Water buffalo mastitis represents a major issue in terms of animal health, cost of therapy, premature culling and decreased milk yeld. The emergence of antibiotic resistance has led to investigate strategies to avoid or reduce antibiotics' based therapies, in particular during subclinical mastitis. The use of Generally Regarded As Safe bacteria (GRAS) such as Lactobacillus rhamnosus to restore the unbalance in mammary gland microbiota could provide potential corrective measures. The aim of this study was to investigate the changes in milk microbiota after the intramammary treatment with inactivated cultures of Lactobacillus rhamnosus of mammary gland quarters naturally affected by subclinical mastitis as compared to antibiotic therapy.A number of 43 quarters affected by subclinical mastitis with no signs of clinical inflammation and aerobic culture positive for pathogens were included in the study. The experimental design was as follows: 11 quarters were treated with antibiotics, 15 with inactivated cultures of Lactobacillus rhmnosus and 17 with PBS as negative control, by means of intrammary injection. Samples were collected at eight time points, pre- (T-29, T-21, T-15, T-7, T0 days) and post- treatment (T1, T2, and T6 days). Microbiological culture and Somatic Cell Count (SCC) were perfomed on all the samples, and microbiota was determined on milk samples collected at T0 and T6 by amplifying the V4 region of 16S rRNA gene by PCR and sequencing using next generation sequencing technique. Treatment with Lactobacillus rhamnosus elicited a strong chemotactic response, as determined by a significant increase of leukocytes in milk, but did not change the microbiological culture results of the treated quarters. For what concerns the analysis of the microbiota, the treatment with Lactobacillus rhamnosus induced the modification in relative abundance of some genera such as Pseudomonas and 5-7N15. As expected, antibiotic treatment caused major changes in microbiota structure with an increase of Methylobacterium relative abundance. No changes were detected after PBS treatment. In conclusion, the present findings demonstrated that the in vivo intrammmary treatment with Lactobacillus rhamnosus has a transient pro-inflammatory activity by increasing SCC and is capable to modify the microbiota of milk after six days from inoculation, albeit slightly, even when the bacterial cultures were heat inactivated. Further studies are necessary to assess the potential use of this GRAS as supportive therapy against mastitis.