Application of Commercial Pectinase as a Biocatalyst During Self-Induced Anaerobic Fermentation of Coffee (Coffea arabica L. var. Typica)

Abstract

This study investigated the impact of enzyme treatment on the physicochemical parameters and volatile and bioactive composition of Arabica coffee beans during self-induced anaerobic fermentation (SIAF). The physicochemical parameters of the beans treated with the enzyme solution were monitored over a 120 h fermentation period. The results showed that increasing enzyme concentration reduced the levels of reducing sugars and phenolic compounds, leading to decrease in antioxidant activity. Pectin lyase activity was highest in beans treated with 10 U.·mL⁻¹, while polygalacturonase activity fluctuated throughout fermentation. The highest caffeine content (722.09 ± 3.7 mg·100g⁻¹) was found in beans treated with 5 U.mL⁻¹ after 72 h of fermentation. In contrast, trigonelline (1028.75 ± 31.4 mg·100g⁻¹) and 5-O-caffeoylquinic acid (5CQA) (423.46 ± 40.3 mg·100g⁻¹) were more prominent in unfermented beans. Volatile formation showed a positive correlation with enzyme concentration, with beans treated with 10 U·mL⁻¹ exhibiting a more diverse volatile profile in the first 24 h. These findings suggest that enzymatic treatment modulates coffee’s volatile and bioactive composition, enhancing levels of aromatic compounds that are directly linked to the sensory quality of the coffee beverage.

1. Introduction

Pectin is a heteropolysaccharide composed of monomeric units of galacturonic acid, rhamnose, arabinose, and galactose, linked to structural proteins found in the cell walls of fruits and vegetables [1]. In coffee, it is present in the mucilage (1–3%) as a complex polysaccharide and serves as a substrate for fermentation processes [2]. During fermentation, pectin degradation occurs through the action of microbial pectinases, particularly polygalacturonase and pectin lyase, which are essential for coffee fermentation and are often used as key parameters for selecting starter cultures [3].

Polygalacturonase catalyzes the hydrolysis of α1-4 glycosidic bonds in D-galacturonic acid, leading to pectin depolymerization and the softening of plant tissue [4]. Additionally, pectin lyase cleaves these α1-4 glycosidic bonds via β-elimination [5]. The action of these enzymes during coffee fermentation serves the functional purpose of removing the mucilage layer and reducing bean moisture by inhibiting fungal growth [6]. Furthermore, enzymatic hydrolysis releases aromatic compounds such as methoxypyrazines and terpenes, key contributors to primary aromas [3].

In this context, coffee fermentation is based on the action of enzymes responsible for physical and chemical changes so that enzyme treatment can contribute to bioprocessing development. Thus, coffee fermentation can be improved by changing processing conditions and has aroused the interest of producers due to its commercial value in the international market [7]. Therefore, several technologies improving coffee fermentation have been studied, yielding promising results. While the selection of starter cultures remains a widely explored avenue, it is still evolving. Other approaches, such as carbon supplementation, pH and temperature adjustments, oxygen demand control, and sugar addition, have been mentioned but remain underexplored [3].

Simultaneously, the use of exogeneous enzymes in the fermentation medium is an emerging technology. Tai et al. [8] assessed the impact of adding polygalacturonase and feruloyl esterase, derived from Aspergillus tubingensis, on the quality of Robusta coffee beans subjected to submerged fermentation. Their study found that enzyme addition reduced fermentation time and improved flavor. Similarly, Elhalis et al. [9] observed that fermentation of pectinases in wet coffee significantly shortened the mucilage degradation time due to the action of polygalacturonase and pectin lyase. According to Febrianto and Zhu [10], adding enzymes to the coffee fermentation process modifies the chemical composition and sensory quality, particularly through enhanced volatile profiles. Thus, with coffee mucilage being a matrix rich in pectin, treatment with pectinases reduces fermentation time, favors the synthesis of volatile compounds, and improves the quality of the coffee beverage.

Currently, no studies are present in the literature on the addition of enzymes as a catalyst for the solid-state fermentation of coffee. However, this technology has advantages such as the possibility of using the whole fruit, dispensing with the pulping step, and adding by spraying, which significantly reduces the amount of water required for the process. Based on this background, this study aims to explore the effects of commercial pectinase on the solid-state fermentation of Arabica coffee, focusing on its impact on the volatile and bioactive composition of the beans.

2. Materials and Methods

2.1. Coffee Sample

Coffee fruits (Coffea arabica L.), of the variety ‘Typica’, were harvested at an altitude of 850 m above sea level from the Varzea Grande and Florentina family farms in Taquaritinga do Norte (7°54′56.2″ S, 36°20′59.2″ W), Pernambuco (PE), Brazil. Harvesting was performed manually, followed by a washing stage in which the fruits were submerged in water. During this process, impurities and floating fruits remained on the surface, facilitating their removal, while the denser green and ripe fruits settled at the bottom of the container. Afterward, immature fruits were removed by manual selection, and only the ripe fruits, with an average of 17 °Brix, were directed to the fermentation process.

2.2. Coffee Fermentation Processing

Approximately 100 g of pre-selected coffee cherries (coffee fruits with similar stages of maturity and sizes) was used for each treatment, including fermented coffee without enzyme application. In the enzymatic treatment, the samples were sprayed with 10 mL of a commercial enzyme preparation derived from Aspergillus aculeatus at concentrations of 1, 5, and 10 U·mL⁻¹ (Pectinex Ultra SP-L (P2611), Sigma-Aldrich, St. Louis, MO, USA). The samples were manually shaken to ensure uniform contact between the enzymatic solution and the entire surface of the cherries. A volume of 10 mL of enzymatic solution was sufficient to avoid liquid accumulation inside the container, thus maintaining the characteristics of solid-state fermentation. Coffee cherries with and without enzyme treatment were packed into cylindrical polyethylene containers with a capacity of 0.5 L (diameter: 8 cm; height: 17 cm), resulting in the cherries occupying approximately 10 cm in height, leaving a headspace of 7 cm. The containers were carefully sealed and equipped with an airlock valve to release CO₂ and a thermo-hygrometer to monitor the temperature and moisture levels of the fermentation environment. The process was evaluated every 24 h over 120 h of fermentation.

2.3. Drying, Roasting, and Grinding Coffee

The fermented coffee beans were dried in an oven with air circulation at 40 °C until they reached 11% moisture content, followed by removing the husks and remaining pulp to obtain green beans (dry, pulped, and unroasted endosperm). Roasting was conducted at temperatures between 180 and 190 °C for 8 min to achieve a medium roast, following the method proposed by Ribeiro et al. [11]. After roasting, the beans were ground using an electric grinder (Di Grano MDR302, JCS Brazil Home Appliances Ltd. Feira de Santana, BA, Brazil) to a medium/fine granulometry with a 32-mesh standard (0.50 mm). The beans were previously frozen to facilitate the grinding of the green beans and prevent residues from adhering to the equipment.

2.4. Obtaining Extracts from the Pulp, Green, and Roasted Beans

The pulp extract was obtained using the methodology proposed by the Adolfo Lutz Institute [12], with minor modifications. Two grams of coffee (3 coffee cherries) from each treatment was added to 10 mL of distilled water. The fruit peel was broken using a glass rod, and the pulp attached to the endosperm was manually removed, followed by filtration through Whatman No. 2 filter paper. To obtain extracts of green and roasted beans by solid–liquid extraction, 1 g of each coffee sample (dry, milled and green, or roasted coffee) was extracted with 100 mL of distilled water at 90 °C and left to stand at room temperature (27 °C ± 0.5 °C) for 20 min, followed by filtration through Whatman No. 2 filter paper, as described by Kim et al. [13].

Extracts were obtained for chromatographic analysis using a modified methodology from Perrone et al. [14]. A total of 0.5 g of ground coffee (green and roasted beans) was suspended in 60 mL of boiling water and stirred at room temperature (25 °C) for 15 min. The mixture was filtered through Whatman No. 2 filter paper and washed with approximately 30 mL of water. The final volume was adjusted to 100 mL with water. The extract was filtered through a 0.45 μm cellulose ester membrane (MilliporeSigma, São Paulo, SP, Brazil).

2.5. Monitoring Mass Loss (ML), pH, Total Soluble Solids (TSSs), Instrumental Color, and Total Reducing Sugar

Mass loss (ML) during the fermentation period was determined by the difference between the coffee samples’ initial and final mass (g). The pH was measured using a digital pH meter (PA200, Marconi Laboratory Equipment Ltd., Piracicaba, SP, Brazil) in extracts obtained from the coffee pulp. Total soluble solids (TSSs) were determined by measuring the refractive index with a portable refractometer (OXD336-1010K, Oxford Precision Plc., London, UK), using the coffee pulp, and were expressed in °Brix. The methods for determining ML, pH, and TSSs followed the AOAC [15] guidelines. The color parameters L* (brightness), a* (red/green intensity), and b* (yellow/blue intensity) of the fermented coffee were measured using a digital colorimeter (CR-10, Konica Minolta Inc., Osaka, FU, Japan) equipped with a C illuminant, an 8 mm aperture, and a standard 10° observer.

The total reducing sugar content in the pulp extract was determined using the DNSA (3,5-dinitrosalicylic acid) method [16], using glucose as a standard. To perform the assay, 100 µL of the pulp extract was added to 1 mL of DNSA reagent. The mixture was homogenized and then subjected to a water bath at 100 °C for 5 min, which was cooled in an ice bath. Absorbance was measured at a wavelength of 540 nm.

2.6. Enzymatic Activity

2.6.1. Polygalacturonase Activity

Polygalacturonase activity was determined by measuring the release of reducing groups. A 1.0% (w/v) citrus pectin solution (Sigma-Aldrich, St. Louis, MO, USA) prepared in 0.1 M acetate buffer (pH 4.5) was incubated at 40 °C for 15 min to stabilize the temperature. Next, 500 µL of enzymatic extract was added to the substrate, and the reaction was incubated at 40 °C for 40 min. After this incubation, 100 µL of the reaction mixture was added to 1 mL of DNSA solution and boiled for 5 min. It was then cooled in an ice bath, and 5.0 mL of distilled water was added. Absorbance was measured at 540 nm using a spectrophotometer, using the method initially proposed by Miller [16]. One unit of activity was defined as the amount of enzyme required to release 1 μmol of galacturonic acid per minute, based on a standard curve established with α-D-galacturonic acid (Molecular mass 212.16, Fluka Chemie AG., Buchs, St. Gallen, Switzerland) as the reducing sugar.

2.6.2. Pectin Lyase Activity

Pectin lyase activity was determined by measuring the increase in absorbance resulting from the formation of 4,5-unsaturated uronic products, as proposed by Albersheim [17]. The reaction mixture consisted of 1.0 mL of a 0.5% (w/v) citrus pectin solution (Sigma-Aldrich, USA) in 0.2 M acetate buffer (pH 4.5) and 1.0 mL of enzyme extract. The mixture was incubated at 40 °C for 1 h. The reaction was stopped by adding 3.5 mL of 0.5 M HCl. One enzyme unit (U) was defined as the amount of enzyme that released one µmol of unsaturated uronic products per minute.

2.7. Determination of Phenolic Content and Antioxidant Capacity Assays

Phenolic compounds in green and roasted coffee beans were quantified using the Folin–Ciocalteu method [18], with gallic acid as the standard. A 0.2 mL aliquot of the sample was added to 1.0 mL of Folin–Ciocalteu reagent and 0.8 mL of sodium carbonate solution (75 g·L⁻¹). The mixture was incubated in a water bath at 37 °C for 30 min. Absorbance was measured at 765 nm. Results were expressed as gallic acid equivalents (mg GAE·g⁻¹).

Antioxidant activity was determined using the DPPH assay, according to the methodology described by Rufino et al. [19]. A 0.1 mL aliquot of the sample was mixed with 3.9 mL of 60 μM DPPH solution, kept at room temperature (25 °C), and protected from light for 30 min. Absorbance was measured at 515 nm. A standard curve was prepared with a Trolox solution, and results were expressed as μM Trolox·g⁻¹.

The ABTS radical cation scavenging assay was performed according to the methodology described by Re et al. [20]. A 50 μL aliquot of the diluted sample was mixed with 950 μL of the diluted ABTS solution (1:90 ABTS: ethanol). The reaction mixture was incubated for 6 min at room temperature (23 °C). Absorbance was measured at 734 nm, and the results were expressed as μM Trolox·g⁻¹.

2.8. Bioactive Compound Assay

The identification and quantification of caffeine, trigonelline, and chlorogenic acid (5-O-caffeoylquinic acid, 5-CQA) were performed by high-performance liquid chromatography (HPLC) with a UV/Vis molecular absorption detector and an Acclaim^® 120 Dionex C-18 analytical column (250 nm × 4.6 mm, 5 µm). The mobile phase consisted of 0.3% aqueous formic acid (eluent A) and methanol (eluent B), delivered at a flow rate of 1.0 mL·min⁻¹. The mobile phase composition was 30% B until the end of the run at 15 min. The column was maintained at 40 °C, and the detection wavelengths were 272 nm for caffeine and trigonelline and 325 nm for chlorogenic acid and caffeic acid [14].

2.9. Volatile Compounds by GC-MS-HS

Roasted coffee beans were subjected to gas chromatography–mass spectrometry (GC-MS) to determine their volatile constituents. The analysis was performed using a Shimadzu GC-MS QP2010 SE system with a mass-selective detector, operating at an electron impact of 70 eV, a scan interval of 0.5 s, and fragment ions ranging from m/z 40 to 550 Da. The system was equipped with a Shimadzu OAC 6000 Headspace automatic injector for direct injection of the headspace, and a nonpolar DB-5 fused silica capillary column (30 m × 0.25 mm × 0.25 μm) (J&W Scientific, Folsom, CA, USA). Helium gas was used as the mobile phase, with a flow rate of 1.0 mL.min⁻¹.

The injection system was set with the following parameters: incubation time of 5 min at 80 °C, 250 rpm, syringe pre-purge time of 60 s, injection flow rate of 25 mL.min⁻¹, and a post-injection residence time of 60 s. The temperature was programmed from 40 to 60 °C at a rate of 5 °C.min⁻¹, held at 60 °C for 2 min, and then increased to 230 °C at a rate of 20 °C.min⁻¹, with a final hold at 230 °C for 10 min. The injector and detector temperatures were set to 250 °C. The amounts of each compound were calculated from the GC peak areas in the order of elution from the DB-5 column and expressed as a relative percentage of the total chromatogram area.

The volatile components were identified by comparing the linear retention indices (LRI) calculated relative to a homologous series of C8-C40 n-alkanes using the equation of Van Den Dool and Kratz [21], along with mass spectra. Additionally, the obtained mass spectra were computerized compared with those in the GC-MS data system’s mass spectral library (WILEY 21). The area percentages were obtained from the GC-FID response.

2.10. Statistical Analysis

A two-way ANOVA was applied to evaluate the effect of the interaction between fermentation time and enzyme solution concentration. Means were compared using the Tukey test, with a significance level set at 95% (p < 0.05). Statistical analyses were performed using the SPSS software (v. 21.0), and significance was determined at the 0.05 level. All analyses were conducted in triplicate, and results are expressed as mean ± standard error. Additionally, principal component analysis (PCA) was performed to identify similarities in the bioactive and volatile composition and assess the correlation of physicochemical parameters across different fermentation times and enzyme solution concentrations.

3. Results

3.1. Physicochemical Monitoring During Coffee Fermentation

Monitoring of the fermentation environment revealed a significant increase in the moisture content of the medium as a function of enzyme treatment. The average moisture percentages 60.6 ± 1.29%, 70.6 ± 0.24%, 72.4 ± 0.24%, and 73.6 ± 0.40% for enzyme concentrations of 0, 1, 5, and 10 U·mL⁻¹, respectively, corroborate the gradual mass loss observed over the fermentation period, primarily due to water release. Accordingly, fermented coffee beans without enzyme treatment exhibited the least mass reduction (8.8 ± 0.2%), while those treated with 1, 5, and 10 U·mL⁻¹ of the enzyme solution showed significantly higher losses of 9.11 ± 0.73%, 9.38 ± 0.76%, and 9.40 ± 0.97%, respectively. The internal temperature did not differ significantly between the fermented coffee samples, presenting an average of 27.2 ± 0.8 °C. However, an average increase of 0.9 ± 0.04 °C was observed when comparing the temperature of the fermentation medium with the external environment (approximately 26.5 °C).

Fruit coloration was significantly affected (p < 0.05) by both fermentation time and enzyme solution concentration (

Table 1. Variations in the colorimetric and physical–chemical parameters of fermented coffee cherries as a function of time and concentration of the enzyme solution.

Physical–Chemical Parameter	Time (h)	Enzyme Solution Concentration (U·ML−1)
		0	1	5	10
Colorimetric parameter L	0	10.17 ± 0.07E	-	-	-
	24	12.17 ± 0.43DAB	13.43 ± 1.44AA	11.60 ± 0.26AAB	10.97 ± 0.12AB
	48	18.40 ± 1.15BA	13.60 ± 0.83AB	11.27 ± 0.15ABC	11.07 ± 0.09AC
	72	21.10 ± 2.40AA	10.63 ± 0.18BB	11.27 ± 0.19AB	11.00 ± 0.06AB
	96	11.30 ± 0.35DA	12.57 ± 0.80ABA	11.50 ± 0.38AA	11.17 ± 0.18AA
	120	14.40 ± 1.11CA	11.43 ± 0.03ABB	11.97 ± 0.94AB	11.30 ± 0.95AB
Colorimetric parameter a*	0	3.17 ± 0.81C	-	-	-
	24	3.30 ± 0.23CAB	5.07 ± 0.42AA	3.23 ± 0.34AAB	2.50 ± 0.40AB
	48	5.80 ± 1.56BCA	5.07 ± 1.41AA	2.20 ± 0.21AB	0.67 ± 0.32AB
	72	11.07 ± 1.66AA	6.30 ± 0.21AB	2.07 ± 0.41AC	0.40 ± 0.15AC
	96	4.17 ± 0.97CB	6.90 ± 0.52AA	3.57 ±0.60ABC	1.67 ± 0.29AC
	120	7.23 ± 0.52BA	4.80 ± 0.78AB	2.73 ± 0.58ABC	1.13 ± 0.37AC
Colorimetric parameter b*	0	1.47 ± 0.66B	-	-	-
	24	−4.50 ± 0.67DD	2.47 ± 0.37BA	1.93 ± 0.95ABB	0.87 ± 0.41ACC
	48	5.57 ± 0.50AA	4.47 ± 0.72AA	2.83 ± 0.19AB	−1.23 ± 0.07DC
	72	6.77 ± 0.54AA	2.77 ± 0.17BB	1.17 ± 0.66BC	−1.47 ± 0.07DD
	96	1.07 ± 0.34BCB	4.03 ± 0.38ABA	1.10 ± 0.97BB	−0.17 ± 0.09CDB
	120	−0.47 ± 0.52CC	3.83 ± 0.24ABA	0.60 ± 0.70BC	0.97 ± 0.07AB
TSSs (°Brix)	0	15.20 ± 0.15A	-	-	-
	24	14.53 ± 0.26ABA	13.40 ± 0.21ABB	11.00 ± 0.29AC	10.13 ± 0.19AD
	48	13.97 ± 0.50BCA	13.43 ± 0.07AA	10.53 ± 0.48ABB	10.00 ± 0.06AB
	72	13.33 ± 0.67CDA	12.67 ± 0.17ABA	10.50 ± 0.29ABB	10.13 ± 0.15AB
	96	13.00 ± 0.06DA	12.60 ± 0.15BA	10.33 ± 0.17ABCB	9.67 ± 0.34AB
	120	12.00 ± 0.29EA	12.67 ± 0.18ABA	9.67 ± 0.09CB	9.63 ± 0.22AB
pH	0	5.23 ± 0.03A	-	-	-
	24	4.49 ± 0.01BA	4.58 ± 0.05AA	4.28 ± 0.08AB	4.19 ± 0.07AB
	48	4.11 ± 0.02CA	4.16 ± 0.01BA	4.13 ± 0.04BA	4.20 ± 0.01AA
	72	4.13 ± 0.02CAB	4.07 ± 0.01BB	4.15 ± 0.02BAB	4.17 ± 0.04AA
	96	3.98 ± 0.01DB	4.07 ± 0.02BAB	4.10 ± 0.06BA	4.04 ± 0.02BAB
	120	3.99 ± 0.01DAB	3.95 ± 0.01CB	4.06 ± 0.03BA	3.90 ± 0.05CB

Different capital letters in the same column indicate statistical differences as a function of fermentation time. Different lowercase letters in the same row indicate statistical differences as a function of the concentration of the enzyme solution. TSSs: Total soluble solids.

). Coffee fermented without enzyme addition and samples treated with 1 U·mL⁻¹ of enzyme solution exhibited significant fluctuations in the L* parameter (lightness) throughout the fermentation process. In contrast, coffee treated with enzyme concentrations of 5 and 10 U·mL⁻¹ showed a reduction in lightness after 24 h of fermentation, which then remained stable, resulting in a darker appearance early in the process. These samples also exhibited similar trends in the a* parameter (green/red), showing the lowest a* values (indicating a reduction in red coloration) that remained stable throughout fermentation. All samples differed significantly depending on enzyme concentration regarding the b* parameter (yellow/blue). Once again, coffee fermented without enzyme treatment showed considerable fluctuations over time, whereas treated samples demonstrated stable behavior, with a direct correlation between increasing enzyme concentration and decreasing b* values.

3.2. Reducing Sugar and Enzyme Activity

Enzyme treatment strongly impacted the reduction in total soluble solids (TSSs) levels. Fermented coffee cherries without enzyme treatment showed a gradual decrease in TSSs, and those treated with 1 U·mL⁻¹ of enzyme solution exhibited similar reductions. In contrast, samples treated with 5 U·mL⁻¹ experienced a significant decrease (p < 0.05) in TSSs within the first 24 h of fermentation, followed by stability until 96 h and a further significant reduction at 120 h. Samples receiving 10 U·mL⁻¹ of enzyme reached the lowest TSSs levels within the first 24 h and remained stable thereafter. Fermentation time had a stronger influence on pH reduction, with all treatments showing significant decreases (p < 0.05) over the 120 h. However, fruits treated with 5 and 10 U·mL⁻¹ of enzyme solution exhibited more pronounced pH reductions within the first 24 h. Complete TSSs and pH data throughout the coffee fermentation process are presented in Supplementary Table S1.

Reductions in reducing sugar concentrations were proportional to the increase in enzyme solution concentration (Figure 1A). Non-treated coffee exhibited less pronounced reductions, decreasing from 7.05 ± 0.08 to 5.57 ± 0.02 mg glucose·mL⁻¹ between the beginning and end of fermentation. Treatment with 1 U·mL⁻¹ of enzyme solution reduced from 5.60 ± 0.09 to 1.57 ± 0.03 mg glucose·mL⁻¹, with more marked declines observed at 72 h (2.60 ± 0.04 mg glucose·mL⁻¹). In contrast, fruits treated with 5 and 10 U·mL⁻¹ of enzyme solution showed significant reductions within the first 24 h, reaching 3.81 ± 0.14 and 3.78 ± 0.03 mg glucose·mL⁻¹, respectively. These samples did not differ statistically at the end of fermentation, with final concentrations of 1.22 ± 0.05 and 1.16 ± 0.03 mg glucose·mL⁻¹, respectively.

All samples showed significant reductions (p < 0.05) in polygalacturonase activity throughout fermentation (Figure 1B). The highest activity was observed in coffee fermented for 24 h without enzyme treatment (1.47 ± 0.15 U·mL⁻¹), while the lowest activity was recorded after 120 h of fermentation (0.30 ± 0.02 U·mL⁻¹) in fruits treated with 1 U·mL⁻¹ of enzyme solution. At the end of 120 h, no significant differences were observed between treatments. In contrast, pectin lyase activity increased throughout fermentation (Figure 1C), particularly in samples with higher enzyme concentrations. The peaks of pectin lyase activity were 6.02 ± 0.02 and 5.88 ± 0.01 U·mL⁻¹ for the 5 and 10 U·mL⁻¹ concentrations, respectively, and were reached after 120 h of fermentation. The complete data for reducing sugar, polygalacturonase, and pectin lyase activities are presented in Supplementary Table S1.

3.3. Phenolic Compounds and Antioxidant Activity

Unfermented green beans had the highest total phenolic content (TPC) of 29.43 ± 0.15 mg GAE·g⁻¹, while fermentation resulted in significant reductions in TPC (p < 0.05) for all treatments (Figure 2A). For roasted beans (Figure 2B), an inversely proportional relationship was observed between enzyme solution concentration and phenolic compound content at the end of 120 h of fermentation, with values of 25.31 ± 0.09, 24.07 ± 0.07, 19.15 ± 0.03, and 14.93 ± 0.07 mg GAE·g⁻¹ for concentrations of 0, 1, 5, and 10 U·mL⁻¹, respectively. In parallel, the highest antioxidant activity values were observed in non-fermented beans, with DPPH values of 1593.81 ± 19.22 and 1736.67 ± 12.60 µM trolox·g⁻¹ for green and roasted beans, respectively (Figure 2C,D). The highest ABTS values were 3701.90 ± 6.30 and 4028.10 ± 8.25 µM trolox·g⁻¹ for green and roasted beans, respectively (Figure 2E,F). Consistent with TPC values, fermentation also reduced the antioxidant potential of the samples, with the lowest values observed for roasted beans treated with 10 U·mL⁻¹ and fermented for 120 h: DPPH (1146.19 ± 20.25 µM trolox·g⁻¹) and ABTS (2342.38 ± 12.16 µM trolox·g⁻¹) radicals. Complete data for TPC and antioxidant activity can be found in Supplementary Table S2.

3.4. Bioactive Compounds by HPLC

The bioactive compounds caffeine, trigonelline, and chlorogenic acid (5-O-caffeoylquinic acid, 5CQA) were influenced by fermentation time, enzyme concentration, and roasting (

Table 2. Concentration of caffeine, trigonelline and chlorogenic acid (5CQA) in non-fermented and fermented coffee samples before and after roasting. * U·mL⁻¹. Different capital letters in the same column denote significant differences between different enzyme concentrations, fixing the fermentation time. Different lowercase letters in the same column denote significant differences between different fermentation times, fixing the enzyme concentration. nd: not detected.

Fermentation Time (Hours)	Enzymatic Concentration *	Caffeine (mg·100g−1)		Trigonelline (mg·100g−1)		Chlorogenic Acid (5CQA) (mg·100g−1)
		Green	Roasted	Green	Roasted	Green	Roasted
0	0	592.93 ± 27.2a	873.63 ± 32.5a	1028.75 ± 31.4a	914.79 ± 44.0bc	423.46 ± 40.3a	185.85 ± 2.5a
24	0	506.98 ± 36.1Cb	894.40 ± 52.4Aa	859.25 ± 8.4Bb	1047.48 ± 35.2Aa	374.24 ± 15.9Ab	196.19 ± 16.1Aa
	1	645.07 ± 24.0Bab	902.38 ± 10.3Aa	910.88 ± 11.9Aa	911.75 ± 37.7Ba	262.97 ± 7.9Ca	95.55 ± 7.9Ba
	5	742.51 ± 16.7Aa	890.18 ± 19.1Aa	847.97 ±11.2Ba	790.36 ± 37.8Ca	302.38 ± 4.7Ba	63.81 ± 1.3BCa
	10	648.60 ± 9.2Ba	795.68 ± 12.1Ba	716.84 ± 9.8Ca	687.30 ± 32.5Dab	168.96 ± 5.6Da	55.96 ± 15.0Ca
48	0	593.25 ± 13.6Ba	866.01 ± 46.2Aab	857.71 ± 8.9Ab	1069.03 ± 38.8Aa	335.80 ± 2.1Ac	189.08 ± 14.4Aa
	1	568.26 ± 19.3Bc	826.88 ± 7.6Ab	820.93 ± 4.9Ab	897.84 ± 15.7Ba	220.28 ± 1.4Bc	92.84 ± 7.6Bab
	5	685.36 ± 25.8Aab	857.57 ± 3.0Ab	768.28 ± 29.6Bb	769.96 ± 4.3Cab	310.80 ± 9.5Aa	69.38 ± 0.2Ba
	10	595.80 ± 2.1Bab	693.24 ± 21.8Bb	686.66 ± 0.1Ca	737.90 ± 11.2Ca	156.54 ± 7.3Cab	26.83 ± 11.4Ca
72	0	546.88 ± 8.9Bab	832.24 ± 13.0Aab	844.96 ± 2.1Ab	993.97 ± 5.9Aab	364.41 ± 0.7Abc	181.43 ± 12.5Aa
	1	694.90 ± 6.2Aa	792.92 ± 2.5Ab	808.48 ± 20.0Abc	847.72 ± 9.7Bab	256.71 ± 11.3Cab	81.18 ± 1.7Bab
	5	722.09 ± 3.7Aa	773.55 ± 5.5Ab	801.71 ± 21.9Aab	691.26 ± 38.0Cbc	312.83 ± 4.8Ba	49.87 ± 11.BCab
	10	592.95 ±14.7Bab	683.22 ± 9.0Bb	676.15 ± 8.6Ba	738.86 ± 42.9Ca	148.20 ± 5.0Dabc	42.77 ± 9.9Ca
96	0	540.69 ± 19.3Bab	808.61 ± 24.3Ab	742.48 ± 9.6Ac	936.50 ± 3.3Abc	236.88 ± 1.4Ad	145.33 ± 23.2Ab
	1	627.60 ± 40.3Ab	806.52 ± 7.6Ab	767.68 ± 26.5Ac	813.87 ± 11.0Bb	226.75 ± 8.1Abc	65.75 ± 7.5Bab
	5	642.11 ± 10.2Abc	786.39 ± 14.7Ab	787.81 ± 5.2Ab	716.09 ± 29.5Cabc	259.23 ± 7.8Ab	47.87 ± 7.1Bab
	10	556.41 ± 11.3Bb	635.56 ± 20.4Bbc	625.58 ± 1.5Bb	564.26 ± 54.6Dc	125.58 ± 11.2Bbc	nd
120	0	557.99 ± 5.6Bab	856.90 ± 20.9Aab	752.06 ± 4.3Ac	871.57 ± 15.4Ac	243.91 ± 1.8Ad	105.58 ± 1.0Ac
	1	643.91 ± 32.9Aab	808. 46 ± 18.3Bb	783.03 ± 27.8Abc	770.99 ± 21.1Bb	228.25 ± 13.3Abc	62.89 ± 6.7Bb
	5	616.28 ± 2.9Ac	701.59 ± 16.3Cc	691.76 ± 17.8Bc	644.68 ± 1.3Cc	213.22 ± 6.7Ac	30.53 ± 2.6Bb
	10	487.83 ± 21.5Cc	605.49 ± 21.9Dc	614.45 ± 12.7Cb	609.95 ± 5.5Cbc	121.76 ± 6.5Bc	nd

). The highest caffeine content in green beans was observed in samples treated with 5 U·mL⁻¹ after 24 h of fermentation (742.51 ± 16.7 mg·100g⁻¹). Enzyme treatment at a concentration of 10 U·mL⁻¹ resulted in the lowest caffeine content among green beans (487.83 ± 21.5 mg·100g⁻¹) after 120 h of fermentation. In contrast, roasting positively affected caffeine levels, with roasted beans showing higher values than green beans across all treatments. For roasted beans, the highest caffeine levels were reached within the first 24 h of fermentation, with no significant differences (p > 0.05) observed among samples treated with enzyme concentrations of 0, 1, and 5 U.mL⁻¹, yielding values of 894.40 ± 52.4, 902.38 ± 10.3, and 890.18 ± 19.1 mg·100g⁻¹, respectively.

Trigonelline decreased significantly with fermentation time, with the highest trigonelline concentration in green beans observed in unfermented coffee (1028.75 ± 31.4 mg·100g⁻¹). Treatment with an enzyme solution at a concentration of 10 U·mL⁻¹ resulted in the greatest reduction in trigonelline concentration throughout fermentation, reaching the lowest level after 120 h (614.45 ± 12.7 mg·100g⁻¹). Depending on the enzyme concentration, roasting influenced trigonelline levels either positively or negatively. For roasted beans, the highest trigonelline values were obtained in beans fermented for 24 and 48 h without enzyme treatment (1047.48 ± 35.2 and 1069.03 ± 38.8 mg·100g⁻¹, respectively).

The 5CQA levels decreased significantly (p < 0.05) with fermentation time, with the highest values observed in unfermented green beans (423.46 ± 40.3 mg·100g⁻¹). The influence of enzyme treatment showed dynamic behavior in green beans, with the lowest 5CQA values found in beans treated with 10 U·mL⁻¹ throughout the 120 h of fermentation. However, beans with enzyme treatment at 5 U·mL⁻¹ exhibited higher 5CQA concentrations than those treated with 1 U·mL⁻¹. Additionally, a direct relationship was observed for roasted beans between increasing enzyme solution concentration and a reduction in 5CQA levels. Thus, the combination of roasting and enzyme treatment at concentrations of 5 and 10 U·mL⁻¹ reduced 5CQA levels to undetectable levels after 120 h of fermentation.

3.5. Volatile Composition

Considering the impact of the Maillard reaction on the formation of volatile compounds, the volatile composition of roasted beans was studied. Thirty-one compounds were identified in both fermented and unfermented coffee beans, primarily distributed across the pyrazine, furan, acid, alcohol, terpene, and ester groups. Enzymatic treatment positively influenced the formation of the volatile profile, with coffee treated at a concentration of 10 U·mL⁻¹ exhibiting the greatest variety of compounds (

Table 3. Volatile compounds in fermented coffee beans treated with 10 U·mL⁻¹ of pectinase solution.

Volatile Compounds	Treatment (RPA *)
	C10H24	C10H48	C10H72	C10H96	C10H120
Amide
l-Methioninamide	nd	1.11	0.59	0.45	nd
Pyrazines
2-Methylpyrazine	3.70	5.04	5.50	6.00	5.36
2,5-Dimethyl-Pyrazine	2.75	nd	nd	1.49	1.13
2-Ethyl-6-methylpyrazine	0.15	0.22	0.76	0.71	0.67
2-Ethylpyrazine	nd	1.89	2.39	1.86	2.03
Pyrroles
1-furfuryl-Pyrrole	0.39	0.55	0.94	1.11	1.39
Furans
Furfural	17.36	15.19	15.68	10.15	12.17
5-Methyl-2-furaldehyde	8.44	8.20	7.48	3.83	5.54
2-Pentylfuran	0.29	0.40	0.46	0.37	nd
Furfuryl acetate	5.19	6.77	6.41	9.98	8.53
Dihydro-2-methyl-3-furanone	1.68	1.55	1.06	0.74	nd
Esters
Pentyl acetate	nd	nd	0.72	nd	nd
3-Methyl-1-butyl propanoate	nd	nd	nd	0.69	0.73
Methyl salicylate	0.20	0.35	0.45	0.44	0.55
Acids
3-methyl-Butanoic acid	2.20	1.93	2.76	2.11	1.55
2-methyl-Hexanoic acid	nd	nd	0.62	nd	0.43
2-methyl-Butanoic acid	43.60	40.62	40.04	45.31	41.06
Ethyl ester Hexanoic acid	1.06	1.35	1.79	1.66	1.82
Cyclohexyl ester 2-Butenoic acid	0.26	0.61	0.60	0.54	0.40
Pyridine
1-Acetyl-1,4-dihydropyridine	0.75	0.86	0.98	0.64	0.57
Alcohol
Benzeneethanol	0.36	0.37	0.66	0.55	0.92
Benzenemethanol	0.18	nd	0.31	nd	0.28
3-Methyl-2-buten-1-ol	0.54	0.56	0.96	1.67	1.74
Propanoate-2-Buten-1-ol	0.42	0.55	nd	nd	nd
1-Hexanol	1.51	1.72	1.55	1.71	2.04
5-methyl-1-Hexanol	0.90	1.20	0.78	1.45	1.38
2,4-Dimethyl-3-heptanol	nd	0.33	0.29	0.34	0.67
Terpenes
Linalool	0.18	0.85	nd	nd	nd
Trans-Linalool oxide	0.18	0.21	0.24	0.24	0.34
Linalyl butanoate	0.21	0.24	0.62	0.93	1.04
Imidazole
1-(3H-Imidazol-4-yl)-ethanone	3.72	5.48	4.45	3.87	4.03

* RPA (%) = (Peak area/total) × 100%; RPA shown is calculated based on all identified compounds. nd: Not detected. C10H24 (enzymatic concentration of 10 U·mL⁻¹ and 24 h of fermentation); C10H48 (enzymatic concentration of 10 U·mL⁻¹ and 48 h of fermentation); C10H72 (enzymatic concentration of 10 U·mL⁻¹ and 72 h of fermentation); C10H96 (enzymatic concentration of 10 U·mL⁻¹ and 96 h of fermentation); C10H120 (enzymatic concentration of 10 U·mL⁻¹ and 120 h of fermentation).

). Notably, treatments C10H48 (enzymatic concentration of 10 U·mL⁻¹ and 48 h of fermentation), C10H72 (enzymatic concentration of 10 U·mL⁻¹ and 78 h of fermentation), and C10H96 (enzymatic concentration of 10 U·mL⁻¹ and 96 h of fermentation) presented 26, 27, and 26 volatile compounds, respectively. The volatile profile of the fermented beans treated with 10 U·mL⁻¹ was enriched with the alcohols benzeneethanol, benzenemethanol, 3-methyl-2-buten-1-ol, propanoate-2-buten-1-ol, and 2,4-dimethyl-3-heptanol; the terpene trans-linalool oxide; and the furan 2-pentylfuran.

For the beans treated with enzyme solution at a concentration of 5 U.mL⁻¹ (

Table 4. Volatile compounds in fermented coffee beans treated with 5 U·mL⁻¹ of pectinase solution.

Volatile Compounds	Treatment (RPA *)
	C5H24	C5H48	C5H72	C5H96	C5H120
Amide
l-Methioninamide	1.08	0.89	1.03	0.96	0.51
Pyrazines
2-Methylpyrazine	4.94	6.66	6.24	6.22	7.84
2,5-Dimethyl-Pyrazine	2.3	nd	nd	nd	nd
2-Ethyl-6-methylpyrazine	0.65	0.88	0.84	1.95	1.08
2-Ethylpyrazine	nd	2.09	2.19	3.08	3.72
Pyrroles
1-furfuryl-Pyrrole	0.26	0.43	0.55	0.60	0.75
Furans
Furfural	13.94	13.67	14.41	17.64	11.26
5-Methyl-2-furaldehyde	8.01	7.41	8.00	8.90	5.39
2-Pentylfuran	nd	0.33	0.36	5.25	0.55
Furfuryl acetate	5.71	7.14	8.75	1.02	8.49
Dihydro-2-methyl-3-furanone	1.86	1.95	1.74	1.78	1.18
Esters
Pentyl acetate	nd	nd	0.58	nd	nd
3-Methyl-1-butyl propanoate	0.77	nd	nd	nd	0.65
Methyl salicylate	0.18	nd	0.33	0.46	0.61
Acids
3-methyl-Butanoic acid	2.02	2.28	2.53	2.91	2.71
2-methyl-Hexanoic acid	nd	nd	nd	nd	nd
2-methyl-Butanoic acid	45.51	42.73	38.92	37.07	37.47
Ethyl ester Hexanoic acid	1.69	1.99	1.99	nd	2.71
Cyclohexyl ester 2-Butenoic acid	0.41	0.18	0.43	0.56	0.85
Pyridine
1-Acetyl-1,4-dihydropyridine	0.57	0.89	0.63	0.55	0.64
Alcohol
Benzeneethanol	0.30	0.28	0.37	0.60	nd
Benzenemethanol	nd	nd	nd	0.36	nd
3-Methyl-2-buten-1-ol	nd	nd	nd	nd	1.68
Propanoate-2-Buten-1-ol	0.33	nd	nd	0.57	nd
1-Hexanol	1.83	1.76	1.82	1.61	1.37
5-methyl- 1-Hexanol	1.08	0.97	1.14	0.66	1.12
2,4-Dimethyl-3-heptanol	nd	nd	nd	nd	nd
Terpenes
Linalool	nd	0.20	nd	nd	nd
Trans-Linalool oxide	0.22	0.19	0.23	0.25	0.35
Linalyl butanoate	0.45	0.51	0.61	nd	1.26
Imidazole
1-(3H-Imidazol-4-yl)-ethanone	4.92	5.59	5.46	5.69	5.85

* RPA (%) = (Peak area/total) × 100%. nd: Not detected. C5H24 (enzymatic concentration of 5 U·mL⁻¹ and 24 h of fermentation); C5H48 (enzymatic concentration of 5 U·mL⁻¹ and 48 h of fermentation); C5H72 (enzymatic concentration of 5 U·mL⁻¹ and 72 h of fermentation); C5H96 (enzymatic concentration of 5 U·mL⁻¹ and 96 h of fermentation); C5H120 (enzymatic concentration of 5 U·mL⁻¹ and 120 h of fermentation).

), the treatments C5H72 (enzymatic concentration of 5 U·mL⁻¹ and 72 h of fermentation) and C5H120 (enzymatic concentration of 5 U·mL⁻¹ and 120 h of fermentation) stood out, each exhibiting 23 volatile compounds. In contrast, for beans treated with enzyme solution at a concentration of 1 U·mL⁻¹ (

Table 5. Volatile compounds in fermented coffee beans treated with 1 U·mL⁻¹ of pectinase solution.

Volatile Compounds	Treatment (RPA *)
	C1H24	C1H48	C1H72	C1H96	C1H120
Amide
l-Methioninamide	0.94	0.93	0.82	0.94	0.89
Pyrazines
2-Methylpyrazine	4.92	5.72	4.51	4.43	6.72
2,5-Dimethyl-Pyrazine	1.63	nd	0.93	1.86	nd
2-Ethyl-6-methylpyrazine	0.69	0.99	0.68	0.71	1.07
2-Ethylpyrazine	nd	3.13	1.06	nd	2.18
Pyrroles
1-furfuryl-Pyrrole	0.33	0.43	nd	0.35	0.42
Furans
Furfural	19.49	14.18	18.06	12.47	19.19
5-Methyl-2-furaldehyde	10.01	8.47	8.71	7.96	10.39
2-Pentylfuran	nd	nd	nd	nd	nd
Furfuryl acetate	4.26	6.76	3.21	6.90	6.94
Dihydro-2-methyl-3-furanone	2.11	nd	1.74	1.30	2.29
Esters
Pentyl acetate	nd	nd	nd	nd	1.06
3-Methyl-1-butyl propanoate	nd	nd	0.50	0.81	nd
Methyl salicylate	0.49	0.34	nd	0.24	0.50
Acids
3-methyl-Butanoic acid	3.01	2.65	2.81	nd	2.55
2-methyl-Hexanoic acid	nd	nd	nd	nd	0.69
2-methyl-Butanoic acid	40.20	40.34	47.81	51.72	32.09
Ethyl ester Hexanoic acid	1.55	2.21	1.05	1.66	1.85
Cyclohexyl ester 2-Butenoic acid	0.38	0.67	0.46	0.33	0.26
Pyridine
1-Acetyl-1,4-dihydropyridine	0.58	0.52	0.54	nd	0.37
Alcohol
Benzeneethanol	nd	0.40	nd	nd	nd
Benzenemethanol	nd	0.23	nd	nd	nd
3-Methyl-2-buten-1-ol	nd	nd	nd	nd	nd
Propanoate-2-Buten-1-ol	nd	1.69	nd	nd	nd
1-Hexanol	1.66	1.78	1.52	1.99	1.59
5-methyl- 1-Hexanol	0.86	0.76	0.57	0.75	0.82
2,4-Dimethyl-3-heptanol	nd	nd	nd	nd	0.25
Terpenes
Linalool	nd	nd	nd	nd	nd
Trans-Linalool oxide	nd	nd	nd	nd	nd
Linalyl butanoate	0.42	0.50	nd	0.56	0.54
Imidazole
1-(3H-Imidazol-4-yl)-ethanone	6.19	5.65	4.63	4.55	6.26

* RPA (%) = (Peak area/total) × 100%. nd: Not detected. C1H24 (enzymatic concentration of 1 U·mL⁻¹ and 24 h of fermentation); C1H48 (enzymatic concentration of 1 U·mL⁻¹ and 48 h of fermentation); C1H72 (enzymatic concentration of 1 U·mL⁻¹ and 72 h of fermentation); C1H96 (enzymatic concentration of 1 U·mL⁻¹ and 96 h of fermentation); C1H120 (enzymatic concentration of 1 U·mL⁻¹ and 120 h of fermentation).

), the treatments C1H48 (enzyme concentration of 1 U·mL⁻¹ and 48 h of fermentation) and C1H120 (enzyme concentration of 1 U·mL⁻¹ and 120 h of fermentation) stood out, with a volatile profile composed of 21 and 22 compounds, respectively. These results demonstrated an increase in the diversity of volatile compounds as the enzyme solution concentration increased.

Unfermented beans (C0H0) and beans fermented without enzyme treatment (

Table 6. Volatile compounds in fermented coffee beans treated with 0 U·mL⁻¹ of pectinase solution.

Volatile Compounds	Treatment (RPA *)
	C0H0	C0H24	C0H48	C0H72	C0H96	C0H120
Amide
l-Methioninamide	0.24	nd	0.88	0.58	0.87	1.02
Pyrazines
2-Methylpyrazine	4.45	4.60	5.07	4.21	3.96	4.69
2,5-Dimethyl-Pyrazine	nd	nd	nd	1.15	2.51	1.74
2-Ethyl-6-methylpyrazine	0.80	1.46	0.61	1.40	nd	nd
2-Ethylpyrazine	nd	nd	nd	nd	nd	nd
Pyrroles
1-furfuryl-Pyrrole	0.26	0.26	0.24	0.28	0.71	0.42
Furans
Furfural	24.24	22.52	14.28	25.39	19.75	14.16
5-Methyl-2-furaldehyde	12.17	10.63	6.81	11.3	10.18	8.35
2-Pentylfuran	0.43	nd	nd	nd	nd	nd
Furfuryl acetate	3.67	2.38	4.30	2.33	6.24	12.77
Dihydro-2-methyl-3-furanone	2.55	nd	1.90	2.01	1.49	1.90
Esters
Pentyl acetate	nd	nd	nd	nd	nd	nd
3-Methyl-1-butyl propanoate	nd	1.76	nd	0.39	1.02	nd
Methyl salicylate	0.41	0.30	nd	0.61	0.6	0.56
Acids
3-methyl-Butanoic acid	3.32	4.11	2.41	3.94	3.40	2.90
2-methyl-Hexanoic acid	nd	nd	nd	nd	nd	nd
2-methyl-Butanoic acid	38.16	41.89	51.48	38.44	38.54	37.43
Ethyl ester Hexanoic acid	1.26	1.41	1.75	1.58	1.92	3.00
Cyclohexyl ester 2-Butenoic acid	0.43	0.54	0.37	0.55	0.68	0.84
Pyridine
1-Acetyl-1,4-dihydropyridine	0.54	0.47	0.94	0.38	0.47	nd
Alcohol
Benzeneethanol	nd	nd	nd	nd	nd	nd
Benzenemethanol	nd	nd	nd	nd	nd	nd
3-Methyl-2-buten-1-ol	nd	nd	nd	nd	nd	nd
Propanoate-2-Buten-1-ol	nd	nd	nd	nd	nd	nd
1-Hexanol	1.29	1.16	1.61	1.08	1.53	1.80
5-methyl-1-Hexanol	0.70	0.63	0.82	nd	0.80	1.09
2,4-Dimethyl-3-heptanol	nd	nd	nd	nd	nd	nd
Terpenes
Linalool	nd	nd	nd	nd	nd	nd
Trans-Linalool oxide	nd	nd	nd	nd	nd	0.35
Linalyl butanoate	nd	nd	0.49	nd	1.10	nd
Imidazole
1-(3H-Imidazol-4-yl)-ethanone	4.39	3.88	3.86	4.36	4.19	5.40

* RPA (%) = (Peak area/total) × 100%. nd: Not detected. C0H0 (enzymatic concentration of 0 U·mL⁻¹ and 0 h of fermentation); C0H24 (enzymatic concentration of 0 U·mL⁻¹ and 24 h of fermentation); C0H48 (enzymatic concentration of 0 U·mL⁻¹ and 48 h of fermentation); C0H72 (enzymatic concentration of 0 U·mL⁻¹ and 72 h of fermentation); C0H96 (enzymatic concentration of 0 U·mL⁻¹ and 96 h of fermentation); C0H120 (enzymatic concentration of 0 U·mL⁻¹ and 120 h of fermentation).

) presented a less diverse volatile profile, with 18 compounds for the C0H0 and C0H72 treatments (enzyme concentration of 0 U·mL⁻¹ and 72 h of fermentation) and 19 compounds for the C0H96 treatment (enzyme concentration of 0 U·mL⁻¹ and 96 h of fermentation), representing the best results for untreated beans. Additionally, principal component analysis (PCA), explaining 67.76% of the total data variation, revealed a correlation between the distribution of volatile compounds (Figure 3A) and the different treatments applied (Figure 3B). The overlapping of the graphs places the samples C1H48, C1H120, C5H72, C5H120, C10H48, and C10H72 in the regions of highest compound density, reinforcing the greater impact of these treatments on the formation of volatile compounds in fermented coffee.

4. Discussion

The increase in moisture within the fermentation medium is closely related to the degradation of the water and pectin-rich mucilage [22]. In this context, the depolymerization of pectin in the cell wall leads to the release of water and softening of the plant tissue [4]. Treatment with pectinases, therefore, favored the degradation of pectin, resulting in a greater release of water and a higher mass loss as the enzyme solution concentration increased, with average percentages of 8.8, 9.1, 9.3, and 9.4% for beans treated with 0, 1, 5, and 10 U·mL⁻¹ of enzymatic solution, respectively. Additionally, the microbial metabolism of CO₂ [2] further contributed to the mass loss of the beans. According to Haile and Kang [23], mucilage degradation helps reduce the drying time required for coffee beans. In previous studies evaluating the influence of fermentation time using the SIAF method, an average percentage reduction of 6.1% was observed in the mass of coffee fermented for 20 days [24]. Thus, enzyme treatment presents technological improvements with the potential to reduce processing time.

Small increases in the temperature of the fermentation medium can be attributed to enzymatic activity and microbial metabolism, which promote exothermic reactions responsible for the temperature rise during the bioprocess [25]. However, the temperature of the fermentation medium did not exceed 28.5 °C, remaining within the growth range of mesophilic microorganisms, characteristic of spontaneous coffee fermentation. Similar results were obtained by Cassimiro et al. [26]. Evaluating coffee, the authors observed variations of up to 2 °C, with the maximum temperature reaching 26 °C. Additionally, the observed temperatures were within the activity range of pectin lyase and polygalacturonase. According to Yadav et al. [5], pectin lyase from Aspergillus sp. is stable in the 10–50 °C range. Similarly, Souza et al. [27] also observed stability in the range of 10–50 °C when evaluating the stability of fungal polygalacturonase.

In general, coffee fruits darken during fermentation. Pigment degradation is a natural process resulting from fruit metabolism and the action of endogenous enzymes, with enzymatic darkening primarily caused by polyphenol oxidases, leading to melanoidin formation [28]. Moreover, the loss of the red and yellow hues of ripe coffee, resulting from pigments such as anthocyanins, carotenoids, and flavonoids, may indicate the action of pectinases on these compounds. Pectinolytic enzymes can be used for clarification due to pigment degradation [4]; in this context, the degradation of anthocyanins, flavonoids, and carotenoids may occur alongside the formation of dark pigmentation. This behavior seems to be promoted by increasing enzyme concentration.

A positive correlation was observed between the increase in enzyme solution concentration and the reduction in TSSs levels. These results suggest the degradation of sugars by enzyme treatment and the action of the epiphytic microbiota involved in mixed fermentation. The synergistic action between the endogenous pectinases of the fruit, enzymes from the mixed microbiota, and enzyme treatment may enhance sugar degradation, reducing the levels of total soluble solids. Significant variations in soluble solids contents were also observed by Cassimiro et al. [26] with reductions from 16 to 8 °Brix as a function of coffee fermentation. Consequently, the breakdown of sugars leads to the production of organic acids, which lower the pH of the fermentation medium. Additionally, consistent with the pH and TSSs results, reducing sugar levels decreased during fermentation. The action of pectinases releases galacturonic acid residues, which should increase sugar concentrations throughout fermentation. However, according to Elhalis et al. [29], during coffee bioprocessing, reducing sugars are consumed by microorganisms, producing organic acids and reducing pH. Corroborating the results of this study, Elhalis et al. [9] observed significant reductions in the pH of wet-fermented coffees added with pectinase.

Although the temperature and pH conditions of the fermentation medium allow the stability of the enzymes studied, they are not optimal for enzymatic activity. Aspergillus aculeatus pectinases show optimal activity at 50 °C and pH 5 [30]. Additionally, polygalacturonase has an optimal temperature range of 30–50 °C, with a near-neutral pH, while pectin lyase functions optimally between 40 and 50 °C at a pH of 5.5 [31]. Thus, combined with the competitive characteristics of bioprocess, this may explain the reductions in polygalacturonase activity. At the time of enzyme treatment, the pH of the fruit was 5.23 ± 0.03, a value close to the optimal pH for pectin lyase (Figure 1). As a result, pectin lyase activity was significantly higher compared to polygalacturonase activity. Consequently, the high activity of pectin lyase may reduce substrate availability, contributing to the inhibition of polygalacturonase activity.

The analysis of phenolic compounds demonstrated dynamic synthesis and degradation reactions throughout the fermentation of coffee subjected to enzymatic treatment, directly influencing the antioxidant potential. The action of pectinases on polysaccharides breaks down the cell wall matrix, releasing polyphenols and increasing antioxidant activity [32]. In this context, the higher TPC values observed in unfermented beans and in those fermented for 24 h corroborate the higher polygalacturonase activity observed in the early stages of fermentation. However, subsequent reductions in TPC levels may be associated with the oxidation of phenolic substances due to microbial metabolism [33]. Evaluating the influence of different microbial strains on coffee fermentation, Abduh et al. [34] reported a similar behavior, with higher phenolic compound levels in the first moments of fermentation, followed by consecutive reductions. The authors attribute this behavior to different species presenting different optimal fermentation times.

Furthermore, the availability of phenolic compounds may be limited by their interaction with pectin fragments released by the action of pectinases during fermentation [32]. Given the direct correlation between phenolic content and antioxidant activity, the oscillatory behavior observed for the DPPH and ABTS variables can be understood. The action of pectinases alters the structure of phenolics, which can exist in free or bound forms, influencing the antioxidant potential [35]. Also, the thermolabile nature of some phenolic compounds may result in lower antioxidant activity after coffee roasting. According to Silva et al. [36], some phenolic compounds are unstable under processing conditions, such as increased temperature. Furthermore, the metabolism of specific crops may reduce the total phenolic content but increase the concentration of volatile phenolics [35], which would be lost during roasting.

Analysis of bioactive compounds revealed that enzyme treatment can contribute to an increase or decrease in caffeine concentration, depending on the fermentation time and the concentration of the enzyme solution. Pectinase acts extracellularly in plant tissue by breaking down cell wall structures, while caffeine synthesis occurs intracellularly as part of a secondary metabolic pathway [37], so there is no evidence that pectinases interfere with caffeine synthesis. Thus, a possible indirect influence could be a metabolic response to stress caused by the action of exogenous pectinases.

During fermentation, reductions in caffeine concentrations occur due to the action of bacteria capable of converting caffeine into xanthines and dimethylxanthines [38]. On the other hand, the mechanisms of caffeine biosynthesis during fermentation remain unclear in the scientific literature [26]. However, possible pathways include the possibility that some microorganisms possess the genetic information necessary for caffeine biosynthesis or that enzymes secreted by the fermentative microbiota act as catalysts for caffeine production [39]. Additionally, roasted coffee presented higher levels of caffeine, which agrees with the results obtained by Cassimiro et al. [26] and Bressani et al. [40]. According to these authors, the thermostable nature of caffeine is responsible for its high levels after roasting. Thus, roasting results in mass loss without caffeine degradation, so that caffeine is more concentrated in roasted beans. The presence of caffeine is associated with one of the primary flavor descriptors of coffee and may be responsible for up to 30% of the bitterness of the beverage [41].

Reductions in the concentrations of non-volatile bioactive compounds, such as trigonelline and chlorogenic acid, showed a strong relationship with the increase in the enzyme concentration. Thus, the results suggest that the enzymatic treatment favors the degradation of these compounds, corroborating the results obtained by Tai et al. [8], where the authors observed that the addition of a crude pectinase extract during coffee fermentation significantly reduced the concentrations of 5CQA. However, the mechanism of action of pectinase on chlorogenic acid has not yet been elucidated. However, the action of pectinases can accelerate the degradation processes of the plant cell wall, favor the release of water, and cause losses of soluble compounds such as caffeine and trigonelline. In addition, the degradation of the cell wall can facilitate the action of oxidative enzymes capable of partially degrading phenolic compounds such as 5CQA.

Additionally, the presence of different yeasts promotes reductions in the concentrations of trigonelline and chlorogenic acid [42]. These compounds are crucial for the formation of the sensory profile of coffee, as they act as precursors of several volatile compounds [43]. Therefore, the influence of enzymatic treatment in the reduction in trigonelline and chlorogenic acid concentrations represents a promising result with potential benefits for the sensory quality of coffee. According to Borém et al. [43], lower levels of trigonelline and chlorogenic acids correlate with higher sensory scores in coffee analysis. During coffee roasting, several reactions simultaneously co-occur with the Maillard reaction, including the degradation of bioactive compounds such as trigonelline and chlorogenic acids. As the temperature increases, trigonelline can be converted into vitamin B3 and volatile compounds, while chlorogenic acids can be converted into lactones or incorporated into melanoidins [38].

Thus, the sum of factors such as enzymatic action, fermentation time with the action of various microbial strains, and the influence of temperature on thermolabile compounds can explain the reduction to non-detectable levels in chlorogenic acids in roasted beans treated with 10 U·mL⁻¹ and fermented for 96 and 120 h. Corroborating these results, Tai et al. [8] obtained more significant reductions in 5CQA concentrations in wet-fermented coffees with added enzymatic extract when compared to fermentation without enzymatic treatment over 24 h of fermentation.

Enzyme treatment at a concentration of 10 U·mL⁻¹ proved promising in enriching the volatile compound profile, mainly contributing to the formation of alcohols. Alcohol synthesis during fermentation is associated with yeast metabolism, which degrades sugars or triggers reactions favored by the activity of FAD+ [44]. The presence of benzoic alcohols (benzeneethanol and benzenemethanol) in beans treated with an enzyme solution at 10 U·mL⁻¹ contributes to the development of fruity, sweet, floral, and chocolate aromas [44]. The correlation between samples C10H48 and C10H72 and the region of highest volatile density, observed in the PCA, also indicates the importance of pyrazine group compounds (2-ethylpyrazine and 2-methylpyrazine) in the volatile composition of these samples. Pyrazines, primarily formed during the Maillard reaction and Strecker degradation, contribute to aromas such as green grass, raw vegetables, potatoes, or pineapple [45]. The diversified volatile composition in coffees treated with an enzyme solution at 10 U·mL⁻¹ corroborates the results for bioactive compounds such as trigonelline and chlorogenic acid (5CQA), where higher degradation rates favor the formation of volatiles.

The overlay of the PCA graphs (Figure 3) also shows that the samples C5H72 and C5H120 are highly correlated with the region of most significant clustering of volatile compounds. Similarly, samples treated with enzyme solutions at a concentration of 5 U·mL⁻¹ showed significant degradation of trigonelline and chlorogenic acid, influencing the volatile composition. These samples differed from beans fermented without enzyme treatment by the presence of volatile compounds such as 2-ethylpyrazine, which imparts nutty, buttery, and woody aromas [46], and trans-linalool oxide, which contributes to floral aromas [47]. Compared to beans treated with an enzyme solution at 1 U·mL⁻¹, the 5 U·mL⁻¹ treatment presented 2-pentylfuran, a compound associated with floral, spicy, cereal, nutty, and vegetable aromas [48].

For the beans treated with 1 U·mL⁻¹, alcohols such as benzeneethanol, benzenemethanol, and propanoate-2-buten-1-ol, along with compounds like 2-methyl hexanoic acid and ethyl-pyrazine, were identified. These were not present in the beans fermented without enzyme treatment. These results suggest that the use of exogenous pectinases, even at the lowest concentrations tested, promotes the formation of compounds of interest for the volatile composition of coffee. In contrast, unfermented beans (C0H0) and beans fermented for 24 h without enzyme treatment (C0H24) had higher levels of furfural, associated with sweet, caramel, and nutty aromas [49], and 5-Methyl-2-furaldehyde, which is linked to spicy and sweet aromas [50]. Additionally, all samples exhibited high levels of 2-methyl-Butanoic acid, a volatile compound that contributes to fruity, acidic, and fermented aromas [51].

Several studies have reported the influence of fermentation on the modulation of volatile composition. Mota et al. [7] observed that the SIAF method modified the concentration of volatiles, and the sensory score was strongly related to the presence of aldehydes. Tang et al. [52] observed that solid-state fermentation can modulate the volatile composition, favoring the formation of pyrazines, pyrroles, furan, and pyridines. These results corroborate the impact of the SIAF method on the volatile composition of coffee beans. However, the unprecedented findings presented in this study point to a positive contribution due to the enzymatic treatment. Therefore, the different fermentation times and various enzymatic treatments applied in this study have the potential to enhance the sensory quality of coffee, enriching the volatile composition directly related to the sensory attributes of the beverage.

5. Conclusions

Enzymatic treatment effectively accelerated changes in the fermentation medium, catalyzing the breakdown of reducing sugars, water release, acidification of the medium, and degradation of coffee fruit pigments. Simultaneously, it influenced the antioxidant potential of the beans, promoting the degradation of phenolic compounds and, consequently, reducing their antioxidant activity. In addition, supplementation with more concentrated enzymatic solutions favored the degradation of bioactive compounds, such as trigonelline and chlorogenic acid (5CQA).

The different treatments resulted in several volatile compound profiles as a result of supplementation with higher concentrations of enzymatic solutions. The variable volatile profiles suggest that enzymatic treatment with different concentrations allows modulation of the chemical composition of the beans, offering potential for coffee fermentation processes that aim to obtain specific volatile profiles that meet diverse consumer preferences. Furthermore, the application of enzymes through spraying has shown promise, allowing the action of endogenous enzymes as a catalyst for coffee bioprocessing.