From virus to inflammation, how influenza promotes lung damage

Abstract

Despite seasonal vaccines, influenza related hospitalization and death rates have remained unchanged over the past five years. Influenza pathogenesis has two crucial clinical components; First, influenza causes acute lung injury that may require hospitalization. Secondly, acute injury promotes secondary bacterial pneumonia, a leading cause of hospitalization and disease burden in the United States and globally. Therefore, developing an effective therapeutic regimen against influenza requires a comprehensive understanding of the damage-associated immune-mechanisms to identify therapeutic targets for interventions to mitigate inflammation/tissue-damage, improve antiviral immunity, and prevent influenza-associated secondary bacterial diseases. In this review, we highlight the pathogenic immune mechanisms implicated in acute lung injury and the possibility of using lung inflammation and barrier crosstalk for developing therapeutics against influenza.

INTRODUCTION

Besides the fatal pandemics that plagued the 20^th century, seasonal influenza strains have been responsible for infecting over 35 million people and have resulted in over 30 thousand deaths during the 2018 and 2019 influenza seasons in the United States [1, 2]. Currently, seasonal tetravalent influenza vaccines allow for moderate protection against a range of influenza A (IAV) and influenza B virus strains [3, 4]. However, in the most recent influenza season (2018–2019), the Center for Disease Control and Prevention (CDC) estimated the vaccine efficacy to be 47%, and collectively, the effectiveness of seasonal influenza vaccines range between 10–60% [5]. Additionally, several antiviral treatments exist with varying efficacies ranging from 60–90%, but this estimate is based on providing antiviral therapy within the first 48 hours of disease onset [6, 7]. Still, despite seasonal vaccines and antiviral treatments, influenza hospitalization and death rates have remained unchanged over the past five years (Figure 1).

Figure 1.

Incidence of Influenza virus infections for the years 2014 to 2019. Data courtesy of the CDC.

Influenza is an enveloped virus that belongs to the family, Orthomyxoviridae [8]. The IAV genome contains eight, negative-sense, single-stranded RNA segments that encode between 11 and 14 proteins depending on the strain of virus [8, 9]. The envelope contains two major glycoproteins, hemagglutinin (HA) and neuraminidase (NA), which surround the virion’s genome [8, 9]. HA is the major glycoprotein responsible for viral entry into cells of the upper respiratory tract by attaching to sialic acid residues [10, 11]. Influenza infections initiate in the upper respiratory tract, where epithelial cells are infected, resulting in producing mucous and inducing the innate immune response [12]. Viral-damaged airway epithelial cells recruit immune effector cells such as neutrophils, monocytes, and macrophages that contribute to hyper-inflammation and local tissue-damage [12, 13]. In immune-competent hosts, the virus spread is contained within the upper respiratory tract, resulting in mild diseases. Viral dissemination from nasal tissues to different lung cavities depends on several factors, including respiratory conditions, age, and immune status [14–17], where the virus spreads to the alveoli and manifests as viral pneumonia. Together, epithelial cell death, hyper-inflammation, and increased airway permeability result in acute lung injury that may require hospitalization [18].

While both host response and viral antigens contribute to dysregulated lung inflammation and tissue-damage, upper respiratory (nasopharyngeal or NP) commensal bacteria may also have a significant role in exacerbating inflammation and causing secondary bacterial pneumonia, that could potentially develop into life-threatening bacterial sepsis [19, 20]. While Streptococcus pneumoniae (Spn) and Staphylococcus aureus (SA) are the most commonly associated bacteria with a secondary bacterial infection in influenza-infected hosts, other respiratory bacteria such as Haemophilus influenzae, Legionella pneumophila, and Pseudomonas aeruginosa, have also been reported to cause severe bacterial pneumonia in a co-infection setting with influenza virus [19]. While commensal bacteria often colonize the NP mucosal surfaces, the lung microbiome has also received significant attention in recent times [21]. Recent data suggest that the host response to influenza and Spn co-infection is likely modulated by lung microbiome in a model of chronic allergic inflammation [21]. Additionally, the lung microbiome can induce IgG and IgA-specific responses, which protect the respiratory epithelium from secondary bacterial infections after IAV infection. Therefore, given the rampant use of antibiotics, the complex dynamics of lung microbiome and host response may alter host defense against co-infecting pathogens.

Spn is most commonly associated with community-acquired bacterial pneumonia and sepsis, which account for the highest morbidity and mortality globally [22–24]. Individuals over 65 years are more likely to experience bacterial pneumonia with influenza acting as a significant risk factor for the development of these diseases [25]. While efficacious against bacteremia, the effectiveness of the 23-valent pneumococcal vaccine (23vPPV or Pneumovax) against pneumococcal pneumonia in people over 65 years is variable [25, 26]. Due to the lack of a vaccine, SA is increasingly being reported to cause severe bacterial pneumonia in a co-infection setting with influenza. Over the past decade, methicillin-resistant S. aureus (MRSA) has become a major cause of pneumonia in the US particularly associated with influenza. Necrotizing pneumonia due to community-acquired MRSA carries a mortality of up to 30% [27–29]. Overall, influenza mediated dysregulation of inflammation and modulation of lung physiological changes are significant contributors to secondary bacterial pneumonia, in general.

Therefore, developing an effective therapeutic regimen for influenza requires a comprehensive understanding of the damage-associated immune-mechanisms to identify therapeutic targets for interventions to mitigate inflammation/tissue-damage, improve antiviral immunity, and prevent influenza-associated bacterial diseases.

MECHANISMS OF LUNG DAMAGE

Influenza-mediated cell death.

Influenza can directly damage epithelial cells through various viral escape strategies and viral recognition by intracellular recognition molecules [9, 30, 31]. Apoptosis and necroptosis of infected epithelial cells represent the primary form of damage to the epithelium during the early stages of influenza infection [12, 32, 33] (Figure 2). Infected host cells attempt to prevent viral replication by inducing caspases to begin the apoptotic cascade. Effector caspases activate proteolytic enzymes to fragment cellular DNA and further inhibit producing regulatory proteins [12, 32–34]. However, apoptosis limits activating the innate immune response as the recognition of pro-cell death receptors on apoptotic cells drives the inflammatory response to apoptosis [34, 35]. Without recognizing cell-to-cell pro-apoptotic receptors, like the FS-7-associated surface antigen (Fas), inflammation from apoptotic cells is slow to signal the immune response that a viral infection has manifested [30, 34]. Necroptosis, or cell-programmed necrosis, is the phenomenon by which cells uncontrollably die and release their intracellular content into the periphery [36–38]. The DNA-dependent activator of interferon (IFN) regulatory factors (DAI) is required for receptor-interacting protein kinase 3 (RIPK3) induction, which leads to type I IFN production [39]. During IAV infection, viral RNA is recognized through DAI and leads to either apoptotic or necroptotic cell death [39]. When DAI recognizes IAV RNA, it activates RIPK3, which subsequently recruits mixed lineage kinase domain-like pseudokinase via phosphorylation to create the necrosome [39]. Conversely, receptor-interacting protein kinase 1 in the presence of caspase 8 and Fas-associated protein with death domain leads to apoptosis rather than necroptosis [39, 40]. The influenza virus can mediate necrosome formation and necroptosis as the primary means of cell death during infection by limiting the presence of caspase 8 [40]. The matrix 1 (M1) protein of IAV specifically binds to caspase 8, thereby skewing the cell to undergo inflammatory necroptosis rather than apoptosis [40].

Figure 2.

Tumor-necrosis-factor receptor apoptosis-inducing ligand (TRAIL) and Fas Ligand interactions with their respective receptors on epithelial cells in the lungs, along with type I and Type III interferons (IFN), result in the induction of either apoptosis or necroptosis as a means of epithelial cell death. Influenza A virus (IAV) predisposes the epithelial cells in the lungs towards cell death suggesting the epithelial cell-monocyte interactions further drive lung damage.

Although epithelial cells act as an initiator and architect of host response in the lung, epithelial inflammation and cell death following influenza infection upregulates bacterial adherence receptors, such as PAF-r [41], CEACAM1 [42], and ICAM-1 [43], which promote bacterial colonization and invasion to establish disease [42]. Both host and viral factors contribute to epithelial inflammation making influenza infected cells permissive for bacterial co-infections. For instance, viral neuraminidase activity cleaves sialic acid residues when influenza virions are released from the cell and results in the exposure of cellular receptors making the lung environment favorable for Spn adherence and invasion [44]. Influenza cytotoxin, PB1-F2, is associated with a potent inflammatory response and susceptibility for secondary bacterial infection in a murine model of influenza and Spn [45]. PB1-F2 elicits an initial proinflammatory response followed by increased IL-10 levels making the IAV infected tissues susceptible to bacterial adherence and disease [46, 47]. Additionally, immune cells that experience prolonged or transient interactions with IAV antigens may induce cellular anergy and a more permissible environment for bacterial adherence and colonization as the immune response becomes less active [48]. Therefore, the anti-viral epithelial response is key to regulating lung inflammation, function, and control of influenza-associated bacterial diseases.

Innate Immune Response and Lung Damage.

The lungs generate a robust inflammatory response to the infection when the influenza-infected alveolar epithelial cells activate a range of pattern recognition receptors such as TLRs, C-type lectin receptors, nod-like receptors (NLRs), and Rig-I-like receptors (RLRs) [9, 12, 49]. The virally infected epithelial cells release chemokines, which recruit an overwhelming number of myeloid and lymphoid cells to the lungs [31, 50]. In some cases, the hyperinflammatory response in influenza-infected lungs can lead to developing acute respiratory distress syndrome (ARDS), resulting in significant lung damage and respiratory failure [51, 52]. ARDS is an incredibly debilitating disease where lung function can be severely impaired as both the epithelial and endothelial layers of the respiratory epithelium, along with the air sacs responsible for gas exchange, become damaged. The quality of life for people afflicted by ARDS decreases drastically [53].

Monocytes and M1 macrophages have a significant role in manifesting airway pathology [54–56]. Influenza triggers the release of chemokines, C-C motif ligand 2 (CCL2), CCL5, and CCL7, which recruit monocytes to the lungs via interaction with the proinflammatory receptor, C-C chemokine receptor type 2 (CCR2) [55]. Monocytes and macrophages have three main antiviral properties: phagocytosis, cytokine production, and antigen presentation [57, 58]. Monocytes and macrophages establish direct contact with virally infected epithelial cells and induce epithelial apoptosis. This process is accomplished by presenting pro-apoptotic receptors (termed death receptors) on virally infected cells that interact with pro-death ligands on the surface of monocytes and macrophages [56, 59]. The primary ligands on the surface of the monocytes and macrophages that induce death are the Fas ligand and the tumor-necrosis-factor receptor apoptosis-inducing ligand (TRAIL) [59].

Inflammatory monocytes and M1 macrophages are also exorbitant producers of inflammatory cytokines that further amplify the immune response by recruiting additional inflammatory cells around the virally infected epithelial cells and mucosal membranes [54, 60]. Monocyte interactions with the influenza-infected epithelium are central to killing the virally infected epithelium. In a pre-adaptive early stage of viral resolution, these cell-cell interactions where monocytes promote the death of epithelial cells, consequently, results in viral resolution [55, 56, 59]. However, the hyper-inflammatory response promotes epithelial inflammation and dysregulates tight junction proteins, leading to increased vascular permeability and disrupted physical barriers in the lungs [55, 61]. Several lines of evidence indicate that barrier integrity is crucial to regulating inflammation [61] and that disrupted barrier integrity is associated with exacerbated inflammation and increased susceptibility to secondary bacterial infections [62].

Neutrophils are professional phagocytic cells and are the most abundant circulating leukocyte in the blood [63]. Neutrophils developed in the bone marrow from myeloblasts are the first responder cells to most infections. However, the neutrophil’s role in influenza infection pathophysiology is conflicting and not clearly understood. Some studies indicate that influenza suppresses the lung’s antibacterial response by promoting neutrophil death [64], as well as impairing neutrophil recruitment [65]. Tissue-pathology and compromised antibacterial defense in influenza-infected lungs result in creating a permissive environment for bacterial colonization, growth, and invasion. However, other studies have demonstrated the significant role of neutrophilic reactive oxygen species (ROS) in promoting cell death of immune and non-immune cells of the influenza-infected lung [66]. Neutrophil extracellular trap formation (NETosis) also contributes to acute lung damage during IAV infection, albeit indirectly [67]. In the context of influenza infection, NETosis serves to produce mucous and cellular debris from dead neutrophils. In areas of tissue consolidation, the DNA from NETosis becomes entangled with the small blood vessels of the respiratory epithelium, suggesting NETosis contributes to damage of the thin alveolar capillaries [67]. Also, when myeloperoxidase (the enzyme responsible for generating hydroxyl radicals from hydrogen peroxide) is inhibited, NETosis formation increased in influenza-infected lungs, suggesting that ROS is important in NETosis, again indicating the neutrophil’s function in influenza immunopathology [67–69]. Together, the published data imply that while neutrophils constitute a significant effector cell type in bacterial infections of the lung, neutrophils may not be the primary drivers of lung pathology during influenza infection. Instead, monocytes and M1 macrophages appear to have a more significant role in viral clearance and lung pathology.

Natural Killer (NK) cells lack antigen/major histocompatibility complex I (MHC I) molecules on their surface and induce apoptosis of virally infected cells, which results in an early and robust cytotoxic response by the innate immune system [70]. The NK cells achieve this cytotoxic response by acquiring cytotoxic function secreting perforins and granzymes close to their target cell [70, 71]. During an early infection phase, NK cells are recruited into the influenza-infected lungs and begin to mount an immune response starting within three days of the initial infection [71]. Although an early NK cell response is crucial for the control of influenza, the virus evades NK cell-mediated killing in several ways, such as through secreting unbound HA [72] and HA attaching to sialic acid residues on the surface of NK cells [73]. Additionally, HA internalization decreases NK cell cytotoxicity mediated by zeta chain downregulation. Consequently, HA diminishes the extracellular receptor kinase (ERK) pathway, which is essential for granule production and exocytosis from the NK cells [73, 74]. Despite the role of NK cells in viral clearance, NK cells’ harmful function in influenza mouse models is evident [75, 76]. More IL-15⁻/⁻ mice survive when NK cells are lacking [77]. In addition to the direct role of NK cells in influenza pathology, NK cells negatively influence generating influenza-specific CD8⁺ T cells, thus improving the survival of the NK cell-deficient mice [78]. Since NK cells are early cytokine-producing cells in influenza mouse models, they could facilitate tissue pathology by promoting immune cell recruitment to influenza-infected tissue [72, 77]. Therefore, NK cells have a double-edged role by the early control of the virus while also manifesting pathology that directly kills influenza-infected epithelial cells and modulates the inflammatory response in the lungs.

Although several immune cells have a role in resolving influenza and contributing to pathology, inflammatory monocytes and macrophages comprise the predominant group of innate cells recruited during the early stages of influenza infection in the lung. M1 macrophages and inflammatory monocytes can cause virally infected epithelial apoptosis via apoptosis-inducing ligands and secrete proinflammatory cytokines that further the inflammatory damage. Additionally, monocytes and macrophages can establish crosstalk with noninfected epithelial cells to manifest non-specific bystander damage. The early innate immune response not only shapes the adaptive immune response but also determines the extent of the damage to the lungs and upper respiratory tract. The initial damage caused by the innate immune response may have adverse effects after viral clearance like increased susceptibility for bacterial superinfections.

T-cell responses.

The reduced viral clearance in RAG⁻/⁻ and MHC I knockout mice highlight the significance of adaptive immune response and CD8 T cells in influenza resolution [79, 80]. While CD8⁺ T cells could directly control primary influenza infection [81], antibodies and the role of B-cells primarily control re-infections [82]. CD8⁺ T cells produce effector molecules such as granzymes, perforins, and serine proteases, that induce the apoptosis of virally infected cells, thereby controlling viral infection and spread [83, 84]. CD8⁺ T cells also express Fas ligand and TRAIL, which interact with their corresponding receptors on the surface of virally infected cells, leading to activating procaspases into apoptotic effector caspases [84–86]. Therefore, CD8⁺ T cells control influenza through effector cytotoxic function. Although CD8⁺ T cells contribute to influenza virus clearance, emerging evidence highlights their role as conditionally protective with prospects of T cell-dependent lung pathology [83, 87].

While the immune response may successfully clear the virus, the indirect and bystander damage could lead to significant airway pathology. Excessive cell death could further the proinflammatory immune response resulting in damage to lung epithelial cells [87]. The damaged barrier could cause excessive epithelial cell proliferation impacting the lungs’ respiratory function. Over-accumulated M1 macrophages and inflammatory monocytes promote rapid clonal CD8⁺ T cell expansion in the lungs that develop effector and cytotoxic function in the first week of influenza infection [78]. Besides cytotoxic function, the CD8⁺ T cells also produce proinflammatory cytokines, which could significantly modulate immune and non-immune cells function in the lungs [86–88]. The emerging role of IFN-γ in lung pathology [89] suggests that an excessive T cell response could be central to immune pathology and, therefore, warrants re-evaluating CD8⁺ T cells as a protective component of host immunity vs. their pathogenic role in the context of influenza.

Additionally, CD4⁺ T cells can modulate CD8⁺ T cells’ effector and cytotoxic function, and influenza-specific memory CD4⁺ T cells can mediate protective immune responses to a secondary viral challenge [90]. Follicular helper CD4⁺ T cells are also required to promote forming germinal centers as well as influenza-specific B cell and antibody responses. Data from 2010 in humans indicates that CD4⁺ T lymphocytes activated during previous infections can limit disease severity in the absence of specific antibodies [90], suggesting that CD4⁺ T cells can mediate protective function independent of CD8⁺ T cells and B cells. Therefore, strategies enhancing CD4⁺ T cells’ ability to promote protective antibody responses to both seasonal and potentially pandemic influenza strains can be useful in controlling influenza strains.

INFLAMMATION AND EPITHELIAL BARRIER CROSSTALK.

The epithelial barrier is the first line of defense against infectious agents in the respiratory tract [49, 91] and has a meaningful role in modulating the immune response against viral infection. This immune response modulation is accomplished by recognizing viral components via pathogen recognition receptors and secreting chemokines. In the last two years, evidence suggests that leukocytes-epithelial barrier crosstalk results in acute injury of epithelial cells and, therefore, constitutes another potential target for therapeutic intervention [92, 93].

The major epithelial cell types in the respiratory tract are ciliated, columnar, secretory, basal, and undifferentiated [94, 95]. The upper respiratory epithelium is composed of ciliated, pseudostratified, and columnar epithelial cells anchored to the basement membrane of the extracellular matrix [95]. The lung epithelial cells change from columnar to cuboidal and then to squamous epithelial cells in the alveolar ducts and alveoli, respectively [49, 95]. The alveolar epithelium is comprised of alveolar type (AT) I and ATII cells [96]. The ATII cells primarily produce pulmonary surfactants essential for efficient gas exchange and for maintaining alveoli’s structural integrity [97, 98]. ATII cells repair epithelial cells upon injury and contribute to lung defense by secreting antimicrobial factors, cytokines, and chemokines [99, 100]. Additionally, ATI and ATII cells form a physical and biological barrier and regulate the flow of solutes, water, and immune cells/molecules across the epithelium [96, 100]. Therefore, the respiratory epithelium’s role is not limited to gas exchange but is the lung’s central defense against pathogens.

The respiratory epithelium recognizes influenza by using TLRs, RLRs, and NLRs (41, 57). When TLRs, NLRs, and RLRs recognize intracellular IAV, an antiviral interferon response ensues by recruiting inflammatory cells to the respiratory mucosal sites and polarizing the resident immune cells in the lungs towards an inflammatory and damaging phenotype [13]. Early influenza recognition provokes epithelial cells to produce chemokines. Epithelial inflammasomes’ activation and the release of mature IL-1β and IL-18 promote epithelial inflammation and local inflammatory response in the lungs [101]. Deleting epithelial inflammasomes impairs viral clearance and exacerbates disease severity in murine influenza models, suggesting the protective role of epithelial inflammasomes in controlling influenza [102]. Epithelial cells and resident immune cells produce the major monocyte chemoattractant, CCL2, early in influenza infection [18, 103, 104], releasing CCR2 bone marrow monocytes into circulation and consequently recruiting them to the lungs [55, 56]. The early monocyte recruitment in the lungs also orchestrates viral-specific cytotoxic T cell responses [105]. The crosstalk between CCR2⁺ monocytes and influenza-infected epithelia involves the direct physical interaction between the two cell types as well as simultaneous lung epithelial cell interactions with inflammatory cytokines. Together, the leukocytes-epithelial crosstalk modulates the components of epithelial inflammation and barrier integrity, i.e., tight junction proteins, apoptosis, proliferation, reactive oxygen species, and repair, leading to exacerbated vascular permeability and lung inflammatory response (Figure 3).

Figure 3.

Inflammation induced by epithelial cell recognition of influenza virus or interactions with proinflammatory immune cells in the lungs results in weakening tight junction barriers and thereby increasing vascular permeability. These cell-cell interactions also result in epithelial cell death which further weakens the epithelial cell barriers and leads to a more permissive state for bacterial infections.

THERAPEUTIC INTERVENTION TO LUNG BARRIER DYSREGULATION

In addition to regulating inflammation, the epithelial cell’s barrier integrity is central to containing the bacteria in the lungs and preventing bacterial dissemination to the bloodstream. Given the lack of a vaccine against SA, as well as the many Spn serotypes that the current pneumococcal vaccine does not cover, therapeutically restoring barrier integrity is a path forward in controlling influenza pathology and associated bacterial diseases in the respiratory tract. However, developing effective therapeutics will require a greater understanding of the host response at multiple levels, starting with initiation, amplification, and regulation. Also, several key aspects require broader consideration, such as targeting inflammatory cell ablation, inflammatory molecule neutralization, and directly targeting the molecules of signaling pathways involved in the crucial inflammation-barrier crosstalk. For instance, deleting NLRP3 inflammasome in myeloid cells dysregulated the alveolar barrier in the lungs and made the lungs more tolerant of Spn infection [13, 61, 106]. Therefore, explicitly inducing NLRP3 inflammasome in myeloid cells could preserve the epithelial barrier’s integrity and promises to regulate influenza pathology and barrier integrity therapeutically.

Pathogenic inflammatory mediators produced by immune and non-immune cells during influenza are also therapeutic intervention targets. Chemokines, CCL2, CCL5, and CCL7, dysregulate respiratory epithelial barrier by targeting inflammatory monocyte and macrophage recruitment, which establish pathogenic crosstalk with barrier cells in the lungs. Therefore, targeting one or a combination of these molecules could lessen the damage to the respiratory barrier. However, a substantial investment is required to establish the targeted intervention’s feasibility to benefit the host’s response in pathology containment with improved viral resolution.

ABBREVIATIONS

ARDS

Acute respiratory distress syndrome

CCL2

C-C motif ligand 2

CCR2

C-C chemokine receptor type 2

CDC

Center for Disease Control and Prevention

DAI

DNA-dependent activator of interferon regulatory factors

Fas

FS-7-associated surface antigen

HA

Hemagglutinin

IAV

Influenza A virus

IFN

Interferon

M1

Matrix 1 protein

MHC I

Major histocompatibility complex I

NETosis

Neutrophil extracellular trap formation

NK

Natural killer cells

NLR

Nod-like receptors

RIPK3

Receptor-interacting protein kinase 3

RLR

Rig-I-like receptors

ROS

Reactive oxygen species

SA

Staphylococcus aureus

Spn

Streptococcus pneumoniae

TLR

Toll-like receptor

TRAIL

Tumor-necrosis-factor receptor apoptosis-inducing ligand