Murta fruit (10 g, fresh or dried by different methods) samples were ground and placed into a solution (aqueous ethanol 40%) of 300 ml at a solid/liquid ratio of 1:30, stirred for 24 h at 40°C (Memmert bath, WNB 22, Schwabach, Germany), centrifuged for 10 min (2,870 g, 10°C), filtrated (#1 filter, Whatman International Limited, Kent, England), and concentrated (Büchi R-210 rotary evaporator) at 40°C in near vacuum. The resulting extract was lyophilized and stored until analysis.

According to Vega-Gálvez et al. (2016), Folin–Ciocalteu reagent colorimetric assay was used for the murta extracts for total phenolic content (TPC) determination. The results were expressed as the standard gallic acid equivalent per 100 g of dry matter (mg GAE/100 g d.m.).

The identification of phenols was performed by a high-performance liquid chromatography (HPLC) system (Santa Clara, Agilent 1200 series, CA, USA) equipped with a diode array detector (DAD). A column Kromasil 100-5C18 at 25°C was used (4.6 mm; Eka Chemical, Bohus, Sweden), and it was controlled by the software ChemStation. The chromatographic conditions were according to Rodríguez et al. (2014). The extracts were eluted at 0.7 ml/min, with 280 and 310 nm detection. The mobile phase was of 0.1% formic acid (A) and 100% acetonitrile (B)—from 87% A and 13% B, linearly graded to increase solvent B to 55% over 18 min; to 60%, from 18 to 23 min; and then a final return to initial conditions within 2 min (minutes 23 to 25). These conditions were repeated for the standard phenols (i.e., gallic, salicylic, chlorogenic, pyrogallol, protocatechuic, syringic, caffeic, vanillic, trans-sinapic, p-coumaric, ellagic, trans-ferulic, and trans-cinnamic acids, as well as pyrogallol, 3-hydroxytyrosol, catechin, tyrosol, and quercetin) and dissolved in methanol–formic acid (99:1). The concentration of each standard curve was within a range of 0.3 to 100 ppm, and the retention times and the absorption spectra to identify the phenolic compound in samples were used. The quantification was expressed as mg/100 g d.m. HPLC reagents were obtained from Merck KGaA (Darmstadt, Germany) and the standards were obtained from Sigma (St. Louis, MO, USA).

DPPH assay was carried out according to Rodríguez et al. (2014). A 50 µM DPPH (2,2-diphenyl-1-picryl-hydrazyl) methanol solution was prepared, and 100 μl of murta extracts (in 3.9 ml of DPPH) was incubated (30 min) under darkness. Assays were measured at 517 nm, and Trolox reagent (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) was used as a standard (0.1 to 1.0 mM standard calibration curve). Results were expressed in µmol Trolox Equivalents (TE)/g d.m.

ORAC assay was carried out according to Zhang et al. (2010) in a multiplate reader (Victor3 Multilabel Plate Reader, PerkinElmer, Turku, Finland). Extracts (40 µl) and 200 μl of fluorescein (100 µmol/L) were added in a well (96-well, polystyrene, OptiPlateTM-96 F HB, PerkinElmer, Turku, Finland) and incubated (37°C, 20 min). Thirty-five microliters of AAPH (0.36 mol/L) was added to start reactions. Excitation and emission wavelengths (485 nm and 535 nm) were used until fluorescence was reduced by 95% of the first reading and quantified by the difference between the sample and blank areas under the kinetic fluorescein decay curve and a calibration curve of Trolox. Results were expressed as µmol TE/g d.m. (standard calibration curve Trolox 6.25 to 200 µM).

This assay was evaluated using an in vivo test described by Schmeda-Hirschmann et al. (2014). Murta extracts were tested in male adult mice (20–25 g each mouse, nomenclature CF-1, ISP Chile, Instituto de Salud Pública de Chile), in adherence with published guidelines and with the University of Chile IRB and ISP approval (Code CBE-2012-16). To identify the anti-inflammatory action mechanism of murta extracts, inflammatory agents TPA (phorbol 12-myristate 13-acetate) and AA (arachidonic acid) were tested against extracts and non-steroidal anti-inflammatory drugs (indomethacin and nimesulide) as control. For each of the samples under study, the anti-inflammatory activity was evaluated in two groups: one group consisting of 8 treated mice and another group composed of 16 control mice. Briefly, mice were treated with murta extracts, controls TPA (5 mg in 20 ml acetone) or AA (2 mg in 20 ml acetone) at the same concentration. Treatments (10 µl each) were applied over the inner and outer folds of the right ear while the other ear was swabbed with acetone. Following 6 h of TPA and 1 h of AA, and then cervical dislocation, ear samples were taken (6 mm diameter) and characterized. The topical anti-inflammatory effect (E) was determined following Equation 1:

where Wc and Ws are the median weights of control and treatment ear samples, respectively; all murta extracts were evaluated at 3 mg/ear doses.

Nine compounds (pyrogallol, vanillic, catechin, 3-hydroxytyrosol, gallic, protocatechuic, tyrosol, quercetin, and trans-cinnamic acids) were detected in the phenolic fractions of the fresh and dried murta fruits ( Table 2 ). Other authors have also identified the same compounds, showing them to be characteristic of murta berries (Alfaro et al., 2014; Augusto et al., 2014; Junqueira-Gonçalves et al., 2015; Jofré et al., 2016; Rodríguez et al., 2016; Fredes et al., 2020; Gómez-Pérez et al., 2022). Trans-cinnamic acid was least present in all drying methods and not quantifiable in fresh murta fruit. Similarly, protocatechuic acid, vanillic acid, and quercetin were undetectable in fresh murta. For FD products, protocatechuic and vanillic acids were not detected. In all the drying methods, there was a significant decrease (p < 0.05) of 3-hydroxytyrosol, catechin, and tyrosol compared to the fresh sample, and higher values of dried samples of pyrogallol, quercetin, gallic, and vanillic acids are observed compared to the fresh sample. Notably, vanillic acid was not measured in FD samples.

The most present phenol in fresh murta berries was catechin (172.72 mg/100 g d.m.), a flavanol (Grzesik et al., 2018). In contrast, the lowest were gallic acid (4.57 mg/100 g d.m.) and 3-Hydroxytyrosol (24.73 mg/100 g d.m.). In all dehydrated samples, pyrogallol was the most present phenolic compound, most significant in the CD drying method (506.89 mg/100 g d.m.), over 7.3 times more than fresh murta berries. The second phenol most common in all dehydrated samples was catechin, which, although significantly decreased compared to fresh berries, suggests resistance to thermal degradation, as mentioned in López et al. (2019). Of the phenolic compounds found in lesser concentrations, pyrogallol—falling into the categories of gallocatechin (Iglesias et al., 2013), a flavonoid (flavan-3-ol) (Du et al., 2012), and trihydroxybenzene (Florence et al., 2018)—has shown to have antibacterial, antioxidant, and anti-inflammatory activity (Furuno et al., 2002; Nicolis et al., 2008; Lima et al., 2016).

Then, drying methods alter the phenolic profiles of murta berries. It may also be presumed that the changes occurring during the distinct drying and temperature processes may be associated with compound instability. Indeed, Ortiz et al. (2021) suggested that processes above 60°C may induce oxidization in more susceptible compounds.

Speisky et al. (2022) mentioned the effects that the oxidative and non-oxidative modification of the phenolic groups of flavonoids might have on the ability of the resulting metabolites to promote direct or indirect antioxidant reactions and include other molecules with reducing capacity. Moreover, according to Fabani et al. (2017), the content of individual phenolic compounds increases with the drying process, mainly due to the concentration effect and the hydrolysis of polymerized phenolic compounds. Consequently, phenol content values vary widely, either fresh or dried fruits.

Figure 1 shows the results of TPC for fresh and dehydrated murta by different drying methods. Fresh murta contained 3,327 mg GAE/100 g d.m. of phenols, much higher than in other murta berry studies: López et al. (2019) reported 1,068.59 mg GAE/100 g d.m.; Augusto et al. (2014), 1,935 mg GAE/100 g d.m.; and Reyes et al. (2011), 1,460 mg GAE/100 g d.m. Additionally, Ramirez et al. (2015) reported 924 mg GAE/100 g d.m. in myrtle berry (Myrtus communis). The differences observed can be attributed to the different ecotypes of murta, raw material management, the location where they grow, harvest time, and soil conditions, among others, whether they are intrinsic or extrinsic conditions of the raw material (López et al., 2018; Fredes et al., 2020).

The TPC in dehydrated murta showed significant differences (p < 0.05) across all drying methods over fresh murta, ranging from 1,579 to 3,570 mg GAE/100 g d.m. for SD and FD, respectively. SD decreased 52% over fresh murta berries. Moreover, freeze-dried berries retained better phenol content than other drying methods. There was no significant difference between VD and IRD or CD and SD (p > 0.05).

These decreases in results may be due to oxidation, degradation, enzymatic inactivation, or volatilization during dehydration processes (Henríquez et al., 2014), especially since phenolic compounds are generally thermolabile (Fredes et al., 2020). Indeed, previous studies (Pap et al., 2021) have reported additional effects from activated polyphenol oxidase (PPO) on phenol structure.

The antioxidant potential of phenolic fractions of the fresh and dried murta fruits was determined by DPPH and ORAC assays in Figure 2 . In the fresh murta berries, the DPPH assay yielded 220 µmol TE/g d.m. and the ORAC assay yielded 4,851 µmol TE/g d.m. ( Figure 2 ). While ORAC values differ from previously reported values—perhaps related to the extraction method—the fresh murta DPPH assay is consistent with previous reports (Alfaro et al., 2013; Alfaro et al., 2014; Augusto et al., 2014; Rodríguez et al., 2014), yielding 21 to 457 µmol TE/g d.m. for different varieties of murta fruits.

All drying methods led to significant losses in the antioxidant potential, almost certainly due to enzymatic or thermal damage or inhibition of phytochemicals (Gao & Mazza, 1994). Notwithstanding, FD murta berries retained 78% of the total antioxidant potential by DPPH of the fresh berries, the highest among drying methods. These ranged between 33% and 44%, similar to previous reports (Alfaro et al., 2014).

In the ORAC assay, VD, IRD, and FD samples showed no significant differences; moreover, all samples exhibited significant decreases over fresh murta, between 45% and 65%, encompassing similar ranges reported by López et al. (2019).

In drying methods, temperature and time are principal contributing factors to the loss of antioxidant potential.

CD drying samples showed the lowest total antioxidant potential in both assays, likely due to the relatively intense and prolonged exposure to oxygen and subsequent enzymatic degradation of some of the phytochemicals. Despite losses in all methods, ORAC suggests relatively high antioxidant activity. This result may be due to pyrogallol content (CD, VD, SD, and IRD) ( Table 2 ), which has previously (López-Alarcón and Lissi, 2005) shown to react highly with peroxyl radicals.

However, for quantitative antioxidant analysis of natural materials, these assays should be replaced with new antioxidant assays, including lipid oxidation assays, with clearly identifiable reaction mechanisms, thoroughly tested reaction conditions, and kinetic and stoichiometric analyses (Schaich et al., 2015).

All murta fruit extracts obtained from the different drying methods exhibited inhibition of topical inflammation induced by TPA and AA (phorbol 12-myristate 13-acetate and arachidonic acid, respectively). Typical anti-inflammatory agents such as indomethacin (IND) and nimesulide (NIM) were used as controls to treat TPA- and AA-induced inflammation, showing 92.9% and 48.8% reductions, respectively ( Table 3 ). The effect of fresh murta fruit extracts obtained against TPA- and AA-induced inflammation was similar (18% and 15% edema reduction, respectively). FD murta sample extracts showed the highest anti-inflammatory response in both models of inflammation (65% and 41.3% edema reduction in the TPA and AA models, respectively), followed by VD extracts (63.2% and 34.2%, respectively). The IRD and CD methods showed a significant effect (51.2% and 47.5%, respectively) against TPA-induced inflammation. No significant topical anti-inflammatory effect was induced by IRD, CD, and SD murta samples’ extracts on AA-inflamed tissues. Previous reports (Delporte et al., 2007; Goity et al., 2013) show strong correlations between bioactive compounds in plant matter and inhibition of topical inflammation from plant extracts. Thus, the potent inhibition responses against TPA (>60% of FD and VD murta extracts over fresh fruit extracts) and AA (FD murta extracts) not insignificantly of the same order as the reference drug (NIM) are almost certainly caused by the presence of highly concentrated phenolic acids and flavonoids after FD and VD processes ( Table 3 ).

Furthermore, previous research (Inoue et al., 1988; Griswold et al., 1991) suggests that phenolic compounds may act on free radicals and inhibit some inflammation events, and, indeed, phenolic compounds identified in the murta extracts, quercetin, and gallic acid are known to elicit an anti-inflammatory response. Quercetin inhibits interleukin-1β (IL-1β), IL-6, and TNF-α, as well as STAT-1 and NF-κB (Hämäläinen et al., 2007). Next, two mechanisms of action have been proposed for inhibiting inflammation by gallic acid: neutralization of Dioxide(1−) radicals and suppression of MPO, and mediation in active NADPH-oxidase (Kroes et al., 1992).

The literature has an ample discussion on other bioactive compounds extracted from berries and their beneficial effects. Thus, Folmer et al. (2014) mentioned in their review the health benefits of bioactive compounds (flavonoids, ellagitannins, phenolic acids, and stilbenoids) from various types of berries (strawberries, red raspberries, black raspberries, blackberries, mulberries, and blueberries) concerning their anti-inflammatory properties and anti-carcinogenic effects. Similarly, Martin et al. (2014) found that polyphenols present in extracts of Aronia inhibited mechanisms leading to rheumatological and broad inflammation disorders in the neurological, joint, and digestive systems.

Arancibia-Radich et al. (2016) studied the effects of murta leaves across several variants and found associations between the phenolic compounds and triterpenoids present and the anti-inflammatory activity.

Cytotoxic activity on highly proliferative cell lines of the prepared extracts of murta berries was determined following standard procedure (Adenan et al., 2018) by subjecting human and mouse cell lines to MTT assays, which measure cellular metabolic activity and estimate cell survival.

Human cancer line NCI-H1975 and mouse line HT-22 cells were significantly repressed after 48 h of murta extract at a 0.25 mg/ml concentration. Figure 3 compares the dehydration method, controls, and measured cell viability.

Our data showed that no extract tested reduced cell viability below 71% of control in the two treated cell lines. Analyzing murta fruit extracts on the NCI-H1975 cell line, the fresh extract decreased cell viability by approximately 13% compared to the control. Comparing different drying methods, the VD and IRD are the most efficient methods to maintain the anti-tumoral effect on this type of cancer cells. This effect may be due to the phenolic components of murta extracts, especially given their potential oxidative capacity in the presence of specific metal ions (Caprioli et al., 2016). In the case of HT-22, an immortalized cell line with a high proliferation rate, the FD murta samples demonstrated the most significant activity, with 71.56% of cell survival, followed by fresh (77.34%), VD samples (77.59%), and CD (78.41%). These results suggest that this immortalized cell line is more susceptible to decreased cell survival in the presence of murta samples.

Chemical characterization of murta extracts showed the presence of polyphenols, with catechin as the most abundant ( Table 2 ). Catechin has been reported to have anticancer effects across many mammalian systems, e.g., digestive components, the endocrine system, circulatory and respiratory systems, and many others (Sheng, 2020). Bragança De Moraes et al. (2012) reported that catechin inhibited the proliferation of GRX liver cells, attributing the action to the inhibition of inflammation and the cell cycle. Additionally, Abd-Rabou et al. (2016) found anti-carcinogenic ellagic and gallic acids to be cytotoxic mediators against HepG2 and HCT 116 cell lines and decreasing viable human colon tumor cells; Flis et al. (2012) showed the effectiveness of antioxidant murta berry extracts—correlated with its phenols—against colon cancer cell lines; Avello et al. (2020) found that total aqueous murta leaf extract had synergistic effects in inhibiting cell viability against gastric adenocarcinoma cells (AGS) and that the metabolites of some elements even inhibited the viability of Caco-2 cells.