Antifungal and antibiofilm activity of crude hydroalcoholic extract and fractions of Myrcia neoobscura leaves against a clinical isolate of Candida krusei

Infections caused by biofilm-forming Candida sp. Are responsible for high morbidity and mortality rates. With the scarcity of therapeutic options and the increase in resistance to antifungal drugs, new treatment options are needed. Myrcia neoobscura is a tree endemic to the Atlantic Rainforest, belonging to the genus Myrcia, several of whose species have demonstrated important antifungal activities. The aim of this study was to evaluate the antifungal and antibiofilm activity of the crude hydroalcoholic extract and fractions of Myrcia neoobscura, as well as the mechanisms of action using in vitro methods. Strong inhibitory results were observed against one clinical isolate of Candida krusei with the aqueous fraction standing out (MIC of 7.81 μg/mL), followed by crude hydroalcoholic extract, ethyl acetate, insoluble fractions (15.62 μg/mL), hexane and dichloromethane fractions (62.50 μg/mL). Regarding the mechanism of action, the increase of MIC values on both Ergosterol and Sorbitol assays suggest that extract and fractions of M. neoobscura acts on fungal membrane and cell wall. Candida krusei showed moderate biofilm formation and significant antibiofilm activity were observed for all tested samples against the mature biofilm, with the dissolution of 48h mature C. krusei biofilm ranging from 45.6% up to 60.2%. These findings highlight M. neoobscura as a promising source of antifungal and antibiofilm agents, particularly against C. krusei.

Candida spp. comprises a genus of yeast-like fungi that are commensals of humans and animals and are the etiological agents of candidiasis, a condition responsible for most fungal-related nosocomial infections (Botelho et al., 2022). Although C. albicans is the main species causing candidiasis, the number and variety of non-albicans Candida (NAC) species are continuously emerging, affecting around 1.6 billion individuals. It represents a growing concern in healthcare settings, especially among high-risk immunocompromised patients. NAC species often display reduced or intrinsic susceptibility to azoles, contributing to higher rates of therapeutic failure. The formation of biofilms by Candida species is well known, especially by Candida albicans and Candida krusei. Biofilms are an important virulence factor and play an important role in antifungal resistance and protection against host defenses (Atiencia-Carrera et al., 2022; Kumar et al., 2022; Botelho et al., 2022; Kaur and Nobile, 2023; Li et al., 2026). The mortality rate associated with biofilm-forming Candida infections increases almost two-fold when compared to planktonic infections (Atiencia-Carrera et al., 2022). The lack of therapeutic options, the increase of incidence and the high mortality rate associated with these pathogens make it necessary to find new treatment alternatives (Botelho et al., 2022; Li et al., 2026).

One of the several obstacles to the development of new antifungal drugs is the similarity between human and fungal cells; as both are eukaryotic cells, drugs that inhibit the synthesis of proteins, DNA or RNA can have the same effect on host cells, potentially causing side effects. The fungal cell membrane is the target for several antifungal drugs, such as azoles and polyenes. The presence of ergosterol, a sterol found in the fungal membrane, which is some similar to sterols from mammal cells, these drugs can bind to cholesterol and damage human cells. Therefore, drugs that act on another specific fungal structures, such as the cell wall, a structure found only in fungal cells and not in human cells, are generally less harmful (Liu et al., 2020; Curto et al., 2021). Plant extracts have demonstrated important antifungal activities, where the complex mixture of bioactive compounds can act on different targets in the fungal cell, being a vital part of their therapeutic efficacy (Vaou et al., 2021; Vaou et al., 2022). Plant derivatives are also an alternative to the use of conventional antifungals as they demonstrate fewer side effects associated with the use of synthetic compounds (Sharma and Kumar, 2020).

The Myrcia genus is one of the largest in the Myrtaceae family and is present in all regions of Brazil (Lingner et al., 2015). Some Myrcia species have been used in folk medicine for therapeutic purposes, such as a group of species called ‘pedra-hume-caá’ or ‘vegetal insulin’, used mainly for the treatment of diabetes (Cascaes et al., 2015). Studies on Myrcia species indicate that they have antifungal activity against a variety of fungal pathogens, from yeast-like to filamentous fungi (Silva et al., 2017; Pontes et al., 2018; Ferreira et al., 2021). Myrcia plants and its isolated compounds have demonstrated strong antifungal activity against C. albicans and Candida parapsilosis, such as the crude ethanolic extract, fractions and flavonoids avicularin, juglanin and guaijaverin isolated from Myrcia tomentosa (Da Silva Sa et al., 2017); the methanolic, hexane, ethyl acetate and acetone extracts of Myrcia oblongata against C. albicans (Santana et al., 2021) and the methanolic extract of Myrcia rufipila against strains of C. albicans resistant to amphotericin B and C. parapsilosis resistant to nystatin (Cascaes et al., 2021).

Myrcia neoobscura is a tree endemic to the Atlantic Rainforest, present in the South, Southeast and Northeast regions of Brazil (Lucas et al., 2016), and there have been few published studies on its chemical compounds and biological activities. Previous studies conducted by our research group revealed that the hydroalcoholic extract of Myrcia neoobscura exhibited high concentrations of total phenolic compounds, particularly flavanols and flavonoids. Ten phenolic compounds were identified, with gallic acid being the major constituent, followed by chlorogenic acid and p -coumaric acid. Other identified compounds included catechin, isoorientin, taxifolin, resveratrol, naringenin, and eriodictyol (Krepsky et al., 2025).

Phenolic compounds have been described in literature as having important antifungal activity, including against different Candida species (Özçelik et al., 2011; Alves et al., 2014; Martins et al., 2015; Singulani et al., 2017; Li et al., 2017; Carvalho et al., 2018; Rhimi et al., 2020; Carevic et al., 2022). Therefore, this study aimed to evaluate the antifungal and antibiofilm activity, due phenolic chemical composition found on M. neoobscura. To address this subject, it was used the crude hydroalcoholic extract (CHE) of M. neoobscura and its fractions, as well as investigate the mechanisms of action involved in the antifungal activity. Also, the lack of studies evaluating the antifungal activity of this species motivated the present work, making it the first to assess the antifungal and antibiofilm potential of M. neoobscura against Candida spp. isolates.

Plant material

The leaves of M. neoobscura were collected in May 2022, in Blumenau-SC, in Parque Natural Municipal São Francisco de Assis (26′55′22‘S; 49'4'39’W). The plant material was identified by Prof Dr André Luis de Gasper, from the Department of Natural Sciences at FURB. An exsiccate was deposited in the Dr Roberto Miguel Klein Herbarium at FURB under the number 72435. After collection, the leaves were separated from the branches and dried in an oven with air circulation at a temperature of approximately 40 °C. The dried material was ground in a knife mill and extracted by maceration in a 70% hydroalcoholic solution for 7 days, after which the maceration procedure was carried out once more. The extract was filtered and concentrated in a rotary evaporator under reduced pressure at a temperature of less than 60 °C until the crude extract was obtained. Once concentrated, the crude extract was resuspended in a 20% hydroalcoholic solution and filtered. The resulting precipitate was separated from the sample and called the insoluble fraction (IF). With the supernatant, liquid-liquid partitioning was carried out in solvents of increasing polarity (Hexane > dichloromethane > ethyl acetate), giving rise to the Hexane (HF), dichloromethane (DF) and ethyl acetate (EAF) fractions and, finally, the aqueous fraction (AF) (Hostettmann et al., 2008). The CHE and its fractions used in this study were obtained from the same batch of plant material previously phytochemically characterized by Krepsky et al. (2025).

Clinical isolates

Four clinical isolates of the genus Candida that showed resistance or decreased susceptibility to some antifungal drugs were provided by the Santa Luzia laboratory in Florianópolis/SC. The species selected were C. albicans (n = 1), C. glabrata (synonym Nakaseomyces glabrata) (n = 1), C. tropicalis (n = 1) and C. krusei (synonym Pichia kudriavzevii) (n = 1). The isolates of C. albicans and Candida tropicalis are resistant to fluconazole and itraconazole and intermediate to ketoconazole. The Candida glabrata isolate is intermediate resistant to ketoconazole and itraconazole. The C. krusei isolate is resistant to fluconazole and intermediate resistant to ketoconazole and itraconazole (Botelho et al., 2022). The isolates were previously identified using automated mass spectrometry with Matrix Assisted Laser Desorption Ionisation Time-of-Flight (MALDI-TOF) technology, VITEK MS (bioMérieux) and the software (Myla™) with bioMérieux analysis database. These isolates were stored at 2–8 °C and cultured on Sabouraud dextrose agar (SDA; Himedia).

Antifungal evaluation test of crude hidroalcoolic extract (CHE) and fractions

The antifungal Amphotericin B was used as a positive control for the experiments. The antifungal was prepared in dimethylsulphoxide (DMSO), as established in document M27-A4 (Clinical and Laboratory Standards Institute, 2017). The working solutions were diluted at the time of use in the test medium to obtain a maximum concentration of 2% DMSO in the experiments.

The tests followed the methodology described by CLSI - Protocol M27-A4 (Clinical and Laboratory Standards Institute, 2017). A standard ATCC strain of C. albicans 14,053 (American Type Culture Collection) was included in the test. The fungus was cultivated in potato dextrose agar (BDA) and then colonies were transferred to saline solution and suspended to obtain the inoculum. The cell concentration was adjusted by comparison with the 0.5 McFarland Ladder using spectrophotometry at 530 nm to achieve a cell concentration of between 1 × 10 6 and 5 × 10 6 CFU/mL.

The culture medium used to carry out the microdilution was RPMI 1640 (RPMI) (Gibco ™) buffered with MOPS (0.165 mol/L). In 96-well microtiter plates (Costar - Corning Incorporated), 90 µL of the culture medium was added to the wells of the first column and 50 µL to the others. Then, also in the wells of the first column, 10 µL of crude extract and fractions were added and serial dilution was carried out, passing 50 µL from the first well to the second and so on.

Afterwards, 50 µL of the fungal inoculum diluted in RPMI-1640 culture medium in a ratio of 1:50 (e.g., 4.9 mL of RPMI 1640 + 100 µL of inoculum) was added to all the wells. The medium and fungal inoculum were also used as a positive control, while the culture medium alone (RPMI) was used as a negative control. DMSO was used as vehicle control. After incubation for 24 h at 37 °C, the plates were read visually, and the minimum inhibitory concentration (MIC) was determined. The MIC is the lowest concentration of the test material that inhibits fungal growth, verified by the absence of turbidity. The results were interpreted considering the MIC values, where results of 100 μg/mL or less were considered strong, from 100 to 500 μg/mL moderate, from 500 to 1000 μg/mL weak and above 1000 μg/mL inactive (Santos et al., 2008).

Sorbitol protection assay

The effect on fungal cell wall integrity was evaluated by the sorbitol protection assay using the C. krusei isolate, which showed sensitivity to CHE and M. neoobscura fractions. The MICs of CHE and fractions were determined by the broth microdilution technique (CLSI M27-A4) in the absence and presence of 0.8 M sorbitol (Sigma-Aldrich) added to the assay culture medium as an osmoprotectant. The experiments were carried out in duplicate. Caspofungin (Sigma-Aldrich) was used as a positive control, since this antifungal has an action on the cell wall. DMSO was used as vehicle control. The MICs were evaluated after 2 and 7 days of incubation at 35 °C (Escalante et al., 2008). Antifungal agents that inhibit the fungal cell wall are identified when the MIC values obtained in the presence of sorbitol are higher than in its absence after 7 days of incubation (Junior et al., 2009).

Ergosterol binding assay

The ability of CHE and fractions of M. neoobscura to complex with ergosterol present in the fungal membrane was evaluated by the ergosterol binding assay using the C. krusei isolate, which showed sensitivity to CHE and fractions. The MICs of CHE and fractions were determined using the broth microdilution technique (CLSI M27-A4) in the absence and presence of 250 μg/mL of external ergosterol (Sigma-Aldrich) added to the test medium. The commercial ergosterol was prepared at the time of the test, where it was first pulverised using a sterilised grater and pistil, dissolved in DMSO (Sigma-Aldrich) for no more than 10% of the final volume and 1% tween 80. The emulsion was homogenized and diluted in RPMI 1640 culture medium so that the final solvent concentration was 0.1%. The experiments were carried out in duplicate. Amphotericin B (União Química, São Paulo, Brazil) was used as a positive control, since this antifungal has an affinity for membrane ergosterol and acts by the mechanism mentioned above. DMSO and 1% Tween 80 were used as vehicle controls. The MICs were evaluated after 2 days of incubation at 35 °C (Escalante et al., 2008). Antifungal agents that complex with membrane ergosterol are identified when the MIC values obtained in the presence of exogenous ergosterol are higher than in its absence (Junior et al., 2009).

Biofilm formation capacity

The C. krusei isolate was chosen for the biofilm formation test as it was the only one susceptible to CHE and M. neoobscura fractions.

The biofilm formation capacity was assessed using the methodology described by Cedeño-Pinargote et al. (2025), with minor modifications. The C. krusei isolate was grown on Sabouraud dextrose agar at 30 °C for 24 h, then repitched on 2% Sabouraud dextrose broth (Himedia) and incubated at 35 °C for 18 h; the cultures were then centrifuged at 5,000 rpm for 5 minutes at room temperature. The pellet formed from each culture was washed twice in 5 mL of PBS (phosphate-buffered saline solution, pH 7.2, free of Ca2+ and Mg2+), under the same centrifugation conditions, then resuspended in 5 mL of PBS and the cell concentration was adjusted using a spectrophotometer at 530 nm to reach an optical density of 0.38 absorbance, corresponding to approximately 10 7 cells/mL. The biofilm was produced in a sterile 96-well polystyrene flat-bottomed microtiter plate. For the fixation phase, 100 μL of the cell suspension was transferred to each well. The solution without cells was used as a negative control. The plate was incubated at 37 °C for 90 min, under orbital agitation at 75 rpm, so that the cells could adhere to the surfaces of the wells. Subsequently, the non-adhered cells were gently removed using a pipette, and then 100 μL of fresh Sabouraud broth was added. The plate was incubated at 37 °C for 72 h, shaking at 75 rpm to allow the biofilm growth.

After the incubation period for biofilm formation, the biofilm was quantified. The culture medium was removed from the microtiter plate, washed once with 150 µL sterile PBS, and the wells were left to dry at room temperature. Then 110 µL of 0.4% aqueous crystal violet solution was added to each well for 45 min. Afterwards, the crystal violet was removed, and the wells were washed three times with 200 µL of distilled water and then decolorized with 200 µL of absolute ethanol for 45 min. After this procedure, 100 µL from each well was transferred to a new microtiter plate of the same type and measured at 560 nm in plate readers.

The cut-off point for considering biofilm formation was established by the average of the absorbance values based on the optical density (OD) of the Candida isolate, based on that of the negative control. Considering the possible results, the biofilm formation can be classified as follows (Cedeño-Pinargote et al., 2025).

Non-producer: OD strain < OD control.

Weak producer: OD control < OD strain ≤ (2 x OD control).

Moderate producer: (2 x OD control) < OD strain ≤ (4 x OD control).

Strong producer: (4 x OD control) < OD strain.

NOTE: OD ctrl = Average optical density of the negative control + (3x standard deviation of the average).

Effect of CHE and M. neoobscura fractions on mature biofilm

The C. krusei isolate was chosen for the biofilm dissolution test as it was the only one susceptible to CHE and M. neoobscura fractions.

Biofilms were formed as described in the previous section. After 24 h of biofilm growth, the wells of the microtitre plate were washed twice with PBS solution and the plate was inverted on absorbent paper to remove residual amounts of PBS. The effect of M. neoobscura CHE and fractions was evaluated using the methodology described by Cedeño-Pinargote et al. (2025), with minor modifications. Each sample of M. neoobscura was subjected to a series of double dilutions in 2% Sabouraud dextrose broth, obtaining 10 dilutions, with decreasing concentrations. 200 μL of each of the concentrations of the challenge components were added to the respective wells of the microtitre plate and then the microtitre plate was incubated for 48 h at 35 °C, under orbital shaking at 75 rpm. The wells containing biofilm, but without the challenge components, were used as positive controls for the isolate tested. Amphotericin B (União Química, São Paulo, Brazil) was used as a positive control. There was no vehicle control included in the test.

After the incubation period, the medium was removed and the biofilms were washed twice with PBS solution. Biofilm activity was then measured using the crystal violet solution assay (described above). Biofilm formation was considered inhibited when the absorbance concentration of the crystal violet solution was equal to or lower than the control film incubated with PBS solution during the film growth phase. The tests were carried out in duplicate.

Statistical analysis

Statistical analyses were performed using the R programming language (R Core Team, 2026). To evaluate the differences between crude extract and its fractions, and among positive and negative controls, it was employed a Bayesian modeling framework. The model was fitted using the brms package (Bürkner, 2017), which implements Markov Chain Monte Carlo (MCMC) sampling. The model was specified using a log-normal distribution family. The response variable (Absorbance) was modeled as a function of the experimental treatments (predictive variable). To ensure the validity and robustness of the Bayesian inferences, comprehensive model diagnostics were performed. Convergence of the Markov Chain Monte Carlo (MCMC) chains was assessed using the Gelman-Rubin diagnostic (values of R ^ ≤ 1.01) for all estimated parameters confirmed that the chains successfully converged to the same target posterior distribution. Furthermore, Bulk and Tail Effective Sample Sizes (ESS) were evaluated to guarantee adequate independent sampling and precision of the posterior estimates. Finally, model adequacy was verified via Posterior Predictive Checks (PPCs), which demonstrated that the simulated data generated from the posterior predictive distribution accurately replicated the observed empirical distribution of the absorbance data, showing no discrepancies.

Following model fitting, Bayesian Estimated Marginal Means (EMMs) were extracted to perform pairwise comparisons among the experimental groups and controls using the emmeans package (Lenth, 2024). Statistical credibility (significance) for differences between treatments was determined using 95% credibility intervals. Pairwise contrasts, adjusted by Bonferroni, were considered statistically significant if their corresponding 95% credibility intervals strictly excluded zero.

Antifungal activity

Four clinical isolates of Candida spp. were used to evaluate the antifungal action of M. neoobscura. These species are clinical isolates that have shown resistance or decreased susceptibility to the main antifungal drugs used in clinical practice, such as azoles, as described previously in our Material and Methods section (item 2.2). The CHE and M. neoobscura fractions showed strong inhibition against C. krusei isolate, with AF standing out, showing a Minimum Inhibitory Concentration (MIC) of 7.81 μg/mL, followed by CHE and EAF and IF (15.62 μg/mL) and HF and DF (62.50 μg/mL) (Table 1). No antifungal activity was found for CHE and fractions of M. neoobscura against isolates of C. albicans, C. glabrata and C. tropicalis (>1000 μg/mL).

Mechanism of action

In both analyses, using the C. krusei isolate, the MIC of all the samples increased up to 128-fold (from 7.81 to 1000 μg/mL) in the presence of ergosterol or sorbitol, compared to their absence, suggesting that the antifungal activity occurs through concomitant action on the membrane and the fungal cell wall (Table 2).

Biofilm formation and inhibition

The C. krusei isolate was chosen for the biofilm formation test because it was the only one susceptible to CHE and M. neoobscura fractions. After the assays, C. krusei was classified as a moderate biofilm producer with the average optical density of the formation of 0.636 (Figure 1). CHE and all M. neoobscura fractions were able to dissolve the mature C. krusei biofilm by up to 60.2% (Table 3) in all concentrations tested, when compared to the negative control (Figure 1).

The CHE and all evaluated fractions exhibited a significant inhibitory effect on C. krusei growth when compared to the negative control (Figure 1). Statistical significance was confirmed as the 95% credibility intervals (CI) for all pairwise contrasts strictly excluded zero (Supplementary Table S2). For instance, the HF MICx2 fraction demonstrated substantial inhibition (Estimate = −0.692 [95% CI: −0.808, −0.575]), followed closely by the DF MIC fraction (Estimate = −0.670 [95% CI: −0.794, −0.551]). In comparison with positive control, CHE and all evaluated fractions exhibited higher C. krusei biofilm growth, demonstrating lower antifungal performance in relation to Amphotericin B (Figure 1; Supplementary Table S2). The less effective extract, with higher biofilm growth was CHE MICx2 in comparison AB MIC (Estimate = 1.305 [95% CI: 1.161, 1.446]) and AB MICx2 (Estimate = 1.292 [95% CI: 1.157, 1.441]). Pairwise comparisons among the extract and fractions revealed distinct clusters of inhibitory performance (Supplementary Table S2). The fractions HF and DF, both for MIC and MICx2, emerged as the most potent group, with no credible statistical differences observed among them. However, HF, DF and AF demonstrated significantly higher inhibitory activity when compared to CHE, EAF, and IF (Figure 1; Supplementary Table S2).

Although C. albicans remains the main cause of candidiasis, in recent years there has been an increase in infections by non-albicans Candida (NAC), including C. krusei. Candida krusei is a dimorphic yeast, with potential to cause skin, mucous membrane and systemic infections, that has intrinsic resistance to fluconazole, as well as rapidly developing acquired resistance to other antifungal drugs (Jamiu et al., 2021; Li et al., 2026). The susceptibility of the C. krusei isolate used in this study to antifungals was assessed previously and was characterized as intermediate to ketoconazole and itraconazole (Botelho et al., 2022), azole antifungals, that work by inhibiting the synthesis of ergosterol, a vital component of the fungal cell membrane (Li et al., 2017).

The phytochemical characterization of the crude hydroalcoholic extract and fractions of M. neoobscura leaves used in this study was previously carried out and published by our research group, where the quantities of total phenolic compounds, flavonoids, and flavanols were assessed (Krepsky et al., 2025). The AF contained the highest amounts of flavanols (302.9 ± 24.8 mg of catechin/g) and phenolic compounds (297.0 ± 11.0 mg of gallic acid/g), followed by the EAF fraction, which contained the second highest amount of phenolic compounds (233.5 ± 18.46 mg of gallic acid g -1) and the highest amount of flavonoids (39.68 ± 9.61 mg quercetin g -1). Concurrently, the phenolic compounds in the CHE were identified, with 10 compounds being identified, of which gallic, chlorogenic and p-coumaric acids were the most abundant. Other phenolic compounds found in smaller quantities include protocatechuic acid, catechin, isoorientin, taxifolin, resveratrol, naringenin and eriodictyol (Krepsky et al., 2025). Interestingly, the pronounced antifungal activity observed for the AF and EAF against C. krusei may be associated with their phytochemical composition. The AF exhibited the highest levels of total phenolics and flavanols, while the EAF concentrated high levels of phenolic compounds and the highest flavonoid content, reinforcing the potential role of phenolic metabolites in the antifungal effect.

Phenolic compounds represent a broad class of secondary metabolites found in plants and are closely associated with defense against pathogens, inhibiting their growth and development (Daayf et al., 2012). Phenolic acids have shown important antifungal properties, but these properties can vary according to the quantity and variety of its structure. In general, compounds with more hydroxyl groups, such as gallic acid, tend to show stronger antimicrobial activity, due to their ability to interact with and disrupt microbial cell membranes. Hydroxycinnamic acids, such as p -coumaric acid, may demonstrate advantages in certain contexts due to their unsaturated side chain, which can enhance interactions with microbial targets. However, they are not universally more potent than hydroxybenzoic acids, whose activity can also be significant depending on the microorganism and environmental conditions (Teodoro et al., 2015; Jaiswal and Kumar, 2022; Oliveira et al., 2025).

Although several mechanisms underlying the antifungal activity of phenolic compounds and flavonoids have already been described, others remain to be fully elucidated due to methodological variability and the complexity of interpreting results. Current evidence suggests that phenolic compounds may act through inhibition of ergosterol biosynthesis, penetration of the cell membrane and alteration of its hydrophobicity, induction of apoptosis, and production of reactive oxygen species (Fakhim et al., 2025). Likewise, flavonoids have been associated with multiple antifungal mechanisms, including inhibition of cell wall formation, disruption of the cell membrane, inhibition of cell division and efflux pump activity, suppression of protein, DNA and RNA synthesis, and induction of mitochondrial dysfunction (Jin, 2019). Supporting this evidence, flavonoids such as baicalein, myricetin, quercetin, and kaempferol have demonstrated antifungal activity against clinical Candida strains, particularly C. glabrata (Salazar-Aranda et al., 2015).

The predominance of gallic acid (GA) in the CHE of M. neoobscura raises the possibility that this compound contributes to the antifungal effects observed, potentially through mechanisms previously described in the literature. GA is a naturally occurring polyphenolic compound widely distributed in plants, exhibits diverse biological activities, including antimicrobial effects, and has previously demonstrated antifungal activity against Candida species. The MIC of GA against C. glabrata, C. albicans, and Candida troplicalis was reported to be between 12 and 100 μg/mL (Li et al., 2017), against C. albicans, C. krusei, C. tropicalis, C. glabrata, Myrcia canis and Trichophyton sp., ranged between 0.25 and 1 mg/mL (Carvalho et al., 2018) and against Candida auris ranging from 1.6 to 3.2 mg/mL (Akhtar et al., 2023). Other phenolic compounds found in M. neoobscura leaves also have been reported as having antifungal activities against Candida species, such as chlorogenic acid (C. albicans, C. parapsilosis, C. tropicalis, C. krusei), p -coumaric acid (C. albicans, C. parapsilosis, C. tropicalis, C. glabrata, C. krusei, Candida lusitaniae), catechin (C. albicans, C. glabrata, C. parapsilosis), naringenin (C. albicans, C. glabrata), resveratrol (C. albicans), and taxifolin (C. albicans, C. krusei, C. glabrata, C. tropicalis, C. parapsilosis) (Özçelik, Kartal, Orhan, 2011; Alves et al., 2014; Martins et al., 2015; Singulani et al., 2017; Li et al., 2017; Carvalho et al., 2018; Rhimi et al., 2020; Carevic et al., 2022).

All M. neoobscura samples exhibited increased MIC values in the presence of exogenous ergosterol, suggesting interference with antifungal activity. This effect may indicate interaction with fungal membrane sterols, since ergosterol added to the medium can sequester bioactive compounds, reducing their interaction with membrane-associated ergosterol and consequently increasing MIC values (Junior et al., 2009). Ergosterol is the specific and most abundant sterol in fungal cells and is essential for cell growth and viability. It is considered the main target of antifungal drugs such as azoles, polyenes and allylamines (Li et al., 2017).

Antifungal agents can act in two ways when they come into contact with ergosterol: (i) by binding to the ergosterol in the membrane and forming pores in this structure, thereby altering its permeability, or (ii) by inhibiting enzymes involved in ergosterol synthesis (Carvalho et al., 2018). Therefore, the M. neoobscura activity on C. krusei cellular membrane can be associated with one of these pathways, and the differentiation is important in future research to better understand the mode of action. However, an obstacle to the production of new antifungals is the similarity between human and fungal cellular membrane, as they are both eukaryotic cells, leading to higher toxicity (Curto et al., 2021).

The fungal cell wall is an essential component for cell viability, as it gives the fungal cell its morphology, provides the turgor to resist to osmotic pressure and the elasticity required to allow modification and expansion of the cell surface at sites of budding. It also has a role in the initiation of biofilm formation in some fungal species due to its hydrophobic nature and the production of adhesins, which function to anchor and invaginate fungal cells into the host’s epithelial or endothelial surfaces (Gow and Lenardon, 2023). Candida krusei cell wall contains the same polysaccharides found in the C. albicans wall: chitin, β-glucan, and mannans. Although both species have similar levels of cell wall β-glucan, C. krusei have a higher chitin content, and less mannan content (Gómez-Gaviria, Mora-Montes, 2020).

Myrcia neoobscura CHE and fractions also increased MIC values in the presence of sorbitol, used in the assay as an osmoprotector, indicating that these samples act on the integrity of the fungal cell wall. Fungal cells without a cell wall, or with a compromised cell wall, cannot survive or proliferate in the absence of sorbitol. Because it is a structure present only in fungal cells and not in human cells, antifungal agents that inhibit cell wall synthesis are generally less harmful. Therefore, the addition of a second mechanism of action, such as the cell wall, can reduce toxic effects, as found in M. neoobscura leaves (Junior et al., 2009; Curto et al., 2021). Also, no cytotoxic effects were detected, as found previously by our research group, through the MTT assay in fibroblasts (L929) (Krepsky et al., 2025).

GA was found to cause damage to the cell membrane of Trichophyton rubrum, a filamentous fungus, increasing fungal necrosis, reducing cell metabolism, biosynthesis and resistance to antifungal drugs (Fan et al., 2023). In prior studies the potential mode of action of GA have been reported as inhibiting the enzymes sterol 14α-demethylase and squalene epoxidase (Li et al., 2017), reduction in CYP51 enzyme activity, essential for the biosynthesis of sterols (Carvalho et al., 2018), binding with the active site residues of carbonic anhydrase (CA) protein from C. auris and C. albicans affecting their catalytic activities (Akhtar et al., 2023) and causing cell death trough the apoptosis mechanism in C. albicans strains (Liberato et al., 2022).

Chlorogenic acid was found in a previous study to have the ability to inhibit and destroy the hyphae of C. albicans, an important virulence factor of this species, while also injuring cellular membranes, causing the leakage of intracellular compounds as well as forming pores and dissipating the electrical potential of the membrane (Sung and Lee, 2010). Another known mechanism of chlorogenic acid is through apoptosis, which causes a reduction of cell viability, enhanced likelihood of mitochondrial membrane depolarization, elevated generation of reactive oxygen species, DNA damage with fragmentation, and externalization of phosphatidylserine on the cell surface (Silva et al., 2022). Other studies shows that resveratrol also acts through the apoptosis mechanism leading to cell death, and that p -coumaric acid acts through generating oxidative stress in fungal cells and interfering with ergosterol biosynthesis (Lee and Lee, 2015; Jaiswal and Kumar, 2025).

Therefore, the antifungal activity of natural products is generally not due to a single mechanism, but to a combination of them, mainly due to the wide variety of compounds found in an extract (Da Silva et al., 2024). Although there are other Myrcia species with antifungal activity, there were no mechanism of action against Candida species determined in previous studies for this genus.

Yeasts of the Candid a genus can adhere to surfaces, such as medical devices on a patient, and subsequently colonize them, where the biofilm formation being one of the genus’ main virulence factors (Malinovská, Čonková, Váczi, 2023). Biofilm formation is associated with increased mortality in candidemias, where deaths related to infections caused by non-biofilm-forming Candida were 37.9% while deaths related to biofilm formation can reach 70%. Although C. krusei is not the species most found in infections related to biofilm formation, it is associated with high mortality on biofilm-related infections, which can vary from 40%–59% (Atiencia-Carrera et al., 2022).

Although C. krusei is among the most frequently identified NAC species, it remains one of the least understood. Candida krusei can form biofilms on several surfaces, such as polyethylene, polyvinylchloride, and glass (Gómez-Gaviria and Mora-Montes, 2020). In this research, we identified C. krusei as a moderate biofilm former, similar to reported by Sachivkina, Podoprigora, Bokov (2021). Several phenolic compounds exhibit antibiofilm activity against species of the genus Candida, such as gallic acid, caffeic acid, epigallocatechin gallate, thymol and sesamol (Jaiswal and Kumar, 2022). These compounds, may be able to reduce initial adhesion, making it difficult for Candida cells and other fungi to attach to surfaces, thereby blocking the initial stage of biofilm formation; weaken the extracellular matrix, reducing the production of polysaccharides and proteins that make up the matrix, making the biofilm more permeable; increase the permeability of the cell wall and membrane, interacting with lipids and proteins, leading to a loss of cellular integrity; induce oxidative stress, structural damage and cell death within the biofilm; and interfere with quorum-sensing signals wich can modulate regulatory molecules that control biofilm maturation and dispersal (Slobodníková et al., 2016; Singh et al., 2015; Shahzad et al., 2014; Davidova, Galabov, Satchanska, 2024).

All M. neoobscura samples exhibited noteworthy inhibition of C. krusei biofilm formation, with inhibition percentages ranging from approximately 45%–60%, whereas amphotericin B showed inhibition values above 80%. It is important to observe that although AF displayed antibiofilm activity similar to that of DF and HF, these effects were achieved at concentrations approximately threefold lower than those required for the latter fractions. This result suggests a greater potency of AF and reinforces the potential contribution of its higher phenolic and flavanol content to the antibiofilm activity observed, indicating that these compounds may play a key role in the inhibition of Candida biofilms. Otherwise, no clear relationship was observed between the previously characterized phytochemical composition and antibiofilm activity, as DF, HF and AF chemical composition are substantially different, suggesting that, in addition to phenolic compounds, other constituents or synergistic interactions among metabolites may contribute to the observed effects.

To the best of our knowledge, this is the first study to demonstrate the antibiofilm activity of the hydroalcoholic crude extract and fractions of a Myrcia species, specifically M. neoobscura, against C. krusei biofilm. Previous studies involving species from the Myrtaceae family have reported reductions of up to 55% in C. albicans biofilm viability, as observed for Eugenia leitonii seed extract and Eugenia brasiliensis leaf extract (Sardi et al., 2017). Notably, the inhibition achieved in the present study reached up to 60.2% against C. krusei biofilm, highlighting the antibiofilm potential of Myrtaceae species and reinforcing the therapeutic potential of Myrcia species as sources of antifungal agents.

Studies on virulence factors of C. krusei are scarce when compared to C. albicans, so it is difficult to define the mechanism related to antibiofilm activity. In a study carried out on the action of gallic acid on C. albicans biofilm, the action occurred after 48 h of maturation, suggesting that the action may be due to the action of GA on biofilm matrix compounds, which are more present in mature biofilms after 48 h (Teodoro et al., 2018). Recently, phenolic acids, such as chlorogenic and p -coumaric acids, showed promising antibiofilm activity against C. albicans. Chlorogenic acid promoted significant disruption of the biofilm, indicating a partial inhibitory effect. In a similar manner, p -coumaric acid prevented the formation of mature biofilms, suggesting reduced biofilm integrity accompanied by limited interference in hyphal growth (Jaiswal and Kumar, 2025).

The results obtained in this study showed strong antifungal and antibiofilm activity of CHE from M. neoobscura leaves and its fractions against C. krusei. It was also identified that it is possible that M. neoobscura acts simultaneously on the cell wall and the fungal membrane. Considering that the toxicity of most antifungals is associated with the similarity between human and fungal cells, finding that these compounds can act through both mechanisms can mean that they may be less toxic to human cells than the ones that act solely on fungal membranes. These findings, together with the previously reported absence of cytotoxic effects (Krepsky et al., 2025), highlights the potential of M. neoobscura. Nevertheless, comprehensive safety evaluation, including hemolysis and in vivo toxicity studies, is still required before clinical translation.

A limitation of this study is the relatively small number of clinical isolates evaluated. Considering the substantial genotypic and phenotypic diversity among Candida clinical isolates, including differences in virulence traits, biofilm-forming capacity, and antifungal susceptibility, the present findings should be interpreted with caution regarding their broader translational applicability. It was not possible to include a reference C. krusei strain because it was not available in our laboratory at the time o of our research. The inclusion of a reference C. krusei strain would further strengthen the comparative analysis and should be considered in future studies. Biofilm disruption was evaluated only at the MIC and MICx2 concentrations. A broader range of doses would be necessary to more comprehensively characterize any dose-response relationship. Although the observed antifungal and antibiofilm activities are encouraging, future studies involving a larger and more genetically diverse collection of clinical isolates are necessary to validate the reproducibility and clinical relevance of these results. Furthermore, in future studies, the inclusion of assays to determine whether the activity observed is fungistatic or fungicidal are necessary.

There is a scarcity of studies on the activity of natural products against C. krusei. This is the first study to demonstrate the antibiofilm activity of the CHE and its fractions from a species of the Myrcia genus and the M. neoobscura species against C. krusei biofilm. These results suggest that there is a promising prospect of using natural products to develop new therapeutic options for infections caused by C. krusei. Thus, knowledge about the antifungal properties of a plant native to the Atlantic Rainforest was made possible, demonstrating the importance of bioprospecting studies and the sustainable use of native flora.