OmpK36-mediated Carbapenem resistance attenuates ST258 Klebsiella pneumoniae in vivo

Carbapenem-resistance in Klebsiella pneumoniae (KP) sequence type ST258 is mediated by carbapenemases (e.g. KPC-2) and loss or modification of the major non-selective porins OmpK35 and OmpK36. However, the mechanism underpinning OmpK36-mediated resistance and consequences of these changes on pathogenicity remain unknown. By solving the crystal structure of a clinical ST258 OmpK36 variant we provide direct structural evidence of pore constriction, mediated by a di-amino acid (Gly115-Asp116) insertion into loop 3, restricting diffusion of both nutrients (e.g. lactose) and Carbapenems. In the presence of KPC-2 this results in a 16-fold increase in MIC to Meropenem. Additionally, the Gly-Asp insertion impairs bacterial growth in lactose-containing medium and confers a significant in vivo fitness cost in a murine model of ventilator-associated pneumonia. Our data suggests that the continuous selective pressure imposed by widespread Carbapenem utilisation in hospital settings drives the expansion of KP expressing Gly-Asp insertion mutants, despite an associated fitness cost.

Remarks to the Author: Summary In this work the authors show that carbapenem resistance in the clinical Klebsiella pneumoniae (KP) ST258 strain is mediated by the presence of plasmid encoded carbapenemase genes such as KPC-2 and OXA-48 and the loss/truncation of OmpK35 and modification of OmpK36. Their studies also show that modification of these Omp proteins confers a significant fitness cost in vivo (reduced ability to replicate in lungs and disseminate to the blood compared to WT).The presence of a GD insertion in loop 3 OmpK36ST258 not seen in OmpK36WT results in a conformational change in loop 3 allowing it to extend into the pore. This is further stabilized by the formation of an Asp-Arg dyad and contributes to a reduction of the OmpK36ST258 pore diameter. This pore constriction is thought to confer carbapenem resistance. Carbapenem diffusion studies show a reduction in antibiotic diffusion when proteins were reconstituted in liposomes. In addition, the authors show that carbapenem resistance in ST258 KP requires both porin modification and the expression of carbapenemase enzymes. In the absence of either KPC-2 or OXA-48 the strains tested remained sensitive to carbapenems. Apart from a few critiques and corrections that should be addressed by the authors, this is a strong body of work.
Supplementary Figure 2 Though antibiotic diffusion is reduced in OmpK36ST258 we see an increase in fitness cost probably due to restricting the influx of nutrients etc. These porins are non-selective and allow transport of small molecules, as is seen in the experiments with glucose. Glucose is a small molecule with a molecular weight ~180 g/mol while the carbapenem ertapenem is 475.5 g/mol. Why wasn't a molecule with a molecular weight comparable to that of ertapenem used as a control to show how pore restriction affects diffusion? Moreover, this might also give insight into how a decrease in nutrient influx may affect fitness cost. Classical liposome swelling assays would give a better description of nutrient size limitations.
Correct caption. Change D to F in second line from the bottom.
Supplementary Table 5 • Fix alignment in column 3, 3rd line from the bottom • The values in the highest resolution shell for I/ of OmpK36WT+GD and OmpK36WT and the CC1/2 of OmpK36WT are low. Adjust the resolution cutoff. • Authors should be consistent with significant figures reported for resolution in the text as well as in the refinement statistics table.
Results and Discussion Figure 1C. Why is the protein band for OmpK35 not seen in the WT strain ICC8001 while OmpK36 is seen?  (Table 2B). What are the authors referring to?
Page 9, line 191: the authors discount the effects of the L4 insertions on pore reduction because the conformational change seen was away from the constriction zone. Were studies done with an OmpK36WT+LSP construct? Or did the authors just assume that because this conformational change was away from the constriction zone it did not play a role in conferring resistance? These studies should be added.
The Asp-Arg salt bridge seems to be important in stabilizing loop 3. Do the authors have any thoughts on whether disrupting this Asp-Arg dyad by altering pH for example would affect the protein structure in the OmpK36ST258 or OmpK36WT+GD? Does charge play a role in nutrient uptake, and is this changed in the mutant?
At what pH were these crystals obtained? Grammar • Page 5 line 97 of Introduction: correct font size of "usage" • Table S1, caption A: correct the spelling of carbapenem • Methods, page 1, last paragraph: correct font size Reviewer #2: Remarks to the Author: Wong et al present a study examining the role of outer membrane porin mutations in conferring carbapenem resistance in Klebsiella pneumoniae clinical isolates. The authors determine the effect of the porins in mediating clinical levels of drug resistance through analysis of isogenic strains. Examining the structure and diffusion activities of purified porin isoforms, the authors show that the variant conferring drug resistance in clinical isolates restricts the influx of carbapenem antibiotics due to a constricted pore. Sensitive tests of competitive fitness in a mouse infection model reveal fitness costs of the pore-constricted variant that lie in a middle ground between WT and complete loss of the porin. The study reveals the mechanism behind the resistance of a clinically successful and highly prevalent porin mutant. The study was carefully performed and conclusions were justified, and the findings will be of broad interest to the field.
The manuscript can be improved by addressing details outlined here: Table S2 is insufficient given the number of Klebsiella strains tested in this study, and should list each strain described in the text (for example, of the isogenic strains only ICC8001 is listed, but the table should list ICC8002, 3, etc and other engineered strains, along with the respective genotype, to aid in analysis of the paper).   Table 1 legend the acronyms should be checked for accuracy. CZA is listed as referring to ceftazidime --please verify. "CAZ" is used in the table but is not defined in the legend.
line 448, Fig. 3A legend should indicate exact time post-infection that the signal was measured.
Line 261: how was significance determined when analyzing difference in fitness between mutant and parent bacteria in blood?
Based on the structure the authors determine that the loop3 insertion is stabilized by a salt-bridge between two residues which were present in the WT porin --has the role of the salt bridge in conferring resistance/carbapenem restriction been examined?
Response: We calculated 95% confidence intervals around the mean (n=10 individual mice for each condition tested). Graphically this is plotted as the mean with 95% confidence interval error bars (as described in the legend). In addition, we provide the reader with competition ratio from each animal tested below the summary point. The inference of this method is that, according to our experimental data, we are 95% confident that the true competition value lies within these ranges. If this range traverses 50% then there is no advantage (or disadvantage) to either strain tested in the assay. The chance of the true value lying outside the 95% confidence interval range is 5%. This is equivalent to the accepted p<0.05 (i.e. 1 in 20 chance of 5%) significance cut-off. In addition, when ICC8001 is competed against ICC8004 no ICC8004 was recovered, resulting in 100% competition (100% favouring ICC8001 and 0% favouring ICC8004) with a 95% confidence interval of 100% to 100%. In these experiments, we are therefore 100% confident that the true result is 100%.
Comment 6: Based on the structure the authors determine that the loop3 insertion is stabilized by a salt-bridge between two residues which were present in the WT porin --has the role of the salt bridge in conferring resistance/carbapenem restriction been examined?