Introduction

Klebsiella pneumoniae is a gram-negative bacterium that is widely distributed in the environment such as surface water, sewage, soil, and plants, and it may also exist in the intestinal tract and nasopharynx of humans (Podschun and Ullmann 1998). CRKP is an increasingly important pathogen of hospital-acquired infections that may cause a wide range of infections, including urinary tract infections, pneumonia, bacteremia, liver abscesses, and brain abscesses, and may be life-threatening in severe infections (Paczosa and Mecsas 2016). In clinical practice, great success has been achieved in the treatment of pathogenic infections with antimicrobials, but in recent years, K. pneumoniae has rapidly developed resistance to all known antimicrobials, including carbapenems, and this phenomenon can be seen all over the world. Due to its highly transmissible nature, the emergence of carbapenem-resistant K. pneumoniae in endemic areas of Klebsiella pneumoniae may lead to a dangerous local increase in the number of carbapenem-resistant K. pneumoniae, and the emergence of carbapenem-resistant K. pneumoniae has clearly become a world public health problem that may be faced by all the human beings (Holden et al. 2018; Doi 2019; Karampatakis et al. 2016). The main resistance mechanisms of carbapenem-resistant K. pneumoniae are currently shown as follows: (1) K. pneumoniae can be prevented by encoding genes (e.g., New Delhi metallo-β-lactamase [NDM], K. pneumoniae carbapenemase [KPC], Verona integron-encoded metallo-β-lactamase [VIM], oxacillin-hydrolyzing β-lactamase [OXA], and imipenem-hydrolyzing β-lactamase [IMP]) to hydrolyze carbapenemases produced by the carbapenemase, which is an enzyme with a positive effect on the hydrolysis of carbapenems. Carbapenemases are catalytically efficient endocannabinoid enzymes that may result in elevated minimum inhibitory concentrations of carbapenem and include classes A, B, and D. The three major families of class-A serine carbapenemases include NMC/IMI, SME, and KPC enzymes; the most common families of class-B metallo-lactamases include blaNDM, blaVIM, blaIMP, blaGIM, and blaSIM enzymes; class-D serine carbapenemase includes blaOXA β-lactamase. (2) The expression of efflux pump gene allows bacterial cells to actively and rapidly squeeze out carbapenems, avoiding the bacterial cell wall from being damaged by carbapenems. (3) The permeability of the outer membrane is reduced through the production of β-lactamases (AmpC) and the alterations in the bacterial cell membrane (pore protein mutations in OmpK35 and OmpK36), thus blocking carbapenem drugs from entering into the bacteria (Alizadeh et al. 2018; Durante-Mangoni et al. 2019; Girlich et al. 2019; Queenan and Bush 2007). The synthesis of carbapenemase-encoding genes has caused numerous difficulties for clinicians, and the drugs do not work as well as they should, making it difficult to develop treatment plans for patients. The common carbapenemase-encoding genes in China are mainly KPC, NDM, and IMP n (Zhang et al. 2020a). A weak correlation exists between the expression of carbapenem antimicrobial resistance in CRKP and the virulence genes they possess, and these strains may carry multiple virulence genes that may have the ability to encode capsules (magA, wcaG), express high viscosity, mucus phenotype regulators (magA, rpmA), produce mycobacterial hair adhesion elements (fimH), synthesize lipopolysaccharides (wabG, uge, ycf), and construct iron acquisition systems (iutA, iroN, entB, KfuB), iron carriers (aerobactin), associated with allantoin metabolism (allS), yersin bacteriocin (ybtS), and other virulence factors (vatD); the functions possessed by these virulence genes allow them to overcome host defenses (Hartman et al. 2009; Yu et al. 2008, 2006, 2007; Remya et al. 2019; Candan and Aksöz, 2015). Almost all CRKPs prevalent worldwide are multi-locus sequence ty** (ST) 258 and its related variants; for example, ST11, which is widely prevalent in China, differs from ST258 in only one gene locus, but it is not yet clear why ST258 and its related variants spread so widely and successfully (Qi et al. 2011; Chen et al. 2014).

In this study, the antimicrobial susceptibility test was performed on 87 carbapenem-resistant K. pneumoniae strains. Multi-locus sequence ty**, drug resistance genes, virulence genes, and capsular serotypes were detected using PCR amplification; clonal relationship detection was performed using pulsed-field gel electrophoresis (PFGE) technique to clarify the molecular epidemiological characteristics of CRKP in this hospital.

Materials and methods

Sample collection and bacterial isolates

A total of 87 non-replicated carbapenem-insensitive K. pneumoniae strains were collected from January 2016 to December 2020 in a tertiary care hospital in Shijiazhuang, Hebei Province, China. All isolates of specimens (sputum, secretions, urine, alveolar lavage, puncture fluid, and blood) were obtained from the clinical departments of this tertiary hospital, and there was no restriction on age, sex, disease type, and clinical department of the patients who provided the specimens. Eighty-seven isolates were identified using a MALDI-TOF mass spectrometer (AUTOFLEXSpeed, Germany). The MIC values of imipenem (MIC ≥ 4 μg/mL), meropenem (MIC ≥ 4 μg/mL), and ertapenem (MIC ≥ 2 μg/mL) were determined using the microbroth dilution method with a fold reference (CLSI M100, 2021). According to the guidelines in the Clinical and Laboratory Standards Institute (CLSI) (2019, 29), the inclusion criteria for the guidelines were reduced susceptibility to at least one carbapenem (imipenem, meropenem, and ertapenem) as carbapenem resistance. All 87 isolates were stored in a − 80℃ refrigerator.

Antimicrobial sensitivity experiments

Antimicrobial susceptibility testing is conducted using the microdilution method (MIC method) according to the manufacturer ‘s instructions. To put it simply, a fresh bacterial strain is prepared into a 0.5 mcf suspension, added to a 96-well plate following the instructions, and incubated in a constant temperature incubator for 24 h. The number of negative wells for each antimicrobial is then recorded and entered into the result analysis system provided by the manufacturer to obtain the corresponding antimicrobial ‘s minimum inhibitory concentration. Finally, these figures are entered into the manufacturer ‘s result analysis system to obtain the minimum inhibitory concentration. Thirty-one antimicrobials used in antimicrobial trials included ampicillin, ampicillin-sulbactam, tetracycline, chloramphenicol, trimethoprim-sulfamethoxazole, cefazolin, cefotaxime, ceftazidime, cefoxitin, gentamicin, imipenem, Nalidixan, azithromycin, sulfisoxazole, ciprofloxacin, amoxicillin-clavulanate, cefotaxime-clavulanate, ceftazidime-clavulanate, colistin, polymyxin B, minocycline, amikacin, aztreonam, cefepime, meropenem, ertapenem levofloxacin, doxycycline, kanamycin, streptomycin, and gemifloxacin. The breakpoint of measured antimicrobial MIC values was obtained according to the corresponding standards of the American Committee for Clinical Laboratory Standardization (CLSI M100, 2021) for the corresponding sensitive (S), moderately sensitive (I), and resistant (R) results. The antimicrobial sensitivity test was performed using E.coli ATCC25922 and K. pneumoniae ATCC700603 as the reference standard.

Drug resistance gene, capsular serotype, and virulence gene detection

The polymerase chain reaction was performed using specific primer sequences with TaKaRa ‘s Premix Taq configuration system using PTC-200 and Veriti. Twelve resistance genes (blaKPC, blaNDM, blaVIM, blaOXA-48, blaOXA-1-like, blaIMP, blaGES, blaSHV, blaCTX-M-65, ompK35, ompK36, ompK37), 16 common virulence genes (WcaG, ureA, wabG, fimH, entB, iutA, ybtS, rmpA, IroN, KfuB, aerobactin, alls, uge, vatD, magA, and ycf), and seven capsular serotype genes (K1, K2, K3, K5, K20, K54, and K57) were selected (Kaczmarek et al. 2006; Dallenne et al. 2010; Zhan et al. 2017; Zhang et al. 2018). The reaction program for the total reaction system of 25 μL was set as follows: pre-denaturation temperature 94 °C for 2 min, denaturation temperature 94 °C for 20 s, annealing temperature depending on different primers (blaKPC, blaVIM, blaIMP, ompK35, ompK36, and ompK37 55 °C; blaGES and blaOXA-48 57 °C; blaOXA-1-like and uge 60 °C; blaSHV, K20, entB, and ybtS 56 °C; blaCTX-M-65, blaNDM, K1, and WcaG 54 °C; ureA, wabG, ycf, rmpA, aerobactin, IroN, KfuB, alls, K3, K5, K54, and K57 58 °C; K2 64 °C; fimH 62 °C; iutA 66 °C; vatD and magA 48 °C) (Kaczmarek et al. 2006; Dallenne et al. 2010; Zhan et al. 2017; Zhang et al. 2018), time 30 s, extension temperature 72℃, number of amplification cycles 35, final extension temperature 72℃, time 5 min. PCR products were analyzed using a fully automated electrophoresis analyzer (QIAXEL) and Sanger sequencing was performed (sent to Bei**g DynaBio), and the sequencing results were analyzed using NCBI BLAST (https://blast.ncbi.nlm.nih.gov/blast.cgi) for comparative analysis.

Detection of efflux pump gene

7500 Real-Time PCR System (Applied Biosystems) was used to detect OqxA and OqxB efflux pump genes, primers were synthesized by Bei**g DynaBio, the primer system used was TaKaRa ‘s Premix Ex TaqTM (Probe qPCR), and the amplification procedure was as follows: initial incubation at 95 °C for 2 min, followed by 40 cycles of 10 s at 95 °C, 30 s at 60 °C, and 10 s at 72 °C (Zhong et al. 2014). The strains that tested positive for DNA were extracted and reverse-transcribed into cDNA using TaKaRa ‘s PrimeScript™ II 1st Strand cDNA Synthesis Kit s. The strains were then configured using the Premix Ex Taq™ system and subjected to real-time fluorescent quantitative PCR. Amplification conditions are as described above. The 2^-ΔΔCT of each strain was calculated as the relative expression of the efflux pump gene for each strain, using the rpoB gene as the reference gene and strain number 20021762; the CRKP strain as the reference. Each sample was processed in triplicate.

Multi-locus sequence ty** (MLST)

PCR amplification of seven housekee** genes of K. pneumoniae (gapA, infB, mdh, pgi, phoE, rpoB, and tonB) was also performed using TaKaRa ‘s Premix Taq configuration system, polymerase chain reaction using PTC-200 and Veriti, and total reaction system 50 μL. The reaction program was set as follows: pre-denaturation at 94 °C for 2 min, denaturation at 94 °C for 20 s, annealing temperature depending on different primers (gapA 60 °C, TonB 45 °C, infB, mdh, pgi, phoE, and rpoB all 50 °C) for 30 s, extension at 72 °C for 30 s, number of amplification cycles 35, final extension at 72 °C for 5 min. The PCR products were analyzed using a fully automated electrophoresis analyzer (QIAXEL), and Sanger sequencing was performed (sent to Bei**g DynaScience Biotech). The sequencing results were typed for STs using the BIGSdb-Pasteur database (https://bigsdb.pasteur.fr/klebsiella/klebsiella.html).

Pulsed-field gel electrophoresis (PFGE)

PFGE assays were performed on all collected carbapenem-resistant K. pneumoniae strains. Briefly, the genomic DNA of carbapenem-resistant K. pneumoniae was prepared by embedding K. pneumoniae cells in Seakem Gold Agarose and then digesting them using XbaI enzymes at 37 °C for 4 h. The genomic DNA of carbapenem-resistant K. pneumoniae was then stained using a Bio-Rad CHEF III system (120° angle, 6 V/cm, switching time 6 s, 36 s) at 14 °C for 18.5 h. After completion of electrophoresis, DNA bands were stained using GelRed and imaged with a gel imaging system (BIO-RAD Molecular Imager® Gel DocTM XR + with Image LabTM software). Salmonella serovar Brendaup strain H9812 was used as a molecular marker. DNA profiles were interpreted using the criteria of Tenover et al. (1995) (Zhan et al. 2017). PFGE patterns were analyzed using Bionumerics software (Applied Mathematics, Sint-Martens-Latem, Belgium) using dice similarity coefficients. The isolates with a similarity of more than 80% were defined as identical PFGE clusters.

Results

Sample types and sources of strains isolated from K. pneumonia

The number of CRKPs isolated per year from 2016 to 2020 in this tertiary care hospital was 14, 5, 6, 28, and 34 strains, respectively. The vast majority of the isolated CRKP specimens were originated from sputum 78.16% (68/87), followed by burn secretions 9.20% (8/87), urine 3.45% (3/87), alveolar lavage fluid 3.45% (3/87), puncture fluid 3.45% (3/87), and blood 2.30% (2/87). The detection rate of CRKP in the lower respiratory tract was high; CRKP was also widely distributed in clinical departments of hospitals, with the majority of carbapenem-resistant K. pneumoniae coming from the respiratory medicine department 21.84% (19/87), intensive care unit 21.84% (19/87), burns department 16.09% (14/87), rehabilitation department 13.79% (12/87), geriatrics department 9.20% (8/87), neurosurgery department 4.60% (4/87), thoracic surgery department 2.30% (2/87), nephrology department 2.30% (2/87), hematology department 2.30% (2/87), general surgery department 2.30% (2/87), neurology department 1.15% (1/87), gastroenterology department 1.15% (1/87), and endocrinology department 1.15% (1/87). Respiratory and Critical Care Medicine remain the main clinical departments prevalent with carbapenem-resistant K. pneumoniae. Although there are relatively fewer carbapenem-resistant K. pneumoniae in other departments, there is a small amount of CRKP epidemic transmission.

Antibacterial drugs, carbapenem-resistant genes, and efflux pump genes

Thirty-one antimicrobials commonly used in clinical practice were selected according to the microbroth dilution method (MIC method), which was developed according to the manufacturer ‘s method. Ampicillin, ampicillin-sulbactam, ceftazidime clavulanic acid, ceftazidime clavulanic acid, cefazolin, ceftazidime, cefoxitin, cefepime, amoxicillin-clavulanic acid, gemifioxacin, levofloxacin, ciprofloxacin, nalidixic acid, meropenem, imipenem, and azithromycin are antimicrobial drugs with a 100% resistance rate. The resistance rate is 98.85% for amineptide and ertapenem. The resistance rates for gentamicin, kanamycin, sulfisoxazole, trimethoprim, amikacin, chloramphenicol, tetracycline, doxycycline, streptomycin, and minocycline were 90.80%, 89.66%, 79.31%, 77.01%, 72.41%, 71.59%, 70.11%, 66.67%, 21.84%, and 10.31%. Colistin and polymyxin B’s resistance rate is 3.45%. The specific MIC values of the 31 antimicrobials are shown in the attached information. The resistance rate of CRKP to 20 antimicrobials in 31 antimicrobial sensitivity assays exceeded 90%, with an alarming 100% resistance rate of CRKP to 17 antimicrobials. All CRKPs were multi-drug-resistant (MDR), with one strain being insensitive to β-lactams, chloramphenicol, macrolides, fluoroquinolones, aminoglycosides, sulfonamides, tetracyclines, carbapenems, and glycopolypeptide antimicrobials. The distribution of resistant genotypes among 87 carbapenem-resistant K. pneumoniae strains was blaKPC-2 type 96.55% (84/87), blaOXA-1 family group 14.94% (13/87), blaNDM-1 type 2.30% (2/87), and blaNDM-5 type 1.15% (1/87), while blaVIM, blaIMP, and blaOXA-48 resistance genes were not detected; the extended-spectrum-beta-lactamase genes blaSHV 95.40% (83/87), blaCTX-M-65 75.86% (66/87); the outer membrane genes OmpK35 100% (87/87), OmpK36 22.99% (20/87), and OmpK37 100% (87/87); and the efflux pump genes OqxA 63.22% (55/87) and OqxB 63.22% (55/87). The detailed distribution of drug resistance genes is shown in Table 1. No mutation sites were found in all detected CRKPs of OmpK35, while all detected OmpK37 genes had four mutation types of p.I70M, p.I128M, m233_None234insQ, and p.N230G. The GenBank ID of the pore protein sequence of the reference Klebsiella pneumoniae strain of OmpK37 was AJ011502. Eleven types of mutation were also detected in the OmpK36 gene, including p.F207W, p.D224E, p.L59V, p.L228V, p.N49S, p.Q227_None679del, p.L191S, p.A190_None568del, p.N218H, p. E232R, and p.A217S. The GenBank accession number of the pore protein sequence of the Klebsiella pneumoniae strain referenced by OmpK36 was Z33506.1. Detailed results of carbapenem resistance genes, outer membrane protein genes, efflux pump genes, extended-spectrum-beta-lactamase genes, and antimicrobial sensitivity test of each isolated strain are shown in Fig. 1. We selected 19 CRKP strains of KPC-2 type for mRNA expression assay of exocytosis pump gene. Expression of mRNA for the efflux pump gene was detected in all 19 CRKP strains. The relative expression of the efflux pump genes is detailed in Fig. 2.

Table 1 Distribution of drug resistance genes and virulence genes of 87 isolates
Full size table
Fig. 1
figure 1

The test results of ST type, carbapenem resistance genes, outer membrane genes, efflux pump genes, and antimicrobial susceptibility for 87 CRKP strains

Full size image
Fig. 2
figure 2

Relative expression of 19 strains of KPC-2 CRKP efflux pump genes. Note: the blue bars represent OqxA gene test results and the brown bars represent OqxB gene test results. The vertical coordinate represents the 2^-ΔΔCT of the oqxA/B gene and the horizontal coordinate represents the strain number

Full size image

Capsular serotypes and virulence genes

The results were verified by amplification of specific primers, the amplified products were verified by electrophoresis, and the products verified by electrophoresis were sent to the company for sequencing, and the sequencing results were verified by BLAST. Only K54 1.15% (1/87) of the 87 isolated carbapenem-resistant K. pneumoniae were detected in the capsular serotype, whereas K1, K2, K3, K5, K20, and K57 capsular serotypes were not detected. Virulence genes were detected in 100.00% (87/87) of WabG, 100.00% (87/87) of ycf, 100.00% (87/87) of entB, 94.25% (82/87) of urea, 91.95% (80/87) of ybtS, 88.51% (77/87) of uge, and 83.91% (73/87) of fimH. Detailed distribution of virulence genes is shown in Table 1. The most common combination of virulence genes in this study was urea-wabG-fimH-entB-ybtS-uge-ycf; the virulence gene combination in the CRKP strain with capsular serotype K54 was WcaG-urea-wabG-fimH-entB-KfuB-alls-uge-ycf. The detailed results are shown in Fig. 3.

Fig. 3
figure 3

ST type, capsular serotype, and virulence gene detection results of 87 CRKP strains

Full size image

Molecular characteristics of K. pneumoniae isolates

Among the 87 CRKP strains, five STs were identified, including ST11 93.10% (81/87), ST15 3.45% (3/87), ST273 1.15% (1/87), ST340 1.15% (1/87), and ST626 1.15% (1/87); the results of PFGE clustering analysis showed that 87 CRKP strains could be divided into five clusters: cluster A comprised of ST15 3.45% (3/87), cluster B comprised of ST11 10.34% (9/87), cluster C comprised of ST11 55.17% (48/87), cluster D comprised of ST11 4.60% (4/87), cluster E comprised of ST11 18.39% (16/87), and ST273 1.15% (1/87). The other six CRKP strains were not clustered with other strains, including ST11 4.60% (4/87), ST626 1.15% (1/87), and ST340 1.15% (1/87). The detailed results are shown in Fig. 4.

Fig. 4
figure 4

Pulse-field gel electrophoresis results of 87 strains of CRKP. Note: the red line is the dividing line with 80% similarity, the light green box represents cluster A; the sky blue box represents cluster B; the pink box represents cluster C; the light yellow box represents cluster D; the dark blue box represents the E cluster

Full size image

Discussion

The globally first case of K. pneumoniae type blaKPC was discovered in the USA in 1996. Over time, carbapenem-resistant bacteria capable of producing blaKPC have spread to most areas in the USA, Israel, and southern European countries (especially Greece and Italy), as well as to the South American continent and China (Munoz-Price et al. 2013). The first blaKPC-2 strain of K. pneumoniae was discovered in Zhejiang Province of China in 2007, and since then blaKPC-2 pneumoniae has spread rapidly to other provinces and cities such as Jiangsu (Qi et al. 2011), Anhui (Qi et al. 2011), Fujian (Chen et al. 2005). While in the USA and European countries, K. pneumoniae ST258 contributes significantly to the spread of carbapenem resistance, in the Chinese region, ST11 dominates; ST11 is a single motif variant (TonB) of ST258 and they are closely related. It was reported that ST11 and ST258 belonged to the clonal complex of CC258, which included ST11, ST258, and five other STSs (ST270, ST340, ST379, ST407, and ST418) (Qi et al. 2011; Wang et al. 2018). The CRKP isolated from this tertiary care hospital had a predominant ST11 type, and the number of ST11 type of CRKP in this tertiary care hospital showed a significant annual increase over the 5-year period from 2016 to 2020. This is consistent with the results of other Chinese researchers (Zhao et al. 2021; Hu et al. 2020). All CRKPs could be clearly divided into 5 taxa in the PFGE clustering analysis, and taxon A was full of ST15 clonal CRKP but had a low number of 3 strains compared to the results of a study in a medical center in northeastern China (Chen et al. 2021). CRKP of taxa B, C, D, and E were all dominated by ST11, but there were still some differences even among the same ST11 clonotype, which was probably because multiple incompatibilities between plasmids could lead to different resistance genes carried on the plasmids of each isolate, thus resulting in individual differences (Kim et al. 2021).

It was found in the virulence gene test results that virulence genes were more abundant in some CRKPs compared with the results of other researchers, especially the detection rate of rmpA, magA, and iutA genes was significantly higher than that of isolates from a tertiary hospital in Chongqing area (Zeng et al. 2021), but the results were closer to those of a tertiary care hospital in the Zhejiang area (Zhan et al. 2017). The rmpA, magA, and iutA virulence genes tend to be common in HVKP, which may be a sign that the CRKP within this hospital is evolving toward high virulence. The virulence plasmids which were considered non-conjugated in most previous studies are present only in HVKP. Now there has been a worldwide spread of virulence plasmids from HVKP into CRKP, in which the plasmids are increasingly associated with the spread of virulence factors, leading to the spread of important pathogenic characteristics, with CRKP possessing highly virulent plasmids being considerably more pathogenic than before (Kopotsa et al. 2020; Russo et al. 2014; Zhang et al. 2020b). Interestingly, a blaNDM-1 carbapenemase-type CRKP strain had capsular serotype K54, which was the first time in the region that a CRKP isolate was found to have capsular serotype K54; however, according to related reports, K54 was one of the supervirulent capsular serotypes of K. pneumoniae (Turton et al. 2018). The virulence gene combination of this blaNDM-1 isolate was WcaG-urea-wabG-fimH-entB-KfuB-alls-uge-ycf with two additional virulence genes of KfuB-alls compared with the virulence gene combinations of other isolates; the results of the correlation study concluded that there was a strong correlation between kfuB, alls, and K1 capsular serotype isolates, and that all K1 strains were considered positive for kfuB and alls (Yu et al. 2008). However, the podosome serotype K54 of the CRKP strain isolated in this study possessed the combination of KfuB-alls, which is a very special finding. The virulence plasmid expressing K54 serotype might be accidentally obtained during the transmission of this isolate, which meant that this isolate had both high resistance of ST11 and high virulence of K54. Subsequent studies may be conducted to verify the plasmid horizontal transfer ability of this strain and to obtain the complete genetic information of its plasmid to determine the cause of this phenomenon (Yang et al. 2021).

The resistance genotypes of CRKP collected during our study were mainly the prevalent genotype of blaKPC-2, with blaNDM-1 and blaNDM-5 resistance genotypes accounting for very few proportions. The types of outer membrane proteins were dominated by OmpK35 and OmpK37. The vast majority of isolates contained both OqxA and OqxB efflux pump genes. ST types were mainly ST11, with ST15, ST273, ST626, and ST340 accounting for a few proportions. The strains with blaKPC and blaNDM resistant genotypes showed very strong resistance to common antimicrobials, and there were also individuals with abundant virulence genes in these CRKPs, which means that some CRKP strains may gradually become novel K. pneumoniae with mixed characteristics of super-resistance and high virulence, so it is a very necessary task to timely monitor, prevent, and control CRKP epidemic disease in hospitals.