Recent Advances in Lung Immunobiology

Recommended Reading from the Emory University Pulmonary and Critical Care Medicine Research Fellows

David A. Schulman, M.D., M.P.H., Program Director, and F. Eun-Hyung Lee, M.D.

Allie SR, et al. The Establishment of Resident Memory B Cells in the Lung Requires Local Antigen Encounter. Nat Immunol (1)

Reviewed by Richard P. Ramonell

Memory B cells differentiate from naive B cells after antigen encounter and are capable of producing immunoglobulins with a high antigenic specificity in response to a second antigenic challenge. These cells are critical to a secondary immune response because they migrate to mucosal surfaces, where antigen is more likely to be encountered, and elicit a humoral immune response to reinfection more rapidly than during primary infection (2). Influenza-specific tissue-resident memory T cells can reside in the lung, but it is uncertain if memory B cells are drawn into sites of active inflammation to proliferate or if they can exist in a tissue-resident phenotype (3). CD27⁺ B cells can be found in multiple human tissues, but to date no experiment has shown whether a distinct influenza-specific resident memory B-cell (BRM) population exists and, if so, to what degree it contributes to local and systemic humoral immunity (4, 5).

In their article published in Nature Immunology (1), Allie and colleagues demonstrate not only that BRM cells are distinct from circulating memory B cells but also that these cells’ differentiation into antigen-specific antibody-secreting cells (ASCs) after secondary infection requires local antigen encounter. They also show that BRM cells play an important role in protective immunity with secondary viral infection. These investigators designed a model of influenza infection in C57BL/6 mice using influenza nucleoprotein and hemagglutinin protein tetramers conjugated to fluorochromes that allowed detection in flow cytometric assays.

With this platform, noncirculating BRM cells isolated from mouse lung tissue expressed lower concentrations of the lymph node homing receptor CD62L and higher concentrations of CXCR3, suggesting that the antigen-specific, noncirculating lung BRM cells are programmed to home to the respiratory parenchyma and reside there after antigen encounter. These BRM cells rely on germinal center CD40-dependent mechanisms before homing to lung parenchyma and appear as early as 10 days after infection.

Next, to see if BRM cells recirculate, CD45.2⁺ mice were surgically paired with CD45.1⁺ mice 44 days after influenza infection. After circulating lymphocytes were equilibrated, the authors could not find CD45.1 influenza-specific BRM cells in CD45.2 lungs, showing that BRM cells in the lung do not recirculate. Surgically paired mice were then infected with different strains of influenza (H1N1 and H3N2). Although both groups of mice had similar numbers of influenza nucleoprotein-specific BRM cells, which would be the same with either infection, the lungs of H1N1-infected mice had a greater abundance of H1N1-specific BRM cells and those of H3N2 mice had a greater abundance of H3N2-specific BRM cells, further confirming that antigen-specific pulmonary BRM cells do not migrate after local antigen encounter in the lung.

Finally, mice infected with influenza virus either in the peritoneal cavity or intranasally were then subjected to a second influenza challenge infection 30 days later. Intranasally primed mice exhibited decreased symptoms of infection with lower viral titers and also possessed higher numbers of antigen-specific ASCs. Addition of an inhibitor of lymphocyte recirculation had no effect on the antigen-specific ASC response in the lung, implying that BRM cells require local antigen encounter and will rapidly differentiate into ASCs after secondary infection.

After the discovery of tissue-resident memory T cells in 2001, the existence of tissue-resident memory B cells remained unknown (6, 7). This new study elucidates the characteristics, origins, and function of a novel layer of protective memory B cells localized in infected tissues, the BRM cells. This early local ASC differentiation for immediate local antibody production is protective and important, but whether vaccination can induce BRM cells and whether BRM cells occur in humans will need further evaluation.

Yao Y, et al. Induction of Autonomous Memory Alveolar Macrophages Requires T Cell Help and Is Critical to Trained Immunity. Cell (8)

Reviewed by Zohra Prasla

The classical view of the immune system as divided into the innate and adaptive components has been challenged in recent years. The innate system consists of the early, nonspecific response to infection and is required, after initial priming, for development of adaptive memory B and T cells in vertebrates (9, 10). However, more recent data have shown that cells of the innate immune system, including dendritic cells, macrophages, and monocytes, can be programmed to augment host defenses by increasing cytokine production upon restimulation in a process known as “trained immunity” (11). In particular, alveolar macrophages (AMs) play a role in adaptive immunity through CD4⁺ T-cell–mediated processes, especially in response to vaccinations (12), but the mechanisms by which they are programmed into memory AMs involved in trained immunity remain elusive.

In their recent paper in Cell (8), Yao and colleagues identified mechanisms that prime memory AMs to acquire a program of trained immunity using various murine models. First, they infected BALB/c mice intranasally with an adenoviral vaccine vector and analyzed AMs from the BAL fluid. Memory AMs had higher levels of major histocompatibility complex II expression and upregulation of genes involved in neutrophil-recruiting chemokines, such as MIP-2 (macrophage inflammatory protein 2) and keratinocyte chemoattractant. To understand if the AMs were of lung-resident or monocytic origin, Siglec-F and CCR2 were examined, respectively. Similarly high levels of Siglec-F expression were noted after infection, whereas expression of CCR2 was minimal, showing that the origin of memory AMs was lung resident. These findings were corroborated in infected CCR2^−/− mice, confirming that memory AM generation remained intact without CCR2. Then, through parabiosis experiments, the authors showed that circulating monocytes did not contribute to memory AM generation or maintenance after infection. Finally, using bone marrow chimeras, the authors demonstrated that memory AMs are not of bone marrow origin.

To investigate the role of adaptive immunity on memory AM generation, CD4⁺ and CD8⁺ T cells were depleted after viral infection. The lack of both CD4⁺ and CD8⁺ T cells led to loss of memory AMs, whereas depleting only CD4⁺ cells had no effect. In contrast, CD8⁺ T cells played a profound role in the priming of memory AMs. Interestingly, late depletion of CD8⁺ T cells did not decrease memory AMs, suggesting that CD8⁺ T cells affect generation but not maintenance of memory AMs. Furthermore, memory AM formation depended on IFN-γ produced by CD8⁺ T cells in a direct cell–cell contact mechanism demonstrated by Transwell experiments.

To understand the function of memory AMs, the authors assessed survival of adenovirus-exposed BALB/c mice after a lethal dose of Streptococcus pneumoniae infection. Compared with control animals, adenovirus-infected mice survived with decreased bacterial loads, indicating that memory AMs conferred antibacterial immunity. Depletion of neutrophils attenuated this protection, as did injection of anti–MIP-2 antibodies, demonstrating that mechanisms of memory AM protection occurred through MIP-2– and keratinocyte chemoattractant–mediated increase in neutrophil chemotaxis.

This study shatters the classical view of a unidirectional influence of innate to adaptive immunity and elucidates novel reciprocal mechanisms between innate and adaptive responses. The authors demonstrate the functional importance of immunologic memory, or trained immunity, of innate immune cells using mucosal memory AMs generated by IFN-γ stimulation from CD8⁺ T-cell–cell contact. Recently, IFN-γ–induced priming of macrophages was shown to augment immune response upon restimulation in cryptococcal infection (13). Yao and colleagues’ work is unique in that a respiratory viral infection led to trained immunity against subsequent bacterial pneumonia in mice. Thus, these results have important implications for new vaccine strategies and therapeutic approaches to infection. Whether these results are universal to other respiratory viruses, such as influenza, rhinovirus, and respiratory syncytial virus, remains undetermined, but the prospect of attenuating the effects of microbial infection by inducing untargeted trained immunity is intriguing because it suggests that a patient’s viral infection history may be protective against other respiratory infections.

Jin C, et al. Commensal Microbiota Promote Lung Cancer Development via γδ T Cells. Cell (14)

Reviewed by Charles R. Terry

γδ T cells, CD4⁺ regulatory T cells, and CD4⁺ T-helper cell type 17 (Th17) cells partially mediate the immune responses to lung cancer that promote or inhibit tumor growth depending on their cytokine expression profile in the tumor microenvironment (15–17). γδ T cells have unique capacities that allow them to function as part of both the innate and adaptive immune responses (18). They are particularly enriched at mucosal surfaces and are capable of directly killing tumor cells via major histocompatibility complex–independent antigen activation via γδ T–cell receptors and natural killer receptor–mediated activation (16). Often, regulatory T cells and Th17 cells counterbalance antitumor effects by inhibiting immune-mediated tumor cytotoxicity, promoting tumor invasion, and angiogenesis via local IL-17–mediated inflammation (15). Whether this protumor inflammation is a consequence of tumor-mediated signaling or rather an effect of extratumoral host or environmental factors modifying the tumor microenvironment is unknown.

In a recent paper in Cell (14), Jin and colleagues showed that tumor growth correlated with lung bacterial burden and less local biodiversity through γδ T-cell–dependent signaling. They used an autochthonous conditional KRAS^LSL-G12D driver mutation and p53^fl/fl-knockout mouse as a model of lung adenocarcinoma. Intratracheal infection with an adenovirus vector expressing Cre-recombinase under the Sftpc (surfactant protein C) promoter activated KrasG12D and deleted tumor suppressor p53 in lung epithelial cells to induce lung adenocarcinoma. Results showed that local lung microbiota-dependent inflammation promoted tumor progression via γδ T-cell–dependent signaling. Depleting local microbiota with broad-spectrum antibiotics or inhibiting γδ T cells or γδ T-cell–dependent chemokines suppressed tumor growth. Mice raised in the pathogen-free environment (i.e., presence of commensal microbes but free of pathogenic microbes) had significantly more tumor growth than mice raised in germ-free environments (i.e., free of all microbes). Moreover, when germ-free mice were placed in cages after tumor induction with pathogen-free littermates, their tumor growth accelerated. Conversely, broad-spectrum antibiotic administration in pathogen-free mice effectively suppressed tumor growth, and higher bacteria levels in the lungs were associated with larger tumors.

The local lung tumor environment was enriched for γδ T cells in these mice raised in pathogen-free environments. These local γδ T cells expressed IL-17 and RORγt receptor, an isoform of the retinoic acid orphan receptor that mediates Th17 and Th17-like differentiation. Exposing γδ T cells to peptidoglycan and LPS increased IL-17 expression as well. Conversely, inhibiting IL-17 or γδ T cells successfully suppressed tumor growth and neutrophil recruitment. Together, these findings suggest that bacteria-mediated inflammation in the tumor microenvironment of lung adenocarcinoma promotes differentiation of local γδ T cells to produce IL-17, leading to faster tumor growth.

This study offers insights into the interactive role of IL-17, the respiratory microbiome, and γδ T cells in the growth and development of lung adenocarcinoma. IL-17 remains essential for protection against extracellular defense (19); yet, overexpression or chronic IL-17 activation leads to tumor progression. Although IL-17 inhibitors are currently used in practice for psoriatic arthritis and ankylosing spondylitis, their therapeutic potential in lung cancer remains promising yet unexplored (20). Chang and colleagues found similar results with IL-17 upregulation via local Th17 cells driving tumor growth in a similar KrasG12D mouse model. Their study used a different inflammatory stimulus (lysate of nontypeable Haemophilus influenzae) that did not increase expression of IL-17–producing γδ T cells (21). In contrast, exogenous IL-17 can have opposite effects and suppress tumor growth in a Lewis lung carcinoma model of a mouse lung cancer (22). Nonetheless, Jin and colleagues’ results support microbial inflammation in the tumor microenvironment as a novel mechanism of lung cancer progression. Collectively, these studies provide new approaches to exploit the lung inflammatory responses, especially by inhibiting IL-17 to its local microbiome with its diversity, to help shape effective tumor immunity. In summary, this study illustrates potentially innovative targets that could affect human lung cancers and lung metastasis, but more studies are needed to determine the role of IL-17 in promoting or inhibiting tumor growth in localized and metastatic models of lung cancer.