Interactions between components of the essential oil of Melaleuca alternifolia

Journal of Applied Microbiology 2001, 91, 492±497 Interactions between components of the essential oil of Melaleuca alternifolia S.D. Cox, C.M. Mann and J.L. Markham Centre for Biostructural and Biomolecular Research, University of Western Sydney, Hawkesbury, New South Wales, Australia 710/1/01: received 10 January 2001, revised 5 April 2001 and accepted 5 April 2001 S.D. C O X, C. M. M A N N A N D J. L. M A R K H A M. 2001. Aims: This study compared the antimicrobial activity of Melaleuca alternifolia (tea tree) oil with that of some of its components, both individually and in two-component combinations. Methods and Results: Minimum inhibitory concentration and time-kill assays revealed that terpinen-4-ol, the principal active component of tea tree oil, was more active on its own than when present in tea tree oil. Combinations of terpinen-4-ol and either c-terpinene or p-cymene produced similar activities to tea tree oil. Concentration-dependent reductions in terpinen-4-ol activity and solubility also occurred in the presence of c-terpinene. Conclusions: Non-oxygenated terpenes in tea tree oil appear to reduce terpinen-4-ol ef cacy by lowering its aqueous solubility. Signi cance and Impact of the Study: These ndings explain why tea tree oil can be less active in vitro than terpinen-4-ol alone and further suggest that the presence of a non-aqueous phase in tea tree oil formulations may limit the microbial availability of its active components. INTRODUCTION The essential oil of Melaleuca alternifolia (tea tree oil) is a membrane-active biocide (Cox et al. 2000) with broad spectrum activity (see Markham 1999 for a review). In recent years tea tree oil has gained widespread acceptance and it is now incorporated as the principal antimicrobial or preservative in a range of pharmaceuticals or cosmetics for external use, such as face and hand washes, pimple gels, vaginal creams, foot powders, shampoos, conditioners and veterinary skin care products. The chemical composition of tea tree oil is well de ned (Brophy et al. 1989) and previous investigations have identi ed that oxygenated terpenoids are the main active components (Southwell et al. 1993; Carson and Riley 1995). A large variation in oil composition occurs naturally with two signi cant components, terpinen-4-ol and 1,8-cineole being present in inverse proportions (Southwell et al. 1993). Terpinen-4-ol is considered to be the principal active component and it constitutes up to about 40% of some tea tree oils with non-oxygenated terpenoid hydrocarbons Correspondence to: S.D. Cox, Building L9, Centre for Biostructural and Biomolecular Research, University of Western Sydney, Bourke Street, Richmond, 2753, New South Wales, Australia (e-mail: s.cox@uws.edu.au). accounting for approximately 50%. In general, hydrocarbon monoterpenes are signi cantly less active than oxygenated monoterpenes (Carson and Riley 1995; Grif n et al. 1999). Numerous studies have addressed the antimicrobial effects of essential oils and their individual terpenoid components. However, to date there has been no speci c investigation into the possibility of interactions between the different types of monoterpene components found in essential oils. In this study we have examined the antimicrobial effects of tea tree oil and some of its individual components against four different microorganisms. We have also examined effects of combining the less active nonoxygenated terpenoid hydrocarbons, c-terpinene and p-cymene, with the principal active component, terpinen-4-ol. MATERIALS AND METHODS Tea tree oil and monoterpenes The tea tree oil (batch 6081) used in all assays was provided by Main Camp, Ballina, NSW, Australia. Monoterpenes were purchased from Sigma-Aldrich (Castle Hill, NSW, Australia) and used without further puri cation. The composition of the tea tree oil sample was determined by gas chromatography following the procedure described in Grif n et al. (1999). ã 2001 The Society for Applied Microbiology

INTERACTIONS BETWEEN MONOTERPENES IN TEA TREE OIL 493 Growth of test organisms The organisms used were Escherichia coli strain AG100, Staphylococcus aureus NCTC 8325, Pseudomonas aeruginosa NCTC 6749 and Candida albicans KEM H5. Samples of the E. coli and C. albicans cultures are maintained in the University of Western Sydney culture collection. Bacterial stock cultures were maintained on nutrient agar slopes and stored at 4 C. C. albicans was stored on malt extract agar. Bacterial cells used in all experiments were twice passaged in Iso-sensitest Broth (Oxoid, Basingstoke, UK) at 37 C. C. albicans was twice passaged in Malt Extract Broth (Oxoid) at 37 C. Minimum inhibitory and minimal cidal concentration and microbial viability Minimum inhibitory/minimum cidal concentrations and microbial viability were determined using procedures described in Mann et al. (2000). Speci c details of individual assays are provided in the gure legends. Terpinen-4-ol solubility determinations The aqueous terpinen-4-ol solubility was determined using the method described by Grif n et al. (1999). An excess of the terpene mixtures (10 ll) was added to 1 ml of milliq water in 2 ml GC sample vials. The vials were sealed, vortexed for 5 min and sonicated (Brandson 1200 bath sonicator) for 30 min. The vials were then equilibrated in a 25 C water bath for 1 h, inverted and centrifuged at 6000 r.p.m. for 15 min at 25 C. Five hundred microlitre samples were withdrawn by syringe and transferred to fresh GC sample vials and terpinen-4-ol concentrations were determined using GC analysis as described by Grif n et al. (1999). RESULTS Analysis of tea tree oil composition The tea tree oil used in this study (Main Camp batch 6081) conformed to the international standard for Oil of Melaleuca, terpinen-4-ol type (International Standards Organization 1996). The major terpenes present were terpinen-4-ol, c-terpinene, a-terpinene, 1,8-cineole and a-pinene. Relative Table 1 Identity, retention times and percentage peak areas of the predominant components of tea tree oil sample MC 6081 as determined by gas chromatography Retention (min) Compound % Peak area 6á308 a-thujene 0á8 6á611 a-pinene 2á1 7á815 sabinene 0á4 8á081 b-pinene 0á7 8á162 b-myrcene 0á8 8á975 a-phellandrene 0á4 9á341 a-terpinene 8á3 9á679 p-cymene 2á3 9á796 limonene 1á1 9á930 b-phellandrene 0á7 10á006 1,8-cineole 4á5 10á897 c-terpinene 17á8 11á294 a-terpinolene 3á3 13á628 sabinene hydrate 0á4 16á008 terpinen-4-ol 39á8 16á515 a-terpineol 3á4 25á381 aromadendrene 1á2 27á140 viridi orene 1á2 27á965 d-cadinene 1á5 30á464 globulol 0á5 30á774 viridi orol 0á4 compositions, expressed as percentage peak areas, for these and minor constituents are given in Table 1. Minimum inhibitory and cidal concentrations (MIC/MCC) of tea tree oil and some of its main components E. coli, Staph. aureus and C. albicans showed similar susceptibilities to tea tree oil and the monoterpenes tested in the MIC/MCC assays (Table 2). On a volume basis, tea tree oil and terpinen-4-ol were most active, followed by 1,8-cineole. c-terpinene was inactive (up to 8% v/v) against the two Gram-negatives, E. coli and Ps. aeruginosa NCTC 6749, but was active against Staph. aureus and C. albicans. p-cymene was inactive against all four species. Ps. aeruginosa NCTC 6749 tolerated exposure of up to 8% Table 2 Minimum inhibitory concentrations (MIC) and minimum cidal concentrations (MCC) expressed as percentage v/v for tea tree oil and some of its main constituents against Staph. aureus NCTC 8325, E. coli AG100, Ps. aeruginosa NCTC 6749 and C. albicans KEM H5 Staph. aureus E. coli Ps. aeruginosa C. albicans MIC MCC MIC MCC MIC MCC MIC MCC Tea tree oil 0á25 0á5 0á25 0á5 > 8 > 8 0á125 0á25 Terpinen-4-ol 0á125 0á25 0á125 0á25 0á5 > 8 0á125 0á5 c-terpinene 0á25 0á5 > 8 > 8 > 8 > 8 0á125 0á5 p-cymene > 8 > 8 > 8 > 8 > 8 > 8 > 8 > 8 1,8-cineole 0á5 1 1 1 > 8 > 8 0á125 1

494 S.D. COX ET AL. v/v tea tree oil and all the monoterpenes tested, except for terpinen-4-ol, which showed growth inhibitory but not bactericidal activity. Effects of tea tree oil and mixtures of some of its main components on microbial viability To investigate further the antimicrobial action of terpinen- 4-ol in tea tree oil we examined the effects of varying monoterpene concentration on viability after xed exposure times. On a weight per volume basis, tea tree oil was less active than terpinen-4-ol on its own against Ps. aeruginosa (Fig. 1a) and Staph. aureus (Fig. 1b). Furthermore, the effect of tea tree oil against these two species was similar to that produced by a mixture composed of 50% (w/w) terpinen-4-ol and 50% c-terpinene. For E. coli (Fig. 1c) and C. albicans (Fig. 1d), tea tree oil, terpinen-4-ol and the mixture of c-terpinene and terpinen- 4-ol had similar activities during the 30 min experiment. Overall, in terms of effects on viability, E. coli and C. albicans were more susceptible than Staph. aureus (Fig. 1b) and Ps. aeruginosa (Fig. 1a). Closer investigation into the effects of mixing hydrocarbon and oxygenated terpenes revealed that combining 0á1% w/v of c-terpinene or p-cymene with an aqueous solution (0á1% w/v) of terpinen-4-ol produced more activity against E. coli than each component on its own at 0á1% (Fig. 2). A similar effect was observed with C. albicans (data not shown). However, no decline in Staph. aureus or Ps. aeruginosa viability occurred within the 60 min assay time (data not shown). In a separate experiment E. coli cells were pre-exposed to 0á1% w/v terpinen-4-ol for 30 min, washed twice and then re-suspended in ISB. When these cells were added to ISB containing either c-terpinene or p-cymene at 0á1% w/v there was no decline in viability after 2 h (data not shown). Varying the concentration of c-terpinene added to the 0á1% w/v terpinen-4-ol solution also showed that activity was signi cantly enhanced against E. coli and C. albicans, even with relatively low levels of hydrocarbon (Fig. 3). However, the effect was progressively reduced when c-terpinene levels were increased above 0á5% w/v. Effect of c-terpinene on aqueous terpinen-4-ol solubility in a two-phase system Measuring the solubility of terpinen-4-ol in a simple twophase system consisting of terpinen-4-ol dispersed in water, and in a more complex system containing terpinen-4-ol and c-terpinene, showed that increasing the c-terpinene content reduced the aqueous solubility of terpinen-4-ol (Fig. 4). (a) (c) Viability (log 10 cfu ml 1 ) (b) (d) % (w/v) monoterpene Fig. 1 Effect of tea tree oil and some of its main components on microbial viability. (a) Ps. aeruginosa NCTC 6749 after 60 min exposure, (b) Staph. aureus NCTC 8325 after 60 min, (c) E. coli AG 100 viability after 30 min and (d) C. albicans KEM H5 after 30 min c-terpinene (j), terpinen-4-ol/c-terpinene 50/50 w/w mixture (h), terpinen-4-ol (d), tea tree oil (s), p-cymene (r), -terpinen- 4-ol/p-cymene 50/50 w/w mixture (e). The limit of detection was 2 log units. Error bars represent standard deviations calculated from replicate experiments (n ˆ 2)

INTERACTIONS BETWEEN MONOTERPENES IN TEA TREE OIL 495 (a) Viability (log 10 cfu ml 1 ) Viability (log10cfu ml 1 ) (b) γ-terpinene concentration (%w/v) Fig. 3 Effect of c-terpinene concentration on microbial viability in the presence of 0á1 w/v terpinen-4-ol. E. coli AG 100 after 60 min exposure (j); C. albicans KEM H5 after 45 min (d). The limit of detection was 2 log units. Error bars represent standard deviations (n ˆ 3) Terpinen-4ol soulbility (p.p.m.) Time (min) Fig. 2 Effect of adding c-terpinene (a) and p-cymene (b) to a 0á1% w/v solution of terpinen-4-ol against E. coli AG 100. Control (m); 0á1% terpinen-4-ol (d), 0á1% c-terpinene (j), 0á1% c-terpinene + 0á1% terpinen-4-ol (h), p-cymene (r), 0á1% p-cymene + 0á1% terpinen-4-ol (e). The limit of detection was 2 log units DISCUSSION The data presented here are consistent with previous studies which demonstrate that terpinen-4-ol is the principal active component of tea tree oil (Southwell et al. 1993; Carson and Riley 1995). The activities measured for the individual monoterpenes tested are similar to those published previously for these compounds (Carson and Riley 1995; Grif n et al. 1999), except that in this study the non-oxygenated terpenoid hydrocarbon, c-terpinene was active against % w/w γ-terpinene in mixture Fig. 4 Aqueous solubility of terpinen-4-ol (n ˆ 3) in a two-phase system consisting of terpinen-4-ol and c-terpinene mixed in different proportions Staph. aureus (MIC/MCC 0á25/0á50 and C. albicans (MIC/MCC 0á125/0á5). c-terpinene was not active against the two Gram negative bacteria included in the study. Previously, c-terpinene has been reported to be inactive in a broth microdilution assay (Carson and Riley 1995) and an agar dilution assay (Grif n et al. 1999). In this present study the MIC assays were performed in glass test tubes on an orbital shaker set at 200 r.p.m., which may have increased the microbial availability of c-terpinene. Given that c-terpinene constitutes 17á8% of the total oil content in the sample tested, its activity against Staph. aureus and C. albicans in MIC/MCC assays suggests that it may contribute to the overall action of tea tree oil. However,

496 S.D. COX ET AL. clearly c-terpinene must be relatively slow-acting because there was no decline in either Staph. aureus or C. albicans viability during the exposure times used for the time-kill assays (up to 1 h), even when assayed at 5% w/v. In the MIC/MCC assays, Ps. aeruginosa tolerated exposure to 8% v/v tea tree oil, the highest concentration assayed. This corresponds to 3á2% v/v terpinen-4-ol. However, terpinen-4-ol was growth inhibitory for this organism at 0á5% v/v, indicating that against Ps. aeruginosa it is more active than tea tree oil. Carson and Riley (1995) also reported inhibitory activity for terpinen-4-ol, but not tea tree oil, against Ps. aeruginosa and our ndings support their suggestion that antagonism occurs between components of the oil. Antagonistic effects were also evident in the viability measurements obtained for Ps. aeruginosa and Staph. aureus after 60 min exposure to monoterpenes. Terpinen-4-ol on its own was clearly more active than a dispersion of tea tree oil containing an equivalent amount of terpinen-4-ol. The lower activity produced by the mixture (50 : 50 w/w ratio) of terpinen-4-ol and c-terpinene (or p-cymene) identi es the hydrocarbon components of tea tree oil as the most probable antagonistic principals. Antagonism was not seen with E. coli or C. albicans. This is due most probably to their being generally more sensitive to the effects of monoterpenes. Ps. aeruginosa NCTC 6749 is intrinsically resistant to tea tree oil (Mann et al. 2000). While Staph. aureus is sensitive to tea tree oil in MIC assays at levels comparable to E. coli and C. albicans, time-kill studies also reveal that it dies off more slowly (Cox et al. 2000). Therefore, it appears that the antagonistic effects of the monoterpene hydrocarbons are likely to be most signi cant against microorganisms that are either resistant to or are not rapidly killed by tea tree oil. In this study, it was also apparent that non-inhibitory levels of either c-terpinene or p-cymene could enhance the antimicrobial effects of aqueous terpinen-4-ol (0á1% w/v) against E. coli and C. albicans. c-terpinene on its own produced minimal effects on E. coli or C. albicans viability in the assay time for the experiment (30 and 45 min), even when assayed at 5% w/v (not shown here). However, in the terpinen-4-ol solution, an increase in activity occurred at 0á03% w/v c-terpinene, the lowest concentration assayed. The progressive loss in activity that occurred when the amount of c-terpinene added was increased above 0á25% w/v again re ects the antagonistic in uence of this compound. Previously, we have shown that p-cymene and c-terpinene become active against Ps. aeruginosa NCTC 6749 cells treated with polymyxin B nonapeptide, a Gram negative bacterial outer membrane (OM) permeabilizing agent (Mann et al. 2000). This indicates that c-terpinene and p-cymene are probably inactive against Ps. aeruginosa because they are unable to penetrate the OM. Helander et al. (1998) reported that the phenolic monoterpenes, carvacrol and thymol permeabilize and irreversibly damage the Gram-negative OM to the extent that lipopolysaccharide and OM protein is released to the supernatant. However, if terpinen-4-ol exerts OM permeabilizing action it would clearly have to be reversible, since E. coli cells pre-exposed to terpinen-4-ol were able to tolerate exposure to p-cymene and c-terpinene. Furthermore, the fact that a combination of c-terpinene and 0á1% w/v terpinen-4-ol produces similar effects against C. albicans means that any permeabilizing action would also have to be non-speci c. For this study, tea tree oil and monoterpenes were dispersed in culture medium without surfactant using a magnetic stirrer. This produced a two-phase system of tea tree oil droplets dispersed in an aqueous phase that could be expected to have become enriched with a more soluble, oxygenated monoterpene fraction. In the case of terpinen- 4-ol dispersions, we have also shown that adding c-terpinene both reduced antimicrobial ef cacy and lowered terpinen-4-ol aqueous solubility. For preservatives such as chlorocresol and p-hydroxybenzoates it is known that antimicrobial activity in solubilized and emulsi ed dispersed systems is dependent upon the free (unbound) concentration in the aqueous phase (Kazmi and Mitchell 1978; Kurup et al. 1991). In the case of monoterpenes, antimicrobially active compounds such as terpinen-4-ol are more watersoluble than inactive compounds (Grif n et al. 1999). Therefore, it seems likely that the level of terpinen-4-ol present in the aqueous phase of tea tree oil dispersions is an important determinant of microbial availability and the antimicrobial ef cacy of the system. Hydrophobic compounds can be toxic to microorganisms because they preferentially partition into and damage the structure and function of cell membranes (Sikkema et al. 1995). However, hydrophobicity is not the only factor that in uences toxicity. For organic solvents, the octanol±water partition coef cient (P ow ) is considered to be an important toxicitydetermining factor and solvents with log P ow values > 5 are not generally toxic to microorganisms because their low aqueous solubility is thought to prevent them from reaching lethal levels in cell membranes (Aono and Inoue 1998). Therefore, an alternative explanation for our observation may be that the presence of terpinen-4-ol in solution increases the microbial availability of c-terpinene or p-cymene when the latter are present at low concentrations. Shaking a terpinen-4-ol solution produces an unstable foam, suggesting that terpinen-4-ol has weak surfactant properties. Possibly, this weak amphiphathic nature results in increased solubilization and increased uptake of the hydrocarbon monoterpenes. Increasing the level of insoluble hydrocarbons would decrease the concentration of soluble terpinen- 4-ol and the effect would be reduced.

INTERACTIONS BETWEEN MONOTERPENES IN TEA TREE OIL 497 CONCLUSIONS The ndings of this study demonstrate that the inherent physical properties of aqueous tea tree oil dispersions will signi cantly in uence the actions of its individual components. Different terpenoid components of tea tree oil can interact in an aqueous dispersion to either reduce or increase antimicrobial ef cacy, depending upon their relative concentrations and the overall susceptibility of the target micro-organism. Non-oxygenated monoterpene hydrocarbons such as c-terpinene and p-cymene appear to produce antagonistic effects against more tolerant micro-organisms. The most probable explanation for this antagonism is that the non-aqueous monoterpene hydrocarbon phase reduces aqueous terpene solubility and, therefore, the microbial availability of the active components. Such effects may have signi cant implications with regard to the ef cacy of formulations containing tea tree oil as the principal antimicrobial or preservative constituent. ACKNOWLEDGEMENTS Thanks to S. Grif n for performing the gas chromatographic analyses and J. Edlin for performing microbiological assays included in this study. REFERENCES Aono, R. and Inoue, A. (1998) Organic solvent tolerance in microorganisms. In Extremophiles: Microbial Life in Extreme Environments eds Horikoshi, K. and Grant, W.D., pp. 287±309. New York: Wiley- Liss. Brophy, J.J., Davies, N.W., Southwell, I.A., Stiff, I.A. and Williams, L.R. (1989) Gas chromatographic quality control for oil of Melaleuca terpinen-4-ol type (Australian tea tree). Journal of Agricultural and Food Chemistry 37, 1330±1335. Carson, C.F. and Riley, T.V. (1995) Antimicrobial activity of the major components of the essential oil of Melaleuca alternifolia. Journal of Applied Bacteriology 78, 264±269. Cox, S.D., Mann, C.M., Markham, J.L. et al. (2000) The mode of antimicrobial action of the essential oil of Melaleuca alternifolia (tea tree oil). Journal of Applied Microbiology 88, 170±175. Grif n, S.G., Wyllie, S.G., Markham, J.L. and Leach, D.N. (1999) The role of structure and molecular properties of terpenoids in determining their antimicrobial activity. Flavour and Fragrance Journal 14, 322±332. Helander, I.M., Alakomi, H.-L., Kyosti, L.-K. et al. (1998) Characterization of the action of selected essential oil components on Gramnegative bacteria. Journal of Agricultural Food Chemistry 46, 3590±3595. International Standards Organization (1996) Oil of Melaleuca, terpinen-4-ol type (tea tree oil). ISO 4730, 1996 (E). Geneva: International Standards Organization, Kazmi, S.J.A. and Mitchell, A.G. (1978) Preservation of solubilized and emulsi ed systems. I: correlation of mathematically predicted preservative availability with antimicrobial activity. Journal of Pharmaceutical Sciences 67, 1260±1265. Kurup, T.R.R., Wan, L.S.C. and Chan, L.W. (1991) Availability and activity of preservatives in emulsi ed systems. Pharmaceutica Acta Helvetiae 66, 76±81. Mann, C.M., Cox, S.D. and Markham, J.L. (2000) The outer membrane of Pseudomonas aeruginosa NCTC 6749 confers tolerance to the essential oil of Melaleuca alternifolia (tea tree oil). Letters in Applied Microbiology 30, 294±297. Markham, J.L. (1999) Biological activity of tea tree oil. In Tea Tree, the Genus Melaleuca eds Southwell, I. and Lowe, R., pp. 169±190. Netherlands: Harwood Academic Publishers. Sikkema, J., de Bont, J.A.M. and Poolman, B. (1995) Mechanisms of membrane toxicity of hydrocarbons. Microbiological Reviews 59, 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. Acta Horticulturae 334, 265±275.