Introduction

Tuberculosis (TB), a common respiratory infectious disease, presents serious threats to public health. Each year, TB infects 2.5–3.2 million population worldwide which results in about 3 million deaths (1,2). Accounting for more than 80% of all TB cases, pulmonary TB is characterized by the participation of alveolar macrophages (AMs) harboring a large number of M. tuberculosis (3). Intracellular parasites such as M. tuberculosis are taken up by AMs via phagocytosis and secrete molecules which could protect them from phagosome-lysosome fusion (4). AMs thus become incubators for TB which resists digestion by lysosomal enzyme (5–7). Mycobacterium tuberculosis (MTB), hidden in AMs, often stages a comeback and leads to secondary tuberculosis (5).

To produce a significant antibacterial effect against MTB, effective concentration of antibiotics, higher than the minimum inhibitory concentration, is often required (8). Clinically, antibiotics are generally administered by the oral and parenteral route for the treatment of TB. Accordingly, antibiotics mostly likely distribute nonspecifically to tissues or organs via the systemic circulation, undesirable effects are thus frequently induced (9–11). Moreover, the amount of antibiotics that accumulate in AMs is insufficient to efficiently inhibit the delitescent MTB. Hence, it is urgent to develop a new delivery system of antibiotics for improving the therapeutic efficacy while minimizing side effects.

Recently, cationic drug delivery vectors, such as cationic polymer, proteins and lipids, have been widely used in gene delivery and tumor specific drug delivery (12). Cationic vectors were shown to interact with negatively charged complexes and serum components via “opsonization effect”, which results in a rapid elimination from the blood circulation (13). Nevertheless, this effect was confirmed as an alternative strategy for lung targeting, which is based on the filtering effect of capillary beds (14–16). Microparticles with sizes larger than the pore size of pulmonary capillaries may achieve selective accumulation at the lung, thus providing a unique opportunity to treat lung disorders (17,18).

To treat the intracellular MTB in AMs, antibiotics loaded nanoparticle systems targeting to AMs have been extensively studied in the past years (8,9,19–21). The sugar modified nanoscale vectors appeared to be a promising platform system in the targeted delivery of antibiotics to AMs (8). Mannose receptors were demonstrated to be specifically expressed on the AM surfaces, which have high affinity with mannose terminal molecules (22). In addition, the study on the interaction between AMs and different sugar-modified derivatives found that α-d-mannose derivatives can be rapidly adsorbed by AMs (22). Therefore, using α-D-mannosylated NLCs as carriers for antibiotics may increase the cellular uptake by AMs.

In the study, we developed a mannosylated nanostructured lipid carrier system for specific delivery of antibiotics to AMs for TB treatment. Rifampicin, a first-line antibiotics for TB, was loaded in mannosylatged NLCs and unmodified NLCs. The biodistribution profiles of both formulations were investigated to evaluate the lung targeting efficiency. Cell specific targeting was further confirmed by intracellular uptake in NR8383 cells in vitro and AMs via intravenous administration. The toxicity and inflammatory response of REP-Man-NLCs were further assessed in vitro and in vivo.

Materials and Methods

Rifampicin (RFP) was obtained from HongJing Chemical Co., Ltd (Hubei, China). Purified ovolecithin (EPC) (purity 80%) was purchased from Lipoid GmbH (Ludwigshafen, Germany). Medium chain triglyceride (MCT) was purchased from Beiya Medical oil CO., Ltd (Tieling, China). Octadecylamine (OCT) was offered by Shengke (Shengke Chemical Co, Ltd, Jining, China). Dimethyldioctadecylammonium bromide (DDAB) and mannan were obtained from Sigma-Aldrich (USA). Fluorescein isothiocyanate (FITC) labeled DHPE was offered by Invitrogen (Carlsbad, CA, USA). The 3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyltetrazolium bromide (MTT) and trypsin were obtained from Sigma Chemical Co. (St. Louis, MO). All other chemicals were of analytical reagent grade or better.

Cell Lines and Animals

NR8383 and L929 cell lines were purchased from Shanghai Institutes for Biological Sciences (SIBS; Shanghai, China). Wistar rats were provided by the Experimental Animal Center of Sichuan University (Chengdu, China). Institute of Cancer Research (ICR) mice were supplied by Chengdu Dossy biological technology Co., Ltd (Chengdu, China). All the animal experiments were approved by the Institutional Animal Care and Ethic Committee (People’s Republic of China).

Synthesis of Mannosylated Cholesterol Derivative

The mannosylated cholesterol derivative (Man-C6-Chol) was synthesized as previously described (23).

Preparation of RFP-NLCs and REP-Man-NLCs

RFP-NLCs and RFP-Man-NLCs were prepared by thin film homogenization method (24). RFP, EPC, MCT, OCT and DDAB (mass ratio: 2:10:6:1:10) were dissolved in chloroform. The organic solvent was completely evaporated at 40°C under vacuum to form the film, which was then hydrated in water solution. Subsequently, NLCs were obtained by ultrasonic probe at 200 W for 4 min. The mannose coating was carried out in the first step of preparation by adding Man-C6-Chol into the lipid matrix at a molar ratio of 5% (8,23). FITC-labeled RFP-NLCs and RFP-Man-NLCs were prepared similarly by adding 1% FITC-DHPE (w/w) into the formulation.

Characterization of RFP-NLCs and RFP-Man-NLCs

The particle size, polydispersity index (PDI) and zeta potential of both RFP-NLCs and RFP-Man-NLCs, were measured in water at 25°C by dynamic light scattering (Zetasizer Nano ZS90, Malvern Instruments Ltd, Worcestershire, UK). The morphology of both RFP-NLCs and RFP-Man-NLCs, freshly prepared and stained with 1% (w/v) phosphotungstic acid, was viewed and photographed using a transmission electron microscopy (TEM, H-600, Hitachi, Japan). The encapsulation efficiency of RFP-NLCs and RFP-Man-NLCs was determined using ultra filtration method (25). Nanosep® Centrifugal Filtration Devices (Mw cut-off 300 kDa; PALL Life Science, USA) was utilized to separate the free Rifampicin from the NLCs solution. The collected filtrate, i.e. unencapsulated rifampicin (Wu) was diluted with methanol by 100 times and quantified by an ultraviolet spectrophotometer at the wavelength of 474 nm. The equivalent total rifampicin in the nanoparticles (Wt) was measured by the same method. The encapsulation efficiency was calculated as follows: encapsulation efficiency = (1-Wu/Wt) × 100%.

Pretreatment and Analysis of Rifampicin in Plasma and Tissue

The plasma and tissue samples were treated as previously described (26). Samples (100 μL) and carbamazepine (8 μL, 1 mg/mL) were mixed and 150 μL acetonitrile was added for protein precipitation. After centrifugation at 11,000g for 5 min, 20 μL the supernatant was directly injected for HPLC analysis.

The prepared samples were determined by high performance liquid chromatography (HPLC) which consisted of a reverse-phase column (C18, 250 × 4.6 mm, 5 μm) with a security guard column (C18, 10 × 4 mm, 5 μm). The columns were kept at 35°C. The mobile phase consisted of methanol–water (67:33, v/v). The flow rate was maintained at 0.6 mL/min and the detection wavelength was set at 333 nm.

Using afore-mentioned HPLC method condition, there was no endogenous peaks interfering with RFP or carbamazepine. In addition, RFP peaks can be separated well from carbamazepine peaks without any interference. The linear standard curve showed good linearity over the concentration range of 0.1–12 μg/mL. This method then applied to the pharmacokinetic and biodistribution studies of RFP.

Pharmacokinetics in Rat Plasma

Wistar rats (male, 200 ± 20 g) were randomly divided into three groups (n = 5). After fasted but had free access to water over 12 h, the three groups were given the commercial drug solution (RFP Solution), RFP-NLCs and RFP-Man-NLCs via tail veins injection with a dose equivalent of 5 mg/kg rifampicin, respectively (27). 0.3 mL blood samples were collected at 5, 15, 30, 45, 60, 120, 240, 360, 480, 720 min post injection and centrifuged at 5000g for 10 min to obtain 100 μL plasma for HPLC analysis (9,26).

Biodistribution Studies in Mice

To evaluate the delivery efficiency of RFP-NLCs and RFP-NLCs in vivo, FITC-labeled RFP-NLCs and REP-Man-NLCs were intravenously injected into ICR mice. 30, 60, 120 and 240 min after injection, mice were immediately sacrificed for intact major organs collection. The fluorescence signals were recorded with in vivo imaging system (Quick View 3000, Bio-Real, Austria).

For quantitative determination of RFP, ICR mice (male, 20 ± 5 g) were randomly assigned into three groups (n = 5). The mice were fasted but watered for 12 h and were injected with RFP Solution, RFP-NLCs and RFP-Man-NLCs at a dose of 7.2 mg/kg, respectively (27). Blood and tissues, including brain, liver, lung, heart, spleen and kidney, were collected at 30, 60, 120 and 240 min post injection. Tissue samples, based on weight, were added with two volumes of 0.9% NaCl solution and homogenized. 100 μL plasma or tissue samples were collected for HPLC analysis (9,26).

Interaction of RFP-NLCs and REP-Man-NLCs with Serum (16)

RFP-NLCs and RFP-Man-NLCs were mixed with murine serum (1:5, v/v). Turbidity of the mixture were determined by the Varioskan Flash (Thermo Scientific, USA) at a wavelength of 600 nm at the indicated times. Serum alone served as blank control. For morphology evaluation of the aggregates, after mixing RFP-Man-NLCs with murine serum, samples were observed using TEM as described before.

In Vitro Uptake Study in NR8383 Cells

The cellular uptake study was carried out as previously reported (23,28,29). In brief, NR8383 cells were seeded on culture plates for 24 h in the complete medium and switched to serum free medium for 1 h before treated with FITC-labeled RFP-NLCs and RFP-Man-NLC. Then, 20 μL FITC-labeled RFP-NLCs and REP-Man-NLCs were added and incubated for 1 h at 37 or 4°C. One of the groups at the temperature of 37°C was preconditioned with 0.125 mg/mL mannan for 30 min prior to the addition of FITC-labeled RFP-NLCs and REP-Man-NLCs. After treatment, cells were washed three times with phosphate-buffered saline (PBS) and analyzed by flow cytometry using a Cytomics FC500 (Beckman Coulter, USA). In addition, compared with serum free medium, cell uptake in the complete medium at 37°C was also studied as described above.

In Vivo Uptake Experiments in Rats

Wistar rats (male, 200 ± 20 g) were randomly divided into three groups (n = 5). The animals in RFP-NLCs group and REP-Man-NLCs group were injected with FITC-labeled RFP-NLCs and REP-Man-NLCs, respectively. For the inhibition study, the animals in mannan group were intratracheally instilled with mannan (5 mg/kg) before the injection of FITC-labeled RFP-Man-NLCs(8). At 2 h, AMs were collected according to the protocol previously reported and subjected to flow cytometry analysis (8,30).

For qualitative analysis, the lungs were collected 2 h after administration. After fixed in paraformaldehyde (4%), cryosections of the lungs were prepared, and stained with DAPI for fluorescence imaging (Quick View 3000, Bio-Real, Austria). Besides, AMs from three treatment groups were collected as described above and processed for fluorescence imaging.

Cytotoxicity Assay

After seeded in 96-well culture plates for 24 h, L929 cells were treated with RFP-NLCs and RFP-Man-NLCs with various DDAB concentrations (n = 3) in 100 μL serum-free medium. 1 h later, samples were replaced with fresh complete medium, followed by a further incubation for 20 h. The cytotoxicity was assessed by MTT assay (16). Cell viability was calculated as the following expression: Cell viability = (Absorbance of cells exposed to RFP-NLCs or REP-Man-NLCs)/ (Absorbance of blank cells cultured without any drug) × 100%.

Inflammatory Response Assay

ICR mice (male, 20 ± 5 g) were randomly assigned into four groups (n =5). Three groups were injected with RFP Solution, RFP-NLCs and RFP-Man-NLCs at a dose equivalent of 7.2 mg/kg, respectively. Additionally, lipopolysaccharide (LPS, 30 μg per mouse) was used as a positive control. Serum were collected after 2 h and evaluated for the concentration of IL-6 using ELISA kit (R&D systems, USA) at 450 nm. For histological evaluation, lungs and livers were collected at 24 h after injection and fixed in 10% (v/v) formaldehyde. After embedded in paraffin, the samples were cut and stained with hematoxylin-eosin (H&E) for analysis.

Statistical Analysis

All experiments were repeated at least three times. Data were expressed as mean ± SD and analyzed by DAS 2.0 (Mathematical Pharmacology Professional Committee of China, Shanghai, China) for non-linear regression analysis. After the analysis of Student’s t-test of the two groups, a p value <0.05 was considered to be statistically significant.

Results

Characterization of RFP-NLCs and RFP-Man-NLCs

As shown in Table I, the size and zeta potential of RFP-NLCs and RFP-Man-NLCs showed no significant differences. NLCs displayed a size distribution around 160 nm (PDI <0.30) with positive charges of approximately 30 mV. The encapsulation efficiency of RFP was above 90%. Furthermore, TEM images of RFP-NLCs and RFP-Man-NLCs presented that the particles were round and uniform with particle size ranged from 100 nm to 200 nm, which was in accordance with the DLS measurement (Fig. 1).

Table I Characterization of RFP-NLCs and RFP-Man-NLCs. Data Represent Mean ± S.D. (n =3)
Full size table
Fig. 1
figure 1

Transmission electron micrograph images. A. RFP-NLCs; B. REP-Man-NLCs. Scale bar: 200 nm.

Full size image
Fig. 2
figure 2

Mean plasma concentration-time curve of RFP after intravenous injection of RFP Solution, RFP-NLCs and RFP-Man-NLCs in rats (equivalent to 5 mg/kg RFP). Data represent mean ± SD (n =5).

Full size image

Pharmacokinetics in Rat Plasma

The mean plasma concentration–time curves of RFP after intravenous administration of RFP Solution, RFP-NLCs and RFP-Man-NLCs to rats were shown in Fig. 2. The results indicated that RFP Solution, RFP-NLCs and RFP-Man-NLCs shared the same trend. The main pharmacokinetic parameters in rats did not exhibit statistically significant differences among three groups (Table II) (p >0.05), indicating that the three formulations were bioequivalent.

Fig. 3
figure 3

RFP-Man-NLCs mediated lung accumulation of RFP in vivo. (a) Representative in vivo fluorescence image of major organs of FITC-labeled RFP-NLCs and RFP-Man-NLCs from mice. (b) Dynamic changes in the turbidity of RFP-NLCs and RFP-Man-NLCs incubated with murine serum. (c) TEM images of the aggregate complex of RFP-Man-NL Cs interacted with serum protein. Scale bar: 200 nm (d). Tissue distribution in mice after intravenous injection of RFP Solution, RFP-NLCs and RFP-Man-NLCs in rats. Data represent mean ± S.D. (n =5). i.v. 7.2 mg/kg. *p < 0.05; ***p < 0.01.

Full size image
Table II Pharmacokinetic Parameters of RFP Solution, RFP-NLCs and RFP-Man-NLCs. Data Represent Mean ± S.D. (n = 5)
Full size table

Biodistribution Studies in Mice

To investigate the delivery efficiency of RFP-NLCs and RFP-Man-NLCs in vivo, the fluorescence of major organs was shown in Fig. 3a. FITC-labeled RFP-NLCs and REP-Man-NLCs were notably accumulated in the lung at the indicated time points. About 4 h after injection, the fluorescence intensity in the lung appeared to decrease and liver accumulation started to increase over time.

Fig. 4
figure 4

Cell uptake of RFP-Man-NLCs and RFP-NLCs in vitro and in vivo. (a) Binding and uptake of RFP-Man-NLCs and RFP-NLCs by NR8383 Cells after 1 h incubation at either 4°C or 37°C. (n = 3). (b) Flow cytometry analysis showing uptake of RFP-Man-NLCs and RFP-NLCs by alveolar macrophages after intravenous administration in rats (n = 5). (c) Confocal laser scanning microscopy images of lungfrozen sections after intravenous administration in rats. Scale bars represent 100 μm. Arrows indicate cell uptake by alveolar macrophages in vivo. (d) Confocal laser scanning microscopy images of isolated alveolar macrophages after intravenous administration in rats. Scale bars represent 20 μm. Green fluorescence as FITC-labeled RFP-Man-NLCs and RFP-NLCs and blue fluorescence as DAPI observed by confocal laser scanning microscope, as well as the merged image were shown. *p <0.05, ***p <0.01, RFP-Man-NLCs compared with RFP-NLCs. ###p <0.01, RFP-Man-NLCs compared with + mannan.

Full size image

Quantitative determination of RFP in major organs was shown in Fig. 3d. As for RFP-Man-NLCs, the concentration of RFP in the lung was 10.48 times at 30 min, 5.81 times at 60 min, 9.17 times at 120 min, and 3.34 times at 240 min higher than that of RFP solution. Moreover, the distribution in kidney decreased compared to that of RFP solution. On the contrary, the distribution in heart exhibited a slight increase. The concentration of RFP in liver, spleen and plasma showed no significant differences among the three groups.

The passive lung targeting of cationic vectors may be attributed to serum-induced aggregation. The turbidity of RFP-NLCs and RFP-Man-NLCs mixed with murine serum was time-dependent (Fig. 3b). The turbidity increased gradually and reached the maximum in 1 h, indicating aggregation happened upon mixing RFP-NLCs and RFP-Man-NLCs with murine serum. After 1 h, the decrease of the turbidity revealed a dissociation of RFP-NLCs and RFP-Man-NLCs with murine serum. In addition, REP-Man-NLCs and murine serum formed complexes, which were typically observed by TEM. Figure 3c showed a loose structure of NLC-serum complexes with a size ranged from 1 to 2 μm, which may result in an efficient retention of RFP-Man-NLCs in the lung capillaries.

Cell Uptake of RFP-Man-NLCs and RFP-NLCs In Vitro and In Vivo

NR8383 AMs derived from bronchio-alveolar lavage of a normal male SD rats, are a continuous cell line (31). To evaluate the potency of targeted delivery of RFP-Man-NLCs to NR8383 cells, the cell internalization of fluorescently labeled NLCs was determined using flow cytometry. The quantified mean fluorescence intensity (MFI), indicated that intracellular uptake of RFP-Man-NLCs was significantly higher (2.6-fold) than that of RFP-NLCs (p <0.01) at 37°C (Fig. 4a). Intracellular uptake of RFP-Man-NLCs also dramatically exceeded that of RFP-NLCs in the complete medium. However, the introduction of mannan (0.125 mg/ mL) significantly decreased the intracellular uptake of RFP-Man-NLCs (p <0.01). At 4°C, cell binding efficiency of RFP-Man-NLCs and RFP-NLCs (p >0.05) showed no significant differences. Besides, the cellular uptake of RFP-Man-NLCs and RFP-NLCs was temperature-sensitive.

Fig. 5
figure 5

Relatively low toxicity and immunogenicity of RFP-NLCs and RFP-Man-NLCs. (a) MTT assay showed the cytotoxicity of RFP-NLCs and RFP-Man-NLCs. L929 cells were seeded in 96-well cell plate at a cell density of 1 × 105/well for 1 d before treated with RFP-NLCs and RFP-Man-NLCs with various DDAB concentration (n = 3) in serum free medium (100 μL). ***p <0.01 indicates the concentration of DDAB, up to 18 μg/mL, reflected significantly cytotoxity. (b) IL-6 concentration of mouse serum after i.v. injection of RFP-NLCs, RFP-Man-NLCs and RFP-Solution to ICR mice, serum was collected and tested for IL-6 level by ELISA method with Lipopolysaccharide (LPS) (30 μg per mouse) as positive control (n = 5). ***p <0.01, compared with LPS group. (c) Histological evaluation of lung and liver after i.v. injection of RFP-NLCs, RFP-Man-NLCs and RFP Solution (100×). Arrows indicate mild inflammatory cell infiltration in hepatic lobules.

Full size image

To verify the cell target ability of RFP-Man-NLCs in vivo, AMs were isolated at 2 h after intravenous administration in mice. Compared with RFP-NLCs, RFP-Man-NLCs significantly increased the uptake in AMs by 3 fold (Fig. 4b). Furthermore, the inhibition of mannose receptor (mannan) dramatically decreased the uptake of RFP-Man-NLCs by 5 fold. Cryosections of lungs (Fig. 4c) and isolated AMs (Fig. 4d) also confirmed fluorescence accumulation in AMs in the RFP-Man-NLCs treated group.

Cytotoxicity and Inflammatory Response

To evaluate the safety of RFP-Man-NLCs, L929 cells in 100 μL serum-free medium were seeded in 96-well plates and treated with RFP-NLCs and RFP-Man-NLCs with various DDAB concentrations. Obvious cytotoxicity was observed when the concentration of DDAB increased to 18 μg/mL, which was much higher than the concentration used in the in vitro uptake study (10 μg/mL) (Fig. 5a). Moreover, DDAB was shown to induce a significant cell death at a concentration of 27 μg/mL with a relative cell viability of less than 70% (Fig. 5a).

Fig. 6
figure 6

Structure model diagram of REP-Man-NLCs.

Full size image

To evaluate the inflammatory response, the serum concentration of IL-6 was used as the indicator for RFP-NLCs and RFP-Man-NLCs induced immune responses in vivo (Fig. 5b). Results showed no obvious upregulation of cytokines was found after i.v. injection of RFP-NLCs, RFP-Man-NLCs and RFP solution in ICR mice. On the contrary, lipopolysaccharide (LPS) strongly induced IL-6 expression (p <0.01). In addition, the histological evaluation of lung and liver sections from four experimental groups showed no inflammatory or pathological changes (Fig. 5c). According to IL-6 results, the histopathological damage caused by LPS was observed microscopically as indicated by mild inflammatory cell infiltration in hepatic lobules.

Discussion

Rifampicin, as the first choice drug, was widely used in the treatment of tuverculosis. To develop rifampicin delivery systems that could facilitate the selective shuttling of drugs to the site of tuberculosis infection, novel delivery devices have been characterized by researchers in recent years, such as liposomal aerosols, solid lipid nanoparticles, biodegradable polymeric nanoparticles (4,9,32,33). The present study reported the successful delivery of RFP loaded Man-cationic NLCs to AMs via intravascular administration for the first time. Mannosylated NLCs were fabricated using DDAB and octadecylamine. The fabrication of NLCs were shown to have following advantages such as the absence of organic solvents, large-scale production, low toxicity, improvement of drug incorporation and controlled release properties by adding spatially incompatible liquid lipids (34,35). Herein, the proposed schematic model is shown in Fig. 6. To achieve high rifampicin encapsulation efficiency, the addition of octadecylamine was necessary which contained a long fatty chain and an ammonium group acting as a bridge linking drugs with the acid group (36). Rifampicin had a phenolic hydroxyl group with weak acidity, which could interact with octadecylamine via ionic bonds. Thereby, rifampicin, with high lipophicity, could be encapsulated in NLC carriers with high efficiency. In addition, DDAB, as a commonly used cationic material, was used in fabricating NLCs to introduce cationic properties to the NLCs and to improve the distribution of drug in the lung. Mannosylated cholesterol (Man-C6-Chol), which could insert into the NLCs via cholesterol residues, was used to modify NLCs to accomplish cell-specific targeting. In brief, we prepared cationic NLCs with an original size distribution around 100 nm and a zeta potential of about 30 mV.

Interactions between cationic vectors and serum components were reported to be a complex process which might influence the in vivo fate of drug cargo (16). In this study, the aggregation behavior between cationic vectors and serum components occurred rapidly after adding RFP-Man-NLCs and RFP-NLCs to murine serums. 1 ~ 2 μm NLC-serum complexes with a hydrophilic protein shell were shown to be retained in pulmonary capillaries thus inducing a rapid accumulation of RFP in the lung tissue as verified by both fluorescence image and HPLC analysis. Generally, particles with size over 7 ~ 8 μm could be retained by lung capillary beds (37,38). In addition, particles with smaller sizes, e.g. 1 ~ 2 μm liposomes, were also proven to achieve lung accumulation (39,40). According to the “Vroman-effect” as previously observed on solid lipid nanoparticles, the protein adsorption and desorption is a dynamic process during which proteins with higher abundance and low affinity could be displaced by less abundant proteins with higher affinity (41,42). A dissociation of NLCs and serum complexes could be triggered by the higher affinity proteins contained in serum. After 1 h incubation, the decrease in the turbidity revealed the destabilization of complexes. Therefore, we proposed that once injected into mouse tail vain, RFP-Man-NLCs interacted with serum proteins thus generating large and loose complexes which resulted in the specific accumulation in the lung. Moreover, these complexes might undergo disintegration as triggered by higher affinity proteins. As a result, the ligand of RFP-Man-NLCs would be exposed thus achieving AMs targeting via mannose receptor mediated uptake (Fig. 4b, c and d). However, a small proportion of RFP was likely to be released in the blood stream that could be rapidly metabolized in liver and excreted in kidney (Fig. 3d). Also, cationic nano carriers were reported to be eliminated rapidly from blood circulation (13,43,44). For pharmacokinetic studies, no significant differences were observed in the pharmacokinetic behaviors among groups administrated with RFP-Man-NLCs, RFP-NLCs or RFP Solution. The in vivo distribution study of nano carriers by fluorescent labeling was proven an efficient and convenient method. FITC-PE used in this study represented NLCs but free RFP which was not encapsulated in the NLCs or released from the NLCs in the blood. Hence, FITC-labeled RFP-NLCs and RFP-Man-NLCs could be detected to stand for most of RFP capsulated in the nano carriers (16). RFP, released from the NLCs, was detected in most major organs according to the quantitative determination of RFP by HPLC. In addition, RFP-Man-NLCs and RFP-NLCs significantly improved the lung accumulation of RFP in comparison with RFP solution (Fig. 3d), while RFP solution displayed no tissue selectivity with most drugs accumulating in the liver which might induce severe liver toxicity (45). Consequently, lower doses of RFP-Man-NLCs and RFP-NLCs could be used to decrease nonspecific distribution in off-target organs including liver while achieving comparable level of accumulation at the target site compared to RFP solution.

AMs were found to be the incubators and hideout of Mycobacterium, and thus, it is important for delivery systems to selectively shuttle drugs to the site of infection in lung (9). In the in vitro uptake study, RFP-Man-NLCs exhibited specific targeting to NR8383 cells as expected. The cellular uptake of RFP-Man-NLCs that surpassed NLCs without mannosylation could be inhibited by the addition of free mannan. Also, a significantly higher level of cellular uptake of RFP-Man-NLCs at 37°C was observed as compared to 4°C, which indicated that the cellular uptake of RFP-Man-NLCs was energy-dependent endocytosis mediated by the interaction between mannose and corresponding receptors (28,29). In addition, cell uptake in the complete medium did not change the cell-specific targetability, which indicated that the interaction between cationic vectors and serum proteins did not affect the AM targetability of RFP-Man-NLCs. This result could explain the efficient targeting to AMs of RFP-Man-NLCs in vivo. A continuous absorption and dissociation of RFP-Man-NLCs and serum complexes may help expose Mannose-residuals which could primarily contact with AM surface receptors thus leading to an efficient cellular delivery.

The safety of RFP-Man-NLCs and RFP-NLCs were evaluated by cytotoxicity and inflammatory response. Despite the cytotoxicity of RFP-Man-NLCs and RFP-NLCs slightly increased along with the rising of the concentration of DDAB, RFP-Man-NLCs and RFP-NLCs were still low toxic vectors for rifampicin delivery. In our study, especially in cell uptake experiments, the concentration was within the low toxicity range. In addition, unlike traditional microspheres with solid structure, the aggregation complexes formed by interaction of RFP-Man-NLCs and RFP-NLCs with serum components were loose and dynamic possessing “Vroman effect”, as discussed above (41,42). Morphological assessments and tissue histology of lungs and livers revealed no inflammation and celluler damage induced by NLCs, comfirmed by Fig. 5c and d. These results revealed that RFP-Man-NLCs and RFP-NLCs accumulation occurred in the lung without inducing inflammatory response, thus suggesting that RFP-Man-NLCs and RFP-NLCs are not toxic under these conditions.

Conclusion

In summary, we designed and developed mannosylated cationic nanostructured lipid carriers for AM targeted delivery of rifampicin. When administered systemically, cationic RFP-Man-NLCs were demonstrated to interact with serum proteins and form NLC-protein complexes with significantly larger sizes which exhibited a primary lung accumulation. Furthermore, RFP-Man-NLCs exhibited AM targetability both in vitro and in vivo. The optimized RFP-Man-NLCs formulation was proven to have minimum cytotoxicity which was safe for systemic administration. Our results demonstrated that RFP-Man-NLCs could be a promising carrier for the treatment of delitescent TB.