Introduction

Polygonum minus Huds is a common medicinal herb grown as a native plant in the Southeast Asian countries. In Malaysia, it is locally known as kesum or daun laksa. It has been widely used as a natural flavoring agent due to the significant amount of aldehydes that contributes to its flavor (Baharum et al. 2010). Various antioxidant, anti-inflammatory, antimicrobial, antiulcer and analgesic properties of P. minus had been reported, which were contributed by its high level of flavonoid and phenolic compounds (Suhailah et al. 2012; Christapher et al. 2015). A study has reported that P. minus showed anti-proliferative activity in several cancer cell lines (Mohd Ghazali et al. 2014). Polygonumins A, a novel compound that was isolated from the stem of P. minus was also reported to be able to inhibit cancer cells without harming the normal cells (Ahmad et al. 2018). In addition, P. minus also exhibit notable ROS scavenging activities (Maizura et al. (2011), which make it as one of natural medicinal resources that has been traditionally used by the locals to treat digestive problems, dandruffs, and body ache (Christapher et al. 2015). P. minus was also reported to be able to improve cognitive function and exhibit neuroprotective effect (You et al. 2018). Most recently, P. minus was reported to be able to reduce anorexia and improves the anhedonia among stressed mice (Bashir et al. 2022a). It was also reported that aqueous extract of P. minus exhibited antidepressant effects, decreased immobility time, increased spatial memory, reduced corticosterone, increased the brain-derived neurotrophic factor level, and reduced monomamine oxidase- A level, coupled with increasing monoamines (serotonin and norepinephrine) in the hippocampus of a chronic ultra-mild stress-induced mice model (Bashir et al. (2022b).

Various studies have reported on the use of different light intensities to influence both the morphological and physiological properties of plants. Shading was reported to be able to increase the leaf mass per unit area and leaf thickness due to the elongation of palisade cells in Acer pseudoplatanus L. and the mature leaves may re-acclimate anatomically to increase irradiance (Wyka et al. 2022). A study reported that shading results in higher synthesis of essential oils in lemon balm, mint and sweet basil (Ilić et al. 2022). The same study also reported that shading significantly increased the antioxidant activities of basil leaves (Ilić et al. 2022). In a different study, Ghasemzadeh et al. (2010) reported that light can stimulate the production of secondary metabolites such as gingerol and zingiberine in Zingiber officinale. This result is in contrast with the study by Ibrahim et al. (2014), where low light intensity along with high CO2 level were found to increase the secondary metabolites content in Labisia pumila. This indicates that the effects of irradiance levels vary with different plant species. In the study of P. minus, intermediate shading onto the plant increases its pigment contents, improve photosynthesis rate, anthocyanin and the phenolic compounds (Mohd Yusof et al. 2021).

Although various reports can be found on the chemical profiles and antioxidative effects of P. minus (Baharum et al. 2010; Suhailah et al. 2012), very few studies had been conducted to assess the benefits of micro-environmental manipulation (such as the effects of shading) on the production of secondary metabolites and antioxidant activity in P. minus and subjected to different treatment durations. By incorporating improvements and another variable from the previous study by Mohd Yusof et al. (2021), this study will be able to yield valuable knowledge on the role of shading levels and durations on the production of secondary metabolites and antioxidant activities of Polygonum minus Huds.

Materials and Methods

Experimental Design

This study was designed based on the outcomes obtained from our previous study (Mohd Yusof et al. 2021), where we had reported that shading levels (T1 = 0%, T2 = 50% and T3 = 70%) positively affected the growth, physiological properties, and the production of secondary metabolites of P. minus. The current study was carried out to further improve the research parameter, where the plants were grown under different shade levels (T1 = 0%, T2 = 30%, T3 = 50% and T4 = 70%) and subjected to two treatment durations (10 weeks and 20 weeks). The effects of these treatments on plant growth, physiological properties and production of bioactive secondary metabolites were thoroughly and carefully monitored throughout the duration of the treatments. Under each treatment, the light intensities were also continuously monitored by using a LI-COR LI-250A light meter, 3 times a day throughout the 12 h daytime (7 a.m. to 7 p.m.) at the location where the study was carried out. The average light intensity (in µmol m−2 s−1) recorded under T1 (0% shading; control) was 160.254 ± 8.770a, T2 (30% shading) recorded 50.922b ± 3.603b, T3 (50% shading) recorded 39.423 ± 2.781bc and T4 (70% shading) recorded 23.328 ± 2.302c (the different letters on top indicate significant difference at p ≤ 0.05 according to Tukey’s test). Instead of looking at the single factor of different shade levels and treatment durations, the interaction or relationship effects between the different shade levels and treatment durations exerted on the plants were also studied. Figure 1 shows the set up of the study.

Fig. 1
figure 1

The glasshouse and the wall-to-wall installation of the black nylon netting, forming horizontal shading above the plants in each treatment row

Full size image

Media Preparation and Treatment Details

This study was carried out in a greenhouse at Rimba Ilmu Botanical Garden, Universiti Malaya. P. minus plants were propagated for two weeks in small pots before being transplanted into small sized polyethylene bags filled with a soilless mixture of burnt rice husk and coco peat (ratio 3:1), following the same procedures reported in Mohd Yusof et al. (2021). The shade levels were prepared in rows, where the shade levels were created by black nylon netting canvas with different mesh percentages. The canvas was installed from wall to wall of the greenhouse, forming horizontal shading above the plants in each row. Inner walls were also installed around the shaded compartments, to make sure plants were shaded from all directions. The light intensities under each shading levels were continuously monitored using a LI-COR LI-250A light. The plants were arranged in RCBD with 4 shade levels × 2 treatment durations × 4 blocks × 4 replicates.

Fertilization and Plant Maintenance

All the plants were watered every morning and fertilized by using NPK green once per week and organic fertilizer twice per week. In order to prevent the attack from fungus and mold, chalk powder was scattered around the plots area (Mohd Yusof et al. 2021).

Data Collection

Plant Growth

Plant height (PH) and number of leaves (LN) were measured every two days until the 25th day after transplant. Plant height measurements were taken on the main stem starting from the reference point on the soil surface towards the tip of the stem by using a measuring tape with an accuracy of ± 1 mm. The number of total leaves were recorded per plant.

Biomass and Root–Shoot Ratio

The plants were harvested and used to obtain the biomass by dividing the plant parts into three different categories which were the leaves, stems and roots. All three parts were subjected to the process of freeze-drying using Labconco freeze dryer (Labconco Corporation, MO 64132 USA) at − 50 °C. The dry weight of the plants was recorded as the biomass dry weight; leaf dry weight (LDW), stem dry weight (SDW) and root dry weight (RDW). Root-shoot ratio was calculated by dividing root dry weight with the above ground dry weight.

Leaf Gas Exchange

The measurements of photosynthesis rate (A), intercellular CO2 (Ci), leaf temperature (Tleaves), transpiration rate (E) and stomatal conductance (Gs) were obtained by a closed infra-red gas analyser LICOR 6400 Portable Photosynthesis System (IRGA, Licor Inc., Lincoln, NE, USA). The measurements used optimal condition set of 400 µmol mol−1 CO2 flux, 30 °C standard cuvette temperature and PAR of 1,200 µmol m−2 s−1. Gas exchange measurements were carried out between 10:00 a.m. to 11:00 a.m., on fully expanded young leaves numbered 3 from the plant apex. Data were recorded in triplicates from each treatment within each block.

In-situ Chlorophyll Content

In-situ chlorophyll content (SPAD) was assessed using the Chlorophyll Meter SPAD-502 (Minolta, Japan) once every three days on a mature expanded leaf of each plant. The equipment was calibrated prior to the measurements. The data were recorded in triplicates from each treatment within each block.

Chlorophyll Fluorescence

Measurements of chlorophyll fluorescence were taken from fully expanded second leaves by using Handy-PEA (Hansatech, Kings Lynn, UK). Leaves were darkened for 15 min by attaching light-exclusion clips to the central region of the leaf surface. Measurement of F0 (initial fluorescence), Fm (maximum fluorescence) and Fv (variable fluorescence) were obtained from this procedure. The use of minimal (F0) and maximal (Fm) fluorescence leads to the well accepted expression for the maximum quantum yield of primary photochemistry (TR0/ABS) equals to 1 − (F0/Fm) = Fv/Fm.

Leaf Chlorophyll and Carotenoid Analysis

The chlorophyll a (Chl a), chlorophyll b (Chl b) and carotenoid (Car) were analysed from mature expanded leaves. Leaf samples (0.1 g) from each treatment from each block were crushed in liquid nitrogen by using a mortar and pestle. The sample was ground in 10 mL absolute methanol using a chilled mortar and pestle before incubated under -20 °C in the dark for 24 h. The chlorophyll and carotenoid levels were measured in triplicates with a spectrophotometer model UV160U at 665.2, 652.4, and 470 nm wavelengths after tenfold dilution. The pigment concentrations of chlorophyll a, chlorophyll b and carotenoid were calculated following the methods by Lichtenthaler and Buschmann (2001) and expressed as mg chlorophyll g−1 tissue fresh weight (mg g−1 FW).

$${\text{Chl a }}\left( {{{\mu {\text{g}}} \mathord{\left/ {\vphantom {{\mu {\text{g}}} {{\text{ml}}}}} \right. \kern-0pt} {{\text{ml}}}}} \right):{ 16}.{72 }\left( {{\text{A665}}.{2}} \right) - {9}.{16 }\left( {{\text{A652}}.{4}} \right)$$
$${\text{Chl b }}\left( {\mu {\text{g}}/{\text{ml}}} \right):{ 34}.0{9 }\left( {{\text{A652}}.{4}} \right) - {15}.{28}\left( {{\text{A665}}.{2}} \right)$$
$${\text{Car}}\left( {\mu {\text{g}}/{\text{ml}}} \right): (1000\left( {{\text{A}}470} \right) - {{\left( {1.63 \left( {\text{Chl a}} \right) - 104.96 \left( {\text{Chl b}} \right)} \right)}/221} $$

Anthocyanin Content

The total anthocyanin content (TAC) was determined using the pH differential method. The methanolic extracts were diluted separately with two types of buffers: potassium chloride (0.025 M) at pH 1.0 and sodium acetate (0.4 M) at pH 4.5 using the ratio 1:4 (1-part test portion and 4 parts buffer). The absorbance of the samples was measured at 510 and 700 nm. The TAC was calculated using the following formula:

$${\text{Total anthocyanin content }}\left( {{\text{mg cyd 3}} - {\text{glu}}/{\text{L}}} \right) = {{\left( {{\text{Ab }} \times {\text{MW }} \times df \, \times {1}000} \right)} \mathord{\left/ {\vphantom {{\left( {{\text{Ab }} \times {\text{MW }} \times df \, \times {1}000} \right)} {\left( {\varepsilon \, \times L \, \times \, m \, } \right)}}} \right. \kern-0pt} {\left( {\varepsilon \, \times L \, \times \, m \, } \right)}}$$

where: Ab = (A510–A700) pH1.0 -(A510–A700) pH4.5, MW = Molecular weight of cyd 3 − glu (449.2 g/mol), df = the dilution factor, ε = Extraction coefficient (29,600 mol/g).

Total Phenolics and Total Flavonoids Quantification

Folin-Ciocalteu reagent (10%) was used to determine the total phenolics content of the leaf samples. Approximately 0.1 mL of the sample extract was mixed with Folin–Ciocalteau reagent (0.75 mL). Then, 2% aqueous sodium carbonate (0.75 mL) was added into a test tube incubated in the dark for 45 min. Blanks were prepared by the addition of 70% methanol instead of the sample. The absorbance was measured at 765 nm. The TPC was then determined from the standard calibration curve prepared with a series of gallic acid standards (0.01, 0.02, 0.03, 0.04, 0.05, and 0.06 mg/mL). The results were represented as mg gallic acid equivalent per 100 g dry weight (mg GAE/100 g dry sample).

For flavonoid quantification, the sample (0.5 mL) was mixed with 1.5 mL of 70% methanol in a test tube covered with aluminum foil, and then was left for 5 min. Next, 10% 0.10 mL of AlCl3 (AlCl3.6H2O) was added followed by addition of 1 M NaOH (0.1 mL) and 2.8 mL of distilled water. Later, the absorbance was measured at 415 nm using a spectrophotometer after 40 min of incubation with quercetin as the standard and the results were expressed as mg g−1 dry sample.

DPPH Radical Scavenging Assay

The DPPH free radical scavenging activity of each sample was determined according to the standard protocol. 50 µL of extract at six different concentrations (0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 mg/mL) was added to 150 µL of DPPH solution (60 mM) in each well of a 96-well plate. The change in absorbance at 515 nm was measured after 30 min incubation at room temperature. The obtained data were then used to determine the concentration of the sample required to scavenge 50% of the DPPH free radicals (IC50). The percentage of inhibition was plotted against the concentration and the IC50 was obtained from the fitted linear curve. A lower IC50 denotes a more potent antioxidant.

Ferric Reducing Antioxidant Power (FRAP)

FRAP assay was performed based on the method described by Benzie and Strain (1999) with slight modifications. About 300 μL of extract or methanol (control) was mixed with 10 μL FRAP reagent and was incubated in microplate wells at room temperature in the dark for 30 min. The absorbance was recorded at 593 nm. A series of stock solution at 100, 10, 1, 0.1 and 0.01 mM of aqueous ascorbic acid were prepared to generate a standard curve (r2 = 0.9944). The results obtained were expressed as mM of ascorbic acid equivalent per gram of dried samples.

Carbohydrates Estimation

The carbohydrates were estimated following an earlier described method (Sharma et al. 2022). Briefly, 100 mg of variously treated fresh samples were extracted overnight with methanol. The supernatant isolated from the extract was used for carbohydrate estimation. The reaction mixture was prepared by adding 5% aqueous phenol and concentrated H2SO4 to the supernatant, followed by incubation, and boiling. The carbohydrates were quantified by UV–Visible absorbance measured at 490 nm.

Activity of Antioxidant Enzymes

The sample (100 mg) was extracted with a chilled extraction buffer constituted of 100 mM potassium phosphate pH 7.0 and 0.1 mM EDTA to estimate the enzyme activity of superoxide dismutase (SOD) (Sharma et al. 2022). The reaction was initiated by adding 50 mM phosphate buffer (pH 7.8), 13 mM methionine, 0.1 μM EDTA, 75 μM nitroblue tetrazolium, and 2 μM riboflavin to 50 μl of the supernatant acting as crude enzyme extract. The reaction mix was illuminated with 20-W fluorescent tubes and the absorbance was recorded at 560 nm.

100 mg sample was also extracted in 50 mM phosphate buffer (pH 7.8) to evaluate the enzyme activity of ascorbate peroxidase (APX) (Sharma et al. 2022). The reaction was initiated by adding the crude extract to a reaction mix comprising 50 mM phosphate buffer, 0.2 mM EDTA, 0.5 mM ascorbic acid, and 0.25 mM hydrogen peroxide. The absorbance was recorded at the start and end of the reaction at 290 nm. The molar extinction coefficient used for the calculation was 2.8 mM−1 cm−1.

Determination of Lipid Peroxidation

Lipid peroxidation was estimated in terms of malondialdehyde (MDA) equivalent accumulation in the samples. The sample was homogenized in 0.1% trichloroacetic acid (TCA) followed by centrifugation. To the supernatant, 20% TCA containing 0.5% thiobarbituric acid was added, heated for 30 min at 95 °C, and cooled quickly. After centrifugation, the absorbance of the supernatant was recorded at 532 nm and 600 nm. Malondialdehyde (MDA) was quantified using extinction coefficient of 155 mM−1 cm−1 (Sharma et al. 2022).

Ascorbic Acid Estimation

The dried sample (100 mg) was crushed into 6% trichloro acetic acid and extracted. In the reaction mixture, 2% 2,4-dinitrophenylhydrazine and 10% thiourea was added. The reaction mixture was incubated at 37 °C for 15 min. After addition of 80% sulphuric acid to the reaction mixture, its absorbance was read at 530 nm (Panwar et al. 2016).

Hydrogen Peroxide Determination

100 mg of sample was crushed in 5 mL of 0.1% trichloro acetic acid and centrifuged to yield the extract. To the 0.5 mL of supernatant, 0.5 mL of 10 mM postassium phosphate buffer and 1 mL KI was added and incubated in the dark for 1 h. The absorbance of the reaction mix was measured at 390 nm using UV–Visible spectrophotometer (Wang and Jiao 2000).

HPLC Quantification of Flavonoids

A HPLC analysis was carried out to determine the amounts of flavonoid under different irradiance, following the methods by Qader et al. (2012). The phytochemical analysis was conducted using the high-performance liquid chromatography (HPLC) method, using an LC-20AD apparatus (Shimadzu Corp, Kyoto, Japan).

The experiments were carried out using a reversed phase C18 Hypersil GOLD column (250 mm × 4.6 mm; Thermo Fisher Scientific). The samples were eluted with a stepwise elution of the mobile phase. The mobile phase consisted of solvent A [H2O + 0.1% trifluoroacetic acid (TFA)] and solvent B (methanol + 0.1% TFA), at a unique flow rate of 1 mL/min. The gradient elution program was as follows: 0-min 90% A, 10% B; 5 min 55% A, 45% B; 20 min 50% A, 50% B; 21 min 40% A, 60% B and 22 min 90% A, 10% B. The sample injection volume was 20 μL followed by equilibration run of 15 min and was monitored with an SPD-20 A (UV) at 280 nm.

Statistical Analysis

Data obtained from the study was analyzed using two-way analysis of variance (two-way ANOVA) and means separation tests between treatments were performed using Duncan multiple range test (DMRT) for the significant interaction effect, while Tukey’s post hoc test was used to determine the significance of single effect. The standard error of differences between means was calculated with the assumption that data were normally distributed and equally replicated. The relationships between parameters were analyzed by using Pearson’s correlation. A multivariate analysis was also done to determine two principle relationships: the relationship between the growth performance and physiological components of the plant and the relationship between the antioxidant activities with the secondary metabolites content. These relationships were determined by using the statistical method of Partial Least Squares Regression (PLSR) using XLSTAT version 2021.1 (Addinsoft Inc., Paris, France). The importance of the physiological components and the secondary metabolites content were expressed by Variable Importance in Projection (VIP) scores, where a VIP score of more than 0.8 (90% confidence interval; CI) was rated as important, whereas a VIP score of more than 1 (95% CI) was deemed as the most important. The said relationships were shown by using a PLSR correlation plot and the summary of the two-way ANOVA table with shading level and treatment durations as the main factors is also provided.

Results

Plant Growth

Plant height (PH) and number of leaves (LN) of P. minus were both significantly influenced by the single effect of shade levels. There was no significant interaction observed in PH and LN between the shade levels and treatment durations (Supplementary Table S1). After the 29th day, the highest PH was obtained under T1 (29.384 cm ± 2.381) followed by T3 (22.797 cm ± 2.411) > T2 (21.394 cm ± 2.103) and the lowest plant height was found under T4 (15.969 cm ± 3.221) (Fig. 2a). Similarly, T1 recorded the highest LN compared to other treatments (Fig. 2b). This confirms that high light intensity resulted in high growth of leaves. LN under T3 and T4 were observed to constantly decrease from day 11 until the 29th day. LN was also strongly correlated with PH (r = 0.674, p ≤ 0.01). This shows that the increase in PH will also increase the number of leaves.

Fig. 2
figure 2

a Effect of shade levels on plant height; b effect of shade levels on number of leaves. [T1 = 0% Shaded; T2 = 30% Shaded; T3 = 50% Shaded and T4 = 70% Shaded. Data are means of treatments; Small bars represent standard error]

Full size image

Biomass and Root–Shoot Ratio

Data analysis showed that there was no significant interaction between the treatment durations and different shade levels on plant biomass (Supplementary Table S1). However, shade levels were found to significantly influence the LDW, SDW and RDW (Fig. 3a). T1 was observed to yield the highest leaf dry weight (LDW), stem dry weight (SDW), root dry weight (RDW) and root-shoot ratio (RSR) compared to other treatments (Fig. 3a). In comparison, lower irradiance level (under 50% and 70% shade levels) showed low biomass dry weight than T1. Meanwhile, 30% shade level (T2) produced LDW of approximately 0.511 g, which was not significantly different compared to the control (0.595 g).

Fig. 3
figure 3

a Effects of different shade levels on leaf dry weight (LDW), shoot dry weight (SDW), root dry weight (RDW) and root-shoot ratio (RSR); b effects of treatment durations on leaf dry weight (LDW), shoot dry weight (SDW), root dry weight (RDW) and root-shoot ratio (RSR). [T1 = 0% Shaded; T2 = 30% Shaded; T3 = 50% Shaded and T4 = 70% Shaded. Data are means of treatments; Small bars represent standard error]

Full size image

Meanwhile, the SDW was observed to be significantly influenced by the single factor of treatment durations, but not LDW, RDW and RSR (Fig. 3b). Treatment duration of 20 weeks was observed to produce higher biomass (dry weight), with lower RSR compared to 10 weeks duration. Longer treatment duration was also found to increase the plant biomass under all treatments (Fig. 3b).

Leaf Gas Exchange, Leaf Pigments and Chlorophyll Fluorescence

The highest photosynthesis rate (A) was observed under T2 (30% shade level) (21.916 ± 0.419 µ mol CO2 m−2 s−1) followed by T3 > T4 = T1 (Table 1). Meanwhile, the lowest intercellular CO2 (Ci) was observed under 50% shade level (T3), followed by T2 > T4 > T1 (Table 1). Stomatal conductance (Gs) were also found to be significantly influenced by different shade levels (p < 0.01). The highest Gs (2.012 ± 0.033 mol H2O m–2 s–1) were recorded under T1 > T2 > T4 > T3.

Table 1 The effects of the different shade levels on the leaf gas exchange of P. minus
Full size table

Besides, the interaction between shade levels and treatment durations was found to significantly influence the transpiration rate (E), leaf resistance (RL) and leaf pigments content (p < 0.05) (Supplementary Table S1). The transpiration rate (E) was found to be the highest in T3 at 20 weeks, followed by T4 at 10 weeks > T3 at 10 weeks = T1 at 10 weeks > T2 at 10 weeks, while the lowest E value was obtained by T4 at 20 weeks (Fig. 4a). RL was found highest under T4 with treatment duration of 20 weeks, followed by T1 at 20 weeks > T2 at 20 weeks and the lowest was recorded by T4 with treatment duration of 10 weeks (Fig. 4a).

Fig. 4
figure 4

a Effects of different shade levels on transpiration rate (E) and leaf resistance (RL); b effects of different shade levels on the leaf pigments (Chl a, Chl b, Car, Chl a + b, Chl a/b ratio and SPAD value). [T1 = 0% Shaded; T2 = 30% Shaded; T3 = 50% Shaded and T4 = 70% Shaded. Data are means of treatments; Small bars represent standard error]

Full size image

The highest chlorophyll a (Chl a) (2.431 ± 0.224 mg g−1 FW), chlorophyll b (Chl b) (1.919 ± 0.156 mg g−1 FW) and total chlorophyll (total Chl) (4.349 ± 0.376 mg g−1 FW) were obtained under T4 at 20 weeks of treatment duration, followed by T3 at 10 weeks = T2 at 20 weeks, while the lowest was recorded by T1 with treatment duration of 20 weeks (Fig. 4b). Meanwhile, chlorophyll a/b (Chl a/b) ratio was found to be highest under T1 with treatment duration of 20 weeks, followed by T3 at 20 weeks = T4 at 20 weeks, while the lowest a/b ratio was recorded by T1 with treatment duration 10 weeks (Fig. 4b).

The highest carotenoid (Car) value was obtained under T3 after 20 weeks (1.225 ± 0.103 mg g−1 FW), parallel to the highest amount of chlorophyll obtained. The highest in-situ chlorophyll (SPAD) value was obtained under T3 with treatment duration of 20 weeks (35.717 ± 0.841), but the value was not significantly different compared to T4 with treatment duration of 20 weeks (35.042 ± 1.167) (Fig. 4b). These were parallel with the chlorophyll content that was found to be the highest under T4 under the same treatment duration. On the other hand, chlorophyll fluorescence shows non-significant difference on the interaction effect, neither the single effect (Supplementary Table S1). The results of chlorophyll fluorescence are presented in Table 2.

Table 2 The effects of the different shade levels and treatment durations on leaf fluorescence of P. minus
Full size table

Relationships Between Growth Performance and Physiological Components of Plant

The significant Pearson’s correlations between the plant growth and physiological parameters described in earlier paragraphs are summarized in Table 3. Moreover, a multivariate analysis (PLSR) was also conducted to evaluate the contributions or relationships between the physiological parameters of the plant (X variables) on plant growth performance (Y). As observed in the PLSR correlation plot in Fig. 5a, the chlorophyll a (Chl a), chlorophyll b (Chl b), total chlorophyll a + b (total a + b), in-situ chlorophyll content (SPAD), carotenoid (Car), photosynthesis rate (A), transpiration rate (E), maximal fluorescence (Fm) and maximum quantum yield of PSII (Fv/Fm) are all negatively loaded along Dimension 1. Meanwhile, the intercellular CO2 (Ci), chlorophyll a/b ratio (Chl a/b), relative humidity (RH), stomatal conductance (Gs), minimal fluorescence (Fo), and ratio of variable fluorescence (Fo/Fm) are positively loaded along Dimension 1 (Fig. 5a). Additionally, based on the computed VIP values, the pigment content (Chl a, Car, total a + b and SPAD), Gs and Ci recorded VIP scores of > 1.0, which indicate that these are the parameters that gave major contribution to the growth performance of P. minus. The VIPs values are illustrated in Fig. 5b.

Table 3 Significant Pearson’s correlation between parameters measured in this phase
Full size table
Fig. 5
figure 5

Partial Least Squares Regression (PLSR) analysis to determine the relationship between the growth performance and the physiological parameters of P. minus; a PLSR correlation plot; b VIPs score. [Small bars represent standard error, dotted line indicates threshold values of 0.8 and 1.0]

Full size image

Total Anthocyanin, Phenolic and Flavonoid Contents

The total anthocyanin content (TAC), total phenolic content (TPC) and total flavonoid content (TFC) of P. minus leaves were all significantly influenced by the interaction effect of shade levels and treatment durations (Supplementary Table S2). For plants that were harvested after 10 weeks, T3 recorded the highest TAC (306.592 ± 7.13 μg g−1 DW) followed by T1 (287.221 ± 6.92 μg g−1 DW) and the lowest was obtained by T4 (203.822 ± 6.59 μg g−1 DW) (Fig. 6a). However, when the plants were exposed with longer treatment duration of 20 weeks, T2 was observed to yield the highest TAC (360.695 ± 18.07 μg g−1 DW) followed by T4 > T3.

Fig. 6
figure 6

Effects of different shade levels on the a total anthocyanin; b total phenolic and; c total flavonoid contents of P. minus. [T1 = 0% Shaded; T2 = 30% Shaded; T3 = 50% Shaded and T4 = 70% Shaded. Data are means of treatments; Small bars represent standard error]

Full size image

In terms of the total phenolic content, T1 was observed to yield the highest TPC (395.491 ± 10.37 mg g−1 DW) after 10 weeks of treatment, followed by T3 > T4 > T2. Meanwhile, for plants harvested after 20 weeks, the highest TPC was recorded under T2, with values of 643.013 ± 25.64 mg g−1 DW (Fig. 6b), similar to that recorded for TAC (Fig. 6a). On the other hand, T4 recorded the lowest TPC after 20 weeks of treatment duration (248.163 ± 19.83 mg g−1 DW) (Fig. 6b). Overall, the TPC in all treatments was observed to increase with treatment duration (Fig. 6b). The same trend was observed for TFC, where the TFC of all treatments was observed to also increase with treatment duration (Fig. 6c). After 10 weeks of treatment, the highest TFC was observed in T1 (201.904 ± 1.42 mg g−1 DW), followed by T3 > T4 > T2. These observations were similar to that obtained in TPC after 10 weeks of treatment (Fig. 6b). Meanwhile, the TFC of T2 and T4 at 10 weeks were observed to be not significantly different. In addition, after 20 weeks of treatment, T1 was also observed to produce the highest TFC (387.606 ± 17.78 mg g−1 DW), followed by T2 > T4 > T3. Nevertheless, the TFC of T1, T2 and T4 was not significantly different.

Antioxidant Assays

The highest FRAP reducing power was recorded under T4 of 10 weeks (7.145 ± 0.01 mg g−1 DE) and under treatment duration of 20 weeks (7.628 ± 0.44 mg g−1 DE). T4 also showed excellent DPPH radical scavenging activities at both treatment durations, where T4 leaf extract subjected to 10 weeks of treatment needed only 1.873 mg mL−1 of the extract to inhibit 50% of the DPPH radicals, and 2.507 mg mL−1 for T4 leaf extracts subjected to 20 weeks treatment. These results are shown in Table 4.

Table 4 The effects of the different shade levels on antioxidant activities of P. minus extract
Full size table

A correlation analysis was also conducted to determine the relationships between the secondary metabolites and antioxidant activities observed (Table 5). Significantly strong correlations were observed between Chl a with Chl b (r = 0.862, p < 0.01), Car (r = 0.982, p < 0.01) and between Chl b with Car (r = 0.923, p < 0.01). Meanwhile, significantly moderate correlations were observed between these three pigments with TAC (Table 5), indicating that the increase in Chl a, Chl b and Car pigments will lead to an increase in TAC. Similarly, a moderate but significant correlation was also observed between TFC and TPC (r = 0.465, p < 0.01). Nevertheless, only weak but significant correlations were observed between FRAP with Chl a (r = 0.196; p < 0.05) as well as between DPPH with TPC (r = -0.240, p < 0.01) and TFC (r = -0.230, p < 0.01).

Table 5 Significant Pearson’s correlation between secondary metabolites and antioxidant activities
Full size table

Carbohydrates Estimation

Total carbohydrate content (TCC) was recorded to be significantly affected by the interaction of different shade levels and treatment durations (Supplementary Table S2). The highest TCC was recorded under shade level of T2 of 20 weeks (49.712 ± 0.611 mg g−1 DW) followed by T4 of 20 weeks (41.882 ± 0.015 mg g−1 DW) which are not significantly different with the total carbohydrate content recorded under T2 of 10 weeks (41.346 ± 0.415 mg g−1 DW) and the lowest was recorded under T3 of 20 weeks (28.327 ± 0.158 mg g−1 DW). Under treatment duration of 10 weeks, the TCC recorded under 0% shade levels or T1 (31.634 ± 0.5 mg g−1 DW) was not significantly different compared to the data obtained under 70% shade levels (T4) (34.458 ± 0.151 mg g−1 DW). These data are represented in Fig. 7.

Fig. 7
figure 7

Effects of different shade levels on the total carbohydrate content of P. minus. [Data are means of treatments; Control, T1 = 0% Shading; T2 = 30% Shading; T3 = 50% Shading; and T4 = 70% Shading. Means with different letters on top of each bar are significantly different at p ≤ 0.05]

Full size image

Activity of Antioxidant Enzymes

The SOD and APX enzyme activities were all recorded to be significantly affected by the interaction effect of shade levels and treatment durations (Supplementary Table S2). Under treatment durations of 10 weeks, the SOD obtained between all treatments were not significantly different with each other, where the highest reading obtained under T1 (35.586 ± 0.433 units mg−1 DW) and the lowest was obtained under T4 (33.634 ± 0.173 units mg−1 DW). However, under the treatment durations of 20 weeks, the highest SOD enzyme activity was recorded under T1 (38.288 ± 0.433 units mg−1 DW) but the value was not significantly different with the one recorded under T4 (36.036 ± 1.040 units mg−1 DW). The lowest SOD enzyme activity was recorded under T2 (33.633 ± 0.347 units mg−1 DW). These results are shown in Fig. 8a.

Fig. 8
figure 8

Effects of different shade levels on a superoxide dismutase (SOD); b ascorbate peroxidase (APX) enzyme activities of P. minus. [Data are means of treatments; Control, T1 = 0% Shading; T2 = 30% Shading; T3 = 50% Shading; and T4 = 70% Shading. Means with different letters on top of each bar are significantly different at p ≤ 0.05]

Full size image

Moreover, significant interaction effects between the shade levels and treatment duration were also observed in the ascorbate peroxidase (APX) enzyme activity (Supplementary Table S2), where under the treatment duration of 10 weeks, the same trend as observed in the SOD enzyme activity was obtained. The highest APX enzyme activity was recorded under T1 (1.277 ± 0.145 nmole min−1 mg−1 DW) followed by T2 > T3 and the lowest was recorded under T4 (0.461 ± 0.247 nmole min−1 mg−1 DW). However, when the treatment duration was prolonged to 20 weeks, the highest enzyme activity was recorded under T4 (1.631 ± 0.034 nmole min−1 mg−1 DW) followed by T1 > T3 and the lowest was observed under T2 (0.725 ± 0.183 nmole min−1 mg−1 DW). The results of APX enzyme activities are shown in Fig. 8b.

Ascorbic Acid Estimation

The ascorbic acid (AsA) content was significantly affected by the interaction of shade levels and treatment durations (Supplementary Table S2). The highest AsA content was observed under T4 of treatment duration of 10 weeks (19.984 ± 0.212 µg mg−1 DE), followed by T2 (17.224 ± 0.069 µg mg−1 DE) and the lowest was recorded under T1 (7.904 ± 0.055 µg mg−1 DE). Meanwhile, under treatment duration of 20 weeks, the highest AsA was recorded under T3 (17.808 ± 0.055 µg mg−1 DE) and the lowest was recorded under T4 (12.592 ± 0.055 µg mg−1 DE), completely in contrast to when the plant was exposed to treatment duration of 10 weeks, however the values were not significantly different with the AsA recorded under T1 (12.736 ± 0.018 µg mg−1 DE). The effects of the treatments and treatment durations on the AsA contents were shown in Fig. 9.

Fig. 9
figure 9

Effects of different shade levels on the ascorbate acid content activity of P. minus

Full size image

Hydrogen Peroxide and Lipid Peroxidation Determination

The concentration of hydrogen peroxide (H2O2) in P. minus was found to be significantly affected by the interaction effect of shade levels and treatment durations (Supplementary Table S2). Under the treatment duration of 10 weeks, the highest H2O2 concentration was recorded under T1 (30.959 ± 0.118 µmol mg−1 DE) followed by T3 (30.745 ± 0.133 µmol mg−1 DE), T2 (23.882 ± 0.059 µmol mg−1 DE) and the lowest concentration was found under T4 (19.933 ± 0.015 µmol mg−1 DE). These results contrasted with the values of APX enzyme activity recorded under the treatment duration of 10 weeks, where the highest was recorded under T1. Under the treatment duration of 20 weeks, the highest concentration of H2O2 was recorded under T4 (38.215 ± 0.163 µmol mg−1 DE) but the value was not significantly different than the one recorded under T1 (37.908 ± 0.104 µmol mg−1 DE). The lowest concentration of H2O2 recorded under the treatment duration of 20 weeks was under T2 (24.574 ± 0.163 µmol mg−1 DE). The effects of the treatments and treatment durations on H2O2 concentration in P. minus are recorded in Fig. 10. P. minus was also subjected to the analysis of malondialdehyde concentration (MDA) to study the effects of shade levels and treatment duration on the lipid peroxidation of P. minus. However, the values recorded for MDA levels were too low to be detected.

Fig. 10
figure 10

Effects of different shade levels on the H2O2 concentration in P. minus

Full size image

Relationships Between Antioxidant Activities and Secondary Metabolites Content

The relationships or contributions of the secondary metabolites content and antioxidant enzymes (TAC, TPC, TFC, TCC, AsA, APX and SOD as the X variables) on the antioxidant activities (Y) of the extracts were elucidated using PLSR. Figure 11a shows that the TAC, TPC, TFC, SOD and APX are all negatively loaded along Dimension 1, while the intercellular TCC and AsA are positively loaded along Dimension 1. Meanwhile, based on the computed VIP values, the antioxidant enzymes (APX and SOD) and AsA recorded VIP scores of > 1.0, which indicated that these three variables were the most important factors that influenced the antioxidant activities of the plants. The VIPs values are illustrated in Fig. 11b.

Fig. 11
figure 11

Partial least squares regression (PLSR) analysis to determine the relationship between the secondary metabolites content and antioxidant activity of P. minus; a PLSR correlation plot, b VIPs score. [Small bars represent standard error, dotted line indicates threshold values of 0.8 and 1.0]

Full size image

HPLC Quantification of Individual Flavonoids

The extracts from different plant parts (leaf, stem and roots) were also subjected to HPLC analysis to screen and quantify the amount of selected flavonoids. The leaf extracts were observed to contain kaempherol, kaempherol 3-glucoside, myricetin, quercetin and rutin, with myricetin and rutin found in a significant amount (Table 6). However, the amounts of flavonoids in the leaf extracts were observed to reduce after a longer treatment duration (20 weeks). The content of kaempherol 3-glucoside in the leaf samples were also observed to diminish after 20 weeks of treatment duration (Table 6). This contrasts with the observation found on the stem extracts, where synthesis of kaempherol 3-glucoside was only observed at 20 weeks of treatment duration and was not detected at 10 weeks of treatment. Not only that, the stem extracts were also observed to contain very little amounts of flavonoids, in contrast to the amounts recorded in the leaf extracts (Table 7). Similar observations were also recorded in the root extracts, where root extracts obtained from plants harvested after 10 weeks of treatment mostly consists of kaempherol 3-glucoside (Table 8). However, its synthesis was found to be stopped at longer treatment duration (20 weeks). Besides that, root extracts were also observed to be lacking in quercetin and rutin compounds, at both treatment durations (10 and 20 weeks). Two-way ANOVA analysis showed that quercetin production in leaves and stem of P. minus is significantly influenced by the interaction effect of shading levels and treatment durations (Supplementary Table S2). Meanwhile, myricetin is recorded to be significantly influenced by the interaction effect of the two factors in root of P. minus (Supplementary Table S2).

Table 6 The effects of the different shade levels on individual flavonoid composition of P. minus leaf extracts
Full size table
Table 7 The effects of the different shade levels on individual flavonoid composition of P. minus stem extracts
Full size table
Table 8 The effects of the different shade levels on individual flavonoid composition of P. minus root extracts
Full size table

Discussion

The data obtained in this study was analyzed for the interaction effects between the shading levels and the treatment durations exposed onto the plants. It was reported that both PH and LN were significantly influenced by single factor shade levels. Both PH and LV have the highest value under T1 and the lowest under T4, after 20 days of exposure. This result is in contrast with the one observed in the study of Bhatt et al. (2002) on 17 tropical grasses and legumes where the plant height was high under low light condition, at maximum of 50% shading. The growth of P. minus was also observed to reduce at low irradiance level (under 50% and 70% shade levels) compared to under T1, which indicates that low light intensity indeed limits the growth of P. minus. Under 30% shade level (T2), the LDW recorded was not significantly different compared to the LDW recorded under control or 0% shade level (T1). In line with this observation, the study on Castana dentata seedlings also showed that 32% light intensity yielded high weight of leaves and were not significantly different with 100% light intensity (Wang et al. 2006). This was different to the one reported in the study on Mitragyna speciosa Korth. where it was shown that low light conditions would most likely maximize plant biomass and total leaf alkaloid production (Zhang et al. 2022). Analysis showed that LDW was positively correlated with SDW (r = 0.546, p ≤ 0.01). Besides that, LDW also showed significant correlation with A (r = 0.371, p ≤ 0.01), Ci (r = 0.371, p ≤ 0.01), E (r = 0.405, p ≤ 0.01) and Gs (r = 0.416, p ≤ 0.01). The association between plant yield and photosynthetic performance have long been studied especially under controlled growth conditions, as light conditions under natural environments are often unpredictable and changed rapidly. However, plant’s ability to respond and adapt to the differences in light intensities are also highly demonstrated in many studies and some proved that the ability to increase photosynthetic performance has significant potential in increasing plant yield (Lawson et al. 2012).

Significant interaction effects between shading levels and treatment durations were observed in E, RL, leaf pigment contant, i.e. Chl a, Chl b and Car, TAC, TPC and TFC. The intercellular CO2 concentration of the leaves is a critical indicator in photosynthesis as the mesophyll cells of leaves consume CO2 during photosynthesis (Tominaga et al. 2018). This is parallel to the significant correlation shown between photosynthesis rate and internal CO2 concentration (r = 0.946, p < 0.01) in Table 3. Higher CO2 assimilation rates and stomatal conductance under lower irradiance level had been reported in a study by Medina et al. (2002). Correlation between Gs and A was significantly high (r = 0.826, p < 0.01) (Table 3), whereby the high Gs also indicated the increase of photosynthesis rate. Meanwhile, the leaf temperatures (Tleaves) were not significantly influenced by shade levels in this experiment. The transpiration rate (E) that was found highest under 70% shade levels and being exposed for 20 weeks indicate the importance of shade levels and its significant influence on transpiration rate, where it will aid in reducing the transpiration rate as the plants became adapted to the growth conditions and also due to maturing leaves. Transpiration rate showed a strong correlation with Gs (r = 0.927, p < 0.01) (Table 3) due to the opening and closing of stomata which would directly influence the transpiration rate and stomatal conductance, where both were greatly depending on leaf characteristics and features (Miyashita et al. 2005), along with differences in mechanical and performance characteristics that may change throughout the treatment duration (Franks and Farquhar 2007). Meanwhile, low but significant correlation was recorded between RL and E (r = 0.305, p ≤ 0.01) and between RL and Gs (r = 0.269, p ≤ 0.01) (Table 3). This shows that the increase in transpiration rate should progressively lower the RL recorded. RL was also influenced by the leaf characteristics and features, and with transpiration rate and stomatal conductance.

Low chl a/b ratio was attributed to the increase in chlorophyll b proportion compared to chlorophyll a and usually obtained under shaded environment, to enhance the efficiency of blue light absorption (Beneragama and Goto 2010). The lowest a/b ratio obtained under T1 of treatment duration 10 weeks may be due to the reduction in chlorophyll b due to interconversion of chlorophyll a and chlorophyll b which is believed to facilitate the regulation of the chlorophyll composition in a fluctuating light environment and throughout developmental stages (Sato et al. 2015). A strong positive correlation was observed between Chl a and Chl b (r = 0.862, p ≤ 0.01) which showed that the increase of Chl a and Chl b occurred at a similar particular time and the decrease of Chl a would also reduce Chl b pigment. Car showed a strong correlation with Chl a (r = 0.982, p ≤ 0.01) and Chl b (r = 0.923, p ≤ 0.01), which proved that the carotenoid composition of P. minus was influenced by the production of chlorophylls in this study.

According to Walters (2005), during the determination of changes in the constituents of the photosynthetic apparatus at a cellular level, it may be more useful to express photosynthesis and chloroplast components relative to chlorophyll, therefore it could reflect the balance in the investment between the capture and utilization of light in photosynthesis process. Degradation of chlorophyll b under high light condition was also reported in a study by Sato et al. (2015). Plants may develop acclimation over time to cope with different irradiance levels in its anatomical, morphological, physiological and biochemical aspects (Zervoudakis et al. 2012). According to Walters (2005), acclimation acts as a homeostatic mechanism, where it is able to reverse the changes from environmental challenges and therefore, maintaining efficient photosynthesis. The present study shows that P. minus may have adapted to the low irradiance condition with the increase of treatment duration by developing more photosynthetic pigments in order to utilize more sunlight for photosynthesis process. Again, in understanding acclimation of plants, Walters (2005) also stated that plants with strong acclimation are more likely to succeed in variable conditions, under both exposed and shaded environments. The present study suggests that for 10 weeks of treatment exposure (or treatment duration), 50% shade level was the most optimal condition in the leaf gas exchange traits and the production of chlorophyll in P. minus plants. In contrast, 70% was the most optimum shade level for P. minus plants exposed to 20 weeks of treatment.

Chlorophyll fluorescence showed non-significant difference on the interaction effect neither the single effect (Supplementary Table S1). However, the chlorophyll fluorescence showed significantly high correlations with A, Ci, E and Gs with r values of more than 0.600, p < 0.01 (Table 3). These correlations are common, as chlorophyll fluorescence would indeed provide the information on the efficiency of the phytochemistry process and the process of heat dissipation happened in the plants (Murchie and Lawson 2013).

PLSR analysis showed that pigments content in the leaves is one of the major parameters that impacted the growth development of P. minus positively (Fig. 5b). This means that the higher the content of the chlorophyll and carotenoid pigments in the leaves, the higher the growth performance of the plants. Stomatal conductance was also seen to highly impact the growth of P. minus with a VIP score value of > 1.0. Many studies have proven the indirect affect of stomatal conductance activities that will affect the growth or yield of plant. A study showed that the blocking of stomatal conductance had caused the reduction of yield in cotton plant by 28% (Zia-Khan et al. 2014). This was probably due to the direct effect of stomatal conductance and the photosynthesis rate of plant (Inoue et al. 2021), which was supported by the Pearson’s correlation results with r of 0.826, p < 0.01 (Table 3).

Overall, the TPC in all treatments was observed to increase with treatment duration (Fig. 6b). The same trend was observed for TFC, where the TFC of all treatments was observed to also increase with treatment duration (Fig. 6c). The results of TPC, TFC and TAC suggested that the shading level of 30% (T2) for 20 weeks was the best treatment to produce high amounts of TPC, TFC and TAC in P. minus.

However, the results of FRAP reducing potential of the T1 extracts after 10 weeks of treatment were in direct contrast to that observed in its TPC and TFC contents, where it exhibited the highest TPC and TFC contents after 10 weeks of treatment but showed lower FRAP reducing potential than other treatments. Nevertheless, the high TPC and TFC contents in T1 (after 10 weeks of treatment) was observed to result in the highest scavenging activity against DPPH radicals (Table 4). On the other hand, T4 resulted in high FRAP reducing potential and DPPH scavenging activities after 10 and 20 weeks of treatment duration, parallel to its high contents of Chl a, Chl b and Car, which suggested that the high antioxidant properties of the extracts were also influenced by the plant photosynthetic pigments, not just due to its phenolic contents, as has been reported by Mohd Yusof et al. (2021). Significantly strong correlations were observed between Chl a with Chl b (r = 0.862, p < 0.01), Car (r = 0.982, p < 0.01) and between Chl b with Car (r = 0.923, p < 0.01). These observations were due to the fact that these three pigments work hand in hand in capturing sunlight for photosynthesis. Meanwhile, significantly moderate correlations were observed between Chl a, Chl b and Car with TAC (Table 5), indicating that the increase in Chl a, Chl b and Car pigments will lead to an increase in TAC. Similarly, a moderate but significant correlation was also observed between TFC and TPC (r = 0.465, p < 0.01). Nevertheless, only weak but significant correlations were observed between FRAP with Chl a (r = 0.196; p < 0.05) as well as between DPPH with TPC (r = -0.240, p < 0.01) and TFC (r = -0.230, p < 0.01). These suggests that there were other compounds that were present in the extracts that could also contributed to its high antioxidant properties. A study by Nam et al. (2012) has demonstrated that there is an abundance of antioxidant compounds that can present in a plant, while Wei and Shibamoto (2007) demonstrated significant antioxidant activities from essential oil in plants such as aldehydes and carboxylic acid.

Plants provide one of the biggest sources of carbohydrates, and it is one of the elicitors of plant defenses involved in plant immunity (Trouvelot et al. 2014). The sugar content in plants is heavily affected by the photosynthesis fixation of CO2 by the leaves, which in turn is affected by other leaf gas exchange processes such as the opening and closing of stomata (Krasavina et al. 2014). In the study, it was observed that the highest carbohydrate content (TCC) recorded was under T2, for both duration of 10 and 20 weeks (Fig. 7). This result is parallel to the photosynthesis rate which was observed to be the highest under T2 of 30% shade levels. The relationship between photosynthetic activity and carbohydrate content in plants can also be seen in a study on Chrysanthemum morifolium Ramat., which shows that lighting conditions affect the epinastic movement of chrysanthemum leaves, thus increasing the light absorption area which in turn increased the content of carbohydrates due to the higher photosynthetic activity in the plant (Yang et al. 2022). A study on Rhynchelytrum repens also showed that natural irradiance increased the content of soluble carbohydrates compared to the plants exposed to low irradiance (Souza et al. 2004). High TCC observed under T2 of 20 weeks was also similar with the trend exhibited by TPC, which was found to be the highest under T2 of 20 weeks (Fig. 6b). It is said that carbohydrates are the basic compounds for the production of phenolic compounds via the shikimic acid pathway, where extra carbohydrates derived from glycolysis and the penthose-phosphate pathway are converted into aromatic amino acids (Mattson et al. 2005).

The increase of H2O2 concentration was reported to happen when plants are exposed to biotic and/or abiotic stresses, such as pathogen attack, wounding, UV irradiation, exposure to intense light, drought, salinity, and/or chilling (Sofo et al. 2015). The concentration of H2O2 along with other ROS are a good marker of the oxidative stress occurrence (Sofo et al. 2015). The highest concentration of H2O2 was recorded under both low (T4) and high irradiance level (T1) when the planting period of P. minus was prolonged until 20 weeks. Meanwhile, under the treatment duration of 10 weeks, the lowest concentration of H2O2 was recorded under low irradiance level of 70% shade level (T4). This result is parallel to the antioxidant enzyme activities recorded under 10 weeks of treatment duration, where the lowest APX and SOD enzyme activities were the observed under T4. The APX response had been reported to involve directly in protecting the plants from environmental damages and it plays a key role in catalyzing the conversion of H2O2 into H2O (Caverzan et al. 2012). High APX enzyme activities were also recorded under T4 of 20 weeks (Fig. 8b), parallel to the high concentration of H2O2 recorded. Sofo et al. (2015) also stated in his study that the balance of APX enzyme activity is crucial for the suppression of toxic H2O2 levels in plants. This result is in contrast with the one recorded on the purple pak choi seedlings where the antioxidant activities were the highest under low light stress at 5 days, but when the study was prolonged to 15 days, the antioxidant activities of the seedlings decreased significantly (Zhu et al. 2017). Meanwhile, the superoxide dismutase (SOD) enzyme activities were recorded to be the highest under no shading environment (T1) when exposed for a longer treatment duration of 20 weeks (Fig. 8a). SOD activity is said to be the first line of defense in plants as it functions to catalyze the dismutation of superoxide radicals into O2 and H2O2 to stop cell injury and the dysfunctional of plant tissues and therefore reducing the oxidative stress in plants, as high light condition may induce stress, resulting in high production of ROS. In addition, PLSR analysis showed that there were positive relationships between APX and SOD with H2O2 (Fig. 11a) which indicated that these enzymes were rapidly employed to scavenge against ROS (H2O2) when P. minus plants were under stress.

Ascorbic acid (AsA) synthesis in plants is also one of the response indicators of abiotic stresses and pathogen infections, where it plays multiple roles in metabolism of plants (Ishikawa et al. 2006). This plant-derived antioxidant is also one of the major sources of Vitamin C in human’s diet. It is also an important chemical that is involved in a wide range of crucial function in plants from antioxidant defense system and photosynthesis to growth regulation (Smirnoff 1996). AsA synthesis has also been studied in tomato in response to different lighting conditions (Zushi et al. 2020). This study showed that low light intensity under planting duration of 10 weeks can enhance the production of AsA in P. minus, in contrast to the result obtained by Lado et al. (2015) which showed that light avoidance would reduce and inhibit the accumulation of AsA in the peels of Citrus fruits. A study on basil plant also showed that high light intensity applied before harvest managed to increase the total ascorbic acid content in the plant (Larsen et al. 2022). Nevertheless, different results were obtained when the treatment duration was prolonged to 20 weeks, where under intermediate light intensity (50% shading), the AsA produced was found to be the highest. A study on the Fragaria x ananassa or strawberry shows that the harvest time of the plant will influence its antioxidant capacity, which appeared to be more important than genotype factor (Pincemail et al. 2012). However, a study by Kevers et al. (2011) on apples and pears showed that harvest time (at normal ripeness) had only limited impact, but the year to year variations were significant. These studies showed that the effect of light conditions and harvest time exposed onto the plants will influence the production of AsA.

The extracts from different plant organs (leaf, stem and roots) were also subjected to HPLC analysis to screen and quantify the amount of selected flavonoids. The leaf extracts were observed to contain kaempherol, kaempherol 3-glucoside, myricetin, quercetin and rutin, with myricetin and rutin found in a significant amounts (Table 6). Similar results were also observed in a study by Tzanova et al. (2018) where the largest amount of rutin was found in Betonica bulgarica. The study also suggested that specific growth conditions would significantly affect the flavonoid content in different parts of plants. Meanwhile, high level of myricetin was found together with low level of kaempherol. A study has reported that high level of kaempherol which produced upstream in the metabolic pathway may prevent the accumulation of myricetin (Carvalho et al. 2010). However, the amounts of flavonoids in the leaf extracts were observed to reduce after a longer treatment duration (20 weeks). The content of kaempherol 3-glucoside in the leaf samples were also observed to diminish after 20 weeks of treatment duration (Table 6). These contrast with the observation found on the stem extracts (Table 7), where synthesis of kaempherol 3-glucoside was only observed at 20 weeks of treatment duration and was not detected at 10 weeks of treatment. Not only that, the stem extract was also observed to contain very little amounts of flavonoids, in contrast to the amounts recorded in the leaves. Similar observations were also recorded in the root extracts, where root extracts obtained from plants harvested after 10 weeks of treatment mostly consists of kaempherol 3-glucoside (Table 8). However, its synthesis was found to be stopped at longer treatment duration (20 weeks). Besides that, root extracts were also observed to be lacking in quercetin and rutin compounds, at both treatment durations (10 and 20 weeks). Overall, the results recorded in this study were similar to the one reported by Karimi et al. (2013) on Labisia pumila and Sembiring et al. (2018) on Caesalpinia bonduc L., where the highest amount of flavonoid was significantly found in leaves followed by the stems and then roots. Besides specific growth condition that affect the production of phytochemicals in plants, a study by Madike et al. (2017) on Tulbaghia violacea highlights on other factors that could affect the composition and amounts of phytocompounds such as extraction methods and types of solvent used.

Conclusions

From the data recorded in the study, it was observed that the interaction effect of shade levels and treatment duration would significantly affect the growth, production of secondary metabolites, antioxidant activities and the antioxidant defense system of P. minus. After 10 weeks, the growth of P. minus was observed to strive the best under high light condition, but its antioxidant activities and antioxidant defense system performed better under low light condition (70% shading). However, when the duration was prolonged until 20 weeks, data analysis showed that P. minus adapt better under high light condition (no shading and 30% shading) in terms of its production of secondary metabolites and its antioxidant activities. This can be seen from the high production of phenolic, flavonoid and anthocyanin compounds in P. minus, as well as its high FRAP reducing power and high superoxide dismutase enzyme activity. Low H2O2 concentration was also recorded under 30% shading level when the treatment duration was prolonged until 20 weeks.