Bioherbicidal Properties of Parthenium hysterophorus , Cleome rutidosperma and Borreria alata Extracts on Selected Crop and Weed Species

: Natural product-based herbicides could be the effective alternatives to synthetic chemical herbicides for eco-friendly weed management. This research, therefore, was conducted to identify the phytotoxic properties of Parthenium hysterophorus L., Cleome rutidosperma DC. and Borreria alata (Aubl.) DC. with a view to introducing them as a tool for natural herbicide development. The methanol extracts of these plants were examined on the germination and growth of Zea mays L., Oryza sativa L., Abelmoschus esculentus (L.) Moench and Amaranthus gangeticus L., Oryza sativa f. Spontanea Roshev. (Weedy rice), Echinochloa colona (L.) Link., Euphorbia hirta L., and Ageratum conyzoides L. under laboratory and glasshouse conditions. A complete randomized design (CRD) with ﬁve replications and randomized complete block design (RCBD) with four replications were laid out for laboratory and glasshouse experiments, respectively. In the laboratory experiment, three plant extracts of 0, 6.25, 12.5, 50, and 100 g L − 1 were tested on survival rate, hypocotyl, and radicle length of eight test plant species. No seed germination of A. conzyoides , E. hirta , and A. gangeticus were recorded when P. hysterophorus extract was applied at 50 g L − 1 . C. rutidosperma had the same effect on those plants at 100 g L − 1 . In the glasshouse, similar extracts and concentrations used in thelaboratory experiments were sprayed on at the 2–3 leaf stage for grasses and 4–6 for the broadleaf species. Tested plants were less sensitive to C. rutidosperma and B. alata compared to P. hysterophorus extract. Among the weeds and crops, A. conyzoides , E. hirta , A. esculentus and A. gangeticus were mostly inhibited by P. hysterophorus extract at 100 g L − 1 . Based on these results, P. hysterophorus was the most phytotoxic among the tested plant extracts and could be used for developing a new natural herbicide for green agriculture.


Introduction
Parthenium hysterophorus L., Cleome rutidosperma DC., and Borreria alata (Aubl.) DC. belong to the family Asteraceae, Cleomaceae, and Rubiaceae, respectively, and are invasive weeds in Malaysia. Their damage includes threats to biodiversity, exacerbation of allergies and dermatitis, mutagenesis in humans and livestock, and interference (competition and allelopathy) with field crops and rangelands [1][2][3]. Parthenium hysterophorus is native to Mexico and has been spreading like wildfire in different countries. At present, ten states of Malaysia are invaded by this weed, and the state Kedah is the worst infested area [4]. Yield C. rutidosperma, and B. alata methanol extract was assessed for their suppressive effects on the growth and development of Weedy rice, E. colona, E. hirta, A. conyzoides, Z. mays, O. sativa, A. esculentus, and A. gangeticus. Pre-germinated seeds of test plants were sown in each pot (15 cm diameter × 12 cm height) and then covered with soil at a depth of 1 cm, and finally, the soil was moistened with tap water. After germination, only five equal-sized healthy seedlings of O. sativa, E. colona, E. hirta, A. conyzoides, and weedy rice, and one seedling (equal-sized healthy) of Z. mays, A. gangeticus, and A. tricolor were kept in each pot. The pots were arranged in a randomized complete block design with four replications. Methanol extracts of P. hysterophorus, C. rutidosperma, and B. alata were sprayed with 6. 25, 12.5, 25, 50, and 100 g L −1 concentration at the 2-3 leaf stage (2 weeks old) for grasses and 4-6 leaf stage for broadleaf species (3 weeks old) with the help of a 1 L multipurpose sprayer (Deluxe pressure sprayer). Spray volume (100 mL m −2 ) was prepared using distilled water [25]. Plants in the control treatment were sprayed with 200 mL water without extract at two days intervals or when needed.

Data Collection
Three weeks after spray, the individual plant was separated into the root, shoot, and leaf fractions. Plant height (PH) and root length (RL) were measured using a 1 m ruler. The leaf area was determined using a leaf area meter (LI-3000, Li-COR, Lincoln, NE, USA) and expressed as cm 2 plant −1 . Fresh and dry weights were determined using a digital balance. Samples were dried in an oven at 60 • C for 72 h to take the dry weight of the samples. Total chlorophyll content indicated as SPAD value was measured by a chlorophyll meter, SPAD-502 (Menolta, Japan), as described by Mahdavikia et al. [26]

Statistical Analysis
A two-way analysis of variance (ANOVA) was performed to determine any significant differences between each treatment and the control for both experiments. The differences between the treatment's means were pooled using the Tukey test with a 0.05 probability level. SAS (Statistical Analysis System) software, version 9.4 (Cary, NC, USA) was used to conduct the analysis. Probit analysis based on the percentage of inhibition of survival rate or radicle and hypocotyl length was used to measure ECs50, ECr50, and ECh50. ECs50, ECr50, and ECh50 were the effective doses capable of inhibiting 50% of survival rate, radicle, and hypocotyl length respectively. The most active extracts were determined as the index (Re) using the equation given below for each extract tested: Rank (Re) = ECs50n (survival rate) + ECr50n (radicle) + ECh50n (hypocotyl) where Re is the rank of the extract and ECs50n, ECr50n, and ECh50n are the concentrations of treatments that cause 50% inhibition on germination, root, and hypocotyl growth, respectively. The extract with the lowest Re values was considered as the most phytotoxic treatment and the least phytotoxic effect of the extract was observed for the highest Re value.

Effect of Methanol Extracts on Survival Rate and Initial Growth of Weeds
The results showed that P. hysterophorus, C. rutidosperma, and B. alata extracts significantly influenced the survival rate as well as the hypocotyl and radicle length of the tested weed species (p < 0.05). The magnitude of inhibition increased with an increase in extract concentration (Table 1).  No survival rate was recorded in A. conyzoides and E. hirta when P. hysterophorus extract was applied at 50 g L −1 . Meanwhile, no significant inhibition on the survival rate of all weed species was observed when C. rutidosperma and B. alata extracts were applied at a low concentration of 6.25 g L −1 . However, C. rutidosperma and B. alata inhibited seed survival rate at the highest concentrations by 100% and 90%, respectively. The survival rate of A. conyzoides and E. hirta seeds was more sensitive to the extracts compared to Weedy rice and E. colona seeds.
The radicle length of the target weed was significantly reduced (p < 0.05) by P. hysterophorus extract at a concentration equal to or higher than 6.25 g L −1 . The root growth of A. conyzoides, E. hirta, E. colona, and Weedy rice was reduced by 89.3%, 85.4%, 89.5%, and 88.7% when treated with C. rutidosperma extracts at the concentration of 50 g L −1 . No radicle development of the test species was observed when P. hysterophorus extract was applied at the highest concentration, whereas up to 90% inhibition was observed by B. alata extract (Table 1). All extracts decreased the hypocotyl elongation of the target weeds. At the concentration of 50 g L −1 , P. hysterophorus, C. rutidosperma, and B. alata extracts reduced the hypocotyl length of all tested weeds by 95-100%, 84-90%, and 76-81%, respectively. Therefore, the extent of inhibition of P. hysterophorus extract was higher compared to C. rutidosperma and B. alata extracts.

Effect of Methanol Extracts on the Survival Rate and Initial Growth of Crops
The survival rate, hypocotyl and radicle length of the tested crops were also significantly influenced by the methanol extract of P. hysterophorus, C. rutidosperma, and B. alata. The decrement of these parameters increased with the increase of the extract concentration when compared to the control ( Table 2).
The reduction in radicle length ranged from 93 to 100% for P. hysterophorus extract, 78 to 100% for C. rutidosperma extract, and 74 to 88% for B. alata extract at the highest concentration (100 g L −1 ). Extracts of P. hysterophorus, C. rutidosperma, and B. alata differed from each other in reducing the radicle length of tested crops compared to the control ( Table 2). The P. hysterophorus extract exerted a higher effect in reducing the radicle length of the target crops. For instance, at 50 g L −1 of P. hysterophorus extract, the radicle growth of A. gangeticus was completely suppressed (100%) while in C. rutidosperma and B. alata extracts it was reduced by 89.2% and 77.4%, respectively.
Hypocotyl growth of all tested crops responded differently to P. hysterophorus, C. rutidosperma, and B. alata extracts. The highest concentration (100 g L −1 ) of P. hysterophorus extract resulted in a reduction of 80 to 100% in hypocotyl length of the tested species. On the other hand, 100 g L −1 of C. rutidosperma and B. alata extracts resulted in 67 to 100% and 65 to 82% hypocotyl length reduction, respectively. At the lowest concentration (6.25 g L −1 ), C. rutidosperma and B. alata did not show any significant effect on the hypocotyl growth of Z. mays. The hypocotyl length of test crops was reduced by arange of 10.8 to 100%, 1.4 to 100%, and 1.4 to 82.0% when treated with P. hysterophorus, C. rutidosperma and B. alata extracts, respectively.

Comparison of Methanol Extracts on Examined Initial Growth Parameters and Plants
The half inhibitory concentrations of each extract for all test species are shown in Table 3. The effectiveness of the P. hysterophorus extract was higher than the C. rutidosperma and B. alata extract, as the rank value of C. rutidosperma extract (598.3 g L −1 ) and B. alata extract (876.9 g L −1 ) were more than the P. hysterophorus (393.9 g L −1 ). The obtained EC 50 showed differences among the response of test plants to the inhibitory effect of P. hysterophorus, C. rutidosperma, and B. alata ( Table 3). The differences in the sensitivity of species to the extracts were also evident from the rank values of plants. Zea mays was the only species affected at higher concentrations i.e., 157.7, 206.3, and 329.3 g L −1 of P. hysterophorus, C. rutidosperma, and B. alata extracts, respectively. In other words, the Z. mays plant showed more tolerance, which indicates that only a high concentration of extracts could suppress this plant. The second less sensitive test plant (after Z. mays) was A. esculentus. The rank value of A. conyzoides, E. hirta, E. colona, Weedy rice, O. sativa, and A. gangeticus was 22.0, 22.6, 25.7, 26.6, 30.9, and 33.8 g L −1 , respectively, when they were treated with P. hysterophorus extract. The rank values for these tested plant species were 39.9, 44.2, 39.9, 39.9, 54.3, and 50.9 g L −1 , respectively for C. rutidosperma and 50.6, 51.6, 53.6, 55.0, 66.9, and 79.8 g L −1 , respectively for B. alata extracts. Therefore, it was apparent that these tested plant species were most sensitive to P. hysterophorus extract. Data regarding the effect of the foliar spray of methanol extracts of P. hysterophorus, C. rutidosperma, and B. alata on control (%), plant height, and root length of tested weeds are presented in Table 4. Among all the tested weeds, only A. conyzoides was controlled by 80% and 100% when sprayed with P. hysterophorus at a concentration of 50 g L −1 and 100 g L −1 , respectively. A similar trend was observed where an increase in the concentration of each treatment resulted in a remarkable reduction in plant height. Among thetreatments, P. hysterophorus showed a more phytotoxic effect on the plant height of tested weeds compared to C. rutidosperma and B. alata at the highest concentration (100 g L −1 ). At the same concentration, P. hysterophorus extract caused 100%, 60.0%, 20.4%, and 19.2% reduction in plant height of A. conyzoides, E. hirta, Weedy rice, and E. colona, respectively. On the other hand, 8.1to 11.1% and 5.6 to 8.6% plant height reduction was achieved by C. rutidosperma and B. alata extracts, respectively for all tested weeds. Among the tested weeds, the root length of A. conyzoides was inhibited completely (100%) when sprayed with 100 g L −1 P. hysterophorus extract. The inhibition of root length of all tested weeds ranged from 35.3 to 100%, 15.9 to 22.4%, and 14.6 to 16.5% at the same concentration (100 g L −1 ) of P. hysterophorus, C. rutidosperma, and B. alata, respectively.
Declined leaf area and total chlorophyll content of all tested weeds werealso observed with an increase in the foliar spray of methanol extracts of P. hysterophorus, C. rutidosperma, and B. alata. Similar to plant height and root length, the leaf area and total chlorophyll of A. conyzoides were most affected by the foliar spray of P. hysterophorus extract compared to the others (Table 5). Leaf area inhibition of A. conyzoides, E. hirta, Weedy rice, and E. colona ranged from 15 to 100%, 12 to 70%, 5.3 to 42.0%, and 5.6 to 35.3%, respectively when sprayed with an increased amount of P. hysterophorus extract. A similar trend was also observed for total chlorophyll. The inhibition percentage for leaf area and total chlorophyll of all tested weeds ranged from 15.3 to 19.4 and 12.0 to 18.90 when sprayed with the highest concentration of C. rutidosperma and B. alata extracts, respectively.

Effect of Methanol Extract on Total Fresh and Dry Weight of Weeds
Total fresh and dry weights of all tested weeds were significantly influenced by the foliar spray of P. hysterophorus extract in a concentration-dependent pattern compared to C. rutidosperma and B. alata extract ( Table 5). The control obtained the highest fresh and dry weight. However, the reduction differed among the targeted species and the treatments. Parthenium hysterophorus extract reduced the fresh and dry weight of tested weeds from 35.3 to 100% and 43.0 to 100% at 100 g L −1 compared to the control, respectively. At 50 g L −1 concentration, the foliar spray of P. hysterophorus extract reduced 52.8% and 87.1% total fresh weight and 56.0% and 90% total dry weight of E. hirta and A. conyzoides, respectively. On the other hand, among the treatments, the highest fresh and dry weight was recorded when different concentrations of C. rutidosperma and B. alata extracts were applied on the tested weeds (Table 5).

Phytotoxic Effect of Methanol Extracts on Plant Height and Root Length of Crops
The effect of treatments on the development of tested crops at the maturity stage is shown in Table 6. The result indicated that the suppressive magnitude of applied extracts was species-dependent. Plant height of all tested crops except Z. mays was significantly influenced by the extract of P. hysterophorus compared to C. rutidosperma and B. alata extract. There was no significant difference between the activities of C. rutidosperma and B. alata at the lowest concentration. The highest plant height reduction (62.2%) occurred at 100 g L −1 concentration of P. hysterophorus for A. gangeticus, and only 20.4% plant height of O. sativa was reduced at the same concentration (Table 6). Table 5. Effect of methanol extract of P. hysterophorus, C. rutidosperma, and B. alata on leaf area (cm 2 ), total chlorophyll (SPAD), total fresh weight (g pot −1 ), and total dry weight (g pot −1 ) of A. conyzoides, E. hirta, weedy rice, and E. colona.    Root lengths of all tested plants were significantly decreased by all the applied extracts. Among the species, root length was more reduced in A. gangeticus at 100 g L −1 concentration of P. hysterophorus with an inhibition index of 72.2% followed by 61.9%, 38.9%, and 38.6% in A. esculentus, Z. mays, and O. sativa, respectively. This indicates that the effects caused by the P. hysterophorus extract on the plant height and root length were more prominent at higher concentrations across the species. The C. rutidosperma and B. alata extracts were less phytotoxic on the plant height and root length of the tested crops compared to P. hysterophorus extract. The extract of C. rutidosperma inhibited the plant height and root length of the tested crops by 8.2 to 14.1% and 9.2 to 21.0%, respectively.

Phytotoxic Effect of Methanol Extracts on Leaf Area, Total Chlorophyll, Fresh and Dry Weight of Crops
Foliar spray of P. hysterophorus, C. rutidosperma,and B. alata extract had a significant effect on leaf area and chlorophyll content of the test species (Table 7). The effects of P. hysterophorus extracts showed a decline from 6.3 to 61.0% at the lowest (6.25 g L −1 ) to the highest (100 g L −1 ) concentrations on the leaf area of A. esculentus, while 3.88 to 37.97% was recorded in O. sativa. The chlorophyll content of all tested crops except O. sativa was significantly affected by the foliar spray of P. hysterophorus extract at the concentration of 6.25 g L −1 . The test crop A. esculentus showed a 17.3% decrease in chlorophyll content compared to O. sativa when sprayed with P. hysterophorus extract at the highest concentration (100 g L −1 ). Leaf area and chlorophyll content of A. gangeticus was inhibited by 22.1% and 18.3% by C. rutidosperma extract and 16.9% and 19.0% by B. alata extract, respectively at 100 g L −1 concentration.   might be suppressed by allelochemicals stress. Aslam et al. [39] reported the phytotoxic effect of Calatropis procera, Peganum harmala, and Tamarix aphylla on the shoot and root length of mustard and wheat, and wheat was sensitive to all three extracts at all the concentrations. Mulberry aqueous leaf extract suppressed shoot and root length, shoot and root dry matters of Bermuda grass by 90% and 80% at 100% concentrations, respectively [40].
Hassan et al. [41] also observed a decrease in shoot and root length of Zea mays and Vigna unguiculata treated with increased concentrations of Jatropha curcas extract. Foliar spray of P. hysterophorus extract reduced dry weights and leaf area as the level of concentration increased across species although the species responded independently. The reduction in total dry weight was observed to be associated with a decrease in plant height and leaf area. Total dry weight and leaf area were mostly decreased in A. conyzoides and E. hirta, respectively. Leaf area reduction was higher in A. conyzoides and lower in Z. mays at 21 days after spraying with P. hysterophorus extract. This type of species-dependent inhibitory activity was also reported by several studies. For example, phytotoxins have an adverse impact on the growth of certain plants while having little or no inhibition in other plants at certain concentrations [42][43][44]. Several studies reveal a decline in leaf area of certain plant species using different extracts [45,46].
Chlorophyll is a determinant factor in photosynthesis and it was found to be lower in A. conyzoides among all tested species. The leaves of the tested plants appeared partially folded and this may lead to a decrease in photosynthetic activity [47]. Reduction of chlorophyll content in plants due to application of allelopathic plant extracts was also reported by Kamal [48], Siyar et al. [49], and Abdel-Farid [50].
It was also observed in the present study that the application of plant extracts in a foliar spray in laboratory conditions caused more inhibition compared to glasshouse conditions. Similar findings were also reported by Al-Humaid and El-Mergawi [21]. The inhibition by foliar spray may occur through various mechanisms such as suppressed hormone activity, a decreased rate of ion absorption, enzyme activity inhibition, reduce cell membrane permeability and also inhibit certain physiological processes such as photosynthesis, respiration, and protein formation. Thus, the seedling stage and the more mature stage of target plants vary in their sensitivities to plant extracts.

Conclusions
The study demonstrated that all the methanol extracts from three Malaysian invasive weeds (P. hysterophorus, C. rutidosperma, and B. alata) have allelopathic potential on the seed germination, growth, and development of tested plants. P. hysterophorus appeared as the most phytotoxic plant extract among the three. Moreover, the phytotoxic effect of the extracts was dependent on the target species, extract concentrations, and the extracted plant species. The growth and development of the tested plant species in the glasshouse were less affected compared to seed germination and growth under laboratory conditions. The only phytotoxic impact was provided by P. hysterophorus on the tested plant species in the glasshouse trial. Among the test species, A. conyzoides was more sensitive to P. hysterophorus extract. Taking into account the promising result of P. hysterophorus extract, this weed could be used for further study to develop a natural product-based herbicide for sustainable green agriculture. Identification and characterization of the most active phytotoxic compounds of the P. hysterophorus extract will be the first step of future studies.