Pathogenesis of Respiratory Viral and Fungal Coinfections.

Individuals suffering from severe viral respiratory tract infections have recently emerged as "at risk" groups for developing invasive fungal infections. Influenza virus is one of the most common causes of acute lower respiratory tract infections worldwide. Fungal infections complicating influenza pneumonia are associated with increased disease severity and mortality, with invasive pulmonary aspergillosis being the most common manifestation. Strikingly, similar observations have been made during the current coronavirus disease 2019 (COVID-19) pandemic. The copathogenesis of respiratory viral and fungal coinfections is complex and involves a dynamic interplay between the host immune defenses and the virulence of the microbes involved that often results in failure to return to homeostasis. In this review, we discuss the main mechanisms underlying susceptibility to invasive fungal disease following respiratory viral infections. A comprehensive understanding of these interactions will aid the development of therapeutic modalities against newly identified targets to prevent and treat these emerging coinfections.

INTRODUCTION

Fungal infections are major causes of human morbidity and mortality. These infections range from superficial mucosal and dermal infections to life-threatening disseminated infections that can involve virtually any organ (1, 2). Opportunistic fungi, including Aspergillus, Pneumocystis, and Cryptococcus, can cause severe fungal infections in the lungs that can lead to invasive disease and dissemination to other tissues (3). Invasive fungal infections in the lungs, including invasive pulmonary aspergillosis (IPA), Pneumocystis pneumonia (PCP), and cryptococcosis, account for more than one million cases worldwide annually and primarily affect immunocompromised individuals, such as patients with HIV/AIDS, those with malignancies and undergoing bone marrow transplantations, and patients receiving immunosuppressive therapies (1, 3). Over the last decade, it has become evident that patients with severe viral respiratory tract infections are highly susceptible to developing a fungal coinfection, in particular pulmonary aspergillosis.

Lower respiratory tract infections cause nearly 4 million deaths annually, with influenza accounting for up to half a million of them (4). Severe bacterial pneumonia following influenza infection, most commonly caused by Streptococcus pneumoniae, Staphylococcus aureus, Haemophilus influenzae, and Streptococcus pyogenes, is well recognized and known to increase disease severity and mortality in these patients (5–7). It is estimated that around 25% of all influenza-related deaths are associated with bacterial coinfections, particularly during seasonal outbreaks (8, 9). More recently, there is increased recognition of the importance of fungal coinfections, primarily caused by Aspergillus, in the severity and mortality of patients suffering from influenza (10, 11). Coinfections are well known complications in other respiratory viral diseases like severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), parainfluenza virus, cytomegalovirus (CMV), respiratory syncytial virus (RSV), rhinovirus, and adenovirus, although assessing their precise impact on disease severity and attributable mortality has proven difficult (12–20). Previous studies estimated that 25 to 30% of SARS survivors experienced secondary infections (21). In the current coronavirus disease 2019 (COVID-19) pandemic, early studies have suggested that 50% of patients who died due to COVID-19 experienced a secondary infection (22). Susceptibility to bacterial coinfections in influenza patients is thought to be attributed to damage and dysfunction of the epithelial barriers, inability to mount an effective primary immune response, and/or incapacity to develop disease tolerance to infection. Even though most of these mechanisms could be playing a role during fungal coinfections, a detailed understanding of the interactions between respiratory viruses and fungal pathogens is lacking. In this review, we evaluate what is currently understood about the immunopathological mechanisms underlying susceptibility to invasive fungal infections following severe viral pneumonia with an emphasis on influenza-associated pulmonary aspergillosis (IAPA) and COVID-19-associated pulmonary aspergillosis (CAPA).

FUNGAL COINFECTIONS IN RESPIRATORY VIRAL DISEASE

The realization that fungal coinfections complicate viral respiratory disease has only recently emerged. In retrospect, earlier case reports described the association of Aspergillus and influenza coinfection, but the significance was not appreciated (8). During the 2009 H1N1 pandemic, increasing numbers of cases with IAPA were described in the literature, resulting in proposed disease definitions and clinical management guidelines (10, 11). Furthermore, the recent update of the U.S. clinical practice guidelines regarding diagnosis, treatment, chemoprophylaxis, and institutional outbreak management of seasonal influenza has included fungal coinfection as a complication of influenza (10). Based on recent clinical observations and diagnostic test results in patients with IAPA, a new clinical algorithm has been proposed to better classify the certainty of the diagnosis of invasive Aspergillus disease (23–25). A similar pattern has been observed during the current COVID-19 pandemic, with hundreds of cases of fungal coinfections being described (26, 27). International efforts to better classify CAPA has also led to proposed novel disease definitions and research and clinical guidelines (2, 28, 29). The recognition that patients with severe viral pneumonia have an increased susceptibility to developing fungal coinfections asks for an analysis of the reported clinical epidemiology to provide insights into the clinical importance of fungal-viral coinfections. Underlying medical conditions resulting in lung injury, such as asthma, cystic fibrosis, and chronic obstructive pulmonary disease (COPD), are considered independent risk factors for developing invasive fungal infections (30) and therefore are covered in this review.

Influenza and Aspergillus

Influenza viruses are classified into types A, B, C, and D based on genetic and antigenic differences (4). Influenza A viruses are predominantly responsible for seasonal annual epidemics, and they have caused a number of pandemics in humans (4, 31). Worldwide, 3 to 5 million people develop severe influenza infection, leading to up to half a million deaths every year. Of the hospitalized patients, 5 to 10% need admission to an intensive care unit (ICU). A feared complication is the development of acute respiratory distress syndrome (ARDS), which is associated with high mortality rates (32). The first report of pulmonary aspergillosis following influenza bronchopneumonia dates back to 1952 (33). After that, several clinical cases have been reported (34–38), but it was not until the 2009 H1N1 pandemic that the number of reports started to increase dramatically (24, 39). One of the first detailed descriptions of the association between influenza H1N1 and Aspergillus coinfection showed that 23% of critically ill patients with influenza admitted to the ICU developed IPA (39). A large number of other centers have reported comparable experiences, but the incidence of Aspergillus- influenza virus coinfection shows huge variation, with incidences reported between 7% and 32% (24, 32, 40–51). Despite such variations, a number of important observations have been made supporting the fact that fungal coinfections do play a significant role in the disease severity and outcome of influenza pneumonia. First, one of the larger clinical studies that included more than 400 patients admitted to the ICU over a period of 7 years identified influenza as an independent risk factor for the development of IPA (24). Second, pulmonary fungal coinfections outnumbered the cases of bacterial coinfection among influenza patients admitted to two ICUs in the Netherlands (16 versus 13 out of 45 patients, respectively) (32). Increased mortality rates of 51% to 66% have been reported in patients with IAPA compared to 15% to 28% in patients without and with bacterial coinfections (24, 49). Even though the majority of reported cases have been associated with influenza A virus, cases associated with influenza B virus have also been described (52, 53). In the last decade, novel avian influenza viruses have emerged in Asia associated with mortality rates of up to 50%. A study from China collected data from 335 patients with avian influenza H7N9 between 2013 and 2018, and 5.4% of those were diagnosed with IPA (54).

SARS-CoV and Aspergillus

The ongoing COVID-19 pandemic, caused by SARS-CoV-2, has affected millions of people and caused more than four and a half million deaths worldwide by September 2021. About 5% of COVID-19 patients infected with SARS-CoV-2 require ICU management; these patients might be at high risk of developing secondary infections, including IPA (55–58). Reports from the previous SARS epidemic in 2003, caused by SARS-CoV-1, pointed to the occurrence of Aspergillus infection in those patients treated with corticosteroids for virus-induced inflammation (59–61); however, no systematic studies were performed to determine rates of incidence. During the current pandemic, secondary bacterial (22) and fungal coinfections, mainly due to Aspergillus spp. (Table 1), are increasingly being reported (62–97). Overall, hundreds of patients with CAPA have been reported from many countries in Europe, Asia, Australia, and America (27). As with IAPA, incidences exhibit great variability, with some studies reporting incidences of CAPA as low as 1% of ICU cases and others reporting extremely high incidences of up to 35% in ICU settings in Europe (56, 98–100). In a case series from the Netherlands, the mortality rate among patients with CAPA was 67% compared to 32% in patients with severe COVID-19 without signs of IPA (101). Importantly, in a recent prospective study from Germany, COVID-19 was independently associated with IPA (95). Some studies suggest that CAPA might be underdiagnosed due to difficulties obtaining respiratory samples. Concerns over aerosolization of respiratory secretions and the SARS-CoV-2 virus have restricted the number of invasive procedures performed, such as bronchoalveolar lavage (102, 103). Moreover, there are inherent difficulties in obtaining a clear diagnosis of Aspergillus infection, whereas others have suggested that the incidence of Aspergillus infection in COVID-19 patients is not as high as previously predicted (104–107). Whether these discrepancies are associated with the extent to which specific fungal diagnostic tests are employed is at present unclear.

TABLE 1

Large clinical studies (with 10 or more patients) reporting cases of CAPA

Study (reference)	Country	Incidence [no. (%)]	Mortality [no. (%)]
Alanio et al., 2020 (56)	France	9/27 (33)	4/9 (44)
Bardi et al., 2021 (58)	Spain	4/140 (3)	NR
Bartoletti et al., 2020 (73)	Italy	30/108 (28)	13/30 (44)
Dupont et al., 2021 (86)	France	19/106 (18)	8/19 (42)
Falces-Romero et al., 2020 (71)	Spain	NR	7/10 (70)
Fekkar et al., 2020 (82)	France	7/145 (5)	4/7 (57)
Fu et al., 2020 (87)	China	1/101 (1)	NR
Gangneux et al., 2020 (28)	France	7/45 (16)	2/7 (29)
Garcia-Vidal et al., 2021 (88)	Spain	7/989 (0.7)	3/7 (43)
Helleberg et al., 2021 (85)	Denmark	2/25 (8)	2/2 (100)
Koehler et al., 2020 (98)	Germany	5/19 (26)	3/5 (60)
Lahmer et al., 2021 (95)	Germany	11/32 (34)	4/11 (36)
Lamoth et al., 2020 (69)	Switzerland	3/118 (3)	1/3 (33)
Machado et al., 2021 (89)	Spain	6/239 (2.5)	6/6 (100)
Nasir et al., 2020 (65)	Pakistan	5/23 (22)	3/5 (60)
Roman-Montes et al., 2021 (90)	Mexico	14/144 (10)	8/14 (57)
Rutsaert et al., 2020 (99)	Belgium	7/20 (35)	4/7 (57)
Segrelles-Calvo et al., 2021 (91)	Spain	7/215 (3)	5/7 (71)
van Arkel et al., 2020 (101)	The Netherlands	6/31 (19)	4/6 (67)
Van Biesen et al., 2020 (92)	The Netherlands	9/42 (21)	2/9 (22)
Velez Pintado et al., 2021 (96)	Mexico	16/83 (19)	5/16 (31)
Wang et al., 2020 (76)	China	8/104 (8)	NR
White et al., 2020 (2)	Wales	19/135 (14)	11/19 (58)
Yang et al., 2020 (75)	China	2/52 (4)	NR

CAPA, COVID-19-associated invasive pulmonary aspergillosis; NR, not reported.

SARS-CoV-2 and Mucormycosis

Recently, thousands of COVID-19-associated mucormycosis (CAM) cases have been reported in the literature, mostly from India (108, 109). Importantly, mucormycosis associated with influenza has also been described (110). Major risk factors include patients receiving systemic corticosteroid treatment or suffering from uncontrolled diabetes (111, 112). Hyperglycemia increases the expression of the glucose-regulated protein (GRP78), which acts as a receptor for the CotH protein kinase present in Rhizopus spores, helping the fungus to adhere and invade endothelial and nasal epithelial cells (113–115). In addition, beyond their immunosuppressive function, treatment with corticosteroids can cause diabetic ketoacidosis, further increasing susceptibility to CAM (116, 117). A recent review of the literature described that CAM exhibits a mortality of up to 49%, with rhino-orbital cerebral mucormycosis being the most common manifestation of the disease, followed by pulmonary mucormycosis (109). Notably, a significant proportion of surviving patients suffered life-changing morbidities, including loss of vision. Diagnosis of CAM is challenging, as the clinical and radiological signs of pulmonary and disseminated mucormycosis are nonspecific and may overlap with findings associated with COVID-19. Furthermore, mucormycosis is caused by a variety of Mucorales species (Rhizopus arrhizus is the predominant species in India), with some of them exhibiting poor susceptibility to antifungal therapy (118). Recently, the European Confederation of Medical Mycology (ECMM) and the International Society for Human and Animal Mycology (ISHAM) provided a comprehensive guideline of recommendations for the clinical management of CAM patients, including diagnosis, treatment, and prevention (119). Early recognition, better diagnostics, and more effective antifungals are required to improve the outcome of these patients, especially in low and middle-income countries.

Viral Respiratory Tract Infections and Other Fungi

Fungal coinfections complicating viral infections with pathogens other than Aspergillus have also been reported. PCP caused by the fungal pathogen Pneumocystis jirovecii is the most common AIDS-defining disease, with up to half a million cases worldwide annually (3). Influenza complicated by PCP has been observed in HIV-infected patients (120–122). Therefore, influenza vaccination has been suggested as a prophylactic measure to reduce the risk of developing PCP secondary to influenza in HIV patients (123). Patients with deficiencies in adaptive immunity or individuals undergoing immunosuppression therapy are also at risk of developing PCP associated with influenza infection (124, 125). Recently, cases of COVID-19 and Pneumocystis coinfections have been reported (81, 126, 127), some of them associated with HIV comorbidity (79, 128–130). Since SARS-CoV-2 and PCP have common clinical and radiological features, coinfection with Pneumocystis is likely underappreciated in patients with SARS-CoV-2 (128, 131–138) and there are several clinical case studies reporting misdiagnosis (139, 140). Cryptococcus infections affect primarily immunocompromised hosts like HIV-infected individuals and solid organ transplant recipients. Infection starts in the lungs, and it can then disseminate into the central nervous system, causing meningitis that accounts for more than 200,000 cases worldwide annually (3). Only a few cases of influenza with concomitant cryptococcal infection have been reported in the literature (141–143), all of them with either the H1N1 or H7N9 strain (142). The scarcity of case studies could be a result of underdiagnosis, so special attention is needed in regions with a high prevalence of HIV/AIDS, including sub-Saharan Africa. Candida auris is a multidrug-resistant fungal pathogen classified as an “urgent threat” by the U.S. Centers for Disease Control and Prevention (CDC) due to its ability to cause life-threatening systemic infections in critically ill patients. Several outbreaks of C. auris infection in COVID-19 patients have been reported (144–149), in some cases associated with corticosteroid treatment (150–154). C. auris is difficult to identify by standard laboratory methods, which can lead to misidentification, causing outbreaks in health care settings often associated with high mortality. Therefore, advancing diagnostic methods is essential for early detection and control of this emerging pathogen.

COPATHOGENESIS OF RESPIRATORY VIRAL-FUNGAL COINFECTIONS

Immune responses against one pathogen can significantly influence immunity to a secondary nonidentical pathogen. This phenomenon, termed heterologous immunity, has been studied mainly in the context of viral infections and vaccines but could also play a role during viral-fungal coinfections (155, 156). Studies on the copathogenesis between viral and fungal coinfections are scarce, unlike studies regarding viral and bacterial coinfections. Most reports have attributed destruction of the airway epithelium and suppression of cellular immunity (including defective antigen-specific cytotoxic T lymphocyte responses and impaired phagocyte activities such as phagocytosis, production of cytokines, and reactive oxygen species [ROS], formation of neutrophil extracellular traps [NETs], and killing abilities) as the causes responsible for fungal coinfections (157–159). Several of the mechanisms that account for fungal susceptibility in individuals suffering from influenza could also be at play during SARS-CoV-2 infection, including the effects on tissue integrity and functionality and the dysregulation of immune responses and effector functions (160). Despite these similarities, the pathophysiology of SARS-CoV-2 infection is different from that of influenza at numerous levels, including viral tropism, viral replication, and incubation period as well as the effects on the host defense (161–165).

The outcome of host-pathogen interactions depends on numerous factors, including dose, route of infection, and virulence properties of the pathogen, as well as several host factors that include innate and adaptive immunity. Initiation of protective antiviral immunity depends on the recognition of viral RNA in the endosomal or cytosolic compartment by Toll-like receptor 3 (TLR3) and TLR7 or by retinoic acid-inducible gene (RIG)-I-like receptors (RLRs) (RIG-I and melanoma differentiation-associated protein 5 [MDA5]), respectively. Viral recognition by innate immune cells, including dendritic cells (DCs) and macrophages, triggers a signaling cascade leading to both NF-κB-mediated induction of proinflammatory cytokines (interleukin 6 [IL-6], tumor necrosis factor [TNF], IL-1), and IFN regulatory factor 3 (IRF3) and IRF7-mediated induction of type I (IFN-α and IFN-β) and type III (IFN-λ) interferons (IFNs). IFNs are crucial for effective antiviral immunity; for example, in epithelial cells, IFN signaling inhibits viral replication and orchestrates an effective adaptive antiviral immune response (4, 166, 167). Conversely, antifungal immunity strongly relies on C-type lectin receptors (CLRs) that play a key role in the recognition of fungal glucans, glycolipids, and glycoproteins by phagocytes (mainly macrophages and neutrophils) and in the activation of innate host defense mechanisms, including phagocytosis, respiratory burst, formation of NETs, autophagy, and chemokine and cytokine production (168). These mechanisms promote fungal killing but also influence activation of the adaptive immune system (169–173). Understanding how these mechanisms interact in a synergistic or antagonistic manner is fundamental to dissecting their roles during viral-fungal coinfections. In the next section, the major mechanisms that mediate susceptibility to fungal coinfections in individuals suffering from severe viral pneumonia are discussed: from innate immune mechanisms, including the role of the epithelium, phagocytes, and antigen-presenting cells (APCs), to T cell responses and adaptive humoral and cytotoxic responses.

Barrier Integrity

The respiratory epithelium is composed of a variety of cells, including a pseudostratified epithelium of ciliated and secretory cells lining the trachea and most proximal airways. A cuboidal epithelium lines the small airways, and squamous type I alveolar cells (involved in the process of gas exchange) and cuboid type II alveolar cells (which secrete pulmonary surfactant) form the alveoli (174). In healthy individuals, inhaled airborne fungal conidia are easily trapped in the mucus and eliminated mechanically by ciliated cells from the upper respiratory tract. However, due to their small size (2 to 3 μm), Aspergillus conidia (asexual spores) can reach the lower respiratory tract and interact with the airway epithelium, at either the bronchial or alveolar level (175). Upon reaching airway epithelial cells, fungal conidia are taken up and trafficked through the endosomal system, culminating with the formation of the phagolysosome by fusion of late phagosomes with lysosomes. This organelle has an acidic pH and contains many degradative enzymes that facilitate destruction and clearance of fungal conidia from the host. However, upon injury or disease (disrupting barrier integrity), conidia may escape this process and eventually germinate, facilitating tissue invasion (175). Different respiratory viruses preferentially bind and infect specific epithelial cells expressing specific receptors along the respiratory tract. For instance, cell entry of influenza virus is mediated by the binding of the viral hemagglutinin to terminal sialic acids that are attached via either an α2,3 or α2,6 linkage. Human influenza virus, such as H1N1, binds preferentially to α2,6-linked sialyloligosaccharide receptors, which predominate in nonciliated epithelial cells from the upper respiratory tract, whereas avian influenza virus, such as H5N1 and H7N9, binds to α2,3 linkages, which are more prevalent in ciliated epithelial cells from the lower respiratory tract (31, 176–179). Lower respiratory tract infection enables deep lung infection by other pathogens, including fungal pathogens. Angiotensin-converting enzyme 2 (ACE2) is the cellular receptor for SARS-CoV and the new SARS-CoV-2 (180–184). Both SARS-CoV and SARS-CoV-2 primarily target type II pneumocytes, consistent with their ACE2 expression (161); however, SARS-CoV-2 replicates abundantly in upper respiratory epithelia and is efficiently transmitted (185, 186). Of note, SARS-CoV and SARS-CoV-2 can also infect alveolar macrophages that support viral replication (185, 187–189). Both influenza virus and SARS-CoV-2 can cause pneumonia, which occurs when infection and inflammation involve the alveoli and lung parenchyma. Therefore, productive viral infection of specific respiratory epithelial cells along the respiratory tract will determine the clinical symptoms as well as the susceptibility to fungal infections (179).

Respiratory viral infection causes multiple changes in the lungs that can weaken antifungal defenses, facilitating secondary fungal invasion. These effects can be grouped into three major aspects, including changes to the extracellular matrix components that facilitate adhesion, compromise of epithelial cytoskeletal machinery that modifies the dynamics of internalization, and damage of the epithelium that compromises barrier integrity (190). Disruption of tracheal epithelial integrity after influenza infection affects the mechanical removal of subsequent pathogens, facilitating secondary infections (191, 192). In the more severe cases, damage to the epithelium can alter the surface display of numerous transmembrane proteins, exposing sites for fungal adherence in the tracheobronchial tree. For instance, injured cells or cells in an intermediate state of differentiation may express apical receptors such as α5β1 integrin (expressed upon lung inflammation and injury to control cell migration during wound healing [193]), to which Aspergillus fumigatus can adhere (194, 195). Moreover, airway fibrinogenolysis and disruption of epithelial tight junctions by Aspergillus-secreted proteases such as alkaline protease 1 (Alp1) during and after germination can trigger allergic inflammation and further contribute to epithelial cell pathology in the context of coinfections (196–199). All of the above suggest that during influenza infection, the chronicity of the disease might determine susceptibility to secondary fungal infections. Lung tissue disruption during SARS-CoV-2 infection has also been shown to be an important factor in determining the severity of the disease, which likely contributes to susceptibility to coinfections (200, 201). Moreover, COVID-19-driven inflammation affects alveolar epithelial regeneration and induces the expansion of pathological fibroblasts that promote fibrosis and impair regeneration (202, 203). Of note, using a model of influenza infection, mice experiencing an acute inflammatory response with limited bystander tissue damage do not show susceptibility to a secondary bacterial infection (204). Whether the same holds true for secondary fungal infections is unknown.

Exposure of the substratum during severe viral pneumonia presents additional opportunities for fungal cells to adhere. In addition, changes in the airways during tissue damage and repair may provide adherence sites during recovery (205). During tissue remodeling, exposure of basement membrane and extracellular matrix components, such as fibronectin, laminin, or collagen, in areas of incomplete healing or where fibrin and fibrinogen deposition have taken place (observed during SARS-CoV-2 infection [206]) could facilitate fungal adhesion to the basal lamina (175, 207). All of the above might be significant in patients suffering from idiopathic pulmonary fibrosis, a condition characterized by the thickening and stiffening of the tissue surrounding the alveoli that shares several risk factors with COVID-19 (208) and has been independently associated with both influenza and Aspergillus infections (209, 210).

Several respiratory viruses, including influenza virus, can hijack the cytoskeletal system to their benefit in order to direct the cellular machinery to the production of viral particles. This could be particularly relevant during Aspergillus infections, since upregulation of several genes involved in cytoskeleton reorganization has been observed during Aspergillus infection (211, 212). Actin polymerization has been suggested to be crucial for internalization of conidia (213). In this context, in vitro studies have shown that Aspergillus-derived mycotoxins, such as gliotoxin, can promote actin cytoskeleton dynamics and internalization of A. fumigatus (214). Importantly, influenza infection alters the levels, structures, and functions of F-actin and microtubules in host cells (215). In addition, influenza infection downregulates the levels and/or activities of proteins involved in the regulation of F-actin and microtubule dynamics, such as Arp2/3 (involved in actin polymerization and the formation of branched actin networks) (216). Notably, components of the Arp2/3 complex have been shown to be upregulated in response to Aspergillus conidia and to mediate internalization of conidia (212). Another example is phospholipase D, which plays a fundamental role in lipid metabolism and cytoskeleton rearrangement and whose activity is stimulated following influenza infection (217). This enzyme mediates rapid endocytosis of the virus and at the same time can promote A. fumigatus internalization (218, 219). The complex interaction between influenza viruses and cytoskeleton components, including actin microfilaments, intermediate filaments, and microtubules, could therefore underlie mechanisms of susceptibility to fungal pathogens.

The epithelium is very important in orchestrating innate immune responses to both viral and fungal infections. The lung epithelium produces several soluble factors that form the first barrier of defense against fungal infections. Airway epithelial cells, particularly from the upper airways, secrete mucins that act as a barrier against fungal invasion. Binding of conidial lectins (FleA) to glycan moieties on gel-forming mucins (MUC5AC and MUC5B) allow them to become trapped within the mucus barrier, unable to reach the underlying epithelium and subsequently cleared from the airways by mucociliary clearance (220, 221). Surfactant proteins (SP-A and SP-D) and ficolins are involved in fungal opsonization and phagocytosis. Chemokines promote neutrophil recruitment and antimicrobial peptides (AMPs; including LL-37 and defensins) and are involved in fungal killing (222, 223). Influenza virus itself can interact with some of these glycosylated proteins, affecting their mode of action. For instance, neuraminidase helps the virus to gain access to airway epithelial cells by catalyzing the cleavage of sialic acids presented by decoy receptors, such as mucins (224–226). Fungal recognition through sialylated mucins is a critical step in the mucociliary clearance and macrophage killing that prevents Aspergillus pneumonia (220). SARS-CoV-2 can inhibit protein translation, which abolishes innate immune responses by the epithelium (i.e., IFN-dependent induction of IFN-stimulated genes) (227). Cross talk between alveolar epithelial cells and other immune cells, including DCs, is crucial for effective pathogen clearance and recovery from injury (Fig. 1) (228). Importantly, epithelial cells can control overinflammation by expressing anti-inflammatory mediators such as the tryptophan catabolizing enzyme indoleamine 2,3-dioxygenase (229). Disruption of these immune-regulatory mechanisms as a result of viral infection could lead to uncontrolled exacerbated inflammation, increasing tissue damage, immunopathology, and susceptibility to fungal coinfections as mentioned above.

FIG 1

Progression of respiratory viral-fungal coinfections into the alveolar space determines disease severity. (A) Progression of a viral infection into the alveolar space. (1) The virus infects airway epithelium. (2) Alveolar macrophages recognize the virus and in response produce cytokines. (3) Cytokines attract more immune cells, including neutrophils and monocytes, which in turn produce more cytokines, creating a cycle of inflammation that damages the lung tissue. (4) Damage can further occur through the formation of fibrin and scar tissue. (5) Weakened blood vessels allow fluid to seep in and fill the lung cavities, leading to respiratory failure. (B) Progression of a fungal infection into the alveolar space following severe viral pneumonia. (1) When Aspergillus enters the airways, damaged epithelium facilitates adhesion of fungal conidia and subsequent invasion. (2) Phagocytosis, fungal killing, and cytokine production by alveolar macrophages are impaired. (3) Recruitment of neutrophils and their cross talk with macrophages are also affected. (4) Loss of neutrophils compromises their cytokine production and neutrophil extracellular trap (NET)-mediated fungal killing. (5) The release of fibrinous material can cause the obstruction of the small airways, decreasing oxygen and carbon dioxide diffusion capacities and creating a hypoxic milieu that changes Aspergillus virulence properties and the outcome of host-Aspergillus interaction. (This figure was created with BioRender.)

In severe cases of influenza, obstruction of the small airways caused by the sloughing of cells and the release of fibrinous material into the airways leads to the decrease of oxygen and carbon dioxide diffusion capacities. This hypoxic milieu can significantly influence the course of Aspergillus infection by affecting fungal virulence and host immune responses (230) (Fig. 1). Importantly, oxygen tension has notable effects on the macroscopic and biofilm morphotypes of Aspergillus fumigatus (colony furrowing and percentage of vegetative nonconidiating mycelia), leading to increased host inflammation, rapid disease progression, and mortality in a murine model of invasive aspergillosis (231). Aspergillus-derived secondary metabolites (i.e., gliotoxin) can further contribute to increased localized and systemic hypoxia by inhibiting angiogenesis and tissue repair (232). Therefore, the metabolic adaptability of Aspergillus spp. to low-oxygen environments could be critical for the ability to cause infection following influenza (175, 233–235). Interestingly, hypoxia inducible factor 1α (HIF1α), a transcription factor that controls immune cell metabolism and function during hypoxic conditions, has been shown to be important in controlling influenza virus replication (236) and for protection against pulmonary Aspergillus infection (237). Beyond the epithelium, hypoxia contributes to endothelial cell activation with the release of several soluble mediators, including proinflammatory cytokines, platelet-activating factor, and adhesion molecules, all of which amplify tissue destruction and inflammation into the small airways (238).

Damage of alveolar integrity can enable fungal spores to reach blood vessels (Fig. 1). Recently, it was shown that a major receptor involved in sensing of Aspergillus 1,8-dihydroxynaphthalene (DHN)-melanin, named melanin-sensing lectin (MelLec), is highly expressed on endothelial cells that line the internal surfaces of vessels (239). Therefore, destruction of alveolar epithelial integrity could enable deep penetration of fungal conidia and invasion through MelLec-expressing endothelial cells. Compromised endothelial sensing of conidia through MelLec could also facilitate Aspergillus infection, as has been shown in murine models of infection (239).

Interferons

Type I and type III IFNs induce an IFN-stimulated gene signature that has the capacity to interfere with every step of viral replication (4, 166, 167). Besides their role during viral infections, IFNs play an essential role in driving antifungal responses in the lungs. Type I and type III IFNs are expressed with distinct kinetics during IPA, and both are essential for the activation of neutrophils (240, 241). Upon Aspergillus recognition, recruited monocytes (via CCR2) are an important early source of type I IFNs that induce optimal expression of IFN-λ. Type III IFN production by hematopoietic and nonhematopoietic cells at the mucosa acts on neutrophils to activate their antifungal response, including ROS production (240, 241). Controlled IFN signaling may be a crucial factor in determining whether secondary fungal infections are cleared at mucosal sites (5, 242, 243). Sustained uncontrolled IFN production can lead to tissue damage and immunopathology; e.g., type I IFNs cause lymphopenia (244), which has been associated with severe cases of influenza and SARS-CoV-2 infection, increasing susceptibility to secondary infections (245–253). Furthermore, not only type I but also type III IFNs can impair microbial control during coinfections (254, 255). Excessive or prolonged production of IFN-λ can interfere with lung repair during influenza recovery, which reduces epithelial proliferation and differentiation, increasing disease severity and susceptibility to coinfections (256, 257) (Fig. 2).

FIG 2

Interplay between cellular mechanisms underlying respiratory viral-fungal coinfections. (A) Effector cellular mechanisms against fungal infections. (1) Fungal recognition by the airway epithelium leads to the production of proinflammatory cytokines that can activate other immune cells, including macrophages and neutrophils. (2) Monocytes and alveolar macrophages play pivotal roles during fungal infections, including phagocytosis and cytokine and chemokine production. (3) Neutrophils form neutrophil extracellular traps (NETs) and produce reactive oxygen species (ROS) that contribute to fungal killing. (4) Lung dendritic cells recognize, ingest, and kill Aspergillus conidia, acquire a fully mature state, and then migrate to draining lymph nodes. (5) Antigen presentation of fungus-derived peptides to naive CD4+ and CD8+ T cells occurs in the draining lymph nodes. (6) T cell activation leads to Th17 differentiation, which is pivotal for control of fungal infections at the airways. In particular, Th17 cells support neutrophil activation and the production of antimicrobial peptides (AMPs) by epithelial cells. (B) Mechanisms responsible for increased susceptibility to fungal infections in patients suffering severe viral pneumonia. (1 to 3) The lung epithelium undergoes different changes over the course of respiratory viral infections, including tissue disruption that facilitates secondary fungal invasion (1) and expression and/or exposure of receptors to which fungal pathogens can adhere (2). In addition, germinated fungal spores themselves release molecules with the potential to increase permeability and tissue damage, such as proteases and mycotoxins (3). (4) The airway epithelium produces type I and type III interferons (IFNs), which have a significant impact on antifungal immunity at different levels. (5) IFN-α/β are also produced by alveolar macrophages. IFNs reduce epithelial cell proliferation and differentiation, increasing susceptibility to coinfections. IFNs suppress monocyte, macrophage, and neutrophil recruitment and effector responses that are essential for fighting fungal infections. IFNs act as negative regulators of inflammasome activation in response to fungal pathogens, thus affecting fungal clearance. IFNs dampen Th17 responses, leading to attenuation of AMP production and neutrophil recruitment that are required for antifungal clearance. (6) Desensitization of pattern recognition receptors (PRRs), which are essential for fungal recognition and antifungal immunity, contributes to susceptibility to coinfections. (7) Viral infections also interfere with antigen-presenting cell functionalities, affecting the subsequent immune response to fungal antigens. For instance, viral infection affects antigen presentation through interference with any of the three signals required for T cell activation, namely, MHC presentation, expression of cosignaling molecules, and/or production of cytokines. (8) Regulatory T cells (Tregs) induced during the recovery and resolution phase of a viral infection persist for long enough to interfere with immunity (i.e., neutrophil functionalities) during subsequent fungal infections. Some questions remain, including the role of the NADPH oxidase 2 (NOX-2) complex in the context of viral-fungal coinfections. (This figure was created with BioRender.)

Type II IFNs can be detrimental during influenza infection (258, 259) and contribute to susceptibility to secondary bacterial infections by depleting alveolar macrophages and suppressing their phagocytic capacity (260–265). IFN-γ also impacts memory Th17 responses that attenuate bacterial clearance following influenza infection (266). Therefore, blocking IFN-γ has been exploited as a therapeutic strategy in several experimental models (267–269). However, type II IFNs might have a protective role during fungal coinfections. IFN-γ production by Th1 cells and invariant natural killer T (iNKT) cells (innate-like lymphocytes that express a conserved αβ T-cell receptor [TCR] chain) is required for the activation of phagocytes (270) and restraining of inflammation during fungal clearance (271). Moreover, postinfluenza Cryptococcus coinfection has been associated with reduced levels of IFN-γ (272). Consequently, IFN-γ therapy has proven to be beneficial as an adjunctive therapy in patients with chronic granulomatous disease (CGD) (who are deficient in the NOX complex) and undergoing transplantation (273), but its beneficial effects during fungal coinfections remain to be investigated.

Understanding the immune response that underpins influenza and COVID-19 is important for identifying potential mechanisms of susceptibility to fungal coinfections. Both are characterized by a cytokine release syndrome; nevertheless, they exhibit distinct immune response profiles (253). In comparison to other highly pathogenic coronaviruses and common respiratory viruses, including influenza A virus, SARS-CoV-2 drives an imbalanced inflammatory response. Studies have shown that both SARS-CoV-2 and SARS-CoV infection drive a lower antiviral transcriptional response that is marked by low type I and III IFN levels and elevated chemokine expression, all of which might contribute to COVID-19 (274–280). Low levels of IFNs have been associated with the presence of autoantibodies against type I IFNs (281–285) or inborn errors in genes involved in the regulation of type I and type III IFN immunity (286). Conversely, other reports have shown upregulation of a type I IFN response in peripheral blood (287) that coexists with a proinflammatory TNF/IL-1β/IL-6-driven response in patients with COVID-19 compared to severe influenza patients (160, 250, 252). Similar results have been observed at mucosal sites (288–290). These discrepancies have been attributed to two differing outcomes (200, 291), with a delayed type I IFN response causing enhanced viral persistence and pathological inflammation but an early robust type I IFN response controlling viral replication that results in a mild disease (275, 280, 292–297). In this context, a type I IFN response is critical for the development of ARDS and increased lethality during severe SARS-CoV infection (275, 292, 298). More recently, studies have shown that the location where IFNs are being produced is also relevant. High levels of type III IFNs, and to a lesser extent type I IFNs, characterize the upper airways of patients with a mild pathology, while severe COVID-19 patients exhibit strong production of IFNs in the lower airways compared to subjects with other infectious or noninfectious lung pathologies (299, 300). This suggests that hyperinflammation and dysregulation of the IFN pathway are likely important factors that contribute to the development of fungal coinfections in COVID-19 patients.

Phagocytes and Effector Mechanisms

Effective elimination of pathogens relies on the recruitment and functions of several immune cells. These effector cells regulate important processes that are pivotal for controlling viral persistence and preventing fungal invasion. Impairment of phagocyte functions following influenza infection is among the most damaging consequences that increase susceptibility to secondary fungal infections (301). Neutrophils are one of the most important innate effector cells for the control of fungal infections. Humans suffering from neutropenia or neutrophil dysfunction exhibit a dramatic increase in susceptibility to major fungal pathogens, including A. fumigatus (168). This is particularly evident in CGD patients, whose neutrophils exhibit impaired fungal killing abilities (302). Suppressed neutrophil recruitment (due to reduced chemokine production) and dysfunction (reflected in impaired myeloperoxidase [MPO], ROS, and NET formation) during influenza virus infection increase susceptibility to secondary bacterial infection (303, 304) and contribute to the development of IPA (301, 305–308). Following influenza, IFN production and signaling through STAT1 impair neutrophil recruitment into the lungs and airways, augmenting fungal burdens (242, 301).

Monocytes and macrophages are key effector cells in controlling fungal infection through direct killing and the production of proinflammatory mediators (168). Influenza infection can result in depletion of alveolar macrophages and affect their functionalities, including inflammasome activation (263, 309, 310), thereby increasing disease severity and susceptibility to coinfections. Cross talk between neutrophils and monocytes/macrophages is important to combat respiratory infections, as neutrophils can drive macrophage inflammasome activation during respiratory viral infection (311–313) and can prevent macrophage depletion during S. pneumoniae coinfection (314). As macrophages are an important source of neutrophil chemoattractants, such as keratinocyte-derived cytokines (KC) and macrophage inflammatory protein-2 (MIP-2) (315), their depletion as seen in IAPA impairs neutrophil recruitment into the lungs (301). As described above, monocytes can trigger neutrophil activation and ROS production through type I IFN production (240, 241). Furthermore, both monocytes and neutrophils can control maturation and expansion of DCs in the lung, which in turn activates neutrophil oxidative burst, which is essential for host defense against Aspergillus fumigatus (316–318). This suggests that reduced numbers of neutrophils, monocytes, and macrophages in the lung tissue caused by influenza infection increases susceptibility to secondary fungal infections (Fig. 2).

Single-cell technologies employed on blood samples from severe COVID-19 patients have revealed defective monocyte activation and dysregulated myelopoiesis with release of immature dysfunctional neutrophils into the circulation (251, 252, 277, 287, 319–323). More recently, high-dimensional flow cytometry analyses have identified a redistribution of monocyte subsets toward intermediate monocytes (a transitional population between classical and nonclassical monocytes that exhibits a hyperinflammatory signature) and the appearance of monocytic myeloid-derived suppressor cell-like cells (324–326). Defective monocyte and neutrophil responses render these patients highly susceptible to invasive fungal infections. However, despite their protective role, excessive phagocyte activation and/or recruitment can also cause lung damage and immunopathology, leading to increased susceptibility to coinfections (168, 327–334). Elevated levels of plasma granulocyte-macrophage colony-stimulating factor (GM-CSF) are observed in fatal COVID-19 cases, but not in influenza cases, and may explain the excessive monocyte and neutrophil recruitment leading to tissue destruction (335). Furthermore, a substantial induction of monocyte/macrophage and neutrophil-associated chemokines has been observed in the lungs of patients with severe COVID-19 (201, 251, 289, 336–339). While mechanistic studies on CAPA and IAPA have not been undertaken with the same level of resolution, the impact of these inflammatory processes on fungal secondary infection is evident from an in vivo model of influenza and Cryptococcus gattii coinfection. Increased neutrophil and macrophage recruitment into the lungs during influenza infection predisposed mice to more severe lung damage and increased fungal burden in the brain, resulting in increased morbidity and mortality (272). In addition, viral infection has been shown to strongly augment macrophage expulsion of Cryptococcus via a nonlytic mechanism (vomocytosis) which could potentially influence cryptococcal dissemination in the host (340).

The mononuclear phagocyte system and, in particular, tissue-resident monocytes and macrophages expressing the fractalkine receptor CX3CR1 are important in controlling fungal growth and dissemination at different tissue locations in mice and humans (341, 342). In the context of influenza infection, CX3CR1+ lung macrophages mediate pulmonary immune pathology and mortality through production of high levels of TNF and nitric oxide synthase 2 (NOS2) (343). However, under certain conditions, such as those involved in airway bacterial colonization, these cells could acquire an anti-inflammatory phenotype that controls influenza-mediated immunopathology (344, 345), suggesting that specific tissue environmental factors might affect the phenotype and function of these mononuclear phagocytes (MNPs). Importantly, CX3CR1+ MNPs are essential in sensing intestinal microbial dysbiosis and in shaping immune responses in the airways during homeostasis and inflammation (346–350). For instance, fungal dysbiosis and colonization with specific fungi in the gut can exacerbate the development of allergic airway diseases through fungal sensing by gut-resident MNPs (346–355). Bacterial dysbiosis as a result of antibiotic use has been associated with augmented severity of influenza infection and increased risk of developing secondary infections (345, 356, 357). Influenza infection itself can cause intestinal dysbiosis that contributes to secondary infections through alterations of the killing activities of alveolar macrophages (345, 357, 358) (Fig. 3). Alterations in the mycobiome have also been reported in COVID-19 patients (359), but whether this affects the functionalities of tissue-resident MNPs in a way comparable to that observed for influenza is unknown. Even though most of these mechanisms are still uncertain, these studies do suggest that MNPs could play a significant role in driving susceptibility to fungal coinfections in individuals suffering from severe viral pneumonia.

FIG 3

Gut-lung axis in the context of respiratory viral-fungal coinfections. The mycobiota plays a significant role in immunity and homeostasis in the intestine, which can influence immune responses at the airways. C. albicans-specific Th17 cells can confer protection against Aspergillus infection in the lungs. Importantly, intestinal microbial dysbiosis could affect functionalities of immune cells in the gut-lung axis, such as CX3CR1+ mononuclear phagocytes (MNPs), which in turn could influence susceptibility to fungal coinfections. Additional important questions remained unanswered. For instance, do these immune responses develop locally or traffic from the gut, or both? If they do migrate, what is their route of migration? What signals control it? (This figure was created with BioRender.)

Nucleotide-binding oligomerization domain (NOD)-like receptor family pyrin domain-containing 3 (NLRP3) inflammasome plays a critical role in the innate immune antiviral response (313, 360, 361) and in shaping adaptive immune responses (360, 362–364). The NLRP3 inflammasome is activated by tissue/cellular damage in a two-step process that is dependent on microbial and host-derived signals. First, NF-κB signaling is induced (i.e., through activation of TLR or TNF receptor signaling), resulting in increased expression of pro-IL-1β and pro-IL-18. Next, signal two (i.e., potassium efflux, uric acid, and mitochondrial damage among others) leads to complex assembly, activation of caspase-1, and secretion of IL-1β and IL-18, causing the recruitment of monocytes, macrophages, and neutrophils to the site of infection (364). Influenza viruses (312, 313, 365–367), as well as most respiratory viruses, including SARS-CoV-2, activate the NLRP3 inflammasome using viroporins (virally encoded hydrophobic proteins that oligomerize in the membrane of host cells, leading to the formation of hydrophilic pores) (157, 368–370). Nevertheless, several influenza virus-derived components (i.e., nonstructural proteins NS1 and PB1-F2) can inhibit the inflammasome, causing viral pathogenicity and immunopathology and increasing the susceptibility to bacterial coinfections (366, 371–377). Additionally, during influenza-associated bacterial coinfection, IL-1 signaling plays a protective role by preventing alveolar macrophage depletion (314) and supporting Th17 immunity (378). Notably, type-I IFNs can act as negative regulators of IL-1β expression and inflammasome activation in response to fungal pathogens, thus affecting fungal clearance (379, 380). Inflammasome activation, through the polysaccharide galactosaminogalactan, is important for protective responses during fungal infection (381–386), such as neutrophil recruitment. Therefore, this mechanism might be playing a crucial role in facilitating IAPA (Fig. 2). Excessive inflammasome activation can lead to uncontrolled inflammation (202, 387, 388), facilitating bacterial coinfections (389, 390) and potentially fungal coinfections. Tight regulation of the inflammasome is important to avoid hyperinflammation and immunopathology that might increase susceptibility to IAPA and/or CAPA. The importance of a balanced regulation of the inflammasome has been shown in patients with cystic fibrosis or CGD suffering from inflammasome-driven immunopathology and at risk for developing invasive fungal infections (271, 383, 384).

One of the most important effector mechanisms in host defense against A. fumigatus is the NADPH oxidase (NOX) complex. This is highlighted by the increased susceptibility to IPA in patients with CGD (302, 391). Conversely, ROS production during viral infection, including influenza infection, promotes virus pathogenicity and immunopathology. Therefore, regulation of ROS production could constitute a synergistic copathogenesis mechanism during viral-fungal coinfections. Oxidative stress during influenza infection induces formation of oxidized phospholipids that can result in acute lung injury and cytokine production by lung macrophages through TLR4 signaling (392). In fact, inhibition of NOX2 reduced lung injury and dysfunction, as well as lowering influenza burdens, suggesting that NOX2-derived ROS production promotes viral infection (392–397). Single-stranded RNA viruses, such as influenza virus, activate NOX2 in endocytic compartments of alveolar macrophages, resulting in endosomal hydrogen peroxide generation, which suppresses antiviral and humoral signaling networks (398). These data correlate with studies in mice deficient in NOX2 and in CGD patients, who have elevated circulating type I IFNs and autoantibodies, supporting the notion that low levels of ROS result in an enhanced immune response to viruses (399, 400). Modulation of NOX2-derived ROS production during influenza infection increased susceptibility to bacterial infections (304, 401). A fine balance of NOX2 activity and ROS production is therefore required to control viral infections and to improve coinfection outcomes. However, more studies are required to better clarify their precise roles during fungal coinfections.

Emerging Cellular Players

Other immune cells are emerging as important players during IAPA and CAPA. For instance, NK cells are crucial for direct killing of fungal pathogens as well as controlling the fungicidal activity of other immune cells such as neutrophils (402). NK cell-depleted mice have increased susceptibility to fungal pathogens, including A. fumigatus (168). NK cell functions are impaired during influenza and SARS-CoV-2 infection (250, 251, 321, 328, 403–406), which has been linked with increased susceptibility to coinfections (407). Of note, recent studies have shown that asymptomatic COVID-19 patients or those who have recovered have elevated levels of NK cells, which was not observed in patients with severe COVID-19, suggesting an important role in controlling disease severity (321, 404, 408, 409). Furthermore, circulating NKT cell frequency (a subset that features characteristics of both T cells and NK cells) was identified as a predictive biomarker for patient outcome (322).

Platelets are also emerging as key mediators of immune responses. Thrombocytopenia often coincides with neutropenia in patients at high risk for developing IPA (410). Importantly, an increased platelet count has been suggested as one of the main predictors of coinfections in patients suffering from severe influenza (411) and recently has been associated with disease severity in COVID-19 patients (412, 413). Interestingly, platelets express ACE2, and in vitro exposure to SARS-CoV-2 potentiates platelet activation and aggregation (414), which might be an important mechanism leading to the vascular complications observed in COVID-19 patients and increasing susceptibility to coinfections.

Dendritic Cells and Antigen Presentation

There are two major types of DC lineages, myeloid conventional DCs (cDCs) and lymphoid plasmacytoid DCs (pDCs) which both arise from DC precursors. During influenza infection, pDCs are a major source of type I IFNs, causing the expansion of antigen-specific T cells (415–418). pDC-derived type I IFNs also inhibit viral replication in airway epithelial cells following SARS-CoV-2 infection (419–421). They play a nonredundant role in host defense against Aspergillus infection. Recognition of Aspergillus hyphae by pDCs results in the release of proinflammatory cytokines, including TNF and IFN-α, and the formation of extracellular traps (422–424). More recently, it was shown that recruitment of pDCs into the lungs activates neutrophil NADPH activity to promote clearance of inhaled conidia (316). Importantly, severe COVID-19 patients show gene expression signatures of apoptosis in pDCs that correlate with reduced pDC frequency (278, 405, 419, 421). Dysregulation of pDCs might contribute to depressed IFN signatures affecting susceptibility to fungal coinfections.

In contrast, cDCs are important players in antigen presentation and activation of T cells that underpin adaptive immune responses. Upon antigen recognition, uptake, and processing, DCs acquire a fully mature state and migrate to the draining lymph nodes, where they present antigen-derived peptides in the context of major histocompatibility complex (MHC) molecules to CD8+ T cells or CD4+ T cells (425). Impairments of DC functionalities following severe viral pneumonia can be classified as having short-lived and long-term effects. Short-lived effects are reversible as soon as the viral infection is cleared and include modulation of DC antigen presentation capabilities and interference with signaling pathways. It is well recognized that some viral pathogens, in particular DNA viruses such as herpesviruses, can interfere with the antigen presentation pathway (426, 427), while RNA viruses, including influenza virus, by an unknown mechanism seem to preferentially target the cross-presentation pathway (which occurs when exogenous antigens, normally loaded into major histocompatibility complex class II [MHC-II] molecules, are shuttled into the MHC-I pathway) (428–432) (Fig. 2). DCs are susceptible to SARS-CoV-2 infection, which attenuates the IFN response via viral antagonism of STAT1 phosphorylation (433). However, whether these mechanisms have implications during secondary fungal infections is unknown.

Heterologous immunity can significantly impact DC phenotype and the ability of DCs to activate optimal immune responses to fungi. For instance, desensitization of pattern recognition receptors (PRRs) (i.e., TLR2, TLR4, and TLR5) associated with reduced chemokine production and NF-κB activation has been reported during viral infections (434, 435). In vitro studies have suggested that exposure to a viral infection affects DC cytokine production to a subsequent secondary challenge (436, 437). In addition, upregulation of inhibitory signals, such as CD200R, desensitizes APCs (DCs and macrophages), increasing their threshold of activation, a mechanism shown to contribute to bacterial coinfections (204) (Fig. 2). Nevertheless, in COVID-19 patients, CD200R expression was shown to be reduced in peripheral blood DCs, which could contribute to the expression of proinflammatory cytokines, tissue damage, disease severity, and mortality (324). More recently, it has been suggested that SARS-CoV-2 infection results in significantly reduced numbers of DCs, with functional impairment reflected in reduced maturation and cytokine production necessary to perform antigen presentation to activate T cells (324, 337, 438–441). Some of these long-lasting mechanisms could persist for weeks or even months after recovery, considerably increasing susceptibility to fungal coinfections. Remarkably, preexposure to Pneumocystis results in enhanced antigen processing, maturation, and trafficking abilities of DCs, which causes an accelerated influenza virus-specific primary immune response and viral clearance (442).

CLRs expressed by myeloid cells, including DCs and macrophages, are crucial for tailoring immune responses to pathogens. A recent study showed that several CLRs, including dendritic cell-specific intercellular adhesion molecule 3-grabbing nonintegrin (DC-SIGN), liver/lymph node-specific intercellular adhesion molecule 3-grabbing integrin (L-SIGN), liver/lymph node sinusoidal endothelial cell C-type lectin (LSECtin), and macrophage galactose-type lectin (MGL), participate in SARS-CoV-2 recognition and induction of proinflammatory mediators (IL-1β, IL-8, CXCL10, CCL2, and CCL3) that correlate with disease severity (443, 444). DC-SIGN also mediates binding and internalization of A. fumigatus conidia by DCs (445), and it is expressed by alveolar macrophages in the lung (446). However, it is not clear whether DC-SIGN is required for activation of innate immune signaling or if it is involved in the initial stages of Aspergillus pulmonary infection and dissemination (446). Importantly, DC-SIGN polymorphisms are associated with the development of IPA, suggesting that it might be involved in the pathogenesis of this infection (447). Therefore, the potential convergence of CLR-driven innate and/or adaptive immune responses in the setting of SARS-CoV-2 and Aspergillus coinfection should be further explored, as it might influence copathogenesis and disease progression.

Resolution of lung inflammatory disease after influenza virus infection sets a different threshold for innate immune activation (204). This altered homeostasis could have a significant impact on the threshold of responsiveness to the next pathogen. Engagement of PRRs on the surface of APCs induces epigenetic functional reprogramming that affects their sensitivity to a second challenge, a process referred to as “trained immunity” (448). Even though myeloid cells are typically short-lived during inflammation, these phenomena can occur in bone marrow progenitors (448–452) and impact monocytes, macrophages, and DC populations during fungal infections (453). In particular, TNF production during invasive cryptococcosis induces a stable state of DC phenotypic programming (DC1/M1-like), rendering the DCs resistant to both antigen- and cytokine (IL-4)-induced alternative activation (DC2/M2-like). This reprogramming was also shown in bone marrow DC precursors and was demonstrated to be essential for Th1/Th17 immune protection (454). Interestingly, DC differentiation from bone marrow precursors is impaired during the course of a viral infection, leading to susceptibility to secondary infections (455, 456). In addition, recent studies have shown that DCs undergo a metabolic reprogramming early during influenza infections that results in significant changes in innate immune functions of DCs, including reduced motility and T cell activation (457). How long this reprogramming persists and whether it may impact fungal coinfections are unknown and will require more study.

Several functional studies have established the role for specific DC subsets (pDCs, cDCs, and inflammatory DCs) in immunity against influenza virus infection (458–462). However, we still know very little about the role of different pulmonary DC subsets during fungal infections (463). Therefore, substantially more work is needed to separate which subsets are needed at which phase of the response to prevent IAPA or CAPA.

T Cell Responses

CD4+ T cells produce several antimicrobial soluble factors that control the spread of viral and fungal infections. They are also required for class switching of antibodies and for optimal CD8+ T cell memory responses (4). The type I IFN response during influenza infection can directly affect adaptive immunity to fungi by dampening Th17 responses via the suppression of IL-23, a cytokine that is crucial for the expansion and maintenance of Th17 cells (464). Defective Th17 responses can lead to attenuation of the AMP production and neutrophil recruitment required for antifungal clearance (168, 465). This suggests that defective Th17 responses are a significant factor during IAPA, as has been shown during bacterial coinfections (464–467) (Fig. 2). Th1 responses are important in controlling systemic fungal dissemination (169), but they might be detrimental during respiratory fungal coinfections. STAT1 knockout (KO) mice, which have defective Th1 cell differentiation and increased Th17 immune activation, are less susceptible to coinfections than wild-type controls (467). Remarkably, it has been shown that COVID-19 patients with moderate disease displayed a progressive reduction of both antiviral (characterized by the dominance of the transcription factor T-bet and the expression of IFN-γ) and antifungal (orchestrated by the RORγt-induced cytokines IL-17 and IL-22) responses whereas patients with severe disease maintained these elevated responses throughout the course of the disease (200). This suggests that dysregulation of T cell responses could be an important factor that contributes to the development of fungal coinfections in COVID-19 patients.

Another important T cell population that plays a significant role during recovery and resolution of inflammation is Foxp3+ regulatory T cells (Tregs). Influenza virus-specific Tregs can be detected for a prolonged time after viral clearance (468) and can reduce expansion of CD4+ and CD8+ effector T cells (469–471) while suppressing neutrophil-driven cytokine release into the airways, contributing to the resolution of disease (472). This Treg-mediated dampening of inflammation likely impacts immunity to subsequent infections such as Aspergillus infections (5). Indeed, Tregs can control immunity and tolerance to Aspergillus at different stages of the immune response (473, 474). However, early during infection, they might have a detrimental role by suppressing neutrophil functions through secretion of IL-10 and expression of cytotoxic T lymphocyte-associated protein 4 (CTLA-4) (which acts to inhibit T cell activation) (475) (Fig. 2). Tregs can also facilitate processes of tissue repair that can further sustain this anti-inflammatory state (306). Functional assessments of how the altered dynamics of T cell populations during viral infection affects antifungal capabilities could help to understand their role during fungal coinfections.

Adaptive Humoral and Cytotoxic Responses

B cell and CD8+ T cell responses are critical for pathogen neutralization and clearance and play a major part in the memory response that prevents reinfection (4, 476, 477). Respiratory viral infections could influence susceptibility to subsequent fungal infections by targeting adaptive immune components at different levels, including cross-presentation, CD8+ T cell activation, and B cell activation and antibody production. The contribution of CD8+ T cells in host immunity against fungal infections in not well understood, and several studies have suggested that they might play a protective role in the setting of CD4+ T cell deficiency (478–480). Studies have suggested that influenza viruses might interfere with the cross-presentation pathway (428–432). In particular, DCs that capture dead cells containing influenza virus are unable to activate CD8+ T cell clones specific to cell-associated antigens of captured dead cells (431). Moreover, influenza virus-infected DCs exhibit impaired cross-presentation of influenza virus-derived and other exogenous antigens (430). Although it has not yet been demonstrated, these mechanisms could have significant implications during fungal coinfections.

The role of protective humoral immunity against fungal infections is not well defined, and it is still very controversial (481–485). Therefore, studies focusing on understanding the role of humoral immunity during viral and fungal coinfections are scarce. Of note, monoclonal antibodies targeting Candida albicans can confer protection against lethal pulmonary infections by Gram-negative bacteria in mice (486), while Pneumocystis infection can protect mice against subsequent influenza infection due to the enhancement of the influenza virus-specific antibody response (487). Susceptibility mechanisms could include the activation of low-affinity cellular responses and/or production of nonneutralizing antibodies that could facilitate coinfections (488–490). Interestingly, meningeal IgA plasma cells that are dependent on the presence of gut commensals can confer protection against fungal invasion into the brain (484). Therefore, alteration of these fungus-specific antibodies could facilitate fungal dissemination during coinfections (Fig. 3).

Cross-Reactive Immunity

The immune response to a pathogen is greatly influenced by the individual’s immune history (491, 492). Cross-reactivity of adaptive immune lymphocytes can either confer protection or drive susceptibility to subsequent infections. Heterologous immunity refers to the ability of one pathogen to modify the immune response to a related or unrelated pathogen that could boost or weaken protective immunity, break tolerance, or induce immunopathology (488). This phenomenon has been shown for closely related but also unrelated pathogens, including parasites, protozoa, bacteria, and viruses (493–495). Literature on heterologous immunity to fungi is limited. One study shows that a modified heat-labile bacterial toxin (LTK63) improved the immune response to subsequent infection with Cryptococcus neoformans by increasing pathogen-specific CD8+ T cell and IgA responses in the nasal mucosa (496). Furthermore, segmented filamentous bacterial colonization during pulmonary Aspergillus infection augments antifungal Th17 immunity (172). Interestingly, heterologous immunity to a single, ubiquitous member of the fungal microbiota is fundamental for systemic induction of protective antifungal Th17 responses and immunopathology (355). In particular, the commensal C. albicans is the main inducer of Th17 responses in peripheral blood that later can be expanded in the lungs by cross-reactive airborne A. fumigatus (355). Further studies suggest that priming of these fungal cross-reactive T cells by gut commensal fungi and selective recruitment of these cells to the lungs may be important factors in the pathogenesis of inflammatory airway diseases (355, 497) (Fig. 3). Therefore, it is plausible that immune responses to commensal fungi may be altered during chronic viral infections, facilitating fungal coinfections. Interestingly, C. albicans colonization does not confer protection against influenza virus infection and rather exacerbates allergic airway inflammation susceptibility, indicating that fine-tuning of T cell responses is required to control immunity versus immunopathology (355, 497). In this context, “pathogenic” Th17/Th1 versus “anti-inflammatory” Th17/Treg mixed responses could play differential roles during influenza-associated coinfections (170).

FROM CLINICAL OBSERVATIONS TO COPATHOGENESIS

Clinical studies have shown a huge variation in the incidence of IAPA and CAPA. This variation in incidence might be explained by the interplay of a number of factors with the copathogenesis of respiratory viral and fungal coinfections, including differences in environmental and/or genetic factors, type of circulating viral strain, treatment modalities for the critical illness, and the use of and access to fungal diagnostic tools. Environmental conditions can modulate host immune responses, including mucociliary clearance, tissue repair functions, and innate immune defenses (498), as well as outbreaks of viral respiratory diseases by influencing virus stability and transmission rates (499–501). Seasonal fluctuations in airborne fungal spore levels have also been determined for different genera, with the dominant genera varying depending on geographical location (502–510). In some regions, seasonal variations in total airborne fungal counts have been shown to correlate with different environmental factors, including temperature, humidity, rainfall, and wind speed (502–507). Whereas most studies have suggested that A. fumigatus is present at low but persistent levels in the outdoor environment (502), there are possible geographical links that need to be more fully explored. A study from Brazil showed that Aspergillus spp. were among the dominant species found in both indoor and outdoor environments (506), while a study from the Netherlands found that Aspergillus was present all year round and prevailed in the autumn and winter months (507). Interestingly, a study from Canada described a positive association between Aspergillus hyphal fragments and wind speed (504).

Genetic factors associated with increased susceptibility to influenza viral infection include genes involved with viral recognition and IFN signaling (Ifitm3, Irf7, Ifnar1, Ifnl1, Stat1, Sfpta1) (511). Another set of genes are associated with disease exacerbation during influenza infection, including genes involved in inflammation (Par1, Tnfaip3, Nos2, Ptges2, and Ifi35) and tissue homeostasis (Epg5, Atg14, and Atg7) (511). This is consistent with the idea of immunopathology contributing to disease severity and susceptibility to coinfections. Recent studies have identified several host factors critical for SARS-CoV-2 infection, including genes involved in cholesterol biosynthesis, autophagy, viral entry, and phosphatidylinositol biosynthesis, among others (512–517). Genes involved in fungal recognition and effector mechanisms (Ptx3, Tlr4, clec7a, Clec1a, Plg, Cxcl10, Ifng, Il10) are associated with increased susceptibility to invasive aspergillosis in patients undergoing hematopoietic stem cell transplant. A second set of genes (S100b, Rage, Nod2) are involved with hyperactivity of innate recognition pathways (518–520). These susceptibility mechanisms have also been reiterated in studies of other vulnerable populations, including patients with COPD, solid organ transplant recipients, and patients with hematological malignancies. Nevertheless, genetic studies are required to determine their role in a population of patients suffering from severe viral pneumonia.

Emergence of highly pathogenic viral strains increases susceptibility and modifies the kinetics of coinfections (521). Viral polymorphisms that alter the tropism of influenza viruses from the upper to the lower respiratory tract and facilitate bacterial coinfections (5) could potentially increase susceptibility to fungal coinfections. Major subtypes of influenza virus strains linked to IAPA include H1N1, H5N1, and H7N9. Influenza virus strain H5N1 increases the production of proinflammatory cytokines and enhances viral replication in the lung, causing immunopathology and pulmonary fibrosis (209, 376, 522–524), which are important contributing factors that drive secondary infections (168, 245, 327–334, 366, 371–377). In contrast, strain H7N9 can inhibit the inflammasome (372), an important effector mechanism against viral and fungal infections. In the case of CAPA, many novel SARS-CoV-2 variants which have different clinical effects are emerging. The B.1.1.7 (alpha, United Kingdom), B.1.351 (beta, South Africa), P.1 (gamma, Brazil), B.1.427/29 (epsilon, USA), and, most recently, B.1.617.2 (delta, India) variants have all shown to be highly transmissible, with some studies suggesting an association with higher mortality and escape from natural and vaccine-induced immunity (525–527). However, whether these novel variants are associated with increased susceptibility to fungal coinfections is unknown. Of note, some authors have suggested an association between the delta variant and the emergence of mucormycosis in India (528, 529); however, so far there are no specific data to support this hypothesis.

Treatment modalities for severely ill patients could also increase susceptibility to respiratory fungal coinfections following severe viral pneumonia. This includes the use of antibiotic, antiviral, and/or immunomodulatory treatment. For instance, the use of antibiotics to prevent secondary bacterial infection can cause dysbiosis, a condition that has been linked to increased severity during respiratory viral infections and susceptibility to secondary infections (345, 356, 357). The use of neuraminidase inhibitors has been suggested to increase susceptibility to fungal coinfections following influenza infection (44) and has been recently demonstrated to increase the susceptibility of mice to invasive aspergillosis (530). The use of the corticosteroid dexamethasone as an immunosuppressive drug to treat ARDS, which was shown to reduce mortality in seriously ill COVID-19 patients (531), is one of the major risk factors for developing CAPA and negatively affects immunity to Aspergillus (532–534). Several studies have reported a relationship between the use of steroidal immunosuppressant (corticosteroids) and the incidence of IPA in critically ill COVID-19 patients (56, 73, 82, 89, 535–539). A prospective study from ICUs in Wales showed that the use of high-dose systemic corticosteroids increased the likelihood of developing CAPA (16 out of 22 patients [72%] compared to 32 out of 57 patients without CAPA [56%]) (2). Similar findings were observed in studies focused on patients with severe influenza (24, 540). However, some studies have found no association between the use of corticosteroids and the incidence of hospital-acquired fungal infections in patients with COVID-19 (541, 542), which has been suggested to be due to an early administration of low-dose corticosteroids for a short period (541). Importantly, in some of these studies, additional risk factors might be at play, including a history of chronic respiratory disease. Recent clinical trials have suggested that combining corticosteroids with anti-IL-6 treatment (tocilizumab) improves the outcomes in terms of morbidity and mortality in severe COVID-19 patients (543, 544). Nevertheless, COVID-19 patients receiving tocilizumab are reported to be at higher risk of developing CAPA (65, 69, 82, 545, 546). As IL-6 is essential for inducing protective Th17 responses (170–172) and controlling the effector functions of phagocytes (547), anti-IL-6 therapy increases the susceptibility for fungal infections. Earlier studies have documented an increased risk for developing bacterial and fungal infections in patients receiving tocilizumab in combination with corticosteroids for the treatment of rheumatoid arthritis (548). A retrospective study conducted in Chicago involving 111 COVID-19 patients found that those receiving tocilizumab had a higher risk of developing fungal infections and increased mortality (549). Strikingly, in a prospective study from Spain that includes 2,723 patients with COVID-19, all CAPA patients who received tocilizumab and corticosteroids had a fatal outcome (8 out of 8 patients) (89). Prospective monitoring of these patients is needed to shed light on whether tocilizumab negatively impacts the susceptibility to and outcome of respiratory fungal coinfections. Furthermore, metanalyses of retrospective studies could help to elucidate the role of immunosuppressive agents in predisposing COVID-19 patients to fungal coinfections.

The epidemiology of, and the mortality associated with, coinfections complicating viral respiratory tract infections is difficult to assess with confidence. This is partly due to the fact that diagnosis of sequential infections is challenging, as the primary pathogen is often no longer detectable by the time the secondary infection presents itself (550). Alternatively, if both infections present themselves at the same time, one could quickly override the other. Furthermore, the presence of a specific pathogen may be part of the airway commensal community rather than an indication of infection and disease (551). Therefore, the specific attribution of the primary viral infection and coinfections to mortality is complex to unravel. In comparison to bacterial coinfections in viral respiratory disease, differentiation of fungal colonization versus infection and disease is even more challenging. Aspergillus spp. are ubiquitous in the environment, and as a consequence, our airways are exposed on a daily basis to its spores (conidia). In immunocompetent healthy individuals, these spores are cleared without infection and disease, while in immunocompromised patients and those with chronic lung disease, spore inhalation can lead to pulmonary aspergillosis (235). Diagnosis of pulmonary aspergillosis is based on a combination of criteria that includes host factors, clinical and radiological features, and mycological studies (11, 552). However, cultures from respiratory samples do not differentiate colonization from disease, antigen testing in serum has a low specificity, and specific changes on chest imaging and invasive diagnostics (e.g., bronchoscopy) are often not feasible due to the critical clinical condition of the patient and the risk of aerosolization in cases of viral pneumonia. The absence of rapid, sensitive, and specific fungal diagnostic tools is a major challenge. Optimal availability of, access to, and implementation of fungal diagnostic tools in routine clinical care will affect incidences reported and provide a real insight into the clinical epidemiology and burden of disease.

FUTURE PERSPECTIVES

Viral-fungal coinfections are increasingly being recognized by the scientific and medical communities. The urgent need to obtain insight into the epidemiology, pathogenesis, and underlying immune mechanisms is driven by the additional mortality among patients with IAPA and CAPA (288). Comprehensive epidemiological data are lacking, and therefore data from larger cohorts of patients are required to better assess the incidence, clinical features, and detailed characteristics of secondary fungal infections in influenza and COVID-19 patients, especially with the highly pathogenic emerging strains (553). Early diagnosis of fungal infection is critical for effective treatment. Current diagnostic methods lack sensitivity and specificity to differentiate colonization from infection and invasive disease. Systematic prospective studies that employ uniform diagnostic testing and criteria are urgently required to optimize the management of patients in the ICU, and first initiatives are been undertaken in this area (119, 133, 554, 555). Azole-resistant A. fumigatus causing infections in COVID-19 (556–558) and influenza (47, 559, 560) patients is emerging, highlighting the importance of early causative diagnosis and surveillance during antifungal therapy. Developing treatment modalities in which both the virus and the fungus are being targeted could also be proven useful. Using unbiased approaches, studies have identified several immunotypes in hospitalized COVID-19 patients that could predict clinical outcome and disease trajectory (200, 561, 562). Similar longitudinal studies could prove of value in identifying comparable profiles predicting the presence of fungal infections in these patients, which could have significant implications for therapeutic interventions.

Our understanding of how host-pathogen interactions are affected during polymicrobial infections is limited. We have summarized a number of immune pathways and mechanisms that may play a crucial role in the copathogenesis of viral and fungal lung infections. Interindividual heterogeneity of the immune system that is shaped by divergent exposure of immune cells to infections, vaccination, and lifestyle-related stimuli (diet, physical activity, and stress) will influence an individual’s risk for acquiring fungal infections. Improved tools and models to study coinfections are needed to obtain better insight into the copathogenesis and immune pathways driving disease. Preliminary studies have shown a strong link between high viral replication and increased susceptibility to fungal coinfection, which suggest that the timing of coinfection is important in determining susceptibility and disease outcome (272). Influenza infection induces several long-lasting changes at molecular levels, as shown by transcriptome, proteome, and metabolome analyses, which could affect susceptibility to fungal infections (563). Studies with convalescent-phase samples from COVID-19 patients have suggested that the immune system does not fully recover after infection (250). As with postacute viral syndromes described in survivors of other SARS epidemics, there are increasing reports of persistent and prolonged effects after acute COVID-19 beyond 1 month from the onset of symptoms (564, 565). Long-term follow-up studies are needed to investigate the consequences on fungal immunity. Importantly, long-term metabolic dysregulation influences disease trajectory and immune response to COVID-19 (566–568) and might also affect susceptibility to secondary fungal infections. Host genetic factors such as polymorphisms in key immune receptors and signaling molecules involved in fungal sensing could also be playing a key role. Increasing our understanding of the copathogenesis of respiratory viral-fungal coinfections and the impact of the microbiomes in this interplay could help to develop better diagnostics and therapeutic modalities against newly identified targets.