Effect of Encapsulation on Antimicrobial Activity of Herbal Extracts with Lysozyme

The last decade has been characterized by a growing interest in natural antioxidants. Natural herbal sources contain a diverse array of compounds such as phenolic acids, fl avonoids, tannins, vitamins and terpenoids that account for their biological properties. Their antioxidant and antimicrobial abilities and health-promoting properties are mainly att ributed to phenolic compounds (1,2). Polyphenols are secondary metabolites ubiquitously distributed in all higher plants, where they have important roles as defense against plant pathogens and other unfavourable environmental conditions. Several thousand plant polyphenols are known, encompassing a wide variety of molecules that contain at least one aromatic ring with one or more hydroxyl groups in addition to other substituents. They have antioxidant, anticancer and anti-infl ammatory eff ects (3). The activity of polyphenols from medicinal plants against a wide range of microorganisms has been extensively investigated. The plant polyphenols represent a source of anti-infective agents against antibiotic-resistant pathogens (3,4).


Introduction
The last decade has been characterized by a growing interest in natural antioxidants.Natural herbal sources contain a diverse array of compounds such as phenolic acids, fl avonoids, tannins, vitamins and terpenoids that account for their biological properties.Their antioxidant and antimicrobial abilities and health-promoting properties are mainly att ributed to phenolic compounds (1,2).Polyphenols are secondary metabolites ubiquitously distributed in all higher plants, where they have important roles as defense against plant pathogens and other unfavourable environmental conditions.Several thousand plant polyphenols are known, encompassing a wide variety of molecules that contain at least one aromatic ring with one or more hydroxyl groups in addition to other substituents.They have antioxidant, anticancer and anti--infl ammatory eff ects (3).The activity of polyphenols from medicinal plants against a wide range of microorganisms has been extensively investigated.The plant polyphenols represent a source of anti-infective agents against antibiotic-resistant pathogens (3,4).
Recent years have witnessed an increase in the interest in plants (mainly plant aqueous extracts) used for wound healing.Skin, the fi rst barrier between body and the outer environment, protects the body from the harsh external conditions.Drug application to the skin surface as topical route of administration has high potential in therapy, and allows delivery and controlled release of active substances into internal environments (5,6).Many of the antimicrobials available on the market are transported with diffi culties through cell membranes and exhibit low intracellular activity, resulting in reduced intracellular potency.Additional drawback for eff ective antimicrobial therapy is the fact that increasing number of microorganisms appear to be developing resistance to potent antimicrobials.Thus, many of the currently available broad--spectrum antibiotics are losing their eff ect.In addition, many antibiotics suff er from poor bioavailability, due to their limited solubility (7).
Resistance to antimicrobial agents has become an increasingly important and pressing global problem (8).The antibacterial activity of phenolic components is being increasingly documented (1)(2)(3)(4).Crude extracts from plants with a history of use in folk medicine have been screened in vitro for antibacterial activity (5).Some researchers have reported synergistic eff ect of naturally occurring fl avonoids and other antibacterial agents against resistant strains of bacteria (8,9).Due to the problem of microbial resistance to antibiotics, att ention is given to biologically active components isolated from plant species commonly used in herbal medicine, as they can be used as antimicrobial preparations (10)(11)(12).
Many studies focus on the development of nanoparticle systems, with dimensions between 1 and 100 nm, for antimicrobial drug delivery (8).Nanoparticles have unique physicochemical properties such as ultrasmall and controllable size, large surface area to mass ratio, high reactivity and functionalizable structure (13).These properties can be applied to facilitate the administration of antimicrobial drugs, thereby overcoming some of the limitations in traditional antimicrobial therapeutics.In recent years, encapsulation of antimicrobial drugs in nanoparticle systems has emerged as an innovative and promising alternative that enhances therapeutic eff ectiveness and minimizes undesirable side eff ects of drugs (14).Extensive studies have demonstrated that nanoparticles such as liposomes, polymeric nanoparticles, solid lipid nanoparticles and others can be used for antimicrobial drug delivery and controlled release (15).Encapsulation provides a means to control stability, solubility, and bioavailability, as well as release of bioactive components.Various techniques of encapsulation of natural bioactive components have been proven as an eff ective method to increase their absorption in vitro and in vivo (16).
Liposomes are currently the most widely studied antimicrobial and clinically established nanoscale systems for drug delivery.They have been extensively studied in pharmaceutical and cosmetic industries (15)(16)(17).Their bilayer structure, resembling cell membrane, enables their easy fusing with infectious microorganisms.Their excellent biocompatibility, biodegradability and possibility to manipulate their size and surface properties makes them ideal particles in drug delivery (17,18).Liposomal antimicrobial delivery systems have also several advantages such as improved solubility, bioavailability and effi cacy, reduced toxicity, and increased product stability (17).Be-sides their unique benefi ts, liposomes show some disadvantages such as low stability, low encapsulation effi ciency, high cost of manufacturing, degradation by hydrolysis or oxidation, sedimentation, aggregation or fusion during storage (18).Thus, other natural polymers based on polysaccharides have been tested as delivery systems for antimicrobial agents (19,20).
The aims of the present study are to assess and compare the antimicrobial activities of selected herb and spice extracts.Possibilities of encapsulation of these antimicrobial herb and spice extracts alone and in mixture with lysozyme and nisin were tested.Long-term stability of particles and antimicrobial eff ect of encapsulated compounds was studied under model conditions.
Herbs and spices were bought from the local market in Brno, Czech Republic.Three diff erent solvents (water, ethanol and citric acid) were used separately to extract each spice and herb.Water extracts were prepared by adding 10 g of spice or herb to 100 mL of hot sterilized distilled water.These were allowed to stand for 15 min and then fi ltered through a cheesecloth.The fi ltrates were kept in sterilized vials.Other extracts were prepared analogously, where the water was replaced with 10 % citric acid and/or 20 % ethanol solution.Total phenolic and fl avonoid contents, antioxidant activity and antimicrobial potential of the spice and herb extracts were analysed.Finally, these extracts were used for encapsulation into liposome and polysaccharide particles.

Determination of antioxidant activity using ABTS
Total antioxidant activity was determined by ABTS radical cation decolourization assay (21).ABTS was dis-solved in water to a 7-mM concentration.The ABTS radical cation (ABTS •+ ) was produced by reacting ABTS stock solution with 2.45 mM potassium persulfate and allowing the mixture to stand in the dark at room temperature for 12-16 h before use.The ABTS •+ solution was diluted with ethanol to an absorbance of 0.70±0.02at 734 nm.Then, 1 mL of diluted ABTS •+ solution was added to 0.010 mL of antioxidant compound or Trolox standard.The absorbance of the samples was read at 734 nm using a UV-Vis spectrophotometer (Thermo Spectronic Helios TM α, Thermo Fisher UK Ltd., Hemel Hempstead, UK) exactly 10 min aft er the initial mixing.Antioxidant activity was calculated as a change in the absorbance.Results were expressed in mg of Trolox equivalent (TE) per gram of dry sample.

Total phenolic and total fl avonoid contents
The total phenolic content of the three extracts of each spice and herb was measured using the Folin-Ciocalteu colourimetric method (22).Readings were quantifi ed using a standard curve of gallic acid and the results were expressed in mg of gallic acid equivalent (GAE) per g of dry sample.The total fl avonoid content was measured by the aluminium chloride colorimetric method (23).Readings were quantifi ed using a standard curve of catechin and the results were expressed in mg of catechin equivalent (CE) per g of dry sample.The measurements were performed using UV-Vis spectrophotometer (Thermo Spectronic Helios TM δ; Thermo Fisher UK Ltd.).

Liposome preparation
Liposomes can be prepared from many lipid species and by a variety of methods.Phospholipids from soya bean lecithin are widely used in liposomal active component delivery systems due to their safety and wide availability at relatively low cost for upscale production, which makes unpurifi ed soya bean phospholipids a good alternative and att ractive choice.The major phospholipids in soya bean lecithin are phosphatidylcholine, phosphatidylethanolamine and phosphatidylinositol (17).In this study, active components were encapsulated into liposome nanoparticles prepared using soya bean lecithin.Three diff erent methods for liposome preparation were tested: ethanol injection, thin fi lm evaporation and ultrasonication.For the formation of liposome, a solution of lecithin of 8 to 45 mg/mL with the addition of cholesterol in the ratio of lecithin/cholesterol of 9:1 was used.The lipid dispersion was sonicated (80 W, 20 kHz) in an ice bath using ultrasonic homogenizer (BANDELIN Electronic GmbH & Co. KG, Berlin, Germany) for a few minutes.During the ethanol injection the lecithin solution of 100 mg/mL was used.Multilamellar vesicles were prepared using the thin fi lm evaporation.In brief, the lipid phase (lecithin 13.5 mg/mL with the addition of cholesterol in the ratio of lecithin/cholesterol of 8:1) was dissolved in chloroform (with or without the encapsulated lipophilic component), which was then removed under reduced pressure in a rotary evaporator (IKA ® -Werke GmbH & Co. KG, Staufen, Germany), thus obtaining a thin fi lm of dry lipids on the fl ask wall.The fi lm was then hydrated by adding distilled water (with or without the encapsulated hydrophilic component) under vigorous stir ring in order to stimulate the vesicle formation (17).

Preparation of polysaccharide particles
Alginate, chitosan and starch particles were prepared by the emulsifi cation method based on gelation and cross-linking of polymers (15,19,20).Sodium alginate is a water-soluble polymer that forms a gel in the presence of multivalent cations such as calcium.The preparation of alginate nanoparticles was achieved in aqueous sodium alginate solution (with an encapsulated component) in which gelation was induced by the addition of calcium chloride solution under intensive stirring.This leads to the formation of invisible clusters of calcium alginate gel beads (15).Chitosan particles were prepared in a similar way as the alginate ones.Methods proposed to prepare chitosan nanoparticles are based on the spontaneous formation of complexes between chitosan and polyanions such as tripolyphosphate (19).Starch nanoparticles were formed using simple, fast and easy method of nanoprecipitation in absolute ethanol under controlled conditions (20).

Particle size and stability analysis
Particle size and size distribution are the key parameters used for the evaluation of the physical stability of nanoparticles.The dynamic light scatt ering (DLS) is widely used to determine the size and size distribution of small particles suspended in liquid medium.The mean particle size and size distribution indicated as polydispersity index (PDI) are the typical parameters of this technique.A PDI value between 0.1 and 0.25 indicates a narrow size distribution, while a PDI greater than 0.5 shows a broad size distribution (15).Although DLS techniques provide rapid measurement of particle size and size distribution, they do not have the capability of evaluating particle morphology similarly to direct visualization techniques such as microscopy.Laser Doppler electrophoresis is commonly used to measure zeta potential.This technique evaluates electrophoretic mobility of suspended particles in the medium.It is a general rule of thumb that an absolute value of zeta potential above 60 mV yields excellent stability, value of 30 mV generally results in good stability, 20 mV is acceptable short-term stability and less than 5 mV means fast particle aggregation (15).In this work, particle size distribution, average size of particles and polydispersity index were analysed by colloidal DLS analyzer Zetasizer Nano ZS (Malvern Instruments Ltd., Malvern, UK).Morphology of the prepared particles was observed by light microscope (Labomed Lx 500; Labomed Inc., Los Angeles, CA, USA) and scanning electron microscope (EVO LS 10; Carl Zeiss AG, Oberkochen, Germany), respectively.For other applications the unifi ed particle size (100 nm) was obtained using membrane extruder (LiposoFast, AVESTIN Europe GmbH, Mannheim, Germany).

Encapsulation effi ciency
Nisin, lysozyme and the above-mentioned herb and spice extracts were used for encapsulation.Encapsulation effi ciency was measured spectrophotometrically (UV-Vis spectrophotometer, Thermo Spectronic Helios TM δ; Thermo Fisher UK Ltd.) and/or by high-performance liquid chromatography with photodiode array (Thermo Fisher Scientifi c, Waltham, MA, USA).Nisin and lysozyme were separated in Aeris peptide XB-C18 column (Phenomenex Inc., Torrance, CA, USA) using isocratic elution with acetonitrile/water mixture (8:2) and the addition of trifl uoroacetic acid (0.1 %).The content of encapsulated components was measured before and aft er encapsulation.The percentage of encapsulation effi ciency (EE) was then calculated according to the following equation:

Long-term stability of particles
Stability of the prepared particles was tested under physiological conditions and in food models.Artifi cial stomach juice was prepared from 0.25 g of pepsin dissolved in 100 mL of distilled water.To this solution 0.84 mL of 35 % hydrochloric acid was added.Final pH was adjusted to 0.9.Artifi cial pancreatic fl uid was prepared with 0.25 g of pancreatin and 1.5 g of natrium hydrogen carbonate in 100 mL of water (pH=8.9).Bile fl uid was composed of 0.8 g of bile acid salts dissolved in 200 mL of phosphate buff er.
Incubation of particles in a ratio of 1:1 was performed at 37 °C for 20 min in stomach fl uid and pancreatic juice and for 40 min in bile fl uid.Aft er incubation, the mass fractions of total phenolics and lysozyme released from particles were measured.
Food models of four main types were prepared with the following compositions: 3 % solution of acetic acid was used as a model of acidic food (pH<4.5),neutral food with pH>4.5 was simulated by distilled water, 10 % ethanol solution was prepared as a model of alcoholic beverages and food containing alcohol, and fatt y food was simulated by 25 % emulsion of oil in water.Incubation mixture in a ratio of 1:3 (liposome/food model) was incubated for 1, 7 and 30 days at 5 °C.Mass fractions of released phenolics and lysozyme were determined in regular intervals.

Microorganisms and cultivation media
The antibacterial activities of all extracts and other active components were determined against several bacteria.For preliminary testing of antimicrobial activity, two Gram-positive bacteria (Bacillus subtilis CCM 2794 and Micrococcus luteus CCM 1569) and two Gram-negative bacteria (Escherichia coli CCM 7395 and Serratia marcescens CCM 8587) were used.All bacteria tested in this study were supplied by the Czech Collection of Microorganisms in Brno.Bacterial cultures were grown in commercial Luria-Bertani (LB) medium (Himedia Laboratories Pvt. Ltd., Mumbai, India) for E. coli and in commercial meat peptone medium (Himedia Laboratories Pvt.Ltd.) for M. luteus, B. subtilis and S. marcescens.Temperature for B. subtilis cultivation was 30 °C and for E. coli, M. luteus and S. marcescens 37 °C.Viability of bacteria was followed by fl ow cytometry (Apogee Flow Systems, Hemel Hempstead, UK).

Determination of antimicrobial activity
The antimicrobial properties of extracts from a wide variety of plants have been assessed and reviewed previously (3,4).In this work, two most widely used methods for determining the antimicrobial properties of active components, agar diff usion method and broth dilution method (24), were used.
The antimicrobial activity of extracts was evaluated using a slightly modifi ed agar disc or well diff usion method.A bacterial culture grown for 24 h (approx.5•10 8 CFU per plate) was cultivated on the surface of agar medium in Petri dishes.For disc diff usion test, an aliquot (10 μL) of herb or spice extract was applied on a sterile paper disc on the agar surface.An additional negative control disc was impregnated with 10 μL of sterile distilled water.The plates were inverted and incubated for 24 h at 30 °C (for B. subtilis) or 37 °C (for E. coli, M. luteus and S. marcescens).For well diff usion test, an aliquot (10-80 μL) of herb or spice extract was applied directly into the wells with 24-hour bacterial culture grown on Petri dishes.The same volume (10-80 μL) of sterile distilled water was plated into wells as negative control.The plates were incubated as described above.Microbial inhibition was determined by measuring the diameter of the clear zone (in mm) of growth inhibition around each disc or well.Experiments were repeated three times and the results were expressed as average values.Experiments were done at least twice.
In the broth dilution method, bacteria are inoculated into a liquid growth medium in the presence of diff erent concentrations of an antimicrobial agent.The bacterial cell concentration was determined turbidimetrically using ELISA reader (BioTek Instruments GmbH, Bad Friedrichshall, Germany) and expressed as absorbance at 630 nm (A 630 nm ).The results are usually presented as values of minimal inhibitory concentration (MIC) or/and minimal bactericidal concentration (MBC).MIC is defi ned as the lowest concentration of an antimicrobial that prevents the growth of microorganism aft er a specifi c incubation time, while MBC is defi ned as the lowest concentration of an antimicrobial agent needed to kill 99.9 % of the initial inocula (24).
The microbroth dilution assay was used to determine antimicrobial activity and the MIC of some free and encapsulated samples.A volume of 150 μL of 2•10 5 CFU/mL of bacterial culture was added to sterile 96-well microtiter plates (Thermo Scientifi c TM , Thermo Fisher Scientifi c, Inc.) followed by 50 μL of the antimicrobial sample.The negative control was prepared by adding 150 μL of bacterial culture (2•10 5 CFU/mL) followed by 50 μL of the sterile distilled water or prepared particles without encapsulated antimicrobial component.The cell concentration change in each well was determined turbidimetrically.Absorbance of the samples in the individual wells was measured using ELISA reader (BioTek Instruments GmbH) at 630 nm, before and aft er a 24-hour incubati on at 30 °C (for B. subtilis) or 37 °C (for E. coli, M. luteus and S. marcescens).The MIC was defi ned as the concentration of antimicrobial agent that allowed an increase in absorbance less than or equal to 0.05 aft er 24 h of incubation.

Statistical analysis
All analytical determinations were performed at least in triplicates.Results were expressed as mean value± standard deviation.Statistical data analysis was conducted using a Microsoft Excel spreadsheet (Microsoft Corporation, Redmond, WA, USA) and STATISTICA v. 12 software (StatSoft CR, s.r.o., Prague, Czech Republic), and the comparison of average values of each treatment was based on the analysis of variance (one-way ANOVA) according to Tukey's test at signifi cance level of 5 % (p≤0.05).

Antioxidant activity and the total phenolic content
The antioxidant activity of 18 tested herb and spice extracts is given in Table 1.The antioxidant activity (expressed as TE) was the highest in the clove extract (46.7 mg/g) followed by conefl ower (45.6 mg/g), marjoram (37.7 mg/g) and sage (34.8 mg/g) extracts.Extracts of thyme, plantago, St. John's wort, ginger and garlic had similar antioxidant activity (around 30 mg/g on average).The antioxidant activity was lower in the elderberry extract (24.0 mg/g).Other tested herb extracts had average antioxidant activity around 10 mg/g.The lowest antioxidant activity was measured in the chamomile extract.
The highest total phenolic content expressed as GAE (Table 1) was observed also in clove (60.8 mg/g), followed by marjoram (44.7 mg/g), St. John's wort (44.1 mg/g), plantago (43.0 mg/g), elderberry (38.8 mg/g) and sage (34.3 mg/g).Statistical diff erence in the total phenolic content of plant extracts can be seen.The lowest total phenolic content was found in the extracts of garlic and ginger, although they exhibited excellent antioxidant activity.These results may indicate that the phenolic components are not a sole source of antioxidant activity of herbal extracts.Finally, values of the total fl avonoid content in the tested plants are also shown in Table 1.These results, in most cases, refl ect the total polyphenol content, except in clove extract.The highest total fl avonoid content (expressed as CE) was observed in St. John's wort (39.3 mg/g), followed by elderberry (32.9 mg/g), marjoram (31.3 mg/g), thyme (29.0 mg/g) and plantago (24.7 mg/g) extracts.From these results it can be seen that there is a statistically signifi cant dependence between the total phenolic content and the total fl avonoid content of plant extracts.No correlation was found between the antioxidant activity and phenolic content.

Encapsulation effi ciency of natural extracts
Three types of plant extracts (aqueous, alcoholic and citric acid) were encapsulated into liposomes.Encapsulation was successful in all types of extracts (Table 2).Generally, phenolic components of herbal extracts were preferably encapsulated as aqueous extracts.Conversely, the worst encapsulation effi ciency was determined in citric acid extracts.Thus, type of extraction agent has an important infl uence on the encapsulation effi ciency.Encapsulation effi ciency of plant extracts in citric acid was in most samples signifi cantly lower than in other solutions, except for common nett le extract (61.2 %).In alcohol extracts the highest encapsulation effi ciency was found in clove extract (55.3 %), followed by the extracts of liquorice (54.3 %), rosemary (46.5 %) and marjoram (42.5 %).
Another important factor of encapsulation effi ciency is material used for particle preparation.Bett er results of encapsulation effi ciency of plant water extracts were achieved more frequently in liposomes than in polysaccharide particles (Table 3).Some extracts, for example that of ginger, exhibited considerably higher encapsulation effi ciency in polysaccharide particles (81.3 % in chitosan and 68.4 % in alginate) than in liposome (28.9 %).
Higher encapsulation effi ciency in polysaccharide particles than in liposomes was also determined in sea buckthorn, plantago, wormwood and clove extracts.
Other antimicrobial compounds encapsulated into particles were enzyme lysozyme (1-10 mg/mL) and bacterial peptide nisin (0.03-0.3 mg/mL).Similarly to the analyzed plant extracts, both these substances exhibited antimicrobial activity and they belong to Generally Rec- ognized as Safe (GRAS) compounds that can be used in food and cosmetics.Diff erent concentrations of lecithin (8-45 mg/mL) and methods of liposome preparation (sonication, thin fi lm evaporation and ethanol injection) were tested for liposome preparation.High encapsulation efficiency of lysozyme (Table 4) was determined in alginate and starch particles (about 90 %), while in chitosan particles encapsulation effi ciency of 21.9 % was achieved.In liposome particles prepared by sonication, high encapsulation effi ciency was found in lysozyme.
In further experiments, encapsulation of mixtures of fi ve selected herbal water extracts and lysozyme (0.5-1 mg/mL) was tested.Plant aqueous extracts were chosen according to their phenolic content, antioxidant activity and encapsulation effi ciency.The following plant extracts were co-encapsulated with lysozyme in a ratio of 1:1 (by volume): thyme, sage, lavender, elderberry and marjoram.Encapsulation effi ciencies of these extracts in liposomes are shown in Fig. 1.Negative infl uence of co-encapsulation on encapsulation effi ciency of individual com ponents Three diff erent parameters were used for characterization of organic particles prepared from diff erent material.Statistical analysis was done for each of these parameters individually.Lowercase lett er in superscript denotes statistical evaluation of encapsulation effi ciency, capital lett er stands for statistical evaluation of zeta potential and number represents statistical evaluation of particle diameter.Values with the same lett er or number in superscript are not statistically diff erent (p>0.05).Methods for liposome preparation: thin fi lm evaporation (TFE) and ethanol injection (EI); d=diameter of particle was not found; moreover, some increase of encapsulation effi ciency was observed when compared with individual extracts and lysozyme.

Stability and size of polysaccharide particles
The average size of polysaccharide particles containing nisin or lysozyme was larger than liposomes containing nisin or lysozyme (Table 4).Polydispersity index of polysaccharide particles was in a range from 0.2 to 0.6 (data not shown).Because of substantially higher encapsulation effi ciency of lysozyme than nisin in all types of particles, further experiments of co-encapsulation of antimicrobial herbal extracts and peptide were done with lysozyme only.
Generally, it can be said that most stable particles containing lysozyme and nisin (Table 4) were prepared by sonication and then by evaporation on a thin layer, regardless of the encapsulated active component.Zeta potential of prepared liposomes was in almost all cases in the range from -30 to -60 mV.Zeta potential of prepared polysaccharide particles was good, too.Chitosan particles were the most stable, while starch particles (-12.4 mV) were determined as the least stable (Table 4).
Size of the prepared liposomes containing plant extracts (Table 5) was on average 180 nm (121.0-358.3nm) and a polydispersity index value was not greater than 0.25, except for particles with garlic extract (data not shown).Zeta potential of these liposomes was in almost all cases in the range of -30 to -60 mV, which means that particles (except thyme) were stable.Statistically significantly higher values of zeta potential were found in liposomes with encapsulated plantago, lavender, clove, sea buckthorn and St. John´s wort extracts (Table 5).Particles containing ginger, thyme and wormwood extracts were less stable than empty liposome capsule under the same conditions.Zeta potential was also used for evaluation of colloid stability of liposomes with co-encapsulated lysozyme and selected plant extracts and it showed that particles were stable below -30 mV (Fig. 2).

Long-term stability of prepared particles
Studies of the stability of liposome particles with encapsulated plant extracts, antimicrobial peptides and their mixtures were performed in diff erent model food and physiological conditions as described in Materials and Methods.
The eff ect of temperature on the long-term stability of particles containing nisin or lysozyme was analyzed (Table 6).The mass fraction of released antimicrobial components was the highest in acidic medium.Fatt y medium also led to greater release of active components.Conversely, particles were stable in aqueous medium.In the ethanol medium the release of active substances depended mainly on the temperature.At 28 °C, signifi cant mass fraction of lysozyme was released, while at 5 °C minimum mass fraction of lysozyme was found outside the particles (Table 6).Thus, the results imply that storage at higher temperatures leads to signifi cant release of active ingredients.
The long-term stability of particles containing an antimicrobial herb and spice extract was analyzed as well (Table 7).The particles were added to the aqueous model   7).Conversely, long-term stability of liposome containing a plant extract was signifi cantly higher in particles with rosemary, marjoram, clove and thyme in most environments (Table 7).However, variability of data in dependence on the combination of active component and environment was relatively high.
These fi ndings were confi rmed during long-term storage experiments with mixtures of plant extracts and lysozyme in liposomes (Table 8).During one week of storage, the highest stability of encapsulated phenolics was found in water medium.Particles stored in fatt y medium were less stable.Water medium was confi rmed as the most suitable for long-term storage of particles with mixed antimicrobial compounds.Stabilization eff ect of herbal extracts on lysozyme and decreased lysozyme release was observed mainly in particles with sage and thyme stored in water (Table 8).
The stability of particles containing a mixture of an antimicrobial herb and spice extracts was also tested in Values with the same lett er in the same column are not statistically diff erent (p>0.05);n.d.=not detected physiological fl uid models, i.e. artifi cial stomach, bile and pancreatic juices (Fig. 3).The particles were added to the fl uids and incubated as described in Materials and Methods.Aft er incubation, samples were taken to determine the mass fraction of released phenolics and lysozyme.While most of phenolic compounds are released from liposomes mainly in artifi cial pancreatic and bile juices, lysozyme was released intensively also in acidic stomach juice.It can be concluded that co-encapsulation led to better stability of the active compound in the particles and to the bett er control of its release during digestive processes.

Antimicrobial activity of herb and spice extracts
In the present study, the antimicrobial activity of pure herb and spice extracts was determined before and aft er encapsulation into liposome or polysaccharide particles.Firstly, antimicrobial properties of the plant extracts alone were tested.Signifi cant variations in the antibacterial activity of each tested spice or herb sample were observed among water, ethanol and acid extracts (Table 9).Water extracts showed a signifi cantly higher overall inhibition against all bacterial strains tested than the other extracts.Thus, water extracts were used for encapsulation of plant antimicrobial agents.Antimicrobial activity was followed aft er 24-hour incubation with tested strain as MIC value as described in Materials and Methods.Results were expressed as A 630 nm of bacterial culture (turbidity decrease) aft er incubation with plant extract in free and/or encapsulated form for 24 h.Antimicrobial activity against Bacillus subtilis, Micrococcus luteus, Serratia marcescens and Esche richia coli is shown in Table 9.
All tested pure herbal extracts showed at least a partial antimicrobial activity against all tested strains (Table 9).Interestingly, some of the herbal extracts exhibited relatively good antimicrobial eff ect against tested Gram-negative strains, mainly E. coli (clove, sea buckthorn, conefl ower, wormwood, thyme, liquorice, pot marigold and chamomile).Antimicrobial eff ect against the Gram-negative strains was observed in all tested extracts except rosemary, ginger and elderberry.In the clove extract almost 100 % inhibitory eff ect was measured during 24 h.S. marcescens was less sensitive to the antimicrobial eff ect of plant extracts than E. coli.Inhibitory eff ect was found in extracts of sea buckthorn, pot marigold, common nett le, clove, conefl ower, plantago, marjoram and ginger.
Lysozyme and extracts of clove, conefl ower, liquorice, pot marigold, common nett le and plantago exhibited very high inhibitory activity against Gram-positive strain B. subtilis (Table 9).M. luteus was less sensitive, with extracts of sea buckthorn, rosemary, sage, clove, conefl ower, chamomile, wormwood and thyme having the highest antimicrobial eff ect.In some extracts a highly positive relationship between antioxidant activity, antibacterial activity and total phenolic content was confi rmed (Tables 1  and 9).Extracts with the highest antioxidant eff ect like  clove, conefl ower, sage or thyme also had the most pronounced antimicrobial activity.However, a very good antibacterial activity was detected in some herbs with lower antioxidant activity (for example liquorice, sea buckthorn, common nett le, wormwood, pot marigold and chamomile).On the other hand, in marjoram extract, with high antioxidant activity (Table 1), only minimum antimicrobial eff ect was found (Table 9).Some non-phenolic constituents of the extract also have the capability to act as antimicrobial agents (14).
Furthermore, antimicrobial activities of extracts encapsulated in diff erent materials (liposome, alginate and chitosan particles) were compared.High antimicrobial eff ect against all tested strains, mainly Gram-positive, was observed in extracts encapsulated in chitosan particles and then in liposome.Except rosemary, all tested extracts exhibited a statistically signifi cant increase of antimicrobial activity against all tested strains aft er encapsulation into liposome particles (Table 9).The best antimicrobial activity (predominantly against Gram-positive bacteria) was detected in liposomes containing extracts of clove and lysozyme.These particles showed a 100 % antimicrobial protection against all tested strains throughout 24 h.A very high antimicrobial eff ect against all strains was measured in liposome containing sage, sea buckthorn or chamomile extract.Liposomes containing wormwood and conefl ower extracts had almost 100 % inhibitory eff ect against S. marcescens.Liposomes contain-ing pot marigold and liquorice extracts also had high antimicrobial activity (Table 9).
Chitosan particles exhibited statistically signifi cant antimicrobial activity against all tested strains, particularly particles with encapsulated extracts of clove, sea buckthorn, sage or liquorice.High antimicrobial eff ect against Gram-negative strains was observed in chitosan particles, comparable with liposomes.In Gram-negative strains combination of test strains and type of encapsulated extract was more important than in Gram-positive strains.Generally, Gram-negative strains were sensitive mainly against particles containing wormwood and clove extracts.S. marcescens was highly sensitive to sage extract, while in E. coli antimicrobial eff ect was confi rmed also using the extracts of liquorice and sea buckthorn (Table 9).In conclusion, encapsulation of plant extracts into liposomes and chitosan particles led in most cases to an increase and prolongation of antimicrobial eff ect of the extracts.This eff ect could be caused by increased stability of active substances entrapped in polymer capsule.
It is interesting that chitosan particles had the highest inhibitory eff ect, but the lowest encapsulation effi ciency.These results (Table 9) confi rm the fi ndings of another study (19) about the possible application of antimicrobial activity of chitosan.Chitosan due to its polycationic nature can interact with negatively charged microbial cell walls and plasma membranes, resulting in decreased osmotic stability, membrane disruption and eventual leak- age of intracellular elements.In addition, chitosan is able to inhibit mRNA and protein synthesis by binding to microbial DNA.Nanoscale chitosan that has a higher surface-to-volume ratio, resulting in higher surface charge density, leads to increased affi nity towards bacteria and fungi and greater antimicrobial activity.Therefore, it is suitable for encapsulation of antimicrobial ingredients (19).
By testing the antimicrobial activity of lysozyme, high eff ect against all tested strains was reported (Fig. 4).However, the inhibitory eff ect against E. coli was very slight and at lower concentrations of lysozyme, no eff ect was observed.In particular, the highest eff ect was observed against Gram-positive strains (B.subtilis and M. luteus).Interestingly, relatively high antimicrobial eff ect was achieved against Gram-negative S. marcescens.The minimum inhibitory concentration of lysozyme was determined to be more than 500 μg/mL against B. subtilis, 1000 μg/mL against M. luteus and more than 1000 μg/mL against Gram-negative strains (Fig. 4).
Liposomes with co-encapsulated lysozyme (at concentrations of 1 and 0.5 mg/mL) and selected plant extracts were subjected to testing the antimicrobial activity as well (Fig. 5).High antimicrobial activity, particularly against Gram-positive strains was detected in all particles.In liposome particles containing a mixture of plant extracts and lysozyme, a statistically signifi cant increase of antimicrobial eff ect compared to liposomes containing plant extracts only was observed (Fig. 5, Table 9).The particles with mixed antimicrobial substances exhibited signifi cantly higher antimicrobial activity against all tested strains (Fig. 5).Higher concentration of lysozyme led to increased antimicrobial eff ect.From the obtained results it is clear that some plant extracts, such as sage or thyme, are able to enhance lysozyme eff ect mainly against Gram--negative strains.These preparations can be used as antimicrobial agents with broader antimicrobial activity applicable in food and cosmetics industries.
The antimicrobial activity of particles containing lysozyme and plant extract was studied aft er 7 and 30 days of storage (Table 10).The results showed that aft er the application of model conditions, partial loss of antimicrobial activity was observed.Antimicrobial tests confi rmed the assumption that free lysozyme in diff erent media over time degrades or denatures and loses more than 50 % of its activity.The obtained results therefore confi rm that free lysozyme is relatively unstable and its encapsulation in combination with plant extract may signifi cantly aff ect its stability.The antimicrobial activity of the prepared liposomes with antimicrobial substances was maintained for a relatively long time.B. subtilis was sensitive to particles containing lysozyme, thyme and sage stored for more than one month, while the antimicrobial activity of these complex extracts against other tested strains decreased to about 50 % of the original activity.Therefore, particles with co-encapsulated herbs and lysozyme are more active against diff erent types of bacteria, and are more stable and more eff ective during long-term storage.
Relatively good stability of particles that have low resistance under certain conditions is a suitable basis for the development of preparations with a targeted transport of antimicrobial substances and their controlled release.An important factor for the maintenance of antimicrobial acti vity is the ability to release antimicrobial agents in a suffi ciently short time and at the required minimum inhibitory concentration.All of the tested particles with coencapsulated herb extracts and lysozyme had very good inhibitory eff ect against all tested strains.Based on the results, it can be concluded that the most appropriate environment for storage of prepared particles is an aqueous medium.The prepared particles are suitable for application in various food industries and also in the form of water-based gels, especially for application in cosmeceuticals (18).

Conclusions
In this study the antimicrobial activities of lysozyme, nisin and extracts of various spices and herbs before and aft er encapsulation were tested.Antimicrobial components were packaged into liposome and polysaccharide particles (alginate, chitosan or starch).Encapsulation of phenolic components was successful with all types of extracts.Encapsulation effi ciency of herbal extracts depended on the particle material and extract composition, while encapsulation effi ciency of lysozyme was relatively good in all particles.All prepared particles exhibited very good colloid stability (zeta potential).Antimicrobial tests (agar well diff usion method and broth dilution assay) were done using two Gram-negative and two Gram-positive bacterial strains.This study demonstrated that some extracts of the tested herbs and spices have a potential for a very good antimicrobial activity against both Gram-po sitive and Gram-negative bacterial strains.Lysozyme and nisin exhibited high antimicrobial eff ect, mainly against the tested Gram-positive bacteria.Chitosan particles with plant extracts also exhibited very high inhibitory eff ect.This could be useful to achieve their combined eff ect; unfortu-nately, chitosan particles were less stable than those of liposomes.Long-term stability and the mass fraction of released components in food and models of physiological conditions were also monitored.Prepared particles could be stored predominantly in an aqueous medium, where they are stable for more than one month and retain their antimicrobial activity.The particles containing encapsulated antimicrobial herb and spice extracts and lysozyme can be used for various food applications.They can also be used as antimicrobial hydrogel formulation containing antimicrobial nanoparticles with controlled release.Antibacterial eff ect of the gel together with the antioxidant activity of herb and spice extracts could be a very promising tool for disinfection or wound healing therapy.Other potential applications of the prepared particles are in food preservation and in pharmaceutical antimicrobial products.

Fig. 2 .
Fig. 2. Colloid stability of liposome with co-encapsulated plant extracts and lysozyme measured as zeta potential

Fig. 3 .
Fig. 3. Long-term stability of co-encapsulated plant extracts and lysozyme in liposomes incubated in diff erent physiological fl uid models: a) released phenolics from liposomes with co-encapsulated plant extracts and lysozyme, b) released lysozyme from particles with mixture of plant extract and lysozyme

Table 1 .
Antioxidant activity (AA), total phenolic content (TPC) and total fl avonoid content (TFC) of tested aqueous extracts of herbs and spices aResults are presented as mean value±standard deviation.Values with the same lett er in the same column are not statistically different (p>0.05).TE=Trolox equivalent, GAE=gallic acid equivalent, CE=catechin equivalent

Table 2 .
Encapsulation effi ciency (EE) of diff erent herb and spice extracts in liposomes expressed in percentage of entrapped phenolic components

Table 3 .
Encapsulation effi ciency (EE) of diff erent herb and spice water extracts in liposomes and in polysaccharide particles expressed in percentage of entrapped phenolic components

Table 4 .
Encapsulation (EE) effi ciency and characterization of particles with lysozyme and nisin encapsulated in liposomes and polysaccharides

Table 5 .
Average value of diameter (d) and zeta potential of liposomes with encapsulated plant antimicrobial component, based on DLS measurement mile (44.1 %) and pot marigold (48.1 %) extracts.In liposomes containing extracts of wormwood, liquorice, pot marigold, sea buckthorn, lavender and garlic, the increase of phenolics was recorded aft er 30 days of storage.Liposomes containing St. John's wort, chamomile and ginger extracts were less stable aft er 30 days of storage in water medium (Table

Table 6 .
Eff ect of temperature on long-term stability of prepared liposomes

Table 7 .
Eff ect of time on long-term stability of prepared liposomes

Table 8 .
Long-term stability of co-encapsulated plant extracts and lysozyme in diff erent food models

Table 9 .
Antimicrobial activity determined by turbidimetric analysis of water extracts of herbs and spices with and without encapsulation in diff erent types of particles, expressed as absorbance (A) at 630 nm Results are expressed as mean values of two incubation experiments (t(incubation)=24 h).Mean values with diff erent lett er in superscript are statistically diff erent (p<0.05).Values with any lett er in superscript have statistically signifi cant antimicrobial eff ect (p<0.05).PE=plant extracts without encapsulation; encapsulation particles: L=liposomes, A=alginate, CH=chitosan

Table 10 .
Stability of antimicrobial activity of liposomes with co-encapsulated lysozyme and plant extracts during long-term storage compared with non-encapsulated lysozyme (γ=1 mg/mL) Antimicrobial activity was measured by well diff usion assay.Inhibition zones represent the average values of two measurements.Mean values with diff erent lett ers in superscript are statistically diff erent (p<0.05)