Macrophages are a versatile population of immune cells that display remarkable phenotypic and functional diversity. In most tissues, including the lung , macrophages play a role in almost every aspect of tissue homeostasis by acting as sentinels and rapidly responding to inhaled challenges and physiological stresses [8, 9]. The diverse functional states exhibited by macrophages are often loosely classified as classically or alternatively activated. Classically activated macrophages, are induced by IFNγ or toll-like receptor ligands (e.g., LPS), express pro-inflammatory cytokines, (e.g., TNFα, IL-1, IL-12, and IL-6) and have enhanced antimicrobial functions, including upregulation of inducible nitric oxide synthase (iNOS). Alternatively activated macrophages include regulatory macrophages that are induced by IL-10 and produce IL-10 and CCL1; as well as wound-healing macrophages that are induced by IL-4, IL-13, and TNFα and express increased Ym1 (chitinase-like protein), RELMα (resistin-like molecule α) and arginase [2]. However, this is not a rigid classification and there is considerable evidence showing that there is extensive overlap of these functional states. It is likely that macrophages change over time and express temporally restricted phenotypes that instruct other cells through the different phases of repair [10].

Another way of classifying macrophages is based on their ontogeny and tissue residency. Several studies have recently revealed that tissue-resident macrophages (CD11c^hi CD11b^neg/low CSF1R^high CCR2^neg/low CD169^pos) are derived from embryonic yolk sac progenitors and self-renew locally in tissues in which they reside throughout adult life, whereas inflammatory macrophages (CD11c^neg/low CD11b^hi CSF1R^neg/low CCR2^high CD169^neg) are derived from bone marrow-derived hematopoietic progenitors [11–13]. In the context of tissue regeneration , evidence is accumulating to suggest that self-renewing tissue-resident macrophages are assigned the bulk of the wound-healing duties, whereas pro-inflammatory bone marrow-derived macrophages are engaged in host defense [9].

In the adult lung , macrophages display distinct location-specific phenotypes in the airway lumen, alveoli , and interstitial space. Studies in mice have shown that alveolar macrophages (AMs) comprise more than 90 % of macrophages in the adult lung [14] and belong to the tissue-resident lineage of macrophages which are derived from fetal monocytes within the first week of life and later self-renew in the adult tissue [15–17]. Mature AMs phagocytose dead cells and debris, coordinate the activation of innate and adaptive immune responses and contribute to immunosuppression associated with tissue repair . The local proliferation of AMs during homeostatic repletion is dependent on both granulocyte macrophage colony stimulating factor (GM-CSF) and colony stimulating factor-1 (CSF-1) [12].

In response to acute lung injury , the initial inflammatory response is necessary to eliminate inhaled pathogens before tissue regeneration can succeed. AMs are considered to be major effector cells in first-line innate host defense against inhaled pathogens and irritants by virtue of their phagocytic ability [18]. They mediate antimicrobial defenses through expression of receptors for immunoglobulin (F_cR), complement, β-glucan, mannose, and several types of scavenger receptors which together facilitate phagocytosis [19]. In addition, classically activated pro-inflammatory macrophages , which are recruited to the lung from the circulation, are able to generate reactive nitrogen and oxygen intermediates to defend against microbial infection [20, 21]. These defense mechanisms are boosted by epithelial-derived surfactant protein A, which stimulates macrophage phagocytosis and the production of reactive oxygen–nitrogen intermediates [22]. Recent studies using targeting strategies to selectively ablate different monocyte/macrophage populations have shown that CCR2^hi bone marrow-derived pro-inflammatory macrophages are also responsible for neutrophil emigration after acute lung injury [23].

The maintenance of the resident AM population is also important for maintaining the first-line of defense against viral and bacterial pathogens. Studies have shown that viral replication is restricted in AMs. Consequently, viral infection of AMs is nonproductive: no infectious virus is released, thereby reducing viral load [24–26]. Attesting to the central role of macrophages in defense against virus, studies have shown that depletion of resident lung macrophages from mice prior to influenza virus infection leads to increased virus titers in the lung [26–28]. However, the downside of this is that virus infection induces cell death of AMs [29]. Hence, the depletion of AMs following a significant viral infection can facilitate secondary bacterial infection by altering early innate cellular immunity. In a mouse model of secondary bacterial infection, the initial depletion of AMs as a consequence of influenza infection rendered the host susceptible to Streptococcus pneumonia colonization and systemic invasion [29]. Importantly, innate host protection to Streptococcus pneumonia was not restored until the self-renewing resident AMs population returned to normal levels 2 weeks after influenza infection [29].

Not surprisingly, given their potent antimicrobial function, impaired AM function is thought to be central to high bacterial colonization rates and increased susceptibility to exacerbations in chronic obstructive pulmonary disease (COPD). This may, at least in part, be due to excessive oxidative stress from chronic cigarette smoke exposure [30], which causes a deficiency in phagocytosis of bacteria [31] and efferocytosis of apoptotic cells [32]. Treatment with antioxidants such as procysteine can significantly improve efferocytic function of AMs isolated from experimental models of COPD [33]. Other studies have suggested that the increase in resident and inflammatory macrophage numbers in the lungs of COPD patients plays a central role in mediating collateral tissue damage [34–36]. AMs secrete elastolytic enzymes (proteases), including matrix metalloprotease (MMP)-2, MMP-9, MMP-12, cathepsin K, L, and S in response to irritants and infection, which together may be responsible for destruction of lung parenchyma [37]. Increased concentrations of these enzymes have been observed in patients with emphysema [38, 39].

Although inflammation plays an important role in eliminating pathogens, it can also cause collateral damage and limit epithelial wound healing . To progress from the inflammatory phase to the next phase of tissue repair , inflammation needs to be dampened. It is now becoming apparent that alternative activation of resident AMs plays a key role in mediating the resolution of inflammation after acute lung injury through their ability to engulf apoptotic neutrophils and debris [40, 41]. Efficient clearance of cellular debris prevents the persistence of potentially toxic and immunostimulatory material in the lung and is an important step in resolution of the inflammatory process. The active phagocytosis of dead cells also activates AMs to secrete factors like IL-1β, IL-8, TNFα, and GM-CSF that perform downstream immunomodulatory functions [42, 43]. The inability to efficiently remove exhausted neutrophils has damaging implications in COPD as accumulation of necrotic neutrophils can lead to the indiscriminate release of granule protease pools including neutrophil elastase. Neutrophil elastase localizes to lung elastic fibers in emphysema patients and degrades extracellular matrix components [44].

In addition to their immune functions, alternatively activated macrophages can also adopt a wound-healing phenotype and release a vast array of cytokines involved in tissue repair [3]. Following resolution of inflammatory responses, restoration of the cellular architecture of the lung requires the coordinated proliferation and differentiation of epithelial and mesenchymal progenitor cells . Inferring a role for macrophages in this phase of lung regeneration , several studies have shown that AMs produce cytokines that may influence the fate of adult lung epithelial and mesenchymal cells. AMs are a source of IGF-1, which can prevent the apoptosis of lung mesenchymal stromal cells [45] and stimulate their proliferation and migration [46, 47]. We have shown that resident mesenchymal stromal cells create a niche for epithelial progenitor cells , providing many of the cues necessary for their proliferation and differentiation [48–50]. Other studies have shown that following Streptococcus pneumonia infection in mice, AMs that phagocytose apoptotic neutrophils produce hepatocyte growth factor (HGF) [51]; and that the elevated levels of HGF induced by viral stimulation of AMs is associated with alveolar epithelial hyperplasia [52]. Significantly, our studies have shown that HGF is a potent epithelial mitogen that acts in synergy with fibroblast growth factor 10 (FGF-10) released by mesenchymal stromal cells to support the proliferation of lung epithelial progenitor cells [50]. IL-13, produced by alternatively activated macrophages has also been shown to stimulate release of TGF-α from lung epithelial cells, which in turn stimulates epithelial cell proliferation via binding to the epidermal growth factor receptor (EGFR) in an autocrine mechanism [53]. Interestingly, studies have also shown that GM-CSF, which is critical for self-renewal of AMs, is involved in epithelial repair in the lung [54–56]. Alternatively activated AMs also secrete large amounts of transforming growth factor-β (TGF-β), which exhibits immunosuppressive activity while also promoting myofibroblast differentiation and extracellular matrix production [57].

The persistence of alternatively activated AMs may therefore support unrestrained proliferation of mesenchymal stromal niche cells without apoptotic clearance, and thus contribute to pathologic tissue remodeling [58]. Conditional depletion of resident lung macrophages expressing a diphtheria toxin receptor under control of the CD11c promoter (CD11c-DTR) after the establishment of OVA-induced asthma, was sufficient to prevent the advancement of airway remodeling [59]. Increasing evidence also suggests that release of profibrotic factors like IGF-1, IL-13, and TGF-β by wound-healing AMs may play a key role in the development of pulmonary fibrosis , which is thought to involve the dysregulated proliferation and myofibroblastic differentiation of mesenchymal stromal cells that are involved in the regeneration and/or resolution phases of epithelial repair [60]. The fact that corticosteroids are also capable of activating the wound-healing phenotype in macrophages, may also explain why these drugs are not effective against fibrosis and may even be disadvantageous [58]. AMs in COPD also display a unique wound-healing phenotype that is thought to contribute to deleterious remodeling in the lung. Evidence suggests that the relative balance between macrophage polarization states can have a profound impact on disease progression [61]. The induction of CD163 is commonly recognized as a marker for alternatively activated macrophages involved in wound healing [62, 63] and CD163 positive macrophages are highly prominent in the lungs of current and ex-smokers with COPD [64].

In recent years, ILCs, which are newly identified members of the lymphoid lineage, have emerged as important effectors of epithelial homeostasis [7]. ILCs are primarily defined by the absence of recombination activating gene (RAG)-dependent rearranged antigen receptors, a lack of mature hematopoietic lineage markers and their lymphoid morphology. They can be divided into three subsets based on the cytokines they produce. Group 1 ILCs include conventional NK cells and are defined by the production of IFNγ. Group 2 ILCs require IL-7 for their development and secrete IL-5 and IL-13 in response to stimulation with IL-25 or IL-33. Group 3 ILCs are defined by their production of IL-17A and/or IL-22 [65, 66].

In the context of tissue regeneration , recent studies have highlighted multiple roles for ILC2s in regulating immunity and tissue homeostasis [67]. Notably, ILC2s have been shown to be a major source of IL-13 driving allergen-induced and influenza-induced airway hyperactivity [68, 69]. Moreover, intranasal administration of IL-25 or IL-33, which promotes IL-13 secretion by ILC2s, induced an asthma phenotype in naïve mice that was associated with an increase in the number of ILC2s in the lungs [70]. The release of IL-13 by ILC2s may be one of the first events driving polarization of AMs towards a wound-healing phenotype.

A number of studies have demonstrated a critical role for ILC2s in regulating lung epithelial homeostasis and tissue remodeling after inflammation or infection. For example, depletion of ILC2s after influenza virus infection using an antibody reagent targeting CD90.2 expressed on the surface of ILC2s, resulted in a breakdown in epithelial integrity and a decrease in lung function [71]. Similarly, blockade of IL-33 led to a reduction in ILC2 number and disruption to epithelial repair processes. However, epithelial repair in ILC2-depleted mice could be restored by adoptive transfer of ILC2s. Furthermore, this study also identified amphiregulin, not IL-13, as the key ILC2-derived factor driving epithelial homeostasis after influenza virus infection [71]. Importantly, amphiregulin is a ligand for the EGFR, and has also been shown to be significantly upregulated during naphthalene-induced epithelial regeneration in the lung [72].

It is also interesting to note that a recent study has shown that IL-33 is upregulated in basal cells in the inflamed or viral infected lung [73]. Basal cells are well-characterised as airway epithelial progenitor cells in the upper airways, and this observation suggests that they may serve as a renewable source of IL-33, which can in turn activate ILC2s to initiate the wound healing process and restore epithelial integrity and tissue homeostasis following influenza virus infection. These studies demonstrate a critical role for ILC2s in regulating epithelial homeostasis and tissue remodeling after inflammation or infection.

In this review, we have discussed the emerging evidence that innate immune cells play a critical role in the regulation of lung regeneration and repair . In addition to their role as first-line defenders against inhaled pathogens, innate immune cells function as niche cells in the adult lung by providing direct and indirect support for tissue repair and regeneration. During the resolution phase of wound healing , resident AMs play a crucial role in the phagocytosis of apoptotic neutrophils, and the release of cytokines that enhance the regenerative capacity of the epithelial–mesenchymal trophic unit. There is also emerging evidence that a relatively newly discovered population of mucosal immune cells (ILC2s), may also play a key role in stimulating the wound-healing process after inflammation or infection. Although we still have much to learn about the mechanisms by which innate immune cells interact with regenerative stem and progenitor cells in the lung, it is clear from these limited studies that they are not simply onlookers and have central roles in regulating tissue homeostasis. Moreover, if left unchecked they may be directly responsible for driving pathogenic remodeling in lung diseases.