Retrospective Study of Listeria monocytogenes Isolated in the Territory of Inner Eurasia from 1947 to 1999

Listeriosis is one of the most significant humans and animals foodborne infectious diseases. Here, we characterized 48 Listeria monocytogenes strains isolated in the territory of inner Eurasia during the second half of the 20th century. A total of 23 strains (52.3%) were susceptible to the nine antibiotics tested, 30.43%, 15.22%, and 8.7% were resistant penicillin G, ampicillin, and enrofloxacin, respectively. We applied the multilocus sequence typing (MLST) scheme to determine the phylogenetic positions of the strains. All but one strain belonged to the II phylogenetic lineage, and the majority of the strains belonged to one of the previously described clonal complexes (CCs). More than 60% of the strains belonged to the clonal complex CC7 that prevailed among all sources, including cattle (58%), small ruminants (64%), rodents (71%), and humans (50%). Further, CC7, CC101, and CC124 were found among human isolates. The MLST scheme was supplemented with virulence gene analysis. In total, eight inlA, six inlB, and six inlC allelic variants were found, and all but one strain carried one of the two inlE alleles. Most strains (62.5%) belonged to the same multivirulence locus sequence typing (MvLST) type, which includes CC7, inlA allele 4, inlB allele 14, inlC allele 6, and inlE allele 8.


Introduction
Listeriosis is an infectious disease of humans and animals that results in high hospitalization and fatality rates [1,2]. The annual incidence of listeriosis ranges from 0.2 to 0.5 cases per 100,000 population, while average fatality rates are estimated to be 20%-30% of hospitalized patients and can reach 40% in outbreaks, including elderly persons or immunosuppressed patients [3,4].
Listeriosis is currently considered to be one of the most significant food infections [5,6]. The worldwide epidemic situation continues to deteriorate for a number of reasons, including specific biological features

L. monocytogenes Virulence
When tested in guinea pigs, 40 L. monocytogenes strains (83%) caused progressive conjunctivitis and keratitis with generalized infection. In one (2%) case, the generalized infection ended in death (strain 211). Seven strains (15%) were incapable of causing conjunctivitis and keratitis.
A total of 17 strains were additionally tested in mice. Strains were selected by random sampling from various sources. Fourteen strains (82%) were virulent for mice, for which the LD 50 ranged from 10 4 to 10 7 CFU per mouse ( Table 2). Two strains (11.8%), including strain 2598 isolated from rabbit and strain 39 isolated from guinea pig, were moderately virulent for mice, although they caused purulent keratoconjunctivitis in guinea pigs. Strain "A," isolated from ticks, was avirulent for mice and unable to cause keratoconjunctivitis in guinea pigs.
Besides CC7, four more clonal complexes were represented by more than one sequence type. The clonal complex CC18 was represented by three sequence types, and CC124, CC101, and CC177 were represented by two sequence types. At least one sequence type found in strain 45 from sheep detected in Kazakhstan did belong to new ST1549 and clonal complex CC307. In total, eight new sequence types were described. The studied strains information is presented at the L. monocytogenes MLST database at https://bigsdb.pasteur.fr/.
With the exception of CC7 strains, strains isolated from small ruminants, cattle, and humans were quite heterogeneous, although all but one belonged to the II phylogenetic lineage. Five isolates from small ruminants belonged to CC124, CC18, CC101, CC21, and a new ST which has not been previously described. CC177 (n = 3), CC124 (n = 2), CC18 (n = 1), and CC21 (n = 1) were found in strains isolated from cattle. Besides isolates from cattle and small ruminants, CC124 was also found in strains isolated from a rabbit and ticks. New STs profiles belonging to CC18 and CC19 were found in strains isolated from pig and rabbit, respectively ( Figure 1). Strains isolated from humans belonged to CC7 (n = 3), CC89 (n = 1), CC177 (n = 1), and CC1 (n = 1, I phylogenetic lineage).
With the exception of CC7 strains, strains isolated from small ruminants, cattle, and humans were quite heterogeneous, although all but one belonged to the II phylogenetic lineage. Five isolates from small ruminants belonged to CC124, CC18, CC101, CC21, and a new ST which has not been previously described. CC177 (n = 3), CC124 (n = 2), CC18 (n = 1), and CC21 (n = 1) were found in strains isolated from cattle. Besides isolates from cattle and small ruminants, CC124 was also found in strains isolated from a rabbit and ticks. New STs profiles belonging to CC18 and CC19 were found in strains isolated from pig and rabbit, respectively ( Figure 1). Strains isolated from humans belonged to CC7 (n = 3), CC89 (n = 1), CC177 (n = 1), and CC1 (n = 1, I phylogenetic lineage).

Distribution of the L. monocytogenes Clonal Complexes on the Territory of Inner Eurasia
All L. monocytogenes strains were obtained from inner Eurasia territories. To simplify analysis, the regions were divided into (1) the European region, which unites the territories of current eastern European countries (Ukraine and Belarus) and central regions of Russia (Moscow, Voronezh, Kursk, Ryazan, Yaroslavl, Tula, and Novgorod regions); (2) the Asia region, which includes territories of Central Asian countries (Tajikistan, Kazakhstan, and Uzbekistan) and territories of the Russian Federation  All but one L. monocytogenes strain isolated in the European region belonged to the II phylogenetic lineage and included the main clonal complexes CC7 (n = 15), CC124 (n = 4), CC177 (n = 1), CC18 (n = 4), CC89 (n = 1), and CC19 (n = 1). Only one strain, isolated in Moscow in 1988, belonged to the I phylogenetic lineage. Strains isolated in the Asia region belonged to CC7 (n = 11), CC177 (n = 2), CC124 (n = 1) CC21 (n = 1), and CC101 (n = 1), and one strain belonged to a new sequence type that was not associated with any described clonal complex. In the far eastern and Siberia region, strains of CC7 (n = 3), CC177 (n = 1), and CC101 (n = 1) were found. Therefore, CC7 was the most frequent in the whole territory in inner Eurasia.

Internalin Gene Diversity in L. monocytogenes Isolates
In this study, previously described internal fragments of the internalin genes inlA, inlB, inlC, and inlE [39] were sequenced to supplement the MLST analysis with virulence gene analysis.
In total, 8 inlA allelic variants were found. Alleles 1, 4, 6, 8, 9, and 12 have been described in previous studies while alleles 14 and 15 were detected for the first time. The inlA gene was not detected in strain 45 isolated from sheep in Kazakhstan, which belongs to a newly described ST. It is noteworthy that each allele of the inlA gene correlated with a specific CC (Table 4).

Internalin Gene Diversity in L. monocytogenes Isolates
In this study, previously described internal fragments of the internalin genes inlA, inlB, inlC, and inlE [39] were sequenced to supplement the MLST analysis with virulence gene analysis.
In total, 8 inlA allelic variants were found. Alleles 1, 4, 6, 8, 9, and 12 have been described in previous studies while alleles 14 and 15 were detected for the first time. The inlA gene was not detected in strain 45 isolated from sheep in Kazakhstan, which belongs to a newly described ST. It is noteworthy that each allele of the inlA gene correlated with a specific CC (Table 4).
All studied L. monocytogenes strains contained one of six alleles of the inlB gene, and the dominant allele 14 was detected in 35 (73%) strains belonging to CC7, CC177, CC21, and CC19. Strains belonging to CC1, CC124, CC101, and CC89 carried alleles 1, 12, 15, and 17, respectively (Table 4). Strain 45, which belongs to a new sequence type, did not correlate with any clonal complex carrying inlB allele 16 (data not shown). The variability of inlC, and especially inlE, was low. In total, 6 inlC alleles were found, among which allele 6 was predominant and found in 40 of 48 (83%) strains belonging to CC7, CC21, CC19, CC124, CC177, and CC18. Other inlC alleles were distributed in a CC-specific manner and were represented in single strains. As for inlE, all but one strain carried one of two alleles, either 6 or 8. The only exception was the phylogenetic lineage I strain K-23, which carried alleles distinct from the other studied genes. Interestingly, all strains carrying inlE allele 8 were isolated in the Asian part of inner Eurasia, and all but one strain carrying allele 6 of inlE was isolated in the European territory. The exception was the strains of CC7 that carried allele 8 and were widely spread over the whole inner Eurasia territory.
Polymorphism among housekeeping genes included in the MLST scheme and virulence genes included in the internalin profile (IP) was compared only for strains of lineage II, to avoid the introduction of false variability which would have been caused by the addition of a single lineage I strain into the analysis because of a high percentage of conservative substitutions between lineages [16,38]. It was shown that for the studied collection, the polymorphism of virulence genes was lower or comparable with the polymorphism of housekeeping genes. Particularly, the virulence gene inlE had the lowest number of alleles relative to all studied genes, and the housekeeping gene ldh had the highest number of alleles (2 vs. 18) ( Table 5). A phylogenetic tree was constructed on the basis of concatenated sequences of the MLST and internalin gene alleles (Table 4 and Figure 3). The major cluster, Cluster I, was identified from phylogenetic analysis, and included strains containing CC7, inlA allele 4, inlB allele 14, inlC allele 6, and inlE allele 8. Other clusters (II-IX) were observed and clustered according to clonal complexes of L. monocytogenes strains and internalin gene alleles profiles (IP). Cluster II included two strains from sheep and cow belonging to CC101 with the internalin profile (IP) 14, 20, 15, 8 for inlA, inlB, inlC, and inlE, respectively. Cluster III was represented by five strains isolated from different sources in Europe and belonged to CC124 and IP 6, 12, 6, 6. Clusters IV-VI were represented by one strain each belonging to distinct clonal complexes (CC21, CC19, CC89, respectively), and carrying distinct inlA alleles (alleles 12, 9, and 15, respectively). These three clusters had the same alleles of inlB and inlE (14 and 8, respectively). Clusters IV and V shared the inlC allele (6) while Cluster VI carried the inlC allele 17. Four strains from the cattle had Cluster VII profile and belonged to CC177 with IP 8, 14, 6, 8 that differed from IP of Clusters IV and V only by the inlA allele. Cluster VIII was represented by three strains obtained from different sources in Europe and belonged to CC18 with IP 12, 13, 6, 6. The only one strain isolated from the human case belonged to the 1 phylogenetic lineage, Cluster IX, CC1 and IP 1,9,1,3.

Discussion
Here, we performed a retrospective study of L. monocytogenes strains isolated in the territory of inner Eurasia in the second part of the 20th century, when countries of the former Soviet Union were to certain extent isolated from the rest of the world. Animal isolates prevailed in the studied collection while only six human isolates were included that is in line with the rare occurrence of human listeriosis until the 1980s [41][42][43]. Food isolates were not included into the study as food products began to be considered as a source of listeriosis in the 80s, and a mandatory check of products for listeriosis was introduced in Russia in 2002, so food isolates dated before 1990 are absent from the National Collection of the Federal Center for Virology and Microbiology (Volginsky, Russia), which was the strain source.
The analysis of antibiotic resistance showed that about half of the strains were sensitive to all antibiotics tested. This is consistent with the rarer use of antibiotics during that period, and is in stark contrast with the modern prevalence of antibiotic-resistant strains among animal and environmental L. monocytogenes isolates [44][45][46].
The obtained results suggested the prevalence of lineage II strains among animal isolates obtained in the territory of inner Eurasia from 1947 to 1999. These data are in line with previously obtained results which demonstrated the prevalence of lineage II L. monocytogenes strains among wild animals inhabiting natural foci of infection in the European part of Russia [16]. In general, the comprehensive analysis of isolates obtained in European countries showed the prevalence of lineage I among clinical isolates, and lineage II among food and environmental isolates [20,44,47]. Both lineage I and II strains were isolated from wild animals that inhabited different regions of Germany and Austria [48]. Prevailing of Lineage II was described in some studies performed in Sweden [49,50], although comparison with these studies was restricted by inconsistence of isolation sources. A more consistent study from the point of view of the isolation source study showed longitude prevalence of Lineage II strains at three Finnish dairy farms during 2013 to 2016 [15].
The current study included only five strains isolated in the far eastern part of Eurasia, all of which belong to lineage II. The phylogenetic lineage I prevailed among isolates obtained from invasive diseases in humans and wild animals in the far eastern region of Russia at the beginning of the 21st century [9]. Other studies demonstrated similar representations of the I and II phylogenetic lineages among L. monocytogenes strains isolated from the intestinal contents and feces of wild animals in Japan and China in this century [25,51]. Taken together, the available data suggest that the distribution of the I and II lineages might be different for the European and Central Asian parts of Eurasia and its far eastern part. More representative studies are required to clarify the phylogenetic pattern of L. monocytogenes circulating in this region.
Further, recent as well as earlier studies have demonstrated the prevalence of the clonal complex CC7 among isolates obtained from the European and Central Asian parts of inner Eurasia (see Table 4 and Figure 1, [16,38]). CC7 is one of the largest-spreading clonal complexes worldwide. It includes the epidemic clone VII (ECVII), which is associated with the foodborne outbreak of listeriosis that was associated with the consumption of cantaloupe during 2011 in the United States [52]. The distribution of L. monocytogenes clones varies across countries around the world. Clonal complexes CC1 and CC2 of phylogenetic lineage I have prevailed in all regions of the world, with the exception of North Africa for CC1. CC2 clones have been found in 30 countries. CC3 is one of the most common clones in all regions, while CC9 of the lineage II was shown to be the third largest in Europe and the Western Hemisphere [31]. The predominant clonal complexes in China are CC9 and CC8 in food and ST87 in clinics [53,54]. In the last decade, CC6 noticeably increased in significance as a causative agent of human listeriosis in Africa, Europe, and North America [31,33,[55][56][57]. In Russia, current human listeriosis is associated with the CC1, CC2, and CC20 strains [38].
Analysis of virulence gene variability revealed a difference between the distribution of inlA and inlB respectively encoding two major invasion factors, InlA and InlB. While inlA is distributed according to a phylogenetic position and is well correlated with clonal complexes, the inlB allele 14 was found in strains belonging to four of the nine revealed clonal complexes. This prevalence might be due to some selective advantage of this allele. Recently, we demonstrated that inlB allele 14 provided the highest level of L. monocytogenes dissemination to the internal organs of mice when infection was performed via intragastric inoculation, which is the best model of the natural infection process [58]. All lineage II strains isolated from internal mouse organs in this and previous studies carried the inlB allele 14, and mice might be an important host for L. monocytogenes in the natural focus of infection [38]. Variability analysis of two other IP genes, inlC and especially inlE, showed that their sequences were as conserved as the most conservative housekeeping genes.
Taken together, the obtained results demonstrate that L. monocytogenes strains belonging to CC7 prevailed among isolates obtained in the territory of inner Eurasia in the second half of the 20th century. Resistance to antibiotics occurred among these strains, but about 50% of the isolates were sensitive to all tested antibiotics with the exception of polymyxin B. Analysis of virulence gene variability demonstrated different patterns of distribution for genes encoding proteins of the internalin family among these strains.

Bacterial Strains and Culturing Conditions
A total of 48 L. monocytogenes strains were obtained from the National Collection of the Federal Research Center for Virology and Microbiology (Volginsky, Russia) for further analysis (Table 6). Independent isolates were chosen on the basis of having been isolated from humans and animals in different regions of inner Eurasia from 1947 to 1999. Strains isolated from animal insect and human parasites were included. All strains from biological sources (with the exception of rodents and insects) were isolated from postmortal material taken from animals who died from listeriosis, daveric (sectional) material. Isolates were obtained from animals which died from listeriosis. Serotypes were characterized with antisera. Strains were kept lyophilized at −18 • C with bovine serum or gelatin agar medium as a stabilizer.  Lyophilized L. monocytogenes cultures were restored using 1 mL of tryptone soya yeast extract broth (TSB-YE, HiMedia, India). The strain suspension was immediately plated on tryptone soya yeast extract agar (TSA-YE, HiMedia, India). Plates were incubated at 37 • C for 24 h; then, colonies were counted and up to 10 typical colonies were replated for further identification and characterization. The remainder of the initial suspension was supplemented with sterile TSA-YE and incubated at 37 • C for 24 h without shaking to guarantee culture viability in the absence of colony growth on agar plates.

Antibiotic Resistance Test
L. monocytogenes strains were tested for their resistance to antibiotics using the disc diffusion method on Mueller-Hinton agar (HiMedia, India), as described by the performance standards for antimicrobial disk and dilution susceptibility test for bacteria isolated from animals (CLSI) [59]. Penicillin G (10 µg/disc), enrofloxacin (5 µg/disc), ampicillin (10 µg/disc), tetracycline (30 µg/disc), levomycetin (chloramphenicol) (30 µg/disc), kanamycin (30 µg/disc), tylosin (15 µg/disc), streptomycin (30 µg/disc), polymyxin B (30 µg/disc), and neomycin (30 µg/disc) were used (NICF, LLC; St. Petersburg, Russia). These antibiotics were chosen for use or preference in veterinary medicine for the treatment of infectious diseases in animals in Russia. After 24 h of incubation, the zones of inhibition were measured (mm), and the strains were categorized as susceptible, intermediate, or resistant to specific antibiotics as per the criteria of CLSI [59]. The breakpoints of Staphylococcus, Streptococcus, and Enterococcus species resistance were considered since no resistance criteria exist in the CLSI guidelines for Listeria susceptibility testing [25].

Anton's Eye Test
Anton's eye test was performed on guinea pigs [60]. Strains were recognized as pathogenic of purulent keratoconjunctivitis when it developed in guinea pigs within 3-5 days.

L. monocytogenes Virulence in Mice
Virulence was estimated by the 50% lethal dose (LD 50 ), determined by the probit method, using white laboratory outbred Swiss mice weighing 14-16 g strains were randomly selected among isolates obtained from a particular source. Bacteria were grown on TSA-YE slopes (HiMedia, India) for 18 h at 37 • C. The culture was standardized turbidimetrically and diluted appropriately in phosphate-buffered saline (PBS, pH 7.3). For each strain, five out of six mice were injected intraperitoneally with 0.5 mL of dilution. Dilutions were 1.0 × 10 9 , 2.0 × 10 8 , 4.0 × 10 7 , 8.0 × 10 6 , and 1.6 × 10 6 CFU per animal. Each inoculum was checked by determining the viable counts on TSA-YE plates.
The LD 50 was calculated using the following formula: where lgDN is the logarithm of the maximum infecting dose, lgδ is the logarithm of the multiplicity of dilution (the ratio of each subsequent dose to the previous one), Li is the ratio of the number of animals who died from a given dose of the inoculation to the total number of animals inoculated in this dose, and Li is the sum of the Li values calculated for all doses tested. All strains with LD 50 s from 1.0 × 10 4 to 1.6 × 10 7 CFU per mouse were considered as virulent. A strain was nonvirulent if it did not cause death in the injected group of mice 7 days after injection with 1.0 × 10 9 CFU [44]. Animal experiments were performed according to the European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific Purposes (Strasbourg, 1986). All procedures involving animals were approved by the Ethic Committee of the Federal Research Center for Virology and Microbiology (Volginsky, Russia) (ref. no. 5 on February 28, 2019). After infection, the surviving mice were euthanized by CO 2 asphyxiation.

PCR
Lysates of overnight L. monocytogenes cultures were obtained as previously described [38]. The Encyclo Plus PCR kit (Evrogen, Russia) and primers (Syntol, Russia) were used for DNA amplification. The MLST scheme based on the sequences of seven housekeeping genes that was developed by Ragon et al. was used [28]. The housekeeping gene fragments were amplified as previously described, with modifications [16]. The following temperature conditions were used: 94 • C, 2 min; (92 • C, 30 s; 55 • C, 30 s; 72 • C, 2 min) × 30; 72 • C, 10 min; melting temperatures (Tm) were 60 • C for bglA and ldh. PCR products were evaluated by electrophoresis in 1% agarose gel. Amplified DNA was eluted from the reaction mixture using Cleanup Standard (Evrogen, Russia).

PCR Product Sequencing
PCR products were sequenced on both strands according to the BigDye Terminator 3.1 Cycle Sequencing protocol for the Genetic Analyzer 3130 of Applied Biosystems/Hitachi. Electrophoretic DNA separation was performed in 50-cm capillaries with POP7 polymer.