Evolutionary trajectories of beta-lactamase NDM and DLST cluster in Pseudomonas aeruginosa: finding the putative ancestor

ABSTRACT Pseudomonas aeruginosa has different antibiotic resistance pathways, such as broad-spectrum lactamases and metallo-β-lactamases (MBL), penicillin-binding protein (PBP) alteration, and active efflux pumps. Polymerase chain reaction (PCR) and sequencing methods were applied for double-locus sequence typing (DLST) and New Delhi metallo-β-lactamase (NDM) typing. We deduced the evolutionary pathways for DLST and NDM genes of P. aeruginosa using phylogenetic network. Among the analyzed isolates, 62.50% of the P. aeruginosa isolates were phenotypically carbapenem resistance (CARBR) isolates. Characterization of isolates revealed that the prevalence of blaNDM, blaVIM, blaIMP, undetermined carbapenemase, and MexAB-OprM were 27.5%, 2%, 2.5%, 12.5%, and 15%, respectively. The three largest clusters found were DLST t20–105, DLST t32–39, and DLST t32–52. The network phylogenic tree revealed that DLST t26–46 was a hypothetical ancestor for other DLSTs, and NDM-1 was as a hypothetical ancestor for NDMs. The combination of the NDM and DLST phylogenic trees revealed that DLST t32–39 and DLST tN2-N3 with NDM-4 potentially derived from DLST t26–46 along with NDM-1. Similarly, DLST t5–91 with NDM-5 diversified from DLST tN2-N3 with NDM-4. This is the first study in which DLST and NDM evolutionary routes were performed to investigate the origin of P. aeruginosa isolates. Our study showed that the utilization of medical equipment common to two centers, staff members common to two centers, limitations in treatment options, and prescription of unnecessary high levels of meropenem are the main agents that generate new types of resistant bacteria and spread resistance among hospitals.


Introduction
P. Pseudomonas aeruginosa infection is highly prevalent in hospital-acquired infections.P. aeruginosa can cause various infections, particularly in immunocompromised people.P. aeruginosa has different antibiotic resistance mechanisms, making it one of the most antibiotic resistant bacteria.The emergence of multidrug-resistant (MDR) P. aeruginosa has become a difficult public health issue [1,2].Beta-lactam resistance is one of the most important types of resistance in P. aeruginosa since carbapenems (as effective antipseudomonal drugs) were used as the last line in the treatment of MDR-P.aeruginosa infections.Penicillin and cephalosporin resistance among P. aeruginosa strains has become a worldwide clinical problem [3].In 2017, the World Health Organization (WHO) designated carbapenemresistant Acinetobacter, Pseudomonas, and Enterobacteriaceae strains as the most critical group of MDR bacteria.These bacteria are regarded as a high priority, demanding urgent attention [4].
P. aeruginosa has numerous mechanisms of betalactam resistance, such as broad-spectrum beta-lactamases and metallo-β-lactamases (MBL), penicillinbinding protein (PBP) alteration, porin mutations, and active efflux pumps.Carbapenemases constitute three classes of β-lactamases, including groups A and D (serine carbapenemases), and B of Ambler (zincdependent beta-lactamases).The group B Ambler enzymes require zinc as a co-factor, and their function is inhibited by metal chelators, such as EDTA and thiolbased products.Accordingly this class is called metallo-β-lactamases (MBLs) [5].The class B of Ambler can hydrolyze a wide range of beta-lactams such as penicillins, cephalosporins, and carbapenems.Metallo-β-lactamase-producing Pseudomonas isolates have been responsible for many nosocomial outbreaks in hospital centers in diverse world regions [4][5][6].Except for monobactams, MBL enzymes are capable of hydrolyzing all β-lactam antibiotics.These enzymes are encoded by bacterial chromosomes and mobile genetic elements [7].Imipenemase (IMP), Verona integron-borne metallo-β-lactamase (VIM), Guiana extended-spectrum β-lactamase (GES), and NDM are the most important enzymes that have attracted attention for clinical and epidemiological studies [8,9].MBLs producing P. aeruginosa were first reported in Japan in 1991 and have since been detected in various world regions, including Asia, Europe, Australia, South America, and North America [10].Another critical mechanism of MDR in P. aeruginosa is the efflux pump system.The genome of this bacterium includes at least 12 genes encoding RND efflux systems [11].The MexAB-OprM efflux pump, a common efflux pump, provides high levels of resistance to multiple antibiotics [12].The MexAB-OprM efflux pump can extrude various dyes, detergents, and antibiotics such as tetracyclines, beta-lactams, fluoroquinolones, macrolides, novobiocin, and trimethoprim [2,13].The epidemiological analysis methods for P. aeruginosa, such as multilocus sequence typing (MLST) and pulsed-field gel electrophoresis (PFGE), are timely, costly, and require technical expertise [14,15].The partial sequencing of two extremely inconsistent loci for P. aeruginosa typing (DLST) has been used to obtain a systematic and simple typing description [13][14][15].DLST has remarkable discriminatory power and reproducible capacity [15,16].There are also some other typing methods for epidemiological study; however, none of these methods are suitable to answer two questions: 1 -Where does the sample originate from? 2 -What is the origin of resistance?Accordingly, the following questions are addressed in this study: (1) Is there a strategy to discover the ancestor type (e.g. which other samples originated from it) among collected samples from patients?(2) Can we discover the evolutionary dependency among isolated P. aeruginosa by DLST and NDM typing?(3) Is it possible to create a system or algorithm to establish the association between prescription antibiotics, hospital departments, bacterial resistance, and bacterial type?
This study answers these concerns using the network phylogenetic tree (median joining network), and parsimony approaches of the Mesquite program.Our approach for analyzing bacterial epidemiology and resistance origins may be interesting for other scientists.

Bacterial isolation and identification
In a cross-sectional study, a total of 200 clinical samples from the two teaching hospitals (Al-Zahra and Burns centers) in Isfahan were isolated within 7 months, from September 2019 to March 2020.A total of 40 P. aeruginosa non-duplicate isolates were obtained from different medical specimens using traditional microbiologic methods and the genotypic approach (gyrB gene presence) [17,18].

Efflux phenotypic method
In this step, the efflux pump inhibitor carbonyl cyanide 3-chlorophenylhydrazone (CCCP) was performed to evaluate the antibiotic resistance phenotypes due to efflux pumps [20,21].In 96-well plates, the minimum inhibitory concentration (MIC) of antibiotics was measured using the broth micro-dilution technique based on CLSI 2022 [19].According to the manufacture guidelines, meropenem antibiotic powder was mixed in sterilized deionized water or a suitable solvent.The concentrations of 256, 128, 64, 32, 16, 8, 4, 2, 0.5, and 0.25 µg/ml were then created.A concentration of 1 × 10 5 CFU was obtained by diluting the bacterial suspension with sterile saline.Each test was performed with triplicate.The plates were incubated at 37°C for 18-24 hours.This method was repeated with a similar concentration of antibiotics and bacteria plus primary 100 µL of CCCP treatment at the first well and serial dilution to final well.After using the CCCP inhibitor, the fourfold decrease in the MIC was supposed to imply strong efflux pump activity.Table S2 shows the antibiotic list which should be tested for the MexAB-OprM efflux pump.

Efflux pump activity by EtBr cartwheel method
This method employed a set of trypticase soy agar (TSA) plates containing EtBr concentrations ranging from 0.0 to 2.5 mg/L.Overnight cultures of the bacterial isolates were organized in agar media.From overnight fresh bacterial culture, their concentration was adjusted to 0.5 McFarland tube.The TSA plates were then divided into as many as several sectors with the aid of radial lines, forming a Cartwheel pattern.The injected bacterial cultures were then swabbed at the EtBr-TSA plates, which began to grow from the center of the plate to the margin.
The TSA plates were then incubated at 37ºC for 16 hours.After this time, the TSA plates were examined below a gel-imaging gadget (or a UV Transil -Luminator); the minimum attention of EtBr that fluorescence of the bacterial mass was recorded; and the TSA plates were photographed [22].

Double disk synergy tests
This procedure was done to phenotypically investigate the metallo-β-lactamases existence.Imipenemresistant isolates were examined using the DDST, including imipenem and imipenem-EDTA, as reported by Lee et al. 2003 [23].The imipenem disc (10 µg) and EDTA (750 µg) treated disc with 10 mm distance were placed in MHA plate lawn culture of the Pseudomonas based on the CLSI guidelines.Then, the plates were incubated at 37°C for 24 h.An expansion inhibition zone in the imipenem and the EDTA discs compared with the inhibition zone of the imipenem (alone) would confirm the presence of metallo-β-lactamases.

DNA extraction
Bacterial strains were grown overnight at 37°C on blood agar, and DNA was extracted prior to PCR analysis by phenol-chloroform method for PCR reagent from new colonies that were described previously [24].

Identification of specific gyrB gene for P. aeruginosa
PCR assay and gel electrophoresis were applied to genotypically confirm the presence of P. aeruginosa.Table 1 displays gyrB primer for specific gyrB gene of Pseudomonas aeruginosa.The isolation and identification of Pseudomonas aeruginosa were proved by PCR amplification that were performed with gyrB specific primers.The amplicon length for gyrB was 222 bp that were confirmed by gel electrophoresis methods [25].

Identification of MexAB-OprM efflux pump genes
PCR amplification was carried out with a series of primer pairs to confirm the existence of the gyrB, MexA, MexB, and oprM genes.These primers have amplified a particular fragment length of genes (Table 1).The amplification program was as follows: initial denaturation at 94°C for 5 min, then 30 cycles of denaturation at 94°C for 45 sec, annealing at 58°C for 45 sec, extension at 72°C for 1 min, and then a final extension at 72°C for 7 min [27,28].

PCR amplification to detect metallo-β-lactamases genes
PCR amplification was carried out for the detection of bla genes using specific primers (Table 1) [27,28].The PCR assay in a thermal cycler was used in this analysis (Eppendorf AG, Germany).The amplification program was as follows: initial denaturation at 94°C for 4 min, then 30 cycles of denaturation of 95°C for 60 sec, annealing 58°C for bla IMP, 54°C for bla VIM , bla NDM for 60 sec, extension at 72°C for 1 min, and then a final extension at 72°C for 7 min [27,28].

Double-locus sequence typing method
DLST typing was done as described previously (Basset and 2014) [16].The full procedure for typing P. aeruginosa with DLST is available on the DLST website (http://www.dlst.org/).In brief, DNA extract has been used for PCR amplification from both ms172 and ms217 loci using specific primers.The PCR products have been purified and sequenced by Niagenenoor Corporation (Niagenenoor, Iran).Each strain was given two numbers and represented its DLST type, and if no result has been obtained for allele selection, null was to be considered [16].

Analysis of the amplicon sequencing
Sequencing outcomes for DLST and full-length NDM amplicons were analyzed with APE [29] program and QIAGEN CLC Genomics workbench 20.0 (https://digita

NETWORK phylogenetic tree generating
This analysis was performed to find the cluster and the ancestral Pseudomonas type.In the cases of the DLST, all DLST sequences were placed in the one.fasta file.This file was aligned in the MEGA program with the ClustalW multiple sequence alignment parameter [31]

Metallo-β-lactamases and efflux phenotypic methods
Among 40 P. aeruginosa isolates, 25 isolates were phenotypically resistant to meropenem, from which 14 of them were MBL producers, based on the synergy test among meropenem/imipenem with EDTA disc.2) were sensitive in the DDST assay, suggesting the presence of MBL in these isolates.Table 3 displays six isolates that were phenotypically susceptible to CCCP and meropenem.Figure 1 demonstrates the cartwheel results.The presence of efflux pump does not imply that the beta-lactam resistance was due to these genes, so the efflux pump activity should be further investigated; therefore, we tested the EtBr cartwheel method and efflux inhibitory effect with CCCP.Outcomes supported that the meropenem resistance in those six types, including Ps 13, Ps 17, Ps 35, Ps 47, Ps 67, and Ps 89 (with DLST t1-21, DLST t20-105, DLST t12-105, DLST t20-105, DLST t32-52, and DLST t32-39), belonged to the efflux pump activity.
According to efflux phenotypic inhibition with CCCP/meropenem, six isolates had efflux pump resistance (Table 4).PCR detected the MexAB-OprM genes in these samples.

Double-locus sequence typing method
For 36 isolates, DLST was able to determine the existing alleles number; however, four of the isolates, 20, 21, 55, and 58 in Table 2, five new loci, and four new DLST were identified (one allele for ms172 and four alleles for ms217).

Network phylogenic tree generating
All the alignments for the NDM sequences are displayed in Figure S1, S2, S3, S4, and S5.The location of the mutation within the NDM structure, as determined by the alignment procedure, is shown in Table 3S.Figure 2 demonstrates the network phylogenic tree of the studied samples.This tree revealed the phylogenic relationship of the DLST types.Network phylogenic tree was also able to display the potential ancestral DLST.Based on this tree, three major clusters were obtained (Figure 2).Moreover, this tree exhibited that DLSTt26-46 perhaps was the potential ancestral types that produced other DLST types and their clusters (Figure 2).The evolutionary background of NDM enzymes is explored.
In the present study, we found the presence of bla NDM-1,4,5,7 primarily in isolates from the ICU and surgery wards.According to network phylogenic tree, it sounded like bla NDM-1 was a hypothetical common ancestor.Bla NDM-4 was derived from bla NDM-1 , and bla NDM-5 diversified from bla NDM-4 Network phylogenic tree revealed that bla NDM-7 was produced by bla NDM-4 .The evolutional tree exhibited bla NDM-1 is the ancestral species for other bla NDM and furthermore, it stated that bla NDM-1,4 and bla NDM-5 produced other bla NDMs .On the other hand, the network software reconstruction (Figure 3a) suggested that single mutations were mostly responsible for the diversification of the bla NDM-1 , bla NDM-4 , and bla NDM-5 nodes.In the case of bla NDM-4 phylogroup bla NDM-7 variant derived directly from it, carrying only one-point mutation.With a single mutation, twelve variants of bla NDM-1 , six variants of bla NDM-4 , and three variants of bla NDM-5 were directly derive.The variants of bla NDM-10 and bla NDM-18 diversified from bla NDM-1 with five mutations.
The evolutionary directionality from bla NDM-1 to bla NDM-4 and bla NDM-4 to bla NDM-5 was supported by the parsimony approach of Mesquite program and further confirmed by analyses of the observed evolution in adaptive natural situation offering benefits to the harboring strains in a challenging clinicaltherapeutic circumstances (Figure 3b).By reconstructing ancestral states through phylogenic analysis, we were able to propose a potential evolutionary theory for the diversification of bla NDM .
Interestingly, evolutionary orientation pathways of NDM were correlated with the network phylogenic tree of DLST types.The majority of the  meropenem resistance belonged to the ICU and surgery departments.All of the efflux pumps' meropenem resistance depended on the ICU and surgery departments.Four efflux pumps were in the cluster I.

Discussion
P. aeruginosa is one of the most prevalent opportunistic pathogens related to nosocomial infections, such as pneumonia, urinary tract infections, and wound  infections.Carbapenems are the most commonly used antibiotics in clinical practice to treat bacterial infections, and P. aeruginosa isolate resistance to these antibiotics is a major concern for clinicians.The meropenem resistance in Pseudomonas is associated with carbapenemase and the MexAB-OprM efflux pump.Some of the most common and powerful carbapenemases include bla NDM , bla VIM , and bla IMP , and the prevalence of their genes has increased among healthcare system patients [33][34][35][36].
A change in the epidemiology of carbapenamses has been noticed, which is characterized by the explosive diversification and rise in the frequency of NDM under various circumstances.This offers a unique chance to explore a protein evolutionary radiation by analyzing precise point mutations, improving the protein efficiency and fitness simultaneously.The presence of antibiotic compounds promoting protein divergence has been studied using evolutionary bioinformatics techniques.Phylogenetic reconstruction employing all of the NDM variants published so far produced a hypothetical evolutionary scenario with at least one diversification event.
In total, 62.5% of P. aeruginosa isolates were phenotypically carbapenem-resistant (CARBR) isolates.Characterization of isolates revealed that the prevalence of bla NDM , bla VIM , bla IMP , undetermined carbapenemase, and MexAB-OprM were 27.5%, 2%, 2.5%, 12.5%, and 15%, respectively.Most of the samples were exclusively isolated from the upper respiratory tract (URT) of hospitalized patients in ICU and surgery departments, and due to the exclusive consumption of meropenem efficacy in the mentioned hospitals, it was not surprising.In this study, bla NDM-1 , bla NDM-4 , bla NDM-5 , and bla NDM-7 were found in the isolates.
Further analysis of the identified resistance profile revealed a novel phenotype, the FEPR-CAZS, in 99 isolates.This isolate was also CARBR-FEPR-CAZS with undetermined carbapenemase, which based on relevant studies in the literature, this resistance perhaps generated from oxacillinases (class D carbapenemase) bla OXA genes.No efflux mechanism or activity for cefepime was indicated in CARBR-FEPR-CAZS [37].A phenotype has been observed in collected P. aeruginosa from URT, and it has been attributed to an extensive efflux pump system (MexAB-OprM).
Seven, three, and one sample of DLST type 20-105 were collected from ICU, surgery, and internal departments, respectively.Six of the DLSTt20-105 were CARBR.We consider that this DLST type spread in both AL-Zahra and Burns centers (and various wards), providing evidence for putative intra-and inter-hospital transmission.There are patient transfers and hospital equipment transfers (special ICU requisite equipment) among these two centers.Also, there are common staff members, particularly physicians, serving both centers; accordingly, the resistance transferred between the two centers.
Likewise, five isolates with DLST 32-52 were obtained from hospitalized patients in various medical departments, with three, one, and one isolated from ICU, emergency, and surgery departments, respectively.All P. aeruginosa with DLST type 32-39 infected patients were in the Burns center, five of whom were in the ICU department and one in the surgery ward.Three of the DLST t32-39 were CARBR and isolated from patients in ICU.Even in limited numbers of examined isolates, the data indicates the emergence of 'high-risk' clones in other hospitals, like DLST t32-39.The three most common DLST types in this study were CARBR and bla NDM producers.
DLST t12-105, sample 35, with a higher resistance rate of utilized antibiotics (due to the activity of the efflux pump), was isolated from a catheter in ICU.This strain was isolated from the urinary tract infection edema.Overall, the data revealed that the frequency of bla NDM-1 producing P. aeruginosa isolates was 27.5%, and it was the first time that bla NDM-4 , bla NDM-5 , and bla NDM-7 were detected in Isfahan.The bla NDM-1 producing P. aeruginosa isolates have also been detected in Isfahan and Iran.In Isfahan, Shokri et al. reported that the prevalence of bla NDM-1 producing P. aeruginosa was 6% in 2017 [38], considerably lower than the outcome of the present study in 2022.Additionally, Dogonchi et al. discovered only one bla NDM-1 producing P. aeruginosa isolate from the north of Iran [39].In another study, Azimi et al. detected a lower percentage of bla NDM-1 producing P. aeruginosa (7%) in carbapenem-resistant P. aeruginosa in 2021 [40].Furthermore, Riahi Rad et al. demonstrated the presence of the bla NDM-1 gene in Iran [41].In the study of Sheikh et al., DLST type 25-11 was the major cluster, and DLST 5-91 was a high-risk clone with resistance to all used antibiotics [42].The present study results are correlated with those studies, including their findings that exhibited DLST 5-91 as a high-risk clone (also DLST t32-39 was a high-risk clone) [42].However, in this study, DLST t20-105 was the significant cluster.
In the study of Pascal Cholley et al., a comparison of double-locus sequence typing (DLST) and multilocus sequence typing (MLST) for the investigation of Pseudomonas aeruginosa populations showed that combining partial sequence data from only two DLST loci (ms172 and ms217) resulted in highly congruent results to those of MLST.Their goal in using DLST was to establish a technique that could quickly identify an increase in the prevalence of a particular P. aeruginosa clone, a sign of a possible epidemic.They convincingly showed that DLST was sufficient for surveillance purposes of P. aeruginosa [15].The authors concluded that adapted bioinformatics tool development is more beneficial to discover the origin of pathogens [15].
Comparison of the results of the mentioned studies with the results of the present study reveals that the prevalence of bla NDM has remarkably progressed.Moreover, it has been proven that a mutation in the NDM non-active site diminishes its dependency on zinc and improves its potency to degrade the carbapenems [43][44][45].
Our study showed four types of bla NDM , three of which, based on previous studies, had more robust activity against carbapenems than bla NDM-1 (particularly in the zinc starvation) [44,46].Moreover, it has been confirmed that a point mutation created new NDM types with higher potency in beta lactam degradation and diminished level dependency on zinc cofactor [43].Therefore, the presence of bla NDM-4 , bla NDM-5 , and bla NDM-7 in the present study suggest that we should expect more variety of bla NDM , and it would be a considerable threat to health in the near future.It appears that the main factors behind the mentioned phenomenon must be antibiotic overuse, inappropriate use of antibiotics, and originated selective pressure from excessive carbapenem prescriptions in our hospitals.
One of the most common acquired resistance mechanisms is the production of MBLs bla VIM and bla IMP , which were also detected in the present study.These MBLs, along with bla NDM , were perhaps obtained from isolates of ICU, and these results are basically comparable with previous Iranian investigations.In regard to common staff members, medical equipment transfers, and phylogenetic trees, our results indicate that the ICU may be a potential risk and a key source of resistance gene dissemination in Iranian hospitals.The present study findings demonstrated that bla NDM-1 positive isolates were extremely resistant to widely used antibiotics in the clinic, consistent with the outcomes of previous investigations [37,38].
In the case of local phylogenetic research, DLST markers are thought to be relatively stable; however, over a long-term research period, they are likely to experience genetic modifications.In the current study, the evolutionary route of P. aeruginosa was explored to determine the origin of the resistance in every Pseudomonas type by DLST and NDM typing.Subsequently, a network phylogenetic tree for DLST and bla NDM was made to reveal the potential Pseudomonas ancestor and the meropenem resistance diversification.Based on the phylogenetic tree, three major clusters for DLST, I, II, and III, were obtained, and DLST t26-46 as a hypothetical ancestor could be inferred.The NDM network phylogenetic tree supported the theory that DLST t26-46 was a potential ancestor since this tree revealed that the bla NDM-1 evolutionary pathway was congruent with DLST diversification.bla NDM-4 could produce bla NDM-5 and bla NDM-7 .When the outcomes of the NDM phylogenetic tree were implemented in the DLST phylogenetic tree (Figure 3c), it was stated that maybe DLST t32-39 and DLST tN2-N3 with bla NDM-4 derived from DLST t26-46 with bla NDM-1 .Consequently, DLST t5-91 with bla NDM-5 (bla NDM-4 in DLST tN2-N3 mutated to bla NDM-5 in DLST t5-91) was successively diversified from DLST tN2-N3 with bla NDM-4 .Interestingly, meropenem effluxresistant isolates, consisting of DLST t32-39, DLST t1-21, and DLST t20-105, were sequentially placed in the network tree.The majority of the meropenemresistance types, created through NDM or efflux pumps, were found in surgery and ICU departments.In both hospitals, the therapeutic option to treat infections was meropenem, which could have exerted selective pressure for antibiotic resistance mechanisms.
We created a relationship between in silico and in vitro assessments to explore evolutionary trajectories.The present study, with comprehensive attention paid to hospital circumstances, antibiotic prescription, and bacterial diversification pathway, revealed that the spreading of P. aeruginosa and MBLs belongs to one outbreak (not multiple outbreaks).Furthermore, it discovered that antimicrobial drugs should be noted not only as recruiters for efficient resistance mechanisms but also as diversifying agents of evolutionary trajectories.

Conclusion
The presence of bla NDM-1 in P. aeruginosa isolates poses a significant therapeutic challenge to people's health worldwide.To the best of our knowledge, this is the first time that the DLST evolutionary route has been investigated by a network phylogenetic tree.This method allowed us to discover the evolutionary pathway of the NDM enzymes.The network phylogenetic tree revealed that DLST t26-46 was the hypothetical ancestor for the other DLST.The network for NDM also corroborated this hypothesis.Furthermore, for the first time in Iran, a phenotypic resistance called FEPR -CAZS was explored in our study.Moreover, the network phylogenetic tree displayed the circulation of the resistant genes among the departments and hospitals.The present study had three clusters for DLST and three major DLST types, where DLST t32-39 in the Burns center was regarded as a 'high-risk' clone, particularly in ICU and surgery wards.The NDM detection types displayed four types.
Our study displayed that common utilization of medical equipment among two centers, common staff members among the two centers, limitations in treatment options, and prescribing unnecessary high levels of meropenem are the main agents that generate new types of resistant bacteria and spread resistance among hospitals.We suggest some solutions to combat this threat, including designing a new class of antibiotics to combat MBLs resistance, using high-level disinfectants and antiseptics to remove bacterial presence on the surfaces of medical equipment and biological surfaces.Additionally, common usage of equipment and hospital staff members should undergo utmost surveillance.

Figure 1 .
Figure 1.Determination of efflux pump activity with the Cartwheel Ethidium Bromide (ETBR) method.Different ETBR concentrations are added to the agar media and the loopful of fresh culture inoculated.The plate is incubated at 37°C for overnight.Plates were visualized in UV Illuminator and documented.a: 0 mg/L, b: 0/5 mg/L, c: 1 mg/L, and d: 2 mg/L.a, b, c, and d exhibit P. aeruginosa 13, 17, 47, and 89, respectively.In the all plates, negative control was cultured that had no florescent shining.Activity of efflux pump led to thrown out of the EtBr, resulting in florescent shining.

Figure 2 .
Figure 2. a. Network phylogenic tree of the DLST, b. three major clusters of DLST types, and evolutionary orientation pathway of the isolated sample.
. The aligned.meg file was converted to .nex file in alignment transformation environment online server (http://www.sing-group.org/ALTER/).The resulting file was applied to generate .rdffile in Network 10.2.0.0 Software, phylogenic Network tree with Median Joining algorithm with default parameter.Also, the network tree for New Delhi metallo-β-lactamase was generated to investigate the evolutional pathway of the Pseudomonas.The reference sequences of 28 NDM types were downloaded from NCBI sequence database (http://www. [32].nlm.nih.gov).The accession number of the downloaded NDM reference sequences is displayed in the supplementary file.The CluastualW multiple sequence alignment of NDM sequences was applied to create the network tree.Then, based on the presence of NDM types in our study, the evolutionary path and relationship among NDM enzymes were investigated.Using parsimony approaches, Mesquite version 2.75 was performed to determine the ancestral states for NDM (https://mesquiteproject.org/ )[32].

Table 2 .
Characteristics, resistance patterns, and genotypes of the 40 P. aeruginosa isolates.

Table 3 .
MIC values with and without CCCP treatment for phenotypically meropenem resistant of isolated P. aeruginosa.

Table 4 .
PCR outcomes for MBL and DLST typing in P. aeruginosa samples.