Atypical Serogroup IVb-v1 of Listeria monocytogenes Assigned to New ST2801, Widely Spread and Persistent in the Environment of a Pork-Meat Producing Plant of Central Italy

In this study, we characterized 84 Listeria monocytogenes (Lm) strains having an atypical IVb-v1 profile and isolated in a meat producing plant of Central Italy. They were assigned to the new MLST type ST2801 (CC218). The new ST was widespread in the food-producing environment where it was able to persist for over a year even after cleaning and sanitation. Cluster analysis identified three main clusters genetically close to each other (0–22 allelic differences and 0–28 SNPs) from two different cgMLST types, suggesting a common source. The coexistence of closely related clusters over time could be the result of a different evolution path starting from a common ancestor first introduced in the plant and/or the consequence of the repetitive reintroduction of closely related clones probably by raw materials. All the strains presented several determinants for heavy metals resistance, stress response, biofilm production, and multidrug efflux pumps with no significant differences among the clusters. A total of 53 strains carried pLI100 and the j1776 plasmids, while in one strain, the pLM33 was found in addition to pLI100. Only the strains carrying plasmids presented cadA and cadC for cadmium resistance and the mco gene encoding a multicopper oxidase and gerN for an additional Na+/H+-K+ antiporter. All the strains presented a virulence profile including a full-length inlA gene and the additional LIPI-3. The isolation of a new ST with a large pattern of stress-adaptation genes and able to persist is an important contribution to deepening the current knowledge on the uncommon IVb-v1 and in general on the genomic diversity of Lm.


INTRODUCTION
Listeria monocytogenes (Lm) is the foodborne pathogen causing human listeriosis, the most serious foodborne disease under EU surveillance with the highest proportion of hospitalized cases and fatality rate (13%). Invasive forms of the disease mainly affect people at risk causing abortion and stillbirth in pregnant women and meningitis septicemia and death in the elderly, immunocompromised people, and newborns (European Food Safety Authority, and European Centre for Disease Prevention and Control, 2021). Lm is a genetically heterogeneous species including hypo-and hypervirulent clones (Maury et al., 2016(Maury et al., , 2019 and strains able to survive and persist in food-producing environments (FPE) even for years, due to their adaptation to different environmental stresses such as cold temperatures, high salinity, low pH, oxidation, and desiccation (Palma et al., 2017;Pombinho et al., 2017;Maury et al., 2019;Parsons et al., 2020;Guidi et al., 2021;Palaiodimou et al., 2021).
Although the first studies published on this atypical IVb-v1 profile reported its isolation from milk products and meat products (Leclercq et al., 2011), the most recent works show a particular association with vegetable matrices such as caramel apple, stone fruits, leafy green, and radicchio as well as the associated processing environments (Torresi et al., 2020;Yang et al., 2020;Chen et al., 2022). Scientific reports on different IVb-v1 clinical isolates are also available (Leclercq et al., 2011;Lee et al., 2012;Scaltriti et al., 2020).
To date, very few in-depth studies have been performed on the virulence genetic profile of Lm strains belonging to IVb-v1 and their genetic determinants involved in stress response. Chen et al. (2022), reported the presence of different stress 1 http://bigsdb.pasteur.fr/listeria response determinants in Lm IVb-v1 belonging to ST554 (low pH,cold,etc.) in the only published study describing the persistence of IVb-v1 strains in a food-processing plant. The same authors also defined the virulence profile for internalins and Listeria Pathogenicity Islands (LIPI) identifying the presence of inlA/B/C/E/F/H/J/K/P and the LIPI-3. The latter was also reported by de Tavares et al. (2020) in IVb-v1 Lm isolates belonging to ST218. In their report on a IVb-v1 Lm isolated from a vegetable matrix, Torresi et al. (2020) described the presence of a plasmid which carried heavy metal-resistance genes, but no information about the virulence profile was provided.
During the extensive environmental sampling plan for Lm performed in a pork meat-producing plant of Central Italy between 2020 and 2021, we isolated several strains having the atypical IVb-v1 serogroup profile and belonging to a new ST assigned by the Institute Pasteur (ST2801).
In this study, we characterized the new ST2801 isolates to (i) evaluate the genomic correlation existing among them, (ii) assess their persistence in the meat producing plant, (iii) investigate the presence of genetic determinants involved in environmental stress adaptation, and (iv) define virulence profiles.

Environmental Sampling and Listeria monocytogenes Detection
During the period between July 2020 and September 2021, an extensive environmental sampling for Lm was performed in a pork meat-producing plant in Central Italy. Three different sampling sessions were performed during production, in particular, on July 2020 (Production 1), May 2021 (Production 2), and September 2021 (Production 3). A sampling scheme of 63 surfaces including both food contact (FCS) and nonfood contact surfaces (NFCS) was defined focusing on the main niches for Lm presence (Supplementary Table 1). In each sampling session, the same surfaces were sampled using commercial sterile sponges.
In accordance with the European Union Reference Laboratory for Lm (EURL) guidelines (Carpentier and Barre, 2012), the total sampled area varied depending on the sampling site but was as large as possible to improve the probability of detecting Lm. The samples were tested according to ISO 11290-1:2017 for Lm detection.
If positive surfaces were found, they were sampled again after extraordinary cleaning and sanitation.

Strains Collection
Up to five Lm colonies from each positive sample were randomly selected and screened for their belonging to one of the five major serogroups (IIa, IIb, IIc, L, and IVb), using a multiplex PCR assay according to the EURL method (Doumith et al., 2004;Kérouanton et al., 2010). At least one isolate for each serogroup found in each sample was selected to be subjected to whole genome sequencing (WGS). In this study, we focused on Lm strains presenting the atypical serogroup IVb-v1 (Lee et al., 2012).

Whole Genome Sequencing and Bioinformatic Analysis
DNA extraction was performed according to Portmann et al. (2018), with minor modifications, using the QIAamp DNA Mini Kit (Qiagen Hilden, Germany) according to the manufacturer's protocol.
The purity of the extracts was evaluated using NanoDrop2000 (ThermoFisher Scientific, Wältham, MA, United States). Starting from 1 ng of input DNA, the Nextera XT DNA chemistry (Illumina, San Diego, CA, United States) for library preparation was used according to the manufacturer's protocols. WGS was performed on the NextSeq 500 platform (Illumina, San Diego, CA, United States) with the NextSeq 500/550 mid output reagent cartridge v2 (300 cycles, standard 150-bp paired-end reads).
Genome de novo assembly of paired-end reads was performed using SPAdes version 3.11.1 4 (Bankevich et al., 2012) with the parameters suggested by the manual for the Illumina platform 2_150 chemistry (-only-assembler -careful -k 21, 33, 55, 77). Then, the genome assembly quality check was performed using QUAST version 4.3 5 (Gurevich et al., 2013). All the genomes that met the quality parameters recommended by Timme et al. (2020) were used for the subsequent analysis steps.
The genome assemblies were deposited at DDBJ/ENA/GenBank under the BioProject PRJNA821663.
Multilocus Sequence Typing Analysis, Core Genome Multilocus Sequence Typing, and Single-Nucleotide Polymorphisms Analysis The sequence type (ST) and the clonal complex (CC) were deducted in silico using the specific tool available on the BIGSdb-Lm database (accessed on October 2021) 6 and based on the MLST scheme including the seven housekeeping genes abcZ, bglA, cat, dapE, dat, ldh, and lhlA (Ragon et al., 2008).
To verify the relatedness among the isolates, identifying genomic clusters, a cgMLST analysis was performed using the chewBBACA 7 allele calling algorithm (Silva et al., 2018) and the Pasteur Institute cgMLST scheme of 1,748 loci (Moura et al., 2017). According to the guidelines for Lm cgMLST typing (Moura et al., 2017), only the genomes with at least 1,660 called loci (95% of the full scheme) were considered. The software GrapeTree 8 (Zhou et al., 2018) was used for the visualization of the minimum spanning tree (MSTreeV2 method).
A core single-nucleotide polymorphism (SNPs) analysis was performed using the reference-free tool KSNP3 9 with a kmer size of 21 (Morganti et al., 2016). The resulting neighbor-joining (NJ) tree was visualized using the interactive tree of life (iTOL). 10 Genetic Determinants Involved in Stress Adaptation, Biofilm Formation, and Virulence Potential All the genome assemblies were manually screened for the presence/absence of loci encoding for disinfectants and metal resistance and stress survival islets (SSIs) using the "Metal and Detergent Resistance" and the "Stress Islands" tools available on the BIGSdb-Lm platform (accessed on January 2022).
The virulence profile of the strains was deduced using the "Virulence" tool provided by the BIGSdb-Lm platform, also investigating the presence of premature stop codon mutations in the inlA gene (accessed from October 2021 to January 2022).

Environmental Sampling and Listeria monocytogenes Strain Collection
During the period between July 2020 and September 2021, 189 environmental samples were collected in the pork meatproducing plant of the study during production and 40 after cleaning and sanitation. A total of 35 samples of those collected during production (18.5%) and 7 samples of those collected after cleaning and sanitation (17.5%) tested positive for the presence of Lm. A total of 147 Lm strains were isolated from these samples and screened for serogroup. Among them, 84 Lm strains (Supplementary Table 2), isolated from 19 environmental surfaces, presented the atypical IVb-v1 serogroup profile carrying the lmo0737 target gene in addition to the four targets normally characterizing the serogroup IVb (ORF2110, ORF2819, prs and prfA). These atypical strains were selected for further genomic analysis.

Multilocus Sequence Typing Analysis, Core Genome Multilocus Sequence Typing, and Single-Nucleotide Polymorphisms Analysis
For all the Lm strains IVb-v1, the MLST analysis found exact allele matches for only six of the seven genes of the scheme. After submitting the genomes to the BIGSdb-Lm database (21 October 2021), a new allelic id was defined for the locus abcZ resulting in a new MLST profile (named ST2801 and belonging to clonal complex CC218).
The cgMLST analysis showed that the microbial population associated with the new ST was heterogeneous and allelic differences among the strains ranged from 0 to 22 (Figure 1). Based on a seven allele threshold, three main clusters were identified (Figure 1). Cluster A (cgMLST type L1-SL218-ST2801-CT10448) included 56 strains and was isolated in the FPE after all the sampling sessions performed during production and after two consecutive cleaning and sanitation sessions subsequent to Production 1. This cluster was isolated from 15 different surfaces: n • 1, 2, 6,9,30,32,36,37,38,43,51,56,57,62,and 63 (Supplementary Table 2). Among these surfaces, n • 9, 30, 37, 38, and 63 were found to be contaminated with cluster A more than once over the course of time.
A total of 18 Lm strains, isolated during Production 1 and Production 3 and after two consecutive cleaning and sanitation sessions subsequent to Production 1 (I-Post sanitation 1 and II-Post sanitation 1), were grouped in cluster B (cgMLST type L1-SL218-ST2801-CT10448). No strains isolated during Production 2 were grouped in this cluster. A total of 4 surfaces of the scheme, n • 6, 30, 43 and 54, were found to be contaminated with cluster B, with surface n • 6 resulting positive more than once (Figure 1 and Supplementary Table 2).
The core SNPs analysis was performed to deepen the genetic relationships between the strains and the results were concordant with those of the cgMLST analysis (Figure 2). The number of SNP differences among all the isolates ranged from 0 to 28 (Supplementary Table 3). In particular, Lm strains grouped in cluster A differed by a number of SNPs ranging from 0 to 20 with a median of 4. Among strains of cluster B and cluster C, SNPs differences ranged from 0 to 5 (median 1) and from 0 to 11 (median 6), respectively.

Genetic Determinants Involved in Stress Adaptation and Biofilm Production
Using the BIGSdb-Lm platform together with the annotation results, several determinants for different heavy metal resistance were detected, as well as genes for multidrug efflux pumps and response to environmental stresses ( Table 1).
In all the strains, the determinants for heavy metal resistance included czcD, arsB, arsC, and the csoR-copA-copZ copper-resistance operon. The zosA gene for Zn(II) uptake was also found.
The same genetic determinants for stress responses were detected in all the strains. In particular, for tolerance to acid stress, in addition to the gadB-gadC operon, gadD, was also found in all the strains and represented the only gene of the SSI-1.
Several determinants associated with monovalent cation/proton antiporters were found in all the ST2801 strains. These genes included the multiple resistance and pH operons mrpABCDEFG and mdrP.
A total of 26 strains from cluster A and all the strains from cluster B and cluster C carried two different plasmids and in particular, the pLI100 (Acc. Number AL592102) together with the j1776 plasmid (Acc. Number CP006612) or the pLM33 (Acc. Number GU244485; Figure 2).
Only the strains that carried these plasmids presented cadA and cadC for cadmium resistance, the mco gene for copper detoxification and the additional Na+/H+-K+ antiporter GerN.
All the strains carried the genes luxS, recO, lmo2504, and lmo0673 involved in biofilm formation.
FIGURE 2 | Single-nucleotide polymorphisms (SNPs) analysis of Lm strains belonging to ST2801. The first, the second, and the third columns indicate the CT, the cluster, and the sampling date, respectively. The presence/absence matrix represents, from left to right, plasmids (j1776 plasmid, pLI100, and pLM33) and genes involved in stresses adaptation.

DISCUSSION
In this study, we reported the widespread isolation of Lm strains having an atypical IVb-v1 serogroup profile from the FPE of a meat-producing plant in Central Italy. To date, very little is known about the circulation of this uncommon variant in Italy, where only two studies reported the isolation of Lm strain IVb-v1 (Scaltriti et al., 2020;Torresi et al., 2020).
The first studies published by authors of other countries on this atypical IVb-v1 profile reported its isolation from milk products and meat products (Leclercq et al., 2011). However, the most recent works, including the one of Torresi et al. (2020), also show the association with vegetable matrices such as caramel apple, stone fruits, leafy green, and radicchio (Torresi et al., 2020;Yang et al., 2020;Chen et al., 2022). Moreover, IVb-v1 Lm strains have been previously isolated from FPEs. In more detail, Lee et al. (2012) typed three IVb-v1 strains isolated from a turkey processing plant of the United States during a 2-year sampling performed by Mullapudi et al. (2008) between 2017 and 2018. These strains were isolated from two drains and a chiller re-work table. Chen et al. (2022) used WGS to type IVb-v1 strains isolated from several environmental surfaces of an apple packinghouse facilities mostly during a 2-year environmental survey performed between 2016 and 2017 by Simonetti et al. (2021). In this FPE, Lm isolates belonging to IVb-v1, counted for more than 90% of all isolates and spread across the facility mostly contaminating floors and other NFCSs and persisting throughout the sampling year (Chen et al., 2022). These results were very similar to those obtained in our study and seemed to indicate that once strains belonging to this uncommon serogroup are introduced into an FPE, they are able to spread and persist mainly on NFCSs such as drains, floors, and cleaning materials.
Lm IVb-v1 strains have been identified among the isolates from human cases of listeriosis occurred in France, Brazil, Switzerland, and United States (Leclercq et al., 2011;Lee et al., 2012). These strains were responsible for maternal-neonatal forms, infection of central nervous system, and bacteremia. Also in Italy, a human case of listeriosis, caused by a IVb-v1 strain in 2016 was reported by Scaltriti et al. (2020).
All the IVb-v1 strains previously isolated worldwide, from both food and humans, belonged to four main STs and in particular to ST218, ST240, ST382, and ST554 (Chen et al., 2017(Chen et al., , 2022Kim et al., 2018;Yang et al., 2020). In particular, all the Italian isolates reported before belonged to ST218 within the CC218.Very interestingly, the IVb-v1 strains described in this study presented a new MLST allelic profile also included in the CC218 and assigned as ST2801. As a future perspective, the evolutionary linkage between these STs isolated in Italy could be investigated.
In this study, the ST2801 was isolated from different environmental surfaces, mostly NFCS, representing the main source of environmental contamination in the FPE. Most of these NFCS, represented ideal niches since they were difficult to clean and sanitize because of inherent inaccessibility (e.g., drains), harborage sites (e.g., porous wall-floor connections), wear surfaces (e.g., door seals), and cleaning tools (e.g., water pullets). On some of these NFCS, such as the wall floor connection of a cold room, the wheels of meat trolleys, the door handle and seal of another cold room, and the water puller of the washing area, the ST2801 was able to persist over the time even after cleaning and sanitation. Since the surfaces were all identified, these results were reported to the FBOs in order to increase their level of attention in terms of more accurate cleaning and sanitation or replacement of worn materials. The persistence of this ST probably indicated an efficient environmental adaptation also suggested by the presence of several genetic determinants for stress response, as discussed below. In their previous study, Chen et al. (2022) also reported the isolation of Listeria monocytogenes IVb-v1 strains from several NFCS and their ability to persist in the FPE.
CgMLST analysis identified three main clusters from two different types (CT10448 and CT11418), genetically close to each other. The results of SNPs analysis confirmed the presence of the same three close-to-clonal groups.
In particular, cluster A was the most represented and was isolated during all the sampling sessions performed both during production and after sanitation, persisting in the plant from July 2020 to September 2021. The same persistence period was observed for the smaller cluster B, which, however, unlike cluster A, was not isolated during the intermediate Production 2. Cluster C, instead, was isolated during Production 1 and Production 2 but not during Production 3. These three near-clonal clusters were found to be coexistent in the meat-producing plant until Production 2, while during Production 3, only cluster A and cluster B were isolated. As observed by Knudsen et al. (2017) and Chiaverini et al. (2021), the coexistence of closely related clusters over the time could be the result of a different evolution path starting from a common ancestor first introduced in the plant and/or the repetitive reintroduction of closely related clones probably by raw materials.
All the ST2801 strains carried several determinants for different heavy metals resistance, stress response, and multidrugs efflux pumps with no significant differences among the clusters.
Genes for heavy metals resistance included czcD encoding a cation diffusion facilitator protein family transporter reducing Cd 2+ , Zn 2+ , and Co 2+ accumulation in the cytoplasm (Osman et al., 2021), arsB and arsC encoding an ATPase and an arsenate reductase involved in arsenic resistance (Parsons et al., 2018), the csoR-copA-copZ copper resistance operon (Corbett et al., 2011), and zosA for Zn(II) uptake also contributing to oxidative stress resistance (Gaballa and Helmann, 2002). For tolerance to acid stress, in addition to the gadB-gadC operon, for a glutamate decarboxylase and a glutamate/GABS antiporter (Cotter et al., 2005;Liu et al., 2019), all the studied strains presented gadD, for a homologous decarboxylase. This last gene was the only one detected within the SSI-1 (Ryan et al., 2010), while no determinants belonging to the SSI-2 were found.
Several genes associated with monovalent cation/proton antiporters were found in all the strains and included the multiple resistance and pH operon mrpABCDEFG encoding for Na + /H + antiporters and mdrP for a Na + Li + K + /H + antiporter, both responsible for sodium and alkali resistance (Abdel-Motaal et al., 2018;Fang et al., 2018;Xu et al., 2018;Yan et al., 2022).
The crcB gene, associated with the riboswitch responsible for tolerance to the anion fluoride, was also detected in the studied genomes (Baker et al., 2012;Johnston and Strobel, 2020;Chellaiah et al., 2021).
Although all the ST2801 strains mostly exhibited the same genetic pattern associated with environmental adaptations, there were some differences between the plasmid-carrying strains and those without. In more detail, two different Listeria plasmids were carried by 54 strains; in 53 of them pLI100 (Glaser et al., 2001;Kuenne et al., 2010) and the j1776 plasmid Mafuna et al., 2021;Maggio et al., 2021) were detected, while in one of them, the pLM33 (Kuenne et al., 2010) was found in addition to pLI100. Only the strains carrying these plasmids presented cadA and cadC for cadmium resistance (Parsons et al., 2018), the mco gene encoding a multicopper oxidase (MCO), and the additional Na+/H+-K+ antiporter GerN (Southworth et al., 2001;Wu et al., 2020). This finding suggests that the above determinants were carried on these plasmids. We used the nucleotide Basic Local Alignment Search Tool (BLASTn) to verify the alignment between the sequences of cadA, cadC, mco, and gerN as annotated by Prokka in the studied strains, and the ones found in the GenBank file of each plasmid on NCBI. A 100% score for coverage and identity was observed between the cadA gene carried by the strains and the one present in both pLI100 and j1776 plasmids. The same result was obtained for the cadC and gerN carried by the strains compared with the respective genes carried by pLI100. A maximum alignment score was also found comparing the mco gene with the homologous gene carried by the j1776 plasmid. All these findings were in agreement with previous authors reporting these plasmids as carrying determinants for cadmium, arsenic and copper resistance (Canchaya et al., 2010;Parsons et al., 2018;Chiaverini et al., 2021;Lachtara et al., 2021;Schmitz-Esser et al., 2021). Moreover, according to several studies, the specific plasmids found in the strains of this new ST are highly conserved between Lm strains on a large scale and across other STs, suggesting that they provide important advantages for survival in food and FPEs (Canchaya et al., 2010;Parsons et al., 2018;Chiaverini et al., 2021;Guidi et al., 2021;Lachtara et al., 2021;Maggio et al., 2021;Schmitz-Esser et al., 2021).
Moreover, the coexistence of Lm strains carrying and not carrying plasmids, even belonging to the same genetic cluster, indicated horizontal transfer dynamics within the microbial population studied, most probably driven by selective pressure.
Although not presenting a Premature Stop Codon Mutation in the inlA gene nor the SSI-1, both associated with an increased production of biofilm, as reported in previous studies (Keeney et al., 2018;Upham et al., 2019;Mahoney et al., 2022), all the strains carried several genetic markers for biofilm production. In particular luxS, for a S-ribosylhomocysteinase, recO encoding the DNA repair protein RecO and lmo2504 for a cell wall-binding protein were found in all the genomes (Pasquali et al., 2018;Gorski et al., 2022). The lmo0673 gene, encoding an hypothetical protein and reported as marker of biofilm production by Pasquali et al. (2018), was also found. These findings indicated the potential of biofilm formation of these strains. Regardless the effective amount of biofilm produced, if it is formed in niches that are difficult to reach during sanitation procedures, it can represent a persistent source of contamination. This could be another factor responsible for the survival of these ST2801 strains to disinfection.
Concerning the virulence profile, all the ST2801 strains presented the same identical pattern consistent with belonging to a IVb serogroup, albeit atypical and presenting hypervirulence features. Indeed, all the strains carried a full length inlA, encoding for a functional inlA, which is one of the essential virulence factors for Lm to cross the intestinal barrier. Other members of the internalin gene family were present in the strains, including inlC, inlD, inlE, inlF, inlH, inlJ, and inlK, which were known to contribute to the pathogenicity of the pathogen with different mechanisms (Su, 2019;Maćkiw et al., 2021). The aut_IVb, gltA, and gltB genes were also important virulence markers. These genes are involved in invasion and teichoic acid biosynthesis respectively, in serogroup IVb isolates, including IVb-v1, and are absent in other Lm (Moura et al., 2017).
The virulence profile of these ST2801 strains also included the additional LIPI-3, encoding a biosynthetic cluster involved in the production of listeriolysin S (LLS), a haemolytic, and cytotoxic factor conferring a greater virulence to Lm (Su, 2019;de Tavares et al., 2020). This finding was in agreement with previous studies reporting the presence of LIPI-3 in atypical IVb-v1 strains (Moura et al., 2017;de Tavares et al., 2020).
All these findings indicated this new ST, as presenting several determinants for stress response and environmental adaptation, probably responsible for its persistence in the meat-producing plant, together with virulence features. The isolation of a new MLST type with a large pattern of stress adaptation genes and able to persist over a year in the same FPE, also surviving cleaning and sanitation, is considered noteworthy.

CONCLUSION
In this study, we discovered a new MLST allelic profile of Lm designed as ST2801, never reported before, and belonging to the atypical serogroup IVb-v1. Moreover, only one study existing before ours witnessed the isolation of IV-v1 Lm strains from FPEs and their ability to persist over time.
The isolation of this new ST occurred during an intensive environmental sampling plan for Lm performed in a pork meat-producing plant of Central Italy, during production and, for positive surfaces, also after cleaning sanitation. We found widespread circulation of the ST2801 in the studied environment, where it was able to persist over a year and survive sanitation. Based on genomic characterization results, this new ST presents several determinants for stress response, environmental adaptation, and biofilm production, probably responsible for its persistence in the meat-producing plant, together with virulence features.
The isolation of a new ST with a large pattern of stress adaptation genes and the ability to persist in the same FPE for a year is an important contribution to deepen the current knowledge on the uncommon IVb-v1 and in general on genomic diversity of Lm. At the same time, the sampling approach adopted allowed us to provide specific recommendations to the FBO in order to improve the control of the pathogen, minimizing the risk of food contamination. All these findings pointed out how the application of intensive environmental sampling plans, which are specific to each FPE, would ensure improved surveillance and would provide the opportunity to increase knowledge about Lm.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm. nih.gov/bioproject/PRJNA821663.

AUTHOR CONTRIBUTIONS
All authors listed have made a substantial, direct, and intellectual contribution to the work, and approved it for publication.

FUNDING
This study was funded by the Italian Ministry of Health (IZSUM RC0012019).