Introduction

Pseudomonas aeruginosa (P. aeruginosa) is a Gram-negative bacillus, which is frequently isolated from hospital environments, especially from medical equipments used in intensive care units (ICUs) [1,2,3]. P. aeruginosa is an opportunistic human pathogen that is associated with various community-acquired and nosocomial infections such as ventilator-associated pulmonary infections, skin and soft-tissue infections, catheter-related urinary tract infections, eye and ear infections, bloodstream infections, endocarditis and surgical site/transplantation infections [1,2,3]. Therefore, it seems that hygiene control of surfaces and medical equipments in the hospital settings in terms of P. aeruginosa contamination is important because nosocomial infections are considered as a growing global threat in terms of economical and public health [3]. The most common antiseptic and disinfectant biocides used in clinical settings are chlorhexidine digluconate (a biguanide, disrupting the cell membrane), benzalkonium chloride (a quaternary ammonium compound, disrupting the cell membrane), triclosan (a bisphenol, blocking of lipid biosynthesis) and formaldehyde (an aldehyde, alkylating agent) [4, 5]. However, several studies have reported bacterial resistance to various biocides due to the presence of resistance genes [6]. On the other hand, the risk of P. aeruginosa infections in ICU-hospitalized patients is high (up to 30%) despite applying hygiene programs [5]. Such a high prevalence can be attributed to the reduced susceptibility of P. aeruginosa strains against a variety of antiseptics and disinfectants over time [7]. Another problem is the emergence of multi-drug resistant (MDR) P. aeruginosa strains, which can lead to treatment failure [3]. Interestingly, P. aeruginosa strains have shown cross-resistance to biocides and antibiotics with probably similar mechanisms, thereby making bacterial elimination from hospital environments difficult [5, 8]. Therefore, obtaining information on the prevalence and mechanisms of bacterial resistance to antimicrobial agents and the selection of suitable biocides and antibiotics can be helpful for controlling hospital-acquired infections caused by P. aeruginosa. Several mechanisms of resistance to antibiotics and tolerance to biocides have been identified in P. aeruginosa strains including: 1) simple growth requirements and ability of biofilm formation, 2) enzymatic degradation, 3) target site modification, 4) outer membrane impermeability, and 5) presence of efflux pumps. These properties are involved in bacterial persistence in medical environments and development of nosocomial infections [2, 9,10,11,12,13,14]. Among these resistance mechanisms to biocides, efflux pumps encoded by qacEΔ1, qacE, qacG and cepA genes play an important role in P. aeruginosa resistance to benzalkonium chloride and chlorhexidine [14]. On the other hand, fabV gene confers high-level triclosan resistance. Various compounds are used in hospitals and non-hospital environments in Ardabil city but there is no exact data on the rate of biocide effectiveness against local isolates of P. aeruginosa. Moreover, the mechanisms of biocide resistance in P. aeruginosa clinical isolates of Ardabil are still unclear. Therefore, the aim of the current study was to assess the distribution of biocide resistance genes, namely qacEΔ1, qacE, qacG, fabV, cepA and fabI, and to determine the minimum inhibitory concentration (MIC) and the epidemiological cut-off (ECOFF) values of various biocides against antibiotic-resistant P. aeruginosa strains isolated from clinical samples in Ardabil.

Methods

Data on P. aeruginosa isolates

A total of 76 confirmed clinical P. aeruginosa isolates were obtained from various specimens in Ardabil hospitals and then used to assess the distribution of biocide resistance genes as well as determination of the MIC and the ECOFF of four biocides. Drug resistance characteristics of P. aeruginosa isolates was evaluated using the disk diffusion method based on the Clinical and Laboratory Standards Institute (CLSI, 2018) guideline [15]. Data on the prevalence of class I integron, harboring resistance genes to biocides and antibiotics, was used in this study in order to evaluate the association with resistance to biocides. It is noteworthy, the prevalence of antibiotic resistance along with class I integrin rate were previously determined by authors [16, 17].

Preparation of biocide solutions

Biocides used in this study were chlorhexidine digluconate (20%) (Sigma-Aldrich, USA), benzalkonium chloride (>95%) (Sigma-Aldrich, USA), triclosan (98%) (Bio Basic, Canada) and formaldehyde (37%) (Thermo Fisher Scientific, USA). Stock solutions of antimicrobial agents were prepared in distilled water or water-alcohol for water-insoluble antimicrobial agents. All antibacterial solutions were sterilized using sterile syringe filters (0.22 µm) before use.

Determination of the MICs of biocides

The MICs of antiseptics and disinfectants including chlorhexidine digluconate, benzalkonium chloride, triclosan and formaldehyde were determined by agar dilution technique and according to the CLSI guideline [15]. The CLSI and similar organizations have not been defined a standard protocol describing bacterial resistance or susceptibility against non-therapeutic antimicrobials in susceptibility tests. Therefore, estimation of the ECOFF value for each biocide was done based on the MIC distributions in vitro. For this purpose, at first a range of biocide concentrations (0.125-1024 μg/mL) was prepared in Mueller-Hinton agar medium and then a 0.5 McFarland standard concentration of P. aeruginosa isolates (1.5 × 108 CFU/mL) was prepared in normal saline. Finally, diluted bacterial inoculum (1:10) (1.5 × 107 CFU/mL) was poured onto medium containing different biocides as a spot (104 CFU per spot). The plates were incubated at 37 ˚C overnight and then checked in terms of bacterial growth. In the current study, the ECOFF value for the antibacterial susceptibility testing against four biocides was determined equal to MIC95 (95% rule). Therefore, P. aeruginosa strains with higher MIC value compared with ECOFF value were considered as non-wild-type isolates, i.e. organisms with detectable resistance and reduced susceptibility for each biocide.

Detection of biocide resistance genes

Extraction of total genomic DNA was done using a simple boiling method according to previous instructions [18]. DNA quantity and quality was controlled using NanoDropTM 2000/2000c Spectrophotometer (Thermo Fisher Scientific, USA) and then validated using sequencing of amplified genes. Extracted DNA was stored at -20 ˚C until use for detection of biocide resistance genes. The presence of qacEΔ1, qacE, qacG, fabV, cepA and fabI genes was detected by specific primers in polymerase chain reaction (PCR). According to the thermal cycling condition for amplification and the oligonucleotide primer sequences presented in Table 1, PCR was performed in a volume of 25 μL containing 20 μL of master mix (Ampliqon, Denmark), 1 μL of each primers (10 μmol/L) and 3 μL of extracted DNA. PCR products were detected by electrophoresis on 1% agarose gel in a TBE 0.5x buffer and then confirmed by sequencing technique. In addition, a fabI gene positive Acinetobacter baumannii clinical isolate was used as positive control. Also, in this study, identified positive isolates for qacEΔ1, qacE, qacG, fabV and cepA genes were used for quality control.

Table 1 Used primers and PCR programs in the present study
Full size table

Data analysis

All data on the MICs of biocides against P. aeruginosa strains, the presence of resistance genes to biocides and P. aeruginosa resistance to various antibiotics were collected and their correlation analyzed by the SPSS software version 16. The Chi-square test was used to analyses and a p value of <0.05 was considered statistically significant.

Results

In the present study, among 76 P. aeruginosa clinical isolates, the prevalence of qacEΔ1, qacE, qacG, fabV, cepA and fabI genes were 73.7% (n=56), 26.3% (n=20), 11.8% (n=9), 84.2% (n=64), 81.5% (n=62) and 0% (n=0), respectively. As shown in Table 2, the MIC range for various biocides was as follows: benzalkonium chloride 256-1024 μg/mL, formaldehyde 32-512 μg/mL, triclosan 32-512 μg/mL and chlorhexidine digluconate 4-64 μg/mL. In total, the highest MIC90 was observed for benzalkonium chloride (MIC90=1024 μg/mL), followed by formaldehyde (MIC90=512 μg/mL), triclosan (MIC90=512 μg/mL) and chlorhexidine digluconate (MIC90=64 μg/mL), Therefore, it seems that chlorhexidine digluconate and benzalkonium chloride had the highest and the lowest effects, respectively, in terms of growth inhibition of P. aeruginosa isolates in this study. As shown in Table 3, a significant association was observed between the presence of biocide resistance genes and MICs (p<0.05). Additionally, biocide resistance gene profiles revealed that isolates simultaneously harboring qacEΔ1, cepA and fabV genes are more prevalent (n=35, 46%), while 8 isolates (10.5%) did not harbor any biocide resistance genes (Table 4). Moreover, strains containing biocide resistance genes had higher MIC50 and MIC90 values compared with isolates without these genes. In this study, ECOFF value for the reduced susceptibility to benzalkonium chloride, triclosan, formaldehyde and chlorhexidine digluconate were determined as 1024, 512, 512 and 64 μg/mL, respectively.

Table 2 Minimal inhibitory concentration of biocides against clinical isolates of P. aeruginosa
Full size table
Table 3 Association between biocide resistance genes and MIC
Full size table
Table 4 Gene profile and minimal inhibitory concentration of biocides
Full size table

Isolates with and without biocide resistance genes were compared in terms of resistance rate to the following antibiotics: piperacillin, piperacillin-tazobactam, ticarcillin-clavulanate, ceftazidime, cefepime, aztreonam, doripenem, imipenem, meropenem, gentamicin, tobramycin, amikacin, netilmicin, ciprofloxacin, levofloxacin, norfloxacin, lomefloxacin and ofloxacin. There was no significant association between the presence of biocide resistance genes and antibiotic resistance (p>0.05), except for levofloxacin and norfloxacin antibiotics and qacE and qacG genes (p<0.05). The prevalence of integron I positive P. aeruginosa strains harboring qacEΔ1 gene was 32 out of 76 (42.1%). No significant association was observed between the presence of class I integron and biocide resistance genes (qacEΔ1, qacE, cepA and fabV), except for qacG gene (p = 0.00).

Discussion

Biocides are basic compounds to control microorganism dissemination and ensuing infections and are frequently used as preservative, disinfectant and sterilizer against various microorganisms, in particular P. aeruginosa [20, 21]. One of the most useful biocides against microbes, especially Gram-positive bacteria, is triclosan, which is widely used in toothpastes, soaps and other daily products [20]. Triclosan is an anionic and lipophilic compound, which its anti-bacterial function stems from inhibition of enoyl-acyl-carrier protein reductase (ENR), an enzyme involved in fatty acid synthesis [22,23,24]. However, P. aeruginosa strains are inherently resistant to this biocide (MIC>2000 μg/mL) [24]. ENR enzymes show diversity among different bacteria in terms of sequence and structure, and contain four isozymes including FabI (triclosan- sensitive ENR), FabL, FabV and FabK (triclosan-resistant ENRs) [22,23,24]. The FabV isozyme is involved in swimming motility, energy metabolism, protein secretion and adherence, and is responsible for P. aeruginosa resistance to triclosan biocide [24]. Genes encoding FabI and FabV enzymes are found in most bacterial chromosomes such as P. aeruginosa [22,23,24]. Bacterial resistance to triclosan is associated with mutation in the active site of fabI gene and the presence of fabV gene [20, 24]. In the current study, the prevalence fabI resistance gene among P. aeruginosa isolates was 0%. Unlike fabI gene, the frequency of fabV gene was high in the present study (84.2%). Zhu et al. showed that deletion of fabV gene confers extremely high susceptibility to triclosan (>2,000 folds) in P. aeruginosa isolates [22]. Similar result was reported by Huang et al. [24]. In this study, the MIC50 and MIC90 values of triclosan for fabV resistance gene- harboring P. aeruginosa strains was higher than fabV gene-negative strains (Table 4).

Chlorhexidine digluconate, an antiseptic, disinfectant and preservative, is a bactericidal biocide, which has higher antibacterial activity against Gram-positive compared with Gram-negative bacteria [25]. This biocide is used in oral health antiseptics, hand washes and other hygienic solutions. The antibacterial mechanism of chlorhexidine digluconate is via the bacterial cell membrane [20]. However, P. aeruginosa is intrinsically resistant to this biocide due to the presence of an outer membrane [25]. Adaptive resistance to chlorhexidine biocide is mediated by a membrane protein encoded by Acinetobacter chlorhexidine efflux gene (aceI). The AceI protein identified in Acinetobacter baumannii is involved in chlorhexidine efflux via an energy-dependent mechanism [26]. However, genes encoding this protein were not identified in P. aeruginosa strains in the current study (data not shown). The antiseptic resistance gene cepA, an efflux pump gene, is associated with chlorhexidine resistance in Gram-negative bacteria causing high chlorhexidine MICs [27, 28]. In our study, 62 (81.5%) cepA-positive strains were found, which is higher than those reported by Mendes et al. (44.5%) and Vijayakumar et al. (63.6%) [27, 28]. According to MIC results, chlorhexidine digluconate is more effective than other biocides against P. aeruginosa isolates (MIC range=4–64 μg/mL) (Table 2). In this study, the presence of cepA gene had variable effects on the MIC50 and MIC90 values of chlorhexidine (Table 4).

A major biocide resistance mechanism in Gram-negative bacteria including P. aeruginosa is the action of efflux pumps such as the small multidrug resistance family (SMR) [14, 21]. Biocide resistance genes qacEΔ1, qacE and qacG encode multidrug efflux pumps, which confer resistance to quaternary ammonium compounds like benzalkonium chloride [14, 21]. In our study, the qacEΔ1 gene was observed in 73.7% of clinical isolates of P. aeruginosa, while in studies conducted by Subedi et al., Roma˜o et al., Kücken et al., Helal et al. and Mahzounieh et al. the qacEΔ1 gene was detected in 46.1%, 48%, 10%, 48% and 91.5% of the isolates, respectively [14, 21, 29,30,31]. According to the reports of Subedi et al., Kücken et al., Helal et al. and Mahzounieh et al., 100%, 2.7%, 13.5% and 50% of P. aeruginosa strains, respectively, had the qacE gene [14, 29,30,31], while we detected this gene in 26.3% of isolates. The frequency of qacG gene in the present study was 11.8%, which is higher compared to the frequency reported by Subedi et al. (0%) [14]. The MIC50 and MIC90 values of benzalkonium chloride were significantly high for qacEΔ1-, qacE- and qacG-positive P. aeruginosa strains compared with the negative strains (Table 4).

Class I integron carries qacEΔ1 and antibiotic resistance genes in clinical isolates of P. aeruginosa [14]. Therefore, P. aeruginosa strains harboring class I integron are resistant to benzalkonium chloride and various antibiotics [29]. Comparison of our current and previous study [17] showed that the frequency of integron I-positive P. aeruginosa strains harboring qacEΔ1 gene was 32 out of 76 (42.1%). No significant association was observed between the presence of class I integron and biocide resistance genes (qacEΔ1, qacE, cepA and fabV), except for qacG gene (p = 0.00).

Formaldehyde is an organic electrophilic biocide, which its mechanism of action involves cross-linking of macromolecules (proteins, RNA and DNA) [20, 32]. Our results indicated that the MIC50 and MIC90 values of formaldehyde were high for biocide resistance genes-positive P. aeruginosa strains compared with the negative strains (Table 4).

A study by Chuanchuen et al. showed a cross-resistance between biocide and antibiotic resistance. They demonstrated a link between P. aeruginosa exposure to triclosan biocide and efflux-mediated resistance to ciprofloxacin [13]. In the present study, there was no significant association between biocide resistance genes and antibiotic resistance, except for levofloxacin and norfloxacin antibiotics and qacE and qacG genes. However, more studies are needed to substantiate the existence of biocide-antibiotic cross-resistance.

Conclusion

Our results revealed that the frequency of resistance genes to benzalkonium chloride, chlorhexidine digluconate, triclosan and formaldehyde was high in clinical isolates of P. aeruginosa. Furthermore, P. aeruginosa isolates harboring resistance genes had higher MIC values compared with those lacking these genes. On the other hand, chlorhexidine digluconate was the most effective biocide against P. aeruginosa isolates in Ardabil hospitals. We recommend continuous monitoring of the antimicrobial activity of biocides and biocide-associated resistance genes in order to prevent microorganism dissemination and infection control in hospitals.