Virulence Genes in Pseudomonas Aeruginosa Strains Isolated at Suez Canal University Hospitals with Respect to the Site of Infection and Antimicrobial Resistance

1Associate Professor of Medical Microbiology, Department of Medical Microbiology &Immunology Faculty of Medicine Suez Canal University; Ismailia, Egypt 2Professor of Medical Microbiology, Department of Medical Microbiology & Immunology Faculty of Medicine, Suez University, Suez, Egypt 3Professor of Medical Microbiology, Department of Medical Microbiology & Immunology Faculty of Medicine, Suez Canal University, Ismailia, Egypt 4Lecturer of Medical Microbiology, Department of Medical Microbiology & Immunology Faculty of Medicine, Suez Canal University, Ismailia, Egypt 5Lecturer of Medical Microbiology Department of Medical Microbiology & Immunology Faculty of Medicine, Suez Canal University, Ismailia, Egypt


Introduction
Pseudomonas aeruginosa (P. aeruginosa) is considered one of the most prevalent nosocomial organisms associated with high mortality rates and one with the highest prevalence of antibiotic resistance. It is one of the top ive pathogens causing Healthcare Associated Infection (HAI) [1]. In Egypt, it can be responsible for HAI in Intensive Care Unit (ICU) and Burn Unit: 17 % and 21.6% respectively [2,3]. P. aeruginosa is considered multi-drug resistant (MDR) bacteria. As by de inition, P. aeruginosa MDR was de ined bacteria resistant at least to three drugs mainly aminoglycosides, antipseudomonal penicillins, cephalosporins, carbapenems and luoroquinolones [4]. The mechanisms of drug resistance are intrinsic and acquired. This resistance is mediated through several mechanisms including multidrug ef lux systems, enzyme production, outer membrane protein loss and target mutations [5]. For virulence, P. aeruginosa possesses extracellular virulence factors controlled by a complex regulatory circuit involving oquorum sensing (QS) thus producing these factors in a coordinated manner [6].
The bio ilm formation which re lects a community of cells attached to either a biotic or an abiotic surface and enclosed in a complex exopolymeric substance is nowadays a major problem as it increases the potential of the pathogen to be resistant to antibiotics and disinfectants [7], is dif icult to be eradicated and contributes to localized or systemic in lammation, which prolongs wound healing [8]. P. aeruginosa produces at least three polysaccharides {alg (alginate), Pel (pellicle), and Psl} which plays an important role in the stability of the bio ilm structure [9]. To overcome this phenomenon, different strategies have been proposed in order to (i) avoid microbial attachment to a surface, (ii) disrupt bio ilm development to increase the penetration of antimicrobials; and (iii) affect bio ilm maturation [10].
The aim of this study is to examine the prevalence of virulence genes in clinical isolates of P. aeruginosa isolated from Suez Canal University Hospitals (SCUH) in respect to the site of infection and antimicrobial resistance of the strains.

Materials and Methods
A cross-sectional descriptive study was carried out from December 2015 to August 2017 at SCUH on 47 P. aeruginosa strains collected from hospitalized patients who were suffering from Urinary Tract Infection (UTI), Respiratory Tract Infection (RTI),burn infection, bed ulcers, wound infections and bacteremia in Suez Canal University Hospitals(SCUHs) in Ismailia.
Various clinical specimens were collected from hospitalized patients and processed in the Medical Microbiology and Immunology department SCUHs for the isolation and identi ication of P. aeruginosa.
The collected specimens were inoculated onto blood agar, MacConkey's agar and Pseudomonas agar P plates. Then the plates were incubated aerobically at 35± 2˚C for 24 hours. Colonies on blood agar and MacConkey's agar suspected to be P. aeruginosa (by their colonial morphology, being non-lactose fermenter on MacConkey's agar and gram negative bacilli by gram stain) were con irmed to be P. aeruginosa by oxidase test and by production of the blue phenazine pigment pyocyanin on Pseudomonas Agar P which is absolute con irmation of a strain as P. aeruginosa [11].
To detect bio ilm formation, we used Tissue Culture Plate Method (TCP) [13] as follows: overnight isolates from fresh agar plate (24 hours at 37°C) were diluted 100 folds in trypticase soy broth. Then 200 μL of this suspension were inoculated into a 96-well lat-bottomed polystyrene plate, covered and incubated overnight at 37°C. Each well was washed four times with 200 μl of phosphate buffer saline (pH 7.2) in order to remove free-loating planktonic bacteria. For the non-adherent bacteria, the plates were severely shaken to get rid of it. After drying the plates, the wells were stained with 200 μL of crystal violet for 15 minutes (dye was dissolved with 200 μL of ethanol 95%). The optical density (OD) at 630nm was recorded and the results were interpreted [14]. Table 1 shows the mean Optical density (OD) to detect the bio ilm formation by using TCP method.
The virulence genes (toxA, algD, nan1, pslA and pelA) were ampli ied by PCR using a speci ic set of primers listed in table 2.
Bacterial DNA for the PCR analysis was prepared using the bacterial DNA extraction kit (Sigma) following the manufacturer's instructions. PCR was carried out with 2 μL template DNA, 0.25 μM of each primer, 0.2 mM deoxyribonucleoside triphosphates, 1x reaction buffer, 2 mM MgCl 2 and 1.5U Prime Taq DNA polymerase in a total volume of 25 μL.
For pelA and pslA genes, the DNA was ampli ied using the following protocol: initial denaturation (5 minutes at 94°C) followed by 30 seconds-35 cycles of denaturation at 94°C, 40 seconds of annealing at 52°C and 50 seconds of extension at 72°C.The ampli ied products were held at -20°C until analysis [15].

Ethical considerations
The study work obtained approval from the Ethics Committee of Faculty of Medicine, Suez Canal University (FOMSCU), Ismailia, Egypt.

Statistical analysis
The data collected were entered into a database ile. Statistical analysis was performed by using the SPSS 22 software statistical package. Qualitative data was summarized in frequencies. For a subsequent analysis of data, chi square test was used to detect the difference between qualitative data. The outcome variables included sex, type of specimen, hospital wards, bio ilm formation, drug susceptibility and MDR. Statistical signi icance was considered at p. value ≤ 0.05.

Results
This study was carried out on 47 P. aeruginosa isolates that were collected from 296 patients who had HAIs after 24-48 hours of admission in SCUHs in Ismailia during the period from December 2015 to August 2017. P. aeruginosa was de ined. In regard to its percentage in relation to sex, it was found out that it was higher among males than females (55.3% versus 44.7%). With reference to its percentage among different age groups, it was the highest among age group ≥50 years (21.3%) and the lowest from the age group 10 -40 years (8.5%).
With regard to the percentage of P. aeruginosa in different hospital wards, the highest percentage was found in cases in the ICU (29.8%), while the lowest percentage was in pediatrics department (6.4%).
The highest percentage of P. aeruginosa was isolated from pus of the wounds and burns (38.3%), then 31.9% from urine, 19.1% from sputum and 10.6% from blood samples.
Testing the antibiotic susceptibility pattern of the isolated strains showed that the highest sensitivity was to Imipenem and Cipro loxacin (85.1% and 68.1% respectively), while the highest prevalence of resistance was to Cefepime and Ceftazidime (68.1% for each of them) ( Table 3).
Testing the resistance pattern of the MDR and non MDR strains showed that; the highest prevalence of resistance for MDR strains was to Azetronam and Cefepime (92.85% for each) and the highest prevalence of resistance for non MDR strains was to Levo loxacin and Meropenem (47.36% for each) ( Table 4).
Testing the relation between bio ilm production and different sites of infection showed that the percentage of bio ilm formation is higher among sputum and blood specimens than others (Table 5).
According to the antibiotic susceptibility pattern of both bio ilm-producing and nonbio ilm-producing isolates, it was found that bio ilm-producing strains had high prevalence of resistance to Cipro loxacin (80%), followed by Azetronam (70%), Gentamicin (66.6%), Amikacin (64.7%), Pipracillin-Tazobactam (64%) and Cefepime and Ceftazidime (59.4% for each). The resistance pattern to Cipro loxacin, Gentamicin, Cefepime, Ceftazidime, Azetronam and Pipracillin-Tazobactam was signi icantly higher (p. value ≤0.05) among bio ilm producers than non-bio ilm producers as shown in table 6.  Table 7).     The percentage of virulence genes in different hospital wards were determined. It was higher in ICU, surgery and Burn Unit than other wards with statistically insigni icant P. value (Table 8).
The percentages of virulence genes in terms of the site of infection were determined. It was higher in pus of the wounds and followed by urine, sputum and blood (Table 9).
It was found out that certain bio ilm-producing strains were speci ically connected to certain virulence genes. Ninteen strains out of 23 (82.6%) express pelA gene, while all the strains express pslA gene with statisticaly signi icant P. value (P ≤0.05) ( Table  10).
In regard to the relation between MDR, bio ilm production and virulence genes of P. aeruginos, MDR and bio ilm producer strains were found the highest groups which carried virulence genes (Table 12).

Discussion
P. aeruginosa is a Gram-negative bacterium possessing pili, lagella, (lipopolysaccharide) LPS [16]. It is dif icult to be eradicated due to its ability to produce bio ilm [17]. It infects the pulmonary tract, urinary tract, burn and becomes a major cause of HAI worldwide [18]. Eradication of P. aeruginosa has become increasingly dif icult due to its remarkable capacity to resist antibiotics. P. aeruginosa strains are known to utilize their high levels of intrinsic and acquired resistance mechanisms to counter most antibiotics. In addition, adaptive antibiotic resistance of P. aeruginosa is a recentlycharacterized mechanism (4) which includes bio ilm-mediated resistance and formation of multi-drug-tolerant cells, and is responsible for the relapse of infections. The discovery and development of alternative therapeutic strategies that present novel avenues against P. aeruginosa infections demand a more increasing attention [19].
This study aimed at determining the prevalence of certain virulence genes in clinical isolates of P. aeruginosa and to correlate the presence of these genes in different sites of infection with antimicrobial resistance.
A total of 296 specimens were collected from patients with nosocomial infections in SCUH. From the specimens, 47 P. aeruginosa strains were isolated (15.9%). In the study of Mahmoud et al. [20] at Meno ia University hospital, P. aeruginosa was found to account for 19.8% of nosocomial infections. Wassef et al. [21] in Cairo, Egypt, isolated P. aeruginosa with a prevalence rate of 20.7%. Lower isolation rate (6.67%), was reported by a number of studies such as Khan et al. [22] in Pakistan. The percentage of P. aeruginosa is variable in various studies in literature. This might be attributed to drug overuse and hospital policy in management of such cases. Moreover, geographic climatic and hygienic factors may also be correlated with the relative variability of results among different areas [19].
In this study, the highest percentages of P. aeruginosa were from ICU, Surgery Department and Burn Unit (29.8%, 21.3% and 14.5% respectively). This is comparable with several studies such as Ikeno et al., Gad et al. and Poursha ie et al [23][24][25], which can be seen as ringing danger alarms for the widespread organism. This can be interpreted in terms of anaerobic growth of the bacteria obtaining energy from oxidation of sugars thus rendering dif iculty of eradication [21]. Amany et al. 2017 [26], found that acquired infection rate in ICU was higher than other hospital wards. The ubiquitous nature, including the ability to survive in a moist environment and resistance to many antibiotics, makes P. aeruginosa a common pathogen in the ICUs of hospitals.  From a different perspective, the cause of the widespread existence of P. aeruginosa in the Burn Unit was due to impairment of the skin barrier in burn victims, debridement and manipulation of the burn site [27,28,29,30,31]. Also, it can be attributed to the production of proteases that can alter the host's physical barriers by splitting proteins with the production of amino-acids that allow the deep in iltration of the bacteria. Exotoxin A halts the synthesis of proteins and the hemolysins break down lipids in epithelial cells in order to permit the bacteria for more penetration and spreading [30].
Infections caused by P. aeruginosa are often severe due to limited antibiotic susceptibility and emergence of antibiotic resistance [30]. NNIS data (i.e. within the period from 1998 to 2003) [33] showed the highest prevalence of resistance rates of P. aeruginosa against antibiotics was to Imipenem, Cipro loxacin, and Ceftazidime by 15%, 9%, and 20%, respectively. Also, there was evidence that the highest rates were against Cefepime and Ceftazidime (68.1% for each) which was in convenience with Mahmoud et al. and Oni et al [20,34]. This is explained by Cefepime which has reliable activity against P. aeruginosa because of the drug chemical structure allowing binding to penicillin-binding proteins and penetrating through the outer membrane of Gramnegative bacteria more rapidly than most Cephalosporins. Moreover, Cefepime is also stabler to β-lactamase hydrolysis [35].
From the previous explanation, we had expected to ind high sensitivity level to Cefepime, but we found a high level of susceptibility to the drug. This may be attributed to the production of high levels of AmpC β-lactamases by some strains that become fully Cefepime-susceptible. This phenotype is usually found among ICU patients who frequently receive multiple treatment courses of expanded-spectrum β-lactam antibiotics for prolonged periods.
For the Ceftazidime that has a C=N-OCH 3 group in its chemical structure which provides stability against β-, acts as a penicillin-binding proteins inhibitor (37) . P. aeruginosa resistance against Ceftazidime arise from the horizontal acquisition of β-lactamases, altered expression of class C β-lactamase AmpC [36].
In the current study, Imipenem and Cipro loxacin were the most effective drugs against P. aeruginosa. The sensitivity of Imipenem and Cipro loxacin were 85.1% and 68.1% respectively due to their ability in producing several different porins as outer membrane porin D (OprD) so they can cross the outer membrane of P. aeruginosa [37].
The variations in the results of the antibiotic resistance might be referred to the difference in the pattern of drug use in different parts of the world and due to the several mechanisms that have been reported for P. aeruginosa, including: 1) Reduced expression or loss of OprD porin causing reduced antibiotic permeability, 2) Overexpression of MexAB-OprM pump which increases antibiotic ef lux, 3) Production of β-lactams and aminoglycosides inactivating enzymes, 4) Mutations of gyrases and topoisomerases which cause luoroquinolone resistance. These mechanisms in combination lead to multiple drug resistance [38,39].
What adds to the problem of P. aeruginosa causing HAIs is the emergence of MDR strains. In this study, a high prevalence of MDR P. aeruginosa strains (59.6%) was reported, and the highest prevalence of resistance for MDR strains was to Azetronam and Cefepime (92.85% for each). Similarly, a high rate of MDR was reported in a number of studies. For instance, in Turkey, Ünan and Gnsern [40] reported that 60% of their P. aeruginosa isolates were MDR; in Egypt, Mahmoud et al. [20] found that MDR P. aeruginosa were (52%) among their isolates.
The evolution of numerous MDR P. aeruginosa can be explained by the ability of the bacteria to acquire antibiotic resistance through horizontal gene transfer and spontaneous mutation [41].
The TCP assay is a simple and rapid method to quantify bio ilm formation. We found that 13 strains out of 47 (27.7%) were strong bio ilm producers, 10 (19.1%) moderate and 24 (51.1%) weak or non-bio ilm producers.
It is noteworthy that literature shows results quite consistent with the present ones. In Egypt, Hisham et al. [42], found that 16 isolates (80%) were strong bio ilm producers; 2 isolates (10%) were moderate and another 2(10%) were weak. Also, Abd El-Galil et al. [43]. found that 42 isolates (84%) were strong bio ilm producers; 4 isolates (8%) were moderate ones and 4 isolates (8%) were weak ones. With no much difference from the present study results, Maita and Boonbumrung found that 60% of strains were strong bio ilm producers from a total of 136 strains; 11% were moderate and 22% non-producers.
In the present study, bio ilm production was higher among blood and sputum than other specimens. The cause can be ascribed by the observation that bio ilm-colonizing devices implanted inside the body or forming a connection between inner and outer surface of the body where a normal microbial lora is present, are to blame. This type of infections is particularly associated with orthopedic devices and intravenous catheters.
Statistical analysis of this study showed signi icant association (P value ≤0.05) between bio ilm production and MDR. 40.7% were MDR and bio ilm producers and 19.1% MDR and non-producers.
Previous studies have shown that bio ilm formation is higher in MDR strains [44][45][46]. This may be referred to the protective nature of the bio ilm that makes the bacteria (i.e. growing intrinsically) resistant to many antibiotics up to 1000 times higher than normal levels. Another reason is the slow growth rate of the bacteria in the presence of antibiotic degradation mechanisms.
It is worth mentioning that Maita and Boonbumrung found that the antibiotic resistance to Amikacin, Gentamicin, Ceftazidime, Cefepime, Imipenem, Meropenem, Cefoperazone/Sulbactam and Piperacillin/Sulbactam was higher among bio ilms producing P. aeruginosa than that which was non-producers. However, Levo loxacin and Cipro loxacin were found to exhibit similar resistance in both bio ilm producers and non-producers. In the case of the bio ilm-producing strains, relating results in the present study found an increase more than 50% of the resistance to Ceftazidime (52.8%), Levo loxacin (51.9%), Cipro loxacin (51.9%) and Cefoperazone/Sulbactam (55.6%).
In the present study, the percentages of a number of P. aeruginosa virulence genes (toxA, nan1, algD, pelA and pslA genes) and their relation to the site of infection were detected. It is known toxA gene encodes exotoxin A that acts as a major virulence factor of P. aeruginosa. The gene was detected in 45 isolates (95.7%). Other studies reported the same results as Qin et al., .
As for algD gene, it encodes GDP-mannose 6-dehydrogenase enzyme which catalyzes the oxidation GDP-D-mannose to GDP-D-mannuronic acid, a precursor for alginate polymerization. The alginate layer causes a mucoid phenotype and provides a protective barrier against host immune defenses and antibiotics. In the present study, evidently, it was detected in 42 isolates (89.4%). Al-Dahmoshi et al. [14], and Ra'oof revealed that all isolates had the algD gene and showed high capacity of alginate bio ilm formation which interfered with response of the P.aeruginosa isolates to antibiotics.
PelA gene, which is necessarily associated to the polysaccharide stage of bio ilm development and maintenance, was detected in 41 isolates (87.2%) and in 19 bio ilm-producing strains out of 23 (82.6%). This result corresponds to Sharma and Choudhury's [50].
The percentages of all virulence genes were high in ICU, surgery and Burn Unit. The differences in the distributions of virulence factor genes in the populations strengthen the probability that some P. aeruginosa strains are better adapted to the speci ic conditions found in speci ic infectious sites [53].
The study results were limited to a sample of 47 strains. The reason behind this was due to insuf icient inancial support and the short duration of the study period. However, we anticipate the expansion of the work on a large sample size in a longer period.
We conclude that P. aeruginosa is seen an extremely versatile micro-organism. It will continue to surprise us yet with unappreciated modes of niche adaptation, lifestyle, and pathogenicity. We conclude that there is relationship between virulence genes and bio ilm formation in P. aeruginosa. We advise the expansion of work on a large sample size in a longer period of time in order to study other virulence genes.