Subtyping of Listeria monocytogenes isolates recovered from retail ready-to-eat foods, processing plants and listeriosis patients in Sweden 2010

Highlights

• 42% of the pulsotypes displayed by human isolates were also found in food isolates.

• 93% of food isolates were found on at least one occasion among human isolates.

• Persistence of one or more specific L. monocytogenes strains was indicated in plants.

Abstract

Identification and prioritisation of food safety interventions requires an understanding of the relationship between food, pathogens and cases. Such understanding can be gained through different approaches, e.g. microbial subtyping to attribute cases of foodborne disease to food vehicles or other sources of illness. In this study, Listeria monocytogenes isolates (n = 166) from (i) three categories of ready-to-eat (RTE) foods, (ii) food processing plant environments, and (iii) human listeriosis cases, all sampled during 2010 in Sweden, were subtyped. In addition, 121 isolates from human listeriosis cases, collected 2005–2009, were subtyped. Subtyping consisted of both serotyping (conventional method and PCR) and genotyping using pulsed-field gel electrophoresis (PFGE). Serotype 1/2a dominated in all three groups of isolates (range 73–96%). Eighteen percent of the human isolates (2010) belonged to serotype 4b, but only 1.4% of the food isolates. The food isolates differentiated into 19 pulsotypes (ID = 0.843), the human isolates collected 2010 into 31 pulsotypes (ID = 0.950) and the processing plant isolates into 22 pulsotypes (ID = 0.991). Six of the pulsotypes were shared between the food and human isolates. These pulsotypes comprised 42% of the human isolates and 59% of the food isolates. For some processing plants, there was suggested persistence of one or more specific L. monocytogenes strains, as indicated by repetitive isolation of the same pulsotype from food. This study indicated the presence of L. monocytogenes in the processing plant environment as a likely source of contamination of gravad and cold-smoked fish, and this food category as an important source of human exposure to the pathogen.

Introduction

Listeria monocytogenes is a food-borne pathogen that can cause listeriosis, a severe infection in humans recognised worldwide as a significant public health problem. There are two variants of listeriosis. One is an invasive form which primarily affects risk groups, i.e. pregnant women, infants, elderly and immune-suppressed individuals (Liu et al., 2007, Swaminathan and Gerner-Smidt, 2007), and typically presents as septicaemia, meningitis or meningoencephalitis, with most cases ending up being hospitalised (Todd and Notermans, 2011). The other is a self-limiting non-invasive variant, febrile gastroenteritis, which from a health perspective is not considered so serious and therefore not notifiable according to the Communicable Disease Act in Sweden (SMI, 2013a). There has been an increasing trend of reported cases of listeriosis in Sweden since the 1980's with a peak in 2009 and 2012 of 0.77 and 0.75 cases per 100 000 inhabitants (SMI, 2013b). An increasing trend has also been observed in several other European countries during the past 10 years, mainly in older age groups, owing to reasons that are still unknown (Goulet et al., 2008). Based on data from 2004–2007, the public health burden in Sweden due to listeriosis is an estimated 59–285 disability adjusted life years (DALY) per year (Lindqvist et al., 2011). Most of the public health burden (50–275 DALY) is associated with years of life lost due to premature deaths (YLL), and 4–25 DALY due to years lived with the disease (YLD). These ranges reflect uncertainty due to variation in the input data, but overall the data indicate a need for strategies to minimise illness due to contamination of food by L. monocytogenes.

L. monocytogenes is widespread in the natural environment and can in principle occur on all raw food products. Its ability to produce biofilms and persist in the plant environment for years (Wilks et al., 2006) makes it a frequent hazard detected in food production environments (Blatter et al., 2010, Carpentier and Cerf, 2011, Gandhi and Chikindas, 2007). Several studies have shown that the processing of ready-to-eat (RTE) food poses a high risk of contamination by L. monocytogenes (Kathariou, 2002, Miettinen and Wirtanen, 2006), indicating that operator awareness and control of this microorganism in production plants is of great importance. In addition, as low temperatures, vacuum packaging and modified atmosphere packaging permit growth of L. monocytogenes (Swaminathan and Gerner-Smidt, 2007), control along the food chain, including the consumer stage, is necessary. Consequently, awareness of the microbial hazard among consumers is also of great importance. However, several studies have indicated that consumers in general lack knowledge of the importance of observing certain refrigerator temperatures and storage times (Marklinder et al., 2004, Redmond and Griffith, 2003), highlighting the necessity to introduce effective measures to reduce L. monocytogenes at an early stage of food production. In the EU, the main L. monocytogenes concerns relate to RTE foods and criteria for L. monocytogenes in foodstuffs are specified in Regulation (EC) No. 2073/2005 (http://eur-lex.europa.eu/).

The incidence of listeriosis in Scandinavian countries is consistently higher than in other European countries (EFSA, 2011). This has been attributed to higher consumption of RTE fish in these countries (Todd and Notermans, 2011). Since most reported listeriosis cases in Sweden are sporadic and outbreaks are rare, the source of infection is most often unknown. An outbreak investigation (Ericsson et al., 1997) and a risk assessment (Lindqvist and Westöö, 2000) indicated RTE fish as an important source of listeriosis in Sweden, but more recent evidence is limited (Parihar et al., 2008, Peiris et al., 2009). However, in 2010, a national survey was conducted to examine three categories of RTE foods, gravad and smoked fish; soft and semi-soft cheese; and packaged meat products, with all samples collected at the retail level (Lambertz et al., 2012). Subtyping and comparison of the food and human isolates obtained may reveal potential sources of contamination of foods. Subtyping results on L. monocytogenes isolated from RTE foods at the retail level have been investigated extensively (Gilbreth et al., 2005, Gray et al., 2004, Liu, 2006, Martinez et al., 2003, Sauders et al., 2004). An important aspect of the study design when comparing subtyping results of food and human isolates is that the isolates must be collected during the same period and in the same geographical region (Gilbreth et al., 2005).

Various genotyping methods have been used successfully to distinguish closely related L. monocytogenes strains (Wiedmann, 2002). An indispensable tool for tracing L. monocytogenes contamination is molecular subtyping by pulsed-field gel electrophoresis (PFGE), whereby DNA profiles (pulsotypes) of isolates can be compared and matches revealed. Due to its highly discriminatory power (Gilbreth et al., 2005) and reproducibility (Brisabois et al., 2007, Martin et al., 2006), PFGE has proven to be one of the most useful subtyping methods for L. monocytogenes and has become the gold standard for subtyping, both within the European Union and in the USA (Gerner-Smidt et al., 2006). As regards serotyping, despite its low discriminatory power and labour intensive nature, conventional serotyping is still widely used. However, an alternative method involving PCR analysis, resulting in serogroups rather than serotypes, has recently been developed and is recommended by the EU-RL Lm to be used by EU Member States in parallel with conventional serotyping (Hong et al., 2007, Kérouanton et al., 2010).

The objectives of this study were to attempt to attribute listeriosis cases in Sweden to possible food vehicles, contamination sources and routes by microbial subtyping of L. monocytogenes strains isolated from three categories of RTE foods, human sporadic cases of listeriosis, and food processing plant environments in Sweden in 2010. In addition, to allow comparison with previous years, human listeriosis isolates collected in Sweden in the period 2005–2009 were subtyped.