Design of a culture medium for optimal growth of the bacterium Pseudoxanthomonas indica H32 allowing its production as biopesticide and biofertilizer

Culture medium composition is one of the most important parameters to analyze in biotechnological processes with industrial purposes. The aim of this study was to design of a culture medium for optimal growth of the bacterium Pseudoxanthomonas indica H32 allowing its production as biopesticide and biofertilizer. The influence of several carbon and nitrogen sources and their molar ratios on P. indica H32 growth was investigated. The effect of different micronutrients such as mineral salts and vitamin on P. indica H32 growth was determined as well. A mixture design based on Design-Expert 10.0 Software was performed to optimize the culture medium concentration. Finally, in the designed medium, an attribute of the biological mechanism of action of the P. indica H32 against nematodes, was evaluated: the hydrogen sulfide production. It was found that tested carbon/nitrogen ratios were not a significant influence on P. indica H32 growth. Growth of P. indica H32 was favored with use of sucrose, yeast extract and phosphate buffer without the addition of any tested micronutrients. An optimal concentration of 10 g/L sucrose and 5 g/L yeast extract were obtained at a cost of 0.10 $/L. In this concentration, the specific growth rate (µ) and maximal optical density (Xmax) were equal to 0.439 h− 1 and 8.00 respectively. It was evidenced that under the culture conditions used, P. indica H32 produced hydrogen sulfide. The designed medium led to a 1.08 $/L reduction of costs in comparison to LB medium. These results were critical to carry on with biotechnological development of P. indica H32 as a bioproduct.


Introduction
Plant parasitic nematodes attack almost all major crops worldwide, causing damages of over US$100 billion annually (Shah et al. 2017). Although reports about nematode-related losses in Cuba are scarce, a wide range of crops are known to have been attacked by these microorganisms, namely sugar cane, tobacco, and coffee (Díaz-Silveira and Herrera 1998). Today, the utilization of biological controls to fight plant nematodes is an attractive alternative to conventional chemical pesticides due to their persistence in the soil and proven environmentally-friendly features (Tranier et al. 2014). Previous studies, not yet published, indicate that the strain Pseudoxanthomonas indica H32 is the active component for a promising bioproduct candidate, with the capacity to control plant-parasitic nematodes, to control fungi, and promote plant growth (Lugo et al. 2019).
Pseudoxanthomonas indica, was isolated and incubated in Luria-Bertani (LB) medium by (Kumari et al. 2011). However, until now there is not a culture medium for optimal growth of the bacterium P. indica H32 allowing its production as biopesticide and biofertilizer. Other species of this genus have been isolated and studied, which has allowed us to know about the physiology and nutritional requirements of these bacteria (Jackson et al. 1998;Mohapatra et al. 2018;Nayaka et al. 2019). Pseudoxanthomonas indica does not hydrolyze urea as nitrogen source and assimilates sucrose and glucose as carbon sources (Kumari et al. 2011), so peptone and yeast extract contained in LB medium could be suitable nitrogen sources for the P. indica H32 growth. In addition, the LB medium contains a total mass concentration of carbon and nitrogen supplements equal to 15 g/L (10 g/L tryptone and 5 g/L yeast extract) (Protocols 2006). Therefore, this would be a criterion to consider as a starting point to design a suitable culture medium that allows optimal growth of P. indica H32.
The study of composition of culture medium for growth of P. indica H32 is critical during its biotechnological development stage, since the culture medium accounts for 30-40% of estimated production costs (Batista and Fernandes 2015). Besides, determining the nutritional, energetic, and environmental requirements of microorganisms is a critical step to develop bioprocesses (Rajendran and Thangavelu 2007). The nutritional, cellular, and biochemical environment of a bioreactor is strongly affected by the conditions of the culture and the composition of the medium, namely the carbon and nitrogen sources, mineral salts, oligoelements, peptides, amino acids, and vitamins (Sampaio et al. 2010). Culture media ensure similar nutritional conditions to the naturally existing ones, allowing for proper functioning of cell metabolism, which means an adequate balance of components (Gómez and Batista 2006).
Several methods can be used to optimize culture medium composition, including the single-factor method, though it is time-consuming, painstaking, and does not guarantee optimal condition determination (Batista and Fernandes 2015;Eswari et al. 2012;Venkata et al. 2006). Likewise, experiments based on all the possible combinations of test factors are impractical, since a large number of experiments are required (Eswari et al. 2012). One choice to increase product yields, cut down culture time and variability, as well as costs, is the application of statistical experimental design techniques (Batista and Fernandes 2015;Mu et al. 2009). Therefore, the main objective of this study was to design a culture medium for optimal growth of the bacterium P. indica H32 allowing its production as biopesticide and biofertilizer, through application of statistical experimental design techniques.

Bacterial strain
Pseudoxanthomonas indica H32 (Culture Collection of Microorganisms of Center of Genetic Engineering and Biotechnology (CCCIEB), Cuba, 771) stored at − 70 °C in 20% (v/v) glycerol was used in this study.

Media screening experiments Evaluation of different carbon and nitrogen sources, as well as their molar ratios on P. indica H32 growth
The initial composition of culture media was based on the M9 mineral medium (Geerlof 2010) (without ZnCl 2 and CuCl 2 ), which was supplemented with different carbon and nitrogen sources to study its influences on P. indica H32 growth. Moreover, the molar ratios of carbon (C) and nitrogen (N) sources were tested same time (C/N equal to 5, 10 and 20). Ammonium chloride (Scharlab, Spain), yeast extract (Angel, China), and bacteriological peptone z (BioCem, Cuba) were tested as nitrogen sources, whereas sucrose (TECNOAZUCAR, Cuba) and glucose (Scharlab, Spain) were evaluated as carbon sources. In the experiment, the independent variables were C, N, and the molar ratios, whereas specific growth rate (µ) and maximum optical density (Xmax) were dependent variables. The experimental design was composed by 18 point (Table 1), which were made by triplicate (n = 3). In the culture medium, the sum of the mass concentrations of the carbon and nitrogen sources studied was fixed at 15 g/L (C + N = 15), taking into consideration the composition of the LB medium (Protocols 2006).

Study of the influence of mineral salts and vitamins on P. indica H32 growth
Seven culture media were employed to study the influence of mineral salts and vitamins present in the M9 medium on P. indica H32 growth. According to results of the first experiment, each culture medium contained as base 7.93 g/L sucrose, 7.07 g/L yeast extract and 100 mL/L phosphate buffer (47.8 g/L Na 2 HPO 4 and 30 g/L KH 2 PO 4 , adjusted to pH 7.2 with NaOH and sterilized for 15 min at 121 °C). This way, the medium 1 was the base of the other six media, which were composed by 2: medium 1 + CaCl 2 , 3: medium 1 + MgSO 4 , 4: medium 1 + NaCl, 5: medium 1 + trace salts solution, 6: medium 1 + vitamins and 7: medium 1 + CaCl 2 + MgSO 4 + NaCl + trace salts solution + vitamins. The micronutrients were adding on equal proportion that on M9 medium (Geerlof 2010).

Optimization of culture medium concentration
An optimal mixture-design based on Design-Expert 10.0 Software was used to optimize culture medium concentration. The dependent variables were µ (h − 1 ), Xmax and unitary cost ($/L). The independent variables were sucrose and yeast extract concentration (0-15 g/L). The design was composed by a total of 17 experimental runs.

Culturing conditions
From the cryopreserved culture of P. indica H32, 25 µL were taken and added to test tubes containing 5 mL of LB medium (previously sterilized at 121 °C during 15 min). It was incubated at 37 °C and 250 rpm on a tube rotator (Boekel Scientific Tube Spinner, USA) during 18-20 h. After this time, the OD was determined to be able to inoculate 50 mL of culture medium contained in 250 mL Erlenmeyer shake starting off initial OD equal to 0.1. Subsequently, the culture was incubated at 37 °C and 250 rpm during 52 h in a shaker incubator (New Brunswick G 25, USA). The samples were taken at 0, 4, 8, 24, 28, 32, 48, and 52 h. Bacterial growth was quantified by spectrophotometry at 530 nm (Biochrom Libra S80 spectrophotometer, UK). Absorbance was read by triplicate in each sample and OD was calculated by multiplying the absorbance value by the dilution factor. Finally, the cell growth curves were built, and µ was determined according to Raina et al. (2009).

Assessment of the hydrogen sulfide production in the designed medium as an attribute of the biological mechanism of action of the P. indica H32 against nematodes
The method to evaluate the hydrogen sulfide production was adapted from García, Vázquez and Campos (2003) and Padilla et al. (2017): In sterile 24-well plates (Nunc, USA), 100 µL of culture at 24 h of growing were applied by triplicate in the wells of plate destined for the samples. Then, 100 µL of 15 mmol/L cysteine (Sigma-Aldrich, USA) solution were added and mixed gently with circular movements of the plate. Subsequently, a filter paper was moistened with 10 µL of 0.1 mol/L Lead (II) acetate (Sigma-Aldrich, USA) solution in the center of each circle Table 1 Experimental design for the evaluation of different carbon and nitrogen sources, as well as their molar ratios on P. indica H32 growth * All design points was contained M9 mineral medium as basis (Geerlof 2010)   marked with the plate cover, to evidence the presence of the color intensity on the filter paper produced by the formation of lead sulfide. The plate was capped placing a filter paper on the cover and incubated at 37 °C during 20 h in an incubation room. After this time, the plate was carefully uncapped to stop the reaction and the results were interpreted. The culture medium was used as negative control and Brevibacterium celere C924, active component of the HeberNem ® registered biological product for the nematode control and plant growth promoting, was used as positive control. The positive and negative controls were processed in the same way as the culture samples.

Statistical analysis
STAT GRA PHICS Centurion XVI.I was used for analysis of variance (ANOVA, α = 0.05) of all data gathered during the evaluation of the carbon and nitrogen sources tested and for study of the influence of M9 medium micronutrients on P. indica H32 growth. Furthermore, Software Design-Expert 10.0.1 was used for optimization of medium concentration, regression analysis, ANOVA of experimental data (α = 0.05), evaluation of goodness-of-fit, predictive model accuracy, equation of the mathematical model, R 2 correlation coefficients, and adjusted R 2 .

Influence of carbon and nitrogen sources and their molar ratios on P. indica H32 growth
The carbon and nitrogen sources were selected according to the nutritional requisites of the microorganism, including availability and cost of the same. The experimental results showed that the carbon sources tested had a significant influence on µ (P = 0.0163), but not so on Xmax (P = 0.6315). With the use of sucrose and glucose as carbon sources, the µ was equal to 0.357 h − 1 and 0.321 h − 1 respectively. On the other hand, the nitrogen sources had a significant effect on µ (P = 0.0000) and Xmax (P = 0.0084). The µ was equal to 0.431 h − 1 and 0.441 h − 1 with use of yeast extract and bacteriological peptone respectively, without differences between their homogeneous groups. However, with the use of NH 4 Cl were obtained the lower values of µ (0.147 h − 1 ). The highest Xmax values were also obtained with the use of yeast extract and bacteriological peptone (9.35 and 8.58 respectively). Nevertheless, tested molar ratios were not a significant influence in the analyzed response variables (P = 0.8640 and P = 0.2192). Therefore, only factors with a significant influence (P ≤ 0.05) were analyzed through of the diagrams of Means and 95.0 Percent LSD intervals (Fig. 1). Consequently, the carbon and nitrogen sources chosen for the composition of the culture medium were sucrose and yeast extract, respectively.

Influence of mineral salts and vitamins on P. indica H32 growth
The growth of P. indica H32 was influenced significantly by the composition of the culture medium. This was corroborated through ANOVA, in which the components had a significant effect on µ and Xmax, (P = 0.0000 and P = 0.0019). Each component was analyzed individually in the diagrams of Means and 95.0 Percent LSD intervals (Fig. 2). The values observed for µ were between 0.440 and 0.462 h − 1 except with the medium 4 (µ = 0.418 h − 1 ) (Fig. 2a). The higher Xmax values were 11.12, 11.0 and 10.93 with medium 1, medium 2 and medium 7 respectively (Fig. 2b). Medium 1 proved has only has three components (sucrose, yeast extract and phosphate buffer) allowing the simplification of the scale up fermentation process without compromising P. indica H32 growth, which constitutes an advantage from the technicaleconomic point of view. Therefore, the medium 1 was chosen to carry on with culture medium design, being named as H medium.

Optimization of culture medium concentration
Optimization of the culture medium concentration (H medium) was made with aim to increase the specific growth rate (µ) and maximum optical density (Xmax), and to minimize the culture medium cost. The experiment results showed that with only yeast extract it was sufficient to increase the P. indica H32 productivity (Table 2). Nevertheless, since yeast extract was the most expensive component of the culture medium, this influenced the optimization criteria. Thus, the Xmax values raised were more adjusted to a quartic mixture model (R 2 = 95.32%, and adjusted R 2 = 93.76%). Factors A, B, AB, AB (A-B), AB (A-B) 2 were considered significant terms of the model (P < 0.05) ( Table 3). The mathematical model equation obtained for Xmax was: Likewise, the µ values achieved, fit a cubic mixture model better with R 2 = 99.21%, and adjusted R 2 = 99.02%. Factors A, B, AB, AB (A-B), were considered significant terms of the model (Table 4). The mathematical model equation obtained for µ was: The ANOVA of cost variable, manifest that only factors A and B were significant terms of the model since ( Table 5). The cost values obtained, fit a linear mixture order model, with R 2 = 100% and adjusted R 2 = 100%. The mathematical model equation obtained for cost was: The solution of the optimization for the H medium concentration had a desirability of 0.656. The optimal concentration predicted was 9.28 g/L of sucrose and 5.72 g/L of yeast extract, respectively (Fig. 3). For validation, 10 g/L of sucrose and 5 g/L of yeast extract were used. The observed value was between the percentage of interval (PI) predicted with 95% confidence ( Table 6).
The H medium is 11.84-fold less costly than the LB medium (Table 7). Consistently, similar and higher values of µ and Xmax than LB medium were achieved, which demonstrated the suitability of the culture medium designed for the P. indica H32 growth and scale up fermentation process. Fig. 2 Influence of mineral salts and vitamins on P. indica H32 growth. a Values of µ obtained with the tested media. b Values of Xmax obtained with the tested media. 1: medium 1 (7.73 g/L sucrose, 7.07 g/L yeast extract and 100 mL/L phosphate buffer). 2: medium 1 + CaCl 2 . 3: medium 1 + MgSO 4 . 4: medium 1 + NaCl. 5: medium 1 + trace salts solution. 6: medium 1 + vitamins. 7: medium 1 + CaCl 2 + MgSO 4 + NaCl + trace salts solution + vitamins. All bars represent the mean values ± 95.0 percent LSD intervals Hydrogen sulfide production in the designed medium as an attribute of the biological mechanism of action of the P. indica H32 against nematodes As a result of this experiment, the production of hydrogen sulfide by the bacterium P. indica H32 was evidenced, through the qualitative determination of the dark spots on the filter paper generated by lead sulfide. The intensity of the spots on the filter paper corresponding to the production of hydrogen sulfide by P. indica H32 were higher than the samples of the negative control (H medium) and similar to the samples of the positive control (Brevibacterium celere C924) (Fig. 4).

Discussion
Process optimization is an essential procedure in biotechnology industry to guarantee higher yield with a minor cost. The design and optimization of medium nutritional parameters such as composition and nutrients ratios, is rather effective to improve fermentation process. Therefore, the use of cheap and available sources of nitrogen and carbon is desirable for the optimization of fermentation as this can significantly reduce the cost of production. This study was conducted to develop a culture medium for optimal growth of the bacterium P. indica H32 allowing its production as biopesticide and biofertilizer.
The influence of different carbon and nitrogen sources on P. indica H32 growth was study. Regarding the effect of organic nitrogen sources, yeast extract and bacteriological peptone, exhibited the most increase of µ. The effect exerted by ammonium chloride on µ was negligible in comparison with yeast extract and bacteriological peptone. The unlike on µ according the use of different nitrogen sources can be explained by their chemical composition. Yeast extract and bacteriological peptone have amino acids and protein, among other micronutrients (Podpora et al. 2016), which are essential for growth, whereas ammonium chloride mostly provide nitrogen.
Meanwhile, the evaluated carbon sources also had significant influence on µ of P. indica H32, although the use of sucrose as carbon source allowed reaching a value of µ of 1.11 times higher than using glucose. Other studies about of growth requirements for production of the bioherbicidal by bacterium Xanthomonas campestristhe (phylogenetically related genus with the P. indica H32) show that the substitution of sucrose with glucose did not significantly alter cell densities (Jackson et al. 1998). Many bacteria possess catalytic machinery to metabolize sucrose and the genes comprising sucrose catabolic operons are strictly regulated (Saier et al. 1995). The fact that the higher value of µ was achieved with sucrose (being glucose in most of cases a preferential carbon source), suggests that P. indica H32 possess a functional and regulable system for sucrose uptake and catabolism, capable of using sucrose efficiently when it is present.
The evaluation of mineral salts and vitamins on P. indica H32 growth demonstrate that only with carbon, nitrogen and phosphate buffer supplementation in the culture medium (medium 1, named as H medium) is sufficient to increase P. indica H32 growth, without addition  of any micronutrient. Similar results were found by Jackson et al. (1998), who showed that omission of the vitamin mixture in the culture medium did not significantly alter cell densities. They report that the presence of sucrose or glucose as the carbon source and various organic nitrogen sources in the culture medium, supported optimal X. campestris growth and cell yield.