Comparison of the BACGene Real-Time PCR and BAX® System PCR Methods for the Detection of Salmonella in Pet Food
Introduction
A head to head study was conducted comparing the performance of the BACGene® Salmonella real-time qPCR method with the BAX® System Salmonella method in spiked pet food (dry kibble). Both systems have been validated/certified (AOAC, AFNOR) for pet food testing.
Materials and Methods
375G pet food composites (previously screened negative for Salmonella) were prepared from kibble provided by a pet food manufacturer and spiked with Salmonella at three different levels (low, med, high) in an attempt to obtain fractional positives at one or more spike levels by either or both methods. Negative controls were also included to which no Salmonella was added. Salmonella Senftenberg (UV-biotag GFP labeled) was used as the test organism. All samples, irrespective of PCR result, were taken to culture confirmation by FDA-BAM and further confirmed by examination of fluorescence.
Results
Fractional results were obtained for both methods with 11/20 (aggregate of low and medium spike) samples positive for each. All presumptive positives for both methods were confirmed by culture, however, the BAX System method had one sample that was positive by culture but negative by PCR (false negative). No false positives were reported by either method.
Conclusions
- Fractional positive recovery was achieved for both methods (21/30 aggregate of all three inoculation levels) and overall the two methods were identical in performance in terms of the number of presumptive positives returned
- Both methods appropriately called all negative controls negative
- 2/10 presumptive low inoculation level, both confirmed. One culture positive that was negative on BAX (false negative)
- 9/10 presumptive medium inoculation level, all nine confirmed positive and the negative confirmed as negative
- 10/10 presumptive at high inoculation level, all ten confirmed
- The BAX System method had one false negative. Examination of the raw data showed an atypical melt peak in the target region which the algorithm “missed” and called negative. The BACGene method had no false positives or negatives
- It cannot be ruled out that the Salmonella strain used in this study (UV-biotag) has sequence changes due to the GFP modification that effected the target region resulting in an atypical melt profile.
- The strain has always performed appropriately as our daily process control for BAX but those are always tested at a high concentration of target.
- The atypical peak may only become apparent when the organism is present at low concentrations (near the LLOD) in the enrichment.
- Both systems are approved for use in dry kibble pet food testing at Eurofins microbiology laboratories
Questions on how this information is applicable to your food process?
