The Role of OmpK35,OmpK36 Porins,and Production of β-Lactamases on Imipenem Susceptibility in Klebsiella pneumoniae Clinical Isolates,Cairo,Egypt

Abstract

Background: OmpK35 and OmpK36 are the major outer membrane porins of Klebsiella pneumoniae. We aimed to study the effect of combined porin loss and production of extended-spectrum β-lactamases (ESBLs) on imipenem susceptibility among K. pneumoniae clinical isolates. Materials and Methods: This study included 91 suspected ESBL-producing K. pneumoniae clinical isolates, isolated from different patient specimens at the Cairo University hospital from January to June 2010. All isolates were subjected to genotypic analysis of the outer membrane protein gene expression using reverse transcription-PCR (RT-PCR) and analysis of OmpK35/36 of 38 isolates by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). Results: By RT-PCR, loss of Omp35 was detected in 78 (85.7%) isolates, loss of Omp36 was detected in 64 (70.32%), and loss of both porins was detected in 62 (68.1%). Out of 91 isolates, 45 (49.5%) were resistant to cefoxitin, and 17 (18.7%) were confirmed as derepressed AmpC producers. Omp35 was lost in all FOX-resistant isolates, whereas Omp36 was lost in 42 (93.3%) (p-value 0.002). The mean of ceftazidime inhibition zone diameter was significantly decreased among ESBL-producing isolates with loss of Omp35/36 (p-value 0.041 and 0.006), respectively. The mean of imipenem minimal inhibitory concentration (MIC) was markedly increased to 8.55 μg/ml among AmpC-producing isolates with Omp35/36 loss, while the mean of imipenem MIC among the 66 confirmed ESBL producers was 0.32 μg/ml. Conclusion: Imipenem MIC was markedly increased among K. pneumoniae isolates showing AmpC production with loss of both porins OmpK35/36. Meanwhile, the association of porin OmpK35/36 loss with ESBL production was not a direct cause of resistance to imipenem.

Introduction

The therapeutic options for treatment of infections caused by extremely drug-resistant (XDR) Klebsiella pneumoniae have become limited as a result of progressive accumulation of different mechanisms of resistance, including high-level production of extended-spectrum β-lactamases (ESBLs), overproduction of efflux pumps, and porin loss.^7,8 K. pneumoniae has three well-known porins in the outer membrane: OmpK35, -36, and-37, OmpK37 is a small porin linked to OmpN of Escherichia coli and is not normally expressed.¹¹ OmpK35 and OmpK36 play a major role in the passage of antibiotics into the cell because loss of the outer membrane proteins may lead (reword this) to resistance to β-lactams, for example, ceftazidime and cefoxitine.⁵ Susceptibility to carbapenams decreases in strains producing Ambler group A, B, C, or D β-lactamases in association with porin loss.¹⁰ The aim of this study was to detect the effect of combined porin loss and ESBL production on resistance of K. pneumoniae to imipenem.

Materials and Methods

The study was conducted in the Central Clinical Microbiology Laboratory of Cairo University Hospitals, Cairo, Egypt in the period from January to June 2010; all K. pneumoniae clinical isolates that were suspected to be ESBL producers by initial screening tests according to the CLSI (2010) guidelines were collected.³

K. pneumoniae isolates were identified by the routine biochemical reactions and were further confirmed by API 10 E (biomérieux^®SA). All 91 K. pneumoniae isolates were subjected to the following:

Antimicrobial susceptibility testing by the disc diffusion method (Kirby-Bauer technique) using Muller-Hinton agar; aerobic incubation at 35°C for 16–18 hr. Antimicrobial discs: imipenem (10 μg), meropenem (10 μg), gentamicin (10 μg), ciprofloxacin (5 μg), amikacin (30 μg), cotrimoxazole (25 μg), cefepime (30 μg), cefotaxime (30 μg), cefotaxime+clavulanic acid (30/10 μg), aztreonam (30 μg), ceftazidime (30 μg), ceftazidime+clavulanic acid (30/10 μg), amoxicillin–clavulanic acid (20/10 μg), and cefoxitin (30 μg) were obtained from Oxoid limited Basingstoke. Phenotypic screening and confirmation of ESBL production were done according to CLSI, 2010, using the combined disc tests.³

Screening for AmpC production was done by susceptibility to cefoxitin disc followed by confirmation of AmpC production by the combined disc using cefotaxime alone and in combination with 300 μg boronic acid (3-aminophenylboronic acid) and the double-disc synergy testing.⁴

Imipenem minimal inhibitory concentration (MIC) was determined by the E-test strips (biomérieux SA) according to the manufacturer's recommendations; the results were interpreted according to the CLSI (2010) interpretive criteria.³

Analysis of outer membrane proteins (OmpK35–OmpK36)

Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) was performed in 38 isolates according to Hernández-Allés et al.⁹ The initial optical density at 600 nm (OD600) was of ∼0.02. Cultures were grown to the early exponential phase till the OD600 reached 0.6 in pH 7.4.

Analysis of outer membrane protein (OmpK35–OmpK36) gene expression

All K. pneumoniae isolates were subjected to analysis of mRNA using reverse transcription-PCR (RT-PCR). RNA was extracted from early exponential phase cultures using an RNeasy Protect Bacteria Mini Kit (QIAGEN) according to the manufacturer's recommendations.¹⁵ RT-PCR was performed using the OneStep RT-PCR Kit (QIAGEN). The 16S rRNA sense primer 5′AATACCGCATAATGTCGC3′ and reverse primer 5′CCCTCGTTTGTCCATATCT3′¹⁵ were used to normalize the levels of mRNA expression in different strains. The target genes included OmpK35 sense primer 5′CAGACACCAAACTCTCATCAATGG3′, reverse primer 5′AGAATTGGTAAACGATACCCACG3′ and OmpK36 sense primer 5′CCGTAACTCTGATTTCTTCG3′, reverse primer 5′TTAGTTGGACGACCTGCT3′.¹¹ Amplification was done using thermal cycler (Gene Amp^® PCR System 9700; Applied Biosystems). The thermal cycler conditions were as follows: RT lasted for 30 min at 50°C, while the initial PCR activation lasted for 15 min at 95°C. The DNA amplification cycle started at 94°C for 0.5–1 min followed by annealing for 0.5–1 min at 50–68°C, and then the initial extension started at 72°C for 1 min (25–40 cycles), whereas the final extension lasted for 10 min at 72°C.

Every PCR was considered negative after failure to amplify the target genes of the expected size for at least three times.

The K. pneumoniae ATCC 13883 strain was used in SDS-PAGE and RT-PCR as a positive control for OmpK35 and OmpK36.

Statistical methods

Data were statistically described in terms of mean±standard deviation (±SD), median and range, or frequencies (number of cases) and percentages when appropriate. Comparison of quantitative variables between the study groups was done using the Mann–Whitney U test for independent samples. For comparing qualitative variables, the chi-square (χ²) test was performed. Fisher's exact test was used when the expected frequency was less than 5. Agreement was calculated using kappa statistic. Accuracy was represented using the terms sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy. p-Values less than 0.05 were considered statistically significant. All statistical calculations were done using computer programs SPSS (Statistical Package for the Social Science; SPSS, Inc.) version 15 for Microsoft Windows.

Results

Antimicrobial drug resistance patterns of 91 K. pneumoniae isolates are shown in Table 1.

Table 1.

Antimicrobial Susceptibility Testing Results of 91 Isolates of Klebsiella pneumoniae Collected from Cairo University Hospitals

Antibiotic	Sensitive, n (%)	Intermediate, n (%)	Resistant, n (%)
Imipenem (10 μg)	73 (80.2)	5 (5.5)	13 (14.3)
Meropenem (10 μg)	73 (80.2)	1 (1.1)	17 (18.7)
Gentamicin (10 μg)	29 (31.9)	0 (0.0)	62 (68.1)
Ciprofloxacin (5 μg)	26 (28.6)	2 (2.2)	63 (69.2)
Amikacin (30 μg)	57 (62.6)	2 (2.2)	32 (35.2)
Cotrimoxazole (25 μg)	20 (22)	0 (0.0)	71 (78)
Cefepime (30 μg)	9 (9.9)	0 (0.0)	72 (79.1)
Cefotaxime (30 μg)	0 (0.0)	0 (0.0)	91 (100)
Cefotaxime+clavulanic acid (30/10 μg)	66 (72.5)	0 (0.0)	25 (27.5)
Aztreonam (30 μg)	23 (25.3)	4 (4.4)	64 (70.3)
Ceftazidime (30 μg)	0 (0.0)	0 (0.0)	91 (100)
Ceftazidime+clavulanic acid (30/10 μg)	66 (72.5)	0 (0.0)	25 (27.5)
Amoxicillin–clavulanic acid (20/10 μg)	1 (1.1)	0 (0.0)	90 (98.9)
Cefoxitin (30 μg)	46 (50.5)	0 (0.0)	45 (49.5)
Pipracillin–tazobactam (100/10 μg)	31 (34.1)	0 (0.0)	60 (65.9)

Confirmed ESBL producers were 66 isolates, among which loss of OmpK35 and OmpK36 was detected by RT-PCR in 86.45% and 65.2%, respectively, whereas loss of both porins was detected in 63.6% of isolates (p-value 0.028), as shown in Fig. 1. Concerning imipenem (IMI) susceptibility among the confirmed ESBL producers, 65 of 66 (98.5%) isolates were sensitive and only one isolate (1.5%) was intermediate to IMI.

FIG. 1.

Reverse transcription-polymerase chain reaction (RT-PCR) results of OmpK35 and OmpK36. The expected products are at 778 for OmpK35 bp (upper panel) and 588 bp for Omp36 (lower panel), respectively. Lane 1, DNA ladder. Lane 2, QC strain Klebsiella pneumoniae ATCC 13883, carries Omp35 and Omp36. Lane 3, positive Omp35 with loss of Omp36, lanes 4 and 6 are negative for both genes (loss of Omp35 and Omp36). There is no expression of OmpK35 with positive Omp36 in lanes 5 and 8.

Confirmed AmpC producers were 20 isolates from 45 cefoxitin-resistant isolates, 17 of 91 (18.7%) isolates showed a derepressed AmpC pattern, while 2 of 91 (2.2%) isolates showed both inducible and derepressed patterns of resistance and only 1 of 91 isolates (1.1%) showed an inducible pattern. Loss of both Omp35 and Omp36 porins was found in all 20 (100%) AmpC-producing isolates. Concerning IMI susceptibility among the 20 AmpC-producing isolates, 13 of 20 (65%) isolates showed elevated IMI MIC (11 were resistant and two isolates were intermediate to IMI).

The mean of IMI MIC among all tested K. pneumoniae clinical isolates was 2.24 μg/ml, whereas it was 8.55 μg/ml among AmpC-producing isolates. The IMI MIC mean values among different groups of tested isolates are shown in Table 2.

Table 2.

The Imipenem MIC Mean Values Among K. pneumoniae-Tested Isolates

K. pneumoniae tested groups	Mean of IMI MIC (μg/ml)	p-Value
All 91 K. pneumoniae isolates+presence of OmpK35	0.28	0.05
All 91 K. pneumoniae isolates+loss of OmpK35	2.75
All 91 K. pneumoniae isolates+presence of OmpK36	0.98	0.269
All 91 K. pneumoniae isolates+loss of OmpK36	2.75
Within ESBL-producing isolates+OmpK35	0.29	0.498
Within ESBL-producing isolates+loss of OmpK35	0.33
Within ESBL-producing isolates+OmpK36	0.29	0.789
Within ESBL-producing isolates+loss OmpK36	0.33

ESBL, extended-spectrum β-lactamases; IMI, imipenem; MIC, minimal inhibitory concentration.

Elevation of imipenem MIC was detected in 18 (19.8%) isolates; 13 isolates were resistant to IMI by the E-test (12 of 13 showed combined porin loss and confirmed AmpC production, whereas 1 of 13 isolates had lost both Omp35 and Omp36 without showing confirmation of either ESBL or AmpC production).

Within five IMI intermediate isolates, the loss of both porins was detected in all isolates, whereas two of five (40%) isolates were AmpC producers. The remaining three of five (60%) isolates had lost both Omp35 and Omp36 without showing confirmation of either ESBL or AmpC production.

Among 45 FOX-resistant isolates, OmpK35 was lost in 45 (100%), whereas OmpK36 was lost in 42 (93.3%), and both porins were lost in 42 (93.3%).

Regarding susceptibility to ceftazidime, loss of OmpK35 increased the estimated mean of CAZ zone of inhibition (p 0.025), whereas OmpK36 loss elevated the estimated mean of CAZ zone of inhibition significantly (p 0.006).

Analysis of porin loss by SDS-PAGE compared to porin gene analysis by RT-PCR showed 0.015 agreement, a 63.64% positive predictive value, a 77.78% negative predictive value, 53.85% sensitivity, and the specificity was 84%.

Discussion

Infection due to XDR K. pneumoniae is a growing problem in tertiary medical centers such as the Cairo University Hospitals. According to a study conducted in the surgical ICU of the same hospital, susceptibility to IMI was 88.7% versus 73% in our study.¹ The decreasing IMI susceptibility raised our attention to understand the different mechanisms of IMI resistance in our hospital.

In our study, among the confirmed ESBL producers, the absence of Omp35/36 (p-value 0.498 and 0.789, respectively) does not affect the estimated mean of imipenem (IMI) MIC (0.3 μg/ml). Yang et al. (2009) found that ESBL production and reduced expression of OmpK35/36 may not result in clinical resistance to imipenem, but does correlate with increasing imipenem MIC.¹⁵ Kaczmarek et al. (2006) added that in case of OmpK35–OmpK36 gene disruption and production of β-lactamases, the upregulation of other porin expressions may recompense for porin loss and conferred carbapenem susceptibility.¹¹

The association of both Omp35/36 loss with AmpC production markedly increased the IMI MIC to 8.55 μg/ml, an observation that was also reported by Frazee et al.⁶

Elevation of IMI MIC within four isolates showed OmpK35/36 loss without production of ESBL or AmpC production was noticed; hence KPC production was tested and confirmed in two isolates.¹³

In our study, the mean of ceftazidime inhibition zone diameter was significantly decreased among ESBL-producing isolates showing Omp35 or Omp 36 loss (p-value 0.041 and 0.006), respectively. Similarly, it was reported that loss of OmpK35 played a major role in the emergence of ceftazidime resistance.^{15, 12}

Loss of both porins Omp35 and Omp36 was detected in all FOX-resistant isolates; other authors concluded that one of the mechanisms of cefoxitin resistance in K. pneumoniae is the reduced permeability of porins to β-lactam antibiotics.^15,2 Tsai et al. concluded that a double deletion of both porins further increased cefoxitin MIC to high-level resistance in K. pneumoniae.¹⁴ Hence, we can conclude that loss of Omp35 and Omp36 may contribute to cefoxitin resistance among ESBL-producing isolates.

Conclusion

The association of OmpK35/36 porin loss with ESBL production in K. pneumoniae may not be a direct cause of resistance to imipenem, although the combined porin OmpK35/36 loss and AmpC production may be a decisive factor in K. pneumonia resistance to imipenem. Continuous complementary research is recommended nationwide for the better understanding of the different mechanisms of antibiotic resistance.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Hasanin

, Eladawy

, Mohamed

, Salah

, Lotfy

, Mostafa

, Ghaith

, and Mukhtar

2014. Prevalence of extensively drug-resistant gram negative bacilli in surgical intensive care in Egypt. Pan Afr. Med. J., 19:177.

Ananthan

, and Subha

2005. Cefoxitin resistance mediated by loss of a porin in clinical strains of Klebsiella pneumoniae and Escherichia coli. Indian J. Med. Microbiol., 23:20–23.

Clinical and Laboratory Standards Institute. 2010. Performance standards for antimicrobial susceptibility testing: 19th informational supplement. Document M100-S29. Clinical and Laboratory Standards Institute, Wayne, PA.

Doi

, and Paterson

D.L.

2007. Detection of plasmid-mediated Class C β-lactamases. Int. J. Infect. Dis., 11:191–197.

Doménech-Sánchez

, Martínez-Martínez

, Hernández-Allés

, Conejo

M.C.

, Á. Pascual, Tomás

J.M.

, Albertí

, and Benedí

V.J.

2003. Role of Klebsiella pneumoniae OmpK35 porin in antimicrobial resistance. Antimicrob. Agents Chemother., 47:3332–3335.

Frazee

B.W.

, Hansen

, and Lambert

2009. Invasive infection with hypermucoviscous Klebsiella pneumoniae: multiple cases presenting to a single emergency department in the United States. Ann. Emerg. Med., 53:639–642.

García-Sureda

, Doménech-Sánchez

, Barbier

, Juan

, Gascó

, and Albertí

2011. OmpK26, a novel porin associated with carbapenem resistance in Klebsiella pneumoniae. Antimicrob. Agents Chemother., 55:4742–4747.

García-Sureda

, Juan

, Doménech-Sánchez

, and Albertí

2011. Role of Klebsiella pneumoniae LamB porin in antimicrobial resistance. Antimicrob. Agents Chemother., 55:1803–1805.

Hernández-Allés

, Albertí

, Alvarez

, Doménech-Sánchez

, Martínez-Martínez

, Gil

, Tomás

J.M.

, and Benedí

V.J.

1999. Porin expression in clinical isolates of Klebsiella pneumoniae. Microbiology, 145: 673–679.

10.

Jacoby

G.A.

, Mills

D.M.

, and Chow

2004. Role of β-lactamases and porins in resistance to ertapenem and other β-lactams in Klebsiella pneumoniae. Antimicrob. Agents Chemother., 48:3203–3206.

11.

Kaczmarek

F.M.

, Dib-Hajj

, Shang

, and Gootz

T.D.

2006. High-level carbapenem resistance in a Klebsiella pneumoniae clinical isolate is due to the combination of blaACT-1 β-lactamase production, porin OmpK35/36 insertional inactivation, and down regulation of the phosphate transport porin PhoE. Antimicrob. Agents Chemother., 50:3396–3406.

12.

Shakib

, Ghafourian

, Zolfaghary

M.R.

, Hushmandfar

, Ranjbar

, and Sadeghifard

2012. Prevalence of OmpK35 and OmpK36 porin expression in beta-lactamase and non-beta lactamase-producing Klebsiella pneumoniae. Biologics, 6:1–4.

13.

Helal

S.F.

, El-Rachidi

N. G.

, AbdulRahman

E.M.

, and Hassan

D.M.

2014. The presence of blaKPC-mediated resistance in Enterobacteriaceae in Cairo University hospitals in Egypt and its correlation with in vitro carbapenem susceptibility. J. Chemother., 26:125–128.

14.

Tsai

, Fung

, Lin

, Chen

, Chang

, Chen

, and Siu

L.K.

2011. Klebsiella pneumoniae outer membrane porins OmpK35 and OmpK36 play roles in both antimicrobial resistance and virulence. Antimicrob. Agents Chemother., 55:1485–1493.

15.

Yang

D.Y.

, Guo

, and Zhang

2009. Combined porin loss and extended spectrum beta-lactamase production is associated with an increasing imipenem minimal inhibitory concentration in clinical Klebsiella pneumoniae strains. Curr. Microbiol., 58:366–370.