Synergistic Role of Biofilm-Associated Genes and Efflux Pump Genes in Tigecycline Resistance of

Background

Acinetobacter baumannii is a tremendous pathogenic factor in clinical infection, showing a high mortality risk [1,2]. Several studies have recognized that biofilm formation was the notable pathogenesis in A. baumannii infections and antibiotic resistance [3–5]. Eze et al [6] pointed out that there is a high correlation between antimicrobial resistance and biofilm formation and that drug-resistant strains have stronger biofilm-formation abilities. Biofilm is a highly organized multicellular community formed by the extracellular matrix secreted by bacteria. The initial attachment to the abiotic surface is mediated by pili production. There are 8 genes related to biofilm formation in A. baumannii, namely bap, bfs, abaI, bfmR, bfmS, csuA, csuB, and csuE.

At present, it is mainly considered that the increased expression of the RND efflux pump gene is one of the important mechanisms of tigecycline resistance in A. baumannii. The resistance of A. baumannii to tigecycline is related to the overexpression of the efflux pump AdeABC. As a bipedal regulator system, AdeABC is regulated by the 2-component regulatory system AdeRS. The AdeRS operon is located upstream of the AdeABC gene and encodes the sensing protein AdeS and the response regulator AdeR. When AdeS mutation causes the high expression of AdeABC, the result is an increase in the minimum inhibitory concentration (MIC) value of tigecycline. At the same time, it has been reported that the efflux pump has a synergistic effect with other drug resistance mechanisms, causing a high level of bacterial resistance.

Greene et al [7] found that environmental isolates produced more biofilm than did clinical patient isolates. In clinical isolates, the multi-drug resistance phenotype increases their resistance to the environment, and the high biofilm phenotype synergizes to improve tolerance. In environmental isolates, the multi-drug resistance phenotype carries a fitness cost of reducing desiccation tolerance, which is buffered by the high biofilm phenotype. The expression of biofilm genes can vary according to environmental conditions, suggesting that the potential of A. baumannii biofilm formation and multi-drug resistance may be related to the survival ability of A. baumannii in the environment. A. baumannii has multiple drug resistance mechanisms, and there is synergy or resistance between the different drug resistance mechanisms. The drug resistance of each A. baumannii may be the result of the synthesis of multiple drug resistance mechanisms, and the different antibiotics used in the treatment of A. baumannii in various departments may lead to this difference.

Tigecycline is a new class of tetracycline anti-infective drugs for intravenous injection, developed by Wyeth Pharmaceuticals and derived from minocycline [8,9]. Tigecycline is the first ammonia acyl antibiotic approved by the FDA for clinical use for complex abdominal cavity infection, skin infection, and the treatment of community-acquired pneumonia [10,11]. Tigecycline has an excellent effect on A. baumannii, but its drug resistance is becoming increasingly severe [12]. Several potential mechanisms that led to the drug resistance in A. baumannii have been disclosed by previous studies, in which biofilm-related genes and efflux pump genes were the primary reasons for antibiotic resistance [13–18]. Other studies have shown that the high expression of efflux pump genes and the improvement of biofilm formation ability lead to the resistance of tigecycline. Despite this, the genetic characteristics of tigecycline-resistant A. baumannii in China remain unclear. Moreover, the relevance of antibiotic resistance and biofilm formation is still controversial. In the present study, we investigated the potential associations between biofilm-associated genes and efflux pump genes in tigecycline-resistant A. baumannii in Xinjiang, China.

In this study, we investigated 78 clinical samples at the Clinical Center of People’s Hospital of Xinjiang Uygur Autonomous Region from October 2018 to October 2019 and isolated 72 A. baumannii strains. Then, all the strains were divided into an tigecycline-resistant A. baumannii group (TR-AN) and an tigecycline-sensitive A. baumannii group (TS-AN), and the efflux pump gene and the expression of common resistance genes in the region were detected. Next, the biofilm growth ability of the 2 groups of strains was compared, and the biofilm related genes of the 2 groups of strains were detected. We focused on the drug-resistant efflux pump genes and fimbriae synthesis genes in tigecycline-resistant A. baumannii. Finally, we analyzed the role and effect of the drug-resistant gene AdeB and fimbriae gene bfs in biofilm formation. Our study results provide updated data for further monitoring A. baumannii resistance levels and clinical infection control in Xinjiang.

Material and Methods

ETHICAL APPROVAL:

All procedures in this study were conducted and approved by the Ethics Committee of the People’s Hospital of Xinjiang Uygur Autonomous Region. Written informed consent was obtained from the patients for their anonymized information to be published in this article.

BACTERIAL STRAINS AND CLINICAL SAMPLES:

Seventy-two A. baumannii strains were collected from clinical specimens between October 2018 and October 2019 at the Clinical Laboratory Department of Xinjiang Regional Hospital, as identified by the VITEK Rapid Bacterial Identification card and VITEK Mass Spestrometry MALD-TOF (bioMérieux Clinical Diagnostics, France) [19,20], according to the Database from the American Society for Microbiology. The specimens then grew on standard MacConkey agar (Neogen, USA) at 37°C for 48 h. The strains were obtained from blood, 21 (29.2%); urine, 13 (18.1%); and 38 (52.8%) sputum. For all selected strains, duplicate strains of the same patient, samples of unknown type, and samples mixed with other Acinetobacter genera were excluded.

ANTIMICROBIAL SUSCEPTIBILITY TESTING:

The sensitivity of tigecycline was determined using VITEK2 Compat (bioMerieux Inc., France) by standard operating procedures, referring to recommended standards of the US FDA and Clinical Laboratory Standards Institute (CLSI), and was interpreted by the CLSI M100 2021 guidelines [21,22]. The MIC of tigecycline was detected by the broth microdilution method. When the MIC is less than 2 mg/L, it is considered sensitive; between 2 and 8 mg/L is considered intermediary; greater than 8 mg/L is considered drug resistant [23,24].

BIOFILM FORMATION ABILITY TESTING:

A. baumannii strains were streaked, incubated on blood agar plates, and cultured at 35°C for 18 h. A single colony was picked and suspended in 0.85% (w/V) saline buffer to prepare a bacterial suspension of 1 McFarland. Then, 200 μL of lysogeny broth medium and 10 μL of bacterial suspension were added into each well of a 96-well plate, with 3 parallel wells per strain, while a negative control was set. After 48 h of incubation at 35°C, the culture medium was aspirated and washed with 0.85% (w/V) saline buffer. An amount of 200 μL of 99% (v/V) methanol was added to each well to fix the strains for 20 min. The methanol was aspirated and then washed twice with 0.85% (w/V) saline buffer. An amount of 200 μL of 1% (w/V) crystal violet staining solution was added to the wells for 10 min. After dyeing, it was washed with distilled water twice and let air dry. Then, 200 μL of 95% (v/V) ethanol was added to each well to fully dissolve the crystal violet and read the absorbance value (A₅₇₀) at 570 nm with a microplate reader. The absorbance value of the experimental strain was greater than the mean value of the negative control well. The biofilm formation was positive, and the result of biofilm formation ability was the mean absorbance value of the 3 parallel wells minus the mean value of the negative control well.

GENOME DNA EXTRACTION:

The genomic DNA of the A. baumannii colonies was extracted by the Bacteria Genomic DNA assay (Beyotime, China). The purified genome DNA was quantified by the OD260/OD280 method on a microplate reader (NanoQuant, Tecan, Japan) and stored at −20°C for future use.

Genomic DNA preparations of A. baumannii colonies on blood agar plates were collected, and genomic DNA was isolated using the Qia Symphony DSP virus/pathogen kit in the Qia Symphony system, according to the manufacturer’s instructions (Qiagen, USA). The purified genomic DNA solution was stored at −20°C for future use. The amplification time was denaturation at 95°C for 2 min, denaturation at 95°C for 30 s, annealing at 56°C for 30 s, extension at 72°C for 1 min, and extension at 72°C for another 10 min, for a total of 35 cycles. DNA templates were amplified in all. The amplified products were analyzed by Tris-acetate-EDTA(TAE) buffer 1% (w/v) agarose gel electrophoresis.

The tigecycline-sensitive and tigecycline-resistant groups were detected, the differences were compared, and the expression was detected by QT-PCR. An amount of 1000 mL 0.5× TAE buffer was prepared. The glue making tank and comb were prepared with buffer and rinsed with distilled water. An amount of 1.0 g agarose powder was weighed and poured into a clean conical flask; 100 mL 0.5× TAE buffer was added, and the mixture was heated in a microwave oven for 3 min to fully melt it into a uniform and transparent shape. The conical flask was rinsed with running water to quickly cool it to about 60°C, and 3 μL DNA green nucleic acid dye was added and mixed well. The comb was inserted into the glue making tank, while slowly pouring the agarose gel into the glue making mold so no bubbles were generated (bubbles removed with pipette). The gel was placed in the refrigerator at 4°C for 40 min to completely solidify it. The rubber block was put into the refrigerator with 0.5× TAE buffer in the electrophoresis tank, with attention paid to not generate bubbles at the sampling hole. An amount of 5 μL DNA marker was added in the first sampling well, and the PCR amplification products were added into each well in sequence for 5 μL. The positive and negative electrodes of the electrophoresis tank were correctly connected, and electrophoresis at 90 V constant voltage was conducted for 40 min. After electrophoresis, the power was turned off, the results in the gel imager were observed, and photos were taken.

BIOFILM-RELATED GENE AND EFFLUX PUMP GENES DETECTION:

The 8 biofilm synthesis-related genes and efflux pump genes were assessed by using the conventional PCR method. The primers of the target genes are listed in Table 1. The PCR products were further electrophoresed on a 1.5% (w/V) agarose gel, and grayscale was valued by Image J software.

MOLECULAR EVOLUTIONARY TREE ANALYSIS:

Gene amplification was performed by enterobacterial recombinant sequencing PCR (Enterobacterial repetitive intergenic consensus PCR [ERIC]) using the ERIC®-2 single primer sequence 5′-AAGTAAGTGACTGGGGTGAGCG-3′ [25]. Twenty-three A. baumannii colonies with genotypes of bfs were selected for homologous analysis using bioinformatics software. According to the Tenover criteria [26], the patterns were the same. For different subtypes with the same band, a difference of 2 to 3 bands is considered closely related, differences of 4 to 6 bands are considered possible correlations, and difference of more than 7 strips is considered not related.

STATISTICAL ANALYSIS:

GraphPad Prism9.0.0 software was used for statistical data processing. Average distribution data is represented by the mean and standard deviation (mean±SD). The comparison of biofilm formation ability of different groups showed a skewed distribution, and the results are expressed by the median and quartile range [M (P25, P75)]. A nonparametric rank-sum test was used to analyze the biological characteristics of strains. Biofilm formation ability and tigecycline resistance were analyzed using the chi-square test. All tests were 2-tailed. Medcal15.2 software was used to draw the receiver operating characteristic (ROC) curve and get the best cut-off value. P<0.05 indicates a statistically significant difference.

Results

BASELINE CHARACTERISTICS:

As shown in Table 2, 45 men and 33 women were enrolled in the study. The patients’ average age was 60±21 years. Seventy-two A. baumannii strains were isolated. Among the 72 clinically isolated and cultured A. baumannii, 45 showed resistance to tigecycline. Additionally, 43.06% (31/72) of patients were admitted to the Intensive Care Unit, and 30.56% (22/72) had chronic obstructive pulmonary disease.

TIGECYCLINE SUSCEPTIBILITY:

The MICs of tigecycline were determined by the broth microdilution method. The tigecycline MIC of the 72 strains isolated ranged from 0.04 to 17.26 mg/L. There were 45 samples (62.5%) in the TR-AN (resistant) group and 27 (37.5%) in the TS-AN (sensitive) group (Table 3).

EFFLUX PUMP GENES ANALYSIS:

Transcription level analysis of blaOXA-23, blaOXA-58, blaOXA-51, AdeABC, and AdeRS genes showed AdeB expression in the TR-AN group was 6.1-fold higher than in the TS-AN group. AdeA, AdeC, and AdeRS genes increased 1.5, 2.1, and 0.7-fold, respectively, in the TR-AN group compared with in the TS-AN group (Figures 1, 2). AdeB gene expression of the TR-AN group was considerably higher than that of the TS-AN group (chi-square test, P<0.05).

Further transcriptional analysis of the AdeB gene in the TR-AN group with or without biofilm strains showed that the AbeB gene was 94.6% positive in the TR-AN group with biofilm formation and 37.5% positive in the TR-AN group without biofilm.

BIOFILM GROWTH DETECTION:

After 48 h of culture, the optical density value was measured and observed by laser confocal microscope; 37 strains of the 45 in the TR-AN group formed biofilms in the in vitro experiment, while in the TS-AN group, only 4 strains formed (Figure 3).

BIOFILM-RELATED GENE ANALYSIS:

In the 72 specimens, the detection rates of abaI, bfmS, bfmR, csuA, csuB, and csuE were all less than 20%, and those of bap, bfmR, and bfs were 34.7%, 41.6%, and 68.1%, respectively. A total of 37 strains of the 45 strains in the TR-AN group formed biofilms in the in vitro experiment, while in the TS-AN group, only 4 strains formed. By studying the transcriptional level of the bfs gene in the membrane-forming and the non-membrane-forming TR-AN groups, it can be concluded that the expression level of bfs in the biofilm-forming TR-AN group was 9.3-fold higher than that in the non-biofilm-forming TR-AN group. The results of biofilm-associated genes are presented in Table 4 and Figure 4.

GROWTH ABILITY OF BIOFILM WAS SIGNIFICANTLY DIFFERENT BETWEEN THE TR-AN AND TS-AN GROUPS:

Among the 72 clinically isolated A. baumannii strains, 57 formed biofilms in vitro. Regarding the ability of bacterial biofilm formation, the biofilm formation level of the TR-AN group was significantly higher than that of the TS-AN group (P<0.01; Table 5).

DIAGNOSTIC VALUE OF BIOFILM GROWTH ABILITY FOR TIGECYCLINE RESISTANCE:

The ROC curve was drawn, and the MIC value of tigecycline drug sensitivity was used as the standard group. The resistance of tigecycline was judged by the growth ability of biofilm. The upper and lower limits, group distance, and cut-off points of the measured values were determined by analyzing the ROC curves. The cumulative frequency distribution table was listed according to the selected group distance interval, and the sensitivity, specificity, and false positive rate (1-specificity) of all cut-off points were calculated respectively. The sensitivity as the ordinate represented the true positive rate, the 1-specificity as the abscissa represented the false positive rate, and the ROC curve was drawn.

The ROC curve was used to assess the A. baumannii biofilm formation based on the sensibility of tigecycline. The area under the ROC curve (AUC) for the biofilm formation was 0.92 (95% CI 0.864–0.981), optimal cut-off level was 1.13, sensitivity was 90.2%, and specificity was 82.1% (Figure 5).

ERIC-PCR HOMOLOGY ANALYSIS:

Cluster analysis was performed using Bionumics software, as shown in Figure 6. Correlation coefficient similarity of 0.5–0.8 is moderately correlated, and similarity of 0.80–1 is highly correlated. A total of 23 A. baumannii isolates, composed of the TS-AN group (G1, n=12) and TR-AN group (G2, n=11), could be clustered into 5 classes (ET1–ET5). It was suggested that some of the collected specimens may have come from the same clone strain and there was clonal transmission in the hospital. The drug resistance of the same clone strain to antibiotics was different, which may have been related to the different antibiotic selection pressure of the clone strain in different patients.

Discussion

Acinetobacter ssp. are oxidase-negative non-fermenting sugar gram-negative pathogenic bacteria [27,28]. Currently, the genus can be divided into 31 species according to genome, and 17 species have been named, of which A. baumannii is the most significant [29,30]. A. baumannii is a critical pathogen of nosocomial infections and can easily cause serious illnesses, such as ventilator-related pneumonia, sepsis, surgical wounds, and urinary tract infections. A. baumannii can adapt to the environment and acquire exogenous drug-resistance genes [31,32]. It shows a high degree of natural inherent drug resistance and acquired resistance to commonly used antibacterial drugs of various chemical structures [33,34]. Research on the tigecycline resistance mechanism of A. baumannii mainly focuses on the production of antibiotic-inactivating enzymes, high expression of efflux pumps, loss of porins, changes in drug targets, and the formation of biofilms [35–37].

A total of 38 strains (38/72, 52.8%) isolated in this study were derived from respiratory tract specimens, indicating that A. baumannii mainly caused respiratory tract infection. A. baumannii, which was colonized in hospitalized patients in Xinjiang Autonomous Region People’s Hospital, was possibly initiated by respiratory exogenous bacterial infections. Judging from the distribution of strains in the department, the ICU of this hospital is a department with a high incidence of A. baumannii infection [38–40]. This may be related to the critical condition of ICU patients, low immune function, invasive operations, indwelling of various tubes, and wide-spectrum use [38,41. Therefore, the infection rate of A. baumannii in the ICU is higher than that of other departments. Moreover, homologous strains have been detected in patients admitted to the ICU from different departments at different times, suggesting that tigecycline-resistant A. baumannii is susceptible to cross-infection in ICU wards, which is associated with severe illness, low-level autoimmunity, and more invasive procedures in ICU patients.

Through this study, we revealed the resistance of A. baumannii to tigecycline and the expression of efflux pump genes in Xinjiang, and found the high expression of AdeB, which was discovered for the first time in Xinjiang. At the same time, we found that the high expression of AdeB bacteria was highly correlated with the biofilm-forming ability of drug-resistant strains. We selected 8 genes related to biofilm-forming ability for screening. csuA, csuB, csue, and bap were expressed, but the expression of bfs was significantly different in the drug-resistant group, (P<0.05), which also showed that bfs was one of the factors of tigecycline resistance of A. baumannii. At present, there is controversy about the relationship between the production function of A. baumannii biofilm and drug resistance. At the same time, it is reported that the biofilm forming ability of A. baumannii is transmitted by the quorum-sensing bap element; however, but there was no difference in the expression of the quorum-sensing element bap between the 2 groups of A. baumannii in this study.

In recent years, researchers have found that biofilm, which can enhance adherence to equipment and cause disease, is an important virulence factor of A. baumannii [42]. Biofilm is a bacterial community formed by several bacteria that irreversibly adhere to the surface of organisms or non-living organisms and are wrapped by a self-secreted matrix, making bacteria have a high immune escape ability [43,44]. The relevance of biofilm-forming ability and drug resistance in A. baumannii is still controversial. Studies have shown that the resistance of A. baumannii biofilm formation-positive strains to ciprofloxacin, cefotaxime, and aztreonam was significantly enhanced [45–48]. Amin et al also proved that the resistance rate of biofilm-positive strains to common antibiotics was higher than that of negative strains [43]. However, Espinal et al found that biofilm formation ability was negatively correlated with drug resistance [49]. The biofilm synthesis-related gene bfs exists in the whole genome of A. baumannii, and its amino acid sequence contains functional domains related to biofilm formation [50]. However, there are few reports about its role in biofilm formation and its correlation with drug resistance in China [51]. The function of bfs is not clear, but bfs may increase the stability in the process of biofilm formation. In the present study, 8 biofilm-related genes and several significant drug-resistant efflux pump genes were amplified by PCR from A. baumannii isolated from clinical samples to explore the potential correlation between tigecycline resistance and biofilm-forming ability. Our results revealed that the tigecycline-resistant group increased the bfs gene transcription level and the corresponding biofilm formation ability (Table 4). The bfs gene is closely related to biofilm formation. The drug resistance rate of the bfs gene-positive group was higher than that of the bfs gene-negative group (chi-square test, P<0.05). In line with that, the expression level of efflux pump gene AdeB in the biofilm-forming group of A. baumannii was notably more upregulated than that of the non-biofilm-forming group (Figure 1). Therefore, it can be considered that the tigecycline resistance of clinical isolates in our hospital was significantly related to the bfs gene. Tigecycline resistance was mainly due to efflux pump genes AdeB and bfs. However, the potential regulatory mechanisms of the 2 genes at the transcriptional level need to be further explained.

Some limitations should be considered in this study. First, the association mechanism between biofilm-associated gene bfs and extracellular pump gene AdeB has yet to be further explored. Some studies have shown that inhibiting carbapenem-resistant Klebsiella pneumoniae extracellular pump genes can significantly affect biofilm formation [52–54]. Whether AdeB gene inhibition or knockout can affect the expression of bfs and inhibit biofilm formation deserves further investigation.

Conclusions

By detecting the efflux pump gene of multidrug-resistant A. baumannii in tigecycline resistance, we identified that the main efflux pump gene of tigecycline-resistant A. baumannii in our hospital is AdeB. At the same time, our study showed the correlation between biofilm forming ability and drug resistance, and the biofilm forming gene bfs showed a higher expression intensity. In conclusion, we speculate that the high expression levels of AdeB and bfs are the main factors causing the resistance of multidrug-resistant A. baumannii to tigecycline in our hospital, and there may be some correlation between them. This information can be used to better prevent the infection of A. baumannii and guide the rational use of antibiotics.