Stress-induced Antibiotic Susceptibility Testing on a Chip

The overall goal of the following experiment is to rapidly determine the susceptibility of bacteria to the antibiotics of interest. This is achieved by first immobilizing log phase bacteria to the bottom of a microfluidic channel. In the second step, media containing the antibiotic and a alive dead fluorescent stain is then pushed through the channel at a high flow rate, stressing the bacteria and activating their antibiotic targeted biochemical pathways.

Next phase contrast and fluorescence images are automatically collected every two minutes for one hour to monitor bacterial viability, normalized bacterial death percentages are calculated from the acquired images at each time point of the experiment. Ultimately, the antibiotic susceptibility phenotype of the bacteria is determined from an increase in the normalized cell death percentage over time. The main advantage of this technique over existing phenotypic susceptibility testing methods such as broth, dilution, or microculture, is that our method does not require the monitoring of bacterial division.

Rather, we can access their susceptibility profile directly from the existing small population. The implications of this technique extent toward the diagnosis of resistant bacterial infections as the method is rapid, can be easily multiplexed and automated To create the PDMS layer begin by degassing a freshly prepared PDMS viscous mixture in a vacuum chamber at room temperature after an hour, use a scale to fill the aluminum mold with a predefined mass of DGAs PDMS mixture. Pour the PDMS slowly over the center of the mold, taking care to keep the mold leveled and the pins uncovered.

Cure the PDMS overnight in a 37 degree Celsius oven. The next morning, release the top mold plate and carefully remove the cured PDMS from the mold to assemble the flow cell. First, place the glass window into the flow cell pocket.

Then lay a coated glass slide over the glass window inside the pocket of the flow cell with the active side up place. The freshly created PDMS layer with the channels facing down on top of it such that the channel inputs align with the through holes in the metal plate. Gently push the air out from between the layers and then flip the PDMS glass slide assembly so that the PDMS faces the glass window.

Overlap the PDMS channel inputs with the through holes in the metal plate, and then place the pressure plate on top and tighten the screws. Now place the assembled flow cell under the microscope, set the microscope magnification to 60 x and pre align the channel positions to prepare log phase bacteria for loading into the flow cell mixed 50 microliters of the overnight cultured bacterial colony of interest into 50 milliliters of fresh MH two broth. Incubate for three hours with shaking at 250 RPM and 37 degrees Celsius after the subculture centrifuge 10 milliliters of the culture for two minutes at 1, 650 times G and room temperature.

Then remove the supernatant and resuspend the pellet in one milliliter of fresh MH two media. Next, attach a short length of tubing to a one milliliter syringe and flush the tubing with one milliliter of media, leaving a small amount of media in the tubing to avoid air bubbles when drawing in the bacterial suspension. Then load 0.7 milliliters of the bacteria into the syringe and fill two channels of the flow cell with the bacterial cell culture.

The transparency of the channel changes as it is filled with the bacteria. Then incubate the flow cell for 45 minutes at 37 degrees Celsius to allow the bacteria to attach to the slide surface. To prepare the solutions for the experiment.

First, fill two 60 milliliter syringes with freshly prepared control solution and two 60 milliliter syringes with freshly prepared antibiotic solution. Flick the syringes to remove any air bubbles and then attach the input tubing pushing the respective solutions through the end of the tubing. Keep the solutions wrapped in aluminum foil to avoid light induced degradation of the reagents.

Now mount the syringes with experimental solution onto the pump, one at a time. Squeeze their plungers as necessary to fit them onto the pump. Lock the syringes in place with syringe holding crossbar.

Set the pump speed to one milliliter per minute and the pump volume to 60 milliliters and flush the pump until a steady liquid flow is seen from all syringes. Now start the experiment by removing the flow cell from the incubator and mounting it onto the pre malaligned microscope stage. Next, connect one input tube coming from a syringe and one output tube going to a waste bottle to each of the flow cell channels.

At this point, the experiment is fully set up and is ready to begin. Set the phase contrast acquisition time to 10 milliseconds and the fluorescence acquisition time to 1600 milliseconds, and then obtain phase contrast and fluorescence images for each position before initiating the flow. Please note that before the start of the flow image, focusing on the bottom of the channel may be unachievable due to the high density of loaded bacteria within the channel.

Now start the liquid flow. Immediately check that the microscope is focused on the bottom of the channels. Then take phase contrast and fluorescence images of the target areas within the first minute of flow.

Acquire images every two minutes after the first set of images until 60 minutes of flow has occurred. Refocusing as necessary to ensure that the experiment is proceeding properly, one can examine acquired phase contrast and fluorescence data. In this data sample, a susceptible strain was loaded into channels one and two, while a resistance strain was in channels three and four.

At the end of the experiment, run a cleaning cycle with sequential application of bleach and water. Fill four 20 milliliter syringes with 10 milliliters of a 10%bleach solution and four 60 milliliter syringes with di water. After de bubbling, attach the bleach filled syringes to the flow cell and set the pump speed to one milliliter per minute and the pump volume to three milliliters.

Run the pump for three minutes, monitor the cleaning of the channel on the screen. After three minutes, replace the bleach syringes with water syringes and run for 30 minutes at one milliliter per minute. Finally, count the number of bacteria in each image using image processing software.

Calculate the normalized bacterial cell death percentage as a function of time using this formula where NF equals the number of dead bacteria as counted in the fluorescence image count. And NP equals the total number of bacteria as determined from the phase contrast image. The images shown here illustrate the time-dependent response of a susceptible staphylococcus aria strain to oxacillin inside a microfluidic flow cell.

These quadrants show phase contrast images acquired one minute after starting the experiment and at one hour with the corresponding fluorescence images shown here. A significant increase in the number of fluorescent bacteria is observed by the end of the experiment indicating cell death within the channel and the susceptibility of the strain to the antibiotic due to the stochastic non-uniformity of individual epoxy coated slides. There may be bacterial cell loss throughout the experiment under sustained sheer stress to account for this variation, each fluorescence image of a dead cell population is normalized with the co acquired phase contrast image giving the total cell population to within one second of the same time point.

These enlargements demonstrate that the counting algorithm is more successful at accurately counting individual bacteria in fluorescence images than in densely populated phase contrast images. The dead cell stain gives a high fluorescence contrast for individual dead bacteria. Since the number of fluorescent bacteria rarely surpasses 5%of the total number, each bacterium is very bright compared to the background.

Therefore, the fluorescent bacteria can be easily counted with a high degree of accuracy. For example, in this experiment of the 5, 828 total bacteria detected, highlighted in red 174 were determined to be dead here. Normalized MSSA and Mr.RSA data for three different experiments are illustrated as expected for resistant strains, the normalized cell death is low in magnitude less than 0.5%and does not change over the course of the experiment.

Susceptible strains however, show a steady increase in cell death and a higher value greater than 1%by the end of the experiment. The initiation of cell death for susceptible strains varies slightly between experiments, but usually occurs between 10 and 30 minutes. Don't forget that working with antibiotic resistant pathogens can be extremely hazardous, and precautions such as a septic technique should always be taken while performing this procedure.

This technique will pave the way for researchers to explore how bacteria respond to stress and will impact the development of novel rapid diagnostics and the discovery of new drugs.