Introduction

Klebsiella pneumoniae is a ubiquitous bacterium that can be found in a wide range of ecological niches including water, soil, vegetation, and insects [1,2,3]. It also can be recovered from the stools, nasopharyngeal swabs, or urine samples collected from companion and farm animals, such as dogs, cats, horses, cows, chickens, and pigs. To these animals, it is an essential cause of pneumonia, metritis, septicemia, and mastitis [4,5,6]. The infection leads to losses in milk production, food contamination, and high mortalities [7,8,9]. Meanwhile, the K. pneumoniae in animal hosts and along the food chain can be transmitted to humans and cause infection [10,11,12]. Treatment with antibiotics is an important method against bacterial pathogens. However, more than one-third of K. pneumoniae isolates in Europe are resistant to at least one class of the antimicrobial agents, and multidrug-resistant (MDR) strains have been recovered in clinical isolates not only from humans but also from companion animals and livestock [13,14,15].

Given the increasing multidrug resistance, new strategies are needed to control and prevent infection by MDR K. pneumoniae in animals and humans. Vaccination is regarded as another effective method to prevent infectious diseases. To date, there are no vaccines available against K. pneumoniae infection and the development of vaccines against K. pneumoniae has become a priority in many countries. The schemes for vaccine development against K. pneumoniae include whole cell vaccines, ribosomal vaccines, outer membrane vesicles, and virulence factors including capsular polysaccharides (CPS), lipopolysaccharide, and protein-based vaccines [16,17,18,19,20,21]. Live attenuated vaccines are a mode of whole cell vaccines that are more immunogenic and can elicit long-lasting immune responses than the subunit vaccines, and they have a relatively low production cost. Although the problem of biological safety limits the use of live vaccines, they are suitable to immunizing large communities or herds. The main vaccines used for animals or humans against the infection of Brucella melitensis, Yersinia pestis, and Mycobacterium tuberculosis are still live attenuated vaccines [22,23,24,25,26]. For K. pneumoniae, a tonB deletion mutant strain can induce a protective immune response against challenge with a wild-type strain [27].

KbvR is a regulator that has been found to reduce biofilm formation and the synthesis of CPS in the K. pneumoniae NTUH-K2044 strain. The virulence of the kbvR mutant was significantly decreased in the murine model by intraperitoneal injection [28]. RNA-seq results show that the expression levels of type 1 fimbriae genes such as fimA and fimH were increased in K. pneumoniae kbvR mutant strain [28]. Fimbriae are filamentous structures present on the cell surface of bacteria. The components of fimbriae have been shown to be promising vaccine candidates [29, 30]. The selected T- and B-cell specific epitopes of type 1 fimbriae FimA, FimF, FimG, and FimH proteins have been estimated to be used as potential vaccine candidates against K. pneumoniae infection, and six epitopes confirmed by molecular dynamics can be applied for vaccine development [31, 32]. Thus, the ΔkbvR strain might be a potentially viable option for the preparation of vaccines in preventing K. pneumoniae infections in humans and animals. In this study, the influence of KbvR on the synthesis of type I fimbriae was investigated. Then, experiments on the safety, antibody test, and protective efficacy of the ΔkbvR strain in mice were performed to evaluate the potential of the kbvR mutant strain as a live attenuated vaccine.

Materials and methods

Bacterial strains and growth conditions

The capsular serotype K1 hypervirulent K. pneumoniae strain NTUH-K2044 (WT), the NTUH-K2044 kbvR gene deletion strain (ΔkbvR), and the capsular serotype K3 standard K. pneumoniae strain ATCC 13,883 were preserved in our laboratory. All strains from frozen stocks were inoculated into Luria–Bertani (LB) broth and grown overnight with shaking at 37 °C. Then, the cultures were diluted 1:100 into fresh LB media the following day and cultured to the desired phase of growth.

RNA isolation and RT-PCR

Total RNA were extracted from mid-log phase K. pneumoniae WT and ΔkbvR strains using an RNeasyMidi kit (Qiagen). Purified RNA was DNase-treated with RNase-free DNase I (Qiagen) to eliminate the residual DNA. The quality and quantity of RNA were determined via 1% agarose gel electrophoresis and with a Nano Drop 2000 UV–Vis Spectrophotometer (Thermo Scientific). The first-strand cDNA synthesis was conducted using the first-strand cDNA synthesis kit (Promega) and quantified. The relative expression levels of fimA and fimH genes in WT and ΔkbvR strains were analyzed by agarose gel electrophoresis after amplifying cDNA by PCR. 16 S rRNA was used as the normalized gene. The relative mRNA levels were quantified by densitometry using ImageJ software and then calculated by dividing the densitometry of fimA/fimH gene by the corresponding densitometry of 16 S rRNA.

Yeast agglutination assay

The presence and activity of type I fimbriae at the bacterial cell surface were assessed using the baker’s yeast suspended in phosphate-buffered saline (PBS) as previously described [33]. The mid-log phase of statically cultured WT and ΔkbvR strains were washed twice and resuspended in PBS with 108 colony-forming units (CFU)/mL. Equal cell number of 50 µL yeast cell suspension and 50 µL different bacterial suspensions were mixed on a glass slide. After shaking at room temperature for 3 min, the aggregation was assessed by visual inspection.

Transmission electron microscopy

The WT or ΔkbvR strain was statically cultured in LB broth for 10 h and suspended in 1 mL of PBS for negative staining to observe the fimbriae directly [34]. A glow-discharged formvar-coated copper grid was floated on a drop of each bacterial suspension liquid for 3 min. Then, the excess liquid was drained off with filter paper, and the copper grid was stained with uranyl acetate for another 3 min followed by air-drying. The specimens were viewed under a transmission electron microscope (HT7800, Japan) operated at an accelerating voltage of 120 kV.

Mouse subcutaneous infections

Five- to six-week-old female BALB/c mice were obtained from HUNAN SJA laboratory animal Co., Ltd. to determine the virulence of ΔkbvR in mice by subcutaneous route. The NTUH-K2044 WT and ΔkbvR strains were cultured to OD600 = 1.2 in LB broth and serially 10-fold diluted with sterile PBS. Appropriate dilutions were plated onto LB agar plates to calculate the numbers of CFU. A total of 10 mice for each group gathered from two experiments were injected subcutaneously with 105, 106 CFU WT or 105, 106, 107 CFU ΔkbvR. The mice were monitored daily for 14 days, and the survival rates were calculated to analyze the virulence of ΔkbvR by subcutaneous route.

Mouse immunizations and challenge experiments

Two different groups of BALB/c mice were respectively immunized with a low dose (105 CFU) or high dose (107 CFU) ΔkbvR strain in 100 µL PBS two times for 4 weeks at 2-week intervals by subcutaneous route. The mice immunized with PBS were used as a normal control. All mice were monitored for skin reaction and weighed daily. At 28 days after the first immunization, the serum samples were collected from immunized (n = 10 for 105 CFU dose from two experiments and n = 5 for 107 CFU dose) and normal control (n = 10 from two experiments) mice to evaluate antibody production by enzyme-linked immunosorbent assay (ELISA). The liver, spleen, and lungs of the mice were processed, paraffin-embedded, microtome sectioned, and stained with hematoxylin and eosin to determine histopathological changes using an Olympus DP74 camera.

Two weeks after the second immunization, the mice (105 CFU ΔkbvR-immunized group and PBS control group) were infected with 104 or 105 CFU log-phase K. pneumoniae WT strain via intraperitoneal route. At the 12 h post-infection, five immunized or control mice challenged by 105 CFU WT strain were sacrificed. The blood was collected and plated onto LB plate to calculate the number of bacteria. Then, 3 mL LB medium was injected into the abdominal cavity of the mice. A total of 10 µL peritoneal lavage fluid was collected and serially diluted to calculate the bacterial number, while another 10 µL peritoneal lavage fluid was coated onto the slides and stained via the gram stain method to be observed by optical microscopy. Thereafter, the liver, spleen, and lungs were removed aseptically, weighed, and homogenized in PBS using a tissue homogenizer. The samples were serially diluted and spread on LB agar plate containing 100 µg/mL ampicillin to determine the bacterial counts. The remaining mice challenged by 104 or 105 CFU WT (n = 7 for each PBS control group and n = 8 for each immunized group) were observed every day for the next 14 days to evaluate the survival rate.

Enzyme-linked immunosorbent assay

Relative antibody levels in mice were determined by ELISA [35]. First, 96-well flat-bottom plates were coated with 100 µL lysate of whole NTUH-K2044 WT cells. This WT strain was grown in LB broth to OD600 = 1.2, collected by centrifugation, resuspended in PBS, and lysed by sonication. After overnight incubation at 4 °C, plates were washed three times with PBS including 0.05% Tween-20 (PBST), blocked with 5% nonfat dry milk for 1 h, and washed three times with PBST. The serum samples of immunized or normal control mice were diluted 1:400. The resulting titer was corrected at an optical density of 450 nm to about 0.1 of control serum, and 100 µL was applied per well followed by 1.5 h incubation at 37 °C. After washing with PBST, 100 µL of 1:3000 horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G (IgG) antibodies was added and incubated at 37 °C for 1 h. The color was developed for 10 min in the dark by addition of 100 µL 3, 30, 5, 50-tetramethylbenzidine, and the reaction was stopped by adding 100 µL 2 M H2SO4. The optical density of the reaction was recorded at 450 nm using an ELISA plate reader and corrected by subtracting the optical density of a blank control.

Opsonophagocytic assay

For an opsonophagocytic assay, plasmid pLac-EGFP with a gene encoding green fluorescence protein was electroporated into K. pneumoniae NTUH-K2044 and ATCC 13,883 strains. The bacteria were cultured in LB at 37 °C to OD600 = 1.2, washed with PBS, and adjusted to approximately 107 CFU/mL in RPMI 1640 medium. RAW264.7 macrophage cells were seeded in 24-well glass bottom culture plates and cultured in RPMI 1640 medium containing 2 mM glutamine and 10% (v/v) fetal calf serum [36]. For the opsonophagocytic assay, macrophages were infected with bacteria (NTUH-K2044 or ATCC 13,883) at a multiplicity of infection (MOI) of 50 bacteria/macrophage and 1 µL pooled immunized or normal mice sera (n = 5 for each group) in a 1100 µL reaction at 37 °C. After 120 min of contact, cells were washed twice with PBS and incubated for an additional 90 min with RPMI 1640 medium in the presence of 500 µg/mL gentamicin to eliminate the remaining extracellular bacteria. Then, the cells were washed, fixed, and stained with Alexa Fluor 594 dye (red) for actin cytoskeleton and DAPI (blue) for host cell nuclei. Finally, the cells were observed by confocal microscopy. For the opsonophagocytic activity against NTUH-K2044, the sum of the intracellular bacteria in the macrophages in different fields were calculated and quantified within 300 cells. For the opsonophagocytic activity against ATCC 13,883, the washed cells were lysed with 0.5% Triton X-100 and serial dilutions were plated on LB agar to quantify intracellular bacteria. The mean opsonophagocytic ratio of sera to ATCC 13,883 was determined using the formula = 100 ⋅ CFU of intracellular phagocytized bacteria/CFU of input bacteria. Experiments were conducted on three independent occasions [36].

Complement-mediated antibacterial test

The complement-mediated antibacterial test was performed as previously described [36, 37]. The pooled immunized or normal mice sera (n = 5) were heat-inactivated in a 56 ℃ water bath for 1 h. A total of 106 bacterial cells (K. pneumoniae strain NTUH-K2044 or ATCC 13,883) were co-incubated with 25% (v/v) human serum (providing the complement) donated by healthy volunteers with or without 5 µL heat-inactivated mice sera in a 50 µL reaction at 37 ℃ for 2 h. The colony count was calculated by the serial dilution method. The mean survival ratio was determined using the formula = 100 ⋅ CFU of experimental wells with mice sera/CFU of control wells without mice sera.

Mouse passive immunization and challenge assay

For passive immunization experiments, the sera from 5 mice immunized with two doses of 105 CFU K. pneumoniae ΔkbvR strain (immunized sera) or from 5 mice injected with PBS (normal sera) were collected and pooled respectively. Then, pooled sera from immunized or normal mice were passively transferred to groups of 5 naïve female BALB/c mice in the dose of 1 µL respectively along with 100 µL 104 CFU WT by intraperitoneal route. Mice were observed daily for 14 days, and survival was monitored [21].

Statistical analyses

The data were presented either as mean ± standard error (SD). A Mantel–Cox (log-rank) test was used to compare the survival curves, and the Student t-test was used for the statistical comparisons between the two groups.

Results

Type 1 fimbriae biosynthesis is increased in kbvR mutant strain

In our previous study, the RNA-seq results show that the transcription levels of gene cluster associated with type I fimbriae synthesis were increased in the ΔkbvR strain (NCBI GenBank: No. SAMN17192750-SAMN17192755) [28]. The RT-PCR confirmed that the transcription levels of fimA and fimH, the genes that encode the major subunit protein and the adhesion of type I fimbriae, respectively, were increased in the kbvR mutant strain. The relative mRNA levels of fimA and fimH in ΔkbvR and WT were 0.233 versus 0.072 and 0.127 versus 0.074, respectively (Figure 1A). The yeast agglutination assay and transmission electron microscopy were used to directly observe the activity and production of type I fimbriae. As shown in Figure 1B, the ΔkbvR strain gave obvious clumps in a clear solution, whereas little agglutination was observed for WT at the same number of cells, indicating more type I fimbriae on the surface of ΔkbvR than on that of WT, which mediated stronger yeast agglutination. The transmission electron microscopy images also show the existence of a large number of fimbriae on the surface of ΔkbvR compared with that of WT (Figure 1C). These results suggest that KbvR negatively regulated the synthesis and activity of type I fimbriae.

Figure 1
figure 1

KbvR negatively regulates the expression of type I fimbriae. A mRNA levels of fimA and fimH genes were detected by RT-PCR. The overnight cultures of WT and ΔkbvR strains were diluted 1:100 into fresh LB media and cultured to mid-log phase at 37 °C. Total RNA were extracted and RT-PCR was conducted with 16 S rRNA as the normalized gene. The relative mRNA levels of fimA and fimH genes were analyzed by agarose gel electrophoresis, quantified by densitometry using ImageJ software, and calculated by dividing the densitometry of fimA/fimH gene by the corresponding densitometry of 16 S rRNA. B Yeast agglutination assay. The mid-log phase WT and ΔkbvR cultures (108 CFU/mL, 50 µL) were mixed with equal cell numbers of 50 µL yeast cell suspension and shaken at room temperature for 3 min. Large clumps (with arrows indicated) were observed in the ΔkbvR strain compared with those in WT. C Transmission electron microscopy images of K. pneumoniae NTUH-K2044 and its ΔkbvR mutant, and arrows indicate the fimbriae.

Full size image

ΔkbvR strain injected subcutaneously is avirulent in the mouse model

The 50% lethal dose (LD50) was equal to 5 × 102 CFU for WT and more than 5 × 105 CFU for ΔkbvR by intraperitoneal infection [28]. To assess the virulence of the ΔkbvR strain infected by the subcutaneous route, the possible immunization route of live vaccine, the survival rate of mouse model infected subcutaneously by 105, 106 CFU WT and 105, 106, 107 CFU kbvR mutant strains were calculated for 14 days. The ΔkbvR strain was also highly attenuated via the subcutaneous route. All mice infected by 106 CFU WT died in 5 days, and the survival rate was decreased to 40% in the mice infected by 105 CFU WT. On the contrary, no deaths occurred in the ΔkbvR-infected groups which were even challenged by the 107 CFU dose (Figure 2).

Figure 2
figure 2

Survival curves for the K. pneumoniae WT and ΔkbvR strains by subcutaneous infection. A total of 5 groups of 10 five- to six-week-old female BALB/c mice were injected subcutaneously with 105, 106 CFU WT or 105, 106, 107 CFU ΔkbvR, and the survival rate was measured for 14 days (P < 0.0001).

Full size image

Subcutaneous immunization with kbvR strain stimulates humoral immune response

Given that ΔkbvR could be a live attenuated vaccine candidate, the concentration of IgG in the mice immunized with two doses of 105 or 107 CFU ΔkbvR at 2-week intervals were quantified by ELISA assay. All immunized mice mounted IgG antibody response to whole K. pneumoniae lysates at day 28 post-immunization, and the concentration was significantly higher than that of the PBS control group (P < 0.0001). The results indicate a positive effect of the ΔkbvR strain on antibody induction. The 107 CFU-immunized group had even higher antibody concentration than the 105 CFU-immunized group (P < 0.05), suggesting that the increase in the immunized ΔkbvR dose might improve vaccine efficacy (Figure 3A).

Figure 3
figure 3

Immune responses to inoculation with 10 CFU or 107 CFU ΔkbvR. Mice were immunized with 105 CFU (n = 10), 107 CFU (n = 5) ΔkbvR or PBS (n = 10) by the subcutaneous route for two doses on days 0 and 14. A Blood was collected on day 28 post-inoculation, and the relative levels of serum IgG antibody against whole cell lysates of NTUH-K2044 were determined by ELISA. The optical density of the horseradish peroxidase reaction was recorded at 450 nm. B Daily weights of mice inoculated with 105 CFU ΔkbvR, 107 CFU ΔkbvR, or PBS over 28 days. C Histopathological examination of the liver, spleen, and lung of 105, 107 ΔkbvR or PBS immunized mice on day 28 post-inoculation. The section was counterstained with hematoxylin and eosin. Scale bars = 100 μm. *: P < 0.05; ****: P < 0.0001.

Full size image

The weights and the skin changes in the injection sites of mice throughout the immunization were measured as indicators of side effects. No skin changes at the sites of injection and weight loss were observed in all mice inoculated with 105 CFU ΔkbvR. However, mice initially inoculated with 107 CFU ΔkbvR had significantly lower weights (about 12%) than PBS-inoculated controls from days 1–6 post the first inoculation (P < 0.05), and local adverse reactions such as skin lesion and a local hair loss at the injection site were observed (Figure 3B). These differences were significant when analyzed by mixed ANOVA study. The mice recovered their body weight on day 8, and no significant weight differences from PBS control were observed after the second administration. The liver, spleen, and lungs of PBS control, 105, or 107 CFU ΔkbvR-inoculated mice were examined histologically at 28 days after the first injection to further investigate the virulence and safety of ΔkbvR in mice. No histological lesions and inflammatory infiltrate were observed in any examined organ (Figure 3C). These data demonstrate that, although the 107 CFU-immunized dose induced higher humoral response, it also might have the local side effect. A dose of 105 CFU ΔkbvR which was well within the range of safe and immunogenic doses was chosen for further evaluation.

ΔkbvR immunization induces protection against K. pneumoniae intraperitoneal challenge

The attenuated virulence and stimulated antibody production of kbvR mutant strain made it a candidate for consideration as a live attenuated vaccine. Thus, the ability of ΔkbvR strain to confer protection against lethal intraperitoneal challenge with hypervirulent K. pneumoniae strain NTUH-K2044 (WT, LD50 = 500 CFU) was tested. BALB/c mice were subcutaneously immunized with 105 CFU ΔkbvR or PBS control at two doses, 14 days apart. A total of 2 weeks after the second vaccination, mice were intraperitoneally challenged with 104 or 105 CFU WT. At 12 h after the challenge, 5 mice of immunized or normal control mice infected with 105 CFU WT were sacrificed. The peritoneal washes were smeared, stained, and observed with a microscope. No bacteria were observed in the peritoneal lavage fluids of immunized mice. However, plenty of K. pneumoniae were found in the abdominal cavities of control mice (Figure 4A and B). The blood, peritoneal washes, liver, lungs, and spleen were harvested to calculate the bacterial number. Immunized mice exhibited significant reductions in organ colonization in the peritoneal washes (1.86-log reduction), liver (1.9-log reduction), spleen (1.6-log reduction), lungs (1.74-log reduction), and blood (3.45-log reduction) compared with the bacterial number of PBS controls (Figure 4C and D). Three of five immunized mice exhibited no detectable bacterial colonization in the blood. By 2 weeks post-challenge, none of the control mice survived beyond 48 h. By contrast, the immunized group had survival rates of 100% and 87.5% for 104 CFU and 105 CFU challenge, respectively (Figure 4E). Overall, these results suggest that the immunization with 105 CFU ΔkbvR conferred significant protection in mice against K. pneumoniae challenge.

Figure 4
figure 4

Δ kbvR immunization induces protection against K. pneumoniae intraperitoneal challenge. Two doses of 105 CFU ΔkbvR strain immunized mice (n = 21) or PBS (n = 19) inoculated normal mice were intraperitoneally infected with 104 (n = 15) or 105 (n = 25) CFU log-phase wild-type strain 2 weeks after the second immunization. A, B Peritoneal lavage fluids of immunized (n = 5) or normal mice (n = 5) were collected, stained, and observed with a microscope at 12 h after being infected with 105 CFU WT. Arrows indicate the bacteria. C, D Number of bacteria in peritoneal washes, liver, spleen, lung, and blood of immunized or normal mice were determined at 12 h after infection with 105 CFU WT. E Survival curves of immunized or normal mice challenged with 104 or 105 CFU WT (n = 7 for each PBS group and n = 8 for each ΔkbvR immunized group) (P < 0.0001).

Full size image

Antisera from immunized mice increase uptake of K. pneumoniae by macrophages in vitro

We measured the ability of the antibodies in immunized sera to promote K. pneumoniae uptake by opsonophagocytic assays for determining their functional activity. Phagocytic assay revealed that NTUH-K2044 was rarely presented inside macrophages without adding sera (Figure 5A). Although the addition of normal sera from PBS control mice increased the phagocytic activity of macrophages, the pooled sera from ΔkbvR immunized mice show statistically significant enhancement in opsonophagocytic activity compared with the pooled normal sera, whose anti-Klebsiella titers were 1:3200 and 1:400 respectively. The average opsonophagocytized NTUH-K2044 were about 80 and 30 within 300 macrophages respectively (P < 0.0001) (Figure 5B). The opsonophagocytic assay of the sera against the K. pneumoniae serotype K3 strain ATCC 13,883 was also tested to evaluate whether the sera antibodies generated by immunization with the serotype K1 NTUH-K2044 ΔkbvR cross-reacted with the other K. pneumoniae serotype. The anti-phagocytosis ability of ATCC 13,883 against macrophage was much lower than that of NTUH-K2044 (Figure 5A), which may be partly due to the difference in CPS. Meanwhile, the pooled sera from mice immunized with two doses of serotype K1 ΔkbvR strain were also able to significantly increase macrophage uptake of ATCC 13,883 over the normal sera from PBS immunized animals (7.175% versus 4.575%, P < 0.05) (Figure 5C). The antisera provided the opsonophagocytic activity not only to the K1 serotype but also to the K3 serotype strain, which implied that immunization with the kbvR gene deletion strain of K1 serotype might confer cross-protective immunity against other serotypes of K. pneumoniae.

Figure 5
figure 5

Determination of opsonophagocytic activity with immunized sera against NTUH-K2044 and ATCC 13,883 by confocal micrograph. A RAW264.7 macrophage cells were infected with the serotype K1 NTUH-K2044 or the serotype K3 ATCC 13,883 containing the plasmid pLac-EGFP (green) at an MOI of 50 with or without 1:1100 final dilutions of sera from PBS-inoculated normal mice or 105 CFU ΔkbvR-immunized mice for 120 min, followed by treatment with 500 µg/mL gentamicin and staining with Alexa Fluor 594 dye (red) for actin cytoskeleton and DAPI (blue) for host cell nuclei. The opsonophagocytic activities of sera were observed under a confocal microscope. Arrows denote the intracellular bacteria. Scale bar = 5 μm. B The opsonophagocytic activities of sera against NTUH-K2044 were counted under a confocal microscope. Data are presented as the intracellular bacterial numbers in 300 cells from three independent trials. C The opsonophagocytic activities of sera against ATCC 13,883 were determined by the phagocytic rate. The mean opsonophagocytic ratio was calculated by the formula = 100 ⋅ CFU of intracellular phagocytized bacteria/CFU of input bacteria. *: P < 0.05; **: P < 0.01; ****: P < 0.0001.

Full size image

Antisera from immunized mice increase complement-mediated lysis of K. pneumoniae

Complement activation is an important component of the innate immunity of the antibacterial immunity. We characterized the complement activation by mice immunized sera in vitro to test whether antisera from immunized mice can increase complement-mediated lysis of K. pneumoniae. The K. pneumoniae NTUH-K2044 or ATCC 13,883 strain was incubated with the heat-inactivated immunized or normal mice sera, along with human sera as a complement. Compared to the lysis induced by normal sera, the survival rate induced by the immunized sera against NTUH-K2044 and ATCC 13,883 were both significantly decreased (66.4% versus 95.9%; and 63% versus 90%) (Figure 6). The above-mentioned results suggest strong complement activation after immunization with ΔkbvR, which might contribute to improving the antibacterial ability of the host immune system.

Figure 6
figure 6

Immunized sera reduce survival rate of complement-mediated antibacterial test against K. pneumoniae. The K. pneumoniae strain NTUH-K2044 or ATCC 13,883 (106 CFU) were co-incubated with human complement with or without 10% heat-inactivated sera from pooled immunized or normal mice for 2 h at 37 °C. The survival ratio was determined using the formula = 100 ⋅ CFU of experimental wells with mice sera/CFU of control wells without mice sera. **: P < 0.01; ****: P < 0.0001.

Full size image

Passive transfer of antisera from immunized mice provides protection against K. pneumoniae infection

We determined whether passive transfer of immunized sera to naïve mice could provide protection against infection of K. pneumoniae to examine the protective role of antibodies in vivo. Pooled sera (n = 5) were obtained from BALB/c mice immunized with 105 CFU ΔkbvR (immunized sera) or PBS control (normal sera). The 1 µL pooled immunized or normal sera were intraperitoneally injected into naïve mice (n = 5 per group) together with 100 µL K. pneumoniae WT (approximately 104 CFU). All mice that were passively transferred with normal sera succumbed to infection within 2 days after injection. On the contrary, all mice treated with immunized sera resolved the infection for 8 days and 40% mice survived for 14 days (Figure 7). These results support the idea that passive transfer of immunized sera is protective against K. pneumoniae lethal infection in mice.

Figure 7
figure 7

Protective effect of passive transfer of mice sera against lethal K. pneumoniae infection. The survival rates of BALB/c mice (n = 5/group) injected intraperitoneally with the mixture of 104 CFU WT and immunized or normal mice sera were determined for 14 days (P < 0.01).

Full size image

Discussion

Klebsiella pneumoniae is an emerging zoonotic pathogen that causes various severe infections in humans and animals, and these infections become untreatable due to the emergence of MDR strains [11, 38]. Therefore, vaccines against K. pneumoniae that prevent infection in veterinary practices and humans need to be developed urgently. Among the efforts, live attenuated vaccines are still an appealing approach for animals because of their high immunogenic potential, a greater number of different immunogens, and ability to elicit long-lasting protection. The basic requirements for live vaccines are to reduce the virulence and ensure they contain the highest number of antigenic determinants necessary for inducing protective immunity [39]. In K. pneumoniae, the CPS, fimbriae antigens (type I fimbriae and type III fimbriae), and outer membrane proteins (OmpK36 and OmpA) were proven to be potential antigenic determinants for the development of subunit vaccines [17]. Our previous study found that KbvR was associated with the virulence and the synthesis of CPS in K. pneumoniae NTUH-K2044 [28]. The decreased expression of CPS in NTUH-K2044 ΔkbvR contributes to the attenuation of strain, and the maintained CPS on the surface of ΔkbvR strain should still stimulate the protective immune response against the infection of serotype K1 strain. Except for CPS, the synthesis of type I fimbriae was found to be negatively regulated by KbvR and the expression level of type I fimbriae was increased in the ΔkbvR strain. Therefore, the weakened physical shielding effect of CPS on the fimbriae along with the reduction in CPS, and the increased synthesis of type I fimbriae on the surface, ultimately enhanced the activity and the chance of type I fimbriae to contact with the immunity system. The highly conserved fimbriae components have been generally shown to be promising candidates for inclusion in vaccines [40,41,42]. In addition, the structural proteins of type 1 fimbriae have been confirmed as effective protein carriers and immunogens in conjugates [43, 44]. Overall, the maintained CPS and the increased synthesis of type 1 fimbriae made ΔkbvR strain a promising live attenuated vaccine that can provide broad antigen exposure and serve as intrinsic adjuvant properties to prime durable immune responses.

Given that ΔkbvR could be a live attenuated vaccine candidate, we chose to further characterize the safety and protective effect of this strain using the murine model by subcutaneous route, a used delivery method for live attenuated vaccine. The findings indicated a positive effect of ΔkbvR on antibody induction and that the immunization of the ΔkbvR strain can stimulate the dose-dependent humoral immune response. The 107 CFU dose-immunized group had higher antibody concentration than the 105 CFU dose-immunized group. Although the mice immunized by 105 CFU ΔkbvR strain could provide 100% protection against the 104 CFU WT strain by intraperitoneal challenge, one mouse died among 105 CFU-challenged mice. When the higher dose 107 CFU was used for immunization, 100% protection against 105 CFU challenge at 14 days or even at 42 days after the second immunization were provided (data not shown). However, mice initially inoculated with 107 CFU ΔkbvR strain had significantly lower weights and had a local skin side effect. Therefore, the optimum immune dose of ΔkbvR with the best protective effect and safety will be fully evaluated in the future.

Klebsiella pneumoniae especially K1/K2 serotype strains resist to phagocytosis and complement-mediated killing through the thick capsule at the cell surface [45]. The serum from patients with recurrent K. pneumoniae liver abscess increased the opsonizing and killing efficacy, suggesting that the humoral immune response plays an important role in eliminating bacterial infection [46, 47]. In this study, the opsonophagocytic assay and in vitro serum bactericidal activity demonstrated that the mice antisera from 105 CFU ΔkbvR-immunized group prompted efficient phagocytosis by macrophages and induced complement-mediated lysis not only against the homologous (K1) serotype strain but also against the heterologous (K3) serotype strain. Thus, the broad immunogens of the live NTUH-K2044 ΔkbvR strain might provide a cross humoral protection to other serotype strains. The passive transfer of antisera from ΔkbvR immunized mice to naïve mice provided protection against K. pneumoniae infection in vivo. The immunization with ΔkbvR strain at 105 CFU dose induced an effective antibody protection against infection.

In conclusion, we evaluated the safety and protective effect of KbvR regulator mutant strain. The results suggest that the ΔkbvR strain can provide protection against the same serotype and different serotypes, which implies its potential as a live attenuated vaccine against K. pneumoniae infection in animals or humans. In the future, the selection of an appropriate immunization dose of the ΔkbvR strain which balances between attenuation and immunogenicity, and the cross-protective effect of ΔkbvR against additional K. pneumoniae serotype strains and MDR strains should be studied subsequently to provide a theoretical basis for developing a commercial vaccine.