Edited by: Alicia Rodríguez, University of Extremadura, Spain
Reviewed by: Chiara Montanari, University of Bologna, Italy; Svetoslav Todorov, University of São Paulo, Brazil
This article was submitted to Food Microbiology, a section of the journal Frontiers in Microbiology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Coffee harvested in the Caparaó region (Minas Gerais, Brazil) is associated with high-quality coffee beans resulting in high-quality beverages. We characterize, microbiologically and chemically, fermented coffees from different altitudes through target NGS, chromatography, and conventional chemical assays. The genera
The Caparaó is a region located in a mountainous territory shared by two Brazilian states, Minas Gerais and Espírito Santo (
Crop growing environment, plants genetic traits, and post-harvesting processes are among the essential drivers of coffee quality (
During fermentation, microorganisms consume carbohydrates or other organic compounds and proliferate (
Microbial communities usually change in response to the environmental conditions where fermentations are carried out and affect coffee quality. Those conditions include temperature, moisture, and altitude (
Further research regarding altitude vs. compounds content variation is needed because they directly influence the beverage flavor. For example, organic acids mainly affect the sweet flavor (
Recent advances in Next-Generation Sequencing (NGS) are now allowing a deep microbiota characterization during the fermentative process under different conditions in several countries (
The present study aimed to characterize the dominant microbial communities of bacteria and fungi present in self-induced anaerobic fermentations containing different altitudes coffees performed in the Caparaó region through a metataxonomic approach. Moreover, this study aimed to evaluate the effect of altitude and microbiota profile on the biochemical compounds profile (organic acids, bioactives, and volatiles) during the fermentative process.
Ripe fruits of
The fermentative processes for all coffees from different altitudes were carried out simultaneously in close batches at a farm located at 1,200 m to avoid any environmental interference and favor controlled conditions. The bioreactors were placed under an open storage house built with fences for fermentation and suspended terraces. Before filling the bioreactors with coffee, portable data loggers (INKBIRD) were placed inside the bioreactors to register the mass temperature during fermentation. Fermentation lasted 72 h, and sub-samples of approximately 100 g were taken after 48 h of fermentation for dominant microbiota profiling and metabolites evaluation. Fruits’ initial sugar content (Brix degree-°Bx) was measured with a refractometer (Sigma-Aldrich, St. Louis, MO, United States).
Total DNA was extracted from 48 h fermented coffee fruits collected in fields at 800, 1,000, 1,200, 1,400 m of altitude. One hundred grams of coffee fruits were vortexed in 50 ml of sterilized Milli-Q water for 10 min to detach the fruits’ microbial cells. Then the resulted suspension was transferred to another tube and centrifuged (12,745 RCF for 10 min at 4°C) to separate the supernatant and obtain a pellet. After the supernatant was discarded, 30 mg of the remaining pellet was used for DNA extraction with the QIAamp DNA Mini Kit, following the “DNA Purification from Tissues” protocol (Qiagen, Hilden, Germany). The purity of the extracted DNA was checked with a Nanodrop Lite spectrophotometer (Nanodrop Technologies, Wilmington, DE, United States) (260/280 nm ratio), and it was quantified by Qubit® 4.0 fluorometer using the dsDNA HS Assay kit (InvitrogenTM) according to the manual. The DNA integrity was also confirmed by electrophoresis in a 0.8% agarose gel with 1 X TAE buffer.
The NGS Soluções Genômicas performed sample preparation for sequence and sequencing in Piracicaba-São Paulo, Brazil. The V3-V4 regions of the 16S rRNA gene of bacteria and the ITS1 and ITS2 regions of fungi were amplified from the total DNA extracted. We used the primers 341F (5′-CCTACGGGNGG CWGCAG-3′) and 806R (5′-GACTACHVGGGTATCTAATCC-3′) (
The raw.fastq files were used to build a table of amplicon sequence variants (ASVs) with dada2 version 1.12 (
Organic acids of coffee fruits were evaluated after 48 h of fermentation. Three grams of coffee fruits were vortexed in Falcon tubes containing 20 mL of 16 mM perchloric acid and Milli-Q water at room temperature (25°C) for 10 min. The resulted suspension (without the fruits) was transferred to another tube, centrifuged at 12,745 RCF for 10 min at 4°C to obtain the supernatant. The supernatant was transferred to a new tube, and then its pH value was adjusted to 2.11 using perchloric acid and recentrifuged under the same conditions. The supernatant from the second centrifugation was filtered through a 0.22 μm cellulose acetate membrane (Merck Millipore, Germany) and directly injected (20 μL) chromatographic column.
The samples were analyzed using a high-performance liquid chromatography (HPLC) system (Shimadzu Corp., Japan) equipped with a detection system consisting of an UV–Vis detector (SPD 10Ai) and a Shimpack SCR-101H (7.9 mm 30 cm) column operating at 50°C, which was used to achieve chromatographic separation of water-soluble acids at a flow rate of 0.6 mL min–1. The acids were identified by comparison with the retention times of authentic standards. The quantification was performed using calibration curves constructed with standard compounds [malic and citric acid were purchased from Merck (Darmstadt, Germany), lactic and tartaric acid were purchased from Sigma-Aldrich (Saint Louis, MO, United States), acetic and succinic acids were purchased from Sigma-Aldrich, isobutyric and butyric acid were purchased from Riedel-deHaen (Seelze, Germany)]. All analyses were performed in duplicate.
The identification of caffeine, chlorogenic acid [5-CGA], and trigonelline was made using a Shimadzu liquid chromatography system (Shimadzu Corp., Japan) equipped with a C18 column, following the protocol proposed by
Coffee samples were defatted following the methodology described by
The polyphenols and antioxidants were extracted, according to
The total polyphenol content (TPC) was determined by a spectrophotometric assay (UV-VIS Spectrum SP-2000 UV, Biosystems) following the Follin—Ciocalteau methodology (
Two different methodologies were applied to measure the antioxidant activity of coffee extracts. In the first one, the 2,2-diphenyl-1-picryl-hydrazyl-hydrate (DPPH) radical scavenging assay was performed as follows: 0.1 mL of coffee extract was added to 3.9 mL of the DPPH radical solution (0.06 mM) and incubated at room temperature (25°C), in the dark for 120 min, then the absorbance was measured at 515 nm (Spectrophotometer UV-Vis Spectrum® SP-2000UV, Shanghai, China). Trolox was used as a standard. A calibration curve (
The second assay was performed with a 2,2′-azinobis (3-ethylbenzothiazoline-6-suslfonic acid) (ABTS) stock solution reaction (7 mM) with potassium persulfate (140 mM). The mixture was left in the dark at room temperature (25°C) for 16 h before use. The ABTS solution was diluted in ethanol to an absorbance of 0.70 ± 0.05 at 734 nm. Thirty microliters of the coffee extracts were added to 3.0 mL of the ABTS radical solution, and after 6 min, the absorbance was measured. Trolox was used as a standard. A calibration curve (
Volatile compounds were extracted from 48 h fermented coffee fruits using a headspace solid-phase microextraction (HS-SPME). Coffee fruits (2 g) were macerated with liquid nitrogen and placed in a 15 ml hermetically sealed vial. After equilibration at 60°C for 15 min, the volatile compounds were extracted at 60°C for 30 min. The desorption time on the column was 2 min.
The compounds were analyzed using a Shimadzu QP2010 GC model equipped with mass spectrometry (MS) and a silica capillary Carbo-Wax 20M (30 m × 0.25 mm × 0.25 mm) column. The operation conditions of analysis consisted of maintaining the oven temperature at 50°C for 5 min, then raised to 200°C at 8°C min–1 and maintained for 15 min. The injector and detector were kept at 230 and 200°C, respectively, and He carrier gas was maintained at a flow rate of 1.9 ml min–1. The volatile compounds were identified by comparing their mass spectra against those available in the NIST11 library. The retention Index (RI) for each compound was calculated using an alkane series (C10–C40) compared with those found in the literature.
Alpha and beta diversity analyzes were performed for the evaluated microbial communities. Each altitude richness and abundance were used to calculate the bacterial and fungal Shannon and Simpson diversity indices. Moreover, the relative abundance was calculated, and ASVs profiles were clustered for each altitude using the XLSTAT software (Addinsoft, version 2020.1.3). Bray-Curtis-based non-metric multidimensional scaling (NMDS) was used to evaluate the dissimilarities between the fungal community and organic acids and volatile compounds with the XLSTAT software (Addinsoft, version 2020.1.3).
The raw data normal distribution for statistical analysis was evaluated with the Shapiro-Wilk and Anderson-Darling tests. All values in the figures are expressed as averages. Standard deviations were calculated using the XLSTAT software (Addinsoft, version 2020.1.3). The Tukey test was run with
The initial °Brix value from coffee fruits was between 18 and 19.3 (
Fruits initial brix, coffee mass temperature, and microbial diversity indices.
800 | 18.6 ± 0.6 | 18 ± 0 | 22 ± 0.06 | 23 ± 0.15 | 2.281 ± 0.2 | 7.973 ± 1.7 | 3.334 ± 0.1 | 14.766 ± 2.0 |
1,000 | 18.6 ± 1.5 | 18 ± 0 | 24 ± 0.12 | 25 ± 0.06 | 2.018 ± 0.5 | 5.574 ± 2.2 | 2.721 ± 0.1 | 5.666 ± 0.3 |
1,200 | 19.3 ± 0.6 | 18 ± 0 | 20 ± 0.10 | 21.6 ± 0.06 | 2.005 ± 0.4 | 5.004 ± 2.0 | 0.924 ± 0.3 | 1.417 ± 0.2 |
1,400 | 18 ± 3.0 | 18 ± 0 | 22 ± 0.12 | 23 ± 0.15 | 1.661 ± 0.5 | 4.330 ± 1.3 | 2.983 ± 0.2 | 9.372 ± 1.0 |
A total of 63.966, 16.346, 42.238, and 19.727 filtered 16S rRNA partial gene sequences and 104.719, 194.033, 263.884, and 119.571 filtered ITS sequences were obtained for the altitudes 800, 1,000, 1,200, and 1,400 m, respectively.
Among the altitudes, 800 m had the highest bacterial richness with 18 genera assigned, and 1,000 m had the highest fungal richness with 166 species assigned.
The altitudes ASVs profiles were clustered, as illustrated in
Clustering of ASVs profile from amplicon sequence of the 16S and ITS region.
A total of 31 genera were assigned in the bacterial community, as shown in
Relative abundance of the microbial communities at different altitudes.
The genera
Regarding the fungal community, a total of 223 species were assigned, showing a yeast predominance (
Each altitude had a broad range of distinctive fungal species, which included
We also identified frequently described species grouped in the species below 1% of abundance (
Filamentous fungi species such as
Acetic, malic, citric, lactic, succinic, and tartaric acid concentrations were statistically different among the altitudes (
The PCA showed that 1,200 m and 1,400 m altitudes were correlated with citric, succinic, malic, and acetic acid (
The Pearson correlation between acid content and bacterial community is depicted in
The NMDS plot (
A total of 67 volatile compounds were detected. These compounds were classified as alcohols (19), esters (13), aldehydes (10), ketones (6), furans (5), phenols (4), pyrans (3), acids (3), alkanes (2), lactones (1), and pyrazines (1).
The total relative concentration of each chemical group was statistically significant. The following chemical groups had the most abundant relative concentrations: alcohols, phenols, aldehydes, lactones, and ketones. The alcohols 2-heptanol and 1,6-octadien-3-ol, 3,7-dimethyl- were the most significant among the other compounds with relative concentrations varying from 70.6 to 26.3 mg. g–1 and 37.5 to 11.8 mg. g–1 (
The PCA graph in
The Pearson correlation between the volatile compounds and bacterial profile is depicted in
The NMDS plot (
Caffeine total concentration was higher (44.5%) than those of trigonelline (29.9%) and chlorogenic acid (25.6%), mainly at 1,200 m with a significant value of 11.64 g. Kg–1 (
The PCA on
The aim was to characterize the microbial community and compounds profiles associated with fermented natural coffee from different altitudes. Out of the detected fungal community, yeast species were the most abundant possibly to the region’s high relative moisture and temperatures, which were within our ranges. Furthermore, this region has rainy summers (November to January) and cold and dry winters (June to August). The average annual rainfall ranges from 1,000 to 1,500 mm, and the average annual temperature ranges from 19 to 22°C (
The coffee’s microbiota in this work varied at different altitudes. Lower altitudes favored bacterial richness, meaning that altitude is a factor that affects this microbial group. The high
The LAB
For the first time, we report a high abundance of the Proteobacteria
Similarly, it is the first time
According to
The bacterial communities were also correlated with volatile compounds because they contribute to their production. In this work,
Filamentous fungi and yeasts also compose the microbial communities of coffee during fermentation. The fungal diversity varied depending on the altitude and diversity index. Genus
37.7% fungi were not present at all coffee growing altitudes, from which 17.16% represented the species that were only identified in coffee from 800 m, and 25.31, 12.29, and 5.74% in coffees from 1,000, 1,200, and 1,400 m. Therefore, even if coffee belongs to the same region, the altitude’s influence on the niches was evident. Since most abundant fungi species were yeast, those who are culturable can be isolated, studied, and use as inoculants for future fermentations in the Caparaó region. For this purpose, yeasts in high abundance such as
The species below 1% abundance were clustered together, meaning they were not as influential as higher abundance species. Consequently, the NMSD plots showed that high abundance microbiota influences acids and volatiles contents, which was also confirmed in the Pearson correlations. The same behavior was seen for tartaric acid but with low abundance species.
The relative abundance of most filamentous fungi was within the 1%, which was expected since their populations usually dominate after several drying days due to reduced water activity (
As expected, the alcohol group in this work had the highest number of compounds and content, possibly due to the high yeast abundance. Yeast uses the nitrogen compounds from amino acids to produce a pool of volatile alcohols (
In coffee, either alcohols or esters are significant because they confer the most sensed odor descriptors. In this study, low altitudes and microbiota are strongly associated with volatile alcohols; these were also the altitudes with the highest bacterial and fungal richness and probably influenced the alcohol quantity. Simultaneously, high altitudes and their microbiota are strongly associated with high contents of aldehydes and esters.
Caffeine, chlorogenic acid, and trigonelline concentrations in our work were in the same range as those previously detected at 800 m in
Hence, the antioxidant activity depends on time, temperature, nature of the substance, and concentration of antioxidants or other compounds (
This work microbial and chemical characterization revealed a new perspective of why coffee from the Caparaó region is different from other Brazilian regions. The altitude and other region characteristics drive shifts in the microbiota profile and abundance, favoring yeast communities during fermentation. Moreover, altitude and high abundance of microbiota affect acetic and citric acid concentration and volatile compounds. 800 m coffee favors bacterial richness, and 1,000 m favors fungal richness during fermentation under SIAF conditions. Yeast that resists low temperatures dominates the Caparaó region coffee’s (mainly from genus
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: NCBI SRA BioProject, accession no:
RS contributed to the experimental design and revision and edition of the manuscript. JS contributed to the study design and data collection. VP supervised and contributed to the manuscript revision. SM collected and analyzed the data, performed the statistical analysis, and wrote the manuscript’s first draft. All authors contributed to the article and approved the submitted version.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We thank the coffee producers from the Caparaó region for the raw coffee and the Caparaó Junior students for their help during our experiments. Also, Ana Paula Pereira Bressani for her collaboration and the Brazilian Microbiome Project (
The Supplementary Material for this article can be found online at: