Lung Tissue-Resident Memory T cells: The Gatekeeper To Respiratory Viral (Re)-Infection

Abstract

The discovery of lung tissue-resident memory T (T_RM) cells and the elucidation of their function in antiviral immunity have inspired considerable efforts to leverage the power of T_RM cells, in defense to the infections and reinfections by respiratory viruses. Here, we have reviewed lung T_RM cell identification, molecular regulation and function after influenza and SARS-CoV-2 infections. Furthermore, we have discussed emerging data on T_RM responses induced by systemic and mucosal vaccination strategies. We hope that our current outstanding of T_RM cells in this review could provide insights towards the development of vaccines capable of inducing highly efficacious mucosal T_RM responses for protection against respiratory viral infections.

Introduction

Respiratory viral infections are a leading cause of mortality, causing an estimated 2 million deaths per year[1]. Respiratory viruses can cause catastrophic viral pandemics evidenced by the 1918 H1N1 pandemic and the current COVID-19 pandemic, which have claimed more than 40 and 6 million lives respectively[2,3]. Long term protection against reinfection by the same or similar viruses is mediated by memory T and B cells, as well as antibodies produced by long-lived plasma cells. Pathogens like influenza virus and SARS-CoV-2 have the ability to change their surface antigens frequently, escaping antibody-mediated viral neutralization. In this case, memory T cells are believed to play a crucial role in providing effective protection against viral dissemination and the development of severe diseases[4].

The respiratory mucosal interface is the front line encountering pathogens that spread via airborne transmission. Following infection, a subset of memory T cells termed tissue-resident memory T (T_RM) cells, are retained in the respiratory tract. T_RM cells are phenotypically and functionally different from the circulating central memory T (T_CM) and effector memory T (T_EM) cells[5]. Numerous reports have already demonstrated that compared to the circulating T_CM and T_EM cells, T_RM (both CD4⁺ and CD8⁺) play a pivotal, if not dominant, role in the protection against secondary viral infection in the respiratory tract[5]. As the local sword, T_RM cells rapidly respond upon pathogen invasion, killing virus-infected cells directly and activating a series of downstream immune events to impede viral replication and dissemination. Despite the important roles of T_RM cells in protection against respiratory infection and re-infection, induction of strong antigen-specific T_RM cells is not usually exploited by most available or candidate vaccines. It is thus needed to make more efforts on the development of T_RM–based mucosal vaccines against the infection and transmission of respiratory viruses, in particular influenza and SARS-CoV-2[6]. In the following sections, we will mainly focus on the regulation and protective functions of CD8⁺ and CD4⁺ T_RM in the contexts of influenza and SARS-CoV-2 natural infections or vaccinations.

Lung T_RM identification and regulation

More than a decade ago, pioneering work on vesicular stomatitis virus (VSV) and listeria monocytogenes revealed the presence and persistence of non-circulating resident memory T cells in non-lymphoid organs after the resolution of the primary infection[7]. Almost at the same time, the presence of antigen-specific memory T cells was identified in the lungs following Sendai virus, influenza and respiratory syncytial virus (RSV) infections in mice[8,9]. Since then, the development of lung T_RM cells has been observed in the context of almost all major respiratory viral, bacterial, fungal and parasitic infections. Aside from flow cytometric staining of T_RM surface markers, techniques such as intravascular labeling, parabiosis, and tissue transplantation have greatly facilitated T_RM research in animal models[10]. The identification and study of human respiratory T_RM cells are emerging but remain challenging due to technical difficulties and the availability of human tissues.

Upon respiratory infection, naïve T cells are generally primed in the lung draining mediastinal lymph node (mLN), where they encounter antigen carried by antigen presenting cells (APCs) migrated from the lungs. Primed effector T cells then egress from the mLN and enter the lungs to combat the infection. After viral clearance, T_RM cells emerge and persist in the lungs following contraction of the majority of effector T cells. Compared to the circulating counterparts (T_CM and T_EM cells), T_RM cells highly express the transcriptional factors Blimp1, Hobit, Runx3, Bhlhe40, Notch, while downregulate the expression of KLF2, T-bet, Eomes and TCF1[11]. As a result, surface markers that maintain tissue localization and retention are upregulated. Most notably, the expression of CD69, which antagonizes the function of tissue-exiting molecule S1PR1, is uniform between mouse and human lung T_RM cells, rendering its expression as the most distinctive feature of T_RM cells compared to circulating memory cells. A group of T_RM cells also express integrins CD103 and/or CD49a, further supporting their retention by interaction with epithelial E-cadherin and collagen in extracellular matrix respectively. Respiratory T_RM cells also express the chemokine receptor CXCR3, allowing them to enter the lungs and be trapped by the local chemotactic ligand CXCL9/10/11 gradient[11]. Immunohistochemistry studies revealed the location of CD8⁺ T_RM cells, mostly outside but also neighboring the area of inducible bronchus-associated lymphoid tissue (iBALT) structures in the lung parenchyma[12]. In addition, CXCR6 expression by T_RM cells further promotes their localization into the airway[13,14] (Figure 1).

Figure 1. Lung T_RM cell generation, maintenance and function.

After respiratory viral infection (such as influenza and SARS-CoV-2) or intranasal vaccination, dendritic cells carrying viral antigen migrate from lung to draining mediastinal lymph nodes (mLNs), where naive T cells are activated. Primed effector T cells then egress from the mLNs and migrate to the lung for viral clearance. During this process, T_RM precursors highly express transcriptional factors Blimp1, Hobit, Runx3, Bhlhe40 and Notch, while downregulate KLF2, T-bet, Eomes and TCF1. As a result, CD69, CD103 (selectively) and CXCR3 are upregulated to assist T_RM cells localization and retention in the lung. In addition, CXCR6 expression further promotes T_RM cell localization into the airway. In this context, CD4₊ T_RM cells can differentiate into IFN-γ- producing CD4₊ T_RM1 and IL-21-producing CD4₊ T_RH cells, with the latter help IFN-γ-producing CD8₊ T_RM cell maintenance and local B cell responses. CD8₊ T_RM cells can also kill the infected cells.

Besides chemotactic and retention signals, lung T_RM cell programming and/or maintenance are also modulated by a variety of local tissue signals. For instance, cytokines such as IL-33, TGF-β, type I interferons and TNF are important to program lung T_RM cell development, while IL-7 and IL-15 have been shown to be important for the maintenance of lung T_RM cells[11]. Of note, it is also required for T_RM cell development to re-encounter antigen locally after effector T cell migration into the lungs[15,16]. Additionally, the maintenance of a subset of lung PD-1⁺ T_RM cells has been shown to be dependent on the persistent antigen stimulation in the memory phase long after the clearance of infectious virus[17]. Interestingly, the oxygen-rich and nutrient-deprived local environment in the respiratory tract imposes high levels of cellular stress in T_RM cells[18]. Lung T_RM cells are relatively short-lived compared to T_RM cells in other organs such as the skin[18,19]. The limited interstitial space, enhanced T_RM cell death, removal by mucociliary elevator and/or transportation via retrograde migration are likely the reasons for the steadily diminishment of T_RM cells in the respiratory tract compared to those in other organs [20]. As a result, CD8⁺ T_RM in the lung can last for approximately 200 days in the mice after viral infection, while variable data shows over a year in human following transplantation[21,22]. Such a short lifespan of lung T_RM cells poses unique challenges for the development of T_RM-based vaccines for combating respiratory viral infection.

Lung CD8⁺ T_RM function following influenza and SARS-CoV-2 infection

The roles of lung CD8⁺ T_RM cells in protection against respiratory viral infections have most extensively been studied in the murine influenza infection model. It is now clear that CD8⁺ T_RM cells are vital for protection against influenza re-infection, particularly in the context of heterotypic influenza viruses that escape pre-existing antibodies[23]. CD8⁺ T_RM cells are able to produce large quantities of cytokines such as IFN-γ conferring this protective function. IFN-γ blockade significantly decreased the survival rate in mice[24]. Similarly, these virus-specific T_RM cells were also observed in human lungs[25], exhibiting high proliferative capacity and poly-functionality after ex vivo stimulation. Further T cell receptor (TCR) repertoire analysis indicated that the diversity of TCRαβ was maintained within the pool of lung CD8⁺ T_RM cells, compared to other memory T cell populations in the lung [26]. Importantly, these virus-specific T_RM cells also showed cross-reactivity against different influenza strains as observed in mice[24,26,27].

Regarding SARS-CoV-2 infection, the unprecedented efforts during the COVID-19 pandemic have shed considerable insight into human CD8⁺ T_RM responses following infection. Early single-cell RNA sequencing of cells in bronchoalveolar lavage (BAL) fluid from patients with acute COVID-19 have identified the presence of effector CD8⁺ T cells and potential CD8⁺ T_RM precursors after infection[28]. Later, our group found that increased antigen-specific CD8⁺ T_RM cells can be detected for at least 2–3 months in the respiratory tract after the recovery from severe COVID-19[29]. The presence of CD8⁺ T_RM cells in the lung and potentially other organs after acute SARS-CoV-2 infection were subsequently observed in a number of studies[30–34]. Strikingly, these CD8⁺ T_RM cells in the lung produce IFN-γ after antigenic stimulation and can persist up to 10 months after initial infection[30], suggesting a potential role for T_RM cell-mediated long-term protection against SARS-CoV-2 reinfection in COVID-19 convalescents. Emerging animal studies have shown that antigen-specific CD8⁺ T cells capable of producing effector cytokines are maintained in the lungs after SARS-CoV-2 viral clearance[35,36]. Notably, lung CD8⁺ T cells can be detected up to 120 days post infection with a mouse-adapted SARS-CoV-2 strain (MA10) in mice[37]. The protective function of CD8⁺ T_RM cells in the context of COVID-19 remains to be firmly established. One study using the K18-hACE2 mouse model concluded that CD8⁺ T_RM cells are dispensable for SARS-CoV-2 protection[38]. However, SARS-CoV-2 mortality in K18-hACE2 mice may not be dependent on the pulmonary pathology[39], which puts the utility of the model for studying T_RM-mediated protection into the question. To this end, CD8⁺ T_RM cells were shown to provide substantial protection from lethal infection by an attenuated mouse-adapted SARS-CoV strain (MA15) in mice[40].

In certain circumstances, CD8⁺ T_RM cells are not always “the more, the better”[41,42]. In aged COVID-19 convalescents, the levels of CD8⁺ T_RM cell number in the respiratory tract appear positively correlated with the post COVID-19 lung sequelae [29,34]. Such a phenomena is consistent with the case of influenza-infected aged mice[43]. Late depletion of CD8⁺ T_RM cells after viral clearance decreased chronic lung pathology and improved lung function in influenza-infected aged mice[29]. Hence, it is important to promote CD8⁺ T_RM cell-mediated protective function, while minimizing their pathogenic potential to avoid chronic lung pathology.

Lung CD4⁺ T_RM function following infection

Unlike CD8⁺ T_RM cells, CD4⁺ T_RM cells are generally more diverse, serving divergent functions against a variety of infections. Similar to effector CD4⁺ T cells, CD4⁺ T_RM cells can be divided into different subsets based on their cytokine production, including IFN-γ-producing T_RM1, IL-4/5/13-producing T_RM2, IL-17-producing T_RM17 cells and IL-21-producing resident helper (T_RH) cells[44,45]. Additionally, Foxp3-expressing regulatory T (T_REG) cells can also gain lung residency and become resident tissue T_REG cells[44,45]. In the context of respiratory viral infection, T_RM1 and T_RH cells are usually the dominant CD4⁺ T_RM cell responses in the lung[16,46].

CD4⁺ T_RM cells have been shown to provide protection via a variety of potential mechanisms in response to heterosubtypic influenza infection[46–48]. For instance, CD4⁺ T_RM1 derived IFN-γ promotes an antiviral state in the lung and directly contribute to influenza clearance[49,50]. T_RH cells localizing within the iBALT can assist in CD8⁺ T_RM cell maintenance, providing the optimal local T cell mediated protection[46]. Moreover, T_RH cells also aid local humoral responses in the lung by augmenting the development of tissue resident memory B (B_RM) cells and local antibody production[16,46].

Our current understanding of CD4⁺ T_RM cell response following SARS-CoV-2 infection is relatively limited. Data from both human and mouse studies have demonstrated CD4⁺ T cells can persist in the respiratory tract for a period of time after infection, even up to 120 days in the lung post infection with MA10 in mice[28–33,37,51]. Occasionally, pre-existing CD4⁺ T_RM cells could be also detected from respiratory tract of unexposed individuals[52], although the function of these cells has not been determined. In this regard, previous studies on SARS-CoV infection showed that CD4⁺ T_RM cells can confer protective immunity against re-infection via IFN-γ production[53]. It is thus reasonable to assume that CD4⁺ T_RM cells may mediate similar function after SARS-CoV-2 infection. Whether SARS-CoV-2 infection promotes the development of T_RH cells is still unknown. Notably, SARS-CoV-2 infection could induce IL-17A- and GM-CSF-producing CD4⁺ T_RM17 cells in the BAL, which may potentially contribute to pulmonary hyperinflammation observed in severe COVID-19 individuals[51].

Lung CD4⁺ and CD8⁺ T_RM interaction

Pioneering work two decades ago have established an essential role of CD4⁺ T cell help in establishing CD8⁺ memory T cells[54,55]. Similarly, it was demonstrated that early CD4⁺ T cell help is important for the formation and localization of CD8⁺ T_RM cells in the lung following influenza infection[48]. In the absence of CD4⁺ T cell-derived IFN-γ, “unhelped” CD8⁺ T_RM cells downregulate CD103 expression and mis-positioned away from the airway epithelium, failing to provide optimal protection against heterosubtypic viral challenge[48]. Recently, we found that late local CD4⁺ T cell help, mediated mainly by T_RH cells, is critical for long-term maintenance of a subset of PD-1^hi CD8⁺ T_RM cells in the lung[46]. This help is mainly dependent on T_RH cell-derived IL-21, and IL-21R blockade after viral clearance resulted in decreased PD-1^hi CD8⁺ T_RM response[46]. In addition, T_RH ablation also impaired B_RM responses in the lung and decreased antibody secreting cells (ASCs) following viral rechallenge[16,46]. Interestingly, T_RH development also requires B cell presence[16]. Altogether, these data suggest an interactive resident lymphocyte network co-operates to promote optimal local immune memory development and/or maintenance after viral reinfection.

Induction of lung T_RM response after vaccination

Given the power of lung T_RM cells in conferring protection, vaccine strategies that can promote robust T_RM responses are expected to provide superb protection against respiratory pathogens[6]. Unfortunately, the current intramuscular inactivated influenza vaccines fail to generate notable resident immunity in the lung[56]. Conceptually, intranasal live-attenuated influenza virus vaccines (LAIV, FluMist) are thought to be better in generating T_RM and mucosal antibody responses, which would result in higher protective efficacy compared to inactivated vaccines in children[57]. In spite of this observation in children, the efficacy of LAIV in adults decreases, evidenced by the inefficient induction of immune responses. This is likely due to diminished LAIV infection in adults as a result of pre-existing anti-influenza immunity[58]. The drawback of LAIV might be potentially overcame by future mucosal protein- or DNA/mRNA-based nanocarriers vaccines, which can be adjuvanted specifically to enhance the immunogenicity in the respiratory tract. The small size of nanocarriers with positive charge may potentially enhance the adhesion of antigens to the mucosa, avoid the physical and biological elimination[59,60], and facilitate the transfection into or taking up by targeted cells including APCs. These strategies are expected to provide better mucosal protection via the induction of strong T_RM responses.

Current approved COVID-19 mRNA vaccines were demonstrated to provide high protection against symptomatic infections, mediated by strong circulating neutralizing antibody (nAb), as well as memory T and B cell responses[61]. However, within the respiratory tract, these immune responses were found to be suboptimal compared to COVID-19 convalescents[62,63]. Consistently, recent studies demonstrated resident T cells and nAb in nasal mucosa could be only detected after SARS-CoV-2 breakthrough infection (infection in vaccinated individuals), but was extremely rare in vaccinees alone[64–66], suggesting that intramuscular immunization alone might not prevent infection and transmission efficiently. Remarkably, low dose of intramuscular mRNA immunization in mice is also inefficient in generating resident T cell response in the respiratory tract[62,67,68]. However, high doses of intramuscular mRNA immunization may generate antigen-specific resident T cell in the lung[69,70], indicating that the dose of peripheral immunization may dictate T_RM generation in the lung, but potential side effects with high dose mRNA vaccination should be considered.

Recently, two COVID-19 intranasal vaccines have been approved in China and India. Both vaccines use the adenoviral vector-based vaccine platform, and appear to be well-tolerated[71]. Intranasal adenoviral immunization promotes strong mucosal humoral and T_RM responses in animal models[62,72]. Therefore, it is tempting to speculate that these two COVID-19 intranasal vaccines may provide more effective protection against SARS-CoV-2 variants than the intramuscular ones, although the real-world efficacy remain to be determined. More recently, results from one small-scale clinical trial of intranasal delivery of approved adenoviral vector-based vaccine ChAdOx1 nCoV-19 (AZD1222) have also shown acceptable tolerability, but exhibited relatively weak immunogenicity even as an intranasal booster [73]. Thus, different formulations or subtypes of adenoviral vectors may have differential immunogenicity in the respiratory mucosa. In addition to adenoviral vectors, many other types of mucosal vaccine formulations are ongoing in clinical development including protein subunits, attenuated viruses and other viral vector-based vaccines [6]. Nevertheless, a “hybrid” systemic prime and mucosal boost strategy is expected to induce both long-lived circulating memory T cells and robust mucosal immunity (which are likely shorter-lived than circulating immunity), ultimately providing highly efficacious and long-lasting immunity against future viral infections.

Conclusion

Since the discovery of lung T_RM cells, great advances have been made in the last decade regarding their characteristics, molecular regulation and protective function. Emerging studies have also begun to elucidate mechanisms modulating T_RM heterogeneity and longevity in the lung. It is the time for the field to take advantage of what we have learned and devote even greater efforts for rational design of future vaccination strategies capable of harnessing the power of T_RM cells for viral protection. Meanwhile, we also should not ignore the other side of the coin that excessive T_RM cell activation may cause lung immunopathology and tissue destruction. Therefore, it is critical to harness T_RM protective function while minimizing their pathogenic potential during the development of mucosal vaccines aiming to promote robust respiratory T_RM responses.

Highlights.

• Tissue-resident memory T (T_RM) cell development following influenza or SARS-CoV-2 infection.

• Lung T_RM cell characteristics, mechanisms of regulation and function.

• Robust T_RM cell response confers optimal protection against heterosubtypic viral infection.

• Intranasal booster immunization is a promising strategy to augment mucosal immunity.