Effect of Solar Pre-drying and Yeast Starter Inoculation Treatments on the Chemical Composition of Cocoa (Theobroma cacao L.) Beans from Southwestern Colombia

1. Introduction

The highest production of cocoa (Theobroma cacao L.) in the world is located in the equatorial region of West Africa (about 75%), followed by the equatorial regions of Central and South America (about 20%) and the tropical parts of Asia (about 5%) [1]. According to the National Federation of Cacao, cocoa bean production in Colombia was 70,205 tons in 2021 [2]. In the region of Tumaco (Nariño, Colombia), in Southwestern Colombia, cocoa clones are cultivated in which productivity characteristics predominate over bean quality [3]. Such is the case of CCN-51 (Castro Naranjal Collection). This is a cocoa clone of Ecuadorian origin that is more resistant to common diseases, and it is able to produce yields greater than 4,000 kg of dry beans/ha under high-density planting conditions and complete exposure to solar radiation [4].

Therefore, it is a crop of great importance worldwide. The pods obtained from these trees contain the seeds (commonly known as cocoa beans) from which chocolate is manufactured after a process of microbial fermentation and drying, which are the two main steps in the processing of cocoa beans [5]. The mucilaginous pulp enclosing cocoa beans, which accounts for approximately 40% of the bean fresh weight, has a high content of fermentable sugars (glucose, fructose, sucrose) and a very acidic pH (3.0-3.5, mainly because of the presence of citric acid) [6]. This makes it an ideal medium for the growth of lactic acid bacteria and yeasts [7]. It is well known that both the quality of the fermented beans and the quality of the chocolate made from them are affected by the nature and distribution of these microorganisms present in the cocoa pulp [8]. Alcoholic fermentation of the pulp sugars by yeasts leads to the production of ethanol and its subsequent conversion into acetic acid. This causes endogenous changes in the beans that are fundamental in the development of a wide variety of other compounds with desirable and undesirable aromas, which are decisive in the quality of cocoa [9]. Furthermore, yeasts also act by degrading the pulp through the action of pectolytic enzymes, also contributing to the reduction of pulp and grain acidity through the utilisation of citric acid [10].

The drying process of cocoa beans, carried out after fermentation, aims to reduce the moisture content (from 55-60 to 8%) and the volatile acidity content to stop undesirable reactions and oxidation of polyphenolic compounds [11]. In this sense, sun drying is one of the most widespread drying methods in the rural areas of Colombia [12]. It is considered to be an adequate method for obtaining maximum flavour development [11]. Rodríguez-Campos et al. [13] reported that alcohols, esters, and pyrazine contents increased during the sun drying process. In contrast, acids, aldehyde, and ketone contents decreased. However, this method can result in the production of cocoa of heterogeneous quality due to climatic variations during drying and the increased labour required for processing [6].

The objective of this study was to evaluate the effect of the solar pre-drying in the open air and the inoculation of yeast starter treatments on the physico-chemical composition of cacao (Theobroma cacao L.) beans (CCN51) cultivated in the Nariño-Tumaco region of Colombia.

2. Materials and Methods

2.1. Cocoa sampling and processing

For this study, cocoa fruits of the clone CCN-51 were harvested by traditional methods in Tumaco (Nariño, Colombia) (01°29’03,6’’ N latitude and 78°39’15,4’’ W longitude), a geographical area with an altitude of 2 m a.s.l. Precipitation varies between 2,265 and 6,238 mm, with 168 to 264 days of rain per year, and the average annual temperature and relative humidity are 24 ºC and 84.5%, respectively. Random sampling in farms from the above-mentioned area was carried out, and then the beans were extracted from the pod. Only healthy beans (seeds and placenta-free pulp) without any infection or physical damage were fermented in 40×40×50 cm wooden boxes divided into two compartments (each with 18.5 kg capacity) to facilitate the turning of the beans. The first turning was done after 48 h (anaerobic phase), and then it was repeated every 24 h to promote aerobic conditions until the end of the fermentation. The drawers presented bottom perforations to facilitate drainage of the exudations of the mucilaginous part of the fresh cocoa beans. Four different fermentation treatments were assayed: (1) Control, the beans were traditionally fermented at room temperature and spontaneously for 120 h; (2) Pre-drying treatment (PD), the beans were previously sun-dried for 24 h under a greenhouse-type infrastructure. The beans were arranged in a layer thickness of 2 cm over a perforated surface 1 m above the floor. Then, the beans were traditionally fermented for 96 h; (3) Yeast starter treatment (YS), the beans were inoculated with a mixed starter culture composed of Saccharomyces cerevisiae, Wickerhamomyces anomalus, Candida tropicalis and Pichia kudriavzevii, previously isolated from spontaneous cocoa fermentations in the Tumaco region. Yeasts were propagated in YPD broth (10 g yeast extract, 20 g bactopeptone, 20 g dextrose, per litre distilled water). Following propagation, cell densities were determined according to McFarland method. Each experimental unit (wooden box) was inoculated with 15 ml of the liquid starter culture (cell density was 1.2 × 109 cell/ml) and fermented for 120 h; (4) PD+YS treatment, the beans were previously sun-dried for 24 h and then treated with yeast starters as described before and fermented 96 h more. In the four fermentation treatments, the beans were finally dried naturally in the sun in a wooden dryer, where the cocoa was arranged in a thin layer for 5 days (until reaching a final moisture content under 7%). The dried beans were vacuum packed and shipped to Spain, where they were maintained at –80 ºC until analyses. In this study, 4 replicates per treatment were performed, and the mean values reported.

2.2. Chemical and fatty acid composition analysis

Moisture (oven air-drying method), total protein (Kjeldahl nitrogen), total fat (Soxhlet method), total dietary fibre (enzymatic-gravimetric determination) and ash (muffle furnace) were analysed following official methods [14]. Carbohydrates were calculated by difference. Fatty acid methyl esters from the lipids obtained were prepared by acidic-trans-esterification in the presence of sodium metal (0.1 N) and sulphuric acid (5% sulphuric acid in methanol) [15], and they were analysed by gas chromatography, using a Hewlett-Packard HP-4890 Series II gas chromatograph equipped with a split/splitless injector and a flame ionization detector (FID). Fatty acids were separated on a nitroterephthalic acid modified polyethylene glycol (HP-FFAP) semi-capillary fused silica column (30 m long, 0.53 mm i.d., 1 μm film thickness) maintained at 230 °C for 25 min. Injector and detector temperatures were 250 °C. The carrier gas was nitrogen at 1.8 ml/min. Individual fatty acids were identified by comparison of their retention times with those of standard reference mixtures (Sigma Chemical Co., St. Louis, Missouri, USA). Results were expressed as the percentage of total fatty acids present.

2.3. Determination of total polyphenol content

The total polyphenol content (TPC) of cocoa extracts was determined spectrophotometrically using the Folin-Ciocalteau reagent according to the method of Cros et al [16]. Briefly, 0.5 g of ground cacao were mixed with 40 mL of acetone:water mixture (80:20) and filtered through Whatman No. 1 filter paper. The residue was washed with 20 mL of the same acetone:water mixture and the total volume of filtrate was made up to 100 mL in a volumetric flask. One mL of the previously prepared extract, 7 mL of distilled water, and 500 μL Folin– Ciocalteu's reagent were mixed. After 3 min, 1 mL of saturated Na₂CO₃ solution was added. Then, the mixture was incubated for 2 h and the absorbance was measured at 760 nm against a blank sample using a digital spectrophotometer (Thermo Fisher Scientific Inc., USA). A calibration curve with gallic acid as the standard with six known concentrations was prepared. The Least Squares method was used to calculate the linear regression and correlate absorbance measures in cocoa samples with polyphenol content. The TPC was expressed as milligram equivalent of gallic acid (GAE) per gram of dried cocoa sample (mg GAE/g dry cocoa). All measurements were performed in duplicate.

2.4. Analysis of methylxanthines and lactic acid

The extract for theobromine, caffeine and lactic acid determination by HPLC was obtained according to the method described by Ramos et al [9], with modifications. One gram of dried and ground cocoa sample was mixed with 20 mL distilled water by vigorous vortexing for 5 min. The mixture was centrifugated at 7,000 rpm at 4 ºC during 10 min. The supernatant was collected, and the precipitate was resolved with 5 mL distilled water, vortexed, and centrifugated as previously described. The two supernatant fractions were mixed and made up to 25 mL and then 2 mL were taken and filtered through 0.22 µm Millipore filter to carry out the HPLC analysis.

HPLC analyses were carried out by high-performance liquid chromatography using an Agilent 1100 Series (Agilent Technologies, Palo Alto, CA, USA) system with a diode array detector (Agilent 1200 series) set at 280 nm. Separation was performed on a Lichrospher 100 RP-18 column (250 mm long, 4 mm id, and 5 µm particle size) with a mobile phase of methanol:acetic acid:water (79:1:20), an injection volume of 50 µL, and a flow rate of 1.0 mL/min. Theobromine, caffeine and lactic acid were identified by comparing the retention times to those of standards. Quantitative analyses were carried out using the individual standard curves for each compound determined.

2.5. Statistical analysis

Data from chemical, fatty acid, total polyphenol, methylxanthines and lactic acid compositions are presented as the mean and standard deviation (SD). A one-way Analysis of Variance (ANOVA) was performed to check the differences between cocoa bean treatments (Control, PD, YS and PD+YS). The differences between the means of each treatment were significant for less than 5% of the total variance (p<0.05) and were compared using the Tukey test. The statistical analyses were performed by using the Minitab 17 statistical software (version 17.2.1, Minitab Pty Ltd, Sydney, Australia).

3. Results and Discussion

3.1. Chemical composition of cocoa bean samples

The chemical composition and the effect of processing treatment of unroasted CCN-51 cocoa bean samples, after fermentation and drying processes, are presented in Table 1. Moisture content ranged from 4.9 to 5.2 g/100g, while the contents of the two main components of the samples, fat and carbohydrate components, varied from 39.0 to 41.4 and 38.2 to 39.2 g/100 g, respectively. Results in the literature confirm that fat is the main component of cocoa beans, followed by carbohydrates [17]. With respect to protein content, the results obtained varied between 12.6 and 14.0 g/100g. Finally, ash content was between 2.7 and 2.9 g/100g. These results agree with previous data obtained in other studies of cocoa beans [17,18]. In the present experiment neither the PD, YS nor the PD+YS treatment had any significant effect (p>0.05) on the chemical composition of cocoa bean, compared to Control treatment. It is well documented that both fermentation and subsequent drying of the beans (artificially or by sun-drying) affect their physico-chemical composition [19]. In this sense, Barrientos et al [20] observed a decrease in the total fat content of cocoa beans fermented and sun-dried for 132 h. In the same way, protein content decreases during pod storage and cocoa bean fermentation. This is caused by the endogenous breakdown of cocoa bean proteins to oligopeptides and free amino acids [21]. This reduction is reported to be due to the action of protease enzymes in the pods during storage, and these enzymes initiate the process of proteolysis [22]. De Brito et al. [23] also pointed out a decrease in protein content during the cocoa bean fermentation process, but it remained constant during the subsequent sun-drying process. In our study, there was no observed effect from the pre-drying treatment on the physico-chemical composition of the beans. This was probably due to the shorter pre-drying process (24 h) carried out in our experiments compared to other studies [20,23]. In the same way, the chemical composition of cocoa beans was not affected by yeast inoculation in the present experiments. Other authors revealed differences in both visual and other sensory characteristics, as well as chemical composition, between beans fermented in the absence and presence of yeasts [24]. In our study, no effect of yeast treatment was observed on the carbohydrate content of cocoa beans. However, it is well known that pulp sugars are the main substrates for microbial fermentation during the fermentation process of cocoa beans. This creates environmental conditions that produce flavour precursors within the beans [7]. So, yeast growth or inoculation are essential to develop cocoa bean fermentation [24].

3.2. Fatty acid composition of pre-dried and yeast starter inoculated cocoa bean samples

The results related to fatty acid composition of cocoa beans are shown in Table 2. The main fatty acids were C18:0 (33.6-33.8 %), C18:1 (32.2-32.4%) and C16:0 (29.2-29.3%). These results are in accordance with those obtained by other authors [17,18] who also reported that these three fatty acids account for more than 90% of the total fatty acids identified in cocoa butter. Total saturated fatty acids (SFA) showed the highest content (up to 64%), followed by total monounsaturated fatty acids (MUFA) (up to 32%), and finally total polyunsaturated fatty acids (PUFA) (less than 3%). The ratio saturated/unsaturated (S/U) varied from 1.82 to 1.85. In this study, no significant differences (p>0.05) in any of the fatty acids identified (nor in total SFA, MUFA, PUFA, and S/U ratio) were found between the Control and the different treatments conducted. Fatty acid profiles in cocoa beans are mainly affected by the variety and the growing conditions of the cocoa beans [18]. However, other studies confirm very small differences in the fatty acid composition of cocoa from different geographical origins [17,18]. In accordance with our results, no studies found changes in the fatty acid profile of cocoa beans because of different fermentation or pre-drying processes. So, our results confirm that during pre-drying or fermentation of cocoa, by adding a yeast culture, no biochemical reactions are generated that could modify the fatty acid profile of the dried cocoa beans. Differences in fatty acid composition, mainly MUFA and PUFA, between treatments can influence the final cocoa butter quality. In this sense, the presence of oleic acid in high quantities plays an important role because it influences the type of triacylglycerol formed, affecting the melting point of cocoa butter and therefore the final quality of the manufactured product [17].

3.3. Total polyphenol, methylxanthine, and lactic acid content of pre-dried and yeast starter inoculated cocoa bean samples

Cocoa fermentation conditions, as well as other post-harvest processing operations such as roasting of cocoa beans, affect both the concentration of polyphenols and methylxanthines, thus affecting the quality of the final product [25]. The kernel or cotyledon of cocoa bean is rich in polyphenols and alkaloids including caffeine and theobromine, stored in a single vacuole from polyphenolic cells [26]. Polyphenols and alkaloids are mainly responsible for the flavour of the bean (polyphenols contribute to the flavour and colour of cocoa beans, and alkaloids to the bitterness) [27], and therefore they confer the characteristic flavours and aromas of chocolate [26]. The results of total polyphenol, theobromine, caffeine, and lactic acid contents of unroasted cocoa bean samples from the different treatments studied are shown in Table 3. The total polyphenol content ranged from 16.6 to 19.3 mg gallic acid/g. These contents are in accordance with those reported by Batista et al [28] but slightly lower than those reported in beans by other authors (34-60 mg/g) [29]. These differences could be related to the effect that both genotypic characteristics and the geographical area of cocoa production have on the total polyphenol concentration of the cocoa bean [30]. In the present study, significant differences (p<0.05) between treatments were detected. Control and PD treatments showed higher contents (18.37 and 19.25 m/g, respectively) than YS treatment (16.62 m/g), and PD+YS showed intermediate contents (17.74 m/g). These results show that the pre-drying treatment, prior to the start of the cocoa fermentation process and carried out under natural conditions for 24 h, did not affect the final polyphenol content of the cocoa bean. Temperature is a determining factor in the degradation of polyphenols. Polyphenol content decreases with increasing temperature [5]. This may be caused by oxidation during the drying process. However, the temperatures reached during pre-drying in our study and the short time during which drying takes place did not seem to be sufficient to contribute to a decrease in the total polyphenol content. In contrast, inoculation of selected culture yeast on cocoa beans significantly decreased the total polyphenol content at the end of the fermentation process, when compared to those cocoa beans fermented spontaneously. In agreement with our results, several studies have reported an increase in the total phenolic content of cocoa spontaneously fermented compared to cocoa beans fermented after culture yeast inoculation [28]. This seems to be related to the effect that the starter culture may have on the microorganisms responsible for spontaneous fermentation. Although polyphenols are valuable antioxidants, their presence has been linked to astringency and bitterness in cocoa and cocoa products [27]. Consequently, the inoculation of cocoa beans with defined yeast starter cultures could affect the flavour quality of cocoa.

With respect to mehtylxanthines analysed, no significant differences were observed in theobromine content between treatments, with an average value of 7.38 mg/g. This content agrees with that reported by other authors [28]. The theobromine content in cocoa beans seems to be unaffected by cocoa fermentation conditions, including spontaneous fermentation or inoculation with starter cultures [28], in accordance with our findings. However, the caffeine content exhibited significant differences due to treatment, showing PD lower (p<0.05) content (0.68 mg/g) than Control, YS and PD+YS treatments (1.02, 0.93 and 0.98 mg/g, respectively). The present results showed that pre-drying treatment of cocoa beans in the sun for 24 h decreased the caffeine content. In contrast, inoculation of cocoa beans with a yeast culture did not affect the caffeine content. These results agree in part with other studies which indicated that caffeine content in cocoa beans is affected by genotype, processing conditions and spontaneous or inoculated fermentation [28]. Finally, the lactic acid amount varied between 0.60-1.40 mg/g, with Control showing higher content than YS, and this last one higher than PD and PD+YS treatments (p<0.001). These results evidenced that both PD and YS treatments cause a decrease in the lactic acid concentration of cocoa beans, as a consequence of a partial reduction of cocoa pulp, compared to traditional and spontaneous fermentation. In our study, the mucilage lost about 20% of the initial weight (data not shown) during the solar pre-drying period. This is in accordance with other results reported previously [10]. It has been well documented that excessive cocoa bean pulp leads to an increase in the sourness of the cocoa beans and thus a decrease in the flavour quality, mainly due to the high production of acid during fermentation [7]. Both treatment processes investigated in this study, solar pre-drying and yeast starter inoculation, are responsible for the reduction of part of the cocoa bean pulp by two ways-namely, mechanically and enzymatically, respectively [22]. Kongor et al. [31] reported that mechanical depulping by spreading beans onto a flat surface and exposing them to the sun for several hours leads to an increase in the sweating produced during fermentation. This is due to bruising of the beans and their cellular structures, which activates the cellular enzymes. In the same way, the reduction of bean acidity during fermentation in cocoa beans inoculated with a defined mixed starter culture, compared to those from spontaneous fermentation, could be related to the enzymatic activity of the yeasts [32].

4. Conclusions

This study showed that the chemical and fatty acid composition of cocoa beans were not affected by a pre-drying treatment and a subsequent fermentation process with a starter culture of selected yeasts, compared to the traditional process, by spontaneous fermentation, carried out by regional farmers in Tumaco-Nariño region in Colombia. Fermentation carried out after inoculation of a defined starter culture of yeasts caused a decrease in total polyphenol content, which could contribute to improving the flavour quality of cocoa. There was a noticeable decrease of lactic acid in cocoa beans. This was caused by fermentation treatment with selected yeasts and especially by a pre-drying treatment with sun. This would reduce the final acidity of the cocoa and improve the final quality of the cocoa manufactured into chocolate, given that excessive acidification of the cocoa beans during fermentation is associated with a lower sensory quality of the cocoa.