Volume 2, Issue 1 p. 50-56
Open Access

Emphysema Mediated by Lung Overexpression of ADAM10

Hiroki Saitoh

Hiroki Saitoh

Department of Genetic Medicine

Search for more papers by this author
Philip L. Leopold

Philip L. Leopold

Department of Genetic Medicine

Search for more papers by this author
Ben-Gary Harvey

Ben-Gary Harvey

Department of Genetic Medicine

Division of Pulmonary and Critical Care Medicine

Search for more papers by this author
Timothy P. O'Connor

Timothy P. O'Connor

Department of Genetic Medicine

Search for more papers by this author
Stefan Worgall

Stefan Worgall

Department of Genetic Medicine

Department of Pediatrics, Weill Cornell Medical College, New York, New York, USA.

Search for more papers by this author
Neil R. Hackett

Neil R. Hackett

Department of Genetic Medicine

Search for more papers by this author
Ronald G. Crystal

Ronald G. Crystal

Department of Genetic Medicine

Division of Pulmonary and Critical Care Medicine

Search for more papers by this author
First published: 18 February 2009
Citations: 8
RG Crystal ([email protected])

Abstract

Cigarette smoking is the major risk factor for emphysema, a disorder of the lung parenchyma characterized by destruction of the alveolar walls. Current concepts of the pathogenesis of emphysema hold that the destruction of the lung parenchyma results, in part, from a local imbalance of proteases and antiproteases. Based on the knowledge that human alveolar macrophages express ADAM 10, a protease capable of destroying basement membrane collagen but not previously implicated in emphysema, we used adenovirus-mediated lung expression of ADAM 10 in a mouse model to assess whether an increased burden of ADAM 10 was capable of inducing emphysema. To assess this, the human ADAM 10 cDNA under control of a constitutive promoter was inserted into an adenovirus gene transfer vector (AdhADAMlO), and the vector (1011 particle units) administered to the respiratory tract of wild type C57BI/6 mice. Lung levels of ADAM 10 mRNAand protein were upregulated following AdhADAMlO administration. After 8 weeks, quantitative morphometry of the lung parenchyma demonstrated that AdhADAMlO administration induced emphysema (mean linear intercept of 60.6 + 1.3 μm compared with 55.6 + 0.8 in mice treated with a control vector, p < 0.003). These results suggest a role of ADAM 10 in the pathogenesis of emphysema, adding to the list of proteases expressed in the lung that are capable of contributing to the development of lung destruction.

Introduction

Emphysema is a form of chronic obstructive pulmonary disease (COPD) characterized by enlargement of air spaces and destruction of the lung parenchyma distal to the terminal bronchioles, without associated inflammation or scarring.1 The major risk factor for emphysema is cigarette smoking of >20 pack/year.2 Current concepts of the pathogenesis of emphysema hold that the destruction of the lung parenchyma results, in part, from a local imbalance of proteases and anti-proteases that permit proteases to destroy the fragile alveolar walls.1–7

The primary sources of the proteases relevant to the pathogenesis of emphysema are alveolar macrophages (AM), the resident inflammatory cells of the lung,8,9 and neutrophils.4 The number of AM are increased in the lungs of smokers, and cigarette smoke is known to activate AM.8,9 Consistent with this concept, several proteases have been shown to be overexpressed by AM recovered from smokers and/or from individuals with emphysema, who were smokers or ex-smokers, including matrix metalloproteases, cathepsins, and plasminogen activator.6,10–26

Based on the increasing evidence that emphysema is a complex disorder, likely involving multiple proteases released by cigarette smoke-activated AM that function in concert to destroy the lung parenchyma,1,2,4,6,8,9,27 using microarray analysis, we identified ADAM10, an additional protease released by human AM.27 ADAM10 is a 64 kDa (catalytically active form) single chain member of the disintegrin and metalloprotease domain gene family, known to function as a type IV collagenase.28 To assess whether ADAM 10 is capable of mediating in the destruction of the lung parenchyma if present in elevated amounts, as would be the case with the increased numbers of AM found in the lung of smokers,8,9 an adenovirus gene transfer vector coding for human ADAM10 (AdhADAMlO) was administered intratracheally to mice, and the lungs were evaluated for the presence of emphysema by quantitative morphometry. Assessment of the lungs 2 months later demonstrated enlargement of the airspaces indicative of emphysema, supporting a possible role of ADAM10 in the pathogenesis of smoking-induced emphysema.

Methods

Adenovirus vectors

The recombinant adenovirus vector AdhADAMlO and AdNull used in the study were Ela-, partial Elb-, and partial E3-, based on the Ad5 genome, with the expression cassette in the El position. The expression cassette includes the cytomegalovirus (CMV) early/intermediate enhancer promoter, an artificial splice signal, the relevant transgene, and an SV40 stop/poly (A) signal. The AdhADAMlO vector contains the human ADAM 10 cDNA obtained by PCR-amplification using RNA extracted from the human THP-1 macrophage cell line (American Type Culture Collection, Manassas, VA, USA) as a template. The fidelity of the human ADAM 10 cDNA in the Ad vector was confirmed by sequencing. The AdNull vector, used as a control, is identical to the AdhADAMlO vector, except that it carries no transgene.29 The vectors were propagated, purified, and stored at -70°C, as previously described.30,31 The vector doses are expressed in particle units (pu) determined spectrophotometrically as described by Mittereder et al.32

TaqMan mRNA analysis

TaqMan real-time reverse transcriptase (RT) PCR analysis was used to quantify the levels of expression of human ADAM 10 mediated by the AdhADAMlO vector in transduced A549 cells. First strand cDNA was synthesized from 2 μg of RNA in a 100 :1 reaction volume, using the TaqMan Reverse Transcriptase Reaction Kit (Applied Biosystems, Foster City, CA, USA), with random hexamers as primers. The cDNA was diluted 1 : 100 or 1:10, and 5 mL was used for a total of 50 mL final reaction volume, using the Universal Master Mix (Applied Biosystems). The probe and primers specific for human ADAM 10 mRNA were designed using the PrimerExpress software (Applied Biosystems). For each individual sample, two conditions were used: 1:10 and 1:100 dilution of the cDNA reaction and each dilution was done in triplicate wells. Relative expression levels were calculated using the DDCt method (Applied Biosystems), using ribosomal RNA (rRNA) as the internal control (Human Ribosomal RNA Kit, Applied Biosystems) and a cocktail consisting of equal parts of mRNA samples from control cells, as the calibrator. The rRNA probe was labeled with VIC™ and the ADAM10 probe with FAM™. The PCR reactions were run in an Applied Biosystems Sequence Detection System 7700. The threshold cycles (Cts) were calculated as an average of the triplicate reactions for each condition, and the ΔCt was calculated as the CtFAM/CtVIC for each sample. Then the DDCt is calculated using the control-cell mRNA cocktail as calibrator, by subtracting the calibrator from the DCt in each individual sample. The relative quantity was calculated using the algorithm provided by Applied Biosystems.

Western analysis

To demonstrate expression of the human ADAM 10 protein mediated by the AdhADAMlO vector, 2 × 106 A549 cells were infected with AdhADAMlO or AdNull (2 × 1010 pu, 90 minutes, 37°C) in serum-free Dulbecco's Modified Eagle Medium (DMEM). The medium was aspirated and replaced with 2 mL of DMEM supplemented with 10% fetal calf serum, 50 U/mL penicillin, 50 μm/mL streptomycin and 0.25 μg/mL fungizone. The cells were cultured for 48 hours, 37°C. Mock-infected (naive) A549 cells were treated exactly as infected cells. After washing three times with phosphate buffered saline (PBS, pH 7.4), the cells were resuspended in 500 mL sample buffer (100 mM Tris-HCl, pH 6.8, 4% sodium dodecyl sulfate (SDS), 20% glycerol, 10% 2-mercaptoethanol) and sonicated. The samples were centrifuged at 12000 g for 10 minutes, supernatant was collected, and total protein amount was measured using the BCA Protein Assay Reagent Kit (Pierce Biotechnology, Inc., Rockford, IL, USA). The total protein sample (10 mg) was mixed with Tris-Glycine-SDS Sample Buffer (Invitrogen) and analyzed by SDS-polyacrylamide gel electrophoresis (PAGE; 8%) and electroblotted onto nitrocellulose membranes using 8% Tris-Glycine-SDS and PVDF membrane (Invitrogen, Carlsbad, CA, USA). The membranes were blocked with 50 mL of blocking solution (10% skim milk, 0.2% Tween 20,3% goat-serum) in PBS at 4°C overnight, washed once with 0.2% Tween 20 in PBS, two times with 0.1% Tween 20 in PBS (BT-PBS), and incubated with human ADAM10 polyclonal antibody (1 : 500, Chemicon International, Temecula, CA, USA) at 23°C, 1 hour. The membranes were then washed three times with BT-PBS and incubated with horseradish peroxidase (HRP)-conjugated secondary antibody (BD Bioscience Pharmingen, San Diego, CA, USA) for 1 hour. After washing with BT-PBS, the membranes were treated with SuperSignal West Pico Chemiluminescent Substrate (Pierce Biotechnology, Rockford, IL, USA) and exposed with X-ray film.

Administration of Ad vectors into mouse lungs

Male C57B1/6 mice, 8–10 weeks old, were obtained from The Jackson Laboratories (Bar Harbor, ME, USA). Animals were housed under specific pathogen-free conditions and treated accordingto National Institutes of Health guidelines. AdhADAMlO or Ad Null (1011 particle units in 50 mL PBS) was administered intratracheally to mice anaesthetized by intraperitoneal inj ection of ketamine (100 mg/kg) and xylazine (10 mg/kg). The trachea was exposed by midline incision and a 22-gauge catheter (Becton Dickinson, Franklin Lakes, NJ, USA) was inserted under direct vision through the tracheal wall into the lumen at the level of the mid-portion of the trachea. The vectors were slowly instilled through the angiocatheter over a period of 1–2 minutes.

Human AdhADAMlO expression in the murine lung

To assess the expression of human ADAM10 mRNA and protein resulting from intratracheal administration of AdhADAMlO into mouse lungs, human ADAM10 expression in whole lung was evaluated by RT-PCR and Western analysis. Following administration of the AdNull or AdhADAMlO vectors (5 × 10” pu), the mice were sacrificed at 1–7 days after vector administration. The sternum was cut and the diaphragm removed to expose the lungs. Blood in the pulmonary circulation was removed using 1 mL PBS by injection into the right ventricle, and the whole lung was removed from the body and transferred to a tube containing 500 mL of PBS. The tissue was homogenized and 50 mL of the resulting solution dissolved in 1 mL TRIzol for total RNA extraction as described above. Total RNA (1 μg) was converted to first strand cDNA, and a 1 : 100 dilution was amplified by 33 cycles of PCR using the following primers: 5′-GCGGAAGATGGTGTTGCTGAGAGT-3′ (forward primer, starting at nucleotide 453 of the human ADAM10 cDNA sequence, GenBank accession number NM_001110) and 5′-AGGGAACACGGGGCACATAATAATA-3′ (reverse primer, hybridizing to the 3′ untranslated region of the human ADAM10 cDNA, ending at nucleotide 3004 of GenBankNM_001110 (http://www.ncbi.nlm.nih.gov/)). The amplified material was electrophoresed in a 1.5% agarose gel. For Western analysis, 20 mL of homogenized lung solution was assessed as described above.

Lung histology and quantitative morphometry

To assess the effects of human ADAM10 overexpression in mouse lungs, AdNull or AdhADAMlO vectors (1011 pu) were administered intratracheally, and the mice were sacrificed at 8 weeks after vector administration. The diaphragms were removed and the sternum cut to expose the lungs. A 22-gauge catheter was inserted into the trachea and the lungs were inflated and fixed with 4% paraformaldehyde at 25 cm H20 transpulmonary pressure for 4 hours. After pressure-dependent fixation, the trachea was ligated, the lungs removed and immersed in 4% paraformaldehyde for 4 hours, 4°C, followed by immersion in 70% ethanol overnight at 4°C. The lungs were embedded in paraffin and 4 μηι sections were prepared and stained with hematoxylin and eosin. Inflammation was graded on a scale from 0–4 with: 0, no inflammation; 1, minimal inflammatory infiltrates affecting less than 25% of bronchiolar and alveolar areas per visual field (magnification 40x); 2, moderate inflammation affecting 25–49%; 3, pronounced inflammation affecting 50–75%, and 4, severe inflammation affecting >75%. Twenty fields were scored (10 left and 10 right) for each experimental animal and then averaged. Alveolar images were captured by Nikon microscope and digital camera operating on Adobe Photoshop.

The integrity of alveolar structure was determined by calculating the mean linear intercept using the method of Dunnill et al.33 adapted for digital image acquisition and analysis. Samples were masked, and 200 images in each experimental group were acquired for analysis and labeled according to a randomized, blinded code. Forty fields were acquired at random per lung. For each experimental condition, five mice were analyzed. Paraffin-embedded sagittal “butterfly” sections were deparaffinized in xylene followed by graded series of ethanol solutions in decreasing concentration of ethanol from 100% to 50%. Sections were stained with hematoxylin and eosin and viewed using a 20x objective lens. When the images were viewed using illumination from a 100 watt mercury arc lamp with a red fluorescence filter set (580 nm excitation, 600 nm dichroic mirror, 630 nm emission filter), the eosin stain emitted a fluorescence signal that was recorded by a Photometrix Quantix cooled charge coupled device (CCD) camera (Roper Scientific, Trenton, NJ, USA). Image analysis was performed using MetaMorph image analysis software (Universal Imaging, Downingtown, PA, USA). All fluorescence images were acquired using the same exposure conditions, and a uniform threshold was applied to the image to mark all image regions with a fluorescence intensity less than or equal to 5% of the dynamic range of the camera corresponding to the dark background of the empty alveoli between the bright alveolar walls. The thresholded fluorescence image was converted to a binary image in which all of the thresholded areas (empty alveoli) received a pixel value of 1 (white) whereas non-thresholded areas (alveolar walls) received a value of 0 (black). To determine the number of alveolar wall intercepts along an arbitrary line of a defined length (corresponding to 250 μm in the micrographic image), the binary image was passed through two masks containing either three horizontal or vertical line segments (with a combined length of 1500 μm per field). The masked image contained white line segments corresponding to the alveolar spaces that intersected the line segments in the mask. Using the image analysis software, the total number of alveolar spaces per field was determined by automated counting. For a single field, the mean linear intercept was calculated as follows: mean linear intercept = 1500 μl /[(No. of horizontal intercepts – 3) + (No. of vertical intercepts – 3)], where 1500 μιτι equals the total length of the six line segments analyzed per field and the reduction of the number of vertical and horizontal linear intercepts by 3 was a correction that was introduced due to the fact that alveolar spaces rather than alveolar walls were counted. Due to the likelihood of a line segment both starting and ending in an alveolar space, the number of alveolar spaces encountered would usually overestimate the number of wall intersections by exactly one per line segment. The mean linear intercept for each lung was calculated as the arithmetic mean of the mean linear intercepts for each field.

Statistical analysis

All statistical analysis except microarray analysis was carried out using Microsoft Excel or StatView. Comparison of the inflammation grades and mean linear intercept (Lm) values in mice treated with AdhADAMlO or AdNull was done by the two-tailed Student's t-test. Unless otherwise noted, all data are expressed as mean ± standard error.

Results

Ad-mediated expression of human ADAM10 in vitro

The ability of AdhADAM10 to direct the expression of ADAM10 mRNA and protein in A549 cells was evaluated by TaqMan realtime RT-PCR and Western analysis, respectively (Figure 1). After 24 hours, ADAM10 mRNA levels were increased 14.5-fold in A549 cells transduced with AdhADAMlO compared with ADAM10 endogenous expression levels in naive cells or in cells transduced with the AdNull vector carrying no transgene (Figure 1A). Expression of ADAM10 protein in AdhADAMlO-transduced A549 cells was evaluated by Western analysis using a polyclonal antibody against a human ADAM 10 cytoplasmic domain epitope (amino acids 732–748). Two specific bands, the precursor of ADAM10 (approximately 98 kDa) and the mature active shed form (approximately 64 kDa), were observed in naive and AdNull-infected and AdhADAMlO-infected A549 cells (Figure 1B). The observed molecular weight of the precursor and active forms of human ADAM10 are in agreement with previously published observations of endogenous ADAM10 in the human monocytic cell line THP-1.34 Increased levels of both the approximately 98 and 64 kD forms of ADAM10 were observed in AdhADAMlO-transduced cells 48 hours following infection (Figure 1B).

Details are in the caption following the image

Expression of human ADAM10 mRNA and protein mediated by Ad-gene transfer in the A549 cells. (A) TaqMan RT-PCR relative quantification of Ad-mediated ADAM10 gene expression. Total RNA was extracted from A549 cells 24 hours after infection with 2×1010 pu of AdhADAMlO, or AdNull, and subjected to quantification by TaqMan RT-PCR. The data are shown relative to the endogenous expression of ADAM10 in naive A549 cells. (B) Expression of ADAM10 protein in A549 cells 48 hours postinfection with 2 × 1010 pu of AdhADAMlO or AdNull (control). Cell lysates of naive A549 cells were used as an additional control. ADAM10 was detected by Western analysis of the cell lysates using an anti-ADAM10 polyclonal antibody. Two forms of ADAM10 are observed in naive, AdNull infected, or AdhADAMlO infected A549 cells (approximately 98 and 64 kDa), representing the precursor and mature form, respectively.

Ad-mediated expression of human ADAM10 In Vivo

To determine the potential role of ADAM10 overexpression in the pathogenesis of emphysema, the adenovirus vector AdhADAMlO was used to mediate gene transfer of this candidate emphysema-mediating protease gene into the lungs of experimental animals, and emphysematous changes were quantified. The expression of human ADAM10 mRNA and protein directed by the AdhADAMlO vector was evaluated in C57B1/6 mouse lungs in vivo following intratracheal administration of 5 × 1011 pu of AdhADAMlO, 5 × 1011 pu of AdNull, or naive. After 1, 3, 5, or 7 days, animals were sacrificed and the whole lungs assessed for human ADAM 10 mRNA and protein by RT-PCR and by Western analysis, respectively (Figure 2). A full-length, 2541 bp human ADAM 10 cDNA was observed at every time point postinfection, with the highest levels at days 1–3 (Figure 2A, lanes 4,5). Western analysis showed that the active approximately 64 kDa form of mouse ADAM10 was predominantly observed in naive and AdNull-infected mouse lungs, indicating that the polyclonal antibody against human ADAM 10 cross-reacts with the murine protein (Figure 2B, lanes 1,2). Both precursor (approximately 98 kDa) and increased amounts of active (approximately 64 kDa) forms of human ADAM 10 were evident at high levels at 5–7 days following AdhADAMlO administration (Figure 2B, lanes 5, 6).

Details are in the caption following the image

Time course of expression of human ADAM10 mRNA and protein in whole mouse lung following intratracheal administration of AdhADAMlO vector or the AdNull control vector (5 × 1011 pu each). Total RNA was extracted from whole mouse lungs to serve as template for RT-PCR using human ADAM10-specific primers, at days 1, 3, 5, and 7 postinfection. Whole cell lysates were also prepared at days 1, 3, 5, and 7 postinfection for Western analysis using an anti-ADAMlO polyclonal antibody. (A) Time course of human ADAM10 mRNA expression monitored by RT-PCR. Naive and Ad Null-infected were used as negative controls. Lane 1 = naive; lane 2 = AdNull, day 1; lane 3 = AdNull, day 3; lanes 4–7 = AdhADAMlO, days 1, 3, 5, and 7, respectively. Expression of human ADAM10 mRNA (2541 bp band) was detected at days 1 through 7 postinfection, peaking at days 1 and 3. Murine b-actin primers were used as a control for template amount. (B) Time course of ADAM10 protein expression detected by Western analysis. Lane 1 = naive; lane 2 = AdNull; lanes 3–6 = ADhADAMlO, days 1, 3, 5, and 7 respectively. Since the antibody used is not specific for human ADAM10 but cross-reacts with the murine protein, an approximately 64 kDa band corresponding to endogenous mouse ADAM10 can be seen in Ad Null-transduced cells. At days 5 and 7 postinfection, a more intense approximately 64 kDa band can be seen in the AdhADAMlO-transduced cells; the approximately 98 kDa band observed in the AdhADAMlO-infected A549, as shown, can also be observed at those same time points, indicating the presence of the precursor protein.

Emphysema induced by AdhADAMlO

To assess the effects of Ad-mediated overexpression of human ADAM 10 in the mouse lung, C57B1/6 mice were infected by intratracheal administration of either 1011 pu of AdhADAMlO, 1011 pu of AdNull, or PBS (naive). Eight weeks after vector administration, lung histology was evaluated by light microscopy for inflammation and emphysematous changes (Figure 3). Lung inflammation grades, rated on the 0–4 point scale (and presented as mean ± S.D.), for the naive, AdNull, and AdhADAMlO treated mice were 0.09 ± 0.09, 0.60 ±0.12, 0.67 ±0.11, respectively (p > 0.3 for AdNull vs. AdhADAMlO). Compared with naive (panel A) and AdNull-treated mice (panel B), emphysematous changes with enlargement of alveoli were observed in AdhADAMlO-treated lungs (panels C and D). To quantify the observed changes, the integrity of alveolar structure was determined by calculating the Lm adapted for digital image acquisition and analysis quantification of alveolar size. Samples were masked and images were acquired, quantified, and analyzed in a blind fashion, using a randomized code for the lung histology preparations. A significant increase in Lm was observed in the lung histology preparations of mice receiving the AdhADAMlO vector in comparison with the AdNull-treated mice (Figure 4; p < 0.003). These data demonstrate the presence of measurable emphysematous changes in mouse lungs overexpressing human ADAM 10.

Details are in the caption following the image

Lung histology of mice 8 weeks following intratracheal administration of AdhADAM! 0 or AdNull. PBS (50 μl), AdNull or AdhADAM10 (1011 pu, each in 50 μl) were delivered by intratracheal administration to 8 week old C57BI/6 mice. At week 8 posttreatment, the mice were sacrificed and the lungs inflated with 4% buffered paraformaldehyde to 25 cm water pressure, for 4 hours. The lungs were then removed and immersed in 4% paraformaldehyde, replaced by 70% ethanol, and stored overnight at 4°C. The fixed lungs were embedded in paraffin and 4μm-thick sections were prepared and stained with hematoxoxylin and eosin. Shown are representative histological lung sections. (A) Naive (PBS); (B) AdNull-treated; (C) AdhADAM10-treated animals (low power); and (D) AdhADAM 10 (high power). (A)-(C), bar = 500 μηη; (D), bar = 200 μηη.

Details are in the caption following the image

Quantitative morphometric analysis of emphysema in the lungs of mice receiving intratracheal AdhADAM 10. Shown is the quantitative morphometric (Lm = mean linear intercept) analysis of alveolar septae of the lungs of five Ad Null-treated and five AdhADAM 10-treated mice. The lungs were prepared as described in Figure 3.

Discussion

Cigarette smoking is the major risk factor for emphysema, a disorder of the lung parenchyma characterized by destruction of the alveolar walls.1,2,4,7,35 In the context that AM are a major source of proteases with the potential to destroy protein components of alveolar walls, a number of studies have focused on assessing AM from humans and experimental animals as a source of proteases that might play a role in the pathogenesis of emphysema. In general, the investigative strategy has been to assess AM for the expression of known proteases with properties that could contribute to the pathogenesis of emphysema or to use knockout or transgenic mice to evaluate if under- or overexpression of proteases and their cognate antiproteases are associated with the development of emphysema. Using these collective strategies, a number of AM-derived proteases have been linked to the pathogenesis of emphysema, including MMP1, MMP2, MMP9, MMP12, MMP14, cathepsins L and S, and plasminogen activator.6,8,10–26

Based on the hypothesis that there may be additional proteases released by AM in cigarette smokers that have not been identified, using microarray technology we observed expression of ADAM 10 in human AM.27 The potential for ADAM 10 to induce emphysema was assessed by creating AdhADAMlO, an adenovirus gene transfer vector expressing the human ADAM 10 cDNA, and using AdhADAMlO to over express ADAM10 in the lungs of mice. The data demonstrate that ADAM10 overexpression results in significant alveolar destruction, characteristic of emphysematous changes. As with other potential mediators of emphysema, this effect could be direct, via destruction of lung tissue by the protease, or indirect, by causing inflammation and/or apoptosis. Taken together, the data support the concept that ADAM10 is capable of playing a role in the pathogenesis of emphysema.

ADAM10

ADAM10 is a membrane-anchored protein that is a member of a distintegrin and metalloprotease (ADAM) gene family. ADAM10 is known to be expressed in human monocytes.36 The cell surface form of AD AM 10 on 293 cells has been shown to be catalytically active.37 The catalytic activity of ADAM10 has been investigated in vitro using purified protein.28,34,37 ADAM10 has a collagenase activity that degrades basement membrane type IV collagen, the most abundant of the non-fibril-forming collagen molecules within lung tissue that is present in both alveolar and capillary basement membranes, in the endothelial basement membranes of pulmonary vessels, and epithelial basement membranes of the tracheobronchial tree.25 Another matrix metalloprotease, MMP-9, also has collagenase activity against type IV collagen and mediates the degradation of rat amnion basement membrane type IV collagen.38

Animal models of emphysema

The classic model of emphysema was the observation by Gross et al.39 that intratracheal administration of papain led to alveolar destruction. Similar models of instillation of proteases into the lung led to the acceptance of the concept of an excess of neutrophil elastase as a mediator of emphysema.40–51 The tight skin mouse, resulting from mutations in the fibrillin-1 gene, is a natural model of emphysema.52,53

Knockout and transgenic mice have been used to investigate the role of the matrix metalloproteases in emphysema. MMP-12 knockout mice (a protease with elastin degrading activity) do not show any emphysematous change after exposure to cigarette smoke, while wild type mice do.44 Transgenic mice overexpressing MMP-1, a secreted enzyme that breaks down the interstitial collagens, types I, II, and III54 and knockout mice lacking TIMP-3, a member of a family of natural inhibitors of the MMPs, developed spontaneous emphysema.55

In the present study, we used Ad-mediated overexpression of a candidate gene to investigate the role of a potential candidate gene in the pathogenesis of emphysema. It is known that expression of Ad-encoded transgenes has a duration of 2–4 weeks after Ad vector administration,56 suggesting that 2–4 weeks of ADAM10 overexpression in the lung are sufficient to result in emphysematous changes, albeit of a smaller magnitude than those reported for transgenic mouse models.9 Compared with transgenic or knockout mouse models, gene delivery to the lung by using an Ad vector has the potential advantage that results can be obtained in a shorter amount of time and are less labor intensive. In addition, interpretation of the emphysema of transgenic and knockout mouse models is complicated by the possibility of the relevant gene influencing the embryonic development if the targeted gene affects development. For example, ADAM10 knockout mice die in an early embryonic stage (day 9.5 of embryogenesis) with multiple defects of the central nervous system, somites, and cardiovascular system.57

In the present study we have not investigated the cell types infected with the AdhADAMlO vector by intratracheal administration. Transduction of AM by Ad vectors, both ex vivo and in vivo, has been demonstrated by our lab as well as by others.26,58 We have previously demonstrated that Ad vectors can effectively transfer and express genes in murine AM recovered by bronchoalveolar lavage fluid,58 and that the transduced AM continue to express the transgene once returned to the respiratory tract. Intratracheal administration of Ad vectors to mice showed that AM were rapidly transduced by the viral vector. Ad vectors administered intratr ache ally to mice transfer genes not only to the respiratory epithelium but also to AM, by immunohistochemistry59 Based on these observations, it is likely but not proven that intratracheal administration of the AdhADAMlO vector likely transduced both airway epithelial cells and AM.

Proteases, antiproteases, and emphysema

A prominent hypothesis regarding the mechanisms of smoking-induced emphysema is that an imbalance of proteases and antiproteases in the lungs of smokers in favor of the proteases leads to the development of emphysema.1,2,47,22,45,46 The definitive example of this concept is the autosomal recessive inherited deficiency of αl-antitrypsin (αlAT), an inhibitor of serine proteases, that accounts for 1–2% cases of emphysema.51 In the more common form of emphysema not associated with al AT deficiency, the major focus has been on AM as a source of excess proteases in the lung. In this regard, elevated expression of several metalloproteases, including MMP-1 (also known as collagenase), MMP-2 (gelatinase A), and MMP-9 (gelatinase B), has been observed in AM of individuals with emphysema.6,8,15,16,18–21,25 In addition, elevated expression of MMP-12 (macrophage elastase), an enzyme that degrades soluble and insoluble elastin, and MMP-9, an enzyme that degrades type IV and V collagens, has been shown in the AM of disease-free smokers versus normal non-smokers.4,14,17 The present study suggests that ADAM10 should be added to the list of AM proteases capable of inducing emphysema.

Conclusions

Emphysema is a complex disorder likely involving multiple proteases released by AM, which function in concert to destroy the lung parenchyma in cigarette smokers. ADAM10 is known to be released from human AM but has not previously been implicated in the pathogenesis of emphysema.28 To assess whether ADAM10 is capable of mediating in the destruction of the lung parenchyma if present in elevated amounts, as would be expected with the increased numbers of AM found in the lung of smokers, an adenovirus gene transfer vector coding for human ADAM10 (AdhADAMlO) was administered intratracheally to mice, and the lungs were evaluated for the presence of emphysema. Quantitative morphometry demonstrated enlargement of the airspaces indicative of emphysema at 2 months following administration of AdhADAMlO, supporting a possible role of ADAM10 in the pathogenesis of smoking-induced emphysema.

Acknowledgments

We thank Adam Cieciuch and Karsta Luettich for expert technical assistance; R. Kaner, A. Heguy, and H. Kobayashi for helpful discussions; and N. Mohamed for help in preparing this manuscript. These studies were supported, in part, by P50 HL084936.