Founding of the culture collection of antibiotic-resistant strains of zoonotic bacteria in the Russian Federation

Background and Aim: The main purpose of a national bioresource center is to standardize, centralize, preserve, and ensure accessibility of microbial bioresources that accumulate there because of state research programs. The establishment of national bioresource centers for antibiotic-resistant microorganisms allows to solve practical problems in the field of veterinary service, as well as to develop effective chemotherapeutic and disinfectant drugs to overcome the mechanisms of resistance. This study aimed to outline the process of forming a national culture collection of antibiotic-resistant strains of zoonotic bacteria in the Russian Federation using two microbial strains. Materials and Methods: The object of research was isolates of Salmonella spp., Escherichia coli, Enterococcus spp., Campylobacter spp., Listeria monocytogenes, and Staphylococcus spp., all of which were obtained from biomaterials of farm animals, feed samples, bedding, water from livestock buildings, washouts from environmental objects, and food products. The resistance of bacterial isolates was determined using microbiological and molecular-genetic research methods. Results: During monitoring studies, 1489 bacterial isolates were isolated. In total, 408 bacterial isolates were tested for sensitivity to antimicrobial agents, including E. coli (47.6%), Salmonella spp. (30.4%), Enterococcus spp. (11.3%), and Campylobacter spp. (10.8%). For genetic characterization, 95 isolates of Salmonella enterica, E. coli, Campylobacter spp., L. monocytogenes, Staphylococcus spp., Enterococcus spp. were chosen from the research collection, which was formed as part of the monitoring program for antibiotic resistance. Conclusion: Deposited isolates that underwent whole-genome analysis can be used as positive control samples both in the development and use of methods or test systems for the detection of various resistance genes in zoonotic bacteria. In addition, such isolates can also be used for microbiological studies related to determining the sensitivity of microorganisms to antibacterial drugs, for phenotypic studies in the diagnosis of various bacterial infections in animals and birds, and retrospective analysis of strains from numerous collections.


Introduction
The main purpose of a national bioresource center is to standardize, centralize, preserve, and ensure the accessibility of microbial bioresources that accumulate there because of state research programs. Its activity plays an important role in ensuring biosafety and technological independence of the state and is also one of the main elements in the structure of harmonization of the quality control system for raw materials and final products [1]. As part of monitoring studies of bacterial resistance to antibacterial drugs for veterinary use, isolates with genes encoding specific proteins that provide various resistance mechanisms, as well as mutations in genes encoding targets for antibiotic action, are of great interest [2]. In accordance with the data of the World Health Organization (WHO), the most relevant studies are on resistant strains of Salmonella spp., Escherichia coli, Enterococcus spp., Campylobacter spp., and Listeria monocytogenes, Staphylococcus spp., which were the causative agents of 5098 outbreaks of food poisonings in the countries of Eurasia in 2021, the mortality rate reached 0.9% [3]. The development of antibiotic-resistant bacteria occurs due to the presence of genes that code specific proteins programmed to destroy antibiotics or protect target of action, or which provide an active efflux of antibiotics. Moreover, the resistance may be linked with the mutations in chromosomal genes that encode targets of action for antibiotics [4][5][6]. Antibiotic resistance genes are usually associated with the mobile elements of the bacterial genome: plasmids, transposons, integrons, genomic islands, etc. This leads to the possibility of horizontal gene transfer even between taxonomically and ecologically distant microorganisms. Thus, pathogenic microorganisms can obtain resistance genes from the environment. The reverse route is also likely [7][8][9].
Identifying and studying multidrug-resistant zoonotic bacteria circulating in the agro-industrial complex is one of the priorities of the public health veterinary service, and it emphasizes the importance of the veterinary surveillance system in monitoring the resistance of microorganisms to antibacterial drugs [10,11]. The priority here is to disclose scientific knowledge in the field of fundamental research of ecological plasticity and adaptation of microorganisms to the effects of antibacterial drugs for establishing national bioresource centers. This will solve applied problems of the veterinary service and help to develop effective chemotherapeutic and disinfectant drugs that overcome resistance mechanisms.
In this regard, this study aimed to discuss the process of formation of a research collection of antibiotic-resistant strains of zoonotic bacteria in the Russian Federation, using the example of two microbial strains, E. coli and Salmonella enterica.

Ethical approval
This study was conducted using isolated microorganisms and does not require permission from the ethics committee.

Study period and location
The study was conducted from September 2021 to November 2022 at the Division of Biotechnology of Federal State Budgetary Institution, The Russian State Center for Animal Feed and Drug Standardization and Quality, Moscow, Russia.

The object of research
The object of research was isolates of Salmonella spp., E. coli, Enterococcus spp., Campylobacter spp., L. monocytogenes, and Staphylococcus spp. obtained from farm animals, including washouts of the mucous membrane of the nasal cavity, rectum, and contents of cloaca and feces. Samples of feed, food products, litter, water from livestock buildings, and washouts from environmental objects were also investigated. All studied samples were taken on the territory of agricultural enterprises and retail trade facilities of the Russian Federation.
According to the recommendations of the European Committee on Antimicrobial Susceptibility Testing (EUCAST) and the Clinical and Laboratory Standards Institute, the following reference strains of microorganisms were used as controls: Escherichia coli ATCC 25922, Campylobacter jejuni ATCC 33560, and Enterococcus faecalis ATCC 29212.

Microbiological methods
The cultivation of microorganisms was carried out at 35 ± 2-45 ± 2°C for 24-48 h in liquid and solid nutrient media. Microorganisms were identified using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry "Microflex LT" (MALDI-TOF-MS) (Bruker Daltonik Inc., Germany), method of direct application. The results were confirmed by conventional microbiological methods in accordance with the classification system "Bergy's manual 1984-1989", as well as using commercial test systems [12]. The ability of microorganisms to form biofilms in vivo was determined by conventional methods [13]. Morphometric studies were performed with a representative sampling of an accurate frequency of occurrence of ≥90.0% of the field of view of the "Leica DMRB microscope" ("Leica", Germany).
The study of the sensitivity of microorganisms to 28 antibacterial drugs of 12 classes (Supplementary Table-

Molecular-genetic methods, utilized software and resources
DNA isolation was carried out using the DNAsorb-B reagent kits (Federal Scientific Research Institute of Experimental Engineering, Russia) in accordance with the manufacturer's instructions. The DNA library was prepared using the Nextera XT DNA Sample Preparation Kit (Illumina, Inc., USA) according to the manufacturer's instructions. Whole genome sequencing was performed on the MiSeq system (Illumina, Inc) according to the standard operating procedure.
The identification of genetic factors that provide bacterial resistance to various antibiotics was carried out using the ResFinder server [21]. The virulence factor database (VFDB) was used to search for the main virulence factors in bacterial genomes [22]. The search for integrons was conducted using the IntegronFinder v5 software [23] and Salmonella pathogenicity islands -SPIFinder online service (v. 2.0) [24].

Statistical analysis
The results of experimental studies were processed by variational statistics, using the software "Statistical Analysis Software" ("Statistics Solutions", Clearwater, USA), also considering the criterion of reliability (Student's t-test) at a confidence level of 95%.

Results and Discussion
As part of a monitoring program for antibiotic resistance in zoonotic bacteria isolated from biomaterial from farm animals, feed samples, bedding, water from livestock buildings, and washouts from environmental objects, as well as from food samples in the period from 2021 to 2022, 2942 strains of bacteria were identified and deposited; the list of microbial strains in the research collection is presented in Table- In total, the collection fund of microorganisms has 187 authenticated and characterized reference strains, the safety of which is ensured by the use of long-term storage methods: Lyophilization and cryopreservation at low temperatures (−70.0°C).
Based on the examples of two microbial isolates, the deposition process is demonstrated. At the first stage, the indication and identification of microorganisms was carried out, and then the phenotypic and genotypic resistance of the pure bacterial culture to antibacterial drugs was studied. Isolates that showed resistance to several antibiotics of different classes were subjected to deposition for several reasons: Conduct quality control studies for nutrient media, drugs for veterinary use, and different agricultural products; to study the growth properties of microorganisms when using various combinations of antibiotics; to use isolates for the diagnosis of pathogens of infectious animal diseases; to use isolates as a reference for quality control studies in veterinary monitoring of antibiotic resistance; and to provide food safety control using microbiological indicators.

Indication and identification of microorganisms
During monitoring studies, we isolated 1489 bacterial isolates from cows, sheep, pigs, horses, chickens, geese, ducks, and environmental objects ( Figure-1). It has been established that the total number of Campylobacter isolates significantly exceeded the number of Salmonella isolates. The above results confirm the literature data that campylobacteriosis occupies a significant place among acute intestinal infections due to its wide prevalence, the multiplicity of reservoirs, and a trend toward an increase in incidence. Microorganisms of the genus Campylobacter as an etiological factor of intestinal infections are more common than Salmonella. The significance of the problem of the spread of campylobacteriosis is confirmed by the WHO, that has included this infection in the national program of combating infectious diseases in 108 countries, including the Russian Federation [3].
The highest indicator of resistance in Campylobacter spp. was established in relation to antibiotics of the tetracycline group: tetracycline (50.0%), doxycycline (50.0%), as well as to fluoroquinolones (50.0%) (Figure-4). Fluoroquinolone-resistant  Campylobacter is in a high-priority group for developing new antibiotics on the WHO list [3]. Bacteria of Enterococcus spp. were resistant to the fluoroquinolones group (24.66%); tetracyclines (25.45-26.73%); macrolide (29.06-72.75%); polypeptides (91.35%) (Figure-5). In the study of the formation of general patterns of a three-dimensional multilayer heterogeneous structure of biofilms in vitro by multidrug-resistant isolates, it was found that all the studied microorganisms were strong producers of biofilms, and the optical density (OD) of the sample exceeded the ODs of the control by more than 4 times (ODs = 0.528 ± 0.31). High ODs of biofilms (ODs ≥ 0.400) are linked to multidrug resistance [13]. A direct correlation has been established between the ability of bacteria to form biofilms and the profile of resistance to antibacterial drugs: Multidrug-resistant strains were classified as strong biofilm producers -91.07% [28].

Genetic characteristics of isolates of zoonotic bacteria from the research collection of the Federal State Budgetary Institution and The Russian State Center for Animal Feed and Drug Standardization and Quality
For genetic characterization, 95 isolates of S. enterica, Escherichia coli, Campylobacter spp., L. monocytogenes, Staphylococcus spp., and Enterococcus spp. were obtained from the research collection as part of antibiotic resistance monitoring. The selection was carried out on the basis of microbiological data on phenotypic resistance: All 95 isolates were resistant to several antibiotics of various classes. Whole genome sequencing of selected isolates and bioinformatics analysis of the obtained data was then carried out. In addition, 18 were deposited to the All-Russian State Collection of Microorganisms Strains Used in Veterinary Medicine and Livestock.
Genetic characterization aims to evaluate the prevalence of genetic determinants of resistance    among zoonotic bacteria isolated from productive animals and food and feed products. The whole-genome sequencing data make it possible to reveal the presence of resistance genes and establish their localization in mobile elements, the structure and features of gene cassettes. The combination of conventional microbiological methods with molecular-genetic methods allows to obtain the most complete information about various characteristics of bacteria, as well as to confirm the phenotypic properties of isolates. In the example of four isolates (Table-2), we present an algorithm for performing genetic characterization and a data's format for isolates from the research collection of the VGNKI. The quality of sequencing data (FASTQ files) was assessed using the FastQC_0.11.17. Removal of technical sequences and low-quality nucleotides was performed in the Trimmomatic v.0.36. Following parameters were used: NexteraPE-PE.fa: 2:30:10, SLIDINGWINDOW: 4:15. De novo assembly of bacterial genomes was performed using the SPAdes 2.11.1 assembler with sequencing error correction and automatic selection of the k-mer length (21,33,55,77,99). Contigs <500 bp were excluded from further analysis.
An assembly with a combination of the following criteria was chosen as the best: the smallest number of contigs, the largest value of N50, the total length of the contigs and the correct GC content (the proportion of guanine and cytosine bases in the DNA molecule) composition for the analyzed microorganism. The main characteristics of the assembly were obtained using QUAST 4.6.3 and are presented in Table-3.
To determine a bacterial species, we used the search for common k-mers implemented in the KmerFinder. Multilocus typing of the S. enterica samples was performed using MLST at loci aroC, dnaN, hemD, hisD, purE, sucA, thrA. Typing of the E. coli samples at loci adk, fumC, gyrB, icd, mdh, purA, and recA. MLST allele sequence and profile data were obtained from PubMLST.org. Genotyping data (Table-4) confirmed the results that were obtained by conventional microbiological methods (tinctorial, morphological, and biochemical).
The contigs were ordered using the MAUVE v.20150226 by the nucleotide sequence of the: Salmonella enterica strain FSIS1502916 (CP016408.1); E. coli strain AR_452 (CP030331.1). The annotation was performed using the RAST server on the open platform SEED for comparative analysis of genomes [29].
Identification of antibiotic resistance genes was carried out by the ABRicate software [30] using BLASTN and BLASTX against nucleotide and amino acid sequences from ResFinder and NCBI BARRGD [31]. The following criteria were used to analyze contig sequences: >95% identity, >80% coverage. The assortment of the identified genes is largely  consistent with the multidrug resistance phenotype of isolates (Table-5). Point mutations in DNA-gyrase (gyrA) and topoisomerase IV (parC, parE) genes were found in phenotypically ciprofloxacin-and enrofloxacin-resistant isolates. The classification of contigs into chromosomal and plasmid ones was carried out using our own algorithm at Python 2.7 [32]. In isolate S. Infantis VGNKI-2108, a pESI-like mega-plasmid was identified that carried genes for resistance to cephalosporins (CTX-M-14), aminoglycosides (aadA), sulfonamides (sul1), trimetoprim (dfrA), and tetracycline (tetA/R) [33]. The IntegronFinder v5 was used with default settings to search of integrons in bacterial genomes. Each of them was localized on plasmid contigs. The compositions of Class 1 and 2 integrons are presented in Figure-6.
The search for virulence factors was carried out in the VFDB database using the following criteria: >95% identity and >80% coverage. The results are presented in Table-6. The following criteria were used to search for pathogenicity islands in Salmonella spp. using the SPIFinder online service: >95% identity and >60% coverage. The results are presented in Table-7.  Available at www.veterinaryworld.org/Vol.16/July-2023/11.pdf

Conclusion
This study demonstrates the process of establishing a research collection of antibiotic-resistant strains of zoonotic bacteria on the examples of two types of microorganisms. The stages of indication and identification of microorganisms are described, followed by the determination of the phenotypic and genotypic resistance of a pure bacterial culture to antibacterial drugs. The deposited genome-wide characterized isolates can be used as positive control samples both in the development and in the use of methods or test systems to detect various zoonotic bacteria resistance genes. In addition, such isolates can also be used for microbiological studies related to determining the sensitivity of microorganisms to antibacterial drugs and for phenotypic studies in the diagnosis of various bacterial infections in animals and birds. Using well-characterized isolates, a retrospective analysis of strains from different collections becomes also possible: Assessment of antibiotic resistance, pathogenicity, and other properties.