Editorial: RSV, dendritic cells, and allergens—a bad combination

Abstract

Discussion of the central role of DC in how RSV primes the respiratory tract to bias subsequent pulmonary mucosal responses.

Respiratory viral infections in early childhood, in particular, with RSV, have been associated with the later development of asthma [1]. Pulmonary disease, characterized by inflammation, airway hyper-reactivity, and mucus hyperplasia, can be induced by combining RSV with allergic sensitization, and RSV can prime the lung for a more severe response to subsequent exposure to model antigens [2]. A most-severe form of this pulmonary response to RSV occurred in infants infected with RSV following vaccination with FI-RSV more than 40 years ago. The mechanism(s) for this inflammatory asthma-like phenotype are still incompletely understood [3] but seem critical to untangle how viral respiratory infections are linked to atopy and asthma and also for vaccine development. The study by Jang et al. [4]in the Journal of Leukocyte Biology further explores the contribution of DCs to this RSV-triggered pulmonary disease by studying the effect of RSV to subsequent allergic sensitization. Although a Th2-biased pulmonary response is associated with experimental RSV infection [5], the overall immune response triggered by RSV is complex, involving the mucosal and systemic innate and adaptive immune system [3]. Airway neutrophilia and dysregulated lymphocyte responses may play the dominant role in the lung disease induced by RSV [6, 7]. However, lung DCs link adaptive to innate immune responses and play a critical role for RSV pulmonary disease [2, 8]. The study demonstrates that RSV infection primes for an enhanced pulmonary response to subsequent sensitization to cockroach antigen in two mouse strains with varying susceptibility to RSV infection. This pulmonary inflammatory disease phenotype resembles pulmonary disease triggered by FI-RSV. The role of DCs and DC trafficking in this process is demonstrated: (1) by a blunted inflammatory response to allergic sensitization, as well as RSV infection, in CCR6 and CCR7 knockout mice and (2) by the observation that a similar disease phenotype could be induced by intratracheal instillation of GM-CSF-cultured syngeneic bone marrow-derived DCs. The fact that pulmonary disease is elicited by ex vivo exposure of DCs to RSV as well as in vivo administration of the virus to the respiratory tract suggests similar underlying mechanism(s). It cannot be concluded from the presented data whether significant infection of the DC by RSV was necessary to trigger the pulmonary disease. There was only a small amount of viral RNA present in the DC, and the authors suggest that the majority of the effects was secondary to skewed antigen presentation rather than infection. Interestingly, the pulmonary disease seen with RSV and allergen was not triggered by influenza and allergen. The authors suggest that differences in the ability of the viruses to induce type 1 IFNs could be responsible. Their findings with influenza, however, are in contrast to studies demonstrating enhanced pulmonary disease, triggered in mice following influenza respiratory infection and allergic sensitization with house dust mites [9]. Differences in the allergen could be responsible for the different results. However, infection of mice with Sendai virus, the murine parainfluenza virus type 1, followed by sensitization with OVA, also resulted in pulmonary disease, characterized by enhanced IgE production, airway hyper-reactivity, and mucus metaplasia in a partial FcϵRI-dependent mechanism [10] that depended on expression of FcϵR1 on lung DCs [11]. It can be speculated that similar DC-related mechanisms of viral respiratory infection and allergic sensitization can occur with other common viral respiratory infections associated with asthma, such as metapneumovirus, adenovirus, and rhinovirus. Infection of the lower respiratory tract at a young age with those viruses or RSV leads to damage of the airways and airway remodeling and triggers subsequent recurrent wheezing. Whether these infections early in life cause asthma directly or simply affect infants who are genetically predisposed to develop asthma is still being debated [1]. As a causal relationship has not been established, it seems more likely that genetic factors favor both—susceptibility to severe infection and to asthma. This is difficult to model in mice.

As acknowledged by the authors, their study has only begun to address the question of how anti-RSV, together with nonviral antigen immune responses, alters the overall pulmonary immune response. A caveat may be that only bone marrow-derived DCs were studied here. Various DC subtypes populate the lung (Fig. 1) that varies in their immune-stimulatory capacity. Besides their location, where mucosal DCs seem most potent in antigen uptake and presentation, lung DCs are categorized based on surface marker expression into epithelial and subepithelial cDCs, pDCs, and an inducible inflammatory subtype, primarily derived from monocytes. Relevant to enhanced pulmonary disease following RSV and nonviral allergen exposure, overall, lung DCs seem primarily biased to promote Th2 responses [2]. This is then altered toward a Th1 or Th17 response, based on concomitant inhalation of adjuvants, such as particulate matter and bacterial contaminants. Lung pDCs may have a protective role in the enhanced RSV pulmonary disease, as depletion of pDC worsens bronchial hyper-reactivity and airway eosinophilia [12]. GM-CSF-induced bone marrow-derived DCs, as used by Jang et al. [4], are probably most similar to the inflammatory DC subtype. It would be interesting to study how transfer of specific DC subtypes would alter the subsequent virus/allergen response.

Figure 1. Infection of the respiratory tract with RSV or transfer of RSV-infected prime for a similar phenotype of enhanced lung disease.

Lung DC subtypes that interact with RSV or RSV-infected cells include cDCs expressing CD11b or CD103, pDCs, and monocyte-derived inflammatory DC. Subsequent exposure to an allergen or RSV results in a Th2-dominant cytokine response, bronchial hyper-reactivity, and enhanced mucus production. Transfer of RSV-infected bone marrow-derived DC (BMDC) results in a similar phenotype that seems dependent on trafficking the DCs to local LNs.

Besides DCs, other resident immune cells, in particular, mast cells and macrophages, are critical in shaping the type of antiviral and allergic pulmonary immune responses [13]. Also, the contribution of nonimmune lung cells, in particular, epithelial and endothelial cells on DC responses, needs to be studied further. Their roles for providing critical cues to lung DC responses to viral antigens and allergens are becoming increasingly recognized [14]. Epithelial cells produce CCL20 and ligands for CCCR6. Triggers for this can be, among others, allergen inhalation and RSV infection [8]. CCR2 seems to be the relevant chemokine receptor responsible for the accumulation of DCs following allergic sensitization and challenge, mainly via release of monocytes from the bone marrow. The role of CCR6 seems to direct DC transit to the airways from the interstitium. Jang et al. [4] show that DCs, instilled directly into the trachea, ended up in the local LNs. It is unknown where and how that exactly occurred. The fact that the transfer to CCR7-deficient DCs did not elicit an enhanced response suggests that intact cells were trafficking directly and that uptake of cellular debris and antigens by resident lung DCs played a major role in the pulmonary disease. Rapid trafficking of labeled resident lung DCs to local LNs has been described following influenza [15] and Sendai virus infection in mice, which was partially dependent on CCR5 [16]. Antigen transport through DCs to mediastinal LNs also seems dependent on CCR8 and CCRL2 [17]. Interestingly, when DCs from CCR7 knockout mice, exposed to RSV in vitro, were instilled into the trachea, the exaggerated RSV response was blunted, suggesting that CCR7 plays a critical role in this process. The fact that enhanced pulmonary disease could still be elicited 3 weeks following transfer of the RSV-infected bone marrow-derived DC, suggests that T memory was induced. This is consistent with the potential of lung DC to induce T effector memory cells [18].

Overall, the findings by Jang et al. [4] provide more insight in the complex and still puzzling power of the interaction of RSVs (and other respiratory viruses) with the respiratory tract to bias subsequent pulmonary inflammatory and remodeling responses. They further confirm a central role for DCs in this problematic process and help better understand enhanced RSV lung disease, which could be critical in the design of new, antiviral strategies and vaccines. More work needs to be done to better understand the contribution of DC subtypes—other lung immune cells, in particular, macrophages, basophils, and mast cells, as well as epithelial and endothelial cells—in this complex process.