The açai palm (Euterpe oleracea) is native to the Amazon basin, a humid tropical forest. High levels of total mesophilic bacteria with high diversity have been consistently reported in açai fruits. As local consumers have few digestive problems, the results of the present study reveal the lactic acid bacteria (LAB) recovered from açai fruits with characteristics that suggest they are possible candidates for probiotics and antagonistic potential against pathogens for the first time. Açai fruits were sampled from five different locations in the Eastern Amazonia floodplains. Sixty-six isolates were recovered from fruits and tested for some probiotic characteristics following FAO/WHO guidelines. Approximately 65% of the isolates showed no catalase or oxidase activity, Gram-positive staining or cocci and bacilli cell morphology. Furthermore, 48% of the isolates demonstrated preliminary characteristics that suggest safety for use, as they presented no coagulase enzyme activity or gamma-hemolysis. These strains were identified as belonging to the genera Lactiplantibacillus and Pediococcus, and 32 strains also presented resistance to vancomycin, ciprofloxacin and streptomycin. In addition, 28 isolates showed a survival rate, expressed as log cycle reduction, higher than 0.9 under gastric conditions (pH 2). All strains tested positive in bile salts deconjugation tests and showed a survival rate higher than 0.8 in the presence of this salt. Regarding antimicrobial activity against pathogens, all strains were able to inhibit Salmonella Typhimurium (ATCC^® 14028^TM) and 97% were capable of inhibiting Escherichia coli (ATCC^® 25922^TM). Concerning the results of in vitro antagonistic assays, three isolates (B125, B135, and Z183 strains) were selected for antagonistic tests using açai juice contaminated with these two pathogens. All tested LAB strains were able to inhibit pathogen growth in açai juice. In summary, açai fruits are a potential source of LAB isolates to be investigated as probiotics.

Reliable analytical methods to authenticate high-quality and economically valuable cocoa beans are highly desirable, and NIR spectroscopy stands out as a rapid and non-destructive alternative. This study employs NIR for the authentication and differentiation of 19 whole Forastero cocoa beans from Eastern Amazon (Pará, Brazil), based on their fermentation status and genetic profiles. Partial Least Squares Discriminant Analysis (PLS-DA) models in wavelength ranges of 400-700 nm, 1400-1600 nm, and 1900-2500 nm, demonstrated high sensitivity and specificity in discriminating fermented from unfermented beans, regardless of genotype. Various compounds, including proteins, lipids, carbohydrates, anthocyanins, and theobromine, provide crucial insights into the spectral regions essential for distinction. The variable importance in projection (VIP) score value greater than 1 was used to select relevant variables, and Linear Discriminant Analysis (LDA) was performed in both the visible range (472 nm and 636 nm) and the infrared range (2096 nm and 2278 nm), demonstrating that absorbances at two specific wavelengths are sufficient for discrimination. The t-distributed stochastic neighbor embedding (t-SNE) indicated a segregation trend of the genotypes based on classification by major genetic groups in unfermented beans, suggesting that the biochemical characteristics shared by them are more prominent before fermentation. The PLS-DA models based on complete vis/NIR spectra showed comparable results in discriminating the 19 cocoa genotypes in both fermented (0.14% prediction error) and unfermented beans (0.16% prediction error). The model's classification errors can be attributed to shared genetic ancestry among the samples, primarily in unfermented beans. This research corroborates the effectiveness of vis/NIR spectroscopy as a straightforward tool for whole cocoa bean authentication, providing rapid insights into genetic diversity regardless of their fermentation state.

The adsorption of five phenolic compounds of four different classes from aqueous batch solutions onto four styrene-divinylbenzene and acrylic resins (EXA 90, EXA 118, XAD 7, and XAD 16) was investigated regarding their equilibria, kinetics and surface-energy heterogeneity, and mechanisms of adsorption. The experimental equilibrium data were very well fitted to Langmuir and Freundlich models (R ² > 0.900). Three kinetic models (pseudo-first-order, pseudo-second-order, and an intra-particle diffusion model) were suitable for describing the experimental data, the pseudo-second-order kinetic model being the best one (p < 0.001). The adsorption energy values were low (<19.00 kJ/mol), suggesting a physical adsorption process. Driving forces involved in the adsorption of the phenolic compounds onto the resins were hydrogen bonding, π − π stacking and hydrophobic interactions.

The cocoa tree (Theobroma cacao L.) has seeds that after processing become a valuable agricultural commodity in the world. In Brazil, the state of Pará located in the Eastern Amazon is the main producer, accounting for more than 53% of the national production. Although the region is the largest producer, no studies are found in the literature containing data on the sensory quality of the beans. Thus, the purpose of this study is to establish the sensory profile of Amazonian cocoa from the main cocoa-producing regions of the Eastern Amazon (Lower Tocantins River, Northeast, West, Southeast and Trans-Amazon). The sensory profile was established from the Quantitative Descriptive Analysis (QDA), in addition to using an affective test to verify consumer preference for the chocolates produced. Physical, physicochemical and bioactive compound evaluations were carried out. Beans from different regions had a similar sensory profile; however, it was possible to observe some differences in certain descriptor terms. In the physical evaluation, the samples presented an acceptable commercial standard, and the humidity was within the values established by the legislation. It was found that the post-harvest and/or processing parameters had no influence on the quality attributes of cocoa beans.