Published ahead of print on December 10, 2004, doi:10.1164/rccm.200407-970OC
© 2005 American Thoracic Society doi: 10.1164/rccm.200407-970OC
Quantitative and Qualitative Analysis of Rhinovirus Infection in Bronchial TissuesDivision of Allergy and Immunology, Departments of Medicine and Pediatrics, University of Wisconsin–Madison, Madison, Wisconsin Correspondence and requests for reprints should be addressed to Dr. Anne G. Mosser, Ph.D., University of Wisconsin, K4/928 CSC-9988, 600 Highland Avenue, Madison, WI 53792. E-mail: am2{at}medicine.wisc.edu
Although rhinovirus (RV) infections can cause asthma exacerbations and alter lower airway inflammation and physiology, it is unclear how important bronchial infection is to these processes. To study the kinetics, location, and frequency of RV appearance in lower airway tissues during an acute infection, immunohistochemistry and quantitative polymerase chain reaction analysis were used to analyze the presence of virus in cells from nasal lavage, sputum, bronchoalveolar lavage, bronchial brushings, and biopsy specimens from 19 subjects with an experimental RV serotype 16 (RV16) cold. RV was detected by polymerase chain reaction analysis on cells from nasal lavage and induced sputum samples from all subjects after RV16 inoculation, as well as in 5 of 19 bronchoalveolar lavage cell samples and in 5 of 18 bronchial biopsy specimens taken 4 days after virus inoculation. Immunohistochemistry detected RV16 in 39 and 36% of all biopsy and brushing samples taken 4 and 15 days, respectively, after inoculation. Infected cells were primarily distributed in discrete patches on the epithelium. These results confirm that infection of lower airway tissues is a frequent finding during a cold and further demonstrate a patchy distribution of infected cells, a pattern similar to that reported in upper airway tissues.
Key Words: bronchus • common cold • epithelium • virus Viral respiratory infections, particularly with rhinovirus (RV), are the major cause of asthma exacerbations in children and adults (1, 2). RV infections have been believed to be confined primarily to the upper respiratory tract, thus creating a paradox with asthma exacerbations, which are lower airway events. There is, however, growing evidence to indicate that the lower airway can also be infected. In in vitro models, RV replicates in epithelial cells isolated from lower airway tissues (3, 4), and, in fact, viral growth curves and the percentage of infected cells are quite similar in epithelial cells derived from either bronchial or adenoidal tissue (5). Moreover, after experimental infection, virus has been cultured from sputum (6), identified by a sensitive polymerase chain reaction (PCR) technique from cells derived from bronchoalveolar lavage (BAL) fluids (7) and by in situ hybridization in bronchial biopsy specimens (8). Although these findings support the hypothesis that RV infects lower airway tissue, and may be an important step in causing airway obstruction, coughing, and wheezing in patients with asthma, studies to evaluate the quantity and distribution of RV in lower airway cells and secretions during RV infections have not been performed. Furthermore, although RV infection produces a spotty pattern of infected cells in the upper airway, it is not clear that the pattern of infection in lower airway tissues is similar. It is also not clear whether subjects with asthma are more likely than normal subjects to have RV infections that extend into the lower airway. Corne and coworkers (9), in a prospective study of 76 cohabiting couples of which one partner had asthma, found that the partners with asthma and healthy partners had similar rates of RV infections and RV-associated upper respiratory tract illness, but the partners with asthma had increased rates of lower respiratory tract symptoms. This finding could result from an increased tendency for infections to extend into the lower airway in subjects with asthma, or from increased bronchial symptoms resulting from the same rate of infection. To define more precisely the appearance of RV in the lower airway of allergic patients with asthma and normal, nonallergic subjects, quantitative PCR (qPCR) analysis was used to determine the amounts of virus in cells derived from nasal lavage, sputum, BAL, and bronchial biopsy samples taken before and over the 15 days after experimental inoculation with serotype 16 RV (RV16). In addition, multiple mucosal biopsy samples and brushings of cells of lower airways were obtained and then stained with an RV16-specific monoclonal antibody to determine the frequency of virus detection and the distribution of virus-infected cells in these samples. The present article presents data from repetitive sampling from different parts of the airway to more comprehensively describe the progress, distribution, and locations of RV infection. Some of the results of these studies have been previously reported in the form of abstracts (10–13).
Study Population Thirteen subjects with allergic asthma and six normal nonallergic subjects were enrolled in this study (Table 1) from November 2002 until January 2004. Allergy was determined by skin-prick tests to cat, house dustmite, and local pollens and molds. All subjects were seronegative for RV16, and none had signs or symptoms of respiratory infections in the 2 weeks before inoculation. Subjects with asthma had a physician diagnosis of asthma based on at least a 6-month history of symptoms in accordance with the American Thoracic Society guidelines (14, 15), which included reversibility to ß-agonists (12% increase in FEV1 after inhaled albuterol) and/or a concentration of methacholine (mg/ml) causing a 20% fall in FEV1 (PC20) of less than 8.0 mg/ml.
The study was approved by the University of Wisconsin–Madison's human subjects committee, and written informed consent was obtained from all subjects.
Study Design
Nasal Lavage, Sputum Induction, and Processing of Samples Nasal lavage and sputum induction were performed and processed as reported earlier (18). Aliquots of total cells from these samples were stored in Trizol reagent (Invitrogen, Carlsbad, CA) for RNA extraction. Analysis by qPCR was based on 500,000 white blood cells plus accompanying epithelial cells per sample, and nasal lavage samples were rejected if they had fewer cells. Before inoculation, 42% of the nasal lavage samples were rejected. Sputum samples were rejected for qPCR analysis if more than 80% of the cells present were squamous epithelial cells (8.5% of all sputum samples). All samples included in the study had 11% or fewer squamous epithelial cells.
Bronchoscopy
Tissue Preparation and Immunohistochemical Staining The monoclonal antibody R16-7, used to identify RV16-infected cells, has been shown to bind to the VP2 capsid protein of RV16 and RV1A, a closely related RV that belongs to the minor receptor group but not to RV2, 14, or 49 (5). Slides were evaluated by three observers who were blinded to the subject information. Only red-purple staining that was clearly cytoplasmic was considered to be positive for RV16. Discrepancies in tissue scoring (5% of the slides) were resolved by two additional viewers, who were also blinded to visit and subject number, who reviewed the disputed slides to determine the scoring.
Titration of RV
Determination of Viral RNA by qPCR
Statistical Methods
Subject Characteristics Thirteen allergic subjects with asthma and six nonallergic normal subjects were recruited for the study (Table 1). There were no statistical differences between the ages of the two groups of subjects or their baseline FEV1 (% predicted). All subjects had an infection as indicated by recovery of RV16 from the nasal lavage after inoculation. Symptoms peaked between the first and fifth days after inoculation (Table 1), with maximum total daily symptom scores between 3 and 16 (median 8.0, interquartile range 6.25–13.00); there were no significant differences between subjects with asthma and normal subjects in the timing or amount of peak symptoms. All subjects except one answered "yes" to the question "Do you have a cold?" on at least 2 days; the subject who never answered with "yes" did have an increase in total reported cold symptoms from 0 to 6. All of the subjects with asthma except one used a ß agonist at least once for symptomatic relief; four were frequent users. One subject (No. 14) did not complete the last two study visits by personal choice, and another (No. 2) did not undergo the final bronchoscopy because she experienced increased asthma symptoms 4 days after the previous bronchoscopy; she recovered within 2 days by increased use of inhaled albuterol. During the acute cold, only one subject (No. 6) had a mild exacerbation of asthma symptoms (a 13% fall in peak flow rate compared with precold peak flow values and an increase in asthma symptom scores and albuterol use).
Virus in the Upper Airway
Virus in the Lower Airway Viral RNA was also found for all subjects by qPCR in cells from sputum samples taken on Days 3 and 7 after inoculation (Figure 2, upper right). Some subjects had equal or greater amounts of virus in their sputum cells compared with nasal lavage cells; for others, the amounts were up to 100-fold less. There was a trend (p = 0.07) for the ratio of RV RNA in the sputum versus nasal lavage to increase between Days 3 and 7 (medians 0.23 and 1.02). Virus was still found in sputum cells from 14 of 18 subjects (10 of 13 subjects with asthma and 4 of 5 normal subjects) 14 days after infection. There were no significant differences between subjects with asthma and normal subjects. Viral RNA was found less frequently in BAL cells and bronchial biopsy specimens than in nasal lavage and sputum during the acute cold (Figure 2, lower panels). Seven of the 19 subjects had detectable virus either in lavage cells (38–1,363 qPCR units) or biopsy samples (25–3,907 qPCR units); four subjects had virus detectable in both specimens, one subject had virus only in BAL cells, and two subjects had virus only in a biopsy sample. No viral RNA was detected in biopsy samples taken 15 days after inoculation, but viral RNA was detected from 1 of 17 BAL cell samples (49 qPCR units) from that timepoint. There were no significant differences between subjects with asthma and normal subjects.
Immunohistochemical Staining for Virus in Bronchial Samples
Samples taken 4 days after RV16 inoculation showed that 10 of the 13 subjects with asthma and three of six control subjects had evidence of virus in the lower airway by immunohistochemistry (Tables 2 and 3). Figure 3 shows typical staining patterns for uninfected (Figure 3A) and infected (Figure 3B) HeLa cells, which were used as controls. It also shows a typical negative area of epithelium from biopsy (Figure 3C) and brushing cells (Figure 3E), and a typical positive area from biopsy (Figure 3D) and brushing (Figure 3F). In biopsy samples, positively staining cells were found mainly in the epithelium, but rarely an unidentified subepithelial cell was positive for virus. In general, the staining for RV was patchy; positive cells often occurred in groups, with uninfected cells in adjacent epithelium. In some samples, only one or two cells were positive for virus. Although sample numbers were small, there were no significant differences between biopsy samples taken from the subjects with asthma and normal subjects. Also, immunohistochemistry and qPCR gave similar rates of positive samples (39% of all samples by immunohistochemistry and 28% of samples by qPCR), although they occasionally identified different subjects as positive.
In the samples collected 15 days after inoculation, 6 of the 13 subjects with asthma still had positive samples by immunohistochemistry (Table 2); this result was not statistically different from detection rates at the acute cold (p = not significant by McNemar's test comparing subjects with positive and negative samples). All of the control subjects had at least one positive sample (Table 3), but with samples for only five control subjects available, there were too few for a significant comparison to the subjects with asthma. No positive biopsy samples were found by qPCR from subjects with asthma or normal subjects at this timepoint. It is not clear whether this finding reflects a difference in the persistence of viral RNA compared with protein, or whether this difference was a chance occurrence. Positively staining cells were still located primarily in the epithelium.
A principal finding of this study is that, after upper airway inoculation with RV16, similar amounts of this virus were detected in cells from nasal lavage and sputum in most infected individuals. This result was especially true during the latter stages of infection, as indicated by the elevated sputum/nasal lavage viral RNA ratio at Day 7 compared with Day 3 and a surprising persistence of viral RNA in sputum cells (12 of 18 individuals) 2 weeks after inoculation. Moreover, 17 of 19 subjects had virus detected either by immunohistochemistry or qPCR in lower airway epithelial cells obtained by brushing or biopsy. Thus, significant infection of the large lower airways appears to be a common, if not universal, occurrence following RV inoculation of the nose. In addition, these findings suggest that RV initially infects the upper airway and then, in many individuals, spreads to the lower airway. Because we frequently found virus in lower airway samples, we must consider the possibility that our inoculation technique may have led to direct exposure of the lower airway to virus. We deliberately chose an atomizer that generates coarse droplets to maximize deposition in the nasopharynx, but even so, it is likely that very small amounts of infectious virus were delivered to the bronchial tree. During natural colds, transmission of colds has been documented via aerosol and direct contact with infected secretions. During natural aerosol transmission, it is likely that the upper airway is the primary site of inoculation, although data are lacking. It seems certain, however, that once infection in the nose is established through either natural or experimental inoculation, there is repeated generation of virus-containing droplets through coughing and sneezing that are inhaled and deposited in the lower airway. This process is likely to be an important source of lower airway inoculation. The concept that RV infections spread from the upper to the lower airway is supported by clinical observations that asthma exacerbations tend to follow upper respiratory infection symptoms by 2 to 3 days (2). Furthermore, in our inoculation study, peak viral shedding in the sputum in many cases followed peak shedding in the nasal secretions. The large amount of virus in sputum cells, which in some subjects exceeded quantities in the upper airway, also argues against the possibility that virus in the lower airway is only there as a result of contamination during sampling procedures, as is the visualization of infected cells in the biopsy tissues. Analysis of bronchial biopsy samples gave evidence that infection in this part of the airway was irregularly distributed. When several biopsy samples from a single timepoint and a single subject were examined, frequently only a subset of lower airway tissues were positive for virus, and only a small portion of the epithelium was infected as detected by histologic methods. This patchy pattern of infection has been noted previously and consistently in studies of the upper airway (24, 25). Our observations indicate that a similar distribution of infection exists in the upper and lower airway. A recent study by Pitkaranta and coworkers (26) used in situ hybridization to study RV infection of middle turbinate tissue during naturally occurring colds. They found infection of epithelial cells and mucosal inflammatory cells in 69% of the biopsy samples taken from subjects 7 days after symptoms appeared. Some biopsy samples showed a large number of infected cells, but, as we also noted, the infected cells were still patchily distributed, with adjacent epithelium uninfected. We used qPCR and immunohistochemical staining to detect virus in bronchial biopsy specimens taken 4 days after inoculation, and the rates of positive biopsy samples were similar (28 and 39%, respectively) for both techniques. In biopsy samples taken 15 days after infection, we continued to find similar evidence of viral protein by immunohistochemistry; however, all 16 biopsy samples studied by qPCR from this timepoint were negative. This finding raises the possibility that viral capsid protein persists in infected tissues longer than viral RNA. However, viral RNA was still detectable by qPCR in sputum cells and in one BAL sample, so the lack of viral RNA in biopsy samples may represent a sampling problem. Although qPCR results correlated very closely with infectious virus titers when nasal lavage samples were tested in both assays (see online supplement), it should be noted that this assay measures the presence of a portion of the viral RNA, and not infectious virus. Small amounts (< 10 qPCR units) of viral RNA were occasionally found in upper or lower airway samples taken before inoculation (Figure 2). This finding presumably represents background noise in the assay, although it might possibly represent RV RNA from an earlier cold caused by a related virus. Because this finding suggests background noise for the qPCR assay, we did not consider samples with fewer than 10 qPCR units as positive in our analysis of the samples. We also found one biopsy sample with cells that stained positively for virus protein in a preinoculation sample. Because the antibody is known to stain tissue infected with RV1A as well as RV16, the positive sample could be the result of a prior infection with RV of an unknown serotype, or could represent a low level of artifactual staining. Our results from this study extend those reported by Papadopoulos and colleagues (8), who demonstrated viral RNA by in situ hybridization in the epithelium of infected subjects in bronchial biopsy samples taken 3 days after an experimental RV inoculation, as well as significant hybridization in the subepithelial layer. They found positive hybridization in samples from 5 of 10 subjects taken during the infection. Two of 10 subjects had positive samples 6 to 8 weeks after inoculation, but it is not clear if these were from the experimental virus or subsequent infection with another RV. The probes used could have hybridized to RV of multiple serotypes, and one of their subjects had another clinical cold at the time of bronchial sampling.
RV was easily detected by qPCR in nasal secretions and sputum but was less often detectable in biopsy samples of large airways or BAL cells, which are a mixture of cells from smaller airways and alveoli. Detection of viral RNA in biopsy samples may have been limited by the patchy distribution of infected cells; this was not a problem for the other samples, because infected cells that had appeared in the airway were concentrated for analysis. These findings suggest that RV infections commonly involve the nasopharynx and proximal lower airways, but are less likely to infect distal airways or the alveoli, which is in accordance with observations that RV infections are an uncommon cause of pneumonia except in immunosuppressed individuals (27). The spread of RV to the distal airways is also likely to be inhibited by temperature gradients, because most serotypes of RV grow best at relatively cool ( In this study, we did not find evidence of enhanced or attenuated RV detection in the airways of subjects with asthma compared with normal subjects. However, there are important limitations in interpreting these data. First, in our study, only one subject had a mild clinical exacerbation of asthma. This finding is typical of the results we have had with experimental inoculation and limits our ability to describe virologic or immunologic changes that are associated with airflow obstruction. Second, the number of normal subjects in this study was limited, thus reducing the power of between-group comparisons. Finally, only subjects with mild and well-controlled asthma were enrolled in this protocol. Subjects with a known history of viral infection–induced asthma exacerbations were excluded for safety reasons. The group we studied, therefore, would be considered at low risk for an exacerbation with a cold. Some of these shortcomings could be overcome by studying subjects with naturally acquired colds and exacerbations of asthma.
Despite these limitations, our study has demonstrated that RV infection of the lower airways is common after experimental inoculation. Moreover, the amount and pattern of infection closely resemble those that have been determined in studies of upper airway tissue. There is also evidence of some differences as well: the appearance of lower airway infection may be delayed after upper airway inoculation, and not all individuals with vigorous RV replication in the upper airway go on to develop high-level replication in the lower airway as well. These findings suggest that there may be important differences in the individual susceptibility to replication of RV in lower airway tissues. Sources of this variable susceptibility may include immune factors, such as the production of IFN-
The authors thank Andrea Tweedie, RN, BSN, and Mary Jo Jackson, RN, BSN, for subject recruitment and screening, the General Clinical Research Center staff, and Michael Evans for help with statistics.
Supported by grants AI34891 and M01 RR03186 from the National Institutes of Health. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Conflict of Interest Statement: A.G.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; R.V. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; L.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; W.-M.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; C.R.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; E.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; D.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; C.A.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; R.D.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; K.C.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; N.N.J. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; W.W.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J.E.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form July 27, 2004; accepted in final form December 3, 2004
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
35°C) temperatures (
(