The molybdate transport protein ModA regulates nitrate reductase activity to increase the intestinal colonization and extraintestinal dissemination of Klebsiella pneumoniae in the inflamed gut

ABSTRACT

The mammalian intestine is a major site of colonization and a starting point of severe infections by Klebsiella pneumoniae. Inflammatory bowel disease (IBD) is an inflammatory disorder of the gut, and host-derived nitrate in IBD confers a luminal growth advantage upon Escherichia coli and Salmonella typhimurium through nitrate respiration in the inflamed gut. However, the impact of nitrate on the growth and pathogenicity of K. pneumoniae in this microenvironment is poorly understood. In this study, we used oral administration of dextran sodium sulphate to induce IBD in mouse models. We then analysed the colonization levels of K. pneumoniae wild-type (WT), the nitrate reductase gene mutant strains (ΔnarG, ΔnarZ and ΔnarGΔnarZ), and the molybdate uptake gene mutant strain (ΔmodA) in the inflamed intestinal tract. Results showed that the growth, intestinal colonization, and extraintestinal dissemination of K. pneumoniae were increased in the intestines of dextran sulphate sodium (DSS)-treated mice. Nitrate in the inflamed bowel conferred a growth advantage to K. pneumoniae through nitrate respiration. The molybdate transport protein ModA regulated nitrate reductase activity to increase the growth, intestinal colonization, and extraintestinal dissemination of K. pneumoniae. Tungstate will be a promising antibacterial agent to tackle K. pneumoniae infections in IBD patients.

Introduction

Klebsiella pneumoniae is a facultative anaerobe that can colonize the mucosal epithelium of the human intestinal tract and oropharynx. The intestinal tract serves as the starting point of severe infections, such as pyogenic liver abscesses, meningitis, and endophthalmitis, caused by K. pneumoniae [1–4]. To survive and establish colonization in the host’s gastrointestinal tract, bacteria must overcome biological stresses and adapt metabolically to the host intestinal environment. The intestine is a hypoxic environment, and alternative electron acceptors (AEAs) such as dimethyl sulphoxide (DMSO), trimethylamine N-oxide (TMAO), and nitrate (NO₃⁻) are required to fuel oxidative phosphorylation for the growth of facultative anaerobes [5]. These AEAs are by-products of gut inflammatory responses and confer a growth advantage to some facultative anaerobes through anaerobic respiration.

Inflammatory bowel disease (IBD) is a chronic recurrent intestinal inflammatory disease, and the nitrate level is elevated during gut inflammation. Host-derived nitrate can be utilized by Escherichia coli and Salmonella typhimurium for anaerobic nitrate respiration to confer a fitness advantage in the inflamed gut [6–8]. Nitrate reductases are the key enzymes involved in nitrate respiration, and their activities require the incorporation of an essential molybdenum cofactor (MoCo) into the active site [9–11]. In most bacteria, the high-affinity uptake of molybdate from the environment to form MoCo depends on the molybdate uptake system, ModABC [12–14]. In E. coli and Pseudomonas aeruginosa, loss of Mo uptake results in reduced nitrate reductase activity and impaired growth on nitrate-containing media [6,15]. Compared to a ModABC system and three nitrate reductases in E. coli, the K. pneumoniae genome contains a ModABC molybdate uptake system and two nitrate reductases encoded by the narGHI and narZYV genes [16]. Little is known about the roles of nitrate in K. pneumoniae growth and pathogenicity in the intestinal microenvironment.

In patients with IBD, the prevalence of K. pneumoniae gastrointestinal carriage and the incidence of pyogenic liver abscess infections have increased compared to those in the control cohort [17–19]. Given the rising multidrug resistance, new strategies are needed to control the colonization and prevent infection of K. pneumoniae. The mechanisms by which IBD increases K. pneumoniae colonization in the gastrointestinal tract need to be better understood. This study used oral administration of dextran sodium sulphate (DSS) to induce IBD in mouse models. Moreover, we analysed the colonization levels of K. pneumoniae wild-type (WT), the nitrate reductase gene mutant strains (ΔnarG, ΔnarZ and ΔnarGΔnarZ), and the molybdate uptake gene mutant strain (ΔmodA) in the inflamed intestinal tract. This study aims to explore the roles of nitrate respiration and the molybdate uptake protein ModA in the growth and pathogenicity of K. pneumoniae in IBD patients. Due to the structural similarity between tungstate and molybdate, tungsten can inactivate nitrate reductase by competitively binding to ModA and replacing Mo in the MoCo [15,20]. It has been shown that tungstate can eliminate the fitness advantages of MoCo-dependent anaerobic respiration and selectively inhibit the expansion of Enterobacteriaceae to ameliorate colitis [21]. Therefore, the effect of tungstate as an antibacterial agent in decreasing K. pneumoniae infections in IBD patients will be studied.

Methods

Bacteria and culture conditions

Bacteria, plasmids, and primers used in the present study are summarized in Table 1. All K. pnemoniae strains were cultured overnight in Luria-Bertani (LB) broth at 37°C and 200 rpm, then diluted 1:100 into fresh LB broth and cultured to an optical density at 600 nm (OD₆₀₀) of 1.2 for standardization. For growth assays in vitro, the standardized cultures were diluted 1:100 into AT minimal medium containing 2 g (NH₄)₂SO₄, 0.078 g MgSO₄, 7.6 mg CaCl₂, 5 mg FeSO₄-7 h₂O, 2.2 mg MnSO₄-H₂O, 10.7 g KH₂PO₄ (pH 7.3), 1 mL 20 mm FeCl₃ and 2 mL freshly prepared glycerol per litre [22]. For anaerobic growth, 20 mm KNO₃ (Aladdin) was added to AT medium, and 100 μM Na₂MoO₄ or 10 mm Na₂WO₄ (Aladdin) was added as needed, followed by incubation in a compact anaerobic workstation DG250 (Don Whitley Scientific Ltd, Shipley, Hampshire, UK) for continuous culture. For intestinal infection, the standardized cultures were diluted 1:100 into fresh LB broth and cultured to an OD₆₀₀ of 2.0 with a 10-fold enrichment. E. coli DH5α was used as a cloning host and grown in LB medium.

Table 1.

Bacterial strains, Plasmids, and primers used in this study.

Bacteria and plasmids	Description	Reference or Source
Bacteria
K. pneumoniae
NTUH-K2044 (WT)	Wild-type strain; Apr; parent strain for generation of isogenic mutant strains	[23]
ΔmodA	Deletion of modA from WT; Apr	Present study
C-modA	Complemented modA mutant; Apr; Kmr	Present study
ΔnarG	Deletion of narG from WT; Apr	Present study
ΔnarZ	Deletion of narZ from WT; Apr	Present study
ΔnarGΔnarZ	Deletion of narG,narZ from WT; Apr	Present study
Plasmids
pKO3-Km	pKO3-derived plasmid, with an insertion of Km resistance cassette from pUC4K into AccI site	[25]
pGEM-T-easy-km	pGEM-T easy with an insert of Km cassette from pUC4K into NdeI site	[26]
Primers
modA Mutant constructs
modA-A	GTATGCGGCCGCATCCACCAACGGATGTTCGC
modA-B	GCCCTTTCAGGTAGTCATAGCTTACTCTCCTGTCAATGCG
modA-C	CGCATTGACAGGAGAGTAAGCTATGACTACCTGAAAGCGC
modA-D	GTATGCGGCCGCCAACGGCAACGTAATGGTC
modA-F	ATGGCAGGTTCTTGGTTACG
modA-R	ACTTTCTGATGCGAGGCTTC
Complemented modA mutant
C-modA-F	ATGGGCCCGAGATGGCCGCCAGCAGG
C-modA-R	TAAAGCGGCCGCTCAGCGGGTGGTAAATCCGT
narG Mutant constructs
narG-A	GTATGCGGCCGCTTCGGTATCTCGCTGGAT
narG-B	TTACGCTCTCCTGTACCTGGCGGGTTTCTCCTGATATGG
narG-C	CCATATCAGGAGAAACCCGCCAGGTACAGGAGAGCGTAA
narG-D	GTATGCGGCCGCAGGATCTCTTCCCAGTTCG
narG-F	ATGAAGATGTGGCGGGAAG
narG-R	GCAGAGTTTGGCGATTTCG
narZ Mutant constructs
narZ-A	GTATGCGGCCGCCTCCACCTTCCAGATGATCG
narZ-B	GTACCTGATCGCGACCTTGACATTTCTCCTGCTCCGAT
narZ-C	ATCGGAGCAGGAGAAATGTCAAGGTCGCGATCAGGTAC
narZ-D	GTATGCGGCCGCCCATCTTCCTCACGCTTGTAGA
narZ-F	CCAGCAGTTGTATCAGGACC
narZ-R	ACCGAGTTATGAATGCCG

Construction of deletion mutants and complemented strain

The deletion of modA, narG, and narZ from K. pneumoniae NTUH-K2044 (WT) [23] were carried out using the homologous recombination method described previously [24]. The left and right flanking sequences of modA, narG, and narZ were cloned into the temperature-sensitive suicide plasmid pKO₃-km [25] (with the sucrose-lethal gene SacB), and then electrotransformed into the WT strain. The successful integration of the flanking sequences into the chromosome was selected by growing the transformants on LB agar at 43°C. Several positive transformants were subcultured at 30°C with sucrose to eliminate the plasmid. The double knockout strain ΔnarGΔnarZ was generated by deleting narZ from the ΔnarG mutant. A 1,594-bp DNA sequence containing the promoter, coding region, and terminator of modA was cloned into pGEM-T-easy-km [26] and electrotransformed into ΔmodA to construct the complemented strain C-modA.

Nitrate reductase activity assay

The nitrate reductase activity of the K. pnemoniae strain was assayed as described by Hughes et al. with minor modifications [27]. Briefly, bacterial cultures were grown anaerobically in an AT medium supplemented with or without 20 mm KNO₃ for 3 h. Sixty μL of the cultures were transferred into 96-well plates, and a BioTek microplate reader (BioTek, Winooski, Vermont, USA) was used to measure OD₆₀₀. Then, 60 μL of 1% (w/v) sulfanilic hydrochloride (prepared in 18% HCl) and 0.1% N-(1-Naphthyl) ethylenediamine dihydrochloride were added, respectively. The samples were incubated for 10 min in the dark after mixing. OD₅₄₀ and OD₄₂₀ were measured, and the nitrate reductase activity was calculated using the following formula:

Nitrate reductase activity = (OD₅₄₀ − 0.72 × OD₄₂₀)/(0.18 × 3 × OD₆₀₀).

DSS-induced colitis model, K. pneumoniae intestinal infection, and tungstate treatment

Female 7- to 9-week-old C57BL/6 mice were purchased from HUNAN SJA Laboratory Animal Co., Ltd. Groups of C57BL/6 mice received either sterilized water (Mock) or a 3% (w/v) dextran sulphate sodium (DSS; molecular weight 36,000–50,000; MP Biomedicals) solution (dissolved in sterilized water and then filtered) for 8 d. For tungstate treatment, mice were administered 3% DSS and 0.2% (w/v) Na₂WO₄ solution. The experiment involved daily observation of mouse faeces and recording body weight. Drinking water was switched to sterilized water 24 h before the end of the experiment for DSS-treated mice. Mice were inoculated intragastrically with 0.1 mL of LB broth containing 1 × 10⁹ CFU of K. pneumoniae strains on day 5. Faecal samples were collected at 24 and 48 h post-inoculation from 4 to 5 mice per group, weighed, and evenly suspended in 1 mL of sterilized phosphate-buffered saline (PBS). Serial 10-fold dilutions were plated on LB agar plates to determine the bacterial loads. At 72 h post-inoculation, mice were euthanized and dissected. The liver, small intestine, and large intestine were collected, weighed, and homogenized in 2 mL of PBS using a tissue homogenizer. Serial 10-fold dilutions were plated on LB agar plates to calculate the bacterial loads.

Determination of nitrate concentrations in the cecal mucus layer

The nitrate concentrations in the caecal mucus layer were determined as described by Winter et al. with minor modifications [6]. Two groups of C57BL/6 mice received either water (n = 10) or 3% DSS (n = 12). Once successful induction of the colitis model was confirmed, the mice were euthanized on day 5. The caecum segments were dissected from the abdomen and weighed after carefully squeezing out the luminal contents. The caecum was then longitudinally cut, washed with PBS three times, and the mucus layer was gently scraped with a surgical blade. The caecal mucus was collected with 0.2 mL deionized water and centrifuged at 20,000 × g for 2 min at 4°C to remove larger particles. If not analysed immediately, the supernatant was sterilized by filtration and stored at −80°C. The obtained samples were mixed with a modified Griess reagent (0.1 g sulphanilamide, 0.25 g vanadium (III) chloride, 0.005 g N-(1-Naphthyl) ethylenediamine dihydrochloride in 100 mL of 0.5 M HCl) in a 1:1 ratio and derivatized at room temperature for 8 h. The absorbance of the reaction solution was measured at 540 nm. Meanwhile, serial 2-fold dilutions of sodium nitrate were prepared and treated as above to generate a standard curve. The nitrate concentration in the caecal mucus layer was calculated based on the standard curve.

Competitive index assay

The plasmid pLac-EGFP, containing an enhanced green fluorescent protein gene, and pBBR1MCS2-Tac-mCherry, containing a red fluorescence protein gene, were transformed into K. pnemoniae WT and the deletion mutants (ΔmodA, ΔnarG, ΔnarZ or ΔnarGΔnarZ). An equal mixture of WT and the respective deletion mutant was then inoculated into AT broth supplemented with or without 20 mm KNO₃, and incubated anaerobically at 37°C for 24 h. The cultures were serially diluted 10-fold and spread on LB agar plates. The colonies were identified and counted using a fluorescence microscope (General Electric Company, Boston, Massachusetts, USA). The competitive index was calculated by dividing the CFU of the recovered WT by the CFU of the deletion mutant. In a separate experiment, an equal mixture of WT and ΔmodA mutant was inoculated into AT broth containing 20 mm KNO₃ supplemented with or without 10 mm Na₂WO₄ for tungstate treatment.

Ethics statement

All animal experiments were approved by the Animal Care and Use Committee of Hubei University of Medicine (Reference Number: HBMU 2019–061) and complied strictly with the ARRIVE guidelines (https://arriveguidelines.org/).

Statistical analyses

The bacterial counts were log-transformed before analysis. The statistical differences between two groups were determined using an unpaired two-tailed Student’s t-test, while differences between multiple groups were assessed using one-way analysis of variance (ANOVA). For the competitive index assay, a paired two-tailed Student’s t-test was used.

Results

The ability of intestinal colonization and extraintestinal dissemination of K. pneumoniae are increased in the inflamed gut

Oral administration of DSS was used to induce the IBD mouse model. The DSS-treated mice exhibited bloody, loose faeces and significant weight loss on day 4 (Figure 1(a)), suggesting the successful induction of the IBD model. After intragastric inoculation with 1 × 10⁹ CFU of the K. pneumoniae WT strain, bacterial numbers in the faecal contents of DSS-treated mice were higher than those in placebo-treated mice (24 h: 10-fold, p = 0.035; 48 h: 1000-fold, p = 0.0027) (Figure 1(b)). Furthermore, the bacterial loads in the liver, large intestine, and small intestine of DSS-treated mice at 72 h post-inoculation were approximately 14, 60, and 100 times as those of the control group, respectively (Figure 1(c)). These results suggest that K. pneumoniae is more likely to proliferate and breach the intestinal barrier in DSS-treated mice, thus infecting the extraintestinal tissues such as the liver. Overall, these findings indicate that K. pneumoniae exhibits enhanced intestinal colonization and extraintestinal dissemination in the inflamed gut.

Figure 1.

The intestinal colonization and extraintestinal dissemination of K. pneumoniae are increased in the dss-induced IBD mouse model. Groups of SPF C57BL/6 mice were administered 3% DSS (DSS group) or water (mock group), and then inoculated intragastrically with 10⁹ CFU of K. pneumoniae NTUH-K2044 wild-type strain (WT) on day 5. (a) Body weight changes of the mice were monitored throughout the experiment. (b) Fecal bacterial counts of mice were determined at 24 and 48 h post-inoculation with the WT strain. (c) Bacterial loads in the liver, large intestine, and small intestine tissues were calculated 72 h after inoculation. Data are represented as means ± standard deviations (error bars). *,p < 0.05; **,p < 0.01 (unpaired two-tailed Student’s t-test).

Nitrate concentrations in the intestinal lumen are elevated in the inflamed gut

Nitrate, a by-product of the host inflammatory response, serves as a terminal electron acceptor to boost the growth of E. coli in the inflamed gut [6]. To assess the impact of intestinal nitrate on the development and pathogenicity of K. pneumoniae, the nitrate concentrations in the caecal mucus layer of mock-treated and DSS-treated mice were measured. The nitrate levels in DSS-treated mice were significantly higher than those in mock-treated mice (~2-fold, p = 0.0001, Figure 2), suggesting that the increased susceptibility of mice with IBD to intestinal infection by K. pneumoniae may be related to elevated nitrate production during gut inflammation.

Figure 2.

Nitrate concentration (NO₃⁻) in the caecal mucus layer of mock and dss-treated mice. C57BL/6 mice were divided into two groups: one receiving water (mock, n = 10) and the other 3% DSS (DSS, n = 12). On day 5, mice were euthanized, and caecum segments were dissected from the abdomen. The caecal mucus layer was collected, and nitrate concentrations were measured. ***,p < 0.001 (unpaired two-tailed Student’s t-test).

The in vitro anaerobic nitrate environment promotes the growth of K. pneumoniae

To investigate the effect of intestinal nitrate on the growth of K. pneumoniae, 20 mm KNO₃ was added to AT medium containing glycerol as the carbon source to simulate the nitrate-rich environment of gut inflammation in vitro. The aerobic growth of the WT strain was not affected by the addition of nitrate (Figure 3(a)), while its anaerobic growth was significantly enhanced throughout the growth stage (Figure 3(b)). Furthermore, nitrate reductase activity was undetectable after 3 h of anaerobic culture of the WT strain in AT medium without KNO₃. In contrast, the addition of nitrate significantly increased nitrate reductase activity (Figure 3(c)). These results suggest that nitrate promotes anaerobic growth of K. pneumoniae by inducing nitrate reductase expression, providing a growth advantage in the inflamed gut through nitrate respiration.

Figure 3.

Impact of intestinal nitrate on aerobic and anaerobic growth of K. pneumoniae in vitro. (a) The WT strain was grown aerobically in at medium supplemented with or without 20 mm KNO₃ for 12 h, and the growth was monitored by the change in OD₆₀₀ every 2 h. (b) Anaerobic growth curve of the WT strain in at medium supplemented with or without 20 mm KNO₃. (c) Nitrate reductase activity of the WT strain was measured after 3 h of anaerobic culture. Data are represented as means ± standard deviations (error bars) of three independent experiments. n.D., not detectable.

Nitrate respiration provides growth and colonization advantages for K. pneumoniae in the inflamed gut

The genome of K. pneumoniae encodes two nitrate reductases, NarGHI and NarZYV [28]. To determine whether nitrate respiration confers a fitness advantage upon K. pneumoniae, we constructed the single mutant strains ΔnarG, ΔnarZ, and double mutant ΔnarGΔnarZ. In vitro analysis of these mutants cultured anaerobically in the presence of nitrate showed that ΔnarG and ΔnarGΔnarZ exhibited severe growth defects and lacked nitrate reductase activity. In contrast, no significant difference in anaerobic growth or nitrate reductase activity was observed between ΔnarZ and WT (Figure 4(a-b)). Additionally, the competitive anaerobic growths of ΔnarG and ΔnarGΔnarZ were strongly outcompeted by the WT strain in the presence of nitrate (p < 0.01), whereas the competitive anaerobic growth of ΔnarZ was not significantly affected by the addition of nitrate (Figure 4(c)). The competitive indexes of WT and ΔnarG, ΔnarZ, and ΔnarGΔnarZ in the absence of nitrate were approximately 1.73, 0.79, and 2.43, respectively, suggesting that the mutants ΔnarG and ΔnarGΔnarZ exhibit slight intrinsic growth defects (Figure 4(c)). These data confirm that the nitrate reductase NarG is the major contributor to nitrate reduction in K. pneumoniae.

Figure 4.

Nitrate respiration enhances the fitness of K. pneumoniae during gut inflammation. (a) Strains WT, ΔnarG, ΔnarZ, and ΔnarGΔnarZ were grown anaerobically in AT medium supplemented with 20 mm KNO₃ for 12 h, and the anaerobic growth was monitored by the change in OD₆₀₀. (b) Nitrate reductase activity of the WT and mutant strains (ΔnarG, ΔnarZ, and ΔnarGΔnarZ) after 3 h of anaerobic culture. (c) The competitive index of K. pneumoniae WT (green fluorescence) and the deletion mutants ΔnarG, ΔnarZ, or ΔnarGΔnarZ (red fluorescence) after 24 h anaerobic growth in AT medium with or without 20 mm KNO₃. The index was determined as the CFU ratio of recovered WT to deletion mutants. (d) Fecal bacterial counts at 24 and 48 h post-inoculation of 10⁹ CFU of WT, ΔnarG, or ΔnarGΔnarZ strains into DSS-treated SPF C57BL/6 mice. (e) Bacterial 27 loads in the liver, large intestine, and small intestine tissues 72 h after inoculation. Data are represented as means + standard deviations (Error Bars) of three independent experiments. n.d., not detectable; *,P < 0.05; **,P < 0.01; ****,P < 0.0001; ns, not statistically significant. (unpaired two-tailed Student’s t-test).

Next, we assessed whether nitrate respiration provides a colonization advantage for K. pneumoniae in the inflamed gut using IBD mouse models. Mice were inoculated with 1 × 10⁹ CFU of WT, ΔnarG, or ΔnarGΔnarZ strains. The WT strain was found in significantly higher numbers in the faeces of DSS-treated mice compared to the mutant ΔnarG (24 h: 8-fold, p = 0.043; 48 h: 160-fold, p = 0.0428) and the double mutant ΔnarGΔnarZ (24 h: 4-fold, p = 0.0367; 48 h: 120-fold, p = 0.0068) (Figure 4(d)). In addition, the bacterial loads of WT in the liver, large intestine, and small intestine 72 h after inoculation were approximately 190, 2600, and 190 times higher than those of the mutant ΔnarG with highly significant differences. Similar results were obtained for the double mutant ΔnarGΔnarZ (Figure 4(e)). These results confirm that nitrate respiration, particularly through the narG gene, provides both growth and colonization advantages for K. pneumoniae in the inflamed gut.

Molybdate transport protein ModA increases the intestinal colonization and extraintestinal dissemination of K. pneumoniae in the inflamed gut through regulating nitrate reductase activity

Nitrate reductases are molybdoenzymes containing a molybdenum (Mo) atom complexed with an organic pyranopterin cofactor at their active site [9]. To investigate the role of Mo in nitrate-dependent anaerobic respiration and the pathogenicity of K. pneumoniae, we created a mutant lacking the gene encoding the molybdate-binding protein ModA. Anaerobic growth curves of the ΔmodA mutant were compared to the WT strain in AT medium supplemented with or without 20 mm KNO₃. The ΔmodA did not exhibit significant anaerobic growth and nitrate reductase activity when supplemented with 20 mm KNO₃, while restored in C-modA (Figure 5(a-b)). The study showed that lower nitrate reductase activity could hinder K. pneumoniae‘s anaerobic growth due to a lack of intracellular Mo accumulation. To confirm this, we assessed the anaerobic growth and nitrate reductase activity of WT and ΔmodA strains in AT broth containing 20 mm KNO₃, with or without 100 µM Na₂MoO₄. The results showed that sufficient extracellular Mo restored the intracellular Mo accumulation and facilitated the nitrate-dependent anaerobic growth of ΔmodA (Figure 5(c-d)). The results indicate that the molybdate transport protein ModA associates with the anaerobic growth of K. pneumoniae through influencing molybdate acquisition from the environment.

Figure 5.

ModA regulates the nitrate reductase activity to enhance the fitness of K. pneumoniae during gut inflammation. (a) Strains WT, ΔmodA and C-modA were grown anaerobically in at medium supplemented with or without 20 mm KNO₃ for 12 h, and the anaerobic growth was monitored by the change in OD₆₀₀. (b) Nitrate reductase activity of WT, ΔmodA, and C-modA strains after 3 h of anaerobic culture in at broth containing 20 mm KNO₃. (c) Anaerobic growth of WT and ΔmodA strains in at broth containing 20 mm KNO₃ with or without 100 µm Na₂MoO₄ for 24 h. (d) Nitrate reductase activity of WT and ΔmodA strains after 3 h of anaerobic growth in at broth containing 20 mm KNO₃ with or without 100 µm Na₂MoO₄. (e) Fecal bacterial counts of dss-treated SPF C57BL/6 mice at 24 and 48 h post-inoculation of 10⁹ CFU of WT, ΔmodA or C-modA strains. (f) Bacterial loads in the liver, large intestine, and small intestine tissues of dss-treated mice 72 h after inoculation with WT, ΔmodA or C-modA strains. Data are represented as means ± standard deviations (error bars). n.d., not detectable; *, p < 0.05; ***, p < 0.001; ****, p < 0.0001; ns, not statistically significant (b-d: unpaired two-tailed Student’s t-test; e-f: one way ANOVA).

Next, the growth, colonization and pathogenicity of WT, ΔmodA and C-modA strains were assessed in DSS-treated mice through intragastric inoculation with 1 × 10⁹ CFU of each strain. At 24 and 48 h post-inoculation, the faecal bacterial counts of the WT strain were significantly higher than those of the ΔmodA mutant (24 h: 35-fold, P  = 0.0261; 48 h: 420-fold, P  = 0.0002). No significant differences were observed between WT and C-modA (24 h: 1-fold, p = 0.7058; 48 h: 0.8-fold, P  = 0.8569) (Figure 5(e)). At 72 h, the cells of ΔmodA were nearly undetectable (20 CFU/mL) in the liver and most large intestine and small intestine contents of DSS-treated mice (Figure 5(f)). Taken together, these findings suggest that the molybdate transport protein ModA regulates nitrate reductase activity to increase the intestinal colonization and extraintestinal dissemination of K. pneumoniae in the inflamed gut.

Tungstate treatment abrogates the fitness advantage of K. pneumoniae in the inflamed gut by inhibiting nitrate respiration

Since ModA is required for anaerobic nitrate reduction for in vitro growth, in vivo robust intestinal colonization and extraintestinal dissemination of K. pneumoniae, we speculate that ModA may be a promising drug target. Tungsten competes with Mo for ModA binding and can replace Mo in the MoCo, thus inhibiting MoCo-dependent processes during gut inflammation [21]. To investigate this, we analysed the competitive anaerobic growth and nitrate reductase activity of WT, ΔmodA and C-modA strains in AT broth supplemented with tungstate. The addition of tungstate reduced the competitive index of WT and ΔmodA after 24 h of anaerobic growth, from 4.8 to 0.89 (Figure 6(a)). Furthermore, tungstate significantly decreased the nitrate reductase activities of WT and C-modA strains (Figure 6(b)), demonstrating that tungstate inhibits the growth advantage of K. pneumoniae conferred by nitrate respiration in vitro.

Figure 6.

Effect of tungstate treatment on nitrate respiration and fitness advantage of K. pneumoniae during gut inflammation.(a) The competitive index of K. pneumoniae WT (green fluorescence) and ΔmodA (red fluorescence) strains after 24h of anaerobic growth in at broth containing 20mM KNO₃ with or without 10mM Na₂WO₄. The competitive index was calculated as the ratio of CFU recovered for WT to that of ΔmodA.(b) Nitrate reductase activity of WT, ΔmodA, and C-modA strains after 3h of anaerobic growth in at broth supplemented with 20mM KNO₃, with or without 10mM Na₂WO₄.(c) Groups of SPF C57BL/6 mice were administered 3% DSS (DSS) or 3% DSS plus 0.2% Na₂WO₄ (DSS+WO₄^2-) and then were inoculated intragastrically with 10⁹ CFU of K. pneumoniae WT strain on day 5. Fecal bacterial counts were determined at 24h and 48h post-inoculation.(d) Bacterial loads in the liver, large intestine, and small intestine tissues of dss-treated mice 72h post- inoculation with WT strain, in the presence or absence of Na₂WO₄. Data are represented as means±standard deviations (error bars). n.d., not detectable; *, p<0.05; **, p<0.01; ****, p<0.0001 (a: paired two-tailed Student’s t-test; b-d: unpaired two-tailed Student’s t-test)

The effect of tungstate were further evaluated in the DSS-induced IBD mouse model. Tungstate-treated mice (DSS+WO₄^2-) exhibited significantly reduced WT bacterial counts in faecal contents compared to the control group (DSS) (24 h: 3-fold, p = 0.0418; 48 h: 30-fold, p = 0.0036) (Figure 6(c)). Additionally, tungstate treatment markedly reduced bacterial loads in the liver, large intestine, and small intestine of mice at 72 h post-inoculation (Figure 6(d)). These findings suggest that tungstate administration in the DSS-induced-colitis model abrogates the fitness advantage of K. pneumoniae during gut inflammation by inhibiting nitrate respiration, thereby highlighting ModA as a promising therapeutic target.

Discussion

IBD is a group of chronic, patchy, and remittent gastrointestinal inflammatory diseases with an elusive aetiology caused by a combination of genetic, immunological, microbial, and environmental factors [29,30]. Over the past decades, Adherent-invasive E. coli, Salmonella species, Clostridium difficile and K. pneumoniae have been implicated in the pathogenesis of IBD [31–33]. At the same time, patients with IBD exhibit a significantly higher incidence and severity of infections caused by E. coli and S. typhimurium compared to healthy individuals [6].

K. pneumoniae is a gram-negative bacterium found in diverse ecological niches and the mouth, skin, and intestine of mammals. The prevalence of gastrointestinal carriage of K. pneumoniae and the incidence of pyogenic liver abscess infections have increased in patients with IBD [34,35]. Several studies have revealed that K. pneumoniae infection can induce intestinal inflammation in a gut microbiota-dependent manner and exacerbate colitis in murine models of IBD [36–39]. Our study confirmed that the K. pneumoniae number was raised in the caecal content, liver, large intestine, and small intestine of DSS-induced-colitis mice inoculated with bacteria than in mock-treated mice. The results prove that IBD is a risk factor for K. pneumoniae infection and changes in the intestinal microenvironment of IBD patients promote the growth, intestinal colonization and extraintestinal dissemination of K. pneumoniae.

Nitrate is one of the by-products of gut inflammatory response, and its levels were increased in the caecal mucus layer of DSS-treated mice. The effect of increased nitrate levels in the gut varies among bacterial species depending on their ability to utilize it. Previous studies have shown that nitrate in the inflamed gut boosts the growth of E. coli and S. Typhimurium, but provides no apparent benefit for Vibrio cholerae [6,40,41]. For K. pneumoniae, the addition of nitrate to the medium significantly increased the anaerobic growth and nitrate reductase activity in vitro, suggesting that K. pneumoniae could use nitrate and might confer a growth advantage in the inflamed gut. Unlike E. coli, K. pneumoniae has only NarGHI and NarZYV nitrate reductases, lacking the periplasmic nitrate reductase NapABC. The anaerobic growth curve and the competitive anaerobic growth of the single mutant strains ΔnarG, ΔnarZ, and double mutant ΔnarGΔnarZ with wild-type strain in AT broth in the presence of nitrate found the nitrate reductase NarG is the major contributor to nitrate reduction in K. pneumoniae. Strains with an inactivated narG gene exhibited a complete loss of nitrate reductase activity and had significantly more colonization defects in DSS-treated mice compared to the wild-type strain. In conclusion, nitrate respiration contributes to the intestinal lifestyle, the ability of intestinal colonization and extraintestinal dissemination of K. pneumoniae.

K. pneumoniae is one of the bacteria that is easily resistant to the antibiotics used today [42–44]. How to impair growth and colonization in the intestinal tract is a problem in mitigating severe infections of K. pneumoniae in IBD patients. The reduction of nitrate reductase activity can decrease K. pneumoniae in the gut. Metal-based antibacterial agents are emerging as novel antimicrobial strategies. Nitrate reductases are molybdoenzymes that contain a single Mo atom at their active site [45–47]. For efficient uptake of molybdate from the gastrointestinal microenvironment, bacteria rely on the high-affinity Mo uptake system ModABC. In S. typhi and P. aeruginosa, ModABC has been identified as essential for host interactions [15,48,49]. Our study demonstrates that the modA gene, which encodes the molybdate-binding protein of ModABC, is crucial for K. pneumoniae‘s in vitro growth under anaerobic conditions through nitrate reduction. ModA increases the intestinal colonization and extraintestinal dissemination of K. pneumoniae in the inflamed gut by regulating the nitrate reductase activity.

Tungsten is an element that competes with Mo to bind ModA and could inhibit MoCo-dependent processes in the inflamed gut to improve colitis [21]. Tungstate supplementation has been shown to inhibit the anaerobic growth of E. coli and P. aeruginosa [15,21]. Similarly, the addition of tungstate to nitrate-containing AT medium decreased the growth of K. pneumoniae. Administration of tungstate in the DSS-induced-colitis model abrogated the fitness advantage of K. pneumoniae, suggesting that tungstate may serve as an alternative selection to decrease the K. pneumoniae growth, colonization and infection in the gut. However, sodium tungstate primarily localizes to bone and spleen following oral administration, which induces DNA damage in the bone marrow [50,51]. To overcome these limitations, advanced tungsten-based delivery systems have been developed. Colonic mucus-accumulating tungsten oxide nanoparticles improve therapeutic efficacy against colitis by increasing the adherence of tungsten with Enterobacteriaceae, compared to sodium tungstate [52]. A calcium tungstate microgel encapsulated probiotics effectively treats colitis by releasing tungsten selectively to inhibit Enterobacteriaceae while promoting probiotics colonization [53]. Additionally, recent research reveals that tungstate exerts potentiating effects on gentamicin and ciprofloxacin against E. coli [54]. The effect of tungstate-based materials or tungstate-antibiotics combinations on the treatment of K. pneumoniae infections needs to be further studied.

Collectively, our findings demonstrate that nitrate confers a significant growth advantage to K. pneumoniae in the intestine of DSS-treated mice, and the molybdate transport protein ModA plays a critical role in regulating nitrate reductase activity to increase the growth, intestinal colonization and extraintestinal dissemination of K. pneumoniae. Tungstate emerges as a promising antibacterial agent to tackle K. pneumoniae infections in IBD patients. This study will contribute to elucidating the impact of nitrate on the growth and pathogenicity of K. pneumoniae in the inflamed gut, providing a theoretical basis for the treatment of K. pneumoniae infection in IBD patients in the future.

Funding Statement

This work was supported by Hubei Province’s Outstanding Medical Academic Leader Program; the Innovative Research Program for Graduates of Hubei University of Medicine [YC2023025]; the Cultivating Project for Young Scholar at Hubei University of Medicine [2021QDJZR021] and the Advantages Discipline Group (Public health) Project in Higher Education of Hubei Province [2021–2025].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions

Conception and design of study: Jichen Xie, Hui Wang.

Acquisition of data: Jichen Xie, Hui Wang, Renhui Ma.

Data analysis and/or interpretation: Renhui Ma, Qiuhang Quan, Zhiqiang Zhang.

Drafting of manuscript and/or critical revision: Jichen Xie, Jinming Fan, Moran Li, Bei Li.

Approval of final version of manuscript: Moran Li, Bei Li.

All authors agree to be responsible for all aspects of their work.

Data availability statement

All data in this article will be obtained through http://doi.org/10.57760/sciencedb.12381.