Next Article in Journal
Pea-Wheat Rotation Affects Soil Microbiota Diversity, Community Structure, and Soilborne Pathogens
Next Article in Special Issue
Clinical Metagenomics Is Increasingly Accurate and Affordable to Detect Enteric Bacterial Pathogens in Stool
Previous Article in Journal
Phage Genome Diversity in a Biogas-Producing Microbiome Analyzed by Illumina and Nanopore GridION Sequencing
Previous Article in Special Issue
Correlation between Phenotypic and In Silico Detection of Antimicrobial Resistance in Salmonella enterica in Canada Using Staramr
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Microorganisms
Volume 10
Issue 2
10.3390/microorganisms10020369
microorganisms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Application of a High-Throughput Targeted Sequence AmpliSeq Procedure to Assess the Presence and Variants of Virulence Genes in Salmonella

by
Ruimin Gao
1,2,*,
Hongsheng Huang
1,
Jérémie Hamel
3,
Roger C. Levesque
4,
Lawrence D. Goodridge
5 and
Dele Ogunremi
1,5,*
1
Ottawa Laboratory Fallowfield, Canadian Food Inspection Agency, Ottawa, ON K2J 4S1, Canada
2
Department of Food Science and Agricultural Chemistry, McGill University, Ste Anne de Bellevue, QC H9X 3V9, Canada
3
Merinov, 96 Montée Sandy Beach, Gaspé, Quebec, QC G4X 2V6, Canada
4
Institut de Biologie Intégrative et des Systèmes, Université Laval, Quebec, QC G1V 0A6, Canada
5
Department of Food Science, University of Guelph, Guelph, ON N1G 2W1, Canada
*
Authors to whom correspondence should be addressed.
Microorganisms 2022, 10(2), 369; https://doi.org/10.3390/microorganisms10020369
Submission received: 1 January 2022 / Revised: 24 January 2022 / Accepted: 26 January 2022 / Published: 5 February 2022
(This article belongs to the Special Issue Genomic Tools for the Surveillance and Investigation of Foodborne Disease and Enteric Zoonoses)
Browse Figures
Review Reports Versions Notes

Abstract

:
We have developed a targeted, amplicon-based next-generation sequencing method to detect and analyze 227 virulence genes (VG) of Salmonella (AmpliSeqSalm_227VG) for assessing the pathogenicity potential of Salmonella. The procedure was developed using 80 reference genomes representing 75 epidemiologically-relevant serovars associated with human salmonellosis. We applied the AmpliSeqSalm_227VG assay to (a) 35 previously characterized field strains of Salmonella consisting of serovars commonly incriminated in foodborne illnesses and (b) 34 Salmonella strains with undisclosed serological or virulence attributes, and were able to divide Salmonella VGs into two groups: core VGs and variable VGs. The commonest serovars causing foodborne illnesses such as Enteritidis, Typhimurium, Heidelberg and Newport had a high number of VGs (217–227). In contrast, serovars of subspecies not commonly associated with human illnesses, such as houtenae, arizonae and salame, tended to have fewer VGs (177–195). Variable VGs were not only infrequent but, when present, displayed considerable sequence variation: safC, sseL, sseD, sseE, ssaK and stdB showed the highest variation and were linked to strain pathogenicity. In a chicken infection model, VGs belonging to rfb and sse operons showed differences and were linked with pathogenicity. The high-throughput, targeted NGS-based AmpliSeqSalm_227VG procedure provided previously unknown information about variation in select virulence genes that can now be applied to a much larger population of Salmonella for evaluating pathogenicity of various serovars of Salmonella and for risk assessment of foodborne salmonellosis.

1. Introduction

Among members of the genus Salmonella, 227 virulence genes (VGs) have been described in the scientific literature, as the determinants of the four critical stages of Salmonella pathogenesis, namely: attachment, invasion, intramacrophage survival, and systemic dissemination [1,2]. Observations on the roles of VGs in Salmonella pathogenesis have mostly been obtained from experimental murine infection studies using Salmonella mutants, while only a few publications on pathogenesis have resulted from human NTS infection [3]. Despite some limitations, including result reproducibility [4,5], rodent models, in general, have formed the basis of the current understanding about the mechanism of action. Probably, a more important limitation of the mouse model is that it may not accurately mimic a human Salmonella infection. For instance, the serovar Typhimurium causes a typhoid-like illness in mice, but in the overwhelming number of human cases, it results in gastroenteritis. Many other models including monocytic macrophage cells from humans (U937 and THP-1; [6]) and non-macrophage cell lines from mice (RAW264.7) and humans (Caco-2 and INT-407; [7,8]), amoeba [8] and in vitro gastrointenstinal systems [9] have provided further illumination of Salmonella virulence. Collectively, many VGs have been well-characterized and proven to affect Salmonella behaviour and by extension the epidemiology of the disease. The application of genomics and bioinformatics has continued to shed increasing light on the biology of Salmonella, beyond that which can be obtained using experimental models and thus provide information that can be translated for the control of this rather important foodborne pathogen. To that end, the Salmonella Foodborne Syst-OMICS research database (SalFoS, https://salfos.ibis.ulaval.ca/, accessed on 30 December 2021), which was developed as part of a project on the diagnostics, surveillance, and control of Salmonella, contains a total of 3143 draft genomes of Salmonella (as of December 2021) and is a very useful repository of Salmonella genomes to identify biomarkers for surveillance and food outbreak investigations and to facilitate the development of tools to control salmonellosis [10]. In addition, the bioinformatics analysis of 500 Salmonella genomes completed by Rakov and colleagues (2019) led to the identification of 70 allelic variants of in silico expressed VGs associated with pathogenesis either manifesting as gastrointestinal or invasive disease [11]. Nevertheless, there still remains a gap in our current ability to infer the pathogenic potential or virulence of a strain based on the presence or absence of known virulence genes. Thus, as a first step in an attempt to interpret the clinical, epidemiological or experimental observations following Salmonella infection, the association between the VGs profile and the pathogenic potential of a strain needs to be established. This should help to embark on an overall assessment of the virulence potential of a strain of interest.
Despite the overall genetic similarity of Salmonella enterica organisms, vastly different diseases and distinct immune responses are elicited in humans by the different serovars [12]. To that end, comparative genomic analyses of multiple Salmonella serovars have conclusively identified genetic factors of the bacterium as an underlying cause of, or contributor to, the various disease manifestations caused by the pathogen [12]. For instance, the epidemiologically important NTS serovars linked with a high burden of foodborne Salmonella outbreaks in humans worldwide include Typhimurium, Enteritidis, Heidelberg, and Newport [13,14], yet a vast number of serotypes are rare causes of outbreaks [15] or were never incriminated in foodborne disease outbreaks in jurisdictions with wholesome public health records. Globally, a relatively small proportion of the thousands of known Salmonella serovars is associated with disease outbreaks in humans [16]. In this study, we sought to use a targeted NGS AmpliSeqTM procedure by focusing on the 227 VGs of Salmonella recently described after a comprehensive evaluation of the scientific literature [1]. The assay was designed with the aid of 80 publicly available closed Salmonella genomes as references. We have demonstrated differences in the profiles of 227 VGs between strains and also displayed VGs variants of amplicons for certain field strains.

2. Materials and Methods

2.1. Salmonella Strains and Genomes

Eighty Salmonella reference genomes were downloaded from NCBI Batch Entrez (https://www.ncbi.nlm.nih.gov/, accessed on 15 October 2019; Table 1) and used to construct a primer set for the development of the AmpliSeqSalm_227VG procedure which is a targeted sequencing method for detecting and characterizing virulence genes of Salmonella. The AmpliSeqSalm_227VG procedure was applied to two groups of Salmonella organisms. The first group consisted of 35 strains retrieved from the pathogen inventory of the Ottawa Laboratory Fallowfield, Canadian Food Inspection Agency, and included 10 Salmonella Typhimurium strains of wild bird origin (see below—pathogenicity assessment of 10 S. Typhimurium strains in a chicken model). Phage typing of the organisms was carried out at the Salmonella Reference Centre, National Microbiology Laboratory, Public Health Agency of Canada, Guelph, ON, Canada. The second group was made of 34 Salmonella strains sourced from the SalFoS database (SalFoS; https://salfos.ibis.ulaval.ca/, accessed on 27 January 2020) [10], which was developed as part of the Genome Canada Salmonella Foodborne Syst-OMICS project and was provided by Dr. R. C. Levesque (Laval University, Quebec City, QC, Canada). The serovar designations of the strains belonging to the second group as well as the metadata of the organisms were not provided at the time of testing (blind testing), which allowed us to infer the serovar designation using the AmpliSeqSalm_227VG procedure (see below), and the results were compared with the widely used Salmonella genome serovar designation tool—the Salmonella In Silico Typing Resource (SISTR) [17].

2.2. AmpliSeqSalm_227VG Procedure

The AmpliSeq technology (Figure 1) is a targeted sequencing procedure of amplicons derived from ultra-high multiplexing of a large number of oligonucleotide primers that bind to areas of interest in a genome, which are then amplified and used to construct a DNA library [18,19]. We have employed the AmpliSeq technology to develop an assay which amplifies, determines the nucleotide sequences, and identifies DNA fragments coding for 227 Salmonella VGs. In designing our AmpliSeq assay for all known Salmonella VGs [5], which we term AmpliSeqSalm_227VG, sets of primers able to bind each virulence gene present in all the reference genomes were developed and pooled (ThermoFisher Scientific, South San Francisco, CA, USA). To that end, three principles were followed: (a) to increase coverage for the conserved and more variable VG sequences, multiple rounds of aligning at progressively different locations on each gene were carried out and the generated primers were pooled; (b) majority of the rounds of alignment focused on conserved areas but subsequently homology was relaxed to allow one single nucleotide change at the 5′ in up to half of primers with the aim of increasing coverage of targets with variable sequence; (c) stringent primer specificity filters were applied for the bulk of the design and the specificity parameters were relaxed in later rounds to increase coverage. Using our 80 reference genomes (see above under Salmonella strains and genomes) and 227 VGs, our expected pool primer would have ranged from 454 oligos (if all the VGs were completely conserved among the 80 genomes) to 36,320 oligos (if there was no conservation among all 227 genes, requiring the use of distinct primer pair for each of the VGs among all 80 genomes). Because of bioinformatics scan of target fragments to maximize the use of conserved areas of each VG across all 80 genomes, we identified a total of 1200 unique oligos capable of in silico generation of 16,729 amplicons (instead of a theoretical number of 18,160 amplicons based on 227 VG × 80 genomes); the expected amplicon sizes ranged from 125 bp to 375 bp. The degeneracy of the primer pool resulted in a large number of amplicons which was expected to contain all the variant forms within the 80 reference genomes (Table S1), and ensured that VGs would almost certainly be amplified and identified if present in a strain of Salmonella. Detailed information of the amplicons including size and start/stop locations are provided in Table S2.

2.3. Library Preparation and Sequencing

High-quality DNA was extracted using the Wizard genomic DNA purification kit (Promega, Madison, WI, USA) and assessed with a spectrophotometer (DU 730; Beckman Coulter, Mississauga, ON, Canada) and Qubit 2.0 fluorometer (Life Technologies, Carlsbad, CA, USA). Libraries were constructed with 20 ng of Salmonella DNA each using the Ion AmpliSeq library kit plus protocol (ThermoFisher, Mississauga, ON, Canada) and PCR amplified using the following parameters: primer pool, 20 ng of template DNA and 15 cycles of PCR, consisting of a denaturation step at 99 °C for 15 s and an annealing/extension step at 60 °C for 8 min. Following the PCR step, DNA amplicons were partially enzyme-digested and ligated with barcode adapters, according to manufacturer’s instructions. To enable sequencing of the different samples on a single Ion Torrent chip and save costs, the Ion Xpress™ Barcode kit was used to tag each of the 23 sample libraries. The barcoded libraries were quantified using the Ion library TaqMan quantification kit (Cat. No. 4468802) and were diluted to give a final template concentration of 100 pM. Enriched beads were prepared with the Ion OneTouch 2 System using the Hi-Q View OT2 Kit and sequencing was performed on the Ion Torrent Personal Genome Machine (PGM) machine using Ion PGM 400 Sequencing kit and an Ion 318™ chip. Sequence data were generated by the PGM (ThermoFisher, Waltham, MA, USA) over a period of 7 h.

2.4. Bioinformatic Analysis

Sequence data analysis was automated as a streamlined NGS bioinformatics workflow using the Ion Torrent Suite™ Software v5.8 (ThermoFisher, Mississauga, ON, Canada) (Figure 1). Specifically, we used three plugins namely, FileExporter (version 5.12.0.0), CoverageAnalysis (version 5.12.0.0) and VariantCaller (version 5.12.0.4) with default parameters to generate VGs reads output which were transferred to an Excel document (Microsoft Corporation, Redmond, WA, USA) to illustrate the presence or absence of the 227 VGs and the variants for each Salmonella strain (Table S3). Positive reads indicating the detection of VGs in the AmpliSeq panel was designated “1”, whereas “0” indicated an absence (Table S4). The data were fed as inputs into R (RStudio, Boston, MA, USA), and four different R libraries consisting of ComplexHeatmap, magrittr, grid, and vegan were used to produce a heatmap depicting VG presence or absence (Figure 2).

2.5. Detection of Variants Using Ion Torrent Suite

Customized Ion Torrent variant caller plugin (version 5.12.0.4) was used to detect variants among amplicons of VGs. For each strain, the Ion Torrent suite automatically mapped amplicon sequence to the most closely matching, pre-loaded reference genomes reads which is often the sequence that corresponded to the correct or assigned serovar (See Table 2 legend). Thereafter, sequence variation in each VG was detected by the variant caller by comparing the respective VGs leading to the identification of single nucleotide polymorphism (SNP), multiple nucleotide polymorphism (MNP), deletion (DEL) and insertion (INS) in each amplicon. The variants, serovar names, gene names, variants types and frequency of detected variants were compiled and summarized in an Excel table. Cut-off value for frequency was rigorously set at 90% to exclude spurious variants.

2.6. Pathogenicity Assessment of 10 S. Typhimurium Strains in a Chicken Model

All experiments involving animals were performed in accordance with Canadian Council for Animal Care guidelines and the experimental protocols were approved by the Canadian Food Inspection Agency Animal Care Committee (Approved Protocol ACC#08-06). Ten strains of S. Typhimurium used in our animal infection experiments were isolated from wild birds (sparrow, gull, cormorant and chicken) in Saskatchewan, Canada and were provided by Dr. Musangu Ngeleka of Prairie Diagnostic Services Inc. (Saskatoon, Saskatchewan, Canada). Frozen glycerol stocks of Salmonella were retrieved from storage, inoculated into Brain Heart Infusion (BHI) media and grown at 37 °C in a rotating shaker overnight. Bacteria counts in log phase culture were estimated by determining the optical density value at A600 and enumerated by plating inoculum used for animal infection on BHI plate agar followed by an overnight culture. Two-week-old white leghorn chicks were wing-banded and subcutaneously inoculated with 0.2 mL of Salmonella Typhimurium bacterial suspension containing ~1 × 108 colony forming units. To avoid cross infection, the group of inoculated chicks exposed to the same inoculation dose was housed in a single isolator. Birds were inspected twice daily and weighed daily from the time of inoculation to the end of the experiment; clinical signs (lameness, ruffled feathers, loss of weight) and mortalities were recorded.

2.7. Pulsed-Field Gel Electrophoretic (PFGE) Analysis

The ten strains of Salmonella Typhimurium of wild bird origin selected for in vivo evaluation of intraserovar comparative pathogenesis experiment were subjected to PFGE using the restriction enzymes XbaI and BInI according to the PulseNet Canada method “General Pulsed Field Gel Electrophoresis for Escherichia coli, Salmonella and Shigella”.
(http://www.pulsenetinternational.org/assets/PulseNet/uploads/pfge/PNL05_Ec-Sal-ShigPFGEprotocol.pdf, accessed on 19 June 2019). PFGE patterns were designated using the BioNumerics software v6.01 (Applied Maths, Sint-Martens-Latem, Belgium), after comparison to the PulseNet Canada database at the Public Health Agency of Canada’s National Microbiology Laboratory (Winnipeg, MB, Canada).

2.8. Statistical Analysis and Data Availability

The normal distribution of VGs was assessed using the statistical function in Excel (Microsoft Corporation). Result was judged significant if p < 0.05. Raw sequence reads of amplicons generated using PCR primers directed at 227 virulence genes for each of the 69 Salmonella enterica strains (see Table S4) and determined using the Ion Torrent PGM have been deposited in the NCBI databank under the BioProject Number PRJNA748885 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA748885, accessed on 1 January 2022).

3. Results

3.1. Characteristics of 80 Salmonella Reference Genomes

Eighty Salmonella enterica genomes representing four subspecies and 75 serovars obtained from the GenBank were used as reference genomes to develop the AmpliSeqSalm_227VG tool designed to evaluate the potential pathogenicity of Salmonella based on the presence/absence of 227 VGs (Table 1). Genome sizes ranged from 4,482,117 bp (serovar Pomona) to 5,395,280 bp (serovar India). The distribution of the VGs followed a normal distribution (mean = median = mode = 210 VGs; p = 1.6 × 10−28). In general, strains known to cause foodborne illnesses had an average or higher number of VGs. Serovars Enteritidis, Typhimurium, Heidelberg and Newport which are the commonest causes of foodborne illnesses had a high number of VGs (217–227). In contrast, the less virulent subspecies houtenae, arizonae and salame and the other serovars of subspecies enterica not commonly associated with foodborne illnesses tended to have fewer VGs (177–195).

3.2. AmpliSeqSalm_227VG Procedure for Evaluating Salmonella Virulence Genes

The in silico simulation of the AmpliSeqSalm_227VG on the 80 reference genomes led to the generation of VG targets presented as a heatmap that serves as a confirmation of the presence or absence of the 227 VGs in each of the 80 reference genomes (Figure 2). The results showed that 151 genes were present in all the genomes and are thus described as core80ref VGs (data not shown in Figure 2). In contrast, the Variable80refs VGs consisted of 76 genes, which were absent in at least one of the reference genomes.

3.3. AmpliSeqSalm_227VG to Detect the Target VGs in 35 Salmonella Field Strains

We evaluated the occurrence of the 227 VGs among the first set of 35 field strains of S. enterica, which belonged to 23 different serovars. A representative result of one sequencing run on PGM using a 318 chip is shown in Figure 3. The percentage of VGs amplicon reads mapping to a reference genome ranged from 96–99% (Table 2). Sequence reads from 31 of the 35 strains matched to the respective reference genome and allowed an accurate inference of the serovar of the organism. Two of the remaining four strains belonged to serovar Worthington for which there was no reference genome available at the start of our study. Consequently, the two strains matched to the closest serovar Mbandaka at 96% agreement, which was the least observed in this study (Table 2). These mismatches were highlighted, unsurprisingly, by an enormous number of variants (371 and 372). The two other mismatches involved the pair of serovars Enteritidis and Nitra, which are known to have genetic similarities [20] and surprisingly, serovars Gallinarum and Indiana (Table 2). For all 33 strains with a corresponding reference genome, the number of VSGs automatically deduced using the mapping procedure agreed well with the predicted VGs for the reference genome (Table 2). The AmpliSeqSalm_227VG procedure also detected variants of amplicons between a test strain and its corresponding reference genome. The total number of variants in all the 29 tested strains, which mapped to their respective reference genomes are summarized in Table 3 and variant details including serovars, gene names, nucleotide changes, frequency and location are listed in Table S3. Out of the 227 VGs, 96 had variants. The serovar showing the highest number was S. Montevideo at 277 variants (Table 2 and Table 3). The six most variable genes were safC, sseL, sseD, sseE, ssaK and stdB having 26, 13, 11, 10, 10 and 10 variants, respectively (Table 3). Two of these genes, safC and stdB, are involved in attachment, while the remaining 4, namely sseD, sseE, sseL and ssaK are involved in intramacrophage survival of Salmonella (Table 4).

3.4. Virulence Genes among the Test Panel of 35 Salmonella enterica Strains

Among the 35 field strains tested with the AmpliSeqSalm_227VaG procedure, 190 VGs were found to be present in all strains while 37 VGs (16%) were absent in at least one of the genomes (Variable35OLF VGs = 37). Many of the variable genes belonged to five operons, namely lpfABCDE, stfACDEFG, stiABC, sseIJ and rfbBDFGHIJKMNOUVX. In particular, the rfbP (n = 16), rfbV (n = 17), rfbJ (n = 18) rfbX (n = 18) and STM2231 (n = 19) were not detected at a high frequency. Phylogenetic analysis of the sequence reads showed that out of 35 strains which consisted of 23 serovars, two S. Kentucky, two S. Worthington and 11 S. Typhimurium strains tended to aggregate together, respectively, as shown by the presence of three distinct clusters (Figure 4A; see orange dot, blue triangle and red dot respectively). Thus, a strong serovar clustering occurred based on the presence or absence of the VGs in Salmonella strains. However, one Enteritidis strain (SE864) was noticeably separated from the other (PT8) by Nitra and Pullorum, and this was attributable to the loss of four rfb genes in Enteritidis_SE864 (Figure 4A).

3.5. AmpliSeq Analysis in the Validation Panel of 34 Salmonella Strains

To validate the AmpliSeq procedure, we comprehensively examined another group of Salmonella strains made up of 34 organisms retrieved from the SalFoS database, but provided to the laboratory without the disclosure of any metadata regarding identity or virulence of the organisms (Table 5). Following blind analysis using the AmpliSeqSalm_227VG procedure, the results were analyzed incorporating the metadata especially the serovar designations which were subsequently provided by the SalFos database coordinator. The heatmap displayed the presence/absence of VGs and served as an illustration of the overall pattern as well as the diversity of the VGs distribution among these 34 SalFoS analyzed strains/serovars (Figure 4B). A total of 71 genes or 31% of 227 genes was absent in at least one of the strains (Variable34SalFoS VGs = 71). Because the AmpliSeq results obtained for the 34 SalFoS strains were comparable to those obtained for the initial batch of 35 OLF known strains, we combined the presence/absence profiles for all 69 strains as shown in Table S4. Phylogenetic analysis resulted in the clustering of the serovars (Figure 5). Three different serovars—Typhimurium (n = 12; red dot), Kentucky (n = 3; orange dot), Worthington (n = 2; blue triangle)—tended to aggregate into three distinct clusters, respectively (Figure 5), which agreed with results from Figure 4A. The only two strains from a group of 34 SalFoS (i.e., UL3_Kentucky and UL34_Typhimurium, which shared the same serovars as other strains from the group of 35 OLF known strains, were precisely clustered in their expected groups namely, S. Kentucky and S. Typhimurium, respectively (Figure 5, either red or orange dot with green border). For the different pathogenesis stages, the 155 Core69_strains VGs were made of 25, 54, 32 and 44 genes for the attachment, invasion, intramacrophage survival and systemic dissemination steps, respectively. The 72 Variable69_strains VGs included 24, 19, 7 and 22 genes in the attachment, invasion, intramacrophage survival and systemic dissemination stages, respectively (Table 6).

3.6. Pathogenicity Evaluation for 10 OLF Salmonella Typhimurium Strains in a Chicken Infection Model

Selected from the 69 strains tested for their VGs profiles (Figure 3A and Figure 5), 10 Salmonella Typhimurium strains were used to examine pathogenicity in a chicken model to assess virulence at the intra-serovar level by observing clinical signs and survival (Table 7) [21]. Following subcutaneous inoculation of chickens, birds inoculated with ST07_22495 (No. 8) showed the highest pathogenicity (lowest survival rate of 40% and earliest clinical signs at 2 dpi), whereas chickens inoculated with strain ST07_22792 (No. 7) displayed the least pathogenicity (all survived at 13 days post inoculation) (Figure 6A). The two strains with distinct pathogenicity attributes also showed contrasting properties in terms of host, phage typing and PFGE patterns (Table 7, Figure S1).
A comparison of the full complement of the VGs of the most pathogenic strain (ST07-22495) with the other nine Typhimurium strains revealed varying degree of genetic divergence which correlated with pathogenicity (Figure 6A,B and Table 7 and Table S5). Two strains (ST07_24355, ST07_29216) which were the least divergent from the most pathogenic strain had only 1 synonymous variant, i.e., no amino acid change. These three most pathogenic strains shared an ochre mutation which inserted a stop codon in the sseL, a VG involved in intramacrophage survival of Salmonella, resulting in a truncated open reading frame of 309 amino acids. This mutation was not found in any other strain as the remaining strains including the least pathogenic had a sseL gene coding for the entire 340 amino acids. The three least diverged strains (i.e., with no amino acid change among the VGs—ST07-22495, ST07_24355, ST07_29216) showed the highest pathogenicity as shown by fewer chickens surviving infection and most severe clinical signs (lameness, ruffled feathers, loss of weight). The strain bearing sequences with the next degree of similarity to the most pathogenic strains (i.e., ST07-35522, 9 variants) also showed a degree of pathogenicity next to the three most pathogenic strains as measured by a combination of mortality and severity of clinical signs among survivors, especially lameness. In contrast, the strain most genetically divergent from the reference strain (namely, ST07-22792) had 17 loci with single nucleotide changes that were different from the most pathogenic strain. The most divergent strain was very clearly the least pathogenic and the only one in which none of the inoculated chickens died during the time monitored. Of the remaining 6 strains whose divergence lie between the three fairly uniform strains on the basis of genetics and virulence on one hand (i.e., ST07-22495, ST07_24355, ST07_29216) and the most divergent and least pathogenic strain (ST07-22792) on the other hand, one strain with a genetic variation closer to the three pathogenic strains (ST07-35522, Figure 6B) had 9 variant loci consisting of both synonymous and non-synonymous mutations. This strain was also the fourth most virulent strain after the earlier described three most virulent strains (Figure 6A). The middle group of five strains had 10 variant loci made up of synonymous and non-synonymous changes when compared to ST07_22495 and all the strains had the milder disease (Figure 6A,B and Table S5). One of the strains (ST07-12335) had a unique variation in the fes gene and behaved in a slightly more virulent fashion that the other 4 strains which had a very uniform variation at the 10 loci which did not include the fes gene. However, it was not as virulent as the three most pathogenic strains described above (ST07-22495, ST07_24355, ST07_29216) or the one strain (ST07-35522) that followed them in virulence behavior and genetics. In all, these observations clearly illustrate that genetic differences among the VGs can be linked to the varying degrees of pathogenesis in the serovar Typhimurium which generally tended to be pathogenic.

4. Discussion

The functions and regulations of a restricted number of VGs have been historically studied in a limited number of serovars mainly Typhimurium and Enteritidis [2,3,11]. Consequently, the overall contribution of the VGs to pathogenesis in Salmonella, either in the well-studied serovars or the others remains elusive [22]. This study used a targeted AmpliSeqSalm_227VG procedure to interrogate Salmonella strains for the purpose of determining the presence of up to 227 VGs and their sequence variation. The approach is analogous to carrying out hundreds of PCR reactions for each strain and sequencing each amplicon. Achieving these steps by simultaneous analysis of many strains at the same in a single assay provides an unprecedented power backed by automated bioinformatics analysis. The same objective can be achieved by extracting and analyzing the VGs present in an available, good-quality whole genome sequence however the AmpliSeqSalm_227VG procedure is targeted and therefore faster, with a higher throughput and the bioinformatics analysis is automated and can therefore be carried out by an analyst will minimal bioinformatics skills, given that no command line procedure is required. We compiled our list of 227 genes following a rigorous literature review. Salmonella virulence has been the focus of many studies and many genes are known to contribute to Salmonella pathogenicity [13,23]. However, the vast majority of studies typically analyze a few genes [24,25,26] and occasionally, a few dozen genes [11], to evaluate the ability of the organisms to cause disease. The only study that we are aware of that evaluated a high number of VGs used 219 genes, from a total of 233 of which 14 were absent among 59 Salmonella strains [27]. Because of its extracellular and intracellular phases, it is not surprising that Salmonella has many more virulence genes than other enteric bacteria such a Escherichia coli which may have fewer than 100 VGs [28,29]. Even though Salmonella and E. coli both colonize the same part of the gastrointestinal tract, have similar genome sizes and cause fairly similar disease phenotype, i.e., gastroenteritis, E. coli only has an extracellular phase while Salmonella inhabits both extracellular and intracellular locations and will need more VGs to survive and thrive. Despite our effort to identify as many Salmonella VGs as possible, our list is probably not exhaustive as more virulence genes are observed with new studies. At the onset of our investigation, we established the sequences of the 227 VG among 80 reference genomes which represented a diverse set of epidemiologically important serovars. We tested our hypothesis that the higher the number of VGs present in a Salmonella strain the greater its ability to cause disease in humans and found that representative genomes of serovars responsible for frequent outbreaks, such as Enteritidis, Typhimurium, Heidelberg and Newport, tended to have higher number of VGs.
We were also interested in evaluating the distribution of the VGs among the strains studied and found 151 of 227 VGs or 66.5% (core80refs VGs) to be present in all the 80 reference genomes (Figure 2). The predominance of the core VGs among Salmonella serovars was further supported by the insightful results of tests conducted on 34 Salmonella strains (Figure 4B). The occurrence of a majority of VGs in many Salmonella strains including those that do not normally cause infections in humans suggest that the variability in the pathogenic potential of Salmonella strains may be attributed to the variable segment of the VGs. We demonstrated an association between variable VGs and distinct phases of Salmonella results as shown by preponderant variation among operons and genes playing a role in the attachment (lpf, stf and sti operons), invasion (STM2231), intramacrophage survival (sse operon) and systemic dissemination (rfb operon). The use of 80 reference genomes which included the most commonly reported serovars contaminating food, was adequate for the AmpliSeqSalm_227VG tool to furnish an insight in the distribution of VGs in all tested Salmonella strains. Even when only 18 of the 34 tested strains had matching serovars among the 80 reference genomes (Figure 4B), information obtained on the distribution of VG and variants among all strains was adequate to evaluate their potential pathogenicity.
A chicken infection model further shed light on the effect of the rfb operon in virulence. Among the 10 S. Typhimurium strains used in experimental infections, five strains (ST07_29216, ST07_12335, ST07_22495, ST07_24355, and ST07_35522) possessed all the 227 VGs and a full complement of the rfb operon and demonstrated the highest virulence based on mortality or short incubation period in infected chickens (Figure 6A). The nucleotide variations among three of the five strains were very similar and the remaining two strains showed a different pattern of variation which could be distinguished from the other, five less pathogenic strains (Figure 6B). The less pathogenic strains did not have a full complement of the rfb operon and were missing the rfbP gene (ST07_12345, ST07_7670, and ST07_22792) or the rfbJ gene (ST07_32529) (Figure 4A and Figure 5). One strain, ST07_7666, was missing the rfbP gene but behaved as an outlier combining symptoms associated with low virulence as the other non-pathogenic strains (delayed onset of clinical symptoms at day 3) but with reduced survival similar to the pathogenic strains (Table 7 and Figure 6A). The behaviour of strain ST07-7666 is most likely influenced by other yet to be identified genes. Two additional Typhimurium strains not used for experimental infection either had a full rfb operon (LT2) or was missing rfbJ (UL34). We found that the sseL gene, which codes for a deubiquitinase and contributes to pathogenesis by macrophage killing [30] was similarly truncated among the three most pathogenic Typhimurium strains but had a full-length version among the remaining less pathogenic Typhimurium strains. How a truncated deubiquitinase may cause increased virulence is not clear but may be similar to other observations where a truncation is required for full protein activity [31] or increased enzymatic activity [32].
The least pathogenic S. Typhimurium strain (ST07_22792; nil clinical signs and mortality for 13 days post infection) was not only devoid of a full rfb operon but showed a very marked nucleotide variation and the greatest divergence, i.e., 17 single nucleotide changes, from the most pathogenic strains (ST07_22495; first clinical sign at 2 days post infection and 40% mortality at 13 dpi; Figure 6A and Table 7 and Table S5). Sequence divergence among strains belonging to the Typhimurium serovar mirrors published observation on the serotype Montevideo where marked variations were found in VGs belonging to different clades [33]. Further attributes of the avirulent ST07-22792 highlighted its distinct behaviour, namely: it is the only strain from a cormorant used in the study, had a unique phage type, 1, as well as a unique PFGE pattern for both primary and secondary enzymes (Table 7). Unlike the observations of Kuijpers and colleagues [27], we observed a serovar-based clustering of the Salmonella strains dictated by the VGs detected with using the AmpliSeqSalm_227VG procedure (Figure 5). Our observation suggests that the VGs analysis may have field relevance: nucleotide sequences of the same target of more related organisms will be expected to show a higher degree of similarities compared to less related organism (e.g., different serovars). At the same time, the AmpliSeq results have provided a very high resolution of the differences among members of a serovar that is usually pathogenic based on the presence/absence of specific VG and VG variants. We previously reported a robust experimental chicken infection model used to demonstrate differences in the virulence of strains of Salmonella Typhimurium strains [21], and which was similar to that used by other workers to evaluate virulence for Salmonella Enteritidis [34]. We were able to use the infection model to demonstrate very convincingly that variations among VG translated to different virulence potential in vivo.
The presence of 155 core69strains VGs distributed over the four phases of Salmonella pathogenesis in all the strains tested (Table 6), including Salmonella organisms known to have low virulence and no or rare report of human infections [3], suggests that this set of genes may not contribute markedly to clinical outcomes in a “normal” human population. This perspective is reinforced by the observations that only a small proportion of Salmonella serovars—perhaps as few as 100 serovars—is responsible for a high proportion of foodborne human outbreaks. Yet, our observations from this study suggest that the core VG may well be present in all Salmonella serovars. We suggest that the core genes could confer virulence in an immunologically compromised, vulnerable population who may be susceptible to a large number of serovars [35]. Put differently, it appears that all Salmonella strains carry an inherent but not commonly expressed ability to be pathogenic, in which case the risk management of Salmonella contamination in food among specific demographics, especially consumers that are advanced in age may require a more rigorous oversight than presently done.
In conclusion, we have demonstrated a targeted NGS-based AmpliSeq procedure to simultaneously assess the presence/absence of 227 VGs and variants of amplicons in Salmonella in a high-throughput way. Focus on the variable VG could help infer risk posed by a Salmonella strain including a prediction of the ability of the organism to cause a foodborne outbreak and the severity of such an incidence.

5. Conclusions

Salmonella enterica is a common bacterial contaminant of food, however, only a relatively small number of serovars of the organisms are associated with disease outbreaks in humans, despite the description of a large number of virulence genes in the species. A comprehensive tool to evaluate the virulence potential of Salmonella enterica could provide an objective risk assessment of Salmonella contamination in food. We developed a sequence-based AmpliSeqSalm_227VG tool for the simultaneous interrogation of many Salmonella isolates for the presence/absence of all the virulence genes. We demonstrated the presence of core virulence genes, detectable in all organisms tested, and variable virulence genes which appeared important in conferring the disease-causing status. Using in vitro experiments and bioinformatics tools, this study revealed a number of important variable virulence genes that play critical roles in the attachment (lpf, stf and sti operons), invasion (STM2231), intramacrophage survival (sse operon), and systemic dissemination (rfb operon) of Salmonella.

Supplementary Materials

The following supporting information can be downloaded at: www.mdpi.com/article/10.3390/microorganisms10020369/s1, Figure S1: Pulsed-field gel electrophoresis pattern of XbaI- and BlnI- digests of DNA fragments of 10 Salmonella Typhimurium strains that were used for pathogenicity assessment in a chicken infection model. Restriction fragments are shown following the digestion of DNA of each strain with XbaI (A) and BInI (B), and electrophoresis on 1% agarose gel followed by staining for visualization. PFGE patterns (XAI No. and BNI No.) were designated using BioNumerics software after comparison to the PulseNet Canada database. Table S1: List of all 1200 primers used in the designed AmpliSeqSalm_227VG panel. Table S2: List of 16,729 amplicons including size, start/stop locations of in silico generated amplicons used for the designed AmpliSeqSalm_227VG panel. Table S3: List of variants of virulence genes detected among 29 out of 35 Salmonella enterica strains bioinformatically assigned to the correct serovar. Table S4: Presence and absence of the 227 virulence genes among all the tested 69 Salmonella strains. Table S5: Variants of the virulence genes found in 10 Salmonella Typhimurium strains used for an infection experiment based on No. 8_ST07-22495 as reference.

Author Contributions

Conceptualization, D.O.; Data curation, R.G. and J.H.; Formal analysis, R.G.; Funding acquisition, R.C.L., L.D.G. and D.O.; Investigation, H.H.; Methodology, R.G., H.H., J.H., R.C.L. and L.D.G.; Project administration, H.H., R.C.L., L.D.G. and D.O.; Resources, H.H., R.C.L., L.D.G. and D.O.; Supervision, D.O.; Writing—original draft, R.G.; Writing—review & editing, H.H., J.H., R.C.L., L.D.G. and D.O. All authors have read and agreed to the published version of the manuscript.

Funding

R.G. was financially supported by the Genome Canada project titled “A Syst-OMICS approach to ensuring food safety and reducing the economic burden of salmonellosis” which was led by L.D.G. and R.C.L. DO’s research program has received funding support from Genome Research and Development Initiative of the Government of Canada, Ontario Ministry of Agriculture, Food and Rural Affairs, Canadian Security and Science Program of the Department of National Defence and the Canadian Food Inspection Agency. L.D.G. is the Leung Family Professor of Food Safety in University of Guelph, Guelph.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Gao, R.; Wang, L.R.; Ogunremi, D. Virulence determinants of non-typhoidal salmonellae. In Microorganisms; Blumenberg, M., Shaaban, M., Elgaml, A., Eds.; IntechOpen: London, UK, 2020; pp. 167–192. [Google Scholar]
  2. Fabrega, A.; Vila, J. Salmonella enterica serovar Typhimurium skills to succeed in the host: Virulence and regulation. Clin. Microbiol. Rev. 2013, 26, 308–341. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Cheng, R.A.; Eade, C.R.; Wiedmann, M. Embracing diversity: Differences in virulence mechanisms, disease severity, and host adaptations contribute to the success of nontyphoidal Salmonella as a foodborne pathogen. Front. Microbiol. 2019, 10, 1368. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  4. Bustin, S.A. The reproducibility of biomedical research: Sleepers awake! Biomol. Detect. Quantif. 2014, 2, 35–42. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  5. Velazquez, E.M.; Nguyen, H.; Heasley, K.T.; Saechao, C.H.; Gil, L.M.; Rogers, A.W.L.; Miller, B.M.; Rolston, M.R.; Lopez, C.A.; Litvak, Y.; et al. Endogenous enterobacteriaceae underlie variation in susceptibility to Salmonella infection. Nat. Microbiol. 2019, 4, 1057–1064. [Google Scholar] [CrossRef] [PubMed]
  6. Kortmann, J.; Brubaker, S.W.; Monack, D.M. Inflammasome activation in primary human macrophages Is dependent on flagellin. J. Immunol. 2015, 195, 815–819. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  7. Birhanu, B.T.; Park, N.-H.; Lee, S.-J.; Hossain, M.A.; Park, S.-C. Inhibition of Salmonella Typhimurium adhesion, invasion, and intracellular survival via treatment with methyl gallate alone and in combination with marbofoxacin. Vet. Res. 2018, 49, 101. [Google Scholar] [CrossRef] [Green Version]
  8. Crouse, A.; Schramm, C.; Emond Rheault, J.-G.; Herod, A.; Kerhoas, M.; Rohde, J.; Gruenheid, S.; Kukavica-Ibrulj, I.; Boyle, B.; Greenwood, C.M.T.; et al. Combining whole-genome sequencing and multimodel phenotyping to identify genetic predictors of Salmonella virulence. mSphere 2020, 5, e00293-20. [Google Scholar] [CrossRef]
  9. Cavestri, C.; Savard, P.; Fliss, I.; Emond-Rheault, J.G.; Hamel, J.; Kukavica-Ibrulj, I.; Boyle, B.; Daigle, F.; Malo, D.; Bekal, S.; et al. Salmonella enterica subsp. enterica virulence potential can be linked to higher survival within a dynamic in vitro human gastrointestinal model. Food Microbiol. 2022, 101, 103877. [Google Scholar]
  10. Emond-Rheault, J.G.; Jeukens, J.; Freschi, L.; Kukavica-Ibrulj, I.; Boyle, B.; Dupont, M.J.; Colavecchio, A.; Barrere, V.; Cadieux, B.; Arya, G.; et al. A Syst-OMICS approach to ensuring food safety and reducing the economic burden of Salmonellosis. Front. Microbiol. 2017, 8, 996. [Google Scholar] [CrossRef] [Green Version]
  11. Rakov, A.V.; Mastriani, E.; Liu, S.L.; Schifferli, D.M. Association of Salmonella virulence factor alleles with intestinal and invasive serovars. BMC Genom. 2019, 20, 429. [Google Scholar] [CrossRef]
  12. Gal-Mor, O.; Boyle, E.C.; Grassl, G.A. Same species, different diseases: How and why typhoidal and non-typhoidal Salmonella enterica serovars differ. Front. Microbiol. 2014, 5, 391. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  13. Jajere, S.M. A review of Salmonella enterica with particular focus on the pathogenicity and virulence factors, host specificity and antimicrobial resistance including multidrug resistance. Vet. World 2019, 12, 504–521. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  14. Jackson, B.R.; Griffin, P.M.; Cole, D.; Walsh, K.A.; Chai, S.J. Outbreak-associated Salmonella enterica serotypes and food commodities, United States, 1998–2008. Emerg. Inf. Dis. 2013, 19, 1239–1244. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  15. Chattaway, M.A.; Chandra, N.; Painset, A.; Shah, V.; Lamb, P.; Acheampong, E.; Lo, J.; Patel, B.; Larkin, L.; Sergeant, M.; et al. Genomic approaches used to investigate an atypical outbreak of Salmonella Adjame. Microb. Genom. 2019, 5, e000248. [Google Scholar] [CrossRef]
  16. Vieira, A.; Jensen, A.R.; Pires, S.M.; Karlsmose, S.; Wegener, H.C.; Wong, D.L.F. WHO global foodborne infections network country databank—A resource to link human and non-human sources of Salmonella. In Proceedings of the 12th Symposium of the International Society for Veterinary Epidemiology and Economics, Durban, South Africa, 10–14 August 2009. [Google Scholar]
  17. Yoshida, C.E.; Kruczkiewicz, P.; Laing, C.R.; Lingohr, E.J.; Gannon, V.P.; Nash, J.H.; Taboada, E.N. The Salmonella in silico typing resource (SISTR): An open web-accessible tool for rapidly typing and subtyping draft Salmonella genome assemblies. PLoS ONE 2016, 11, e0147101. [Google Scholar] [CrossRef] [Green Version]
  18. Buzolin, A.L.; Moreira, C.M.; Sacramento, P.R.; Oku, A.Y.; Fornari, A.; Antonio, D.S.M.; Quaio, C.; Baratela, W.R.; Mitne-Neto, M. Development and validation of a variant detection workflow for BRCA1 and BRCA2 genes and its clinical application based on the Ion Torrent technology. Hum. Genom. 2017, 11, 1–9. [Google Scholar] [CrossRef] [Green Version]
  19. Mustafa, A.E.; Faquih, T.; Baz, B.; Kattan, R.; Al-Issa, A.; Tahir, A.I.; Imtiaz, F.; Ramzan, K.; Al-Sayed, M.; Alowain, M.; et al. Validation of Ion TorrentTM inherited disease panel with the PGMTM sequencing platform for rapid and comprehensive mutation detection. Genes 2018, 9, 267. [Google Scholar] [CrossRef] [Green Version]
  20. Franklin, K.; Lingohr, E.J.; Yoshida, C.; Anjum, M.; Bodrossy, L.; Clark, C.G.; Kropinski, A.M.; Karmali, M.A. Rapid genoserotyping tool for classification of Salmonella serovars. J. Clin. Microbiol. 2011, 49, 2954–2965. [Google Scholar] [CrossRef] [Green Version]
  21. Ogunremi, D.; Blais, B.; Huang, H.; Wang, L.; Elmufti, M.; Allain, R.; Hazelwood, J.; Grenier, C.; Amoako, K.; Savic, M.; et al. Draft genome sequences of two strains of Salmonella enterica serovar Typhimurium displaying different virulence in an experimental chicken model. Genome. Announc. 2017, 5, e01526-16. [Google Scholar] [CrossRef] [Green Version]
  22. Wang, M.; Qazi, I.H.; Wang, L.; Zhou, G.; Han, H. Salmonella virulence and immune escape. Microorganisms 2020, 8, 407. [Google Scholar] [CrossRef] [Green Version]
  23. Fierer, J.; Guiney, D.G. Diverse virulence traits underlying different clinical outcomes of Salmonella infection. J. Clin. Investig. 2001, 107, 775–780. [Google Scholar] [CrossRef] [Green Version]
  24. Gonzalez, C.D.; Espinoza, L.L.; Huamán, D.C.; Rodríguez, G.S.; Alcántara, R.R.; Hernández, L.M. Evaluation of virulence factors of strains of Salmonella Typhimurium isolated from diseased and healthy guinea pigs (Cavia porcellus). Rev. Investig. Vet. Peru 2021, 32, e21331. [Google Scholar]
  25. Higgins, D.; Mukherjee, N.; Pal, C.; Sulaiman, I.M.; Jiang, Y.; Hanna, S.; Dunn, J.R.; Karmaus, W.; Banerjee, P. Association of virulence and antibiotic resistance in Salmonella—Statistical and computational insights into a selected set of clinical isolates. Microorganisms 2020, 8, 1465. [Google Scholar] [CrossRef] [PubMed]
  26. van Asten, A.J.A.M.; van Dijk, J.E. Distribution of “classic” virulence factors among Salmonella spp. FEMS Immunol. Med. Microbiol. 2005, 44, 251–259. [Google Scholar] [CrossRef] [PubMed]
  27. Kuijpers, A.F.A.; Bonacic Marinovic, A.A.; Wijnands, L.M.; Delfgou-van Asch, E.H.M.; van Hoek, A.H.A.M.; Franz, E.; Pielaat, A. Phenotypic prediction: Linking in vitro virulence to the Genomics of 59 Salmonella enterica strains. Front. Microbiol. 2019, 9, 3182. [Google Scholar] [CrossRef] [PubMed]
  28. Chapman, T.A.; Wu, X.-Y.; Barchia, I.; Bettelheim, K.A.; Driesen, S.; Trott, D.; Wilson, M.; Chin, J.J.-C. Comparison of virulence gene profiles of Escherichia coli strains isolated from healthy and diarrheic swine. Appl. Environ. Microbiol. 2006, 72, 4782–4795. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  29. Government of Canada. Review of The State Of Knowledge on Verotoxigenic Escherichia coli and the Role of Whole Genome Sequencing as an Emerging Technology Supporting Regulatory Food Safety in Canada; Gill, A., Ed.; Health Canada: Ottawa, ON, Canada, 2019; 66p, Catalogue No A104-184/2019E-PDF; ISBN 978-0-660-33244-4.
  30. Rytkönen, A.; Poh, J.; Garmendia, J.; Boyle, C.; Thompson, A.; Liu, M.; Freemont, P.; Hinton, J.C.D.; Holden, D.W. SseL, a Salmonella deubiquitinase required for macrophage killing and virulence. Proc. Natl. Acad. Sci. USA 2007, 104, 3502–3507. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  31. Towne, J.E.; Renshaw, B.R.; Douangpanya, J.; Lipsky, B.P.; Shen, M.; Gabel, C.A.; Sims, J.E. Interleukin-36 (IL-36) ligands require processing for full agonist (IL-36α, IL-36β, and IL-36γ) or antagonist (IL-36Ra) activity. J. Biol. Chem. 2011, 286, 42594–42602. [Google Scholar] [CrossRef] [Green Version]
  32. Jin, N.; Yin, X.; Gu, J.; Zhang, X.; Shi, J.; Qian, W.; Ji, Y.; Cao, M.; Gu, X.; Ding, F.; et al. Truncation and activation of dual specificity tyrosine phosphorylation-regulated kinase 1A by calpain I—A molecular mechanism linked to tau pathology in Alzheimer disease. J. Biol. Chem. 2015, 290, 15219–15237. [Google Scholar] [CrossRef] [Green Version]
  33. Nguyen, S.V.; Harhay, D.M.; Bono, J.L.; Smith, T.P.L.; Fields, P.I.; Dinsmore, B.A.; Santovenia, M.; Wang, R.; Bosilevac, J.M.; Harhay, G.P. Comparative genomics of Salmonella enterica serovar Montevideo reveals lineage-specific gene differences that may influence ecological niche association. Microb. Genom. 2018, 4, e000202. [Google Scholar] [CrossRef]
  34. Cui, L.; Wang, X.; Zhao, Y.; Peng, Z.; Gan, P.; Cao, Z.; Feng, J.; Zhang, F.; Guo, K.; Wu, M.; et al. Virulence comparison of Salmonella enterica subsp. enterica isolates from chicken and whole genome analysis. Microorganisms 2021, 9, 2239. [Google Scholar] [PubMed]
  35. Gurtler, J.B.; Doyle, M.; Kornacki, J.L. Foodborne Pathogens: Virulence Factors and Host Susceptibility; Springer: Cham, Switzerland, 2017. [Google Scholar] [CrossRef]
Microorganisms 10 00369 g001 550
Figure 1. Workflow for the adaptation of AmpliSeq technology to develop and construct the AmpliSeqSalm_227VG panel. VGs: virulence genes.
Figure 1. Workflow for the adaptation of AmpliSeq technology to develop and construct the AmpliSeqSalm_227VG panel. VGs: virulence genes.
Microorganisms 10 00369 g001
Microorganisms 10 00369 g002 550
Figure 2. Distribution of 227 virulence genes in 80 reference Salmonella genomes used in designing the AmpliSeqSalm_227VG panel. Red and blue colors represent gene presence and absence, respectively. The x axis contains the list of 7680refs variable genes which are absent in at least one genome and the y axis shows the 80 Salmonella enterica reference genomes.
Figure 2. Distribution of 227 virulence genes in 80 reference Salmonella genomes used in designing the AmpliSeqSalm_227VG panel. Red and blue colors represent gene presence and absence, respectively. The x axis contains the list of 7680refs variable genes which are absent in at least one genome and the y axis shows the 80 Salmonella enterica reference genomes.
Microorganisms 10 00369 g002
Microorganisms 10 00369 g003 550
Figure 3. Analysis metrics for Ion Torrent Personal Genome Machine (PGM) amplicon sequencing for virulence genes present in a pool of 23 Salmonella strains. (A) Heatmap showing an optimal Ion Sphere Particle (ISP) density loading of 81% on the surface of PGM 318 chip. Red and blue colors represented the highest and lowest loading density areas, respectively. (B) A total of 6.9 million reads was obtained from the sequencing run and 82% reads were clonal. (C) Read length distribution of this representative run, with mean, median, mode length of 247 bp, 265 bp and 298 bp, respectively.
Figure 3. Analysis metrics for Ion Torrent Personal Genome Machine (PGM) amplicon sequencing for virulence genes present in a pool of 23 Salmonella strains. (A) Heatmap showing an optimal Ion Sphere Particle (ISP) density loading of 81% on the surface of PGM 318 chip. Red and blue colors represented the highest and lowest loading density areas, respectively. (B) A total of 6.9 million reads was obtained from the sequencing run and 82% reads were clonal. (C) Read length distribution of this representative run, with mean, median, mode length of 247 bp, 265 bp and 298 bp, respectively.
Microorganisms 10 00369 g003
Microorganisms 10 00369 g004 550
Figure 4. Heatmap illustration showing presence/absence of variable virulence genes (VGs) of Salmonella. (A) The 35 OLF strains representing 23 serovars. (B) The 34 SalFos strains. The x axis contains the list of variable VGs out of 227 used to design primers for the AmpliSeqSalm_227VG tool. The y axis depicts the Salmonella strains and their identities. Red and blue colors represent presence and absence of a VG, respectively. The following serovars are highlighted because more than one strain was used in this study: Typhimurium ●, Enteritidis ▲, Worthington ▲, Kentucky ●. The remainder of the 227 VGs which were found in all strains (core VGs) are not shown in this illustration.
Figure 4. Heatmap illustration showing presence/absence of variable virulence genes (VGs) of Salmonella. (A) The 35 OLF strains representing 23 serovars. (B) The 34 SalFos strains. The x axis contains the list of variable VGs out of 227 used to design primers for the AmpliSeqSalm_227VG tool. The y axis depicts the Salmonella strains and their identities. Red and blue colors represent presence and absence of a VG, respectively. The following serovars are highlighted because more than one strain was used in this study: Typhimurium ●, Enteritidis ▲, Worthington ▲, Kentucky ●. The remainder of the 227 VGs which were found in all strains (core VGs) are not shown in this illustration.
Microorganisms 10 00369 g004
Microorganisms 10 00369 g005 550
Figure 5. Heatmap illustration of reads presence/absence for 72 variable virulence genes (variable69strains VGs) from a total of 69 Salmonella strains (35 OLF known and 34 SalFos strains). The x axis shows the 72 variable69strainsVGs while the y axis represents all the tested 69 Salmonella strains. Red and blue colors represent presence and absence of a VG, respectively. For annotation purpose: Typhimurium ●, Enteritidis ▲, Worthington ▲, Kentucky ●. The illustration is a pool of the data from Figure 4A,B to show the clustering of the Salmonella serovars based on analysis of VGs sequences.
Figure 5. Heatmap illustration of reads presence/absence for 72 variable virulence genes (variable69strains VGs) from a total of 69 Salmonella strains (35 OLF known and 34 SalFos strains). The x axis shows the 72 variable69strainsVGs while the y axis represents all the tested 69 Salmonella strains. Red and blue colors represent presence and absence of a VG, respectively. For annotation purpose: Typhimurium ●, Enteritidis ▲, Worthington ▲, Kentucky ●. The illustration is a pool of the data from Figure 4A,B to show the clustering of the Salmonella serovars based on analysis of VGs sequences.
Microorganisms 10 00369 g005
Microorganisms 10 00369 g006 550
Figure 6. Survival rate plot of chickens inoculated with Salmonella Typhimurium strains and virulence gene variants detected among the strains, n = 10. (A) Survival rate plot of chicken inoculated with 10 S. Typhimurium strains. The sample name is labeled with ID: strain name, day post-infection when clinical signs were first observed, for instance, the number 1 sample was labeled as “1: ST07-7666, 3d”; the sample name was highlighted with the same color as its survival plot line color. The most virulent strain was ST07-22495 No. 8 strain (red color line with red short horizontal line highlighted) whereas the least virulent strain was No. 7 ST07-22792 strain (green color line with green square highlighted). (B) Detected variants among the 10 S. Typhimurium strains used for the infection experiment, using No. 8_ST07-22495 as index. For easy identity, the same color was used for strain names as well in panel (A). The variants causing non-synonymous amino acid changes are highlighted in red.
Figure 6. Survival rate plot of chickens inoculated with Salmonella Typhimurium strains and virulence gene variants detected among the strains, n = 10. (A) Survival rate plot of chicken inoculated with 10 S. Typhimurium strains. The sample name is labeled with ID: strain name, day post-infection when clinical signs were first observed, for instance, the number 1 sample was labeled as “1: ST07-7666, 3d”; the sample name was highlighted with the same color as its survival plot line color. The most virulent strain was ST07-22495 No. 8 strain (red color line with red short horizontal line highlighted) whereas the least virulent strain was No. 7 ST07-22792 strain (green color line with green square highlighted). (B) Detected variants among the 10 S. Typhimurium strains used for the infection experiment, using No. 8_ST07-22495 as index. For easy identity, the same color was used for strain names as well in panel (A). The variants causing non-synonymous amino acid changes are highlighted in red.
Microorganisms 10 00369 g006
Table
Table 1. Salmonella reference genomes used in the design of AmpliSeqSalm_227VG panel and their distribution of 227 known virulence genes.
Table 1. Salmonella reference genomes used in the design of AmpliSeqSalm_227VG panel and their distribution of 227 known virulence genes.
IDSubspecies/Serovara StrainSize
(bp)		b VGs (Max = 227)
CP006602.14,[5],12:i:- 08-1736 4,822,189227
CP007532.1Abaetetuba ATCC 356404,547,600199
CP007534.1Abony 00144,737,447225
CP001138.1Agona SL4834,798,660222
CP019177.1Albany ATCC 519604,805,448216
CP007531.1Anatum ATCC BAA-15924,706,101211
CP019116.1AntsalovaS01-05114,648,086196
CP019403.1Apapa SA20060561 4,801,658206
CP006693.1arizonae 62:z36:-RKS29834,574,846177
CP006053.1Bareilly CFSAN0001894,730,612210
CP019405.1Bergen ST350 4,801,835207
CP019406.1Blegdam S-1824 4,693,979222
CP019407.1Borreze SA20041063 4,777,558223
HF969015.2Bovismorbificans31144,680,283216
CP022490.1Braenderup SA20026289 4,734,880209
CP007533.1Bredeney CFSAN0010804,603,849211
CP012833.1Cerro CFSAN0015884,651,400200
CP019178.1ChesterATCC 119974,660,922214
AE017220.1CholeraesuisSC-B674,755,700210
CP019408.1Crossness1422-74 4,847,468227
CP006055.1Cubana CFSAN0020504,977,480209
CP022494.1Derby SA20035215 4,850,334223
CP019409.1Djakarta S-1087 4,668,861195
CP001144.1Dublin CT_020218534,842,908222
NC_011294.1EnteritidisP125109 4,685,848222
NC_011274.1Gallinarum 287/91 4,658,697220
CP024165.1Gaminara CFSAN070644 4,801,841194
CP017719.1Hayindogo CFSAN050752 4,765,719210
CP001120.1Heidelberg SL4764,888,768225
CP019410.1Hillingdon N1529-D3 4,618,056194
CM001471.1houtenaeATCC BAA-1581 4,672,567175
CP022503.1HvittingfossSA200149814,940,239210
CP022015.1India SA20085604 5,395,280192
CP022450.1Indiana D90 4,779,514213
LN649235.1InfantisNA4,710,675209
CP019181.1InvernessATCC 10720 4,865,682211
LT571437.1Java NCTC5706 4,756,780221
CP004027.1Javiana CFSAN0019924,634,161207
CP019411.1JohannesburgST203 4,651,794196
CP022500.1Kentucky SA20030505 4,782,363216
CP019412.1KoessenS-1501 4,566,169208
CP019413.1Krefeld SA20030536 4,942,273216
CP022117.1MacclesfieldST-1643 4,822,139202
CP019414.1ManchesterST278 4,532,753204
CP022497.1Manhattan SA20084699 4,732,484216
CP019183.1Mbandaka ATCC 519584,905,181198
CP022489.1Mbandaka SA20026234 4,796,292207
CP019184.1MinnesotaATCC 492844,592,393195
CP007530.1Montevideo 507440-204,694,375195
CP019415.1MoscowS-18434,690,402222
CP019201.1Muenster4204756014203
NC_011080.1Newport SL2544,827,641217
CP019404.1Newport SA20113257 4,849,139218
CP019416.1Nitra S-1687 4,691,807222
CP022034.1OnderstepoortSA200600864,774,926194
CP022116.1Ouakam SA20034636 4,874,915212
CP012346.1Panama ATCC 73784,555,576208
CP000026.1Paratyphi A ATCC 91504,585,229210
CP000886.1Paratyphi B SPB74,858,887227
CP000857.1Paratyphi C RKS45944,833,080210
CP019186.1Pomona ATCC 107294,482,117196
CP019189.1Poona ATCC BAA-1673 4,876,720192
CP012347.1Pullorum ATCC 91204,694,842221
CP019192.1RubislawATCC 10717 4,572,929198
CP022491.1Saintpaul SA20031783 4,775,303226
CP022139.1salamae 55:k:z39NA4,859,044187
CP001127.1Schwarzengrund CVM196334,709,075210
LN868943.1Senftenberg NCTC103843,746,274NA
CP012349.1Sloterdijk ATCC 157914,817,791218
CP017723.1Stanleyville SARB61 4,888,463226
CP007505.1TennesseeTXSC_TXSC08-194,864,410207
CP006717.1ThompsonRM68364,707,648210
NC_003198.1Typhi CT184,809,037207
HF937208.1Typhimurium DT1044,933,631227
NC_003197.2Typhimurium LT24,857,450227
CP006048.1Typhimurium var. 5- CFSAN0019214,859,931227
CP019417.1WandsworthSA200920954,916,040211
CP022138.1WaycrossSA200416084,812,886198
LN890520.1WeltevredenNA5,129,845211
CP019418.1Yovokome S-1850 4,640,929215
a NA: not available; b VGs = virulence genes.
Table
Table 2. Results of AmpliSeq analysis performed on strains of Salmonella with known serovar designation.
Table 2. Results of AmpliSeq analysis performed on strains of Salmonella with known serovar designation.
Serovars/StrainsPredicted No.Detected No.Mapped ReadsTotal Readsa On TargetAssigned
Serovar		b Variant
Number
Agona 07D669 3-5222224331,304337,2610.98Agona1
Enteritidis SE864222221306,812312,3390.98Enteritidis8
Enteritidis PT8222222313,320319,1430.98Enteritidis7
Gallinarum SA20103326220226285,325290,2750.98Indiana28
Heidelberg D1005841225224298,529305,7420.98Heidelberg0
Indiana 11SU006 5-3213217330,510335,6800.98Indiana1
Infantis 39209226276,145282,1070.98Infantis4
Javiana OLF16127-1207211292,639298,8290.98Javiana47
Kentucky D0827783216215278,679283,0950.98Kentucky2
Mbandaka 06D1004 8-11207208257,337264,1290.97Mbandaka1
Montevideo 08D107 18-3195196310,953317,6270.98Montevideo277
Newport S443211216254,846258,5660.99Newport21
Newport S447218219301,846308,7100.98Newport23
Nitra SA20103329222222260,241265,4580.98Enteritidis390
Paratyphi A SA20114389210214313,073318,5530.98Paratyphi A5
Pullorum ATCC19945221222283,317288,6000.98Pullorum2
S.Kentucky216217273,264279,5340.98Kentucky1
S.Typhimurium 07-12345227226248,137252,3510.98Typhimurium11
S.Typhimurium07-12335227227257,763264,1950.98Typhimurium11
S.Typhimurium07-22495227227333,463342,4270.97Typhimurium15
S.Typhimurium07-22792227227292,454300,7210.97Typhimurium5
S.Typhimurium07-24355227227344,674355,5650.97Typhimurium14
S.Typhimurium07-29216227227258,620263,5170.98Typhimurium14
S.Typhimurium07-32529227226263,450271,4110.97Typhimurium11
S.Typhimurium07-35522227227305,560313,9800.97Typhimurium11
S.Typhimurium07-7666227226289,621298,3970.97Typhimurium11
S.Typhimurium07-7670227226294,378302,4600.97Typhimurium11
S.Worthington_43209209303,102314,9340.96Mbandaka371
S.Worthington_70209209304,260317,6650.96Mbandaka372
Schwarzengrund ADRI40210212290,699295,3690.98Schwarzengrund3
Senftenberg OLF13070217217304,695313,4910.97Senftenberg-
Tennesse D1019361207209291,490298,9760.97Tennesse1
Thompson 08D021 13-6210213292,588297,5760.98Thompson0
Typhimurium LT2 SA1930227227278,222284,9380.98Typhimurium0
Weltevreden OLF18006-1211213280,421286,7780.98Weltevreden0
a On target: the proportion of sequenced amplicon reads mapping to the target genes in the reference genome. b Two rounds of reads mapping were performed for the amplicons of each strain. First round of mapping was an automated real time alignment of sequence reads to the preloaded 80 Salmonella reference genomes resulting in serovar identification from the highest reads match. The identified serovar is depicted as the “Assigned Serovar” in this table. Subsequently, using the “Assigned Serovar” genome as reference, the second round amplicon reads mapping was carried out to find the number of variants present between the tested strain (listed in column named “Serovars/strains”) and its respective reference genome. Except for the serovar Montevideo, high variant number was predictive of wrong serotype designation and could be used as a means of assessing suspicious results. For instance, during the 1st round mapping, sequence reads from Worthington, Nitra and Gallinarum incorrectly mapped to the serovars Mbandaka, Enteritidis and Indiana, respectively and showed relatively high to very high number of variants.
Table
Table 3. Number of virulence gene variants present in each of 29 Salmonella enterica strains identified with the correct serovar designation as a result of automated, real time mapping of sequence reads to preloaded reference genomes by the Ion Torrent sequencer.
Table 3. Number of virulence gene variants present in each of 29 Salmonella enterica strains identified with the correct serovar designation as a result of automated, real time mapping of sequence reads to preloaded reference genomes by the Ion Torrent sequencer.
GeneNo.GeneNo.GeneNo.GeneNo.GeneNo.GeneNo.
safC26pagC5fhuE2ymdA2fhuC1sapB1
sseL13StiC5invA2barA1fimD1sapC1
sseD11sptP4iroB2BcfE1fimF1sicA1
ssaE10invJ3lpfA2BcfF1hilC1sifA1
ssaK10mgtA3lpfC2crp1hofB1sipC1
stdB10ompR3pagD2csgB1orgB1smpB1
cirA9rfbN3rfaY2csgC1pipC1ssaJ1
csgE9rfbV3rpoE2csgD1rfaB1ssaM1
fimH9sipA3ssaI2csgG1rfaP1SseA1
hilD9StbC3ssaO2entA1rfbM1stfF1
iacP9fepD2ssaT2entC1rfbX1stfG1
rfbG6fepG2sseI2fepE1rpoS1sthA1
csgF5fes2sspH22fhuB1sapA1ybdB1
SafC26pagC5fhuE2ymdA2fhuC1sapB1
sseL13stiC5invA2barA1fimD1sapC1
sseD11sptP4iroB2bcfE1fimF1sicA1
Table
Table 4. Virulence genes involved in the four known steps of Salmonella pathogenesis.
Table 4. Virulence genes involved in the four known steps of Salmonella pathogenesis.
VG TypeNo.Gene Names
Attachment49barA, bcfA, bcfB, bcfC, bcfD, bcfE, bcfF, bcfG, csgA, csgB, csgC, csgD,
csgE, csgF, csgG, fimC, fimD, fimF, fimH, FimI, fimW, FimY, FimZ, hofB,
hofC, lpfA, lpfB, lpfC, lpfD, lpfE, ppdD, safC, stbC, stdB, stfA, stfC, stfD,
stfE, stfF, stfG, sthA, sthB, sthD, stiA, stiB, stiC, STM0551, STM4595, yibR
Invasion/intracellular survival73crp, hilA, hilC, hilD, hnr, iacP, iagB, invA, invB, invC, invE, invF, invG, invH, invI,
invJ, ompR, orgA, orgB, orgC, pagC, pagD, pagP, pipB, pipB2, pipC, pipD, prc,
prgH, prgI, prgJ, prgK, rpoE, rpoS, sapA, sapB, sapC, sapD, sapF, sicA, sicP,
sifA, sipA, sipB, sipC, sipD, slrP, slyA, sodA, sodB, sodC, sopA, sopB, sopD,
sopD2sopE2, spaO, spaP, spaQ, spaR, spaS, sprB, sptP, sspH2,
STM1056 (msgA), STM2231, STM2244, STM4315 (rtsA), yejA, yejB, yejE, yejF, ymdA
Intramacrophage survival39csrA, hfq, mgtA, mgtB, mgtC, smpB, ssaB, ssaC, ssaD, ssaE, ssaG, ssaH, ssaI
ssaJ, ssaK, ssaL, ssaM, ssaN, ssaO, ssaP, ssaQ, ssaR, ssaS, ssaT, ssaU,
ssaV, sscA, sscB, sseA, sseB, sseC, sseD, sseE, sseF, sseG, sseI, sseJ,
sseL, STM1410 (ssaX)
Dissemination66cirA, entA, entB, entC, entD, entE, entF, fepA, fepB, fepC, fepD, fepE, fepG,
fes, fhuA, fhuB, fhuC, fhuD, fhuE, foxA, fruR, fUR, iroB, iroC, iroD, iroE, iroN,
msbA, msgA, rfaB, rfaC, rfaD, rfaF, rfaG, rfaH, rfaI, rfaJ, rfaK, rfaL, rfaP, rfaQ, rfaY,
rfaZ, rfbB, rfbD, rfbF, rfbG, rfbH, rfbI, rfbJ, rfbK, rfbM, rfbN, rfbP, rfbU, rfbV, rfbX,
rfc, STM0719, udhA, wzxC, wzxE, wzzB, wzzE, ybdA, ybdB
Total227
Where virulence genes are also known by alternate names, these are indicated in parenthesis, namely: STM1056 (msgA), STM4315 (rtsA), and STM1410 (ssaX).
Table
Table 5. Validation of AmpliSeqSalm_227VG based on blind assessment of virulence and serovar designation for 34 Salmonella strains.
Table 5. Validation of AmpliSeqSalm_227VG based on blind assessment of virulence and serovar designation for 34 Salmonella strains.
Strain IDSalFoS IDSerovara VGs No.
UofLaval23S158NA196
UofLaval4S78511:b:e,n,x182
UofLaval10S77458:d:z6183
UofLaval20S361Adelaide199
UofLaval32S1393Apapa214
UofLaval7S603Bareilly216
UofLaval16S333Berta223
UofLaval29S256Bovismorbificans218
UofLaval12S209Braenderup209
UofLaval33S325Bredeney213
UofLaval14S364Cerro200
UofLaval30S1426Cubana222
UofLaval5S229Derby218
UofLaval31S718Derby223
UofLaval9S354Gaminara198
UofLaval8S219Hadar215
UofLaval24S1603Havana208
UofLaval1S1288Javiana206
UofLaval17S551Johannesburg220
UofLaval3S246Kentucky216
UofLaval19S267Kiambu211
UofLaval25S334Kottbus216
UofLaval6S494Lockleaze192
UofLaval22S238Mbandaka201
UofLaval13S1395Mississippi207
UofLaval2S1925Muenchen218
UofLaval15S206Muenchen215
UofLaval27S317Ohio203
UofLaval28S307Poona197
UofLaval21S212Stanley223
UofLaval18S761Stanleyville218
UofLaval26S193Thompson210
UofLaval34S164Typhimurium226
UofLaval11S277Uganda209
a VG = Virulence gene.
Table
Table 6. Relative distribution of core and variable virulence genes and their participation in the different stages of pathogenesis in Salmonella enterica (n = 69 strains).
Table 6. Relative distribution of core and variable virulence genes and their participation in the different stages of pathogenesis in Salmonella enterica (n = 69 strains).
Pathogenesis StageVirulence Genes
Core69strainsVariable69strainsTotal
Attachment252449
Invasion541973
Intramacrophage survival32739
Systemic dissemination442266
Total15572227
Table
Table 7. Pathogenicity of 10 Salmonella Typhimurium strains in a chicken infection model and number of variant virulence genes detected with the AmpliSeq tool using Number 8 strain ST07-22495 as index.
Table 7. Pathogenicity of 10 Salmonella Typhimurium strains in a chicken infection model and number of variant virulence genes detected with the AmpliSeq tool using Number 8 strain ST07-22495 as index.
ID/GroupHostAntigensPhage TypingPFGE PatternsSurvival (7d)Survival (>13d)1st Signb Variant Number
1: ST07-7666Sparrow4,5:12160XAI 0280/BNI 007150%40%3d10
2: ST07-7670Sparrow4,5:12160XAI 0280/BNI 007190%70%3d10
3: ST07-12345Sparrow4,5:12160XAI 0280/BNI 0071100%90%3d10
4: ST07-32529Sparrow4,5:12160XAI 0280/BNI 029660%a ND2d10
5: ST07-12335Sparrow4,5:12146XAI 0021/BNI 009670%a ND2d10
6: ST07-35522Sparrow4,5:1240XAI 0075/BNI 029760%a ND2d9
7: ST07-22792Cormorant4,5:121XAI 0654/BNI 0295100%100%13d17
8: ST07-22495Gull4,5:1241XAI 0269/BNI 008160%40%2dIndex
9: ST07-24355Gull4,5:12125XAI 0269/BNI 008170%40%2d1
10: ST07-29216Chicken4,5:12126XAI 0269/BNI 0081100%40%3d1
a ND: not determined. b AmpliSeq tool results: variant number represents each instance of sequence variation among the virulence genes as detected following the mapping of the amplicon reads to the Salmonella Typhimurium reference genome.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Gao, R.; Huang, H.; Hamel, J.; Levesque, R.C.; Goodridge, L.D.; Ogunremi, D. Application of a High-Throughput Targeted Sequence AmpliSeq Procedure to Assess the Presence and Variants of Virulence Genes in Salmonella. Microorganisms 2022, 10, 369. https://doi.org/10.3390/microorganisms10020369

AMA Style

Gao R, Huang H, Hamel J, Levesque RC, Goodridge LD, Ogunremi D. Application of a High-Throughput Targeted Sequence AmpliSeq Procedure to Assess the Presence and Variants of Virulence Genes in Salmonella. Microorganisms. 2022; 10(2):369. https://doi.org/10.3390/microorganisms10020369

Chicago/Turabian Style

Gao, Ruimin, Hongsheng Huang, Jérémie Hamel, Roger C. Levesque, Lawrence D. Goodridge, and Dele Ogunremi. 2022. "Application of a High-Throughput Targeted Sequence AmpliSeq Procedure to Assess the Presence and Variants of Virulence Genes in Salmonella" Microorganisms 10, no. 2: 369. https://doi.org/10.3390/microorganisms10020369

APA Style

Gao, R., Huang, H., Hamel, J., Levesque, R. C., Goodridge, L. D., & Ogunremi, D. (2022). Application of a High-Throughput Targeted Sequence AmpliSeq Procedure to Assess the Presence and Variants of Virulence Genes in Salmonella. Microorganisms, 10(2), 369. https://doi.org/10.3390/microorganisms10020369

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop