Modelling growth of two monocytogenes strains, persistent and non-persistent: Effect of temperature.

Better growth is a phenotypic trait that can contribute to persistence of in food processing environments. To test the hypothesis objectively, persistent and non-persistent strains were selected and grown in different media to gain reliable quantitative growth characteristics. In this study, the effect of temperature in the range from 6 °C to 43 °C on the planktonic growth of genotypically and phenotypically different strains LM9611-19 (LM-P, persistent) and LM120/5 (LM-S, sporadic - potentially non-persistent) in Tryptone Soy Broth (TSB) and in semi-synthetic cheese medium (SCM) was investigated. Two steps of growth modelling were applied to primary growth data and growth parameters using Baranyi and cardinal temperature models (CM), respectively. No statistically significant differences were found between the growth rates of the strains within the temperature range of 6 °C-37 °C in both media. However, the average growth rates were significantly higher (p < 0.05) for LM-P than for LM-S at 40 °C and 43 °C in both media. Regardless of whether calculated on or basis, in TSB or SCM, for LM-P strain ranged from -1.2 to 0.7 °C with an average of 0.0 ± 0.9 °C (mean ± SD). Other averages of cardinal values were in TSB (a = 0.995; pH 7)  = 37.8 ± 2.0 °C,  = 43.6 ± 0.5 °C and  = 1.27 ± 0.2 h. In SCM (a = 0.970, pH 7), the averages of , , and were 38.0 ± 1.2 °C, 45.2 ± 2.9 °C and 0.92 ± 0.04 h, respectively. Generally, the parameters of the CM model for the growth rate of sporadic strain in cheese medium were lower than for the persistent strain. This includes also , which reflects lower experimental growth rates in the range from to . However, based on the results found in the suboptimal temperature range, it seems that the growth rate did not play an important role in the persistency characteristics. It should be noted that the study was accompanied not only by low errors in model parameters but also by acceptable external validation indices for values in SCM. To consider the ComBase Predictor data (n = 8), the bias factors ( ) of 1.08 and 1.05 and accuracy factor ( ) of 1.09 and 1.07 were calculated for the LM-P strain and LM-S strain, respectively. The approach used in this study revealed different growth responses in the range of temperatures higher than . It can be extended also for mild inactivation temperature range as similar differences in behaviour between persistent and non-persistent strains might also be expected.