Quantifying Bacterial Surface Swarming Motility on Inducer Gradient Plates

Summary

Here, we describe the use of inducer gradient plates to evaluate bacterial swarming motility while simultaneously obtaining multiple concentration responses.

Abstract

Bacterial swarming motility is a common microbiological phenotype that bacterial communities use to migrate over semisolid surfaces. In investigations of induced swarming motility, specific concentration of an inducer may not be able to report events occurring within the optimal concentration range to elicit the desired responses from a species. Semisolid plates containing multiple concentrations are commonly used to investigate the response within an inducer concentration range. However, separate semisolid plates increase variations in medium viscosity and moisture content within each plate due to nonuniform solidification time.

This paper describes a one-step method to simultaneously test surface swarming motility on a single gradient plate, where the isometrically arranged test wells allow the simultaneous acquisition of multiconcentration responses. In the present work, the surface swarming of Escherichia coli K12 and Pseudomonas aeruginosa PAO1 were evaluated in response to a concentration gradient of inducers such as resveratrol and arabinose. Periodically, the swarm morphologies were imaged using an imaging system to capture the entire surface swarming process.

The quantitative measurement of the swarm morphologies was acquired using ImageJ software, providing analyzable information of the swarm area. This paper presents a simple gradient swarm plate method that provides qualitative and quantitative information about the inducers' effects on surface swarming, which can be extended to study the effects of other inducers on a broader range of motile bacterial species.

Introduction

Bacterial swarming motility refers to the collective migration of bacterial cells across the surface of a substance. In addition to semisolid agar plates specially prepared in the laboratory¹, this phenotype is also observed on some soft substrates such as animal tissues², hydrated surfaces³, and plant roots⁴. While a semisolid surface is considered one of the fundamental conditions for bacterial swarming, some species also require an energy-rich medium to support their swarming motility⁵. Flagella rotation powers both swimming and swarming motility-swimming describes the unicellular motility within a liquid environment, whereas swarming is the synchronous movement of a microbial population across semisolid surfaces.

Substrate viscosity influences bacterial motility; studies of pathogenic microbes, such as Helicobacter pylori, have shown that the pathogen's motility changes depending on the mucin layer viscosity, which is influenced by environmental acidification in the human host⁶. To replicate these environments, earlier studies using agar concentration above 0.3% (w/v) restrict bacterial swimming motility to effect a gradual shift into surface swarming. The use of agar concentration above 1% (w/v) prevents the swarming motility of many species⁷. The colony patterns formed on the surface are diverse, including featureless mat⁸, bull's eye⁹, dendrites¹⁰, and vortex¹¹.

Although the relevance of such patterns remains unclear, those patterns seem to be dependent on environmental and chemical cues¹². Environmental cues cover different aspects, including temperature, salinity, light, and pH, whereas chemical cues include the presence of microbial quorum sensing molecules, biochemical byproducts, and nutrients. Autoinducer quorum sensing signaling molecules such as AHL (N-hexanoyl-L homoserine lactone) can impact surface swarming by regulating the production of surfactant^13,14. Resveratrol, a phytoalexin compound, could restrict bacterial swarming motility¹⁵.

In the present work, we investigate the effect of gradient concentrations of resveratrol on wild-type Escherichia coli K12 strain and investigate arabinose-inducible swarming motility of engineered E. coli K12-YdeH and Pseudomonas aeruginosa PAO1-YdeH species. The production of the YdeH enzyme is induced by arabinose via the araBAD promoter, resulting in cellular c-di-GMP perturbation and affecting bacterial swarming motility^16,17. This inducible swarming behavior is studied using arabinose gradient swarm plates with E. coli K12-YdeH and P. aeruginosa PAO1-YdeH strains.

The gradient swarm plates are prepared by successively solidifying double-layer medium (Figure 1B). The bottom layer comprises the medium added with the inducer, poured on one side of a propped-up Petri dish. Upon the solidification of the bottom layer, the Petri dish is returned to a flat surface, where the upper layer containing the medium without the inducer is added from the other side of the plate. After the swarm plates are completely solidified, isometrically arranged holding-wells are produced by boring holes on the swarm plates following a fixed layout (Figure 1C) or by imprinting the wells using a 3D printed model of the plate cover containing pegs during the medium curing process (Supplemental Figure S1). A gel imaging system is used to capture the swarming morphologies at different time points (Figure 2). Analysis of surface swarming using ImageJ software (Supplemental Figure S2) provides quantitative results of the surface swarming process (Figure 3).

Thus, we propose a simple method to test surface swarming motility within a concentration range of inducers. This method can be used to test multiple concentration responses of other inducers by mixing the inducer into the bottom-layer medium and can be applied to other motile species (e.g., Bacillus subtilis, Salmonella enterica, Proteus mirabilis, Yersinia enterocolitica). This approach could provide reliable qualitative and quantitative results for screening a single chemical inducer, and separate plates may be employed to evaluate more chemical inducers.

Protocol

1. Preparation of gradient swarm plates

1. Preparation of swarm medium
   NOTE: See the discussion section for a brief comparison of different medium viscosities; 0.7% (w/v) agar concentration of swarm medium was used in this protocol.
   1. Prepare Lysogeny broth (LB) powder with agar in two conical flasks; each flask contains 2 g of tryptone, 2 g of sodium chloride, 1 g of yeast extract, and 1.4 g of agar. Add double-distilled water (ddH₂O) and stir the suspension using a magnetic stir bar. Adjust the final volume to 200 mL by adding additional ddH₂O.
   2. Autoclave the solution at 121 °C for 20 min. Use an air-permeable cap or bottle sealing film with an air vent.
      NOTE: Agar will dissolve when heated in the autoclave.
   3. When the temperature drops to 65 °C, mix the solution to ensure homogeneity, and transfer the medium to a 65 °C incubator or water bath for short-term usage.
2. Preparation of bottom-layer swarm medium
   NOTE: The bottom-layer medium is the mixture of swarm medium and inducer stock solutions. The formulation of inducer gradient swarm plates is shown in Table 1.
   1. Prepare a 100 mM resveratrol stock solution by dissolving 114.12 mg of lyophilized resveratrol powder into 5 mL of dimethyl sulfoxide (DMSO), and store the solution at -20 °C.
   2. Prepare a 20% (w/v) arabinose stock solution by dissolving 6 g of arabinose powder in 30 mL of ddH₂O; wait for 10-15 min to allow the arabinose to dissolve; and store the solution at room temperature.
   3. Take out the medium from the 65 °C incubator and place it at room temperature; allow the swarm medium to cool until the Erlenmeyer flask is cool enough to hold (~50 °C). Do not place the swarm medium at room temperature for long periods, as this will cause the solidification of agar.
   4. Add the required volume of the inducer stock solution to the swarm medium at 50 °C (Table 1). Use a pipette to dispense the inducer solution instead of pouring it. Gently swirl to mix the inducer with the swarm medium.
      NOTE: This step is for inducers that cannot be autoclaved. Be careful not to introduce bubbles into the medium.
3. Preparation of double-layer gradient swarm plates
   NOTE: The upper-layer medium is LB medium containing 0.7% (w/v) agar.
   1. Label 13 x 13 cm square Petri dishes with inducer name and strains, and prop the dishes up over the edge of the lids (Figure 1B).
   2. Add 40 mL warm bottom-layer medium (50 °C) using a 50 mL pipette or a 50 mL centrifuge tube.
      NOTE: Alternatively, for 13 x 13 cm square Petri dishes, 40 mL bottom-layer and upper-layer medium is suitable; for 10 x 10 cm square Petri dishes, 25 mL bottom layer and upper-layer medium is suitable.
   3. Allow the bottom-layer medium to cure uncovered for 1 h inside a laminar flow hood. Do not disturb square Petri dishes while the medium solidifies.
      NOTE: While curing the bottom layer, swarm medium not containing inducers should be maintained in a 65 °C incubator or water bath.
   4. Once the bottom layer is completely solidified, remove the lids and lay the square Petri dishes inside a laminar flow hood.
   5. Add 40 mL of warm upper-layer medium (50 °C) using a 50 mL pipette or 50 mL centrifuge tube.
      NOTE: The upper-layer medium does not contain inducers.
   6. Cure the double-layer plates on the benchtop covered and undisturbed for 1 h. Store the prepared plates at 4 °C for up to 24 h.
      ​NOTE: Longer curing times would reduce the moisture content and restrict swarming motility.

2. Growth of E. coli K12 and P. aeruginosa PAO1

1. Prepare 500 mL of LB medium by adding 5 g of tryptone, 5 g of NaCl, and 2.5 g of yeast extract into ddH₂O, and top up the solution to 500 mL. Autoclave the solution on liquid cycle for 20 min at 121 °C, and store it at 4 °C.
2. Prepare 100 mL of 1.5% (w/v) LB-agar medium by adding 1 g of tryptone, 1 g of NaCl, 0.5 g of yeast extract, and 1.5 g of agar into ddH₂O and top up the solution to 100 mL. Autoclave the solution on liquid cycle for 20 min at 121 °C. Transfer the medium to a 50 °C water bath to prevent the agar from solidifying.
3. When the LB-agar medium flask is comfortable to hold, add 20 mL of LB-agar medium into a Petri dish (10 cm in diameter) using a 25 mL pipette. Leave the plate at room temperature overnight, and store the LB-agar plate at 4 °C.
4. Take stock cultures stored at -80 °C, streak E. coli K12, E. coli K12-YdeH, P. aeruginosa PAO1, and P. aeruginosa PAO1-YdeH strains on LB-agar Petri dishes using disposable inoculation loops. Incubate the Petri dishes inverted overnight at 37 °C.
5. Pick single colonies for different strains from the Petri dishes, inoculate each colony into 5 mL of LB medium, and incubate the culture at 37 °C in a laboratory orbital shaker set at 220 rpm.
6. When the culture density reaches OD_600nm ~1.0, remove the culture from the shaker and place it at room temperature. Adjust the culture density to OD_600nm = 1.0, as described in step 3.2.1.

3. Inoculation and incubation of gradient swarm plates

1. Preparation of inoculation wells
   NOTE: 3D printing cover models capable of generating wells separated by a standard distance can be used instead of the method described below (Supplemental Figure S1).
   1. Mark the well positions on A4 paper shown in Figure 1C. Set three test concentrations in one square Petri dish with two or three replicates.
   2. Place the marked A4 paper under a solidified gradient plate. Push the broader side of a 100 µL pipette tip into the semisolid medium surface at the marked position. Press the pipette tip until it reaches the bottom of the upper-layer medium.
   3. When the tip touches the bottom, apply no vertical force to the tip; gently rotate the tip to isolate the content of the cylindrical well.
   4. Horizontally move the pipette tips along a very small distance to allow airflow into the narrow place set aside. Press the tip with the index finger to block the gas flow inside the tip while holding the pipette using the thumb and middle finger.
   5. Pull the tip out vertically, keeping the well content in the tip while pulling it out.
      NOTE: If the well content slips, apply slightly more pressure with the index finger to completely seal the tip.
   6. Repeat steps 3.1.2 to 3.1.5 in every marked position. Cover the swarm plate before inoculation.
2. Gradient plate inoculation and incubation
   1. Adjust the overnight growth culture density to OD_600nm = 1.0.
   2. Pipette 80 µL of the overnight growth culture into every well. Do not spill the bacterial culture outside the wells.
   3. Wrap the plates with sealing film. For long-term observation (3-5 days), wrap the plates with sterile laboratory rubber tape.
      NOTE: Rubber tape is less likely to break.
   4. Place a beaker filled with ddH₂O in the incubator to maintain humidity inside the incubator. Incubate the gradient swarm plates at 37 °C.
      NOTE: Do not incubate the swarm plates upside-down; this will cause the bacterial culture to leak from the wells.
   5. Image the swarm plate immediately after inoculation, recording this as the 0 h time point.

4. Imaging bacterial surface swarming

1. Take the swarm plates out, one at a time, from the incubator every 12 h, holding the plate horizontally, and place them in the gel imaging system (see the Table of Materials).
   NOTE: Do not leave fingerprints on the surface of plates; hold the side of the swarm plate with clean gloves.
2. Select gel imaging mode; expose the swarm plate to white light; and adjust the focal length to give the clearest view of swarms.
   NOTE: Use the same focal length for all plates in a given batch.
3. Enhance the brightness of the swarms for clear observation by adjusting the exposure time to 300 ms. Adjust the threshold to minimize interference from the background light.
   NOTE: Threshold is adjusted on the operating interface of the gel imaging system. Increase values on the left to minimize interference from background light; decrease values on the right to enhance the brightness of the swarms. In this protocol, the region is usually between 6,000 and 50,000.
4. Save the image file for further analysis. Record the imaging time, inducer type, gradient orientation, and strains in a .txt file.

5. Quantify the swarm area using ImageJ software

1. Import the image file acquired using the gel imaging system.
2. Set the scale bar using ImageJ software and apply it to all the images.
   1. Create a line segment marking the length of the board by clicking on the line tool.
   2. Click Analyze | Set Scale to open the Set Scale window.
   3. Type the actual length in Known distance and Unit of length.
      NOTE: Because 13 x 13 cm square Petri dishes were used in this work, the actual length is '130', and the unit of length is 'mm.'
   4. Check the Global box.
   5. Insert a scale bar by clicking Analyze | Tools | Scale Bar, type Width in mm, Height in pixels, Font size, and select Color, Background, and Location from the dropdown menu. Alternatively, choose Bold Text, Hide Text, Serif Font, and Overlay by ticking the checkboxes.
      NOTE: The choice of those parameters is determined by users and the properties of the images. In this protocol, Width in mm was set to 25, Height in pixels to 20, Font size to 80, and the scale bar was placed in the lower right corner by selecting Location | Lower Right. Other parameters can be chosen by the user.
3. Click Process | Shadows to enhance the sharpness of the image, especially the boundaries (Supplemental Figure S2A). Click Process | Batch to process images.
   NOTE: The purpose of this step is to provide more precise boundary demarcation.
   1. Process an image as a reference by clicking on Process | Shadows | South.
   2. Click Process | Batch | Macro to open the Batch Process window. Look for the following commands displayed in the window:
      run("South");
      run("Save");
      ​close();
   3. Type the folder address of the original images and the output file address by clicking Process in the Batch Process window.
      NOTE: It is recommended to export images with shadows to another folder and retain a copy of the original images.
4. Use Wand (tracing) tool to select swarms individually and adjust the tolerance (double-click Wand (tracing) tool) until the generated line fits the swarm boundary correctly (Supplemental Figure S2B).
   NOTE: First, click Wand (tracing) tool and select a swarm on one image. If the boundaries have not been depicted correctly, double-click the Wand (tracing) tool to open the Wand Tool windows, where the Tolerance can be adjusted.
5. Click on Analyze | Measure to export the area value.
6. Repeat steps 5.1.1-5.1.5 until all swarms are measured, save the results to a .csv file for further analysis.

Representative Results

The workflow consisting of the preparation of gradient swarm plates, inoculation, and incubation is shown in Figure 1B. To generate gradient swam plates, the bottom-layer medium is poured into propped-up dishes at ~4.3° from the horizontal plane (Supplemental Figure S3), followed by pouring the upper-layer medium after the bottom layer is completely solidified. The composition of the double-layer medium is shown in Table 1. Then, bacterial culture cultured overnight is pipetted into the test wells and incubated at 37 °C, maintaining appropriate levels of humidity. Multiple test concentrations are set in one gradient swarm plate with two or three replicates (Figure 1C). The alternative option of a 3D printing model of the cover lid with the columnar protrusion on the test points is shown in Supplemental Figure S1.

Two engineered species, E. coli K12-YdeH and P. aeruginosa PAO1-YdeH, were tested on arabinose gradient plates. Figure 2A shows the surface swarming process test in five wells, with overlap occurring between adjacent wells. Three test wells were set in one plate, as shown in Figure 2B,C, which enabled the formation of nonoverlapping boundaries. Bacterial swarming motility was promoted with an increase in the arabinose concentration from the lowest concentration but was gradually restricted with higher arabinose concentrations. E. coli K12 wild-type strain was tested on resveratrol gradient plates (Figure 2C), within the concentration range of 0-400 µM. A modest restriction of the swarming motility was observed with increasing resveratrol concentration. Swarm areas were quantified by ImageJ software (Supplemental Figure S2). The swarming curves display the multiple concentration responses (Figure 3).

Figure 1: Schematic of inducer gradient swarm plate preparation, inoculation, and incubation. (A) Overnight growth of bacterial culture at 37 °C. (B) Workflow of double-layer inducer gradient swarm plate. i) Prop up square Petri dishes. ii) Pour bottom-layer medium and solidify at room temperature. iii) Lay the square Petri dishes flat, and pour upper-layer medium. iv) Plate curing. v) Make wells corresponding to the marked positions. vi) Pipette bacterial culture into wells. vii) Wrap plates using sealing film or rubber tape. viii) Place swarm plates in a 37 °C incubator. C) Sketch map of wells on A4 paper or 3D printing model. Three wells with I) three replicates and II) two replicates. Please click here to view a larger version of this figure.

Figure 2: Surface swarming processes on inducer gradient plates. Bacterial surface swarming processes are captured using a gel imaging system within 5 days of inoculation. (A) Arabinose induced surface swarming process of E. coli K12-YdeH. (B) Arabinose induced surface swarming process of P. aeruginosa PAO1-YdeH. (C) Surface swarming process of E. coli K12 wild-type strain on resveratrol gradient plate. Please click here to view a larger version of this figure.

Figure 3: Surface swarming curves represent multiple concentration responses on inducer gradient plates. Every quantifiable data point consists of three replicates. (A) Arabinose-induced swarming motility of P. aeruginosa PAO1-YdeH; approximate concentrations within test wells are 0.17% (w/v), 0.25% (w/v) and 0.42% (w/v). (B) E. coli K12 wild-type strain swarming on resveratrol gradient swarm plate; approximate concentrations within test wells are 67 µM, 200 µM, and 335 µM. Please click here to view a larger version of this figure.

Upper layer medium/lysogeny broth medium (per 100 mL)
Tryptone: 1 g
Sodium chloride: 1 g
Yeast extract: 0.5 g
Agar: 0.7 g
Bottom layer medium/Inducer-containing medium (per 100 mL)
Tryptone: 1 g	Working concentration:	Stock solution concentration:
Sodium chloride: 1 g
Yeast extract: 0.5 g	- Resveratrol: 400 μM	- Resveratrol: 100 mM
Agar: 0.7 g
Inducer:	- Arabinose: 0.5% (w/v)	- Arabinose: 20% (w/v)
- Resveratrol stock solution: 400 μL	or 1% (w/v)
- Arabinose stock solution: 2.5 mL or 5 mL

Table 1: Double-layer swarm medium specifications.

Supplemental Figure S1: 3D printing models of swarm plate lid. (A) 3 x 3 wells including three wells and three replicates. (B) 3 x 2 wells including three wells and two replicates. (C) Make wells with 3D printing models. Please click here to download this File.

Supplemental Figure S2: Quantification of surface swarm area using ImageJ software. (A) Add 'shadows' to original images (Process | Shadows) to generate quantifiable swarms with clear boundaries. (B) Set scale bar (Analyze | Set Scale), select swarms using Wand  (tracing) Tools, and export swarm area (Analyze | Measure). Please click here to download this File.

Supplemental Figure S3: Propped-up square Petri dish for pouring bottom-layer medium. The angle of inclination is ~4.3 °. Please click here to download this File.

Discussion

Investigation of bacterial swarming motility on semisolid gradient plates can be a challenging task^18,19,20, as it involves multiple factors such as substrate viscosity, humidity, and medium components. Among these factors, agar concentration plays a decisive role in determining microbial reversion to either swimming or swarming motility. As the agar concentration increases from 0.3% (w/v) to 1% (w/v), bacterial swimming motility will be restricted and gradually shift to surface swarming, and agar concentration above 1% (w/v) will prohibit both swimming and swarming motility⁷. The agar concentration was fixed at 0.7% (w/v) based on preliminary experiments, as it showed better performance than other concentrations.

This agar concentration was also previously employed to study microbial chemotaxis²¹. Decreased agar concentration results in a larger swarm area, accompanied by overlaps between adjacent wells, increasing the difficulty of quantifying swarm area due to unclear boundary demarcation. Relatively high agar concentration results in a small swarm area, decreasing the possibility of overlaps. However, excessive agar (>1.0%) can prevent bacterial surface swarming. Hence, it is essential to select an appropriate agar concentration that can be applied to all test species to generate comparable results with a standard viscosity.

Isometrically arranged wells provide equal and appropriate space for bacteria to swarm. The arrangement of wells can be different depending on the needs of gradient swarm plates. Surface swarm can overlap due to insufficient distance between test wells (Figure 2A), hindering the quantification of the swarm area, especially in a prolonged study. Three test wells were set within one swarm plate to test multiple concentration responses while preventing colony overlap (Figure 2B,C).

Compared to an inoculating needle²², holding-wells prepared in these semisolid plates provide a standardized inoculation volume. Holding-wells were also found to sustain the bacterial culture, preventing bacterial spread observed in other methods such as pipetting ²³. Holding-wells made by inserting 3D print models present clearer and standardized inoculation start sites. Although these wells require additional preparation, they reduce the deviation between the test points by marking inoculation positions on the template. Additionally, the exact bacterial counts in each holding-well can be calculated, improving the reproducibility of the data. As a precaution, hasty or careless preparation of wells could result in the cracking of the wells, resulting in variation of the surface swarming during migration, as the microbes are inclined to move through the cracks.

Care must be taken to avoid splattering of the microbial starter culture during the preparation of the loading well and sample loading, especially in low-viscosity swarm plates. Further, the volume of the bacterial culture loaded must be optimized to minimize the time needed for the microbes to scale the inner walls of each holding-well while preventing the possibility of spillage during plate transportation (for imaging purposes). In this work, we decreased the bacterial culture loading volume to minimize the possibility of spillage, resulting in a delay in bacterial migration that is commonly mistaken for swarm lag²⁴.

It is necessary to adapt the medium formulation due to the differences in viscosity and required nutrient sources between motile species. For some species (e.g., Bacillus subtilis²⁵), surface swarming can be rapid; therefore, the imaging interval should be shortened. Computer-assisted swarm area quantification gives more precise information than distance measurement of radius²⁶. To generate clear boundaries for calculating swarm area by ImageJ software, a built-in method was used in this protocol that adds shadows to the original images acquired using the gel imaging system. If a boundary of the swarm merges with the adjacent one, migration toward the border will be inhibited, where these communities prefer to migrate toward the unoccupied areas of the swarming plates (Figure 2A). These limiting factors resulting from the overlapping boundaries present a significant challenge in determining the migratory distance of the surface swarming, where these values cannot be integrated into the free swarming process.

Materials

Name	Company	Catalog Number	Comments
Agar	Sigma-Aldrich	V900500	500 g
Ampicillin	Solarbio	A8180	25 g, ≥ 85% (GC)
Centrifuge tube	Corning	430790	15 mL
Cryogenic vial	Corning	430488	2 mL
Dimethyl sulfoxide (DMSO)	Aladdin	D103272	AR, > 99% (GC)
L(+)-Arabinose	Aladdin	A106195	98% (GC), 500 g
Petri dishes	Bkman	B-SLPYM90-15	Plastic Petri dishes,circular,90 mm x 15 mm
Resveratrol	Aladdin	R107315	99% (GC), 25 g
Sodium chloride	Macklin	S805275	AR, 99.5% (GC), 500 g
Square Petri dishes	Bkman	B-SLPYM130F	Plastic Petri dishes, square, 13 mm x 13 mm
Tryptone	Thermo Scientific Oxoid	LP0042	500 g
Yeast extract	Thermo Scientific Oxoid	LP0021	500 g
Equipments
Biochemical incubator	Blue pard	LRH-70
Tanon 5200multi imaging system	Tanon	5200CE
Thermostatic water bath	Jinghong	DK-S28