This study aimed to examine the antibacterial characteristics of five commonly used probiotic bacteria strains, namely Lactobacillus rhamnosus American Type Culture Collection (ATCC) 52103, Pediococcus acidilactici ATCC 8042, Bacillus coagulans ATCC 31284, Bacillus subtilis ATCC 21335, and B. licheniformis ATCC 21415, against the reference strain of MDR A. baumannii ATCC BAA-747. These reference strains were developed by the Microbiological Standardization Department of the Pasteur Institute in Iran.

The Kirby-Bauer method was employed to evaluate the susceptibility of 20 different antibiotic disks, in accordance with the guidelines established by the Clinical and Laboratory Standards Institute (CLSI). This study included different classes of antibiotics, such as beta-lactams, aminoglycosides, tetracyclines, and macrolides [39].

The antibiotics utilized in the study included ceftriaxone (30 µg), ceftazidime (30 µg), gentamicin (10 µg), amikacin (30 µg), ciprofloxacin (5 µg), piperacillin (100 µg), piperacillin-tazobactam (100/10 µg), imipenem (10 µg), meropenem (10 µg), aztreonam (30 µg), cefepime (30 µg), cefotaxime (5 µg), tobramycin (10 µg), ticarcillin-clavulanic acid (75 µg), polymyxin B (75 µg), ampicillin-sulbactam (20 µg), levofloxacin (5 µg), cotrimoxazole (40 µg), doxycycline (40 µg), and colistin (10 µg). These antibiotics were sourced from MAST Co. (Merseyside, UK).

To assess the sensitivity or resistance of each antibiotic using the Kirby-Bauer method, antibiotic discs were positioned at specific intervals on a Mueller Hinton Agar (MHA) (Merck, Darmstadt, Germany) culture medium plate. Once the bacteria had adequately multiplied, the diameter of the zone of growth inhibition beneath the antibiotic discs was evaluated. The interpretation of the results was then done based on the CLSI breakpoints [40].

After collecting the supernatant from each probiotic strain, we examined the antimicrobial characteristics of these strains using the agar well diffusion technique. To perform this method, a fresh culture suspension of the A. baumannii standard strain ATCC BAA-747 was prepared at a half McFarland dilution. The suspension was then cultured on the MHA plate using a sterile swab. Then, using a Pasteur pipette, 6 mm wells were created on the plate. 50 µL of the supernatant from each strain of probiotics were individually transferred into each well. The metabolite obtained from B. licheniformis showcased an outstanding antibacterial impact, surpassing that of other probiotics.

The presence of two genes, lanA and lanM, in B. licheniformis was determined through polymerase chain reaction (PCR) technology. To achieve this, the genome was extracted by the boiling method and added to a final volume of 20 µL for the reaction. A Nanodrop spectrophotometer (Thermo Scientific, Waltham, MA, USA) at A260/A280 nm was used to quantify the extracted genome. In our work, the ratio of 260/280 was 1.85, which indicated that the DNA had a good purity [41]. The reaction mixture consisted of 1 µL each of forward and reverse primers (Table 1), 12.5 µL of master mix, 1 µL of the extracted genome, and 4.5 µL of sterile distilled water.

The thermal program employed in the PCR reaction involved an initial denaturation step at 94 °C for 4 minutes, followed by denaturation at 94 °C for 1 minute. The annealing temperature for the specific primer targeting the lanA gene was set at 53 °C, while the annealing temperature for the specific primer targeting the lanM gene was set at 52 °C, both for a duration of 1 minute. The amplification step occurred at 72 °C for 30 seconds, repeated for a total of 30 cycles. Lastly, there was a final amplification step at 72 °C for 5 minutes. Subsequently, the PCR product was evaluated using a 1% agarose gel analysis [42].

The supernatant extracted from B. licheniformis was mixed with ethyl acetate in a decanter flask, keeping a 1:1 volume ratio (V/V), and the upper phase was filtered through a 0.45 µm filter. Next, the ethyl acetate was dried using a rotary evaporator (Boeco, Hamburg, Germany). Then, high-performance liquid chromatography (HPLC) and fourier-transform infrared spectroscopy (FTIR) methods were utilized to analyze and determine the structure [36]. The desired lipopeptide was assessed for its minimum inhibitory concentration (MIC) by preparing an initial concentration of 32 µg/mL from the resulting powder of the purified probiotic product, which was powdered using the Zirbus freeze-dryer (ZIRBUS technology GmbH, Bad Grund, Germany).

For this purpose, a germ suspension was prepared from a logarithmically growing standard strain of A. baumannii with a concentration of 10${}^6$ CFU/mL. Initially, 100 µL of Muller’s Hinton Broth (MHB) medium from Merck in Germany was added to each well of a 96-well microplate. Gradient dilutions were then performed using 100 µL of metabolites extracted from the B. licheniformis supernatant, resulting in concentrations of 32, 16, 8, 4, 2, 1, 0.5, 0.25, 0.125, and 0.0625 µg/mL, which were added to each well. Subsequently, 50 µL of the microbial suspension at the specified concentration was added to each well. Positive control wells were filled with culture medium and microbial suspension, while negative control wells only contained broth. The microplates were subsequently placed in an incubator set at 37 °C and left for 24 hours [35].

To begin, a concentration of 10${}^6$ CFU/mL bacterial suspension was prepared from logarithmically growing A. baumannii strain. Subsequently, 100 µL of the prepared microbial suspension was inoculated into microplate wells containing nutrient medium, while three wells containing only Luria broth nutrient medium were used as controls for each pathogenic strain. The microplates were incubated at 37 °C for 24 hours. Afterward, the microplates were washed three times using physiological serum and left to dry at room temperature.

For staining, 200 µL of a 1% crystal violet solution was added to the wells and allowed to sit for 15 minutes. Following this, the dye was removed; the wells were rinsed three times with physiological serum, and then dried once more at room temperature. Next, 200 µL of 95% ethanol was added to the wells. The absorbance of each well was measured at a wavelength of 578 nm using Eliza Reeder (Biotech, Winooski, VT, USA). Furthermore, we measured the state of biofilm formation in each well by examining the optical absorbance (OD) and the optical absorbance control (ODC), so that if OD $\le$ ODC indicates the absence of biofilm formation, ODC OD $\le$ (2 $\times$ ODC) is equal to weak biofilm, (2 $\times$ ODC) OD $\le$ (4 $\times$ ODC) represents moderate biofilm, and (4 $\times$ ODC) OD infers strong biofilm [46].

A microbial suspension was prepared from logarithmic growth phase standard strains of A. baumannii, with a concentration of 10${}^6$ CFU/mL. Subsequently, half of the MIC of the metabolite derived from B. licheniformis was introduced into every well and left to incubate at 37 °C for a duration of 24 hours. After 15 minutes, the wells were stained with crystal violet dye and then washed with 95% ethanol. The absorbance of the wells was assessed at a wavelength of 595 nm. The percentage of biofilm inhibition was calculated using the equation “%inhibition = control OD-sample OD/control OD $\times$ 100”. Furthermore, the qualitative investigation of biofilm inhibition was performed by utilizing field emission-scanning electron microscopy (FE-SEM) with a TESCAN Model MIRA3 instrument from Brno, the Czech Republic [47]. Biofilm imaging analysis of a standard drug-resistant strain, including A. baumannii, was conducted before and after the treatment with metabolites of B. licheniformis (probiotic) using FE-SEM electron microscopy. For this purpose, the contents of the desired wells were first dried and then the dried biofilm at the bottom of the wells was used for FE-SEM imaging with special magnifications and under vacuum conditions [48]. This analysis illustrated the proliferation and buildup of bacteria pre- and post-treatment.

The genome of the standard strain of A. baumannii was extracted using the boiling method and transferred into a microtube, which was then stored at a temperature of –20 °C. The PCR reaction consisted of 1 µL of forward and 1 µL of reverse primer (Table 2), 12.5 µL of master mix (dNTP, Mgcl2, and a Taq DNA polymerase enzyme), 1 µL of the extracted genome of A. baumannii strain, and 4.5 µL of sterile distilled water in a final volume of 20 µL.

The thermal program utilized in the PCR involved an initial denaturation cycle at 95 °C for 5 minutes, followed by 30 cycles. Each cycle consisted of denaturation at 94 °C for 1 minute. Specific primer binding occurred at different temperatures: 53 °C for the pgaA gene, 57 °C for the pgaB gene, 54 °C for the pgaC gene, and 52 °C for the pgaD gene, each for 1 minute. Subsequently, there was an amplification step at 72 °C for 1 minute specifically targeting the pga operon genes. The process was concluded with a final amplification phase at 72 °C for 5 minutes. A holding temperature of 20 °C was maintained throughout the procedure [42].

The expression of pga operon genes, involved in biofilm formation in MDR A. baumannii, was investigated using this method. Specifically, a relative qPCR test was conducted to compare the expression of genes before and after 24-hour treatment with antibiotics and metabolite derived from the supernatant of B. licheniformis. For RNA extraction and cDNA synthesis, a kit from Biofact (Celbridge, Ireland) was employed.

Following RNA extraction and cDNA synthesis, the desired gene expression was examined using the qPCR technique. The reaction mixture comprised 12.5 µL of 2X Sybergreen master mix, 1 µL of each forward & reverse primers (Table 3, Ref. [49, 50]), and 5.5 µL of cDNA, resulting in a final volume of 20 µL. The temperature program for the reaction involved an initial denaturation step at 95 °C for 4 minutes, denaturation at 94 °C for 10 seconds, primer binding at 58 °C for 20 seconds, and extension at 72 °C for 25 seconds. This cycle was repeated 30 times.

The antimicrobial property of the supernatant obtained from strains of L. rhamnosus, P. acidilactici, B. subtilis, B. coagulans, and B. licheniformis against the MDR standard strain of A. baumannii was examined utilizing the agar well diffusion technique. Among the probiotic strains tested, only the B. licheniformis strain demonstrated the ability to inhibit the growth of antibiotic-resistant strain, resulting in a non-growth zone including a well with a diameter of 10 mm (Fig. 1).

The PCR reaction created bands of 278 bp for the lanM gene and 724 bp for the lanA gene in the target probiotic strain (Fig. 2).

The metabolite was examined on a C18 column using the HPLC method, with the mobile phase rinsed using a blend of 1% acetic acid and acetonitrile (V/V) in a ratio of 60:40. Apart from the solvent peak (2.96 minutes with a frequency of 54.5%), the metabolite was detected at 5.96 minutes with a frequency of 6.64%, and its lipopeptide nature was determined by using the calibration curve and comparison with the Sigma international standard. The chromatogram obtained from the HPLC is shown in Fig. 3.

The chemical nature of the separated metabolites was confirmed through FTIR spectra analysis (Fig. 4). The IR spectrum exhibited a stretching frequency of N-H at 3300 cm${}^{-1}$, indicating the presence of this functional group. Additionally, a peak observed at 2960 cm${}^{-1}$ suggested the existence of long alkyl bonds.

Furthermore, the peak detected at 1657 cm${}^{-1}$ indicated the presence of N-CO stretches within the peptide bonds of the lipopeptide molecule. Overall, the FTIR spectra confirmed that the extracted metabolites from the B. licheniformis supernatant exhibit characteristics of lipopeptides.

The Kirby-Bauer disc diffusion agar method was utilized to perform an antibiotic sensitivity test on MDR A. baumannii. In this study, these bacteria demonstrated resistance to 10 out of the 20 antibiotics that were tested. The antibiotics to which they exhibited resistance included Ceftriaxone (CRO), Ceftazidime (CAZ), gentamicin (GM), amikacin (AK), ciprofloxacin (CIP), piperacillin (PRL), piperacillin-tazobactam (PTZ), imipenem (IMI), meropenem (MEM), treonam (ATM), cefepime (CFM), Cefotaxime (CTX), ampicillin-sulbactam (SAM), tobramycin (TN), ticillin-clavulanic acid (TCC), polymyxin B (PB), Cotrimoxazole (TS), Doxycycline (DXT), Cefepime (CPM), and Levofloxacin (LEV). It should be emphasized that intermediate outcomes were also regarded as resistant [42, 51]. The detailed results can be found in Supplementary file (Supplementary Table 1).

The MIC values of the probiotic product and several ineffective antibiotics, as mentioned in the table above, were determined and compared against the specified MDR standard strain using microdilution broth. When the MIC test results were compared, it was observed that the lipopeptide derived from the probiotic exhibited a relatively low concentration, effectively suppressing the growth of drug-resistant bacterial strain. Specifically, the MIC value for the lipopeptide derived from B. licheniformis supernatant against the MDR A. baumannii ATCC BAA-747 strain was found to be 4 µg/mL. The MIC results is observable in Supplementary file (Supplementary Table 2).

Through optical absorption measurements and utilizing the aforementioned formula, it has been demonstrated that A. baumannii’s standard strain exhibits a robust biofilm formation capability prior to its proximity to the probiotic product. The OD of the sample showed an absorption value of 0.788, while the well ODC was determined to be 0.131. Following treatment with the extracted lipopeptide, the light absorbance of the sample decreased to 0.212. Based on these findings, a 73% inhibition of biofilm formation was achieved the standard strain of A. baumannii.

The FE-SEM electron microscopy images (Fig. 5a,b) at the same magnification (10.00 kx) captured from the biofilm of the MDR bacterial strain revealed the creation and inhibition of the biofilm, both prior to and following treatment with lipopeptide extracted from B. licheniformis.

The A. baumannii genome was extracted and utilized as a template for PCR amplification. Electrophoresis of the amplified products on an agarose gel revealed bands at 605 bp for the pgaA gene, 400 bp for the pgaB gene, 518 bp for the pgaC gene, and 224 bp for the pgaD gene (Fig. 6).

The relative qPCR results were assessed prior to and following treatment a standard A. baumannii strain with lipopeptide extracted from B. licheniformis as a probiotic.

To examine and compare the expression of the mentioned genes, we require two conditions:

(A) Samples treated with different antibiotics.

(B) Samples treated with the lipopeptide extracted from B. licheniformis.

Conducting an investigation into gene expression patterns under varying conditions, such as no treatment, treatment with one of the selected antibiotics from Supplementary Table 2 (MEM), and treatment with a probiotic product. To assess the expression of the target genes, the following steps were carried out: total RNA extraction, cDNA synthesis, addition of specific primers for each target genes (pgaA, B, C, D) and reference gene (16sDNA), addition of Cybergreen master-mix, and adjustment of the temperature program on the Real-time device. During the operation of the device, the acquired data, images, graphs, and Cqs could be reviewed and analyzed. The obtained data and $\mathrm{\Delta }$$\mathrm{\Delta }$Cq values of genes were utilized to generate graphs using GraphPad Prism software (Version 9.4.1, San Diego, CA, USA) in order to assess gene expression (Fig. 7). Analysis of the p-values for the genes of interest revealed a significant decrease in their expression (p-value 0.05).

Furthermore, the melting curves analysis was conducted on the target genes following the conclusion of the qPCR process. This analysis served to validate the optimal efficiency of the primers utilized in the reaction and confirmed the specific amplification of the pga operon genes (pgaA, B, C, D) in the standard strain of A. baumannii (Supplementary file (Supplementary Fig. 1)).

Drug resistance presents a significant challenge in the treatment of microbial infections, leading researchers to explore new strategies for combating resistant microbes [52]. Certain probiotic substances have been identified as having the ability to combat bacteria and assist in overcoming drug resistance [53]. In addition to their antibacterial properties, these substances offer favorable benefits for the body by promoting microbial balance and boosting the immune system [54].

Bacterial biofilms are composed of various substances, including proteins, polysaccharides, and phospholipids, which enhance bacterial resistance to antimicrobial agents. As previously mentioned, PNAG is a polysaccharide that promotes adhesion between bacteria within the biofilm and is considered a virulence factor for bacteria. The genes of the pga operon are involved in its synthesis in A. baumannii [55]. Choi et al. [16] conducted a study on the pgaABCD locus in A. baumannii and discovered that this locus codes for proteins known as PNAG, which play a crucial role in biofilm formation. They performed PCR analysis on 30 strains of A. baumannii obtained from hospitals. Out of these 30 strains, 14 exhibited robust PNAG production, 14 displayed weak PNAG production, and two strains lacked any PNAG product. By comparing two mutant strains of A. baumannii in terms of PNAG production, they observed that biofilm formation did not occur in strains lacking this particular locus [16].

In 2022, an Egyptian study examined the impact of L. rhamnosus probiotic products on Enterococcus faecalis using the gel diffusion technique. The study demonstrated the antibacterial potential of L. rhamnosus products. However, in our study, the effect of B. licheniformis probiotic products on A. baumannii was found to be more significant when compared to L. rhamnosus products [56].

Different mechanisms likely play a role in the antibiotic activity of linear and cyclic lipopeptides. This is because they are distinct amphiphiles that not only interact with the charged groups of membrane lipids through electrostatic interactions, but also with the hydrophobic regions of the lipid bilayer in membranes [57]. These interactions can result in electrostatic and mechanical changes, decreased surface tension, enhanced metal ion absorption, and disruption of lipid bilayers [58]. In simpler terms, these lipopeptides generally have a tendency to interact with membranes and form pores within them [59]. By entering the cell cytoplasm through membrane pores, these metabolites ultimately impact the transcription and expression processes of genes associated with microbial cell virulence, altering the expression of genes of interest [60]. Hence, employing advantageous microbial metabolites to minimize or hinder the activation of genes responsible for biofilm formation in A. baumannii could present a feasible approach in tackling antibiotic-resistant strains.

In our research, we have validated the lipopeptide composition of the probiotic product through the analysis of HPLC and FTIR data. Moreover, this probiotic has shown a lower MIC value against the targeted MDR bacteria when compared to other antibiotics. Electron microscopy images have illustrated the prevention of bacterial biofilm formation. Additionally, the outcomes of this research suggest that the probiotic produced by B. licheniformis influences the downregulation of pga genes, leading to the dispersion and inhibition of PNAG synthesis, subsequently resulting in defects in the bacterial biofilm.

In 2020, Porbaran et al. [61] collected 120 hospital samples from different clinical patients. The samples comprised of 50 strains of P. aeruginosa and 70 strains of A. baumannii. Similar to our study, they utilized PCR testing to validate the existence of pga operon genes and analyzed the frequency of these resistance genes. They concluded that the gene frequencies of the pgaABC operon in A. baumannii were 45.71% for pgaA, 32.85% for pgaB, and 42.85% for pgaC [61].

In their 2021 research, Gupta et al. [62] evaluated the antibiofilm potential of glycolipid obtained from B. licheniformis against Candida glabrata biofilm. They employed a scanning electron microscope (FE-SEM) to evaluate the anti-biofilm characteristics of the derived glycolipid, which inhibited and eradicated up to 80% of C. glabrata biofilm [62].

The MIC value of our probiotic product in the present study was determined to be 4 µg/mL, indicating its strong potency as an antimicrobial agent. In a 2022 study by Roberts et al. [63], investigated the antibacterial properties of a synthetic lipopeptide against multidrug-resistant Gram-negative pathogens, including A. baumannii, and MICs were reported approximately high. It is clear that the potency of the natural probiotic product derived from B. licheniformis is better than the material used in the aforementioned study.

Lipopeptides, essentially short-chain fatty acids, when consumed by bacteria, modify bacterial metabolism. They reduce the production of adhesive substances essential for biofilm formation while facilitating an anti-adhesion mechanism [64, 65]. Lactobacilli employ diverse mechanisms to secrete antimicrobial compounds, which may involve the liberation of organic acids [66]. These substances create differences in proton electrochemical gradients and membrane permeability, resulting in increased intracellular acidity and protein denaturation [67]. Moreover, Lactobacilli produce H${}_2$O${}_2$ in the presence of oxygen, which exhibits antimicrobial effects by denaturing specific enzymes, peroxidizing membrane lipids, increasing membrane permeability, and generating antimicrobial free radicals [68].

Various research have demonstrated that, in addition to probiotics, there are other compounds with antimicrobial properties. For instance, a study conducted in 2021 on the effects of cannabidiol in managing drug resistance revealed its effectiveness against highly resistant bacteria such as S. aureus and Streptococcus pneumoniae, as well its excellent activity against biofilms. Furthermore, this substance, derived from the non-psychoactive component of marijuana, possesses the ability to act locally on drug-resistant infections [69].

In 2022, another study discovered an antibacterial peptide called lycosin II, which is derived from the venom of the spider Lycosa singoriensis. Lycosin II exhibited potent antibacterial and biofilm-inhibiting properties against both Gram-positive and Gram-negative bacteria, including resistant strains like S. aureus. The study revealed that lycosin II binds to lipoteichoic acid and lipopolysaccharide present in bacterial membranes, leading to their destruction. The results indicate that lycosin II has the potential to serve as a remedy for infections induced by strains resistant to multiple drugs [70]. Unlike our research, which focused on the effects of probiotic products on resistant A. baumannii, these two studies also found that antimicrobial agents targeting other bacteria had positive effects on reducing drug resistance.

Selvaraj et al. [71] carried out research in 2020, where they utilized a natural compound called myrtenol derived from plants to inhibit genes linked to the formation of biofilm in both standard and clinical strains of A. baumannii. They effectively inhibited the activity of various target genes including bfmR, csuA/B, bap, ompA, pgaA, pgaC, and katE. Similar to our own research, their findings demonstrated a reduction in gene expression, resulting in thinner biofilms and increased vulnerability of A. baumannii strains to commonly used antibiotics like CIP and GM [71].

The current study provided evidence that probiotic products exhibit antimicrobial characteristics, rendering them effective and safe substitutes for antibiotics when combating drug-resistant bacteria. The identified lipopeptide was found to suppress the expression of genes that contribute to biofilm formation in MDR bacterial strains.

Building on these results, probiotic products produced by B. licheniformis have the potential to affect the pga operon, leading to defects in bacterial biofilms and reduced resistance. Therefore, these materials can be used to combat resistant strains of A. baumannii, providing a potential treatment option for patients with antibiotic-resistant bacterial infections. Additionally, the lipopeptide’s mechanism of action also reduces the expression of biofilm formation genes in MDR bacterial strains.

One limitation of this study is the small number of standard strains evaluated, as well as the lack of evaluation of other probiotic Bacillus products. However, considering the remarkable capacity of B. licheniformis in producing antimicrobial and anti-biofilm compounds, it is advisable to utilize them as low-risk biological agents to combat pathogenic biofilms. In the future, promising solutions for combating drug-resistant pathogenic microorganisms could be achieved through further research and advancements in probiotic applications. The rise of antibiotic resistance highlights the need for further exploration into new alternative treatments to address the growing challenge of MDR bacteria.