2.5.1. Evaluation of Metabolic Activity
Free drugs (INH and RFB), raw material (LBG) as obtained commercially, unloaded LBG microparticles and drug-loaded LBG microparticles were incubated with cultures of the two cell lines. A shorter (3 h) and a more prolonged period of exposure (24 h) were tested, and three concentrations of materials were assessed (0.1, 0.5 and 1 mg/mL). Considering that the theoretical drug loading of microparticles was 10%, free drugs were evaluated at 10x lower concentrations comparing with the other materials (0.01, 0.05 and 0.1 mg/mL). In all cases samples were presented as solutions/suspensions prepared in pre-warmed cell culture medium (CCM). For the discussion of results, it was considered that a material has cytotoxic potential when cell viability after exposure to the material decreases below 70% (indicated with a dashed line in all figures), as designated by the ISO 10993-5 [39
The exposure of cells to the free drugs revealed two important aspects. INH did not show any detrimental effect on cell viability, which remained around 90%–100% in all cases, irrespective of the cell line, tested concentration and time of exposure (Figure S5
). On the contrary, time- and concentration-dependent effects were observed for RFB (p
< 0.05). After 3 h of exposure to this drug, cell viabilities remained above 80% (Figure S6
), but a strong decrease was observed after 24 h (Figure 5
). This decrease was particularly noticeable for the highest concentration tested (0.1 mg/mL). There was also a trend indicating lower viabilities obtained in A549 cells, suggesting higher sensitivity of this cell line comparing with differentiated THP-1 cells upon contact with the free drugs.
LBG and LBG-based microparticles were evaluated separately, in order to disclose an effect from the carrier structure [40
]. Testing LBG is not only a relevant control of the work, but also contributes to the state of the art, as no similar evaluation is, to our knowledge, described in the literature.
Again, while no overt toxicity was found in THP-1 cells (cell viability >85% in all conditions), A549 cells demonstrated to be more sensitive, showing 40% cell viability after 24 h of exposure independently of the concentration (Figure S7
). Interestingly, this detrimental effect completely reverted after spray-drying, as unloaded LBG microparticles exhibited a very mild effect even after 24 h (A549 cell viability >75% in all cases, Figure 6
The described effect of LBG as raw material after the contact with A549 cells could be attributed to either of two reasons: (i) the occurrence of a partial hydrolysis of commercial LBG in the acidic solution either prior to spray-drying or potentiated by the heat and shear forces of the process, leading to lower molecular weight polymer chains; or (ii) a slower solubilisation of the microparticles when compared to the commercial powder, which delays the increase of viscosity in the solution to be tested. This translates directly to an effect at the level of the viscosity of dispersions prepared at the same concentration using LBG commercial powder and unloaded LBG microparticles. As the assay is performed by incubating a dispersion of polymer/microparticles with cells, higher viscosity of the dispersion possibly makes gaseous exchanges between cells, medium and air a difficult task. This could lead to higher cell death. In fact, the dispersion obtained from LBG microparticles was perceived as more fluid than that prepared from commercial LBG polymer.
Concerning macrophage-like THP-1 cells, no relevant variation of cell viability was observed upon exposure to either a dispersion of commercial LBG (Figure S7
) or unloaded LBG microparticles (Figure 7
). This observation is again in line with a higher sensitivity of A549 cells to the materials, comparing with THP-1 cells.
The results of the cytotoxic evaluation of drug-loaded LBG microparticles are presented in Figure 6
and Figure 7
, representing the 24 h exposure of A549 and macrophage-differentiated THP-1 cells, respectively. The results obtained after a short contact time (3 h) are available as Supplementary Materials (Figures S8 and S9)
and show similar tendencies, although in some cases the determined cell viabilities were higher than those at 24 h, indicating a time-dependent effect.
The general trend indicates that both cell lines responded to the presence of drug-loaded formulations according to a similar pattern, that is, INH-loaded microparticles induced low effect on cell viability and RFB-loaded microparticles elicited considerable cytotoxicity. More specifically, INH-loaded microparticles induced cell viabilities above 70% in all tested conditions (exposure times, concentrations and cell lines) with the exception of the 24 h exposure of A549 cells to concentrations of 0.5 and 1 mg/mL, which resulted in viabilities of 66%–68% (Figure 6
). It was also observed an absence of concentration-dependent effect for this formulation, at 3 h and 24 h. The only exception occurs in A549 cells after 24 h exposure, where 0.1 mg/mL of INH-loaded microparticles induced 92% cell viability, contrasting with 66%–68% when testing 0.5 and 1 mg/mL (p
< 0.05). Worth mentioning is the fact that responses to INH-loaded microparticles were very similar to those generated by unloaded LBG microparticles in both cell lines, giving a clear indication on the absence of toxicity of INH itself. No overt cytotoxic effect was, thus, considered to occur for INH-loaded microparticles. These observations are in line with the literature, as the IC50
of INH is reported as 1000 mg/mL in alveolar macrophages (isolated from albino rats) [41
]. The referred study was performed in primary cells, thus different from those used in this study, but they mimic in vivo
conditions in a closer way and are supposedly more sensitive than established cell lines [38
The response to RFB-loaded microparticles contrasted well with the previous results. After a contact of 3 h, a reduction of cell viability in both lines to values around 40%–50% and even 20% (THP-1 cells), particularly for the concentrations of 0.5 and 1.0 mg/mL (Figures S8 and S9
), was observed. After 24 h, cell viabilities were generally not very different from those at 3 h, with the relevant exception corresponding to the highest concentration of microparticles (1 mg/mL), namely for LBG:RFB = 10:1 (w
) microparticles (Figure 6
and Figure 7
). In these conditions, cell viabilities of 10%–15% were determined (p
< 0.05). In A549 cells, a time-dependent effect was observed as a whole (p
< 0.05), although it was more pronounced for microparticles LBG:RFB 10:1 (w
). A significant difference was also perceived between the various RFB-loaded microparticles (p
< 0.05), in the order LBG:RFB = 10:1 > 10:0.5~10:0.2, reinforcing that the cytotoxic effect is due to RFB. THP-1 cells also revealed a time-dependent effect (p
< 0.05), which was particularly visible for microparticles LBG:RFB = 10:1 and 10:0.5 (w
) and for the two highest concentrations tested (0.5 and 1 mg/mL). A dose-dependent effect was also visible for RFB-loaded microparticles (p
< 0.05), which was more pronounced than that observed for A549 cells.
The above mentioned trend, indicating higher susceptibility of A549 cells when compared with THP-1 cells, was thus not followed when testing microparticles, where relatively similar responses are observed between both cell lines. The literature reports opposite effects, either demonstrating higher resistance of differentiated THP-1 cells [42
] or establishing lower susceptibility for A549 cells [44
], clearly indicating that the generated responses are strongly dependent on the assessed materials. What is clearly seen in our study is that there was a difference of susceptibilities when testing free drugs and polymer, which are exposed as solutions, comparing with microparticles. This different outcome is probably related with specific endocytic–exocytic mechanisms of phagocytic and non-phagocytic cells, along with the specialized physiological role of each particular cell type. THP-1 cells are phagocytes with significant endocytic and exocytic activity and natural ability to uptake particulate matter [43
]. Therefore, they possibly respond with higher intensity to the ingestion of particulates [44
]. On the contrary, epithelial cells possibly have more intimate contact with dissolved solutes than with the corresponding particulates.
2.5.2. Evaluation of Cell Membrane Integrity
As a complementary study, cell membrane integrity was evaluated upon exposure to the different materials. Taking into account the results obtained in the MTT assay, released LDH was determined after 24 h of exposure to the highest concentration tested, 1 mg/mL. For RFB-loaded microparticles, the concentration of 0.5 mg/mL was also tested. RFB was tested as free drug at the concentrations of 0.05 and 0.1 mg/mL while free INH was only tested at 0.1 mg/mL.
The results of free drugs are available as Supplementary Materials (Figure S10)
, being in agreement with those of the MTT assay for both cell lines and reinforcing the observations of RFB cytotoxicity. While the contact with INH did not increase significantly the release of LDH, RFB induced an increase to 150%–170%, which was particularly noticeable for the highest concentration of the drug (0.1 mg/mL). The assessment of the effect of LBG microparticles was also in line with MTT results. Figure 8
and Figure 9
show the observations performed in A549 and macrophage-like THP-1 cells, respectively. The first remarkable observation was related with the effect of LBG as raw material. The MTT assay had shown very different outcomes between the raw material and the unloaded microparticles. However, these differences did not appear at the level of LDH release, as the amount of released enzyme was similar in both cases, for both cell lines. Therefore, although cell death was observed after the MTT assay, it was not related with events at the level of membrane integrity. Additionally, it is important to mention that the amount of enzyme that was released was comparable or even lower than that observed for the control (cells incubated with culture medium).
As such, it was also verified that the amount of released LDH induced by the contact with INH-loaded microparticles was, in both cell lines, similar to that induced by unloaded microparticles, raw material and the control, again evidencing an absence of toxicity of INH.
The observations were very different for RFB-loaded microparticles, justifying the assessment of two concentrations of these microparticles. In A549 cells, the three RFB-loaded microparticles induced similar effect when tested at the lower concentration (0.5 mg/mL), with only a slight increase in LDH release to 110%–140%. Interestingly, the cells responded in a similar manner to a doubled concentration (1 mg/mL) of LBG:RFB 10:0.5 and 10:0.2 (w/w) microparticles, with no significant alterations in LDH release. However, LBG:RFB 10:1 (w/w) microparticles elicited a stronger increase in LDH release (p < 0.05) to approximately 400%, indicating a clear concentration-dependent effect. The exposure to the lysis buffer, indicating the highest LDH amount possibly released, induced 666%.
The trend was approximately similar in THP-1 cells. LBG:RFB 10:0.2 (w/w) microparticles induced 125%–150% LDH release at both concentrations tested. A clear concentration-dependent effect was observed for the formulation 10:0.5 (w/w), with LDH release increasing from 139% to 223% with the increase of concentration (p < 0.05). LBG:RFB 10:1 (w/w) microparticles induced similar release at both concentrations (220%–230%). The lysis buffer induced a value around 350%.
The general observation from the whole set of results of cytotoxicity is that an increased toxic effect is seen when RFB is included in the microparticles. The high in vivo
toxicity of RFB is well reported [45
]. Although the mechanism is still not well established, the cytotoxic behavior might be due in part to the lipophilic character of RFB. It presents a high membrane lipid tropism, resulting in high penetration into the cells, which can impose increased toxicity [46
Notwithstanding the determinations that were made in this study, there is the expectation that the toxicity does not translate to such a severe level in vivo
. This belief is based on the assumption of a relatively even distribution of the dry powder in the alveolar zone upon inhalation. The highest dose tested in the described assays (1 mg/mL) corresponds to 303.03 μg/cm2
. The area of the epithelial surface of alveolar zone is about 70 m2
]. If an even distribution of microparticles is assumed and it is considered that a nominal proportion (e.g., a third) of an inhaled dose (considering the approximate 160 mg of powder delivered by the TOBI®
Podhaler in one dose) deposits in the alveolar region, the dose estimated across this area is 0.08 μg/cm2
. This dose is remarkably lower than the 303.03 μg/cm2
used in our study. However, one should bear in mind that the lung of tuberculosis patients possibly has a much lower area, apart from certainly being variable among patients. Still, even if only 10% of the alveolar area is considered functional (7 m2
), the dose will be 0.76 μg/cm2
, very far from that used in this study. Taking this into account, the effects will be much closer to those of the lower dose of 0.1 mg/mL than to those of the higher dose or even the 0.5 mg/mL. Unfortunately, we could not meet the conditions permitting weighing such a low amount of dry powder that could resemble in a better way the in vivo
conditions. Another reason contributing to a possible decrease of in vivo
toxicological effects associated with the microparticles is their capacity to form complexes with polar heads of groups of the phospholipids present in the pulmonary surfactant, allowing reaching high concentrations without damaging the epithelium [48
Several other issues related to the toxicity need to be discussed and addressed experimentally in the near future in order to verify the real possibilities of using LBG microparticles for the proposed application. LBG has been clearly referred as biodegradable when administered orally, owing to the presence of β-mannosidase in the human intestine [49
]. Importantly, the enzyme has also been detected in the lung, although in lower concentration comparing with other organs [50
]. As long-term dosing is needed in the application focused in this work, it is very important to ensure the biodegradability of the microparticles and, although the presence of β-mannosidase is a promising indication, more studies are needed in this regard. In parallel, it is important to unveil the immunological effects of these microparticles, as LBG is a novel polymer in lung delivery and polysaccharides are particularly susceptible in this regard. We are currently performing in vivo
assays to verify the immunological response to the administration of LBG microparticles by inhalation.