13 Melaleuca Alternifolia

Journal of Applied Microbiology 2000, 88, 170–175

The mode of antimicrobial action of the essential oil ofMelaleuca alternifolia (tea tree oil)

S.D. Cox1, C.M. Mann1, J.L. Markham 1, H.C. Bell 2, J.E. Gustafson 3, J.R. Warmington 3 andS.G. Wyllie 1

1Centre for Biostructural and Biomolecular Research, University of Western Sydney, Hawkesbury, New SouthWales, 2Australian Tea Tree Oil Research Institute, Lismore, New South Wales and 3GeneticaBiotechnologies, Bentley, Western Australia

7236/5/99: received 14 May 1999, revised 16 august 1999 and accepted 16 August 1999

S.D. COX, C.M. MANN, J.L. MARKHAM, H.C. BELL, J.E. GUSTAFSON, J.R. WARMINGTON AND S.G.

WYLLIE. 2000. The essential oil of Melaleuca alternifolia (tea tree) exhibits broad-spectrumantimicrobial activity. Its mode of action against the Gram-negative bacterium Escherichiacoli AG100, the Gram-positive bacterium Staphylococcus aureus NCTC 8325, and the yeastCandida albicans has been investigated using a range of methods. We report that exposingthese organisms to minimum inhibitory and minimum bactericidal/fungicidal concentrationsof tea tree oil inhibited respiration and increased the permeability of bacterial cytoplasmicand yeast plasma membranes as indicated by uptake of propidium iodide. In the case of E.coli and Staph. aureus, tea tree oil also caused potassium ion leakage. Differences in thesusceptibility of the test organisms to tea tree oil were also observed and these are interpretedin terms of variations in the rate of monoterpene penetration through cell wall and cellmembrane structures. The ability of tea tree oil to disrupt the permeability barrier of cellmembrane structures and the accompanying loss of chemiosmotic control is the most likelysource of its lethal action at minimum inhibitory levels.

INTRODUCTION

The essential oil of Melaleuca alternifolia, commonly knownas tea tree oil, has a long history of use as a topical antiseptic(Markham 1999). In recent times it has gained a reputationas a safe, natural and effective antiseptic. This has led to aresurgence in popularity and currently it is incorporated asthe principal antimicrobial or as a natural preservative inmany pharmaceutical and cosmetic products intended forexternal use.

The chemical composition of tea tree oil has been welldefined and consists largely of cyclic monoterpenes (Brophyet al. 1989) of which about 50% are oxygenated and about50% are hydrocarbons. Tea tree oil exhibits broad-spectrumantimicrobial activity (see Markham 1999 for a review) whichcan be principally attributed to terpinen-4-ol (Southwell et al.1993; Carson and Riley 1995).Correspondence to: Dr S.D. Cox, Building L9, Faculty of Science andTechnology, UWS Hawkesbury, Bourke Street, Richmond, 2753, New SouthWales, Australia (e-mail: [email protected]).

© 2000 The Society for Applied Microbiology

A wide variety of essential oils are known to possessantimicrobial properties and in many cases this activity is dueto the presence of active monoterpene constituents (Knoblochet al. 1988; Beylier 1979; Morris et al. 1979). Several studieshave also shown that monoterpenes exert membrane-dam-aging effects (reviewed by Sikkema et al. 1995). Examinationof Escherichia coli cells using electron microscopy afterexposure to tea tree oil revealed a loss of cellular electron-dense material and coagulation of cytoplasmic constituents,although it was apparent that these effects were secondaryevents that occurred after cell death (Gustafson et al. 1998).Tea tree oil also stimulates leakage of cellular potassium ionsand inhibits respiration in E. coli cell suspensions, providingevidence of a lethal action related to cytoplasmic membranedamage (Cox et al. 1998).

Here we report the further investigation of the antimicro-bial activity of tea tree oil against three clinically significantmicro-organisms, E. coli, Staphylococcus aureus and Candidaalbicans.

MODE OF ACTION OF TEA TREE OIL 171

MATERIALS AND METHODS

Tea tree oil

The tea tree oil used in all assays was from a sample (Batch6081) donated by Main Camp, Ballina, NSW, Australia.

Growth of test organisms

Cells used in all assays were twice passaged, in Iso-sensitestBroth (Oxoid, Basingstoke, UK) in the case of E. coli strainAG100, a K-12 derivative (George and Levy 1983) in thecase of Staph. aureus NCTC 8325, and in Malt extract broth(Oxoid) in the case of C. albicans KEM H5 at 37 °C.

Minimum inhibitory concentrations (MIC) and minimumbactericidal/fungicidal concentrations (MBC)

MIC/MBC assays were performed as described in Gustafsonet al. (1998) with the following exceptions. In the case of C.albicans, Malt extract broth (Oxoid) was substituted for Iso-sensitest broth (Oxoid). Tween-80 was omitted from thedilution/assay mixture. Minimum bactericidal/fungicidalconcentrations were determined by sampling 100ml fromeach tube that showed no growth into a neutralising brothwhich contained 30 g l−1 Tryptone soy broth (Oxoid), 30 g l−1

neutralized liver digest (Oxoid) and 10 g l−1 lecithin (DefianceMilling Co., Acacia Ridge, QLD). Following a 10-min roomtemperature incubation, 10ml Iso-sensitest was added to eachtube (malt extract broth was used in the case of C. albicans)and they were then incubated at 37 °C for 72 h. The minimumbactericidal/fungicidal concentration was determined as thelowest concentration of tea tree oil that showed no growth.

Viability assays

An overnight culture was used to inoculate an Iso-sensitestbroth (Malt extract broth was used for C. albicans). Cellswere grown at 37 °C to exponential phase (4–5 h), washedonce and resuspended in sterile 100-ml conical flasks con-taining 20ml of cell suspension and the required volume oftea tree oil. The flask contents were continually stirred on amagnetic stirrer to ensure uniform oil dispersion throughoutthe assays. Aliquots (1ml) were removed at the required timeintervals into 9ml of neutralising broth and allowed to standat room temperature for 10min. Serial 10-fold dilutions ofthe neutralising broths were then prepared in 0·1% peptoneand pour plates prepared using Tryptone soy agar (Oxoid).Colonies were counted after a 3 d incubation at 37 °C and theviable cell number reported as colony-forming units (CFU)per ml.

© 2000 The Society for Applied Microbiology, Journal of Applied Microbiology 88, 170–175

Measurement of respiration

Microbial respiration rates were determined using an oxygenelectrode as previously described in Cox et al. (1998). For E.coli and C. albicans, cell suspensions were preincubated for5min in the indicated tea tree oil concentration prior tomeasuring the respiratory activity. In the case of Staph.aureus, cells were preincubated in the presence of tea tree oilfor 10min prior to measurement.

Efflux of potassium ions

Potassium ion concentration in cell suspensions was measuredusing a potassium ion selective electrode, as previouslydescribed in Cox et al. (1998). The concentration of total freepotassium for Staph. aureus suspensions was measured afterincubation in lysostaphin (100mgml−1) at 37 °C for 60min,followed by sonication. To measure total free potassium inC. albicans, cells were lysed by incubating in chitinase(1mgml−1) and lyticase (1mgml−1) at 37 °C for 60min,followed by sonication. Complete lysis in each case was con-firmed by microscopic examination.

Propidium iodide uptake

Cells (100ml culture) were grown overnight as describedabove, washed and resuspended in 50mmol l−1 sodium phos-phate buffer, pH 7·1. Aliquots (1ml) were added to stirredconical flasks containing 19ml buffer and the requiredamount of tea tree oil. The inoculum density was¼ 108 CFUml−1. Following a 30-min incubation at roomtemperature, 50ml aliquots were transferred into Eppendorfscontaining 950ml phosphate buffer in FACS tubes (BectonDickinson, Immunocytometry Systems, Mountain View,California). These tubes were stored on ice and 5ml of stainingsolution, consisting of 2·5mgml−1 propidium iodide (Molec-ular Probes, Eugene, Oregon) dissolved in milliQ water, wasadded to give a final propidium iodide concentration of10mgml−1. Immediately following this, the percentage ofpropidium iodide stained cells was determined using a FAC-Scan Flow cytometer (Becton Dickinson).

Assay of tea tree oil-induced carboxyfluoresceinleakage

Multilamellar lipid vesicles were prepared following the pro-cedure of New (1990). Phospholipids (14mg phos-phatidylethanolamine, 4mg phosphatidylglycerol and 2mgcardiolipin) were dissolved in chloroform in a 100-ml round-bottom flask and evaporated to dryness. Following this, thedried phospholipid mixture was resuspended in 2ml of50mmol l−1 sodium phosphate buffer, pH 7·0, containing aself-quenched concentration of carboxyfluorescein

172 S.D. COX ET AL.

(100mmol l−1), by gentle shaking with glass beads. Theresulting suspension of liposomes (multilamellar lipid ves-icles) was then dialysed overnight to remove unencapsulatedcarboxyfluorescein. Liposome suspension (100ml) was addedto an Eppendorf tube, followed by phosphate buffer and therequired amount of tea tree oil, to give 1ml final volume.The mixture was then vortexed and incubated for therequired time interval with repeated mixing at 5min intervals.When the incubation period had elapsed, 50ml of the lipo-some suspension was sampled into 2ml phosphate buffer.Fluorescence was measured in a glass cuvette using a flu-orescence spectrophotometer (Hitachi F-4500, Hitachi, SanJose, CA, USA; lex � 470 nm; lem � 520 nm). One hundredper cent carboxyfluorescein leakage was determined byadding triton-X100, 1·0% (v/v).

RESULTS

Minimum inhibitory concentrations (MIC) and minimumbactericidal/fungicidal concentrations (MBC) of tea treeoil

The MIC and MBC of tea tree oil were 0·25% and 0·5%(v/v), respectively, for both E. coli AG100 and Staph. aureusNCTC 8325. MIC and MBC values for C albicans KEM H5were a factor of two lower, at 0·125% and 0·25% (v/v),respectively.

Effects of tea tree oil on cell viability

The effects of tea tree oil exposure on the viability of E. coli,Staph. aureus and C. albicans are shown in Fig. 1(a,b,c). Eachfigure is representative of three separate experiments thatgave similar results. At minimum inhibitory and minimumlethal concentrations, E. coli was most susceptible to theeffects of tea tree oil, followed by C. albicans and then Staph.aureus.

Effects of tea tree oil on respiration

Tea tree oil inhibited respiration in cell suspensions of E.coli, Staph. aureus and C. albicans (Fig. 2). Inhibition of E.coli respiration commenced in 0·25% (v/v) tea tree oil andwas complete at 0·5% (v/v). Respiration in C. albicans cellswas inhibited at 0·125% (v/v), which was the lowest con-centration assayed and corresponds to the MIC for this organ-ism. Inhibition of Staph. aureus cell respiration (after 10minof tea tree oil exposure) commenced at a concentration of0·5% (v/v).


Fig. 1 Effects of tea tree oil on viability of test organisms. (a) E. coliAG 100: (�) no tea tree oil, and (Ž) 0·50% v/v tea tree oil. (b)Staph. aureus NCTC 100: (�) no tea tree oil, (ž) 0·25% v/v teatree oil, and (Ž) 0·50% v/v tea tree oil. (c) C. albicans KEM H6:100:(�) no tea tree oil, (ž) 0·125% v/v tea tree oil, (�) 0·25% v/vtea tree oil; and (�) 0·50% v/v tea tree oil

Effects of tea tree oil on membrane integrity

Exposing cell suspensions of E. coli, Staph. aureus and C.albicans to 0·25% (v/v) tea tree oil for 30min increased cellpermeability to the fluorescent nucleic acid stain, propidiumiodide (Fig. 3), relative to control suspensions that did notcontain tea tree oil. The inability of propidium iodide topenetrate cells with intact cytoplasmic or plasma membranes(see Brul et al. 1997; Mason et al. 1997; Wenisch et al. 1997;Lebaron et al. 1998) was confirmed by the low level of uptakeobserved in cells not exposed to tea tree oil (Fig. 3).

Tea tree oil at 0·25% (v/v) induced leakage of potassiumions from E. coli and Staph. aureus cells (Fig. 4). The data,which is representative of triplicate experiments that gavesimilar results, shows that leakage from E. coli cells com-menced immediately upon addition of tea tree oil and the


Fig. 2 Effects of tea tree oil concentration on O2 consumption ratesin cell suspensions of E. coli AG100 (�), Staph. aureus NCTC 8325(ž) and C. albicans KEM H5 (�). Error bars represent the standarddeviation (n � 3) of data from replicate experiments. In some casesthe error bars are small enough to be obscured by data symbols

Fig. 3 Uptake of propidium iodide in cell suspensions of E. coliAG100, Staph. aureus NCTC 8325 and C. albicans KEM H5. Cellswere exposed to 0·25% v/v tea tree oil for 30 min (�) and comparedwith control flasks containing no added tea tree oil (Ž). Error barsrepresent standard deviations calculated from separate assays(n � 3)

extent approached 100% of total cellular free potassium afterabout 30min. Efflux from Staph. aureus cells, by comparison,commenced after about 5min exposure to tea tree oil andcontinued at a slower rate, reaching 20% after 30min. Leak-age of potassium ions from C. albicans in the presence 0·25%(v/v) tea tree oil over a period of two h was no greater thanbackground levels (data not shown). However, after exposureto 2·5% (v/v) for 60min the amount of potassium in cellsupernatants was 23·1% of that in supernatants from totalcell lysates.


Fig. 4 Effects of 0·25% v/v tea tree oil on potassium ion efflux incell suspensions of E. coli AG100 and Staph. aureus NCTC 8325.(�) E. coli, 0·25% v/v tea tree oil; (ž) E. coli, no tea tree oil; (�)S. aureus, 0·25% v/v tea tree oil; (Ž), Staph. aureus, no tea tree oil

Tea tree oil (0·25% v/v) also stimulated the leakage ofencapsulated carboxyfluorescein from a suspension of mul-tilamellar lipid vesicles (Fig. 5).

DISCUSSION

In this study, tea tree oil inhibited respiration in E. coli,Staph. aureus and C. albicans cells at minimum inhibitorylevels. The possibility that tea tree oil directly inhibits aspecific respiratory enzyme or metabolic event cannot beeliminated. However, our findings also reveal that minimuminhibitory levels of tea tree oil altered cell membrane struc-

Fig. 5 Stimulation of leakage of carboxyfluorescein frommultilamellar lipid vesicles exposed to tea tree oil. (�) No tea treeoil; (ž) 0·25% v/v tea tree oil. Error bars represent standarddeviations (n � 2)

174 S.D. COX ET AL.

ture. Increased uptake of the nucleic acid stain propidiumiodide, to which the cell membrane is normally impermeable,was observed. Also, leakage of potassium ions commencedimmediately upon adding tea tree oil to suspensions con-taining E. coli and within 5min for Staph. aureus cells.

In the case of C. albicans, we did not detect the appearanceof potassium ions in cell supernatants containing 0·25% (v/v)tea tree oil. However, the propidium iodide staining of C.albicans cells exposed to tea tree oil is a clear indication ofdamage to the plasma membrane. It may be that potassiumions do not appear in cell supernatants (after up to 2 hexposure) because they remain incorporated in the thick layerof the C. albicans cell wall. Given the increased permeabilityto propidium iodide, it seems unlikely that the plasma mem-brane would have remained impermeable to potassium ions.Further confirmation of the general toxicity of tea tree oil tomembrane structures is provided by its permeabilising effecton multilamellar liposomes.

Previously, we have shown that tea tree oil inhibits res-piration and causes leakage of cellular potassium in E. coli atminimum inhibitory levels (Cox et al. 1998). These effects,along with the findings presented here, indicate that tea treeoil damages cell membrane structure in E. coli, Staph. aureusand C. albicans. The cytoplasmic membranes of bacteria andthe plasma and mitochondrial membranes of yeast provide abarrier to the passage of small ions such as H+, K+, Na+ andCa2+ and allow cells and organelles to control the entry andexit of different compounds. This permeability barrier roleof cell membranes is integral to many cellular functions,including the maintenance of the energy status of the cell,other membrane-coupled energy-transducing processes, sol-ute transport, regulation of metabolism and control of turgorpressure (Booth 1985; Poolman et al. 1987; Trumpower andGennis 1994).

Toxic effects on membrane structure and function havegenerally been used to explain the antimicrobial action ofessential oils and their monoterpenoid components (Andrewset al. 1980; Uribe et al. 1985; Knobloch et al. 1988). Sikkemaet al. (1994) showed that, as a result of their lipophilic charac-ter, cyclic monoterpenes will preferentially partition from anaqueous phase into membrane structures. This resulted inmembrane expansion, increased membrane fluidity and inhi-bition of a membrane-embedded enzyme. In yeast cells andisolated mitochondria, a-pinene and b-pinene destroy cellularintegrity, inhibit respiration and ion transport processes andincrease membrane permeability (Andrews et al. 1980; Uribeet al. 1985). More recently, Helander et al. (1998) havedescribed effects of different essential components on outermembrane permeability in Gram-negative bacteria. The factthat tea tree oil-induced damage to cell membrane structureaccompanied the decline in viability for all three micro-organ-isms included in this study confirms it as the most likelycause of cell death.


In spite of similar MIC/MBC values, the micro-organismsstudied here showed obvious differences in their sus-ceptibility to tea tree oil. The rate of viability decline of C.albicans in 0·25% (v/v) tea tree oil was less than that seen forE. coli in the same concentration and for Staph. aureus therate of inactivation was slower than that of either E. coli orC. albicans. The relative inhibition of respiration and theextent of membrane damage of these different micro-organ-isms follow the same pattern. Given the broad spectrumactivity of tea tree oil and its general membrane-damagingeffect, it is likely that this variability reflects the rate at whichits active components diffuse through the cell wall and intothe phospholipid regions of cell membrane structures.

In conclusion, our observations confirm that the antimicro-bial activity of tea tree oil results from its ability to disruptthe permeability barrier of microbial membrane structures.This mode of action is the same against E. coli, Staph. aureusand C. albicans and is similar to that of other broad-spectrum,membrane-active disinfectants and preservatives, such asphenol derivatives, chlorhexidine (see McDonnell and Rus-sell 1999) and parabenzoic acid derivatives (Sox 1997).

ACKNOWLEDGEMENT

This work was wholly funded by the Australian Tea Tree OilResearch Institute (ATTORI), Lismore, New South Wales,Australia.

REFERENCES

Andrews, R.E., Parks, L.W. and Spence, K.D. (1980) Some effectsof Douglas fir terpenes on certain microorganisms. Applied andEnvironmental Microbiology 40, 301–304.

Beylier, M. (1979) Bacteriostatic activity of some Australian essentialoils. Perfumer and Flavourist 4, 23–25.

Booth, I.R. (1985) Regulation of cytoplasmic pH in bacteria Micro-biological Reviews 49, 359–378.

Brophy, J.J., Davies, N.W., Southwell, I.A., Stiff, I.A. and Williams,L.R. (1989) Gas chromatographic quality control for oil of Mel-aleuca terpinen-4-ol type (Australian tea tree) Journal of Agri-cultural and Food Chemistry 37, 1330–1335.

Brul, S., Nussbaum, J. and Dielbandhoesing, S.K. (1997) Flu-orescent probes for wall porosity and membrane integrity infilamentous fungi. Journal of Microbiological Methods 28, 169–178.

Carson, C.F. and Riley, T.V. (1995) Antimicrobial activity of themajor components of the essential oil of Melaleuca alternifolia.Journal of Applied Bacteriology 78, 264–269.

Cox, S.D., Gustafson, J.E., Mann, C.M., Markham, J.L., Liew,Y.C., Hartland, R.P., Bell, H.C., Warmington, J.R. and Wyllie,S.G. (1998) Tea tree oil causes K+ leakage and inhibits respirationin Escherichia coli. Letters in Applied Microbiology 26, 355–358.

George, A.M. and Levy, S.B. (1983) Amplifiable resistance to tetra-cycline, chloramphenicol, and other antibiotics in Escherichia coli:


involvement of a non-plasmid-determined efflux of tetracycline.Journal of Bacteriology 155, 531–540.

Gustafson, J.E., Liew, Y.C., Chew, S., Markham, J.L., Bell, H.C.,Wyllie, S.G. and Warmington, J.R. (1998) Effects of tea tree oilon Escherichia coli. Letters in Applied Microbiology 26, 194–198.

Helander, I.M., Alakomi, H.-L., Kyosti, L.-K., Mattila-Sandholm,T., Pol, I., Smid, E.J., Gorris, G.M. and von Wright, A. (1998)Characterization of the action of selected essential oil componentson Gram-negative bacteria. Journal of Agricultural Food Chemistry46, 3590–3595.

Knobloch, K., Pauli, A., Iberl, B., Weis, N. and Weigand, H. (1988)Antibacterial activity and antifungal properties of essential oilcomponents. Journal of Essential Oils Research 1, 119–128.

Lebaron, P., Catala, P. and Parthuisot, N. (1998) Effectiveness ofSYTOX Green stain for bacterial viability assessment. Appliedand Environmental Microbiology 64, 2697–2700.

Markham, J.L. (1999) Biological activity of tea tree oil. In Tea Tree,the Genus Melaleuca. ed. Southwell, I. and Lowe, R., pp. 169–190. Amsterdam: Harwood Academic Publishers.

Mason, D.J., Dybowski, R., Larrick, J.W. and Gant, V.A. (1997)Antimicrobial action of rabbit leukocyte CAP18106–137. Antimicro-bial Agents and Chemotherapy 41, 624–629.

McDonnell, G. and Russell, A.D. (1999) Antiseptics and dis-infectants: activity, action and resistance. Clinical MicrobiologyReviews 12, 147–179.

Morris, J.A., Khettry, A. and Seitz, E.W. (1979) Antimicrobialactivity of aroma chemicals and essential oils. Journal of the Amer-ican Chemical Society 56, 595–603.

New, R.R.C. (1990) Characterization of liposomes. In Liposomes: A


Practical Approach, ed. New, R.C.C., pp. 105–162. Oxford: IRLPress.

Poolman, B., Driessen, A.J.M. and Konings, W.N. (1987) Regu-lation of solute transport in Streptococci by external and internalpH values. Microbiological Reviews 51, 498–508.

Sikkema, J., de Bont, J.A.M. and Poolman, B. (1994) Interactionsof cyclic hydrocarbons with biological membranes. Journal ofBiological Chemistry 269, 8022–8028.

Sikkema, J., de Bont, J.A.M. and Poolman, B. (1995) Mechanismsof membrane toxicity of hydrocarbons. Microbiological Reviews59, 201–222.

Southwell, I.A., Hayes, A.J., Markham, J.L. and Leach, D.N. (1993)The search for optimally bioactive Australian tea tree oil. ActaHorticulturae 334, 265–275.

Sox, T.E. (1997) Mechanisms of action of cosmetic preservatives.In Cosmetic Microbiology: a Practical Handbook. ed. Brannan,D.K. pp. 163–176, Boca Raton, FL: CRC Press.

Trumpower, B.L. and Gennis, R.B. (1994) Energy transduction bycytochrome complexes in mitochondrial and bacterial respiration:the enzymology of coupling electron transfer reactions to trans-membrane proton translocation. Annual Reviews in Biochemistry63, 675–716.

Uribe, S., Ramirez, T. and Pena, A. (1985) Effects of b-pinene onyeast membrane functions. Journal of Bacteriology 161, 1195–1200.

Wenisch, C., Linnau, K.F., Parschalk, B., Zedtwitz-Lebenstein, K.and Georgopolous, A. (1997) Rapid susceptibility testing of fungiby flow cytometry using vital staining. Journal of Clinical Micro-biology 35, 5–10.

Documents

13 Melaleuca Alternifolia