Genetic characterization and in vitro activity of antimicrobial combinations of multidrug-resistant Acinetobacter baumannii from a general hospital in China.

The present study aimed to develop a rational therapy based on the genetic epidemiology, molecular mechanism evaluation and in vitro antibiotic combinations activity in multidrug-resistant Acinetobacter baumannii (MDRAB). MDRAB was screened by the Kirby-Bauer method. The random amplified polymorphic DNA technique was used to establish genetic fingerprinting, and a series of resistance genes were detected by polymerase chain reaction. Antimicrobial agents including amikacin (AK), cefoperazone/sulbactam (SCF I/II), meropenem (MEM), minocycline (MINO) and ciprofloxacin (CIP) were used to determine the minimum inhibitory concentrations (MICs) and interactions between antibiotics by the broth microdilution method and chequerboard assays. In total, 34 MDRAB strains were isolated and classified into 8 phenotypes A-H, according to their general drug susceptibilities. A total of 4 major genotypes (I-IV) were clustered at 60% a genotypic similarity threshold. High positive rates of β-lactamase TEM-1, topoisomerase IV, oxacillinase (OXA)-23, AdeB family multidrug efflux RND transporter adeB, β-lactamase AmpC, class 1 integrons (Int-1), 16S rRNA methylase rmtA, phosphotransferase aph(3), 16S rRNA methyltransferase armA were presented to exceed 90%, acetylyltransferase aac(3)-I, aac(6'-I, ant(3″)-I, 16S rRNA methylase rmtB, oxacillinase OXA-24 and metallo-β-lactamase IMP-5 genes demonstrated positive rates of 29.4-85.29%, while adeRS two-component system was not observed in any strain. MEM+SCF I or SCF II primarily exhibited synergistic effects. AK+SCF I, AK+SCF II, MINO+SCF I, MINO+SCF II, MINO+CIP and MINO+MEM primarily presented additive effects. AK+CIP demonstrated 70.59% antagonism. The antibacterial activity of SCF I was superior compared with that of SCF II. The results indicated the polyclonal genetic epidemiological trend of MDRAB in the Second Xiangya Hospital, and verified the complexity of genetic resistance. In addition, combinations suggested to be efficacious were MEM+SCF I and MEM+SCF II, which were more effective compared with other combinations for the management of MDRAB infection.

Introduction

Acinetobacter baumannii, a leading nosocomial pathogen, has been identified to induce serious infections and high mortality rates in intensive care units (ICUs) (1,2). Multidrug-resistant Acinetobacter baumannii (A. baumannii) (MDRAB), with resistance to at least three classes of antibiotics among cephalosporins, carbapenems, β-lactamases, aminoglycosides and quinolones (3), is only susceptible to certain agents such as tigecycline, and polymyxins due to intrinsic and acquired resistance (4,5). In addition, several pandrug-resistant (PDR) A. baumannii strains with resistance to almost all available antibiotics have been identified in previous decades (6,7).

Abuse of broad-spectrum antibiotics has been demonstrated to be a major cause for the development of drug resistance of A. baumannii. At present, a number of genes responsible for drug resistance have been identified though long-term studies on the mechanisms of bacterial resistance (7–10). Production of β-lactamase is suggested to be associated with the bacterial resistance to penicillin, cephalosporins and carbapenems (9). At present, four major categories have been available for the β-lactamase protein-encoding genes, including narrow-spectrum β-lactamase, extended-spectrum β-lactamase (ESBLs), metallo-β-lactamases (MBLs) and oxacillinase (OXA)-type carbapenemases (10). Bacterial resistance to aminoglycoside usually results in the production of aminoglycoside-modifying enzymes. Aminoglycoside resistance genes, including acetyltransferase (aac), phosphotransferase (aph) and adenylyltransferase (aad), have been frequently identified in MDRAB (11). For example, MDRAB has been demonstrated to acquire antimicrobial resistance genes via class 1 integrons (Int-1), which contain single or multiple gene cassettes (12). Carbapenemase and aminoglycoside resistance genes were localized within Int-1 (13). In addition, 16S rRNA methylases may confer resistance to aminoglycosides (14). The increased production of fluoroquinolone-resistant A. baumannii was demonstrated to be markedly associated with spontaneous mutations in the quinolone resistance-determining regions (QR-DRs) in DNA gyrase (gyrA) or topoisomerase IV (parC) (15).

MDRAB remains a challenge for the clinical management of life-threatening infections, including bacteremia, pneumonia and wound infections. At present, MDRAB is a lethal threat to public health due to the lack of effective antimicrobial agents available. Currently, combination therapy has been considered to be a promising method for the management of infection. Previous studies revealed that tigecycline and polymyxins were active against MDRAB (4,5), but their application is inhibited due to high toxicity and low commercial availability in China. Consequently, the effective combinations of clinical drugs may be an improved choice for treating MDRAB infection. In the present study, the genotypes and encoding resistance genes of MDRAB were determined. Based on the phenotypic analysis, the gene structure and molecular determinants that confer alternative MDRAB phenotypes were investigated. Furthermore, five drugs were used to evaluate the in vitro activity of various antibiotic combinations against MDRAB, in order to provide reliable data to support novel clinical combination therapies.

Materials and methods

Bacterial isolates

A total of 34 consecutive and non-repetitive MDRAB strains were identified using the MicroScan WalkAway-96 system (Siemens AG, Munich, Germany) from the Second Xiangya Hospital (Changsha, China) between February 2011 and May 2011. Kirby-Bauer antibiotic susceptibility testing (K-B test) was utilized to determine the susceptibility to several clinically significant antimicrobial agents. MDRAB was defined as the presence of resistance to at least three classes of antibiotics, including: Cephalosporins, carbapenems, β-lactamases, aminoglycosides and quinolones. In the present study, a total of 34 strains were collected (Table I), 30 of which were identified from sputum samples from patients with nosocomial pneumonia, referring to criteria for radiologically-confirmed pneumonia occurring ≥48 h after hospitalization in non-intubated patients (16), 3 were isolated from wound secretion and one was isolated from fluid drainage. The samples were primarily collected from the ICU Respiratory and Cardiothoracic Surgery departments. The interpretation of susceptibility test and breakpoints was performed according to the Clinical and Laboratory Standard Institute (CLSI) criteria (17). The present study was approved by the Ethics Committee of The Second Xiangya Hospital, Central South University (Changsha, China). Written informed consent was obtained from all patients.

Table I.

Characteristics of study isolates.

				Clinical antimicrobial susceptibility
NOx	PT	Date of strain separated	Ward	TZP	CAZ	SCF	ATM	IPM	AK	FEP	MEM	TIM	CIP	SXT	CTX	LEV	CN	SAM
1AI	51M	3/9	3	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
2AI	13M	3/9	4	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
3BI	46M	3/23	1	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
4AI	64M	3/22	10	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
5AI	34M	3/11	8	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
6BI	1M	3/17	3	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
7CI	65M	3/12	1	R	R	R	R	R	R	R	R	R	R	R	R	I	R	R
8AI	7M	5/14	4	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
9DI	88F	3/31	6	R	R	S	R	R	R	R	R	R	R	R	R	R	R	R
10BI	33F	3/28	3	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
11BII	25F	3/30	3	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
12BII	48F	3/28	1	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
13BII	30F	3/28	1	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
14BII	47F	4/2	2	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
15BII	55M	4/21	3	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
16BIV	25M	4/21	1	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
17AII	63M	3/20	1	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
18BI	72F	3/23	1	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
19CI	57M	3/12	1	R	R	R	R	R	R	R	R	R	R	R	R	I	R	R
20AI	78M	3/18	1	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
21AII	41M	3/10	4	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
22EII	80F	3/30	1	R	R	I	R	R	R	R	R	R	R	R	R	R	I	R
23FI	46M	4/5	4	R	R	I	R	R	R	I	R	R	R	R	R	R	R	R
24BII	43M	4/3	9	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
25AII	55M	4/5	5	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
26AII	55M	4/6	5	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
27BII	57F	4/7	2	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
28AI	70F	4/7	2	R	R	R	R	R	R	R	R	R	R	R	R	R	R	R
29DII	69M	4/5	2	R	R	S	R	R	R	R	R	R	R	R	R	R	R	R
30GII	87F	2/28	1	R	S	R	R	R	R	R	R	R	R	R	R	R	S	S
31BIII	56M	2/28	1	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
32BI	6F	2/28	7	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
33BI	27M	3/4	1	R	R	I	R	R	R	R	R	R	R	R	R	R	R	R
34HI	79M	2/28	1	R	S	I	R	R	R	R	R	R	R	R	R	R	R	R

PT, patient age and sex. Wards: 1, Intensive care unit; 2, Department of Respiratory; 3, Cardiothoracic surgery; 4, Orthopedics; 5, Neurology; 6, Geriatric Ward; 7, Department of Minimally Invasive Surgery; 8, Neurosurgery; 9, Urology; 10, Digestive system department. Antimicrobials: TZP, piperacillin/tazobactam; CAZ, ceftazidime; SCF, cefoperazone/sulbactam; ATM, aztreonam; IPM, imipenem; AK, amikacin; FEP, cefepime; MEM, meropenem; TIM, ticarcillin/clavulanic acid; CIP, ciprofloxacin; SXT, trimethoprim/sulfamethoxazole; CTX, cefotaxime; LEV, levofloxacin; CN, gentamicin; SAM, ampicillin/sulbactam. NOx: The isolates were classified into 8 phenotypes (A-H) according to their susceptibility to the tested clinical antimicrobials; A, resistant to all the aforementioned drugs; B, only intermediate to SCF; C, only intermediate to LEV; D, only susceptible to SCF; E, intermediate to SCF and CN; F, intermediate to SCF and FEP; G, susceptible to CAZ, CN and SAM; H, susceptible to CAZ and intermediate to SCF; R, resistance; I, intermediary; S, sensitivity.

Random amplified polymorphic DNA (RAPD) genotyping and detection of drug resistance genes by polymerase chain reaction (PCR)

Bacterial DNA was extracted and purified by using Tiangen UltraClean Microbial DNA Isolation kit (Tiangen Biotech Co., Ltd., Beijing, China) according to the manufacturer's instructions. MDRAB genotyping was conducted using RAPD (Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA) with an AP2 primer (5-GTTTCGCTCC-3) designed by Primers Express software (version 2.0; Applied Biosystems; Thermo Fisher Scientific, Inc.) and synthesized by Sangon Biotech Co., Ltd. (Shanghai, China), as previously described (18). The genes encoding resistance, including β-lactamase TEM-1, AmpC, metallo-β-lactamase IMP-5, oxacillinase (OXA)-23, OXA-24, acetylyltransferase aac(3)-I, aac(6′)-I, ant(3″)-I, 16S rRNA methyltransferase armA, 16S rRNA methylase rmtA, rmtB, phosphotransferase aph-(3), AdeB family multidrug efflux RND transporter adeB, adeRS two-component system, Int-1 and ParC genes, were detected by PCR. All primers were designed based on the sequences in GenBank (19) using Primer Express software v2.0 (Applied Biosystems; Thermo Fisher Scientific, Inc., Foster City, CA, USA) (Table II). The PCR total reaction volume of 20 µl, containing 10 µl 2X TaqMan PCR master mix, 1 µl forward and reverse primers (10 µmol/l), 7 µl ddH2O and 1 µl DNA template. The amplification conditions of the target genes were presented in Table III. Following amplification, 3 µl of products were electrophoresed on a 1.5% agarose gel (Oxoid; Thermo Fisher Scientific, Inc.) and visualized using ethidium bromide (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) for 20 min at a voltage of 150 V.

Table II.

Primers of resistance genes.

	Primer sequences (5′-3′)
Target genes	Forward	Reverse	Size, bp
TEM-1	TTCGTGTCGCCCTTATTC	ACGCTCGTCGTTTGGTAT	512
IMP-5	CTACCGCAGCAGAGTCTTTG	AACCAGTTTTGCCTTACCAT	587
OXA-23	TGTCATAGTATTCGTCGTT	TTCCCAAGCGGTAAA	453
OXA-24	TTTGCCGATGACCTT	TAGCTTGCTCCACCC	175
AmpC	CGACAGCAGGTGGAT	GGTTAAGGTTGGCATG	510
aac(3)-I	ACCTACTCCCAACATCAGCC	ATATAGATCTCACTACGCGC	158
aac(6′)-I	TATGAGTGGCTAAATCGA	CCCGCTTTCTCGTAGCA	395
ant(3″)-I	TGATTTGCTGGTTACGGTGAC	CGCTATGTTCTCTTGCTTTTG	284
armA	GGGGTCTTACTATTCTG	TTCCCTTCTCCTTTC	503
rmtA	CCTAGCGTCCATCCTTTCCTC	AGCGATATCCAACACACGATGG	315
rmtB	ATGAACATCAACGATGCCCTC	TTATCCATTCTTTTTTATCAAGTATAT	756
aph(3)	ATACAGAGACCACCATACAGT	GGACAATCAATAATAGCAAT	234
adeB	GTATGAATTGATGCTGC	CACTCGTAGCCAATACC	1,000
adeRS	CTCAGACTCCCGTGATCATGTTG	CGTAAGTCTTCGACTAAGTGAGA	1,115
Int-1	GCACCGCCAACTTTC	CCTTGATGTTACCCGAGA	433
ParC	CTGAACAGGCTTACTTGAA	AAGTTATCTTGCCATTCG	200

Table III.

Polymerase chain reaction conditions of target genes.

		Amplification
Target genes	Initialdenaturation (°C, min)	Denaturation (°C, sec)	Annealing (°C, sec)	Extension (°C, sec)	Cycles (n)	Final extension (°C, min)
AP2	95, 6	95, 45	33, 45	72, 120	45	72, 10
TEM-1	94, 5	94, 60	55, 60	72, 50	30	72, 7
IMP	94, 5	94, 60	55, 60	72, 50	30	72, 7
OXA-23	94, 5	94, 30	48, 30	72, 35	30	72, 10
OXA-24	94, 5	94, 30	48, 30	72, 35	30	72, 10
AmpC	94, 5	94, 30	50, 30	72, 50	30	72, 10
aac(3)-I	94, 5	94, 30	55, 30	72, 30	35	72, 10
aac(6′)-I	94, 5	94, 30	55, 30	72, 30	35	72, 10
ant(3″)-I	94, 5	94, 30	55, 30	72, 30	35	72, 10
rmtA	93, 2	93, 20	55, 30	72, 30	30	72, 5
aph (3)	93, 2	93, 20	55, 30	72, 30	30	72, 5
armA	94, 5	94, 30	47, 30	72, 35	30	72, 10
rmtB	94, 5	94, 30	55, 30	72, 60	30	72, 10
adeB	95, 5	95, 30	53, 60	72, 60	30	72, 7
adeRS	95, 5	95, 30	53, 40	72, 60	30	72, 7
Int-1	94, 5	94, 30	53, 30	72, 60	30	72, 10
ParC	94, 4	94, 30	53, 30	72, 40	30	72, 7

Antimicrobial agents and minimum inhibitory concentration (MIC) assays

Antimicrobial agents used in the present study were amikacin (AK), cefoperazone/sulbactam [SCF, SCFI 1:1 (cefoperazone:sulbactam) and SCFII 2:1 (cefoperazone:sulbactam)], meropenem (MEM), minocycline (MINO) and ciprofloxacin (CIP). MIC assays were performed in 96-well microtiter plates by the broth microdilution method, according to the CLIS protocol (17). Bacteria were cultured in 10% horse blood agar (Oxoid; Thermo Fisher Scientific, Inc.) for 20–24 h until cells reached the exponential phase. The inoculums were adjusted with fresh Cationic adjustment of Mueller-Hinton Broth [Oxoid; Thermo Fisher Scientific, Inc.; CAMHB, containing Ca2+ (10–25 mg/l) and Mg2+ (10–12.5 mg/l)] to produce solutions with ~5×105 colony forming units (CFUs)/ml in a final volume of 100 µl. Subsequently, the bacteria were cultured using various concentrations of drugs: AK and SCF, 256, 128, 64, 32, 16, 8 µg/ml; MEM and MINO, 64, 32, 16, 8, 4, 2 µg/ml; CIP, 16, 8, 4, 2, 1, 0.5 µg/ml, for 18–20 h at 37°C. The average MIC (MICG), the concentration that inhibited 50% of growth (MIC50) or 90% of strains (MIC90) were calculated. All tests were performed in triplicate, and growth and sterility controls were conducted simultaneously.

Chequerboard assay

A chequerboard assay was used to determine the potential interactions between antibiotics. In each assay, a combination of two antibiotics randomly chosen from the total five was used, and the range of drug concentration was identical to the MIC assays. The drugs in the 96-well plates were diluted with CAMHB by checkerboard method as previously described (20). The broth microdilution plates were inoculated with each MDRAB isolate (initial concentration of bacteria was 0.5 McFarland) for 18–24 h at 37°C, to yield ~5×105 CFU/ml in a 100 µl volume. The effect of the combinations was analyzed through measuring the fractional inhibitory concentration index (FIC) with the following formula: FICA + FICB, where FICA was the ratio of MIC of drug A in combination compared with that of drug A used alone, and FICB was the ratio of MIC of drug B in combination compared with that of drug B used alone. The interaction was defined as synergy (FIC ≤0.5), addition (0.5< FIC ≤1), indifference (1< FIC ≤2) or antagonism (FIC >2), respectively.

Results

Antimicrobial susceptibility

The MDRAB phenotype was determined according to the susceptibility results. In total, 34 strains were isolated, among which 11 strains (29.41%) were PDR A. baumannii with resistance to almost all clinically significant agents. All the strains were resistant to piperacillin/tazobactam, imipenem, meropenem, amikacin, cefotaxime, ticarcillin/clavulanic acid and ciprofloxacin, while only 14 strains (41.18%) were resistant to cefoperazone/sulbactam. The isolates were also resistant to other tested drugs including ceftazidime (97.06%), aztreonam (97.06%), trimethoprim/sulfamethoxazole (97.06%), cefepime (94.12%), gentamicin (94.12%), ampicillin/sulbactam (94.12%) and levofloxacin (94.12%). The strains were classified into 8 phenotypes (A-H) based on the resistance to the aforementioned primary clinical drugs (Table I).

Genotypic diversity and molecular determinants for MDR

To detect the extent of genotypic diversity of the tested MDRAB, RAPD was performed and the fingerprint images were analyzed by NTsys 2.10e (Exeter Software; Exeter Publishing Ltd., East Setauket, NY, USA) using dice similarity index for cluster analysis and unweighted pair group average for dendrogram construction. The isolates were clustered into four major genotypes (I–IV) at 60% genotypic similarity threshold (Fig. 1).

Figure 1.

RAPD fingerprinting of MDRAB strains. The gel image displayed diversity RAPD genotyping pattern of each MDRAB isolates representing various phenotypes from different wards. The genotypic similarities were calculated by NTsys 2.10e using dice similarity index and UPGMA, presented in the left of the glue image with coefficients and lines. Four genotypes (I–IV) were formed at a 60% similarity level. RAPD, randomly amplified polymorphic DNA; MDRAB, multidrug resistant Acinetobacter baumannii; UPGMA, unweighted pair group average.

Table IV summarizes the distribution of resistance genes in the strains. TEM-1 and ParC were identified in all strains. A total of seven genes demonstrated high positive rates, including OXA-23 (97.06%), AdeB (97.06%), AmpC (94.12%), Int-1 (94.12%), rmtA (94.12%), aph(3) (91.18%) and armA (91.18%). The other genes, including aac(3)-I, aac(6′-I, ant(3″)-I, rmtB, OXA-24 and IMP-5, demonstrated positive rates of 29.41, 32.35, 76.47, 41.18, 85.29 and 64.71%, respectively.

Table IV.

Distribution of positivity of various resistance genes.

No.	aac(3)-I	aac(6′)-I	ant(3”)-I	OXA-23	OXA-24	TEM-1	IMP-5	AmpC	ArmA	RmtA	RmtB	Aph(3)	AdeRS	AdeB	Int1	ParC
1	N	N	P	P	P	P	N	P	P	P	N	P	N	P	P	P
2	N	N	P	P	P	P	P	P	N	P	P	P	N	P	P	P
3	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
4	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
5	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
6	N	N	P	P	P	P	P	P	P	N	N	P	N	P	P	P
7	N	N	P	P	P	P	P	P	N	P	N	P	N	P	P	P
8	N	N	P	P	P	P	P	P	P	P	P	P	N	P	P	P
9	N	N	P	P	P	P	P	P	P	P	P	P	N	P	P	P
10	N	N	P	P	P	P	N	P	P	P	N	N	N	P	P	P
11	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
12	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
13	P	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
14	N	N	P	P	P	P	P	P	P	P	P	P	N	P	P	P
15	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
16	N	N	P	P	P	P	P	P	P	P	P	P	N	P	P	P
17	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
18	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
19	N	N	P	P	N	P	P	P	P	P	N	P	N	P	P	P
20	N	P	N	P	P	P	P	P	P	P	P	P	N	P	P	P
21	N	N	P	P	P	P	P	P	P	P	N	P	N	P	P	P
22	N	N	P	P	P	P	P	P	P	P	P	P	N	P	P	P
23	P	P	N	P	N	P	N	P	P	P	N	P	N	P	P	P
24	P	P	N	P	N	P	N	P	P	P	N	P	N	P	P	P
25	N	N	N	N	N	P	N	P	N	P	P	P	N	P	P	P
26	P	N	N	P	P	P	N	N	P	P	P	P	N	P	N	P
27	P	P	P	P	P	P	N	P	P	P	N	P	N	P	P	P
28	P	P	N	P	P	P	N	P	P	P	P	P	N	P	N	P
29	P	P	P	P	P	P	N	P	P	P	P	P	N	P	P	P
30	N	P	N	P	N	P	N	N	P	P	P	N	N	P	P	P
31	P	P	P	P	P	P	P	P	P	P	P	P	N	P	P	P
32	P	P	P	P	P	P	P	P	P	P	P	P	N	P	P	P
33	N	P	N	P	P	P	N	P	P	P	N	P	N	P	P	P
34	P	P	P	P	P	P	N	P	P	P	N	N	N	N	P	P
P (%)	29.41	32.35	76.47	97.06	85.29	100.00	64.71	94.12	91.18	97.06	41.18	91.18	0.00	97.06	94.12	100.00

P (%), percentage of strains positive for each gene; P, positive; N, negative.

MIC and the interaction of drug combinations

The antibiotic susceptibility levels, expressed as MIC of AK, SCF I, SCF II, MEM, MINO and CIP, were preliminarily determined for the 34 MDRAB isolates. The distribution of MIC50, MIC90 and MICG are presented in Table V. The majority of isolates were resistant to CIP (91.18%), SCF II (91.18%), amikacin (85.29%), SCF I (82.35%) and MEM (73.53%), while less isolates (5.88%) were resistant to MINO.

Table V.

MIC parameters of drugs alone use or in combination.

Antibiotics usage	MIC50 (µg/ml)	MIC90 (µg/ml)	MICG (µg/ml)
AK alone	>256.00	>256.00	>222.24
SCF I alone	64.00	>256.00	>90.83
SCF II alone	128.00	>256.00	>130.35
MEM alone	33.60	>64.00	>46.39
MINO alone	3.50	>25.00	>5.50
CIP alone	>16.00	>16.00	>14.13
AK/SCF I
AK	8	8	8
SCF I	32	256	29.88
AK/SCF II
AK	8	>256	>24.24
SCF II	64	>256	>57.41
MEM/SCF I
MEM	2	32	2.82
SCF I	8	128	17.18
MEM/SCF II
MEM	2	>64	>6.14
SCF II	8	>256	>26.12
CIP/SCF I
CIP	0.5	>16	>0.96
SCF I	32	>256	>48.24
CIP/SCF II
CIP	0.5	>16	>1.85
SCF II	64	>256	>87.76
MINO/SCF I
MINO	2	2	2
SCF I	8	8	8
MINO/SCF II
MINO	2	4	3.24
SCF II	8	8	8
AK/MEM
AK	8	8	>37.18
MEM	16	>64	>89.82
AK/CIP
AK	>256	>256	>176.99
CIP	>16	>16	>12.25
CIP/MEM
CIP	>16	>16	>14.40
MEM	16	>64	>24.65
CIP/MINO
CIP	0.5	1	0.53
MINO	2	4	2.47
MEM/MINO
MEM	2	2	2
MINO	2	2	2

MIC, minimum inhibitory concentration; MIC50, the concentration that inhibits the growing of 50% of strains; MIC90, the concentration that inhibits the growing of 90% of strains; MICG, the average MIC; AK, amikacin; SCF, SCFI 1:1 and SCFII 2:1, cefoperazone/sulbactam; MEM, meropenem; MINO, minocycline; CIP, ciprofloxacin; FIC, fractional inhibitory concentration index.

A chequerboard assay was performed with random combinations of two drugs (Table V). Fig. 2 demonstrates the percentage of isolates inhibited at various MIC of antibiotics when use alone or in combination. The majority of the drug combinations exhibited superior inhibitory effects compared with each used alone. All the combinations demonstrated synergism, with the exception of CIP with MINO. MEM in combination with SCF I or SCF II primarily demonstrated synergy, while the combination of AK+SCF I, AK+SCF II, MINO+SCF I, MINO+SCF II, MINO+CIP, and MINO+MEM primarily exhibited additive effects. Concurrently, the combination of AK+CIP demonstrated evidence of antagonism (Fig. 3).

Figure 2.

Cumulative percentages of MDRAB strains that were inhibited by increasing concentrations of various types of drugs used alone or combination. (A) AK+SCFI; (B) AK+SCFII; (C) AK+MEM; (D) AK+CIP; (E) MEM+SCFI; (F) MEM+SCFII; (G) CIP+SCFI; (H) CIP+SCFII; (I) MINO+SCFI; (J) MINO+SCFII; (K) CIP+MEM; (L) CIP+MINO; and (M) MEM+MINO. There were two cumulative percentage lines overlapped (to 100% bacteriostasis) in parts I and M of MINO combining with SCFI and MEM. AK, amikacin; SCF, SCFI 1:1 and SCFII 2:1, cefoperazone/sulbactam; MEM, meropenem; MINO, minocycline; CIP, ciprofloxacin.

Figure 3.

The distribution of FIC of various combinations of antimicrobial drugs. (A) AK+SCFI; (B) AK+SCFII; (C) AK+MEM; (D) AK+CIP; (E) MEM+SCFI; (F) MEM+SCFII; (G) CIP+SCFI; (H) CIP+SCFII; (I) MINO+SCFI; (J) MINO+SCFII; (K) CIP+MEM; (L) CIP+MINO; and (M) MEM+MINO. Synergy, FIC ≤0.5; addition, 0.5< FIC ≤1; indifference, 1< FIC ≤2; antagonism, FIC >2. AK, amikacin; SCF, SCFI 1:1 and SCFII 2:1, cefoperazone/sulbactam; MEM, meropenem; MINO, minocycline; CIP, ciprofloxacin; FIC, fractional inhibitory concentration index.

Discussion

Antibiotic resistance patterns observed in A. baumannii exhibits the capacity to cause an epidemic globally. MDRAB has been demonstrated to lead to serious hospital-acquired infection, as limited therapeutic options are available (20). In the present study, the features of 34 MDRAB strains isolated from the Second Xiangya Hospital were investigated, including antimicrobial susceptibility, genotypes, screening of antibiotic resistance genes, MIC assay and antibiotic interactions.

In the antimicrobial susceptibility study, eight phenotypes (A-H) were classified using the K-B test. The results of the drug susceptibility were in accordance with previous studies (21,22) that highlighted the efficiency of cefoperazone/sulbactam against MDRAB. However, the expression of characteristic bacterial phenotypes may be easily affected by various environmental factors. Therefore, only determining the phenotype was not sufficient for the complete epidemiological typing of various strains. Organism identification based on the genotype is considered to be more reliable, as the genotype of each organism is unique and invariable. In the present study, a AP2 primer known to be efficient in RAPD genotyping was used to amplify the genomic DNA of MDRAB strains (23). In total, four genotypes (I–IV) were formed at a 60% similarity level. The genotypic distribution was polyclonal, which was opposite to a previous study (24). In general, the ‘classic’ outbreaks of MDRAB may be more frequently induced by a single clone spread among people, whereas for the prevalence of polyclones in the Second Xiangya Hospital, it may be associated the existence of mobile genetic elements, for example integrons. The variety of patient wards and isolate origin was hypothesized to be responsible for the transmission of different subtypes. When comparing the phenotypes with diverse fingerprinting profiles, it is noteworthy to select isolates of the same phenotype with different genotypes, as it manifested that the environment may affect the discrepancy between genotype and phenotype.

To additionally investigate the resistant mechanisms of MDRAB, the expression of the genes associated with drug resistance was detected among the 34 isolates. Differences were observed in the genetic characteristics of β-lactam, aminoglycoside and quinolones resistance. Previously, the carbapenem resistance associated with class D β-lactamase genes had been suggested to cause serious therapeutic problems in clinical practice (25,26). The high positive rate of OXA-23 (33/34) in the present study was consistent with a previous study (27), while the proportion of OXA-24-positive strains (29/34) was increased compared with a previous study (28). The production of OXA-23 and OXA-24 β-lactamase may be the major cause for the selected MDRAB representing 100% resistance to imipenem and meropenem. Other β-lactamase genes, including TEM-1 (class A), IMP-5 (class B) and AmpC (class C) were also identified in the present study, which may be associated with the resistance of MDRAB to various types of β-lactams including aztreonam, ceftazidime, cefepime, and cefoperazone/sulbactam. At present, aminoglycoside-modifying enzymes and 16S rRNA methylases have been suggested to be the most important mechanism for bacterial resistance against aminoglycosides (14). In the present study, the co-existence of aph(3) (91.18%), ant(3″)-I (76.47%), aac(3)-I (29.41%), aac(6′)-I (32.35%), armA (91.18%), rmtA (94.12%) and rmtB (41.18%) resistance genes confers the high resistance to amikacin and gentamicin in MDRAB. Mutations in ParC were observed in all isolates, which may assist in explaining the 100% resistance to ciprofloxacin. The result was similar to a previous study (29). Among the isolates, the positive rates of Int1 and adeB were 94.1 and 97.1%, respectively, while the adeRS was completely negative. Integron and efflux pump genes are non-specific resistance genes. Concurrently, integrons are widely present in MDRAB, particularly Int-1, which provides A. baumannii with a gene capture system adapted to circumvent the challenges of multiple-antibiotic treatment regimens (30). The results of the present study demonstrated that Int1 serves a crucial role in multiple drug resistance. In addition, the identification of the co-existence of Int-1 and the majority of the β-lactamase and aminoglycoside genes was notable, which is in accordance with a previous study (31). This highlighted the importance of the roles of Int-1 in the horizontal spreading of antibiotic resistance genes, which may finally result in the polyclonal prevalence of MDRAB in the Second Xiangya Hospital. AdeB is a vital component of the AdeABC efflux pump, and the expression of adeABC is regulated by the adeRS two-component regulatory system (32). The sophisticated feedback between them may account for the high positive ratio of adeB and absolute absence of adeRS observed in the present study.

The results of the present study confirmed that several genes were associated with MDR. Abuse of antibiotic chemotherapeutics affords major challenges in treating MDRAB infections. At present, traditional therapy regimens are not efficient to manage these life-threatening infections. Tigecycline and polymyxins have been considered as the last resort for treating MDRAB infections (4,5). However, they are still widely used across mainland China due to the lack of any other qualified commercial products. On this basis, the present study aimed to identify an effective regimen to manage MDRAB infections through a combination of antibiotics that are frequently used in clinical practice. In the present study, 6 drugs (iAK, SCF I, SCF II, MEM, MINO and CIP) were selected to study the in vitro activity of drug combinations against MDRAB. The lower MIC of SCFI compared with SCF II demonstrated a comparatively increased rate of in vitro activity of SCF I against MDRAB, which was incompatible with previous studies (33,34). This discrepancy may arise from the geographical and biological evolutional differences. In addition, a high percentage (76.47%) of MDRAB was susceptible to minocycline, indicating the potential of this drug for the treatment of this fatal infection (35). The chequerboard assay indicated synergism for all the tested combinations, particularly the combination of MEM+SCF I, and MEM+SCF II, with the exception of CIP+MINO. Conversely, the combinations of MINO+SCF I, MINO+SCF II, CIP+MINO and MEM+MINO demonstrated additive effects. In summary, the combination of cefoperazone-sulbactam or meropenem-minocycline has been indicated to be more active compared with ciprofloxacin-amikacin, which is similar to the recent surveillance data (16,21). The present study also demonstrated that meropenem and cefoperazone-sulbactam were generally active in MDRAB. In a previous study, the combined utilization of meropenem and sulbactam was considered a therapeutic option for A. baumannii infection (36). However, only a small number of studies have been focused on the study of the efficiency of a MEM+SCF combination. The present study attempted to investigate the in vitro activity of two types of SCF (SCF I 1:1; SCF II 2:1) combined with MEM against 34 strains of MDRAB. The results indicated a marked synergistic interaction in the majority tested isolates. Although no significant differences were observed in the activity of cefoperazone-sulbactam combined with meropenem, it revealed a novel potential option for clinical combination therapy. Meropenem belongs to the family of β-lactam antibiotics, while cefoperazone-sulbactam is a type of the third-generation cephalosporin and β-lactamase inhibitor. When meropenem is combined with cefoperazone-sulbactam, they may bind to different types of penicillin bonding proteins, executing their bactericidal effects. Concurrently, sulbactam may irreversibly inhibit β-lactamase activity. This may be the most probable explanation for the synergism observed. MINO was active against MDRAB whenever it is used alone or combination. Although it is only a second-line antibiotic for the majority of clinical bacterial infections, its potential antibacterial activity against MDRAB should not be neglected. In the present study, the combination of AK+CIP produced an antagonism of 70.59%. Therefore, the combined use of these drugs should be avoided in clinical practice.

In conclusion, the identification of fingerprinting diversity highlights the issues with the polyclonal and horizontal spread of MDRAB in the Second Xiangya Hospital. Although the co-occurrence of numerous resistance-encoding genes presented a completely threaten for the active therapy, the determination of efficacious combinations among minocycline, meropenem and cefoperazone-sulbactam, particularly MEM+SCFI and MEM+SCFII, provides improved choices for the rational clinical combination therapy for MDRAB infections. Additionally, MINO may be the alternative choice to overcome the critical resistance of A. baumannii. The present study failed to depict the pharmacokinetics and pharmacodynamics of these drug combinations. Future studies should focus on updating these data and proceed to additionally identify the clinical effects of combination therapy.