The crude extract of A. maurorum plant was prepared with hydroalcoholic method as described in our previous study (Mirzaei et al., 2022a). The extraction solvent utilized for the crude extract was water.

The minimum inhibitory concentration (MIC) of the AME was determined using the microbroth dilution method, following a previously described protocol (Veiga et al., 2019). In summary, stock solution of AME (2 mg/mL) was subjected to serial two-fold dilution in sterile U-bottom 96-well microplates containing Mueller Hinton Broth (MHB). Subsequently, 100 μL of MHB was added to all wells. A 100 μL of the extract (2 mg/mL) was poured into each well of the first row. Inoculum of 10 μL with a concentration of 10⁸ CFU/mL was added to each test well. The plates were then covered and incubated at 37°C for 24 h. Following incubation, the MIC was determined as the lowest concentration of the tested material that visibly inhibited bacterial growth. In the negative control well, no substances were added.

Forty P. mirabilis isolates were collected from catheters at the intensive care units (ICUs) of reference hospitals in Isfahan, Iran, and identified by conventional biochemical method followed by PCR-sequencing of the ureG gene (Mirzaei et al., 2021). The local ethical committee of Isfahan University of Medical Sciences approved this study (IR.MUI.MED.REC.1401.405). Also, the participants provided written informed consent to participate in this study.

P. mirabilis ATCC7002 was used as a reference strain. The antimicrobial resistance pattern and biofilm formation of isolated bacteria was performed by disk diffusion and microtiter plate according to the Clinical and Laboratory Standards Institute (CLSI₂₀₂₂) guidelines and the O'Toole study (O’Toole, 2011). MDR isolates were classified as those resistant to at least one agent from three or more antimicrobial classes, according to Magiorakos et al. (2012). Twelve MDR and strong biofilm-forming clinical isolates were chosen for further investigation (Supplementary Table S1).

Pm Isf-1 and Pm Isf-2, lytic phages of P. mirabilis ATCC7002, were isolated from Municipal wastewater. The phage isolation and characterization are reported in our previous study (Mirzaei et al., 2022b).

In order to test the direct effect of the extract on phage titer and plaque-forming ability, a coincubation assay was performed (Pimchan et al., 2018). Briefly, 1 mL of phage lysates (Pm Isf-1: 10⁸ PFU/mL, Pm Isf-2: 10⁸ PFU/mL) were sampled and supplemented with 1 mL of two-fold serial dilutions of plant extract (62.5–1,000 μg/mL). Phage lysate alone was used as positive control. The microtubes were incubated at 37°C for 24 h. The samples were centrifuged at 5,000 × g for 5 min, and the supernatants (phages) were titered by the double-layer agar method (Clokie et al., 2009).

The anti-QS test was performed based on the ability to inhibit the production of the purple pigment violacein by Janthinobacterium lividum (J. lividum) ATCC 12472 according to Chenia’s method (Chenia, 2013).

To study the anti-biofilm activity of phage-extract on bacteria, a 96-well microplate dilution method was performed. A hundred microliter of the overnight culture of P. mirabilis (OD_620nm of 0.1) suspension were added to a 96-well polystyrene flat-bottomed plate and incubated at 37°C for 72 h. Then, the media was discarded, and the biofilms were washed with phosphate-buffered saline (PBS) (pH 7.2). Serial dilutions of AME (125–1,000 μg/mL) were prepared and were added to each biofilm-coated well in combination with the phage cocktail (1:1) at different MOIs (0.1, 1, 10). As a positive control, 100 μL of nutrient broth was added to the original biofilm of the P. mirabilis isolates. The treated plates were incubated for 18 h at 37°C. The biofilm biomass was stained by 0.1% crystal violet method and was quantified by measuring the OD₆₂₀ using a microtiter plate reader (Infinite F50, Tecan) (Mirzaei et al., 2021). The experiment was conducted in triplicates.

Vero cells (Pasteur Institute of Iran), seeded in 24-well plates (0.5 × 10⁵ cells/well), was overlaid with 200 µL of AME (125–1,000 μg/mL) and 200 µL of phage cocktail (MOIs 0.1–10). Then, 40 µL of the P. mirabilis suspensions (OD_620nm = 0.1) were added to the wells, followed by incubation at 37°C (5% CO₂) for 3–6 h. The wells were washed three times with PBS to remove non-adherent bacteria. Cells were then treated with 400 μL of 0.025% Triton X-100 for 5 min at 37°C to detach and lyse the cell monolayer and suspended in 1.6 mL phosphate-buffered saline (PBS). Bacterial colonies were counted, and the number of bacterial colonies in the treated plates were compared to the control (Rocha et al., 2007).

In vitro bladder models, as originally described by Stickler and Morgan (2006), were established and maintained at 37°C using a double-walled glass chamber. Briefly, after sterilization through of phantom bladder model, size 18 all-silicone Foley catheters were inserted, and retention balloons were inflated with 10 mL of sterile water. A sterile urine sample from a healthy male volunteer, free from urinary tract infections, was filtered and pumped into the bladder, creating a residual volume of 30 mL. P. mirabilis ATCC 7002 cell suspensions (10⁸ CFU/mL) were added to this residual urine to simulate late-stage infection and allowed to incubate for 1 h. After 45 min, the models received a single dose of a two-phage cocktail (2 × 10¹⁰ PFU/mL) and AME at the optimal concentration (125 μg/mL) in a volume of 1 mL, after which flow was restored. Viable cell counts in the residual bladder medium were measured at the beginning and end of the experiments, and pH levels were checked. Additionally, calcium deposition on the catheters was assessed using flame photometry 18 h after treatment. In this procedure, 1-cm catheter sections were placed in a solution of 95% ammonium oxalate and 5% oxalic acid, mixed vigorously for 3 min, and incubated at room temperature for 30 min. After removing the catheter sections, the solution was centrifuged at 3,000 × g for 10 min, and the supernatant was discarded. The remaining pellets were resuspended in 5 mL of 0.05 M perchloric acid, mixed well, and centrifuged again at 3,000 × g for 2 min to collect the supernatants. The calcium concentration in these supernatants was then measured using a flame photometer (Nzakizwanayo et al., 2016).

The biofilm of P. mirabilis (10⁶ CFU/mL) was formed in a 6-well microtiter plate containing a glass coverslip and incubated at 37°C for 72 h. The coverslips were subsequently exposed to 1 mL of the respective phage cocktail (final 2 × 10¹⁰ PFU/mL of each phage) in combination with 1 mL of AME (optimal concentration 125 μg/mL) and incubated for 18 h. Then, the coverslips were rinsed two times with PBS and air-dried in an incubator set to 37°C for 20 h. The biofilms were fixed with 2.5% glutaraldehyde and dehydrated using a graded ethanol series (30%–100%) for 5 min. Subsequently, the biofilms were subjected to critical point drying, sputtered with gold, and observed under SEM (FEI, Tecnai G-2S Twin) using an operating voltage of 15 kV.

Quantitative real-time PCR (qRT-PCR) assay was performed to determine the effect of the combination of phage - AME on the expression of QS and adhesion genes (mrpA, pmfA, luxS, rsmA, and rsbA) of P. mirabilis. Overnight pooled urine of P. mirabilis 7,002 was transferred to fresh urine, treated with an optimal concentration of AME (125 μg/mL) and phage cocktail (MOI 1:1), and incubated at 37°C for different time intervals (4, 16, and 48 h). Total RNA was extracted from bacterial cells using an RNA extraction kit (Jena Bioscience, Germany). The cDNA was synthesized according to the kit protocol (Jena Bioscience, Germany). qRT-PCR was performed on an ABI system (Applied Biosystems StepOne Plus™, United States). Each 20 µL reaction contained 2 × Master Mix (SYBR Green Ampliqon, Denmark), diluted cDNA (5 ng/L), primers (10 p.m. of each forward and reverse primer), and RNase-free ddH₂O. The thermocycling conditions were as follows: 10 min, 95°C, followed by 40 cycles of 15 s at 95°C, 54 °C for 60 s 16 s rRNA was used as an internal control. The primers used in this study are mentioned in our previous study (Mirzaei et al., 2022b). All samples were run in triplicate. The conventional 2^−ΔΔCT method calculated the relative expression of target genes (Livak and Schmittgen, 2001).

Vero cells (0.5 × 10⁴ cells/well) were seeded in 96-well microtiter plates in the presence of Dulbecco’s Modified Eagle Medium (DMEM, Gibco, United States) supplemented with 5% FBS (Gibco, United States) and incubated for 12 h in a humidified atmosphere with 5% CO₂ at 37°C. A 100 μL of ten-fold serial dilutions of AME (62.5–1,000 μg/mL) and Phage (MOI_s 0.01–10) were added to each well. Cell viability was assessed 24 h post incubation by MTT assay, as previously described (Mirzaei et al., 2022b).

All statistical analyses were carried out using GraphPad Prism9 (GraphPad Software, San Diego, CA, United States). All results were presented as mean and standard deviation (SD). A two-way ANOVA plus Dunnett’s multiple comparisons tests was used to analyze the phage statistically–extract coincubation assay results. P-values< 0.05 were considered statistically significant.

The antimicrobial activity of AME was previously assessed against strong biofilm-forming MDR P. mirabilis isolates. The results showed that the bulk plant extract had the best biofilm inhibition at a 125 μg/mL concentration against all isolate (Supplementary Figure S1) (Mirzaei et al., 2022a). A MIC assay on 12 selected isolates treated with AME demonstrated no antibacterial activity in any extract concentration. The lytic phages Isf-Pm1 (Casjensviridae, Lavrentievavirus, accession number OL741431) and Isf-Pm2 (Straboviridae, Tequatrovirus, accession number OL741432) have been previously shown to be active against the biofilm of the isolates efficiently. The optimal efficacy of phage cocktail was determined at MOI1 (Supplementary Figure S2) (Mirzaei et al., 2022b).

When the AME was tested for antiphage activity, no significant reduction of the phage count was observed at bulk concentrations of 1,000 mg/mL or lower (p > 0.05; Supplementary Figure S3).

Lower concentrations of the plant extract (125–1,000 μg/mL) clearly caused an increase in the anti-QS activity of MDR isolates. Overall, the combination of AME and phage cocktails exhibited no significant difference in anti-QS activity when compared to AME individual control (p > 0.05, Figure 1). However, AME alone and in combination with the phage cocktail significantly reduced QS activity in all cases.

The biomass of MDR P. mirabilis biofilm was significantly reduced by 73% after treatment with the combination of phage cocktail (MOI 1) and AME at 125 μg/mL compared to the control (p < 0.0001), and remained at a similar level for the extract-only therapy (Figure 2). Compared to combination therapy and control, the phage cocktail did not significantly alter bacterial multiplication at a high concentration (MOI 10). The combined treatment at higher concentrations did not show a significant effect on decreasing the bacterial biomass.

The adhesion assay on Vero cells showed that the combined treatment of phage cocktail (MOI 1) and the extract (125 μg/mL) decreased the adhesion of the bacteria to the Vero cells with 1.85 and 1.34 log as compared to the control and phage alone, respectively (P < 0.0001, p = 0.0002) (Figure 3).

As demonstrated in Figure 4, the phage cocktail alone, AME alone, and the combined therapy all significantly reduced levels of encrustation as compared to the control. There was no difference in urine pH after treatment with the extract alone, or in combination with phage cocktail when compared to the control.

SEM micrograph revealed that the treatment led to changes in disruption of the polysaccharide matrix of biofilm and bacterial morphology (Figures 5 B, C) as compared to the untreated control (Figures 5A). Combined phage cocktail-AME treated biofilm induced more cell lysis and biofilm destruction compared to phage or AME alone treatment (Figures 5D).

The effect of the combined phage cocktail - extract at optimum concentrations on the expression of QS- and adhesion-related genes, including mrpA, pmfA, luxS, rsmA, and rsbA, was evaluated for P. mirabilis ATCC after 4, 16, and 48 h. Expression of all examined genes was significantly downregulated after 16 h post-treatment with a combination of phage cocktail and AME ranging from 63% to 98% in comparison with the negative control (P< 0.0001) (Figure 6).