INTRODUCTION

TOP ABSTRACT INTRODUCTION HOST RESPONSE TO VIRAL... LATENT ADENOVIRAL INFECTION AND... LATENT ADENOVIRAL INFECTION AND... SUMMARY REFERENCES

Keywords: adenovirus; cigarette smoke-induced lung inflammation; adenoviral E1A gene

It is well recognized that an inflammatory process is responsible for the lesions that cause both chronic obstructive pulmonary disease (COPD) and asthma. An important difference in the nature of this response is that the majority of patients with asthma are effectively managed by a therapeutic regimen that includes inhaled steroids (1) whereas the progressive decline in lung function that characterizes COPD is unaffected by steroid therapy (2). The steroid resistance that occasionally develops in asthma (3) and is a constant feature of COPD (2, 4) can result from adenoviral infection (3, 5). This presentation begins with a brief review of the nature of the respiratory epithelial defense against viral respiratory tract infections as a background for a discussion of the role of latent viral infection in the pathogenesis of COPD and the development of steroid resistance in asthma.

Acute viral respiratory tract infections increase the airway response to nonspecific stimuli in both normal subjects and patients with asthma (6). They are well known to precipitate attacks of asthma in children (7) and adults (8) and have been implicated in the development of steroid-resistant airway obstruction (5). The respiratory viruses primarily target airway epithelial cells and gain entry by attachment to cell membrane proteins such as intercellular adhesion molecule 1 (ICAM-1) in the case of the rhinoviral infection (9). After entering the cell the virus releases its nucleic acid and uses the protein-generating machinery of the host to replicate itself. This occurs through a complex series of steps that are different for each virus and usually begin with the production of a viral protein that can activate the host transcriptional machinery (10). In the case of adenoviral infection the trans-activating protein is referred to as E1A, where the E stands for the fact that it is one of a group of genes transcribed early in the viral life cycle (11). The E1A protein has the ability to initiate a cycle of viral replication that results in either lysis of the host cell with release of large numbers of viral particles to infect other cells, or the shedding of viral particles from the surface of a living host cell. Low levels of viral replication allow the virus to establish a persistent infection in the host that can be long lasting after the initial viral illness has cleared. Viral persistence after an acute replicating infection may also occur if viral DNA forms a circular chromosome or plasmid within the host cell or actually integrates into the host cell genome. During a latent infection, viral proteins are produced without replication of a complete virus but in some cases, notably with the herpesvirus, certain stimuli initiate replication of complete viral particles in latently infected nerve cells (8). These particles then migrate down the nerve to produce the familiar herpes eruption in the skin supplied by that nerve (12). The expression of viral proteins by the host cell without replication of a complete virus is the defining feature of a latent infection and there is evidence that this type of infection can influence the inflammatory response to stimuli that are relevant to both COPD and asthma. Although the work presented here focuses on latent adenoviral infections, the principles involved apply to other intracellular pathogens capable of establishing this form of infection.

	HOST RESPONSE TO VIRAL INFECTION

TOP ABSTRACT INTRODUCTION HOST RESPONSE TO VIRAL... LATENT ADENOVIRAL INFECTION AND... LATENT ADENOVIRAL INFECTION AND... SUMMARY REFERENCES

The response of the host to an infecting virus has both an innate component, which is fully functional before the virus enters the host airway epithelium (13), and an adaptive component that develops in response to the presence of the virus and has memory, specificity, and diversity (14, 15). The components of the innate immune response include the barrier provided by the epithelium, natural killer cells that destroy virus-infected cells in a nonspecific fashion, a special group of intraepithelial lymphocytes, and the polymorphonuclear neutrophils (PMNs), monocytes (MONOs), and macrophages (MACs) that are recruited to the infected site. The epithelial cells and the PMNs produce peptides with a natural antibiotic function, of which the best known are the defensins. The subset of intraepithelial T lymphocytes that are part of the innate immune system differ from standard T cells in that they encode receptors from their germ line rather than by gene rearrangement. This severely restricts their repertoire of receptor specificity compared with the T cells that are involved in the adaptive immune response, which use gene rearrangement to express a much broader range of receptors. The unusual nature of their receptors and the limited species of antigens that they recognize provide the intraepithelial T cells with a low level of specificity for a restricted number of antigens. The innate response is triggered before the adaptive immune response and generates cytokines that initiate both a local response at the site of injury and a systemic response. In the case of viral infection one of the molecules capable of initiating the innate response is double-stranded viral RNA. This molecule is produced during viral replication and activates a host cell kinase that initiates the production of cytokines. The cytokines produced by the infected epithelial cells, the intraepithelial lymphocytes, and natural killer cells as well as by the PMNs and macrophages attracted to the infected site initiate the sequence of rolling followed by firm attachment and subsequent transmigration of circulating leukocytes into the infected tissue. This series of events results in a rapid accumulation of PMNs followed by a slower and more prolonged accumulation of MONOs, MACs, and lymphocytes at the infected site. In addition to destroying virus-infected cells the natural killer cell provides a source of interferon  (IFN-) to activate macrophages to destroy any microbes they may have phagocytosed. The mediators generated at the site of infection and in the blood stimulate a systemic response that increases cardiac output, initiates acute-phase protein production by the liver, stimulates the bone marrow, elevates the circulating leukocyte count, and increases body temperature. Collectively these events are part of the acute-phase response to the initial site of infection and tissue injury.

In addition to providing the first line of defense, the innate immune system also helps to initiate the proliferation and differentiation of the T and B cells that participate in the adaptive immune response. This initial warning to the immune system to mount a specific response to the invading microbe is followed by both the production of antibody by the B cell response and activation of the cell-mediated host response against receptors that recognize class II and class I MHC molecules expressed by other host cells. When macrophages phagocytose microbes they process them in their phagolysosomes and express microbial antigens on the cell surface with MHC class II proteins. This complex is recognized by the T cell receptor on the CD4⁺ lymphocytes and the interaction between the CD4⁺ T lymphocyte and the microbial antigen-MHC class II protein complex expressed on the macrophage surface results in activation of the macrophage in an attempt to kill the microbes they have phagocytosed (14). The CD8⁺ cytolytic lymphocytes that proliferate in response to a viral infection are able to recognize and kill any nucleated cell that expresses viral protein in association with class I MHC molecules on their surface. This is an important and specific process for the removal of virally infected cells and involves three different mechanisms. These include the perforation of the target cell membrane by the molecule perforin, which results in osmotic lysis of the cell, the entry of serine proteases (granzymes) into the target cell through the perforin channel followed by granzyme-triggered induction of apoptotic cell death, and the expression by the CD8⁺ cell of Fas ligand (FasL), which engages Fas on the target cell to initiate apoptosis (14).

Studies of transgenic mice have shown that the CD8⁺ cells, directed against viral proteins in host airway epithelial cells, influence the host response to the virus in other ways. In these transgenic mice the type 2 alveolar epithelial cells were made to express an influenza viral protein by placing its gene under the control of the promoter for the host gene encoding a lung surfactant protein (16). T cell clones were then sensitized to those viral proteins in vitro and adoptively transferred into these transgenic mice. As expected, the transferred CD8⁺ cells recognize and kill off some of the epithelial cells by cytolytic mechanisms, but in addition they also initiate a separate MHC-restricted inflammatory response that attracts large numbers of host macrophages in the absence of perforin, granzyme, or Fas expression. Interestingly, they also showed that T cell-derived tumor necrosis factor  (TNF-) stimulated the target epithelial cells and caused them to secrete monocyte chemoattractant protein 1 (MCP-1) and macrophage inflammatory protein 2 (MIP-2). These experiments demonstrate CD8⁺ cell stimulation of target alveolar epithelial cells expressing a viral protein can initiate an inflammatory response that contains large numbers of macrophages by a mechanism that is quite separate from those that the CD8⁺ cell uses to kill the target cell. Although these experiments are highly artificial, in that all the type 2 epithelial cells expressed the viral protein in these transgenic animals, they suggest that disease can result from viral genes that persist in a much smaller number of host epithelial cells after an acute replicating viral infection has cleared.

DNA viruses such as the adenovirus target airway epithelial cells (17) and their viral DNA can persist in the genome of the host epithelial cells in a latent form (18). The adenoviral E1A gene has been demonstrated in the airway epithelial cells lining the human bronchial lumen, bronchial gland ducts, glands, bronchiolar lumen, and alveolar surface long after viral replication has ceased and the infection has cleared (19). When guinea pigs with an established latent adenoviral infection were exposed to cigarette smoke the inflammatory response was greater than that produced by either the latent infection or cigarette smoke alone (20). Similar studies of allergen- induced lung inflammation (5) showed no effect of latent viral infection on the allergen-induced inflammatory response but the reduction in allergen-induced lung tissue eosinophil infiltration that normally results from the administration of steroids was prevented by latent infection. Collectively these results suggest that the presence of a latent adenoviral infection is capable of both amplifying the cigarette smoke-induced lung inflammatory response and inducing steroid resistance in allergen-induced lung inflammation.

	LATENT ADENOVIRAL INFECTION AND COPD

TOP ABSTRACT INTRODUCTION HOST RESPONSE TO VIRAL... LATENT ADENOVIRAL INFECTION AND... LATENT ADENOVIRAL INFECTION AND... SUMMARY REFERENCES

The adenovirus is a double-stranded DNA virus originally isolated from adenoid tissue (21). There are more than 40 known serotypes that are responsible for a wide range of human and animal diseases affecting the eye, the upper and lower airways, and the gastrointestinal tract. It is responsible for 5% of acute respiratory disease in children under the age of 5 years and for several well-documented epidemics of respiratory tract disease in military recruits (21). The adenovirus adheres to the cell surface through both its fiber protein and penton base and is internalized by receptor-mediated endocytosis (11). After cell entry the viral genome is translocated to the nucleus and transcribed in early and late groups of genes. The first early gene to be expressed is the E1A gene, which activates the host cell transcriptional apparatus and is therefore said to be trans-activating in nature (21). The E1A protein initiates cell division by competing key transcription factors away from the retinoblastoma gene protein product to allow the cell to enter its replicative cycle (22). The E1A protein also interacts with the DNA-binding sites of several other transcription factors that contribute to the expression of a large number of cellular genes (23). This generalized activation of the cell allows the virus to gradually take over host cell protein generation to manufacture viral particles, but it also initiates the production of other molecules that contribute to host defense.

Studies of human tonsils (24), peripheral blood lymphocytes (25), and lung (26) have shown that the adenoviral DNA persists after an acute infection and that the viral E1A protein is expressed in lung epithelial cells long after the virus has stopped replicating and all clinical signs of infection have cleared (18, 19). Matsuse and coworkers found excess amounts of adenoviral E1A DNA in patients with COPD (26) and Elliott and coworkers (19) reported that the E1A protein is expressed in the epithelial cells lining the bronchi, the bronchial glands, gland ducts, the cells lining the bronchioles, and the type 2 cells lining the alveolar surface. Vitalis and coworkers (18) developed a model of human adenovirus 5 infection in the guinea pig and showed that the E1A DNA persisted and E1A protein could be demonstrated in lung epithelial cells. Furthermore, this latent infection was associated with chronic lung inflammation 7 wk after the initial inoculation of virus and 6 wk after the virus had stopped replicating in guinea pigs lungs (18). The lung inflammatory response associated with this form of latent adenoviral infection includes an increase in CD8⁺ cells in the airway wall and an increase in B cells, CD4⁺ and CD8⁺ lymphocytes, as well as monocytes and macrophages in the lung parenchyma (5). When these animals were exposed to a single dose of cigarette smoke the inflammatory response was markedly enhanced by the presence of latent adenoviral infection (20) and chronic exposure to cigarette smoke over 13 wk produced both excess inflammation and more emphysematous lung destruction in the animals with latent adenoviral infection (27).

It is now well established that cigarette smoke exposure is the major risk factor for COPD but only a minority of smokers actually develops this condition. The hypothesis that genetic differences between cigarette smokers account for this increase in susceptibility to develop COPD is an obvious possibility but environmental factors may also be important. A report from Ratemales and coworkers (28) showed a marked amplification of the cigarette smoke-induced inflammatory process in patients with advanced emphysema compared with control subjects with similar smoking histories and preserved lung function. This excess inflammation was associated with an increase in the number of alveolar epithelial cells expressing the adenoviral E1A protein and is consistent with the hypothesis that latent viral infection can amplify cigarette smoke-induced lung inflammation. Keicho and coworkers (29) have also shown that a human lung epithelial cell line (A549 cells) that expresses the adenoviral E1A protein after transfection with the gene produces greater amounts of interleukin 8 (IL-8) and ICAM-1 when challenged. These studies of humans and a human epithelial cell line have been supported by studies of an animal model (18) showing that animals with a latent infection develop excess inflammation after a single exposure to cigarette smoke (20) and increased inflammation with more extensive emphysema when chronically exposed to cigarette smoke (27). These latter studies also showed that the guinea pigs developed an inflammatory response that was similar to that observed in humans (28). Meshi and colleagues were able to show that the CD-4 lymphocyte response was related to the smoke exposure and the CD8⁺ response to the latent virus infection (27).

	LATENT ADENOVIRAL INFECTION AND ALLERGEN-INDUCED LUNG INFLAMMATION

TOP ABSTRACT INTRODUCTION HOST RESPONSE TO VIRAL... LATENT ADENOVIRAL INFECTION AND... LATENT ADENOVIRAL INFECTION AND... SUMMARY REFERENCES

Respiratory tract viral infections, particularly rhinoviral infections, are well known to precipitate attacks of asthma (32), and steroid-resistant asthma has been associated with viral infections. In a follow-up study of a large group of children with seasonal bronchiolitis, Macek and coworkers (3) reported that a proportion of the children continued to wheeze in the recovery period. In the majority of these children this wheezing cleared spontaneously and most of those that required treatment responded promptly to a regimen that included inhaled steroids. Further investigation of the small subgroup that continued to wheeze in spite of treatment revealed that adenoviral capsid protein was present in either the bronchoalveolar lavage (BAL) fluid or bronchial biopsies from these cases. This finding implicated the adenovirus in the development of steroid-resistant airway obstruction and stimulated us to use our animal model of latent adenoviral infection to determine (1) whether latent adenoviral infection could amplify allergen-induced lung inflammation in the same way that it had amplified cigarette smoke-induced lung inflammation and (2) whether the presence of latent adenoviral infection influenced the response of allergen-induced lung inflammation to steroid therapy.

Experiments performed by Yamada and coworkers (5) confirmed that the presence of latent adenoviral infection caused an increase in CD8⁺ cells in the airways and an increase in B cells, macrophages, and CD4⁺ and CD8⁺ cells in the lung parenchyma. Sensitization and challenge of the lung with aerosolized albumen, on the other hand, produced a marked increase in eosinophils, B cells, macrophages, and CD4⁺ cells in both the airways and parenchyma without effect on the CD8⁺ cells (Figure 1). Animals with a latent adenoviral infection that were also sensitized with ovalbumin and challenged showed a mixture of both the adenoviral and antigen-induced effects. But a two-way analysis of variance failed to demonstrate any interaction between the inflammatory reaction produced by latent adenoviral infection and that produced by antigen sensitization followed by aerosolized antigen challenge. The major difference was found when the animals with allergen-induced lung inflammation were treated with steroids. These results showed that the sham-infected animals had a clear reduction in eosinophils in the airways and this did not occur in the animals with latent viral infection (Figure 2a). Other differences that followed steroid treatment included a reduction in CD8⁺ cells in the airways (Figure 2a) and a reduction in CD8⁺, CD4⁺, and B cells in the parenchyma of the animals with latent adenoviral infection but not the sham-infected controls (Figure 2b). The possibility that a shift in cytokine production associated with those changes in the lymphocyte population is responsible for the persistence of eosinophils after steroid treatment in the animals with latent adenoviral infection remains to be investigated.

View larger version (123K): [in this window] [in a new window]   Figure 1.   Eosinophil and CD8+ cell infiltration in the airways, demonstrated with Hansel's stain and immunohistochemistry, respectively. (A) Airway eosinophlia (arrows) induced by OA sensitization and challenge of a sham-infected guinea pig. (B) Airway of a sham-infected guinea pig sensitized and challenged with OA and treated with steroids, by which the number of eosinophils is reduced. (C) Airway eosinophilia (arrows) of an adenovirus-infected, OA-sensitized and challenged guinea pig. (D) Airway of an adenovirus-infected, OA-sensitized and challenged guinea pig, showing that steroid treatment failed to reduce airway eosinophilia (arrows). (E) CD8+ cell infiltration (arrowheads) in the airway of an adenovirus-infected guinea pig sensitized and challenged with OA. (F) Airway of an adenovirus-infected, OA-sensitized and challenged guinea pig, showing that treatment reduced CD8+ cells. Magnification bars: 50 µm. Reproduced by permission from Reference 5.

View larger version (29K): [in this window] [in a new window]   Figure 2.   Effects of steroid treatment on inflammatory cells in (a) the airways and (b) the parenchyma of the sham-infected groups (open columns) and adenovirus-infected groups (solid columns). p Values shown here from unpaired t tests were corrected for multiple comparisons by using the Bonferroni method. Note that steroid treatment decreased the eosinophilic response in the airways of the sham-infected but not the adenovirus-infected group. See text for further explanation. OA = Ovalbumin challenge; S = steroid treatment. Reproduced with permission from Reference 5.

	SUMMARY

TOP ABSTRACT INTRODUCTION HOST RESPONSE TO VIRAL... LATENT ADENOVIRAL INFECTION AND... LATENT ADENOVIRAL INFECTION AND... SUMMARY REFERENCES

The adenovirus targets lung epithelial cells in acute infections (17) and portions of its genome can remain latent in these cells after the infection has cleared (18, 19). These latently infected cells continue to produce viral proteins without replicating a complete virus and the host response to them results in a specific increase in CD8⁺ T cells (5, 27). The presence of this latent viral protein amplifies cigarette smoke-induced inflammation (27) and is associated with a greater amount of emphysema (27, 28) and causes allergen-induced eosinophilic inflammation to become steroid resistant (5). Transfection of a lung epithelial cell line with E1A has also shown that expression of this protein can amplify the production of relevant inflammatory mediators by nonimmune mechanisms (29). Transgenic mouse experiments suggest that when the lung epithelium is induced to express a latent influenza viral protein and targeted by adoptive transfer of CD8⁺ cells sensitized to this protein the epithelium is induced to produce cytokines that drive the inflammatory process (16). These data suggest that a better understanding of both immune and nonimmune mechanisms associated with host microbe interactions during a latent infection with intracellular pathogens could provide insight into both the increased risk of cigarette smoking in some individuals and the development of steroid resistance in others.