Our group has focused on the heat shock protein produced by Mycobacterium leprae (hsp65) as a vaccine antigen against several pathologies including TB, leishmaniasis , diabetes, arthritis [3, 4, 27–29] and cancer. The success of hsp65 vaccination in these different diseases reflects its effectiveness as an immunodominant antigen, and also suggests that it possesses hsp-dependent properties, such as causing the formation of peptide-hsp complexes . When DNA-hsp65 was used to vaccinate mice intramuscularly it elicited an effective immune response only at high DNA doses, which we postulate were necessary because of the loss of a considerable amount of injected pDNA via deoxyribonuclease degradation. To circumvent this problem, drug carriers such as microspheres or liposomes have been developed which protect DNA from extracellular enzymes and could improve DNA vaccination results.
Here we showed that cationic liposomes could be used as delivery system for low doses of DNA-hsp65, causing protection equivalent to that of naked DNA against M. tuberculosis. Most important were the findings that the use of liposomes resulted in a 16-fold reduction in the amount of DNA used in the immunization and that this could be administered in a single dose through a non-invasive route.
The physico-chemical characterization of liposomes containing DNA entrapped or complexed on the surface demonstrated that these structures have positive zeta potential. This cationic characteristic supports DNA delivery to the cells since the driving force for the binding of lipid structures to the negatively charged cell membrane is electrostatic . Furthermore, different DNA-liposome association methods produced liposomes with different characteristics of mean diameter and morphology. Our findings are in accordance with the studies of Perrie and colleagues , which described the DNA localization in dehydrated-hydrated vesicles obtained by entrapping or complexing methods. The entrapping process distributes the DNA along the bilayers and the complexing concentrates the DNA on the external surface of the vesicle. However the mean size obtained in our study with complexed DNA on the liposome surface was smaller than previously reported  due to our modified protocol, rigid temperature control (4°C) and intense vortexing.
In vitro studies showed that our liposome preparations are safe with low cytotoxicity even at high concentrations. Also, in vivo studies showed that the immunization with liposomes skewed the antibodies production to IgG2a without IgG1 production, suggesting a polarization to Th1 pattern of immune response. This profile is similar to previous data of our group injecting PLGA microspheres carrying DNA-hsp65 plus trehalose 6,6'-dimycolate (an adjuvant derived from cell walls of mycobacteria) by intramuscular route . Results from immunization with naked hsp65 showed a mixed antibody response and levels of IgG2a antibodies than seen with the liposome delivery system. These differences could be related to the amount of DNA, the different doses or to the immunization route. However, the liposomes are the delivery vehicle of choice as they elicited Th1 polarization without the use of additional adjuvant.
Our analysis of different immunization protocols showed that one intramuscular dose (which delivers 50 μg of DNA) did not promote a significant reduction in CFU number (Figure 4). However, all groups immunized twice with liposome exhibited a slight reduction in bacterial growth, suggesting that the constituents of liposomes may have intrinsic immunostimulatory properties. To analyze a new route of delivery, we used the immunization through the nasal mucosa that provides a simple, non-invasive route to deliver DNA vaccines stimulating mucosal immunity . While this worked effectively for our liposome bearing DNA-hsp65, the intranasal delivery of naked DNA was ineffective. Possibly naked DNA was destroyed by physical and chemical mucosal barriers that abrogate the gene transfection including endonucleases, enzymes present on mucus, low pH and fast elimination by ciliate epithelium [33, 34], a scenario that can be circumvented by liposomal carriers. Furthermore, cationic liposomes have been proved to enhance immune response against pathogens [32, 35] including M. tuberculosis  when administered intranasally. COMP-hsp65 intranasal instillation conferred a significant impairment in bacilli growth that was comparable to 400 μg of naked DNA injected by intramuscular route.
Since the lipids and molar ratio of both liposomes were similar, the reason that only COMP-hsp65 immunization achieved effective protection could be due to where the DNA resides. The higher exposure of DNA on the liposome surface in COMP-hsp65 may favor the interaction of CpG motifs present in plasmid DNA with receptors such as endogenous toll-like receptor 9 (TLR9) leading to an activation of antigen-presenting cells . Another feature to be considered is that the diameter of the liposomes could influence the macrophage phagocytosis process, as showed by Volle et al. , which established a relation between the size of synthetic particles and phagocytosis efficacy, a result confirmed in rat alveolar macrophages . Related, we observed that mice immunized with larger particles (COMP-hsp65) caused a significant reduction of CFU from infected lung tissue, suggesting that COMP-hsp65 delivery improved the DNA capture by immune competent cells. However, additional studies are required to test this hypothesis.
In addition to protecting against bacterial growth, new vaccines need to be evaluated for their ability to maintain the integrity of the tissues involved in the pathology. In our study, COMP-hsp65 vaccination preserved the pulmonary parenchyma (Figure 5) and reduced tissue inflammation (Figure 6). Therefore, liposomes complexed with DNA-hsp65 showed the ability to stimulate the immune system in a manner that resulted in bacilli reduction and lung preservation. These effects could be due to cytokines present in the pulmonary parenchyma, as high levels of IFN-γ were detected in mice immunized with both naked hsp65 and COMP-hsp65 (Figure 7). This finding suggests that COMP-hsp65 is a promising vaccine, since a Th1 immune response could be important for the development of protective immunity against M. tuberculosis infection [29, 40–42]. On the other hand, production of the regulatory cytokine IL-10 was reduced in vaccines that conferred effective protection against M. tuberculosis infection (naked hsp65 and COMP-hsp65), when compared to non-immunized mice. This is particularly interesting because the role of IL-10 in TB is controversial, since it has been reported that IL-10-/- mice are not protected from infection [43, 44] while others showed that increases of IL-10 expression leads to a down regulation of the immune response against TB [45, 46]. Indeed, our group recently showed that the evaluation of many immunological parameters is necessary to understand the mechanisms underlying the immune response against TB [47, 48]. In our model, it seems that a combination of the increase of IFN-γ expression and a decrease of IL-10 expression improves the immune response leading to bacilli reduction and lung parenchyma preservation.
The utilization of cationic liposomes as a drug delivery system represents an advance for vaccine technology, which has been growing in recent years. Liposomes delivering amphotericin B is used in the therapy of visceral leishmaniasis patients with great efficacy . Okada and colleagues showed 100% of survival in a cynomolgus monkey model infected with M. tuberculosis, using BCG plus DNA-hsp65 vaccine carried by liposomes . Moreover, several articles pointed out the effects of DNA-hsp65 immunization in different protocols and a study has been done to achieve an effective, cost-convenient and safest vaccination scheme (Souza et al., in press). Additionally, two DNA vaccines for veterinary use (horse and salmon) are in the market and there are several reports showing the safety of DNA vaccines when tested in preclinical and clinical trials . We recently tested DNA-hsp65 vaccine in a phase I/II clinical trial in cancer patients and the results showed no toxicity or autoimmunity reactions (Michaluart et al., in press). Considering that liposomes are currently used in medical products and DNA vaccines are under development, their combination in TB vaccination offers a promising approach to fighting this disease.