The role of viruses in development or exacerbation of atopic asthma.

Respiratory viral infections in early childhood have been linked to the development of persistent wheezing and asthma. Epidemiologic data indicate that, for the majority of children, virus-induced wheezing is a self-limited condition, with no long-term consequences. For a substantial minority, however, virus-induced wheezing is associated with persistent asthma and the potential for enhanced allergic sensitization. For the most part, this subset of patients is genetically predisposed; they are atopic children in whom respiratory viral infections trigger the early development of asthma by mechanisms that have not been fully elucidated. Both inflammatory and noninflammatory mechanisms may be involved. It does not appear that viral infection per se in early life is responsible for the induction of atopic asthma. Data from animal models provide support for the concept that enhanced allergic sensitization caused by increased uptake of allergen during infection may play a critical role, as well as T-cell-mediated immune responses to viral infection, which may favor eosinophilic inflammatory responses and the development of altered airway function to inhaled methacholine. Recent advances in our understanding of the interactions between respiratory viruses and the development of reactive airway disease offer new possibilities for preventive treatment in children at risk for developing persistent wheezing and asthma exacerbation as a result of viral infection.

EPIDEMIOLOGY OF VIRAL RESPIRATORY INFECTIONS AND ASTHMA DEVELOPMENT

An association between the development of asthma in childhood and viral respiratory tract infections has been recognized for several decades. Acute viral infections are important triggers of wheezing in children and asthma exacerbation in both children and adults. 7, 16, 55 Using polymerase chain reaction-based methods of detection, recent studies found that more than 80% of wheezing episodes in school children were associated with viral respiratory tract infections. In more than 60% of these patients, rhinovirus (RV) was detected. In infants and toddlers under the age of 2 years, the virus most frequently isolated during wheezing episodes is respiratory syncytial virus (RSV). 39, 52 Parainfluenza viruses (PIV), corona virus, adenovirus, influenza virus, and enteroviruses have all been implicated in the development or exacerbation of wheezing. 39, 57 Some of the epidemiologic evidence suggests that respiratory viruses may both trigger asthma exacerbation and contribute to the enhancement of allergic sensitization and subsequent development of allergic asthma. An association between viral respiratory tract infections and the onset of allergic sensitization in children born into allergic families has been proposed.

The close link between virus-induced bronchiolitis and development of asthma has been recognized in several studies. 21, 53, 73, 74, 76, 82 In a prospective cohort study with matched controls, RSV bronchiolitis in infancy was identified as the most important risk factor for the development of asthma and sensitization to common allergens by the age of 3 years. This risk was further increased when there was a family history of asthma or atopy. By the age of 7 years, asthma was still more prevalent in the group of children who had RSV bronchiolitis in the first year of life than in the control population. Some studies indicate that RSV bronchiolitis at an early age results in bronchial hyperreactivity persisting for more than 10 years and may even be associated with chronic pulmonary disease persisting into adulthood. 3, 4 A criticism of these epidemiologic studies has been that by selecting children with RSV disease severe enough to require hospitalization, the selection might also favor those with an atopic predisposition or intrinsic abnormalities of airway function. Further, a number of observations indicate that atopy may alter the response to respiratory viral infections, resulting in more severe disease. 18, 51 An atopic predisposition may not be required for the development of bronchial hyperresponsiveness following RSV bronchiolitis, however. 73, 74, 82, 83

Data from the Tucson Children's Respiratory Study, a large prospective birth cohort study, has provided additional important information indicating that a majority of children with wheezing episodes in the first 3 years of life did not develop persistent asthma. In this group of transient wheezers, no significant association with allergic sensitization was detected, but reduced airway function, indicated by a low Vmax forced residual capacity measured before the first viral respiratory infection was a significant risk factor for the development of wheezing. 48, 49, 51 For these children, who may have smaller airways than children who never wheeze, wheezing in the first 3 years of life seems to be a self-limiting condition, without long-term sequelae. In a substantial minority who were “persistent wheezers”—i.e., children who developed wheezing in the first 3 years of life that persisted beyond the age of 6 years, wheezing was associated with allergic sensitization and a history of maternal asthma. These children also had elevated serum levels of total immunoglobulin (Ig)E during the acute phase of their first lower respiratory tract infection as well as eosinophilia in the peripheral blood, in contrast to transient wheezers and children without asthmatic symptoms during acute infection. Furthermore, these children demonstrated reduced airway function at the age of 6 years, having had normal lung function after birth. At the age of 11 years, this group showed similar characteristics.

A stratification of a birth cohort according to the viruses detected during respiratory tract infection confirmed that the majority of infections before the age of 3 years were associated with RSV. For this group of children with RSV infection, an increased odds ratio for wheezing, both infrequent and frequent, was demonstrated at the age of 6 years. Interestingly, this risk diminished with age and was not significant by 13 years of age. Furthermore, the rate of allergic sensitization did not differ between children with or without respiratory tract infection at any age and the authors concluded that RSV infection was not associated with allergic sensitization and that both atopy and RSV infection are independent risk factors for the development of asthma. The reversible reduction in airway function suggested that RSV infection may induce a dysregulation of airway tone but provides no evidence for fixed abnormalities of airway function as a consequence of infection.

Viruses other than RSV and PIV infecting the respiratory tract also may increase the risk for persistent wheezing, which does not decrease with age. The sum of the epidemiologic data available indicates that viral respiratory tract infections in the very young may cause transient asthmatic symptoms or may trigger the early development of asthma in children with a genetic predisposition to atopy, rather than inducing allergic sensitization in and of itself. This raises the possibility of synergistic interactions between respiratory tract viral infections, predisposition to atopy, and sensitization to aero-allergens, resulting in the development of persistent airway inflammation and asthma. Such interactions have been demonstrated in children presenting to an emergency room, where the combination of atopy and concurrent respiratory tract viral illness was associated with the greatest risk for asthma symptoms. Overall, the relationships between viral respiratory infections and development of persistent wheezing or asthma are complex and still ill-defined. Nonetheless, the prevailing view is that viral infections, including RSV, are not a major factor in the induction of atopic asthma.

In contrast, the role of viral infection in asthma exacerbations appears much clearer; such interactions have also been confirmed in a human model of experimental RV infection. Adult volunteers suffering from allergic rhinitis were infected with RV16 prior to bronchoprovocation with allergen. Rhinovirus infection altered the pattern of response to allergen exposure, enhancing the asthmatic late-phase response.

INFECTION AND INFLAMMATION BY RESPIRATORY VIRUSES

The mechanisms by which respiratory viruses could contribute to the development of asthma are slowly being defined (Fig. 1) . Development of allergic asthma reflects the interplay between strong genetic predisposition and the development of an airway inflammatory response following allergen or other exposures, most often characterized by infiltration of eosinophils and neutrophils. The consequence of a respiratory virus infection may be dictated by the host's immune response to the virus; in most cases, this response results in the relatively rapid curtailment of the infection. In some individuals, however, the response to infection results in exacerbation of underlying changes in airway function, resulting in more persistent wheezing. Investigations in human and animal models have revealed possible clues to pathogenesis, some of which have been reviewed recently. 23, 56, 81 Respiratory viruses, especially RSV, often initially infect the respiratory epithelium of the upper airways but the infection can spread to the lower airways. Following RSV bronchiolitis, viral antigen can be detected in the bronchioles. Rodent and other animal models also show evidence of lower airway infection. 43, 61 Rhinovirus, the classical cold virus, which was thought to infect upper airways exclusively, also is capable of infecting the lower airways. Rhinovirus mRNA has been detected in cell pellets from bronchoalveolar lavage fluid following experimental RV16 infection.

Figure 1

The interplay between respiratory viral infections, asthma, and allergic sensitization is complex. Recent advances suggest that the host response to the virus may be a major factor in dictating the outcome.

Respiratory viral infections trigger an inflammatory response in the airways. Bronchial biopsies following experimental RV16 infection show increased inflammation in the lower airways, including increases in numbers of submucosal lymphocytes and of epithelial eosinophils. In lung tissue samples from children with RSV bronchiolitis, the mucous membranes were demonstrated to be inflamed, with cellular debris and fibrin forming plugs within the bronchioles, resulting in atelectasis and hyperinflation. Destruction of respiratory epithelium, necrosis of lung parenchyma, and hyaline membrane formation also occur. The cellular inflammatory response is dominated by interstitial mononuclear cell infiltrates and neutrophil-rich exudation in the airway lumen. Increased numbers of eosinophils have also been observed in many tissue samples. In patients with RSV bronchiolitis, increases in the numbers of peripheral blood eosinophils have been detected, suggesting that eosinophils are recruited to the airways and are activated. Increased levels of eosinophilic cationic protein have been reported. Indeed, activation of eosinophils by RSV has been demonstrated in vitro, resulting in increases in superoxide production and priming for increased leukotriene C4 release.

Another inflammatory mechanism potentially involved in the development of asthmatic symptoms is the increase in production of IL-11 by epithelial cells following viral infection. In children with viral upper respiratory tract infection and in those with wheezing, IL-11 levels are elevated in nasal secretions. Administration of recombinant IL-11 into the lungs of mice results in increased airway responsiveness to methacholine. The role IL-11 plays in virus-induced lung disease remains to be determined.

THE ROLE OF EOSINOPHILS AND M2-RECEPTOR DYSFUNCTION IN VIRUS-INDUCED WHEEZING

The inflammatory response elicited following viral infection and, in particular, the infiltration of eosinophils, are postulated to play essential roles in the development of wheezing during acute infection. The authors recently reported in a murine model that the eosinophilic component of the inflammatory response to acute RSV infection and the associated development of airway hyperresponsiveness (AHR) to methacholine provocation were dependent on the presence of IL-5. Blockade of the eosinophil adhesion molecule VLA-4 in this model, in the presence of IL-5, prevented both eosinophil migration into the airways and the associated development of AHR. These data extend the clinical observations that development of RSV-induced AHR in this model is associated with the presence of eosinophils and may well be dependent on this eosinophilic response.

Dependence of virus-induced AHR on IL-5 has also been reported in a guinea pig model of PIV infection but eosinophil influx to the airways appeared to be independent of IL-5. Studies in guinea pigs revealed a mechanism by which eosinophils can influence airway tone and reactivity. Cationic proteins released by eosinophils are capable of binding to presynaptic M2 muscarinic receptors on postganglionic parasympathic airway nerves. The resulting blockade interrupts an inhibitory feedback mechanism, resulting in increased release of acetylcholine and in increased airway muscle tone and reactivity. This mechanism has been demonstrated both in models of allergic sensitization and following acute viral infection. 29, 30, 31 Parainfluenza virus neuraminidase can also bind to M2 muscarinic receptors directly and may be responsible for the effects described in the absence of eosinophilic inflammation. In addition, viral infection and interferon (IFN)-γ downregulate M2-receptor gene expression.

NONINFLAMMATORY MECHANISMS IN VIRUS-INDUCED WHEEZING

There are also noninflammatory mechanisms that may contribute to the development of wheezing following viral respiratory tract infection. Viral infection of respiratory epithelium results in reduced nitric oxide production associated with AHR in guinea pigs. Nitric oxide is the putative bronchodilator agonist of the nonadrenergic, noncholinergic inhibitory (NANCi) system. This system can be defective during and following respiratory viral infection, resulting in AHR, demonstrated in RSV infection of cotton rats. A reduced barrier function of the respiratory epithelium may expose sensory C fibers to enhanced stimulation. This results in release of neuropeptides such as substance P and neurokinin A, both agonists of the nonadrenergic, noncholinergic, activating system and induces a brainstem reflex, leading to bronchoconstriction. Neuropeptides can also contribute to airway obstruction by causing increased leukotriene synthesis, release of mast cell mediators, 13, 42 and increased mucous secretion. In addition, infected epithelial cells produce smaller amounts of neutral endopeptidase, an enzyme that degrades neuropeptides. 5, 19, 37, 47 The role of sensory C fibers in virus-induced asthma exacerbations in humans remains controversial. Bradykinin provocation following experimental RV16 infection in mild asthmatics did not result in increased bronchial hyperresponsiveness. Bradykinin is a strong stimulator of sensory C fibers and would be expected to cause increased bronchial hyperresponsiveness if this system plays a major part in virus-induced asthma.

PERSISTENCE OF INFECTION

It is unclear how changes induced by acute respiratory tract virus infection can impact the development of asthma well after the infection has resolved. Some of the pathologic changes may simply persist for long periods after the acute infection. A defect in NANCi function has been demonstrated to last for up to 24 weeks following RSV infection in ferrets. Persistence of infection, resulting in chronic alterations of epithelial cell function and chronic inflammation, has also been suggested. This hypothesis is supported by findings in guinea pigs and calves, in which RSV antigen can be detected in the lung 6 and 12 weeks after resolution of the infection. 65, 79 In guinea pigs, this persistence is associated with persistent AHR. Respiratory syncytial nucleic acid is also detectable in postmortem lung tissue from infants who died long after an RSV epidemic, supporting the possibility of virus persistence.

INTERACTION BETWEEN VIRAL RESPIRATORY TRACT INFECTION AND ALLERGIC SENSITIZATION

To define potential mechanisms of interaction between viral respiratory tract infection and allergic sensitization to inhaled allergens, rodent and bovine models have been developed. 26, 35, 45, 46, 63 The majority of these models showed increased allergic sensitization following respiratory virus infection that resulted in eosinophilic airway inflammation and AHR. 26, 35, 46, 63 Respiratory syncytial virus infection has been shown to prolong methacholine-induced AHR in mice sensitized and challenged to ovalbumin. In these models, animals were first exposed to allergen during the acute infection, followed by subsequent allergen challenges, resulting in increased allergic sensitization, with elevated serum levels of allergen-specific IgE. In these experimental approaches, enhanced allergic sensitization was thought to be caused by increased allergen uptake across inflamed mucous membranes. Indeed, in both a guinea pig and a mouse model, exposure to ovalbumin aerosol caused increased levels of serum ovalbumin if administered during acute virus infection. 26, 63

The authors recently reported on a murine model of RSV infection and subsequent sensitization to aerosolized ovalbumin. In this model, exposure to allergen over 10 days was begun only after complete resolution of the acute (RSV) infection. This resulted in enhanced responses to allergen and, as a consequence, airway inflammation, with the influx of neutrophils and eosinophils. This was associated with altered airway responsiveness to inhaled methacholine. In contrast to many of the models discussed previously, allergen-specific IgE serum levels were not higher in the group that was infected with RSV prior to allergic sensitization. This may indicate that mechanisms other than increased allergen uptake are responsible for the effects of RSV infection on the subsequent consequences of exposure to allergen. As demonstrated following acute RSV infection, sensitization following infection triggers eosinophilic inflammation and associated AHR. Anti–interleukin (IL)-5 treatment during the allergen exposure phase prevented lung eosinophilia and the development of AHR.

To define the role of IL-5 and, specifically, of eosinophils in mediating the effects of RSV infection on subsequent airway sensitization, two approaches were pursued: (1) anti–IL-5 treatment during the infection phase but not during the period of allergen exposure, and (2) infection and sensitization in genetically IL-5–deficient mice and evaluation of the effects of IL-5 reconstitution during the different phases. Anti–IL-5 treatment during RSV infection significantly reduced lung eosinophilia and AHR following subsequent allergen exposure via the airways. Mice genetically deficient in IL-5 did not develop lung eosinophilia or AHR following RSV infection and allergen sensitization. Both eosinophilia and AHR were reconstituted if IL-5 was administered during acute infection. In contrast, administration of IL-5 only during the allergen sensitization phase, but not during infection, did not reconstitute AHR, despite increases in lung eosinophils.

These data demonstrate that the presence of IL-5 and, concomitantly, of eosinophils, during acute infection is critical to the expression of the effects of RSV infection on subsequent allergen exposure. To further define the underlying mechanisms, the authors evaluated the role of IL-4 and IFN-γ. Genetic deficiency of IL-4 prevented the development of RSV-induced effects on subsequent allergen exposure. This deficit could be compensated for by administration of IL-5, indicating that IL-4 may be required for sufficient IL-5 production. The presence of IFN-γ was not necessary for the effects of RSV infection to develop. On the contrary, lung eosinophilia and AHR were highest in the absence of IFN-γ.

T lymphocytes are believed to play a pivotal role in the regulation of immune responses to viral infection and allergens. The authors tested the hypothesis that RSV infection induces a T-cell response that mediates the consequences of infection on subsequent allergic sensitization. Following RSV infection, peribronchial lymph nodes, the regional lymph nodes of the lung, were harvested. T cells were isolated and adoptively transferred into noninfected mice, which were then exposed to aerosolized allergen over 10 days. Adoptive transfer of T cells from RSV-infected mice (but not noninfected mice) resulted in lung neutrophilia and eosinophilia as well as AHR following airway allergen exposure. Transfer of isolated CD8+ T cells, but not CD4+ T cells, resulted in similar effects.

Interestingly, the effect of transfer was dependent on the interval between the onset of infection and the day of harvest of T cells. Transfer of T cells harvested 14 days postinfection—i.e., after resolution of the acute phase of infection—resulted in positive effects on the response to allergen when cellular infiltration and airway responsiveness were monitored. In contrast, adoptive transfer of cells 7 days postinfection—i.e., during the acute phase of infection—had no effect on subsequent allergen exposure.

Using T-cell depletion, the authors further confirmed that CD8+ T cells were, indeed, essential both for the development of eosinophilic inflammation and AHR following RSV infection and for the T-cell–mediated enhancement of the response to subsequent allergen exposure. The depletion of CD8+ T cells was associated with an absence of the increases in IL-5 concentrations in the bronchoalveolar lavage fluid following RSV infection in normal mice. These studies implied that RSV infection triggers CD8 T-cell activation—not just that of cytotoxic, IFN-γ–producing, T cells, a dominant T-cell response following viral infection, but also noncytotoxic, IL-5–producing, CD8+ T cells (Tc2 cells). These cells are activated during viral infection and can orchestrate airway eosinophilia.

It is highly likely that different viral antigens can differentially induce specific immune responses. Some antigens—for example, the RSV G protein—may result in the selective induction of Th2 and Tc2 (CD8+ T cells producing Th2-like cytokines) responses, leading to airway eosinophilia as demonstrated following vaccination against G protein and subsequent RSV challenge. Other components such as RSV F protein may favor development of Tc1 responses. Interleukin-5–producing CD8+T cells may be primarily responsible for the recruitment of eosinophils during acute RSV infection. These cells may persist following the acute infection and possibly expand as the infection resolves and numbers of cytotoxic CD8+ T cells diminish. This could explain the different results observed following transfer of peribronchial lymph node (PBLN) T cells obtained either 7 or 14 days postinfection. It is conceivable that these Tc2-like cells, which persist and have a stable phenotype over time, on reactivation, favor an IL-5–mediated eosinophilic inflammatory response, promoting the development of AHR. CD4+ T cells or the same Tc2 cells may be capable of IL-4 production, which facilitates allergic sensitization when allergens are subsequently encountered.

STRATEGIES FOR PREVENTION OF VIRUS-INDUCED ASTHMA

Definition of these pathophysiologic pathways opens the possibilities for strategizing and developing preventive therapies for children at risk for early asthma development induced by respiratory viruses. Several novel approaches may be entertained: anti–IL-5 antibody treatment during severe viral lower respiratory tract infection, induction and sustaining Th1 and Tc1 immune responses during and following viral infection using local Th1-type cytokine treatment, 6, 75 Th1-type cytokine gene transfection, or treatment with CpG oligonucleotides. Immunomodulatory vaccines for inducing protective responses have recently been reported for an IFN-γ gene-coupled RSV vaccine.

SUMMARY

Respiratory viral infections in early childhood have been linked to the development of persistent wheezing and asthma. Epidemiologic data indicate that, for the majority of children, virus-induced wheezing is a self-limited condition, with no long-term consequences. For a substantial minority, however, virus-induced wheezing is associated with persistent asthma and the potential for enhanced allergic sensitization. For the most part, this subset of patients is genetically predisposed; they are atopic children in whom respiratory viral infections trigger the early development of asthma by mechanisms that have not been fully elucidated. Both inflammatory and noninflammatory mechanisms may be involved. It does not appear that viral infection per se in early life is responsible for the induction of atopic asthma. Data from animal models provide support for the concept that enhanced allergic sensitization caused by increased uptake of allergen during infection may play a critical role, as well as T-cell–mediated immune responses to viral infection, which may favor eosinophilic inflammatory responses and the development of altered airway function to inhaled methacholine. Recent advances in our understanding of the interactions between respiratory viruses and the development of reactive airway disease offer new possibilities for preventive treatment in children at risk for developing persistent wheezing and asthma exacerbation as a result of viral infection.