Evaluation of phenotypic and genotypic patterns of aminoglycoside resistance in the Gram-negative bacteria isolates collected from pediatric and general hospitals

The purpose of the current study was to evaluate the phenotypic and genotypic patterns of aminoglycoside resistance among the Gram-negative bacteria (GNB) isolates collected from pediatric and general hospitals in Iran. A total of 836 clinical isolates of GNB were collected from pediatric and general hospitals from January 2018 to the end of December 2019. The identification of bacterial isolates was performed by conventional biochemical tests. Susceptibility to aminoglycosides was evaluated by the disk diffusion method (DDM). The frequency of genes encoding aminoglycoside-modifying enzymes (AMEs) was screened by the PCR method via specific primers. Among all pediatric and general hospitals, the predominant GNB isolates were Acinetobacter spp. (n = 327) and Escherichia coli (n = 144). However, E. coli (n = 20/144; 13.9%) had the highest frequency in clinical samples collected from pediatrics. The DDM results showed that 64.3% of all GNB were resistant to all of the tested aminoglycoside agents. Acinetobacter spp. and Klebsiella pneumoniae with 93.6%, Pseudomonas aeruginosa with 93.4%, and Enterobacter spp. with 86.5% exhibited very high levels of resistance to gentamicin. Amikacin was the most effective antibiotic against E. coli isolates. In total, the results showed that the aac (6')-Ib gene with 59% had the highest frequency among genes encoding AMEs in GNB. The frequency of the surveyed aminoglycoside-modifying enzyme genes among all GNB was found as follows: aph (3')-VIe (48.7%), aadA15 (38.6%), aph (3')-Ia (31.3%), aph (3')-II (14.4%), and aph (6) (2.6%). The obtained data demonstrated that the phenotypic and genotypic aminoglycoside resistance among GNB was quite high and it is possible that the resistance genes may frequently spread among clinical isolates of GNB. Supplementary Information The online version contains supplementary material available at 10.1186/s40348-022-00134-2.


Introduction
Antimicrobial resistance is being increasingly recognized as a serious public health threat worldwide [1][2][3][4]. Antimicrobial resistance is highly noticeable among Gram-negative bacteria (GNB), and therefore, clinical isolates resistant to multiple or almost all available antibiotics have been consistently emerging [5,6]. The three broad-spectrum classes of antimicrobial agents including β-lactams (especially β-lactam antibiotics and β-lactamase inhibitor combinations, cephalosporins, and carbapenems), aminoglycosides, and fluoroquinolones are the most common classes of antibiotics used to treat different infections caused by GNB [1,7,8]. Aminoglycosides as broad-spectrum antibiotics have an important role under clinical settings and are used for the treatment of severe and life-threatening hospital-acquired Open Access Molecular  infections caused by GNB [9]. The aminoglycosides including tobramycin, gentamicin, and amikacin play a bactericidal role against a wide range of GNB such as Acinetobacter baumannii (A. baumannii), Pseudomonas aeruginosa (P. aeruginosa), Escherichia coli (E. coli), and Klebsiella pneumoniae (K. pneumoniae) [10,11]. However, in recent years, resistance to aminoglycosides, especially its association with other antibiotic classes such as β-lactams and fluoroquinolones, has been increasingly reported. Resistance to aminoglycoside antibiotics is present virtually all over the world, particularly in Asia and Latin America [12]. The resistance mechanisms against aminoglycosides in GNB mainly result from the (1) production of aminoglycoside-modifying enzymes (AMEs), inactivating antibiotics classified in several families such as aminoglycoside nucleotidyltransferases (ANTs), aminoglycoside acetyltransferases (AACs), and aminoglycoside phosphoryltransferases (APHs); (2) methylation of 16S rRNA by a family of ribosomal methyltransferase enzymes; (3) mutation in the 30S ribosomal subunit; (4) active expulsion of antibiotics from the bacterial cells by efflux pumps; and (5) alteration of cell membrane permeability and reduction in intracellular concentration of aminoglycosides [13][14][15]. Among these factors, AMEs represent the most common mechanism by which clinical isolates of GNB and Gram-positive bacteria (GPB) can enzymatically modify the hydroxyl or amino groups of the drug, inhibiting their binding to ribosomes and hence allowing the bacteria to survive [16,17]. According to the several condition such as partial immune system, neonates and children are a susceptible group to bacterial infections. Neonates can acquire bacteria from families or mothers within horizontally and vertically transmission ways, respectively. Antibiotic-resistant GNB can cause severe infections in neonates and children and are considered the main concern for physicians [18]. In Iran, although some authors have reported a high prevalence rate of aminoglycoside resistance among the GNB isolated from clinical samples [19][20][21], the overall prevalence of aminoglycoside-resistant genes among clinical GNB isolates has not been determined widely. Therefore, the present study follows several objectives: (1) evaluation of the phenotypic resistance patterns of GNB; (2) determination of the frequency of common aminoglycoside-resistance genes including genes encoding AMEs; and (3) characterization of the correlation between aminoglycoside-resistance genes and the phenotypic resistance.

Samples and bacterial isolation
The present study investigated the clinical isolates of GNB that were collected from pediatric and general hospitals throughout Iran during January 2018 to the end of December 2019 (Supplementary file).
Bacterial identification was carried out using conventional biochemical tests including growth on MacConkey (MAC) agar (Merck, Darmstadt, Germany), chocolate agar, and blood agar plates; evaluation of colony morphologies and Gram stain characteristics; oxidase test; catalase test; citrate utilization test; nitrate reduction test; indole production; motility; lactose fermentation; H 2 S production; urease activity; Methyl Red (MR) test; and Voges Proskauer (VP) test. After identification, all bacterial strains were preserved in Tryptic Soy Broth (Merck, Darmstadt, Germany) containing 20% glycerol at -70 °C for further analysis.

Antibiotic susceptibility testing
The susceptibility and resistance patterns of GNB to aminoglycoside agents were determined by the disk diffusion method (DDM) on Mueller Hinton agar medium (Merck, Darmstadt, Germany). The tested antimicrobial agents included amikacin (30 μg/disk), gentamicin (10 μg/disk), and tobramycin (30 μg/disk). The isolates resistant to at least one of these aminoglycoside agents were considered as aminoglycoside-resistant strain. The results of the DDM method were interpreted based on Clinical and Laboratory Standards Institute (CLSI) criteria. E. coli ATCC 25922 and P. aeruginosa ATCC 27853 were used as quality control strains in every test run.

Molecular detection of aminoglycoside-resistant genes
Total genomic DNA was extracted from GNB isolates grown in Mueller-Hinton broth (Merck, Darmstadt, Germany) overnight at 37 °C. The bacterial cell pellets were resuspended in 250 μl of phosphate-buffered saline (PBS), and then, DNA extraction was performed using High Pure PCR Template Preparation Kit (Roche Diagnostics, Mannheim, Germany) in line with the manufacturer's instructions. The purity and concentration of the extracted DNA were evaluated by Nanodrop (DeNovix Inc., USA). Extracted DNA was stored at -80 °C for further analysis. The frequency of six main genes encoding AMEs including aac (6')-Ib, aph (3')-VIe, aph (3')-II, aadA15, aph (3')-Ia, and aph (6) was screened by polymerase chain reaction (PCR) method via specific primers. The sequence of the primers used for performing PCR is shown in Supplementary

Statistical analysis
All data regarding the prevalence of isolated bacteria, resistance profile to aminoglycoside agents, and frequency of genes encoding AMEs were added to the statistical package SPSS v.23.0 (SPSS Inc., Chicago, IL, USA) and analyzed using descriptive statistical tests. Among GNB, Acinetobacter spp. (n = 186; 56.9%) and P. aeruginosa (n = 55; 40.4%) were isolated frequently from tracheal samples. On the other hand, most isolates of E. coli (n = 119/144; 82.6%) and K. pneumoniae (n = 65/140; 46.4%) were isolated from urine samples. Moreover, Enterobacter spp. (n = 68/89; 76.4%) were isolated frequently from blood samples. The distribution of the isolated bacteria in clinical samples is shown in Supplementary Table 2.

The frequency of GNB by age groups
The frequency of the GNB isolated from pediatric and general hospitals among different age groups is shown in Supplementary Table 3. In general, GNB had the highest and lowest frequency rates among patients in > 50-year and 5-to 14-year age groups, respectively. P. aeruginosa (59%), Acinetobacter spp. (53.7%), and E. coli (43.7%) showed the highest frequency in the age groups > 50 years. The results showed that Acinetobacter spp., P. aeruginosa, and E. coli isolates exhibited the lowest frequency in < 1-year, 1-to 4-year, and 5-to 14-year age groups. On the other hand, K. pneumoniae (27.1%) and Enterobacter spp. (25.8%) revealed the highest frequency in < 1-year age groups. Our analyses revealed that the frequency of K. pneumoniae (6.4%) among patients in the 5-to 14-year age group was low. Moreover, Enterobacter spp. had the lowest frequency in patients belonging to 15-to 29-year age groups.

Frequency of aminoglycoside-resistant genes among GNB
The current study evaluates the distribution of AME genes among phenotypically aminoglycoside-resistant GNB. The distribution of aminoglycoside-resistant genes among GNB is shown in

Discussion
In recent years, the incidence of both phenotypic and genotypic aminoglycoside resistance has been surging around the world [22,23]. The corresponding high resistance rate can severely complicate combination therapy for aminoglycoside agents with broad-spectrum β-lactams including cephalosporins and carbapenems against severe Gram-negative infections, particularly in case of nosocomial infections [24,25]. Since aminoglycoside agents are the first choice of clinicians for infection treatments [22] and given that aminoglycosides are the most commonly prescribed antimicrobial agents for treating serious GNB in Iran, an attempt is made here to characterize the aminoglycoside resistance by means of phenotypic and genotypic methods among five important GNB isolated from pediatric and general hospitals in Iran. The results of antimicrobial susceptibility screening revealed that about half of the GNB were fully resistant to at least one of the tested aminoglycosides, including gentamicin, tobramycin, and amikacin. Previously, resistance to gentamicin, amikacin, and tobramycin, which are considered as newer aminoglycosides with a broader spectrum of antibacterial activities, was generally found to be less common than resistance to older aminoglycosides such as streptomycin, kanamycin, and neomycin [26]. In the current study, approximately, more than 90% of Acinetobacter spp. exhibited resistance to all the tested aminoglycosides. In a previously published study by Aliakbarzade et al., 103 clinical A. baumannii strains were collected from Imam Reza Hospital of Tabriz University of Medical Sciences. They showed that A. baumannii strains were isolated from different clinical samples such as urine, sputum, tracheal secretion, bronchial lavage, wound, and blood. The findings of their study revealed that the rates of resistance to gentamicin, amikacin, and tobramycin were 86%, 81%, and 63%, respectively [27].
Compared to the above studies, we witnessed a significant increase in the number of aminoglycoside-resistant Acinetobacter isolates. In this study, the rate of resistance to gentamicin, amikacin, and tobramycin was 93.6%, 90.2%, and 90.5%, respectively. The high-level resistance to aminoglycoside is a serious issue for combination therapy of aminoglycoside with broad-spectrum β-lactams including carbapenems and cephalosporins against Acinetobacter infections [19]. According to our findings, in comparison to other tested aminoglycoside agents, amikacin caused less resistance in GNB, especially in E. coli isolates, and is still an extremely useful drug in treating Table 3 Aminoglycoside resistance gene profiles of the GNB in Iran  severe E. coli infections. Several studies have pointed out that increased resistance against amikacin could be associated with the unrestricted use of this compound by some clinicians [28,29]. In 2018, Nasiri et al. surveyed the molecular epidemiology of aminoglycoside resistance in clinical isolates of K. pneumoniae. They collected 177 K. pneumoniae strains from the patients admitted to intensive care units (ICUs) as well as infectious diseases, internal medicine, and surgery wards. K. pneumoniae strains were isolated from different clinical specimens such as urine, wound, sputum, trachea, and blood. The above authors reported that amikacin was a more active antimicrobial agent than other aminoglycosides toward clinical isolates of K. pneumoniae with a resistance rate of 61% [20]. Nevertheless, our findings show that the rate of resistance to amikacin in fact increased to 93.6% among K. pneumoniae isolates, compared to the mentioned studies. Overall, the high frequency of aminoglycoside-resistant GNB suggests that extensive use of these antimicrobial agents in clinical settings of Iran has led to the emergence of resistant isolates.
This study surveyed the prevalence of six main aminoglycoside-resistant genes in GNB. Results showed that a majority of aminoglycoside-resistant GNB (about threequarters of isolates) harbored at least one AME gene. In total, the aac (6')-Ib (59%) and aph (3')-VIe (48.7%) genes were the most prevalent AME genes among all the aminoglycoside-resistant GNB. Related reports from different parts of the world have illustrated that aac (6´)-II and ant (2´´)-I genes are the most prevalent AME genes in Europe. Moreover, it has been revealed that aph (3′)-VIe, ant (2´´)-I, and aac (6´)-I genes have the highest frequency among AME genes in Korea [11,30]. On the other hand, aac (6´)-31/aadA15 and aadA2 genes were also detected frequently in the GNB isolated from nosocomial infections in Mexico and Brazil [31,32]. AMEs are the most important sources of aminoglycoside resistance among bacteria. The corresponding genes are highly mobile and can be transported by integrons, transposons, plasmids, and other transposable gene elements, often along with other resistant genes (such β-lactamases genes). As a matter of fact, the most threatening GNB acquire AME genes through horizontal gene transfer [33,34].
In Iran, a high prevalence rate of AMEs was previously reported in GNBs such as P. aeruginosa, A. baumannii, and K. pneumoniae [20,37,38]. The aph (3')-VIe was the most common AME in Acinetobacter isolates (52.6%), followed by aadA15 (37.9%), aph (3')-Ia (35.5%), and aac (6')-Ib (34.3%). In a study conducted by Aghazadeh et al. in Iran, aph(3')-VIe and aph(3')-II were the most prevalent AME genes in A. baumannii with prevalence rates of 90.6% and 61.8%, respectively [39]. In another study in Iran, Asadollahi et al. reported that AME genes including aadA12, aacC1, and aadB were the most prevalent ones among A. baumannii [40]. Altogether, these data indicate that aph (3')-VIe and aadA15 genes contribute to aminoglycoside resistance among clinical isolates of A. baumannii in different geographic locations of Iran. Our results were similar to those found by Nasiri et al. in Iran who reported aac(6′) as the most dominant AME among clinical isolates of K. pneumoniae [20]. However, Ghotaslou et al. previously reported that ant(3″)-Ia and aph(3″)-Ib were the most prevalent AME genes in Enterobacteriaceae isolates in the northwest of Iran with frequency rates of 35.9% and 30.5%, respectively [14]. In another research by Soleimani et al., aac (3)-IIa and ant(2″)-Ia genes were identified as the most common AMEs in uropathogenic E. coli isolated from an Iranian hospital [41]. These data suggested that various reasons such as diversity of specimen type, geographic regions, sample size, bacterial sources, usage of antibiotics, and applied detecting methods would affect the distribution patterns of AME. Liang et al. previously reported that aac (3)-II, aac (6′)-Ib, and ant (3″)-I genes were the most common AME genes in K. pneumoniae isolates in China [42]. In Norway, Lindemann et al. indicated that the majority of E. coli and K. pneumoniae isolates in their study harbored aac(3)-IIa and aac(6′)-Ib genes [42]. The significant variation of the results may be attributed to geographical factors. Regarding P. aeruginosa isolates, we found aac (6')-Ib and aph (3')-II as the most prevalent AME genes. These findings are consistent with those reported by Aghazadeh et al. in Tabriz, Iran [39]. In general, results demonstrated that more than 90% of GNB were resistant to one of the antibiotics. However, the results of the molecular method revealed that 59% of GNB harbored the aac (6')-Ib gene. Our analyses revealed that 78% of GNB were positive for at least one of the examined genes.