Aerial parts of Stevia: extraction model, adjuvant potential in metabolic syndromes and bacteriostatic effect

Abstract

This study investigated hydroethanolic extracts (70:30 v/v) from leaves and stems of Stevia rebaudiana (UEM-13 variety), obtained by Soxhlet and Shaker extraction, focusing on bioactive properties. UHPLC-MS/MS and FTIR characterization identified phenolic and sweetening compounds. Both extracts exhibited significant antioxidant activity in DPPH, TBARS, OxHLIA, and ORAC assays. Soxhlet extraction was more effective in preserving phytochemicals and inhibiting α-amylase, whereas Shaker extraction showed greater pancreatic lipase inhibition. Both extracts demonstrated antimicrobial activity against pathogenic and foodborne bacteria, as well as antifungal effects. These findings highlight Stevia biomass as a promising source of compounds with antioxidant, antidiabetic, antilipid, and antimicrobial potential.

Key words
Stevia TBARS; Stevia UHPLC-MS/MS Hydroethanolic Extract; Stevia ORAC; Stevia Antimicrobial Activity

INTRODUCTION

Stevia rebaudiana has been extensively studied due to its bioactive properties, especially for its phenolic compounds and glycosides present in its leaves, which confer antioxidant, antimicrobial, and antidiabetic potential (Cabeça et al. 2023). NEPRON, Natural Products Research Center of the State University of Maringá (UEM), achieved a significant breakthrough by developing a variety of this plant, called Stevia UEM-13, which can be reproduced through seeds. The efficient extraction of these compounds depends on several factors, including the method used and the composition of the solvent. Conventional methods such as extraction to exhaustion using soxhlet apparatus or with control of time, agitation and temperature using equipment or shaker apparatus have been used to optimize the obtaining of extracts rich in bioactive compounds, aiming pharmacological and nutraceutical applications (Milani et al. 2017a).

Another aspect that deserves to be highlighted is the predominant focus of studies with S. rebaudiana on the fragmentation of extracts from its leaves. The biomass of the Stevia plant (Figure 1) can be subdivided into leaves, stems, roots, flowers, and twigs, the leaves being often used in obtaining various extracts for commercial purposes. The stems, representing approximately 30% of the total biomass, however, have been the target of limited studies. In a pioneering approach, Nogueira et al. (2024) investigated antihyperglycemic and antilipidic activity of methanolic and hydroalcoholic extracts from the stems of Stevia UEM-13. Despite this, the stems are not industrially directed as a crucial source of sweeteners, so this biomass is considered a residue of the production chain, with reports of limited use, especially intended for enrichment in animal feed. The presence of potent antioxidants represents a relevant research task (Yu et al. 2017).

Figure 1
Aerial parts of Stevia.

Conventional and unconventional methods commonly used to extract phenolic compounds and Stevia flavonoids exclude sweeteners in an attempt to obtain richer, purified extracts. However, sweeteners have insulinotropic properties (Milani et al. 2017b) that can potentiate these extracts. Thus, among the objectives of this work are to obtain functional extracts with sweetening potential, which can increase the range of application of this product.

In this context, the present pioneering study investigates the chemical constituents and the functional effects of hydroethanolic extracts through the total exploitation of the aerial parts of interest of the plant Stevia, variety UEM-13, obtained by different extraction methods. The quantitative analysis of bioactive compounds, as well as the evaluation of antioxidant and antimicrobial properties, and possible antidiabetic and antilipid effects of extracts, will provide information on the possibility of full use of the plant, what can contribute to its greater use, increase in added value, reducing costs, avoiding the separation of parts, and also for sustainability of the production chain of S. rebaudiana.

The findings of this study also compared two conventional methods of extraction which can contribute to decision making of experimental protocols, providing for the development of new natural products with improved functional properties, offering safe and effective alternatives for applications in the food, cosmetic and pharmaceutical industries (Yu et al. 2022, Rahmani et al. 2024).

MATERIALS AND METHODS

Samples and reagentes

Stevia rebaudiana plants of the variety Stevia UEM-13 were grown in NEPRON - UEM (23°24’21.2”S 51°56’21.8”W). The plants were collected under the same conditions described by Formigoni et al. (2018). The stems, with their leaves, were harvested and the biomass was measured, being the composition of 80% leaves and 20% stems. This material was then previously dried in an air circulation oven at 60°C until the moisture content reached levels below 10% and subsequently ground. All solvents and samples are analytical grade or HPLC-MS. The absolute ethanol is from Merck (Londrina, Paraná, Brazil). Deionized water (18 MΩ cm) of the Milli-Q plus system were acquired from the company Induslab (Londrina, Paraná, Brazil). All reference compounds were provided by Sigma-Aldrich (Brazil).

The standards used in the UHPLC-MS/MS analysis were: gallic acid, caffeic acid, 4-hydroxybenzoic acid, vanillic acid, chlorogenic acid, naringenin, apigenin, luteolin, kampferol, quercetin, coumaric acid, 3-O-caffeoylquinic acid, 5-O-caffeoylquinic acid, quercetin-3-O-rutinoside, quercetin-O-hexoside, quercetin-3-O-glucoside, kaempferol-o-hexoside, kaempferol-rhamnosyl-glucoside, quercetin-O-pentoside, 3,4-O-dicaffeoylquinic acid, 3,5-O-dicaffeoylquinic acid, quercetin-O-rhamnoside, 4,5-O-dicaffeoylquinic acid, kaempferol-O-deoxyhexoside, quercetin dihexosyl deoxyhexoside, tricaffeoylquinic acid.

Obtaining the hydroethanol extracts.

Two extracts were obtained from the stems with leaves that were called extract soxhlet and extract shaker. In the sohxlet apparatus, the extraction of 100 grams of plant (leaf + stem) was carried out until exhaustion (10h), at a temperature of 55-60°C, using 700 mL of solvent (water: ethanol 70:30 v/v). The extraction in shaker equipment was carried out in the same mass (100g), temperature and solvent (water: ethanol 70:30 v/v), but were performed 3 cycles of 2 hours each and with agitation of 140 rpm. Subsequently, both extracts were filtered and dried in the rotary evaporator.

Characterization of hydroethanolic extracts

Quantification of total phenolic compounds and flavonoids

The total phenolic compounds result, following the methodology adapted described in Nogueira et al. (2024), was expressed in grams of galic acid equivalent per 100 g of sample, using a gallic acid standard curve (10-500 μg/mL; R2 = 0.9998). The quantification of flavonoids was determined and data were expressed in quercetin equivalent, using a quercetin standard curve (25-500 μg/mL; R2 = 0.9998), Nogueira et al. (2024).

1. Analysis of Phenolic Compounds by UHPLC-MS/MS

For the chromatographic analysis of phenolic compounds, a UHPLC-MS/MS system composed of an UPLC Acquity H-CLASS coupled to a triploquadrupole Xevo TQD mass spectrometer, equipped with a Z sprayTM ESI interface operating in positive and negative modes was used (Waters, Milford, MA, USA). The instrument control and data processing were performed with the software MassLynx and QuanLynx version 4.1 (Waters). The chromatographic separation of the phenolic compounds was conducted in a 50 mm 2.1 mm (internal diameter) Waters Acquity UPLC® BEH C18 column with a particle size of 1.7 μm. The column temperature was maintained at 30 ± 1 °C, and the total analysis time was 16 minutes, with a flow rate of 0.150 mL/min and an injection volume of 1.5 μL. The mobile phase A consisted of water with 0.1% formic acid, while the mobile phase B was methanol. Elution was performed by gradient: the gradient started at 10% of B and increased to 60% of B in 3 minutes.

Then phase B was elevated to 80% in 1 minute and maintained for 1 minute. After that, the gradient increased linearly to 100% of B in 5.5 minutes and was maintained for 1.5 minutes. Finally, the gradient was reduced to 10% of B in 4.5 minutes, and the column was balanced in this state for another 4.5 minutes. All experiments were carried out using negative electrospray ionization (ESI-). The ESI source block and dissolvation temperatures were adjusted to 130 and 550 °C, respectively, with a capillary voltage of 3 kV. The dissolvation and nebulization gas was nitrogen, operating at 700 and 50 L/min, respectively. Collision-induced dissociation was performed with argon at 3.0 × 10⁻³ mbar.

Quantitative analysis was performed using the standard addition method. Analytical calibration curves were established at five concentration levels, with concentration ranges defined according to the detectability of each target phenolic compound in the hydroethanolic extracts. This approach allowed for the evaluation of the linear dynamic range, limit of detection (LOD), limit of quantification (LOQ), precision, and matrix effect (ME). The LOD, defined as the lowest analyte concentration that produces a response distinguishable from system noise, and the LOQ, defined as the lowest analyte concentration that can be measured with acceptable precision and accuracy, were determined as the amounts of analyte that generate chromatographic peaks three and ten times larger than the background noise, respectively, in chromatograms of unfortified samples after endogenous concentration estimation. For phenolic compounds not endogenously detected in the sample extracts, the LOD and LOQ were calculated based on the peak signals obtained from known concentrations of the respective standards added to the extracts. The precision of the method was evaluated as intraday and interday precision, assessed over two consecutive days at the endogenous concentration of each phenolic compound, and expressed as relative standard deviation (RSD). The matrix effect (ME, %) for each target compound was determined by comparing the slopes of the calibration curves obtained by the standard addition approach (with matrix matching) and those prepared in solvent at identical concentration levels. Positive or negative ME values indicate increased or suppressed ionization, respectively (Carina et al. 2022).

Centesimal Composition

1. Quantification of glycosides

Glycoside content was quantified using the standards of stevioside, rebaudioside A and C according to Dacome et al. (2005). The mobile phase used was a mixture of acetonitrile and water (85:15 v/v ratio). After 5 minutes of the methodology, the samples were subjected to ultrasonification and filtration and injected using a high-performance liquid chromatograph (HPLC), Gilson brand, model 307 isocratic, coupled to an IR detector, NH2 column, flow rate of 0.5 mL/min. The total analysis time will be 30 minutes.

2. Quantification of total proteins, lipids, carbohydrates and moisture

The analyses were performed using different methodologies: Total proteins by Kjeldahl; Total carbohydrates using the methodology described by Dubois et al. (1956) and following the methodology of the Adolfo Lutz Institute (2008).

Identification of functional groups by infrared spectroscopy with Fourier transform (FTIR)

The functional groups of the samples were identified by infrared spectroscopy with Fourier transform (FTIR) (Shimadzu, IRTracer-100). For the analysis, the samples were mixed with KBr (1:100 m/m) and pelletized. An FTIR -BOMEN 100 spectrometer with 21 minimum scans 1 and a resolution of 4 cm 1. The FTIR spectra were in the range of 4000 400 cm 1 in the absorbance mode.

Particle Size

Particle size was measured in a laser light diffraction instrument, Litesizer (Anton Paar, USA), using a small sample of microparticles that were suspended in ethyl alcohol (99.9%). The particle size was expressed as nm.

Evaluation of bioactivity and functionality of extracts

Antioxidant properties

1. DPPH method (radical 2,2-diphenyl -1-picrylhydrazyl

Antioxidant activities were determined using DPPH as a free radical, according to the method adapted described in Nogueira et al. (2024). Trolox standard curve was used as a reference (0.5 - 250 μg/ml; R2 = 0.9991) and the results were expressed in g equivalent of Trolox/100g extract.

2. TBARS (Thiobarbituric acid reactive substances) assay

The assay was performed in extracts. Swine brains (Sus scrofa) were obtained from official slaughter animals, dissected and homogenized with Tris HCl buffer (20 mM, pH 7.4) to produce a homogenization of brain tissue 1:2 w/v, which was centrifuged at 3000 g for 10 minutes. An aliquot (100 μl) of the supernatant was incubated with the different concentrations of sample solutions (200 μl) in the presence of FeSO₄ 78 (10 mM; 100 μl) and ascorbic acid (0.1 mM; 100 μl) at 37 °C for 1 h. The reaction was interrupted by the addition of trichloroacetic acid (28% w/v, 500 μl), followed by thiobarbituric acid (TBA, 2%, w/v, 380 μl), and the mixture was then heated to 80 °C for 20 minutes. After centrifugation at 3000 g for 10 minutes to remove the precipitated protein, the color intensity of the malondialdehyde complex (MDA)-TBA in the supernatant was measured by its absorbance at 532 nm. The inhibition ratio (%) was calculated using the following formula: Inhibition ratio (%) = [(A B)/A] x100, where A and B were the absorbances of the control and sample solution, respectively. The results are expressed as EC₅₀ values (mg/mL), which means the extract concentration can inhibit 50% lipid oxidation.

3. Test OxHLIA (Oxidative Hemolysis Inhibition Assay)

The OxHLIA assay was performed using a 48-well flat-bottom microplate. For this, 200 μL of erythrocytes from sheep blood in PBS (2.8% w/v) were combined with 400 μL of the extract samples prepared in successive dilutions. PBS was used as a control, while milli-Q water was used to induce complete hemolysis. After pre-incubation at 37 °C for 10 minutes with shaking, AAPH (200 μL, 160 mM in PBS) was added, and the optical density was measured at 690 nm at approximately 10-minute intervals using a microplate reader (Bio-Tek Instruments, ELX800) until hemolysis was complete. The results were expressed as IC₅₀ values, representing the extract concentration (µg/mL) required to protect 50% of erythrocytes from oxidative hemolysis over 60 and 120 minutes.

4. ORAC test (oxygen radical absorption capacity)

The reaction was performed in phosphate buffer 75 mM (pH 7.0), and the final mixture of the reaction was 200 μL. The antioxidant (20 μL) and â-phycoerythrin (120 μL; final concentration of 16.7 nM) solutions were placed in the microplate well. The mixture was pre-incubated for 15 minutes at 37 °C. The AAPH solution (60 μL; 32 mM, final concentration) was added quickly using a multichannel pipette.

The microplate was immediately placed in the reader and fluorescence was recorded every minute for 80 minutes. The microplate was automatically stirred before each reading. A blank using phosphate buffer instead of antioxidant solution and 10 calibration solutions using Trolox (final concentration of 3-40 μM) as an antioxidant were also performed in each assay. All reaction mixtures were prepared in duplicate, and at least three independent tests were performed for each sample. The results were expressed in equivalent of Trolox (R2 = 0.9734).

Antimicrobial activity

1. Antibacterial activity

Antibacterial activity was performed according to Pires et al. (2018). The extracts were tested against food contaminants (bacteria): five Gram-negative bacteria, namely Enterobacter cloacae (ATCC 49741), Escherichia coli (ATCC 25922), Pseudomonas aeruginosa (ATCC 9027), Salmonella enterica (ATCC 13076), Yersinia enterocolitica (ATCC 8610) and three Gram-positive bacteria, Bacillus cereus (ATCC 11778), Listeria monocytogenes (ATCC 19111) and Staphylococcus aureus (ATCC 25923). All these microorganisms were acquired in Frilabo, in the city of Porto, Portugal. The bacteria were incubated at 37 °C in an appropriate fresh medium for 24 h before analysis to maintain the exponential growth phase.

The samples were also tested using clinical bacteria: the bacterial strains came from clinical isolates obtained from patients hospitalized in various Hospital Center of Trás-os-Montes and Alto Douro (Vila Real, Portugal). Five Gram-negative bacteria (Escherichia coli (VRU12881), Proteus mirabilis (VRU17684), Klebsiella pneumoniae (VRI17214), Pseudomonas aeruginosa (VRU14123) and Morganella morganii (VRU14272) and three Gram-positive bacteria Enterococcus faecalis (VRU14041), Listeria monocytogenes (VRU17684) and Methicilin Resistant Staphylococcus aureus (MRSA) (VRI17654).

The MIC determinations in all bacteria were performed using the colorimetric test according to the described by Pires et al. (2018). The samples were first dissolved in 5% (v/v) dimethylsulfoxide (DMSO) and 95% autoclaved distilled water to give a final concentration of 20 mg/mL for the stock solution. 100 μL of this concentration was added to the first well (96-well microplate) in duplicate with 90 μL of tryptic soybean broth (TSB). In the remaining wells, 90 μL of TSB medium was added. Then, the 80 samples were diluted in series to obtain the concentration ranges (0.03125 to 10 mg/mL). To finish, 10 μL of inoculum (standardized at 1.5x106 colony-forming units (CFU) /mL) were added in all wells, ensuring the presence of 1.5x105 CFU. Two negative controls were prepared, one with TSB and the other with extract. Two positive controls were prepared with TSB and each inoculum and one with medium, antibiotics and bacteria. Ampicillin and Streptomycin were used for all bacteria tested, ampicillin and streptomycin were selected as positive controls due to their broad-spectrum activity and widespread use in antimicrobial susceptibility assays, allowing comparison with previously published studies and Methicillin was also used for Staphylococcus aureus. The microplates were incubated at 37°C for 24 h. The MIC of the samples was detected after the addition (40 μl) of 0.2 mg/mL p-chloride of iodonitrotetrazolium (INT) and incubation at 37°C for 30 min. MIC was defined as the lowest concentration that inhibits visible bacterial growth, determined by the change of color from yellow to pink, if the microorganisms are viable. For the determination of MBC, 10 μL of liquid from each well that did not show color change were plated in solid medium, blood agar (7% sheep’s blood) and incubated at 37°C for 24 h. The lowest concentration that did not produce growth determined the MBC. MBC was defined as the lowest concentration needed to kill bacteria.

2. Antifungal activity

Antifungal activity was performed according to the description by Heleno et al. (2013). Aspergillus fumigatus (ATCC 204305) and Aspergillus brasiliensis (ATCC 16404) were used. The organisms were obtained from Frilabo, Porto, Portugal. The micromycetes were kept in agar malt and the cultures were stored at 4 °C and were subsequently placed in a new medium and incubated at 25°C for 72h. In order to investigate the antifungal activity, the fungal spores were washed from the surface of agar plates with sterile saline solution at 0.85% containing 0.1% of Tween 80 (v/v). The spore suspension was adjusted with sterile saline solution to a concentration of approximately 1.0x105 in a final volume of 100 μL per well. The samples were first dissolved in 5% (v/v) dimethylsulfoxide (DMSO) and 95% autoclaved distilled water to obtain a final concentration of 20 mg/mL for the stock solution. Subsequently, 100 μL of this concentration were added in the first well (96-well microplate) duplicated with 90 μL of malt extract broth (MEB). In the remaining wells, 90 μL of MEB medium was placed. Then, the samples were diluted in series to obtain the concentration ranges (10 to 0.03125 mg/mL). The determinations of the minimum inhibitory concentration (MIC) were performed by a series dilution technique using 96-well microplate. The lowest concentrations without visible growth (at binocular microscope) were defined as MIC. The fungicidal concentration (MFC) was determined by a serial subculture of 2 μL of tested compounds dissolved in medium and inoculated for 72 h, in microplates containing 100 μL of MEB per well and additional incubation of 72 h at 25 °C. The lowest concentration without visible growth was defined as MFC indicating 99.5% death of the original inoculum. The commercial fungicide ketoconazole (Frilabo, Porto, Portugal) was used as a positive control.

3. Cytotoxicity activity

The cytotoxicity potential of the extracts on human tumor cells was assessed using the Sulforhodamine B (SRB) assay, following a protocol described by Vaz (2018). This colorimetric method indirectly quantifies cell number by staining cellular proteins with SRB dye. Four human cancer cell lines were tested: Caco-2 (colorectal adenocarcinoma), MCF-7 (breast adenocarcinoma), HeLa (cervical adenocarcinoma), and AGS (gastric adenocarcinoma). Cells were cultured in RPMI-1640 medium with 5% heat-inactivated FBS at 37 °C in a 5% CO₂ humidified incubator. Each cell line was seeded in 96-well plates at a 5.0 × 10³ cells/well density and left to adhere for 24 hours. Extracts were applied at different concentrations for 48 hours, after which cells were fixed with 10% trichloroacetic acid, washed with 1% acetic acid, and stained with SRB. The bound dye was dissolved using 10 mM Tris, and absorbance was measured at 490 nm. Ellipticine served as the positive control, and the results were reported as GI₅₀ values, representing the sample concentration (μg/mL) required to inhibit 50% of net cell growth.

4. Anti-inflammatory activity

The nitrite concentration resulting from lipopolysaccharide (LPS) stimulation was assessed in the murine macrophage RAW 264.7 cell line. Cells were exposed to varying concentrations (0.125-0.4 mg/mL) of extract samples dissolved in ultrapure water. Nitric oxide (NO) production was quantified using the Griess Reagent System kit from Promega (Madison, WI, USA), following the method previously described by Silva et al. (2020). Dexamethasone (50 μM) served as a positive control. The findings were expressed as IC₅₀ values (μg/mL), representing the extract concentration required to inhibit NO production by 50%.

Evaluation of antidiabetic activity

1. Inhibition of α-amylase

The inhibition of α-amylase enzyme was evaluated following the methodology adapted described in Nogueira et al. (2024), with adaptations. The absorbance read in spectrophotometer at 540 nm, and the enzymatic kinetics calculated on a standard maltose curve and the results will be expressed as inhibition percentage.

Evaluation of lipase inhibition (antilipid activity)

The lipase inhibition evaluation was made following methodology according adapted described in Nogueira et al. (2024). The samples were prepared in different concentrations. p-nitrophenyl palmitate will be used as substrate for the enzyme. The absorbance reading in spectrophotometer at 410 nm. A curve with p-nitrophenol will also be performed. The results were expressed by percentage of inhibition.

Statistical analysis

All tests were performed in triplicate the results were expressed as mean ± standard error of the mean (SEM). Data were analyzed using one-way analysis of variance (ANOVA), followed by Tukey’s post hoc test for multiple comparisons. Statistical significance was considered at p < 0.05. Statistical programs such as GraphPadPrism and R 4.3.2 were used.

RESULTS AND DISCUSSION

Obtaining the hydroethanolic extract

The extract called E1 obtained in soxhlet showed a mass yield of 37%. The extract obtained in the shaker in 3 cycles, called E2, showed a mass yield of approximately 46%. This difference in yield may be associated with the shaking during extraction in the shaker apparatus and the difference in time and cycle. Applications of stirring techniques can induce a greater penetration of the solvent in the cellular matrix, and therefore better mass transfer.

Some studies have demonstrated the direct relationship between the influence of the solvent used and the content of secondary metabolites, as well as time and temperature are important factors in obtaining the total yield of extracts (Tiwari et al. 2011). Hydroethanolic extracts obtained by agitator-assisted methods are increasingly recognized for their efficiency in the extraction of bioactive compounds from plant materials.

Characterization of the hydroethanolic extract and evaluation of antioxidant potential

Total phenolics and flavonoids

A concentration of 20.7% and 21.3% for total phenolics and 4.5% and 5.1% of flavonoids were found in soxhlet and shaker, respectively, which showed that there was no significant difference in the amount of these extracted compounds (p > 0.05). These results are similar to those found by Chakma et al. (2023). In this study, no significant difference was reported in the extraction method of phytochemicals for testing the antioxidant and antibacterial activity of Stevia rebaudiana leaf extracts.

1. Analysis by UHPLC-MS/MS (Ultra-High Performance Liquid Chromatography)

The analysis by UHPLC-MS/MS quantified several phenolic compounds found in both extracts presented in Table I. It is possible to observe that extraction performed in soxhlet showed significantly higher concentrations of phytochemicals, even presenting a mass yield similar to the shaker. Gallic acid, caffeic acid, chlorogenic acid, apigenin, luteolin, kampferol and quercetin were found.

Table I
Quantification of phenolic compounds found in analysis UHPLC-MS/MS.

Phenolic compounds	Soxhlet extract (mg/g)	Shaker extract (mg/g)
Gallic acid	5.0 ± 0.0006	3.0 ± 0.0001
caffeic acid	1.800 ± 0.004	1.080 ± 0.001
4-hydroxybenzoic acid	ND	ND
Vanillic acid	ND	ND
Chlorogenic acid	2.610 ± 0.005	1.560 ± 0.001
Naringenin	ND	ND
Apigenin	8.0 ± 0.0001	0.4 ± 0.004
Luteolin	2.0 ± 0.001	0.2 ± 0.001
Kaempferol	7,0 ± 0,005	4.0 ± 0.002
Quercetin	10.0 ± 0.002	6.0 ± 0.001
Coumaric acid	ND	ND
3-O-Caffeoylquinic acid	2.21 ± 0.3	1.85 ± 0.1
5-O-Caffeoylquinic acid	17.74 ± 1.7	14.85 ± 0.4
Quercetin-3-O-rutinoside	0.60 ± 0.1	0.03 ± 0.005
Quercetin-O-hexoside	0.51 ± 0.002	0.50 ± 0.005
Quercetin-3-O-glucoside	0.47 ± 0.003	0.47 ± 0.003
kaempferol-O-hexoside	0.51 ± 0.003	0.50 ± .002
kaempferol-rhamnosyl-glucoside	0.72 ± 0.02	0.67 ± 0.008
Quercetin-O-pentoside	0.50 ± 0.004	0.48 ± 0.004
3,4-O-Dicaffeoylquinic acid	1.61 ± 0.02	0.98 ± 0.003
3,5-O-Dicaffeoylquinic acid	11.77 ± 1.31	20.54 ± 0.82
Quercetin-O-rhamnoside	1.39 ± 0.09	1.22 ± 0.016
4,5-O-Dicaffeoylquinic acid	15.47 ± 1.69	8.7 ± 0.05
Kaempferol-O-deoxyhexoside	0.68 ± 0.02	0.64 ± 0.003
Quercetin dihexosyl deoxyhexoside	0.49 ± 0.007	2.8 ± 3.2
Tricaffeoylquinic acid	ND	ND

• ND: Not detectable - Above the method’s limit of detection.

The soxhlet method uses continuous heating cycles which may result in increased solubility and release of phenolic compounds. In addition, the solvent is continuously renewed, promoting a more efficient extraction of poorly soluble compounds. Being a particularly effective method for exhaustive extraction, ensuring that even low concentrations of compounds are completely extracted (Torres-Rodriguez et al. 2024).

In the study conducted by Zainol et al. (2023), investigating the influence of agitation and solvent percentage on phytochemical extraction of Asystasia gangetica, it was observed that agitation can significantly influence phytochemical extraction, with ideal conditions at 204 rpm. However, the percentage of solvent was considered insignificant for total extraction of phenolic compounds, suggesting that only a stronger stirring can increase the release of strongly bound compounds.

Research conducted by Pacifico et al. (2019) reported 33 phenolic compounds of Stevia being 24 chlorogenic acids with their TOF-MS data. The chlorogenic acid, being the major phenolic in both extracts is known for its excellent antioxidant activity helping to neutralize free radicals, reducing oxidative stress (Zheleva-Dimitrova et al. 2024). In addition, its presence has been associated with several health benefits, including anti-inflammatory, antimicrobial effects, inhibition of enzymes related to glucose metabolism, which may help control diabetes (Zheleva-Dimitrova et al. 2024).

Yu et al. (2022), conducted a population intervention study on the effects of chlorogenic acid intake on various health parameters such as fasting blood sugar and weight control. Seeking to analyze the dose-effect relationship. According to the authors, the estimated dietary intake of chlorogenic acid varies from 5 to 1000 mg/day, indicating a wide variability in consumption between different populations. The study highlights that daily doses between 13.5 mg and 1200 mg/day can reduce fasting blood sugar, improve glucose tolerance, promote weight loss or prevent weight gain, as well as improve blood pressure in hypertensive patients.

In addition, it states that daily intake of more than 200 mg shows a reduction in blood glucose, with a dose-effect ratio in the range of 13.5 to 500 mg/day. Thus, a specific level proposed to improve fasting blood glucose would be 200 mg/day. However, insufficient data did not allow the determination of a maximum tolerable intake (UL) for chlorogenic acid.

Glycoside quantification and centesimal composition

Table II shows the contents of crude protein, lipids, carbohydrates, glycosides, and moisture present in the plant (stems with leaves) and extracts (E1 and E2). The chemical composition values found are similar to those reported by Shalaby et al. (2019) who conducted a study of chemical composition and nutritional evaluation of S. rebaudiana leaves. It was also observed that the plant of this study presented 14% in the content of total glycosides. The extracts showed a concentration of these compounds, being 34% and 28% in soxhlet and shaker, respectively.

Table II
Chemical composition of Stevia whole plant. Soxhlet extract. Shaker extract in g/100g.

Composition	Whole plant (g/100g)	Soxhlet extract (g/100g)	Shaker Extract (g/100g)
Crude protein	10.6 ± 0.0009	7.5 ± 0.001	9.9 ± 0.001
Total lipids	1.01 ± 0.002	10.3 ± 0.0001	18.7 ± 0.0001
Total Carbohydrates	66.51 ± 0.02	42.14 ± 0.05	37.25 ± 0.04
Glycosides	14.3 ± 0.04	34.9 ± 0.01	28.8 ± 0.01
Humidity	7.58 ± 0.002	5.16 ± 0.002	5.35 ± 0.0005

Identification of functional groups by infrared spectroscopy with Fourier transform (FTIR)

The effects of the phytochemical extraction techniques of Stevia leaf extracts obtained by shaker and soxhlet extraction were evaluated and the profile of phenolic compounds obtained by UHPLC -MS/MS and mass spectrometry was corroborated by FTIR analysis of the extracts hydroethanols shown in Figure 2.

Figure 2
FTIR spectra.

By means of the absorption spectra of both extracts it is possible to observe a high degree of similarity of the present functional groups, which are part of the phytochemical compounds extracted. The band present in 3393 _cm-1 may be associated with the stretch vibration -OH of portions of alcohols and phenols and/or stretch by bending of phenols and hydrogen-bound alcohols in the molecules, binding these characteristics of caffeic and chlorogenic acids. Peaks at 2931 and 2851 _cm-1 can be attributed to the elongation vibration of C-H (An et al. 2024), and the peaks at 1720 _cm-1 and 1599 _cm-1 are attributed to the elongation vibration of the -C-O band, which is due to the presence of steol glycosides present in hydroethanol extracts. In addition, there are also the peaks present at 1516 and 1381 cm-1 that can be attributed to C-C of the benzene ring characteristic of aromatic compounds present in both the molecules of rebaudiosides, steviosides and acids found mostly in the studied fractions (Liu et al. 2013), suggesting that phenolic acid is present in the extract. The peaks at 1168 - 1078 cm-1 correspond to the C-O elongation of aliphatic ethers present in stevioside and rebaudioside molecules. Finally, the band at 603 cm-1 indicates the existence of the C-H bond belonging to aromatic functional groups. Through the intensity of the peaks evaluated we can conclude that the profile of phytochemical constituents of the hydroethanolic extract of Stevia leaf obtained by soxhlet, which had a greater amount of phytochemical compounds found in the extract.

Based on the published reports, a total of sixteen compounds were identified in Stevia leaf extracts (Myint et al. 2021). Most of the phytochemicals in Stevia leaf extracts were phenolic acids (neochlorogenic acid, chlorogenic acid, cafeoylkinic acid and dicafeoylkinic acid) and their derivatives.

Particle size

The Dynamic Light Scattering (DLS) measures the Brownian motion of particles allowing the calculation of their hydrodynamic diameter. Hydroethanolic extracts usually contain a variety of nanoparticles; DLS can help to evaluate their size distribution, which influences their biological activity (Urquidi et al. 2023). Figure 3 shows the particle size of the extracts. In Figure 3a, we can observe a bimodal distribution with a lower peak at 10 nm to 60 nm and a higher peak at 80 nm to 750 nm. Tonon et al. (2009), describes that the bimodal distribution is a particularly interesting phenomenon in the case of powders, since the smaller particles can occupy the spaces between the larger ones, reducing the total volume occupied. The presence of larger particles is often attributed to an initial agglomeration process, in which the formation of irreversible bonding bridges results in larger particles.

Figure 3
Particle size distribution. a) Soxhlet extract. b) Shaker extract.

In Figure 3b, a more comprehensive peak is observed going from 20 nm to 700 nm. Feihrmann et al. (2024) observed a similar behavior of the microcapsules of leaf extract of Tradescantia zebrina, such dimensions are interesting for application in food. Particle sizes smaller than 100,000 nm are acceptable for use in food products as they are not perceived by the consumer. Also, studies suggest that the particular of smaller extracts (10-50 nm) demonstrate better dissolution rate, higher antioxidant activity antibacterial effects, indicating a greater bioavailability of bioactive compounds.

Evaluation of antioxidant, antimicrobial and enzymatic inhibition activity

Antioxidant capacity

The extracts showed antioxidant activity when tested for the methods: DPPH, TBARS, OxLHIA, and ORAC (Figure 4).

Figure 4
Antioxidant activity in vitro. DPPH (TEAC/g): soxhlet: 5.5g; Shaker 4.0g. TBARS: EC₅₀ soxhlet: 25.1 μg/mL; EC₅₀ Shaker: 41.2 μg/mL. OxLHIA: EC₅₀ soxhlet: 178 ± 8 µg/mL; EC₅₀ Shaker: 168 ± 9 µg/mL. ORAC (TEAC/g): soxhlet: 18.4 μmol/g; Shaker: 13.8 μmol/g. Different letters indicate statistical differences among samples, specifically were observed for DPPH, TBARS and ORAC (p ≥ 0.01), while no difference was detected for OxHLIA (p = 0.642).

There are many different antioxidants present in plants; however, it is very difficult to measure each antioxidant component separately. DPPH radicals are widely used as a model system to evaluate the elimination activities of natural antioxidant compounds (Zayova et al. 2013). The results presented in Figure 4a, expressed in equivalent of Trolox, indicate that for the radical DPPH the soxhlet extract showed greater activity, being 5.5 ± 0.01 g TE/100. On the other hand the extract shaker content of 4.0 g. These data did not corroborate the work of Chakma et al. (2023), who reported that the extracts obtained by the agitator method demonstrated higher antioxidant activity compared to those obtained by soxhlet extraction.

Reactive Oxygen Species (ROS) are oxidizing molecules generated as part of the metabolism of several cell types. When overproduced, ROS can react with lipids, resulting in the formation of lipid peroxidation products. The extent of this process can be evaluated by the Thiobarbituric Acid Reactive Substances (TBARS) assay. In Figure 4b it is possible to observe that for this test the soxhlet extract presented EC₅₀ values of 25.1 ± 0.02 μg/ml and the shaker an EC₅₀ of 41.2 ± 0.04 μg/ml. Lower TBARS values suggest effective antioxidant action. Thus, soxhlet extract indicated higher protection potential against oxidative damage.

Phenolic compounds can interact with lipid bilayers, affecting their structure and biophysical properties. This interaction is important for its biological activity, including its antioxidant effects. Such interactions are important to mitigate oxidative stress, which is related to various diseases. Phenolic compounds, due to their chemical structure, can donate electrons to free radicals, stabilizing them and preventing cell damage. This antioxidant activity is a key mechanism through which phenolic extracts exert their biological effects in the context of OxHLIA. In Figure 4c it is possible to observe that both methods of extraction demonstrate a significantly higher antioxidant capacity than the standard Trolox. In addition, for such analysis, there is no significant difference between the two extraction methods.

The ORAC (Oxygen Radical Absorption Capacity) value of Stevia extract is indicative of its antioxidant potential, which is linked to several health benefits and disease prevention mechanisms. In this study, the results were expressed in equivalent of Trolox, the extract soxhlet presented 18.4 ± 0.04 μmol/g, as the shaker it was observed 13.8 μmol/g. The results of a meta-analysis indicated that ORAC values correlate positively with the content and total phenolic profile, suggesting that higher concentrations of phenolics increase antioxidant activity (Papaefthimiou et al. 2024).

By analyzing the results of the antioxidant capacity tests, we noticed that the soxhlet extract presented the most promising results. Such behavior may be associated with the extraction efficiency of phenolic compounds, as well as the particle size (Prasedya et al. 2021).

Antimicrobial capacity

The results in Tables III and Tables IV show us that both extracts exhibited bacteriostatic capacity. For food bacteria (Table III), the soxhlet extract presented bacterial growth inhibition from the concentration of 0.6 mg/ml for bacteria of the genus Bacillus cereus, 2.5 mg/ml for Pseudomonas aeruginosa and Listeria monocytogenes, while the shaker inhibition occurred from 1.25 and 5 mg/ml, respectively. For the other bacteria tested antimicrobial activity was observed from 5 and 10 mg/ml for both extracts.

Table III
Antibacterial activity. Food bacteria. The maximum concentration tested was 10 mg/mL of the hydroethanolic extracts.

							Positive Control
	Soxhlet extract		Extract Shaker		Streptomicin 1mg/mL		Methicilin 1mg/mL		Ampicillin 10mg/mL
	MIC	MBC	MIC	MBC	MIC	MBC	MIC	MBC	MIC	MBC
Gram-negative bacteria
Enterobacter Cloacae	5	>10	5	>10	0.007	0.007	n.t.	n.t	0.15	0.15
Escherichia coli	10	>10	10	>10	0.01	0.01	n.t.	n.t.	0.15	0.15
Pseudomonas aeruginosa	2.5	>10	5	>10	0.06	0.06	n.t.	n.t.	0.63	0.63
Salmonella enterica	5	>10	5	>10	0.007	0.007	n.t.	n.t.	0.15	0.15
Yersinia enterocolitica	10	>10	10	>10	0.007	0.007	n.t.	n.t.	0.15	0.15
Gram-positive bacteria
Bacillus cereus	0.6	>10	1.25	>10	0.007	0.007	n.t.	n.t.	n.t.	n.t.
Listeria monocytogenes	2.5	>10	5	>10	0.007	0.007	n.t.	n.t.	0.15	0.15
Staphylococcus aureus	5	>10	5	>10	0.007	0.007	0.007	0.007	0.15	0.15

• *MIC: minimum inhibitory concentration; MBC: minimum bactericidal concentration.

Many bacteria are directly related to several diseases, being responsible for more than 80% of cases of Foodborne Diseases (FBD). It is estimated that these diseases affect about 600 million people annually (Pontello & Gori 2023).

Studies have confirmed that Stevia extracts can effectively inhibit food-spoilage bacteria such as Salmonella and B. cereus (Shojaee et al. 2017). Therefore, it is essential for the food industry to adopt and develop effective methodologies to prevent the proliferation of these microorganisms, ensuring food safety.

Hydroethanolic extracts are rich in polyphenols, such as flavonoids. These compounds have considerable antimicrobial activity. Flavonoids present such activity possibly due to their ability to complexate with extracellular and soluble proteins and to complexate with bacterial cell walls. It has been reported that polyphenols, including chlorogenic acids, are the main component of Stevia leaves that have antibacterial activity against typical food-borne gram-negative pathogenic bacteria (Kabir et al. 2014). The literature has shown its antibacterial efficiency against several bacteria and inhibitory effect on biofilm formation (Hochma et al. 2021). However, it is not toxic against probiotic bacteria, which makes this compound suitable for use in the food industry (Puupponen-Pimiä et al. 2001). The antimicrobial and natural antioxidant properties of Stevia extracts make them suitable alternatives to synthetic preservatives, aligning with consumer preferences for natural ingredients.

For the clinical bacteria tested (Table IV), it was observed that the shaker extract showed antibacterial activity from a concentration of 1.25 mg/ml for the genera E. faecalis and L. monocytogenes and for M. morganii from 5 mg/ml. For the gram-negative bacteria K. pneumoniae the extract did not show antimicrobial activity. The soxhlet extract showed activity from 5 mg/ml to 10 mg/ml, and maximum concentration was tested. These results are corroborated by Chakma et al. (2023) who observed that the extract obtained by shaker showed a higher inhibition zone against gram-positive bacteria than extracts obtained by the extraction technique in apparatus soxhlet.

Table IV
Antibacterial Activity. Clinical Bacteria. Maximum concentration tested was 10mg/mL of hydroethanolic extracts.

							Positive Control
	Soxhlet extract		Extract Shaker		Ampicillin (10mg/mL)		Imipenem (1mg/mL)		Vancomycin (1mg/mL)
	MIC	MBC	MIC	MBC	MIC	MBC	MIC	MBC	MIC	MBC
Gram-negative bacteria
Escherichia coli	>10	>10	>10	>10	<0.15	<0.15	<0.0078	<0.0078	n.t.	n.t.
Klebsiella pneumoniae	10	>10	>10	>10	10	>10	<0.0078	<0.0078	n.t.	n.t.
Morganella morganii	5	>10	5	>10	>10	>10	<0.0078	<0.0078	n.t.	n.t.
Proteus mirabilis	10	>10	10	>10	<015	<0.15	<0.0078	<0.0078	n.t.	n.t.
Pseudomonas aeruginosa	10	>10	10	>10	>10	>10	0.5	1	n.t.	n.t.
Gram-positive bacteria
Enterococcus faecalis	5	>10	1.25	>10	<0.15	<0.15	<0.0078	<0.0078	n.t.	n.t.
Listeria monocytogenes	5	>10	1.25	>10	<0.15	<0.15	n.t.	n.t.	0.25	0.5
MRSA	5	>10	5	>10	<0.15	<0.15	n.t.	n.t.	0.25	0.5

• MIC: minimum inhibitory concentration; MBC: minimuml bactericidal concentration.

As for antifungal activity (Table V), it was observed that at 10 mg/ml the soxhlet extract presented growth inhibitory effect in both species tested. Nonetheless, it did not show a fungicidal effect. The shaker extract showed growth inhibition of A. brasiliensis at the maximum concentration tested. For Aspergillus fumigatus, growth inhibition was observed from 5 mg/ml of the sample. No fungicidal activity was observed at any concentration of the extracts tested.

Table V
Antifungal Activity (MIC/MFC, mg/mL). (Maximum concentration tested was 10 mg/mL).

	Aspergillus brasiliensis	Aspergillus fumigatus
Soxhlet extract	10/>10	10/>10
Extract Shaker	10/>10	5/>10
Ketoconazole	0.06/0.125	0.5/1

• *MIC: minimum inhibitory concentration; MFC: minimum fungicidal concentration.

Aspergillus fumigatus and other species of the genus Aspergillus are widely distributed in various environments, which increases the likelihood of exposure and infection in humans. Patients with chronic lung diseases, cancer or immunosuppression are especially vulnerable to serious infections, which can have mortality rates between 30% and 90% (Punia et al. 2022). Therefore, it is interesting to promote research that seeks methods of controlling and preventing such genders.

Cytotoxicity and anti-inflammatory activities

In the cytotoxicity assay, the various tumor cell lines were tested at a maximum concentration of 400 µg/mL, but no significant cytotoxic activity was observed, suggesting a low cytotoxic potential for the analyzed extracts. Similarly, the highest tested concentration did not exhibit toxicity in non-tumor cells, indicating a broad safety profile.

Cytotoxic responses in non-tumor cells can differ from those in tumor cells, as they involve distinct mechanisms, including apoptosis through caspase activation and immune-mediated pathways. Futhermore, when evaluating anti-inflammatory activity at the same concentration, no significant effects were detected in immune cells, further supporting the extracts’ limited cytotoxic and anti-inflammatory impact under the tested conditions.

Inhibition of pancreatic α-amylase

Regarding the inhibition of α-amylase (Figure 5a) this study showed higher results for samples at 0.3 mg/ml, showing an inhibition of 82.4 ± 0.008% of soxhlet extract and 54 ± 0.001% of shaker extract (12.35 μmol starch). However, it is possible to observe that the soxhlet extract maintained higher percentages of inhibition at substrate concentrations at 24.7 μmol (57% inhibition) and 37.0 μmol (9.8% inhibition). The data show that as the starch concentration increases, a decrease in inhibition percentage occurs. Cabeça et al. (2024) reported that similar concentrations obtained 10% inhibition of the enzyme with its ethanolic extract of Stevia leaf.

Figure 5
a) Inhibition of pancreatic α-amylase. b) Inhibition of pancreatic lipase.

Due to its essential activity in the metabolism of carbohydrates impact on hyperglycemia, enzymatic inhibition of α-amylase is studied for the possible antidiabetic profile of natural products. Although the presented data indicate a low inhibition at higher concentrations of substrate, this does not invalidate its functionality or its potential as an enzyme inhibitor. On the contrary, it suggests a viable natural alternative, with profit and industrial viability. Since conventional and non-conventional extraction processes used industrially to obtain phenolics lead to the exclusion of sweeteners, producing bitter extracts which makes their application difficult.

When we performed the green extractions (water + ethanol) with the whole plant, we obtained an extract with considerable content of phenolic compounds, appreciable levels of glycosides, besides being a sustainable and economically viable approach. Since separating the stems from the leaves, commonly intended for disposal or animal feed, entails costly processes for industry. Thus, the integral use of the plant can represent an efficient and sustainable solution (Milani et al. 2017b).

Inhibition of pancreatic lipase

The pancreatic lipase inhibition activity was evaluated at various concentrations of the tested samples (50-500 μg/ml), substrate at 200μg/ml, along with a control curve of gallic acid inhibitor (50-500 mg/ml). For the extract soxhlet no inhibition was observed, and a better investigation is needed. The best inhibition rate was observed in the shaker extract with 90 ± 0.04% at a concentration of 100μg/ml. In contrast, at the same concentration the standard inhibitor presented 50 ± 0.02%. With a concentration of 200μg/ml, the sample showed inhibition of 87 ± 0.01% and gallic acid 18 ± 0.001%.

In the study by Nogueira et al. (2024), they found comparable results when they tested two extracts of Stevia stems the ethyl acetate fraction of the Stevia stem extracts also showed significant lipase inhibitory activity. The authors highlight that hydroalcoholic extracts of Stevia have 17% phenolic compounds, and the ethyl acetate fraction has 26% phenolic compounds, suggesting that the concentration of these compounds can directly influence the ability to inhibit lipase. Due to controversial data we suggest further investigations.

The mechanisms of inhibition of pancreatic lipase are still being studied, however, the literature suggests that extracts rich in polyphenols, including 3,5-dicaffeoquinic acid, epicatechin and isoquercetin, are associated with its inhibition if binding to key amino acid residues (HIS-263, PHE-77 and SER-152) at the active site of the enzyme (Cao et al. 2024).

CONCLUSIONS

This study investigated in a pioneering way extracts from whole plant of commercial and research interest, that is, extracts from leaves with stems. The chemical composition and functional effects of hydroethanolic extracts of Stevia UEM-13, obtained by soxhlet and shaker methods were published. Both methods were effective in the extraction of phenolic compounds and flavonoids, however, the soxhlet despite not agitating, presented a superior performance in obtaining phytochemicals such as chlorogenic acid, caffeic acid and quercetin. These compounds are associated with a high antioxidant activity, which reinforces the potential of these extracts for applications in functional foods, cosmetics and pharmaceuticals. In addition, FTIR spectroscopic analysis confirmed the presence of characteristic functional groups of these bioactive compounds. The extracts also demonstrated antimicrobial activity against various pathogenic and food-borne bacteria. The extract obtained by soxhlet stood out in antioxidant activity and inhibition of α-amylase, while the shaker extract showed better inhibition of pancreatic lipase. These findings suggest that the choice of extraction method may influence the functional profile of the extract, allowing its application directed to different applications and health benefits. In conclusion, the hydroethanolic extracts of Stevia UEM-13 have great bioactive potential. In addition, future research should evaluate the bioavailability and safety of these compounds to ensure their effectiveness and viability as a natural alternative in different applications.

Acknowledgements

This work was supported by national funds through FCT/MCTES (PIDDAC): CIMO, UIDB/00690/2020 (DOI: 10.54499/UIDB/00690/2020) and UIDP/00690/2020 (DOI: 10.54499/UIDP/00690/2020); and SusTEC, LA/P/0007/2020 (DOI: 10.54499/LA/P/0007/2020). And the UEM Research Support Centers Complex – COMCAP.

Data availability

All the data used in the manuscript are available in the tables and figures. Code Availability not applicable.

References

• ADOLFO LUTZ INSTITUTE. 2008. Physicochemical methods for food analysis. 4th ed., São Paulo: Instituto Adolfo Lutz, p. 1020.
• AN FK, LI MY, LUO HL, LIU XL, FU Z & REN MH. 2024. Structural properties and antioxidant capacity of different aminated starch-phenolic acid conjugates. Food Chem 460: 140592.
• CABEÇA CL ET AL. 2024. Stevia By-product Fraction has Antioxidant Capacity and has been Evaluated for its Antihyperglycemic and Antilipid Potential. Curr Bioact Compd 21(4). DOI: 10.2174/0115734072322861240927105151.
• CABEÇA L, NOGUEIRA NC, ROSA M, DACOME AS, MADRONA GS, MAREZE E & MILANI PG. 2023. Microencapsulated antioxidant Stevia fraction fortifies whey protein and enhances its antidiabetic activity. J Food Sci Technol 60(8): 2275-2285.
• CAO Q, MEI S, MEHMOOD A, SUN Y & CHEN XB. 2024. Inhibition of pancreatic lipase by coffee leaves derived polyphenols: A mechanistic study. Food Chem 444: 138514.
• CARINA ET AL. 2022. μ-QuEChERS method combined with UHPLC-MS/MS for the analysis of phenolic compounds in red pepper varieties. J Food Compos Anal 112: 104647.
• CHAKMA A, AFRIN F, RASUL MG, MAEDA H, CHUNMEI, YUAN A & SHAH. 2023. Effects of extraction techniques/solvents on antioxidant and antibacterial activity of Stevia rebaudiana Bertoni leaf extracts. Food Chem Adv 3: 100494.
• DACOME AS, SILVA CC, COSTA CE, FONTANA JD, ADELMANN J & COSTA SC. 2005. Sweet diterpenic glycosides balance of a new cultivar of Stevia rebaudiana (Bert.) Bertoni: Isolation and quantitative distribution by chromatographic, spectroscopic, and electrophoretic methods. Process Biochem 40(11): 3587-3594.
• DUBOIS M, GILLES KA, HAMILTON JK, REBERS PA & SMITH F. 1956. Colorimetric method for determination of sugars and related substances. Anal Chem 28(3): 350-356.
• FEIHRMANN AC, SILVA NM, MARINS AR, MATIUCCI MA, NUNES KC, NAKAMURA CV & GOMES RG. 2024. Ultrasound-assisted extraction and encapsulation by spray drying of bioactive compounds from Tradescantia zebrina leaves. Food Chem Adv 4: 100621.
• FORMIGONI M, MILANI PG, SILVA AVÍNCOLA A, SANTOS VJ, BENOSSI L, DACOME AS, PILAU EJ & COSTA SC. 2018. Pretreatment with ethanol as an alternative to improve steviol glycosides extraction and purification from a new variety of stevia. Food Chem 241: 452-459.
• HELENO AS, FERREIRA ICFR, ESTEVES AP, ĆIRIĆ A, GLAMOČLIJA J, MARTINS A, SOKOVIĆ M & QUEIROZ MJRP. 2013. Antimicrobial and demelanizing activity of Ganoderma lucidum extract, p-hydroxybenzoic and cinnamic acids and their synthetic acetylated glucuronide methyl esters. Food Chem Toxicol 58: p. 95-100. DOI: 10.1016/j.fct.2013.04.025.
  » https://doi.org/10.1016/j.fct.2013.04.025
• HOCHMA E, YARMOLINSKY L, KHALFIN B, NISNEVITCH M, BEN-SHABAT S & NAKONECHNY F. 2021. Antimicrobial Effect of Phytochemicals from Edible Plants. Processes 9(11): 2089.
• KABIR F, KATAYAMA S, TANJI N & NAKAMURA S. 2014. Antimicrobial effects of chlorogenic acid and related compounds. J Korean Soc Appl Biol Chem 57: 359-365.
• LIU J, LU J, KAN J, TANG Y & JIN C. 2013. Preparation, characterization and antioxidant activity of phenolic acids grafted carboxymethyl chitosan. Int J Biol Macromol 62: 85-93.
• MILANI PG ET AL. 2017b. Fortification of the whey protein isolate antioxidant and antidiabetic activity with fraction rich in phenolic compounds obtained from Stevia rebaudiana (Bert.) Bertoni leaves. J Food Sci Technol 54(7): 2020-2029.
• MILANI PG, FORMIGONI M, DACOME AS, BENOSSI L, DA COSTA CEM & DA COSTA SC. 2017a. New seminal variety of Stevia rebaudiana: Obtaining fractions with high antioxidant potential of leaves. An Acad Bras Cienc 89: 1841-1850. DOI: 10.1590/0001-3765201720170174.
  » https://doi.org/10.1590/0001-3765201720170174
• MYINT KZ, ZAR Z, ZHOU Z, XIA Y, YUN F, WU M, ZHU S & SHEN J. 2021. Stevia polyphenols: a stable antioxidant that presents a synergistic effect with vitamin C. J Food Process Preserv 45(4): e15317.
• NOGUEIRA NC, CABEÇA CLS, SICILIANO PLM, PEREIRA BC, ZORZENON MRT, DACOME AS & FERNANDES PGM. 2024. Methanolic and hydroalcoholic extract of Stevia stems have antihyperglycemic and antilipid activity. Food Biosci 58: 103690.
• PACIFICO S, PICCOLELLA S, NOCERA P, TRANQUILLO E, DAL POGGETTO F & CATAURO M. 2019. New insights into phenol and polyphenol composition of Stevia rebaudiana leaves. J Pharm Biomed Anal 163: 45-57.
• PAPAEFTHIMIOU M, KONTOU PI, BAGOS PG & BRALIOU GG. 2024. Integration of Antioxidant Activity Assays Data of Stevia Leaf Extracts: A Systematic Review and Meta-Analysis. Antioxidants 13(6): 692.
• PIRES TCSP, DIAS MI, BARROS L, ALVES MJ, OLIVEIRA MBPP, SANTOS-BUELGA C & FERREIRA ICFR. 2018. Antioxidant and antimicrobial properties of dried Portuguese apple variety (Malus domestica Borkh. cv Bravo de Esmolfe). Food Chem 240: 701-706.
• PONTELLO M & GORI M. 2023. Foodborne diseases. In: Global Health Essentials. Cham: Springer International Publishing, p. 149-154.
• PRASEDYA ES, FREDIANSYAH A, MARTYASARI NWR, ILHAMI BTK, ABIDIN AS, PADMI H, JUANSSILFERO AB, WIDYASTUTI S & SUNARWIDHI AL. 2021. Effect of particle size on phytochemical composition and antioxidant properties of Sargassum cristaefolium ethanol extract. Sci Rep 11(1): 1-9.
• PUNIA A, AWANA CHOUDHARY P, SHARMA N, DAHIYA S, GULIA P & CHHILLAR AK. 2022. Therapeutic approaches for combating Aspergillus-associated infection. Curr Drug Targets 23(16): 1465-1488.
• PUUPPONEN-PIMIÄ R, NOHYNEK L, MEIER C, KÄHKÖNEN M, HEINONEN M, HOPIA A & OKSMAN-CALDENTEY KM. 2001. Antimicrobial properties of phenolic compounds from berries. J Appl Microbiol 90: 494-507.
• RAHMANI Z, KARIMI M, SAFFARI I, MIRZAEI H, NEJATI M & SHARAFATI CHALESHTORI R. 2024. Nanoemulsion and nanoencapsulation of a hydroethanolic extract of Nettle (Urtica dioica) and Wormwood (Artemisia absinthium): comparison of antibacterial and anticancer activity. Front Chem 12: 1266573.
• SHALABY RA, BADR SE, SAKR DM, ABDEL-RAHMAN A, EL-REFFAEI W & ABDEL-SALAM A. 2019. Nutritional evaluation on Stevia rebaudiana leaves. Food Nutr J 4: 206.
• SHOJAEE MF, MAHOONAK ARS, KHOMEIRI M & GHORBANI M. 2017. Antibacterial Effects of Stevia rebaudiana Bertoni Extract on Pathogenic Bacteria and Food Spoilage. J Food Biosci Technol 7(1): 57-64.
• SILVA AR, PINELA J, DIAS MI, CALHELHA RC, ALVES MJ, MOCAN A, GARCÍA PA, BARROS L & FERREIRA ICFR. 2020. Exploring the phytochemical profile of Cytinus hypocistis (L.) L. as a source of health-promoting biomolecules behind its in vitro bioactive and enzyme inhibitory properties. Food Chem Toxicol 136: 111071.
• TIWARI P, KUMAR B, KAUR M & KAUR G. 2011. Phytochemical screening and Extraction: a review. Int Pharm Sci 1: 98-106.
• TONON RV, BRABET C, PALLET D, BRAT P & HUBINGER MD. 2009. Physicochemical and morphological characterisation of açaí (Euterpe oleracea Mart.) powder produced with different carrier agents. Int J Food Sci Technol 44(10): 1950-1958.
• TORRES-RODRIGUEZ A, DARVISHZADEH R, SKIDMORE AK, FRÄNZEL-LUITEN E, KNAKEN B & SCHUUR B. 2024. High-throughput Soxhlet extraction method applied for analysis of leaf lignocellulose and non-structural substances. MethodsX 12: 102644.
• URQUIDI O, BARBOSA N, BRAZARD J & ADACHI TBM. 2023. Toward time resolved dynamic light scattering microscopy: Retrieving particle size distributions at high temporal resolutions. Rev Sci Instrum 94(8): 083101.
• VAZ IMM. 2018. Avaliação in vitro do efeito de antioxidantes e inibidores da apoptose/autofagia na nefrotoxicidade induzida por catinonas sintéticas. Master’s Thesis. Universidade Fernando Pessoa, Portugal. (Unpblished).
• YU H, YANG G, SATO M, YAMAGUCHI T, NAKANO T & XI Y. 2017. Antioxidant activities of aqueous extract from Stevia rebaudiana stem waste to inhibit fish oil oxidation and identification of its phenolic compounds. Food Chem 232: 379-386.
• YU Y, ZHANG Z & CHANG C. 2022. Chlorogenic acid intake guidance: Sources, health benefits, and safety. Asia Pac J Clin Nutr 31(4): 602-610.
• ZAINOL N, AZIZ NH & BAHARUDIN AS. 2023. Influence of agitation and solvent percentage on the extraction of phytochemical compound from Asystasia gangetica. Food Chem Adv 3: 100538.
• ZAYOVA E, STANCHEVA I, GENEVA M, PETROVA M & DIMITROVA L. 2013. Antioxidant activity of in vitro propagated Stevia rebaudiana Bertoni plants of different origins. Turk J Biol 37: 106-113.
• ZHELEVA-DIMITROVA D, ZENGIN G, BOUYAHYA A, AHMED S, PEREIRA CG, SHARIFI-RAD M & CUSTODIO L. 2024. Deciphering the chemical constituents of Phlomis monocephala extracts using UHPLC-HRMS and their antioxidant, neuroprotective, antidiabetic and toxic potentials. Food Biosci 59: 104183.

Correspondence

Correspondence to: Paulo Leonardo Marotti Siciliano [e-mail]

Handling editor

Gisele da Rocha

Publication Dates

• Publication in this collection
  30 Oct 2026
• Date of issue
  2026

History

• Received
  26 May 2025
• Accepted
  29 Apr 2026