Risk factors favoring the presence of Listeria monocytogenes in Colombian pork-meat processing plants

Laboratorio Microbiología de Alimentos, Grupo de Biotecnología Ambiental e Industrial (GBAI). Departamento de Microbiología. Facultad de Ciencias. Pontificia Universidad Javeriana. Bogotá, D.C., Colombia. Laboratorio de Biotecnología Molecular, Grupo de Biotecnología Ambiental e Industrial (GBAI). Departamento de Microbiología. Facultad de Ciencias. Pontificia Universidad Javeriana. Bogotá, D.C., Colombia.

Listeriosis is a foodborne disease caused by L.
monocytogenes that is acquired in 99% of the cases by consuming contaminated food.The illness can manifest itself as invasive or non-invasive disease.The invasive form of the disease is characterized by spread of a severe-and/or local-infection in the central nervous system (CNS), which can result in death.In addition, it involves complications such as meningitis, and perinatal infections (Korkeala and Siitonen, 2003).On the other hand, the non-invasive form is associated with fever and gastrointestinal symptoms with no complications (Aureli et al., 2000;Pichler et al., 2009).Up to date 13 potentially pathogenic serovars of L. monocytogenes have been identified, with 4b (37 -64%), 1/2b and 1/2a (15 -25%), and 1/2c (0 -4%) the most frequent outbreaks and sporadic cases of the disease (Jacquet et al., 1995;Jay, 1996;Miyasaki et al., 2009;Markkula et al., 2011).The population identified at risk for this disease include immunocompromised individuals, pregnant women, infants and adults over 65 years (López et al., 2007;Garmendia et al., 2008;Mammina et al., 2009;Taillefer et al., 2010).
Despite the low incidence of listeriosis and far of surpassing other foodborne pathogens such as Salmonella, listeriosis outbreaks are associated with high mortality (20 and 30%), (Mead et al., 1999;FAO/OMS, 2004).The European Union and the United States of America report an annual incidence of 0.3 per 100,000 and 0.4 per 100,000 inhabitants, respectively (Centers for Disease Control and Prevention Morbidity and Mortality CDC, 2008;EFSA, 2008).Reports from Africa, Asia and South America have been scarce (Rocourt et al., 2003).However, recent reports indicate an increase in the number of cases, possibly associated with an increasing population at risk, in addition to changing trends in food consumption (Taillefer et al., 2010;Scallan et al., 2011).
In Colombia reports on foodborne disease incidence is scarce.The epidemiological data for L. monocytogenes shows few cases, probably associated with subregistration (Vanegas et al., 2009).However, the "Instituto Nacional de Vigilancia de Medicamentos y Alimentos" (INVIMA), the Colombian equivalent of the US FDA, include a surveillance for the pathogen in their programs of food Inspection, Monitoring and Control (IVC).Recent studies have demonstrated the presence of this bacterium in the Colombian pork-industry.Furthermore, studies identified 4b as the most common serovar in contaminated RTE products.However, there is few data on contamination sources or microorganism control measures (Muñoz et al., 2011).
Considering the necessity to provide information on the country's status with respect to L. monocytogenes in meat processing plants, our objective was to evaluate and document the risk factors favoring the presence of L. monocytogenes in Colombian meat processing plants (pressed ham (sandwich type), standard sausage, and sausage "chorizo").

Selection of meat products plants
Five meat processing plants were selected based on the following criteria: 1. Surveyed by the "Instituto Nacional de Vigilancia de Medicamentos y Alimentos" (INVIMA) as a registered plant.2. Located in areas one and two, where the largest number of pork meat processing plants are (Figure 1). 3. Production of at least one of the following meat products: pressed ham (sandwich type), standard sausage, and sausage "chorizo".4. Use of pork-meat as an ingredient in other processed meat products.

Polls
Each plant was assessed to establish factors that might favor the presence of L. monocytogenes.At question were aspects related to: technology level, infrastructure, type of utensils and processing equipment, blanching time and temperature, use of antimicrobials, and frequency of cleaning and disinfection, water quality, raw meat storage temperature, finished product and operator training, among others.

Sampling numbers
Two samplings were conducted for each of the plants included in this study.However, plant 4 included four samplings due to its complexity.

Microbiological analysis
Samples were transported to the laboratory under aseptic conditions and processed immediately.The method for processing environmental samples was described by the United States Department of Agriculture, (USDA-FSIS, 2002).For meat cuts, the standard method ISO11290-1 was used (ISO, 1996).Isolates presumptive of L. monocytogenes were subjected to molecular identification (genus and species).

Genomic DNA purification and quantification
For genomic DNA extraction 100 l of each isolate stock was cultivated in BHI supplemented with 0.5% (w/v) glucose during 12 h at 37°C and 250 rpm (Ruiz-Bolivar et al., 2011).One millilitre of culture was taken for DNA purification using the Wizard® Genomic DNA Purification Kit (Promega, Madison, WI USA).DNA purity and concentration were determined (NanoDrop 2000c, Thermo Scientific, Waltham, MA, USA) (Sambrook and Russell, 2001).

Characterization of the risk factors
Prevalence of L. monocytogenes risk characterization in each type of tested surface was analyzed, and the following scale was assigned: 0% insignificant, 1-10% low, 11-30% medium and high with values above 30%.Additionally, contamination of the final product was estimated by using a 2 x 2 matrix following the model proposed by the World Health Organization (WHO).Severity was determined based on the possibility of product contamination with L. monocytogenes, by using the following scale.
a) Low: Contaminated surface was not in direct contact with the finished product.b) Medium: Contaminated equipment or raw meat used before thermal processing.c) High: Equipment, surfaces or utensils were in direct contact with the finished product after subjected to thermal processing.

RESULTS AND DISCUSSION
For the five plants analyzed our results pointed out the following risks: use of wooden baskets (plant 2), use of equipment with additions made out of iron (plant 2), weekly monitoring instead of a daily inspection of residual chlorine in the water (plant 3), raw meat storage temperatures higher than the country's mandatory temperature (plant 3), and finished product storage temperature higher than 4°C (plants 1 and 5).
The most important extrinsic parameter to control L.
monocytogenes is temperature, considering its ability to grow in refrigeration temperatures.This aspect is critical for meat products.Even at low initial contamination concentrations, the organism can multiply during storage in the processing plant or the market, reaching levels above the minimum infective dose (1000 CFU/g).It has been shown that these levels can be achieved during the lifetime of the product, when stored at temperatures ranging between 4 -9°C.From our results we observed that plants 1 and 5 maintain products at storage temperatures within this range, and are therefore at risk (Sergelidis et al., 1997).
A total of 36 isolates of L. monocytogenes were molecularly identified and serotyped as follows (Figure 2): plant 1: 13/53 samples were positive (24.5%), plant 2: 10/71 samples were positive (14%); plant 3: 4/61 samples were positive (6.5%), plant 4: 9/89 samples were positive (10%) and plant 5: 0/40 were positive (0%), demonstrating that the presence of this organism due to external factors, for an overall prevalence of 11.46%.Our results agree with studies by Gudbjörnsdóttir et al., (2004) who reported a prevalence ranging between 0 and 15.1% and are inferior to those obtained by Chasseignaux et al., (2002) who reported 23.7%.These differences may be related to the technology level and equipment used.Isolates were recovered from: work surfaces (33%), followed by raw meat (29%) and knives (16.7%).As with other reports the most frequent site of contamination were floors (Thévenot et al., 2006).Noncontact surfaces represented 14% and 10% in processing equipment, demonstrating that any environment within the meat industry may be contaminated with this bacterium.
For many years pork-meat derived products have been associated with foodborne outbreaks around the world (Goulet et al., 1998;CFIA, 2008).When analyzing the causes of outbreaks, it has been determined that contamination in meat processing plants occurs at the post-processing stage or by the presence of contaminated raw meat entering the plants.The latter one are contaminated during the deboning process, because the prevalence of this organism during the benefit has been low 3.7 to 4%, (Gamboa-Marín et al., 2012), 4% (Kanuganti et al., 2002).Several studies indicate that contamination of raw meat occurs mainly in deboning-rooms (Van den Elzen and Snijders, 1993;Nesbakken et al., 1996).
In the present study we found L. monocytogenes in four out of five plants, which confirms the ability of this microorganism to enter into meat processing plants, despite established programs for cleaning and disinfection.The presence of L. monocytogenes showed a wide spread in meat processing plants, which is consistent with studies by various authors (Chasseignaux et al., 2002;Thévenot et al., 2006).L. monocytogenes can survive industrial environments because it can grow isolates of serotype 4 that were not amplified with MAMA-C, therefore classified as serotypes 4b/4d/4e, Lane27: the same DNA as in lane 20, Lane 28: the same DNA as in lane 14. (a 6 ) amplification with CLM1/CLM2, Lane 29: L. monocytogenes ATCC 19115, Lanes 30 and 31: serotypes 4b/4d/4e isolates).(a 7 ) Restriction Enzyme Analysis of amplification products obtained with CLM1/CLM2, Lane 32: same DNA as in lane 30, Lane 33 same DNA as in lane 31, Lane 34: isolate of serotype 1/2a, 3a showing the expected restriction pattern with HindIII, therefore serotype 3a, Lane 35: isolate of serotype 4b/4d/4e, showing the expected restriction pattern with HindIII, therefore serotype 4b.B: Distribution of serovars of L. monocytogenes in meat processing plants.at low temperatures, adheres easily to and some strains are tolerant to disinfectants (Ruiz-Bolivar et al., 2011).According to Kathariou (2002) the main source of food contamination by L. monocytogenes before arriving to the final consumer seems to be the processing environment.The RTE food can become contaminated by contact with raw meat or ingredients, equipment and utensils, or post-processing contamination.This issue was confirmed by our results as we demonstrated L. monocytogenes on meat choppers.Gandhi and Chikindas (2007), emphasize that food contamination can occur at any stage of processing, and the most important source of re-contamination are raw products.In the present study two plants were found to be contaminated with raw meat.Other authors indicate that L. monocytogenes can survive in different types of surfaces, including stainless-steel (material commonly used in the meat industry for equipment, tools and work tables).Previous studies have demonstrated L. monocytogenes' ability to persist and form biofilms (Mattos de Oliveira et al., 2010).All these risk factors favor the cross-contamination between the environment and all processing equipment (Jemmi et al., 2002).
Figure 2A shows the identification and molecular characterization of isolates and Figure 2B depicts the serovars isolated.Serovar 4d/4e was detected in all plants positive for L. monocytogenes.Serovar 4b was isolated from plants 1 and 2; in this latter one it was isolated from the chopper.Serovar 3a was recovered in samples from three different plants.Some of the isolates only amplified bands for genus (938 bp) and species (750 pb), but no amplification products were generated with primers used for serotyping.Thus they were grouped as members of serotypes 4ab or 7, these results remain to be confirmed experimentally (Figure 2B).
Although Orsi et al., (2011), indicated that serotypes 1/2 a, 1/2 c, 3a and 3c are often found in food processing plants because of their greater ability to adapt to the environment, this study also identified isolates of serotype 4. We assert the finding of serovar 4b, as it has been associated to primary causes of foodborne outbreaks.
Regarding disinfectants evaluation, no isolates were tolerant to per-acetic acid at the concentrations studied.Nine isolates (25%) from: floors (3), walls (3) and sewage (3) displayed tolerance to ACDA at concentrations of 0.5 and 2.0% (v/v).When analyzing the isolates' tolerance to disinfectants, it was found that those isolated from sewage, floors and walls showed certain tolerance degree to ACDA, which suggests that contact with organic matter and high humidity, promotes the formation of a biofilm, increasing the tolerance to disinfectants (Denyerae and Stewart, 1998).Additionally, L. monocytogenes' presence in sewage received sublethal concentrations of disinfectant, allowing it to gradually adapt to these substances, favoring the persistence of some strains (Mattos de Oliveira et al., 2010).
Our results are consistent with other studies that demonstrated an increase in the tolerance of L. monocytogenes to quaternary ammonium (Mereghetti et al., 2000), traditionally used disinfectants in the food industry.This data suggest the possible presence of efflux pumps as an adaptation mechanism in sensitive strains (To et al., 2002).
Risk factors classification for each meat processing plant is described in Table 2.As depicted in this table risk varied among plants.None the less, some surfaces and equipment had increased contamination, with chopper and cutter at the highest risk.Furthermore, we found this organism in raw meat.
Traditionally studies to assess the risk factors of L. monocytogenes in the industry are established through the sampling of different sites where the organism may be present and correlated with environmental factors (Giovannacci et al., 1999;Chasseignaux et al., 2002).However, this appraisal cannot determine the impact that contamination could have on the final product.In this study we propose to qualify the risk factors.This model proposes to estimate the probability level of contaminated foods leaving the factory.
This tool allowed to demonstrate that the contamination vary depending on the processing plant, agreeing with several authors (Chasseignaux et al., 2002;Thévenot et al., 2006).Moreover, this tool allowed establishing that the number of contaminated sites along the production process is proportional to the risk of final product contamination.It was also demonstrated that the critical risk on plant 3 was the chopper.If this equipment is contaminated with L. monocytogenes, it increases the possibility of releasing contaminated product to the customer.Additionally it demonstrated the low risk that sewage represents to the final product.However, it remains an issue to be considered.Last, this model allowed determining which and how areas within the plant should be treated to reduce contamination of the final product.

Conclusions
Finally, one of the most important practices for the control of L. monocytogenes in processing plants should be to implement a cleaning and disinfection program under strict monitoring.As part of this program it should be included cleaning of sewage.It is decisive during this cleaning to avoid the formation of aerosols and the use of high pressure systems.
The general prevalence of L. monocytogenes in pork meat processing plants was 11.46%.The main risk factors were: raw meat, cutter, mixer, sewage and choppers.The best strategy for reducing L. monocytogenes in meat processing plants is the implementation and monitoring of cleaning and disinfection programs.

Figure 1 .
Figure 1.Map of Colombia with the two main areas of pork meat derivate product production.