In vitro Synergistic Antimicrobial Activity of Romanian Propolis and Antibiotics against Escherichia coli Isolated from Bovine Mastitis

The study was aimed to characterize the chemical composition and the antimicrobial activity of Romanian propolis ethanolic extracts (EEP) against antibiotic-sensitive and antibiotic-resistant E. coli strains isolated from bovine mastitis. The preliminary antimicrobial screening was performed by a disk diffusion method, followed by determination of minimum inhibitory concentrations (MIC) and minimum bactericidal concentrations (MBC) based on broth microdilution assay; further, the synergistic action of propolis with antimicrobial drugs was assessed by a disk diffusion method on agar containing subinhibitory concentrations of propolis. For the chemical characterisation of EEP, the flavonoids (flavones/flavonols, flavanones/dihydroflavonols) and total phenolics were evaluated by spectrophotometric methods. The phenolic compounds of these extracts were also determined using HPLC. The results indicated for Romanian propolis ethanolic extracts the typical poplar composition profile with flavonoids and phenolic acids as main biological active compounds, with chromatographic analysis data confirmed also spectrophotometrically. In addition, positively correlated with the chemical composition, a strong antimicrobial efficacy was exhibited towards E. coli strains, along with interesting synergistic interaction with antibiotics that can be further investigated to obtain propolis-based formulation with antibacterial properties. Subsequent in vitro and in vivo studies evaluating the safety and efficacy are intended to consider propolis in veterinary therapeutic protocols.


Introduction
Bovine mastitis is described as one of the most significant diseases affecting dairy herds, a pathology leading to considerable financial losses to the bovine industry due to the costs associated with diagnostics, treatment, redundancy (milk production losses, discarded milk) and animal culling (Halasa et al., 2007;Viguier et al., 2009;Down et al., 2013;Hegazi et al., 2014).Since the etiology of bovine mastitis involves bacterial pathogens, such as Escherichia coli, Staphylococcus aureus and Streptococcus uberis (Bradley, 2002;Hegazi et al., 2014), the use of antimicrobials (intramammary infusion and systemic administration) represents a fundamental part of the therapeutic protocols.Unfortunately, as one of the most important consequences of the intensive and/or inappropriate use of antimicrobials, elevated levels of antimicrobial resistance are currently reported in cases of mastitis etiological agents, with particular reference to E. coli (Bradley, 2002;Viguier et al., 2009;Hegazi et al., 2014).
Worldwide, the emergence of antibiotic-resistant microorganisms triggered the search for alternatives such as natural products with antimicrobial activity (Nweze and Eze, 2009;Hegazi et al., 2014).The stringent need of research into natural alternatives to antimicrobials is emphasized also by the current farm animal health and welfare policies (Ruegg, 2009).
The protocol for flavanones/dihydroflavonols evaluation was adapted after Popova et al. (2004): 1 ml of propolis ethanolic extract was added to 2 ml dinitrophenilhydrazine (1 g dinitrophenilhydrazine was mixed with 2 ml H2SO4 96% and diluted to 100 ml with methanol).The obtained solution was heated at 50 °C for 50 min.After cooling at room temperature, the solution was diluted to 10 ml with 10% KOH in methanol.An aliquot of 0.5 ml was transferred into a volumetric flask and the volume was adjusted to 25 ml methanol.Blank solution was prepared by replacing the amount of sample with methanol and carried out through all steps of the procedure.Standard solution of pinocembrin (1 mg/ml) was prepared for calibration curve (equation Y = 0.11034 * X -0.00416; r 2 = 0.99910).The absorbance was measured at 486 nm against blank.
Total phenolics were determined by Folin Ciocalteu method (Popova et al., 2004;Laslo, 2007).Briefly, 1 ml of EEP was added to 4 ml Folin-Ciocalteu reagent and 6 ml 20% sodium carbonate and the volume was adjusted to 50 ml with distilled water.The test solutions were kept in the dark for 2h and then the absorbance was measured at 760 nm against blank.Standard mixture of pinocembrin:galangin (2:1, w/w) was used for calibration curve (equation Y = 0.00709 * X -0.00109; r 2 = 0.99932).
Scientific data regarding Romanian propolis are documented in the literature, mostly about the chemical composition, quality criteria for standardization and certain biological properties (Laslo, 2007;Rindt et al., 2009aRindt et al., , 2009b;;Stan et al., 2011;Stan et al., 2012;Mărghitaş et al., 2013).Still, the growing interest in the veterinary apitherapy requires more research studies to substantiate the therapeutic use of propolis.Few studies investigated the synergistic effects between propolis and antibiotics (Stepanovic et al., 2003, Orsi et al., 2006) and no studies regarding the synergistic effect of Romanian propolis with other drugs have been done.
Therefore, taking into consideration the importance of new scientific research relating Romanian propolis, this study was aimed to investigate in vitro antibacterial activity of propolis ethanolic extracts tested alone and in combination with five antibiotics against E. coli strains isolated from bovine presenting clinical mastitis.

Extraction of active principles from propolis
Extraction of active principles from propolis (finely grounded) was performed by maceration with 70% ethanol (final concentration 1:100, w/v) with continuous stirring at 400 rpm for 24 h.The ethanolic extract was further diluted in optimal concentrations needed for quantitative analysis according to methods described by Popova et al. (2004).Extraction was realized in triplicate for all samples.All ethanolic extracts of propolis (EEP 1-5) were kept in the dark until anti E. coli activity analysis was performed.
Spectrophotometric method from Popova et al. ( 2004) was adapted to determine the flavones/flavonols: 1 ml of EEP was added to 0.5 ml of 5% aluminium chloride and adjusted to 25 ml enrofloxacin and florfenicol were selected for this assay and Escherichia coli ATCC® 25922 was also tested as a quality control organism.These microorganisms were cultivated on Mueller Hinton agar and 24h pure colonies were used to prepare 1.5 x 10 8 cfu/ml inoculum.

The evaluation of the propolis ethanolic extracts antimicrobial potential
The antimicrobial potential of each propolis ethanolic extract was evaluated using an agar diffusion protocol similar to the standard Kirby-Bauer method according to the Clinical and Laboratory Standards Institute (CLSI, 2013) guidelines.The bacterial strains prepared as inoculum were inoculated on Mueller Hinton agar plates; sterile filter paper discs impregnated with 20 μl EEP were applied and the inhibition zone diameters (IZD, mm) were recorded after 24 h incubation at 37 °C.The testing was performed in duplicate.Positive controls (antibiotics) and 70% ethanol (EEP solvent) were included for each determination.

Minimum inhibitory and minimum bactericidal concentrations determination
The minimum inhibitory (MIC) and minimum bactericidal (MBC) concentrations were established using a broth microdilution method, with twofold serial dilutions of each EEP, ranging from 4% to 0.125% (v/v), mixed with an equal volume of bacterial inoculum and incubated for 24h at 37 °C, when the MICs values were determined considering the lowest concentrations of EEP able to inhibit the visible growth of bacteria (no turbidity), when compared to the control.Afterwards, 10 μl of each EEP dilution were cultured on Mueller agar plates for 24h at 37 °C and the lowest concentrations associated with no visible bacterial growth on the agar plates (no colonies) were recorded as the MBCs.The effect type on the E. coli strains was also investigated based on the value of MBC/MIC (bactericidal if MBC/MIC < 4 or bacteriostatic for MBC/MIC ≥ 4) according to method previously described (Pavithra et al., 2010).

Evaluation of the interaction between the EEP and antibiotics
In order to evaluate the anti E. coli efficacy of EEP and antibiotic combinations, the agar diffusion method was carried out as previously described by Nweze and Eze (2009).Briefly, the bacterial strains prepared as inoculum were cultured on Mueller Hinton agar mixed with sub-inhibitory concentration of EEP and tested against five antimicrobials: amoxicillin/clavulanic acid (AMC, 20/10 µg), tetracycline (TE, 30 µg), gentamycin (CN, 10 µg), enrofloxacin (ENR, 5 µg) and florfenicol (FFC, 30 µg).After 24 h of incubation at 37 °C, the inhibition zone diameters (IZD) were measured.The results of the assay were expressed as percentage change in IZD compared to controls (IZD for antibiotics) and recorded as synergistic effect for IZD ≥ 19%, additivity for IZD between 0-19% and antagonism for IZD < 0 (Nweze and Eze, 2009).

Statistical analysis of results
Statistical analysis of data was performed using the Microsoft Office Excel 2010 program.The results were expressed as average ± standard deviation.Comparisons between antibiotics and EEP combinations were assessed by analysis of variance (ANOVA).In addition, Pearson's

Chemical characterization of ethanolic extracts of propolis
The analysed propolis ethanolic extracts had a dark reddishbrown color, with a specific taste and smell of aromatic resins from Populus sp.To evaluate both total phenolics and flavonoids (flavones/flavonols and flavanones/dihydroflavonols), spectrophotometric methods were applied according to Popova et al. (2004), Bankova (2005) and Laslo (2007); spectrophotometric methods are fast and reliable methods compared to chromatographic which have the disadvantage to be more expensive.Total phenolics (Table 1) in the studied EEP ranged from 30.61 mg/g propolis to 48.11 mg/g propolis.The lowest amount of phenolics was recorded for EEP 1 and 2, which were collected from mountain area, where fir is the main vegetal source of resins for propolis.EEP 4 and 5 presented the highest amount of total phenolics (over 45 mg/g propolis) and this is due to abundance of other vegetal sources of resins like pine, poplar and chestnut.These results are in agreement with previous reports on Romanian propolis composition (Laslo, 2007;Stan et al., 2011).
The smallest amount of flavonoids was recorded for EEP 3 (about 5 mg/g propolis), although total phenolics were in average range (34.96 ± 9.36 mg/g propolis) compared to the other EEP (38.02 ± 8.30 mg/g propolis).This lead to assumption that EEP 3 had a higher concentration of phenolic acids, hypothesis that was further confirmed by HPLC results (Table 2), where data clearly showed that the highest concentration of p-coumaric acid and ferrulic acid was found in this case.
These compounds are considered typical for poplar type of propolis, which is mainly found in European temperate zone (Bankova, 2005;Laslo, 2007).All EEP presented most of aforementioned compounds.At the same time, specific phenolics like caffeic acid, galangin and pinostrobin were missing from sample 3.
The most common compounds were represented by pcoumaric acid, ferrulic acid and chrysin.P-coumaric acid was found in all EEP in concentration ranging from 0.46 (sample 2) to 2.45 mg/g propolis (sample 3), while siringic acid and sinapic acid were least present and found only in EEP 3 with concentration of 0.22 mg/g propolis and 0.03 mg/g propolis, respectively.
Flavonoid chrysin was the main compound in propolis EEP in concentration ranging from 0.59 mg/g propolis (sample 3) to 2.75 mg/g propolis (sample 4), while galangin was completely absent in EEP 3, 4 and 5 and in very low amount in EEP 1 and 2 (0.02-0.58 mg/g propolis).
Compared to Croatian propolis analysed by Barbarić et al. (2011) that had the ferrulic acid as the most commonly found phenolic acid (0.03-0.9 mg/g propolis), the EEP in our experiment identified the same compound, but in higher concentration (0.13-2.39 mg/g propolis).Other components identified in Croatian propolis were: pcoumaric acid in 16 samples out of 20, with concentration ranging from 0.0023 mg/g to 0,156 mg/g propolis, chrysin in only 8 samples and in the range between 0.7-4.1 mg/g propolis and galangin in 17 samples and with higher concentration (0.37-47.48 mg/g propolis) (Barbarić et al., 2011).

Antimicrobial activity of ethanolic extracts of propolis
The antimicrobial potential of EEP was expressed in vitro against all tested E. coli strains, but with lower inhibition zone diameters in case of the antibiotic-resistant ones.The obtained diameters varied from 20.5 ± 0.7 mm (EEP 2) to 25.0 ± 1.4 mm (EEP 5) in case of antibioticsusceptible strains and 16.5 ± 0.7 mm (EEP 2) to 19.5 ± 0.7 mm (EEP 4) for the antibiotic-resistant strains, respectively.All five EEP inhibited the growth of E. coli; the most intense inhibitory effect was recorded for EEP 4 and 5 (Table 3).In case of antibiotic-susceptible E. coli strains, the values of IZD were similar to those determined by the antibiotics, the positive controls of the assay, while for the antibioticresistant strains significant differences were noticed when comparing to antibiotics -EEP combinations (Fig. 2) as follow: P < 0.0001 for AMC, CN, ENR and FFC and P < 0.001 for TE (P value determined by ANOVA analysis).The extracts solvent (70% ethanol) had not inhibitory activity on the studied strains indicating the antimicrobial efficacy of propolis against E. coli.
The minimum inhibitory concentrations (MICs) of the EEP ranged between 0.125 and 2% (v/v) when tested against E. coli antibiotic-susceptible strains, while those obtained for antibiotic-resistant strains were between 2 and 4% (v/v).As for the minimum bactericidal concentrations (MBCs), the values were similar or two times higher than MICs for both types of E. coli strains.The EEP effect against E. coli strains was considered bactericidal based on the interpretation previously established (Pavithra et al., 2010) for the ratio MBC/MIC < 4 (Table 3).
A percentage change in the inhibition zone diameter ≥ 19% was noticed for 92% (23/25) of the EEP and antibiotics combinations; thus, according to the interpretation given by Nweze and Eze (2009), in vitro synergistic interactions (Table 4) were established between the EEP and five antibiotics frequently used in bovine pathology, including also E. coli induced mastitis.The synergism between EEP and antibiotics was observed for all the tested antimicrobials, except for florfenicol (Table 4) that in combination with EEP 1 and 3 displayed additive effect against the E. coli strains.
With regard to propolis antibacterial activity against E. coli, as opposed to certain data found in literature presenting weak or no activity against this bacterium (Bankova et al., 1999;Kosalec et al., 2005;Gonsales et al., 2006;Seidel et al., 2008), all tested EEP were found active in vitro against E. coli strains, both antibiotic-susceptible and antibioticresistant strains isolated from bovine mastitis.The Romanian propolis ethanolic extracts efficacy on E. coli was reported also by Mărghitaş et al. (2013), with inhibition zone diameters ranging between 7 -12 mm and MIC of 0.625% (v/v).These results are consistent also with those reported by Hegazi et al. (2000), which evaluated European propolis extracts obtained from France, Austria and Germany and observed antimicrobial activity against Staphylococcus aureus, Escherichia coli and Candida albicans.German propolis displayed the highest antimicrobial potential against Staphylococcus aureus and Escherichia coli and possessed significant high concentrations of flavonoids (Hegazi et al., 2000).Thus, the variations noticed in the antibacterial activity of propolis from distinct areas can be explained taking into consideration the chemical composition complexity of this natural product.
The antimicrobial potential of propolis was previously considered as a possible alternative for the bovine mastitis treatment, but such in vitro studies were focused on Staphylococcus aureus (Rindt et al., 2009a;Santana et al., 2012).A recent study pointed out the in vitro efficacy of Egyptian propolis on several Gram-positive bacteria isolated from bovine mastitis (Staphylococcus aureus, coagulasenegative staphylococci, Streptococcus agalactiae, Streptococcus dysgalactiae) and the lack of activity against Gram-negative bacteria (E. coli and Pasteurella spp.) (Hegazi et al., 2014).
To the best of authors' knowledge, this is the first study aimed to investigate the antibacterial activity of Romanian propolis and antibiotics combinations against antibioticsensitive and antibiotic-resistant E. coli strains isolated from bovine mastitis.Synergistic effects of tested EEP and five Fig. 2. Inhibition zone diameter (IZD) (mm) determined for EEP and antibiotics combination against E. coli antibiotic-resistant strains (n = 5) modulation of β-lactam resistance (Cushnie and Lamb, 2011).
Given the great variability of the chemical composition of propolis and the propolis compounds role in the expression of antimicrobial potential, Popova et al. (2005) stated that a complete characterization of this property should involve qualitative and quantitative chemical analysis.Additionally, scientific data demonstrated that quantification of propolis active principles as groups of compounds correlated better with biological activity, especially the antimicrobial action, than the quantification of individual constituents (Popova et al., 2010).Therefore, following the chemical characterisation of EEP, with the flavonoids and total phenolics quantitative determination and the chromatographic identification of phenolic compounds, and the antibacterial potential evaluation, Pearson correlation coefficients were calculated between these parameters (Table 5).
Flavanones/dihydroflavonols were significantly negatively correlated with MICs for both antibioticsensitive E. coli strains (r 2 = -0.922)and antibiotic-resistant E. coli strains (r 2 = -0.888).The results also indicated a very strong negative correlation between total phenolics with MICs for antibiotic-sensitive E. coli strains (r 2 = -0.864)and antibiotic-resistant E. coli strains (r 2 = -0.827)and a similar pattern in case of total phenolics with MBCs (r 2 = -0.806for antibiotic-sensitive E. coli strains and r 2 = -0.511forantibiotic-resistant E. coli strains).Previous study (Bankova, 2005) had already demonstrated strong negative correlation between the concentration of total phenolics in propolis and MIC the greater the concentration, the lower the MIC (P = 0.003).This data supports the concept that 332 antibiotics (amoxicillin/clavulanic acid, tetracycline, gentamycin, enrofloxacin and florfenicol) against E. coli strains of bovine origin were observed.
The synergistic activity between propolis and antibiotics was reported by other authors (Stepanović et al., 2003;Orsi et al., Scazzocchio et al., 2006;Orsi et al., 2012), but most of these studies included Staphylococcus spp., mainly Staphylococcus aureus strains (Krol et al., 1993;Stepanović et al., 2003;Fernandes Júnior et al., 2005;Onlen et al., 2007).From the group of Gram-negative bacteria, Salmonella Typhi manifested in vitro an enhanced susceptibility towards combinations between both Brazilian and Bulgarian propolis and antibiotics such as amoxicillin, ampicillin and cephalexin (Orsi et al., 2006;Orsi et al., 2012), with similar MICs (9.90 and 10.0%, respectively), but with different types of action: bacteriostatic activity in case of Brazilian propolis, while the Bulgarian one acted bactericidal (Orsi et al., 2006).
The ability of propolis extracts to potentiate the antimicrobial activity of other substances was reported also in case of essential oils (Probst et al., 2011), honey (Noori et al.,2012), lysozyme (Ramanauskienė et al., 2009).
According to scientific literature, certain mechanisms associated with such synergistic effects may encompass flavonoids from propolis conferring several antibacterial properties: a decreased resistance of the bacterial wall that becomes more susceptible to antibiotics (Pascoal et al., 2014), direct inhibitory effect on ribosomes (Sforcin and Bankova, 2011;Orsi et al., 2012;Pascoal et al., 2014), inhibition of several bacterial enzymes (Daglia, 2012), alteration of bacterial protein expression (Daglia, 2012), measuring the concentrations of groups of active compounds instead of that of individual components is the right approach in the case of propolis (Bankova, 2005;Popova et al., 2007).
Phenolic compounds such as flavonoids represent a key element of propolis samples characterization in connection with the biological activity of this natural product (Bankova et al., 1995;De Castro, 2001;Marcucci et al., 2001;Kosalec et al., 2005;Gonsales et al., 2006;Alencar et al., 2007).Thus, the identification and quantification of phenolic compounds in Romanian propolis ethanolic extracts indicate not only the authenticity and the quality of tested EEP, but also underline the complex antimicrobial potential manifested against E. coli and moreover in the form of synergism with antibiotics.

Conclusions
The study indicated for Romanian propolis ethanolic extracts the typical poplar composition profile with flavonoids and phenolic acids as main biological active compounds, with chromatographic analysis data confirmed also spectrophotometrically.Furthermore, a strong antimicrobial efficacy positively correlated to the chemical composition was exhibited against E. coli strains isolated from bovine mastitis, along with interesting synergistic interaction with antibiotics that can be further investigated to obtain propolis-based formulation with antibacterial properties.Subsequent in vitro and in vivo studies evaluating the safety and efficacy are intended to consider propolis in veterinary therapeutic protocols.

Table 1 .
Spectrophotometric quantitative determination of specific flavonoid groups and total phenolics in analyzed EEP

Table 2 .
Phenolics identified by HPLC in analyzed EEP

Table 3 .
In vitro antibacterial efficacy of tested propolis samples against

Table 4 .
Aspects of the interaction between EEP and E. coli antibiotic resistant strains (n = 5)