Effects of Fermentation Temperature on the Physicochemical Properties, Bioactive Compounds, and In Vitro Digestive Profile of Cacao (Theobroma cacao) Seeds

Abstract

This study investigates the impact of fermentation temperature on the physicochemical properties, bioactive compound retention, and in vitro digestion profile of cacao seeds (Theobroma cacao L.). Three fermentation conditions were evaluated: low (F40, 40 °C), medium (Control, 50 °C), and high (F60, 60 °C). The study assessed macronutrient composition, phenolic compound retention, antioxidant activity, enzymatic activity, structural changes, and glucose release during in vitro digestion. Fermentation temperature significantly influenced cacao seed quality and functionality. F40 preserved the highest levels of phenolic compounds (61% reduction compared to raw seeds) and antioxidant activity (73% reduction), offering a pronounced hypoglycemic effect through enzyme inhibition. In contrast, F60 facilitated extensive enzymatic activity, particularly protease and lipase, promoting flavor precursor formation and structural changes like cracking. However, this high-temperature treatment resulted in significant losses of phenolic compounds (76%) and antioxidant capacity (88%). Structural analysis revealed that higher fermentation temperatures enhanced cellular breakdown, increasing enzymatic access and glucose bioavailability. Digestion studies confirmed that roasted cacao fermented at higher temperatures released more glucose, driven by enzymatic hydrolysis and structural modifications. Conversely, the cacao from F40 exhibited slower glucose release due to the retention of bioactive compounds that inhibit carbohydrate-hydrolyzing enzymes. This research underscores the trade-offs in cacao processing: fermentation temperature significantly modulates cacao seed properties. At higher temperatures (60 °C), enhanced enzymatic activity (protease, lipase) facilitates the release of flavor precursors and structural modifications, increasing digestibility and glucose bioavailability, making it ideal for chocolate production. Conversely, fermentation at lower temperatures (40 °C) preserves bioactive compounds, including phenolics and antioxidants (with 61% retention compared to raw seeds), which may offer functional food applications for glycemic control. Roasting reversed some fermentation effects, reducing phenolic retention while increasing glucose bioavailability. This work tailors cacao fermentation for diverse end uses, from premium chocolate to nutraceutical products aimed at glycemic control.

1. Introduction

Cacao (Theobroma cacao L.) is a tropical crop cultivated in various world regions, primarily for chocolate production. It has three main varieties: Criollo, Forastero, and Trinitario, each with distinct characteristics. The Criollo variety, originating in Mexico and parts of South America, is prized for its delicate flavor and aroma. In contrast, Forastero, native to the Amazon, is more bitter and acidic, while Trinitario, a hybrid of both, exhibits intermediate sensory and agronomic traits [1].

The cacao seeds, which develop inside pods, reach optimal harvest two to three weeks after ripening. Each pod contains 30 to 40 seeds, typically 1 to 3 cm long and weighing 0.5 to 2.0 g [2,3,4,5]. These seeds are enclosed in a white mucilage layer (Figure 1A), constituting nearly 40% of the fresh bean’s weight. This mucilage is particularly rich in sugars (glucose, fructose, sucrose) and organic acids (citric acid, pectin, hemicelluloses), making it a crucial substrate for microbial activity during fermentation because it influences microbial succession and dictates the biochemical transformations that occur within the cotyledons during processing [6,7,8].

Once harvested, cacao seeds undergo a fermentation process that enhances their flavor precursors and reduces bitterness. During this stage, various fermentation techniques can be employed, including box fermentation, heap fermentation, and alternative methods such as basket or tray fermentation [8,9]. Although the primary objective is to remove mucilage and prevent seed germination, this step also promotes the breakdown of intrinsic macromolecules, developing chocolate flavor compounds [6,10].

Among these methods, wooden box fermentation, commonly practiced in regions like Chiapas, Mexico, is particularly efficient in temperature regulation. As depicted in Figure 1B, traditional fermentation boxes are stacked in a ladder-like structure, allowing seeds to be transferred between compartments every two to four days. This setup ensures progressive microbial activity, fostering optimal fermentation conditions. Furthermore, the use of “hoja blanca” (Calathea lutea) leaves as a cover provides natural insulation, helping to maintain fermentation temperatures [11,12]. Figure 1 presents a region-specific adaptation of the ladder-box fermentation system commonly used in Chiapas, Mexico. This setup ensures controlled temperature regulation and microbial succession, optimizing fermentation efficiency. Unlike standard fermentation methods, this approach facilitates staged seed transfer, allowing progressive aeration and better uniformity in enzyme activity and bioactive compound retention.

The fermentation process itself unfolds in distinct microbial phases. During the first 12–24 h, native yeast ferments the pulp’s sugars, producing CO₂; and ethanol (2.5%) under anaerobic conditions at a pH below 4. As oxygen permeates the system, lactic acid bacteria (LAB) thrive between 38 and 48 h, further metabolizing sugars and organic acids to generate lactic acid and ethanol. This microbial shift raises the temperature to approximately 40 °C, initiating enzymatic activity within the beans. By 72–96 h, the acetic acid bacteria (AAB) become dominant, oxidizing ethanol into acetic acid, an exothermic reaction that increases temperatures to nearly 50 °C by the fifth to seventh day [13,14].

As fermentation progresses, key biochemical and structural transformations occur inside the cacao cotyledons. Proteins (albumins and vicilins) are degraded by aspartic endoprotease (pH 3.5), generating hydrophobic peptides that later convert into hydrophilic peptides and free amino acids through carboxypeptidase activity (pH 5.8). Simultaneously, invertases hydrolyze sucrose into glucose, fructose, and mannitol, all contributing to forming aroma precursors. In parallel, polyphenol oxidase (PPO) catalyzes the oxidation of phenolic compounds, reducing astringency and bitterness while modulating antioxidant activity [15].

Fermentation also affects cacao lipids, which undergo structural rearrangements while maintaining their overall proportion. This leads to the formation of microstructural modifications such as holes and ridges, impacting the final texture of fermented cacao [16]. Ultimately, fermentation concludes with drying, a crucial step in stabilizing the seeds by preventing microbial overgrowth. Notably, the fermentation temperature exhibits a linear relationship with pH reduction, which plays a key role in activating crucial enzymes such as aspartic endoprotease, carboxypeptidase, and PPO [17].

Overall, fermentation is a critical determinant of cacao quality, influencing sensory characteristics, structural integrity, and bioactive compound retention. Understanding the intricate relationships between temperature, enzymatic activity, and microbial succession allows for precise optimization of fermentation conditions to enhance cacao’s suitability for different industrial applications, from fine-flavor chocolate to functional food products.

2. Materials and Methods

2.1. Sample Collection and Inoculation Process

Mature cacao pods (Trinitario variety) were harvested in November 2023 from the RAYEN Cooperative in Tapachula, Chiapas, Mexico, selecting pods with no external damage; shell colors varying between green, yellow, and reddish tones; and lengths of 12–19 cm. The pods were transported to the laboratory under refrigeration (4–8 °C) to preserve native microorganisms. In the laboratory, the pods were opened using sterile gloves and tools, and 1500 g of seeds were weighed and placed into a sterile 45 × 35 cm polyethylene bag. Pieces of the pod shells, which harbor native microorganisms, were cut into small sections (approximately 2–3 cm) and added to the bag to inoculate the seeds. The bag was sealed to maintain moisture and incubated under controlled temperature conditions (≈30 °C); after 24 h, the pod shells were removed from the bags.

2.2. Fermentation Profiles

Three fermentation treatments were carried out to evaluate the effect of temperature on cacao seed composition, enzymatic activity, and bioactive compound retention. The Control treatment was designed to emulate the fermentation conditions observed in the RAYEN cooperative, where the fermentation temperature remains at ~30 °C after 72 h, gradually increases to 40 °C by 96 h, and reaches 50 °C by the end of fermentation at 144 h (Figure 2).

To examine the impact of lower and higher fermentation temperatures, two experimental treatments were implemented, as detailed below.

F40 (low-temperature treatment): The fermentation temperature was maintained at 40 °C from 72 h until 144 h to assess the effects of a milder temperature profile on bioactive compound retention and enzymatic activity.

F60 (high-temperature treatment): The temperature gradually increased from 72 h, reaching 60 °C at 120 h, and remained at this level until the end of fermentation at 144 h. This condition was selected to study the impact of high-temperature fermentation on structural modifications, enzyme activity, and flavor precursor formation.

The selection of these temperature profiles, rather than intermediate values (e.g., 45 °C or 55 °C), was based on the need to maximize the contrast between distinct fermentation conditions. Lower temperatures (40 °C) help preserve bioactive compounds, while higher temperatures (60 °C) promote enzymatic hydrolysis and seed matrix transformation. Intermediate temperature gradients might have resulted in incremental changes, making it difficult to discern the specific effects of mild- vs. high-temperature fermentation.

To ensure homogeneity in the fermentation conditions, two aeration stages were conducted at 48 and 96 h, involving a bag change to replicate the ladder-box fermentation setup used in traditional processes. A temperature-controlled incubator (Thermo-Scientific, 32100A, Waltham, MA, USA) maintained a ±3 °C temperature range, ensuring experimental precision. Seed samples (100 g) were collected at 0, 72, and 144 h for analysis.

2.3. pH of Pulp and Cotyledons

The pH of the pulp was measured at 0, 72, and 144 h; for this, a pH probe (HM Digital, PH-2000, Henderson, NV, USA) was used, which was inserted into three different points of the polyethylene bag, verifying that it had direct contact with the surface of the beans. The pH of the cotyledons was measured from detached seeds. For this, 10 g of pure cotyledon was ground in 90 mL of distilled water, and the measurement was taken.

2.4. Chemical Analysis

The chemical composition of the cacao samples was analyzed using standardized methodologies. Moisture content was determined by oven-drying at 105 °C until constant weight. Total fat content was measured by Soxhlet extraction, using hexane as the solvent. Protein content was assessed using the Kjeldahl method, which applied a nitrogen-to-protein conversion factor of 6.25. Total carbohydrates were calculated by difference, subtracting the sum of moisture, fat, protein, and ash from the total weight according to AOAC methods 972.20, 968.06, 2001.11, and 923.03, respectively (AOAC, 2025) [18].

2.5. Free Sugars

Cacao samples were first freeze-dried for 48 h to remove residual moisture while preserving bioactive compounds. The dried samples were then ground using a stainless steel blade mill (IKA A11 Basic, Staufen im Breisgau, Germany), operating at 18,000 rpm for 30 s per cycle, with a 5-min cooling interval between cycles to prevent heat degradation. The resulting fine powder was passed through a 60-mesh (250 µm) sieve to obtain a uniform particle size suitable for chemical and structural analyses. The homogenate of 50 mg of the pulverized sample with nanopure water was centrifuged at 10,000× g for 15 min, and the supernatant was filtered through a 0.45 µm nylon membrane. The filtrate was subjected to high-performance liquid chromatography (HPLC) analysis using a Waters HPLC system equipped (Milford, MA, USA) with a MetaCarb 87C column and a guard column. Deionized water served as the mobile phase, and the flow rate was maintained at 0.5 mL/min. Soluble sugars were detected using a Waters 410 differential refractometer detector (Milford, MA, USA) and quantified with Empower 3.7 software. Commercial fructose, glucose, and sucrose standards were used for calibration; the retention times were approximately 8.5, 9.7, and 10.3 min, respectively [19].

2.6. Fatty Acid Determination

The fatty acid composition was determined using a gas chromatograph. Fatty acid methyl esters (FAMEs) were identified by comparison with the retention times of authentic standards. The lipidic extract obtained with chloroform was converted into fatty acid methyl esters (FAMEs) by esterification, according to the AOAC 963–22 method [20], and subsequently injected into the chromatography system. The equipment conditions were those used by Moreira and Mancini Filho (2004) [21] with a TSQ DUO GC system, and the result was calculated using Chromeleon V 3.1 (Fischer, Wisconsin, WA, USA) software. The column used was SUPELCOWAX 10 fused silica (polyethylene glycol, 30 × 0.25 × 0.25 µm). The temperature gradient started at 170 °C and increased to 225 °C at a rate of 1 °C/min, the vaporization temperature was 250 °C, the detector temperature was 270 °C, and helium (He) was used as the carrier gas at a flow rate of 1 mL/min, with a distribution ratio of sample to injector of 1/150. Finally, Sigma Aldrich standards (St. Louis, MO, USA) were used for comparison.

2.7. Free Amino Nitrogen (FAN)

The Ninhydrin method was used [22]. For this, 2 mL of defatted cacao seeds with distilled water were placed in test tubes, other tubes with known amounts of glycine were placed, and 1 mL of ninhydrin reagent at pH 6.6 was added. The mixtures were placed in a water bath for 15 min and cooled to 20 °C. After 20 min, the absorbance at 570 nm was measured against a reagent blank in a spectrophotometer (Genesys 10 S, Maryland, MD, USA), and the absorbance calculated from the sample tests was divided by the measurement in the assays with the glycine solution and multiplied by the dilution factor.

2.8. Total Phenolic Compounds

The defatted cocoa bean was mixed with an 80% methanol solution and magnetically stirred for 2 h at 50 °C with a stirrer. Then, the sample was passed through Whatman No.1 filter paper, and the extract was capped in a volumetric flask. The Folin–Ciocalteu method was used for phenolic compound determination, where a 100 mL beaker was used to mix, in order, 1 mL of diluted extract, 50 mL of deionized water, 5 mL of Folin reagent, and 20 mL of a volumetric 20% (w/v) sodium carbonate solution. After 30 min, the absorbance was measured at a wavelength of 750 nm (Genesys 10 S, Maryland, MD, USA). The standard curve was determined with gallic acid (GA) in 0 to 600 mg/L concentrations.

2.9. Antioxidant Activity

A mixture of 60 µM of a methanolic solution of DPPH (2940 µL) with 60 µL of extract in a polystyrene cuvette was used to measure the absorbance at 515 nm on two occasions, before adding the extract and 60 min after adding it, with a spectrophotometer (Genesys 10 S, Maryland, MD, USA). The standard curve was determined with Trolox, and the results are expressed as equivalents in mmol.

2.10. Enzyme Activity

Enzyme activity was measured using Sigma Aldrich kits. Protease uses casein as a substrate; when consumed, the amino acid tyrosine is released and later reacts with Folin’s reagent, resulting in a blue chromophore quantified in the spectrophotometer (LIBDH-RO kit, Chnspec Technology, Hangzhou, China). The lipase kit follows the Bulletin technique, which consists of a couple of reactions that result in a colorimetric product proportional to the enzyme’s activity using glycerol as the standard (MAK482). Moreover, amylase activity was assessed under the reaction of alpha-amylase hydrolyzing starch to produce maltose (MAK478).

2.11. Seed Structural Analysis

To quantify the size and percentage of the area that was cracked in the cotyledons of the bean through fermentation, 10 beans were selected, which were divided in half using a knife, and a photograph was taken in a leaflet. The ImageJ (v. 1.38e, Java, USA) program was used to process the images, convert them into 8-bit images, and select the sink area to apply thresholds and quantify it.

2.12. In Vitro Available Carbohydrate Digestibility

The protocol of Grandfeldt et al. [23] was employed in dried and roasted samples. Briefly, they were incubated in boiling water for 30 min with constant stirring. Afterward, porcine pancreatic α-amylase was used, and the hydrolyzed starch was measured at 0, 30, 60, 90, 120, and 180 min with glucose oxidase/peroxidase reagent. From the calculated area under the starch hydrolysis curve, the hydrolysis index (HI) was calculated and compared against a fresh white bread sample that was assumed to consist of 100% digestion-available starch. The Predicted Glycemic Index (pGI) was calculated from this value with the following equation, reported by Goñi et al. [24]:

pGI = 39.71 + 0.540 (HI)

(1)

2.13. In Vitro Protein Digestibility

The protocol of Hsu et al. [25] was used to estimate the protein digestibility of the cacao samples. First, 50 mL of an aqueous suspension of the materials, considering a protein amount of 6.25 mg of protein/mL, was prepared. The solution was adjusted to pH 8.0 with 0.1 N HCl or NaOH. A multienzyme solution of trypsin at 1.6 mg/mL (15 units/mg), chymotrypsin at 3.1 mg/mL (60 units/mg), and peptidase at 1.3 mg/mL (40 units/mg) was adjusted to pH 8.0 and maintained in an ice bath until use. The multienzyme solution (5 mL) was added to the protein suspension and incubated at 37 °C in a water bath with continuous magnetic stirring for 10 min. The pH was monitored for 10 min, and the recorded values were used to estimate the in vitro protein digestibility.

2.14. Statistical Analysis

All procedures were performed in triplicate unless otherwise specified. Data from these replicates were analyzed using one-way ANOVA and Tukey’s post-hoc test (p ≤ 0.05), ensuring statistical significance and reproducibility. Standard deviations are reported to reflect experimental variation. Additionally, a principal component analysis (PCA) was conducted to evaluate the interactions between variables. Statistical analyses were conducted using Minitab (version 19.20201.0, Minitab, LLC, State College, PA, USA).

3. Results and Discussion

3.1. Effect of Fermentation Dynamics on Seed Characteristics

Figure 3 illustrates the dynamics of the crackle area, pulp, and cotyledon pH during cacao fermentation. Across all treatments, the crackle area progressively increased as fermentation progressed, with the F60 treatment exhibiting the most significant crackle percentage by the sixth day, driven by the higher temperature promoting enzymatic activity and cellular breakdown (Figure 3A). The Control treatment demonstrated moderate structural changes, while F40 showed minimal cracking, which correlated with lower enzymatic and microbial activity under milder temperatures [15,26]. The pulp pH initially decreased in all treatments due to the activity of microorganisms such as yeasts and lactic acid bacteria consuming citric acid and producing organic acids [6,27]. By the end of fermentation, the pulp pH stabilized in the Control and F40 treatments (Figure 3B). In contrast, the cotyledon pH declined significantly in the F60 treatment due to intensified enzymatic breakdown of proteins into free amino nitrogen (FAN) and other flavor precursors (Figure 3C). The pulp pH ranged between 3 and 3.5 outside the bean due to the citric acid content, creating an ideal environment for yeast development [28,29,30]. Previous studies have shown that specific strains, such as Pichia kluyveri and Kluyveromyces marxianus, hydrolyze pectin, facilitating oxygen diffusion and pH increases as citric acid is consumed [31,32]. Internally, the cotyledon pH in the Control and F60 treatments dropped from 6.4 to close to 5, characteristic of fermented cacao, whereas in F40, it remained at 5.73 by the end of the treatment.

3.2. Cacao Seed Proximal Analysis

The chemical composition changes observed in the cacao samples during fermentation and processing are presented in

Table 1. Chemical composition of fermented cacao seeds.

Sample	Moisture (%)	Protein (%)	Lipids (%)	Carbohydrates (%)	Free Sugars (%)	Ash (%)
Control.0	65.23 ± 0.69 a	17.23 ± 0.36 a	53.23 ± 0.59 a	24.21 ± 0.64 a	20.11 ± 0.62 a	5.33 ± 0.56 a
Control.72	63.45 ± 0.05 a	16.52 ± 0.96 a	54.21 ± 0.07 a	23.81 ± 0.85 a	19.82 ± 0.58 a	5.46 ± 0.06 a
Control.144	62.73 ± 0.78 a	16.56 ± 0.48 a	53.98 ± 0.43 a	24.25 ± 0.00 a	15.33 ± 0.23 c	5.21 ± 0.32 a
Control.D	10.51 ± 0.30 b	15.36 ± 0.16 a	53.41 ± 0.85 a	25.57 ± 0.32 a	11.43 ± 0.70 dd	5.66 ± 0.90 a
Control.R	6.23 ± 0.84 c	15.55 ± 0.93 a	54.26 ± 0.67 a	24.41 ± 0.47 a	9.86 ± 1.00 d	5.78 ± 0.42 a
F40.0	65.57 ± 0.76 a	16.54 ± 0.01 a	50.57 ± 0.72 c	23.24 ± 0.66 a	18.10 ± 0.32 b	5.12 ± 0.40 a
F40.72	64.61 ± 0.36 a	15.52 ± 0.30 a	50.96 ± 0.18 c	23.91 ± 0.28 a	18.23 ± 0.13 b	5.24 ± 0.43 a
F40.144	63.97 ± 0.47 a	15.56 ± 0.42 a	48.04 ± 0.41 d	23.04 ± 0.07 a	14.72 ± 0.64 c	5.00 ± 0.81 a
F40.D	9.18 ± 0.67 b	14.28 ± 0.32 b	50.74 ± 0.31 c	23.27 ± 0.66 a	10.52 ± 0.31	5.43 ± 0.86 a
F40.R	5.68 ± 0.28 c	14.13 ± 0.28 b	51.55 ± 0.03 b	22.72 ± 0.47 a	9.27 ± 0.31 d	5.49 ± 0.14 a
F60.0	64.27 ± 0.54 a	17.23 ± 0.08 a	54.23 ± 0.37 ba	23.00 ± 0.17 a	19.51 ± 0.04 a	5.43 ± 0.09 a
F60.72	62.71 ± 0.74 a	15.03 ± 0.34 a	54.31 ± 0.31 a	23.10 ± 0.23 a	19.23 ± 0.71 a	4.31 ± 0.11 a
F60.144	62.21 ± 0.55 a	14.57 ± 0.12 b	54.21 ± 0.89 a	23.52 ± 0.92 a	14.56 ± 0.17 c	3.39 ± 0.20 b
F60.D	10.09 ± 0.32 b	14.28 ± 0.72 b	52.14 ± 0.22 b	22.80 ± 0.73 a	11.09 ± 0.89 d	3.55 ± 0.34 b
F60.R	6.04 ± 0.38 c	14.15 ± 0.49 b	51.12 ± 0.77 b	22.68 ± 0.04 a	9.47 ± 0.81 d	2.97 ± 0.00 b

F40: fermentation treatment set at 40 °C; F60: fermentation treatment set at 60 °C; 0, 72, and 144, the sampling hours; D: dried; R: roasted. Different letters in the same columns indicate statistically significant differences (α = 0.05).

. The moisture content significantly declined from the raw state (65.23% in Control.0) to the roasted stage (6.23% in Control.R). Such processes enhance shelf stability and reduce water activity. Across treatments, F40 and F60 followed similar moisture reduction trends, with roasted values of 5.68% and 6.04%, respectively. This decrease is essential for preventing microbial spoilage and ensuring long-term quality. The protein content experienced a moderate reduction, with initial values of 17.23% in raw samples (Control.0 and F60.0), decreasing after roasting to 14.15% in F60.R and 14.93% in F40.R. This could be attributed to proteolytic activity during fermentation, which generates free amino nitrogen (FAN), a precursor for flavor compound synthesis. Among the treatments, F60 exhibited the highest protein loss (15.4%), suggesting that elevated temperatures during fermentation enhance enzymatic activity, accelerating protein breakdown and FAN production. This trend aligns with prior studies emphasizing FAN’s role in flavor development during cacao fermentation [33,34]. The lipid content, crucial for cacao fat quality and chocolate production, remained relatively stable across most treatments, such as in Control samples (53.23% in Control.0 and 54.26% in Control.R). However, F40 showed a notable lipid decline during fermentation (50.57% in F40.0 to 48.04% in F40.144), possibly due to incomplete cell structure destabilization under milder conditions. This effect may cause lipids to form hydrophilic dispersions, limiting their availability and extraction efficiency [17]. By contrast, the F60 and Control treatments maintained lipid integrity, supporting the hypothesis that traditional fermentation temperatures preserve lipid proportions while influencing molecular conformation [35,36]. The carbohydrate content demonstrated a dynamic pattern, increasing during fermentation due to polysaccharide breakdown and sucrose hydrolysis by enzymes like invertases and α -amylases, providing substrates for microbial activity. However, carbohydrate levels declined significantly during roasting, from 24.21% in Control.0 to 24.41% in Control.R and from 23.00% in F60.0 to 23.68% in F60.R, driven by Maillard reactions. These reactions are critical for developing chocolate’s characteristic aroma and flavor complexity [16,37]. Free sugars followed a similar trend, with a decline during fermentation and a more pronounced reduction during roasting. For instance, free sugars in Control samples decreased from 20.11% (Control.0) to 9.86% (Control.R), while F40 and F60 showed similar trends, indicating sugar utilization by microbes during fermentation and thermal degradation during roasting. The reduction in sugars highlights their conversion into essential flavor precursors through fermentation and Maillard reactions. The ash content, representing the mineral fraction, remained relatively stable throughout fermentation and slightly increased in roasted samples due to moisture loss and mineral concentration. For instance, the ash content rose from 5.33% in Control.0 to 5.78% in Control.R, with similar patterns observed in the F40 and F60 treatments. The F40 treatment provides a controlled environment for balanced flavor development, while F60 accelerates biochemical reactions, enhancing flavor precursor formation but with potential trade-offs in lipid content stability. The stable lipid profile, reduced free sugars, and increased flavor complexity show the importance of understanding these transitions to optimize cacao for diverse products or applications.

3.3. Fatty Acid Determination

The consistent presence of lipids during the fermentation and drying stages reflects their high stability, which is critical for maintaining cocoa butter quality—a key determinant of chocolate’s texture and melting behavior [33,38,39]. As shown in Figure 4, the fatty acid profiles of the cacao seeds across the different fermentation temperature treatments revealed that oleic acid consistently accounted for the highest proportion of total fatty acids (40.33% for Control and ≈38% for F40 and F60), followed by stearic acid and palmitic acid as the next most abundant fatty acids. The lipid composition, particularly the balance between saturated fatty acids (e.g., palmitic and stearic acids) and the unsaturated fatty acid oleic acid, is critical for achieving the desired properties in the final chocolate product. Saturated fatty acids, such as palmitic acid (melting point: 62 °C) and stearic acid (melting point: 68 °C), contribute to chocolate’s firmness and solid state at room temperature. Meanwhile, the lower melting point of oleic acid (16 °C) ensures a smooth melting experience in the mouth, creating a sensory profile highly valued by consumers [33,40]. The stability of the fatty acid profile across treatments also reflects the limited impact of fermentation temperature on the structural integrity of cacao lipids. This aligns with the earlier observation in Table 1, where the lipid content remained relatively unchanged in most cases. However, a slight reduction in lipid content was observed in F40 during fermentation, which may be related to hydrophilic lipid dispersions under milder conditions.

3.4. Free Amino Nitrogen (FAN), Phenolics, and Antioxidants

The phenolic content and antioxidant activity declined significantly throughout these stages (

Table 2. Antioxidant profiles of fermented, dried, and roasted cacao seeds.

Sample	Phenolics (mg/g)	Antioxidant Activity (μM Trolox eq)	Cracking Area (%)	FAN (mg/g)
Control.0	180.00 ± 4.41 a	795.00 ± 5.00 a	2.83 ± 0.96 f	3.36 ± 0.39 c
Control.72	86.00 ± 3.08 c	243.00 ± 6.10 c	16.76 ± 0.40 e	8.33 ± 0.91 b
Control.144	42.00 ± 8.38 e	196.00 ± 1.41 f	39.7 ± 1.50 b	11.41 ± 0.38 a
Control.D	33.14 ± 5.33 f	112.00 ± 4.51 h	ND	11.26 ± 0.69 a
Control.R	22.32 ± 4.69 g	85.00 ± 10.26 j	ND	9.31 ± 0.75 b
F40.0	181.3 ± 5.35 a	790.00 ± 10.03 a	2.74 ± 0.16 f	3.29 ± 0.64 c
F40.72	101.12 ± 12.02 b	291.00 ± 2.68 b	17.05 ± 0.03 e	7.58 ± 0.54 b
F40.144	66.43 ± 7.50 d	204.00 ± 1.65 e	32.24 ± 1.14 c	10.38 ± 0.23 b
F40.D	54.23 ± 6.40 e	188.00 ± 5.22 g	ND	10.02 ± 0.05 b
F40.R	33.14 ± 8.76 f	101.00 ± 4.36 i	ND	8.38 ± 0.89 b
F60.0	179.14 ± 7.96 a	797.00 ± 2.16 a	2.81 ± 0.33 f	3.36 ± 0.16 c
F60.72	84.33 ± 11.20 c	214.00 ± 9.38 d	19.67 ± 0.12 d	10.75 ± 0.40 b
F60.144	22.41 ± 6.03 g	91.00 ± 4.85 j	48.96 ± 0.31 a	12.50 ± 0.65 a
F60.D	15.21 ± 4.17 g	77.13 ± 2.60 j	ND	11.21 ± 0.31 a
F60.R	8.25 ± 1.36 h	31.14 ± 7.13 k	ND	11.57 ± 0.85 a

F40: fermentation treatment set at 40 °C; F60: fermentation treatment set at 60 °C; 0, 72, and 144, the sampling hours; D: dried; R: roasted; FAN: free amino nitrogen; ND: not detected. Different letters in the same columns indicate statistically significant differences (α = 0.05).

). The initial phenolic content was highest in the raw cacao samples (≈180 mg/g) but decreased markedly during fermentation, with roasted samples showing drastic reductions (22.32 mg/g in Control.R). The antioxidant activity followed a similar trend, dropping from 795 μM Trolox eq in Control.0 to just 85 μM Trolox eq in Control.R, primarily due to polyphenol breakdown during fermentation and thermal degradation during roasting [41]. Despite this general loss, the F40 treatment retained more phenolic compounds than the Control or F60 treatments, with a 61% reduction compared to 76% in the Control, as the lower pH observed in F40 inhibits cell wall breakdown, aiding in polyphenol retention. The antioxidant capacity decreased similarly, with reductions of 75% for the Control treatment, 73% for F40, and 88% for F60, showing the thermosensitivity of antioxidant compounds. Interestingly, F40 showed better preservation of antioxidant activity relative to its phenolic loss; this could be due to the presence of catechin and epicatechin condensation into tannins with proteins and flavonoids, which, while not identified within phenolic compounds, retain antioxidant potential [42,43]. The cracking area progressively increased during fermentation, peaking in F60.144 (48.96%). This is due to enzymatic activity and structural transformations in the cotyledons, facilitated by high temperatures that promote cell wall breakdown. The extent of cracking reflects seed fermentation and flavor precursor release, which is critical for chocolate production [17]. Treatments such as F60, emphasizing high-temperature fermentation, promote extensive structural modifications at the expense of phenolic retention. The FAN levels increased during fermentation, peaking in F60 (12.50 mg/g), indicative of heightened proteolytic activity and the release of hydrophobic amino acids due to the action of exoproteases like carboxypeptidase enzymes [17]. F40 showed the lowest FAN levels (10.38 mg/g), reflecting milder proteolytic activity. The FAN levels plateaued or decreased during roasting (11.57 mg/g in F60.R), reflecting the thermal degradation of amino acids.

3.5. Enzymatic Activity

The enzymatic activity profiles reveal significant variations influenced by temperature conditions and fermentation progress. Amylase activity peaked during the intermediate fermentation stages in F40 (Figure 5A), correlating with its optimal temperature range of 30–40 °C [15]. This activity facilitates the hydrolysis of starch or dextrins into fermentable sugars, which microorganisms metabolize to produce flavor precursors. In the F60 treatment, amylase activity significantly decreased due to the elevated temperature, exceeding the enzyme’s optimal functional range. In contrast, moderate amylase activity persisted in the Control treatment (50 °C), balancing sugar breakdown and preserving complex carbohydrate structures. Notably, amylases from cacao exhibited their highest activity in F40, measured at 53.31 μmol maltose/min⁻¹ by the end of fermentation, aligning with their reported maximum activity temperature of 37 °C [44]. Conversely, F60 showed the lowest activity at 21.16 μmol maltose/min⁻¹, with the Control treatment registering 43.41 μmol maltose/min⁻¹. Despite this, amylase has been less studied in cacao due to its limited accessibility to starch substrates and potential inhibition during fermentation, drying, and roasting. Protease activity was markedly higher in F60 during the later stages of fermentation (Figure 5B), aligning with its higher temperature conditions [8,26]. This increased activity leads to extensive protein degradation, producing free amino nitrogen (FAN), a key precursor for aroma and flavor compounds. The FAN levels were highest in F60, reflecting the activity of carboxypeptidase, a highly active exoprotease at elevated temperatures. By comparison, F40 showed the highest protease activity among all treatments at moderate temperatures (30–40 °C), suggesting that its microenvironmental conditions were optimal for protease enzymes. The Control treatment supported moderate protease activity but lagged behind F40, likely due to the slightly higher temperatures. The lipase activity demonstrated a more consistent profile across treatments, peaking prominently in F40 (Figure 5C). This activity is well-suited to moderate temperatures, leading to the release of free fatty acids, which are crucial for flavor development during fermentation and roasting [34,45]. The elevated temperature in F60 inhibits lipase activity, potentially through enzyme denaturation, limiting fatty acid release and impacting flavor quality. F40′s conditions enable optimal lipase functionality, highlighting its importance in flavor formation; for this, we hypothesize that moderate temperatures favor the enzymatic processes essential for flavor and bioactive compound retention, while high temperatures drive faster biochemical transformations suited for industrial-scale chocolate production.

3.6. In Vitro Carbohydrate and Protein Digestion

Figure 6A shows the glucose release kinetics during in vitro digestion of the cacao samples. The Control, F40, and F60 fermentation treatments show that higher fermentation temperatures (F60) enhance glucose release due to structural modifications like cell wall disruption and increased enzymatic accessibility [17]. Roasting further degrades complex carbohydrates into soluble sugars, enhancing sweetness and digestibility, but reduces phenolic compound retention, which diminishes inhibitory effects on carbohydrate-hydrolyzing enzymes [41]. In contrast, lower-temperature fermentations (Control, F40) retain more phenolic compounds, which inhibit enzymes like α-amylase and α-glucosidase, reducing glucose absorption and providing potential hypoglycemic benefits [44,46]. This phenolic retention aligns with functional food applications for glycemic control. In contrast, higher-temperature fermentation and roasting are better suited for flavor-driven products like chocolate, as they enhance glucose release and sweetness [44]. In addition to carbohydrate digestibility, the protein digestion profiles (Figure 6B) highlight how fermentation and processing influence protein bioavailability in cacao samples. Across the treatments, the dried samples (F40.D, F60.D) demonstrated slightly higher protein digestibility than their roasted counterparts (F40.R, F60.R), likely due to the preservation of native protein structures in dried conditions. While beneficial for flavor development, roasting may induce Maillard reactions and protein cross-linking, reducing protein digestibility. Among all treatments, samples subjected to higher fermentation temperatures (F60) showed enhanced protein digestion, indicating that higher temperatures may facilitate protein denaturation and increased enzymatic access. Casein was used as a reference standard and demonstrated the highest protein digestibility, serving as a benchmark for comparison [2,3]. High-temperature treatments could enhance bioavailability and reduce bioactive compound retention, emphasizing the need for tailored processing strategies based on whether the target application prioritizes flavor, nutritional value, or functional benefits.

3.7. Data Processing by Multivariate Analysis

The first dendrogram group observations were based on physicochemical properties, bioactive compound retention, enzymatic activities, and stages of fermentation and processing (Figure 7A,B). Notably, samples are clustered by processing stage rather than fermentation conditions, emphasizing the significant impact of post-fermentation processes, such as roasting, on the final product’s characteristics. For instance, roasting reduces phenolic content and antioxidant activity, driving the clustering of roasted samples. Similarly, F60 samples show distinct grouping, driven by higher enzymatic activities and greater structural changes, as reflected in the increased cracking area and free amino nitrogen (FAN) levels. The second dendrogram highlights variable correlations, grouping moisture, total sugars, phenolics, and antioxidant activity, aligning with their similar decline during the fermentation and roasting stages. In contrast, enzymatic activities such as protease and lipase activities form distinct clusters, underscoring their critical and independent roles in flavor precursor development and lipid transformation. These findings align with previous observations in the manuscript that phenolic retention and antioxidant activity are higher at F40, while F60 promotes enzymatic hydrolysis and flavor enhancement. This statistical analysis reinforces the trade-offs in cacao processing: lower fermentation temperatures retain bioactive compounds, while higher temperatures and roasting enhance flavor precursors, which are critical for chocolate production [17,46].

4. Conclusions

This study underscores the critical influence of fermentation temperature on cacao’s physicochemical properties, affecting seed structure, enzymatic activity, polyphenol retention, and pH evolution—key determinants of quality for chocolate and functional foods. Fermentation at 40 °C (F40) preserved bioactive compounds by minimizing enzymatic degradation, whereas fermentation at 60 °C (F60) accelerated structural changes and flavor precursor formation, enhancing sensory characteristics. The Control treatment provided a balanced outcome, reflecting traditional practices. The observed correlation between temperature, pH, and enzymatic activity highlights the importance of optimizing fermentation conditions based on the desired end-product, whether prioritizing functional properties or sensory enhancement.