Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A guinea pig model of human respiratory syncytial virus lung infection Hegele, Richard G. 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_spring_phd_hegele_richard.pdf [ 23.02MB ]
JSON: 831-1.0086367.json
JSON-LD: 831-1.0086367-ld.json
RDF/XML (Pretty): 831-1.0086367-rdf.xml
RDF/JSON: 831-1.0086367-rdf.json
Turtle: 831-1.0086367-turtle.txt
N-Triples: 831-1.0086367-rdf-ntriples.txt
Original Record: 831-1.0086367-source.json
Full Text

Full Text

THE UNIVERSITY OF BRI ISH COLUMBIAA GUINEA PIG MODEL OF HUMAN RESPIRATORY SYNCYTIALVIRUS LUNG INFECTIONbyRICHARD GEORGE HEGELEMD, The University of Toronto, 1984A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESDepartment of PathologyWe accept this thesis as conformingto the required standardSeptember, 1992© Richard George Hegele, 1992In presenting this thesis in partial fulfilment of the requirements for an advanced degreeat The University of British Columbia, I agree that the Library shall make it freelyavailable for reference and study. I further agree that permission for extensive copyingof this thesis for scholarly purposes may be granted by the Head of my Department orby his or her representatives. It is understood that copying or publication of this thesisfor financial gain shall not be allowed without my written permission.Department of PathologyThe University of British Columbia2075 Wesbrook PlaceVancouver, CanadaV6T 1W5Date: 23 December 1992ABSTRACTChildren who have an episode of acute respiratory syncytial virus (RSV) bronchiolitis areat increased risk for developing asthma, a disorder characterized by reversible airwayobstruction, airway hyperresponsiveness and airway inflammation. Because latent or persistentpulmonary viral infections have been implicated in the pathogenesis of the airway inflammationunderlying asthma, a guinea pig model of human RSV lung infection was developed to test thehypothesis that RSV persists within the lung following resolution of acute bronchiolitis. Onemonth old, "juvenile" guinea pigs, intranasally inoculated with human RSV, were compared onday 6 post-inoculation to two month old, "adolescent" RSV-inoculated animals and uninfectedcontrols by clinical examination, gross lung examination and lung histopathologicalexamination using a semi-quantitative scoring system for bronchiolar inflammation. The naturalhistory of intrapulmonary RSV was studied in juvenile animals on days 6, 14, 60 and 125using viral culture to test for replicating virus, transmission electron microscopy (TEM) to testfor assembled virus, immunohistochemistry to test for RSV antigens and the reversetranscriptase-polymerase chain reaction (RT-PCR) to test for RSV genomic RNA. Theexperiments showed that juvenile guinea pigs inoculated with human RSV developedstatistically significant bronchiolar inflammation by day 6 that resolved by day 14, similar tohuman acute bronchiolitis. Intrapulmonary RSV was documented by culture, TEM,immunohistochemistry and RT-PCR during acute infection on day 6; in the longer termstudies, replicating RSV was cultured up to 14 days post-inoculation, RSV antigens wereidentified within alveolar macrophages up to day 60 and RSV genomic RNA was identified onday 125. In conclusion, human RSV produced a self-limited acute bronchiolitis in juvenileguinea pigs, with evidence of subsequent persistent infection of alveolar macrophages.Persistent RSV infection of alveolar macrophages may represent a mechanism by which hostpulmonary defense mechanisms are compromised against other, non-specific environmentalagents implicated in the pathogenesis of asthma.TABLE OF CONTENTSAbstractTable of ContentsList of Tables^ viiList of Figures viiiAcknowledgments^ ixChapter 1 General Introduction^1.1^Definition of Asthma1.2^Characteristics of Asthma121.2.1^Reversible Airflow Obstruction 21.2.2^Airway Hyperresponsiveness 31.2.3^Airway Inflammation 41.3 Atopic Allergy and Asthma 51.4 Viral Respiratory Tract Infections as Signals forChronic Airway Inflammation in Asthma 71.4.1^Epidemiological Data 71.4.2^Clinical Data 81.4.3^Physiological Data 91.4.4^Histopathological Data 91.4.5^Biochemical Data 101.5 Summary 11Chapter 2 Respiratory Syncytial Virus2.1 Historical Background 132.2 Classification of RSV 132.3 Structure of RSV 142.3.1 Nucleocapsid (N) Protein 142.4 Life Cycle of RSV 142.52.6RSV Infection in vitro^2.5.1^Lytic Infection^2.5.2^Nonlytic InfectionRSV Infection in vivo161617182.6.1^Predilection of RSV for the Lower RespiratoryTract in Children 182.7 Immunity to RSV Infection 202.7.1^Humoral Immunity 202.7.2^Cell-mediated Immunity 202.8 Animal Models of RSV Lung Infection 212.9 The Guinea Pig as a Possible Animal Model of HumanAcute RSV Bronchiolitis 222.10 Summary 23Chapter 3 Working Hypothesis, Specific Aims and Strategy3.1 Working Hypothesis 253.2 Specific Aims 253.3 Strategy 253.4 Summary 28Chapter 4 Materials and Methods4.1 Preliminary Study 304.2 Species 324.3 Virus 324.4 Inoculation Procedure 334.5 Viral Plaque Assay 334.6 Clinical Evaluation 364.7 Gross Lung Examination and Tissue Processing 364.8 Light Microscopy 37iv4.9^Viral Culture^ 384.10 Transmission Electron Microscopy^ 474.11 Immunohistochemistry^ 474.12 Reverse Transcriptase Polymerase Chain Reaction (RT-PCR)^514.12.1 Preliminary Studies of RT-PCR^ 514.12.2 Selection of Oligonucleotides 514.12.3 Purification of Oligonucleotides^ 524.12.4 Total Cellular RNA Extraction 534.12.5 Reverse Transcription^ 544.12.6 PCR Amplification 574.12.7 Agarose Gel Electrophoresis^ 574.12 .8 Southern Transfer^ 574.12.9 Preparation of Oligonucleotide Probe^584.12.10 Filter Hybridization and Autoradiography 594.13 Photography^ 594.14 Statistical Analyses 604.14.1 Clinical and Gross Lung Examination^604.14.2 RSV Bronchiolitis Histological Scoring System^604.14.3 Light Microscopic Evaluation^ 614.15 General Technical Note^ 62Chapter 5^Results5.1^Preliminary Experiments^ 645.2^Specific Aim 1: Clinical-Pathological Evaluation^695.2.1^Clinical and Gross Lung Examination 695.2.2^Lung Histopathological Evaluation^69v5.3 Specific Aim 2: Natural History of Intrapulmonary RSV 765.3.1^Viral Culture 765.3.2^Transmission Electron Microscopy 805.3.3^Immunohistochemistry 805.3.4^RT-PCR 835.4 Summary 95Chapter 6 Discussion6.1 Specific Aim 1: Production of Acute RSV Bronchiolitis inthe Guinea Pig 966.1.1^Clinical-Pathological Evaluation 966.2.2.^Histological Scoring System 996.2 Specific Aim 2: Natural History of Intrapulmonary RSV 1026.2.1^Viral Culture 1026.2.2^Transmission Electron Microscopy 1036.2.3^Immunohistochemistry 1046. 2.4^RT-PCR 1066.3 Concluding Remarks 109Bibliography 111Appendix A Experimental Data for All Guinea Pigs Studied 130Appendix B Statistical Analyses of Airway Histological Scores 143viLIST OF TABLESTable 1 Human RSV genes and gene products^ 15Table 2 Techniques for the guinea pig model of RSV lung infection^27Table 3^Final protocol for inoculation of human RSV into guinea pigs 31Table 4 Guinea pig body weights, lung wet weights and lung to body weight ratios^70Table 5 Documentation of intrapulmonary RSV in virus-inoculated guinea pigs^77viiLIST OF FIGURESFigure 1 Preliminary study of guinea pig inoculation with human RSV(Experiment #1742)^ 30Figure 2 Photomicrographs of Hep-2 cell monolayers (inverted microscope)^34Figure 3 RSV bronchiolitis histological scoring system: standard photomicrographs^39Figure 4 Electron micrograph of RSV-infected HEp-2 cells^ 48Figure 5 Agarose-formaldehyde gel electrophoresis of total cellular RNA^55Figure 6 Low power photomicrograph of guinea pig lung, day 6 followingintranasal RSV inoculation (preliminary experiment)^ 65Figure 7 Intraobserver variation of the RSV bronchiolitis histological scoring system^67Figure 8 Interobserver variation of the RSV bronchiolitis histological scoring system^68Figure 9 Scores from the RSV bronchiolitis histological scoring system^71Figure 10 HEp-2 cell culture five days after addition of digested lung froman RSV-inoculated guinea pig (day 6 study)^ 78Figure 11 Electron micrographs of RSV-inoculated guinea pig lung (day 6 study)^81Figure 12 Immunohistochemical staining of RSV-inoculated guinea pig lung(day 6 study)^ 84Figure 13 Immunohistochemical staining of RSV-inoculated guinea pig lung(day 60 study) 87Figure 14 Agarose gel electrophoresis and autoradiography of Hep-2 cell culturesundergoing RT-PCR^ 89Figure 15 Agarose gel electrophoresis and autoradiography for day 6 and day 125RT-PCR studies 92viiiACKNOWLEDGMENTSI wish to thank Dr. J.C. Hogg for his inspiration, guidance and unwavering support; mysupervisory committee, Drs. G. Krystal, P.D. Pare, L. Haley, W. Jefferies and C. Sherlock,for their constructive input; Drs. P. Robinson, A. Bramley and R.R. Schellenberg for theircollaboration in the physiological and morphometric aspects of the guinea pig model; Ms. J.Tanch and staff of the St. Paul's Hospital Histology Laboratory for processing histologicalsections; Mr. T. Littlewood and staff at the U.B.C. Virology Laboratory for assistance in viralculture methods; Ms. F. Chu for preparation of TEM sections; Ms. Margaret McLean and Ms.J. Hards for assistance in immunohistochemistry; Drs. S. Hayashi, G. Bondy and T. Bai,Messrs. S. Bicknell and S. Granleese for advice and assistance in RT-PCR; Ms. L. Carter andMs. D. Minshall of the U.B.C. Pulmonary Research Laboratory Animal Care Facility; Mr. S.Greene for his photographic expertise; Mr. A. MacKenzie for his administrative expertise; Mr.B. Wiggs for statistical advice; and to my wife, Heather and my children, Arden and Paul, fortheir tolerance, faith and love.ix1CHAPTER 1: GENERAL INTRODUCTIONAccording to the Canada Health Survey of 1978-79, asthma affects approximately 500,000Canadians, one third of whom are under the age of 15 years (1). Epidemiological studies (2, 3)have reported marked increases in hospitalization rates and mortality from asthma in Canadasince the early 1970s and have shown that these increases are not attributable to artifacts ofdiagnosis or sampling. In 1989, the latest year for which Canadian statistics are available,children under 15 spent approximately 120,000 days in hospital due to asthma (with an averagehospital stay of 3.7 days) and 12 deaths attributable to asthma were reported in this age group(Dr. Y. Mao, Canadian Laboratory Centre for Disease Control, personal communication).While the costs of asthma to the Canadian health care system are unknown, in the UnitedStates, where there is also increased morbidity and mortality from asthma (4), the annual healthcare costs are estimated at $6 billion (5).There is currently no satisfactory explanation for these increases in the severity of asthma.This Chapter reviews contemporary ideas about the role of airway inflammation in thepathogenesis of asthma and discusses evidence that pulmonary viral infections may stimulatethe production of airway inflammation.1.1 DEFINITION OF ASTHMA. Asthma, the Greek word for "panting" (6) was used as asynonym for "shortness of breath" until the late 1800s (7), when Osler (8) recommended thatthe term be limited to denote "bronchial asthma" in contrast to so-called "cardiac asthma" or"renal asthma" associated with pulmonary edema. The Ciba Guest Symposium of 1959 (9)used clinical criteria to classify asthma as either "extrinsic" (ostensibly related to allergy') or"intrinsic" (ostensibly of non-allergic, "endogenous" origin). A poor understanding of the1 In the context of bronchial asthma, "allergy" refers to so-called "atopic allergy" or "typeI hypersensitivity"— immunologically defined as a predilection for localized IgE-mediatedanaphylactic reactions and clinically manifested by eczema, rhinitis and positive skin tests toextrinsic allergens (10). The clinical association between allergy and many cases of asthma hasbeen recognized for over 800 years (11). Atopy is defined as a familial predisposition for type Ihypersensitivity and recent genetic linkage studies purport the existence of an "atopy gene" onhuman chromosome 1 lq (12, 13).2etiology and pathogenesis of asthma resulted in the inability of a second Ciba Symposium (14)to further refine the definition. The American Thoracic Society currently defines asthma as "aclinical syndrome characterized by increased responsiveness of the tracheobronchial tree to avariety of stimuli" (15): a potential pitfall of such a broad, empirical definition is that patientswith causally unrelated conditions might be grouped inappropriately under the commonheading, "asthma" (16). Such inappropriate grouping of patients may impede the studiesneeded to achieve a more precise definition of asthma based on etiology and/or pathogenesis(17). The lack of a satisfactory definition of asthma also has important practical implications:for example, current medical therapies for asthma focus on relief of symptoms rather thanreversal of underlying cause(s). An improved understanding of the cause(s) and pathogenesisof asthma, even if applicable to only a subset of patients, may permit the design of newtherapies for cure of disease rather than control of clinical symptoms.1.2 CHARACTERISTICS OF ASTHMA. Clinically, an asthma "attack" is characterized byepisodic cough (which may be productive of sputum), wheezing and shortness of breath thatusually resolves spontaneously or following medical therapy (15). However, the widespectrum of clinical presentations often precludes the use of clinical criteria alone to diagnoseasthma2. Instead, the diagnosis of asthma may depend on the documentation of several"hallmarks" of the condition: reversible airflow obstruction, airway hyperresponsiveness andairway inflammation. Each of these hallmarks will be described below.1.2.1 REVERSIBLE AIRFLOW OBSTRUCTION. Airflow obstruction is considered presentwhen a patient forcibly exhales lower volumes of air at lower flow rates than normal subjects(15). Lung spirometry may be used to assess the extent of airflow obstruction through suchpulmonary function tests as forced expiratory volume in one second (FEV 1) and forcedexpiratory flow rates at different lung volumes (19). Although airflow obstruction is a hallmark2^The heterogeneity of asthma symptoms is reflected in the classification of Bates et al.,where asthma is classified into five categories ranging from class I (asymptomatic, inremission) to class V (unremitting symptoms, so-called status asthmaticus) (18).3of all chronic obstructive pulmonary diseases (asthma, chronic bronchitis, emphysema andperipheral airways disease), asthma is distinguished by the reversibility of airflow obstruction(e.g., improvements in pulmonary function indices) that occurs either spontaneously orfollowing the administration of bronchodilator agents such as salbutamo13. However,asthmatics "in remission" may have normal lung spirometry and further evaluation for airwayhyperresponsiveness may be required for diagnosis (18).1.2.2 AIRWAY HYPERRESPONSIVENESS. Airway hyperresponsiveness refers to theairways of asthmatic patients having a lower threshold ("increased sensitivity") and greaterextent ("increased maximal response") of narrowing following exposure to a variety ofphysical, chemical and biological stimuli4 (15). Airway hyperresponsiveness may be assessedin a pulmonary function laboratory where the subject is challenged with a known dose of a"provocative" bronchoconstrictor agent such as histamine or methacholine (an analog ofacetylcholine) and evaluated for such indices as decreased 1-EV1 or increased pulmonaryresistance. Although histamine or methacholine may elicit bronchoconstriction in normalsubjects, the increased sensitivity and increased maximal response of asthmatic patients to theseagents over a range of doses (21, 22) produce a "left-shift" of the dose-response curve thatusually permits distinction between asthmatic and normal individuals (21).Increased airway smooth muscle contraction has long been considered to be the underlyingmechanism of airway hyperresponsiveness and airflow obstruction in asthma, in part because3 While no single pulmonary function test is diagnostic of asthma, various guidelineshave been developed to standardize the interpretation of pulmonary function tests (15). Forexample, a 15% improvement in FENI following bronchodilator administration is generallyconsidered to be indicative of reversible airflow obstruction (20) since improvements in FEVi(usually less than the 15% guideline) may also occur post-bronchodilator in the other chronicobstructive pulmonary diseases and in many "normal" subjects (15).4 Examples of physical stimuli include cold air and exercise. Chemical stimuli include avariety of medications (e.g., B-adrenergic receptor blockers and non-steroidal anti-inflammatory drugs), noxious gases, inhaled and systemic allergens, inflammatory mediatorssuch as histamine and neurotransmitters such as acetylcholine. Biological stimuli includerespiratory tract infections by viruses, bacteria and fungi. Acute exacerbations of asthma mayalso be induced by psychological factors, e.g., anxiety.4drugs that relax bronchial smooth muscle (e.g., 132-adrenergic receptor agonists,anticholinergics and methyl xanthines) are efficacious in the treatment of asthma symptoms(23). However, increased airway smooth muscle contraction may represent a sequela ratherthan the underlying process in the pathogenesis of asthma because several studies have failed todemonstrate differences in airway smooth muscle mechanics (either in vitro or in vivo) betweenasthmatic patients and normal subjects (24-26). In addition, morphometric studies (27) havesuggested that the thickened submucosal layer of the airway wall in asthma potentially plays agreater role in the pathogenesis of airway narrowing than increased airway smooth musclecontraction (28, 29). The thickening of the airway wall submucosa may result from a chronicinflammatory process (30) and an improved understanding of the factor(s) which stimulate thechronic inflammation of airways may provide clues into the etiology and pathogenesis ofasthma.1.2.3 AIRWAY INFLAMMATION. Although Curschmann described asthma as "a specialform of inflammation of the small bronchioles" over a century ago (8), his idea was largelyforgotten until 1965, when Nadel reintroduced the concept of inflammatory injury to the airwaymucosa being related to airway hyperresponsiveness (31). Two factors contributing to thedelay of studying inflammatory phenomena in asthma have been the technical difficulties intissue sampling from living asthmatic patients and the prevailing view of increased bronchialsmooth muscle constriction being the underlying abnormality (23). Histological studies frompostmortem (32-37), open lung biopsy (38) and transbronchial biopsy (39) specimens havedescribed features and sequels; of inflammation in the airway walls of asthmatic patientsincluding cellular (eosinophilic, lymphocytic and polymorphonuclear) and fluid exudates,epithelial necrosis with repair (e.g., squamous and goblet cell metaplasia), bronchial glandenlargement with intraluminal mucus, smooth muscle hypertrophy, increased submucosalextracellular matrix and thickening of the bronchial subepithelial basement membrane. Thesehistopathological abnormalities reflect the presence of chronic inflammation in the airways ofasthmatic patients but none indicates the underlying cause(s) of the inflammatory process (30).5The cellular and biochemical events comprising the airway inflammation of asthma havebeen intensively studied in recent years (40, 41). The next section will describe themechanisms of "extrinsic" asthma associated with atopic allergy (section 1.1) because atopicallergy is prevalent in children with asthma.1.3 ATOPIC ALLERGY AND ASTHMA. Depending on the age group studied and the criteriaused for diagnosis, epidemiological studies report a prevalence of atopy from 23-80% amongasthmatic patients, with an especially high prevalence of atopy in asthmatic children (42).Patients with atopy develop type I hypersensitivity reactions to a variety of extrinsic antigens("allergens") such as ragweed, various fungi and, of particular epidemiological importance inchildhood asthma, the house-dust mite, Dermatophagoides pteronyssinus (43, 44). Type Ihypersensitivity consists of an initial "sensitization" step in which a predisposed individual,upon first exposure to an allergen, produces allergen-specific IgE. Upon subsequent exposureto the allergen, complexes of allergen and allergen-specific IgE bind to mast cells (via receptorsfor the F, portion of IgE) and stimulate the degranulation of mast cells. The release of chemicalmediators such as histamine (section 1.2.2) stimulates the contraction of airway smooth muscle("immediate reaction"). This reaction is followed by the "late-phase reaction", characterized byedema and inflammatory cell infiltrates (section 1.2.3). Clinically, the immediate reaction isbelieved to correspond to the early events in an acute exacerbation of asthma while the late-phase reaction is believed to correspond to the prolongation of clinical symptoms (45).Type I hypersensitivity responses appear to be regulated by both genetic and environmentalfactors. Concerning genetic factors, further characterization of the putative atopy gene onhuman chromosome 1 lq (section 1.1) may provide new insights regarding the predilection ofcertain individuals to become sensitized to various allergens. Concerning environmentalfactors, type I hypersensitivity reactions appear to be regulated by a hierarchy of lymphocytesubsets and macrophages. For example, the so-called TH2 cell (46, 47), a subset of CD4+"helper" T-lymphocytes (originally described in mice), secretes a variety of chemical mediatorsimplicated in the pathogenesis of both immediate reactions (e.g., interleulcin-4 (IL-4)) and late-6phase (e.g., IL-5, granulocyte-macrophage colony stimulating factor (GM-CSF)) reactionss.Lymphocytes with characteristics of TH2 cells are the predominant lymphocyte subsetrecovered at bronchoalveolar lavage (BAL) from human atopic asthmatic patients (49). Apossible mechanism for the increased TH2 cell activity in atopic asthma is the decreased abilityof alveolar macrophages to exert regulatory control on TH2 cells (50-52)6 . Environmentally-induced defects of alveolar macrophage function may play a crucial role in the pathogenesis ofextrinsic asthma.The above discussion has highlighted some recent advances in our understanding of thevarious cellular and biochemical events underlying the pathogenesis of airway inflammation inthe vast majority of children with asthma. Despite these advances, Nadel has emphasized thatlittle is known about the factors which "signal" the production of chronic airway inflammation(54). Evidence that persistent or latent respiratory tract viral infections 7 may possibly signal theproduction of chronic airway inflammation in asthma is presented below. Testing thispossibility is important because if respiratory tract viral infections are involved in thepathogenesis of airway inflammation in asthma (even if applicable to only a minority ofpatients), then treatment with anti-viral drugs such as Ribavirin (1-B-D-ribofuranosy1-1,2,4-triazole-3- carboxamide) (57) could potentially cure these patients of their disease.5^Among its many activities, IL-4 stimulates mast cell differentiation and promotes B celldifferentiation into IgE-secreting plasma cells. IL-5 and GM-CSF stimulate the proliferationand activation of eosinophils in late phase reactions (48).6^Alveolar macrophages phagocytose, process and present inhaled antigens to a variety ofcell types, including T-cells (53). Following experimental depletion of alveolar macrophages,there is increased TH2 cell activity (52). The mechanisms by which alveolar macrophagesinteract with antigens and regulate other cell types in vivo are not understood.7^For the purposes of this thesis, persistent infection is defined as a state in whichreplicating virus can be reproducibly and continuously recovered from the host well past theusual period of acute illness (with or without lysis of infected cells) and latency is defined as astate in which a replication-competent virus remains within the host in a reversiblynonproductive form, with replicating virus being detected only intermittently, often inassociation with clinical recurrence of disease (55, 56).71.4 VIRAL RESPIRATORY TRACT INFECTIONS AS SIGNALS FOR CHRONICAIRWAY INFLAMMATION IN ASTHMA. Although the clinical observation of upperrespiratory tract infections triggering acute exacerbations of asthma had been well recognizedfor decades, Moll (58) postulated that the presence of microorganisms within the lung wasresponsible for the development of airway hyperresponsiveness to methacholine challenge.Subsequent efforts to confirm a possible role for various bacteria (59, 60) and fungi (61) (orthe host immune response to these microorganisms) in the pathogenesis of asthma failed suchthat by the mid-1950s, a primary role for either bacteria or fungi was largely discounted (62,63). The possibility that viral infections might play a role in the pathogenesis of asthma wasinitially based on circumstantial evidence: for example, Salk showed that the extent ofapparently "allergic" responses to influenza vaccines was proportional to the concomitantamount of influenza virus within the host (64) and Stuart-Harris's group (65) observed thatbacteria were rarely cultured from asthmatic patients' sputa, in marked contrast to thefrequently positive sputum cultures from patients with chronic bronchitis or emphysema. Intheir discussion of a possible viral etiology to nasal polyps (inflammatory lesionscharacteristically affecting atopic individuals), Weille and Gohd speculated that "virus proteinor virus infection in nasal and sinus polyps and polyposis may be important in the etiology ofasthma" (66). The discovery of new respiratory tract viruses and technological improvementsin laboratory methods for their detection has produced a growing body of epidemiological,clinical, physiological, histopathological and biochemical evidence implicating viral respiratorytract infections in the pathogenesis and possibly etiology of asthma.1.4.1 EPIDEMIOLOGICAL DATA. Viral infections are implicated in asthma in at least threeways: (a) as contributors to the initial sensitization process to extrinsic allergens (section 1.3)(67); (b) as stimuli for acute exacerbations of asthma (68, 69); (b) as stimuli for ongoingairway inflammation in some asthmatic patients (69). Frick et al. prospectively studied 13children with a biparental history of atopic allergy and reported that 11 of these children had aviral upper respiratory tract infection within two months of becoming sensitized to a variety of8allergens (67). In 10/11 children, the sensitization process coincided with rising titres of virus-specific IgG, and this striking correlation led the investigators to speculate that the host'shumoral immune response contributed to the development of sensitization. Concerning virusesstimulating acute exacerbations of asthma, prospective studies have clearly demonstrated thatantecedent upper respiratory tract infections by adenovirus (70), influenza virus (70-72),parainfluenza virus (70, 73, 74), respiratory syncytial virus (RSV) (73, 74) and rhinovirus(71, 72) provoke wheezing in many asthmatic patients. Welliver's group reported thatcirculating titres of RSV-specific IgE independently predict airway hyperresponsiveness inchildren (75) and suggested that virus-specific IgE was responsible for acute exacerbations ofasthma. However, a preliminary study from a European group has not reproduced Welliver'sfindings (76).Concerning viruses possibly being signals for ongoing airway inflammation in asthma, thestrongest epidemiological association is the development of asthma in children who have hadan episode of acute respiratory syncytial virus (RSV) bronchiolitis (68). This association wasfirst reported in a retrospective analysis of 100 asthmatic children, of whom 32 haddocumented acute bronchiolitis during infancy (77). Subsequent long-term prospective studies(78-81) have confirmed the association between acute RSV bronchiolitis during earlychildhood and the subsequent development of asthma, with the reported incidence rangingfrom 21 (80) to 92 (81) per cent 8 . Although clinical, physiological, hi stopathol ogi cal andbiochemical studies suggest that RSV lung infection may play a role in the pathogenesis ofasthma in these children (see below), the biological basis of this compelling epidemiologicalassociation has not been elucidated.1.4.2 CLINICAL DATA. RSV causes over 85% of cases of acute bronchiolitis, a commonpediatric disease that accounts for approximately 1% of hospital admissions during the first twoyears of life (82). The clinical illness shares several features with an acute exacerbation of8^For comparison, the estimated prevalence of asthma is 2-5% of the general population(1, 4).9asthma, including abrupt onset of cough, dyspnea, wheezing and chest wall retractions (83).Ventilatory support may be required in severely affected patients and the case fatality rate isapproximately 1 per cent. While most patients recover within 10-14 days, RSV shedding maycontinue for up to 27 days in immunocompetent patients (84) and for up to 47 days inimmunocompromised patients (85). Intriguingly, "acute bronchiolitis" is rarely diagnosed inpatients over 4 years of age (86) and given the clinical similarities between an episode of acuteviral bronchiolitis and an acute exacerbation of asthma, investigators have recently speculatedthat the distinction between the two conditions may reflect diagnostic bias on the basis of thepatient's age rather than differences in etiology or pathogenesis (7, 54, 87).1.4.3 PHYSIOLOGICAL DATA. Numerous studies in humans (88-93) and experimentalanimals (94-100) have shown that viral respiratory tract infections provoke airwayhyperresponsiveness. Pulmonary function testing of children with acute RSV bronchiolitis hasrevealed decreased forced expiratory flow rates and increased respiratory resistance (90, 91)which are reversible following administration of bronchodilator drugs (91). Theseabnormalities of pulmonary function are consistent with reversible airflow obstruction, ahallmark of asthma (section 1.2.1). Furthermore, these abnormalities have persisted for up toeight months following clinical resolution of acute RSV bronchiolitis (92). Whilebronchoconstrictor challenge testing for airway hyperresponsiveness is unethical in humaninfants with acute RSV bronchiolitis, bronchoconstrictor challenges of RSV-infected sheep(100) and parainfluenza virus-infected rats (99) have documented both acute and persistent(over weeks to months) airway hyperresponsiveness.1.4.4 HISTOPATHOLOGICAL DATA. The histopathological features of acute bronchiolitishave been primarily determined from autopsy material (101-104) because it is not feasible tobiopsy infants and young children with acute bronchiolitis. The histopathological features offatal cases of acute bronchiolitis are common to all lower respiratory tract viral pathogens andinclude: (a) necrosis and sloughing of bronchiolar epithelium; (b) peribronchiolar mononuclearinfiltrates consisting primarily of small lymphocytes; (c) edema of the bronchiolar submucosa10and adventitia; (d) mucus and necrotic cellular debris within the airway lumen; (e) goblet cellmetaplasia of the bronchiolar epithelium; (f) in rare instances, intracellular viral inclusions.Increased amounts of bronchiolar smooth muscle and thickening of the subepithelial basementmembrane have also been observed in a case of fatal acute RSV bronchiolitis (87). Epithelialnecrosis, airway wall inflammation and edema, intraluminal mucus and metaplastic epithelialrepair are also histopathological features of asthma (section 1.2.3) and the histopathologicaldifferences between acute bronchiolitis and asthma (e.g., prominent eosinophilic infiltrates,smooth muscle hypertrophy and subepithelial basement membrane thickening in asthma) mayreflect the chronicity of the inflammatory process in asthma.1.4.5 BIOCHEMICAL DATA. There are at least three biochemical mechanisms by whichrespiratory viruses possibly contribute to the pathogenesis of asthma: (a) increased constrictionof airway smooth muscle in response to inflammatory mediators; (b) enhanced cellularsecretion of inflammatory mediators; (c) alterations of pulmonary defense mechanisms to non-specific environmental agents that trigger acute exacerbations of asthma. With respect toviruses increasing the constriction of airway smooth muscle in response to inflammatorymediators, Saban et al. (96) demonstrated that guinea pigs infected with parainfluenza virushad increased bronchial smooth muscle contraction to substance P, a neuropeptide secreted byintrinsic nerves within airway smooth muscle and implicated as a bronchoconstrictor in humanasthma (105). Busse has speculated that the damaged airway epithelium in asthma results inexposure of nerve fibres in the airway wall to non-specific environmental factors that stimulatesubstance P production (68).With respect to viruses enhancing the secretion of inflammatory mediators, Lin et al. havedocumented increased IL-2 production in vitro from peripheral blood lymphocytes of asthmaticchildren who have concurrent influenza A viral upper respiratory tract infections (106). IL-2promotes the proliferation of T-lymphocytes and these cells may secrete additional mediators(e.g., IL-4, IL-5) that stimulate the recruitment and activation of other effector cells such asmast cells and eosinophils (48, 107). Alternatively, enhanced secretion of histamine-releasing1 1factor (HRF), a cytokine that induces mast cell degranulation and histamine release (108, 109),has been described in cultured mononuclear cells infected with either influenza virus (110) orRSV (111). A third possibility is the binding of virus-IgE complexes to induce histaminerelease, as shown by Ida et al. for adenovirus and influenza virus (112). Busse's group hasextended these observations by showing that influenza virus induces histamine release incultured human leukocytes via both IgE-dependent and IgE-independent mechanisms (113).These investigators also have documented a synergistic effect on histamine release fromconcomitant virus infection and ragweed antigen exposure, consistent with a virus-allergeninteraction.With respect to viruses altering pulmonary defense mechanisms to non-specificenvironmental agents, Slauson et al. have shown that parainfluenza virus infection of alveolarmacrophages decreases the ability of these cells to phagocytose inhaled cobalt oxide particles(114). Witten et al. (115) have speculated that airway inflammation may result from the delayedphagocytosis and clearance of non-specific environmental agents by virus-infected alveolarmacrophages. Furthermore, since RSV-infected macrophages in vitro display defects ofimmunological function such as fewer cell membrane receptors for the Fc portion ofimmunoglobulin (116) and increased production of an inhibitor to IL-1 (117), these defectsmight also compromise pulmonary defense mechanisms against non-specific environmentalagents (115).Whether virus infection of alveolar macrophages results in increased TH2 cellactivity (section 1.3) is unknown.1.5 SUMMARY. Asthma is a condition characterized by reversible airflow obstruction, airwayhyperresponsiveness and chronic inflammation of airways. Epidemiological data suggest thatthe morbidity and mortality of asthma have increased in Canada over the last two decades,especially in children. There is increasing evidence that chronic inflammation of airways is theprimary abnormality responsible for the production of reversible airflow obstruction andairway hyperresponsiveness. While much is known about the various inflammatory cell typesand biochemical mediators involved in the pathogenesis of asthma, much less is understood12about the "signals" which stimulate chronic inflammation of airways. Persistent or latent viralpulmonary infections possibly signal this chronic inflammatory process, based onepidemiological, clinical, physiological, hi stopathological and biochemical studies. Ofparticular epidemiological importance is the strong association between acute RSV bronchiolitisand the development of asthma in children, a situation where persistent RSV lung infectionmay play a role in the pathogenesis of disease.13CHAPTER 2: RESPIRATORY SYNCYTIAL VIRUS2.1 HISTORICAL BACKGROUND. In 1956, a novel viral pathogen, the "chimpanzee coryzaagent", was discovered by serendipity at the Walter Reed Army Medical Center in Washington,D.C., when a group of laboratory chimpanzees developed a cluster of upper respiratory tractinfections (118). The following year, Chanock and associates (119) isolated an antigenicallyidentical virus from two infants9 with lower respiratory tract infection and the name"respiratory syncytial virus" (RSV) reflected the virus's tendency to produce fusion of infectedcells into syncytia in vitro. Subsequent epidemiological studies demonstrated that RSV was theinfectious agent most frequently isolated from children with acute bronchiolitis and pneumonia(120-122) and today RSV is regarded as "the most important respiratory pathogen of infancyand early childhood" (83).2.2 CLASSIFICATION OF RSV. Human RSV, bovine RSV and pneumonia virus of micecomprise the genus, Pneumovirus within the family, Paramyxoviridce (123). RSV lunginfection has also been described in wild mountain goats (124) and sheep (125) but these so-called "caprine" (126) and "ovine" (100) variants of RSV have not been well characterized. TheParamyxoviridce family also includes the Paramyxoviruses (e.g., parainfluenza and mumpsviruses) and Morbilliviruses (e.g., measles virus) and the common features of these virusesare: (a) single-stranded, negative polarity RNA genome; (b) assembled virions 80-500 nm indiameter; (c) 12-21 nm diameter nucleocapsids with helical symmetry; (d) assembly ofnucleocapsids in the cytoplasm of infected cells (measles virus can also assemble in thenucleus); (e) envelopment of virus on the plasma membrane of infected cells (127). RSV isdistinguished from the other Paramyxoviridce by its smaller diameter (12 nm) nucleocapsid andpropensity for infection limited to the respiratory tract and middle ear.9^The so-called "Long" strain of RSV was one of Chanock's original isolates and namedafter the patient from whom it was recovered. The Long strain is the prototypic "type A" RSV(section 2.3).142.3 STRUCTURE OF RSV. The single-stranded, non-segmented RNA of RSV encodes tengenes, all of which have been cloned and sequenced (128, 129). Table 1 lists the 3'-5'transcription order of these ten genes and the known properties of the gene products (123, 128,130). Two subtypes of human RSV (A and B) (131) have been distinguished by antigenicdifferences between the F, G and P proteins (132) and type A virus is considered to producemore severe clinical disease (133). In contrast to influenza virus, the genome of human RSV isrelatively stable (83, 123) 10 .2.3.1 NUCLEOCAPSID (N) PROTEIN. The RSV N protein is a 42 kD product of a gene1197 bases long. Between subtypes A and B of human RSV, there is 86% sequence identity atthe nucleotide level and 96% identity at the amino acid level (134). The N protein is closelyassociated with the viral genomic RNA (Table 1), has a regulatory role in viral replication(section 2.4) and is antigenic in terms of the host humoral and cellular immune responses(section 2.7). These structural and functional aspects of the RSV N protein make it an attractivetarget for the study of the natural history of RSV following acute lung infection.2.4 LIFE CYCLE OF RSV. The natural hosts of RSV are humans, cows and chimpanzees(83). Virus-containing droplets or fomites enter the host via the nasal, or ocular routes.Adsorption of RSV occurs via binding of the viral G protein to a receptor on the host cellmembrane (as yet uncharacterized), with 60-90% of virus being adsorbed by 30 minutes at37°C. Following cleavage of the viral F protein into Fl and F2 fragments by a host-derivedprotease, an exposed hydrophobic region on the F2 fragment produces holes in the viralenvelope and cell membrane to result in fusion and release of free viral nucleocapsid into thehost cell's cytoplasm.10 The further subclassification of RSV (e.g., Al, A2 strains within type A) likely reflectsantigenic drift (i.e., point mutations of genomic RNA). The major differences in severalproteins between type A and type B RSV are speculated to represent antigenic drift (i.e.,exchanges of portions of genomic RNA between distinct viruses) (131, 132).15TABLE 1: HUMAN RSV GENES AND GENE PRODUCTSGENE # RNABASES# AMINOACIDSMW(10)LOCATION FUNCTION/PROPERTIES OFGENE PRODUCT1C (NS1) 528 139 15.6 not known non-structural; acidic protein1B (NS2) 499 124 14.7 not known non-structural; basic proteinN 1197 391 42 nucleocapsid encases RNA genome in a flexiblehelixP 907 241 27.1 nucleocapsid ? component of polymerase;phosphorylatedM 952 256 28.7 inner aspectof viralenvelopecontacts N protein and internal endof viral G protein; hydrophobic,basic proteinSH(1A,N53)405 64 7.5 viral envelope non-structural; binds antibody andCD4+ T-lymphocytes; smallhydrophobic proteinG 918 298 32.6 full thicknessof viralenvelopeattachment protein of virus to hostcell membrane receptor; extensivepost-translational glycosylation(MW 84-90 kD post-glycosylation)F 1899 574 63.5 full thicknessof viralenvelopefusion of viral envelope to host cellmembrane; ? causes formation ofsyncytia. The Fl subunit has MW—20 kD and the F2 subunit has MW—43 kD.M2 957 194 22.2 inner aspectof viralenvelopematrix protein of unknown function;basic proteinL —6500 ? —200 nucleocapsid viral RNA polymerase16A viral-associated RNA polymerase modulates replication into progeny viruses within thehost cell's cytoplasm. The RNA polymerase has two distinct activities: (a) transcription of 10individual mRNAs and translation into the corresponding proteins for viral assembly; (b)transcription of the complementary mRNA of the entire RSV genome (i.e., full lengthtranscript) which serves as a template for the further full length transcription into progeny RNAgenomes by the same polymerase. Which of these two activities predominates at any given timedepends on the relative cytoplasmic concentrations full length mRNA transcripts — N proteincomplexes vs. free N protein. For example, if there is a relatively high cytoplasmicconcentration of these complexes, the polymerase will preferentially transcribe ten individualmRNAs for translation into proteins and thus increase the cytoplasmic concentration of free Nprotein. Conversely, if there is a low cytoplasmic concentration of these complexes, thepolymerase will preferentially transcribe more full length complementary mRNA. AssembledRSV matures by budding through the host cell membrane to produce cell-to-cell spread ofinfectious virus. Cellular infection with RSV produces characteristic effects in vitro and in vivoas will be discussed below.2.5 RSV INFECTION IN VITRO.2.5.1 LYTIC INFECTION. In the diagnostic virology laboratory, RSV is propagated onpermissive cell lines derived from human epithelial malignant neoplasms such as HEp-2 cellsand HeLa cells (135). The characteristic "cytopathic effect" (CPE) of RSV infection in vitro isthe formation of syncytial giant cells produced by fusion of cell membranes from adjacentinfected cells. Cell lysis occurs when the replication of progeny viruses overwhelms theinfected cell's capacity for homeostasis.Lyric RSV infection in vitro has recently been described in primary cultures of adult humannasal epithelial cells, bronchial epithelial cells and alveolar macrophages (136), three cell typessusceptible to RSV infection in vivo (104). Of the three cell types, nasal epithelial cells weremost susceptible to RSV infection in vitro, underwent the greatest amount of cell lysis andreleased the most virus into culture supernatants; alveolar macrophages were least susceptible17to infection, underwent the least amount of cell lysis and retained the most virus intracellularly.These experiments suggest that the release of RSV from nasal epithelial cells allows viralspread to neighboring cells to produce a clinical upper respiratory tract infection (section 2.6)while the retention of RSV by alveolar macrophages without cell death may result in persistentRSV infection in vivo.2.5.2 NON-LYTIC INFECTION. Acute RSV infection of alveolar macrophages in vitro hasbeen studied as a model of acute non-lytic infection (136, 137), with descriptions of virus-induced alterations in cell structure and function (section 1.3.5). In addition to acute non-lyticRSV cellular infections, several empirical studies have implicated the existence of cellular andviral factors that permit persistent non-lytic RSV infections in vitro. Concerning cellularfactors, persistent non-lytic RSV infections have been described in cultures of HEp-2 cells athigh passage number (138) and in cultures of a cell line derived from the Balb/c mouse embryo(139). Concerning viral factors, persistent non-lytic infection has been described whentemperature sensitive (ts) mutants of RSV are cultured in a variety of cell types (140). Forexample, at 31°C, three epithelial cell types" infected with ts mutants of RSV expressed viralantigens on the cell membrane and had ultrastructural features of cultured cells transformed byoncogenic nuclear viruses, including overlapping of adjacent cells and agglutination by a lowconcentration of concanavalin A (141). When the temperature was increased to 39°C, RSVantigen expression was confined to the cytoplasm and the "pseudo-transformed" phenotypewas reversed. These results showed that RSV, a cytoplasmic virus, apparently did not have tointegrate into the host cell's genome to produce persistent cellular infection or induce atransformed phenotype. The presence of viral antigens on the host cell membrane wasassociated with an altered cellular phenotype but the molecular mechanisms of these changeshave not been elucidated12.The cell lines were derived from human embryonic lung (MRC-5), feline embryo(FEA) and mink lung (Mvi Lu).12^Although persistent RSV infection has not been described in vivo, there are welldocumented examples in which persistent non-lytic infection by RNA viruses produces182.6 RSV INFECTION IN VIVO. In contrast to RSV infection in vitro, where acute lytic, acutenon-lytic and persistent non-lytic infections have been described, only acute lytic RS Vinfections are known to occur in vivo. Acute lytic RSV infections produce a spectrum ofclinical diseases ranging from the trivial common cold to the potentially life-threateningcondition, acute bronchiolitis. Children under the age of two years are at particular risk fordeveloping acute RSV bronchiolitis (123) and may go on to become asthmatic (Chapter 1). Inadults, RSV infection is usually limited to the nasopharynx as a common cold (145) but maycause pneumonia in patients who are immunocompromised, institutionalized or who haveunderlying chronic obstructive pulmonary disease (146). Several possibilities for the apparentpredilection of RSV for the lower respiratory tract of children are discussed below.2.6.1 PREDILECTION OF RSV FOR THE LOWER RESPIRATORY TRACT INCHILDREN. There is experimental evidence for at least three factors to account for theapparent predilection of RSV to the lower respiratory tract of children: (a) anatomical andfunctional characteristics peculiar to the airways of children; (b) "intrinsic" cellular factorsimparting variable susceptibilities to RSV infection in different cell types; (c) immunologicalfactors related to viral clearance and/or the possibility that the host immune response isresponsible for the pathogenesis of clinical disease. Concerning anatomical and functionalcharacteristics of children's airways, Hogg et al. (147) used a retrograde catheter technique(148) to show that the peripheral airways resistance is much higher in children than in adults,implying that a similar amount of bronchiolar inflammation would produce clinical symptomsin children but not in adults. Furthermore, Reid has speculated that the small dimensions anddeleterious effects on infected cells without known integration into host genome. For example,infection of the mouse pituitary gland with lymphocytic choriomeningitis virus (LCMV), asingle-stranded RNA Arenavirus, produces growth retardation without evidence of cell injuryor inflammation (142, 143). Mumps virus, like RSV a member of the family, Paramyxoviridce,may cause persistent infection within the central nervous system (144). In contrast, latentinfections appear to be a property of a variety of DNA viruses (e.g., herpes viruses) that canintegrate into the host's genomic DNA, although so-called "preintegration" latency (i.e., beforereverse transcription into DNA provirus) has been described for human immunodeficiencyvirus, an RNA retrovirus (56). In the absence of definitive experimental data supporting theoccurrence of latent RSV infection, only the possibility of persistent RSV infection wasconsidered in this thesis.19physical geometry of the children's airways facilitate the spread of viruses to the bronchioles(149). The combination of high baseline peripheral airways resistance with lung anatomyconducive to viral spread places children at particular risk for developing acute RSVbronchiolitis.Concerning "intrinsic" cellular factors conferring varying susceptibilities to RSV infectionin different cell types, empirical observations have confirmed that RSV preferentially infectsepithelial cell lines in vitro (135). A recent report (136) has extended these observations byshowing a differential susceptibility to RSV infection between various types of epithelial cellsfrom the same person: cultured nasal epithelial cells were readily infected by RSV infectioncompared to cultured bronchial epithelial cells. No studies to date have specifically examinedwhether respiratory epithelial cells from children have a greater susceptibility to RSV infectionthan similar cells from adults. The failure to characterize the nature and distribution of thecellular receptor(s) for RSV has greatly impeded this area of study.Concerning the role of the host immune response to RSV, it is controversial whether RSV orthe host immune response to RSV is responsible for the pathogenesis of clinical disease. Forexample, congenitally athymic mice and irradiated mice (150) and cotton rats renderedimmunodeficient by cyclophosphamide treatment (151) have been used as animal models ofpersistent, lytic RSV pulmonary infection and these models suggest that the virus itself isresponsible for the pathogenesis of inflammatory lesions. In contrast, early field trials ofpotential RSV vaccines (derived from formalin-inactivated virus) showed that, uponsubsequent natural RSV infection, vaccinated persons developed more severe pulmonarydisease than unvaccinated control subjects (152-154), suggestive of an exaggerated hostimmune response being responsible for the production of clinical diseaseo. The role of the13 Several deaths occurred in vaccinated subjects. These findings have subsequently beenconfirmed in RSV-inoculated cotton rats (155-157) in which vaccinated animals developedmore extensive pulmonary inflammation than controls. Although the mechanism for theuntoward effects of vaccination remains unexplained, the process of formalin inactivation mayhave altered viral antigens to elicit the production of antibodies deleterious to the host (152).No safe, effective RSV vaccine is currently available.20host immune response to the eradication of RSV and in the possible pathogenesis of RSV-induced disease is discussed further in section IMMUNITY TO RSV INFECTION.2.7.1 HUMORAL IMMUNITY. In humans, numerous RSV-specific antibodies of differentclasses (IgM, IgG, IgE and IgA) to various RSV surface (F, G, 1A) (158-160) and structural(N, P, L, M) (161) proteins have been identified but none of these antibodies shows areproducible correlation between circulating titres and the neutralization, clearance andprotective immunity against RSV (162, 163). Similarly, studies of humoral immunity to RSVin experimental animals (161, 164-172) have been inconclusive. As mentioned in section2.6.1, the unfortunate clinical experience of early RSV vaccines suggests that host-derivedantibodies may be involved in the pathogenesis of acute bronchiolitis. However, thiscontention has been disputed for several reasons: for example, there is no epidemiological orclinical evidence of previous RSV sensitization in children who develop acute RSVbronchiolitis and, in general, successive RSV infections tend to produce increasingly mildersymptoms in the host (162, 173).2.7.2 CELL-MEDIATED IMMUNITY. Studies in humans (163, 168, 174, 175) andexperimental animals (150, 160, 169, 176-178) suggest that virus-specific cytotoxic T-lymphocytes (CTL) play a crucial role in the host immune response to RSV 14. Exogenouslyadministered CTL can eradicate pulmonary RSV infection in both normal (178) andimmunodeficient mice (150) but giving a large number (3 x 10 6) virus-specific CTL to RSV-infected mice produces more extensive airway inflammation, increases the severity of clinicalsymptoms and may be fatal (181).In summary, there appears to be a precarious balance in the host immune response to RSVbetween eradication of the virus and exacerbation of disease. One possibility to explain whetherthe immune response is beneficial or deleterious to the host is the contribution of concomitant14^There are descriptions of CTL which recognize the RSV surface 1A protein (160, 179)and structural N protein (180) but their importance in vivo is not known.21viral factors to the host-virus system: in studies of immunity to RSV, comparatively littleattention has been paid to documenting or localizing RSV itself (section 1.4). The technical andethical considerations of tissue sampling in humans has limited the ability of investigators totest for the presence of persistent RSV lung infection. Consequently, RSV has beenexperimentally inoculated into a number of animal species to address questions that would notbe possible in humans.2.8 ANIMAL MODELS OF RSV LUNG INFECTION. Inoculation of RSV into variousexperimental animals has been attempted for over thirty years (182). Chimpanzee and monkeyare the only known "natural hosts" (besides humans) in which human RSV produces acutelower respiratory tract illness and bronchiolar inflammation (183-185). Inoculation of bovineRSV into cows also produces clinically significant symptoms and bronchiolar inflammation(186-188). Lambs inoculated with bovine RSV (94, 189-192) or so-called "ovine" RSV (100,193) develop minimal clinical signs of lower respiratory tract illness and mild bronchiolarinflammation but may develop persistent airway hyperresponsiveness (100). The high costsand specialized facilities required to maintain these large species have limited their widespreaduse (194).Smaller laboratory animals, including ferrets (195), mice (196-198) and cotton rats (199),given up to 107 plaque forming units (pfu) of human RSV, do not develop acute lowerrespiratory tract disease or histologically impressive bronchiolitis. Furthermore, these speciesdo not model the increased susceptibility of human infants to acute RSV bronchiolitis becausepulmonary infection occurs only in mature mice (198) while cotton rats of all ages are equallysusceptible (199). Despite these limitations, the cotton rat and the mouse have been favored instudies of humoral and cell mediated immunity to RSV (150, 168, 169, 172, 176, 179, 181,200, 201) and in animal trials of new vaccines (156, 157, 202, 203).To circumvent the difficulties of establishing small animal models of acute RSVbronchiolitis, some investigators have opted to infect experimental animals with other virusesthat produce clinical disease in the host: for example, canine parainfluenza virus type 2 (97) and22canine adenovirus (98) infection of beagle puppies and parainfluenza type I (Sendai) virusinfection of rats (99). While the production of acute respiratory disease and airwayhyperresponsiveness have been reported in these animal models, their major limitation is thatthe human counterparts of the animal parainfluenza and adenoviruses are epidemiologically ofminor significance (82). One small laboratory animal that has not been used as a model of acuteRSV bronchiolitis is the guinea pig, a species that potentially has advantages over the cotton ratand the mouse (section 2.9).2.9 THE GUINEA PIG AS A POSSIBLE ANIMAL MODEL OF HUMAN ACUTE RSVBRONCHIOLITIS. Previous studies of guinea pigs challenged with human RSV havedocumented the development of otitis media following instillation into the middle ear (204) andproduction of RSV-specific neutralizing and complement-fixing antibodies following intranasalinstillation (164, 182). Review of Index Medicus dating from the discovery of the chimpanzeecoryza agent until the present time fails to reveal any reports examining for pulmonary lesionsin guinea pigs inoculated with human RSV; however, the guinea pig has been used as a modelof virus-induced airway hyperreactivity to another Paramyxovirus, parainfluenza type 3 virus(95, 205).Compared to other small laboratory animals, the guinea pig particularly resembles thehuman in terms of its reproductive physiology, corticosteroid metabolism and type I(immediate hypersensitivity) and type IV (delayed hypersensitivity) immune responses (206).Because of its well characterized airway physiology, the guinea pig has long been favored asan animal model of asthma (207) 15. Factors that potentially limit the guinea pig as an animalmodel of acute RSV bronchiolitis, a pediatric disease, include the "precociousness" of newbornanimals (fully furred, teethed and open-eyed), early weaning by postnatal day 33 and earlysexual maturity at approximately 60-90 days of age (body mass > 450 g) (209). Secondly,15^With the knowledge that no species other than humans spontaneously develops anillness resembling asthma, some investigators prefer to use the expression, "experimentally-induced airway hyperresponsiveness" in reference to animal models of asthma (208).23although guinea pigs produce IgE (210), type I hypersensitivity is primarily mediated by IgGi.A third limitation is the relatively small number (compared to the mouse) of leukocyte antigenscharacterized in guinea pigs (211), thus rendering difficult studies of host lymphocyteresponses to RSV. In addition, although guinea pigs are not overly susceptible to viral lunginfections, they spontaneously develop lung infections to such bacterial pathogens asStreptococcus pneumonia? and Bordetella species (213). While one may be confident that aguinea pig inoculated with human RSV will probably not develop a lung infection from anunrelated virus or transmit RSV to another guinea pig, the presence of bacterial lung infectioncould confound the interpretation of results. Precautions to avoid undesired infections inexperimental animals (without having to resort to so-called "pathogen-free" animals (214)requiring specialized breeding, transport and housing conditions) include randomly assigningguinea pigs into RSV-inoculated and control groups, housing animals in closed rooms with noother traffic and investigators employing isolation procedures (e.g., gowns, masks, hats,gloves and shoe covers) during animal handling. With these potential advantages andlimitations in mind, in this thesis the guinea pig was used as a new animal model of acute RSVbronchiolitis to test for the possibility of persistent RSV infection in vivo within a geneticallyheterogeneous population.2.10 SUMMARY. Human RSV is a member of the Pneumovirus genus within the family,Paramyxoviridce. Although its life cycle, ultrastructure, proteins and nucleic acid sequenceshave been well characterized, there is no unifying explanation for the propensity of children todevelop lower respiratory tract infections and only limited understanding of the host immuneresponse to RSV. Other than studies of severely immunodeficient animals, investigators havepaid little attention to the possibility of persistent RSV lung infections in vivo. Animal modelsof acute RSV bronchiolitis in species that are not natural hosts are limited by such factors asdifferent age susceptibilities to lung infection compared to humans, absence of clinical featuresof acute lower respiratory tract disease and development of mild bronchiolar inflammationdespite inoculation with large amounts of virus. Previous experiments in guinea pigs have24shown that human RSV produces otitis media and a virus-specific humoral immune response:this thesis examined whether human RSV produces acute bronchiolitis and/or persistent lunginfection in guinea pigs.25CHAPTER 3: WORKING HYPOTHESIS, SPECIFIC AIMS AND STRATEGYThe first two Chapters of this thesis provided the background and rationale for the workinghypothesis and experimental portion. In Chapter 1, the concept of asthma as an inflammatoryairway disease was discussed and evidence for latent or persistent viral pulmonary infectionscontributing to this chronic airway inflammation was presented from the level of epidemiologyto biochemistry. In Chapter 2, the biology of RSV was discussed in terms of the pathogenesisof acute bronchiolitis and the circumstances under which persistent cellular infection mayoccur. This Chapter also presented the available information concerning current animal modelsof RSV lung infection and discussed the potential advantages and limitations of using theguinea pig as a new animal model to test the working hypothesis of this thesis.3.1 WORKING HYPOTHESIS. The working hypothesis of this thesis was based primarilyon the epidemiological association between acute RSV bronchiolitis and subsequent asthma inchildren and previous descriptions of persistent RSV cellular infection in vitro:Persistent cellular RSV infection may occur within the lung following resolutionof acute bronchiolitis and this chronic persistence of virus places the host at riskfor developing chronic inflammation of airways, a hallmark of asthma. The significance of this working hypothesis was that, for the first time, an attempt was made toexamine a plausible biological basis for the epidemiological association between acute RSVbronchiolitis and asthma.3.2 SPECIFIC AIMS.(1) To test whether human RSV produces acute bronchiolitis in the guinea pig;(2) To document the natural history of RSV in the guinea pig lung following primaryinfection.3.3 STRATEGY. Specific Aim 1 was a study of the comparative biology between acute RSVlung infection of humans and guinea pigs. Outbred guinea pigs (modeling a geneticallyheterogeneous pediatric human population) were evaluated by clinical and pathological criteriabased on human acute RSV lung infection. Specific Aim 2 was a study of the natural history of26intrapulmonary RSV from the time of a known primary infection. Because of possibledifferences between the human and guinea pig immune response to RSV (section 2.9), weused methods that tested directly for evidence of intrapulmonary RSV. Table 2 summarizes thetechniques used in this thesis and refers to representative previous studies of these techniquesfor the study of RSV infection.Clinical-pathological evaluation consisted of clinical examination, gross and histologicallung examination that permitted comparison of the guinea pig model to human disease (103)and the cotton rat (199) and mouse (198) models of acute RSV bronchiolitis. In addition, thisthesis tested for a possible age susceptibility of guinea pigs to develop acute RSV bronchiolitis,examined the extent of airway inflammation following infection with a larger amount of virusand examined for long-term clinical or pathological sequelx of primary RSV lung infection.The documentation of intrapulmonary RSV included viral culture to test for replicatingvirus, transmission electron microscopy (TEM) to test for assembled viral particles and viralnucleocapsids, immunohistochemistry to test for viral proteins and the reverse transcriptasepolymerase chain reaction (RT-PCR) to test for viral genomic RNA. Viral culture onpermissive cell lines in vitro is widely considered to be the "gold standard" of viral diagnosisbecause the identification of a characteristic cytopathic effect (CPE) in vitro confirms theviability of the virus (135). However, the absence of CPE does not imply absence of virusbecause persistent non-lytic infections (section 2.5) may produce "negative" cultures due tointracellular retention of virus (144).Transmission electron microscopy (TEM) is considered to be a specific technique (191,222) to document intracellular virus particles and nucleocapsids but technical aspects of tissuesampling (a typical "ultrathin" TEM section is a 60-100 nm cut through a 1 mm3 specimen)limit the sensitivity of TEM to only 5-30% (228-231). Recently, the specificity of TEM in theidentification of Paramyxoviruses has been questioned (232) because these viruses tend toassemble incompletely (220) and their tubular and filamentous nucleocapsids (127) might beconfused with normal cellular structures such as intermediate filaments and microtubules.27TABLE 2: TECHNIQUES FOR THE GUINEA PIG MODEL OF RSV LUNG INFECTIONSPECIFIC AIM (1)^CLINICAL-PATHOLOGICAL EVALUATIONTECHNIQUE PURPOSE PREVIOUSAPPLICATIONSREFERENCE(S)clinical examination(coryza, cough,tachypnea, ruffled fur,poor weight gain,decreased activity)evidence of acuterespiratory diseaseand "failure to thrive"cotton rat and mousemodels of acute RSVbronchiolitis(198, 199)gross lung examination(wet weight, grossappearance)evidence of lungconsolidation oredemahuman autopsy lungs (215, 216)airway histologicalscoring systemevaluate extent ofairway inflammationin RSV-challengedanimals vs. controlshuman chronicobstructive lungdisease(217, 218)SPECIFIC AIM (2)^DOCUMENTATION OF INTRAPULMONARY RSVTECHNIQUE PURPOSE PREVIOUSAPPLICATIONSREFERENCE(S)viral culture documentation ofreplicating viruses- humannasopharyngeal cells- human autopsylungs- experimental animallungs(219)(104)(198, 199)transmission electronmicroscopy (TEM)documentation ofassembled viralparticles and viralnucleocapsids- ultrastructure ofhuman RSV- ultrastructure ofbovine RSV(220, 221)(191, 222-224)immunohistochemistry documentation ofviral antigens- humannasopharyngeal cells- human autopsylungs(219, 225)(104)reverse transcriptase -polymerase chainreaction (RT-PCR)documentation ofviral nucleic acidsequences- human otitis media- humannasopharyngeal cells(226)(227)28In response to these concerns, Phillips has used the term "virus-related particles" to specifystructures that have features of Paramyxoviruses but are not completely assembled (233). Withthese technical limitations in mind, TEM was used as an ancillary method to examine forevidence of viral assembly within the lung.Studies employing immunohistochemistry to diagnose human acute RSV infection report asensitivity of anti-RSV antibodies from 67 to 76 percent and a specificity approaching 100%compared to viral culture (104, 219, 225). Since the working hypothesis of this thesis includedthe possibility of persistent, non-lytic RSV lung infection with potentially negative cultures,immunohistochemistry was particularly important to document evidence of intrapulmonaryRSV after resolution of acute bronchiolitis.An advantage of the reverse transcriptase polymerase chain reaction (RT-PCR) is thepotential to "amplify" cDNA (following reverse transcription of genomic RNA) to detectablelevels in specimens containing insufficient target RNA to be detected by conventional Northernblotting (234). To date, RT-PCR for RSV has been limited to specimens from acute humaninfections: middle ear effusions from patients with otitis media (226) and nasopharyngealaspirates from children with acute RSV bronchiolitis (227). In nasopharyngeal aspirates, RT-PCR had a sensitivity of 94% and a specificity of 97% compared to viral culture (227) but thesensitivity of RT-PCR to detect a known quantity of RSV RNA was not tested. While theutility of RT-PCR has been established for acute RSV infections, there are no published reportsof RT-PCR in the study of persistent RSV infections.3.4 SUMMARY. The guinea pig was used as a new animal model of human RSV lunginfection to test the working hypothesis that RSV persists within the lung following resolutionof acute bronchiolitis. The susceptibility of the guinea pig for human RSV infection wasassessed by clinical examination for evidence of acute respiratory disease, gross lungexamination for evidence of pulmonary consolidation or edema and histopathologicalexamination for features of human acute bronchiolitis. The natural history of intrapulmonary29RSV was assessed by viral culture, TEM, immunohistochemistry and RT-PCR to documentstructural aspects of RSV from the level of replicating virus to the viral genome.30CHAFFER 4: MATERIALS AND METHODS4.1 PRELIMINARY STUDY. Figure 1 summarizes the protocol of a preliminary studyrelating to Specific Aim 1 that was designed to evaluate whether human RSV produces acutebronchiolitis in guinea pigs.18 guinea pigs, body weight 270 ± 6 g (mean ± SD)— 4000 pfu of human RSV per inoculated animalJrintranasal inoculation^intratracheal inoculation^controls (no inoculation)(n = 6) (n = 6)^ (n = 6)lungs histologically examined (1 animal/group) on days 4, 5, 6, 7, 10 and 14post-inoculationFigure 1. Preliminary study of guinea pig inoculation with human RSV (Experiment #1742).The details of guinea pig handling, virus preparation and lung tissue processing forhistological examination are given below.Based on the results of this preliminary study, Specific Aim 1 was completed by using theintranasal route of RSV inoculation for the protocol outlined in Table 3.For the selection of study days post-RSV inoculation, day 6 was chosen because thegreatest amount of bronchiolar inflammation was observed on this day in the preliminarystudy. Final studies on day 6 also included evaluation of a possible age susceptibility of guineapigs to human RSV lung infection and clinical-pathological evaluation of juvenile guinea pigsgiven sevenfold more RSV. Day 14 was chosen as the "convalescent" phase of acute RSVbroncitiolitis analogous to human infection. For chronic studies, day 60 was chosen because itis approximately twice the greatest reported time that RSV has been cultured from animmunocompetent human (84, 235) and day 125 was chosen because it is more than31TABLE 3: FINAL PROTOCOL FOR INOCULATION OF HUMAN RSVINTO GUINEA PIGSSTUDY(EXPT #)*RSV(n)CONTROL(n)GUINEAPIGAGE**RSVDOSE***CULT TEM IMM RT-PCRDAY 6(1755) 10 10 juvenile low yes yes yes noDAY 6(1791) 10 10 juvenile high yes no yes noDAY 6(1830) 10 9 adolescent low yes no yes noDAY 6RT-PCR(1979)4 2 juvenile low yes yes yes yesDAY 14(1779) 10 9 juvenile low yes yes yes noDAY 60(1889) 10 11 juvenile low yes no yes noDAY 125(1950) 4 1 juvenile low yes no yes yes"EXPT #" refers to the U.B.C. Pulmonary Research Laboratory Experiment Number.**^"Juvenile" guinea pigs had body weights < 450 g (22-29 days old) when inoculatedwith human RSV while "adolescent" guinea pigs had body weights > 450 g (64-70 days old) atthe time of RSV inoculation, based on published values for these age groups (209, 236).*** "Low dose" RSV refers to inoculation of guinea pigs with 3.9 ± 0.3 x 103 pfu Longstrain human RSV (mean ± SD); "high dose" RSV refers to inoculation of guinea pigs with 2.8± 0.4 x 104 pfu Long strain human RSV.Abbreviations: CULT = viral culture of lung digests on HEp-2 cells; TEM = transmissionelectron microscopy of ultrathin sections of lung; IMM = immunohistochemistry on lung tissuesections; RT-PCR = reverse transcriptase polymerase chain reaction on homogenized lungtwice the greatest reported time that RSV has been cultured from an immunosuppressed human(85). The control animals in the final protocol received uninfected cell culture supernatant(section 4.4) in order to make the results comparable to the cotton rat and mouse models ofacute RSV bronchiolitis (198, 199).All animals had clinical-pathological evaluation (Specific Aim 1) consisting of clinicalexamination, gross lung examination and histological lung examination. For the documentation32of the natural history of intrapulmonary RSV (Specific Aim 2), animals were studied assummarized in Table 3.4.2 SPECIES. Female Cam Hartley guinea pigs (Charles River Laboratories, Montreal, PQ)were randomly assigned into either RSV-inoculated or control groups and housed in separaterooms under identical conditions of plastic cages containing Bed o'Cobs corn cob bedding(The Andersons, Industrial Products Division, Maumee, OH), access to Purina guinea pigchow (Ralston Purina Company, St. Louis, MO), alfalfa hay cubes and water and 12 houralternating light-dark cycles.4.3 VIRUS. All virus handling occurred in a Biohood Model 1148 biological safety cabinet(Forma Scientific, Marietta, OH). The Long strain of subgroup A human RSV (American TypeCulture Collection, Rockville, MD) was propagated at multiplicities of infection from 0.01-0.1(237) on HEp-2 cell monolayers (gift of U.B.C. Virology Laboratory) in 75 cm3 plastic flasks(Corning Laboratory Science Products, Mississauga, ON) at 34°C in a Fisher Model CO2 610incubator (Fisher Scientific, Napean, ON) containing humidified air with 5% CO2. The cellculture medium was Dulbecco's minimal essential medium (MEM) (Grand Island BiologicalCompany, Grand Island, NY) supplemented with 5% fetal bovine serum (FBS, heat-treated(56°C for 30 minutes) to inactivate complement) (GIBCO), 0.292 mg/mL L-glutamine(GIBCO), vitamins (GIBCO), 100 U/mL penicillin G (Sigma Chemicals, St. Louis, MO), 100pg/mL streptomycin (Sigma) and 10 pg/mL amphotericin B (GIBC0)16. To passage virus,RSV-inoculated HEp-2 cell monolayers were disrupted into 20 mL of cell culture medium bythe addition of autoclaved 3 mm diameter solid glass beads (Propper Mfg. Co. Inc., Germany)to the flask and placing the flask on a high speed Genie-2 vortex (Fisher Scientific) for 10seconds. One and one-half mL of this cell lysate were added to a 75 cm3 flask of subconfluentHEp-2 cells in 18.5 mL of fresh cell culture medium at 34°C. Formation of small syncytia, the16^For studies of days 60 and 125 post-inoculation and day 6 for RT-PCR, antibiotics andamphotericin B were not used in the cell culture medium.33characteristic RSV cytopathic effect (CPE) (Figure 2), was typically observed within 5-6 daysunder a Nikon TMS-F inverted microscope.4.4 INOCULATION PROCEDURE. On the day of inoculation of guinea pigs (designatedstudy day 0), RSV-infected HEp-2 cell monolayers showing extensive CPE were disrupted asdescribed above. Lysed cell suspensions underwent centrifugation at 800 x g on a benchtopcentrifuge (International Equipment Company (IEC) Model Centra-8, Needham Heights, MA)at 4°C for 4 minutes and the supernatant was transferred to a sterile Falcon 2057 tube (BectonDickinson, Lincoln Park, NJ) using an autoclaved cotton-plugged Pasteur pipette (Kimble,Toledo, OH). Supernatant from disrupted uninfected HEp-2 cells was obtained in a similarmanner.Guinea pigs were anesthetized with 3-5% halothane (Ayerst Laboratories, Montreal, PQ) inoxygen delivered by a Medishield Inhalational Anesthetic Machine (Ohio Medical, Rexdale,ON). Depending on the group to which they were stratified, anesthetized animals receivedeither 100 pL of RSV-containing supernatant or 100 /AL uninfected supernatant deliveredintranasally using an Eppendorf pipetter with autoclaved 0-100 pL tips (Elkay Products, Inc.,Shrewsbury, MA) 17.4.5 VIRAL PLAQUE ASSAY. In order to quantify the amount of viable RSV deliveredintranasally to each animal, viral plaque assays were performed in duplicate as described byLennette and Schmidt for the identification of RSV syncytia in infected HEp-2 cells (238).Supernatants remaining from the inoculation of guinea pigs were serially diluted (1:103 to1:105) in MEM/5% FBS and 0.5 mL of these diluents were added to confluent HEp-2 cells inFalcon 3046 six well culture plates (Becton Dickinson). After allowing 90 minutes for viraladsorption at 37°C, the diluent was removed and replaced by17^In the preliminary experiments, RSV was also instilled intratracheally to 6 guinea pigsvia a flexible plastic tube attached to a 1 mL tuberculin syringe (Becton Dickinson).34Figure 2.^Photomicrographs of HEp-2 cell monolayers (inverted microscope). Panel ashows RSV-infected Hep-2 cells with the formation of syncytia (arrowheads) characteristic ofRSV cytopathic effect (final magnification: x208). Panel b shows uninfected HEp-2 cells withthe normal "cobblestone" pattern in vitro (final magnification: x104).35362.5 mL of a liquid mixture of 2 parts "double strength" dsMEM (i.e. MEM prepared in half thevolume of dH2O)/10% FBS to 3 parts autoclaved 1% agarose (GIBCO BRL, Gaithersburg,MD) in dH2O. After the dsMEM-agarose mixture hardened (usually within one minute), theculture plates were transferred to the CO2 incubator at 34°C and kept for 5-7 days until syncytiacould be discerned under a Nikon TMS-F inverted microscope.Cells were fixed over 30 minutes at 34°C by the diffusion of 2 mL 10% neutral bufferedformalin (BDH Chemicals, Toronto, ON) through the dsMEM-agarose. The outer aspect of thedsMEM-agarose was then cut with a spatula blade and removed (taking care to avoid damage tounderlying cells). Cells were stained with 0.5% neutral red (Fisher Scientific) for 1 minute andgently rinsed several times in tap water. The number of "plaques" (syncytia) were counted ineach well with the mean being expressed as the number of plaque forming units (pfu) per 100pL of original supernatant given to guinea pigs.4.6,CLINICAL EVALUATION. For the day 6 and 14 studies, guinea pigs were examineddaily for signs of respiratory disease (coryza, coughing or tachypnea) and general signs ofillness (ruffled fur, decreased activity) (198). Body weights were recorded on the applicablestudy day. For several studies (Appendix A), animals had "initial" (day of inoculation) and"final" (study day) body weights recorded to evaluate changes in body weight over the studyinterval.4.7 GROSS LUNG EXAMINATION AND TISSUE PROCESSING. Guinea pigs wereanesthetized by intraperitoneal injection with 0.5 mL euthanyl forte® (MTC Pharmaceuticals,Cambridge, ON) and exsanguinated via right ventricular needle puncture. The chest wasopened and the heart and lungs were removed en bloc. Following suture ligation of the rightmainstem bronchus, the right lung was isolated and weighed in a sterile vessel.The left lung was inflated through the trachea with 5 mL of 4% paraformaldehyde (BDHChemicals) in phosphate buffered saline (PBS) (Oxoid, Basingstoke, UK) containing 0.1%diethylpyrocarbonate (DEPC) (Aldrich Chemical Company, Milwaukee, WI). After overnight37fixation at 4°C, serial 3 mm sagittal sections of the left lung were examined for gross evidenceof consolidation.Midsagittal slices (239) of paraformaldehyde-fixed lungs were processed in the usual wayfor paraffin embedding (240) by an Histomatic Automated Tissue Processor Model 266 MP(Fisher Scientific). Serial 4 j4L sections were prepared by Histology Laboratory staff at St.Paul's Hospital and stained with hematoxylin and eosin (H&E) and periodic acid-Schiff (PAS)stains.The right middle lobe was dissected and processed for viral culture and TEM (sections 4.9and 4.10). For day 6, day 14 and day 60 studies, the remainder of the right lung was frozen inliquid nitrogen (with or without inflation with cryoprotective O.C.T. Compound (MilesScientific, Elkhart, IN) and stored at -70°C in an UT 1786 freezer (Revco Scientific Inc.,Asheville, NC). In the day 125 study, a portion of the right middle lobe was processed forviral culture (section 4.9), the remainder of the right lung was inflated with 4% phosphatebuffered paraformaldehyde through the pleura as described by Churg (241) and processed forlight microscopy while the left lung was prepared for RT-PCR (section 4.12).4.8 LIGHT MICROSCOPY. The variability of bronchiolar inflammation induced by intranasalinoculation of RSV to that induced by uninfected cell culture supernatant was compared using asemi-quantitative scoring system based on previous reports of RSV infection in experimentalanimals (145, 192) and human chronic obstructive pulmonary disease (217, 218). Bronchioleswere scored for six histological features including epithelial necrosis, mononuclear cellinfiltrates and edema because they are features of human acute viral bronchiolitis (103);polymorphonuclear (PMN) cell infiltration as an index of concomitant acute inflammatoryphenomena (242); hyperplasia of bronchus-associated lymphoid tissue (BALT) as an index ofthe mononuclear cell response (171) and goblet cell metaplasia as an index of respiratoryepithelial repair (243). For a given airway, each feature was scored from 0 (normal) to 2(moderate to severe changes) by comparison to photomicrographs of guinea pig airways(Figure 3). Ten bronchioles covering all regions of the lung section were examined per slide38and the observed score for each histological feature was expressed as the sum of individualairway scores (maximum score of 2 x 10 = 20 per parameter). The criteria for selection ofairways to be scored included: (a) evaluation limited to membranous bronchioles (i.e.muscular, non-cartilaginous airways); (b) scoring a given airway only once; (c) avoidance ofnearby cuts of an already scored airway.A Zeiss Photomicroscope II was used for scoring. Slides were coded such that the originof a particular section (control or RSV-inoculated animal) was not known to the microscopist.Upon completion of all scoring, the code was broken and statistical analysis performed (section4.14).4.9 VIRAL CULTURE. One half of the fresh right middle lobe was minced into fine pieceswith a sterile razor blade, transferred to a 120 mL sterile specimen jar (Fisher Scientific)containing an autoclaved magnetic stir bar and digested into single cells by addition of 10 mLfiltered 0.25% trypsin (GIBCO) in PBS and placing the jar on a Corning Model PC353magnetic stirrer (Corning, NY) for 90 minutes at 37°C. The single cell suspension wastransferred to a sterile Falcon 2057 tube and centrifuged at 800 x g (IEC Model Centra-8) atroom temperature for 4 minutes. Following centrifugation, the supernatant was removed andreplaced with 1 mL sterile cell culture medium. The resulting suspension was added tosubconfluent HEp-2 cell monolayers growing in 25 mL Corning flasks containing 4 mL cellculture medium at 34°C (135). Flasks were examined daily under the Nikon invertedmicroscope for signs of RSV cytopathic effect (CPE). Cells which did not show CPE werepassaged at weekly intervals for up to one month using the protocol described above for thepassage of uninfected HEp-2 cells. A culture was classified as "positive" when CPE wasobserved, regardless of the number of syncytia. A culture was classified as "negative" when noCPE was observed over the one month interval.To quantify the number of replicating viruses in the guinea pig lung parenchyma at day 6post-inoculation, viral plaque assays (section 4.5) were performed using trypsin-digested39Figure 3.^RSV bronchiolitis scoring system: standard photomicrographs. Unlessotherwise specified, all panels are hematoxylin and eosin stained sections of guinea pig lungwith final magnification: x208.Panel a: Grade 0 (normal) membranous bronchiole.Panel b: Grade 1 respiratory epithelial cell necrosis. Some epithelial cells aresloughed into the airway lumen.Panel e:^Grade 2 respiratory epithelial cell necrosis. There is more extensiveepithelial cell sloughing with intraluminal necrotic debris.Panel d: Grade 1 mononuclear cell infiltrates. Mononuclear cells have a peri-bronchiolar distribution.Panel e: Grade 2 mononuclear cell infiltrates. Mononuclear cells extendthrough the bronchiolar smooth muscle into the subepithelial space.Panel f: Grade 1 polymorphonuclear (PMN) cell infiltrates. PMNs havepredominantly a peri-bronchiolar distribution with occasional cellsin smooth muscle and subepithelial space (final magnification: x260).Panel g: Grade 2 PMN infiltrates. There is more extensive PMN infiltrationinto smooth muscle and subepithelial space than in Grade 1 (finalmagnification: x260).Panel h: Grade 1 airway wall edema. There is mild dilatation of the adventitia.Panel i:^Grade 2 airway wall edema. Congestion of adventitia1 vessels isapparent.Panel j: Grade 1 BALT. The bronchiole contains a small proportion ofBALT.Panel k: Grade 2 BALT. The bronchiole has prominent BALT.Panel I:^Grade 1 goblet cell metaplasia. PAS-positive goblet cells constitute25-50% of bronchiolar epithelial cells (PAS stain).Panel m: Grade 2 goblet cell metaplasia. PAS-positive goblet cells constituteover 50% of bronchiolar epithelial cells (PAS stain).40414243454^1°• •I••P t1, •4 tri • -n• •s‘r,k^..b?i /1 ' )^a 1re^i * N. .• •4,ge4647lung homogenates on 4 RSV-inoculated guinea pigs and 2 controls from the day 6 RT-PCRstudy. The results were expressed as the number of pfu/g wet weight of fresh lung.4.10 TRANSMISSION ELECTRON MICROSCOPY (TEM). The remaining half of the rightmiddle lobe was inflated with 2% glutaraldehyde (Electron Microscopy Services (EMS), FortWashington, PA) in 0.1 M sodium cacodylate buffer (EMS) through a 25 gauge needle(Becton Dickinson) as previously described (241) and immersed in this fixative for 2 hours at4°C. Fixed lung tissue was cut with a razor blade into 1 mm3 cubes, washed three times in 0.1M cacodylate buffer, postfixed for 1 hour in 1% °sat (EMS) at room temperature, washed for15 minutes in dH20, dehydrated by serial 10 minute immersions in 30%, 50%, 70%, 90% and100% ethanol and infiltrated with LR White (Polysciences, Warrington, PA). Specimens wereplaced into plastic beam capsules containing LR White which polymerized overnight at 65°C.Multiple semi-thin (0.5 jim) sections were prepared, stained with toluidine blue (FisherScientific) and those sections containing foci of bronchiolar inflammation were selected forpreparation of ultrathin (60-80 nm thickness) sections using an ultramicrotome (Ultracut,Reichert, Austria). Ultrathin sections were mounted on copper grids (EMS) and examinedunder a Philips 400 transmission electron microscope.Lung sections from 14 RSV-inoculated guinea pigs at day 6 and 2 RSV-inoculated guineapigs at day 14 were examined by TEM. Ultrathin sections from RSV-infected HEp-2 cells werepositive controls (Figure 4) and sections from uninfected HEp-2 cells were negative controls.4.11 IMMUNOHISTOCHEMISTRY. The protocol was adapted from Neilson et al., whodocumented RSV antigens in paraffin-embedded sections of human lung obtained at autopsy(104). Five pm thick sections of paraformaldehyde-fixed, paraffin-embedded guinea pig lungwere incubated with 0.1% protease, type XIV (Sigma) in 0.5 M TBS, pH 7.6 (TBS) (BDHChemicals) at 37°C for 10 minutes to disrupt protein crosslinks induced by fixation. Followingbrief rinses in tap water and 95% Et0H, sections were incubated with 0.9% H202 (BDH48Figure 4. Electron micrograph of RSV-infected HEp-2 cells. There are completely assembledRSV seen in cross-section (arrowheads) and in longitudinal section (arrow). Cross-sections show the trilaminar membrane with projections ("fuzzy" coat) and centrallylocated electron dense nucleocapsids (bar represents 100 nm).4950Chemicals) in methanol (Baxter Health Care Corporation, Muskegan, MI) for 25 minutes atroom temperature to eliminate endogenous peroxidase activity. This incubation was followedby a tap water rinse and 5 minute wash in TBS at room temperature.To prevent non-specific IgG binding, sections were preincubated in normal swine serum(DAKO) diluted 1:20 in primary antibody diluting buffer (Biomeda) for 30 minutes at roomtemperature. Incubation with primary rabbit anti-RSV antibody B344 (DAKO, Denmark) 18diluted 1:300 in TBS/2% bovine serum albumin fraction V (BSA) (BDH Chemicals)/1%human AB serum was performed for 90 minutes at room temperature. Negative control slideswere incubated in parallel with TBS/2% BSA/1% human AB serum in the absence of anti-RSVantibody. Following incubation with primary antibody, sections were washed in TBS for 5minutes at room temperature.Sections were next incubated with biotinylated swine anti-rabbit secondary antibody(DAKO) diluted 1:300 in TBS/2% BSA/1% human AB serum for 45 minutes at roomtemperature, followed by a 5 minute wash in TBS. A 45 minute incubation at room temperaturein peroxidase-conjugated streptavidin (DAKO) diluted 1:600 in TBS/2% BSA/1% human ABserum was followed by a 10 minute wash in TBS.The colorimetric peroxidase reaction consisted of developing sections in 20 jAL workingAEC solution (1 drop 3-amino-9-ethylcarbazole (Sigma), 1 drop 3% H202 in 3 mL 0.1Msodium acetate, pH 5.2 (Fisher Scientific)) for 15 minutes at room temperature. Following arinse in dH2O, sections were counterstained with Mayer's hematoxylin for 1 minute at roomtemperature and rinsed with dH2O. Coverslips were mounted using Immu-Mount aqueousmounting medium (Shandon Labs, Pittsburgh, PA).18 The polyclonal rabbit anti-RSV antibody recognizes epitopes from several viralproteins, including the surface F protein and the nucleocapsid N protein (P.C. Grauballe,DAKO Patts, Denmark, personal communication). A mouse monoclonal anti-RSV N proteinantibody (SeroTec MCA 491, Oxford, UK) cross-reacted with normal guinea pig lungparenchyma and was therefore not used.51Formalin-fixed, paraffin-embedded lung tissue from two fatal cases of human acute RSVbronchiolitis (courtesy of Dr. J. Dimmick, B.C. Children's Hospital and Dr. L. Holloway,New Zealand) were used as positive controls and a formalin-fixed, paraffin-embedded blockfrom an adult autopsy lung (St. Paul's Hospital, Vancouver, BC) was used as a negativecontrol.4.12 REVERSE TRANSCRIPTASE-POLYMERASE CHAIN REACTION (RT-PCR).4.12.1 PRELIMINARY STUDIES OF RT-PCR. The RT-PCR method was initially developedusing total cellular RNA extracted from RSV-infected HEp-2 cells because these cells were aknown source of viral genomic RNA. The sensitivity of the PCR amplification step was testedby using serial dilutions of a known amount of plasmid DNA (pGEM 3) containing the fulllength cDNA of the RSV N gene (generous gift of Dr. P.L. Collins, National Institutes ofHealth, Bethesda, MD). The specificity of RT-PCR was tested by using several controls: (a)no nucleic acid template; (b) uninfected HEp-2 cells; (c) RSV-infected HEp-2 cells which didnot undergo reverse transcription prior to PCR; (d) RSV-infected HEp-2 cells whichunderwent pretreatment with 5 units RNase A (Sigma) at 37°C for 30 minutes prior to reversetranscription. The first control tested whether solutions were contaminated by either RSV orplasmid DNA. Uninfected HEp-2 cells were a source of non-specific total cellular RNA and thetwo positive controls were a source of viral genomic RNA. Once the optimal conditions forRT-PCR were established, the protocol was used in two studies of juvenile guinea pigsinoculated with low dose RSV: day 6 (n = 4 RSV-inoculated, 2 controls) and day 125 (n= 4RSV-inoculated, 1 control).4.12.2 SELECTION OF OLIGONUCLEOTIDES. For reasons given in section 2.3.1, the1197 base N gene of human RSV (129) was chosen as a target for PCR amplification. Threeoligonucleotides were synthesized at the DNA Synthesis Laboratory, University of Calgary:(1) 5' GCG ATG TCT AGG TTA GGA AGA A 3' (bases 223 to 244 of the N genemRNA sense strand);(2) 5' GCT ATG TCC TTG GGT AGT AAG COT 3' (the complement of bases 632to 609 of the N gene mRNA sense strand);(3) 5' TAG CTC CAG AAT ACA GGC ATG ACT C 3' (bases 449 to 473 of the Ngene mRNA sense strand) .52The first two oligonucleotides were used as flanking PCR primers and the third was used as aprobe of an internal sequence of the predicted PCR product. The oligonucleotides were chosenby the following criteria: (a) sequences toward the "middle" of the target N gene; (b) lengthgreater than 20 bases (to ensure high specificity); (c) numerical balance between A+T (doublehydrogen bonds) and G+C (triple hydrogen bonds); (d) no predicted secondary structure thatwould result in significant "folding" of the oligonucleotide onto its own complementary basesequences; (e) no significant sequence homology to the other two oligonucleotides or to otherknown viral, human or rodent genomic sequences available on the Genbank® database; (f)sufficient distance between the flanking primers to yield a reasonably-sized PCR product of410 bp.4.12.3 PURIFICATION OF OLIGONUCLEOTIDES. Crude oligonucleotide pellets weredissolved in 1.5 mL 0.5 M ammonium acetate (BDH Chemicals). Sep-Pak C18 cartridges(Millipore Corporation, Milford, MA) were prepared for chromatography by passing 10 mL100% acetonitrile (Fisher Scientific) through the cartridge, followed by passage of 10 mLdH2O. The dissolved oligonucleotides were transferred to a sterile 3 mL syringe (BectonDickinson) and injected into the Sep-Pak cartridge. Ten millilitres of dH2O were put throughthe cartridge and equal fractions were collected in seven 1.5 mL microfuge tubes. One millilitreof room air was then injected into the cartridge through a syringe to remove remaining dH2O.Three millilitres of 20% acetonitrile were passed through the cartridge and equal 1 mLfractions were collected in separate microfuge tubes. The DNA content of each fraction wasestimated by measuring the optical density at a wavelength of 260 nm with a Perkin ElmerUV/VIS Lambda 2 Spectrophotometer (Uberlinger, Germany) and Hellma quartz glasscuvettes (Concord, ON). Oligonucleotides were dried by evaporation of 20% acetonitrile on arotary Speedvac SC 100 with refrigeration unit RT 100 and vacuum pump VP 100 (SavantInstruments Inc., Farmingdale, NY) over 2 hours at 65°C and the pellets were resuspended inautoclaved dH2O to a final concentration of 20 J4M and stored at -70°C.534.12.4 TOTAL CELLULAR RNA EXTRACTION19. Total cellular RNA from guinea pig lungwas extracted by a modification of the method of Chomczynski and Sacchi (244). Three mL ofsolution D (4M guanidinium isothiocyanate (Sigma); 25 mM sodium citrate, pH 7 (BDHChemicals); 0.5% sarcosyl (Sigma); 0.1 M 2-mercaptoethanol (Sigma)) was instilled into theguinea pig lung via a plastic tube inserted through a tracheostomy. The lung was excised,frozen in liquid nitrogen for 30 minutes, transferred to an autoclaved 50 mL Oakridgecentrifuge tube (Nalge Company, Rochester, NY) filled with 2 mL solution D. The lung washomogenized into solution D by three 15 seconds pulses on a Polytron (Kinematic GMBH,Lucerne, Switzerland) set at full speed. To the homogenized lung was added 0.5 mL 2 Msodium acetate, pH 4.1 (BDH Chemicals), 5 mL water-saturated molecular biology gradephenol (Bethesda Research Laboratories (BRL), Gaithersburg, MD) and 1 mL of a 24:1mixture of chloroform (BDH Chemicals) and isoamyl alcohol (Sigma). After vigorouslyagitating the lung homogenate, the Oakridge tube was placed on ice for 15 minutes andunderwent centrifugation at 10000 g (Beckman Model J 21-C) at 4°C for 20 minutes. Fivehundred microlitre aliquots of the upper aqueous phase (containing the RNA fraction) wereaspirated with an Eppendorf pipetter using autoclaved aerosol-resistant tips (ContinentalLaboratory Products, Seattle, WA) and transferred to autoclaved 1.5 mL microfuge tubes(National Scientific Supply Company, Inc., San Rafael, CA). An equal volume of isopropanol(BDH Chemicals) was added to each aliquot and the RNA precipitated over one hour at -20°Cin an Amana Model 17 freezer (not frost-free).A pellet of RNA was obtained by centrifugation at 10000 g (MSE MicroCentaur, JohnsScientific, UK) at 4°C for 10 minutes and the supernatant was aspirated with an aerosolresistant pipette tip as described above. The RNA pellet was dissolved in 0.3 mL of solution D19 For cultured cells grown in 75 cm3 Corning flasks, the volumes of all solutions werereduced by 80%. Cell monolayers were washed with sterile PBS. PBS was aspirated out of theflask with an autoclaved cotton-plugged Pasteur pipette, solution D was poured directly ontocell monolayers and the mixture was transferred into an autoclaved Oakridge tube. ThePolytron was not used.54and precipitated with 0.3 mL isopropanol at -20°C over one hour. A pellet of RNA wasobtained by centrifugation at 10000 g at 4°C for 10 minutes; the supernatant was aspirated andthe pellet was stored in 250 JAL 95% Et0H at -70°C until use for RT-PCR20.4.12.5 REVERSE TRANSCRIPTION. The conditions for reverse transcription of total cellularRNA from guinea pig lung homogenates into cDNA were similar to the method of template-specific RT-PCR recently described by Shulinder et al. (234). Each RNA sample was preparedby aspiration of 95% Et0H from the pellet, washing briefly in 250 pL 70% Et0H containing0.1% DEPC and drying at room temperature in a vacuum desiccator (Wheaton Dry Seal,Millville, NJ). The dried RNA pellet was dissolved in 25 pL 0.1% DEPC-dH20, heated for 10minutes at 65°C to remove secondary RNA structure and quick chilled on ice.The reverse transcriptase reaction occurred at 37°C over 75 minutes in a total volume of 20pL per sample consisting of 5 pL RNA template, 50 mM KC1 (Fisher Scientific), 10 mM Tris,pH 8.3 (BDH Chemicals), 8 mM MgC12 (BDH Chemicals), 200 /AM of pooled ultrapuredeoxynucleotide triphosphates (dNTPs) (Pharmacia, Montreal, PQ), 2 pM of theoligonucleotide, 5' GCG ATG TCT AGG TTA GGA AGA G 3' (i.e. the primer for reversetranscription in the direction complementary to genomic viral RNA), 40 units of RNasin(Boehringer Mannheim, Montreal, PQ) and 16 units of Moloney murine leukemia virus reversetranscriptase (MoMuLV RT) (Pharmacia). The reverse transcriptase was inactivated by heating20 One aliquot of each specimen was run on a formaldehyde-containing agarose gel (245)to evaluate the degree of RNA preservation. Only those samples which had well-defined 28Sand 18S bands indicative of good RNA preservation (Figure 5) were used for RT-PCR.Formaldehyde gel electrophoresis consisted of preparing a 1% agarose gel in DEPC-dH20containing 1X running buffer (5X running buffer is 0.1 M 34N-morpholino)-propane sulfonicacid (MOPS), pH 7 (Sigma), 40 mM Na acetate and 5 mM EDTA) and 2.2 M formaldehyde(BDH Chemicals). RNA samples were resuspended in 0.1% DEPC-dH20 as described insection 4.12.5, of which 4.5 pL was added to 2 pL 5X running buffer, 3.5 pL formaldehydeand 10 pL deionized formamide (Fisher Scientific). After heating each specimen at 65°C for 10minutes and quick chilling on ice (section 4.12.5), 1 pL of 1 mg/mL ethidium bromide inDEPC-dH20 and 2 pi, formaldehyde gel loading buffer (50% glycerol (BDH Chemicals), 1mM EDTA, pH 8, 0.25% bromophenol blue (Fisher Scientific), 0.25% xylene cyanol) wereadded to each tube. In the Fisher Biotech minigel apparatus (section 4.12.7), the gel wasimmersed in 1X running buffer, pre-run for 5 minutes at 30 volts, and samples were added tothe wells. After a 1 hour run at 30 volts, the gel was removed and photographed as describedin section 5. Agarose-formaldehyde gel electrophoresis of total cellular RNA. The distinct 28Sand 18S bands of 30 pg samples of total cellular RNA from the Day 6 RT-PCRstudy (Lane 1) and the Day 125 study (Lane 2) are indicative of good RNApreservation compared to the markedly denatured RNA from the Day 60 study(Lane 3).57the specimen to 99°C in an Isotemp Dry Bath 147 (Fisher Scientific) for 5 minutes. Specimenswere kept on ice until PCR amplification.4.12.6 PCR AMPLIFICATION. Each PCR amplification reaction occurred in a total volume of50 pL consisting of 514L reverse transcriptase cDNA product, 50 mM KCI, 10 mM Tris, pH8.3, 5 mM MgC12, 200 pM of pooled ultrapure dNTPs, 5 mM of both flanking primers and,after initially heating the mixture to 100°C for 10 minutes, addition of 2.5 units (0.5 pL) of Taqpolymerase (GIBCO BRL) to each tube. A GeneAmp PCR System 9600 (Perkin Elmer Cetus)was used for 35 cycles of PCR amplification, each cycle consisting of denaturation (94°C, 1minute) and annealing/extension (70°C, 2 minutes) 21 . Following completion of the last cycle,tubes were kept at 4°C until required for agarose gel electrophoresis.4.12.7 AGAROSE GEL ELECTROPHORESIS (246). Eight well, 1.5% agarose (BRL) gelsin lx TBE (10X TBE is 0.9 M Tris, pH 8; 0.9M boric acid (BDH Chemicals); 20 mM EDTA)and 0.01% ethidium bromide (BRL) were cast for electrophoresis using a Fisher Biotech FB103 minigel apparatus (Fisher Scientific) with 1X TBE as the electrophoresis buffer.Twenty-five microlitres of each PCR product were added to 5 pL of loading buffer (15%Ficoll 400 (Pharmacia); 0.1 M EDTA, pH 8; 1% SDS (Fisher Scientific); 0.25% xylene cyanol(Eastman Kodak Company, Rochester, NY)). Samples were loaded into designated wells andelectrophoresis performed at 100 volts over 45 minutes. Gels illuminated over a Fisher BiotechFBTI 816 UV light source (wavelength: 312 nm) were photographed using a Polaroid MP-3Land Camera (Polaroid Corporation, Cambridge, MA) (f 5.6, shutter speed 0.5 seconds) withPolaroid 667 ISO 3000 Professional Print Film.4.12.8 SOUTHERN TRANSFER (247). Following photography, gels were immersed indenaturing solution (1.5 M NaCl (BDH Chemicals); 0.5 M NaOH (Fisher Scientific)) for 3021^In the first cycle, the denaturation time was 5 minutes and in the last cycle, theannealing/extension time was 10 minutes to ensure completeness of the extension (234).58minutes at room temperature, washed briefly in dH20 and immersed in neutralizing solution(1.5 M NaCl; 1 mM EDTA; 0.5 M Tris, pH 7.5) for 30 minutes at room temperature.Each gel was then placed on top of a moist strip of Whatman 3MM Chromatography Paper(Whatman International Ltd., Maldstone, UK) whose ends were immersed in a baking dishcontaining 20X SSC transfer buffer (20X SSC is 3M NaCI; 0.3 M Na citrate.2H20, pH 7(BDH Chemicals)). A Hybond-N membrane (Amersham, Arlington Heights, IL), cut so thatits length and width were each 2 mm greater than the gel's, was placed against the uppersurface of the gel, covered with a sheet of Whatman 3MM paper (cut to the same size as theHybond-N) and Saran Wrap was placed around the outer 2 mm edges. Paper towels wereplaced on top of the Whatman paper and a 500 g weight was placed on top of the paper towelsto facilitate capillary transfer of DNA from the gel to the Hybond-N membrane over 18 hours.Following Southern transfer, DNA was crosslinked to the Hybond-N membrane byexposure to UV light (wavelength: 250 nm) (UVP Inc., San Gabriel, CA) for 4 minutes atroom temperature and stored in a sealed plastic bag at -70°C.4.12.9 PREPARATION OF OLIGONUCLEOTIDE PROBE. The 5' labeling of the thirdsynthetic oligonucleotide (section 4.12.2) with gamma 32P-ATP (Amersham) wasaccomplished by the polynucleotide kinase reaction followed by DEAE-cellulosechromatography (248). In a 1.5 mL microfuge tube, 1 pL (20 pmoles) of oligonucleotide wasmixed with 2 yL 10X kinase buffer (1 M Tris, pH 8; 0.1 M MgCl2; 1 mM EDTA; 1 mMspermidine (Sigma)), 2 pL 0.1 M dithiothreitol (DTT) (Sigma), 5 pL of gamma 32P-ATP(2000 Ci/mmole; 10 pCi/pL) and 10 pL dH20. One microlitre (4.5 units) of T4 polynucleotidekinase (Pharmacia) was added to the tube, the contents briefly spun on a benchtop centrifugeand incubated in a Magni-Whirl water bath (Blue M Electric Company) at 37°C for 45 minutes.The polynucleotide kinase reaction was stopped by heating the tube in a New BrunswickScientific Model G76 Gyrotory Waterbath (Edison, NJ) at 65°C for 10 minutes. Two hundredmicrolitres of TE (10 mM Tris, pH 8; 1 mM EDTA, pH 8) and 10 pL of 10 mg/mL E. coiltRNA were then added to the tube.59A 0.7 cm x 10 cm Bio-Rad column was packed with 0.5 mL of DEAE-cellulose (Sigma),rinsed with 50 mL of 1 M NaCI in TE and equilibrated with 50 mL of TE. The sample wasloaded into the column, rinsed with 1 mL of TE and the effluent was collected. Theunincorporated nucleotide was eluted with 1 mL 0.2 M NaCI in TE, repeated four times. Thelabeled oligonucleotide was eluted into two 0.5 mL fractions with 1 M NaC1 in TE. Thefraction containing the highest amount of radioactivity (measured on a 5 pL aliquot in aBeckman LS 7500 scintillation counter) was used as a probe in filter hybridization.4.12.10 FILTER HYBRIDIZATION AND AUTORADIOGRAPHY. In a sealed plastic bag,the Hybond-N membrane was incubated with 40 pL/cm 2 prehybridization solution consistingof 6X SSC, 5X Denhardt's solution (100X Denhardt's stock solution is 10 g Ficoll 400, 10 gpolyvinylpyrrolidone (Sigma) and 10 g BSA Fraction V per 500 mL dH2O), 0.5% SDS, 0.05M sodium phosphate buffer, pH 6.8 (Fisher Scientific) and 20 pg/mL E. coli tRNA(Boehringer Mannheim) in a gyrotory water bath at 65°C for 2 hours. Followingprehybridization, the radiolabeled probe was added to the bag for an 18 hour hybridization at65°C.Posthybridization washes included three 5 minute washes in 6X SSC at room temperatureand, to increase stringency, a 60 minute wash in 6X SSC at 65°C. For autoradiography, themembrane was taken to a darkroom, covered with Saran Wrap, apposed to a sheet of KodakEktascan IR Diagnostic Xray Film and kept inside a Kodak X-OMatic Film Cassette withregular intensifying screens. Xray films were exposed from 4 hours (at room temperature) to72 hours (at -70°C) and developed in a Kodak RP X-OMat Processor.4.13 PHOTOGRAPHY. Light microscopic photomicrographs (H&E, PAS andimmunoperoxidase slides) were shot on an Olympus AH-2 photomicroscope using 100 ASAKodak Tmax film. This film was also used for photography of Polaroid prints from ethidiumbromide-stained agarose gels and autoradiographs. Negatives were developed in Kodak Tmaxdeveloper, printed by an Agfa-Gevaert Rapidoprint DD3700 printer (Germany) onto Ilfordmultigrade paper (Ilford, Cheshire, UK). Electron micrographs were shot on a Philips 40060transmission electron microscope using Kodak 4489 3 1/4" x 4" EM plate film. Negatives weredeveloped in Kodak D19 developer and printed as described above.4.14 STATISTICAL ANALYSES. Statistical analyses were performed on clinical, grosspathological and histological data (Specific Aim 1) from the day 6, 14 and 60 studies usingSY STAT® Version 5.1 software (Systat, Inc., Evanston, IL). No statistical analysis wasperformed for the day 125 study because there was only one control animal.For the purposes of documenting the natural history of intrapulmonary RSV (Specific Aim2), viral cultures, TEM, immunohistochemistry and RT-PCR were interpreted as either positive(unequivocal signal observed) or negative (no signal observed). An animal was considered tohave evidence of RSV lung infection whenever a positive signal was observed by any of thefour methods, as opposed to reliance on a "gold standard" such as viral culture. This approachwas used for three reasons: (a) the possibility of persistent, non-lytic infection producing "falsenegative" viral cultures on HEp-2 cells (sections 2.5.2 and 3.3); (b) the use of known positiveand negative control specimens for each method (permitting concomitant evaluation of themethod's sensitivity and specificity during each experiment); (c) the protocols were designed tobe highly specific: the inevitably lowered sensitivities made quantitative or semi-quantitativeanalyses inappropriate.4.14.1 CLINICAL AND GROSS LUNG EXAMINATION. The Student's t-test was used tocompare mean body weights, lung wet weights and lung to body weight ratios (all continuousvariables) between RSV-inoculated and control groups. A p value s 0.05 was considered to bestatistically significant.4.14.2 RSV BRONCHIOLITIS HISTOLOGICAL SCORING SYSTEM. The intraobservervariation and interobserver variation of the RSV bronchiolitis histological scoring system wereevaluated by calculating the Pearson coefficient of mean-square contingency (R2) for eachhistological feature and expressing R2 as a fraction of the maximum possible value, R2„,ax(249). The Pearson coefficient of mean-square contingency is an extension of the Pearson chi-square test (the null hypothesis being that the rows and columns of a matrix are independent).61The histological scores from two analyses of the same airways were expressed as a 3 x 3matrix, with 100% agreement being found along the diagonal (a score of "0" from run 1 alsobeing scored "0" on run 2, etc.). However, in contrast to the null hypothesis of the Pearsonchi-square test, the histological scoring system was intended to be highly reproducible (i.e.,scores ideally being identical between and within observers); therefore, the rows and columnswere expected to show marked dependence. The Pearson coefficient of mean-squarecontingency was used to evaluate dependence between rows and columns in the matrix by thecalculation of R2:R 2 = if:1_:-,, ,where T = Pearson chi-square statistic and N = number of observations. The interpretation ofR2is facilitated when R2 is expressed as a proportion of the maximum possible value, R2max:iicijTR 2 max = ^, qwhere q = the number of rows (or columns) in a square matrix. Concerning the histologicalscoring system, R2max would represent the value of the Pearson chi-square coefficient if thereis 100% agreement in the airway scores: for a 3 x 3 matrix, R2 max = z ,. 0.82.The intraobserver variation was assessed by rescoring 50 airways from randomly selectedslides three months after the initial scoring and the interobserver variation was assessed by twopathologists (Dr. Sergio Gonzalez Bombardier (SGB), Catholic University, Santiago, Chileand the author (RGH)) independently scoring 30 airways on an Olympus BH-2 microscopewith a teaching head attachment. For analysis of both intra- and interobserver variation, anR2/R2max ratio 0.75 was considered to represent acceptable reproducibility.4.14.3 LIGHT MICROSCOPIC EVALUATION. For each study, the nonparametric MannWhitney U test (250) was used to compare observed scores (ordinal variables) between RSV-inoculated and control groups for each histological feature. In addition, this test was used to62compare histological scores from 9 control animals at day 6 (given uninfected cell culturesupernatant) to 6 "historical" (unmanipulated) controls from the preliminary studies describedin section 4.1. The rationale for using the Mann Whitney U test was twofold: (a) this test doesnot require particular assumptions about the guinea pig populations' airway scores (incomparison, the underlying assumption of the Student t-test would be that the airway scores ofthe guinea pig populations are normally distributed with equal variances); (b) this test permitscomparison between categories (as represented by the standard photomicrographs). Theconcomitant scoring of six histological features represented a scenario of multiple comparisons;to account for multiple comparisons, the sequential rejective Bonferroni procedure (251) wasused to determine statistical significance at sequential p values: 0.0083 (0.05/6); then 0.01(0.05/5) etc. until all six histological features had been analyzed. A sequential rejectiveBonferroni procedure was used in preference to unmodified Bonferroni analysis (i.e., setting asignificant p value at 0.0083 for all histological features) because the latter method is anoverly conservative analysis with low statistical power (252).4.15 GENERAL TECHNICAL NOTE. Distilled H2O was prepared through a Milli-Q FilteringApparatus (MilliPore Corporation, Bedford, MA) and autoclaved in an Amsco Vacamatic S(Amsco, Erie, PA) for 20 minutes at 121°C under 20 lb/in2 steam pressure. For RNA work,DEPC was added to dH20 at a concentration of 0.1% (by volume), vigorously shaken intosolution and autoclaved as above. Solutions containing Tris buffer were prepared using apreviously unopened container of fresh crystals added to dH20 that did not contain DEPC(253). Pyrex® glassware (Corning) and metal utensils were either autoclaved for 30 minutes at132°C under 30 lb/in2 steam pressure or, for RNA work, baked overnight at 200°C in a Single-Wall Transite Oven (Blue M Electrical Company, Blue Island, IL). Reagents were weighed onMettler Model AE-50 or PC 4400 balances (Mettler Instruments AG, Zurich, SW) and pHadjustments of solutions were made using a Beckman Model 4500 digital pH meter (BeckmanCorporation, Irvine, CA). Latex gloves (Smith & Nephew Perry, Massillon, OH) were worn63for all experiments involving virus and animal handling (sections 4.1-4.7), viral culture(section 4.9) and RT-PCR (section 4.12).64CHAPTER 5: RESULTS 225.1 PRELIMINARY EXPERIMENTS. Inoculation of approximately 4000 pfu of Long strainhuman RSV into guinea pigs did not produce clinical evidence of acute respiratory disease andno animal died during the course of the study. Histological evaluation of lung parenchymafrom the RSV-inoculated group showed a patchy inflammatory process, centered aroundbronchioles from all lung lobes, evident by day 4 post-inoculation and maximal on day 6.Respiratory epithelial necrosis, mononuclear cell peribronchiolar infiltrates, PMN bronchiolarwall infiltrates and airway wall edema were present (Figure 6) and the extent of inflammationwas similar between intranasally inoculated animals and intratracheally inoculated animals.Neither alveolitis nor bronchopneumonia was observed in RSV-inoculated animals. By day 14the inflammatory process had substantially resolved.Some bronchioles in control animals had one or more of the histopathological features ofacute bronchiolitis. The patchy nature of bronchiolar inflammation in the RSV-inoculated groupand the presence of inflammatory background "noise" in the control group necessitated thedevelopment of the semi-quantitative, pictorially-based histological scoring system (section4.8) to discriminate between bronchiolar inflammation attributable to RSV inoculation and non-specific inflammatory background. Figure 7 shows the results and statistical analysis forintraobserver variation and Figure 8 shows the results and statistical analysis for interobservervariation. There was substantial agreement in airway scoring within and between observers forall six histological features examined: in particular, there were no instances of any airway beingscored "0" on one occasion and "2" on another. The acceptably low intra- and interobservervariation of this relatively simple scoring system were encouraging for airway scoring in thelarger studies of the final protocol.22^Appendix A shows all data collected for each guinea pig studied in this thesis.Appendix B shows the statistical analyses for airway histological scores.65Figure 6. Low power photomicrograph of guinea pig lung, day 6 following intranasal RSV-inoculation (preliminary experiment). There is marked peri-bronchiolarinflammatory cell infiltration and edema (arrowheads) with relative sparing of thealveoli. The bronchiole at top left has epithelial cell necrosis with sloughing into theairway lumen. (H&E stain, final magnification: x52).6667run2EPITHELIALNECROSISrun 1sr.,ore 0 1 20 24 2 01 3 14 02 0 1 6R2: 0.76R2/R2max: 0.93MONONUCLEARINFILTRATESrun 1srere1 20 35 2 01 1 10 02 0 1 1R2: 0.73R2/R2max: 0.90PMNINFILTRATESrun 1score1 20 21 2 01 2 21 32 0 1 3R2: 0.76R2/R2max: 0.94run2run2run2run2run2AIRWAY WALLEDEMArun 1.%re 0 1 20 13 3 01 2 26 02 0 0 6R2: 0.78R2/R2max : 0.96BALTHYPERPLASIArun 1s%re 0 1 20 40 0 01 1 6 02 0 1 2R2: 0.77R2/R2max: 0.94GOBLET CELLMETAPLASIArun 1seore 1 20 20 1 01 0 12 22 0 0 15R2: 0.79R2/R2max : 0.97Figure 7: Intraobserver variation of the RSV bronchiolitis histological scoring system: n = 50airways; run 1 = airway score on first run; run 2 = airway score on repeat evaluation. Boldfacenumbers indicate the score (0, 1 or 2). Each box within the 3 x 3 matrix contains the numberof observations (plain type) corresponding to each score from a given run. The Pearson chi-square coefficient of mean square contingency (R2) is expressed on its own and as a ratio ofthe maximum possible value (R2/R2x) as described in section 1secire 0 1 20 11 2 01 1 14 02 0 1 1R2: 0.78R2/R2max : 0.89MONONUCLEARINFILTRATESobs 1score 0 1 20 16 0 01 1 12 02 0 0 1R2: 0.81R2/R2max : 0.99PMNINFILTRATESobs 1score0 1 20 23 1 01 1 5 02 0 0 0R2: 0.62R2/R2max : 0.76obs2obs2obs2AIRWAY WALLEDEMAobs 1seore 1 20 16 3 01 0 11 02 0 0 0R2: 0.63R2/R2max :0.77BALTHYPERPLASIAobs 1'col.e 0 1 20 28 1 01 1 0 02 0 0 1R2 : 0.71R2/R2max : 0.87GOBLET CELLMETAPLASIAobs 1seore 0 1 20 12 1 01 2 11 02 0 0 4R2: 0.78R2/R2max : 0.96obs2obs2obs2Figure 8: Interobserver variation of the RSV bronchiolitis histological scoring system: n = 30airways; obs 1 = score given by RGH; obs 2 = score given by SGB. Boldface numbersindicate the score (0, 1 or 2). Each box within the 3 x 3 matrix contains the number ofobservations (plain type) corresponding to the scores given by the two observers. The Pearsonchi-square coefficient of mean square contingency (R2) is expressed on its own and as a ratioof the maximum possible value (R2/R2max) as described in section SPECIFIC AIM 1: CLINICAL-PATHOLOGICAL EVALUATION.5.2.1 CLINICAL AND GROSS LUNG EXAMINATION. No RSV-inoculated guinea pigdeveloped clinical evidence of coryza, cough, tachypnea, ruffled fur or decreased activity.Table 4 summarizes the results of body weights, lung wet weights and lung to body weightratios which revealed no statistically significant differences between RSV-inoculated animalsand uninfected controls. In summary, RSV-inoculation into guinea pigs did not produceclinical or gross pathological evidence of pulmonary disease.5.2.2 LUNG HISTOPATHOLOGICAL EVALUATION. Figure 9 shows the results ofscoring using the histological bronchiolitis scoring system. On day 6 post-inoculation, juvenileguinea pigs inoculated with low dose (4000 pfu) RSV had statistically significantly greaterrespiratory epithelial necrosis (p s 0.002), bronchiolar PMN infiltrates (p s 0.008) andmononuclear cell infiltrates (p s 0.01) compared to controls (Figure 9a). One control animalwas omitted from the analysis due to severe aspiration pneumonia. For juvenile guinea pigsinoculated with high dose (28000 pfu) RSV (Figure 9b), there was also statisticallysignificant epithelial necrosis, mononuclear infiltrates and PMN infiltrates compared touninfected controls but none of these features was significantly different from the low doseRSV group. In contrast to juvenile guinea pigs on day 6 post-inoculation, adolescent guineapigs inoculated with RSV had no statistically significant differences in airway histologicalscores compared to controls (Figure 9c). In the day 6 RT-PCR study (Figure 9d), there was atrend for the RSV-inoculated group to have higher scores for epithelial necrosis, mononuclearinfiltrates, PMN infiltrates and goblet cell metaplasia but the small number of animals in eachgroup did not reveal any statistically significant differences.At days 14 (Figure 9e) and 60 (Figure 9f) there were no statistically significant differencesin histological scores between the RSV-inoculated group and the control group.23^Histological scores between the control animals of these groups were also similar(Appendix B).70TABLE 4. GUINEA PIG BODY WEIGHTS, LUNG WET WEIGHTS ANDLUNG TO BODY WEIGHT RATIOSSTUDY(EXPT #)GROUP nINITIALBODYWEIGHT*(g)(mean ± SD)FINALBODYWEIGHT(g)(mean ± SD)LUNGWETWEIGHT(g)(mean ± SD)LUNG TOBODYWEIGHTRATIO (%)(mean ± SD)DAY 6 RSV 10 - 395 ± 26 1.61 ± 0.24 0.38 ± 0.04(1755) CONTROL 10 - 412 ± 26 1.62 ± 0.29 0.41 ± 0.05p value** NS NS NSDAY 6 RSV 10 303 ± 16 335 ± 18 1.23 ± 0.15 0.37 ± 0.05(1791) CONTROL 10 308 ± 19 359 ± 18 1.20 ± 0.11 0.36 ± 0.04p value NS NS NS NSDAY 6 RSV 10 - 519 ± 15 2.32 ± 0.13 0.44 ± 0.05(1830) CONTROL 10 - 515 ± 18 2.38 ± 0.21 0.46 ± 0.03p value NS NS NSDAY 6(1979) RSV 4 292 ± 5 326 ± 7 1.16 ± 0.04 0.36 ± 0.01(RT-PCR) CONTROL 2 318 ± 13 355 ± 34 1.25 ± 0.21 0.36 ± 0.02p value NS NS NS NSDAY 14 RSV 10 - 469 ± 24 1.70 ± 0.32 0.38 ± 0.08(1779) CONTROL 9 - 463 ± 41 1.90 ± 0.32 0.39 ± 0.11p value NS NS NSDAY 60 RSV 10 364 ± 19 650 ± 53 3.12 ± 1.37 0.48 ± 0.22(1889) CONTROL 11 360 ± 20 661 ± 63 2.89 ± 1.34 0.43 ± 0.18p value NS NS NS NSDAY 125 RSV 4 388 ± 24 826 ± 48 3.07 ± 0.61 0.37 ± 0.07(1950) CONTROL 1 340 722 3.26 0.45* -: not done** NS: p value > 0.05 by Student t-test71Figure 9. Scores from the RSV bronchiolitis histological scoring system. Each pointrepresents the sum of individual scores for 10 airways in a single animal. Legendfor all panels: EPITH NECROSIS = respiratory epithelial necrosis; MONO =mononuclear cell infiltrates; PMN = polymorphonuclear cell infiltrates; EDEMA =airway wall edema; BALT = BALT hyperplasia; GOBLET = goblet cell metaplasia.Detailed statistical analyses appear in Appendix B.Panel a: Day 6 study: juvenile guinea pigs, low dose RSV.Panel b: Day 6 study: juvenile guinea pigs, high dose RSV.Panel c: Day 6 study, adolescent guinea pigs, low dose RSV.Panel d: Day 6 RT-PCR study, juvenile guinea pigs, low dose RSV.Panel e: Day 14 study, juvenile guinea pigs, low dose RSV.Panel f: Day 60 study, juvenile guinea pigs, low dose RSV.Panel g: Day 125 study, juvenile guinea pigs, low dose RSV.Panel h: Day 6 control guinea pigs (same controls as panel a) compared tounmanipulated "historical" controls of the preliminary study.72o CONTROLS• RSV-INOCULATED* * p c 0.002* p t 0.00873•0•0OO• •^ O es• • ••^111111^OD^ 0^IPSIMO^OS^MI o •• coo^0300 • •••••^0 • ••^SO^•^000000 OD 0 • 00 •^woo so^•^co ow^o •000 ••• 03X:00• CXX) •••^00003 ••74e2015•• oo00•SOCC0(I)inv •0 0000 ••••o •••• 0 •• •• 0 •• 0 Me00 00 0 •CO • 0 COO ••••000 0 OD COD 0IMO 0••CCO • 00 000•••• MOOO coo • • o • 0 •EP ITH^MONO^PM4^EDEMA^BALT^GOBLETNECROSISo CONTROLS• RSY-INOCULATED15cc0 10U)f20EP ITH^MONO^NAN^EDEMA^BALT^GOBLETNECROSISO CONTROLS• RSV-INOCULATEDwcc0O^Eij000000 0O CO ^0 0 ^ ^000 1333 0 ^ 0 ^0 0 ^ 0 0 0 003 ^00 133 ^ 0 ^ 0000CO ^ 00 ^ 0 1333•• • • • •• • • • • • •0 10^75EP ITH^MONO^PMV^EDEMA^BALT^GOBLETNECROSISo CONTROLS• RSV-INOCULATEDhEP ITHNECROSISMONO PM EDEMA BALT GOBLETO DAY 6 CONTROLS• HISTORICAL CONTROLS76On day 125, the control animal had a comparable score to the RSV group for each histologicalfeature (Figure 9g). Finally, there were no statistically significant differences for any of the sixhistological features between the control groups given uninfected cell culture supernatant andthe "historical" (unmanipulated) controls from the preliminary study (Figure 9h).In summary, juvenile guinea pigs inoculated with as few as 4000 pfu of RSV developedhistological features of human acute RSV bronchiolitis on day 6 which resolved by day 14.Secondly, on day 6 post-inoculation, no significant bronchiolar inflammation developed inadolescent guinea pigs given a similar amount of RSV. Thirdly, the intranasal delivery ofuninfected cell culture supernatant did not elicit significant bronchiolar inflammation in controlguinea pigs compared to "historical", unmanipulated controls.5.3 SPECIFIC AIM 2: NATURAL HISTORY OF INTRAPULMONARY RSV. Table 5shows the results of studies performed on juvenile guinea pigs given low dose RSV. Theseresults will be presented in further detail below.5.3.1 VIRAL CULTURE. Whether inoculated with low dose or high dose RSV, 9/10 juvenileguinea pigs on day 6 showed evidence of viral infection by the development of thecharacteristic RSV cytopathic effect (CPE) in HEp-2 cells (Figure 10). RSV CPE wasobserved in 2/10 RSV-inoculated juvenile guinea pigs on day 14 but not on days 60 or 125.No HEp-2 cell culture showed CPE characteristic of other viruses. In the day 6 study ofadolescent guinea pigs, the cell culture medium was contaminated with yeast which killed theHEp-2 cell monolayers; thus, viral cultures in this group were not completed. There were nosporadic bacterial or fungal infections in HEp-2 cell cultures containing digested guinea piglung (note that the cell culture medium for the day 60, day 125 and day 6 RT-PCR studies didnot contain antibiotics or amphotericin B). In the day 6 RT-PCR study, viral plaque assayrevealed a mean ± SD of 1.7 ± 0.3 x 103 pfu RSV/g wet weight fresh lung.77TABLE 5: DOCUMENTATION OF INTRAPULMONARY RSV INVIRUS-INOCULATED GUINEA PIGS *STUDY(EXPT #)CULTURE TEM IMMUNOHISTO-CHEMISTRYRT-PCRDAY 6 9/10 2/10 7/10 -(1755)DAY 6 9/10 - 8/10 -(1791)DAY 6 ** - 6/10 -(1830)DAY 6 414 1/4 2/4 4/4(RT-PCR) 1.7±0.3 x 103 pfu/g(1979)DAY 14 2/10 0/2 6/10 -(1779)DAY 60 0/10 - 1/10 -(1889)DAY 125 0/4 - 0/4 3/4(1950)* Results are presented for RSV-inoculated guinea pigs only because no control animalsstudied had a positive result by any method.** Yeast contamination of culture medium precluded analysis for RSV CPE in this group.-: not done.78Figure 10.^HEp-2 cell culture five days after addition of digested lung from an RSV-inoculated guinea pig (day 6 study). Note the syncytial formation characteristic of RSV CPE(inverted microscope, final magnification: x800).79805.3.2 TRANSMISSION ELECTRON MICROSCOPY. In ultrathin sections of RSV-inoculated guinea pig lung, only rare bronchial epithelial cells contained structures consistentwith completely assembled RSV (Figure 11a) (191), incompletely assembled "virus-relatedparticles" (Figures 1 lb and 11c) (220) and viral nucleocapsids (Figure 11d) (216). The virus-related particles did not have the ultrastructural features of cilia, microvilli or phagocyticvacuoles (254). The viral nucleocapsids were discernible from intermediate filaments (10 nmdiameter), microtubules (25 nm diameter) associated with cilia, actin microfilaments (6 nmdiameter) associated with microvilli, rough endoplasmic reticulum, immunoglobulinaggregates, lamellar bodies of type II pneumocytes or granules from Clara cells,neuroendocrine cells, neutrophils, eosinophils or mast cells (254).No virus-related particles were observed in bronchiolar epithelial cells, type I or IIpneumocytes, Clara cells, neuroendocrine cells or alveolar macrophages. No virus-relatedparticles or nucleocapsids were identified in two RSV-inoculated animals in the day 14 studyand consequently TEM was not performed for either the day 60 or day 125 studies.5.3.3 IMMUNOHISTOCHEMISTRY. For juvenile guinea pigs on day 6, the sensitivity ofimmunohistochemistry to detect RSV antigens in lung sections ranged from 50 to 80 per cent(Table 5). Most positive staining was observed within the cytoplasm of bronchiolar andbronchial epithelial cells (Figure 12), with substantially fewer type II pneumocytes and alveolarmacrophages staining positively. A similar cellular distribution of staining was observed in6/10 adolescent guinea pigs on day 6. However, on day 14, most staining was observed withinalveolar macrophages, with only scattered bronchiolar and bronchial epithelial cells stainingpositively. Figure 13 shows unequivocal cytoplasmic staining within an alveolar macrophageof an RSV-inoculated animal on day 60. No positive staining was observed in the mid-sagittallung sections examined from the day 125 study. The control lung sections from two fatal casesof human acute RSV bronchiolitis stained positively during each run and no staining was everobserved in the negative control from human autopsy lung.81Figure 11.^Electron micrographs of RSV-inoculated guinea pig lung (day 6 study).Panel a: Budding viruses (arrowheads) are larger than normal microvilli (my)of the bronchial epithelial cells but smaller than the cilium (c) seen inoblique section (white bar represents 250 nm).Panel b: Two incompletely assembled "virus-related particles" showingtrilaminar membrane with projections (spikes) but paucity ofnucleocapsid material (bar represents 100 nm).Panel c: Cross section of a virus-related particle showing central electron densenucleocapsid material (bar represents 100 nm).Panel d: Intracytoplasmic cluster of filamentous RSV nucleocapsids(bar represents 100 nm).82835.3.4 RT-PCR. Figure 14 shows the results of agarose gel electrophoresis and correspondingautoradiography for HEp-2 cell samples which underwent RT-PCR. A band of the predicted410 by size was observed in samples containing as few as 5 copies of plasmid DNA with theRSV N gene insert (Figure 14a, Lane 8) and was the most prominent band in samples fromRSV-infected HEp-2 cells (Figure 14a, Lanes 6 and 7). Of the less prominent, smaller bandspresent on ethidium-bromide stained agarose gels from RSV-infected HEp-2 cells (Figure 14a,Lanes 6 and 7), one band was faintly positive by autoradiography.The specificity of RT-PCR for the RSV genomic RNA target sequence was confirmedwhen no bands were observed in samples that did not contain a nucleic acid template, insamples of uninfected HEp-2 cells and in samples of RSV-infected HEp-2 cells whichunderwent RNase digestion before reverse transcription. The sensitivity and specificity of RT-PCR was confirmed on multiple runs.Figure 15 shows the ethidium bromide-stained agarose gels and correspondingautoradiographs of Hybond membranes hybridized with the 32P-labeled oligonucleotide probefrom an internal sequence of the predicted PCR product. A specific band of 410 by was presentin 4/4 RSV-inoculated guinea pigs at day 6 and in 3/4 RSV-inoculated guinea pigs at day 125.RT-PCR was not done on RNA extracted from frozen guinea pigs lungs in the original day 6,day 14 and day 60 studies because of extensive RNA degradation in these samples.84Figure 12.^Immunohistochemical staining of RSV-inoculated guinea pig (day 6 study).Panel a:Panel b:Panel c:Panel d:Low power photomicrograph showing an inflamed membranousbronchiole with a few epithelial cells staining positively with polyclonalanti-RSV antibody (arrowhead). (Hematoxylin counterstain; finalmagnification: x208).High power view of RSV-positive cells from the airway of panel a,confirming intracytoplasmic staining for RSV antigens. (Hematoxylincounterstain; final magnification: x416).Same airway as panel a without incubation with anti-RSV antibody.There is absence of non-specific staining in this control section.(Hematoxylin counterstain; final magnification: x208).High power view of airway from panel c, confirming the absence ofnon-specific staining (Hematoxylin counterstain; final magnification:x416).858687Figure 13.^Immunohistochemical staining of RSV-inoculated guinea pig lung (day 60study). There is intracytoplasmic staining in the alveolar macrophage at centre.(Hematoxylin counterstthn; final magnification: x800).8889Figure 14.^Agarose gel electrophoresis and autoradiography of HEp-2 cell culturesundergoing RT-PCR. Top: agarose gel stained with ethidium bromide; Bottom:autoradiograph after 24 hour exposure.Panel a: Lane 1: PUC 18-Hinf I digest (size markers).Lane 2: blankLane 3: no template, RI and PCR stepsLane 4: no template, PCR step onlyLane 5: uninfected HEp-2 cellsLane 6: RSV-infected HEp-2 cellsLane 7: RSV-infected HEp-2 cells (duplicate of sample in lane 6)Lane 8: RSV N gene cDNA in pGEM3, 5 copiesThe agarose gel shows a number of bands in Lanes 6, 7, and 8, with the predictedPCR product of 410 bp being most prominent. The autoradiograph shows aspecific band of the predicted 410 bp PCR product in Lanes 6, 7 and 8corresponding to positive samples. There is also a specific band from a samplecontaining 5 copies of target cDNA (Lane 8).Panel b: Lane 1: PUC 18-Hinf I digest (size markers).Lane 2: uninfected HEp-2 cellsLane 3: RSV-infected HEp-2 cells, PCR step only (no RT step)Lane 4: duplicate of sample in lane 3Lane 5: blankLane 6: RSV N gene cDNA in pGEM3, 500 copiesLane 7: RSV-infected HEp-2 cells with RNase pretreatment prior tothe RI stepLane 8: RSV-infected HEp-2 cellsThe agarose gel shows that performing PCR without prior reverse transcription didnot result in amplification of specific DNA from RSV-infected 1-lEp-2 cells (Lanes 3and 4) and that RNase pretreatment did not result in amplification of PCR product(Lane 7).90a2 76543I92Figure 15.^Agarose gel electrophoresis and autoradiography for day 6 and day 125 RT-PCR studies. Top: agarose gel stained with ethidium bromide; Bottom:autoradiograph after 72 hour exposure.Panel a: Day 6 RT-PCR study.Lane 1: PUC 18-Hinf I digest (size markers).Lane 2: no template (negative control).Lane 3: 1979-3 (RSV-inoculated)Lane 4: 1979-5 (RSV-inoculated)Lane 5: 1979-6 (RSV-inoculated)Lane 6: 1979-7 (RSV-inoculated)Lane 7: 1979-19 (control)Lane 8: 1979-20 (control)The agarose gel shows a number of non-specific bands in each of the lanes; theautoradiograph shows a specific band of the predicted 410 bp PCR product inLanes 3, 4, 5 and 6 corresponding to 4/4 guinea pigs inoculated 6 days previouslywith RSV.Panel b: Day 125 study.Lane 1: PUC 18-Hinf I digest (size markers).Lane 2: 500 copies of RSV N gene cDNA in pGEM 3 (positivecontrol of PCR step).Lane 3: no template (negative control).Lane 4: 1950-2 (RSV-inoculated)Lane 5: 1950-3 (RSV-inoculated)Lane 6: 1950-4 (RSV-inoculated)Lane 7: 1950-5 (RSV-inoculated)Lane 8: 1950-1 (control)The agarose gel shows a number of non-specific bands in each of the samples; theautoradiograph shows a specific band of the predicted 410 bp PCR product inLanes 4, 5 and 6 corresponding to 3/4 guinea pigs inoculated 125 days with RSV.9394955.4 SUMMARY. Inoculation of juvenile guinea pigs with human RSV resulted in significantbronchiolar epithelial necrosis, mononuclear cell bronchiolar infiltrates and PMN bronchiolarinfiltrates on day 6 post-inoculation in the absence of detectable clinical disease or grosspulmonary lesions. The extent of bronchiolar inflammation was similar whether juvenile guineapigs were given —4000 pfu or —28000 pfu of RSV. By day 14, the bronchiolar inflammationhad substantially resolved and there was no chronic bronchiolar inflammation on either day 60or day 125. In contrast to the juvenile group on day 6, no significant bronchiolar inflammationwas observed in a group of adolescent guinea pigs inoculated with —4000 pfu of RSV.RSV was cultured from the lung parenchyma in 9/10 juvenile guinea pigs inoculated with—4000 pfu of RSV on day 6, from 2/10 animals on day 14 and none of the RSV-inoculatedanimals on days 60 or 125. TEM revealed completely assembled virus, virus-related particlesand viral nucleocapsids in rare airway epithelial cells on day 6. On day 6, RSVimmunohistochemistry was positive in 7/10 juvenile and 6/10 adolescent guinea pigs inoculatedwith low dose RSV, in 8/10 juvenile animals given high dose RSV and in 2/4 RSV-inoculatedanimals in the RT-PCR study. RSV antigens were identified in lung sections from 6/10juvenile animals on day 14 and 1/10 juvenile animals on day 60. On day 6, positiveimmunostaining was present predominantly within bronchiolar epithelial cells; on day 14,alveolar macrophages were the predominant RSV-positive cell type and on day 60, positiveRSV staining was present only within alveolar macrophages. RSV genomic RNA wasdocumented in the lung of 4/4 infected guinea pigs at day 6 and in 3/4 infected guinea pigs atday 125.In summary, while human RSV did not produce clinical evidence of acute lower respiratorytract disease in guinea pigs, younger animals had statistically significant bronchiolarinflammation on day 6 which resolved by day 14. Despite resolution of the inflammatoryprocess, evidence of intrapulmonary RSV was documented up to 125 days after primaryinoculation.96CHAPTER 6: DISCUSSION6.1 SPECIFIC AIM 1: PRODUCTION OF ACUTE RSV BRONCHIOLITIS IN THEGUINEA PIG. The results of these experiments showed that intranasal inoculation of —4000pfu of human RSV into anesthetized, one month old juvenile guinea pigs produced maximalhistological evidence of acute bronchiolitis on day 6, without concomitant clinical signs ofacute respiratory disease or gross pulmonary lesions. Bronchiolar inflammation resolved byday 14 and no chronic inflammatory sequelx were observed on days 60 or 125. Increasing theinoculated dose of virus to —28000 pfu did not increase the severity of bronchiolarinflammation on day 6 and inoculation of two month old adolescent guinea pigs with —4000pfu of RSV produced no evidence of acute bronchiolitis.These experiments established that this protocol of human RSV infection produced a self-limited histological bronchiolitis in juvenile guinea pigs which had features of human disease.Similarly to the previously described cotton rat (199) and mouse (198) models, RSVinoculation of guinea pigs did not produce clinical evidence of acute respiratory disease orgross pulmonary lesions. However, the guinea pig model better resembled human disease thaneither the cotton rat or mouse models in terms of the age susceptibility of animals to developbronchiolar inflammation. The details of these results are discussed below.6.1.1 CLINICAL-PATHOLOGICAL EVALUATION. Bronchiolar inflammation in theabsence of clinical disease has been observed in cotton rats inoculated intranasally with 104 pfuhuman RSV (199) and in young mice given 10 7 pfu (198) (in contrast to humans, clinical signsof respiratory disease "paradoxically" developed in older mice given human RSV and noexplanation was offered for this observation) 24 . The extent of bronchiolar inflammationappeared similar in the guinea pig, cotton rat and mouse, based on review of publishedphotomicrographs and comparison to results from another histological scoring system24 In human infants, the amount of RSV necessary to produce clinical signs of acutebronchiolitis is unknown because it is unethical to perform the experiment. However, aprevious study of adult human volunteers showed that intranasal inoculation with 10 2 . 7 pfuproduced upper respiratory infection in 16/16 subjects (255).97developed for cotton rats (156). In this thesis, one month old guinea pigs were used as thejuvenile group because the laboratory did not have breeding facilities. Consequently, the effectsof intranasal inoculation of human RSV into very young guinea pigs (analogous to infection ofhuman infants) remain unknown. In summary, human RSV produces similar degrees ofbronchiolar inflammation in the guinea pig, cotton rat and mouse, but the apparent predilectionof younger guinea pigs to develop significant bronchiolar inflammation more closelyapproximates human infections.Concerning adolescent guinea pigs, the absence of statistically significant bronchiolarinflammation on day 6 post-RSV inoculation may have been due to a "dilutional" effect fromthe instillation of a similar viral inoculum into larger lungs (in comparison to juvenile animals).However, the extent of RSV-induced airway inflammation was probably not dependent on theamount of virus instilled because a similar degree of bronchiolar inflammation was observed injuvenile guinea pigs given either —4000 pfu or —28000 pfu of human RSV (Appendix B). Todefinitively address this issue would require a separate study of adolescent guinea pigs giventhe larger amount of human RSV.Other possibilities to account for the apparent predilection of younger RSV-inoculatedguinea pigs to develop significant bronchiolar inflammation include structural characteristics ofairways conducive to viral-spread to the bronchioles, a higher intrinsic susceptibility of airwayepithelial cells to RSV infection or differences in immune response to RSV (section 2.6.1).Although examination for differences in airway geometry between juvenile and adolescentanimals would be difficult experimentally (149), the other two possibilities are amenable tostudy by the experimental approaches previously used for humans: comparison ofsusceptibility to RSV infection in different respiratory cell types in vitro (136), RSV serology(section 2.7.1) and assays of RSV-specific cytotoxic T-lymphocytes (section 2.7.2).However, the differences between the human and guinea pig immune systems (section 2.9)may limit the interpretation of experiments examining virus-specific immune responses.Collaborative studies (256) using acetylcholine challenge have shown that the RSV-98inoculated juvenile guinea pigs develop airway hyperresponsiveness by day 6 that resolves byday 14, in parallel with the histological lesions 25. These observations contrast with those of arecent study (99) of parainfluenza type I (Sendai) virus-infected rats challenged withmethacholine: in this study, the authors documented both acute and persistent airwayhyperresponsiveness in otherwise asymptomatic animals (unfortunately, histological evaluationto document the duration of airway inflammation was not performed). These differences in theduration of airway hyperresponsiveness may be attributable to intrinsic differences betweenRSV and Sendai virus, differences in host susceptibilities to viral infection (the rat is a naturalhost to Sendai virus) or immune responses to these viruses.The ability of RSV to produce concomitant histological and physiological abnormalities inthe setting of subclinical infection may be relevant to the majority of acute RSV lung infectionsin humans since hospitalization is rarely required (257). Should this be the case, theepidemiological implications are significant because RSV-induced airway inflammation andairway hyperresponsiveness (two hallmarks of asthma) may affect far more individuals thanpreviously suspected, as studies of acute RSV bronchiolitis tend to be limited to hospitalizedpatients (90, 91). An hypothesis generated from the experiments of this thesis is that, duringacute RSV lung infection, asymptomatic or mildly symptomatic children develop airwayinflammation and airway hyperresponsiveness. Furthermore, the study of Sendai virus-infected rats suggests that airway hyperresponsiveness might persist (99). While theidentification and recruitment of asymptomatic or mildly symptomatic RSV-infected childrenwould be difficult in practice, serial tissue sampling (e.g., nasopharyngeal aspirates to test forpresence of RSV) and lung function testing of these children could yield valuable information25 Increased airway inflammation and increased airway hyperresponsiveness wereobserved for the RSV-inoculated group as a whole. For a given animal, no histological featuresfrom the RSV bronchiolitis scoring system independently predicted the presence of coexistentairway hyperresponsiveness, in part due to the number of animals studied (data not shown).There was no augmentation of airway hyperresponsiveness to acetylcholine by using high doseRSV in juvenile guinea pigs or after low dose RSV given to adolescent animals. Additionalmorphometric analyses revealed no significant airway submucosal thickening or smoothmuscle thickening in RSV-inoculated guinea pigs on either days 6 or 14.99about the possible role of persistent RSV infection in the pathogenesis of persistent airwayhyperresponsiveness.6.1.2 HISTOLOGICAL SCORING SYSTEM. The histological scoring system for acutebronchiolitis was simple, reproducible and permitted distinction between RSV-specificbronchiolar inflammation from non-specific inflammation induced by uninfected cell culturesupernatant. The results from the scoring system also showed that the uninfected cell culturesupernatant did not induce significant airway inflammation because airway histological scoresin this group were similar to those of unmanipulated, historical control animals26. On day 6post-inoculation, the statistically significant epithelial necrosis and bronchiolar mononuclearcell infiltrates in juvenile RSV-inoculated guinea pigs confirmed the presence of acute lyticinfection resembling human disease (103). Curiously, despite extensive study of the hostimmune response to RSV in many species (section 2.7), there are no published reportsdescribing the characteristics of the lymphocytes comprising the bronchiolar wall infiltrates.Immunocytochemical characterization of these cells may provide information concerning thehost's local immune response to intrapulmonary RSV. Preliminary experiments usingantibodies which cross-react with guinea pig B-lymphocyte (anti-CDw75) and T-lymphocyte(anti-CD3) epitopes suggest that T-cells constitute that vast majority of infiltrating mononuclearcells on day 6 (data not shown). Further experiments are planned to further subclassify thesecells into CD4A- (helper-inducer phenotype) and CD8+ (cytotoxic-suppressor phenotype) withantibodies directed against guinea pig T-cell antigens. A predominance of CD8+ cells would beconsistent with a local cytotoxic T-cell response to RSV while a predominance of CD4+ cellswould be consistent with either humoral (i.e., TH2 cells) or cell-mediated (i.e., TH1 cells)immunity (258).26 In general, the unmanipulated historical control group had "near-normal" airwayhistology. Some focally striking bronchiolar inflammation was observed in a few of theanimals inoculated with uninfected cell culture supernatant; however, these findings were notstatistically significant for the group as a whole.100A second unresolved issue concerning the mononuclear cell bronchiolar wall infiltrates isthe specificity of lymphocytes toward RSV. One method to screen for RSV-specificlymphocyte memory is the in vitro 3 H-thymidine incorporation assay that documentslymphocyte proliferation induced by microbial crude antigen preparations (259). This utility ofthis assay has been previously shown in several rheumatological disorders, in which T-lymphocytes aspirated from inflamed joints proliferate (i.e., incorporate 3H-thymidine) in amore specific fashion (following exposure to a panel of crude microbial antigen preparations)than the circulating lymphocytes from the same patients (260). If lymphocytes isolated from theguinea pig lung are exposed to a panel of crude microbial antigen preparations in a similarmanner, one may screen for RSV-specific lymphocyte proliferation from the inflamed lung.Limitations of this technique include the loss of morphological information regarding thelocation(s) of the responding lymphocytes within the lung (e.g., bronchiolar wall vs. BALT)and the majority of proliferating lymphocytes having a CD4+ phenotype consistent withmemory T-cells rather than cytotoxic T-cells (261). The presence of RSV-specific cytotoxic T-lymphocytes within the lungs of RSV-inoculated guinea pigs may be assessed by in vitrochromium release assays (174-178).The presence of statistically significant bronchiolar wall PMN infiltrates reproduced recentobservations of cow lungs infected with bovine RSV (224) and extended previousobservations of increased neutrophil adherence to RSV-infected HEp-2 cells induced by RSV-specific antibody in vitro (242). Circulating anti-RSV antibodies may play a role in therecruitment of neutrophils to bronchioles in vivo because the guinea pig is known to produceRSV-specific antibodies following intranasal RSV inoculation (164, 182). Testing thispossibility would require serology on day 6 post-inoculation, but the interpretation of resultswould be limited by species differences in humoral immunity (section 2.9).Airway wall edema, BALT hyperplasia and goblet cell metaplasia were not statisticallysignificant between RSV-inoculated guinea pigs and controls. While the lack of significantairway wall edema might be related to species differences in the host inflammatory response to101RSV, another possibility is that extensive airway wall edema may be a feature of fatal cases ofhuman acute RSV bronchiolitis (since histological descriptions of human airway lesions havebeen primarily derived from postmortem specimens) (101-104). Although the experiments ofthis thesis confirmed the existence of BALT in guinea pigs (similar to rabbits, rats and humans(262)), the lack of BALT hyperplasia following RSV may have been due to several factors.For example, BALT may not be an important constituent of the pulmonary immune system, asrecently suggested for human BALT by Pabst (263). Secondly, if cell-mediated immunity is ofgreater importance than humoral immunity in the host response to acute RSV lung infection,then BALT hyperplasia might not occur since (in humans) BALT consists mostly of B-lymphocytes associated with secretory IgA immunity (264). A third possibility is thatlymphocyte traffic through BALT (264) to the bronchioles achieves an equilibrium during acuteRSV infection that does not result in detectable BALT hyperplasia. Immunocytochemicallycharacterizing the lymphocytes (see above discussion for bronchiolar wall lymphocyticinfiltrates) within guinea pig BALT is a first step toward clarifying any role for BALT in theguinea pig's immune response to RSV.The lack of significant goblet cell metaplasia of the bronchiolar epithelium may have beendue to RSV inducing insufficient epithelial cell lysis to stimulate metaplastic repair.Alternatively, if normal guinea pig bronchioles contain a greater proportion of goblet cells thanhuman airways, subtle goblet cell metaplasia might have been obscured. Another possibility isthat, in contrast to humans (243), goblet cell metaplasia is not an important epithelial repairmechanism in guinea pig airways. These possibilities could be addressed by counting gobletcells in guinea pig airways which have undergone extensive experimental epithelial cell injury(e.g., acid injury), comparing the number of goblet cells in normal guinea pig vs. humanairways, and possibly re-scoring lung sections for alternate mechanisms of epithelial repairsuch as squamous cell metaplasiar.27^Squamous cell metaplasia as mechanism of airway epithelial repair following acute lyticRSV infection is not favored because inspection of histological sections from RSV-inoculated1026.2 NATURAL HISTORY OF INTRAPULMONARY RSV. The combination of viral culture,TEM, immunohistochemistry and RT-PCR provided new insights into the natural history ofintrapulmonary virus (within both lungs) up to 125 days from the time of acute infection—approximately 25% of the expected life span of a guinea pig (209). The details of these studiesare summarized below.6.2.1 VIRAL CULTURE. The ability to culture replicating RSV from the lung in the majorityof animals during acute infection has also been reported in the cotton rat (199) and the mouse(198) models. In particular, on day 6 post-inoculation, a mean ± SD of 1.7 ± 0.3 x 103 pfuRSV/g wet weight fresh lung in guinea pigs was similar to levels previously reported for cottonrats and mice. An apparent discrepancy between the guinea pig and the other two species wasthe maximum period of ability to isolate virus post-inoculation: in the guinea pig, RSV CPEwas observed in 2/10 RSV-inoculated animals at day 14, while virus was only isolated up to 7days post-infection in the cotton rat and 8 days in the mouse. However, there was a distinctpossibility of false negative cultures in both the cotton rat and mouse models because in bothmodels, excised lungs were frozen and stored for days to weeks prior to processing for cultureon HEp-2 cells. In contrast, fresh lungs were used in this thesis because freezing and thawingmay inactivate RSV (265)28. Another important difference in the viral culture protocol of thisthesis was the weekly passaging of samples for up to one month before being callednegative29: HEp-2 cell monolayers were examined only at 5 days after plating of lung digestsin the cotton rat model and after 4 days in the mouse model. In summary, the ability to cultureintrapulmonary RSV from the guinea pig substantially longer than the cotton rat or the mouseanimals did not reveal evidence of squamous cell metaplasia. This observation was the rationalefor not including squamous cell metaplasia as part of the histological scoring system.28^If freezing and thawing caused viral degradation in these studies, then both cotton ratsand mice may have actually had greater amounts of intrapulmonary RSV than observed inguinea pigs on day 6.29^One of the two positive RSV cultures in the day 14 study occurred during the secondpassage of HEp-2 cells.103may have reflected differences in viral culture methods rather than intrinsic species differencesin the susceptibility to longer term human RSV infection.The inability to culture virus in the day 60 and day 125 studies was probably a consequenceof insufficient release of RSV from persistently infected alveolar macrophages to infect HEp-2cells in vitro. This explanation is favored because infected alveolar macrophages have apropensity to retain RSV (136) and the immunohistochemistry studies of this thesis haveshown evidence of RSV only within alveolar macrophages on day 60. A potential mechanismfor the persistent, non-lytic infection of alveolar macrophages is through the production of so-called "defective interfering particles" (237) of RSV following acute infection. Defectiveinterfering particles, although missing portions of the viral genome required for replication,remain capable of producing viral proteins that may be deleterious (without necessarily beinglethal) to the infected cell.The endpoint of viral culture was the documentation of syncytial formation in HEp-2 cellmonolayers, the characteristic CPE of human RSV. A potential confounding factor was that"pathogen-free" guinea pigs (214) were not used in the experiments of this thesis such that theobserved CPE may have been due to other viruses that form syncytia in HEp-2 cells, e.g.,parainfluenza virus or mumps virus (135). However, pulmonary infection by eitherparainfluenza or mumps viruses was not likely because the guinea pig is not a natural host toeither virus (213) and no syncytia were observed in any HEp-2 cell cultures from controlanimals. In addition, precautions were taken to decrease the likelihood of undesired infectionsin guinea pigs (section 2.9). The room housing RSV-inoculated guinea pigs was reservedexclusively for the experiments of this thesis and was cleaned prior to the delivery of animals.Furthermore, isolation procedures (gowns, hats, masks, gloves and shoe covers) wereemployed to protect both guinea pigs and investigators against undesired infections.6.2.2 TRANSMISSION ELECTRON MICROSCOPY. As discussed in section 3.3, the lowsensitivity of TEM in the diagnosis of viral infections resulted in TEM being used as an104ancillary technique in this thesis30. Particular effort was made to exclude the possibility of otherstructures that may have been confused with RSV nucleocapsid or assembled virus. The mainutility of TEM was the confirmation that viral replication and assembly took place withininfected bronchial epithelial cells, as shown by the presence of free intracytoplasmicnucleocapsids and assembled virus particles budding from the cell membrane. The TEM resultsindicate that positive viral cultures, immunohistochemical staining and RT-PCR were not solelyattributable to free virus left over from the inoculation procedure. The inability to findunequivocal virus-related particles by TEM within bronchiolar epithelial cells, type I and type IIpneumocytes or alveolar macrophages has also been reported in cows infected with bovineRSV and in that instance was attributed to sampling error (222).Immunogold electron microscopy (267, 268) has advanced the capability of TEM toidentify the presence of specific antigens in ultrathin sections; however, no protocol usingimmunogold electron microscopy for the identification of RSV antigens has been published todate. The development of a suitable protocol would permit the unequivocal identification ofvirus-related structures that might otherwise have been missed or misinterpreted.6.2.3 IMMUNOHISTOCHEMISTRY. During the acute phase of RSV lung infection on day6, the polyclonal anti-RSV antibody detected RSV antigens in paraformaldehyde-fixed,paraffin-embedded midsagittal sections of guinea pig lung with 50-70% sensitivity and 100%specificity compared to viral culture. Both the sensitivity and specificity of this antibody werecomparable to a previous report that documented RSV antigens in formalin-fixed, paraffinembedded sections of human autopsy lungs (104).In adolescent guinea pigs, the immunohistochemical demonstration of RSV antigens in5/10 infected animals confirmed that RSV reached the peripheral airways following intranasal30 Neither the cotton rat nor mouse models of RSV lung infection used TEM to documentintrapulmonary viral particles. The difficulty in finding virus-related particles in the guinea piglung prompted a review of Index Medicus: since 1973, there have been only 6 reports of TEMdocumenting intrapulmonary pulmonary RSV in ultrathin sections of lung (104, 191, 222-224,266). These reports have primarily dealt with the ultrastructural identification of bovine RSV incows or sheep.105inoculation. Therefore, the lack of statistically significant airway inflammation in adolescentguinea pigs was not due to absence of virus. As discussed in section 4.14, no attempts weremade to quantify the number of positive cells with airway inflammation, etc. because theimmunohistochemistry protocol was designed be to highly specific rather than highlysensitive31 . Concerning adolescent guinea pigs, if the lack of significant airway inflammationduring acute RSV infection resulted from a lower "burden" of intrapulmonary replicating virus,then quantitative viral plaque assays may have revealed differences. Unfortunately, no viralcultures were completed on the lung digests of adolescent guinea pigs because of inadvertentcontamination of the cell culture medium by yeast.The major new findings by immunohistochemistry related to the changing distribution ofRSV antigens in different lung cell types over time. On day 6, RSV antigens were identifiedprimarily in airway epithelial cells (consistent with the cotton rat and mouse models) but by day14 were present mostly within alveolar macrophages. On day 60, RSV antigens were identifiedexclusively within alveolar macrophages. These results suggest that persistent non-lytic RSVinfection of alveolar macrophages may occur in vivo and extend recent observations of RSV-infected human alveolar macrophages in vitro remaining viable, with little shedding of RSVinto culture supernatants (136). However, these observations do not entirely exclude thepossibility of RSV infection of other lung cell types (undetectable by immunohistochemistry),with scavenging of free virus by alveolar macrophages. Alternatively, repeated episodes ofRSV aspiration from chronic or recurrent upper respiratory tract infections may have beenresponsible for the presence of RSV antigens within alveolar macrophages. This mechanism isnot favored because repeated aspiration of RSV from the upper respiratory tract wouldprobably have also resulted in infection of other susceptible lung cell types such as airway31 Western blotting was attempted to quantify RSV antigens in protein extracts of frozenguinea pig lung. However, multiple runs were negative (data not shown), probably because theamount of virus was small compared to the amount total lung protein, thus rendering detectionof RSV beyond the sensitivity of the technique (269). A positive signal by Western blottingrequires at least 20 pg of target protein—equivalent to over 10 7 nucleocapsid molecules.106epithelial cells (136). Presumably anti-RSV antibody would have detected RSV antigens in celltypes in addition to alveolar macrophages.The lack of statistically significant bronchiolar inflammation on days 14 and 60, despitedocumentation of intrapulmonary RSV antigens, suggests that the virus itself is insufficient tostimulate chronic airway inflammation. However, if persistent RSV infection occurs in humanalveolar macrophages in vivo, one may speculate that virus-induced alterations of macrophagefunction (section 1.3.5) could play a role in the pathogenesis of chronic airway inflammation tonon-specific inhaled environmental agents (115): for example, the lack of "down regulation" ofTH2 cells by alveolar macrophages in atopic asthma. The possibility of virus-infectedmacrophages permitting the production of allergen-induced airway inflammation was notexamined in this thesis because extensive efforts were made to prevent guinea pigs from beingexposed to non-specific environmental agents.Finally, the documentation of RSV antigens within alveolar macrophages on day 60provides new information regarding the fate of these cells. Previous studies have suggestedthat alveolar macrophages are cleared from the lung at the rate of 3.5%/day—the implicationbeing that a population of alveolar macrophages would be completely cleared from the lungwithin one month (270). In contrast, Brain et al. have speculated that alveolar macrophagesundergo several possible fates, from clearance up the tracheobronchial tree to migration into thepulmonary interstitium and even migration into regional pulmonary lymph nodes (271). Sincealveolar macrophages are not believed to undergo appreciable mitotic activity in vivo, theexperiments of this thesis suggest that the clearance of RSV-infected alveolar macrophages isdelayed from the peripheral lung. The possibility of RSV infection of pulmonary lymph nodeswas not examined; regarding other pulmonary lymphoid tissue, there was no evidence of RSVantigens within BALT.6.2.4 RT-PCR. The RT-PCR methodology developed for this thesis introduced three newaspects to the investigation of pulmonary viral disease. First, this protocol represented the firstknown attempt to document pulmonary RSV infection using lung specimens rather than107exfoliated cells from the upper respiratory tract. Secondly, the sensitivity and specificity of thisRT-PCR protocol were evaluated such that one could be confident that a positive signal wouldresult if a sample contained at least 5 copies of target cDNA after reverse transcription. Thirdly,RT-PCR not only revealed the presence of RSV genomic RNA in the guinea pig lung duringacute RSV infection but was also positive in 3/4 guinea pigs studied 125 days post-inoculation—twice as long as previously documented in immunodeficient humans and fourtimes as long as previously documented in immunocompetent humans. The long-termpersistence of RSV within the lung is a novel finding which supports the working hypothesisof this thesis that RSV (or at least a portion of its genome) may persist within the lungfollowing resolution of acute bronchiolitis.In samples of RSV-infected HEp-2 cells, the presence of bands in addition to the main 410by PCR product may have represented amplified non-specific human DNA or single-strandedDNA from the viral PCR amplification. Two of the three bands were negative onautoradiography and a viral origin for these bands is essentially excluded. One band was faintlypositive by autoradiography and likely represented single-stranded DNA from the viral PCRproduct (272). Importantly, the absence of extra bands from PCR products in samples fromRSV-infected guinea pig lungs confirmed the specificity of the RT-PCR method in thissituation.The RT-PCR technique did not quantify the amount of RSV genomic RNA within a givensample: to do this rigorously would have required concomitant reverse transcription andamplification of a guinea pig mRNA target sequence with known cellular levels. This was notpursued in this thesis because the additional variables inherent in quantitative RNA PCR (e.g.,the efficiency of the reverse transcription step; factors influencing expression of the mRNAsequence chosen to normalize the data) are yet to be resolved in other, more establishedsystems (273-275)32 .32^A relatively easy approach for quantitative RT-PCR for RSV genomic RNA is to usespecimens of RSV-infected HEp-2 cells containing known amounts of plaque forming units.108Finally, the RT-PCR protocol developed for this thesis did not localize RSV genomic RNAwithin the guinea pig lung. Although in situ hybridization is a technique that wouldtheoretically permit cellular localization RSV RNA, a sensitivity of only 50% has been reportedfor conventional in situ hybridization in the detection of RSV nucleic acid in culture-positivenasopharyngeal secretions (219, 276). The low sensitivity of in situ hybridization is related tothe requirement for a minimum of 10-20 copies of target RNA per infected cell for a positivesignal and the propensity of RNA to be digested by ubiquitous intracellular and exogenousRNases (277). Repeated attempts to demonstrate RSV by in situ hybridization (based on themethod of Murphy et al. (278) for the identification of RSV RNA in cotton rat lung sections)were uniformly negative for sections of RSV-infected guinea pig lung despite consistentpositivity in sections of RSV-infected HEp-2 cells (data not shown). The negative results mayhave been related to low copy numbers of RSV within infected cells and/or the apparentpropensity for RNA within the guinea pig lung to degrade easily33.Recently, protocols combining in situ hybridization with PCR have successfully been usedto detect human papilloma viral DNA in formalin-fixed, paraffin embedded tissue sections(279) and human immunodeficiency virus proviral RNA in peripheral blood monocytes (280).These results indicate that morphological correlation to PCR products is now possible.However, the combined use of in situ hybridization and RT-PCR in tissue sections has notbeen reported and a further technical advance would therefore be required to apply thiscombined approach to the guinea pig model of RSV lung infection.However, the relationship between plaque forming units and the corresponding copy numbersof RSV genomic RNA would still have to be established.33^Recall that there was extensive degradation of total cellular RNA extracted in the firstthree day 6 studies, the day 14 study and in the day 60 study. Two principal investigators inthe U.B.C. Pulmonary Research Laboratory (Drs. T. Bai and G. Bondy) have independentlyexperienced problems in extracting intact RNA from the guinea pig lung, despite the use ofprotocols intended for RNA extraction in organs rich in endogenous RNase such as pancreas(244). A literature review yielded no information concerning this apparent tendency for RNA inthe guinea pig lung to degrade easily.1096.3 CONCLUDING REMARKS. The experimental results presented in this thesis support theworking hypothesis that human RSV may chronically persist within the lung followingresolution of acute bronchiolitis, although the presence of RSV per se was insufficient toproduce chronic airway inflammation. In the future the guinea pig model could be extended toinclude studies of: (a) re-infections with RSV; (b) genetic factors influencing the phenomenaobserved in primary RSV lung infection; (c) the specificity of the host immune response toRSV during acute lung infection; (d) the effects of non-specific inhalational irritants on alveolarmacrophage function, airway inflammation and airway hyperresponsiveness in guinea pigswith coexistent RSV lung infection. The guinea pig model is now sufficiently well establishedto examine whether second or third RSV infections produce progressively milder disease (as isapparently the case for humans). Concerning the role of genetic factors of guinea pigs inprimary RSV lung infection, the study of airway histology and lung mechanics using inbredanimals such as the BE strain (this strain mounts a poor immune response to most antigens) orthe PCA strain (a model of cutaneous anaphylaxis) (206) may yield new informationconcerning the relative roles of RSV itself vs. the host immune response to RSV in thepathogenesis of acute bronchiolitis. Unfortunately, inbred strains of guinea pigs tend to breedpoorly and thus may not be readily available.Concerning whether the host immune response to RSV is virus-specific in guinea pigs, athreefold approach testing for RSV-specific cytotoxic T-lymphocytes, RSV-specific antibodiesand RSV-specific T-cell memory (the 3H-thymidine incorporation assay) may be used. Incontrast to previous reports, the evaluation of guinea pig specimens on day 6 post-inoculationwill provide a new perspective about the nature and extent of virus-specific immunity duringthe acute phase of primary RSV lung infection, when bronchiolar inflammation is maximal.The possible deleterious effects of RSV on alveolar macrophage function may beinvestigated as previously described for parainfluenza virus (114). Of potentially greaterinterest is testing the extent of airway inflammation and airway hyperresponsiveness in RSV-infected guinea pigs following challenge with such non-specific environmental agents as110cigarette smoke or inhaled allergens because the combination of RSV infection with non-specific environmental agents may be synergistic (113). Since it is not feasible to perform suchstudies on humans, the guinea pig model is appropriate for such investigations.Perhaps the most exciting application of the methods and results of this thesis is a rationaleto investigate human asthmatic patients for evidence of persistent RSV lung infection. Inparticular, the immunohistochemistry and RT-PCR techniques could be used to study humanalveolar macrophages obtained by bronchoalveolar lavage (BAL) (281, 282). A less invasiveapproach would involve testing cells obtained from nasopharyngeal aspirates for evidence ofpersistent RSV upper respiratory tract infection in children with asthma. In contrast to theguinea pig, it is plausible that repeated episodes of aspiration of RSV from the upperrespiratory tract (either as recurrent or relapsing infections) may play a role in the pathogenesisof chronic airway inflammation in childhood asthma because humans are natural hosts to RSV.Although sampling of nasopharyngeal aspirates is less direct than sampling lung cells by BAL,the technical ease and high patient tolerance of nasopharyngeal sampling would facilitate therapid acquisition of a sufficient number of specimens to permit epidemiologically relevantinterpretations, e.g., the prevalence of RSV in the upper respiratory tract in asthmatic patientsvs. normal subjects.In conclusion, investigators have long postulated a role of viral lung infections in thepathogenesis of asthma; however, it is only with the advent of improved, highly sensitive andspecific techniques such as RT-PCR that this possibility can realistically begin to be examined.This thesis used the guinea pig to show how the combination of "classical" and contemporaryexperimental approaches may yield intriguing new information about the natural history ofintrapulmonary virus from the time of known primary infection. The application of thesetechniques to the study of human asthma is now feasible and may culminate in an improvedunderstanding of the role of pulmonary viral infections in the pathogenesis of asthma.111BIBLIOGRAPHY1. Wigle DT. Prevalence of selected chronic diseases in Canada, 1978-79. Chr Dis Can1982; 3: 8-9.2. Mao Y, Semenciw R, Morrison H, MacWilliam L, Davies J, Wigle D. Increased ratesof illness and death from asthma in Canada. Can Med Assoc J 1987; 137: 620-624.3. Wilkins K, Mao Y. Trends in rates of hospitalization and death from asthma amongchildren and young adults in Canada during the 1980s. Can Med Assoc J 1992 (inpress).4. Buist AS. Report of a symposium: Asthma mortality: trends and determinants. Am RevResp Dis 1987; 136: 1037-1039.5. Weiss KB, Gergen PJ, Hodgson TA. An economic evaluation of asthma in the UnitedStates. New Engl J Med 1992; 326: 862-866.6. Friel JP, ed. Dorland's Illustrated Medical Dictionary, 25th edition.  Philadelphia, PA:W.B. Saunders Company, 1974.7. Matthys H. Definition and assessment of asthma. Lung 1990; 168 (Suppl.): 51-56.8. Osler W. Bronchial asthma. In: The Principles and Practice of Medicine, 1st edition. New York, NY: D. Appleton and Company, 1892: 497-501.9. Ciba Guest Symposium. Terminology, definitions and classification of chronicpulmonary emphysema and related conditions. Thorax 1959; 14: 286-299.10. Roitt IM. Atopic allergy. In: Essential Immunology, 7th edition.  Oxford, UK:Blackwell Scientific Publications, 1991: 255-260.11. Maimonides M. Treatise on asthma. (Munter S, ed.) Philadelphia, PA: J.B. LipponcottCompany, 1963.12. Cookson WOCM, Sharp A, Faux JA, Hopkin JM. Linkage between immunoglobulinE responses underlying asthma and rhinitis and chromosome 1 lq. Lancet 1989; 1:1292-1295.13. Lympany P, Welsh K, MacCochrane G, Kemeny DM, Lee TH. Genetic analysis usingDNA polymorphism of the linkage between chromosome 11q13 and atopy andbronchial hyperresponsiveness to methacholine. J Allergy Clin Immunol 1992; 89:619-628.14. Porter R, Birch J., eds. Identification of asthma. Ciba Foundation Study Group 1971;38: 35.15. American Thoracic Society. Standards for the diagnosis and care of patients withchronic obstructive pulmonary disease (COPD) and asthma. Am Rev Resp Dis 1987;136: 225-244.16. Holgate ST. Reflections by the chairman. Am Rev Resp Dis 1991; 143: 1175-1176.11217. Samet JM. Epidemiologic approaches for the identification of asthma. Chest 1987; 91(Suppl.): 74S-75S.18. Bates DV, Macklem PT, Christie RV. Spasmodic asthma. In: Bates DV, Macklem PT,Christie RV, eds. Respiratory function in disease: an introduction to the integratedstudy of the lung, 2nd edition.  Philadelphia, PA: W.B. Saunders Company, 1971:111-132.19. West JB. Lung function tests and what they mean. In: Pulmonary Pathophysiology,3rd edition.  Baltimore, MD: Williams and Wilkins, 1987: 1-55.20. Editorial. Airflow limitation—reversible or irreversible? Lancet 1988; 1: 26-27.21. Sterk PJ, Bel EH. The shape of the dose-response curve to inhaled bronchoconstrictoragents in asthma and in chronic obstructive pulmonary disease. Am Rev Resp Dis1991; 143: 1433-1437.22. Woolcock AJ, Anderson SD, Peat JK, et al. Characteristics of bronchialhyperresponsiveness in chronic obstructive pulmonary disease and in asthma. Am RevResp Dis 1991; 143: 1438-1443.23. Holgate ST, Wilson JR, Howarth PH. New insights into airway inflammation byendobronchial biopsy. Am Rev Resp Dis 1992; 145: S2-S6.24. Vincenc KS, Black JL, Yan K, Armour CL, Donnelly PD, Woolcock AJ. Acomparison of in vivo and in vitro responses to histamine in human airways. Am RevResp Dis 1983; 128: 875-879.25. Armour CL, Lazar NM, Schellenberg RR, et al. A comparison of in vivo and in vitrohuman airway reactivity to histamine. Am Rev Resp Dis 1984; 129: 907-910.26. Cerrina J, Ladurie ML, Labat C, Raffestin B, Bayol A, Brink C. Comparison ofhuman bronchial muscle responses to histamine in vivo with histamine andisoproterenol in vitro. Am Rev Resp Dis 1986; 134: 51-61.27. James AL, Hogg JC, Dunn LA, Pare PD. The use of internal perimeter to compareairway size and to calculate smooth muscle shortening. Am Rev Resp Dis 1988;138:136-139.28. Jongejan RC, de Jongste JC, Kerrebijn KF. The increased responsiveness to inhaledmethachoine in asthma: combination of causative factors. Am Rev Resp Dis 1991; 143:1194-1196.29. Pare PD, Wiggs BR, James A, Hogg JC, Bosken C. The comparative mechanics andmorphology of airways in asthma and in chronic obstructive pulmonary disease. AmRev Resp Dis 1991; 143: 1189-1193.30. Hogg JC, James AL, Pare PD. Evidence for inflammation in asthma. Am Rev RespDis 1991; 143: S39-S42.31.^Nadel JA. Structure and function relationship in the airways. Medicinia thoracalis 1965;22: 231-243.11332. Huber HL, Koessler KK. The pathology of bronchial asthma. Arch Int Med 1922; 30:689-760.33. Unger L. The pathology of bronchial asthma. Southern Med J 1945; 38: 513-522.34. Bullen SS. Correlation of clinical and autopsy findings in 176 cases of asthma. J AllergClin Immunol 1952; 23: 193-203.35. Dunnill MS. The pathology of asthma with special reference to changes in the bronchialmucosa. J Clin Path 1960; 13: 27-33.36. Messer J, Peters GA, Bennet WA. Cause of death and pathological findings in 304cases of bronchial asthma. Dis Chest 1960; 38: 616-624.37. Takizawa T, Thurlbeck WM. Muscle and mucous gland size in the major bronchi ofpatients with chronic bronchitis, asthma and asthmatic bronchitis. Am Rev Resp Dis1971; 104: 331-336.38. Cutz E, Levison H, Cooper DM. Ultrastructure of airways in children with asthma.Histopathology 1978; 2: 407-421.39. Salvato G. Some histological changes in chronic bronchitis and asthma. Thorax 1968;23: 168-172.40. Barnes PJ. New concepts in the pathogenesis of bronchial hyperresponsiveness andasthma. J Allergy Clin Immunol 1989; 83: 1013-1026.41. Austin KF. Airway inflammation in asthma. The proceedings of a roundtablediscussion, July 1990. Am Rev Resp Dis 1992; 145: S1-S58.42. Weiss ST, Sparrow D, eds. Airway Responsiveness and Atopy in the Development ofthe Obstructive Airways Diseases. New York NY: Raven Press, 1989.43. Sears MR, Herbison GP, Holdaway MD, Heweitt CJ, Flannery EM, Silva PA. Therelative risks of sensitivity to grass pollen, house duse mite and cat dander in thedevelopment of childhood asthma. Clin Exp Allergy 1989; 19: 419-424.44. Sporik R, Holgate ST, Platts-Mills TAE, Cogswell JJ. Exposure to house-dust miteallergen (Der pi) and the development of asthma in childhood. New Engl J Med 1990;323: 502-507.45. Kay AB, Henson PM, Hunninghake GW, Irwin C, Lichtenstein LM, Nadel JA.Cellular Mechanisms. In: Holgate ST, ed. The Role of Inflammatory Processes in Airway Hyperresponsiveness. Oxford, UK: Blackwell Scientific Publications, 1989:151-178.46. Mosmann TR, Cherwinski H, Bond MW, Gieldlin MA, Coffman RL. Two types ofmurine helper T cell clone. I. Definition according to profiles of lymphokine activitiesand secreted proteins. J Immunol 1986; 136: 2348-2357.47. Mosmann TR, Coffman RL. TH1 and TH2 cells: different patterns of lymphokinesecretion lead to different functional properties. Annual Rev Immunol 1989; 7: 145-173.11448. Roitt IM. The acquired immune response. II. Production of effectors. In: Essential Immunology, 7th edition. Oxford, UK: Blackwell Scientific Publications, 1991: 129-151.49. Robinson DS, Hamid Q, Ying S, et al. Predominant TH2-like bronchoalveolar T cellpopulation in atopic asthma. New Engl J Med 1992; 326: 298-304.50. Aubas P. Crosso B, Goddard P, Michel FB, Clot J. Depressed suppressor cell activityby PAM in bronchial asthma. Am Rev Resp Dis 1984; 130: 875-878.51. Holt PG. Down regulation of immune responses in the lower respiratory tract: the roleof alveolar macrophages. Clin Exp Immunol 1986; 63: 261-270.52. Thepen T, van Rooifen N, Kraal G. alveolar macrophage elimination in vivo isassociated with an increase in pulmonary immune response in mice. J Exp Med 1989;170: 499-509.53. Roitt TM. The acquired immune response. I. Consequences of antigen recognition. In:Essential Immunology, 7th edition.  Oxford, UK: Blackwell Scientific Publications,1991: 105-128.54. Hogg JC. Persistent and latent viral infections in the pathology of asthma. Am RevResp Dis 1992; 145: S7-S9.55. Kauffman RS, Fields BN. Pathogenesis of viral infections. In: Fields BN, Knipe DM,Chanock RM, Melnick JL, Roizman B, Shope RE, eds. Fields Virology. New York,NY: Raven Press, 1985: 163-169.56. Garcia-Blanco M, Cullen BR. Molecular basis of latency in pathogenic human viruses.Science 1991; 254: 815-820.57. Snell NJC. Economic and long-term benefits of ribavirin therapy on respiratorysyncytial virus infection. Lung 1990; 168 (Suppl.): 422-429.58. Moll HH. The action of parasympathetic-mimetic drugs in asthma. Quart J Med 1940;9: 229-240.59. Heise HA. The role of streptococci in bronchial asthma. Ann Allergy 1949; 7: 250-249.60. Curry JJ. Pathogenesis of bronchial asthma. Med Clin N America 1950; 34: 1829-1838.61. Dutton LO. Mold fungi in the etiology of respiratory allergic diseases. Ann Allergy1949; 7: 585-591.62. Chobot R, Uvitsky IH, Dundy H. The relationship of the etiologic factors in asthma ininfants and young children. J Allergy 1951; 22: 106-110.63. Spain WC, Fontana VJ. The role of infection in asthma. New York State J Med 1955;55: 3096-3102.64.^Salk JE. Reaction to influenza virus vaccines. J Am Med Assoc 1947; 134: 393.11565. Brown CC, Coleman MB, Alley RD, Stranahan A, Stuart-Harris CH. Chronicbronchitis and emphysema: significance of the bacterial flora in the sputum. Am J Med1954; 17: 478-484.66. WeiIle FL, Gohd RS. The virus theory of nasal polyp etiology and its practicalapplications. Ann Otol Rhinol Laryngol 1956; 65: 443-449.67. Frick OL, German DF, Mills J. Development of allergy in children. I. Association withvirus infections. J Allergy Clin Immunol 1979; 63: 228-241.68. Busse WW. The contribution of viral respiratory infections to the pathogenesis ofairway hyperreactivity. Chest 1988; 93: 1076-1082.69. Busse WW, Vrtis RF, Dick EC. The role of viral infections in intrinsic asthma:activation of neutrophil inflammation. Agents Actions 1989; 28 (Suppl.): 41-56.70. Berkovich S, Millian SJ, Snyder RD. The association of viral and mycoplasmainfections with recurrence of wheezing in the asthmatic child. Ann Allergy 1976; 28:43-49.71. Minor TE, Dick EC, DeMeo AN, Ouellette JJ, Cohen M, Reed CE. Viruses asprecipitants of asthmatic attacks in children. J Am Med Assoc 1974; 227: 292-298.72. Minor TE, Dick EC, Baker JW, Ouelette JJ, Cohen M, Reed CE. Rhinovirus andinfluenza type A infections as precipitants of asthma. Am Rev Resp Dis 1976; 113:149-153.73. Freeman GL, Todd RH. The role of allergy in viral respiratory tract infections. Am JDis Child 1962; 104: 330-334.74. McIntosh K, Ellis EF, Hoffman LS, Lybass TG, Eller JJ, Fulgniti VA. The associationof viral and bacterial respiratory infections with exacerbations of wheezing in youngasthmatic children. J Pediatr 1973; 83: 578-590.75. Welliver RC, Sun M, Rinaldo D, Ogra PL. Predictive value of respiratory syncytialvirus-specific IgE responses for recurrent wheezing following bronchiolitis. J Pediatr1986; 109: 776-780.76. Halonen M, Stern D, Wright A, Taussig L, Martinez FD. Total and specific IgE levelsduring and after lower respiratory illness in infancy and early childhood. Am Rev RespDis 1992; 145: A434.77. Wittig HJ, Cranford NJ, Glaser J. The relationship between bronchiolitis andchildhood asthma. J Allergy 1959; 30:19-23.78. Rooney JC, Williams HE. The relationship between proved viral bronchiolitis andsubsequent wheezing. J Pediatr 1971; 79: 744-747.79. Pullan CR, Hey EN. Wheezing, asthma, and pulmonary dysfunction 10 years afterinfection with respiratory syncytial virus in infancy. Br Med J 1982; 284: 1665-1669.80. Hall CB, Hall WJ, Gala CL, MaGill FB, Leddy JP. Long-term prospective study inchildren after respiratory syncytial virus infection. J Pediatr 1984; 105: 358-364.11681. Sly PD, Hibbert ME. Childhood asthma following hospitalization with acute viralbronchiolitis in infancy. Pediatr Pulmonol 1989; 7: 153-158.82. Nicolai T, Pohl A. Acute viral bronchiolitis in infancy: epidemiology and management.Lung 1990; 168 (Suppl.): 396-405.83. Hall CB. Respiratory syncytial virus. In: Feigin RD, Cherry JD, eds. Textbook ofPediatric Infectious Diseases, 2nd edition.  Philadelphia PA: W.B. Saunders Company,1987: 1653-1676.84. Hall CB, Douglas RGJr, Geiman JM. Resiratory syncytial virus infections in infants:quantitation and duration of shedding. J Pediatr 1976; 89: 11-15.85. Hall CB, Powell KR, MacDonald NE, et al. Respiratory syncytial viral infection inchildren with compromised immune function. New Engl J Med 1986; 315: 77-81.86. High RH. Bronchiolitis. Pediatr Clin N Amer 1957; 1: 183-191.87. Hogg JC. Asthma as a bronchiolitis. Sem Resp Med 1992; 13: 114-118.88. Lemanske RFJr, Dick EC, Swenson CA, Vrtis RF, Busse WW. Rhinovirus upperrespiratory infection increases airway hyperreactivity and late asthmatic reactions. JClin Invest 1989; 83: 1-10.89. Busse WW, Swenson CA. The relationship between plasma histamine concentrationsand bronchial obstruction to antigen challenge in allergic rhinitis. J Allergy ClinImmunol 1989; 84: 658-666.90. Seidenberg J, Masters IB, Hudson I, Olinsky A, Phelan PD. Disturbance in respiratorymechanics in infants with bronchiolitis. Thorax 1989; 44: 660-667.91. Smith DW, Rector DM, Derish MT, Frankel LR, Ariagno RL. Pulmonary functiontesting in infants with respiratory syncytial virus bronchiolitis requiring mechanicalventilation. Pediatr Infect Dis J 1990; 9: 5108-5111.92. Caswell SJ, Thomson AH, Ashmore SP, Beardsmore CS, Simpson H. Latentsensitisation to respiratory syncytial virus during acute bronchiolitis and lung functionafter recovery. Arch Dis Childhood 1990; 65: 946-952.93. Laitinen LA, Elkin RB, Empey DW, Jacobs L, Mills J, Nadel JA. Bronchialhyperresponsiveness in normal subjects during attenuated influenza virus infection. AmRev Resp Dis 1991; 143: 358-361.94. Al-Darraji AM, Cutlip RC, Lehmkuhl HD, Graham DL, Kluge JP, Frank GH.Experimental infection of lambs with bovine respiratory syncytial virus and Pasteurellahaemolytica: clinical and microbiologic studies. Am J Vet Res 1982; 42: 236-240.95. Buckner CK, Songsiridej V, Dick EC, Busse WW. In vivo and in vitro model of theuse of the guinea pig as as model for virus-provoked airway hyperreactivity. Am RevResp Dis 1985; 132: 305-310.96. Saban R, Dick EC, Fishleder RI, Buckner CK. Enhancement by parainfluenza 3infection of contractile responses to substance P and capsaicin in airway smooth musclefrom the guinea pig. Am Rev Resp Dis 1987; 136: 586-591.11797. Lemen RJ, Quan SF, Witten ML, Sobonya RE, Ray CG, Grad R. Canineparainfluenza type 2 bronchiolitis increases histamine responsiveness in beaglepuppies. Am Rev Resp Dis 1990; 141: 199-207.98. Quan SF, Witten ML, Grad R, et al. Acute canine adenovirus 2 infection increaseshistamine airway reactivity in beagle puppies. Am Rev Resp Dis 1990; 141: 414-420.99. Sorkness R, Lemanske RFJr, Castleman WL. Persistent airway hyperresponsivenessafter neonatal viral bronchiolitis in rats. J Appl Physiol 1991; 70: 375-383.100. Wagner MH, Evermann JF, Gaskin J, McNicol K, Small P, Stecenko AA. Subacuteeffects of respiratory syncytial virus infection on lung function in lambs. PediatrPulmonol 1991; 11: 56-64.101. Engle S. Newns GH. Proliferative mural bronchiolitis. Arch Dis Child 1940; 15: 219-229.102. Hubble D, Osborne GR. Acute bronchiolitis in children. Br Med J 1941; 1: 107-110.103. Aherne W, Bird T, Court SDM, Gardner PS, McQuillin J. Pathological changes invirus infections of the lower respiratory tract in children. J Clin Path 1970; 23: 7-18.104. Neilson KA, Yunis EJ. Demonstration of respiratory syncytial virus in an autopsyseries. Pediatr Pathol 1990; 10: 491-502.105. Barnes PJ. Airway neuropeptides. In: Barnes PJ, Rodger IW, Thomson NC, eds.Asthma: Basic Mechanisms and Clinical Management.  London, UK: Academic Press,1988: 395-413.106. Lin C-Y, Kuo Y-C, Liu W-T, Lin C-C. Immunomodulation of influenza virus infectionin the precipitating asthma attack. Chest 1988; 93: 1234-1238.107. Ford-Hutchinson AW. The neutrophil and lymphocytes. In: Barnes PJ, Rodger IW,Thomson NC, eds. Asthma: Basic Mechanisms and Clinical Management.  London,UK: Academic Press Limited, 1988: 131-142.108. Theuson DO, Speck LS, Lett-Brown MA, Grant JA. Histamine releasing activity(HRA): I. Production by mitogen- or antigen-stimulated human mononuclear cells. JImmunol 1979; 123: 626-632.109. Theuson DO, Speck LS, Lett-Brown MA, Grant JA. Histamine releasing activity(HRA): II. Interaction with basophils and physicochemical characterization. J Immunol1979; 123: 633-639.110. Clementsen P, Norn S, Kristensen KS, Hannoun C. Influenza A virus enhancesbasophil histamine release and the enhancement is abolished by carbohydrates. Allergy1990; 45: 471-476.111. Chonmaitree T, Lett-Brown MA, Grant JA. Respiratory viruses induce production ofhistamine-releasing factor by mononuclear leukocytes: a possible role in the mechanismof virus-induced asthma. J Infectious Dis 1991; 164: 592-594.118112. Ida S, Hooks JJ, Siraganian RP, Notkins AL. Enhancement of IgE-mediated histaminerelease from human basophils by viruses: role of interferon. J Exp Med 1977; 145:892-905.113. Busse WW, Swenson CA, Borden EC, Treuhaft MW, Dick EC. Effect of influenza Avirus on leukocyte histamine release. J Allergy Clin Immunol 1983; 71: 382-388.114. Slauson DO, Lay JC, Castleman WL, Neilsen NR. Acute inflammatory lung injuryretards pulmonary particle clearance. Inflammation 1990; 13: 185-199.115. Witten ML, Grad R, Quan SF, Lantz RC, Sobonya RE, Lemen RJ. Effects ofrespiratory viruses on pulmonary alveolar macrophages. Pediatr Pulmonol 1992; 12:105-112.116. Trigo FJ, Breeze RG, Liggitt HD, Evermann JF, Trigo E. Interaction of bovinerespiratory syncytial virus and Pasteurella haemolytica in the ovine lung. Am J Vet Res1984; 45: 1671-1678.117. Roberts NJ, Prill AH, Mann TN. Interleukin 1 and interleukin 1 inhibitor productionby human macrophages exposed to influenza or respiratory syncytial virus. Respiratorysyncytial virus is a potent inducer of inhibitor activity. J Exp Med 1986; 163: 511-519.118. Morris JA, Blount RE, Savage RE. Recovery of cytopathogenic agent fromchimpanzees with coryza. Proc Soc Exp Biol Med 1956; 92: 544-549.119. Chanock RM, Roizman B, Myers R. Recovery from infants with respiratory illness ofa virus related to chimpanzee coryza agent (CCA). I. Isolation, properties andcharacterisation. Am J Hygiene 1957; 66: 281-290.120. Beem M, Wright FH, Hamre D, Egerer R, cEhme M. Association of the chimpanzeecoryza agent with acute respiratory disease in children. New Engl J Med 1960; 263:523-530.121. Holzel A, Parker L, Petterson WH, et al. Virus isolations from throats of childrenadmitted to hospital with respiratory and other diseases. Br Med J 1964; 1: 614-619.122. Chanock RM, Parrott RH. Acute respiratory disease in infancy and childhood: presentunderstanding and prospects for prevention. Pediatr 1965; 36: 21-39.123. Stott ET, Taylor G. Respiratory syncytial virus: brief review. Arch Virol 1985; 84: 1-52.124. Dunbar MR, Foreyt WJ. Serologic evidence of respiratory syncytial virus infection infree-ranging mountain goats (Oreamnos americanus). J Wildlife Dis 1986; 22: 415-416.125. Sparker TR, Collins JK. Isolation and serologic evidence of respiratory syncytial virusin bighorn sheep from Colorado. J Wildlife Dis 1986; 22: 416-418.126. Trudel M, Nadon F, Simard C, et al. Comparison of caprine, human and bovine strainsof respiratory syncytial virus. Arch Virol 1989; 107: 141-149.119127. Doane FW, Anderson N. Paramyxoviridae. In: Electron Microscopy in DiagnosticVirology. A Practical Guide and Atlas. Cambridge, UK: Cambridge University Press,1986: 149-154.128. Collins PL, Dickens LE, Wertz GW. cDNA cloning, mapping and translation of tenrespiratory syncytial virus mRNAs. In: Nonsegmented Negative Strand Viruses. SanDiego, CA: Academic Press, Inc., 1984: 21-34.129. Collins PL, Anderson K, Langer SJ, Wertz GW. Correct sequence for the majornucleocapsid protein mRNA of respiratory syncytial virus. Virology 1985; 146: 69-77.130. Lopez JA, Penas C, Garcia-Barreno B, Melero JA, Portela A. Location of a highlyconserved neutralizing epitope in the F glycoprotein of human respiratory syncytialvirus. J Virol 1990; 64: 927-930.131. Mufson MA, Orvell C, Rafnar B, Norrby E. Two distinct subtypes of humanrespiratory syncytial virus. J Gen Virol 1985; 66: 2111-2124.132. Norrby E, Mufson MA, Sheshberadaran H. Structural differences between subtype Aand B strains of respiratory syncytial virus. J Gen Virol 1986; 67: 2721-2729.133. McConnochie KM, Hall CB, Walsh EE, Roghmann KJ. Variation in severity ofrespiratory syncytial virus infections with subtype. J Pediatr 1990; 117: 52-62.134. Johnson PR, Collins PL. The 1B (N52), 1C (NS1) and N proteins of humanrespiratory syncytial virus (RSV) of antigenic subgroups A and B: sequenceconservation and divergence within RSV genomic RNA. J Gen Virol 1989; 70: 1539-1547.135. Talis A, McIntosh K. Respiratory syncytial virus. In: Balows A, Hausler WJ,Herrmann KL, Isenberg HD, Shadomy HJ, eds.  Manual of Clinical Microbiology, 5thedition. Washington, DC: American Society for Microbiology, 1991: 883-886.136. Becker S, Soukup J, Yankaskas JR. Respiratory syncytial virus infection of humanprimary nasal and bronchial epithelial cell cultures and bronchoalveolar macrophages.Am J Resp Cell Mol Biol 1992; 6: 369-375.137. Panuska JR, Cirino NM, Midulla F, Despot JE, McFadden ERJ, Huang YT.Productive infection of isolated human alveolar macrophages by respiratory syncytialvirus. J Clin Invest 1990; 86: 113-119.138. Baldridge P. Senterfit LB. Persistent infection of cells in culture by respiratorysyncytial virus. Proc Soc Exp Biol Med 1976; 151: 684-688.139. Fernie BF, Ford EC, Gerin JL. The development of Balb/c cells persistently infectedwith respiratory syncytial virus: presence of ribonucleoprotein on the cell surface. ProcSoc Exp Biol Med 1981; 167: 83-86.140. Pringle CR, Shirodaria PV, Cash P, Chiswell DJ, Malloy P. Initiation and maintenanceof persistent infection by respiratory syncytial virus. J Virol 1978; 28: 199-211.141. Parry JE, Shirodaria PV, Pringle CR. Pneumoviruses: the cell surface of lytically andpersistently infected cells. J Gen Virol 1979; 44: 479-491.120142. Oldstone MBA. Viruses can cause disease in the absence of morphological evidence ofcell injury: implication for uncovering new diseases in the future. J Infect Dis 1989;159: 384-389.143. Oldstone MBA. Viral persistence. Cell 1989; 56: 517-520.144. Oldstone MBA. Viral alteration of cell function. Sci Amer 1989; 261: 42-48.145. Vikerfors T, Grandien M, Olcen P. Respiratory syncytial virus infections in adults. AmRev Resp Dis 1987; 136: 561-564.146. Guidry GG, Black-Payne CA, Payne DK, Jamison RM, George RB, Bocchini JAJr.Respiratory syncytial virus infection among intubated adults in a university medicalintensive care unit. Chest 1991; 100: 1377-1384.147. Hogg JC, Williams J, Richardson JB, Macklem PT, Thurlbeck WM. Age as a factor inthe distribution of lower-airway conductance and in the pathologic anatomy ofobstructive lung disease. New Engl J Med 1970; 282: 1283-1287.148. Hogg JC, Macklem RT, Thurlbeck WM. Site and nature of obstruction in chronicobstructive lung disease. New Engl J Med 1968; 278: 1355-1360.149. Reid L. Influence of the pattern of structural growth of lung on susceptibility to specificinfectious diseases in infants and children. Pediatr Res 1977; 11: 210-215.150. Cannon MJ, Stott EJ, Taylor G, Askonas BA. Clearance of persistent respiratorysyncytial virus infections in immunodeficient mice following transfer of primed T cells.Immunol 1987; 62: 133-138.151. Johnson RA, Prince GA, Suffin SC, Horswood RL, Chanock RM. Respiratorysyncytial virus infection in cyclophosphamide-treated cotton rats. Infection Immunity1982; 37: 369-373.152. Kim HW, Canchola JG, Brandt CD, et al. Respiratory syncytial virus disease in infantsdespite prior administration of antigenic inactivated vaccine. Am J Epidemiol 1969; 89:422-434.153. Fulginiti VA, Eller JJ, Sieber OF, et al. Respiratory virus immunization: I. A field trialof two inactivated respiratory virus vaccines: an aqueous trivalent parainfluenza virusvaccine, and an alum-precipitated respiratory syncytial virus vaccine. Am J Epidemiol1969; 89: 435-448.154. Chin J, Magoffin RL, Shearer LA, et al. Field evaluation of a respiratory syncytialvirus vaccine and a trivalent parainfluenza virus vaccine in a pediatric population. Am JEpidemiol 1969; 89: 449-463.155. Prince GA, Jenson AB, Hemming VG, et al. Enhancement of respiratory syncytialvirus pulmonary pathology in cotton rats by prior intramuscular inoculation offormalin-inactivated virus. J Virol 1986; 57: 721-728.156. Piedra PA, Faden HS, Camussi G, Wong DT, Ogra PL. Mechanism of lung injury incotton rats immunized with formalin-inactivated respiratory syncytial virus. Vaccine1989; 7: 34-38.121157. Murphy BR, Sotnikov AV, Lawrence LA, Banks SM, Prince GA. Enhancedpulmonary histopathology is observed in cotton rats immunized with formalin-inactivated respiratory syncytial virus (RSV) or purified F glycoprotein and challengedwith RSV 3-6 months after immunization. Vaccine 1990; 8: 497-502.158. Toms GL, Webb MSC, Milner PD, et al. IgG and IgM antibodies to viral glycoproteinsin respiratory syncytial virus infections of graded severity. Arch Dis Child 1989; 64:1661-1665.159. Wagner DK, Muelenaer P, Henderson FW, et al. Serum immunoglobulin G antibodysubclass response to respiratory syncytial virus F and G glycoproteins after first,second and third infections. J Clin Microbiol 1989; 27: 589-592.160. Nicholas JA, Mitchell MA, Levely ME, et al. Mapping an antibody-binding site and a Tcell-stimulating site on the lA protein of respiratory syncytial virus. J Virol 1988; 62:4465-4473.161. Popow-Kraupp T, Lakits E, Kellner G, Kunz C. Immunoglobulin-class-specificimmune response to respiratory syncytial virus structural proteins in infants, children,and adults. J Med Virol 1989; 27: 215-223.162. Hall CB, Walsh EE, Long CE, Schnabel KC. Immunity to and frequency of reinfectionwith respiratory syncytial virus. J Infectious Dis 1991; 163: 693-698.163. Askonas BA, Openshaw P. Cannon M. T-cells in respiratory syncytial virus (RSV)infection. Adv Exp Med Biol 1989; 257: 9-17.164. Hambling MH. Antibody response in guinea-pigs following intranasal inoculation ofrespiratory syncytial virus. J Path Bact 1966; 91: 625-629.165. Levine S, Dajani A, Klaiber-Franco R. The response of infants with bronchiolitis to theproteins of respiratory syncytial virus. J Gen Virol 1988; 69: 1229-1239.166. Stewart RS, Gershwin U. Role of IgE in the pathogenesis of bovine respiratorysyncytial virus in sequential infections in vaccinated and nonvaccinated calves. Am JVet Res 1989; 50: 349-355.167. Kimman TG, Westenbiink F, Strayer PJ. Priming for local and systemic antibodymemory responses to bovine respiratory syncytial virus: effect of amount of virus,virus replication, route of administration and maternal antibodies. Vet ImmunolImmunopathol 1989; 22: 145-160.168. Cannon MJ, Bangham CRM. Recognition of respiratory syncytial virus fusion proteinby mouse cytotoxic T cell clones and a human cytotoxic T cell line. J Gen Virol 1989;70: 79-87.169. Piedra PA, Camussi G, Ogra PL. Immune response to experimentally induced infectionwith respiratory syncytial virus: possible role in the development of pulmonarydisease. J Gen Virol 1989; 70: 325-333.170. Stewart RS, Gershwin 11. Systemic and secretory antibody responses to sequentialbovine respiratory syncytial virus infections in vaccinated and nonvaccinated calves.Am J Vet Res 1990; 51: 1596-1602.122171. Nadal D, Ogra PL. Development of local immunity: role in mechanisms of protectionagainst or pathogenesis of respiratory syncytial viral infections. Lung 1990; 168(Suppi.): 379-387.172. Anderson JJ, Norden J, Saunders D, Toms GL, Scott R. Analysis of the local andsystemic immune responses induced in BALB/c mice by experimental respiratorysyncytial virus infection. J Gen Virol 1990; 71: 1561-1570.173. Glezen WP, Taber LH, Frank AL, Kasel JA. Risk of primary infection and reinfectionwith respiratory syncytial virus. Am J Dis Children 1986; 140: 543-546.174. Chiba Y, Higashidate Y, Suga K, Honjo K, Tsutsumi H, Ogra PL. Development ofcell-mediated cytotoxic immunity to respiratory syncytial virus in human infantsfollowing naturally acquired infection. J Med Virol 1989; 28: 133-139.175. Domurat FM, Keng P, Mock DJ, Roberts NJJ. Early identification and retrieval ordeletion of human lymphocyte subpopulations responding to influenza virus orrespiratory syncytial virus challenge. Cell Biophysics 1989; 15: 173-188.176. Sun C-S, Wyde PR, Wilson SZ, Knight V. Cell-mediated cytotoxic responses in lungsof cotton rats infected with respiratory syncytial virus. Am Rev Resp Dis 1983; 127:460-464.177. Openshaw PJM, Anderson K, Wertz GW, Askonas BA. The 22,000-kilodalton proteinof respiratory syncytial virus is a major target for Kd-restricted cytotoxic Tlymphocytes from mice primed by infection. J Virol 1990; 64: 1683-1689.178. Munoz JL, McCarthy CA, Clark ME, Hall CB. Respiratory syncytial virus infection inC57BL/6 mice: clearance of virus from the lungs with virus-specific cytotoxic T cells.J Virol 1991; 65: 4494-4497.179. Nicholas JA, Levely ME, Mitchell MA, Smith CW. A 16-amino acid peptide ofrespiratory syncytial virus 1A protein contains two overlapping T cell-stimulating sitesdistinguishable by class II MHC restriction elements. J Immunol 1989; 143: 2790-2796.180. Bangham CRM, Openshaw PJM, Ball LA, King AMQ, Wertz GW, Askonas BA.Human and murine cytotoxic T cells specific to respiratory syncytial virus recognize theviral nucleocapsid (N), but not the major glycoprotein (G), expressed by vaccinia virusrecombinants. J Immunol 1986; 137: 3973-3977.181. Cannon MJ, Openshaw PJM, Askonas BA. Cytotoxic T-cells clear virus but augmentlung pathology in mice infected with respiratory syncytial virus. J Exp Med 1988; 168:1163-1168.182. Coates HV, Chanock RM. Experimental infection with respiratory syncytial virus inseveral species of animals. Am J Hyg 1962; 76: 302-312.183. Belshe RB, Richardson LS, London WT, et al. Experimental respiratory syncytialvirus infection of four species of primates. J Med Virol 1977; 1: 157-162.184. Richardson LS, Belshe RB, Sly DL, et al. Experimental respiratory syncytial viruspneumonia in cebus monkeys. J Med Virol 1978; 2: 45-59.123185. Prince GA, Suffin SC, Prevar DA, et al. Respiratory syncytial virus infection in owlmonkeys: viral shedding, immunological response, and associated illness caused bywild-type virus and two temperature-sensitive mutants. Infection Immunity 1979; 26:1009-1013.186. Mohanty SB, Ingling AL, Lillie MG. Experimentally induced respiratory syncytial viralinfection in calves. Am J Vet Res 1975; 36: 417-419.187. McNulty MS, Bryson DG, Allan GM. Experimental respiratory syncytial viruspneumonia in young calves: microbiologic and immunofluorescent findings. Am J VetRes 1983; 44: 1656-1659.188. LeBlanc PH, Baker JC, Gray PR, Robinson NE, Derksen FJ. Effects of bovinerespiratory syncytial virus on airway function in neonatal calves. Am J Vet Res 1991;52: 1401-1406.189. Lehmkuhl HD, Cutlip RC. Experimentally induced respiratory syncytial viral infectionin lambs. Am J Vet Res 1979; 40: 512-514.190. Al-Darraji AM, Cutlip RC, Lehmkuhl HD, Graham DL. Experimental infection oflambs with bovine respiratory syncytial virus and Pasteurella haemolytica: pathologicstudies. Am J Vet Res 1982; 42: 224-229.191. Al-Darraji AM, Cutlip RC, Lehmkuhl HD. Experimental infection of lambs with bovinerespiratory syncytial virus and Pasteurella haemolytica: immunofluorescent andelectron microscopic studies. Am J Vet Res 1982; 42: 230-235.192. Sharma R, Woldehiwet Z. Pathogenesis of bovine respiratory syncytial virus inexperimentally infected lambs. Vet Microbiol 1990; 23: 267-272.193. Bryson DG, Evermann JF, Liggitt HD, Foreyt WJ, Breeze RG. Studies on thepathogenesis and interspecies transmission of respiratory syncytial virus isolated fromsheep. Am J Vet Res 1988; 49: 1424-1430.194. Tyeryar FJ, Richardson LS, Belshe RB. Report of a workshop on respiratory syncytialvirus and parainfluenza viruses. J Infectious Dis 1978; 137: 835-846.195. Prince GA, Porter DD. The pathogenesis of respiratory syncytial virus infection ininfant ferrets. Am J Pathol 1976; 82: 339-352.196. Prince GA, Horswood RL, Berndt J, Suffin SC, Chanock RM. Respiratory syncytialvirus infection in inbred mice. Infection Immunity 1979; 26: 764-766.197. Taylor G, Stott El, Hughes M, Collins AP. Respiratory syncytial virus infection inmice. Infection Immunity 1984; 43: 649-655.198. Graham BS, Perkins MD, Wright PF, Karzon DT. Primary respiratory syncytial virusinfection in mice. J Med Virol 1988; 26: 153-162.199. Prince GA, Jenson AB, Horswood RL, Camargo E, Chanock RM. The pathogenesisof respiratory syncytial virus infection in cotton rats. Am J Pathol 1978; 93: 771-792.124200. Anderson JJ, Serin M, Harrop J, Amin S, Toms GL, Scott R. Natural killer cellresponse to respiratory syncytial virus in the BALB/c mouse. Adv Exp Med Biol 1989;257: 211-220.201. Graham BS, Bunton LA, Wright PF, Karzon DT. Role of T lymphocyte subsets in thepathogenesis of primary infection and rechallenge with respiratory syncytial virus inmice. J Clin Invest 1991; 88: 1026-1033.202. Norrby E, Akerlaind B, Mufson MA. The immunobiology of respiratory syncytialvirus: prospects for a vaccine. Adv Exp Med Biol 1989; 257: 147-153.203. Murphy BR, Sotnikov A, Paradiso PR, et al. Immunization of cotton rats with thefusion (F) and large (G) glycoproteins of respiratory syncytial virus (RSV) protectsagainst RSV challenge without potentiating RSV disease. Vaccine 1989; 7: 533-540.204. Berglund B, Kortekangas AE, Lauren P. Experimental inoculation of guinea pig'smiddle ear with respiratory syncytial virus. Acta Otolaryngol 1966; 224 (Suppl.): 268-271205. Folkerts G. Virus-induced airway hyperresponsiveness [PhD]. Rijksuniversiteit toUtrecht, 1991.206. Altman PL, Katz DD. IV. Guinea pig. In: Altman PL, Katz DD, eds. Inbred andGenetically Defined Strains of Laboratory Animals.  Bethesda, MD: Federation ofAmerican Societies for Experimental Biology, 1979: 505-563.207. Patterson R, Kelly JF. Animal models of the asthmatic state. Ann Rev Med 1974; 25:53-68.208. Wanner A, Abraham WM, Douglas JS, Drazen JM, Richerson HB, Sri Ram J. Modelsof airway hyperresponsiveness. Am Rev Resp Dis 1990; 141: 253-257.209. Harkness JE, Wagner JE. The Guinea Pig. In: Harkness JE, Wagner JE, eds. TheBiology and Medicine of Rabbits and Rodents, 2nd edition.  Philadelphia, PA: Lea &Febiger, 1983: 17-24.210. Karol M, Jin R, Bennedsen M, Vaughan F. Production and isolation of guinea pig IgEantibody. J Immunol Methods 1991; 139: 123-134.211. Schafer H, Burger R. Identification and functional characterization of guinea pig CD4:antibody binding transduces a negative signal on T-cell activation. Immunol 1991; 72:261-268.212. Haslov K, Bentzon MW, Moller S. Estimation of guinea pig antigen-specific and non-specific suppressor cell activity. J Immunol Methods 1989; 120: 85-91.213. Castleman WL. Spontaneous pulmonary infections in animals affecting studies oncomparative lung morphology and function. Am Rev Resp Dis 1983; 128: S83-S87.214. Wagner JE, Foster HL. Germfree and specific pathogen-free. In: Wagner JE, ManningPJ, eds. The Biology of the Guinea Pig.  New York, NY: Academic Press, Inc., 1976:21-28.125215. Thurlbeck WM. Examination of the lung: autopsy. In: Thurlbeck WM, ed. Pathologyof the Lung. New York, NY: Thieme Medical Publishers, Inc., 1988: 95-100.216. Miller RR. Viral infections of the respiratory tract. In: Thurlbeck WM, ed. Pathologyof the Lung. New York, NY: Thieme Medical Publishers, Inc., 1988: 147-179.217. Cosio M, Ghezzo H, Hogg JC, et al. The relations between structural changes in smallairways and pulmonary-function tests. New Engl J Med 1977; 298: 1277-1281.218. Wright JL, Cosio M, Wiggs B, Hogg JC. A morphologic grading scheme formembranous and respiratory bronchioles. Arch Path Lab Med 1985; 109: 163-165.219. Cubie HA, Inglis JM, McGowan AM. Detection of respiratory syncytial virus antigenand nucleic acid in clinical specimens using synthetic oligonucleotides. J Virol Methods1991; 34: 27-35.220. Baal T, Howe C. Morphogenesis and ultrastructure of respiratory syncytial virus. JVirol 1973; 12: 1173-1180.221. Fuchs H, Bachi T. Scanning electron microscopical demonstration of respiratorysyncytial virus antigens by immunological markers. J Ultrastruct Res 1975; 52: 114-119.222. Castleman WL, Chandler SK, Slauson DO. Experimental bovine respiratory syncytialvirus infection in conventional calves: ultrastructural respiratory lesions. Am J Vet Res1985; 46: 554-560.223. Bryson DG, Platten MF, McConnell S, McNulty MS. Ultrastructural features oflesions in bronchiolar epithelium in induced respiratory syncytial virus pneumonia ofcalves. Vet Pathol 1991; 28: 293-299.224. Bryson DG, McConnell S, McAliskey M, McNulty MS. Ultrastructural features ofalveolar lesions in induced respiratory syncytial virus pneumonia in calves. Vet Pathol1991; 28: 286-292.225. Halstead DC, Todd S, Fritch G. Evaluation of five methods for respiratory syncytialvirus detection. J Clin Microbiol 1990; 28: 1021-1025.226. Okamoto Y, Shirotori K, Kudo K, et al. Genomic sequences of respiratory syncytialvirus in otitis media with effusion. Lancet 1991; 338: 1025-1026.227. Paton AW, Paton JC, Lawrence AJ, Goldwater PN, Harris RJ. Rapid detection ofrespiratory syncytial virus in nasopharyngeal aspirates by reverse transcription andpolymerase chain reaction amplification. J Clin Microbiol 1992; 30: 901-904.228. Joncas JH, Berthiqume L, Williams R, Beaudry P. Pavilanis V. Diagnosis of viralrespiratory infections by electron microscopy. Lancet 1969; 1: 956-959.229. Naeem K, Goyal SM. Comparison of virus isolation, immunofluorescence and electronmicroscopy for the diagnosis of animal viruses. Microbiologica 1988; 11: 355-362.230. Hyatt AD. The application of electron microscopy to veterinary virus diagnosis. AustrVet J 1989; 66: 445-449.126231. Schulze P. Predictive value of electron microscopic studies in virological diagnosis.Zeit Ges Hyg Grenzg 1990; 36: 250-254.232. Yunis E, Agostini R. Syncytial giant cell hepatitis [letter]. New Engl J Med 1992; 327:130.233. Phillips MJ. Syncytial giant cell hepatitis. New Engl J Med 1992; 327: 130-131.234. Shulinder AR, Tanner K, Moore CA, Roth J. RNA template-specific PCR: animproved method that dramatically reduces false positives in RT-PCR. BioTechniques1991; 11: 560-563.235. Hall CB. The shedding and spreading of respiratory syncytial virus. Pediatr Res 1977;11: 236-239.236. Festing MFW. Growth and body composition. In: Wagner JE, Manning PJ, eds. TheBiology of the Guinea Pig. New York, NY: Academic Press, Inc., 1976.237. Treuhaft MW, Beem MO. Defective interfering particles of respriatory syncytial virus.Infection Immunity 1982; 37: 439-444.238. Lennette EH, Schmidt NJ.  Diagnostic Procedures for Viral, Rickettsial and Chlamydial Infections, 5th edition. Washington, DC: American Public Health Association, 1979:100-104.239. Wright JL, Wiggs B, Pare PD, Hogg JC. Ranking the severity of emphysema onwhole lung slices: concordance of upper lobe, lower lobe, and entire lung ranks. AmRev Resp Dis 1986; 133: 930-931.240. Drury RAB, Wallington EA. Carleton's Histological Technique. London, UK: OxfordUniversity Press, 1967: 48-137.241. Churg A. An inflation procedure for open lung biopsies. Am J Surg Pathol 1983; 7:69-71.242. Faden H, Hong JJ, Ogra PL. Interaction of polymorphonuclear leukocytes and virusesin humans: adherence of polymorphonuclear leukocytes to respiratory syncytial virus-infected cells. J Virol 1984; 52: 16-23.243. Karpick RJ, Pratt PC, Asmundsson T, Kilburn KH. Pathological findings inrespiratory failure: goblet cell metaplasia, alveolar damage, and myocardial infarction.Ann Int Med 1970; 72: 189-197.244. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidiniumthiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162: 156-159.245. Sambrook J, Fritsch EF, Maniatis T. Electrophoresis of RNA through agarose gelscontaining formaldehyde. In:  Molecular Cloning: A Laboratory Manual, 2nd edition. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989: 7.43-7.45.246. Sambrook J, Fritsch EF, Maniatis T. Agarose gel electrophoresis. In: Molecular Cloning: A Laboratory Manual, 2nd edition.  Cold Spring Harbor, NY: Cold SpringHarbor Laboratory Press, 1989: 6.3-6.16.127247. Sambrook J, Fritsch EF, Maniatis T. Analysis of genomic DNA by Southernhybridization. In:  Molecular Cloning: A Laboratory Manual, 2nd edition.  Cold SpringHarbor, NY: Cold Spring Harbor Laboratory Press, 1989: 9.31-9.45.248. Zoller MJ, Smith M. Oligonucleotide-directed mutagenesis of DNA fragments clonedinto M13 vectors. Methods in Enzymology 1983; 100: 468-500.249. Conover WJ. Measures of dependence. In: Practical Nonparametric Statistics.  NewYork, NY: John Wiley and Sons, 1980: 178-188.250. Siegel S, CasteIlan NJ. Nonparametric Statistics for the Behavioral Sciences, 2ndedition. New York, NY: McGraw-Hill, 1988.251. Sachs L. Applied Statistics: A Handbook of Techniques.  New York, NY: Springer,Verlag, 1982: 537.252. Godfrey K. Comparing means of several groups. In: Bailar JCIII, Mosteller F, eds.Medical Uses of Statistics. Waltham, MA: NEJM Books, 1986: 205-234.253. Sambrook J, Fritsch EF, Maniatis T. Extraction and purification of RNA. In:Molecular Cloning: A Laboratory Manual, 2nd edition.  Cold Spring Harbor, NY: ColdSpring Harbor Laboratory Press, 1989: 7.4.254. Henderson DW, Papadimitriou JM.  Ultrastructural Appearances of Tumours. A Diagnostic Atlas. Edinburgh, UK: Churchill Livingstone, 1982.255. Mills J, van Kirk JE, Wright PF, Chanock RM. Experimental respiratory syncytialvirus infection in adults. J Immunol 1971; 107: 123-130.256. Robinson PJ, Hegele RG, Schellenberg RR. Increased airway responsivenessassociated with respiratory syncytial-virus induced bronchiolitis in the guinea pigs (inpreparation).257. Glezen WP, Denny FW. Epidemiology of acute lower respiratory disease in children.New Engl J Med 1973; 288: 498-505.258. Roitt IM. T -cell regulation. In: Essential Immunology, 7th edition. Oxford, UK:Blackwell Scientific Publications, 1991: 154-159.259. Ford DK. Synovial lymphocyte responses in the spondyloarthropathies. In: Ziff M,Cohen SB, eds. Advances in Inflammation Research, Volume 9: The Spondyloarthropathies.  New York, NY: Raven Press, 1985: 189-202.260. Ford DK, da Roza DM, Schulzer M. Lymphocytes from the site of disease but notblood lymphocytes indicate the cause of arthritis. Ann Rheu Dis 1985; 44: 701-710.261. Keat AC, Knight SC. Do synovial fluid cells indicate the cause of reactive arthritis? JRheumatol 1990; 17: 1257-1259.262. Pabst R, Gehrke I. Is the bronchus-associated lymphoid tissue (BALT) an integralstructure of the lung in normal mammals, including humans? Am J Resp Cell Mol Biol1990; 3: 131-135.128263. Pabst R. Is BALT a major component of the human lung immune system? ImmunolToday 1992; 13: 119-122.264. Bienenstock J. Bronchus-associated lymphoid tissue. In: Bienenstock J, ed.Immunology of the Lung and Upper Respiratory Tract. New York, NY: McGraw-Hill, 1984: 96-118.265. Tannock GA, Hierholzer JC, Bryce DA, Chee C-F, Paul JA. Freeze-drying ofrespiratory syncytial viruses for transportation and storage. J Clin Microbiol 1987; 25:1769-1771.266. Morrell RE, Marks MI, Champlin R, Spence L. An outbreak of severe pneumonia dueto respiratory syncytial virus in isolated Arctic populations. Am J Epidemiol 1975; 101:231-237.267. Bendayan M, Zollinger M. Ultrastructural localization of antigenic sites on osmium-fixed tissues applying the protein A-gold technique. J Histochem Cytochem 1983; 19:101-109.268. Bendayan M. Protein-A gold electron microscopic immunocytochemistry: methods,applications, and limitations. J Electron Microsc Tech 1984; 1: 243-270.269. Amersham Corporation. ECL. Detection of Nucleic Acids and Proteins with Light. Amersham International PLC, 1990.270. Masse R, Fritsch P, Nolibe D, Sedaghat B. Evaluation quantitative et homeostasie de lapopulation des macrophages alveolaires. Pathol Biol 1975; 23: 464469.271. Brain JD, Godleski JJ, Syroken SP. Quantification, origin and fate of pulmonarymacrophages. In L'Enfant C, ed. Respiratory Defense Mechanisms, Part II. NewYork, NY: Marcel Dekker, 1977: 849-889.272. Knecht H, Odermatt BF, Bachmann E, et al. Frequent detection of Epstein-Barr virusDNA by the polymerase chain reaction in lymph node biopsies from patients withHodgkin's disease without genomic evidence of B- or T-cell clonality. Blood 1991; 78:760-767.273. Grillo M, Margolis FL. Use of reverse transcriptase polymerase chain reaction tomonitor expression of intronless genes. Biotechniques 1990; 9: 262-268.274. Hoof T, Riordan JR, Tummler B. Quantitation of mRNA by the kinetic polymerasechain reaction assay: a tool for monitoring P-glycoprotein gene expression. AnalytBiochem 1991; 196: 161-169.275. Buck KJ, Harris RA, Sikela JM. A general method for quantitative PCR analysis ofmRNA levels for members of gene families: application to GABA receptor subunits.Biotechniques 1991; 11 :636-641.276. van Dyke RB, Murphy-Corb M. Detection of respiratory syncytial virus innasopharyngeal secretions by DNA-RNA hybridization. J Clin Microbiol 1989; 27:1739-1743.277. Singer RH, Lawrence JB, Villnave C. Optimization of in situ hybridization usingisotopic and non-isotopic detection methods. Biotechniques 1986; 4: 232-246.129278. Murphy BR, Prince GA, Lawrence LA, Croen KD, Collins PL. Detection ofrespiratory syncytial virus (RSV) infected cells by in situ hybridization in the lungs ofcotton rats immunized with formalin-inactivated virus or purified RSV F and Gglycoprotein subunit vaccine and challenged with RSV. Virus Res 1990; 16: 153-162.279. Nuovo GJ, MacConnell P. Forde A, Delvenne P. Detection of human papillomavirusDNA in formalin-fixed tissues by in situ hybridization after amplification bypolymerase chain reaction. Am J Pathol 1991; 139: 847-854.280. Bagasra 0, Hauptman SP, Lischner HW, Sachs M, Pomerantz RJ. Detection of humanimmunodeficiency virus type 1 provirus in mononuclear cells by in situ polymerasechain reaction. New Engl J Med 1992; 326: 1385-1391.281. Reynolds HY. Bronchoalveolar lavage. Am Rev Resp Dis 1987; 135: 250-263.282. Fick RBJr, Richerson HB, Zavala DC, Hunninghake GW. Bronchoalveolar lavage inallergic asthmatics. Am Rev Resp Dis 1987; 125: 1204-1209.130APPENDIX A: EXPERIMENTAL DATA FOR ALL GUINEA PIGS STUDIEDLEGEND:ANIMALNote: Blank spaces indicate that no data are available.Unique guinea pig identifier. The first four digits refer to the Experimentnumber of the UBC Pulmonary Research Laboratory. The next two digitsrefer to the animal number within the experiment. For example,"174203.000" designates Experimental number 1742, Animal 3.Experiment numbers are as follows:17421755179118301779188919501979"historical" controlsday 6day 6day 6day 14day 60day 125day 6 (RT-PCR)juevnile guinea pigsjuvenile guinea pigsjuvenile guinea pigsadolescent guinea pigsjuvenile guinea pigsjuvenile guinea pigsjuvenile guinea pigsjuvenile guinea pigsunmanipulatedlow dose RSVhigh dose RSVlow dose RSVlow dose RSVlow dose RSVlow dose RSVlow dose RSVGROUPRSVBODY WTIBODY WTFLUNGWTLBRATIOEPITHMONOPMNEDEMABALTGOBLETCULTURETEMIMMUNORT-PCRDesignates unique groups of guinea pigs (i.e., handled in the same manner forthe same experiment).RSV inoculation status: (1=uninfected controls; 2= RSV-inoculated animals)Body weight (g) on the day of the inoculation procedure.Body weight (g) on the study day.Lung wet weight (g) on the study day.Lung to body weight ratio on the study day (= LUNGWT/BODYWTF)Histological score for epithelial necrosis (n=10 airways)Histological score for mononuclear cell infiltrates (n=10 airways)Histological score for polymorphonuclear cell infiltrates (n=10 airways)Histological score for airway wall edema (n=10 airways)Histological score for hyperplasia of bronchus-associated lymphoid tissue(n=10 airways)Histological score for goblet cell metaplasia (n=10 airways)Results of one month viral culture^(1=no CPE; CPE,xxx = number of days post-plating onto HEp-2 cells until CPE was observed).For Expt # 1979, values express the number of pfu/g wet weight fresh lung.Results of transmission electron microscopy (1=negative; 2=positive)Results of immunohistochemistry^(1=negative; 2=positive)Results of RT-PCR for RSV genomic RNA (1=negative; 2=positive)131ANIMAL GROUP RSV BODYwn BODYWTF174203.000 1.000 1.000174206.000 1.000 1.000174211.000 1.000 1.000174212.000 1.000 1.000174218.000 1.000 1.000174219.000 1.000 1.000175501.000 2.000 1.000 433.000175502.000 2.000 1.000175503.000 2.000 1.000 438.000175504.000 3.000 2.000 441.000175505.000 2.000 1.000 383.000175506.000 2.000 1.000 380.000175507.000 3.000 2.000 422.000175508.000 3.000 2.000 428.000175509.000 3.000 2.000 387.000175510.000 2.000 1.000175511.000 2.000 1.000 427.000175512.000 3.000 2.000 382.000175513.000 3.000 2.000 370.000175514.000 3.000 2.000 382.000175515.000 2.000 1.000 399.000175516.000 2.000 1.000 429.000175517.000 2.000 1.000 407.000175518.000 3.000 2.000 378.000175519.000 3.000 2.000 367.000175520.000 3.000 2.000 394.000179101.000 4.000 1.000 281.000 321.000179104.000 5.000 2.000 302.000 324.000179107.000 4.000 1.000 289.000 322.000179110.000 5.000 2.000 277.000 307.000179113.000 4.000 1.000 294.000 315.000179114.000 4.000 1.000 288.000 327.000179115.000 5.000 2.000 292.000 331.000179116.000 5.000 2.000 289.000 316.000179117.000 4.000 1.000 323.000 337.000179118.000 4.000 1.000 324.000 355.000179119.000 5.000 2.000 316.000 351.000179120.000 5.000 2.000 296.000 328.000179128.000 4.000 1.000 316.000 350.000179129.000 4.000 1.000 326.000 352.000179132.000 5.000 2.000 312.000 345.000179133.000 5.000 2.000 324.000 362.000179136.000 4.000 1.000 331.000 374.000179137.000 4.000 1.000 312.000 341.000179138.000 5.000 2.000 326.000 359.000179139.000 5.000 2.000 294.000 333.000183001.000 6.000 1.000 518.000183002.000 6.000 1.000 527.000183003.000 6.000 1.000 497.000183004.000 6.000 1.000 503.000183005.000 6.000 1.000 533.000132ANIMAL GROUP RSV BODY WTI BODY WTF183006.000 6.000 1.000 526.000183007.000 6.000 1.000 501.000183008.000 6.000 1.000 488.000183009.000 6.000 1.000 539.000183019.000 7.000 2.000 527.000183020.000 7.000 2.000 506.000183021.000 7.000 2.000 541.000183022.000 7.000 2.000 521.000183023.000 7.000 2.000 522.000183032.000 7.000 2.000 545.000183033.000 7.000 2.000 514.000183034.000 7.000 2.000 505.000183035.000 7.000 2.000 513.000183036.000 7.000 2.000 499.000177901.000 8.000 1.000177902.000 8.000 1.000177903.000 8.000 1.000 4. 95.000177904.000 9.000 2.000 473.000177905.000 9.000 2.000 427.000177906.000 8.000 1.000 475.000177907.000 8.000 1.000177908.000 9.000 2.000 532.000177909.000 9.000 2.000 469.000177910.000 8.000 1.000 491.000177911.000 8.000 1.000 449.000177912.000 9.000 2.000 438.000177913.000 9.000 2.000 434.000177914.000 9.000 2.000 487.000177915.000 8.000 1.000 470.000177916.000 8.000 1.000 434.000177917.000 9.000 2.000 518.000177918.000 9.000 2.000 407.000177919.000 9.000 2.000 451.000188901.000 10.000 1.000 .3 83.000 563.000188902.000 10.000 1.000 336.000 643.000188903.000 10.000 1.000 352.000 643.000188904.000 10.000 1.000 340.000 611.000188905.000 10.000 1.000 380.000 584.000188906.000 10.000 1.000 366.000 655.000188907.000 11.000 2.000 342.000 576.000188908.000 11.000 2.000 358.000 703.000188909.000 11.000 2.000 365.000 660.000188910.000 11.000 2.000 357.000 635.000188911.000 11.000 2.000 369.000 648.000188912.000 10.000 1.000 746.000188913.000 10.000 1.000 675.000188914.000 10.000 1.000 768.000188915.000 10.000 1.000 704.000188916.000 10.000 1.000 681.000188917.000 11.000 2.000 627.000188918.000 11.000 2.000 .3 66.000 699.000188919.000 11.000 2.000 343.000 746.000133ANIMAL GROUP RSV BODYWTI BODYWTF188920.000 11.000 2.000 366.000 609.000188921.000 11.000 2.000 409.000 600.000195001.000 12.000 1.000 340.000 722.000195002.000 13.000 2.000 383.000 845.000195003.000 13.000 2.000 360.000 757.000195004.000 13.000 2.000 390.000 869.000195005.000 13.000 2.000 419.000 835.000197903.000 15.000 2.000 299.000 331.000197905.000 15.000 2.000 292.000 317.000197906.000 15.000 2.000 286.000 324.000197907.000 15.000 2.000 292.000 331.000197919.000 14.000 1.000 328.000 379.000197920.000 14.000 1.000 309.000 331.000134ANIMAL LUNGWT LBRATIO EPITH MONO174203.000 2.000 0.000174206.000 4.000 2.000174211.000 1.000 0.000174212.000 2.000 0.000174218.000 7.000 1.000174219.000 4.000 2.000175501.000 1.610 0. .372 2.000 0.000175502.000 1.250175503.000 1.620 0.370 4.000 0.000175504.000 1.730 0.392 7.000 14.000175505.000 1.600 0.418 1.000 1.000175506.000 1.630 0.429 4.000 2.000175507.000 2.180 0.512 4.000 0.000175508.000 1.600 0.374 8.000 18.000175509.000 1.510 0.390 9.000 12.000175510.000 2.300 1.000 0.000175511.000 1.820 0.426 2.000 4.000175512.000 1.640 0.429 17.000 13.000175513.000 1.500 0.405 10.000 9.000175514.000 1.640 0.429 8.000 13.000175515.000 1.470 0.368 4.000 1.000175516.000 1.4.40 0.336 1.000 2.000175517.000 1.420 0.349 2.000 0.000175518.000 1.290 0.341 5.000 0.000175519.000 1.450 0.395 5.000 9.000175520.000 1.520 0.386 2.000 2.000179101.000 1.270 0.396 3.000 1.000179104.000 1.200 0.370 9.000 4.000179107.000 1.110 0.345 5.000 2.000179110.000 1.130 0.368 7.000 4.000179113.000 1.070 0.340 5.000 4.000179114.000 1.330 0.407 4.000 3.000179115.000 1.110 0.335 9.000 4.000179116.000 1.190 0.376 6.000 2.000179117.000 1.160 0.344 2.000 1.000179118.000 1.120 0.315 2.000 1.000179119.000 1.370 0.390 8.000 3.000179120.000 1.120 0.341 8.000 5.000179128.000 1.050 0.300 1.000 0.000179129.000 1.220 0.347 7.000 3.000179132.000 1.200 0.348 8.000 3.000179133.000 1.160 0.320 8.000 5.000179136.000 1.320 0.353 2.000 0.000179137.000 1.360 0.399 3.000 2.000179138.000 1.240 0.364 7.000 5.000179139.000 1.610 0.483 10.000 5.000183001.000 2.000 0.386 1.000 1.000183002.000 2.260 0.429 3.000 1.000183003.000 2.720 0.547 3.000 0.000183004.000 2.560 0.509 4.000 0.000183005.000 2.300 0.431 1.000 0.000183006.000 2.410 0.458 6.000 10.000135ANIMAL LUNGWT LBRATIO EPITH MONO183007.000 2.350 0.469 1.000 0.000183008.000 2.510 0.514 9.000 6.000183009.000 2.290 0.425 4.000 1.000183019.000 2.180 0.414 8.000 8.000183020.000 2.190 0.405 10.000 8.000183021.000 2.130 0.409 2.000 0.000183022.000 2.540 0.488 1.000 0.000183023.000 2.480 0.475 1.000 4.000183032.000 2.370 0.435 1.000 0.000183033.000 2.310 0.449 1.000 0.000183034.000 2.420 0.479 1.000 2.000183035.000 2.300 0.448 1.000 0.000183036.000 2.290 0.459 6.000 3.000177901.000 5.000 1.000177902.000 2.000 1.000177903.000 2.490 0..503 5.000 3.000177904.000 2.140 0.452 2.000 3.000177905.000 2.360 0.553 4.000 3.000177906.000 2.670 0.562 2.000 0.000177907.000 2.360 2.000 3.000177908.000 1.640 0.308 2.000 1.000177909.000 1.460 0.311 4.000 3.000177910.000 1.310 0.267 4.000 1.000177911.000 1.550 0.345 4.000 2.000177912.000 1.400 0.320 7.000 3.000177913.000 1.760 0.406 2.000 0.000177914.000 1.710 0.351 7.000 0.000177915.000 1.420 0.302 1.000 0.000177916.000 1.470 0.339 6.000 0.000177917.000 1.590 0.307 2.000 0.000177918.000 1.510 0.371 12.000 6.000177919.000 1.430 0.317 4.000 2.000188901.000 2.310 0.410 2.000 1.000188902.000 1.700 0.264 5.000 0.000188903.000 1.570 0.244 1.000 0.000188904.000 1.500 0.245 3.000 2.000188905.000 1.930 0.330 7.000 2.000188906.000 1.500 0.229 3.000 0.000188907.000 1.600 0.278 4.000 1.000188908.000 1.700 0.242 6.000 2.000188909.000 1.700 0.256 6.000 0.000188910.000 3.670 0.578 6.000 1.000188911.000 3.730 0.576 4.000 3.000188912.000 3.900 0.523 6.000 4.000188913.000 3.960 0.587 3.000 1.000188914.000 4.200 0.547 6.000 2.000188915.000 4.700 0.668 3.000 1.000188916.000 4.500 0.662 4.000 2.000188917.000 5.700 0.909 0.000 1.000188918.000 4.000 0.572 9.000 2.000188919.000 3.100 0.416 2.000 0.000188920.000 1.900 0.312 5.000 6.000136ANIMAL LUNGWT LBRATIO EPITH MONO188921.000 4.100 0.683 4.000 6.000195001.000 2.870 0.397 2.000 1.000195002.000 3.310 0.392 3.000 0.000195003.000 2.640 0.349 3.000 0.000195004.000 2.520 0.290 2.000 0.000195005.000 3.800 0.455 4.000 0.000197903.000 1.190 0.360 9.000 4.000197905.000 1.180 0.372 6.000 0.000197906.000 1.110 0.343 5.000 2.000197907.000 1.170 0.353 7.000 3.000197919.000 1.410 0.372 3.000 1.000197920.000 1.120 0.338 2.000 0.000137ANIMAL PMN EDEMA BALT GOBLET174203.000 4.000 1.000 0.000 10.000174206.000 3.000 6.000 0.000 5.000174211.000 2.000 2.000 0.000 8.000174212.000 5.000 0.000 1.000 10.000174218.000 5.000 7.000 1.000 4.000174219.000 5.000 5.000 1.000 6.000175501.000 3.000 0.000 0.000 11.000175502.000175503.000 6.000 2.000 2.000 5.000175504.000 9.000 7.000 6.000 14.000175505.000 3.000 1.000 5.000 5.000175506.000 8.000 6.000 2.000 10.000175507.000 5.000 4.000 0.000 13.000175508.000 14.000 10.000 7.000 13.000175509.000 7.000 7.000 11.000 12.000175510.000 4.000 7.000 0.000 10.000175511.000 7.000 7.000 2.000 16.000175512.000 14.000 13.000 6.000 8.000175513.000 11.000 6.000 6.000 13.000175514.000 10.000 11.000 8.000 9.000175515.000 5.000 4.000 2.000 9.000175516.000 5.000 5.000 1.000 10.000175517.000 5.000 1.000 0.000 9.000175518.000 8.000 8.000 1.000 12.000175519.000 9.000 11.000 3.000 11.000175520.000 4.000 1.000 1.000 7.000179101.000 3.000 5.000 0.000 7.000179104.000 9.000 7.000 1.000 8.000179107.000 4.000 5.000 1.000 9.000179110.000 4.000 4.000 2.000 9.000179113.000 6.000 7.000 4.000 4.000179114.000 2.000 6.000 3.000 6.000179115.000 5.000 6.000 2.000 7.000179116.000 6.000 7.000 2.000 8.000179117.000 2.000 3.000 0.000 7.000179118.000 0.000 3.000 1.000 8.000179119.000 6.000 6.000 1.000 8.000179120.000 10.000 7.000 3.000 8.000179128.000 0.000 4.000 1.000 7.000179129.000 4.000 9.000 2.000 8.000179132.000 8.000 4.000 1.000 6.000179133.000 8.000 6.000 2.000 8.000179136.000 2.000 5.000 0.000 7.000179137.000 4.000 6.000 1.000 8.000179138.000 8.000 4.000 2.000 9.000179139.000 11.000 8.000 3.000 12.000183001.000 1.000 2.000 3.000 7.000183002.000 2.000 3.000 0.000 4.000183003.000 0.000 3.000 0.000 3.000183004.000 2.000 1.000 3.000 5.000183005.000 0.000 1.000 0.000 2.000138ANIMAL PMN EDEMA BALT GOBLET183006.000 7.000 6.000 5.000 13.000183007.000 0.000 1.000 1.000 4.000183008.000 5.000 9.000 3.000 6.000183009.000 2.000 4.000 4.000 4.000183019.000 8.000 6.000 7.000 5.000183020.000 6.000 7.000 2.000 9.000183021.000 4.000 1.000 2.000 3.000183022.000 2.000 2.000 1.000 3.000183023.000 4.000 4.000 3.000 2.000183032.000 1.000 2.000 0.000 1.000183033.000 3.000 1.000 2.000 1.000183034.000 4.000 4.000 3.000 2.000183035.000 3.000 1.000 1.000 3.000183036.000 5.000 7.000 0.000 6.000177901.000 5.000 6.000 2.000 11.000177902.000 0.000 1.000 2.000 10.000177903.000 4.000 11.000 4.000 13.000177904.000 7.000 7.000 3.000 15.000177905.000 6.000 6.000 3.000 5.000177906.000 2.000 3.000 2.000 12.000177907.000 3.000 3.000 4.000 11.000177908.000 1.000 4.000 2.000 7.000177909.000 7.000 6.000 1.000 9.000177910.000 6.000 4.000 3.000 11.000177911.000 5.000 7.000 5.000 9.000177912.000 10.000 7.000 3.000 10.000177913.000 3.000 1.000 0.000 10.000177914.000 6.000 4.000 2.000 10.000177915.000 2.000 0.000 1.000 12.000177916.000 4.000 3.000 4.000 15.000177917.000 4.000 4.000 2.000 13.000177918.000 8.000 15.000 4.000 13.000177919.000 4.000 6.000 2.000 10.000188901.000 1.000 1.000 1.000 13.000188902.000 1.000 2.000 0.000 17.000188903.000 1.000 1.000 0.000 11.000188904.000 1.000 1.000 0.000 10.000188905.000 5.000 5.000 1.000 7.000188906.000 1.000 2.000 1.000 12.000188907.000 6.000 1.000 1.000 5.000188908.000 7.000 5.000 2.000 9.000188909.000 2.000 1.000 2.000 14.000188910.000 6.000 4.000 0.000 2.000188911.000 4.000 9.000 1.000 6.000188912.000 5.000 5.000 1.000 3.000188913.000 3.000 6.000 0.000 4.000188914.000 5.000 5.000 0.000 2.000188915.000 1.000 5.000 0.000 0.000188916.000 3.000 3.000 1.000 4.000188917.000 4.000 4.000 0.000 7.000188918.000 10.000 7.000 4.000 10.000139ANIMAL PMN EDEMA BALT GOBLET188919.000 1.000 1.000 0.000 7.000188920.000 7.000 7.000 2.000 10.000188921.000 8.000 10.000 3.000 5.000195001.000 1.000 0.000 0.000 0.000195002.000 2.000 1.000 0.000 0.000195003.000 1.000 1.000 0.000 0.000195004.000 2.000 1.000 1.000 0.000195005.000 3.000 0.000 1.000 0.000197903.000 8.000 5.000 1.000 10.000197905.000 8.000 2.000 0.000 7.000197906.000 7.000 5.000 1.000 8.000197907.000 6.000 5.000 2.000 8.000197919.000 3.000 2.000 0.000 6.000197920.000 2.000 1.000 0.000 5.0001.0001.0001.0002.0001.0001.0001.0001.0002.0001.0001.0001.0002.0002.0001.0001.0001.0002.0002.0002.0001.0002.0001.0002.0001.0001.0001.0002.0001.0001.0002.0001.0001.0001.0002.0002.0001.0001.0002.0002.0001.0001.0001.0001.0001.0001.0001.0001 .0001 .0001.0002. .0001.0001.0001.0002.0001.000140ANIMAL^CULTURE^TEM^IMMUNO^RT-PCR^174203.000^1.000174206.000 1.000174211.000^1.000174212.000 1.000174218.000^1.000174219.000 1.000175501.000^1.000175502.000 1.000175503.000^1.000175504.000 2.007175505.000^1.000175506.000 1.000175507.000^2.006175508.000 2.006175509.000^2.004175510.000 1.000175511.000^1.000175512.000 2.005175513.000^2.005175514.000 2.005175515.000^1.000175516.000 1.000175517.000^1.000175518.000 2.007175519.000^1.000175520.000 2.007179101.000^1.000179104.000 2.004179107.000^1.000179110.000 2.006179113.000^1.000179114.000 1.000179115.000^2.005179116.000 2.007179117.000^1.000179118.000 1.000179119.000^1.000179120.000 2.006179128.000^1.000179129.000 1.000179132.000^2.006179133.000 2.006179136.000^1.000179137.000 1.000179138.000^2.005179139.000 2.005183001.000183002.000183003.000183004.000183005.000183006.000141ANIMAL CULTURE INIMUNOTEM183007.000 1.000183008.000 1.000183009.000 1.000183019.000 2.000183020.000 2.000183021.000 2.000183022.000 2.000183023.000 1.000183032.000 1.000183033.000 1.000183034.000 2.000183035.000 1.000183036.000 1.000177901.000 .1.000 1.000177902.000 1.000 1.000177903.000 1.000 1.000177904.000 2.006 1.000 1.000177905.000 1.000 2.000177906.000 1.000 1.000177907.000 1.000 1.000177908.000 1.000 2.000177909.000 2.011 .1.000 2.000177910.000 1.000 1.000177911.000 1.000 1.000177912.000 1.000 2.000177913.000 1.000 1.000177914.000 1.000 2.000177915.000 1.000 1.000177916.000 1.000 1.000177917.000 1.000 1.000177918.000 1.000 1.000177919.000 1.000 2.000188901.000 1.000 1.000188902.000 1.000 1.000188903.000 1.000 1.000188904.000 1.000 1.000188905.000 1.000 1.000188906.000 1.000 1.000188907.000 1.000 1.000188908.000 1.000 1.000188909.000 1.000 1.000188910.000 1.000 2.000188911.000 1.000 1.000188912.000 1.000 1.000188913.000 1.000 1.000188914.000 1.000 1.000188915.000 1.000 1.000188916.000 1.000 1.000188917.000 1.000 1.000188918.000 1.000 1.000188919.000 1.000 1.000188920.000 1.000 1.000RT-PCR142ANIMAL CULTURE IMMUNO RT-PCRTEM188921.000 1.000 1.000195001.000 1.000 1.000 1.000195002.000 1.000 1.000 2.000195003.000 1.000 1.000 2.000195004.000 1.000 1.000 2.000195005.000 1.000 1.000 1.000197903.000 2023.000 1.000 2.000 2.000197905.000 1298.000 1.000 1.000 2.000197906.000 1554.000 1.000 2.000 2.000197907.000 1813.000 2.000 1.000 2.000197919.000 0.000 1.000 1.000197920.000 0.000 1.000 1.000143APPENDIX B: STATISTICAL ANALYSES OF AIRWAY HISTOLOGICAL SCORESDAY 6 STUDY: JUVENILE GUINEA PIGS, LOW DOSE RSV (EXPT # 1755)RANK SUM, n EPITH MONO PMN EDEMA BALT GOBLETRSV 10 139 129 132 126 125.5 117.5CONTROL , 9 51 61 58 64 64.5 72.5Mann-WhitneyU statistic6 14 13 16 19.5 27.5,^p value 0.001 0.01 0.008 0.018 0.035 0.15DAY 6 STUDY: JUVENILE GUINEA PIGS, HIGH DOSE RSV (EXP'T # 1791)RANK SUM n EPITH MONO PMN EDEMA BALT GOBLETRSV 10 153 145.5 150.5 118 124.5 127CONTROL 10 57 64.5 59.5 92 85.5 83Mann-WhitneyU statistic2 9.5 4.5 37 30.5 28p value- 0.0001 0.002 0.001 0.32 0.13 0.09DAY 6 STUDY: ADOLESCENT GUINEA PIGS, LOW DOSE RSV (EXPT # 1830)RANK SUM n EPITH MONO PMN EDEMA BALT GOBLETRSV 10 91.5 102 123.5 103 97 80.5CONTROL 9 98.5 88 66.5 87 93 109.5Mann-WhitneyU statistic53.5 43 21.5 42 48 64.5,^p value 0.46 0.86 0.053 0.80 0.80 0.11144DAY 14 STUDY: JUVENILE GUINEA PIGS, LOW DOSE RSV (EXPT # 1779)RANK SUM n EPITH MONO PMN EDEMA BALT GOBLETRSV 10 107 111.5 122.5 116.5 104.5 85CONTROL 9 83 78.5 67.5 73.5 85.5 105Mann-WhitneyU statistic38 33.5 22.5 28.5 59.5 60p value 0.55 0.33 0.056 0.17 0.22 0.21DAY 60 STUDY: JUVENILE GUINEA PIGS, LOW DOSE RSV (EXPT # 1889)RANK SUM n EPITH MONO PMN EDEMA BALT GOBLETRSV 10 121.5 119.5 146 123.5 136 111.5CONTROL 11 109.5 111.5 85 107.5 95 119.5Mann-WhitneyU statistic43.5 45.5 19 41.5 29 53.5p value 0.41 0.49 0.015 0.33 0.051 0.92DAY 6 CONTROL JUVENILE GUINEA PIGS (EXPT # 1755) VS. "HISTORICAL"CONTROLS (EXPT # 1742)RANK SUM n EPITH MONO PMN EDEMA BALT GOBLETDAY 6 10 77.5 89 94 86.5 96 89.5HISTORIC 6 58.5 47 42 49.5 40 46.5Mann-WhitneyU statistic37.5 26 21 28.5 19 25.5p value 0.40 0.64 0.32 0.87 0.21 0.62145DAY 6 JUVENILE GUINEA PIGS: HIGH DOSE RSV VS. LOW DOSE RSV(RSV-INOCULATED ANIMALS FROM EXPT #s 1791 AND 1755)RANK SUM n EPITH MONO PMN EDEMA BALT GOBLETHIGH DOSE 10 115.5 84.5 90.5 84.5 83 73LOW DOSE 10 94.5 125.5 119.5, 125.5 127 137Mann-WhitneyU statistic39.5 70.5 64.5 70.5 72 82p value 0.42 0.12 0.27 0.12 0.09 0.015DAY 6 JUVENILE GUINEA PIGS: CONTROLS FROM HIGH DOSE VS. LOW DOSERSV STUDIES (CONTROL ANIMALS FROM EXFT #s 1791 and 1755)RANK SUM n EPITH MONO PMN EDEMA BALT GOBLETHIGH DOSE 10 126.5 122.5 80 124 105.5 84LOW DOSE 10 83.5 87.5 130 86 104.5 126Mann-WhitneyU statistic28.5 32.5 75 31 49.5 71p value 0.10 0.18 0.06 0.15 0.97 0.11


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items