Antibiofilm and antivirulence potential of silver nanoparticles against multidrug-resistant Acinetobacter baumannii

We aimed to isolate Acinetobacter baumannii (A. baumannii) from wound infections, determine their resistance and virulence profile, and assess the impact of Silver nanoparticles (AgNPs) on the bacterial growth, virulence and biofilm-related gene expression. AgNPs were synthesized and characterized using TEM, XRD and FTIR spectroscopy. A. baumannii (n = 200) were isolated and identified. Resistance pattern was determined and virulence genes (afa/draBC, cnf1, cnf2, csgA, cvaC, fimH, fyuA, ibeA, iutA, kpsMT II, PAI, papC, PapG II, III, sfa/focDE and traT) were screened using PCR. Biofilm formation was evaluated using Microtiter plate method. Then, the antimicrobial activity of AgNPs was evaluated by the well-diffusion method, growth kinetics and MIC determination. Inhibition of biofilm formation and the ability to disperse biofilms in exposure to AgNPs were evaluated. The effect of AgNPs on the expression of virulence and biofilm-related genes (bap, OmpA, abaI, csuA/B, A1S_2091, A1S_1510, A1S_0690, A1S_0114) were estimated using QRT-PCR. In vitro infection model for analyzing the antibacterial activity of AgNPs was done using a co-culture infection model of A. baumannii with human fibroblast skin cell line HFF-1 or Vero cell lines. A. baumannii had high level of resistance to antibiotics. Most of the isolates harbored the fimH, afa/draBC, cnf1, csgA and cnf2, and the majority of A. baumannii produced strong biofilms. AgNPs inhibited the growth of A. baumannii efficiently with MIC ranging from 4 to 25 µg/ml. A. baumannii showed a reduced growth rate in the presence of AgNPs. The inhibitory activity and the anti-biofilm activity of AgNPs were more pronounced against the weak biofilm producers. Moreover, AgNPs decreased the expression of kpsMII , afa/draBC,bap, OmpA, and csuA/B genes. The in vitro infection model revealed a significant antibacterial activity of AgNPs against extracellular and intracellular A. baumannii. AgNPs highly interrupted bacterial multiplication and biofilm formation. AgNPs downregulated the transcription level of important virulence and biofilm-related genes. Our findings provide an additional step towards understanding the mechanisms by which sliver nanoparticles interfere with the microbial spread and persistence.

Synthesis and characterization of silver nanoparticles (AgNPs). AgNPs (1000 µg/ml) were prepared via chemical reduction of silver nitrate in presence of polyvinylpyrrolidone (PVP) as a stabilizer according to 25 . The obtained samples were stirred, forming AgNPs. AgNPs formation was proved by Ultraviolet-visible spectral analysis. The absorbance spectra were determined using UV spectroscopy at a wavelength of 300-700 nm. The obtained NPs were characterized using transmission electron microscopy (TEM), X-ray diffraction methods (XRD) and Fourier transform infrared (FTIR) spectroscopy 26,27 . In TEM, a drop of NPs dispersion was placed on a carbon-coated copper grid and dried at room temperature and the micrograph of the preparation was taken.
In case of XRD, crystalline structure of AgNPs powder was identified using the X-ray diffractometer with CuKα radiation (λ = 1.5405 Å) in the 2θ range of 10°-80°. For FTIR, samples were prepared by KBr-discs technique and the spectra were recorded in the range of 400-4000 cm −1 .
Bacterial isolates and identification of A. baumannii. Two hundred isolates of A. baumannii were identified from wound infections. Phenotypic identification included culture onto the MacConkey agar (OXOID, Basingstoke, England), Leed Acinetobacter agar (Hardy Diagnostics, Santa Maria, CA, USA), and CHROMagar Acinetobacter/MDR medium, and incubated under aerobic conditions at 37 C for 48 h. Gram staining, motility, oxidase, peroxidase, and oxidative-fermentation reactions were performed according to standard techniques. API 20NE bacterial identification system (Biomerieux, Marcy-l'Étoile, France) was used for species identification. Molecular identification of isolates was also conducted using PCR for detection of recA specific gene and bla -oxa51 gene as previously described 28 . The sequence of the primers are listed in supplemental Table 1.
Antimicrobial susceptibility patterns. Following Clinical and Laboratory Standards Institute guidelines 29 , bacterial cultures were plated onto the surface of Muller Hinton agar plates (HiMedia Laboratories, Mumbai, India) and susceptibility was measured by the disc-diffusion methods. The following antibiotics were tested: cephalothin (30 μg), cotrimoxazole (23.75/1.25 μg), nitrofurantoin (300 μg), trimethoprim (5 µg), ceftazidime (30 µg), tobramycin (10 µg), amikacin (30 µg), tetracycline (30 μg), gentamicin (10 µg), streptomycin (10 µg), erythromycin (15 µg), levofloxacin (5 µg), rifampicin (5 μg), chloramphenicol (30 μg), azithromycin (15 µg), imipenem (10 μg) and ciprofloxacin (5 μg Testing the biofilm strength using the microtiter plate method. The biofilm-forming abilities of the tested strains were carried out using the microtiter plate method as previously described 32 . Briefly, isolates were cultivated in Muller Hinton Broth (HiMedia Laboratories, India) containing 2% w/v sucrose (Sigma-Aldrich, USA) and incubated for 24 h at 37 °C in 96 well microtiter plates (1 × 10 5 CFU/ml, 200 μl /well). After incubation, wells were gently washed twice with PBS solution. Biofilm formed by adherent organisms was stained with crystal violet (0.2% w/v) and the excess dye was washed with distilled water. After drying, 95% ethanol was added to the wells and the optical density (OD) was measured at 570 nm using a microplate reader. Average OD values were determined for all tested isolates and the negative controls. Cut-off value was defined as the value of mean OD of the negative control plus 3X standard deviation of the negative control reads. Isolates were considered weak biofilm producers if they have a mean OD values ≤ two times the cut-off value, moderate biofilm producer if OD ≤ four times the cut-off value and strong biofilm producer if OD > four times the cut-off value.
In some experiments, the effect of AgNPs on biofilm formation was tested. To do that, the isolates were treated with 100 µl of 25 µg/ml AgNPs in BHI medium and analyzed as mentioned before. The ability of AgNPs to reduce biofilm formation was assessed by comparing the OD of the AgNPs treated wells versus control untreated wells. To test the ability of AgNPs to disperse the pre-formed biofilm layers, bacterial suspensions were allowed to grow overnight in the 96-well plates, then different aliquots of AgNPs were added to the overnight bacterial cultures. Plates were incubated for additional 24 h, and finally, the strength of the biofilms exposed to the AgNPs was re-evaluated. The experiment was performed in triplicate for each isolate.
Effect of AgNPs on microbial growth kinetics. Using the 96-well plate, we compared the effect of AgNPs on the growth of 3 representative A. baumannii strong, moderate and weak biofilm producers. Bacterial suspensions at OD595 in Muller Hinton broth were incubated at 37 °C in the presence of 5 μg/ml AgNPs (sub-MIC concentration) and OD595 was measured at 6, 24, 48 and 72 h using a microplate reader (Epoch, biotek, USA). Then, growth curves were plotted using the mean of three replicates for each strain ± SD 33 .
Quantitative detection of the effect of AgNPs virulence and biofilm-related genes using RT-PCR. The effect of AgNPs (25 µg/ml) on the expression of the virulence-related genes (kpsMII and afa/ draBC genes) and biofilm-related genes (bap, OmpA, abaI, csuA/B, A1S_2091, A1S_1510, A1S_0690, A1S_0114) was evaluated using RT-PCR. RNA was extracted from bacteria treated with AgNPs using TRIzol reagent (Sigma-Aldrich, Switzerland) and untreated bacteria were used as controls. cDNA was synthesized subsequently by Superscript III kit (Invitrogen Inc., USA) according to the manufacturer protocol. PCR reactions were carried out using ePower SYBR Green PCR Master Mix (Applied Biosystems, USA) in a 7500 Sequence Detection System (CFX 96 biorad, USA). Samples were normalized to the expression of the 16S rRNA gene and the fold change in the expression of the target genes was calculated using 2 −ΔΔCT method 34 . Expression level of each gene in AgNps treated cells was compared to the expression of untreated bacteria as a control.

In vitro infection model for analyzing the antibacterial activity of AgNPs. The ability of AgNPs
to kill extracellular and intracellular A. baumannii was determined using a co-culture infection model of A. baumannii and HFF or Vero cells, as described previously 35,36 . Briefly, 1 × 10 5 HFF or Vero cells were seeded in a 24-well plate and incubated overnight at 37C with 5% CO 2 . Log phase bacterial suspension was prepared by incubating fresh colonies of A. baumannii overnight in Mueller-Hinton broth at 37C. Then, the bacterial suspension was diluted with broth and incubated for another 2 h at 37 C to reach the log phase of growth.
First, the extracellular killing activity was evaluated as previously described 35 . For this purpose, HFF or Vero cells were challenged with 2 × 10 5 CFU/mL of A. baumannii then AgNPs (25 µg/ml) were added in a total volume of 0.5 mL. Control samples treated with PBS were included. Kinetic studies were conducted at different time points to evaluate the extracellular killing activity at different contact times (5, 15, 30 and 60 min). Post-treatment samples were serially diluted in sterile PBS, plated on Mueller-Hinton agar plates, incubated for 24 h at 37C and CFUs were counted. The extracellular killing activity of AgNPs was evaluated by determining the percentage killing at the different time points, which was calculated by the following formula: (CFU of the control untreated samples − CFU of the AgNPs treated samples/CFU of the untreated control samples) × 100.
Second, the ability of AgNPs to kill the intracellular A. baumannii was carried out using the same co-culture infection model of A. baumannii and human fibroblast skin cell line HFF-1or Vero cells 35  www.nature.com/scientificreports/ X-100 (0.1%) and samples were diluted in PBS, plated on Mueller-Hinton agar plates, incubated for 24 h at 37C and CFUs were counted. The Intracellular killing activity of AgNPs was evaluated by determining the percentage killing at different time points relative to the control untreated cells as described previously. All experiments were performed in triplicate and the mean ± standard deviation was presented.
Analysis of in vitro cytotoxicity of AgNPs. The cytocompatibility of AgNPs was evaluated using two cell lines; the human fibroblast skin cell line HFF-1 and Vero cell line (kindly provided by VACSERA tissue culture laboratory, Dokki, Giza, Egypt) as previously described 33

Results
Characterization of synthesized AgNPs. Shape and size of AgNPs, which were prepared via chemical reduction of Ag salts, were characterized using TEM. AgNPs had a spherical shape with particle size in the range of 10-50 nm (Fig. 1A). XRD measurement for a dried film from the concentrated suspension confirmed the crystalline structure of AgNPs (Fig. 1B). Figure 1C shows FTIR spectra for PVP and PVP/AgNPs. Spectrum of AgNPs stabilized with PVP was identical to that of pure PVP, showing PVP characteristic peaks at 1290 cm −1 , www.nature.com/scientificreports/ 1670 cm −1 and 2900 cm −1 , which correspond to CN bond, carbonyl group and C-H stretching, respectively 37,38 . Slight shifts in the spectrum of AgNPs indicate the formation of coordination bonds between oxygen or nitrogen of PVP with Ag atoms.
Antibacterial susceptibility testing. Antibacterial susceptibility of the isolated A. baumannii isolates were tested against 17 antibiotics. These isolates were 100% resistant to 12 antibacterial agents (cephalothin, cotrimoxazole, trimethoprim, ceftazidime, tobramycin, amikacin, tetracycline, streptomycin, erythromycin, levofloxacin, chloramphenicol and ciprofloxacin). Only azithromycin, nitrofurantoin, rifampicin, imipenem and gentamycin displayed limited degrees of resistance. The resistance profile of the 200 A. baumannii isolates and distribution of the virulence-related genes are shown in supplemental file 2.

Evaluation of the antibacterial and anti-biofilm activities of AgNPs.
Agar-well diffusion was used to test the inhibitory activity of AgNPs on bacterial growth. Exposure to 50 µg/mL AgNPs produced marked inhibition zones in all tested bacterial strains (mean = 16 mm and range = 6-27 mm). However, the biofilm formation seems to interfere with the formulation's inhibitory activity. The inhibitory activity was more pronounced on weak biofilm producers, where AgNPs induced a mean inhibition zone diameter of 22 ± 5 mm, compared to 17 ± 4 mm and 10 ± 4 mm in moderate and strong A. baumannii biofilm producers ( Fig. 2A). The growth curve of A. baumannii strains was determined in the presence of silver nanoparticles or its absence as a control. The growth rate was estimated by measuring the OD595 at different time points (6, 24, 48 and 72 h). Normal cells showed a marked increase in OD595 after 6 h and reached a stationary phase over the duration of 24-72 h. Generally, A. baumannii showed a reduced growth rate in the presence of AgNPs (at 24, 48 and 72 h), compared to untreated cells. Consistent with the resistance phenotype of the strong biofilm producers, the effect was more obvious with moderate and weak biofilm producing bacteria (Fig. 2B).
A similar effect was observed when MIC values of the formulated AgNPs against the isolated A. baumannii were determined. MIC values of the AgNPs against A. baumannii were in the range of 4-25 µg/mL. Lowest MIC values were observed for the weak biofilm producers, followed by the moderate and strong biofilm forming isolates (8 ± 4 µg/mL, 15 ± 4 µg/mL and 19 ± 6 µg/mL, respectively). Significant difference was observed between strong and moderate biofilm producers vs the weak group as shown in Fig. 2C.

Modulation of virulence and biofilm-related gene expression. The expression of kpsMII and afa/
draBC genes were assessed with and without exposure to AgNPs. In presence of AgNPs, the expression of kpsMII and afa/draBC adhesin genes decreased by 4.6-folds (p = 0.001) and 3.4-folds (p = 0.001), respectively. RT-PCR analysis of total RNA isolated from strong biofilm producers treated with AgNPs showed that the transcription of bap, OmpA, and csuA/B was significantly decreased by 4.5-, 3.1-and 3.1-folds, respectively, when compared with the untrerated bacteria (Table 1). In contrast, the transcriptional expression of the A1S_2091, A1S_1510, A1S_0690 and A1S_0114 was not affected.

In vitro infection model for analyzing the antibacterial activity of AgNPs. An in vitro infection
model was conducted to evaluate the antibacterial activity of the synthesized AgNPs against extracellular and intracellular A. baumannii. The bacterial killing efficacy increased with increasing the contact time of the treatment. The 25 μg/ml AgNPs displayed strong anti-Acinetobacter activity against extracellular bacteria, showing 10 ± 3% killing efficacy after 5 min and reaching complete killing after 60 min for Vero cells with a similar trend in case of HFF cell line (Fig. 3A). In addition, the killing efficacy against intracellular A. baumannii was investigated. There was an increase in killing efficacy with time; the percentage of killed intracellular bacteria was 29 ± 13% after 1 h, which increased to complete killing after 16 h in case of Vero cells and a similar trend was also observed in case of HFF cells (Fig. 3B).

Evaluation of cytocompatibility of AgNPs on human foreskin fibroblast (HFF-1) and Vero cell lines.
The biocompatibility of AgNPs was evaluated using the human foreskin fibroblast (HFF) and Vero cell lines. Cell viability test showed that exposure of the human fibroblasts to different AgNPs concentrations, starting from 0.25 to 25 µg/mL (the highest MIC value), did not affect the viability of the cells compared to untreated control cells, indicating that the nanoparticles are not harmful and biocompatible (Fig. 4). www.nature.com/scientificreports/

Discussion
Severe infections associated with the multidrug-resistant bacteria, have increased significantly in the last decades [39][40][41][42][43][44] . The ability of bacteria to form biofilms is one of the main virulence factors that interferes with antibiotic activity and immune defense response mechanisms [45][46][47][48] . A. baumannii has developed many virulence factors and is responsible for severe life threatening infections [49][50][51][52][53] . The bacterium affects different sites, including wound infections, pneumonia, UTI. Biofilm formation abilities and various adhesins participate in the pathogenesis of their infections and resistance to antimicrobial drugs 54 . Herein, we observed that the majority of A. baumannii were resistant to most of antibiotics and could produce strong biofilms (82.5%). Interestingly, all the isolates which contained fimH and afa/draBC could produce strong biofilms. A. baumannii from wound infections mostly harbored the fimH, afa/draBC, cnf1, csgA and cnf2 55,56 . It is worth considering that all of these strains were also MDR. The most effective antibiotic against MDR isolates included nitrofurantoin (12%) and imipenem (23%). Our results are in line with other studies which reported a high rate of biofilm formation among MDR-A. baumannii 57 .
In the last years new approaches have been developed to enhance the antimicrobial treatment such as the use antibiotic combinations, bacteriophage therapy, usage of antimicrobial NP-based formulations of old antimicrobial agent [58][59][60] . Recently, several reports have described the improved antimicrobial activity of AgNPs to be used for topical administration against multidrug-resistant bacteria 61 .
The anti-Acinetobater activity of the formulated AgNPs was shown by producing marked zones of inhibition and delay in microbial growth curves. Similar results were reported by previous studies using 14-27 µg/ mL AgNPs 62 . The inhibitory activity of AgNPs on Gram-negative bacteria is more efficient than their activity on the Gram-positive bacteria 63,64 . AgNPs exerts potent antibacterial activity through different mechanisms. They have been reported to adsorb in the form of ionic silver (Ag + ) onto the cytoplasmic membranes of Gramnegative bacteria which leads to destruction of the cell membrane, leakage of bacterial contents and death 65 . In addition, AgNPs also inhibit synthesis of bacterial cell wall 66 . They cause bacterial death through interaction with sulfur groups in essential proteins, generation of reactive oxygen species leading to bacterial death 67 . The  www.nature.com/scientificreports/ high anti-Acinetobacter activity of the AgNPs could be attributed to the small size of the particles and the larger surface area, which allows more contact with the bacterial surface 68 .
Our results showed a potent ability of silver in inhibiting the strong biofilms produced by A. baumannii. Also, overnight incubation of sliver to well-formed biofilms lead to their partial removal. This allows the potential use of AgNPs as a biofilm-disrupting agent 69 . Consistent with our results, previous studies have shown efficient effect of AgNPs in inhibiting biofilms produced by different microorganisms 70,71 . Different investigators have evaluated the ability of silver nanoparticles to inhibit biofilm formation by A. baumannii, Pseudomonas aeruginosa (synergy with tobramycin), Staphylococcus aureus (such as synergy with vancomycin), Escherichia coli and Klebsiella pneumonia at similar concentrations 23,72-78 . Importantly, this is the first report that shows that ability of AgNPs to inhibit the expression of different virulence-related genes (kpsMII and afa/draBC) and biofilm-related genes (bap, OmpA, and csuA/B), which adds to the known mechanism by which silver interferes with microbial growth. We observed that 25 µg/mL of AgNPs significantly decreased the expression of virulence and biofilm-related genes.
Studies have shown that the presence of Csu gene enables A. baumannii strains to attach and form strong biofilms through formation of pili 79 . When Csu is inactivated, the ability of pili production and subsequently the ability of bacteria to format biofilms is inhibited 80 . Finally, we concluded that decreasing the expression of bap, OmpA, and csuA/B by AgNPs is a main approach for decreasing biofilm activity of A. baumannii that will help in reducing antibiotic resistance mechanisms 81 . The limitations of our study included lack of an in vivo study, which we plan to study in the future work.

Conclusion
This study has enhanced our understanding of the characteristics of clinical A. baumannii isolates. The analysis of the isolates revealed that most isolates showed high level of resistance to antibiotics, carry bundle of virulence-related genes and different abilities to produce biofilms. Furthermore, treating the bacteria with AgNPs significantly interrupted bacterial growth and multiplication. The activity of AgNPs on A. baumannii growth kinetics, biofilm inhibition and dispersion was affected markedly by the strength of the produced biofilms. AgNPs downregulated the transcription level of important virulence and biofilm-related genes. Our findings provide an additional step towards understanding the mechanisms by which AgNPs interfere with the microbial spread and persistence.

Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. www.nature.com/scientificreports/ Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.