A novel Klebsiella pneumoniae diguanylate cyclase contributes to intestinal cell adhesion, biofilm formation, iron utilization, and in vivo virulence by gastrointestinal infection

ABSTRACT

Klebsiella pneumoniae is responsible for various infections such as bacteremia, urinary tract infections, pneumonia, and liver abscesses. Multidrug-resistant K. pneumoniae infections pose a critical public health threat, often associated with high mortality rates. The emergence of hypervirulent K. pneumoniae has also raised global health concerns due to its invasive disease potential. Clinical studies suggest intestinal colonization by K. pneumoniae as a risk factor for subsequent infections, but the underlying mechanisms remain unclear. Cyclic di-GMP (c-di-GMP), a bacterial signaling molecule synthesized by diguanylate cyclases (DGCs), controls various cellular processes and is absent in higher organisms, making it an attractive target for antimicrobial development. In this study, we identified a novel DGC, designated as DgcG, in K. pneumoniae, which plays a pivotal role in gastrointestinal colonization and pathogenesis. Using genetic deletion and complementation analyses in a bacteremia and liver abscesses-inducing strain Ca0437, we observed that DgcG promoted intestinal adherence, biofilm formation, iron utilization, and in vivo virulence. RT-qPCR showed that DgcG regulated genes involved in type 3 fimbrial expression and iron transport. In a gastrointestinal infection model of female BALB/cByl mice, dgcG deletion significantly reduced host mortality and bacterial load in the liver, spleen, and intestines, underscoring its role in enhancing bacterial survival and dissemination. Additionally, dgcG gene was found highly conserved and prevalent among diverse K. pneumoniae isolates. These findings provide new insights into c-di-GMP-mediated virulence regulation in K. pneumoniae and highlight DgcG as a potential therapeutic target for controlling K. pneumoniae infections, especially amidst the growing global antimicrobial resistance crisis.

Introduction

Klebsiella pneumoniae is a significant human pathogen associated with a range of diseases. The World Health Organization (WHO) has designated K. pneumoniae as a critical priority among healthcare-associated pathogens [1]. It is a leading cause of hospital-acquired infections globally, including urinary tract infections (UTIs), pneumonia, wound infections, and septicemia, particularly in immunocompromised individuals [2–4]. K. pneumoniae is one of the “ESKAPE” pathogens, a group of six major bacterial species driving the global antimicrobial resistance crisis in healthcare settings [5]. According to 2019 global burden studies on bacterial antimicrobial resistance, K. pneumoniae ranks among the leading pathogens responsible for drug-resistant-related fatalities [6]. A distinct infection, known as community-acquired pyogenic liver abscess (PLA), caused by K. pneumoniae has emerged in Taiwan and other regions [7–10]. These severe, invasive infections are primarily attributed to hypervirulent K. pneumoniae (hvKp), which expresses acquired virulence factors. The emergence and global spread of multidrug-resistant (MDR) and hypervirulent clones have made K. pneumoniae a serious clinical and public health concern [11,12].

Cyclic di-GMP (c-di-GMP) is a global second messenger in bacteria, regulating diverse cellular processes [13]. Diguanylate cyclases (DGCs) and phosphodiesterases (PDEs) have opposing roles in c-di-GMP synthesis and degradation. Synthesis is mediated by DGCs with GGDEF domains (characterized by a conserved GG(D/E)EF motif), while degradation occurs via PDEs containing either EAL domains (with a conserved E(A/V)L motif) or HD-GYP domains [13]. The c-di-GMP molecule links environmental or intracellular signals to changes in bacterial motility, development, biofilm formation, exopolysaccharide synthesis, and virulence gene expression [13]. As c-di-GMP signaling is absent in higher eukaryotes, it presents an attractive target for antimicrobial development [14–17]. Small-molecule DGC inhibitors have been screened to create anti-biofilm agents against pathogens such as Pseudomonas aeruginosa [15], Acinetobacter baumannii [15], and Vibrio cholerae [16]. Strategies targeting c-di-GMP signaling or essential DGCs are gaining attention as potential therapies for antibiotic-resistant bacteria [17].

Many bacterial genomes encode numerous putative dgc and pde genes, suggesting complex, species-specific c-di-GMP regulation [18,19]. C-di-GMP interacts with various effectors and receptors, modulating downstream targets through intricate regulatory mechanisms [20]. In K. pneumoniae, c-di-GMP regulates biofilm formation, type 3 fimbriae, capsular polysaccharide production, and oxidative stress responses [21–23]. Despite the diversity of putative DGCs and PDEs in its genome [24], their roles and regulatory networks in pathogenesis remain largely uncharacterized.

The intestine serves as a major reservoir for K. pneumoniae in humans. Clinical studies suggest infections such as liver abscess and bacteremia may be preceded by gastrointestinal (GI) colonization [25–28]. Cohort studies report that intestinal colonization is a risk factor for subsequent infection [25,26]. Capsular polysaccharides and type 1 and 3 fimbriae contribute to attachment and colonization [29]. However, the molecular mechanisms and regulation of K. pneumoniae-GI interactions remain unclear.

To investigate these interactions, our previous work screened a transposon mutant library and identified a mutant with disruption in kpn02265, a gene essential for adherence to Caco-2 intestinal epithelial cells [30]. Gene kpn02265 encodes a putative DGC containing a GGDEF domain. In this study, we demonstrate that kpn02265 encodes a functional DGC, designated DgcG (diguanylate cyclase for gastrointestinal interaction), which plays a crucial role in host interaction. Through genetic deletion and complementation experiments, we show that DgcG contributes to adherence, type 3 fimbrial expression, biofilm formation, and iron utilization. A mouse model of gastrointestinal infection revealed that DgcG enhances bacterial survival and in vivo virulence. Moreover, the dgcG gene is universally conserved among clinical K. pneumoniae isolates.

Materials and methods

Bacterial strains and genetic manipulation

K. pneumoniae Ca0437 strain is a clinical isolate with K2 capsular type originally obtained from the blood of a patient with septicemia [31]. Ca0437 exhibited a great ability of adhesion to and invasion into intestinal epithelial cells [32], and induced liver abscess through gastrointestinal infection in a murine model [30]. Ca0437 wild type and dgcG deletion mutant (ΔdgcG) was gifted from Professor Jin-Town Wang from the National Taiwan University (NTU). Ca0437 isogenic mutant with unmarked deletion of gene was generated using the temperature-sensitive pKO3-Km plasmid in a method as described previously [30]. Genetic complementation of dgcG in ΔdgcG strain was constructed by chromosomal integration of a single copy of dgcG using the pKO3-Km vector according to a previously reported method [30]. Gene deletion and complementation were confirmed by PCR amplification and sequencing. K. pneumoniae isolates for prevalence analysis were collected from the E-Da Hospital. The bacterial strains used in this study are listed in supplementary Table S1. The sequence types (STs) and capsule locus types (KL types) of 70 gut isolates, determined by whole genome sequencing analysis (unpublished data), are provided in Dataset 1 of supplementary files.

Determination of growth kinetics

Bacterial single colonies were inoculated into Luria-Bertani (LB) broth and grown overnight at 37°C with shaking at 200 rpm. Overnight cultures were diluted in fresh LB broth to an initial optical density at 600 nm (absorbance value A₆₀₀) of 0.05. Growth was monitored at 37°C by measuring A₆₀₀ at each time point: 0, 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 24 h.

Cell adherence assays

Four cell lines were tested: Caco-2 (intestinal epithelial), T24 (bladder epithelial), ARPE-19 (retinal epithelial), and RAW 264.7 (murine macrophages). T24, ARPE-19, and RAW 264.7 cells were purchased from the Bioresource Collection and Research Center (BRCR), Taiwan (https://www.bcrc.firdi.org.tw/). Caco-2 cells were gifted from Professor Jin-Town Wang (NTU). Cells were cultured in appropriate media supplemented with 10% fetal bovine serum (FBS). Bacteria at mid-log phase were suspended in serum-free media and added to cells at a multiplicity of infection (MOI) of 50. After centrifugation, cells were incubated, washed with PBS, and treated with Triton X-100 to release adhered bacteria. Quantification was performed by plating and counting colony forming units (CFUs). The adherence rate was calculated as the proportion of the inoculum adhering to cells. Three independent experiments were conducted. Statistical significance was assessed using analysis of variance followed by Bonferroni multiple comparisons test. Details are described in the Supplementary Methods.

Prevalence and sequences analysis of dgcG in K. pneumoniae

Ca0437 dgcG was sequenced based on Sanger sequencing by commercial service (Genomics BioSci & Tech. Co. Ltd., Taiwan). Protein domain was predicted using SMART program (http://smart.embl-heidelberg.de/). Amino acid sequence alignments, similarities, and phylogeny were analyzed using Clustal Omega multiple sequence alignment program (https://www.ebi.ac.uk/jdispatcher/msa/clustalo). The tBLASTn program was used to identify DgcG in K. pneumonia genomes. The publicly available K. pneumonia genomes were obtained from the NCBI database (https://www.ncbi.nlm.nih.gov/datasets/genome/). The list of 636 K. pneumonia genomes for DgcG comparison was in Dataset 1 of supplementary files. K. pneumonia STs and KL types were determined using Kleborate program (https://github.com/klebgenomics/Kleborate).

Expression of dgcG and c-di-GMP quantification

To verify the function of Ca0437 dgcG, the gene was cloned into the pJET1.2 vector and expressed in Escherichia coli BL21 (DE3). Expression was induced with 0.4 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) for four hours at 37°C. Cellular c-di-GMP was extracted and quantified using high-performance liquid chromatography (HPLC), as detailed in the Supplementary Methods.

Biofilm assays and scanning electron microscopy (SEM)

K. pneumoniae biofilm was quantified using the crystal violet (CV)-based biofilm assays [33]. In brief, K. pneumoniae overnight culture was diluted to A₆₀₀ = 0.05 in LB and incubated statically at 37°C for 24 h to allow biofilm formation. Bacterial growth was measured at A₆₀₀ to estimate total cell biomass. The culture medium was carefully removed and biofilms were fixed with methanol and stained with 0.1% of CV for 15 min, followed by wash with sterile distilled water three times to remove excess dye and allowed to dry at room temperature. The CV was dissolved in 200 μL of 95% ethanol and the optical density at 570 nm was measured. The amount of biofilm formed was determined from the A₅₇₀/A₆₀₀ ratio, to compensate for variations due to differences in bacterial growth [34]. Three independent experiments were conducted. Statistical significance was assessed using analysis of variance followed by Bonferroni multiple comparisons test.

To validate potential differences of biofilm, SEM was used to observe the structures and morphology of K. pneumoniae biofilm according to previously described methods [35]. K. pneumoniae overnight culture was diluted to A₆₀₀ = 0.05 in LB and seeded on sterile coverslips in the wells of a 6-well plate, followed by static incubation at 37°C for 24 h. After removal of culture, biofilms were washed with PBS, fixed with 2.5% glutaraldehyde at 4°C for 4 h, and further subjected to a gradient series of ethanol (30, 50, 70, 90, and 100% v/v), each for 10 min. The samples were dried for 2 h and observed using SEM (Hitachi, S-3400N).

Iron utilization and serum resistance assays

For iron utilization analysis [36], bacteria were grown in iron-free chemically defined DIS medium and then tested with or without Fe³⁺ (500 µM FeCl₃) or Fe²⁺ (500 µM FeSO₄). K. pneumoniae was grown overnight on LB agar at 37°C. A single bacterial colony in 5 mL iron-free DIS medium was equilibrated to an optical density at 600 nm (absorbance value A₆₀₀) of 0.1 for iron-starvation at 37°C for 24 h. Iron-starved bacteria were diluted with DIS medium appropriately to match cultures to 1 × 10⁷ CFU/mL. Iron-starved bacteria (4 µL) were added to the wells of 96-well polystyrene microplate containing 100 µL of DIS supplemented with 0 or 500 µM FeCl₃ (for Fe³⁺)/FeSO₄ (for Fe²⁺), in triplicate. The plate was incubated at 37°C, and A₆₀₀ was measured at various time points to assess growth.

Serum resistance was determined by incubating bacteria with human serum for two hours, followed by CFU enumeration. A K. pneumoniae inoculum of 2.5 × 10⁴ CFU in 25 μL were mixed with 75 μL human serum from the healthy volunteers. The mixtures were incubated at 37 °C for 2 h. After serial dilution and plating, the numbers of CFU were determined. The bacterial survival ratio was expressed as recovered CFUs/inoculum CFUs. The survival ratios ≥1 indicated serum resistance [37].

RNA-sequencing (RNA-seq) and real-time reverse-transcription PCR (RT-qPCR)

The transcriptomes of K. pneumoniae were analyzed using RNA-seq conducted by a commercial service (AllBio Science., Inc, Taiwan). Total RNA was extracted from log-phase bacteria, ribosomal RNA was removed, and libraries were prepared for Illumina HiSeq sequencing. Differential expression analysis was performed using DESeq2, and enriched pathways were identified through Gene Ontology (GO) and KEGG analysis. For real-time reverse-transcription PCR (RT-qPCR), RNA was extracted and treated with DNase I. Reverse transcription and amplification were performed using SYBR Green. For each gene, the calculated threshold cycle (Ct) was normalized to the Ct of the 23S ribosomal RNA gene. The relative RNA expression was calculated based on the ΔΔCt value. RT-qPCR primers used in this study are listed in supplementary Table S2.

Mouse experiments

All animal experiments were approved by the Institutional Animal Care and Use Committee of National Taiwan Ocean University (IACUC-109060). The study has adhered to the ARRIVE guidelines. Mouse survival, bacterial load, and in vivo competition assays were performed in a murine gastrointestinal infection model using five-week-old female BALB/cByl mice purchased from the National Laboratory Animal Center, Taiwan.

For survival analysis, mice (8 per group) were intragastrically (IG) administered 6 × 10⁶ CFUs of mid-log-phase K. pneumoniae in 100 μL of saline solution and monitored daily for 30 days. Mouse survival was analyzed by means of Kaplan-Meier analysis using a log-rank test with Bonferroni correction.

Bacterial load experiments involved IG administration of 1 × 10⁷ CFUs in 100 μL of saline solution to mice (10 per group), followed by organ harvesting (liver, spleen, and colorectum) at 72 h post-infection. Recovered CFUs from organ homogenates were standardized to 0.1 g wet organ weight. Statistical significance was assessed using analysis of variance followed by Bonferroni multiple comparisons test.

For in vivo competition assays [35], the test strain (Ca0437 WT, ΔdgcG or ΔdgcG::dgcG) was mixed with the fully virulent isogenic lacZ promoter deletion mutant (Ca0437ΔplacZ) at a 1:1 ratio. A per-mouse dose of 1 × 10⁸ CFU in 100 μL of saline solution was inoculated IG into each BALB/cByl mouse (8 per group). At 24 h post-inoculation, the liver and spleen were removed and homogenized in 1× PBS; bacteria were recovered by plating appropriate dilutions on LB plates containing 1 mmol/ml IPTG and 50 µg/ml X-Gal. The number of LacZ-positive (blue) and LacZ-negative (white) colonies were counted. The competitive index (CI) was defined as (output_{test strain}/output_ΔplacZ)/(input_{test strain}/input_ΔplacZ); the resulting ratio was interpreted as the in vivo colonization ability [38]. Statistical significance was assessed using analysis of variance followed by Bonferroni multiple comparisons test.

Statistical analyses

Data are presented as means with standard errors of the mean (SEM). Statistical significance was assessed using analysis of variance followed by a post hoc multiple comparisons test performed with Prism 5 software (GraphPad). Differences were considered significant at p < 0.05.

Results

Identification of a novel K. pneumoniae DGC promoting adherence to intestinal cells

Previously, we identified dgcG (kpn02265) through transposon-insertion mutagenesis screening of a bacteraemia-inducing clinical strain K. pneumoniae Ca0437 [30]. This strain exhibited a great ability of intestinal cell adhesion [32] and induced liver abscess by gastrointestinal infection in a murine model [30]. To verify the dgcG effects on cell interactions, we further analysed the in-frame unmarked gene deletion mutant (ΔdgcG) and chromosomal gene complementation (ΔdgcG::dgcG) strains for cell adherence (Figure 1a). Since K. pneumoniae can cause infections in various tissues [2–4,7–10], we evaluated its adherence to different cell types, including intestinal epithelial cells, urinary bladder epithelial cells, retinal pigment epithelial cells, and macrophages. Cell adherence assays showed that deletion of dgcG led to a significant reduction in adherence to Caco-2 intestinal epithelial cells (62% reduction compared to the wild type) and T24 urinary bladder epithelial cells (42% reduction compared to the wild type). No significant impact was observed on adherence to APRE-19 cells or RAW 264.7 macrophages. Genetic complementation significantly restored the defect in adherence to Caco-2 cells, resulting in an 88% increase compared to the mutant. It also rescued adherence to T24 bladder epithelial cells, with a 55% increase relative to the mutant. These data confirmed that dgcG was important for K. pneumoniae adherence to intestinal cells. Deletion of dgcG did not cause changes in bacterial growth in the nutrient broth (Figure 1b).

Figure 1.

K. pneumoniae DgcG promoted adherence to human intestinal cells. (a) Cell adherence assays. Adherence of K. pneumoniae Ca0437 wild type (WT), dgcG deletion mutant (ΔdgcG), and complementation strain (ΔdgcG::dgcG) to different types of cells were analysed: caco-2 intestinal epithelial cells, T24 urinary bladder epithelial cells, ARPE-19 retinal pigment epithelial cells and RAW264.7 macrophages. Relative adherence rates were normalized to that of the WT parent strain (defined as 100%). Bars represent the mean and standard error of the mean from ≥ 3 independent experiments. *, p < 0.05; ns, not significant; analysis of variance followed by Bonferroni multiple comparisons test. (b) Growth kinetics. Separate overnight cultures of each K. pneumoniae strain were inoculated into fresh LB broth and grown at 37°C. The growth of bacteria was monitored by measuring absorbance at 600 nm (A₆₀₀).

Sequence-based functional prediction of the K. pneumoniae Ca0437 dgcG gene indicated that it encoded a putative diguanylate cyclase with a GGDEF/GGEEF domain and four transmembrane regions (Figure 2a). Ca0437 dgcG sequences showed 100% identical to KPN2242_02265 (accession number AEJ96373.1) in K. pneumoniae KCTC2242 complete genome from NCBI database. Comparative analysis revealed conservation of Ca0437 DgcG and other well-characterized DGCs from various bacterial species (Figure 2b). Phylogenetic analysis demonstrated that K. pneumoniae DgcG shares the closest relationship with Yersinia pestis HmsT (Figure 2c). Sequence similarities of K. pneumoniae DgcG to these well-characterized DGCs are shown in Table 1. Amino acid similarities ranged from ~21% to 46.84%. We also compared K. pneumoniae DgcG to the designated DGCs in E. coli since systematic nomenclature of DGCs has been established for E. coli [39]. The amino acid similarities between DgcG and14 E. coli DGCs were all below 30% (Table 1), suggesting that DgcG is distinct from these known E. coli DGCs.

Figure 2.

Domain prediction and comparison of K. pneumoniae Ca0437 DgcG to other well-characterized DGCs.(a) Domain prediction of K. pneumoniae DgcG by SMART program (http://smart.Embl-heidelberg.De/). DgcG contained a predicted GGDEF/GGEEF domain and four transmembrane (TM) domains (indicated in blue bars). (b) Amino acid sequence alignments of K. pneumoniae DgcG and other well-characterized DGCs, by Clustal Omega multiple sequence alignment program and MView (https://www.ebi.ac.uk/jdispatcher/msa/clustalo). The red box indicated the GGDEF/GGEEF domain. Accession No.: AAD25088.1 for HmsT_Yersinia pestis, NP_415544.1 for YcdT (DgcT)_Escherichia coli, WP_002914116.1 for YfiN (DgcN)_Klebsiella pneumoniae, AAA87378.1 for PleD_Caulobacter vibrioides, ELX61125.1 for AdrA_Salmonella enterica, AAL71852.1 for WspR_Pseudomonas fluorescens, ACP10914.1 for CdgA_Vibrio cholera. (c) Phylogenetic tree of K. pneumoniae DgcG and other DGCs.

Table 1.

Comparison of K. pneumoniae DgcG to other well-characterized DGCs and known E. coli DGCs.

DGCs	Accession number	Other name	Amino acid number	Amino acid sequence similarities to DgcG (%)a
Experimentally characterized
HmsT [Yersinia pestis]	AAD25088.1		390 aa	46.84
PleD [Caulobacter vibrioides]	AAA87378.1		454 aa	26.73
AdrA [Salmonella enterica]	ELX61125.1	DgcC [E. coli]; YaiC [E. coli]	370 aa	26.28
WspR [Pseudomonas fluorescens]	AAL71852.1		333 aa	30.47
CdgA [Vibrio cholerae]	ACP10914.1		366 aa	23.55
YcdT [E. coli]	NP_415544.1	DgcT [E. coli]	452 aa	25.81
YfiN [Klebsiella pneumoniae]	WP_002914116.1	DgcN [E. coli]; TpbB [Pseudomonas aeruginosa]	407 aa	21.30
Other DGCs in E. coli
DgcE [E. coli]	NP_416571.1	YegE [E. coli]	1105 aa	26.89
DgcF [E. coli]	NP_416039.1	YneF [E. coli]	315 aa	26.55
DgcI [E. coli]	NP_415355.1	YliF [E. coli]	442 aa	23.36
DgcJ [E. coli]	NP_416300.2	YeaJ [E. coli]	496 aa	25.00
DgcM [E. coli]	NP_415857.2	YdaM [E. coli]	410 aa	28.31
DgcO [E. coli]	NP_416007.3	DosC [E. coli]; YddV [E. coli]	460 aa	29.43
DgcP [E. coli]	NP_416308.4	YeaP [E. coli]	341 aa	25.16
DgcQ [E. coli]	NP_416465.2	YedQ [E. coli]	564 aa	24.40
DgcX [E. coli]	CAU96679.1		443 aa	26.25
DgcY [E. coli]	ACB15888.1		349 aa	29.79
DgcZ [E. coli]	NP_416052.1	YdeH [E. coli]	296 aa	26.83
Others
DgcA [Treponema denticola]	Q73RG3		371aa	27.48
DgcB [Bdellovibrio bacteriovorus]	Q6MPU8		320 aa	30.03

a. Compared with K. pneumoniae Ca0437 (348 aa, accession number PQ435165). Sequence alignment and similarity was determined using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/).

To validate the diguanylate cyclase function, K. pneumoniae Ca0437 dgcG was cloned and expressed in E. coli BL21 (DE3). Intracellular c-di-GMP levels were quantified by HPLC following IPTG-induced expression of dgcG, which led to a significant increase in c-di-GMP levels (supplementary Table S3). This confirmed the diguanylate cyclase activity of DgcG to synthesize c-di-GMP molecules.

K. pneumoniae DgcG contributes to biofilm formation and iron utilization

We examined whether dgcG influenced virulence-associated traits of K. pneumoniae, including biofilm formation, capsular polysaccharide (CPS) production, serum resistance, and iron utilization (Figure 3a–d). Interestingly, we observed that dgcG contributed to K. pneumoniae biofilm formation and iron utilization. The crystal violet-based biofilm assays (Figure 3a) showed that deletion of dgcG caused the decrease of biofilm formation, and the defects were rescued by genetic complementation. The differences of biofilm formation were observed using scanning electron microscope (SEM) (Figure 3d). Consistent with the results of the crystal violet assay, the ΔdgcG mutant exhibited a lack of biofilm formation compared to the wild type (WT), and this defect was restored by genetic complementation.

Figure 3.

K. pneumoniae DgcG contributed to biofilm formation and iron utilization. Virulence-associated traits and growth of K. pneumoniae Ca0437 wild-type (WT), dgcG-deletion mutant (ΔdgcG), and complementation strain (ΔdgcG::dgcG) were determined and compared. (a) Quantification of biofilm formation. Biofilm levels were measured by crystal violet-based biofilm assays (see methods) and quantified as the A₅₇₀/A₆₀₀ ratio to normalize for differences in bacterial growth. Each bar indicates the mean ± SEM from three independent experiments. (b) Capsular polysaccharide (CPS) production. The levels of extracted CPS from indicated strains were determined and relative production were normalized to WT (defined as 100%). Each bar indicates the mean±SEM from three independent experiments. (c) Serum resistance. Indicated strains were treated with human serum (75%) at 37°C for 2 hours. The numbers of recovered CFUs were determined, and the survival ratio was expressed as recovered CFUs/inoculum CFUs. (d) Scanning electron microscopy (SEM) images of biofilm formation. Equivalent amount of K. pneumoniae strains were grown on sterile coverslips for 24 h. After removal of culture, biofilms were washed with PBS and fixed with 2.5% glutaraldehyde for SEM observation at 2,000x (upper graph) and 5,000x (lower graph). (e) Iron utilization with Fe³⁺ (500 µM FeCl₃, left graph) or Fe²⁺ (500 µM FeSO₄, right graph). Bacterial strains were starved for iron for 24 hours and then shifted to DIS cultures supplemented with iron (+ Fe³⁺ or Fe²⁺) or without iron. Growth was assessed by monitoring the increase in culture absorbance at 600 nm (A₆₀₀) at the time points indicated. Data in (a), (b) and (c) are presented as the mean±SEM from at least three independent experiments. Throughout figure, *, p < 0.05; **, p < 0.01; ns, not significant; analysis of variance followed by Bonferroni multiple comparisons test.

Iron utilization assays (Figure 3e) showed that the ΔdgcG mutant was deficient in iron utilization during the 0–8 h period, including the use of Fe^{3 +} (left graph, FeCl₃) and Fe^{2 +} (right graph, FeSO₄). While iron-starved WT bacteria failed to grow in iron-depleted medium, their growth was restored in the presence of either Fe³⁺ or Fe²⁺. In contrast, the iron-starved ΔdgcG mutant was unable to resume growth with either form of iron by 8 h. This defect in iron utilization was rescued by genetic complementation. By 24 h, no differences in growth were observed among the WT, ΔdgcG, and ΔdgcG::dgcG strains. Therefore, deletion of dgcG caused delayed kinetics and appeared to reduce the efficiency of iron utilization, particularly during the exponential growth phase.

Production of CPS (Figure 3b) and serum resistance (Figure 3c) were not significantly influenced by deletion of dgcG. In addition, the string tests and low-speed centrifugation assays to analyze K. pneumoniae mucoviscosities both showed no significant differences between WT, ΔdgcG, and ΔdgcG::dgcG strains (data not shown), in agreement with observation that K. pneumoniae DgcG might not regulate CPS biosynthesis.

K. pneumoniae DgcG regulates expression of type 3 fimbriae and iron transport-related genes

To dissect the underlying mechanisms, RNA-sequencing (RNA-seq) was conducted to compare the transcriptomes of WT and ΔdgcG strains, revealing that dgcG deletion affected genes involved in metabolism, transmembrane transport, cell adhesion, and iron acquisition (Figure 4a). Among the 5211 genes analyzed, expression of 47 genes was significantly altered, including 27 downregulated genes (log₂ fold change ranging from −0.49 to −2.87) and 20 upregulated genes (log₂ fold change ranging from 0.48 to 0.75). Twenty-nine of these genes showed expression changes greater than ~ 1.5-fold (log₂ fold change >0.58 or <−0.58), including 18 downregulated and 11 upregulated genes (Table 2). Notably, several type 3 fimbrial genes (mrk) and iron transport-related genes bfd and fecD were downregulated upon deletion of dgcG. The raw RNA-seq gene expression data are provided in Dataset 2 of the supplementary files.

Figure 4.

K. pneumoniae DgcG promoted expression of type 3 fimbrial and iron transport genes. (a) RNA-seq transcriptomic analysis to compare K. pneumoniae Ca0437 wild-type (WT) and ΔdgcG mutant. left graph: gene ontology (GO) categories showing differentially expressed genes between WT and ΔdgcG strains; right graph: venn diagram analysis of differentially expressed genes between WT and ΔdgcG strains. Data from two independent experiments. (b) RT-qPCR to assess potential effects on expression of specific gene (as indicated). K. pneumoniae Ca0437 wild-type (WT), dgcG-deletion mutant (ΔdgcG), and complementation strain (ΔdgcG::dgcG) were determined and compared. Data represent the mean and standard error of the mean from 3 independent experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001; analysis of variance followed by Bonferroni multiple comparisons test.

Table 2.

DgcG-regulated genes identified in RNA-seq transcriptomic analysis.

Gene ID	Predicted function (gene)a	Accession number	Fold change (log2)b
Down
Gene468	Diguanylate cyclase (dgcG)	WP_002887653.1	−2.873678777
Gene4012	Type 3 fimbria adhesin subunit MrkD (mrkD)	WP_004900381.1	−1.700913247
Gene4008	Type 3 fimbria protein MrkH (mrkH)	WP_004152886.1	−1.176812053
Gene4013	Type 3 fimbria usher protein MrkC (mrkC)	WP_000813718.1	−1.131645233
Gene1475	Hypothetical protein	WP_002895851.1	−0.97164089
Gene4014	Molecular chaperone	WP_014599251.1	−0.835499923
Gene4009	LuxR family transcriptional regulator	WP_004149657.1	−0.801099084
Gene734	tRNA-Ala		−0.778722138
Gene4011	Type 3 fimbria minor subunit MrkF (mrkF)	WP_002916122.1	−0.732081349
Gene4493	Bacterioferritin-associated ferredoxin (bfd)	WP_004900993.1	−0.729630352
Gene4015	Type 3 fimbria major subunit MrkA (mrkA)	WP_002916128.1	−0.7220177
Gene2783	Iron ABC transporter permease (fecD)	WP_004148676.1	−0.698754303
Gene397	Hypothetical protein	WP_002887388.1	−0.664076289
Gene334	Hypothetical protein	WP_004192273.1	−0.617954192
Gene3138	tRNA-Ser		−0.60136311
Gene2858	D-amino acid dehydrogenase	WP_004180373.1	−0.601344184
Gene214	Primosomal replication protein N	AEJ96132.1	−0.592762114
Gene2022	Transcriptional regulator	AEJ97850.1	−0.584059681
Up
Gene4265	Dihydroxyacetone kinase subunit DhaK (dhaK)	WP_002917682.1	0.752825101
Gene4319	Alpha-dehydro-beta-deoxy-D-glucarate aldolase	WP_004900848.1	0.751904498
Gene1587	Oxidoreductase	WP_004147771.1	0.684258867
Gene3875	Glucarate dehydratase	WP_002915223.1	0.669223088
Gene208	3-dehydro-L-gulonate-6-phosphate decarboxylase UlaD (ulaU)	WP_002885684.1	0.605177494
Gene378	Hypothetical protein	WP_002887282.1	0.600854841
Gene1588	Hydroxylamine reductase	AEJ97439.1	0.59937008
Gene1486	Putative periplasmic binding protein/LacI transcriptional regulator	AEJ97336.1	0.59760055
Gene4875	Permease	AEK00571.1	0.591763952a
Gene4266	Dihydroxyacetone kinase subunit DhaL(dhaL)b	AEJ99993.1	0.5904176
Gene4334	Protease	AEK00063.1	0.588401486

^aAnalysis using NCBI-BLAST.

^bTwenty-nine genes with expression changes more than ~ 1.5 folds (log2 fold change > 0.58 or < −0.58), including 18 genes downregulated and 11 genes upregulated.

RT-qPCR was further preformed to examine the expression of individual fimbrial and iron-related genes (Figure 4b). We verified reduced expression of type 3 fimbrial genes mrkA, mrkC, mrkH, and iron transport gene fecD in ΔdgcG, with genetic complementation restoring the defects. For comparison, other genes relating to K. pneumoniae colonization and virulence were also determined (Figure 4b). Expression of type 1 fimbrial genes (fimA and fimC), siderophore genes (iucA and entC), and CPS gene galF were not significantly altered. These findings indicate that DgcG promotes the expression of genes relating to type 3 fimbriae production and iron transport.

K. pneumoniae DgcG enhances in vivo virulence by gastrointestinal infection

The impact of DgcG on in vivo virulence of K. pneumoniae was evaluated using a murine gastrointestinal infection model [30]. We observed that bacterial loads in the liver, spleen, and colon were significantly reduced in ΔdgcG-infected mice (Figure 5a). In vivo competition assays showed that ΔdgcG had a reduced competitive index compared to WT in both the liver and spleen, indicating that the attenuated ability of ΔdgcG to survive and spread in the host (Figure 5b). These results suggest that DgcG plays a critical role in promoting gastrointestinal colonization, bacterial survival, and dissemination within the host. By intragastrical inoculation, mice infected with ΔdgcG exhibited significantly attenuated virulence, with a 30-day survival rate of 87.5% (7/8) compared to 12.5% (1/8) in the wild-type (WT) group (p = 0.0062, log-rank test) (Figure 5c). This indicated that DgcG contributed to in vivo virulence of K. pneumoniae and influenced host mortality.

Figure 5.

K. pneumoniae DgcG contributed to in vivo virulence in a gastrointestinal infection model. (a) Bacterial loads in mice. K. pneumoniae WT, ΔdgcG or ΔdgcG::dgcG was administered IG to each of four BALB/cByl mice (ten mice per group). Bacterial levels in liver, spleen, and colon were determined at 72 h post-infection. Bacterial numbers (expressed as log₁₀ CFU) were standardized per 0.1 g of wet organ weight. *, p < 0.05; ns, not significant; analysis of variance followed by Bonferroni multiple comparisons test. Abbreviation: ns, not significant. (b) in vivo competition assays. Each test strain (WT, ΔdgcG or ΔdgcG::dgcG) was compared with the ΔplacZ mutant strain in BALB/cByl mice (eight mice per group) by IG inoculation. Following bacterial recovery on IPTG/X-Gal plates, the ratio of LacZ-positive (blue) to LacZ-negative (white) colonies in the liver or spleen of each mouse was determined. The competitive index (CI) was defined as (outputΔplacZ/output_{test strain})/(inputΔplacZ/input_{test strain}). Each symbol represents CI for each inoculum, with the medians shown by bars. *, p < 0.05; **, p < 0.001; ns, not significant; Wilcoxon signed rank test. (c) Mouse survival analysis. BALB/cByl mice (eight mice per group) were infected intragastrically (IG) with K. pneumoniae Ca0437 wild type (WT) or isogenic ΔdgcG mutant. Survival of mice was monitored for 4 weeks. *, p < 0.05; **, p < 0.01; ns, not significant; log-rank test.

High prevalence and conservation of dgcG in diverse K. pneumoniae isolates

The prevalence of dgcG in K. pneumoniae strains from clinical and gastrointestinal colonization sources was determined (Table 3). The overall prevalence of dgcG was 98.88% (177/179) in 179 tested strains, with 98.17% (107/109) in clinical isolates and 100% (70/70) in gut isolates. In clinical strains, lower prevalence rates were observed in wound (9/10, 90%) and CSF (16/17, 94.12%) isolates. The sequence types (STs) and capsule locus (KL) types of gut isolates were compared (supplementary Table S4). Despite the diversity in STs and KL types, all gut isolates carried dgcG. The amino acid sequence similarity of the putative DgcG ranged from 83.54% to 100%, with the majority of strains (67.1%, 47/70) possessing DgcG with 100% identity. These findings indicate a high prevalence and conservation of dgcG among genetically diverse K. pneumoniae gut isolates.

Table 3.

Detection of dgcG in K. pneumoniae strains from different sources.

Sources	Number of test strains	Positive of dgcG	Prevalence (%)
Gut colonization strains	70	70	100.00
Clinical strains	109	107	98.17
Blood isolates	23	23	100.00
CSF isolates	17	16	94.12
Sputum isolates	29	29	100.00
Urine isolates	21	21	100.00
Wound isolates	10	9	90.00
Bile isolates	9	9	100.00
Total	179	177	98.88

Notes: PCR-based detection. Gut colonization strains were stool isolates from the subjects of health examination. Clinical strains were collected from different specimens of the patients as indicated.

The putative DgcG in 636 K. pneumoniae genomes from publicly available NCBI databases [40–42] was further analyzed (Table 4). The overall prevalence of the dgcG gene was 100%. The amino acid sequence similarity of the putative DgcG ranged from 95.98% to 100%, with 80.5% of strains (514/636) carrying DgcG with 100% identity. While a few strains encoded DgcG proteins of varying lengths, most strains (98.6%, 627/636) carried a 348-amino-acid DgcG. The presence or conservation of DgcG was not associated with hvKp, MDR-Kp, or specific ST or KL types. Notably, even within the same ST-KL group (e.g. ST23-KL1), variations in DgcG sequence similarity or protein length were observed. Conversely, DgcG with 100% identity was found across distinct ST-KL types. These results indicate again that dgcG is highly prevalent and conserved across genetically diverse K. pneumoniae clones. Table S5 showed putative DgcGs in the well-known reference strains. The putative DgcG was highly conserved among 9 K. pneumoniae reference genomes. Seven strains, including hypervirulent reference strains NTUH-K2044, CG43, SGH10 and multidrug-resistant strains HS11286 and NJST258_1, carried DGCs with 100% amino-acid sequence similarities to Ca0437 DgcG. The other two strains, MGH 78,578 and Kp342, harboured a putative DgcG with the sequence similarities of 99.63% and 95.98%, respectively. These findings suggest that dgcG is ubiquitous and conserved across K. pneumoniae strains. This gene likely encodes an essential and core DGC involved in the regulation of important physiological functions.

Table 4.

Comparison of dgcG in K. pneumoniae genomes from the public available database.

STa	KLb	Nc	dgcG-positive N (prevalence rate)	DgcG lengthd (N)	Amino acid sequence identitye (N)
HvKpfrelated
ST23	KL1	53	53 (100%)	348 a.a. (51)	100% (51)
				269 a.a. (2)	100% (2)
ST86	KL2	24	24 (100%)	348 a.a. (24)	100% (22), 99.71% (2)
ST65	KL2	6	6 (100%)	348 a.a. (6)	100% (6)
ST29	KL54	6	6 (100%)	348 a.a. (4)	99.71% (3), 100% (1)
				347 a.a. (2)	99.71% (2)
ST412	KL57	4	4 (100%)	348 a.a. (4)	100% (4)
ST218	KL57	3	3 (100%)	348 a.a. (3)	100% (2), 99.71% (1)
MDR-Kpgrelated
ST11	KL24	9	9 (100%)	348 a.a. (9)	100% (9)
	KL47	9	9 (100%)	348 a.a. (9)	100% (9)
	KL64	5	5 (100%)	348 a.a. (5)	100% (5)
	KL105	3	3 (100%)	348 a.a. (3)	100% (3)
	KL125	2	2 (100%)	348 a.a. (2)	100% (2)
	KL103	1	1 (100%)	348 a.a. (1)	100% (1)
ST15	KL24	9	9 (100%)	348 a.a. (9)	100% (9)
	KL112	8	8 (100%)	348 a.a. (7) 180 a.a. (1)	100% (7) 99.44% (1)
	KL19	6	6 (100%)	348 a.a. (6)	100% (6)
	KL39	2	2 (100%)	348 a.a. (2)	100% (2)
	KL28	1	1 (100%)	348 a.a. (1)	100% (1)
ST258	KL107	14	14 (100%)	348 a.a. (13) 269 a.a. (1)	100% (13) 99.71% (1)
	KL106	10	10 (100%)	348 a.a. (10)	100% (10)
	KL74	2	2 (100%)	348 a.a. (2)	100% (2)
	KL63	1	1 (100%)	348 a.a. (1)	100% (1)
ST512	KL107	13	13 (100%)	348 a.a. (13)	100% (13)
ST307	KL102	12	12 (100%)	348 a.a. (12)	100% (12)
ST147	KL64	11	11 (100%)	348 a.a. (11)	99.43% (10), 100% (1)
	KL10	2	2 (100%)	348 a.a. (2)	99.43% (2)
	KL20	1	1 (100%)	348 a.a. (1)	99.43% (1)
	KL111	1	1 (100%)	348 a.a. (1)	99.43% (1)
	KL125	1	1 (100%)	348 a.a. (1)	99.42% (1)
ST101	KL17	11	11 (100%)	348 a.a. (11)	100% (11)
	KL106	2	2 (100%)	348 a.a. (2)	100% (2)
	KL62	1	1 (100%)	348 a.a. (1)	100% (1)
	KL64	1	1 (100%)	348 a.a. (1)	99.43% (1)
ST14	KL2	10	10 (100%)	348 a.a. (10)	100% (8), 99.71 (2)
	KL16	3	3 (100%)	348 a.a. (3)	100% (3)
	KL64	1	1 (100%)	348 a.a. (1)	99.71% (1)
ST48	KL62	6	6 (100%)	348 a.a. (6)	100% (6)
ST111	KL63	4	4 (100%)	348 a.a. (4)	99.71% (4)
ST29	KL30	2	2 (100%)	348 a.a. (2)	100% (2)
Other dominant type
ST20	KL28	7	7 (100%)	348 a.a. (7)	100% (7)
ST45	KL24	6	6 (100%)	348 a.a. (6)	100% (6)
ST16	KL51	6	6 (100%)	348 a.a. (6)	100% (6)
ST231	KL51	6	6 (100%)	348 a.a. (6)	100% (6)
ST340	KL15	6	6 (100%)	348 a.a. (6)	100% (6)
ST17	KL25	5	5 (100%)	348 a.a. (5)	100% (5)
ST268	KL20	5	5 (100%)	348 a.a. (5)	100% (5)
ST336	KL25	5	5 (100%)	348 a.a. (5)	100% (5)
ST25	KL2	4	4 (100%)	348 a.a. (4)	99.71% (4)
ST35	KL22	4	4 (100%)	348 a.a. (4)	100% (4)
ST299	KL7	4	4 (100%)	348 a.a. (4)	99.71% (4)
ST437	KL36	4	4 (100%)	348 a.a. (4)	100% (4)
Other strains with different length of DgcG
ST1754	KL62	1	1 (100%)	234 a.a. (1)	99.57% (1)
ST321	KL3	1	1 (100%)	269 a.a. (1)	99.63% (1)
ST660	KL16	1	1 (100%)	288 a.a. (1)	99.65% (1)
ST1741	KL104	1	1 (100%)	391 a.a. (1)	98.56% (1)
The others		310	310 (100%)	348 a.a. (310)	100% (233), 99.71% (65), 99.43% (9), 99.72% (1), 97.70% (1), 95.98% (1)
Total		636	636 (100%)	348 a.a. (626)	95.98% − 100%
				269 a.a. (4)	99.63% − 100%
				347 a.a. (2)	99.71%
				180 a.a. (1)	99.44%
				234 a.a. (1)	99.57%
				288 a.a. (1)	99.65%
				391 a.a. (1)	98.56%

^aST: sequence type.

^bKL: K locus type.

^cN: number of strains.

^da.a.: amino acid number.

^eSequence identity determined using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/), compared with Ca0437 DgcG (348 a.a.).

^fHvKp: hypervirulent K. pneumoniae.

^gMDR-Kp: multidrug-resistant K. pneumoniae.

Discussion

C-di-GMP, though absent in higher eukaryotes, is ubiquitous in bacteria where it regulates various cellular processes. DGCs have been identified as promising drug targets for developing therapeutics that interfere with the second messenger-signaling networks in pathogenic bacteria [14–17]. Most bacteria possess multiple DGCs and PDEs to respond to diverse environmental signals and regulate a variety of cellular functions [18,19]. The multiplicity of these enzymes, along with their structural diversity, raises questions about the specificity and complexity of c-di-GMP regulatory networks. In this study, we identified an essential DGC, DgcG, playing a critical role in K. pneumoniae pathogenesis. This K. pneumoniae DGC regulates cell interactions, biofilm formation, and iron utilization. Based on the study of a bacteremia and liver abscess-inducing strain Ca0437, we demonstrated the multiple functions of DgcG in controlling virulence-related traits in K. pneumoniae. Given the global threat posed by MDR and hypervirulent K. pneumoniae, DgcG could be a promising target for developing new therapeutic agents. For example, small-molecule libraries can be screened for drug discovery to identify DGC inhibitors as the novel antimicrobials.

Our findings expand the understanding of c-di-GMP signaling in K. pneumoniae-host interactions. Adherence to host cells and subsequent tissue colonization is often a crucial first step in establishing infection. Deletion of dgcG led to significant reductions in K. pneumoniae adherence to both intestinal (Caco-2) and bladder (T24) epithelial cells, suggesting that DgcG contributes to epithelial colonization in both gastrointestinal and urinary tracts. These results are consistent with previous studies highlighting the role of c-di-GMP signaling in modulating fimbrial expression and host cell attachment in various Gram-negative pathogens [21,43–45]. We found that DgcG did not significantly affect adherence to retinal pigment epithelial cells and macrophages. Our previous study also demonstrated no impact on adherence to A549 lung epithelial cells and HepG2 hepatocytes [30]. Therefore, DgcG seems to promote K. pneumoniae interactions with specific cell types, highlighting the importance of this DGC in selective tissue colonization. Notably, RNA-seq and RT-qPCR data showed that DgcG regulated type 3 fimbriae but not type 1 fimbriae, indicating a pathway-specific regulation mechanism. In K. pneumoniae, type 1 fimbriae (encoded by the fim operon) was reported essential for UTI initiation [46], while type 3 fimbriae (encoded by the mrk operon) promoted biofilm formation and binding to collagen and abiotic surfaces [47,48]. In agreement with our findings, previous research has shown that MrkH, a PilZ-domain transcriptional activator, regulated mrk expression in response to c-di-GMP [21]. A potential regulatory pathway could be that DgcG influences cell adherence through the c-di-GMP-dependent protein MrkH, which further regulates mrk operon and therefore type 3 fimbrial expression.

Biofilm formation, another critical virulence factor regulated by c-di-GMP, was impaired in the absence of DgcG. This phenotype was validated by both crystal violet staining and SEM imaging. Several studies have linked intracellular c-di-GMP levels to enhanced biofilm formation in K. pneumoniae and other enteric pathogens [21,23,49–52]. In K. pneumoniae, type 3 fimbriae has been considered as the primary factor to influence biofilm formation under c-di-GMP control [21,53]. On the other hand, we did not observe changes in capsule production or mucoviscosity upon dgcG deletion, in agreement with studies showing that capsule biosynthesis in K. pneumoniae may be more closely linked to RmpA, RcsAB, or other capsule regulators [54]. Consistent with our findings, a study of a K. pneumoniae UTI strain (AJ218) reported a distinct K. pneumoniae diguanylate cyclase, YfiN (DgcN), upregulating biofilm formation and type 3 fimbriae expression [21]. Sequence alignment between DgcG and YfiN with a similarity of only 21.30%. This indicates that K. pneumoniae employs multiple diverse DGCs to regulate similar functions, such as biofilm formation and expression of fimbriae, potentially allowing the bacterium to finely tune these critical processes in response to varying environmental signals.

DgcG played a role in iron acquisition mainly in the growth phase, as ΔdgcG mutants were deficient in utilizing Fe^{3 +} or Fe^{2 +} between 0 and 8 hours but not by 24 hours. Lack of dgcG caused the lower efficiency and the delayed kinetics to resume growth in the presence of iron by the iron-starved strain. Some possible reasons could contribute to the difference between the 8-hour and 24-hour time points. For example, DgcG may regulate the timely activation of iron acquisition systems or be involved in growth phase-dependent regulation [55]. There could also be compensatory mechanisms such as activation of alternative iron uptake pathways [56]. Our data indicate that while DgcG might not be essential for iron uptake, it enhances the rate and efficiency of iron utilization during periods of rapid growth. RNA-seq results during the log phase revealed downregulation of iron transport genes such as fecD and bfd, supporting the idea that c-di-GMP signaling modulates the expression of genes involved in iron metabolism. Iron is essential for K. pneumoniae survival and pathogenicity, and iron uptake systems are tightly regulated in response to host-imposed nutritional immunity. To our knowledge, this is the first report of a DGC regulating iron metabolism in K. pneumoniae. Studies in Pseudomonas aeruginosa demonstrated that c-di-GMP modulates siderophore production and iron acquisition in response to environmental signals [57]. Our data suggest a similar role for c-di-GMP in fine-tuning iron metabolism in K. pneumoniae, although further studies are needed to define the mechanistic basis of this regulation.

The in vivo mouse model of gastrointestinal infection further validated the role of DgcG in virulence of K. pneumoniae. To our knowledge, this is the first study to demonstrate the contributions of a DGC and c-di-GMP signaling to K. pneumoniae GI infection. The ΔdgcG mutant exhibited significantly reduced gastrointestinal colonization and attenuated virulence to affect host mortality, underscoring the importance of intestinal interactions in K. pneumoniae pathogenesis. Bacterial burden and in vivo competition analyses indicated that DgcG was important for bacterial survival and spread within the host from the gastrointestinal tract to cause systemic infection. Our data suggest that DgcG functions as an upstream regulator of both type 3 fimbriae and iron acquisition, two key traits implicated in gastrointestinal colonization and systemic dissemination [12,58,59]. By study of a bacteremia and liver abscess-inducing strain (Ca0437), we revealed c-di-GMP involvement in GI interaction and systemic infection, contributing to the knowledge about the role of c-di-GMP signaling in K. pneumoniae pathogenesis. The c-di-GMP signaling has been reported to regulate Salmonella gastrointestinal colonization and in vivo virulence [60]. In K. pneumoniae, previous studies about c-di-GMP in pathogenesis mainly focused on respiratory infections [61,62]. One study demonstrated that increased c-di-GMP levels by expression of a heterologous DGC in a cystitis isolate (TOP52) attenuated bacterial virulence in a murine pneumonia model, likely through enhanced type 1 fimbrial expression [61]. Another study showed that local intranasal (i.n.) or systemic subcutaneous (s.c.) administration of synthetic c-di-GMP in mice induced protective immunity against lung infections by K. pneumoniae [62]. The roles of K. pneumoniae DgcG in distinct types of infections warrant further exploration.

The high prevalence and conservation of dgcG among gut-derived and clinical K. pneumoniae isolates, including hypervirulent and multidrug-resistant lineages, suggest that DgcG encodes a core signaling enzyme that is under strong evolutionary conservation. The presence of highly conserved DgcG in diverse ST and KL types, and its absence of linkage to specific lineages, highlight its potential as a broad-spectrum target for therapeutic development.

In conclusion, this study characterized a novel K. pneumoniae DGC and demonstrated its contributions to gastrointestinal interactions and pathogenesis. DgcG functions as a diguanylate cyclase that promotes to intestinal adherence, biofilm formation, and iron metabolism. Given its influence on bacterial virulence and conservation among K. pneumoniae isolates, DgcG represents a potential therapeutic target, especially in the context of the growing antimicrobial resistance crisis. Our findings provide new insights into the c-di-GMP-mediated regulation of K. pneumoniae pathogenicity, particularly in its interactions with the host GI tract. Future research should investigate the potential stimuli and downstream effectors of DgcG involved in c-di-GMP signaling networks that regulate bacterial virulence.

Funding Statement

This work was supported by the National Science and Technology Council (NSTC) in Taiwan; E-Da Hospital under grant [EDAHT112016, EDAHT113011].

Disclosure statement

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

Data availability statement

The data that support the findings of this study are available in this article, supplementary files, and Figshare (https://figshare.com) with DOI: https://doi.org/10.6084/m9.figshare.28517942. The sequencing data of K. pneumoniae Ca0437 dgcG is available through NCBI GenBank with the accession number PQ435165.

Supplemental data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/21505594.2025.2544882