Phenotypic profiling of complex soil communities and bacterial isolates with RhizoPlate N and P

Phenotypic profiling of complex soil communities
and bacterial isolates with RhizoPlate N and P

Measuring the capacity of microbes to fix nitrogen, or convert gaseous nitrogen into ammonia, has historically been achieved using indirect techniques such as the acetylene reduction assay (ARA). The reduction of acetylene by nitrogen-fixing bacteria requires the same nitrogenase enzyme needed to reduce atmospheric nitrogen to ammonia, only in this reaction, acetylene is reduced to ethylene. The conversion of acetylene to ethylene can be quantified using gas chromatography. While this method approximates the relative ability to fix nitrogen, it is not a direct measurement. ARA assays also require the incubation of plates in bags containing flammable gas which must be periodically sampled over time in a labor-intensive process to monitor nitrogenase activity. Biolog’s RhizoPlate N
enables direct probing of nitrogen fixation ability of microbes, as atmospheric nitrogen is the only available source of nitrogen in the assay. The plate contains 31 pre-selected carbon substrates in triplicate which are commonly available within the rhizosphere and which have been shown to influence the ability to fix nitrogen (Hameeda 2006 and Charyulu 1981). Single nitrogen-fixing isolates or complex communities can be inoculated onto RhizoPlate N and their growth monitored to directly measure the ability to fix atmospheric nitrogen, as only those organisms with nitrogenase activity will
be able to grow.

Inorganic phosphate solubilization is another critical function of rhizosphere bacteria as it is the process by which phosphorus can be made available for uptake by roots. Assessing the ability of microbes to perform this function has traditionally been done through the use of the Pikovskaya (PVK) medium which is an agar containing insoluble phosphate. Bacteria are plated and incubated on the PVK medium and monitored for the formation of a zone of clearing, indicating the solubilization of the inorganic phosphate which is often accomplished through the secretion of acid. While simple to perform, this assay is inherently qualitative as it depends on the passive diffusion of compounds through the agar matrix which can be confounded by bacterial motility and other factors. Biolog RhizoPlate P is similar to RhizoPlate N where 30 relevant carbon substrates are included in triplicate, however instead of lacking nitrogen, the plate contains only pre-plated insoluble tri-calcium phosphate which is commonly found in soil. The only possibility for growth in the well by bacteria is through the solubilization of the insoluble phosphate.

In this study we used RhizoPlate N and RhizoPlate P
(Figure 1) to assess the relative nitrogen fixation and phosphate solubilization ability of the microbial communities present in the rhizosphere in three different soil samples. Soil 1, leaf litter-fertilized planted soil, Soil 2, a commercial garden soil, and
Soil 3, unplanted soil, were all compared to single isolate control strains. Average well growth development across each RhizoPlate was used to compare and rank each of the soil communities and control strains for their relative ability to fix nitrogen and solubilize inorganic phosphate. Using this method, we determined that the diverse microbial communities tested had a relatively higher phenotypic output than purified strains, likely due
to combinatorial effects.

Methods:

Soil Sample Preparation
Soil samples were collected from 2-4 inches below the surface level in sterile 50 mL conical tubes.
5 g of dry soil for each type were transferred to a fresh sterile 50 mL conical tube and sterile 0.8% NaCl solution was added to a final volume of 40 mL.
Samples were shaken horizontally at 120 rpm for 30 minutes to dislodge microbes and allowed to settle vertically for 30 minutes. The supernatant was transferred to a fresh sterile 50 mL conical tube and fresh 0.8% NaCl was added to achieve a volume of 40 mL, and the tubes were spun down at 500 x g for 10 minutes to remove heavier clay and silt particulates. 35 mL of supernatant was transferred to a fresh sterile 50 mL conical tube and spun down at 4000 x g for 10 minutes to pellet the microbes present in solution. The supernatant was removed, and the pellets resuspended in 35 mL 0.8% NaCl; three 10 mL aliquots were transferred into fresh sterile 15 mL conical tubes and spun down at 4000 x g for 10 minutes and the supernatant was discarded. One aliquot for each sample was resuspended in
15 mL IF-RN and the other in IF-RP inoculation fluids while the third aliquot was resuspended in 0.8% NaCl, spun down and supernatants were discarded to remove any excess nitrogen and phosphorus before resuspending one final time in 15 mL fresh IF-RN, IF-RP, and 0.8% NaCl for a final cell density of <95% T. 100 µL of the IF-RN, IF-RP, and 0.8% NaCl soil microbe suspensions were inoculated onto one RhizoPlate N, RhizoPlate P, and EcoPlate respectively per sample. Plates were incubated in the Odin™ platform at 26 °C for 100 hours and automatically read every 20 minutes at 590 nm.

Single Strain Sample Preparation
Cultures of Azospirillium brasilense (positive control for nitrogen fixation), Acinetobacter pittii and Pseudomonas putida (positive controls for phosphate solubilization) and Staphylococcus epidermidis and Escherichia coli (negative controls for both assays) were grown on Biolog Universal Growth medium + 5% sheep’s blood (BUG+B) at 30°C for 48 hours until colonies formed. Each of the 5 control strains were inoculated onto RhizoPlate N and RhizoPlate P according to standard Biolog protocols. In brief, cells were transferred from the agar plate to 10 sterile glass test tubes; 5 containing IF-RN and 5 containing IF- RP until a density of 80% T was reached. IF-RN and IF-RP suspensions were inoculated onto RhizoPlate N and RhizoPlate P respectively at 100 µL per well. The plates were transferred to Odin for incubation at 26 °C for
100 hours and read automatically every 20 minutes at 590 nm.

Data Analysis
RhizoPlate N and RhizoPlate P data were analyzed using the Odin software Community Analysis tool to rank samples by average well growth development (AWGD) for all of the different substrates on the plate (Sofo 2019). Background subtraction was applied to each sample and AWGD was calculated and plotted over time. The second inflection point of each curve, indicative of the beginning of growth during nitrogen fixation or phosphate solubilization was determined, and the latest inflection point of 60 hours was used as the x-axis minimum for all samples. Data were normalized to eliminate background growth signal attributed to nutrient carryover from stored nitrogen and phosphorus so that only growth during nitrogen fixation or phosphate solubilization phases was considered. Samples were then ranked according to their final AWGD absorbance level. EcoPlate data were analyzed using the Odin Software Community Analysis tool to aggregate replicates for each substrate and generate kinetics curves and calculate Maximum Curve Height (MCH) for each substrate. Growth curves and MCH for each substrate were compared for Soils 1, 2, & 3 to determine unique substrate utilization patterns among the soil samples.

Unique Substrate Use – EcoPlate
	Can’t use	Can use
Soil 1	Pyruvic acid methyl ester Phenylethylamine	–
Soil 2	α-D-Lactose	α-Cyclodextrin Glycogen L-Threonine α-Ketobutyric acid
Soil 3	D-xylose	–

Unique Substrate Use – RhizoPlate P
	Can’t use	Can use
Soil 1	D-Gluconic acid α-Keto-valeric acid	L-Arabinose
Soil 2	D-Mannose D-Galactose L-Phenylalanine L-Proline	–
Soil 3	Sucrose	–

Table 1. Carbon Source Utilization Profiles for EcoPlate and RhizoPlate P

Unique Substrate Use – Soil 1 vs Soil 3 – EcoPlate
Soil 1	D-Xylose
Soil 3	Pyruvic acid methyl ester

Unique Substrate Use – Soil 1 vs Soil 3 – RhizoPlate P
Soil 1	L-Arabinose D-Mannitol Sucrose
Soil 3	Acetic acid Citric acid

Table 2. Carbon Source Utilization Profiles for EcoPlate and RhizoPlate P

Results

We began by challenging three different soil samples for their ability to utilize a variety of carbon sources included on the EcoPlate, RhizoPlate P, and RhizoPlate N. This was accomplished by extracting microbial communities from the soil samples, growing them on each type of plate, and monitoring growth over time. Each of the three soil samples’ microbial communities were readily able to metabolize a variety of carbon sources even when restricted for the availability of phosphate. Interestingly, however, only one carbon substrate, glycogen, was sufficient to drive growth on RhizoPlate N. Comparing each of the three soil samples to one another, we found that Soil 2, the commercial garden soil, had the highest number of uniquely usable carbon sources as found in the EcoPlate including α-cylcodextrin, glycogen, L-threonine, and α-ketobutyric acid (Table 1). Taken together, with the RhizoPlate N result, this suggests that there might be a relationship between glycogen metabolism and the availability of nitrogen. We then compared Soil 1 (leaf litter fertilized) and Soil 3 (unplanted soil) and found that there are carbon sources which are uniquely utilized as well (Table 2). Soil 3 was more effective at metabolizing organic acids (citric and acetic acids) while Soil 1 showed higher growth on sucrose, D-mannitol, and L-arabinose in phosphate limiting conditions on RhizoPlate P (Table 2).

We then compared the three different soil samples alongside known nitrogen fixing and phosphate solubilizing species of bacteria for their ability to fix atmospheric nitrogen and solubilize inorganic phosphate. With the array of 30 different compounds provided in triplicate in each well, the RhizoPlates simultaneously probe the ability of each community to grow in the presence of a range of carbon sources. The average well growth development (AWGD) provides a plate-wide parameter for comparing the functional diversity of each of the communities. After normalization to eliminate growth signal due to reserves of nitrogen and phosphorus stored in the cells, we found that each of the soil samples and positive control strains demonstrated some ability to grow using atmospheric nitrogen and tricalcium phosphate as the sole sources of nitrogen and phosphorus respectively (Figures 2 and 3). For both nitrogen fixation (Figure 2) and phosphate solubilization (Figure 3), all three soil samples ranked well above the positive controls with Soil 2, the commercial garden soil, ranking highest in both assays. 

Figure 2. RhizoPlate N AWGD curves show the highest growth in Soil 2 (commercial garden soil) followed by Soils 1 and 3 (planted and unplanted soil respectively) and the A. brasilense positive control. 

Figure 3. RhizoPlate P AWGD curves show the highest growth in Soil 2 (commercial garden soil) followed by Soils 1 and 3 (planted and unplanted soil respectively) and the P. putida and A. pitti positive controls. Interestingly, A. brasilense demonstrated similar growth to the positive controls indicating additional acidification of the media.

It is likely that the soil samples were ranked the highest due to the diverse communities present. While the cell titer for any given species within the community was likely relatively low compared to the pure cultures, the composite effect of the community was enough to outpace a single species’ ability to grow under the RhizoPlate conditions. This demonstrates the importance of having a healthy, mixed community of microbes present in the soil to efficiently cycle critical nutrients such as nitrogen and phosphorus. Another factor which may have increased the AWGD value for the soil samples relative to the purified isolates was the increase in diversity of carbon sources utilized by the soil community due to the larger pool of microbes. The commercial garden soil, Soil 2, was the best performer of the three soils for both phenotypes. The reduced relative growth capacity in the nitrogen and phosphorus-limited conditions of the planted soil and unplanted soil (Soils 1 and 3) can likely be attributed to lower availability of decaying plant material and moisture commonly found in commercial garden soils which are designed to
foster rapid and robust growth of plants. 

Conclusions

Through the use of the Odin platform, RhizoPlates, EcoPlates, and the Community Analysis software package, we were able to effectively and efficiently rank soil samples based on their relative ability to fix nitrogen and solubilize inorganic phosphate and found that the commercial garden soil had the most robust phenotypic output in both cases. This work also demonstrates that having a complex community of microbes within the rhizosphere is critical to the turnover of nutrients essential to plant growth, whereas single isolates were less effective likely due to less flexibility in their metabolic profile relative to a vibrant community. Finally, the combination of Odin and RhizoPlates significantly reduces hands-on time and provides relative quantitation of nitrogen fixing and phosphate solubilization ability for many samples in parallel without continuous user input or complex sample preparation.

References
Hameeda, B., Reddy, YH., Rupela, K., Reddy,G. Effect of carbon substrates on rock phosphate solubilization by bacteria from composts and macrofauna. Curr.Microbiol. 53, (2006)

Charyulu, P., Rajaramamohan, V. Influence of carbon substrates and moisture regime on nitrogen fixation in paddy soils. Soil Biol. Biochem. 39, (1981)

Sofo, A.; Ricciuti, P. A Standardized Method for Estimating the Functional Diversity of Soil Bacterial Community by Biolog® EcoPlates™ Assay—The Case Study of a Sustainable Olive Orchard. Appl. Sci. 9, 4035, (2019). https://doi.org/10.3390/app9194035