3.1. Physicochemical Characterization
The aim of this work was to develop a thermosensitive nanodevice for localized antibiotic delivery. Unspecific antibiotic delivery has, indeed, important side effects, such as sub-lethal concentrations at the site of infection, a lack of targeting, causing the disruption of intestinal flora, and the appearance of multidrug resistance. The encapsulation of antibacterial agents in niosomes has been widely pursued in recent years to improve their stability and biocompatibility, prolong their release and enhance their antibacterial efficacy [
27].
Niosomes are formed through the self-assembly of non-ionic surfactants in water, creating closed bilayers that can encapsulate both lipophilic and hydrophilic substances within the bilayer membrane or the aqueous layer, respectively. This assembly process depends on the hydrophilic–lipophilic balance (HLB), critical packing parameter values, and gel–liquid transition temperature of nonionic surfactant, and typically requires energy input, such as physical agitation or heat. Compared to liposomes, niosomes are more stable, and the presence of non-ionic surfactants allows for their prolonged circulation, potentially enhancing therapeutic efficacy and targeted delivery [
28]. Additionally, the phospholipids used in liposome manufacturing are both expensive and prone to degradation. These factors make niosomes a more stable and cost-effective option for drug delivery systems [
29].
Considering the recent success of stimuli-responsive nanocarriers to improve antibiotic efficacy, we focused on the design of thermoresponsive niosomes for the control of TC release. For this goal, a eutectic mixture of lauric and stearic acid is used as a thermoresponsive gatekeeper of the vesicle bilayer. The choice of this type of fatty acid was mainly due to the fact that, when combined at a weight ratio of 4:1, it has a well-defined melting point at 39 °C.
Niosomes were prepared via a film hydration approach by changing the molar ratios of non-ionic surfactant, CHOL and PCM (
Table 2) in accordance with the goal of obtaining a stable formulation at physiological temperatures and a higher release in hyperthermic conditions.
When visually observed, the niosomes appeared as milky and turbid dispersions, which are the typical aspect of vesicular formulations. The physicochemical properties of vesicles are reported in
Table 2. The mean particle sizes ranged between 244.9 and 348.3 nm, with a P.I. lower of 0.3 indicating a homogeneous and narrow size distribution (
Figure 1).
The increase in cholesterol content resulted in a decrease in vesicle size, since surface free energy usually decreases with increasing hydrophobicity, and therefore resulted in smaller vesicles [
30]. A similar trend was also observed with the increase in PCM content in vesicle bilayers, which further raises nanocarrier hydrophobicity. The formulations developed presented a good ability to encapsulate TC in an aqueous core, as suggested by the high values of EE% that ranged between 53.44 and 81.29 (
Table 2).
The drug loading led to an important variation in particle size: a significant increase in the hydrodynamic diameter was indeed observed (
Table 3). This trend is already reported in the literature and is due to the fact that TC has a negative charge and, when it was located in the aqueous compartment, was free to move and cause an electrical repulsion to each other, leading to an increase in vesicle size [
31,
32].
Despite its non-ionic character, the SP60 niosomes showed a large negative Z-potential value. The incorporation of the fatty acid mixture gave rise to lower Z-potential values due to the negative charge of these compounds. Additionally, all formulations exhibited large negative Z-potential values, which help to increase the stability of these formulations due to electrostatic interactions. The TC-loaded niosomes (
Table 3) had higher values, but they were still negative enough to maintain their stability. All these niosomes (free and TC-loaded) presented low polidispersion indexes (0.175–0.281), which can also contribute to enhancing the stability of these aggregates.
The size and morphology of the SP60CHPCM2 niosomes have also been studied using TEM. These niosomes showed the best in vitro cumulative release profile of TC (see the next section) because it was chosen to carry out these studies. As shown in
Figure 2, the morphology of SP60CHPCM2 niosomes results is spherical (
Figure 2A), and the shape was homogeneous throughout the whole population (
Figure 2B,C). Moreover, the sizes were concordant with that evaluated by DLS.
Next, we used differential scanning calorimetry (DSC) to evaluate the response of the developed formulation to temperature, and the melting profiles for all the samples are shown in
Figure 3.
Figure 3A shows the DSC profile corresponding to the pure fatty acid and its mixture. As expected, the melting point of the PCM is lower than that corresponding to the two components. The melting point of the SPAN60 and SPAN60CH was around 55 °C, and the introduction of the PCM in the vesicle bilayer of these niosomes conferred in them a melting point of around 40 °C (
Figure 3B,C).
3.2. In Vitro Release Studies
The thermoresponsive properties of the developed nanodevices were investigated by performing drug release experiments at two different temperatures, corresponding to physiological (37 °C) and hyperthermic temperature (42 °C). Ideal drug delivery systems for antibiotic delivery should ensure no premature drug leakage during blood circulation, but quick release at the target sites.
Preliminary studies were conducted to achieve the best surfactant/PCM ratio that produces a formulation with suitable thermoresponsive properties. The inclusion of PCMs in vesicles prepared without cholesterol conferred to the systems a thermoresponsiveness dependent on the amount of eutectic mixture added in the bilayer. As shown in
Figure S1, SP60PCM2TC, with the higher amount of PCM, achieved a total release of the entrapped drug after only 4 h at hyperthermic temperatures, while SP60PCM released 93% of the TC after 7 h. The release observed at 37 °C for these formulations was slower, at about 60% at 24 h for both systems. These results confirm our hypothesis that the inclusion of a eutectic mixture in a vesicle bilayer with a melting point of 39 °C allows us to achieve a control of release dependent on temperature. Additionally, a significant leakage of TC at 37 °C was already observed, making these devices less efficient and increasing the risk of systemic toxicity. Therefore, we decided to incorporate cholesterol in the bilayer, because of its ability to increase niosomes’ stability and decrease the leakage of encapsulated content in the bloodstream [
33,
34].
Figure 4 shows the TC release experiments corresponding to the niosomes containing cholesterol. We investigated if the cholesterol incorporation affects the thermoresponsive properties of the developed niosomes. As shown in
Figure 4, the TC release profile from SP60CH vesicles shows only a little increase with the temperature change from 37 to 42 °C. The cumulative release of TC is, respectively, 45.21 and 59.72% at 37 °C and 42 °C after 24 h. On the contrary, a significant increase in release was observed for vesicles that contain a PCM mixture in their bilayer. SP60CHPCM and SP60CHPCM2 achieved a release equal to 65.40 and 54.04% in 24 h at physiological temperatures. When the temperature increased to 42 °C, a fast and rapid release occurred, resulting in the release of the total amount of drug encapsulated in only 7 and 2 h, respectively, for SP60CHPCM and SP60CHPCM2. Consequently, a clear temperature-dependent release pattern was observed, highlighting the nanocarrier’s stability at physiological temperatures and the increased release ability at phase-transition temperatures, enhancing the local drug concentration at the infection site. So, cholesterol incorporation in the bilayer did not influence the temperature-triggered release properties of the developed vesicle, but limited the TC release under physiological conditions. The best release profile was observed for SP60PCM2 that was more stable at 37 °C and released the drug rapidly when the temperature was higher than the phase-transition temperature. Because of the obtained results, stability studies and antimicrobial activity were conducted only considering SP60PCM2 vesicles.
The results of TC release at the two different temperatures were then fitted into several kinetics models and linear forms for understanding the mechanism of drug release. We summarized all the kinetic parameters obtained in
Table S1, and the results demonstrated that only the Peppas–Sahlin and Weibull models show acceptable fitting results. Moreover, it was found that TC release from non-thermoresponsive systems (SP60 and SP60CH) was best explained by the Peppas–Sahlin model. The values of Adj. R-Square, equal to 0.99647, and the residual sum of squares, equal to 0.00176, were, respectively, higher and smaller than those of the other models, as described in
Table S1. Conversely, the pattern of release from vesicles that present in their bilayer the thermoresponsive PCM mixture at 42 °C, in addition to the Peppas–Sahlin model, were also best fitted by the Weibull model, as suggest by the high value of Adj. R-Square and the very low residual sum of squares data.
3.3. Stability Studies
The stability of niosomes is important for assessing their potential applications. In fact, one of the many problems with liposomes is that they are not stable aggregates, given the low chemical stability of phospholipids [
35]. Stability studies of the developed vesicles stored at room temperature were carried out for 3 months and highlighted that the vesicles were stable at room temperature, and no sedimentation, creaming or flocculation was observed. To gain more detailed information, the changes in particle size, PDI values and zeta potential were monitored. As reported in
Table 4, only small variations in particle size and EE% throughout 3 months of storage occurred, suggesting that these formulations are suitable for drug delivery applications
The stability studies of niosome formulations were also assessed in various medium simulant biological conditions (
Figure 5,
Figures S2 and S3), since complex mixtures with a certain pH, ionic strength and often the presence of organic matter can influence the stability and, thereby, alter the efficacy.
No significant alterations in dimensions and ζ-potential within 120 h were observed in the different conditions tested. However, niosome diameter in PBS and NaCl 0.9% was slightly larger than that in water, which can be ascribed to the salt absorption on the particle surface. So, it is possible to conclude that niosomes are stable in physiological conditions and the culture medium used for antibacterial assays. This suggests that the system is stable, and is therefore suitable for antibacterial applications.
3.4. Antimicrobial Activity
The antimicrobial activity (MIC and MBC) of empty and TC-loaded niosomes were evaluated at two different temperatures (37 °C and 42 °C) against some representative Gram-positive and Gram-negative bacteria. This study was carried out to investigate the antimicrobial performance of the developed nanodevices and to identify if there would be some enhancement of antibacterial efficacy with temperature.
The experiments were carried out only on SP60CHPCM2, which, as reported above, presents better thermoresponsive properties, and the results are shown in
Table 5 and
Table 6. For comparison purposes, the MIC and MBC values of free TC and empty niosomes were also measured.
At the higher concentration tested, 2970 μM, the empty niosomes did not show antimicrobial activity against any of the tested bacteria. Only a weak activity was observed at 42 °C against some of the Gram-positive microorganisms at very high concentrations, possibly due to the lauric acid presence in the vesicle bilayer, which has been reported to have antimicrobial properties [
36]. Notice that, at 42 °C, the fatty acid state has changed; this can give rise to a better bioavailability of this compound and consequently to a better antimicrobial activity. The experimental results in our study confirmed the antimicrobial activity of free TC against Gram-positive and Gram-negative bacteria with MIC values in the range of 0.6–30 μM, in accordance with that reported in the literature [
37,
38]. A very big MIC value against
S. epidermidis was recorded, which agrees with previous studies that observed that, compared with other antibiotics, tetracyclines exhibited low antimicrobial effectivity against different
S. epidermidis strains [
39]. In fact, Aelenei et al. [
40] reported that
S. epidermidis ATCC 12228, the strain used in this work, is resistant to this antibiotic. The MBC values of TC are much higher than the MICs. The mode of action of tetracycline implies a binding to the bacterial ribosome, and thereby interference with protein translation. The result of this type of interaction is generally bacteriostatic rather than bactericidal [
41]. At 37 °C, it was observed that TC encapsulated in niosomes maintained its antimicrobial efficiency against the tested bacteria. Even a slightly better inhibition effect was observed against BS and SE strains, with an MIC reduction from 1.62 to 0.77 and from 207.94 to 112.81, respectively (
Table 5). Instead, a significant decrease in MBC values was observed for TC-loaded niosomes, unlike the ones obtained with free drugs against all of the tested bacteria, except for LM, indicating a good ability to completely kill bacteria (
Table 6). For example, the MBC value of TC-loaded niosomes at 37 °C against
S. aureus (7.05) was considerably lower than that of the free drug, equal to 103.97 μM. This enhanced antibacterial activity may be due to a change in the mode of action of the encapsulated TC against bacteria due to the better permeability and fusogenic properties of niosomes and/or the interaction of the other niosome components. Indeed, it has been demonstrated that the encapsulation of antibiotics in vesicles can improve their pharmacokinetic profiles and increase their accumulation at the infection site, minimizing their cytotoxicity and protecting them from peripheral degradation [
42].
Moreover, the developed niosomes showed an interesting temperature-dependent antibacterial activity against some bacterial strains. In fact, the temperature increase to 42 °C further improves the activity of the niosomes against S. Epidermis, with a lower MIC value equal to 14.10 μM, unlike that obtained at 37 °C, which was equal to 112.81 μM. A moderate temperature-dependent trend in antibacterial activity was also observed for EC, EF and AB strains.
For example, MIC values for E. faecalis, E. coli and A. baumannii at 37 °C were 28.20, 7.05 and 7.05, and were reduced to 14.01, 3.52 and 3.52, respectively, by temperature increase. It is noticeable that in using this niosomial formulation, one resistant bacteria has become a sensitive one. This shows that through this type of technology, it is possible to reuse antibiotics that were no longer effective against certain resistant bacteria; so, this could be an interesting approach to combat the emergence of resistant bacteria. This shows again that the encapsulation of TC in thermoresponsive niosomes also permitted the antimicrobial activity of the drug to be exerted at a lower concentration than that of the free drug, without affecting the treatment outcomes and reducing the side effects.
The experimental results demonstrate that the inclusion of a eutectic mixture in the vesicle bilayer not only affects drug release but also affects antibacterial activity. The enhanced activity at 42 °C of the TC-loaded niosomes for some of the tested bacteria can be ascribed to a better ability to release their content at temperatures above the melting point of the eutectic mixture, increasing drug concentrations at the infection site and offering a better killing effect. Taken together, encapsulating the antibiotic into a thermoresponsive nanocarrier provides good biosafety, prevents premature release, allows for a more precise release of antibiotics and eliminates bacteria more effectively, which makes these devices appealing for further development into a clinical agent to prevent serious bacterial infections.
In this regard, the literature contains a lot of formulations containing antibiotics encapsulated in liposomial formulations that improve the antimicrobial activity of the drug [
43]. In this respect, a TC-liposomial formulation has been already approved by the FDA to treat the
Chlamydia trachomatis [
44]. However, liposomes show certain disadvantages: low solubility, a short half-life, low chemical phospholipid stability and high production cost. Gentamicin-loaded niosomes were also prepared using a non-ionic surfactant (Tween 60, tween 80 or Brij 35), cholesterol and a negative charge inducer, dicetyl phosphate. Niosomes composed of Tween 60, cholesterol and dicetyl phosphate were the most effective in terms of prolonged in vitro drug release [
45].
In this work, we have prepared, for the first time, a stable thermosensitive tetracycline-niosomial system using economical starting materials. It has been assessed that the release of the antibiotic at 42 °C is higher than that at 37 °C, which could improve the effectivity of antibiotics against some resistant bacteria.
To evaluate the biosafety of the newly developed nanocarriers (empty and TC-loaded), a hemolysis assay was performed on human erythrocytes. It was found that both empty and TC-loaded vesicles were hemocompatible, causing a negligible release of hemoglobin from erythrocytes, even at the highest concentration tested. We observed only a slight hemolytic activity, equal to 11.37 ± 1.99% and 19.97 ± 1.99% of empty and drug-loaded systems, at the concentration of 119 μg/mL of LA. These values indicate that the therapeutic index of these systems (hemolysis/MIC ratio) is very high, which indicates the high biocompatibility of these formulations. The non-toxicity of free and TC-loaded niosomes makes these new antimicrobial systems highly desirable as safe nanocarriers for biomedical applications.