INTRODUCTION

Among the bacteria that cause foodborne diseases, L. monocytogenes, according to the WHO (dated April 30, 2020), belongs to the third risk group, second only to Salmonella, Campylobacter, and enterohemorrhagic E. coli (EHEC) (https://www.who.int/NEWS-ROOM/FACT-SHEETS/DETAIL/FOOD-SAFETY).

Microbiological control of food products in our country is organized from the stage of production and transportation to storage in the distribution network and is carried out in accordance with the regularly updated Technical Regulations of the Customs Union TR CU 021/2011 “On Food Safety” (as amended on August 8, 2019) and SanPiN 3.3686-21 “Sanitary and epidemiological requirements for the prevention of infectious diseases.” Meat and meat products are controlled by a separate document: TR CU 034/2013 Technical Regulations of the Customs Union “On the safety of meat and meat products.” Isolation and identification of Listeria is determined by GOST (State Standard) 32031–2012 (with Amendment IUS N 6-2019, http://docs.cntd.ru/document/1200105310), which is an interstate standard for the countries of the Eurasian Union.

It should be emphasized that the control of the production environment is a key point in the technological chain, since the persistence of Listeria in production increases the likelihood of cross-contamination of finished products.

For EU countries, the control of the working environment in relation to L. monocytogenes is a subject of special attention. Two other species of the Listeria sensu strictu cluster [1], L. innocua and L. welshimeri, are also subject to control as L. monocytogenes contamination/colonization indicators of the production environment [2]. Bacteria of genotypes detected several times over 6 months or more are defined as persistent contaminants, and genotypes detected once for 2 or more years are defined as nonpersistent ones [2]. Molecular methods to determine a genotype were introduced by the ECDC (European Center for Disease Prevention and Control) in the surveillance of foodborne pathogens in the EU/EEA (European Union/European Economic Area) in 2012. In March, 2019 The EC-DC introduced whole-genome sequencing (WGS) for surveillance of invasive listeriosis in the EU/EEA. In 2021, in the midst of the COVID-19 pandemic, the ECDC conducted the seventh round of external quality control of national reference laboratories for Listeria serology and molecular testing (https://www.ecdc.europa.eu/en/publications-data/listeria-monocytogenes-typ-ing-seventh-external-quality-assessment-scheme/).

In the United States, the first preventive food-control regulations, including monitoring of L. monocytogenes in food-processing plants, were adopted in 2011 (https://www.fda.gov/regulatory-information/search-fda-guid-ancedocuments/draft-guidance-industry-control-listeria-monocytogenes-ready-eat-foods). In 2019, the Food and Drug Administration (FDA) published food-safety regulations that define the control strategy from animal rearing to food production, including control of equipment, tools, and facilities (https://www.fda.gov/food/fbod-safetymodernization-act-fsma/fsma-final-rule-produce-safety). The presence of the PulseNet system, which has a database of genomic data of bacteria that have caused foodborne diseases, as well as a network of laboratories that perform WGS, allows one to quickly conduct studies of the genetic relationship of Listeria isolated from infected people and from products. In 2021, the effectiveness of this approach was demonstrated by an investigation of a listeriosis outbreak in 4 states caused by the use of fresh and soft cheeses of the El Abuelito brand (https://www.cdc.gbv/listeria/butbreaks/hispanic-soft-cheese-02-21/index.html).

Whole-genome sequencing of Listeria has become one of the tools in a polycentric study, which included the Gamaleya Research Institute of Epidemiology and Microbiology at the end of 2018 [3]. WGS supplemented the methods of MultiLocus Sequence Typing (MLST) and Multivirulentlocus Sequence Typing (MvLST), which determine the genotype (Sequence Type, ST) of Listeria and the profile of internalins (Internaline Profile, IP), proteins that ensure Listeria invasion into eukaryotic cells. Molecular genetic data made it possible to characterize the clinical and food isolates of L. monocytogenes in Moscow region in the period before COVID-19 and during the pandemic [4]. We applied the accumulated experience in a pilot study of isolates from two meat processing plants in Moscow. The objective of this work is to determine the genetic characteristics of Listeria isolated from swabs performed at different sites of the production, as a basis for further monitoring of Listeria of persistent genotypes.

MATERIALS AND METHODS

Organization of Sampling at Meat-Processing Plants

Two meat-processing plants in Moscow were examined in 2019–2020. Both enterprises used pork and poultry (produced in Russia), as well as beef (produced in Russia and Belarus), as raw materials. The objects of control of the production environment were selected in accordance with the principles of zoning. The characteristics of the zones and the identified isolates are presented in Table 1. Sterile sponges impregnated with Letheen broth (3M Hydrated-Sponge with 10 mL Letheen Broth, 3M Science. Applied to Life, United States) were used for washings. Components of the broth (lecithin and polysorbate 80) neutralize quaternary ammonium compounds and hexachromophene, which is used to disinfect the working environment, as well as phenolic disinfectants and formalin, which are used for disinfection in medical institutions. At Plant 1, we collected 41 swabs; at Plant 2, 38 swabs.

Table 1. Characteristics of Listeria isolates obtained from swabs at meat-processing plants. Distribution by sampling area
Full size table

Examination of Swabs for the Presence of Listeria spp. Enrichment

Sponges used for washings were placed in 225 mL LPT broth (bioMérieux, France) intended for selective enrichment of Listeria spp., mixed using an AES MIX2 Lab Blender (American Laboratory Trading, United States) and incubated in a Binder BD 115 Avantgarde (BINDER GmbH, Germany) at 37 ± 1°C for 18–24 h.

Real-time PCR (RT-PCR) was performed using reagents and a GENE-UP device (bioMérieux, France). Twenty microliters of the enriched culture was lysed using the GENE-UP Lysis Kit. Ten microliters of the sample were used for testing with reagents and according to the protocol kit for GENE-UP L. monocytogenes 2 (LMO 2).

Positive samples were confirmed according to GOST (State Standard) 32031–2012.

Molecular and Genetic Characteristics of Listeria Isolates

Multilocus sequencing, which included analysis of seven housekeeping genes and four virulence genes (MLST, MultiLocus Sequence Typing, and MvLST, Multi-virulent-Locus Sequence Typing), was performed as described previously [5]. Analysis of MLST alleles and allelic profiles (ST, Sequence Type) was performed using the resources of Bacterial Isolate Genome Sequence Database for L. monocytogenes (BIGSdb-Lm) (https://bigsdb.pasteur.fr/listeria/) [6]. The analyzed isolates and new allelic profiles were deposited in the site’s database, ID 49188-49201, 49206.

Amplification and sequencing of the abcZ locus (ABC transporter) for L. welshimeri was carried out using the abcZ F216 (TTG-GATGCTGATTCTCTGCT) and abcZ R897 (YGAATATT-GAACGAACATAACG) primers that we had developed, since the primers proposed by M. Ragon et al. [7] for L. monocytogenes were not complementary to the sequences of the abcZL gene of L. welshimeri.

MvLST and IP alleles (Internalin genes Profile (in-lA, inlB, inlC, inlE)) were determined using published sequences as references [3, 5, 8–11]. New variants of the inlA and inlE alleles identified in this study were registered in GenBank (Accession Numbers: MT812686, MT812687).

Whole-genome sequencing of two L. welshimeri isolates, GIMC2049:LmcC11 and GIMC2051: LmcC15, obtained at Plant 2 in 2020 was performed on the MiSeq Illumina platform (MiSeq v2 Reagent Kit 300 cycles cartridge, Illumina, United States). Fragment libraries were prepared using the KAPA HyperPlus kit (Roche, Switzerland). The quality and size of the libraries were assessed by electrophoresis on High Sensitivity DNA Chips on a 2100 Bioanalyzer System (Agilent, United States). Sequencing results (Sequence Read Archive, SRA) were deposited with the GenBank NCBI (BioProject ID: PR-JNA658237).

Genomes were assembled using CLC Genomic Workbench v. 20.0.4 (QIAGEN, United States) and SPAdes v. 3.13.0 0 (St. Petersburg genome assembler, Russia, URL: http://cab.spbu.ru/software/spades/). CGView Server (http://stothard.afns.ualberta.ca/cgview_server/) [12] was used to check the assembly results. Genome annotation was performed using the RAST (Rapid Annotations using Subsystems Technology) server [13, 14]. Prophage sequences were searched using PHASTER (PHAge Search Tool Enhanced Release, https://phaster.ca/) [15]. MLST of the core genome (cgMLST) of the isolates was determined using the BIGSdb-Lm resources [6].

RESULTS AND DISCUSSION

Variety of Genotypes of Listeria Isolates Obtained from Meat-Processing Industries

Listeria was isolated from 29% of swabs at the first plant and from 11% of swabs at the second plant. The Listeria isolates obtained from the production environment were predominantly (81%) L. monocytogenes (see Table 1). L. welshimeri isolates made up 19%. The following genetic diversity of L. monocytogenes was characteristic for Plant 1: four genotypes of the phylogenetic lineage II were determined in the isolates (ST8, ST121, ST321, and ST2330). The ST8 isolates predominated, which were identified in swabs from the surfaces and cart wheels, joints of a deboning table, and a clipper machine. The next most frequent ST321 isolates are marked on the conveyors. ST121 and ST2330 (CC9, Clonal Complex 9) isolates were obtained from the cart wheel and door frame. Plant 2 was dominated by L. welshimeri, represented by genotypes 1050 and 2331 and found in swabs from the body of the mixer, the surface of the trolleys, and the plastic container. The L. monocytogenes ST8 isolate was obtained from a trolley wash.

Comparison of Isolates Obtained with Russian Isolates from Other Sources

Currently, the Listeria Pasteur MLST database (https://bigsdb.pasteur.fr/, July 27, 2022) contains information on 288 isolates from Russia. Analysis of these data and publications shows that ST8, which was predominant in the work environment in all sources considered (clinical human isolates, food and environmental isolates), did not occur prior to the COVID-19 pandemic. However, representatives of clonal complex 8 (CC8) differing from ST8 in one (SLV, single locus variant) or two (DLV, double locus variant) loci were noted in single cases in all sources: ST758, in the environment [8]; ST2096, in a clinical isolate in meningitis [3]; and ST16, in food products [16]. In December 2021–January 2022, four cases of invasive listeriosis caused by L. monocytogenes ST8 (ID 82490–82492, 82494) were noted in Moscow clinics. The isolate that is next largest by size, ST321, at the plant was discovered in Russia for the first time.

ST121 isolates are the most abundant in food products (meat, poultry, and fish), both according to our data for 2018–2019 [3, 5] and according to the State Research Center for Applied Biotechnology and Microbiology in 2017 [16]. ST121 has not been found in other sources.

ST2330 belongs to CC9 and is SLV ST9 at the dat locus. ST9 isolates were obtained in Russia from a variety of food products [3, 5, 16, 17], as well as from clinical meningitic material in 2015 and 2016 [16]. However, CC9 isolates were not found in the sample of clinical isolates in 2018–2019 [3]. L. welshimeri was noted among Russian isolates for the first time.

Analysis of Internalin Profiles (IPs) of Obtained L. monocytogenes Isolates

Internalin profiles (IPs) are presented in Table 1. All ST8 isolates had the same IP53; identical IPs were determined both in clinical ST8 isolates and in isolate ST2096 (CC8) [3], which is an SLV by the cat gene fragment. The internalin profile of the ST121 isolate did not differ from the IP48 isolates of this genotype previously obtained from food products [5]. For isolates of the new ST321, the profile of internalins also turned out to be new; the 21 inlA allele was determined for the first time. In isolate ST2330 (CC9), IP differed by the inlE allele from the profile of ST9 isolates obtained from food products [5].

The inlA, inlB, inlC, and inlE internalin genes were absent in L. welshimeri isolates. It should be noted that, while 32 internalin genes were identified in the genome of L. monocytogenes ST7 (clinical isolate GIMC2009:LmcUH4, GenBank Accession Number CP060435), only eight internalin genes were found in the genomes of L. welshimeri GIMC2049:LmcC11 and GIMC2051:LmcC15 (ST2331 and ST1005, respectively). All eight translated polypeptides corresponded to the characteristics of internalins of the LPXTG class [18]: they had a signal peptide at the N-terminus, a region of leucine-rich repeats (LRRs), and a C-terminal sorting signal that determines covalent interaction with peptidoglycan. These polypeptides were similar to L. monocytogenes internalins in 49–89% of cases. Of the remaining domains found in eight polypeptides, the MucBP (mucin-binding protein) domain, which determines adhesion to mucins of intestinal mucus, is well characterized [18]. In three internalins of L. welshimeri, we identified from one to four MucBP domains.

Genomic Characteristics of L. welshimeri Isolates

L. welshimeri detected at Plant II are represented by two genotypes in the context of the MLST genes, 1005 and 2331 (see Table 1). Analysis of BIGSdb-Lm isolates showed that the database, together with the three isolates that we entered, contains 19 L. welshimeri isolates of seven different genotypes, 42% of which belong to ST1005. The isolates were obtained mainly (84%) from the environment of food production. ST2331 was new on this list.

Comparison of the genomes of the sequenced isolates GIMC2049:LmcC11 and GIMC2051:LmcC15 with the genome of the reference strain L. welshimeri NCTC11857 (GenBank Accession Number LT906444) isolated from rotting plant remains showed that the genomes are very close in chromosome size and smaller than the chromosome of L. monocytogenes EGD-e at 110706145573 bp, which is associated with the loss of genes of the main pathogenicity island of Listeria, LIPI-1 [19]. However, the ST2331 isolate was distinguished by the presence of a plasmid. The 57530 bp plasmid, according to the BLAST results, was homologous to eight plasmids deposited in the Gen Bank NCBI (Table 2).

Table 2. Plasmids of Listeria homologous to the ST2331 plasmid of L. welshimeri
Full size table

As can be seen from Table 2, most of the plasmids were obtained from L. monocytogenes isolates of two phylogenetic lineages, both from patients with listeriosis and from foods and the environment. One plasmid was found in the L. innocua isolate. It should be noted that four isolates represented CC3 of the first phylogenetic line, and, of three isolates of phylogenetic line II, two isolates were clinical. Most of the plasmids were 57.6–57.8 kb in size, excluding pN1-011A, which was significantly larger than the other plasmids of the group. Similarity of the plasmids with the L. welshimeri ST2331 plasmid was 99.88–99.94%. In the L. welshimeri ST2331 plasmid, we found residues of the Listeria phage A006, genes for the CRISPR-associated Cas5 protein, Clp protease, NADH peroxidase, PemI/PemK toxin and antitoxin (MazE/MazF), betaine ABC transporter, ATPase, and transcription regulator efflux systems of heavy-metal ions (cadmium, zinc, lead, mercury). All these proteins are necessary to withstand environmental stresses [20].

Genomes of the obtained L. welshimeri ST2331 and ST1005 isolates were characterized by the presence of an extended region including the genes responsible for the metabolism of cobalamin, ethanolamine, and propanediol. Ethanolamine and propanediol are products of anaerobic utilization of sugars in the intestines of vertebrates and are the only sources of carbon, as well as a source of nitrogen for bacteria that persist in the gastrointestinal tract; cobalamin serves as a cofactor for these two metabolic pathways [21]. The presence of such a genome locus is a characteristic feature of Listeria sensu strictu, to which cluster L. welshimeri belongs. Note that the operons responsible for these metabolic pathways in the sequenced genomes of L. welshimeri also contain genes encoding proteins that form a selectively permeable icosahedral protein shell of microcompartments in which enzymatic reactions occur [22, 23]. The presence of such microcompartments, or polyhedral structures 100–200 nm across, is also a distinguishing feature of bacterial cells that are capable of surviving in the intestines of vertebrates.

In addition, the genomes of L. welshimeri ST2331 and ST1005 contain the agrABCD operon, which encodes a system of proteins (accessory gene regulator protein A, B, C, D) involved in the direct and indirect regulation of more than 650 genes associated with the survival and advantages in various ecological niches, adhesion to surfaces, biofilm formation, cell invasion, virulence, and global changes in gene expression [24, 25]. Thus, the genomic characteristics of the L. welshimeri isolates obtained confirm the broad adaptive capabilities of this Listeria species.

CONCLUSIONS

A pilot study of Listeria in swabs obtained at two meat processing plants in Moscow showed the presence of Listeria in the production environment and made it possible to describe their genetic characteristics, which are necessary for the current comparison with domestic and international data, as well as for the subsequent tracking of persistent L. monocytogenes genotypes: ST8, ST321, and ST121 and CC9(ST2330), which we detected at the production facilities in Moscow. According to the data of French researchers obtained on 7342 isolates of the French National Reference Center for Listeria, L. monocytogenes CC121 and CC9 are strongly correlated with food products, CC8 and CC321 are intermediate, i.e., occur both in products and in clinical specimens [26]. At the same time, in Austria, L. monocytogenes CC8 and CC9 were more often found in meat products, while CC121 was found less often [27], and, in a joint study of collection strains from Canada and Switzerland, L. monocytogenes CC8 and CC9 prevailed in food products, CC321 was detected less often, and CC121 was found only in individual cases [28].

In the second food-production plant, we isolated L. welshimeri of two genotypes, ST2331 and ST1005, the genomic characteristics of which indicated their adaptive capabilities both to production conditions and to the metabolic peculiarities of the gastrointestinal tract of animals. A study conducted on food-production facilities in the United Kingdom and Ireland also identified L. welshimeri, which contains genes for resistance to disinfectants, cadmium, and plasmids, among persistent contaminants [2]. During the implementation of a project of Cornell University (United States) [29], it was shown that L. welshimeri persisting in the production environment, according to the results of whole-genome sequencing, differs from isolates from the soil, which confirms that L. welshimeri evolves when adapting to production conditions.

Thus, the use of molecular-genetic methods contributed to the detection of the diversity of Listeria in meat-processing industries, which laid the foundation for monitoring the safety of the food-production environment.