Natural Extracts from Eugenia brasiliensis Lam Leaves to Improve the Shelf-Life of Fresh Tomatoes

In this work, hydroalcoholic extracts were obtained from Eugenia brasiliensis leaves as a source of bioactive compounds of potential application to fresh tomatoes for preservation. Analyzes of total phenolics, flavonoids, antioxidant and antibacterial activity, and toxicity were performed. Tomatoes were treated with the extracts by brief immersion in an aqueous solution at concentrations of 200, 600, and 1000 mg L−1, and evaluated weekly for 28 days for total soluble solids, titratable acidity, maturation index, ascorbic acid content, weight loss, and visual test. The presence of gallic acid, rutin, and quercetin was identified in all extracts. The extracts obtained by water, ethanol at 25% e 50% showed better results regarding the concentration of phenolic compounds, antioxidant and antimicrobial activity. However, the 25 and 50% ethanol extracts were the only ones that showed toxicity. The treatments proposed for the conservation of tomatoes, in concentrations superior to 600 mg L−1, showed the most promising results, mainly in protecting the fruits from water loss and, consequently, visual depreciation due to wrinkles. Thus, most of the proposed treatments demonstrated viability by conserving fresh tomatoes organoleptic properties.


Introduction
The Eugenia brasiliensis Lamarck (EBL) is a fruit-bearing plant native from the Atlantic Forest, popularly known as "grumixama" or "Brazilian cherry". The EBL specie has dense treetop and dark green leaves, which are used in folk medicine for therapeutic and pharmacological purposes [1]. The interest in these leaves concerns their high content of biologically active compounds [2,3], mainly flavonoids, such as catechin, myricetin, isoquercetin, rutin, quercetin, galangin, and apigenin [4,5]. Among the benefits conferred to the bioactive compounds from EBL leaves have already been described: antioxidant and antilipemic activity [6], anti-inflammatory [7], and antidepressant [8], as well as have been reported antifungal [9] and antibacterial activity [10].
Besides the healthy benefits, the phenolic compounds from natural sources are potential replacements for some synthetic preservatives used in the food industry, reducing oxidative and microbiological deterioration of the food products, for example [11]. Such effects are essential for maintaining the nutritional and sensory quality of foods, considering that products of plant origin sold in nature present greater problems due to their perishability when compared to processed foods.
Natural extracts are potentially interesting in maintaining the nutritional and sensorial quality of fresh products, as fruits and vegetables, which have significantly lower shelflife than processed food. These extracts can be powerful allies to the conventional postharvest technology, extending the fresh food's shelf-life and contributing to their availability to a more significant part of the population [12][13][14][15][16]. However, there are few studies on the application of extracts obtained from the leaves of E. brasiliensis in foods.
The bioactives' profile on the EBL extracts resulted from the employed extraction method, which is the preferred solid-liquid extraction (SLE) technique due to their low cost and simple operation [17]. Water and ethanol stand out as low cost and biocompatible solvents, besides the readily removed [7,18].
Considering the exposure, this work aims to evaluate the use of hydroalcoholic solvents (water, ethanol, and its mixtures) in the extraction of bioactive compounds from the leaves of Eugenia brasiliensis Lam. (EBL). For this, the processes conditions were determined regarding the content of the extracts' total phenolics, total flavonoids, and antioxidant activity. Sequentially, was determined the phenolic profile by high performance liquid chromatography (HPLC) and the toxicity by the Artemia salina method. Finally, the extracts were evaluated concerning the ability to improve the shelflife of fresh tomatoes.

Materials
The chemicals used in this work were: ethanol (C 2

Preparation of Extracts from EBL Leaves
The EBL extracts were obtained by solid-liquid extraction at 45 °C with constant stirring for 1 h, and a solid-solvent ratio of 1:30. The evaluated solvents were water, ethanol, and its mixtures (v/v): 75% of water + 25% of ethanol (25% EtOH); 50% of water + 50% ethanol (50% EtOH); and 25% of water + 75% of ethanol (75% EtOH). Subsequently, the mixtures were centrifuged (Thermo Scientific, Microcentrifuge Heraeus™ Fresco™ 21) at 3500 rpm for 15 min at 20 °C, and the supernatant was collected. The solvents 1 3 removal was carried out in a rotary vacuum evaporator at 40 °C for ethanol extracts and an L101 manual lyophilizer (Liotop, Brazil) for the aqueous extracts. The dry samples were stored at 4 °C until use.

Total Phenolic Content, Total Flavonoids, and Antioxidant Activity Assays
The dry extracts were resuspended in water and characterized by the following analysis: total phenolic content by the Folin Ciocalteu method [19], total flavonoids content [20], and antioxidant activity through the radical 2,2-diphenyl-1-picrylhydrazyl (DPPH) inhibition and of the iron reduction (FRAP) methods [21,22]. All analyses were carried out in triplicate. A detailed description of the methods carried out can be found in the Supplementary Material.

Identification of Phenolic Compounds by HPLC-DAD
The phenolic compounds of the extracts from EBL leaves were identified and quantified by a high-performance liquid chromatography equipment equipped with a photodiode detector (DAD) described by Belmiro et al. [23] with some modifications. The ACE 5 C18 column (Bath V13-7473) (250 mm × 4.6 mm, 5 µm particle size, 110 Å particle porosity) was used for the chromatographic separation at 35 °C with a constant flow rate of 1.2 mL min −1 . The sample injection volume was 20 µL. Mobile phase A comprised 0.1% (v/v) water: formic acid, and mobile phase B of acetonitrile. The column was eluted following a gradient system: 0-15 min, 8% to 30% B (5 min); 22-35 min, 30% to 70% B (10 min); and 35-40 min, B from 70 to 8%. DAD recorded spectra from 200 to 400 nm, and phenolic compounds were quantified using standard curves for gallic acid, rutin, and quercetin.

Antibacterial Activity
The antimicrobial activity of the extracts was evaluated by the disk diffusion method as described by Bauer et al. [24]. The standard strains evaluated were: Bacillus cereus (ATCC® 10876™) (Gram-positive), Escherichia coli (ATCC® 25922™) (Gram-negative), Staphylococcus aureus (ATCC® 25923™) (Gram-positive) and Salmonella enteritidis (ATCC® 13076™) (Gram-negative). The microorganisms were reactivated in Muller Hinton broth (Kasvi) at 37 °C for 24 h, then transferred to the saline solution (0.9%) until reaching approximately 10 8 CFU mL −1 , equivalent to a standard McFarland solution of 0.5 absorbance units recorded by spectrophotometer at 600 nm. Microbial cultures were seeded evenly onto sterile Petri dishes on Mueller Hinton agar (Kasvi). After complete absorption of the inoculum, sterile filter paper discs of 5 mm diameter were deposited and impregnated with 10 μL of dry extracts previously diluted in sterile distilled water at concentrations of 100, 50, and 20 mg mL −1 . The antibiotic Chlorophenicol (3%) was used as a positive control, and sterile distilled water, and ethanol as a negative control. The plates were incubated at 37 °C for 24 h, and in sequence, the diameters of the inhibition halos were measured and reported as millimeters of the inhibition zone.

Artemia salina Toxicity Test
The extracts of EBL leaves were evaluated in a toxicity test with Artemia salina larvae as described by Fonseca et al. [25]. For this, 0.2 g of Artemia salina eggs were added to 200 mL of 3.5% (w/v) sea salt solution and incubated at 37 °C for 48 h under constant lighting. After hatching, the larvae were captured with a Pasteur pipette and transferred to test tubes containing 5 mL of saline solution, totaling 10 individuals of Artemia in each tube. The extracts, previously diluted in sterile distilled water, were added to the tubes at concentrations of 0.05, 0.1, 0.25, 0.50, and 1 mg mL −1 . A potassium dichromate solution (K 2 Cr 2 O 7 ) (mg mL −1 ) was used as a positive control, and only salt water was used as a negative control. The tubes were incubated at 37 °C for 24 h, and the dead and alive individuals were counted to calculate the survival percentage. The lethal doses of each extract for 50% of the population Artemia salina larvae (LD50) were calculated with logarithmic curve fitting, using the Statística 10 software. Dose-response curves were constructed using linear regression through the graph of sample concentration versus the percentage of dead larvae.

Preservative Effect on Fresh Tomatoes
The tomatoes used for the preservation trial were purchased from a vegetable supply center (CEASA) in Curitiba, Paraná. Fruits of uniform red color and free from any visible injury were chosen. The tomatoes were washed with a 2% (v/v) hypochlorite solution, in sequence with running water, and air dried.
For the treatment, the five types of dry extracts from EBL leaves were diluted in distilled water at 200 mg L −1 , 600 mg L −1 , and 1000 mg L −1 . Tomatoes were immersed in the different solutions for 5 min and then dried at room temperature. Tomato samples used as control were immersed only in distilled water. All samples remained exposed on a bench at room temperature for 28 days, and analyses to assess tomatoes' preservation were carried out on the first day (day 0) and after 7, 14, 21, and 28 days of storage.

Tomatoes Preservation Assessments
The parameters of total soluble solids (°Brix) were determined using a bench-top refractometer (Abbemat 500, Germany). The total acidity expressed as citric acid content, maturation index (°Brix/total acidity), and ascorbic acid content, were determined according to the methods proposed by Lutz et al. [26] for fruit analyses. According to Kamel [27], Tomato weight loss was evaluated from the difference between the initial and final fruit weight and was expressed as a percentage. The visual test was performed as described by Ishkeh et al. [28], with 10 raters, who selected scores on a scale of 1-10 for the visual attributes of tomatoes. Attributes include 1-2 = Excellent (Firm and bright in color), 3-4 = Good (Only firm), 5-6 = Acceptable (Slightly shrunk), 7-8 = Poor (Shrunk and with other depreciations such as black spots and cracks), 9-10 = Unacceptable (Very compromised surface with the appearance of rot). The results of the visual analysis were presented in terms of the reported mean score.

Statistical Analysis
All experimental results were obtained in triplicate and analyzed statistically using the Statistica 10 software. Data were analyzed by one-way analysis of variance (ANOVA) and then by Duncan's mean comparison test (p < 0.05).

Extracts Characterization
The dry extracts were resuspended in water and were characterized concerning their total phenolic content (TPC), total flavonoid content (TFC), and antioxidant activity (DPPH and FRAP assays), as shown in Fig. 1. The TPC of the extracts decreases in the opposite way of the ethanol concentration in the mixture (Fig. 1A). No statistical difference was observed between the extracts obtained with water, 25% EtOH, and 50% EtOH as solvents, which achieved a TPC yield up to 370.55 mg gallic acid equivalents g −1 dry extract (25% EtOH solvent).
On the other hand, the TFC increased with the higher ethanol amount on the mixtures, achieving an amount of 17.19 mg quercetin equivalents g −1 of dry extract where ethanol was the solvent (Fig. 1B). Flavonoids, such as quercetin and rutin, which are the significant bioactives on EBL leaves [5], have a hydrophobic character being the extraction favored by low polar solvents, such as ethanol (in comparison to water). This behavior can also be represented as a function of the logarithm of the biomolecules' octanol-water partition coefficient (log KOW). For example, the log KOW values for quercetin and rutin are 1.82 and 0.85, respectively Rothwell et al. [29] and Pedriali et al. [30], positive values of log K OW related to hydrophobic compounds, the non-polar behavior increases with higher values of log K OW.
Regarding the antioxidant activity of the extracts, those obtained using 25% EtOH as solvent showed the highest values for both DPPH and FRAP analysis, namely, 586.82 mg EtOH, 50% EtOH, 75% EtOH, and 100% EtOH. GAE gallic acid equivalents, QE Quercetin equivalents. Different letters represent significant differences among extracts (Duncan's test; p < 0.05) Trolox g −1 of dry extracts and 610.40 mg of ferric sulfate g −1 of the dry extract, respectively. The higher values obtained by the FRAP assay are related to the different mechanisms of both analyses. Compounds such as gallic acid in EBL leaves tend to have a higher ability to reduce the ferric complex; the same was observed by Rotili et al. [31].

Bioactive Compounds Identification by HPLC-DAD
The significant compounds on the extracts of EBL leaves were gallic acid, rutin, quercetin, cf. Table 1. The concentration of these compounds was determined through calibration standard curves, which showed good linearity according to regression analysis (R 2 > 0.9972).
The solvent 25% EtOH stands out in the gallic acid and rutin extraction, in agreement with the TPC analysis. In its turn, Quercetin, the most hydrophobic between the identified compounds, was found in higher concentrations for the 50% EtOH and 75% EtOH solvents.

Antibacterial Activity
Investigations regarding the antimicrobial activity of plant extracts have fomented interest in its application in food processing to replace chemical additives. Taking this into consideration, the disc diffusion method evaluated the EBL leaves extracts. Concerning its ability to inhibit four different bacteria strains commonly associated with food industry problems.
According to Oliveira et al. [32], no single evaluation criterion for the size of inhibition halos was observed for plant extracts. However, when applying plant extracts to foods, inhibition halos between 8 and 13 mm can be considered moderately inhibitory, while halos above 14 mm can be highly inhibitory. The inhibition halos of the extracts on the evaluated microorganisms are shown in Table 2. In general, inhibition halos were found from 5.87 to 14.11 mm, which varied as a function of the concentration of the resuspended extract; the most appropriate for this purpose is 100 mg mL −1 .
Those obtained when 25% EtOH and 50% EtOH were the solvents, at a concentration of 100 mg mL −1 , inhibited the growth of the four bacteria tested. It agrees with the high bioactive compounds present in these extracts. Additionally, the finds of this work reinforced the assumption that an extract's antimicrobial activity is directly related to its antioxidant activity.
All the extracts effectively inhibited the gram-positive microorganisms Staphylococcus aureus, Bacillus cereus, and gram-negative Salmonella sp. (Table 2). It is known that gram-positive bacteria are more susceptible to the mechanisms of action of antimicrobial agents due to the absence of the lipid barrier in their cellular structure. Gram-negative bacteria, in turn, have a complex cell structure composed of peptidoglycan, which usually makes it difficult for antimicrobials to penetrate their nucleus [33].
The inhibition activity can occur in different ways, such as impairing the enzymatic action of bacterial cells by directly linking with the enzyme, interfering in the cell's metabolic processes due to complexing essential metal ions, or modifying metabolic pathways by intercepting electrons [34,35]. So, it is believed that the inhibitory effect of the extracts on Salmonella sp. (negative gram) was caused by the action of the flavonoids, catechin, and quercetin. It agrees with the literature described, where these flavonoids showed inhibitory activity against Salmonella sp. In its turn, Escherichia coli strains had no effect in most of the evaluated concentrations [36,37].

Extracts' Toxicity
Considering the interest in applying the EBL leaves extracts to preserve fresh tomatoes, a preliminary test was carried out concerning its toxicity on Artemia salina larvae, as seen in Fig. 2. This test is based on the sensibility of the Artemia salina larvae to physicochemical changes caused when exposed to toxic components, being a simple and low-cost technique to evaluate plants' extracts [9,25,38].
As can be seen, the extracts from water, 75% EtOH, and 100% EtOH showed similar percentage survival compared to the control solution (salt water). The lethal concentration (LC 50 ) was also calculated for all solvents. The LC 50 values and respective regression lines are shown in Table 3. According to the scale proposed by Clarkson et al. [39], extracts from plants with LC 50 above 1000 µg mL −1 are not considered toxic, and from 500 to 1000 µg mL −1 are low-toxic.
The non-toxicity evidenced when water was the solvent explains the use of leaves of the EBL species by traditional medicine in the form of an infusion due to the report of their therapeutic and pharmacological properties, which are associated with the presence of the phenolic compounds revealed in this study. On the other hand, the lower concentration of phenolics found for the 75% EtOH and 100% EtOH extracts reflect a lower antioxidant activity, which consequently did not exhibit significant toxicity against A. salina.
There is related to the use of EBL species by the folkmedicine as infusion, as a route of consumption to take advantage of this species's therapeutic and pharmacological of this specie. In the agreement, the extract obtained using water didn't present toxicity on A. Salina, similar to the negative control. It is important to note that when 25% EtOH and 50% EtOH was the solvents, the extracts (1000 ugL −1 ) presented some toxicity. The results reveal that these extracts may show toxicity when used in high concentrations; however, the application proposed in this work concerning the application on surface of fruits or vegetables, which require sanitation practices before consumption, seems to be a promising alternative.

Effect of Treatments with Extracts of EBL Leaves on the Shelf Life of Tomatoes
The preservative effect of the extract on the tomatoes shelflife was evaluated through the following parameters: weight loss, visual analysis of fruit appearance, total soluble solids (Brix°), acidity titratable in citric acid, maturation index, and ascorbic acid content. The treatments were performed  According to Table 4, the weight loss of all samples increases over the days, achieving up to 27.33% for the control sample (without any treatment). Regarding the treated samples, which higher the extract concentration better was the protective barrier, and lower weight loss was observed. The weight loss directly affects the sensory characteristics of the tomatoes and is associated with the water loss from the fruits. As a consequence, the fruits' surface of the fruits takes on a wrinkled appearance. Besides, changes in the firmness profile were also observed. These factors directly interfere with consumer acceptance and market value since visual attributes measure the quality of plant-based products.
Ten evaluators analyzed the tomatoes' sensorial analysis by visual analysis (Table 4). After 28 days, most of the treated fruits had lower values in the evaluation of the decay of the attributes when compared with the control sample (9.70). The dehydrated aspect, the presence of stains, and the visible microbial deterioration were determining factors in the classification of the attributes of the tomatoes. Part of the samples are shown in Fig. 3; the complete figure can be found in the Supplementary Material. The resuspend extracts on a concentration of 1000 mgL −1 extract contributed to the most negligible reduction in weight loss (7.76%) and decay of the visual attributes of the fruits (4.10).
The titratable acidity of the tomatoes was expressed in citric acid, and the results obtained after the different treatments with the EBL leaf extracts can be seen in Fig. 4.
Beyond the visual changes, during the fruit ripening, chemical reactions such as changes on the titratable acidity (AT) also occur. In the case of tomatoes, AT tends to decrease since most organic acids are secondary metabolites consumed during respiration and in the enzymatic activity of the fruits during storage [12]. In addition, AT is an important parameter to evaluate the fruit quality being desirable to delay these reactions. A 600 mg L −1 was enough for the purposed extracts to reduce the AT changes during all the storage time. Another important parameter to evaluate the tomatoes' preservation is the concentration of total soluble solids (SST), as seen in Fig. 5. In general, the TSS content showed a tendency to increase during storage. However, it was observed that tomatoes treated with extracts at concentrations higher than 600 mg L −1 showed more constant SST levels from the 14th day of storage. The control sample, in its turn, showed a continuous increase in the SST values, indicating the fruit senescence. It occurs due to the combustion of organic acids during respiration, which is also converted into sugars due to the polysaccharides' hydrolysis. In addition, tomatoes are characterized as climacteric fruits, as at the end of maturation, they present an increase in respiratory rate and ethylene production [40].
After the AT and TSS determination, the ripening index was calculated (SST/AT), and the results are shown in Fig. 6. The results indicated that these treatments delayed the ripening similarly for all evaluated extracts. Tomatoes treated  The positive effects of immersion in the tomatoes on the EBL extracts reflected favorable ripening rates. It results from the phenolics on the extracts and their antioxidant and antimicrobial activities, which contribute directly to the shelf-life of the products. According to Moshari-Nasirkandi et al. [16], immersion treatments of fruits in solutions containing phenolic compounds can preserve the natural phenolic content of the fruits for a long time since these compounds are lost during storage.
Regarding ascorbic acid concentration, a significant decrease can be observed over the 28 days of storage (Fig. 7). The extracts obtained with the solvents 25% EtOH, 75% EtOH, and 100% EtOH, at concentrations of 600 and 1000 mgL −1 , were the most effective in delaying the losses of ascorbic acid during storage. According to Nair et al. [41], ascorbic acid protects against oxidative damage caused by the ascorbate peroxidase reaction. The results indicated the natural extracts' capacity to reduce fruit degradation after a simple immersion treatment.
Lastly, the SST and TA index was used to determine the half-life of tomatoes (Table 5), where it is possible to note an expressive increase in half-life (t 1/2 ) when 25% EtOH and 50% EtOH were the solvents, achieving a t 1/2 up to 2.45 times longer than the control sample.
Therefore, based on the results obtained, it is possible to affirm that the extracts efficiently preserved the organoleptic characteristics of fresh tomatoes, demonstrating beneficial bioactive properties for prolonging the shelf life of vegetable foods.

Conclusions
Hydroalcoholic extracts from Eugenia brasiliensis Lam leaves were obtained by a conventional extraction technique, and the extracts were evaluated considering their phytochemical characteristics and potential natural preservatives of fresh tomatoes. The extractions conducted with water solvents and 25% EtOH showed better yields in terms of total phenolic compounds when compared with 50% EtOH, 75% EtOH, and 100% EtOH. The same trend was observed in evaluating antioxidant activity in both methods (DPPH and FRAP). In addition, the presence of gallic acid, rutin, and quercetin was identified by HPLC in the five extracts evaluated. The extracts obtained by solvents water, 25% EtOH, and 50% EtOH stood out concerning the antimicrobial activity of the four bacteria tested when they were used at a concentration of 100 mg mL −1 , promoting inhibition halos considered moderately inhibitory by the literature consulted. However, it is suggested that the evaluated extracts present greater action when placed in contact with microbial loads lower than 10 8 (McFarland), considering that the results of this study can optimize future investigations.
The toxicity evaluation indicated that the extracts from water, 75% EtOH, and 100% EtOH have low or no toxicity.
In addition, the proposed treatments' feasibility was demonstrated to preserve fresh tomatoes. The results showed that the immersion treatments of the fruits, in solutions with the extracts of water, 25% EtOH, and 50% EtOH, in concentrations higher than 600 mgL −1 , were responsible for preserving organoleptic properties determinants in the acceptance of consumers' by-products of vegetable origin.