Content uploaded by Christine Frances Carson
Author content
All content in this area was uploaded by Christine Frances Carson
Content may be subject to copyright.
Abstract. Objective and Design: To evaluate potential anti-
inflammatory properties of tea tree oil, the essential oil steam
distilled from the Australian native plant, Melaleuca alterni-
folia.
Material and Methods: The ability of tea tree oil to reduce
the production in vitro of tumour necrosis factor-
a
(TNF
a
),
interleukin (IL)-1
b
, IL-8, IL-10 and prostaglandin E
2
(PGE
2
)
by lipopolysaccharide (LPS)-activated human peripheral
blood monocytes was examined.
Results: Tea tree oil emulsified by sonication in a glass tube
into culture medium containing 10% fetal calf serum (FCS)
was toxic for monocytes at a concentration of 0.016% v/v.
However, the water soluble components of tea tree oil at con-
centrations equivalent to 0.125% significantly suppressed
LPS-induced production of TNF
a
, IL-1
b
and IL-10 (by ap-
proximately 50%) and PGE
2
(by approximately 30%) after
40 h. Gas chromatography/ mass spectrometry identified terpin-
en-4-ol (42%),
a
-terpineol (3%) and 1,8-cineole (2%, respec-
tively, of tea tree oil) as the water soluble components of tea tree
oil. When these components were examined individually, only
terpinen-4-ol suppressed the production after 40 h of TNF
a
,
IL-1
b
, IL-8, IL-10 and PGE
2
by LPS-activated monocytes.
Conclusion: The water-soluble components of tea tree oil can
suppress pro-inflammatory mediator production by activated
human monocytes.
Key words: Tea tree oil – Monocytes – Interleukin-1 –
Tumour necrosis factor-
a
– Prostaglandin E
2
Introduction
Tea tree oil is the essential oil steam distilled from the Aus-
tralian native plant, Melaleuca alternifolia. Tea tree oil con-
tains over 100 components, the majority being monoterpene
and sesquiterpene hydrocarbons and their alcohols. The anti-
bacterial properties of tea tree oil have now been well docu-
mented, and there are susceptibility data on a wide range of
bacteria [1–9]. There is also considerable information iden-
tifying the components of tea tree oil active against bacteria
and yeasts [6]. Until recently there have only been anecdotal
claims about tea tree oil’s anti-inflammatory activity. The
only specific report on anti-inflammatory properties of tea
tree oil has been of an in vitro study published as an abstract
[10] in which the addition of tea tree oil to lipopolysac-
charide-primed neutrophils inhibited superoxide release by
approximately 85%. The concentration of tea tree oil used
was 0.05% v/v.
Several studies have investigated the anti-inflammatory
properties of compounds also found in tea tree oil in a rat car-
rageenan-induced hind paw oedema model [11–14]. These
studies investigated the immunomodulatory effects of oils
from Bupleurum gibraltaricum, Bupleurum fruticescens,
Zingiber cassumunar and Salvia sclarea, respectively.
Collectively, these studies concluded that
a
-pinene [11, 12],
a
-terpinene [13], terpinen-4-ol [13],
a
-terpineol [14] and
linalool [14] may have direct or indirect anti-inflammatory
activity. The implications of these studies for the activity of
tea tree oil are uncertain since the composition of the oils and
the concentrations used were generally in excess of those
found in tea tree oil. However, they go some way to identify-
ing essential components of tea tree oil which possess anti-
inflammatory activity. Whether there are synergistic or anta-
gonistic interactions of several potential immunomodulators
when they are in the same chemical mixture is not known.
Some of the components of tea tree oil previously shown to
have anti-bacterial properties, e. g. terpinen-4-ol and
a
-terpi-
neol, are also anti-inflammatory in vivo [6].
In this study of the anti-inflammatory activity of tea tree
oil in vitro, human peripheral blood monocytes were used as
Inflamm. res. 49 (2000) 619–626
1023-3830/00/110619-08 $ 1.50+0.20/0
© Birkhäuser Verlag, Basel, 2000
Inflammation Research
Terpinen-4-ol, the main component of the essential oil of
Melaleuca alternifolia (tea tree oil), suppresses inflammatory
mediator production by activated human monocytes
P. H. Hart
1
,C.Brand
1
,C.F.Carson
2
, T. V. Riley
2
,R.H.Prager
3
and J. J. Finlay-Jones
1
1
Department of Microbiology and Infectious Diseases, School of Medicine, Flinders University, GPO Box 2100, Adelaide, 5001, Australia,
Fax: 6188276 8658, e-mail: Prue.Hart@flinders.edu.au
2
Department of Microbiology, University of Western Australia, Nedlands, Australia 6907
3
School of Chemistry, Physics and Earth Sciences, Flinders University, GPO Box 2100, Adelaide, 5001, Australia
Received 1 January 2000; returned for revision 11 May 2000; accepted by R. O. Day 21 June 2000
Correspondence to: P. H. Hart
a model for tissue macrophages. Upon activation with mole-
cules such as LPS, these cells produce many mediators in-
cluding the central mediators of inflammation, TNF
a
and
IL-1
b
. Other important monocyte/macrophage-derived
mediators of inflammation include IL-8, IL-10 and PGE
2
.
Together with other products of activated macrophages,
these molecules can damage tissue or, in turn, activate other
cells to produce pro-inflammatory mediators. It was hypo-
thesised that if anti-inflammatory, tea tree oil would reduce
the production in vitro of TNF
a
, IL-1
b
, IL-8, IL-10 and
PGE
2
by LPS-activated monocytes.
Materials and methods
Tea tree oil and its major components
Tea tree oil was kindly provided by Australian Plantations (Wyrallah,
NSW, Australia). The levels of 15 major components of the batch of tea
tree oil used in all experiments (97/03) was examined by Wollongbar
Agricultural Institute, Wollongbar, Australia, according to the inter-
national standard for tea tree oil [15] and were as follows: terpinen-4-
ol (41.6%),
g
-terpinene (21.5%),
a
-terpinene (10.0%), terpinolene
(3.5%),
a
-terpineol (3.1%),
a
-pinene (2.4%), 1,8-cineole (2.0%), p-
cymene (1.8%), aromadendrene (1.1%),
d
-cadinene (1.0%), limonene
(0.9%), ledene (0.9%), globulol (0.5 %), sabinene (0.4%) and viridi-
florol (0.2%) (Table 1). For individual study, terpinen-4-ol and
a
-ter-
pineol were obtained from Fluka, Buchs, Switzerland and 1,8-cineole
from Sigma Chemical Co., St Louis, MO.
Monocyte isolation and culture
Human monocytes were isolated from peripheral blood as published
[16, 17] to >93% purity by countercurrent centrifugal elutriation and
cultured in RPMI-1640 medium (Cytosystems, Castle Hill, Australia)
supplemented with 13.3 mM NaHCO
3
, 2 mM glutamine, 50 mM
b
-
mercaptoethanol, 100 U/ml penicillin, 100 mg/ml streptomycin, and
2 nM 3-(N-morpholino)propanesulphonic acid with an osmolality of
290 mmol/kg H
2
O (‘complete RPMI’). During isolation and subsequent
culture of all cells, extreme care was taken to limit LPS contamination
of buffers and culture fluids [16, 17].
For measurement of regulation of cytokine production, freshly
isolated monocytes were cultured at 10
6
cells/ml with tea tree oil or its
water-soluble components and, unless otherwise indicated, 1% fetal calf
serum (FCS). LPS from Escherichia coli 0111:B4 (Sigma) was added
to give a final concentration of 500 ng/ml. Triplicate cultures for each
test variable were incubated at 37°C in 5% CO
2
. After 20 or 40 h,
cultures were centrifuged and the supernatants harvested for measure-
ment of TNF
a
, IL-1
b
, IL-8, IL-10 and PGE
2
. For cultures of adherent
cells, polystyrene 24-well tissue culture plates (Falcon, Becton Dickin-
son, Lincoln Park, NJ) were used.
Measurement of toxicity of tea tree oil and its water soluble
components
Metabolically active cells were enumerated using the CellTiter 96
AQ
ueous
non-radioactive cell proliferation assay (Promega Corporation,
Madison, WI). Monocytes (2 ¥ 10
5
/200 ml) were cultured under ad-
herent conditions for 17 or 37 h with tea tree oil or its water-soluble
components, with or without LPS. The MTS/PMS substrate (20 ml) was
added and the amount of formazan product measured spectrophoto-
metrically at 490 nm hourly for the next 4 h.
ELISAs for TNF
α
, IL-1
β
, IL-8 and IL-10
Culture supernatants were stored at –20°C until used. TNF
a
, IL-8 and IL-
10 were measured by sandwich ELISA using mAbs to human TNF
a,
IL-
8 and IL-10 purchased as antibody pairs from PharMingen, San Diego,
CA. The antibody pair for measuring IL-1
b
was purchased from Endo-
gen, Woburn, MA. One antibody from each pair was purchased as a bio-
tinylated antibody. The assays were sensitive to levels of >40 pg/ml.
Radioimmunoassay for PGE
2
The levels of PGE
2
in the culture supernatants were determined by
radioimmunoassay using competitive adsorption to dextran-coated
charcoal (3H-PGE
2
, Amersham, Bucks, UK; PGE
2
antiserum, Sigma).
The assay was sensitive to levels of > 30 pg/ml.
Solubility of tea tree oil in water
A sample of tea tree oil (500 mg) was rapidly stirred in water (50 ml)
for 20 h at 20°C. The layers were separated and the aqueous layer fil-
tered through filter paper, then extracted with ether (3 ¥ 25 ml). The
ether extracts were dried (Na
2
SO
4
), and the solvent carefully removed
by distillation at 50°C in a closed system. The residue weighed 80 mg
giving a solubility of 1.6 g/l in water. The organic phase and the oil
adhering to the filter paper were extracted into ether, dried (Na
2
SO
4
)
and evaporated as above to give 400 mg oil. Gas chromatographic
analysis as above showed the composition to be 84% terpinen-4-ol, 7%
a
-terpineol and 3% 1,8-cineole.
For separation of tea tree oil preparations in culture medium into
water and oil soluble components, very similar procedures were adop-
ted. Immediately before separation, samples of 250 ml were vigorously
shaken before separation of the aqueous and oil layers in a separating
funnel. The aqueous fraction was filtered through paper, extracted with
ether (2 ¥ 25 ml), dried with sodium sulphate and the ether removed by
distillation. The oily fraction, and the oil adhering to the filter paper,
was extracted in a similar manner. The residues from both fractions were
analysed by gas chromatography/mass spectrometry (GC/MS) (Table 1).
Expression of results
Cytokine measurements were performed on samples from triplicate cul-
tures and the LPS-induced level was normalised to 100%. The mean
values from each set of replicates were used to determine the mean
+ SEM for n donors. For comparison of responses by cell populations
from a number of different donors, the Student’s paired t test was used.
A value of P < 0.05 was considered significant.
Results
Regulation of monocyte mediator production by tea tree oil
emulsions prepared in glass tubes
To assess the regulatory properties of tea tree oil on inflam-
matory mediator production by monocytes, it was necessary
to prepare emulsions of tea tree oil in culture medium. If tea
tree oil was dissolved in propylene glycol or ethanol, the tea
tree oil separated from its diluent before a dilution of 1 in 100
required for monocyte culture could be made. Instead, dilu-
tions of tea tree oil were performed in glass tubes with
medium containing 10% FCS, followed by sonication for
20 seconds immediately before use. Emulsions of tea tree oil
were toxic to adherent monocytes to some extent at con-
centrations greater than, or equal to 0.004 % (v/v). For mono-
cytes from 4 donors, tea tree oil at 0.004% caused a mean
620 P. H. Hart et al. Inflamm. res.
toxicity of 9 ± 5% (not significant), at 0.008% a mean toxi-
city of 20 ± 11% (not significant) and at 0.016%, a mean
toxicity of 69 ± 17% (P = 0.03). No viable cells were detec-
ted after incubation of higher concentrations of tea tree oil
with adherent monocytes for 20 h. As shown in Fig. 1, the
suppressive properties of tea tree oil on LPS-induced TNF
a
production paralleled the toxic properties of tea tree oil.
Similar results were detected for regulation of LPS-induced
IL-1
b
, IL-8, IL-10 and PGE
2
(data not shown).
Removal of components of tea tree oil toxic to monocytes by
dilution in plastic tubes
We next examined the effect of tea tree oil emulsions prep-
ared in polystyrene plastic, rather than glass tubes. It was
noted that the plastic adsorbed a considerable proportion of
the oil; we hypothesised that this approach would allow a
separation of the water soluble components into the culture
medium. Initially, 10% FCS was included in the diluting
medium; however, concentrations of tea tree oil greater than,
or equal to 0.06% destroyed all monocytes over a 20 h in-
cubation period. Removal of serum from the culture medium
used to serially dilute the tea tree oil was then evaluated. Tea
tree oil mixtures were vortexed for 1 minute immediately
before further dilution. Although the diluting medium was
serum-free, the monocytes were cultured in complete RPMI
supplemented with 1% FCS for 20 and 40 h. Under these con-
ditions, the monocyte toxic components in tea tree oil did not
partition into the culture medium; no toxicity was detected in
the cultures of monocytes harvested after 40 h (Fig. 2).
Regulation of monocyte mediator production by tea tree oil
solutions prepared in plastic tubes
The tea tree oil solutions had no effect on inflammatory
mediator production by monocytes incubated in the absence
of LPS. Addition of LPS stimulated mediator production to
levels shown in Table 2; in Fig. 2, the mediator production
induced by LPS has been normalised to 100%. The dose-
dependent suppressive effects of tea tree oil are shown in
Fig. 2. As shown in Table 3, the suppressive properties of tea
tree oil generally increased with longer incubation times.
Components of tea tree oil responsible for the anti-inflam-
matory activity of tea tree oil
The data presented in Figure 1 indicated that tea tree oil con-
tained components that were toxic to monocytes in culture.
When these were absent, other components of tea tree oil
suppressed the production of inflammatory mediators by
stimulated monocytes. The components of tea tree oil that
Vol. 49, 2000 Anti-inflammatory properties of tea tree oil 621
Table 1. Components of tea tree oil separating into aqueous and oil phases under different conditions.
Glass + Sonication Plastic
10% FCS 0% FCS 10% FCS
Component Concentration in
tea tree oil Aqueous % Oil % Aqueous % Oil % Aqueous %
1.
a
-pinene 2.4 0.2 2.8 – 2.8 –
2. sabinene 0.4 – 0.9 – 0.9 –
3.
a
-terpinene 10.0 0.8 9.1 – 9.3 –
4. limonene 0.9 – 1.1 – 1.0 –
5. p-cymene 1.8 0.7 6.1 – 6.8 0.8
6. 1,8-cineole 2.0 2.8 1.7 3.3 1.8 2.5
7.
g
-terpinene 21.5 2.3 23.0 – 21.0 1.3
8. terpinolene 3.5 – 3.8 – 3.5 –
9. terpinen-4-ol 41.6 80.4 34.3 83.8 37.0 80.8
10.
a
-terpineol 3.1 6.3 2.7 6.5 2.9 6.3
11. aromadendrene 1.1 – 1.4 – 1.3 –
12. ledene 0.9 – 1.2 – 1.1 –
13.
d
-cadinene 1.0 – 1.2 – 1.1 –
14. globulol 0.5 – 0.5 – 0.5 –
15. viridiflorol 0.2 – 0.4 – 0.4 –
Fig. 1. The effect of tea tree oil diluted in glass tubes with medium con-
taining 10% FCS on TNF
a
production by monocytes in culture for 20 h.
Monocytes from 4 donors were incubated in triplicate for 20 h with LPS
(500 ng/ml) and decreasing amounts of tea tree oil. The LPS-induced level
was normalised to 100%; the mean result from each experiment was used
to calculate the mean + SEM. Regulation of LPS-induced TNF
a
levels is
shown by histograms (left axis). The mean percentage + SEM of viable
cells in the cultures is shown by the line (right axis). An asterisk indicates
a significant reduction in TNF
a
production or monocyte viability.
622 P. H. Hart et al. Inflamm. res.
Fig. 2. The effect of tea tree oil diluted in polystyrene plastic tubes with serum-free medium on monocytes in culture for 40 h. Monocytes from 5
donors were incubated in triplicate under conditions allowing monocyte adherence for 40 h with LPS (500 ng/ml) and decreasing amounts of tea tree
oil. The LPS-induced level was normalised to 100 %; the mean result from each experiment was used to calculate the mean + SEM. Regulation of
LPS-induced A. TNF
a
, B. IL-1
b
, C. IL-8, D. IL-10 and E. PGE
2
levels is shown by histograms. For A, the mean percentage + SEM (if sufficiently
large) of viable cells in the cultures is shown by the line (right axis). An asterisk indicates a significant reduction in mediator production.
Vol. 49, 2000 Anti-inflammatory properties of tea tree oil 623
Fig. 3. The effect of the water-soluble components of tea tree oil diluted in glass tubes with serum-free medium on monocytes in culture for 40 h.
Monocytes from 3 donors were incubated in triplicate under conditions allowing monocyte adherence for 40 h with LPS (500 ng/ml) and concentra-
tions of terpinen-4-ol,
a
-terpineol and 1,8-cineole equivalent to those found in 0.125% (shown as a) and 0.062 % (shown as b) tea tree oil. The LPS-
induced level was normalised to 100%; the mean result from each experiment was used to calculate the mean + SEM. Regulation of LPS-induced A.
TNF
a
, B. IL-1
b
, C. IL-8, D. IL-10 and E. PGE
2
levels is shown by histograms. For A, the mean percentage + SEM (if sufficiently large) of viable
cells in the cultures is shown by the line (right axis). An asterisk indicates a significant reduction in mediator production.
624 P. H. Hart et al. Inflamm. res.
Table 2. Induction of cytokines and PGE
2
by LPS-activated adherent
human peripheral blood monocytes.
Mediator production (mean ± SEM, ng/ml)
20 h (n = 10) 40 h (n = 5)
TNF-
a
6.2 ± 1.8 6.8 ± 1.6
IL-1
b
8.2 ± 1.8 8.1 ± 2.3
IL-8 250.8 ± 28.2 448.5 ± 58.0
IL-10 0.3 ± 0.1 0.6 ± 0.2
PGE
2
6.1 ± 0.8 6.7 ± 1.9
Table 3. Suppressive effect of the water-soluble components of tea tree
oil on LPS-induced cytokine and PGE
2
production by adherent human
peripheral blood monocytes.
Water-soluble components Terpinen-4-ol
of tea tree oil
20 h (n = 10) 40 h (n = 5) 20 h (n = 6) 40 h (n = 3)
TNF-
a
23% 58% 58 % 43%
IL-1
b
27% 57% 30 % 38%
IL-8 19% – 31% 20 %
IL-10 – 58% – 36%
PGE
2
– 32% 25 % 32%
– = no significant decrease.
P < 0.05 for all changes shown.
partitioned into the culture medium when diluted in glass or
polystyrene plastic tubes, in the absence or presence of 10%
FCS, were determined. Table 1 shows that under the initial
conditions of dilution in glass tubes with medium containing
10% FCS, the aqueous phase contained detectable levels of
a
-pinene,
a
-terpinene, p-cymene, 1,8-cineole,
g
-terpinene,
terpinen-4-ol and
a
-terpineol. With dilution in polystyrene
plastic tubes with medium containing 10% FCS, the aqueous
phase (which still contained significant toxic activity for
monocytes) contained p-cymene, 1,8-cineole,
g
-terpinene,
terpinen-4-ol and
a
-terpineol. Using plastic tubes and
serum-free diluting medium, three components of tea tree oil
were detected in the aqueous phase, namely 1,8-cineole, ter-
pinen-4-ol and
a
-terpineol, with terpinen-4-ol representing
84% of the recovered material. The identity and relative
proportion of these components were identical to those in the
aqueous phase when 500 mg tea tree oil was rapidly mixed
with 50 ml water in a glass vessel.
Effects of terpinen-4-ol,
a
-terpineol and 1,8-cineole on
monocyte mediator production
The data in Figure 2 and Table 3 suggested that water-soluble
components of tea tree oil can suppress inflammatory media-
tor production by monocytes in vitro. The GC/MS analyses
suggested that terpinen-4-ol,
a
-terpineol and 1,8-cineole
were the components of tea tree oil that were partially soluble
in tissue culture medium. It was therefore hypothesised that
these three components were not toxic and could suppress
inflammatory mediator production by monocytes activated
with LPS in vitro.
The amounts of terpinen-4-ol,
a
-terpineol and 1,8-cineo-
le investigated were calculated according to their concentra-
tion in a 0.125% solution of tea tree oil. A concentration
equal to their level in a solution of 0.062% tea tree oil was
also investigated. To maximise the amount of terpinen-4-ol,
a
-terpineol and 1,8-cineole that remained in the aqueous
phase, dilutions were performed in glass tubes. As for the
cultures of Figure 2 and Table 3, dilutions were performed
in the absence of serum and monocytes were cultured in
medium containing 1% FCS.
When monocytes were incubated with 0.052 and 0.026%
terpinen-4-ol, 0.004 and 0.002%
a
-terpineol and 0.0025 and
0.0013% 1,8-cineole (concentrations equivalent to those in
0.125 and 0.062% tea tree oil, respectively), no toxicity for
monocytes was detected (Fig. 3). After incubation for 20 h,
0.052% terpinen-4-ol significantly suppressed LPS-induced
TNF
a
, IL-1
b
, IL-8 and PGE
2
production (Table 3). The sup-
pression of LPS-induced IL-10 was not significant. After
incubation for 40 h, the lower concentration of terpinen-4-ol
significantly suppressed LPS-induced IL-1
b
and PGE
2
, as
well as TNF
a
(Fig. 3). After incubation for 20 or 40 h, the
other water soluble components of tea tree oil, namely
a
-ter-
pineol and 1,8-cineole, were without effect (Fig. 3).
Discussion
The potential for tea tree oil to have anti-inflammatory activ-
ity (as anecdotally reported) was investigated in vitro using
activated human monocytes. However, when tea tree oil was
emulsified in tissue culture medium containing 10 % FCS in
glass tubes and incubated at a dilution of 0.016% with
adherent monocytes for 20 h, approximately 30% of the cells
retained viability. Tea tree oil components toxic for human
cells in culture have been reported previously. In a Swedish
study [18], tea tree oil at concentrations of approximately
0.03% (300 mg/ml) and 0.05% (500 mg/ml) caused 50% kil-
ling of gingival epithelial cells and fibroblasts, respectively,
after 24 h incubation. In another Australian study [19], the
toxicity of tea tree oil and its main water-soluble components
(terpinen-4-ol,
a
-terpineol and 1,8-cineole) on five human
cell lines was examined after 4 and 24 h incubation. Suscep-
tibility of the cell lines to the toxicity of tea tree oil varied
with 50% killing after 4 h at a concentration of 0.28 % for
HeLa (epithelioid), K562 (chronic myelogenous leukaemia)
and Hep G2 (hepatocellular carcinoma) cells. A concentra-
tion of 0.06% tea tree oil over 4 h caused 50% killing of
CTVR-1 (B cell-derived from bone marrow of a patient with
acute myeloid leukaemia) and Molt-4 (lymphoblastic leu-
kaemia) cell lines. After 24 h, the IC
50
for HeLa cells was
0.27%, for K562 cells 0.03 %, CTVR-1 cells 0.03 %, Molt-4
cells 0.03% and Hep G2 cells 0.002 % [19]. Thus, monocytes
cultured in vitro have a susceptibility to the toxic effects of
tea tree oil similar to that of K562, CTVR-1 and Molt-4 cells.
When tea tree oil was diluted to a concentration which was
not toxic or minimally toxic to monocytes (0.008 % or lower),
there were some suggestions of reduced production of
inflammatory mediators. However, it was difficult to rule out
the effects of lowered LPS-stimulated production of media-
tors by dying cells. In subsequent experiments, partitioning
of tea tree oil components into aqueous and oil phases was
encouraged with the oil components adhering to the poly-
styrene plastic tubes. Inclusion of 10% serum provided suf-
ficient emulsifying activity such that tea tree oil was still
cytotoxic to monocytes. A similar effect was seen if lower
concentrations of serum were used in the presence of 0.001%
Tween 20 (data not shown). Only when tea tree oil was dilu-
ted with serum-free medium in polystyrene plastic tubes
were the components toxic for monocytes not dissolved.
Analysis of the components of tea tree oil remaining in the
culture medium suggested that either terpinen-4-ol (84% of
compounds identified),
a
-terpineol (7% of compounds iden-
tified) or 1,8-cineole (3 % of compounds identified) were
regulatory for inflammatory mediator production by LPS-
activated monocytes in vitro. Further studies suggested that
terpinen-4-ol was responsible in large part for the regulatory
effects of tea tree oil. The effects of terpinen-4-ol after 40 h
closely paralleled the effects of tea tree oil as a whole, sug-
gesting that if there were any other water-soluble anti-inflam-
matory components in tea tree oil, they were minor.
The plastic-non-adherent, water soluble components of a
concentration of tea tree oil of 0.125% were examined in the
experiments of Fig. 2 and Table 3. As it is recommended that
a 100% solution of tea tree oil is applied to skin, this con-
centration represents approximately one thousandth of that
used and is a level less than the water solubility of tea tree oil
(1.6 g/l). The ability of tea tree oil components to reach the
epidermal and dermal tissues and enter the systemic circula-
tion has not been measured. However, tea tree oil contains
several components known to enhance skin penetration of
other compounds, e.g. 1,8-cineole [20], limonene [21], terpi-
nen-4-ol [22] and
a
-terpineol [22]. The exact concentrations
of the water soluble components of tea tree oil in the culture
medium were not determined for each experiment but were
assumed to be similar to those of Table 1 (84% terpinen-4-
ol, 7%
a
-terpineol and 3% 1,8-cineole). To test the activity
of terpinen-4-ol,
a
-terpineol and 1,8-cineole in isolation,
calculations were made according to their concentration in
whole tea tree oil and therefore their approximate level in tea
tree oil solutions of 0.125%. Thus, in experiments in which
the independent activity of the components was analysed,
concentrations higher than in the aqueous phase of tea tree oil
were examined. It should be noted that even at these higher
concentrations, no toxicity was observed. 1,8-Cineole, the
main component of eucalyptus oil (85%) was not regulatory
in our study. We are uncertain why in a previous study it sig-
nificantly suppressed LPS-induced TNF
a
production in vitro
by monocytes at a concentration of 0.00001% [23] and fol-
lowing administration to asthmatic patients caused reduced
leukotriene B4 and prostaglandin E
2
production by mono-
cytes ex vivo [24]. Our study identified terpinen-4-ol, the
major component of tea tree oil (>30 %), as the component
with the ability to suppress the production of TNF
a
, IL-1
b
,
IL-8, IL-10 and PGE
2
by LPS-activated monocytes.
It was apparent from our study that all mediators released
by activated monocytes are not similarly regulated by the
water-soluble components of tea tree oil. In particular, little
regulation of LPS-induced IL-8 production was detected at
20 or 40 h. TNF
a
and IL-1
b
production was regulated at 20
h by the water-soluble components in tea tree oil and this
regulation was enhanced after 40 h incubation. In contrast,
regulation of LPS-induced IL-10 and PGE
2
production was
not detected until the later time point. This difference re-
flects, in part, the production of mediators with autocrine
regulatory activity. The greater regulation of TNF
a
is signi-
ficant as it is widely recognised that control of TNF
a
produc-
tion/activity can limit further production of other inflamma-
tory cytokines [25]. TNF
a
is regarded as central to the deve-
lopment and maintenance of chronic inflammation; this was
evidenced by the ability of anti-TNF
a
antibodies to reduce
inflammatory diseases such as rheumatoid arthritis [26].
Our study suggests that tea tree oil may potentially con-
trol inflammatory responses to foreign antigens in the skin.
With application of tea tree oil to skin, toxicity would be
limited and the anti-inflammatory water-soluble components
may penetrate into the vascularised dermis and regulate
inflammatory processes. This study supports the selection of
superior trees for propagation after identification of produc-
tive clones which produce tea tree oil high in terpinen-4-ol
[27]. Alternatively, an aqueous extract of tea tree oil which
produces a fraction very high in terpinen-4-ol, may be used
as an anti-inflammatory agent. As we are uncertain of the
concentrations of terpinen-4-ol that may penetrate beyond
the stratum corneum, the potential of tea tree oil as a topical
anti-inflammatory agent will only be confirmed by docu-
mentation of a reduction of inflammatory cells and mediators
in skin after application of tea tree oil.
Acknowledgments. This work was supported by Thursday Plantation
Pty. Ltd. and grants from the Rural Industries Research and Develop-
ment Corporation, Australia (UWA-43A and UF-5A).
References
[1] Walsh LJ, Longstaff J. The antimicrobial effects of an essential oil
on selected oral pathogens. Periodontology 1987; 8: 11–5.
[2] Altman PM. Australian tea tree oil. Aust J Pharm 1988; 69: 276–8.
[3] Altman PM. Australian tea tree oil – a natural antiseptic. Aust J
Biotechnol 1989; 3: 247–8.
[4] Carson CF, Riley TV. The antimicrobial activity of tea tree oil. Med
J Aust 1994a; 160: 236.
[5] Carson CF, Riley TV. Susceptibility of Propionibacterium acnes to
the essential oil of Melaleuca alternifolia. Lett Appl Microbiol
1994b;19: 24–5.
[6] Carson CF, Riley TV. Antimicrobial susceptibility of the major
components of the essential oil of Melaleuca alternifolia. J Appl
Bacteriol 1995a;78: 264–9.
[7] Carson CF, Riley TV. Toxicity of the essential oil of Melaleuca
alternifolia or tea tree oil. J Toxicol Clin Toxicol 1995b; 33: 193–4.
[8] Carson CF, Cookson BD, Farrelly HD, Riley TV. Susceptibility of
methicillin-resistant Staphylococcus aureus to the essential oil of
Melaleuca alternifolia. J Antimicrob Chemother 1995 c; 35:
421–4.
[9] Hammer KA, Carson CF, Riley TV. Susceptibility of transient and
commensal skin flora to the essential oil of Melaleuca alternifolia.
Am J Infect Control 1996; 24: 186–9.
[10] Pippin MA, Pabst KM, Pabst MJ, Haney L, Dabbous MKH.
Superoxide release by neutrophils is inhibited by tea tree oil. J
Dent Res 1994; 73: 259.
[11] Ocete MA, Risco S, Zarzuelo A, Jimenez J. Pharmacological
activity of the essential oil of Bupleurum gibraltaricum: anti-
inflammatory activity and effects on isolated rat uteri. J Ethno-
pharmacol 1989; 25: 305–13.
Vol. 49, 2000 Anti-inflammatory properties of tea tree oil 625
[12] Martin S, Padilla E, Ocete MA, Galvez J, Jimenez J, Zarzuelo A.
Anti-inflammatory activity of the essential oil of Bupleurum
fruticescens. Planta Med 1993; 59: 533–6.
[13] Pongprayoon U, Soontornsaratune P, Jarikasem S, Sematong T,
Wasuwat S, Claeson P. Topical anti-inflammatory activity of the
major lipophilic constituents of the rhizome Zingiber cassumunar.
Part I: The essential oil. Phytomedicine 1996/97; 3: 319–22.
[14] Moretti MDL, Peana AT, Satta M. A study on anti-inflammatory
and peripheral analgesic action of Salvia sclarea oil and its main
components. J Essent Oil Res 1997;9:199-204.
[15] Essential oils – oil of Melaleuca, terpinen-4-ol (tea tree oil). ISO-
4730 (1996) International Organisation for Standardisation,
Geneva, Switzerland.
[16] Hart PH, Vitti GF, Burgess DR, Whitty GA, Piccoli D, Hamilton
JA. Potential antiinflammatory effects of interleukin 4: suppres-
sion of human monocyte TNF, interleukin 1 and prostaglandin E
2
.
Proc Natl Acad Sci USA 1989; 86: 3803–7.
[17] Hart PH, Ahern MJ, Jones CA, Jones KL, Smith MD, Finlay-Jones
JJ. Synovial fluid macrophages and blood monocytes differ in their
responses to interleukin-4. J Immunol 1993; 151: 3370–80.
[18] Soderberg TA, Johansson A, Gref R. Toxic effects of some conifer
resin acids and tea tree oil on human epithelial and fibroblast cells.
Toxicology 1996; 107: 99–109.
[19] Hayes AJ, Leach DN, Markham JL. In vitro cytotoxicity of Aus-
tralian tea tree oil using human cell lines. J Essent Oil Res 1997;
9: 575–82.
[20] Obata Y, Takayama K, Machida Y, Nagai T. Combined effect of
cyclic monoterpenes and ethanol on percutaneous absorption of
diclofenac sodium. Drug Des Delivery 1991; 8: 137–44.
[21] Okabe H, Obata Y, Takayama K, Nagai T. Percutaneous absorption
enhancing effect and skin irritation of monocyclic monoterpenes.
Drug Des Delivery 1990; 6: 229–38.
[22] Magnusson BM, Runn P, Koskinen LOD. Terpene-enhanced
transdermal permeation of water and ethanol in human epidermis.
Acta Dermato-Venereol 1997; 77: 264–7.
[23] Juergens UR, Stober M, Vetter H. Inhibition of cytokine pro-
duction and arachidonic acid metabolism by eucalyptol (1,8-
cineole) in human monocytes in vitro. Eur J Med Res 1998a; 3:
508–10.
[24] Juergens UR, Stober M, Schmidt-Schilling L, Kleuyer T, Vetter H.
Antiinflammatory effects of eucalyptol (1,8-cineole) in bronchial
asthma: Inhibition of arachidonic acid metabolism in human blood
monocytes ex vivo. Eur J Med Res 1998b; 3: 407–12.
[25] Vassalli P. The pathophysiology of tumor necrosis factors. Annu
Rev Immunol 1992; 10: 411–36.
[26] Elliott MJ, Maini RN, Feldmann M, Long-Fox A, Charles P, Kat-
sikis P, et al. Treatment of rheumatoid arthritis with chimeric
monoclonal antibodies to tumor necrosis factor
a
. Arth Rheum
1993; 12: 1681–90.
[27] Williams LR. Clonal production of tea tree oil high in terpinen-4-
ol for use in formulation for the treatment of thrush. Complement
Ther Nurs Midwif 1998; 4: 133–6.
626 P. H. Hart et al. Inflamm. res.
