The metallic nanoparticles were prepared by chemical reduction at a molar ratio of Bi_0.8Ag_0.2. All reagents used were of analytical grade. For the synthesis, Bi(NO₃)₃.5H₂O (catalog number NBP100 from Chemika reagents) 3 mM and AgNO₃ (catalog number NIPL50 from Chemika reagents) 1 mM solutions were used as metallic precursors, and ascorbic acid (1% w/v; catalog number A7506 from Sigma, St Louis, MO, United States) was used as a reducing agent. In addition, citric acid (10% w/v; catalog number C-0759 from Sigma, St. Louis, MO, USA), tartaric acid (0.6% w/v; catalog number A3125 from Jalmek, San Nicolás de los Garza, México), and medium molecular-weight chitosan (0.5% w/v; catalog number 448877 from Sigma, St. Louis, MO, USA) were added to the previous mixture solution at a disposable polystyrene Petri dish. The reaction in aqueous solution was accomplished by UV irradiation of 900,000 μJ/cm² for 20 min in a UV Crosslinker (model CL-1000 from UVP, Cambridge, UK).

The surface plasmon resonance (SPR) of bismuth-silver nanoparticles (Bi/Ag NPs) was analyzed by ultraviolet–visible (UV–VIS) spectroscopy. The measurements were performed in a NanoDrop™ 2000 spectrophotometer (Thermo Fisher Scientific, USA). In disposable polystyrene cuvettes, in the range of 200–700 nm. The dynamic light scattering (DLS) technique was used to determine the average hydrodynamic diameter in dilution of 1:10,000 (NPs:water); the electric charge and zeta potential were established using a ZetaSizer Nano ZS90 instrument (Malvern Instruments, Spain). The individual size of metallic nanoparticles was evaluated by a histogram of size distribution from measurements of 100 Bi/Ag NPs from images obtained by scanning electronic microscopy (SEM) with 100,000 magnifications. Nanoparticle shape was determined by transmission electronic microscopy (TEM) with 500,000 magnifications. Nanoparticle elemental composition was determined with a TEM device equipping, an energy-dispersive X-ray spectroscopy (EDS) apparatus, and the compositional analysis was carried out using X-ray photoelectron spectroscopy (XPS) (ULTRA DLD, Shimadzu Ltd., Kyoto, Japan).

The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of Bi/Ag NPs for Escherichia coli ATCC 11229 and Staphylococcus aureus ATCC 29213 were measured by the broth microdilution method, according to Clinical and Laboratory Standards Institute (Clinical and Laboratory Standards Institute, 2013). In brief, a sterile, disposable, polystyrene, curved-bottom 96-well plate was inoculated with 10⁵ CFU/mL of each of the bacterial strains in Mueller–Hinton broth with different concentrations of Bi/Ag NPs (0.22, 0.43, 0.86, 1.2, 1.5, 1.72, 2.3, 2.9, 3.44, 6.88, 13.76, 27.52, and 55.05 μg/mL). The plate was then incubated at 37°C for 24 h. The optical density was measured at 600 nm wavelength. The MIC was considered as the concentration at which no bacterial growth was observed. The inoculum with the same number of bacteria, without Bi/Ag NPs but treated with gentamicin (10 μg/mL), was used as the control to inhibit the growth of microorganisms. This assay was performed in triplicate. For the MBC assay, from each Bi/Ag NP concentration where no apparent bacterial growth was observed (based on turbidimetry), an aliquot (10 μL) was taken to inoculate in a Petri dish with Mueller–Hinton agar. The plate was incubated at 37°C for 24 h, and the concentration in which there was no growth of CFU was taken as MBC.

The membrane integrity was evaluated by the lactate dehydrogenase (LDH) release using CytoTox 96® Non-Radioactive Cytotoxicity Assay (PROMEGA with catalog number: G1780), following the manufacturer’s recommendations. In summary, 10⁵ CFU/mL of each strain (E. coli ATCC 11229 and S. aureus ATCC 29213) were inoculated separately in a sterile, disposable, polystyrene, curved-bottom 96-well plate with subinhibitory nanoparticle concentrations of Bi/Ag NPs (0.22, 0.43, 0.86, and 1.72 μg/mL for S. aureus and 0.22, 0.43, 0.86, 1.72, and 3.44 μg/mL for E. coli) and incubated at 37°C for 24 h. Thereafter, the 96-well plates were centrifuged at 250 xg for 4 min, and aliquots of 50 μL of supernatant were transferred from all wells to new 96-well flat-bottom plates. Then, 50 μL of CytoTox 96 reagent was added to each well (the plate was protected from light for 30 min at room temperature). Next, 50 μL of stop solution were added to each well and incubated for 1 h at room temperature. Finally, the LDH release was determined by measuring OD at 490 nm. To determine LDH maximum release, 10 μL of lysis solution provided by the kit was added (9% v/v Triton® X-100) to 10⁵ CFU/mL of each strain (E. coli ATCC 11229 and S. aureus ATCC 29213), which was grown in Mueller–Hinton broth for 24 h. This measurement was considered 100% LDH release control. Each treatment measure was performed in duplicate.

The static microtiter plate method was used for the semiquantitative determination of biofilm formation using S. aureus ATCC 29213 by crystal violet staining. The biofilm assay was only performed in S. aureus due to its strong biofilm producer, and this characteristic is a virulence factor that is important for the establishment of infection (Ou et al., 2020). On the other hand, E. coli is a moderate-weak biofilm producer (Risal et al., 2018). For this assay, S. aureus was passaged by streaking on blood agar plate and incubating at 37°C for 24 h two times after thawing. The strain was adjusted to 0.5 MacFarland Scale and diluted to 1:100 (1.5×10⁶ UFC/ml) in tryptic soy broth (TSB) supplemented with 1% of glucose. Then, 100 μL of bacterial dilution were added per well in a 96-well plate (non-treated flat-bottom); subinhibitory concentrations of Bi/Ag NPs (0.22, 0.43, 0.86, and 1.72 μg/mL) were added per well to sterile TSB with 1% of glucose. Then, the plate was incubated for 24 h at 37°C without agitation. After that, the supernatant was removed, and adhered cells were washed twice with sterile distilled water. Next, the formed biofilm was fixed with 200 μL of methanol for 15 min, and then, 200 μL of 0.1% crystal violet was used to stain it for 15 min. Then, wells were washed five to ten times with sterile distilled water to remove the crystal violet excess. Finally, the plate was dried at 60°C, and the dyed biofilm was dissolved in 200 μL of ethanol-acetone (30:70) solution, to measure its OD at 595 nm. An aliquot of TSB with 1% glucose was used as sterilized control, and other aliquots of inoculum of strain with sterile TSB with 1% glucose and gentamicin (10 μg/mL) were used as negative control. The biofilm formation of S. aureus in TSB supplemented with 1% glucose was considered as positive control (100% biofilm formation). The Bi/Ag NP biofilm formation inhibition capacity was evaluated on 3 different days, depending on the time the nanoparticles had been synthesized (7, 14, and 48 days), with the goal of settling the time in which the NPs were more effective. Each assay (per day) was performed in duplicate.

The effect of Bi/Ag NPs on cell viability was determined using the Alamar Blue assay. For this, the endothelial (HUVEC with catalog number CRL-1730), fibroblast (NIH/3 T3 with catalog number CRL-1658), and mammary epithelial (MCF7 with catalog number HTB-22) cell lines were acquired from the American Type Culture Collection (ATCC, Manassas, VA, United States). Cell lines were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% antibiotic-antimycotic solution. Cells were grown to 80% confluence at 37°C, with a humidified atmosphere of 5% CO₂ and 95% air in an incubator. The cells were plated with density cell of 2×10⁴, 1×10⁴, and 5×10³ on 96-well, flat-bottom plates and incubated with the aforementioned conditions for 24 h. After incubation, the culture medium was removed, and cells were washed with PBS and then treated with different concentrations of Bi/Ag NPs (1.72 and 3.44 μg/mL), and the treatment was prepared in DMEM medium. The HUVEC cell line was also evaluated at other Bi/Ag NP concentrations (0.22, 0.43, 0.86, 1.72, 3.44, 6.88, 13.76, 19.52, and 27.52 μg/mL). The plate was incubated for 24 h at 37°C and 5% CO₂ atmosphere. Then, the supernatant was removed, and the cells were washed twice with DMEM medium. The percentage of relative viability was compared with cells from each cell line without NP treatment. For this, 100 μL of DMEM with 20% Alamar Blue (v/v) were added to each well, and the plate was incubated for 2 h at 37°C in a humidified incubator with 5% CO₂ (protected from light). Finally, fluorescence intensity was measured on a plate reader using excitation and emission wavelengths of 530 and 590 nm, respectively. Each concentration was performed in triplicate. The cell line without treatment was considered to have a relative viability of 100%. Blank subtraction control was included in each plate.

6–8-week-old female BALB/c mice were used for in vivo experiments. The animals were kept in 12-h light and dark cycles, with water and food ad libitum. Five animals were used per experimental group for all models. All experiments were performed according to the Mexican Official Norm of the technical specifications for the production, welfare, and use of laboratory animals (NOM-062-ZOO-1999) and approved by the internal Comité de Ética de Investigación y Bienestar Animal (CEIBA).

The murine-infected wound model was performed according to Adibhesami et al. (2017) with slight modifications; mice were anesthetized using an intraperitoneal injection of ketamine (75 mg/kg) and xylazine (15 mg/kg). The dorsal surface was shaved and wiped with 70% ethanol; next, an excision of 5–6 mm diameter was made on the skin to induce the wound. Mice were distributed randomly into four groups: Group 1 as the control of infected wound with S. aureus without treatment, Group 2 as the infected wound with S. aureus treated with 22.02 μg Bi/Ag NPs, Group 3 as the control of infected wound with E. coli without treatment, and Group 4 as the infected wound with E. coli treated with 22.02 μg Bi/Ag NPs. Immediately after the wound was made, it was inoculated with 5×10⁵ CFU of each bacterial strain. The bacterial inoculum was maintained for 2 h at the wound site for infection establishment, and then topical treatment with Bi/Ag NPs was started and applied daily for 5 days. The sacrifice of mice was performed on the 7th day after infection, and the bacterial load at the wound site was determined from the infected wound biopsy. The biopsy was placed in 5 mL of sterile saline solution, and a bacterial suspension was obtained. This suspension was 10-fold serially diluted and plated by spread plate technique in mannitol salt agar for S. aureus and EMB plates for E. coli, to quantify the bacterial load. Additionally, the biopsies were fixed in 10% buffered formaldehyde and subjected to several steps of histological processing including staining by Masson’s trichrome. The collagen expression was quantified as the percentage of the blue-stained area in the photography using ImageJ software with the color deconvolution function by the FIJI plugin. In addition, groups that are similar to the aforementioned ones were monitored until the wound was healed clinically (Day 10).

Mice were separated randomly into three groups: Group 1 as the control, mice without infection only treated with intraperitoneal 22.02 μg Bi/Ag NP injection; Group 2 as mice with sepsis treated with intraperitoneal sterile PBS injection; Group 3 as mice with sepsis treated with intraperitoneal 22.02 μg Bi/Ag NP injection. Mice were kept in polypropylene cages under constant conditions with a temperature of 24°C, 50% relative humidity, and a control light and dark cycle (12 h:12 h). Feed and water were supplied ad libitum for 24 h. Group 2 was intraperitoneally injected with 3.6×10⁹ CFU E. coli ATCC 11229, and simultaneously sterile PBS was applied. Group 3 was injected intraperitoneally with 3.6×10⁹ CFU E. coli ATCC 11229, and 22.02 μg Bi/Ag NPs were applied simultaneously. A second dose of sterile PBS (Group 2) or Bi/Ag NPs (Group 3) was administrated after 2 h of inoculation with E. coli. The mice were monitored for 24 h. The sepsis model is induced with E. coli because this model has been reported to be highly reproducible in small mammals and displays many features of human sepsis (Poli-de-Figueiredo et al., 2008).

Bovine raw milk was obtained from the dairy farm of Facultad de Agronomía of the Universidad Autónoma de Nuevo León. The milk sample was transported on ice to the laboratory and kept at 4°C until the assay was performed. In total, 10 mL of milk per flask were inoculated with 1.5×10⁵ CFU/ml of S. aureus ATCC 29213 and treated with Bi/Ag NPs at concentrations of 1.72 or 3.44 μg/mL Bi/Ag NPs, 100 μg/mL of gentamicin were used as positive control of inhibition and untreated milk was used as negative control. The milk was maintained for 24 h at 37°C, and the average bacterial load was determined by the spread plate technique of 10-fold serial dilutions in nutritive and mannitol salt agar. This assay was only evaluated in S. aureus infection, as this one is the main pathogen causing bovine mastitis (Heikkilä et al., 2018).

The data of membrane integrity and cell viability assays were analyzed using an ordinary one-way ANOVA test and Tukey’s multiple comparisons between means. For the Biofilm formation assay, the data were analyzed by the two-way ANOVA test, and the comparison between means was carried out using Tukey’s multiple comparisons. The data of collagen expression were analyzed by Welch’s t-test. The means were considered significantly different with a value of p < 0.05. All statistical analyses were performed using GraphPad Prism 8 Software (San Diego, CA, United States).

Bismuth and silver nanoparticles were obtained from a single reaction in aqueous synthesis by a chemical method. The SPR of these nanoparticles was characterized by UV–VIS spectroscopy and showed maximum absorbance bands at 257 nm and 411 nm wavelengths. The nanoparticles were stable for at least 28 days after their synthesis (Figure 1).

We then determined the average hydrodynamic size using the DLS technique: 10.15 nm for Bi NPs and 8.06 nm for Ag NPs. The zeta potential was +39.3 and + 39.4 mV for the Bi and Ag NPs, respectively, and the polydispersity index was 0.370 and 0.357 for the Bi and Ag NPs, respectively, (Table 1 along 21 days) achieving stability on 7th day after synthesis.

The average individual size was determined by scanning electron microscopy (SEM) with 100,000 magnifications; a histogram of size distribution was built using ImageJ software version 1.53 from the measurement of 100 Bi/Ag NPs 7 days after synthesis (when the Bi/Ag NPs achieve stability, as shown in Table 1), obtaining an average size of 18.39 ± 7.49 nm. Through the TEM analysis, we determined that Bi/Ag NPs were quasi-spherical (Figure 2).

The elemental analysis of Bi/Ag NPs was performed to determine atomic concentrations. In the Bi/Ag NPs synthesis, three different populations were obtained: silver (11%), bismuth (33.4%), and bimetallic bismuth and silver NPs (55.6%) (Figure 3). To confirm the existence of bimetallic nanoparticles, the EDS spectrum was measured by TEM, which was also used to determine the intensity line profile extracted from individual bimetallic NPs (Figure 4). The chemical state of bismuth and silver present in the nanoparticles was determined by XPS. The XPS survey exhibits the presence of Bi, Ag, C, and O (Figure 5). High-resolution XPS spectra at the Bi(4f) electron code level showed two asymmetrical peaks at 164.58 and 159.48 eV corresponding to Bi(4f_5/2) and Bi(4f_7/2), respectively. The separation between the peaks of Bi(4f) regions was 5.1 eV, which was consistent with the Bi₂O₃, and the presence of not metallic bismuth was observed since metallic bismuth is characterized by a separation between peaks of 5.3 eV and the presence of a peak in the Bi4f region of 157 eV (Ling et al., 2010; Sun et al., 2017). In the case of the Ag(3d), the presence of silver metallic (Δ = 6.0 eV) was confirmed.

In addition, the Bi/Ag NPs have an antibacterial effect on E. coli ATCC 11229 and S. aureus ATCC 29213 (Figure 6) with MIC and MBC of 3.44 for E. coli and 1.72 μg/mL for S. aureus (Table 2).

LDH was released in a dose-dependent manner (p < 0.05; Figure 7), indicating that a higher concentration of bismuth and silver nanoparticles increased the rupture of the plasmatic membrane in E. coli and S. aureus.

The evaluated nanoparticles can inhibit the formation of S. aureus biofilm, at higher concentrations, greater the inhibition, but this capacity is decreased in a time-dependent manner (p < 0.05), except for the concentration of 0.43 μg/mL. The Bi/Ag NPs are more effective 7 days after being synthesized (Figure 8).

The previously established nanoparticle concentrations of MBC for S. aureus (1.72 μg/mL) and E. coli (3.44 μg/mL) were evaluated in HUVEC, NIH/3 T3, and MCF7 cell lines. It was observed that the viability of these cells was not affected at different cell densities (p > 0.05; Figure 9). In addition, to corroborate that nanoparticles did not have any cytotoxic effect on HUVEC cells, a cell viability curve was performed, and it was observed that viability began to decrease from a concentration of 13.76 μg/mL (p < 0.05), eight and four times higher than the MBC used for S. aureus and E. coli, respectively (Figure 10).

An in vivo experiment was performed where wounds were infected with inoculums of different strains (S. aureus and E. coli). The results showed that treatment of 22.02 μg Bi/Ag NPs achieved a reduction in the infection in S. aureus (Figure 11) and increased collagen expression (p < 0.05). The treatments also eliminated E. coli infection (Figure 12). This way, we demonstrated that the formulation of our nanoparticles proved to be effective in combating and eliminating superficial infections.

In this model, the Bi/Ag NPs did not show effectiveness in eliminating this type of infection, given that groups 2 and 3 (sepsis treated and non-treated with Bi/Ag NP) died 24 h after infection, and there was no difference between them. However, group 1 (treated only with Bi/Ag NPs) survived, so the Bi/Ag NPs had no toxic effect.

The results showed that Bi/Ag NPs did not eliminate the infection in milk, since it was not possible to eliminate the bacteria present in it (Table 3), whereas those were already present in the milk sample or were inoculated.