**1. Introduction**

Dramatic increases in food production have been observed in the latter half of the twentieth century owing to the use of agro-chemicals, mechanization, irrigation, high yielding varieties, and post-harvest technology. The production of wheat in Pakistan has increased to ~25 m ton from 4.55 m ton in 1965 [1,2]. The pest attacks continue to incur losses to crop production owing to the diversity of pests and their resistance to prevailing control practices. The use of pesticides has increased from 15 to 20-fold over the last fifty years [3]. Chemical herbicides have gained importance in crop production in the face of a shortage of labor and limited application of mechanical control [4]. The mechanical control is known to contribute to soil erosion and its degradation [5]. Herbicides have led to the emergence of resistant biotypes of weeds, making the herbicide compounds useless to control these weeds [6]. Hence, the discovery of new compounds with novel modes of action is needed to replace these herbicides with more effective compounds to control such weeds. The discovery of such compounds, having herbicidal properties, has reduced over time. Further, the control of one type of weeds with herbicides has provided space to the proliferation of other weed species, which were less problematic for crop production in the past [7]. They have caused losses of biodiversity in the environment. It has deprived the ecosystems of some of their vital functions. Herbicides have aggravated the loss of biodiversity by killing the susceptible species, restricting the growth of others and the degradation of natural resources [8]. Poisoning, growth retardation, sterility, and deaths of wildlife owing to herbicide exposure have been reported by [9]. The residues of herbicides, apart from polluting the natural resources and destroying life forms, may also accumulate in the edible portions of plants, which facilitate their entry to the food chain and bodies of humans. It causes poisoning and chronic diseases in human beings, leading to deaths [10]. Human health disorders caused by herbicides include disorders of the nervous system, malformation of the embryo, loss of fertility, loss of immunity, kidney disorders, and liver disorders [11].

Farmers pay only the costs of manufacturing and marketing of herbicides, which provides economic access to farmers to adopt chemical weed control. The additional costs incurred on the treatment of human illnesses, degradation of natural resources and environment, and loss of biodiversity also need to be paid by farmers, society, or governments. Hence, the scenario of economic, environmental, and biological costs of chemical weed control pushes the researchers towards finding out safer weed control techniques. The importance of biological control has dramatically increased in the present situation. It presents a safer, inexpensive, and easier solution to the above-discussed issues of other control practices. It relies on increasing the strength, population, and activities of the organisms, resulting in growth reduction of weeds [12].

The past efforts in this area were focused on pathogens causing diseases in weeds [13] and insects feeding on weeds [14]. The success of insect biocontrol agents is limited by the existence of multiple hosts of insects in nature, which may cause the emergence of new pests of crops [15]. The pathogens of weeds used for biocontrol wait for suitable environmental conditions to cause infections and diseases in weed plants [13]. It may usually lead to delayed disease development, even after the weeds have caused economic losses of crops. Plant allelochemicals have also been investigated for biological weed control [16]. Their efficacy for weed control is reduced owing to the soil reactions, biodegradation, and mobility. It reduces their bioavailability and phytotoxicity on weeds [17]. These limitations of conventional biological weed control have discouraged researchers of this field, and the popularity of chemical weed control has increased dramatically.

The low success rate in conventional biological weed control has driven scientists to explore the characteristics of the rhizosphere inhabiting bacteria of weeds and crops for the development of novel weed biocontrol techniques. However, researchers have made efforts to explore the type of rhizobacteria, which produce substances inhibitory to the growth of weeds and are the least explored candidates for biological weed control. They release their secondary metabolites (phytotoxic in nature) in the rhizosphere, which is followed by their absorption in weeds. It results in a growth reduction of these weeds. The nature of this interaction between plants and microorganisms may be termed as plant-microbe allelopathy, and the bacteria responsible for these interactions may be called as allelopathic bacteria (AB) [18]. The discovery of host specificity in such microbial interactions with plants by [19] has opened ways for their potential application in crops for weed control. It reflects

the properties of non-inhibition or even promotion of growth of crops among these rhizobacteria [20]. Therefore, the present study was conducted to explore such bacteria from the rhizosphere of weeds and wheat growing in fields facing weed invasions chronically, characterize them for the biological weed control, and evaluate their effects on the growth of wheat and weeds species of wheat.

#### **2. Materials and Methods**

#### *2.1. Isolation of Rhizobacteria*

We collected a large pool of samples of wheat and five weeds along with earth ball across the District of Faisalabad, Punjab, Pakistan. The sampling field was selected based on chronic weed invasions over the last 5 years. The weed species sampled included field bindweed, little seed canary grass, common lambs' quarter, wild oat, and broad-leaved dock. The scientific names of these weeds are *Convolvulus arvensis, Phalaris minor*, *Chenopodium album*, *Avena fatua,* and *Rumex dentatus*, respectively. These samples were transferred to the laboratory in an icebox and stored at 4 ◦C. The rhizosphere soil of these samples was used for the isolation of rhizobacteria using the dilution plating technique. A hundred microliters of each of the serial dilutions (10<sup>−</sup>1–10<sup>−</sup>8) were spread on the sterilized King's B agar media in Petri plates aseptically. This media was prepared by adding 1.5-g K2HPO4, 10 mL glycerol, 20 gm proteose peptone, 1.5 gm MgSO4.7H2O, and 20-g agar and making up the volume of one liter with distilled water following King et al. [21]. The growth of rhizobacterial colonies was obtained after 48 h of incubation of these plates at 28 ± 1 ◦C. The fast-growing colonies were picked and transferred to other Petri plates containing sterilized King's B agar media. These colonies were, hence, purified after some streaking. In this way, 393 strains were purified and preserved at −20 ◦C in 40% glycerol.

### *2.2. Cyanide Production Assay on Strains of Rhizobacteria*

The method given by Bakker and Schipper [22] was followed for the qualitative determination of the production of hydrogen cyanide (HCN) by the isolated strains of rhizobacteria. The pieces of filter paper to the sizes of Petri plates were made, autoclaved for sterilization, and soaked in a 1% solution of picric acid for 12 h. These soaked filter papers were dried aseptically. Glycine amended media was prepared by adding 0.35 gm K2HPO4, 2.5 mL glycerol, 5 gm proteose peptone, 0.35 gm MgSO4·7H2O, 5 gm glycine, and 20-g agar and making up the volume to one liter with distilled water. It gave out quarter strength media with glycine amendment. It was autoclaved and poured in Petri plates. The fresh culture of the strains was used to make a layer on the surface of the media and placing the picric acid-soaked paper on the inner side of the Petri plate lid. The paper was fastened with the help of a 10% solution of Na2CO3. The plates were closed and tightened with parafilm to avoid the leakage of gas. The plates were incubated at 28 ◦C and periodically observed for a change in the color of filter paper. The turning of color to brown indicated the production of HCN, while the intensity of brown color indicated the level of its production (Figure 1).

**Figure 1.** Pictorial view of cyanide production by rhizobacteria.

#### *2.3. Antimetabolite Assay on E. coli*

The bacterial production of toxic metabolites in extracellular spaces can be tested in a simple test based on the growth retardation of sensitive bacteria, *E. coli* [23]. All the strains (393) were tested for HCN production, while this assay was performed on only those strains that produced HCN to any level in step 1. These were 89 strains. Strain K12 of *E. coli* was cultured on LB agar media and placed in an incubator at 28 ◦C. After 2 days, the gentle rubbing of the surface and mixing with sterilized 0.01 M MgSO4 solution formed the culture suspension of *E. coli.* The population of cells of bacteria in the suspension was maintained at 108 cells mL−<sup>1</sup> through the measurement of optical density at 600 nm and the addition of 0.01 M MgSO4 to get the value around 0.55–0.6. A layer of the harvested cell suspension was made on the Petri plates containing sterilized media (King's B). The culture of strains of cyanogenic rhizobacteria was spot inoculated at 3 points of equal distance on the plates pre-inoculated with *E. coli*. The plates were placed in an incubator at <40 ◦C. The production and release of toxic substances by the strains were evident from the zone of clearing around the spot of inoculation of strain. It indicated that the extracellular release of toxic compounds by the strains killed the growth of *E. coli* around its growth. The diameters of the zone of the clearing were recorded.

#### *2.4. Antimetabolite Assay on Lettuce (Lectuca sativa L.) Seedlings*

Nineteen strains restricted the growth of *E. coli* in the previous test. These strains were tested on the seedlings of lettuce as lettuce is considered sensitive to any type of phytotoxic substances and, hence, can be used as an indicator plant [24]. The fresh culture of the selected strains was prepared in Petri plates on KB media. This culture was suspended with the help of a sterilized buffer solution of MgSO4 (0.01 M) by shaking gently. The suspension was collected in test tubes, and the cell population was maintained using optical density measurement at 600 nm with a value of 0.33. It established the population at 10<sup>6</sup> cells mL<sup>−</sup>1.

The seeds of lettuce were disinfected on their surface in a parallel activity. The surface disinfection process comprised of seed dipping in ethanol for a moment, followed by the treatment with sodium hypochlorite (5%) for three minutes and complete rinsing of the seed with autoclaved water [25]. These seeds were allowed to germinate in the growth chamber.

Water agar was used as a medium for the growth of lettuce seedlings, where agar was added into the water at the rate of 1%. It was sterilized and poured in large-sized Petri plates, having a diameter of 15 cm. Seeds with good germination were picked up and transferred to the surface of these plates aseptically. Twenty germinating seeds of lettuce were placed on each plate.

Thirty microliters of the bacterial cell suspension were dispensed to each seed for inoculation. Three Petri plates were prepared for each strain in the same way. The control plates were treated with 30 μL buffer (0.01 M MgSO4) per seed. The plates were placed at ambient temperature in the dark for 4 days. Then, the seedlings were removed from the plates and blotted. The measurements of masses and lengths of roots and shoots were done. The data were analyzed statistically to determine the significant differences [26].

#### *2.5. Antimetabolite Assay on Weeds Using Presumed Allelopathic Bacteria*

The strains of rhizobacteria obtained after the above-mentioned steps of the screening process were now called as presumed allelopathic bacteria. These strains were, now, used for testing on weeds. We selected four weeds of wheat for this assay i.e., wild oat, broad-leaved dock, common lambs' quarter, and little seed canary grass. These weeds cause maximum economic losses in the wheat crop in Pakistan. Nineteen strains were used to conduct this study in an experimental set up similar to the one used for bioassay on lettuce seedlings in Section 2.4. The culture of each strain was prepared in King's B broth. The culture was centrifuged to get the supernatant and form the bacterial pellets. These pellets were mixed in a sterilized buffer (0.01 M MgSO4) to adjust the optical density value of 0.55 at 600 nm. It gave out the bacterial cell population at 108 cells mL<sup>−</sup>1.

Water agar was prepared by adding 10 g of agar in 1 L distilled water and sterilizing in an autoclave at 121 ◦C and 15 PSI pressure for 20 min. The water agar was poured on large-sized Petri plates. It served as a medium for the growth of seedlings (Figure 2).

**Figure 2.** Flow chart of isolation and large-scale screening of allelopathic bacteria for the biocontrol of wheat-associated weeds.

The seeds of the selected weeds were surface disinfected by washing with ethanol (70%) momentarily, followed by washing with sodium hypochlorite (5%) and rinsing of seeds in plenty of sterilized water [25]. These seeds were placed in the growth chamber for germination.

Twenty germinated seeds were placed inside each prepared Petri plates aseptically. The culture suspension of each strain was applied at the rate of 30 μL per seed. For the control treatment, the sterilized buffer (0.01 M MgSO4) was applied at the same rate. The plates were placed at ambient temperature in the dark. Each treatment in the experiment was replicated four times. After 7 days, the seedlings were uprooted from the water agar plates and blotted. These seedlings were measured for the lengths and weights of roots and shoots. The data were analyzed statistically to determine the significant differences following Steel et al. [26].

#### *2.6. Antimetabolite Assay on Wheat Using Presumed Allelopathic Bacteria*

The same nineteen strains were also tested for their effects on the growth of seedlings of wheat in a similar agar bioassay (Figure 3). The culture suspension of the strains was prepared following the same method as above. The large-sized Petri plates containing water agar were prepared as in previous bioassays. The surface of seeds of wheat was disinfected following Abd-Alla et al. [25]. Then, the seeds were placed for germination. The germinated seeds were placed on the already prepared Petri plates aseptically. The culture suspension of each strain was dispensed at the rate of 30 μL per seed. For the control treatment, the sterilized buffer (0.01 M MgSO4) was dispensed to each seed at the rate of 30 μL. Each treatment was replicated four times. The seedlings were uprooted after five days and blotted. The data of lengths and weights of roots and shoots were taken and analyzed statistically to determine the significant differences following Steel et al. [26]. These analyses were carried out using *Statistix 8.1* software. All the data were first subjected to analysis of variance (ANOVA) test in this software, followed by multiple comparisons of means using the linear model. The least significant difference (LSD) test was then applied to determine the significant difference among treatments at *p* < 0.05.

**Figure 3.** Cyanogenic rhizobacteria of weeds and wheat-producing metabolites against *E. coli* in the antimetabolite assay.
