Emergence of High-level Gentamicin Resistance among Enterococci Clinical Isolates from Burn Patients in South-west of Iran: Vancomycin Still Working.

Enterococcus faecalis and Enterococcus faecium are among the main agents associated with nosocomial infections with high mortality in immunocompromised patients. Antibiotic resistance, especially against gentamicin and vancomycin among Enterococci , is a risk factor that could increase the morbidity and mortality rate. 179 Enterococci isolates from burn patients were included in this study. Antibiotic susceptibility testing was done using the disk diffusion test and minimum inhibitory concentration (MIC) was evaluated by agar microdilution. Vancomycin and gentamicin resistance associated genes including vanA , vanB , vanC , aac (6')-Ie aph(2''), aph(3')-IIIa and ant(4')-Ia were detected by PCR and their statistical relation with antibiotic resistance was evaluated. E. faecalis was the more prevalent strain among our local isolates and showed a higher antibiotic resistance in comparison to E. faecium . Vancomycin had a good antibacterial effect on the Enterococcus spp. isolates; however, resistance to this antibiotic and a high-level gentamicin resistance (HLGR) phenotype were observed. Among van operon genes, vanA was the most prevalent gene and among the gentamicin resistance genes, aph (3')-IIIa was more frequent. The HLGR Enterococci are a real challenge in nosocomial infections. Vancomycin is a key antibiotic to treat such infections but emergence of VRE in our region could be a real concern and, therefore, phenotypic and molecular surveillance must be considered. Enterococcus faecalis and Enterococcus faecium are among the main agents associated with nosocomial infections with high mortality in immunocompromised patients. Antibiotic resistance, especially against gentamicin and vancomycin among Enterococci, is a risk factor that could increase the morbidity and mortality rate. 179 Enterococci isolates from burn patients were included in this study. Antibiotic susceptibility testing was done using the disk diffusion test and minimum inhibitory concentration (MIC) was evaluated by agar microdilution. Vancomycin and gentamicin resistance associated genes including vanA, vanB, vanC, aac (6’)-Ie aph(2’’), aph(3’)-IIIa and ant(4’)-Ia were detected by PCR and their statistical relation with antibiotic resistance was evaluated. E. faecalis was the more prevalent strain among our local isolates and showed a higher antibiotic resistance in comparison to E. faecium. Vancomycin had a good antibacterial effect on the Enterococcus spp. isolates; however, resistance to this antibiotic and a high-level gentamicin resistance (HLGR) phenotype were observed. Among van operon genes, vanA was the most prevalent gene and among the gentamicin resistance genes, aph (3’)-IIIa was more frequent. The HLGR Enterococci are a real challenge in nosocomial infections. Vancomycin is a key antibiotic to treat such infections but emergence of VRE in our region could be a real concern and, therefore, phenotypic and molecular surveillance must be considered.

Introduction

Burn injury is a common cause of morbidity and mortality. In Iran, approximately 30 000 people with burns present to the emergency departments each year. Among these, 3 000 dies and others either have minor burn injuries that are treated primarily in the emergency department or sustain major burn injuries that require hospital admission. Both impose the health systems a burden of cost. These numbers are eight times larger than the world average and therefore are a source of concern (Karimi et al. 2015).

In most cases, the bacterial infection of burn wounds is an unquestionable phenomenon because of the skin destruction, which plays a role of the major barrier to bacterial access to the internal tissues. Gram-positive bacteria including Staphylococcus aureus, β-hemolytic Streptococci and Enterococci, and Gram-negative bacteria such as Pseudomonas aeruginosa, Acinetobacter baumannii and Enterobacteriaceae are among the most frequent etiological agents of burn patients infections (Norbury et al. 2016).

The facultative anaerobic Gram-positive Enterococcus spp. normally colonize the gastrointestinal tract, oral cavity, and vaginal tract. Enterococci are among the major agents associated with nosocomial infections particularly in burn patients presenting with bacteremia, urinary tract infections and endocarditis (Hashem et al. 2017). The US National Nosocomial Infection Surveillance (NNIS) system, has ranked Enterococci among the top three most common pathogens of nosocomial infections and the leading cause of nosocomial infections in burn patients (Pan 2012).

Enterococci are intrinsically resistant to multiple antibiotic agents: cephalosporins, penicillinase-resistant penicillins, and low concentrations of aminoglycosides (Hollenbeck and Rice 2012). Vancomycin remains the drug of choice to treat enterococcal infections but nowadays vancomycin-resistant Enterococci (VRE) isolates are a global challenge. Such infections are treated by a combination of cell wall-active agents with aminoglycosides that achieve synergistic bactericidal activity. Aminoglycoside antibiotics are positively charged, carbohydrate-containing molecules that find a clinical use for the treatment of infections caused by both Gram-negative and Gram-positive bacteria (Miller et al. 2014). Two antibiotic resistance phenotypes among Enterococci including VRE and HLGR have emerged as important nosocomial causes throughout the world as well as in Iran (Sakoulas et al. 2013a; Emaneini et al. 2016; Osuka et al. 2016). Operon associated genes (van operon) are responsible for VRE emergence and are the best studied antibiotic resistance operon (Faron et al. 2016).

In contrast to the rest of aminoglycosides, gentamicin is not inactivated by target insensitivity and enzymatic modification and, therefore, remained for many years the aminoglycoside often used to achieve synergistic killing of Enterococci. However, there are some variants of aminoglycosides modifying enzymes (AMEs) that can affect gentamicin (Garneau-Tsodikova and Labby 2016).

There are limited studies regarding the simultaneous resistance to vancomycin and gentamicin in hospitalized burn patients in Iran and thereby, we aimed to follow up and monitor the antibiotic resistance pattern and the genes encoding resistance to these two antibiotics among Enterococcus spp. strains isolated from burn patients.

Experimental

Materials and Methods

Bacterial isolates. One hundred seventy-nine out of 628 bacterial isolates confirmed as Enterococcus spp. were isolated from different clinical specimens (wound biopsies and blood) of burned patients referred to Taleghani burn hospital (the only referral burn center in Ahvaz, Khuzestan province, Iran) during January 2015 to 2016. All isolates were primary identified by conventional microbiological methods and confirmed by specific tests such as bile esculin hydrolysis and PYR (Pyrrolidinyl Aminopeptidase) Test. Sugar fermentation (arabinose and sorbitol) were used for characterization of Enterococcus species (Emaneini et al. 2016). Molecular confirmation of the species was done by screening for the ddl (d-alanine-d-alanine ligase) gene (Dutka-Malen et al. 1995).

Antimicrobial susceptibility testing. Susceptibility to antimicrobial agents was determined by disk diffusion method according to CLSI criteria using commercially available disks (Mast, UK) including vancomycin, teicoplanin, gentamicin, chloramphenicol, linezolid, ciprofloxacin, and amoxicillin. The Minimum Inhibitory Concentration (MIC) against gentamicin and vancomycin were determined using two-fold serial agar dilution method with Mueller Hinton agar (Difco, USA) according to CLSI guidelines (CLSI 2015). E. faecalis ATCC 29212 and E. faecium IP 4107 (The Collection of Institut Pasteur, France) were used as quality control reference strains.

DNA extraction and PCR. DNA was extracted using the AccuPrep® Genomic DNA Extraction Kit (Bioneer, South Korea). The oligonucleotide primers used in this study are listed in Table I. For each sample, the PCR assay was performed to identify vanA, vanB, vanC genes for vancomycin resistance and aac(6’)-Ie aph(2’’), aph(3’)-IIIa and ant(4’)-Ia genes for gentamicin resistance. DNA amplification was carried out in a peqSTAR thermal cycler system using a defined protocol as described previously (Dutka-Malen et al. 1995; Vakulenko et al. 2003).

Table I

Primers used in this study.

Target gene	Oligonucleotide sequences (5’-3’)	Size of product (bp)	Reference
van A	GGGAAAACGACAATTGC GTACAATGCGGCCGTTA	732	Dutka-Malen et al. 1995
vanB	ATG GGA AGC CGA TAG TC GAT TTC GTT CCT CGA CC	638	Dutka-Malen et al. 1995
vanC	AAT CGT CAA TTC CTG CAT GT TAA TCG TGG AAT ACG GGT TTG	299	Dutka-Malen et al. 1995
aac(6’)-Ie aph(2’’)	AGGAATTTATCGAAAATGGTAGAAAAG CACAATCGACTAAAGAGTACCAATC	369	Vakulenko et al. 2003
aph(3’)-IIIa	GGCTAAAATGAGAATATCACCGG CTTTAAAAAATCATACAGCTCGCG	523	Vakulenko et al. 2003
ant(4’)-Ia	CAAACTGCTAAATCGGTAGAAGCC GGAAAGTTGACCAGACATTACGAACT	294	Vakulenko et al. 2003

Statistical analysis. We used absolute and relative frequency to present descriptive statistics. Chi-Square test and also Fisher exact test (if it was necessary) used to explain analytical statistics. Data were analyzed by SPSS version 22.

Results

The patients referred to the hospital had different degrees of burn. From 628 bacterial isolates, isolated from patient referred to our hospital, 28.5% (179 isolates) were confirmed as Enterococci and among them 108 (60.3%) were Enterococcus faecalis and 71 (39.7%) were Enterococcus faecium. These species were isolated from blood (29.6%) and wound (70.4%) of burn patients. The drug resistance pattern of the isolates against eight antibiotics including chloramphenicol, linezolid, ciprofloxacin, amoxicillin, teicoplanin, vancomycin, and gentamicin is shown in Fig. 1. Overall, drug resistance of E. faecalis was significantly higher than that of E. faecium (p < 0.03). Among the aminoglycosides, the highest resistance rate was against gentamicin, and was observed in 45% of the E. faecalis and 15% of the E. faecium isolates, respectively. The prevalence of resistance to amoxicillin and tetracycline came in the second and third positions. However, a good sensitivity was observed toward vancomycin.

Fig. 1.

Antibiotic resistance pattern of E. faecalis and E. faecium isolated in this study against eight antibacterial agents.

The frequency of vancomycin resistance associated genes: vanA, vanB, vanC and gentamicin resistance associated genes: aph(3’)-IIIa, ant(4’)-Ia and aac(6’)-Ie aph(2’) in the two Enterococcus species is reported in Table II.

Table II

Frequency of vancomycin and gentamicin resistance genes among our local Enterococci isolates.

Genus	van gene				Gentamicin gene
	vanA	vanB	vanA, vanB	vanC	3’	4’	6’	3’ + 4’	3’ + 6’	4’ + 6’	3’ + 4’ + 6’
E. faecalis	14	3	3	1	74	8	67	3	26	6	4
E. faecium	9	0	3	2	48	10	42	6	50	2	1
Total (n)	23	3	6	3	122	18	109	9	76	8	5

Abbreviations are as follows: 3: aph (3’)-III;, 4’: ant (4’)-I;, 6’: aac (6’)-Ie aph (2’’)

The vanA gene was the most frequent vancomycin associated gene and it was detected in 23 positive cases (12.8%). As shown in Table II, the vanA and vanB genes were detected simultaneously in 6 isolates. The vancomycin MICs among VRE isolates ranged between 64 mg/l and 1024 mg/l. There was not a significant correlation between the presence of van operon genes and the vancomycin MICs.

Among the gentamicin resistance genes, the highest frequency was observed for aph(3’)-IIIa in 68% (n = 122) and aac(6’)-Ie aph(2’) in 61% (n = 109) of the cases.

The prevalence of aph(3’)-IIIa, ant(4’)-Ia and aac(6’)-Ie aph(2’) in HLGR isolates were 10.8%, 65.5% and 60.1% respectively.

The simultaneous presence of at least two gentamicin resistance associated genes also were observed and the most frequently, the simultaneous presence of aph(3’)-IIIa and aac(6’)-Ie aph(2’) genes wad detected on 109 cases (60.8%). There was no significant correlation between aminoglycoside resistance emergence and the prevalence of these genes.

The relation between the aminoglycoside resistance genes and the MIC of aminoglycosides antibiotics is reported in Table III. Of 122 isolates that were harboring the aph(3’)-IIIa gene, 7.3% (n = 9) had the gentamicin MIC of ≥ 512 µg/ml. The gentamicin MICs for isolates harboring ant(4’)-Ia and aac(6’)-Ie aph (2’) were 3 µg/ml and 11 µg/ml respectively.

Table III

The association between presence of gentamicin resistance genes and MICs against gentamicin.

Gene target		MIC (µg/ml)
		256		512		1024
		E. faecalis	E. faecium	E. faecalis	E. faecium	E. faecalis	E. faecium
Gentamicin resistance gene	aph(3’)-IIIa	0	0	3	1	2	3
	ant(4’)-Ia	0	0	3	1	0	0
	aac(6’)-Ie aph(2’’)	0	0	5	2	2	2

The HLGR phenotype was reported in 82.7% of the isolates in this study (57.4% of E. faecalis and 42.6% of E. faecium isolates, respectively). There was no significant correlation between HLGR rate and the species of Enterococci (p = 0.2).

Discussion

We studied 179 Enterococci isolates from patients, which were referred to a burn center in Ahvaz, south-west of Iran and this number is higher than other similar studies that could be a point for our study. The morbidity and mortality associated with nosocomial infections due to antimicrobial resistant Enterococci demonstrated a crude mortality rate of 17–100% in case of enterococcal bacteremia in different hospitals around the world (Edmond et al. 1996).

The clinical importance of Enterococci is directly related to their antibiotic resistance, which contributes to the risk of colonization and infection. The species of the greatest clinical importance are E. faecalis and E. faecium (Kajihara et al. 2015). In our study, E. faecalis was more prevalent, which is in concordance with other reports (Olawale et al. 2011; Komiyama et al. 2016).

The increase of antimicrobial resistance among Enterococcus spp. is a serious health problem globally and there are several reports of antimicrobial resistance among Enterococci isolated from hospitalized patients in Iran and other countries (Emaneini et al. 2016; Lan et al. 2016). In this study, antibiotic resistance rate in E. faecalis isolates was significantly higher than E. faecium. However, it has been reported that notwithstanding of its lower frequency, E. faecium has a more ability to develop antibiotic resistance (Werner et al. 2008).

Although there are an increasing number of reports on VRE emergence in other countries, in the present study most of the isolates were susceptible to vancomycin. This could be due to lower usage of vancomycin in the first place because treatment is done with other antibacterial agents. Moreover, the VRE phenotype is more often associated with E. faecium (Werner et al. 2008; Arias et al. 2010), which was not prevalent in our study. The vanA and vanB genes are the most frequently vancomycin resistance associated genes among Enterococcus spp. Enterococci which harbor the vanA gene, are resistant to vancomycin (MIC ≥ 64 µg/ml) and teicoplanin (MIC ≥ 8 µg/ml) at a high concentration. Resistance is induced by the presence of these drugs (Eliopoulos and Gold 2001). In our study the vanA gene was more prevalent but totally the vancomycin resistance was observed only in 10.6% of the isolates, which accordingly to the ability of the resistance induction in the presence of the antibiotic, could be associated with lack of exposure to vancomycin and teicoplanin, as it was mentioned previously. The vanB harboring Enterococci are resistant to a range of vancomycin concentrations: from 4 to over 1024 µg/ml. Such strains remain susceptible to teicoplanin. The vanC gene was reported in E. gallinarum and E. casseliflavus, which are intrinsically resistant to vancomycin at concentrations typically lower than or equal to 32 µg/ml (Eliopoulos and Gold 2001). In this study, we observed the vanC gene in three isolates and this gene could has been transmitted from these organisms to the clinically important E. faecalis and E. faecium spp.

In addition to the costs imposed to health systems, the importance of the VRE emergence is that these strains could serve as a van genes reservoir for other organisms, especially Staphylococcus aureus. This could be a real problem because vancomycin is the therapeutic agent of choice for methicillin-resistant S. aureus (Gardete and Tomasz 2014).

For the first time, the HLGR phenotype was reported in E. faecalis in 1979 in France, followed by some U.S. healthcare institutions, in which 25% of E. faecium isolates displayed the HLGR phenotype and a decade later, more than 60% of E. faecium isolates demonstrated this phenotype (Kobayashi et al. 2003).

Nine genes that encode enzymes targeting eight different aminoglycosides have been identified. In most cases, the aac(6’)-Ie aph (2’) gene has been found to be associated with resistance to aminoglycosides (Ramirez and Tolmasky 2010).

Like in this study, the HLGR was reported high in other similar studies from Iran, (Emaneini et al. 2016; Heidari et al. 2017). The studies conducted in other countries have also been reported a high rate of the HLGR phenotype. Almost in all these studies, high rate of HLGR was more prevalent in E. faecalis which could be related to its higher prevalence in the clinic (Wendelbo et al. 2003; Adhikari, 2010). Thus, the prevalence of aph(3’)-IIIa, ant(4’)-Ia and aac(6’)-Ie aph(2’) was higher in E. faecalis isolates in comparison to E. faecium isolates.

We could not find any report that study the relation between the presence of aminoglycosides resistance genes and the HGLR rate. The prevalence of the HLGR phenotype and the aminoglycosides resistance genes varied in the studies from all over the world (Udo et al. 2004). Nowadays, there are several genes conferring aminoglycosides resistance among Enterococci including the aph(2’’)-Ic, aph(2’’)-Id, aph(2’’)-Ib genes, and the aac(6’)-Ie-aph(2’)-Ia gene is no longer the only gentamicin resistance gene (Chow et al. 1997; Tsai et al. 1998; Kao et al. 2000). Almost all these genes are located on transposable elements, which contribute to their easy dissemination and this poses a challenge for health systems. Treatment of HLGR Enterococci is a real problem. Recently, few reports suggested using an anti-peptidoglycan active agent in combination with a membrane active agent. Accordingly, a successful treatment of endocarditis caused by E. faecalis of HLGR phenotype was demonstrated after administration of daptomycin plus ceftaroline (Sakoulas et al. 2013b). Although the HLGR Enterococci are predominant in our region, vancomycin keeps its antimicrobial effect on such strains. However, we reported VRE isolates and this could be a “red alarm” to our health system, and thus a continuous surveillance using both genetically and phenotypically methods should be done for this type of resistance.