Insertion Sequence-Dependent OmpK 36 Mutation Associated Ertapenem Resistance in Clinical Klebsiella pneumoniae

Extended-spectrum β-lactamases (ESBLs) and/or AmpC enzymes combined with deficiency of porins OmpK35 and OmpK36 are important for the development of carbapenem-resistant Klebsiella pneumoniae. We characterized the clinical K. pneumoniae human isolates and investigated the effect of meropenem induction on the ompK35 and ompK36 mutation to develop carbapenem resistance from six carbapenem-susceptible ESBL-producing K. pneumoniae strains. 163 clinical K. pneumoniae isolates were grouped mostly into the ESBL + AmpC (44.2%) and ESBL (42.9%) phenotypes. The resistance rate differed between cephalosporins (52.1% for cefepime 97.5% for cefotaxime) and carbapenems (16% for meropenem 28.2% for imipenem) (P < 0.001). The ESBL group showed the lowest resistance to cefoxitin and cefepime and all carbapenems, whereas the AmpC group exhibited the lowest resistance to cefepime and the highest resistance to all carbapenems. PCR amplification identified blaTEM, blaSHV, blaCTX-M-3-like, and blaCTX-M-14-like of AmpA β-lactamase genes and blaDHA and blaCMY of AmpC β-lactamase genes. Compared to all 163 clinical isolates, the 56 carbapenem-resistant isolates carried less frequently of blaTEM, blaCTXM-14-like, and blaCTXM-3-like and more frequently of blaDHA-1 and blaCMY-2. The carbapenem-resistant isolates differed in prevalence against imipenem, ertapenem, and meropenem and lacked OmpK35 more frequently than OmpK36, but abnormal PCR amplicons were detected fewer in the Omp K35-deficient group than in the OmpK36-deficient group (32.5% vs. 68.4%, *Equal contribution. How to cite this paper: Lee, J.-J., Huang, Y.-C., Hsiao, Y., Lee, C.-H., Liu, C.-S. and Chu, C. (2018) Insertion Sequence-Dependent OmpK36 Mutation Associated Ertapenem Resistance in Clinical Klebsiella pneumoniae. Advances in Microbiology, 8, 253-269. https://doi.org/10.4236/aim.2018.84017 Received: February 20, 2018 Accepted: April 23, 2018 Published: April 27, 2018 Copyright © 2018 by authors and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/ Open Access

The mechanism of carbapenem resistance involves mutations in penicillin-binding proteins (PBPs) to prevent carbapenem binding [8].The incorporation of carbapenases as NDM-1 carbapenemase can be through plasmid transfer.
The aim of this study was to investigate the effect of low-dosage meropenem treatment on the induction of carbapenem resistance in five carbapenemsusceptible K. pneumoniae strains with ESBL production with expression of OmpK35 and OmpK36 and one isolate without Ompk35 expression.

Determination of ESBL, AmpC Enzyme and Carbapenase
ESBL isolates were identified by the differences in diameter of the inhibition zones between cefotaxime and cefotaxime/clavulanic acid or between ceftazidime and ceftazidime/clavulanic acid.Active AmpC enzyme was identified using disks with 300 µg of 3-aminophenylboronic acid and carbapenemase production was detected by the Modified Hodge Test (MHT) based on the Centers for Disease Control and Prevention guidelines [21].

Sequence Analysis
PCR products were purified using a DNA purification kit (ProTECH, Taiwan) and sequenced.DNA sequences were aligned and compared using Lasergene v7.1 software (DNASTAR Inc, USA) and the BLAST program of the National Center for Biotechnology Information.
Except that isolate KP15 lacked OMPK15 due to a deletion in the promoter region of ompK35, all isolates carried normal PCR size and protein expression of ompK35 and ompK36.These bacteria were cultured firstly in Muller-Hinton Broth (MHB) containing 0.1 μg/ml meropenem for 16 hours at 100 rpm, followed by subculture in MHB supplemented with 0.5 μg/ml, 1 μg/ml, and 2 μg/ml meropenem, subsequently.The bacterial solution was plated onto Muller-Hinton Agar (MHA), and meropenem discs were plated thereafter.Colonies in the inhibition zone were selected, and outer membrane proteins were analyzed using a previously described method [26].Proteins were separated by 12% SDS-PAGE at 100 V, 100 mA, and 400 W. Protein profiles were recorded and analyzed using ImageJ software (National Institutes of Health, USA).

Statistical Analysis
Duncan's multiple range test and SPSS software (version 18, Chicago, IL, USA) were used to analyze differences among groups.

Antimicrobial Resistance and Prevalence of AmpA and AmpC Genes in ESBL and AmpC Phenotypes
The 163 isolates were separated into four phenotypic groups: the ESBL + AmpC group (44.2%), the ESBL group (42.9%), the AmpC group (7.4%), and neither group (5.5%).The prevalence was 87.1% (142 isolates) for ESBL-producing isolates, and 50.3% (82 isolates) for AmpC isolates, with the different resistance patterns among the ESBL and AmpC phenotypes (Table 2).The resistance rate differed between cephalosporins (52.1% for cefepime -97.5% for cefotaxime) and carbapenems (16% for meropenem -28.2% for imipenem) (P < 0.001).The ESBL + AmpC group with respect to resistance revealed no differences among ceftriaxone, cefotaxime, and cefoxitin as well as lower resistance to cefepime in cephalosporins and the highest resistance for imipenem, followed by ertapenem and r meropenem in carbapenem.The ESBL group showed the lowest resistance to cefoxitin and cefepime and all carbapenems, whereas the AmpC group exhibited the lowest resistance to cefepime and the highest resistance to all carbapenems.Nine isolates of the none group differed in resistance to cephalosporin, but some were resistant to carbapenems.

Characterization of the 56 Carbapenem Resistant Isolates
Of the 56 carbapenem-resistant isolates, 42.9% of the isolates were resistance to all three carbapenems and one carbapenem, respectively, while 14.3% isolates were resistant to two carbapenems (Table 3).The highest carbapenem resistance was observed in the ESBL + AmpC group (71.4%), which exhibited the highest prevalence in the isolates with single resistance to imipenem (87.5%, ertapenem (75.0%), and the AmpC group (14.3%), which showed the highest rate of resistance to all three carbapenems (25%).Although no PCR products were detected  for bla VIM , bla IPM-1 -like, bla IPM-2 -like, bla SPM , bla SIM , bla GIM , and bla KHM , of the AmpB β-lactamase genes bla KPC were identified in three isolates (Supplementary Figure 1).

Meropenem-Induced Carbapenem Resistance Is Associated with the Insertion Sequence and Antimicrobial Variations
Although no isolates altered antimicrobial resistance genes, the MIC value increased the highest against ertapenem, followed by meropenem and imipenem.
Exception of KP08-15 with IS5 insertion in 5'-end ompK36, all KP08 derivatives were resistant to ertapenem and exhibited reduced susceptibility to meropenem and imipenem (Table 5).Such antimicrobial susceptibility was also confirmed by the disc diffusion method.For the ompK36 mutation-associated change in susceptibility to cephalosporins, all mutant isolates exhibited a reduced inhibition zone compared to the parental isolates.Importantly, isolates KP08-2 and KP15-4 developed resistance to cefepime, Instead, KP08-1, KP08-12, KP08-15 and KP16-22 developed an intermediate phenotype.

Discussion
The prevalence of ESBL-producing K. pneumoniae differs among regions and sources in Taiwan: increasing from 3.4% -10.3% in 1995 to 11.3% in 2000 [27] [28], 20.9% in 2007 [29], 40.5% in 2010-2012 [30], 69.7% in 2006-2007 [31] and 87.1% in the current study (Table 2).The addition of 3-aminophenylboronic acid can enhance the accuracy to determine the ESBL-producing and the AmpC phenotypes [32] [33].In our study, 50.7% of ESBL-producing isolates carried AmpC enzymes, which is higher than that reported previously (35.3%) (Table 2) [30] Furthermore, the prevalence differed between the carbapenem-resistant isolates and the total clinical isolates, with a low prevalence of AmpA genes and a high prevalence of AmpC genes (especially bla DHA-1 , 91.1%) in the carbapenem-resistant isolates (Table 2 and Table 3).For the bla DHA and bla CMY genes of the AmpC enzyme, bla CMY can be located on a chromosome with a nonsense mutation or in a plasmid to reduce the susceptibility to third-generation cephalosporins [34], while bla DHA can increase resistance to cefoxitin (Table 4) [35] [36].In our study, ESBL-producing isolates and AmpC enzyme-producing isolates exhibited different resistance patterns with higher resistance rate to ceftriaxone and cefepime vs. higher resistance rates to ceftazidime and cefoxitin, respectively (Table 2).Regardless of whether the isolate was ESBL-or non-ESBL-producing, the carbapenem-resistant isolates carried bla DHA-1 and an insertion in both ompK35 and ompK36 genes, exhibiting the highest MIC to carbapenems (Table 4).These results indicate that porin loss and the presence of bla DHA-1 are important for carbapenem resistance.However, we also observed carbapenem-resistant isolates without the ESBL and AmpC phenotypes, indicating the involvement of other mechanisms for carbapenem resistance.
In the present study, only three isolates carried bla KPC .Furthermore, carbapenem resistance can be developed by isolates with the ESBL-type bla SHV or bla CTX-M and/or AmpC enzymes in combination with deficiency of outer membrane proteins OmpK35 and OmpK36 [42] [43].Here, we observed one isolate with bla SHV and bla CTX-M14-like with no OmpK35 and OmpK36 expression, 13 isolates with bla SHV and bla CTX-M14-like with OmpK35 loss, and two isolates with bla SHV and bla CTX-M3-like with OmpK35 loss.In K. pneumoniae, porins OmpK35 and OmpK36 play multiple roles in the development of antibiotic resistance and virulence.Isolates with deficiency in both OmpK35 and OmpK36 showed a significant decrease in virulence (i.e., a slower growth rate), an increase of susceptibility to neutrophil phagocytosis [44] and a significant decrease in proinflammatory cytokine levels [45].Further, the loss of OmpK35 is associated with resistance to lipophilic (benzylpenicillin) and large (cefepime) compounds [46].In contrast to ompK35 deletion, ompK36 deletion caused an increased resistance to cefoxitin and cefepine associated with bla gene [47].Additionally, the deletion of both ompK35 and ompK36 led to 8-Advances in Microbiology and 16-fold increases in the MICs of meropenem and cefepime, respectively.
Although imipenem resistance is not directly associated with porin OmpK35/36 loss combined with ESBL production [48].ertapenem resistance can be acquired by the loss of OmpK35/36 alone or the loss of a single porin combined with bla CTX-M-15 or bla DHA-1 expression [47].
Altered susceptibility to carbapenem is associated with carbapenem type, insertion sequence type, insertion location, and the MIC of the parental strain (Table 5).A previous study reported that the insertion sequences IS1, IS5, IS10, and IS903 insert into the promoter and coding regions of ompK35 and ompK36 (Supplementary Table 1).Imipenem-resistant isolates containing bla SHV-12 and bla DHA-1 are obtained from IS5 insertion in ompK36 from Korea and Taiwan [15] [49].IS26, IS5, IS903, and IS1 insertion in ompK36 increases resistance to cefoxitin [50].Following meropenem induction, all isolates with an insertion sequence in ompK36 exhibited reduced susceptibility to imipenem (2-to 7.9-fold increase) and meropenem (16-to 43.5-fold increase) and increased resistance to ertapenem (42.6-to 173.9-fold increase) (Table 5), suggesting that only OmpK36 is responsible for ertapenem resistance due to its small molecular weight.Despite the presence of IS1, IS5, and IS903 in all isolates, only KP08, KP15, and KP16 developed ertapenem resistance from a single insertion sequence in each ertapenem-resistant strain.These results imply a strain-dependent activation of the insertion sequence.Furthermore, the MIC levels of ertapenem and the inhibition zone of cephalosporins are associated with the insertion type and site.

Conclusion
In clinical isolates, ESBL or AmpC-producing isolates associated with carbapenem resistance were more common with deficiency in OmpK35, not OmpK36.
The isolate dependent IS1, IS5, and IS903 were able to insert into ompK36 to cause resistance to ertapenem and reduced susceptibility to imipenem and carbapenem.

Table 1 .
Primer sequences and PCR conditions.
a TA: annealing temperature.

Table 2 .
Resistance to cephalosporins and carbapenems and AmpA and AmpC genes in different ESBL and AmpC pnenotypes of clinical K. pneumonia.
a-dDifferent letters indicate significant difference between different ESBL and AmpC types, and x-z Different letters indicate significant difference between antibiotics in each ESBL and AmpC phenotypes, P < 0.05.

Table 4 .
Characteristic of ESBL-and non-ESBL-producing clinical isolates with insertion in ompK35 and ompK36.

Table 5 .
Insertion in OmpK36MIC and diameter of inhibition zone of carbapenem in clinical and carbapenem-induced resistance isolates.