ADAM33 Expression in Asthmatic Airways and Human Embryonic Lungs Hans Michael Haitchi, Robert M. Powell, Timothy J. Shaw, Peter H. Howarth, Susan J. Wilson, David I. Wilson, Stephen T. Holgate, and Donna E. Davies Roger Brooke Laboratories and Histochemistry Research Unit, Allergy and Inflammation Research, Division of Infection, Inflammation, and Repair, and Human Genetics Division, School of Medicine, University of Southampton, Southampton General Hospital, Southampton, United Kingdom

Correspondence and requests for reprints should be addressed to Hans Michael Haitchi, M.D., M.Med., Allergy and Inflammation Research, Division of Infection, Inflammation, and Repair, School of Medicine, University of Southampton, Tremona Road, Mail Point 810, Southampton General Hospital, Southampton SO16 6YD, UK. E-mail: [email protected] or [email protected]

by an increase in BHR that is only partially or nonresponsive to treatment with corticosteroids (2). Asthma is a complex, genetically inherited disease that requires environmental interactions to be fully expressed; however, the biochemical and molecular mechanisms resulting in asthma are uncertain. ADAM33, a member of the ADAM (A Disintegrin And Metalloprotease) gene family, was identified as an asthma susceptibility gene through genetic linkage analysis and association studies of families with asthma in an outbred white population (3). Several ADAM33 single-nucleotide polymorphisms (SNPs) and their haplotypes have been shown to be strongly associated with asthma and BHR (3, 4). In support of a key role for ADAM33 in regulating BHR is the finding that the quantitative trait locus for airway hyperresponsiveness (bhr1) maps to a region on mouse chromosome 2 (74cM) that is very close to the mouse ortholog of ADAM33 (73.9cM) (5). Although the function of ADAM33 is mostly unknown, its selective expression in mesenchymal cells (3, 6) suggests that its function is linked to airway remodeling, an important component of asthma pathophysiology (7–9). The replication of the association between ADAM33 and asthma in ethnically diverse populations (10–13) and the finding that SNPs in ADAM33 are associated with more rapid decline in FEV1 over 20 years in a Dutch population (14) support its importance in the development and progression of asthma. Recent studies have shown that SNPs in ADAM33 predict low lung function in children aged 3 and 5 years (15, 16), suggesting that the influence or influences of ADAM33 commence early in life. ADAM proteins have a complex structure comprising the pro-, catalytic (metalloprotease), disintegrin, cysteine-rich, epidermal growth factor–like, transmembrane, and cytoplasmic domains. These domains may be involved in growth factor shedding, cell migration, and membrane fusion (17). A number of ADAM proteins, including those most closely related to ADAM33, exist as alternatively spliced transcripts. These have been shown to produce protein products with distinct functions, either by the deletion of functional domains, or by altering the cellular localization through the production of secreted and intracellular isoforms (18–24). Several alternatively spliced variants of ADAM33 have been detected in bronchial fibroblasts (25), but there is no information as to whether these differ quantitatively or qualitatively in tissue from normal subjects and subjects with asthma. Therefore, we quantified ADAM33 splice variants in bronchial biopsies obtained from normal volunteers and volunteers with asthma and human embryonic lungs and performed immunolocalization to determine the cellular origin of ADAM33 protein. Some of this work has been presented in abstract form (26).

This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org

METHODS

Am J Respir Crit Care Med Vol 171. pp 958–965, 2005 Originally Published in Press as DOI: 10.1164/rccm.200409-1251OC on February 11, 2005 Internet address: www.atsjournals.org

Additional details of the following methods are provided in the online supplement.

Rationale: Polymorphic variation in ADAM33 (A Disintegrin And Metalloprotease) is strongly associated with asthma and bronchial hyperresponsiveness in different populations. Objective and Methods: To study the role of ADAM33 in asthma, we investigated its expression in normal, asthmatic, and embryonic airways using reverse transcriptase–quantitative polymerase chain reaction and immunochemistry. Results: Several ADAM33 mRNA splice variants were detected in bronchial biopsies and embryonic lung; however, the ␤-isoform and variants encoding the metalloprotease domain were rare transcripts. Western blotting of bronchial biopsies confirmed the presence of multiple isoforms of ADAM33, which had molecular weights of 22, 37, 55, and 65 kD. Immunohistochemistry and laser confocal microscopy of adult bronchial biopsies showed that ␣–smooth muscle actin and ADAM33 immunoreactivity were mostly colocalized to smooth muscle and isolated cells in the submucosa. There was no significant difference in ADAM33 mRNA amplicons or protein in subjects with asthma compared with control subjects. In developing lung, ADAM33 was found around bronchial tubes; however, immunoreactivity was more widely distributed than ␣–smooth muscle actin within undifferentiated mesenchyme; on Western blots, an additional 25-kD ADAM33 variant was detected. Conclusions: Several ADAM33 protein isoforms occur in adult bronchial smooth muscle and in human embryonic bronchi and surrounding mesenchyme, strongly suggesting its importance in smooth muscle development and/or function, which could explain its genetic association with bronchial hyperresponsiveness. The occurrence of ADAM33 in embryonic mesenchymal cells suggests that it may be involved in airway wall “modeling” that contributes to the early life origins of asthma. Keywords: bronchial hyperresponsiveness; epithelial mesenchymal trophic unit; lung development; mesenchymal cells; remodeling

The prevalence of asthma in children and adults is increasing, with more than 100 million people affected worldwide (1). It is characterized by variable airflow obstruction and bronchial hyperresponsiveness (BHR) caused by airway inflammation and remodeling. In chronic, severe asthma, airway inflammation and structural changes both become more intense and are paralleled

(Received in original form September 21, 2004; accepted in final form January 26, 2005) Supported by the Asthma, Allergy, and Inflammation Research Charity (UK), HOPE Wessex Medical Research (UK), the British Lung Foundation (UK), and the Medical Research Council (UK).

Haitchi, Powell, Shaw, et al.: ADAM33 Localization in Human Lungs

Subjects Collection and use of bronchial biopsies, induced sputum, and human embryonic lung were performed after ethical approval from the Southampton and South West Hampshire Joint Local Research Ethics Committee or Newcastle Health Authority and informed consent from the donor. Bronchial biopsies were obtained from 21 normal subjects (8:13, men to women; mean age, 36 years [range, 21–54 years]; FEV1 [%predicted] ⫽ 107 ⫾ 10%) and 19 subjects with asthma (12:7, men to women; mean age, 34 years [range, 21–65 years]; FEV1 [%predicted] ⫽ 90 ⫾ 10%). All subjects with asthma were using ␤2-agonists as required, and 14 were using additional inhaled corticosteroids (median dose, 670 [range, 200–2000] ␮g/day). Human embryonic tissue was collected, staged, and processed as described previously (27). Gestational age was between 8 and 10 weeks.

Reverse Transcription Quantitative Polymerase Chain Reaction Total RNA was extracted from tissue and cell lysates using Trizol reagent, which was DNase-treated (Ambion, Huntingdon, UK) and then reverse-transcribed using random hexamers and Moloney murine leukemia virus (MMLV) reverse transcriptase (Promega, Southampton, UK). The 5⬘ nuclease assay (TaqMan) primers and probes for reverse transcriptase–quantitative polymerase chain reaction were designed to target different regions of ADAM33, as previously described (Figure 1A) (25). Expression levels in bronchial biopsies for each amplicon were calculated using the ⌬⌬CT and expressed relative to the 3⬘-untranslated region (3⬘UTR) amplicon, which was assigned a value of 1. Samples were measured in duplicate.

Western Blotting Airway smooth muscle cells (Clonetics, Wokingham, UK), bronchial biopsies, bronchial brushings, and embryonic lung tissue were solubi-

959

lized, separated by sodium dodecyl sulfate gel electrophoresis and Western blotted using an affinity purified rabbit antibody raised against the cytoplasmic domain of ADAM33 (RP3; Triple Point Biologics, Inc., OR), as previously described (25). The specificity of the antibody for ADAM33 was confirmed using lysates of COS7 cells transfected with full-length ADAM33 or control, mock-transfected cells, using antibody alone or preadsorbed with immunizing peptide.

Immunohistochemistry of ADAM33 in Lung Tissue Bronchial biopsies and human embryonic lungs were acetone-fixed and embedded in glycol methacrylate resin. Consecutive 2-␮m sections were cut and immunostained for ADAM33 (RP3 antibody) and ␣–smooth muscle actin (␣-SMA) using standard protocols (28). To confirm specificity, control staining was performed using the ADAM33 antibody after preadsorption with a fivefold molar excess of the immunizing peptide. Image analysis was performed using an Open Zeiss KS 400 image analysis system (Carl Zeiss, Welyn Garden City, UK).

Colocalization Studies of ADAM33 and ␣-SMA Using Confocal Microscopy Pieces of bronchial biopsies or embryonic lung were fixed in 4% paraformaldehyde and then cleared in dimethyl sulfoxide. Immunostaining was performed using ADAM33 antibody (RP3) directly labeled with Zenon Alexa Fluor 546 (Molecular Probes/Invitrogen, Paisley, UK) and fluorescein isothiocyanate–conjugated mouse monoclonal anti ␣-SMA antibody (Sigma-Aldrich, Gillingham, UK). Nuclei were counterstained with TO-PRO-3 iodide (Molecular Probes). Control pieces of lung tissue were immunostained with anti-ADAM33 antibody preadsorbed with a 20-fold molar excess of the immunizing peptide. Microscopy was performed using a TCS SP2 confocal microscope (Leica, Mannheim, Germany).

Figure 1. (A ) The exon structure and domain organization of ADAM33 and location of the TaqMan amplicons. ADAM33 mRNA splice variant expression relative to 18S ribosomal RNA (B ) and ␣–smooth muscle actin (␣-SMA) (C ), respectively, in bronchial biopsies from normal subjects (n ⫽ 11–13) and subjects with asthma (n ⫽ 8–14). Note that expression data are plotted on a log scale. Data were analyzed by the MannWhitney U test. Dis ⫽ disintegrin amplicon; EGF ⫽ epidermal growth factor like domain; MP ⫽ metalloprotease domain; Sol ⫽ soluble amplicon; SS ⫽ signal sequence; TM ⫽ transmembrane domain; 3⬘UTR ⫽ 3⬘-untranslated region.

960

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 171 2005

Figure 2. ADAM33 splice variant expression relative to ␣-SMA and in human embryonic lung tissue (n ⫽ 3).

Statistics Because data were not normally distributed, significances of differences between groups were analyzed by the nonparametric Mann-Whitney U test (p ⭐ 0.05 was significant).

RESULTS Analysis of ADAM33 mRNA Amplicons in Bronchial Biopsies and Embryonic Lung Tissue

Using quantitative polymerase chain reaction amplicons that recognize alternatively spliced variants of ADAM33 (Figure 1A), both the ␣- and ␤-isoforms of ADAM33 were detected in bronchial biopsy tissue, although the ␤-isoform that lacks exon Q was less abundant compared with that previously found in primary airway fibroblasts (25). The metalloprotease (MP) domain (exons FGHI) and the putative soluble form of ADAM33 (generated by a 37-bp deletion in exon R) were both rare transcripts relative to the 3⬘UTR (Figure 1B), similar to findings in airway fibroblasts (25). Analysis of ADAM33 mRNA in biopsies from 13 normal subjects and 14 subjects with asthma did not reveal any significant difference in the level of expression of any splice variant between these two groups. Furthermore, within the asthmatic group, treatment with steroids did not have any significant effect on the pattern of ADAM33 expression. Because ADAM33 is expressed in mesenchymal cells (3), we also measured mRNA expression of ␣-SMA, a marker of smooth muscle cells and myofibroblasts. We found no significant difference in ␣-SMA expression between normal subjects and subjects with asthma, and analysis of ADAM33 splice variants relative to ␣-SMA did not reveal any significant disease-related differences when comparing subjects with asthma with normal control subjects (Figure 1C). A similar pattern of splice variant expression was also found in embryonic lungs (n ⫽ 3), with relatively few transcripts encoding the MP domain (FGHI; Figure 2). No expression of ADAM33 could be detected in either bronchial epithelial brushings or cells harvested from induced sputum of subjects with asthma or normal subjects (n ⫽ 3; data not shown), even though very strong signals were present for the housekeeping genes, 18S rRNA and ␤-actin. The sputum cell samples contained, on average, 65.3% macrophages, 26.4% neutrophils, 1.9% eosinophils, and 6.4% epithelial cells. Western Blot Analysis of ADAM33 Expression in Bronchial Biopsies, Smooth Muscle Cells, Bronchial Brushings, and Human Embryonic Lung

Western blot analysis was undertaken to characterize ADAM33 protein expression in bronchial biopsies and human embryonic

Figure 3. Western blots of lysates from COS7 cells transfected with fulllength ADAM33 (Cos-FL), airway smooth muscle cells (SMC), bronchial biopsies (BBx), bronchial brushings (BBr), and human embryonic lungs (HEL) were stained with ADAM33 antibody or ADAM33 antibody preadsorbed with the immunizing peptide (P3). Black arrowheads indicate specific ADAM33 bands. The white arrowheads show the nonspecific band with a molecular weight of 48 kD in BBx, BBr, and HEL that could not be blocked with the immunizing peptide and which was putatively identified as cytokeratin.

lungs and to enable comparison with splice variants in airway smooth muscle cells. Using an antibody to the cytoplasmic domain that recognizes all predicted variants of ADAM33, with the exception of the putative secreted isoform, several specific bands with molecular weights of 22, 37, 55, and 65 kD were detected in lysates of airway smooth muscle cells, bronchial biopsies, and human embryonic lungs, but not bronchial epithelial brushings; a strong band at 25 kD was also evident in embryonic lung samples. Although the immunoreactivity of these bands was blocked in the presence of the immunizing peptide, a strong band at around 50 kD whose reactivity could not be blocked was seen in bronchial biopsies, embryonic lung, and bronchial brushings (Figure 3). On the basis of a Western blot using a pan cytokeratin antibody (not shown) and on its molecular weight and epithelial origin (i.e., presence in bronchial brushings), we suggest that this is a nonspecific interaction with epithelial cytokeratins, which have molecular weights in this range. Taken together with the mRNA data, these findings suggest that the majority of ADAM33 isoforms in bronchial biopsies and airway smooth muscle cells exist as smaller alternatively spliced variants that lack the MP domain, as previously reported in bronchial fibroblasts (25). Based on their recognition by an antibody to the carboxyl tail of ADAM33, the 22- and 25-kD proteins are most likely to contain the carboxyl tail and a short region upstream into epidermal growth factor and cysteine-rich domain. The lack of mRNA expression and a corresponding lack of any specific immunoreactivity in bronchial brushings confirm previous reports that ADAM33 is not expressed in epithelial cells (3). Immunohistochemical Analysis and Localization of ADAM33 in Bronchial Biopsies and Human Embryonic Lung

We next localized ADAM33 protein expression in bronchial biopsies by immunohistochemical staining using the same ADAM33 antibody. Through use of serial thin sections (2 ␮m), it was possible to demonstrate localization of ADAM33 in ␣SMA–positive cells (Figure 4). Immunostaining for ADAM33 (Figure 4B) was predominantly within the ␣-SMA smooth muscle bundles (Figure 4A); however, single cells in the submucosa and around vessels were also positively stained for ADAM33 (Figure 4B). Although immunostaining of the muscle and sub-

Haitchi, Powell, Shaw, et al.: ADAM33 Localization in Human Lungs

961

Figure 4. Immunohistochemical analysis (A–C ) of serial sections taken from a bronchial biopsy showing the pattern of immunostaining for ␣-SMA (A ), ADAM33 (B ), or ADAM33 plus immunizing peptide (C ). Colocalization of ␣-SMA and ADAM33 by laser immunofluorescence confocal microscopy (D–I ) using whole mounts of bronchial biopsy tissue. Tissue samples were immunostained with fluorescein isothiocyanate (FITC) conjugated ␣-SMA (green fluorescence; D and G ) and Alexa Fluor 546–labeled ADAM33 (red fluorescence, E ). Overlay of red and green channels (F ) showed that ␣-SMA and ADAM33 were usually colocalized (yellow) within the same cell; however, some cells stained positively for ADAM33 but not for ␣-SMA (arrowhead). ADAM33 immunostaining was markedly reduced when the antibody was preadsorbed with the immunizing peptide (H and I ). Nuclei are counterstained with TO-PRO-3 iodide (purple-blue).

mucosal cells was blocked in the presence of the immunizing peptide (Figure 4C), the weak staining of the bronchial epithelial cells was not fully blocked by the immunizing peptide and probably represents nonspecific cross-reactivity with epithelial cytokeratins, as demonstrated previously. Analysis of the ADAM33 and ␣-SMA–positive stained area relative to the total area of smooth muscle bundles in serial sections showed no significant difference in expression of ADAM33 and ␣-SMA protein in subjects with asthma compared with control subjects (Figure 5). Similarly, no significant difference (p ⬎ 0.05) was observed between normal subjects and subjects with asthma if ADAM33 or ␣-SMA immunostaining was expressed relative to submucosal area (data not shown). To further explore the distribution of ADAM33 and ␣-SMA

Figure 5. Quantitation of ADAM33 and ␣-SMA (ASMA) immunostaining in normal and asthmatic bronchial biopsies.

in bronchial biopsies, tissue pieces were processed for immunofluorescence confocal microscopy. Using a fluorescein isothiocyanate–conjugated antibody, ␣-SMA–positive myofibroblasts or migratory smooth muscle cells could be easily visualized by the green immunostaining of the actin filaments (Figures 4D and 4G). These cells were also ADAM33-positive (Figure 4E) and showed yellow colocalization when the green ␣-SMA and red ADAM33 channels were overlaid (Figure 4F). In addition to its presence in the ␣-SMA–positive cells, ADAM33 immunostaining was also occasionally detected in cells with no obvious ␣-SMA staining (Figure 4F, arrowhead); these may be fibroblasts or primitive smooth muscle precursors. No immunostaining was detected when the ADAM33 antibody was preadsorbed with the immunizing peptide (Figures 4H and 4I). Immunohistochemical staining of human embryonic lung tissue in the pseudoglandular stage of lung development around Weeks 8 to 10 of gestation also showed specific staining around the embryonic bronchial tree for ␣-SMA (Figure 6A) and ADAM33 (Figure 6B). However, in contrast with the adult lung where ADAM33 and ␣-SMA immunoreactivity showed a very similar distribution, there were many ADAM33-positive cells in the undifferentiated mesenchyme surrounding the bronchial ducts that were not positive for ␣-SMA (Figure 6B). To further explore the relationship between ADAM33 and ␣-SMA distribution, embryonic lung pieces were analyzed by immunofluorescent confocal microscopy. Using the fluorescein isothiocyanate–conjugated ␣-SMA antibody, the three-dimensional structure of the embryonic airways could be visualized easily through the green immunostaining of the actin filaments, and there was no detectable ␣-SMA immunostaining outside of these tubular structures (Figure 6D). By analyzing z-series taken through individual airways, the ␣-SMA immunostaining clearly delineated the boundary between the epithelial cells within the airway and the undifferentiated mesenchyme

962

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 171 2005

Figure 6. Immunohistochemical analysis (A–C ) of serial sections taken from glycol methacrylate–embedded human embryonic lung showing the pattern of immunostaining for ␣-SMA (A ), ADAM33 (B ), or ADAM33 plus immunizing peptide (C ). Colocalization of ␣-SMA and ADAM33 by laser immunofluorescence confocal microscopy (D–L ) using whole mounts of human embryonic lung tissue. Human embryonic lung pieces were immunostained with FITC-conjugated ␣-SMA, Alexa Fluor 546–labeled ADAM33, and the nuclei counterstained with TO-PRO-3 iodide (purple-blue). D shows a three-dimensional image of an embryonic airway constructed from the z-series (5 ␮m apart) shown in E–L (within the white frame). In each panel, red and green channels are overlaid so that yellow pixels indicate areas of colocalization of ␣-SMA and ADAM33. Data are representative of three independent experiments.

around the developing airway (Figures 6E–6L). ADAM33 immunoreactivity, which was detected using a red Alexa Fluor conjugate, was absent from the airway epithelial cells, but showed prominent immunostaining in the surrounding mesenchymal cells (Figures 6E–6L). Analysis of the pattern of red ADAM33 and green ␣-SMA immunostaining confirmed that, whereas ADAM33 did show some yellow colocalization within the overlay image, the majority was present in undifferentiated mesenchymal cells that showed no detectable ␣-SMA expression (Figures 6D–6L). ADAM33 staining was not detected when the ADAM33 antibody was preadsorbed with the immunizing peptide (Figure E1 in the online supplement).

DISCUSSION Asthma is a complex disease affecting children and adults, but the biochemical and molecular mechanisms underlying asthma pathogenesis are currently uncertain. Multiple genomic regions are linked to asthma, and there is now strong evidence that polymorphic variation in ADAM33, located at the short arm of chromosome 20, is a genetic determinant for the development of asthma and BHR (3). This finding has been replicated in several other populations (10–13); however, a study in a Puerto

Rican and Mexican population failed to replicate an association (29), and a recent study found only weak association (30), which is a common occurrence in diseases with complex genetic trails (31, 32). This may be because of insufficient statistical power, genotypic and phenotypic heterogeneity across study groups, as well as environmental interactions (31, 33). We report the first demonstration that ADAM33 occurs in smooth muscle bundles and around embryonic bronchi, strongly suggesting that it might play an important role in smooth muscle development and function. In addition to its presence in ␣-SMA–positive cells, the occurrence of ADAM33 in undifferentiated mesenchymal cells, which are much more abundant in embryonic lung, suggests that it may also have a role in development of bronchial, vascular, and neuronal structures of the airways and in differentiation of mesenchymal cells, as such fibroblasts and smooth muscle cells, in these structures. Although it would be of great interest to follow ADAM33 expression during lung development, it was impossible to obtain data at later stages during lung development in utero or postpartum. However, our studies in murine lung development show that ADAM33 mRNA expression increases during embryonic lung development and remains into adulthood (34). A significant number of significant SNPs have been mapped to introns within ADAM33, suggesting that they may regulate

Haitchi, Powell, Shaw, et al.: ADAM33 Localization in Human Lungs

the splicing of ADAM33 transcripts. We have previously shown that a number of novel ADAM33 transcripts are expressed in human airways fibroblasts and have measured their relative prevalence using specifically designed reverse transcription quantitative polymerase chain reaction assays (25). This study used the same assays to analyze ADAM33 transcripts in bronchial biopsies from normal subjects and subjects with asthma and human embryonic lung tissue and confirmed the findings originally made in primary airway fibroblasts. Although the biopsy donors were derived from the Wessex region in the United Kingdom (i.e., they had similar genetic backgrounds and/or the relevant environmental exposures as the cohort that led to the original identification of ADAM33 as an asthma susceptibility gene), we could not show a significant difference in either the overall amount of ADAM33 mRNA or the relative abundance of individual splice variants when we compared 14 subjects with asthma with 13 control subjects (Figures 1B and 1C). This suggests that simple up- or downregulation of ADAM33 expression levels is unlikely to account for its role in asthma pathogenesis. Furthermore, in contrast with a recent report for another new asthma susceptibility gene, GPRA (35), there is no obvious alternatively spliced variant of ADAM33 associated with asthma. However, interpretation of these findings should be viewed with some caution because of the relatively small number of subjects in this study, which may not be sufficient to take into account the influence of genetic variation on ADAM33 protein expression. Of the patients studied, only 12 subjects (9 normal and 3 with asthma) gave consent for genotyping, where analysis of nine SNPs resulted in identification of 11 haplotypes (some more common than others; data not included). Although this did not allow ADAM33 polymorphisms and their haplotypes to be related to mRNA and protein expression, plans are underway for a prospective study to analyze larger study groups where it will be possible to relate expression levels with genotype. Given that some of the disease-related SNPs in ADAM33 encode amino acid changes, the effect of polymorphic variation on disease pathogenesis may be functional, and may involve effects in the regulatory cytoplasmic domain of the molecule. Alternatively, SNPs in untranslated regulatory regions of ADAM33 may affect long-range chromosomal interactions, as has recently been reported for alleles of the tumor necrosis factor ␣/lymphotoxin ␣ locus (36). Further genetic studies are still required to establish the exact location of the genetic lesion or lesions in ADAM33 that predispose to asthma before genotypic considerations can be integrated into molecular analyses performed on clinical samples, especially those requiring invasive procedures, such as bronchoscopy. The ADAM33 mRNA data were also confirmed at the protein level, with no significant difference between ADAM33 expression levels in bronchial biopsies of normal subjects and subjects with asthma. Furthermore, four specific ADAM33 protein bands were detected in airway smooth muscle cells and bronchial biopsies by Western blot analysis. These bands were all smaller than the predicted full-length molecule, consistent with our finding that the majority of mRNA transcripts lack the MP domain. On the basis of our previous work (25), the protein bands between 50 and 70 kD are consistent with isoforms that lack the MP domain but express the domains downstream of the disintegrin domain in conjuction with a variable N-terminal region. The protein band at approximately 37 kD is the approximate size of a variant expressing half of the cysteine-rich domain and the downstream C-terminal portion of ADAM33. The smallest band at around 22 kD most likely contains the carboxyl tail and a short region upstream of the epidermal growth factor and cysteine-rich domains. Although it is possible that these bands are the result of proteolytic degradation of ADAM33, which may occur during handling of the biopsy samples, it is more likely that the proteins

963

represent ADAM33 isoforms for three reasons: first, the protein bands are consistent with the mRNA data; second, the biopsies were processed immediately after bronchoscopy, thus handling time was minimal; and third, similar bands were found in cultured smooth muscle cells, which were solubilized immediately into sodium dodecyl sulfate–containing buffer with protease inhibitors (i.e., under conditions where proteolysis was kept to an absolute minimum). Although our antibody, which reacts with the cytoplasmic domain of ADAM33, would fail to recognize any secreted form of ADAM33, the possibility that a soluble form of ADAM33 is produced in any substantial amount seems unlikely, because mRNA for the putative secreted variant of ADAM33 was of even lower abundance than the MP domain. The absence of the MP domain suggests that some of the functions of ADAM33 in smooth muscle and primitive mesenchymal cells may be linked to other domains, such as the disintegrin, cysteine-rich, or epidermal growth factor domains, which may play roles in adhesion. It has only recently been shown that the isolated disintegrin domain of ADAM33 supports ␣9␤1 integrin–dependent cell adhesion of leucocytes (37). Because this integrin is also expressed on mesenchymal cells, including fibroblasts and smooth muscle– like cells (38), these data together suggest a role for ADAM33 in cell–cell adhesion. In addition, the cysteine-rich and epidermal growth factor domains might also mediate cell spreading by binding to extracellular matrix as well as migration and membrane fusion, as has been demonstrated for other ADAMs (17). All these functions might play an important role in remodeling or functioning of the smooth muscle layer of the epithelial mesenchymal trophic unit (8) in asthmatic airways. We also demonstrated that cells harvested from induced sputum (predominantly inflammatory cells with a small number of epithelial cells) lack ADAM33 expression and confirmed previous observations that epithelial cells lack ADAM33 expression (3), because no mRNA splice variant was detected in bronchial brushings. Although we observed weak immunoreactivity in bronchial epithelial cells using immunohistochemistry applied to bronchial biopsies, this was shown to be nonspecific and is probably caused by unintentional immunization of the rabbits with cytokeratin-containing human-skin scales shed by animal husbandry workers during care of the animal. Outcomes in adult asthma are determined primarily in early childhood. It is now clear that genetic and environmental factors can affect BHR and lung function independent of allergic sensitization (39). Furthermore, BHR can be detected as early as 4 weeks of age and its presence is a predictor of asthma at 6 years of age independent of allergic sensitization (40, 41). In a recent study involving 302 3-year-old and 504 5-year-old children born of parents with asthma and allergy (the Manchester Allergy and Asthma Study), we have shown that 10 asthma-related SNPs in ADAM33 predict reduced lung function measured as specific airway resistance, with the strongest association at both ages being to the F ⫹ 1 SNP (15, 16). Biopsy evidence in severe childhood asthma has revealed that remodeling events occur at disease inception associated with activation of the epithelial mesenchymal trophic unit (42–44). Consistent with our postulate that ADAM33 is contributing to the early life origins of asthma is the current observation that ADAM33 mRNA is present in human embryonic lungs. Furthermore, the occurrence of ADAM33 protein in both differentiated and undifferentiated embryonic mesenchymal cells suggests that it may be involved in airway wall “modeling” that may contribute to the early life origins of asthma. Of particular note was the occurrence of a distinct 25-kD isoform of ADAM33 detected by Western blotting of lysates from the embryonic lung tissue. Further studies are now required to dissect the function of each ADAM33 splice variant,

964

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 171 2005

both in developing and adult lung, and to determine how genetic variation has the potential to modify function leading to BHR, a cardinal feature of asthma. Conflict of Interest Statement : H.M.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; R.M.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; T.J.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; P.H.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.J.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; D.I.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.T.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; D.E.D. received £686,371 as a research grant from AstraZeneca, £167,000 as a research grant from Novartis, and $60,000 as a research grant from Aventis, and also consults for Synairgen. Acknowledgment : The authors thank Peter Wark, M.D., Ph.D., Ian Yang, M.D., Ph.D., Suresh K. Babu, M.D., Timothy Howell, M.B. Ch.B., and research nurses from Allergy and Inflammation Research (AIR) for their assistance with bronchoscopy to obtain bronchial biopsies at the Wellcome Trust Clinical Research Facility; Anton Page, Ph.D., and Roger Alston from the Biomedical Imaging Unit for their assistance with image capture and analysis; Jon Ward, B.Sc., Janet Underwood, B.Sc., and Helen Rigden, B.Sc., from the Histochemistry Research Unit for their assistance processing and cutting GMA resin-embedded tissue; and Julie A. Cakebread, B.Sc., and John W. Holloway, Ph.D., from AIR Asthma Genetics Group for genotype analysis. For personal ethical reasons, R.M.P. was not involved in any work with embryonic tissue.

References 1. National Institutes of Health and National Heart, Lung and Blood Institute. Global strategy for asthma management and prevention guidelines. Global Initiative for Asthma; 2002. Report number 02-3659, a1-176. 2. Juniper EF, Kline PA, Vanzieleghem MA, Ramsdale EH, O’Byrne PM, Hargreave FE. Effect of long-term treatment with an inhaled corticosteroid (budesonide) on airway hyperresponsiveness and clinical asthma in nonsteroid-dependent asthmatics. Am Rev Respir Dis 1990;142: 832–836. 3. Van Eerdewegh P, Little RD, Dupuis J, Del Mastro RG, Falls K, Simon J, Torrey D, Pandit S, McKenny J, Braunschweiger K, et al. Association of the ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 2002;418:426–430. 4. Holgate ST, Davies DE, Murphy G, Powell RM, Holloway JW. ADAM 33: just another asthma gene or a breakthrough in understanding the origins of bronchial hyperresponsiveness? Thorax 2003;58:466–469. 5. De Sanctis GT, Merchant M, Beier DR, Dredge RD, Grobholz JK, Martin TR, Lander ES, Drazen JM. Quantitative locus analysis of airway hyperresponsiveness in A/J and C57BL/6J mice. Nat Genet 1995;11:150–154. 6. Umland SP, Garlisi CG, Shah H, Wan Y, Zou J, Devito KE, Huang WM, Gustafson EL, Ralston R. Human ADAM33 mRNA expression profile and post-transcriptional regulation. Am J Respir Cell Mol Biol 2003;29:571–582. 7. Shapiro SD, Owen CA. ADAM-33 surfaces as an asthma gene. N Engl J Med 2002;347:936–938. 8. Davies DE, Wicks J, Powell R, Puddicombe S, Holgate S. Airway remodeling in asthma: new insights. J Allergy Clin Immunol 2003;111:215– 225. 9. Cakebread JA, Haitchi HM, Holloway JW, Powell RM, Keith T, Davies DE, Holgate ST. The role of ADAM33 in the pathogenesis of asthma. Springer Semin Immunopathol 2004;25:361–375. 10. Howard T, Postma D, Jongepier H, Moore W, Koppelman G, Zheng S, Xu J, Bleecker E, Meyers D. Association of a disintegrin and metalloprotease 33 (ADAM33) gene with asthma in ethnically diverse populations. J Allergy Clin Immunol 2003;112:717–722. 11. Sakgami T, Hasegawa T, Yoshizawa H, Suzuki E, Koshino T, Gejyo F. ADAM33 polymorphisms are associated with aspirin intolerant asthma in Japanese population [abstract]. Am J Respir Crit Care Med 2003;167:A750. 12. Werner M, Herbon N, Gohlke H, Altmuller J, Knapp M, Heinrich J, Wjst M. Asthma is associated with single-nucleotide polymorphisms in ADAM33. Clin Exp Allergy 2004;34:26–31. 13. Lee JH, Park HS, Park SW, Jang AS, Uh ST, Rhim T, Park CS, Hong SJ, Holgate ST, Holloway JW, et al. ADAM33 polymorphism: association with bronchial hyper-responsiveness in Korean asthmatics. Clin Exp Allergy 2004;34:860–865.

14. Jongepier H, Boezen HM, Dijkstra A, Howard TD, Vonk JM, Koppelman GH, Zheng SL, Meyers DA, Bleecker ER, Postma DS. Polymorphisms of the ADAM33 gene are associated with accelerated lung function decline in asthma. Clin Exp Allergy 2004;34:757–760. 15. John S, Jury F, Custovic A, Holloway JW, Woodcock A, Cakebread J, Lowe L, Ollier W, Holgate ST, Simpson PJ. ADAM33 polymorphisms predict early-life lung function: a population based cohort study [abstract]. Am J Hum Genet 2003;73:A237. 16. Simpson A, Jury F, Cakebread J, Lowe L, Holgate ST, Woodcock A, Ollier W, Custovic A, Holloway J, John S. Manchester Asthma and Allergy Study: polymorphisms in ADAM33 predict lung function at age 5 years. J Allergy Clin Immunol 2004;113:S340. 17. Seals DF, Courtneidge SA. The ADAMs family of metalloproteases: multidomain proteins with multiple functions. Genes Dev 2003;17: 7–30. 18. Gilpin BJ, Loechel F, Mattei MG, Engvall E, Albrechtsen R, Wewer UM. A novel, secreted form of human ADAM 12a´(Meltrin alpha) provokes myogenesis in vivo. J Biol Chem 1998;273:157–166. 19. Kurisaki T, Wakatsuki S, Sehara-Fujisawa A. Meltrin beta mini, a new ADAM19 isoform lacking metalloprotease and disintegrin domains, induces morphological changes in neuronal cells. FEBS Lett 2002; 532:419–422. 20. Cerretti DP, DuBose RF, Black RA, Nelson N. Isolation of two novel metalloproteinase-disintegrin (ADAM) cDNAs that show testisspecific gene expression. Biochem Biophys Res Commun 1999;263: 810–815. 21. Hotoda N, Koike H, Sasagawa N, Ishiura S. A secreted form of human ADAM9 has an alpha-secretase activity for APP. Biochem Biophys Res Commun 2002;293:800–805. 22. Roberts CM, Tani PH, Bridges LC, Laszik Z, Bowditch RD. MDC-L, a novel metalloprotease disintegrin cysteine-rich protein family member expressed by human lymphocytes. J Biol Chem 1999;274:29251–29259. 23. Sagane K, Yamazaki K, Mizui Y, Tanaka I. Cloning and chromosomal mapping of mouse ADAM11, ADAM22 and ADAM23. Gene 1999; 236:79–86. 24. Yavari R, Adida C, Bray-Ward P, Brines M, Xu T. Human metalloprotease-disintegrin Kuzbanian regulates sympathoadrenal cell fate in development and neoplasia. Hum Mol Genet 1998;7:1161–1167. 25. Powell RM, Wicks J, Holloway JW, Holgate ST, Davies DE. The splicing and fate of ADAM33 transcripts in primary human airways fibroblasts. Am J Respir Cell Mol Biol 2004;31:13–21. 26. Haitchi HM, Powell RM, Shaw TJ, Babu KS, Howarth PH, Holgate ST, Davies DE. Analysis of alternatively spliced variants of ADAM33 in bronchial biopsies from normal and asthmatic subjects [abstract]. Am J Respir Crit Care Med 2004;169:A471. 27. Hanley NA, Hagan DM, Clement-Jones M, Ball SG, Strachan T, SalasCortes L, McElreavey K, Lindsay S, Robson S, Bullen P, et al. SRY, SOX9, and DAX1 expression patterns during human sex determination and gonadal development. Mech Dev 2000;91:403–407. 28. Britten KM, Howarth PH, Roche WR. Immunohistochemistry on resin sections: a comparison of resin embedding techniques for small mucosal biopsies. Biotech Histochem 1993;68:271–280. 29. Lind DL, Choudhry S, Ung N, Ziv E, Avila PC, Salari K, Ha C, Lovins EG, Coyle NE, Nazario S, et al. ADAM33 is not associated with asthma in Puerto Rican or Mexican populations. Am J Respir Crit Care Med 2003;168:1312–1316. 30. Raby BA, Silverman EK, Kwiatkowski DJ, Lange C, Lazarus R, Weiss ST. ADAM33 polymorphisms and phenotype associations in childhood asthma. J Allergy Clin Immunol 2004;113:1071–1078. 31. Ioannidis JP, Ntzani EE, Trikalinos TA, Contopoulos-Ioannidis DG. Replication validity of genetic association studies. Nat Genet 2001;29: 306–309. 32. Raby BA, Weiss ST. ADAM33: where are we now? Am J Respir Cell Mol Biol 2004;31:1–2. 33. Ioannidis JP, Trikalinos TA, Ntzani EE, Contopoulos-Ioannidis DG. Genetic associations in large versus small studies: an empirical assessment. Lancet 2003;361:567–571. 34. Haitchi HM, Powell RM, Wilson DI, Holgate ST, Davies DE. ADAM33 expression in embryonic mouse lung [abstract]. Am J Respir Crit Care Med 2003;167:A377. 35. Laitinen T, Polvi A, Rydman P, Vendelin J, Pulkkinen V, Salmikangas P, Makela S, Rehn M, Pirskanen A, Rautanen A, et al. Characterization of a common susceptibility locus for asthma-related traits. Science 2004;304:300–304. 36. Knight JC, Keating BJ, Rockett KA, Kwiatkowski DP. In vivo characterization of regulatory polymorphisms by allele-specific quantification of RNA polymerase loading. Nat Genet 2003;33:469–475.

Haitchi, Powell, Shaw, et al.: ADAM33 Localization in Human Lungs 37. Bridges LC, Sheppard D, Bowditch RD. ADAM disintegrin-like domain recognition by the lymphocyte integrins alpha4beta1 and alpha4beta7. Biochem J 2005;387:101–108. 38. Wiester LM, Giachelli CM. Expression and function of the integrin alpha9beta1 in bovine aortic valve interstitial cells. J Heart Valve Dis 2003;12:605–616. 39. Lawrence S, Beasley R, Doull I, Begishvili B, Lampe F, Holgate ST, Morton NE. Genetic analysis of atopy and asthma as quantitative traits and ordered polychotomies. Ann Hum Genet 1994;58:359–368. 40. Joseph-Bowen J, de Klerk NH, Firth MJ, Kendall GE, Holt PG, Sly PD. Lung function, bronchial responsiveness, and asthma in a community cohort of 6-year-old children. Am J Respir Crit Care Med 2004;169: 850–854.

965 41. Palmer LJ, Rye PJ, Gibson NA, Burton PR, Landau LI, LeSouef PN. Airway responsiveness in early infancy predicts asthma, lung function, and respiratory symptoms by school age. Am J Respir Crit Care Med 2001;163:37–42. 42. Warner JO, Marguet C, Rao R, Roche WR, Pohunek P. Inflammatory mechanisms in childhood asthma. Clin Exp Allergy 1998;28:71–75. 43. Payne DNR, Rogers AV, Adelroth E, Bandi V, Guntupalli KK, Bush A, Jeffrey PK. Early thickening of the reticular basement membrane in children with difficult asthma. Am J Respir Crit Care Med 2003; 167:78–82. 44. Jenkins HA, Cool C, Szefler SJ, Covar R, Brugman S, Gelfand EW, Spahn JD. Histopathology of severe childhood asthma: a case series. Chest 2003;124:32–41.