Salmonella enterica subsp. enterica bacteria are important foodborne pathogens with major economic impact. Some isolates exhibit increased heat tolerance, a concern for food safety. Analysis of a finished-quality genome sequence of an isolate commonly used in heat resistance studies, S. enterica subsp. enterica serovar Senftenberg 775W (ATCC 43845), demonstrated an interesting observation that this strain contains not just one, but two horizontally acquired thermotolerance locus homologs. These two loci reside on a large 341.3-kbp plasmid that is similar to the well-studied IncHI2 R478 plasmid but lacks any antibiotic resistance genes found on R478 or other IncHI2 plasmids. As this historical Salmonella isolate has been in use since 1941, comparative analysis of the plasmid and of the thermotolerance loci contained on the plasmid will provide insight into the evolution of heat resistance loci as well as acquisition of resistance determinants in IncHI2 plasmids.
IMPORTANCE Thermal interventions are commonly used in the food industry as a means of mitigating pathogen contamination in food products. Concern over heat-resistant food contaminants has recently increased, with the identification of a conserved locus shown to confer heat resistance in disparate lineages of Gram-negative bacteria. Complete sequence analysis of a historical isolate of Salmonella enterica serovar Senftenberg, used in numerous studies because of its novel heat resistance, revealed that this important strain possesses two distinct copies of this conserved thermotolerance locus, residing on a multireplicon IncHI2/IncHI2A plasmid. Phylogenetic analysis of these loci in comparison with homologs identified in various bacterial genera provides an opportunity to examine the evolution and distribution of loci conferring resistance to environmental stressors, such as heat and desiccation.
Heat-resistant bacterial contaminants have increasingly become a source of concern in the last several years especially with regard to food safety and human health. Although many members of the family Enterobacteriaceae exhibit variable heat resistance after heat shock, relatively few strains have been found to demonstrate an innate thermotolerant phenotype. Recently, however, a genetic locus found to confer heat resistance has been identified in certain strains of Cronobacter sakazakii, Escherichia coli, and Klebsiella species (1–3). Using a top-down proteomic approach, Williams et al. identified a protein biomarker sequence in Cronobacter correlating with thermotolerance to reverse engineer the DNA sequence of the protein that was homologous to a protein found in thermotolerant Methylobacillus flagellatus (4). Gajdosova and coworkers expanded upon this work and sequenced an 18-kbp region containing this open reading frame (ORF) and determined that the full locus is required for highest thermotolerance (1). Independently, Bojer et al. (3) isolated several Klebsiella pneumoniae strains over the course of a 2-year surveillance study of patients in a Danish hospital; this study demonstrated multiple antibiotic resistance phenotypes (3). These nosocomial isolates were associated with persistence in contaminating reusable endoscopes, and Bojer et al. (3) hypothesized that the adaptability of the isolates was due to their resiliency to harsh environmental stressors such as heat. Bojer et al. then identified a locus encoding small heat shock proteins and a novel Clp ATPase termed ClpK (3). This locus showed high similarity to the same chromosomal region from Methylobacillus that Williams et al. (4) identified and Gajdosova later sequenced (1). Subsequently, Bojer and coworkers reported that the clpK gene was cotransferred with plasmids encoding multiple drug resistance hosted by emerging K. pneumoniae clones associated with nosocomial outbreaks (5). More recently, Mercer et al. (2) and Dlusskaya et al. (6) sequenced an E. coli beef isolate that was able to survive in beef patties grilled to 71°C. Genomic comparisons of highly heat-resistant E. coli isolates resulted in the identification of a 14-kbp locus of heat resistance with greater than 99% identity to the aforementioned loci in Cronobacter and Klebsiella (2). Lee et al. expanded on the thermotolerance genetic module, noting that other ORFs on the loci are homologous to proteins involved in protein homeostasis and proposed that the loci be called transmissible loci for protein quality control (TLPQC) (7, 8). This locus appears to have been horizontally acquired in some bacterial lineages, as it is found not only in a number of clinically relevant members of the family Enterobacteriaceae but also more broadly in other Proteobacteria species (7).
A striking example of innate thermotolerance in the foodborne pathogen Salmonella enterica subsp. enterica has been reported for a particular strain of S. enterica serovar Senftenberg (9). This isolate, S. Senftenberg ATCC 43845 (originally referred to as strain 775W), has been used extensively in thermal resistance studies over the last several decades (9–12); it was first reported to be extremely thermotolerant in 1946 by Winter et al. (13), having been isolated previously by E. Beckler in 1941 from Chinese egg powder at the Massachusetts Department of Public Health (14). While dried egg products are not commonly sold in modern times, dried egg powder was used in rations in many Western countries, especially England, during the onset of World War II. Indeed, multiple reports of salmonellosis in the 1940s were associated with consumption of powdered eggs (15, 16).
Despite being used in numerous studies examining thermotolerance, the genetic determinants conferring heat resistance in S. Senftenberg ATCC 43845 have not been identified previously. Analysis of the complete genome of S. Senftenberg ATCC 43845 revealed the presence of two closely related loci with more than 98% identity to the thermotolerance loci previously studied in Cronobacter, Escherichia, Klebsiella, and Pseudomonas (1–3, 8) on a potentially conjugable IncHI2 plasmid. As the synteny of the genetic loci is highly conserved broadly across diverse members of the phylum Proteobacteria, here we designate the putative thermotolerance loci in S. Senftenberg TLPQC-1 and TLPQC-2 (Fig. 1) in keeping with the nomenclature proposed by Lee et al. (7). Comparative analysis of the S. Senftenberg TLPQCs, as well as other known homologs, provides a unifying framework for understanding the origin and evolution of loci conferring mechanisms of increased protein maintenance and reveals the prevalence and conservation of these loci within the Proteobacteria.
RESULTS AND DISCUSSION
Genomic features of S. Senftenberg ATCC 43845.S. Senftenberg ATCC 43845 has been used extensively in thermal resistance studies over the last several decades, having been identified as an extremely thermotolerant strain of Salmonella in 1946 by Winter et al. (13). A draft genome assembly consisting of 114 contigs (139-kb contig N50; accession number NZ_AOXX00000000) and a total length of 5,184,857 bases have previously been reported for this strain (17). Here we present the complete and closed, finished-quality genome sequence of S. Senftenberg ATCC 43845 based on long-read sequencing. The circularized chromosome consists of 4,920,660 bp with a chromosomal GC content of 52.2%, comparable to other finished Salmonella (Fig. 2). Annotation by the NCBI Prokaryotic Genome Annotation Pipeline (version 3.3) indicated the presence of 5,105 total genes, with 4,980 coding sequences (CDS) and 125 genes encoding RNAs (GenBank accession number CP016837). This includes 3,930 predicted proteins with homology to known proteins, in addition to 979 hypothetical or “miscellaneous” predicted proteins. In addition, the chromosome contains potential restriction-modification (RM) systems, including a DNA (cytosine-5)-methyltransferase motif, as well as phage integration-related N6-adenine methylation systems. Genome analysis using PHASTER (18) revealed the presence of two intact prophages and one questionable prophage (Fig. 2). Manual inspection revealed a Mu-like prophage, with up to 77% nucleotide identity but only 11% coverage of phage D108 by PHASTER at nucleotide positions 30,172 to 82,470. The Mu-like prophage is integrated downstream of mnmE (also known as thdF or trmE) which is a hot spot integration site for other mobile genetic elements in Salmonella (19). The other intact phage identified by PHASTER is a prophage related to the Haemophilus influenzae HP1 phage at positions 738,135 to 768,638. Manual inspection of the phage region identified as questionable by PHASTER analysis showed what appears instead to be an intact prophage related to the Salmonella phage vB_SosS_Oslo at nucleotides 1,549,983 to 1,599,675. No known virulence factors were associated with any of the prophages. A putative integrative conjugative element (ICE) identified by BLASTn is integrated into tRNAPhe at nucleotide positions 4,324,610 to 4,433,343. BLAST analysis of the ICE showed no virulence factors; however, the element appears to encode an arsenic resistance locus and a copper homeostasis and silver resistance island (CHASRI) (20). Significantly, there were no predicted thermotolerance loci on the chromosome of the isolate revealed by BLAST analysis.
Features of the pSSE ATCC-43845 plasmid.The finished assembly of S. Senftenberg ATCC 43845 revealed the presence of a large 341,373-bp circular plasmid, not previously described for this strain, and is here designated pSSE ATCC-43845. Combining the plasmid length with the chromosome takes the total genome complement of the strain to 5,262,033 bases, slightly exceeding that predicted by the short-read draft assembly (17). Annotation of this plasmid sequence via the same method as the chromosome predicts 336 genes on the plasmid, including two complete thermotolerance loci with distinct sequences between nucleotides 189,272 to 211,036 and 134,033 to 148,118 of GenBank accession number CP016838 (Fig. 3). These two loci are designated TLPQC-1 and TLPQC-2. Both thermotolerance islands have characteristic high average GC content (61%), contain the canonical clp protease gene, and are flanked by small heat shock proteins and predicted DNA-binding genes, in keeping with previously reported heat tolerance islands (Fig. 1) (1–3, 7). The GC content of the islands suggests that the origins of TLPQC-mediated heat resistance lie in Proteobacteria with higher average GC content than Salmonella species.
BLASTn comparison of the entire plasmid sequence to microbial sequences in GenBank indicates that it is closely related to the lncHI2 R478 plasmid, sharing up to 98% identity with more than 61% coverage. In silico analysis of the plasmid sequence with PlasmidFinder (21) identified two replicon regions, IncHI2 and IncHI2A. Comparative analyses further revealed that the pSSE ATCC-43845 plasmid has a comparable GC content value of 48.0% to the GC content of R478 at 45.5% (Fig. 3) (22, 23). Further evidence that pSSE ATCC-43845 is a member of the IncHI2 plasmid family is found by comparing transposon insertion sites which have been used as markers to track closely related plasmids of the same family. In R478 and closely related IncHI2 plasmids with antibiotic resistance, Tn10 and the linked tetracycline resistance genes have been used for this purpose (24), and in pSSE ATCC-43845, the Tn10 transposition site (BFF41_25185 [Fig. 4]) was found to be intact. As with R478, several heavy metal resistance determinants are present on pSSE ATCC-43845, including a tellurite resistance island and another copy of CHASRI (23, 25). As illustrated in Fig. 4, the variable regions of R478 and pSSE ATCC-43845 are bracketed by the tellurite resistance operon and CHASRI. This variable region in R478 encodes mercury resistance and chloramphenicol (cat) and kanamycin (aphA) resistance, while in pSSE ATCC-43845, this region contains the two TLPQC loci. DNA sequence analyses showed that pSSE ATCC-43845 does not contain genes encoding drug resistance determinants. In keeping with this finding, antimicrobial sensitivity phenotyping (as evaluated by broth microdilution testing methods; TREK, Sensititre CMV3AGNF plates) showed S. Senftenberg ATCC-43845 to be pansusceptible to 15 antimicrobial agents tested on the panel. This finding is novel and suggests that pSSE ATCC-43845 is a relic of IncHI2 plasmids before the era of therapeutic antibiotic use, as present-day examples of IncHI2 plasmids have frequently been identified as carriers of antibiotic resistance genes (26–28).
Methylation analysis.Sequence analysis using the Pacific Biosciences platform produces data that can detect base modification in the source DNA (29). Detecting these modifications is dependent on both the depth of sequence coverage and the magnitude of the effect of the modification on the kinetics of nucleotide incorporation. Nucleotide modification is predicted from the interpulse distance data (see Materials and Methods) with N6-adenine methylation producing a strong signal that is easily detected at low coverage, while N5-cytosine methylation demonstrates a weaker signal that requires deeper coverage. Methylation motif analysis of the S. Senftenberg sequence data and comparisons to established motifs in REBASE (30) showed adenine N6-methylation of the sites GATC, CAGAG, and ATGCAT (the adenine methylated is shown underlined), as observed previously with other Salmonella serovars (31). Additional motifs were identified in S. Senftenberg and are summarized in Table 1. Of particular interest was methylation of the CGCAN8TCNG motif, as the presence of a cluster of three of these motifs was observed in each of the novel clp protease genes found in TLPQC-1 and TLPQC-2. This is illustrated in Fig. 3, where modification values (Qmod) indicating methylation of this site on the positive and negative strands are visualized as the outermost and innermost tracks, respectively, on pSSE ATCC-43845. Depiction of the methylation state of this site on the plasmid also reveals a skew of the presence of these motifs, which is likely due to the chimerism of the R478 backbone. The presence of these motifs in TLPQC-1 and -2 is noteworthy and hints at the possibility that this motif may play some regulatory role in the expression of genes contained in the TLPQC loci. Additional base modification motif maps of the plasmid are visualized in the supplemental figures (see Fig. S1 and S2 in the supplemental material).
Copyright © 2017 Nguyen et al.
This content is distributed under the terms of the Creative Commons Attribution 4.0 International license.
Copyright © 2017 Nguyen et al.
This content is distributed under the terms of the Creative Commons Attribution 4.0 International license.
Comparison of TLPQC loci.A phylogenetic analysis (see Materials and Methods) was performed to determine the relationship of the Salmonella heat tolerance islands to those in Klebsiella plasmids, Pseudomonas chromosomes, and other bacteria. Heat tolerance islands were identified by searching the literature (previous reports) or by searching for similarity of nucleotide sequence records in GenBank to representative TLPQC sequences by BLASTn (see Materials and Methods). A total of 91 heat tolerance islands, including 48 reported to be located on chromosomes, 42 reported to reside on plasmids, and one presumably on a plasmid (Table 2), as well as the two TLPQC islands from plasmid pSSE ATCC-43845, were included in the analysis. For each species, the thermotolerance island was manually curated from the reported sequence (Table 2). Previously, two clades (A and B) of heat tolerance islands, which were termed “locus of heat resistance” (LHR), were identified (2). We chose to use instead the nomenclature of TLPQC proposed and adopted by Lee et al. (7), since a number of ORFs within the LHR islands have been shown or predicted to encode protein quality control functions involved in surviving a range of stressors beyond heat, including antibiotic exposure, desiccation, and oxidative stress. However, we propose to amend the TLPQC classification by Lee et al. (7) and incorporate phylogenetic analysis from our work and Mercer et al. (2) for multiple clades (Fig. 5). Thus, the broader phylogeny incorporating the pSSE ATCC-43845 islands (Fig. 5) indicates three clades, TLPQC-1, -2, and -3, with the third clade comprised of a reduced TLPQC island primarily observed in plasmids of Klebsiella. There is high genetic conservation within clades and less genetic conservation between clades, and genetic synteny is maintained across multiple genera (Fig. 1 and 5). The two pSSE ATCC-43845 islands fall within clades 1 and 2, consistent with the names assigned here, and are no more closely related to other plasmid-borne islands than to chromosome-borne islands.
Members of the TLPQC-1 clade are broadly distributed within the Proteobacteria (Fig. 5) with examples found in environmental isolates such as Methylobacillus flagellatus and Dechlorosoma suillum (32, 33), as well as within clinically relevant bacteria such as Pseudomonas, Pandorea, and Burkholderia (8). The majority of TLPQC-1 were found to be chromosomally located, with the exception of S. Senftenberg TLPQC-1 and another on plasmid 1 of Klebsiella pneumoniae J1 (GenBank accession number CP013712). Comparative analyses of these chromosomally bound TLPQCs suggest that they are inserted into larger mobile genetic elements (data not shown). TLPQC-1 genes in strains marked with black stars (Fig. 5) indicate that the loci are found within an integrative conjugative element (ICE) that is integrated into tRNAGly. Attachment repeats flanking the putative ICE were identified with the core att nucleotide sequence 5′ TGGAGCGGGCGATGGGAA. Members of the TLPQC-1 lineage contain the longest islands of the three lineages (~15-kbp average), with core genes predicted to encode a phage-like transcriptional regulator, two small heat shock protein variants, a novel Clp protease, cardiolipin synthase, the membrane-associated ATP-dependent protease FtsH, two variants of YfdX protein family with potential chaperone activity, thioredoxin, a glutathione regulated K+ efflux pump related to KefB, a phosphate starvation-inducible PsiE family protein, a Zn-dependent protease, and a periplasmic serine protease (7) (Fig. 5).
Although most of the ORFs in TLPQC-1 have not been experimentally verified, one of the small heat shock proteins, sHsp20c, was recently characterized and found to aggregate in a 24-mer that plays a role in ameliorating stress resistance as a protein chaperone in Pseudomonas (8). Additionally, the novel Clp protease, ClpGG1, has been shown to confer thermal resistance in Cronobacter, Escherichia, Pseudomonas, and other Enterobacteriaceae (3, 8). The rest of the ORFs in TLPQC-1 have not been studied in depth; however, it has been shown that the full TLPQC is needed for the greatest thermal protection (1). The conservation of TLPQC and protein domains within the uncharacterized ORFs to known proteases, chaperones, and other quality control proteins suggests that these additional ORFs may contribute to survival in other adverse conditions such as osmotic shock or desiccation resistance.
The second lineage observed, TLPQC-2, is the best-studied TLPQC thus far, having first been identified in Cronobacter sakazakii contaminants of powdered infant formula (1). Islands in this lineage are shorter than TLPQC-1 (~14 kb) primarily because of the truncation of genes predicted to encode cardiolipin synthase (cls) and FtsH (hflB). Although these truncated ORFs are conserved across various TLPQC-2s, it is unknown whether they are transcribed and expressed or what function they might serve. TLPQC-2 islands were observed on both chromosomal and plasmid locations and were predominantly found in members of the Enterobacteriaceae, including Escherichia, Obesumbacterium, Klebsiella, Cronobacter, Salmonella, and Yersinia (Fig. 5) (1–3). Bacterial species that possess TLPQC-2 are of great interest to food industry and human health communities, as many are relevant to food safety such as Obesumbacterium proteus as brewery yeast contaminants (34) and C. sakazakii as infant formula contaminants (1). Given that a number of TLPQC-2 loci are found on plasmids, the potential exists for transmission and subsequent conferring of heat resistance to foodborne pathogens. In S. Senftenberg, the presence of the two TLPQCs on a multireplicon conjugative plasmid highlights the potentially mobile nature of these environmental stress resistance determinants.
The third lineage, TLPQC-3, is the shortest of the three loci (mean length of 6.5 kb), and composed only of the ORFs that encode a heat shock protein, the Clp protease, a truncated version of cardiolipin synthase, the Zn-dependent protease, and the periplasmic serine protease (Fig. 5). Found primarily within Klebsiella pneumoniae plasmids, this TLPQC has been associated with the emerging multidrug-resistant K. pneumoniae ST16 clone (5, 35). The Clp protease in TLPQC-3 has been characterized and was named ClpK, sharing 98.3% amino acid identity with the pSSE ATCC-43845 ClpGG1 (3). The high degree of conservation of the core elements within these islands underscores their apparent importance to host fitness.
Conclusions.We describe the presence of two broadly disseminated but uncommon TLPQCs present in the important pathogen Salmonella enterica. The discovery of two TLPQCs in S. Senftenberg ATCC 43845 underscores its uniqueness and suggests a likely mechanism for the unusually thermotolerant phenotype repeatedly observed for this strain (9–13). As this strain of S. Senftenberg has been maintained in collections since at least 1941, S. Senftenberg ATCC 43845 is also one of the oldest known isolates to harbor TLPQCs. The presence of TLPQC-1 and TLPQC-2 on a single plasmid raises questions about the selective pressures and genetic events that led to their evolution. As the original source of S. Senftenberg ATCC 43845 was powdered eggs, it is tempting to suggest a correlation between the acquisition of the TLPQC loci and exposure to conditions likely encountered in the production of powdered egg. Regardless of how it evolved, the complete genome sequence and methylome data of this unusually thermotolerant strain, as well as the phylogenetic analysis of these broadly distributed islands conferring enhanced mechanisms of protein maintenance, will facilitate future studies on the role of TLPQC loci in extreme heat and desiccation resistance within Gram-negative bacteria important to human health and food safety. In addition, the complete sequence of an IncHI2 R478-like plasmid isolated at the dawn of the era of therapeutic antibiotic use will aid in furthering our understanding of the evolution of resistance plasmids.
MATERIALS AND METHODS
DNA isolation and genome sequencing.S. Senftenberg ATCC 43845 was acquired from ATCC and cultured statically at 37°C in Trypticase soy broth (Becton, Dickinson, Franklin Lakes, NJ) for 18 to 20 h. DNA was purified using Qiagen genomic-tip 100/G columns and the DNeasy blood and tissue DNA isolation kit (Qiagen, Valencia, CA) using the manufacturer’s recommended protocol. Sequencing libraries were prepared following the recommended Pacific Bioscience protocols. Single molecule real-time sequencing (SMRT) was performed using P6/C4 chemistry on a Pacific Bioscience (PacBio) RS II instrument (Pacific Biosciences, Menlo Park CA), resulting in average subreads of >7 kbp and mean coverage of 172.59×. The HGAP3 protocol in smrtanalysis v2.3 was used to assemble sequence reads and polish the contig. To circularize the contigs, a self/self dot plot of the contig sequences was generated in Geneious 9.1.5 (Biomatters Ltd., New Zealand) and used to identify duplicated sequence at the contig ends. The duplicated sequence was trimmed to generate a circularized sequence (36). OriFinder was used to determine the origin of replication and to reset base position 1 of the chromosome. The position of circularization was confirmed by mapping all the reads to the renumbered contig using the resequencing protocol in smrtanalysis v2.3, which also provided a second round of polishing via the included quiver routine. Genome and plasmid sequence data were annotated using the NCBI Prokaryotic Genome Annotation Pipeline and deposited into GenBank (GenBank accession numbers CP016837 and CP016838, respectively).
Genome analysis.The sequences of the thermoresistance cluster originally identified in C. sakazakii (GenBank accession number FR714908) (1) and the thermotolerance loci in S. Senftenberg were used to identify related islands in the GenBank nr database using BLASTn. Genomes with strong similarity to these sequences were identified from 20 genera, and genome sequences were used to extract 91 additional thermotolerance islands by manual inspection to determine the island boundaries (Table 2). Sequence inspection and manipulation were performed in Geneious vR9.1.5. Manual inspection included BLAST comparisons of flanking mobile element genes, which revealed little homology between the different TLPQCs so these flanking sequences were trimmed and discarded to define the boundaries of the TLPQC. The extracted TLPQCs were aligned by MAFFT (37) in Geneious 9.1.5, and the resulting aligned sequences were analyzed by jModelTest 2 to determine the parameters for MrBayes (36, 38, 39). The best model for the full-aligned sequence of the TLPQC was determined to be TRN+I+G (Tamura-Nei model with invariant sites and discrete gamma distribution) using the Akaike information criterion (40). MrBayes 3.2.6, utilized as a Geneious plug-in, was used to calculate a Bayesian inference tree of the various TLPQCs. As the TRN+I+G model is not possible in MrBayes 3.2.6, the GTR+I+G model (generalized time reversible model with invariant sites and discrete gamma distribution) was used, as overparameterizeration gives substantially less bias than underparameterizeration (41), and 3,000,000 iterations were performed with these defined parameters. The first 25% of the iterations were discarded as burn-in, as the Markov chains have not reached stationary and may alter the final result. Visualization of the tracer output and examining the standard deviation of the split frequencies were used to assess the quality of the tree. As the standard deviation of the split frequencies reached only <0.04, the chains have truly not converged but were acceptable, given the extensive computing resources to run the calculations. Additional maximum likelihood trees generated by PhyML (42) gave further support for the Bayesian consensus tree (not shown). Representatives of TLPQCs derived from the Bayesian phylogenetic tree were extracted and visualized in an Easyfig BLASTn comparison (43). CGView Server was used to visualize chromosome and plasmid sequence data with default tBLASTx settings (44).
Base modification analysis.Genome analysis of nucleotide base modifications were detected using the RS_Modification Motif_Analysis.1 protocol in smrtanalysis v2.3 (Pacific Biosciences, Menlo Park, CA) with the default threshold quality value (QV) of 30. The resulting motif_summary.csv files were uploaded to the restriction enzyme database (REBASE) and assessed for the presence of novel restriction modification systems (30). Additionally, kinetic Qmod values [defined by the log-transformed P value from the t test, −10log(P value)] with a threshold value of 55 were extracted and processed by BaseModFunctions v2.1.R (45) for visualization in Circos (46). The threshold value of 55 was based on the bimodal distribution of modification values observed in the kinetic detection map. Base modification data have been uploaded to GenBank.
Accession number(s). Accession numbers for the strains used in this study can be found in Table 2.
We are grateful to Kerry Brader, Sandy Bradley, and Kristen Kuhn (USDA-ARS) for technical assistance and Jody Gallagher for secretarial support.
We declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The mention of a trade name, proprietary product, or specific equipment does not constitute a guarantee or warranty by the USDA and does not imply approval to the exclusion of other products that might be suitable. The USDA is an equal opportunity employer.
Citation Nguyen SV, Harhay GP, Bono JL, Smith TPL, Harhay DM. 2017. Genome sequence of the thermotolerant foodborne pathogen Salmonella enterica serovar Senftenberg ATCC 43845 and phylogenetic analysis of loci encoding increased protein quality control mechanisms. mSystems 2:e00190-16. https://doi.org/10.1128/mSystems.00190-16.
- Received December 9, 2016.
- Accepted January 28, 2017.
- Copyright © 2017 Nguyen et al.
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license.