Introduction

Salmonella is a major foodborne pathogen worldwide that can cause self-limiting gastroenteritis or life-threatening systemic disease in a wide range of animals (Fàbrega & Vila, 2013; Ohl & Miller, 2001). Salmonella infection remains a significant global public health concern. An estimated 93.8 million cases of gastroenteritis and 27 million cases of systemic diseases caused by Salmonella species occur annually worldwide, with 355,000 deaths (Kim et al., 2019; Majowicz et al., 2010). The ability to survival and replicate in host macrophages is a key determinant for Salmonella to induce systemic infection (Fields, Swanson, Haidaris, & Heffron, 1986; LaRock, Chaudhary, & Miller, 2015; Leung & Finlay, 1991). After internalization by macrophages, Salmonella delivers a set of more than 30 effector proteins to the macrophage cytoplasm, mainly through a type III secretion system (T3SS) encoded by Salmonella pathogenicity island-2 (SPI-2) (Pillay et al., 2023). SPI-2 effectors manipulate diverse cellular processes to promote the formation of a membrane-bound compartment, termed the Salmonella-containing vacuole (SCV), a niche where Salmonella resides and grows (Castanheira & García-Del Portillo, 2017; Rosenberg, Riquelme, Prince, & Avraham, 2022; Steeb et al., 2013). SCV protects Salmonella from contact with antimicrobial agents in macrophages (Figueira & Holden, 2012; Li et al., 2023).

As the SCV of macrophages is a nutrient-poor environment, to effectively replicate in the SCV, Salmonella needs to acquire a wide range of host nutrients or host-derived metabolites and synthesize metabolites de novo that cannot be sufficiently accessed from the host (Dandekar et al., 2014; Röder, Felgner, & Hensel, 2021; Tuli & Sharma, 2019). Nutrients/metabolites are used by intracellular Salmonella either as carbon sources to generate energy or for the synthesis of fatty acids and proteins (Dandekar et al., 2014; Steeb et al., 2013). Moreover, several metabolites were found to be employed by Salmonella as environmental cues to induce the expression of virulence genes (Jiang et al., 2021; X. Wang et al., 2023). In recent years, an increasing number of studies have focused on the intracellular nutrition of Salmonella; however, the nutrients that are required for Salmonella replication in macrophages remain largely unknown.

β-Alanine, also known as 3-aminopropionic acid (3-AP), is the only naturally occurring β-type amino acid and is found in all living organisms. β-Alanine can be synthesized de novo by bacteria, fungi, and plants, whereas animals need to obtain it from food or generate it via the catabolism of cytosine and uracil. In bacteria, β-alanine is synthesized via the decarboxylation of L-aspartate, a reaction catalyzed by L-aspartate decarboxylase (PanD) (Begley, Kinsland, & Strauss, 2001; Schmitzberger, Smith, Abell, & Blundell, 2003). The panD gene is conserved among most bacteria. Although β-alanine is a nonprotein amino acid that is not incorporated into proteins, it has important physiological functions in the metabolism of organisms (Yuan et al., 2022). First, β-alanine forms a part of pantothenate (vitamin B5), which is the key precursor for the biosynthesis of coenzyme A (CoA) (Webb, Smith, & Abell, 2004; White, Gunyuzlu, & Toyn, 2001). CoA is an essential cofactor involved in many metabolic pathways, including the synthesis and degradation of fatty acids, pyruvate oxidation through the tricarboxylic acid (TCA) cycle, and the production of secondary metabolites (Davaapil, Tsuchiya, & Gout, 2014; Gout, 2019; Sibon & Strauss, 2016; Theodoulou, Sibon, Jackowski, & Gout, 2014). Second, β-alanine is a limiting precursor of carnosine, a nonenzymatic free radical scavenger and a natural antioxidant, with anti-inflammatory and neuroprotective effects in animals (Boldyrev, Aldini, & Derave, 2013; Hoffman, Varanoske, & Stout, 2018). In the past 15 years, β-alanine has become one of the most commonly used sports supplements worldwide (Bellinger, 2014; Hoffman et al., 2018; Huerta Ojeda, Tapia Cerda, Poblete Salvatierra, Barahona-Fuentes, & Jorquera Aguilera, 2020). Although both Salmonella and host cells are capable of producing β-alanine, whether β-alanine contributes to the pathogenicity and intracellular growth of Salmonella remains unknown.

In this work, using targeted metabolic profiling, in vitro and in vivo infection assays, and many other molecular techniques, we demonstrated that the utilization of β-alanine is essential for Salmonella replication in host macrophages and virulence in mice. Salmonella acquires β-alanine both via the uptake of β-alanine from host macrophages and the de novo synthesis of β-alanine. Further investigation revealed the molecular mechanism underlying the contribution of β-alanine to Salmonella intracellular replication and pathogenicity, wherein β-alanine promotes the expression of zinc transporter genes to facilitate the uptake of the essential micronutrient zinc by intracellular Salmonella, therefore promoting Salmonella replication in macrophages and subsequent systemic infection. Taken together, these findings demonstrate a correlation between Salmonella β-alanine utilization and zinc uptake during intracellular infection and provide new insights into the intracellular nutrition of Salmonella. The rate-limiting gene (panD) in the β-alanine synthesis pathway of Salmonella might be a future target for the prevention and treatment of this pathogen.

Results

Host-derived β-alanine promotes Salmonella replication inside macrophages

To explore changes in the levels of different amino acids inside macrophages upon Salmonella infection, we performed targeted metabolomics analysis of mouse RAW264.7 macrophages that were mock-infected or infected with wild-type Salmonella (Salmonella enterica serovar Typhimurium ATCC 14028s, STM) for 8 h using liquid chromatography-tandem mass spectrometry (LC‒MS/MS) (Figure 1A). Principal component analysis (PCA) demonstrated a clear separation between the mock- and Salmonella-infected groups (Figure 1B). A total of 26 free amino acids were analyzed, and 8 showed significant differences in abundance between the two groups (VIP > 1 and P < 0.05; fold change > 1.5 or < 0.667) (Figure 1C). Compared with those in the mock-infected group, the concentrations of 3 amino acids (L-hydroxyproline, L-citrulline and L-cysteine) were upregulated, and 5 amino acids (L-asparagine, L-serine, L-aspartate, β-alanine and γ-aminobutyric acid) were downregulated in the Salmonella-infected group (Figure 1C). Consistent with previous findings, intracellular serine concentrations were downregulated due to the reprogramming of macrophage glucose metabolism during Salmonella infection (Jiang et al., 2021). Salmonella can use host-derived aspartate and asparagine for growth in macrophages (Popp et al., 2015); therefore, the decrease in intracellular aspartate and asparagine upon Salmonella infection is likely due to their utilization by bacteria. Interestingly, β-alanine concentrations were also downregulated in the Salmonella-infected group (Figure 1C, D), suggesting that intracellular Salmonella may use host-derived β-alanine for growth.

Figure 1.

To investigate whether host-derived β-alanine can promote intracellular Salmonella replication, we added an additional 1 mM β-alanine (Schneider, Krämer, & Burkovski, 2004) to the culture medium (RPMI) of RAW264.7 cells and then infected them with Salmonella to test the influence of β-alanine addition on the ability of Salmonella to replicate in macrophages. The results showed that the replication of Salmonella in RAW264.7 cells significantly (P < 0.01) increased with the addition of β-alanine (Figure 1E), suggesting that host-derived β-alanine promotes Salmonella replication inside macrophages. We then investigated whether β-alanine-mediated Salmonella growth promotion is due to the changes in antimicrobial activity of the macrophages, via quantitative real-time PCR (qRT‒PCR) assays and flow cytometry analysis. We observed that the addition of 1 mM β-alanine did not influence the expression of pro- (Il1b, Tnf) and anti-inflammatory genes (Il10, Il4ra) of Salmonella-infected RAW264.7 cells (Figure 1F), and did not affect the percentage of pro-inflammatory M1 macrophages (CD86+) and anti-inflammatory M2 macrophages (CD163+) during Salmonella infection (Figure 1G), implying that the addition of β-alanine to macrophages does not change their immune response. Combining these results, we can further infer that Salmonella use host-derived β-alanine for intracellular growth.

Direct validation of Salmonella using host-derived β-alanine for intracellular growth requires a mutant that has a defect in β-alanine uptake. Escherichia coli uptakes β-alanine via the transporter protein CycA (Schneider et al., 2004). However, the Salmonella ΔcycA mutant was able to use β-alanine as the sole carbon source for growth in minimal medium (Supplementary Figure 1A), indicating that CycA is not a transporter for β-alanine in Salmonella and that other unknown β-alanine import mechanisms are involved. Consistent with these results, mutation of cycA did not influence the replication of Salmonella in RAW264.7 cells (Supplementary Figure 1B) or colonization in mouse systemic tissues (liver and spleen; Supplementary Figure 1C).

De novo β-alanine synthesis is critical for Salmonella replication inside macrophages

Salmonella can de novo synthesize β-alanine via the decarboxylation of L-aspartate, which is catalyzed by L-aspartate decarboxylase (PanD) (Figure 2A) and is reportedly the rate-limiting step of β-alanine generation (Begley et al., 2001; Schmitzberger et al., 2003). To further assess the role of β-alanine in Salmonella intracellular replication, we analyzed the expression level of the Salmonella panD gene in macrophages and the impact of panD mutation on the ability of Salmonella to replicate in macrophages. Quantitative real-time PCR (qRT‒PCR) assays revealed that the expression level of panD was significantly (P < 0.01) greater in RAW264.7 cells than in RPMI-1640 medium (Figure 2B). Increased expression of panD was also observed in N-minimal medium, a widely used medium that mimics the conditions inside macrophages, as revealed by qRT‒PCR and bioluminescent reporter assays (Figure 2C, D). These results demonstrate that panD expression is enhanced during Salmonella growth inside macrophages, suggesting a relationship between panD expression and intracellular Salmonella growth.

Figure 2.

We then constructed the panD mutant strain ΔpanD and compared the replication ability of the ΔpanD strain and the wild-type (WT) strain in RAW264.7 cells. Gentamicin protection assays showed that the replication of ΔpanD in RAW264.7 cells decreased 2.4-fold at 20 h postinfection compared with that of the WT strain (P < 0.01), while complementation of ΔpanD with the panD gene restored the replication ability of the mutant strain in RAW264.7 cells (Figure 2E). Immunofluorescence analysis revealed that the number of ΔpanD in each infected RAW264.7 cell was comparable to that of the WT strain at the initial infection stage (2 h), but at 20 h postinfection, the number of ΔpanD in each infected RAW264.7 cell was significantly (P < 0.0001) lower than that of the WT strain (Figure 2F, G). These results indicate that panD contributes to Salmonella replication in macrophages. The growth rates of ΔpanD in LB medium and RPMI medium resembled those of the WT (Supplementary Figure 2A, B), indicating that the impaired intracellular replication ability of the mutant was not due to a growth defect. Moreover, the replication defect of ΔpanD in RAW264.7 cells was relieved by the addition of 1 mM β-alanine to the RPMI medium (Figure 2H). These data collectively suggest that β-alanine synthesis is critical for Salmonella replication inside macrophages.

De novo β-alanine synthesis is critical for systemic Salmonella infection in mice

As replication in macrophages is a key determinant of systemic Salmonella infection, we reasoned that β-alanine synthesis also influences Salmonella virulence in vivo. BALB/c mice were infected by i.p. injection of 5,000 CFU of WT, ΔpanD, or the complemented strain cpanD. The survival rate, body weight, bacterial burden in the liver and spleen, and liver histopathological alterations of the infected mice were measured (Figure 3A). The WT-infected mice exhibited high lethality and marked loss of body weight within 5 days, and all mice died within 9 days of infection (Figure 3B, C). In contrast, the ΔpanD-infected mice displayed significantly improved survival rates and body weights, and no mice died within the 10-day surveillance period (Figure 3B, C). Consistent with these results, the bacterial burden in the liver and spleen of the ΔpanD-infected mice was significantly decreased, and the body weight was significantly increased compared with that of the WT-infected mice on day 3 postinfection (Figure 3D). Complementation of ΔpanD with panD significantly decreased the survival rate and body weight of infected mice but significantly increased the bacterial burden in the liver and spleen of the infected mice (Figure 3B-D). Furthermore, H&E staining revealed increased aggregation of inflammatory cells and pyknosis in the livers of the WT-infected mice on day 5 postinfection, while these histopathological alterations were obviously reduced in the livers of ΔpanD-infected mice (Figure 3E). Taken together, these results reveal that β-alanine synthesis is critical for systemic Salmonella infection in mice.

Figure 3.

β-Alanine is involved in the regulation of several metabolic pathways in Salmonella

To explore the mechanism(s) associated with β-alanine-mediated promotion of Salmonella replication in macrophages and in vivo virulence, we performed RNA sequencing (RNA-seq) to reveal the differences in gene transcripts between Salmonella WT and ΔpanD strains grown in N-minimal medium. PCA plot of the global transcriptomic profiles clearly demonstrated separation between the WT and ΔpanD strains (Figure 4A). Remarkable transcriptional changes were observed due to the mutation of panD. Compared with those in the WT strain, 1379 genes were differentially expressed in the ΔpanD strain, with 561 upregulated genes and 618 downregulated genes (fold change ≥ 2 and P value < 0.05; Figure 4B). Gene Ontology (GO) enrichment analysis revealed that the differentially expressed genes (DEGs) were mainly involved in the metabolism and biosynthesis of several amino acids (including arginine, leucine, histidine, and branched amino acids), carboxylic acid metabolism, small molecule biosynthesis, and aerobic respiration (Figure 4C). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis also revealed a high frequency of terms related to metabolism, including amino acid metabolism, lipid metabolism, carbohydrate metabolism, energy metabolism, and nucleotide metabolism (Figure 4D). These data collectively indicate that β-alanine is involved in the regulation of a series of metabolic pathways in Salmonella.

Figure 4.

Further analysis of the downregulated DEGs (activated by PanD) revealed that mutation of panD decreased the expression of genes involved in 7 pathways that are associated with the virulence of Salmonella or other bacterial pathogens, including methionine metabolism, fatty acid β-oxidation, histidine biosynthesis, and the transport of zinc, galactose, potassium, and polyamine (Figure 4E). Zinc and potassium uptake are associated with the virulence of Salmonella (zinc acquisition promotes Salmonella Typhimurium virulence in mice, and potassium acquisition promotes Salmonella Enteritidis virulence in chickens) (Ammendola et al., 2007; Battistoni, Ammendola, Chiancone, & Ilari, 2017; Ilari et al., 2016; Liu et al., 2013), while the remaining 5 pathways are involved in the pathogenicity of other pathogens (Basavanna et al., 2013; de Paiva et al., 2016; Feldman et al., 2016; Lauriano et al., 2004; Martínez-Guitián et al., 2019). In addition, the expression of the LysR-type transcriptional regulator LeuO, which activates the expression of the leuABCD leucine synthesis operon and numerous virulence genes in Salmonella Typhimurium (Dillon et al., 2012; Guadarrama, Villaseñor, & Calva, 2014; Hernández-Lucas et al., 2008), was also downregulated in the ΔpanD strain (Figure 4E). In line with the decreased expression of leuO, the expression of leuABCD was downregulated in the ΔpanD strain (Figure 4E).

We selected 16 downregulated DEGs (including the regulatory gene leuO and genes from the above 7 pathways) for qRT‒PCR analysis. The results showed that the expression of all 16 genes significantly (P < 0.05) decreased in the ΔpanD mutant compared with the WT strain (Figure 4F), and complementation of ΔpanD with panD restored the gene expression to the WT level (Figure 4F), thus confirming the positive regulation of these pathways and LeuO by β-alanine.

Although SPI-2 gene expression is essential for Salmonella replication in macrophages and systemic infection, the expression of SPI-2 genes was not influenced by panD mutation (Supplementary Figure 3A). In addition, the gene expression of 4 other virulence-associated pathogenicity islands, namely, SPI-1, SPI-3, SPI-4, and SPI-5, was not influenced by panD mutation (Supplementary Figure 3B, C).

Taken together, these data suggest that β-alanine might promote Salmonella intracellular replication and virulence by activating virulence-associated pathway(s) or activating the virulence-associated regulator LeuO, rather than by activating the expression of virulence genes encoded within pathogenicity islands.

β-Alanine promotes Salmonella virulence in vivo by increasing the expression of zinc transporter genes

Next, we inactivated the 7 downregulated pathways mentioned above, as well as the regulatory gene leuO in Salmonella, to uncover the mechanism(s) by which β-alanine promotes Salmonella virulence in vivo. Mouse infection assays revealed that mutations in fadAB, metR, hisABCDFGHL, kdpABC, mglABC, and potFGHI, which are associated with fatty acid β-oxidation, methionine metabolism, histidine biosynthesis, potassium uptake, galactose uptake, and polyamine uptake, respectively, did not influence Salmonella colonization in the mouse liver or spleen (Figure 5A, B) or the body weight of infected mice (Figure 5C). Interestingly, although LeuO has been reported to be associated with the regulation of a diverse set of virulence factors (Dillon et al., 2012; Guadarrama et al., 2014), mutation of the regulatory gene leuO did not influence Salmonella colonization in the mouse liver or spleen (Figure 5A, B) or the body weight of infected mice (Figure 5C). In contrast, mutation of the zinc transporter gene znuA significantly decreased Salmonella colonization in the mouse liver and spleen (Figure 5D, left and middle panels); this result is consistent with previous studies (Ammendola et al., 2007; Battistoni et al., 2017; Ilari et al., 2016). Accordingly, the body weight of ΔznuA-infected mice was significantly (P < 0.001) greater than that of the WT-infected mice (Figure 5D, right panel). These results indicate that β-alanine might promote Salmonella virulence in vivo by promoting zinc uptake.

Figure 5.

To test this hypothesis, we constructed a double mutant, ΔpanDΔznuA, and compared colonization of the mouse liver and spleen of the double mutant to that of the single mutant, ΔznuA. The results showed that colonization of the liver and spleen of infected mice by ΔpanDΔznuA was significantly lower than that of infected mice colonized by ΔznuA (Figure 5D, left and middle panels). In agreement with these results, the body weight of ΔpanDΔznuA-infected mice was greater than that of ΔznuA-infected mice (Figure 5D, right panel), suggesting that the contribution of panD to the virulence of Salmonella is partially dependent on znuA.

Collectively, these data indicate that β-alanine promotes in vivo virulence of Salmonella partially by increasing the expression of zinc transporter genes.

β-Alanine promotes Salmonella replication within macrophages by increasing the expression of zinc transporter genes

To determine whether β-alanine influences Salmonella intracellular replication by acting on zinc transporters, we analyzed the ability of ΔznuA and ΔpanDΔznuA to replicate in RAW264.7 macrophages via gentamicin protection assays. The results showed that the replication of ΔpanDΔznuA in RAW264.7 cells was significantly reduced compared with that of the single mutant ΔznuA (Figure 6A), implying that the contribution of panD to the intracellular replication of Salmonella is partially dependent on znuA. The addition of 100 μM Zn²⁺ to RPMI medium increased the replication of ΔpanD in RAW264.7 cells (Figure 6B), while the addition of 1 mM β-alanine to RPMI medium did not increase the replication of ΔznuA (Figure 6C), suggesting that the impaired replication due to the decrease in β-alanine can be relieved by zinc supplementation. Taken together, these data indicate that β-alanine promotes Salmonella replication within macrophages by increasing the expression of zinc transporter genes.

Figure 6.

Discussion

Replication within host macrophages is a crucial step for Salmonella to cause life-threatening systemic infection in the host (Bomjan, Zhang, & Zhou, 2019; Lathrop et al., 2015), while the crosstalk between Salmonella and macrophages at the metabolic interface is critical for intracellular Salmonella replication (Dandekar et al., 2014; Herrero-Fresno & Olsen, 2018; Lynch & Lesser, 2021; Rosenberg et al., 2021; Thompson, Fulde, & Tedin, 2018). Emerging evidence suggests that several metabolites affect the replication of Salmonella in macrophages. The promotion of intracellular replication by metabolites is possibly achieved in three ways: i) metabolites are utilized by Salmonella as nutrients for intracellular growth (Bowden, Rowley, Hinton, & Thompson, 2009; Eisenreich, Dandekar, Heesemann, & Goebel, 2010; J. Wang et al., 2021); ii) Salmonella senses metabolites as environmental cues to activate the expression of virulence genes (Jiang et al., 2021; X. Wang et al., 2023); and iii) metabolites can regulate the immune responses of macrophages (Michelucci et al., 2013; Peace & O’Neill, 2022; Yang & Cong, 2021). In this study, we demonstrated that Salmonella promotes its replication inside macrophages by utilizing both host- and bacterial-derived β-alanine (Figure 6D). We showed that β-alanine promotes Salmonella intracellular replication and systemic infection partially by increasing the expression of zinc transporter genes and therefore the uptake of zinc by intracellular Salmonella (Figure 6D). Therefore, this work identified another metabolite that can influence the replication of Salmonella in macrophages and illustrated the mechanism by which β-alanine promotes intracellular Salmonella replication.

We observed that Salmonella-infected macrophages contained lower β-alanine levels than mock-infected macrophages, while β-alanine supplementation in the cell medium increased the replication of Salmonella in macrophages, revealing that Salmonella uptakes host-derived β-alanine to promote intracellular replication. In addition, a deficiency in the biosynthesis of β-alanine (via mutation of the rate-limiting gene panD) reduced Salmonella replication in macrophages and systemic infection in mice, suggesting that Salmonella also utilizes bacterial-derived β-alanine to promote intracellular replication and pathogenicity. Therefore, we speculate that Salmonella relies on a large amount of β-alanine to efficiently replicate in macrophages, thereby highlighting the importance of β-alanine for Salmonella intracellular growth. Nevertheless, unlike the closely related species E. coli, which takes up β-alanine via the transporter protein CycA (Schneider et al., 2004), Salmonella does not use CycA to uptake β-alanine, implying the presence of other yet unknown β-alanine import mechanisms in Salmonella.

Several amino acids, including lysine, proline, arginine, aspartate and asparagine, have previously been reported to be associated with the pathogenicity of Salmonella (Popp et al., 2015; Steeb et al., 2013). These amino acids are involved in the synthesis of proteins in Salmonella. In contrast, β-alanine is not incorporated into proteins but can participate in the regulation of bacterial activity through the synthesis of pantothenate and CoA (Webb et al., 2004; White et al., 2001; Yuan et al., 2022). Accordingly, our transcriptome data showed that a deficiency in β-alanine biosynthesis affected the expression of Salmonella genes involved in a series of important metabolic pathways. Importantly, although β-alanine does not influence the gene expression of SPIs, it activates methionine metabolism; fatty acid β-oxidation; histidine biosynthesis; and the transport of zinc, galactose, potassium, and polyamine, which have been previously known to be associated with the virulence of Salmonella and other bacterial pathogens. Further analysis revealed that β-alanine promotes Salmonella intracellular replication and systemic infection partially by promoting the uptake of zinc. As for the other virulence-related metabolic pathways activated by β-alanine, methionine metabolism, fatty acid β-oxidation, and histidine biosynthesis contribute to the virulence of Streptococcus pneumoniae (Basavanna et al., 2013), Yersinia pestis (Feldman et al., 2016), and Acinetobacter baumannii (Martínez-Guitián et al., 2019), respectively; the uptake of galactose and polyamine influences the pathogenicity of Francisella tularensis (Lauriano et al., 2004) and the avian pathogenic Escherichia coli (de Paiva et al., 2016), respectively; and the uptake of potassium is associated with the virulence of Salmonella Enteritidis in chickens (Liu et al., 2013). However, blocking these pathways did not influence the systemic Salmonella Typhimurium infection in mice, implying that different bacterial pathogens adopt different virulence strategies to establish infection. Determining other mechanism(s) by which β-alanine promotes the intracellular replication and systemic infection of Salmonella require further investigation.

We observed that β-alanine also activates the expression of the LysR-type transcriptional regulator LeuO, which is known to regulate the expression of a wide variety of Salmonella genes that impact the stress response and virulence (Dillon et al., 2012; Guadarrama et al., 2014; Hernández-Lucas et al., 2008). Typically, LeuO activates the synthesis of the quiescent porins OmpS1 and OmpS2, which are required for Salmonella virulence in mice (De la Cruz et al., 2007; Fernández-Mora, Puente, & Calva, 2004; Rodríguez-Morales et al., 2006). Consistent with the positive regulation of OmpS1 and OmpS2 by LeuO, lack of leuO in Salmonella also attenuated virulence in a mouse model (Rodríguez-Morales et al., 2006). However, the attenuated phenotypes of the leuO mutant in mice were not evident after i.p. injection relative to oral infection, as a previous report showed that the competitive index for the leuO mutant indicated approximately 1,000-fold reduced colonization in mouse systemic tissues after oral infection but much less reduced colonization after i.p. injection (less than 10-fold) (Rodríguez-Morales et al., 2006). These results imply that LeuO might be predominantly associated with the invasion and intestinal infection of Salmonella but is weakly implicated in Salmonella intracellular replication and systemic infection. Therefore, it is not surprising that lack of leuO did not significantly affect Salmonella colonization in the systemic tissues of mice after i.p. injection, as revealed by our results.

Zinc is an essential micronutrient for all living organisms and is used as a cofactor for various enzymes and proteins (Bock, Müller, & Blankenfeldt, 2016; Ilari et al., 2016; Tan, Bramlett, & Lindahl, 2004). In bacteria, zinc-binding proteins account for approximately 5% of the bacterial proteome and play crucial roles in bacterial metabolism and virulence (Andreini, Banci, Bertini, & Rosato, 2006). Knockout of the zinc transporter ZnuABC reduces the virulence of Salmonella, Campylobacter jejuni, Haemophilus ducreyi, Moraxella, and urinary tract pathogenic Escherichia coli (UPEC) in the host (Ilari et al., 2016). Moreover, zinc is also utilized by Salmonella to subvert the antimicrobial host defense of macrophages by inhibiting NF-кB activation and impairing NF-кB-dependent bacterial clearance (Jennings, Esposito, Rittinger, & Thurston, 2018; Wu et al., 2017). In this work, we found that β-alanine facilitates the pathogenicity of Salmonella by promoting the expression of zinc transporter genes. The results demonstrate a correlation between Salmonella β-alanine utilization and zinc uptake during intracellular infection and provide evidence that β-alanine can influence the macrophage immune response by acting on zinc uptake.

Overall, our findings suggest a model in which Salmonella exploits host- and bacterial-derived β-alanine to efficiently replicate in host macrophages and cause systemic disease. We propose that Salmonella requires a large amount of β-alanine during intracellular infection. The utilization of β-alanine promotes Salmonella uptake of the essential micronutrient zinc, which was previously shown to be required for the metabolic needs of intracellular Salmonella and to subvert the antimicrobial defense of macrophages by Salmonella. These observations provide new insight into Salmonella pathogenesis and the crosstalk between Salmonella and macrophages during intracellular infection. Considering that the panD gene is present in the genomes of all Salmonella strains and that mutation of panD markedly reduced Salmonella replication ability in macrophages, as well as virulence in the mouse model, this gene may be used as a potential target to control systemic Salmonella infection.

Methods

Ethics statement

Six-week-old female BALB/c mice were obtained from Beijing Vital River Laboratory Animal Technology (Beijing, China). Mice were housed in barrier facilities under specific pathogen-free conditions with a 12 h light/dark cycle at a temperature of 24 ± 2 °C and a relative humidity of 50 ± 5%. Mice were fed a standard mouse chow diet, and they consumed food and water ad libitum throughout the experiment. All animal experiments were conducted in accordance with the policies of the Institutional Animal Care Committee of Nankai University (Tianjin, China) and performed under protocol no. 2021-SYDWLL-000029.

Cell culture

The RAW264.7 mouse macrophage-like cell line (ATCC TIB-71) was obtained from the Shanghai Institute of Biochemistry and Cell Biology of the Chinese Academy of Sciences (Shanghai, China). Cells were cultured in RPMI-1640 medium (Gibco #11879020) supplemented with 10% (v/v) fetal bovine serum (FBS, Gibco #10100147) at 37 ℃ with 5% CO₂. Cells were seeded in 24-well tissue culture plates at 1×10⁵ cells per well 24 h before infection.

Bacterial strains, plasmids, and growth conditions

The bacterial strains and plasmids used in this study are listed in Table S1. The Salmonella enterica serovar Typhimurium (STM) strain ATCC 14028s was used as the WT strain throughout this study and for the construction of the mutants. Mutant strains were generated using the λ Red recombination system with the plasmid pSIM17 (Jiang et al., 2017). To construct the complemented strain of ΔpanD, the amplified DNA fragments of the panD ORF and its upstream promoter were digested and inserted into the BamHI–EcoRI site of the low-copy-number plasmid pBR322. To generate the panD-lux reporter fusion, the amplification products of the panD promoter region were digested and cloned into the XhoI–BamHI site of the plasmid pMS402, which carries a promoter-less luxCDABE reporter gene cluster (Liang, Li, Dong, Surette, & Duan, 2008). The sequences of primers used for the construction of the strains are listed in Table S2. All the strains were verified by PCR amplification and sequencing.

Bacterial strains were conventionally grown overnight in Luria–Bertani (LB) medium (10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl) or in N-minimal medium (10 μM MgCl₂, 110 μM KH₂PO₄, 7.5 mM (NH₄)₂SO₄, 0.5 mM K₂SO₄, 5 mM KCl, 38 mM glycerol, and 0.1% [w/v] casamino acids) supplemented with appropriate antibiotics at 37 °C with shaking at 180 rpm or on LB agar plates. All antibiotics were used at their standard concentrations (chloramphenicol, 25 μg/mL; kanamycin, 50 μg/mL; ampicillin, 100 μg/mL; gentamicin, 10 or 100 μg/mL) unless otherwise mentioned.

Growth curve

Bacterial strains were conventionally grown overnight in LB medium. The next day, they were subcultured (1:100) in new LB medium and RPMI-1640 medium or subcultured in N-minimal medium supplemented with glycerol or β-alanine as the sole carbon source at 37 °C with shaking at 180 rpm. To measure the growth of bacteria, 200 μL of the bacterial cultures were transferred to the microplate wells. The absorbance (OD₆₀₀) of the bacteria was measured every half hour for 12 h with a Spark multimode microplate reader (Tecan).

Bioluminescent reporter assays

STM WT carrying the panD-luxCDABE fusion plasmid was conventionally grown overnight in LB medium, and the next day, the cells were subcultured (1:100) in new LB medium or N-minimal medium for 8 h. The luminescence of the cultured bacteria (200 μl) was measured with a Spark multimode microplate reader (Tecan). Moreover, the cultured bacteria (100 μl) were serially diluted and plated on LB agar plates to estimate bacterial CFUs. Bacterial CFUs were used to normalize luminescence values.

Salmonella infection of macrophages

Bacterial strains were conventionally grown overnight in LB medium to the late stationary phase, and the next day, the bacteria were diluted to 2 × 10⁶ CFUs/mL and opsonized in RPMI-1640 medium supplemented with 10% FBS for 15 min. The macrophage monolayers were infected with the opsonized bacteria culture (0.5 mL/well, multiplicity of infection (MOI) = 10) and centrifuged at 800×g for 5 min to synchronize infection. The infected cells were incubated for 30 min at 37 °C in 5% CO₂ and then washed three times with 1× PBS. Fresh RPMI-1640 medium containing 100 μg/mL gentamicin was added to the infected cells to kill extracellular bacteria. After 1 h, fresh RPMI-1640 medium containing 10 μg/mL gentamicin was added to the infected cells for the remainder of the experiment. To assess the intarcelluar growth of Salmonella, the infected cells were lysed with 1% Triton X-100 at 2 hpi and 20 hpi, and the abundance of the intracellular bacteria CFUs was estimated on LB agar plates. The relative fold replication of intracellular bacterial strains was denoted as the CFUs recovered at 20 hpi relative to those at 2 hpi. The relative fold change in replication was normalized to the number of RAW264.7 cells. When indicated, 1 mM β-alanine or 100 μM ZnSO₄ were added after 1 h of gentamicin treatment.

Targeted metabolomics analysis of amino acids in macrophages

RAW264.7 cells were mock-infected or infected with STM WT for 8 h. The cells were harvested and washed with precooled PBS solution to remove the medium. Cellular metabolites were extracted using ice-cold extraction solvent (40:40:20 vol/vol/vol acetonitrile:methanol:water, 0.1 M formic acid), incubated at −20 ℃ for 20 min, and then centrifuged for 10 min at 12,000×g and 4 ℃ to obtain the supernatant. Subsequently, the supernatant was transferred to an LC–MS vial and analyzed using ultrahigh-performance liquid chromatography (Acquity; Waters, Milford, MA, USA) coupled with mass spectrometry (Q Exactive Hybrid Quadrupole-Orbitrap; Thermo Fisher Scientific, Waltham, MA, USA). Metabolites were separated with a Luna NH₂ column (2 mm × 100 mm, 3 µm particle size; Phenomenex). Mobile phase A was 20 mM ammonium acetate (pH 9.0), and mobile phase B was acetonitrile containing 0.1% formic acid. The flow rate was 0.4 ml/min. Xcalibur 4.0 software (Thermo Fisher) was used for data acquisition and processing. Metabolite identification was achieved by high-resolution mass and retention time matching to authentic standards. The absolute quantification of amino acids was performed using the standard curve method, and the values were normalized to the cell number. Four biological replicates of each sample were analyzed.

Immunofluorescence staining

RAW264.7 cells were infected with STM WT or ΔpanD and the complemented strain cpanD as described above. After 2 and 20 h of cultivation, the infected cells were fixed for 15 min in 4% paraformaldehyde, washed in PBS, and permeabilized with 0.1% Triton X-100 in PBS for 15 min. The fixed samples were blocked in 5% bovine serum albumin for 30 min, followed by staining with a FITC-conjugated anti-Salmonella antibody (1:100 dilution, Abcam #ab20320) for 1 h at room temperature in the dark. The nuclei were then stained with DAPI (Invitrogen #D21490) for 2 min. A confocal laser scanning microscope (Zeiss LSM800) and ZEN 2.3 software (blue edition) were used to acquire and analyze the cell images (Objective lense: 40×; The number of intracellular bacteria per infected cell was estimated in random fields by Fiji-ImageJ).

RNA isolation

RNA was extracted from Salmonella strains cultured in N-minimal medium or LB medium. To investigate the expression of the Salmonella panD gene inside macrophages, we obtained RNA from intracellular bacteria in RAW264.7 cells at 8 h postinfection and from bacteria in RPMI medium. RNA was extracted using an EASYspinPlus bacterial RNA rapid extraction kit (Aidlab #RN0802) according to the manufacturer’s protocol. RNA quantity and purity were determined using a NanoDrop 2000 spectrophotometer (NanoDrop Technologies). RNA samples were stored at −80 °C before use.

Quantitative real-time PCR (qRT‒PCR)

According to the manufacturer’s protocols, qRT‒PCR was performed using 2× RealStar Power SYBR qPCR Mix (Genstar #A304) in a QuantStudio 5 Real Time PCR system (Applied Biosystems). cDNA was synthesized using a StarScript III RT Kit (Genstar #A232). Each sample was subjected to qRT‒PCR in triplicate. The expression level of the 16S rRNA gene was used to normalize that of the target genes. We estimated the expression of each target gene using the 2^−ΔΔCt method.

RNA sequencing and analyses

The STM WT and ΔpanD strains were conventionally grown overnight in LB medium, subcultured (1:100) in N-minimal medium for 8 h and then collected by centrifugation for RNA extraction. Sequencing libraries were generated using the NEBNext Ultra RNA Library Prep Kit for Illumina (New England Biolabs) according to the manufacturer’s instructions, and sequencing was conducted using the Illumina HiSeq 2000 platform at Shanghai Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China). The sequencing data were deposited in the NCBI Sequence Read Archive under accession number (SRA, PRJNA1124637). The clean reads were mapped to the STM ATCC 14028 reference genome (CP001363 and CP001362) by using the short-sequence alignment software Bowtie 2. Gene expression was evaluated using the fragments per kilobase of transcript per million mapped reads (FPKM) method. DEGs in the panD mutant relative to the WT were determined using the R statistical package software EdgeR. The thresholds for statistically significant differences were set to a fold change ≥ 2 or ≤ 0.5 and a false discovery rate (FDR) ≤ 0.05. P values were adjusted using the Benjamini–Hochberg procedure for controlling the FDR. Enrichment analysis of DEGs was conducted using GO and KEGG enrichment analyses.

Mouse infection

Salmonella strains were conventionally cultured overnight in LB medium, and the next day, they were subcultured (1:100) in new LB medium and grown at 37 °C with shaking at 200 rpm to stationary phase (OD₆₀₀∼2). The bacteria were diluted to 5 × 10⁴ CFUs/mL in 0.9% NaCl. Groups of BALB/c mice were infected i.p. with 0.1 mL of the NaCl suspension. For survival assays, we recorded and monitored the mortality and body weight of the infected mice daily for 10 days. To analyze the bacterial burden of the mouse liver and spleen, we weighed the infected mice first on day 3 postinfection and then harvested the liver and spleen. The liver and spleen of infected mice were homogenized in ice-cold PBS, serially diluted, and plated on LB plates containing the appropriate antibiotics to determine bacterial CFUs. H&E staining of the mouse liver was performed to investigate the histopathological alterations in the liver of infected mice. To evaluate the histopathological alterations in the mouse liver, we harvested the liver of the infected mice on day 5 postinfection. The mouse liver was washed with 0.9% NaCl, fixed with 10% neutral formalin for 48 h and subsequently processed for routine paraffin embedding. Paraffin-embedded tissues were sectioned at a thickness of 5 μm and then stained with hematoxylin (Sangon Biotech #E607318) and eosin (Sangon Biotech #E607318) for histopathological examination. The stained sections were then examined by light microscopy (Leica DM2500 LED).

Flow cytometry

RAW264.7 cells were infected with Salmonella WT for 8 h, in the absence or presence of 1 mM β-alanine, which was added to the infected cells at 1 h post-infection. The infected cells were fixed for 15 min in 4% paraformaldehyde, washed in PBS, and permeabilized with 0.1% Triton X-100 in PBS for 15 min. The fixed samples were blocked in 5% bovine serum albumin for 30 min, followed by staining with an anti-CD86 antibody (Abcam, ab119857), an anti-CD163 antibody (Abcam, ab182422) for 30 min, and a goat anti-rat IgG H&L (Alexa Fluor® 488) (Abcam, ab150165), a donkey anti-rabbit IgG H&L (Alexa Fluor® 647) (Abcam, ab150075) for 30 min in the dark. The infected cells were analyzed using a BD FACSAria Flow Cytometer (BD Biosciences).

Statistical analysis

The data are presented as the mean ± SD. Statistical analyses were performed using GraphPad InStat software (version GraphPad Prism 9.5.1, San Diego, CA, USA) with two-sided Student’s t tests, one-way ANOVA, two-way ANOVA, log-rank Mantel–Cox tests, or Mann‒Whitney U tests according to the test requirements (as stated in the figure legends). A P value < 0.05 indicated a statistically significant difference. ns represents no statistical significance.

Data availability statement

The RNA-seq data generated in this study have been deposited in the NCBI Sequence Read Archive (SRA) database under the accession number PRJNA1124637. All other data associated with this study are available in the main text and supplementary materials. Source data are provided in Supplementary Data 1.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (NSFC) Program (grant no. 32170110, 32270191), the Natural Science Foundation of Tianjin (grant no. 22JCYBJC01060), and the Fundamental Research Funds for the Central Universities, Nankai University (grant no. 63241588, 63243161).

Additional information

Author contributions

LJ designed the research; SM, YS, XW, HG, RL, TY, CK, and JC performed the research; BY provided technical support and insights; SM and BY analyzed the data; and LJ and SM wrote the manuscript.

Disclosure statement

The authors report there are no competing interests to declare.