European Journal of Pharmaceutical Sciences 66 (2015) 20–28

Contents lists available at ScienceDirect

European Journal of Pharmaceutical Sciences

journal homepage: www.elsevier .com/ locate /e jps

Spray-by-spray in situ cross-linking alginate hydrogels delivering a teatree oil microemulsion

http://dx.doi.org/10.1016/j.ejps.2014.09.0180928-0987/� 2014 Elsevier B.V. All rights reserved.

⇑ Corresponding authors.E-mail addresses: quaglia@unina.it (F. Quaglia), paola.laurienzo@ipcb.cnr.it

(P. Laurienzo).1 These authors contributed equally to the work.

O. Catanzano a,1, M.C. Straccia b,1, A. Miro a, F. Ungaro a, I. Romano c, G. Mazzarella d, G. Santagata b,F. Quaglia a,⇑, P. Laurienzo b,⇑, M. Malinconico b

a Drug Delivery Laboratory, Department of Pharmacy, University of Napoli Federico II, Via D. Montesano 49, 80131 Naples, Italyb Institute for Polymers, Composites and Biomaterials (IPCB), CNR, Via Campi Flegrei 34, 80078 Pozzuoli (Naples), Italyc Institute of Biomolecular Chemistry (ICB), CNR, Via Campi Flegrei 34, 80078 Pozzuoli (Naples), Italyd Institute of Food Science (ISA), CNR, Via Roma 64, 83100 Avellino, Italy

a r t i c l e i n f o a b s t r a c t

Article history:Received 5 August 2014Received in revised form 19 September2014Accepted 21 September 2014Available online 30 September 2014

Keywords:MicroemulsionsAlginateSpray-by-sprayTea tree oilAdvanced dressing

In this paper we propose an in situ forming ionically cross-linked alginate (Alg) hydrogel delivering a TeaTree Oil microemulsion (MeTTO) and potentially useful as an advanced dressing for infected wounds. Alghydrogels were prepared by a spray-by-spray deposition method with the aim to minimize the discom-forts during application. From pseudoternary phase diagrams, it was found that proper combination ofTTO, water, polysorbate 80 and ethanol gave stable spherical MeTTO with good antimicrobial activity.On this basis, MeTTO at 20% TTO was selected for further inclusion in an Alg hydrogel prepared by alter-nating sprays of Alg/MeTTO and calcium chloride solutions. Homogeneous dispersion of MeTTO insidecross-linked Alg was assessed by different macroscopic and microscopic methods demonstrating thesuperior propensity of MeTTO to be integrated in the water-based hydrogel as compared to TTO. Antimi-crobial effect of Alg/MeTTO hydrogels on Escherichia Coli strains was remarkable, highlighting the potentialof the system as bioactive wound dressing.

� 2014 Elsevier B.V. All rights reserved.

1. Introduction

Hydrogels are a network of polymers filled with water that maybe applied to absorb wound exudates and to protect wounds fromsecondary infection. Hydrogel dressings can be applied to thewound either as pre-formed solid gels or as liquids that crosslinkafter application. An in situ–forming hydrogel (ISH) is initially fluidat room temperature but becomes a hydrogel in situ due to specificconditions, including ionic cross-linking, pH, or temperaturechange (Ruel-Gariepy and Leroux, 2004). ISHs have some meritsover traditional dressings, including conformability without wrin-kling or fluting of the wound bed, ease of application, good patientcompliance, and comfort. Thus, ISHs are an excellent option asmultifunctional wound dressings suitable for superficial wounds,burns, skin grafts and abrasions that cover a large area.

Alginates (Alg) are a family of polyanionic copolymers derivedmainly from brown sea algae (Laurienzo, 2010), and are of growingimportance in the healthcare and pharmaceutical industry

(Boateng et al., 2008). The ability of Alg to form crosslinks in pres-ence of divalent ions has allowed the development of biocompati-ble hydrogels that helps to maintain the lesion at an optimummoisture content and healing temperature (Peng et al., 2012). Alg,in combination with other biopolymers or active agents, is widelyused as base material in film dressings to improve wound healingrate and to prevent burn infection (Brachkova et al., 2011; Dantaset al., 2011). ISHs based on polysaccharides, obtained by alternatingspray of polyanionic and polycationic polymers, are gaining interestin wound management since they allow quick formation of highlyuniform thin films over a large surface area (Cado et al., 2012;Schlenoff et al., 2000). Alternate polysaccharide deposition on asolid substrate represents an appealing option to give ISHs also inview of simple industrial scale-up (Schaaf et al., 2012).

Integration of bioactive molecules in ISH is rather simple forhydrophilic compounds but much more challenging for hydropho-bic or oily actives. Amid actives with a high potential in wound care,Tea Tree Oil (TTO), a natural essential oil steam-distilled from theAustralian native plant Melaleuca alternifolia, is gaining aconsiderable relevance. In fact, TTO is a very promising agent forthe treatment of dermatologic diseases due to its antimicrobialeffects against a wide spectrum of microorganisms (Carson et al.,2006; Pazyar et al., 2013) and its minimal impact on developing

O. Catanzano et al. / European Journal of Pharmaceutical Sciences 66 (2015) 20–28 21

resistance (Hammer et al., 2012). Furthermore, it has potent activityagainst many fungi (Hammer et al., 2003, 2004), protozoa (Carsonet al., 2006; Mikus et al., 2000), and certain viruses, including herpessimplex and influenza viruses (Carson and Riley, 2001; Garozzoet al., 2011). Besides the well-known antimicrobial activities, TTOhas been shown to possess a number of other therapeutic properties,including anti-inflammatory (Hart et al., 2000; Koh et al., 2002;Pearce et al., 2005) and anti-tumor properties (Bozzuto et al.,2011), especially in skin cancer (Greay et al., 2010; Ireland et al.,2012). Hydrogels containing TTO were already tested in a burnwound model and suggested to increase the rate of wound healing(Jandera et al., 2000). Furthermore, it was recently shown that TTOaccelerated wound healing in humans (Chin and Cordell, 2013). Asall essential oils, TTO is a lipophilic liquid that shows poor miscibilitywith water-based products, while its low surface tension demandsfor methods able to provide an adequate barrier to volatilization.

Microemulsions (MEs) have recently emerged as novel deliveryvehicles for hydrophobic drugs by different administration routes(Fanun, 2012). MEs appears as an attractive and competitivesystem due to many benefits such as easy manufacturing, smalldroplet size (20–200 nm), high thermodynamic stability, andenhanced solubilization of hydrophobic ingredients. In recentyears, MEs have been investigated as potential drug deliveryvehicles for transdermal and dermal delivery of several com-pounds, especially hydrophobic, in order to avoid clinical adverseeffects associated with oral administration (Shakeel et al., 2012).Furthermore, MEs can be easily incorporated into Alg hydrogelsto create composite hydrogels with a controlled release profile(Josef et al., 2013, 2010). Formation of MEs can represent a viableand efficient approach also to increase physical stability ofessential oils, protecting them from undesired interactions andincreasing their bioactivity. As demonstrated by Donsi et al.(2011, 2012), MEs can enhance the antimicrobial activity ofencapsulated essential oils increasing their water solubility andthe consequent capacity to interact with cell membranes.

The objective of the present study was to developantimicrobial in situ-forming Alg wound dressings incorporatingTTO microemulsions (MeTTO). Alg hydrogels were prepared bya layer-by-layer spray deposition method with the aim tominimize the discomforts, especially during dressing application.After a thorough characterization of MeTTO and MeTTO-loaded Alghydrogels, their antimicrobial effect was tested.

2. Experimental

2.1. Materials

Pharmaceutical grade alginic acid sodium salt (Alg) extractedfrom Laminaria hyperborea (viscosity 360 cps) and Tea Tree Oil(TTO) were supplied by Farmalabor (Italy). Polysorbate 80 (Tween�

80), calcium chloride dihydrate (CaCl2 ⁄ 2H2O), sodium chloride(NaCl), potassium chloride (KCl), sodium phosphate dibasic (Na2-

HPO4), calcium chloride (CaCl2), and nile red (NR) were obtained fromSigma–Aldrich (USA). Analysis-grade ethanol was provided by CarloErba Reagenti (Italy). Yeast extract, Tryptone and Sodium Chloride,used for prepare Luria–Bertani broth (LB) medium, and Agar bacte-riological were purchased from Oxoid Ltd. (Hampshire, UK). Strep-tomycin sulfate was supplied by Applichem (Germany). The spraypump used for the spray deposition (Classic line equipped with SLpump) was kindly gifted by Aptar Pharma (France). The water usedthroughout this study was depurated and filtered (Milli Q filter).

2.2. Pseudoternary phase diagrams

The boundaries of the ME domains were determined, withthe aid of pseudo ternary phase diagrams, for the polysorbate

80:ethanol (surfactant:co-surfactant), TTO and water phase. Thetitration method was employed for the construction of phase dia-grams. Different mixtures of polysorbate 80 and ethanol (1:1, 2:1and 3:1) were weighed in a dark-brown, screw-cap glass vial, mixedusing a magnetic bar on a stirring plate for 1 h and subsequentlystored overnight at room temperature. TTO was then added at ratiosranging from 9:1 to 1:9 to different vials. Finally, aqueous phase wasslowly added (under vigorous stirring) with a graduated syringe upto clouding of homogenous mixture of oil and surfactants/co-surfactants. Approximately 10 data points were obtained todetermine each pseudoternary phase diagram. No attempts weremade to completely identify the other regions of the phase diagramsin detail, and these have been described in terms of their visual andexternal appearance. To graphically show the phase variations ofpolysorbate/ethanol–water–TTO system, a pseudo-ternary phasediagram was built using Microsoft excel software.

2.3. Preparation of microemulsions

Once the MEs region was identified, TTO was dissolved in a sur-factant/co-surfactant mixture previously prepared using a mag-netic stirring plate. An oil-in-water ME was prepared by slowlyadding water to the oily phase (oil plus surfactants) under contin-uous magnetic stirring. To prepare the NR-loaded ME, the probewas dissolved directly into TTO at the concentration of 20 lg/mL.All the formulations were prepared at room temperature andtested after 24 h. Resulting MEs were visually inspected and tur-bidity was measured using a spectrophotometer.

2.4. Characterization of microemulsions

Droplet size and polydispersity index (PDI) of MEs was mea-sured by Dynamic Light Scattering (DLS) using a Zetasizer NanoZs(Malvern instruments, UK). MEs were diluted 1:10 with deionizedwater prior the experiment to avoid the effect of viscosity and totrim down multiple scattering effects. Particle size and PDImeasurements were performed at a scattering angle of 90� andat a temperature of 25 �C.

ME droplet morphology was visualized by TransmissionElectron Microscopy (TEM). Samples for TEM analysis were pre-pared by placing one drop of MEs onto a copper grid. After approxi-mately 1 h, images were captured. TEM analysis was performedwith a FEI Tecnai G12 (LAB6 source) equipped with a FEI Eagle 4 KCCD camera (USA) operating with an acceleration voltage of 120 kV.

Viscosity of the formulated MEs was measured using a Brook-field Viscometer (model LVF 69726) supplied with UL-adapter.All the experiments were performed in triplicates at 25� C.

Turbidity of the formulated MEs was evaluated on a UV/VISspectrophotometer (UV 1800, Shimadzu, Japan) at a wavelengthof 502 nm, fitted out with a 1 cm quartz cell (Hellma, Germany).The turbidity was calculated as turbidity � path length = 2.303 �absorbance (Fletcher and Morris, 1995).

The pH of each formulation was evaluate using a Crison basic 20pHmeter equipped with an electrode 50 10T (Crison, Spain).

2.5. Microemulsion stability studies

2.5.1. Centrifugation studyThe formulated MEs were studied for their stability to centrifu-

gation by placing the sample at 13,000 rpm for 30 min (Mikro 20centrifuge, Hettich, Germany) and observing phase separation,creaming or cracking (if any).

2.5.2. Heating–cooling cycleThis study was performed to check the effect of temperature

variations on the stability of MEs. Samples were stored between

22 O. Catanzano et al. / European Journal of Pharmaceutical Sciences 66 (2015) 20–28

4 and 40 �C for a period of 48 h at each temperature. The heating–cooling cycle was repeated four times. MEs that did not showinstability such as cracking, creaming and phase separation werechosen and subjected to freeze–thaw stress cycles.

2.5.3. Freeze–thaw cycleMEs formulations were subjected to freeze–thaw stress between

�21 and +25 �C, with storage at each temperature for a minimum of48 h. Three freeze–thaw cycles were performed and the formula-tions that resulted to be stable were selected for application studies.

2.5.4. Kinetics stabilityMEs were kept at room temperature for checking out the intrin-

sic stability. The emulsion formulations were observed for phaseseparation, creaming and cracking with respect to prolonged stor-age time period. Kinetic stability was investigated by measuringdroplet size of the MEs at different intervals of time.

2.6. Preparation and characterization of Alg hydrogels

2.6.1. Stability of MEs in Alg solutionThe stability of MEs in Alg solution was assessed by checking

the variation of their size by DLS. MeTTO20 was dispersed in 1%w/v Alg aqueous solution, at a 20% v/v concentration. Before mea-surements, the Alg/ME solution was diluted 1:10 with distilledwater. Mean size and polydispersity index were analyzed at differ-ent time intervals.

2.6.2. Spray-by-spray hydrogel formationHydrogels were obtained by a spray-by-spray crosslinking

method (SbS). The selected pharmaceutical grade spray pump dis-penses exactly 140 lL of an Alg solution at 1% w/v at each actua-tion. The SbS procedure was standardized as follows: a 1% (w/v)Alg aqueous solution was sprayed twice on a lightly wet glass petridish (36 mm diameter), followed by a single spray of a 2% w/vCaCl2 aqueous gelling solution. A thin layer of hydrogel formedimmediately. Three layers were deposited one above the othersfor each preparation. As concerning the preparation of hydrogelsincluding MEs, only the MeTTO20 formulation was selected andincorporated in Alg hydrogel by pre-dispersing 20% (v/v) ofMeTTO20 in Alg solution (1% w/v) under magnetic stirring.

2.6.3. Hydrogel characterizationThe bulk morphology of Alg/ME hydrogels was examined

through Cryo-Scanning Electron Microscopy (Cryo-SEM) using acryo-system Gatan Alto 1000 installed on a FEI Quanta 200 FEGSEM. The sample was placed on the holder, mounted on thecryo-transfer rod, slam-frozen in nitrogen slush and transferredto the cryo-chamber, where it was cryo-fractured and sputtercoated with gold/palladium. Sample was finally moved to theSEM chamber and the fracture surface was analyzed at �140 �C,using an acceleration voltage of 5–10 kV.

Homogeneity of SbS hydrogels was determined by the dry to wetweight ratio method (Alexander et al., 2011; Kuo and Ma, 2001).Hydrogels were cut in four slices (labelled a-d) and weighed on an ana-lytical balance. The specimens were dried to constant weight (60� C for24 h under vacuum) and weighed again. The calculated dry to wetweight ratio provides an indication of gel homogeneity. The reporteddata are the average of three samples ± standard deviations.

Water content was determined gravimetrically on hydrogelsobtained by spray deposition. The initial weight of each samplewas recorded using an analytical balance. Successively, thehydrogels were frozen overnight at �40 �C, and then lyophilizedat 0.01 atm and �60 �C in a Modulyo apparatus (Edwards, UK).After freeze-drying the samples were reweighed and the watercontent (W%) was calculated using the equation:

W% ¼W �W0

W0� 100 ð1Þ

where W is the mass of the freeze-dried sample and W0 is the massof the initial wet sample. The weight loss, expressed as %, corre-sponds to the water content.

Degradation of hydrogels containing MEs in a simulated biolog-ical fluid was visually followed through a modified Enslin appara-tus designed to mimic the wound bed (Rossi et al., 2007).Hydrogels were placed on a porous glass filter which is in contactwith an hydration medium, to simulate biological fluids in tissueinjuries. Phosphate Buffer Saline (PBS, NaCl 120 mM, KCl 2.7 mM,Na2HPO4 10 mM, pH 7.4) and citrate buffer solution (citric acidmonohydrate, 0.1 M; trisodium citrate dihydrate, 0.1 M; pH 4.8)were employed. All the experiments were performed in triplicate.

2.7. Confocal laser scanning microscopy (CLSM)

The distribution of fluorescently labelled MEs within the Alghydrogel was examined by confocal laser scanning microscopy(CLSM). An hydrophobic fluorescent probe, Nile Red (NR), was dis-solved in TTO (20 lg/mL) and used to stain MEs (coded MeTTO20-NR). The Alg/MeTTO20-NR hydrogel was deposited by SbS proceduredirectly on microscope slide and imaged by CLSM (Leica TCS-SP,Germany) using a 20� Plan-Neofluar objective. The illuminationsource was from a HeNe ion laser (543 nm).

2.8. Antibacterial assay

The minimum inhibitory concentrations (MIC) of MEs weredetermined using the broth dilution methods (Amsterdam, 1996;Wiegand et al., 2008) against an E. Coli wild type (DSM 478) pur-chased from Deutsche Samlug von Mikroorganismen und Zellk-ulturen GmbH (DSMZ Germany). The bacterial strain was grownin Luria–Bertani broth (LB) medium, in aerated incubator, at37 �C, for 18–24 h. MEs were serially diluted with sterile LB med-ium: 1 mL of ME was added to LB, mixed by vortexing for 1 min,in order to obtain a concentration range spanning from 50% to0.39%. One mL of each diluted ME was inoculated with 10 lL of107 CFU/mL (Colony Forming Units/mL) microbial suspension andincubated at 37 �C for 24 h. Inoculum assay and streptomycin sul-fate were evaluated as negative and positive controls, respectively.The MIC value was defined as the lowest concentration of ME thatinhibited the visible growth of the bacterium and was verifiedmeasuring the optical density of bacteria at 600 nm using a spec-trophotometer. Each measurement was replicated three times.

Antibacterial activity of the Alg/MeTTO20 hydrogel was evaluatedfollowing the ASTM standard test method E2149-01 (ASTM, 2001)used in case of immobilized antimicrobial agents, under dynamiccontact condition. The bacterial concentration was determinedspectrophotometrically by measuring the optical density at600 nm. The bacteria were grown in LB-medium for 18 h at 37 �C,to obtain isolated colonies. After growing and harvesting, bacterialcells were washed with Ringer Solution (RS), a sterile buffer salinesolution at pH 7, (0.150 g KCl, 2.25 g NaCl, 0.05 g NaHCO3 and 0.12 gCaCl2, per liter of solution). The bacterial cells were diluted in RS toobtain assays inoculum of about 105 CFU/mL. Alg/MeTTO20 samplewas added to 10 mL of bacterial suspension for testing antibacterialactivity, whereas Alg hydrogel was used as negative control. Afterfixed contact times (0, 3, 6 and 12 h) surviving cells were evaluatedby a standard plate count method: 100 lL of the diluted bacterialsuspensions were spread on solid medium in petri dishes andincubated at 37 �C for 24 h. The average colony count of duplicateplates was used to calculate the CFU/mL. Inoculum assay was usedas control. All experiments were performed in triplicate andaverage data were used for calculations. The percentage reductionwas calculated using the following formula:

O. Catanzano et al. / European Journal of Pharmaceutical Sciences 66 (2015) 20–28 23

Reduction % ðCFU=mLÞ ¼ B� AB� 100 ð2Þ

where A is the CFU/mL for the corning flask containing the sampleafter specified contact time and B represents the CFU/mL at thebeginning of the experiment.

3. Results and discussion

3.1. Phase diagrams

Polysorbate 80 is a non-ionic surfactant (HLB = 15) giving MEswithout the aid of a co-surfactant only in a very limited range ofconcentrations. In preliminary experiments we found that themaximum amount of TTO that could be incorporated in the MEswas 10% (w/w) with a total surfactant content of 80–85% (w/w)(data not shown). Indeed, the use of a non-ionic surfactant alonehampers the spontaneous formation of the zero mean curvaturelipid layer, necessary for ME formation (Bhargava et al., 1987).Thus, to achieve the ultra-low interfacial tension necessary forthe formation of the small ME droplets, a high amount of surfac-tant is required. To extend ME area, a co-surfactant, such as a shortchain alcohol, has been traditionally employed because alcoholsare able to increase the hydrophilic–lipophilic balance of polysor-bate 80 by decreasing the hydrophilicity of the polar solvent

Fig. 1. Phase diagrams obtained at different S:CoS volume ratios. W = water phase, O = othe water content). Shaded area refers to regions where a stable ME is formed.

Fig. 2. Transmission electron micrographs

(external aqueous phase) (Moreno et al., 2003). The presence of ashort chain co-surfactant imparts sufficient flexibility to the inter-facial film which becomes able to adopt the different curvaturesrequired to form MEs.

Pseudo-ternary phase diagrams composed of polysorbate 80 (S),ethanol (used as co-surfactant, CoS), TTO (termed the oily phase)and water (W) were constructed at room temperature to showthe relationship between composition and phase behavior of thesamples (Fig. 1).

The construction of phase diagrams allowed the selection ofoptimized concentration ratio of components to form MEs. In allthe cases, the areas of stable ME formation extended over a moreor less limited zone in the S:CoS-rich part of the phase diagram.At increasing S:CoS ratio, a smaller zone of stable ME arose dueto the effect of ethanol on interface fluidity and consequentlypolysorbate 80 hydrophilic–lipophilic balance (HLB). HLBdescribes the simultaneous attraction of the surfactant mixturefor the oil and aqueous phases, so when it is similar to therequired HLB of the ME oily phase, the system provides the min-imum energy condition for ME formation. The polysorbate80:ethanol 1:1 (v/v) ratio gave stable MEs with the largest MEformation area. In this particular region four formulations wereselected in order to obtain oil-in-water systems. The remainingregion of the phase diagram allowed the formation of conven-tional emulsions. The red dots shown in Fig. 2 correspond to

il phase (TTO), S:CoS = Surfactant:Co-Surfactant (top apex of the triangle represents

of MeTTO20 at different magnifications.

Table 1Properties of MEs at different TTO loading.

Formulation W/S:CoS (%) TTO (%) q 25 �C g/mLa Size (nm ± D.S.)b PDI Turbidity (%) Viscosity (cps) pH

MeTTO20 30/50 20 0.989 21.95 ± 0.20 0.232 1.74 45.50 ± 0.71 5.6 ± 0.1MeTTO15 35/50 15 0.975 18.31 ± 0.67 0.209 3.39 55.50 ± 0.58 5.7 ± 0.1MeTTO10 40/50 10 0.963 14.06 ± 0.33 0.206 2.82 76.50 ± 0.71 5.8 ± 0.1MeTTO5 45/50 5 0.934 11.92 ± 0.04 0.211 3.56 102.00 ± 0.40 5.9 ± 0.1

a q 25 �C for TTO is 0.889 g/mL.b After 1:10 dilution in water.

24 O. Catanzano et al. / European Journal of Pharmaceutical Sciences 66 (2015) 20–28

the specific compositions of the formulations at different TTOconcentrations. The data of the pseudo-ternary phase diagramshow that it is possible to add a large volume of aqueous phaseand a limited volume of the oily phase maintaining the thermo-dynamic stability of the systems.

3.2. Characterization and stability of MEs

Based on the phase diagrams, four MEs were selected for furtherinvestigation differing in the amount of TTO included in the inter-nal phase. The properties of the selected MEs are shown in Table 1.

Dynamic Light Scattering (DLS) measurements were performedon MEs, which were diluted with aqueous buffer in order to avoidexperimental errors due to short interparticle space between glob-ules (Orthaber and Glatter, 2000).

TTO-loaded MEs (MeTTO) showed a very small particle size andnarrow size distribution. The final TTO concentration influencedthe average size, which increased as TTO amount increased. Lowpolydispersity (�0.2) was indicative that homogeneous monodi-spersed particles were formed. The pH values were within therange accepted for skin application. Morphology of MEs was visu-alized by TEM (Fig. 2). Emulsion droplets were spherical and thedroplet size was confirmed to fall within the range found with DLS.

Determination of ME viscosity is necessary to characterize thesystem and to control its physical stability (Cilek et al., 2006). Asreported in Table 1, low viscosity values at room temperature werefound for all TTO-loaded MEs with a trend to increase as TTO con-tent decreases. Low viscosity could be very useful to obtain an effi-cient nebulization from a spray device.

MEs are thermodynamically stable by definition but even soaccelerated testing methods are required for evaluating long termstability. Thermodynamic stability of ME formulations prepared atdifferent oil concentrations were investigated by visual inspectionand size determination after centrifugation, freeze–thaw and heat-ing–cooling cycles. The centrifugation test allowed the evaluationof the absence of phase separation under the effect of mechanicalstress. All MEs were found to be stable to centrifugation, as nochange in any of their properties was observed. ME stability to

0 1 2 3 4 5 60

5

10

15

20

25

Hyd

rody

nam

ic d

iam

eter

(nm

)

Time (Months)

MeTTO20 MeTTO15 MeTTO10 MeTTO5

Fig. 3. Kinetic stability of MeTTO stored at room temperature.

temperature was observed after rapid temperature changes. Allthe systems undergoing freeze–thawing and heating–coolingcycles, remained stable since no significant change in droplet sizewere observed after the thermal treatments (data not shown).Kinetic stability of MEs was studied by storing the formulationfor 6 months at room temperature. During this period the sizeremained constant and no sign of phase separation was found(Fig. 3).

3.3. Determination of MIC values of MEs

Due to the their subcellular size, nanometric delivery systemsmay increase the passive cellular absorption reducing mass trans-fer resistance and increasing antimicrobial activity, especially inmicroorganism with a thicker cell wall structure, characteristic ofgram-positive bacteria (Donsi et al., 2011). Thus, it was worth toevaluate the in vitro antimicrobial activity of MeTTO formulationsagainst E. Coli, a model bacterium strain sensitive to TTO (Carsonet al., 2006). The MIC values of TTO encapsulated in the variousMEs are reported in Table 2.

Results evidenced that MeTTO exhibited antibacterial activityhigher as compared to TTO and that, as expected, MIC decreasedas TTO content in the ME increased. The increase of antimicrobialactivity was probably due to the enhanced TTO transport insidebacteria. Nevertheless, TTO can alter membrane properties bydecreasing polarity and increasing permeability in a time- and con-centration-dependent manner (Hammer and Heel, 2012). Since theantimicrobial activity of lipophilic compounds is severely limitedby poor water solubility, it can be hypothesized that an increaseof equilibrium concentration of TTO components in water signifi-cantly contributes to decrease MIC. MEs did not lose theirefficiency to inactivate bacteria upon dilution, suggesting that norelevant change of their properties occurred, as previouslyreported in the literature (Ghosh et al., 2014). On this basis,MeTTO20 showing the lowest MIC value was selected for inclusionin spray hydrogels.

3.4. Stability of ME in Alg solution

Incorporation of lipophilic liquids, such as TTO, in a hydrophilicplatform is a challenging task. To obtain Alg/ME hydrogels,different amounts of MeTTO20 were dispersed in a 1% (w/v) Algaqueous solution, and then the solution was sprayed and cross-linked by CaCl2 as described in the experimental part. A 20% (v/

Table 2MIC values of METTO

⁄.

MIC (mg/mL)

MeTTO20 3.092MeTTO15 4.570MeTTO10 6.019MeTTO5 5.838TTO 6.943

⁄ Positive control was streptomycin sulfate. MIC = 15 lg/mL.

Fig. 4. Size of MeTTO20 diluted in Alg solution. (A) MeTTO20 in water (Red), MeTTO20 in Alg aqueous solution (Green). (B) MeTTO20 in Alg aqueous solution at t = 0 (Red), t = 2 days(Green), t = 7 days (Blue). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

O. Catanzano et al. / European Journal of Pharmaceutical Sciences 66 (2015) 20–28 25

v) mixture of MeTTO20 in Alg aqueous solution was clear, with noevident sign of phase separation. Since dilution of MeTTO20 in theAlg solution could change or break its structure, the presence ofintact MEs in the Alg solution was verified by DLS. As shown inFig. 4A, only a slight difference in ME size was detected (the meansize changes from 20 to 26 nm). These data are in agreement withthose reported by Josef et al. (2010) and suggest that the structureof the ME is unchanged upon mixing with Alg. The stability ofMeTTO20 in Alg solution with time was checked (Fig. 4B). After7 days, the mean size was reduced to 18.53 nm and the polydisper-sity index showed only a small increase (0.358). After 2 weeks(curve not reported) no further changes in dimension weredetected, proving a good shelf life of MeTTO20 in Alg solution.

3.5. Preparation and characterization of SbS hydrogels

ISH were formed by alternate spraying of Alg and CaCl2 solu-tions on a glass support. The SbS procedure was standardized as

Fig. 5. Hydrogels obtained by SbS. (A) Alg; (B) Alg/MeTTO20; (C) Alg/NR-TTO; (D)Alg/MeTTO20-NR.

a function of two parameters: (i) Alg and CaCl2 concentration inwater, and (ii) number of relative Alg/CaCl2 pump actuations. Algconcentrations higher than 1% w/v were not suitable for the SbStechnique due to their high viscosity. Different CaCl2 concentra-tions were tested (1%, 2%, 4%, 6%, 8%), being 2% the most suitableto give adequate gelation time and homogeneous gels. As a matterof fact, 1% CaCl2 was not enough to give a hydrogel, while at con-centration >2% a too fast gelation occurred causing surface shrink-ing. Furthermore, a too rapid gelation generated an irregularsurface upon which successive spray layers gave rise to a stratified,not homogeneous final hydrogel. The optimal conditions were setas follows: two sprays of 1% Alg followed by one spray of 2% CaCl2.The number of consecutive layers was set to three for further

Fig. 6. LSCM images of: (A) TTO-NR in Alg solution; (B) MeTTO20-NR in Alg solution;(C) TTO-NR in Alg SbS hydrogel; (D) MeTTO20-NR in Alg SbS hydrogel.

26 O. Catanzano et al. / European Journal of Pharmaceutical Sciences 66 (2015) 20–28

experiment. The gels were let to complete gelation for 30 minbefore further characterization.

The SbS technique allows to realize thin coatings as well asthick hydrogels by regulating the cumulative number of pumpactuations. SbS formation of wound dressing is fast, clean, ‘‘touch-less’’ (hands free) and simple. Application is feasible on any shapeor size of wound using the same device, thereby reducing the needto prepare several pre-formed size wound dressings. Because itforms in situ, the dressing easy adapts to the wound bed conforma-tion ensuring that the active agents are more efficiently deliveredalso to cavernous wounds. Meanwhile, penetration of the hydrogelinto the wound bed may aid in debridement of the bed duringdressing changes without removal of epithelial cells, which isknown to accelerate the wound healing process (Boateng et al.,2008). Another advantage of the spray-by-spray technique is thepossibilities of easily prepare sterile hydrogels starting from twosterile solutions. A sterile calcium chloride solution can be easilyprepared using classic sterilization methods while a sterile alginatesolution can be prepared from a commercially available sterilealginate powder. Filling of spray devices can be carried out in aclass A room.

Fig. 7. Cryo-SEM micrographs of SbS hydrogels i

Hydrogels obtained by SbS showed an irregular top surface buta quite regular thickness and diameter (Fig. 5). Transparency wasdrastically reduced as a consequence of the spray procedure used,that implies entrapment of air bubbles inside the gel. Nevertheless,it was still possible look through the gel in order to inspect thewound bed (Fig. 5A and B). Structural uniformity is crucial for bio-medical applications not only to achieve distribution of the deliv-ered active in the matrix, but especially to get constantproperties in the entire hydrogel. A nile red loaded MeTTO20

(MeTTO20-NR) was prepared and used to evaluate the ME distribu-tion inside the Alg hydrogel (Fig. 5C). As a comparison, NR was dis-solved in TTO (NR-TTO) at the same concentration used in MeTTO20-NR preparation (20 lg/mL) (Fig. 5D). From a macroscopic point ofview, despite the same amount of NR present in the hydrogel, adifferent color of NR-TTO and MeTTO20-NR loaded hydrogel wasobserved, probably due to a finer dispersion of the probe whichallows greater transparency to light.

CLSM was used to investigate the distribution of MEs within thehydrogel on a micrometric scale. NR-TTO dispersed in the Alg solu-tion at 20% (v/v) formed irregular large droplets comprising in avery large size range (Fig. 6A) and well separated from the Alg

ncluding MeTTO20 at different magnification.

ALG/ME TTO20

0 2 4 6 8 100.0

2.0x104

4.0x104

6.0x104

8.0x104

1.0x105

1.2x105

1.4x105

CFU

/mL

Time (hours)

ALG

Fig. 9. Time kill curve of Alg and Alg/MeTTO20 SbS hydrogels.

O. Catanzano et al. / European Journal of Pharmaceutical Sciences 66 (2015) 20–28 27

matrix (black area). A nanometric distribution of TTO in Alg solu-tion (Fig. 6B) was evident in the case of MeTTO20. The distributionof TTO-NR and MeTTO-NR in Alg hydrogel obtained by spray depo-sition (single layer) advantageously demonstrated a rough distri-bution with the persistence of large TTO droplets in case of NR-TTO dispersed in Ca2+ cross-linked Alg (Fig. 6C), while using theMeTTO20-NR a homogeneous distribution was attained (Fig. 6D).

3.6. Hydrogel characterization

3.6.1. Morphology, water content and homogeneityIn Fig. 7, cryo-SEM micrographs of hydrogels including MeTTO20

are reported. As shown in Fig. 7A and B, the Alg network appearedregular and the partial removal of surface water allowed to exposethe porous structure. The pore walls were thin, and the pore sizeranged from few micron to 18 lm, with an estimated averagevalue of around 10 lm. Micrographs taken at higher magnificationare shown in Fig. 7C and D. The presence of numerous pores of sub-micron dimension is likely related to the spray technique whilenanosized aggregates are evident. Finally, Fig. 7E show the imprint-ing of a trapped air bubble of around 100 lm in size.

Weight, thickness, diameter and mean water content of freshly-prepared hydrogels are reported in Table 3. The decrease of watercontent in ME-loaded hydrogel was consistent with the reducedwater content of the initial Alg solution.

Homogeneity of Alg and Alg/MeTTO20 SbS hydrogel was esti-mated by the dry/wet weight ratio method. A constant weight ratioacross the constituent slices was indicative of a homogeneoushydrogel. No relevant variation between the slices was observed(Fig. 8), indicating that the spray-deposition process here describedcan be applied to produce homogeneous Alg hydrogels suitable forwound dressing. Moreover, the presence of MEs did not signifi-cantly affect gel homogeneity in the concentration range studied.

3.6.2. SbS hydrogel stabilityIn order to ensure a prolonged and durable protection of the

wounds, hydrogels are expected to be stable when in contact withthe wound bed for a reasonable time, in order to avoid frequent

Table 3Properties of Alg and Alg/MeTTO20 SbS hydrogels (±SD of five specimen).

Hydrogels Weight (g) Thickness(mm)

Diameter(cm)

Water content(%)

Alg 1.250 ± 0.025 1.41 ± 0.43 2.53 ± 0.05 72.56 ± 2.35Alg/MeTTO20 1.154 ± 0.05 0.96 ± 0.12 2.73 ± 0.06 61.14 ± 1.29

Fig. 8. Dry/wet weight ratios of Alg and Alg/MeTTO20 SbS hydrogel slices accordingto the schematic representation.

dressing change. Using a modified Enslin apparatus to simulate awound bed, the degradation, and consequently the residence time,of cross-linked gels on wound could be estimated by visual inspec-tion. As the exudate pH in chronic wounds is reported to fall in the4.8–9.8 pH range (Britland et al., 2011), solutions with different pH,namely 4.8 and 7.2, were tested. Hydrogels were placed in contactwith the solution only by the lower surface to better mimic thewound environment, and allow the exudate penetration intohydrogels. Since Alg presents resistance to degradation in a prote-olytic environment, as in the case of chronic wound, physical deg-radation is due mainly to ions exchange. Hydrogels swelled in thefirst 24 h and began to dwindle gradually only after 48 h, givingrise to loss of material. No significant differences were appreciatedat different pH values.

3.6.3. Antibacterial activity of Alg/MeTTO20 SbS hydrogelThe antibacterial activities of Alg and Alg/MeTTO20 SbS hydrogels

were carried out against an E. Coli wild type strain. Results (Fig. 9)showed that Alg/MeTTO20 SbS hydrogel caused a strong bacterialinactivation already after 6 h, while a complete killing of bacteriawas reached after about 12 h. The control hydrogel (Alg) did notcause any detectable microbiological inactivation after 12 h.

4. Conclusions

TTO, a natural compound with well recognized antimicrobialand anti-inflammatory properties, was formulated as microemul-sion and incorporated within an in situ forming cross-linked Alghydrogel. In a first part of the study, microemulsions comprisingTTO, water and polysorbate 80/ethanol with good thermodynamicand kinetic stability, as well as improved antibacterial activitywere developed. In a second part, the most promising microemul-sion formulation was incorporated in cross-linked Alg hydrogels bya newly developed spray-by-spray procedure. In specific condi-tions, the spray technique allowed formation of transparent andregular hydrogels where TTO microemulsion was homogeneouslydistributed. Finally, antibacterial assays evidenced a total inhibi-tion of E. Coli proliferation after 12 h. All together, these findingssuggest that spray-by-spray procedure represents a novel optionto obtain in situ forming hydrogels delivering microemulsions.Application of this concept to TTO can hold great potential in themanagement of different types of wounds.

Acknowledgments

The authors gratefully acknowledge the project DIATEME, in theframe of National Operative Program (PON 2007-2013), for finan-cial support. The authors wish to thank Mrs. Maria Cristina Del Bar-one from Laboratory of Electron Microscopy ‘‘LaMEST’’ of IPCB fortechnical assistance in cryo-SEM analysis, and Mr. Clemente Mec-

28 O. Catanzano et al. / European Journal of Pharmaceutical Sciences 66 (2015) 20–28

cariello from ISA for technical assistance in CLSM analysis. Theauthors are also indebted to dr. Gennaro Gentile from IPCB for sci-entific assistance in cryo-SEM analysis.

References

Alexander, B.R., Murphy, K.E., Gallagher, J., Farrell, G.F., Taggart, G., 2011. Gelationtime, homogeneity, and rupture testing of alginate-calcium carbonate-hydrogen peroxide gels for use as wound dressings. J. Biomed. Mater. Res. BAppl. Biomater.

Amsterdam, D., 1996. In: Loman, V. (Ed.), Susceptibility Testing of Antimicrobials inLiquid Media. Williams and Wilkins, Baltimore, MD, pp. 52–111.

Bhargava, H.N., Narurkar, A., Lieb, L.M., 1987. Using microemulsions for drugdelivery. Pharm. Tech. 11, 46–52.

Boateng, J.S., Matthews, K.H., Stevens, H.N., Eccleston, G.M., 2008. Wound healingdressings and drug delivery systems: a review. J. Pharm. Sci. 97, 2892–2923.

Bozzuto, G., Colone, M., Toccacieli, L., Stringaro, A., Molinari, A., 2011. Tea tree oilmight combat melanoma. Planta Med. 77, 54–56.

Brachkova, M.I., Marques, P., Rocha, J., Sepodes, B., Duarte, M.A., Pinto, J.F., 2011.Alginate films containing Lactobacillus plantarum as wound dressing forprevention of burn infection. J. Hosp Infect. 79, 375–377.

Britland, S., Smith, A., Finter, W., Eagland, D., Vowden, K., Vowden, P., Telford, G.,Brown, A., Pritchard, D., 2011. Recombinant Lucilia sericata chymotrypsin in atopical hydrogel formulation degrades human wound eschar ex vivo.Biotechnol. Progr. 27, 870–874.

Cado, G., Kerdjoudj, H., Chassepot, A., Lefort, M., Benmlih, K., Hemmerle, J., Voegel,J.C., Jierry, L., Schaaf, P., Frere, Y., Boulmedais, F., 2012. Polysaccharide films builtby simultaneous or alternate spray: a rapid way to engineer biomaterialsurfaces. Langmuir 28, 8470–8478.

Carson, C.F., Riley, T.V., 2001. Safety, efficacy and provenance of tea tree (Melaleucaalternifolia) oil. Contact Dermatitis 45, 65–67.

Carson, C.F., Hammer, K.A., Riley, T.V., 2006. Melaleuca alternifolia (Tea Tree) oil: areview of antimicrobial and other medicinal properties. Clin. Microbiol. Rev. 19,50–62.

Chin, K.B., Cordell, B., 2013. The effect of tea tree oil (Melaleuca alternifolia) onwound healing using a dressing model. J. Altern. Complement Med. 19, 942–945.

Cilek, A., Celebi, N., Tirnaksiz, F., 2006. Lecithin-based microemulsion of a peptidefor oral administration: preparation, characterization, and physical stability ofthe formulation. Drug Deliv. 13, 19–24.

Dantas, M.D., Cavalcante, D.R., Araujo, F.E., Barretto, S.R., Aciole, G.T., Pinheiro, A.L.,Ribeiro, M.A., Lima-Verde, I.B., Melo, C.M., Cardoso, J.C., Albuquerque Junior, R.L.,2011. Improvement of dermal burn healing by combining sodium alginate/chitosan-based films and low level laser therapy. J. Photochem. Photobiol. B105, 51–59.

Donsi, F., Annunziata, M., Sessa, M., Ferrari, G., 2011. Nanoencapsulation of essentialoils to enhance their antimicrobial activity in foods. Lwt-Food Sci. Technol. 44,1908–1914.

Donsi, F., Annunziata, M., Vincensi, M., Ferrari, G., 2012. Design of nanoemulsion-based delivery systems of natural antimicrobials: effect of the emulsifier. J.Biotechnol. 159, 342–350.

Fanun, M., 2012. Microemulsions as delivery systems. Curr. Opin. Colloid InterfaceSci. 17, 306–313.

Fletcher, P.D.I., Morris, J.S., 1995. Turbidity of oil-in-water microemulsion dropletsstabilized by nonionic surfactants. Colloids Surfaces A-Physicochem. Eng.Aspects 98, 147–154.

Garozzo, A., Timpanaro, R., Stivala, A., Bisignano, G., Castro, A., 2011. Activity ofMelaleuca alternifolia (tea tree) oil on Influenza virus A/PR/8: study on themechanism of action. Antiviral Res. 89, 83–88.

Ghosh, V., Mukherjee, A., Chandrasekaran, N., 2014. Eugenol-loaded antimicrobialnanoemulsion preserves fruit juice against, microbial spoilage. Colloids Surf. BBiointerfaces 114, 392–397.

Greay, S.J., Ireland, D.J., Kissick, H.T., Heenan, P.J., Carson, C.F., Riley, T.V., Beilharz,M.W., 2010. Inhibition of established subcutaneous murine tumour growthwith topical Melaleuca alternifolia (tea tree) oil. Cancer Chemother. Pharmacol.66, 1095–1102.

Hammer, K.A., Heel, K.A., 2012. Use of multiparameter flow cytometry to determinethe effects of monoterpenoids and phenylpropanoids on membrane polarity

and permeability in staphylococci and enterococci. Int. J. Antimicrob. Agents 40,239–245.

Hammer, K.A., Carson, C.F., Riley, T.V., 2003. Antifungal activity of the componentsof Melaleuca alternifolia (tea tree) oil. J. Appl. Microbiol. 95, 853–860.

Hammer, K.A., Carson, C.F., Riley, T.V., 2004. Antifungal effects of Melaleucaalternifolia (tea tree) oil and its components on Candida albicans, Candidaglabrata and Saccharomyces cerevisiae. J. Antimicrob. Chemother. 53, 1081–1085.

Hammer, K.A., Carson, C.F., Riley, T.V., 2012. Effects of Melaleuca alternifolia (teatree) essential oil and the major monoterpene component terpinen-4-ol on thedevelopment of single- and multistep antibiotic resistance and antimicrobialsusceptibility. Antimicrob. Agents Chemother. 56, 909–915.

Hart, P.H., Brand, C., Carson, C.F., Riley, T.V., Prager, R.H., Finlay-Jones, J.J., 2000.Terpinen-4-ol, the main component of the essential oil of Melaleuca alternifolia(tea tree oil), suppresses inflammatory mediator production by activatedhuman monocytes. Inflamm. Res. 49, 619–626.

Ireland, D.J., Greay, S.J., Hooper, C.M., Kissick, H.T., Filion, P., Riley, T.V., Beilharz,M.W., 2012. Topically applied Melaleuca alternifolia (tea tree) oil causes directanti-cancer cytotoxicity in subcutaneous tumour bearing mice. J. Dermatol. Sci.67, 120–129.

Jandera, V., Hudson, D.A., de Wet, P.M., Innes, P.M., Rode, H., 2000. Cooling the burnwound: evaluation of different modalites. Burns 26, 265–270.

Josef, E., Zilberman, M., Bianco-Peled, H., 2010. Composite alginate hydrogels: aninnovative approach for the controlled release of hydrophobic drugs. ActaBiomater. 6, 4642–4649.

Josef, E., Barat, K., Barsht, I., Zilberman, M., Bianco-Peled, H., 2013. Compositehydrogels as a vehicle for releasing drugs with a wide range of hydrophobicities.Acta Biomater. 9, 8815–8822.

Koh, K.J., Pearce, A.L., Marshman, G., Finlay-Jones, J.J., Hart, P.H., 2002. Tea tree oilreduces histamine-induced skin inflammation. Br. J. Dermatol. 147, 1212–1217.

Kuo, C.K., Ma, P.X., 2001. Ionically crosslinked alginate hydrogels as scaffolds fortissue engineering: Part 1. Structure, gelation rate and mechanical properties.Biomaterials 22, 511–521.

Laurienzo, P., 2010. Marine polysaccharides in pharmaceutical applications: anoverview. Mar. Drugs 8, 2435–2465.

Mikus, J., Harkenthal, M., Steverding, D., Reichling, J., 2000. In vitro effect ofessential oils and isolated mono- and sesquiterpenes on Leishmania major andTrypanosoma brucei. Planta Med. 66, 366–368.

Moreno, M.A., Ballesteros, M.P., Frutos, P., 2003. Lecithin-based oil-in-watermicroemulsions for parenteral use: pseudoternary phase diagrams,characterization and toxicity studies. J. Pharm. Sci. 92, 1428–1437.

Orthaber, D., Glatter, O., 2000. Synthetic phospholipid analogs: a structuralinvestigation with scattering methods. Chem. Phys. Lipids 107, 179–189.

Pazyar, N., Yaghoobi, R., Bagherani, N., Kazerouni, A., 2013. A review of applicationsof tea tree oil in dermatology. Int. J. Dermatol. 52, 784–790.

Pearce, A.L., Finlay-Jones, J.J., Hart, P.H., 2005. Reduction of nickel-induced contacthypersensitivity reactions by topical tea tree oil in humans. Inflamm. Res. 54,22–30.

Peng, C.W., Lin, H.Y., Wang, H.W., Wu, W.W., 2012. The influence of operatingparameters on the drug release and anti-bacterial performances of alginatewound dressings prepared by three-dimensional plotting. Mater. Sci. Eng. C-Mater. Biol. Appl. 32, 2491–2500.

Rossi, S., Marciello, M., Sandri, G., Ferrari, F., Bonferoni, M.C., Papetti, A., Caramella,C., Dacarro, C., Grisoli, P., 2007. Wound dressings based on chitosans andhyaluronic acid for the release of chlorhexidine diacetate in skin ulcer therapy.Pharm. Dev. Technol. 12, 415–422.

Ruel-Gariepy, E., Leroux, J.C., 2004. In situ-forming hydrogels – review oftemperature-sensitive systems. Eur. J. Pharm. Biopharm. 58, 409–426.

Schaaf, P., Voegel, J.C., Jierry, L., Boulmedais, F., 2012. Spray-assisted polyelectrolytemultilayer buildup: from step-by-step to single-step polyelectrolyte filmconstructions. Adv. Mater. 24, 1001–1016.

Schlenoff, J.B., Dubas, S.T., Faehat, T., 2000. Sprayed polyelectrolyte multilayers.Langmuir 16, 9968–9969.

Shakeel, F., Shafiq, S., Haq, N., Alanazi, F.K., Alsarra, I.A., 2012. Nanoemulsions aspotential vehicles for transdermal and dermal delivery of hydrophobiccompounds: an overview. Expert Opin. Drug Deliv. 9, 953–974.

Wiegand, I., Hilpert, K., Hancock, R.E., 2008. Agar and broth dilution methods todetermine the minimal inhibitory concentration (MIC) of antimicrobialsubstances. Nat. Protoc. 3, 163–175.