Juçara fruits were harvested from Euterpe edulis palm trees grown on the Bimini Farm (Rolândia, Paraná, Brazil). Lactobacillus reuteri LR92 (DSM 26866) and Lactobacillus plantarum BG112 (LMG 23520) were purchased from Clerici-Sacco (Cadorago, Italy), maltodextrin 10 DE was purchased from Ingredion (Westchester, IL, USA), gum arabic from Nexira SAS (São Paulo, Brazil), and bovine gelatin 220 Bloom from PB Leiner (Acorizal, Mato Grosso, Brazil). Glycerol, sodium hydroxide, ethanol, acetate buffer, DPPH solution, Trolox, luminol and phorbol myristate acetate (PMA) were all obtained from Sigma-Aldrich, Merck (São Paulo, Brazil).

Juçara fruit pulp was prepared as by Guergoletto et al. (17). In short, the fruits were sanitized for 30 min in water and chlorine (200 mg/kg), then the grains were rinsed and pulped with clean water (1:1) (fruit/water) using a pulper (Macanuda DM-Ji-05; Joinville, Brazil). The pulp was pasteurized in a water bath at 80 °C for 1 min and frozen at -20 °C until further use.

The bacterial strains (0.01% m/V) were frozen in pasteurized juçara pulp containing 20% (V/V) sterile glycerol. Prior to the assay, inoculum was activated in the juçara pulp and incubated at 37 °C for 18 h. The juçara pulp was fermented for 30 h at pH=5.7 and 5.0 (adjusted with 0.1 M NaOH) and 32 °C and 25 °C for L. reuteri and L. plantarum, respectively, as described by Guergoletto et al. (18). After fermentation, the phenolic content and antioxidant capacity were determined according to the methods described in Analytical procedures.

Three drying agents (gum arabic, maltodextrin and gelatin) were evaluated separately as carriers for the spray-drying process. Each was added to the fermented juçara pulp (with L. reuteri or L. plantarum) under aseptic conditions at 10% (m/V) (19), obtaining six samples. The samples containing maltodextrin or gum arabic were stirred for 10 min at room temperature, and the samples with gelatin were stirred for 20 min at 45 °C (16). Samples were spray-dried with a laboratory-scale spray dryer (LabPlant SD-05; Huddersfield, UK) equipped with a 0.7-mm nozzle and the main chamber (500 mm×215 mm). Prior to drying, the dryer conditions were uniformly maintained for 10 min with distilled water. The six samples (two bacteria strains×three drying agents) were continually stirred with a magnetic stir bar and fed into the main chamber via a peristaltic pump with a drying air flow of 60 m³/h and compressor air pressure of 0.11 MPa. The feed flow rate was 0.52 L/h, and the inlet temperature was (150±1) °C. Low input temperature conditions were selected for higher bacterial survival during drying, as described by Mestry et al. (20) and Anekella and Orsat (21), and the cellular viability results by Guergoletto et al. (16). Total phenolics, antioxidant capacity measured by DPPH^• radical scavenging assay, and oxidative stress in the neutrophil burst were measured before and after spray-drying.

The juçara pulp fermented by L. reuteri was chosen for storage analyses based on the neutrophil burst results. For these analyses, the samples were packed in biaxially oriented polypropylene (BOPP), with a light barrier and under vacuum and stored at 25 °C (representing ambient temperature) for 90 days. Every 30 days, total phenolic content, DPPH^•, and prooxidant effects from the generation of free radicals was measured (i.e. measurements on day 0, 30, 60 and 90). These parameters were chosen because the powdered products are usually stored at room temperature for longer than one month (22).

The total phenolic content and the antioxidant capacity determined by the DPPH^• method were expressed as mean value±standard deviation (S.D.) and compared using a Tukey’s post hoc test with a 95% confidence level in software Statistica v. 7.0 (28). The results were considered statistically significant if p<0.05. The results of the respiratory burst were first analyzed using a Gaussian distribution (Shapiro-Wilk test) and homogeneity of variance (Levene’s test). In the absence of a normal distribution and homogeneity of variance, the data were analyzed via a non-parametric Kruskal-Wallis test, complemented with Dunn’s test. The statistical analyses were conducted in GraphPadPrism^® software v. 3.02 (29).

Table 1 shows the results of the total phenolic content and DPPH^• radical scavenging analyses. Fermentation (with both bacterial strains) increased the total phenolic content and scavenging of free DPPH radicals in the juçara pulp (p<0.05). This is consistent with the results of our previous study (18). However, spray-drying caused a significant decrease in the observed values in both assays. Total phenolic content of fermented juçara pulp spray-dried with gum arabic, maltodextrin and gelatin was 65, 78 and 85% lower than the fermented and non-dried samples, respectively. Interestingly, total phenolic content of the juçara pulp fermented with either bacterial strain and spray-dried with gum arabic was higher than those spray-dried with maltodextrin or gelatin.

The reductions of fermented juçara pulp spray-dried with gum arabic, maltodextrin and gelatin were also significant for the DPPH^• method (Table 1); however, there were bacterial strain-dependent effects. The activities of L. reuteri fermented pulp dried with gum arabic and maltodextrin were 34 and 8% higher, respectively, than the samples fermented with L. plantarum and dried with the same carriers. There were no significant differences in spray-drying with gelatin between the bacterial strains.

The reactive oxygen species (ROS) released during the reaction can conveniently be detected by the chemiluminescent reagent luminol, and after every minute, the equipment detects these species. When samples are added to phorbol myristate acetate (PMA) and neutrophil cells, it is possible to detect a reduction of the absorbance readings due to the production of fewer species in the presence of antioxidant compounds. When the neutrophils were stimulated with PMA in the presence of the all tested samples, there was a significant inhibitory effect on ROS production (p<0.001), as demonstrated by the significantly lower peak values of the samples than of the control (Fig. 1). Among the samples, the ones fermented with L. reuteri and spray-dried with gum arabic were more effective in inhibiting ROS produced by the neutrophils (0-10 000 cpm). These results indicate that the fermented pulp and the spray-dried pulp had an antioxidant effect in vitro. This effect occurred through the elimination of the dioxide anion (O₂ˉ) in the respiratory burst of the neutrophils via electron donation, measured by DPPH^• method.

The antioxidant capacity of the juçara pulp may be associated with the presence of phenolic compounds and anthocyanins, whose chemical structures provide reducing power (7, 26). The bacterium-specific fermentation differences observed in this study may stem from the characteristics of each species. According to Kullisaar et al. (30), some species of lactic acid bacteria possess antioxidant enzymes such as superoxide dismutase (SOD), NADH oxidase, NADH peroxidase and catalase, and L. reuteri has a greater ability to reduce oxidative damage during endotoxic shocks than other Lactobacillus species (31).

The bioactive compounds of plants can degrade during spray-drying, but in this study, the fermented and spray-dried juçara pulp inhibited the neutrophil oxidative burst better than the pulp extracts. This result can be explained by the potential protective effects of the carriers (gum arabic, maltodextrin and gelatin) on the antioxidant activity and bacterial cells. Thus, our results suggest that the spray-drying does not affect the inhibitory effects on ROS production.

The fermented pulp and that dried with gum arabic had higher contents of phenolic compounds and greater protective effects on the antioxidant components of the fermented juçara pulp powder. Oliveira et al. (32) found that the powders produced from these materials contain high bulk density particles with less air, which may lower the rate of oxygen diffusion and, consequently, decrease the oxidation of the compounds. Tonon et al. (22) evaluated the use of carrier agents (maltodextrin, gum arabic and tapioca starch) in the drying of açaí pulp and found no differences in the anthocyanin content and antioxidant capacity when using maltodextrin or gum arabic.

Juçara pulp was fermented with L. reuteri, spray-dried, and then stored at 25 °C for 90 days. The samples were evaluated every 30 days, due to their high antioxidant capacity observed in the previous experiment. The total phenolic content and antioxidant capacity (measured by the DPPH^• method) of the samples decreased significantly over the 90 days (p<0.05; Table 2). These changes were less pronounced when the samples were dried with maltodextrin (24% decrease of total phenolic content and 19% of DPPH scavenging), followed by gum arabic (decrease of 50.7 and 69%) and gelatin (decrease of 61 and 78%). Tonon et al. (22) observed higher anthocyanin retention and higher antioxidant activity during storage with maltodextrin 10 DE. According to the authors, higher stability is associated with lower water activity, which is consistent with our previous study of fermented juçara pulp spray-dried with maltodextrin, gum arabic and gelatin (16). Rodríguez-Hernandez et al. (33) attributed higher stability to the better binding properties of maltodextrin 10 DE, which has a higher polymerization degree.

In the electronic paramagnetic resonance (EPR) analysis, a small free radical signal (Fig. 2) was detected in all of the samples and the unfermented and lyophilized juçara pulp. Maltodextrin, gum arabic, gelatin and the lyophilized bacteria were also analyzed to clarify the origins of the free radicals; however, none presented EPR signals. The EPR analysis of free radicals in the dried and stored pulp revealed similarities across the samples, with an increase in the number of species per g sample after 90 days of storage (Table 3 and Fig. 2).

High-temperature dehydration may degrade anthocyanins (34) and oxidize some lipid compounds, leading to the formation of or increase in the free radicals in dehydrated samples (35). However, in this study, the characteristics of the spectroscope factor (g) and the peak-to-peak line width (ΔH_pp) of the EPR spectra were similar for both the fermented and spray-dried juçara pulp and the lyophilized (control) unfermented pulp, suggesting that the atomization process did not influence the formation of free radicals, and that its origin can be attributed to the juçara itself or to the method of extraction of its pulp. In addition, the antioxidants present in juçara pulp might have acted to prevent the formation of more free radicals or the formation of a less active radical (36), because phenolic compounds are present in plants as a form of protection against ultraviolet and visible light (37). Similar values ​​of the ΔH_pp factor suggest as a strong indicator that the same radical was present in all the juçara pulp samples and the low g-factor values ​​are characteristic of unpaired electrons allocated to oxygen and carbon present in the pulp of juçara with a low connection to the spin orbit (38).

The weak singlet signal obtained (g=2.004) and peak-to-peak line width (ΔH_pp=0.4000±0.0005 mT) are common in plants and result from their metabolic activities (39). Although their origin is not fully understood, it has been suggested that these compounds are semiquinone or lignin-like free radicals, remnants of plant metabolism or produced by the oxidation of polyphenols or plant fatty acids (40). Studies suggest that the lignin is degraded by extracellular enzymes in smaller portions, which makes it susceptible to absorption by microbial cells, where it is then converted into phenols and quinones that are released with oxidative enzymes into the environment to be polymerized by the mechanisms of free radical formation (41-43). Alternatively, the process of pulping açaí and juçara in a depulper (or mixer) with water introduces excess air and may encourage the oxidation of fruit lipids. Further study is needed to elucidate the origins of free radical formation in juçara pulp.

Spray-drying requires high temperatures and is widely used for the dehydration of food and pharmaceutical products, due to its short time requirements and applicability to products with sensitive components (11). Our results suggest that spray-drying of fermented juçara pulp does not influence the production or proliferation of free radicals. We observed similar free radical content across juçara samples and the increase in the number of species after 90 days of storage was still lower than the content of radicals observed in herbal studies (10¹⁸ spin/g) (9) and sterilized mulberry leaves (10¹⁸ spin/g) (38). The increase in free radicals may be related to reduced antioxidant capacity and phenolic contents after storage. Katsube et al. (34) reported that the anthocyanins present in juçara are very sensitive to storage, particularly the temperature and contact with oxygen, which can directly influence the antioxidant capacity of the fermented and dehydrated juçara pulp. Even though the fermented and dehydrated juçara pulp showed reduced antioxidant capacity and a higher number of free radical species after storage, it is still noteworthy that both processes were effective in maintaining the beneficial properties, as evidenced by the lack of an increase in the intensity and the types of free radicals (as observed by EPR soon after drying). This result is supported by the protective effects observed in the neutrophil burst assay.