Characterization of the immune response driven by novel nanoparticle-based vaccines for tuberculosis

Abstract

Tuberculosis (TB) is an infectious disease that has become one of the leading causes of death and worldwide, claiming around 1.5 million deaths annually. It is caused by the mycobacterium Mycobacterium tuberculosis (Mtb), a small bacillus which is transmitted in droplets generated by sick patients while talking or coughing. In addition, the bacteria is capable of entering into a quiescent state and remaining dormant in the body for years. New threats such as the development of multi drug resistant strains urges the development of new strategies to control the disease, including an effective vaccination strategy. To date, the only authorized vaccine against TB is BCG (Bacille Calmette-Guérin), generated from an attenuated strain of Mycobacterium bovis. However, this vaccine has a heterogeneous efficacy against pulmonary TB in adults, the most common form of the disease. For this reason, multiple attempts have been made to develop new vaccines that improve the protection conferred by BCG. Among them, one of the most popular approaches is to boost BCG (normally administered subcutaneously during infancy), with new vaccines administered via mucosa to enhance the local response in the lungs, the first site of infection. New materials, such as nanoparticles, have opened the door to the development of new types of TB vaccines. Nanoparticles are elements up to 1000 nm in diameter, and they have unique properties due to their size. For example, they are easily recognized by the immune system as they are similar in size to many microorganisms, and they have a high surface/volume ratio, which favour the adsortion of multiple molecules. Nonetheless, one of the factors hampering the development of effective TB vaccines is the lack of knowledge of the mechanisms that confer complete and lasting protection against the disease. Mtb is an intracellular bacterium, whose favourite niche are the alveolar macrophages. Macrophages are innate cells designed to phagocyte and destroy pathogens, but Mtb is able to escape this fate through different mechanisms. T cells have a crucial role in TB by producing interferon (IFN)-gamma and activate macrophages, but they are not enough to achieve complete protection. The study of new vaccines may help discerning the mechanisms underlying TB protection and the development of new vaccine targets. In this thesis, we tested two new models of TB nanoparticles-based vaccines in a mouse model. Nanovaccines were used used as a mucosal (intranasal) boost of BCG. The work developed in this thesis was part of the European Horizon 2020 project "Eliciting Mucosal Immunity to Tuberculosis (EMI-TB)". The two candidate vaccines, Nano-FP1 and Nano-FP2, were designed and provided by collaborators of this consortium. The nanovaccines were composed of lipid nanoparticles of yellow carnauba palm wax with sodium myristate, polyinosinic-polycytidylic acid as adjuvant, and two different fusion proteins (FP1 and FP2, respectively). The fusion proteins were formed by three Mtb antigens: Acr, Ag85B and HBHA for FP1, and Acr, MPT64 and HBHA for FP2. The Nano-FP1 vaccine showed good protective results in previous analyses performed by our collaborators. We aimed to define the immune profiles triggered by the candidate nanovaccines and correlate those changes with the degree of protection obtained against Mtb infection. For this purpose, we used C57BL/6 mice, one of the most used strains for the study of the immune system. 7-week-old mice were subcutaneously administered BCG. 12 weeks later, two dosis of the novel nanovaccines, either Nano-FP1 (group called BCG /Nano-FP1) or Nano-FP2 (BCG/Nano-FP2), were administered intranasally, separated by a 2-weeks interval. Mice vaccinated only with BCG and mice without any vaccination (Naive) were also included. Furthermore, some mice were administered a third intranasal dose of the nanovaccines at 11 weeks after the second dose, to study the effect of a third dose. In the first chapter of this thesis, mice were immunized and subsequently infected with Mtb at different time points after vaccination: 2-3, 7 and 11 weeks, and 3 weeks after receiving the third dose (14 weeks). After thirty days of infection, mice were sacrificed, and lungs and spleens were harvested. Whereas lung represent the first site of infection, spleen was used to study the systemic response. One fraction of the organs was used to measure the bacterial load, and other fraction was used to analyse the immune populations. The bacterial load study revealed that mice vaccinated with BCG/Nano-FP1 had a lower bacterial load in the lung than those vaccinated with BCG alone 2-3 weeks after the booster, but not those vaccinated with BCG/Nano-FP2. However, this increase in protection conferred by BCG was rapidly lost, and 7 weeks after the boost there were no longer significant differences between the BCG/Nano-FP1 and BCG groups. Administration of the third booster of Nano-FP1 at 11 weeks again improved protection against Mtb, but the effect was slightly less than that seen after the first two doses. Surprisingly, 11 weeks after the second dose, mice vaccinated with BCG/Nano- FP1 significantly increased their bacterial load in lungs compared to mice vaccinated with BCG alone. In spleen, no big differences in the bacterial load were found among groups. To complement this analysis, a lung lobe was fixed and processed to analyse the percentage of area affected by cellular infiltration, to compare the percentage of area affected by the infection. However, we did not find statistically significant differences between vaccination groups or at different time points. To characterize the immune cell types participating in the infection, we analysed by flow cytometry multiple immune cell populations. In the first cytometry panel, we studied the main cellular components of the innate system: dendritic cells, macrophages, neutrophils, eosinophils and NK cells. In a second panel, we analysed the proportion of T and B lymphocytes, and the main phenotypes of T cells: central memory, effector memory, effector or resident memory. Lastly, we designed a third panel to study the percentage of T lymphocytes secreting the cytokines IFN-gamma, tumor necrosis factor (TNF)-alpha, interleukin (IL)-2 and IL- 17. We also assessed the existence of polyfunctional lymphocytes (that secrete more than one type of cytokine at the same time), since they have been proposed as mediators of protection in TB and other diseases. Nonetheless, we found no big differences among vaccination groups in none of the three panels analysed. We hypothesized that maybe after thirty days of infection it was too late to ob- serve any differences among groups. Infection may be already controlled in all groups, and many immune cells could also be exhausted. Because of that, the second chapter of this thesis focused on the study of the immune response induced by vaccines without infection, to characterize the immunological changes induced solely by the nanovaccines. We performed similar analyses than those performed on chapter one. We studied the innate and adaptive populations by flow cytometry, as well as the phenotype of T lymphocytes and their ability to produce the cytokines IFN-gamma, TNF-alpha, IL-2 and IL-17. We selected again four time points (2, 7, 11 and 14 (3rd boost) weeks), and two organs: lung and spleen. Without infection, we found several differences between vaccination groups. Mice vaccinated with BCG/Nano-FP1 slightly increased the percentage of interstitial macrophages at 2 weeks, and the number of neutrophils at 11 weeks after the second boos. Interestingly, shortly after the second and third doses (2 and 14 weeks, respectively), it seemed to be a decrease in the number of neutrophils, but it was not statistically significant. It was described that neutrophils may play a dual role in TB: on the one hand, they help fight infection, but on the other hand, they can become large reservoirs of the bacteria and favour its spread. Because of that, the observed increase in neutrophils at 11 weeks could be correlated with the increase of bacterial load in mice vaccinated with BCG/Nano-FP1 observed in the previous chapter. Both nanovaccines, but especially Nano-FP1, had a strong effect on the adaptive system. The BCG/Nano-FP1 vaccine increased the population of total and activated CD4+ T cells (CD44+) in the lung. In addition, both BCG/Nano-FP1 and BCG/Nano-FP2 vaccines increased the percentage of activated CD8+ T cells. However, all these changes were observed at short term after the boosts (2 and 14 weeks), but they were lost from 7 weeks on. Regarding the immunophenotype of the lymphocytes, we found that BCG/Nano-FP1 increased the fraction of memory-resident CD4+ and CD8+ lymphocytes, especially at 2 and 14 weeks. Vaccination with BCG/Nano-FP2 also increased the number of CD8+ cells with that phenotype. Resident memory lymphocytes are a subset of lymphocytes that do not recirculate, remaining in the organ where the first infection occurred. It is a population of growing interest, especially in TB research, as these cells constitute a rapid line of defence in case of reinfection. However, the resident memory cells triggered by the nanovaccines also decreased 7 weeks after the boost. The two nanovaccines also induced multiple cytokine-secreting CD4+ cells, but especially by BCG/Nano-FP1 vaccination. BCG/Nano-FP1 induced monofunctional cells secreting IFN-gamma, TNF-alpha and IL-17, and polyfunctional lymphocytes, secreting the following combinations: IFN-gamma+ TNF-alpha+, TNF-a+ IL-17+, IFN-gamma+ TNF-alpha+ IL-2+, and IFN-gamma+ TNF-alpha+ IL-17+. Similarly to the previous sections, all these cells were lost at 7 weeks until they completely disappeared at 11 weeks, but they reappeared with the third boost of the vaccine at 14 weeks. The third and last chapter of the thesis was dedicated to the gene expression analysis of the complete transcriptome, by using the technique of ribonucleic acid sequencing (RNA-Seq). We selected this method to obtain an unbiased and complete vision of the processes occurring in the lung, to define better the mechanisms induced by the nanovaccines and to identify new correlates of protection. In addition, in this chapter we differentiate two compartments of the lung: bronchoalveolar lavage and lung parenchyma. Bronchoalveolar lavage is performed through washing the respiratory tract with saline phosphate buffer, collecting the cells located in the airways, while the lung parenchyma represent the immune fraction of the rest of the lung. We performed a differential expression analysis, comparing each vaccination group (BCG, BCG/Nano-FP1 and BCG/Nano-FP2) to their respective Naive control group, at the different time points studied: 2, 11 and 14 weeks. The boosts altered the expression of a large number of genes at short-term. Big differences were also found among the two nanovaccines and, more surprisingly, among the second and the third boost of each regimen. Next, we analysed the pathways related to the altered genes. Multiple pathways were shared between all vaccination groups, such as immunoregulatory interactions, generation of messenger molecules or antigenic presentation. Focusing on the pathways induced uniquely by BCG/Nano-FP1 at 2 weeks (the time-point of maximum protection), we found the following pathways: IFN-gamma signaling and nucleotide receptors in bronchoalveolar lavage, costimulation by the CD28 family, production of reactive oxygen and nitrogen species, TNFs and their receptors or sphingolipid metabolism in lung parenchyma. We also analysed the expression of multiple genes related to immune cells (especially lymphocytes) and other immune processes. We found an increase in genes related to the Complement system and multiple genes that could be participating in protection against TB, such as interferon-inducible genes, antimicrobial or signalling-related genes. Genes that code for the T cell receptor (TCR) and immunoglobulins were also inspected. A higher number of both TCR and immunoglobulins genes were induced at 2 weeks compared to 14 weeks, especially in bronchoalveolar lavage. Our results showed big differences between the two lung compartments, and between the second and third boosts of the nanovaccines. Further research will be necessary to explain the differences caused by the repeated boosting. To obtain a reduced list of candidate genes correlates of protection, we selected those genes uniquely expressed in the BCG/Nano-FP1 group at 2 weeks. For that, we performed a differential expression analysis of the BCG/Nano-FP1 group com- pared to the other three vaccination groups (BCG/Nano-FP1 vs Naive, BCG/Nano- FP1 vs BCG, and BCG/Nano-FP1 vs BCG/Nano-FP1) at each time point (2, 11 or 14 weeks). Then, we selected those genes that were differentially expressed by BCG/Nano-FP1 in the three comparisons. We compared the results obtained in the 3 time points and selected those genes that were only present at 2 weeks. Following this strategy, we obtained a list of 22 candidate genes in bronchoalveolar lavage and 29 genes in lung parenchyma. Some of the upregulated bronchoalveolar lavage genes included: Nos2, which encodes an enzyme important in the production of nitric oxide in macrophages, genes coding for Complement proteins C1s1 and C3, Itgam genes or Cd38. Some of the upregulated genes obtained in lung parenchyma included: multiple genes coding for immunoglobulin chains (Ighg2b, Ighg2cIgkv1-135, Ighv9-3, Igkv1-133 Igkv1-117), the chemokine Ccl8, the macro- phage marker Angptl2, or the Toll-like receptor Tlr12. We also found some genes inhibited in BCG/Nano-FP1, such as the gene coding for cathepsin K, Cstk, the alternatively activated macrophage marker Mgl2 or the chemokine Ccl17. As conclusions of this thesis, we showed that vaccination with two doses of Nano- FP1 as an intranasal boost of BCG improves protection against TB in mice, but only at short term. We defined the immune populations mobilized by the vaccination in the lung, which include an increase in CD4+ lymphocytes, the activation of CD4+ and CD8+ lymphocytes, and diverse monofunctional and polyfunctional cytokine-secreting CD4+ lymphocytes. Furthermore, we observed a decrease in protection in mice vaccinated with BCG/Nano-FP1 at 11 weeks, that could be related to an increase in neutrophils in the lung. Last, we described the transcriptional changes induced by the vaccine, such as multiple genes related to lymphocyte activation and recruitment, macrophage activation, and immune signalling, and we obtained two lists of candidate genes (for bronchoalveolar lavage and parenchyma, respectively) that could serve as TB correlates of protection. Our results highlight the importance of designing complete and long-term vaccine characterization studies. Furthermore, the genes and pathways proposed as correlates of protection could be explored in future works, to assess their roles in the disease.