Shedding light on Klebsiella pneumoniae virulence: Engineering of broad host range bioluminescence reporter vectors for transcriptional analysis in drug resistant pathogens

Abstract

In this work, we report the construction of four bacterial luciferase-based promoter probe vectors with an expanded set of selectable markers, designed to facilitate their use in antibiotic-resistant bacteria. These vectors contain the low-copy-number, broad-host-range pBBR origin of replication and an origin of transfer, allowing efficient conjugative transformation into various bacterial genera. The broad host range origin also enables their use in bacterial strains that harbor other plasmids, as the pBBR origin is compatible with a wide variety of other plasmid replication systems. The utility of these vectors was demonstrated by quantifying capsule gene expression in both classical and hypervirulent Klebsiella pneumoniae strains lacking tolC, which encodes the outer membrane pore protein for tripartite transport systems. Our results revealed that the tolC mutation reduced capsule gene expression, highlighting a critical role for tolC in K. pneumoniae pathobiology and the utility of bioluminescence for studying gene expression in real time. These new vectors provide a flexible platform for circumventing antibiotic resistance phenotypes and studying gene expression across diverse bacterial species, including strains containing additional plasmids.

1. Introduction

Transcriptional reporters have laid the foundation for analyzing regulatory circuits involved in modulating bacterial adaptation to environmental and genetic perturbations. As such, they have become indispensable tools for examining conditional gene expression in bacteria. Reporter systems, based on enzymatic and fluorescent proteins such as β-galactosidase (LacZ) and green fluorescent protein (GFP), have been widely used as surrogate reporters to quantify gene expression (Kunkle et al., 2017). However, the use of these reporters has limitations. Enzyme-based systems, like LacZ, often require genetic modifications for use in organisms and requires time-consuming processing steps that involve destructive sample processing, and require the addition of reactant chemicals, making their use laborious and hindering their use in vivo (Miller, 1972). Conversely, GFP is a non-destructive reporter that does not require extensive processing for quantification, but the requirement for light excitation and interference from background autofluorescence can limit GFP applicability in some environments (Silva-Rocha and de Lorenzo, 2012). In contrast, bioluminescent reporters based on the bacterial lux operon circumvent these issues.

The lux operon, initially isolated from Vibrio fischeri (Engebrecht et al., 1983) and later from Photorhabdus luminescens (Schmidt et al., 1989, Szittner and Meighen, 1990), encodes both the luciferase enzyme and the proteins necessary for generating the luciferase substrate (i.e., luciferin). For this reason, the lux operon has been developed for use as a promoter-probe reporter for quantifying gene expression based on bioluminescence production (Engebrecht, Nealson and Silverman, 1983, Schmidt, Kopecky and Nealson, 1989, Szittner and Meighen, 1990). Utilizing bioluminescence as a surrogate for gene expression eliminates the need for sample processing and the addition of reactants, making quantification less laborious and more efficient. The lux operon, approximately 5.8 kb in length, consists of five genes (luxCDABE): luxA and luxB encode the luciferase enzyme, while luxC, luxD, and luxE form a fatty acid reductase complex responsible for producing luciferin, the substrate required for luciferase-dependent bioluminescence (Brodl et al., 2018).

Bacterial bioluminescence reporters offer advantages over enzymatic and GFP-based fluorescent transcriptional reporters, including increased sensitivity, a broad dynamic range for signal detection, and minimal background bioluminescence (Hutchens and Luker, 2007). Unlike enzymatic reporters, lux reporters do not require additional substrates or cell processing steps. Further, luminescence-based imaging systems enable non-invasive, real-time kinetic and quantitative assessment of bioluminescence production in animals and tissues.

Antimicrobial resistance is a rapidly expanding global health issue, particularly in Klebsiella pneumoniae, where multiple drug resistance contributes to raising the morbidity and mortality rates of those infected (de Man et al., 2018, Li et al., 2018, Marsh et al., 2019). The rise in antibiotic resistance also constrains basic bacterial research by diminishing the pool of effective selectable genetic markers that are indispensable for genetic engineering research. Addressing this expanding issue requires the development of genetic tools that encode alternative selective markers to facilitate their use in antibiotic-resistant organisms.

In this study, we report the construction and validation of new lux-based promoter reporter vectors that incorporate alternative selectable markers. The vectors are based on pBBRlux (Hammer and Bassler, 2007), a promoter probe vector extensively used in Vibrio cholerae research. They feature a low-copy-number and broad-host-range origin of replication derived from the Bordetella pertussis plasmid pBBR (Antoine and Locht, 1992, Jahn et al., 2016). Additionally, the vectors contain Mob for efficient transformation into recipient cells by conjugation and the thermostable P. luminescens lux operon (Schmidt, Kopecky and Nealson, 1989, Szittner and Meighen, 1990) located downstream from a multiple cloning site to facilitate ease during promoter cloning. The development of these reporters expands the genetic toolbox for analyzing gene expression in a wide variety of drug-resistant organisms.

2. Materials and Methods

Bacterial strains and growth conditions.

The bacterial strains used in this study are listed in Table 1. Hypervirulent K. pneumoniae strain KPPR1S (Palacios et al., 2017), a streptomycin and rifampicin-resistant variant of ATCC 43816 (Budnick et al., 2021), served as the wild-type (WT) reference. Classical pathotype K. pneumoniae MKP103 and its isogenic Tn∷tolC mutant (Ramage et al., 2017) were kindly provided by Dr. Colin Manoil (University of Washington). Escherichia coli strains EC100λpir and SM10λpir were used for cloning and conjugation of plasmids into K. pneumoniae, respectively. All strains were routinely propagated in lysogeny broth (LB) or on LB agar. When necessary, antibiotics were included in the media at the following concentrations: 20 μg/ml for chloramphenicol (Cm) and 5 μg/ml tetracycline (Tet), 200 μg/ml for hygromycin (Hyg), 50 μg/ml for kanamycin (Kan), 10 μg/ml neomycin sulfate (Neo), 100 μg/ml for streptomycin (Sm), and 100 μg/ml for carbenicillin (Cb).

Table 1.

Strains, plasmids and primers.

Bacteria, plasmid or primer	Genotype, characteristics, or sequence	Source
Klebsiella pneumoniae
KPPR1S	Rifampicin and streptomycin resistant variant of hypervirulent strain ATCC 43816. RifR SmR	(Broberg et al., 2014)
KPPR1S Δbla	Lab stock DC29. Δbla mutant of hypervirulent strain KPPR1S. AmpS RifR SmR.	Bina paper in review
KPPR1S ΔtolC	ΔtolC derivative of hypervirulent strain ATCC 43816 that is resistant to streptomycin and rifampicin. RifR SmR.	(Bina, Weng, Budnick, Van Allen and Bina, 2023)
MKP103	ST238 Classical strain KPNIH1ΔKPC-3.	(Ramage, Erolin, Held, Gasper, Weiss, Brittnacher, Gallagher and Manoil, 2017)
MKP103 tolC::Tn	KPNIH1ΔKPC-3 library strain tnkp1_lr150124p02q109 containing a Tn insertion into tolC.	(Ramage, Erolin, Held, Gasper, Weiss, Brittnacher, Gallagher and Manoil, 2017)
Escherichia coli
EC100 λ pir	F− mcrA Δ(mrr-hsdRMS-mcrBC) &80dlacZΔM15 AlacX74 recAl endAl araD139 Δ(ara, leu)7697 galU galKλ- rpsL (StrR) nupG pir+(DHFR)	Epicenter
SM10 λ pir	thi thr leu tonA lacYsupE recA::RP4–2-Tc::Mu Km λpir. KmR	Lab stock
Additional Strains
Acinetobacter baumannii	Multiple drug resistant strain 17978, Provided by Dr. Vaughn Cooper, University of Pittsburgh	(Adams et al., 2008)
Proteus mirabilus	Strain K2644, provided by Dr. Robert Shanks, University of Pittsburgh	(Brothers et al., 2019)
Pseudomonas aeruginosa	Strain PAO1, laboratory collection	Lab stock
Serratia marcescens	Wild-type strain PIC3611, provided by Dr. Robert Shanks, University of Pittsburgh	(Stella et al., 2015)
Stenotrophomonas maltophilia	Strain K1852, provided by Dr. Robert Shanks, University of Pittsburgh	(Brothers et al., 2018)
Vibrio cholerae	Seventh pandemic O1 El Tor strain C6706, laboratory collection	Lab stock
Plasmids
pBBR1-MCS2	Expression vector used as the source for the neomycin resistance marker, NeoR	(Kovach et al., 1994)
pBBR1-MCS3	Expression vector used as the source for the tetracycline resistance marker, TetR	(Kovach, Phillips, Elzer, Roop and Peterson, 1994)
pMQ300	Expression vector used as the source for the hygromycin resistance marker, HygR	(Kalivoda, Horzempa, Stella, Sadaf, Kowalski, Nau and Shanks, 2011)
pMON39	ESBL encoding plasmid used as the source of the SHV-2 gene.	(Huletsky et al., 1990)
pBBRlux	pBBR1-based lux reporter plasmid. CmR	(Hammer and Bassler, 2007)
pDC111	pBBRlux derivative containing PstI site flanking the Cm resistance gene.	This work
pDC112	pDC111-Hyg, Hygromycin resistant derivative of pDC111, HygR	This work
pDC156	pDC111-Tet; Tetracycline resistant derivative of pDC111, TetR	This work
pDC158	pDC111-Neo; Neomycin resistant derivative of pDC111, NeoR	This work
pDC175	pDC111-SHV-2; ESBL (SHV-2) resistant derivative of pDC111	This work
pDC202	pDC112 containing a manC-lux transcriptional fusion, HygR	This work
pDC307	pDC112 containing a carA-lux transcriptional fusion, HygR	This work
pBAD33	Arabinose regulated expression vector, CmR	(Guzman et al., 1995)
pXB506	pBAD33 expressing K. pneumoniae tolC gene from the araC promoter. CmR	(Bina, Weng, Budnick, Van Allen and Bina, 2023)
Primers (5’ - 3’)
Lux_QC_F	GAAGTGATCTGCAGTCACAGGTATTTATTCG
Lux_QC_R	CCTGTGACTGCAGATCACTTCGCAGA
Hyg_PacI_F	GCTTAATTAAAGCCGATCTCGGCTTGAACG
Hyg_PstI_R	GGCTGCAGTGATAAATGCTTCAATAATATTGAAAAAGGAAGAGTAT
shv2_fwd	CTTCGAATAAATACCTGTGACTGCAGATTACGCCAAGCTTGGTAC
shv2_rev	TTCGATAAGCAGCATCGCCTTTAATTACTTATAAAAATGGGAATTTAATTATGGTG
tet_fwd	CTTCGAATAAATACCTGTGACTGCATAAGAAACCATTATTATCATGACATTAAC
tet_rev	TTCGATAAGCAGCATCGCCTTTAATTGTTTCCTGTGTGAAATTGTTATC
kan_fwd	CTTCGAATAAATACCTGTGACTGCACTAGACTGGGCGGTTTTATG
kan_rev	TTCGATAAGCAGCATCGCCTTTAATCATGCATAAAAACTGTTGTAATTC
manC-F	CGGCCGCTCTAGAACTAGTGCGCGATTATTTTGGTGCGCACACC
manC-R	TTTTGCGGCCGCAACTAGAGATCGGCCAGAGACGACTGCCGG
carA-F	CGGCCGCTCTAGAACTAGTGCGTATGACATTTGCTAACGGCGCG
carA-R	TTTTGCGGCCGCAACTAGAGCACCCTCCAGAGAATATTCACTCAC

Plasmids were transformed into E. coli by chemical transformation (Swords, 2003). E. coli SM10λpir was used to mobilize plasmids into the target cells by conjugation. Briefly, E. coli SM10λpir bearing the desired plasmid was cultured overnight on an LB agar plate containing the appropriate antibiotic to select for each plasmid at 37°C. Simultaneously, recipient strains were cultured overnight on LB agar plates at 30°C. The next day, colonies of each strain were collected with a disposable 10 μL plastic loop and mixed in a ~1 cm square on the surface of a fresh LB agar plate lacking antibiotics. The plate was then incubated for ~1 hour at 37°C before samples of the conjugation mixture were collected and streaked for single colonies on fresh LB agar plates containing Sm to counter-select E. coli, and the appropriate antibiotic to select for the presence of the conjugated reporter plasmid. The plates were then incubated at 37°C overnight before individual transconjugants were isolated.

Plasmid construction.

pBBRlux encodes a unique PacI site at the 3’ end of the Cm^R gene (Fig. 1). To facilitate the replacement of the Cm^R gene with alternate selective markers, we engineered a unique PstI site flanking the 5’ end of the Cm^R gene by site directed mutagenesis. Briefly, mutagenic primers Lux_QC_F and Lux_QC_R (Table 1) were used in a PCR reaction with pBBRlux as a template. The resulting PCR amplicon was digested with DpnI restriction endonuclease to remove template DNA before it was transformed into E. coli EC100λpir and transformants were selected on LB-Cm plates. Several Cm^R colonies were retained, and the resulting plasmids were screened for incorporation of the PstI restriction site by restriction digest with PstI and PacI endonucleases. One PstI positive clone, named pDC111, was retained and used as the entry vector for the generation of variants containing alternative selective markers.

Fig. 1.

Map of pDC111-Cm and derivatives.

Schematic of the five new promoter reporter vectors containing alternative selective markers as indicated. (A) Map of five reporter plasmids showing relevant genes and regions. (B) Schematic of the multiple cloning site (MCS) of each respective plasmid showing the unique restriction sites.

Derivatives of pDC111 expressing alternative selective markers were constructed as follows. The hygromycin-resistant variant was generated by amplifying and isolating the hygromycin resistance gene from pMQ300 (Kalivoda et al., 2011) using the Hyg_PacI_F/Hyg_PstI_R primers (Table 1). The resulting PCR amplicon was digested with PacI and PstI restriction endonucleases, gel purified, and ligated into similarly digested pDC111 to generate pDC112. The tetracycline, kanamycin, and SHV-2-marked variants were constructed by recombination-based cloning using the NEBuilder HiFi DNA Assembly kit (NEB, Beverley, MA, USA) according to the manufacturer’s instructions. Briefly, the marker-specific PCR primers (Table 1) were used to amplify the indicated resistance marker from pBBR1-MCS2 (Km^R), pBBR1-MCS3 (Tet^R), and pMON39 (SHV-2), respectively. Each of the resulting PCR amplicons was then cloned into PacI/PstI digested pDC111 using the NEBuilder Assembly kit before being transformed into chemically competent E. coli EC100λpir. Transformants were selected on LB agar plates containing the respective antibiotics to select for each individual selective marker. Individual colonies from each recombination reaction were retained and the DNA sequence of the plasmids was determined by whole plasmid DNA sequencing at Plasmidsaurus.com (Portland, OR).

Construction of the manC promoter reporter plasmid was performed as follows. The manC F/R primer pair (Table 1) was used in a PCR amplification with KPPR1S genomic DNA as the template to amplify the manC promoter. The resulting amplicon was gel purified and then cloned into BamHI-restricted pDC112 using the NEBuilder HiFi DNA Assembly kit according to the manufacturer’s instructions. The resulting plasmid, pDC202 (P_manC-lux), was sequenced before use. Construction of the E. coli carA promoter reporter plasmid (pDC307) was done similarly using carA F/R primer pair in a PCR amplification reaction with E. coli EC100 chromosomal DNA as template (Table 1). The E. coli carA promoter is constitutively expressed and controls the expression of the carA gene, which encodes the small subunit of carbamoyl phosphate synthetase (Shimada et al., 2014).

Gene expression analysis.

Bacterial strains bearing the indicated reporter plasmids were cultured in 150 μL LB medium in 96-well black plates with clear bottoms. Luminescence production as relative light units (RLU) was assayed at the indicated times using a Biotek Synergy 4 plate reader or a Berthold Sirius tube luminometer and normalized according to the optical density at 600 nm. Test samples were assayed in triplicate and the results averaged. For the kinetic experiments, overnight LB cultures of the test strains were used to inoculate fresh LB medium (1:1000) LB supplemented with the appropriate antibiotics and 0.2% L-arabinose when needed to induce the expression from the araC promoter in pBAD33 and pXB506 (pBAD33-tolC). The microtiter plates were then incubated at 37°C in Biotek Synergy 4 plate reader with shaking and luminescence production and the optical density at 600 nm was recorded every thirty minutes.

DNA sequencing and plasmid availability.

Whole Plasmid Sequencing was performed by Plasmidsaurus.com using Oxford Nanopore Technology with custom analysis and annotation. The resulting DNA sequences were annotated by hand. The annotated DNA sequence of each plasmid in GenBank format is provided in the supplementary data.

Statistical analysis.

Statistical analysis was performed using GraphPad Instate version 3.06 and GraphPad Prism version 10.1.1.

3. Results and Discussion

The global dissemination of antibiotic resistance traits among bacterial pathogens is complicating more than just therapeutic interventions. The increase in antibiotic resistance impedes basic microbiology research by rendering many selectable genetic markers ineffective. Circumventing this issue requires the development of new genetic tools containing additional selectable markers. In this work, we created new lux-based promoter reporters that express alternative markers to facilitate their use in antibiotic-resistant bacteria. We used pBBRlux (Hammer and Bassler, 2007), a bacterial lux-based promoter probe vector, as a template to create a new entry vector, named pDC111, to simplify the introduction of new selectable genetic markers (Fig. 1). This was achieved by engineering a unique PstI restriction site at the 5’ end of the Cm^R gene, which along with the existing PacI restriction site on the 3’ side of the Cm^R gene, facilitates easy removal and replacement of the Cm^R cassette. pDC111 was then used to generate four variants encoding resistance to kanamycin, tetracycline, extended-spectrum β-lactams (SHV-2), and hygromycin.

Generating a diverse set of resistance cassettes extends applicability of these vectors in bacteria with varying and antibiotic resistance profiles. For example, use of the hygromycin resistance gene is advantageous as hygromycin is not used clinically and does not confer cross-resistance to other antibiotics. Because hygromycin resistance is uncommon in bacteria, it becomes a valuable selectable marker for circumventing pre-existing antibiotic resistance. pDC158 utilizes the APH(3’)-II allele, which confers resistance to both kanamycin and neomycin. While many clinical isolates are resistant to kanamycin, these isolates often remain sensitive to neomycin, allowing pDC158 to be used in kanamycin-resistant strains. Additionally, the SHV-2 allele encodes an extended-spectrum β-lactamase (SHV-2), which serves as a selectable marker in strains inherently resistant to ampicillin but susceptible to cephalosporins, such as K. pneumoniae KPPR1S with its chromosomal β-lactamase providing resistance to ampicillin. Moreover, SHV-2 and the other markers provide alternatives for use in pathogens where the specific selectable antibiotic markers are restricted. For example, SHV-2 was engineered for use in type I Francisella tularensis strains (Bina et al., 2010), where many antibiotic resistance markers are restricted due to potential interference with the treatment of this highly virulent pathogens.

The pBBR1 origin of replication offers several advantages for use in reporter plasmids (Antoine and Locht, 1992). Notably, it has an exceptionally broad host range, allowing its use in numerous bacterial genera. This broad host range makes the pBBR1 replicon particularly useful for studying gene expression across diverse bacterial genera. Plasmid incompatibility can be a significant roadblock in genetic studies. However, pBBR1, is also compatible with other commonly used plasmid replicons such as IncP, IncW, and ColE1 (Kovach et al., 1995). This compatibility allows pBBR1-based reporters to be used in bacteria that already harbor other plasmids or carry complementation plasmids. For example, in studies investigating transcription factor binding specificity, a separate plasmid expressing a transcription factor (e.g., pBAD vectors) can be used alongside pBBRlux promoter reporter plasmids in a heterologous host like E. coli. This setup enables researchers to determine whether the recombinantly expressed transcription factor functions directly at the promoter to activate gene expression or mediates gene expression indirectly by quantifying lux production (Weng et al., 2021, Xu et al., 2010). Further, the Mob region in pBBRlux facilitates easy and efficient transformation of target cells by conjugation (Smillie et al., 2010). This is a significant benefit for bacteria that are refractory to transformation using other approaches (e.g., electroporation or chemical transformation), thus expanding the use of the reporters to other bacterial hosts.

The utility of the pBBR origin of replication and the mob region was demonstrated by cloning the constitutive E. coli carA promoter into pDC112, generating pDC307. Both pDC112 and pDC307 (carA-lux) were then conjugated into multiple bacterial species, including Acinetobacter baumannii 17978, Escherichia coli EC100, K. pneumoniae KPPR1, Proteus mirabilis K2644, Pseudomonas aeruginosa PAO1, Serratia marcescens PIC3611, Stenotrophomonas maltophilia K1852, and Vibrio cholerae C6706 (Table 1), with selection for hygromycin resistance. The resulting transconjugants were cultured to mid-log phase in LB medium, and bioluminescence was quantified using a Berthold Sirius single-tube luminometer. Strains carrying pDC307 exhibited significantly higher bioluminescence compared to those with the empty vector (Fig. 2). These results demonstrate the functionality of the mob region for conjugation across multiple genera and confirm that pDC112 is a useful tool for assessing promoter activity in diverse bacterial species where antibiotic resistance is a critical concern.

Fig. 2.

Validation of pDC112 use across multiple different bacterial genera.

The indicated strains harboring either pDC112 or pDC307 (carA-lux) were cultured to middle logarithmic phase in LB medium at 37°C when bioluminescence was quantified using a Berthold Sirius tube luminometer as described in the Materials and Methods. Data represent the means ± SD from three experimental replicates. Statistical significance was determined using a t-test, comparing bioluminescence relative to the empty vector control (*, P < 0.05).

Bioluminescence offers advantages over enzymatic and fluorescent protein reporters for studying gene expression (Bazhenov et al., 2023). This includes a high signal-to-noise ratio and the ability to monitor gene expression in real-time with minimal processing. This enables real-time monitoring using a plate reader, luminometer, or imaging system, facilitating the assessment of gene induction kinetics under varying times and conditions, including different growth media, phases, and environments including in animals (Cronin et al., 2008). The non-destructive monitoring of bioluminescence allows for easy detection, and minimizes processing steps, leading to simpler and more efficient experimental approaches.

Despite the advantages of using bioluminescence as a transcriptional reporter, some limitations exist (Bazhenov, Novoyatlova, Scheglova, Prazdnova, Mazanko, Kessenikh, Kononchuk, Gnuchikh, Liu, Al Ebrahim, Zavilgelsky, Chistyakov and Manukhov, 2023). The luciferase reaction requires oxygen, theoretically limiting its use in anaerobic environments. However, successful application of lux reporters in the low-oxygen environment in both murine and human colons has been reported, suggesting that oxygen availability in vivo may not be a significant concern for the use of the lux operon as a reporter (Cronin, Sleator, Hill, Fitzgerald and van Sinderen, 2008, Denoel et al., 1997). Another limitation is the detection threshold in animal imaging, which can be affected by bacterial cell density and tissue depth in animals (Close et al., 2011). This can be partially offset by using highly sensitive luminescence detection equipment like an IVIS Spectrum in vivo imaging system (Caliper Life Sciences, USA).

To validate the use of the lux reporter plasmids, we assessed the contribution of tolC to capsule production in hypervirulent K. pneumoniae (hvKp) and classical K. pneumoniae (cKp) strains. Capsule production is a key virulence factor for K. pneumoniae (Paczosa and Mecsas, 2016), with hypervirulent strains producing polysaccharides that contribute to the formation of “hypercapsules” that directly contribute to the hypermucoviscosity phenotype that is a hallmark of hvKp pathotypes (Paczosa and Mecsas, 2016, Rendueles, 2020). Previous studies demonstrated that the capsule is essential for the K. pneumoniae virulence (Paczosa and Mecsas, 2016) and that tolC mutation in hvKp strain KPPR1S attenuated both capsule production and virulence in the Galleria mellonella infection model, both phenotypes were restored when tolC was ectopically expressed from the arabinose promoter in pBAD33 (Bina et al., 2023). The reduction in capsule biosynthesis and attenuated virulence in the tolC mutant correlated with decreased expression of capsule biosynthesis genes manC and wzc, though the mechanism responsible for this downregulation remains undefined (Bina, Weng, Budnick, Van Allen and Bina, 2023). The role of tolC in capsule production in classical K. pneumoniae strains has not been investigated. To address this we created a transcriptional fusion of the manC promoter to the lux operon in pDC112 to generate pDC202 (manC-lux). The resulting reporter was introduced into the hypervirulent pathotype strain KPPR1S, its isogenic ΔtolC mutant, as well as multidrug-resistant classical pathotype strain MKP103 and its isogenic Tn∷tolC mutant. MKP103 is a member of the pandemic ST258 clade that is responsible for ongoing global nosocomial outbreaks (Ramage, Erolin, Held, Gasper, Weiss, Brittnacher, Gallagher and Manoil, 2017).

The resultant strains were cultured in LB medium to mid-log phase (OD₆₀₀ ~1.0) and gene expression was assessed. High expression levels of manC were observed in KPPR1S, a result consistent with high-level capsule gene expression observed under these conditions (Fig. 3A). In contrast, manC expression was reduced in the ΔtolC mutant, confirming a previous study showing that tolC was required for high-level manC expression and capsule production in hvKp strain KPPR1S (Bina, Weng, Budnick, Van Allen and Bina, 2023). In cKp strain MKP103, expression of manC phenocopied the results observed in hvKp strain KPPR1S with the MKP103 tolC∷Tn mutant showing attenuated manC expression relative to WT (Fig. 3A). These results collectively suggest that the link between tolC and capsule gene expression is likely conserved across the two pathotypes.

Fig. 3.

Effects of tolC mutation on K. pneumoniae capsule gene expression.

(A) Expression of manC in wild-type and isogenic tolC mutants of the hypervirulent K. pneumoniae strain KPPR1S and the classical K. pneumoniae strain MKP103. Strains harboring either pDC202 (manC-lux) or the empty vector (pDC112) were cultured in LB medium at 37°C to mid-log phase when bioluminescence production was measured as described in the Materials and Methods. Data represent the means ± SD from three independent experiments, each performed in triplicate. Statistical significance was determined using one-way ANOVA with Tukey-Kramer multiple comparisons test (***, P ≤ 0.001). (B) Ectopic expression of tolC restores manC-lux expression in the KPPR1S ΔtolC mutant. Wild-type KPPR1 and its isogenic tolC mutant containing the indicated plasmids were cultured in LB medium with chloramphenicol and hygromycin in the presence of 0.2% arabinose to induce expression from the araC promoter in either the pBAD33 empty vector or pXB506 (pBAD33-tolC). Bioluminescence production was measured every 30 minutes over a 15-hour period on a Biotek Synergy 4 plate reader. Data represent the means ± SD of three replicates. Statistical significance was determined using two-way ANOVA with Tukey’s post-hoc test; * p < 0.05 relative to WT-pBAD33-pDC202.

One of the advantages of using bioluminescence as a transcriptional reporter is that its non-destructive nature facilitates real-time kinetic analysis of gene expression. To demonstrate this, as well as the utility of the broad-host-range origin of replication in cells harboring additional plasmids, we conducted complementation studies using K. pneumoniae strain KPPR1S ΔtolC and monitored manC-lux expression over 15 hours (Fig. 3B). Wild-type (WT) and ΔtolC mutant strains were independently transformed with either pBAD33 or pBAD33-tolC, in combination with pDC112 (empty vector control) or pDC202 (manC-lux). Cultures were grown in LB medium supplemented with 0.2% arabinose to induce expression from the araC promoter in pBAD33, and bioluminescence was measured every 30 minutes over the course of 15 hours, as described in the Materials and Methods.

The results presented in Fig. 3B showed that in the ΔtolC mutant harboring pBAD33 and pDC202, manC-lux expression began to diverge from WT at around 3 hours and remained significantly reduced throughout the experiment (closed orange triangles vs. closed black circles). In contrast, complementation of the ΔtolC mutant with pBAD33-tolC restored manC-lux expression to levels exceeding those observed in WT (closed red triangles vs. closed black circles). Strains transformed with pDC112 (empty vector) produced only background levels of bioluminescence, confirming that the observed increase in bioluminescence in strains bearing pDC202 (manC-lux) was due to activation of the manC promoter. Collectively, the results are consistent with the individual time point assessments and confirm that tolC is necessary for WT expression of manC. The observation that ectopic tolC expression significantly increased manC expression relative to WT suggests that TolC may be limiting for maximal expression under the test conditions.

The development of these new lux-based transcriptional reporter vectors expands the genetic toolbox for analyzing gene expression in a wide range of antibiotic-resistant bacteria. Lux reporters enable real-time, non-invasive monitoring of bacterial gene expression that is effective both in vitro and in vivo. Their successful application in hypervirulent and classical strains of K. pneumoniae showcased their use by demonstrating a conserved function for tolC in capsule production in both K. pneumoniae pathotypes.

Highlights.

• Broad-host-range lux reporters enable gene expression analysis in diverse bacteria.

• Vectors with alternative resistance markers work in antibiotic-resistant strains.

• Lux reporters allow real-time, non-invasive gene expression monitoring in bacteria.

• Validated in drug-resistant Klebsiella, revealing TolC’s role in capsule production.