Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Foliar Essential Oil Glands of Eucalyptus Subgenus Eucalyptus (Myrtaceae) Are a Rich Source of Flavonoids and Related Non-Volatile Constituents

  • Jason Q. D. Goodger ,

    Contributed equally to this work with: Jason Q. D. Goodger, Samiddhi L. Seneratne

    jgoodger@unimelb.edu.au

    Affiliation School of BioSciences, The University of Melbourne, Melbourne, Victoria, Australia

  • Samiddhi L. Seneratne ,

    Contributed equally to this work with: Jason Q. D. Goodger, Samiddhi L. Seneratne

    Affiliation School of BioSciences, The University of Melbourne, Melbourne, Victoria, Australia

  • Dean Nicolle,

    Affiliation Currency Creek Arboretum, PO Box 808 Melrose Park, Currency Creek, SA, 5039, Australia

  • Ian E. Woodrow

    Affiliation School of BioSciences, The University of Melbourne, Melbourne, Victoria, Australia

Foliar Essential Oil Glands of Eucalyptus Subgenus Eucalyptus (Myrtaceae) Are a Rich Source of Flavonoids and Related Non-Volatile Constituents

  • Jason Q. D. Goodger, 
  • Samiddhi L. Seneratne, 
  • Dean Nicolle, 
  • Ian E. Woodrow
PLOS
x

Correction

9 May 2016: Goodger JQD, Senaratne SL, Nicolle D, Woodrow IE (2016) Correction: Foliar Essential Oil Glands of Eucalyptus Subgenus Eucalyptus (Myrtaceae) Are a Rich Source of Flavonoids and Related Non-Volatile Constituents. PLOS ONE 11(5): e0155568. https://doi.org/10.1371/journal.pone.0155568 View correction

Abstract

The sub-dermal secretory cavities (glands) embedded within the leaves of Eucalyptus (Myrtaceae) were once thought to be the exclusive repositories of monoterpene and sesquiterpene oils. Recent research has debunked this theory and shown that abundant non-volatile compounds also occur within foliar glands. In particular, glands of four species in subgenus Eucalyptus contain the biologically active flavanone pinocembrin. Pinocembrin shows great promise as a pharmaceutical and is predominantly plant-sourced, so Eucalyptus could be a potential commercial source of such compounds. To explore this we quantified and assessed the purity of pinocembrin in glands of 11 species of E. subg. Eucalyptus using Electro-Spray Ionisation Liquid Chromatography Mass Spectrometry of acetonitrile extracts and Gas Chromatography Mass Spectrometry analyses of hexane extracts of isolated glands which were free from other leaf tissues. Our results showed that the glands of subgenus Eucalyptus contain numerous flavanones that are structurally related to pinocembrin and often present in much greater abundance. The maximum concentration of pinocembrin was 2 mg g-1 dry leaf found in E. stellulata, whereas that of dimethylpinocembrin (5,7-dimethoxyflavanone) was 10 mg g-1 in E. oreades and that of pinostrobin (5-hydroxy-7-methoxyflavanone) was 12 mg g-1 in E. nitida. We also found that the flavanones are exclusively located within the foliar glands rather than distributed throughout leaf tissues. The flavanones differ from the non-methylated pinocembrin in the degree and positions of methylation. This finding is particularly important given the attractiveness of methylated flavonoids as pharmaceuticals and therapeutics. Another important finding was that glands of some members of the subgenus also contain flavanone O-glucosides and flavanone-β-triketone conjugates. In addition, glands contain free β-triketones, β-triketone heterodimers and chromone C-glucosides. Therefore, the foliar glands of this taxonomically distinct group of plants are a rich source of a range of flavonoids and other biologically active compounds with great commercial potential.

Introduction

One of the most distinctive features of the foliage of Eucalyptus (Myrtaceae) trees is the presence of numerous sub-dermal secretory cavities (glands). Until recently, Eucalyptus leaf glands were thought to exclusively contain essential oils (a mixture of monoterpenes and sesquiterpenes), but they are now known to house additional non-volatile constituents, only a few of which have been structurally elucidated [13]. Remarkably, it has been estimated that the non-volatile compounds in Eucalyptus glands can comprise up to 50% of gland lumen volume in some species [4]. In addition, the chemical classes of the non-volatile compounds present in the glands appear to show a level of taxonomic segregation between the five major subgenera of this large genus (>800 species). For example, in a survey of the glandular content of 19 eucalypts from three subgenera, members of the subgenera Symphyomyrtus and Eudesmia were shown to contain monoterpene acid glucose esters as the most abundant non-volatiles, whereas members of the subgenus Eucalyptus were found to contain the flavanone pinocembrin (5,7-dihydroxyflavanone). Interestingly, pinocembrin was not detected at all in Symphyomyrtus and Eudesmia [3].

This finding that flavonoids occur within the glands of at least some members of subgenus Eucalyptus is particularly noteworthy for two reasons. Firstly, research on the biosynthesis, transport and storage of these biologically active compounds will be considerably advantaged by being able to study such processes in isolated, functional glands that are free of other leaf tissues (see [4] for protocol). Secondly, localization of flavonoids within the gland lumen may offer some advantages to rapid selection of high yielding Eucalyptus genotypes, which is important given the growing interest in the potential therapeutic and pharmacological applications of the compounds, and a parallel interest in economic production systems (see recent reviews [59]). For example, pharmacological studies have shown the flavanone pinocembrin to exhibit properties that may make it amenable for treating cardiovascular diseases and cancer [10]. In addition, pinocembrin has been shown to have the ability to reduce reactive oxygen species and inflammation, and also to have neuroprotective effects such as protecting the brain against damage from ischemic stroke [11]. Furthermore, pinocembrin is readily absorbed following oral administration [12], and in vitro studies have shown it to be capable of passing through the blood-brain barrier in a passive transport process [13]. Indeed, the potential therapeutic utility of pinocembrin is so great that in 2008 it was approved by the State Food and Drug Administration of China for clinical trials in patients with ischemic stroke and participants are currently being recruited for phase II clinical trials [14].

At present, flavonoids are largely isolated from plant sources [15]. This is particularly the case for pinocembrin, where it is almost exclusively purified from whole leaf, flower, root or seed extracts from a wide assembly of 16 plant families, but surprisingly in a total of less than 50 species [10]. Pinocembrin can be synthesized chemically [16], but byproducts of excessive hydrogenation are produced during synthesis that may reduce its efficacy [17]. Plant families with the greatest number of pinocembrin-containing species are Myrtaceae with six Eucalyptus and one Syzygium species, Zingiberaceae with five Alpinia species, Piperaceae with three Piper species, Verbenaceae with two Lippia species and Lauraceae with two Cryptocarya species [10]. Within Myrtaceae, the compound has been found in fruit extracts of Syzygium samarangense [18], and in addition to the aforementioned foliar gland extracts of Eucalyptus gregsoniana, E. muelleriana, E. pauciflora and E. olsenii [3], pinocembrin has also been detected in whole leaf extracts of E. sieberi [19] and E. fraxinoides [20]. Furthermore, reanalysis of previously published mass spectral data from whole leaf extracts of a further 11 species of Eucalyptus [21] tentatively identified the matching mass of pinocembrin in all species, albeit at low levels in some [22]. Interestingly, all 17 pinocembrin-containing Eucalyptus species are members of the subgenus Eucalyptus, supporting the notion that members of this subgenus show promise as commercial sources of flavonoids with pharmaceutical and therapeutic potential.

If we are to progress consideration of Eucalyptus as possible commercial sources of flavonoids then two initial research questions need to be answered. First, are flavonoids such as pinocembrin largely or exclusively located in the oil glands? Given the ease with which glands can be visualized within intact, fresh leaves [23], an affirmative answer to this question will allow relatively rapid screening for high-yielding plants at both the species and population level, as has been successfully achieved in identifying high-yielding plants for Eucalyptus essential oil production [24]. It is noteworthy that in three species from subgenus Symphyomyrtus, non-volatile monoterpene acid glucose esters were shown to consistently comprise a constant proportion of gland volume (41, 48 and 49%, respectively) across a range of gland sizes within each species [4]. The second important question involves the purity of individual flavonoids such as pinocembrin: do individuals and species exist in which pinocembrin comprises the entire glandular non-volatile fraction or if not, what types and amounts of other flavonoids co-occur with pinocembrin? Purity at the gland level is clearly an important consideration because it will influence the ease with which pure whole leaf extracts can be obtained. Alternatively, the presence of other flavanones or more broadly, other flavonoids with similar biological activities may also make Eucalyptus species attractive targets for commercial harvesting of those compounds.

In this paper we address the issue of flavonoid content and purity within the leaf glands of plants from the subgenus Eucalyptus. We selected 11 species of E. subg. Eucalyptus based on the presence of large and abundant glands in their leaves and the ability to isolate sufficient amounts of these glands for chemical analyses.

Materials and Methods

Plant Material

Fully expanded leaves were collected from Eucalyptus species growing in Currency Creek arboretum, South Australia (35°25’45”S, 138°45’46”E) in April 2014. Leaves were harvested using a pole pruner and immediately sealed in snap-lock plastic bags and returned to the laboratory on dry ice. Samples were stored at -80°C until analysed.

Gland isolation and extraction

Foliar glands were isolated from the leaves of collected Eucalyptus species following the method described elsewhere [4] using pectinase P-4716 (Sigma-Aldrich, St. Louis, USA) enzyme from Aspergillus niger. The enzyme was diluted with distilled water (1:2 v/v). Each leaf was cut into approximately 4 × 20 mm strips and immersed in 1500 μl of diluted pectinase enzyme solution and incubated at 30°C with agitation at 600 rpm for 30 s at 1-minute intervals. The incubation time was altered from 12–24 h for different species depending on the time taken for complete digestion. Incubation was deemed complete when the leaf cuticle with attached epidermal layer could be easily removed with forceps, and mesophyll-free glands separated from vasculature and mesophyll cells via agitation. All glands isolated from leaf strips of each species were collected and two aliquots of 200 glands were randomly selected and immediately ground using vial micropestles, with one dissolved in 200 μl acetonitrile (3 days at 25°C) for extraction of non-volatiles and the other dissolved in 200 μl hexane (3 days at 50°C) for extraction of volatile terpenoids. All extracts were passed through 0.45 μm membrane filters (Phenex PTFE, Phenomenex) to remove any cellular debris.

A separate subset of leaves were scanned for leaf area determinations and leaf mass per unit area calculated by weighing scanned leaves after drying in an oven at 60°C. Gland density was determined as described elsewhere [23]. Average gland lumen volume for each species was estimated by imaging 30 isolated glands and measuring gland diameters using Image J (Version 1.49, NIH, Bethesda, MD, USA). Lumen volumes were then calculated assuming an ellipsoid shape as described elsewhere [4]. The assumption of an average glandular constituent density of 1 μg per nl was made based on the densities of known abundant constituents: 1,8-cineole, 0.92 μg nl-1; p-cymene, 0.86 μg nl-1; β-eudesmol, 0.95 μg nl-1; and pinocembrin, 1.39 μg nl-1.

To test if the key flavanones were exclusively localized to glands or if they were also present in other leaf tissues, six fully expanded E. nitida leaves were randomly selected and divided into halves by cutting along the midrib. The area of each leaf half was measured before one half was ground and extracted in acetonitrile, and the other half subjected to enzymatic digestion with only the subsequent isolated glands ground and extracted in acetonitrile. All samples were passed through a 0.45 μm membrane filter to remove any cellular debris. The amount of pinocembrin and pinostrobin per unit leaf area in the leaf and isolated gland extracts were determined using HPLC and statistically compared using paired t-tests (IBM SPSS Statistics version 22, IBM Armonk, NY, USA).

Chromatographic analyses

Non-volatile extracts of ground glands were fractionated using a Shimadzu Prominence High-Performance Liquid Chromatography (HPLC) system with photo diode array UV detection (190–500 nm). The column used was a Gemini C18 (5 μm, 150 × 4.6 mm; Phenomenex) analytical column eluted at a flow rate of 0.8 ml min-1. The eluent system was a gradient of acetonitrile and water (both acidified with acetic acid, 0.1%) from 30–50% acetonitrile over 7 min, followed by 50–95% over 7 min and then isocratic at 95% for a further 12 min, and finally a gradient of 95–100% over 0.25 min. Standards of flavonoids were purchased from Sigma-Aldrich (pinocembrin and pinostrobin; St. Louis, MO, USA) and from Indofine Chemical Company (dimethyl pinocembrin and dimethyl chrysin; Hillsborough, NJ, USA). Standard series were produced for quantification at the following wavelengths: 289 nm (pinocembrin), 288 nm (pinostrobin), 283 nm (dimethylpinostrobin), 263 nm (dimethylchrysin). Alpinetin was quantified at 288 nm using the standard series of pinostrobin.

Electro-Spray Ionisation Liquid Chromatography Mass Spectrometry (ESI-LCMS) of non-volatile extracts was conducted on an Agilent 6520 QTOF MS system (Santa Clara, CA, USA) with a dual spray ESI attached to an Agilent 1200 series HPLC with a diode array detector using the same column and eluent system as above. The MS was operated in both positive and negative modes. Positive mode was operated using the following conditions: nebulizer pressure 37 psi, gas flow-rate 12 l min-1, gas temperature 350°C, capillary voltage 4000V, fragmentor 150 and skimmer 65 V. In negative ion mode, a nebulizer pressure of 45 psi and a capillary voltage of 3500 V were used. Mass spectra were collected in the range of 70–1700 m/z. The analytical method comprised two scan experiments: the first was a full scan with a dwell time of 493 ms spectrum-1, followed by a second MS/MS experiment with a dwell time of 493 ms spectrum-1 where a collision energy of 20 V was applied. Analytes were also monitored with diode array UV detection at specific wavelengths of 278 nm and 340 nm with the bandwidth set at 4 nm. Chromatograms and mass spectra were evaluated using MassHunter software (Agilent).

Volatiles were extracted from ground glands in hexane containing 100 mg l-1 tridecane as an internal standard and incubated at 50°C for 3 days. Samples were then dehydrated with anhydrous Na2SO4. Hexane samples were analysed by Gas Chromatography with Flame Ionization Detection (GC-FID) using a Perkin Elmer Autosystem XC (Perkin Elmer, Melbourne, Australia) fitted with a Zebron ZB-5 low polarity column (30 m × 0.25 mm i.d., Phenomenex). Helium was used as the carrier gas at a flow rate of 1 ml min-1. The injector temperature was 250°C and detector temperature was 220°C. The column temperature was held at 120°C for 1 min following injection, then ramped at 7°C min-1 to 180°C and held at that temperature for a further 10 min. GC-FID constituent identification and quantification was based on standard series of key monoterpenes and sesquiterpenes as described elsewhere [25]. β-triketones were quantified based on a standard series of the sesquiterpene caryophyllene oxide (Sigma-Aldrich) due to the unavailability of commercial β-triketone standards.

Gas Chromatography Mass Spectrometry (GC-MS) was also performed on the hexane extracts using an Agilent 7890A GC and 5975C MS (Agilent Technologies, Santa Clara, USA). Samples (1 μl) were injected in splitless mode into a GC-MS system comprised of a Gerstel 2.5.2 autosampler, a 7890A Agilent gas chromatograph and a 5975C Agilent quadrupole MS (Agilent, Santa Clara, USA). The MS was adjusted according to the manufacturer’s recommendations using tris-(perfluorobutyl)-amine (FC-43). The following MS source conditions were used: injection temperature 250°C, transfer line 280°C, ion source 230°C, quadrupole 150°C, 70 eV (EI mode), 2.66 scans s-1, and scanning range m/z 50–600. The GC fractionation was performed on a 30 m VF-5MS column with 0.2 μm film thickness and a 10 m Integra guard column (J & W, Agilent). The carrier gas and temperature program were the same as for the GC-FID. Mass spectra were evaluated using Agilent MSD ChemStation E.02.02.1431 for GC-MS and monoterpenes, sesquiterpenes and β-triketones identified using either the NIST 11 or Adams 2012 mass spectra libraries.

Results and Discussion

Glands that were free from other tissues were successfully prepared from 11 E. subg. Eucalyptus species in this study (Table 1, Fig 1). From each species, 400 glands were randomly collected and half the glands were ground and extracted in acetonitrile and the other half were extracted and ground in hexane. All compounds identified in either solvent were therefore constituents of the gland lumen and/or the secretory cells bounding the lumen, not other leaf tissues.

thumbnail
Table 1. Taxonomic classification of the Eucalyptus species sampled in this study.

Species were chosen based on gland size and abundance and ease of isolation. Average leaf and gland parameters are presented. LMA, leaf mass per unit area.

https://doi.org/10.1371/journal.pone.0151432.t001

thumbnail
Fig 1. Representative foliar glands isolated from E. subg. Eucalyptus.

A) E. apiculata and B) E. dendromorpha. Glands were isolated free from other leaf tissues prior to solvent extraction and chromatographic analyses. Scale bars represent 450 μm.

https://doi.org/10.1371/journal.pone.0151432.g001

Glandular Flavanones

From the acetonitrile extracts, we were able to identify and quantify flavonoids based on comparison with a series of commercially available compounds. We found the flavanone pinocembrin to be a key constituent of the non-volatiles in foliar glands of all species from E. subg. Eucalyptus sect. Eucalyptus (see Table 1), a result in agreement with previous analyses of the gland lumen content of four E. subg. Eucalyptus species [3]. Interestingly, pinocembrin was not the only flavonoid detected in the gland extracts, and it was the most abundant glandular flavonoid in only two species—E. muelleriana and E. gregsoniana (Fig 2). We show here for the first time that the related di-methoxylated flavanone dimethylpinocembrin (5,7-dimethoxyflavanone), the structurally similar mono-methoxylated flavanones pinostrobin (5-hydroxy-7-methoxyflavanone) and alpinetin (5-methoxy-7-hydroxyflavanone), and the related dimethoxylated flavone dimethylchrysin (5,7-dimethoxyflavone) are also localized to foliar glands of E. sect. Eucalyptus species and that some of these compounds are in remarkably high abundance (Fig 2). Interestingly, we were unable to detect any flavonoids in the glands of the two E. subg. Eucalyptus species examined here that are not part of E. sect. Eucalyptus—E. suberea and E. brevistylis. Flavonoids have previously been shown to be compartmentalised in analogous plant secretory structures in select plants from other families. In particular, flavones have been detected together with monoterpenes in extracts of glandular trichomes collected from the leaf surface of species in the Lamiaceae (oregano [26] and mint [27]) and the Fabaceae (alfalfa [28]).

thumbnail
Fig 2. Total quantifiable flavonoids extracted from leaves of E. subg. Eucalyptus plants.

Concentration per unit leaf mass was calculated using flavonoid amount per gland (from HPLC analyses), estimated average gland number per leaf and average leaf dry mass.

https://doi.org/10.1371/journal.pone.0151432.g002

The co-occurrence of different methoxylated forms of unsubstituted B-ring flavanones has previously been observed in plants. In particular, pinocembrin, alpinetin, and dimethylpinocembrin were reported in whole leaf extracts of E. sieberi from E. subg. Eucalyptus [19]. Similarly, the flavanone dimethylpinocembrin and the flavone dimethylchrysin have been found together in whole leaf extracts of the Myrtaceous shrub Leptospermum scoparium [29] and the liverwort Tylimanthus renifolius [30], and also in rhizome extracts of Boesenbergia pandurata [31].

When the amount of flavonoids extracted from glands is presented on a leaf dry mass basis (Fig 3), it is clear that some species contain much higher concentrations of flavonoids in their leaves than others and that the relative proportion of constituent flavonoids also varies. For example, E. nitida and E. oreades both contained the highest concentration of total flavonoids (>1.2% w/leaf dw), but E. nitida contained mostly pinostrobin whereas E. oreades contained mostly dimethylpinocembrin. Indeed, in six of the species, the latter dimethylated form was the most abundant flavonoid, whereas in all species alpinetin was either the flavanone of lowest abundance or it was not detected (Fig 3).

thumbnail
Fig 3. Chemical structures of flavonoids quantified from foliar secretory cavities of 11 species within E. subg. Eucalyptus.

Dashed lines denote characteristic fragmentation patterns of flavanones and flavones observed in ESI-LCMS/MS.

https://doi.org/10.1371/journal.pone.0151432.g003

On a per gland basis E. oreades glands possessed the most total flavonoids (0.7 μg gland-1), followed by E. nitida and then E. apiculata (0.4 μg gland-1; Table 2). Dimethylpinocembrin was the most abundant flavonoid ranging from 60–84% of total flavonoids in six species (E. oreades, E. stellulata, E. falciformis and the three species from E. ser. Strictae—E. approximans, E. apiculata and E. dendromorpha). Pinocembrin was dominant in the two species E. muelleriana and E. gregsoniana (76 and 90%, respectively), although it should be noted that total flavanones were very low in both species. Pinostrobin was dominant in E. nitida (91% of flavanones) and the highest amount of alpinetin was in E. muelleriana (23%). Interestingly, the two species from E. ser. Radiatae differed in terms of constituent flavanones with E. falciformis dominated by dimethylpinocembrin (69%) and E. nitida dominated by pinostrobin (91%), although both had around 10% pinocembrin.

thumbnail
Table 2. Abundant, quantifiable constituents extracted from foliar glands of trees in E. subg. Eucalyptus.

Numbers in parentheses represent percentage abundance of a constituent within each compound class.

https://doi.org/10.1371/journal.pone.0151432.t002

Despite pinocembrin not being the dominant flavonoid in the glands of the majority of species sampled in this study, the maximum concentration of 2 mg g-1 leaf dw detected in E. stellulata remains comparable with commercial plant sources. For example, the concentration of pinocembrin extracted from seeds of the Chinese medicinal plant Alpinia katsumadai ranges from as low as 0.02 mg g-1 [32] to 2.5 mg g-1 [33], and Indian Curcuma ecalcarata rhizomes contain up to 4 mg g-1 pinocembrin [34]. The pinocembrin abundances described here are also similar to those found recently in whole leaf extracts of the E. subg. Eucalyptus species E. fraxinoides and E. sieberi which contained 2.4 and 3.3 mg g-1 pinocembrin, respectively [20]. Interestingly, in the studies on A. katsumadai, the monomethylated flavanone alpinetin was found in equal [33] or much greater amounts [32] than pinocembrin. This was not the case in any of the Eucalyptus species studied here, with only very low levels of alpinetin detected, but it is reminiscent of the high levels of pinostrobin and particularly dimethylpinocembrin we observed to co-occur with pinocembrin. It should be noted that only one tree per species was sampled for this study and given glandular monoterpenes in Eucalyptus show quantitative variation in natural populations [35], it is likely that flavanones will show similar intra-specific variability. Future work will aim to identify individuals of species such as E. stellulata with even greater amounts of pinocembrin and other flavanones in their foliar glands.

A comparison of flavonones extracted from isolated glands of E. nitida with those extracted from intact leaves (including glands) showed no statistical difference between the concentrations of the abundant flavonoids pinostrobin (t = -0.73, P = 0.498) or pinocembrin (t = -1.09, P = 0.33). This result suggests that the flavanones are exclusively housed within the glands, rather than distributed throughout different leaf tissues. A similar result was found for the localization of monoterpene acid glucose esters to glands of Eucalyptus species in other subgenera [2], and may represent a means of avoiding potential autotoxicity from these biologically active flavonoids.

Other gland constituents

Comparison of the acetonitrile and hexane extracts enabled quantification of total constituents per gland and the relative proportion of different classes of compounds within the gland lumena. In general, monoterpenes were the most abundant gland constituents in the E. subg. Eucalyptus species (Table 2). The exceptions to this were E. stellulata, in which flavanones were the most abundant gland constituents, E. gregsoniana and E. brevistylis, in which sesquiterpenes were the most abundant, and E. suberea, in which β-triketones predominated. As noted above, glands of E. suberea and E. brevistylis did not contain detectable flavonoids, but were the only species examined that had β-triketones in their glands. This result suggests that the presence of flavonoids and possibly the absence of free β-triketones in foliar glands may be characteristic of E. sect. Eucalyptus species. Glands of E. falciformis possessed the most monoterpenes (1.3 μg gland-1), followed by E. approximans, E. oreades and E nitida (1.0 μg gland-1). Overall, p-cymene, α-pinene and 1,8-cineole were the most abundant monoterpene constituents found in glands. In general, glands of all species contained low levels of sesquiterpenes, the highest being E. gregsoniana (0.1 μg gland-1). The major sesquiterpene observed was β-eudesmol ranging from 19–50% of total sesquiterpenes.

The sum of all quantifiable constituents per gland (Table 2) was plotted against the average gland lumen volume for each species (Fig 4). In some species, the total quantifiable constituents accounted for most of the gland lumen volume (assuming an average constituent density of 1 μg per nl), but in all species there was a portion of volume unaccounted for, and indeed in species such as E. gregsoniana, E. approximans and E. apiculata, the known constituents accounted for only a low proportion of gland lumen volume. This suggests that glands of the species studied contained other non-volatile compounds. This notion was reinforced by the number of unknown PDA peaks observed with HPLC fractionation of the acetonitrile extracts.

thumbnail
Fig 4. Relationship between quantifiable constituents per gland and the estimated gland lumen volume for 11 species within E. subg. Eucalyptus.

Quantifiable constituents were flavanonoids, monoterpenes, sesquiterpenes and triketones. Lumen data are the means (± 1 SE) of 30 cavities. Dotted line represents an idealised 1 to 1 relationship between lumen volume and total gland constituents, assuming an average constituent density of 1 μg per nl.

https://doi.org/10.1371/journal.pone.0151432.g004

We used ESI-LCMS/MS to identify unknown constituents in glands, focusing on the characteristic fragmentation patterns of particular compound classes and structures known to occur in Eucalyptus or more broadly within Myrtaceae. Firstly, unsubstituted B-ring flavanones characteristically produce fragments with neutral losses of 104 and 42 Da via heterocyclic C-ring fragmentation (Fig 2; [36, 37]). Searching for compounds with these characteristic losses in the mass spectra of the acetonitrile extracts showed many putative flavanones and more complex compounds containing flavanone moieties. In particular, a number of C-methyl flavanones were observed (Fig 5, S1 Fig), and this class of compound was particularly abundant in the E. muelleriana gland extract. Many of these compounds have previously been detected in the family Myrtaceae. For example: (1) strobopinin (5,7,dihydroxy-6-methylflavanone) has been found in Corymbia torelliana fruit resins [38]; (2) cryptostrobin (5,7,dihydroxy-8-methylflavanone), strobopinin, dimethylstrobopinin (5,7,dimethoxy-6-methylflavanone), and demethoxymatteucinol (6,8-dimethylpinocembrin) together with pinostrobin and pinocembrin have been found in extracts of whole Campomanesia adamantium leaves [39]; (3) strobopinin and desmethoxymatteucinol have been found in Heteropyxis canescens leaves [40]; (4) strobopinin, cryptstrobin, demethoxymattuecinol and 7-hydroxy-5-methoxy-6,8-dimethylflavanone together with pinocembrin have been found in Sygium samarangense leaves [41]; and (5) strobopinin together with dimethylpinocembrin have been found in Leptospermum scoparium leaves and stems [42]. It is particularly noteworthy that recent work on L. scoparium and L. morrisonii used Raman spectroscopy on intact leaves to show that both the β-triketone grandiflorone and unidentified C-methyl flavanones were compartmentalized within embedded foliar glands [43].

thumbnail
Fig 5. Compounds identified from extracts of isolated glands of E. subg. Eucalyptus.

Dashed lines denote characteristic fragmentation patterns observed in ESI-LCMS/MS. Identifications are based on known compounds from genus Eucalyptus or family Myrtaceae.

https://doi.org/10.1371/journal.pone.0151432.g005

Further analysis of glandular extracts using LC-MS/MS showed the presence of larger, more complex compounds that contained flavanone moieties. For example, E. stellulata glands contained two flavanone-glucosides. These compounds were identified based on the characteristic fragmentation of unsubstituted B-ring flavanones, together with the characteristic neutral loss of glucose (162 Da) from glucosides [44]. The first compound eluted at 6.4 min with a m/z of 419.13 [M + H]+ and 417.13 [M − H] in the respective ionization modes, corresponding to a molecular formula of C21H22O9. The loss of glucose from this compound via MS2 fragmentation produced a flavanone moiety with m/z 257.08 [M + H − 162]+ and 255.07 [M − H − 162] corresponding to pinocembrin, which then underwent subsequent fragmentation resulting in neutral losses of 104 Da in both ionization modes (S2 Fig). The second compound was of lower abundance based on PDA and ion current and eluted at 8.6 min with a m/z of 433.10 [M + H]+ corresponding to a molecular formula of C22H24O9. This compound fragmented with the loss of glucose to pinostrobin (m/z 271.09 [M + H − 162]+) which further fragmented with a neutral loss of 104 Da to m/z 167.04 [M + H − 162–104]+. The likely structures of these compounds are pinocembrin-7-O-glucoside and 5-methoxy-pinocembrin-7-O-glucoside (Fig 5), which have been shown to co-occur in aerial parts of Penthorum chinense (Penthoraceae; [45]). Moreover, both compounds have been found individually in many different plant families, but not previously in Myrtaceae. Interestingly, neither of these flavanone-glucosides was detected in any of the other Eucalyptus species in this study. It is noteworthy that flavonoid glucosides have previously been identified from extracts of monoterpene producing glandular trichomes of Phillyrea latifolia (Oleaceae; [46]).

LC-MS/MS analysis of the acetonitrile extract from E. meulleriana showed its glands to contain a suite of other compounds containing flavanone moieties. The parent masses and MS2 fragments of these compounds strongly suggest they are flavanone-triketone conjugates linked by an iso-butyl moiety. In particular, abundant compounds were detected at 16.9 min, with a m/z of 491.21 [M − H] corresponding to C29H32O7 and fragmenting to m/z [M − H] 181.09, 309.11 and 255.07 (pinocembrin; S3A Fig), and at 18.0 min, with a m/z of 505.23 [M − H] corresponding to C30H34O7 and fragmenting to m/z [M − H] 181.09, 323.13 and 269.08 (pinostrobin; S3B Fig). There is a consistent difference of 55 Da between the fragment ion m/z 181 and the loss of 236 Da (m/z 255 fragment) from the first compound, and the fragment ion m/z 181 and the loss of 236 Da (m/z 269 fragment) from the second compound suggesting fragmentation occurs either side of an iso-butyl bridge in both compounds (as shown in the fragmentation scheme in Fig 5, S3 Fig). A compound matching the parent mass and fragmentation pattern of compound 2 is known from the myrtaceous species Baeckia frutescens [47].

Similar flavanone-triketone conjugates are known from the literature, but none perfectly matched the mass spectra of the other compounds detected in E. muelleriana. For example, structurally similar compounds have been isolated from extracts of aerial parts of the myrtaceous shrubs Kunzea ambigua and K. baxterii in which the flavanone and β-triketone moieties are linked by an iso-pentyl, rather than iso-butyl moiety [48]. Kunzeanones A and B have also been isolated from leaves of K. ambigua and have flavanone and β-triketone moieties linked via two bridges: an iso-pentyl moiety and an oxygen bridge through the double bonded oxygen of the β-triketone and 7-hydroxyl of flavanone [49]. It should be noted that these single and double bridge compounds may be tautomers. Similarly, leaves from the myrtaceous shrub Luma chequen contain lumaflavanones A-C, structures with a flavanone bonded to a β-triketone through both an iso-propyl bridge and an oxygen bridge [50]. Finally, leucadenones A-D extracted from leaves of Melaleuca leucadendron have a β-triketone and a flavanone linked by a phenyl moiety [51]. Despite none of these structures being detected in either E. muelleriana or the other species examined here, the mass spectral data and fragmentation patterns suggest many similar compounds are likely to be present in the glands of E. subg. Eucalyptus trees and may be present in relatively high abundances.

Non-flavonoid glandular constituents were also detected in E. subg. Eucalyptus based on characteristic mass spectral fragmentation patterns. In addition to the free β-triketones detected in the hexane extracts of E. brevistylis and E. suberea glands, and the flavanone-triketone conjugates of E. muelleriana glands, the acetonitrile extract of E. gregsoniana also contained β-triketone heterodimers linked by an iso-pentyl moiety (Fig 5). These include compounds with high abundance (according to absorbance at 278 nm) detected at 16.8 min with a m/z of 431.21 corresponding to C24H32O7, 17.6 min with a m/z of 445.22 [M − H] corresponding to C25H34O7, 18.5 min with a m/z of 459.24 [M − H] corresponding to C26H36O7, and at 21.3 min with a m/z of 473.25 [M − H] corresponding to C27H38O7. These compounds are likely semimyrtucommulones identified from the Myrtaceous shrubs Myrtus communis [52, 53] and Kunzea sinclairii and K. ericoides [54] in which the syncarpic acid and phloroglucinol moieties fragment characteristically on either side of the iso-pentyl bridge (Fig 5; S4 Fig). A number of unknown compounds with a similar fragmentation pattern were observed in E. gregsoniana and other E. subg. Eucalyptus species, suggesting additional β-triketone heterodimers likely occur in the glands of the subgenus.

A final group of non-flavonoid glandular constituents exhibited the highly characteristic mass spectral fragmentation pattern of C-glucosides. In contrast to the abundant neutral loss of glucose (162 Da) commonly observed with the cleavage of the O-link in O-glucosides, ESI-MS/MS fragmentation of C-glucosides is restricted almost exclusively to the glycan resulting in fragments with characteristic neutral losses of 18, 66, 90, 96, 120 and 150 Da, which may be associated with additional water losses to give further neutral losses of 36, 54 and 138 Da (Fig 5, S5 Fig; [55]). Moreover, the relative abundance of fragments is diagnostic of the presence of 6-C or 8-C-glucosides. In particular, 6-C-glucosides show an increased abundance of 0,2X+: [M + H − 120]+ relative to 0,1X+: [M + H − 150]+ (Fig 5, S5 Fig; [56]), whereas 8-C-glucosides show losses of ≥96 Da without the smaller fragments associated with 6-C-glucoside fragmentation [55]. Based on this, and the calculated mass of the aglycone, the Eucalyptus glands are highly likely to contain up to four 6-C chromone glucosides, differing in the hydrocarbon chains attached to the chromone moiety.

The 6-C chromone glucoside isobiflorin was detected at 2.2 min with a m/z of 355.09 [M + H]+ corresponding to C16H18O9 (S5 Fig). The compound was detected in gland extracts of all species examined, albeit at lower levels (based on EIC) in E. approximans, E. dendromorpha and E. suberea. Isobiflorin appears to be relatively common in Myrtaceae and has previously been detected in leaves of Eucalyptus globulus [57], E. cypellocarpa [58], Baeckea frutescens [59] and Eugenia caryophyllata [60], and in buds of Syzygium aromaticum [61] and bark of S. johnsonii [62].

A related 6-C chromone glucoside was detected at 2.4 min with a m/z of 383.13 [M + H]+ corresponding to C18H22O9. This compound was found in at least trace amounts in all species except E. stellulata and was particularly prevalent (based on EIC) in E. brevistylis, E. gregsoniana, and E. muelleriana. As with isobiflorin, it has previously been observed in a number of myrtaceous species including the bark of Eucalyptus globulus [57], E. grandis, E. x urograndis and E. maidenii [63], and leaves of Baeckea frutescens [59, 64] and Kunzea ambigua [65].

Finally, two 6-C chromone glucoside isomers were detected at 2.6 and 4.3 min with m/z of 397.15 [M + H]+ corresponding to C19H24O9. The two isomers were only detected in E. gregsoniana glandular extracts, but have previously been found individually in Eucalyptus with the C-glucoside possessing a methlpropyl chain on the chromone moiety found in E. grandis, E. x urograndis and E. maidenii bark [63], and the C-glucoside with an isopropyl chain on the chromone found in E. maidenii branches [66].

Conclusions

The results presented here show that the foliar oil glands of Eucalyptus species from the subgenus Eucalyptus contain numerous flavanones that are structurally related to pinocembrin and often present in much greater abundance. We also show that these flavanones are exclusively located within the foliar glands rather than distributed throughout leaf tissues. The flavanones differ from the non-methylated pinocembrin in the degree and positions of methylation of both the A-ring hydroxyls and carbons. This finding is particularly important given the attractiveness of methylated flavonoids as pharmaceuticals and therapeutics due to their improved intestinal absorption [67], enhanced bioactivities [68], and increased metabolic stability [69], relative to non-methylated forms. We show for the first time glands of some E. subg. Eucalyptus members also contain flavanone O-glucosides and flavanone-β-triketone conjugates. In addition, glands contain free β-triketones, β-triketone heterodimers and chromone C-glucosides, together with their well characterized complements of monoterpene and sesquiterpene oils. Therefore, the foliar glands of this taxonomically distinct group of plants are a rich source of a range of flavonoids and other biologically active compounds with great commercial potential.

Supporting Information

S1 Fig. Representative mass spectra of putative flavanones and C-methyl flavanones detected from E. meulleriana glands.

https://doi.org/10.1371/journal.pone.0151432.s001

(PDF)

S2 Fig. Representative mass spectra of a putative flavanone O-glucoside from E. stellulata glands.

https://doi.org/10.1371/journal.pone.0151432.s002

(PDF)

S3 Fig. Representative mass spectra of putative flavanone β-triketone conjugates from E. muelleriana glands.

https://doi.org/10.1371/journal.pone.0151432.s003

(PDF)

S4 Fig. Representative mass spectra of putative β-triketone heterodimers from E. gregsoniana glands.

https://doi.org/10.1371/journal.pone.0151432.s004

(PDF)

S5 Fig. Representative mass spectra of a putative chromone C-glucoside from E. gregsoniana glands.

https://doi.org/10.1371/journal.pone.0151432.s005

(PDF)

Acknowledgments

We thank Metabolomics Australia for MS analyses.

Author Contributions

Conceived and designed the experiments: JG IW. Performed the experiments: SS DN JG. Analyzed the data: JG. Contributed reagents/materials/analysis tools: DN IW JG. Wrote the paper: JG IW.

References

  1. 1. Goodger JQD, Cao B, Jayadi I, Williams SJ, Woodrow IE. Non-volatile components of the essential oil secretory cavities of Eucalyptus leaves: discovery of two glucose monoterpene esters, cuniloside B and froggattiside A. Phytochem. 2009;70:1187–94.
  2. 2. Goodger JQD, Woodrow IE. α, β-unsaturated monoterpene acid glucose esters: Structural diversity, bioactivities and functional roles. Phytochem. 2011;72:2259–66.
  3. 3. Heskes AM, Goodger JQD, Tsegay S, Quach T, Williams SJ, Woodrow IE. Localization of oleuropeyl glucose esters and a flavanone to secretory cavities of Myrtaceae. PLoS One. 2012;7(7):e40856. pmid:22911712
  4. 4. Goodger JQD, Heskes AM, Mitchell MC, King DJ, Neilson EH, Woodrow IE. Isolation of intact sub-dermal secretory cavities from Eucalyptus. Plant Methods. 2010;6:20. pmid:20807444
  5. 5. Agrawal AD. Pharmacological activities of flavonoids: a review. International Journal of Pharmaceutical Sciences and Nanotechnology. 2011;4(2):1394–8.
  6. 6. Costa DC, Costa HS, Albuquerque TG, Ramos F, Castilho MC, Sanches-Silva A. Advances in phenolic compounds analysis of aromatic plants and their potential applications. Trends Food Sci Technol. 2015;45(2):336–54.
  7. 7. Peluso I, Miglio C, Morabito G, Ioannone F, Serafini M. Flavonoids and immune function in human: a sytematic review. Crit Rev Food Sci Nutr. 2015;55(3):383–95. pmid:24915384
  8. 8. Solanki L, Parihar P, Mansuri ML, Parihar MS. Flavonoid-based therapies in the early management of neurodegenerative diseases. Advances in Nutrition. 2015;6(1):64–72. pmid:25593144
  9. 9. Chen ZJ, Zheng SR, Li LP, Jiang HD. Metabolism of flavonoids in hueman: a comprehensive review. Curr Drug Metab. 2014;15(1):48–61. pmid:24588554
  10. 10. Rasul A, Millimouno FM, Ali Eltayb W, Ali M, Li J, Li X. Pinocembrin: a novel natural compound with versatile pharmacological and biological activities. BioMed Research International. 2013;379850:1–9.
  11. 11. Lan X, Wang W, Li Q, Wang J. The natural flavonoid pinocembrin: molecular targets and potential therapeutic applications. Mol Neurobiol. 2015:1–8.
  12. 12. Sayre CL, Takemoto JK, Martinez SE, Davies NM. Chiral analytical method development and application to pre-clinical pharmacokinetics of pinocembrin. Biomed Chromatogr. 2013;27(6):681–4. pmid:23212747
  13. 13. Yang ZH, Sun X, Qi Y, Mei C, Sun XD, Du GH. Uptake characteristics of pinocembrin and its effect on p-glycoprotein at the blood-brain barrier in in vitro cell experiments. J Asian Nat Prod Res. 2012;14(1):14–21. pmid:22263589
  14. 14. Phase II study of pinocembrin injection to treat ischemic stroke [Internet]. ClinicalTrials.gov U.S. National Institutes of Health. 2015 [cited 03/11/2015]. Available from: http://www.clinicaltrials.gov/ct2/show/NCT02059785?term=pinocembrin&rank=1.
  15. 15. Terahara N. Flavonoids in foods: a review. Natural Product Communications. 2015;10(3):521–8. pmid:25924542
  16. 16. Fowler ZL, Koffas MAG. Biosynthesis and biotechnological production of flavanones: current state and perspectives. Appl Microbiol Biotechnol. 2009;83(5):799–808. pmid:19475406
  17. 17. Yang Q-Y, Tong Y-F, Chen F, Qi Y, Li W, Song W. Identification and synthesis of impurities in pinocembrin-a new drug for the treatment of ischemic stroke. Chinese Journal of Chemistry. 2012;30(6):1315–9.
  18. 18. Simirgiotis MJ, Adachi S, To S, Yang H, Reynertson KA, Basile MJ, et al. Cytotoxic chalcones and antioxidants from the fruits of a Syzygium samarangense (Wax Jambu). Food Chem. 2008;107(2):813–9. pmid:22359426
  19. 19. Bick IRC, Brown RB, Hillis WE. Three flavanones from leaves of Eucalyptus sieberi. Aust J Chem. 1972;25(2):449–51.
  20. 20. Saraf I, Choudhary A, Sharma RJ, Dandi K, Marsh KJ, Foley WJ, et al. Extraction of pinocembrin from leaves of different species of Eucalyptus and its quantitative analysis by qNMR and HPTLC. Natural Product Communications. 2015;10(3):379–82. pmid:25924509
  21. 21. Eschler BM, Pass DM, Willis IR, Foley WJ. Distribution of foliar formylated phloroglucinol derivatives amongst Eucalyptus species. Biochem Syst Ecol. 2000;28:813–24. pmid:10913843
  22. 22. Tucker DJ, Wallis IR, Bolton JM, Marsh KJ, Rosser AA, Brereton IM, et al. A metabolomic approach to identifying chemical mediators of mammal–plant interactions. J Chem Ecol. 2010;36:727–35. pmid:20556637
  23. 23. Brooker MIH, Nicolle D. Atlas of leaf venation and oil gland patterns in the eucalypts. Collingwood, Australia: CSIRO Publishing; 2013. 226 p.
  24. 24. Goodger JQD, Heskes AM, King DJ, Gleadow RM, Woodrow IE. Micropropagation of Eucalyptus polybractea selected for key essential oil traits. Funct Plant Biol. 2008;35(3):247–51.
  25. 25. Goodger JQD, Connolly CA, Woodrow IE. Examination of the consistency of plant traits driving oil yield and quality in short-rotation coppice cultivation of Eucalyptus polybractea. For Ecol Manage. 2007;250(3):196–205.
  26. 26. Bosabalidis A, Gabrieli C, Niopas I. Flavone aglycones in glandular hairs of Origanum × intercendens. Phytochem. 1998;49(6):1549–53.
  27. 27. Voirin B, Bayet C, Coulso M. Demonstration that flavone aglycones accumulate in the peltate glands of Mentha × piperita leaves. Phytochem. 1993;34:85–7.
  28. 28. Aziz N, Paiva NL, May GD, Dixon RA. Transcriptome analysis of alfalfa glandular trichomes. Planta. 2005;221:28–38. pmid:15578217
  29. 29. Häberlein H, Tschiersch K-P. On the occurrence of methylated and methoxylated flavonoids in Leptospermum scoparium. Biochem Syst Ecol. 1998;26:97–103.
  30. 30. Feld H, Zapp J, Becker H. Secondary metabolites from the liverwort Tylimanthus renifolius. Phytochem. 2003;64:1335–40.
  31. 31. Jaipetch T, Reutrakul V, Tuntiwachwuttikul P, Santisuk T. Flavonoids in the black rhizomes of Boesengergia pandurata. Phytochem. 1982;22(2):625–6.
  32. 32. Li A, Sun A, Liu R. Prepatative isolation and purification of three flavonoids from the Chinese medicinal plant Alpinia katsumadai Hayata by high-speed counter-current chromatography. J Liq Chromatogr Rel Technol. 2012;35:2900–9.
  33. 33. Yamamoto N, Kawabata K, Sawada K, Ueda M, Fukuda I, Kawasaki K, et al. Cardamonin stimulates glucose uptake through translocation of glucose transporter-4 in L6 myotubes. Phytother Res. 2011;25:1218–24. pmid:21305634
  34. 34. Rameshkumar KB, Alan Sheeja DB, Nair MS, George V. Curcuma ecalcarata—new natural source of pinocembrin and piperitenone. Nat Prod Res. 2015;29(13):1276–9. pmid:25553726
  35. 35. Goodger JQD, Woodrow IE. Genetic determinants of oil yield in Eucalyptus polybractea R.T. Baker. Trees. 2012;26:1951–6.
  36. 36. Ma YL, Li QM, Van den Heuvel H, Claeys M. Characterization of flavone and flavonol aglycones by collision-induced dissociation tandem mass spectrometry. Rapid Communications in Mass Spectrometry. 1997;11:1357–64.
  37. 37. Porter B, Fabre N, Rozenberg R, Habib-Jiwan J-L, Moulis C, Quetin-Leclercq J. Analysis of minor flavonoids in Piper hostmannianum var. berbicense using liquid chromatography coupled with atmospheric pressure chemical ionization mass spectrometry. J Chromatogr. 2008;1210:45–54.
  38. 38. Massaro CF, Katouli M, Grkovic T, Vub H, Quinn RJ, Heard TA, et al. Anti-staphylococcal activity of C-methyl flavanones from propolis of Australian stingless bees (Tetragonula carbonaria) and fruit resins of Corymbia torelliana (Myrtaceae). Fitoterapia. 2014;95:247–57. pmid:24704551
  39. 39. Coutinho ID, Coelho RG, Kataoka VMF, Honda NK, Silva JRM, Vilegas W, et al. Determination of phenolic compounds and evaluation of antioxidant capacity of Campomanesia adamantium leaves. Eclética Química. 2008;33(4):53–60.
  40. 40. Mohammed AMA, Coombes PH, Crouch NR, Mulholland DA. Non-volatile isolates of two Heteropyxis species: a first chemotaxonomic assessment of subfamily Psiloxyloideae (Myrtaceae). Biochem Syst Ecol. 2009;37:241–3.
  41. 41. Kuo YC, Yang LM, Lin LC. Isolation and immunomodulatory effect of flavonoids from Syzygium samarangense. Planta Med. 2004;70(12):1237–9. pmid:15643565
  42. 42. Mayer R. Flavonoids from Leptospermum scoparium. Phytochem. 1990;29(4):1340–2.
  43. 43. Killeen DP, van Klink JW, Smallfield BM, Gordon KC, Perry NB. Herbicidal b-triketones are compartmentalized in leaves of Leptospermum species: localization by Raman microscopy and rapid screening. New Phytol. 2015;205(1):339–49. pmid:25103692
  44. 44. Domon B, Costello CE. A systematic nomenclature for carbohydrate fragmentations in FAB-MS/MS spectra of glycoconjugates. Glycoconjugate Journal. 1988;5:397–409.
  45. 45. Guo W, Jiang Y, Chen X, Yu P, Wang MY, Wu X, et al. Identification and quantitation of major phenolic compounds from Penthorum chinense pursh. by HPLC with tandem mass spectrometry and HPLC with diode array detection. Journal of Separation Science. 2015;38:2789–96. pmid:26037645
  46. 46. Tattini M, Gucci R. Ionic relations of Phillyrea latifolia L. plants during NaCl stress and relief from stress. Can J Bot. 1999;77:969–75.
  47. 47. Tsui W-Y, Brown GD. Unusual metabolites of Baeckea frutescens. Tetrahedron. 1996;52(29):9735–42.
  48. 48. Khambay BPS, Beddie DG, Simmonds MSJ. An insecticidal mixture of tetramethylcyclohexenedione isomers from Kunzea ambigua and Kunzea baxterii. Phytochem. 2002;59:69–71.
  49. 49. Ito H, Iwamori H, Kasajima N, Kaneda M, Yoshida T. Kunzeanones A, B and C: novel alkylated phloroglucinol metabolites from Kunzea ambigua. Tetrahedron. 2004;60:9971–6.
  50. 50. Labbe C, Coll J, Connolly JD, Faini F, Farrugia LJ, Rycroft DS. Bioactive flavanones from Luma chequen. Collection of Czechoslovak Chemical Communications. 2002;67(1):115–23.
  51. 51. Lee C-K. Leucadenone A-D, the novel class flavanone from the leaves of Melaleuca leucadendron L. Tetrahedron Lett. 1999;40:7255–9.
  52. 52. Appendino G, Bianchi F, Minassi A, Sterner O, Ballero M, Gibbons S. Oligomeric acylphloroglucinols from myrtle (Myrtus communis). J Nat Prod. 2002;65(3):334–8. pmid:11908974
  53. 53. Fiorini-Puybaret C, Aries M-F, Fabre B, Mamatas S, Luc J, Degouy A, et al. Pharmacological properties of Myrtacine and its potential value in acne treatment. Planta Med. 2011;77(14):1582–9. pmid:21452108
  54. 54. Bloor SJ. Antiviral phloroglucinols from New Zealand Kunzea species. J Nat Prod. 1992;55(1):43–7. pmid:1318355
  55. 55. March RE, Lewars EG, Stadey CJ, Miao X-S, Zhao X, Metcalfe CD. A comparison of flavonoid glycosides by electrosprary tandem mass spectrometry. International Journal of Mass Spectrometry. 2006;248:61–85.
  56. 56. Waridel P, Wolfender J-L, Ndjoko K, Hobby KR, Major HJ, Hostettmann K. Evaluation of quadrupole time-of-flight tandem mass spectrometry and ion-trap multiple-stage mass spectrometry for the differentiation of C-glycosidic flavonoid isomers. J Chromatogr. 2001;926:29–41.
  57. 57. Al-Sayed E, Hamid HA, Abu El Einin HM. Molluscicidal and antischistosomal activities of methanol extracts and isolated compounds from Eucalytus globulus and Melaleuca styphelioides. Pharm Biol. 2014;52(6):698–705. pmid:24824322
  58. 58. Ito H, Koreishi M, Tokuda H, Nishino H, Yoshida T. Cypellocarpins A-C, phenol glycosides esterified with oleuropeic acid, from Eucalyptus cypellocarpa. J Nat Prod. 2000;63:1253–7. pmid:11000030
  59. 59. Kamiya K, Satake T. Chemical constituents of Baeckia frutescens leaves inhibit copper-induced low-density lipoprotein oxidation. Fitoterapia. 2010;81(3):185–9. pmid:19699785
  60. 60. Kim HJ, Lee JS, Woo E-R, Kim MK, Yang BS, Yu YG, et al. Isolation of virus-cell fusion inhibitory components from Eugenia caryophyllata. Planta Med. 2001;67(3):277–9. pmid:11345703
  61. 61. Hah A-R, Paik Y-S. Identification and PEP inhibitory activity of acetophenone glucosides from the clove buds (Syzygium aromaticum). Journal of the Korean Society for Applied Biological Chemistry. 2010;53(6):847–51.
  62. 62. Harris SM, McFeeters H, Ogungbe IV, Cruz-Vera LR, Setzer WN, Jackes BR, et al. Peptidyl-tRNA hydrolase screening combined with molecular docking reveals the antibiotic potential of Syzygium johnsonii bark extract. Natural Product Communications. 2011;6(10):1421–4. pmid:22164773
  63. 63. Santos SAO, Villaverde JJ, Freire CSR, Dominigues MRM, Pascoal Neto C, Silvestre AJD. Phenolic composition and antioxidant activity of Eucalyptus grandis, E. urograndis (E. grandis x E. urophylla) and E. maidenii bark extracts. Ind Crops Prod. 2012;39:120–7.
  64. 64. Adib AM, Jamaludin F, Kiong LS, Hashim N, Abdullah Z. Two-dimensional correlation infrared spectroscopy applied to analyzing and identifying the extracts of Baeckea frutescens medicinal materials. Pharm Biol. 2014;52(6):698–705.
  65. 65. Ito H, Kasajima N, Tokuda H, Nishino H, Yoshida T. Dimeric flavonol glycoside and galloylated C-glucosylchromones from Kunzea ambigua. J Nat Prod. 2004;67(3):411. pmid:15043420
  66. 66. Tian L-W, Xu M, Li Y, Li X-Y, Wang D, Zhu H-T, et al. Phenolic compounds from the branches of Eucalyptus maidenii. Chem Biodivers. 2012;9:123–30. pmid:22253109
  67. 67. Wen X, Walle T. Methylated flavonoids have greatly improved intestinal absorption and metabolic stability. Drug Metab Disposition. 2006;34(10):1786–92.
  68. 68. Wilson MA, Rimando AM, Wolkow CA. Methoxylation enhances stilbene bioactivity in Caenorhabditis elegans. BMC Pharmacol. 2008;8:15. pmid:18700960
  69. 69. Wen X, Walle T. Methylation protects dietary flavonoids from rapid hepatic metabolism. Xenobiotica. 2006;36(5):387–97. pmid:16854778