Abstract
Bush tea (Athrixia phylicoides DC.) is a popular medicinal South African indigenous plant and it has been used for many decades as a health beverage and medicine. The objective of the study was to profile metabolites for assessment of quality of bush tea (A. phylicoides DC.) subjected to different pruning levels. Treatments consisted of untreated control, top-branch pruning, middle pruning, and basal pruning arranged in a randomized complete block design (RCBD) using 10 single trees as replications. The liquid chromatography quadrupole time-of-flight mass spectrometry (LC–QTOF–MS) was carried out to annotate the bush tea metabolites present in bush tea. Orthogonal partial least square-discriminatory analysis (OPLS-DA) from 1H nuclear magnetic resonance (NMR) revealed a separation between the basal, middle, top pruning, and the unpruned bush tea plants. The pruned (top) and unpruned tea plants, exhibited higher levels of metabolites than the basal and middle pruned. Pruning bush tea showed a significant effect on accumulation of secondary metabolites and thus could enhance bush tea quality. The study successfully annotated 28 metabolites (compounds), which elucidated canonical differences in pruning treatment of bush tea, as validated through multivariate analysis. Top pruning (apically pruned) resulted in improved metabolite accumulation than other treatment and can be recommended in bush tea cultivation. Future studies to enhance vegetative enhancement after pruning will be evaluated.
Bush tea (A. phylicoides DC.) is a South African indigenous and traditional herbal tea rich in secondary metabolites, which have therapeutic effects (Padayachee, 2011), pharmacological properties (McGaw et al., 2007), and different phenolic and flavonoid compounds (Maudu et al., 2010). These secondary metabolites are compounds synthesized from primary metabolites as the plant interacts with its environment for adaptation or defense (Ramakrishna and Ravishankar, 2011). The compounds are used by the plants against herbivores and pathogen attack and also environmental stress. Thus, concentrations of different metabolites are highly dependent on the growing conditions and the type of stresses the plant is subjected to viz., pathogen attack, ultraviolet irradiation, high light, wounding, nutrient deficiencies, temperature, and herbicides (Ramakrishna and Ravishankar, 2011).
In tea cultivation, cultural practices such as mineral nutrition (Maedza, 2015), irrigation (Bandara, 2012), harvesting methods (Mphangwe, 2012), and processing (Hlahla et al., 2010) have been reported to influence the chemical composition of tea (Njogu et al., 2014; Venkatesh et al., 2007). Moreover, agronomic practices such as nutrient application was found to enhance accumulation of carbohydrates for plant growth and to increase photosynthetic rates (Haukioja et al., 1998), which results in the biosynthesis of carbon-based secondary metabolites, such as flavonoids, phenolic acids, and tannins, known as total polyphenols (Chabeli et al., 2008). Pruning is also one of the most important cultural practices that has been reported to improve productivity and quality of tea (Camellia sinensis) (Yilmaz et al., 2004).
Although several studies have investigated phytochemicals in bush tea, there is lack of literature on the effect of pruning on accumulation of metabolites in bush tea. Herbal tea extracts have gained popularity and their chemical analysis has also increased, nonetheless, there are no studies which reported the comprehensive chemical characterization of bush tea using the NMR and liquid chromatography linked to mass spectometry (LC–MS) techniques to quantify differences in metabolic profiles of bush tea. Therefore, the objective of this study was to profile metabolites for assessment of quality of bush tea subjected to different pruning types.
Materials and Methods
Experimental site.
The trial was conducted at the Agricultural Research Council, Vegetable and Ornamental Plant Institute, situated ≈25 km north of Central Pretoria on the Moloto/KwaMhlanga Road (R573), GPS coordinates 25°59″S; 28°35″E. The farm covers ≈4000 ha, of which only 650 ha is under irrigation. The bush tea was grown under 40% shade net and drip irrigation at frequency of 5 mm per 2 h throughout the study.
Treatments and experimental design.
Treatments consisted of untreated control, top-branch pruning, middle pruning, and basal pruning arranged in a RCBD using 10 single plants as replications. The untreated control involved leaving individual bush tea plants intact with no pruning from the start to end of the study, top branch (apical) pruning involved pruning of all top branches and stems of individual bush tea plants at the top up to 15 cm length, middle pruning involved pruning of all branches and stems of individual plants right in the middle of individual plants, whereas basal pruning involved pruning individual plants at the base just above the soil surface. Treatments were imposed once per the duration of the study, with all plant variables collected 90 d after starting the treatments.
Leaf area index (LAI).
Leaf area index was measured nondestructively, using a LAI 2200 plant canopy analyzer (LI-COR Bioscience, Lincoln, NE). The instrument uses measurements made above and below the canopy to calculate light interception at five zenith angles, from which LAI is computed using a model of radiative transfer in vegetative canopies.
Stomatal conductance (gS).
Stomatal conductance was measured using SC-1 leaf porometer (Decagon Devices, Pullman, WA). Stomatal conductance was described as a function of the density, size, and degree of opening of stomata. The measurements were taken on the abaxial (bottom) side of a fully expanded mature leaf, during midday when the environmental factors were at their peak.
Chlorophyll content.
Leaf chlorophyll content was measured with a chlorophyll content meter (SPAD 502 plus; Konica Minolta, Chiyoda, Tokyo, Japan).
Sample preparation.
Two-year mature leaf and twig samples were harvested from bush tea plants. Harvested leaves were air-dried at room temperature in the shade. Samples were then ground to a powder using a benchtop grinder and stored in glass vials below −50 °C until extraction.
1H-NMR data acquisition.
NMR spectral data were obtained using a 600 MHz 1H NMR spectrometer (Varian, Inc., Palo Alto, CA), with 40 scans recorded. The sample preparation, data acquisition, analysis, mining, and processing were carried out as described by Maree and Viljoen (2012). The powdered bush tea samples of 10 mg were weighed in 2-mL Eppendorf tubes for extraction and analysis. A solvent of 1.0 mL of solvent, which was made by a ratio of 1:1 methanol D4 and deuterium oxide (pH 6.0), was added. The mixture was vortexed at room temperature for 1 min and ultrasonicated for 15 min to break the cell membranes, and centrifuged for 5 min. About 0.5 mL of the supernatant was transferred into NMR tubes for analysis.
The phasing and baseline corrections were performed using MestreNova software (10.0; Mestrelab Research, Santiago de Compostela, Spain) with consistent settings for all sample spectra. The chemical shift ranges 4.70–4.90 and 3.23–3.36 representing water and methanol, respectively, were excluded from further analysis (Mediani et al., 2012). SIMCA-P (software 14.0; Umetrics, Umeå, Sweden) with Pareto scaling method was used to perform the multivariate data analysis using the principal component analysis (PCA) and OPLS-DA. Contribution plots and scatter plots from PCA were constructed to identify and evaluate the groupings, trends, and strong outliers.
Liquid Chromatography Linked to Mass Spectometry Analysis
Extraction of bush tea samples for LC–MS.
Ten milligrams of each sample (five replications for each) were weighed in a 2-mL Eppendorf tube. One milliliter of solvent (0.1% formic acid in methanol) was added to dissolve the samples and vortexed for 1 min then placed in a sonicator waterbath for 30 min. Samples were then centrifuged for 5 min at 10,000 rpm on a benchtop centrifuge. About 700 μL of the supernatant was pipetted into high performance liquid chromatography (HPLC) vials and metal caps with rubber septa secured with a crimper.
LC–MS analysis.
The supernatant was injected into the LC–MS instrument from Bremen, Germany. Metabolites were separated using a gradient of H2O with 0.1% formic acid (solvent A) and acetonitrile (solvent B), using a Dionex UltiMate 3000 ultra performance liquid chromatography (UPLC) at a flow rate 0.3 mL·min−1 on a Waters BEH C18, 2.1 × 100-mm column (Ireland). Mass spectrometry was obtained on a TOF Bruker Impact II (Germany) using electron spray ionization running in positive mode, which scans between 50 and 1600 m/z, with nebulizer at 1.8 bar and dry gas as 8 L/min.
Liquid chromatography–MS data were analyzed using the Compass data analysis tool version 4.3.110 and converted into buckets after peak integration and Pareto scaling. Principal component analysis was performed using the Profile Analysis software version 2.1 (Bruker®, Bremen, Germany) on buckets imported from the data analysis tool on a minimum of three samples for each tea. Peaks were identified according to actual mass, MS/MS, and retention time. Accurate mass and MS/MS spectral data were compared with the Kyoto Encyclopedia of Genes and Genomes, Chemical Entities of Biological Interest, Chemspider, and Pubchem online databases.
The LC–QTOF–MS data were preprocessed using XCMS for relative quantitation and statistical analysis, and were subjected to parametric analysis of variance (ANOVA). The XCMS uses sample class information during preprocessing to help decide which groups of peaks are significant. The unsupervised PCA and the supervised partial least squares–discriminant analysis, the most well-known classification procedures in chemometrics, were carried out using MetaboAnalyst. The quality of the model was described by R2X and Q2 values.
Statistical analysis.
Data were subjected to one-way ANOVA, carried out using STATISTICA (StaSoft, Inc., Tulsa, OK) package (2011). Mean separation was carried out using Tukey’s honest significant difference at P < 0.05.
Results and Discussion
Information on the influence of pruning on accumulation of primary and secondary metabolites in herbal teas is limited to allow a balanced discussion of some of the variables collected. The present study is critical in tea industry because periodic harvesting of shoots generally imposes significant perturbation of carbon-based secondary metabolites, which are important in general tea plant growth and development. Metabolites accumulate more when plants were subjected to stresses for adaptation (Ramakrishna and Ravishankar, 2011).
Physiological measurements.
Basal-pruned bush tea plants exhibited low (0.2 m2·m−2) LAI while unpruned and middle pruned showed high LAI (0.7 m2·m−2). Chlorophyll content significantly increased in the unpruned bush tea plants with 64.5 μmol·m−2 and was low (29.4 μmol·m−2) in the basal-pruned bush tea plants. Current findings are consistent with a study by Mahanta and Baruah (2006), who reported that unpruned black tea exhibited higher chlorophyll content than pruned black tea (C. sinensis). Moreover, the unpruned bush tea exhibited high gS (18.2 m2·s−1) whereas basal pruning resulted in low gS (9.1 m2·s−1) (Table 1).
Leaf area index (LAI), chlorophyll, and stomatal conductance (gs) of bush tea in response to different pruning types.
1H-NMR profiling data.
The untargeted metabolites profiling using 1H NMR were subjected to a supervised method, the OPSL-DA, which revealed distinct groupings between the basal, middle, top pruning, and the unpruned control (Fig. 1). The basal-pruned treatments (green) clustered to the bottom left of the ellipse whereas the top-pruned treatments (yellow) grouped to the top left. On the right side of the ellipse the middle-pruned treatments clustered to the bottom whereas the unpruned (control) grouped to the top right. The discriminating peaks in the bush tea samples were from the aromatic region 6–8.5 ppm. The unpruned bush tea plant showed more intense peaks in the phenolic region from 4 to 7 ppm chemical shifts compared with other pruning treatments. However, top-pruned bush tea had more intense peaks throughout the 1H NMR.
Orthogonal partial least square discriminatory analysis results of bush tea samples from nuclear magnetic resonance spectra, with ellipse representing Hotelling within 95% confidence.
Citation: HortScience horts 53, 7; 10.21273/HORTSCI13023-18
The unpruned bush tea plant exhibited more peaks in the aromatic region from 6 to 8.5 chemical shifts than the basal-pruned plants, which did not exhibit any peak under the aromatic region. Similar trends have been observed with the phenol region (4–7 ppm chemical shifts). The basal-pruned plants showed more intense peaks under the sugar region (chemical shifts) and aldehyde region (9.5–9.6 ppm). However, top-pruned bush tea plants exhibited more intense peaks throughout the 1H NMR spectra regions from 0.04 to 9.96 ppm compared with other pruning levels. The control or unpruned plants showed more peaks in the aromatic (7.3 ppm), alcohol and amine (0.5–5 ppm), and aniline (3–5 ppm) regions more than all the pruned bush tea plants.
The supervised comparison of bush tea pruned at different levels revealed a distinct grouping among the treatments (Fig. 2). The unpruned (green) treatment clustered to the middle left, whereas top-pruned (light blue) clustered to the top center. The basal-pruned (red) treatment clustered to the bottom center of the graph, whereas the middle-pruned (dark blue) was scattered through the middle center to the right side of the graph.
Partial least squares–discriminant analysis from liquid chromatography linked to mass spectometry data of samples from 40 A. phylicoides tea samples.
Citation: HortScience horts 53, 7; 10.21273/HORTSCI13023-18
R2X is the proportion of variance in the data explained by the model and it designates the goodness of fit test which in the study was 0.126. R2Y value of the model was 0.198.
The dendrogram presented in Fig. 3 demonstrated biochemical differences in bush tea with respect to pruning treatments. There were two main clusters with the first group represented by basal and unpruned treatments. In the fourth samples, eight were basal and five were unpruned dominated in the first cluster was dominated, whereas top pruning was four and only one middle pruned sampled grouped with the latter. In the second cluster, there were four main subclusters with middle-pruned treatment dominating (60%), whereas the unpruned treatment had 50%. However, the top-pruned bush tea samples grouped together in the fourth subcluster denote large proportion of chemical compositions.
Dendrogram for the clustering of A. phylicoides tea pruning treatments.
Citation: HortScience horts 53, 7; 10.21273/HORTSCI13023-18
There were significant differences among the pruning treatments regarding compound concentration or metabolite accumulation (Tables 2 and 3). Tables 2 and 3, denote the exact masses and concentration based on LC–MS peak areas and variation of compounds attributed to pruning techniques. This aspect has not been investigated in commercial cultivation of bush tea in spite of the fact that these compounds have been reported to have health attributes (Malongane et al., 2017). Most found peak area responses occurred in the top-pruned bush tea and exhibited 22 out of 28 compounds. Compounds included coumarin, 6-cis-docosenamide, 5-p-coumaroylquinic acid, (6′Z,8′E,10E,10′E,12E,12′E,14 + E,14′E)-3,4,4′,5′-tetradehydro-5′,9-dihydro-betabetacaroten-9-ol, integrastatin A, 1,3-dicaffeoylquinic acid, magnoshinin, okanin, (2S)-5-hydroxy-7-methoxy-6,8-dimethylflavanone, (9Z,12Z,15Z)-octadecatrienoic acid, 2″-deamino-2″-hydroxy-6″-dehydroparomamine, O-butanoylcarnitine, myricitrin, gorlic acid, tetracenomycin X, D-tryptophan, linoleamide, L-beta-phenylalanine, pheophorbide A, stearidonic acid, gibberellin A14 aldehyde, N-(2-hydroxyheptadecanoyl)-1-O-beta-D-glucosyl-15-methylhexadecasphing-4-enine. In addition, middle-pruning yielded 14 out of 28 compounds and they were chlorogenate, coumarin, naringenin 7-O-beta-D-glucoside, luteolin 7-O-(6-O-malonyl-beta-D-glucoside), okanin, 2″-deamino-2″-hydroxy-6″-dehydroparomamine, O-butanoylcarnitine, myricitrin, gorlic acid, D-tryptophan, linoleamide, L-beta-phenylalanine, pheophorbide A, N-(2-hydroxyheptadecanoyl)-1-O-beta-D-glucosyl-15-methylhexadecasphing-4-enine. Basal-pruned bush tea plants exhibited 17 out of 28 compounds and they were chlorogenate, 6-cis-docosenamide, integrastatin A, magnoshinin, okanin, 2″-deamino-2″-hydroxy-6″-dehydroparomamine, O-butanoylcarnitine, myricitrin, gorlic acid, sakuranin, D-tryptophan, linoleamide, laricitrin 7-monoglucoside, L-beta-phenylalanine, pheophorbide A, gibberellin A14 aldehyde, N-(2-hydroxyheptadecanoyl)-1-O-beta-D-glucosyl-15-methylhexadecasphing-4-enine. Similar compounds were evident in the research reported by Malongane et al. (2018) of bush tea metabolites using untargeted approach by LC–MC. However, the compounds exhibited by the unpruned bush tea plants were 15 out of 28 and they consistently appeared in the other pruned treatments.
Response of concentrations of identified bush tea compounds to different pruning types (LC–QTOF–MS).
Response of concentrations of identified bush tea compounds to different pruning types (LC–QTOF–MS).
In this study, different pruning heights significantly enhanced accumulation of secondary metabolites that are not only important in plant stress physiology for adaptation but also important as quality indicators of herbal teas. It was observed from the spectra that control (unpruned) plants exhibited more secondary metabolites than the other plants that were middle pruned or basal pruned, with basal pruning exhibiting much less metabolites. Consistent with a finding by Maudu et al. (2010) which demonstrated that basal pruning was not viable in potted bush tea growing under 50% shade nets as the treatment revealed highly reduced total phenolics. Although unpruned bush tea exhibited higher chlorophyll content and higher gS (Table 1), the concentrations of the secondary metabolites were not enhanced (Tables 2 and 3). This could be a consequence of pruning, possibly due to concentrations of various secondary plant products accumulated when the plants were subjected to stress (Ramakrishna and Ravishankar, 2011). Hence, secondary metabolites were improved in the pruned than the unpruned bush tea plants. Our results are consistent with a study by Yilmaz et al. (2004) who reported that pruning enhances tea quality. The top pruning (apically pruned) showed intensive peaks throughout the spectra compared with other pruning levels. Conversely, Maudu et al. (2010) reported that top-pruned plants exhibited the lowest total polyphenols whereas the highest total polyphenol concentration was observed in unpruned plants. However, the improved metabolites accumulation by top pruning (apically pruned) bush tea in this study could be attributed to the fact that cutting of the apical bud exposed the bush tea plants to severe stress, hence, the top-pruned plant exhibited more metabolites with improved concentrations (Tables 2 and 3). This study revealed 28 metabolites (compounds) that explained significant differences in pruning treatments of bush tea.
In conclusion, pruning bush tea showed a significant effect on accumulation of secondary metabolites and thus could enhance bush tea quality. The study successfully annotated 28 metabolites (compounds) that elucidated canonical differences in pruning treatments of bush tea, as validated through multivariate analysis. Top pruning (apically pruned) resulted in improved metabolites than the control (no pruning) and can be recommended in bush tea cultivation.
Literature Cited
Bandara, S.N. 2012 Agronomy of irrigated tea in low elevation growing areas of Sri Lanka. Univ. Adelaide School Agr., Food Wine, Doctoral Diss
Chabeli, P.M., Mudau, F.N., Mashela, P.W. & Soundy, P. 2008 Effects of nitrogen, phosphorus and potassium nutrition on seasonal tannin content of bush tea (Athrixia phyliciodes DC.) S. Afr. J. Plant Soil 25 9 83
Haukioja, E., Ossipov, V., Koricheva, J., Tuija Honkanen, T., Larsson, S. & Lempa, K. 1998 Biosynthetic origin of carbon-based secondary compounds: Cause of variable responses of woody plants to fertilization Chemoecology 8 133 139
Hlahla, L.N., Mudau, F.N. & Mariga, I.K. 2010 Effect of fermentation temperature and time on the chemical composition of bush tea (Athrixia phylicoides DC.) J. Med. Plants Res. 4 824 829
Maedza, K.V. 2015 Effects of micronutrients on growth and quality of bush tea (Athrixia phylicoides DC.). Univ. South Africa, Johannesburg, South Africa, MSc Diss
Mahanta, P.K. & Baruah, S. 2006 Changes in pigments and phenolics and their relationship with black tea quality J. Sci. Food Agr. 59 21 26
Malongane, F., MacGaw, L.J. & Mudau, F.N. 2017 The synergetic potential of various teas, herbs and therapeutic drugs in health improvement: A review J. Sci. Food Agr. 97 4679 4689
Malongane, F., McGaw, L.J., Nyoni, H. & Mudau, F.N. 2018 Metabolic profiling of four South African herbal teas using high resolution liquid chromatography-mass spectrometry and nuclear magnetic resonance Food Chem. 257 90 100
Maree, J.E. & Viljoen, A.M. 2012 Phytochemical distinction between Pelargonium sidoides and Pelargonium reniforme—A quality control perspective S. Afr. J. Bot. 82 83 91
Maudu, M.E., Mudau, F.N. & Mariga, I.K. 2010 The effect of pruning on growth and chemical composition of cultivated bush tea (Athrixia phylicoides D.C) J. Med. Plants Res. 4 2353 2358
McGaw, L.J., Steenkamp, V. & Eloff, J.N. 2007 Evaluation of Athrixia bush tea for cytotoxicity, antioxidant activity, caffeine content and presence of pyrrolizidine alkaloids J. Ethnopharmacol. 110 16 22
Mediani, A., Abas, F., Ping, T.C., Khatib, A. & Lajis, N.H. 2012 Influence of growth stage and season on the antioxidant constituents of Cosmos caudatus Plant Foods Hum. Nutr. 67 344 350
Mphangwe, N.I.K. 2012 Lung pruning: A review of practice. Tea Research Foundation of Central Africa (TRFCA) News. pp 18–23
Njogu, R.E.N., Kariuki, D.K., Kamau, D.M. & Wachira, F.N. 2014 Effects of foliar fertilizer application on quality of tea (Camellia sinensis) grown in the Kenyan highlands Amer. J. Plant Sci. 5 2707 2715
Padayachee, K. 2011 The phytochemistry and biological activities of Athrixia phylicoides. Univ. Witwatersrand, Johannesburg, South Africa, MSc Diss
Ramakrishna, A. & Ravishankar, G.A. 2011 Influence of abiotic stress signals on secondary metabolites in plants Plant Signal. Behav. 6 1720 1731
Venkatesh, P., Jaiprakash, M., Prasad, P., Pillai, B., Sahale, P.P. & Sinkare, V.P. 2007 Flavonoid biosynthesis in tea (Camellia sinensis). Original Research Report [PDF]. 13 Mar. 2017. <http://www.teascience.org/pdf/chapter-2-6-2>
Yilmaz, G., Kandemir, N. & Kinalioglu, K. 2004 Effects of different pruning intervals on fresh shoot yield and some quality properties of tea (Camellia sinensis L. Kuntze) in Turkey Pak. J. Biol. Sci. 7 1208 1212
