Abstract Salmonella enterica serovar Heidelberg (SH), often isolated from broiler chicken samples, damages the entire production chain due to its high resistance in the environment. The search for sustainable disinfectant agents has intensified, focusing on the action of essential oils. This work's objective was to evaluate the in vitro, effect of Thymus vulgaris essential oil (TEO) and thymol in SH isolated from broilers, and the cytotoxicity in animal cells. Four different concentrations of TEO (0.1%, 0.2%, 0.4%, and 0.8%, (v/v)) were used against five SH isolates obtained from broilers and ATCC 8326. Thymol was evaluated at the concentrations of 0.023%, 0.047%, 0.071%, and 0.094% (v/v). In vitro antibacterial assays were performed by quantifying viable planktonic cells in the broth microdilution test. The MTT technique was used to assess the cytotoxicity in IEC-6 intestinal cells and NCTC fibroblasts. IC50 was assessed testing the concentrations of 0.0625%, 0.125%, 0.25%, and 0.5% of TEO and thymol in 24 hours. The bacterial activity was observed from the 0.2% concentration of TEO since no colony-forming units were observed. Thymol in the concentration of 0.094% controlled 83.33% of the bacteria. TEO presented an IC50 of 0.14% and 1.22%, while the IC50 for thymol was 0.068% and 0.001% for the IEC-6 and NCTC cells, respectively. On the other hand, TEO showed low cytotoxicity in fibroblasts and a potential action capacity to eliminate SH strains in vitro.

Spectroscopy associated with chemometrics is a non-destructive method of the samples for the prediction of phenolic compounds. The objective of the work was to apply the near infrared spectroscopy (NIR) technique for the prediction of phenolic compounds and flavonoids in aerial part samples of cover crops plants Raphanus sativus L., Secale cereale L. and Avena strigosa L. Samples were collected at 60, 80 and 100 days after sowing (DAS) of the plants and, after lodging (DAA), at 15 and 30 days, in an experiment implanted in the field, in the city of Ituporanga, Santa Catarina. The samples were homogenized, identified, transported on dry ice, lyophilized, crushed, sieved and stored at -20 °C for analysis. For the calibration, 84 samples were used and for the validation 10 samples. The combination of NIR spectroscopy and multivariate analysis by partial least squares regression allowed the development of prediction models of phenolics and flavonoids with associated errors averaging 10%. NIR spectroscopy is recommended for the development of calibration models because it allows to non-destructive infer phenolic and total flavonoid contents, with high analytical frequency, without the use of reagents and other inputs required by the reference method, with results very close to those obtained in the laboratory.