Importance of T-cell location rekindled: implication for tuberculosis vaccination strategies

Pulmonary TB remains one of the leading causes of death worldwide. The bacillus Calmette–Guérin (BCG) vaccine has been administered to infants via the skin for more than 85 years. However, the protective immunity offered by BCG only lasts for 10–15 years. This perhaps explains why BCG is effective in protecting against severe forms of childhood TB but fails to protect people into adulthood. It is currently believed that BCG or an improved BCG vaccine will continue to be used as a prime vaccine for childhood immunization, but its protective immunity ought to be boosted/prolonged by using a nonmycobacterial vaccine platform [1–3]. Our quest, therefore, is to come up with effective, safe and affordable booster vaccines for use after the initial BCG immunization. A successful boosting strategy entails the consideration of not only vaccine formulation and immunization schedule but also the route of delivery.

TB vaccines: where to administer?

In addition to the immunogenicity and immunization regimen (schedule), the route of vaccine delivery is critical to the efficacy of any given immunization program [4]. The last 10 years have seen more than a dozen promising TB vaccine candidates completing or entering early-phase clinical trials [5,6]. Some of these vaccines have entered or will soon enter more advanced clinical trials and represent those standing out from more than 200 novel TB vaccines tested in experimental models in the last two decades or so. Apart from mycobacterial organism-based platforms, the majority of these candidate vaccines are amenable only to parenteral applications (e.g., intramuscular or cutaneous injection). Although the mycobacterial organism-based vaccines can be used effectively via the respiratory mucosal route in experimental animals [7,8], they may not be suitable for respiratory mucosal application in humans due to safety considerations. Except for in a few cases [9,10], most bacterial plasmid DNA and protein-based TB vaccine formulations are poorly immunogenic when administered to the respiratory tract. The vaccine platform most amenable to respiratory mucosal application is the recombinant, replication-defective virus-vectored TB vaccine [11–17].

TB is primarily a respiratory mucosal infectious disease. Respiratory mucosal immunization was found to be generally more immune protective than parenteral immunization against pulmonary Mycobacterium tuberculosis challenge. This phenomenon alone is not surprising as the current conviction is that mucosal immunization at the site of pathogen entry confers the best level of prophylactic protection. Based on this understanding and at least from an immunologic point of view, TB boosting immunization post-parenteral BCG priming should be carried out via the respiratory tract [4,18,19].

Linking TB vaccine efficacy to T-cell location

The immunologic basis for better protection by mucosal immunization is that it elicits and retains more antigen-specific immune cells, such as T cells, at the same mucosal site of pathogen entry [18–22]. Thus, upon the deposition of M. tuberculosis bacilli and subsequent infection of alveolar macrophages and dendritic cells in the lung, antigen-specific memory T cells residing in the vicinity of infection can undergo rapid activation and act on infected cells. Since M. tuberculosis antigens take approximately 8–10 days to be available to the lymphoid tissue draining the lung [22], and recruitment of systemically located memory T cells to the lung interstitium and airway lumen requires the expression of adhesion molecules, chemokines and chemokine receptors [20], which M. tuberculosis infection may counter, having antigen-specific T cells directly elicited at the site of action helps bypass these hurdles. Such T cells operate as the on-site ‘firefighters’ who bring down the ‘fire’ (TB infection) before it becomes too violent to control.

Indeed, the current understanding is that in a naive host, it takes 8–10 days for naive T cells to be primed by mycobacterial antigens in the local draining lymph nodes following pulmonary M. tuberculosis exposure, and it takes another 5–8 days for the activated effector T cells to appear in the lung [22]. Thus, a full-blown T-cell response in the lung of naive hosts is not normally seen until 21–25 days postprimary mycobacterial infection. Although parenteral TB immunization (BCG) saves some time by skipping the T-cell priming stage, the anti-TB memory or effector memory T cells are mostly distributed in the peripheral lymphoid and nonlymphoid tissues. Thus, compared with a naive host, the T-cell responses in the lung of a BCG-immunized host to pulmonary M. tuberculosis challenge are only accelerated by a few days [22,23]. Notwithstanding, in murine models the parenteral BCG immunization is able to reduce lung M. tuberculosis infection consistently by approximately 1 log, which is being used as a ‘gold standard’ in efficacy studies for any new vaccine in question. This highlights the importance of the timing of T-cell appearance in the lung and suggests that even the acceleration by just ‘a few’ days may translate into a significant reduction in mycobacterial load in the lung.

T-cell location in the lung: the interstitium or airway lumen?

Following parenteral TB vaccination, the memory T cells populate the peripheral lymphoid and nonlymphoid tissues, including the lung interstitium. Indeed, in most instances of parenteral vaccination, the magnitude of lung interstitial T-cell responses often mirrors that of the spleen [24–27]. However, the lung can be divided into two compartments: the airway lumen and lung interstitium. The former includes both the conducting airway luminal and alveolar spaces. The latter encompasses the alveolar septum and peribronchal/perivascular connective tissues. As the murine airway hardly has the type of submucosa that its human counterpart does, the subepithelial dense connective tissue is also part of the lung interstitium. To date, the vast majority of TB vaccine studies have not yet studied the T-cell response in the airway lumen and lung interstitium separately. Given their important role in the first line of host defense against intracellular infection in the lung [20,21], we ought to pay greater attention to airway luminal memory T cells.

Airway luminal T cells, but not lung interstitial or splenic T cells, are critical to protection

Experimental studies using genetic or genetically modified cell-based TB vaccines have provided compelling evidence that a robust level of anti-TB protection in the lung is always correlated with the presence of antigen-specific T cells within the airway lumen [24–27]. And if such T cells are only present in the lung interstitium, spleen and circulation, even when in large quantities, the vaccinated host is poorly protected from pulmonary M. tuberculosis challenge. Thus, the lack of airway luminal T cells may explain why parenteral immunization with plasmid DNA, virus-vectored or cell-based TB vaccine is poorly protective [10,15,17,24,26,27]. By contrast, a single respiratory mucosal inoculation of virus-vectored or cell-based TB vaccine induces a strong presence of antigen-specific memory T cells within the airway lumen and provides robust protection from pulmonary M. tuberculosis challenge [24–27]. The maintenance of, and protection by, airway luminal CD8 T cells can be independent of systemic T-cell recruitment [Jeyanathan M et al., Unpublished Data]. While respiratory mucosal immunization also triggers the lung interstitial distribution of T cells, whether these T cells are immunologically imprinted differently from the lung interstitial T cells triggered by parenteral immunization, remains to be understood. Respiratory mucosal immunization, but not parenteral immunization, is also expected to trigger the formation of inducible bronchus-associated lymphoid tissue in the lung.

While the results from aforementioned studies promote the notion of respiratory mucosal TB immunization, they provide an important immunologic clue to developing much needed novel strategies for improving the protective efficacy of parenteral immunization [17,24,27–30]. Such new strategies will aim to mobilize the systemically located antigen-specific T cells activated by parenteral immunization into the airway lumen and subsequently sustain them. The most effective and innocuous way is perhaps to inoculate purified soluble M. tuberculosis proteins to the airway of parenterally immunized hosts [24,27]. Such proteins must be those expressed by the parenteral TB vaccine, either a genetic-based or BCG vaccine. As such soluble M. tuberculosis proteins are not adjuvant-formulated and they on their own are unable to activate T cells, this approach possesses the advantage of being ‘vaccine-less’.

The fact that the presence of systemic antigen-specific T cells serves as a poor correlate of lung protection poses a challenge to the current TB vaccine trials where the circulating T-cell reactivity is a primary readout following parenteral immunization. The only trial that evaluates the intranasal delivery of a fusion protein TB vaccine, led by scientists at the Statens Serum Institut of Denmark, shall provide viable information on the prospect of respiratory mucosal immunization in humans.

When after pulmonary M. tuberculosis exposure should the T cells be present in the airway lumen?

Ideally, the antigen-specific T cells should be in the airway lumen at the time of pulmonary M. tuberculosis exposure, as is the case for respiratory mucosal immunization. Based on the understanding that T cells will not begin converging to the lung (it is unknown whether the airway lumen was separately examined) until 14–18 days post-M. tuberculosis exposure in a naive host, whereas parenteral BCG immunization only accelerates this process by a few days thus being able to enhance protection in experimental models, any strategy that could activate and mobilize T cells into the airway lumen within perhaps the first 10 days of M. tuberculosis infection is expected to enhance immune protection. It is not yet known whether a significant number of T cells must be in the airway lumen within the first 5-day window in order to accomplish what is considered a significant level of protection (1 log) in parenterally immunized hosts. However, it was observed that pulmonary mycobacterial exposure failed to elicit a significant recruitment of systemically (peripherally) activated T cells by intramuscular virus-based TB immunization within the first 5 days [24], reaffirming the importance of early T-cell presence within the airway luminal space before mycobacterial replication within alveolar macrophages is full blown.

Closing remarks

The critical role of airway luminal T cells in TB protection has only just begun to be recognized. The presence of T cells in the lung interstitium, similar to those located in the peripheral lymphoid tissue, is a poor correlate of protection. While recent murine studies have provided the most clear-cut evidence linking the lack of airway luminal T cells to the lack of protection in parenterally genetic-immunized animals, bovine and guinea pig studies suggest the existence of interspecies difference in the property of parenterally activated T cells [31,32]. It also remains to be answered why parenteral BCG- or protein-based immunization can be more effective than parenteral genetic-based immunization. Furthermore, it remains a riddle why, contrary to the airway luminal T cells, the presence of vaccine-activated lung interstitial T cells is a poor correlate of protection in the lung. The role of these lung interstitial memory T cells in protection and their relationship with the maintenance of airway luminal counterparts are still poorly understood. It is likely that the lung interstitial T cells are unable to migrate into the airway lumen due to the lack of chemotactic signals in the early stage of M. tuberculosis infection. Alternatively, these cells may serve to only guard the interstitium and do not partake in the initial fight within the airway lumen. Lastly, the mechanisms of the maintenance of airway luminal anti-TB memory T cells also remain poorly understood.

Acknowledgements

The author thanks Mangalakumari Jeyanathan and Carly Horvath for helpful comments.

Financial & competing interests disclosure

The work cited from the author’s laboratory is supported by funds from the Canadian Institutes for Health Research. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.