Comparison of identification systems for psychrotrophic bacteria isolated from raw bovine milk.

Psychrotrophic bacteria in raw milk produce heat-resistant extracellular proteases, resulting in spoilage and shelf-life reduction of ultrahigh temperature treated milk and milk products. Controlling of these spoilage microbes requires rapid and reliable identification systems for screening of raw milk. This study aimed to compare commercial bacterial identification systems with a genetic method (considered as the 'gold standard' method) for the identification of heat-resistant protease producing bacteria in raw milk. Five bacterial identification systems were compared based on typability, discrimination power (i.e. Simpson's Index of Diversity), reproducibility and speed of analysis. The accuracy of 16S rRNA gene sequencing, Biolog, MALDI-TOF MS, API, and Microbact for the identification of Gram negative bacilli at the species level was 100.0%, 86.8%, 63.2%, 60.5% and 57.9%, respectively. The Gram positive bacilli were identified by 16S rRNA gene sequencing, Biolog, MALDI-TOF MS, and API with accuracies at the species level of 100.0%, 85.0%, 95.0% and 90.0%, respectively. The 16S rRNA gene sequencing and phylogenetic analysis discriminated Pseudomonas fluorescens, Pseudomonas syringae, Hafnia alvei, Bacillus cereus, Bacillus pumilus and Bacillus licheniformis to the subspecies level. The Simpson's Index of Diversity scores were 0.966, 0.711, 0.496, 0.472, and 0.140, for 16S rRNA gene sequencing, Biolog, MALDI-TOF MS, API and Microbact, respectively. Limited reference profiles in the databases of Biolog, MALDI-TOF MS, API and Microbact systems reduced their accuracy in bacterial identification, compared to 16S rRNA gene sequencing. The rapidity of each assay is in the following order; MALDI-TOF MS>16S rRNA gene sequencing>Biolog>Microbact>API. The reproducibility of the assays is in the order of 16S rRNA gene sequencing>API>Microbact>MALDI-TOF MS>Biolog. Thus, 16S rRNA gene sequencing appears to be the most reliable and robust system for the identification of dairy spoilage bacteria. The Biolog system is suitable for the identification of Gram negative spoilage bacteria, while MALDI-TOF MS and API systems are suitable for the identification of Gram positive spoilage bacteria isolated from raw milk. The commercial systems used in this study have been developed and extensively used for the identification of clinical microbes but only a limited number of studies used those systems to identify the environmental microorganisms that often contaminate raw milk. Therefore, comparison of those systems for the identification of spoilage microbes in raw milk would provide better understanding of their suitability for routine dairy microbiology and more extensive dairy research.