Understanding buruli ulcer : elucidation of the cellular and molecular mechanisms underlying infection with mycobacterium ulcerans

Buruli ulcer (BU) is an emerging neglected tropical skin disease caused by Mycobacterium ulcerans and remains the third most common mycobacteriosis worldwide, exceeded only by tuberculosis and leprosy. Infection with M. ulcerans starts off as a nonulcerative cutaneous lesion and can eventually break down into an ulcer if left untreated. Development of severe disabling deformities is extremely frequent in BU patients, having a significant socioeconomical impact and undermining the potential development of the affected populations. BU disease is closely related with the low temperature requirement of M. ulcerans (30- 33ºC), which favors the development of cutaneous lesions; with its slow growth rate that translates into slowly progressing lesions; and, most importantly, with the secretion of a lipidic exotoxin known as mycolactone, which is responsible for the induction of cell death and tissue necrosis. All these particular features impact on the mycobacterial nature of this intriguing pathogen. Mycolactone has been attributed cytotoxic and immunosuppressive properties and is suggested to inhibit both local and systemic protective immune responses. Indeed, the histological hallmarks for advanced BU lesions are extensive necrosis and minimal or absent inflammatory infiltrates, in the presence of abundant extracellular bacilli. In contrast, infections with the genetically related pathogenic mycobacteria M. tuberculosis and M. marinum have been consistently associated with the recruitment of acute and chronic inflammatory infiltrates to the site of infection. Despite the histological pattern associated to BU, several studies have associated cell-mediated immunity (CMI) and delayed-type hypersensitivity (DTH) responses as being protective against M. ulcerans infections, suggesting that the predominantly described histopathology of BU lesions may not reflect the actual host-parasite interaction at the active site of infection. Therefore, to more extensively reassess the dynamics of the inflammatory response we performed a comprehensive histological analysis of footpads of mice infected with M. ulcerans, throughout the whole course of the infectious process. We found that experimental infections with M. ulcerans induced persistent, virulence-dependent inflammatory responses. A mycolactonenegative avirulent strain induced an initial acute neutrophilic inflammatory process that was gradually replaced by a predominantly chronic mononuclear infiltrate with granuloma-like organization. On the other hand, infection with mycolactone-positive strains was associated with a persistent influx of a mixed inflammatory infiltrate composed of neutrophils and macrophages with the frequent presence of intraphagocytic bacilli. The recruited cells were later depleted in the center of the infectious focus, by apoptosis and lytic secondary necrosis, which led to the generation of characteristic necrotic acellular areas with abundant extracellular bacilli. At the periphery of this necrotic area cellular infiltrates with intraphagocytic bacilli were consistently observed. At later stages of infection, these necrotic areas devoid of inflammatory cells expanded by progressive invasion of healthy tissues reproducing at a lower scale the scenario typical of human biopsies. Our results led to the proposal that M. ulcerans would be, like the other pathogenic mycobacteria, a facultative intracellular pathogen with phases of intramacrophage and extracellular multiplication, a hypothesis confirmed by subsequent studies in our laboratory. This interpretation allows the reconciliation of the cytological, histopathological, microbiological and immunological features of M. ulcerans infection with the above highlighted particular traits of this peculiar mycobacterium. We have provided evidence for the development of a local cellular immune response at the site of M. ulcerans infection. However, several reports have suggested that mycolactone inhibits local as well as systemic protective immune responses. In this study, we found that M. ulcerans infection triggered the early development of a T cell response in the draining lymph node (DLN) characterized by an increase in the number of mycobacteria-specific IFN-γ-producing cells. However, mice remained susceptible to infection with a highly virulent strain, and disease progression was accompanied by a depletion of CD4+ cells, not only in the active infection foci, but also in the DLN, contributing to the loss and thus ineffective immune response. The dissemination of M. ulcerans to the DLN and its ultimate destruction further impaired a sustained protective immune response. Despite this, advanced infection with M. ulcerans did not render mice more susceptible to a systemic co-infection with Listeria monocytogenes, suggesting that the local and regional immunosuppression associated to M. ulcerans infection is not accompanied by systemic immunosuppression. Given our previous observations, and bearing in mind the reported association of CMI, DTH and IFN-γ with host resistance to BU, we next focused on the role of this cytokine in M. ulcerans infections. In vivo and in vitro experimental infections demonstrated that IFN-γ is required for the control of both an avirulent and an intermediate virulent strain of M. ulcerans. However, for a highly virulent strain, no differences were found in bacterial control, regardless the presence of IFN-γ. When cultured macrophages were infected with different multiplicities of infection (MOI), we found that the lack of protection was due to a virulence-dependent/MOI dependent impairment of the macrophage antimicrobial mechanisms triggered by IFN-γ, such as phagosome maturation, phagolysosome fusion, and generation of reactive nitrogen intermediates. This inhibition of macrophage antimicrobial activity was due to the production of mycolactone, since the addition of increasing concentrations of purified toxin to cultured macrophages infected with avirulent mycolactone-negative M. ulcerans had a dose-dependent inhibitory effect on phagosome acidification/maturation and nitrite production. The fact that adaptive immunity is initially triggered against M. ulcerans infection suggests that the development of improved vaccines that boost this type of response could lead to an early recall immune response before the build-up of mycolactone compromised T cell differentiation/maintenance. Therefore, we then characterized the immunological mechanisms underlying the transient protection conferred by vaccination with M. bovis BCG or with a mycolactone-negative strain of M. ulcerans, in the footpad murine model of experimental BU. We observed that BCG vaccination was effective in delaying the onset of disease and bacterial proliferation, regardless the experimental dose of M. ulcerans footpad infection. At the site of infection, previous BCG vaccination triggered a strong chronic inflammatory response with increased levels of IFN-γ expression. In the DLN, a faster, heightened and longer T cell and IFN-γ responses were also mounted. However, in vaccinated mice, the eventual increase in bacterial loads and the development of extensive areas of tissue necrosis in footpads resulted in a delayed progression to ulceration and DLN necrosis. Immunization with the mycolactone-negative M. ulcerans strain also significantly delayed the progression of infection by virulent M. ulcerans, but pathology in the DLN and in the footpad also eventually prevailed with emergence of ulceration. The findings presented here help clarify several aspects of host response to M. ulcerans infection and have important implications for the development of new prophylactic and therapeutic strategies against BU.

Name thesis:degree_name

Tese de Doutoramento em Ciências da Saúde

Year dc:date.issued

2010

Author dc:creator

• Fraga, Alexandra G.

Advisors dc:contributor.advisor

• Pedrosa, Jorge
• Castro, António G.

Statement dc:rights

• restrictedAccess

Language dc:language.iso

por

Handle dc:identifier.uri

https://hdl.handle.net/1822/11812

OAI record oai:identifier

http://repositorium.sdum.uminho.pt/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Arepositorium.uminho.pt%3A1822%2F11812