Content uploaded by Malcolm Douglas
Author content
All content in this area was uploaded by Malcolm Douglas on Nov 20, 2017
Content may be subject to copyright.
Essential oils from New Zealand manuka: triketone and other
chemotypes of Leptospermum scoparium
q
Malcolm H. Douglas
b
, John W. van Klink
a,*
, Bruce M. Smallfield
b
, Nigel B. Perry
a
,
Rosemary E. Anderson
b
, Peter Johnstone
c
, Rex T. Weavers
d
a
Department of Chemistry, Plant Extracts Research Unit, New Zealand Institute for Crop & Food Research Limited, University of Otago,
P.O. Box 56, Dunedin, New Zealand
b
New Zealand Institute for Crop & Food Research Limited, Invermay Agricultural Centre, Private Bag 50034, Mosgiel, New Zealand
c
AgResearch Limited, Invermay Agricultural Centre, Private Bag 50034, Mosgiel, New Zealand
d
Department of Chemistry, University of Otago, P.O. Box 56, Dunedin, New Zealand
Received 17 December 2003; accepted 17 March 2004
Dedicated to the memory of our seven Crop & Food Research colleagues killed in a plane crash, 6 June 2003
Available online 23 April 2004
Abstract
The triketone chemotype of manuka, Leptospermum scoparium (Myrtaceae), is commercially important because of its antimi-
crobial activity. Oils from 36 individual plants on the East Cape of New Zealand all showed similar high triketone contents (>20%
total triketones) with little seasonal variation. Analyses of oils from 261 individual manuka plants collected from 87 sites throughout
New Zealand showed that the high triketone chemotype was localised on the East Cape, although oils with triketone levels up to
20% were found in the Marlborough Sounds area of the South Island. Cluster analysis revealed other chemotypes localised on other
areas. Ten further chemotypes are described: a-pinene; sesquiterpene-rich with high myrcene; sesquiterpene-rich with elevated
caryophyllene and humulene; sesquiterpene-rich with an unidentified sesquiterpene hydrocarbon; high geranyl acetate; sesquiter-
pene-rich with high c-ylangene + a-copaene and elevated triketones; sesquiterpene-rich with no distinctive components; sesquiter-
pene-rich with high trans-methyl cinnamate; high linalol; and sesquiterpene-rich with elevated elemene and selinene. Some of the
chemotypes contained aroma compounds at relatively high levels, with a geranyl acetate-rich oil being most notable. Possible origins
for this complex array of chemotypes are proposed.
Ó2004 Elsevier Ltd. All rights reserved.
Keywords: Leptospermum scoparium; Myrtaceae; Manuka; Essential oil; Chemotype; Triketones; Sesquiterpenes; Monoterpenes; Geranyl acetate;
Methyl cinnamate
1. Introduction
Manuka, Leptospermum scoparium J.R. et G. Forst.
(Myrtaceae), grows as a shrub or small tree throughout
New Zealand (Wardle, 1991) and in eastern Australia
(Brophy et al., 1999b). In New Zealand it is valued for its
essential oil (Porter, 2001). There is growing interna-
tional interest in triketone-rich manuka oils due to their
activity against Gram-positive bacteria including antibi-
otic-resistant strains (Christoph et al., 2000, 1999;
Harkenthal et al., 1999; Kim and Rhee, 1999; Lis-Balchin
et al., 2000; Porter and Wilkins, 1998). However, there
are major variations in the chemical composition of oils
from this one species, leading to potential confusion in
the marketplace.
In our previous work on Leptospermum essential oils
we found that oils from Australian L. scoparium had
higher monoterpene levels than most New Zealand
L. scoparium, and low or no triketones (Perry et al.,
1997). Brophy et al. (1999b) also found no triketones in
oils from Australian L. scoparium var. scoparium and
q
Supplementary data associated with this article can be found, in the
online version, at doi:10.1016/j.phytochem.2004.03.019.
*
Corresponding author. Tel.: +6434798357; fax: +6434798543.
E-mail address: vanklinkj@crop.cri.nz (J.W. van Klink).
0031-9422/$ - see front matter Ó2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2004.03.019
Phytochemistry 65 (2004) 1255–1264
www.elsevier.com/locate/phytochem
PHYTOCHEMISTRY
var. eximum but did find triketones (Brophy et al.,
1999a, 2000) in the closely aligned Leptospermum sub-
groups of Thompson (1989). Within New Zealand, our
original investigations were based on analyses of plants
grown from seed collected from 15 sites around the
country and grown at one site. Therefore, environmental
effects on oil composition were eliminated and it was
possible to sample all the plants on the same day to
eliminate possible seasonal effects. From this work we
identified three regional chemotypes: high a-pinene
(mean 22.5%) in the far North, high triketones (mean
total 32.5%) on the East Cape, and a type containing a
complex of sesquiterpenes (mean total 64.7%) over the
rest of the North and South Islands (Perry et al., 1997).
This study showed that manuka oil composition was
largely genetically controlled since oil compositions of
plants grown at the study site were similar to those of
plants growing at the seed source site (Perry et al., 1997).
Table 1
Composition of L. scoparium oils from the East Cape seasonal studya
GC peak Compound name Mean (SD) Minimum Maximum
1a-Pinene 0.7 (0.9) 0.0 3.6
2b-Pinene 0.3 (0.4) 0.0 1.9
3 Myrcene 0.3 (0.3) 0.0 1.3
4p-Cymene 0.3 (0.2) 0.0 0.9
5 1,8-Cineole 0.7 (0.4) 0.0 1.8
6b-Ocimene 0.2 (0.2) 0.0 0.7
7c-Terpinene nd – –
8a-Terpinolene nd – –
9 Linalol 0.3 (0.1) 0.1 0.8
10 Terpinene-4-ol 0.1 (0.1) 0.0 0.2
11 a-Terpineol 0.1 (0.1) 0.0 0.4
13 Citronellol 0.0 (0.0) 0.0 0.1
14 Citronellyl formate 0.2 (0.2) 0.0 0.8
15 Methyl citronellate nd – –
16 cis-Methyl cinnamate 0.0 (0.1) 0.0 0.1
17 Methyl geranate 0.0 (0.1) 0.0 0.1
18 Citronellyl acetate nd – –
19btrans-Methyl cinnamate/a-cubebene 2.8 (1.2) 0.4 5.3
20+21cc-Ylangene + a-copaene 5.7 (3.0) 2.0 13.6
22 Geranyl acetate 0.0 (0.0) 0.0 0.1
23 b-Elemene 0.4 (0.4) 0.0 1.6
24 a-Gurjunene 0.8 (0.7) 0.1 4.1
25 b-Caryophyllene 1.5 (1.2) 0.0 3.9
26 C15H24 0.3 (0.2) 0.0 1.1
27 Aromadendrene 1.7 (1.3) 0.0 5.5
28 C15H24 1.3 (2.3) 0.0 7.9
29 a-Humulene 3.6 (3.3) 0.4 11.9
30 C15H24 0.6 (0.4) 0.0 1.4
31 a-Amorphene 2.6 (1.5) 0.0 6.1
32 b-Selinene 1.8 (1.3) 0.0 5.8
33 C15H24 2.4 (2.3) 0.0 9.1
34da-Selinene/viridiflorene 3.0 (1.6) 1.1 8.2
35 a-Muurolene 0.9 (0.2) 0.5 1.4
36 c-Cadinene 0.2 (0.2) 0.0 0.7
37 trans-Calamenene 15.6 (3.7) 7.9 22.7
38 d-cadinene 4.5 (1.2) 2.7 8.6
39 Flavesone 8.2 (2.0) 4.9 12.3
40 Cadina-1,4-diene 0.0 (0.0) 0.0 0.1
41 Calacorene 0.2 (0.1) 0.0 0.4
42 Not identified 0.4 (0.1) 0.0 1.3
43 Caryophyllene epoxide 0.7 (0.4) 0.3 1.9
44 Not identified 0.2 (0.2) 0.0 1.2
45 Isoleptospermone 6.2 (1.9) 2.7 11.5
46 Leptospermone 16.6 (4.4) 10.6 29.4
47 b-Eudesmol 0.8 (0.2) 0.4 1.3
48 a-Eudesmol 0.0 (0.0) 0.0 0.2
50 Grandiflorone nm – –
a
GC peak areas as % of total peak area (nd, not detected; nm, not measured).
b
Not resolved but 1H NMR spectra showed no trans-methyl cinnamate in East Cape samples.
c
Resolution varied so peaks summed.
d
Not resolved.
1256 M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264
Porter and Wilkins (1998) proposed a fourth chemotype
rich in linalol and eudesmols, found in the Nelson area
of the South Island. They also proposed a fifth chemo-
type high in myrcene and eudesmols, but considered it a
variant of the linalol–eudesmol chemotype.
Leptospermum scoparium grows throughout New
Zealand in habitats ranging from lowland to sub alpine
areas (Wardle, 1991) and our previous sampling (Perry
et al., 1997) was not representative of this range. The
aim of this research was to determine if the triketone-
rich chemotype was present in other New Zealand lo-
cations and define the boundaries of this chemotype
around the East Cape. Before undertaking this survey
we needed to know whether seasonal variation in es-
sential oil composition was important, as it would be
logistically impossible to collect samples in a national
survey all at the same time. Work on manuka in the
Nelson area of the South Island showed seasonal
variation in the leaf oil composition of young plants
(<6 year old); significantly higher relative levels of
monoterpenes were present during the spring/summer
period of foliage growth (Porter et al., 1998). We
Fig. 1. Sampling sites and triketone levels in New Zealand L. scoparium.
M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264 1257
therefore investigated the seasonal variation of the
triketone-rich oil from East Cape manuka before
sampling countrywide. We now report here the results
of this seasonal study and of a national survey of the
essential oil composition of manuka, covering many
habitats throughout the North and South Islands of
New Zealand.
2. Results
Manuka foliage samples were collected and oils ex-
tracted using standardised methods. Forty-eight GC
peaks were quantified, with two reference peaks used for
retention time correction (Perry et al., 1997). We have
now identified most of the peaks (Table 1) by GC–MS
and correlations with the results of Porter and Wilkins
(1998).
2.1. Seasonal and individual variation within the East
Cape triketone chemotype
An East Cape population (close to site 19, Fig. 1) was
sampled at approximately monthly intervals from Oc-
tober 1996 to September 1997, with foliage from four
different individual plants harvested each time. The
composition of the oils from 36 individual plants is
summarised in Table 1.
The mean levels of total triketones ranged from
about 25% to about 35%, with leptospermone the main
triketone in all the oils. Triketone levels did not show
any obvious trends with the seasons (Fig. 2). Statistical
analysis using a SAS General Linear Model showed
only a few significant (P<0:05) differences in triketone
composition over the study period. For example, total
triketone levels in the November samples were signifi-
cantly (P<0:05) higher than in the March and
September samples. The mono- and sesquiterpene
components of the oil samples also did not show any
clear-cut seasonal patterns. We found significant dif-
ferences in oil yield over the study period (data not
shown) due primarily to seasonal differences in leaf/
twig ratio.
2.2. Triketone levels throughout New Zealand
In the regional variation study, 43 sites in the South
Island were sampled during January–February 1999 and
January–February 2000, and 44 sites in the North Island
during February–March 2001. The 87 collection sites
are indicated on Fig. 1, which also summarises triketone
levels, with one segment for each of the three oils
analysed from each site.
There were major differences in triketone levels, both
between sampling sites and between individuals at a
given site. As expected, all three plants at the East Cape
site 19 had total triketones >20% (Fig. 1), as found
previously (Perry et al., 1997) and in the seasonal study
(see above). These were the highest levels of triketones
found and there was a rapid drop-off in triketone levels
at neighbouring sites. For example, 75 km South at site
25 all three plants had <10% total triketones, though
one of the plants from site 18 (50 km W) had total
triketones >20%.
The South Island sampling also showed a very lo-
calised ‘‘hot spot’’ for high triketone oils in the Marl-
borough Sounds region, with three plants on D’Urville
Island (sites 51 and 52) having 15–20% total triketones
(Fig. 1).
2.3. Other chemotypes in New Zealand
To search for other chemotypes within this complex
dataset of 261 oils with 48 GC peak levels quantified, we
used non-hierarchical cluster analysis. This is an ex-
ploratory technique used to identify and separate like
groups by computation, based on Euclidean metric
(Krzanowski and Marriott, 1995). The number of clus-
ters separated within the data set is arbitrary, and we
have chosen 10 because these correspond to distinct
chemotypes, as shown in Table 2 and Fig. 3 and sum-
marised below:
Cluster 1,a-Pinene Chemotype, oils were found in
the North of both islands. Cluster 2,Sesquiterpene/
Myrcene Chemotype, included only five of the plants
surveyed and was very localised in the Waikato region.
Cluster 3,Caryophyllene/Humulene Chemotype, oils
contained mostly sesquiterpenes. Cluster 4,Sesquiter-
pene 33 Chemotype, was found only in the North
Island. Cluster 5,Geranyl Acetate Chemotype, included
only three plants in the survey and is a previously
unreported manuka chemotype. Cluster 6,c-Ylangene/
Fig. 2. Triketone levels in L. scoparium oils from the East Cape sea-
sonal study.
1258 M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264
a-Copaene Chemotype, includes the 15–20% triketone
oils from the Marlborough Sounds, but the high
ylangene/copaene level distinguishes it from the East
Cape high-triketone chemotype. Cluster 7,Sesquiter-
pene plus East Cape Triketone Chemotype, included the
East Cape triketone chemotype and a chemotype with
Table 2
Compositional clusters for L. scoparium oils (mean GC peak areas as % of total peak area)
Compound Cluster number (number of plants in each cluster)
1 (26) 2 (5) 3 (51) 4 (28) 5 (3) 6 (22) 7 (46) 8 (18) 9 (35) 10 (27)
a-Pinene 21.5a0.6 1.4 1.9 1.2 1.0 1.8 1.6 4.1 1.5
b-Pinene 5.8 0.1 0.7 0.8 0.1 0.8 0.2 0.4 3.7 0.3
Myrcene 1.3 20.0 1.5 0.8 0.9 0.8 1.2 1.6 2.3 1.5
p-Cymene 1.1 0.2 0.3 0.5 0.1 0.3 0.3 0.6 0.4 0.3
1,8-Cineole 8.5 0.3 0.7 1.2 0.1 1.9 0.7 3.2 3.0 1.9
b-Ocimene 0.1 0.0 0.0 0.1 0.0 0.3 0.8 0.3 0.3 0.3
c-Terpinene 1.1 0.3 0.5 0.6 0.2 0.3 0.5 0.7 0.4 0.3
a-Terpinolene 0.2 0.0 0.0 0.0 0.0 0.1 0.1 0.2 0.1 0.1
Linalol 8.5 2.2 1.1 0.8 1.0 3.5 3.3 3.7 12.6 3.2
Terpinene-4-ol 0.9 0.1 0.2 0.3 0.1 0.3 0.2 0.4 0.5 0.3
a-Terpineol 1.9 0.1 0.2 0.4 0.1 0.6 0.2 0.8 0.9 0.5
Citronellol 0.3 0.0 0.0 0.0 0.1 0.2 0.1 0.5 2.1 0.5
Citronellyl formate 0.7 0.0 0.2 0.2 0.0 2.7 0.6 1.0 1.8 2.3
methyl citronellate 0.5 0.2 0.1 0.1 0.3 0.7 0.1 0.2 1.2 0.2
cis-Methyl cinnamate 0.4 0.2 1.1 1.4 0.8 0.6 1.0 0.4 0.4 0.2
Methyl geranate 3.4 1.0 0.8 0.2 0.9 1.5 0.5 1.1 2.1 1.5
Citronellyl acetate 0.5 0.2 0.0 0.0 0.1 0.4 0.1 1.3 1.4 0.6
trans-Methyl cinnamate/a-cubebene 2.1 6.0 5.3 5.3 3.4 7.3 6.0 17.0b9.0 10.6
c-Ylangene + a-copaene 4.6 0.8 2.3 2.1 0.1 25.6 5.4 4.0 5.1 6.0
Geranyl acetate 0.1 0.5 0.4 0.5 48.6 0.2 0.1 0.5 0.3 0.7
b-Elemene 1.1 5.9 4.4 2.6 1.0 2.6 0.6 6.1 4.1 10.8
a-Gurjunene 0.2 0.6 1.0 0.5 0.3 0.6 1.2 0.9 0.6 0.6
b-Caryophyllene 5.2 7.7 8.3 2.7 0.9 3.2 5.4 4.0 4.7 3.2
C15H24 0.1 0.1 0.3 1.4 1.9 0.2 0.5 0.3 0.2 0.5
Aromadendrene 0.3 0.2 0.7 0.9 0.2 2.4 3.9 1.0 0.6 1.2
C15H24 0.8 0.4 1.0 2.4 0.8 1.5 2.2 1.0 0.6 0.8
a-Humulene 1.6 6.6 9.1 3.1 2.5 3.1 4.8 6.4 3.3 2.6
C15H24 0.3 0.3 0.5 2.0 1.3 0.8 1.1 1.1 0.9 0.9
a-Amorphene 1.1 1.2 3.1 2.0 0.9 2.0 5.5 1.9 1.4 2.8
b-Selinene 3.5 3.2 6.0 1.3 3.1 4.4 5.5 3.6 3.6 10.0
C15H24 (sesquiterpene JJ) 0.6 1.5 1.8 14.5 4.4 0.8 1.4 2.1 0.9 1.3
a-Selinene/viridiflorene 3.3 5.0 8.6 10.8 2.3 5.8 7.0 7.3 5.6 13.5
a-Muurolene 0.4 1.9 1.7 4.5 3.8 1.0 1.0 1.9 1.4 3.1
c-Cadinene 0.9 0.5 0.8 1.6 0.8 0.8 1.6 0.8 0.5 0.5
trans-Calamenene 2.7 4.1 7.2 4.5 2.5 3.9 7.0 5.1 4.3 3.3
d-Cadinene 1.1 2.1 3.7 4.1 0.9 2.1 5.2 2.6 1.7 1.1
Flavesone 0.9 1.4 3.1 2.7 1.3 1.6 3.6 1.7 1.1 0.7
Cadina-1,4-diene 0.2 0.1 0.4 1.1 0.6 1.0 1.4 0.2 0.2 0.1
Calacorene 0.2 0.1 0.2 0.3 0.1 0.4 0.3 0.4 0.5 0.3
Not identified 0.1 0.0 0.1 0.1 0.1 0.1 0.2 0.2 0.3 0.4
Caryophyllene epoxide 0.2 0.3 0.4 0.6 0.1 0.3 0.5 0.2 0.1 0.2
Not identified 0.4 0.2 0.4 0.3 0.3 0.4 0.5 0.3 0.2 0.3
Isoleptospermone 0.3 0.5 0.9 0.5 0.1 1.3 1.8 0.3 0.4 0.3
Leptospermone 1.0 3.1 1.7 0.8 0.3 3.4 4.5 0.8 1.2 0.7
b-Eudesmol 1.1 3.9 2.2 1.1 0.3 0.7 0.8 0.8 1.3 0.5
a-Eudesmol 1.0 3.8 1.6 0.7 0.6 0.5 0.5 0.5 1.2 0.9
Grandiflorone 0.0 0.6 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0
Tm 56.3 25.5 8.2 8.4 53.8 15.3 11.0 18.0 37.3 15.9
Ts 30.9 50.6 65.6 65.1 29.5 64.3 63.6 52.5 43.3 64.7
Tt 2.2 5.0 5.8 4.0 1.6 6.3 9.9 2.9 2.8 1.7
Oil yieldc0.5 0.3 0.3 0.3 0.3 0.3 0.2 0.3 0.4 0.4
Tm, total monoterpene; Ts, total sesquiterpenes; Tt, total triketones.
a
Bold numbers refer to the distinguishing features of each cluster.
b
Identified as trans-methyl cinnamate by 1H NMR in all cluster 8 oil samples.
c
ml per 100 g dry weight.
M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264 1259
a complex of sesquiterpenes at similar levels. The ses-
quiterpene plants were mostly in the North of the
South Island. Cluster 8,Methyl-Cinnamate/Sesquiter-
pene Chemotype, oils were analysed by 1H NMR to
confirm the presence of high levels of trans-methyl
cinnamate since this compound co-eluted with d-cub-
ebene on GC. Cluster 9,Linalol Chemotype, oils also
had higher levels of citronellol, citronellyl formate and
methyl citronellate that contributed to a noticeable
citrus aroma. Cluster 10,Elemene/Selinene Chemotype,
oils were found predominantly in the South Island and
only one site in the North Island.
2.4. Inter-relationship of clusters
Based on the cluster means, a dendrogram was con-
structed that allowed the similarity of each cluster to its
nearest neighbour to be placed in context (Fig. 4). Of
interest is the close alignment of the three mostly South
Island clusters, 8, 9 and 10, which are linked with cluster
6, also predominantly in the South Island. North Island
clusters 2 and 3 are closely aligned, while cluster 7,
which is present in both islands, is similar to clusters 2, 3
and 6. Clusters 1 and 5, the two high monoterpene
chemotypes, are the most different.
Fig. 3. Cluster analysis of essential oil composition of L. scoparium in New Zealand.
1260 M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264
3. Discussion
We have shown that the volatile oil composition of
manuka, L. scoparium, in New Zealand is extremely
variable. We propose 11 chemotypes with intriguing
patterns in their geographic distribution (Figs. 1 and 3).
These chemotypes are distinguished by different levels of
four classes of natural products: monoterpenes and
sesquiterpenes, both products of the isoprenoid bio-
synthetic pathway; methyl cinnamate, from the phenyl-
propanoid pathway; and triketones, from an as yet
uncharacterised biosynthetic pathway.
We have shown that the high triketone (>20% total
triketones) chemotype previously identified (Perry et al.,
1997; Porter and Wilkins, 1998) has a very limited dis-
tribution, being found only on the East Cape of the
North Island (Fig. 1). There were only slight seasonal
differences in triketone levels (Fig. 2). This means that
commercial producers of antibacterial manuka oils
could harvest throughout the year, as long as this was
consistent with maximising foliage yield and regrowth of
the plants.
The only other area of New Zealand that yielded
relatively high triketone manuka oils was the Marlbor-
ough Sounds (Fig. 1), although the level of triketones
was lower than on the East Cape and generally below
the standard expected for good antimicrobial activity
(Christoph et al., 2000). We do not know why both the
East Cape, and the Marlborough Sounds should have
manuka with higher levels of triketones. One potential
link could be by Maori travelling to D’Urville Island in
the Marlborough Sounds (Thomson, 1918), since Maori
sometimes took plants and seed on their journeys
(Brailsford, 1997). Another, perhaps stronger, linkage is
that both high-triketone regions are known to have been
plant refugia, being less affected by past environment
obliteration (as discussed by McGlone (1985) and col-
leagues (McGlone et al., 2001)). As noted, Brophy et al.
(Brophy et al., 1999b) did not find triketones in oils from
their Australian samples of L. scoparium, but did find
leptospermone in an oil from L. glabrescens (Brophy
et al., 1999a).
A high a-pinene chemotype in manuka from the
North of the North Island has previously been proposed
(Perry et al., 1997) and we now confirm this chemotype
and report that it is also present in an area on the West
Coast of the South Island (cluster 1, Fig. 3). Porter and
Wilkins (1998) also identified a pinene-rich chemotype
type III on the eastern side of the South Island, but
because the a-pinene level was so high (63%), this was
probably a sample of Kunzea ericoides that had been
misidentified as manuka (N. Porter, personal commu-
nication). Volatile monoterpenes have been related to
fire ecology (Owens et al., 1998), and we suggest that the
presence of high monoterpene oils in Northland and on
the Westland fringe could be explained by frequent fir-
ing and regeneration cycles (Ogden et al., 1998; Burrows
et al., 1979). In contrast to Australia, fire adaptation is
not a common feature of the New Zealand flora (Ogden
et al., 1998), although manuka has serotinous capsules
Fig. 4. Dendrogram of the cluster analysis.
M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264 1261
that release seeds en masse after fire, and of all the native
woody plants manuka (together with kanuka (K. erico-
ides)) regrows most strongly after disturbance (Wardle,
1991). In Northland, the L. scoparium examined by
Harris (2002) had strongly serotinous capsules, but his
Westland samples did not.
We have discovered a new monoterpene chemotype
that is high in geranyl acetate and very localised in the
South of the North Island (cluster 5, Fig. 3). Monoter-
pene acetates are less flammable than monoterpene hy-
drocarbons (Owens et al., 1998), so this is less likely to
relate to fire adaptation. Geranyl acetate is a perfumery
ingredient (Arctander, 1969), so this chemotype might
be commercially useful. Brophy et al. (2000) found
geranyl acetate (4–11%) in oils from L. variabile.We
have also identified a mixed monoterpene + sesquiter-
pene chemotype (cluster 9, Fig. 3) with elevated levels of
linalol, another aroma compound. This chemotype is
similar to the type-II north-west Nelson chemotype de-
scribed by Porter and Wilkins (1998).
The general sesquiterpene chemotype previously
described (Perry et al., 1997) is here subdivided into
different chemotypes based on particular major com-
pounds. These chemotypes are not likely to have any
commercial use, because these sesquiterpene hydrocar-
bons are not generally distinguished by biological ac-
tivity or aroma. However, our discovery of relatively
high levels of methyl cinnamate in many South Island
samples, at levels up to 30%, is interesting because this
compound is a perfume and flavour ingredient (Arct-
ander, 1969). Methyl cinnamate has been noted in sev-
eral Australian Leptospermum species, with the highest
levels in an oil from L. riparium (Brophy et al., 1999a).
Overall, the volatile oils of L. scoparium in New
Zealand are complex mixtures showing a very high de-
gree of infraspecific variation with some geographical
patterns. The oils from Australian Leptospermum spe-
cies are similarly complex mixtures, variable between
species and occasionally within species (Brophy et al.,
1999b), but the infraspecific variation has not been ex-
plored in the same detail as reported here. There is also
large variability in morphological characters of L.
scoparium, both in Australia (Thompson, 1989) and in
New Zealand (Yin Ronghua et al., 1984; Wardle, 1991).
Both morphological and metabolic variability might
be explained by polyphyletic origins for New Zealand’s
L. scoparium.
The first myrtaceous pollens in New Zealand and
Australia that are similar to Leptospermum are recog-
nisable in the Palaeocene some 60 million yeas ago (Lee
et al., 2001). Since that time about 82 species have
evolved, predominantly in eastern Australia (McGlone
et al., 2001; Thompson, 1989). Thompson, in her de-
finitive study of the Leptospermum genus (Thompson,
1989), notes that L. scoparium in New Zealand is not a
primitive Leptospermum and cannot be an ancient
Myrtaceae plant of the New Zealand Palaeocene as
detailed by Fleming (1975). During the Cenozoic period,
the land shape and environment of New Zealand
changed dramatically with tectonic uplift, volcanism,
sea inundation and long term shifts in climate (Lee et al.,
2001; McGlone, 1985; McGlone et al., 2001). Consid-
ering the Pleistocene extinction of such Australian gen-
era as Eucalyptus and Acacia, there is a possibility that
the Palaeocene Leptospermum was replaced by more
recent genetic stock following long distance dispersal
(Pole, 1994; McGlone et al., 2001). A phylogenetic study
could help to explain chemotype differences in
L. scoparium as well as giving answers to the evolution
and dispersal of the genus Leptospermum.
4. Experimental
4.1. Plant materials
In October 1996 a population of L. scoparium near Te
Araroa on the East Cape of New Zealand was chosen
for a seasonal study. Each month foliage (leaf and stem)
samples were collected from four individual plants.
Samples were not collected in January, May and August
of 1997 due to extreme weather conditions at those
times.
For the regional variation study, three manuka plants
were sampled for foliage (leaf and stem) at each site. The
South Island survey was undertaken from January to
April 1999, (site nos. 45–47 and 63–87), with nine ad-
ditional sites (nos. 48–62) harvested in the Marlborough
Sounds in January 2000 (Fig. 2). The North Island
sampling was undertaken in February–March 2001 (site
nos. 1–44, Fig. 2). The plants at each site were generally
within 50 m of each other, but at four sites (1, 12, 17 and
34; Fig. 2) a single plant was at a greater distance.
Foliage samples (3–5 kg) from both seasonal and
regional studies were bagged, numbered and sent by
courier post to the Invermay Research Centre. Voucher
samples were retained, while the bulk samples were
slowly dried at 3 °C and 17% relative humidity for
20 days prior to distillation
4.2. Distillation method
Each dried foliage sample was chopped into ap-
proximately 25 mm lengths and steam distilled under
standardised conditions for 2 h. The distillation was
essentially the same as that method reported previously
(Perry et al., 1997). However, due to frothing of the
condensate, the set-up was modified to include a hot-
water jacketed separating funnel, which improved oil
recovery and measurement by helping the oil particles
coalesce. Oil samples were stored at )20 °C in glass
vials.
1262 M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264
4.3. Essential oil analyses and component identifications
GC analyses were performed on a Perkin–Elmer
Autosystem gas chromatograph (under the control of
PE-Omega software) equipped with a split injector
(100:1, 260 °C) and a 10 m J &W DB-1 column with a
0.25 mm ID and 0.25 lm film. Oil samples (10 ll) were
diluted in cyclohexane (1 ml) containing 0.5% n-dode-
cane (C12) and n-octadecane (C18 ), and 0.5 ll subsam-
ples were injected using an autosampler. The carrier gas
used was hydrogen with a linear velocity of 50 cm/s. For
the seasonal studies, the column temperature was pro-
grammed 80–160 °Cat5°C/min. For the regional var-
iation study the upper limit on the temperature program
was extended to 180 °C in order to detect grandiflorone.
Peaks were detected by a flame ionisation detector
(350 °C). Levels of 48 peaks were recorded (Table 2).
The alkanes C12 and C18 were used as reference peaks to
correct for retention time fluctuations and to help in
peak matching.
GC–MS data were collected on a Finnigan GC8000,
70 eV, detector voltage 250 V, fitted with a 30 m J &D B-
1 column (0.25 mm ID, 1 lm film thickness) and helium
carrier gas with a flow rate of 1.5 ml/min, split ratio 10:1.
The oven temperature program was set at 50–100 °Cat3
°C/min, then 100–180 °Cat5°C/min then 180–250 °Cat
30 °C/min, followed by a 10 min hold at 250 °C. Peak
scans were compared with those in a series of libraries
(NBS, WILEY, LIBTX) on the Finnigan MASLAB
data system, using the reverse fit factor for peak
matching. Peaks were matched to those reported earlier
(Porter and Wilkins, 1998). To determine of Kovats
retention indices (RIs), an altered temperature program
(50–250 °Cat5°C/min) was used. Indices were calcu-
lated by comparison with a separate injection of a
standard n-alkane (even number) mixture.
Total monoterpenes in each oil were calculated by
summing the levels of peaks 1–11, 13–15, 17, 18 and 22;
total sesquiterpenes by summing peaks 20 + 21, 23–38,
40–44, 47 and 48; and total triketones by summing
peaks 39, 45, 46 and 50.
4.4. Statistical analyses
The data for the seasonal trial was analysed using
SAS (SAS Institute, version 6.15) software. The data for
the regional variation study was subjected to non-hier-
archical classification (k-means clustering) in which the
Euclidean distance between classes is maximised
(Krzanowski and Marriott, 1995). From the cluster
means a dendrogram was drawn from the minimum
spanning tree obtained from single linkage cluster
analysis. The implementation of the clustering algo-
rithms used was in GenStat release 6.1 (Genstat, 2003).
Supplementary material (Tables 5 and 6) is available
which details sampling sites, the mean levels of mono-
terpenes, sesquiterpene and triketones, and the cluster
placement for each plant oil analysed.
Acknowledgements
We thank J.-P. Dufour for GC–MS analyses; N.
Porter, J. Douglas, J. Follett, N. and H. Fulton, H., A.
and J. Douglas, and M. Kerr and staff at Tairawhiti
Pharmaceuticals for field assistance; and N. Porter, J.
Douglas and W. Harris for discussions. This research
was supported by the New Zealand Foundation for
Research, Science and Technology, the Ministry of
Agriculture and Fisheries Sustainable Farming Fund
and New Zealand Manuka Products Ltd.
References
Arctander, S., 1969. Perfume and flavor chemicals (aroma chemicals).
Arctander, S., Montclair.
Brailsford, B., 1997. The Tattooed Land. Stoneprint Press, Hamilton.
Brophy, J.J., Goldsack, R.J., Bean, A.R., Forster, P.I., Lepschi, B.J.,
1999a. Leaf essential oils of the genus Leptospermum (Myrtaceae)
in Eastern Australia. Part 3. Leptospermum arachnoides and allies.
Flavour and Fragrance Journal 14, 85–91.
Brophy, J.J., Goldsack, R.J., Bean, A.R., Forster, P.I., Lepschi, B.J.,
1999b. Leaf essential oils of the genus Leptospermum (Myrtaceae)
in Eastern Australia. Part 5. Leptospermum continentale and its
allies. Flavour and Fragrance Journal 14, 98–104.
Brophy, J.J., Goldsack, R.J., Bean, A.R., Forster, P.I., Lepschi, B.J.,
2000. Leaf essential oils of the genus Leptospermum (Myrtaceae) in
eastern Australia. Part 6. Leptospermum polygalifolium and its
allies. Flavour and Fragrance Journal 15, 271–277.
Burrows, C.J., McQueen, D.R., Esler, A.E., Wardle, P., 1979. New
Zealand heathlands. In: Specht, R.L. (Ed.), Heathlands and
Related Shrublands. Descriptive Studies. Elsevier, Amsterdam,
pp. 339–364.
Christoph, F., Kaulfers, P.-M., Stahl-Biskup, E., 2000. A comparative
study of the in vitro antimicrobial activity of tea tree oils s.l. with
special reference to the activity of b-triketones. Planta Medica 66,
556–560.
Christoph, F., Kubeczka, K.-H., Stahl-Biskup, E., 1999. The compo-
sition of commercial manuka oils from New Zealand. Journal of
Essential Oil Research 11, 705–710.
Fleming, C.A., 1975. The geological history of New Zealand and its
biota. In: Kuschel, G. (Ed.), Biogeography and ecology in New
Zealand. Junk, The Hague, pp 1–86.
GenStat for Windows. Release 6.1, sixth ed. VSN International Ltd.
Harkenthal, M., Reichling, J., Geiss, H.-K., Saller, R., 1999.
Comparative study on the in vitro antibacterial activity of
Australian tea tree oil, cajuput oil, niaouli oil, manuka oil, kanuka
oil, and eucalyptus oil. Pharmazie 54, 460–463.
Harris, W., 2002. Variation of inherent seed capsule splitting in
populations of Leptospermum scoparium (Myrtaceae) in New
Zealand. New Zealand Journal of Botany 40, 405–417.
Kim, E.H., Rhee, G.J., 1999. Activities of ketonic fraction from
Leptospermum scoparium alone and synergism in combination with
some antibiotics against various bacterial strains and fungi.
Yakhak Hoechi 43, 716–728.
Krzanowski, W.J., Marriott, F.H.C., 1995. Multivariate Analysis. Part
2. Classification, Covariance Structures and Repeated Measure-
ments. Arnold, London.
M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264 1263
Lee, D.E., Lee, W.G., Mortimer, N., 2001. Where and why have all the
flowers gone? Depletion and turnover in the New Zealand
Cenozoic angiosperm flora in relation to palaeogeography and
climate. Australian Journal of Botany 49, 341–356.
Lis-Balchin, M., Hart, S.L., Deans, S.G., 2000. Pharmacological and
antimicrobial studies on different Tea-tree oils (Melaleuca alterni-
folia,Leptospermum scoparium or Manuka and Kunzea ericoides or
Kanuka), originating in Australia or New Zealand. Phytotherapy
Research 14, 623–629.
McGlone, M.S., 1985. Plant biogeography and the late Cenozoic history
of New Zealand. New Zealand Journal of Botany 23, 723–749.
McGlone, M.S., Duncan, R.P., Heenan, P.B., 2001. Endemism, species
selection and the origin and distribution of the vascular plant flora
of New Zealand. Journal of Biogeography 28 (2), 199–216.
Ogden, J., Basher, L., McGlone, M., 1998. Fire, forest regeneration
and links with early human habitation: evidence from New
Zealand. Annals of Botany 81, 687–696.
Owens, M.K., Lin, C.D., Taylor, C.A., Whisenant, S.G., 1998.
Seasonal patterns of plant flammability and monoterpenoid con-
tent in Juniperus ashei. Journal of Chemical Ecology 24, 2115–2129.
Perry, N.B., Brennan, N.J., van Klink, J.W., Harris, W., Douglas,
M.H., McGimpsey, J.A., Smallfield, B.M., Anderson, R.E., 1997.
Essential oils from New Zealand manuka and kanuka: chemotax-
onomy of Leptospermum. Phytochemistry 44, 1485–1494.
Pole, M., 1994. The New Zealand flora – entirely long-distance
dispersal? Journal of Biogeography 21, 625–635.
Porter, N., 2001. Manuka. The good oil from New Zealand.
HerbalGram 53, 26-30.
Porter, N.G., Smale, P.E., Nelson, M.A., Hay, A.J., van Klink, J.W.,
Dean, C.M., 1998. Variability in essential oil chemistry and plant
morphology within a Leptospermum scoparium population. New
Zealand Journal of Botany 36, 125–133.
Porter, N.G., Wilkins, A.L., 1998. Chemical, physical and antimicro-
bial properties of essential oils of Leptospermum scoparium and
Kunzea ericoides. Phytochemistry 50, 407–415.
Thompson, J., 1989. A revision of the genus Leptospermum (Myrta-
ceae). Telopea 3, 301–448.
Thomson, J.A., 1918. Maori rock-quarries on D’Urville Island. New
Zealand Journal of Science and Technology 1, 321–322.
Wardle, P., 1991. The Vegetation of New Zealand. Cambridge
University Press, Cambridge, UK.
Yin, R., Mark, A.F., Wilson, J.B., 1984. Aspects of the ecology of the
indigenous shrub Leptospermum scoparium (Myrtaceae) in New
Zealand. New Zealand Journal of Botany 22, 483–507.
1264 M.H. Douglas et al. / Phytochemistry 65 (2004) 1255–1264
