Streptomycin sulfate and phosphate buffer solution (PBS) were purchased from Bio Basic, Canada. Cholesterol, polyoxyethylene sorbitan monostearate (Tween 60), sorbitan monostearate (Span 60), dimethyl sulfoxide (DMSO), and chloroform were bought from Merck, Germany. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, penicillin/streptomycin 100X, Trypsin-EDTA, Trypan blue, RPMI 1640 medium, Dulbecco’s modified Eagle medium (DMEM), fetal bovine serum (FBS) were obtained from Gibco, United States. A dialysis membrane (MWCO 12,000 Da) and MTT (dimethylthiazol-2-yl-)-2,5 were received from Sigma-Aldrich (United States). Mueller Hinton broth, Mueller Hinton agar, barium chloride, and H₂SO₄ were received from Merck, Germany. Staphylococcus aureus ATCC 6538, Escherichia coli ATCC 25922, and Pseudomonas aeruginosa ATCC 15442 were obtained from the Pasteur Institute of Iran.

One of the most known methods for preparing niosomes is the thin-layer hydration method (Hülsermann et al., 2009; Akbarzadeh et al., 2021b). Cholesterol and surfactants (Span 60 and Tween 60) with a 1:1 M ratio were dissolved in 10 ml chloroform, which was evaporated using the rotary evaporator for 1 h at 60°C and 120 rpm. Afterward, the dried thin films were hydrated using streptomycin sulfate solution in PBS (10 ml, 1.5 mg/ml) at 30°C for 1 h with stirring at 120 rpm. Subsequently, the samples were sonicated (Hielscher UP50H ultrasonic processor, Germany) for 5 min and stored at 4°C in a refrigerator. Different formulations of niosomes were prepared, as shown in Table 1.

The purpose of applying the design of experiments is to identify the factors influencing the experiment process and determine the optimal values. The design method used in this study is D-optimal design using Design-Expert 7.0.10 software (Stat-Ease Inc., United States). This technique can identify the variables that have the most impact on output and evaluate the most optimal conditions in terms of effective factors (Gunst, 1996; Bernkop-Schnürch et al., 2006). For this purpose, two factors were considered: the surfactant-to-cholesterol ratio and the Span 60-to-Tween 60 M ratio as test variables, and the nanoparticle size, polydispersity index (PDI), and encapsulation efficiency (% EE) as test responses. These variables were selected on the basis of information obtained from previous studies (Moghtaderi et al., 2021) and initial screening tests.

The morphology of optimized niosomes was characterized by the field emission scanning electron microscope (SEM). For imaging, the nanoparticle suspension was diluted 1:100 in deionized water; a drop of sample was spread on a conductor film such as aluminum and dried at room temperature.

To investigate storage stability of the synthesized niosomes, 1 ml of the streptomycin sulfate–loaded niosome solution with 1 mg/ml drug concentration was poured into glass vials and stored at 4 and 25 °C, respectively, for 1 month. Samples were evaluated for the particle size and EE% at different time intervals.

To investigate cytotoxicity of free streptomycin sulfate, free niosome, and streptomycin sulfate–loaded niosome towards the human foreskin fibroblast (HFF) normal cell line, the colorimetric MTT [(3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl-tetrazolium bromide] assay was used. In brief, the HFF cells were seeded into 96-well plates for 24 h at 37°C. Then, various concentrations of free streptomycin sulfate, free niosome, and streptomycin sulfate–loaded niosome were added into each well. After incubation time, 100 µl of MTT dye was added to the wells and incubated for 4 h at 37°C. Subsequently, 100 µl of DMSO was added, and the absorbance was measured at 570 nm using a microplate reader (AccuReader, Metertech, Taiwan), and the cell survival rate was calculated by the formula: Cell viability ( % ) =   Optical density of sample / Optical density of control  × 100 .

For control, HFF cells were incubated with the Dulbecco’s modified Eagle medium (DMEM) without the test sample (Ali et al., 2010).

A Statistical analysis was performed using Design-Expert 7.0.10 software (Stat-Ease Inc., United States) and IBM SPSS Statistics version 24. ANOVA was used to compare multiple samples, and the p value < 0.05 was considered significant.

In the structure of the niosomes, the size of the vesicles and the efficiency of streptomycin sulfate encapsulation are highly dependent on the type of surfactant and the ratio of surfactant to cholesterol. For a suitable drug delivery system, it is desired to have a small size and a high encapsulation efficiency (EE) (Moghaddam et al., 2021; Moghtaderi et al., 2021). A variety of niosomal formulations synthesized with different ratios of surfactant to cholesterol and Span 60 to Tween 60 were prepared and compared (Table 1). It was found that with the same Span 60-to-Tween 60 ratio, an increasing surfactant-to-cholesterol ratio led to a larger nano-vesicle size and lower EE. The minimum size was at low levels of the Span 60-to-Tween 60 and surfactant-to-cholesterol ratios as shown in the three-dimensional graph (Figure 1A). The maximum EE of nano-vesicles was found at the high Span 60-to-Tween 60 ratio and low levels of the surfactant-to-cholesterol ratio (Figure 1B). The lowest PDI was found at low Span 60-to-Tween 60 and high surfactant-to-cholesterol ratios (Figure 1C).

As mentioned, Span and Tween are non-ionic surfactants that have many advantages such as improved stability, broad compatibility, and flexibility of formulation. Due to the weak rigidity of Tween 60 and the high lipophilicity of Span 60, proper confinement of cholesterol and surfactant (Span 60:Tween 60) in a 1:1 M ratio can lead to the density of niosome films (Junyaprasert et al., 2012; Ghafelehbashi et al., 2019). By the combination of Span and Tween in different ratios, systems with a wide range of hydrophobic–lipophilic balance (HLB) are produced. In addition, studies have shown that high transition temperatures of Span and Tween 60 provide high levels of drug encapsulation (Bharti et al., 2012; Taymouri and Varshosaz, 2016). Increasing the amount of cholesterol increases the lipid profile and stability of the two layers and results in reduced permeability, so that the drug can be trapped more effectively in the vesicles. However, excessive amounts of cholesterol make the drug and cholesterol compete for the space between the two layers, and consequently, the drug cannot enter the structure (Balakrishnan et al., 2009). It was reported that a 1:1 M ratio of cholesterol to surfactant results in a high EE formulation (Rochdy Haj-Ahmad et al., 2015). The PDI value is an estimate of the particle distribution and their heterogeneity, which is measured between 0 and 1 (Moghassemi et al., 2015). The uniform particles have better distribution and less tendency to accumulate (Waddad et al., 2013).

Based on the data shown in Table 2, multi-criteria optimization was performed by using the desirability function to obtain the optimal formulation (Derringer and Suich, 1980; Esfahani et al., 2019; Shabani et al., 2020). According to the desired parameters, the predicted optimal formulation was calculated and compared with the experimentally obtained one (Table 3). It was found that both predicted and empirically obtained formulation showed similar values, with the latter having a size of 97.8 nm, a polydisperse index of 0.27, and an encapsulation efficiency of 86.7%.

A field emission scanning electron microscope (FE-SEM) was used to investigate the morphology of the synthesized niosomes. It was observed that the streptomycin sulfate–containing niosomes were perfectly spherical in morphology, with a smooth surface (Figure 2). The average particle size for the synthesized niosomes is approximately 20–40 nm, which is less than the size obtained by the light scattering method. This difference could be because the FE-SEM shows the nanoparticle size in the dried form (actual nanoparticle size), while DLS measures the hydrodynamic diameter, which may include any molecule (such as like ions or water molecules) attached to the nanoparticle surface.

Multilayer niosomes were observed using an FE-SEM, which has been reported previously. Niosomes prepared by thin-layer hydration is usually multilayer vesicles below 100 nm with suitable particle distribution that can confine a large amount of drug and slower drug release (Hope et al., 1986; Akbari et al., 2015; El-Sayed et al., 2017).

During storage, the niosomes can swell/break down or are affected by steric/repulsion forces. Here, we investigated the stability of the synthesized niosomes at 4 and 25°C for 60 days. It was found that samples stored at 4°C had better stability in terms of size, PDI, and EE than those stored at 25°C during the 60-day storage (Figure 3). There was a significant difference in the size of the niosomes kept at two temperatures, and the size increase for the samples stored at 4°C was slower than the corresponding sample at 25°C, which could be due to less mobility of the niosomes at 4°C (Lawrence et al., 1996; Balasubramaniam et al., 2002; Akbarzadeh et al., 2020b). Studies have also shown that the size of formulations can affect the stability of the system because, according to the theory of thermodynamics, smaller niosomes contain excess energy, which makes them unstable. The high leakage of the drug at 25°C can also be caused by the higher fluidity of the lipid vesicles at high temperature (Pardakhty et al., 2011; Akbarzadeh et al., 2020b; Akbarzadeh et al., 2020c).

The drug release rate is an essential factor for upgrading drug delivery systems. The release of encapsulated drugs within the niosomes can be optimized for controlled drug release over the long term (Tarekegn et al., 2010). Here, we compared the release profile of the streptomycin sulfate–soluble form and the encapsulated niosome form in the PBS medium for 72 h. It was found that the streptomycin sulfate release from the nano-carrier (66.4 ± 1.3%) was lower than the drug solution (97.8 ± 1.12%) during 72 h of release (Figure 4). Thus, encapsulation of streptomycin sulfate in the niosome reduced the release burst and allowed more sustainable and prolonged release. It has been previously reported that the release profile of niosomes can have two steps: the first is faster and the second is slower (Paolino et al., 2008; Akbarzadeh et al., 2020c). The rapid release of the drug in the first stage is due to the excretion of the drug from the outer surface of the niosome, and the slower release in the second stage is due to the penetration of the drug through the niosome (Manosroi and Bauer, 1989). Other factors contributing to the release can be components of the niosomes. As the amount of cholesterol in the niosomes increases, the amount of drug release from the vesicles decreases because, at higher cholesterol levels, the movement of the bilayer cannot eliminate osmotic changes and lead to the released drugs absorbed to the niosomal surface (Liang et al., 2004; Ruckmani and Sankar, 2010; Hedayati Ch et al., 2020; Sadeghi et al., 2020). In addition, as the surfactant chain length increases, drug release can continue for a longer period because the transfer temperature can affect the surfactants and make them completely fluid, thereby providing greater penetration of the drug at 37°C (Ruckmani et al., 2000; El-Ridy et al., 2018).

Different models were evaluated to fit the release kinetics of streptomycin sulfate from the optimal niosomal formulation (Table 4). Based on the model parameters and the coefficient of determination (R2) for each model, the release for niosomal formulation was found to follow the Korsmeyer–Peppas model, where the N obtained values (n < 0.45) indicate that the Fickian diffusion mechanism determines the release of streptomycin sulfate molecules from the niosomal formulation (Korsmeyer et al., 1983).

We further investigated the antimicrobial activity of the synthesized niosomes by measuring minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC). Free niosome, free streptomycin sulfate, and streptomycin sulfate–loaded niosomes against S. aureus, E. coli, and P. aeruginosa were tested. The streptomycin sulfate–loaded niosomes showed a higher antibacterial effect against all studied pathogenic bacteria than free streptomycin sulfate, with the MIC values decreased between 4- and 8-fold (Figure 5). Furthermore, lower MBC was found for the streptomycin sulfate–containing niosomes than for the free streptomycin sulfate. These results suggest that lower concentrations of niosomal streptomycin sulfate are needed to inhibit bacterial growth than free streptomycin sulfate. This could be caused by the possibility that niosomes can protect drug against the effects of bacterial enzymes and facilitate niosome fusion with the bacterial membrane, as reported previously (Mugabe et al., 2005; Moammeri et al., 2021; Moghtaderi et al., 2021).

To investigate the killing profile of the synthesized niosomes, we next performed the time-kill assay against S. aureus, E. coli, and P. aeruginosa with sublethal concentrations (half of the MIC shown in Figure 5). During the 72-h test, the loaded niosomes displayed highest antibacterial activity compared to the unloaded niosomes and free streptomycin (Figure 6). The results further demonstrate that the direct interaction of the niosomal carrier with bacteria (likely cell membrane) could be a reason for the greater antibacterial property in the niosomes, as reported previously (Zille et al., 2015; Raza et al., 2016; Ghafelehbashi et al., 2019; Moghtaderi et al., 2021).

Since biofilm is a prevalent factor causing antimicrobial resistance and accounts for 65–80% of all infections (Macià et al., 2018), the fabricated niosomes here were investigated for their efficacy against biofilms of S. aureus, E. coli, and P. aeruginosa with MIC shown in Figure 5. It was revealed that streptomycin-loaded niosomes reduced significantly the preformed biofilm in comparison to the free streptomycin (Figure 7). Previously, it has been reported that niosomal vesicles, due to their cationicity, interact electrostatically with the negatively charged biofilms; the drug can be released into the biofilm structure. Thus, niosomes are excellent carriers for delivery of antimicrobial drugs for eradication of biofilms. Previously, Kashef et al. studied the anti-biofilm effects of ciprofloxacin-containing niosomes against S. aureus biofilm and showed that niosome encapsulation reduced the minimum biofilm eradication concentration of ciprofloxacin by 2- to 4-fold compared to free ciprofloxacin (Kashef et al., 2020). In this work, even with 4- to 8-fold lower MIC of the free streptomycin (Figure 5), encapsulation allowed more efficient removal of the biofilm (Figure 7). This result once more demonstrates the power of niosomes.

The free and encapsulated streptomycin was evaluated for their cytotoxicity toward the HFF using the MTT assay. The cells exposed to medium only were used as control, and their viability was set to 100%. The cytotoxic cutoff was set as 70% of the viable cells in the control. It can be noticed that the niosome-encapsulated streptomycin had low toxicity (cell viability above 70%) to the HFF cells within the tested concentrations up to 1,300 μg/ml after 24 h, whereas free streptomycin exhibited toxicity with a concentration at 1,300 μg/ml (Figure 8). The lower toxicity of the niosome-encapsulated streptomycin than the free drug may be due to the use of surfactants (Span and Tween) in the niosomes, which are highly biocompatible (Marianecci et al., 2010; Shaker et al., 2015). The lower toxicity can also be caused by the lower release of streptomycin from niosome than the free form (Figure 4): after 24-h interaction, only about half of streptomycin was released from niosomes compared with that from the free form. Consequently, slightly higher cytotoxicity was observed for the free form than the niosome form (Figure 8), with about 70% cell viability for the former and 80% for the latter at the highest tested concentration. The low toxicity of niosomal carriers can be an ideal proposition in clinical applications.