Identification of Methylxanthines and Phenolic Compounds by UPLC-DAD-ESI-MS OTOF and Antioxidant Capacities of Beans and Dark Chocolate Bars from Three Trinitario×Forastero Cocoa (Theobroma cacao L.) Hybrids

Interest on Trinitario cocoa has continuously increased due to the fact that some genotypes of this group had inherited some characters of vigorosity from Forastero and the flavor grade of Criollo. The pronounced incompatibility between Trinitario clones orientates research on the crossing of Trinitario clones with other varieties of cocoa in order to increase productivity and cocoa beans quality. Polyphenols and methylxanthines are bioactive compounds responsible for the health benefits of cocoa and cocoa based products. Cocoa is a crop with a high content of bioactive compounds in the plant kingdom. This study aims at correlating genotypes, methylxanthines and polyphenols as well as antioxidant activity of beans and dark chocolate derived from Trinitario×Forastero hybrids. Total polyphenol content and total condensed tannin contents were determined by spectrophotometric methods. Individual bioactive compounds were identified by UPLC-DAD-ESI(+)-MS both in beans and dark chocolate bar. Results showed differences within beans of dark chocolate and between beans and dark chocolate. Beans from the hybrid (♀)SNK10×(♂)IMC67 recorded the highest polyphenol content (49.18±1.55mg CatE/g) considering the two matrices. The highest concentration of condensed tannins (22.81±0.69 mgCatE/g) was recorded in beans obtained from the hybrid (♀)ICS40×(♂)UPA134. Dark chocolate bar from beans of the hybrid (♀)ICS40×(♂)UPA134 was the richest in condensed tannins (18.25±0.71 mg CatE/g). The total polyphenol and total condensed tannin contents, the chemical composition as well as the antioxidant activity could be genotype-dependent and were affected negatively during roasting. In fact, roasting decreased the polyphenol content and consequently the antioxidant activity. (-)-epicatechin, theobromine, ferulic and chlorogenic acids and their derivatives, procyanidin C1, caffeine and salicylic acid 3-O-glucose/galactose were identified in beans and/or dark chocolate.

consumed frequently for its pleasant, stimulant, and euphorizing effects (Torres-Moreno et al., 2012). The primary chocolate categories are dark, milk and white chocolate (Afoakwa et al., 2007). The widely enjoyed chocolate-flavor makes it a favorite ingredient in bakery, ice cream, beverage, syrup production and as confection in itself (Lecumbersi et al., 2007).
Polyphenols and methylxanthines are two main active metabolites in cocoa and chocolate. Polyphenols represent about 10 to 18% of unfermented dry cocoa beans weight (Gómez-Juaristi et al., 2019;Martí nez-López et al., 2014a;Niemenak et al., 2006;Wollsgast & Anklam, 2000). Stored in so-called cotyledons, polyphenols confer astringent and bitter sensations and contribute significantly to the green and fruity flavors of cocoa liquors (Noor-Soffalina et al., 2009). Also, polyphenols are responsible for the positive health benefits associated with the consumption of cocoa and cocoa based products. Three main groups of polyphenols have been identified in cocoa/chocolate. They include catechins (flavan-3-ols, ∼37%), anthocyanins (∼4%), and proanthocyanidins (∼58%). Monomers account for 5-10% of the total cocoa polyphenols and polymers account for approximately 90% of the total cocoa polyphenols (Khan & Nicod, 2012). According to some studies, (-)-epicatechin appears to be the most abundant catechin representing about 35% of the total phenolic content (Gonzá lez- Barrio et al., 2020;Oracz et al., 2015). Besides the different health effects, studies underscored the role of polyphenols in modulating eicosanoid synthesis as well as increasing nitric oxide synthesis which in turn inhibits the oxidation of low density lipoprotein (LDL). This inhibits platelet activation thereby stimulating the production of anti-inflammatory cytokines and inhibiting pro-inflammatory cytokine production (Rimbach et al., 2009;Niemenak et al., 2006). Although most of the studies attribute the health benefits of cocoa or cocoa based products to polyphenols, it should be noted that cocoa and its products are not only rich in polyphenols, but also contain methylxanthines that account for about 3.2% of defatted unsweetened chocolate composition (Camillo & Benitez, 2000). These purine alkaloids are present in a variety of pharmaceutical products and drugs since they are known to stimulate the central nervous system, induce gastric secretions, and act as diuretics (Gómez-Juaristi et al. (2019); Camillo & Benitez, 2000;Evans & Griffiths, 1999). Theobromine was found as the major alkaloid in Theobroma cacao. Cocoa seeds contain between 1 and 4 % weight of this compound. It is therefore a plant with tremendous economic importance because of its use in beverages and chocolates (Ashihara & Suzuki, 2004;Caudle et al., 2001).
Cocoa (Theobroma cacao L.) is a perennial crop widely cultivated in Africa, America and Asia and whose beans are used to produce cocoa powder and chocolate (Cuatrecasas, 1964). According to the international cocoa committee (ICCO, 2019), over 4.7 million tons of cocoa were produced during the 2018/2019 crop year. World leaders in cocoa beans production include Ivory Coast, Ghana, Indonesia, Cameroon, Ecuador, Nigeria and Brazil. In the last decade, the African continent accounted for about ¾ of the total annual cocoa production (ICCO, 2019;Patras et al., 2014). Three major groups of cocoa are known: Criollo, Forastero and Trinitario. Criollo cocoa is known for its fine aroma. Forastero, in its majority is classified as bulk cocoa. Trinitario cocoa is a hybrid of Forastero and Criollo groups. Research on this cocoa group highlighted that some of them can be classified as fine/flavor cocoa. Meantime, the problem of incompatibility between Trinitario clones limits production of pure Trinitario cocoa (Loor Solórzano, 2007). Crossing between other groups of cocoa may increase the production of Trinitario cocoa thereby ensuring cocoa quality. In Cameroon, Trinitario cocoa is represented by SNK and ICS clones (Efombagn et al., 2008). It is well known that the quality of cocoa and cocoa based products depends on the genotype and post-harvest practices as well as technological processes. This quality is a function of both chemical composition and concentration of cocoa beans and chocolate. Kongor et al. (2016) mentioned that genotypes of cocoa beans are significantly related to the chemical composition. Methylxanthine and polyphenol content in cocoa is highly dependent on the cocoa genotype (Urbanska et al., 2019;Herná ndez-Herná ndez et al., 2018;Aprotosoaie et al., 2016). Diversity of polyphenol and methylxanthine composition have been reported in cocoa beans within the same clone/variety (Noor-Ariefandie & Fan Zhu, 2019). In fact, the total phenol content and specifically the type of polyphenol is associated with antioxidant capacity. Presently, no study correlates the total polyphenol content, the chemical composition as well as the antioxidant activity of dried fermented cocoa beans and their corresponding dark chocolate from specific Trinitario×Forastero hybrids.
This study seeks to assess variations in polyphenol content, antioxidant activity as well as identification of polyphenols and purine alkaloids in fermented dry cocoa beans from three Tinitario×Forastero hybrids and their corresponding dark chocolate bars.

Materials
Beans of (♀)SNK16×(♂)T60/887, (♀)ICS40×(♂)UPA134, and (♀)SNK10×(♂)IMC67 hybrids obtained by hand-pollination during two consecutive crop years (2017 and 2018) were fermented for 144 h periodically spread after every 48 h. After ten days' sun-drying, the cocoa beans were sent to the laboratory of Food Technology of the Institute of Agricultural Research for Development (IRAD), Yaounde station for chocolate production. Dark chocolates (100 g and 400 g chocolate bars) were produced (67% liquor, 23% grey cane sugar and 10% cocoa butter). Fermented and dried beans were deshelled and roasted in a vacuum oven (Heratherm, Thermo scientific) at 120°C for 20 min. After roasting, cocoa liquor was obtained using a plenary mixer (Varjomatic, England) at 45°C (heat gun) for 4 h. The dark chocolates were obtained by mixing cocoa butter and cane sugar for 15 min at 45°C. Dark chocolate samples were molded and stored at 4°C for 24 h before being wrapped with aluminum paper. For bioactive compounds analysis, the dark wrapped chocolates were kept in refrigerator (Liebehrr, Poland) at 6°C before analysis Folin-Ciocalteu phenol was purchased from Sigma-Aldrich (St. Louis, Mo, USA). Na 2 CO 3 and caffein were purchased from Merk (Darmstadt, Germany). Methanol, chlorogenic acid, theobromine, quercetin, catechin, ferulic acid, caffeic acid were obtained from Sigma-Aldrich. Acetonitrile was of HPLC grade purity (Romyl, Teknokroma, Barcelona, Spain) and n-hexane was obtained from VWR international (Fontanay-sous-bois, France). All chemicals were of analytical and HPLC grades.

Total Polyphenols and Methylxanthines Extraction
Approximately 20 fermented dry cocoa beans were randomly selected from 1 kg of the sample. Prior to extraction, samples of beans and chocolate (20 g) were first defatted. The samples of fermented dry cocoa beans of each hybrid and its corresponding dark chocolate were manually ground to fine particle size of 0.5 μm for extraction. The cocoa mass and chocolate mass samples were defatted with n-hexane (1:5 m/v) for five hours using a Soxhlet apparatus. The extraction procedure was done as described by Herná ndez-Herná ndez et al. (2018) with light modifications. Briefly, 500 mg of defatted bean powder or dark chocolate were mixed with 5 ml methanol: water: hydrogen chloride (80/19.5/0.5) and vortexed for 5 min. The mixture was then centrifuged at 3500 rpm (5702 R Centrifuge, Eppendorf) at 4°C for 10 min. The supernatant was kept and the filtrate was mixed with 2 mL of the above solution and centrifuged in the same conditions. The second supernatant was mixed with the first one and the solution obtained was used as bioactive compounds extract.

DPPH Radical Scavenging Activity
Free-radical scavenging activity of fermented dry cocoa beans and dark chocolate extracts was measured using the method described by Scherer and Godoy (2009) with slight modifications, where the DPPH removes a hydrogen atom from the molecule under study resulting to a decrease in absorbance at 517 nm. For analysis, 500 µL of the solution of DPPH (Sigma-Aldrich, Steinheim, Germany) at a concentration of DPPH (400 µM) was added to 50 µL of each chocolate or beans extract prepared at four different concentrations: 50; 100; 200 and 500 µg/mL in methanol, protected from light. The samples were allowed to stand at 25°C for 0.5 h after which the absorbance was read at 517 nm using a spectrophotometer (UV-1605 Shimadzu spectrophotometer). All analyses were performed in triplicate. Gallic acid was used as control. The antiradical activity was expressed as a percentage of reduced polyphenols and calculated as follows: Antiradical activity (%) = [1-(Abs sample 517/Abs control 517)] ×100 Antioxidant activity index (AAI) was calculated as follows: AAI = (final concentration of DPPH in the reaction)/IC 50 , where the final concentration of the reaction was 143.38mg· mL −1 . The concentration for 50% inhibition (IC 50 ) was calculated by the linear regression equation between the extract concentration and the corresponding scavenging effect. Scherer and Godoy (2009) established the following criteria of AAI values for plant extracts: poor activity < 0.05 < moderate < 1.0 < strong < 2.0 < very strong.

Ferric Reducing/Antioxidant Power (FRAP)
The ferric reducing/antioxidant power (FRAP) assay was carried out according to a standard procedure by Benzie & Strain (1996) with some modifications. FRAP reagent was prepared as follows: 3 mL of TPTZ (2,4,6-Tripyridyl-s-Triazine) solution (10 mM) in 50 mM HCl added to 3 mL of FeCl 3 .7H 2 O and 25 mL of acetate buffer (0.25 mM pH 3.8). The solution was incubated at 37 °C for 30 minutes. For antioxidant activity evaluation, 1000 µL of the FRAP reagent was mixed with 400 μL of distilled water and 100 μL of samples or standard. Samples and standards were homogenized and incubated at 37 °C for 30 minutes, and the reading was performed at 593 nm on the spectrophotometer (UV-1605 Shimadzu spectrophotometer). Gallic acid solution was used as a standard. Results were expressed as gallic acid equivalent (micrograms) (GAE)/milligrams (mg) of dry fat-free matter (ffm). All analyses were performed in triplicate.

Determination of Total Polyphenols and Total Condensed Tannins
Total phenolic content was determined by the spectrophotometric method with Folin-Ciocalteu reagent according to (Singleton & Rossi, 1965). Folin-Ciocalteu reagent (1 mL) and 20% (w/v) sodium carbonate (25 µL) were added to the proper amount of sample (estimated to achieve absorbance reading in the range of 0.1-0.8). Subsequently, the mixture was vortexed using STUART vortex mixer (3000 x g; 30 s; Biocote, UK) and then the samples were left in a dark place at ambient temperature for 60 min. After that time, the absorbance was read using GENWAY 6305 spectrophotometer (Germany) at the wavelength of 760 nm. Total phenolic content was read against calibration curve prepared for the MeOH/H 2 0/HCl (80/19.5/0.5) solution of chlorogenic acid.
Proanthocyanidins (condensed tannins) were analyzed by the procedure described by Hagermann (2002) with some modifications. Briefly, ButOH/HCl assay was carried out by mixing 0,5 mL of dark chocolate or beans extract with 3 mL of a solution of n-ButOH-conc. HCl (95:5, v/v) and 0.1 mL of a 2 g/100 g solution of NH 4 Fe(SO 4 ) 2 in 2 mol/L HCl. The solution was capped and thoroughly mixed and heated for 45 min at 95°C in a water bath. The sample was cooled and the visible spectrum recorded at λ =550 nm. The blank value of the ButOH-HCl-FeIII solvent was subtracted. All measurements were performed in triplicate. The quantity of condensed tannins was determined from a standard curve of catechin treated with ButOH-HCl-FeIII mixture, and expressed as mg catechin equivalents (CatE)/g of sample.

UPLC-DAD-ESI(+)-MS of Individual Phenolic and Methylxanthine Compounds
All the bioactive compound extracts were prepared in a concentration of 50 mg/mL then filtered through a syringe-filter-membrane. Aliquots of 5.00 µL were injected into the UPLC-DAD-MS Dionex Ultimate 3000 HPLC (Germany) were used to perform the analyses. High Resolution Electro Spray Ionization -Mass Spectra (HRESI-MS) were obtained with an OTOF Spectrometer (Bruker, Germany) equipped with a HRESI source and a UV-visible absorbance detector. The spectrometer was operated in positive mode (m/z range: 100-1200, with a scan rate of 2.5 Hz) with automatic gain control to provide high-accuracy mass measurements within 2 ppm deviation using Na Formate as calibrant. Mass spectra were simultaneously acquired using electrospray ionization in the positive ionization mode. The following parameters were used for experiments: spray voltage of 4.5 kV, capillary temperature of 200°C. Nitrogen was used as sheath gas (10 L/min). The spectrometer was attached to an Ultimate 3000 (Thermo Fisher, USA) HPLC system consisting of LC-pump, UV traces were measured between 200-600 nm, auto sampler (injection volume 5 L) and column oven (30.0°C). The separations were performed using a Poroshell 120A RP-C18 (50x3.0 mm, 2.7 µm particle size) with H 2 O (+0.1% HCOOH, A/acetonitrile (+0.1% HCOOH, B) gradient (flow rate 200 μL/min). Samples were analyzed using a gradient program as follows: 5% B isocratic for 0.5 min, followed by a linear gradient up to 95% B for 4.5 min, after that an isocratic system containing 5% B was spent for 4 min. the system returned to its initial condition (95% A) within 1 min, and was equilibrated for 1 min. Eight standard compounds were used for identification of bioactive compounds using LC-MS apparatus. The eight standard molecules were (+)-catechin, (-)-epicatechin, theobromine, caffeine, quercetin, caffeic acid, chlorogenic and ferulic acids.

Statistical Analysis
Results were obtained by performing analysis of variance (ANOVA) followed by Tukey HSD at P<5% to compare means with the assistance of SPSS 25.0 for windows. Gaphpad 5.0 software was used to perform the scavenging activity of extracts at different concentrations. Cluster analysis with antioxidant activity index (AAI) of cocoa beans and chocolate, using the unweighted pairwise group methods with arithmetical average (UPGMA) on the basis of Nei's (1978) genetic distance were performed with the assistance of SPAD 5.5.
The presence of polyphenols and their main subclasses in cocoa estimated by spectrophotometric methods showed many differences among hybrids and matrixes. Some hybrids in the same conditions of postharvest practices and technological processes are richer in bioactive compounds than others. It is well known that the bioactive compound content of cocoa beans depends on many factors amongst which the genotype/variety of cocoa. Several authors highlighted that different cocoa genotypes/varieties have different contents of polyphenols in the same conditions (Noor Ariefandie et Fan Zhu, 2019;Oracz et al., 2015;Niemenak et al., 2006). In a general point of view, the content of polyphenols seemed to be in accordance with data in literature. Many studies have already presented the concentration of polyphenols in cocoa. It has been demonstrated that the amount of polyphenols represents 5% in fermented dry beans of Forastero cocoa type (Albertini et al., 2015). Our results recorded approximatively the same quantity in fermented dry beans of (♀)ICS40×(♂)UPA234 hybrid. It has been shown that polyphenol content is higher in Forastero cocoa than in Trinitario and Criollo groups (Tomas-Barberan, 2012). On the contrary, the proanthocyanidins content was higher in Criollo classified as fine/flavour cocoa than in Trintario and Forastero cocoas (Giacometti et al., 2015). The content of proanthocyanidin in cocoa was recognized for discriminating cocoa varieties. Results of procyanidin concentrations in different hybrids showed that both beans and bar from the hybrid (♀)ICS40×(♂)UPA134 could be discriminated by others. Many health benefits of polyphenols from cocoa are well described. In a meta-analysis of eight short-term trials, consumption of cocoa was found to reduce the LDL cholesterol content (Jia et al., 2010). In addition, an epidemiological study revealed a consistent increase in blood flow velocity from 8% after 1 week of cocoa consumption to 10% following 2-week consumption of flavanol-rich cocoa (Sorond, 2008). The antioxidant activity of cocoa was mainly attributed to this group of bioactive compounds. The processing of beans led to a reduction in bioactive compound content from fermented dry beans to chocolates. This observation agrees with the data in literature. Certain authors highlighted that Polyphenols are thermolabile molecules whose content reduces in high temperatures and prolonged roasting (Urbanska et Kowalska, 2019;Kothe et al., 2013;Ramli et al., 2006). On the other hand, it was considered that the roasting at low temperatures for a short time better preserves polyphenol content (Ioannone et al., 2015). Temperatures ranging from 120°-140°C for 5 to 35 min were recommended. From our results, it appears that there is not a fix percentage of reduction of polyphenol content from beans to bar. Also, beans with high polyphenol content produce chocolate with similar polyphenol content as well.

Identification of Phenolic and Methylxanthine Individual Compounds by UPLC-DAD-ESI(+)/MS OTOF
Among the eight standard molecules, except (+)-catechin, caffeic acid and quercetin, five of them were detected in cocoa beans or dark chocolate. We noted the presence of methylxanthines (theobromine and caffeine) with high peaks.  (Fig. 1). Several studies had mentioned the presence of (-)-epicatechin in cocoa and cocoa based products. (-)-epicatechin was found to be the second major individual compound in cocoa (Herná ndez-Herná ndez et al., 2018; Noor-Ariefandie & Fan Zhu, 2019). There was a difference between beans and bar intensities of The (-)-epicatechin UV and (+)HRESI-MS (Fig. 1) spectra. These peaks were higher in beans than bars and the intensity was genotype-dependent ( Fig. 1A and 1B).   (Fig. 2B). The presence of methylxanthines specifically theobromine and caffeine help to confirm these compounds in cocoa and cocoa based products. It has been noted that theobromine (C 7 H 8 N 4 O 2 , 3,7-dimethyl-xanthine) is the main methylxanthine in cocoa that represents approximatively 4% of cocoa butter, or between 0.8 and 2% of total cocoa bean dry weight (Barišic et al., 2019;Herná ndez-Herná ndez et al., 2018;Carrillo et al., 2014). Methylxanthines, mainly caffeine, enhance physical and intellectual performance, mitigate fatigue, and cause a feeling of alertness (Gómez-Juaristi et al., 2019;Martí nez-López et al., 2014b;Aprotosoaie & Stanescu, 2010). Theobromine has a weaker effect than caffeine and in the case of cocoa products, its content is even lower or different from caffeine (Franco et al., 2013). Studies have shown that theobromine appears to be even safer for humans than caffeine (Franco et al., 2013).   (Simirgiotis et al., 2015). The maximum wavelengths of ferulic acid were 246 and 310 nm respectively. The presence of ferulic acid in these cocoa beans and their dark chocolate was an important indicator for consumer's health. It had been shown that ferulic acid can inhibit ultraviolet C-induced oxidative DNA damage, being as effective as glutathione, αtocopherol, and vitamin C. In humans, flavanol-rich cocoa products increase the total plasma antioxidant capacity (Sarriá et al., 2020;Keen et al., 2005). The chlorogenic acid also had 222 (325) nm as wavelength and 6.5 min as retention time.  Faisal et al. (2013) identified chlorogenic acid in cocoa using HPLC-DAD-ESI-MS/MS optimized method. The authors obtained two maximum wavelengths in negative mode ESI for this compound 226 and 300 nm respectively. The acidity and antioxidant activity of chlorogenic acid were reported and this phenolic compound was richer in coffee than in cocoa (Rao & Fuller, 2018)   Salicylic acid-3-O-glucoside/galactoside and isorhamnetin-3-O-glucoside were identified using the same procedure. Like procyanidin C1, these two polyphenols were also reported in cocoa.
Three peaks were recorded at 4.8, 5 and 5.1 min respectively in beans of the three hybrids considered. All these peaks had the wavelength of 218 nm. On the contrary, in dark chocolates, only one peak appeared at 5 min. The wavelength and the retention time of each peak and the different m/z ratios obtained were given in Table 2. With respect to total polyphenol and condensed tannins contents, we observed a reduction in peak intensities from beans to bars. The chronological appearance of different individual polyphenol and methylxanthine compound permits their classification into groups. We obtained first flavan-3-ols (epicatechin) and methylxanthine. The second group was composed of phenolic acids (chlorogenic, ferulic and salicylic acids). A similar order was obtained with these individual compounds (Faisil et al., 2013).

DPPH Antiradical Test
Considering the four concentrations (50, 100, 200 and 500 µg/g of free fat matter), the activity of extracts increased when concentration increased. In addition, antiradical activity was specific of each extract and the different genotype expressed different antioxidant activities. Results of antiradical activity of bioactive extracts showed that fermented dry beans had higher DPPH scavenging power than the dark chocolate. This corroborates the observations made by Urbanska et Kowalska (2019). This was confirmed by DPPH in vitro test. This is easily explained by the reduction in polyphenol content during roasting prior to chocolate production. This can be clearly seen with the high content of oligomeric procyanidins in beans than chocolate which exhibit higher antioxidant capacity than monomeric polyphenols. Therefore, beans of cocoa have excellent antioxidant properties than their corresponding dark chocolate bars (Di Mattia et al., 2017). Indeed, in beans, the highest DPPH was obtained with samples whose concentration of polyphenols was highest. The same results were obtained by Caporaso et al. (2018) and Tomas-Barberan (2012). It was observed that antioxidant activity was not directly link to the polyphenol content in chocolate. According to Urbanska et Kowalska (2019), the antioxidant character of chocolate is an extremely complex phenomenon which depends on many factors. It is therefore not possible to determine and correlate the polyphenolic content and antioxidant activity unequivocally. The beans of the hybrid (♀)SNK10×(♂)IMC67 showed the highest percentage of DPPH antiradical activity (85.17%) considering the concentration of 500 µg/ml of extract solution (Fig. 4). The classification of antioxidant activity according to Scherer and Godoy (2009) showed four extracts with strong antioxidant activity and two others with moderate antioxidant activity (Table 3). Only one dark chocolate sample obtained from (♀)ICS40×(♂)UPA134 hybrid displayed a strong antioxidant activity. At 5% heterogeneity, the cluster below was obtained with antioxidant activity index (AAI) of cocoa beans/chocolate on DPPH radicals with three groups identified (Fig. 6). The first group comprised samples with strong antioxidant activity. This group is divided into two subgroups specifying by the matrix. The second group comprised beans from (♀)ICS40×(♂)UPA134 hybrid which displayed the strongest antioxidant activity . The third group comprised chocolate bar from (♀)SNK16×(♂)T60/887 hybrid was classified as having the moderate antioxidant activity. The hierarchical classification demonstrated the fact that the antiradical activity is higher in cocoa beans than dark chocolate bars. The ferric reducing/antioxidant power (FRAP) showed different results in each matrix with the three considered hybrids. The reducing/antioxidant power of beans was higher than that of chocolate bars. The highest reducing/antioxidant activity was obtained with the hybrid (♀)SNK10×(♂)IMC67 in beans and (♀)SNK16×(♂)T60/887 in bars (93.05±1.91 µg GAE/mg ffm and 75.47±2.08 µg GAE/mg ffm respectively). There was a significant difference between dried fermented beans and dark chocolate bar ferric reducing capacity (Fig. 5). The hybrid (♀)ICS40×(♂)UPA134 showed a small difference with respect to the reducing power between beans and dark chocolate. The antioxidant capacity evaluated by the reactions of FRAP with extracts of beans and dark chocolate bars of each genotype varied according to the matrix and the genotype. According to the results displayed in Figures 5 and Table 1 as opposed to total polyphenol content (TPC), the antioxidant capacities of beans are higher compared to the antioxidant capacities of dark chocolate. This was confirmed by FRAP in vitro test. As earlier mentioned, the reduction in polyphenol content during roasting prior to chocolate production directly affected the antioxidant activity of chocolate. It is well-known that cocoa polyphenols are the major contributing factors to the overall antioxidant potential of cocoa and cocoa based Group 1 Group 2 Group 3 products (Urbanska et al., 2020;Urbanska et Kowalska, 2019;Komes et al., 2013).

Conclusion
The identification of methylxanthines and phenolic compounds in an extract is an interesting approach in predicting the antioxidant activities of the extract. The antioxidant activities of cocoa and cocoa based products especially dark chocolate is correlated to the content of bioactive compounds, mainly polyphenols. The content and the activity of each extract depend on the nature of the extract. This study shows that there is a progressive reduction in polyphenol content from fermented dry beans to dark chocolate resulting to a significant decrease in peak intensities of monomeric compounds. The UPLC-DAD-ESI-MS allows the identification of individual bioactive compounds in cocoa beans or cocoa based products. This method could be a rapid way of identifying major individual bioactive compounds in cocoa and cocoa based products. This study therefore shows a relative difference between cocoa hybrids with respect to polyphenol and proanthocyanidin contents, the composition in individual compounds and their antioxidant properties. The antioxidant activity of cocoa is determined by the polyphenol content in cocoa beans. In dark chocolate, the antioxidant activity was a function of polyphenol content which is directly related to the thermic processes employed. Future works will focus on the identification and quantification of individual bioactive compounds on a large scale sampling of genotypes using different liquid chromatographic methods.

Conflict of Interest
The author(s) declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.