Efflux pump genes and chlorhexidine resistance : Clue for Klebsiella pneumoniae infections in intensive care units , Egypt

One of the serious and growing challenges for infection control programs worldwide is hospital acquired infections. Klebsiella pneumoniae is considered the second most common cause of hospitalacquired Gram-negative blood stream infections. Chlorhexidine is a topical antiseptic agent, widely used in different applications in hospitals. Healthcare associated infection rates were lower after chlorhexidine-containing detergent hand washing when compared with using plain soap or an alcoholbased hand rinse. Intensive exposure of hospital pathogens to biocides may result in the emergence of resistance not just to the biocides but also to antibiotics. We aimed to investigate the susceptibility of multi-drug resistant K. pneumoniae isolates to chlorhexidine and to correlate chlorhexidine susceptibility and its association with both the efflux pump genes (cepA, qacDE, qacE), and resistance to later-generation anti-Gram negative antibiotics. Fifty-six strains of K. pneumoniae were isolated from blood specimens in intensive care units, Suez Canal University Hospital, Ismailia, Egypt. Antibiotic sensitivity profiles were determined by disc diffusion method. Minimal inhibitory concentration (MICs) of 1% chlorhexidine was assessed by the agar dilution method. The effect of efflux pumps was determined by repeating the susceptibility in the presence of the efflux pump inhibitor carbonyl cyanide m-chlorophenyl hydrazone (CCCP) (10 mg/L). Polymerase chain reaction (PCR) was used for identifying efflux pump genes. The MICs of chlorhexidine ranged from 4 256 mg/L. Most isolates carried the cepA gene. The MICs of chlorhexidine was significantly reduced on addition of CCCP. Carriage of efflux pump gene cepA affect chlorhexidine susceptibility in ICU related K. pneumoniae infections.


INTRODUCTION
Health care associated infections (HAI) have direct or indirect extensive health effects and large economic burden (Valbona et al., 2004).Despite the lower number of intensive care units (ICU) patients as compared to those in other clinics, the ICUs have the highest rate of HAI among all the units.This can be attributed to fre-quent use of antibiotics and invasive techniques such as central venous catheterization, mechanical ventilation and urinary catheterization for long periods (Akalın, 2001).
Klebsiella pneumoniae is considered the second most common cause, after Escherichia coli, of hospital-acquired Gram-negative bloodstream infection (BSI) (Meatherall et al., 2009).It causes about 8% of hospital-acquired infections including pneumonia, wound infections, diarrhoea and urinary tract infections (Gupta, 2002).Furthermore, it can lead to a poor prognosis in critically ill patients with bacteremic community-acquired pneumonia (Lin et al., 2010).
The severity of BSI has been associated with multi drug resistant (MDR) strains producing extended-spectrum beta-lactamase enzymes (ESBLs).This is associated with increased morbidity and mortality especially in intensive care units (ICUs), neonatal units and surgical wards (Khanfar et al., 2009).BSI is considered a major problem in critically care units as it is associated with increased rates of hospitalization, and increased length of stay and hospital costs (Pien et al., 2010).
Hand contamination of Health Care Workers (HCWs) is one of the most common modes of transmission of K. pneumoniae in ICUs.Others include direct patient contact and respiratory-tract care; all of which contaminate the fingers of caregivers.Whereas, Staphylococcus aureus account for 11% of isolates, Gram-negative bacilli accounted for about 15%.Duration of patient-care activity was strongly associated with the intensity of bacterial contamination of HCWs' hands (Samore et al., 1996).
In order to combat these infections, the staff frequently wash their hands with cationic biocides (such as chlorhexidine), which produces a 98-100% reduction in the number of patients infected with K. pneumoniae (Casewell and Phillips, 1977).Chlorhexidine is a topical antiseptic agent with a broad spectrum of actions, widely used in different applications in hospitals such as surface cleaning, hand disinfection and skin preparation before invasive procedures (Milstone et al., 2008).HAIs rates were lower after antiseptic hand washing using a chlorhexidine-containing detergent when compared with hand washing with plain soap or use of an alcohol-based hand rinse (Doebbeling et al., 1992).
Even with the considerable reductions in K. pneumoniae infections with the wide use of biocides, infections caused by multi-resistant K. pneumoniae are increasing, and this raises concerns that intensive exposure of hospital pathogens to biocides may result in the emergence of resistance not just to the biocides but also to antibiotics (Buehlmann et al., 2011;Chong et al., 2011).The concentrations of the biocides used clinically are often higher than those required to inhibit the organism.However, many factors, such as the local dilution policy, presence of organic matter, formation of biofilms and length of exposure, could comprise the effective concentration of the biocide so that resistance may become a problem (Smith and Hunter, 2008;Tumah, 2009).
As there is little information on reduced activity of chlor-hexidine toward K. pneumoniae and its relationship with resistance to antibiotics, this study aimed to investi-gate the activity of chlorhexidine on clinical isolates of MDR K. pneumoniae.And also study the correlation bet-ween chlorhexidine activity and its association with both the efflux pump genes (cepA, qacDE and qacE), which have been linked to antiseptic 'resistance', and resistance to later-generation anti-Gram negative antibiotics (Kucken et al., 2000).

Bacterial strains
Fifty-six strains of K. pneumoniae were isolated from blood specimens in patients with BSIs at intensive care units in Suez Canal University from November 2012 to June 2013.All blood cultures were examined by the hospital's microbiology laboratory using the BacT/ALERT ® system (bioMe´rieux, France) and the isolated strains were confirmed by the API 20E strips (bioMe´rieux, Basingstoke, UK).

Antibiotic susceptibility
Antibiotic susceptibility testing was performed using the Kirby-Bauer Disk Diffusion test, according to the criteria of the National the Clinical and Laboratory Standards Institute (CLSI) (National Committee for Clinical Laboratory Standards, 2002).The antibiotics used were cefotaxime, cefoxitin, colistin, chloramphenicol, gentamicin, polymyxin B, ciprofloxacin, ceftriaxone, chloramphenicol, amikacin, trimethoprim, piperacillin-tazobactam, ceftazidime, imipenem and meropenem (Sigma, Poole, UK).These are the commonly used generation of antibiotics for the Gram negative strains in the infirmary.

Minimum inhibitory concentration (MIC) of chlorhexidine
The common hospital biocide used was 1% chlorhexidine gluconate, a member of the biguanide family (Sigma, St. Louis, MO, USA).The MIC of chlorhexidine was determined using agar dilution in accordance with the 2002 guidelines of the CLSI.Serial two fold dilution of chlorhexidine digluconate on Muller-Hinton (MH) agar containing 0-256 mg/mL was used.K. pneumoniae ATCC13883 (MIC of chlorhexidine 16 mg/mL) and Escherichia coli ATCC25922 (MIC of chlorhexidine 2 mg/mL) were included in each experiment as control strains.

Efflux pumps inhibitor and reduced biocide susceptibility
In order to determine the activity of efflux pump, 10 mg carbonyl cyanide m-chlorophenyl hydrazone (CCCP) (Sigma aldrich, USA) was dissolved in 10 ml of dimethyl sulphoxide (DMSO) to make 10 µmol concentration.This was then added to Muller Hinton agar at 10 mg/L in plates containing increasing concentrations of chlorhexidine for minimum inhibitory concentration (MIC) determination.Decrease in chlohexidine MIC indicated the presence of an efflux

Polymerase chain reaction (PCR)
The antiseptic resistance genes cepA, qacDE and qacE were screened by PCR.The primer sequences, specificities and predicted amplicon sizes are listed in Table 1 Positive and Non-template controls were included in each experiment (positive control was a known resistant K. pneumoniae isolate from a surgical wound previously tested as described before by Fang et al. (2002) and Khazama et al. (1998).The optimal cycling conditions were as follows: 5 min at 95°C, 95°C for 45s, (66 °C for 30s for cepA , 49 ̊ C for 40s for both qacE and qacDE), 72°C for 1 min for 35 cycles and final extension step 72°C for 10 min (Fang et al., 2002;Khazama et al., 1998) using Eppendorf Mastercycler® nexus PCR thermal cycler.PCR products were analyzed on agarose gel 1.5% (Promega, Madison, USA).The gel was stained with ethidium bromide and photos were taken using the documentation system syngene G.Box (Syngene, UK).

Antiseptic resistance genes
Forty two isolates contained the cepA gene and, in every case, the size of gene fragment was 1051 bp.All samples were negative for both qacE and qacDE genes by PCR (Figures 1 and 2).Table 2 shows the individual levels of chlorhexidine, susceptibility with the presence of antiseptic resistance genes cepA, qacE and qacDE.High MICs (64-128 mg/L) of chlorhexidine were found in 15 and 12% isolates, respectively.On the other hand, qacDE and qacE antiseptic resistance genes were not   found in any isolate.

Efflux pumps inhibitors and reduced biocid susceptibility
On addition of CCCP (10 mg/L) to the MICs of the chlorhexidine, we found that the MICs of chlorhexidine reduced in 39 isolates (Tables 2).This was associated with the presence of the cepA gene in 35 isolates.No reduction of MICs was observed at concentration 4-8 mg/L.In association with cepA gene, reduction on MICs occurred in 3 strains at concentration 16 mg/L, 8 strains at concentration 32 mg/L, 13 strains at concentration 64 mg/L, 10 strains at concentration 128 mg/L and one strain at concentration 256 mg/L.In the absence of cepA gene, the reduction on MICs on addition of CCCP occurred in only 4 strains (2 of them from concentration 64 to 16 mg/L, one from concentration 32 to 16 mg/L and one from concentration 16 to 4 mg/L).

Antibiotic sensitivity pattern
The susceptibility of 56 isolates of K. pneumoniae was determined by disc diffusion methods.All isolates were sensitive to imipenem and meropenem.The MICs of chlorhexidine ranged from 4 -256 mg/L as shown in Table 3 which were higher than the control strains K. pneumoniae ATCC13883 (MIC of chlorhexidine 16 µg/ml) and E. coli ATCC25922 (MIC of chlorhexidine 2 µg/ml).
The MIC for all used antibiotics in the ICU was determined for the isolates.Increased MIC range for most of the antibiotics was noted (Table 4).The resistance of some isolates to ceftazidime, cefotaxime and ciprofloxa-cin was noted in traditional disc diffusion method.Most isolates were sensitive to third generation cephalosporines.A similar consistent antibiotic sensitivity pattern for the 4 isolates (with decreased MIC for chlorehexidine and absent resistant genes) was shown.

DISCUSSION
Chlorhexidine is a widely used topical antiseptic that has been used worldwide since the fifties of the last century.Its safety has been proved with high efficacy in many clinical applications.These include preoperative skin preparation, vaginal antisepsis and body washes to prevent neonatal sepsis.Its major role in infection control has been followed over years with high satisfactory results (Hugo and Longworth, 1964;McDonnell and Russell, 1999).
Reduced susceptibility to disinfectants among bacteria often involves the action of active or over expressed efflux pumps (Russell et al., 1998).Many products contain chlorhexidine in different concentrations and forms.Chlorhexidine affects membrane integrity at low concentrations, whereas at high concentrations it precipitate cytoplasmic contents, resulting in cell death (Hugo and Longworth, 1964;McDonnell and Russell, 1999).
Unlike Abuzaid and his co-workers (2012), we did not examine other biocides as they are not commonly used in our hospital.In our hospital, chlorhexidine is considered to be an essential component of infection control strategies and the most widely used disinfectant in hospitals, so high MICs of chlorhexidine may offer an advantage to nosocomial pathogens.So this study evaluates the relationship between reduced susceptibility to chlorhexidine and the carriage of antiseptic resistance genes, cepA, qacDE and qacE, as well as identify the role of efflux pumps in conferring reduced susceptibility.Moreover, this may be of great help in optimizing the best concentration for this antiseptic.
Increasing frequency of hospital infection leads to overuse and pressure of biocides, similarly to antibiotics.The linkage between bacterial resistance and the use of biocides has been suggested (Block and Furman, 2002).They have observed a significant inverse relationship between the intensity of chlorhexidine use and overall susceptibility of organisms (S. aureus, coagulase-negative staphylococci, K. pneumoniae, P. aeruginosa, Acinetobacter baumannii and Candida albicans) to this antiseptic (Smith et al., 2008;Abuzaid et al., 2012).We examined the most common pathogenic species in the ICU (K.pneumoniae followed by E. coli and S. aureus) according to the infection control unit records.In our study, MICs of chlorhexidine ranged from 4 -256 mg/L.
The MIC assays performed in the presence of CCCP showed a considerable decrease in the MICs of chlorhexidine for almost all the strains (39 isolates) and this was usually associated with the presence of cepA except in four cases (35 isolates).These results were in agreement with that of Abuzaid and his co-workers (2012) who examined the effect of CCCP on the MIC of different hospital biocides.They found that it reduced the MICs of chlorhexidine considerably with presence of cepA gene in most of the isolates (except five strains).
On the other hand, no strains carried the other 2 efflux pump genes (qacE and qacDE) unlike Abuzaid et al. (2012) who detected qacDE gene in 34 isolates; and qacE w in only one isolate out of 64 isolates.
The chlorehexidine susceptibility was notably reduced in most isolates similar to the decrease reported by Abuzaid el al. (2012).However, a different pattern of antiseptic resistance genes carriage was noted.This may be explained by the exposure to a significant amount of chlorehexidine in the ICU.Moreover, K. pneumoniae isolates were obviously resistant to the commonly used antibiotics in contrast with those of Abuzaid et al. (2012) and in accordance with Koljalg et al. (2002).
Thus, measures were applied in line with the decisions of the infection control committee.Intensive care unit staff was trained regularly and frequently.The staff was trained on hand hygiene and personal protective equipment to limit the transmission of infections.Also, we recommend evaluating and applying the possibility of predicting chlorhexidine susceptibility as a routine work with multidrug resistant organisms.However, the clinical importance of this is considered when other factors are taken into account, such as biofilm formation, which already compromises the activity of the biocide.
The presence of decreased MIC for chlorehexidine in 4 isolates without any resistant gene was an interesting finding.This was incoherent with previous results reported by Abuzaid et al. (2012).However, a similar consistent antibiotic sensitivity pattern with the lowest MIC for most of the antibiotics used was shared by these isolates.

Conclusion
The carriage of efflux pump gene cepA affects chlorhexidine activity on K. pneumoniae infections in the ICU related infections.However, the study should be extended to include other species in the same ward.Also, other antibiotics should be tested (other than the commonly used antibiotics in the hospital).Further evaluation against biofilm formation should be planned with the hospital team.
*Corresponding author: E-mail: m.fouad452006@yahoo.com.Tel: +201025099921.Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License

Table 2 .
Effect of carbonyl cyanide m-chlorophenyl hydrazone (CCCP in 10 mg/L) on the minimum inhibitory concentration (MIC) of chlorhexidine in association with CepA gene.

Table 3 .
Minimal inhibitory concentrations (MICs) for chlorhexidine in different K. pneumoniae strains in the presence of different efflux pumps genes.

Table 4 .
The minimum inhibitory concentration (MIC) of chlorhexidine correlated with antibiotics sensitivity pattern among K. pneumoniae isolates.