HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Kolayli, F. Articles by Vahaboglu, H. Search for Related Content PubMed PubMed Citation Articles by Kolayli, F. Articles by Vahaboglu, H. Agricola Articles by Kolayli, F. Articles by Vahaboglu, H.

Effect of carbapenems on the transcriptional expression of the oprD, oprM and oprN genes in Pseudomonas aeruginosa

^1–3,5Mikrobiyoloji ve Klinik Mikrobiyoloji AD¹, Tibbi Biyoloji AD², Biyofizik AD³ and Enfeksiyon Hastaliklari & Klinik Mikrobiyoloji AD⁵, KOU Tip Fakultesi, Sopali Ciftligi, 41900 Derince Kocaeli, Turkey ⁴IU, Deneysel Tip Arastirma Enstitusu, Istanbul, Turkey

Correspondence Haluk Vahaboglu vahabo{at}hotmail.com

Received April 7, 2004
Accepted May 27, 2004

The effects of imipenem and meropenem on the transcriptional expression of resistance-related genes oprD, oprM and oprN in Pseudomonas aeruginosa were studied by quantitative real-time PCR. Four strains were examined: the type strain PT5 (PAO1), its derivatives M7 and PT149, and a clinical isolate, PaKT3. The derivative M7 is a nalB mutant, overexpressing the MexAB–OprM pump, and the derivative PT149 is a nfxC-type mutant, overexpressing the MexEF–OprN pump while it is down-regulated for the OprD protein. After 18 h incubation in broth, the cultures were divided into three portions. Two were supplemented with antibiotics and the other was left antibiotic-free as the control. After a further 45 min incubation, total RNA was isolated from the strains by guanidine denaturation and acid-phenol/chloroform extraction. DNA-free total RNAs were immediately reverse-transcribed by MMuLV reverse transcriptase. Concentrations of mRNAs obtained by quantitative PCR were expressed relative to uninduced portions of the strains. The results showed that oprD was relatively stable against carbapenem antibiotics. oprM was induced significantly by imipenem in only one strain and oprN was induced by imipenem in most of the strains. The responses at the mRNA level found here were unexpected and suggested a chaotic, unpredictable regulatory mechanism.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Pseudomonas aeruginosa poses a significant health threat because of its ability to develop resistance to structurally unrelated antibiotics, leading to multiple drug resistance (MDR). In P. aeruginosa, MDR is often linked to the altered regulation of specific efflux pumps and porins (Poole, 2000, 2001). So far, four efflux pumps, all of the Resistance-Nodulation-Division (RND) type, namely MexAB–OprM, MexCD–OprJ, MexEF–OprN and MexXY–OprM, and an outer-membrane porin (OprD) have been shown to be strongly associated with antibiotic resistance (Poole, 2001).

Pumps belonging to the Mex family exhibit a broad range and substrate-specific affinity for antibiotics (Aires et al., 2002). Tetracycline, chloramphenicol, quinolones, most penicillins, most cephems (except ceftazidime) and meropenem are among the substrates of the so-called pumps (Masuda et al., 2000). Porin D, on the other hand, behaves as a portal for the entry of imipenem and is linked with resistance to this antibiotic (Ochs et al., 1999; Yoneyama & Nakae, 1993).

These pumps and porins play a vital role in the adaptation of P. aeruginosa to environmental conditions as well. For example, porin D facilitates the uptake of basic amino acids (Ochs et al., 1999), and efflux pumps from the Mex family regulate quorum sensing and a number of other virulence determinants (Evans et al., 1998; Kohler et al., 2001). In other words, these pumps and the outer-membrane porin OprD are engaged in extremely important functions, affecting both antimicrobial resistance and potential to cause disease.

The regulation of these operons and how they respond to antibiotic challenge at the mRNA level have not been fully explained. To this end, we studied the effect of two carbapenem antibiotics, imipenem and meropenem, on the transcriptional expression of the oprD, oprM and oprN genes of four selected strains of P. aeruginosa.

	METHODS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Strains.

Four strains were used in the study: the type strain P. aeruginosa PT5 (PAO1), two mutants of this, PT149 and M7 (Kohler et al., 1997), and a clinical isolate, PaKT3, that is fully susceptible to ß-lactam antibiotics. According to protein expression studies, PT149 is a nfxC-type mutant strain which overproduces MexEF–OprN and is down-regulated for OprD, while M7 is a nalB mutant overproducing the pump MexAB–OprM.

The MICs of imipenem and meropenem were determined by the E-test method (Oxoid) and are as follows: PT5, 1/0.38 (imipenem/meropenem, µg ml^–1); PT149, 2/0.5; M7, 1/0.5; and PaKT3, 1/0.19.

DNA and RNA studies.

All plastic materials were made RNase-free by treatment with diethyl pyrocarbonate (DEPC). To protect it from degradation, purified RNA was transcribed to cDNA immediately or was stored at 4 °C for a maximum of 1 h before reverse transcription. Cultures were grown in Mueller–Hinton broth (Oxoid) at 37 °C.

Induction by carbapenems was as follows. Bacteria were incubated in 30 ml broth for 18 h and cultures were divided into three equal portions. Imipenem and meropenem were added to two separate tubes and the remaining tube was left antibiotic-free as the control. Both the antibiotic-supplemented and control tubes were incubated for an additional 45 min before RNA isolation.

Induction experiments were carried out at low and high antibiotic concentrations. The low concentrations for imipenem and meropenem were 1 and 0.5 µg ml^–1, respectively; the high concentrations were 16 and 8 µg ml^–1, respectively. Experiments were repeated three times and the results were evaluated together.

In the second stage of the study, induction with imipenem and meropenem was tested using exponentially growing bacteria. First, growth plots in similar incubation conditions were obtained and showed that in these conditions the strains grew actively up to 13 h of the incubation period. As a result, inductions at this stage were tested at 10 h. Incubation and induction conditions were exactly the same as for the first stage experiment. However, imipenem and meropenem were incorporated at sub-MIC levels (imipenem, 0.5 µg ml^–1; meropenem, 0.25 µg ml^–1) and the induction periods were 45 min and 3 h.

Total RNA was isolated from bacteria using a commercial kit (EZ-RNA, Biological Industries) according to the manufacturer's instructions. The method is based on guanidine denaturation and acid-phenol/chloroform extraction of total RNA. An additional chloroform extraction step, however, was included to improve purity. The integrity of the RNA was confirmed by agarose gel (2.5 %) electrophoresis and, finally, total RNA was dissolved in 40 µl DEPC-treated double-distilled water. Reverse transcriptions were performed using random hexamer primers, 100 IU MMuLV (MBI Fermentas) and 8 µl RNA in 20 µl final volume at 42 °C for 2 h.

Real-time PCR.

The sequences of the studied genes were obtained from the gene bank of the National Library of Medicine, and primers were designed with the aid of OLIGO software (version 5.0) (Molecular Biology Insights). The sequences of the primers are shown in Table 1.

PCR was performed using a LightCycler instrument (Roche Diagnostics), in capillary glass tubes with the LightCycler FastStart DNA Master SYBR Green I kit (Roche). All work was carried out on desktop coolers (pre-cooled to 4 °C). Master mixtures were prepared in accordance with the manufacturer's recommendations but the final concentrations of Mg and primers were 2.5 mM and 50 pmol (each), respectively.

Control cDNA was obtained from P. aeruginosa PT5 and the primers of the control reactions were proC gene specific. Six controls obtained by twofold dilutions were always included. The concentration of the highest control was approximately 1 µg ml^–1.

The concentrations of the studied cDNA preparations were adjusted on the LightCycler instrument. Briefly, for every sample, 1 µl cDNA and 9 µl SYBR Green I (same concentration as indicated for the PCR assay by the manufacturer) were mixed in capillary tubes. After an incubation step of 95 °C for 5 min, the fluorescence emissions were read at 55 °C using the real-time fluorometry (RTF) facility of the LightCycler instrument to enable comparison of the amounts of cDNA. Subsequently the cDNA concentrations were adjusted to a level between the highest and the first dilution of the control cDNA. This adjustment was critical to obtain good amplification.

An arbitrary value of 1.600 E+3 was given to the highest concentration of control cDNA and twofold dilutions down to 0.050 E+3 were included. After 5 min of an initial ‘activation and denaturation’ step at 95 °C, PCR was accomplished in 13 s at 95 °C, 10 s at 58 °C, 15 s at 72 °C and with a single read after 5 s at 85 °C for 45 cycles.

Evaluation of real-time PCR.

Melting curve analysis was employed to check for the existence of primer dimers and other artefacts. Consequently, only dimer- and artefact-free amplifications were included in the analysis. The Second Derivate Maximum algorithm of the LightCycler software was used to calculate the concentrations. The software drew a standard curve by plotting the crossing cycle number versus the logarithm of given concentrations for each control sample.

For every sample, six genes were studied, namely proC, rpoD, oprB, oprD, oprM and oprN, and proC and rpoD genes served as the internal controls of the test strains (Savli et al., 2003). Raw concentration values were corrected according to the geometric means of the proC and rpoD genes.

Outer-membrane proteins.

Cells growing in Mueller–Hinton broth for 18 h were harvested by centrifugation at 1500 g for 30 min at 8 °C, suspended in 30 mM Tris/HCl (pH 8.0), and then broken in a sonicator for 2 min. Unbroken cells were removed by centrifugation at 4 °C and membranes were pelleted by centrifugation at 100 000 g for 1 h at 4 °C and suspended in the same buffer. The inner membrane was solubilized by adding sodium N-lauroylsarcosinate to the suspension at a final concentration of 1 % and incubating for 30 min at room temperature. The outer membrane was pelleted by centrifugation at 40 000 g for 40 min at 4 °C and then suspended in the buffer. The proteins were analysed by SDS-PAGE with 10.7 % (w/v) acrylamide and 0.3 % (w/v) N,N'-methylene-bis-acrylamide in the running gel. Samples for SDS-PAGE were treated with 2 % SDS–5 % 2-mercaptoethanol at 100 °C for 5 min and then subjected to electrophoresis at a constant 60 V for 2 h.

Study design and statistical analysis.

Twelve independently obtained cDNA samples of PT5 were diluted twofold up to six dilutions and used as controls in different experiments. First, the raw concentration data of the control dilutions were evaluated statistically to assess the magnitude of sampling errors and all other application deviations. Second, housekeeping genes proC and rpoD were statistically analysed for stability to validate their use as internal controls. Third, the normalized values of induced target genes were compared against uninduced controls for the relative expression levels of mRNA.

Statistical analyses were performed with the aid of the statistical package SPSS (version 9.0). Gene stability was calculated by Spearman correlation coefficients on the raw concentration data (Savli et al., 2003).

Relative comparisons of the resistance genes were made after correcting the crossing point (CP) values by the geometric mean of proC and rpoD genes with the aid of a freely distributed Excel applet REST (Pfaffl et al., 2002). This applet calculates the relative expression of a gene, which is defined as the expression ratio of a target gene against its control by the Pair-Wise Fixed Reallocation Randomization test. The details are described elsewhere (Pfaffl et al., 2002).

	RESULTS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Descriptive statistical data from the observed versus expected values of the serial dilutions are shown in Table 2. In this analysis, the standard deviations of the observed values were within ±0.5 of the means. Therefore, higher variations in the relative comparisons were accepted as significant. The second analysis was performed on correlations between the raw concentrations of the studied genes. This comparison revealed that proC and rpoD had the highest correlation coefficients (0.911). This high correlation validated the use of these genes as internal controls in these experiments. Subsequently, in relative comparisons, raw data were normalized using geometric means of these genes.

Relative comparisons of the mRNA from uninduced strains were obtained against the type strain PT5 (Fig. 1). This revealed the up-regulation of oprM in M7 and up-regulation of oprN in PT149, as expected (Table 3). However, the mRNA level of oprD was not down-regulated in PT149. On the other hand, the outer-membrane fractions in the SDS-polyacrylamide gel of PT149 and M7 were consistent with the previous findings (Kohler et al., 1997), such as the OprD protein was down-regulated in PT149 while the OprM protein was up-regulated in M7 (nalB mutant). Interestingly, both bands were absent in PaKT3 (Fig. 2).

View larger version (11K): [in this window] [in a new window]   Fig. 1. Anchor bar chart (anchor set point, 1) of oprB, oprD, oprM and oprN genes relative to the corresponding genes of the type strain PT5 (PAO1).

View larger version (68K): [in this window] [in a new window]   Fig. 2. SDS-PAGE analysis of the outer-membrane proteins of the studied strains; the outer membranes were prepared and solubilized in 2 % SDS–5 % 2-mercaptoethanol at 100 °C for 5 min. Lanes: M, molecular mass marker; 1, PT5 (type strain); 2, PT149 (nfxC-type mutant); 3, M7 (nalB mutant); 4, PaKT3.

The relative expressions of mRNAs from induced versus uninduced bacteria are shown in Table 4. This showed three important features: first, imipenem especially at the high concentration, induced oprN in most instances; second, imipenem, and to a lesser extent meropenem, induced oprM in PaKT3 but not in others; and last, oprD was not significantly affected by antibiotic pressure.

These unexpected results prompted us to repeat the experiments under sub-MIC levels of antibiotics over two time periods, i.e. 45 min and 3 h (Table 5). The ampC gene was strongly induced by imipenem as expected, and oprD again was not down-regulated under imipenem pressure. However, imipenem and meropenem in most instances increased oprM and more strikingly increased oprN transcriptions.

	DISCUSSION

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
We tested the responses of the oprB, oprD, oprM and oprN genes of P. aeruginosa to carbapenem antibiotics at the mRNA level. Genes oprD, oprM and oprN are antimicrobial resistance related and oprB is a carbohydrate gateway (Wylie & Worobec, 1995). The latter was included to serve as a control for cellular status. We showed that oprB was not significantly affected by the inducing conditions even with the high levels of carbapenems. In other words, it is highly probable that the inducing conditions in this study did not cause sharp reactions in the cellular metabolism of P. aeruginosa within the initial 45 min.

It has been shown in outer-membrane diffusion studies that OprD is the principal portal of entry for imipenem (Huang & Hancock, 1996; Perez et al., 1996). Confirming these experiments, imipenem-resistant isolates, in some studies, have been found to be deficient in the OprD protein (Trias et al., 1989; Yoneyama & Nakae, 1993). Therefore, a relation between the deficiency of OprD and imipenem resistance is evident. To our knowledge, this relationship has not been previously studied at the mRNA level. We tested this relationship in four selected strains, one of which, PT149, is a mutant with down-regulation of the OprD protein. Interestingly, oprD mRNA was not down-regulated in this strain. The disagreement between protein and mRNA levels of oprD in PT149 has been shown by indirect methods previously and was explained by the existence of a post-transcriptional regulation pathway (Kohler et al., 1997). However, one might expect a negative response (down-regulation) of oprD mRNA among the other three strains during imipenem pressure, which did not appear to occur. These experiments failed to show a significant relationship between oprD mRNA levels and imipenem or meropenem induction in the tested strains. This study, therefore, provided further evidence for the existence of a post-transcriptional regulatory pathway for OprD.

Another significant finding was the absence of the band corresponding to OprD in PaKT3 despite the slightly increased level of its mRNA. This strain was a fully susceptible clinical isolate and so the absence of the OprD band needs to be explored in further studies.

The highly inducible nature of OprM in PaKT3 is worthy of note. In uninduced conditions, OprM mRNA was down-regulated by a factor of 0.003. In other words, oprM was almost undetectable in PaKT3, which is consistent with the absence of the corresponding protein band in the SDS-PAGE analysis. When induced by imipenem it was up-regulated significantly. Imipenem failed to induce OprM in other test strains. The fact that oprM, in contrast to the other strains, was highly inducible in PaKT3 could be explained by the existence of a defective repressor gene, mexR. Sequence analysis, however, failed to show any insertion or deletion in the mexR region (data not shown) of PaKT3.

The mRNA of oprN was induced by imipenem in all strains, except M7. This was also unexpected, because it has been demonstrated that imipenem is not a substrate of the MexEF–OprN pump (Maseda et al., 2000).

Induction experiments under sub-MIC levels of antibiotics at the exponentially growing phase further revealed the unpredictable and unexpected reaction of these porin and pump genes in P. aeruginosa.

This study provided interesting data on the mRNA responses of P. aeruginosa to carbapenem antibiotics. Imipenem induced oprM in one strain but not in others, oprD stayed relatively stable, and the responses of oprN were variable. In other words, transcriptional responses in P. aeruginosa were unpredictable and chaotic. Further studies are therefore needed to understand the pathways of mRNA regulation of resistance-related genes in P. aeruginosa during antibiotic pressure.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
We are grateful to T. Kohler for the generous provision of the type strains and moreover for his kind help in interpreting Fig. 2. Also we wish to thank the Research Council of Kocaeli University for supporting this study.

	REFERENCES

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES