Adaptation mechanisms of Brucella abortus to low magnesium ion stress

Background

Brucella abortus, a facultative intracellular pathogenic bacterium that usually causes diseases under animals and humans, can survive and replicate within phagocytic cells. Within the host cells, B. abortus has to adapt to low cytosolic magnesium ion (Mg²⁺) environment, which is critical for bacterial survival and replication. To understand the fitness of B. abortus under the low Mg²⁺ environment, transcriptome analysis was performed by RNA-seq. Results: 262 differentially expressed genes (DEGs, fold-change > 1.5 and p < 0.05) of B. abortus, 123 significantly upregulated genes and 139 significantly downregulated genes, were identified under Mg²⁺ starvation environment, highlighting that B. abortus probably employed large amounts of factors to support the adaptation of low Mg²⁺ stress responses. Amongst them, two key genes, BAB_RS26550 (encoding putative protein, abbreviated as HP3) and BAB_RS26555 (encoding MgtC/SapB family protein, abbreviated as MgtC), was associated with the ATP hydrolysis to maintain the growth and metabolism of B. abortus under Mg²⁺ starvation environment. Furthermore, the HP3 supported B. abortus to resist bactericidal polycations and polymyxin B, as well as influenced the biofilm formation of B. abortus. However, HP3 does not appear to have an appreciable effect on the B. abortus virulence. Conclusions: In this study, a first description of the pattern of B. abortus genetic expression in response to low Mg²⁺ stress response provides insights into the intracellular behavior of B. abortus at the genetic level.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12917-025-04831-8.

Introduction

Brucella abortus, a Gram-negative and facultative intracellular pathogenic bacterium, usually causes brucellosis, a global zoonotic disease characterized by reproductive disease in domestic animals and chronic debilitating disease in humans [1, 2]. The most important feature of Brucella virulence is the ability to invade and replicate within professional and non-professional phagocytes, such as macrophages, dendritic cells (DCs), placental trophoblasts and epithelial cells [3]. During the course of infection, Brucella will experience different stress environments, such as acid conditions, nutrient limitation, and then modulate their gene expression assisting Brucella to survive and replicate within host cells [4]. Within the macrophage phagosome, the intracellular bacterial pathogen also encounters a host-imposed Magnesium ion (Mg²⁺) limitation [5, 6].

Mg²⁺, the most abundant multivalent cation in all living cells with a concentration ranging from 5 to 20mM [7], is required for assembly, structure and function of the ribosome [8, 9], stabilization of RNA and DNA [10–12]. Moreover, Mg²⁺ counteracted the negative charges on adenosine triphosphate (ATP) [8], and neutralized the negative charges in outer membrane lipopolysaccharide (LPS) [13]. These results indicated that Mg²⁺ is a key ion for living cells. Furthermore, Mg²⁺ is an extracellular signal that controls the PhoP/PhoQ two-component regulatory system of Salmonella typhimurium [14]. Upon the Mg²⁺ starvation environment, the expression of many factors including the Mg²⁺ transporters MgtA and MgtB [15], the enzymes that modify LPS [16], the F₁F_o ATP synthase inhibitory protein MgtC [17] and polyamines [5], were directly activated. Brucella strains were predicted to contain Mg²⁺ transport system based on surveys of the publicly available genome sequences, and require Mg²⁺ as an essential micronutrient for optimal growth in vitro [18, 19]. Homologs of two genes associated with Mg²⁺ transport in Salmonella typhimurium have been genetically linked to the virulence in Brucella strains. One Mg²⁺ transporter, MgtB, was involved in Brucella melitensis pathogenesis in experimentally infected mice [20]. In Salmonella, MgtC inhibits the activity of the F₁F_o ATP synthase by direct interaction, hindering ATP-driven proton translocation and NADH-driven ATP synthesis in inverted vesicles [17]. Brucella suis MgtC was associated with its growth in a Mg²⁺-restricted medium and its virulence in the murine macrophage-like J774 cell line [21]. B. abortus employed MgtC to maintain its metabolism through supporting the ATP hydrolysis in low Mg²⁺ environment [26]. However, the research about adaptation of B. abortus under the low Mg²⁺ stress response is missing. Therefore, it is of interest to identify the resistance determinants responsible for the adaptation of B. abortus to low Mg²⁺ stress responses.

A description of the pattern of genetic expression in response to low Mg²⁺ stress response would possibly provide insights into the intracellular behavior of Brucella at the genetic level. In the present study, we reported that five continuous genes upregulated in B. abortus under low Mg²⁺ conditions were identified by RNA-seq. Among them, two genes, hypothetical protein gene (BAB_RS26550) and mgtC (BAB_RS26555), were associated with the bacterial growth under low Mg²⁺ conditions by influencing bacterial ATP concentrations. The obtained results provide relevant information in the adaptation of B. abortus to low Mg²⁺ stress responses.

Materials and methods

Strains, plasmids, macrophages, and culture conditions

All strains and plasmids used in this study are listed in Table 1. B. abortus 2308 from the National Center for Veterinary Culture Collection (CVCC, China) and its derivatives were cultured in tryptic soy broth (TSB) or on tryptic soy agar (TSA) (Difco, Franklin Lakes, NJ, USA) at a temperature of 37 °C with 5% CO₂ in a BSL3 biosafety laboratory. E. coli DH5α employed in recombination and transduction experiments were cultivated in Luria-Bertani (LB) medium at a temperature of 37 ℃. When necessary, the culture medium was supplemented with 100 µg/mL ampicillin and 25 µg/mL of chloramphenicol. B. abortus 2308 and its derivatives were grown in M9 medium supplemented with 0.1% casamino acids, 1% yeast extract, 1% arabinose and 10 µM MgCl₂ (low Mg²⁺ conditions) or in M9 medium supplemented with 0.1% casamino acids, 1% yeast extract, 1% arabinose and 10 mM MgCl₂ (high Mg²⁺ conditions). RAW264.7 cells (ATCC, Manassas, VA, USA) were cultured at a temperature of 37 °C with 5% CO₂, using Dulbecco’s modified Eagle’s medium (DMEM) supplemented with either 10% or 1% heat-inactivated fetal bovine serum (Gibco, Thermo Scientific Grand Island, NY, USA).

Table 1.

Strains and plasmids used in the present study

Names	Description	Source
Strains
B. abortus	Wild-type strain 2308	CVCC
S2308ΔHP1	HP1 deleted mutant	This study
S2308ΔMgtA	MgtA deleted mutant	This study
S2308ΔHP2	HP2 deleted mutant	This study
S2308ΔHP3	HP3 deleted mutant	This study
S2308ΔMgtC	MgtC deleted mutant	This study
S2308ΔHP1-MgtC	HP1, MgtA, HP2, HP3, MgtC deleted mutant	This study
E. coli DH5α	F − φ80lacZ∆M15∆(lacZYA-argF)U169 recA1 endA1 hsdR17(rk−,mk+) phoA supE44 thi-1 gyrA96 relA1 λ−	Invitrogen
Plasmids
pUC19-sacB	AmpR; pUC19 derived plasmid containing sacB gene	[43]
pUC19-sacB-ΔHP1	AmpR; pUC19-sacB containing the upstream and downstream fragments of the HP1 gene	This study
pUC19-sacB -ΔMgtA	AmpR; pUC19-sacB containing the upstream and downstream fragments of the MgtA gene	This study
pUC19-sacB -ΔHP2	AmpR; pUC19-sacB containing the upstream and downstream fragments of the HP2 gene	This study
pUC19-sacB -ΔHP3	AmpR; pUC19-sacB containing the upstream and downstream fragments of the HP3 gene	This study
pUC19-sacB -ΔMgtC	AmpR; pUC19-sacB containing the upstream and downstream fragments of the MgtC gene	This study
pUC19-sacB-ΔHP1-MgtC	AmpR; pUC19-sacB containing the upstream and downstream fragments of the five consecutive genes HP, MgtA, HP2, HP3, MgtC gene	This study
pBBR1MCS1	Cmr; Broad-host-range cloning vector; parental plasmid	[44]
pBBR1MCS1- HP1	Cmr; pBBR1MCS1 plasmid containing the HP1 gene	This study
pBBR1MCS1-MgtA	Cmr; pBBR1MCS1 plasmid containing the MgtA gene	This study
pBBR1MCS1- HP2	Cmr; pBBR1MCS1plasmid containing the HP2 gene	This study
pBBR1MCS1- HP3	Cmr; pBBR1MCS1plasmid containing the HP3 gene	This study
pBBR1MCS1-MgtC	Cmr; pBBR1MCS1 plasmid containing the MgtC  gene	This study
pBBR1MCS1-HP1-MgtC	Cmr; pBBR1MCS1 plasmid containing the five consecutive genes HP, MgtA, HP2, HP3, MgtC gene	This study

Plasmids and mutant construction

Primers for construction were designed using the sequence of the target genes HP1, MgtA, HP2, HP3, MgtC or HP1-MgtC in the B. abortus S2308 genome (GenBank Code: NC_007618.1). All primers used in this study are listed in Table S1. The mutant strain was constructed according to the previously published protocol [22]. Suicide plasmids were constructed, using an overlap PCR assay. Briefly, the upstream and downstream fragments of the target genes HP1, MgtA, HP2, HP3, MgtC or HP1-MgtC were amplified in two independent PCR reactions using PrimeSTAR Max Mix (TaKaRa, Dalian, China) with primer pairs, and then recovered PCR products were used as templates for overlap PCR to produce joint sequences with primer pairs. The product of overlap PCR that contained joined flanking sequences was purified by gel extraction, digested with XbaI and SacI, and ligated into a pUC19-sacB plasmid to generate the suicide plasmids. The suicide plasmids were propagated in E. coli DH5α cells (Invitrogen Corp., Carlsbad, CA, USA) and then extracted to construct the mutant strains. These vectors were sequenced to confirm its accuracy. Allelic replacement was employed to delete the target genes HP1, MgtA, HP2, HP3, MgtC or HP1-MgtC from the wild-type strain S2308. S2308 was cultured and collected by centrifugation at the exponential phase. After an ice bath for 15 min, S2308 was washed twice with ice-cold sterile water. Bacteria were resuspended in 10% (v/v) glycerin water and 3–5 µg suicide plasmid was added on ice. Then the suicide plasmids were transferred to S2308 by electroporation. After electroporation, transformed S2308 were immediately transferred to prewarmed TSB media and cultured overnight. The first exchanged recombinants were selected by plating on TSA containing ampicillin at 100 µg/mL. The second round of exchanged recombinants was selected by plating on TSA containing 5% sucrose, then the sucrose-resistant colonies were respectively culture on the TSA plate and TSA plate supplemented with 100 µg/mL ampicillin. At least ten colonies that grown on the TSA plate and did not grow on the TSA plate supplemented with 100 µg/mL ampicillin were collected for identification with PCR using outer primer pair of the target genes HP1, MgtA, HP2, HP3, MgtC or HP1-MgtC up-F/down-F, and inner primer pair of the target genes HP1, MgtA, HP2, HP3, MgtC or HP1-MgtC inner-F/inner-R (Table S1). Colonies with length-reduced fragment from up-F/down-F pair and no fragment from inner-F/inner-R were selected as the target genes HP1, MgtA, HP2, HP3, MgtC or HP1-MgtC deleted mutant.

To complement the function of each deleted gene, a complement plasmid was constructed using the broad host-range plasmid pBBR1MCS1. The target gene fragment was amplified by PCR using the primers and gel purified. The recovered fragment was digested with the restriction enzymes BamHI and HindIII, and then ligated into the BamHI- and HindIII- digested plasmid pBBR1MCS1. The recombinant complement plasmid was propagated in DH5α cells and extracted to construct the complement strain. These vectors were sequenced to confirm its accuracy. Then the complementary plasmids were transferred to the corresponding mutant by electroporation. The complementary strains were selected by plating on TSA containing chloramphenicol. Analyses of PCRs were carried out to identify clones.

Determination of bacterial growth curves in high or low Mg²⁺ conditions

B. abortus 2308 and its derivatives were cultured in TSB to optical density 600 nm (OD₆₀₀) at 0.6, and then the bacteria were washed twice with phosphate buffered saline (PBS) (HyClone, GE Lifesciences, Logan, UT, USA) to remove the TSB medium. The bacteria were centrifuged at 8000 × g for 5 min to precipitate bacteria, and the pellets of strains were resuspended in 1mL M9 medium. The cell suspension was serially diluted 10-fold with PBS and spread onto TSA plates to calculate viable bacteria (CFU). To analyze bacterial responses to Mg²⁺, 1 × 10⁸ CFU of each strain were added to 20 mL M9 medium containing 0.1% casamino acids, 1% yeast extract, 1% arabinose supplemented with 10 µM Mg²⁺ (low Mg²⁺ conditions) or 10 mM Mg²⁺ (high Mg²⁺ conditions). Each strain was incubated at 37 °C with 200 rpm, and OD₆₀₀ of bacteria under different conditions were measured at the same time points. Bacterial growth curves were recorded by measurement of OD₆₀₀ throughout the incubation period.

RNA extraction and Real-Time PCR

As previously reported [22], total RNA was extracted from bacteria using the RNA simple Total RNA Extraction Kit (TianGen, Beijing, CN) according to the manufacturer’s protocol. Genomic DNA contamination was removed through treatment with a ReverTra Ace^® qPCR RT Master Mix with gDNA Remover kit (TOYOBO, Shanghai, CN). The RNA quantity and quality were evaluated by calculation of RNA concentration and OD₂₆₀/OD₂₈₀ ratio using the Thermo Scientific NanoDrop One (Thermo Fisher Scientific, Waltham, MA, USA), and the RNA integrity was assessed by calculation of RNA integrity number using standard denaturing agarose gel electrophoresis. 1 µg RNA was reverse transcribed into cDNA, using a ReverTra Ace^® qPCR RT Master Mix with gDNA Remover kit (TOYOBO, Shanghai, CN) according to the manufacturer’s instructions. A 20 µL RT-PCR mixture was made comprising 10 µL 2×ChamQ Universal SYBR qPCR Master Mix (Vazyme, NanJing, CN), 1 µL cDNA, 0.5 µL (each) forward and reverse primers (10 µM each), and 8 µL double-distilled water (ddH₂O). The mixture was incubated at 95 °C for 2 min, and then subjected to 40 cycles at 95 °C for 15 s, followed by 60 °C for 1 min using the Applied Biosystems™ 7500 (Thermo Fisher Scientific). All samples were analyzed in triplicate and relative transcription levels of each gene were determined by the 2^−∆∆Ct method, using 16 S RNA as an internal control for data normalization. All primers used for RT-PCR are listed in Table S1.

RNA-Seq

Total RNA was extracted from B. abortus 2308 cultured under high or low Mg²⁺ minimal medium M9 supplemented with 0.1% casamino acids, 1% yeast extract and 1% arabinose at the mid-phase of cell growth (OD₆₀₀ = 0.6–0.8), respectively, as described above. Subsequently, the cell samples were subjected to RNA-seq global transcriptomic analysis by Shanghai Majorbio Bio-pharm Technology Co.Ltd, and all samples were analyzed in triplicate. The data were analyzed via the online Majorbio I-Sanger CloudPlatform (https://cloud.majorbio.com). To ensure the quality of sequencing data, SeqPrep (https://github.com/jstjohn/SeqPrep) and Sickle (https://github.com/najoshi/Sickle) with the default settings were used to filter raw data. The genome of B. abortus 2308 was used as a reference, and the cleaned reads were aligned to the reference transcriptome using Bowtie 2 (http://bowtie-bio.sourceforge.net/index.shtml). To identify the differentially expressed genes (DEGs) between the samples, gene level transcripts per million reads (TPM) were generated using RSEM Version 1.3.1 (http://deweylab.github.io/RSEM) and DEGs were screened by DESeg2 (http://bioconductor.org/packages/stats/bioc/DESeq2) with FC > 1.5 and p-adjust < 0.05. Then DEGs were further subjected to function annotation and pathway enrichment analysis by hypergeometric distribution testing using Gene Ontology (GO) (http://www.geneontology.org/) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (http://www.genome.jp/kegg/).

Determination of intracellular ATP

ATP concentrations in the bacteria were detected using the ATP content detection kit (Solarbio Lifesciences, Beijing, China) according to the manufacturer’s protocol. Briefly, B. abortus 2308 and its derivatives were cultured under high or low Mg²⁺ minimal medium M9 supplemented with 0.1% casamino acids, 1% yeast extract and 1% arabinose at the mid-phase of cell growth (OD₆₀₀ = 0.6–0.8), and then the CFU of bacteria were assessed by serially diluting and plate counting. 1 × 10⁷ CFU bacteria were collected by centrifuging at 8,000 × g for 5 min to precipitate bacteria before the experiment. The pellet of bacteria resuspended in the buffer. After treatment, the bacteria were subjected to ultrasonic disruption and 10,000 × g for 10 min at 4 ℃, supernatant was collected for detection. The supernatant was mixed with chlorine and vortexed for 15 to 20 s, followed by centrifuging at 10,000 × g at 4 °C for 3 min. The absorbance of the supernatant at 340 nm was measured.

Stress resistance assays

Stress resistance assays were performed to determine the sensitivity of B. abortus 2308 and its derivatives to oxidative stress, cationic bactericidal peptides, acidic stress, Fe²⁺ stress, osmolarity stress, and nitrosative stress, as previously reported [23].

2.5 mM or 5 mM H₂O₂, 200 µg/mL or 500 µg/mL polymyxin B medium, pH 4.5 or 5.5 acid medium were used to determine sensitivity of B. abortus 2308 and its derivatives to oxidative stress, cationic bactericidal peptides, and acidic stress. Briefly, B. abortus 2308 and its derivatives were cultured to the mid-logarithmic phase (OD₆₀₀ = 0.6–0.8) in TSB, and then the bacteria were diluted with PBS to a concentration of 5 × 10⁸ CFU/mL. 5 µL of bacteria in PBS was mixed with 200 µL of the appropriate medium. After incubation at 37 ℃ with 5% CO₂ for 2 h, the bacterial survival percentages were calculated as: CFU obtained from bacteria treated with different factors / CFU obtained from bacteria in PBS × 100. The calculation is normalized by the volume used in the experiment (CFUs/ml). The results are expressed as the mean percentage of triplicate samples ± standard deviation (SD), and three independent experiments were performed.

4mM 2, 2-bipyridine (Fe²⁺ chelating agent), 200 mM sodium chloride (NaCl), 0.5 mM sodium nitroprusside (SNP) was used to assess sensitivity to Fe²⁺ stress, osmolarity stress, and nitrosative stress. Briefly, B. abortus 2308 and its derivatives were cultured to the mid-logarithmic phase in TSB, and then the bacteria were diluted with PBS to a concentration of 5 × 10⁹ CFU/mL. The suspension was serially diluted by 10-fold and 5 µl of the solution was blotted on the TSA containing the appropriate medium and cultured for 3–5 days at 37 °C. TSA without SNP was used as a normal growth control.

Crystal Violet staining

Biofilm formation assay was performed using crystal violet staining according to published protocols [24]. Briefly, overnight cultures of B. abortus 2308 and its derivatives were diluted to 1 × 10⁶ CFU/mL in TSB. 200 µL aliquots were transferred into the corresponding wells of 96 well microtiter plate and incubated at 37 °C for 48 h and 96 h. After that, the culture supernatant was discarded and cells were rinsed with PBS. 200 µL crystal violet solution (0.1%, v/v) was added to each well and incubated at 37 °C for 20 min. Subsequently, crystal violet solution was removed, and wells were washed with PBS 4 times in order to remove unattached cells and residual crystal violet. 200 µL of 95% v/v EtOH was added to each well and optical density of released crystal violet was measured at 600 nm.

Bacterial adherence, invasion and cellular survival assay

Bacterial adherence, invasion and intracellular survival were tested using RAW 264.7 cells. Cells were plated in the 24-well plates (Corning, NY, USA) and infected with B. abortus S2308 or its derivatives at a multiplicity of infection (MOI) of 100 CFU per cell. To synchronize the infection, the infected plates were centrifuged at 400 × g for 5 min, and cells were then incubated at 37 °C with 5% CO₂ for 1 h. Cells were washed twice with PBS to remove nonadherent bacteria.

For adherence assays, wells of infected cells were incubated with 500 µL of 0.1% (v/v) Triton X-100 in sterile water for 15 min at 37 °C and the solution of lysed cells was serially diluted 10-fold and the CFU of bacteria was determined by plating on TSA for 3–5 days at 37 °C.

For invasion assays, the infected cells were washed twice with PBS and cultured in a medium containing gentamicin (100 µg / mL) to remove extracellular bacteria, and incubated at 37 °C with 5% CO₂ for 1 h. Cells were washed twice with PBS to remove gentamicin. CFU were counted to the invading bacteria.

For intracellular survival assays, after killing extracellular bacteria, the cells were incubated with DMEM containing gentamicin (20 µg / mL) and 1% FBS after being washed twice with PBS. Infected cells were washed thrice with PBS and lysed with 500 µL of 0.1% (v/v) Triton X-100 in sterile water for 15 min at 1, 4, 24, 48, and 72 h post-infection (h.p.i.). The solution of lysed cells was serially diluted 10-fold and the CFU of bacteria that survived in cells was determined by plating on TSA for 3–5 days at 37 °C. All samples were analyzed in triplicate.

Statistical analysis

Data were analyzed using GraphPad Prism 10.0 software (Graph Pad Software, San Diego, CA). Statistical significance was determined using the student’s t-test. One-way or two-way analysis of variance (ANOVA) followed by Dunnett’s multiple comparisons test was used for group analysis. A probability (p) value of < 0.05 was considered statistically significant.

Results

Transcriptome analysis of B. abortus S2308 under low Mg²⁺ conditions

Mg²⁺ is an essential micronutrient for optimal growth in vitro. To evaluate the effects of Mg²⁺ on the growth of B. abortus S2308, the growth curve of wild-type strain in a medium supplemented with 10 mM Mg²⁺ or in a medium with 10 µM Mg²⁺ were plotted by measuring the OD₆₀₀. Under the 10 µM Mg²⁺ medium, B. abortus S2308 survived well, but it exhibited a slow rate of growth compared to 10 mM Mg²⁺ medium at the same time points (Fig. 1A).

Fig. 1.

Transcriptome analysis of B. abortus S2308 under low Mg²⁺ conditions. B. abortus 2308 cultured under high (10 µM) or low (10 mM) Mg²⁺ minimal medium M9 supplemented with 0.1% casamino acids, 1% yeast extract and 1% arabinose at the mid-phase of cell growth (OD₆₀₀ = 0.6), Total RNA was extracted from bacteria and the expression of differentially expressed genes (DEGs) in B. abortus S2308 under high or low Mg²⁺ conditions were identified by RNA-seq. (A) The growth curve of B. abortus S2308 in high or low Mg²⁺ conditions. All the results are representative of three repeated experiments. Data were shown as the mean ± SD. The statistical significance of differences among groups was analyzed by two-way multivariate ANOVA with Tukey’s post hoc test: ns, no significance. ***: p < 0.001, and ****:p<0.0001. (B) Heatmap and hierarchical clusters analysis of DEGs in different comparison groups, low Mg²⁺ conditions (abbreviated to “SL”) vs. high Mg²⁺ conditions (abbreviated to “SH”). Each small box indicates the expression status of a certain gene in one sample, red for up-regulated and blue for down-regulated expression. (C) Volcano map of DEGs in different comparison groups. The genes are colored if they pass the thresholds for P value < 0.05 and fold change ≥ 1.5, red for up-regulated and blue for down-regulated. (D) GO enrichment–based cluster analysis of DEGs in different comparison groups, S2308_Low_MG vs. S2308_High_MG (mRNA: red column. up-mRNA: pale blue column. down-mRNA: green column) (E) KEGG enrichment–based cluster analysis of DEGs in different comparison groups, S2308_Low_MG vs. S2308_High_MG (mRNA: red column. up-mRNA: pale blue column. down-mRNA: green column)

To explore how Mg²⁺ concentrations regulated gene expression profiles of B. abortus, a comparative RNA-seq analysis of B. abortus S2308 under high or low Mg²⁺ conditions was conducted. The total RNA was extracted from B. abortus S2308 under high or low Mg²⁺ conditions at the mid-phase of cell growth (OD₆₀₀ = 0.6), and then the transcriptomes were sequenced to characterize differentially expressed genes (DEGs). The transcript data have been uploaded to the NCBI Sequence Read Archive (SRA) database (https://www.ncbi.nlm.nih.gov/sra), and the accession number is PRJNA1244925. A total of 3215 genes (99.6% of total genes in the B. abortus S2308 genome) were identified in this RNA-seq study, 262 DEGs (fold-change > 1.5 and p < 0.05) were found under low Mg²⁺ conditions (Supplementary File 2). Of these DEGs, 123 genes were significantly upregulated, 139 genes were significantly downregulated. According to the sequencing results, the hierarchical clustering heatmap (Fig. 1B) and the volcano plot (Fig. 1C) were plotted, displaying the significantly upregulated and downregulated DEGs in the pairwise comparisons.

GO enrichment–based cluster analysis was performed on DEGs, showing that most of them were related to the biological processes, cellular components and molecular function, such as metabolic process, cellular anatomical entity, binding, and catalytic activity (Fig. 1D). KEGG enrichment–based cluster analysis showed that the DEGs were associated with metabolism, genetic information processing, and environmental information processing, such as energy metabolism, signal transduction, and translation (Fig. 1E).

Five consecutive genes identified from RNA-seq analysis were upregulated in B. abortus S2308 under low Mg²⁺ conditions

Based on transcriptome analysis results, five consecutive genes in the B. abortus genome, BAB_RS26535 (encoding putative protein, abbreviated as HP1), BAB_RS26540 (encoding magnesium-translocating P-type ATPase, abbreviated as MgtA), BAB_RS26545 (encoding putative protein, abbreviated as HP2), BAB_RS26550 (encoding putative protein, abbreviated as HP3), and BAB_RS26555 (encoding MgtC/SapB family protein, abbreviated as MgtC), were significantly upregulated under low Mg²⁺ conditions (Fig. 2A). To confirm the authenticity of RNA-seq results, transcriptional levels of the five consecutive genes were measured in B. abortus S2308 under high or low Mg²⁺ conditions by RT-PCR. The results of RT-PCR for the five consecutive genes were all upregulated (Fig. 2B), consistent with the RNA-seq results.

Fig. 2.

Five consecutive genes identified from transcriptome analysis were upregulated in B. abortus S2308 under low Mg²⁺ conditions. (A) Schematic diagram of five consecutive DEGs and its fold change (FC) values obtained from RNAseq in B. abortus S2308 under high (10 µM) or low (10 mM) Mg²⁺ conditions. (B) qPCR verification of five consecutive DEGs in B. abortus S2308 under high or low Mg²⁺ conditions. (C&D) Growth curves of B. abortus S2308 (WT), S2308∆HP1-MgtC (in-frame deletion strain of five consecutive genes) and its complemented strain S2308∆HP1-MgtC(pBBR1MCS-HP1-MgtC) under high or low Mg²⁺ conditions. All the results are representative of three repeated experiments. Data were shown as the mean ± SD. The statistical significance of differences among groups was analyzed by one-way or two-way multivariate ANOVA with Tukey’s post hoc test: ns, no significance. *: p < 0.05. **: p < 0.01. and ****:p<0.0001. Only the statistical significance of differences between the wild-type strain and its mutant strain were shown in the figure, and there is no significance between the wild-type strain and the complemented strain

To determine the role of five consecutive genes in the adaptation of low Mg²⁺ conditions, an in-frame deletion strain of B. abortus S2308 five consecutive genes (from BAB_RS26535 to BAB_RS26555), named as S2308∆HP1-MgtC, and its complemented strain S2308∆HP1-MgtC(pBBR1MCS-HP1-mgtC) were constructed respectively, and the growth curve of the mutant and its complemented strain under high or low Mg²⁺ conditions was plotted. The results showed that under the high Mg²⁺ conditions, there is no significant difference among the growth of S2308, S2308∆HP1-MgtC and S2308∆HP1-MgtC(pBBR1MCS-HP1-mgtC) (Fig. 2C). However, under the low Mg²⁺ conditions, S2308∆HP1-MgtC exhibited a slow rate of growth compared to wild-type strain S2308 (Fig. 2D), and the growth curve of S2308∆HP1-MgtC(pBBR1MCS-HP1-mgtC) is similar to the wild-type strain S2308 (Fig. 2D).

HP3 and MgtC play a critical role in the adaptation of B. abortus to low Mg²⁺ stress responses

To determine whether HP1, MgtA, HP2, HP3 or MgtC of B. abortus is necessary for the growth of bacteria under Mg²⁺ starvation, we constructed the in-frame deletion of gene HP1, MgtA, HP2, HP3 or MgtC in B. abortus to avoid any polar effect on downstream genes and complemented the mutation with a very low-copy plasmid. The following B. abortus mutant strains were constructed: S2308∆HP1, S2308∆MgtA, S2308∆HP2, S2308∆HP3, S2308∆MgtC, S2308∆HP1(pBBR1MCS-HP1), S2308∆MgtA(pBBR1MCS-MgtA), S2308∆HP2(pBBR1MCS-HP2), S2308∆HP3(pBBR1MCS-HP3), S2308∆MgtC (pBBR1MCS-MgtC), respectively. Subsequently, the growth curves of the mutant and its complemented strain under high or low Mg²⁺ conditions were plotted. Under the high Mg²⁺ conditions, the growth of mutant S2308∆HP1, S2308∆MgtA, S2308∆HP2, S2308∆HP3 or S2308∆MgtC is similar to the wild-type strain S2308 (Fig. 3A-E). Under the low Mg²⁺ conditions, the growth of mutant S2308∆HP1, S2308∆MgtA, S2308∆HP2 is similar to the wild-type strain S2308 (Fig. 3F-H), and S2308∆HP3 or S2308∆MgtC exhibited a slow rate of growth compared to the wild-type strain S2308 (Fig. 3I and J).

Fig. 3.

HP3 and MgtC play a critical role for the adaptation of B. abortus to low Mg²⁺ stress responses. B. abortus 2308 and its derivatives were cultured under high (10 µM, up panel) or low (10 mM, down panel) Mg²⁺ conditions. Growth curves of B. abortus 2308 and its derivatives were plotted. (A) and (F) Growth curves of B. abortus S2308, S2308∆HP1 and its complemented strain S2308∆HP1 (pBBR1MCS-HP1). (B) and (G) Growth curves of B. abortus S2308, S2308∆MgtA and its complemented strain S2308∆MgtA(pBBR1MCS-MgtA). (C) and (H) Growth curves of B. abortus S2308, S2308∆HP2 and its complemented strain S2308∆HP2 (pBBR1MCS-HP2). (D) and (I) Growth curves of B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3). (E) and (J) Growth curves of B. abortus S2308, S2308∆MgtC and its complemented strain S2308∆MgtC(pBBR1MCS-MgtC). All the results are representative of three repeated experiments. Data were shown as the mean ± SD. The statistical significance of differences among groups was analyzed by one-way or two-way multivariate ANOVA with Tukey’s post hoc test: ns, no significance. *: p < 0.05. **: p < 0.01. ***: p < 0.001. and ****:p<0.0001. Only the statistical significance of differences between the wild-type strain and its mutant strain were shown in the figure, and there is no significance between the wild-type strain and the complemented strain

HP3 deletion affects ATP concentrations of B. abortus S2308 under low Mg²⁺ conditions

Mg²⁺ represents a powerful signal arising from interconversions of adenylates (ATP, ADP and AMP) [8, 25]. In this study, we found that HP3 and MgtC were associated with the adaptation of low Mg²⁺ environment in B. abortus S2308. Lavigne et al. have reported that MgtC was associated with Brucella suis growth in a Mg²⁺-restricted medium [21], and we have recently reported that B. abortus S2308 employed MgtC to maintain its metabolism through supporting the ATP hydrolysis in low Mg²⁺ environment [26]. Next, the connection between interconversions of adenylates and HP3 was detected in B. abortus S2308. To determine the HP3 function in bacteria ATP hydrolysis, ATP amounts of S2308, S2308∆HP3 and S2308∆HP3(pBBR1MCS-HP3) under high or low Mg²⁺ conditions were conducted. The results showed that under high Mg²⁺ conditions, there was no significant difference in ATP amounts among S2308, S2308∆HP3 and S2308∆HP3(pBBR1MCS-HP3) (Fig. 4), and under low Mg²⁺ conditions, the ATP amounts of S2308∆HP3 were significantly increased compared to S2308 and S2308∆HP3(pBBR1MCS-HP3) (Fig. 4), highlighting that HP3 deletion affects ATP concentrations of B. abortus S2308 under low Mg²⁺ conditions.

Fig. 4.

HP3 deletion affects ATP concentrations of B. abortus S2308 under low Mg²⁺ conditions. ATP concentrations in B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3) were detected under high (10 µM) or low (10 mM) Mg²⁺ conditions. ns: no significant difference. All the results are representative of three repeated experiments. Data were shown as the mean ± SD. The statistical significance of differences among groups was analyzed by one-way or two-way multivariate ANOVA with Tukey’s post hoc test: ns, no significance. ***: p < 0.001

S2308∆HP3 reduced the biofilm formation and its ability to resist polymyxin B, but not hydrogen peroxide, acidic pH, sodium Nitroprusside (SNP), ferric ion (Fe²⁺), and osmolarity (NaCl)

To assess the ability of S2308∆HP3 to resist bactericidal factors, 2.5mM or 5mM hydrogen peroxide medium, 200 µg/mL or 500 µg/mL polymyxin B medium, an acidic medium, Fe²⁺ medium, NaCl medium, and SNP medium were used to evaluate its sensitivity compared to that of S2308. The results showed that compared to the wild type strain S2308, the mutant S2308∆HP3 exhibited similar sensitivity to the hydrogen peroxide (Fig. 5A), acidic pH (Fig. 5C), Fe²⁺ (Fig. 5D), osmolarity (NaCl) (Fig. 5E), and SNP (Fig. 5E), suggesting that HP3 did not affect Brucella’s ability to resist killing by oxidative stress, acidic stress, Fe²⁺ stress, osmolarity stress, and nitrosative stress. However, compared to wild type strain S2308, the mutant S2308∆HP3 had a significantly reduced survival rate under the medium supplemented with 200 µg/mL or 500 µg/mL polymyxin B, suggesting that HP3 affects Brucella’s capacity to tolerate cationic bactericidal peptides (Fig. 5B). One of the most successful strategies of microbes against environmental fluctuations might be their ability to form biofilms [27]. The biofilm-forming capacity of the mutant S2308∆HP3 was examined via crystal violet assay, showing that the mutant S2308∆HP3 exhibited a decreased biofilm-forming capacity compared to the wild type strain S2308, and the decreased biofilm-forming capacity in the S2308∆HP3 was completely restored by HP3 complementation (Fig. 5F).

Fig. 5.

Stress resistance. (A) Oxidative stress. The sensitivity of B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3) to 2.5mM or 5mM H₂O₂. (B) Cationic bactericidal peptides. The sensitivity of B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3) to 200 µg/mL or 500 µg/mL polymyxin B medium. (C) Acidic stress. The sensitivity of B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3) to pH = 4.5 or 5.5 acid medium. (D) Fe²⁺ stress. The sensitivity of B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3) to 4mM 2, 2-bipyridine (Fe²⁺ chelating agent). (E) Osmolarity stress and nitrosative stress. The sensitivity of B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3) to 200 mM sodium chloride (NaCl) and 0.5 mM sodium nitroprusside (SNP). “A” represents S2308; “B” represents S2308∆HP3; “C” represents S2308∆HP3(pBBR1MCS-HP3); (F) Biofilm formation assay was performed in B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3) using crystal violet staining. All the results are representative of three repeated experiments. Data were shown as the mean ± SD. The statistical significance of differences among groups was analyzed by one-way or two-way multivariate ANOVA with Tukey’s post hoc test: ns: no significant difference. *: P < 0.05. **: p < 0.01

HP3 did not affect B. abortus S2308 adherence, invasion and survival in RAW264.7 cells

In this study, the ability of S2308, S2308∆HP3 and S2308∆HP3(pBBR1MCS-HP3) to adhere to and invade RAW264.7 cells was evaluated. The adherence and invasion of the S2308∆HP3 to RAW264.7 cells was comparable to that of S2308 (Fig. 6A). B. abortus is a intracellular bacterium and its virulence relies mainly on its ability to invade and replicate within professional and non-professional phagocytes [22]. To evaluated the function of HP3 in B. abortus virulence, the survival ability of S2308, S2308∆HP3 and S2308∆HP3(pBBR1MCS-HP3) in RAW264.7 cells was evaluated at 1, 4, 24, 48, and 72 h after infection. No statistical differences in intracellular survival among S2308, S2308∆HP3 and S2308∆HP3(pBBR1MCS-HP3) were detected (Fig. 6B), indicating that HP3 did not affect B. abortus S2308 virulence in RAW264.7 cells.

Fig. 6.

HP3 did not affect B. abortus S2308 adherence, invasion and survival in RAW264.7 cells. Intracellular survival within RAW 264.7 macrophages. RAW264.7 cells were infected with B. abortus S2308, S2308∆HP3 and its complemented strain S2308∆HP3(pBBR1MCS-HP3) at a MOI of 100. (A) Adherence and invasion detection. After incubating at 37 °C with 5% CO₂ for 1 h, adherent bacteria were enumerated. After culturing in a medium containing gentamicin (100 µg / mL) to remove extracellular bacteria, invasion bacteria were enumerated. (B) Cellular survival detection. The infected cells were incubated with 0.2% (v/v) Triton X-100 at 1 h, 4 h, 24 h, 48 h and 72 h p.i. Then, intracellular CFU of bacteria were enumerated. All the results are representative of three repeated experiments. Data were shown as the mean ± SD. The statistical significance of differences among groups was analyzed by one-way or two-way multivariate ANOVA with Tukey’s post hoc test: ns, no significant difference

Discussion

Brucella is a facultative intracellular bacterium that was characterized by persisting for long periods within the host cells and establishing a chronic infection [28]. Within the host cells, Brucella will experience low cytosolic Mg²⁺. Mg²⁺ acts as a specific signaling molecule implicated in numerous intracellular functions [29]. RNA sequencing (RNA-seq) has become an indispensable tool for transcriptome-wide analysis of differential gene expression [30]. Here, we showed the adaptation of B. abortus to low Mg²⁺ stress by a comprehensive analysis of global changes in the transcriptome levels.

Bacteria must acquire Mg²⁺ in a coordinated manner to achieve balanced growth and avoid loss of viability [31]. In this study, we found that the growth rate of B. abortus S2308 significantly decreased under low Mg²⁺ conditions at the same time points (Fig. 1A), suggesting that in vitro, Mg²⁺ is essential in B. abortus growth and metabolism. When Salmonella was exposed to Mg²⁺ starvation environment, the expression of many genes directly activated upon low extra-cytoplasmic Mg²⁺ [32, 33]. Similarly, When B. abortus faces Mg²⁺ starvation environment, 262 DEGs (fold-change > 1.5 and p < 0.05), 123 significantly upregulated genes and 139 significantly downregulated genes, were found in this study (Fig. 1B and C), suggesting that some of these genes was probably associated with the B. abortus growth and metabolism in the low Mg²⁺ stress responses. Mg²⁺ is involved in a multitude of essential biological processes including energy production, translation, replication, transcription, protein function and stability, through its interactions with most polyphosphate compounds, such as ATP, guanosine triphosphate (GTP), as well as other nucleotide triphosphates (NTPs), and its roles in enzyme catalysis [8, 25, 34–36]. Indeed, GO enrichment–based cluster analysis and KEGG enrichment–based cluster analysis showed that most DEGs were also associated with metabolic processes, cellular anatomical entity, binding, catalytic activity, energy metabolism, signal transduction, and translation (Fig. 1D and E), suggesting that Mg²⁺ is involved in a multitude of essential processes in B. abortus, which well consisted with the previous reports [7–10].

Interestingly, amongst 123 significantly upregulated DEGs identified from RNA-seq, five consecutive DEGs, HP1, MgtA, HP2, HP3, MgtC, were significantly upregulated in Brucella under Mg²⁺ starvation environment (Fig. 2). MgtA and MgtB, the members of the broad family of P-type ATPases, are responsible for transporting numerous biologically important transition metals including Ca²⁺, Na⁺/K⁺, H⁺/K⁺, Mg²⁺, Co²⁺, Zn²⁺, as well as lipids, using ATP hydrolysis to fuel transport [36, 37]. Here, we found that the in-frame deletion of gene MgtA did not affect the growth of B. abortus under the high or low Mg²⁺ conditions (Fig. 3B and G). The possible explanation is that another Mg²⁺ transporter in the B. abortus MgtA mutant was exploited to present the compensation function. Indeed, another Mg²⁺ transporter, mgtE, was significantly upregulated in B. abortus under Mg²⁺-deprived environments among the DEGs identified from transcriptome analysis (Supplementary File 2). In the future studies, generating the single (mgtE) and double (mgtE and mgtA) mutants to provide a more comprehensive understanding of the observed phenotypes will be further conducted. MgtC, first described in S. typhimurium, is a virulence factor required for survival inside macrophages, full virulence in mice and growth in a low Mg²⁺ medium [38]. Consistent with the reports in S. typhimurium, B. abortus requires MgtC to grow in Mg²⁺-deprived environments (Fig. 3J). Additionally, three putative proteins, HP1, HP2, HP3, were all significantly upregulated in B. abortus under Mg²⁺ starvation environment, and among them, only HP3 is responsible for the B. abortus growth in Mg²⁺-deprived environments (Fig. 3I). However, the functions of the upregulated HP1 and HP2 need to be uncovered in future studies.

Mg²⁺ is an essential cofactor in all phosphoryl transfer reactions that involve ATP, GTP, as well as NTPs [25]. In Salmonella, MgtC interacts with a subunit of the F1Fo ATP synthase, hindering ATP-driven proton translocation and NADH-driven ATP synthesis in inverted vesicles leading to lower intracellular ATP levels [17]. We have recently reported that B. abortus employed MgtC to maintain its metabolism by supporting the ATP hydrolysis in a low Mg²⁺ environment [26], which is consistent with the results that B. abortus requires MgtC to grow under Mg²⁺-deprived environments in this study. Next, our study focused on another factor HP3 which is also responsible for the B. abortus growth in Mg²⁺-deprived environments. Here, the direct link between growth status and ATP levels in B. abortus was detected. We found that the ATP levels in the B. abortus HP3 deletion strain were significantly increased under low Mg²⁺ conditions (Fig. 4), demonstrating that similar to MgtC, factor HP3 was also associated with the ATP hydrolysis to maintain the growth and metabolism of B. abortus under low Mg²⁺ conditions. Considering that HP3 was associated with the adaptation of B. abortus to low Mg²⁺ stress responses, the ability of HP3 in B. abortus to resist bactericidal factors was accessed, showing that the HP3 deletion decreased the sensitivity of B. abortus to bactericidal polycations, polymyxin B (Fig. 5B). Moreover, biofilm-forming capacity was also decreased in the B. abortus HP3 mutant (Fig. 5F). Mg²⁺ plays crucial roles in protein synthesis by maintaining the structure of ribosomes, participating in the biochemistry of translation initiation and functioning as a counterion for ATP [45]. We hypothesized that HP3 deletion maybe result in a damaged component of the cell membrane by affecting the Mg²⁺ concentration in B. abortus, and the damaged cell membrane changed the tolerance of B. abortus to polymyxin B and biofilm formation of B. abortus. Additionally, we found that HP3 did not affect Brucella ability to resist killing by oxidative stress, acidic stress, Fe²⁺ stress, osmolarity stress, and nitrosative stress.

Bioinformatics analysis showed that HP3 protein contains a transmembrane region (amino acid positions: 21–44) and a signal peptide with cleavage occurring between amino acid positions 49 and 50. The signal peptide located behind the transmembrane region is a characterization of membrane proteins. Similarly, GO annotation analysis indicated that the HP3 protein is localized to the cell membrane. We hypothesize that the HP3 protein may be involved in sensing changes of external Mg²⁺ concentrations and transmit signals into the cell, which is similar to the function of ion channels, membrane receptors, or other membrane-associated proteins.

B. abortus is an intracellular parasite, and its virulence is mainly reflected in its intracellular survival ability. After invading cells, B. abortus forms Brucella-containing vacuole (BCV), and various virulence factors will assist BCV break through various intracellular pressure environments, such as acidic environment, nutrient deficiency environment, and low Mg²⁺ environment, to migrate to the endoplasmic reticulum for replication [3, 38]. Among these virulence factors, MgtC is a key factor for bacteria to adapt to low Mg²⁺ environment and was responsible for the survival within macrophage in Salmonella enterica, Yersinia pestis, Yersinia pseudotuberculosis, Mycobacterium tuberculosis, Burkholderia cenocepacia, and Brucella suis [14, 21, 39–42]. In this study, the factor MgtC were identified from RNA-seq and play a critical role for in the adaptation of B. abortus to low Mg²⁺ stress responses. However, unlike MgtC, another factor HP3 identified from RNA-seq in this study did not affect the B. abortus adherence, invasion and intracellular survival (Fig. 6), suggesting that HP3 does not appear to have an appreciable effect in the B. abortus virulence. Further, the B. abortus HP3 mutant is inhibited in liquid culture at 10µM Mg²⁺ environment but not during macrophage infection, suggesting that 10µM is perhaps lower than the intracellular macrophage environment.

In this study, there are some limitations that need to be further addressed in order elucidate the functions of other DEGs identified from RNA-seq in the adaptation of B. abortus to low Mg²⁺ stress.

Conclusions

Altogether, this is a first description of the pattern of B. abortus genetic expression in response to low Mg²⁺ stress, providing insights into the behavior of B. abortus at the genetic level. Two factors, MgtC and HP3, played a key role in supporting the adaptation of B. abortus to low Mg²⁺ stress and its biological functions were elucidated, providing evidence for its possible mechanisms.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Peng Li, Nan Wang and Jianxia Jiang conceived and designed the experiments; Hengtai Wang mainly performed the experiments and analyzed the data; Lang Lv, Yike Huang, Liangquan Zhu, Huai-Ming Sang and Lei Xu helped to perform some experiments; Peng Li and Hengtai Wang wrote the manuscript; Hui Jiang, Xiaowen Yang, Jianxia JiangNan Wang and Jiabo Ding revised the manuscript and coordinated the research. All authors have read and approved the manuscript.

Funding

This work was supported by funds from the National Key Research and Development Program of China (2023YFD1800703), Innovation Program of Chinese Academy of Agricultural Sciences (CAAS-CSAB-202403), the Beijing Natural Science Foundation (6244054).

Data availability

All data supporting the findings of this study are available within the paper.

Declarations

Ethical approval

The experimental protocol was approved by Experimental Ethical Committee of Institute of Animal Science, Chinese Academy of Agricultural Sciences. B. abortus 2308 and its derivatives were manipulated in the Biosafety Level-3 Laboratory (BSL-3) of National/WOAH Reference Laboratory for Brucellosis, FAO Reference Centre for Brucellosis, China Institute of Veterinary Drug Control, Beijing, China.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.