The various sub-species of Salmonella enterica cause a range of disease in human hosts. The human-adapted Salmonella enterica serovar Typhi enters the gastrointestinal tract and invades systemic sites to cause enteric (typhoid) fever. In contrast, most non-typhoidal serovars of Salmonella are primarily restricted to gut tissues. Across Africa, invasive non-typhoidal Salmonella (iNTS) have emerged with an ability to spread beyond the gastrointestinal tract and cause systemic bloodstream infections with increased morbidity and mortality. To investigate this evolution in pathogenesis, we compared the genomes of African iNTS isolates with other Salmonella enterica serovar Typhimurium and identified several macA and macB gene variants unique to African iNTS. MacAB forms a tripartite efflux pump with TolC and is implicated in Salmonella pathogenesis. We show that macAB transcription is upregulated during macrophage infection and after antimicrobial peptide exposure, with macAB transcription being supported by the PhoP/Q two-component system. Constitutive expression of macAB improves survival of Salmonella in the presence of the antimicrobial peptide C18G. Furthermore, these macAB variants affect replication in macrophages and influence fitness during colonization of the murine gastrointestinal tract. Importantly, the infection outcome resulting from these macAB variants depends upon both the Salmonella Typhimurium genetic background and the host gene Nramp1, an important determinant of innate resistance to intracellular bacterial infection. The variations we have identified in the MacAB-TolC efflux pump in African iNTS may reflect evolution within human host populations that are compromised in their ability to clear intracellular Salmonella infections.
Salmonella Typhimurium will generally cause acute gut infections in humans. However, S. Typhimurium strains causing severe, systemic infections have emerged in sub-Saharan Africa and are phylogenetically distinct from other S. Typhimurium strains. Our comparative genomic analysis revealed S. Typhimurium sequence-type 313 (ST313) from Africa have notable sequence variations within the macA and macB genes. These genes are already known to play a role in Salmonella pathogenesis and are otherwise conserved in Salmonella and many other Gram-negative bacteria. We show that regulation of macAB transcription depends, in part, on the key Salmonella virulence system PhoP/Q and that expression of MacAB improves Salmonella resistance to an antimicrobial peptide. African macAB variants interfere with this antimicrobial peptide resistance function and can alter Salmonella replication within macrophages. Using competitive infection experiments in mice, we see that these macAB variants influence fitness in the mammalian gut and systemic sites, with African S. Typhimurium reliant upon its macAB genotype for systemic infection of susceptible hosts. These results suggest that the evolution of African S. Typhimurium has been shaped by human populations with impaired ability to control intracellular Salmonella infections.
Citation: Honeycutt JD, Wenner N, Li Y, Brewer SM, Massis LM, Brubaker SW, et al. (2020) Genetic variation in the MacAB-TolC efflux pump influences pathogenesis of invasive Salmonella isolates from Africa. PLoS Pathog 16(8): e1008763. https://doi.org/10.1371/journal.ppat.1008763
Editor: Andreas J. Baumler, University of California Davis School of Medicine, UNITED STATES
Received: May 1, 2020; Accepted: June 30, 2020; Published: August 24, 2020
Copyright: © 2020 Honeycutt et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: Research reported in this publication was supported by grant R01-AI116059 from NIAID (DM) and the Paul Allen Stanford Discovery Center on Systems Modeling of Infection. This work was supported by a Wellcome Trust Senior Investigator award to JCDH (Grant 106914/Z/15/Z). JDH was supported the Lucile P. Markey Stanford Graduate Fellowship and NIH T32AI007290. SWB was supported by NIGMS T32GM007276, NIAID 5T32AI00729035, and the Dean’s Postdoctoral Fellowship – Stanford School of Medicine. SMB was supported by a Stanford Graduate Fellowship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: I have read the journal's policy and the authors of this manuscript have the following competing interests: Rocío Canals currently works for GSK Vaccines Institute for Global Health. All other authors have no competing interests.
Salmonella infections continue to be a significant challenge for human health. With an estimated 95 million annual cases, non-typhoidal Salmonella (NTS) infection is typically characterized by severe but self-resolving gastroenteritis in otherwise healthy people [1–3]. Typhoid and paratyphoid fever cases number more than 14 million annually and are characterized by invasive, bloodstream infection by Salmonella serovars Typhi and Paratyphi, respectively . Risk for typhoid disease remains high in geographic areas with inadequate sanitation infrastructure as the infectious cycle relies on human-to-human transmission. While some humans can become asymptomatic chronic carriers, untreated typhoid fever is often fatal . With appropriate treatment, the 1% case mortality of typhoid fever is similar to that of gastroenteritis associated with NTS [2,4,5].
Invasive non-typhoidal Salmonella (iNTS) isolates belonging to the serovars Typhimurium and Enteritidis have caused major disease outbreaks within sub-Saharan Africa [6–9]. These African iNTS isolates are associated with systemic infections, and particularly high case mortality in children less than 5 years of age, the elderly, and those with comorbidities such as HIV and malaria [6,10]. Although typhoid infections outnumber iNTS infections by over 25-fold globally, iNTS caused nearly half as many deaths with overall mortality rates above 14% . In Africa, iNTS was responsible for 49,600 deaths in 2017 . There is great need to further understand iNTS pathogenesis and epidemiology in order to improve diagnosis and clinical outcomes for these increasingly antibiotic resistant infections [11–13].
By multi-locus sequence typing, many Salmonella enterica serovar Typhimurium (S. Typhimurium) gastrointestinal isolates are classified as sequence-type 19 (“ST19”), while African S. Typhimurium associated with invasive disease belong mainly to sequence-type 313 (“ST313”) [7,9]. ST313 isolates have also been observed in the UK and Brazil, though African ST313 form a distinct lineage [14–16]. Comparative genomic analysis has identified numerous changes in African ST313 lineages. While they have acquired unique prophages, plasmids, and antibiotic resistance genes, African ST313 also display gene degradation events that impair the ability of these isolates to survive outside of mammalian hosts [13,17–23]. Further gene degradation events have been shown to alter invasive and immune stimulating behavior in experimental animal models, supporting the view that African iNTS strains are evolving from causing strictly enteropathogenic disease to causing invasive disease in human hosts [15,24–26].
In our comparative analysis of S. Typhimurium ST313 lineage isolates with other Salmonella genomes we observed specific changes at the macAB locus in ST313 lineages. In Gram-negative bacteria, MacAB forms a tripartite channel with the outer membrane protein TolC to efflux various antimicrobial compounds as well as endogenous molecules and toxins [27–31]. As an ABC-type efflux pump, hydrolysis of cytoplasmic ATP by MacB directly drives movement of MacA and TolC to translocate molecules from the periplasm into the extracellular space [32,33]. The naming of macA and macB (previously annotated as ybjY and ybjZ) references their ability to confer resistance to macrolide antibiotics when overexpressed together from a plasmid , though in standard laboratory culture conditions macAB is not expressed in Salmonella [35,36]. In clinical isolates of other bacteria, increased expression of macAB homologues increases resistance to antimicrobial peptides such as polymyxins . The fact that MacAB has a virulence role in animal models of oral Salmonella infection [35,38] prompted us to explore how African ST313-associated macAB gene variants might influence pathogenesis.
Lineage-specific variation of the macAB locus in invasive African S. Typhimurium ST313 isolates
Comparative analysis of all currently available genomes of ST313 isolates of Salmonella enterica serovar Typhimurium identified several genomic changes at the macAB locus that differed from other gastroenteritis-associated S. Typhimurium (Fig 1A). Although some UK and Brazilian ST313 isolates carry an indel in the macA gene, this variant is not present in African isolates from ST313 lineages (1, 2) or sublineages (2.1, 2.2) that are associated with invasive disease (Fig 1A). Instead, all other ST313 lineage isolates in our analysis carry a C➝T non-synonymous SNP within macA (Fig 1A), which replaces Serine (S174) with Leucine (L174) (Fig 1B and 1C). Alignment of this sub-region of MacA shows that the Serine residue is highly conserved amongst other Gram-negative bacterial genera (Fig 1D). The structure of the Escherichia coli (E. coli) MacAB-TolC complex has been solved ; assuming an analogous overall structure in S. Typhimurium, the hydrophilic Serine174 residue faces the channel interior, residing beyond the proposed gating ring and in series with other hydrophilic residues that form the interior surface of the MacA channel  (Fig 1E). We hypothesized that the ST313-associated mutation of this conserved Serine residue to the hydrophobic amino acid Leucine altered the function of the MacAB-TolC channel, particularly since this mutation would be repeated around the interior of the fully-assembled hexameric MacA channel (Fig 1E, right).
(A) Phylogeny emphasizing all available ST313 isolates from Africa, the UK, and Brazil, with sequence type and lineage membership indicated by first vertical colored bar. Salmonella Typhi CT18 was used as an outgroup for tree construction. Presence of macAB variants are shown by corresponding colored bars. Representative isolates of each lineage are highlighted in red. (B) The macAB genomic locus with SNP locations and their effects on MacA and MacB proteins. The proximal upstream region of the macA start codon indicating the PhoP-box characterized by Nishino et al  and the 5’-UTRmacALin2.1 SNP identified by Van Puyvelde et al  (B, bottom). 5’-UTR = 5’ untranslated region. TSSmacAB = macAB transcription start site, 504 bases upstream of macA start codon. RBS = ribosome binding site. (C) macAB genomic locus highlighting variant combinations that pertain to each lineage. (D) Alignment showing conservation of the amino acid sequence surrounding the S174 residue of MacA with the ST313 S174L mutation boxed in red. (E) Overlay of 4/74 MacA predicted structure onto E. coli MacAB-TolC (PDB 5NIK), with residues colored blue and tan for hydrophilic and hydrophobic side chains, respectively. S174 highlighted in magenta (E, left and middle top) and L174 in yellow (E, middle bottom). S174 highlighted in all chains of the MacA hexamer, with the putative channel gating ring  in boxed outline (E, right). (F) E. coli MacB domain architecture from  (E, top) with macBW262Stop and macBindel truncations of MacB (E, bottom). Overlay of truncated MacB structure (orange) onto E. coli MacB (blue) (F, inset). 4/74 MacA prediction by Phyre2 . Structural diagrams generated with CCP4  using the operation superpose .
S. Typhimurium ST313 lineage 1 isolates contain a nonsense mutation in macB  changing the codon for W262 of MacB to a stop codon; this change lies within the amphipathic helix that precedes the first transmembrane domain , leading to the production of a truncated MacB (Fig 1F) . Furthermore, a two-nucleotide insertion (an indel) that results in a frameshift in the macB gene was found in about half of the S. Typhimurium ST313 lineage 2 isolates, including the reference strain D23580. This pseudogenization event created a stop codon that truncates the MacB protein (Fig 1F). This truncation interrupts the N-terminal ATP-binding domain and prevents translation of the transmembrane domains that extend from the cytoplasm into the periplasmic space (Fig 1F); assuming an analogous structure to E. coli MacB, the truncated MacB protein in ST313 Lineage 2 is predicted to be unable to interact with MacA in the context of the MacAB-TolC efflux pump .
The recently described African S. Typhimurium ST313 sublineage 2.1  retains the macAC➝T SNP associated with other ST313 while also having an A➝G SNP in the 5’ untranslated region (5’-UTR) of macA (Fig 1B). Sublineage 2.1 and 2.2 isolates do not carry the macB mutations associated with lineage 1 or 2 isolates (Fig 1A).
The presence of multiple variations at the macAB locus of African ST313 suggested reductive or adaptive evolution has occurred . We focused on characterizing macAB variants found in the most recently isolated African S. Typhimurium ST313 lineages 2 and sublineage 2.1 that are associated with invasive disease (Fig 1C). For clarity, these mutations are designated: macAC➝T for the non-synonymous SNP shared across all ST313 lineages; macBindel for the SNP present in ST313 lineage 2 that introduces multiple stop codons due to a frameshift and truncates the MacB protein; and 5’-UTRmacALin2.1 for the A➝G SNP within the 5’-UTR of macA of ST313 sublineage 2.1 isolates. We use macAST19 and macBST19 to refer to the alleles that are carried by S. Typhimurium ST19.
Expression of macAB is promoted by PhoP in Salmonella Typhimurium
The macAB (ybjY-ybjZ) genes are operonic with the transcriptional start site (TSS) located 504 nucleotides upstream of the translational start of the macA gene . This is a particularly long 5’-UTR, and regions of this type have previously been shown to play important regulatory roles in Salmonella [42,43].
Since previous studies suggested that macAB is important for S. Typhimurium virulence [35,38], we wanted to clarify how expression of this locus is regulated. In Salmonella the two-component system PhoP/Q senses low magnesium, acidic pH, and antimicrobial peptide disturbance of the inner membrane. PhoP up-regulates a set of genes that increase cellular resistance to antimicrobial peptides and promote survival in macrophages [44–48]. Previously, PhoP was shown to physically bind to a PhoP-box upstream of the macA coding sequence, with the authors concluding that PhoP represses macAB transcription .
We previously published the RNA-seq-based transcriptomic profiles of S. Typhimurium ST19 strain 4/74 during growth in multiple in vitro conditions and during intramacrophage replication [36,49]. The major pathogenicity locus SPI-2 is important for survival of Salmonella in mammalian phagocytic cells, and expression of SPI-2 genes can be induced by defined media that mimic some conditions of the vacuolar environment (InSPI2). Our published data show that S. Typhimurium 4/74 increases transcription of macA and macB in low magnesium InSPI2 medium, consistent with a role for stimuli sensed by PhoP/Q in promoting macAB expression (Fig 2A). More recently, we published the transcriptomic profiles of the macAB genes of S. Typhimurium ST313 strain D23580 using the same environmental conditions, including intra-macrophage replication , summarized at https://tinyurl.com/macAB-SalCom474-D23. In data sets from both ST19 isolate 4/74 and ST313 lineage 2 isolate D23580, the macAB transcript is upregulated during replication in RAW264.7 macrophage-like cells, compared to early stationary phase (ESP) growth in LB medium (Fig 2B). Furthermore, we previously observed that deletion of phoPQ reduced macAB transcription by about half in InSPI2 medium .
(A) Gene expression in InSPI2 media in response to environmental stimuli, fold-change relative to InSPI2 medium alone. (B) S. Typhimurium intramacrophage gene expression measured from RAW264.7 macrophages after 8 hours of infection, fold-change relative to expression in LB at early stationary phase. Panels A and B show RNA-seq data extracted from the SalComD23580, SalComMac, and SalComRegulon databases reported previously [18,36,49]. (C) ST19 4/74 and 4/74 phoP null mutants (phoP::Tn10) with chromosomally integrated lacZY transcriptional fusions for either phoN and macAB were grown to mid-exponential phase in N minimal medium, pH 7.4 and high (10mM) MgCl2 before transfer to the same or low (10μM) MgCl2 media and growth for 90 minutes. β-galactosidase production was measured by a kinetic Miller assay. (D) At mid-exponential phase, transcriptional fusion strains were shifted from pH 7.4 and 10mM MgCl2 to media buffered to the indicated pH with high (10mM) or low (10μM) MgCl2 and grown for 90 minutes. (E) Transcriptional fusion strains at mid-exponential phase were exposed to the antimicrobial peptide C18G (5μg/mL) in N minimal medium at pH 7.4 with 1mM MgCl2 and grown for 90 minutes. Data are from three repeat experiments. ANOVA with Tukey post-test (C, D), or t-test (E); bar = mean; error = standard deviation.
To examine macAB transcriptional regulation more closely, we created chromosomal lacZY transcriptional fusions in S. Typhimurium ST19 isolate 4/74 driven by the endogenous phoN or macAB promoters. We grew these reporter strains overnight in defined N minimal medium at pH 7.4 with high (10mM) magnesium, a condition that represses PhoP activity . After subculture and growth to mid-exponential phase, we shifted cells to low (10μM) magnesium medium, a treatment that up-regulates expression of the PhoP-regulated gene phoN . We found that β-galactosidase levels increased for both phoN::lacZY (positive control) as well as macAB::lacZY when cells were shifted to low magnesium medium (Fig 2C). Furthermore, the up-regulation of these genes was impaired when phoP null mutants were shifted to low magnesium medium (Fig 2C). In addition, when cells with the phoN and macAB transcriptional fusions were shifted to increasingly acidic conditions (from pH 5.8 to pH 4.9), levels of β-galactosidase activity were significantly increased (Fig 2D), in agreement with previous reports that acidic pH stimulates expression of PhoP/Q-dependent genes [53,54].
Cationic antimicrobial peptides induce expression of PhoP-regulated genes  and play bacteriostatic and bactericidal roles during Salmonella infection of macrophages . To test whether macAB gene expression is upregulated by antimicrobial peptides, we grew the phoN and macAB transcriptional fusion strains in moderate (1mM) magnesium followed by treatment with a sub-inhibitory concentration of the salt-insensitive, cationic antimicrobial peptide C18G . Exposure to C18G induced higher β-galactosidase activity in the phoN and macAB transcriptional fusion strains (Fig 2E). These experiments collectively show that macAB transcription is facilitated by PhoP under biologically relevant conditions.
African S. Typhimurium ST313 macAB variants influence replication in macrophages
We next wanted to determine if the genetic changes in the macAB locus of the ST313 lineages commonly associated with invasive disease in Africa had functional consequences during infection. We focused our experiments on lineage 2 and sublineage 2.1 macAB variants, reasoning that the lineage 1 macBSTOP would have effects that are similar to the lineage 2 macBindel. Furthermore, ST313 lineage 1 isolates are no longer causing a clinical problem in Africa . To ensure otherwise native regulation of the macAB genes, we introduced marker-less nucleotide changes directly into the macAB locus of the ST19 isolate 4/74 or the ST313 lineage 2 isolate D23580. The 4/74 macAC➝T mutant was made to test the impact of this SNP alone, the 4/74 macAC➝T macBindel mutant to represent the lineage 2 macAB genotype, and the 4/74 5’-UTRmacALin2.1 macAC➝T mutant to represent the lineage 2.1 macAB genotype (Fig 1C). We additionally made a 4/74 5’-UTRmacALin2.1 mutant to test whether this SNP alone affects pathogenesis.
Similarly, we modified D23580 to test the role of individual macAB SNPs in modulating virulence. The ST313 lineage 2 macBindel was first removed, yielding D23580 macAC➝T macBST19. This was followed by alteration of the macAC➝T SNP, creating a D23580 strain with the full ST19 genotype (D23580 macAST19macBST19). We confirmed that our engineered strains carried the desired nucleotide modifications, with no unintended mutations elsewhere in the chromosome, by whole genome sequencing (see S2 Table and Materials and Methods).
Previous studies have shown that S. Typhimurium ST19 strains 14028S and SL1344 with deletions of macAB replicate poorly within mouse macrophages [38,56]. We infected the murine RAW264.7 macrophage cell line with a range of our marker-less macAB mutants to assess Salmonella intracellular replication. As expected, we found that ST19 4/74 phoP and macAB null mutants showed significantly reduced replication in macrophages compared to parental 4/74 (Fig 3A). Replication of the 4/74 macAC➝T mutant was significantly lower than the parental 4/74 and similar to replication of the macAB null mutant (Fig 3A) showing this ST313-associated SNP likely impairs MacAB-TolC functionality during intramacrophage replication. The 4/74 macAC➝T macBindel mutant (ST313 lineage 2 genotype) also showed lower replication than parental 4/74 (Fig 3A). Interestingly, the 5’-UTRmacALin2.1 SNP alone significantly reduced replication of 4/74, while the 4/74 5’-UTRmacALin2.1 macAC➝T mutant with the full ST313 sublineage 2.1 genotype replicated at a level similar to the macAB null mutant (Fig 3A).
RAW264.7 cells were infected at an MOI of 10 with bacteria from overnight stationary phase cultures. Replication was assessed by plating bacteria at the indicated timepoints with fold-replication calculated relative to CFU/well at t = 0 as described in Materials and Methods. (A) S. Typhimurium ST19 4/74 with derived mutants, and (B) ST313 lineage 2 isolate D23580 with derived mutants. Two-way ANOVA (time, strain) with Dunnett’s post-test comparing each mutant to the parent strain; bar = geometric mean; error bars = geometric standard deviation. ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001.
We also assessed replication of S. Typhimurium D23580, the reference ST313 lineage 2 African isolate, in murine RAW264.7 macrophages. D23580 replicated extensively in RAW cells to a higher level than 4/74 (Fig 3A and 3B), as previously reported . We made marker-less point mutations in D23580 to change the macAC➝T and macBindel SNPs to the respective ST19 alleles. The D23580 macAC➝TmacBST19 and D23580 macAST19macBST19 mutants showed no change in fold-replication in this experimental setting when compared to the parent D23580 (Fig 3B).
Taken together, these results show the macAB-associated SNPs of S. Typhimurium ST313 lineages do have phenotypic consequences in macrophage replication, though the replication effect may be epistatic. The ST19 strain 4/74 shows reduced levels of intra-macrophage replication when macAB is modified to the various ST313 genotypes, suggesting that these SNP mutations impair MacAB-TolC function. In contrast, an African ST313 lineage 2 isolate maintains its level of intra-macrophage replication irrespective of macAB genotype. This implies that D23580 has a macAB-independent mechanism for its enhanced replication phenotype in RAW macrophage-like cells.
ST19 macAB provides resistance to the cationic antimicrobial peptide C18G
Previous studies have shown that MacAB homologues can contribute to increased resistance to antimicrobial peptides in other bacteria . In E. coli, the TolC-dependent secretion of the helical, amphiphilic peptide enterotoxin II is facilitated by MacAB, but not by other TolC-interacting partners . Additionally, another ABC-type efflux pump (MtrCDE) improves Neisseria resistance to both macrolide antibiotics and antimicrobial peptides . Since we showed macAB expression in S. Typhimurium is regulated by the PhoP/Q system that is important for antimicrobial peptide resistance, we hypothesized that MacAB of Salmonella would also support replication in the presence of amphiphilic cationic antimicrobial peptides.
We took a reductive approach to quantify the impact of macAB genotype on antimicrobial peptide resistance. We reasoned that the contribution of MacAB to antimicrobial peptide resistance could be obscured in vitro by the profound PhoP-induced membrane modifications that dramatically slow antimicrobial peptide interactions with the cell envelope . Furthermore, the constitutively expressed AcrAB-TolC pump effluxes a wide variety of compounds and can mask the contributions of other efflux pumps like MacAB during in vitro tests of antibiotic resistance [34,60]. Thus, we compared the contributions of MacAB variants to Salmonella growth in the presence of the antimicrobial peptide C18G using 4/74 phoP acrAB macAB null mutants harboring low-copy plasmids constitutively expressing macAST19macBST19 (the ST19 genotype), macAC➝TmacBST19 (the ST313 macA SNP alone), or macAC➝TmacBindel (the ST313 lineage 2 genotype). We found that the macAST19macBST19 plasmid facilitates growth of the 4/74 phoP acrAB macAB null mutant in minimal medium in the presence of 2μg/mL C18G (Fig 4A), while cells expressing macAC➝TmacBST19, macAC➝TmacBindel, or carrying empty plasmids show much longer lag times (Fig 4A). With a subinhibitory level of C18G treatment (1μg/mL), all strains grew equally well, indicating the differences in lag time are not due to toxicity from expression of MacAB variants (Fig 4B). Comparison of lag times (Fig 4, inset table) suggests that the macAC→T SNP impairs resistance to C18G while the macBindel further disables the MacAB-TolC channel. These data demonstrate that the ST19 macAB genotype assists growth in the presence of inhibitory concentrations of C18G and can do so independently of other PhoP-induced genes.
4/74 phoP acrAB macAB null mutants with low-copy pACYC177 plasmids constitutively-expressing macAB variants (pJH14-17, see S3 Table) were grown in N minimal medium, pH 7.4 and 1mM MgCl2. Overnight stationary phase cells were washed and normalized to OD600 = 1 before 1:200 final dilution into fresh N minimal medium with 2μg/mL C18G (A) or 1μg/mL C18G (B). OD600 was monitored over time using a BioTek Synergy HTX plate reader. Growth curves presented here are from one experimental run and representative of three independent experiments. Plotted data points are the geometric mean of quadruplicate or triplicate microplate wells. Lag time (inset table) was determined as time to reach OD600 = 0.15.
MacAB contributes to the ability of the S. Typhimurium ST19 isolate 4/74 to outcompete the iNTS ST313 isolate D23580 in the gut
Previous work has shown that MacAB promotes survival of Salmonella in the mouse gut after oral infection of C57BL/6 mice . In addition, gut colonization is known to induce transcription of PhoP/Q regulated genes in Salmonella . We thus hypothesized that the macAB mutations acquired by African S. Typhimurium ST313 isolates may impact fitness in the gut. To test this, we pretreated C57BL/6J mice with streptomycin and orally infected the next day with an equal mixture of 4/74 and the ST313 lineage 2 isolate D23580. We measured cecum and colon CFUs at day 2 after infection to calculate a competitive index (CI) and found that the ST19 strain 4/74 outcompeted D23580 by ~20-fold (Fig 5A). To test whether macAB contributed to this fitness difference, we competed macAB null mutants of 4/74 and D23580. The relative fitness of D23580 in gut tissues improved by approximately 5-fold when both isolates lacked macAB (Fig 5A), indicating that the ability of 4/74 to outcompete D23580 in the gut was partly macAB-dependent. To test whether the ST19 macAB genotype modulated D23580 fitness in the gut, we competed unmodified 4/74 with the D23580 macAST19macBST19 mutant. We found that altering macAB locus SNPs to the macAST19macBST19 genotype in D23580 improved its relative fitness in the cecum, colon, and feces when in competition with 4/74 (Fig 5B). These data show that the lower fitness of lineage 2 isolate D23580 in the mouse gut is partly due to its macAB genotype.
C57BL/6J mice (Nramp1-/-) were orally gavaged with streptomycin one day prior to oral infection with 5x107 CFU each of a mixture of 4/74 and D23580 in PBS. Tissues were isolated for CFU determination and calculation of competitive index at day 2 after infection. (A) Competitive index of unmodified D23580 (“+”) versus unmodified 4/74 (“+”) compared to the competitive index of D23580 macAB null (“‒”) versus 4/74 macAB null (“‒”). (B) Competitive index values of 4/74 versus D23580, or 4/74 versus D23580 macAST19macBST19. Two-way repeated measures ANOVA on log-transformed data with Sidak’s post-test within organ, p-values as indicated in (A, B). (C) Competitive index of D23580 versus 4/74 and their T3SS-1 (orgA null) or T3SS-1/2 (orgA ssaV null) mutants. Two-way repeated measures ANOVA on log-transformed data with no significant differences in (C). (D) Competitive index of D23580 versus 4/74 T3SS-1/T3SS-2 mutants (orgA ssaV null) compared to competition of T3SS-1/2 mutants with full deletions of macAB (orgA ssaV macAB null) in each isolate. t-test on log-transformed data from the cecum in (D). Dots are competitive index values for individual mice with lines plotted at the geometric mean. See S1 Fig for CFU/g values.
For S. Typhimurium, the type-3 secretion systems (T3SS) encoded by the SPI-1 and SPI-2 pathogenicity islands are important for inducing colitis during gut infection [62,63]. Previous work has shown S. Typhimurium ST313 isolates from Africa induce less host inflammation due to lower SPI-1-mediated invasion activity and reduced levels of flagellin expression when compared to ST19 strains [19,26,64]. Since macAB was shown to play a role in the streptomycin pre-treatment model of colitis , we suspected that 4/74 could be using SPI-1 and/or SPI-2-induced inflammation in the gut to outcompete D23580. Accordingly, we competed 4/74 and D23580 strains that lacked T3SS-1 (orgA null mutants) or T3SS-1/T3SS-2 (orgA ssaV null mutants). The competitive advantage of 4/74 over D23580 was not dependent upon the presence of T3SS-1 or T3SS-1/2 to induce gut inflammation (Fig 5C). However, there was a significant ~2.9-fold improvement in relative fitness of D23580 orgA ssaV macAB null versus 4/74 orgA ssaV macAB null mutants in the cecum compared to their orgAssaV null mutants with unmodified macAB (Fig 5D). These data demonstrate that macAB can influence relative fitness of D23580 versus 4/74 in the gut in the absence of T3SS-1/T3SS-2 induced inflammation.
Host genetics shape the utility of macAB genotypes for D23580 systemic infection
Since we saw no impact of the ST19 macAB genotype on D23580 replication in RAW264.7 cells, we predicted that ST313 macAB-associated SNPs might be dispensable or otherwise have no impact on the ability of D23580 to spread to systemic tissues. To exclude the contributions of known differences between ST19 and African ST313 in their inflammatory and disseminating behaviors in vivo [19,24,26,64], we competed D23580 with isogenic D23580 macAB mutants. To calculate a competitive index, we created a marked D23580 strain by placing a kanamycin resistance cassette at an intergenic site on the D23580 chromosome (see Materials and Methods) that was chosen for its low transcriptional activity when assessed by RNA-seq under a variety of growth conditions. After oral infection of C57BL/6J mice with the marked D23580-KanR strain and the parent D23580, we observed a competitive index of 1 in the gut and systemic sites, showing that the intergenic kanamycin marker insertion did not affect fitness in this model (Fig 6A). We found that the D23580 strain modified to the macAST19macBST19 genotype was outcompeted more than 4-fold by the isogenic D23580-KanR strain in systemic sites (liver and spleen), while showing no difference in the cecum (Fig 6A). This implies the macAC➝TmacBindel genotype provides an advantage to D23580 during systemic infection.
Isogenic competition of a kanamycin-marked D23580 with parent D23580 or D23580 macAST19macBST19 in (A) Nramp1-/- C57BL/6J mice or (B) Nramp1+/+ C57BL/6J mice, day 3 post oral infection after streptomycin pre-treatment. Two-way ANOVA with repeated measures on log-transformed data using Sidak’s post-test within organ, p values as indicated. Dots are competitive index values for individual mice with lines plotted at the geometric mean. See S2 Fig for CFU/g values.
One important hypothesis regarding the emergence of African iNTS lineages is that the immune status of certain human populations in sub-Saharan Africa provides a permissive niche for systemic S. Typhimurium infection, thus uniquely shaping Salmonella evolution [15,65–67]. In our infection experiments with RAW264.7 macrophages (BALB/c origin) and C57BL/6J mice, the host gene Nramp1 is not functional. BALB/c and C57BL/6J mice have a Glycine to Aspartic Acid (G169D) mutation in Nramp1 (i.e. a genotype of Nramp1D169/D169) which inactivates Nramp1 to yield a more permissive environment for Salmonella replication within macrophages . Nramp1 dramatically restricts intracellular bacterial infection at systemic sites through the removal of magnesium and other divalent cations from the vacuolar environment [68–72]. However, Nramp1 has further influence on the host immune response. For example, Nramp1+/+ mice have more rapid innate responses than isogenic Nramp1-/- mice, with higher levels of interferon γ (IFN-γ) and increased influx of neutrophils during the streptomycin pretreatment model of Salmonella-induced colitis  and in the dextran sodium sulfate (DSS)-induced colitis model . Furthermore, Nramp1+/+ dendritic cells produce more inflammatory cytokines than Nramp1-/- dendritic cells during Salmonella infection , an important route for the early, rapid dissemination of D23580 into the mesenteric lymph nodes . These published data suggest that, in addition to the role of Nramp1 in control of intramacrophage replication of Salmonella, many relevant parameters of the immune response differ between Nramp1-/- and Nramp1+/+ mice, especially in the amount of IFN-γ produced . Furthermore, IFN-γ, in conjunction with other stimuli, induces maximal upregulation of Nramp1 transcription . Thus, Nramp1 genotype has pleiotropic effects on the course of Salmonella infection. Given our results in a mouse background with defective Nramp1 (Fig 6A), we thus sought to test the effects of D23580 macAB genotypes during infection in mice with a more robust immune response.
We performed competitive oral infections after streptomycin pretreatment in resistant Nramp1G169/G169 C57BL/6J mice (hereafter, Nramp1+/+ mice). When we competed D23580-KanR with either the parent D23580 or D23580 macAST19macBST19 in Nramp1+/+ mice, systemic loads in liver and spleen at day 3 were equivalent, regardless of macAB genotype (Fig 6B). However, in this Nramp1+/+ host environment the macAST19macBST19 genotype did confer a fitness advantage to D23580 in the cecum (Fig 6B). This fitness advantage for the macAST19macBST19 genotype on the D23580 background is analogous to our observations in gut tissues when competing D23580 with 4/74 in C57BL/6J mice that are Nramp1-/- (Fig 5B).
Overall, our murine infection and competition experiments show that S. Typhimurium ST313-associated macAB SNPs and indels influence Salmonella fitness depending, in part, on the host Nramp1 genotype. During gut infection, the ST19 macAB genotype provides a fitness benefit to D23580 in competition with an ST19 isolate in permissive Nramp1-/- mice (Fig 5). We also show that the impact of the ST313 lineage 2 macAB genotype upon systemic colonization by D23580 is Nramp1-dependent (Fig 6A).
We have provided new insights into the function of the MacAB-TolC channel in Salmonella Typhimurium pathogenesis by exploring the consequences of variations in the 5’-UTR and coding sequence of macAB that are found within ST313 lineages associated with invasive disease in Africa. We show that macAB expression is influenced by PhoP and that the ST19 macAB genotype improves functional resistance to the antimicrobial peptide C18G. ST313-associated variations in macAB genotype can affect antimicrobial peptide resistance as well as replication in a permissive macrophage-like cell line. In mice, the ST313-associated macAB gene variants contribute to the lower relative fitness of ST313 lineage 2 isolate D23580 in competition with the ST19 strain 4/74 in inflamed and uninflamed settings in the gut.
Previously, macAB mutants of S. Typhimurium were reported to show impaired oxidative stress resistance . However, S. Typhimurium macAB null mutants of 4/74 did not show survival defects after peroxide treatment when compared to the parent strain (S3A Fig). Furthermore, sub-lethal concentrations of hydrogen peroxide did not induce transcription of macAB after short-term in vitro exposure (Fig 2A) or in time course assays with the 4/74 macAB::lacZY transcriptional reporter strain (S3B Fig). Poly-specificity allows efflux pumps to take on various functional roles since net cellular resistance to toxic compounds results from redundancies in efflux pumps and dynamic properties of the outer membrane . Given the poly-specificity of efflux channels like MacAB-TolC, it is likely that, in addition to antimicrobial peptides, some bacterially-generated molecules are exported to directly quench reactive oxygen species (ROS), as proposed by Bogomolnaya and colleagues [38,78]. However, this oxidative stress resistance function of MacAB may depend upon experimental conditions and the bacterial genetic background.
We thus do not exclude a direct role for macAB in oxidative stress resistance in Salmonella but propose that an additional function is to counteract the effects of antimicrobial peptides encountered in both the gut and systemic sites. Indeed, bacterial responses to oxidative stress and antimicrobial peptide exposure are tightly linked by the PhoP/Q two-component regulatory system [79,80]. Antimicrobial peptide exposure induces transcription of genes in the RpoS regulon that assist bacterial adaptation to oxidative stress [81,82]. We have shown that macAB variants confer distinct levels of antimicrobial peptide resistance independently from other PhoP-mediated effects. Notably, on its own the macAC➝T SNP which leads to the S174L mutation in MacA reduces the utility of MacAB in resistance to C18G (Fig 4). The role played by the PhoP/Q system in upregulating transcription of macAB is consistent with the MacAB-TolC channel of Salmonella countering antimicrobial peptide exposure, either through direct efflux or by assisting translocation of other bacterial factors important to cope with antimicrobial peptide stress.
Although one publication suggested PhoP repressed macAB transcription in Salmonella after low magnesium treatment , other published data suggest that macAB is positively regulated by PhoP. We have previously noted that phoPQ null mutants have lower expression of macAB when grown in InSPI2 medium , while others have seen that S. Typhi production of MacA protein under low magnesium conditions is PhoP-dependent . Furthermore, macAB is important for S. Typhimurium replication in macrophages [38,56], a niche where PhoP/Q signaling is active and important for Salmonella survival [55,61]. In addition to the different Salmonella isolates used across studies, our results suggesting PhoP promotes macAB transcription may arise from differences in experimental conditions, the importance of regulatory kinetics such as PhoP turnover , changes in ribosomal RNA levels in response to low magnesium , or coordinated activity with other transcriptional regulators . We conclude that PhoP-inducing signals generally serve to maximize macAB transcription in Salmonella.
The macAB locus contains additional regulatory complexity that requires further investigation. While we show that the 5’-UTRmacALin2.1 SNP is sufficient to impair replication of 4/74 in macrophages (Fig 3A), the mutation did not affect the level of macAB transcription in low magnesium medium (S4 Fig). The mapped transcriptional start site (TSS) of macA is 504 nucleotides upstream of the start codon and is located within the divergently-transcribed gene ybjX (also known as somA) (see Fig 1B and https://tinyurl.com/macAB-Jbrowse). Long 5’-UTRs have been shown to have trans-acting regulatory roles in expression of virulence genes in both S. Typhimurium [43,85] and S. Typhi . Such 5’-UTR elements can post-transcriptionally regulate gene expression through formation of RNA secondary structures that control interactions with small regulatory RNAs, or by directly influencing transcript stability . The 5’-UTRmacALin2.1 SNP in particular is near the ribosome binding site of the macA transcript which could impact translation initiation (see Fig 1B). We observed that expression of MacAB from a high-copy plasmid (pUC19) or a plasmid with an arabinose-inducible promoter is toxic to cells, a lethality phenomenon reported by others . Additionally, MacA exhibits high binding affinity for TolC in vitro, suggesting it may outcompete other TolC-interacting efflux pumps whenever MacAB is expressed . Taken together, these observations support the conclusion that the tight transcriptional regulation of macAB functions to provide utility without compromising other cellular functions.
We have not demonstrated the direct structural consequences for the MacAB-TolC channel that result from the macAB SNPs present in ST313 isolates. Such changes might include assembly kinetics of the MacA hexamer, stability or mechanical function of the tripartite complex, or perhaps altered association with other important constituents of the Gram-negative cell envelope . The E. coli MacA protein can bind tightly to rough LPS via residues on the N-terminus with affinity that is higher than that of polymyxin B, prompting the suggestion that rough LPS may be a cargo of the channel . Given that PhoP is required for maximal macAB transcription and that MacAB-mediated resistance to C18G did not require other PhoP-regulated genes (Fig 2B and 2C), we speculate that rough LPS-binding by MacA is a structural feature important for its antimicrobial peptide resistance function. While the ST313 macAC➝T SNP resulting in the S174L mutation impairs MacAB-mediated resistance to C18G, it is possible this SNP variant improves efflux of other substrates not tested here and thus might serve a direct role in the pathogenic adaptation of African iNTS. Further structural investigation of the Salmonella MacAB-TolC channel will provide new insights into the role of this important efflux pump in Salmonella physiology and host-pathogen interactions.
It is not immediately clear why S. Typhimurium ST313-associated macAB gene variants lead to different infection outcomes depending on strain background; however, some possibilities may be proposed based on current knowledge. In the case of D23580 and other ST313 lineage 2 isolates, the mutation responsible for the marked increase in expression of the PhoP-regulated protease PgtE  could sufficiently compensate for antimicrobial peptide resistance otherwise provided by the ST19 macAB genotype. In this case, the ATP-dependent MacAB-TolC channel might siphon energy or even directly interfere with other genes or behaviors unique to the D23580 strain background. Such loss-of-function mutations that favor the activity of other pathogenesis mechanisms have been noted in Salmonella evolution. For Salmonella Typhi, a stop codon that interrupts fepE prevents very-long O-antigen production , which in turn permits the horizontally-acquired Vi capsule to have maximal immune evasive effects during systemic infection . Finally, it has been shown that point mutations in the highly expressed AcrAB-TolC efflux pump can alter expression of a variety of Salmonella pathogenic genes indirectly without affecting expression of other efflux pumps . Further research will be required to identify other Salmonella genes that might interact with the macAB genotype to influence pathogenesis.
In single infections, S. Typhimurium D23580 and other ST313 isolates readily colonize the intestinal tracts of a variety of animals including primates, rodents, and chickens [23,24,92–94]. A recent study of competitive fitness in the streptomycin-pretreatment model of mouse colitis showed that D23580 can outcompete the ST19 isolate IR715, a strain that is derived from the common laboratory reference isolate 14028S , further confirming the ability of ST313 to infect multiple animal species. We show that differences in macAB genotype are partly responsible for a relative fitness defect of D23580 during competition with the ST19 isolate 4/74 in the inflamed and uninflamed gut (Fig 5A, 5B and 5E). Several factors could contribute to these differences in competitive fitness of D23580 versus certain ST19 strains. Both D23580 and IR715 do not express sopE, a SopEΦ prophage-encoded virulence factor delivered by the T3SS of SPI-1 [19,26]. In contrast, sopE is present in 4/74 and the closely-related ST19 isolate SL1344 . sopE is known to drive substantial cecal inflammation in the streptomycin pretreatment model of colitis . Further, the ST19 strains SL1344 and 4/74 contain the pCol1b plasmid and can produce colicin Ib, potentially helping Salmonella outcompete other Enterobacteriaceae in the inflamed gut . We found that ST19-associated macAB gene variants increase fitness of D23580 in scenarios with robust gut inflammation, such as in competition with a S. Typhimurium ST19 isolate that has sopE  and in mice with Nramp1 functionality . However, when D23580 infects a more permissive host, such as Nramp1-/- C57BL/6J mice, its ST313 lineage 2 macAB variant is advantageous for systemic infection (Fig 6B). Thus, D23580 shows a degree of specialization for systemic infection that depends upon both its macAB genotype and parameters of the host innate immune response. Our experiments reinforce the value of testing Salmonella isolates in various host genotypes to identify potentially important host-pathogen gene interactions.
Although no studies have specifically linked NRAMP1/SLC11A1 polymorphisms to the susceptibility of humans to invasive Salmonella, several NRAMP1 polymorphisms found within West African populations are strongly linked to susceptibility to intracellular infection with Mycobacterium tuberculosis [98,99]. We note that numerous parameters in mammalian hosts that are related to susceptibility to invasive Salmonella disease are also compromised in Nramp1-deficient mice, as summarized below. For example, macaques that were previously infected with simian immunodeficiency virus have diminished IL-17 production that reduces the influx of neutrophils into the gut, thus compromising control of Salmonella dissemination from the gut ; similarly, neutrophil influx into the gut during Salmonella colitis is delayed in Nramp1-/- mice . Nramp1 also regulates the oxidative burst capacity of phagocytes . A reduced oxidative burst response is observed in Gambian children recovering from malaria infection  and is likely a consequence of malaria-induced hemolysis as revealed by experiments in mice .
In humans, STAT4 variants have been linked to increased risk for invasive Salmonella infection. Specifically, a genome-wide-association study identified a STAT4 polymorphism in children in Kenya and Malawi which correlated with reduced IFN-γ responses of ex vivo stimulated immune cells . The importance of IFN-γ to control of Salmonella infection at both acute and chronic stages has been clearly demonstrated , and upregulation of NRAMP1 in macrophages is part of the response to IFN-γ . Malaria, malnutrition, and HIV can compromise innate resistance to intracellular bacterial infection through a wide variety of mechanisms, including alterations in macrophage responsiveness to IFN-γ . Based on outcomes of our competitive infections using Nramp1-/- and Nramp1+/+ mice, we speculate that the S. Typhimurium ST313-associated macAB gene variants represent adaptive evolution for systemic infection when restriction of Salmonella dissemination is compromised in the host.
Oral infection of mice with Salmonella Typhimurium after streptomycin pretreatment causes substantial inflammation and colitis, which depends on the function of SPI1 and SPI2-encoded type three secretion systems . Although D23580 exhibits less T3SS-1 mediated invasion in vitro and in animal infections [19,26], we found that D23580 was outcompeted by 4/74 in the mouse gut whether or not both isolates lacked T3SS-1 or T3SS-1/2, and this fitness defect was partly macAB-dependent (Fig 5D and 5E). Although all of these experimental models vary in degree and type of inflammation, they share a diverse and dynamic set of host- and microbe-generated antimicrobial peptides. We suggest that MacAB contributes to Salmonella pathogenesis by countering antimicrobial peptide stress as part of the bacterial response orchestrated by the PhoP/Q two-component system.
The S. Typhimurium ST313-associated macAB variants we have characterized suggest a pattern of evolutionary convergence toward a degraded function of the MacAB-TolC efflux pump. While MacAB remains functional amongst S. Typhimurium lineages associated with gastroenteritis, we conclude that inactivation of the MacAB system within African ST313 lineages represents a unique adaptation that may facilitate systemic infection of permissive hosts.
Materials and methods
Vertebrate animal ethics statement
All animal experiments were approved by the Stanford University Administrative Panel on Laboratory Animal Care (APLAC) with oversight by the Institutional Animal Care and Use Committee (IACUC) under local Protocol ID 12826. Animals were housed at specified-pathogen free (SPF) level in University facilities accredited by the Association of Assessment and Accreditation of Laboratory Animal Care (AAALAC) International.
Phylogenetic tree construction and macAB status visualization
The sources of all the genomic data files used are listed in S1 Table. The assembled genomes of 18 S. Typhimurium strains, including the ST19 representative strain 4/74  and ST313 representative strain D23580 , were obtained from GenBank, while the raw sequencing data of 267 S. Typhimurium ST313 strains derived from previous publications [12,13,15,16,109] were downloaded from SRA (https://www.ncbi.nlm.nih.gov/sra/) and EMBL-EBI (https://www.ebi.ac.uk/) databases. To root the phylogenetic tree, the genome of Salmonella Typhi CT18 was downloaded from GenBank and used as the outgroup.
SNIPPY v4.4.0  was used to map the sequencing data against 4/74, call the SNPs, construct pseudo-genomes, and make a genome alignment. SNIPPY used Freebayes  as the variant caller, the default parameter of minimal coverage was 10, and the minimal fraction was 0. For the assemblies from GenBank, SNIPPY used contigs as the input. Recombinant regions were detected and removed from the alignment using Gubbins v2.4.1 . RAxML-NG v0.9.0  was used to build a phylogenetic tree, with substitution model GRT+G. The phylogenetic tree was visualized on iTol  (https://itol.embl.de/). Based on the phylogeny and prior publications, the ST313 strains were classified into UK ST313 , Brazil ST313 , lineage 1 , lineage 2 , lineage 2.1 , and lineage 2.2 .
The macA 5'-UTR, macA, and macB sequences of strain 4/74 were used to generate BLAST databases with BLAST 2.9.0+ . Raw reads of all the strains were assembled using Unicycler v0.4.8 . The quality of assembly was checked by Quast v5.0.2 . The N50 value of all assemblies was >20kb, and the number of contigs was <600. The genome assemblies were queried against the databases using the BLASTn algorithm.
Bacterial strains, plasmids, primers and growth conditions
See S2 Table for bacterial strains used in this study. Where applicable, P22 HT105/1 int-201 phage was used to move marked mutations from previously-generated S. Typhimurium mutants into the 4/74 strain according to standard protocols. Salmonella were routinely grown in LB Lennox (10g/L tryptone, 5g/L yeast extract, 5g/L NaCl) for cloning manipulations or in LB Miller (10g/L NaCl) for infection assays with the following antibiotic concentrations: streptomycin, 200μg/mL; chloramphenicol, 25μg/mL; tetracycline, 15μg/mL; kanamycin, 40μg/mL; and gentamicin, 25μg/mL (GoldBio).
InSPI2 and intramacrophage RNA-seq data
Lambda (λ) red recombination for marked mutants
Marked mutants were made based on the λ red procedure  using the temperature sensitive pSIM5-Tet plasmid that contains a temperature shock inducible promoter driving recombinase expression . Briefly, primers with 40bp homology targeting the flanking regions of gene to be deleted were used to amplify the FRT-flanked kanamycin cassette from pKD4. 4/74 or D23580 / pSIM5-Tet cells were grown with streptomycin and tetracycline at 30°C overnight in LB Lennox before subculturing 1:100 into fresh mediumand growing to an OD of 0.4. The culture was then incubated at 42°C shaking for 15 minutes before placing on ice make cells competent for electroporation. Cells were washed with chilled double-distilled water, followed by electroporation of 800ng of purified PCR product. Cells were recovered in SOC for 1.5 hours shaking at 30°C before pelleting and plating on LB kanamycin plates. After overnight growth at 37°C, individual colonies were struck across fresh plates to purify single colonies, followed by colony PCR to confirm correct insertion of the kanamycin cassette. For 4/74, marked mutants were moved into a clean 4/74 strain background using P22 HT105/1 int-201 phage, and the correct insertion was confirmed by PCR.
For construction of D23580-KanR we amplified the kanamycin resistance cassette from pKD4 using primers del_23_F and del_23_R for insertion by λ red recombination as described above. These primers target the KanR cassette to the intergenic region between STMMW_41451 and STMMW_41461 between coordinates 4441510 and 4441511 in D23580 (GenBank: FN424405.1). This region is part of the remnant prophage Def4  and not transcribed in RNA-seq datasets in variety of in vitro conditions . Correct insertion of the KanR cassette was confirmed by primers Fw_STM4196-7_ext and Rv_STM4196-7_ext. D23580-KanR was confirmed to be cured of temperature sensitive pSIM5-Tet after growth at 37°C by testing isolated colonies for tetracycline sensitivity.
Construction of lacZY reporter strains
Chromosomal β-galactosidase (lacZY) transcriptional fusions were made using the method of Ellermeier and colleagues . 4/74 marked mutants with FRT-flanked kanamycin cassettes derived from pKD4 were electroporated with 200ng pCP20, recovered for 1 hour in SOC at 30°C before plating dilutions on LB streptomycin chloramphenicol plates, followed by growth at 30°C overnight. Colonies were re-struck to purify for single clones, then patched onto LB kanamycin and LB streptomycin chloramphenicol plates to confirm loss of kanamycin by FLP-recombinase activity. Two kanamycin sensitive colonies were picked and grown in LB Lennox at 30°C, followed by electroporation with 200ng pCE36 plasmid purified from the Salmonella pir+ strain JS198/pCE36 . Cells were recovered in SOC at 37°C and plated on LB kanamycin to select for pCE36 integration. Purified colonies were screened by PCR using P1 and Lac primers for integration of pCE36 into the FRT site .
Scarless mutant generation
In order to study the influence of these nucleotide changes on Salmonella pathogenesis without otherwise altering macAB regulation or introducing other changes in the genome, we used scarless mutagenesis to make nucleotide changes in the genome of a given Salmonella isolate. We used the suicide plasmid pEMG system described by Martínez-García and de Lorenzo  and as applied to Salmonella Typhimurium by Owen and colleagues . Genomic DNA from 4/74 or D23580 was used as template for Phusion (ThermoFisher) PCR of 1600bp flanking the polymorphism to be transferred. For marker-less deletions, 800bp flanking each side of the region to be deleted were amplified to generate two PCR products. Primers were designed to include ~20nt overhangs to permit Gibson Assembly of a single 1600bp fragment (for marker-less nucleotide changes) or the two 800bp flanking regions (for marker-less deletions) into pEMG that was previously digested with XbaI and KpnI-HF (New England Biolabs, NEB). Purified PCR product (Qiagen) and digested pEMG backbone were assembled with the DNA HiFi Assembly MasterMix (NEB). The 5’-UTRmacALin2.1 SNP was incorporated into pEMG::macAST19 or pEMG::macAC➝T by PCR using primer pairs 1309,1310 and 1311,1312; the two PCR products were assembled with HiFi DNA Assembly MasterMix with pEMG previously digested with XbaI and KpnI-HF. A given pEMG::X plasmid was mobilized from E. coli S17-1 λpir into recipient Salmonella by conjugation, followed by selection for transconjugants on M9 minimal agar plates formulated with 0.2% glucose, 1mM MgSO4, and 40μg/mL kanamycin. Merodiploid transconjugants were resolved by electroporation of pSW-2 as previously described , followed by colony PCR and standard Sanger sequencing of PCR product to identify clones with the intended point mutations. Confirmed mutants were cured of unstable pSW-2 by several passages in LB before patch plating to confirm loss of pSW-2 by gentamicin sensitivity.
To confirm that key strains contained the intended engineered nucleotide(s) and no unintended mutations in other parts of the genome, the four strains 4/74 macAC➝T, 4/74 macAC➝TmacBindel, D23580 macAC➝TmacBST19 and D23580 macAST19macBST19 were genome-sequenced using Illumina paired-end sequencing (SNPsaurus, Oregon) as indicated in S2 Table and aligned to reference genomes (NCBI) using CLC Genomics Workbench (Qiagen).
The macAB sequence was amplified from genomic DNA templates with Phusion polymerase (ThermoFisher) using primers 1223 and 1224. The PCR product and pBAD33.1 (Addgene #36267) were digested with NdeI and HindIII-HF (New England Biolabs, NEB) followed by T4 ligation and transformation into NEB 10beta competent cells. The cassette inclusive of the T7 ribosome binding site, macAB, and the two transcriptional terminators was amplified with Phusion polymerase using primers 1272 and 1273 from pBAD33.1-macAB templates. The PCR products and destination pACYC177 were digested with ScaI and PstI, purified using the PCR purification kit (Qiagen), and ligated with T4 polymerase before transformation into NEB 10beta competent cells, selecting on kanamycin. The promoter for the AmpR gene thus drives expression of the macAB gene inserted at the ScaI site within the AmpR sequence of pACYC177.
Stimulation conditions for PhoP-regulated gene expression
The procedures here are based on methods described by Bader and colleagues . For low-magnesium treatment, cells were grown at 37°C overnight in N minimal medium with 100mM Tris-HCl pH 7.4, 0.2% deferrated Casamino acids (Chelex treated), 0.2% glycerol, and 10mM MgCl2 (high magnesium) with relevant antibiotics. Cells were subcultured 1:100 into fresh medium and grown to OD600 ~0.2, or about 4 hours. Cells were washed 3x in the same medium or medium with 10μM MgCl2 (low magnesium) and returned to the incubator, growing for a further 90 minutes. Where applicable, N minimal medium was buffered with 10mM MES at pH 5.8 or 4.9. For β-galactosidase assays, 1 mL of cell culture was then transferred to a microcentrifuge tube and placed on ice, followed by one wash in 1mL pre-chilled 100mM phosphate buffer, pH 7. A 150μL aliquot was removed and the OD600 was checked on a microplate reader (Synergy HTX, BioTek) and recorded. The aliquot was returned to the original tube and cells were pelleted again in a chilled centrifuge. The supernatant was removed and the pellet was frozen immediately on dry ice and transferred to -80°C storage.
For C18G stimulation, cells were grown overnight in N minimal medium, pH 7.4 (see above) formulated with 1mM MgCl2. We titrated C18G (Anaspec) to confirm a concentration that did not inhibit growth, identifying 5μg/mL as optimal, similar to what was reported by Bader and colleagues . Cells were subcultured 1:100 and grown to OD ~0.2 before pelleting and resuspending in fresh media with or without 5μg/mL C18G. Cells were grown for a further 90 minutes and processed as described above.
We used a modified lysis protocol and a kinetic microplate assay to measure β-galactosidase activity based on the methods described by Schaefer and colleagues  and Thibodeau and colleagues . Bacterial pellets were removed from -80°C to room temperature, thawed briefly, then resuspended in 200μL of 100mM phosphate buffer, pH 7.0. A 200μL mixture of 20% PopCulture (Millipore) and 8U/μL of rLysozyme (Millipore #71110–4, Lot 3277983, 30U/μL) was added to each sample and vortexed for 5 seconds. After 5 minutes at room temperature, the sample was vortexed again and incubated for a further 5 minutes. During lysis optimization tests, this freeze/thaw and lysis buffer treatment achieved a >95% reduction in OD405 within two minutes, and the suspension was visually clear within 20 seconds. In each well of a 96 well microplate, 70μL of Z-buffer with 0.05M β-mercaptoethanol was added. After lysis, 80μL of sample was added in quadruplicate wells, with control wells having lysis buffer only; a lysed pellet of wild-type 4/74 cells with no lacZY reporter served as an additional control. Using a multichannel pipette, 30μL of ONPG (Sigma) previously dissolved in 100mM phosphate buffer at 4μg/mL was added to each well. The microplate was transferred immediately to a Synergy HTX (BioTek) plate reader that was pre-equilibrated to 28°C. OD420 was measured every minute for 90 minutes, incubating at 28°C with shaking between reads. The change in absorbance at 420nm over time was calculated from the slope of the reads and used in the standard Miller unit calculation, where OD600 is the density of cells in 1mL before pelleting and freezing; 2.5 corrects for the pellet resuspension in a 400μL lysis volume; and 0.08 is the volume of sample (in mL) assayed per well.
Antimicrobial peptide sensitivity assay
4/74 phoP acrAB macAB null mutants with indicated pACYC177 plasmids were grown aerobically overnight in N minimal media, pH 7.4 (100mM Tris-HCl) with 1mM MgCl2, 0.2% Casamino acids, and 0.2% glycerol with appropriate antibiotics. Cells were normalized to OD600 of 1 = 1x109 cells/mL, then diluted 1:100 into fresh medium without antibiotics. 75μL of fresh medium alone or 2x final concentration of C18G (Anaspec, Fremont, CA) in fresh medium were added to wells of a sterile, polypropylene 96-well flat-bottomed plate (Griener BioOne, Product 655261). 75μL (~7.5x104 cells) of the 1:100 bacterial suspension was added to wells in quadruplicate for a final 1:200 dilution of cells, with final μg/mL C18G as indicated. C18G was titrated previously, with higher concentrations (≥3μg/mL) preventing growth of all strains while treatment at concentrations ≤1.5μg/mL showing no differences in growth kinetics compared to untreated wells. The plate was sealed with a Breathe-Easy gas permeable film (Diversified Biotech) and OD600 was measured every ten minutes while growing at 37°C with linear shaking using a Synergy HTX plate reader (BioTek). Lag time was calculated as the time to reach OD600 = 0.150.
Macrophage infection assays
RAW264.7 cells were passaged in DMEM + 10% FBS with 4.5g/L glucose and 110mg/L sodium pyruvate and L-glutamine. For infection, cells were seeded at 2.5x104 cells per well in a 96 well plate starting 24 hours prior to infection. Single Salmonella colonies were picked from recently struck LB plates and grown overnight in LB Miller with appropriate antibiotics. After 16–18 hours of culture, cells were pelleted, washed once in PBS, and concentration was determined by OD600 of 1 = 1x109 CFU/mL. Bacteria were diluted to 1x108/mL and added to DMEM complete medium for a final 2.5x106 CFU/mL concentration. Medium was aspirated from the RAW cells with a multichannel and replaced with 50μL of fresh media. 100μL of each inoculum (2.5x105 CFU, MOI = 10) was added in quadruplicate for each time point (0, 8 and 20 hours). Plates were spun at 300g for 10 minutes at room temperature (21–25°C), then incubated at 37°C with 5% CO2 for 30 minutes to allow phagocytosis. Meanwhile, the inoculum was diluted 10-fold in PBS and plated to confirm the MOI. After 30 minutes, the medium was then aspirated and RAW cells were gently washed twice with 100μL medium, replacing finally with 100μL of DMEM complete with 100μg/mL gentamicin. The RAW cells were incubated at 37°C for 60 minutes, then the high gentamicin medium was removed and replaced with medium with 20μg/mL gentamicin to suppress extracellular growth of Salmonella for the remainder of the culture period (this was t = 0). At each timepoint, medium was aspirated and RAW cells were washed 3x with 150 μL PBS before lysis in 30μL of 1% Triton X-100 at room temperature. After 5 minutes, the wells were pipetted vigorously with a multichannel, then each well was topped with 120μL of PBS. The well suspensions were mixed, serially diluted in PBS, and spot-plated to calculate CFUs per well.
C57BL/6J (Stock Number 000664), mice were purchased from Jackson Laboratories. C57BL/6J Nramp1G169/G169 mice were previously described .
Streptomycin pre-treatment model of colitis
Mice 7–12 weeks old were deprived of food briefly for 4 hours prior to gavage with 20mg streptomycin in sterile water. The next day, overnight stationary phase cultures of Salmonella strains grown with appropriate antibiotics were pelleted and washed twice in PBS before cell quantification by OD600. Twenty hours after streptomycin treatment, mice were again briefly deprived of food for 4 hours, followed by gavage of an equal mixture in PBS of 5x107 CFU of each strain for a total inoculum of 1x108 CFU. The inoculum was diluted and plated onto LB agar plates with appropriate antibiotics to differentiate strains and calculate the input ratio. Mice were euthanized by CO2 asphyxiation at a given timepoint, tissues were removed and homogenized in sterile PBS, then diluted and plated onto LB agar plates with different antibiotics to distinguish Salmonella strains. Total Salmonella CFUs were determined from streptomycin plates, while the D23580 proportion was calculated from streptomycin and chloramphenicol plates. In the isogenic D23580 competition, the addition of kanamycin identified the proportion of D23580-KanR colonies from the total D23580 population. The output ratio in tissues was divided by the input ratio of the inoculum to compute the competitive index.
S1 Fig. CFU counts corresponding to Competitive Index (CI) values of cecal tissue samples plotted in Fig 5.
Fig S1A-D correspond to Fig 5A–5D, respectively. Connecting lines show paired values of CFU per gram from the cecum of an individual mouse.
S2 Fig. CFU counts corresponding to Competitive Index (CI) values of spleen and cecal tissue samples plotted in Fig 6.
Fig S2A-B correspond to Fig 6A and 6B, respectively. Connecting lines show paired values of CFU per gram from the spleen or cecum of an individual mouse.
S3 Fig. macAB does not assist in vitro peroxide resistance in the ST19 strain 4/74.
(A) Survival of 4/74 macAB null mutant after peroxide treatment. 4/74, 4/74 macAB null and 4/74 rpoS::Kan were grown in LB Miller overnight with appropriate antibiotics, normalized to OD600 = 1 before 1:100 dilution into fresh LB with or without 1mM H2O2, growing at 37°C while rotating. Cells were removed hourly and serial dilutions plated to calculate percent survival in reference to CFUs at t = 0. (B) Transcriptional induction after peroxide exposure. 4/74 parent and the 4/74 macAB::pCE36 lacZY transcriptional fusion were normalized to OD600 = 1 after overnight culture in LB, followed by 1:100 dilution into fresh LB medium and growth while shaking at 37°C. At OD600 = 0.5 (~2 hours of growth), mid-exponential cells were pelleted and resuspended in the same volume of fresh LB with or without 0.5mM H2O2. Cells were removed every 30 minutes and assayed for β-galactosidase production as described in Materials and Methods.
S4 Fig. 5’-UTRmacALin2.1 SNP does not alter transcriptional response of macAB to low Mg2+.
Two clones of 4/74 macAB::pCE36 transcriptional fusion strains with the 5’-UTRmacALin2.1 SNP preceding macA were grown to mid-exponential phase in N minimal medium pH 7.4 with high Mg2+ (10mM) then shifted to the same or low Mg2+ (10μM) media and grown for 90 minutes. β-galactosidase activity was measured using a kinetic Miller assay as described in Materials and Methods. The 5’-UTRmacALin2.1 SNP was incorporated by λ red recombination using the primer pair 1288b, 1289 to amplify the KmR cassette from pKD4. Transcriptional fusions generated with pCE36 include an internal, independent ribosome binding site for translation of lacZY from the transcript.
We thank both current and recent members of the Monack, Hinton, and Amieva labs who offered helpful comments and critiques at various stages of this project. Yueh-hsiu Chien, Justin Sonnenburg, and Ami Bhatt provided insightful suggestions throughout stages of experimental design and data analysis. We extend thanks to Igor Brodsky and Rina Matsuda for technical advice on working with antimicrobial peptides. We additionally thank Sean-Paul Nuccio for P22 phage reagents and protocols for mutant generation. We are grateful to the animal husbandry staff and veterinarians at the Veterinary Service Center at Stanford University School of Medicine for their support.
- 1. Majowicz SE, Musto J, Scallan E, Angulo FJ, Kirk M, O’Brien SJ, et al. The Global Burden of Nontyphoidal Salmonella Gastroenteritis. Clin Infect Dis. 2010 Mar 15;50(6):882–9. pmid:20158401
- 2. Roth GA, Abate D, Abate KH, Abay SM, Abbafati C, Abbasi N, et al. Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980–2017: a systematic analysis for the Global Burden of Disease Study 2017. The Lancet. 2018 Nov 10;392(10159):1736–88.
- 3. Crump JA, Sjölund-Karlsson M, Gordon MA, Parry CM. Epidemiology, Clinical Presentation, Laboratory Diagnosis, Antimicrobial Resistance, and Antimicrobial Management of Invasive Salmonella Infections. Clin Microbiol Rev. 2015 Oct 1;28(4):901–37. pmid:26180063
- 4. Stanaway JD, Reiner RC, Blacker BF, Goldberg EM, Khalil IA, Troeger CE, et al. The global burden of typhoid and paratyphoid fevers: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Infect Dis. 2019 Apr 1;19(4):369–81. pmid:30792131
- 5. Stanaway JD, Parisi A, Sarkar K, Blacker BF, Reiner RC, Hay SI, et al. The global burden of non-typhoidal salmonella invasive disease: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Infect Dis. 2019 Sep 24;19(12):1312–24. pmid:31562022
- 6. Balasubramanian R, Im J, Lee J-S, Jeon HJ, Mogeni OD, Kim JH, et al. The global burden and epidemiology of invasive non-typhoidal Salmonella infections. Hum Vaccines Immunother. 2018 Aug 6;0(0):1–6.
- 7. Feasey NA, Dougan G, Kingsley RA, Heyderman RS, Gordon MA. Invasive non-typhoidal salmonella disease: an emerging and neglected tropical disease in Africa. The Lancet. 2012 Jul 6;379(9835):2489–99.
- 8. Feasey NA, Hadfield J, Keddy KH, Dallman TJ, Jacobs J, Deng X, et al. Distinct Salmonella Enteritidis lineages associated with enterocolitis in high-income settings and invasive disease in low-income settings. Nat Genet. 2016 Oct;48(10):1211–7. pmid:27548315
- 9. Kingsley RA, Msefula CL, Thomson NR, Kariuki S, Holt KE, Gordon MA, et al. Epidemic multiple drug resistant Salmonella Typhimurium causing invasive disease in sub-Saharan Africa have a distinct genotype. Genome Res. 2009 Dec 1;19(12):2279–87. pmid:19901036
- 10. Marks F, von Kalckreuth V, Aaby P, Adu-Sarkodie Y, El Tayeb MA, Ali M, et al. Incidence of invasive salmonella disease in sub-Saharan Africa: a multicentre population-based surveillance study. Lancet Glob Health. 2017 Mar;5(3):e310–23. pmid:28193398
- 11. Kariuki S, Okoro C, Kiiru J, Njoroge S, Omuse G, Langridge G, et al. Ceftriaxone-Resistant Salmonella enterica Serotype Typhimurium Sequence Type 313 from Kenyan Patients Is Associated with the blaCTX-M-15 Gene on a Novel IncHI2 Plasmid. Antimicrob Agents Chemother. 2015 Jun 1;59(6):3133–9. pmid:25779570
- 12. Okoro CK, Kingsley RA, Connor TR, Harris SR, Parry CM, Al-Mashhadani MN, et al. Intra-continental spread of human invasive Salmonella Typhimurium pathovariants in sub-Saharan Africa. Nat Genet. 2012 Nov;44(11):1215–21. pmid:23023330
- 13. Van Puyvelde S, Pickard D, Vandelannoote K, Heinz E, Barbé B, de Block T, et al. An African Salmonella Typhimurium ST313 sublineage with extensive drug-resistance and signatures of host adaptation. Nat Commun. 2019 Sep 19;10(1):4280. pmid:31537784
- 14. Panzenhagen PHN, Paul NC, Conte CA, Costa RG, Rodrigues DP, Shah DH. Genetically distinct lineages of Salmonella Typhimurium ST313 and ST19 are present in Brazil. Int J Med Microbiol. 2018 Mar 1;308(2):306–16. pmid:29396155
- 15. Ashton PM, Owen SV, Kaindama L, Rowe WPM, Lane CR, Larkin L, et al. Public health surveillance in the UK revolutionises our understanding of the invasive Salmonella Typhimurium epidemic in Africa. Genome Med. 2017 Oct 31;9:92. pmid:29084588
- 16. Almeida F, Seribelli AA, da Silva P, Medeiros MIC, dos Prazeres Rodrigues D, Moreira CG, et al. Multilocus sequence typing of Salmonella Typhimurium reveals the presence of the highly invasive ST313 in Brazil. Infect Genet Evol. 2017;41–4.
- 17. Canals R, Chaudhuri RR, Steiner RE, Owen SV, Quinones-Olvera N, Gordon MA, et al. The fitness landscape of the African Salmonella Typhimurium ST313 strain D23580 reveals unique properties of the pBT1 plasmid. PLOS Pathog. 2019 Sep 27;15(9):e1007948. pmid:31560731
- 18. Canals R, Hammarlöf DL, Kröger C, Owen SV, Fong WY, Lacharme-Lora L, et al. Adding function to the genome of African Salmonella Typhimurium ST313 strain D23580. PLOS Biol. 2019 Jan 15;17(1):e3000059. pmid:30645593
- 19. Okoro CK, Barquist L, Connor TR, Harris SR, Clare S, Stevens MP, et al. Signatures of Adaptation in Human Invasive Salmonella Typhimurium ST313 Populations from Sub-Saharan Africa. PLOS Negl Trop Dis. 2015 Mar 24;9(3):e0003611. pmid:25803844
- 20. Owen SV, Wenner N, Canals R, Makumi A, Hammarlöf DL, Gordon MA, et al. Characterization of the Prophage Repertoire of African Salmonella Typhimurium ST313 Reveals High Levels of Spontaneous Induction of Novel Phage BTP1. Front Microbiol. 2017;8(1):235.
- 21. Parsons BN, Humphrey S, Salisbury AM, Mikoleit J, Hinton JCD, Gordon MA, et al. Invasive Non-Typhoidal Salmonella Typhimurium ST313 Are Not Host-Restricted and Have an Invasive Phenotype in Experimentally Infected Chickens. PLOS Negl Trop Dis. 2013 Oct 10;7(10):e2487. pmid:24130915
- 22. MacKenzie KD, Wang Y, Musicha P, Hansen EG, Palmer MB, Herman DJ, et al. Parallel evolution leading to impaired biofilm formation in invasive Salmonella strains. PLOS Genet. 2019 Jun 24;15(6):e1008233. pmid:31233504
- 23. Singletary LA, Karlinsey JE, Libby SJ, Mooney JP, Lokken KL, Tsolis RM, et al. Loss of Multicellular Behavior in Epidemic African Nontyphoidal Salmonella enterica Serovar Typhimurium ST313 Strain D23580. mBio. 2016 May 4;7(2):e02265–15. pmid:26933058
- 24. Carden S, Walker GT, Honeycutt J, Lugo K, Pham T, Jacobson A, et al. Pseudogenization of the Secreted Effector Gene sseI Confers Rapid Systemic Dissemination of S. Typhimurium ST313 within Migratory Dendritic Cells. Cell Host Microbe. 2017 Feb 8;21(2):182–94. pmid:28182950
- 25. Tanner JR, Kingsley RA. Evolution of Salmonella within Hosts. Trends Microbiol. 2018 Dec 1;26(12):986–98. pmid:29954653
- 26. Carden S, Okoro C, Dougan G, Monack D. Non-typhoidal Salmonella Typhimurium ST313 isolates that cause bacteremia in humans stimulate less inflammasome activation than ST19 isolates associated with gastroenteritis. Pathog Dis. 2015 Jun 1;73(4):ftu023. pmid:25808600
- 27. Boinett CJ, Cain AK, Hawkey J, Do Hoang NT, Khanh NNT, Thanh DP, et al. Clinical and laboratory-induced colistin-resistance mechanisms in Acinetobacter baumannii. Microb Genomics. 2019;5(2):e000246.
- 28. Henry R, Crane B, Powell D, Deveson Lucas D, Li Z, Aranda J, et al. The transcriptomic response of Acinetobacter baumannii to colistin and doripenem alone and in combination in an in vitro pharmacokinetics/pharmacodynamics model. J Antimicrob Chemother. 2015 May 1;70(5):1303–13. pmid:25587995
- 29. Lu S, Zgurskaya HI. MacA, a Periplasmic Membrane Fusion Protein of the Macrolide Transporter MacAB-TolC, Binds Lipopolysaccharide Core Specifically and with High Affinity. J Bacteriol. 2013 Nov 1;195(21):4865–72. pmid:23974027
- 30. Rouquette-Loughlin CE, Balthazar JT, Shafer WM. Characterization of the MacA–MacB efflux system in Neisseria gonorrhoeae. J Antimicrob Chemother. 2005 Nov 1;56(5):856–60. pmid:16162665
- 31. Turlin E, Heuck G, Brandão MIS, Szili N, Mellin JR, Lange N, et al. Protoporphyrin (PPIX) efflux by the MacAB-TolC pump in Escherichia coli. MicrobiologyOpen. 2014 Dec 1;3(6):849–59. pmid:25257218
- 32. Crow A, Greene NP, Kaplan E, Koronakis V. Structure and mechanotransmission mechanism of the MacB ABC transporter superfamily. Proc Natl Acad Sci. 2017 Nov 21;114(47):12572–7. pmid:29109272
- 33. Fitzpatrick AWP, Llabrés S, Neuberger A, Blaza JN, Bai X-C, Okada U, et al. Structure of the MacAB–TolC ABC-type tripartite multidrug efflux pump. Nat Microbiol. 2017 Jul;2(7):17070.
- 34. Kobayashi N, Nishino K, Yamaguchi A. Novel Macrolide-Specific ABC-Type Efflux Transporter in Escherichia coli. J Bacteriol. 2001 Oct;183(19):5639–44. pmid:11544226
- 35. Nishino K, Latifi T, Groisman EA. Virulence and drug resistance roles of multidrug efflux systems of Salmonella enterica serovar Typhimurium. Mol Microbiol. 2006 Jan 1;59(1):126–41. pmid:16359323
- 36. Kröger C, Colgan A, Srikumar S, Händler K, Sivasankaran SK, Hammarlöf DL, et al. An Infection-Relevant Transcriptomic Compendium for Salmonella enterica Serovar Typhimurium. Cell Host Microbe. 2013 Dec 11;14(6):683–95. pmid:24331466
- 37. Greene NP, Kaplan E, Crow A, Koronakis V. Antibiotic Resistance Mediated by the MacB ABC Transporter Family: A Structural and Functional Perspective. Front Microbiol. 2018;9:950. pmid:29892271
- 38. Bogomolnaya LM, Andrews KD, Talamantes M, Maple A, Ragoza Y, Vazquez-Torres A, et al. The ABC-Type Efflux Pump MacAB Protects Salmonella enterica serovar Typhimurium from Oxidative Stress. mBio. 2013 Dec 31;4(6):e00630–13. pmid:24169575
- 39. Kelley LA, Mezulis S, Yates CM, Wass MN, Sternberg MJE. The Phyre2 web portal for protein modeling, prediction and analysis. Nat Protoc. 2015 Jun;10(6):845–58. pmid:25950237
- 40. Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P, Evans PR, et al. Overview of the CCP4 suite and current developments. Acta Crystallogr D Biol Crystallogr. 2011 Apr 1;67(4):235–42.
- 41. Krissinel E, Henrick K. Secondary-structure matching (SSM), a new tool for fast protein structure alignment in three dimensions. Acta Crystallogr D Biol Crystallogr. 2004 Dec;60(Pt 12 Pt 1):2256–68.
- 42. Dong F, Xia L, Lu R, Zhao X, Zhang Y, Zhang Y, et al. The malS-5′UTR weakens the ability of Salmonella enterica serovar Typhi to survive in macrophages by increasing intracellular ATP levels. Microb Pathog. 2018 Feb 1;115:321–31. pmid:29306008
- 43. Zwir I, Latifi T, Perez JC, Huang H, Groisman EA. The promoter architectural landscape of the Salmonella PhoP regulon. Mol Microbiol. 2012;84(3):463–85. pmid:22435712
- 44. Band VI, Weiss DS. Mechanisms of Antimicrobial Peptide Resistance in Gram-Negative Bacteria. Antibiotics. 2015 Mar;4(1):18–41. pmid:25927010
- 45. Miller SI, Kukral AM, Mekalanos JJ. A two-component regulatory system (phoP phoQ) controls Salmonella typhimurium virulence. Proc Natl Acad Sci U S A. 1989 Jul;86(13):5054–8. pmid:2544889
- 46. Groisman EA. The Pleiotropic Two-Component Regulatory System PhoP-PhoQ. J Bacteriol. 2001 Mar 15;183(6):1835–42. pmid:11222580
- 47. Fields PI, Groisman EA, Heffron F. A Salmonella locus that controls resistance to microbicidal proteins from phagocytic cells. Science. 1989 Feb 24;243(4894):1059–62.
- 48. Thompson JA, Liu M, Helaine S, Holden DW. Contribution of the PhoP/Q regulon to survival and replication of Salmonella enterica serovar Typhimurium in macrophages. Microbiology. 2011;157(7):2084–93.
- 49. Srikumar S, Kröger C, Hébrard M, Colgan A, Owen SV, Sivasankaran SK, et al. RNA-seq Brings New Insights to the Intra-Macrophage Transcriptome of Salmonella Typhimurium. PLOS Pathog. 2015 Nov 12;11(11):e1005262. pmid:26561851
- 50. Colgan AM, Kröger C, Diard M, Hardt W-D, Puente JL, Sivasankaran SK, et al. The Impact of 18 Ancestral and Horizontally-Acquired Regulatory Proteins upon the Transcriptome and sRNA Landscape of Salmonella enterica serovar Typhimurium. PLOS Genet. 2016 Aug 26;12(8):e1006258. pmid:27564394
- 51. Véscovi EG, Soncini FC, Groisman EA. Mg2+ as an Extracellular Signal: Environmental Regulation of Salmonella Virulence. Cell. 1996 Jan 12;84(1):165–74. pmid:8548821
- 52. Bader MW, Sanowar S, Daley ME, Schneider AR, Cho U, Xu W, et al. Recognition of Antimicrobial Peptides by a Bacterial Sensor Kinase. Cell. 2005 Aug 12;122(3):461–72. pmid:16096064
- 53. Rathman M, Sjaastad MD, Falkow S. Acidification of phagosomes containing Salmonella typhimurium in murine macrophages. Infect Immun. 1996 Jul 1;64(7):2765–73. pmid:8698506
- 54. Choi J, Groisman EA. Acidic pH sensing in the bacterial cytoplasm is required for Salmonella virulence. Mol Microbiol. 2016;101(6):1024–38. pmid:27282333
- 55. Rosenberger CM, Gallo RL, Finlay BB. Interplay between antibacterial effectors: A macrophage antimicrobial peptide impairs intracellular Salmonella replication. Proc Natl Acad Sci. 2004 Feb 24;101(8):2422–7. pmid:14983025
- 56. Reens AL, Crooks AL, Su C-C, Nagy TA, Reens DL, Podoll JD, et al. A cell-based infection assay identifies efflux pump modulators that reduce bacterial intracellular load. PLOS Pathog. 2018 Jun 7;14(6):e1007115. pmid:29879224
- 57. Yamanaka H, Kobayashi H, Takahashi E, Okamoto K. MacAB Is Involved in the Secretion of Escherichia coli Heat-Stable Enterotoxin II. J Bacteriol. 2008 Dec 1;190(23):7693–8. pmid:18805970
- 58. Shafer WM, Qu X-D, Waring AJ, Lehrer RI. Modulation of Neisseria gonorrhoeae susceptibility to vertebrate antibacterial peptides due to a member of the resistance/nodulation/division efflux pump family. Proc Natl Acad Sci. 1998 Feb 17;95(4):1829–33. pmid:9465102
- 59. Hong X, Chen HD, Groisman EA. Gene expression kinetics governs stimulus-specific decoration of the Salmonella outer membrane. Sci Signal. 2018 May 8;11(529):eaar7921. pmid:29739882
- 60. Blair JMA, Piddock LJV. How to Measure Export via Bacterial Multidrug Resistance Efflux Pumps. mBio. 2016 Sep 7;7(4):e00840–16. pmid:27381291
- 61. Richards SM, Strandberg KL, Conroy M, Gunn JS. Cationic antimicrobial peptides serve as activation signals for the Salmonella Typhimurium PhoPQ and PmrAB regulons in vitro and in vivo. Front Cell Infect Microbiol. 2012;2:102. pmid:22919691
- 62. Hapfelmeier S, Stecher B, Barthel M, Kremer M, Müller AJ, Heikenwalder M, et al. The Salmonella Pathogenicity Island (SPI)-2 and SPI-1 Type III Secretion Systems Allow Salmonella Serovar typhimurium to Trigger Colitis via MyD88-Dependent and MyD88-Independent Mechanisms. J Immunol. 2005 Feb 1;174(3):1675–85. pmid:15661931
- 63. Stecher B, Robbiani R, Walker AW, Westendorf AM, Barthel M, Kremer M, et al. Salmonella enterica Serovar Typhimurium Exploits Inflammation to Compete with the Intestinal Microbiota. PLOS Biol. 2007 Aug 28;5(10):e244.
- 64. Ramachandran G, Perkins DJ, Schmidlein PJ, Tulapurkar ME, Tennant SM. Invasive Salmonella Typhimurium ST313 with Naturally Attenuated Flagellin Elicits Reduced Inflammation and Replicates within Macrophages. PLoS Negl Trop Dis. 2015 Jan 8;9(1):e3394. pmid:25569606
- 65. Cunnington AJ, de Souza JB, Walther M, Riley EM. Malaria impairs resistance to Salmonella through heme- and heme oxygenase-dependent dysfunctional granulocyte mobilization. Nat Med. 2012 Jan;18(1):120–7.
- 66. MacLennan CA, Gilchrist JJ, Gordon MA, Cunningham AF, Cobbold M, Goodall M, et al. Dysregulated Humoral Immunity to Nontyphoidal Salmonella in HIV-Infected African Adults. Science. 2010 Apr 23;328(5977):508–12. pmid:20413503
- 67. Feasey NA, Everett D, Faragher EB, Roca-Feltrer A, Kang’ombe A, Denis B, et al. Modelling the Contributions of Malaria, HIV, Malnutrition and Rainfall to the Decline in Paediatric Invasive Non-typhoidal Salmonella Disease in Malawi. PLoS Negl Trop Dis. 2015 Jul 31;9(7):e0003979. pmid:26230258
- 68. Vidal SM, Pinner E, Lepage P, Gauthier S, Gros P. Natural resistance to intracellular infections: Nramp1 encodes a membrane phosphoglycoprotein absent in macrophages from susceptible (Nramp1 D169) mouse strains. J Immunol. 1996 Oct 15;157(8):3559–68. pmid:8871656
- 69. Blackwell JM, Goswami T, Evans CAW, Sibthorpe D, Papo N, White JK, et al. SLC11A1 (formerly NRAMP1) and disease resistance. Cell Microbiol. 2001 Dec;3(12):773–84. pmid:11736990
- 70. Nairz M, Fritsche G, Crouch M-LV, Barton HC, Fang FC, Weiss G. Slc11a1 limits intracellular growth of Salmonella enterica sv. Typhimurium by promoting macrophage immune effector functions and impairing bacterial iron acquisition. Cell Microbiol. 2009;11(9):1365–81. pmid:19500110
- 71. Fritsche G, Nairz M, Libby SJ, Fang FC, Weiss G. Slc11a1 (Nramp1) impairs growth of Salmonella enterica serovar typhimurium in macrophages via stimulation of lipocalin-2 expression. J Leukoc Biol. 2012;92(2):353–9. pmid:22706314
- 72. Cunrath O, Bumann D. Host resistance factor SLC11A1 restricts Salmonella growth through magnesium deprivation. Science. 2019 Nov 22;366(6468):995–9. pmid:31753999
- 73. Valdez Y, Grassl GA, Guttman JA, Coburn B, Gros P, Vallance BA, et al. Nramp1 drives an accelerated inflammatory response during Salmonella-induced colitis in mice. Cell Microbiol. 2009 Feb 1;11(2):351–62. pmid:19016783
- 74. Jiang H-R, Gilchrist DS, Popoff J-F, Jamieson SE, Truscott M, White JK, et al. Influence of Slc11a1 (formerly Nramp1) on DSS-induced colitis in mice. J Leukoc Biol. 2009;85(4):703–10. pmid:19116231
- 75. Valdez Y, Diehl GE, Vallance BA, Grassl GA, Guttman JA, Brown NF, et al. Nramp1 expression by dendritic cells modulates inflammatory responses during Salmonella Typhimurium infection. Cell Microbiol. 2008 Aug 1;10(8):1646–61. pmid:18397382
- 76. Hedges JF, Kimmel E, Snyder DT, Jerome M, Jutila MA. SLC11A1 is expressed by innate lymphocytes and augments their activation. J Immunol Baltim Md 1950. 2013 Apr 15;190(8):4263–73.
- 77. Zgurskaya HI, Rybenkov VV, Krishnamoorthy G, Leus IV. Trans-envelope multidrug efflux pumps of Gram-negative bacteria and their synergism with the outer membrane barrier. Res Microbiol. 2018 Sep 1;169(7):351–6.
- 78. Bogomolnaya LM, Tilvawala R, Elfenbein JR, Cirillo JD, Andrews-Polymenis HL. Linearized Siderophore Products Secreted via MacAB Efflux Pump Protect Salmonella enterica Serovar Typhimurium from Oxidative Stress. mBio. 2020 Jun 30;11(3):e00528–20. pmid:32371597
- 79. Dalebroux ZD, Miller SI. Salmonellae PhoPQ regulation of the outer membrane to resist innate immunity. Curr Opin Microbiol. 2014 Feb 1;17:106–13. pmid:24531506
- 80. Groisman EA, Mouslim C. Sensing by bacterial regulatory systems in host and non-host environments. Nat Rev Microbiol. 2006 Sep;4(9):705–9. pmid:16894339
- 81. Bader MW, Navarre WW, Shiau W, Nikaido H, Frye JG, McClelland M, et al. Regulation of Salmonella typhimurium virulence gene expression by cationic antimicrobial peptides. Mol Microbiol. 2003;50(1):219–30. pmid:14507376
- 82. Tu X, Latifi T, Bougdour A, Gottesman S, Groisman EA. The PhoP/PhoQ two-component system stabilizes the alternative sigma factor RpoS in Salmonella enterica. Proc Natl Acad Sci. 2006 Sep 5;103(36):13503–8. pmid:16938894
- 83. Charles RC, Harris JB, Chase MR, Lebrun LM, Sheikh A, LaRocque RC, et al. Comparative Proteomic Analysis of the PhoP Regulon in Salmonella enterica Serovar Typhi Versus Typhimurium. PLOS ONE. 2009 Sep 10;4(9):e6994. pmid:19746165
- 84. Pontes MH, Yeom J, Groisman EA. Reducing Ribosome Biosynthesis Promotes Translation during Low Mg2+ Stress. Mol Cell. 2016 Nov 3;64(3):480–92. pmid:27746019
- 85. Choi E, Han Y, Cho Y-J, Nam D, Lee E-J. A trans-acting leader RNA from a Salmonella virulence gene. Proc Natl Acad Sci. 2017 Sep 19;114(38):10232–7. pmid:28874555
- 86. Hör J, Matera G, Vogel J, Gottesman S, Storz G. Trans-Acting Small RNAs and Their Effects on Gene Expression in Escherichia coli and Salmonella enterica. EcoSal Plus. 2020 Mar 20;9(1):
- 87. Tikhonova EB, Dastidar V, Rybenkov VV, Zgurskaya HI. Kinetic control of TolC recruitment by multidrug efflux complexes. Proc Natl Acad Sci. 2009 Sep 22;106(38):16416–21. pmid:19805313
- 88. Hammarlöf DL, Kröger C, Owen SV, Canals R, Lacharme-Lora L, Wenner N, et al. Role of a single noncoding nucleotide in the evolution of an epidemic African clade of Salmonella. Proc Natl Acad Sci. 2018 Feb 27;201714718.
- 89. Bravo D, Silva C, Carter JA, Hoare A, Álvarez SA, Blondel CJ, et al. Growth-phase regulation of lipopolysaccharide O-antigen chain length influences serum resistance in serovars of Salmonella. J Med Microbiol. 2008;57(8):938–46.
- 90. Crawford RW, Wangdi T, Spees AM, Xavier MN, Tsolis RM, Bäumler AJ. Loss of Very-Long O-Antigen Chains Optimizes Capsule-Mediated Immune Evasion by Salmonella enterica Serovar Typhi. mBio. 2013 Aug 30;4(4):e00232–13. pmid:23860765
- 91. Wang-Kan X, Blair JMA, Chirullo B, Betts J, Ragione RML, Ivens A, et al. Lack of AcrB Efflux Function Confers Loss of Virulence on Salmonella enterica Serovar Typhimurium. mBio. 2017 Sep 6;8(4):e00968–17. pmid:28720734
- 92. Ramachandran G, Panda A, Higginson EE, Ateh E, Lipsky MM, Sen S, et al. Virulence of invasive Salmonella Typhimurium ST313 in animal models of infection. PLoS Negl Trop Dis. 2017 Aug 4;11(8):e0005697. pmid:28783750
- 93. Yang J, Barrila J, Roland KL, Kilbourne J, Ott CM, Forsyth RJ, et al. Characterization of the Invasive, Multidrug Resistant Non-typhoidal Salmonella Strain D23580 in a Murine Model of Infection. PLoS Negl Trop Dis. 2015 Jun 19;9(6):e0003839. pmid:26091096
- 94. Lacharme-Lora L, Owen SV, Blundell R, Canals R, Wenner N, Perez-Sepulveda B, et al. The use of chicken and insect infection models to assess the virulence of African Salmonella Typhimurium ST313. PLoS Negl Trop Dis. 2019 Jul 26;13(7):e0007540. pmid:31348776
- 95. Friebel A, Ilchmann H, Aepfelbacher M, Ehrbar K, Machleidt W, Hardt W-D. SopE and SopE2 from Salmonella typhimurium Activate Different Sets of RhoGTPases of the Host Cell. J Biol Chem. 2001 Sep 7;276(36):34035–40. pmid:11440999
- 96. Hapfelmeier S, Ehrbar K, Stecher B, Barthel M, Kremer M, Hardt W-D. Role of the Salmonella Pathogenicity Island 1 Effector Proteins SipA, SopB, SopE, and SopE2 in Salmonella enterica Subspecies 1 Serovar Typhimurium Colitis in Streptomycin-Pretreated Mice. Infect Immun. 2004 Feb 1;72(2):795–809. pmid:14742523
- 97. Nedialkova LP, Denzler R, Koeppel MB, Diehl M, Ring D, Wille T, et al. Inflammation Fuels Colicin Ib-Dependent Competition of Salmonella Serovar Typhimurium and E. coli in Enterobacterial Blooms. PLOS Pathog. 2014 Jan 2;10(1):e1003844. pmid:24391500
- 98. Bellamy R, Ruwende C, Corrah T, McAdam KPWJ, Whittle HC, Hill AVS. Variations in the NRAMP1 Gene and Susceptibility to Tuberculosis in West Africans. N Engl J Med. 1998 Mar 5;338(10):640–4. pmid:9486992
- 99. Awomoyi AA, Marchant A, Howson JMM, McAdam KPWJ, Blackwell JM, Newport MJ. Interleukin-10, Polymorphism in SLC11A1 (formerly NRAMP1), and Susceptibility to Tuberculosis. J Infect Dis. 2002 Dec 15;186(12):1808–14. pmid:12447767
- 100. Raffatellu M, Santos RL, Verhoeven DE, George MD, Wilson RP, Winter SE, et al. Simian immunodeficiency virus–induced mucosal interleukin-17 deficiency promotes Salmonella dissemination from the gut. Nat Med. 2008 Apr;14(4):421–8. pmid:18376406
- 101. Barton CH, Whitehead SH, Blackwell JM. Nramp transfection transfers Ity/Lsh/Bcg-related pleiotropic effects on macrophage activation: influence on oxidative burst and nitric oxide pathways. Mol Med. 1995 Mar;1(3):267–79. pmid:8529105
- 102. Cunnington AJ, Njie M, Correa S, Takem EN, Riley EM, Walther M. Prolonged Neutrophil Dysfunction after Plasmodium falciparum Malaria Is Related to Hemolysis and Heme Oxygenase-1 Induction. J Immunol. 2012 Oct 24;1201028.
- 103. Gilchrist JJ, Rautanen A, Fairfax BP, Mills TC, Naranbhai V, Trochet H, et al. Risk of nontyphoidal Salmonella bacteraemia in African children is modified by STAT4. Nat Commun. 2018 Mar 9;9(1):1014. pmid:29523850
- 104. Monack DM, Bouley DM, Falkow S. Salmonella typhimurium Persists within Macrophages in the Mesenteric Lymph Nodes of Chronically Infected Nramp1+/+ Mice and Can Be Reactivated by IFNγ Neutralization. J Exp Med. 2004 Jan 19;199(2):231–41. pmid:14734525
- 105. Atkinson PG, Blackwell JM, Barton CH. Nramp1 locus encodes a 65 kDa interferon-gamma-inducible protein in murine macrophages. Biochem J. 1997 Aug 1;325(Pt 3):779–86.
- 106. Gilchrist JJ, MacLennan CA. Invasive Nontyphoidal Salmonella Disease in Africa. EcoSal Plus. 2019 Jan 18;8(2):
- 107. Barthel M, Hapfelmeier S, Quintanilla-Martínez L, Kremer M, Rohde M, Hogardt M, et al. Pretreatment of Mice with Streptomycin Provides a Salmonella enterica Serovar Typhimurium Colitis Model That Allows Analysis of Both Pathogen and Host. Infect Immun. 2003 May 1;71(5):2839–58. pmid:12704158
- 108. Richardson EJ, Limaye B, Inamdar H, Datta A, Manjari KS, Pullinger GD, et al. Genome Sequences of Salmonella enterica Serovar Typhimurium, Choleraesuis, Dublin, and Gallinarum Strains of Well- Defined Virulence in Food-Producing Animals. J Bacteriol. 2011;193(12):3162–3. pmid:21478351
- 109. Msefula CL, Kingsley RA, Gordon MA, Molyneux E, Molyneux ME, MacLennan CA, et al. Genotypic Homogeneity of Multidrug Resistant S. Typhimurium Infecting Distinct Adult and Childhood Susceptibility Groups in Blantyre, Malawi. PLOS ONE. 2012;7(7):e42085. pmid:22848711
- 110. Seemann T. Snippy: rapid haploid variant calling and core SNP phylogeny [Internet]. 2019. (Available). Available from: https://github.com/tseemann/snippy
- 111. Garrison E, Marth G. Haplotype-based variant detection from short-read sequencing. ArXiv12073907 Q-Bio [Internet]. 2012 Jul 20 [cited 2020 Jun 17]; Available from: http://arxiv.org/abs/1207.3907
- 112. Croucher NJ, Page AJ, Connor TR, Delaney AJ, Keane JA, Bentley SD, et al. Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins. Nucleic Acids Res. 2014;43(3):e15–e15. pmid:25414349
- 113. Kozlov AM, Darriba D, Flouri T, Morel B, Stamatakis A. RAxML-NG: a fast, scalable and user-friendly tool for maximum likelihood phylogenetic inference. Bioinformatics. 2019;35(21):4453–5. pmid:31070718
- 114. Letunic I, Bork P. Interactive Tree Of Life (iTOL): an online tool for phylogenetic tree display and annotation. Bioinformatics. 2006;23(1):127–8. pmid:17050570
- 115. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC Bioinformatics. 2009 Dec 15;10(1):421.
- 116. Wick RR, Judd LM, Gorrie CL, Holt KE. Unicycler: Resolving bacterial genome assemblies from short and long sequencing reads. PLOS Comput Biol. 2017;13(6):e1005595. pmid:28594827
- 117. Gurevich A, Saveliev V, Vyahhi N, Tesler G. QUAST: quality assessment tool for genome assemblies. Bioinformatics. 2013 Apr 15;29(8):1072–5. pmid:23422339
- 118. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci. 2000 Jun 6;97(12):6640–5. pmid:10829079
- 119. Ellermeier CD, Janakiraman A, Slauch JM. Construction of targeted single copy lac fusions using λ Red and FLP-mediated site-specific recombination in bacteria. Gene. 2002 May 15;290(1):153–61.
- 120. Martínez-García E, de Lorenzo V. Engineering multiple genomic deletions in Gram-negative bacteria: analysis of the multi-resistant antibiotic profile of Pseudomonas putida KT2440. Environ Microbiol. 2011 Oct 1;13(10):2702–16. pmid:21883790
- 121. Schaefer J, Jovanovic G, Kotta-Loizou I, Buck M. Single-step method for β-galactosidase assays in Escherichia coli using a 96-well microplate reader. Anal Biochem. 2016 Jun 15;503:56–7. pmid:27036618
- 122. Thibodeau SA, Fang R, Joung JK. High-throughput β-galactosidase assay for bacterial cell-based reporter systems. BioTechniques. 2004 Mar 1;36(3):410–5. pmid:15038156
- 123. Jacobson A, Lam L, Rajendram M, Tamburini F, Honeycutt J, Pham T, et al. A Gut Commensal-Produced Metabolite Mediates Colonization Resistance to Salmonella Infection. Cell Host Microbe. 2018 Aug 8;24(2):296–307.e7. pmid:30057174