Phylogenetic Aspects of Antibiotic Resistance and Biofilm Formation of P. aeruginosa Isolated from Clinical Samples

Introduction Biofilm production and drug resistance phenomenon play a critical role in P. aeruginosa infections. Several genes, including psl, pel, brlR, and mex, are involved in the phenomenon. The aim of this study was to find the relationship between the mentioned genes and the sources of P. aeruginosa infections. Materials and Methods Fifty-nine P. aeruginosa isolates detected from clinical specimens were used to determine antibiotic susceptibility patterns, prevalence of the genes using PCR, biofilm formation, biofilm eradication concentration assay (MBEC), and epidemiological characteristics using pulsed-field gel electrophoresis (PFGE). Results The results showed that 35.6% and 16.94% of all the samples were isolated from urine and wounds, 81.33% of the isolates were biofilm producers, 27.11% were multidrug-resistant (MDR), and 100% of the main biofilm former genes belonged to pslA. 94.91% of the isolates possessed brlR and mexA, and 91.5% of them expressed pslA. It was also indicated that neither ciprofloxacin nor imipenem could eradicate the formed biofilms. Moreover, we could identify 81.4% distinctive restriction profiles among the isolates, using an 80% similarity cutoff point; brlR and pel genes were significantly (P=0.032; P=0.044) related to phylogenetic pulsotypes. Comparison of the dendrogram in the isolates revealed that the detected isolates from urine were present in 12 different pulsotypes. Conclusion It was found that there was a relationship between MDR, biofilm production, and brlR and pel genes among the isolates. It is distinguished there were similar genetic patterns between detected isolates from urine and could be concluded that the urinary tract played a critical role in maintaining and transferring biofilm drug-resistant genes of P. aeruginosa in clinical sites. The study highlights the importance of urine in distribution of clinical biofilm formation and drug-resistant P. aeruginosa isolates.


Introduction
P. aeruginosa is a nosocomial MDR pathogen that poses a serious threat to public health [1,2].Bioflm-associated infections caused by P. aeruginosa are particularly challenging, as bacteria embedded in bioflms are more resistant to antibiotic treatments.Te bioflm matrix of P. aeruginosa is composed of proteins, DNA, lipids, alginate, and essential exopolysaccharides such as pel and psl.Ma et al. demonstrated that the psl gene plays a critical role in adherence of cell surfaces [3].Bioflm production by P. aeruginosa facilitates its persistence in the environment and can lead to severe bacterial infections, especially in hospitalized patients [4].However, in vitro bioflm production among clinical strains varies, and P. aeruginosa can be classifed as strong, moderate, weak, or nonbioflm producers [5,6].
Bioflm production and MDR P. aeruginosa infections are particularly problematic in immunocompromised individuals with respiratory, urinary tract, or chronic wound infections [7,8].Te brlR gene contributes to drug tolerance and efux pump in the formed bioflm of P. aeruginosa, while the mex operon is responsible for MDR P. aeruginosa [9,10].
To investigate the relationship between drug resistance and bioflm formation in clinically recovered P. aeruginosa isolates, there is a need to fnd epidemiological similarity among the isolates.Molecular epidemiological typing techniques, such as PFGE, can be used for following the "transmission route" and "clonal analysis" of clinical bacterial infections [11].
In this study, the clinically recovered P. aeruginosa isolates were examined for susceptibility to common antibiotics, recognition of MDR strains, bioflm formation, the in vitro efect of diferent antibiotic concentrations on bioflm formation, and the prevalence of fve genes: pslA, pelA and pelB, brlR, mexA, and mexB, and fnally epidemiological relationship between the isolates.

Materials and Methods
2.1.Bacterial Isolates.A total of 59 P. aeruginosa isolates were collected from clinical samples using standard diagnostic bacteriological procedures [12].Te isolates were recovered from infectious sites such as blood, body fuids, sputum, urine, and wounds of hospitalized patients.

Molecular Identifcation of P. aeruginosa Isolates 2.2.1. DNA Extraction.
To confrm the molecular identifcation of the isolates, genomic DNA was extracted from all isolates using a simple boiling method.Briefy, a few bacterial colonies were suspended in 200 µL of molecular grade water and heated for 15 min at 95 °C, followed by centrifugation at 14000 × g for 10 min.Te supernatant containing DNA was used for PCR amplifcation.

Molecular Confrmation of the Isolates as P. aeruginosa.
PCR primers designed in this study were employed to amplify the sequence of the genes of the isolates.Amplifcation was carried out in a total reaction volume of 25 μL containing 2.5 μL 10x PCR bufer, 1.5 mM MgCl 2 , 0.2 mM deoxynucleotide, 0.4 pmol/µL of each primer (Bioneer, Seoul, South Korea), 0.2 U Taq DNA polymerase, and 3 µL DNA.Te thermocycler (Analytik Jena, model: Flex Cycler 96G) was set with the following conditions: initial denaturation for 2 min at 95 °C, followed by 25 cycles of denaturation for 20 s at 94 °C, annealing for 20 s at 58 °C, an extension for 40 s at 72 °C, and fnal extension for 1 min at 72 °C.Electrophoresis was performed on a 1.5% agarose gel along with GeneRuler 100 bp DNA Ladder (Fermentas, Lithuania) and stained with 0.5 μg/mL ethidium bromide.
P. aeruginosa ATCC 27853 was used as a positive control in all experiments.

Identifcation of Virulence Genes of Pseudomonas aeruginosa.
To determine the pathogenic potential of P. aeruginosa strains among the isolates, PCR amplifcation was carried out using specifc primers for fve virulence genes pslA, pelA and pelB, brlR, mexA, and mexB (Table 1).

Broth Dilution
Method: MIC Determination.Te minimum inhibitory concentration (MIC) of ciprofoxacin and imipenem was determined using the microbroth dilution method as described by NCCLS [14].MIC was defned as the lowest concentration of antibiotic giving complete inhibition of visible growth.

P. aeruginosa Isolates' Bioflm.
In vitro bioflm formation of the P. aeruginosa isolates, wildtype PAO1 and ATCC 27853 were used for culture, bioflm formation, and antibiotic-resistant bioflm experiments.Bioflm formation was assessed using a simple microtiter-plate assay in triplicates, and optical density was measured at 570 nm with an established cutof value (ODc) for demonstrating strong, moderate, and weak bioflm formation [2,15].

Bioflm Eradication Concentration Assay (MBEC) of
Imipenem and Ciprofoxacin against Formed Bioflm.Te minimum bioflm eradication concentration (MBEC) values of ciprofoxacin and imipenem against P. aeruginosa bioflms were determined [16].Briefy, the isolates were grown in 96well microplates for 24 h at 37 °C in the Mueller-Hinton (MH) medium and then washed twice with 0.9% NaCl to remove the planktonic cells.Te attached bioflms were then subjected to treatment with diferent concentrations of ciprofoxacin and imipenem at 2-fold dilutions from 2048 μg/mL to 2 μg/mL for each antibiotic.After incubation for 24 h at 37 °C, the microplates were washed with 0.9% NaCl to remove any antimicrobial residue.Te bioflm was scraped from the microplate wells and transferred into a sterile tube containing tryptic soy broth (TSB).Five microliters of the suspensions was cultured on the Mueller-Hinton agar plate and then incubated overnight at 37 °C.MBEC values were determined to be the lowest antibiotic concentrations that prevented the bacterial growth from the treated bioflm.

Phylogenetic Assay: DNA Genomic Typing by Pulsed-Field Gel Electrophoresis (PFGE).
Using the highly discriminating method and according to the criteria developed by Durmaz et al. [17], briefy, examined "cell suspension preparation, plug preparation, lysis of cells in agarose plugs, restrictive digestion of DNA in agarose plugs, casting of the agarose gel, electrophoresis," and restriction analysis; then, PFGE (PFGE CHEF-DR III System, Bio-Rad) of DNA was performed, and the band profles of DNA were analyzed by using BioNumerics software (version 7.0, Applied Maths, Sint-Martens-Latem, Belgium).We analyzed 59 clinical isolates of P. aeruginosa recovered from patients, and one P. aeruginosa ATCC 27853 strain was used as a control.For molecular 2 Canadian Journal of Infectious Diseases and Medical Microbiology analysis of P. aeruginosa strain isolates, PFGE was performed according to the mentioned protocol.Salmonella serotype Branderup strain (H9812) ladder (Bio-Rad Laboratories) restricted with XbaI was used as a universal size marker.SpeI (MBI Fermentas, Vilnius, Lithuania) restriction enzyme was used to generate the banding patterns for the P. aeruginosa clinical isolates as well as the standard controls.Te protocol included cell suspension preparation, plug preparation, lysis of the cells in agarose plugs, restrictive digestion of DNA in agarose plugs, casting of the agarose gel, electrophoresis, and restriction analysis.Te band profles of DNA were analyzed using BioNumerics software (version 7.0, Applied Maths, Sint-Martens-Latem, Belgium), and the developed criteria by Durmaz were followed.

Statistical Analysis.
Statistical analysis was performed using the chi-square test through SPSS 16.P values <0.05 were considered statistically signifcant.Te results, after then, were compared with PFGE band profles gained from the similarity of examined specimens.

Characteristics of P. aeruginosa Isolates.
Te collected isolates were confrmed as P. aeruginosa by specifc PCR.Of 59 clinical samples, 21 (35.6%) and 10 (16.94%)P. aeruginosa isolates were recovered from urine and wound specimens, respectively.Te remaining isolates were obtained from other sites of infection (Table 2).Te antibiotic susceptibility patterns of P. aeruginosa isolates indicated that 16 (27.11%)were multidrug-resistant, with the highest number MDR recovered from the wound (11.76%) and urine (8.48%) samples (Table 2).

Te Most Common Combination of Resistance Was
Found to be CP, IPM, and CRO.Sixteen isolates showed resistance to three or more antibiotics with 8 (50%) isolates exhibiting resistance to three antibiotics and the other 8 (50%) isolates exhibiting resistance to more than fve antibiotics.Overall, 16 isolates were classifed as multiresistant as they were resistant to three or four diferent antibiotic classes.

Prevalence of Virulence Genes among P. aeruginosa
Isolates.Te distribution of fve virulence genes, namely, pslA, pelA and pelB, and brlR of P. aeruginosa bioflms, as well as mexA and mexB, varied depending on the site of infection (Table 3).Table 3 shows that the frequencies of virulence gene occurrence in all the isolates were as follows: mexA, brlR, pslA, pelA and pelB, and mexB, respectively.
When compared with the isolates from other specimens, the majority of the strains producing a strong bioflm belonged to wound infection (70%) and urinary tract infection (52.38%).Furthermore, 100% of wound isolates and 85.71% of urine isolates were bioflm producers (Table 4).
During the EMBEC test, we found that ciprofoxacin was unable to eliminate bacterial bioflms at any concentration ranging from 2 to 2048 μg/mL.While for some strains, a reduction in bioflm formation was observed when using imipenem at a concentration of >512 µg/mL, and it was still unable to eradicate the established bioflm.

Genotyping by PFGE.
Using this method, we could identify 13 distinctive restriction profles, each with unique banding patterns among 48 P. aeruginosa isolates.Tis corresponded to a cutof value of 80% identity for PFGE, and the PFGE profles typically contained 12-17 bands.Te distribution of the number of fragment diferences is shown in Figure 1.However, the reaming 11 isolates had their own unique patterns that were not categorized into any of 13 distinctive profles (Table 5).Te P. aeruginosa ATCC 27853 control strain was also typed.Te distribution of virulence genes among the 13 distinctive clones and 11 single clones is shown in Table 5.

Discussion
Drug-resistant pathogenic bacteria have created numerous problems in hospitals and patient care centers.Not only can opportunistic pathogenic bacteria, such as P. aeruginosa, can be drug-resistant and infective but also drug-resistant environmental Pseudomonas sp. can infect many accidental or traumatic patients [18].Meng et al. expressed concern about   4 Canadian Journal of Infectious Diseases and Medical Microbiology antibiotic resistance in Pseudomonas spp.detected from "bulk-tank milk."Surprisingly, they found that the bacteria were highly resistant to several important clinical antibiotics, such as imipenem (95.3%) and aztreonam (60.5%).Teir mainly intrinsic resistance virulence and drug efux pump genes in the bacteria [19] may pose a signifcant threat to healthcare units and hospitals that admit susceptible patients.Currently, P. aeruginosa is the most signifcant challenge in the treatment of hospitalized Pseudomonas infections.Moreover, a relationship is believed to exist between drug resistance and the bioflm formation phenomenon in the P. aeruginosa isolates.Tere are many genes in the bacterium that play a role in drug resistance and/or bioflm formation.In the present study, we investigated several genes such as mex and brlR involved in drug resistance and such as pslA, pelA and pelB, and brlR genes of clinical P. aeruginosa isolates in bioflm formation.
Of 59 clinical isolates of P. aeruginosa obtained from several specimens, 35.6% of the isolates were recovered from urine, 25.42% from sputum, 16.64% from wound, 10.16% from blood, and 11.88% from other specimens.In another research, Karruli reported the frequency ratio of isolates from sputum, urine, surgery sites, and blood specimens separately.
As shown in Table 2, 27.11% of our isolates were MDR, which was similar to Sezadehghani's study which was 29.35% and Arta Karruli's report which was 33.1%.According to the European Antimicrobial Resistance Surveillance Network (EARS-Net), 9.7% of P. aeruginosa isolates were resistant to at least three of fve antimicrobial groups and 3.1% were resistant to all fve antimicrobial groups under surveillance [20].
Moreover, in our study, 70% of all the isolates from wound specimens were MDR; in a similar report from India, it was shown that 76.8% of the isolated P. aeruginosa were detected from burn patients [21].However, in a report from the United States, MDR P. aeruginosa isolates, with prevalent carbapenem resistance, were detected from sputum and blood of patients.Furthermore, a relatively high proportion of resistance to ceftazidime and cefepime (60%) was detected in our isolates from wound infections as well.It is almost diferent from Batt's fndings that show that 76.79% and 71.4% of the burns' isolates were resistant to ceftazidime and cefepime separately [21].Overall, in our study, most resistance was against ceftriaxone (45.76%) and then imipenem (45.76%), while in India, it was against ceftazidime (76.70%) and tobramycin (75%) [21]; in Europe, it was seen that most resistance was against fuoroquinolones like ciprofoxacin (19.6) and then piperacillin-tazobactam (18.8%) [22].
In the presence of environmental stresses, or under prolonged exposure to subinhibitory antibiotic concentrations, the majority of bacteria exist in surface-adherent bioflm.Bioflm formation, as an important virulence factor, plays a critical role in the pathogenesis of P. aeruginosa.It can protect bacteria from immune system clearance and from antibiotics [23].In our study, analyses of in vitro bioflm production by clinical isolates of P. aeruginosa indicated that more than 81.33% of the isolates were bioflm producers in strong, moderate, and weak forms.Interestingly, 100% of the detected P. aeruginosa isolates from the wound specimen and 85.71% from the urine specimens were bioflm producers compared with the bioflm producers isolated from other infection sites in this study.Overall, the results of our study indicated that 50.84%, 23.72%, and 6.77% of the bioflm forming P. aeruginosa strains produced strong, moderate, and weak bioflms separately.Dolatshah and Tabatabaei in their investigation found that 69% of the P. aeruginosa isolates were bioflm producers, but only 9% of them formed strong bioflm, and the other strains were moderate and weak bioflm producers [24].
Tere are various environmental and genetic factors that can infuence the bioflm formation of P. aeruginosa, but certainly psl and pel genes contribute to bioflm formation [25,26].Also, there are several genes such as mexA and mexB involved in drug resistance in P. aeruginosa, which act as the drug efux pump, and brlR contributes to the highlevel drug tolerance of P. aeruginosa bioflms [9,10], as followed in our study.We found that pslA gene was detected in 91.5% of the isolates, while pelA and pelB genes were found in 69.49% of the P. aeruginosa isolates.In a similar study, Dolatshah and Tabatabaei reported that "92% of the isolates possessed the pslA gene" [24].However, Ghadaksaz showed that 83.7% of their isolates had psl, and 45.2% of them possessed pelA genes, which was diferent from our research.Based on our fnding, we can conclude that the pslA gene was more effective in the bioflm formation of the isolated P. aeruginosa compared to the pel gene.
Although Márió Gajdács et al. found no relationships between bioflm formation and virulence factors of P. aeruginosa, in our study, the researchers investigated the relationship between bioflm formation and several virulence factors such as pigment production and motility, not the pslA gene under study.
Płókarz et al. have expressed that pslA and pelA play 60.1% and 38.7% roles in bioflm formation of P. aeruginosa in animals, respectively [27].Ghadaksaz et al.' study also showed a signifcant relationship between pslA and pelA genes and bioflm formation [28].Tese recent reports validate our fndings.
In this study, the frequency of mexA, mexB, and brlR genes, which are involved in resistance phenomenon, effux pump, and bioflm drug resistance in P. aeruginosa, was investigated.Table 3 shows that the distribution of the three genes was 94.91%, 50.84%, and 94.91%, respectively.Bhandari et al. reported that 70.6% of the isolates from all of the examined P. aeruginosa isolates possessed mexA and mexB which are involved in the efux pump [29].Based on our data, we can conclude that the mexA gene in our study played a critical role in causing drug resistance compared to mexB.
Te brlR gene, as a member of the MerR family, "is expressed only in bioflms" and plays a signifcant role in drug tolerance in P. aeruginosa.It is found that the brlR gene acts as a drug efux pump to resistance of antibiotics against P. aeruginosa [30].
In our study, brlR (94.91%) played an essential role in causing resistance to the formed bioflms.Our examined microbroth dilution tests for determining MIC and MBC levels against the two important antibiotics, imipenem and ciprofoxacin, revealed that none of the antibiotics were able to prevent bioflm formation at a concentration of <512 µg/mL.For some strains, a decrease in bioflm formation was observed when imipenem at a concentration of >512 µg/mL was used, but it was not able to eradicate bioflm formation.Musafer et al. showed that there was a strong link between bioflm formation and imipenem resistance in P. aeruginosa isolates [31].
Ciprofoxacin was not able to eradicate the bacteria that formed bioflms at any concentration in our study.In a similar report by Takrami and Ranji, it was observed that the P. aeruginosa isolates were resistant to ciprofoxacin due to the overexpression of mexAB operon [32].
For P. aeruginosa phenotyping and genotyping, diferent techniques have been developed; however, PFGE is still considered to be the gold standard molecular typing method for genotyping P. aeruginosa which has shown a high index of diversity [11].
According to Figure 1, the rate of genomic similarity, depending on using an 80% similarity cutof point, reveals that PFGE pattern (pulsotypes) analysis resulted in 13 distinctive pulsotypes (81.35%) and 10 unique singletons (18.6%) of P. aeruginosa isolates.Karami et al. similarly used PFGE to genotype P. aeruginosa isolates and found a high level of diversity, with 22 pulsotypes identifed [33].
Overall, a review of the total results in this study reveals that 35.6% of all the samples were isolated from urine and 16.94% from wound.Among the total MDR isolates (27.11%), 8.48% of the isolates were detected from urine and 11.86% from wound.85.71% of the detected isolates from urine and 100% from the wound specimens were able to form bioflm, so that 52.38% and 70% of the formed bioflms separately were strong.Furthermore, Awan et al. reported a signifcantly higher prevalence of strong bioflm formation in MDR P. aeruginosa isolates when compared to our fndings [34].
Figure 2 and Table 6 demonstrate that in PFGE analysis, pulsotype 2 with 8 (13.6%) isolates was predominant among the others followed by pulsotypes 8, 11, and 7.In Nikbin et al.'s study, 84 resistance banding patterns were detected from two hospitals; there was the dominant pattern of 14 isolates at both hospitals [35].
Comparison of the dendrogram of the isolates revealed that isolated P. aeruginosa detected from urine specimens was present in 12 diferent pulsotypes, while the isolates recovered from wound specimens existed in only 6 separate pulsotypes.Based on the mentioned concepts, it may be concluded that bioflm formation causes a urinary tract infection.In the same direction, Uner-Kayabas announced that P. aeruginosa usually infects the urinary tract by strong adherence to the bladder uroepithelium, and P. aeruginosa usually afects the urinary tract through ascending infection and adheres strongly to the bladder uroepithelium.
Finally, according to our fndings in urine exams, 100% of the main responsible genes in bioflm formation belonged to pslA and 95.23% of the MDR genes were mexA.We can conclude that the urinary tract can play a key role in maintaining and transferring bioflm drug-resistant P. aeruginosa.Tis is in the same line with Cholley's opinion that believed bacterial "cross-transmission" between infected patients was responsible for the isolated MDR presentation [36].
We believe that future studies on UTI among people prone to urinary tract infections, such as young children, pregnant and older individuals, or kidney failure patients, should be focused on the pattern of their genetic exchange, bioflm formation, and drug resistance in their UTI recurrence.It is also important to follow the patients to ensure successful treatment and negative growth of their urine culture to control the recurrence of the infection.For this purpose, it is suggested two urine cultures should be performed in prospective studies [37].Briefy, a urine culture and an antibiotic susceptibility test, for detected bacteria, should be performed four days after starting the prescribed choice of antibiotics and ten days after stopping them, in cases of positive urine cultures.If the antibiotic therapy proves efective in both steps, it signifes a suitable treatment and reduces the chances of recurrent infection.Positive

Limitation.
Te main limitation of this study was the lack of compatibility between in vitro and in vivo experimental conditions.However, considering the urgency in treating drug-resistant infectious diseases in humans, we have no choice but to rely on laboratory experiments.

3. 6 .
Clonal Analysis.A dendrogram was constructed using BioNumerics software (version 7.0, Applied Maths, Sint-Martens-Latem, Belgium) to compare each of the PFGE banding patterns with all other banding patterns, and it demonstrated the rates of genomic similarity (Figure1).Using an 80% similarity cutof point, the analysis of PFGE patterns (pulsotypes) resulted in 13 pulsotypes and 10 unique singleton PFGE types.

6
Canadian Journal of Infectious Diseases and Medical Microbiology

Figure 2 :
Figure 2: Dendrogram and cluster analysis of P. aeruginosa isolates detected in the present study by pulsed-feld gel electrophoresis (PFGE).

Table 6 :
Comparison of the dendrogram of the isolated P. aeruginosa from clinical specimens.
(2), U.C (2), W.C (1), F.C(1) Canadian Journal of Infectious Diseases and Medical Microbiology urine cultures at any stage of the tests can indicate a recurrent urinary tract infection because of bioflm presence centers in the urinary tract, which is hazardous to patients and poses a risk to public health.

Table 2 :
Antibiotic susceptibility patterns of P. aeruginosa isolates recovered from clinical specimens.

Table 4 :
Distribution of bioflm formation of Pseudomonas aeruginosa in clinical specimens.

Table 5 :
Te distribution of the virulence genes among 13 distinctive clones and 11 single clones.