Antifungal and Anti-Inflammatory Potential of Bupleurum rigidum subsp. paniculatum (Brot.) H.Wolff Essential Oil

Fungal infections remain a major health concern with aromatic plants and their metabolites standing out as promising antifungal agents. The present study aims to assess, for the first time, the antifungal and anti-inflammatory potential of Bupleurum subsp. paniculatum (Brot.) H.Wolff essential oil from Portugal. The oil obtained by hydrodistillation and characterized by GC-MS, showed high amounts of monoterpene hydrocarbons, namely α-pinene (29.0–36.0%), β–pinene (26.1–30.7%) and limonene (10.5–13.5%). The antifungal potential was assessed, according to CLSI guidelines, against several clinical and collection strains. The essential oil showed a broad fungicidal effect being more potent against Cryptococcus neoformans and dermatophytes. Moreover, a significant germ tube inhibition was observed in Candida albicans as well as a disruption of mature biofilms, thus pointing out an effect of the oil against relevant virulent factors. Furthermore, fungal ultrastructural modifications were detected through transmission electron microscopy, highlighting the nefarious effect of the oil. Of relevance, the oil also evidenced anti-inflammatory activity through nitric oxide inhibition in macrophages activated with lipopolysaccharide. In addition, the essential oil’s bioactive concentrations did not present toxicity towards macrophages. Overall, the present study confirmed the bioactive potential of B. rigidum subsp. paniculatum essential oil, thus paving the way for the development of effective drugs presenting concomitantly antifungal and anti-inflammatory properties.


Introduction
Fungal infections account for nearly 1.6 million deaths worldwide, thus presenting a huge socio-economic burden, frequently neglected by public health authorities [1]. Most fungal infections affect mucous membranes, especially vaginal and mouth membranes [2] and, despite being primarily superficial, mycosis can rapidly evolve into systemic infections, especially in immunocompromised individuals. These type of infections are mainly caused by Candida albicans, nevertheless other Candida species are becoming important etiological agents such as C. krusei, C. parapsilosis and C. tropicalis [3]. Several features contribute to Candida spp. virulence including yeast-to-hypha morphological transition and biofilm formation [4]. The first, starts with germ tube formation that develop into hypha promoting Among Bupleurum species, B. rigidum L. is one of the most widely represented in Portugal. This species is widely used in Spain for the treatment of topical inflammation, as anti-diarrheic and against furuncles [21]. Two subspecies of B. rigidum have been described, namely subsp. paniculatum and subsp. rigidum. B. rigidum subsp. paniculatum (Brot.) H.Wolff is an endemic species to the south and center of Portugal, Andalucía (Spain) and North Africa [32] and, up to date, no studies on this subspecies have been conducted. Therefore, and as part of our ongoing studies on the valorization of wild Apiaceae species, we aim to characterize the essential oil composition of B. rigidum subsp. paniculatum from Portugal and assess its bioactive potential. A combined antifungal and anti-inflammatory potential for B. rigidum subsp. paniculatum essential oil is reported, with emphasis on its effect against C. albicans virulence factors. Table 1 summarizes the chemical profile of two samples of B. rigidum subsp. paniculatum analyzed by GC-MS. A yield of 0.7% (w/v) was obtained for the essential oil from Coimbra and 0.6% for the oil from Fátima. More than 95% of the compounds were identified in both samples. The chemical analysis showed that both essential oils are characterized by the presence of high amounts of monoterpene hydrocarbons (94.4-96.6%), mainly α-pinene (29.0-36.0%), β-pinene (26.1-30.7%) and limonene (10.5-13.5%). The essential oil from Coimbra region represents a much larger population and, therefore, was selected to assess the antifungal and anti-inflammatory potential of the oil.

Essential Oil Composition
2.2. Antifungal Activity 2.2.1. The Essential Oil Showed a Broad-Spectrum Fungicidal Effect The antifungal potential of B. rigidum subsp. paniculatum essential oil was assessed against several yeast, dermatophyte and Aspergillus strains as depicted in Table 2. Besides collection strains, fungal isolates obtained from the clinic were also assessed, thus providing data on the antifungal potential of the oil against more resistant strains. Overall, the essential oil showed a significant antifungal effect against Cryptococcus neoformans and Trichophytum rubrum with MIC values of 72 µg/mL ( Table 2). For the remaining dermatophytes, the oil was also effective with MIC values ranging from 144 µg/mL to 288 µg/mL. In addition, relevant antifungal effects were observed for Aspergillus and the majority of Candida strains (Table 2). Interestingly for the majority of the tested stains, the oil showed similar MIC and MLC values, thus presenting a fungicidal effect ( Table 2). This is quite relevant since azoles, namely flucanozole, an antifungal drug widely used in the clinic, shows primarily fungistatic effects. Indeed, this antifungal drug, for the majority of the tested strains, is able to inhibit fungal growth but at the MIC it is not lethal to the fungi.

The Essential Oil was Effective against Candida albicans Virulent Factors
Candida albicans is able to switch from yeast-to-hypha morphology promoting tissue invasiveness [33] and to form biofilms, an organized structure more resistant to antifungal agents than planktonic fungi [34]. These features play a pivotal role in C. albicans pathogenesis and virulence, thus constituting promising therapeutic targets. Therefore, the present study aimed to assess the effect of the essential oil on both features (Figures 1 and 2). As shown in Figure 1, the essential oil strongly inhibited germ tube formation in C. albicans at concentrations well below its respective MIC. Interestingly, at concentrations as low as MIC/8 (72 µg/mL) the oil inhibited germ tube formation by ca. 40% and at MIC/4 (144 µg/mL), 88% of inhibition was observed (Figure 1a). Fluconazole failed to accomplish this up to 200 µg/mL, as pointed out in Figure 1b, thus highlighting the relevance of our results.  One-way ANOVA followed by Dunnett's Multiple Comparison Test, ** p < 0.01; *** p < 0.001, compared to control.
The capacity of the essential oil to decrease biofilm mass and biofilm viabilitywas also tested as well as that of fluconazole ( Figure 2). Overall, the oil was effective and decreased biofilm viability at concentrations similar to the respective MIC and at a slightly higher concentration, the oil also disrupted mature biofilms ( Figure 2a). Fluconazole, on the other hand, was only effective in decreasing C. albicans viability at a much higher concentration than its respective MIC and was ineffective in reducing biofilm mass even at concentration 200× higher than its MIC, as shown in Figure 2b Figure 4c-h) presented a structural disorganization, particularly at the cell wall (e.g., Figures 3c and 4c), for C. albicans and T. rubrum respectively. These alterations can be due to an impairment in the deposition of cell wall constituents. In C. albicans cytoplasmic disintegration was also observed (e.g., Figure 3h). Particularly for T. rubrum an increased number of vacuoles was observed with the accumulation of electron dense material (Figure 4c-h), reflecting a defensive mechanism of the fungus. As the number of vacuoles significantly increased with the essential oil treatment, it seems that autophagy is activated [35]. Indeed, autophagic structures were observed (Figure 4h), thus suggesting that the oil activates this pathway as a protective mechanism of the fungi.  Effect of Bupleurum rigidum subp. paniculatum essential oil (a) and fluoconazole (b) on Candida albicans ATCC 10231 biofilm biomass and viability. One-way ANOVA followed by Dunnett's multiple comparison test (performed in separate on each feature), * p < 0.05, compared to biofilm mass control; # p < 0.05 and ## p < 0.01, compared to biofilm viability control; arrows indicate MIC value.

Anti-Inflammatory Activity
Chronic inflammation contributes to the successful colonization of the host tissues by pathogenic fungi, thus compromising disease suppression. Indeed, an antifungal drug with inherent anti-inflammatory activity, would enable a rapid symptomatic relief while providing an effective treatment. Therefore, we also assessed the anti-inflammatory potential of the essential oil using an in vitro model of lipopolysaccharide (LPS)-stimulated macrophages. The effect on NO production, a canonical marker of inflammation, was analyzed by measuring the accumulation of nitrites in the culture medium. Moreover, the effect of the oil on macrophages' viability was also tested ( Figure 5). Expectably, our results show that macrophages without LPS stimuli produced residual levels of nitrites (<1 µM) and this value increased to 35.25 ± 3.91 µM in the presence of LPS (Figure 5a). Interestingly, nitrite production was significantly reduced in the presence of the essential oil, attaining a decrease of more than 51% with 576 µg/mL of the oil (17.15 ± 3.35 µM) and a decrease of 33% with half the concentration (23.53 ± 3.93 µM; Figure 5a). Importantly, all the tested concentrations of the essential oil did not affect macrophages' viability ( Figure 5b), thus confirming a safety profile of the oil at concentrations exhibiting pharmacological activity.    paniculatum essential oil on NO production (a) and cell viability (b). One-way ANOVA followed by Dunnett's Multiple Comparison Test, ### p < 0.001, compared to control; ** p < 0.01; *** p < 0.001, compared to LPS. EO-essential oil.

Discussion
Bupleurum rubrum subp. paniculatum essential oils from Portugal showed high amounts of monoterpene hydrocarbons with α-pinene, β-pinene and limonene standing out as the main compounds. Similar compounds were also reported for essential oils obtained from plants collected in Montes Toledo, Spain, although in these plants myrcene was also found in high amounts (26.2%) [36]. These differences might be due to different growing locations with diverse abiotic factors that may influence the biosynthesis of essential oils. In a previous work, we developed a protocol for the in vitro propagation of Bupleurum species and observed the presence of the major compounds inside secretory ducts of the in vitro propagated plantlets of B. rigidum subsp. paniculatum [37]. Since secondary metabolites signature in spontaneous plants was also found in the in vitro propagated plants, this technique could enable a large-scale production of this species, thus reducing the influence of abiotic factors and enabling the production of uniform essential oils. Furthermore, this approach could allow a year-round plant supply.
To the best knowledge of the authors, no reports on the bioactive potential of B. rigidum subsp. paniculatum have been conducted. Nevertheless, the antimicrobial activity of other Bupleurum species has already been reported. For example, the essential oil isolated from B. marginatum, rich in tridecane, undecane, pentadecane, β-caryophyllene and β-caryophyllene oxide, showed a MIC of 4 mg/mL towards two strains of C. albicans and inhibited several bacterial strains, including methicillin-resistant S. aureus (MIC = 0.063-4 mg/mL) [22]. The antifungal activity of B. gibraltaricum, characterized by sabinene, α-pinene and 2,3,4-trimethylbenzaldehyde, was demonstrated for Plasmopara halstedii through fungal sporulation inhibition at 5 mL/L [30]. The essential oil from B. longiradiatum, having thymol, butylidene phthalide, 5-indolol, heptanal, 4-hydroxy-2methylacetophenone, 4,5-diethyl-octane, borneol and hexanoic acid as main compounds, was able to inhibit the growth of several bacteria with MIC ranging from 250-500 µg/mL [27]. The essential oil herein presented showed a broad antifungal activity, very effective against Cryptococcus neoformans and Trichophyton rubrum (MIC = 72 µg/mL). The antifungal effect of the essential oil might be attributed to the presence of several compounds with strong antifungal activity, namely α-pinene, limonene, β-pinene and myrcene [38][39][40][41][42][43][44][45][46]. Regarding the anti-virulent effects of Bupleurum spp., no studies have been conducted until now. Nevertheless, the major compounds of the essential oil were able to decrease the yeast-to-hypha transition in C. albicans. Indeed, β-pinene [47,48] and limonene [48] were described as germ tube inhibitors. Similarly, the capacity of the oil to disrupt preformed C. albicans biofilms might also be attributed to the presence of limonene and β-pinene as both compounds have been described as effective in disrupting mature biofilms. Nevertheless, the influence of minor compounds cannot be discarded. Indeed, myrcene and p-cymene were also able to inhibit the formation of C. albicans biofilms [48].
Regarding ultrastructural characterizations, studies on dermatophytes are scarce, thus highlighting the relevance of our results. In Candida species, essential oils tend to act on ergosterol synthesis, modify fungal cell morphology, modulate membrane permeability and produce reactive oxygen species [49]. In accordance, our observations showed alterations in the cell wall structure as well as cytoplasmic disorganization, thus contributing to an overall modification of fungal cell morphology. Similar effects were reported for conventional antifungal drugs, namely caspofungin alone or in conjugation with pyroquilon in Alternaria infectoria [50], suggesting a similar mechanism of action for B. rigidum subsp. paniculatum essential oils. Regarding T. rubrum, our results clearly show that the essential oil induces an increase in the number of vacuoles, thus suggesting an activation of autophagic pathways. Similar results were shown for Microsporum gypseum treated with resorcinol derivatives [35].
Furthermore, the present study reports for the first time the anti-inflammatory potential of B. rigidum subsp. paniculatum. Nevertheless, other studies have pointed out promising effects for other Bupleurum species. For example, the essential oil from B. fruticosum, rich in αand β-pinene, thymol and carvacrol, was able to decrease the edema in a carrageenan-induced edema model [23] and that of B. frutiscens, characterized by α-pinene and β-caryophyllene, also showed effects in paw edema induced by carrageenan and prostaglandin (PG) E1 [24]. Moreover, B. gibraltaricum essential oil with high amounts of ∆-3-carene and α-pinene, showed anti-inflammatory effects in both carrageenan-induced paw edema and granuloma-induced inflammation [51,52]. Additionally, the production of PGE2 and the activity of lipoxygenase was decreased in the presence of B. marginatum essential oil [22]. Overall, these results corroborate the anti-inflammatory effects observed in the present study as all species present similar major compounds to those found in B. rigidum subsp. paniculatum, namely pinene isomers. Accordingly, several reports have shown the anti-inflammatory effect of isolated major compounds including αand βpinene as well as limonene [53][54][55]. In addition, minor compounds, such as 1,8-cineole [56], camphor [56,57] and γ-terpinene [58] have also been assessed in a plethora of assays, thus suggesting that these compounds may contribute to B. rigidum subsp. paniculatum anti-inflammatory potential.

Conclusions
Overall, the present study provides evidence, for the first time, of the antifungal and anti-inflammatory potential of B. rigidum subsp. paniculatum essential oil from Portugal. The oil showed high amounts of monoterpene hydrocarbons (94.4-96.6%), namely α-pinene (29.0-36.0%), β-pinene (26.1-30.7%) and limonene (10.5-13.5%) and demonstrated a broad fungicidal effect, particularly against Cryptococcus neoformans and Trichophyton rubrum. Moreover, a significant germ tube inhibition as well as a disruption of mature biofilms for C. albicans was pointed out and for both C. albicans and T. rubrum significant ultrastructural alterations were observed. Finally, B. rigidum subsp. paniculatum essential oil also evidenced anti-inflammatory activity, without affecting macrophages viability, further endorsing its exploitation as a natural source of bioactive compounds for the development of drugs with concomitant antifungal and anti-inflammatory effects. Nevertheless, further in vivo studies are needed to validate the efficacy of the essential oil. Moreover, the clinical relevance of our results depends on the availability of the essential oil on target organs. Therefore, pharmacokinetic studies are essential to link in vitro effects to in vivo efficacy.

Essential Oil Isolation and Chemical Analysis
The essential oils were obtained by hydrodistillation (3 h) using a Clevenger-type apparatus as described in the European Pharmacopoeia [59]. Their chemical characterization was carried out using both gas chromatography (GC) and gas chromatography coupled to mass spectrometry (GC-MS).
A Hewlett-Packard 6890 (Agilent Technologies, Palo Alto, CA, USA) chromatograph with a single injector, two flame ionization detectors and a HP GC ChemStation Rev. A.05.04 data handling system was used for GC analysis. Two columns: polydimethylsiloxane (SPB-1) and polyethylene glycol (SupelcoWax-10) with 30 m × 0.20 mm and a film thickness 0.20 µm were used and GC parameters were as follows: oven temperature: 70-220 Retention indices and mass spectra were used to identify the compounds with the first being compared to samples from the Laboratory of Pharmacognosy database of the Faculty of Pharmacy, University of Coimbra and the later with that of reference spectra from a home-made library or from literature data. GC peak area was used to calculate relative amounts of the identified compounds. The fungal isolates were identified by standard microbiological methods and stored at −80 • C on Sabouraud broth with 20% glycerol, being inoculated on Sabouraud dextrose agar (SDA) to ensure optimal growth and purity, prior to each assay.

Fungal Growth
To assess the effect of the essential oil on fungal growth, two parameters were assessed: the minimal inhibitory concentration (MIC) and the minimal lethal concentration (MLC). A broth macrodilution method was used according to the Clinical and Laboratory Standards Institute (CLSI) reference protocols M27-A3 and M38-A2, for yeasts and filamentous fungi, respectively. Serial dilutions of the oil (up to 0.036 µg/mL) were tested as well as a negative control (medium without fungi) and a positive control (inoculated medium with DMSO 1%).

Fungal Ultrastructure
C. albicans and T. rubrum were selected to perform the transmission electron microscopy studies. Cultures untreated (DMSO 1%) or treated with the essential oil (MIC) were used. Fungi cultures were collected and centrifuged at 3000 rpm for 5 min. Pellet cells were fixed using 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.2) for 2 h and post-fixed in 1% osmium tetroxide during 1 h. Following rinsing, samples were embedded in 2% molten agar, dehydrated in ethanol (30-100%), and impregnated and included in epoxy resin (Sigma Aldrich, St. Louis, MO, USA). Ultrathin sections (70 nm) were obtained and stained with uranyl acetate 2%, for 5 min followed by lead citrate 0.2% for 5 min. Observations were carried out on a FEI Tecnai G2 Spirit Bio Twin at 100 kV.

Candida albicans Germ Tube Formation
To assess the effect of the oil on the yeast-hyphae transition, cell suspensions of C. albicans ATCC 10231, from overnight cultures on SDA, were prepared in NYP medium [N-acetylglucosamine (10-3 mol/L), Yeast Nitrogen Base (3.35 g/L), proline (10-3 mol/L), NaCl (4.5 g/L), and pH 6.7 ± 0.1] [60]. The suspensions were adjusted to a density of 1.0 ± 0.2 × 10 6 CFU/mL and then distributed into glass test tubes (990 µL). Then, subinhibitory concentrations of the essential oil or fluconazole (10 µL) were added and incubated without agitation for 3 h, at 37 • C. The number of germ tubes (germinating protuberance at least as long as the diameter of the blastopore) were recorded using a light microscope (40×) and DMSO 1% (v/v) was used as a positive control.

Candida albicans Biofilm Disruption
Serial dilutions of the essential oil (1125-72 µg/mL) and fluconazole (200-1 µg/mL) were used to assess their effect on mature biofilms, as reported elsewhere [61]. C. albicans density was adjusted to 1 × 10 6 CFU/mL and 100 µL of this suspension added to 96-well microtiter plates and incubated for 24 h at 37 • C for biofilm adhesion and formation, followed by 24 h at 37 • C in the presence or absence of the essential oils.
Biofilm viability was determined using the XTT/menadione metabolic assay. Briefly, menadione (10 mM in acetone) was added to a XTT solution (1 mg/mL in PBS) to a final concentration of 4 µM. A volume of 100 µL of this solution was added to each well and incubated for 2 h at 37 • C in the dark. The absorbance was measured at 490 nm and cell viability expressed as [(AbsT/AbsC) × 100], where AbsT represents the absorbance of the treated cells and AbsC the absorbance of control cells (without essential oil or fluconazole).
Biofilm biomass was assessed using the crystal violet assay. Following the treatment period, performed as referred above, the culture medium was removed and cells were fixed with methanol 99% for 15 min. The supernatant was discarded, and the wells air dried. Then, 100 µL of crystal violet (CV) solution (0.02%) was added to each well and left to stain the biofilm for 15 min. After CV removal, the wells were washed twice with sterile water to remove excessive reagent. Then, 150 µL of acetic acid 33% were added to release the stain from the cells and the supernatant was transferred to a 96-well microtiter plate. The absorbance was then read at 620 nm using a microtiter plate reader. Biomass decrease was determined as [(AbsT/AbsC) × 100], where AbsT represents the absorbance of each treated well and AbsC the absorbance in the control wells. Negative and positive controls comprising biofilm-free and oil-free wells, respectively, were also included.

Cell Culture
The mouse macrophage cell line, Raw 264.7 (ATCC number: TIB-71) was kindly supplied by Dr. Otilia Vieira (CEDOC, Faculdade de Ciências Médicas, Universidade Nova de Lisboa). The cells were cultured on endotoxin-free Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% (v/v) non-inactivated fetal bovine serum, 3.02 g/L sodium bicarbonate, 100 µg/mL streptomycin and 100 U/mL penicillin at 37 • C in a humidified atmosphere with 5% CO 2 . Morphological appearance of macrophages was microscopically monitored during the assays.

Assessment of Nitric Oxide (NO) Production
To evaluate the anti-inflammatory potential of the oils, the in vitro model of lipopolysaccharide (LPS)-stimulated macrophages was used. The cells (0.6 × 10 6 cells/well) were cultured in a 48-well microplate and left to stabilize for 12 h. The cells were then incubated for 24 h in the culture medium alone (control) or with different concentrations of the essential oils (144-576 µg/mL).The production of nitric oxide (NO) was measured by the accumulation of nitrites in the culture medium, using the colorimetric Griess reaction. Briefly, 170 µL of culture supernatants were added to an equal volume of Griess reagent [0.1% (w/v) N-(1-naphthyl)-ethylenediamine dihydrochloride and 1% (w/v) sulphanilamide containing 5% (w/v) H 3 PO 4 ] and maintained for 30 min, in the dark. Absorbance was measured using an ELISA automatic microplate reader (SLT, Austria) at 550 nm and the nitrite concentration determined from a regression analysis prepared with serial dilutions of sodium nitrite.

Assessment of Cell Viability
Cell respiration, an indicator of cell viability, was performed using a colorimetric assay with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT). The cells were cultured and treated as referred above. After cells treatments, 43 µL of MTT solution (5 mg/mL in phosphate buffered saline) were added to each well and the microplates were further incubated at 37 • C for 15 min, in a humidified atmosphere with 5% CO 2 . Supernatants were centrifuged (1000× g during 5 min) to recover viable cells. To dissolve formazan crystals formed in adherent cells in the microplates, 300 µL of acidified isopropanol (0.04 M HCl in isopropanol) were added to each cell and recovered to the respective tube containing the pellet formed after centrifugation. Quantification of formazan was performed using an ELISA automatic microplate reader (SLT, Salzburg, Austria) at 570 nm.

Statistical Analysis
The results presented as mean ± SD are from experiments performed at least in triplicate. The statistical test one-way ANOVA followed by Dunnett's Multiple Comparison Test were performed using GraphPad Prism, version 6 (GraphPad Software, San Diego, CA, USA).