Mapping the risk of evaporated milk spoilage in the Mediterranean region based on the effect of temperature conditions on Geobacillus stearothermophilus growth

Accepted Manuscript Mapping the risk of evaporated milk spoilage in the Mediterranean region based on the effect of temperature conditions on Geobacillus stearothermophilus growth

Myrsini Kakagianni, Konstantinos P. Koutsoumanis PII: DOI: Reference:

S0963-9969(18)30358-2 doi:10.1016/j.foodres.2018.05.002 FRIN 7591

To appear in:

Food Research International

Received date: Revised date: Accepted date:

3 April 2018 2 May 2018 3 May 2018

Please cite this article as: Myrsini Kakagianni, Konstantinos P. Koutsoumanis , Mapping the risk of evaporated milk spoilage in the Mediterranean region based on the effect of temperature conditions on Geobacillus stearothermophilus growth. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Frin(2017), doi:10.1016/j.foodres.2018.05.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Mapping the risk of evaporated milk spoilage in the Mediterranean region based on the effect of temperature conditions on Geobacillus stearothermophilus growth

Myrsini Kakagianni, Konstantinos P. Koutsoumanis*

PT

Laboratory of Food Microbiology and Hygiene, Department of Food Science and

RI

Technology, School of Agriculture, Faculty of Agriculture, Forestry and Natural

SC

Environment, Aristotle University of Thessaloniki, Thessaloniki, Greece 54124.

NU

*Corresponding author: Konstantinos P. Koutsoumanis, Aristotle University of Thessaloniki, Department of Food Science and Technology, School of Agriculture, Faculty of Agriculture,

MA

Forestry and Natural Environment, Thessaloniki, Greece 54124. Phone: +30 2310991647, Fax: +30 2310991647, e-mail: [email protected]

AC

CE

PT E

D

E-mail address: [email protected] (Myrsini Kakagianni)

ACCEPTED MANUSCRIPT Abstract A predictive model for the effect of storage temperature on the growth of Geobacillus stearothermophilus was applied in order to assess the risk of evaporated milk spoilage in the markets of the Mediterranean region. The growth of G. stearothermophilus in evaporated milk was evaluated during a shelf life of one year based on historical temperature profiles

PT

(hourly) covering 23 Mediterranean capitals for five years over the period 2012 - 2016

RI

obtained from the Weather Underground database (http://www.wunderground.com/). In total,

SC

115 scenarios were tested simulating the distribution and storage conditions of evaporated milk in the Mediterranean region. The highest growth of G. stearothermophilus was predicted

NU

for Marrakech, Damascus and Cairo with mean values over the period 2012 - 2016 of 7.2, 7.4 and 5.5 log CFU/ml, respectively, followed by Tunis, Podgorica and Tripoli with mean

MA

growth of 2.8, 2.4 and 2.3 log CFU/ml, respectively. For the rest 17 capitals the mean growth of the spoiler was less than 1.5 log CFU/ml. The capitals Podgorica, Cairo, Tunis and Ankara

D

showed the highest variability in the growth during the 5 years examined with standard

PT E

deviation values for growth of 2.01, 1.79, 1.77 and 1.25 log CFU/ml, respectively. The predicted extent and the variability of growth during the shelf life were used to assess the risk

CE

of spoilage which was visualised in a geographical risk map. The growth model of G. stearothermophilus was also used to evaluate adjustments of the evaporated milk expiration

AC

date which can reduce the risk of spoilage. The quantitative data provided in the present study can assist the food industry to effectively evaluate the microbiological stability of these products throughout distribution and storage at a reduced cost (by reducing sampling quality control) and assess whether and under which conditions (e.g. expiration date) will be able to export a product to a country without spoilage problems. This decision support may lead to a significant benefit for both the competitiveness of the food industry and the consumer.

ACCEPTED MANUSCRIPT

Keywords: Geobacillus stearothermophilus, evaporated milk, risk of spoilage,

AC

CE

PT E

D

MA

NU

SC

RI

PT

Mediterranean countries

ACCEPTED MANUSCRIPT 1.

Introduction Shelf - stable foods have undergone preservation processes that have produced foods that

are considered to be ‘commercially sterile’, i.e., they will not spoil or cause disease under normal conditions of handling and distribution (Zottola, 2003). Shelf - stable foods include foods of a type that, because of their composition (low moisture, high acidity, high salt or

PT

sugar content), do not require any special storage conditions as they generally do not provide

RI

conditions suitable for the growth of microorganisms, or foods which have been processed so that they can be safely stored at room temperature for a long time (NZFSA, 2016). Regarding

SC

the second case, the microbiological stability of these, non – perishable, food products is

NU

based on the intense thermal processing that kills the vegetative cells of microorganisms. Evaporated milk is a product of high consumption belonging to the latter category.

MA

Although the heat processing of evaporated milk eliminates all vegetative bacterial cells, the presence of bacterial spores in the final product is possible (Chen, Coolbear, & Daniel,

D

2004; Cosentino, Mulargia, Pisano, Tuveri, & Palmas, 1997). Indeed, the spore - forming

PT E

bacterium Geobacillus stearothermophilus constitutes one of the most important quality problems for evaporated milk products. The high prevalence and concentration of spores in

CE

the raw material, the adhesive characteristics of spores that enhance their persistence in industrial environments and, most importantly, the extreme heat resistance of the spores are

AC

among the major factors explaining the importance of this bacterium (André, Zuber, & Remize, 2013; Postollec et al., 2012; Yoo, Hardin, & Chen, 2006). At favorable storage conditions, G. stearothermophilus spores that survived the heat process may subsequently germinate and grow out leading to sensory rejection. The resulting spoilage is characterised by the coagulation of milk caused by acid production due to the metabolic activity of the cells (Boor & Murphy, 2002; Hill & Smythe, 2012; Kakagianni, Gougouli, & Koutsoumanis, 2016; Kalogridou-Vassiliadou, 1992; Yoo et al., 2006).

ACCEPTED MANUSCRIPT The microbiological stability of evaporated milk is mainly based on the thermophilic nature of G. stearothermophilus spores that may be present in the final product which does not allow their germination and growth at temperatures below 33 - 40 °C (Burgess, Lindsay, & Flint, 2010; Hill & Smythe, 2012; Kakagianni et al., 2016; Llaudes, Zhao, Duffy, & Schaffner, 2001; Ng & Schaffner, 1997; Oomes et al., 2007). Unlike other parameters

PT

affecting food quality (e.g. pH, water activity, redox potential) however, the temperature

RI

along the supply chain of non – refrigerated products, including transportation and retail

SC

storage, is out of the direct control of the manufacturer and often deviates from the specifications. Moreover, temperature control is completely lacking during transportation

NU

from the retail stores to consumers’ shelf and during domestic storage until the time of preparation and consumption (Koutsoumanis & Gougouli, 2015). Temperature abuses during

MA

any stage of the supply chain may result in an unexpected loss of quality. Thus, recording the temperature conditions of a supply chain and evaluating their effect on the microbiological

D

stability of the products are of paramount importance in a food quality management system

PT E

especially for countries with hot climate (Koutsoumanis, Taoukis, & Nychas, 2005; Nychas, Panagou, & Mohareb, 2016; Nychas, Skandamis, Tassou, & Koutsoumanis, 2008).

CE

The objective of the present study was to apply predictive microbiology tools for mapping the risk of evaporated milk spoilage in the Mediterranean region based on the effect

AC

of storage temperature on G. stearothermophilus growth. Predictive microbiology models can assist the food industry to effectively evaluate the microbiological stability of these products throughout distribution and storage at a reduced cost (by reducing quality control sampling) and assess whether and under which conditions (e.g. expiration date) will be able to export a product to a country without spoilage problems. This decision support may lead to a significant benefit for both the competitiveness of the food industry and the consumer.

ACCEPTED MANUSCRIPT 2.

Materials and Methods

2.1. Prediction of Geobacillus stearothermophilus growth in evaporated milk The predictive model of Kakagianni et al. (2016) describing the effect of temperature on the growth of Geobacillus stearothermophilus ATCC 7953 was applied to assess the risk of spoilage of evaporated milk during distribution and storage in the market of 23

PT

Mediterranean capitals. The prediction of growth was based on the combination of the

RI

secondary Cardinal Model with Inflection (Rosso, Lobry, & Flandrois, 1993) with the

SC

differential equations of the Baranyi and Roberts primary model (Baranyi & Roberts, 1994). The effect of temperature on the maximum specific growth rate (μmax) was predicted by the

( )( (

)( )(

MA

{(

(1)

)

) (

)(

)

PT E

( )

( )

D

( )

NU

secondary model as follows:

(2)

where Tmin = 33.76 °C, Topt = 61.82 °C and Tmax = 68.14 °C are the theoretical minimum, = 2.068/h is the

CE

optimum and maximum temperature for growth, respectively, and

AC

optimum value for the maximum specific growth rate (when T=Topt).

The prediction of growth under dynamic storage temperature conditions was based on the time - temperature profile of evaporated milk storage, T(t), in conjunction with the secondary model (Eq. 1) for the estimation of the “momentary” max and the differential equations of Baranyi and Roberts model (Eq. 3 and 4), which were numerically integrated with respect to time:

ACCEPTED MANUSCRIPT

Where

( )

(

( )

(4)

( )

)

(3)

is the maximum specific growth rate and

is the maximum spopulation

PT

density. The parameter q denotes the concentration of a substance critical to growth and is

NU

SC

RI

related to the physiological parameter a0 as follows:

The model was based on the assumption that after a temperature shift, the growth rate is

MA

adopted instantaneously to the new temperature environment. Hourly temperature data from Mediterranean capitals for five years over the period 2012 - 2016 were used as model data

D

input serving to predict growth and impeding spoilage of the evaporated milk. For the growth

PT E

prediction the maximum population density was fixed at 107.4 CFU/ml based on the value reported by Kakagianni et al. (2016). The a0 was set to 1 representing a worst case scenario

CE

with the spoilage organism growing with no lag phase.

AC

2.2. Collection of temperature data Historical temperature data covering 23 Mediterranean capitals for five years over the period

2012

-

2016

was

obtained

from

the

Weather

Underground

database

(http://www.wunderground.com/). Data was collected from all recording airport weather stations. For the capitals of Slovenia, Spain, Cyprus and Israel, no hourly temperature data was available in the database and therefore temperature data was collected from the next highly populated city.. The temperature data was collected at 1, 3 or 6 hour intervals,

ACCEPTED MANUSCRIPT depending on the station. Τhe extracted information was in the form of time – temperature data in Excel worksheets.

2.3 Mapping the risk of evaporated milk spoilage in Mediterranean capitals The temperature data collected from 23 Mediterranean capitals was introduced into the

PT

developed model, according to the Section 2.1, to predict the growth of G.

RI

stearothermophilus under storage conditions corresponding to five years over the period 2012

SC

– 2016 for each capital. The total growth was predicted for a storage period of one year representing the expiration date currently applied to evaporated milk by a Greek dairy

NU

industry. Box - plots were used to present the growth data, providing the principal measures of central tendency and dispersion. In the Box - plot representation (Fig. 3), the bottom and

MA

top of the box are the 25th and 75th percentiles (Q1 and Q3, respectively), the band is the median and the dots correspond to the minimum and maximum value. The total growth was

D

further used to evaluate the risk that a product exported to a certain Mediterranean capital

PT E

will be spoiled before it expires. For this, a maximum initial contamination level of 101 spores/ml observed by a Greek dairy industry and a spoilage level of 107.4 CFU/ml reported

CE

by Kakagianni et al. (2016) for the products of the same industry were considered. Based on the above, the time - to - spoilage can be predicted as the time required for G.

AC

stearothermophilus population to increase by 6.4 log CFU/ml. Taking into account the variability of the total growth (during 1 year shelf life) over the five years tested, the risk of spoilage was categorized in the levels of high, moderate, low and very low as described in the Section 3 (Results and Discussion section).

3.

Results and Discussion

ACCEPTED MANUSCRIPT The main objective of the present study was to provide information to the dairy industry on the risk of spoilage of evaporated milk exported to Mediterranean capitals. In a previous study (Kakagianni et al., 2016) we have shown that spoilage of this product is due to acid coagulation observed when the pH of evaporated milk decreases to a level around value of 5.2 due to the growth and metabolic activity of G. stearothermophilus. Indeed, G.

PT

stearothermophilus growth in evaporated milk results in acid production enhancing the

RI

formation of protein aggregates which is mainly related to unfolding and gelation of β -

SC

lactoglobulin (Hill & Smythe, 2012; Yoo et al., 2006). Considering the high prevalence of G. stearothermophilus in raw milk and the fact that its spores are extremely heat resistant (André

NU

et al., 2013; Membré and van Zuijlen, 2011; Postollec et al., 2012; Simmonds, Mossel, Intaraphan & Deeth, 2003; Yoo et al., 2006), the spoilage of evaporated milk depends almost

MA

exclusively on the time - temperature conditions at which the product is exposed to during distribution and storage. Research data support that germination, outgrowth and subsequent

D

growth of G. stearothermophilus spores do not occur below 33 – 40 °C (Burgess et al., 2010;

PT E

Hill & Smythe, 2012; Kakagianni et al., 2016; Llaudes et al., 2001; Ng & Schaffner, 1997; Oomes et al., 2007). However, the conditions prevailing during the evaporated milk supply

CE

chain in some Mediterranean countries may exceed the above minimum temperature range for growth. The important question for the dairy industry is whether the climate of a

AC

Mediterranean capital allows sufficient growth of G. stearothermophilus resulting in a high (unacceptable) risk of spoilage, or in other words, the problem is based on the identification of the capitals which allow exporting of evaporated milk with a low (acceptable) risk of spoilage. Predictive microbiology can be used as an effective tool in answering the above questions. A predictive model for G. stearothermophilus growth (Kakagianni et al., 2016) was applied to assess the risk of spoilage of evaporated milk, with 1 year shelf – life produced by

ACCEPTED MANUSCRIPT a Greek dairy industry, when exported to 23 Mediterranean capitals. Since evaporated milk is a non - refrigerated product, predictions of growth during a one year of storage were based on historical

temperature

data

obtained

from

a

meteorological

database

(http://www.wunderground.com/). In order to take into account the year – to - year variability of the temperature conditions, data for five years, over the period 2012 – 2016, was collected

PT

and used. Fig. 1 shows the annual temperature profiles of a representative capital (Tunis) for

RI

the five year period. As expected, temperature data is characterized by a high daily and

SC

seasonable variance. Differences in the annual temperature profiles among the years are also observed. For the five years examined, the minimum, maximum, mean and standard

19.3 to 20.8 and 6.3 to 7.6oC, respectively.

NU

deviation of the annual temperature in the latter country ranges from 2.0 to 4.0, 40.0 to 42.0,

MA

The predictive model for G. stearothermophilus can be used to translate the temperature data to total growth of the spoiler during a storage period of one year (shelf life). It needs to

D

be noted that the efficiency of the model used in predicting G. stearothermophilus growth at

PT E

–dynamic temperature conditions has been previously evaluated with extensive validation experiments. Indeed, Kakagianni et al. (2016) compared the predictions of the model with

CE

observed growth under several dynamic temperature profiles which included temperatures inside and outside the growth region of the microorganism. For all the temperature scenarios

AC

tested in the latter study, the model adequately predicted the growth of G. stearothermophilus in evaporated milk, suggesting a good performance of the model. In Fig. 2, examples of the predicted growth of G. stearothermophilus in evaporated milk are illustrated for three capitals of the Mediterranean region (Athens, Tunis and Damascus) for the years 2012, 2014 and 2015, respectively. As shown, depending mainly on the climate of each city but also on the year, the growth of the spoiler may range from negligible (Athens, 2012) to significant exceeding the spoilage threshold of 107.4 CFU/ml (Damascus, 2015).

ACCEPTED MANUSCRIPT The box – plots displaying the 25th, 50th and 75th percentiles and the extreme values of growth for the five years are plotted in Fig. 3 for the 23 Mediterranean capitals examined. The box - plots illustrate both the extend of G. stearothermophilus growth in evaporated milk during one year of storage but also the variability in the growth among the five years examined. The highest growth of G. stearothermophilus was predicted for Marrakech,

PT

Damascus and Cairo with mean values over the period 2012 - 2016 of 7.4, 7.2 and 5.5 log

RI

CFU/ml, respectively, followed by Tunis, Podgorica and Tripoli with mean growth of 2.8, 2.4

SC

and 2.3 log CFU/ml, respectively. For the rest 17 capitals the mean growth of the spoiler was less than 1.5 log CFU/ml. The capitals with the highest variability in the growth among the 5

NU

years examined were Podgorica, Cairo, Tunis and Ankara with standard deviation values for growth of 2.01, 1.79, 1.77 and 1.25 log CFU/ml, respectively.

MA

The predicted growth of G. stearothermophilus was further used as the basis for ranking the 23 Mediterranean capitals in relation to the risk of spoilage of evaporated milk products

D

distributed and stored within their region. Based on a maximum initial contamination level of

PT E

101 spores/ml observed by a Greek dairy industry and a spoilage level of 107.4 CFU/ml reported by Kakagianni et al. (2016) the time - to - spoilage can be predicted as the time

CE

required for G. stearothermophilus population to increase by 6.4 log CFU/ml. However, apart from the extent of growth, it is important to take into account the variability of growth among

AC

the years in assessing the risk. Given the considerations of a Greek dairy industry about the acceptable (low) level of risk, the following risk categorization was set based on both the mean growth (Mg) during one year of storage and the standard deviation of growth (SDg) among the 5 years examined:



High Risk: Mg > 4 logs CFU/ml and Mg + 1*SDg > 6 logs CFU/ml



Moderate Risk: Mg < 4 logs CFU/ml and Mg + 2*SDg > 6 logs CFU/ml

ACCEPTED MANUSCRIPT 

Low Risk: Mg < 3 logs CFU/ml and Mg + 3*SDg > 4 logs CFU/ml



Very Low Risk: Mg < 2 logs CFU/ml and Mg + 3*SDg < 4 logs CFU/ml

Fig. 4 illustrates the geographical risk assessment for evaporated milk spoilage in the Mediterranean region. As shown Marrakech, Damascus and Cairo present a high risk with

PT

Mg + 1*SDg of G. stearothermophilus growth for the five years exceeding 7 log CFU/ml.

RI

Tunis and Podgorica present a moderate risk Mg + 2*SDg exceeding 6 logs CFU/ml. Tripoli,

SC

Ankara and Amman present a low risk (Mg + 3*SDg > 4.5 log CFU/ml) while all the rest capitals present a very low risk of spoilage (Mg + 3*SDg < 3.5 logs CFU/ml). It needs to be

NU

noted the above categorization of risk is also a risk management decision and should be based on the relation between the information provided by the model and the quality requirements

MA

of the food industry. In addition, the risk depends on the initial contamination level G. stearothermophilus. In this study, a maximum initial contamination level of 101 spores/ml

D

observed by a Greek dairy industry was considered but any changes to this level should be

PT E

taken into account in risk evaluation.

The growth model of G. stearothermophilus was also used to evaluate adjustments of the

CE

evaporated milk expiration date which can reduce the risk of spoilage. Fig. 5 shows the growth of G. stearothermophilus for scenarios in which products with a shelf life of six

AC

months are exported to Damascus for the periods October - March, January - June and March - August based on the temperature data of the years 2014 - 2015. It can be readily seen that for the first two scenarios, which do not include the summer period, the risk of spoilage is significantly reduced since the growth of G. stearothermophilus during distribution and storage of evaporated milk is negligible. However, as shown in Fig. 5c, even with a reduced shelf life, products which are distributed and stored during the summer period present a high risk of spoilage. Based on the above prediction, the dairy industry can manage the shelf life

ACCEPTED MANUSCRIPT of the product in combination of the season of exporting in order to reduce the risk of spoilage.

4.

Conclusions In conclusion, the present study focuses on the prediction of G. stearothermophilus

PT

growth in evaporated milk as a basis for assessing the risk of spoilage for products exported

RI

to Mediterranean capitals. The quantitative data provided is useful for the identification of the

SC

capitals which allow exporting of evaporated milk with a low (acceptable) risk of spoilage and support quality management in the dairy industry. Further studies on the effect of heat

NU

processing on the germination time of G. Stearothermophilus spores or on the growth/no growth interface of the spoiler could improve the accuracy of the predictions.

Future

MA

applications of the predictive microbiology approach applied in this study may also include the evaluation of the effect of climate change on food quality. The results of the study

D

showed that based on the temperature data of the last years, the storage and distribution of

PT E

evaporated milk in most Mediterranean capitals present a low risk of spoilage. Temperature, however, is expected to increase significantly due to the global warming phenomenon.

CE

Recent studies presented climate change projections over the Mediterranean region based on the latest and most advanced sets of global and regional climate model simulations. These

AC

simulations give a collective picture of a substantial warming ranging from 1 to 7 °C (Giorgi & Lionello, 2008; IPCC, 2014; Räisänen et al., 2004). The approach presented in this study can be used to evaluate the consequences of global warming on the quality of non refrigerated foods and help the food industry to design effective quality management and logistic systems.

ACCEPTED MANUSCRIPT

Acknowledgments This research was carried out with the financial support of “Understanding the impact of manufacturing processes in the ecology of microorganisms that spoil-contaminate milk products (ESL, evaporated milk) and fresh fruit juices – Development of molecular

PT

methodologies and mathematical models for the prediction of their shelf-life” within the

RI

framework of the action “Cooperation” (NSRF 2007-2013), that was co-financed by the

SC

European Social Fund (ESF) and National Resources. The authors are also grateful to

AC

CE

PT E

D

MA

NU

Dimitrios Vasileiou for his assistance in compiling the data, through coding.

ACCEPTED MANUSCRIPT References André, S., Zuber, F., & Remize, F. (2013). Thermophilic spore-forming bacteria isolated from spoiled canned food and their heat resistance. Results of a French ten-year survey. International Journal of Food Microbiology, 165(2), 134-143. Baranyi, J., & Roberts, T. A. (1994). A dynamic approach to predicting bacterial growth in food. International Journal of Food Microbiology, 23(3-4), 277-294.

PT

Boor, K. J., & Murphy, S. C. (2002). Microbiology of market milks, In Robinson R. K. (Ed.), Dairy Microbiology Handbook: The Microbiology of Milk and Milk Products (pp. 91-

RI

122). John Wiley & Sons Inc., Hoboken, NJ.

Burgess, S. A., Lindsay, D., & Flint, S. H. (2010). Thermophilic bacilli and their importance

SC

in dairy processing. International Journal of Food Microbiology, 144(2), 215-225. Chen, L., Coolbear, T., & Daniel, R. M. (2004). Characteristics of proteinases and lipases

NU

produced by seven Bacillus sp. isolated from milk powder production lines. International Dairy Journal, 14(6), 495-504.

MA

Cosentino, S., Mulargia, A., Pisano, B., Tuveri, P., & Palmas, F. (1997). Incidence and biochemical characteristics of Bacillus flora in Sardinian dairy products. International Journal of Food Microbiology, 38(2), 235-238.

D

Giorgi, F., & Lionello, P. (2008). Climate change projections for the Mediterranean region.

PT E

Global and Planetary Change, 63(2-3), 90-104. Hill, B. M., & Smythe, B. W. (2012). Endospores of thermophilic bacteria in ingredient milk powders and their significance to the manufacture of sterilized milk products: an

CE

industrial perspective. Food Reviews International, 28(3), 299-312. IPCC (Intergovernmental Panel on Climate Change) (2014). Climate Change 2014: Synthesis

AC

Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Kakagianni, M., Gougouli, M., & Koutsoumanis, K. P. (2016). Development and application of Geobacillus stearothermophilus growth model for predicting spoilage of evaporated milk. Food Microbiology, 57, 28-35. Kalogridou - Vassiliadou, D. (1992). Biochemical activities of Bacillus species isolated from flat sour evaporated milk. Journal of Dairy Science, 75(10), 2681-2686. Koutsoumanis, K. P., Gougouli, M. (2015). Use of Time Temperature Integrators in food safety management. Trends in Food Science & Technology, 43(2), 236-244. Koutsoumanis, K., Taoukis, P., & Nychas, G. (2005). Development of a safety

ACCEPTED MANUSCRIPT monitoring and assurance system for chilled food products. International Journal of Food Microbiology, 100(1), 253-260. Laudes, M. K., Zhao, L., Duffy, S., & Schaffner, D. W. (2001). Simulation and modelling of the effect of small inoculum size on time to spoilage by Bacillus stearothermophilus. Food Microbiology, 18(4), 395-405. Membré, J.-M., & Van Zuijlen, A. (2011). A probabilistic approach to determine thermal

PT

process setting parameters: application for commercial sterility of products. International Journal of Food Microbiology, 144(3), 413-420.

Ng, T. M., & Schaffner, D. W. (1997). Mathematical models for the effects of pH,

RI

temperature, and sodium chloride on the growth of Bacillus stearothermophilus in salty

SC

carrots. Applied and Environmental Microbiology, 63(4), 1237-1243. Nychas, G.-J. E., Panagou, E. Z., & Mohareb, F. (2016). Novel approaches for food safety

NU

management and communication. Current Opinion in Food Science, 12, 13-20. Nychas, G.-J. E., Skandamis, P. N., Tassou, C. C., & Koutsoumanis, K. P. (2008). Meat spoilage during distribution. Meat Science, 78(1), 77-89.

MA

NZFSA (New Zealand Food Safety Authority) (2016). A guide to calculating shelf life in foods. Available online at http://www.foodsafety.govt.nz/elibrary/industry/determine-

D

shelf-life-of-food/.

Oomes, S. J. C. M., van Zuijlen, A. C. M., Hehenkamp, J. O., Witsenboer, H., van der

PT E

Vossen, J. M. B. M., & Brul, S. (2007). The characterisation of Bacillus spores occurring in the manufacturing of (low acid) canned products. International Journal of Food Microbiology, 120(1), 85-94.

CE

Postollec, F., Mathot, A.-G., Bernard, M., Divanac'h, M.-L., Pavan, S., & Sohier, D. (2012). Tracking spore-forming bacteria in food: from natural biodiversity to selection by

AC

processes. International Journal of Food Microbiology, 158(1), 1-8. Räisänen, J., Hansson, U., Ullerstig, A., Döscher, R., Graham, L. P., Jones, C., Meier, H. E. M., Samuelsson, P., & Willén, U. (2004). European climate in the late twenty-first century: regional simulations with two driving global models and two forcing scenarios. Climate Dynamics, 22(1), 13-31. Rosso, L., Lobry, J., & Flandrois, J. (1993). An unexpected correlation between cardinal temperatures of microbial growth highlighted by a new model. Journal of Theoretical Biology 162(4), 447-463. Simmonds, P., Mossel, B. L., Intaraphan, T., & Deeth, H. C. (2003). Heat resistance of Bacillus spores when adhered to stainless steel and its relationship to spore

ACCEPTED MANUSCRIPT hydrophobicity. Journal of Food Protection, 66(11), 2070-2075. Yoo, J.-A., Hardin, M. T., & Chen, X. D. (2006). The influence of milk composition on the growth of Bacillus stearothermophilus. Journal of Food Engineering, 77(1), 96-102. Zottola, E. A. (2003). Bacterial Spoilage. In B. Caballero, P. M. Finglas, & F. Toldra (Eds.),

AC

CE

PT E

D

MA

NU

SC

RI

PT

Encyclopedia of Food Sciences and Nutrition (pp. 5506-5510), Academic Press.

ACCEPTED MANUSCRIPT Figure Captions

Fig. 1. Representative examples of historical hourly temperature data for Tunis for the years 2012 (a), 2013 (b), 2014 (c), 2015 (d) and 2016 (e) obtained from the Weather Underground

PT

database. Time = 0 corresponds to January 1st.

RI

Fig. 2. Representative examples of Geobacillus stearothermophilus growth prediction (red

SC

lines) in evaporated milk with a shelf life of one year in the supply chain of 3 Mediterranean capitals based on the historical hourly temperature data (blue lines) obtained from the

NU

Weather Underground database.

MA

a: Athens, (data from 2012); b: Tunis, (data from 2014); c: Damascus, (data from 2015)

Fig. 3. Box - plot representations of the total predicted growth of Geobacillus

D

stearothermophilus in evaporated milk with a shelf life of one year for the supply chain of 23

PT E

Mediterranean capitals for five years (2012 - 2016). The bottom and top of the box are the 25th and 75th percentiles (Q1 and Q3, respectively), the band is the median and the dots

CE

correspond to the minimum and maximum value.

AC

Fig. 4. Geographical risk assessment for evaporated milk spoilage in the Mediterranean region. Risk is assessed based on the predicted growth of G. stearothermophilus in evaporated milk with a shelf life of one year for the supply chain of the 23 Mediterranean capitals for five years (2012 - 2016)iRed: High risk, Orange: Moderate Risk, Yellow: Low Risk, Green: Very Low Risk

ACCEPTED MANUSCRIPT Fig. 5. Predicted growth of Geobacillus stearothermophilus (red lines) for scenarios in which evaporated milk with a shelf life of six months is exported to Damascus for the periods October - March (a), January - June (b) and March - August (c) based on the hourly

AC

CE

PT E

D

MA

NU

SC

RI

PT

temperature data (blue lines) of the years 2014 - 2015.

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

Graphical abstract

ACCEPTED MANUSCRIPT Highlights:

CE

PT E

D

MA

NU

SC

RI

PT

Growth of G. stearothermophilus in evaporated milk was predicted Prediction was based on temperature data from 23 capitals in Mediterranean region Higher growth predicted for Marrakech, Damascus and Cairo Podgorica, Cairo, Tunis and Ankara showed the higher variability in the growth Spoilage risk map was developed based on the extent and the variability of growth

AC

    

Graphics Abstract

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5