ABSTRACT
We show that a previously described Klebsiella pneumoniae variant that is resistant to ceftazidime-avibactam plus meropenem-vaborbactam, has a ramR plus ompK36 mutation, and produces the V239G variant KPC-3 (V240G per the standard numbering system) exhibits resistance to ceftazidime-avibactam plus aztreonam and imipenem-relebactam but not cefepime-taniborbactam. The V239G variant does not generate collateral β-lactam susceptibility like many KPC-3 variants associated with ceftazidime-avibactam resistance. Additional mutation of ompK35 and production of the OXA-48-like carbapenemase OXA-232 were required to confer cefepime-taniborbactam resistance.
KEYWORDS: KPC, carbapenemase
TEXT
Aztreonam-avibactam (AZT/AVI) is a β-lactam–β-lactamase inhibitor combination currently in clinical trials which has activity against Enterobacterales producing metallocarbapenemases and those with AZT-hydrolyzing enzymes, such as plasmid-mediated AmpCs (pAmpCs), extended-spectrum β-lactamases (ESBLs), and the serine carbapenemase KPC. All these enzymes are increasingly carried in Klebsiella pneumoniae, and yet few studies have been performed to consider mechanisms of AZT/AVI resistance in this species. It was recently reported that among 8,787 Enterobacterales isolates, 17 were AZT/AVI resistant. Of these, three Klebsiella spp. were identified. Production of the pAmpC DHA-1 plus acrA efflux pump gene overexpression and mutation of ompK35 or ompK36 porins were identified in two resistant isolates. The other produced the ESBL PER-2 and carried an ompK35 loss-of-function mutation (1). In one in vitro study, selecting AZT/AVI resistance identified mutations in the pAmpC CMY-16 in a K. pneumoniae strain (2). AVI is currently in clinical use partnered with ceftazidime (CAZ/AVI), and here, mutations in KPC are known to confer resistance. However, such mutations tend to reduce hydrolytic activity to β-lactams other than CAZ, including carbapenems and AZT (3–6). Accordingly, it is conceivable that such mutant KPC enzymes might not confer AZT/AVI resistance.
Another recently licensed β-lactam–β-lactamase inhibitor combination is imipenem-relebactam (IMI/REL). Unlike AZT/AVI, this combination does not have efficacy against isolates producing metallocarbapenemases but is generally efficacious against Enterobacterales producing pAmpC, KPC, and ESBLs (7). Again, analysis of clinical isolates shows that IMI/REL resistance in K. pneumoniae is rare, but resistant isolates have mutations in or reduced expression of ompK35 and/or ompK36 porin genes and/or increased acrA efflux pump gene expression, alongside ESBL production (8). Similar impacts of porin and efflux pump production on IMI/REL susceptibility have been seen in in vitro studies using KPC-producing isolates (9).
Given seeming overlaps between AZT/AVI and IMI/REL resistance mechanisms in K. pneumoniae, we set out to dissect the mechanisms contributing to resistance to each in K. pneumoniae using a bank of clinical isolates and targeted recombinants having fully defined genotypes. Table 1 reports MICs (determined using CLSI broth microdilution methods [10, 11]) of AZT/AVI and IMI/REL against a collection of clinical isolates which have been described (12) and whose β-lactam resistance genotypes have been characterized (13). All isolates, whether producing carbapenemases of class A (KPC-3), B (NDM-1), or D (OXA-232) were AZT/AVI susceptible, but the NDM-1/OXA-232 producer KP4 was, as expected, IMI/REL resistant, as was the OXA-232 producer KP11, though with a lower MIC (Table 1). Notably, KP4 and KP11 have ramR mutations (12), which lead to overproduction of the AcrAB-TolC efflux pump, and reduced production of the OmpK35 porin in K. pneumoniae (14). Nonetheless, a ramR mutant clinical isolate producing KPC-3, KP30, was susceptible to both AZT/AVI and IMI/REL (Table 1), so we conclude that modulating production of these permeability-associated proteins is not sufficient to confer resistance to either β-lactam–β-lactamase inhibitor combination in a KPC-3-positive background.
TABLE 1.
MICs of aztreonam or imipenem with or without avibactam or relebactam for K. pneumoniae clinical isolates and derivatives of isolates KP21 and KP47
| Isolate (relevant genotype) | MIC (μg mL−1)a |
|||
|---|---|---|---|---|
| AZT | AZT/AVI | IMI | IMI/REL | |
| KP31 (wild type) | ≤0.5 | ≤0.5 | ≤0.5 | ≤0.5 |
| KP21 (ramR TEM-1) | ≤0.5 | ≤0.5 | ≤0.5 | ≤0.5 |
| KP11 (ramR OXA-232 CTX-M-15 TEM-1) | >128 | ≤0.5 | 4 | 4 |
| KP30 (ramR ompK35 KPC-3 TEM-1) | >128 | 1 | >128 | 1 |
| KP4 (ramR NDM-1 OXA-232 CTX-M-15 TEM-1) | >128 | ≤0.5 | 64 | 16 |
| KP21(ramR) pUBYT | 0.5 | ≤0.5 | 0.5 | ≤0.5 |
| KP21(ramR) pKPC-3 | >128 | 1 | 64 | 2 |
| KP21(ramR) pKPC-3 D178Y | 1 | ≤0.5 | 1 | 0.5 |
| KP21(ramR) pKPC-3 V239G | >128 | 2 | 32 | 2 |
| KP21(ramR) pUBYT pOXA-232 | 0.5 | ≤0.5 | 2 | 1 |
| KP21(ramR) pKPC-3 pOXA-232 | >128 | 2 | 128 | 2 |
| KP21(ramR) pKPC-3 D178Y pOXA-232 | 16 | 2 | 2 | 2 |
| KP21(ramR) pKPC-3 V239G pOXA-232 | >128 | 4 | 32 | 8 |
| KP21(ramR) ompK36 pUBYT | 0.5 | 1 | 2 | 0.5 |
| KP21(ramR) ompK36 pKPC-3 | >128 | 2 | 128 | 2 |
| KP21(ramR) ompK36 pKPC-3 D178Y | 8 | 2 | 1 | 0.5 |
| KP21(ramR) ompK36 pKPC-3 V239G | >128 | 16 | 128 | 4 |
| KP21(ramR) ompK36 pUBYT pOXA-232 | 1 | 1 | 4 | 2 |
| KP21(ramR) ompK36 pKPC-3 pOXA-232 | >128 | 2 | >128 | 32 |
| KP21(ramR) ompK36 pKPC-3 D178Y pOXA-232 | 8 | 4 | 8 | 2 |
| KP21(ramR) ompK36 pKPC-3 V239G pOXA-232 | >128 | 16 | >128 | 32 |
| KP47 ompK36 pUBYT | 0.5 | ≤0.5 | 0.5 | 0.5 |
| KP47 ompK36 pKPC-3 | >128 | 1 | >128 | 8 |
| KP47 ompK36 pKPC-3 D178Y | 16 | 2 | 1 | 0.5 |
| KP47 ompK36 pKPC-3 V239G | >128 | 4 | >128 | 16 |
Shading indicates resistance based on CLSI breakpoints (11).
To investigate the effect of blaKPC-3 mutations known to be associated with CAZ/AVI resistance (15) on AZT/AVI and IMI/REL susceptibility, we utilized K. pneumoniae clinical isolate KP21, which is a ramR mutant and fully susceptible to AZT and IMI (Table 1). We introduced blaKPC-3 on a plasmid (pKPC-3), either wild type or following site-directed mutagenesis, to create the D178Y or V239G amino acid substitution previously associated with CAZ/AVI resistance (15). Here, we use numbering based on the KPC-3 amino acid sequence (16); these substitutions are frequently referred to in the literature as D179Y and V240G using a standardized numbering system for class A β-lactamases (15). The construction of these variant KPC-3 plasmids has been reported previously (17). Reduced MICs of AZT and IMI were observed for KP21 carrying the D178Y variant, compared with KP21 carrying wild-type KPC-3 (Table 1). This phenomenon of reduced spectrum of β-lactamase activity has been described for other blaKPC-3 mutants associated with CAZ/AVI resistance (3–6). However, in a KP21 background, this reduction in activity was seen to a lesser extent when the KPC-3 V239G variant was present (Table 1). This observation fits with previous reports that K. pneumoniae strains carrying the V239G mutant blaKPC-3 remain meropenem resistant, while those carrying the D178Y mutant are meropenem susceptible (15, 17). However, AZT/AVI and IMI/REL MICs were not greatly elevated for KP21 carrying pKPC-3 V239G in comparison with KP21 carrying pKPC-3, and all these KP21 recombinants remained AZT/AVI and IMI/REL susceptible (Table 1). We conclude, therefore, that mutating blaKPC-3 in a way that gives CAZ/AVI resistance is not sufficient to give AZT/AVI or IMI/REL resistance, even in a ramR mutant K. pneumoniae background.
Addition of an OXA-232 (class D carbapenemase) plasmid (pOXA-232, as described in our previous work [17]) to KP21 carrying pKPC-3 D178Y or pKPC-3 did not confer IMI/REL resistance (Table 1). This is because, despite the fact that REL does not notably inhibit class D β-lactamases (7), pOXA-232 does not confer IMI resistance, even in the absence of REL in KP21 (Table 1). Hence, if KPC-3 is inhibited by REL, OXA-232 cannot confer IMI resistance in KP21 alone (Table 1). However, importantly, adding pOXA-232 to the KP21 recombinant carrying pKPC-3 V239G conferred IMI/REL (but not AZT/AVI) resistance, showing that even the weakly expressed imipenemase OXA-232 can act in synergy with the partially inhibited KPC-3 V239G variant and together they can confer IMI/REL resistance.
Disruption of the ompK36 porin gene in KP21 (as described previously [17]) conferred AZT/AVI and IMI/REL resistance when the recombinant was carrying pKPC-3 V239G but not when it carried pKPC-3 D178Y or pKPC-3. Addition of pOXA-232 to the KP21 ompK36 recombinants further raised IMI/REL MICs against the pKPC-3 V239G recombinant and conferred IMI/REL resistance by acting in synergy with wild-type KPC-3 in the recombinant carrying pKPC-3 but not pKPC-3 D178Y (Table 1). Using the ramR wild-type isolate KP47, engineered to have an ompK36 mutation, we confirmed that ramR mutation is essential for the AZT/AVI (but not IMI/REL) resistance seen in KP21 ompK36 pKPC-3 V239G (Table 1).
We therefore conclude that three steps—mutation of ramR, mutation of ompK36, and carriage of the V239G variant of blaKPC-3—are sufficient for K. pneumoniae to become resistant to both AZT/AVI and IMI/REL. However, prior to clinical approval of AZT/AVI, this combination is usually created clinically by adding AZT to CAZ/AVI therapy. A checkerboard assay confirmed that the AZT/AVI- and IMI/REL-resistant derivative KP21[ramR] ompK36 pKPC-3 V239G is also resistant to CAZ/AVI plus AZT, with MICs of CAZ (>32 μg mL−1) and AZT (16 μg mL−1) for this recombinant (Fig. 1).
FIG 1.
Checkerboard assays for CAZ and AZT in the presence of AVI against K. pneumoniae KP21[ramR] ompK36 producing KPC-3 V239G. The image represents duplicate assays for an 8-by-8 array of wells in a 96-well plate. All wells contained cation-adjusted Mueller-Hinton broth (CA-MHB) containing avibactam (4 μg mL−1). A serial dilution of aztreonam (AZT, x axis) and ceftazidime (CAZ, y axis) was created from 32 μg mL−1 in each plate, as recorded. All wells were inoculated with a suspension of bacteria, made as per CLSI microtiter MIC guidelines (10), and the plate was incubated at 37°C for 20 h. Growth was recorded by measuring optical density at 600 nm (OD600), and growth above background (broth) is recorded as a yellow block, while absence of growth is recorded as a white block. Growth represented by the red-edged block indicates resistance to both AZT and CAZ.
We showed previously that this combination of ramR and ompK36 mutation coupled with acquisition of pKPC-3 V239G also gives resistance to CAZ/AVI and another licensed β-lactam–β-lactamase inhibitor combination, meropenem-vaborbactam (17). Finally, therefore, we tested cefepime-taniborbactam, a combination in late-stage clinical trials (18). Notably, in the KP21[ramR] ompK36 background, pKPC-3 D178Y supported lower cefepime MICs than pKPC-3 and pKPC-3 V239G (Table 2), as seen for the other β-lactams (Table 1), and this was also true for cefepime-taniborbactam MICs (Table 2). In contrast, pKPC-3 V239G supported the same cefepime-taniborbactam MIC as pKPC-3 in KP21[ramR] ompK36, 8 μg mL−1, which is one doubling dilution below the cefepime resistance breakpoint (11) (Table 2). Further addition of pOXA-232 elevated cefepime MICs against the KP21[ramR] ompK36 pKPC-3 D178Y recombinant (Table 2), as expected since OXA enzymes are known to hydrolyze cefepime (19). Even without OXA-232, cefepime MICs for KP21[ramR] ompK36 pKPC-3 KPC-3 and pKPC-3 V239G were >256 μg mL−1, so any additional effect of OXA-232 could not be measured. Nonetheless, cefepime-taniborbactam MICs remained at ≤8 μg mL−1 for all KP21[ramR] ompK36 recombinants, indicating successful inhibition of OXA-232 (Table 2). However, additional insertional inactivation of the porin gene ompK35 (performed as described previously [17]) pushed the cefepime-taniborbactam MIC for the KP21[ramR] ompK36 recombinant carrying pOXA-232 and pKPC-3 V239G (but not pKPC-3 or pKPC-3 D178Y) to 16 μg mL−1, which is classed as cefepime resistant (Table 2).
TABLE 2.
MICs of cefepime/taniborbactam against derivatives of K. pneumoniae clinical isolate KP21
| Isolate | MIC (μg mL−1)a |
|
|---|---|---|
| Cefepime | Cefepime/taniborbactam | |
| KP21(ramR) ompK36 pUBYT | 8 | 1 |
| KP21(ramR) ompK36 pKPC-3 | >256 | 8 |
| KP21(ramR) ompK36 pKPC-3 D178Y | 16 | 2 |
| KP21(ramR) ompK36 pKPC-3 V239G | >256 | 8 |
| KP21(ramR) ompK36 pUBYT pOXA-232 | 8 | 1 |
| KP21(ramR) ompK36 pKPC-3 pOXA-232 | >256 | 8 |
| KP21(ramR) ompK36 pKPC-3 D178Y pOXA-232 | 64 | 1 |
| KP21(ramR) ompK36 pKPC-3 V239G pOXA-232 | >256 | 8 |
| KP21(ramR) ompK36 ompK35 pUBYT | 8 | 1 |
| KP21(ramR) ompK36 ompK35 pKPC-3 | >256 | 8 |
| KP21(ramR) ompK36 ompK35 pKPC-3 D178Y | 16 | 2 |
| KP21(ramR) ompK36 ompK35 pKPC-3 V239G | >256 | 8 |
| KP21(ramR) ompK36 ompK35 pUBYT pOXA-232 | 8 | 2 |
| KP21(ramR) ompK36 ompK35 pKPC-3 pOXA-232 | >256 | 8 |
| KP21(ramR) ompK36 ompK35 pKPC-3 D178Y pOXA-232 | 64 | 2 |
| KP21(ramR) ompK36 ompK35 pKPC-3 V239G pOXA-232 | >256 | 16 |
Shading indicates resistance based on CLSI breakpoints (11).
We conclude, therefore, that while three events (ramR, ompK36, and blaKPC-3 V239G) are sufficient to cause CAZ/AVI/AZT, IMI/REL and, as previously shown, meropenem-vaborbactam resistance in K. pneumoniae, additional events are required to give cefepime-taniborbactam resistance. Furthermore, while many blaKPC-3 mutations leading to CAZ/AVI resistance do come with the collateral effect of increased susceptibility to carbapenems, late-generation cephalosporins, and AZT, KPC-3 V239G does not suffer from this effect to the same degree. This explains why KPC-3 V239G, rather than KPC-3 D178Y, which does suffer from collateral increased susceptibility, is able to confer resistance to multiple β-lactam–β-lactamase inhibitor combinations, provided that their accumulation is slowed. The biochemical basis of the observation that KPC-3 V239G does not behave like KPC-3 D178Y, which has an activity biased toward ceftazidime (20), requires clarification. Nonetheless, the emergence of this blaKPC-3 V239G variant should be watched with caution.
ACKNOWLEDGMENTS
This work was funded by grant MR/S004769/1 to M.B.A. from the Antimicrobial Resistance Cross Council Initiative supported by the seven United Kingdom research councils and the National Institute for Health Research. N.S. received a postgraduate scholarship from the University of Bristol.
We declare no conflicts of interest.
REFERENCES
- 1.Mendes RE, Doyle TB, Streit JM, Arhin FF, Sader HS, Castanheira M. 2021. Investigation of mechanisms responsible for decreased susceptibility of aztreonam/avibactam activity in clinical isolates of Enterobacterales collected in Europe, Asia and Latin America in 2019. J Antimicrob Chemother. 10.1093/jac/dkab279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Niu S, Wei J, Zou C, Chavda KD, Lv J, Zhang H, Du H, Tang YW, Pitout JDD, Bonomo RA, Kreiswirth BN, Chen L. 2020. In vitro selection of aztreonam/avibactam resistance in dual-carbapenemase-producing Klebsiella pneumoniae. J Antimicrob Chemother 75:559–565. 10.1093/jac/dkz468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Poirel L, Vuillemin X, Juhas M, Masseron A, Bechtel-Grosch U, Tiziani S, Mancini S, Nordmann P. 2020. KPC-50 confers resistance to ceftazidime-avibactam associated with reduced carbapenemase activity. Antimicrob Agents Chemother 64:e00321-20. 10.1128/AAC.00321-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hobson CA, Bonacorsi S, Jacquier H, Choudhury A, Magnan M, Cointe A, Bercot B, Tenaillon O, Birgy A. 2020. KPC beta-lactamases are permissive to insertions and deletions conferring substrate spectrum modifications and resistance to ceftazidime-avibactam. Antimicrob Agents Chemother 64:e01175-20. 10.1128/AAC.01175-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Qin J, Feng Y, Lü X, Zong Z. 2021. KPC-12 with a L169M substitution in the Ω loop has reduced carbapenemase activity. Eur J Clin Microbiol Infect Dis 40:1761–1766. 10.1007/s10096-021-04198-2. [DOI] [PubMed] [Google Scholar]
- 6.Findlay J, Poirel L, Juhas M, Nordmann P. 2021. KPC-mediated resistance to ceftazidime-avibactam and collateral effects in Klebsiella pneumoniae. Antimicrob Agents Chemother 65:e00890-21. 10.1128/AAC.00890-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bush K, Bradford PA. 2019. Interplay between β-lactamases and new β-lactamase inhibitors. Nat Rev Microbiol 17:295–306. 10.1038/s41579-019-0159-8. [DOI] [PubMed] [Google Scholar]
- 8.Castanheira M, Doyle TB, Deshpande LM, Mendes RE, Sader HS. 2021. Activity of ceftazidime/avibactam, meropenem/vaborbactam and imipenem/relebactam against carbapenemase-negative carbapenem-resistant Enterobacterales isolates from US hospitals. Int J Antimicrob Agents 58:106439. 10.1016/j.ijantimicag.2021.106439. [DOI] [PubMed] [Google Scholar]
- 9.Balabanian G, Rose M, Manning N, Landman D, Quale J. 2018. Effect of porins and blaKPC expression on activity of imipenem with relebactam in Klebsiella pneumoniae: can antibiotic combinations overcome resistance? Microb Drug Resist 24:877–881. 10.1089/mdr.2018.0065. [DOI] [PubMed] [Google Scholar]
- 10.Clinical and Laboratory Standards Institute. 2015. M07-A10. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard, 10th ed. Clinical and Laboratory Standards Institute, Wayne, PA. [Google Scholar]
- 11.Clinical and Laboratory Standards Institute. 2021. M100-S31. Performance standards for antimicrobial susceptibility testing; thirtieth informational supplement. An informational supplement for global application developed through the Clinical and Laboratory Standards Institute consensus process. Clinical and Laboratory Standards Institute, Wayne, PA. [Google Scholar]
- 12.Wan Nur Ismah WAK, Takebayashi Y, Findlay J, Heesom KJ, Jiménez-Castellanos JC, Zhang J, Graham L, Bowker K, Williams OM, MacGowan AP, Avison MB. 2018. Prediction of fluoroquinolone susceptibility directly from whole-genome sequence data by using liquid chromatography-tandem mass spectrometry to identify mutant genotypes. Antimicrob Agents Chemother 62:e01814-17. 10.1128/AAC.01814-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Takebayashi Y, Wan Nur Ismah WAK, Findlay J, Heesom KJ, Zhang J, Williams OM, MacGowan AP, Avison MB. 2017. Prediction of cephalosporin and carbapenem susceptibility in multi-drug resistant Gram-negative bacteria using liquid chromatography-tandem mass spectrometry. BioRxiv. 10.1101/138594. [DOI]
- 14.Jiménez-Castellanos JC, Wan Nur Ismah WAK, Takebayashi Y, Findlay J, Schneiders T, Heesom KJ, Avison MB. 2018. Envelope proteome changes driven by RamA overproduction in Klebsiella pneumoniae that enhance acquired β-lactam resistance. J Antimicrob Chemother 73:88–94. 10.1093/jac/dkx345. [DOI] [PubMed] [Google Scholar]
- 15.Shields RK, Chen L, Cheng S, Chavda KD, Press EG, Snyder A, Pandey R, Doi Y, Kreiswirth BN, Nguyen MH, Clancy CJ. 2017. Emergence of ceftazidime-avibactam resistance due to plasmid-borne blaKPC-3 mutations during treatment of carbapenem-resistant Klebsiella pneumoniae infections. Antimicrob Agents Chemother 61:e02097-16. 10.1128/AAC.02097-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Woodford N, Tierno PM, Jr, Young K, Tysall L, Palepou MF, Ward E, Painter RE, Suber DF, Shungu D, Silver LL, Inglima K, Kornblum J, Livermore DM. 2004. Outbreak of Klebsiella pneumoniae producing a new carbapenem-hydrolyzing class A β-lactamase, KPC-3, in a New York medical center. Antimicrob Agents Chemother 48:4793–4799. 10.1128/AAC.48.12.4793-4799.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Dulyayangkul P, Douglas EJA, Lastovka F, Avison MB. 2020. Resistance to ceftazidime/avibactam plus meropenem/vaborbactam when both are used together is achieved in four steps in metallo-β-lactamase-negative Klebsiella pneumoniae. Antimicrob Agents Chemother 64:e00409-20. 10.1128/AAC.00409-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Hamrick JC, Docquier JD, Uehara T, Myers CL, Six DA, Chatwin CL, John KJ, Vernacchio SF, Cusick SM, Trout REL, Pozzi C, De Luca F, Benvenuti M, Mangani S, Liu B, Jackson RW, Moeck G, Xerri L, Burns CJ, Pevear DC, Daigle DM. 2020. VNRX-5133 (taniborbactam), a broad-spectrum inhibitor of serine- and metallo-β-lactamases, restores activity of cefepime in Enterobacterales and Pseudomonas aeruginosa. Antimicrob Agents Chemother 64:e01963-19. 10.1128/AAC.01963-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Torres E, López-Cerero L, Rodríguez-Martínez JM, Pascual Á. 2016. Reduced susceptibility to cefepime in clinical isolates of Enterobacteriaceae producing OXA-1 beta-lactamase. Microb Drug Resist 22:141–146. 10.1089/mdr.2015.0122. [DOI] [PubMed] [Google Scholar]
- 20.Compain F, Arthur M. 2017. Impaired inhibition by avibactam and resistance to the ceftazidime-avibactam combination due to the D179Y substitution in the KPC-2 β-lactamase. Antimicrob Agents Chemother 61:e00451-17. 10.1128/AAC.00451-17. [DOI] [PMC free article] [PubMed] [Google Scholar]