Getting the measure of Salmonella

Effective screening of food for microbes requires a fast and accurate approach, Laboratory News finds out how one company think the answer lies in a RABIT

SALMONELLA first caught the UK’s attention on a large scale following the curious incident of Edwina Currie and the politician’s egg in 1989, where she announced that most British eggs contained the bacteria. This led to a sharp increase in demand for Salmonella testing and sample throughput in food and public health testing laboratories.

Dr Bob Madden, project leader in the Food Microbiology Branch of the Agri-Food & Biosciences Institute in Belfast, Northern Ireland, realised the impact this was going to have on his staff’s workload. He identified the need for a timesaving Salmonella screening method that would enable a significantly increased volume to be effectively accommodated. For the past 20 years, his team has been using the Rapid Automated Bacterial Impedance Technique (RABIT) to develop processes to analyse pathogens, such as Salmonella, in food related matrices including animal protein and carcass swabs. In addition, research into factors affecting bacterial growth has also been undertaken using this system.

The indirect impedance method of growth detection for microorganisms relies on their production of carbon dioxide (CO2) as they metabolise carbohydrates and amino acids. The gas dissolves in the aqueous phase of the growth medium and can then diffuse and be absorbed into a solution or gel of potassium hydroxide (KOH). The molten gel is poured into a RABIT cell and changes in the conductance of the KOH occur as the CO2 produced by the organisms, growing in a tube of medium held above the KOH, reacts to form bicarbonate ions whose conductivity is less than that of the original solution. The growth of, for example, Salmonella is detected when the culture has grown enough to produce sufficient CO2 to cause a significant reduction in the conductance of the detecting gel. The RABIT can be easily programmed to detect the rapid fall in conductance and record the time taken for this to occur. This significant drop in conductance normally corresponds to late log/early stationary phase of bacterial cell growth, when the highest numbers of metabolically active cells are present. Given an appropriate selective broth, such as Rappaport-Vassiliadis broth (RV), then most food samples will show no growth and no further work on these samples is required, giving results within 48 hours, and saving the costs of streaking out selective media, incubating and examining them1. The indirect conductimetry method is now approved for use under the Animal By-Products Regulations (Northern Ireland) methods for the detection of Salmonella.

The rapid results obtained with the RABIT also allow the growth rates of salmonellas in different conditions, such as different brands of RV, to be easily compared2.  The very precise and accurate temperature control of the RABIT also allows detailed study of microbial growth rates over narrow temperature ranges – optimum incubation conditions can therefore be determined very quickly, in a manner impossible with conventional plating techniques. Large volumes of data for inclusion in growth modelling studies can thus be generated3.

However, the sensitivity of the RABIT means that it cannot always be used to enumerate bacteria from foodstuffs as a colony on a plate simply shows some cells could grow to a visible size during an incubation of one or two days. Growth detection in the RABIT will depend on both the lag phase of cells, and their growth rate. The latter can vary markedly depending on the history of a product; hence the correlation between a total viable count and RABIT detection time may be too low to be of practical use. Despite this problem Madden and Gilmour showed that coliforms in milk could be enumerated much more rapidly, and cheaply, using the RABIT than the very labour intensive most probable number (MPN) method normally used4.

In a research laboratory, microbial cultures can easily be standardised, therefore the sensitivity of the RABIT becomes a considerable asset which can be applied to the investigation of bacterial growth. In a recent study published in the Journal of Applied Microbiology, Bob Madden’s research group subjected 40 Salmonella enterica isolates to three separate antimicrobial treatments used in food processing: heat, irradiation and high hydrostatic pressure (HHP)5. The indirect technique was used on the RABIT system to quantify the time taken for a sample of the surviving cells to increase in numbers to reach the detection threshold. This time to detection (TTD) was compared with the conventional total viable count (TVC) to assess the suitability of the method for rapid determination of relative stress resistance in Salmonella isolates.

With irradiation and heat treatments, a significant correlation was found between the TVC and TTD. However, the impedance results were obtained within nine hours, as opposed to at least 24 with traditional methods, and required markedly fewer resources. This study would have been impossible using conventional methods for measuring growth, and the results with HHP showed a very surprising result. Using plate counting those isolates showing a small reduction in population after HHP treatment would be classified as resistant, and a large population fall would lead to an isolate being considered as sensitive. The results showed no correlation between TTD and TVC for HPP, and some ‘sensitive’ isolates rapidly grew to high numbers after treatment. The converse was also true. Therefore the ability of salmonellas to grow after HHP treatment could not be predicted from TVC results, but was clearly shown by those from the RABIT. These results have major implications for the design of safe processing conditions as the TVC may apparently show the target organism has been eliminated, but the surviving cells can rapidly grow, although not on solid media.

Overall the RABIT can form the basis of rapid and simple methods which can be used to measure the time taken by a population of organisms to reach an activity threshold. It has shown its worth in determining the relative resistance of microorganisms to processing stressors and its ease of use and economy of materials allows a large number of isolates to be compared, ensuring that the most appropriate strains are selected for detailed study.

Provided the target microbe produces detectable quantities of metabolites, the RABIT system lets the growth of the organism to be quantified on different compounds using precise and accurate temperature control. This allows the temperature for optimal growth to be determined very quickly. For example, Salmonella serovars show exponential growth rates of about one log cycle per hour at 37°C therefore growth rates can be determined within a working day, using a single dilution series. Using conventional counting would require multiple dilution series and the overnight incubation of plates. Also manual counts, for statistical reasons alone, have an inherent lack of precision which will feed through to the growth rates determined. RABIT makes it possible to screen large numbers of bacterial strains that would previously have been unfeasible using traditional methods, and produce much more precise results.