Identifying suitableListeria innocuastrains as surrogates forListeria monocytogenesfor horticultural products

We conducted a laboratory-based study testing nine Listeria innocua strains independently and a cocktail of 11 Listeria monocytogenes strains. The aim was to identify suitable L. innocua strain(s) to model L. monocytogenes in inactivation experiments. Three separate inactivation procedures and a hurdle combination of the three were employed: thermal inactivation (55°C), UV-C irradiation (245 nm) and chemical sanitiser (Tsunami™ 100, a mixture of acetic acid, peroxyacetic acid and hydrogen peroxide). The responses were strain dependent in the case of L. innocua with different strains responding differently to different regimes. L. innocua isolates generally responded differently to the L. monocytogenes cocktail and had different responses among themselves. In the thermal inactivation treatment, inactivation of all strains including the L. monocytogenes cocktail plateaued after 120 minutes. Chemical sanitiser, inactivation could be achieved at concentrations of 10 and 20 ppm with inactivation increasing with contact time up to 8 minutes, beyond which there was no significant benefit. Although most of the L. innocua strains in the study responded similarly to L. monocytogenes when subjected to a single inactivation treatment, when the treatments were applied as hurdle, all L. innocua strains except PFR16D08 were more sensitive than the L. monocytogenes cocktail. PFR16D08 almost matched the resistance of the L. monocytogenes cocktail but was much more resistant to the individual treaments. A cocktail of two L. innocua strains (PFR 05A07 and PFR 05A10) had the closest responses to the hurdle treatment to those of the L. monocytogenes cocktail and is therefore recommended for hurdle experiments. Importance Owing to researcher safety risks it is often difficult to use actual pathogens, such as Listeria monocytogenes , to explore different inactivation procedures under field conditions. Organisms that are closely related to the pathogen but without its virulence are therefore used as surrogates for the actual pathogen. However, this assumes that the surrogate will behave in a similar manner to the pathogen and it is difficult to predict the responses of the surrogate compared to the actual pathogen. This study compares the responses of individual and combined “cocktails” of strains of non-pathogenic Listeria innocua to different inactivation procedures when compared to the response of a cocktail of L. monocytogenes . Our study highlights the importance of evaluating a number of strains when choosing surrogates.