In Vitro Antifungal Activity of Crude Hydro-Alcoholic Extract of Petiveria alliacea L on Clinical Candida Isolates

Background: The global burden of infections due to Candida and the emerging resistance to antifungals has led to the search for new therapeutic alternatives. The aim of the present work it was to evaluate the in vitro antifungal activity of crude hydro-alcoholic extract of Petiveria alliacea L (HAEPAL) versus fluconazole against Candida isolates. Methods: In vitro antifungal activity was evaluated by broth microdilution method in front of 125 Candida isolates (60 C. albicans). Inoculum of 1.5×106 CFU/mL in sterile saline solution were incubated with five dilutions of the extract (128, 64, 32, 16 and 8 μg/mL). The minimum inhibitory concentration (MIC) was defined as the lowest concentration of HAEPAL showing ≤ 50% growth compared with the extract-free growth control estimated by counting the CFU/mL. Fluconazole susceptibility was assessed by ATBTM Fungus 3 and E-test according to manufacturer’s instructions. Results: HAEPAL showed higher antifungal activity compared to fluconazole. Only four isolates (one C. albicans, one C. glabrata and two C. krusei) exhibited high MICs (≥ 64 μg/mL) compared to 34 (19 C. albicans and 15 Candida non-albicans) which showed resistance to fluconazole. Conclusion: These results show the antifungal potentiality of HAEPAL, which could become a potential alternative for Candida treatment.


Introduction
The incidence of fungal infections has drastically increased over the past three decades and has become a major cause of morbidity and mortality [1]. Antifungal therapy failure depends on multiple factors including microbiological resistance. The use of fluconazole as standard prophylaxis in aids patients and the hematopoietic cell transplantation setting was simultaneously accompanied by increased acquired and innate resistance to antifungal drugs. Azole resistance was first noted in Candida species in patients with acquired immunodeficiency syndrome (of 348 isolates tested against fluconazole, and 33% were found to be resistant). With the advent of new azole and echinocandin agents, we have seen the emergence of more azole-resistant and echinocandin-resistant fungi [1,2]. This has stimulated a search for safe and more potent antifungal products. In this sense the plant kingdom has been an important source of large numbers of natural drugs over the years [1,3].
There are approximately 8000 species of Cuban medicinal plants and nearly half are endemic species. This is a rich source for the development of potential therapeutic agents [4]. Petiveria alliacea L, popularly known as anamú, is a native plant from Latin America that belongs to the family Phytolaccaceae. Because fresh leafs contain the similar chemical compounds also present in the rest of the plant, it is considered the richest vegetative organ responsible for different pharmacological activities [5].
There are few reports of antifungal activity of P. alliacea against fungal species [6][7][8]. However, literature on the evaluation of various preparations from this plant against yeasts does not provide conclusive results due to the scarce number of studied isolates [9,10]. Given the presence of substances with proven anti-fungal activity in this plant, this work aimed to determine the in vitro activity of a crude hydroalcoholic extract of P. alliacea against Cuban clinical isolates of Candida and compared it with that of fluconazole, by far the most common antifungal used in clinical practice.
Antifungal activity assay: A broth microdilution method based on M27A3 document of the Clinical and Laboratory Standards Institute (CLSI) [12] was used to determine yeasts susceptibility. Recommended CLSI reference strains Candida krusei ATCC 6258 and Candida parapsilosis ATCC 22019 were included. P. alliacea hydro-alcoholic extract (HAEPAL) (200 g/L) was prepared by the method described in Public Health Branch Standard 311 (NRSP 311; MINSAP, 1992). Briefly, plant leaves were kept at 40°C until dried and powdered using an appropriate mill; the material was macerated with ethanol 70% (1:5) and stirred 24 h at room temperature; finally it was stored in amber flasks with hermetic caps at 4°C until use. The extract was sterilized by filtration (0.45 and 0.22 μm membranes) and diluted in Sabouraud dextrose broth (SDB) to obtain 256, 128, 64, 32 and 16 μg/mL. Each solution was dispensed (100μL/ well) in lines C to G from column 2 to 11. Row B column 2 to 11 was coated with SDB + ethanol 10%. Once the inoculum was added, HAEPAL concentration was diluted to 8, 16, 32, 64 and 128μg/mL (within the range of concentration as set by CLSI for fluconazole) and ethanol decreased to 5% as final concentration.
Candida isolates were transferred to plates with Sabouraud dextrose agar (SDA) and incubated 24-48 h at 37°C. Each culture was suspended in sterile saline solution (SSS). Turbidity was adjusted to McFarland 0.5 scale, equivalent to 1.5×10 6 colony forming units per milliliter (CFU/mL). Suspensions were dispensed (100 μL/well) in column 3 to 11. A reference strain was randomly incorporated in column 2 of each plate as internal control assay. Rows A and H and columns 1 and 12 were coated with SSS in order to reduce the evaporation effects during incubation at 37°C for 24 h.
For reading and interpretation the content of each well was vigorously pipetted and diluted 1:1000 and 1:2000 in SSS. Each dilution (10 μL) was inoculated on SDA and incubated at 37°C for 48-72 h. The CFU/mL was determined for every strain vs each dilution of the extract using a magnifying viewer (Gallenkamp, England) ( Figure 1). Minimum inhibitory concentration (MIC) was defined as the lowest extract concentration able to inhibit at least 50% of growth compared with culture medium without HAEPAL. Each procedure was performed three times. For fluconazole susceptibility ATB TM Fungus 3 (bioMérieux, France) was used. When susceptibility interpretation was inconclusive Etest ® strips impregnated with fluconazole (0.016 to 256 μg/mL) (AB BIODISK, Sweden) were used as an alternative method. Both methods were carried out following manufacturer's instructions.
Statistical analysis: Data analysis was performed with 6.07 XLSTAT Excel. Frequencies and percentages were calculated to compare susceptibility values among isolates according to the species. In addition, the MIC ranges and geometric mean (GM) values were determined. The MIC50 was calculated considering the cumulative percentages of isolates inhibited at different concentrations of HAEPAL and fluconazole respectively.

Results and Discussion
A main concern in medical mycology is the emergence of resistant isolates. This is especially relevant when studying azoles, the most used among the antifungal drugs [13,14]. Fluconazole is one of the widest used for treatment and prophylaxis of fungal infections particularly candidiasis [13,15]. This situation has stimulated a search for new alternatives from natural products of which the herbarium is highly attractive [1,16]. P. alliacea, flourishes in Cuba and has several traditional applications such as anti-inflammatory, analgesic, antirheumatic, antitumor, hypoglycemic, antipyretic, antispasmodic, abortifacient, diuretic, and sedative among others. It has been also proven its antimicrobial activity against virus, bacteria, protozoa and fungi [17][18][19][20][21][22][23]. Most of these activities have been shown for hydroalcoholic extract from root, stem, and/or leaves maybe because the main active substance are not soluble in water, saline solution or phosphate buffer. Nevertheless, the inclusion of ethanol on the growth control at the highest concentration present on the tested dilution of HAEPAL, allows conclude that the observed antifungal activity was due the plant extract itself.
In order to avoid contaminations during the in vitro antifungal activity of the extract, it was filtrated by using two different porosities.
Although the chemical composition of HEAPAL was not monitored during this procedure we assume this did not suffer major change since 0.22 µm (the smallest pore size used) is only able to retain particles larger than such diameter what is far outsized than chemical compounds.
Previous work showed that broth dilution method provides better results than agar diffusion for the evaluated extract [9]. Because there are no standards for determining the susceptibility to natural products, a method was developed based on CLSI standards [12]. The adapted method allowed the evaluation of HAEPAL antifungal activity even though it was used SDB where the extract was previously proved [9] in stated of RPMI media. Table 1 shows the ranges, GM and MIC 50 of HAEPAL and fluconazole against the studied Candida species.
In general, the MICs were higher for HAEPAL. The analysis according to the species indicates that C. albicans was the most represented among fluconazole-resistant strains (13 isolates) followed by C. tropicalis (3 isolates) and C. parapsilosis (2 isolates) while only 2 C. krusei, 1 C. glabrata, and 1 C. albicans required concentrations above 64 μg/mL of HAEPAL to inhibit their growth (data not shown). Fluconazole displayed broad ranges, especially against C. albicans and C. tropicalis whereas GM of the extract also had wider ranges against those species and C. glabrata.  Table 1: Range, geometric mean (GM) and MIC 50 (μg/mL) values of the studied P. alliacea L extract and fluconazole against Candida isolates. The studied formulation did not undergo any fractionating and purification process, therefore all the constituents, with and without antifungal activity, were present in the solution. This implied that the concentrations of active ingredients with possible antifungal activity were "diluted" in the extract and for that reason higher concentrations of the crude extract were necessary which could explains that HAEPAL GM ranges and MIC values were higher than those of fluconazole. Despite this, over 95% of the isolates were inhibited at HAEPAL MIC ≤ 128 μg/mL. Currently there are no established criteria to define the antifungal activity of natural products [3]. However there are several studies that report the association between this activity and MIC values. Holetz et al. suggested that MICs of ≤ 100 μg/mL correspond with good activity, 100 to 500 μg/mL with moderate and >500 μg/mL with weak activity [18]. Duarte et al. proposed that MIC values of <500 μg/mL represent a strong action, 600 -1500 μg/mL moderate and >1500 μg/mL slight action [19]. Webster et al. suggest that plant extracts with MICs of ≤ 1000 μg/mL encompass a strong antifungal potency [20]. Regardless of the various criteria for categorizing the antifungal activity of a natural extract, HAEPAL exhibited relatively low effective MIC values.

Species
Previous studies have shown that P. alliacea is active against certain bacteria, protozoa and viruses [8,10,[21][22][23]. While some authors were unable to detect antifungal effects of this plant after studying hydroalcoholic extracts against C. albicans strains [10,22] others demonstrated activity against Colletotrichum gloeosporides, Cladosporium cladosporiodes, C. sphaerospermum, dermatophytes, six species of Candida and other yeast (Saccharomyces, Rhodotorula, Trichosporon and Cryptococcus) [6,7,9]. These discrepancies could be explained by differences in the method for extract preparation, the solvent, the parts of the plant used, the technique applied in susceptibility studies, and the amount of the active ingredients in preparations. The latter condition can vary depending on the region and weather conditions where the plant is harvested [24].
Former phytochemical screening showed that P. alliacea contains substances with proven antifungal activity such as triterpenes, derived alkaloids, flavonoids and saponins [25]. These metabolites probably act by inhibiting the biosynthesis of ergosterol or other sterols present in the fungal membrane, damaging and altering its permeability resulting in loss of essential intracellular elements [26,27]. Especially the saponins are able to form pores in the lipid membrane, to increase cell permeability and inhibit biofilm formation, rendering fungal cells more susceptible to osmotic stress [28,29]. Another potential mechanism by which chemical components could have antifungal activity is the inhibition of triglycerides and phospholipids biosynthesis as well as oxidative and peroxidative enzymatic activity; this would lead to the accumulation of toxic concentrations of hydrogen peroxide that contribute to the deterioration of subcellular organelles and cell necrosis [27]. However, it is difficult to define a specific action mechanism since the results of the studies done to date are not conclusive. The high concentration of these compounds in leaves, which were used in the preparation of HAEPAL, could explain the obtained results in this study.
In conclusion, the assessment of the P. alliacea L extract against 125 clinical Candida isolates showed a prominent antifungal effect.
Comparison of its activity with that of fluconazole, a major antifungal used in candidiasis management, demonstrated the therapeutic potential of this plant, which could be useful for treatment. Future studies of the extract fractions could help to define the compound(s) responsible for the antifungal activity. Other methods such as lyophylization should also be evaluated for the active compounds obtained from the plant.