Subcellular trafficking of mycobacteria: Implications for virulence and immunogenicity

The aim of this thesis is to determine the properties of the compartment where mycobacteria end up after phagocytosis and which mycobacterial genes play a role in this process. In most cases, bacterial pathogens are taken up by the cell, processed in the endocytic pathway and eventually bacterial derived peptides are presented on MHC class II molecules to CD4+ T-cells. Proteins from viral pathogens in contrast, are degraded in the cytosol and transported into the ER for presentation on MHC class I molecules to CD8+ T-cells. Sometimes, peptides from bacterial pathogens are also presented by MHC class I molecules, a process called cross-presentation. The exact mechanism is still unknown but essential in eliciting a good immune response. Several theories are mentioned; one of these theories is that the phagosome can fuse with the ER, allowing the pathogens and derived peptides to get in contact with MHC class I molecules. This theory raised a lot of questions, and was therefore examined carefully by us and others in chapter 1. Biochemical approaches, together with fluorescence and electron microscopy were used to quantitatively and dynamically assess the contribution of the ER to phagosome formation and maturation. We could not detect continuity between the ER and the plasma membrane, nor were we able to detect a significant contribution of the ER to the formation or maturation of phagosomes in either macrophages or dendritic cells. Instead, our data indicate that the plasma membrane is the main component of phagosomes. During maturation, the phagosomes fuse with endosomal and eventually lysosomal vesicles, to become acidic and degradative organelles. One of the bacterial pathogens whose peptides can be presented via the MHC class I pathway is Mycobacterium tuberculosis. For a very long time, it was not understood how this process takes place, as mycobacteria were reported to replicate in early phagosomes. In chapter 2 we show that after 2 days, M. tuberculosis progressively translocates from phagolysosomes into the cytosol in non-apoptotic cells. Also M. leprae is able to translocate, while the vaccine strain M. bovis BCG or heat-killed mycobacteria remain phagosomal. The translocation process is dependent upon secretion of the mycobacterial gene products CFP-10 and ESAT-6, as mutants that do not secrete these proteins stay in the phagosome. The cytosolic bacterial localization and replication are pathogenic features of virulent mycobacteria, causing significant cell death within a week. This cytosolic localization of virulent mycobacteria may also provide an explanation for the observed MHC class I based antigen presentation which is lacking in current vaccine strains. That it took a long time for this pathway to be discovered can be explained, as the set-up of the experiments is crucial to detect translocation. Translocation takes place later during infection and only with pathogenic species. Furthermore, not every technique is suitable to detect cytosolic mycobacteria. In chapter 3 we discussed why mycobacterial translocation to the cytosol has often been overlooked or has not been detected, but also addressed its importance for pathogenicity and future studies. The mycobacterial translocation pathway was further characterized in chapter 4, where we demonstrate the causality between translocation and pathogenicity. We show that various pathogenic, patient-derived mycobacteria are able to translocate to the cytosol, while non-pathogenic species or mutants can not. We further establish in this chapter that the mechanism of translocation is dependent on the ESX-1 (type VII) secretion system, as only strains that have a functional secretion system translocate, and reintroduction of the ESX-1 gene cluster into the non-virulent vaccine strain M. bovis BCG is sufficient to induce translocation. Several proteins are secreted by this secretion complex. We show that ESAT-6, and specifically the C-terminal region of the secreted ESAT-6 protein, is crucial for translocation. These data establish that the ESX-1 secretion system causes translocation and thereby determining mycobacterial pathogenicity. The compartment the mycobacteria reside in, determines which immune response will be triggered. The translocation to the cytosol gives more clarity on how antigen presentation via MHC class I molecules takes place. But there are also still question marks on the CD4+ T-cell activation, either via the MHC class II or CD1 presentation pathways. MHC class II molecules normally acquire peptide antigens and rapidly traffic from the phagolysosome to the plasma membrane, whereas CD1 molecules instead load lipid antigens, and continually recycle between the phagolysosome and the plasma membrane. In chapter 5 we show that both MHC class II and CD1b localize to bacterial phagosomes early after phagocytosis, but MHC class II traffics from the phagosome to the plasma membrane before the antigens are availabile and loading can take place. In contrast, CD1 antigen presentation is not hampered during M. tuberculosis infection, as a pool of CD1 molecules remains available on the phagolysosomal membrane that is able to acquire lipid antigens and deliver them to the plasma membrane. Thus, rather than facilitating antigen presentation, a lack of coordination in timing may allow M. tuberculosis to use DC maturation as a mechanism of immune evasion. An important and powerful technique, which is used often in this thesis to answer biological questions, is the transmission electron microscope. This microscope can go to much higher resolutions than light microscopy. Although special sample preparation is necessary, there is no need for (antibody) markers to visualize cellular structures or compartments. By morphological distinctive features, organelles can be distinguished, although protein A gold markers can be used as well to detect specific proteins. The advantage of this is that you can visualize in addition the structures in the cell that you were not primarily looking at. In contrast to for example fluorescence microscopy, whereby only the labeled proteins are visible while the rest of the cell remains invisible. Without using the electron microscope, many results described in this thesis would not have been found. I would like to draw special attention to the findings of chapter 2. By examining human dendritic cells and macrophages, which were infected with M. tuberculosis, we unexpectedly found mycobacteria that were not surrounded by a phagosomal membrane. This was a serendipitous finding, as we were not looking for them and were not even aware of the possibility of M. tuberculosis to become cytosolic. If another technique had been used, this important discovery would not have been made. By exploring the cellular localization of mycobacteria in greater detail, also in relation to antigen presenting molecules, we now have more insight in this important mechanism. Not only do we have a better understanding why M. bovis BCG is not very effective as vaccine against tuberculosis, we also know the requirements on how to improve a future vaccine: getting it into the cytosol