Emerging Understanding of Tuberculosis and the Granuloma by Comparative Analysis in Humans,Cattle,Zebrafish,and Nonhuman Primates

The granuloma is the hallmark lesion of tuberculosis in man and animals. Within the granuloma, the host interacts with the pathogen, and disease outcomes are determined. The granuloma is essential for control of mycobacterial pathogens, yet paradoxically, the granuloma also provides a venue for bacterial survival, multiplication, latency, and dissemination. Within the granuloma, microbial offense meets host defense; which side wins is the question addressed in this issue of Veterinary Pathology by A.J. Martinot. ¹⁴

The granuloma is the host’s attempt to both isolate and kill mycobacterial pathogens as well as other persistent pathogens and foreign bodies. In the process, however, inflammation and tissue damage caused by the host response is often more detrimental than the pathogen itself. As necrosis progresses, erosion into airways or vasculature results in dissemination of infection within the host and enables transmission to other susceptible hosts.

Study of the pathogenesis of tuberculosis in humans is restricted by the inability to conduct controlled experimental infection studies. The mouse model of human tuberculosis has enhanced our understanding of the immune response to Mycobacterium tuberculosis, including the critical roles of CD4+ and CD8+ T-cells in restricting disease dissemination and enhancing mycobacterial killing. Martinot ¹⁴ draws our attention to the evolutionary conservation of CD4+ and CD8+ T-cell epitopes by M. tuberculosis, suggesting that this T-cell dominated response is actually favorable and beneficial to this ancient pathogen. ¹⁴ Evolutionary epitope conservation is not limited to M. tuberculosis only but is seen in other members of the M. tuberculosis complex, including Mycobacterium bovis, the cause of tuberculosis in cattle and many other animals. ¹⁴

Granulomas of bovine tuberculosis are remarkably similar to those of human tuberculosis, and cattle have been suggested as a model of human tuberculosis. ²⁶ Similar to human tuberculosis, CD4+ and CD8+ T-cell responses have been shown to be necessary in bovine tuberculosis. ^15,20 Additionally, studies in cattle demonstrate the critical nature of γ/δ T-cells during M. bovis infection, especially in the early response, where they may play a role in bridging the innate and adaptive immune responses. ¹⁶ The large numbers of γ/δ T-cells in cattle, compared to humans or mice, make cattle extremely useful in the study of this important class of T-cells. ⁷

To understand who controls the TB granuloma, Martinot ¹⁴ suggests we focus on immune recognition of highly conserved T-cell epitopes that favor mycobacterial survival. One protein with numerous highly conserved T-cell epitopes is early secreted antigenic target 6 (ESAT-6). ¹⁴ This highly immunogenic protein is used as an antigen in the interferon gamma release assay (IGRA) for the diagnosis of human tuberculosis, Quantiferon. The forerunner of the Quantiferon assay was an IGRA developed for cattle, ²⁸ a perfect example of the mutual benefit of human and veterinary research. ²⁶ A similar IGRA (Bovigam) is now used as an official test for bovine tuberculosis in many countries, including the US. Although the Bovigam assay currently uses M. bovis purified protein derivative (PPD) as antigen, to increase specificity, modified versions have examined the use of ESAT-6 as antigen. ³ Moreover, ESAT-6 is being considered for use in assays to differentiate M. bovis-infected from M. bovis BCG-vaccinated animals (ie, differentiation of infected and vaccinated animals, or DIVA assay). ²³ Studies have shown the potential benefit of BCG vaccination of cattle; however, reluctance to use BCG stems in part from the inability of current diagnostic tests to differentiate vaccinated from infected cattle. ESAT-6 is also used as an antigen in antibody-based assays approved for use in the diagnosis of M. bovis infection in deer and M. tuberculosis infection in elephants. ^1,6,12,13,27

If the granuloma is the playing field where host defense and microbial offense meet, it stands to reason that understanding interactions at the level of the granuloma would be crucial. Animal models of human tuberculosis, most notably the mouse model, have elucidated some details; nevertheless, the mouse model does not recapitulate many features of granuloma formation or maintenance as seen in human tuberculosis. In contrast, granulomas resulting from Mycobacterium marinum infection of zebrafish and especially M. tuberculosis infection of nonhuman primates (NHP) closely resemble those seen in human tuberculosis, and their study has revealed many interesting findings.

Mycobacterium marinum is a natural pathogen of zebrafish, causing systemic disease with granulomas structurally similar to those of human tuberculosis. ¹⁷ The zebrafish model has shown that granulomas are dynamic rather than static, granuloma formation is initiated before adaptive immune responses develop, and the granuloma is exploited by mycobacteria in providing a niche for accelerated bacterial replication. ¹⁷

Zebrafish embryos are semi-transparent; as such, fluorescent markers can be used to visually track both mycobacteria and host cells during experimental infection. Observation of such cellular events is not a feature of other mycobacterial disease models. Injection of M. marinum in the hindbrain ventricle of zebrafish embryos reveals that even a single infected macrophage will stimulate the chemotactic recruitment of noninfected macrophages. Thereafter, granuloma expansion is driven by a continual cycle of apoptotic death of infected macrophages and phagocytosis of membrane-bound cell remnants containing bacilli. ⁴ Within the first 24 hours, approximately 90% of naïve, infiltrating macrophages are infected. Macrophages not only migrate toward the site of infection but move within the granuloma. Importantly, some infected macrophages exit the granuloma to disseminate infection and produce secondary granulomas. ⁴

The NHP model has the distinct advantage of the close genetic relationship between NHP and humans, including similarities in immune responses. Thus, the NHP model most closely mimics the spectrum of granuloma types seen in humans. ¹¹ Elegant studies using the cynomolgus macaque model have added much to our understanding of who controls the TB granuloma, and the answer is complex. Although clinical disease in human patients or animals may be described as active or latent, the NHP model has clearly shown that pulmonary granulomas within the same animal are dynamic and independent from each other. ¹⁰ Although it is the cumulative activity of all granulomas that define the clinical picture, individual granulomas are widely variable in terms of morphology, cellular constituents, cytokine environment, bacterial burden, and bacterial killing. ⁵ A ramification of such findings is that to decipher the pathogenesis of mycobacterial diseases, host-pathogen interactions must be evaluated at the granuloma level, an undertaking ideally suited to pathologists in collaboration with immunologists, microbiologists, and molecular biologists.

Preliminary studies are beginning to characterize the morphologies, cellular constituents, and cytokine patterns of granulomas in bovine tuberculosis. Although results differ in some respects between studies, they support the important roles of CD4+, CD8+, and γ/δ T-cells in granuloma formation and document the production of inflammatory cytokines such as IFN-γ, TNF-α, and IL-17A, as well as anti-inflammatory cytokines such as IL-10 and TGF-β. ^{2,8,9,16,18,21,22,24,25} For example, studies have shown that granulomas of similar microscopic morphology, within the same organ, may differ in cytokine gene expression profiles. ¹⁹ To better understand host-pathogen interactions in the bovine granuloma, studies are needed to correlate cellular and cytokine findings with bacterial burden, bacterial killing, and bacterial gene expression at the granuloma level.

Newer approaches, such as gene expression analysis and proteomics, combined with anatomic pathology, will further the investigation of microbial offense and host defense at the granuloma level. Such work will not only further our understanding of tuberculosis pathogenesis but also aid in developing enhanced diagnostic assays and efficacious vaccines for both humans and animals.