Assessment of tea tree oil skin penetration

The export and subsequent distribution to consumers of Australian products containing essential oils, such as tea tree oil (TTO), is governed by regional regulatory bodies that have the ability to grant or deny access to markets based on their own assessment of perceived threat to end-user safety.

Unfortunately, the safety of topical application of the components of many essential oils is confounded by a lack of transdermal penetration data, which is absolutely necessary to adequately inform the opinions and recommendations of such regulatory bodies. Assessment of toxicity data relies on the existence of a reasonable understanding of the likely exposure levels to the various components of topical formulations containing essential oils, and conversely even the most reliable transdermal penetration data is difficult to assess if there is no understanding of the potential benefit or toxicity of the chemical species crossing the skin. In 2005, the European regulatory body, the Scientific Committee on Consumer Products (SCCP), received a letter of concern from the Danish Environmental Protection Agency claiming that a significant risk to humans existed and subsequently published an opinion on the safety of Australian TTO which concluded that it had insufficient information to determine whether TTO was in fact safe, and required the industry to put together a safety dossier containing data to support its continued topical application to humans. The current scarcity of experimental evidence, combined with the lipophilic nature of many of the components of TTO, has led to the assumption that TTO is readily absorbed through human skin.

A black mark against TTO by the SCCP would obviously have major consequences for the industry in Australia, and even threaten the adoption of the SCCP ruling by other significant export markets. Therefore, the Australian industry responded by forming a technical and safety committee which, with funding from the Rural Industries Research and Development Council, set out to perform a number of definitive studies, including percutaneous absorption assessment to address the concerns of the SCCP. Our role in this work was to assess the skin penetration and retention of TTO components applied topically to human skin in vitro.

Methods

The TTO used in the study, supplied by the Australian Tea Tree Industry Association, was analysed according to International Standard ISO 4730:2004 and shown to contain acceptable levels of each of the 15 defined TTO constituent components. All sample analyses were performed using gas chromatography mass spectrometry by the phytochemistry laboratory at the Department of Primary Industries in NSW due to its familiarity and expertise in these procedures.

In order to assess the penetration of this essential oil, we applied TTO (i) as the neat oil and (ii) as a 20% solution in ethanol, chosen to represent a formulation likely to disrupt the barrier properties of the skin and potentially facilitate the transdermal delivery of TTO. The TTO formulations were applied to the skin of 3 different female donors, n=6 per donor, mounted in Franz-type horizontal, static diffusion cells and dosed using ”in use” application conditions (10 µL/cm2). The penetration of TTO components into a receptor phase containing 4% bovine serum albumin dissolved in phosphate buffered saline, in order to increase the solubility of lipophilic TTO components, followed over a 24 hour period. In addition to skin penetration, we quantified the retention of TTO components in the epidermis at the end of the study and examined the effect of partial occlusion on penetration rates following application of the pure oil. In addition to the above studies, the rate and extent of evaporation of TTO applied to filter paper discs was determined in order to better understand the depletion of TTO from surface applied doses. All statistical analysis was performed using Analysis of Variance (ANOVA) and Tukey’s post hoc testing with Minitab software v13.32, significance was taken at p<0.05.

Results

Terpinen-4-ol and a-terpineol were found to penetrate epidermal membranes and could be quantified in the receptor phase in samples from each of the three different skin donors (Table 1). No other TTO components could be readily detected in the receptor phase samples at any time point following application of either the pure oil or 20% TTO formulation. There was a slight variability in the retention of TTO between the 3 different skin donors. There was a higher retention of terpinen-4-ol in skin donor number 2 when TTO was applied as the pure oil, though no significant differences when applied as the 20% TTO formulation.

Peaks relating to other oil components, a-terpineol and a mixture of other sesquiterpinenes, were also present in the extracted skin samples, but none of these components were found penetrating through the epidermal membrane into the receptor phase.

Following partial occlusion of the donor chamber compartment of the Franz cell, there was a significant increase in the amount of terpinen-4-ol recovered from the epidermis following application of the pure oil. In addition, all of the other 15 TTO components could now be identified in the membrane extraction samples, though in the majority of cases peaks were at or below the limit of confident quantification of the assay. Recovery of TTO components within the experimental system (donor chamber, epidermis and receptor solution) was extremely low following non-occluded application onto human epidermis. The percentage of applied amount of terpinen-4-ol recovered in the system varied between 5.5% and 10% when applied in the neat oil, and 2.4% and 4.4% following application as the 20% oil formulation.

Our evaporation studies showed that oil applied to filter paper at a dose rate of 1.43 mg/cm2 evaporated within 1 hour whereas increasing to 7.37 mg/cm2, approximating the dermal penetration application (8.9 mg/cm2 = 10.0 µl/cm), there was a retention of oil for only 4 hours.

Discussion

The current study has shown that following application of pure TTO under normal “in use” conditions, a small quantity of TTO components, 1.1%-1.9% and 2-4% of the applied amount following application of a 20% TTO solution and pure TTO respectively, were found to penetrate into or through human epidermis. The largest TTO component penetrating the skin was terpinen-4-ol. Previous studies have also reported the penetration of terpinene-4-ol through human skin, for example in the infinite TTO oil dosing study of Reichling et al (2006), with the use of aqueous ethanol as a receptor solution, reported concentrations of approximately 4 µl/cm2 of terpinen-4-ol penetrating human epidermis in vitro over a 24 hour period, equating to, assuming a specific gravity of 0.933, a concentration of 4.28 mg/cm2 absorbed in 24 hours. Our current study found penetration rates under “in use” non-occluded conditions at least ten-fold lower than this in each of the 3 different skin donors we examined. The most likely explanation for the tenfold difference is the use of infinite and finite doses in the two studies. We have found a similar difference for the human skin penetration of oxybenzone (LogP 2.63) which has a similar lipophilicity to terpinen- 4-ol (LogP 2.99) (2001). This difference arises mainly from the much greater exposure associated with infinite dose application relative to “in use” conditions (this study). A further source for the greater exposure is a reduced evaporation in the infinite donor case relative to the “in use” study here, where the recovery was only 5.5% to 10% and was not reported by Reichling et al (2006). The higher penetration of terpinen-4-ol is also likely to be due to the inclusion of 50% ethanol in the receptor solution in the previous study that may have diffused into the epidermal membrane and affected its barrier function (see Brain et al, 1998). Following partial occlusion of the application site, this penetration increased to approximately 7% of the applied TTO. This data should now be used to expand risk assessment profiles of the use of TTO and reduce speculation that all lipophilic components of TTO are assumed to freely penetrate the skin. The evaporation of TTO following topical application supports the suggestion that TTO appears to be of little potential risk of transdermal absorption following topical application as most of the oil applied is lost into the atmosphere. However, the potential toxicity of terpinene-4-ol at the levels found in the epidermis in this study needs to be addressed as a separate issue.

The full details of this paper together with the suite of studies submitted to the SCCP in 2007 can be found on the RIRDC government website (www.rirdc.gov.au).

References

1 Brain K.R., Walters K.A., Watkinson A.C. Investigation of skin permeation in vitro, In: Roberts M.S. and Walters K.A. (Eds), In: Dermal Absorption and Toxicity Assessment , Marcel Dekker Inc. New York, 1998, 161-188.

2 Cross S.E., Jiang R., Benson H.A.E., Roberts M.S. Can increasing the viscosity of formulations be used to reduce the human skin penetration of the sunscreen oxybenzone?, J. Invest. Dermatol. 117 (2001) 147-150.

3 Reichling J., Landvatter U., Wagner H., Kostka K.H., Schaefer U.F. In vitro studies on release and human skin permeation of Australian tea tree oil (TTO) from topical formulations, Eur. J. Pharm. Biopharm. 64 (2006) 222—228.

AUTHORS

Sheree E. Cross and Michael S. Roberts are of the Therapeutics Research Unit, School of Medicine, University of Queensland, Princess Alexandra Hospital, Woolloongabba, QLD 4102 Australia.

Michael Russell and Ian Southwell are of the NSW Department of Primary Industries, Wollongbar Agricultural Institute, Wollongbar, NSW 2477 Australia.