- Email: [email protected]
Effects of mechanical wounding on concentration and composition of essential oil from Melaleuca alternifolia leaves
Biochemical Systematics and Ecology 30 (2002) 399–412 www.elsevier.com/locate/biochemsyseco
Effects of mechanical wounding on concentration and composition of essential oil from Melaleuca alternifolia leaves D. Zabaras a,*, R.N. Spooner-Hart b, S.G. Wyllie a a
b
Centre for Biostructural and Biomolecular Research, University of Western Sydney, Hawkesbury Campus, Richmond, New South Wales, 2753, Australia Centre for Horticulture and Plant Sciences, University of Western Sydney, Hawkesbury Campus, Richmond, New South Wales, 2753, Australia Received 28 April 2000; accepted 3 June 2001
Abstract The effects of mechanical damage on the essential oil obtained from Melaleuca alternifolia leaves were determined by post-wounding field experiments at 24 and 48 h. The rupture of oil glands caused by mechanical damage resulted in a decrease in oil concentration in mature and immature leaves in the first 24 h, but mature leaves were found to recover most of their loss 48 h after damage. Post-wounding changes in the oil composition from mature and immature leaves were also detected. The response of the mature leaves was expressed by different oil constituents for every post-wounding day elapsed. Results also indicated that wounding of immature leaves affects their chemical maturation. The post-wounding response in both leaf types was found to be independent from the pre-wounding levels of the particular compounds expressing the response and the overall leaf oil-composition. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Melaleuca alternifolia; Wounding effects; Oil composition; Oil yield; Monoterpenes; Terpenoids; Principal component analysis
* Corresponding author. Tel.: +61-2-4570-1640; fax: +61-2-4570-1621. E-mail address: [email protected] (D. Zabaras). 0305-1978/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 3 0 5 - 1 9 7 8 ( 0 1 ) 0 0 1 0 1 - 6
400
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
1. Introduction Studies have shown induced defense to be a vital protective mechanism for many economically-interesting agricultural plants such as corn (Turlings and Tumlinson, 1992; Takabayashi et al., 1995), cotton (Loughrin et al., 1994; Pare´ and Tumlinson 1997, 1998), cabbage (Coleman et al., 1997) and potato (Bolter et al., 1997). However, limited information exists on the response of terpene accumulating plants to mechanical wounding or attack by arthropods or microbial pathogens. Work on Grand fir (Abies grandis) oleoresin has shown that stem-wounding induces an increased activity of monoterpene cyclases (Lewinsohn et al., 1991a,b); changes in the profile of the oil obtained from the leaves of bush basil (Ocimum minimum) have also been demonstrated (Zabaras and Wyllie, 2001). This communication reports the effects of mechanical damage on the essential oil obtained from the leaves of Melaleuca alternifolia Cheel (Australian tea tree) (family Myrtaceae) over a two-day period. This perennial shrub produces oil that is known to have germicidal properties mainly due to the its high (+)-terpinen-4-ol (苲40%) content (Southwell, 1988). Chemical varieties affording oils high (⬎15%) in 1,8 cineole also exist however these oils are medicinally undesirable because of the concomitant decrease in the active constituent, terpinen-4-ol (Southwell et al., 1996). Other constituents of the oil include γterpinene (苲20%), α-terpinene (苲10%), α- and β-pinene and sesquiterpenes (Southwell, 1988).
2. Methods and materials 2.1. Plants Healthy, pest-free, 2-yr regrowth trees were used (high terpinen-4-ol, low 1,8 cineole chemotype).The trees are part of a small plantation located at the Precinct of Horticulture, Faculty of Science and Technology, University of Western Sydney Hawkesbury, New South Wales, Australia. 2.2. Experimental design Three leaves (of approximately the same age and size) were cut in half longitudinally at 0 h (T0), combined, weighed and immediately subjected to solvent extraction. After 24 h the other half of the leaves (that had remained on the branch) was taken and treated similarly. The above process was carried out on eight different trees divided into two groups. The two groups were sampled one month apart in order to test the reproducibility of the results. The 48 h treatment was carried out in the same way but this time the second half of each leaf was sampled 48 h after hour zero. The whole procedure was carried out on mature and flush growth (immature) M. alternifolia leaves. Different trees were used for each treatment and each leaf type. The 2nd and 3rd pair of leaves from the apex (yellowish-light green in colour) were used as flush growth.
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
401
2.3. Field conditions The experiments were carried out during spring (temperature 21–26°C, humidity 65–80%). Sampling was performed at midday only during full light conditions. The trees were irrigated daily and each tree was wounded only once. 2.4. Chemical analysis After sampling the leaf material (10–15 mg) was immersed in CH2Cl2 (100 µl) spiked with 400 ppm octane (as an internal standard) for 24 h at room temperature in the dark. This extraction period was deemed to be sufficient as the detection of differences in oil concentration and composition before and after wounding rather than exhaustive extraction was the purpose of this study. The recovery of the internal standard after this period was found to be 97.3±2.4% for the flush growth and 95.1±2.1% for the mature leaves (mean±SEM, n = 8). Extracts were then subjected to GC and GC–MS analysis. GC analysis was carried out using a HP 6890 gas chromatograph equipped with a flame ionisation detector and a HP 7673 GC/SFE auto-injector. The column used was BPX-5 (50 m length×0.22 µm ID×0.25 µm film thickness) (SGE Scientific, Melbourne, Australia). The injection volume was 2 µl, inlet temperature and pressure were 280°C and 20 psi respectively, carrier gas was H2 (40 ml/min), split ratio was 1:10, detector temperature at 280°C, and the oven program was: initial temperature 60°C for 5 min, increased to 180°C at 4°C/min, final temperature maintained for 5 min. GC–MS analysis conditions were as above with He as the carrier gas. The instrument used was a Varian 3800 gas chromatograph connected to a Varian 2000 ion trap detector (0.9 scans/s, 20 µA). Compounds were identified by comparison of their mass spectra and retention indices (based on n-alkanes) with those of authentic standards (Fluka Chemicals, NSW, Australia). The relevant literature was also consulted (Brophy et al., 1989; Adams, 1995). 2.5. Oil concentration determination The concentration of each oil component (Ci) was determined by the equation: Ci = [(Cistd/Aistd)×Ai]×RRF, where Cistd is the concentration of the internal standard, Aistd is the peak area of the internal standard and Ai is the peak area of each component after normalisation (Rouessac and Rouessac, 2000). The relative response factor (RRF) of the internal standard in relation to all oil components was assumed to be 1. Any differences in the detector response between the internal standard and the oil components and also between the various oil components could introduce small bias in the determination but these are known to be consistent that do not affect trends over time (Murtagh and Smith, 1996). 2.6. Headspace experiment Six-month old, potted M. alternifolia plants were used. Part of a small branch (approximately 8–10 pairs of leaves) was placed in a small glass cylinder (20 ml in
402
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
volume) closed at one end with a Teflon cap. The cap was pre-drilled to fit exactly a manual solid-phase microextraction (SPME) holder (Supelco, USA). Glass wool was used to restrict the other end after the branch was in place, however, the system was not air-tight. After 30 min the headspace above the branch was sampled using a polydimethylsiloxane-coated fibre (10 min, 25°C). Thermal desortpion of the fibre in the injector of the GC and GC–MS system followed (220°C, 10 min, other conditions as above). Three of the enclosed leaves were then cut longitudinally in half. The headspace was then monitored for 1 h at 20-min intervals similarly as above. The whole procedure was repeated with unwounded leaves and the results were used to correct for depletion of the headspace effects and other headspace losses in the system. 2.7. Moisture determination Wounded leaves (experimental design as described above) were placed in a glass petri-dish and dried for 10 days at 50°C. Moisture was determined by the weighing of the leaves before and after drying. The experiment was performed three times for both mature and flush growth leaves. 2.8. Statistical analysis The contribution of each monoterpene (and of the sesquiterpenes measured as a group) in the oil was expressed as a percent value based on the peak areas of the integrated chromatograms. This was necessary in order to compare the oil-profiles before and after wounding. If absolute values (e.g., area counts/g) were to be used for the comparison, then changes in moisture, leaf development and/or oil concentration would have been superimposed on the data and the results obtained would be inaccurate and misleading. The difference between the percentage contribution of each compound in the oil before and after wounding was tested for statistical significance using the Wilkoxon signed ranks test (paired samples) (Sall and Lehman, 1996). A correlation matrix was then built (based on the actual percentages) using as variables the compounds found to be statistically significant. PCA was then performed on the correlations and the eigenvalues obtained were used to visualise the relationship between objects (leaves) in two-dimensional score plots. The association between the significant variables (compounds) was examined based on their loading values on the first two principal components. Data from independent samples (e.g., comparison between treatments) was analysed as above; however in this case the Kruskal–Wallis test was applied (Sall and Lehman, 1996). The effect of wounding on the non-parametric measures of association between the major thujanes and p-menthanes based on their percentage contribution to the flush growth extracts was also investigated (Spearman’s test) (Sall and Lehman, 1996). The software used for all data analyses was JMP-IN (version 3 for Windows) (SAS Institute Inc., USA).
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
403
3. Results and discussion 3.1. Oil concentration Fig. 1 shows the change in oil concentration in mature and flush growth M. alternifolia leaves caused by wounding. Although the extraction was carried out on fresh leaves the values are expressed on a dry weight basis due to the substantial variations that can occur with moisture content (Murtagh and Smith, 1996). The observed change was similar for both leaf types during the first 24 h but differed markedly 48 h after injury. After this time the oil concentration of the mature leaves appears to have returned to its pre-wounding level. In contrast, the flush growth leaves show a continuing loss as post-wounding time elapses (Fig. 1) although it is possible that some of this loss reflects the increasing weights of leaf structural features due to the ongoing ontogeny of the flush growth. Results from headspace experiments indicate that the initial post-wounding oil loss is due to the volatilisation of the oil constituents into the atmosphere. As can be seen from Fig. 2 the concentration of the headspace above M. alternifolia leaves increases dramatically immediately upon wounding and then gradually decreases. This is likely to be caused by the rupture of the leaf’s secretory cells although ‘intentional’ release of oil through modified epidermal cells is also possible (List et al., 1995). This increased emission of volatiles may be a preliminary response by the leaf to deter further predation.
Fig. 1. Changes in oil concentration of M. alternifolia leaves as a result of mechanical damage (mean±SE, n = 8). Identical superscripts indicate statistical significance between values (P⬍0.01, Wilkoxon signed ranks test).
404
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
Fig. 2. Changes in the headspace concentration above M. alternifolia leaves as a result of wounding (mean±SE, n = 3). Samples were adjusted to the same scale based on their unwounded (control) values. The immediate increase and then subsequent decrease in headspace concentration over time is evident.
The recovery of the oil concentration observed after 48 h for the mature leaves suggests that biosynthetic processes are active during the post-wounding period and replenish any oil losses caused by wounding. 3.2. Oil composition Studies on corn (Turlings et al., 1995) and cotton (Pare´ and Tumlinson, 1997) have shown that there is a delay of several hours between the start of wounding and the release of induced terpenoids. This suggests that a series of inducible biochemical reactions are required for those compounds to be produced and subsequently released. Similarly, if changes in oil from M. alternifolia were to be induced by mechanical wounding, the process would not be instantaneous. Therefore, the assumption that T0 values represent pre-wounding values appears to be valid in relation to oil composition. The typical T0 oil composition as obtained from the trees used in this experiment (for both leaf types) is given in Table 1. 3.2.1. Flush growth The statistical tests showed a significant change in the percentage composition of several constitutive terpenoids between unwounded and wounded M. alternifolia leaves (Table 1). The 24 h post-wounding results showed that the major p-menthanes (α- and γ-terpinene, terpinolene, terpinen-4-ol) increased in concentration at the expense of the thujanes (cis-sabinene hydrate, sabinene). The negative correlation between the two skeletal groups is better observed 48 h after wounding (Table 1). The conversion of the thujane precursors to the more stable p-menthanes occurs naturally in M. alternifolia flush growth as the leaves mature (Southwell and Stiff, 1989; Cornwell et al., 1995). During the aging process, cis-sabinene hydrate and to
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
405
Table 1 Time-zero (T0) oil composition from flush growth and mature Melaleuca alternifolia leaves used in this study. Range values are given to demonstrate the inter-plant variation observed. Values from both 24- and 48-h experiments are included. Selected components as detected after cold solvent extraction are presented Components
Percentage composition (%) (min.–max.) Flush growth (n = 16) Mature leaves (n = 15)
α-thujene α-pinene sabinene β-pinene myrcene α-terpinene cluster 2a γ-terpinene trans-sabinene hydrate terpinolene cis-sabinene hydrate trans-p-menth-2-en-1-ol cis-p-menth-2-en-1-ol terpinen-4-ol α-terpineol trans-piperitolb Unknown sesquiterpenes
0.50–0.70 1.41–1.82 3.33–7.42 0.61–0.73 0.61–0.94 1.10–2.50 2.01–8.72 1.91–4.62 5.24–6.72 0.42–0.83 30.41–46.32 0.21–0.36 0.23–0.39 2.51–5.74 0.61–1.73 0.32–0.54 0.20–0.37 21.62–34.43
a b
0.36–1.10 2.00–3.18 nd 0.64–1.13 0.79–1.95 3.95–7.80 5.48–9.00 8.60–20.6 nd 1.10–2.90 nd 0.10–0.75 0.10–0.78 27.72–43.91 2.01–6.46 0.50–1.90 0.40–1.30 9.17–20.71
cluster 2= p-cymene, limonene, β-phellandrene, 1,8-cineole Tentative identification
a lesser extent sabinene and trans-sabinene hydrate are known to be implicated in the formation of terpinen-4-ol, terpinolene, α- and γ-terpinene (Cornwell et al., 1995; Southwell, 1999). However, the results obtained in this study indicated that wounding has some effect over and above the normal ontogenetical activities occurring during flush growth maturation. From Table 2 it can be seen that in unwounded flush growth significant linear correlations were obtained only between concentrations of compounds of the same skeletal group (with the exception of the relationship of cis-sabinene hydrate with αterpinene). However, in wounded leaves several significant linear-correlations were observed between p-menthanes and their precursor thujanes. Sabinene and cis-sabinene hydrate appear to be strongly related to terpinen-4-ol and γ-terpinene whilst sabinene itself is also correlated with terpinolene. Also in wounded leaves α-terpinene was not linearly related with any of the major thujanes. It can only be speculated as to how leaf wounding possibly affects the ontogenetical changes in M. alternifolia flush growth. Recent work has established that the thujanes to p-menthanes transformations are pH and water dependent (Southwell and Stiff, 1989; Cornwell, 1999). Therefore it is possible that cell rupture alters pH and
406
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
Table 2 Effect of mechanical damage on oil composition in flush growth of M. alternifolia. Selected components as detected after cold solvent extraction are presented. Asterisks indicate statistical significance (*at P⬍0.02; **at P⬍0.01, Wilkoxon signed-ranks test) Components
α-thujene α-pinene sabinene cluster 1a α-terpinene cluster 2b γ-terpinene trans-sabinene hydrate terpinolene cis-sabinene hydrate terpinen-4-ol α-terpineol a b
Difference in percentage composition between unwounded and wounded leaves (mean±SEM) (n = 8) 24 h after injury ⫺0.01±0.01 ⫺0.03±0.03 ⫺0.44±0.23 ⫺0.06±0.05 0.34±0.06* 0.14±0.08 0.63±0.09* 0.04±0.11 0.29±0.04* ⫺1.54±1.05 1.19±0.23* 0.11±0.09
48 h after injury 0.04±0.03 0.06±0.05 ⫺0.78±0.15** 0.05±0.07 0.71±0.09* 0.18±0.11 0.98±0.25 ⫺0.13±0.10 0.40±0.07** ⫺4.57±1.30* 1.91±0.40** 0.08±0.06
Cluster 1 includes β-pinene and myrcene Cluster 2= p-cymene, limonene, β-phellandrene and cineole
water levels within the leaf thus affecting the chemical maturation of the flush growth. From an ecological perspective, rapid chemical maturation of the flush growth may be interpreted as an attempt to minimise subsequent predation. 3.2.2. Mature leaves Different sets of oil constituents for each treatment are utilised to express the effect of wounding on M. alternifolia mature leaves (Table 3). The monoterpenes α-thujene and cis-menth-2-en-1-ol are the only components that change significantly in both post-wounding measurements (Table 4). The other terpenes that initially change their concentration appear to revert back to their pre-wounding levels; a different set of monoterpenes is then used for the expression of the response. This indicates that mature leaves are biosynthetically active during the post-wounding period so some oil constituents are synthesised at the expense of others. This view is also supported by the fact that certain terpenes (e.g., trans- and cis-menth-2-en1-ol) greatly increased their presence in the oil (⬎60% increase) within 24 h (Tables 1 and 4). The increase in monoterpene levels at the expense of the sesquiterpenes in the first 24 post-wounding hours is also very interesting. Direct biosynthetic interdependence between the two terpene groups is not likely given that they are synthesised in different cellular compartments (Bohlmann et al., 1998). However it is possible that wounding affects the amount of substrate available or the terpenoid synthases responsible for the production of these oil constituents as compared to normal biosynthesis.
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
407
Table 3 Non-parametric correlation coefficients between the major thujanes and p-menthanes in M. alternifolia flush growth for unwounded (n = 14) and wounded leaves (both 24 and 48 h treatments, n = 16). Asterisks indicates statistical significance (*at P⬍0.05; ** at P⬍0.01, Spearman’s test) Variable by variable α-terpinene-sabinene γ-terpinene-sabinene γ-terpinene-α-terpinene trans-sabinene hydrate-sabinene trans-sabinene hydrate-αterpinene trans-sabinene hydrate-γterpinene terpinolene-sabinene terpinolene-α-terpinene terpinolene-γ-terpinene terpinolene-trans-sabinene hydrate cis-sabinene hydrate-sabinene cis-sabinene hydrate-α-terpinene cis-sabinene hydrate-γ-terpinene cis-sabinene hydrate-transsabinene hydrate cis-sabinene hydrate-terpinolene terpinene-4-ol-sabinene terpinen-4-ol-α-terpinene terpinen-4-ol-γ-terpinene terpinen-4-ol-trans-sabinene hydrate terpinen-4-ol-terpinolene terpinen-4-ol-cis-sabinene hydrate
Spearman’s ρ Unwounded leaves
Wounded leaves
⫺0.2478 ⫺0.2156 0.6619** 0.3633 0.0871
⫺0.2789 ⫺0.7574** 0.6966** 0.0324 0.0496
0.3917
0.1468
⫺0.1597 0.9319** 0.6935** 0.0000
⫺0.5403* 0.5583* 0.8657** 0.4122
0.5842* ⫺0.5957* ⫺0.3868 0.5233*
0.7828** ⫺0.5090 ⫺0.7710** 0.3291
⫺0.4286 ⫺0.1079 0.8070** 0.9281** 0.2923
⫺0.4293 ⫺0.7142** 0.6020* 0.9372** 0.3255
0.8574** ⫺0.4570
0.9407** ⫺0.6141*
It is possible that additional oil components were affected by leaf wounding. However, only the components that showed a consistent behaviour across the different trees used were labeled as significant by the statistical analysis. This together with the fact that half the trees were sampled one month apart from the other half ensured the reproducibility of the detected response. An experiment was conducted to test the hypothesis that the changes outlined above originated from the different oil composition that the two leaf halves might possessed due to developmental or other factors. Statistical analysis however found no significant differences (Wilkoxon signed-ranks test at Pⱕ0.05, n ⫽ 8) in the oil composition from the two halves. Also, repetitive injections of two different extracts showed the detected wounding response to be over and above the small errors introduced by the instrument used (Table 5). The reproducibility of the extraction process itself could not be determined due to the existing inter-leaf variation. However, the results obtained when the same extraction process was applied on flush growth indi-
408
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
Table 4 Effect of mechanical damage on oil composition in mature growth of M. alternifolia. Selected components as detected after cold solvent extraction are presented. Asterisks indicate statistical significance (* at P⬍0.05; ** at P⬍0.01, Wilkoxon signed-ranks test) Components
Difference in percentage composition between unwounded and wounded leaves (mean±S.E.M) (n=8) 24 h after injury 48 h after injury
-thujene α-pinene β-pinene myrcene α-terpinene cluster 2a γ-terpinene terpinolene trans-p-menth-2-en-1-ol cis-p-menth-2-en-1-ol terpinen-4-ol α-terpineol trans-piperitolb Unknown sesquiterpenes
0.11±0.01** 0.02±0.02 0.03±0.00 0.02±0.05 ⫺0.25±0.57 0.36±0.54 0.14±0.12 0.02±0.10 0.10±0.01** 0.11±0.01** ⫺0.28±0.35 0.05±0.03 0.40±0.24 0.21±0.14 ⫺1.06±0.35*
a b
0.12±0.01* 0.08±0.02 0.04±0.02 0.02±0.01 0.77±0.40 ⫺0.34±0.13* 0.27±0.15 0.18±0.07* 0.05±0.02 0.12±0.02* ⫺0.18±0.15 ⫺0.01±0.01 ⫺0.16±0.18 ⫺0.09±0.12 ⫺0.81±0.30
Cluster 2=p-cymene, limonene, β-phellandrene, 1,8-cineole Tentative identification
Table 5 Instrument reproducibility for the compounds found to be implicated in the wounding response of M. alternifolia leaves. Results from two extracts are shown (mean±SEM, n = 5). SEM is the standard error of the mean Compound
α-thujene cluster 2a terpinolene trans-p-menth-2-en-1-ol cis-p-menth-2-en-1-ol sesquiterpenes a
% (mean±SEM) Extract 1
Extract 2
0.732±0.009 6.45±0.018 1.94±0.001 0.740±0.014 0.281±0.006 12.6±0.090
0.882±0.010 5.41±0.007 2.11±0.002 0.745±0.011 0.300±0.002 11.7±0.069
Cluster 2= p-cymene, limonene, β-phellandrene, 1,8-cineole
cate its reliability as it was able to detect the well known ‘chemical maturation’ of M. alternifolia immature leaves. PCA two-dimensional score plots were constructed in order to elucidate further the detected wounding response (Fig. 3). Plots defined by the first two principal components in each case were sufficient for that purpose as it could explain most of the variation in the data (⬎88.2%). From Fig. 3 it can be seen that the wounding
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
409
Fig. 3. PCA scatterplots of oil from M. alternifolia mature leaves on the first two principal components (unwounded leaves squared, wounded leaves not squared). Identical numbers indicate same leaf (before– after wounding). (a) 24 h after injury, (b) 48 h after injury. Individuals 5–7 were sampled one month after individuals 1–4 to test the reproducibility of the response.
response in each mature leaf (for both treatments) is independent of the pre-wounding levels of the particular compound expressing the response. Also, the magnitude of the response appears to be similar between replicates within the same treatment, again, despite the large inter-tree variation. Similar behaviour was observed in the case of flush growth (data not shown). It is likely that the different sets of compounds involved in the post-wounding response serve distinct purposes. Differential timing in the production of defense oleoresin components has also been detected in the defense response of Grand fir
410
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
(Abies grandis) (Steele et al., 1998). In that case the monoterpenes formed immediately after wounding act as insect toxins while their later production at solvent levels is involved in wound sealing (Steele et al., 1998). No significant differences were found when the results from the two post-wounding measurements were compared to each other. This was probably due to the high level of inter-tree variation observed when independent samples (leaves from different trees) were considered during the statistical analysis. It is likely that the small compositional changes (if any) were ‘masked’ by the high inter-tree variability which can be seen in Table 1. The oil-profile differences demonstrated above between unwounded and wounded leaves are small in magnitude for both mature and immature leaves. Studies on cabbage (Coleman et al., 1997) and corn (Turlings et al., 1993) seedlings have shown that the wound-response in these plants is much greater when an elicitor (for example enzymes found in the predator’s saliva) is present rather than the response obtained by simple mechanical damage. Work in progress is currently testing the validity of this hypothesis in relation to terpene accumulating plants.
4. Conclusion In this study we were able to demonstrate compositional and quantitative changes in the oil from M. alternifolia mature and immature leaves as a result of mechanical damage. Wounding caused a decrease in the total oil concentration of all leaves. However, mature leaves were able to return to their pre-wounding oil concentration 48 h after damage. Leaf wounding stimulated increased production of certain terpenes at the expense of others. Both flush growth and mature leaves exhibited this response irrespective of total oil concentration, oil composition or pre-wounding concentration of the particular terpenes involved in the response. From a commercial point of view the effects of wounding on leaf total oil concentration are perhaps more significant than the observed compositional changes. However, it is likely that changes in both, oil concentration and composition serve important ecological roles during the leaf’s post-wounding period.
Acknowledgements The authors are grateful to the Faculty of Science and Technology, University of Western Sydney, Hawkesbury for the financial support. D.Z acknowledges receipt of an Australian Postgraduate Award (APA).
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
411
References Adams, R.P., 1995. Identification of Essential Oil Components by Gas Chromatography/Mass Spectrometry. Allured, Illinois. Bohlmann, J., Meyer-Gauen, G., Croteau, R., 1998. Plant terpenoid synthases: molecular biology and phylogenetic analysis. Proc. Natl. Acad. Sci. USA 95, 4126–4133. Bolter, C.J., Dicke, M., van Loon, J.J.A., Visser, J.H., Posthumus, M.A., 1997. Attraction of colorado potato beetle to herbivore-damaged plants during herbivory and after its termination. J. Chem. Ecol. 23, 1003–1023. Brophy, J.J., Davies, N.W., Southwell, I.A., Stiff, I.A., Williams, L.R., 1989. Gas chromatographic quality control for oil of Melaleuca terpinen-4-ol type (Australian tea trea). J. Agric. Food Chem. 37, 1330–1335. Coleman, R.A., Barker, A.M., Fenner, M., 1997. A test of possible indirect mediation of wound-induced resistance in cabbage against Pieris brassicae. Oikos 80, 43–50. Cornwell, C.P., Leach, D.N., Wyllie, S.G., 1995. Incorporation of oxygen-18 into terpinen-4-ol from the H218O steam distillates of Melaleuca alternifolia (Tea tree). J. Essent. Oil Res. 7, 613–620. Cornwell, C.P., 1999. Aspects of the chemistry and origins of the terpenes from Australian Myrtaceae. Ph.D. thesis. University of Western Sydney, Hawkesbury. Lewinsohn, E., Gijzen, M., Croteau, R., 1991b. Defense mechanisms of conifers: differences in constitutive and wound-induced monoterpene biosynthesis among species. Plant Physiol. 96, 44–49. Lewinsohn, E., Gijzen, M., Savage, T.J., Croteau, R., 1991a. Defense mechanisms of conifers: relationship of monoterpene cyclase activity to anatomical specialization and oleoresin monoterpene content. Plant Physiol. 96, 38–43. List, S., Brown, P.H., Walsh, K.B., 1995. Functional anatomy of the oil glands of Melaleuca alternifolia (Myrtaceae). Aust. J. Bot. 43, 629–641. Loughrin, J.H., Manukian, A., Heath, R.R., Turlings, T.C.J., Tumlinson, J.H., 1994. Diurnal cycle of emission of induced volatile terpenoids by herbivore-injured cotton plants. Proc. Natl. Acad. Sci. USA 91, 11836–11840. Murtagh, G.J., Smith, G.R., 1996. Month of harvest and yield components of tea tree. II. Oil concentration, composition and yield. Aust. J. Agric. Res. 47, 817–827. Pare´ , P.W., Tumlinson, J.H., 1997. De novo biosynthesis of volatiles induced by insect herbivory in cotton plants. Plant Physiol. 114, 1161–1167. Pare´ , P.W., Tumlinson, J.H., 1998. Cotton volatiles synthesized and released distal to the site of insect damage. Phytochem. 47, 521–526. Rouessac, F., Rouessac, A., 2000. Chemical Analysis: Modern Instrumental Methods and Techniques. Wiley, Chichester. Sall, J., Lehman, A., 1996. JMP Start Statistics. Duxbury Press, Belmont. Southwell, I.A., 1988. Australian tea tree: Oil of Melaleuca, terpinen-4-ol type. Chem. Aust. 55, 400–402. Southwell, I.A., 1999. Tea tree constituents. In: Southwell, I.A., Lowe, R. (Eds.), Tea Tree. The Genus Melaleuca. Hardwood Academic Publishers, Australia, pp. 29–62. Southwell, I.A., Markham, J., Mann, C., 1996. Is cineole detrimental to tea tree oil? Perfum. Flav. 21, 7–10. Southwell, I.A., Stiff, I.A., 1989. Ontogenetical changes in monoterpenoids of Melaleuca alternifolia leaf. Phytochemistry 28, 1047–1051. Steele, C.L., Katoh, S., Bohlmann, J., Croteau, R., 1998. Regulation of oleoresinosis in grand fir (Abies grandis). Differential transcriptional control of monoterpene, sesquiterpene and diterpene synthase genes in response to wounding. Plant Physiol. 116, 1497–1504. Takabayashi, J., Takahashi, S., Dicke, M., Posthumus, M.A., 1995. Developmental stage of herbivore Pseudaletia separata affects production of herbivore-induced synomone by corn plants. J. Chem. Ecol. 21, 273–287. Turlings, T.C.J., Loughrin, J.H., McCall, P.J., Ro¨ se, U.S.R., Lewis, W.J., Tumlinson, J.H., 1995. How caterpillar-damaged plants protect themselves by attracting parasitic wasps. Proc. Natl. Acad. Sci. USA 92, 4169–4174. Turlings, T.C.J., McCall, P.J., Alborn, H.T., Tumlinson, J.H., 1993. An elicitor in caterpillar oral
412
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
secretions that induces corn seedlings to emit chemical signals attractive to parasitic wasps. J. Chem. Ecol. 19, 411–425. Turlings, T.C.J., Tumlinson, J.H., 1992. Systemic release of chemical signals by herbivore-injured corn. Proc. Natl. Acad. Sci. USA 89, 8399–8402. Zabaras, D., Wyllie, S.G., 2001. The effect of mechanical wounding on the composition of essential oil from Ocimum minimum L. leaves. Molecules 6, 79–86.
Effects of mechanical wounding on concentration and composition of essential oil from Melaleuca alternifolia leaves D. Zabaras a,*, R.N. Spooner-Hart b, S.G. Wyllie a a
b
Centre for Biostructural and Biomolecular Research, University of Western Sydney, Hawkesbury Campus, Richmond, New South Wales, 2753, Australia Centre for Horticulture and Plant Sciences, University of Western Sydney, Hawkesbury Campus, Richmond, New South Wales, 2753, Australia Received 28 April 2000; accepted 3 June 2001
Abstract The effects of mechanical damage on the essential oil obtained from Melaleuca alternifolia leaves were determined by post-wounding field experiments at 24 and 48 h. The rupture of oil glands caused by mechanical damage resulted in a decrease in oil concentration in mature and immature leaves in the first 24 h, but mature leaves were found to recover most of their loss 48 h after damage. Post-wounding changes in the oil composition from mature and immature leaves were also detected. The response of the mature leaves was expressed by different oil constituents for every post-wounding day elapsed. Results also indicated that wounding of immature leaves affects their chemical maturation. The post-wounding response in both leaf types was found to be independent from the pre-wounding levels of the particular compounds expressing the response and the overall leaf oil-composition. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Melaleuca alternifolia; Wounding effects; Oil composition; Oil yield; Monoterpenes; Terpenoids; Principal component analysis
* Corresponding author. Tel.: +61-2-4570-1640; fax: +61-2-4570-1621. E-mail address: [email protected] (D. Zabaras). 0305-1978/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 3 0 5 - 1 9 7 8 ( 0 1 ) 0 0 1 0 1 - 6
400
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
1. Introduction Studies have shown induced defense to be a vital protective mechanism for many economically-interesting agricultural plants such as corn (Turlings and Tumlinson, 1992; Takabayashi et al., 1995), cotton (Loughrin et al., 1994; Pare´ and Tumlinson 1997, 1998), cabbage (Coleman et al., 1997) and potato (Bolter et al., 1997). However, limited information exists on the response of terpene accumulating plants to mechanical wounding or attack by arthropods or microbial pathogens. Work on Grand fir (Abies grandis) oleoresin has shown that stem-wounding induces an increased activity of monoterpene cyclases (Lewinsohn et al., 1991a,b); changes in the profile of the oil obtained from the leaves of bush basil (Ocimum minimum) have also been demonstrated (Zabaras and Wyllie, 2001). This communication reports the effects of mechanical damage on the essential oil obtained from the leaves of Melaleuca alternifolia Cheel (Australian tea tree) (family Myrtaceae) over a two-day period. This perennial shrub produces oil that is known to have germicidal properties mainly due to the its high (+)-terpinen-4-ol (苲40%) content (Southwell, 1988). Chemical varieties affording oils high (⬎15%) in 1,8 cineole also exist however these oils are medicinally undesirable because of the concomitant decrease in the active constituent, terpinen-4-ol (Southwell et al., 1996). Other constituents of the oil include γterpinene (苲20%), α-terpinene (苲10%), α- and β-pinene and sesquiterpenes (Southwell, 1988).
2. Methods and materials 2.1. Plants Healthy, pest-free, 2-yr regrowth trees were used (high terpinen-4-ol, low 1,8 cineole chemotype).The trees are part of a small plantation located at the Precinct of Horticulture, Faculty of Science and Technology, University of Western Sydney Hawkesbury, New South Wales, Australia. 2.2. Experimental design Three leaves (of approximately the same age and size) were cut in half longitudinally at 0 h (T0), combined, weighed and immediately subjected to solvent extraction. After 24 h the other half of the leaves (that had remained on the branch) was taken and treated similarly. The above process was carried out on eight different trees divided into two groups. The two groups were sampled one month apart in order to test the reproducibility of the results. The 48 h treatment was carried out in the same way but this time the second half of each leaf was sampled 48 h after hour zero. The whole procedure was carried out on mature and flush growth (immature) M. alternifolia leaves. Different trees were used for each treatment and each leaf type. The 2nd and 3rd pair of leaves from the apex (yellowish-light green in colour) were used as flush growth.
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
401
2.3. Field conditions The experiments were carried out during spring (temperature 21–26°C, humidity 65–80%). Sampling was performed at midday only during full light conditions. The trees were irrigated daily and each tree was wounded only once. 2.4. Chemical analysis After sampling the leaf material (10–15 mg) was immersed in CH2Cl2 (100 µl) spiked with 400 ppm octane (as an internal standard) for 24 h at room temperature in the dark. This extraction period was deemed to be sufficient as the detection of differences in oil concentration and composition before and after wounding rather than exhaustive extraction was the purpose of this study. The recovery of the internal standard after this period was found to be 97.3±2.4% for the flush growth and 95.1±2.1% for the mature leaves (mean±SEM, n = 8). Extracts were then subjected to GC and GC–MS analysis. GC analysis was carried out using a HP 6890 gas chromatograph equipped with a flame ionisation detector and a HP 7673 GC/SFE auto-injector. The column used was BPX-5 (50 m length×0.22 µm ID×0.25 µm film thickness) (SGE Scientific, Melbourne, Australia). The injection volume was 2 µl, inlet temperature and pressure were 280°C and 20 psi respectively, carrier gas was H2 (40 ml/min), split ratio was 1:10, detector temperature at 280°C, and the oven program was: initial temperature 60°C for 5 min, increased to 180°C at 4°C/min, final temperature maintained for 5 min. GC–MS analysis conditions were as above with He as the carrier gas. The instrument used was a Varian 3800 gas chromatograph connected to a Varian 2000 ion trap detector (0.9 scans/s, 20 µA). Compounds were identified by comparison of their mass spectra and retention indices (based on n-alkanes) with those of authentic standards (Fluka Chemicals, NSW, Australia). The relevant literature was also consulted (Brophy et al., 1989; Adams, 1995). 2.5. Oil concentration determination The concentration of each oil component (Ci) was determined by the equation: Ci = [(Cistd/Aistd)×Ai]×RRF, where Cistd is the concentration of the internal standard, Aistd is the peak area of the internal standard and Ai is the peak area of each component after normalisation (Rouessac and Rouessac, 2000). The relative response factor (RRF) of the internal standard in relation to all oil components was assumed to be 1. Any differences in the detector response between the internal standard and the oil components and also between the various oil components could introduce small bias in the determination but these are known to be consistent that do not affect trends over time (Murtagh and Smith, 1996). 2.6. Headspace experiment Six-month old, potted M. alternifolia plants were used. Part of a small branch (approximately 8–10 pairs of leaves) was placed in a small glass cylinder (20 ml in
402
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
volume) closed at one end with a Teflon cap. The cap was pre-drilled to fit exactly a manual solid-phase microextraction (SPME) holder (Supelco, USA). Glass wool was used to restrict the other end after the branch was in place, however, the system was not air-tight. After 30 min the headspace above the branch was sampled using a polydimethylsiloxane-coated fibre (10 min, 25°C). Thermal desortpion of the fibre in the injector of the GC and GC–MS system followed (220°C, 10 min, other conditions as above). Three of the enclosed leaves were then cut longitudinally in half. The headspace was then monitored for 1 h at 20-min intervals similarly as above. The whole procedure was repeated with unwounded leaves and the results were used to correct for depletion of the headspace effects and other headspace losses in the system. 2.7. Moisture determination Wounded leaves (experimental design as described above) were placed in a glass petri-dish and dried for 10 days at 50°C. Moisture was determined by the weighing of the leaves before and after drying. The experiment was performed three times for both mature and flush growth leaves. 2.8. Statistical analysis The contribution of each monoterpene (and of the sesquiterpenes measured as a group) in the oil was expressed as a percent value based on the peak areas of the integrated chromatograms. This was necessary in order to compare the oil-profiles before and after wounding. If absolute values (e.g., area counts/g) were to be used for the comparison, then changes in moisture, leaf development and/or oil concentration would have been superimposed on the data and the results obtained would be inaccurate and misleading. The difference between the percentage contribution of each compound in the oil before and after wounding was tested for statistical significance using the Wilkoxon signed ranks test (paired samples) (Sall and Lehman, 1996). A correlation matrix was then built (based on the actual percentages) using as variables the compounds found to be statistically significant. PCA was then performed on the correlations and the eigenvalues obtained were used to visualise the relationship between objects (leaves) in two-dimensional score plots. The association between the significant variables (compounds) was examined based on their loading values on the first two principal components. Data from independent samples (e.g., comparison between treatments) was analysed as above; however in this case the Kruskal–Wallis test was applied (Sall and Lehman, 1996). The effect of wounding on the non-parametric measures of association between the major thujanes and p-menthanes based on their percentage contribution to the flush growth extracts was also investigated (Spearman’s test) (Sall and Lehman, 1996). The software used for all data analyses was JMP-IN (version 3 for Windows) (SAS Institute Inc., USA).
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
403
3. Results and discussion 3.1. Oil concentration Fig. 1 shows the change in oil concentration in mature and flush growth M. alternifolia leaves caused by wounding. Although the extraction was carried out on fresh leaves the values are expressed on a dry weight basis due to the substantial variations that can occur with moisture content (Murtagh and Smith, 1996). The observed change was similar for both leaf types during the first 24 h but differed markedly 48 h after injury. After this time the oil concentration of the mature leaves appears to have returned to its pre-wounding level. In contrast, the flush growth leaves show a continuing loss as post-wounding time elapses (Fig. 1) although it is possible that some of this loss reflects the increasing weights of leaf structural features due to the ongoing ontogeny of the flush growth. Results from headspace experiments indicate that the initial post-wounding oil loss is due to the volatilisation of the oil constituents into the atmosphere. As can be seen from Fig. 2 the concentration of the headspace above M. alternifolia leaves increases dramatically immediately upon wounding and then gradually decreases. This is likely to be caused by the rupture of the leaf’s secretory cells although ‘intentional’ release of oil through modified epidermal cells is also possible (List et al., 1995). This increased emission of volatiles may be a preliminary response by the leaf to deter further predation.
Fig. 1. Changes in oil concentration of M. alternifolia leaves as a result of mechanical damage (mean±SE, n = 8). Identical superscripts indicate statistical significance between values (P⬍0.01, Wilkoxon signed ranks test).
404
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
Fig. 2. Changes in the headspace concentration above M. alternifolia leaves as a result of wounding (mean±SE, n = 3). Samples were adjusted to the same scale based on their unwounded (control) values. The immediate increase and then subsequent decrease in headspace concentration over time is evident.
The recovery of the oil concentration observed after 48 h for the mature leaves suggests that biosynthetic processes are active during the post-wounding period and replenish any oil losses caused by wounding. 3.2. Oil composition Studies on corn (Turlings et al., 1995) and cotton (Pare´ and Tumlinson, 1997) have shown that there is a delay of several hours between the start of wounding and the release of induced terpenoids. This suggests that a series of inducible biochemical reactions are required for those compounds to be produced and subsequently released. Similarly, if changes in oil from M. alternifolia were to be induced by mechanical wounding, the process would not be instantaneous. Therefore, the assumption that T0 values represent pre-wounding values appears to be valid in relation to oil composition. The typical T0 oil composition as obtained from the trees used in this experiment (for both leaf types) is given in Table 1. 3.2.1. Flush growth The statistical tests showed a significant change in the percentage composition of several constitutive terpenoids between unwounded and wounded M. alternifolia leaves (Table 1). The 24 h post-wounding results showed that the major p-menthanes (α- and γ-terpinene, terpinolene, terpinen-4-ol) increased in concentration at the expense of the thujanes (cis-sabinene hydrate, sabinene). The negative correlation between the two skeletal groups is better observed 48 h after wounding (Table 1). The conversion of the thujane precursors to the more stable p-menthanes occurs naturally in M. alternifolia flush growth as the leaves mature (Southwell and Stiff, 1989; Cornwell et al., 1995). During the aging process, cis-sabinene hydrate and to
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
405
Table 1 Time-zero (T0) oil composition from flush growth and mature Melaleuca alternifolia leaves used in this study. Range values are given to demonstrate the inter-plant variation observed. Values from both 24- and 48-h experiments are included. Selected components as detected after cold solvent extraction are presented Components
Percentage composition (%) (min.–max.) Flush growth (n = 16) Mature leaves (n = 15)
α-thujene α-pinene sabinene β-pinene myrcene α-terpinene cluster 2a γ-terpinene trans-sabinene hydrate terpinolene cis-sabinene hydrate trans-p-menth-2-en-1-ol cis-p-menth-2-en-1-ol terpinen-4-ol α-terpineol trans-piperitolb Unknown sesquiterpenes
0.50–0.70 1.41–1.82 3.33–7.42 0.61–0.73 0.61–0.94 1.10–2.50 2.01–8.72 1.91–4.62 5.24–6.72 0.42–0.83 30.41–46.32 0.21–0.36 0.23–0.39 2.51–5.74 0.61–1.73 0.32–0.54 0.20–0.37 21.62–34.43
a b
0.36–1.10 2.00–3.18 nd 0.64–1.13 0.79–1.95 3.95–7.80 5.48–9.00 8.60–20.6 nd 1.10–2.90 nd 0.10–0.75 0.10–0.78 27.72–43.91 2.01–6.46 0.50–1.90 0.40–1.30 9.17–20.71
cluster 2= p-cymene, limonene, β-phellandrene, 1,8-cineole Tentative identification
a lesser extent sabinene and trans-sabinene hydrate are known to be implicated in the formation of terpinen-4-ol, terpinolene, α- and γ-terpinene (Cornwell et al., 1995; Southwell, 1999). However, the results obtained in this study indicated that wounding has some effect over and above the normal ontogenetical activities occurring during flush growth maturation. From Table 2 it can be seen that in unwounded flush growth significant linear correlations were obtained only between concentrations of compounds of the same skeletal group (with the exception of the relationship of cis-sabinene hydrate with αterpinene). However, in wounded leaves several significant linear-correlations were observed between p-menthanes and their precursor thujanes. Sabinene and cis-sabinene hydrate appear to be strongly related to terpinen-4-ol and γ-terpinene whilst sabinene itself is also correlated with terpinolene. Also in wounded leaves α-terpinene was not linearly related with any of the major thujanes. It can only be speculated as to how leaf wounding possibly affects the ontogenetical changes in M. alternifolia flush growth. Recent work has established that the thujanes to p-menthanes transformations are pH and water dependent (Southwell and Stiff, 1989; Cornwell, 1999). Therefore it is possible that cell rupture alters pH and
406
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
Table 2 Effect of mechanical damage on oil composition in flush growth of M. alternifolia. Selected components as detected after cold solvent extraction are presented. Asterisks indicate statistical significance (*at P⬍0.02; **at P⬍0.01, Wilkoxon signed-ranks test) Components
α-thujene α-pinene sabinene cluster 1a α-terpinene cluster 2b γ-terpinene trans-sabinene hydrate terpinolene cis-sabinene hydrate terpinen-4-ol α-terpineol a b
Difference in percentage composition between unwounded and wounded leaves (mean±SEM) (n = 8) 24 h after injury ⫺0.01±0.01 ⫺0.03±0.03 ⫺0.44±0.23 ⫺0.06±0.05 0.34±0.06* 0.14±0.08 0.63±0.09* 0.04±0.11 0.29±0.04* ⫺1.54±1.05 1.19±0.23* 0.11±0.09
48 h after injury 0.04±0.03 0.06±0.05 ⫺0.78±0.15** 0.05±0.07 0.71±0.09* 0.18±0.11 0.98±0.25 ⫺0.13±0.10 0.40±0.07** ⫺4.57±1.30* 1.91±0.40** 0.08±0.06
Cluster 1 includes β-pinene and myrcene Cluster 2= p-cymene, limonene, β-phellandrene and cineole
water levels within the leaf thus affecting the chemical maturation of the flush growth. From an ecological perspective, rapid chemical maturation of the flush growth may be interpreted as an attempt to minimise subsequent predation. 3.2.2. Mature leaves Different sets of oil constituents for each treatment are utilised to express the effect of wounding on M. alternifolia mature leaves (Table 3). The monoterpenes α-thujene and cis-menth-2-en-1-ol are the only components that change significantly in both post-wounding measurements (Table 4). The other terpenes that initially change their concentration appear to revert back to their pre-wounding levels; a different set of monoterpenes is then used for the expression of the response. This indicates that mature leaves are biosynthetically active during the post-wounding period so some oil constituents are synthesised at the expense of others. This view is also supported by the fact that certain terpenes (e.g., trans- and cis-menth-2-en1-ol) greatly increased their presence in the oil (⬎60% increase) within 24 h (Tables 1 and 4). The increase in monoterpene levels at the expense of the sesquiterpenes in the first 24 post-wounding hours is also very interesting. Direct biosynthetic interdependence between the two terpene groups is not likely given that they are synthesised in different cellular compartments (Bohlmann et al., 1998). However it is possible that wounding affects the amount of substrate available or the terpenoid synthases responsible for the production of these oil constituents as compared to normal biosynthesis.
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
407
Table 3 Non-parametric correlation coefficients between the major thujanes and p-menthanes in M. alternifolia flush growth for unwounded (n = 14) and wounded leaves (both 24 and 48 h treatments, n = 16). Asterisks indicates statistical significance (*at P⬍0.05; ** at P⬍0.01, Spearman’s test) Variable by variable α-terpinene-sabinene γ-terpinene-sabinene γ-terpinene-α-terpinene trans-sabinene hydrate-sabinene trans-sabinene hydrate-αterpinene trans-sabinene hydrate-γterpinene terpinolene-sabinene terpinolene-α-terpinene terpinolene-γ-terpinene terpinolene-trans-sabinene hydrate cis-sabinene hydrate-sabinene cis-sabinene hydrate-α-terpinene cis-sabinene hydrate-γ-terpinene cis-sabinene hydrate-transsabinene hydrate cis-sabinene hydrate-terpinolene terpinene-4-ol-sabinene terpinen-4-ol-α-terpinene terpinen-4-ol-γ-terpinene terpinen-4-ol-trans-sabinene hydrate terpinen-4-ol-terpinolene terpinen-4-ol-cis-sabinene hydrate
Spearman’s ρ Unwounded leaves
Wounded leaves
⫺0.2478 ⫺0.2156 0.6619** 0.3633 0.0871
⫺0.2789 ⫺0.7574** 0.6966** 0.0324 0.0496
0.3917
0.1468
⫺0.1597 0.9319** 0.6935** 0.0000
⫺0.5403* 0.5583* 0.8657** 0.4122
0.5842* ⫺0.5957* ⫺0.3868 0.5233*
0.7828** ⫺0.5090 ⫺0.7710** 0.3291
⫺0.4286 ⫺0.1079 0.8070** 0.9281** 0.2923
⫺0.4293 ⫺0.7142** 0.6020* 0.9372** 0.3255
0.8574** ⫺0.4570
0.9407** ⫺0.6141*
It is possible that additional oil components were affected by leaf wounding. However, only the components that showed a consistent behaviour across the different trees used were labeled as significant by the statistical analysis. This together with the fact that half the trees were sampled one month apart from the other half ensured the reproducibility of the detected response. An experiment was conducted to test the hypothesis that the changes outlined above originated from the different oil composition that the two leaf halves might possessed due to developmental or other factors. Statistical analysis however found no significant differences (Wilkoxon signed-ranks test at Pⱕ0.05, n ⫽ 8) in the oil composition from the two halves. Also, repetitive injections of two different extracts showed the detected wounding response to be over and above the small errors introduced by the instrument used (Table 5). The reproducibility of the extraction process itself could not be determined due to the existing inter-leaf variation. However, the results obtained when the same extraction process was applied on flush growth indi-
408
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
Table 4 Effect of mechanical damage on oil composition in mature growth of M. alternifolia. Selected components as detected after cold solvent extraction are presented. Asterisks indicate statistical significance (* at P⬍0.05; ** at P⬍0.01, Wilkoxon signed-ranks test) Components
Difference in percentage composition between unwounded and wounded leaves (mean±S.E.M) (n=8) 24 h after injury 48 h after injury
-thujene α-pinene β-pinene myrcene α-terpinene cluster 2a γ-terpinene terpinolene trans-p-menth-2-en-1-ol cis-p-menth-2-en-1-ol terpinen-4-ol α-terpineol trans-piperitolb Unknown sesquiterpenes
0.11±0.01** 0.02±0.02 0.03±0.00 0.02±0.05 ⫺0.25±0.57 0.36±0.54 0.14±0.12 0.02±0.10 0.10±0.01** 0.11±0.01** ⫺0.28±0.35 0.05±0.03 0.40±0.24 0.21±0.14 ⫺1.06±0.35*
a b
0.12±0.01* 0.08±0.02 0.04±0.02 0.02±0.01 0.77±0.40 ⫺0.34±0.13* 0.27±0.15 0.18±0.07* 0.05±0.02 0.12±0.02* ⫺0.18±0.15 ⫺0.01±0.01 ⫺0.16±0.18 ⫺0.09±0.12 ⫺0.81±0.30
Cluster 2=p-cymene, limonene, β-phellandrene, 1,8-cineole Tentative identification
Table 5 Instrument reproducibility for the compounds found to be implicated in the wounding response of M. alternifolia leaves. Results from two extracts are shown (mean±SEM, n = 5). SEM is the standard error of the mean Compound
α-thujene cluster 2a terpinolene trans-p-menth-2-en-1-ol cis-p-menth-2-en-1-ol sesquiterpenes a
% (mean±SEM) Extract 1
Extract 2
0.732±0.009 6.45±0.018 1.94±0.001 0.740±0.014 0.281±0.006 12.6±0.090
0.882±0.010 5.41±0.007 2.11±0.002 0.745±0.011 0.300±0.002 11.7±0.069
Cluster 2= p-cymene, limonene, β-phellandrene, 1,8-cineole
cate its reliability as it was able to detect the well known ‘chemical maturation’ of M. alternifolia immature leaves. PCA two-dimensional score plots were constructed in order to elucidate further the detected wounding response (Fig. 3). Plots defined by the first two principal components in each case were sufficient for that purpose as it could explain most of the variation in the data (⬎88.2%). From Fig. 3 it can be seen that the wounding
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
409
Fig. 3. PCA scatterplots of oil from M. alternifolia mature leaves on the first two principal components (unwounded leaves squared, wounded leaves not squared). Identical numbers indicate same leaf (before– after wounding). (a) 24 h after injury, (b) 48 h after injury. Individuals 5–7 were sampled one month after individuals 1–4 to test the reproducibility of the response.
response in each mature leaf (for both treatments) is independent of the pre-wounding levels of the particular compound expressing the response. Also, the magnitude of the response appears to be similar between replicates within the same treatment, again, despite the large inter-tree variation. Similar behaviour was observed in the case of flush growth (data not shown). It is likely that the different sets of compounds involved in the post-wounding response serve distinct purposes. Differential timing in the production of defense oleoresin components has also been detected in the defense response of Grand fir
410
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
(Abies grandis) (Steele et al., 1998). In that case the monoterpenes formed immediately after wounding act as insect toxins while their later production at solvent levels is involved in wound sealing (Steele et al., 1998). No significant differences were found when the results from the two post-wounding measurements were compared to each other. This was probably due to the high level of inter-tree variation observed when independent samples (leaves from different trees) were considered during the statistical analysis. It is likely that the small compositional changes (if any) were ‘masked’ by the high inter-tree variability which can be seen in Table 1. The oil-profile differences demonstrated above between unwounded and wounded leaves are small in magnitude for both mature and immature leaves. Studies on cabbage (Coleman et al., 1997) and corn (Turlings et al., 1993) seedlings have shown that the wound-response in these plants is much greater when an elicitor (for example enzymes found in the predator’s saliva) is present rather than the response obtained by simple mechanical damage. Work in progress is currently testing the validity of this hypothesis in relation to terpene accumulating plants.
4. Conclusion In this study we were able to demonstrate compositional and quantitative changes in the oil from M. alternifolia mature and immature leaves as a result of mechanical damage. Wounding caused a decrease in the total oil concentration of all leaves. However, mature leaves were able to return to their pre-wounding oil concentration 48 h after damage. Leaf wounding stimulated increased production of certain terpenes at the expense of others. Both flush growth and mature leaves exhibited this response irrespective of total oil concentration, oil composition or pre-wounding concentration of the particular terpenes involved in the response. From a commercial point of view the effects of wounding on leaf total oil concentration are perhaps more significant than the observed compositional changes. However, it is likely that changes in both, oil concentration and composition serve important ecological roles during the leaf’s post-wounding period.
Acknowledgements The authors are grateful to the Faculty of Science and Technology, University of Western Sydney, Hawkesbury for the financial support. D.Z acknowledges receipt of an Australian Postgraduate Award (APA).
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
411
References Adams, R.P., 1995. Identification of Essential Oil Components by Gas Chromatography/Mass Spectrometry. Allured, Illinois. Bohlmann, J., Meyer-Gauen, G., Croteau, R., 1998. Plant terpenoid synthases: molecular biology and phylogenetic analysis. Proc. Natl. Acad. Sci. USA 95, 4126–4133. Bolter, C.J., Dicke, M., van Loon, J.J.A., Visser, J.H., Posthumus, M.A., 1997. Attraction of colorado potato beetle to herbivore-damaged plants during herbivory and after its termination. J. Chem. Ecol. 23, 1003–1023. Brophy, J.J., Davies, N.W., Southwell, I.A., Stiff, I.A., Williams, L.R., 1989. Gas chromatographic quality control for oil of Melaleuca terpinen-4-ol type (Australian tea trea). J. Agric. Food Chem. 37, 1330–1335. Coleman, R.A., Barker, A.M., Fenner, M., 1997. A test of possible indirect mediation of wound-induced resistance in cabbage against Pieris brassicae. Oikos 80, 43–50. Cornwell, C.P., Leach, D.N., Wyllie, S.G., 1995. Incorporation of oxygen-18 into terpinen-4-ol from the H218O steam distillates of Melaleuca alternifolia (Tea tree). J. Essent. Oil Res. 7, 613–620. Cornwell, C.P., 1999. Aspects of the chemistry and origins of the terpenes from Australian Myrtaceae. Ph.D. thesis. University of Western Sydney, Hawkesbury. Lewinsohn, E., Gijzen, M., Croteau, R., 1991b. Defense mechanisms of conifers: differences in constitutive and wound-induced monoterpene biosynthesis among species. Plant Physiol. 96, 44–49. Lewinsohn, E., Gijzen, M., Savage, T.J., Croteau, R., 1991a. Defense mechanisms of conifers: relationship of monoterpene cyclase activity to anatomical specialization and oleoresin monoterpene content. Plant Physiol. 96, 38–43. List, S., Brown, P.H., Walsh, K.B., 1995. Functional anatomy of the oil glands of Melaleuca alternifolia (Myrtaceae). Aust. J. Bot. 43, 629–641. Loughrin, J.H., Manukian, A., Heath, R.R., Turlings, T.C.J., Tumlinson, J.H., 1994. Diurnal cycle of emission of induced volatile terpenoids by herbivore-injured cotton plants. Proc. Natl. Acad. Sci. USA 91, 11836–11840. Murtagh, G.J., Smith, G.R., 1996. Month of harvest and yield components of tea tree. II. Oil concentration, composition and yield. Aust. J. Agric. Res. 47, 817–827. Pare´ , P.W., Tumlinson, J.H., 1997. De novo biosynthesis of volatiles induced by insect herbivory in cotton plants. Plant Physiol. 114, 1161–1167. Pare´ , P.W., Tumlinson, J.H., 1998. Cotton volatiles synthesized and released distal to the site of insect damage. Phytochem. 47, 521–526. Rouessac, F., Rouessac, A., 2000. Chemical Analysis: Modern Instrumental Methods and Techniques. Wiley, Chichester. Sall, J., Lehman, A., 1996. JMP Start Statistics. Duxbury Press, Belmont. Southwell, I.A., 1988. Australian tea tree: Oil of Melaleuca, terpinen-4-ol type. Chem. Aust. 55, 400–402. Southwell, I.A., 1999. Tea tree constituents. In: Southwell, I.A., Lowe, R. (Eds.), Tea Tree. The Genus Melaleuca. Hardwood Academic Publishers, Australia, pp. 29–62. Southwell, I.A., Markham, J., Mann, C., 1996. Is cineole detrimental to tea tree oil? Perfum. Flav. 21, 7–10. Southwell, I.A., Stiff, I.A., 1989. Ontogenetical changes in monoterpenoids of Melaleuca alternifolia leaf. Phytochemistry 28, 1047–1051. Steele, C.L., Katoh, S., Bohlmann, J., Croteau, R., 1998. Regulation of oleoresinosis in grand fir (Abies grandis). Differential transcriptional control of monoterpene, sesquiterpene and diterpene synthase genes in response to wounding. Plant Physiol. 116, 1497–1504. Takabayashi, J., Takahashi, S., Dicke, M., Posthumus, M.A., 1995. Developmental stage of herbivore Pseudaletia separata affects production of herbivore-induced synomone by corn plants. J. Chem. Ecol. 21, 273–287. Turlings, T.C.J., Loughrin, J.H., McCall, P.J., Ro¨ se, U.S.R., Lewis, W.J., Tumlinson, J.H., 1995. How caterpillar-damaged plants protect themselves by attracting parasitic wasps. Proc. Natl. Acad. Sci. USA 92, 4169–4174. Turlings, T.C.J., McCall, P.J., Alborn, H.T., Tumlinson, J.H., 1993. An elicitor in caterpillar oral
412
D. Zabaras et al. / Biochemical Systematics and Ecology 30 (2002) 399–412
secretions that induces corn seedlings to emit chemical signals attractive to parasitic wasps. J. Chem. Ecol. 19, 411–425. Turlings, T.C.J., Tumlinson, J.H., 1992. Systemic release of chemical signals by herbivore-injured corn. Proc. Natl. Acad. Sci. USA 89, 8399–8402. Zabaras, D., Wyllie, S.G., 2001. The effect of mechanical wounding on the composition of essential oil from Ocimum minimum L. leaves. Molecules 6, 79–86.