Antibiogram, Prevalence of OXA Carbapenemase Encoding Genes, and RAPD-Genotyping of Multidrug-Resistant Acinetobacter baumannii Incriminated in Hidden Community-Acquired Infections

Acinetobacter spp. has gained fame from their ability to resist difficult conditions and their constant development of antimicrobial resistance. This study aimed to investigate the prevalence, susceptibility testing, OXA carbapenemase-encoding genes, and RAPD-genotyping of multidrug resistant Acinetobacter baumannii incriminated in hidden community-acquired infections in Egypt. The antimicrobial susceptibility testing was assessed phenotypically using Kirby–Bauer disk diffusion method. Also, Modified-Hodge test (MHT) was carried out to detect the carbapenemases production. Multiplex-PCR was used to detect the carbapenemase-encoding genes. Furthermore, the genetic relationship among the isolated strains was investigated using RAPD fingerprinting. The bacteriological examination revealed that, out of 200 Gram-negative non-fermentative isolates, 44 (22%) were identified phenotypically and biochemically as Acinetobacter spp. and 23 (11.5%) were molecularly confirmed as A. baumannii. The retrieved A. baumannii strains were isolated from urine (69%), sputum (22%), and cerebrospinal fluid (csf) (9%). The isolated A. baumannii strains exhibited multidrug resistance and the production rates of carbapenemases were 56.5, 60.9, and 78.3% with meropenem, imipenem, and ertapenem disks, respectively. The blaOXA-24-like genes were the most predominant among the tested strains (65.2%), followed by blaOXA-23 (30.4%) and blaOXA-58 (17.4%), in addition, the examined strains are harbored IMP, VIM, and NDM genes with prevalence of 60.9, 43.5, and 13%, respectively, while KPC and GES genes were not detected. RAPD-PCR revealed that the examined strains are clustered into 11 different genotypes at ≥90% similarity. Briefly, to the best of our knowledge, this study is the first report concerning community-associated A. baumannii infections in Egypt. The high prevalence of hidden multidrug-resistant (MDR) and extensively drug-resistant (XDR) A. baumannii strains associated with non-hospitalized patients raises an alarm for healthcare authorities to set strict standards to control the spread of such pathogens with high rates of morbidity and mortality.


Acinetobacter calcoaceticus
Male Female Total

Burkholderia cepacia
Male Female Total

Antimicrobial Susceptibility Profiles of A. baumannii
Antimicrobial susceptibility profile was determined using 20 antibiotics, the susceptibility profiles were different in each isolate ( Figure 1). The antimicrobial susceptibility profiles of A. baumannii were represented in Table 3, with the difference between S, R, and I assessed statistically with different antibiotics at 0.05 level. All the examined A. baumannii strains were resistant to a minimum of one antimicrobial agent in three or more antimicrobial categories (multidrug-resistant (MDR)), of which 19 (82.6%) were resistant to a minimum of one antimicrobial agent in all but two or less antimicrobial categories (extensively drug-resistant (XDR)) ( Figure 1). As shown in Table 3, the strains, collectively, were resistant to oxacillin, nalidixic acid, and amoxicillin/clavulanic acid. The resistance rates to quinolones-ciprofloxacin, ofloxacin, levofloxacin, and norfloxacin were 69.6, 60.9, 65.2, and 69.6%, respectively. Resistance rates to aminoglycosides-tobramycin, gentamycin, and amikacin were 65.2, 73.9, and 52.2%, respectively. Cephalosporins resistance rates were 87, 91.3, and 56.5% for ceftazidime, cefotaxime, and cefepime, respectively. Carbapenem resistance showed 52.2, 47.8, 87, and 73.9% against meropenem, imipenem, ertapenem, and doripenem, respectively. The polypeptides polymyxin B and colistin were the most active antimicrobials agents against the studied A. baumannii with a sensitivity of 82.6% and 73.9%, respectively (Table 3).

Prevalence of Carbapenemase Producers
Modified Hodge Test (MHT) was carried out to detect the carbapenemases production in A. baumannii strains (n = 23), the results revealed that the production rates were 56.5, 60.9, and 78.3% with meropenem, imipenem, and ertapenem disks, respectively ( Figure 1). Collectively, all the tested strains were MHT positive to a minimum of one carbapenem species used in the test.

Random Amplification of Polymorphic DNA (RAPD) PCR Fingerprinting
Ten RAPD primers (Table 4) were tested for all the 23 strains, five were chosen according to their reproducible and discriminating powers. The RAPD typing of the 23 community-acquired isolates has been performed by using all five RAPD primers (P1, P2, P3, P4, and P5) ( Table 4). For each strain, five RAPD gels were combined into one single lane (Figure 3a). Each RAPD PCR generated up to 8 fragments per strain. The cumulative patterns yielded approximately 35 bands with fragments extended from 500 to 2000 bp per strain that were used for cluster analysis. Construction of the dendrogram was performed using GelJ cluster analysis software, the Dice coefficient and Unweighted Pair Group Method with Arithmetic Mean (UPGMA) method were used for analysis at ≥90 percentages similarity. In Figure 3b, RAPD discriminated 11 genetically distinct groups for the 23-community acquired A. baumannii. However, antibiotic resistance profiles were different between the isolates within the same group (Figure 1).  Percentages of similarity and primers used as presented in Table 3 on the top line of the dendrogram. RAPD type and strain code as in Table 3 and strain numbers as presented in Figure 3a appear on the right (between parenthesis).

Discussion
The rise and proliferation of antibiotic-resistant pathogenic microorganisms are of incredible worry to clinicians. Since the original depiction of a penicillin-inactivating catalyst in Escherichia coli, the furious 75-year battle against these microbes has been alluded to as an "unwinnable war" [31].  Table 3 on the top line of the dendrogram. RAPD type and strain code as in Table 3 and strain numbers as presented in Figure 3a appear on the right (between parenthesis).

Discussion
The rise and proliferation of antibiotic-resistant pathogenic microorganisms are of incredible worry to clinicians. Since the original depiction of a penicillin-inactivating catalyst in Escherichia coli, the furious 75-year battle against these microbes has been alluded to as an "unwinnable war" [31]. Recent clinical consideration has concentrated on the expanding rate of lactose non-fermenting, Gram-negative pathogens liable for health settings-gained infections [31]. In this gathering, Acinetobacter species are rising as pathogens that now and again cause infections in patients in intensive care units [32]. In this kind of bacteria, protection from different classes of anti-infection agents genuinely bargains with the capacity to treat patients who are tainted with these pathogens. Hence, for the immunocompromised host, current efficient therapy is a matter of survival.
In the existing study, 44 (22%) isolates were preliminarily identified as Acinetobacter spp., using differential growth morphology and biochemical tests, from 200 Gram-negative, non-fermentative strains collected between October 2017 and August 2018 with several types of infections from different private and governmental laboratories through Egypt. Twenty-nine of the 44 (65.9%) strains were identified biochemically as A. baumannii by the API 20NE system. It is widely accepted that identification of non-fermenting, Gram-negative bacilli is usually carried out by commercial identification systems such as the API 20 NE [10,11].
Acinetobacter spp. characterization to the genospecies level using phenotypic techniques is difficult, especially identifying strains belonging to A. baumannii [33]. In this work, PCR detection of A. baumannii specific OXA-51 was used to confirm the identification of A. baumannii, 23 isolates showed the presence of blaOXA-51-like genes and was confirmed by 16s-23s rRNA region sequencing. Turton et al. (2006) provide evidence that A. baumannii can be identified by the detection of blaOXA-51 by PCR [26]. However, a plasmid-mediated blaOXA-51-like carbapenemase has been found in other genomic species rather than A. baumannii, and the accuracy of using blaOXA-51 detection as a tool for differentiating A. baumannii from other Acinetobacter species has been argued [34]. The sequencing of bacterial 16S-23S rRNA intragenic spacer region has proved useful in taxonomic studies to distinguish between different species of common genus [35].
In the current work, A. baumannii was isolated from urine (69.6%), followed by sputum (21.7%) and csf (8.7%). In studies on community-acquired A. baumannii in China, Taiwan, and Australia, 63.8% of cases were pneumonia and 36.2% cases were bacteremia [39]. Falagas [39]. In our results, the highest percentage of isolation was from female patients (91.3%), followed by male patients (8.7%). Contrary to our results, Rebic et al. (2018) observed that A. baumannii infection was higher in male patients (58%) than female patients (42%) in a study that included 622 in-and out-patients in a Bosnian hospital [40]. These differences may be due to the randomized sampling pattern for our collected isolates without the concentration on specific symptoms or cases. Interestingly, Pires et al. (2020) found that female mice were more susceptible to A. baumannii pneumonia [41].
This study highlighted the importance of Acinetobacter spp. as an important community-acquired pathogen in contrast to the study done by Basuda K. and Medhabi S. (2013), which was carried out in a hospital setting and found that the predominant Acinetobacter spp. were A. calcoaceticus (42%) and A. baumannii (34%), and that both developed resistance to commonly-used antibiotics [14]. Recently, an exponential elevated rate of community-acquired A. baumannii cases has been reported that led to increased awareness of the danger of this type of infection with high pathogenicity and death rates, especially in equatorial and sub-equatorial geographical regions [39]. However, Acinetobacter spp. true incidence among Gram-negative, non-fermenting pathogens circulating the community has not been reported, to the best of our knowledge. Importantly, all of Acinetobacter spp. identified in this work were unidentified by the collecting lab, failure, and/or misidentification of Acinetobacter spp. with other gram-negative bacteria is common [12,13,15].
The small size and distribution of A. baumannii outer membrane porins are responsible for the decreased membrane permeability and may explain their effective, de facto higher resistance to certain antimicrobials [42]. Resistance to aminoglycosides like amikacin was 52.2% within the studied strains, which is close to other studies on A. baumannii from Egypt with 68.5% resistance [43]. A. baumannii cephalosporins resistance was similar to other studies in Egypt, Saudi Arabia, and India, where it ranges from 60 to 100% [44][45][46].
The resistance of the strains in this work to quinolones, nalidixic acid, and ciprofloxacin was close to studies from Algeria, Egypt, and Saudi Arabia [44][45][46]. A. baumannii strains' resistance to carbapenems like meropenem and imipenem (65.2 and 56.5%, respectively) were comparable to other studies in Egypt (57.5 and 65%) and Saudi Arabia (50 and 48.1%) [44,45]. The incidence of carbapenem resistance is becoming a serious problem in A. baumannii worldwide. A plethora of reports on carbapenem resistant A. baumannii exists, which have been associated with carbapenemases production [47][48][49].
In this work, carbapenemases production, as measured by MHT, demonstrated that 56.9, 60.9, and 78.3% of the strains produce carbapenemase inhibiting either meropenem, imipenem, or ertapenem, respectively. Therefore, all Acinetobacter isolates were phenotypically positive carbapenem producers for at least one type of carbapenem disk in the MHT test. In another study from Egypt, 60% of A. baumannii were positive for carbapenemases production by MHT assay using meropenem disks [49]. Recently, the accuracy of MHT was urged, suggesting that carbapenem resistance of an isolate is more likely correlated to the type of carbapenem produced rather than the level of production [50][51][52].
The persistence of OXA-51 gene family are found in all tested A. baumannii strains and believed to be intrinsic to this genomic species, but recently this was detected in another species perhaps due to horizontal gene transfer [25,26]. A. baumannii can easily develop resistance to a new lactam ring containing antibiotics due to the presence of this variety of OXA enzymes. In the current study, among the 23 carbapenem-resistant strains, 30.4% strains were detected as positive for the presence of blaOXA-23-like genes, 65.2% for blaOXA-24-like genes, and 17.4% for blaOXA-58-like genes. This differed from the previously reported prevalence of OXA type carbapenemases in 40 carbapenem-insensitive A. baumannii isolates from hospitalized patients collected from Egypt in 2014, which were 50, 7.5, and 5%, respectively [45]. This difference in OXA carbapenemases prevalence may indicate different genetic lineage between community-acquired and hospital-acquired A. baumannii in Egypt [45]. In Taiwan, the prevalence of blaOXA-23-, blaOXA24-, and blaOXA-58-like genes was 28.69, 90.91, and 0%, respectively [53]. The prevalence of other types of carbapenemases, such as blaNDM, blaVIM, blaIMP was similar to other studies in Egypt, where it was found that the incidence of blaNDM was 39.3%, blaIMP 95.7%, and blaVIM 93% [54][55][56].
All profiles recognized in this investigation were ordered into 11 groups that were seen in 23 strains by utilizing RAPD-PCR demonstrating high genotypic varieties. Various examinations have been coordinated on genotyping A. baumannii strain by different genotyping techniques, and various genetic profiles have been represented in various regions. The findings of this study are similar to that of studies reported by Asadian et al. (2019), who were also able to cluster A. baumannii strains based on their genetic relatedness using RAPD-PCR [57]. The assorted variety of genetic patterns among the isolates is probably due to a wide assortment of sampling locations, otherwise several strains with identical genetic pattern and common origin would have been observed. It is also noticeable that strains originated from different infection sites were grouped genetically in the same cluster, suggesting the same origin and increase of the danger of multiple infection capabilities for such genotypes. Distinguishing proof of a satisfactory degree of decent genetic variety among the strains by this strategy utilizing five consolidated RAPD oligos demonstrates that this technique is valuable for examining and typing A. baumannii, and strains with various roots can be arranged into various clusters. This agreed with Koeleman et al. (1998), who proposed using at least 5 RAPD primers to better discriminate genetic diversity between bacterial isolates [58].

Ethical Approval
The research was ethically approved by the Faculty of Science-Suez Canal University (2017 MS518) and is in compliance with the Helsinki Declaration of 1964 and the resulting revisions or comparative moral guidelines. This research was waived from written informed consent/assent/parental permission, and the patients were informed orally.

Study Design and Bacterial Isolates
A sum of 200 non-repetitive, Gram-negative, non-fermentative isolates was selected from 2108 isolates from non-hospitalized (for the last 12 months) symptomatic patients from private medical and microbiological analysis laboratories across Egypt during a period of 11 months from September 2017 to August 2018. These bacteria were originally isolated from sputum, urine, blood, cerebrospinal fluid. Isolates (n = 200) were streaked on MacConkey agar (OXOID, Basingstoke, England), Leed Acinetobacter agar (Hardy Diagnostics, Santa Maria, CA, USA), and CHROMagar (Oxoid, Basingstoke, England), and incubated under aerobic conditions at 37 • C for 48 hours. Gram staining, motility, oxidase, peroxidase, and oxidative-fermentation reactions were performed according to standard techniques [59]. API 20NE bacterial identification system (Biomerieux, Marcy-l'Étoile, France) was used for species identification.

RAPD-Genotyping of A. baumannii
Ten decanucleotide primers obtained from RAPD kit (Eurofins Biodiagnostics) and used in this study (Table 3). RAPD analysis was accomplished in accordance with Zhang et al. (2008) [66]. For cluster analysis, a DNA fingerprint java application analysis software (GelJ) was used (downloaded from https://sourceforge.net/projects/gelj/) [67]. UPGMA method was applied in cluster algorism and dendrogram creation using Dice as a similarity index. A similarity index of ≥90% was used for clustering genetically-related isolates.

Statistical Analysis
Data were checked for normality using Shapiro-Wilk normality testing at p < 0.05. Parametric data were presented as mean followed by standard deviation. Non-parametric data were presented as number and percentages. Differences in parametric data were checked using one-way analysis of variance (ANOVA) at p < 0.05. Non-parametric data were checked using Chi squared, and Kruskal-Wallis test statistic to differentiate between (sensitive, intermediate, resistant). Dendrogram were plotted by hierarchical clustering analysis of isolates and factors, both binary factor (carbapenemase genes) and ordinal factor (antibiotics) using Ward's method and squared Euclidean distances. Statistical analyses were carried by IBM SPSS version 23.0 for Mac OS. (Table 4).

Conclusions
In this work, we identified multidrug-resistant strains of A. baumannii from community-acquired infections in Egypt. The isolated A. baumannii strains exhibited multidrug resistance with the production of carbapenemases. The blaOXA-24-like genes were the most predominant among the tested strains, followed by blaOXA-23 and blaOXA-58. In addition, the examined strains harbored IMP, VIM, and NDM genes. Genetic relatedness studies revealed 11 different genotypes with more than 90% similarity. The current study provided phenotypic and genotypic characterization data of community-acquired A. baumanii. Further studies are needed to assess other resistance genes, the potential virulence factors of these community-acquired strains, and the severity of the clinical outcome associated with these strains.
The findings of this research essentially encourage continual precise monitoring and reporting of the incidences and antimicrobial susceptibility of community-associated Acinetobacter spp. to help clinicians in the containment of these life-threatening infections. Funding: This research received no external funding.

Conflicts of Interest:
The authors declare no conflict of interest.