Characterisation of the Prevailing Multidrug Pseudomonas aeruginosa Strains from Surgical Wound Using 16S rRNA Sequencing Technique

Background Pseudomonas aeruginosa (P. aeruginosa) is prevalent in hospital-acquired surgical wound infections. It exhibits both innate and acquired resistance to a broad range of antimicrobials and remains a principal problem in clinical practice. Methods In total, 284 sterile surgical wound swabs (142 each) were collected from two government hospitals: Central Hospital Benin (CHB) and University of Benin Teaching Hospital (UBTH) in Benin City, Nigeria. Pseudomonas spp. isolated from both hospitals were screened with eight different antibiotics by way of disk diffusion method. Polymerase chain reaction (PCR) amplification of 34 multiple drug-resistant isolates was carried out using genus-specific primer set on extracted genomic DNA for the identification of Pseudomonas spp. and substituent 16S rRNA sequencing to determine the prevailing strains in the two locations. Results Sixty-two Pseudomonas spp. were isolated from the two locations (27 isolates from CHB and 35 isolates from the UBTH). Surgical wound infections screened with regularly used antibiotics revealed that 17 (62.9%) isolates from CHB and 20 (57.1%) isolates from UBTH were multiple drug resistant Pseudomonas spp. PCR identification using Pseudomonas spp. specific primer showed that 16 (94.1%) isolates from CHB and 18 (90%) isolates from UBTH were confirmed. The 16S DNA sequencing revealed that P. aeruginosa strain H25883 was dominant in both locations. Conclusion High antibiotic resistance among P. aeruginosa isolates was established in our study. PCR technique revealed a more reliable method of bacterial identification. H25883 strain of P. aeruginosa is the prevalent strain in both locations and it should be given attention in nosocomial surgical wound infections.


Introduction
Post-surgical wound infection is the major source of nosocomial infection in surgical patients, accounting for 39.9% of all infections. It mainly causes post-operative morbidity, resulting in longer hospital stay, increased hospital bill and incidences of postoperative death. Generally, wound infections are a result of wound contamination caused by endogenous bacteria from the patient's skin, mucous membrane or hollow viscera (1). The development of an infection in any wound is subjective largely to the virulent nature of the microorganism and immunity of the patient. Nevertheless, when pus oozes from a closed surgical opening along with signs of inflammation in the adjoining tissues, it is referred to as wound infection (2,3).
Pseudomonas aeruginosa (P. aeruginosa) is a gram-negative bacterium. It is nonsporous, motile and a facultative anaerobe. It is accountable for a wide range of diseases in both humans and animals (4). Generally, P. aeruginosa is an opportunistic and nosocomial infectious organism that could develop infections in burns, injury, surgical wounds and in immunocompromised subjects (5,6). Incidences of P. aeruginosa infections are on the rise worldwide due to their mechanisms of survival, adaptation and resistance to different types of antibiotics (7). Wound infection caused by P. aeruginosa is considered a major cause of morbidity and mortality (8). The immense use of routine broad-spectrum antibiotics has increased the resistance of P. aeruginosa to clinical drugs, which has led to serious therapeutic problems (9). Thus, timely and precise diagnosis is essential for proper treatment and also to control future disease outbreaks. A wide range of diagnostic methods have been established for P. aeruginosa identification. They include phenotypic methods (10), electrochemical techniques (11) such as enzyme-linked immunosorbent assay (12), and molecular methods such as polimerase chain reaction (PCR) (13), real-time PCR (14,15), and particularly 16S DNA sequencing (16). Despite the existing and extensive reports on the prevalence of P. aeruginosa in hospital environments, there is still a paucity of research finding on molecular identification of multidrug P. aeruginosa strains from surgical wounds particularly in Benin City, Nigeria. Therefore, the present study sought to identify prevailing and phylogenetic tree were constructed using neighbour-joining method as described by Agbonlahor et al. (21).

Statistical Analysis
Percentage multiple drug resistance isolates was calculated using the following equation: Number of MDR isolates from location × 100 Total number of isolates from location 1

Biochemical Characterisation of Bacterial Isolates and Distribution of Etiologic Agents of Surgical Wound Infection
Two hundred and eighty-four postoperative wound swabs specimens were collected from patients in CHB and UBTH both in Benin City and analysed. A total of 99 (35%) of patients studied had wound infections. Phenotypic identification of these bacterial isolates using morphological and biochemical tests revealed rod shaped, Gram negative, motile, catalase, oxidase, glucose and citrate positive isolates as well lactose, urease, mannitol, coagulase negative which was suggestive of Pseudomonas spp. as shown in Table 2

PCR Product Purification and Sequencing
Amplification and sequencing were done as described by Agbonlahor et al. (21), with the following modifications: Purification was done with the Applied Biosystems Incorporation (ABI) V3.1 Big dye kit according to manufacturer's instructions. The labeled products were then cleaned with the Zymo Seq clean-up kit (USA) in accordance with manufacturer's instructions. The ultra-pure DNA was sequenced with ABI3500XL analyser at Functional Bioscience, Madison, USA. Sequences data generated were analysed with Geneious version 9.0.5  (20) transilluminator. In addition, a similar band was also seen for positive control strain with American type culture collection number 27852. As expected, no band was seen in the negative control where nuclease free water was used instead of bacterial DNA as shown in

16S rRNA Sanger Sequencing of Pseudomonas spp.
Sequencing of Pseudomonas spp. isolates were carried out to further identify the MDR Pseudomonas spp. isolates to the strain level. Phylogenetic tree of isolates revealed different P. aeruginosa strains for all 34 Pseudomonas spp. as exemplified in Figures 5-8.

Prevalence of MDR P. aeruginosa Strains from CHB and UBTH
The percentage occurrence of MDR P. aeruginosa strains among sequenced isolates from CHB revealed that P. aeruginosa strains H25883 had the highest percentage occurrence of 18.75% followed by P. aeruginosa strains resistance (85.2%) in isolates from CHB while isolates from UBTH showed highest resistance against nitrofuration (77.1%) followed by (68.6%) observed for gentamycin ( Table 4).

Amplification of Pseudomonas spp.
PCR amplification using Pseudomonas spp. specific primer set indicated that 16 (94.1%) suspected Pseudomonas spp. isolates from CHB and 18 (90.0%) suspected Pseudomonas spp. isolates from UBTH were confirmed to be Pseudomonas spp. with bands at 618 base pair which were clearly visible under UV

Proteus mirabilis Staphylococcus aureus
Notes: +ve = positive; -ve = negative; Group A, B, C, D = different isolates  Ciprofloxacin (5 µg as the microbe has intrinsic resistance to a large number of antimicrobial agents. Furthermore, with the acquisition of antibiotic-resistant genes, it is becoming more difficult to cure infections caused by this organism (24). Despite the existing and extensive reports on the prevalence of P. aeruginosa in hospital environments, there is still a paucity of research finding on molecular identification of multidrug P. aeruginosa strains from surgical wounds, particularly in Benin City, Nigeria. Hence, data from the present study revealed that the P. aeruginosa strain showed the highest antibiotic resistance to ceftazidime in isolates from CHB and nitrofurantoin in isolates from the UBTH. Lowest resistance was observed for AR7-520 and PA006 with 12.5%, respectively (Figure 9). In the same vein, P. aeruginosa strains H25883 also recorded the highest percentage occurrence of (22.22%) in UBTH followed by P. aeruginosa strains KAR21 with 11% as shown in Figure 10.

Discussion
P. aeruginosa, a non-fermentative gramnegative bacterium, is currently the second most widespread nosocomial bacterium, after Acinetobacter species (22). P. aeruginosa broadly exists in hospital environments (23) and medical equipment (22  (94.1%) out of 17 isolates were confirmed to be Pseudomonas spp. in CHB, while 18 (90%) out of 20 were confirmed to be Pseudomonas spp. with bands at 618 base pair for test isolates and positive control strains. The above-mentioned genus-specific PCR assays indicated that three clinical isolates had been misidentified using phenotypic laboratory methods. This signifies the efficiency of the molecular characterisation method over phenotypic characterisation. With regard to a study by Spilker et al. (19), the genusspecific PCR assays indicated that several of the 66 clinical isolates were misidentified by the referring laboratories (19).
ciprofloxacin for both locations. In comparison to previously reported data, the result of the present study corroborates the finding of Carroll et al. (25) and Leone et al. (26) that also reported a high antibiotic resistance rate towards ceftazidime and gentamycin antibiotics in both clinical and environmental isolates. Additionally, the study of Ruiz et al. (27) reported that clinical bacterial isolates are less susceptible to antimicrobial agents than environmental bacterial isolates due to their selective action (27). Molecular characterisation of Pseudomonas spp. isolated from surgical wound infections specimens from both locations showed that 16 wound bacterial isolates neither to evaluate the relative precision of different phenotypic identification systems. Both assays have 100% sensitivity and specificity for their intended targets. We have also established the utility of these PCR assays in precisely identifying P. aeruginosa strains among isolates not correctly identified by phenotypic analyses. These assays should serve as a valuable accessory in the evaluation of gram-negative non-fermenting bacteria recovered from surgical wound isolates.

Conclusion
The results obtained from our study revealed that the 16S rRNA-based PCR and sequencing are highly sensitive, precise and consistent for the identification of P. aeruginosa strains isolated from post-operative surgical wound infections than conventional bacterial phenotypic methods. Our finding further highlights the use of DNA sequencing of the 16S rRNA gene as an effective tool to study bacterial phylogeny and taxonomy associations between bacteria and bacterial detection as well. Thus, early identification and control of this pathogen have become increasingly important.

Ethics of Study
Approval was obtained from the UBTH and CHB ethical committee, and all the patients consented to collection of samples after being educated on the objectives of the study.

Conflict of Interest
None.

Funds
None.
Sequence analysis of 16S rRNA is now being used as a taxonomic 'gold standard' in determining the phylogenies of bacterial species (28). The 16S rRNA gene sequences comprise hypervariable regions with high conservation that can differentiate species-specific signature sequences helpful in the classification of bacteria (29,30). Going forward, 34 (91.9%) Pseudomonas spp. were further examined by 16S rRNA sequence analysis, and in each case, the PCR assay results were consistent. Thus, when this set of isolates was assessed against the 16S rRNA sequence, the sensitivity and specificity of both PCR assays were again 100%. It has also been reported that selective amplification of Pseudomonas 16S rRNA analysis is used to detect and differentiate Pseudomonas species from clinical and environmental samples (30). The present results agree with the finding of Didelot et al. (32) who reported that 16S rRNA gene sequencing is now common in medical microbiology as a quick and inexpensive alternative to phenotypic approaches of bacterial identification.
The result from the phylogenetic trees showed that MDR P. aeruginosa strain H25883 was the predominant strain in both locations (CHB and UBTH) with 18.75% and 22.22%, respectively.
P. aeruginosa strains AR7-520 and PA006 with 12.5% were observed in CHB, making it the second predominant strains. However, many studies reported that P. aeruginosa has been mostly isolated from post-operative surgical wounds regardless of the site of infection and location of samples as a result of its high survival uniqueness in the hospital setting (33,34). It has been ranked second among nosocomial disease-causing microbes. They are isolated from hospitals frequently, contaminating hospital equipment such as sinks used for wound dressing and other surgical tools. Furthermore, many antimicrobialresistant strains continue to exist in apparently sterile hospital equipment, therefore, making it a precarious nosocomial pathogen broadly dispersed in the hospital environments where they are most difficult to eliminate (35).
The 16S rRNA sequence analysis on all identified MDR isolates was carried out and the sequence data confirmed the PCR results. Though our PCR and DNA sequence analyses showed bacterial isolates that were misidentified by phenotypic testing, we must point out that this study was not intended to determine the incidence of misidentification of post-surgical