Integron-Mediated Antibiotic Resistance in Acinetobacter baumannii Isolated from Intensive Care Unit Patients, Babol, North of Iran

Background We investigated the integron types and their relation with antibiotic resistance among A. baumannii isolates collected from intensive care unit patients, Babol, north of Iran. Methods In this cross-sectional study, a total of 73 bronchoalveolar lavage samples were obtained from patients in ICU. Susceptibility testing was performed by disk diffusion method. Types of integrons were identified by an integrase gene PCR. Results In total, 47.9% A. baumannii isolates were recovered from the BAL samples. All isolates were resistant to ceftazidime. 91.4% and 58.3% of isolates were MDR and XDR, respectively. The rate of colistin resistance with the E-test was 5.7%. Molecular analysis of class I, II, and III integrons showed that 25.7%, 88.6%, and 28.6% of the isolates carried the intI, intII, and intIII genes, respectively. Discussion Our results show that different classes of integrons are commonly spread among A. baumannii strains and these genomic segments can play an important role in the acquisition of MDR and XDR phenotypes. So monitoring drug resistance in A. baumannii isolates with the use of int gene PCR is very important to plan specific infection control measures to prevent the spread of MDR-AB and XDR-AB in Iran's hospitals.


Introduction
Multidrug-resistant Acinetobacter baumannii (MDR-AB) is an important opportunistic pathogen responsible for severe hospital-acquired infections (HAIs), particularly patients admitted to the intensive care unit (ICU) [1]. MDR-AB infections especially pneumonia and bacteremia show a high mortality rate (30% to 75%) and require prolonged hospital stays in intensive care units (ICUs) [2]. In the ICU, the mortality rate associated with A. baumannii is 54%. The majority of infections are originated from epidemic outbreaks. Appearance of extensively drug-resistant A. baumannii (XDR-AB) limits the therapeutic options and causes a serious concern to HAI control. For a period of 5 years (2005-2010), the frequency of XDR-AB in clinical isolates was increased from 15% to more than 41%. Control of MDR-AB and XDR-AB infections is an important issue for clinical microbiologists and physicians. Treatment of A. baumannii has become difficult because many isolates are now resistant to a wide range of antimicrobial agents [3,4].
An important factor that influences the development of multiresistance phenotype is the acquisition of mobile genetic elements (MGEs). Antibiotic-resistant elements in Acinetobacter strains are commonly carried on MGEs, such as the R plasmids, transposons (TEs), integrons (Int), and genomic islands (GEIs) [5]. Integrons are conserved sequences (3 -CS and 5 -CS) of DNA that are able to acquire gene cassettes, which can carry antibiotic resistance genes, by site-specific recombination [6]. These genetic segments are considered by the presence of an intI gene (integrase), a recombination site (attI), and a promoter (P C ) [7]. The most common types of integrons are the transportable class I (Tn402 derivatives) integron, followed by class II and class III integrons, respectively. Class I integrons harbor numerous antimicrobial resistance gene cassettes encoding broad-spectrumlactamase, dfr (dihydroflavonol-4-reductase/trimethoprim), qacEΔ1 (disinfectants and tetravalent ammonium compounds), sul1 (sulfonamide), and aminoglycoside-modifying enzymes (AMEs). Integrons of class II are located in Tn7 and 3 -CS which contains five tns genes and are responsible for the mobility of TE. Integrons of class III located in transposons have also been described, but the 3 -CS is still not well defined [8].
MDR phenotype occurs in A. baumannii when different integrin-borne antibiotic resistance elements coexist, giving rise to MDR gene cassettes [9,10]. Due to the high prevalence of this infection, as well as various profiles of drug resistance in different geographical areas, a study on the prevalence and antibiotic resistance pattern in different part of the world is essential. These data would provide useful information on the spread of resistance elements and the possibility of choosing appropriate treatment strategies. So, the aim of this study was to determine the frequency of type I, II, and III integrons in A. baumannii isolates recovered from bronchoalveolar lavage (BAL) samples obtained from hospitalized patients in ICU ward, Babol, north of Iran [11].

Chest Radiograph (Chest X-Ray).
A portable chest radiograph was obtained prior to, and on the same day as, bronchoscopic sampling. Infiltration of lung (chest X-ray) was included as a criterion for hospital admission. The radiograph was reviewed and interpreted by a staff radiologist and pulmonary disease specialist [12].

Procedure and Specimen Collection (BAL Collection).
Patients were commonly sedated with IV midazolam before bronchoscopic procedure. The fraction of inspired oxygen was increased to 100%, and positive end-expiratory pressure was limited to 8 cm H 2 O. The bronchoscope used was the Pentax FB15X (Pentax Instruments, Tokyo, Japan). Bronchoscopic BAL samples were collected by pulmonologist by wedging the tip of a fiber optic bronchoscope in the subsegmental bronchus of the most compromised lobe seen in chest X-ray or, in cases of diffuse radiologic presentation, in the posterior bronchus of the lower lobe. As little topical lidocaine as possible was used so as not to interfere with bacterial growth (never >20 mg per bronchus). Aspiration of secretions by the bronchoscope was avoided. This procedure was repeated in the contralateral lung, and samples were preserved at ambient temperatures or 4 ∘ C before being transported to the microbiology laboratory within 2 h of collection [12,13].

Quantitative Cultures.
Quantitative cultures were processed according to the standard laboratory protocol. Sample pairs collected were diluted in tryptic soy broth (Merck, Co., Germany) plated at final dilutions of 10 −3 and 10 −4 onto eosin methylene blue agar, chocolate agar and blood agar (Merck, Co., Germany) plates and then incubated at 35 ∘ C under 5% CO 2 . Gram staining was performed on 1 drop of undiluted BAL fluid. Final colony counts were determined after 48 h. Potential pathogens present at ≥1 × 10 4 CFU/ml were considered clinically significant, and subsequent identification and antimicrobial susceptibility testing were performed. Isolates detected at ≥1 × 10 3 and <1 × 10 4 CFU/ml were presumptively identified, and antimicrobial susceptibility testing was not performed. The following were considered etiologic agents of pneumonia: (1) bacteria with >10 4 CFU/mL in BAL quantitative culture (according to clinical judgment, >10 3 CFU/mL was considered positive) and (2) [14]. In order to confirm the identity of strains, the presence of gyrB gene was assessed using PCR [15,16]. The isolates were preserved in −80 ∘ C in brain-heart infusion broth (Merck, Co., Germany) containing glycerol 50% v/v until the molecular analysis.  Forty-nine of 73 BALs (67.1%) presented >10 4 CFU/mL in BAL quantitative culture. The three BALs (4.8%) that presented between 10 3 and 10 4 CFU/mL were also considered positive. Total positivity was 52 of 73 (71.2%); 49 of these 52 positive (67.1%) had >10 4 CFU/mL. There were no cases of bacterial growth with <10 3 CFU/mL. Twenty-one cases (28.7%) did not present bacterial growth (Table 2).

Mean
A. baumannii isolates ( = 35; 47.9%) were recovered from the BAL samples ( = 73). The highest resistance rate was related to CAZ (100%) ( Table 3). In the current study, MDR-AB is defined as resistance to more than three classes of antibiotics. XDR-AB is regarded as the isolate that is resistant to 3 classes of antibiotics defined above (MDR) as well   (Figure 1). Molecular detection of class I, II, and III integrons was performed by amplification of intI, intII, and intIII genes in all A. baumannii strains using PCR method. The results revealed that 94.3% ( ; 33/35) of the isolates carried the int gene. Molecular analysis of class I, II, and III integrons showed that 25.7% ( ; 9), 88.6% ( ; 31), and 28.6% ( ; 10) of isolates carried the intI, intII, and intIII genes, respectively. The coexistence of intI/intII and intI/intII/intIII in isolates was in order 22.9% ( ; 8) and 8.6% ( ; 3). Two (5.7%) of all isolates were int-negative. The correlation of class I, II, and III integrons and antibiotic resistance profile are shown in Table 4. Nucleotide sequence data reported in the current study has been deposited in the PubMed/NCBI/GenBank nucleotide sequence databank under the accession numbers KX122025.1, KX122026.1, KX122027.1, KX122028.1, and KX122029.1.

Discussion
A. baumannii is an important opportunistic bacterium causing severe HAIs with high mortality rates due to its wide drug resistance. Integrons are vital in enabling the A. baumannii genome to capture and accumulate many antimicrobial resistance genes. Because of the MDR phenotype and the trend to spread in the hospital environment, A. baumannii has a special clinical significance requiring epidemiological monitoring as a measure to control the spread of HAIs [18].
In line with Gomes et al. [12], Zaccard et al. [13], and Nomanpour et al. [19] studies, considering 10 4 CFU/mL as the cutoff and including the cases with bacterial growth between 10 3 and 10 4 CFU/mL, 28 of all 52 positive BALs (53.8%) presented more than one isolate, which means polymicrobial infections. The most frequently encountered pathogens were associated with two different strains of A. baumannii (48%) and between P. aeruginosa (5.5%) and MRSA (5.5%), with three cases of each association.
Our findings showed that all these strains were resistant to CAZ. DD method results showed 91.4% and 58.3% of the isolates were MDR and XDR, respectively. These results are in agreement with the Ghajavand et al. study [20]. In a parallel study piloted in Taipei, Taiwan, 25 XDR-AB were found during two years of study [21]. This contrast may be due to the geographical distance and/or the different levels of hygiene. The majority of the strains were sensitive to CS, regarded as the last resort against MDR-AB, probably due to the fact that CS has not been broadly used in Iran; thus, the corresponding selection pressure has not been established on this bacterial species. -test MIC value showed that 5.7% ( ; 2) of isolates were resistant to CS. This data are similar to Vakili et al. [14]. All CS-resistant isolates were positive for class II integron. No significant association was observed between CS-resistant isolates and class II integron.
Integrase-encoding genes were found in 33 (94.3%) of the 35 isolates. Our data showed that the prevalence of class I, II, and III integrons was 25.7%, 88.6%, and 28.6%, respectively. In the strains that had all the three classes of integrons, resistance to all the antibiotics tested was observed. Two samples (5.71%) did not contain any integrons but were still resistant to CP and CAZ, which indicates that resistance genes could also be transmitted by other elements such as plasmids, TEs, and bacteriophages. In the study carried by Ramírez et al., the frequency of class I and II integrons was 42% and 68%, respectively [22]. Also, Kamalbeik et al. showed that 7.5% and 67.5% of the strains contained class I and II integrons, respectively [23]. In the studies carried out by Ramírez et al. [24] and Martins et al. [25], the prevalence of class II Integron was 23% and 41.7%, respectively. In the research conducted by Taherikalani [27] the frequency of class I, II, and III integrons was 85%, 14%, and 0%, and 97%, 31%, and 0%, respectively. These findings are in contrast with our results and may be related to the difference in source of samples, level of hygiene, and geographical areas. So, it might be due to stiff search-and-destroy and surveillance policies, as well as control in antibiotic prescription.   Integron-positive A. baumannii isolates showed higher antibiotic resistance rate compared to integron-negative strains. All the isolates with class II integron had a significant relationship with resistance to all the tested antibiotics which could be due to the gene cassettes that encoded resistance to these antibiotics. According to our results, there was a significant correlation between the presence of class I, II, and III integrons and resistance to CAZ. Lin et al. [28] and Mirnejad et al. [29] observed a significant correlation between the presence of integrons and resistance to CP, ofloxacin, cefepime, CAZ, aztreonam, AN, and norfloxacin. Koeleman et al. indicated a significant relationship between the presence of integrons and resistance to AN, CP, and CAZ [30]. The results of Gaur et al. confirmed the results of our study and showed correlation between the presence of integrons and resistance to AN, cefepime, and CP [31]. In cases where significant relationships between the integrons and antimicrobials resistance were not detected, resistance could be attained by resistance elements that reside on MGEs, such as plasmids, TEs, and phages, indicating substantial HGT from other bacteria [32].

Conclusion
In general, our results showed that different classes of integrons are commonly spread among A. baumannii strains and these genomic segments can play an important role in the acquisition of MDR and XDR A. baumannii. In this study, regardless of the presence or lack of resistance genes, strong association was observed between the presence of class II integrons and a decreased sensitivity to the many classes of antibiotics which could be challenging, since these structures can move among strains and consequently become resistant to new antimicrobials. So, monitoring drug resistance in A. baumannii isolates with the use of int gene PCR is very important to plan specific infection control measures to prevent the spread of MDR-AB and XDR-AB in Iran's hospitals.

Conflicts of Interest
No conflicts of interest were reported by the authors of this paper.