Outbreak of Carbapenem Resistant High-Risk Clone ST244 Pseudomonas aeruginosa in Dogs and Cats in Algeria

1. Introduction

Pseudomonas aeruginosa is an important Gram-negative opportunistic pathogen of humans and animals [1]. In dogs, this bacterium can cause various infections, including ulcerative keratitis, otitis, pyoderma, urinary tract infections, skin infection, wound infections, and respiratory tract infections [2]. Infections caused by P. aeruginosa may be linked to immunosuppression in companion animals, including documented cases in dog’s post-kidney transplantation[3] and in connection with cancer treatments [4]. Moreover, it acts as a pathogen in cats, although it is less common than in dogs [5,6] and the reported infections include respiratory tract infections [3,7] as well as ulcerative keratitis and wound infections [8].

Due to its increased resistance to antibiotics, the World Health Organization (WHO) has classified P. aeruginosa carbapenem-resistant as a high-priority bacterium. It was designated as one of the ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa and Enterobacter species) because of its capacity to escape killing by acquiring exogenous genes and developing resistance through a variety of internal pathways, which promotes the emergence of multidrug resistant strains [9,10]. The European Food Safety Authority's Panel on Animal Health and Welfare has classified P. aeruginosa as one of the most significant antimicrobial-resistant bacteria in the European Union (EU), with over 90% certainty [11].

P. aeruginosa infections are worrying, because their treatment poses a major global challenge due to the development of resistant strains in both humans and animals. Antibiotics have become less effective, or even ineffective, against this bacterium, due to its resistance mechanisms, which lead to therapeutic failure. The main resistance mechanisms are beta lactamase production, efflux pumps, induced mutations and biofilm production [12]. Some authors reported that P. aeruginosa strains showed a high rate of resistance to multiple antimicrobial agents [13]. Its infections in pets are currently generally treated with broad-spectrum antibiotics [14] and close interactions between pets and humans make significant opportunities for interspecies transmission of resistant bacteria and horizontal transfer of antibiotic resistance genes in both directions mainly through physical injuries, petting, or licking activities [15]. P. aeruginosa is also recognized for its ability to quickly acquire additional resistances, meaning that the combination of intrinsic and acquired resistance can result in therapeutic failures [16].

Biofilm formation is a crucial survival strategy used by P. aeruginosa to endure challenging conditions such as exposure to antibiotics and host immune defenses [17]. Moreover, it displays various virulence factors, including, exotoxins (toxA, toxR), elastases (lasB), proteases (plcH) and alginate (algD) all of which contribute to the development of severe diseases. A significant virulence factor is the Type 3 Secretion System (T3SS), which delivers four cytotoxins, including exoU [18].

There are few reports documenting the patterns of antimicrobial resistance and virulence factors of P. aeruginosa isolated from companion animals, and until now there are no studies focused on the epidemiology of this pathogen in Algeria. The aim of the study was to investigate the antimicrobial resistance profiles, and the associated resistance and virulence genes of Pseudomonas aeruginosa strains isolated from dogs and cats in some regions of Eastern Algeria. As the first study of its kind in Algeria, it aimed to provide valuable insights into the epidemiological characteristics and potential health risks posed by these strains in companion animals, highlighting their implications for veterinary and public health.

2. Results

2.1. General Population Information

Table 1 summarizes the basic information of each analyzed P. aeruginosa strain. Fifteen were isolated from dogs, mostly from nasal swabs (n=6), followed by rectum (n=5) and middle ear (n=4). Four cats were positive for P. aeruginosa from rectum. One isolate was collected per animal.

2.2. Antimicrobial Susceptibility Testing

Table 2 displays the prevalence of resistance to each antimicrobial agent. All the strains were resistant to AMC in both cats and dogs, 36.8% and 15.8% of the strains were resistant to TC and TCC from dogs (40%) and cats (25%), respectively. The resistance rate to Imipenem was 36.8%, with 20% in dogs and 75% in cats. No resistance towards aminoglycosides and fluoroquinolones was detected. None of the isolates were multidrug resistant.

2.3. Biofilm Formation

Diverse biofilm profiles were identified, 42% (n=8) classified as strong biofilm producers, 25% (n=5) as moderate producers, all isolated from dogs, and 27% (n=5) as weak producers, including four isolates from cats and one from a dog. One strain, a non-producer, was identified from a dog. All isolates identified as strong producers were derived from dogs; 37.5% were isolated from the rectum, with an equal amount from the nasal cavity. Additionally, 60% of the strains exhibiting moderate production in dogs were isolated from the nasal cavity. There is no relation between the sampling site and the ability of biofilm production (p= 0.430).

2.3. General Features of the Genomes

All the strains were uploaded to NCBI database under Bioproject PRJNA1153397. The sequencing data from the Nanopore Mk1C after genome assembly include a mean coverage of 120× with N50 of 6216739 bp and a genome size of 6560953.684 bp. Table 3 summarizes the main statistics of the genomes assembled and polished.

Whole-genome based phylogenetic tree (Figure 1) was built with the online tool Integrated Prokaryotes Genome and pan-genome Analysis service IPGA (v1.09) (accessed on June 13^th 2024) including the reference strain (PAO1) used during the bioinformatic analyses [19].

The pan-genome profile of all the P. aeruginosa strains is reported in figure 2A. In red are represented metabolism genes, in orange those related to information storage and processing, while genes involved in cellular processes and signaling are in blue, finally the grey ones are poorly characterized or unannotated genes. Together these components constitute the core genes shared between the analyzed strains. As shown in Figure 2B, a total of 12,250 pan-gene clusters were identified. The Average Nucleotide Identity (ANI) analysis (Figure 2C) revealed a high (>98%) identity among all the strains. The upset plot (Figure 2D) indicated that the distinct gene clusters in each genome ranged from 15 to 803. The pan-genome profile derived from Clusters of Orthologous Genes (COG) annotation indicated that the core gene clusters comprised 2480 for metabolism, 1102 for information storage and processing, 1521 for cellular functions and signaling, and 718 were poorly characterized or unannotated. Figure 3 represent the pangenomes visualized using ANVI’O to address various aspects of interactive displays.

2.4. Multilocus Sequence Typing (MLST)

MLST analysis revealed 12 different sequence types (ST4160, ST1248, ST1247, ST2788, ST1722, ST189, ST343, ST16, ST1415, ST388, ST244, ST1342). The predominant ST was the high-risk clone ST244 found in 7 strains (2 from cats and 5 from dogs) firstly described in pets in Algeria. These strains were isolated from different sampling sites, rectum in cats and from the rectum, ear, nasal cavity in dogs. On the other hand, two other sequence types were isolated from cats (ST1247 and ST1342) and the others STs were recovered from dogs. One clonal complex (Figure 4) was displayed comprising two STs (2788 and 388) with a double loci variant, and the rest which did not share at least two out of the seven loci, were considered as singletons STs. The isolates from the same ST are clustered together according to the phylogenetic tree. Figure 4 was generated using the eBURST software [20].

2.5. Antibiotic Resistance and Virulence Genes

A total of 76 resistance genes were found in P. aeruginosa genomes (figure 5). Most of them were associated with the efflux pump systems. The genes aph(3’)IIb, bcr-1, catB7 and fosA, conferring resistance to aminoglycosides, bicyclomicin, phenicoles and fosfomycin respectively, were present in all isolates. Furthermore, the peptide resistance genes basR, basS, cprS and cpxR were also identified in all the isolates.

The results from WGS showed that all the strains harbored at least two genes responsible for betalactam resistance. The predominant gene was bla_PAO found in all isolates, followed by bla_OXA-50 and bla_OXA-396 in 73.7% and 57.9% of the strains respectively.

All the strains resistant to imipenem harbored the following genes OXA_396, OXA₄₉₄, OXA₅₀ and bla_PAO. The combination between bla_OXA and bla_PDC was found in 57.9% of the strains. 84.2 % of the strains harbored more than three genes responsible for beta lactam resistance. Aminoglycoside resistance gene aph(3’)IIb was detected in all the isolates. At least one of the bla_PDC variants was present in 11 strains (57.9%).

Figure 5. Phylogenetic tree and distribution of resistance genes among analyzed strains.

Figure 5. Phylogenetic tree and distribution of resistance genes among analyzed strains.

All genomes were screened for virulence factors (Figure 6), resulting in the identification of those responsible for motility and adhesion, quorum sensing, biofilm production, type III secretion system, siderophores production, proteases, toxins, and enzymes.

A total of 80 genes associated with adherence and motility were detected, with variations observed in their distribution among the isolates. All of them contained the gene essential for lipopolysaccharide (LPS) production. Of the 46 genes involved in flagellar assembly, 35 were present across all strains, with fleS gene being the most frequently absent, missing in 63.2% of the genomes. Regarding the genes responsible for type IV pili biosynthesis, pilA and pilB were found in 84.2% and 100 % of the strains, respectively.

The frequency of exoU and exoS was 5.3% (1/19) and 89.5 % (17/19), respectively.

WGS analysis revealed that the predominant type three secretion system (T3SS) virulotypes were exoU - / exoS + found in 17 strains (89.5%), exoU + / exoS – and exoU - / exoS – in one isolate each.

Proteases, toxins and enzymes are among the most effective virulence factors that contribute in the severity of Pseudomonas aeruginosa infection. All the isolates harbored the genes aprA, lasB and prpL coding for proteases. The toxA gene coding for the exotoxine A was also present in all the isolates. Regarding genes encoding enzymes plcH, plcN, plcB were present in all isolates, while pldA was found in only 8 isolates (42.1%)

Additionally, the isolated P. aeruginosa strains exhibited a 100% prevalence of the quorum sensing genes lasR, rhlR, and rhlI, while the lasI gene was identified in 73.7% of the strains.

Alginates and lipopolysaccharide are essential for the biofilm formation and the genes encoding their synthesis algD and LpS were found in all the strains.

3. Discussion

P. aeruginosa is a significant opportunistic and challenging to treat bacterium. This pathogen aligns with the concept of "One Health" [21] due to its unique characteristics, including extensive environmental diffusion [22], intrinsic resistance to several classes of antibiotics [23], high capacity to acquire new resistance mechanisms [24] and numerous virulence factors [25].

In the present study, P. aeruginosa was isolated at a low prevalence of 7.5% and 1.91%, respectively, from the middle ear, rectum, and nasal cavities of canines and cats.. These rates are lower than those documented in an earlier investigation in Thailand by [26] with 79.6% and 20.4%. In dogs, our results are similar to previous reports in Italy (8%) [24] and China (6.7%) [27]. However, prevalence is higher than earlier reports in India and California (3% each) [28,29]. Other studies conducted in Romania (40.84%) [30], South Korea (18.75%) [31] and Brazil (31.62%) [32] had reported high prevalence rates than our findings. In cats, we observed a prevalence of 1.9%, which is close to the results reported by Gentilini et al., (2018) in Italy with 1.41% [33].

With no MDR or XDR isolates, the antibiotic resistance profile of the isolates from this study was lower than that of academic research on P. aeruginosa isolated from dogs and cats. Research in Tunisia revealed that all isolates were susceptible to all antibiotics [34], despite other studies from humans and animals [35] reporting significant antibiotic resistance in P. aeruginosa isolates. According to Valero et al. (2019), the betalactams ceftazidime, cefepime, and piperacillin/tazobactam demonstrated susceptibilities greater than 85% [36]. In a similar study in Romania, strains isolated from dogs with superficial skin infections showed very high resistance rates to aminoglycosides (62.06% for CN, 55.17% for AK, and 91.37% for TOB) [30]. In contrast, our findings showed no resistance to this class of antibiotics. In our study, no resistance to ciprofloxacin was detected, which contrasts with the findings of Feßler et al. (2022), who reported a resistance rate of 16.1% in dogs' isolates, while no resistance was observed in cats, thus aligning with our findings [37]. Rubin et al. (2008) reported that ciprofloxacin was the most effective fluoroquinolone labeled for veterinary use with 16% of canine resistant P. aeruginosa [38].

Imipenem-resistant strains have been identified with a frequency of 20% in dogs and 75% in cats, indicating a significant public health risk. This class of antimicrobials, a key class in human medicine, is not used for treating infections in animals nor licensed for veterinary use [39] and its prescription is limited to treating urinary and respiratory tract infections in cats and dogs [40].Carbapenem resistance was not extensively studied, several reports have indicated a high resistance prevalence [11,41,42]. In Algeria this study is the first report underlying the presence of carbapenem resistant P. aeruginosa isolated from pets while other Algerian reports focused on carbapenem resistant enterobacterales in companion animals [43,44]. The emergence of such high levels of resistance could severely limit therapeutic options and complicate infection management in both animals and potentially humans. Notably, our results show significantly higher resistance rates than those reported by [45] in Japan, where only 3.4% of strains exhibited resistance to imipenem. Nevertheless, our results are approximately similar to those found by [46] at 30%, while [34,47] found one carbapenem resistant isolate among a collection of 181 P. aeruginosa strains in Portugal and 66 in Tunisia, respectively. This discrepancy in resistance may be attributed to the sampling site, the overall health condition of the animals and the impact of antibiotic use as a key contributing factor, particularly when antibiotics are used in an uncontrolled and indiscriminate way, all these factors may be involved. The isolation of carbapenem resistant strains in this context from non-human source is of a great risk to public health and their origin can possibly be the human-pet bond [40].

MLST is an excellent tool for long-term, worldwide epidemiological research. In our study, we identified 12 sequence types (STs) from dogs and cats; 12 STs were isolated, each from a different strain, while ST244 was found in eight strains and one clonal complex was detected (2788 and 388). The presence of a clonal complex indicates a strong resistance-related genetic association between the current isolates [45]. This can affect bacterial resistance, as closely related strains may harbor shared resistance genes or protective mechanisms against antibiotics.CC244 clonal complex was detected in P. aeruginosa in pediatric populations in China with ST244, ST8818, ST1701, and ST1103. Isolates belonging to CC244 demonstrated significantly higher resistance [48].

ST244 is significant for its role in the horizontal acquisition of antibiotic resistance genes through mobile genetic elements [49] and is recognized as the fourth among the top ten high-risk P. aeruginosa epidemic lineages worldwide [50] associated with MDR [51]. This epidemic clone is one the most extensively researched clones, as mentioned in 182 articles, highlighting its significance in the dissemination of antimicrobial resistance genes (ARGs) [49]. The global spread of high-risk P. aeruginosa clones poses a significant public health challenge since their microevolution in aggressive environmental conditions by acquiring new mutations in their genome leading to new antimicrobial resistances [52]. In addition, many difficult to treat infections resulted from high-risk clone strains with higher pathogenicity and virulence levels and increased capacity to colonize and persist within a host [53]. However, our strains did not show multidrug resistance phenotypes.

ST244, which is the second most prevalent Mediterranean P. aeruginosa clone, was frequently reported in the study [54] in hospitals in Annaba and Skikda in northeastern Algeria, as well as by [55] in Batna hospital in eastern Algeria. In Europe ST244 is one of the most prevalent epidemic high-risk genotypes [56]. Since ST244 was recovered from both dogs and cats, and since P. aeruginosa is considered as an important source of both community-and hospital-acquired infections [57] this suggests that these community acquired P. aeruginosa strains may be more prone to disseminate in the surrounding environment. Our study is the first conducted in Algeria on the sequencing of P. aeruginosa in companion animals, making impossible for us to compare our results with previous research.

In this study, we performed WGS focusing on the detection of resistance and virulence-associated genes. Fosfomycin, one of the earliest antimicrobials, has recently been reconsidered for its potential effectiveness against multidrug-resistant strains, including P. aeruginosa [58]. Further research may be required to understand the role of the fosA gene on P. aeruginosa susceptibility to fosfomycin. The detected ARGs aph(3′)-IIb, fosA, and catB7, are worldwide documented in P. aeruginosa and often located on the chromosome [59]. The blaPAO is a cephalosporinase encoded in the chromosome and in P. aeruginosa is prevalent among multidrug resistant strains [60].

Similarly to the ARGs detected in our strains conferring resistance to beta-lactams, fosfomycin, aminoglycosides and chloramphenicol, the complete genome sequence of P. aeruginosa strains from canine skin lesion showed the presence of aph(3’)-IIb, catB7, bla_OXA-488, bla_PAO and fosA as ARGs [1] and bla_PAO, bla_PDC-24, bla_OXA-486, aph (3′)-IIb, fosA and catB7 in a carbapenem resistant P. aeruginosa isolate from red deer [47]. Also a multidrug-resistant P. aeruginosa isolate from a dairy cow with chronic mastitis carried bla_OXA-485, bla_OXA-488, aph(3')-IIb, bla_PAO, fosA and catB7 [61]. bla_PAO and bla_OXA50 presented high prevalence in P. aeruginosa genome [62].

Whole genome sequencing of P. aeruginosa (Figure 2) identified many antibiotic resistance gene sequences, suggesting that this bacterium possesses the ability for spontaneous transformation as suggested from the literature [63].

Regarding the virulence genes, the transcription of many genes is controlled by a mechanism known as quorum sensing (QS). The lasR, rhlR and rhlI were identified in all the strains (100%), unlike the lasI which was found in 73.7% of the strains. So, 73.7% of the strains harbored all the genes of QS together.

The high prevalence of the exoS, exoT, and exoY genes in the present study (89.5%, 89.5%, and 94.7%, respectively) was consistent with the existing literature [46,64]. The most common virulotype identified in our study was exoU-/exoS+ (89.5%), while the least frequent was exoU+/exoS+ which was absent in all the strains. These findings are similar to those reported by Hayashi et al. (2021) with respective rates of 81.3% and 1.3% [64]. These type three secretion system effectors are crucial contributors to mortality [65]. Strains with both exoU and exoS cytotoxins have been found in other studies [66]. exoT is the most frequent effector in genomes of clinical and environmental P. aeruginosa [67,68]. ExoY gene was predominant in the genomes of both urinary and environmental strains [69].

In agreement with our results, the genes toxA, lasB, and plcH were identified in all strains confirming the findings reported in the literature [24,70]. Similarly, in line with our findings, the gene coding for alkaline protease aprA, was found in all strains, while the toxA gene was present in 91.7% confirming previously results [24].

The algD gene encodes for an enzyme that synthesizes the polysaccharide alginate, an important component of the biofilm produced by P. aeruginosa was identified in all the strains. Our finding showed that 95% of all isolates had the ability to form a biofilm. According to [21] biofilm-forming strains made up 90.6% of P. aeruginosa isolates from dogs and 86.4% from cats. Of these biofilm-forming bacteria, 26.3% had poor, 35.0% had intermediate, and 38.7% had strong biofilm formation. However, a study conducted in Portugal, in which the biofilm-forming ability of P. aeruginosa isolated from dogs was investigated, found that all the isolates appeared to be weak biofilm producers [46] and in another study conducted by Hattab et al. (2021) among canine isolates, five isolates (20.8%) were classified as strong biofilm producers, while 8 (33.3%) and 11 (45.8%) isolates were weak and intermediate biofilm producers respectively [24].

4. Materials and Methods

4.1. Population’s Study and Sample Collection

From February 2021 to December 2023, 409 samples were taken from healthy and clinically ill dogs and cats from different breeds in various regions of eastern Algeria; Batna (n=281), Khenchela (n=58), Setif (n=66) and M’sila (n=4) (figure 7). For cats, samples were collected from rectum (n=102), ear (n=30), abscesses (n=29), wound (n=17), uterus (n=12), nasal cavity (n=10), urine (n=1), buccal cavity(n=8) and 4 of other origin. In dogs, the sample sites included: ear (n=88), nasal cavity (n=70), rectum (n=25), wound (n=6), abscesses (n=4), eye (n=2) and vagina (n=3). After sampling the samples were refrigerated and shipped to the laboratory for bacterial investigation.

Figure 7. Map of northern Algeria highlighting the sampling areas (https://www.mapchart.net/).

Figure 7. Map of northern Algeria highlighting the sampling areas (https://www.mapchart.net/).

This study was conducted in accordance with the requirements of the Scientific committee of the Institute of Veterinary and Agricultural Sciences (University of Batna 1), under the certificate of Animal use Protocol N^o: 001/DV/ISVSA/UB1/2025.

4.2. Culture Conditions and Bacterial Identification

Upon arrival at the laboratory, swabs were incubated in BHIB (Brain Heart Infusion broth; HIMEDIA, Nashik, India) for 24 hours at 37°C, then cultivated aerobically on cetrimide agar (MERCK, Germany) at 42°C for 24-48 hours. A culture is considered positive when greenish colonies, oxidase and catalase positive, and Gram-negative rods (after Gram staining) are observed. Cultures were then purified on MacConkey agar (MERCK, Germany), and pure strains were identified using the API 20 NE system^® (Biomérieux, France).

4.3. Antimicrobial Susceptibility Testing

Antimicrobial susceptibility testing was performed using the Kirby-Bauer disc diffusion method according to the EUCAST guidelines [71]. Bacterial suspensions were adjusted to the 0.5 McFarland turbidity standard. Bacterial isolates were tested to a panel of 13 antibiotics: amikacin (AK) 30µg, ceftazidime (CAZ) 30µg (Oxoid, Basingstoke, UK), gentamicin (CN) 15µg (Bioanalyse, Ankara, Turkey) netilmicin (NET) 30µg, tobramycin (TOB) 10µg, levofloxacin (LEV) 50µg, ciprofloxacin (CIP) 5µg, cefepime (FEP) 30µg, aztreonam (ATM) 30µg, ticarcillin (TC) 75µg, ticarcillin-clavulanic acid (TCC) 75/10µg, imipenem (IMP) 10µg and amoxicillin-clavulanic acid (AMC) 20/10µg (Biomaxima, Lublin, Poland). P. aeruginosa ATCC 27853 was used as a quality control strain.

4.4. In Vitro Biofilm Formation Assay

To study the biofilm formation ability, the microtiter assay was performed, using the protocol as previously described [72]. BHIB broth was used to induce biofilm formation, and P. aeruginosa ATCC 27853 was used as a positive biofilm producer.

Bacterial optical densities were measured using a microplate reader (M 960, Metertech), at 550 nm using as blank a 30 % (v/v) solution of acetic acid. Optical density for each isolate (A₅₅₀) were obtained by averaging 8 wells, then compared with the cutoff value (0.102) which was determined arbitrarily by the mean of the negative control (0.042) plus three standard deviations (0.02). The level of biofilm production was classified in four categories: no production A550< 0.101, weak biofilm production 0.101 ≤ A550 < 0.202 (2 x negative control), moderate biofilm production 0.202 ≤ A550 < 0.408 (4 x negative control), and strong biofilm production 0.404 <A550 according to the literature [21].

4.5. Whole Genome Sequencing (WGS) and Bioinformatics Analysis

4.5.1. DNA Extraction

All P. aeruginosa strains were sequenced using a long-reads sequencing (LRS) approach based on MinION Mk1C platform (Oxford Nanopores Technologies, UK). Isolated colonies from a fresh culture of P. aeruginosa on blood agar were resuspended in 500 µL PBS (Phosphate Buffered Saline, Euroclone, Italy), centrifuged 10,000 ×g for 1 minute at room temperature and high molecular weight DNA was extracted with the Quick-DNA^TM HMW MagBead Kit (Zymo Research, Irvine, CA, USA) following the manufacturer's instructions. The DNA quantity and quality were assessed using a NanoReady Touch series Micro Volume (UV-Vis) (Aurogene, Italy), ensuring that the A_260/A₂₈₀ and A_260/A₂₃₀ ratios ranged between 1.8 and 2, respectively. Extracted DNA was also subjected to gel electrophoresis to check its integrity.

4.5.2. Whole Genome Sequencing

The sequencing libraries were prepared with 200 ng, as input DNA, which were subjected to transposase fragmentation with the Rapid Barcoding Sequencing kit (SQK-RBK114.24, Oxford Nanopore Technologies, UK). Then, 12 isolates were multiplexed on a single flow cell (FLO-MIN114, R10.4.1 version), and sequenced in a MinION Mk1C for 72h maximum.

4.5.3. Bioinformatic Analysis

Dorado (v0.8.2) was used to basecall (--dna_r10.4.1_e8.2_400bps_hac@v5.0.0), trim adapters and demultiplex .pod5 files (--config configuration.cfg --barcode_kits SQK-RBK114.24 --trim_barcodes; min_score threshold default 60). Summary statistics were obtained with NanoPlot (v1.44.0) (--verbose --tsv_stats –N50 --fastq) [73]. Reference guided filtration with a 1000 bp threshold was achieved using FiltLong (v0.2.1), and blasting each strain against Pseudomonas aeruginosa reference strains PAO1 (--assembly Ref_PAO1.fasta --trim --min_length 1000 --keep_percent 90) [74]. Genomes were de novo assembled using Flye (v2.8.1-b1676) (--nano-corr --genome-size 5m --asm-coverage 50 --plasmids --trestle) [75]. Assembled contigs were polished with Medaka (v2.0.1) (medaka_consensus -t 8 -m dna_r10.4.1_e8.2_400bps_hac@v4.1.0:variant) [76]. Genomic completeness and contamination were derived by CheckM2 (v1.0.2) (checkm2 predict --threads 30 –x fna) [77].

The NCBI Prokaryotic Genome Annotation Pipeline (PGAP) was used to annotate genomes and find out the total numbers of coding sequences, rRNA and tRNA [78].

Multilocus sequence types (MLSTs) were determined uploading the genomes on Pathogenwatch website (https://pathogen.watch/), while eBURST software was used to visualize clonal complexes [20]. Finally pan-genomes were visualized using both on online tool (IPGA) [19] and a command-line pipeline (ANVI’O) [79].

4.5.4. Bioinformatic Analysis of Antimicrobial Resistance Genes and Virulence Factors

To ascertain the presence of antibiotic resistance genes (ARGs), the Comprehensive Antibiotic Resistance Database (CARD), the National Centre for Biotechnology Information (NCBI) database, the ResFinder database, the Plasmidfinder and the virulence factor database database (all updated on 4 November 2023) were used in the analysis of genomes using Abricate (v1.0.1) (abricate /path/to/fna/*.fna --db card,vfdb,resfinder,ncbi,plasmidfinder --minid 95 --csv > /path/to/output/.xlsx) [80,81,82,83,84,85,86,87]. For all these analyses, a threshold identity ≥95% was set.

4.6. Statistical Analysis

To analyse associations in the distribution of P. aeruginosa strains across sample origins and antibiotic phenotypes comparison, a chi-squared test (Chi²) and Fisher's test were performed at a 95% confidence interval (α = 0.05) to assess if observed species distributions differed significantly across origins. All statistical analyses in this study were performed with SPSS.

5. Conclusions

This study investigated the resistance patterns of P. aeruginosa isolates from cats and dogs. In Algeria, this is the first identification of carbapenem resistant P. aeruginosa (31,56%) in pets with an arsenal of resistance genes mainly those related to aminoglycosides (Aph(3’)IIb), beta-lactams (bla_OXA, bla_PAO, and bla_PDC), phenicols (catB7), fosfomycin (FosA) and bicyclomicin (bcr-1) resistance with the emergence of high-risk clone ST244. High capacity of biofilm production (42% strong producers) and a wide range of virulence genes was associated with third system secretion, quorum sensing and others. The research underscores the need of comprehending P. aeruginosa resistance patterns across diverse populations and areas, promoting judicious antibiotic utilisation and stringent infection control measures to avert the dissemination of resistant strains. Furthermore, the monitoring of antibiotic resistance in companion animals should be intensified to prevent potential transmission of the infections between animals and their owners.