Abstract
Listeria monocytogenes is often present in meat and meat products that are sold in the area of northeast Bosnia and Herzegovina. The major objective of this study was to examine the virulence of L. monocytogenes strains isolated from these types of food in that geographic area. Polymerase chain reaction was used to detect eight genes responsible for virulence of this pathogen, namely, prfA, inlA, inlB, hly, plcA, plcB, actA, and mpl. All examined isolates were confirmed to possess the eight virulence genes. Ten different pulsed-field gel electrophoresis (PFGE) macrorestriction profiles were recognized among 19 L. monocytogenes strains after restriction with two different endonucleases (ApaI and AscI). The pathogenicity of three different PFGE types of L. monocytogenes was confirmed through in vivo tests, which were performed on female white mice (Pasteur strain), and it ranged from 3.55×108 LD50 to 1.58×1010 LD50. All of the three different PFGE types of L. monocytogenes were regarded as moderately virulent in relation to the reference strain L. monocytogenes Scott A. This result might be one of the reasons for the absence of reported listeriosis in northeast Bosnia and Herzegovina, despite the high degree of food contamination with this pathogen.
Introduction
The primary site of entry of L. monocytogenes is the digestive tract: the pathogen penetrates very fast into the intestinal epithelium, where it is phagocytosed by macrophages (in humans, this phase is asymptomatic). Intracellular proliferation occurs, with destruction of macrophages and a consequential septicemia (Cousens and Wing 2000). Certain segments of human population such as the elderly, infants, pregnant women, people with reduced immunity, and individuals undergoing immunosuppressive therapy have an increased risk of infection.
Ingestion of L. monocytogenes is very common considering the omnipresent distribution of these bacteria and the high incidence of contamination of raw and industrially processed food. Yet, the incidence of human listeriosis is very low, usually about 2–8 sporadic cases per annum on a million of inhabitants in Europe and the United States of America (Recourt et al. 2000). Despite the high percentage, 23.3% (42/180) of raw meat samples (Hodžić and i Hukić 2004) and 11.1% (10/90) of meat product samples contaminated by L. monocytogenes in the area of Bosnia and Herzegovina (Hodžić and i Hukić 2006), no cases of human listeriosis were reported in the past 10 years in the same area (Department of Health Statistics, 2009).
Even though occurrence of listeriosis may be underestimated, another plausible reason for such conflicting proofs may be the variability of virulence of L. monocytogenes. In fact, although L. monocytogenes is defined as a pathogen at the level of species, within this species a high degree of genetic diversity and an extremely variable pathogenic potential are present. Main genes responsible for virulence of this pathogen are located on the bacterial chromosome (Vazquez-Boland et al. 2001). Molecular analyses of known virulence genes may be useful (Nexmann-Larsen et al. 2002), although they cannot replace experiments on animals for the determination of the pathogenic potential of L. monocytogenes.
The major objectives of this article were to examine the genetic diversity and pathogenic potential of L. monocytogenes isolates from meat and meat products in northeast Bosnia and Herzegovina.
Materials and Methods
Bacterial isolates
The research included 20 isolates of L. monocytogenes (Table 1). L. monocytogenes Scott A was used as a control virulent strain, which is associated with outburst of invasive listeriosis (Fleming et al. 1985). The L. monocytogenes control avirulent strains used in this research were strain American Type Culture Collection (ATCC) 43248, isolated from guinea pig; strain ATCC 15313, isolated from rabbit; and strain L.285, isolated at the Institute of Health in Rome from a meat sample imported from Brazil. Listeria innocua isolated from beef collected from retail in Tuzla Canton was also included in the study.
GGP, beef product; GSM, raw beef; PFGE, pulsed-field gel electrophoresis; PSM, raw chicken; SGP, pork product; SSM, raw pork.
Isolation and identification of the Listeria species was performed by applying the method ISO 11290–1/204 (Anonymus 2004), and identification was confirmed by a biochemical test 10300 API Listeria (BioMerieux Italia, Florence, Italy). The serotyping was performed using a commercial kit (Listeria Antiserum kit; Denka Seiken, Tokyo, Japan).
PCR detection of virulence genes
Preparation of bacterial cells and isolation of DNA were performed as previously described (Franciosa et al. 1998). Polymerase chain reaction (PCR) experiments to detect eight L. monocytogenes virulence genes (prfA, inlA, inlB, hly, plcA, plcB, actA, and mpl) were performed according to the method described by Franciosa et al. (2005).
Genotyping
L. monocytogenes strain genotyping was performed by application of pulsed-field gel electrophoresis (PFGE), according to the PulseNet standardized protocol (Graves and Swaminathan 2001), by using CHEF-DRII apparatus (Bio-Rad Laboratories, Hercules, CA). The restrictive endonuceases ApaI (Roche Diagnostics, Penzberg, Germany) and AscI (New England BioLabs, Beverly, MA) were used.
Similarity of the macrorestriction profiles was calculated using the program Molecular Analyst Software Fingerprinting (Bio Rad Laboratories, Hercules, CA). Dendrogram was formed by applying a correlation coefficient dice and unweighted pair group method by means of arithmetic mean according to the same program. A degree of homology ≥80% (2% tolerance of position) between the genetic profiles of L. monocytogenes isolates was selected for defining one cluster.
Biologic experiments on the mouse model
The pathogenicity of selected PFGE types of L. monocytogenes was performed on the model of immunocompetent females of white mice (Pasteur strain [Department of Pharmacology and Toxicology at the Veterinary Faculty in Sarajevo], 6–8 weeks old, weighing about 20 g in 5 days). The pathogenicity of selected PFGE types of L. monocytogenes was examined on 20 mice as follows: Inocula of 106, 107, 108, and 109 cells/mL were prepared in physiological solution for each selected L. monocytogenes PFGE type, according to Takeuchi et al. (2003). Groups of five mice were intraperitoneally injected, each with 0.1 mL of the different inocula.
L. monocytogenes Scott A and L. innocua were used as positive and negative controls of pathogenicity, respectively, by the protocol described earlier. As a further negative control, a group of five mice were also used, which was intraperitoneally inoculated with 0.1 mL of physiological solution.
All animal procedures were performed in accordance with the European Directive 86/609/EEC on protection of animals used for experimental and other scientific purposes.
Isolates of L. monocytogenes that caused at least one death in the course of 5 days were considered to be pathogenic. The isolates that did not cause a single death in the course of 5 days were considered nonpathogenic (Takeuchi et al. 2003).
The pathogenic potential of the L. monocytogenes isolates was determined by calculation of lethal dosage for 50% of the inoculated animals (LD50) through graphic linear interpolation (Reed and Muench 1938) and by determining the relative virulence (%) in relation to the reference strain of L. monocytogenes Scott A, according to Dongyou (Dongyou 2004).
Results
Serotyping
Serologic testing was performed in 20 isolates of L. monocytogenes (Table 2). Four serotypes were determined: 1/2a in 50% (10/20) of the isolates, 1/2b in 5% (1/20) of the strains, 1/2c in 30% (6/20) of the isolates, and 4b in 15% (3/20) of the isolates (Table 1).
CFU, colony forming unit.
Virulence genes of L. monocytogenes
PCR results indicated that all examined L. monocytogenes isolates possessed the eight virulence-related genes (prfA, inlA, inlB, hly, plcA, plcB, actA, and mpl) considered in this study.
PFGE types of L. monocytogenes
The analysis of the PFGE macrorestriction profiles obtained with two different endonucleases, AscI and ApaI, distinguished 10 different PFGE types among the tested 19 L. monocytogenes isolates (Table 1). Each PFGE type represents a separate clone of L. monocytogenes. L. monocytogenes serotype 1/2a displayed the broadest genomic variation: in fact, 10 isolates of serotype 1/2a were grouped into six different PFGE types. Five L. monocytogenes isolates of serotype 1/2c were grouped in two different PFGE types. Within L. monocytogenes isolates of the serotype 4b, two different clones were identified. The only L. monocytogenes strain of serotype 1/2b produced a single PFGE type.
Virulence of L. monocytogenes
Three different PFGE L. monocytogenes types, each representative of serotypes 1/2a, 1/2b, and 4b, were selected for the virulence assay. Results are summarized in Table 2.
All three L. monocytogenes isolates (70, 206, and 59) caused the death of some inoculated mice within 5 days of the experiment and were considered to be pathogenic. Death of mice was caused by microbial inocula of 109 and 108 cells/mL, respectively, whereas the microbial inocula of 106 and 107 did not kill any mice.
Five days after inoculating L. monocytogenes isolate 70 (PFGE type 2, serotype 4b), a total of five animals died, of which four animals (4/5) were infected by inoculum of 1.5×109 cells/mL, and one animal (1/5) received a dose of 1.5×108 cells/mL.
Five days upon inoculation of L. monocytogenes isolate 59 (PFGE type 10, serotype 1/2a), a total of five animals died, of which one animal (1/5) was infected by an inoculum of 2×109 cells/mL and four animals (4/5) received a dosage of 2×108 cells/mL.
Five days upon inoculation of L. monocytogenes isolate 206 (PFGE type 4, serotype 1/2b), a total of three animals died, of which two animals (2/5) were infected by an inoculum of 5×109 cells/mL, and one animal (1/5) that received a dose of 5×108 cells/mL.
Five days upon inoculation of L. monocytogenes Scott A (positive control), a total of 10 animals died, of which five animals (5/5) were infected by an inoculum of 3.5×109 cells/mL and five animals (5/5) received a dose of 3.5×108 cells/mL.
All animals inoculated with L. innocua (negative control) were alive after 5 days.
The five animals inoculated with the physiological solution were also alive after 5 days.
Relative virulence was calculated in relation to the reference strain L. monocytogenes Scott A, which caused a 100% death rate in two dosing groups: 109 and 108 cells/mL. LD50 calculated for strain L. monocytogenes Scott A was 1.11×108 cells/mL. LD50 for L. monocytogenes isolate 70 (PFGE type 2) was 4.74×108 cells/mL, and the derived relative virulence was 50%. LD50 for L. monocytogenes isolate 59 (PFGE type 10) was 3.55×108 cells/mL and relative virulence is 50%. LD50 for L. monocytogenes isolate 206 (PFGE type 4) was 1.58×1010 cells/mL and relative virulence is 30%.
Discussion
The L. monocytogenes species include strains with variable pathogenic potential. Although many strains of L. monocytogenes are very pathogenic and sometimes lethal to humans and animals, others are relatively avirulent and do not cause a great damage in the host. An obvious association has been noticed between the antigenic properties and pathogenicity of L. monocytogenes. The proof lies in the fact that of the 13 known serotypes of L. monocytogenes, only 3 (1/2a, 1/2b, and 4b) causeed more than 90% of human and animal cases of listeriosis (Low et al. 1993). Our research, which was based on L. monocytogenes isolates from foods, showed that most of them (14 of 20) belonged to serotypes 1/2a, 1/2b, and 4b, thus representing a potential health risk.
Many typing methods, including the very discriminatory genotyping method of PFGE, have confirmed the presence of genetic clusters in L. monocytogenes. Results of our PFGE analyses showed some genetic diversity among the different serotypes of the L. monocytogenes tested, in accordance with other researchers' observations (Bibb et al. 1998). In addition, consistent with data from the literature (Bibb et al. 1998), certain serotypes were genetically more diversified than others, with serotype 1/2a being the most diversified.
Virulence of L. monocytogenes strains depends on expression of several genes responsible for the pathogen capability to penetrate the host cells and to proliferate and expand. The entry process includes the expression of two proteins, internalin A and B, coded by two genes inlA and inlB. The other gene locus includes genes involved in the functions essential for intracellular survival and proliferation of the bacterium: these genes code listeriosyn O, phospholipase C, and actin. All of these cell surface proteins are regulated coordinately by means of a pleiotrophic transcriptional activator coded by prfA.
In this study, the above mentioned virulence genes were detected in all L. monocytogenes isolates analyzed, conforming with previous results (Franciosa et al. 2005). Moreover, all of the isolates were equally pathogenic after intraperitoneal inoculation in mice.
However, the control avirulent strain ATCC 43248, isolated from a guinea pig, possessed the virulence genes inlA, inlB, hly, plcA, plcB, actA, and mpl, but it lacked the prfA gene. Despite the presence of many genes associated with virulence, as confirmed by PCR, defective regulation by prfA results in avirulence of strain ATCC 43248 for mice. Control avirulent strain L 285, isolated from a meat sample originating from Brazil, is avirulent in mice, although PCR analysis confirms the presence of all examined genes (Franciosa et al. 2005). The control avirulent strain originating from the infected rabbit, ATCC 15313, was initially hemolytic but has become nonhemolytic and avirulent after the next laboratory screening (Kathariou and Pine 1991). Further PCR analysis has shown that this strain had lost genes responsible for virulence (Franciosa et al. 2005).
Analysis of virulence in mice provides a reference standard for in vivo determation of the pathogenic potential of L. monocytogenes. By applying criteria for calculation of the relative virulence described by Liu (2004), the L. monocytogenes isolates tested in this study may be classified as moderately virulent, which may be one of the reason for the absence of listeriosis in Bosnia and Herzegovina.
The results of this study indicate that Listeria is an important emerging pathogen in northeast Bosnia and Herzegovina, thereby presenting potential threat for public health and also for the neighboring countries because of livestock market. Serious epidemiology studies among humans are necessary to give insight on the significance of listeriosis for the population in Bosnia and Herzegovina. However, less-virulent isolates in the northeast part of the country may suggest that less-severe cases may predominate, with certain percentage of unrecognized cases, which are consequently underreported.
Footnotes
Acknowledgment
The authors express gratitude to the Veterinary Faculty of the University of Sarajevo, Department of Pharmacology, for enabling this biological experiment.
Disclosure Statement
No competing financial interests exist.