First report of the isolation and molecular characterisation of Lactococcus garvieae in dairy cattle in Poland

Lactococcus garvieae is a catalase-negative, facultative anaerobic, Gram-positive coccus and is present throughout the world (15). It was identified and described for the first time in the early 1980s as a causative agent of bovine mastitis (3, 14). Previously, it was described as Streptococcus garvieae (3) and was later reclassified to the Lactococcus genus (23). Since then, numerous cases of L. garvieae infections in fish colonies as a highly opportunistic pathogen have been reported frequently in the literature (30).

This microorganism is a nonmotile, nonspore-forming bacterium that occurs in pairs and short chains. It can cause α-haemolysis on blood agar. The optimal growth temperature for this bacterium is 37°C, but it can grow between 10°C and 42°C, and some strains can grow weakly and slowly even at 45°C (17). These bacteria grow at pH 4.5–9.6 in broth containing 4% NaCl, although some strains can grow even at 6.5% NaCl (17). Because of these characteristics, it was misidentified as a new enterococcal species, Enterococcus seriolicida, when it was determined to be the pathogen responsible for septicaemia in yellowtail (Seriola quinqueradiata) (12, 17). However, subsequent phenotypic and genetic analyses comparing characteristics between the new species and isolates of L. garvieae showed that both organisms belonged to the same species, carrying the implication that E. seriolicida should be reclassified as L. garvieae (7, 8, 26).

The wide distribution of L. garvieae is most likely related to the ability of this bacterium to adapt and survive in many environmental conditions. It has been isolated from food and feed products, terrestrial waters and sewage (1), as well as from various aquatic and terrestrial animals (6, 11, 20, 27, 28). Like other widespread microorganisms, L. garvieae is phenotypically and genetically heterogeneous (8, 20, 29). Methods such as analysis of the acidification of sugars and the presence of the enzymes pyroglutamic acid arylamidase and N-acetyl-β-glucosaminidase have been used to identify and differentiate different biotypes of L. garvieae (29). Genotyping studies have also been performed, mainly on trout isolates. They have shown that L. garvieae forms clonal genetic groups (6, 20, 29). However, considerable genetic heterogeneity is observed among isolates from other hosts, environments or foods (20, 27). Recently, genetic and phylogenetic analyses have revealed genotypic differences supported by corresponding phenotypic variations, leading to the proposal of two subspecies: Lactococcus garvieae subsp. garvieae and Lactococcus garvieae subsp. bovis (12, 28)

Lactobacillus garvieae is known to be the causative agent of lactococcosis in various species of wild and farmed fish. Its proliferation peaks during summer, when the water temperature rises, and this also coincides with a rise in the number of L. garvieae infections in fish (30). The bacteria have also been isolated from cows and water buffaloes with subclinical mastitis (1, 3, 6, 13, 18, 21, 24, 26, 30) and pigs with pneumonia (27). In humans, L. garvieae is currently considered an opportunistic pathogen that causes a variety of infections (2, 22, 25). The increasing number of human infections in recent years has raised awareness of L. garvieae as an emerging pathogen associated with the handling or consumption of raw fish and seafood (2). However, the status of L. garvieae as a potential zoonotic bacterium is still contentious. Furthermore, the isolation of L. garvieae from faecal samples of healthy individuals (16) suggests that not all strains are pathogenic to humans or that not all humans are susceptible to L. garvieae infections.

Besides having been present in seafood, L. garvieae has also been found in some dairy products such as cheese and raw cow’s milk. The main objective of this research was to investigate and detect the presence of L. garvieae in milk samples obtained from cows with mastitis in Poland. To the best of our knowledge, this is the first report of mastitis in cows caused by L. garvieae in Poland.

The study was conducted on a dairy farm located in central Poland, keeping 366 Polish Holstein-Friesian dairy cows. The average milk yield of the cows was 11,673 kg of milk which contained 3.64% fat and 3.34% protein. The somatic cell count was 204,000 per mL. The animals were kept in a free-stall box barn and bedded with straw every day. The cows were milked in a 2 × 8 herringbone milking parlour over an average time per group of approximately 45 min. The cattle were fed in the total mixed ration system. The feed ration consisted of the following feeds: corn, grass and legume haylage, brewer’s and fermented corn grain, and soybean, rapeseed and cereal meal. A total of 118 quarter-milk samples which were positive in a California mastitis test (CMT) were collected from 50 cows with clinical (4 cows) and subclinical (46 cows) mastitis, following aseptic procedures as described by the National Mastitis Council (https://www.nmconline.org/nmc-protocols-guidelines-and-procedures). Milk samples were collected from quarters that scored 1, 2 or 3 in the CMT. Sample collection was performed aseptically, and samples were stored in sterile tubes under refrigerated conditions until microbiological and molecular analyses. For classical microbiological analyses, 10 mL of each milk sample was spread on blood agar plates with 5% defibrinated sheep blood. The plates were incubated at 37°C for 72 h, with observation every 24 h. After bacterial growth, colonies with single and pure growth were identified based on Gram stain, morphology and haemolysis patterns. Gram-positive, catalase-negative cocci were further evaluated using aesculin hydrolysis, haemolysis and the Christie–Atkins–Munch-Petersen reaction.

For three isolates suspected of being L. garvieae, additional identification analyses were performed. Milk from three cows suspected of being infected by L. garvieae was sampled again after 30 d and analysed with the procedures previously described. Additionally, analysis with matrix-assisted laser desorption/ionisation– time-of-flight (MALDI-TOF) mass spectrometry, biochemical identification in the VITEK 2 system and a PCR with primers compatible with 16S ribosomal RNA (rRNA) of L. garvieae were performed.

Forty-one bacterial and fungal isolates were identified by classical microbiological analyses. The quantitative distribution of the identified strains was as follows: Lactococcus sp. (L. garvieae) 7.32%, coagulase-negative staphylococci (Staphylococcus spp., S. epidermidis, S. chromogenes and S. xylosus) 43.9%, coagulase-positive staphylococci (Staphylococcus spp. and S. aureus) 12.2%, Streptococcus spp. (Str. spp., Str. dysgalactiae and Str. uberis) 24.39%, Enterococcus sp. (Enterococcus faecalis) 2.44%, Corynebacterium sp. (Corynebacterium bovis) 2.44%, Micrococcus spp. 4.88% and yeasts (Candida sp.) 2.44% (Table 1).

Lactobacillus garvieae was identified biochemically and by MALDI-TOF mass spectrometry. In the VITEK2 system, the probability for all three strains was 95%, which corresponds to very good identification.

In the MALDI Biotyper the scores were 2.16, 2.29 and 2.40, which are trustworthy against the benchmarks: score values higher than 1.99 are indicative of secure to highly probable species identification, those between 1.7 and 1.99 indicative of probable genus identification and those between 0.0 and 1.69 signify unreliable identification.

The results of BLAST showed that in the case of all three bacterial DNA samples, the highest similarity was observed to the 16S rRNA sequence of L. garvieae. The query sequence coverage ranged between 91 and 95%, and the percentages of identical base pairs ranged from 85.16% for the sequence obtained from bacterial strain number 64 to 98.74% for the sequence obtained from bacterial strain number 5 (Table 2). The results of molecular identification confirm that the analysed bacterial isolates belonged to L. garvieae.

Analysis of antibiotic susceptibility with VITEK 2 revealed that all L. garvieae isolates were susceptible to moxifloxacin, erythromycin, linezolid, teicoplanin, vancomycin, tetracycline, tigecycline, chloramphenicol and trimethoprim-sulfamethoxazole. Lactobacillus garvieae had high resistance to benzylpenicillin, clindamycin and rifampicin. One of the isolated strains was additionally resistant to tetracycline and had different MIC values. The results of antibiotic susceptibility are presented in Table 3.

Mastitis in cows is caused by the simultaneous occurrence and overlap of many unfavourable factors, of which the most important are pathogens. The aetiological agents that cause mastitis may vary depending on the climate, species or husbandry practices. These agents include a wide range of Gram-positive and Gramne-gative bacteria, mycoplasmas, fungi, yeasts, algae and viruses. Lactococcus garvieae is among these agents, but in addition is currently recognised as a species of clinical importance to human medicine besides being one to veterinary medicine (3–6).

In addition to being a factor responsible for mastitis in domestic and wild animals, L. garvieae is also an emerging pathogen in aquaculture as well as human endocarditis and sepsis. The pathogenic mechanisms of this bacterium are poorly understood (7), and the world literature on L. garvieae as an aetiological factor of inflammation of the mammary gland is also not extensive compared to the literature on other microorganisms. This may be due to the focus of most studies on major pathogens and the focus of rather limited counterpart studies on minor pathogens (8).

We confirmed the presence of L. garvieae in milk samples from three cows with symptoms of subclinical mastitis using three different methods. The general percentage of L. garvieae detected in milk samples was 2.54 (7.32% of all strains). It was comparable to the results of de Oliveira et al. (9), where 6.9% of 72 isolated bacteria were identified as L. garvieae using MALDI-TOF mass spectrometry.

The isolates identified in the presented study were susceptible to moxifloxacin, erythromycin, linezolid, teicoplanin, vancomycin, tetracycline, tigecycline, chloramphenicol and trimethoprim-sulfamethoxazole. However, they exhibited high resistance to benzylpenicillin, lincomycin and rifaximin, and one strain was resistant to tetracycline. A characteristic feature of L. garvieae is resistance to clindamycin, which distinguishes it from Lactococcus lactis (10). In our study, all three strains were resistant to clindamycin. These results are similar to those obtained by Xie et al. (11), with the difference that in our studies, all strains were resistant to penicillin. According to Scillieri Smith et al. (24), the odds of a successful bacteriological cure for bovine clinical mastitis were eight times higher when the agent was Streptococcus dysgalactiae or Str. uberis than when it was L. lactis or L. garvieae, indicating greater difficulties in treating diseases caused by L. garvieae than treating those caused by other bacteria.

To the best of our knowledge, L. garvieae had not been identified before the time of writing as an aetiological factor of mastitis in cows in Poland. It is possible that Lactococcus strains had already caused mastitis before our investigation, given the possibility that laboratories could have misclassified the cases as infections with Streptococcus spp. A similar hypothesis was presented by Rodrigues et al. (21).

Our findings highlight the presence of L. garvieae as a bovine mastitis pathogen in Poland, a fact that had not been previously reported. The obtained results confirmed subclinical mastitis in cows caused by the identified pathogen. This underscores the importance of accurate identification and characterisation of pathogens to ensure effective treatment and management of mastitis.