Assessment of shade-unshade condition and subsequently pesticide treatment on first flush tea leaf metabolites through GC/MS based metabolomics approach

Abstract All around the world, tea is one of the most popularly consumed non-alcoholic beverages. Tea is a shade loving shrub. However, how the tea metabolites get changed under shade versus unshade condition and also under shade-pesticide treatment versus unshade-pesticide treatment condition have not been determined and reported yet. So, a metabolomics approach was applied to investigate the shade and pesticide treatment condition of the first flushes of Camellia sinensis L. A GC/MS-based metabolomics approach was established encompassing a homogeneous experimental setup for growth, treatment and sampling of tea leaves and their subsequent data analysis using statistical and chemometric tool. A total of 90 metabolites have been identified and semi-quantified. The chemometric analysis elucidates the changes in important metabolites under shade and unshade conditions and also the effect of thiamethoxam on metabolite profile and their responses on tea bushes from shade and unshade conditions. The antioxidant activity and total antioxidant capacity were also assessed and showed a significant change in activity after pesticide treatment both under shade–unshade conditions. Increase in phenols, amino acids, organic acids, sugar alcohols concentrations when plants exposed to shade than unshade treatment was reported. The results suggest that during pesticide treatment on shaded tea plants there is decrease in many important metabolites (phenolics and organic acids), which impart the major flavor, taste, aroma, and quality of the beverage. The increase in amino acids cope the plants to protect them from chemical pesticide treatments under shade conditions.


PUBLIC INTEREST STATEMENT
Tea is the most popularly consumed beverage worldwide. Tea was introduced in India by the British as the common man's drink. Tea industry is one of the oldest industries in India and the teas of India are well appreciated for their unique aromas, flavors, and biological activities. It is well-known that tea has great antioxidant potentiality. The goal of this study is to provide an insight into the changes in metabolic network and antioxidant capacities of first flush tea samples collected from tea garden of North Bengal University campus, exposed to different conditions, namely, shade-unshade (control), shade-unshade (thiamethoxam treatments). Gas chromatography/mass spectrometry-based metabolomic-chemometric approaches identified the distinct changes in chemical composition when exposed to shade than in unshade treatment. Pesticide (Thiamethoxam 25% WG) treatment on shade tea samples showed a decrease in metabolites like important phenolics and organic acids which impart the major flavour, taste, aroma, and superiority in quality of tea and increase in amino acids is also notable.

Introduction
Tea [Camellia sinensis (L.) Kuntze] is the most consumed beverage around the world, principally produced and processed from the top two fresh green leaves and a bud. It contains a wide array of important metabolites (such as tannins, alkaloids,terpenoids, flavonoids and polyphenols), which impart its taste, aroma, flavor, and quality.
Tea is a shade loving, perennial, evergreen shrub. This plant is usually grown as a monocrop, under the covering of shade trees. Various metabolism of tea plants are facilitated by shade conditions, which may influence its vigor, superiority in quality, and yield of the beverage crop (Beddage & Mohotti, 2005). As tea is a characteristic subtropical plant, warm, humid atmosphere, acid soil and shade are required for the growth of tea bushes. An ideal shade formed by the trees is a single-covered foliage canopy, which cuts out a large amount of the infra-red rays but allows adequate visible light to infiltrate through the tea bush and does not compete with the tea for nutrients, nourishments, and soil moisture. Some species of Albizia such as Albizia chinensis (Osbeck) Merr., Albizia procera(Roxb.) Benth., Albizia odoratissima(L.f)Benth., Albizia lebbeck(L.) Benth, Acacia lenticularisBuch.-Ham. ex Benth, Derris robusta(Roxb. ex DC.) Benth, Dalbergia sericeaG. Don, Indigofera teysmanii, Cassia siameaLam., Melia azedarachL., Dalbergia sissooRoxb., Dalbergia assamicaBenth., Acacia auriculiformisA.Cunn. ex Benth. and many others are some common shade trees for shading tea bushes (Barua, 2007).
Planting of shade trees has all along been associated with tea cultivation. Shade trees simulate forest conditions. Plants at high altitude are subjected to enhanced oxidative stress due to high UV influence (resulting from air rarefaction, i.e. reduction in density of air) compared to places at low altitude (Balakrishnan et al., 2005;Lesser, 1996). For survival, plants have developed specific cell defensive metabolites, namely, the flavonoids. Flavonoid derivatives (viz. flavonoid glycosides) may also serve as precursor of tea aroma producing compounds and are thus relevant in studies on plant quality under varying UV influence (Bhattacharya & Sen-Mandi, 2011).
Conversely, an important characteristic of shade is its influence on the development of diseases and pests. Mites are normally adverse to shade, but blister blight and black rot diseases increase well in shade. Red spider and other tea mites thrive well under unshade condition (Das, 1959;Banerjee, 1979). Mite damage is more rigorous on unshade rather than shade tea.
So, control of the tea pest is a major challenge in the management of tea plantations. 1,034 arthropods have been identified in tea plantations, around the world (Ye et al., 2014). Throughout the major tea-producing regions, almost 3% of arthropods are common pests, for instance, Ectropis obliqua, Empoa scaonukii, Euproctis pseudoconspersa and so on. Outbreaks of these pests account for serious yield losses and decline in tea quality. Different pest control methods are employed globally, such as chemical pesticides, sex pheromones, entomo-pathogenic insect pesticides, and many others (Ye et al., 2014). The application of chemical pesticides is the prime method used by farmers to control tea pests. However, extreme uses of chemical pesticides to control pests have serious unpleasant effect to the environment, resulting in ecological harm and reduced biodiversity. Pesticide residues also show negative impact regarding food safety and security and ultimately resulting in various health issues (Chen et al., 2017).
In the current scientific scenario, the development of metabolomics has provided a new measurement in the subject of multi-dimensional biology for the comprehensive study of the global metabolic networks (Allwood et al., 2008;Dunn, 2008;Krishnan, 2005;Spratlin et al., 2009;Vinayavekhin et al., 2010). Metabolomics has been defined as the comprehensive qualitative and quantitative profiling of a large number of metabolites of a biological system (Fiehn et al., 2000;Trethewey et al., 1999). The major advantage of metabolomics is the simultaneous monitoring of metabolic networks in a way that enables the association of changes in such networks with biotic and/or abiotic causal agents and the detection of corresponding markers.
Tea plant metabolites are altered through the influence of environmental factors such as insect attacks (Kessler & Baldwin, 2002;Paré & Tumlinson, 1999), differential shading treatment (Yang et al., 2012), light quality (Fu et al., 2015), and temperature (Katsuno et al., 2014;Zeng et al., 2016;Zhou et al., 2017). Present study has been designed to evaluate the effect of shading and unshading treatment on the first flush of tea leaf metabolites. In addition to this, to determine the changes in the quality and quantity of metabolome composition under thiamethoxam (25% WG) treatment for the management of tea mosquito bug which causes major damage for first flushes, under shade versus unshade condition in comparison with the shade-pesticide-treated and unshade-pesticide-treated tea plants using GC/MS-based metabolomics technology.
Plant constituents form an important source of antioxidants that scavenge free radicals and prevent cell and tissue damage thereafter by terminating the free radical chain reactions (Mruthunjaya et al., 2016). The antioxidant properties of the samples grown under shade, unshade conditions as well as their pesticide-treated samples were also measured spectrophotometrically.

Collection of tea leaf samples
First flushes from Camellia sinensis (L.) Kuntze (cv. TV26) were collected from the tea garden of University of North Bengal campus. Four tea bushes were grown at the same tea garden of university campus (26.7095° N, 88.3542° E) with temperature ranging in an average maximum of 14.4°C. and an average minimum of 6.3°C during the month of March. Two bushes were selected under shade condition under Dalbergia sissoo and the other two bushes were selected under unshade condition. No chemical pesticide was applied to one set of bush under shade and unshade conditions and were selected as control group of shade plants and control group of unshade plants. The other sets of bush distant from the control group, grown under shade and unshade plants were treated with chemical pesticide (Thiamethoxam 25%WG) (0.250 g/100 mL in distilled water) at a permissible dosage as per Plant Protection Code (July, 2019, Ver. 11.0). The soil type, fertilization schedule, and other tea plantation management methods were kept same among the plants. The plucking of two leaves and one bud from each bush was confined during the first flush in the month of March, 2019.

Experimental design
Maintaining the Tea Board of India guidelines, the tea was plucked (two leaves and one bud) from non-pesticide-treated shade, non-pesticide-treated unshade, pesticide-treated shade and pesticide-treated unshade plant bushes.

Sample preparation for GC-MS analysis
After harvesting the first flush, the samples were immediately ground to a fine powder with liquid nitrogen. The freeze ground samples of approximately 100 ± 10 mg were taken in 2 mL eppendorff tubes (x4) and extracted with MeOH (HPLC) and H 2 O (HPLC) in a ratio of 1:1 (v/v) at 65°C. for 30 min in a boiling water bath. 20 µL of 0.2 mg/mL solution of ribitol (adonitol) was added as internal standard. After extraction, MeOH (HPLC) and H 2 O (HPLC) were again added in 1:1 (v/v) ratio, and the extracted sample was cold-centrifuged (REMI C-24) at 14,000 rpm for 20 min. The aliquots were distributed in eppendorff tubes (100 µL x4) and evaporated to dryness. The residue obtained was re-suspended in 5 µLof MOX (20 mg/mL in pyridine) and subsequently agitated for 90 min at 30°C. Thereafter, 45 µL of MSTFA was added and incubated at 37° C with frequent agitation for 30 min for trimethylsilylation to increase the volatility of metabolites. 1 µL of FAME markers (a mixture of IRI markers) were added [prepared using FAMEs of C 8 -C 26 linear chain length dissolved in Chloroform (HPLC)]. GC/MS analysis (Agilent 7890 A GC equipped with 5795 C inert MSD with Triple Axis Detector) was carried out following the method of Kind et al. (2009) with some modifications (Das et al., 2016). Prior to analysis, the samples were preserved at 4ºC. for 10 min to maintain equilibration and sedimentation of particles.

Temperature program for GC/MS analysis
For separation and detection of analytes, DB-5 MS capillary column (Agilent J&W GCcolumns, USA) (30 m × 0.25 mm × 0.25 μm) was used. Injection was made in sandwich mode with fast plunger speed without viscosity delay or dwell time. The analysis was performed under the following oven temperature program: oven ramp 60°C (1-min hold) to 325°C at 10°C/min, 10-min hold before cooling down, 37.5 min run time. The injection temperature was set at 250°C, the MS transferline at 290°C, and the ion source at 230°C. Helium was used as the carrier gas at a constant flow rate of 2.5 mL/min (carrier linear velocity 57.95 cm/s). Samples (1 μL) were injected via the split mode (split ratio 1:5) onto the GC column. Prior to analysis, the method was calibrated with the FAME standards available on the Fiehn GC-MS Metabolomics library (2008) (Agilent ChemStation, Agilent Technologies Inc., Wilmington, USA) following users' guide. AMDIS was used to deconvolute GC-MS results and to identify chromatographic peaks. Auto-tuning of MSD was done at least once a week. All samples were measured in a randomized fashion.
The metabolites were identified by comparing the RT, RI of the metabolites and also by comparing their MS fragmentation patterns of the mass spectra with the entries of compound in Agilent Fiehn GC/MS Metabolomics library (2008) (Agilent Technologies INC., Wilmington, USA) using metabolite database-Automated Mass Spectral Deconvolution and Identification System (AMDIS) using Agilent RTL method. Retention time of some of the compounds were also compared with that of the standards for confirmation of the metabolites.

DPPH radical scavenging activity
Following the method described by Braca et al., (2001), the antioxidant activity of all the four extracts (unshade control, unshade pesticide treated, shade control and shade pesticide treated) were determined based on the scavenging of the stable DPPH free radical. 0.1 mL of aqueous methanolic leaf crude extracts was added to 3 mL of DPPH solution prepared in methanol in a concentration of 0.004%. After 30 min of incubation, the absorbance was measured at 517 nm wavelength and the percentage inhibition activity was calculated using the formula [(A 0-A e )/A 0 ] × 100, where A 0 = Absorbance without extract; A e = Absorbance with extract.

Superoxide (O 2 − ) radical scavenging activity
In the riboflavin-light -NBT (nitro blue tetrazolium) system, superoxide radical scavenging activity was measured with the tea leaf extracts. In a 3 mL volume of total reaction mixture, 50 mM phosphate buffer (pH 7.8), 13 mM methionine, 2 µm riboflavin, 100 µm EDTA, 75 µm NBT, and 1 mL of plant sample solution were added (Beauchamp & Fridovich, 1971;Dasgupta & De, 2004). After 10 min of illumination of a fluorescent lamp, the increase in the production of blue formazan was measured at 560 nm.

TAC
The reaction mixture contained 0.1 mL of tea sample solution and 1 mL of reagent mixtures (mixed phosphate buffer, sulphuric acid, and ammonium heptamolybdate in a ratio of 4:3:3) (Prieto et al., 1999). The reduction of Mo VI to Mo v by the extract and the formation of green phosphate/Mo v complex at acidic pH were assayed. The TAC was measured as equivalent to ascorbic acid.

Statistical analysis
All the antioxidant assays were performed thrice. From the replicates within the experimental assays, mathematical calculations like means and standard errors were performed by Microsoft Excel 2013. The data were statistically analyzed using t-test of Microsoft Excel 2013 (p-values < 0.05 considered significant).
Prior to MVA, the missing values were replaced with half of the detection limit (Fiehn, 2006) followed by normalization of the data. The RRRs of all the compounds identified were considered by normalizing the peak areas of the compounds by sample weight in gram and by peak area of the internal standard. The RRRs of metabolites were correlated with the control group of shade plants, control group of unshade plant, shade plant with pesticide treatment, and unshade plant with pesticide treatment. The RRRs were considered as the independent variables and the different treatment (shade, unshade and chemical pesticide-25% thiamethoxam WG) on plants were taken as the modeled responses.
The normalized data were then converted to comma delimited version (.csv) and then subjected to Metaboanalyst v.4.0. The purpose of this study was to determine the differences in metabolite profiling among a set of treatments [shade, unshade (control group) condition and shade, unshade plant treated with chemical pesticide -Thiamethoxam 25% WG (experimental group)] on first flush of Camellia sinensis cv. TV 26 leaves and also to identify the important metabolites responsible for such differences. The normalized data were subjected to ANOVA and a post-hoc test based on Tukey's HSD test was done to verify the significant variances among the different treatments and determined by p-values lower than 0.05. Based on the generalized logarithm transformed dataset, heatmap clustering of samples as well as metabolites was also developed utilizing squared Euclidean distance and ward linkage.

Metabolite profiling and chemometric study
A total of 90 metabolites have been identified (Table 1) including 14 amino acids, 13 organic acids, 10 fatty acids (2 MUFA and 1 PUFA), 1 inorganic acid, 20 sugar and sugar acids, 3 polyols, 18 Shade pesticide treated and unshade pesticide treated tea sample metabolite profiles also showed distinct groups between the two as observed in PLS-DA and OPLS DA 2-D scores plot in Figure 3.   The top 20 metabolites were detected as the VIP based on PLS-DA among these four sets of treatments on plant ( Figure 6). The top 20 metabolites identified for their separation were: epicatechin, D-glucose-6-phosphate, L-valine, L-glutamic acid, kaempferol, 2-isopropylmalic acid,  phloroglucinol, DL-3,4-dihydroxyphenylglycol, uracil, mucic acid, L-proline, galactinol, D-glucose, 4-isopropylbenzoic acid, D-saccharic acid, catechin, norvaline, piceatannol and alizarin. Differences of the amount of the detected key contributory metabolites (based on VIP scores) were measured as identified by PLS-DA as shown in Figure 6.
The top 10 metabolites identified from s-PLS-DA loadings (Figure 7) were: citric acid, epicatechin, epigallocatechin, kaempferol, L-proline, -(-) epicatechin, cellobiose, L-glutamic acid and D-mannitol. Heatmap (Figure 8) illustrating changes in the relative concentrations of the metabolites between different treatments encoded using a color scale. The two-dimensional hierarchical cluster analyses combined with heat map reveals trends between the different treatments on tea plants and the metabolites (variables).
Overall changes in concentration of the important metabolites during different treatments on tea leaf samples are shown in Figure 9 as developed from VIP scores of PLS-DA ( Figure 6).

Discussion
In this study, a dynamic change in the concentration of metabolites was determined due to shade versus unshade condition on tea plants (Figure 9). Gas chromatography coupled with MS is a powerful tool and a fast and accurate method to determine hundreds of metabolites including sugars, organic acids, polyols, diverse phenolic and cyclic compounds following solvent extraction and derivatization (Rohloff, 2015). Overall, all the phenolic compounds showed increase in response to shade than unshade condition, namely, epicatechin ( Tea plants are adapted to shade conditions under natural habitat, so sensitivity to strong light is predictable. Chemical composition analyses in this study indicated notable increase in phenols, flavonoids, amino acids, and some sugars under shade condition of tea bushes than plants grown under unshade condition, that is, under sunlight-exposed condition. The phenolic compounds, namely, gallic acid, quinic acid, pyrogallol, chlorogenic acid, catechin, epicatechin, and epigallocatechin are derived from the shikimic acid pathway, located upstream of the phenylpropanoid pathway in plants (Hermann, 1995). Whereas when the plants are exposed to direct sunlight, UV radiation cause cell destruction and the UV stress may result in the synthesis of cell-protective compounds (flavonoids, phenol, amino acids) in tea leaves but in lesser concentration than the shade-treated plants.
The luminous intensity of shade and unshade conditions could not be measured as light reaching different parts of tea bushes are different and it also depends on the canopy formed by different shade trees at different seasons and the shade trees are deciduous in most of the cases. But the knowledge of shading effects on tea shoots is important for the improvement of tea quality and productivity. Leaf thickness, leaf mass area, leaf density of new leaves were smaller under shade culture than under open or unshade culture. Also chlorophyll, epicatechin and epigallocatechin are strongly related with shading effects (Sano et al., 2018). From the above study, it can be said that the shading cultivation may be beneficial for high quality and high productivity in tea bushes in tea gardens.
A change in DPPH radical scavenging activity among the unshade control versus unshade pesticide treated samples and shade control versus shade pesticide treated were found significantly (p-value < 0.05) different by t-test (p-values showing 0.000259 and 0.000235) and also in case of superoxide radical scavenging activity (p-values showing 0.0197 and 0.002) among types of samples of plant extracts were noticeable. No statistically significant change was observed in their total antioxidant capacity.

Conclusion
This study provides an insight into the change in metabolite network of first flush leaf samples of Camellia sinensis cv. TV 26 for the first time, whose plants have been exposed to shade, unshade, shade pesticide and unshade pesticide treatments. Using GC/MS based metabolomic and chemometric approaches the changes in metabolite level due to different treatments on tea plants have shown that there is increase in phenols, amino acids, organic acids, sugar alcohol concentrations when plants are exposed to shade treatment than unshade treatment. On the other hand during pesticide treatment on shade tea plants, there is decrease in many important metabolites especially phenolics and organic acids, which impart the major flavor, taste, aroma and quality of the beverage, but it is interesting enough to note the increase in amino acids, which cope the plants to protect them from chemical pesticide treatments under shade conditions.
There is very limited information on antioxidant capacities of shade, unshade, shade pesticide treated, and unshade pestcide treated first flush tea leaf samples. Therefore, from this study, it can be concluded that shade treatment may be beneficial for the tea cultivation than unshade condition as many important metabolites are increasing due to shade conditions and in some cases antioxidant activities are also found higher in shade treated plant samples than unshade condition. However, more studies are required to be done. In this study, we have highlighted not only the changes in metabolites but their antioxidant capacities using multiple treatment procedures on first flush tea leaf samples. Authors' contribution SR and SD together conceived and designed the research project; SR and AG conducted the treatment of tea leaves with pesticide, shade and unshade conditions, collected and prepared the samples and supervised the whole experiment. JC and SD prepared and analyzed the GC/MSbased research. SD analyzed the data and interpreted results chemometrically and wrote the manuscript. All authors read and approved on the final version of the manuscript.