The soils were collected from various gardens in Iran, placed in sterile bags, and transported to the laboratory, where they were refrigerated at 4°C for further experimentation.

To begin with, various soil samples were cultured on a Nutrient Agar medium to isolate bacteria. Subsequently, a pure bacterial culture was suspended in 20 mL of modified M9 medium broth containing 0.1% methionine. This solution was then incubated at 30°C for 72 h at 120 rpm.

L-methioninas activity evaluated by Nessler’s method (Wriston and Yellin, 1973). 72-h bacterial cultures were centrifuged at 20124 RCF for 10 min, and a clear supernatant was kept for enzyme assay. 100 μL of supernatant transported to the sterile 1.5 mL microtube with 900 μL 0.5%W/V L-methionine (substrate) in Tris–HCl buffer (50 mM and pH 7.2). After incubation at 30°C for 60 min, 100 μL 15% trichloroacetic acid was added for stop the reaction and centrifuged at 8944 RCF for 10 min. The ammonia liberation in the supernatant was determined using the colorimetric technique by adding 200 Nessler regent in 200 μL supernatant and distilled water. The sample was first vortexed for 15 s and incubated at room temperature for 10 min, and finally, the absorption (OD) was measured at 425 nm. The ammonia released in the reaction was calculated based on the standard curve obtained using ammonium chloride. One unit of L-methioninase activity is defined as the amount of enzyme that releases 1 μL of ammonia per minute at 37°C.

The morphological, physiological, and biochemical characterizations of the best L-methionine-producing isolates (according to the highest yield of enzyme production at the screening stage) were determined by identification tests according to Bergey’s manual of systematic bacteriology (Sneath et al., 1986). For further identification, the analysis of the 16 s rDNA gene sequence using universal bacterial primers, 5′-CAGCCGCGGTAATAC-3′ as the forward primer and 5′-ACGGGCGGTGTGTAC-3′ as the reverse primer. Trench sequencing was employed to analyze 16S rDNA sequences After that, the phylogenic method for identifying bacterial isolates was used using the UPGMA algorithm tree in the Clustal W server.

The optimization of medium constituents for L-methioninase production by the selected bacterium was performed in two steps. MINITAB software conducted all experimental designs and data analysis (version 20, PA, United States).

The bacteria were cultured in a modified M9 medium, and the screening of bacteria produces L-methionine. We cultured 16 samples from different soil sources in broth-modified M9 medium for 72 h. The experiments were repeated three times, and the average values were used. The results show that we found three isolates that can produce L-methionine, and the B5s9 isolate had higher activity than the other isolate. B5s9 isolate could be considered an excellent L-methioninase producer.

To identify the isolated strain, the results demonstrated that the strain B5s8 was a gram-negative motile bacillus in addition to indole production; however, this isolate could not ferment all the tested carbohydrates. In parallel, the widely used method of 16 S rRNA gene analysis was also applied, demonstrating that the almost complete sequence of 16 S rRNA gene had 95% homology with Pseudomonas mosselii strain MN02. Moreover, the phylogenetic tree was drawn (Figure 1). Therefore, according to the observations and analyses, the strain MN02 was identified as Pseudomonas mosselii. Another strain that we placed as producing the L-methioninase includes B2s8 and NB7s3.B2S8 was a gram-negative andNB7s3 was gram-positive, and we used 16 S rRNA gene was applied, however after a complete sequence of 16 S rRNA gene, respectively, had homology with Ralstonia solanacearum strain MN02 and Cytobacillus kochii strain MN02 and drowned phylogenetic tree (Figures 2, 3).

L-methininase productions were assessed at various pH to discover the influence of pH. Maximum of L-methininase production of 1.5 ± 0.1 U/mL was obtained at a pH of 6. A decrease in the production of L-methioninase is observed in pH’s lower than 5 and higher than 8. Therefore, the optimum pH for Pseudomonas mosselii to produce L-methioninase has attained a pH of 6 (Figure 4A). In this regard, some scientists reported the best pH for L-methioninas production in other strains. Selim et al. reported that the best pH range for L-methioninase producing in Streptomyces sp. was 7–7.5 (Selim et al., 2015), and Mokham et al. reported that the best pH range for L-methioninase producing in Alcaligenes when increased pH from 4 to 6 (Mohkam et al., 2020). To study the influence of temperature for maximum production of L-methioninase was 32°C and L-methioninase production was 2.0 ± 0.1 U/mL. When we increased the temperature to more than 37°C, we could see an apparent decrease in enzyme production (Figure 4B).

Among various nitrogen sources, we tested two different types of nitrogen sources, including inorganic and organic. The review and result of this are in Figure 5A. We found that the best nitrogen source for Pseudomonas mosselii to produce L-methioninase was yeast extract at 1% concentration, and MN02 can produce 3.2 ± 0.1 U/mL L-methioninas. After the yeast extract, the tryptone was placed at a concentration of 0.5%. On the other hand, we can see lower production of the enzyme in (NH4)2SO4; it can be shown that producing of L-methioninase in Pseudomonas mosselii, was not dependent on L-methionine which means that these bacteria can produce the enzyme in the absent of L-methionine (Singh and Kharayat, 2018). Some other bacteria identified in the past, like these bacteria, were not dependent on L-methionine for enzyme production, like Alcaligenes (Mohkam et al., 2020). However, fungi can produce L-methininase when L-methionine is in their culture. This fungi like Trichoderma harzianum and Aspergillus spp. (Salim et al., 2019). Without needing L-methioninase, this feature helps us produce L-methioninas from Pseudomonas mosselii, a cheaper nitrogen source. The properties of P. mosselii, including its faster growth and production compared to fungi, simpler cultivation conditions that are typically easier and more cost-effective, easier genetic engineering capabilities, and fewer side effects, produce L-methioninase without L-methionine, make it a more suitable choice for producing the enzyme L-methioninase.

The influence of different carbon sources, glucose was the best carbon source for L-methioninase production (3.25 ± 0.1 U/mL), followed by 0.5% concentration (Figure 5B). Like other studies that have been done before, it shows that in lack of carbon source, L-methioninase production was inhibited. This is significant for bacteria like Alcaligenes and Acromobacter starkeyi. Despite our findings, some bacteria produce more L-methioninanase with lactose source, which is probably related to the β-galactosidase gene in those bacteria like Alcaligenes (Mohkam et al., 2020).

We tested four measure conditions such as pH, temperature, glucose, and yeast extract, and we tried to optimize these conditions to the optimum. We got help from Box–Behnken design and applied many conditions tabulated in Table 1. The variance analysis (ANOVA) result is tabulated in Table 2. The variable is considered significant when the p value is lower than 0.05. ANOVA showed the significance of all four factors (pH, temperature, glucose, and yeast extract) in producing L-methioninase. However, the interactive effects had no significant impact on L-methioninase production. We analyzed the data in multiple regression and found the model for predicting L-methioninase.

X1, X2, X3, and X4 represent the coded values of pH, temperature, glucose (%), and yeast extract (%), respectively.

According to analysis, the significant p value (<0.001) and high F value showed that the model equation was in excellent agreement with the experimental. The coefficient of determination (R²) was 0.852 for L-methioninase production, which describes a higher correlation between experimental results and their predictions. The closer R² is to 1.0, the more robust the model is and the better its prediction.

After that, the surface plot was immersed in different conditions to determine the best conditions for L-methioninase production (Figures 6A–F). To identify the optimized values of the parameters, we created plots for the pairwise combination of parameters while keeping the other parameters constant at their middle level. Although the response surface plots helped evaluate the parameters, assessing them all simultaneously took work. Hence, we used Minitab’s Response Surface Optimizer function to determine the optimal levels of yeast extract (2.0%), glucose (2.0%), pH (Sahoo and Sahoo, 2021), and temperature (30°C). The model has predicted that L-methioninase will have an activity of 12.56 U/mL. The excellent correlation between the forecasted and examined values of the Box–Behnken design (as presented in Table 2) confirms the validity of the response model and the existence of an optimal point. In a similar study, Kharayat et al. used the RSM technique to optimize L-methioninase production in Bacillus subtilis, achieving 17.4 U/mL under optimized conditions. According to Sundar et al., the optimized conditions from Serratia marcescens resulted in an optimum activity of 13.7 U/mL, matching with Mokhom et al. from Alcaligenes (Mohkam et al., 2020).

We conducted several experiments to examine various factors affecting the production of L-methioninase. Due to the complexity of the process and the large number of experiments, we performed half of the experiments under different conditions. Our experiments found that the highest production of L-methioninase occurred when yeast extract was used at a concentration of 1.5%, glucose at 2.0% at a temperature of 30°C, and a pH level of 6. We used a Multiple Regression software called MiniTab to determine a model for predicting enzyme activity under different conidiation conditions. The model was highly accurate (86.63%) in our experiments. We then utilized this model to identify the optimal conidiation conditions for producing enzymes from Pseudomonas mosselii. According to the model, the best conidiation conditions are pH = 6, temperature = 30°C, glucose = 2%, and yeast extract = 2%, which produced 12.5670 U/mL of L-methioninase.