Medicinal and Aromatic Plants—Industrial Profiles


part of The Gordon and Breach Publishing Group



Yüklə 6,32 Mb.
Pdf görüntüsü
səhifə5/29
tarix27.08.2017
ölçüsü6,32 Mb.
1   2   3   4   5   6   7   8   9   ...   29
part of The Gordon and Breach Publishing Group.

IAN SOUTHWELL
30
varieties. Also 
α-thujene (1), ß-pinene (4), myrcene (5), terpinolene (13), α-terpineol (15)
and aromadendrene (16) were detected in the low cineole variety and dipentene (limonene)
(8) in the high cineole form. These workers were able to provide an estimate of the percentage
contributions made by each component identified that was remarkably close to the gas
chromatographic measurements of recent workers (Southwell and Stiff 1990; Kawakami et
al. 1990) (Table 1). The structures of these major components of tea tree oil are shown in
Figure 1
.
The next thorough investigation of tea tree oil chemistry involved reduced-pressure
spinning-band column distillation and gas chromatographic (GC) analysis of resulting
fractions (Laakso, 1966). This study did not report any new constituents. It does, however,
seem to be the first published report on the GC examination of tea tree oil even though
routine GC quality control of tea tree oil at the Museum of Applied Arts and Sciences in
Sydney had commenced around 1960 (Museum of Applied Arts and Sciences, unpublished
records). Laakso’s study also seemed to be the first to describe a terpinolene(13)-rich chemical
variety of M. alternifolia. Soon after this investigation Guenther (1968) reported a similar
fractionation-GC study from the Fritzsche Brothers laboratories.
Table 1 Estimate of the percentage composition of the oil of the terpinen-4-ol chemical
variety of M. linariifolia (Davenport et al. 1949) compared with the gas chromatographic
results of Southwell and Stiff (1990) and Kawakami et al. (1990)

Clone 1, now known to be M. linariifolia (Southwell et al. 1992). nr, not recorded; tr,
trace; 
a,b,c
estimated collectively.
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

TEA TREE CONSTITUENTS
31
Figure 1 Chemical structures of the significant constituents in tea tree oil
The first gas chromatography-mass spectrometry (GCMS) investigation of tea tree oil
(Swords and Hunter 1978) reported forty eight constituents of which only eight were
unassigned. Assignments were based on GCMS data with preparative GC, liquid
chromatography (LC) and infrared spectroscopy (IR) confirmation used for some
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

IAN SOUTHWELL
32
identifications. It must be noted however that this oil was not a typical commercial oil as
cineole content (16.5%) was above standard limits (Brophy et al. 1989b; Standards
Association of Australia 1985; International Standards Organisation 1996) and p-cymene
(11.4%), a-terpinene (2.7%) and 
γ-terpinene (11.5%) indicated substantial oxidation (Brophy
et al. 1989b). Components identified in tea tree oil above trace levels (>0.1%) included a-
gurjunene (0.23%), ß-terpineol (0.24%), allo-aromadendrene (0.45%), a-muurolene (0.12%)
8-p-cymenol (0.13%) and viridiflorene (1.03%). The short, packed Carbowax 20 M GC
column used for the GCMS analysis had obvious deficiencies. Even the 100m Carbowax
20 M capillary column failed to resolve a-thujene from a-pinene, ß-phellandrene from 1,8-
cineole and had difficulty in resolving myrcene from a-phellandrene and 1,4-cineole and
terpinen-4-ol from ß-elemene.
These authors described the sesquiterpene hydrocarbon viridiflorene for the first time
and proposed structure (21) (Figure 2). This structure was inconsistent with the dehydration
product of viridiflorol (20) which should have structure (17) because of the absolute
configuration assignments of Buchi et al. (1969) following the synthesis of (-)-
aromadendrene (22), the enantiomer of naturally occurring (+)-aromadendrene (16).
Viridiflorene must then have structure (17) which is also the structure of ledene, the
dehydration product of ledol (23), the C10 enantiomer of (+)-viridiflorol (20). This
assignment has been confirmed by nuclear magnetic resonance (NMR) by both Australian
and French workers (Southwell and Tucker 1991; Faure et al. 1991). (+)-Ledene (17) has
been well known as a dehydration product of viridiflorol (e.g. Birch et al. 1959) and also as
a natural constituent of essential oils (e.g. Taskinen 1974). Consequently the name
viridiflorene should be replaced by ledene in tea tree and other essential oil reports.
A more recent GCMS analysis of M. alternifolia oil corrected the viridiflorene structure,
listed a total of 97 constituents and identified several new components (Brophy et al.
Figure 2 Chemical structures of sesquiterpenoids conformationally related to ledene
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

TEA TREE CONSTITUENTS
33
1989b). The most significant findings of this investigation included (a) determining
suitable GC and column conditions for the optimal separation of tea tree oil constituents,
b) the identification of the cis and trans alcohol pairs of sabinene hydrate, p-menth-2-en-
1-ol and piperitol, (c) the composition of atypical and aged oils, (d) the affect of distillation
time on oil composition, (e) the micro-scale analysis of a single leaf (6 mg) for quality
determination and (f) the presence of oil precursors in the ethanolic extracts of tea tree
flush growth.
This last finding was subsequently investigated more thoroughly (Southwell and
Stiff 1989). Although individual mature leaves when immersed in ethanol gave an extract
which accurately reflected the quality of the oil if the leaves were distilled, the same
could not be said for the brighter green flush growth. This flush growth was found to be
rich in cis-sabinene hydrate (24) which was replaced by 
γ-terpinene (12) and terpinen-
4-ol (14) as the leaves matured. Very little of (24) and other minor precursors were
detected in the distilled oil due to the lability of cis-sabinene hydrate (Erman 1985;
Fischer et al. 1987, 1988).
Comprehensive GCMS analyses were performed on eight tea tree clones propagated at
the University of California (Kawakami et al. 1990). These clones, originally claimed to be
M. alternifolia stock, have since been shown to contain at least one M. linariifolia variety
(Southwell et al. 1992). Oils were obtained by both simultaneous purging and extraction
(SPE) to give headspace analysis and steam distillation and extraction (SDE) to give oil
analyses. Only two of the distilled oils were from terpinen-4-ol type clones. The headspace
(SPE) analysis concentrated the more volatile monoterpene hydrocarbons especially
sabinene, 
α-thujene,  α-terpinene and γ-terpinene at the expense of the less volatile
oxygenated terpenoids especially terpinen-4-ol. These compositions were similar to the
composition of the initial vapour cloud that emerges from the still condenser prior to
condensation (Southwell 1988). Although this investigation reported for the first time a
number of new sesquiterpenoids, none exceeded 0.6% of the total oil.
The two commercial terpinen-4-ol type tea tree oils sourced from M. alternifolia
and M. linariifolia are distinguishable on the grounds of oil chemistry (Southwell and
Stiff 1990). The former when analysed by GC gives a smaller peak for a-thujene than
for a-pinene and the latter gives the converse. As these peaks are always the first two
significant chromatographic peaks, a glance at the GC trace will tell which oil has been
analysed as long as 
α-pinene and α-thujene are resolved. The mean a-thujene: a-pinene
ratio was 0.33 (n=521) for M. alternifolia and 1.49 (n=180) for M. linariifolia. Non-
polar or intermediate polarity stationary phases are best for this analysis as resolution
is greater. The 
α-thujene: α-pinene ratio is upset as the percentage of cineole in the oil
increases. As 
α-thujene is associated with the terpinen-4-ol biogenetic pathway, higher
cineole means less terpinen-4-ol and a corresponding lower 
α-thujene: α-pinene ratio.
Hence the test is only suitable for the terpinen-4-ol chemical varieties. A similar way of
distinguishing the two species is to measure the cis-sabinene hydrate (24) to trans-
sabinene hydrate (25) ratio in the ethanolic extract of the flush growth of both species.
The mean cis-sabinene hydrate: trans-sabinene hydrate ratio was 7.1:1 (62 extracts
from 11 trees) for M. alternifolia and 0.7:1 (29 extracts from 6 trees) for M. linariifolia
(Southwell and Stiff 1990).
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

IAN SOUTHWELL
34
Terpinolene varieties of Melaleuca have, from time to time, received passing mention in
the literature (Laakso 1966; Brophy et al. 1989a; Kawakami et al. 1990). Their existence
was formally acknowledged and data documented by Southwell et al. (1992) and Butcher
et al. (1994). Both M. alternifolia and M. trichostachya were found to have varieties with
from 10–57% terpinolene, high proportions of cineole (13–56%) and insufficient terpinen-
4-ol (1–20%) for the tea tree oil market.
With the advent of chiral GC columns, the enantiomeric ratios of tea tree oil constituents
were determined (Russell et al. 1997; Leach et al. 1993; Cornwell et al. 1995). Seven
monoterpenes were resolved in this way and their percentages and enantiomeric ratios shown
in Table 2. These ratios provide valuable criteria for checking the authenticity of tea tree oil
especially blends with (-)-terpinen-4-ol from Eucalyptus dives that have been detected in
the past. Similar chiral resolutions were achieved using lanthanide shift reagents on the
individual tea tree oil constituents terpinen-4-ol and a-terpineol (Leach et al. 1993). The
complexity was however too great for these shifts to be meaningful for an entire oil. The
58% enantiomeric excess obtained for a standard (99%) sample of terpinen-4-ol ([
α]
D
+29°)
was consistent with both the 30% enantiomeric excess for terpinen-4-ol ([
α]
D
+16%)
fractionated from tea tree oil (Russell et al. 1997) and the maximum rotation values (+47°-
+48°) reported on enantiomerically pure samples (Naves and Tullen 1960; Ohloff and Uhde
1965; Verghese 1966).
Because of the efforts of these investigators over the years, the chemistry of tea tree oil
is now, well established. There is a need to clarify some of the minor and trace constituent
assignments and to establish the enantiomeric composition of the sesquiterpenoids. Then
efforts can concentrate on which of these constituents are beneficial or detrimental (to the
commercial uses of the oil) and how to maximise or minimise their contribution to
commercial oils.
a, insufficient resolution to allow accurate quantitation; tr, trace.
Table 2 The enantiomeric composition of the seven tea tree oil
constituents resolved by chiral gas chromatography on ß-cyclodextrin
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

TEA TREE CONSTITUENTS 35
Molecular Perspective
The components in the accompanying Table (
Table 3
) have been reported as volatile
constituents from the leaves of the terpinen-4-ol variety of Melaleuca alternifolia. The list
is based on the comprehensive list of Brophy et al. (1989b) supplemented with the additional
components by Swords and Hunter (1978), Kawakami et al. (1990) and Leach et al. (1993).
Correlation of entries from the different sources is confounded by the use of different GC
column lengths and stationary phases. Entry sequence is based on the non-polar BP1 FSOT
column of Brophy et al. (1989b) with entries from the other sources cited. The assignments
for minor and trace constituents may not be reliable where mass spectra alone were used for
these identifications. Indeed specialists in this field have suggested that “the injection of an
essential oil into a GCMS instrument with a fully automatic library search was not a vigorous
scientific exercise worthy of publication” (Stevens 1996) and indicate that such identifications
be supported by retention indices comparisons on two columns of different polarity.
Furthermore, one regulatory body now insists that identification be confirmed by at least
two methods (Liener 1966). Consequently some of the minor and trace component
identifications listed in 
Table 3 m
ust be considered tentative especially where supporting
evidence is not available.
EXTRACT CONSTITUENTS
Although steam distillation is the preferred method for the isolation of essential oils, some
commercial products are obtained by alternative processes. For example, citrus oils are
isolated from the peels of citrus by cold pressing and many perfumery products (e.g. jasmin,
boronia, acacia) by solvent extraction. This latter method is usually preferred for low-volume
high-value products. More recently carbon dioxide and supercritical fluids have been
replacing conventional solvents especially for important flavour constituents (Kerrola 1995).
little is known of the extractive constituents of M. cajuputi and M. quinquenervia. Some
workers have investigated the alcoholic extraction of M. alternifolia, M. linariifolia, M.
lanceolata and M. uncinata (Brophy et al. 1989b; Southwell and Stiff 1989; Southwell and
Stiff 1990; Jones et al. 1987).
Solvent extraction has never been seriously considered as an alternative procedure for
obtaining tea tree oil. The cost of the process, the scale of the operation and the problems
that would be associated with marketing a chemically different product have been deterrents
for the industry.
On a laboratory scale however, these alternative procedures have been found to be
very useful analytical techniques which have also contributed to our knowledge of leaf
chemistry.
Volatile Constituents
As solvent extraction removes both volatile and non-volatile constituents from leaf, the
volatile constituents are those which can be readily analysed, without derivatisation, by
gas chromatography (GC). Ethanolic extraction of a single tea tree leaf (1–10mg) in a
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

IAN SOUTHWELL
36
Table 3 Volatile constituents of Melaleuca alternifolia leaf
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

TEA TREE CONSTITUENTS
37
Table 3 (Continued)
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

IAN SOUTHWELL
38
A
Retention index from reference superscripted.
B
Yield (%) from reference in italics e.g. a.
C
a, Swords and Hunter (1978); b, Brophy et al. (1989); c, Kawakami et al. (1990); d, Leach et al.
(1993); tr, trace<0.1%; un, unresolved.
D
Identification (ID) by means of comparative mass spectrum (MS), retention time (RT), GC co-injection
(COGC), Nuclear Magnetic Resonance Spectrometry (NMR) and Infrared Spectroscopy (IR).
Table 3 (Continued)
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

TEA TREE CONSTITUENTS 39
0.1ml GC vial insert gave a solution suitable for GC analysis in the usual way (Southwell
and Stiff 1989; Brophy et al. 1989b). The 30 hour room temperature extraction was reduced
to one hour following 10 seconds of microwave irradiation (Russell et al. 1997). The
monoterpene region of the resultant trace (Figure 3) accurately reflected the quality of
the oil obtained by steam distillation of the same leaf material when mature leaf was
extracted.
This microextraction method revealed numerous interesting facts about the chemistry of
tea tree which led to many significant analytical uses.
The first of these was developed in response to the increasing production of and demand
for tea tree oil. Prior to the 1980s, supplies had been adequately sourced from natural
stands growing in specific areas in the New South Wales northern rivers region. Production
could only be increased by either seeking good quality stands growing further afield or
establishing plantations. Both approaches present quality control problems because of
the abundance of the undesirable high cineole chemical variety. The microextraction
analytical procedure provided an easy way to check both the quality of distant natural
stands and the genetic quality of parent trees providing propagation material for plantation
establishment (Brophy et al. 1989b). This extraction method then provides an excellent
way of determining whether the tree under investigation has acceptable levels of terpinen-
4-ol and 1,8-cineole without having to carry out the resource-consuming steam distillation
procedure.
When the brighter green flush growth was extracted, 
γ-terpinene (12) and terpinen-4-ol
(14) were found to be present in their precursor sabinene hydrate forms. These thujane
precursors (
Figure 4
) convert to their more stable end products either during steam distillation
by artifact formation or in vivo as the leaf matures (Southwell and Stiff 1989). Hence accurate
figures from microextraction quality checks are best obtained using the dark green mature
tea tree leaves.
This difference between flush growth and mature leaf also led to many significant
observations. For example, the along-the-branch variation of key constituents or-sabinene
hydrate (24), trans-sabinene hydrate (25), sabinene (3), terpinen-4-ol (14) and 
γ-terpinene
(12) was monitored from the tip to the main stem by the single leaf microextraction method
Figure 3 Gas chromatographic trace of M. alternifolia single leaf extract on a 60m AT35 FSOT column.
Peak numbers are consistent structures shown in 
Figures 1
 and 4
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

IAN SOUTHWELL
40
(Southwell and Stiff 1989). The high cis-sabinene hydrate levels in the flush growth gave
way abruptly to high levels of 
γ-terpinene and terpinen-4-ol as the leaf matured (Figure 5).
The point of inflection occurred at exactly the stage where flush growth ceased and mature
leaf began irrespective of the percentage flush on the branch. Searches for similar
relationships in the closely related Melaleuca bracteata (black tea tree) and Leptospermum
petersonii (lemon-scented tea tree) did not reveal significant along-the-branch differences
(Southwell 1989) other than a minor inverse relationship between citral and citronellal with
the latter species.
This extraction method also provided another way of distinguishing between M.
alternifolia and the closely related M. linariifolia (see above) (Southwell and Stiff 1990).
The ratio of the precursors cis-sabinene hydrate: trans-sabinene hydrate was approximately
Figure 4 Chemical structures of the precursor thujanes from Melaleuca, terpinen-4-ol type flush growth
extracts
Figure 5 Along-the-branch variation in the concentrations of cis-sabinene hydrate, terpinen-4-ol and
γ-terpinene in individual leaves of M. alternifolia
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,
part of The Gordon and Breach Publishing Group.

TEA TREE CONSTITUENTS
41
7.1:1 for the flush growth ethanolic leaf extract of M. alternifolia and approximately 0.7:1
for M. linariifolia. Measuring this ratio in the flush extract provides an alternative method
for distinguishing the two species to measuring the 
α-thujene/α-pinene ratio in the oil
(Table 4).
Microextraction also proved ideal for determining the quality of seedlings before planting
out (Russell et al. 1997). With the existence of a number of chemical varieties of M.
alternifolia and M. linariifolia and the likelihood of cross pollination (Butcher et al. 1992),
plantation establishment using the right genetic material is essential. Freshly emerging leaves
from propagation material including seedling dicotyledon leaves are ideally suited for single
leaf extraction analysis. A preliminary analysis of supposedly good quality terpinen-4-ol
type tea tree seedlings showed higher than expected concentrations of a-pinene (approx.
12%) and terpinolene (approximately 12%) in early stages of development. Terpinen-4-ol
and the precursor sabinene hydrates were not present in the dicotyledon leaves. Single leaf
analysis, on a day by day basis as each successive leaf pair formed showed that, by the time
leaf set 10 was 10 weeks old, terpinolene concentrations had dropped to normal (3%) and
Figure 6 Concentration of tea tree oil constituents in different leaf sets 6 weeks after emergence
Table 4 The minimum, maximum, mean and standard deviation (s.d.) cis: trans sabinene
hydrate (A) and a-thujene : a-pinene (B) ratios for M. alternifolia and M. linariifolia
flush growth leaf extracts from the terpinen-4-ol variety
Copyright © 1999 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint,

Yüklə 6,32 Mb.

Dostları ilə paylaş:
1   2   3   4   5   6   7   8   9   ...   29




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©azkurs.org 2020
rəhbərliyinə müraciət

    Ana səhifə