Detecting Contaminated Poultry

Daniel S. Burgess, Senior Editor

If you live in the US, chances are you’re having chicken for dinner tonight. A significant percentage of that chicken is probably contaminated with pathogens, so governmental agencies and the industry monitor the inspection line to determine if the meat should be discarded, if remediation efforts should be tweaked or if a recall should be issued.

An optical sensor under development by David S. Gottfried’s team at Georgia Tech Research Institute in Atlanta promises to improve this process. Incorporating a commercially available laser diode, an optical waveguide chip and an inexpensive CCD camera, the sensor has demonstrated that it can detect salmonella, and can do so much more quickly than other techniques. It next has its sights on Campylobacter.

Scientists at Georgia Tech Research Institute are working on an optical biosensor that detects the presence of a contaminant by observing the phase of a light beam through the sensing region compared with the light traveling through a reference channel. Pathogens that bind to the chemically selective layer perturb the evanescent electric field of the light in the waveguide, changing the interference pattern at the CCD detector.

It is virtually impossible to prevent the exposure of livestock to salmonella or Campylobacter. Some of those pathogens inevitably make their way into the processing plant. Willie L. Willis, a professor of poultry science at North Carolina Agricultural and Technical State University in Greensboro, suggested a figure of 25 to 40 percent for salmonella contamination. His research indicates that about 66 percent of raw chicken in the US is colonized by Campylobacter.

Remediation techniques, such as rinsing and chilling carcasses in chlorinated water, have reduced salmonella in chicken by two-thirds in recent years, said Richard L. Lobb, director of communications for the Washington-based National Chicken Council. To optimize the effectiveness of these techniques and to minimize water usage, it is necessary to know if, which and how many pathogens are present. The industry and the agriculture department, therefore, continuously monitor the meat for salmonella and generic Escherichia coli. It currently is not mandatory to screen for Campylobacter because there is no fast and accurate means of detection.

Although reliable, today’s sampling techniques are time-consuming, labor-intensive and/or expensive. Conventional testing requires a culturing stage that can take several days, and immunoassays necessitate many hours of enrichment. Sampling methods based on polymerase chain reaction — while much faster, at four to six hours — require a dedicated lab and a specialized staff.

The optical sensor, in contrast, takes a direct approach. Light from a Thorlabs Inc. laser diode is split and travels down two channels in a waveguide chip manufactured by Atlanta-based Microcoatings Technology, one serving as a reference and the other as a sensing region, essentially forming a Mach-Zehnder interferometer. A lens combines the light emerging from these channels and focuses it on a consumer-grade, 320 x 240-pixel CCD for detection.

A chemically sensitive layer is deposited over the sensing channel. If a pathogen is present in the sample, it will bind to this layer, altering the evanescent electrical field of the light traveling through the waveguide. When the light is recombined with the reference beam, the presence of a pathogen thus appears as a shift in the interference fringe pattern, which the researchers convert into a phase change by fast Fourier transform.

In tests using chips prepared with antisalmonella antibodies in the sensing region, the researchers found that they could detect fewer than 10,000 cells per milliliter of poultry chiller water, comparable with other techniques. But detection took less than 30 minutes, demonstrating the potential of the device for process control.

The near-real-time assessment of bacterial levels would enable plant engineers to rapidly identify and address any problems in the system and to maximize remediation while minimizing waste. “In-line process control would benefit the industry and consumer by preventing highly contaminated products from being sold or consumed,” Willis said. “Moreover, it would prevent product recalls and provide insight into better intervention strategies.”

Perhaps the optical sensor’s biggest problem is that it is too sensitive. “In most assays, only viable cells are detected, since a growing step is required,” Gottfried said. “The antibody that we use is sensitive to both live and dead cells, so we get an aggregate count.” The researchers are working to reduce this problem.

Other improvements will include microelectromechanical systems to boost the diffusion of microbes from the sample to the binding region. Because the sensor can detect any microbe with an antibody that can be conjugated to the waveguide, they plan to investigate its use in detecting Campylobacter, Gottfried said.