Transient Transfection of the Respiratory Epithelium with Gamma Interferon for Host-Directed Therapy in Pulmonary Tuberculosis

Abstract

Nebulized gamma interferon (IFN-γ) protein has been studied for clinical safety and efficacy against pulmonary tuberculosis (TB). The protein is expensive, requires a cold chain, and is difficult to deploy in limited-resource, high-incidence settings. We generated a preclinical proof of concept (PoC) for a dry powder inhalation (DPI) containing DNA constructs to transiently transfect the lung and airway epithelium of mice with murine IFN-γ. Bacterial colony-forming units (CFU) in the lungs of mice infected with Mycobacterium tuberculosis (Mtb) reduced from about 10⁶/g of tissue to ~10⁴ after four doses given once a week. Nodular inflammatory lesions in the lungs reduced significantly in number. Immunohistochemistry of infected lung sections for LC3-1 and LAMP-1 indicated autophagy induction between 18 and 48 h after inhalation. ELISA on bronchoalveolar lavage (BAL) fluid showed differences in kinetics of IFN-γ concentrations in the epithelial lining fluid of healthy versus infected mice. Uninfected mice receiving DNA constructs expressing a fluorescent protein were live-imaged. The fluorescence signals from the intracellular protein peaked at about 36 h after inhalation and declined by 48 h. These results establish preclinical PoC of the efficacy of a DPI and dosing regimen as a host-directed and transient gene therapy of experimental pulmonary TB in mice, justifying preclinical development.

Keywords: dry powder inhalation; gamma interferon; gene delivery; gene therapy; host-directed therapy; preclinical proof of concept; pulmonary tuberculosis.

Graphical abstract

Figure 1

Efficacy of Inhaled Ifng against Experimental TB in Mice (A) Mice were infected with Mtb by whole-body exposure in a Glas-Col chamber on day 0 and treated once a week starting at day 28. Two untreated control animals were culled on days 0, 28, and 56. Two treated animals were culled on days 28, 35, 42, and 56. Lungs and spleens were harvested as indicated by arrows. (B) Dashed line connecting open circles (○) show average ± standard error of mean (n=2) of log₁₀–transformed values of CFU/g of lung, and open squares (□) show these values in respect of the spleens. Filled symbols connected by solid lines starting from day 28 indicate CFU counts in these tissues in treated animals.

Figure 2

Lung Pathology (A–E) Photographs (scale graduations, mm) of freshly harvested lungs after the first (A), second (B), third (C), and fourth (D) doses show progressive reduction in nodular lesions, while the lungs of mice receiving no treatment showed a large number of lesions (E). Photographs were subjected to image analysis using ImageJ software. (F) The numbers of nodular lesions (scatter points) reduced progressively, and their dimensions were more diffuse in animals receiving inhalations. Scatter points refer to individual lesions. The average number of lesions at each time point/group is shown as a line, with error bars showing standard deviations. Statistically significant differences (t test, α = 0.05, actual power 0.67) are shown by asterisks. NS, not significant.

Figure 3

Induction of Autophagy (A) Confocal micrographs of representative lung sections of mice that were infected 21 days prior to the experiment but received no treatment, Untreated control stained with DAPI to reveal nuclei, and with antibodies to LC3-1 and LAMP-1 conjugated with green and red fluorescent dyes, respectively. (B) Six hours after inhalation of the plasmid bearing Ifng. (C) Punctate fluorescence (arrows) was observed to initiate at 18 h after the dose due to sporadic co-localization of the two markers. (D) Extensive co-localization of the two markers indicative of autophagic flux was observed at 48 h after inhalation. Scale bars, 75 μm.

Figure 4

Transient Transfection In Vitro (A) IFN-γ secreted in culture supernatant by 3 × 10⁵ A549 alveolar epithelial cells transfected with about 1 ng (filled symbols) or 1.5 ng (open symbols) of plasmid DNA. Means ± SD of three replicates are plotted. (B) A chimeric GFP-IFN-γ protein was not secreted, but it could be detected by flow cytometry. Means ± SD of three replicates are plotted. (C) The green fluorescence attributable to the chimeric protein co-localized with LysoTracker red at 48 h post-transfection to generate yellow fluorescence. (D) Detail of (C).

Figure 5

Pharmacokinetics of IFN-γ in Uninfected and Infected Mice (A–C) Scatter points show values from individual animals (A–C, uninfected mice, n = 3; D, infected mice, n = 2); points with error bars indicate mean ± SEM, and solid lines depict model-fitted values of concentrations of IFN-γ as determined by ELISA in BAL fluid (A), lung homogenate (B), and blood plasma (C) of uninfected mice. (D) The concentration-time profile in the BAL fluid of infected mice showing drastic shift in C_max and t_max (note the logarithmic scale of the y axis; see also Table S1).

Figure 6

In Vivo Imaging. Mice were imaged at different time intervals after inhalation of a plasmid bearing RFP under the CMV promoter (A–E) 6, 12, 18, 24, and 48 h after inhalation. (F) The animal that received control particles without DNA at 6 h. (G) One animal showed signal in the nares and buccal regions and only in the left lung at the 12 h time point (corresponding to the animal shown in B). (H) Control animal, the same as shown in (F), at 6 h.