Antifungal and Antibacterial Metabolites from a French Poplar Type Propolis

During this study, the in vitro antifungal and antibacterial activities of different extracts (aqueous and organic) obtained from a French propolis batch were evaluated. Antifungal activity was evaluated by broth microdilution on three pathogenic strains: Candida albicans, C. glabrata, and Aspergillus fumigatus. Antibacterial activity was assayed using agar dilution method on 36 Gram-negative and Gram-positive strains including Staphylococcus aureus. Organic extracts showed a significant antifungal activity against C. albicans and C. glabrata (MIC80 between 16 and 31 µg/mL) but only a weak activity towards A. fumigatus (MIC80 = 250 µg/mL). DCM based extracts exhibited a selective Gram-positive antibacterial activity, especially against S. aureus (SA) and several of its methicillin-resistant (MRSA) and methicillin-susceptible (MSSA) strains (MIC100 30–97 µg/mL). A new and active derivative of catechin was also identified whereas a synergistic antimicrobial effect was noticed during this study.


Introduction
Propolis is a resinous natural substance collected by honeybees from buds and exudates of various trees and plants, mixed with beeswax and salivary enzymes. Bees generally use it as a sealer, to smooth out the internal walls of the hive, as well as a protective barrier against intruders. Propolis has been used in folk medicine since ancient times due to its pharmacological potential associated with antioxidant [1][2][3], antifungal [4,5], antibacterial [6][7][8], and anti-inflammatory [9] properties.
Propolis is generally composed of 50% of resin and balm (including polyphenolic compounds), 30% of wax and fatty acids, 10% of essential oils, 5% of pollen, and 5% of various organic and inorganic compounds. However, the composition of propolis deeply depends on the vegetation at the site of collection [10]. Indeed, propolis from temperate climatic zones, like in Europe, North America, or nontropical regions of Asia, mainly originates from the bud exudates of Populus species (Salicaceae) and consequently is rich in flavonoids and phenolic acids and their esters [11]; however tropical propolis, originating from regions where neither poplars nor birches grow, is rich in prenylated derivatives of p-coumaric acids, benzophenones, or terpenoids [12,13].
The antifungal, antibacterial properties and chemical composition of propolis from many countries all over the world have been widely studied [6,8,[14][15][16][17][18][19][20] but few reports were already given for European propolis [21,22]. In 1990, Grange and Davey [23] highlighted for the first time the bactericidal activity of a French propolis against Grampositive strains whereas later on, in 2000, Hegazi et al. [22] could associate this antibacterial activity with the presence of benzyl caffeate, pinocembrin, and p-coumaric acid.
During a previous study, we have evaluated the antioxidant and anti-AGEs activities of different solvents extracts [water; 95% EtOH; 70% EtOH; MeOH; dichloromethane (DCM) and DCM/MeOH/H 2 O (31/19/4)] obtained from a French propolis batch and identified their active constituents 2 Evidence-Based Complementary and Alternative Medicine [24]. Here we have investigated the in vitro antifungal and antibacterial activities of these extracts. The antifungal activity was studied on three fungal strains (two yeasts, Candida albicans and C. glabrata, and one filamentous fungus, Aspergillus fumigatus). 36 strains of Gram-positive (including Staphylococcus aureus) and Gram-negative (including Escherichia coli) bacteria were used for the antibacterial assays. During this study, a new secondary metabolite was isolated, namely, 8-[(E)-phenylprop-2-en-1-one]-5-methoxy-(±)-catechin.

Instrumentation.
Optical rotation was measured on a JASCO P-2000 polarimeter. IR spectra were recorded on a Bruker Vertex 70 spectrophotometer. NMR spectra (1D and 2D) were recorded on a Bruker Avance spectrometer at 500 MHz for 1 H and 125 MHz for 13 C. MS analyses were performed on an ESI/APCI Ion Trap Esquire 3000+ from Bruker. UV absorbances were obtained from a Tecan Infinite M200 microplate spectrophotometer.

Propolis Samples.
In order to analyze a typical French batch, that is, exhibiting an average chemical composition, a mixture of samples (10 g of each), collected in apiaries originating from different regions of France, was used for this study. These samples were provided by "Ballot-Flurin Apiculteurs, " a company specialized in organic beekeeping. Indeed, even collected in the same geographical region, propolis profiles may differ between apiaries and even inside the same apiary from one hive to another one [25]. Keeping in mind any potential economic development, it then appeared more appropriate to study an industrial end-product, that is, a mixture, exhibiting an average chemical composition associated with an average antimicrobial activity, rather than a specific sample. Therefore, 24 batches of propolis collected over two years (2010 and 2011) from different places in France (cf. supporting information 1; see Supplementary Material available online at http://dx.doi.org/10.1155/2015/319240) were homogeneously mixed to undergo this study.

Extractions.
The extraction processes have been already described [24]. Briefly, the propolis batch was homogeneously pulverized in the presence of liquid nitrogen and divided into 1 g samples. Four different extractions were then carried out on 1 g samples with water (E1), 95% EtOH (E2), 70% EtOH (E3), and MeOH (E4). Then, two extractions, preceded by a cyclohexane wax elimination, were independently performed on 1 g samples with DCM (E5) and a mixture of DCM, MeOH, and H 2 O (31/19/4) (E6). For E1, a decoction of 1 g of propolis powder was boiled in 20 mL H 2 O at 100 ∘ C for 15 min. After cooling, the solidified wax and the residue were removed by filtration, and the filtrate was evaporated to dryness. For other solvents, 1 g of propolis powder (or residue obtained from a previous extraction) was macerated in 3 × 20 mL of solvent. After stirring for 3 × 2 h at room temperature, the mixture was filtered. The filtrates were gathered and evaporated under vacuum. Extraction yields (dried extract/100 g) were as follows: E1 7%; E2 68%; E3 65%; E4 68%; E5 50%; and E6 59%.
2.5. Antifungal Activity. Antifungal activity was assayed on human pathogenic fungi, including two common yeasts (Candida albicans ATCC 66396 and C. glabrata LMA 90-1085) and an opportunistic mould (Aspergillus fumigatus CBS 11326). The strains were obtained from the Parasitology and Mycology Laboratory at the University Hospital Center of Angers, France. Microorganisms were cultivated at 37 ∘ C on yeast extract-peptone-dextrose-agar (YPDA) containing 0.5 g/L chloramphenicol for two (C. albicans and C. glabrata) or three (A. fumigatus) days. Tests were performed according to a procedure described by Alomar et al. [27], following the guidelines of the approved reference method of the National Committee for Clinical Laboratory Standards (NCCLS) for yeasts [28] and filamentous fungi [29]. Briefly, the yeast suspensions were prepared in RPMI-1640 culture medium and adjusted spectrophotometrically at 630 nm to reach a final concentration of ca. 0.5 × 10 3 to 2.5 × 10 3 cells/mL. The tests were performed using sterile 96 flat shaped well microtiter plates. Serial twofold sample dilutions were made in DMSO. Sample solutions were dispensed at a volume of 5 L in triplicate into the wells to obtain final concentrations from 250 to 1.95 g/mL. After 48 h at 37 ∘ C for C. albicans and C. glabrata and 72 h for A. fumigatus, the spectrophotometric MIC endpoint was calculated from the turbidimetric data as the lowest sample concentration causing a growth inhibition equal to or greater than 80% of the control (MIC 80 ). Amphotericin B was used as a positive control.
2.6. Antibacterial Activity. Antibacterial activity was evaluated on 36 human pathogenic bacterial strains collected by the Laboratory of Bacteriology at the University Hospital Center of Angers, France: seven strains of Acinetobacter baumannii (RCH, SAN008, 12, AYE, CIP7034, 107292, and 5377), two of Escherichia coli (ATCC25922 and a clinical isolate), three of Pseudomonas aeruginosa (ATCC27853 and two clinical isolates), and 4 clinical isolates of Enterobacter cloacae, E. aerogenes, Klebsiella oxytoca, and Salmonella enteritidis (phage type 4) for Gram-negative bacteria; thirteen strains of Staphylococcus aureus (ATCC25923, six methicillinsusceptible clinical isolates, six methicillin-resistant clinical isolates), two clinical isolates of S. epidermidis (methiS and methiR), three clinical isolates of Enterococcus faecalis and one of E. faecium, and one clinical isolate of Corynebacterium striatum for Gram-positive bacteria. Tests were performed using the methodology described by Alomar et al. [30]. Briefly, a stock solution of each sample was prepared in triplicate at 20 mg/mL in DMSO under sterile conditions. Serial dilutions were prepared (sample concentrations: 10, 20, 30, etc., to 100 g/mL) and 0.1 mL of each dilution was added to 19.9 mL of Mueller-Hinton agar (Merck, Germany) and transferred to Petri plates. Bacterial strains (2 × 10 4 CFU/mL) were suspended in sterile NaCl aqueous solution (0.15 M) and inoculated on the different Petri plates using the multipoint inoculator (AQS, England). After 24 h of incubation at 37 ∘ C, the minimum inhibitory concentration (MIC 100 , g/mL) against bacterial strains was defined as the lowest concentration of each sample that inhibited visible growth. A blank was made inoculating the strains on Mueller-Hinton agar without any extract or compound. Oxacillin was used to distinguish the methicillin-resistant from the susceptible staphylococcal strains.

HPLC-DAD and HPLC-MS Procedures.
Dry extracts were dissolved in MeOH (5 mg/mL for the aqueous extract and 10 mg/mL for the organic solvents ones) and centrifuged at 13000 rpm for 10 min prior to injection (10 L) into the HPLC system. Analytical HPLC was run on a 2695 Waters separation module equipped with a diode array detector 2996 Waters. Separation was achieved on a LiChrospher column 100 RP-18 (125 × 4 mm i.d., 5 m) protected with a LiChrocart 4-4 guard cartridge (4 × 4 mm i.d.) at a flow rate of 1 mL/min. The mobile phase consisted of 0.1% formic acid in water (solvent A) and MeOH (solvent B) and the separation was performed using the linear gradient: 25-100% B in 40 min. UV detection was achieved at two wavelengths: 254 and 280 nm.
The mass analyses were performed with an ESI interface coupled to an ion trap mass analyzer in both positive and negative modes, with the following conditions: collision gas, He; collision energy amplitude, 1.3 V; nebulizer and drying gas, N 2 , 7 L/min; pressure of nebulizer gas, 30 psi; dry temperature, 340 ∘ C; flow rate, 1.0 mL/min; solvent split ratio 1 : 9; scan range, m/z 100-1000.

Antifungal and Antibacterial Activities.
Results showed that Gram-negative bacteria were not susceptible to E1-6 at this concentration. In contrast, organic solvents extracts were active on several Gram-positive bacteria such as Corynebacterium striatum (sometimes involved in pleuropulmonary infections) (E2-5) and especially Staphylococcus aureus, including for the latter several methicillinresistant (MRSA) and methicillin-susceptible (MSSA) clinical isolates (E5-6). Sometimes called "golden staph, " S. aureus is the most pathogenic species of Staphylococcus genus. It might cause food poisoning, skin infections, abscesses, and diseases like pneumonia, meningitis, and sepsis. S. aureus is additionally one of the major causes of hospital-acquired infections, and the treatment of some multiresistant strains has become quite problematic. Among them, MRSA appears in France as one of the most commonly multiresistant strains encountered in hospitals.
MIC 100 of E1-6 were determined on the 6 susceptible Gram-positive strains as well as on 8 other MRSA and MSSA strains. Results are given in Table 3.
E1 did not show any interesting activity on the 14 studied strains (MIC 100 > 100 g/mL). E2-6 showed interesting activities against Corynebacterium striatum with MIC 100 ranging from 63 to 90 g/mL. E5 and E6 exhibited the best antibacterial activities against the Staphylococcus strains with MIC 100 up to 57 and 30 g/mL, respectively. Among the alcoholic extracts, only E4 showed a moderate activity (MIC 100 90 g/mL) against S. aureus and one MRSA whereas E2 and E3 appeared as inactive. These overall activities therefore appeared to be better than those reported by Grange and Davey for the antibacterial activity of a French propolis on S. aureus and MRSA (MIC 100 188-375 g/mL) [23]. Our global antibacterial activity against MRSA and MSSA could be compared with those reported for propolis collected in Solomon Islands, exhibiting MIC 100 between  64 and 128 g/mL [6]. Similarly E4 was more active than a methanolic propolis extract from Jordan (585 g/mL against S. aureus and 4700 g/mL against MRSA) [20]. These results suggested that antifungal and antibacterial activities of propolis extracts could be related to their flavonoids contents [24]. Indeed, whereas E1-6 exhibited high total polyphenol contents (239-281 mg GAE/g), only those showing both high flavone/flavonol (FF) and flavanone/dihydroflavonol (FD) contents (i.e., E5-6) were active on the studied strains. In addition the higher the cumulative contents FF+FD were, the stronger the antibacterial activity was, as shown with E5 (254 mg/g) and E6 (236 mg/g) > E2-4 (220-228 mg/g). These results are in agreement with those reported by Velazquez et al. [15] for different Mexican propolis collected in Sonora State where EEP from the areas of Ures (410 mg/g), Caborca (332 mg/g), and Pueblo de Alamos (209 mg/g) showed MIC 100 against S. aureus of 100, 200, and >400 g/mL, respectively. Figure 1 shows the HPLC chromatogram of the DCM extract E5. 48 compounds were identified by comparison with the literature data (UV/MS) and pure standards or, when needed, through 1 H and 13 C (1D and 2D) NMR analysis after compound isolation.
Compound 40 was obtained as a yellow amorphous solid (0.6 g/g of DCM extract). The molecular formula was Evidence-Based Complementary and Alternative Medicine 5 Enterobacter aerogenes (0705A0867) Corynebacterium striatum (56) − Staphylococcus aureus (ATCC25923)   The trans-olefinic proton H at H 8.06 was also correlated with the phenyl quaternary carbon at C 136.2 (C 1 ). This correlation implied the presence of a (2E)-4-phenylprop-2-en-1-one moiety. A correlation between the methoxyle protons ( H 3.92) and the carbon at C 165.1 (C 5 ) proved that the OCH 3 was attached to C 5 . The NOESY spectrum showed that this methoxyle was spatially close to the proton at H 6.15 (H 6 ), whereas a long-range COSY indicated a correlation between H 6 and one of the hydroxyl groups at H 14.49 (OH 7 ). Therefore a (2E)-4-phenylprop-2-en-1-one moiety was located at C 8 ( C 105.9). Finally, it appeared that the aromatic ring B was substituted at C 3 and C 4 by two hydroxyl groups (NMR spectra cf. supporting information 2). 1 H and 13 C NMR data together with 2D NMR correlations for 40 are summarized in Table 4 and Figure 3.      already isolated a similar compound, only differing from 40 by a 1 ,3 ,5 -trisubstituted aromatic ring B, from a Chinese propolis [26] (Figure 4).

Antifungal and Antibacterial Activities of 40.
The new flavan-3-ol 40 did not show any antifungal activity on the three strains studied (Table 5). However, though active neither on S. aureus nor on MSSA, its MIC 100 on MRSA numbers 23 and 24 were lower or equal to 10 g/mL (close to oxacillin: ≥4 g/mL).

Major Compounds Activities.
Antifungal and antibacterial activities were then individually evaluated for the five major compounds identified in E2-6, namely, pinobanksin-3-acetate (28), pinocembrin (25), chrysin (32), galangin (34), and prenyl caffeate (29) [24]. Their MIC 80 towards C. albicans, C. glabrata, and A. fumigatus as well as their MIC 100 towards S. aureus, MRSA, and MSSA are given in Table 6. Pinobanksin-3-acetate (28), chrysin (32), and galangin (34) appeared as inactive. Pinocembrin (25) showed a moderate activity towards Candida albicans, C. glabrata (MIC 80 62-125 g/mL), and S. aureus (MIC 100 100 g/mL). Overall prenyl caffeate (29) exhibited the best activities (MIC 100 up to 16 g/mL against C. glabrata and up to 63 g/mL against S. aureus and MRSA). Even so it appeared that these compounds were not individually as active as it could be expected from E5-6 results (MIC 100 30-97 g/mL). As far as S. aureus and MRSA are concerned, this kind of synergistic effects was recently pointed out by Darwish et al. [20] who evaluated the antibacterial activities of pinobanksin-3-acetate, pinocembrin, and chrysin isolated from a Jordanian propolis. Therefore these results are also in agreement with Kujumgiev et al. stating that, in spite of a great chemodiversity, no specific compounds can be associated with the antimicrobial activities of propolis extracts whereas, obviously, different flavonoid combinations are essential for these activities [7]. The antimicrobial of propolis extracts most probably involves a complex mechanism. It can be attributed to the synergistic effects of phenolic compounds such as cinnamic acid and ester derivatives, including caffeic acid and CAPE, as well as flavonoids including quercetin and naringenin [17,53,54]. Indeed, each of these compounds would be able to increase membrane permeability and inhibit bacterial mobility [54], thus contributing to the antimicrobial activity of propolis but also to its synergism with other antibiotics [53,55,56]. It is the reason why Stepanović et al. could notice the antibacterial and synergistic actions of propolis extracts with ampicillin, ceftriaxone, and doxycycline towards Staphylococcus aureus and with nystatin towards Candida albicans, stating that the bacterial resistance to antibiotics had no influence on the susceptibility to propolis extracts [57]. In vitro studies of synergism carried by Fernandes Jr. et al. also revealed synergistic effects of EEP with chloramphenicol, gentamicin, netilmicin, tetracycline, vancomycin, and clindamycin [58]. Therefore our findings are in total accordance with these results and, now that antibiotic resistance to bacteria has become a major public health concern [59], could bring valuable knowledge to develop new antimicrobial drugs for challenging S. aureus infections.

Conclusions
On the basis of these results, it may be concluded that organic solvents extracts of a French poplar type propolis are 8 Evidence-Based Complementary and Alternative Medicine associated with a good antifungal activity towards Candida albicans and C. glabrata, correlated with high flavonoid contents. However only DCM based extracts (E5-6) showed a significant antibacterial activity against both methicillinresistant and methicillin-susceptible Staphylococcus aureus strains. Unfortunately these extracts are not compatible with a pharmaceutical use because of their toxicity, whereas EtOH based extracts were not as active as expected. Therefore it would be interesting to develop some alternative extraction of propolis using a nontoxic solvent such as subcritical water. In addition, it should be noticed that, as an intrinsic polytherapy, propolis may also circumvent the development of drug resistance by bacteria [60].