Bronchial hyperresponsiveness and cellular infiltration in the lung of guinea-pigs sensitized and...

Clinical and Experimental Allergy. 1991. Volume 21. pages 67-76 ADONIS 09602178910001 IK

Bronchial hyperresponsiveness and cellular infiltrationin the lung of guinea-pigs sensitized and challengedby aerosol

ELISABETH BOICHOT. V. LAGENTE, C. CARRE, P. WALTMANN,J. M. MENCIA-HUERTA and P. BRAQUET

Institut Henri Beaufour, 1, avenue des Tropiques. Les Ulis. France

Summary

We have studied the development of airway hyperresponsiveness and the pulmonary cellinfiltration in a guinea-pig model in which both initial sensitization and subsequentexposure to the antigen were performed by aerosol. Enhanced bronchopulmonaryresponse to aerosol administration of acetyicholine (ACh) and 5-hydroxytryptamine(5-HT) was observed 3 4 hr and 18-24 hr after antigen exposure of sensitized animals.In contrast, when ACh and 5-HT were administered intravenously 3-4 hr after thechallenge, no significant alteration of the dose-response curves was observed. However,18-24 hr after antigen challenge, a marked leftward shift of the dose response curve wasobserved on intravenous injection of ACh or 5-HT. The increased bronchial reactivityto aerosolized ACh in sensitized and challenged guinea-pigs reached a maximum bydays 2-4. was still significantly increased at day 5 and returned to the basal value by day8. No further alteration of the dose-related bronchopulmonary response to aerosol orintravenous administration of ACh was recorded 24 hr after a second antigen challenge,performed 8 days after the initial one. The analysis of bronchoalveolar lavage fluidsshowed a significant increase in the number of polymorphonuclear neutrophils 3-4 hrafter the exposure of sensitized animals to the antigen, which was also associated with asignificant eosinophilia at 18 24 hr. Histologica! examination o'l lung specimensobtained from animals 3-4 hr following challenge demonstrated eosinophil infiltrationin the peribronchial regions and bronchial walls, as well as within the epithelium.Furthermore, as compared to time 3-4 hr, less eosinophils in the peribronchial urea andsubmucosa were counted 24 hr after antigen challenge. However, a role of eosinophil-derived products in the development ofbronchial hyperresponsivenss in this experimen-tal model remains to be established.

Clinical and Experimental Allergy, Vol. 21, pp. 67-76. Submitted 14 March 1990; revised24 July 1990; accepted 17 August 1990.

Introduction

Non-specific increase in airway responsiveness tobronchoconstrictor stimuli, i.e. airway hyperresponsive-ness, is considered a central feature of asthma. Indeed,allergen inhalation resulting in early and late asthmaticresponses has generally been associated with an increasein airway reactivity [1]. Although clinically well charac-terized, the mechanisms involved in the development of

Correspondence: Dr Jean-Michel Mencia-Huerta. Department ofImmunology. Instilut Henri Beaufour, I. avenue des Tropiques. 91952Les Ulis Cedex. France.

airway hyperresponsiveness are poorly determined andthe elucidation of the events underlying this phenomenonare prerequisite to understand better the pathologicalbasis of asthma. Various animal models have beendeveloped in which both early and late airway responsesassociated or not with bronchial hyperresponsivenessafter exposure to antigen are observed. Hutson et al. [2]have demonstrated that 2 hr after antigen challenge,sensitized guinea-pigs present an early fall in airwayconductance followed by a return to the baseline. Inaddition, broncbopulmonary alterations are alsoobserved 17 and 72 hr after challenge. These changes in

67

6S Elisabeth Boichot et al.

bronchopulmonary functions are accompanied by anincrease in the number of neutrophils in the broneho-alveolar lavage, 6 hr after the antigen challenge andfollowed by a dramatic intlux of eosinophils after 17 hrwith a maximum at 72 hr [2]. These changes in the cellcomposition of bronchoalveolar lavages resemble tboseoccurring after allergen challenge in asthmatic patients[3]. However, the role of the cell infiltration in the lungtissue in the development of the bronchial hyperrespon-siveness has not been elucidated.

To assess bronchial reactivity, most investigators haveusually administered the spasmogenic substances by theintravenous route [4.5]. This leads to a systemic responsewhere the pulmonary and vascular components areintricated and is thus difficult to dissociate. For thesereasons, the direct administration of the antigen in theairways would eliminate, or at least markedly reduce, thesystemic effects [reviewed in 6]. Therefore, we developed amodel in which guinea-pigs are sensitized and challengedwith the antigen by aerosol. In the present study, thebronchopulmonary responses indueed by aerosol andintravenous administrations of acetyicholine (ACh) and5-hydroxytryptamine (5-HT) at various time intervalsfollowing antigen challenge without prior pharmacologi-cal modulation were compared. In addition, the cellcomposition of bronchoalveolar lavages and the histo-iogical examination of lung specimens from sensitizedand challenged animals was analysed.

Materials and methods

Materials

The following drugs were used: ovalbumin (OA, chiekenegg. grade 5, Sigma. St Louis, U.S.A.). Uretbane (ethyl-carbamate, Prolabo, Paris, France), pancuronium bro-mide (Pavulon. Organon, Fresnes, France), ACh chloride(Sigma), 5-HT chloride (Sigma), EDTA (Sigma). Hanks"balanced salt solution containing 0 35 g/I sodium bicar-bonate (Flow laboratories, Irvine, U.K.).

Sensitization procedure

Specific pathogen-free male Hartley guinea-pigs (350-400 g) (Charles River, St-Aubin les Elbeuf, France) wereplaced twice, after a 48-hr interval, in a plexiglasschamber (30 x 50 x 30 cm) and exposed to aerosols of a0 9''';i NaCl solution in saline solution containing 2 mg/mlOA for 30 min. modified from [7]. The aerosol wasgenerated by a Devilbiss ultrasonic nebulizer (Ultra-Neb99, Sommerset. Pennsylvania. U.S.A.). Under theseconditions the aerodynamic diameter of the homo-geneated particles averaged 0 5 3 /jm.

Antigen expo.sure

Fifteen to 20 days after the initial sensitization procedure,tbe guinea-pigs were challenged in the plexiglasschamber, by exposure to live successive solutions of OAof. respectively 10 /(g/ml, 100 /Jg/ml. 1 mg/ml, 5 mg/mland 10 mg/ml for 15 min each. The aerosol was generatedby a Devilbiss ultrasonic nebulizer as described above.These increasing concentrations were used to avoid fatalanapbylactic reactions. Indeed, when the concentrationof OA reached I mg/ml in the nebulizer, the animalspresented labored breathing, coughing and rubbing of thenose. In addition, a few guinea-pigs experienced convul-sions, although animal death was never observed. How-ever, 1 hr after the last challenge with the highestconcentration of OA, no apparent sign of respiratoryfailure was observed in the guinea-pigs. As a controlgroup, sensitized guinea-pigs were exposed to an aerosolof the saline solution for 30 min.

Assessment of the in-vivo bronchopulmonary reactivity

In two different sets of experiments the bronchopulmon-ary reactivity was assessed either 3 4 hr or 18 24 hr afterexposure of the animals to the antigen or saline. Theguinea-pigs were anaesthetized with urethane (12 g/kg,i,p.) and placed in a dorsal recumbent position. Thetrachea was cannulated and the lungs mechanicallyventilated with a constant volume of I ml laboratory air/100 g body weight with a respiratory pump (Ugo Basilc,Varese, Italy; 60 brcaths/min). The right jugular vein wascannulated for intravenous administration of the variousdrugs. Spontaneous breathing was abolished with pan-curonium bromide (2 mg/kg, i.v.) and both cervical vaginerves were transected at the level of the neck. The initialresistance to inflation was adjusted to 10 cm H.̂ Oaccording to the method of Konzett und Rossler 18] andthe excess air volume was monitored from a lateral port ofthe ventilator circuit with a bronchospasm transducer(model 7020. Ugo Basile).

After a 10 min equilibration period, four successiveI-min aerosol administrations of ACh (50, 100. 200 und500 /ig/ml) or 5-HT (3. 10, 30 and 60 ;jg/ml) wereperformed at 10 min intervals with constant monitoringof the bronchopulmonary response. The aerosol wasgenerated by a Devilbiss "Pulmosonic" ultrasonic nebu-lizer permanently connected in series with the afferentlimb of the ventilator circuit as modified from [9]. Thebronchopulmonary responses are expressed as percen-tage changes calculated over the lOO'Si obtained byclamping the trachea at the end of the experiment. Thepercentage of the bronchopulmonury response is used asan index of intrathoracic airway calibre. In parallelexperiments, the bronchopulmonary responses induced

Bronchial hyperresponsivene.%s in aero.sol-sensitized guinea-pigs 69

by four successive intravenous administrations of ACh(10, 30,60 and 100 ;jg/kg) or 5-HT(3,6, 10. and 30 /jg/kg)were assessed, 3-4 hr »nd 18 24 hr after the antigenchallenge. Similar bronchopulmonary reactivity to AChand 5-HT was observed when the guinea-pigs were used inthe time range of 18 hr or 24 hr after saline or OAchallenge exposure (data not shown).

Time-cour.se ofthe airway hyperresponsiveness to ACh

In this series of experiments, the dose of 100 /ig/ml AChwas selected because it induced a submaximal broncho-pulmonary response. The response to aerosol administ-ration ofthe single dose of ACh for 1 min was assessed indifferent groups of anaesthetized and ventilated guinea-pigs as described above, and 1, 2,4, 5, 8 and 10 days afterthe antigen challenge. As a control, the bronchopulmon-ary response to ACh in a group of guinea-pigs challengedwith a saline solution was also analysed throughout thetime-course study.

In separate experiments and 8 days after the initialantigen stimulation by aerosol, the animals were placed inthe plexiglass chamber and challenged again with the fourgraded doses of OA. Upon this second stimulation withthe antigen, the guinea-pigs presented respiratory symp-toms similar to those described above for the initialchallenge. The bronchopulmonary response foiiowingsuccessive aerosol or intravenous administration of AChwas assessed as described above, 18-24 hr after thissecond antigen exposure.

Determination of the cell composition of bronchoalveolarlavages

Bronchoalveolar lavages were performed 3-4 hr andIX-24 hr after antigen challenge or saline exposure.Briefly, the animals were anaesthetized with urethane. thetrachea cannulated and 5 ml of a warmed (37 C) salinesolution containing 2 6 mM EDTA was instilled into thelungs with a 5-mI syringe. The lavage fluid was recoveredby gentle aspiration and this procedure was repeated fivetimes. The combined lavage fluids from each animal wereadjusted to 30 ml with Hanks' balanced salt solution.After centrifugation at 250 xj? for 15 min, the super-natant was discarded and 9 ml of distilled water wasadded for 30 sec to the cell pellet in order to lyse theerythrocytes. Three millilitres of 0 6 M KCi solution werethen added to restore isotonicity and the volume wasadjusted to 50 ml with Hanks' solution. After a secondcentrifugation, the cell pellet was resuspended in 4 ml ofthe same solution and smears were prepared with acytocentrifuge (Shandon. Astmoor, U.K.). The smearswere stained according to the May-Grunwald Giemsa

procedure. Differential cell counts on at least 200 cellswere performed using standard morphological criteria toclassify the cells into neutrophils, eosinophils, macro-phages and lymphocytes. The cell composition of bron-choalveolar lavage fluids in a control group correspond-ing to non-sensitized and non-challenged guinea-pigs wasalso analysed.

Histologicai techniques

Lungs from anaesthetized guinea-pigs were removed 10min, 4 hr or 24 hr after antigen or saline exposure. Thelungs were washed by injection of 10 ml of a salinesolution through the pulmonary artery. The lung tissuewas fixed in a buffered lO'Yo formalin solution, embeddedin paraffin and 5-/nTi sections were obtained as previouslydescribed [18]. Small bronchial or bronchiolar walls weresectioned normal to the airway axis and stained withLuna's reagent specific for the eosinophil grimtile content[10]. Histologicai preparations were observed with a LeitzAristoplan microscope (Rueil-Malmaison, France) usinga X 64 magnification. For each guinea-pig, five to sixbronchi or bronchioles were randomly chosen and eosi-nophils were counted in a blinded fashion using acomputerized light analyser (Leitz).

Statistical analysis

All data are expressed as means ± s.e.m. Statistical differ-ences between the dose-response curves to ACh and5-HT in the groups of guinea-pigs receiving the varioustreatment were analysed by two-way analysis of variance.Thus, the whole dose response curve obtained in Ibesedifferent groups of animals was examined. In the othercases, the data were analysed and compared to therespective control by Student's /-test for unpaired data.

Results

Ejfect of aerosol challenge with the antigen on the responsi-veness to ACh and 5-HT

Exposure of anaesthetized guinea-pigs to four successiveaerosols of ACh (50. 100. 200 and 500 /(g/ml) induceddose-related bronchopulmonary responses (Fig. 1).When the animals were previously exposed to OA, thedose-response curve to ACh was significantly shifted tothe left, such enhanced response being observed at bothtime points studied, i.e. 3 4 hr and 18-24 hr (Fig. I). Inanother series of experiments, similar results wereobtained when dose-related bronchopulmonary res-ponses were induced with four successive aerosol admi-nistrations of 5-HT (3, 10, 30 and 60 /<g/ml) (Fig. 2). Withthis agonist, an even more dramatic leftward shift of thedose-response curve was observed, as compared to Ihe

70 Elisabeth Boichot et al.

_ 100 n

80 -

60 -

40 -

20 -

_ 100 n

10 100

Log ACh (|jg/ml)1000

Fig. t. Dosc-rclated bronchopuimonary response induced byfour successive administrations of ACh (50. 100, 200 and 500/ig/ml) by aerosol for I min to anaesthetized sensitized guinea-pigs. Bronchopulmiinary responses were assessed 3 4 hr (• .n = l) and 18-24 hr (A, H = 6) after antigen (OA) challenge. Thedata were compared by two-way analysis of variance to thecontrol obtained by exposure of the guinea-pigs to a salinesolution (• . H = 9). The results are expressed as means + s.e.m. ofpercentages of the maximal bronehopulmonary responseobtained by total clamping of the ventilation system. Statisticaldifferences between saline-exposed and OA-challenged guinea-pigs reached levels of P < 0 05 and P < 0 001 at 3-4 hr and 18-24hr. respectively.

one tioted with ACh. both after 3 4 hr and 18 24 hrfollowing the exposure of the animals to the antigen(Kig. 2). In contrast, the dose response curves establishedwith ACh or 5-HT administered by aerosol were notsignificantly altered 3-4 hr or 18-24 hr after aerosoladministration of saline to sensitized guinea-pigs. Asmore than four intravenous or aerosol administrations ofACh to guinea-pigs gave inconsistent results (possiblydue to lung function alterations), only the concentrationrange giving a linear response in saline-exposed animalswas investigated in the antigen-challenged ones.

In parallel experiments, the dosc-rclatcd bronchopul-monary responses following intravenous administrationsof ACh (10. 30. 60 and 100 /ig/kg) (Fig. 3) or 5-HT (3. 6,10 and 30/(g/kg) (Fig. 4) were analysed. In contrast to theresponses elicited by aerosol challenge no significantalteration of the dose response curves of both agonistswas recorded 3 4 hr after antigen exposure {Figs 3 and 4).A marked and significant leftward shift of the dose-response curve to intravenous administration of ACh(Fig. 3) or 5-HT (Fig. 4) was, however, observed 18 24 hrafter tbe exposure of animals to the antigen. The dose-response curves established with intravenous ACh or

i=oEZ3a.o.cucom

80 -

60 -

40 -

20 -

10

Log 5-HT

100

Fig 2. Dose-related bronehopulmonary response induced byfour successive administrations of 5-HT (3, 10. 30 and 60 ^g/ml)by aerosol for 1 min. to anaesthetized sensitized guinea-pigs.Bronchopulmonary responses were assessed 3-4 hr (• . ;) = 9)and 18-24 hr (A. n = 6) after antigen (OA) challenge. The datawere compared by two-way analysis of variance lo the controlobtained by exposure of the guinea-pigs to a saline solution(• , H—14). The results are expressed as means + s.e.m. ofpercentages of the maximal bronchopulmonary responseobtained by total elamping ofthe ventilation system. Statisticaldifferences between saline-exposed and OA-challengcd guinea-pigs reached levels o f / ' < 0 001. at 3-4 hr and 18-24 hr,

5-HT prior to, and 3-4 br and 18-24 br after aerosoladministration of saline to sensitized guinea-pigs weresimilar.

Time-course of bronchiat hyperresponsivene.ss to AChadministration

The time-course ofthe alterations ofthe broncbopulmon-ary response induced upon inhalation of a single dose ofACh (100 ;/g/ml) following aerosol challenge with OA orsaline solution was investigated. As presented in Fig. 5, nodifference of the bronchopulmonary response followingACh administration was observed in salinc-exposcdguinea-pigs al all time points of the kinetics study. Incontrast., tbe bronchopulmonary response followingaerosolized ACh in antigen-exposed guinea-pigs reacheda maximum by days 2 4 and was still significantlyincreased at day 5. as compared to the value obtained withsaline-challenged guinea-pigs. At days 8 and 10, thebronchopulmonary response of antigen-challengedsensitized guinea-pigs to ACh was comparable to valuesobtained in saline-exposed ones (Fig. 5). Also, no altera-tions of tbe dose-related bronchopulmonary response tointravenous ACb was observed 8 days after the exposureofthe animals to the antigen (Fig. 6a).

Bronchial hyperresponsiveness in aerosol-sensitized guinea-pigs 71

_ 120 n

100 -

ao -

60 H

^ 40 H

20 -

„ 100-1

10 100

Log ACh (pg/kg)

Kig. 3. Dose-related bronehopulmonary response induced byfour successive intravenous administrations of ACh (10. 30, 60and too /ig/kg) in anaesthetized sensitized guinea-pigs. Bron-chopulmonary responses wore assessed 3-4 hr (•./! = 6) and 18-24 hr (A. /I —5) after antigen (OA) challenge. The data werecompared by two-way analysis of variance to the controlobtained by exposure of the guinea-pigs to a saline solution(• , «=12). The results are expressed as means + s.e.m. ofpercentages of the maximal bronchopulmonary responseobtained by total clamping of the ventilation system. Statisticaldifferences between saline-exposed and OA-ehallenged guinea-pigs reached a level of /*<0 0I. at 18-24 hr. No statisticaldiftcrenee was recorded at 3-4 hr.

oo.

80

60 -

a.o

m

20 -

10 100

Log 5-HT (pg/kg)

Fig. 4. Dose-related bronchopulmonary response induced byfour successive intravenous administrations of 5-HT (3. 6. 10and 30 /ig/kg) in anaesthetized sensitized guinea-pigs. Broncho-pulmonary responses were assessed 3 4 hr (•./i = 6) and 18-24hr (A. n = l) after antigen (OA) challenge. The data werecompared by two-way analysis of variance to the controlobtained by exposure of the guinea-pigs to a saline solution( • , « = 6). The results are expressed as means + s.e.m. ofpercentages of the maximal bronchopulmonary responseobtained by total clamping of the ventilation system. Statisticaldifferences between saline-exposed and OA-challenged guinea-pigs reached a level of / '<0-0I. at 18-24 hr. No statisticaldifference was recorded at 3-4 hr.

These results demonstrate that the increase in broncho-pulmonary responsiveness induced by aerosol challengeof actively sensitized guinea-pigs is non-specific and oflimited duration. The possibility that a second antigenchallenge, performed 8 days after tbe initial one couldinduce an increase in the ACh-induced bronchopulmon-ary response was investigated. No further alteration ofthe dose-related broncbopulmonary responses to intra-venous (Fig. 6a) or aerosol (Fig. 6b) administration ofACh was recorded 24 hr after this second antigenchallenge.

Determination of the cell composition of bronchoalveolarlavages

Exposure of sensitized guinea-pigs to a saline solution didnot significantly modify the total cell number, or those ofpolymorphonuclear neutrophils. eosinophils. macro-phages and lymphocytes in the broncboalveolar lavagetluid (Table 1). The total number of recovered cells inbronchoalveolar lavage fluids was already increased after3-4 hr, but only significantly 18-24 hr after antigenchallenge by aerosol. However. 3-4 br after exposure ofthe animals to tbe antigen, a significant increase in tbenumber of neutrophils was observed (Table 1). Tbis

_ 100 n

BO -

40 H

20 -

0

II

I Ji 1

0 1 2 4 5 8 10

Time interval following antigen or saline exposure IDay]

Fig. 5. Time-eourse of the bronchopulmonary response inducedby administration by aerosol for 1 min ofaeetylcholine (ACh.100 /(g/ml) in anaesthetized sensitized guinea-pigs, challengedby saline (D) or OA exposure (E). Signitieanl increases of thebronchopulmonary response were observed until 5 days afteraniigen exposure, as analysed by Student's f-test compared tothe saline-exposed guinea-pigs. * P< 0 05; *• /* < 0 01;***/'<OOOI (« = 5-6 in eaeh group).

72 Etisabeth Boichot ct al.

_ 100-1

80-

6 0 -

40-

2 0 -

(a)

10 100

00 "1

80 -

60 -

20 -

rt -

(b)

Log ACh IMg/kg)

I

to 100 1000

Log ACh

Fig. 6. Dose related bronchopulmonary response induced bylour successive intravenous (a) or aerosol (b) administrations ofacetylcholine (ACh) in anaesthetized sensitized guinea-pigs. Thedose response curves to ACh established 10 days after the firstantigen exposure (•) (a) or 18-24 hr after the second antigenexposure (A) (a and b), were not statistically different from thedose response curve obtained by exposure of the guinea-pigs toa saline solution (•) . as analysed by two-way analysis ofvariance. The results are expressed as means + s.e.m. of thebronchopulmonary response as percentages of the maximalresponse obtained by total clamping of the ventilation system{/I = 5-7).

marked neutropbilia in tbe lavage fluid was still observed18 24 hr after antigen exposure but was accompanied atthis latter time by a significant 10-fold increase in thenumber of eosinopbils. No significant cbange in tbeabsolute numbers of recovered macrophages and lym-phocytes was noted 3 4 hr and 18-24 hr after antigenchallenge (Table I).

Histological examination of cell accumulation in the lung

Ten minutes after OA exposure, no marked cell accumu-lation was observed. Indeed, as detected after colorationwith Luna's stain, eosinophils were present in small

numbers, mostly in the peribroncbial area (Fig. 7a). Incontrast. 4 br after antigen exposure of sensitized animals,small bronchi demonstrated a marked infiltration ofeosinophils and unstained cells (mostly polymorphonuc-lear neutrophils upon morphological criteria) in thebronchial walls and the submucosa. as well as in tbeperibronchia! regions (Fig. 7b). Tbe presence of eosino-phil-rich mucus in tbe bronchial lumen was also observed.Lungs from aerosol sensitized guinea-pigs exposed to asaline solution (Fig. 7c) were histologically similar tothose from control guinea-pigs (data not shown). Whenthe lungs were removed 24 br after OA exposure.cosinopbil infiltration in tbe broncbial wall was stillobserved, althougb to a lower extent compared toinfiltration at 3-4 hr (Fig. 7d). As compared to lungpreparations removed 4 hr after OA exposure, fewereosinophils were observed in the peribroncbial regionsand in tbe submucosa. Eosinophil numbers in the peri-broncbial area 4 hr after antigen challenge by aerosolwere 1264+173 7 cell/mm- (n = 5). compared to189-8 ±34-8 cell/mm-' {n = 5) in saline-exposed guinea-pigs. Twenty-four hours after antigen challenge, eosino-phil numbers in tbe peribronchial regions were8611 +132' eell/mm'{/i-6). compared to 2I0 2±40-6celi/mm- (n = 6) in saline-exposed guinea-pigs.

Discussion

We establisbed a model to study broncbial byperrespon-siveness involving actively sensitized guinea-pigs cbal-lenged with the antigen by aerosol. This protocol wasdeveloped because it allows pulmonary reactions witboutanimal death. Indeed, despite tbe fact that guinea-pigsdeveloped labored breathing during tbe antigen exposure,only a few of tbem experienced convulsions. To preventmortality. Hutson ct al, [2] have developed an experimen-tal model in which guinea-pigs sensitized by aerosol aretreated witb an anti-bistamine H|-receptor antagonistprior to tbe challenge with tbe antigen. Histamine is.however, the major mediator in the lung involved inanaphylactic reaction in guinea-pigs [11.12] and conse-quently may modulate tbe release of secondary mediatorsimplicated in airway byperresponsiveness [5]. Althoughbronchial hyperresponsiveness may be evoked despiteanti-bistamine treatment, we chose to develop an experi-mental model in which pharmacological manoeuvres wereavoided. The data reported here demonstrate tbat thesuccessive administration of increasing doses of theantigen allowed tbe induction of homogeneous reactionsin the absence of pbarmacological modulation.

In this guinea-pig model, antigen challenge inducesafter 3-4 hr. and to a higber extent after 18-24 br. anairway hyperresponsiveness. as demonstrated by a left-

Bronchial hyperrespotisivene.ss in aerosol-sensitized guinea-pigs 73

Table I. Total and differential cell counts {n x 10* cells) in bronchoalveolar lavage from non-sensitized and saline-exposed (Control), sensitized and saline-exposed (Saline) and sensitizedand OA-exposed guinea-pigs

Treatment

Control

Total cell

lt-7±l 6

Neutrophil

0-1 ±0-05

3-4 hr after antigen or saline exposureSalineOA

12-2 + 1-3i5-2±2-l

0-6 + 0-25.4 + 0-8***

18-24 hr after antigen or saline exposureSalineOA

10-1+0419-2 + 29*

0-2 + 0-032.510.2*-

Eosinophil

0-6 ±0-2

1+0-20-9 ±0-3

0-9+0-16-3±2.|*

Macrophage

10-5±l-3

10-1+1-28-2±l-2

8-7 + 0498 ±1

Lymphocyte

0 3 ±0-2

04 ±0-104 ±0-03

02 + 0-020 3 ±0-1

Bronchoalveolar lavages were performed 3-4 hr and 18 24 hr after the exposure of animals tothe antigen. The comparison of cell numbers between bronehoalveolar lavage fluid obtainedfrom OA-exposed guinea-pigs and saline-exposed guinea-pigs were analysed by Student's f-test{/i = 4). *P<0-05; *

ward shift of the dose-related bronchopulmonary res-ponses curve to aerosol ACh. Although this increase inbronchial reactivity is observed wilh either aerosolizedACh or 5-HT, as compared to control values higheralterations are noted with the latter agonist and at bothtime points. When ACh or 5-HT are injected intrave-nously, alterations of the dose-related bronchopulmon-ury responses are only observed 18-24 hr after antigenchallenge. The increase in pulmonary reactivity observedunder these conditions., and underlined by two differentspasmogenic agents, is consistent with a non-specificeffect., characteristic of the airway hyperresponsiveness inhumans [6]. Similar observations were reported for theincreased airway responsiveness when ACh was adminis-tered either by the intravenous or aerosol route aftercigarette smoke exposure [13].

The observed differences in bronchopulmonary reacti-vity may be due to a bronchial inflammatory reactioninduced by the aerosolized antigen. Indeed, in aerosol-sensitized guinea-pigs, no change in blood pressure andcirculatory blood cell counts were recorded after anligenchallenge by aerosol, a result in contrast to the dataobtained when systemic sensitization and intravenouschallenge of the animals is performed [7]. Thus, successiveantigen aerosols might induce a bronchial reaction lead-ing to hyperresponsiveness restricted to the airways. Oneexplanation for this hyperresponsiveness may be analtered permeability of the airway wall. Indeed airwaypermeability is increased after exposure to cigarette

smoke [13,14] or antigen challenge [15], possibly allowinga greater proportion of inhaled ACh or 5-HT lo reach thesmooth muscle.

As allergen challenge induces complex processes in thelung that are considered to beof particular importance inthe inflammatory events underlying hyperrcsponsiveness[16,17], the time-course of airway hyperresponsivenesswas investigated. An increased reactivity to ACh wasobserved for 5 days after antigen challenge whichreturned to pre-challenge values by day 10. When guinea-pigs exhibiting normal bronchopulmonary reactivitywere exposed to the antigen again, no further alteration inthe bronchial response to either aerosol or systemicadministrations of ACh was noted. These guinea-pigs,however, presented respiratory difficulties during thesecond antigen exposure, as in the case of the initialchallenge. Furthermore, in separate experiments weobserved that during the second antigen provocationthese guinea-pigs developed a bronchopulmonary res-ponse of similar magnitude to that noted during the initialchallenge or that of animals exposed to saline and used 10days after the initial exposure (data not shown). This lackof change in bronchopulmonary reactivity to aerosolizedACh following a second antigen challenge is presentlyunexplained and remains to be investigated.

The relationship between the development of airwayhyperresponsivess and the late phase response is unclear.In both asthmatic patients [16] and sensitized sheep [18],late-phase responses are usually resolved by 24 hr. In the

74 Elisabeth Boichot et al.

Bronchial hyperre.sponsiveness in aerosol-sensitized guinea-pigs 75

guinea-pig, puimonary function returns to nearly normalbetween 24 and 48 br [2]. but with a new decrease inairway conductance at 72 hr. suggesting that differentmechanisms are operational in this species. Therefore, tbeseverity and duration of bronchial hyperresponsivenessand of tbe late-phase response are likely to depend on thedegree of sensitization of the animal, the amount ofantigen presented to the airways, and finally tbe severityofthe anapbylactic response.

Asthmatic reactions in ailergic patients bave beenassociated with a bronchoalveolar eosinopbilia followingallergen inbalation [19,20]. In the guinea-pig modelreported here, only a significant neutropbilia in broncho-alveolar lavage fluids is observed 3-4 br after antigenexposure. This increase in neutrophil counts is followedby a marked eosinophilia 18- 24 hr after antigen exposure.an observation similar to that reported by Hutson et al.[2]. In human asthma. Metzger et al. [3] have also shownibal the late airway response is accompanied by an influxof neutrophils followed by eosinophils. In contrast,compared to lungs from saline- or OA-exposed guinea-pigs removed 10 min after the challenge, a marked cellaccumulation, principally eosinopbils. was observedupon microscopic observation of lung tissue, both after3-4 hr and 18-24 hr. Furthermore, as compared to time3 4 hr. less eosinophils in the peribroncbial area andsubmucosa are counted 24 hr after antigen challenge. Amarked cell invasion ofthe bronchial submucosa includ-ing eosinophils, mast cells and macrophages bas beendescribed 1 br after OA aerosol in sensitized guinea-pigs[21]. However, marked differences between the alveolarspace (as reflected by the cell composition ofthe broncho-alveolar lavage fluids) and tbe lung tissue (as assessed byhistological techniques) may exist, leading to invalidconclusions if one takes into account only the former. Tbekinetics of tbe passage of tbe neutrophils and tbeeosinophils tbrougb the airway walls probably explainsthe late appearance of the latter cell type in tbe broncho-alveolar lavage fluids.

Eosinophil migration tbrougb tbe bronchus and up tothe lumen does not appear to be necessary for thedevelopment of broncbial hyperresponsiveness becauseno increase in eosinopbil number in bronchoalveolarlavage fluids is observed 3-4 br after antigen challenge.

After tbis time period, bowever, numerous eosinopbilsare observed upon bistological examination ofthe lungs,In tbis regard, it would be relevant to distinguish betweeneosinophil migration and activation because tbe latterprocess is probably relevant to tbe developrnent ofbronchial hyperresponsiveness. Indeed, lijima el al. [22]have reported that the eosinophils infiltrating around thebronchioli 7 hr after allergen challenge in tbe guinea-pigdid not appear degranulated. Thus, in the initiation ofbroncbial hyperresponsiveness. mediator release fromresident leucocytes, including eosinophils. may be asimportant as tbe subsequent recruitment of circulatingcells.

Although the mechanisms underlying tbe pathogenesisof airway byperresponsiveness are not elucidated at thistime, a major role for eosinopbils and tbeir granuleproteins is suspected [23]. In the present study, airwayeosinophilia and increased bronchial reactivity did notappear to be associated with detectable disruption ofepithelial cell integrity. Electron microscopy observationsofthe lungs that are currently in progress might help toelaborate a final statement regarding possible alterationsofthe epithelial layer.

In summary, we studied the development of broncbialhyperresponsiveness in guinea-pigs sensitized and chal-lenged with tbe antigen by aerosol, a process tbat isassociated witb an increase of polymorphonuclear cellnumbers in lung tissues and bronchoalveolar lavage fluids.,although exhibiting distinct kinetics. In this experimentalmodel, the histological changes were not correlated withthe cell composition in the tavage fluids, suggesting adynamic process of eosinophil infiltration (and probablyactivation) and passage through the bronchial walls.Finally, tbe dissection of tbe relationship between cellinflltration in tbe lung and alterations in bronchopulmon-ary functions may allow a better understanding of thecomplex mechanisms implicated in tbe immunologicalprocesses associated witb byperresponsiveness.

References

1 Cockroft DW. Ruffin DE. Dolovich J. Hargreave KE.Allergen-induced increase in non-altergic bronchial reacti-vity. Clin Allergy 1987; 7:503-13.

Fig. 7. (a) Photomicrograph of a lung obtained 10 min after OA exposure from aerosol-sensiti/ed guinea-pig. Luna stain.Magnification x64. (b) Photomierograph of a lung obtained 4 hr after OA exposure from aerosol-sensitized guinea-pig, showing amarked eosinophil infiltration through the bronchial wall to the airway lumen. Luna stain. Magnification x 64. (c) Photomicrographof a lung obtained 4 hr after saline exposure from aerosol-sensitized guinea-pig. Luna stain. Magnifiealion x 64.(d) Photomicrograph of a lung obtained 24 hr after OA exposure from aerosol-sensitized guinea-pig. Eosinophil accumulation in thebronchial wall was also observed, although as compared to histological preparations of lungs removed 4 hr after OA exposure, lesseosinophils in the peribronchial regions were observed. Luna stain. Magnification x 64.

76 Elisabeth Boichot et al.

2 Hutson PA. Church MK, Clay TP, Miller P. Holgate ST.Early and late phase bronchoconslriction after allergenchallenge of nonancsthetized guinea-pigs. Am Rev RespirDis 1988; 137:548-57.

3 Met/gcr WJ. Richerson, HB, Worden K, Monnick M.Hunninghake GW. Bronchoalvcolar lavage of allergic asth-matic patients following allergen provocation. Chest 1986;89:477-83.

4 Coyle AJ, Urwin SC, PageCP, Touvay C. Villain B, BraquetP. The cfl"ect of the selective PAF antagonist BN 52021 onPAF- and antigen-induced bronchial hyper-reactivity andeosinophil accumulation. Eur J Pharmacol 1988; 148:51-8.

5 DaflTonchio L. Payne AN. Lees IW. Whittle BJR. Airwayhyperreactivity follows anaphylactic microshock in anesthe-tized guinea-pigs. EurJ Phiinnacol 1989; 161:135-42.

6 Boushey HA, Holtzmann MJ, Sheller JR, Nadei JA.Bronchial hyperreactivity. Am Rev Respir Dis 1989:121:389-413.

7 Lagente V, Moncada S, De Nucci G. Payne AN. Sensitiza-tion of guinea-pigs by aerosol: efl"ect of paraquat. Br JPharmacol 1987:90:35.

8 Konzett H, Rossler R. Versuchsanorung zu untersuchungenan der bronchiol muskulatur. Areh Exp Path Pharmacol1940; 135:71-6.

9 Payne AN, De Nucci G. Anaphylaxis in the guinea-pigsinduced by ovalbumin aerosol: in vivo and in vitro methods.J Pharmacol Methods 1987; 17:83 90.

10 Luna LG. Manual of histologic slaining methods of theArmed Forces Institute of Pathology. 3rd Edn. New-York:McGraw-Hill. 1968.

11 Brocklehurst WE. The release of hislamine and formation ofil slow-reacting substance (SRS-A) during anaphylacticshock. J Physiol I960; 151:416-35.

12 Garcez Do Carmo L. Cordeiro R. Lagente V. Lefort J.Randon J, Vargaftig BB. Failure of a combined anti-histamine and anti-leukolriene treatment to suppress pas-sive anaphylaxis in the guinea-pig. Int J Immunopharmacol1986; 8:985-95.

13 James AL. Dirks P. Ohtaka H, Schellenberg RR, Hogg JC.Airway responsiveness to intravenous and inhaled acetyi-choline in the guinea pig after cigarette smoke exposure. AmRev Respir Dis 1987; 136:1158-62.

14 Hulbcrt WC, Walker DC, Jackson A. Hogg JC. Airwaypermeubilily to horseradish peroxidase in guinea-pigs: therepair phase after injury by cigarette smoke. Am Rev RespirDis 1981; 123:320-6.

15 Boucher FC. Pare PD. Hogg JC. Relationship betweenairway hyperreactivity and hyperpermeability in Ascaris-sensitive monkeys. J Allergy Clin Immunol 1979; 64:197-201.

16 Hiirgreave FE. Dolovich J, Robertson DJ, Kerigan AT. Thehitcastbmaticre.sponse. Can Med Assoc J 1974; 110:415-21.

17 Holgate ST. Beesley R. Twentyman OP. The pathogenesisand significance ofbronchial reactivity in guinea-pigs. ClinSci 1987; 73:561-80.

18 Abraham WM, Delehunt JC, Yerger L, Marchette B.Characterization of a late phase pulmonary response follow-ing antigen challenge in allergic sheep. Am Rev Respir Dis1983; 128:839-44.

19 De Monchy JGR. Kaufmann HF, Venge P. et al. Bron-choalveolar eosinophilia during allergen-induced late asth-matic reactions. Am Rev Respir Dis 1985; 131:373-6.

20 Dunn CJ. Elliott GA. Oostveen JA. Richards IM. Develop-ment of a prolonged eosinophil-rich inflammatory leukocyteinfiltration in the guinea-pig asthmatic response to ovalbu-min inhalation. Am Rev Respir Dis 1988; 137:541-7.

21 Lellouch-Tubiana A. Lefori J. Pfister J, Vargaflig BB.Interactions between granulocytes and platelets with theguinea-pig lung in passive anaphylactic shock. Correlationswith PAF-accther-induced lesions. Int Arch Allergy ApplImmunol 1987; 83:198-205.

22 lijima J, Ishii M, Vamauchi K.etal. Bronchoalvcolar lavageand histoiogical characterization of late asthmatic responsein guinea-pigs. Am Rev Rcspir Dis 1987: 136:922 9.

23 Frigas E, Gleicb GJ. The eosinophil and the patbophysio-logy of asthma. J Allergy Clin Immunol 1986; 77:527-37.