In Canada, the incidence of food poisoning is estimated at 2.2 million cases a year. Since most food poisoning comes from bacterial sources, appropriate monitoring is essential to ensure safe food handling all along the food chain. Existing detection methods are long or tedious so we have developed a system able to detect and identify various bacteria simultaneously. Our new method is objective and does not require any particular expertise. Furthermore, the test is simple, quick, and inexpensive.
Available for technology transfer, our solution consists of a microarray reader for micro-organisms detection, such as pathogenic bacteria, in the food industry. This biosensor can read slides on which antibodies are immobilized. The advantage of this reader is that no data analysis is required by the operator since the results of the specific detection of targeted bacteria are displayed on a screen. In addition, this reader is more sensitive and costs considerably less than the microarray readers presently available on the market. Validation tests (see “Plus...” section) allowed us to conclude that the microarray technique is equally effective for the detection of salmonella pathogen bacteria in pure culture, mixed bacterial cultures, or food samples.
In addition to the agrifood sector, bacteria identification and quantification are also of interest in other markets, including security and defense, the environment, and pharmaceuticals.
Three-Phase Comparative Evaluation
We conducted tests to validate the protein microarray approach for pathogen detection, as well as to determine the detection threshold and necessary specificity and amplification. To do so, five commercial anti-salmonella antibodies were printed onto slides. Eight salmonella dilutions, from 1.2 x 106 to 1.2 X 106, were put down in quadruplicate onto the slides. This first step made it possible to determine that the detection limit of our prototype was inferior—1.2 x 106 compared to 1.2 x 106—and that the variation between readings was less than with commercial microarray readers currently on the market, and which are 10 times more expensive on average than the INO reader.
In the second validation phase, other bacteria were added to the known salmonella concentrations before putting the samples on the slides. This test confirmed that the presence of other bacteria did not interfere with the specific detection of salmonella. Given that no crossed reactions were observed, specific salmonella amplification was not necessary.
The third and final phase of validation work consisted of determining the amplification time required for salmonella detection using the biosensor developed at INO. In this test, 10 CFUs (colony-forming units) of salmonella were added to 25 g of chopped chicken and pre-amplified in compliance with a protocol currently used within the agrifood industry. Test results showed that the total analysis time was between 6 and 18 hours, compared to over 48 hours with the current microbiology analysis method.