Critical role of msgA in invasive capacity and intracellular survivability of Salmonella

ABSTRACT

Salmonella enterica serovar Typhimurium, which is a common foodborne pathogen, causes both intestinal and systemic infections in hosts. Salmonella has a complex pathogenic mechanism that involves invasive capacity and intracellular survivability, which hampers research on virulence of Salmonella. The virulence of Salmonella is primarily studied through Salmonella pathogenicity islands (SPIs). However, there are also genes outside these SPIs that significantly impact virulence. Macrophage survival gene msgA is positioned at a region independent of the SPIs and conserved in Salmonella. However, there has been limited research on msgA to date. This study aims to investigate the virulent function of msgA to deepen our understanding of Salmonella virulence. Proteomic and RT-qPCR analyses reveal that MsgA influences multiple metabolic pathways and the expression of SPIs. The depletion of msgA led to the significantly reduced invasive capacity and intracellular survivability, and thus the decreased virulence of Salmonella. In conclusion, our study suggests that MsgA is an important regulator that mainly regulates virulence. Further research into the function of MsgA will enhance the understanding of Salmonella pathogenesis and promote the application of Salmonella for medical treatment.

IMPORTANCE

Salmonella enterica serovar Typhimurium is a common foodborne pathogen, it has a complex pathogenic mechanism that involves invasive capacity and intracellular survivability. The virulence of Salmonella is primarily studied through its pathogenicity islands. In contrast, virulence genes located outside the Salmonella pathogenicity islands (SPIs) have received less attention. Macrophage survival gene (MsgA) is positioned at a region independent of the SPIs and conserved in Salmonella. Our research indicates that MsgA is a novel global regulator influencing the metabolic pathways and SPIs. Further research into the function of MsgA will enhance the understanding of Salmonella pathogenesis and promote the application of Salmonella for medical treatment.

INTRODUCTION

Salmonella enterica serovar Typhimurium (STM), which is a common foodborne pathogen, causes both intestinal and systemic infections in hosts (1). The virulence of STM is closely related to Salmonella pathogenicity islands (SPIs). Different SPIs encode different virulence factors, of which the best-understood SPIs are SPI-I and SPI-II. SPI-I and SPI-II encode different type III secretion systems (T3SSs), the T3SS encoded by SPI-I induces cellular actin rearrangements and membrane ruffling in intestinal epithelial cells, which can be used by STM to invade intestinal epithelial cells and induce extrusion and pyroptosis (2–4). SPI-II is associated with intracellular survival (5–7) and systemic infection (8, 9). After phagocytosis by macrophage, effector proteins encoded by SPI-II induce macrophage to form stable Salmonella-containing vacuoles (SCVs) that do not fuse with lysosomes (2, 10–12), induce the formation of Salmonella-induced filaments (SIFs) to take up nutrients from host cells (13–15), trigger metabolic reprogramming in the host cell and use host metabolites as signals to induce virulence (16, 17), and invade deep tissues, such as the liver and spleen, using macrophages as vectors (18). Furthermore, there are also genes outside the SPI that affect the virulence of Salmonella, such as htrA, phoP/phoQ, and fis (19–22). MsgA is conserved in Salmonella, utilized as a marker for Salmonella detection, and related to intracellular viability (23–25). In addition, the effect of msgA on Salmonella virulence has been preliminarily explored in the highly virulent strain ATCC14028s (23), but the mechanism is still unclear. VNP20009 is an attenuated strain with potential for anticancer treatment. Exploring the mechanism of msgA and its effect on virulence in this genetic context not only deepens the understanding of Salmonella virulence but also advances the research progress of VNP20009 in antitumor applications. Furthermore, it is worth noting that msgA is independent of the virulence island. Hence, we have further explored the function of MsgA to deepen our understanding of Salmonella virulence.

MsgA contains 246 bps and encodes a small-molecule protein consisting of 81 amino acids, which is composed mainly of α-helices, extended strand, and random coil. MsgA belongs to the DinI protein family, which is highly conserved in Escherichia coli. Many studies have shown that DinI regulates the SOS response by competitively binding to ssDNA in E. coli (26, 27). However, the function of the DinI protein family in Salmonella remains unclear, and evolutionary tree analysis indicates a relatively distant evolutionary relationship between the MsgA and DinI; thus, the function of the MsgA protein requires further investigation. A transcriptome analysis suggested that the transcriptional activity of msgA is increased two- to threefold within macrophages (28). Another microarray analysis result indicated that the transcription level of msgA is upregulated in STM SL1344 in human colonic epithelial (Caco-2) cell cytoplasm (29). These studies further validate that MsgA functions in intracellular survival. Based on this (28, 29), we assumed that msgA might be related to the regulation of SPIs. To date, extensive research has indicated that STM regulates the expression of SPI through various mechanisms, such as two-component systems (30, 31). However, the complete network and mechanisms of virulence remain incompletely understood. Exploration of MsgA will help refine the regulatory network of virulence.

In this study, we utilized bioinformatics, particularly proteomic analysis, to elucidate the role of MsgA in STM. The impact of MsgA on the invasive capacity and intracellular survivability of Salmonella was evaluated using RAW cells. Additionally, the effect of MsgA on virulence was studied using a mouse model.

MATERIALS AND METHODS

Physical and chemical analysis, structural prediction, and evolutionary analysis

We obtained the MsgA protein sequence from NCBI, queried the STRING database to identify species in which this protein is present, and subsequently retrieved sequences from these species for multiple sequence alignment and phylogenetic tree construction. The accession numbers for all sequences used are provided in Table S1. We used Mega-11 to compare the MsgA proteins with the ClustW algorithm. Evolutionary tree analysis was performed by maximum likelihood method in Mega-11, and the results were presented in the circular evolutionary tree. Analyzing the physicochemical properties of the protein’s primary structure, including isoelectric point and hydrophobicity, was done using online tools such as ExPASy (https://www.expasy.org/). The secondary and tertiary structures of MsgA were predicted by SOPMA (https://npsa.lyon.inserm.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_sopma.html) and SWISS-MODEL (https://swissmodel.expasy.org/) (32). The presence of specific sites, such as phosphorylation sites, within the protein was assessed by NetPhos (https://services.healthtech.dtu.dk/services/NetPhos-3.1/) (33). The presence of signal peptides was investigated by SignalP (https://services.healthtech.dtu.dk/services/SignalP-5.0/) (34). Potential transmembrane regions in the protein were analyzed by TMHMM (https://services.healthtech.dtu.dk/services/TMHMM-2.0/) (35). The protein–protein interaction network of MsgA was predicted by STRING (https://cn.string-db.org/) (36).

Label-free quantitative proteomic analysis

Salmonella VNP20009 and ΔmsgA were utilized in the current study. VNP20009 and ΔmsgA are sourced from Nanjing University. Salmonella VNP20009 is a purine auxotrophy, which deletes msbB from S. enterica ATCC 14028 (37). ΔmsgA was constructed in our previous studies generated using an efficient and scarless editing system (38).

Glycerol stocks of these strains were streaked onto lysogeny broth (LB) agar plates and incubated overnight at 37°C. Single colonies were selected and cultured in LB until reaching the exponential growth phase. When the bacteria are in the exponential growth phase, the optical density (OD) at 600 nm of the bacterial cultures is between 0.6 and 0.8. OD 600 was measured using a UV spectrophotometer. Bacterial cultures were centrifuged at 4°C and 10,000 rpm for 10 min, and the pellets were collected. Each pellet was solubilized with 100 µL of radioimmunoprecipitation assay buffer and lysed on ice for 30 min with occasional inversion every 5 min. The lysates were treated with non-contact sonication for 10 min, followed by a 10-min centrifugation at 12,000 rpm to collect the supernatant. Protein concentrations were measured and adjusted to 1 µg/µL before proceeding with protein sequencing. The samples were sent to label-free quantitation at the Translational Medicine Center of Nanjing University Medical School.

Differential analysis of the proteomic results was conducted based on criteria involving a fold change of 1.2 or 1/1.2 and a P value less than 0.05. A Volcano plot and a Venn diagram with differentially expressed proteins were mapped by ggplot2-3.4.3 and VennDiagram-1.7.3. Subsequently, enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of the differentially expressed genes were performed using Protein Analysis Through Evolutionary Relationships (PANTHER) (http://pantherdb.org/) (39) and Database for Annotation, Visualization and Integrated Discovery (DAVID) software (https://david.ncifcrf.gov) (40).

Quantitative reverse transcription-PCR

RT-qPCR was used to detect genes related to the pathways of virulence island, ethanolamine degradation. The total RNA was extracted using the FastPure Cell/Tissue Total RNA Isolation Kit (Vazyme). The extracted RNA samples were treated with DNase I to remove any genomic DNA contaminants. The cDNA was prepared from the RNA using Revert Aid first-strand cDNA synthesis kit (Thermo Scientific). The housekeeping gene encoding 16S rRNA was used as internal control (41, 42). All primers used in RT-qPCR are listed in Table S1. Quantitative PCR was performed on StepOne Real-Time PCR System (Applied Biosystems, USA) with AceQ qPCR SYBR Green Master Mix (Vazyme China). Data were analyzed by StepOne Software 2.1 (Applied Biosystems, USA) according to the manufacturer’s specifications. The volume of the reaction system was 20 µL. The Q-PCR reaction program consists of three steps as follows: pre-denaturation (95°C, 5 min), cycling reaction (95°C, 10 s; 60°C, 30 s; 40 cycles), and melting curve (95°C, 15 s; 60°C, 60 s; rise to 95°C at 0.15 °C/s). Image and statistical analysis were performed in Prism-9.5.1. Two-way analysis of variance (ANOVA) was used to determine statistical significance with 95% CI, P value style: 0.12349 (ns), 0.0332 (*), 0.0021 (**), 0.0002 (***), <0.0001 (****).

Gentamycin protection assay

Salmonella cannot only invade cells but also survive in macrophages and utilize macrophages to reach deep tissues. The RAW 264.7 cells are monocyte/macrophage-like cells, originating from Abelson leukemia virus-transformed cell line derived from BALB/c mice. This cell line has been an appropriate model of macrophages (43). So, RAW 264.7 cells were used to explore the invasive and intracellular survival of Salmonella in vitro (44). RAW 264.7 cells were seeded in 12-well plates with a cell density of 10⁵ cells/well. VNP20009 and ΔmsgA were cultured to the logarithmic growth phase and then centrifuged at 5,000 rpm, 4°C, for 5 mins to collect pellets. The bacterial solution was adjusted to the same concentration using sterile phosphate-buffered saline (PBS); the concentration was approximately 4 * 10⁸ CFU/mL (OD600 is 1.0). Cells were infected with a multiplicity of infection of 100:1. The mixture of cells and bacteria was incubated in a cell incubator for 1 h. The supernatant was discarded. and cell monolayers were washed with PBS buffer to remove extracellular bacteria. Then cells were cultured for 0.5 and 2.5 h in the medium containing 100 µg/mL of gentamicin. After discarding the medium and washing it with PBS buffer three times, cells were lysed with 0.5% Triton X-100. The lysates were inoculated on LB plates after a gradient dilution. Single colonies were counted after overnight incubation at 37℃. The number of single colonies was recorded as C0 for cells incubated with gentamicin for 0.5 h; the number of single colonies was recorded as C2 for cells incubated with gentamicin for 2 h, counting and comparing C0 to assess the invasive capacity of the strain. Growth rate of intracellular bacteria was calculated using the following equation: Growth rate of intracellular bacteria (%) = (C2 − C0)/C0. A t-test was used to determine statistical significance with 95% CI, P value style: 0.12349 (ns), 0.0332 (*), 0.0021 (**), 0.0002 (***), <0.0001 (****).

Cytotoxicity detection by flow cytometry

RAW 264.7 cells were utilized in the current study. Cell invasion assay was performed similarly to gentamicin protection assay. After killing the extracellular bacteria, all cells were further incubated overnight. The cells were collected, washed three times with PBS, and resuspended gently in Annexin V–APC diluted with 400 µL 1 × binding buffer. The cells were then incubated on ice in the dark for 30 min. Subsequently, 1 µL of propidium iodide (PI) was added to each sample and gently mixed. Cell viability (A5⁻PI⁻) was assessed using a flow cytometer. One-way ANOVA was used to determine statistical significance with 95% CI.

Chronic toxicity assessment

Glycerol stocks of VNP20009 and ΔmsgA were streaked onto LB agar plates and incubated overnight; then, a single colony of each strain was selected and cultured to the exponential growth phase. The specific methods and conditions are the same as in the “Gentamycin Protection Assay” section. Centrifuge and collect the bacterial precipitate, then resuspend the bacterial precipitate in PBS to 1 * 10⁷ CFU/mL. Female C57BL/6 mice were used as a mouse model to explore the virulence of Salmonella. C57BL/6 mice were randomly divided into three groups, with each group consisting of six individuals. These groups received intraperitoneal injections of 100 µL of either PBS, VNP20009, or ΔmsgA. After intraperitoneal injections, the weight of each mouse was measured every 2 days. Two-way ANOVA was used to determine statistical significance with 95% CI, P value style: 0.12349 (ns), 0.0332 (*), 0.0021 (**), 0.0002 (***), <0.0001 (****). Female C57BL/6, 6 to 8 weeks old, were purchased from Huachuang Sino Company (Nanjing, China) and kept under specific pathogen-free conditions.

RESULTS

Physical and chemical analyses, protein interaction network, structural prediction, and evolutionary analyses

Online tools in ExPASy predicted the molecular weight of MsgA to be 9470.77 Da and the isoelectric point to be 5.12. STRING prediction indicates that MsgA is associated with 10 proteins in which six of them are virulence proteins (Fig. S1). In addition, it has been demonstrated that in Salmonella ATCC14028, the mutation of msgA results in a decrease in LD50 (23). So, MsgA might interact with these virulence proteins. We focused on the levels of these proteins after MsgA knockdown and labeled the relevant protein changes in the volcano plot (Fig. 2a). NetPhos analysis showed that MsgA may be phosphorylated at residues 29, 53, 56, and 58 (Fig. 1a; Fig. S2a). Modeling (Fig. 1b) with Swiss Model was consistent with the secondary structure predicted by SOPMA (Fig. 1a; Fig. S2b), indicating that the protein might be composed mainly of α-helices (60.49%), extended strand (14.84%), and random coil (24.69%). Phosphorylation sites and random coil suggest that MsgA has a biological function. TMHMM analysis, which predicts the Trans-Membrane using Hidden Markov Models, showed that there is no transmembrane region in MsgA, and SignalP analysis indicated that there is no signal peptide in MsgA. It means that MsgA may not be a secreted protein and functions by influencing life activities of Salmonella.

Fig 1.

Bioinformatics analysis results. (a) Sequence of MsgA; different colored fonts represent different secondary structures (blue: α-helices; red: extended strand; orange: random coil), and yellow backgrounds represent amino acid residues that may be phosphorylated. (b) Tertiary structure of MsgA predicted by SWISS-MODEL based on homology modeling approach. (c) Circular evolutionary tree drawn by MEGA11 for msgA in different species. Salmonella are marked in red; DinI is marked in blue.

Based on the cooccurrence analysis in the STRING database, MsgA is conserved in Enterobacterales except Morganellaceae and Thorsellia anopheles (Fig. S3). The evolutionary tree indicates that the evolutionary relationship of MsgA in Salmonella is related to MsgA in Citrobacter, Enterobacter, Escherichia coli, and Klebsiella and is far from MsgA in Cedecea, Erwiniaceae, and Pectobacterium (Fig. 1c). However, MsgA displays a distant evolutionary relationship with DinI in Escherichia coli. The result of the multiple sequence alignment is visualized using Jalview (Fig. S4) and indicates that the alpha helix, which is between the two arrows, exhibits a high degree of conservation.

Differential protein expression between VNP20009 and ΔmsgA

Label-Free was employed to explore the function of MsgA in STM. Proteomic sequencing was performed on both VNP20009 and ΔmsgA. In total, 2,368 proteins (FDR < 1%) were detected by Proteomic sequencing; in ΔmsgA 476, proteins were found to be upregulated [foldchange(ΔmsgA/VNP20009) >1.2 and P < 0.05], while 603 proteins were downregulated [foldchange(ΔmsgA/VNP20009) <0.83 and P < 0.05]. The differently expressed proteins constitute 45.6% of all detected proteins, which suggests that MsgA significantly impacts the gene expression of STM.

A Volcano plot and a Venn diagram with differentially expressed proteins were mapped as shown in Fig. 2a. MsgA-associated proteins predicted by STRING are marked with yellow dots in the volcano plot. Among the differentially expressed proteins, ΔmsgA has 24 unique proteins, while VNP20009 has 61 unique proteins (Fig. 2b). The information of these proteins is shown in Table S2. However, enrichment analysis with these unique proteins in PANTHER did not yield statistically significant results.

Fig 2.

Differentially expressed protein (DEP) between VNP20009 and ΔmsgA. (a) Volcano plots of the DEP; yellow dots represent MsgA-related proteins predicted by STRING. (b) Distribution of differential genes in VNP20009 and ΔmsgA.

Enrichment analysis in differentially expressed proteins between VNP20009 and ΔmsgA

To investigate the roles of the impacted proteins, we performed enrichment analysis on differentially expressed proteins using PANTHER and DAVID. The PANTHER enrichment analysis revealed that upregulated proteins are associated with nine Gene Ontology (GO) terms (Fig. 3a), while downregulated proteins are associated with 14 GO terms (Fig. 3b). Upregulated biological processes included nitrate metabolic, fatty acid biosynthesis, and amino acid metabolism. Upregulated proteins are primarily found in the cytosol and are classified according to their molecular function as either oxidoreductases or enzymes with catalytic activity. Downregulated biological processes encompass the tricarboxylic acid cycle, proton motive force-driven ATP synthesis, ethanolamine catabolic process, and α-amino acid process. Downregulated proteins are mainly present in the proton-transporting ATP synthase complex, plasma membrane respiratory chain complex II, periplasmic space, ethanolamine degradation polyhedral organelle, and cytosol, whose molecular functions include oxidoreductase, metal ion binding, ligase, ammonia binding, and adenyl nucleotide binding. The GO enrichment analyses outcomes from the DAVID database are in overall concurrence with PANTHER (Fig. 3c and d).

Fig 3.

GO enrichment analysis results. (a, b) Enrichment results of up- and downregulated proteins expressed in ΔmsgA using PANTHER; gene ratio represents the proportion of differentially expressed proteins to the total proteins listed in the pathway in the database. BP means biological process, CC means cellular component, and MF means molecular function. (c, d) Enrichment results of up- and downregulated proteins in ΔmsgA using DAVID; count represents the number of differentially expressed proteins in the pathway.

The KEGG analysis results from the DAVID database are displayed in Fig. 4, and they are in general accordance with the GO enrichment analysis results. In the downregulated pathways, sequence-specific RNA polymerase transcriptional activator is identified, which includes TtrR, PhoP, RstA, Fnr, TorR, Crp, BasR, PhoB, HilA, and CreB. Among the above regulatory activators, CrP, HilA, and Fnr are implicated in regulating SPI-I expression (45–47); PhoP and BasR are associated with the regulation of SPI-II expression; and TtrR regulates genes involved in thiosulfate utilization (31, 48, 49). These pathways collectively influence virulence. Therefore, we counted the transcription levels of SPI regulators and T3SSs proteins encoded by SPIs (Fig. 4c and d). The proteins related to SPI-II were not detected in proteome sequencing, maybe because the expression was too low to be detected when Salmonella is cultured in LB (50). Our findings indicate that the transcription activators of SPI-II undergo downregulation in ΔmsgA (Fig. 4c). Therefore, we hypothesize that the expression of SPI-II is also downregulated in ΔmsgA and plan to confirm this through RT-qPCR analysis.

Fig 4.

KEGG enrichment results and protein expression mapping on genomes. (a, b) KEGG enrichment results of up- and downregulated proteins expressed in ΔmsgA using DAVID; count represents the number of differentially expressed proteins in the pathway. (c) Volcano plots of the transcription activator of SPI-I and SPI-II. (d) Volcano plots of proteins encoded by SPI-I.

Confirmation by quantitative reverse transcription-PCR

To further confirm the effect of MsgA on SPI, we detected the transcript level of crucial genes via RT-qPCR. Based on the proteome and previous research findings, for SPI-I, we selected genes encoding central local regulatory factors hilA and hilD, the structural proteins invA and prgH, the accessory proteins sicA and spaN, and the effector proteins sopE2 and sipA. For SPI-II, we selected genes encoding the global regulatory factor phoP and central local regulatory factors ssrA/ssrB, the structural proteins ssaS and ssaV, the accessory proteins sscA and sscB, and the effector proteins sseF. The transcription level of eutR, which not only is the regulatory factor of the ethanolamine pathway but also promotes the expression of ssrB, was also detected using RT-qPCR. The RT-qPCR analysis revealed downregulated transcript levels of the aforementioned genes (Fig. 5 and 6), which aligns with the proteomic results. Although there was no significant difference in transcript level of prgH in ΔmsgA compared to VNP20009, the relative expression was still lower than that of VNP20009.

Fig 5.

The relative expression of the crucial genes of SPI-I and invasive capacity of VNP20009 and ΔmsgA.

Fig 6.

The relative expression of the crucial genes of SPI-II and intracellular survivability of VNP20009 and ΔmsgA.

Gentamicin protection assay and toxicity assessment

Proteomic analysis indicates a potential reduction in the invasive ability and intracellular survivability of ΔmsgA. Therefore, the gentamicin protection assay was employed to assess the invasion and intracellular survival capabilities of ΔmsgA. The results show a significant reduction in invasion and intracellular survival abilities in ΔmsgA compared to VNP20009 (Fig. 5d and 6d).

The invasion capability and intracellular survival of Salmonella are closely linked to its cytotoxicity. To investigate the cytotoxicity of ΔmsgA, flow cytometry was utilized to analyze Salmonella’s impact on host cells. We chose Annexin V and propidium iodide (PI) to assess cell apoptosis and necrosis. The results revealed a significant increase in the proportion of A5⁻PI⁻ cells in RAW cells treated with ΔmsgA compared to VNP20009, indicating that ΔmsgA exhibited significantly reduced toxicity (Fig. 7a). Then, we used C57BL/6 to further evaluate the toxicity of ΔmsgA. The result showed no significant difference in body weight changes between mice injected with ΔmsgA and PBS (Fig. 7b).

Fig 7.

Cytotoxicity assessment. (a) Cell viability detected by flow cytometry. (b) Weight of mice after intraperitoneal injection of Salmonella.

DISCUSSION

To have a basic understanding of the structure and function of MsgA, structural predictions and evolutionary analyses were conducted. Predictions revealed that MsgA is similar to DinI at the tertiary level. Multiple sequence alignments and evolutionary analyses indicated that MsgA shows a high degree of conservation within Enterobacteriaceae but with a distant evolutionary relationship to DinI. It suggests that MsgA may play an important role in Enterobacteriaceae, which is different from DinI. Bioinformatics predictions indicate that MsgA does not contain a signal peptide, which implies that it may function intracellularly. Prediction results of protein interaction networks using the STRING indicated that MsgA co-occurs with virulence proteins. Therefore, further investigation is required to research the function of MsgA, particularly its impact on virulence.

In this study, the function of MsgA was predicted through proteomics. The proteomics analysis results indicate that knockdown of msgA led to changes in the levels of 1,079 proteins, which constitute 45.6% of all proteins detected. It is suggested that MsgA plays a significant role in regulating gene expression in Salmonella. Proteomic findings also demonstrate a significant downregulating in MsgA-related proteins, which were predicted by STRING, specifically the virulence proteins. Enrichment analyses indicate that downregulated pathways include ethanolamine degradation, the tricarboxylic acid (TCA) cycle, and pyruvate metabolism, and upregulated pathways include fatty acid biosynthesis and nitrate metabolism. In addition, differential proteins are predominantly localized in the cytoplasm, which is consistent with the previous predictions. RT-qPCR analysis demonstrates that the transcript levels of SPI-I and II, eutR, and genes encoding regulatory factors of SPIs are lower in ΔmsgA compared to those in VNP20009 (Fig. 5 and 6). These results are consistent with those of the proteomic analysis. These pathways, which are changed in ΔmsgA, are associated with virulence.

Much research has shown that the ethanolamine catabolic process impacts the virulence and organizational distribution of Salmonella (51, 52). The eut operation regulates the ethanolamine catabolic process (53, 54), consisting of 17 genes that encode proteins vital for the degradation of ethanolamine, and enables Salmonella to utilize ethanolamine. Within the ethanolamine degradation pathway, ethanolamine is catabolized to acetyl-CoA (Ac-CoA) and acetate (55), acetate is capable of producing through the enzymatic action of phosphotransacetylase known as EutD (56), acetate interacts with intestinal G-protein-coupled receptor 43, subsequently modulating innate immunity and inflammatory responses (57). It has been shown that STM can induce enteritis and induce the host intestine to produce tetrathionate and use tetrathionate to utilize ethanolamine, thus becoming the dominant flora in the intestinal tract (51). In addition, EutR directly activates the expression of SPI-II in the intracellular environment of macrophages, thereby affecting the STM’s intracellular survival capabilities (58). Simultaneously, studies have pointed out the regulatory role of SpeG in intracellular amine metabolism and virulence factors of STM, and the transcript level of msgA was downregulated after the knockdown of speG (29). Twelve proteins encoded by eut operation were detected in the proteomic analysis. Of these, 10 exhibited reduced expression in ΔmsgA. In conclusion, MsgA may help STM survive intracellularly by promoting the uptake of ethanolamine.

The proteomic data in this study revealed a downregulation of the TCA cycle in ΔmsgA. This suggests that MsgA may be able to influence the important pathways for the supply of energy. Previous research has suggested that FNR and ArcA promote the TCA cycle and oxidative phosphorylation (47, 59). Our proteomic data indicated that the expression of Fnr and ArcA is reduced in the knockout strain, indicating that MsgA may promote the TCA cycle and oxidative phosphorylation by enhancing their expression. Furthermore, the TCA cycle and its intermediates have been shown to influence the virulence of STM: disruption of the TCA cycle results in reduced virulence (60), and exogenous succinate increases intracellular survival and virulence (16). These findings indicate that MsgA participates in the TCA cycle, thus impacting energy metabolism and the expression of virulence genes in STM.

Furthermore, enrichment analysis indicated an upregulation of fatty acid synthesis and a downregulation of fatty acid degradation, indicating a possible increase in fatty acid content. Fatty acids have been shown to influence bacterial membrane synthesis and virulence (61); free long-chain fatty acids can directly bind to HilA and inhibit SPI-I transcription (62). In addition, Ac-CoA serves as a crucial intermediate in fatty acid metabolism, as well as in the TCA cycle and ethanolamine metabolic pathway. Thus, changes in these metabolic pathways may affect the concentration of Ac-CoA. Ac-CoA also has been implicated in regulating the expression of virulence genes. Ac-CoA supplies the necessary acetyl group for the acetylation of PhoP (63). Acetylation of the PhoP protein K88 affects its DNA-binding function, thus impacting the regulation of SPI-II (64). The acetylation of HilD by Pat maintained HilD stability and was essential for the transcriptional activation of HilA (65). The above results indicate that MsgA might have the ability to regulate the expression of virulence genes by affecting different metabolites of Salmonella.

In this study, the proteomic and RT-qPCR results demonstrate a downregulation of SPI-I and its activators in ΔmsgA (Fig. 4 and 5), which may lead to a reduction in the invasion capacity of ΔmsgA. In addition, most reported factors that promote SPI-II expression showed downregulation in ΔmsgA (Fig. 4 and 6a). Moreover, The RT-qPCR results indicate a significant downregulation of structural genes and effector factors of the T3SS encoded by SPI- II (Fig. 6b and C). Interestingly, prior research has suggested that CsrA and Fis positively regulate SPI gene expression (66, 67), but in this study, their expression was upregulated in the knockout strain. The regulatory factor ssrB of SPI-II inhibits the expression of SPI-I by binding to HilA. However, in this study, both SPI-I and SPI-II showed downregulated expression. The above results suggest that MsgA affects the expression of SPI genes through different signaling pathways and has a high priority. Combining the results of proteomics and RT-qPCR, we hypothesized that there is a regulatory network between MsgA and SPI (Fig. 8). The more precise regulatory networks require further exploration.

Fig 8.

Part of the regulatory network between MsgA and SPI; dashed lines represent potential regulatory relationships, solid lines represent proven regulatory relationships, arrows represent facilitation, and blunt ends represent inhibition.

Numerous studies have shown that SPI-I and SPI-II play a significant role in interactions between STM and hosts. SPI-I regulates the ability of Salmonella to invade host cells, which is closely associated with gastrointestinal disease but less involved in systemic infections (68–70). In vitro experiments have demonstrated that invasion genes also induce neutrophil migration through polarized epithelial monolayers and macrophage apoptosis (71, 72). SPI-II controls Salmonella intracellular survival and impacts both localized and systemic infection capabilities (5–8, 10). The reduction in SPI expression levels implies an attenuation of Salmonella invasive capacity and intracellular survivability. In addition, as expected, we demonstrated that msgA plays a key role in the invasive capacity and intracellular survivability of Salmonella by gentamicin protection assays, and the virulence is significantly decreased after knocked down msgA (Fig. 7).

Salmonella Typhimurium VNP20009 possessed a safety profile that included reduced pathogenicity in animal models based on genetically stable deletions of the msbB and purI genes and a high degree of antibiotic susceptibility (73). In clinical phase I trial, Salmonella Typhimurium VNP20009 demonstrated both tumor colonization ability and toxicity in the high-dose group; this means that the dosing is limited by the virulence of Salmonella (74). Therefore, we need to further reduce the virulence to improve the efficacy of Salmonella. Similar to msbB and purI, msgA was also independent of the virulence island, and knockdown of msgA further improved the safety of Salmonella. There was no significant difference in body weight between the mice injected with the ΔmsgA and those injected with PBS. This indicates that a new attenuated strain, which has better safety than VNP20009, can be constructed based on msgA.

DATA AVAILABILITY

The proteome data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the data set identifier PXD051177.

ETHICS APPROVAL

All animal experiments were approved by the Nanjing University Institutional Animal Care and Use Committee (IACUC-2003167).

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/aem.00201-24.

Supplemental figures and tables. aem.00201-24-s0001.docx.

Figures S1 to S5; Table S1.

Table S2. aem.00201-24-s0002.xlsx.

Accession numbers for all sequences used in multiple sequence alignment and phylogenetic tree construction.

Table S3. aem.00201-24-s0003.xlsx.

Proteomic data.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.