Microbioma de búfalas leiteiras: estudo da interação com o bezerro e ambiente ao desenvolvimento de resistência bacteriana.

Abstract

This thesis was divided into two chapters. In the first chapter, the hypothesis that the buffalo milk microbiome could share taxa with samples from its feces, pasture soil, and even with the saliva of its calf was tested. Additionally, a possible correlation across generations was also studied. For this, milk, feces, and saliva samples were collected in triplicate from first and third-parity animals on a farm in the Marajó Archipelago, Amazon, Pará-Brazil. Pasture soil was also collected. The animals were clinically healthy. The somatic cell count in the milk showed a significant difference when evaluated between heifers and cows. The metabarcoding approach using 16S rRNA demonstrated a significant difference in the phylogenetic and ecological structure among all the matrices studied. Beta-diversity showed clustering between the milk samples of first and third-lactation animals, and similar patterns in feces and saliva samples. The most abundant phyla in the biological samples were Firmicutes, Proteobacteria, and Bacteroidota. When shared, these taxa were found simultaneously in at least two biological matrices, but never in all three matrices at the same time. The phylum Firmicutes was shared between the soil and the milk, and between the feces of cows and heifers. The samples studied presented a matrix-dependent pattern. It was possible to observe the sharing of families in the microbiome of buffalo milk and feces and the saliva of their calves, reinforcing the idea of microbial interaction among them. In the second chapter, the hypothesis was assessed that there would be a correlation between the Gram-negative microbiota in buffalo feces when compared across different lactation numbers, calves, and the environment. To this end, fecal samples were collected from first-lactation (heifer - HF) and fourth-lactation (cow - CW) buffaloes, as well as their respective calves (CFH and CFC) from a farm in the Marajó Archipelago, Pará, Brazil. Soil samples (S) were also collected. The samples were diluted and plated on selective media supplemented with Ceftriaxone (CRO) or Tetracycline (TET) for bacterial isolation. Colonies grown on CRO were subjected to selective pressure using the replication method on plates with Cefepime (CFP) and Imipenem (IMP). Counting, isolation, and identification were performed using MALDI-TOF. In total, 91 bacterial strains were identified: 37 were resistant to TET, 27 to CFP, and 27 to IMP. The strains were identified as *Escherichia coli* (TET [33]; IMP [7]), *Ochrobactrum intermedium* (TET [4]; CFP [19]; IMP [10]), *Achromobacter xyloxidans* (CFP [5]), and *Stenotrophomonas maltophilia* (CFP [3]; IMP [10]). Grouping was observed between the resistance profiles for CFC and CW for the antibiotic IMP; CW, HF, and CFC with group S for the antibiotics TET, IMP, and CFP, respectively. The results highlight the importance of constant monitoring of the bacterial resistance profile in these animals and the surrounding soil, as commensal and environmental bacteria can be indicators of selective pressure and may highlight emerging antimicrobial resistance agents commonly used.