Bioherbicidal Activity and Metabolic Profiling of Potent Allelopathic Plant Fractions Against Major Weeds of Wheat—Way Forward to Lower the Risk of Synthetic Herbicides

The productivity of major field crops is highly compromised due to weed infestation. Inefficient weed management practices and undue and excessive use of chemical herbicides have drastically contaminated the environment and human health, in addition to resistance development in weed species. Therefore, utilization of allelopathic plants to explore phytochemicals as potent organic alternatives to such chemical herbicides has become indispensable. The current study evaluates the comparative bio-herbicidal potential of methanolic extracts of castor (Ricinus communis), artemisia (Artemisia santolinifolia), wheat (Triticum aestivum), and sorghum (Sorghum bicolor) to suppress growth of major weeds, i.e., wild mustard (Sinapis arvensis), Italian ryegrass (Lolium multiflorum), and carrot grass (Parthenium hysterophorus). The results demonstrated a concentration-dependent effect on weeds’ growth. Overall, in vitro seed germination was reduced from 60 to 100% in response to 5% (w/v) extract concentration. Significant reduction in radicle length, hypocotyl length, and fresh biomass of the weeds was also observed. A strong inhibitory effect was seen in in vivo pot experiments, revealing that application of 10–20% methanolic extracts induced permanent wilting and substantial reduction in the chlorophyll content of weeds along with 20–80% increase in oxidative stress. Artemisia showed the most significant allelopathic effect, on account of highest phenolic and flavonoid contents, followed by castor, wheat, and sorghum, against S. arvensis, L. multiflorum, and P. hysterophorus, respectively. Phytochemical analysis, through high-performance liquid chromatography (HPLC), also exhibited a correlation between extract’s phytotoxicity and their antioxidant potential due to their major constituents (rutin, quercetin, catechin, gallic acid, vanillic acid, syringic acid, ferulic acid, p-hydroxy benzoic acid, p-coumaric acid, and sinapic acid), among the total of 13 identified in methanolic fractions. Comprehensive profiling of allelochemicals with liquid chromatography–mass spectrometry (LC-MS) determined 120, 113, 90, and 50 derivates of phenolic acids, flavonoids, and alkaloids, reported for the first time through this study, demonstrating significant allelopathic potential of the targeted plant fractions, which can be explored further to develop a sustainable bio-herbicidal formulation.

The productivity of major field crops is highly compromised due to weed infestation. Inefficient weed management practices and undue and excessive use of chemical herbicides have drastically contaminated the environment and human health, in addition to resistance development in weed species. Therefore, utilization of allelopathic plants to explore phytochemicals as potent organic alternatives to such chemical herbicides has become indispensable. The current study evaluates the comparative bio-herbicidal potential of methanolic extracts of castor (Ricinus communis), artemisia (Artemisia santolinifolia), wheat (Triticum aestivum), and sorghum (Sorghum bicolor) to suppress growth of major weeds, i.e., wild mustard (Sinapis arvensis), Italian ryegrass (Lolium multiflorum), and carrot grass (Parthenium hysterophorus). The results demonstrated a concentration-dependent effect on weeds' growth. Overall, in vitro seed germination was reduced from 60 to 100% in response to 5% (w/v) extract concentration. Significant reduction in radicle length, hypocotyl length, and fresh biomass of the weeds was also observed. A strong inhibitory effect was seen in in vivo pot experiments, revealing that application of 10-20% methanolic extracts induced permanent wilting and substantial reduction in the chlorophyll content of weeds along with 20-80% increase in oxidative stress. Artemisia showed the most significant allelopathic effect, on account of highest phenolic and flavonoid contents, followed by castor, wheat, and sorghum, against S. arvensis, L. multiflorum, and P. hysterophorus, respectively. Phytochemical analysis, through high-performance liquid chromatography (HPLC), also exhibited a correlation between extract's phytotoxicity and their antioxidant potential due to their major constituents (rutin, quercetin, catechin, gallic acid, vanillic acid, syringic acid, ferulic acid, p-hydroxy benzoic acid, p-coumaric acid, and sinapic acid), among the total of 13 identified in methanolic fractions. Comprehensive profiling of allelochemicals with
Considering the increased density of weeds among major crops including wheat, there is a consistent need to explore new allelopathic plants and their phytotoxins. For this purpose, screening of plants should be done through easy and rapid bioassays that may efficiently identify the active allelochemicals through chromatographic techniques (Duke, 2015). Allelochemicals can prove to be an eco-friendly and equally efficient weed control approach with minimum environmental hazards due to their structural organizations, short half-life in soils, and synergistic action when released into the environment (Soltys et al., 2013). Diverse classes of allelochemicals, including phenols, terpenes, flavonoids, fatty acids, and steroids, are known to have phytotoxicity against multiple weeds by inhibiting their photosynthetic pathways, disrupting activity of metabolic enzymes, adversely affecting mitochondrial respiration, inhibiting mitotic process, increasing oxidative stress, altering lignin composition, inhibiting cell division and enzyme functions, and inactivating protein synthesis (Li et al., 2010;Golisz et al., 2011;Borella et al., 2014;Wang et al., 2014). Contrary to synthetic herbicides, these natural compounds act by blocking biochemical pathways of weeds in a natural way and, hence, may be considered as a sustainable alternative to chemical herbicides (Dayan et al., 1999). Thus, the present study has been designed to investigate herbicidal potential of R. communis, A. santolinifolia, T. aestivum, and S. bicolor extracts against major weeds like S. arvensis, L. multiflorum, and P. hysterophorus and to identify and characterize potent allelochemicals in them. Characterization of novel allelopathins in such plants may assist in providing a basic platform for organically developed herbicide formulation to manage weed infestation in wheat.

Plant Extract Preparation
Wheat and sorghum accessions (36003 and 9978, respectively) were obtained from National Agricultural Research Centre (NARC), Islamabad, Pakistan. These accessions were grown under glass house conditions and were harvested after 1 month of seed germination. Artemisia plants were collected in July before flowering stage, from Khunjerab National Park (KNP), Pakistan, while castor plants were collected from the local area of Islamabad. For extraction of phenolic compounds, five whole plants of each species were separately washed and shade dried before maceration with liquid nitrogen. Methanol was used as solvent, as it prevents phytochemical degradation by oxidative enzymes, i.e., polyphenol oxidase and peroxidase (POD), which are active in aqueous solutions (Duke, 2015). Macerated plant material (20 g) was soaked in methanol (100 ml) in a conical flask. The flasks were incubated in dark at room temperature for 72 h. Following incubation, the flasks were regularly shaken to allow efficient extraction of phytochemicals. Extracts were filtered through filter paper (pore size 2.5 µm), and the filtrate was subsequently run through filtration assembly (DEKKER, Model #A300, filter pore size 0.45 µm), under gravity, for complete removal of contaminants. Resulting 20% (w/v) castor and artemisia extracts served as stock solution and were stored at -20 • C.
Each methanolic extract was then diluted to prepare 1, 3, and 5% solutions for plate bioassay. For soil (pot) experiments, 10, 15, and 20% concentration dilutions were prepared. The same dilutions were prepared with pure solvent, to treat control seeds/seedlings for plate and soil bio-assays, whereas sterile distilled water was used as a positive control in both experiments. The effect of solvent was normalized to aqueous control before data analysis for each experiment.

Seed Germination Bioassay
Seeds of S. arvensis, L. multiflorum, and P. hysterophorus were collected from the Department of Weed Science, Peshawar, Pakistan, and were surface sterilized with 5% sodium hypochlorite (1 min) to inactivate any microbes and washed three times with autoclaved double-distilled water (Ali et al., 2010). The sterilized seeds (6-10) of each weed species were placed separately in glass Petri plates (90 mm × 10 mm) containing autoclaved filter paper (Whatman no. 42) in triplicates. Subsequently, 1 ml of each extracts was applied to evenly moisten filter paper. The control group received the same amount of methanol (1, 3, and 5%) and sterile water. Plates were sealed with parafilm and incubated in a growth room at 24 • C and 16/8-h light/dark photoperiod. Each treatment was repeated three times at an interval of 48 h each. Seed germination count of individual weed was recorded from 10 to 15 days after germination to calculate germination index and germination percentage of each, as follows (Scott et al., 1984).

In vitro Evaluation of Methanolic Fractions on Seedling Growth
The effect of methanolic fractions (1, 3, and 5%) of castor, artemisia, wheat, and sorghum were investigated through plate bioassays. Several surface-sterilized seeds of S. arvensis, L. multiflorum, and P. hysterophorus were allowed to germinate separately on autoclaved filter papers (Whatman no. 42), moistened with 1 ml of sterile water, contained in Petri plates, under the same conditions as described above. After 7 days of germination, 6-10 similarly sized seedlings of P. hysterophorus, S. arvensis, and L. multiflorum were selected for bioassay. Each extract was applied uniformly (1 ml in each plate) to seedlings, while the control group was inoculated with the same amount of sterile water. The solvent control group was given 1, 3, and 5% methanol. Treatments were repeated thrice on alternate days. After 5 days of each experiment, hypocotyl length (mm), radicle length (mm), and fresh biomass (mg) of each seedling were measured.

Total Phenolic Acid Content Assay
TPC of extracts was determined by Folin-Ciocalteu (FC) method (Kamtekar et al., 2014). Stock solution of gallic acid (100 µg of gallic acid per ml of methanol) was used to prepare the following dilutions: 50, 25, 12.5, and 6.25 µg ml −1 . For TPC estimation, 120 µl of gallic acid or each extract fraction was mixed in 600 µl of FC reagent in a 15-ml Falcon tube, followed by incubation for 5 min. Sodium carbonate (7%) was then added in the solution, volume was adjusted up to 10 ml with sterile distilled water, and the solution was allowed to stand in dark at room temperature for 30 min. Absorbance was read at 760nm wavelength by a spectrophotometer (SPECORD 50). Solution containing only the reagents except the extract or standard was considered as a blank. TPC in samples was estimated by gallic acid calibration curve, and results were expressed as mg of gallic acid equivalent (GAE)/100 g dry mass of plant material (Kamtekar et al., 2014).

Total Flavonoid Content Assay
TFC of each extract was determined by using aluminum chloride colorimetric assay (Kamtekar et al., 2014). Quercetin standard solution (1,000 µg of quercetin per ml of methanol) was used to prepare following dilutions: 100, 200, 400, 600, and 800 µg ml −1 . To estimate TFC, 1 ml of each fraction or quercetin standard was thoroughly mixed in 4 ml of sterile water and 0.3 ml of 5% NaNO 2 in a 15-ml tube. The mixture was allowed to stand at room temperature for 5 min. Subsequently, 0.3 ml of 10% AlCl 3 and 2 ml of 1 M NaOH solution were added at the sixth minute. The volume of the mixture was raised up to 10 ml with sterile distilled water. The final solution was used to measure absorbance at 510 nm on spectrophotometer (SPECORD 50) using visible range. Solution containing all the chemicals except the extract/standard was used as a blank. TFC was determined by quercetin calibration curve, and the results were expressed as quercetin equivalent (QE) in mg/100 g dry mass of plant material (Kamtekar et al., 2014).

In vivo Evaluation of Methanolic Fractions on Seedling Growth
To determine the optimized phytotoxic extract concentration for weed growth suppression in soil, multiple seeds were sown in pots (7.5 × 7.8 cm) containing 100 g mixed compost material and incubated in a growth room at 24 • C temperature and 16/8h light/dark photoperiod. After seedling emergence, thinning was done to maintain uniform plant density of 3, 5, and 15 seedlings of P. hysterophorus, L. multiflorum, and S. arvensis per pot, respectively. In the preliminary screening assessment, lower concentrations of extracts did not affect the weed growth. Therefore, 10, 15, and 20% concentrations were applied at 5 ml per pot of each solution to 1-week-old seedlings on alternate days (days 1, 3, and 5). The same amount of 20% methanol and water was applied to the control.

Chlorophyll Content Assay
Treated weed seedlings (10, 15, and 20% of each extract) were harvested for total chlorophyll content analysis. Leaves were macerated in liquid nitrogen to determine chlorophyll contents. Ground leaves' material (0.5 g) in 1.5-ml tubes was mixed with 1 ml of dimethyl sulfoxide (DMSO). Chlorophyll was allowed to dissolve into DMSO by incubating the tubes at 60 • C temperature for 30 min. Absorbance were recorded in spectrophotometer (SPECORD 50) at 645 and 663 nm for chlorophyll pigments a and b, respectively. Total chlorophyll concentration was estimated according to the formula equation by Barnes et al. (1992).

Biochemical Analysis of Weeds Post Extract Treatment
Biochemical analysis of weeds on the effect of 20% allelopathic plant extracts (CE, AE, TE, and SE) was determined through enzymatic [catalase (CAT), superoxide dismutase (SOD), POD] and non-enzymatic assays [malondialdehyde (MDA)]. All analyses were performed spectrophotometrically. Crude enzyme extract was prepared separately from leaves of S. arvensis, L. multiflorum, and P. hysterophorus after permanent wilting post-treatments. Fifty milligrams of each sample was macerated with liquid nitrogen followed by homogenization with 50 mM of phosphate buffer (pH 7.0). The homogenate was centrifuged at 4 • C for 20 min at 10,000 × g. Supernatant was collected as enzyme extract and subsequently used for antioxidant enzyme assays. The activity of various enzymes in response to allelopathic stress was determined as follows.

Catalase Activity
CAT activity was used to determine the rate of decomposition of H 2 O 2, measuring spectrophotometrically at room temperature by monitoring the decrease in absorbance at 240 nm as per the method of Aebi (1974), with some modifications. The reaction mixture (2 ml) contained 100 mM of phosphate buffer of pH 7.0, 40 mM of H 2 O 2 , and 0.2 ml of enzyme extract. One unit was defined as the amount of enzyme that caused an absorbance change of 0.1 per minute under assay conditions. The extinction coefficient of CAT enzyme is 39.4 mM min −1 , and concentration of H 2 O 2 was expressed as mM of H 2 O 2 oxidized min −1 g −1 fresh weight (FW).

Superoxide Dismutase Activity
SOD activity was determined by measuring its ability to inhibit the photoreduction of nitroblue tetrazolium chloride (NBT) by following Beauchamp and Fridovich (1971) method with some modifications. The reaction mixture (1 ml) comprised 63 µM of NBT, 13 mM of L-methionine, 0.1 mM of EDTA, 13 mM of riboflavin, 0.05 M of sodium carbonate, 50 mM of potassium phosphate buffer of pH 7.8, and 0.125 ml of enzyme extract. Distilled water (0.125 ml) was used instead of extract in the control. The reaction mixture was placed in a fluorescent lamp for 15 min at 25 • C to stimulate the reaction. The reaction was stopped by dark incubation for 15 min. The blue-colored formazan formed by photochemical reduction was measured as increase in absorbance at 560 nm. Appropriate controls devoid of enzyme extract were used as blank controls incubated in the dark.
One SOD unit was defined as the amount of enzyme causing 50% of the NBT photoreduction inhibition. The extinction coefficient of enzyme is 26.6 mM min −1 and expressed as units of enzyme activity mM min −1 g −1 FW.

Peroxidase Activity
POD activity was analyzed by determining the absorbance at 470 nm caused by oxidation of guaiacol to tetraguaiacol as describe by Mandal et al. (2008), with some modifications. The final reaction mixture (3 ml) contained 20 mM of guaiacol (0.1 ml), 0.1 mM of potassium phosphate buffer (pH 5.0) (2.5 ml), 40 mM of H 2 O 2 (0.2 ml), and the enzyme extract (0.2 ml). The extinction coefficient of POD enzyme is 26.6 mM cm −1 . POD unit was taken as the amount of protein required to oxidize 1 mM of tetraguaiacol per minute and calculated as mM min −1 g −1 FW.

Lipid Peroxidase (Malondialdehyde) Activity
The MDA content was determined to test the level of lipid peroxidation as described by Heath and Packer (1968), with some modifications. Fresh leaves of the target weed (0.1 g of each) were homogenized with 1.0% (w/v) trichloroacetic acid (TCA). The homogenate was centrifuged at 10,000 × g for 15 min at 4 • C, and 0.8 ml of the supernatant was added to 2 ml 0.5% (v/v) thiobarbituric acid (TBA) in 20% TCA. The reaction mixture was incubated at 95 • C for 15 min in water bath. The reaction was stopped by incubating samples on ice for 10 min. Mixture was then centrifuged 10,000 × g for 5 min. The absorbance of supernatant was measured at 532 and 600 nm. The MDA concentration was calculated by using the extinction coefficient of 155 mM cm −1 . The concentrations were expressed as mM g −1 FW.

Crude Extract Fractionation Through Solid-Phase Extraction
To remove impurities from crude extracts, fractionations were performed individually using solid-phase extraction (SPE) assembly (Thermo Scientific 24-Port Glass Box Vacuum Manifold) (Głowniak et al., 1996). Methanol was utilized for fractionation, as most phenolic compounds are soluble in it. Initially, the SPE cartridge of 3 ml (CMWBOND, 500 mg) was conditioned once with water under gravity flow, followed by conditioning with methanol. Sample tube was then connected to flushed cartridge. Vacuum pressure and flow rate were adjusted to 10 mmHg and 5-7 ml min −1 , respectively. Waste solvent was collected in waste container. Cartridge was dried under maximum pressure and residues were eluted with methanol.
Column was purged to remove initial mobile phase from the system before application of standards. Flow rate and pressure (maximum) of solvent were set as 1 ml min −1 and 250 kgf cm −2 , respectively. Each standard was injected into the micro injector with a volume of 20 µl. Retention times of standards were noted for reference. Peak detected through UV detector was obtained, and peak area was identified for quantification of phenolic compounds in pure methanolic fractions. Samples were analyzed in three replicates.

Liquid Chromatography-Mass Spectrometric Analysis for Allelochemical Identification
The detailed profiling of allelochemicals present in each allelopathic extract was performed according to protocols as described by Kang et al. (2016), with minor modifications. Liquid chromatography-mass spectrometry (LC-MS) analysis was executed in Agilent triple quadrupole system. Compounds were separated on column compartment (

Statistical Analysis
The experiments for seed germination, seedling growth, and phytochemical analysis were conducted in completely randomized design (CRD) with three replicates. Each experiment was repeated thrice. Treatment values were normalized to methanol prior to analysis. Results were analyzed with two-way multivariate ANOVA using general linear model. Variables were transformed where and when required for data normalization. Percentage data were arcsine transformed, while the rest of the variables were log transformed for normality. Transformed variables were back transformed, and data were presented as mean ± standard errors of means (SEM). Significance of obtained results was analyzed with Tukey's honestly significant difference (HSD) test at probability level P < 0.05 with SPSS v. 23 (IBM, United States).

Inhibitory Effects of Methanolic Extracts on Weed Seed Germination
The potential allelopathic contribution of castor (Ricinus communis), artemisia (Artemisia santolinifolia), wheat (Triticum aestivum L.), and sorghum (Sorghum bicolor) extracts revealed that 3 and 5% concentration significantly reduced seed germination index and germination percentage. The 5% castor extract induced 100, 70, and 40% germination inhibition in Sinapis arvensis, Lolium multiflorum, and Parthenium hysterophorus, respectively (Supplementary Figures 1-3). Artemisia extracts (3 and 5%) were able to completely inhibit the germination of S. arvensis and L. multiflorum, while 90% inhibition of P. hysterophorus was observed. Wheat and sorghum extracts were able to induce 60-90% and 30-80% germination inhibition in all weeds, respectively. Solvent and aqueous controls had no negative effect on germination index or percentage and seeds sprouted in a normal time period (Figures 1A,B).

In vitro Evaluation of Seedling Growth of Weeds
To determine the seedling growth rate, radicle length, hypocotyl length, and biomass of weed seedlings were measured. Results illustrated that 3 and 5% extracts significantly reduced average root and shoot length of all weeds as well as the relative seedling growth. Sorghum extract showed the least significant effect on all weeds. No reduction in seedling growth was observed in control treatments (Supplementary Figures 4-8). Apart from stunting growth attributes, extract treatments also exhibited a bleaching effect in weeds. Any effect of similar concentrations of methanol was normalized to aqueous control (Figures 2A-C).

In vivo Effect on Morphology and Chlorophyll Content of Weeds
In soil pot experiment, all the test species of weeds permanently wilted and showed a faded effect after the treatment's applications. S. arvensis seedlings became flaccid after 2 h of extract application; however, L. multiflorum and P. hysterophorus showed reduction in seedling growth after 24 and 72 h of fraction application, respectively. Permanent growth inhibition was observed after 5 days of experiment with three applications of extracts on alternate days. The most significant effect was demonstrated with artemisia extract against all weeds. Values are presented as mean ± SEM (n = 3).
Frontiers in Plant Science | www.frontiersin.org P. hysterophorus showed the least phenotypic changes in comparison with control seedlings. Since chlorophyll content determines the photosynthetic ability and the growth pattern of the plants; thus, 10 and 15% extract application did not affect the chlorophyll content significantly, whereas 20% of all allelopathic extracts drastically reduced chlorophyll content of the weeds, resulting in dead seedlings (Figure 3). Seedlings treated with methanolic control (20%) and aqueous control demonstrated normal growth pattern (Figures 4A-C).

Biochemical Analysis of Weeds
To evaluate whether the negative effects of 20% extracts were caused by excessive production of reactive oxygen species (ROS) and altered membrane structure, activity of few antioxidant enzymes (CAT, SOD, and POD) that regulate ROS levels along with membrane damage by lipid peroxidation was investigated. Results indicated that activity of these enzymes significantly increased in weeds, where 20% AE depicted the most significant increase (20-80%) in CAT, SOD, POD, and lipid peroxidation, in all tested weeds. CE and TE also altered enzymatic and MDA levels, whereas SE had no significant effect on these enzymes while increasing MDA level in all weeds (Figures 5A-C).

Preliminary Identification of Allelochemicals With High-Performance Liquid Chromatography
Methanolic fraction of extracts (from SPE) revealed the presence of various phenolic compounds and flavonoids against the standards employed. Flavonoids identified in extracts included rutin, quercetin, myricetin, kaempferol, and catechin. Concentration of flavonoids ranged from 2.31 to 3,869.36 µg ml −1 , with rutin (3,869.36 µg ml −1 ) and quercetin (1,114.54 µg ml −1 ) having the highest concentration in wheat and artemisia, respectively.
Identified phenolic acids included gallic acid, vanillic acid, p-hydroxy benzoic acid, chlorogenic acid, p-coumaric acid, caffeic acid, and sinapic acid, with a range of 0.02-856.35 µg ml −1 among all extracts ( Table 2). In phenolic acids, gallic acid had the highest amount in all extracts. Kaempferol, catechin, and syringic  acid were not found in artemisia extract, while p-hydroxy benzoic acid was absent in sorghum extract.

Liquid Chromatography-Mass Spectrometric Analysis for Allelochemical Identification
Scanning and characterization of allelochemicals and their derivates in individual allelopathic fraction were performed by LC-MS (Figure 6).

DISCUSSION
Allelopathy has a pertinent significance for sustainable, ecological, and integrated weed management systems (Jabran et al., 2015). The advantages of allelochemicals lie in their versatility of not only being inhibitor of weed growth but also being eco-sustainable. The secondary phytotoxic metabolites released by plants have been focused in agriculture system as a multidisciplinary science for sustainable production (Amb and Ahluwalia, 2016). There      are no generic systemic of allelochemicals in crops system, and henceforth, phytotoxic potential varies from plant-toplant species. These differences across species and allelopathic activities of different phytochemicals have uncovered many aspects of future investigation. Majority of the allelochemicals are completely or partly soluble in water, which makes their application easier without addition of surfactants (Dayan et al., 2009). The structural chemistry of such natural compounds makes them environment friendly as compared with artificially synthesized. Moreover, these allelochemicals are composed of higher content of oxygen and nitrogen, with a less significant amount of heavy and unnatural rings (Soltys et al., 2013). Such attributes reduce their halflife, preventing their accumulation in the environment and ultimately decreasing their toxicity to non-targeted organisms while increasing their efficacy against weeds. The mechanism of suppression of weed growth as affected by few of such allelochemicals' activity has been summarized (see Figure 7), wherein the most significant compounds including phenolic acids, artemisinin, DIMBOA (2,4-dihydroxy-7-methoxy-1,4-benzoxazin-3-one), and sorgoleone are involved in suppressing photosynthetic activity, elevating water stress, and declining cell division to reduce the overall weed growth and development. In this study, the phytotoxic potential of methanolic extracts of castor (R. communis), artemisia (A. santolinifolia), wheat (T. aestivum), and sorghum (S. bicolor) was used in bioassay-based detection of allelopathins to control the invasion of weeds of wheat in an eco-friendly manner.
Germination index and growth suppression are preliminary determinants of the allelopathic activity of extracts (Puig et al., 2018). Initially, the in vitro phytotoxicity was assessed by the inhibition of weeds' (S. arvensis, L. multiflorum, and P. hysterophorus) germination rate and growth suppression in response to application of methanolic extracts of castor, artemisia, wheat, and sorghum, which was found to be dose dependent (3-5% concentration). Phytotoxicity of artemisia extracts was more pronounced leading up to 100% germination inhibition in S. arvensis and L. multiflorum while 90% in P. hysterophorus. The castor extract produced 100, 70, and 40% germination inhibition in S. arvensis, L. multiflorum, and P. hysterophorus, respectively. Germination reduction may be associated with allelochemical interference due to inhibition of amylases and gibberellins, hence modifying the process of mobilization of reserves for embryo development (Singh et al., 2009). Hypocotyl and radicle growth was reduced significantly in response to 3 and 5% extract application in comparison with solvent and aqueous controls. Aslani et al. (2014) reported that seed emergence and early growth of seven test plants, including tomato, lettuce, cucumber, rice, and its weeds, were inhibited with increasing concentrations of methanolic extract of Tinospora tuberculata. Similar results were demonstrated by Han et al. (2008) and Sodaeizadeh et al. (2009) where phytotoxic effects of ginger (Zingiber officinale) and harmel (Peganum harmala) aqueous extracts were concentration dependent against early growth of Avena fetua and Convolvulus arvensis.
Environmental conditions such as biotic or abiotic factors contribute toward structural modification of allelochemicals, when applied in soil (Jilani et al., 2008); thus, the initial concentrations of extracts play a predominant role in determining its phytotoxicity. For this purpose, we used 10, 15, and 20% plant extracts to evaluate their toxicity level for growth suppression in soil experiments. The 20% concentration of all the extracts demonstrated permanent wilting in S. arvensis, L. multiflorum, and P. hysterophorus after 2, 24, and 72 h, respectively. In response to treatment application, a bleaching effect in leaves of S. arvensis and L. multiflorum was observed. Therefore, total chlorophyll content of all weed seedlings was decreased significantly with increased extract concentration. The reduction of chlorophyll levels in lettuce was found to be dose dependent in response to artemisinin (Yan et al., 2015). Such reduction of chlorophyll content can be an appropriate indicator to monitor oxidative damage to plants (Guidi et al., 1997). In the present study, decrease in chlorophyll level of weed plants may be attributed to excessive production of ROS caused by the bioactive compounds present in extracts. In vivo investigation of 20% extracts against weeds suggested an induced oxidative stress in them as exhibited by enzymatic and nonenzymatic assays. Phytotoxicity of many allelochemicals can largely be characterized by the induction of free radicals, which consequently produce O −2 (Yan et al., 2015). These ROS go through various biochemical pathways to generate even more reactive molecules like OH − or HO 2 . All these free radicals thereafter cause membrane and DNA/protein damage as well as lipid peroxidation that eventually result in cell destruction (Cheng and Cheng, 2015). Ding et al. (2007) also affirmed significant increase in CAT, SOD, and H 2 O 2 concentration in cucumber, in response to cinnamic acid application. Conversely, it has also been reported that some allelochemicals may reduce SOD and POD activity. The phytotoxic allelochemical, secalonic acid F, derived from the fungus Aspergillus japonicus, considerably declined antioxidant enzyme activities in several plants (Zeng et al., 2001). Similarly, rice extracts are also known to inhibit SOD and CAT in barnyard grass (Lin et al., 2000). This indicates that allelochemicals can be directly associated with ROS generation, while elevation of oxidative enzymes is a consequence of production of harmful radicals. In an alternate manner, an allelochemical may also block these enzymes, thus making the plants at high risk of damage.
Phytotoxic effects of bio-extracts on recipient plants are mainly a manifestation of an additive effect of allelopathic compounds that act as toxins when applied exogenously, thus inhibiting weed growth (Afzal et al., 2015). These phytotoxins are mostly characterized as phenolics, as the compounds isolated from one of the allelopathic plants, i.e., Delonix regia, were generally phenolic compounds including chlorogenic acid, 3,4-dihydroxy cinnamic acid, 4-hydroxybenzoic acid, 3,5dinitrobenzoic acid, and gallic acid (Chou and Leu, 1992). Thus, total phenolic and flavonoid content assays were performed on 1, 3, and 5% extracts; and both the phenolic and flavonoid contents were higher in higher concentration (5%). However, the highest concentration of total phenolic acids was found in castor followed by artemisia, wheat, and sorghum, whereas total flavonoids were in the highest amount in artemisia followed by castor, wheat, and sorghum. Secondary metabolites like phenols, flavonoids, alkaloids including terpenes, terpenoids, quinones, isothiocyanates, and benzoxazinoids are widely reported allelochemicals. These bioactive compounds, found abundantly in plant extracts, are pertinently responsible for herbicidal activity (Macías et al., 2019). Differentially expressed metabolites and their qualitative and quantitative analyses through profiling are pivotal to gain insights of species-based differences of donor allelopathic plants. The active phytotoxic metabolites in methanolic fractions prepared from castor, artemisia, wheat, and sorghum extracts were identified and subsequently quantified by HPLC and LC-MS by comparing retention time and area under the peak (abundance) with appropriate standards. HPLC analysis of methanolic fraction of crude extracts showed the presence of phenolic compoundsrutin, quercetin, kaempferol, catechin, gallic acid, vanillic acid, ferulic acid, syringic acid, p-hydroxy benzoic acid, p-coumaric acid, chlorogenic acid, and caffeic in the extracts. Strong allelopathic potential of studied extracts can be linked to a higher contents of phenolics in them and consequent elevated antioxidant potential, as most of these compounds are reported as phytotoxic (El-Gawad et al., 2019;Elshamy et al., 2019). Sytar (2015) demonstrated that high antioxidant potential of Fagopyrum esculentum and Fagopyrum tataricum rotundatum was correlated with higher levels of chlorogenic acid, pcoumaric acid, trans-ferulic acid, p-anisic acid, salicylic acid, methoxycinnamic acid, and vanillic acid in them. Batish et al. (2008) thus attributed phenols in rhizosphere of Ageratum conyzoides L., namely, gallic, coumaric, protocatechuic, benzoic, sinapic, p-hydroxybenzoic, and coumaric acid, and flavonoids like kaempferol and quercetin, to be responsible for growth reduction in chick pea. Caffeic acid, gallic acid, syringic acid, catechol, p-coumaric acid, and p-hydroxy benzoic acid, from Calotropis procera aqueous leaf extract, also disrupted mitotic index and induced chromosomal abnormalities in Cassia sophera L. and Allium sepa L., thus reducing their growth and biomass (Gulzar et al., 2016). Phenolic acids (gallic acid, protocatechuic acid, p-hydroxybenzoic acid, p-hydroxybenzaldehyde, vanillic acid, syringic acid, p-coumaric acid, and ferulic acid) and flavonoids (rutin, luteolin, apigenin, and catechin) from Acacia melanoxylon R. Br. aqueous extract significantly reduced protein content of Lactuca sativa (Hussain et al., 2020). Other phytotoxic effects of phenolics include inhibition of nutrient uptake, respiration, photosynthesis, and enzymatic activities of receiver plants (Li et al., 2010). These investigations verify the correlation between the allelopathic activity of donor plants with their antioxidant potential and their phenol content.
The phytotoxic efficacy of allelopathic bioactive compounds on the target weeds is evident from the results of our studies, proving the natural weed control potential of castor, artemisia, wheat, and sorghum. A combination therapy for weed management may be a step forward by utilizing these allelopathic extracts.

CONCLUSION
Castor (Ricinus communis), artemisia (Artemisia santolinifolia), wheat (Triticum aestivum), and sorghum (Sorghum bicolor) have immense allelopathic potential against Sinapis arvensis, Lolium multiflorum, and Parthenium hysterophorus weeds. Phytotoxic responses of these bioextracts were dose dependent. Increase in the herbicidal activity of the extracts with increased concentration can be attributed to higher amount of allelochemicals. Presence of putative allelopathic compounds in these plant fractions makes them an effective source of a natural bioherbicide formulation.
The plants investigated in this study for allelopathic potential are already largely distributed and cultivated for food, feed, medicinal, and industrial purposes; thus, their large-scale cultivation is already in process. Therefore, these plants as a whole and/or their parts can be utilized for extraction of bioactive compounds having bioherbicidal activity. This may be cost-effective and sustainable production, without disturbing the ecosystem or biodiversity. Further, assessment of allelopathicity of identified derivates of compounds may provide a foundation platform for development of eco-friendly analogs for efficient weed management practices.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
SA performed the experimental procedures and wrote the manuscript. SN helped in experiments planning and execution and reviewed the manuscript. SK helped in experimentation. MA helped in HPLC and LC-MS analyses and reviewed the manuscript. AA reviewed the manuscript. ZA conceptualized, supervised, and executed the research project. All the authors have read and approved the final manuscript.

FUNDING
The financial support provided by the Pakistan Science Foundation (PSF) under the project no. PSF/NSLP/C-CIIT (663) to conduct this study is highly acknowledged.