Experimental Rhinovirus 16 Infection Potentiates Histamine Release after Antigen Bronchoprovocation in Allergic Subjects1- 3

WILLIAM J. CALHOUN, CHERI A. SWENSON, ELLIOT C. DICK, LAWRENCE B. SCHWARTZ, ROBERT F. LEMANSKE, JR., and WILLIAM W. BUSSE

Introduction

\liral upper respiratory infections cause wheezing in many patients with asthma (1, 2). Although the respiratory viruses associated with these infectious asthma exacerbations have been identified in many instances, the mechanisms by which these illnessescause changes in airwayresponsivenessand function have yet to be fully determined (3). Nonetheless, a number of mechanisms have been proposed to explain virus-provoked asthma, including airway injury by the infectious pathogen (4), production of virusspecific IgE antibodies (5), diminished 13-adrenergic function (6, 7), loss of epithelial structure and function (8, 9), and enhanced leukocyte histamine release (10). More recently, we have found that an experimental respiratory viral illness with rhinovirus 16(RVI6) increased both airway responsiveness and the propensity to develop a late asthmatic reaction (LAR) following antigen bronchoprovocation in allergicrhinitis patients (11). Because LAR provides a model for chronic inflammation in asthma, these latter observations suggest possible important relationships among viral respiratory infections, airway inflammation, and airway hyperresponsiveness (12). The LAR has many features in common with chronic asthma, including more persistent airway obstruction, diminished bronchodilator response, and enhanced bronchial responsiveness (12). Although the precise mechanisms of LAR remain to be established, pulmonary mediator release and inflammatory cell recruitment are likely participants in this process. The purpose of this study was to expand on previous observations by determining the profile of histamine and tryptase release into the peripheral circulation as markers of pulmonary mast cell secretion and to establish the relationship of any observed change with

SUMMARY Viral respiratory infections exacerbate asthma in many patients. Wehypothesized that one mechanism by which this effect occurs may Include potentisted or altered mediator release by mast cells and/or basophlls to tavor the development of late-phase asthmatic reactions (LAR). Therefore, we studied eight subjects with allergic rhinitis before and during an experimentally Induced rhinovirus 16 (RV16) Infection. We determined levels of plasms histamine and tryptase, and we observed the associated patterns of airway obstruction that developed following Inhaled antigen challenge. Bronchial responSIYene88 to histamine, methacholine, and antigen were all significantly Increased during the RV16 Illness. Further, the Incidence of LAR was significantly higher (five of eight) during the Infection than before (one of eight; p 0.014).In addition, In those petlents whose pattern of response following antigen challenge converted from an immediate response only before Infection to a dual response (Immediate + late phase) during Infection, plasma histamine eoneentratlons after challenge were significantly greater than In those whose pattern of response did not change. We conclude that one mechanism by which RV16 infection Increases the likelihood of LAR could include enhanced mediator release from pulmonary mast cells or from circulating or recruited basophlls. AM REV RESPIR DIS 1991: 144:1287-1273

=

the development of LAR following antigen challenge during an experimental RV16 infection. Methods Subjects The eight male subjects selected for study had an average age of 33 ± 5 yr (mean ± SEM). All subjects had a positiveimmediate skin test by the prick method (~ 3 mm wheal at 15 min) to an extract of ragweed antigen (1:20 wt/vol; Greer Laboratories, Lenoir, NC) and had no serum neutralizing antibody « 1:1 agonist; 5 to 32 tissue culture infective dose 50070 TCID so) to RVI6. None of the patients smoked, had active clinical asthma, or used any medication at the time of study. All patients gave informed consent before enrolling in the study, which was conducted during periods when the subjects were free of allergic rhinitis symptoms. For reasons outlined in a previous study (11), we chose to evalute the effects of a viral respiratory illness in allergic rhinitis subjects rather than asthma patients. First, airway responsiveness in allergic rhinitis patients is often enhanced compared with normal subjects even though clinical asthma does not exist (13). Second, in allergic rhinitis, it would be unlikely to find peribronchial fibrosis (14)or abnormal airway smooth muscle with characteristics similar to those in asthma (e.g.,

hypertrophy or inflammation) to complicate the effects that a respiratory illness may have on pulmonary function. Finally, none of the patients required medication, that is, antiasthma treatment, that could interfere with changes in pulmonary function or response to the viral illness.

Study Design The patients were studied on three separate occasions, each at least 28 days apart: baseline (screening), precold, and acute RVI6(figure I). At each study period, the patients were evaluated on 2 consecutive days by spirome-

(Received in original form February 15, 1991 and in revised form June 17, 1991) 1 From the Departments of Medicine, Pediatrics, and Preventive Medicine, University of Wisconsin, Madison, Wisconsin, and the Department of Medicine, Medical College of Virginia, Richmond, Virginia. 2 Supported in part by Grant Nos. AI 26609, K08HL-0l828, AI 15685, K07-AI00995, andAI-20487, and Grant No. UW GCRC-RR03186 to the General Clinical Research Unit from the Nationallnstitutes of Health. 3 Correspondence and requests for reprints should be addressed to William J. Calhoun, M.D., Assistant Professor of Medicine, H6/384 Clinical Sciences Center, University of Wisconsin-Madison, Madison, WI 53792.

1267

1268

CALHOUN, SWENSON, DICK, SCHWARTZ, LEMANSKE, AND BUSSE

I

Allergic Rhinitis Patient

I

I Clinically Quiescent

I

ExclUde or Walt

I Inoculate with RV16

I

Wait 48 hours

I BaselineStudies

Acute RY16

Pre-cold

Day 1 Hlstamlne BP Day 2 AntIgen BP to determIne PD20

Day 1 HIstamIne BP MethacholIne BP

Day I HIstamIne BP MethacholIne BP

Day 2 AntIgen BP wIth PD20 dose lrnmedtate » Late Plasma Histamine

Day 2 AntIgen BP wlth PD20 dose trnrnectate » Late Plasma Hlstamlne

Waft for return

of normal airway

reacttvttv (4 weeks)

f---

Walt 4 Weeks

I

Fig. 1. Experimental design for RV16 inoculation. Bp, bronchoprovocation; PD", provocative dose for a 20% fall in FE"".

try (forced expiratory volume in 1 s, FEV., and forced vital capacity, FVC) and determination of nonspecific airway responsiveness. At the initial screening visit, baseline studies of spirometry and airway responsiveness to histamine were obtained. An inhalation antigen challenge was conducted 1day later with an extract of ragweed antigen (Greer Laboratories) to determine the provactive dose necessary to reduce the patient's FEV. by 20010 (PD 20 ) . Approximately 4 wk later, after airway responsivenessreturned to the initial baseline value, "precold" tests of airway responsiveness were determined and served as the comparative values for observations during the acute RV16 illness. Furthermore, measurement of airway responsiveness to histamine was repeated approximately 1 wk before virus inoculation to ensure a consistent baseline level of airway reactivity. On the day of antigen challenge, during both precold and cold phases, an intravenous catheter was placed in an antecubital fossa vein before the inhalation procedures werebegun. At least 10 min later, four 5-ml blood samples were obtained to determine the baseline concentrations of plasma histamine and tryptase, A calculated ragweed antigen PD 20 concentration (determined from the cumulative antigen dose-response curve at baseline testing) was then administered in a five-breath bolus, and replicate samples of blood were drawn approximately every 2 min for 15 min and again at 30, 60, and 360 min. The blood was immediately anticoagulated with 50 /11 of a 0.5 M EDTA solution, placed in ice, and plasma separated by centrifugation. The plas-

rna was frozen (- 80 0 C) and later analyzed for histamine and tryptase.

Bronchial Provocation 'Jesting Airway responsiveness to histamine and methacholine was determined by standard bronchial provocation techniques to increasing concentrations of either histamine or methacholine, which were delivered through a nebulizer (DeVilbiss, Somerset, PA) attached to a French-Rosenthal dosimeter (15). Bronchial provocation was performed by having the subject inhale from functional residual capacity to vital capacity with a lO-s breath hold. Histamine (0.03, 0.06, 0.12, 0.25, 0.5, 1.0, 2.0, 5.0, 10.0, and 25.0 mg/ml) and methacholine concentrations (0.08, 0.16,0.32, 0.64, 1.25,2.5, 5.0, 10.0,and 25.0 mg/ml) were prepared from stock solutions and diluted in buffered saline (pH 7.4). If an initial aerosol of buffered saline did not significantly change the baseline FEV, (none did), increasing concentrations (five breaths each) of either histamine or methacholine were given at 5-min intervals until the FEV, fell at least 20010 from the baseline measurement and was sustained at this level. Spirometry was performed on a spirometer (Puritan-Bennett, Wilmington, MA), with the best of three efforts for the FEV, selected to construct a logarithmic doseresponse plot. From this plot, the dose of either histamine or methacholine required to produce a 20% decrease in FEV, was calculated by linear interpolation and defined as the PD20 for that agonist. Since concomitant measurements of airway mediator release and pulmonary function are

often not technically feasible, we developed methods to assess mediator release with measures of circulating plasma histamine (16).At the initial screening visit, the cumulative dose of antigen required to decrease the FEV, by 20% was determined (antigen PD 2 0 ) . Approximately 4 wk later, the patient was rechallenged, but at this time the entire predetermined cumulative PD 20 ragweed antigen dose was administered in five consecutive inhalations. From an intravenous catheter, sequential blood samples were drawn to measure plasma histamine and tryptase; spirometry was performed at 5, 10, and 15 min after administration of antigen. Using this procedure we found a relationship between the rise in plasma histamine and the development of airway obstruction (16). Furthermore, these studies provided evidence that the rise in plasma histamine probably reflects pulmonary mediator release. Consequently, this procedure permitted evaluation of the effect of RVl6 on the airway response to a predetermined dose of antigen and pulmonary mediator release.

Late-phase Responses Late-phase reactions were defined as a decrease in FEV, ~ 15% from baseline, which occurred 3 h or more following antigen challenge and was preceded by the development and resolution of airway obstruction immediately following challenge. Determination of Plasma Histamine Plasma histamine was measured using a radioenzymatic assay as described previously (17). This assay uses histamine N-methyltransferase and S-adeno-syl-C-['H-methyl)methionine to convert histamine to [3H]-telemethylhistamine, which is isolated from other radiolabeled material and quantitated by liquid scintillation. Standard curves were prepared for each experiment with histamine concentrations from 15to 5,000 pg/ml (linear coefficient of correlation equal to 0.99 in this range) and used to calculate the plasma histamine concentration. Determination of Plasma Tryptase 'Iryptase was measured by a sandwich enzymelinked radioimmunoassay based on modification (18) of a previously described procedure (19). The murine monoclonal antibody G5, for capture, and the biotinylated murine monoclonal antibody G4, for detection, were used. The immunoassay has a lower limit of detection of 0.05 ng tryptase/40 /11 sample. RV16 Inoculation The technique for rhinovirus inoculation followed previously described methods (11, 20). Briefly, all patients recruited for study had no serum neutralizing antibody to RV16(see Subjects). Rhinoviral "colds" were induced in the eight donors by instilling, on 2 successive days, 0.25 ml RV16suspension into each nostril by pipette (320 to 3,200 TCID.o) and then spraying approximately the same amount into each nostril with an atomizer (No. 286,

INCREASED HISTAMINE RELEASE DURING RHINOVIRUS 16 INFECTION

1269

Effects of RV Infection on Pulmonary Function and Airway Responsiveness

TABLE 1 CLINICAL AND LABORATORY RESPONSE TO RV16 INOCULATION Highest Symptom Score

Highest Virus Titer

Highest Mucous Index

Convalescent Titers

There was no difference in FEV 1 values during precold (98 ± 3070 of predicted) 16 31,600 1 0.75 8 and cold periods (98 ± 4%, table 2). 2 6 10,000 4 0.5 However, there was a significant increase 316,000 3 9 2.0 > 45 23 4 1,000,000 3.0 45 in airway responsiveness to both inhaled 8 1,000 5 1.0 2.8 histamine and methacholine (figure 2 and 8 3,160 NT' 22.5 6 table 2). The relationship and signifi13 31,600 7 22.5 1 cance of changes in airway responsive8 6 31,600 45 2 ness to the RVl6 infection is further sub• Not tested. stantiated by the consistency of histamine POlO (cumulative breath units) values in these patients. Histamine POlO values DeVilbiss), poweredby a compressor (561 se- = one part serum and one part diluent, and (geometric means) during the baseline and screening period (40.0) and again imries; DeVilbiss). The coarse spray coated the so on. mediately before virus inoculation (53.9) nasal cavitywell, as visualized bya head lamp. Colds wereevaluated with a questionnaire were identical to precold values (47.6) but Statistical Analysis that wascompleted bythe participant hourly Data were analyzed using a data base and significantly (p < 0.05) different from during waking hours (20). Symptoms evalustatistics package for microcomputers (Ab- histamine POlO at cold. Only one patient ated included cough, nasal discharge, sneezstat® 5; Anderson-Bell,Parker, CO) and were (subject I) failed to show an increase in ing, stuffy nose, headache, malaise, chills,or expressed as the mean ± the standard error bronchial reactivity to both histamine fever. Symptomsweregraded0 = absent, 1 = of the mean. Comparisons between data and methacholine. mild, 2 = moderate, or 3 = severe. Total obtained before cold and during acute RV16 Compared to precold values, there was symptom scoresof ~ 12defined a severe cold, infection were made using a paired t test or a significantly greater fall in FEV 1 to the 7 to 11 a moderate cold, and < 7 a mild cold. the nonparametric analog (Wilcoxon test). same inhaled dose of ragweed antigen Differences were considered significant for during the cold, 35.0 ± 5.2 versus 21.8 Laboratory Confirmation of a p < 0.05 (21). ± 4.5% at the baseline period, p = 0.02 Viral Infection (figure 3). The greater fall in FEV 1 to anVirus shed by the inoculated subjects wasastigen during the infection appeared to sayedby titration of a nasal wash into diploid Results reflect the relative increase in airway re(WI-38)cellcultures. Nasal washes weretakResults of RV16 Inoculation en by instilling 5 ml of Hanks' balanced salt sponsiveness to histamine and methachosolution (HBSS)with 0.50/0 gelatin into each All eight patients inoculated with RVl6 line during RVI6 illness. The single exnostril, holding for 5 s, and then expelling developed a respiratory illness as indicat- ception was Patient I, who showed no the fluid into a sterile petri dish. Nasal wash- ed by the presence of symptoms and RVl6 change in airway responsiveness to either es wereinoculated into WI-38, primary rhe- identified in nasal washes (table I). Furhistamine or methacholine during the sus monkey, and Hep-2cellcultures (20). The thermore, seven of the eight patients had cold; however, his drop in FEV 1 to antietiology of a viral illness was established by a four fold or greater increase in RVI6gen was greater during the active RVI6 at least one identified isolate or by a fourfold or greater increase in neutralizing antibody specific antibody in convalescent sera. illness compared to values before inoculain the convalescent serum specimens. Anti- Based upon these data, all eight subjects ton, 21% versus 4%, respectively. The airbody titers wereexpressedas initial serum di- were considered to have developed a RVI6 way response to inhaled antigen (at both lutions in which 1:1 = undiluted serum, 1:2 illness at the time of the study. precold and cold) was to a calculated dose Patient

TABLE 2 EFFECTS OF RV16 ILLNESS ON PULMONARY FUNCTION AND AIRWAY RESPONSIVENESS Precold Spirometry Patient

Age

FEV, (L)

1 2

24

3 4

23 53 22 55 25 19

4.15 3.65 4.11 3.56 5.19 3.49 3.76 4.26 4.02 0.19

44

5 6 7 6 Mean

SEM

33 6

Cold Antigen'

PO,.

% of Predicted

HIS

98 95 99 101 114 92 89 95 98 3

12.2 5.5 46.5 294 231 121 63.8 17.1 47.6

-t

METH

Imm

Late

6.8 6.7 28.0 63.4 287

4.0 20.1 12.7 20.0 20.0 24.3 24.4 48.9 21.8 4.5

0.0 0.0 4.0 9.0 0.0 3.5 0.0 24.1

44 300 11.4 40.5

_t

Spirometry

Antigen'

PO,.

FEV, (Ll

% of Predicted

4.16 3.61 4.34 3.68 5.30 3.20 3.72 4.35 4.04 0.22

98 94 104 105 116 85

88 97 98 4

HIS

METH

Imm

Late

29.0 2.9 27.5 54.0 108 23.9 38.0 8.2 23.6

13.7 6.5 9.6 20.0 52.0 4.3 44.0 5.0 12.9

21.0 31.4 52.0 27.8 12.8 44.6 35.8 55.0 35.0 5.6

19.0 20.5 15.0 17.0 7.0 3.2 0.0 26.8

_t

-t

.DrItInftion of abbreviations: PO" = Provocative dose to cause a 20% fall in FEV, expressed as cumulative breath units (CaU), where 1 au - 1 breath of 1 mg/ml; HIS = histamine PO", IlII:TH - methacholine PO". cso: Imm = immediate fall in FEV, following antigen challenge; Late = late (4 to 6 h) fall in FEV, following antigen challenge. ; Fall in FEV, following antigen inhalation. Geometric mean: no SEM.

cau;

CALHOUN, SWENSON, DICK, SCHWARTZ, LEMANSKE, AND BUSSE

1270

PRECOLD

(j] I--

O~

2.00

E2000

p·0.025 ,.-----,

Q,

"'I-g;5

W1500

1.50

>!Il

~W 1.00

;'! 1000

0!ii

I

U~

1

:> ~

(/)

~~050 ~

90

Z

~

W