Tuberculosis (TB) is a chronic infectious disease caused by members of the Mycobacterium tuberculosis complex (MTBC). Despite being a preventable and curable disease, TB remains as the world’s leading cause of death from a single infectious agent in humans (1). In animals, a wide range of mammals act as reservoirs of TB, including domestic species such as cattle and goats, as well as wildlife (2). While M. tuberculosis is the main causative agent of human TB, animal TB is primarily attributed to M. bovis, M. caprae, and M. microti, which can also infect humans and vice versa, being then a zoonosis (3).

The Bacillus Calmette-Guérin (BCG), a live-attenuated strain of M. bovis, remains the only licensed vaccine against TB for humans (4, 5), however, it has variable efficacy and does not fully prevent TB infection in adults (6). Similarly, BCG administered parentally and orally has shown a variable degree of protection against MTBC in cattle, goats, and wildlife (7–12). An alternative approach of animal vaccination using a heat-inactivated M. bovis (HIMB) vaccine administered parentally and orally has also shown divergent levels of protection against MTBC challenge in various species (13–17). Given these limitations, there is a clear need to explore novel vaccine candidates and strategies that can elicit stronger and more consistent protection across hosts.

To date, the understanding of protective immunity against TB in both animals and humans remains incomplete and correlates of protection are poorly defined (18), as efficient immunity may comprehend a complex interplay between adaptive and innate responses. Antigen-specific CD4⁺ T-cell subsets producing IFNγ and TNFα have been identified as key components of protective immunity (19, 20). However, BCG and HIMB vaccines, when administered by subcutaneous or intradermal routes, induce systemic T-cell-specific responses that are insufficient to confer full protection against MTBC challenge (16, 21–23), suggesting that additional immune mechanisms are required to effectively prevent infection.

Given that MTBC is primarily transmitted via aerosols and enters the host through the airway mucosa, it is hypothesized that vaccine efficacy could be improved by changing administration route and harnessing mucosal immunity mechanisms (24–27). Mucosal vaccination with BCG via the respiratory tract has been reported to elicit robust antigen-specific memory CD4⁺ and CD8⁺ T-cell responses in the airways, resulting in enhanced protection (28–31). Moreover, mucosal vaccination has also been shown to activate innate immune mechanisms within the lung mucosa, particularly alveolar macrophages (AMs), associated with reduced MTBC infection (32, 33).

In this study, we aimed to assess the immunogenicity of different single-dose and prime-boost vaccination strategies using intranasal BCG or HIMB (with or without adjuvant) vaccines in a goat model, following the experimental design outlined in Figure 1. We assessed MTBC-specific immune responses both in peripheral blood and at the pulmonary mucosal level.

Developing vaccines against MTBC which can elicit an effective mucosal immunity is critical for enhancing early immune responses at the primary site of infection (39, 40). Intranasal (i.n.) delivery platforms offer a promising strategy to elicit both local and systemic immunity, potentially improving upon the limited and variable protection conferred by conventional parenteral BCG and HIMB vaccination, particularly against pulmonary disease (41, 42) which is the most frequent clinical form of active TB in humans as well as in livestock (1, 43, 44). Building on this concept, revaccination represents another strategy to enhance the limited protection conferred by subcutaneous vaccination. However, boosting through the same route as the prime has repeatedly shown little to no additional benefit (45).

In contrast, intranasal delivery as booster may more effectively amplify the pre-existing immunity induced by the initial parenteral vaccination, activating innate and adaptive immunity at a local level unlikely to be achieve by purely parenteral administration. Although BCG is one of the most widely used vaccines globally, and has a well-established safety profile in immunocompetent individuals, a heat-inactivated vaccine could provide several advantages over a live-replicating formulation: improved safety profile, including immunocompromised host; as well as greater stability, which may facilitate storage, distribution and regulatory approval. In previous studies, immunogenicity and efficacy of parenteral HIMB has been investigated in goat and sheep (13, 14, 16). Given the emerging interest in mucosal vaccination—particularly the promising results obtained with intranasal BCG—it is also of consideration to assess HIMB delivered via the same intranasal route.

The lack of animal models that accurately replicate the complex immune responses observed in humans following either M. tuberculosis infection or BCG vaccination hinders the preclinical evaluation and prediction of vaccine efficacy (46, 47). Over the past fifteen years, several studies have proposed the goat model of TB as a feasible and translational platform to resemble key immunopathological features of active human TB (48–50), as well as the immune responses induced by BCG and other mycobacterial vaccines (7, 35, 51, 52). Goats are a suitable TB model as they develop early pulmonary lesions, necrotic granulomas, and disease progression similar to humans and cattle, and they mount comparable cellular and innate immune responses. They also offer practical advantages over cattle, being smaller, easier to handle, more cost-effective, and better adapted to BSL-3 studies. The relevance of this model also relies on the fact that goats are natural hosts of tuberculosis (53), thereby providing not only a robust platform for translational research but also direct applicability of the findings to the target host species.

In this study, we demonstrate that vaccination of goat kids with a single dose of intranasal BCG elicited systemic specific cell-mediated proinflammatory immune responses comparable to those induced by parenteral vaccination with BCG and HIMB (13, 54). Intranasal BCG was able to induce stronger MTBC-specific production of IFNγ and increase the frequencies of CD4⁺IFN γ⁺ and CD8⁺IFNγ T-cells in peripheral blood than intranasal HIMB vaccines at nine weeks after vaccination, regardless of the use of a mucosal adjuvant, suggesting that mucosal delivery of BCG can trigger robust systemic Th1 immune responses. This is consistent with previous studies in non-human primate models where intradermal and pulmonary BCG vaccination induced similar peripheral CD4⁺ T-cell proliferation and IFN-γ responses (55), and aerosol BCG vaccination increased the frequency of systemic PPD-specific IFNγ–secreting cells and elicited Th1- and Th17-type cytokine responses in both CD4⁺ and CD8⁺ T-cells from PBMCs (31, 56). The enhanced systemic responses elicited by intranasal BCG compared to intranasal HIMB may be explained by the fact that BCG is a live attenuated vaccine with replication capacity (57). This allows BCG not only to act at the mucosal surface but also to drain to regional lymph nodes, as supported by histopathological and bacteriological findings. Similarly, prime-boosted groups showed higher frequencies of antigen-specific CD4⁺ and CD8⁺ T-cells producing IFNγ compared to the HIMB-only and non-vaccinated groups.

Antigen-specific IFNγ responses increased notably after the subcutaneous prime and intranasal BCG vaccinations, confirming effective induction of cell-mediated immunity. HIMBpar (s.c.) - HIMBmuc (i.n.)As expected, IFNγ production peaked after s.c. HIMB and BCG vaccination (13). Although overall IFNγ levels were broadly comparable among these vaccinated groups, the HIMBpar (s.c.) – HIMBmuc (i.n.) regimen showed a more sustained response and was the only one to elicit detectable IFNγ against ESAT-6/CFP-10/EspC, suggesting enhanced antigen-specific engagement. In the BCG (s.c.) - HIMBmuc (i.n.) group, however, IFNγ levels gradually declined, consistent with previous findings (7, 57), suggesting that a heterologous boost using HIMBmuc i.n. after subcutaneous BCG does not seem to have an improved effect on IFNγ production. HIMBpar (s.c.) - HIMBmuc (i.n.). By comparison, animals vaccinated with intranasal HIMB alone or with adjuvant, as well as non-vaccinated animals, failed to elicit detectable IFNγ responses, highlighting the limited immunogenicity of HIMB alone when delivered intranasally.

In general, systemic immune responses induced by vaccination over time showed a slight decline in the BCG (s.c.) - HIMBmuc (i.n.)–vaccinated group, both in IFNγ release and in the proportion of IFNγ–producing CD4+ and CD8+ T-cells. In contrast, the responses were maintained or even increased in the HIMBpar (s.c.) - HIMBmuc (i.n.)–vaccinated group. The intranasal BCG group exhibited a marked increase between the two sampling points, likely reflecting that the peak IFNγ response had not yet been reached. Additionally, antigen-specific IFNγ production by PBMCs measured by ELISA showed better direct correlation with CD4⁺ IFN γ⁺ T-cell frequencies than with CD8⁺ T-cell subsets (see Supplementary Figures S4a, b), indicating that IFNγ production is primarily driven by CD4⁺ T-cells, consistent with previous reports (58).

Long lasting protection of effective TB vaccines must relies on durable T-cell memory subsets, such as central memory T-cells (T_CM), which exhibit high proliferative potential, and effector memory T-cells (T_EM), which display robust cytokine secretion capacity (59, 60). When memory CD4⁺ and CD8⁺ T-cell subsets were further characterized using CD62L as memory marker, both antigen-specific IFNγ-producing central memory T-cells (CD62L⁺) and effector T-cells (CD62L⁻) from BCG i.n. and prime-boosted groups produced higher levels of IFNγ than memory and effector cells isolated from non-vaccinated and HIMB i.n. vaccinated individuals. These findings are consistent with previous studies reporting robust proliferation of memory PPDB-specific CD4+ T-cells (CD4⁺CD45RO⁺) in cattle (61) and goats (7) vaccinated and revaccinated with s.c. BCG, respectively; and with Melgarejo et al. (2022), who reported that frequencies of PPDB-specific IFNγ-producing memory T-cells were positively associated with favorable disease outcome in HIMBpar-vaccinated goats.

Even though antibody responses are not considered essential to protective immunity against MTBC, vaccine-induced seroconversion and MTBC-specific IgG levels serve as indicators of vaccine responsiveness at both the group and individual levels (34) It is well established that s.c. HIMB vaccination elicits an earlier and more robust humoral immune response compared to s.c. BCG administration (13, 62). In our study, only HIMBpar (s.c.) - HIMBmuc (i.n.) and BCG (s.c.) - HIMBmuc (i.n.) groups induced detectable levels of antibodies in blood, considerably higher in HIMBpar (s.c.) - HIMBmuc (i.n.) group. These results may reflect that MPB83 is constitutively expressed in virulent M. bovis, whereas BCG strains produce it only at very low levels (63). Additionally, both BCG vaccinated groups, whether intranasally or subcutaneously, showed a boost in antibody levels at week 16 of the study, after skin test was performed, markedly larger in the s.c. BCG (mean ΔOD goes from 0.451 at week 14 to 2.848 at week 16) than the i.n. BCG (mean ΔOD goes from 0.044 at week 14 to 0.307 at week 16) group as previously described (12, 34). This boosting phenomenon of increased MTBC-specific IgG levels has been well characterized before (63). While HIMB intranasally vaccinated groups did not show increment in antibody responses, aligning with the results observed for cell-mediate immunity.

It has been previously shown that mucosal vaccination with BCG (27, 64) and an attenuated M. tuberculosis vaccine (65) can induce local T-cells responses in the respiratory system, and have the ability to provide protection against pulmonary infection. In line with these finding, our study shows that, in addition to systemic responses, BCG i.n. also induced robust proinflammatory immune responses at the lung mucosa, the primary site of infection of MTBC (43). This group exhibited elevated levels of MTBC-specific proinflammatory cytokine production and increased frequencies of T-cell subsets in BALF, as well as polarization of AMs toward a proinflammatory M1 phenotype and increased activation state.

While it is hypothesized that parenteral vaccination may not effectively prime antigen-specific T-cells in the lung mucosa, thus limiting local protection, existing literature presents mixed results. Some studies suggest limited mucosal T-cell recruitment following systemic vaccination (66, 67), while others indicate that parenteral vaccination can still promote mucosal immune responses, including the induction of memory macrophages and trained immunity (68), antigen-specific tissue-resident CD4⁺ T-cell population (69, 70), and mucosal multifunctional T_EM cells in the lung (71). Following this line of evidence, in this study we observed that prime-boosted groups presented higher frequencies of antigen-specific CD4⁺ and CD8⁺ IFNγ producing T-cells isolated from BALF than HIMB-only vaccinated animals and non-vaccinated animals. Moreover, this CD4⁺IFNγ lymphocyte subsets isolated from BALF from BCG i.n. and prime-boosted groups produced higher levels of IFNγ than those from the rest of the groups. This shows that parenteral vaccines were capable of recruiting antigen-specific IFNγ producing T-cells to the lung, in a similar manner to a single-dose of intranasal BCG, while a single-dose or boost with HIMB does not impact lung recruitment of T-cells.

BCG i.n., HIMBpar (s.c.) - HIMBmuc (i.n.) and BCG (s.c.) - HIMBmuc (i.n.) groups showed higher antigen-specific proinflammatory cytokine production in BALF, than HIMB intranasally vaccinated groups, when restimulated in vitro with HIMB. Overall, ex vivo HIMB stimulation of BALF cells led to increased production of IL-1β, IL-6, TNFα, and IL-10 across all groups indicating that it induces the production of these cytokines in BALF cells regardless of vaccination status. In contrast, IL-17A and IFNγ were produced following antigen re-exposure ex vivo, as non-vaccinated and HIMB intranasally vaccinated animals did not produce significant levels of these cytokines. TNFα followed a mixed pattern, showing a slight increase upon ex vivo HIMB stimulation regardless of vaccination status, but a more pronounced, specific increase in BCG i.n. and prime boosted groups. IL-10, an anti-inflammatory cytokine, was also elevated, possibly as a regulatory mechanism to control excessive inflammation.

While most TB vaccine research over the past decades has focused on the adaptive immune response, increasing evidence suggests that innate immune mechanisms within the lung mucosa—particularly AMs, which are the first immune cells to encounter MTBC and serve as its primary host (72, 73) —may play a crucial role in the early control of infection, even before the activation of T-cell responses (32, 33, 74). Classically activated macrophages (M1) are known to have a restrictive phenotype for growth of the bacilli (75, 76), contributing to the control of mycobacterial replication. In our study, all vaccinated groups displayed elevated frequencies of markers for both activated and proinflammatory (M1-polarized) AMs phenotypes compared to non-vaccinated controls, suggesting enhanced innate immune activation induced by vaccination. M1 macrophages are characterized by the production of pro-inflammatory cytokines, their high antigen-presenting capacity, upregulation of co-stimulatory molecules involved in T-cell activation, and increased production of reactive nitrogen and oxygen species, which contribute to the macrophage’s antimicrobial activity (77–79). Here, the results indicated that vaccination promoted all these M1-associated functions, as evidenced by increased production of IL-1β, IL-6 and TNFα, elevated expression of MHC II, CD80 and CD86, and upregulation of iNOS. These findings suggest that vaccination induced a shift toward a more pro-inflammatory, bactericidal macrophage phenotype. This effect was most pronounced in the prime-boosted groups and the intranasal BCG group, where the strongest macrophage activation and M1 polarization was observed. Consistent with previous studies (68, 80–82), these results highlight the potential of mucosal and parenteral BCG vaccination to activate AMs and enhance early immune responses at the site of MTBC entry. Moreover, there was a strong positive correlation between activation and M1 markers at the individual animal level—suggesting that most activated macrophages had also adopted the M1 phenotype.

Regarding the use of the mucosal adjuvant in intranasal vaccination with heat-inactivated M. bovis, our results show no clear differences between the two intranasally vaccinated HIMB groups, suggesting that the adjuvant does not provide a marked benefit under the study conditions. However, a slight increase in CD8⁺ T-cell responses and AMs activation was observed in the HIMBmuc i.n. group compared to HIMB intranasal alone. This may suggest that a different adjuvant dose or repeated administration could be necessary to reveal more pronounced immunogenic effects of the adjuvant. In addition, HIMBmuc i.n. may have contributed to maintaining steady IFNγ levels in the HIMBpar (s.c.) - HIMBmuc (i.n.) group, potentially acting as a boost to the parenteral prime. Nonetheless, the current study design does not allow for definitive conclusions regarding the adjuvant’s boosting capacity and additional studies evaluating increased doses or repeated intranasal administrations of HIMB, with or without adjuvant, are needed.

In conclusion, intranasal BCG vaccination demonstrated strong potential as a mucosal immunization strategy against TB, effectively inducing both systemic and, notably, robust local immune responses in the lung. BCG i.n. elicited systemic antigen-specific cellular immunity comparable to that induced by parenteral BCG and HIMB vaccination. Importantly, BCG i.n. and prime-boost approaches triggered robust pro-inflammatory, antigen-specific responses within the lung mucosa, highlighting its capacity to target the primary site of infection. These results support further investigation into the efficacy of respiratory mucosal BCG vaccination to prevent spreading of early MTBC infection.