Talinum paniculatum: a plant with antifungal potential mitigates fluconazole-induced oxidative damage-mediated growth inhibition of Candida albicans

Aims: This study investigated the bioactivity of the crude leaf extract (CLE) and fractions hexane (HX) and ethyl acetate (EtOAc) from Talinum paniculatum alone and in association with fluconazole (FLC) against reference strain and clinical isolates of FLC-resistant Candida albicans. Furthermore, the antioxidant capability, chemical composition of this plant, and the effect’s underlying mechanisms were evaluated. Methods: The antifungal activity was evaluated using checkerboard assay to establish the minimum inhibitory (MIC) and minimum microbicidal concentrations (MMC). During FLC and plant products challenges, the reactive oxygen species (ROS) generation (hydroxyl radicals [HO•]) were detected in C. albicans cells using the membrane-permeable fluorescent probes APF and HPF. High-performance liquid chromatography (HPLC) profile, quantitative analysis of antioxidant compounds, and free radical scavenging activity (DPPH assay) tests were performed. Results: The CLE and fractions presented outstanding antifungal activity and selectivity against C. albicans cells but had no synergistic effects with FLC. The MIC values for CLE and its fractions against C. albicans reference strain were in the order of HX (31.25 μg ml–1) < EtOAc (62.5 μg ml–1) < CLE (500 μg ml–1), and against FLC-resistant C. albicans HX (125 μg ml–1) = EtOAc < CLE (500 μg ml–1). CLE and its fractions had more potent antifungal activities than FLC against the clinical isolates. Moreover, fungicidal effects for these plant products were demonstrated against FLC-resistant C. albicans, which further confirmed an antifungal potential. Conversely, during association, plant products were shown to cause an increase in FLC MIC anywhere from 2- to 16-fold. FLC exposure led to an increase in the steady-state levels of ROS (HO•) in C. albicans cells. Next, we found that the increases in FLC MICs were owing to action of antioxidants containing-CLE and its fractions in preventing FLC-induced ROS-mediated growth inhibition of C. albicans. Conclusion: T. paniculatum can be a source of bioactive compounds with antifungal potential. However, because of the common use of its edible leaf, caution is advised during therapy with FLC (since it can decrease FLC susceptibility). 


Introduction
Over the last few decades, the relentless increase of antimicrobial resistance and multidrug resistance (AMR/MDR) in microorganisms has been observed in low-, middle-, and high-income countries and has resulted in mortality rates in patients who have hospital-or community-associated infections (HAI/CAI) reaching numbers comparable to the pre-antimicrobial era [1][2][3][4]. The ever-increasing incidence of AMR/MDR combined with a weak pipeline of new antimicrobial agents launching on the market has created a great threat to the successful management of infectious diseases as well as a major global public health problem [5][6][7].
Since the early 1980s, AMR has risen to alarming levels in Candida albicans (i.e., elevated MICs to azoles and cross-resistance to related antifungal agents) -a frequent pathogen of immunologically compromised individuals, but an even more common commensal fungus from the microbiome of healthy humans-, and the severity of candidiasis caused by antimicrobial-resistant C. albicans strains has increased over the years due to clinical interactions, such as the use of corticosteroids, immunosuppressive agents, radiotherapy, and anti-tumoral chemotherapy [1,2,8]. Furthermore, the spread of resistance, emergence of highly virulent and pathogenic strains of C. albicans that commensally colonize 30-70% of healthy individuals, and formation of biofilms underlie chronic and recurrent infections accordingly have increased the ability of this microorganism to cause disease, further aggravating the issue of AMR/MDR. These factors are, at least partially, responsible for the high invasive fungal infections (IFIs) and mortality rates from C. albicans infections [2].
Approximately 400 000 life-threatening infections caused by C. albicans are reported per year worldwide. In the United States, from the estimated 46 000 healthcare-associated C. albicans infections reported each year, 3400 are caused by Fluconazole-resistant C. albicans, with approximately 220 deaths. In this context, regarding level of concern, Fluconazole-resistant C. albicans is today recognized by the centers for disease control and prevention (CDC) as a "serious threat" to people's health, the second major threat into one of three threat categories: urgent, serious, and concerning. In Brazil, the real data are still underestimated, but studies have showed a considerable prevalence of Fluconazoleresistant C. albicans infections, and besides the natural process of AMR/MDR, the inadequate antimicrobial use (most from food production) is the single most important factor leading to antibiotic resistance in this country [9][10][11][12].
The serious threat of AMR/MDR in C. albicans strains represents a current public health problem and poses a burgeoning need for new antifungal agents that can tackle AMR/MDR as well as novel antifungal agent deployment strategies are imperative and urgent, to address the issue of AMR/MDR [6,[9][10][11][12][13]. Moreover, new antifungal mechanisms are a field to be explored against the broad antifungal resistance seen in epidemiology. Among the approaches to cope with AMR, plants have been used for medicinal purposes long before recorded history [14]. Talinum paniculatum (Jacq.) Gaertner (Talinaceae family) is commonly known in Brazil as "Erva-gorda" and "Língua-de-vaca", or Jewels of Opar and Ginseng Java worldwide. Nowadays, this plant is widely spread throughout the world and all Brazilian territories. Its edible leaves are succulent subshrub and make an excellent addition to salads, among other culinary purposes (a non-conventional edible leaf) [15].
In folk medicine, T. paniculatum is used to treat several pathological conditions including headaches, ulcers, and diarrhea; as an emollient for fighting gastrointestinal disorders; and used topically to combat a broad spectrum of wounds and skin infections. Furthermore, it is used to ease digestion, moisten the lungs, as an aphrodisiac, and to promote breast milk production [15,16]. Studies have demonstrated the phytochemical constitution [17] and several bioactivities of this plant and its metabolites, including antibacterial and antifungal [18], estrogenic [19], antifertility [20], antinociceptive [21], and the induction of uterine contractility [16].
Over the past decade, intense investigation revealed that the action of antibiotics with specific targets into microbial cells can be accompanied by oxidative stress (i.e., mainly via an increase in respiratory chain-dependent reactive oxygen/nitrogen species [ROS/RNS] production), which is considered a common mechanism of antibiotic-mediated cell death (referred to as "unified mechanism of killing"), as those seen during the killing of C. albicans, Staphylococcus aureus, Escherichia coli, Mycobacterium tuberculosis, and Pseudomonas aeruginosa [22][23][24][25]. Recently, a great number of antibiotics have been demonstrated to stimulate the production of ROS in microbial cells [22][23][24][25]. More specifically, studies conducted by Kobayashi et al. [26]; Silva et al. [23]; and Mahl et al. [22] have shown a participation of ROS in antifungal mechanism of fluconazole (FLC)-the first-line antifungal treatment agent against C.
albicans infections-and other azoles, resulting in oxidative DNA damage-mediated cell death [9].
Additionally, in recent years, studies have shown that the interaction between plant product and antibiotic can increase the minimum inhibitory concentration (MIC) of some antibiotics, so that their actions could be hampered by antioxidants containing-plant products [27,28]. This effect could be from a previous decrease in oxidants; thus, compromising the mechanisms of antibiotic-induced ROS/RNSmediated cell death, such as ROS-induced guanine pool oxidation that leads to double-strand DNA breaks and cell death. This might also have consequences in vivo, reducing the pharmacokinetics, efficacy, and even safety of antibiotics and leading to microbial resistance, causing high medical costs during treatment and higher risk of death for patients [27,28].
In that way, reductive stress caused by antioxidants has been investigate in the non-communicable diseases, for example, cardiovascular disease and cancer (i.e., via down regulation of Nrf2 pathways), and communicable diseases, for example, infectious diseases (i.e., via a decrease in oxidants generated in phagocytic cells) [29,30]. However, the influence of reductive stress on the susceptibility of microorganisms to antibiotics remain poorly explored. Moreover, to the best of the authors' knowledge, there are no reports about the influence of T. poaniculatum on the susceptibility of C. albicans to FLC.
The elucidation of possible interactions between antioxidants-rich foods and antibiotic is important because of many nutrition products and medicines -including the leaves from T. paniculatum that are often used as a green leafy vegetable for human consumption-may interact with antibiotics during medical treatment and modify their action hence having a great impact on clinical practice and patient outcomes.
Based upon the above, we evaluated the antifungal activity of the crude leaf extract (CLE), fractions hexane (HX), and ethyl acetate (EtOAc) from T. Paniculatum, alone or in association with the azole antifungal FLC, against reference and FLC-resistant C. albicans strains. We also investigated the influence of the CLE, HX, and EtOAc from T. paniculatum during the action of FLC against these strains with focus on oxidative/reductive stress. Furthermore, chemical constituents and antioxidant activity of these plant products were checked.

Ethics statement
This study did not involve any endangered or protected species and no specific permits were required for the described studies. Botanical material from T. paniculatum was collected in an area, with access permitted to researchers. fractions, and a final residue termed the aqueous fraction (this last not used in this study) [18]. At time of their use, the CLE or fractions were solubilized in DMSO (at the concentrations used for solubilization in this study, DMSO does not display antifungal, cytotoxic, antioxidant, or other activities here evaluated) and adjusted at used concentrations.

Antifungal activity
Strains C. albicans (ATCC ® 90028 and ATCC 10231) used in this study were from the American Type Culture Collection (ATCC, Manassas, VA). Antibiotic-resistant C. albicans were isolated from clinical source (patients with orofacial cleft), as shown in table 1 and identified as previously reported [31].  5´ CCA TTA CAG CTG AAC CAG CGA GGG (F) 5´ CGC TAG GTA ACC TAC AGA TTG TGG C (R) MEE: Multilocus Enzyme Electrophoresis; PCR: Polymerase Chain Reaction; *These species-specific primers for C. albicans were used to amplify a DNA fragment of approximately 1,644 bp (PHR1 gene); **Origin: from children with cleft lip and palate; ***Conditions: Amplifications were conducted using an initial program for DNA denaturation at 95 °C for 5 minutes, followed by 30 cycles at 95 °C for 20 seconds (denaturation), 50 °C for 1 minute (annealing), and 72 °C for 1½ minute (extension). The final cycle was conducted at 72 °C for 10 minutes for the final extension; ****(i) CAI: initial program for DNA denaturation at 95 °C for 5 minutes, followed by 30 cycles at 94 °C for ½ minute (denaturation), 60 °C for ½ minute (annealing), and 72 °C for 1 minute (extensio n). The final cycle was conducted at 72 °C for 7 minutes for the final extension, (ii) CDC3: initial program for DNA denaturation at 95 °C for 5 minutes, followed by 35 cycles at 95 °C for 1 minute (denaturation), 52 °C for 1 minute (annealing) and 72 °C for 1 minute (extension). The final cycle was conducted at 72 °C for 7 minutes for the final extension, (iii) ERK1, KRE6, LOC4 and ZNF1: initial program for DNA denaturation at 95 °C for 5 minutes, followed by 35 cycles at 95 °C for 1 minute (denaturation), 55 °C for 1 minute (annealing) and 72 °C for 1 minute (extension). The final cycle was conducted at 72 °C for 5 minutes for the final extension, (iv) CPH1 and MNT2: initial program for DNA denaturation at 94 °C for 5 minutes, followed by 40 cycles at 94 °C for 1 minute (denaturation), 50 °C for 1 minute (annealing) and 72 °C for 1 minute (extension). The final cycle was conducted at 72 °C for 5 minutes for the final extension [31].

Minimum inhibitory concentration and determination of synergistic action
Both plant product (CLE, HX and EtOAc from T. paniculatum) and FLC were tested to determine the MIC values against the two resistant C. albicans strains, as well as against the reference strain of C.
The plates were then incubated at 37 C for 48 hours. After the incubation period, readings were performed visually as previously determined [18], wherein the presence of turbidity in the wells was [w/v] of glucose; Petri plates, 9015 mm) and then maintaining them at room temperature for complete absorption. After that, the plates were incubated at 37 C for 48 hours. Finally, colony counts were performed to determine which concentrations presented microbicidal (fungicidal) or microbistatic (fungistatic) action.

Evaluation of the amount of intracellularly-generated oxidants in C. albicans
The quantification of hydroxyl radicals (HO • ) generated under the different treatments was performed using the probe APF (highly specific toward HO • ) and HPF [33]. First, reference strain of C. albicans (ATCC 10231) was grown overnight on BHI. Next, microbial suspensions (in exponential-phase growth) were centrifuged (900g for 6 min) and the pellet diluted in sodium chloride 0.9% and spectrophotometrically adjusted. Afterwards, C. albicans cells (1  10 4 CFUs/mL -1 ) were added on wells of a 96-well microplate and previously treated (for 10 min) with CLE (100 μg ml -1 ) or fractions (EtOAc

Cell-free assays
Antioxidant activity of the CLE and fractions from T. paniculatum.
The free radical scavenging ability of the CLE and fractions was measured using the 2,2-Diphenyl-1picrylhydrazyl (DPPH) assay [34]. The results were reported in percentage (%) of DPPH • scavenging activity inhibition.

Quantitative analysis
The Folin Ciocalteu reagent was used to determine the total polyphenols content of the CLE and fractions according to Fattahi et al. [34]. As a standard, gallic acid was used. From the calibration curve, total polyphenol content was expressed in terms of gallic acid equivalent per gram of sample (mg/g).

Chemical characterization of T. paniculatum
High-performance liquid chromatography (HPLC) was performed for analysis of the CLE from T. FICA + FICB. "Synergy" was defined as an FICI ≤ 0.5, while "antagonism" was defined as an FICI value > 4.0. A FICI between 0.5 and 1.25 was considered as "Additive effect", while between 1.25 and 4 was considered to have "No interaction" [35][36][37]. According to the ratio of MMC/MIC, we reported the type of antifungal action displayed by the sample. If the ratio of MMC/MIC= 1 or 2, the effect was considered fungicidal, but if the ratio of MMC/MIC= 4 or 16, the effect was defined as fungistatic [38].
The Selectivity Index (SI) was calculated as follows: SI= CC90/MIC99.9, being CC90 adopted from our previous study [18]. The significance of difference was analyzed by one-way ANOVA and a Tukey post-test (BioEstat 5.0, Belém, Pará, Brazil, 2007). Significance was accepted at p<0.05 (α= 5%), unless indicated otherwise. Structural elucidation of compounds from T. paniculatum was performed by interpreting the second mass spectra order upon the fragmentation patter assumption. Structures depicted in this study were constructed on ACD/Labs (Advanced Chemistry Development Inc., version 6.0).

MIC and MBC values for the CLE, EtOAc, and HX from T. paniculatum
As observed in albicans strains. The CLE was more effective than FLC against the two antibiotic-resistant C. albicans, but not against the reference strain of C. albicans. However, when the mixture of extract/fraction and antifungal agent was assayed to determine the type of interaction between the two compounds, FLC MICs increased 2-to 16-fold (table 3), whereas some fractions MICs were shown to be decreased as an effect experienced during these associations. Taken together, data from tables 2 and 3 were interpreted as the FICI values (table 3) in which there was no synergy for any of associations between plant product and antifungal agent. The observed effects for the associations were: "additive effect" (2 associations), "no interaction" (5 associations), or "antagonism" (5 associations).

Antifungal effects/actions for the crude leaf extract (CLE) and fractions
In table 2, it is also shown that the CLE displayed fungicidal action against antibiotic-resistant C.
albicans (sample 1). There was a fungicidal action from both EtOAc and HX against antibioticresistant C. albicans, but only HX presented this effect against the two samples (1 and 2), whereas the effect of EtOAc against sample 1 of the antibiotic-resistant C. albicans and both fractions against the reference strain of C. albicans were fungistatic. Therefore, the EtOAc and HX fractions not only presented the lowest MIC values, but also presented fungicidal action.

Chemical characterization, antioxidant property, and total polyphenol content of T. paniculatum
Here, we identified compounds likely to be related to the bioactivities. To support the increase in MIC values of FLC in the association with plant products, table 5 shows the antioxidant activity and the total polyphenols content of the CLE and fractions. EtOAc and HX fractions presented higher polyphenols contents and the best results as for antioxidant activity. Table 6 shows some compounds identified by HPLC (retention time in minutes, compared to standards) and illustrates the proposed structures. Figure 1 presents the chromatogram of the CLE, which renders the identified structures drawn on it (compounds also depictured in table 6). Antifungal [39] Antibacterial [39] Antioxidant [40] Antibiotic interaction: ----Ascorbic acid CLE: + EtOAc: ----HX: ----

Discussion
In this study, we demonstrated that the CLE, EtOAc, and HX from T. paniculatum present outstanding antifungal activity against C. albicans, with "moderating" or "promising" inhibitory potential [18].
The CLE and its fractions (EtOAc and HX) promoted growth inhibition of the reference strain of C.
albicans (fungistatic effect). Against FLC-resistant C. albicans, the cell death was the principal outcome of the treatments (fungicidal effect). With focus on a selectivity of T. paniculatum against C.
albicans cells, outstanding SI values were observed for the CLE, EtOAc, and HX against both clinical isolates of FLC-resistant C. albicans and reference strain of C. albicans. A previous work has shown notable SI values of T. paniculatum toward C. albicans cells (ATCC 10231) and other microorganisms [18].
The pathogenic potential (overgrowth) of C. albicans arises after a break in homeostasis and depends on factors of both the human host, for example, immunosuppression, and the microorganism, for example, virulence of the strain. Thus, in these cases, the invasion of C. albicans into vital sites of immunocompromised patients causes the so-called IFI (sometimes life-threatening infections) and accounts for the high mortality rates from infectious diseases caused by this fungus [61,62]. Here, the clinical isolates of C. albicans presented resistance to FLC, which is commonly used to treat C.
albicans infections. The LE, HX, and EtOAc fractions from T. paniculatum presented significant inhibitory activity against clinical isolates of C. albicans. Moreover, the activity of terpenes, sterols (compounds found in T. paniculatum), and plant products with high terpenes/sterols contents were also demonstrated against Candida species [18,63].
In vivo, even on antibiotic action, the killing of pathogens in humans requires a competent immune system. The immunocompetent host is usually far better equipped to eliminate C. albicans than an immunosuppressed host. Therefore, it is especially desirable to have a truly microbicidal drug-one that absolutely kills the microorganisms-as a treatment option for immunosuppressed patients [64]. In this context, regarding the type of antifungal effect expected for the CLE and fractions from T.
paniculatum, as seen here, and bearing in mind the clinical considerations above mentioned, the antimicrobial agents with fungicidal action are preferred to with fungistatic action.
Interestingly, the MIC and MBC values may vary for the actions of the CLE, EtOAc, and HX, as well as for FLC, as consequences of different AMR profiles of C. albicans, used in this study. Furthermore, differences between chemical compositions of the fractions or CLE (i.e., in part, from the differences in solvents used to extraction) and the association with antifungal may explain the effects found in this study, since the concentration of metabolites produced may produce a combination of antimicrobial effects or inactivation of FLC [65].
The different compounds in T. paniculatum, here evidenced by HPLC, explain the observed bioactivities, including the effects when there were associations between plant product and antifungal.
As examples, chlorogenic acid, benzoic acid, caffeic acid, and ferulic acid, and phytosterols display antifungal activity [45,46,[50][51][52][53][55][56][57][58][59]. In contrast, isolated ascorbic acid does not have antifungal activity up to 250 μg ml -1 , as reported by Khalil et al. [41],  Over the last few decades, controversies about the participation of intense ROS production during antibiotic action have generated discussion [68,69]. Reinforcing the theory that high levels of ROS are induced during antibiotic action, the presence of some exogenous redox-active compounds with pro-or antioxidant activity, such as polyphenols and other plant products, can change the antibiotic action. It has been shown that antioxidants promote a diminution in the expression of genes related to ROS detoxification systems in microorganisms, reversing expectations of an increase in gene expression-a common consequence of oxidative stress induced by antibiotics- [27], as seen in cases of antibiotic resistance because of a higher oxidative stress tolerance in C. albicans [70] and P.
aeruginosa biofilms [71]. As demonstrated here, we found that these increased This effect was also assigned by the author to an antioxidant action of the compounds/extracts [27].
Of great relevance, in this study, we considered intracellularly-generated and confined oxidants in C.
albicans cells, so that we used membrane-permeable fluorescent probes that readily diffuse through the cell membrane and then are rapidly oxidized to highly fluorescent products. The treatments with the CLE, EtOAc, and HX were performed; and these plant products present antioxidant compounds able to cross a membrane. This is further supported in previous studies, in which non-polar compounds with antioxidant action have been described for this plant [17,18].
In this study, we detected elevated HO • formation in C. albicans cells induced by FLC, which may, at least in part, explain the link between FLC treatment and oxidative damage in DNA (quantifying 8- tropicalis [23]. However, different strains may differ in the sensitivity and response to oxidative stress because of a specific sensitivity to FLC. In addition, azole antifungals act in a time and concentration dependent-manner in these cases, and the associated generation of ROS after antifungal exposure to appear not to be directly related to them but by some by-products generated through their mechanisms of action [22]. Silva et al. [23] and Mahl et al. [22] have shown that resistance to FLC among Candida species can involve increased gene expression of products related to redox homeostasis systems, such as genes for the synthesis of antioxidant enzymes, GPx, Sod, and GST [70], provided that the antifungal activity of FLC has been reported to be dependent of the intensive ROS production, which generate DNA damage (consequence of an excessive farnesyl pyrophosphate formation) beyond the effects in inhibiting ergosterol biosynthesis and consequent changes in the fungal membrane. A recent study has also demonstrated that FLC at subinhibitory concentrations induced oxidative-and nitrosative-responsive genes TRR1, GRE2, and YHB1, and led to AMR profile in C. albicans and resistance to phagocytes [9]. (different antibiotics representing a wide diversity of mechanisms of action) against an experimental model of prokaryote microorganism (bacteria), may indicate a putative mechanism of antioxidants, as ROS scavengers, to interfere with the susceptibility of C. albicans, rather than, a pleiotropic effect.
The limitations of our study should be noted, which are the analysis of oxidative stress response just in exponential growth phase (i.e., yeast cells in exponential and stationary phase can respond in a different manner to oxidative stress) and a verification upon the C. albicans exposure to FLC more indepth, under the influence of treatments with the CLE, HX, or EtOAc, and the correspondent C.
albicans antioxidant response (analyses of the systems of ROS detoxification: GPx, SOD, and GST genes and related enzymes), total glutathione, and oxidative damage in DNA (since FLC-induced oxidative damage upon lipids and proteins appears to be no significant in this case) [22]. These tests could help to shed light whether the treatments are priming/modulating genes expressions/enzymes activities and/or whether they are creating direct antioxidant effects (most probably) or other types of interaction with FLC.
Taken together, our findings suggest that T. paniculatum presents potential for furthers studies to look at it as an antifungal. This plant presents low toxicity at concentrations of optimal antifungal activity (significant SI values). As a recommendation, additional studies are required to isolate new compounds with original antifungal actions through a bioassay-guided approach. Alternatively, a study could be conducted regarding the efficacy, safety, and pharmacokinetics for a herbal drug from T. paniculatum leaves to treat infectious diseases, including those caused by AMR, since the CLE from T. paniculatum alone presents outstanding antifungal activity, its inability to have significant synergic effects with antibiotics notwithstanding.
Conversely, these findings also provide novel insights into redox regulation of C. albicans cells during the association between plant product and antifungal. We used an experimental model of antifungal action (FLC), demonstrating a plausible putative activity of antioxidants containing-plant products in preventing the effects of FLC by decreasing oxidants. Of clinical relevance, our data support the fact that some polyphenols and other antioxidant compounds, being a part of many nutrition products and medicines, may interact with antibiotics during medical treatment and modify their action. Therefore, since the dietary intake of T. paniculatum as a green leafy vegetable is common, a word of warning should be issued regarding the association between T. paniculatum and FLC, in face of the antagonistic effects here demonstrated; thus, during antibiotic therapy, the physician should take all this into account, since patients under FLC treatment parallel to an intake of this plant may cause a decrease in antifungal efficacy.
Nonetheless, these data should also prompt in vivo studies, focusing additional effects of exogenous products with antioxidants on the host/organism to observe resistance phenomenon and/or the failure of therapeutic regimens, since some combinations may increase the MIC of antibiotics in vitro and there is possibility for this also occurs in vivo-in addition to-be worthwhile to target these interactions between plant product and antibiotics and associated effect's underlying mechanisms as a means to enhance the killing efficacy of available antimicrobial agents.