Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

The role of EII complex in the bacterial responses to the glucose-survey in clinical Klebsiella pneumoniae isolates

  • Yu-Tze Horng,

    Roles Writing – original draft, Writing – review & editing

    Affiliation Department of Laboratory Medicine and Biotechnology, College of Medicine, Tzu Chi University, Hualien, Taiwan, R.O.C

  • Novaria Sari Dewi Panjaitan,

    Roles Investigation, Validation

    Affiliations Department of Laboratory Medicine and Biotechnology, College of Medicine, Tzu Chi University, Hualien, Taiwan, R.O.C, Center for Biomedical Research, Research Organization for Health, National Research and Innovation Agency (BRIN), Cibinong Science Center, Cibinong, Bogor, West Java, Indonesia

  • Yi-Jhen Tsai,

    Roles Investigation, Validation

    Affiliation Department of Laboratory Medicine and Biotechnology, College of Medicine, Tzu Chi University, Hualien, Taiwan, R.O.C

  • Pin-Wei Su,

    Roles Investigation, Validation

    Affiliation Department of Laboratory Medicine and Biotechnology, College of Medicine, Tzu Chi University, Hualien, Taiwan, R.O.C

  • Hung-Chi Yang,

    Roles Resources

    Affiliation Department of Medical Laboratory Science and Biotechnology, Yuanpei University of Medical Technology, Hsinchu, Taiwan, R.O.C

  • Po-Chi Soo

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    pcsoo@mail.tcu.edu.tw

    Affiliation Department of Laboratory Medicine and Biotechnology, College of Medicine, Tzu Chi University, Hualien, Taiwan, R.O.C

Abstract

Type 3 fimbriae in Klebsiella pneumoniae are important for bacterial colonization on abiotic and biotic surfaces. The major subunit of type 3 fimbriae (MrkA) is increased by overexpression of EtcABC, an EII complex of phosphoenolpyruvate:carbohydrate phosphotransferase systems (PTSs), through cAMP-cAMP receptor protein (cAMP-CRP) in K. pneumoniae STU1. Here, we further characterized the relations between the amount of etcABC mRNA and MrkA in 78 clinical K. pneumoniae isolates incubated in high levels of glucose. By Western blotting, we observed that MrkA of 29 isolates were not decreased much by high levels of glucose (Group A) but MrkA of other 49 isolates were significantly reduced (Group B) in the same condition. The bacterial biofilms on abiotic surfaces and colonization in the Caenorhabditis elegans of representative isolates in the Group A were not affected by high levels of glucose. However, the biofilm and colonization in the worm of clinical isolates in the Group B were much reduced by high levels of glucose. After quantification by real time RT-PCR, 76% of Group A but just 10% of Group B showed high amount of etcA mRNA. In summary, our results suggested that for most of K. pneumoniae clinical isolates, the amount of etcABC mRNA was positively related to their type 3 fimbriae production in a high level of glucose, thereby to their biofilm formation and colonization in the worm.

Introduction

Klebsiella pneumoniae can be found ubiquitously in natural superficial waters and soils [1] and colonize in the gastrointestinal tracts of humans and animals [24]. K. pneumoniae can be detected in 62.1% of stool specimens from healthy adults [5]. In addition, K. pneumoniae can cause opportunistic infections, including urinary tract infections and pneumonia [1, 6, 7]. Human guts are reservoirs of hypervirulent K. pneumoniae [8]. K. pneumoniae forms biofilms on catheters or ventilators, increasing the risk of urinary tract infections, bloodstream infections, and pneumonia [9]. K. pneumoniae can also form biofilms on tissue surfaces. Furthermore, K. pneumoniae adheres to the surfaces mainly via type 1 and type 3 fimbriae [1]. Most members of the Enterobacteriaceae family express type 1 fimbriae, but just several members of this family produce type 3 fimbriae [10]. In K. pneumoniae, type 3 fimbriae are synthesized by proteins encoded by the mrkABCDF operon [11]. The major subunits of the type 3 fimbrial shaft are encoded by the mrkA gene, while MrkD is an adhesin of type 3 fimbriae that facilitates bacterial binding to the collagen-coated surface or extracellular matrix formed by bronchial cells. In addition, type 3 fimbriae promote the colonization of K. pneumoniae on abiotic surfaces to form biofilms [10, 12]

To translocate various sugars, bacteria possess several phosphoenolpyruvate (PEP): carbohydrate phosphotransferase systems (PTSs) to respond to carbohydrate availability. The phosphate group is sequentially transferred from PEP to the related carbohydrates via PTS, which generally consists of enzyme I (EI), histidine-containing phosphocarrier protein (HPr) and enzyme II (EII) complexes. EI and HPr are general to all PTSs, but EII complexes are sugar-specific. EII complexes are composed of EIIA, EIIB and EIIC (and sometimes EIID) proteins/domains. The membrane-bound EIIC is directly responsible for translocating sugar into bacteria. The phosphate group from HPr is transferred to EIIA, then to EIIB and eventually to the sugar in the cytoplasm [13, 14]. In addition to carbohydrate transportation, EIIA and EIIB are also involved in various physiological processes in bacteria [15]. For example, Crr, which is a glucose-specific EIIA in enteric bacteria, interacts with glycerol kinase, leading to inhibition of glycerol uptake in the presence of glucose. In addition, Crr regulates the synthesis of cAMP via interactions with adenylate cyclase (AC) [13, 16]. We reported that EtcA, EtcB and EtcC in K. pneumoniae STU1 (homologous to KPN00353, KPN00352 and KPN00351 in K. pneumoniae MGH78578, respectively) were novel EIIA, EIIB, and EIIC homologs respectively and encoded from genes in an operon, etcABC [17]. Overexpression of etcABC increased biofilm and type 3 fimbriae in K. pneumoniae [18, 19]. Deficiency of crr in K. pneumoniae STU1 resulted in a low level of cAMP production, but overexpression of etcABC elevated the cAMP level in the ΔcrretcABC double mutant [19]. The capsular polysaccharide (CPS) of the crr mutant is higher than that of the wild type, but deletion of etcABC in the crr mutant decreases CPS synthesis [20].

We previously reported that EtcABC regulated the type 3 frimbriae in K. pneumonniae STU1 via cAMP-cAMP receptor protein (CRP) [19]. Given literature on the activity of CRP regulated by the glucose-specific EIIA in other bacteria [16], we identified the relation between glucose, etcABC and type 3 frimbriae in K. pneumoniae in this study. Since the result from a single strain may not represent the phenomena of most clinical strains, 78 clincal K. pneumoniae isolates were examined in this study.

Materials and methods

K. pneumoniae strains and growth conditions

K. pneumoniae STU1 is a laboratory-maintained strain that was acquired from National Taiwan University (Taipei, Taiwan). Seventy-eight K. pneumoniae clinical strains were isolated from clinical specimens (sputum, blood, wound, urine and others) during 2016–2018 in Tzu Chi Hospital (Hualien, Taiwan) and then transferred to Tzu Chi University (Hualien, Taiwan) through an official transfer. K. pneumoniae STU1/pBSK-Km::ZsGreen, which carries the ZsGreen gene to produce green fluorescence, was constructed in our previous study [20]. K. pneumoniae Clin200 and K. pneumoniae Clin73, two of the 78 clinical isolates mentioned above, were isolated from sputum specimens. In this study, the plasmid pBSK-Km::ZsGreen was transformed into K. pneumoniae Clin200 and K. pneumoniae Clin73 to emit green fluorescence according to methods described previously [20]. K. pneumoniae STU1/pBSK::Gm::etcABC, which carries the promoter and structure of etcABC in the plasmids to overexpress etcABC, and K. pneumoniae STU1/pBSK::Gm, which was the vector control in this study, were constructed in our previous study [19]. K. pneumoniae was routinely cultured in Luria-Bertani (LB) medium (0.5% yeast extract, 1% tryptone and 1% NaCl) at 37°C. For the bacteria carrying pBSK-Km::ZsGreen, pBSK, pBSK::Gm or pBSK::Gm::etcABC, LB was supplemented with 50 μg/mL kanamycin or 20 μg/mL gentamicin in routine culture.

Nematode maintenance

Caenorhabditis elegans N2 (wild type) was a kind gift from Professor Hung-Chi Yang at the Department of Medical Laboratory Science and Biotechnology, Yuanpei University of Medical Technology and was propagated on a nematode growth medium (NGM) plate seeded with Escherichia coli OP50 prior to being fed with K. pneumoniae [21].

Observation of bacteria in the intestine of a nematode

After bacteria with pBSK-Km::ZsGreen were cultured in LB with 50 μg/mL kanamycin at 37°C overnight, the bacteria were resuspended in fresh LB to an OD600 of 9. Fifty microliters of this bacterial suspension were pipetted onto NGM plates without or with 2% glucose. Then, approximately 50~60 C. elegans were cultured on these plates for 3 days. After moving from the plates to microscope glass slides, the worms were washed with 0.9% NaCl three times. Subsequently, the C. elegans was paralyzed with 200 mM sodium azide for 5~10 min. Images of nematodes were captured using an upright fluorescence microscope Nikon Ni-E (Nikon, Japan). In order to quantify the fluorescence levels in worms, 20 C. elegans in a well of microplate (PerkinElmer, USA) were detected at 505 nm after 492-nm excitation by Varioskan Flash microplate reader (Thermo Fisher Scientific, USA).

Western blotting

After overnight culture in LB, bacteria were diluted in LB containing 0%, 0.1%, 0.2%, 0.5%, 1% or 2% glucose respectively to optical density at 600 nm (OD600) of 0.08 and then incubated for 24 hours at 37°C. After determination of bacterial concentration by measuring their OD600, the bacteria were suspended in the SDS sample buffer and lysed at 100°C for 10 min. An aliquot of cell lysis was applied to 12% SDS-PAGE, following the procedures by Sambrook [22]. The MrkA and ManA (mannose 6-phosphate isomerase), were detected using Western blotting according to our previous study [19]. ManA was used as the loading control in Western blotting. The Western blotting signals were detected using the Gel Catcher 2850 chemiluminescence camera system (CLUBIO, Taipei, Taiwan) and analyzed by ImageJ (National Institutes of Health).

Quantification of biofilm

The bacterial biofilms formed in LB without/with 2% glucose for 24 hours at 37°C were quantified according to our previous study [19].

Reverse transcription quantitative real-time PCR (RT-qPCR)

The transcriptional levels of etcA and recA were quantified by RT-qPCR as previously described and normalized to that of 16S rRNA following the 2−ΔΔCT method [20]. The recA, a housekeeping gene encoding recombinase A, was used as a reference to compare the gene expression of etcA. All primers and probes are listed in Table 1.

thumbnail
Table 1. Oligonucleotide primers and probes for RT-qPCR in this study.

https://doi.org/10.1371/journal.pone.0289759.t001

Statistical analysis

For RT-qPCR and quantification of biofilm and fluorescence, the values were expressed as the average +/- standard deviation (SD) from three independent experiments. Data were subjected to analysis using a student’s t-test. Significant difference was considered at p < 0.05.

Results

Discrepancy in the effects of glucose on type 3 fimbriae and biofilms in K. pneumoniae clinical isolates

We observed that 0.5% (and more) glucose inhibited the expression of MrkA in K. pneumoniae STU1 (Figs 1 and 2). This phenomenon is inconsistent with the finding by Lin et al., who reported that 0.5% glucose increased MrkA expression in K. pneumoniae CG43S3 [23]. The different phenotypes of these two K. pneumoniae isolates inspired us to examine the clinical isolates. To determine which concentration of glucose cannot be exhausted by bacteria after overnight culture, 15 K. pneumoniae clinical isolates (including STU1) were incubated in LB medium supplemented with 0.5%, 1% or 2% glucose respectively. The remnant glucose in the supernatant of bacterial culture after incubation for 24 hours was examined by Multistix SG reagent strips (Siemens, Germany). The results showed that 11 clinical isolates (including STU1) depleted 1% glucose, but all 15 isolates did not exhaust the 2% glucose after culture (Fig 3). Subsequently, 2% glucose in LB was chosen to examine MrkA expression in 78 clinical isolates. By Western blotting, MrkA expression in 29 clinical isolates incubated in LB with 2% glucose were not/slightly reduced, compared to those in LB (Group A in Table 2 and Clin73 is shown as a representative in Fig 2). In contrast, MrkA expression in the other 49 bacteria incubated in LB with 2% glucose was less than half of their MrkA expression in LB (Group B in Table 2 and Clin200 is shown as a representative in Fig 2). Thus, we found that glucose reduced the expression of MrkA in 63% of clinical isolates but did not decrease MrkA expression in the other clinical isolates (37%).

thumbnail
Fig 1. The effects of glucose on K. pneumoniae STU1.

The bacteria were incubated in LB (Lane 1) or LB supplemented with glucose from 0.1% (duplicated loading on lane 2 and 3), 0.2% (lane 4 and 5), 0.5% (lane 6 and 7), 1% (lane 8 and 9) to 2% (lane 10 and 11) as the final glucose concentration. After incubation for 24 hours, the cellular proteins were analyzed by SDS-PAGE (i) and Western blotting using an antibody against MrkA (anti-MrkA) (ii). PC: MrkA (21 kDa) as a positive control. The protein marker is on the left lane of SDS‒PAGE. The SDS‒PAGE images showed consistency of loading and general pattern of cell lysate of each sample. Representative images of SDS and Western blotting were from three independent experiments.

https://doi.org/10.1371/journal.pone.0289759.g001

thumbnail
Fig 2. Western blotting analysis of MrkA and ManA.

The bacteria were incubated in LB (-) or LB supplemented with 2% glucose (+) for 24 hours. STU1: K. pneumoniae STU1. STU1/etcABC: K. pneumoniae STU1/pBSK::Gm::etcABC; Clin200 and Clin73: clinical K. pneumoniae isolates. ManA (mannose 6-phosphate isomerase) was the loading control. Representative images were from three independent experiments.

https://doi.org/10.1371/journal.pone.0289759.g002

thumbnail
Fig 3. Examination of the remnant glucose in medium.

(a) Fifteen K. pneumoniae clinical strains were incubated in LB medium supplemented with 0.5%, 1% or 2% glucose for 24 hours. (b) After incubation for 24 hours, the remnant glucose in medium was examined by Multistix SG reagent strips (Siemens, Germany). The images were representatives of 15, 4 and 15 clinical strains respectively in 0.5%, 1% and 2% glucose. (c) The titer of remnant glucose was determined by following the manufacturer’s instruction.

https://doi.org/10.1371/journal.pone.0289759.g003

thumbnail
Table 2. The number of clinical isolates tested in Western blotting and RT-qPCR.

https://doi.org/10.1371/journal.pone.0289759.t002

Since MrkA is the major subunit of type 3 fimbriae, which promotes the attachment of bacteria to abiotic surfaces to form biofilms [1], the effects of glucose on biofilms of K. pneumoniae STU1, Clin200 and Clin73 were examined. The results showed that the biofilms of K. pneumoniae STU1 and Clin200 in LB with glucose were much less than those in LB, while the biofilm of K. pneumoniae Clin73 in LB with glucose did not decrease much, compared to that in LB (Fig 4). Thus, the effects of glucose on fimbriae and biofilms are positively correlated in one strain, but the effects of glucose on the K. pneumoniae population vary.

thumbnail
Fig 4. The effects of glucose on biofilm formation.

The bacteria were incubated in LB (-) or LB with 2% glucose (+) for 24 hours. The amount of biofilm in LB with glucose was compared to that in LB. The error bars indicate the SD from three independent cultures. An asterisk (*) indicates that the data were significantly different at p < 0.05. STU1: K. pneumoniae STU1. STU1/etcABC: K. pneumoniae STU1/pBSK::Gm::etcABC, Clin200 and Clin73: clinical K. pneumoniae isolates.

https://doi.org/10.1371/journal.pone.0289759.g004

The effects of glucose on bacterial colonization in Caenorhabditis elegans

Little is known about the effects of glucose on the colonization of K. pneumoniae in Caenorhabditis elegans. To directly observe the bacteria in the gut of worms, The plasmid, pBSK-Km::ZsGreen, was purposely transformed into K. pneumoniae STU1, Clin200 and Clin73 respectively. After C. elegans was fed a lawn of K. pneumoniae in NGM without/with 2% glucose for 3 days (Figs 5 and 6). The activity and lifespan of the worms fed with anyone of K. pneumoniae isolates (STU1, Clin200 and Clin73) were not different. However, the fluorescence intensities from K. pneumoniae STU1 and Clin200 were decreased in the nematodes from the NGM with glucose, compared to that from the NGM. In contrast, the fluorescence intensity of K. pneumoniae Clin73 in C. elegans on the NGM with glucose addition was not significantly different from that on the NGM (Figs 5 and 6). This result indicated that the colonization number of K. pneumoniae STU1 and Clin200 in C. elegans was reduced by glucose, but glucose did not affect the colonization of K. pneumoniae Clin73 in the worm. To confirm whether glucose influences C. elegans, C. elegans was fed on a lawn of E. coli OP50 on NGM without/with 2% glucose. The results showed that the activity and lifespan of the nematode were not influenced by glucose. Thus, the results showed the various bacterial responses to glucose in colonization in C. elegans, related to bacterial fimbriae production.

thumbnail
Fig 5. The effects of glucose on bacterial colonization in the gut of C. elegans.

K. pneumoniae carrying the pBSK-Km::ZsGreen (ZsGreen) emitted fluorescence. The bacteria were seeded on NGM (-) or NGM with 2% glucose (+) to feed the C. elegans. Representative images of bacteria in the gut of nematodes were acquired by fluorescence microscopy from three independent experiments. STU1: K. pneumoniae STU1. Clin200 and Clin73: clinical K. pneumoniae isolates.

https://doi.org/10.1371/journal.pone.0289759.g005

thumbnail
Fig 6. Quantification of bacteria in the gut of C. elegans.

The bacteria carrying the pBSK-Km::ZsGreen were seeded on NGM (black bar) or NGM with 2% glucose (white bar) to feed the C. elegans. The fluorescence levels of 20 worms were quantified and then expressed as the average + SD from three independent experiments. STU1: K. pneumoniae STU1. Clin200 and Clin73: clinical K. pneumoniae isolates. An asterisk (*) indicates that the data were significantly different at p < 0.05. ns indicates non-significant difference.

https://doi.org/10.1371/journal.pone.0289759.g006

High amount of etcABC mRNA is related to fimbriae resistance to glucose inhibition

In our previous study, overexpression of etcABC in K. pneumoniae STU1 enhanced type 3 fimbriae expression in the LB [19]. Furthermore, K. pneumoniae STU1/pBSK::Gm::etcABC, also showed high levels of the type 3 fimbriae in the LB with glucose, compared to its parent strain (STU1) (Fig 2). In addition, the biofilm of K. pneumoniae STU1/pBSK::Gm::etcABC was not decreased in the LB with glucose, compared with that in the LB (Fig 4B). We did not transform pBSK-Km::ZsGreen into the K. pneumoniae STU1/pBSK::Gm::etcABC to observe it in the gut of the nematode because of the same replication origin of two plasmids.

Herein, we hypothesized that the clinical bacteria in Group A had high mRNA levels of etcABC, resulting in their high production of type 3 fimbriae in the medium with glucose. Therefore, the amount of etcA mRNA in clinical isolates was quantified to examine this hypothesis since etcABC was transcribed as an operon [17]. After RT‒qPCR was performed, the amount of etcA mRNA was normalized to that of recA mRNA, which is a housekeeping gene, in an individual clinical isolate. If the ratio of etcA to recA was higher than one, the amount of etcA mRNA was defined as high level. The results showed that high levels of etcA mRNA were found in 22 isolates in Group A (76%, 22/29). However, 20 isolates were randomly selected from Group B to be tested and showed that 90% (18/20) isolates in Group B expressed low amount of etcA mRNA (Table 2 and Fig 7). The results indicated that amount of etcABC mRNA was positively related to type 3 fimbriae production of K. pneumoniae in the high concentration of glucose.

thumbnail
Fig 7. RT‒qPCR analysis of the etcA and recA transcripts in clinical isolates.

The mRNA level of etcA was compared to that of recA, a housekeeping gene encoding recombinase A, in each K. pneumoniae isolate. MrkA>50%: the 29 clinical isolates in Group A. MrkA<50%: the 20 clinical isolates randomly selected from Group B. Each dot represents one clinical isolate. The error bars indicate SD from the population. Asterisks (***) represent p < 0.05.

https://doi.org/10.1371/journal.pone.0289759.g007

Discussion

Through analyzing clinical K. pneumoniae isolates, the various effects of glucose on K. pneumoniae were observed in this study. For 37% of the clinical K. pneumoniae isolates (Group A), their MrkA was not decreased or was slightly decreased at high glucose concentrations, but MrkA in 63% of the clinical isolates (Group B) was reduced by glucose (Table 2). Most strains in Group A also showed high amount of etcA mRNA. In contrast, most strains in Group B showed low amount of etcA mRNA (Table 2 and Fig 7). The K. pneumoniae isolate Clin73, one of GroupA, also presented a high level of biofilm on the abiotic surface and colonization in the gut of worms in the medium with glucose, compared to that without glucose (Figs 46). In general, the effects of glucose on the expression of type 3 fimbriae, biofilm formation and gut colonization in the worm were positively related to the amount of etcABC mRNA in the most of K. pneumoniae isolates.

We previously reported that etcABC operon is wildly distributed in the most of K. pneumoniae isolates [17]. Overexpression of EtcABC increased the amount of cAMP in K. pneumoniae STU1. The high level of cAMP caused CRP to increase the transcription of the mrk operon, leading to an increase in type 3 fimbriae production (Fig 8A) [19]. In addition, high concentration of glucose results in low amount of intracellular cAMP in E. coli, which does not have etcABC (Fig 8B) [16, 24]. Taken together, we speculated that high concentration of the glucose led to inactivation of Crr (unphosphorylated Crr) and the amount of etcABC mRNA and cAMP levels in 90% of clinical isolates in Group B were low, resulting in reduced MrkA in the bacteria growing in the medium containing high concentration of glucose (Fig 8D). In contrast, even growing in the high concentration of glucose, the 76% of clinical isolates in Group A expressed high amount of etcABC mRNA and cAMP-CRP, resulting in high expression of type 3 fimbriae (Fig 8C).

thumbnail
Fig 8. Schematic diagram of the roles of EtcABC in type 3 fimbriae production by K. pneumoniae growing in the medium containing glucose.

The possible regulatory mechanisms are proposed by combination of the previous research and this study. (A) Overexpression of EtcABC in K. pneumoniae crr mutant increased the activity of cAMP-CRP and then mrk operon, resulting the high production of type 3 fimbriae in LB [19]. (B) In E. coli and several members in Enterobacteriaceae, the activity of cAMP-CRP is increased by Crr in phosphorylated state but inhibited by glucose due to dephosphorylation of Crr in the environment with glucose [16]. (C) In 76% of Group A growing in the LB containing glucose, high amount of etcABC mRNA (EtcABC) activates cAMP-CRP and then mrk operon is increased. (D) In 90% of Group B growing in the LB containing glucose, Crr in unphophorylated state and EtcABC in low amount cannot activate cAMP-CRP, leading to decreased expression of mrk operon and then type 3 fimbriae production. Crr-P: phosphorylated Crr. mrk: mrk operon. Dashed arrow: inhibition. Black arrow: activation. White arrow: expression levels of genes are increased (up) or decreased (down). Forbidden sign: the gene regulatory activity is inhibited or reduced. glu: glucose.

https://doi.org/10.1371/journal.pone.0289759.g008

Thus, we hypothesized that the upstream DNA regions of etcABC may affect the transcriptional level of etcABC in various K. pneumoniae clinical isolates. Then, we examined 400-bp DNA region upstream of etcABC in several clinical isolates and chose some of them to be ligated with the reporter gene, luxCDABE [19]. By measuring the bioluminescence levels from the reporter, the results showed that the transcriptional activities from these 400-bp regions were not different. Therefore, the 400-bp upstream region of etcABC is not the factor in 2% glucose to reduce etcABC transcription in Group B or keep etcABC transcription high in Group A.

Lin et al. reported that MrkA in K. pneumoniae CG43S3 was increased in LB with 0.25%-0.5% glucose, compared to that in LB [23]. Although we did not examine K. pneumoniae CG43S3, any clinical isolate in this study did not show that MrkA was increased in the LB with 2% glucose, compared to that in LB. In addition, 0.25%-0.5% glucose is inferred to be exhausted by bacteria after culture (Fig 3). Therefore, the model of MrkA regulation by glucose that proposed by Lin et al. is rarely applicable in clinical K. pneumoniae isolates. However, the report by Lin et al. also demonstrated that the biofilm of K. pneumoniae CG43S3 in LB with 0.5%-2% glucose was less than that in LB, similar to K. pneumoniae STU1 and K. pneumoniae Clin200 in Fig 4.

A retrospective study reported that 44% of patients with K. pneumoniae bacteremia in Taiwan in 2014 and 2016 were diabetes mellitus (DM) patients [25]. From the findings in this study, we think that high amount of etcABC mRNA and type 3 fimbriae of K. pneumoniae resistance to glucose may play important roles in DM patients that suffer from infection. Therefore, further study on high-etcABC-mRNA K. pneumoniae in DM patients is needed in the future. In addition, DM was defined as a fasting blood glucose level of 126 mg/dL or higher on two separate tests. (World Health Organization. https://www.who.int/data/gho/indicator-metadata-registry/imr-details/2380). The 2% glucose concentration in this study was much higher than the blood glucose concentration. However, glucose was gradually consumed by K. pneumoniae in our experiments. The bacteria may change their metabolic status after glucose was depleted. To maintain K. pneumoniae in the medium with glucose, we chose 2% glucose in this study (Fig 3). In future studies, continuous fermentation to maintain glucose at a certain concentration can be performed to examine the expression of type 3 fimbriae in K. pneumoniae.

In this study, we reported that the expression of type 3 fimbriae in clinical K. pneumoniae isolates growing in the medium with glucose were various and related to the amount of etcABC mRNA. The effects of glucose on bacterial biofilm formation and colonization in the gut of worm are consistent with the expression of type 3 fimbriae, depending on etcABC, in the most of K. pneumoniae isolates.

Supporting information

S1 Raw image. The effects of glucose on K. pneumoniae STU1.

The bacteria were incubated in LB (Lane 1) or LB supplemented with glucose from 0.1% (lane 2 and 3), 0.2% (lane 4 and 5), 0.5% (lane 6 and 7), 1% (lane 8 and 9) to 2% (lane 10 and 11) as the final glucose concentration. The cellular proteins were analyzed by SDS-PABE (A and B). PC means MrkA (21 kDa) as a positive control. The protein markers (M) shows 170 kDa, 130 kDa, 95 kDa, 72 kDa, 55 kDa, 43 kDa, 34 kDa, 26 kDa. The raw images of Western blotting from chemiluminescence camera system (C and D) are inverted to (E and F) by inverting black and white.

https://doi.org/10.1371/journal.pone.0289759.s001

(PDF)

S2 Raw image. Western blotting analysis of MrkA and ManA.

The raw images of Western blotting from chemiluminescence camera system (A) are inverted to (B) by inverting black and white. (-) means LB and (+) means LB supplemented with 2% glucose. STU1: K. pneumoniae STU1. STU1/etcABC: K. pneumoniae STU1/pBSK::Gm::etcABC; Clin200 and Clin73: clinical K. pneumoniae isolates.

https://doi.org/10.1371/journal.pone.0289759.s002

(PDF)

Acknowledgments

We appreciate Miss Huei-Jen Chao and the staff in Tzu Chi Hospital, Hualien, Taiwan for assisting in the specimen collection. We are grateful to the Core Facility Center, Tzu Chi University for their support with upright fluorescence microscope.

References

  1. 1. Guerra MES, Destro G, Vieira B, Lima AS, Ferraz LFC, Hakansson AP, et al. Klebsiella pneumoniae Biofilms and Their Role in Disease Pathogenesis. Front Cell Infect Microbiol. 2022;12:877995. pmid:35646720.
  2. 2. Li Y, Sun Y, Sun S-w, Liang B, Jiang B-w, Feng N, et al. Prevalence of antimicrobial resistance and virulence genes in Klebsiella pneumoniae and Congenetic Raoultella Isolates from captive giant pandas. PLOS ONE. 2023;18(3):e0283738. pmid:36996135
  3. 3. Zadoks RN, Griffiths HM, Munoz MA, Ahlstrom C, Bennett GJ, Thomas E, et al. Sources of Klebsiella and Raoultella species on dairy farms: be careful where you walk. J Dairy Sci. 2011;94(2):1045–51. pmid:21257074.
  4. 4. Saddam S, Khan M, Jamal M, Rehman SU, Slama P, Horky P. Multidrug resistant Klebsiella Pneumoniae reservoir and their capsular resistance genes in cow farms of district Peshawar, Pakistan. PLOS ONE. 2023;18(2):e0282245. pmid:36848367
  5. 5. Lin YT, Siu LK, Lin JC, Chen TL, Tseng CP, Yeh KM, et al. Seroepidemiology of Klebsiella pneumoniae colonizing the intestinal tract of healthy Chinese and overseas Chinese adults in Asian countries. BMC Microbiol. 2012;12:13. pmid:22260182.
  6. 6. Van Hooste W, Vanrentergem M, Nulens E, Snauwaert C, De Geyter D, Mertens R, et al. Infections caused by hypervirulent Klebsiella pneumoniae in non-endemic countries: three case reports and review of the literature. Acta Clin Belg. 2022:1–5. pmid:35904343.
  7. 7. Ku YH, Chuang YC, Chen CC, Lee MF, Yang YC, Tang HJ, et al. Klebsiella pneumoniae Isolates from Meningitis: Epidemiology, Virulence and Antibiotic Resistance. Sci Rep. 2017;7(1):6634. pmid:28747788.
  8. 8. Yang J, Li Y, Tang N, Li J, Zhou J, Lu S, et al. The human gut serves as a reservoir of hypervirulent Klebsiella pneumoniae. Gut Microbes. 2022;14(1):2114739. pmid:36001493.
  9. 9. Percival SL, Suleman L, Vuotto C, Donelli G. Healthcare-associated infections, medical devices and biofilms: risk, tolerance and control. J Med Microbiol. 2015;64(Pt 4):323–34. pmid:25670813.
  10. 10. Ong CL, Beatson SA, Totsika M, Forestier C, McEwan AG, Schembri MA. Molecular analysis of type 3 fimbrial genes from Escherichia coli, Klebsiella and Citrobacter species. BMC Microbiol. 2010;10:183. pmid:20576143.
  11. 11. Allen BL, Gerlach GF, Clegg S. Nucleotide sequence and functions of mrk determinants necessary for expression of type 3 fimbriae in Klebsiella pneumoniae. J Bacteriol. 1991;173(2):916–20. pmid:1670938.
  12. 12. Paczosa MK, Mecsas J. Klebsiella pneumoniae: Going on the Offense with a Strong Defense. Microbiol Mol Biol Rev. 2016;80(3):629–61. pmid:27307579.
  13. 13. Deutscher J, Ake FM, Derkaoui M, Zebre AC, Cao TN, Bouraoui H, et al. The bacterial phosphoenolpyruvate:carbohydrate phosphotransferase system: regulation by protein phosphorylation and phosphorylation-dependent protein-protein interactions. Microbiol Mol Biol Rev. 2014;78(2):231–56. pmid:24847021.
  14. 14. Jeckelmann JM, Erni B. Carbohydrate Transport by Group Translocation: The Bacterial Phosphoenolpyruvate: Sugar Phosphotransferase System. Subcell Biochem. 2019;92:223–74. pmid:31214989.
  15. 15. Rom JS, Hart MT, McIver KS. PRD-Containing Virulence Regulators (PCVRs) in Pathogenic Bacteria. Front Cell Infect Microbiol. 2021;11:772874. pmid:34737980.
  16. 16. Deutscher J, Francke C, Postma PW. How phosphotransferase system-related protein phosphorylation regulates carbohydrate metabolism in bacteria. Microbiol Mol Biol Rev. 2006;70(4):939–1031. pmid:17158705.
  17. 17. Jeng WY, Panjaitan NSD, Horng YT, Chung WT, Chien CC, Soo PC. The Negative Effects of KPN00353 on Glycerol Kinase and Microaerobic 1,3-Propanediol Production in Klebsiella pneumoniae. Front Microbiol. 2017;8:2441. pmid:29375490.
  18. 18. Horng YT, Wang CJ, Chung WT, Chao HJ, Chen YY, Soo PC. Phosphoenolpyruvate phosphotransferase system components positively regulate Klebsiella biofilm formation. J Microbiol Immunol Infect. 2018;51(2):174–83. pmid:28716362.
  19. 19. Panjaitan NSD, Horng YT, Cheng SW, Chung WT, Soo PC. EtcABC, a Putative EII Complex, Regulates Type 3 Fimbriae via CRP-cAMP Signaling in Klebsiella pneumoniae. Front Microbiol. 2019;10:1558. pmid:31354661.
  20. 20. Panjaitan NSD, Horng YT, Chien CC, Yang HC, You RI, Soo PC. The PTS Components in Klebsiella pneumoniae Affect Bacterial Capsular Polysaccharide Production and Macrophage Phagocytosis Resistance. Microorganisms. 2021;9(2). pmid:33567595.
  21. 21. Girard LR, Fiedler TJ, Harris TW, Carvalho F, Antoshechkin I, Han M, et al. WormBook: the online review of Caenorhabditis elegans biology. Nucleic Acids Res. 2007;35(Database issue):D472–5. pmid:17099225.
  22. 22. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 1989.
  23. 23. Lin CT, Lin TH, Wu CC, Wan L, Huang CF, Peng HL. CRP-Cyclic AMP Regulates the Expression of Type 3 Fimbriae via Cyclic di-GMP in Klebsiella pneumoniae. PLoS One. 2016;11(9):e0162884. pmid:27631471.
  24. 24. Notley-McRobb L, Death A, Ferenci T. The relationship between external glucose concentration and cAMP levels inside Escherichia coli: implications for models of phosphotransferase-mediated regulation of adenylate cyclase. Microbiology (Reading). 1997;143 (Pt 6):1909–18. pmid:9202467.
  25. 25. Chen CL, Hou PC, Wang YT, Lee HY, Zhou YL, Wu TS, et al. The High mortality and antimicrobial resistance of Klebsiella pneumoniae bacteremia in northern Taiwan. J Infect Dev Ctries. 2020;14(4):373–9. pmid:32379714.