Lactic Acid Bacteria Isolated from Bovine Mammary Microbiota: Potential Allies against Bovine Mastitis

Abstract

Bovine mastitis is a costly disease in dairy cattle worldwide. As of yet, the control of bovine mastitis is mostly based on prevention by thorough hygienic procedures during milking. Additional strategies include vaccination and utilization of antibiotics. Despite these measures, mastitis is not fully under control, thus prompting the need for alternative strategies. The goal of this study was to isolate autochthonous lactic acid bacteria (LAB) from bovine mammary microbiota that exhibit beneficial properties that could be used for mastitis prevention and/or treatment. Sampling of the teat canal led to the isolation of 165 isolates, among which a selection of ten non-redundant LAB strains belonging to the genera Lactobacillus and Lactococcus were further characterized with regard to several properties: surface properties (hydrophobicity, autoaggregation); inhibition potential of three main mastitis pathogens, Staphylococcus aureus, Escherichia coli and Streptococcus uberis; colonization capacities of bovine mammary epithelial cells (bMEC); and immunomodulation properties. Three strains, Lactobacillus brevis 1595 and 1597 and Lactobacillus plantarum 1610, showed high colonization capacities and a medium surface hydrophobicity. These strains are good candidates to compete with pathogens for mammary gland colonization. Moreover, nine strains exhibited anti-inflammatory properties, as illustrated by the lower IL-8 secretion by E. coli-stimulated bMEC in the presence of these LAB. Full genome sequencing of five candidate strains allowed to check for undesirable genetic elements such as antibiotic resistance genes and to identify potential bacterial determinants involved in the beneficial properties. This large screening of beneficial properties while checking for undesirable genetic markers allowed the selection of promising candidate LAB strains from bovine mammary microbiota for the prevention and/or treatment of bovine mastitis.

Fig 1. Adhesion of lactic acid bacteria to bovine mammary epithelial cells.

Lactic acid bacteria populations that adhered to bMEC were determined after 1 h of interaction at a MOI of 400:1 (A) and 2000:1 (B), respectively. Data are presented as mean population per well (i.e., corresponding to 2.5x10⁵ bMEC) ± standard deviation. Each experiment was done in triplicate and differences between strains were assessed using a one-way analysis of variance, followed by Tukey’s range test. Letters a, b, c and d indicate homogeneous statistical processing groups that were significantly different according to Tukey’s range test.

Fig 2. Internalization of lactic acid bacteria into bovine mammary epithelial cells.

Lactic acid bacteria populations internalized into bMEC were determined after 2 h of interaction at a MOI of 400:1 (A) and 2000:1 (B), respectively. Data are presented as mean population per well (i.e., corresponding to 2.5x10⁵ bMEC) ± standard deviation. Each experiment was done in triplicate and differences between strains were assessed using one-way analysis of variance, followed by Tukey’s range test. Letters a, b, c and d indicate homogeneous statistical processing groups that were significantly different according to Tukey’s range test.

Fig 3. Modulation of cytokine IL-8 production by LAB isolates.

A: modulation of IL-8 production by the PS cell line in the presence of LAB isolates (MOI 100:1). Bars represent the mean IL-8 production ± standard deviation for four assays (two biological and two technical replicates), normalized with regard to IL-8 production by unstimulated PS cells (76 +/-16 pg/mL as a mean); B: modulation of IL-8 production by the E. coli-stimulated PS cell line in the presence of LAB isolates. E. coli was used at MOI 1:1 and LAB at MOI 100:1. Bars represent the mean IL-8 production ± standard deviation for four assays, normalized with regard to IL-8 production by E. coli-stimulated PS cells (368 ± 88 pg/mL as a mean). Differences in IL-8 production with regard to the reference condition were assessed using the Mann-Whitney test (* p < 0.05).