Antibacterial activity of Turkish propolis and its qualitative and quantitative chemical composition.

Abstract

The antibacterial activity of propolis from different regions of Turkey was studied, accompanied by TLC and GC-MS analyses of its chemical composition and spectrophotometric quantification of the most important active principles. All six samples were active against the bacterial test strains used; however, samples 1 (Yozgat), 2 (Izmir) and 3 (Kayseri) were more active than samples 4 (Adana), 5 (Erzurum) and 6 (Artvin). By TLC comparison all samples were found to contain poplar taxonomic markers but in samples 4 (Adana), 5 (Erzurum) and 6 (Artvin), different substances were observed, which were not present in P. nigra L. bud exudate. The typical poplar samples 1 (Yozgat), 2 (Izmir) and 3 (Kayseri) displayed very similar phenolic and flavonoid content. Samples 4 (Adana), 5 (Erzurum) and 6 (Artvin) were characterized by low phenolic and very low flavonoid concentrations. Qualitative analysis by GC-MS revealed that sample 4 (Adana) contained diterpenic acids and high percent of cinnamyl cinnamate, sample 5 (Erzurum)--significant amounts of hydroxy fatty acids and triterpenic alcohols, and sample 6 (Artvin)--phenolic glycerides, characteristic for the bud exudate of Populus euphratica Oliv. The results confirm the importance of phenolics for propolis antibacterial activity, and the significance of P. nigra L. as a propolis source, which provides the hive with the best defense against microorganisms.

[c] 2004 Elsevier GmbH. All rights reserved.

Keywords: Turkish propolis; Antibacterial activity; Bees; P. nigra; Populus euphratica

**********

Introduction

Bees use propolis (bee glue) not only as a building material but also to keep low concentration of bacteria and fungi in the hive. Thus, the action against microorganisms is an essential characteristic of propolis and that's why it has been applied as a remedy by man since ancient times (Ghisalberti, 1978). It is still one of the most frequently used remedies in the Balkan states (Wollenweber et al., 1990), applied for treatment of wounds and burns, soar throat, stomach ulcer, etc. Propolis is on the drug market of some European countries as a medication against prostate hyperplasia (Dos Santos Pereira et al., 2002).

Numerous reports describe the antimicrobial properties of bee glue, and many active components have been identified (Marcucci, 1995; Burdock, 1998; Bankova et al., 2000; Banskota et al., 2001). Due to the variability of plant sources, the chemical composition of propolis is also highly variable and in distinct geographic regions the antibacterial compounds in bee glue are different, for example, flavonoids and cinnamic acid derivatives in European samples, diterpenic acids and prenylated coumaric acids in Brazilian, etc. For this reason, the complete characterization of antibacterial activity of propolis has to involve qualitative and quantitative chemical analysis; however, such studies are rare (Bonvehi and Coll, 1994; Park et al., 1998). Chemical data are especially valuable with respect to the problem of propolis standardization based on active principles, which is of crucial importance for the use of propolis preparations in mainstream medical practice (Bankova and Marcucci, 2000). In this article, we report on the antibacterial activity of propolis from different regions of Turkey, accompanied by TLC and GC-MS analyses of its chemical composition and spectrophotometric quantification of the most important active principles.

Materials and methods

Propolis samples

Geographic origin of the samples is listed in Table 1, see also Fig. 1. All the samples were collected in summer 2001.

TLC identity test for poplar propolis

TLC analysis of the alcohol extract was performed on silica gel (Alufolien Kieselgel Merck [F.sub.254]) with mobile phase petroleum ether/ethyl acetate 7:3. Visualization of the spots--UV light (366 nm); spraying with 60% sulfuric acid in ethanol and heating at 100[degrees]C. Standard substances: pinostrobin ([R.sub.f] 0.82, color at 366 nm: dark yellow, color after charring: gray), pinocembrin [R.sub.f] 0.63, yellow, color at 366 nm: orange, color after charring: yellow-orange), benzyl ferulate ([R.sub.f] 0.57, color at 366 nm: blue, color after charring: gray), galangin ([R.sub.f] 0.52, color at 366 nm: dark yellow, color after charring: yellow orange), chrysin ([R.sub.f] 0.46, color at 366 nm: dark brown, color after charring: yellow), phenethyl caffeate ([R.sub.f] 0.31, color at 366 nm: blue, color after charring: gray), kaempferol ([R.sub.f] 0.20, color at 366 nm: yellow, color after charring: yellow).

GC-MS analysis

Propolis, grated after cooling, was extracted for 24 h with 70% ethanol (1:10, w/v) at room temperature. The extract was evaporated to dryness. About 5 mg of the residue was mixed with 50 [micro]l of dry pyridine and 75 [micro]l bis(trimethylsilyl)trifluoracetamide, heated at 80[degrees]C for 20 min and analyzed by GC-MS. The GC-MS analysis was performed with a Hewlett Packard Gas Chromatograph 5890 Series II Plus linked to Hewlett Packard 5972 mass spectrometer system equipped with a 23 m long, 0.25 mm id, 0.5 [micro]m film thickness HP5-MS capillary column. The temperature was programmed from 100 to 310[degrees]C at a rate of 5[degrees]C/min. Helium was used as a carrier gas, flow rate 0.7 ml/min. Split ratio 1:80, injector temperature 280[degrees]C, ionization voltage 70 eV. The identification was accomplished using computer searches on a NIST98 MS data library. In some cases, when identical spectra have not been found, only the structural type of the corresponding component was proposed on the basis of its mass-spectral fragmentation. If available, reference compounds were co-chromatographed to confirm GC retention times. The components of ethanol extracts of propolis were determined by considering their areas as percentage of the total ion current. Some components remained unidentified because of the lack of authentic samples and library spectra of the corresponding compounds.

[FIGURE 1 OMITTED]

Quantification procedures

Flavone and flavonol content

Total flavone and flavonol content was measured by spectrophotometric assay based on aluminum chloride complex formation (Bonvehi and Coll, 1994; Popova et al., 2003). To 2 ml of the test solution, 20 ml methanol and 1 ml 5% Al[Cl.sub.3] (wt/vol) were added and the volume made up to 50 ml (volumetric flask). After 30 min, the absorbance was measured at 425 nm. Blank: 2 ml methanol instead of test solution. Every assay was carried out in triplicate. Flavone and flavonol content were estimated using a calibration curve of galangin, concentration range of 4-32 [micro]g/ml.

Flavanone and dihydroflavonol content

For flavanones and dihydroflavonols, the colorimetric method from DAB9, modified for propolis (Nagy and Grancai, 1996; Popova et al., 2003). One ml of test solution and 2 ml of DNP solution (1 g DNP in 2 ml 96% sulfuric acid, diluted to 100 ml with methanol) were heated at 50[degrees]C for 50 min. After cooling to room temperature, the mixture was diluted to 10 ml with 10% KOH in methanol (wt/vol). One ml of the resulting solution was added to 10 ml methanol and diluted to 50 ml with methanol. Absorbance was measured at 486 nm. Blank: 1 ml methanol instead of test solution was used in analogous procedure. Every assay was carried out in triplicate. Flavanone and dihydroflavonol content was estimated using calibration curve of pinocembrin, concentration range of 0.18-1.8 mg/ml.

Total phenolic substances

For quantification of total phenolics, the Folin-Ciocalteu method (Waterman and Mole, 1994; Woisky and Salatino, 1998) was used. One ml of the test solution was transferred to a 50 ml volumetric flask, containing 15 ml distilled water, and 4 ml of the Folin-Ciocalteu reagent and 6 ml of a 20% sodium carbonate solution (wt/vol) were added. The volume was made up with distilled water to 50 ml. After 2 h, the absorbance was measured at 760 nm. Blank solution: 1 ml methanol instead of test solution was used in analogous procedure. Every assay was carried out in triplicate. Total phenolics content was estimated using calibration curve of standard mixture pinocembrin-galangin 2:1, concentration range 37-326 [micro]g/ml.

Assay of propolis extract

Propolis (cooled) was grated before extraction. Two successive extractions were performed with 70% EtOH. One gram was dissolved in 30 ml, left for 24 h at room temperature, filtered, and then the procedure was repeated. (A third extraction under the same conditions gave negative reaction to 5% ferric chloride, which means that the extraction was complete.) The extracts were filtered with paper filter, combined and diluted to 100 ml in a volumetric flask. This extract was analyzed for the determination of the total flavanone and dihydroflavonol content.

From this extract, 3 ml were transferred into a 50 ml volumetric flask, filled up with methanol and this solution was further analyzed for total phenolics and for flavones and flavonols.

Antibacterial tests

Propolis samples obtained from different regions of Turkey were grounded, 30% ethanolic extracts of propolis were prepared and protected from bright light, with moderate shaking at room temperature (Sforcin et al., 1995). After a week, extracts were filtered and diluted in culture medium. Staphylococcus aureus ATCC 29213, Escherichia coli ATCC 25922 and their clinical isolates were obtained at the Erciyes University Hospital. The minimal inhibitory concentration (MIC) was defined as the lowest concentrations at which no bacterial growth was observed after incubation at 37[degrees]C for 24 h. Determination of MICs by the agar dilution method was performed, following the National Committee of Clinical Laboratory Standard Guidelines (1985).

Bacterial strains were grown in Sheep Blood Agar (Oxoid) at 37[degrees]C/24 h. After incubation a few colonies of each strain were suspended in 2 ml sterile saline, adjusted to 0.5 Mc Farlands and diluted to yield a final inoculum [10.sup.4] bacteria in 5 [micro]l suspension.

Serial concentrations of propolis from different samples were achieved (%v/v) in plates containing Mueller Hinton Agar, as follows: 0.1%, 0.2%, 0.4%, 0.8%, 1.75%, 3.5%, 7.0% and 14.0%. Each antimicrobial test also included plates containing the culture medium plus ethanol, in order to obtain a control of the solvent antibacterial effect.

Results and discussion

There has been only limited research on antibacterial activity of Turkish propolis (Velikova et al., 2001; Keskin et al., 2001). In this work, we could verify that S. aureus is susceptible to very low propolis concentrations (Table 2). Kujumgiev et al. (1999). Sforcin et al. (2000), and Drago et al. (2000) also showed an efficient propolis antibacterial action on S. aureus. On the other hand, E. coli was more resistant and these results were in agreement with those of Grange and Davey (1990), and Drago et al. (2000). All the samples, originating from different geographic locations in Turkey (Fig. 1), were active against Gram-positive and Gram-negative bacterial test strains; values were similar to those reported by Drago et al. (2000) (8-3 g/l for S. aureus; 250 g/l for E. coli). However, with respect to the magnitude of the MIC, two groups of samples could be observed. Obviously, samples 1-3 were more active against E. coli than samples 4-6, and samples 5 and 6 show lower activity against S. aureus than the other samples. The quantitative and qualitative chemical composition could provide an explanation for the observed differences. The analysis of all the biologically important individual components of propolis is often a tedious, time consuming and expensive procedure. It could be avoided in cases when the plant origin, and respectively, the qualitative composition of the bee glue is known. In such case, the rapid and low-cost determination of the quantitative chemical profile of the solution by measuring the total concentration of active compound groups is convenient and reasonable. Recent studies of Turkish bee glue have shown that its main source is poplar bud exudate (Velikova et al., 2001). It contains pentenyl and aromatic caffeates, pinocembrin, pinobanksin-3-O-acetate and galangin, which are regarded as taxonomic markers for poplars of the section Aigeiros (Greenaway et al., 1991b). These propolis samples could be regarded as "poplar type" propolis. The chemical profile of "poplar" propolis can be characterized by the following parameters: total flavone and flavonol content, total flavanone and dihydroflavonol content, and total phenolics content. Newly, we developed a procedure for proving identity of poplar propolis and the analysis of poplar propolis samples by spectrophotometric methods (Popova et al., 2003). Here we applied it to find out what the chemical differences between the two sample groups were.

[FIGURE 2 OMITTED]

The TLC comparison of a model mixture of typical poplar constituents (see Materials and Methods) with the alcohol extracts of the propolis samples was performed. All samples turned out to contain the poplar markers but in samples 4-6, different substances were observed, which are not present in P. nigra bud exudates (Fig. 2). According to the results of the TLC screening, all the samples could be analyzed for their flavonoid and phenolic content, using spectrophotometric procedures developed for the poplar type propolis. The results are represented in Table 1. The typical poplar samples, originating from Central and Western Anatolia, showed high antibacterial activity and displayed very similar phenolic and flavonoid content. Samples 4 (from Adana, Central Anatolia), 5 and 6 (both from Eastern Anatolia) were characterized by very low flavones and flavanones concentrations. Total phenolics were also low. Undoubtedly, the higher activity of samples 1-3 against E. coli is connected to higher concentration of phenolics, flavones and flavanones. The low activity of samples 5 and 6 against S. aureus might be also connected to low concentrations of these substances, but the high activity of sample 4 needs some explanation.

[FIGURE 3 OMITTED]

Detailed study of the qualitative chemical differences between samples of poplar (1-3) and of mixed origin (4-6) was performed by GC-MS analysis of the alcohol extracts after silylation (see Materials and Methods). The GC-MS analysis confirmed the results obtained by TLC and revealed the specific chemical characteristics of all of samples 4-6. Samples 1-3 were confirmed to contain the typical poplar flavonoid aglycones (pinocembrin 1, pinobanksin 2, pinobanksin 3-O-acetate 3, chrysin 4, galangin 5) phenolic acids (p-coumaric 6, ferulic 7, caffeic) and esters (pentenyl caffeates, benzyl and phenethyl esters of caffeic, ferulic and p-coumaric acids 8, 9, 10). Obviously, their major plant source is black poplar P. nigra, the most widespread Aigeiros poplar in temperate zones (Table 1).

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

In sample 4 (Fig. 3) these compounds were present, however, the main constituent (over 14% of the total ion current) was cinnamyl cinnamate 11 (Fig. 4), found in poplar propolis only as a minor component (Greenaway et al., 1991b). In addition, several diterpenic acids (dihydroabietic, abietic, isopimaric, a total of about 4-5%) were also identified, their plant source remained unknown. Diterpenic acids were found to contribute to the antibacterial activity of propolis (Bankova et al., 1996) and this could explain the high activity of sample 4 against S. aureus.

Sample 5 was characterized, besides the poplar phenolics, by the presence of significant amounts of hydroxy fatty acids (hydroxypalmitic, hydroxystearic), and triterpenic alcohols, which obviously originate from another source plant, growing in the region of Erzurum. It is interesting to note that in a recent study on Turkish propolis, a sample from Erzurum was also found to contain low amounts of flavonoids (Sorkun et al., 2001).

The composition of sample 6 deserves special attention (Fig. 5). Low concentration of flavonoids and very high levels of p-coumaric 6 and ferulic 7 acids were detected by GC-MS. Further, a series of phenolic glycerides were identified: glyceryl-, monoacetylglyceryl- and diacetyilglyceryl- esters of p-coumaric, ferulic and caffeic acid, together with 1,3-di-p-coumarouyl-2-acetyl glycerol, 1,3-diferuloyl-2-acetyl glycerol, 1-p-coumaroyl-3-feruloyl-2-acetyl glycerol and 1-p-coumaroyl-3-caffeoyl-2-acetyl glycerol. Their presence explains the relatively high percentage of total phenolics in this sample (Table 1). This chemical composition was found to be characteristic for the bud exudate of Populus euphratica Oliv., one of the poplar species growing in Turkey (Greenaway et al., 1991a). Obviously, in the region of Artvin P. euphratica was the main source plant for propolis collection. Till now, this tree has not been detected to play the role of propolis source.

In conclusion, our results confirm the importance of the amount of phenolic compounds, flavones and flavanones for the antibacterial activity of poplar propolis. They also stress the significance of P. nigra L. as a propolis source, which provides the hive with the best defense against microorganisms. These findings confirm that the determination of the type of propolis, according to its plant source, is an important issue in standardization of bee glue. They could help to establish criteria for quality control of propolis.

Table 1. Percentage of main biologically active compounds in propolis
from different regions of Turkey

        Geographical                       % in the sample
Sample  origin        Plant origin         Ph (a)  F-F (b)  F-D (c)

1       Yozgat        P. nigra             26.4    8.7      6.0
2       Izmir         P. nigra             30.4    9.6      5.5
3       Kayseri       P. nigra             27.5    5.6      4.8
4       Adana         P. nigra + unknown1   8.2    1.5      2.7
5       Erzurum       P. nigra + unknown2  10.5    2.0      1.5
6       Artvin        P. euphratica        14.5    2.0      3.0

(a) Total phenolics.
(b) Total flavones and flavonols.
(c) Total flavanones and dihydroflavonols.

Table 2. Susceptibility of Gram-positive and Gram-negative bacteria to
ethanolic extract of propolis from different regions of Turkey
(concentration in % (v/v) of 30% propolis extract)

Sample            Range      MI[C.sub.50]   MI[C.sub.90]  Range

1                 <0.1       <0.1           <0.1          0.4-14
2                 <0.1       <0.1           <0.1          0.2-14
3                 <0.1-0.2   <0.1            0.2          3.5-14
4                 <0.1       <0.1           <0.1          3.5 to > 14
5                  0.2-0.4    0.4            0.4          0.4 to > 14
6                  0.2-0.4    0.2            0.4          3.5 to > 14
Tetracycline (a)  <0.1       <0.1           <0.1          0.2-0.4

Sample            MI[C.sub.50]  MI[C.sub.90]

1                   7            14
2                   7            14
3                   7            14
4                  14           >14
5                 >14           >14
6                  14           >14
Tetracycline (a)    0.2           0.4

(a) Concentration in %.

Acknowledgements

The authors wish to thank Mrs. Antonova for technical assistance with GC-MS and Mrs. Veleva for technical assistance with spectrophotometric analyses.

Received 3 April 2003; accepted 2 September 2003

References

Bankova, V., Marcucci, M.C., 2000. Standardization of propolis: present status and perspectives. Bee World 81, 182-188.

Bankova, V., Marcucci, M.C., Simova, S., Nikolova, N., Kujumgiev, A., Popov, S., 1996. Antibacterial diterpenic acids of Brazilian propolis. Naturforsch 51c, 227-280.

Bankova, V., De Castro, S.L., Marcucci, M.C., 2000. Propolis: recent advances in chemistry and plant origin. Apidologie 31, 3-15.

Banskota, A.H., Tezuka, Y., Kadota, Sh., 2001. Recent progress in pharmacological research of propolis. Phytother. Res. 15, 561-571.

Bonvehi, J.S., Coll, F.V., 1994. Phenolic composition of propolis from China and from South America. Z. Naturforsch. 49c, 712-718.

Burdock, G.A., 1998. Review of the biological properties and toxicity of bee propolis (propolis). Food & Chem. Toxicol. 36, 347-363.

Dos Santos Pereira, A., Silva Seixas, F.R.M., Radler de Aquino Neto, 2002. Propolis: 100 anos de pesquisa e suas perspectives futures. Quim. Nova 25, 321-326.

Drago, L., Mombelli, B., De Vecchi, E., Fasina, M.C., Tocalli, L., Gismondo, M.R., 2000. In vitro antimicrobial activity of propolis dry extract. J. Chemother. 12, 390-395.

Ghisalberti, E.L., 1978. Propolis: a review. Bee World 60, 59-84.

Grange, J.M., Davey, R. W., 1990. Antibacterial properties of propolis (bee glue). J. R. Soc. Med. 83, 159-160.

Greenaway, W., Gumusdere, I., Whatley, F.R., 1991a. Analysis of phenolics of bud exudate of Populus euphratica by C-MS. Phytochemistry 30, 1883-1885.

Greenaway, W., May, J., Scaysbrook, T., Whatley, F.R., 1991b. Identification by gas chromatography-mass spectrometry of 150 compounds in propolis. Z. Naturforsch. 46c, 111-121.

Keskin, N., Hazir, S., Baser, S.H., Kurkcuoglu, M., 2001. Antibacterial activity and chemical composition of Turkish propolis. Z. Naturforsch. 56c, 1112-1115.

Kujumgiev, A., Tsvetkova, I., Serkedjeva, Yu., Bankova, V., Christov, R., Popov, S., 1999. Antibacterial, antifungal and antiviral activity of propolis of different geographic origin. J. Ethnopharmacol. 648, 235-240.

Marcucci, M.C., 1995. Propolis: chemical composition, biological properties and therapeutic activity. Apidologie 26, 83-99.

Nagy, M., Grancai, D., 1996. Colorimetric determination of flavanones in propolis. Pharmazie 51, 100-101.

Park, Y.K., Koo, M.H., Abreu, J.A.S., Ikegaki, M., Cury, J.A., Rosalen, P.L., 1998. Antimicrobial activity of propolis on oral microorganisms. Curr. Microbiol. 36, 24-28.

Popova, M., Bankova, V., Butovska, D., Petkov, V., Damyanova, B., Sabatini, A.G., Marcazzan, G.L., Bogdanov, S., 2003. Poplar type propolis and analysis of its biologically active components. Honeybee Science 24 (2), 61-66.

Sforcin, J.M., Fernandes Jr., A., Lopes, C.A.M., Bankova, V., Funari, S.R.C., 2000. Seasonal effect on Brazilian propolis antibacterial activity. J. Ethnopharmacol. 73, 243-249.

Sforcin, J.M., Funari, S.R.C., Novelli, E.L.B., 1995. Serum biochemical determinations of propolis treated rats. J. Venom. Anim. Toxin. 1, 31-37.

Sorkun, K., Suer, B., Salih, B., 2001. Determination of chemical composition of Turkish propolis. Z. Naturforsch. 56c, 666-668.

Velikova, M., Bankova, V., Sorkun, K., Popov, S., Kujumgiev, A., 2001. Chemical composition and biological activity of propolis from Turkish and Bulgarian origin. Mellifera 1, 57-59.

Waterman, P.G., Mole, S., 1994. Analysis of Phenolic Plant Metabolites. Blackwell Scientific Publications, Cambridge, MA.

Woisky, R.G., Salatino, A., 1998. Analysis of propolis: some parameters and procedures for chemical quality control. J. Apicult. Res. 37, 99-105.

Wollenweber, E., Hausen, B.M., Greenaway, W., 1990. Phenolic constituents and sensitizing properties of propolis, poplar balsam and balsam of Peru. Bull. Group Polyphenols 15, 112-120.

M. Popova (a), S. Silici (b), O. Kaftanoglu (c), V. Bankova (a,*)

(a) Institute of Organic Chemistry with Centre of Phytochemistry, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria

(b) Erciyes University, S. Cikrikcioglu Vocational College, Department of Animal Sciences, Kayseri, Turkey

(c) Cukurova University, Faculty of Agriculture, Department of Animal Sciences, Adana, Turkey

*Corresponding author. Tel.: [phone]; fax: [phone].

E-mail address: [e-mail] (V. Bankova).