Nonaqueous Polyethylene Glycol as a Safer Alternative to Ethanolic Propolis Extracts with Comparable Antioxidant and Antimicrobial Activity

We compared the chemical composition, antioxidant and antimicrobial activity of two propolis extracts: one obtained with nonaqueous polyethylene glycol, PEG 400 (PgEP), and the other obtained with ethanol (EEP). We analyzed the total phenolic content (TPC) and the concentrations of ten markers of propolis antioxidant activity with HPLC-UV: caffeic acid, p-coumaric acid, trans-ferulic acid, trans-cinnamic acid, kaempferol, apigenin, pinocembrin, chrysin, CAPE, and galangin. Antioxidant activity was tested using DPPH and FRAP assay, and antimicrobial activity was assessed through minimum inhibitory concentrations (MICs) and minimum biofilm eradication concentration (MBEC) determination. Maceration gave the yield of propolis of 25.2 ± 0.08% in EEP, and 21.5 ± 0.24% in PgEP. All ten markers of antioxidant activity were found in both extracts, with all marker concentrations, except kaempferol, higher in EEP. There was no significant difference between the TPC and antioxidant activity of the PgEP and the EEP extract; TPC of PgEP was 16.78 ± 0.23 mg/mL, while EEP had TPC of 15.92 ± 0.78 mg/mL. Both extracts had antimicrobial activity against most investigated pathogens and Staphylococcus aureus, Acinetobacter baumannii, and Escherichia coli biofilms. EEP was more effective against all tested susceptible pathogens, except E. coli, possibly due to higher content of kaempferol in PgEP relative to other polyphenols. Nonaqueous PEG 400 could be used for propolis extraction. It gives extracts with comparable concentrations of antioxidants and has a good antioxidant and antimicrobial activity. It is a safe excipient, convenient for pediatric and veterinary formulations.


Introduction
Propolis is a honey bee product known for its antioxidant, antimicrobial, and antiinflammatory properties [1,2]. Over the centuries, it had a significant role in traditional medicine and is still used today as an antibiotic alternative due to the growing problem of

Introduction
Propolis is a honey bee product known for its antioxidant, antimicrobial, and antiinflammatory properties [1,2]. Over the centuries, it had a significant role in traditional medicine and is still used today as an antibiotic alternative due to the growing problem of antimicrobial resistance [3]. Besides the antibiotic resistance problem, natural products are more valuable than synthetic ones due to their lower cytotoxicity [4].
Honey bees collect crude propolis from resinous plant material to strengthen wax combs and thermoinsulate the beehive, and at the same time, to protect the colony from pathogens. Various plants produce resinous compounds with anti-putrefaction, waterproofing, and heat-insulating properties to protect their plant organs. Honey bees gather these resins and transform them into propolis by chewing and mixing resins with salivary glands products and bee wax. Propolis ("bee glue") is a strong adhesive for blocking holes and cracks in the beehive, strengthening the nest, and smoothing out the internal walls [2,3]. In addition, since incorporated into beehives and in direct contact with adult bees, it contains other compounds such as essential oils, pollen, enzymes from the honey bee's digestive system, and other organic compounds [2].
Extraction is a crucial and the most demanding stage in the utilization of bioactive compounds. Most in vitro assays of propolis biological activity use aqueous ethanolic (60-80% w/w EtOH) or 96% ethanolic (EtOH) extracts. Although very effective, these extracts are not suitable for all purposes. Ethanol is a relatively aggressive solvent, not appropriate for pregnant and breastfeeding women, children, certain patients, and veterinary medicine use. Furthermore, due to the relatively high content of beeswax in such extracts, they are not suitable for manufacturing various pharmaceutical, cosmetic, veterinary, and other products containing predominantly water. The latter acts as anti-solvent-type precipitation of beeswax when mixing with the rest of the water-based formulations. Thus, various meth-ods and solutions are studied to replace conventional maceration with aqueous ethanol [12]. Although extraction technology is advanced and new solvents are introduced, ethanol remained a golden standard due to its efficacy and affordability [12].
Polyethylene glycol 400 (PEG 400) is a non-toxic, low-molecular-weight type of polyethylene glycol (PEG). It is widely used in different pharmaceutical formulations as an excipient serving as solvent, diluent, and humectant. Several studies compared the composition and activities of aqueous polyethylene glycol propolis extracts with ethanol, but they mostly used only 20% of PEG in water [13][14][15]. Pure, nonaqueous PEG 400 could be a more effective extraction solvent suitable for simple methods of propolis extraction such as maceration. Therefore, this study compared the chemical composition and biological activities of nonaqueous PEG 400 (PgEP) with the ethanolic propolis extract (EEP).

Propolis Collection and Pre-Treatment
Propolis was collected in Croatia from the beehives situated at ten stationary apiaries in Sisak-Moslavina County during one week in late summer. In the vicinity of apiaries, there are mixed oak and poplar tree forests. Samples of propolis were collected from two commercial traps per apiary. Each trap with a sample of the crude poplar-type propolis was chilled in a refrigerator at −20 • C for minimally 1 h, and harvested. Then, all collected propolis samples were mixed in one composite sample (1 kg), which was milled and prepared for extraction.

Propolis Extraction with 96% Ethanol as the Extraction Solvent
The propolis extract was obtained by maceration. First, ethanol (96%; 70.00 g) was added to milled propolis (30.00 g). This mixture was left at room temperature for 72 h with periodical stirring. Then, the mixture was filtered, yielding 60.00 g of isolated liquid propolis extract EEP (85.7% w/w against the mass of starting extraction solvent). This process was repeated three times, and it was estimated that EEP had 25.2% ± of propolis in a deep brown-colored solution with an intensive propolis odor.
The yield (% from theoretical yield) is expressed as weight (% w/w) percent of isolated liquid propolis extract against the mass of starting extraction solvent (EtOH, PEG).

Propolis Extraction with PEG 400 as the Extraction Solvent
Milled propolis (30.00 g) was mixed with PEG 400 (70.00 g). The obtained mixture was left at room temperature to macerate for 72 h with periodical stirring and filtration at the end. The process was repeated three times. This yielded 55.00 g (78.6% w/w) of liquid propolis extract PgEP with approximately 21.5% ± 0.24 propolis, in the form of a deep brown viscous liquid of intensive propolis odor.

Total Amount of Phenolic Compounds Determination
The total amount of phenolic compounds (TPC) was determined using the Folin-Ciocalteu (F-C) assay. First, a 200 µL of 10% F-C reagent was added to 100 µL of each propolis extract and vortexed thoroughly. Then, 800 µL of 700 mM Na 2 CO 3 was added to each tube and incubated at room temperature for 2 h. 200 µL of each sample was transferred to a clear 96-well microplate, and the absorbance was read at 765 nm. A standard curve was calculated from the blank-corrected A 765 of the gallic acid standards (c (gallic acid) = 0.05-0.5 mg/mL). Total phenolics were calculated as gallic acid equivalents using the regression equation between gallic acid standards and A 765 .

Determination of Ferric Reducing/Antioxidant Power (FRAP Assay)
A solution of 10 mM 2,4,6-Tri (2-pyridyl)-s-triazine (TPTZ) and 20 mM ferric chloride was diluted in 300 mM sodium acetate buffer (pH 3.6) at a ratio of 1:1:10. Extracts (20 µL) were added to the 96-well microplate, followed by a working FRAP solution (280 µL). The mixture was shaken and incubated for 30 min at 37 • C in the dark. The final concentration of tested extracts was 1 mg/mL. The absorbance at 593 nm was recorded using microplate reader µQuant TM (BioTek ® Instruments, Inc., Winooski, VT, USA). Ferrous sulfate (FeSO 4 ·7H 2 O) was used to develop a 20-2000 µmol/L standard curve. All results were then expressed as Fe 2+ equivalents (Fe 2+ µmol/mg propolis extract). All tests were carried out in triplicate, and the results were averaged.

DPPH Radical Scavenging Assay
An equal volume of tested phenolic extracts at various concentrations was added to a solution of 2,2-diphenyl-1-picrylhydrazyl (DPPH; final concentration 100 µM in absolute ethanol), in a 96-well microtiter plate. An α-tocopherol solution of 10 mg/mL was a positive control, and ethanol was used as a negative control. After the incubation of 30 min in the dark, the absorbance was recorded at 517 nm on µQuant TM (BioTek ® Instruments, Inc., Winooski, VT, USA) reader at room temperature. All measures were carried out in triplicate.
The percentage scavenging of test samples at each concentration was calculated using the following formula: [(Abs control − Abs compound )/Abs control ] × 100 The EC 50 values for each compound were calculated from dose-response curves using linear regression analysis. The serial microdilution procedure was used to determine the MIC of the extracts. Cell suspensions were prepared from the parent culture in phosphate-buffered saline (PBS) buffer (pH 7.4), and these were adjusted to 0.5 McFarland units by nephelometry. The testing was performed in serial dilution in 96-well microtiter plates in the range from 100 to 0.7125 µg/mL, by addition of 100 µL of the solution of propolis extract dissolved in Mueller Hinton broth. After inoculation 100 µL of each bacterial culture adjusted to 10 5 cfu/mL, plates were incubated for 24 h at 37 • C. The MIC values were determined by the addition of 10 µL of 0.5 mg/mL solution of 2,3,5-triphenyl-2H-tetrazolium chloride (TTC; redox indicator) per single well, and, after incubation for 4 h at 30 • C, the absorbance was determined by spectrophotometry at wavelength 490 nm. The MIC values were expressed as the propolis extract concentration at which 80% reduction in bacteria occurred (MIC 80 ). For fungal species, the MIC values were determined in the RPMI medium with additional glucose, using the same scheme as with bacteria. After incubation (48 h, 37 • C, aerobic conditions, in the dark), 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-carboxanilide-2H-tetrazolium) (XTT; redox indicator) was added in combination with menadion, and the absorbance was determined by spectrophotometry at wavelength 540 nm. The MIC values were determined as the propolis extract concentration at which 80% reduction in bacteria or fungi occurred (MIC 80 ). The negative control contained only the medium and the solvent (without added microorganisms and propolis), while the positive control was exposed to the influence of antibiotics (gentamicine sulphate, norfloxacin, colistin, nystatin) and voriconazole (Pfizer, New York, NY, USA) as antifungal agent (MIC for C. albicans, ATCC 90028, was 0.01 ug/mL).

Determination of Minimum Biofilm Eradication Concentrations (MBEC) of Propolis Extracts
We determined the minimum biofilm eradication concentration (MBEC) of PgEP and EEP on S. aureus, E. coli, A. baumannii, and C. albicans. Wells (96 well plate) were pretreated with FBS (fetal bovine serum; Millipore Sigma, St. Louis, MI, USA) (250 µL per well). In each well, 100 µL of bacterial (10 7 CFUs/mL) or yeast (5 × 10 6 CFUs/mL) suspension was added. Negative controls contained broth only, and gentamycin and amphotericin were used for positive controls. Plates with bacteria were covered and incubated for 24 h, while plates with yeasts were left for 48 h in aerobic environment. After incubation, we aspirated each well. In order to remove all non-adherent bacteria/yeast, each well was washed three times and vigorously shaken. The attached cells that remained were fixed with methanol (15 min), and left overnight to dry. We used 1% crystal violet to stain the formed biofilm for 5 min. The remaining stain was rinsed by placing the plate under running tap water and the plates were left to dry. Adherent cells were solubilized using ethanol. The absorbance was read at 570 nm. The MBEC value represents the lowest dilution of a compound at which bacteria fail to grow.

Statistical Analyses
All extractions and analyses were carried out in triplicates. Mirobiological analyses were performed as technical and biological replicates. Analyses were carried out using GraphPad Prism software 9.1.1. for Windows (GraphPad Software, San Diego, CA, USA, www.graphpad.com, accessed on 2 January 2021). MIC/MBEC 80 were obtained using non-linear regression. The normality of distributions were confirmed with the D'Agostino-Pearson omnibus test and Shapiro-Wilk test. Unpaired t-test was used to test the differences between the two corresponding groups, with the significance set at p ≤ 0.05.

Results
Representative markers of antioxidant activity (Figure 1) were found in both extracts, as presented with chromatograms in Figure 2 and their concentrations in Figure 3. In comparison to Pg-EP, EEP had higher concentrations of all markers except kaempferol (5).

Results
Representative markers of antioxidant activity (Figure 1) were found in both extracts, as presented with chromatograms in Figure 2 and their concentrations in Figure 3. In comparison to Pg-EP, EEP had higher concentrations of all markers except kaempferol (5). as presented with chromatograms in Figure 2 and their concentrations in Figure 3. In comparison to Pg-EP, EEP had higher concentrations of all markers except kaempferol (5).
Total amount of phenolic compounds (TPC) is presented in Figure 4, while antioxidant activities of EEP and PgEP are presented in Figures 5 and 6. There was no significant difference between the TPC and antioxidant activity of the PgEP and the EEP extract. TPC of PgEP was 16.78 ± 0.23 mg/mL, while EEP had TPC of 15.92 ± 0.78 mg/mL (Figure 3). In DPPH assay, EC 50 was 11.71 ± 1.34 µg/mL for PgEp vs. 13.50 ± 0.87 µg/mL for EEP. In the FRAP assay, the reducing ability of PgEP was 41.03 ± 0.91 vs. 40.37 ± 1.28 Fe 2+ µmol/mg propolis extract of EEP (Figures 5 and 6).
Total amount of phenolic compounds (TPC) is presented in Figure 4, while antioxidant activities of EEP and PgEP are presented in Figures 5 and 6. There was no significant difference between the TPC and antioxidant activity of the PgEP and the EEP extract. TPC of PgEP was 16.78 ± 0.23 mg/mL, while EEP had TPC of 15.92 ± 0.78 mg/mL (Figure 3). In DPPH assay, EC50 was 11.71 ± 1.34 μg/mL for PgEp vs. 13.50 ± 0.87 μg/mL for EEP. In the FRAP assay, the reducing ability of PgEP was 41.03 ± 0.91 vs. 40.37 ± 1.28 Fe2 + μmol/mg propolis extract of EEP (Figures 5, 6).  Antibacterial activities of EEP and PgEP are presented in Tables 1 and 2. MIC80 values are expressed as the amount (%) of initial propolis content in the extract. Both extracts had antimicrobial activity against S. aureus, MRSA, MSSA, E. faecalis, E. coli and A. baumannii, but not against VRE, P. aeruginosa, and C. albicans (Table 1). In addition, the extracts had good activity against biofilm formed by S. aureus, E. coli, and A. baumannii, but did not work against C. albicans biofilm ( Table 2). The most sensitive bacteria to EEP were gram positives, especially S. aureus and MSSA. EEP was more effective, with 40× dilution reducing 80% of S. aureus, compared with PgEP, with effective 10× dilution. The remarkable difference was in reducing MSSA, where even 52.5× diluted EEP was effective as 10× Antibacterial activities of EEP and PgEP are presented in Tables 1 and 2. MIC80 values are expressed as the amount (%) of initial propolis content in the extract. Both extracts had antimicrobial activity against S. aureus, MRSA, MSSA, E. faecalis, E. coli and A. baumannii, but not against VRE, P. aeruginosa, and C. albicans (Table 1). In addition, the extracts had good activity against biofilm formed by S. aureus, E. coli, and A. baumannii, but did not work against C. albicans biofilm ( Table 2). The most sensitive bacteria to EEP were gram positives, especially S. aureus and MSSA. EEP was more effective, with 40× dilution re-  (Table 1). In addition, the extracts had good activity against biofilm formed by S. aureus, E. coli, and A. baumannii, but did not work against C. albicans biofilm ( Table 2). The most sensitive bacteria to EEP were gram positives, especially S. aureus and MSSA. EEP was more effective, with 40× dilution reducing 80% of S. aureus, compared with PgEP, with effective 10× dilution. The remarkable difference was in reducing MSSA, where even 52.5× diluted EEP was effective as 10× diluted PgEP. EEP was not more effective than PgEP against E. coli, which was interestingly the most susceptible bacteria to antimicrobial activity of PgEP.
As the technology develops, so do the methods of propolis extraction. In recent years, progress has been made toward more efficient methods regarding the time and solvents used. These methods include Soxhlet, ultrasound-assisted extraction (UAE), microwave-assisted extraction (MAE), supercritical CO 2 extraction (scCO 2 ), and highpressure methods [8]. Still, maceration is the most affordable and accessible extraction method, with ethanol as the relatively most effective solvent. Although there are some new promising solvents, such as natural deep eutectic solvents (NADES) [8], most of the further developed technologies are still using ethanol as the solvent. Ethanolic and aqueous ethanolic propolis extracts are used commercially in most propolis-based products and various in vitro research applications. The highest yields of propolis extract is obtained by aqueous ethanol containing 70 to 96% of ethanol [8]. In addition, Ma et al. (2016) confirmed that ethanol and methanol propolis extracts had greater antioxidant activity than water, ethyl acetate, chloroform, and benzene extracts [29].
On the other hand, several studies showed the contrary. Water extracts of Brazilian and Turkish propolis demonstrated greater scavenging capabilities of DPPH, H 2 O 2 , O 2 •− , and • OH radicals [30]. Rocha et al. (2013) demonstrated that differently prepared extracts of propolis had significant DPPH radical scavenging activity. Still, water extract exhibited a higher antioxidant activity than EEP [31]. Galeotti et al. (2018) demonstrated that propolis from the same source, in the form of aqueous ethanolic, glycolic (1,2-PG), glyceric, oil solutions, and powder extract had a similar chemical composition with differences in TPC [32]: the highest was in 1,2-PG extract (81.2 ± 3.7%), followed by hydroalcoholic extract (69.7 ± 2.0%), while the lowest was in powder, a micronized sample composed of propolis with a minimum of 12% total polyphenols. This form had the highest concentrations of caffeic (1), p-coumaric (2), ferulic (3), and isoferulic acids. All extracts had comparable antioxidant activity, presented as µg Trolox equivalent/mg polyphenols [32].
To the best of our knowledge, the use of PEG as the extraction solvent was mentioned for the first time among other organic solvents by Sosnowski (1981), in his patent. [33]. However, there are no data on the antioxidant's content and activity presented in the patent. Later, other authors used aqueous PEG 400 (AqPg-EP, with 20% PEG 400), but these extracts had lower concentrations of bioactive molecules and antioxidant and antimicrobial activity [13,14,34]. Interestingly, none of the previous studies used the pure, nonaqueous PEG 400 for propolis extraction. According to Kubiliene et al. (2015), it is more difficult to sterilize the extract by filtration when PEG 400 content in the solvent is above 20% [13]. Still, the methods of propolis extract sterilization should be a subject of further research.
In our study, all the ten markers of activity were found in both extracts. There was no significant difference between the TPC (Figure 3) and the antioxidant activity of the PgEP and the EEP extract ( Figures 5 and 6). In comparison to EEP, PgEP had lower concentrations of all marker compounds except kaempferol (5). This selectivity for kaempferol is a result of two facts: (i) PEG 400 is a slightly more polar solvent than EtOH and more effectively extracts more polar flavonoid compounds. Since kaempferol (5) is the most polar propolis flavonoid of all analyzed, significantly higher quantitative flavonoid 5 content was obtained with PEG 400 than with less polar EtOH; (ii) Due to the presence of four phenolic hydroxyl (OH) groups which could act as hydrogen bond donors and one chromene oxygen acting as hydrogen bond acceptor, kaempferol (5) could form a kind of molecular complex with PEG 400, which would be stabilized by five (!) hydrogen bonds. PEG 400, with its average molecular weight (Mw) 380-420, and average 8.7 ethylene oxide (EO) units per molecule, contains two hydroxyl (OH) groups, at the beginning and the end of the chain, as well as 7 or 8 oxygen atoms of ether functions. The structure of possible kaempferol-PEG 400 complex (5a), which might explain the significantly increased chemoselectivity of its extraction with PEG 400 in PgEP over the EEP, is given in Figure 7.
In our study, all the ten markers of activity were found in both extracts. There was no significant difference between the TPC (Figure 3) and the antioxidant activity of the PgEP and the EEP extract ( Figures 5 and 6). In comparison to EEP, PgEP had lower concentrations of all marker compounds except kaempferol (5). This selectivity for kaempferol is a result of two facts: (i) PEG 400 is a slightly more polar solvent than EtOH and more effectively extracts more polar flavonoid compounds. Since kaempferol (5) is the most polar propolis flavonoid of all analyzed, significantly higher quantitative flavonoid 5 content was obtained with PEG 400 than with less polar EtOH; (ii) Due to the presence of four phenolic hydroxyl (OH) groups which could act as hydrogen bond donors and one chromene oxygen acting as hydrogen bond acceptor, kaempferol (5) could form a kind of molecular complex with PEG 400, which would be stabilized by five (!) hydrogen bonds. PEG 400, with its average molecular weight (Mw) 380-420, and average 8.7 ethylene oxide (EO) units per molecule, contains two hydroxyl (OH) groups, at the beginning and the end of the chain, as well as 7 or 8 oxygen atoms of ether functions. The structure of possible kaempferol-PEG 400 complex (5a), which might explain the significantly increased chemoselectivity of its extraction with PEG 400 in PgEP over the EEP, is given in Figure 7. Since there is a correlation between phenolic levels of propolis extracts and their antioxidant and antimicrobial activities [7,13,35], we expected to see that both of our extracts had comparable antimicrobial activity. Tosi et al. [36] were among the first to prove that solvents used for propolis extraction can increase the antimicrobial efficacy of propolis. Their propolis oil extracts had a wider range of antimicrobial activity than ethanol (60%), propylene glycol, and glycerine, but ethanol and propylene glycol extracts had good activity against yeasts [36]. In the previously mentioned study by Kubiliene et al. (2015), propolis extracts made only with water or oil had no antimicrobial effect [13]. EEP and non-alcoholic extracts with PEG 400 (AqPgEP and oAqPgEP) had antimicrobial activity against S. aureus, E. coli, P. aeruginosa, K. pneumoniae, B. cereus, and C. albicans. Interestingly, the antimicrobial activity of propolis non-alcoholic extracts (AqPgEP and oAqPgEP) against all investigated microorganisms was equal or even higher (against P. aeruginosa and K. pneumoniae) than EEP [13]. In our study, EEP was more effective against most of the investigated pathogens except E. coli. This could be due to the chemoselective extraction of kaempferol because it was shown that kaempferol is one of the strongest flavonoid inhibitors of E. coli growth [37]. In that same study carried out by Wu et al. (2013), kaempferol was located deeply in the hydrophobic core of the lipid bilayer of the model membrane, decreasing the membrane fluidity most while exhibiting the highest antibacterial activity against E. coli in comparison to other 10 commercially available flavonoids [37].
Our findings are consistent with studies that confirmed the superior antimicrobial efficacy of ethanolic extracts in comparison to solvents such as hexane [38], water [13,[39][40][41], and propylene glycol [40]. In their comprehensive literature review on antimicrobial properties of propolis, Przybyłek and Karpiński (2019) sort out the antibacterial activity of propolis extracts prepared with different solvents [42]. In almost every pathogen studied (even the E. coli) EEP had lower MIC levels than water extracts. On the other hand, organic solvents such as dichloromethane and methanol propolis extracts were more effective than EEP [42]. However, these solvents are quite toxic and impose a health risk to anyone working with them.

Conclusions
Even though PEG 400 gave a lower yield of propolis extraction than ethanol (21.5 ± 0.24 vs. 25.2 ± 0.08%), it gave very similar TPC and antioxidant activity. Moreover, it extracted all the ten markers of antioxidant activity with chemoselectivity over kaempferol. This could also explain the greater activity of PEG 400 propolis extract against E.coli in comparison to ethanolic extract. PEG400 is a polar, hydrophilic solvent known as a non-toxic, which is very well-tolerated and of low cost. It is acceptable even in the most demanding cases, such as pediatric formulations. With the advent of new green methods of polyethylene glycol production [43], we believe this will become a sustainable and environmentally friendly solvent base for propolis extraction.

Patents
Some of the presented methods and results are published in the WIPO patent No. WO/2020/169425: LIQUID PROPOLIS EXTRACT, ITS FORMULATION AND USE THEREOF [44]: extraction method, HPLC analyses, MIC determination.