rapid than zinc complexes.^2,3 Regreening via the formation of metallo – chlorophyll complexes has been used to preserve the desirable fresh green color of vegetables by the food industry,^4,5 including several patents.^4–7

In the traditional processing of Spanish style of table olives, chlorophylls a and b are degraded completely to pheophytins and pheophorbides during the fermentation process by two different and coexisting mechanisms. In the initial treatment of the olives with NaOH, at alkaline pH, the chlorophyllase is active, but in the fermentation process, pH turns acidic followed by the loss of Mg²⁺. The pigment transformation brought about by lactic fermentation produces olives that are highly valuable for their characteristic golden color.⁸

The innovations introduced in the traditional system of processing (elimination of the short washing, addition of acids, reuse of brines, etc.) have changed the previously described mechanism of chlorophyll degradation, with the detection of oxidative reactions that affects the chlorophyll isocyclic ring and production of allomerized chlorophylls.⁹

Nevertheless, the Gordal variety, used almost exclusively to produce table olives, develops an alteration in color, known as “green staining” (GS) that is undesirable to consumers.¹⁰ This common defect is observed as bluish-green zones distributed over the skin that contrasts with its natural olive color. This alteration most likely occurs through the formation of endogenous metal porphyrin complexes to create GS.¹¹ The formation of metallo–chlorophyll complexes is thought to arise from the fruit’s endogenous Cu and may involve degradation of cell integrity allowing contact of the metal with the chlorophyll derivatives, which under normal conditions are protected in their natural lipophilic environment. Endogenous metal such as Cu may arise from the pectin chain that can serve as a reservoir of copper that is captured by chlorophyll pigments separated from the thylakoid membrane to form the copper chlorophyll complexes.¹²

The most common technique to analyze these pigments is RP-HPLC-DAD, although HPLC–MS has also been employed for the determination of chlorophylls and their derivatives because it provides structural information, and, in addition, it is a very sensitive and selective technique. Methods based on HPLC–MS include atmospheric pressure ionization (APCI)^13,14 and electrospray ionization (ESI)^15,16 interfaces.

Previously, GS alteration has been studied,^8,11 although those studies did not utilize HPLC–MS and tandem MS with accurate mass as an identification tool. In this study, metallo–chlorophyll complexes responsible for green staining are identified by HPLC mass spectrometry and quantified by HPLC in three commercial samples. In addition, their pigment profiles are compared to typical table olives (yellow-brown olive color).

MATERIALS AND METHODS

Samples

The study was carried out with five different commercial green table olives bought at different markets: sample A, “Italian Castelvetrano Whole Green Olives” (product of Italy) packed by Mezzetta Inc.; sample B, “Olives Castelvetrano Divina” packed by Whole Foods Market; sample C, “Green Table Olives” packed by Vincenzo’s; sample D, “Spanish Olives” (var. Manzanilla); and sample E, “Spanish Queen Olives” (var. Gordal) (both Spanish products) distributed by Kroger Co.

Chemicals and Standards

Ammonium acetate was supplied by Fluka (Zwijndrecht, The Netherlands). HPLC/MS and HPLC reagent grade solvents were purchased from Fisher Scientific Inc. (Pittsburgh, PA). Standards of pheophorbide a, Cu-pheophorbide a, rhodin g₇, chlorin e₆, and Cu-chlorin e₆ were supplied by Porphyrin Product Inc. (Logan, UT). Chlorophylls a (chl a) and b (chl b) were supplied by Sigma-Aldrich Co. (St. Louis, MO). Standards of pheophytin a (phy a) and b (phy b) were obtained from their respective chlorophyll by acidification.¹⁷ Standards of Cu-pheophytin a and b were obtained from their respective chlorophyll by reaction with an excess of copper(II) ions as chloride.¹⁰

Extraction of Pigments

Fifty depitted fruits (ca. 40 g) were ground and homogenized. Triplicate extractions were prepared by accurately weighing 5–15 g for each sample. Pigment extraction was performed with N,N-dimethylformamide (DMF), according to the method proposed by Mínguez-Mosquera and Garrido-Fernádez¹⁸ that allows the selective separation of pigments in oil free pigment solutions. Thus, DMF retains chlorophyll and xanthophylls, while the hexane phase retains the carotene fraction. All of the analytical procedures were performed under red lighting to avoid any photooxidation of the chlorophyllic compounds.

HPLC–MS and HPLC–MS/MS

HPLC–MS analysis was carried out on a Waters 2695 gradient HPLC separation module (Waters Corp., Milford, MA) equipped with an auto injector and a 996-photodiode array detector (PDA). Chromatographic separations were performed on a HALO C18 Symmetry column (75 mm × 4.6 mm i.d., 2.7 µm) (MAC-MOD Analytical, Inc.). The mobile phase consisted of (1/1/8) (v/v/v) water/ammonium acetate 1 M/methanol (solvent A), acetonitrile (solvent B), and reagent alcohol (solvent C) at a flow rate of 1.25 mL/min. Column temperature was kept at 25 °C. Table 1 shows the gradient scheme for eluting the pigments.

Table 1.

Gradient Used for Elution of Pigments by HPLC Reverse Phase^a

time (min)	solvent A	solvent B	solvent C	flow (mL/min)	curve
0.00	75.0	25.0	0.0	1.25	1
8.00	25.0	75.0	0.0	1.25	6
10.00	25.0	0.0	75.0	1.25	6
11.00	19.0	0.0	81.0	1.25	6
13.00	14.0	0.0	86.0	1.25	6
15.00	11.0	0.0	89.0	1.25	6
16.00	0.0	0.0	100.0	1.25	6
18.00	0.0	0.0	100.0	1.25	6
20.00	75.0	25.0	0.0	1.25	1

^a(1/1/8) (v/v/v) water/ammonium acetate 1 M/methanol (solvent A), acetonitrile (solvent B), and reagent alcohol (solvent C).

Mass spectrometric studies were performed on a Quadrupole/ time-of-flight mass spectrometer (QToF Premier, Micromass Limited, Manchester, U.K.) equipped with an atmospheric pressure chemical ionization interface (APCI). Positive ion APCI conditions for chlorophyll analyses were: 30 µA corona current, collision energy 8 eV, cone voltage 35 V, source temperature at 110 °C, 50 L/h of cone gas, and the flow of nitrogen desolvation gas was 400 L/h, at 450 °C. Daughter ions spectra were obtained by APCI negative using various collision energies (8–35 eV). Sodium formate solution was used to calibrate the mass spectrometer from m/z 100–1500 with mass accuracy of <2 ppm. Detection and tentative identification of all of the chlorophyll derivatives were accomplished using in-line electronic absorption spectra recorded every 1.2 nm from 200 to 800 nm with a Waters model 996-PDA (Milford, MA). Quantification of all of the chlorophyll derivatives was based on PDA data from calibration curves with external standards at the respective λ_max of each compound.

Color Measurement

The color analysis has been made with a Minolta colorimeter (CR-300) using CIE L*a*b* scale with illuminants C.

Statistical Analysis

Data were analyzed for differences between means using one-way analysis of variance (ANOVA). The Brown and Forsythe test¹⁹ was used as a post hoc comparison of statistical significance (p values < 0.05) using Statistica 6.0 (StatSoft, Inc., 2001).

RESULTS AND DISCUSSION

As color is the variable that should distinguish bright green table olives from regular table olives with a yellow-brown olive color, a Minolta colorimeter using CIE Lab scale was used. Figure 1 displays L*a*b* values of all of the samples showing that bright green table olives are statistically different (p values < 0.05) from the regular table olives with all of the coordinates (−8.4 ± 2.6 vs −2.6 ± 0.3), with coordinate a* (a*, negative values indicate green, while positive values indicate magenta) being the most remarkable. In addition, there are no statistical differences (p values < 0.05) in L*a*b coordinates within the samples of the two sets.

Figure 1.

Color measurement L*a*b* coordinates of five samples of table olives.

Once it was established that color was a determinant variable distinguishing the two sets of samples, the chemical compounds responsible for the color difference were identified and quantified by HPLC–MS/MS. Thus, Figure 2 shows the HPLC chromatograms, recorded at 650 nm, of chlorophylls, chlorophyll derivatives, and metallo–chlorophylls complexes (Table 2) determined in the five samples studied. Table 3 lists all of the pigments identified in Figure 2 with their chromatographic, spectroscopic, and mass characteristics. The information from the spectral data agrees basically with the published information. Although many of these pigments observed herein have been identified^20–22 by mass spectrometry, nine pigment components (peaks numbered 6, 7, 9, 11, 15, 19, 22, 23, and 26) were not yet characterized by MS/MS.

Figure 2.

HPLC chromatograms of five samples, coded A–E, recorded at 650 nm. Note: AU, absorption units; TO, table olives.

Table 2.

Different Structures of Pigments

pigment	I. ringa	R1	R2	R3	R4	R5	M
pheophorbide a	1	H	CH3	H	COOCH3		2H
pyropheophorbide a	1	H	CH3	H	H		2H
Cu-pheophorbide a	1	H	CH3	H	COOCH3		Cu
Cu-pyropheophorbide a	1	H	CH3	H	H		Cu
152-Me-phytol-rhodin g7 ester	3	phytyl	CHO		CH2COOCH3	COOH	2H
Cu-152-Me-phytol-rhodin g7 ester	3	phytyl	CHO		CH2COOCH3	COOH	Cu
chlorophyll b	1	phytyl	CHO	H	COOCH3		Mg
152-Me-phytol-chlorin e6 ester	3	phytyl	CH3		CH2COOCH3	COOH	2H
Cu-152-Me-phytol-chlorin e6 ester	3	phytyl	CH3		CH2COOCH3	COOH	Cu
chlorophyll a	1	phytyl	CH3	H	COOCH3		Mg
pheophytin b	1	phytyl	CHO	H	COOCH3		2H
Cu-pheophytin b	1	phytyl	CHO	H	COOCH3		Cu
pyropheophytin b	1	phytyl	CHO	H	H		2H
Cu-pyropheophytin b	1	phytyl	CHO	H	H		Cu
pheophytin a	1	phytyl	CH3	H	COOCH3		2H
Cu-pheophytin a	1	phytyl	CH3	H	COOCH3		Cu
pyropheophytin a	1	phytyl	CH3	H	H		2H
152-Me-phytol-isochlorin e4	3	phytyl	CH3		CH2COOCH3	H	2H
152-Et-phytol-chlorin e4	3	phytyl	CH3		CH2CH3	COOH	2H
Cu-152-Me-phytol-isochlorin e4	3	phytyl	CH3		CH2COOCH3	H	Cu
Cu-152-Et-phytol-chlorin e4	3	phytyl	CH3		CH2CH3	COOH	Cu
Cu-pyropheophytin a	1	phytyl	CH3	H	H		Cu

^aI. ring: Isocyclic ring (ring V).

Table 3.

Spectroscopic and Chromatographic Information of Pigments Quantified by HPLC–MS and Identified in the Samples^a

			spectral data in the eluent						published data
			maxima (nm)				calcdc	obsd	maxima (nm)
peak no	pigment	K′c	I	II	III	peak ratio	[M + H]+	[M + H]+	I	II	III	peak ratio	solvent	ref
1	pheophorbide a	2.05	410b		666	1.94	593.276	593.277	409		666	1.85	MP	23
2	Cu-pheophorbide a	2.76	400	425b	651	1.10	654.190	654.193	400	424b	654	1.00	MP	34
3	pyropheophorbide a	3.27	410b		666	2.03	535.270	535.272	409	666		1.85	MP	11
4	Cu-pyropheophorbide a	4.63	400	425b	651	1.10	596.184	596.182	404	422b	653		MP	27
5	unknown	9.14	447b	577	626	10.98
6	152-Me-phytol-rhodin g7 ester	9.60	427b	561	649	8.65	903.563	903.565	429b	656			MP	21e
7	Cu-152-Me-phytol-rhodin g7 ester	9.80	429b	565	613	6.54	964.477	964.475	430b		614		MP	21e
8	chlorophyll a	10.10	464b	598	649	2.52	907.522	907.520	453b	593	642	2.89	DE	35
9	152-Me-phytol-chlorin e6 ester	10.20	400b		662	3.90	889.584	889.586	404b	642	660		MP	21e
10	chlorophyll b′	10.35	464b	598	649	2.52	907.522	907.520	453b	593	642	2.89	DE	35
11	Cu-152-Me-phytol-chlorin e6 ester	10.65	407b	504	629	2.46	950.498	950.495	408b		635		MP	21e
12	chlorophyll a	10.95	432b		662	1.14	893.543	893.545	430b	615	661	1.32	DE	35
13	chlorophyll a′	11.29	432b		662	1.14	893.543	893.545	430b	615	661	1.32	DE	35
14	pheophytin b	11.73	437b		651	4.59	885.552	885.550	433b	599	654	4.81	DE	35
15	Cu-pheophytin b	11.97	445b		633	2.52	946.466	946.469	443b	633		2.38	MP	34
16	pheophytin b′	12.01	437b		651	4.59	885.552	885.550	433b	599	654	4.81	DE	35
17	pyropheophytin b	12.35	437b		651	4.34	827.547	827.549	435b		654	3.83	MP	34
18	pheophytin a	12.97	410b	505	666	2.10	871.573	871.575	408b	503	667	2.14	DE	35
19	Cu-pyropheophytin b	13.12	444b		633	2.02	888.461	888.463	443b		633	2.38	MP	34
20	152-Me-phytol-isochlorin e4	13.27	400b		667	2.94	845.594	845.595	399b		662	2.92	MP	27f
21	pheophytin a′	13.39	410b	505	666	2.10	871.573	871.575	408b	503	667	2.14	DE	35
22	Cu-pheophytin a	13.91	400	425b	651	0.97	932.487	932.489	400	424b	654	1.00	MP	34
23	Cu-pheophytin a′	14.57	400	425b	651	0.97	932.487	932.489	400	424b	654	1.00	MP	34
24	pyropheophytin a	14.63	410b	505	666	1.97	813.568	813.565	409b		666	1.85	MP	34
25	Cu-152-Me-phytol-isochlorin e4	15.50	408b		630	2.34	906.508	906.510	406b		630	2.55	MP	21e
26	Cu-pyropheophytin a	15.94	400	425b	651	1.02	874.482	874.484	400	424b	654	1.00	MP	34

^aK′_c = (t_r – t_m)/t_m, where t_r is the retention time of the peak and t_m is the retention time of the unretained component. MP, mobile phase. DE, diethyl ether.

^bSoret band.

^cCalculated mass (the mass given is for the most abundant isotope, i.e., ⁶³Cu).

^dObserved mass, average Δppm < 2.5.

^eCompared to a similar structure without phytol.

^fCompared to formylpheophytin a.

Pigment profiles presented in Figure 2 show that bright green table olive samples (A, Castelvetrano-Mezzetta; B, Vincenzo’s; C, Castelvetrano-WF) can be easily distinguished from Spanish table olives (D, var. Gordal; E, var. Manzanilla). The chromatograms of the samples coded A–C showed the highest concentrations of Cu-15²-Me-phytol-chlorin e₆ ester (peak 11), Cu-pheophytin a (peaks 22–23), Cu-pyropheophytin a (peak 26), and chlorophylls a (peaks 12–13) and b (peaks 8 and 10).

Concerning the other two samples (coded D,E), the more representative compounds are pheophytin a (peaks 18 and 21) and b (peaks 14 and 16), 15²-Me-phytol-rhodin g₇ ester (peak 6), and 15²-Me-phytol-chlorin e₆ ester (peak 9), although minor pigments, such as pheophorbide a (peak 1) and pyropheophorbide a (peak 3), are also present. The most characteristic of the chlorophyll pigments in the Spanish style olive, processed by fermentation, is that all of the chlorophyll pigments have lost magnesium within the porphyrin ring, resulting in table olives with a yellow-brown olive color.

However, allomerized chlorophyll compounds detected in the Spanish style of table olives were identified primarily as phytol-chlorin e₆ and phytol-rhodin g₇ from their chromatographic characteristics and absorption spectral data.⁹ Subsequently, the nominal mass molecular ion obtained by MS led to its identification as 15-glyoxilic acid pheophytin a and 15-glyoxilic acid pheophytin b, compounds with similar chromatographic and spectral characteristics but different molecular mass.⁸ Finally, accurate mass MS/MS data from the ion fragmentation obtained in this work suggested a new structure such as 15²-Me-phytol-rhodin g₇ ester (peak 6) and 15²-Me-phytol-chlorin e₆ ester (peak 9). These pigments could be obtained by solvolysis of the isocyclic ring of the respective pheophytin according to the reaction scheme given by Hyninnen.²³

From the set of newly identified pigments, the most abundant in bright green table olives were Cu-15²-Me-phytol-rhodin g₇ ester, Cu-15²-Me-phytol-chlorin e₆ ester, Cu-pheophytin a, and Cu-pyropheophytin a, and also 15²-Me-phytol-rhodin g₇ ester and 15²-Me-phytol-chlorin e₆ ester in the Spanish table olives. The MS/MS fragmentation patterns of the first two pigments, using a ramped collision energy, revealed the following daughter ions (Figures 3 and 4): m/z 918.5, 886.4, 568.2, and 536.1 for Cu-15²-Me-phytol-rhodin g₇ ester ([M – H]⁻= 964.5) and 904.5, 872.5, 554.2, and 522.1 for Cu-15²-Me-phytol-chlorin e₆ ester ([M –H]⁻ = 948.5). The ions 918.5 and 904.4 are consistent with the loss of a carboxylic group (M − CO₂) from the respective molecular ion, while the loss of a methoxy group (M − CH₃O−H) from ions 918.5 and 904.4 would yield ions with m/z 886.4 and 872.5, respectively. Similarly, the loss of the phytyl ester chain (M − CH₂CH₂COO-phytyl) from 886.4, 872.5, 918.5, and 904.4 would yield ions m/z 536.1, 522.1, 568.2, and 554.2, respectively. These pigments have been identified before as Cu-15-glyoxilic acid pheophytins a and b, respectively.¹¹ However, these pigments differ from the new structures identified herein due to a glyoxilic acid (−COCOOH) in R₄ (Table 4). No fragment such as M−CO or M−COCOOH has been found for the new structures, allowing the possibility of phytol-rhodin or phytol-chlorin structures.

Figure 3.

MS/MS daughter ions of Cu-15²-Me-phytol-rhodin g₇ ester.

Figure 4.

MS/MS daughter ions of Cu-15²-Me-phytol-chlorin e₆ ester.

Table 4.

MS/MS Fragmentation Pattern of Chlorophyll Derivatives

	(m/z)a
pigment	calculatedb	observed	MS/MS (m/z)
152-Me-phytol-rhodin g7 ester	901.5 a	901.5	857.6, 825.6, 507.2, 475.2
152-Me-phytol-chlorin e6 ester	887.6 a	887.6	843.6, 811.6, 493.2, 461.2
Cu-152-Me-phytol-rhodin g7 ester	962.5 a	962.5	918.5, 886.4, 568.2, 536.1
Cu-152-Me-phytol-chlorin e6 ester	948.5 a	948.5	904.5, 872.5, 554.2, 522.1
Cu-pheophytin b	945.5 b	945.5	662.2, 590.1
Cu-pyropheophytin b	887.5 b	887.5	608.1, 564.2, 536.1
Cu-pheophytin a	931.5 b	931.5	652.2, 580.2, 576.2, 565.1
Cu-pheophytin a′	931.5 b	931.5	652.2, 580.2, 576.2, 565.1
Cu-pyropheophytin a	873.5 b	873.5	594.2, 550.2, 522.2

^a[M – H]⁻ is labeled with “a” and [M]^•− is labeled with “b”.

^bThe mass given is for the most abundant isotope, that is, ⁶³Cu.

The MS/MS fragmentation patterns of Cu-pheophytin a and its epimer a′ using a fixed collision energy of 30 eV are shown in Figures 5 and 6. Both epimers yielded the same fragmentation patterns: the loss of phytyl yielding the fragment 652.2 (M − phytyl), the loss of the phytyl ester chain yielding the mass 580.2 (M − CH₂CH₂COO-phytyl), the loss of the phytol chain (M − O-phytyl) plus the loss of the methoxy carbonyl group (M − COOCH₃) of R₄ yields the mass 576.2, and the loss of the methyl group from the methoxycarbonyl group of R₄ yields the mass 565.1. However, the abundance of fragment ions is different for Cu-pheophytin a and its respective epimer under identical conditions. A similar observation has been reported previously with pheophytins a and b.²⁰ The Cu-pheophytin a epimer exhibited an abundant fragment of 580.2 (M − CH₂CH₂COO-Phytyl), approximately 10% greater than for a, whereas the fragment m/z 576.2 occurred with minor intensity.

Figure 5.

MS/MS daughter ions of Cu-pheophytin a.

Figure 6.

MS/MS daughter ions of Cu-pheophytin a′.

Table 4 shows MS/MS fragmentation patterns of 15²-Me-phytol-rhodin g₇ ester and 15²-Me-phytol-chlorin e₆ ester using fixed collision energy of 15 eV. The ions with m/z 857.6 and 843.6 represent the loss of a carboxylic group from the respective molecular ion, while the loss of a methoxy group from ions 857.6 and 843.6 yields the ions 825.5 and 811.6, respectively. The loss of the phytyl ester chain from 825.5, 811.6, 857.6, and 843.6 yields the ions 475.2, 461.2, 507.2, and 493.2, respectively. MS/ MS fragmentation patterns of Cu-pyropheophytin a used a fixed collision energy of 30 eV. The loss of phytyl yielded the fragment 594.2 (M − phytyl), the loss of the carboxy phytyl yielded the mass 550.2 (M − COO-phytyl), and the loss of the phytyl ester chain yielded the mass 522.2 (M − CH₂CH₂COO-phytyl). Less relevant due to their low concentration in bright green table olives are Cu-pheophytin b and Cu-pyropheophytin b. The MS/ MS fragmentation pattern of Cu-pheophytin b is 945.5 and of Cu-pyropheophytin b is 887.5. The main daughter fragments obtained for Cu-pheophytin b (Table 4) are 666.2 (M − phytyl) and 590.1 due to the loss of −COO-phytyl and −COOCH₃ of the R₄, while the main fragments obtained for Cu-pyropheophytin b (Table 4) are 608.1 (M − phytyl), 564.2 (M − COO-phytyl), and 536.1 (M − CH₂CH₂COO-phytyl).

In regard to 15²-Me-phytol-isochlorin e₄ and Cu-15²-Me-phytol-isochlorin e₄, these compounds were identified previously as formylpheophytin a and its Cu derivative by their respective molecular ions.^11,27 However, both mass accurate m/z values obtained herein are consistent with a structure with a −CH₂COOCH₃ group in R₄ and −H in R₅ (15²-Me-phytol-isochlorin e₄ and Cu-15²-Me-phytol-isochlorin e₄, respectively) or with an ethyl group (−CH₂CH₃) and a carboxylic group (−COOH) in R₅ (15²-Et-phytol-chlorin e₄ and Cu-15²-Et-phytol-chlorin e₄, respectively) (Table 4). The two first options (15²-Me-phytol-isochlorin e₄ and Cu-15²-Me-phytol-isochlorin e₄) are more likely to occur from the loss of the carboxylic group of the respective chlorin e₆. Nevertheless, we could not acquire an appropriate MS/MS fragmentation pattern for these pigments that gave us a confident identification. Further MS/MS characterization is needed.

The quantitative analysis of the pigments present in the table olive samples (Figure 7) shows that the range of concentrations of the major pigments of bright green table olives were Cu-15²-Me-phytol-chlorin e₆ ester (26–48%), Cu-15²-Me-phytol-rhodin g₇ ester (2–18%), Cu-pheophytin a (3–18%), pheophytin a (0–18%), Cu-pyropheophytin a (1–5%), chlorophyll a (0–6%), 15²-Me-phytol-rhodin g₇ ester (1–2%), and chlorophyll b (0–2%). The major pigments present in the Spanish table olives, in terms of concentration, were pheophytin a (36–37%), pheophytin b (13–20%), 15²-Me-phytol-chlorin e₆ ester (17–29%), and 15²-Me-phytol-rhodin g₇ ester (8–15%).

Figure 7.

Percent of major pigments present in the five samples, coded A–E. Note: Me-Pr g₇, 15²-Me-phytyl-rhodin g₇ ester; Cu-Me-Pr g₇, Cu-15²-Me-phytyl-rhodin g₇ ester; chl b, chlorophyll b; Me-Pc e₆ ester, 15²-Me-phytyl-chlorin e₆ ester; Cu-Me-Pc e₆, Cu-15²-Me-phytyl-chlorin e₆ ester; chl a, chlorophyll a; Pheo b, pheophytin b; Pheo a, pheophytin a; Cu-pheo a, Cu-pheophytin a; Cu-py a, Cu-pyropheophytin a.

The relative amount of various chlorophyll derivatives, with or without metal within the porphyrin ring, allows differentiation among the samples. Thus, Figure 8 shows the percent of Cu-metallo–chlorophylls (T Cu), Mg-metallo–chlorophylls (T Mg), chlorophylls without magnesium (T-Mg), Cu-metallo–chlorophylls plus Mg-metallo–chlorophylls (T Cu+Mg), and an unknown compound (UID). From this comparison, we have observed that bright green table olives contain 33–99% of Cu-metallo–chlorophyll complexes, while Spanish table olives have approximately 95% of chlorophyll derivatives without the magnesium porphyrin complex and approximately 5% of Cu–porphyrin complex. Furthermore, bright green table olives contain small quantities of Mg-metallo–chlorophylls (0–8% due to chlorophylls a and b) in contrast to Spanish table olives that do not contain detectable levels of these compounds. The differences may be attributed to the different methods of the industrial processing of table olives (Figure 9). The Castelvetrano style (samples coded A and C) is manufactured without fermentation and requires approximately 2 weeks for processing, although the shelf life of its table olives is only a few weeks under warm storage conditions.²⁴ The presence of Mg-metallo–chlorophylls (5–8%) in samples B and C implies that the processing style was Castelvetrano. Sample A, however, does not contain Mg-metallo–chlorophylls, which can be due to a modification of the Castelvetrano processing style to prolong the shelf life such as pasteurization. In contrast, the Spanish style olive, based on a lactic fermentation process (Figure 9), requires 2–5 months^24,25 and results in the disappearance of chlorophylls a and b in less than 120 days.²⁶

Figure 8.

Percents of total Cu-metallo–chlorophylls (T Cu), total Mg-metallo–chlorophylls (T Mg), total chlorophylls without magnesium (T-Mg), and total Cu-metallo–chlorophylls plus Mg-metallo–chlorophylls (T Cu+Mg) and the unknown peak (UID) for each sample studied.

Figure 9.

Scheme of the three different processing styles for table olives. Note: Source, refs 24, 25.

Previous researchers^{10–12,27,28} have reported a GS alteration of the Gordal variety of table olives where the color alteration is due to Cu-metallo–chlorophylls derivatives. One report shows the percent of total Cu-metallo–chlorophylls a with respect to the total of a series after the lactic fermentation with amounts ranging from 2% to 11%.²⁷ These data are lower than our observation for three commercially available bright green table olive samples (36.1–99.8%).

We have found that the total elemental Cu from the quantification of Cu-metallo–chlorophyll complexes present in the three bright green table olive samples varies between 1.3 mg and 4.7 mg/kg, and within the range of mineral Cu (1.7–11.0 mg/ kg) quantified in different varietal table olives²⁹ and fresh olives (3.5–7.2 mg/kg).³⁰ Thus, there appears to be adequate quantities of Cu native to the fruit to allow for regreening to occur.¹⁰

The copper absorbed by the olive fruit could arise from copper in the soil but most likely derives from the content of copper in the pesticides applied to prevent soapy olive (fungus infection by Gloeosporium olivarum) and “repilo” (fungus infection by Cycloconium oleaginum).

As such, the bright green color of table olives is mostly due to the existence of Cu within porphyrin ring, but the presence of Cu-metallo–chlorophylls has not been reported in fresh olives, fruits, or leaves. However, these compounds can be generated using copper salts (CuCl₂, CuSO₄) during the industrial processing of food^4–7 or generated during the commercial processing of table olives with endogenous copper absorbed by the fruit.²⁷ We assume that the Cu–chlorophyll compounds are produced without adding copper salts and that the industrial processing has been carried out according to international rules and regulations.^31,32

Olives with GS alteration might be a defect created by the partial formation of metallo–chlorophyll pigments during the manufacture of green table olives resulting in a color that is not uniform. Although color is important for consumer acceptance, the bright green table olives provide a source of metallo– chlorophyll derivatives (Cu-Me-phytol-chlorin e₆, 26–48%; and Cu-pheophytin a, 3–18%) that have been reported to possess an inhibitory activity against mutagenesis and higher antioxidant activity as well.³³ Further investigation is needed to understand the processing method for regreening table olives that will result in optimal uniformity in color and consumer acceptance.