Characterization of carotenoid and carotenoid esters of astringent persimmon tissues

Carotenoid and carotenoid esters profiles of peel, pulp and whole fruit tissues of astringent persimmon (Diospyrus kaki Thunb., var. Rojo Brillante) have been characterized in detail and quantified for the first time. Carotenoids were determined by HPLC-PDA-MS/MS (APCI), using a reverse phase C30 column. A total of 38 carotenoids were identified and quantified, corresponding to 21 free carotenoids (13 xanthophylls and 8 hydrocarbon carotenes) and a total of 17 carotenoid esters. The qualitative profiles are very similar among tissues, differing only in the carotenoids concentration. The most important identified free xanthophylls were (all-E)-β-cryptoxanthin, (all-E)-antheraxanthin, (all-E)-lutein, (all-E)Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 22 November 2018 © 2018 by the author(s). Distributed under a Creative Commons CC BY license. doi:10.20944/preprints201811.0548.v1


Introduction
Carotenoids are lipophilic compounds found in fruits, vegetables, flowers, and animals with an important role in nutrition due to vitamin A activity, in addition to antioxidant activity, intercellular communication and immune system activity [1].Two classes of carotenoids can be distinguished: carotenes, which possess only carbon and hydrogen atoms, and xanthophylls, which also possess oxygenated groups.Approximately, 750 carotenoids have been described in nature and only about 50 of them are identified in the human diet, the most abundant being β-carotene, α-carotene, lutein, zeaxanthin and β-cryptoxanthin [2].
Recent evidence has emerged that consumption of carotenoids may protect from cardiovascular, skin, eye and bone diseases and delay the onset and development of several types of cancer [3].
The human body is not able to synthesize carotenoids and therefore dietary ingestion is the only source to meet their requirements.Although the existing evidence is insufficient to establish a recommended dietary allowance (RDA) or adequate intake (AI) for provitamin A carotenoids and other carotenoids, several dietary guidelines recommend a RDA = 700-900 retinol activity equivalents (RAE), considering 1 RAE = 6 µg of β-carotene and 12 µg for others provitamin-A carotenoids such as β-cryptoxanthin and α-carotene [4,5].Since the main food sources of carotenoids are fruits and vegetables, knowledge on carotenoid composition in different edible parts and cultivars will be useful for the selection of nutrient-rich plants for food fortification and proper diet recommendation [6].
Persimmon (Diospyros kaki L.) is a widespread fruit that contains high quantities of carotenoids, besides others bioactive compounds including vitamin C, condensed tannins and dietary fiber [7].The carotenoid composition in fruit and vegetables depends on several factors including variety, ripening stage, de-astringency method or processing [8].Spanish "Rojo Brillante" variety is an appreciated astringent persimmon cultivar that needs a deastringency treatment before commercialization to improve its sensorial quality such as exposure to carbon dioxide in high concentrations [9].Different parts of the persimmon fruit also may contain different types and amounts of carotenoids.For example, the quantity of βcryptoxanthin, β-carotene, lycopene or lutein is greater in the peel compared to the flesh [10].
Persimmon is a seasonal fruit and only obtainable in fresh form for a short period of time throughout the year (in Europe, from November to January).In order to prolong its availability, persimmon (including whole fruit, flesh and/or peel) could be processed into derivative products which would naturally provide great amounts of health-promoting compounds [11,12].Nowadays, the development of new fruit-based products in the form of beverages, smoothies or desserts is increasing worldwide since consumers have become more conscious of the importance of healthy and natural food consumption.Persimmon carotenoid profile had been reported by several authors [13,14,9], but in all of these studies only saponified extracts were characterized and in all cases a C18 column was used by HPLC carotenoid separation, being trans-β-cryptoxanthin, trans-zeaxanthin, trans-lutein, trans-βand α-carotene and lycopene the major carotenoids found in these saponified extracts.Only, Hitaka et al. (2013) [15] reported the characterization of carotenoid fatty acid esters from persimmon peels using the isolation of the main carotenoids and their esters by purification via silica gel column chromatography and the characterization of their structures buy using HR-FAB-MS, 1 H-and 13 C-NMR.β-carotene, lycopene, β-cryptoxanthin mono myristic acid ester, zeaxanthin di-myristic acid ester, and small amounts of xanthophyll esters of palmitoleic and oleanolic acid were found.No other previously published studies showed the individual profile of carotenoids and carotenoid esters of persimmon fruits.
On the other hand, functional foods have made use of innovations in food technology.To minimize losses of nutritional and organoleptic quality associated with traditional thermal treatment, non-thermal technologies have been applied extensively in the processing of plant foods.In this framework, high hydrostatic pressure (HHP) processing has been identified as a useful tool for extending the shelf-life and quality as well as for preserving the nutritional and functional characteristics mainly of fruit and vegetable products [16].Recently, the effects of HHP on the retention of potentially health related compounds and antioxidant activity of fruits and vegetables have widely gained attention from researchers [17].One very important benefit of HHP is that it could produce an improvement on the extraction of bioactive compounds due to induction of many changes on plant food structure during food processing [18,19,20,21].A previous study of our group [9] reported that the pressurization of persimmon tissues can contribute to an efficient extraction of carotenoids.These studies were carried out in saponified persimmon extracts.From these data, the pressurization of persimmon tissues at 200 MPa was the most interesting HPP condition to continue the carotenoid studies in no saponified extracts to analyzed also carotenoid esters with a C30 reverse phase column.
Based on the description above, in this study for the first time an almost complete characterization of carotenoid and carotenoid esters profile of astringent persimmon tissues (peel, pulp, and whole fruit) (Diospyros kaki Thunb., var.Rojo Brillante) was made.In addition, the effects of a high pressure treatment and a pasteurization treatment on the composition of individual carotenoid and carotenoid esters of persimmon tissues were evaluated in order to explore the possibility to use them to improve the extraction of these bioactive compounds, carotenoid and carotenoid esters, for the use of persimmon as functional ingredient.

Persimmon samples
Persimmon fruits (Diospyros kaki Thunb., var.Rojo Brillante) were harvested in Ribera del Xúquer (Valencia, Spain) at commercial maturity stage IV, which is based on the external colour according to Salvador et al. (2007) [22].As corroborated by our previous analysis of "Rojo Brillante" persimmon variety from several areas and local supermarkets, the carotenoid profile remains stable through years (unpublished results).After harvest, fruits were not treated to remove astringency (astringent samples).Selection of those fruits with uniform size and no defects was carried out.Physical and physicochemical characteristics of persimmon fruits (Table 1) were evaluated as described before [9].
Astringent persimmon fruits were washed, drained and hand prepared to obtain three types of tissue: whole fruits, flesh and peel (Control samples).Pieces (20 × 20 mm) of persimmon tissue (approximately 200 g) were vacuum packaged in 200 × 300mm plastic bags (Cryovac®, Sealed Air Corporation, Madrid, Spain), frozen in liquid nitrogen and lyophilized by freeze drying for 5 days at −45°C and 1.3× 10−3 mPa (LyoBeta 15, Azbil Telstar SL, Terrasa, Spain).Each type of freeze-dried persimmon sample was ground by pulverizing (GrindomixGM200, Retsch, Germany) to a fine particle size (<2 mm) before being carefully homogenized and vacuum stored at −36°C in Cryovac® bags until carotenoid analysis.

Carotenoid extraction from persimmon tissues
Extraction and saponification procedures were carried out under dim light.Freeze dried samples were turned into a fine powder using a cutter mill. 1 g of freeze-dried sample was combined with 0.5 g of magnesium carbonate and 60 µL of trans-β-apo-8'-carotenal (0.2 mg/mL), as internal standard, and then extracted with 20 ml of tetrahydrofuran (THF) stabilized 0.01 % BHT in a homogenizer (OMNI Macro ES®, OMNI International).The extract was then filtered, and the residue was extracted twice more with 20 mL of THF and filtered.The three filtrates were combined and evaporated to half the volume on a rotatory evaporator at 35 °C, under nitrogen ambient.The concentrated extract was then added to a funnel containing 15 mL of diethyl ether and 25 mL of water saturated with NaCl.Aqueous phase was washed twice with other 15 mL of diethyl ether.The ethereal organic phases were dried with anhydrous sodium sulfate.The dried organic phase (non-saponified extract) was completely evaporated by vacuum and controlled temperature (20⁰C) and nitrogen ambient and then, dissolved to exactly 2 mL with MeOH/MTBE/H2O (45.5:52.5:2,v/v/v), filtered through a 0.45 µm membrane filter and immediately analyzed by HPLC.

Saponification of carotenoid extracts
In the case of saponified extracts, the dried organic phase was combined with 4 mL of 30% methanolic potassium hydroxide and kept under magnetic agitation for 1.5 hours in nitrogen atmosphere in the dark.The saponified extract was added to a funnel containing 15 mL of diethyl ether and was washed five times with 25 mL of water saturated with NaCl, discarding the aqueous phase each time, to obtain a neutral pH.The extract was then dried with anhydrous sodium sulfate, completely evaporated on a rotatory evaporator with controlled temperature (20⁰C) and then, re-dissolved to 2 mL with MeOH/MTBE/H2O (45.5:52.5:2,v/v/v), filtered through a 0.45 µm membrane filter and immediately analyzed by HPLC.

Carotenoid analysis by HPLC-DAD
The Injection volume was 20 µL.Carotenoids were monitored at 450 nm; also chromatograms were recorded at 402nm also for carotenoids, 325nm for retinoids, and 294 for tocopherols [23].Additional UV/Vis spectra were recorded between 300 to 700 nm.
These calibration curves (up to seven concentration levels) were prepared per triplicate with standard stock solutions for each carotenoid in the concentration range 5-100 µg/mL.
Calibration curves were constructed by plotting the peak area at 450nm for all carotenoids.
The (all-E)-β-carotene calibration was used for quantitation of β-carotene, β-caroteneisomers, while (all-E)-violaxanthin, violaxanthin-isomers and (all-E)-antheraxanthin were quantitated by violaxanthin calibration.In addition, (all-E)-lutein calibration was used for lutein-epoxide quantitation.Other carotenoids such as (all-E)-neoxanthin, (all-E)-βcryptoxanthin and lycopene were quantitated by their corresponding standards.Results were expressed micrograms of the corresponding the carotenoid per 100 g of fresh weight.The NAS-NRC conversion factor was used to calculate the vitamin A value [31].

Liquid chromatography-mass spectrometry (LC-MS/MS (APCI + )
LC/MS analyses were performed with the same HPLC system described above coupled on-line to an Agilent mass spectrometry detector with APCI source model G1947B compatible with the LCMS SQ 6120 equipment, according to the procedure described by Breithaupt and Schwack (2000) [24].Positive ion mass spectra of the column eluate of 13000 Th/s (peak width 0.6 Th, FWHM).Nitrogen was used both as the drying gas at a flow rate of 60 L/min and as nebulizing gas at a pressure of 50 psi.The nebulizer temperature was set at 350⁰C and a potential of +2779 kV was used on the capillary.Corona was set at 4000 nA both in positive ion mode, and the vaporizer temperature was set at 400⁰C.Collisium gas was helium and fragmentation amplitude was 0.8-1.2V.The chromatographic conditions were the same as described for quantitative analyses of carotenoids.The HPLC retention times, UV/Vis spectra, and MS spectral data of carotenoids from whole fruit, peel and pulp of persimmon (Diospyros kaki Thunb., var.Rojo Brillante) tissues are showed in Table 2.

Statistical analysis
The compositional data were expressed as mean and standard deviation (SD).The obtained results were evaluated with variance analysis (ANOVA) and the least significant differences (LSD) were calculated at a p<0.05 significance level.The correlation coefficients were determined by Pearson's test at a p<0.05 significance level.The statistic software employed was SPSS version 2.7.All the analysis was done at least in triplicate.

Characterization of the carotenoid and carotenoid esters profile of Astringent persimmon
fruits (Diospyrus kaki Thunb., var.Rojo Brillante) Carotenoids from lyophilized and rehydrated samples were efficiently extracted with THF.
Methanol was not employed in the last extraction step due to the precipitation of tannins which take place due the high amounts of these compounds present in astringent persimmon tissues.Figure 1  The qualitative profiles of carotenoids and their carotenoid esters in different astringent persimmon, var.Rojo Brillante, were very similar among different tissues (whole fruit, peel and pulp) (Figure 2 and Figures S1 (persimmon peel) and S2 (persimmon pulp) from Supplementary material).The MS fragmentation pattern of carotenoid esters of persimmon tissues showed the usual fragmentation described for the xanthophyll esters previously reported for other fruits and vegetables as mango and citrus fruits [32], pepper, wolfberry, sea buckthorn, apple, squash [33] and jackfruit [27].
The identification is discussed according to the elution order in the C30 column and the chromatographic conditions employed in the present study.The use of authentic carotenoid standards as lycopene, (all-E)-lutein, (all-E)-β-carotene, (all-E)-α-carotene, (all-E)-βcryptoxanthin, (all-E)-zeaxanthin, (all-E)-neoxanthin and (all-E)-violaxanthin, and (all-E)-βapo-8'-carotenal as internal standard make easy the identification and quantification of these carotenoids in the extracts.
Also, recently Petry and Mercadante (2016) [32] summarized sources and MS/MS characteristic fragments by APCI (+) for xanthophyll esters.This reported data helped to reinforce the chemical identification of these compounds in the present study about persimmon fruits.
e : "Y" indicates that the presented analytical data were in agreement with an authentic reference compound.
f .%III/II was not calculated because of poor definition of the UV/vis spectrum or because it was not detected.with nine conjugated double bonds and a potential one β-ring and one ε-ring [30].Several authors shown in their works that the APCI (+) mass spectrum of lutein can be used for unmistakable determination of the chemical structure of lutein [34].Also, injection of (all-E)-β-cryptoxanthin standard as well as the chromatogram of saponified extract (Figure 1, B) confirmed this identification.However, the most abundant fragment ion in the positive ion tandem mass spectrum of (all-E)-αcarotene, corresponding to the α-ionone moiety of m/z 123, was not observed in the APCI(+) mass spectrum of (all-E)-β-carotene.Formation of the ion of m/z 123 was facilitated by the position of the double bond in the terminal ring, which helped stabilize the resulting carbocation.Since this ion was not observed in the positive ion APCI tandem mass spectrum of β-carotene, γ-carotene, or lycopene, it may be used to distinguish α-carotene from these isomeric carotenes [35].

Hydrocarbon carotenes
Peak

Xanthophyll esters
In order to identify the nature of each xanthophyll esters from persimmon tissues, a combined information from chromatographic characteristics as elution order, UV/vis spectra (maxima absorption wavelengths (λmax), spectral fine structure (%III/II) and peak cis intensity (%AB/AII), ad mass spectrum (molecular weight and APCI (+) MS fragmentation pattern)) were employed, using the literature references and the available data of these compounds [37].
In the persimmon extracts xanthophyll esters were more retained that hydrocarbon carotenoids (all-E)-αand β-carotene and their isomers (9Z) and (13Z).Mercadante et al. (2017) [37] reported that reliable identification of carotenoid esters based only in the elution order or retention time is not possible and can lead wrong assignments due to co-elution or poor baseline separation.In the present study, no problems were found to assign hydrocarbon carotenoids as showed Fig. Persimmon carotenoid and carotenoid ester composition was characterized by the presence of xanthophylls in free form and in totally esterified form, in contrast with the carotenoid composition of other fruits as citrus fruits which have the presence of free, partially and totally esterified carotenoids [32].In UV-vis spectrum of (9Z) isomer showed the presence of λmax 327 nm, and % AB/AII values were 0 and 16, respectively (Table 2).In the present work, an asymmetric neoxanthin was detected in contrast to the reported results for mango and citrus fruits, where no perceptible differences in regioisomeric forms of this carotenoid were observed in the conditions applied for analysis by Petri and Mercadante (2016) [32].Previous studies about the carotenoid and carotenoid esters composition of persimmon peel [15] reported the presence of β-carotene, lycopene, β-cryptoxanthin mono-myristic acid ester, zeaxanthin di-myristic acid ester, using silica gel chromatography followed by structural characterization by 1 H-NRM and 13 C-NMR.
Other study described the carotenoids and carotenoid esters in persimmon edible pulp [38], using a reverse phase column C18, identifying β-carotene, lycopene, β-cryptoxanthin myristate, β-cryptoxanthin laurate, zeaxanthin laurate-myristate, zeaxanthin dimyristate, zeaxanthin myristate-palmitate, antheraxanthin dimyristate and antheraxanthin myristate-palmitate and an unidentified ester.In the present study, carotenoid esters found in different tissues of astringent persimmon tissues were accord with these studies, but a higher number of carotenoid esters separated by C30 reverse phase column was identified, being 19 all of them in (all-E) form and esterified mainly by butyric, lauric, myristic, palmitic and stearic acids.
These differences could be attributed to the fruit variety studied which is an astringent one, the agronomic conditions and the fruit maturity stage (Table 1).

3.2.
Carotenoid composition in Spanish astringent persimmon fruits (Diospyrus kaki Thunb., var.Rojo Brillante) The quantitation of carotenoids in the mature Spanish persimmon var.Rojo Brillante tissues are showed in Table 3.Also in this table, the carotenoid and carotenoid esters composition of pressurized and pasteurized persimmon tissues are included.Total carotenoid content (the sum of total free xanthophyls + total hydrocarbon carotenoids + total xanthophyll esters) was greater in peel extracts in all samples, being 11610.1±580.5, 12040.5±602.0 and 12676.7±633.8µg/100 g fresh weight for control, pasteurized and pressurized peel samples.This higher concentration of carotenoids in the peel comes as no surprise since carotenoids play an important role in attracting animals so they can act as pollinators and seed dispersion vehicles [39].Total carotenoid content in fruit pulp ranged 2183.3±110.3(control tissue) to 2058.4±65.7 µg/100 g (pasteurized pulp) fresh weight, while pressurized pulp has 22180.3±155.4µg/100 g fresh weight.
No significant differences were observed between total carotenoids in persimmon peels and pulps non saponified (direct analysis) extracts due to the assayed treatments.Related to whole fruit persimmon samples, the higher total carotenoid content was found in control samples (4164.8±208.2µg/100 g fresh weight), being significantly different (p≤0.05)than pasteurized and pressurized whole fruit tissues (3117.6±155.9 and 3237.9±39.5 µg/100 g fresh weight, respectively).
Persimmon peel extracts have 3.8 and 6.0-fold more total free xanthophylls than whole fruit or pulp extracts, respectively (Table 3).The content of individual free xanthophyll in all persimmon tissues (non saponified extracts, direct analysis) was in the following order: (all-E)-lutein ≥ (all-E)-zeaxanthin ≥ (all-E)β-cryptoxanthin ≥(all-E)-antheraxanthin.Other minor abundant free xanthophyll was found in persimmon Total hydrocarbon carotenoids content ranged between 2381.6±119µg/100 g fresh weight in peel extracts to 266.9±7.4 µg/100 g fresh weight in pulp extracts, being (all-E)-β-carotene the most abundant hydrocarbon carotenoid, excluding lycopene, whose content in the different samples is discussed below.
Persimmon fruits are very rich in lycopene (Table 3).As other carotenoids, the persimmon peel is the tissue with the highest content of this carotenoid, control peel samples have 1007.7±50.4µg/100 g fresh weight, followed by whole fruit samples 176.1±8.8 µg/100 g fresh weight and pulp samples 77.6±21.5 µg/100 g fresh weight.Both treatments, pasteurization and pressurization, did not produced any significant differences (p≤0.05) in the lycopene content in all tissues, except for whole fruit samples when treatments reduced the lycopene content a 25% in pasteurized and 56% pressurized samples, probably due to a degradation reactions by heat (pasteurization) and by a high enzymatic actions produced by high pressures [16].
Most recent papers about persimmon composition or persimmon derived products, showed only the content of carotenoids from saponified extracts.Total carotenoid content in persimmon flours from vars.
Rojo Brillante and Triumph analyzed by spectrophotometric method [40] ranged between 1600 to 1920 µg/ g fresh weight), which is a greater content that in fresh fruit due to the dehydration process.Giordani et al.
Processing by high pressures produced no regular effect on the individual carotenoid and carotenoid esters (xanthophyll esters), but attending to the sum of all of them, high pressures only affect the total xanthophyll esters when the whole fruit samples were analyzed.However, pasteurization affects negatively the content of all carotenoid and carotenoid esters in all persimmon tissues.
If the carotenoids of persimmon tissues were analyzed from saponified extracts (Table 4), no significant differences (p ≤ 0.05) can be observed between control and treated samples for both total free xanthopylls

3
*Values are the mean of two independent determinations ± standard deviation.Lowercase letters indicate statistically significant differences (p<0.05) between treatments for the same tissue.

4 1
Represents the algebraic sum of the identified free xanthophylls and hydrocarbon carotenoids, respectively.

5 2
Represents the algebraic sum of the identified carotenoids in each sample.

6 3
Calculated according to the guidelines of the US Institute of Medicine (2001).

Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 22 November 2018 Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 22 November 2018 doi:10.20944/preprints201811.0548.v1
Values are the mean of two independent determinations ± standard deviation.Lowercase letters indicate statistically significant differences (p<0.05) between treatments for the same tissue.Represents the algebraic sum of the identified free xanthophylls, hydrocarbon carotenoids and xanthophyll esters, respectively.Represents the algebraic sum of the identified carotenoids in each sample.Calculated according to the guidelines of the US Institute of Medicine (2001).
*a b c

Table 2 .
HPLC retention times, UV/Vis spectra and MS spectral data of carotenoids from whole fruit, peel and pulp of astringent persimmon (Diospyros kaki Thunb.var.Rojo brillante)

Table 1S .
Sum of carotenoids (free and esterified) organized by carotenoid species and percentage of contribution of each species to total carotenoids in direct extracts of astringent persimmon (Diospyros kaki Thunb.var.Rojo brillante) submitted to pasteurization (85°C, 15 min) and high-pressure processing (HPP; 200 MPa, 25°C, 6 min) Preprints (www.