Listeria monocytogenes is a foodborne pathogen that has wide distribution in the environment and high prevalence in a broad range of foodstuff. Infection with L. monocytogenes can cause from mild, febrile illness to systemic illness in susceptible populations, accompanied by more severe symptoms such as meningitis, meningoencephalitis, septicemia, and abortion as well as high hospitalization and case fatality rates. The cause of illness has been traditionally associated with the intake of contaminated ready-to-eat food, although recent outbreaks have recognized many other transmission vehicles such as crustaceans, shellfish, mollusks and products thereof, dairy products, meat and meat products, vegetables and products thereof, fruits and juices ².

In Argentina, no outbreaks of L. monocytogenes associated with contaminated food have been reported so far. A total of 310 cases of listeriosis have been registered by the Argentinean National Institute of Infectious Diseases of the National Administration of Laboratories and Institutes of Health "Dr. Carlos Malbran" (INEI-ANLIS, for its Spanish acronym) from 1985 to the present, all of them corresponding to invasive disease, mainly sepsis and meningitis. These data suggest that the disease is underreported and its prevalence and real impact on public health is unknown ¹³.

In the global context, a massive outbreak of invasive L. monocytogenes linked to frozen vegetables has been affecting Austria, Denmark, Finland, Sweden and the United Kingdom since 2015, with 47 reported cases including nine deaths as of June 15, 2018 ⁶. In Argentina, as a result of the warning issued by the European Rapid Alert System for Food and Feed, the Argentinean National Administration of Medicines, Food and Medical Technology (ANMAT, for its Spanish acronym) ordered the preventive withdrawal of 128 batches of frozen products based on corn and other vegetables made with raw material presumably contaminated with L. monocytogenes¹.

This situation warns about the importance of reinforcing the isolation, characterization and subtyping of L. monocytogenes from food, clinical and environmental samples. In this sense, molecular typing is fundamental for the epidemiological tracing of isolates, the cross-linked surveillance of isolates in human beings and food, the identification of sources of infections and the performance of outbreak controls.

Pulsed-field gel electrophoresis (PFGE) is considered the gold standard typing method in epidemiological studies of pathogenic organisms and, as such, it is used by the INEI-ANLIS. Ribotyping is a completely automated method based on restriction fragment length polymorphisms associated with ribosomal operons. Enterobacterial repetitive intergenic consensus polymerase chain reaction (ERIC-PCR) is an alternative typing method based on 124-127 base-long elements consisting of highly conserved central inverted repeat sequences found in the extragenic regions of the bacterial genome.

In this study, subtyping of L. monocytogenes by ERIC-PCR, automated ribotyping and PFGE was evaluated in 27 strains isolated from Argentinean meat and environmental samples belonging to the collection of the Institute of Veterinary Genetics, School of Veterinary Sciences, National University of La Plata.

For ERIC-PCR, a single colony of pure L. monocytogenes was inoculated into 5 ml Brain Heart Broth (Biokar Diagnostics, France) and cultured at 37 °C for 24 h. Genomic DNA was extracted using the Wizard Genomic DNA Purification kit (Promega, WI, USA). Primers ERIC 1R (5 -ATGTAAGCTCCTGGGGATTCAC-3) and ERIC 2 (5-AAGTAAGTGACTGGGGTGAGCG-3) ¹⁴ were used. Amplification reactions were performed in 25 μl of a solution containing 1 x Taq buffer with (NH ₄) ₂SO ₄ (Thermo Scientific, MA, USA), 1 ^M of each primer (Embiotec SRL, Argentina), 0.4mM of each deoxynucleoside triphosphate (PB-L, Argentina), 6mMMgCl ₂ (Thermo Scientific, MA, USA), 1.25 U DNA polymerase (Thermo Scientific, MA, USA) and 2 μl DNA extract. Amplifications were performed in a DNA thermal cycler (Life Express, Bioer, China) with the following temperature profiles: 1 cycle at 95 °C for 5 min; 35 cycles at 94°C for 30s, at 40°C for 3 min, and at 72 °C for 2 min; and 1 cycle at 72 °C for 7 min. Finally, PCR products were separated by electrophoresis in 2% agarose (Genebiotech, South Korea) in 1x TBE running buffer at 80V for 3h. Molecular weight marker 1 Kb Plus DNA Ladder (Thermo Scientific, MA, USA) was used as size standard.

EcoRI-Ribotyping was performed on L. monocytogenes isolates grown on Nutritive Agar (Biokar Diagnostics, France) for 24 h at 37 °C in an automated ribotyping system (RiboPrinter®, Qualicon DuPont™) following the manufacturer's instructions.

For PFGE analysis, the one-day standardized laboratory protocol for molecular subtyping of L. monocytogenes was employed ⁴. Restriction digestion of DNA in agarose plugs was carried out with Apal or AscI (Thermo Scientific, MA, USA) during 18 h. MaestroGen slider imager (Maestrogen Inc., Nevada, USA) was used to obtain PFGE images of gels.

Tagged file format (TIFF) images obtained by ERIC-PCR, ribotyping and Apal- and Ascl-PFGE were analyzed with BioNumerics version 6.6 software package (Applied Maths, Sint-Martens-Latem, Belgium) using the Dice coefficient and the unweighted pair-group method with arithmetic mean (UPGMA) to generate dendrograms with 1.5% band matching tolerance. Two or more isolates with identical band pattern (100% similarity) were grouped into a cluster.

The three typing methods were compared using the Simpson's Diversity Index (Dl) and the corresponding confidence intervals (Cls) ¹¹. Statistical differences among Dl values were analyzed by calculating p <-values according to the jackknife pseudo-values resampling method. The Comparing Partitions tool was used to calculate Dl and for the statistical analysis (http://www.comparingpartitions.info/?link=Tool).

All L. monocytogenes strains analyzed (n = 27) were typeable by ERIC-PCR, ribotyping and Ascl/Apal-PFGE ( Fig. 1). Cluster analysis results are shown in Table 1. The number of bands was greater and the band pattern was more defined in the gels obtained by AscI- and ApaI-PFGE than in those obtained by ERIC-PCR and ribotyping. Accordingly, DI showed the following discriminatory power: ApaI-PFGE, AscI-PFGE, ribotyping and ERIC-PCR. DI values were neither significantly different between ApaI- and AscI-PFGE nor between ribotyping and ERIC-PCR ( Table 1).

Figure 1 Dendrogram representing the genetic relationships among 27 L. monocytogenes strains based on ApaI-PFGE profiles. Gel images of ERIC-PCR, EcoRI-ribotypes and AscI-PFGE patterns were included to allow direct visual comparison. 

Table 1 Results of cluster analysis for the different typing methods analyzed 

The number of unique profiles obtained with ApaI was greater than with AscI-PFGE, and only three clusters resulted from the cluster analysis. Strains with equal ApaI profile also presented the same profile with AscI, with the exception of strains 732, 733 and 734. Therefore, a greater discriminatory power was obtained by combining the results of AscI- and ApaI-PFGE. Clusters formed by strains 721, 722 and 724 and strains 813 and 815 were found to be clonal by the combined ApaI and AscI analysis. However, different profiles were obtained for these strains by ribotyping and ERIC-PCR, which could be explained by the presence of faint and diffuse bands in ERIC-PCR and ribotyping profiles, which interfered with complicated band assignment and result interpretation. ERIC-PCR patterns were the most difficult to interpret, as already reported by Harvey ⁹.

In PFGE typing, DIs of AscI- and ApaI-PFGE (0.966 and 0.980, respectively) were similar to those described by Henri et al. ¹⁰, who reported a DI of 0.992 in 687 strains isolated from meat and meat products and 0.987 in 209 strains isolated from the food processing environment.

Our results agree with those reported by Fugett et al. ⁷, who found that the discriminatory power determined by the Simpson's index did not differ between ApaI and AscI (DI 0.992 each) in 495 L. monocytogenes isolates, while the combined analysis of both enzymes was more discriminatory than EcoRI ribotyping (0.995 vs. 0.950, respectively). Grif et al. ⁸ also found a higher discriminatory power of PFGE than of automated ribotyping. However, some PFGE types were differentiated by ribotyping, suggesting that the combination of data obtained by both methods increased the likelihood of strain discrimination. Other authors found that DIs of automated ribotyping and ApaI-PFGE did not differ significantly ^{5 12}.

Since L. monocytogenes serotype 4b is responsible for most hospitalization rates and deaths ³, its sensitive discrimination is particularly crucial for surveillance and monitoring of human listeriosis cases. Previous studies showed the highly clonal nature of L. monocytogenes 4b strains ⁸. In the present study, the four serotype 4b isolates exhibited unique Apal-PFGE patterns, although they presented similar profiles clustering in nearby branches by the other typing methods.

At present, the availability of molecular techniques for L. monocytogenes subtyping is essential to reinforce the development of epidemiological outbreak studies as well as to identify and trace strains in food, human and environmental samples isolated from different countries.

In this study, ERIC-PCR showed the lowest Dl, suggesting the need to optimize the method to obtain accurate results. However, since it is a simple and economic technique that can be carried out in low-complexity laboratories, ERIC-PCR could be used as a first subtyping screening. On the other hand, ribotyping had higher DI than ERIC-PCR; despite being automatic, rapid and less laborious than PFGE, ribotyping is expensive and requires specific equipment. Finally, the high discriminatory power obtained by two-enzyme PFGE makes it a good technique for epidemiological outbreak studies. This method also produced stable and reproducible band patterns and showed high concordance with epidemiological data ⁴. However, it is time-consuming, expensive, laborious and requires specific equipment ⁴. In Argentina, its implementation is restricted to high-complexity research laboratories or centers. While new alternatives for food-borne disease surveillance such as whole genome sequencing (WGS) are gaining importance, the epidemiological surveillance of listeriosis should be reinforced in Argentina, since subtyping techniques alone are not sufficient and should be accompanied by effective food monitoring strategies and reliable clinical and epidemiological studies.