Using bacterial consortia is a sustainable and promising tool to improve the sustainability of fertile land while achieving high production rates (Duncker et al., 2021; Seenivasagan and Babalola, 2021). Combining two or more strains with plant growth-promoting (PGP) traits to enhance the inoculum’s efficacy is a proven method (Deter and Lu, 2022). The bacterial consortium consisting of Herbaspirillum seropedicae Z67, Gluconacetobacter diazotrophicus Pal5, Azospirillum brasilense Cd, and Burkholderia ambifaria PHP7 has been shown to have positive biostimulating effects on various crops, including tomato, carrots, hemp, Apennines genepì, and ancient Triticum genotypes (Del Gallo et al., 2010; Botta et al., 2013; Pagnani et al., 2018; Pagnani et al., 2020; Pellegrini et al., 2020a; Pellegrini et al., 2021a; Pellegrini et al., 2021b).

The favorable results achieved with this consortium are due to the synergy of various plant-growth-promoting qualities found in these strains. A. brasilense is a well-known species of PGPB and has great potential for biofertilization (Bashan and De-Bashan, 2010). Azospirillum spp. and Gluconacetobacter spp. are known to enhance plant growth through various processes. Azospirillum spp. are involved in phytohormone synthesis, N₂-fixation, mineral mobilization, synthesis of small molecules and enzymes, root system proliferation, increased membrane activity, and mineral and water uptake (Bashan, 1986; James et al., 2001; Logeshwarn et al., 2011; Brígido et al., 2015; Suleman et al., 2018; Silva et al., 2019). Gluconacetobacter spp. are also linked to N₂-fixation, synthesis of siderophores and phytohormones, and the solubilization of inorganic forms of zinc and phosphorus (James et al., 2001). Herbaspirillum spp. and Burkholderia species assist in plant growth by producing phytohormones, synthesizing siderophores, and N_2-fixing (Olivares et al., 1997; Tawfik et al., 2010; Xu et al., 2020). They also reduce stressful ethylene levels through 1-aminocyclopropane-1-carboxylate (ACC) deaminase. Furthermore, the biocontrol potential against bacterial and fungal phytopathogens has been documented for each strain (Zahir et al., 2009; Djebaili et al., 2021).

The manufacturing process of this consortium involves growing pure cultures separately and then combining them in the appropriate ratio (Bashan et al., 2014). Researchers and enterprises commonly use this technique. However, this procedure increases operational and production costs by considering multiple strains in the formulation and their individual fermentations. Application of multi-strain fermentation techniques has gained relevance in recent years in an effort to lower total bioprocess costs (Jangra et al., 2016). The co-culture systems could overcome the limitations of monocultures or consortia with the added advantages of exploring allelopathic interactions in several biotechnological applications due to their versatility, robustness, and complexity (Hays et al., 2015). Furthermore, several studies underlined how the complexity of co-culturing fermentation could increase its resilience against contamination (Hathi et al., 2021). By creating a co-culture fermentation, costs can be saved by eliminating the need for several processes. Creating multistrain inoculants with the appropriate ratios of each strain is a complex process that requires extensive research (Hussain et al., 2022; Sobhi et al., 2023). Previous attempts at mixed-culture bioprocesses have utilized this technique to reduce production costs (Kapoore et al., 2022). However, microbial strains interact in various ways, such as mutualism, competition, parasitism, commensalism, and predation (Wu et al., 2023). Therefore, understanding the population dynamics and interactions between microbial strains during growth is crucial for co-culture fermentation development.

This work aimed to develop a biomass-producing process that incorporates a co-culture of the four mentioned PGPB strains. The co-fermentation was developed on the T4 medium, previously designed to maximize the four-strain growth. The co-fermentation procedure was studied in 500 mL flasks. Subsequently, the fermentation process was successfully scaled up to a 5-L bioreactor.

Trails and evaluations were repeated five times (n = 5 independent experiments) and values were expressed as mean ± standard deviations. After collection, the dataset was analyzed for outlier’s presence and distribution. Results of the densities obtained from the co-culture experiments, with no normal distribution, were analyzed by Kruskal-Wallis, followed by Conovan-Iman multiple comparisons of the mean. 5L bioreactor data, with a normal distribution, were analyzed by ANOVA, followed by Tukey’s post-hoc test. Data processing was carried out using XLSTAT 2016 software.

The four-strain growth behavior in the T4 medium was studied to keep the goal of a combined process. Figure 1 displays the growth cultures obtained for each strain. A first comparison shows that the growth curves for A. brasilense and H. seropedicae have very similar kinetics, whereas B. ambifaria has a more rapid growth. For G. diazotrophicus the growth is slower. In particular, A. brasilense and H. seropediace start the exponential phase after a lag phase of 5 hours, which lasts up to 24 hours. B. ambifaria starts the exponential phase after 2 hours and lasts up to 12 hours. G. diazotrophicus has a slow adaptation phase and starts the exponential phase after 12 hours, that lasts up to 48 hours. For all the strains, the maximum density recorded was 10 Log CFU mL^-1.

Analyzing this behavior, it was hypothesized that co-cultures could only be grown by scalar inoculations. Therefore, three different experiments of scalar additions were set up (Experiments B-D). A control co-culture was established as control by adding all strains at time 0 (Experiment A). Figure 2 shows the results obtained for each experiment and the statistical analysis performed for comparison. From the counts performed for each strain, it was observed that the control culture only allowed the growth of the B. ambifaria strain up to 9.5 Log CFU mL^-1 (after 12 hours), which appeared to significantly inhibit the growth of the other three strains (which obtained a maximum density of 10⁷ CFU mL^-1). In Experiment B and Experiment C good densities were obtained for G. diazotrophicus and B. ambifaria (9 Log CFU mL^-1). However, the other two strains were negatively influenced by these set up. The best results were obtained for Experiment D and at the end of the co-fermentation each strain obtained a final density of 10 Log CFU mL^-1 on average.

Experiment D was also scaled up to a 5 L bioreactor. Figure 3 shows the strain’s densities obtained after 80 hours of co-culture fermentation. The strains achieved a similar final density, comparable to that obtained with the 500 mL flask scale (9.8 Log CFU mL^-1). No significant changes in the culture medium were registered during the fermentations.

In the first part of this work, we explored the growth behavior of four Gram-negative bacteria. The observed growth rates are in line with the nature of each bacteria. Studying a single fermentation process is crucial to comprehensively characterize the strain’s bacteriological characteristics and metabolic demands. The strains analyzed in this work have been the subject of numerous studies over the years, which have determined their metabolic demands for optimal development (Eskew et al., 1977; Baldani et al., 1986; Stephan et al., 1991; Coenye et al., 2001; Canini et al., 2006). In addition, Botta et al (Botta et al., 2013). developed a culture medium suitable for all four bacteria. These findings are of great importance for experimental research with potential industrial applications. These aspects may influence a company’s decision to scale up a formulation for industrial use.

After completing the initial stage, it is recommended to carefully analyze the growth process and consider potential bioprocesses that may help reduce production costs. This analysis involves evaluating nutrient levels and the duration of the bioprocess. Moreover, the possible interactions among bacteria must be considered (Wu et al., 2023). Given the growth rates and interactions of these strains with other living organisms, adding the strains in a scalar manner was suggested, beginning with the slowest strain, G. diazotrophicus, and concluding with the one with the highest competitive potential, B. ambifaria. The slow growth of G. diazotrophicus is usually reported. Compared with other bacteria with similar sugar assimilation pathways, it is usually described as slower (Restrepo et al., 2023). In fact, when the cultures were added during G. diazotrophicus growth, this resulted in lower charges, probably given to competition behaviors established among strains. B. ambifaria, instead, is a fast-growing bacterium with competitive behavior by secondary metabolites production (e.g., antibiotics and other secondary metabolites) (Pal et al., 2022). This aspect could be considered as one of the most relevant limits of co-culturing fermentation, mainly related to the competition for nutrients and the production of secondary compounds that interfere with the growth of other organisms (Santos et al., 2014). The different trials showed a good compatibility of A. brasilense and H. seropedicae, which grew together without any interference. This compatibility is probably due to the A. brasilense ability of to easily adapt its metabolism and organization of the cells to respond to environmental changes (Bible et al., 2015). This ability could be essential to facilitate the creation of effective synergism and growth rated of the strains during the early exponential phase. Thus, understanding the ideal combination of growth medium, culture condition, and fermentation technique is essential for promoting development, stability, and preservation of functional activity through downstream procedures and storage (Rodrigues et al., 2006).

Microorganisms interact with each other in various ways. These interactions can be positive, negative, or neutral in simple ecological models. Any close, long-term biological interaction between two different biological organisms, either positive or negative, is called symbiosis (Hector et al., 2022). In both symbioses and transitional associations, interactions between two units can be classified as mutualism, commensalism, amensalism, competition, and predation/parasitism, depending on the directionality of the interaction (Wu et al., 2023). Interactions can become more complex and highly dynamic in systems with multiple units due to interspecific, intraspecific, and environmental factors. As the number of entities involved increases, community complexity also increases due to the combined growth in the number and type of pairwise interactions (Duncker et al., 2021). A third unit can alter the strength or nature of a pairwise association, which a fourth unit can further modify. These interactions, along with their types and intensity, can significantly impact a community’s stability and can be used to predict community composition over time (Bairey et al., 2016).

It is widely recognized that co-culturing microorganisms with different metabolic capabilities can have additive or synergistic interactions, with beneficial effects on the microbial community and, directly or indirectly, on plant growth and protection. For example, Pagnussat et al. (2016) found that the amount of biofilm produced by Azospirillum brasilense Sp245 significantly increased in co-culture with Pseudomonas protegens CHA0 (Pagnussat et al., 2016). Few authors focused on optimizing growth conditions of PGPBs strains in co-culture. Bagheri et al. (2022) studied the interactions between Azospirillum oryzae NBT506 and Bacillus velezensis UTB96 in a co-culture system. Using the adjusted population of each strain (8 Log CFU mL^-1 of A. oryzae NBT506 and 6 Log CFU mL^-1 of B. velezensis UTB96), they observed the maximum growth efficiency (11 Log CFU mL^-1) based on different preculture concentrations for both strains (Bagheri et al., 2022). Similar results were obtained by Masciarelli and collaborators, who tested Bradyrhizobium japonicum E109 strain and Bacillus amyloliquefaciens in a co-fermentation to enhance soybean nodulation (Masciarelli et al., 2014). These findings underline the bacterial activity and ability to increase their growth in an optimized co-culture bioreaction. Creating multi-microbial formulations can be valuable, providing economic and industrial benefits for microbial-based inoculants. This strategy can lead to a reduction in the number of bioprocesses required, resulting in lower material and energy costs and increased efficiency.

The selection of a co-culture is an important factor in optimizing economic resources for product development. It might be useful to enhance biomass yield in several biotechnological processes such as fermentation, biofuels, bioremediation, nutraceutical, and chemical production (Kapoore et al., 2022). Controlled co-culturing enables the synergistic utilization of metabolic pathways of participating microorganisms under industrial, reproducible, and controlled conditions. Significant energy and resource savings can be achieved during industrial production by defining optimal values for process parameters, substrate, and product contents (Bader et al., 2010). The ability to produce multiple strains in a single bioreaction process can result in lower process waste production and limit the need for multiple culture broths to obtain biomass. This approach can lead to less economic effort in management and a lower environmental impact of industrial production. By using fewer resources and producing less waste, higher environmental sustainability for the farm can be achieved (Rosero-Chasoy et al., 2021).

Our results provide a solid basis for developing a product that includes these microorganisms and pave the way for further research. As far as we know, no other studies have co-formulated a four multistrain consortium formed by Gram-negative PGPB. Given the large global market for these products and their increasing use, research in this area is very important (Pellegrini et al., 2020b; Chaudhary et al., 2023). Microbial fermentation is of utmost importance in the development of next-generation bioinoculants. Recent research breakthroughs have provided effective methods for optimizing the fermentation of endophytic microbes, from fundamental to scaling-up production. This progress represents cutting-edge agricultural, food, and nutrition technology, ushering in a new scientific era (Ganeshan et al., 2021). However, there is a lack of integration between scientific results and the industrial sector (Singh et al., 2024). Therefore, such studies should be encouraged. This initial research report sets the stage for future evaluations of any product on an industrial and commercial level, including bioprocessing and technoeconomic feasibility analysis.

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

BF: Data curation, Formal Analysis, Investigation, Visualization, Writing – original draft. FP: Conceptualization, Funding acquisition, Resources, Supervision, Writing – original draft. RD: Formal Analysis, Investigation, Writing – original draft. DS: Conceptualization, Methodology, Validation, Writing – original draft. MD: Methodology, Validation, Writing – review & editing. MP: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – review & editing.