Limits...
Visualization of Biosurfactant Film Flow in a Bacillus subtilis Swarm Colony on an Agar Plate.

Kim K, Kim JK - Int J Mol Sci (2015)

Bottom Line: Here, we developed a visualization method using submicron fluorescent beads for investigating the flow field in a thin layer of fluid that covers a Bacillus subtilis swarm colony growing on an agar plate.The beads were initially embedded in the agar plate and subsequently distributed spontaneously at the upper surface of the expanding colony.At the advancing edge of the colony, both the magnitudes of velocity and vorticity of flow in swarm colony were inversely correlated with the spreading speed of the swarm edge.

View Article: PubMed Central - PubMed

Affiliation: Department of Mechanical and Aerospace Engineering, North Carolina State University, Raleigh, NC 27695, USA. kkim15@ncsu.edu.

ABSTRACT
Collective bacterial dynamics plays a crucial role in colony development. Although many research groups have studied the behavior of fluidic swarm colonies, the detailed mechanics of its motion remains elusive. Here, we developed a visualization method using submicron fluorescent beads for investigating the flow field in a thin layer of fluid that covers a Bacillus subtilis swarm colony growing on an agar plate. The beads were initially embedded in the agar plate and subsequently distributed spontaneously at the upper surface of the expanding colony. We conducted long-term live cell imaging of the B. subtilis colony using the fluorescent tracers, and obtained high-resolution velocity maps of microscale vortices in the swarm colony using particle image velocimetry. A distinct periodic fluctuation in the average speed and vorticity of flow in swarm colony was observed at the inner region of the colony, and correlated with the switch between bacterial swarming and growth phases. At the advancing edge of the colony, both the magnitudes of velocity and vorticity of flow in swarm colony were inversely correlated with the spreading speed of the swarm edge. The advanced imaging tool developed in this study would facilitate further understanding of the effect of micro vortices in swarm colony on the collective dynamics of bacteria.

No MeSH data available.


Related in: MedlinePlus

Swarm sample images of fluorescent beads undergone specific steps for image processing and analysis. (a) Raw image of swarm colony; (b) Smooth; (c) Find edge; (d) Subtract background (rolling = 10); (e) Particle image velocimetry analysis based on cross-correlation method; and (f) Velocity vector and vorticity fields.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4613199&req=5

ijms-16-20225-f006: Swarm sample images of fluorescent beads undergone specific steps for image processing and analysis. (a) Raw image of swarm colony; (b) Smooth; (c) Find edge; (d) Subtract background (rolling = 10); (e) Particle image velocimetry analysis based on cross-correlation method; and (f) Velocity vector and vorticity fields.

Mentions: Figure 6 shows the image processing and analysis procedure used in this study. Each raw image (Figure 6a) was pre-processed by the ImageJ program. We first removed the background noise through “Smooth” process (Figure 6b). Then “Find Edges” and “Subtract Background” processes were applied to highlight the fluorescent beads in the image (Figure 6c,d). We visualized the dynamic flow pattern near the B. subtilis bacterial swarm colony formed on agar gel plate. The PIV analysis based on a cross-correlation method provides the velocity field of the biosurfactant film flow in a swarm colony using a pattern-matching algorithm. When we acquire successive PIV images at time points t0 and t0 + ∆t, small interrogation windows (M × N pixels) taken from two successive images are denoted as IW(i, j) and SW(i, j), respectively. The cross-correlation function, R(p, q), is defined as [22]:(2)R(p,q)=∑j=0N−1∑i=0M−1IW(i−p,j−q)SW(i,j)∑j=0N−1∑i=0M−1IW2(i,j)∑j=0N−1∑i=0M−1SW2(i,j)where i and p represent integers from 0 to M − 1 and likewise j and q represent integers from 0 to N − 1. If the maximum peak of R(p, q) is found at (pmax, qmax), the representative velocity (u, v) at the center point of the interrogation area can be determined by dividing the displacements, pmax and qmax,with the time interval ∆t and the scale factor. The algorithm was applied to the images of fluorescent particle motion induced by the collective dynamics of the bacterial cells. A custom-made program (FlowVision, Seoultech, Seoul, Korea) was used for processing instantaneous velocity vectors with an optimized set of various parameters. The interrogation size was 16 pixels. Overlapping between two adjacent interrogation areas and super-resolution analysis were not used. The computational time for calculating the cross-correlation function was dramatically reduced by applying the fast Fourier transform (FFT). A polynomial interpolation method was used to search for a peak of the cross-correlation function with a subpixel accuracy (Figure 6e). Erroneous vectors were eliminated by both range validation and peak-height validation methods.


Visualization of Biosurfactant Film Flow in a Bacillus subtilis Swarm Colony on an Agar Plate.

Kim K, Kim JK - Int J Mol Sci (2015)

Swarm sample images of fluorescent beads undergone specific steps for image processing and analysis. (a) Raw image of swarm colony; (b) Smooth; (c) Find edge; (d) Subtract background (rolling = 10); (e) Particle image velocimetry analysis based on cross-correlation method; and (f) Velocity vector and vorticity fields.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4613199&req=5

ijms-16-20225-f006: Swarm sample images of fluorescent beads undergone specific steps for image processing and analysis. (a) Raw image of swarm colony; (b) Smooth; (c) Find edge; (d) Subtract background (rolling = 10); (e) Particle image velocimetry analysis based on cross-correlation method; and (f) Velocity vector and vorticity fields.
Mentions: Figure 6 shows the image processing and analysis procedure used in this study. Each raw image (Figure 6a) was pre-processed by the ImageJ program. We first removed the background noise through “Smooth” process (Figure 6b). Then “Find Edges” and “Subtract Background” processes were applied to highlight the fluorescent beads in the image (Figure 6c,d). We visualized the dynamic flow pattern near the B. subtilis bacterial swarm colony formed on agar gel plate. The PIV analysis based on a cross-correlation method provides the velocity field of the biosurfactant film flow in a swarm colony using a pattern-matching algorithm. When we acquire successive PIV images at time points t0 and t0 + ∆t, small interrogation windows (M × N pixels) taken from two successive images are denoted as IW(i, j) and SW(i, j), respectively. The cross-correlation function, R(p, q), is defined as [22]:(2)R(p,q)=∑j=0N−1∑i=0M−1IW(i−p,j−q)SW(i,j)∑j=0N−1∑i=0M−1IW2(i,j)∑j=0N−1∑i=0M−1SW2(i,j)where i and p represent integers from 0 to M − 1 and likewise j and q represent integers from 0 to N − 1. If the maximum peak of R(p, q) is found at (pmax, qmax), the representative velocity (u, v) at the center point of the interrogation area can be determined by dividing the displacements, pmax and qmax,with the time interval ∆t and the scale factor. The algorithm was applied to the images of fluorescent particle motion induced by the collective dynamics of the bacterial cells. A custom-made program (FlowVision, Seoultech, Seoul, Korea) was used for processing instantaneous velocity vectors with an optimized set of various parameters. The interrogation size was 16 pixels. Overlapping between two adjacent interrogation areas and super-resolution analysis were not used. The computational time for calculating the cross-correlation function was dramatically reduced by applying the fast Fourier transform (FFT). A polynomial interpolation method was used to search for a peak of the cross-correlation function with a subpixel accuracy (Figure 6e). Erroneous vectors were eliminated by both range validation and peak-height validation methods.

Bottom Line: Here, we developed a visualization method using submicron fluorescent beads for investigating the flow field in a thin layer of fluid that covers a Bacillus subtilis swarm colony growing on an agar plate.The beads were initially embedded in the agar plate and subsequently distributed spontaneously at the upper surface of the expanding colony.At the advancing edge of the colony, both the magnitudes of velocity and vorticity of flow in swarm colony were inversely correlated with the spreading speed of the swarm edge.

View Article: PubMed Central - PubMed

Affiliation: Department of Mechanical and Aerospace Engineering, North Carolina State University, Raleigh, NC 27695, USA. kkim15@ncsu.edu.

ABSTRACT
Collective bacterial dynamics plays a crucial role in colony development. Although many research groups have studied the behavior of fluidic swarm colonies, the detailed mechanics of its motion remains elusive. Here, we developed a visualization method using submicron fluorescent beads for investigating the flow field in a thin layer of fluid that covers a Bacillus subtilis swarm colony growing on an agar plate. The beads were initially embedded in the agar plate and subsequently distributed spontaneously at the upper surface of the expanding colony. We conducted long-term live cell imaging of the B. subtilis colony using the fluorescent tracers, and obtained high-resolution velocity maps of microscale vortices in the swarm colony using particle image velocimetry. A distinct periodic fluctuation in the average speed and vorticity of flow in swarm colony was observed at the inner region of the colony, and correlated with the switch between bacterial swarming and growth phases. At the advancing edge of the colony, both the magnitudes of velocity and vorticity of flow in swarm colony were inversely correlated with the spreading speed of the swarm edge. The advanced imaging tool developed in this study would facilitate further understanding of the effect of micro vortices in swarm colony on the collective dynamics of bacteria.

No MeSH data available.


Related in: MedlinePlus