Clinical and Experimental Allergy 1992, Volume 22, pages 187-197

REVIEW

Asthma, inflammation, eosinophils and bronchialhyperresponsiveness

H. SMITH

SmithKline Beecham Pharmaceuticals, Epsom, Surrey, U.K.

Introduction

There is currently an emphasis on the importance ofinflammation in asthma and the contribution that theeosinophil makes to this. It is intended to review theevidence for an association between asthma and increasednumbers of eosinophils in the blood, broncho-alveolarlavage (BAL) fluids and lung tissue. An attempt will bemade to estimate the extent to which inflamrnationcontributes to the bronchial hyperresponsiveness of asth-matics. The part that the eosinophil might play inmediating asthma attacks will be considered briefly.

Blood eosinophilia and asthma

It has been known since the beginning of this century thatasthmatics can have an increased number of eosinophilsin their blood [1]. In 1975 Horn et al. [2] studied 14intrinsic asthmatics and measured the number of bloodeosinophils and the forced expiratory volume in 1 second(FEVi) in each patient on an average of five times, over aperiod of time when the values were changed by the use oforal steroids. They found a negative correlation betweenFEVI and the number of blood eosinophils. Since thenothers have shown similar correlations. Bousquet et al. [3]studied 43 chronic asthmatics, 19 of whom had severedisease, and found a negative correlation betweennumber of blood eosinophils and FEV| and a positivecorrelation between number of eosinophils and theclinical severity of the disease. More recently measure-ments in 23 asthmatics, 16 of whom were atopic, showed anegative correlation between number of blood eosino-phils and peak expiratory flow rate (PEF), before treat-ment with steroids. After steroid treatment the negativecorrelation was between levels of eosinophil cationicprotein (ECP) in the blood and PEF, suggesting that abetter negative correlation could be between airwayfunction and the number of activated eosinophils ratherthan the total number [4].

A correlation has been shown between the number ofeosinophils in the blood of asthmatics and their bronchial

Correspondence: Dr H. Smith, SmithKline Beecham Pharmaceuticals,Great Burgh, Yew Tree Bottom Road, Epsom, Surrey KT18 5XQ, U.K.

responsiveness to inhaled histamine [5] or methacholine[6], measured as the inhaled dose required to produce a20% fall in FEV, (PC20FEV,).

A direct relationship between the ability of cortico-steroids to reduce blood eosinophil counts and provideclinical benefit was demonstrated when they were givenorally [2] or by inhalation [7,8] and it has been suggestedthat the ability of inhaled steroids to reduce bloodeosinophil counts in man can be used to assess theirpotential effectiveness as treatments for asthma [7]. In arecently reported study it was shown that inhaled budeso-nide given to 10 asthmatics for 14 days produced areduction in their responsiveness to inhaled methacholinetogether with a fall in number of total eosinophils in theblood and a reduction in the percentage of hypodenseeosinophils (9). The finding that inhaled steroids canreduce blood eosinophil numbers at doses that have few ifany systemic effects is evidence that the lungs can be thesite of formation of eosinopoietic substances and fore-shadows the finding of increased expression of mRNA forinterleukin-5 (IL-5) in mucosal bronchial biopsies ofasthmatics (10).

A blood eosinophilia can occur in many diseases,however, without necessarily being associated withasthma. In some of these diseases, for example theidiopathic hypereosinophilic syndrome, the number ofeosinophils in the blood can be far higher than is found inasthma. In addition a high proportion of these eosino-phils can be in the activated hypodense form (11). Areduction in blood eosinophil number can occur in acutesevere asthma (12) and this is probably related to theblood eosinopenia that can occur 4 hr after inhaledallergen challenge of asthmatics (13). The reason for theeosinopenia is not known but could be due to the effects ofendogenous steroid release, or other factors, producing aredistribution of eosinophils, possibly involving adhesionto the lung vascular endothelium prior to migration intothe lung.

The conclusion must be that to obtain a correlationbetween eosinophil number in the blood and measure-ments of the severity of asthma, will be dependent uponpatient selection. If the eosinophil contributes to thepathology of asthma it will be the eosinophils in the lung

187

188 H. Smith

that will be involved. The number of eosinophils in theblood of asthmatics must therefore reflect changes in thelung.

Cells in BAL fluids

That there is an increased number of eosinophils in theairways of asthmatics was indicated in the middle of lastcentury by the finding of Charcot-Leyden crystals, nowknown to be degradation product of eosinophils, in thesputum of asthmatics [14]. In 1889 the presence ofeosinophils in sputum was shown to be a characteristic ofasthma [15]. There are difficulties in collecting sputumand in quantifying its contents. An important advancewas the introduction in the mid 1970s of small volumebroncho-alveolar lavage (BAL), performed under a localanaesthetic through a fibreoptic bronchoscope, as a

Table 1. Changes of eell numbers in the BAL fluids of asthmaticscompared with normals

Cell

TotalMacrophageLymphocyteNeutrophilEosinophilMast cellEpithelial cell

Change

NilDecreaseIncreaseIncreaseIncreaseIncreaseIncrease

Studies showing significantehange/total studies

In all studies2/102/101/108/106/63/9

Ten studies, 250 asthmatics c.f 136 normals, range 8 to 93asthmatics per study. There was no count of mast cells in fourstudies and of epithelial eells in one. The numbers of cells weremeasured as absolute counts or percentage of total cells [3,18-26].

diagnostic tool [16]. It was shown to be sufficiently safe forit to be recommended as an appropriate part of theroutine investigation of mild to moderate asthmatics [17].

Some of the results in 10 studies, are outlined in Table1. The asthma was of mild to moderate severity in most ofthe patients but 19, in a study of 43 had severe asthma [3].The inflammatory changes were not marked. The totalnumber of cells in the BAL fluids were not differentbetween the asthmatics and the controls in any of thestudies, except for 11 asthmatics on inhaled steroids inwhom the total cells were reduced [22]; data from thesepatients are not included in Table 1. There were, however,changes in the differential cell counts, with a tendency fora decrease in number of macrophages and an increase innumber of lymphocytes, neutrophils, eosinophils andshed epithelial cells. The changes in number of macro-phages and lymphocytes only reached significance in twostudies [ 19,20] and of neutrophils in one [19]. The increasein number of eosinophils reached significance in all thestudies except for two [23,26]. A consistent finding was anincrease in number of mast cells in all the studies in whichthey were evaluated [18,21-25]; however, the number ofmast cells, as a percentage of total cells varied amongstthe studies, from a mean of 0-22 in asthmatics to 0-04 incontrols, in one study [18], to 1-8 and 0-5 in another [22].The increase in number of shed epithelial cells reachedsignificance in two studies [22,26] and was found in somebut not all asthmatics in another [3]. Some of the resultsfrom the study in which the largest number of asthmaticswere subjected to BAL [19] are represented in Table 2. Itshould be emphasized that this is the only study in whichthere was a significant increase in number of neutrophils.

Cells in BAL fluids, airflow obstruction and bronchialhyperresponsiveness

Surprisingly few attempts have been made to correlate thenumber of cells in the BAL fluids with the degree of

Table 2. Number of cells in the BAL fluids of asthmatics and normal subjects

Controls(« = 30)Asthmatics(« = 93)P

Total

X lO^ml

158(±23)

207(±23)

NS

Macrophage

90-0(±0-9)

71-9(±2-3)< 0-001

Differential count %

Lymphocyte

8-7(±0-9)

15-3(±1-9)< 0-001

Neutrophil

0-9(±0-1)

5-7(±1-4)< 0-001

of total

Eosinophil

0-2(±0-1)

6-8(±1-3)< 0-001

Epithelial cell

0-1(±0-06)

0-4(±0-13)

NS

Mean values±s.e.m. [17].

Asthma, inflammation, eosinophils and bronchial hyperresponsiveness 189

Table 3. Determination of an inverse correlation between logPC20 F E V I (histamine or methaeholine) and composition ofBAL fluids of asthmatics in some studies

% cellsMacrophageLymphocyteNeutrophilEosinophilMast cellsEpithelial cell

Number of cellsTotalMacrophageLymphocyteNeutrophilEosinophilMast cellEpithelial cell

ConcentrationMBP

Eound

20201818,2518,21

2427202424,272426

18

Not found

22, 28*22, 28*21,22,28*21,22,28*21,2222

20,2626,2720, 26, 2720,26

20

24

The numbers are the reference number to the appropriate study,* PC|5 F E V I methacholine was used.

Table 4. Patients with > 5% eosinophils in their BAL fluids of1059 tested [30]

Diagnosis

Interstitial lung diseaseAIDS-associated pneumoniaEosinophilic pneumoniaDrug induced lung diseaseHodgkins diseaseAsthmaBronchitisHelminthic infectionBone marrow transplantBacterial pneumonia

n

19876221111

% eosinophils range

5-517-43

21-575-325-66-11

281796

airflow obstruction. In one study no correlation wasfound between FEVi and cellular contents [20] and inanother, whilst the number of macrophages inverselycorrelated with the forced expiratory flow rate at between25 to 75% of maximum (FEF25-75%) there was nocorrelation with this and the counts of other cells [27]. Aninverse correlation was found between FEV|, as a

percentage of predicted, and the percentage of mast cells[25] or number of mast cells, total cells, eosinophils orneutrophils in the BAL fluids [24].

In many studies attempts were made to correlatebronchial responsiveness to inhaled histamine or metha-chohne with the cell content of the BAL fluids. The resultshave been variable (Table 3), The log PC20 FEVi, tohistamine or methacholine, has been reported to nega-tively correlate with: the percentage of lymphocytes,neutrophils [20], eosinophils [18], mast cells [18,25], orepithelial cells [18,21]; the number of total cells [24],macrophages [27], lymphocytes, particularly CD8''" Tcells [20], neutrophils [24], eosinophils [24,27], mast cells[24], or epithelial cells [26]; the concentration of MBP [18];and with the reactivity of BAL macrophages, as measuredby lucigen-enhanced chemiluminescence [20]. However,in other studies no correlation was found betweenbronchial responsiveness and some of these changes(Table 3).

The failure to consistently demonstrate a correlation isnot surprising since most of the patients had mild tomoderate asthma and the changes in the cellular contentsof the BAL fluids were small and, in addition, only smallnumbers of asthmatics were evaluated in most of thestudies. A correlation may merely reflect two difl'erentpopulations, the asthmatics and the controls, with clus-ters at either end of the correlation curve. The relation-ship between bronchial hyperresponsiveness and theclinical state of asthma has been claimed to be insuffi-ciently close to be of practical clinical use [29]. It maytherefore be inappropriate to attempt to correlatemeasures of lung inflammation with responsiveness;better parameters could be measurements of lung airflowor the severity of the disease. The latter was precluded inmost of the studies since the patients had mild asthma andit is of significance that in a study of 43 asthmatics, whichincluded 19 with severe asthma, a correlation was foundbetweeen the severity of the disease and the number ofeosinophils and levels of ECP in their BAL fluids [3].

BAL eosinophilia without asthma

There is a wide range of pulmonary eosinophilic syn-dromes and in some of these, for example the eosinophilicpneumonias, numbers of eosinophils in the BAL fluidscan be higher than those found in asthma (Table 4) [30].In addition there is evidence that the eosinophils areactivated in some of these diseases. In chronic eosinophi-lic pneumonia degranulated eosinophils [31] and MBPhave been found in the lung tissue [32] and a highproportion of the eosinophils in the BAL can be in theactivated hypodense form [33]. Increased levels of ECPhave been found also in the BAL fluids of patients withadult respiratory distress syndrome [34] and with idio-

190 H. Smith

Tabte 5. Changes in bronchial biopsies taken from asthmatics compated with those from normalsubjects

Found Not found

Epithelial damageThickened lamina reticularis below basement membrane

Increased number(a) Epithelium

Total cellsMacrophagesLymphocytesNeutrophilsEosinophilsMast cellsDegranulated mast cellsDegranulated eosinophils

(b) SubmucosaTotal cellsMacrophagesLymphocytesNeutrophilsEosinophilsMast cellsDegranulated mast cellsDegranulated eosinophils

3, 38, 40, 4221,26,42,43,44

21

3

3,4521453,45

21,394744*, 47

21,26,44,454126,4526, 44, 45

21,26,41,44,4639

All studies3845

26

26All studies3,4121,26,45

Thirteen studies, 159 asthmatics c.f. 89 normals, range 4 to 43 asthmatics per study [3, 21,26, 38-47]. Inone study (42) a leucocyte infiltration was apparently present in the bronchial biopsies but did not reachsignificance. * Increase in number of activated T cells, not total lymphocytes. The numbers refer to theappropriate study.

pathic pulmonary fibrosis [35]. Excluding patients withasthma only, there is a wide range of clinical presentationof pulmonary eosinophilic syndromes. The patients havevariable pulmonary infiltrates on chest radiographs andusually, but not invariably, have breathlessness, reducedlung function, weight loss, often with fever and nightsweats. Histology studies showed an alveolitis withleucocytes, mainly eosinophils, in the interstitial spacesand small airways. In asthma there is little alveolitis andthe eosinophils in the lung are mainly localized in thebronchial submucosa and the bronchial airways. Asthmacan occur in every clinical syndrome associated withpulmonary eosinophilia [11]. It would be of interest toknow if the distinguishing feature of those patients withpulmonary eosinophilia and asthma was the presence ofactivated and secreting eosinophils in the bronchialsubmucosa.

Bronchial biopsy

Until recently descriptions of the pathology of asthmahave been based on autopsy studies after death in status

asthmaticus. The changes found have been fully docu-mented, and include mucous plugging of segmentalbronchi and bronchioles, epithelial cell loss, mucus glandhyperplasia, thickening of the epithelial basement mem-brane, oedema of the submucosa, with infiammatory cellinfiltrations rich in eosinophils, and smooth musclehypertrophy [36,37]. There are doubts as to the extent ofthese changes in clinical asthma and it has been claimedthat patients who have suffered lifelong asthma may diewith relatively normal lungs [37].

In the past decade bronchial biopsies have beencollected from asthmatics and normal subjects using thefibreoptic or rigid bronchoscope. These studies haveshown that infiammatory changes can occur in the lungsof some patients with even mild asthma. Some of theresults obtained in 13 recent studies are outlined in Table5 [3,21,26,38-47].

Marked changes in the bronchial epithelium of asth-matics, with shedding of columnar ciliated cells leavingonly basal cells, have been described [3,38,40,42]. How-ever, they were not present in biopsy specimens from all

Asthma, inflammation, eosinophils and bronchial hyperresponsiveness 19t

asthmatics and tended to be associated with increasedseverity [3,42], They were not found in some patients withmild asthma who also had bronchial hyperresponsiveness[41,46] and shedding of the epithelium has been foundalso in normal controt subjects [21,26,44], leading to theclaim that collecting and processing the biopsy candamage the surface epithelium [44], A consistent findingwas a thickening of the lamina reticularis below thebasement membrane [21,26,42-44], although in one studythis did not reach significance [39], There is evidence thatthe thickening is due to collagen deposition produced by anetwork of contractile myofibroblasts beneath the base-ment membrane and that it is not produced by theepithelium [43,48], A cellular infiltration, rich in eosino-phils, into the epithelium and submucosa has beendescribed (Table 5), The presence of degranulated eosino-phils in the epithelium and submucosa in tissue fromasthmatics but not from normal controls was identifiedvisually and with the use of antibodies to the secretedform of ECP (EG2) [3,26,44,45], Increased numbers ofdegranulated mast cells were found in the tissue fromasthmatics [26,45] but this finding has been disputed [49],Increased numbers of neutrophils were not found in anyof these studies and in one there was a significant decreasein numbers of neutrophils in the submucosa [26] and inanother a decrease of neutrophil elastase secreting cells[44], However in a recently reported study increasednumber of neutrophils were found in the bronchialepithelium of asthmatics [50], In one study a cellularinfiltration, into the bronchial mucosa of asthmatictissue, did not reach significance but there was an increasein the percentage of lymphocytes with an irregular shape[42], Since then, an increase in the number of activatedlymphocytes has been found, as evidenced by an increasein the number of IL-2 receptor positive cells [44],increased HLA-DR expression [47] and the presence ofmRNA for IL-5 [10] in lung biopsies from asthmatics.Increased concentrations of IL-lj5, IL-6 and colonystimulating factor (GM-CSF) in the BAL fluids fromasthmatics have also been claimed [51], The evidence hasbeen summarized to support the thesis that activated Tlymphocytes in the lung produce a selective accumulationand activation of eosinophils in the bronchial submucosaof asthmatics [52],

Changes in lung biopsies and bronchial hyperresponsiveness

Attempts have been made to correlate changes in bron-chial biopsies in asthmatics with bronchial responsivenessto inhaled histamine or methacholine rather than withmeasurements of bronchial airway function or the sever-ity of the disease. Comments made previously on theattempts to correlate changes in BAL fluids with bron-

chial responsiveness are applicable and it is not surprisingthat the results have been variable, A negative correlationbetween the PC20 FEV| methacholine and the degree ofepithelial damage has been claimed [42], but not found byothers [38], and asthmatics with no evidence of bronchialepithelial damage were hyperresponsive to inhaled meth-acholine [46], A correlation between bronchial respon-siveness and number of leucocytes in the epithelium hasalso been claimed [2t] but others found no correlationbetween responsiveness and number of eosinophils, neu-trophils or mast cells in the epithelium [41], A correlationwith a good spread of points on the correlation curve, wasfound between mononuclear cell activation, as measuredby HLA-DR expression, and bronchial responsiveness[47], An increase of IL-2 receptor positive lymphocyteswas also associated with hyperresponsiveness [44], Asth-matics with mRNA for IL-5 in their biopsies tended tohave more severe disease and an increased number ofsecreting eosinophils (EG2+) and lymphocytes express-ing IL-2 receptors [10], However, others found that thenumber of EG2+ eosinophils did not correlate withdisease severity [45], In a study of 43 asthmatics, 19 ofwhom had severe disease, there was a correlation betweenthe number of eosinophils in the epithelium and theseverity of the disease [3],

Quantitative assessment of bronchial hyperresponsiveness

Asthmatics versus normals

In asthmatics, compared with normals, there is anincreased responsiveness to the bronchoconstrictoreflects of inhaled histamine or methacholine with a shiftto the left and an increase in slope of the log dose-responsecurve. There is a good correlation between responsivenessto histamine and methacholine, but not with some otherstimuli [53], The responsiveness is conventionally mea-sured as the PC20FEV1 to histamine or methacholine andasthmatics are 10- to 1000-fold more responsive thannormals [54], In Table 6 the responsiveness of asthmaticshas been compared with normals in subjects from whomBAL fluids or bronchial biopsies were collected in thestudies reviewed previously. In most of the studies thenormal subjects did not respond at the highest doses ofhistamine or methacholine used whilst the meanPC20FEV1 for the asthmatics was 30- to 16-fold less. Intwo studies [18,26], a sufficiently large dose of methacho-line or histamine was given to produce a 20% fall in FEViin the normal subjects. In these studies the ratios of thegeometric mean PC20FEV1 values for the normals com-pared with that for the asthmatics were 75 and 47, On thisbasis even mild asthmatics can be at least 40 times moreresponsive than are normals.

192 H. Smith

Table 7. Ratio of geometric mean PC20 FEVj after to that beforetreatment with inhaled steroids

1 able 6. Ratio ot geometric mean PC20 FEVi Normals com-pared with asthmatics

Nature of asthma

Mild to moderateMild to moderateMild remissionStableMild atopicMild atopicMild remission :

Ratio

h m

75>32>21>24

47>16

>16

n

8cf 920 cf 226cf 13

10 cf 104cf 86 cf 116cf 15

Reference

18202124264547

Duration

1 to 5 months2 weeks1 month

1 inonin

6 months4 weeks10 weeks1 to 2 years4 months1 month2 months

h

3-41-82-4

7-08-2

34-08-5

m

1-53-1

1-88-3

n

1417171 1

1 111262613

18

10

Reference

555656enJ /

5758585960*6162

h, histamine. m, methacholine.

Asthmatics after treatment with inhaled steroids

Treatment with inhaled steroids can reduce the hyperres-ponsiveness of asthmatics but the effects, after treatmentfor several weeks, are only partial with a 1 -5- to eightfoldincrease in the dose of inhaled histamine or methacholinerequired to produce a 20% fall in FEV, [55-62] (Table 7).It is not surprising that these small changes were notdemonstrated in some studies [22,63]. Treatment withdoses of inhaled steroids carefully adjusted to symptomsfor 1-2 years did produce a 34-fold reduction in respon-siveness to inhaled histamine [59]. However, six asth-matics treated with inhaled steroids for 10 years were still,on average, 33 times more responsive to inhaled metha-choline than were normal subjects. The inflammatorychanges in bronchial biopsies from these patients werenot different from normal, except for patches of reducedcilia on the epithelium [39]. Inhaled steroids can reducethe number of eosinophils [7-9] and the percentage ofhypodense eosinophils in the blood [9]. When an asth-matic was treated with inhaled steroids for 15 weeks, noeosinophils could be detected in lung biopsy specimensand the number of ciliated cells and inflammatory cellswere returned to baseline values [60]. Therefore treatmentwith inhaled steroids can reduce the number of inflamma-tory cells in the bronchial submucosa to normal but thishas only a minor effect on the bronchial hyperresponsive-ness.

Effects of increasing the number and activity of inflamma-tory cells in the lung

Increased number of eosinophils have been produced inthe lungs of guinea-pigs by the inhalation of antigen [64]and in rats by the intravenous injection of Sephadex

* PC15 FEVi was used.Six asthmatics treated with inhaled steroids for 10 years were 33-fold more responsive to inhaled methaeholine than normals [39].h, histamine. m, methacholine.

Table 8. Ratio of geometric mean PC20 FEVi histamine, beforeto that after inhaled allergen challenge of asthmaties who have adual response

Time after challenge

2hr3hr7hr24 hr

30 hr48 hr2 weeks

Ratio

0-84-42-84-44-21-52-12-1

n

12141214I3o121414

P

NS<0-05<0-05<0-05<0-05<0-05<0-05

NS

Reference

6869686970686969

particles [65]. These treatments also produced a two- tofourfold increase in lung airways responsiveness tohistamine and methacholine in the guinea pig and to 5-hydroxytryptamine in the rat (Fig. 1). This shift in thedose response curve is of a similar magnitude to thetwofold shift to the right of the log dose response curve toinhaled methacholine produced in asthmatics after 4weeks of treatment with budesonide [61] (Fig. 2) whichshould have reduced the number and activity of theeosinophils in their lungs.

The number and activation state of eosinophils can beincreased in the lungs of asthmatics by the inhalation of

Asthma, inflammation, eosinophils and bronchial hyperresponsiveness 193

70 r

1.5 3 6 12 24 48 96

Dose of 5-HT given iniravenousiy (/j.g/kg)

Fig. t. The effects of increasing the number of eosinophils in thelungs of rats on the responsiveness of their lung airways to 5-HT. Sephadex G200,0-5 mg, was given intravenously to the ratson days 0, 2 and 5 and the measurements were made on day 7, atthe time of the peak in lung eosinophilia [65]. Mean + s.e.m.,n = 6 or more. * P<005, ** P<OOl, *** P<000\.

40

30

20

10

0

Budesonide

-

-

!

/X

T .

/

A1

vV/

/ AX/

1 1

I.—

f——f""

I—Jo

f—• 4

1 1

weeks

weeks

0-5 I 2 4 8 16 32 64 128

Methacholine (mg/ml)

Fig. 2. Dose-response curves to inhaled methacholine beforeand after 4 weeks of treatment with budesonide in eight subjectswith mild asthma. Mean ( + s.e.m.) [61].

allergen. This is seen in those asthmatics given sufficientallergen to produce an immediate and late phase responsebut not in those with only an immediate response. In thesedual responders there is an eosinophilia in the BAL fluidssome hours after allergen inhalation, together with anincrease in levels of eosinophil derived basic proteins[66,67]. The dual responders also show a two- to fourfoldincrease in responsiveness to inhaled histamine followingthe inhalation of allergen [68-70] (Table 8).

Conclusion

Reducing the cellular infiltration in the lungs of asth-matics to normal levels produces a trivial reduction inbronchial responsiveness (two- to eightfold). Converselyincreasing the cellular infiltration and activation of theinflammatory cells produces a two- to fourfold increase inresponsiveness. Inflammatory cells in the bronchi ofasthmatics therefore produce a small increase in respon-siveness which is easily reversible. The bulk of thehyperresponsiveness is more persistent and less tractable.Inflammatory cells could be responsible for producingboth types of hyperresponsiveness and early treatmentmay be required to prevent the development of the longlasting hyperresponsiveness.

Mechanisms of bronchial hyperresponsiveness

Reversible hyperresponsiveness

Inflammatory cells, particularly the eosinophil, contain avariety of materials which could be implicated in produc-ing the reversible hyperresponsiveness. For example themajor basic protein (MBP) in human eosinophilsincreased the responsiveness of guinea-pig trachea in vitroand of dog trachea in situ to histamine and acetylcholine[71,72]. The effect was an increase in the slope of the logdose response curve. Bronchiolar strips from an asth-matic [73] and tracheal smooth muscle from sevenasthmatics who died of asthma [74] showed a similarincrease in the slope of the log dose-response curve tohistamine and cholinergic drugs. Platelet activating factorcan be released by eosinophils and it has been reported toproduce a long-lasting bronchial hyperresponsivenesswhen given by inhalation to normal people [75] but thiseffect was not reproduced by others [76]. Other spasmo-genic materials such as histamine and the peptido-leukotrienes could increase the basal tone of the respira-tory smooth muscle. Oedema of the airway walls couldalso contribute to the reversible hyperresponsiveness.

Persistent bronchial hyperresponsiveness

A consistent finding in bronchial biopsies taken fromasthmatics is a thickening of the lamina reticularis belowthe basement membrane. There is evidence that myo-fibroblasts may produce the subepithelial fibrosis. Anexcess of myofibroblasts beneath the bronchial epithe-lium has been demonstrated in asthmatics, with a closecorrelation between the myofibroblast number and thedepth of the collagen layer [48]. The fibrosis may cause adecline in pulmonary function and the presence of anetwork of contractile myofibroblasts beneath the base-ment membrane could contribute to the persistent hyper-responsiveness.

194 H. Smith

The eosinophils could be responsible for these changes.The changes occur at sites where activated eosinophilsand the deposition of ECP has been demonstrated[3,44,45]. Endomyocardial fibrosis occurs in patients withthe idiopathic hypereosinophilic syndrome and activatedeosinophils and ECP have been demonstrated in thefibrotic lesions [77]. Eosinophils can secrete a fibroblastgrowth factor [78] and ECP stimulates the synthesis ofhyaluronan and proteoglycan in human fibroblasts [79].

Symptoms of asthma

Asthma attacks seen as spontaneous increases in bron-chial airways resistance are, to the asthmatic, the mostimportant symptoms of the disease. They may bemimicked by provocation. Recent studies with the newpeptido-leukotriene antagonists, which have appropriatepotency and pharmacodynamics, have shown that theycan protect asthmatics against immediate and late phaseprovoked asthma [80-82]. This is evidence that thepeptido-leukotrienes are important mediators of asthmaattacks. Human neutrophils [83], monocytes [84] and lungmacrophages [85] release mainly LTB4 upon stimulation.Mast cells [86] and eosinophils [87] release LTC4 and cando so by IgE-dependent mechanisms. The eosinophil maytherefore be an important source of the peptido-leuko-trienes and may be the only source during late phaseasthma.

Summary

Asthmatics can have a blood eosinophilia which in somestudies correlates with the severity of the disease. How-ever, an increased number and percentage of activatedeosinophils can be present in the blood without asthma.The eosinophils that contribute to asthma will be those inthe lung. In the BAL fluids collected from asthmaticsthere is usually no change in total cell number, but thereare changes in the differential cell count. A consistentfinding is an increase in percentage of mast cells andeosinophils with a tendency for an increase in lympho-cytes and epithelial cells and a decrease in percentage ofmacrophages. As with the blood eosinophilia, an increasein number of eosinophils can be present in BAL fluidswithout asthma.

The site of localization and activation of the eosino-phils in the lung may be critical. In bronchial biopsies,taken from asthmatics, increased number of mast cells,eosinophils and lymphocytes have been demonstrated inthe bronchial mucosa together with shedding of columnarepithelial cells. However these changes have not beenfound, or have not reached significance, in all studies. Anincrease in number of activated eosinophils and T-

lymphocytes has been demonstrated but an increase innumber of degranulating mast cells has been disputed. Aconsistent finding has been thickening below the base-ment membrane. Attempts to correlate the changes in theBAL or lung biopsies with the severity of asthma, lungairways function or bronchial responsiveness have giveninconsistent results. Treatment of asthmatics with inhaledsteroids can reduce the cellular infiltration in the bron-chial biopsies to normal levels but this produces a trivialreduction in bronchial responsiveness. It is possible thatinfiltration of inflammatory cells into the bronchialmucosa is intermittent, at least in mild asthma, but thisproduces changes leading to a long lasting bronchialhyperresponsiveness.

References

1 Hirsch JG, Hirsch BI. Paul Ehrlich and the discovery of theeosinophil. In: Mahmoud AAF, Austen KF, eds. Theeosinophil in health and disease. New York: Grune andStratton, 1980:2-23.

2 Horn BR, Robin ED. Theodore J, Van Kessel A. Totaleosinophil counts in the management of asthma. N Engl JMed 1975; 292:1152-5.

3 Bousquet J, Chanez P, Lacoste JY, Barneon G, GhavanianN, Enander I, Venge P, Ahlstedt S, Simony-Lafontaine J,Godard P, Michel F-B. Eosinophilic inflammation inasthma. N Engl J Med 1990; 323:1033-9.

4 Griffin E, Hakansson L, Formgren H, Jorgensen K, Peter-son C, Venge P. Blood eosinophil number and activity inrelation to lung function in patients with asthma andeosinophilia. J Allergy Clin Immunol 1991; 87:548-57.

5 Taylor KJ, Luksza AR. Peripheral blood eosinophil countsand bronchial asthma. Thorax 1987; 42:452-6.

6 Durham SR, Kay AB. Eosinophils, bronchial hyperreacti-vity and late phase reactions. Clin Allergy 1985; 15:411-18.

7 Harris DM. Clinical pharmacology of beclomethasonedipropionate. In: Mygind W, Clark TJH eds. Topical steroidtreatment for asthma and rhinitis. London: Bailleire Tindall;180:35-47.

8 Toogood JH, Jennings B, Baskerville J, Letroe NM. Per-sonal observation of the use of inhaled corticosteroid drugsfor chronic asthma. Eur J Respir Dis 1984; 68:321-38.

9 Evans PM, O'Conner BJ, Fuller KF, Chung KF, Barnes PJ.Effect of inhaled budenoside on eosinophil density inasthma. Br J Clin Pharmac 1991; 31:616P.

10 Hamid Q, Azzawi M, Ying S, Moqbel R, Wardlaw AJ,Corrigan CJ, Bradley B, Durham SR, Collins JV, JefferyPK, Quint DJ, Kay AB. Expression of mRNA for inter-leukin-5 in mucosal biopsies from asthma. J Clin Invest1991; 87:1541-6.

11 Spry JF. Eosinophils. A comprehensive review, and guide tothe scientific and medical literature. Oxford: Oxford Univer-sity Press, 1988.

12 Luksza AR, Jones DK. Comparison of whole-blood eosino-

Asthma, inflammation, eosinophils and bronchial hyperresponsiveness 195

phil counts in extrinsic asthmatics with acute and chronicasthma. Clin Research 1982; 285:1229-31.

13 Dahl R, Venge P, Olsson I. Variations in blood eosinophilsand eosinophil cationic protein in serum in patients withbronchial asthma. Allergy 1978; 33:211-15.

14 Sakula A. Historical review. Charcot-Leyden crystals andCurshmann spirals in asthmatic sputum. Thorax 1986;41:503-7.

15 Gollasch. Zur kenntniss der asthmatischen sputums.Fortschritte der Medizin (Berlin) 1889; 7:361-5.

16 Reynolds HY, Newball HH. Analysis of proteins and thoraxrespiratory cells obtained from human lungs by bronchiallavage. J lab clin Med 1974; 84:559-73.

17 Godard P, Aubas P, Calvayrac P, Taib J, Michel FB.Endoscopie et lavage bronchoalveolaire chez l'asthmatiqueallergique. Nouv Presse Med 1981; 10:3141.

18 Wardlaw AJ, Dunnette S, Gieich GJ, Collins JV, Kay AB.Eosinophils and mast cells in bronchoaiveolar lavage insubjects with mild asthma. Am Rev Respir Dis 1988; 137:62-9.

19 Godard P, Bousquet J, Lebel B, Michel FB. Le lavagebronchoalveolaire chez l'asthmatique. Bull Eur Physio-pathol Respir 1978; 23:73-83.

20 Kelly CW, Stenton CS, Bird G, Hendrick DJ, Walters EH.Number and activity of inflammatory cells in bronchoaiveo-lar lavage fluid in asthma and their relation to airwayresponsiveness. Thorax 1988; 43:684-92. Lymphocyte sub-sets in bronchoaiveolar lavage fluid obtained from stableasthmatics and their correlations with bronchial responsive-ness. Clin Exp Allergy 1989; 19:169-75.

21 Foresi A, Bertovelli G, Pesci A, Chetta A, Olivieri D.Inflammatory markers in bronchoaiveolar lavage and inbronchial biopsy in asthma during remission. Chest 1990;98:528-35.

22 Adelroth E, Rosenhall L, Johansson SA, Linden M, VengeP. Inflammatory cells and eosinophilic activity in asthmaticsinvestigated by bronchoaiveolar lavage. Am Rev Respir Dis1990; 142:91-9.

23 Tomioka M, Ida S, Shindoh Y, Ischihara T, Takishima T.Mast cells in bronchoaiveolar lumen of patients withbronchial asthma. Am Rev Respir Dis 1984; 129:1000-5.

24 Kirby JG, Hargreave FE, Gieich GJ, O'Byrne PM. Bron-choaiveolar cell profiles of asthmatic and non-asthmaticsubjects. Am Rev Respir Dis 1987; 136:379-83.

25 Flint KC, Leung KBP, Hudspith BN, Brostoff J, Pearce FL,Johnson NM. Bronchoaiveolar mast cells in extrinsicasthma: a mechanism for the initiation of antigen specificbronchoconstriction. Br Med J 1985; 291:923-6.

26 Beasley R, Roche WR, Roberts JA, Holgate ST. Cellularevents in the bronchi in mild asthma and after bronchialprovocation. Am Rev Respir Dis 1989; 139:806-7.

27 Ferguson AC, Wong WM. Bronchial hyperresponsivenessin asthmatic children. Correlation with macrophages andeosinophils in broncholavage fluid. Chest 1989; 96:988-91.

28 Paggiaro P, Bacci E, Paoletti P, Bernard P, Dente FL,Marchetti G, Talini D, Menconi GF, Giuntini C. Bron-choaiveolar lavage and morphology of the airways after

cessation of exposure in asthmatic subject sensitized totoluene diisocyanate. Chest 1990; 98:536-42.

29 Josephs LK, Gregg I, Mullee MA, Holgate ST. Nonspecificbronchial reactivity and its relationship to the clinicalexpression of asthma. A longitudinal study. Am Rev RespirDis 1989; 140:350-7.

30 Allen JN, Davis WB, Pacht ER. Diagnostic significance ofincreased bronchoaiveolar lavage fluid eosinophils. Am RevRespir Dis 1990; 142:642-7.

31 Gonzalez EB, Swedo JL, Rajaraman S, Daniels JC, GrantJA! Ultrastructural and immunohistochemical evidence forrelease of eosinophilic granules in vivo: cytotoxic potentialin chronic eosinophilic pneumonia. J Allergy Clin Immunol1986; 79:755-62.

32 Grantham JG, Meadows JA, Gieich GJ. Chronic eosinophi-lic pneumonia: evidence for eosinophil degranulation andrelease of major basic protein. Am J Med 1986; 80:89-94.

33 Prin L, Capron M, Gosset P, Wallaert B, Kusnierz JP, BletryO, Tonnel AB, Capron A. Eosinophilic lung disease:immunological studies of blood and alveolar eosinophils.Clin Exp Immunol 1986; 63:249-57.

34 Hallgren R, Samuelsson T, Venge P, Modig J. Eosinophilactivation in the lung is related to lung damage in adultrespiratory distress syndrome. Am Rev Respir Dis 1987;135:639-42.

35 Hallgren R, Bjermer L, Lundgren R, Venge P. The eosino-phil component of the alveolitis in idiopathic pulmonaryfibrosis. Signs of eosinophil activation in the lung are relatedto impaired lung function. Am Rev Respir Dis 1989;139:373-7.

36 Dunhill MS. The pathology of asthma. In: Porter T, Birch J,eds. Identification of asthma. Ciba Foundation StudyGroup 38, London: Churchill Livingstone, 1971:35-46.

37 Lopez-Vidriero MT, Reid L. Pathological changes inasthma, fn: Clark TJH, Godfrey S, eds. Asthma. Cam-bridge: Chapman and Hall, 1983; 2 edn: 79-98.

38 Laitinen LA, Heino M, Laitinen TK, Haahtela T. Damageof the airway epithelium and bronchial reactivity in patientswith asthma. Am Rev Respir Dis 1985; 131:599-606.

39 Lundgren R, Soderberg M, Hostedt P, Sterling R. Morpho-logical studies of bronchial mucosal biopsies from asth-matics before and ten years of treatment with inhaledsteroids. Eur Respir J 1988; 1:883-9.

40 Elia C, Bucca C, Rolla G, Scappaticci E, Cantino D. Afreeze-fracture study of human bronchial epithelium innormal, bronchitic and asthmatic subjects. J SubmicroscCytol Pathol 1988; 20:509-17.

41 Lozewicz S, Gomez E, Ferguson H, Davies RJ. Inflamma-tory cells in the airways in mild asthma. Br Med J 1988;297:1515-16.

42 Jeffery PK, Wardlaw AJ, Nelson FC, Collins JV, Kay AB.Bronchial biopsies in asthma. An ultrastructural, quantita-tive study and correlation with hyperreactivity. Am RevRespir Dis 1989; 140:1745-53.

43 Roche WR, Beasley R, Williams JH, Holgate ST. Subepithe-lial fibrosis in the bronchi of asthmatics. Lancet 1989:520-4.

44 Azzawi M, Bradley B, Jeffery PK, Frew AJ, Wardlaw AJ,

196 H. Smith

Knowles G, Assoufi B, Collins JV, Durham S, Kay AB,Identification of activated T lymphocytes and eosinophils inbronchial biopsies in stable atopic asthma. Am Rev RespirDis 1990; 142:1407-13.

45 Djukanovic R, Wilson JW, Britten KM, Wilson SJ, WallsAF, Roche WR, Howarth PH, Holgate ST, Quantitation ofmast cells and eosinophils in the bronchial mucosa ofsymptomatic atopic asthmatics and healthy control subjectsusing immunohistochemistry. Am Rev Respir Dis 1990;142:863-71,

46 Lozewicz S, Wells C, Gomez E, Ferguson H, Richfrian P,Devalia J, Davies RJ. Morphological integrity of thebronchial epithelium in mild asthma. Thorax 1990; 45:12-15,

47 Poulter LW, Power C, Burke C, The relationship betweenbronchial immunopathology and hyperresponsiveness inasthma. Eur Respir J, 1990; 3:792-9,

48 Brewster CEP, Howarth PH, Djukanovic R, Wilson J,Holgate ST, Myofibroblasts and subepithelial fibrosis inbronchial asthma. Am J Respir Cell Mol Biol 1990; 3:507-11,

49 Heard BE, Jeffery PK, Kay AB, Quantitation of mast cellsand eosinophils in the bronchial mucosa of symptomaticatopic asthmatics and healthy control subjects using immu-nohistochemistry. Am Rev Respir Dis 1991; 143:1200-1,

50 Laitinen A, Laitinen LA. Cellular infiltrates in asthma andchronic obstructive pulmonary disease. Am Rev Respir Dis1991; 143:1159-60,

51 Mattoli S, Mattoso VL, Soloperto M, Allegra L, Fasoli A,Cellular and biochemical characteristics of bronchoalveolarlavage fluid in symptomatic nonallergic asthma, J AllergyClin Immunol 1991; 87:794-802,

52 Kay AB, Lymphocytes in asthma, Resp Med 1991; 85:87-90,

53 Snashall PD, Mechanisms of hypersensitivity. Generalreview. In: Nadel JA, Pauwels R, Snashall PD, eds. Bron-chial hyperresponsiveness. Normal and abnormal controlassessment and therapy. London: Blackwell, 1987:259-314,

54 Jongejan RC, De Jongste JC, Kerrebijn KF, The increasedresponsiveness to inhaled methacholine in asthma; combina-tion of causative factors. Am Rev Respir Dis 1991;143:1194-6,

55 Clarke PS, The effect of beclomethasone dipropionate onbronchial hyperreaetivity, J Asthma 1982; 19:91-3,

56 Kraan J, Koeter GH, Mark Th Wrd, Sluiter HJ, de Vries K,Changes in the bronchial hyperreaetivity induced by 4 weeksof treatment with antiasthmatic drugs in patients withallergic asthma, A comparison between budesonide andterbutaline, J Allergy Clin Immunol 1985; 76:628-36,

57 Kerrebijn KF, van Essen-Zandvliet EEM, Neijens HJ, Theeffect of long term treatment and inhaled corticosteroids onthe bronchial responsiveness in children with asthma, JAllergy Clin Immunol 1987; 79:653-9,

58 Dutoit JI, Salome CM, Woolcock AJ, Inhaled cortico-steroids reduce the severity of bronchial hyperresponsive-ness in asthma but oral theophylline does not. Am RevRespir Dis 1987; 136:1174-8,

59 Woolcock AJ, Yan K, Salome CM, Effect of therapy onbronchial hyperresponsiveness in the long term manage-ment of asthma, Clin Allergy 1988; 18:165-76,

60 Laitinen LA, Laitinen A, Heino M, Haahtela T, Eosinophi-lic airway inflammation during exacerbation of asthma andits treatment with inhaled corticosteroids. Am Rev RespirDis 1991; 143:423-7,

61 Bel EH, Timmers MC, Zwinderman AH, Dijkman JH, SterkPJ, The effect of inhaled corticosteroids on the maximaldegree of airway narrowing to methacholine in asthmaticssubjects. Am Rev Respir Dis 1991; 143:109-13,

62 Boner AL, Piacentini GL, Bonizzato C, Dattoli V, Sette L.Effect of inhaled beclomethasone disproprionate on bron-chial hyperreaetivity in asthmatic children during maximalallergen exposure, Ped Pulmon 1991; 10:2-5,

63 Easton JG, Effect of inhaled corticosteroids on methacho-line airway reactivity, J Allergy Clin Immunol 1981; 67:388-90,

64 Featherstone RL, Hutson PA, Holgate ST, Church MK,Active sensitisation of guinea-pig airways in vivo enhancesin vivo and in vitro responsiveness, Eur Respir J 1988; 1:839-45.

65 Spicer BA, Baker RC, Hatt PA, Laycock SM, Smith H, Theeffects of drugs on Sephadex induced eosinophilia and lunghyper-responsiveness in the rat, Br J Pharmacol 1990;101:821-8,

66 De Monchy JGR, Kauffman HF, Venge P, Koeter GH,Jansen HM, Sluiter HJ, De Vries K, Bronchoalveolareosinophilia during allergen-induced late asthmatic reac-tions. Am Rev Respir Dis 1985; 131:373-6,

67 Diaz PM, Gonzalez MC, Galleguillos FR, Ancic P, Crom-well O, Shepherd D, Durham SR, Gleich GJ, Kay AB,Leukocytes and mediators in bronchoalveolar lavage duringallergen-induced late-phase asthmatic reactions. Am RevRespir Dis 1989; 139:1383-9.

68 Cockroft DW, Murdock KY, Changes in bronchial respon-siveness to histamine at intervals after allergen challenge.Thorax 1987; 42:302-8,

69 Durham SR, Craddock CF, Cookson WO, Benson MK,Increases in airway responsiveness to histamine precedeallergen-induced late asthmatic responses, J Allergy ClinImmunol 1988; 82:764-70,

70 Gibson PG, Manning PJ, O'Byrne PM, Girgis-Gabardo A,Dolovich J, Denburg JA, Hargreave FE, Allergen-inducedasthmatic responses. Relationship between increases inairway responsiveness and increases in circulating eosino-phils, basophils and their progenitors. Am Rev Respir Dis1991; 143:331-5.

71 Flavahan NA, Slifman NR, Gleich GJ, Vanhoute PM.Human eosinophil major basic protein causes hyperreaeti-vity of respiratory smooth muscle. Am Rev Respir Dis 1988;138:685-8,

72 Brofman JD, White SR, Blake JS, Munoz NM, Gleich GJ,Leff AR, Epithelium augmentation of trachealis contractioncaused by major basic protein of eosinophils, J Appl Physiol1989; 66:1867-73,

73 De Jongste JC, Mons H, Bonta IL, Kerrebijn KF, In vitro

Asthma, inflammation, eosinophils and bronchial hyperresponsiveness 197

responses of airways from an asthmatic patient, Eur J RespirDis 1987; 71:23-9,

74 Bai TJ, Abnormalities in airway smooth muscle in fatalasthma. Am Rev Respir Dis 1990; 141:552-7,

75 Cuss FM, Dixon CM, Barnes PJ, Effect of platelet activatingfactor on pulmonary function and bronchial responsivenessin man. Lancet 1986; 2:189-92,

76 Jenkins JR, Lai CKW, Holgate ST, Effect of increasingdoses of platelet-activating factor (PAF) on normal humanairways (abstract), J Allergy Clin Immunol 1989; 83:282A,

77 Tai P-C, Ackerman SJ, Spry CJ, Dunnette S, Olsen EG,Gleich GJ, Deposits of eosinophil granule proteins incardiac tissues of patients with eosinophilic endomyocardialdisease. Lancet 1987; 1:643-7,

78 Pincus SH, Ramesh KS, Wyler DJ, Eosinophils stimulatefibroblast DNA synthesis. Blood 1987; 70:572-4,

79 Sarnstrand B, Westergren-Thorsson G, Hernas J, PetersonC, Venge P, Malmstrom A, Eosinophil cationic protein andtransforming growth factor-A stimulates synthesis of hya-luronan and proteoglycan in human fibroblast cultures,Proc Intern Colloquium on Pulmonary Fibrosis 1988,

80 Manning PJ, Watson RM, Margolskee DJ, Williams VC,Schwarz JI, O'Byrne PM, Inhibition of exercise-inducedbronchoconstriction by MK-571 a potent leukotriene D4-receptor antagonist, N Engl J Med 1990; 323:1736-9,

81 Taylor IK, O'Shaughnessy KM, Fuller RW, Dollery CT,

Effect of a cysteinyl-leukotriene receptor antagonist ICI 204,219 on allergen induced bronchoconstriction and airwayhyper-reactivity in atopic subjects, Proc, Thoracic SocWinter Meeting 1990; 43, P47,

82 Eiser N, Denman W, Hayhurst M, Effect of histamine andleukotriene antagonists on antigen induced asthma, Allergo-logie 1989; 12 (Suppi):Abst FC2601,

83 Feinmark SJ, Cannon PJ, Endothelial cell leukotriene C4synthesis results from intracellular transfer of leukotriene A4synthesised by polymorphonuclear leucocytes, J Biol Chem1986; 261:16466-72,

84 Czop JK, Austen KF, Generation of leukotrienes by humanmonocytes upon stimulation of their beta glucan receptorduring phagocytosis, Proc Natl Acad Sci USA 1985;82:2751-5,

85 Laviolette M, Coulombe R, Picard S, Braquet P, Borgeat P,Decreased leukotriene B4 synthesis in smokers alveolarmacrophages in vitro. J Clin Invest 1986; 77:54-60,

86 Schleimer RP, Macglasham DW, Peters SP, Pinckard PN,Adkinson NF, Lichtenstein LM, Characterisation ofinflammatory mediator release from purified human lungmast cells. Am Rev Respir Dis 1986; 133:614-17,

87 Moqbel R, Macdonald AJ, Cromwell O, Kay AB, Release ofleukotriene C4 (LTC4) from human eosinophils followingadherence to IgE and IgG coated schistosomula of Schisto-soma mansoni. Immunology 1990; 69:435-42,