Methods for Characterizing the Co-development of Biofilm and Habitat Heterogeneity

Summary

Biofilms have complex interactions with their surrounding environment. To comprehensively investigate biofilm-environment interactions, we present here a series of methods to create heterogeneous chemical environment for biofilm development, to quantify local flow velocity, and to analyze mass transport in and around biofilm colonies.

Abstract

Biofilms are surface-attached microbial communities that have complex structures and produce significant spatial heterogeneities. Biofilm development is strongly regulated by the surrounding flow and nutritional environment. Biofilm growth also increases the heterogeneity of the local microenvironment by generating complex flow fields and solute transport patterns. To investigate the development of heterogeneity in biofilms and interactions between biofilms and their local micro-habitat, we grew mono-species biofilms of Pseudomonas aeruginosa and dual-species biofilms of P. aeruginosa and Escherichia coli under nutritional gradients in a microfluidic flow cell. We provide detailed protocols for creating nutrient gradients within the flow cell and for growing and visualizing biofilm development under these conditions. We also present protocols for a series of optical methods to quantify spatial patterns in biofilm structure, flow distributions over biofilms, and mass transport around and within biofilm colonies. These methods support comprehensive investigations of the co-development of biofilm and habitat heterogeneity.

Introduction

Microorganisms attach to surfaces and form biofilms — cell aggregates enclosed in an extracellular-polymer matrix¹. Biofilms behave very differently from individual microbial cells, because biofilms have dramatic spatial heterogeneity resulting from a combination of internal solute transport limitations and spatial variations in cellular metabolism^2,3. Oxygen and nutrient concentrations drastically decrease at the interface between biofilm and surrounding fluid and get further depleted within in the biofilm². Spatial variations in biofilm respiration and protein synthesis can also occur as a response to localized oxygen and nutrient availability².

In aquatic and soil environments, most bacteria dwell in biofilms. Natural biofilms carry out important biogeochemical processes including cycling carbon and nitrogen and reducing metals^4,5. Clinically, biofilm formation is responsible for prolonged pulmonary and urinary infections⁶. Biofilm-associated infections are highly problematic because cells in biofilms have extremely high resistance to antimicrobials compared to their planktonic counterparts⁶. Because biofilms are important in diverse settings, a substantial amount of research has been focused on understanding the environmental factors that control biofilm activities and the spatial heterogeneity in biofilms and the surrounding microenvironment.

Previous studies have found that biofilm development is strongly regulated by a number of environmental factors: biofilms develop different morphologies under various flow conditions; oxygen and nutrient availability influence biofilm morphology; and hydrodynamic shear stress affects the attachment of planktonic cells to surfaces and the detachment of cells from biofilms^7-9. Furthermore, external flow condition influences the delivery of substrates into and within biofilms¹⁰. The growth of biofilms also alters surrounding physical and chemical conditions. For example, biofilm growth leads to local depletion of oxygen and nutrients²; biofilms accumulate inorganic and organic compounds from the surrounding environment¹¹; and biofilm clusters divert flow and increase surface friction^12,13. Because biofilms interact with their surrounding environment in very complex ways, it is critical to simultaneously obtain information on biofilm properties and environmental conditions, and multi-disciplinary approaches need to be used to comprehensively characterize biofilm-environment interactions.

Here we present a series of integrated methods to characterize spatial patterns in microbial growth within mono-species and dual-species biofilms under an imposed nutritional gradient, and to observe the resulting modification of the local chemical and fluid microenvironment. We first describe the use of a recently developed double-inlet microfluidic flow cell to observe biofilm growth under well-defined chemical gradients. We then demonstrate the use of this microfluidic flow cell to observe the growth of two species of bacteria, Pseudomonas aeruginosa and Escherichia coli, in biofilms under a range of nutritional conditions. We show how in situ visualization of fluorescent tracer propagation into biofilm colonies can be used to quantitatively assess patterns of solute transport in biofilms. Finally, we show how microscale particle tracking velocimetry, performed under confocal microscopy, can be used to obtain local flow field around the growing biofilms.

Protocol

1. Flow Cell Setup and Inoculation

NOTE: Use a double-inlet microfluidic flow cell described in Song et al., 2014¹⁴ to grow biofilms. This flow cell is able to create well-defined smooth chemical gradients. The flow cell design is shown in Figure 1 and flow cell fabrication was previously described in Song et al., 2014¹⁴. Here we detail our methods by using P. aeruginosa and E. coli to form biofilms, but other species may be suitable too. We used P. aeruginosa PAO1-gfp, which constitutively expresses a green fluorescent protein (GFP), as a model organism for biofilm development. Additionally, we used E. coli DH5α to form mixed-species biofilms with P. aeruginosa. P. aeruginosa and E. coli strains were grown on LB agar plates.

1. Prepare the flow cell using polydimethylsiloxane (PDMS) bound to a glass coverslip via oxygen plasma treatment as described in¹⁴. The dimensions of the flow cell chamber are 23 mm × 13 mm × 0.24 mm (length × width × depth).
2. Prepare modified FAB growth medium¹⁵. To observe biofilm growth under a nutritional gradient within the flow cell, introduce modified FAB medium¹⁵ with 0.6 μM glucose in one inlet and introduce FAB medium without any carbon source through the other inlet. Filter-sterilize (pore size = 0.2 μm) the glucose stock solution (60 mM) before adding it to the FAB medium. Also autoclave the FAB medium with cycle 1 (liquid, 15 min; 121 °C, 17 psi) before use.
3. Sterilize the flow system. Prior to inoculation, autoclave the entire flow path (medium bottles, tubing, bubble traps, flow cells) using cycle 1, except for the plastic three-way valves (disposable and pre-sterilized) located upstream of the flow cell. (Three-ways valves are used for injecting cell culture, fluorescent tracer and microbeads.) To avoid contamination during assembly, cover all the tubing and connector openings with aluminum foil or autoclave bags before autoclaving.
4. Connect the flow system. Assemble the components of the flow cell system carefully (see video for flow system assembly) and deliver growth medium to the flow cell via a peristaltic pump that precisely controls the flow rate.
5. Prepare cell culture for inoculation by transferring a colony of P. aeruginosa or E. coli from LB plates to 3 ml of LB broth and shake the culture O/N at 225 rpm and 37 °C. Dilute the O/N cell cultures in 1 ml sterilized water to a final OD₆₀₀ = 0.01 as the inoculum. (Culture and dilute the bacteria in a laminar flow hood to avoid contamination.)
6. For the experiments with mixed-species biofilms, dilute the two bacterial cultures to a ratio of 1:1 in 1 ml with an equivalent OD₆₀₀ = 0.01 for each bacterium.
7. Inoculate the flow cell. Inject 1 ml of the inoculum to the flow cell inlet from the three-way valve. After the injection, pause the flow for 1 hr to allow bacterial cells to attach to the cover glass. (Make sure that the flow cell is placed with the coverslip side down to allow suspended cells to settle onto the coverslip.)
8. After 1 hr, resume the flow and pump the growth medium to the flow cell at a constant rate of 0.03 ml/min for each inlet for 3 days.

2. Characterizing Biofilm Development in Response to Nutrient Gradients Using Confocal Microscopy

NOTE: P. aeruginosa can be imaged with constitutively-expressed GFP, but E. coli in mixed-species biofilms must be imaged by counterstaining.

1. Observe the 3-day biofilms using confocal microscopy. For the representative results, observe biofilms using a confocal microscope with a 63X objective. Prior to imaging, mark the double-inlet flow cell with a grid on the cover glass side. (The purpose of this grid is to allow the experimenter to locate the imaging regions within the flow cell chamber.)
2. To counterstain, dilute 10 μl of cell-permeant red fluorescent nucleic acid stain, such as green fluorescent nucleic acid stain such as STYO 62, stock solution (1 mM) in 990 μl sterilized water and then slowly inject the diluted stain using a syringe into the flow cell chamber from the upstream three-way valve. Keep the flow cell stagnant and in the dark for 30 min, then resume the flow to wash out the unbound stain for 15 min.
3. Perform confocal imaging. Activate the 488 nm Argon laser for excitation and the emission collecting channels for GFP (500-535 nm) and nucleic acid stain (for green fluorescent nucleic acid stain, 650-750 nm). Adjust gains and offsets for both channels to have bright and clean images. Select xyz and simultaneous imaging mode. Use the z-control knob to identify the bottom and top of the biofilm in the field. Set the z-step to 0.5 μm for acquiring a 3-D image stack. (To ensure high image quality, use a line average of four for image acquisition.)
4. To map the spatial patterns of biofilm development within the flow cell, image biofilms at three or more longitudinal (x) distances along the flow cell inlet, and at three transverse (y) positions relative to the imposed nutritional gradient, as shown in Figure 1.

3. Characterizing Internal Solute Transport by Injections of Conservative Fluorescent Tracer

NOTE: A conservative fluorescent tracer, such as Cy5, can be used to characterize solute transport patterns within the biofilm.

1. Dissolve Cy5 (Mono-Reactive NHS Ester) in water to a final concentration of 10 mg/ml as a stock solution. Store the Cy5 stock solution in a -20 °C freezer.
   NOTE: As Cy5 is light-sensitive, store, dilute and inject Cy5 solutions in the dark.
2. Prior to the injections of the fluorescent tracer, take a 3-D image stack of the 3-day biofilm with confocal microscopy (with a 63X objective). (Wide regions of biofilm colonies are ideal for time-series observations of dye transport.) Also choose a z-plane with well-separated colonies for imaging (as shown in Figure 5A).
3. For each Cy5 injection experiment, dilute 2 μl of the Cy5 stock solution in 998 μl sterilized water to yield an injection solution having a Cy5 concentration of 20 μg/ml.
4. Stop the pump to pause the flow and inject the Cy5 solution into the upstream of the flow cell using a syringe via the 3-way valve. (It is important to prevent air bubbles from entering the flow path while injecting. Bubble traps should be kept open during the back injection.)
5. After injecting the Cy5 solution, close the bubble traps, adjust the 3-way valve, and restart the pump to deliver the injected Cy5 solution into the flow cell.
6. Switch confocal imaging mode to xyt. Activate Cy5 and GFP channels under simultaneous imaging mode. Adjust confocal settings on Cy5 intensity (laser intensity, gain and offset) to avoid saturating signals, which is critical for quantification. (High temporal resolution is preferred for visualizing dye penetration. However, fast scanning decreases image quality.) Choose proper scanning speed and line average to balance the time resolution of imaging and the image quality. For this protocol use a line average of four and a frame rate of 0.15 Hz.
7. Resume the flow to deliver Cy5 into the flow cell (still at a flow rate of 0.03 ml/min). Simultaneously start time-series confocal imaging.
8. Quantify Cy5 penetration by radially averaging Cy5 intensity within a 2-D circular biofilm colony. To average, first identify the edge of a biofilm cluster, then generate a distance map for a 2-D section of the cluster, and finally average radial patterns of Cy5 intensities to generate a penetration curve (see Figure 6).
   1. To computationally realize step 3.8, use BioSPA (Biofilm Spatial Pattern Analysis) software package (unpublished, software package and manual available upon request) or other image analysis programs.
   2. To use BioSPA, first import an image set into BioSPA. Then select proper threshold for GFP and Cy5 channels to make binary images. In the analysis/calculation menu, select Advanced Analysis and then Diffusion Analysis. Use the GFP channel to define the edge of a cluster and use the polygon tool to select one cluster as the region of interest (ROI).
   3. Double click the selected biofilm cluster to perform diffusion analysis. Save the diffusion analysis results and calibrate Cy5 intensity to Cy5 concentration using a calibration curve.
      1. To generate Cy5 concentration calibration curve, measure Cy5 intensities over a Cy5 concentration gradient under confocal microscopy. (Cy5 intensity and concentration have a linear relationship.)

4. Characterizing the Surrounding Flow Field by Tracking Fluorescent Particles

NOTE: Fluorescent particles, such as fluorescent microbeads, can be used to characterize the flow field around biofilms. Flow field around biofilms can be calculated from the time-series observations of particle positions using particle tracking velocimetry software. The results shown here were obtained with the Streams 2.02 software package¹⁶ (software download and user’s guide available at http://www.civil.canterbury.ac.nz/streams.shtml).

1. Dilute the fluorescent microbeads (diameter ~1 μm) in sterilized water to a final solid concentration of 0.2% and a final volume of 0.5 ml. Vortex to make sure that the microbeads are well dispersed.
2. Inject and deliver the diluted fluorescent microbeads into the flow cell following procedures from steps 3.3 to 3.5. To allow an accurate tracking of the particle path, use a relatively low flow rate (0.01 ml/hr for the particle tracking results in this paper).
3. Switch confocal settings to xyt mode to track particle movement at one z-slice and take time-series images. (A frame rate of 1 Hz was used for the representative results shown in this paper.) Repeat particle injection and image at different z locations to generate 3-D flow field.
4. Pre-process the time-series confocal images by subtracting the background fluorescence signal (the signal in the first acquired image of the image set) using ImageJ (software download and manual available on http://imagej.nih.gov/ij/) or other programs.
5. In ImageJ, open the first image of an image set and then import the image set. Under Process / Image Calculator, subtract the first image from the image set. Save the pre-processed image set.
6. To calculate flow vectors, import time-series confocal images in Streams 2.02. Under “Open process view”, select and execute “Identify particles”. Use proper threshold to identify particles. Use 0.5 and 5 μm as minimum and maximum diameter thresholds for 1 μm particles.
7. Then under “Open process view”, select and execute “PTV analysis pipeline” to generate particle paths. Check the particle paths to see if they represent particle movement. Finally, select and execute “Create velocity field” to generate a flow vector map.

Results

The double-inlet microfluidic flow cell allows observation of biofilm growth under a well-defined chemical gradient formed by mixing of two solutions within the flow chamber. The resulting chemical gradient was formerly observed by dye injection and characterized in detail by Song et al.¹⁴. Smooth concentration gradients were formed in the transverse direction, as shown in Figure 1. The concentration profile was steep near the inlet and got relaxed downstream due to diffusion (

Discussion

We demonstrated a suite of methods to characterize three important biofilm-environment interactions: biofilm response to chemical gradients, effects of biofilm growth on the surrounding flow microenvironment, and biofilm heterogeneity resulting from internal transport limitations.

We first showed the use of a novel microfluidic flow cell to impose a well-defined chemical gradient for biofilm development. To generate a well-defined chemical gradient within the flow cell, it is important to main...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Peristaltic Pump	Gilson	Miniplus 3	Flow cell setup and inoculation
Pump Tubing 0.50 mm OVC, Orange/Yellow	Gilson	F117934	Flow cell setup and inoculation
Three-way Stopcock w/ Swivel Male Luer lock	Smiths Medical	MX9311L	Flow cell setup and inoculation
Sylgard 184 Solar Cell Encapsulation for Making Solar Panels	ML Solar LLC		Flow cell setup and inoculation
Pyrex Medium Bottle, 1 L, GL45	VWR	16157-191	Flow cell setup and inoculation
C-FLEX Tubing	Cole-Parmer	06422-02	Flow cell setup and inoculation
1 ml TB Syringe	BD	309659	Flow cell setup and inoculation
Polymer Tubing	IDEX	1520G	Flow cell setup and inoculation
Sterile Intramedic Luer Stub Adapter	Clay Adams	427564	Flow cell setup and inoculation
PrecisionGlide Needle	BD	305195	Flow cell setup and inoculation
Spectrophotometer	HACH		Flow cell setup and inoculation
Syringe filters - sterile (0.2 μm)	Fisherbrand	09-719A	Flow cell setup and inoculation
MAXQ Shaker	Thermo Scientific		Flow cell setup and inoculation
Ammonium sulfate	Sigma Aldrich	A4418	Growth media
Sodium phosphate dibasic anhydrous	Sigma Aldrich	RES20908-A7	Growth media
Monobasic potassium phosphate	Sigma Aldrich	P5655	Growth media
Sodium chloride	Sigma Aldrich	S7653	Growth media
Magnisium chloride	Sigma Aldrich	M8266	Growth media
Calcium chloride	Sigma Aldrich	C5670	Growth media
Calcium sulfate dihydrate	Sigma Aldrich	C3771	Growth media
Iron(II) sulfate heptahydrate	Sigma Aldrich	215422	Growth media
Manganese(II) sulfate monohydrate	Sigma Aldrich	M7634	Growth media
Copper(II) sulfate	Sigma Aldrich	451657	Growth media
Zinc sulfate heptahydrate	Sigma Aldrich	Z0251	Growth media
Cobalt(II) sulfate heptahydrate	Sigma Aldrich	C6768	Growth media
Sodium molybdate	Sigma Aldrich	243655	Growth media
Boric acid	Sigma Aldrich	B6768	Growth media
Dextrose	Sigma Aldrich	D9434	Growth media
Luria Bertani Broth	Sigma Aldrich	L3022	Growth media
TCS SP2 Confocal Microscopy	Leica		Fluorescent imaging
SYTO 62	Life Technology	S11344	Fluorescent imaging
Cy5	GE Healthcare Life Sciences	PA15100	Fluorescent imaging
Red Fluorescent (580/605) FluoSphere	Life Technology	F-8801	Fluorescent imaging
BioSPA	Packman Lab		Image Processing
ImageJ	NIH		Image Processing
Volocity	PerkinElmer		Image Processing
Streams 2.02	University of Cantebury		Image Processing