Assessment of Innovative Dry Powders for Inhalation of a Synergistic Combination Against Mycobacterium tuberculosis in Infected Macrophages and Mice

Abstract

Background/Objectives: In vitro, vancomycin (VAN) and tetrahydrolipstatin (THL) together have been shown to synergistically inhibit Mycobacterium tuberculosis (Mtb), the world's most infectious killer. The poor oral bioavailability of VAN and THL and predominant tropism of Mtb infection to the lungs and alveolar macrophages make pulmonary administration highly attractive. This study aimed to develop and assess the efficacy of dry powders for inhalation of VAN microparticles embedded with THL. Methods: The dry powders produced by spray-drying, with or without hydrogenated castor oil (HCO), were characterized for their physicochemical properties among others by HPLC-DAD. The fast-screening impactor was used to determine powder aerodynamic properties, and VAN and THL releases were established from the paddle over disk method. Biological activities were assessed in a new M. bovis-infected macrophage model and in Mtb-infected mice. Results and Discussion: The addition of 25% HCO enables co-deposition (fine particle dose) at the desired weight ratio and co-releasing of VAN and THL in aqueous media. Microparticles with 0% to 50% HCO drastically reduced cytoplasmic Mycobacterium bovis survival (99.9% to 62.5%, respectively), with higher efficacy at low HCO concentration. Consequently, VAN/THL with or without 25% HCO was evaluated in Mtb-infected mice. Although no decrease in Mtb lung burden was observed after two weeks of administration, the endotracheal administration of VAN 500 mg/kg and THL 50 mg/kg with 25% HCO administrated three times during five days concomitantly with daily oral rifampicin (10 mg/kg) demonstrated 2-fold bacterial burden reduction compared to the group treated with RIF alone. Conclusions: HCO was crucial for obtaining a fine particle dose at the synergistic weight ratio (VAN/THL 10:1) and for releasing both drugs in aqueous media. With oral administration of the first-line rifampicin, the dry powder VAN/THL/25% HCO was able to exert a potential anti-tubercular effect in vivo in Mtb-infected mice after five days.

Keywords: macrophage infection model; orlistat; pulmonary delivery; tuberculosis; vancomycin.

Figure 1

THL (on left) and VAN (on right) deposition (in % related to the nominal dose in the capsules using 10 capsules filled with 20 mg of dry powder) in the different parts of the device (capsule, DPI, adapter) and FSI (induction port, pre-separator, and filter). A focus was made about VAN and THL deposition in the filter representing the fine particle fraction (d_ae inferior to 5 µm) related to nominal dose. All results are expressed as means of VAN or THL content ± SD (n = 3). Results were very significant for p < 0.01 (**); extremely significant for p < 0.001 (***) and extremely significant with p < 0.0001 (****).

Figure 2

Release of VAN and THL at 37 ± 0.5 °C from inhalable particles of dry powders for inhalation impacted on the FSI filter and recovered with a filter (i.e., the assembly), in a dissolution medium of 500 mL phosphate-buffer saline (PBS) containing 3% of sodium dodecyl sulfate (SDS). All results are expressed as mean percentage of VAN or THL release ± SD (n = 3).

Figure 3

Impact of dry powder treatments on bacterial burden in macrophages infected with the BCG::ESX-1^Mmar strain. Treatment was performed for (A) 24 h or (B) 4 days. All results are expressed as means of CFU/mL ± SD (n = 3). The statistical analyses were performed versus the negative control group (extremely significant for p < 0.0001 (****)).

Figure 4

In vivo efficiency assay of blends based on (A) VAN/THL/25% HCO (VAN 500 mg/kg and THL 50 mg/kg) (n = 20) or HCO (n = 16) and (B) VAN/THL (VAN 500 mg/kg and THL 50 mg/kg) (n = 5), administered by endotracheal route under anesthesia 3 times a week during 2 consecutive weeks. All results are expressed as means of RLU/mL ± SD (n = 5–20). The p-values are higher than 0.05 (non-significant), in comparison to untreated infected mice (CRT = 14 and 8, for A and B) (Table S10).

Figure 5

In vivo efficacy assay of blends based on (A) VAN/THL/25% HCO (VAN 500 mg/kg and THL 50 mg/kg) three times a week in association with oral RIF at 10 mg/kg five times a week after 1 week of treatment without nutritive supplementation (n = 12) and (B) VAN/THL (VAN 500 mg/kg and THL 50 mg/kg) three times a week in association with oral RIF at 7.5 mg/kg five times a week after 2 weeks of treatment with nutritive supplementation (n = 10). All results are expressed as means of RLU/mL ± SD (n = 10–14). The statistical analyses were performed versus the oral RIF control group, at 10 mg/kg (n = 12), panel A, or at 7.5 mg/kg (n = 10), panel B (significant for p < 0.05 (*)).