Molecular Detection of Carbapenemase-Encoding Genes in Multidrug-Resistant Acinetobacter baumannii Clinical Isolates in South Africa.

INTRODUCTION: Carbapenem-resistant Acinetobacter baumannii has been responsible for an increasing number of hospital-acquired infections globally. The study investigated the prevalence of carbapenemase-encoding genes in clinical multidrug-resistant A. baumannii strains.
MATERIALS AND METHODS: A total of 100 nonduplicate multidrug-resistant A. baumannii strains were cultured from clinical samples obtained from healthcare facilities in the O. R. Tambo district. The strains were confirmed by detecting the intrinsic bla OXA-51-like gene. Antimicrobial susceptibility testing was performed by VITEK® 2 and autoSCAN-4 systems. The MIC of imipenem and meropenem was rechecked by E-test. Colistin MIC was confirmed by the broth microdilution method. Real-time PCR was performed to investigate the presence of carbapenemase-encoding genes.
RESULTS: Most strains showed high resistance rates (>80%) to the antibiotics tested. Resistance to amikacin, tetracycline, and tigecycline were 50%, 64%, and 48%, respectively. All strains were fully susceptible to colistin. The bla OXA-51-like was detected in all strains whilst bla OXA-23-like, bla OXA-58-like, bla OXA-24-like, bla IMP-1, bla VIM, and bla NDM-1 were found in 70%, 8%, 5%, 4%, 3%, and 2% of strains, respectively. None of the tested strains harboured the genes bla SIM and bla AmpC. The coexistence of bla OXA-23-like, and bla IMP-1 or bla OXA-58-like was detected in 1% and 2% strains, respectively. A distinct feature of our findings was the coharbouring of the genes bla OXA-23-like, bla OXA-58-like, and bla IMP-1 in 2% strains, and this is the first report in the Eastern Cape Province, South Africa. The intI1 was carried in 80% of tested strains whilst ISAba1/bla OXA-51-like and ISAba1/bla OXA-23-like were detected in 15% and 40% of the strains, respectively. The detection of bla OXA-23-like, ISAba1/bla OXA-51-like, ISAba1/bla OXA-23-like, and bla OXA-23-like, bla OXA-58-like, and bla IMP-1 carbapenemases in strains had a significant effect on both imipenem and meropenem MICs.
CONCLUSIONS: Results showed a high level of oxacillinases producing A. baumannii circulating in our study setting, highlighting the need for local molecular surveillance to inform appropriate management and prevention strategies.

1. Introduction

Acinetobacter baumannii, an opportunistic Gram-negative bacterium, has become a predominant cause of healthcare-associated infections worldwide [1]. This microorganism has emerged to become a major concern for clinicians worldwide causing infections such as septicaemia, meningitis, endocarditis, ventilator associated pneumonia, burns, urinary tract infections, surgical site infections, and wound infections in hospitals due to its intrinsic as well as its remarkable propensity to rapidly acquire resistance determinants to a wide range of antibacterial agents [2, 3]. The World Health Organization (WHO) lists A. baumannii among critical antibiotic-resistant “priority pathogens,” highlighting its serious threats to public health [4]. Risk factors for A. baumannii infections include immune suppression, burns, trauma, mechanical ventilation, catheters, invasive medical procedures, previous antibiotic treatment, an extended hospital stay (>90 days), and underlying diseases such as diabetes [5].

Carbapenem has been the drug of choice used to treat MDR A. baumannii infections over the past decade because they are highly efficacious and have low toxicity [6]. However, the resistance rate against carbapenem has increased dramatically due to its frequent application in recent years and is mainly mediated by the production of carbapenem-hydrolysing enzymes [7]. The most common mechanism of carbapenem resistance in A. baumannii is the production of β-lactamases, including enzymes of Ambler Classes D OXA-type carbapenemases consisting of blaOXA–23-like (OXA–23, OXA–27 and OXA–49), blaOXA–40-like, (OXA–24/40, OXA–25, OXA–26 and OXA–72), blaOXA–58-like (OXA–58 and OXA–96), blaOXA–143-like and blaOXA–51-like enzymes, and B Metallo- β -lactamases (MBLs) mainly blaVIM, blaIMP-1, blaNDM-1, blaSPM, blaGIM, and blaSIM types [8]. Among the multiple types of MBL genes described throughout the world, the blaIMP-1, blaVIM, and blaSIM types are the most common [9]. The upstream of OXA-type class D carbapenemases in Acinetobacter is often associated with insertion sequence (IS), of which ISAba1 is the most commonly detected. ISAba1 and other IS may modulate the expression and transfer of OXA-type carbapenemase genes [10]. Integrons have been demonstrated to play a major role in the dissemination of multidrug-resistance among Gram-negative bacteria through the capture and expression of resistant genes embedded within cassettes [11].

Studies in Africa and the Gulf Cooperation Council (e.g., Saudi Arabia) have reported increasing carbapenem resistance among A. baumannii isolates [12-14]. Since nosocomial A. baumannii infection has generated an abundance of therapeutic obstacles for the treatment of hospitalized patients in South Africa, as well as other countries, the knowledge regarding the prevalence level of antibiotic resistance genes and the resistance pattern of this bacterium to the initial antibiotics is required to control, prevent, and cure A. baumannii driven infections. Therefore, the aim of the current study was to investigate the rates of carbapenem resistance as well as to detect the frequency of carbapenemase-encoding genes in MDR A. baumannii isolates obtained from hospitalized patients receiving treatment in healthcare facilities in O.R. Tambo district in the rural Eastern Cape Province of South Africa.

2. Materials and Methods

2.1. Study Design and Settings

This was a prospective cross-sectional study conducted at healthcare facilities in the OR Tambo district municipality, Eastern Cape, South Africa. The municipality is formed by five local municipalities (King Sabata Dalindyebo, Nyandeni, Mhlontlo, Port St Johns, and Ingquza Hill) with an estimated total population of 1,760,389. Healthcare services are delivered by 1 Academic Central Hospital, 1 Regional Hospital, 12 District Hospitals; 11 Community Health Centers, 49 clinics, 52 Health Posts, and 15 mobiles. Patients' clinical samples collected from all these facilities are sent for culture and susceptibility testing in the Department of Medical Microbiology at the National Health Laboratory Services (NHLS), located in the Nelson Mandela Central Hospital in Mthatha, Eastern Cape. Clinical samples from those various hospitals and clinics were sent as part of the patients' routine standard of care.

2.2. Data Collection, Bacterial Identification, and Storage of A. baumannii Isolates

From August 2016 to July 2017, a total of 100 nonduplicate nonconsecutive MDR A. baumannii strains were collected arbitrarily from blood, pus, urine, sputa, catheter tips, tracheal aspirates, and cerebrospinal fluid of patients from the Microbiology Laboratory at the National Health Laboratory Services (NHLS) at NMAH, Mthatha (see Supplementary Materials). The strains were preliminarily speciated phenotypically using standard microbiological procedures: the VITEK® 2 automated machine with Gram-negative identification (GNI) cards (BioMérieux, France) and the MicroScan autoSCAN-4 System (Dade Behring Inc., Deerfield, IL) with Gram-negative ID type 2 panel. A high percentage (≤95%) was utilized as the acceptance criterion for identification by the MicroScan autoSCAN system. Suspected colonies were also further verified using the Acinetobacter specific primer set Ac436F and Ac676r to amplify the 16S rRNA gene. Confirmation of A. baumannii strains was carried out by polymerase chain reaction analysis of the presence of inherent blaOXA-51-like genes. All of the strains were stored at −80°C in skim milk with 15% glycerol until further use.

2.3. Antimicrobial Susceptibility Testing

A total of 18 clinically relevant antibiotics were tested using MicroScan autoSCAN-4 Gram-negative MIC panel, and results were analyzed and interpreted according to the recommended clinical breakpoints given in the clinical and laboratory standard institutes (CLSI) guidelines v27 [15]. These antibiotics were amikacin (AMK), ampicillin/sulbactam (AMS), cefepime (FEP), cefotaxime (CTX), ceftazidime (CAZ), ceftriaxone (CRO), ciprofloxacin (CIP), gentamicin (GEN), imipenem (IMP), meropenem (MEM), levofloxacin (LVX), tetracycline (TEC), tobramycin (TOB), trimethoprim/sulfamethoxazole (SXT), piperacillin/tazobactam (TZP), piperacillin (PRL), colistin (CST), and tigecycline (TGC), all from Beckman Coulter, South Africa. Determination of MICs to imipenem and meropenem rechecked by E-test minimum inhibitory concentration method using E-test strips as per the manufacturer's guidelines (BioMérieux, France) on Mueller-Hinton agar plates and CLSI breakpoints (strains displaying MICs ≥8 µg/mL for imipenem and meropenem were considered resistant). Quality control strains used in antimicrobial susceptibility testing are Escherichia coli ATCC #25922, Staphylococcus aureus ATCC 29213, and Pseudomonas aeruginosa ATCC #27853 [16]. Colistin susceptibility testing was confirmed using the broth microdilution method according to the CLSI and by ComASP™ Colistin, Liofilchem® (REF. 75001), according to the manufacturer's recommendations. Escherichia coli NCTC 13846 (mcr-1 positive) was used as a positive control for ComASP™ Colistin test. The United States Food and Drug Administration (US-FDA) approved breakpoints for members of the family Enterobacteriaceae were used as interpretative criteria for tigecycline (http://www.fda.org.uk/sitemap.aspx). Nonsusceptibility was defined as a combination of resistance and intermediate resistance. MDR A. baumannii isolates were defined as acquired nonsusceptibility to at least one agent in three or more antimicrobial categories [17]. All results were within quality control ranges.

2.4. Detection of Carbapenem-Resistant Genes

Genomic DNA from an overnight culture on tryptic soy broth of A. baumannii was extracted using the MagNaPure Compact® nucleic acid isolation kit I (ref. no. 03730964001, Roche Diagnostics, Mannheim, Germany) according to the manufacturer's protocol. A final elution volume of 200 µL of pure A. baumannii DNA was used. 2 µL of the DNA extract was used for each PCR analysis. The extracted DNA was stored at −80°C until required for further analysis. All MDR isolates were systematically screened by LightCycler 2.0 real-time PCR (Roche Diagnostics) instrument targeting some of the epidemiologically most relevant resistant genes. The presence of the genes encoding Amber class D Serine-Carbapenemase genes (blaOXA-51-like, blaOXA-23-like, blaOXA-24-like, and blaOXA-58-like), Ambler class B Metallo-β-lactamases (MBLs) (blaIMP-1, blaVIM, blaSIM, and blaNDM-1), and Ambler class C blaAmpC were assessed. MDR A. baumannii strains were subsequently screened for the presence or absence of IS element ISAba1 [18] and class 1 integrons by real-time PCR [19]. To determine the overexpression of blaOXA-23-like and blaOXA-51-like families due to the upstream adjacent ISAba1, PCR was firstly performed for ISAba1, and in cases where blaOXA-like is present, a combination of forward primer of ISAba1 and reverse primer of blaOXA-like family was further used. Appropriate positive and negative controls were run simultaneously (Table 1). The primers used for PCR amplification of the carbapenemase genes are listed in Table 2.

Table 1

Quality control (QC) organisms used for the detection of resistant genes in this study.

Organism	ATCC or NCTC number	Resistant gene(s)
A. baumannii	NCTC 13302	OXA-26 (OXA-24-like; OXA-51-like
A. baumannii	NCTC 13305	OXA-58
Klebsiella pneumoniae	ATCC-BAA 2146	NDM
A. baumannii	NCTC 13301	OXA-23
A. baumannii	NCTC 13303	OXA-26 (OXA-51-like)
A. baumannii	NCTC 13304	OXA-27 (OXA-51-like)
Enterobacter cloacae	NCTC 13405	AmpC
Pseudomonas aeruginosa	NSCC stock	VIM
A. baumannii	ATCC 19606	OXA-51
Escherichia coli	NCTC 13476	IMP-1
Escherichia coli	Se 131 (accession number: AJ238350)	intI1
Pseudomonas aeruginosa	ATCC 27853	OXA-24-like; OXA-51-like
Klebsiella pneumoniae	ATCC 8303	SIM-1
A. baumannii	AYE strain	AmpC, ISAba1, OXA-51-like

Table 2

List of primers used for PCR amplification of resistant genes.

Gene	Oligonucleotide sequence	Fragment size (bp)	Location	References
16SrRNA	F: 5′TTT AAG CGA GGA GGA GG3′R: 5′ATT CTA CCA TCC TCT CCC3′	240	16SrRNA	[20]
OXA-23-like	F: 5′GATCGGATTGGAGAACCAGA3′R: 5′ATTTCTGACCGCATTTCCAT3′	501	bla OXA-23	[21]
OXA-24-like	F: 5′GGTTAGTTG GCC CCC TTA AA3′R: 5′AGTTGAGCGAAAAGGGGATT3′	246	bla OXA-24	[21]
OXA-51-like	F: 5′TAATGCTTT GATCGG CCT TG3′R: 5′TGGATTGCACTT CAT CTT GG3′	353	bla OXA-51	[21]
OXA-58-like	F: 5′AAGTATTGGGGCTTGTGCTG-3′R: 5′CCCCTCTGCGCTCTACATAC3′	599	bla OXA-58	[21]
IMP-1	F: 5′GATGGTATGGTGGCTCTTGT3′R: 5′TTAATTTGCCGGACTTAGGC3′	448	Bla IMP	[22]
NDM-1	F: 5′ATTAGCCGCTGCATTGAT3′R: 5′CATGTCGAGATAGGAAGTG3′	154	bla NDM	[23]
VIM-like	F: 5′ACTCACCCCCATGGAGTTTT3′R: 5′ACGACTGAGCGATTTGTGTG3′	815	bla VIM	[24]
SIM-1-like	F: 5′TAATGCTTT GATCGG CCT TG3′R: 5′TGGATTGCACTT CAT CTT GG3′	353	bla SIM	[21]
AmpC	F: ACAGAGGAGCTAATCATGCGR: GTTCTTTTAAACCATATACC	1243	bla AmpC	[25]
ISAba1	F: 5-CACGAATGCAGAAGTTG-3R: 5-CGACGAATACTATGACAC-3	599	ISAba1	[18]
intI1	F: 5′CAG TGG ACA TAA GCC TGT TC3′R: 5′CCC GAC GCA TAG ACT GTA3′	160	Integrase gene	[19]

2.5. Statistical Analyses

Data analyses were performed using the Statistical Package for Social Sciences Statistics for Windows software package version 23.0 (IBM, Armonk, NY). Results were expressed using frequency and percentages for qualitative variables and as mean (standard deviation) or median (interquartile range: IQR) for quantitative variables. The association between A. baumannii and drug resistance was analyzed by the chi-square test. All P values were two-sided. A P-value <0.05 was considered statistically significant.

2.6. Ethical Considerations

Ethical approval for this project was obtained from the Walter Sisulu University Research Ethics and Biosafety Committee (Reference number: 019/2016). No written informed consent was required because each MDR A. baumannii isolate obtained was delinked with the patient's personal information. Also authors had no direct interactions with patients from whom these isolates were isolated.

3. Results

3.1. Demographic Characteristics of the Study Population

The demographic characteristics of the patients with A. baumannii infection analyzed in our study showed a slight female preponderance of 47% males versus 53% females). The age of the overall patients ranged between 1 day and 91 years (mean 44.4 years, SD ± 36.3 years). A. baumannii were isolated highest from tracheal aspirate (32%) and lowest from cerebrospinal fluid (2%). A. baumannii was most prevalent in the ICU wards 41% followed by neonatal ward 22%, surgical ward 15%, medical ward 13%, and general ward 9%. A. baumannii was isolated from various age groups but was the most prevalent in age group 0–9 years (Supplementary Materials).

3.2. Antimicrobial Susceptibility Testing

Generally, the results for antimicrobial susceptibility testing for the 100 MDR A. baumannii revealed that A. baumannii strains showed a high resistance rate to most of the commonly used antibiotics tested. Overall, ≥81% of all the strains were resistant to extended-spectrum cephalosporins, fluoroquinolones, and carbapenems antibiotics. Resistance to aminoglycosides ranges from 50% to 87% whilst 48% of strains were resistant to tigecycline. All strains were susceptible to colistin. Findings of the antimicrobial susceptibility patterns of all A. baumannii strains in this study are shown in Figure 1.

Figure 1

Antimicrobial resistance profile of 100 A. baumannii isolates. IMP: imipenem; MEM: meropenem; FEP: cefepime; CAZ: ceftazidime; CTX: cefotaxime; CRO: ceftriaxone; AMS: ampicillin/sulbactam; PRL: piperacillin; TZP: piperacillin/tazobactam; AMK: amikacin; TOB: tobramycin; GEN: gentamicin; TEC: tetracycline; CIP: ciprofloxacin; LVX: levofloxacin; SXT: trimethoprim-sulfamethoxazole; TGC: tigecycline; CST: colistin.

3.3. PCR Results of Carbapenemase-Encoding Genes

Results showed that all of the strains investigated in this study had the intrinsic blaOXA-51-like gene. Among them, 70% were positive for the blaOXA-23-like, 8% were positive for blaOXA-58-like, and 5% positive for blaOXA-24-like. For the MBLs, 4% strains were positive for blaIMP-1, 3% carried blaVIM, 2% carried blaNDM-1, and none of the tested isolates harboured the blaSIM. Alleles encoding blaAmpC enzymes were not detected in any strain. The ISAba1 was located upstream the blaOXA-51-like and the blaOXA-23-like carrying strains 15% and 40%, respectively. Some of the strains coharboured different carbapenemase genes mostly yielding very broad-spectrum antibiotic resistance profiles to imipenem and meropenem. Both strains possessing the ISAba1/blaOXA-23-like and blaOXA-23-like, blaOXA-58-like, and blaIMP-1 genes had the highest carbapenem resistance with these strains having imipenem MICs of 32 µg/mL or more, and having meropenem MICs of 32 µg/mL or more. This was followed by the strains carrying the blaOXA-23-like gene with 62% having imipenem MICs of 32 µg/mL or more, and 58% having meropenem MICs of 32 µg/mL or more (Figures 2 and 3). Screening for the presence of integron integrase genes showed the presence of the class 1 integrase gene in 80% MDR A. baumannii strains (Table 3).

Figure 2

Distribution of imipenem MICs in 100 CRAB isolates with the blaOXA-23-like, blaOXA-58-like, ISAba1/blaOXA-23-like, ISAba1/blaOXA-51-like and blaOXA-23-like, blaOXA-58-like and blaIMP-1 genes.

Figure 3

Distribution of meropenem MICs in 100 CRAB isolates with the blaOXA-23-like, blaOXA-58-like, ISAba1/blaOXA-23-like, ISAba1/blaOXA-51-like and blaOXA-23-like, blaOXA-58-like and blaIMP-1 genes.

Table 3

Percentage distribution of carbapenemases-encoding genes in 100 CRAB strains.

Carbapenemases-encoding genes	Number (%)
bla OXA-51-like only	0
bla OXA-51-like and blaOXA-23-like	23
bla OXA-51-like and blaOXA-58-like	4
bla OXA-51-like and blaOXA-24-like	5
bla OXA-51-like and blaIMP-1	1
bla OXA-51-like and blaVIM	3
bla OXA-51-like and blaNDM-1	2
ISAba1/blaOXA-51-like	15
bla OXA-51-like and ISAba1/blaOXA23-like	40
bla OXA-51-like and blaOXA23-like and blaIMP-1	1
bla OXA-51-like and blaOXA-23-like and blaOXA-58-like	2
bla OXA-51-like and blaOXA-23-like and blaOXA-58-like and blaIMP-1	2
ISAba1/blaOXA-51-like and ISAba1/blaOXA-23-like	2
intI1 only	0

4. Discussion

Understanding the fundamental mechanism that underlines A. baumannii infections, including original source of infecting strains, resistance patterns, and genes responsible for the development of resistance is critical for the development of appropriate infection control measures and more efficient treatment strategies [5]. Data from countries that lack routine surveillance, including South Africa, are necessary in order to understand and prevent the dissemination of carbapenem-resistant strains of species that are in epidemiological expansion, like A. baumannii.

Results from the present study revealed a 100% resistant rate of MDR A. baumannii to third-generation cephalosporins (ceftazidime, cefotaxime, and ceftriaxone). Resistance to cefepime was 87%. All strains were resistant to piperacillin, piperacillin/tazobactam, trimethoprim/sulfamethoxazole, and ampicillin/sulbactam. The percentage rates of resistance to remaining antibiotics range from 48% to 89% with the exception of colistin which had no resistant strain detected. Carbapenem resistance of A. baumannii is a major concern since it is the drug of choice in the treatment of A. baumannii infections [26]. The results showed that the majority of A. baumannii strains tested were resistant to both imipenem 81% and meropenem 83%. Lowings et al. found all MDR A. baumannii isolates showed high level resistance (cefotaxime 100%; cefotaxime 100%; ceftazidime 89%; cefepime 90%; imipenem 86%; and meropenem 86%) in a study conducted at the Department of Medical Microbiology, Prinshof Campus, University of Pretoria/National Health Laboratory Service (NHLS) in 2013 [27]. A previous study by Kock et al. in the same setting showed that the overall percentage of resistance to the cephalosporin antibiotics was cefepime (62%) and ceftazidime (45%) [28]. A report on antimicrobial resistance surveillance from sentinel public hospitals in South Africa in 2013 showed a resistance rate of 80% to cefepime and 73% to ceftazidime [29]. Comparing the results of the present study with these earlier studies indicates that the resistant rate of MDR A. baumannii to various antibiotics has not decreased. High levels of resistance to third and fourth generation cephalosporins in this study indicated that cephalosporins are no longer efficacious in the treatment of MDR A. baumannii isolated in South Africa. The full susceptibility of all MDR A. baumannii to colistin in this study indicated that colistin is still a better option of the drug for the treatment of infections caused by A. baumannii in South African hospitals. However, the issue of nephrotoxicity, neurotoxicity, colistin-resistance, and heteroresistance shown by colistin monotherapy is a challenge in the management of this infection [30]. The high resistance rate to other antibiotic categories may probably be due to the heavy selection pressure from overuse of these antibiotics.

Although A. baumannii infections are observed in all body parts, they are mostly observed in the respiratory system. Our study showed that A. baumannii isolates were predominantly found in the respiratory samples (tracheal aspirate) and the most frequent setting for MDR A. baumannii was found in the ICUs (41%). ICUs (Neonatal and general intensive care) are settings where most patients have indwelling devices and invasive procedures. Guckan et al. and Lowings et al. isolated 85.2% and 12% A. baumannii isolates, respectively, from ICUs [27, 31].

Molecular detection of carbapenemase-encoding genes has shown that the production of OXA-type carbapenemases is predominant in A. baumannii [30, 32]. The chromosomally encoded blaOXA-51-like gene was detected in all the strains supporting those of other studies demonstrating that the detection of the blaOXA-51-like gene can be used as a supplementary tool to identify A. baumannii at the species level together with additional methods [33]. It was observed that the expression of blaOXA-23-like was the second dominant carbapenem resistance mechanism (70%) in MDR A. baumannii among the strains investigated aside the intrinsic blaOXA-51-like gene. Its prevalence among isolates was higher than this in previous reports in South Africa. Its prevalence among strains was higher in this study than in the previous report (59%) by Kock et al., whilst Lowings et al. reported a prevalence of 77%. The blaOXA-23-like gene is considered a virulence biomarker and a significant cause of carbapenem resistance worldwide [34]. This gene was first reported in 1985 in Scotland and since then outbreaks of OXA-23 carbapenemase-producing A. baumannii have been reported all over the world and is a dominant genetic determinant in the Asia subcontinent [35]. In Latin America, Brazil was the first country to report OXA-23-like-producing isolates in 2003, followed by Colombia, Argentina, and Mexico some years later [36].

This study also detected blaOXA-24-like and blaOXA-58-like genes in 5% and 8% isolates, respectively. Although the acquired blaOXA-23-like is the dominant genetic determinant reported in most studies in South Africa, the detection of blaOXA-58-like and blaOXA-24-like genes in our study shows that these genes are becoming significant in carbapenem resistance in A. baumannii and may be responsible for the increase in carbapenem resistance in A. baumannii. It also suggested that these oxacillinases may locally spread in Mthatha, South Africa. These findings concur with a study by Ji et al. in China, who reported that blaOXA-23-like are the most frequent carbapenem-resistant A. baumannii, whereas blaOXA-24-like and blaOXA-58-like genes pose potential threats of hospital outbreaks of multidrug-resistant A. baumannii [37]. For class B metallo-β-lactamases, we found that 4%, 3%, and 2% of the isolates had blaIMP-1, blaVIM, and blaNDM-1, respectively, and no isolate was found to produce blaSIM gene. None of the clinical A. baumannii isolates harboured blaAmpC. IMP (primarily detected in South Korea) and NDM-1 (recently reported in many countries, such as India, Israel, Egypt, Germany, Spain, Switzerland, the United Arab Emirates, and China) are the two MBL genes most frequently detected in A. baumannii isolates [7, 33]. The blaNDM-1 producing bacteria are mainly reported in Enterobacteriaceae [38]. This gene was first identified from Klebsiella pneumoniae and E. coli strains from a patient previously hospitalized in India and has recently emerged in A. baumannii isolates [32]. Taking into consideration the diversity of the South African community (prominence of Asian and European communities where blaNDM-I have been reported) and the frequency of travels made by individuals between these countries, it is believed that the spread of blaNDM-1 was likely to occur mostly through A. baumannii rather than Enterobacteriaceae. However, these findings need further investigation.

Interestingly, this study showed that some isolates were found to coharbour carbapenemase genes. Among these isolates include the coharbouring of carbapenemase genes blaOXA-23-like and blaIMP-1 in 1% isolate, blaOXA-23-like and blaOXA-58-like in 2% isolates, and blaOXA-23-like, blaOXA-58-like, and blaIMP-1 in 2% isolates. To the best of our knowledge, this is the first report of the coharbouring of blaOXA-23-like, blaOXA-58-like, and blaIMP-1 in the Eastern Cape province of South Africa. We believe that coharbouring of these resistance genes is due to the presence of different genes in the same strain. This warrants the use of molecular typing methods to determine the clonal types of the isolates. The coharbouring of these major resistance mechanisms seriously limits therapeutic options, raising concerns regarding their transmission to other organisms. For example, we found that all isolates that coharboured blaOXA-23-like, blaOXA-58-like, and blaIMP-1 had imipenem MICs of 32 μg/mL or more and meropenem MICs of 64 μg/mL.

The study showed that carbapenem susceptibility phenotypes of our A. baumannii isolates clearly correlated with the presence or absence, in the genomes, of ISAba1-linked blaOXA-23-like or blaOXA-51-like genes (Figures 2 and 3). For instance, the 100% MDR isolates with ISAba1 elements upstream of the blaOXA-23-like had imipenem and meropenem MICs 32 μg/mL or more, whereas 83.34 and 77.78% of MDR isolates with ISAba1 elements upstream of the blaOXA-51-like had imipenem and meropenem MICs 32 μg/mL, respectively. Presumably, the ISAba1 element likely provides exogenous promoter functions to overexpress the linked genes. Evidence showed that carbapenem resistance mediated by overexpression of blaOXA-like families due to upstream IS elements is more prevalent [39-42]. Our data are consistent with the findings of previous studies in that acquisition ISAba1 are the main mechanism of carbapenem resistance among A. baumannii by acting as a promoter to increase oxacillinase expression [9, 39, 43]. However, there are limited studies in South Africa as most of the studies evaluated the prevalence of OXA-like families in detail. There is, therefore, the need for further investigations to find the contribution of other IS elements (ISAba2, ISAba3) upstream to OXA-like families in different parts of the country. The results suggested that class 1 integrons did not contribute resistance to carbapenems since its presence did not have an effect on the MICs of carbapenems.

4.1. Limitations

The study has some limitations. Firstly, the study was conducted in a single centre, NMAH in the Tambo district municipality, so our findings may not be generalized to other settings. Secondly, the study did not include molecular typing epidemiological study reference methods such as PFGE, MLST, REP-PCR, etc. which would have better evaluated the dissemination of a carbapenemase or some clones to understand CRAB isolates in the clinical setting. These studies are necessary to help highlight the need for area-specific surveillance to inform appropriate management.

5. Conclusion

In conclusion, the results of this study provide direct evidence of the spread of carbapenemase-encoding genes especially blaOXA-23-like in our setting. High presence of the ISAba1 element upstream of the blaOXA-51-like and blaOXA-23-like increases expression of these carbapenemases and presents an emerging threat in this study setting. The results also provided clinically relevant insights into antimicrobial susceptibility and knowledge of the prevalent resistance mechanism among CRAB. This information will help in preventing the development of MDR A. baumannii in hospitals and also helps clinicians in the empirical treatment of patients. To prevent the development of Carbapenem-resistant A. baumannii, routine implementation of simple and cost-effective screening methods to detect carbapenemase production in hospitals is crucial in enhancing infection control practices and finally establishing antimicrobial stewardship programs. Contribution and interplay of different mechanisms to the rapid increase in antibiotic resistance of A. baumannii in South Africa warrant further investigation.