Quorum Sensing and Genetic Lineages in Carbapenem-Resistant Pseudomonas aeruginosa

Pseudomonas aeruginosa is a clinically important opportunistic pathogen with resistance to various antimicrobial agents (1). This microorganism is one of the leading causes of life-threatening infections, particularly in individuals with immune system deficiencies, such as those with cystic fibrosis, AIDS, cancer, burns, wounds, and in intensive care units (ICUs) (2). The emergence of multidrug-resistant isolates has become one of the most concerning global issues today (3). Carbapenems are bactericidal agents that act on the cell wall and are used against severe Pseudomonas infections, often as a last resort (4). These antimicrobials consist of a penem fused to a beta-lactam ring and inhibit peptidoglycan synthesis by binding to and inactivating penicillin-binding proteins (PBPs) (5). However, their effectiveness in infection management has decreased with the emergence of carbapenem-resistant P. aeruginosa (CRPA) strains (6).

Impaired permeability of the porin OprD due to loss or alteration, hyperexpression of an active efflux pump, or production of class B metallo-β-lactamases (MBLs) are the primary mechanisms responsible for the emergence of CRPA strains (7). Metallo-β-lactamases are zinc-dependent enzymes belonging to Ambler class B β-lactamases that hydrolyze all β-lactam antibiotics, especially carbapenems, with the exception of monobactams (aztreonam), and are inhibited by EDTA (ethylenediaminetetraacetic acid; a metal chelator) (8). There are three subclasses (B1, B2, and B3) of Ambler class B β-lactamases, with subclass B1 being the most common worldwide (9). The B1 MBL subclass, including the bla_IMP, bla_VIM, bLa_sIM, bLa_sPM, bla_GIM, bla_DIM, bla_IND, and bla_NDM families, has been identified in P. aeruginosa worldwide. Plasmid-carried MBL genes are particularly important due to their widespread global distribution (10).

A cell density-dependent cell-to-cell signaling mechanism known as quorum sensing (QS) is recognized in P. aeruginosa for controlling the expression of biofilm formation genes and virulence factors (11, 12). las and rhl are two major components of the QS system, encoding N-(3-oxododecanoyl)-L-homoserine lactone (3O-C12-HSL) and N-butanoyl-L-homoserine lactone (C4-HSL) as autoinducers, respectively (13). The rhl system contributes to various phases of biofilm production, including microcolony formation, maintenance of open-channel architecture, and detachment of embedded bacterial cells (14). Additionally, rhl controls P. aeruginosa virulence factors, such as rhamnolipids, pyocyanin, hydrogen cyanide, and elastase (12). The las system regulates other features of the biofilm production process, such as biofilm maturation, and also controls genes encoding elastase, exotoxin A, and alkaline protease (15).

Currently, various typing approaches, such as bio-typing, sero-typing, and pyocin-typing, have been established to determine common clones of P. aeruginosa (16). Identifying these clones and understanding their relationships is crucial in epidemiological studies of P. aeruginosa hospital infections, the global prevalence of these bacteria, and the determination of disease burden (17). However, the discriminatory power of these methods is lower compared to nucleic acid amplification-based typing techniques (18). Therefore, several molecular typing methods have been developed, including pulsed-field gel electrophoresis (PFGE), multiple-locus variable number tandem repeat (MLVA), ribotyping, enterobacterial repetitive intergenic consensus polymerase chain reaction (ERIC-PCR), and random amplification of polymorphic DNA (RAPD) (19).

Although PFGE has high sensitivity and is recognized as the gold standard genotyping technique, it is limited by technical complexity, cost, the need for trained personnel, and long turnaround times for results (20). One disadvantage of MLVA is that it is not a universal method, as primers must be designed for each specific application, and results cannot be directly compared with those from other laboratories due to differences in amplicon band patterns. One of the most widely used repetitive DNA targets in PCR-based genotyping approaches is the ERIC sequence, which is common to gram-negative enteric bacteria (21). ERIC sequences offer better potential for the bacterial interspersed repetitive sequence evolution project because they are longer, more informative in comparative analyses, and found in a wider range of species (22).

This study was conducted to investigate the QS genes and molecular typing of CRPA strains using the ERIC sequence method.

In this study, 57 non-repetitive CRPA isolates were collected from various clinical samples as follows: Sputum (22.8%, n = 13/57), BAL (21.1%, n = 12/57), wound (17.5%, n = 10/57), urine (14.0%, n = 8/57), blood (14.0%, n = 8/57), and stool (10.5%, n = 6/57). The frequency of CRPA strains according to hospital wards was ICU (36.8%, n = 21/57), Emergency (22.8%, n = 13/57), Internal (21.1%, n = 12/57), and Infectious (19.3%, n = 11/57). The carbapenemase test conducted with the mCIM assay showed that 89.5% (n = 51/57) of CRPA strains were positive for carbapenemase production. To differentiate MBLs from serine carbapenemase, both mCIM and eCIM phenotypic tests were used. Among the CRPA isolates, 17.5% (n = 10/57) were positive in the eCIM tests.

The biofilm formation test showed that out of the 57 CRPA isolates, 10.5% (n = 6/57), 19.3% (n = 11/57), and 70.2% (n = 40/57) had weak, moderate, and strong biofilms, respectively. The presence of β-lactamase-encoding genes revealed that 75.4% (n = 43/57), 64.9% (n = 37/57), 12.3% (n = 7/57), and 8.7% (n = 5/57) carried bla_IMP, blaV_IM, bla_NDM, and bla_KPC, respectively. On the other hand, 73.7% (n = 42/57), 7.0% (n = 4/57), and 1.7% (n = 1/57) of CRPA isolates carried bla_OXA-48-like, bla_OXA-23-like, and bla_{OXA-20/40-like}, respectively. No isolates were found to carry bla_OXA-58-like. Table 2 shows the correlation between β-lactamase-encoded genes and biofilm production. No significant correlation was observed between the β-lactamase-encoded genes and biofilm formation ability, except for the bla_KPC gene.

The molecular distribution of QS genes showed that 100% (n = 57/57), 96.5% (n = 55/57), 92.9% (n = 53/57), 89.5% (n = 51/57), 84.2% (n = 48/57), 73.6% (n = 42/57), and 63.2% (n = 36/57) of isolates harbored the lasR, lasI, rhlI, rhlR, aprR, aprA, and RhlAB genes, respectively (Table 3). Table 4 shows that isolates harboring QS genes produced stronger biofilms than those without these genes. However, no significant correlation was observed between the QS genes and biofilm formation ability.

Enterobacterial repetitive intergenic ERIC-PCR analysis of 57 CRPA isolates identified eight distinct clusters (A, B, C, D, E, F, G, and H) using a similarity cut-off of ≥ 60%. The distribution of isolates within these clusters was as follows: A, (3 isolates); B, (2 isolates); C, (12 isolates); D, (7 isolates); E, (7 isolates); F, (6 isolates); G, (17 isolates); and H, (2 isolates). Additionally, one isolate was classified as a singleton. Notably, the G cluster exhibited the highest isolate count, with 17 strains (Figure 1).

Figure 1.

Enterobacterial repetitive intergenic consensus polymerase chain reaction (ERIC-PCR) profiles of carbapenem-resistant P. aeruginosa (CRPA) Isolates. Enterobacterial repetitive intergenic ERIC-PCR patterns for CRPA isolates, grouped into clusters based on ≥ 60% similarity.

It is well established that P. aeruginosa infections are associated with high mortality and morbidity due to its ability to adapt to various environmental conditions, express many virulence factors, and acquire antimicrobial resistance (AMR) (11). In the present study, of the total 57 CRPA strains, 70.2%, 19.3%, and 10.5% were strong, moderate, and weak biofilm formers, respectively. Additionally, 89.5% and 17.5% of the isolates were positive in the mCIM and eCIM tests, respectively. The molecular distribution of β-lactamase-encoded genes revealed that 75.4%, 64.9%, 12.3%, and 8.7% of isolates carried bla_IMP, bla_VIM, bla_NDM, and bla_KPC, respectively. Furthermore, 73.7%, 7.0%, and 1.7% of isolates harbored blaOXA-48-like, blaOXA_-23-like, and bla_{OXA-20/40-like}, respectively. The bla_OXA-58-like gene was not found in any of the isolates. In contrast to our findings, Vaez et al. showed that the prevalence of bla_VIM, bla_IMP, and blaNDM was 20% (18 of 90), 8.9% (8 of 90), and 5.6% (5 of 90) of P. aeruginosa isolates, respectively. They stated that the MBL genes (bla_VIMand bla_IMP) seem to play a critical role in the spread of carbapenem-resistant infections and the emergence of multidrug-resistant strains, leading to therapeutic failure.

In a study conducted by Ramazani et al., out of 117 non-duplicative P. aeruginosa isolates, the distribution of MBL genes was as follows: bla_IMP (62.1%), bla_VIM (31.0%), and bla_NDM (6.8%) (25). This difference may be due to the origin of the samples, as most of the strains in the present study were multidrug-resistant (data not shown) and strong biofilm producers. Esmaeili et al. found that of 150 clinical samples, 75 P. aeruginosa isolates were recovered, and 68.6% were MBL producers, all of which carried the VIM gene (100%) (29). In a study by Doosti et al., 70 P. aeruginosa isolates were analyzed, and 56% (n = 23/41) and 24.3% (n = 10/41) carried the bla_VIM and bla_IMP genes, respectively (30). Dogonchi et al. reported that 90%, 40%, and 5% of CRPA isolates harbored the bla_IMP, bla_VIM, and bla_NDM genes, respectively (31).

In line with our data, Ramazani et al. found that 75.8% (n = 22/29), 10.3% (n = 3/29), and 3.4% (n = 1/29) of CRPA isolates harbored bla_OXA-48-like, bla_OXA-23-like, and bla_{OXA-20/40-like}, respectively (25). The bla_OXA-58-like gene was not found in any isolate. In our study, and in agreement with Devanga Ragupathi et al., the distribution of resistance-coding genes in strains with strong biofilm formation was significantly different compared to other strains (32). Devanga Ragupathi et al. showed that 27.8% (n = 20/72) were strong biofilm producers, with a positive correlation to carbapenem resistance, compared to 22.2% moderate, 19.4% weak, and 30.6% non-biofilm formers (32). In agreement with our results, Rahdar et al. highlighted that the increased carbapenem resistance burden in biofilm-producing strains is a significant concern, and the implementation of basic measures to address this phenomenon is imperative (33).

The distribution of QS genes showed that 100%, 96.5%, 92.9%, 89.5%, 84.2%, 73.6%, and 63.2% of isolates carried the lasR, lasI, rhlI, rhlR, aprR, aprA, and rhlAB genes, respectively. In contrast, Salehi et al. reported that the frequencies of the rhlR, lasR, and lasI genes were 5%, 48.3%, and 60%, respectively, while the rhlI, lasB, apr, and rhlAB genes could not be identified in any of the strains (26). Ghanem et al. found that, out of 125 clinical isolates of P. aeruginosa, the frequencies of the lasB and aprA genes were 89.6% and 85.6%, respectively (11). Baskan et al. showed that two QS system genes were detected in 98.1% (n = 51/52), and co-existence of the four QS system genes (lasI/R and rhlI/R genes) was found in 78.8% (n = 41/52) of the isolates (34).

Interestingly, and in line with Karami et al., the presence of QS genes was significantly higher in strains with strong biofilm formation compared to other strains (P < 0.05) (35). Karami et al. also observed a significant correlation between the rhlA gene and biofilm formation capability (P = 0.02) (35). El Askary showed that, of 62 P. aeruginosa isolates, 70.9% were biofilm producers (36). The QS genes rhlI and lasI were present in 67.7% and 48.4%, respectively. The prevalence of the bla_IMP and bla_VIMgenes was found in 33.9% and 24.2% of CRPA isolates, respectively. In line with our data, there was a high propensity for biofilm production in P. aeruginosa, which increased with the presence of QS genes. Contrary to our study, no significant relationship was observed between biofilm formation and carbapenem resistance. This difference may result from geographical diversity, strain genetics, and differences in methodology.

Enterobacterial repetitive intergenic consensus typing revealed eight distinct clusters. The distribution of isolates within these clusters was as follows: A, (3 isolates); B, (2 isolates); C, (12 isolates); D, (7 isolates); E, (7 isolates); F, (6 isolates); G, (17 isolates); and H, (2 isolates). Additionally, one isolate was classified as a singleton. In line with our data, Eladawy et al. demonstrated that ERIC typing is an effective tool for detecting the similarity percentage between P. aeruginosa isolates (37). Hematzadeh and Haghkhah reported that, according to the Dice similarity coefficient of > 80%, 38 ERIC types were identified (16). Zarei et al. found 14 different ERIC fingerprints, including five unique types and nine common types, among P. aeruginosa isolates collected from clinical, environmental, and cockroach samples (38). In agreement with our study, Zarei et al. stated that ERIC-PCR is inexpensive, easy to perform, and provides reliable, rapid, and discriminatory power for typing P. aeruginosa (38).