Tel.+49 30 838 62618 Fax.+49 30 838 62620
The objective of this study was to proof that bacteria causing mastitis in dairy cows can be identified by their specific odor. Specifically we set out to train dogs to discriminate the odor of Staphylococcus aureus against five other common mastitis causing agents (i.e., Escherichia coli, Streptococcus uberis, Streptococcus dysgalactiae, Pseudomonas aeruginosa and Candida albicans). The study included two experiments. First we tested if dogs can discriminate Staph. aureus in the headspace of blood agar plates growing Staph. aureus colonies. In the second experiment the dogs had to identify the odor of Staph. aureus in milk samples spiked with Staph. aureus or the other infectious agents, respectively.
10 privately owned dogs of various breeds and both sexes were trained in this study. For the first experiment bacteria were isolated from milk samples from cows with clinical mastitis and cultured on Columbia Agar with 5% sheep-blood for 48 h. Identification of infectious agents was by their morphology and was confirmed by MALDI-ToF. Until use, strains were stored at – 20°C in cryotubes. Before dog training or testing the infectious agents were again cultured on Columbia Agar with 5% sheep-blood. A cotton swab was placed in the lit of the plate to absorb the specific odors. For the second experiment only Staph aureus, Strep uberis and Enterococcus spp. were used. Bacteria were concentrated (1012Cfu/g and frozen at -20°C in cryotubes. For dog training and testing the bacterial concentrate was solved in 2 ml fresh bulk tank milk with SCC 120 000/ml. The cotton swabs or milk sample were placed into a 1 liter sterile plastic buckets with perforated lits. The tests for each experiment included 10 test trials. For one trial a dog had to discriminate one sample containing Staph. aureus against nine samples containing other infectious agents. Every dog tested 100 samples.
Eight dogs completed the training and took part in the test of the first experiment. The training duration in minutes varied from 111 – 290 min (163.4 ±70.2min). Overall sensitivity for identifying Staph aureus in the head space of agar plates against five other common mastitis causing agents (i.e., Escherichia coli, Streptococcus uberis, Streptococcus dysgalactiae, Pseudomonas aeruginosa and Candida albicans) was 91.3% and specificity was 97.9 %. Four dogs did not make any mistakes. Breed or sex of the dog had no influence on the performance. Six dogs were included in the second experiment. In bulk tank milk the dogs could identify the samples with Staph aureus, Strep uberis and Enterococcus spp. with a specificity and sensitivity of 82% and 87%, respectively.
In this study we demonstrated that Staph. aureus can be identified by means of a specific odor in two substrates. Further research is warranted to proof that odor diagnostic either with technical methods such as gas chromatography or electronic noses devices or biosensors such as trained animals (e.g., rats, bees or dogs) could be a valid and practical diagnostic tool to identify mastitis pathogens.