Immunological and pathological comparative analysis between experimental latent tuberculous infection and progressive pulmonary tuberculosis

Abstract

Mycobacterium tuberculosis produces latent infection or progressive disease. Indeed, latent infection is more common since it occurs in one-third of the world's population. We showed previously, using human material with latent tuberculosis, that mycobacterial DNA can be detected by in situ PCR in a variety of cell types in histologically-normal lung. We therefore sought to establish an experimental model in which this phenomenon could be studied in detail. We report here the establishment of such a model in C57Bl/6 x DBA/2 F1 hybrid mice by the intratracheal injection of low numbers of virulent mycobacteria (4000). Latent infection was characterized by low and stable bacillary counts without death of animals. Histological and immunological study showed granulomas and small patches of alveolitis, with high expression of tumour necrosis factor alpha (TNFalpha), inducible nitiric oxide synthase (iNOS), interleukin 2 (IL-2) and interferon gamma (IFNgamma). In contrast, the intratracheal instillation of high numbers of bacteria (1 x 106) produced progressive disease. These animals started to die after 2 months of infection, with very high bacillary loads, massive pneumonia, falling expression of TNF-alpha and iNOS, and a mixed Th1/Th2 cytokine pattern. In situ PCR to detect mycobacterial DNA revealed that the most common positive cells in latently-infected mice were alveolar and interstitial macrophages located in tuberculous lesions, but, as in latently-infected human lung, positive signals were also seen in bronchial epithelium, endothelial cells and fibroblasts from histologically-normal areas. Our results suggest that latent tuberculosis is induced and maintained by a type 1 cytokine pattern plus TNFalpha, and that mycobacteria persist intracellularly in lung tissue with and without histological evidence of a local immune response.

Fig. 1

The effects of the dose of Mycobacterium tuberculosis used for intratracheal infection on survival, bacterial counts and pathology. (a) Survival curves after doses of (♦) 1 × 10⁶ and (□) 4 × 10³. (b) Colony-forming units/lung at intervals after infection with 10⁶ (♦) or 4 × 10³ (□) bacteria. (c) The percentage of the lung occupied by pneumonia or (d) alveolitis in mice infected with 10⁶ (♦) or 4 × 10³ (□) bacteria.

Fig. 2

Representative histopathology at 270 days post-infection in experimental progressive pulmonary tuberculosis (PPT) and latent tuberculosis infection (LTI) induced by infection with 10⁶ or 4 × 10³ bacteria, respectively. a. In this late phase of infection, PPT lungs have extensive pneumonia. b. In contrast, LTI showed small and scattered patches of alveolitis. c. The mycobacterial DNA detection in LTI by in situ PCR showed positive labelling (blue dots) in alveolar macrophages (arrow), perivenular fibroblasts (arrow heads) and bronchial epithelial cells (asterisks) from histologically-normal lung areas, and in d. macrophages located in lung hilar lymph nodes.

Fig. 3

Immunostained cell percentages in granulomas from progressive disease () and latent infection (□). Pooled results from two experiments, each with eight mice, are expressed as the mean and standard deviation. Asterisks indicate statistical significance (P < 0·005). (a) IL-2; (b) IL-4; (c) TNF; (d) IL-1; (e) iNOS; (f) BCG.

Fig. 4

Expression of mRNA encoding cytokines and nitric oxide synthase during progressive pulmonary tuberculosis (PPT) and latent tuberculosis infection (LTI) induced by intratracheal infection with 10⁶ or 4 × 10³ bacteria, respectively. Animals suffering PPT (♦) showed higher IL-4 expression, with lower mRNA concentrations of IFN-γ, IL-1α, TNFα and iNOS than mice with LTI (□). (a) IFN; (b) IL-4; (c) TNF; (d) IL-1; (e) iNOS.