The remodelled tracheal basement membrane zone of infant rhesus monkeysafter 6 months of recovery

M. J. Evans, M. V. Fanucchi, G. L. Baker, L. S. Van Winkle, L. M. Pantle, S. J. Nishio, E. S. Schelegle,L. J. Gershwin, L. A. Miller, D. M. Hyde and C. G. PlopperDepartment of Anatomy, Physiology & Cell Biology; Center for Comparative Respiratory Biology and Medicine, School of Veterinary Medicine,University of California, Davis, CA, USA

SummaryBackground In previous studies, we showed that repeated exposure to (1) house dust mite allergen

(HDMA) (Dermatophagoides farinae) caused thickening of the basement membrane zone (BMZ) and

(2) HDMA1ozone (O3) caused depletion of BMZ perlecan and atypical development of BMZ

collagen (irregular thin areaso2.0mm in width).

Objective The purpose of this study was to determine if these remodelling changes were reversible

after 6 months of recovery.

Methods Rhesus monkeys were exposed to a regimen of HDMA and or O3 or filtered air (FA) for 6

months. After the exposure protocol was completed FA and O3 groups were allowed to recover in

FA for 6 months. The HDMA and HDMA1O3 exposure groups recovered in a modified

environment. They were re-exposed to HDMA aerosol for 2 h at monthly intervals during recovery

in order to maintain sensitization for pulmonary function testing. To detect structural changes in the

BMZ, collagen I and perlecan immunoreactivity were measured and compared to data from the

previous papers.

Results The remodelled HDMA group had a significantly thicker BMZ and after 6 months of

recovery the width had not regressed. In the remodelled BMZ of the HDMA1O3 group, perlecan

had returned to the BMZ after 6 months of the recovery protocol, and the thin, irregular, collagen

BMZ had been resolved.

Conclusion In summary, this study has shown that: (1) The width of the remodelled HDMA BMZ

did not regress during a recovery protocol that included a sensitizing dose of HDMA. (2) The

atypical collagen BMZ in the HDMA1O3 BMZ was resolved in the absence of O3. (3) Depletion of

perlecan from the BMZ by O3 was reversed by recovery in the absence of O3.

Keywords basement membrane zone, collagen I, house dust mite allergen (Dermatophagoides farinae),

ozone, perlecan, recovery, remodelling

Submitted 3 November 2003; revised 8 March 2004; accepted 13 April 2004

Introduction

The basement membrane zone (BMZ) is the central structureof the epithelial–mesenchymal trophic unit [1–3]. Exchange ofinformation between the epithelium and fibroblasts occurs viathe BMZ. The BMZ appears as three component layers withtransmission electron microscopy, the lamina lucida, thelamina densa and the lamina reticularis (LR). Together theymake up the basal lamina. The LR is especially pronouncedunder the respiratory epithelium of large conducting airways,where it may be several microns thick. Collagen type I, IIIand V form heterogeneous fibers that account for thethickness of the LR. Attenuated fibroblasts beneath theBMZ are thought to synthesize the collagen components ofthe BMZ [1–3]. The collagen fibers are arranged as a mat of

large fibers oriented along the longitudinal axis of the airway.Smaller fibers are cross-linked with the larger fibers tocomplete this structure [4]. Remodelling of the epithelialBMZ, such as occurs in asthma, involves increased depositionof collagen in the BMZ [5–8]. Thickening of the BMZ isthought to protect against airway narrowing and air trapping[9].The BMZ has a number of functions in the epithelial-

mesenchymal trophic unit. It is specialized for attachment ofepithelium with the extracellular matrix; it also serves as abarrier, binds specific growth factors, hormones and ions, isinvolved with electrical charge and plays a critical role in cell–cell communication [10–12]. Heparan sulphate proteoglycans(perlecan), and chondroitin sulphate proteoglycans (bama-can) are an intrinsic part of the BMZ that are involved withmost of its functions [13–15]. Binding and storage of growthfactors by perlecan (mainly fibroblast growth factor-2) is alsoan important function of the BMZ. It is not known howremodelling affects the various functions of the BMZ.

Correspondence: Michael J Evans, Department of Anatomy Physiology

and Cell Biology, School of Veterinary Medicine, University of California

Davis, CA 95616, USA. E-mail: mevans@ucdavis.edu

Clin Exp Allergy 2004; 34:1131–1136 doi:10.1111/j.1365-2222.2004.02004.x

r 2004 Blackwell Publishing Ltd 1131

In our previous studies, we found that development of theepithelial BMZ in the trachea occurred postnatally in therhesus monkey [16]. The collagen BMZ increased in widthfrom 1 to 6 months, and perlecan was localized in the BMZ atall stages of development. We found significant thickening ofthe tracheal BMZ in rhesus monkeys treated with house dustmite allergen (HDMA) during this period of development.We also found that all tracheal samples expressed thin focalareas of the BMZ associated with leucocyte trafficking. In thethin areas, the collagen BMZ was damaged and depleted ofperlecan. In monkeys exposed to either O3 or O31HDMA,the tracheal BMZ was significantly different during this samedevelopmental period. The collagen BMZ was irregular andthin in many areas, and perlecan was depleted or severelyreduced in BMZ. These changes, were present throughout theentire BMZ and not associated with only areas of leucocytetrafficking, as seen in animals exposed to HDMA [8]. Onlythe epithelial BMZ was depleted of perlecan; the BMZaround blood vessels and smooth muscle cells was noteffected. These results indicate that ozone (O3) can amplifythe allergic and structural remodelling effects of HDMAsensitization and inhalation.The purpose of this study was to determine: (1) if the

remodelling changes seen in the BMZ immediately after anexposure regimen of HDMA regressed after 6 months of amodified recovery protocol that included a monthly sensitiza-tion dose of HDMA, (2) did perlecan return to the BMZ duringrecovery in the absence of O3 and (3) was the atypical BMZ inthe HDMA1O3 group resolved during recovery in the absenceof O3. We found that the thickened BMZ in the HDMA groupdid not regress in the modified recovery protocol but remainedsignificantly wider than the filtered air (FA) group. Howeverdepletion of perlecan and the atypical collagen BMZ in theHDMA1O3 group was resolved during recovery.

Materials and methods

Experimental animals

All monkeys selected for these studies were CaliforniaRegional Primate Research Center colony-born rhesusmacaques (Macaca mulatta). Care and housing of animalscomplied with the provisions of the Institute of LaboratoryAnimal Resources and conforms to practices established bythe American Association for Accreditation of LaboratoryAnimal Care (AAALAC).Sixteen infant rhesus monkeys (30 days old) were exposed

to 11 episodes of either filtered air (FA), HDMA allergenaerosol, O3 or HDMA1O3 (5 days each followed by 9 days ofFA). O3 was delivered for 8 h/day at 0.5 p.p.m. Eight of themonkeys (HDMA and HDMA1O3 groups) were sensitizedto HDM allergen (Dermatophagoides farinae) at age 14 and28 days, by subcutaneous inoculation (s.q.) of HDMA inalum and intramuscular injection of heat-killed Bordetellapertussis cells. HDMA sensitization was confirmed via skintesting with sq HDMA on day 38 of the exposure protocol.Sensitized monkeys were exposed to HDMA aerosol for 2 h/day on days 3–5 of either FA (HDMA, n5 4) or O3

(HDMA1O3, n5 4) exposure. Non-sensitized monkeys wereexposed to either FA (FA, n5 4) or O3 (O3, n5 4). After the

exposure protocol the FA and O3 groups were allowed torecover in FA for 6 months. The HDMA and HDMA1O3

exposure groups were re-exposed to HDMA aerosol for 2 h atmonthly intervals during recovery in order to maintainsensitization for pulmonary function testing. At the end ofthe sixth month, they were challenged with HDMA aerosolfor 2 h/day for 3 days, followed by pulmonary functiontesting. It should be noted that this modified recoveryprotocol is not the same as a true FA recovery protocol.

Preparation of animals

Following the recovery protocol, monkeys were euthanizedwith an overdose of pentobarbital after being sedated withTelazol (8mg/kg i.m.) and anesthetized with Diprivan (0.1–0.2mg/kg/min, i.v.). Monkeys were then necropsied followingexsanguination through the abdominal aorta and the lungsprepared for analysis as previously described [23]. Trachealsamples were sliced into rings perpendicular to the long axisof the airway.

Immunohistochemistry

Tracheal slices were fixed in 1.0% paraformaldehyde for onehour and embedded in paraffin. For routine histology 5mmsections were stained with haematoxylin and eosin (H&E).For immunohistochemistry, 5mm section were deparaffinizedin xylene, hydrated in ethanol and washed in phosphate-buffered saline (PBS). For collagen I staining, sections weretreated with pepsin (1.0mg pepsin/mL 3.0% acetic acid) at37 1C for 2 h, and blocked with bovine serum albumin, andincubated with antibody to collagen I (1 : 250), (rabbit, antihuman polyclonal antibody; Biogensis, Kingston, NH, USA)overnight at 4 1C. For perlecan staining, the sections weretreated with 0.1% pronase in PBS for 30min, rinsed innanopure water followed by PBS, blocked with bovine serumalbumin for 30min, and incubated with an antibody toperlecan (1 : 2000; mouse, anti human monoclonal antibody,clone 7B5; Zymed, San Francisco, CA, USA) overnight at4 1C. Following immunohistochemistry the sections werewashed in PBS, treated with the secondary antibody(1 : 1000; Alexa Fluor 568, Molecular Probes, Inc., Eugene,OR, USA) for 30min, washed in PBS and the coverslipmounted in fluorescent safe media (GelMount, Foster City,CA, USA). Fluorescence was visualized on an Olympus BH-2fluorescent microscope (Olympus America Inc., Melville, NY,USA).

Antibody specificity

The antibody for collagen type I had negligible cross-reactivity with other collagen or non-collagen matrix proteins(per supplier). The antibody for perlecan may cross-react withthe short arm of laminin A and B chains [17].

Quantitation

The width of the tracheal collagen I BMZ was measuredmorphometrically to quantitate the immunohistochemicalresults. An observer blinded to the study groups made themeasurements. In a previous report using human biopsysamples it was demonstrated that 31–45 measurements, at

1132 M. J. Evans et al.

r 2004 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 34:1131–1136

least 20mm apart and covering 1000mm BMZ are necessary togive an accuracy of � 15.0% [18]. In this study eightmicrographs were taken equidistant apart around thecircumference of the tracheas. The width of the BMZ wasmeasured at four points, 50mm apart, on each micrograph. Atotal of 1600mm of BMZ was sampled in each tracheal ring inthis manner. The average width of the BMZ was determinedfrom these measurements for each animal. The percentageof the measurements that were less than 2.0mm were alsodetermined for each animal. The mean and standard de-viation (SD) of BMZ width and the percentage of measure-ments less than 2.0mm in width were then determined for eachgroup of animals and compared with data from a previousstudy [19].

Semiquantitation

The intensity of immunoreactivity for perlecan expression intrachea was graded on a scale of 0–3 for each animal [19, 20].A scale of 0 indicates no staining beyond background, 1indicates isolated areas of staining in the BMZ, 2 indicatesstaining in 25–75% of the BMZ, and 3 indicates strongstaining in most of the BMZ. Analysis of the tissues wasperformed by an observer blinded to the study groups. Themean and SD of perlecan expression for each group ofanimals was then determined and compared with data from aprevious paper [19]. Differences between groups were de-termined by one-way analysis of variance and determinationof significance was based on Holm–Sidak as Po0.05 (SigmaStats SPSS Science, Chicago, IL, USA).

Results

Collagen I

Immunohistochemistry for collagen I is used to demonstratethe collagen framework of the BMZ (Fig. 1a). The widths ofthe collagen I BMZ after 6 months of the recovery protocolare given in Fig. 2. These results are compared with thecollagen I BMZ widths measured immediately after treatmentgiven in a previous paper [8]. The width of the BMZ in theFA group was 4.4 � 0.5mm immediately after treatment.After 6 months of the recovery protocol the width of theBMZ had increased (5.3 � 0.6mm) but not significantly overthe previous group. In the HDMA group the width of theBMZ after treatment (6.3 � 0.8mm) was significantly greaterthan the width of the BMZ in the FA group immediately aftertreatment. After 6 months of the modified recovery protocolthe width of the BMZ (7.4 � 1.1mm) was still significantlygreater than the FA group indicating that the collagenframework of the BMZ had not regressed.In the O3 group, the width of the BMZ was significantly

greater than the FA group immediately after treatment(5.7 � 1.2mm). After 6 months of the recovery protocol, thewidth of the BMZ had increased but was not significantlywider (6.7 � 0.9mm) than the FA group. In the groupexposed to HDMA1O3 the structure of the BMZ was moreirregular immediately after the exposure protocol with manyareas of thin BMZ (Fig. 1b). In the HDMA1O3 group16.7 � 6.1% of the measurements of BMZ width were less

2.0mm indicating atypical development of the BMZ [19]. Dueto the thin areas of the BMZ, the width (4.7 � 1.1mm) wasnot significantly different from the FA group immediatelyafter treatment. However after 6 months of the recoveryprotocol areas of the BMZ less than 2.0mm had declined to0.7 � 1.6% indicating that the thin areas had been filled withcollagen (Fig. 3). The BMZ was now more uniform in width(8.9 � 2.0mm) and it was significantly wider than the FAgroup (Fig. 1). These data suggest that the molecular signal tosynthesize a thick BMZ in the HDMA group persisted in theHDMA1O3 group during the 6 months of the modifiedrecovery protocol when the thin areas were filled withcollagen fibers.

Perlecan

Immunohistochemistry for perlecan is measured to evaluatethe heparan sulfate proteoglycan component of the BMZ

Fig. 1. (a) Example of collagen I immunoreactivity in a thick region ofthe basement membrane zone (BMZ) (arrowheads) of a house dust miteallergen (HDMA)-treated monkey. Both of the surfaces are irregular.(b) Example of collagen I immunoreactivity in the BMZ of an O31HDMA-treated monkey. Some areas of the BMZ appear normal (arrowheads)and other areas are thin (o2.0 mm) and incomplete (arrows).

Recovery of the basement membrane zone 1133

r 2004 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 34:1131–1136

(Figs 4a and b). In our previous study we found immediatelythat after completing the exposure protocol, perlecan waspresent throughout the BMZ of the FA and HDMA groups,however, it was absent in the BMZ of the O3 and HDMA1O3

groups [19]. The lack of perlecan in the BMZ of the O3 groupswas associated with atypical development of the BMZ. Aftersix months of the recovery protocol perlecan immunoreactiv-ity had returned to the BMZ in both of the O3 groups (Fig. 5).

Discussion

The purpose of this study was to determine: (1) if theremodelling changes seen in the BMZ immediately after

exposure to a regimen of HDMA regressed after 6 months ofthe modified recovery protocol and (2) if the atypical BMZ inthe O3 groups is resolved during the modified recoveryprotocol. After 6 months the width of the BMZ in theHDMA group was still significantly greater (Po0.05) thanthe treatment FA group. These findings demonstrate that theremodelled collagen framework of the BMZ did not regress.In the HDMA1O3 group atypical development of the BMZwas characterized by thin areas of the BMZ after treatment.After 6 months of the modified recovery protocol, these areashad filled in and the BMZ was now as wide as the HDMAgroup. These findings indicate that the atypical collagen BMZhad been resolved during the modified recovery protocol.However it was now as thick as the BMZ in the HDMAgroup suggesting that the signal to develop a thick BMZpersisted during the modified recovery protocol.

Fig. 2. Graphic representation of the width of the collagen I basementmembrane zone (BMZ) immediately after treatment and after 6 monthsof the recovery protocol (mean � SD). The width of the BMZ remainedelevated during the recovery protocol. In the hose dust mite allergen(HDMA), and HDMA1O3 groups the width increased significantly duringrecovery indicating that the signal for development of a thick BMZ persistedduring recovery. (*Po0.05 when compared to treatment filtered air (FA);wPo0.05 when compared to 6 months recovery FA.)

Fig. 3. Graphic representation of the percentage of the collagen Ibasement membrane zone (BMZ) that was less than 2.0 mm in width(mean � SD). There was a significant decrease in the number of BMZmeasurements that were less than 2.0 mm in width after 6 months of therecovery protocol in the O3 and house dust mite (HDMA)1O3 groups. In theHDMA group none of the measurements were less than 2.0 mm in width.These data indicate that the thin areas of the BMZ had been resolvedduring recovery. (*Po0.05 vs. the treatment group).

Fig. 4. (a) Perlecan immunoreactivity in the basement membrane zone(BMZ) of a house dust mite allergen (HDMA)-treated monkey. There isstrong immunoreactivity in the epithelial BMZ (arrowheads) and in the wallsof blood vessels (arrows). (b) In the epithelial BMZ of an HDMA1O3-treatedmonkey (arrowheads) there is very little perlecan immunoreactivity,however, in the walls of blood vessels (arrows) it remained strong.Bar5 10 mm.

1134 M. J. Evans et al.

r 2004 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 34:1131–1136

The mechanism of collagen accumulation in the BMZ is notcompletely understood. Some reports suggest an increase inthe number of attenuated fibroblasts/myofibroblasts sur-rounding the BMZ account for the increased thickness ofthe BMZ [2, 3, 6, 19]. The signals that cause the increase incell number and also release of cytokines initating collagensynthesis are likely to come from epithelial cells implicatingthe epithelium in the process of BMZ remodelling [21–28].Another study showed that asthmatic bronchial fibroblastshad a decreased capacity to degrade collagen due to animbalance in metalloproteinase enzyme regulation [29]. Inthis study, it was not determined if the signal to maintain athick BMZ persisted because of permanent molecular changesin epithelial cells, the layer of attenuated fibroblasts or if themonthly sensitizing dose of HDMA in the modified recoveryprotocol was enough to maintain the thickened BMZ.Immediately after treatment perlecan had been depleted

from the BMZ in both O3 groups. During the recoveryprotocol perlecan was re-incorporated into the BMZ of theboth groups. Perlecan depletion in the BMZ has beenreported in perlecan knockout mice [30], lung tumors(squamous-cell carcinoma and adenocarcinoma) [31], andatheroslerotic blood vessels [32]. Also in a previous study,total proteoglycans were shown to be decreased in ratschronically exposed to O3 [33]. In perlecan knockouts, there isBMZ deterioration in blood vessels in areas of mechanicalstress (such as the heart) that leads to bleeding and death ofthe animal [30]. These studies demonstrate that perlecandepletion is associated with abnormal development of theBMZ, similar to what we found in monkeys exposed to eitherO3 or HDMA1O3. In the absence of O3, perlecan returned tothe BMZ, and the thin regions of the collagen BMZ wereresolved. These findings indicate that O3-induced depletion ofperlecan could be reversed in the absence of O3. These resultsalso indicate that perlecan is necessary for normal develop-ment of the BMZ.In summary, this study has shown that: (1) The width of the

remodelled HDMA BMZ did not regress during a recoveryprotocol that included a sensitizing dose of HDMA. (2) The

atypical collagen BMZ in the HDMA1O3 BMZ was resolvedin the absence of O3. (3) Depletion of perlecan from the BMZby O3 was reversed by recovery in the absence of O3.

Acknowledgements

Supported by NIH grants PO1 ES00628, PO1 ES11617,ES04311, ES06700, ES05707, RR000169.

References

1 Evans MJ, Guha SC, Cox RA, Moller PC. Attenuated fibroblast

sheath around the basement membrane zone in the trachea. Am J

Respir Cell Mol Biol 1993; 8:188–92.

2 Evans MJ, Van Winkle LS, Fanucchi MV, Plopper CG. The

attenuated fibroblast sheath of the respiratory tract epithelial–

mesenchymal trophic unit. Am J Respir Cell Mol Biol 1999;

21:655–7.

3 Holgate ST, Davies DE, Lackie PM, Wilson SJ, Puddicombe SM,

Lordan JL. Epithelial–mesenchymal interactions in the pathogen-

esis of asthma. J Allergy Clin Immunol 2000; 105:193–204.

4 Evans MJ, Van Winkle LS, Fanucchi MV et al. Three-dimensional

organization of the lamina reticularis in the rat tracheal

basement membrane zone. Am J Respir Cell Mol Biol 2000;

22:393–7.

5 Bousquet J, Jeffery PK, Busse WW, Johnson M, Vignola AM.

Asthma. From bronchoconstriction to airways inflammation and

remodeling. Am J Respir Crit Care Med 2000; 161:1720–45.

6 Brewster CE, Howarth PH, Djukanovic R, Wilson J, Holgate ST,

Roche WR. Myofibroblasts and subepithelial fibrosis in bronchial

asthma. Am J Respir Cell Mol Biol 1990; 3:507–11.

7 Redington AE. Fibrosis and airway remodelling. Clin Exp Allergy

2000; 30 (Suppl. 1):42–5.

8 Evans MJ, Van Winkle LS, Fanucchi MV et al. Fibroblast growth

factor-2 in remodeling of the developing basement membrane zone

in the trachea of infant rhesus monkeys sensitized and challenged

with allergen. Lab Invest 2002; 82:1747–54.

9 Milanese M, Crimi E, Scordamaglia A et al. On the functional

consequences of bronchial basement membrane thickening. J Appl

Physiol 2001; 91:1035–40.

10 Adachi E HI, Hayashi T. Basement-membrane stromal relation-

ships: interactions between collagen fibrils and the lamina densa.

Int Rev Cytol 1997; 173:73–156.

11 Crouch EC, Martin GR, Brody JS, Laurie GW. Basement

membranes. In: Crystal RG, West JB, Wiebel ER, Barnes PJ, eds.

Lung. Philadelphia: Lippencott Raven, 1997; 769–91.

12 Sannes PL, Wang J. Basement membranes and pulmonary

development. Exp Lung Res 1997; 23:101–8.

13 Roberts CR, Wright TN, Hascall VC. Proteoglycans. In: Crystal

RG, West JB, Wiebel ER, Barnes PJ, eds. Lung. Philadelphia:

Lippincott Raven, 1997; 757–67.

14 Sannes PL, Burch KK, Khosla J. Immunohistochemical localiza-

tion of epidermal growth factor and acidic and basic fibroblast

growth factors in postnatal developing and adult rat lungs. Am J

Respir Cell Mol Biol 1992; 7:230–7.

15 Sannes PL, Burch KK, Khosla J, McCarthy KJ, Couchman JR.

Immunohistochemical localization of chondroitin sulfate,

chondroitin sulfate proteoglycan, heparan sulfate proteoglycan,

entactin, and laminin in basement membranes of postnatal

developing and adult rat lungs. Am J Respir Cell Mol Biol 1993;

8:245–51.

16 Evans MJ, Fanucchi MV, Van Winkle LS et al. Fibroblast growth

factor-2 during postnatal development of the tracheal basement

Fig. 5. Graphic representation of perlecan immunoreactivity expressionscores (mean � SD). In the O3 treatment groups there was a significantincrease in the perlecan scores following 6 months of the recovery protocol.These findings indicate that perlecan depletion in the basement membranezone (BMZ) was resolved during recovery in the O3 groups. (*Po0.05vs. the treatment group).

Recovery of the basement membrane zone 1135

r 2004 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 34:1131–1136

membrane zone. Am J Physiol Lung Cell Mol Physiol 2002;

283:L1263–70.

17 Murdoch AD, Liu B, Schwarting R, Tuan RS, Iozzo RV.

Widespread expression of perlecan proteoglycan in basement

membranes and extracellular matrices of human tissues as detected

by a novel monoclonal antibody against domain III and by in situ

hybridization. J Histochem Cytochem 1994; 42:239–49.

18 Sullivan P, Stephens D, Ansari T, Costello J, Jeffery P. Variation in

the measurements of basement membrane thickness and inflam-

matory cell number in bronchial biopsies. Eur Respir J 1998;

12:811–5.

19 Evans MJ, Fanucchi MV, Baker GL et al. Atypical development of

the tracheal basement membrane zone of infant rhesus monkeys

exposed to ozone and allergen. Am J Physiol Lung Cell Mol

Physiol 2003; 285:L931–9.

20 Kranenburg AR, De Boer WI, Van Krieken JH et al. Enhanced

expression of fibroblast growth factors and receptor FGFR-1

during vascular remodeling in chronic obstructive pulmonary

disease. Am J Respir Cell Mol Biol 2002; 27:517–25.

21 Morishima Y, Nomura A, Uchida Y et al. Triggering the induction

of myofibroblast and fibrogenesis by airway epithelial shedding.

Am J Respir Cell Mol Biol 2001; 24:1–11.

22 Tschumperlin DJ, Shively JD, Kikuchi T, Drazen JM. Mechanical

stress triggers selective release of fibrotic mediators from bronchial

epithelium. Am J Respir Cell Mol Biol 2003; 28:142–9.

23 Sun G, Stacey MA, Bellini A, Marini M, Mattoli S. Endothelin-1

induces bronchial myofibroblast differentiation. Peptides 1997;

18:1449–51.

24 Zhang S, Smartt H, Holgate ST, Roche WR. Growth factors

secreted by bronchial epithelial cells control myofibroblast

proliferation: an in vitro co-culture model of airway remodeling in

asthma. Lab Invest 1999; 79:395–405.

25 Hastie AT, Kraft WK, Nyce KB et al. Asthmatic epithelial cell

proliferation and stimulation of collagen production: human

asthmatic epithelial cells stimulate collagen type III production by

human lung myofibroblasts after segmental allergen challenge. Am

J Respir Crit Care Med 2002; 165:266–72.

26 Richter A, Puddicombe SM, Lordan JL et al. The contribution of

interleukin (IL)-4 and IL-13 to the epithelial-mesenchymal trophic

unit in asthma. Am J Respir Cell Mol Biol 2001; 25:385–91.

27 Nakao A, Sagara H, Setoguchi Y et al. Expression of Smad7 in

bronchial epithelial cells is inversely correlated to basement

membrane thickness and airway hyperresponsiveness in patients

with asthma. J Allergy Clin Immunol 2002; 110:873–8.

28 Xu J, Benyon RC, Leir SH, Zhang S, Holgate ST, Lackie PM.

Matrix metalloproteinase-2 from bronchial epithelial cells induces

the proliferation of subepithelial fibroblasts. Clin Exp Allergy 2002;

32:881–8.

29 Laliberte R, Rouabhia M, Bosse M, Chakir J. Decreased capacity

of asthmatic bronchial fibroblasts to degrade collagen. Matrix Biol

2001; 19:743–53.

30 Costell M, Gustafsson E, Aszodi A et al. Perlecan maintains the

integrity of cartilage and some basement membranes. J Cell Biol

1999; 147:1109–22.

31 Nackaerts K, Verbeken E, Deneffe G, Vanderschueren B, Demedts

M, David G. Heparan sulfate proteoglycan expression in human

lung-cancer cells. Int J Cancer 1997; 74:335–45.

32 Pillarisetti S. Lipoprotein modulation of subendothelial heparan

sulfate proteoglycans (perlecan) and atherogenicity. Trends Car-

diovasc Med 2000; 10:60–5.

33 Radhakrishnamurthy B. Consequences of prolonged inhalation of

ozone on F344/N rats: collaborative studies. Part III: effects on

complex carbohydrates of lung connective tissue. Res Rep Health

Eff Inst 1994: 3–14; discussion 5–23.

1136 M. J. Evans et al.

r 2004 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 34:1131–1136