Contents lists available at ScienceDirect International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro Mathematical modelling of temperature effect on growth kinetics of Pseudomonas spp. on sliced mushroom (Agaricus bisporus) Fatih Tarlak, Murat Ozdemir ⁎ , Mehmet Melikoglu Department of Chemical Engineering, Gebze Technical University, 41400 Gebze, Kocaeli, Turkey ARTICLE INFO Keywords: Predictive microbiology Mushroom spoilage Microbiological change Growth behaviour Shelf-life ABSTRACT The growth data of Pseudomonas spp. on sliced mushrooms (Agaricus bisporus) stored between 4 and 28 °C were obtained and fitted to three different primary models, known as the modified Gompertz, logistic and Baranyi models. The goodness of fit of these models was compared by considering the mean squared error (MSE) and the coefficient of determination for nonlinear regression (pseudo-R 2 ). The Baranyi model yielded the lowest MSE and highest pseudo-R 2 values. Therefore, the Baranyi model was selected as the best primary model. Maximum specific growth rate (r max ) and lag phase duration (λ) obtained from the Baranyi model were fitted to secondary models namely, the Ratkowsky and Arrhenius models. High pseudo-R 2 and low MSE values indicated that the Arrhenius model has a high goodness of fit to determine the effect of temperature on r max . Observed number of Pseudomonas spp. on sliced mushrooms from independent experiments was compared with the predicted number of Pseudomonas spp. with the models used by considering the B f and A f values. The B f and A f values were found to be 0.974 and 1.036, respectively. The correlation between the observed and predicted number of Pseudomonas spp. was high. Mushroom spoilage was simulated as a function of temperature with the models used. The models used for Pseudomonas spp. growth can provide a fast and cost-effective alternative to traditional microbiological techniques to determine the effect of storage temperature on product shelf-life. The models can be used to evaluate the growth behaviour of Pseudomonas spp. on sliced mushroom, set limits for the quantitative detection of the microbial spoilage and assess product shelf-life. 1. Introduction Mushrooms have been consumed as a source of food and medicine for centuries, because of their high amounts of proteins, minerals and bioactive compounds (Wani et al., 2010). The cultivated button mushroom (Agaricus bisporus) is the most common edible mushroom in the world. Agaricus bisporus has a very short shelf-life because it has no cuticle to protect it from physical deterioration or microbial attack (Brennan et al., 2000). Although no outbreak of pathogenic micro- organisms such as E. coli O157:H7 and L. monocytogenes has been re- ported (Guan et al., 2012) for Agaricus bisporus, it is very susceptible to contamination with Pseudomonas spp. which are abundant in nature (González-Fandos et al., 2006; Simón et al., 2005). The Pseudomonas spp. are responsible for causing spoilage, and the initial count of Pseudomonas spp. on cultivated mushrooms is quite high, ranging from 6.9 to 8.1 log 10 CFU/g (Simón et al., 2005; Venturini et al., 2011). Predictive food microbiology aims to estimate the microbial growth using mathematical models under different conditions. These mathe- matical models are generally classified into three main categories known as primary, secondary and tertiary models (Whiting, 1995). Primary models describe the growth data as a function of time under a constant environmental condition. Sigmoidal type models such as the modified Gompertz, logistic and Baranyi models are widely used as primary models for fitting microbial growth data. Secondary models describe the effects of environmental factors, such as temperature, pH and water activity (a w ) on the parameters of the primary models, in- cluding maximum specific growth rate and lag phase duration. One of the most important environmental factors from the food safety point of view is temperature. The most widely used secondary model to de- termine the relationship between temperature and maximum specific growth rate is known as the Ratkowsky or square root model (Ratkowsky et al., 1982). Tertiary models combine both the primary and secondary models with user-friendly application software or expert systems to assess microbial behaviour under specific conditions (Wang et al., 2013; Whiting, 1995). Predictive models are considered as important tools to assess pro- duct shelf-life and food safety, to perform hazard analysis and set cri- tical control points, and to develop risk assessment plans. Predictive https://doi.org/10.1016/j.ijfoodmicro.2017.12.017 Received 31 October 2016; Received in revised form 24 October 2017; Accepted 17 December 2017 ⁎ Corresponding author. E-mail address: ozdemirm@gtu.edu.tr (M. Ozdemir). International Journal of Food Microbiology 266 (2018) 274–281 Available online 18 December 2017 0168-1605/ © 2017 Elsevier B.V. All rights reserved. T