- Email: [email protected]
Matrix metalloproteinase-12 is a therapeutic target for asthma in children and young adults
Matrix metalloproteinase-12 is a therapeutic target for asthma in children and young adults Somnath Mukhopadhyay, FRCPCH, MD, PhD,a* Joseph Sypek, PhD,b* Roger Tavendale, PhD,c Ulrike Gartner, PhD,c John Winter, FRCP,c Wei Li, PhD,d Karen Page, PhD,b Margaret Fleming, PhD,b Jeff Brady, PhD,e Margot O’Toole, PhD,f Donald F. Macgregor, FRCPCH,g Samuel Goldman, PhD,b Steve Tam, PhD,d William Abraham, MD,h Cara Williams, PhD,b Douglas K. Miller, PhD,i and Colin N. A. Palmer, PhDc Brighton and Dundee, United Kingdom, Cambridge, Mass, Miami Beach, Fla, and Collegeville, Pa Background: Matrix metalloproteinase (MMP)-12–mediated pathologic degradation of the extracellular matrix and the subsequent repair cycles influence the airway changes in patients with asthma and chronic obstructive pulmonary disease (COPD). The common serine variant at codon 357 of the MMP12 gene (rs652438) is associated with clinical manifestations consistent with more aggressive matrix degradation in other tissues. Objective: We sought to explore the hypothesis that MMP12 represents a novel therapeutic target in asthma. Methods: The role of the rs652438 variant on clinical phenotype was explored in young asthmatic patients and patients with COPD. Candidate MMP-12 inhibitors were identified on the basis of potency and selectivity against a panel of other MMPs. The role of MMP-12–specific inhibition was tested in vitro, as well as in animal models of allergic airway inflammation.
From aThe Department of Pediatrics, Royal Alexandra Children’s Hospital, Brighton and Sussex Medical School, Brighton; bPfizer Research, Inflammation and Immunology, Cambridge; cthe Population Pharmacogenetics Group, Biomedical Research Institute, University of Dundee, Ninewells Hospital and Medical School, Dundee; dPfizer Research, Chemical and Screening Sciences, Cambridge; eTMRC Laboratory, University of Dundee; fPfizer Research, Biologic Therapeutics, Global Biotherapeutics Technologies, Cambridge; gDirectorate of Paediatrics, NHS Tayside, Ninewells Hospital, Dundee; hMt Sinai Medical Center, Department of Research, University of Miami, Miami Beach; and iPfizer Research, Discovery Translational Medicine, Collegeville. *These authors contributed equally to this work. Supported by the Translational Medicine Research Collaboration (INF-DU-031) and Wyeth Research. The BREATHE study of asthma in children was funded by Scottish Enterprise Tayside, the Gannochy Trust (Perth), and the Perth and Kinross City Council. Disclosure of potential conflict of interest: S. Mukhopadhyay receives research support from MSD UK and Wyeth plc and has provided legal consultation/expert witness testimony in cases related to cystic fibrosis. J. Winter provides personal injury reports per annum income @ £20k. J. Sypek, W. Li, K. Page, M. Fleming, J. Brady, M. O’Toole, S. Goldman, S. Tam, C. Williams, and D. K. Miller are employed by Pfizer Research. D. F. Macgregor receives research support from MSD and has provided legal consultation/expert witness testimony in cases related to pediatric respiratory illness and child protection issues. W. Abraham receives research support from Wyeth, the National Institute of Environmental Health Sciences, Parion, JSJ Pharmaceuticals, and Novartis. C. N. A. Palmer receives research support from Wyeth Pharma. The rest of the authors have declared that they have no conflict of interest. Received for publication April 13, 2009; revised March 22, 2010; accepted for publication March 25, 2010. Available online May 24, 2010. Reprint requests: Somnath Mukhopadhyay, FRCPCH, MD, PhD, Chair of Paediatrics, Academic Unit level 10C, Royal Alexandra Children’s Hospital, Eastern Road, Brighton, BN2 5BE, United Kingdom. E-mail: [email protected]. Joseph Sypek, PhD, Inflammation and Immunology, Pfizer Research, 200 Cambridge Park Dr, Cambridge, MA 02140. E-mail: [email protected]. 0091-6749/$36.00 Ó 2010 American Academy of Allergy, Asthma & Immunology doi:10.1016/j.jaci.2010.03.027
70
Results: The odds ratio for having greater asthma severity was 2.00 (95% CI, 1.24-3.24; P 5 .004) when comparing asthmatic patients with at least 1 copy of the serine variant with those with none. The carrier frequency for the variant increased in line with asthma treatment step (P 5 .000). The presence of the variant nearly doubled the odds in favor of asthmatic exacerbations (odds ratio, 1.90; 95% CI, 1.19-3.04; P 5 .008) over the previous 6 months. The serine variant was also associated with increased disease severity in patients with COPD (P 5 .016). Prior administration of an MMP-12–specific inhibitor attenuated the early airway response and completely blocked the late airway response with subsequent Ascaris suum challenge in sheep. Conclusion: Studies on human participants with asthma and COPD show that the risk MMP12 gene variant is associated with disease severity. In allergen-sensitized sheep pharmacologic inhibition of MMP12 downregulates both early and late airway responses in response to allergic stimuli. (J Allergy Clin Immunol 2010;126:70-6.) Key words: Asthma, allergy, chronic obstructive pulmonary disease, metalloproteinase, genetics, exacerbation, polymorphism, animal studies
There is now strong evidence across species implicating both acute allergen-responsive proinflammatory and chronic airway remodeling roles for the enzyme matrix metalloproteinase (MMP)-12 in the lung.1-4 Mice deficient in MMP-12 exhibit a marked overall reduction in their airway inflammation profile after allergen-induced lung injury.1 Infiltration with eosinophils and macrophages is critical to the airway responses to allergens and the subsequent development of airway hyperresponsiveness (AHR).4-6 MMP-12 is a critical driver of this process, regulating IL-13–induced allergic inflammation in the airways, thus controlling the accumulation of eosinophils and macrophages in response to allergen exposure.4,6 MMP-12 also plays a key role in influencing airway remodeling by degrading a wide range of extracellular matrix proteins, including elastin, type IV collagen, fibronectin, laminin, and gelatin. This activity occurs either directly or through the activation of other MMPs or downstream from IL-13.2-4 Although MMP-12 is found predominantly in macrophages, human bronchial epithelial cells also express and secrete MMP-12.7 Tricyclic and other compounds have been known to inhibit MMP-12 and other metalloproteinases, possibly through the pharmacologic roles of forming chelate complexes and acting as electron donors.8 Animal models of allergen-induced lung inflammation are well characterized9; this offers the potential for testing the role of MMP-12 inhibition in such models.
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
Abbreviations used AHR: Airway hyperresponsiveness BTS: British Thoracic Society COPD: Chronic obstructive pulmonary disease MMP: Matrix metalloproteinase RL: Mean pulmonary resistance SNP: Single nucleotide polymorphism SRL: Specific lung resistance
MMP-12 polymorphic variation in the general population appears to influence clinical phenotype in diseases in which MMP-12 could play a prominent role. Two putative functional polymorphisms have been characterized in the MMP12 gene. A common asparagine to serine substitution at codon 357 of the MMP12 gene (1082A/G, rs652438) is associated with worse survival outcomes in patients with breast cancer,10 indicating faster tumor invasion with this allele, possibly because of a greater risk of degradation of the extracellular matrix. Haplotypes containing the MMP12 Asn357Ser (rs652438) allele are associated with lung function decrease in smokers with chronic obstructive pulmonary disease (COPD).11 An A-82G polymorphism occurs at the site of binding of the transcription factor activator protein 1, the A allele being associated with a greater binding affinity for activator protein 1.10,12 This results in higher MMP12 promoter activity.10,12 The A allele is also associated with a clinically adverse outcome, being linked to smaller coronary arterial diameter in diabetic patients.10,12 These putative functional variants would be predicted to influence increased lung remodeling in patients with asthma and might be associated with asthma outcomes and response to treatment. To explore the hypothesis that MMP-12 inhibitors might have therapeutic potential in patients with asthma, we adopted a number of concurrent approaches. We explored the role of the putative functional variants on asthma clinical outcomes that are likely to be affected by acute allergen-responsive proinflammatory (inhaled b2-agonist use and risk of asthma exacerbations) and chronic remodeling (chronic medication needs) responses in children and young adults, as well as disease severity in patients with COPD. In parallel, we developed specific MMP-12 inhibitors and characterized the role of the most specific inhibitor on the regulation of airway inflammation in a well-characterized animal model of allergic airway bronchoconstriction, airway responsiveness, and allergic airway inflammation.
METHODS Population genetic studies in patients with asthma and COPD Children and young adults (3-22 years) with physician-diagnosed asthma recruited into the BREATHE study formed the group of cases in this investigation (Tables I and II). Participants attended primary or secondary clinics in 14 primary care practices and a secondary care asthma clinic in either Tayside or Dumfries. During 2004-2006, we collected mouthwash samples for DNA analysis from 1313 subjects; demographic and anthropometric data were available for 1017 of these participants. A detailed history was obtained, including information on symptoms, treatment, and exacerbations. We measured pulmonary function by means of spirometry (FEV1, forced vital capacity, and peak expiratory flow rate). The asthma prescribing level was determined in accordance with the British Thoracic Society (BTS) guidelines for physician-led management of asthma,13 with some modification as follows:
MUKHOPADHYAY ET AL 71
step 0, no use of inhaled albuterol on demand within the past month; step 1, inhaled albuterol on demand; step 2, regular inhaled steroids plus inhaled albuterol on demand; step 3, regular inhaled salmeterol plus inhaled steroids with inhaled albuterol on demand; and step 4, regular inhaled salmeterol plus inhaled steroids plus oral montelukast with inhaled albuterol on demand. The use of inhaled short-acting b-agonists (bronchodilators) was categorized as follows: 0, rarely or never required; 1, required few times a week but less than once daily; 2, required daily; and 3, multiple doses over a 24-hour period on a regular basis. An asthma exacerbation was defined as the presence of 1 or more of the following 3 indicators during the period of study: (1) asthmarelated hospital admission; (2) 1 or more days of asthma-related absence from school; and (3) prescribed course of oral steroids. All participants were of Northern European origin. Informed consent was provided by the patient and the parent/guardian, as relevant. A control population consisting of a random sample of 1451 specimens from a previous study of 2,454 schoolchildren from the Tayside area was used to assess the possibility that these MMP12 single nucleotide polymorphisms (SNPs) might be involved in susceptibility to asthma. Informed consent and saliva DNA samples were obtained from 989 patients with physician-diagnosed COPD, and clinical phenotypes were obtained from their medical records. All studies were approved by the Tayside Committee on Medical Research Ethics. DNA extraction and genotyping were performed as described in our previously reported studies in patients with asthma.14-18 DNA in the mouthwash specimens was extracted with QIAamp 96 DNA blood kits (Qiagen, Hilden, Germany). The codon 357 variant and the 282 A/G promoter variant were genotyped in both asthmatic children and young adults and population control subjects (see Tables E1 and E2 in this article’s Online Repository at www.jacionline.org). The allele frequency of both variants was similar in the patients with asthma (serine variant, 0.048; 95% CI, 0.039-0.058; n 5 997) and population control subjects (serine variant, 0.051; 95% CI, 0.0430.059; n 5 1431), demonstrating that they are not associated with an increased susceptibility to asthma per se (see Tables E1 and E2). This is consistent with the prior observations regarding the serine variant and lung cancer susceptibility. We therefore examined the association between these variants and the clinical phenotype within the asthmatic population. We have previously extensively validated and tested the above measures of asthma severity and control in this population of young asthmatic patients and have demonstrated associations between other common gene polymorphisms and clinical phenotype in allergy and asthma.14-18 Because the serine variant occurs at a low frequency, subjects carrying 1 or 2 copies of the serine variant were analyzed as one group. Allelic discrimination assays for the 282A/G (rs2276109) and 1082A/G (rs652438) SNPs in the MMP12 gene were carried out using in-house assays in the asthmatic population (details available on request) and Applied Biosystems–catalogued assays in the control population with a 7700 Sequence Detection System (Applied Biosystems, Foster City, Calif). Results obtained with the 2 assay systems were 100% concordant. The genotype distribution of both MMP12 SNPs was in Hardy-Weinberg equilibrium (P > .05), and genotyping call rates ranged from 96% to 97%. Statistical analysis was carried out with SPSS for Windows version 15 software (SPSS, Inc, Chicago, Ill). Differences in the genotype distributions between the groups of cases and control subjects were evaluated by using the x2 test. Logistic regression analysis, controlling for age and exposure to cigarette smoke, was used to determine odds ratios for the occurrence of any exacerbation. Sex was considered as a covariate but was subsequently removed from all models because it was not a significant covariate for exacerbations or severity in the BREATHE study.
Laboratory studies Development of a specific MMP-12 inhibitor. We subsequently proceeded to screen and identify potential candidate compounds with specific inhibitory properties for MMP-12. A unique class of compounds featured with a dibenzofuran core and a zinc-chelating group were prepared, modified, and profiled for their drug-like properties. Compounds were selected for their potency in both enzymatic and cell-based assays, as well as their selectivity against a panel of 8 other closely related MMPs, as well as
72 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE III. Half-maximal inhibitory concentrations for MMP40820 on MMP-12 and other related enzymes
TABLE I. Summary characteristics of 1,017 asthmatic patients and 1451 population control subjects
Sex, male/female (n) Age (y)* Body mass index (kg/m2)*
Asthmatic patients
Control subjects
608/409 10.43 (7.13-13.12) 19.22 (16.09-21.11)
724/727 7.42 (6.51-8.45) 17.11 (15.66-18.09)
*Mean (25th-75th percentile).
TABLE II. Clinical characteristics of asthmatic patients Lung function
Peak expiratory flow rate (% predicted) FEV1 (% predicted) Forced vital capacity (% predicted) BTS step of treatment (n 5 1001) 0 1 2 3 4 Asthma exacerbation in preceding 6 mo (n 5 971) Any hospital admission Any school absence One or more courses of oral steroids Any exacerbation
Mean (25th-75th percentile)
88.0 95.8 92.3 No. 30 176 569 133 93 No.
(77.0-98.4) (85.8-105.7) (83.2-101.1) (%) (3.0) (17.6) (56.8) (13.3) (9.3) (%)
113 301 205 352
(11.2) (30.8) (20.4) (36.2)
No.
811 814 813
TNF-a converting enzyme and aggrecanase 1 and 2. The procedure followed for the development of the active tricyclic compound has already been described in detail.19,20 One highly potent and specific inhibitor selective for MMP-12, (S)-2-(8 (methoxycarbonylamino)dibenzo[b,d]furan-3-sulfonamido)-3-methylbutanoic acid (MMP408; Table III),19,20 was taken forward for testing in an Ascaris suum–sensitized sheep model of allergic airway disease.9 Test and control articles. A suum extract (Greer Diagnostics, Lenoir, NC) was diluted with PBS to a concentration of 82,000 protein nitrogen units/mL and delivered as an aerosol (20 breaths/min 3 20 minutes). This crude preparation has an endotoxin level of 50 EU/mL, an amount that does not have an effect on pulmonary responses in sheep. Carbamylcholine (Carbachol; Sigma Chemical Co, St Louis, Mo) was dissolved in buffered saline at concentrations of 0.25, 0.50, 1.0, 2.0, and 4.0% wt/vol and delivered as an aerosol. Animal preparation. A well-validated animal model of allergic airway disease was used for these studies.9 Sheep weighing 45 to 55 kg that were naturally sensitized to the nematode A suum were used. All sheep had previously been shown to develop both early and late bronchial responses to inhaled A suum antigen. The sheep were conscious and restrained in a modified shopping cart in the prone position with their heads immobilized. After topical anesthesia of the nasal passages with 2% lidocaine, a balloon catheter was advanced through one nostril into the lower esophagus. The animals were intubated with a cuffed endotracheal tube through the other nostril with a flexible fiberoptic bronchoscope as a guide. Airway mechanics. Breath-by breath determination of mean pulmonary resistance (RL) was measured with the esophageal balloon technique. Pleural pressure was estimated with an esophageal balloon catheter (filled with 1 mL of air), which was positioned 5 to 10 cm from the gastroesophageal junction. With the catheter in this position, the end-expiratory pleural pressure ranged from 22 to 25 cm H2O. Once the balloon was placed, it was secured so that it remained in the same position for the duration of the experiment. Lateral pressure in the trachea was measured with a sidehole catheter (internal diameter, 2.5 mm) advanced through and
Enzyme
Human MMP-12 Monkey MMP-12 Dog MMP-12 Rabbit MMP-12 Sheep MMP-12 Mouse MMP-12 Rat MMP-12 Human MMP-1 Human MMP-2 Human MMP-3 Human MMP-7 Human MMP-8 Human MMP-9 Human MMP-13 Human MMP-14 TNF-a converting enzyme Aggrecanase 1 Aggrecanase 2
IC50 (nmol/L)
2 1.5 3.4 6.7 22.3 160 320 >5000 61 351 >5000 41 1300 120 1100 >5000 >5000 >5000
IC50, Half-maximal inhibitory concentration.
positioned distal to the tip of the endotracheal tube. The tracheal and pleural-pressure catheters were connected to a differential pressure transducer (MP45; Validyne, Northridge, Calif) for the measurement of transpulmonary pressure, which was defined as the difference between tracheal and pleural pressure. Airflow was measured by connecting the proximal end of the endotracheal tube to a pneumotachograph (Fleisch No. 1; Dyna Sciences, Inc, Blue Bell, Pa). The transpulmonary pressure and flow signals were recorded on a multichannel physiologic recorder, which was linked to an 80386 DOS Personal Computer (CCI, Inc, Miami, Fla) for online calculation of RL by dividing the change in transpulmonary pressure by the change in flow at midtidal volume (obtained by means of digital integration). The mean of at least 5 breaths free of swallowing artifact was used to obtain RL in centimeters of H2O per liter per second. Immediately after the measurement of RL, thoracic gas volume was measured in a constant-volume body plethysmograph to obtain specific lung resistance (SRL) as follows: SRL 5 RL 3 Thoracic gas volume. Aerosol delivery systems. All aerosols were generated with a disposable medical nebulizer (Raindrop; Puritan Bennett, Lenexa, Kan) that provided an aerosol with a mass median aerodynamic diameter of 3.2 mm, as determined with an Andersen cascade impactor. The nebulizer was connected to a dosimeter system consisting of a solenoid valve and a source of compressed air (20 psi). The output of the nebulizer was directed into a plastic T-piece, one end of which was connected to the inspiratory port of a piston respirator (Harvard Apparatus, Mills, Mass). The solenoid valve was activated for 1 second at the beginning of the inspiratory cycle of the respirator. Aerosols were delivered at a tidal volume of 500 mL and a rate of 20 breaths/min.
Concentration response curves to carbachol aerosol. To assess bronchial responsiveness, cumulative concentration-response curves to carbachol were generated by means of SRL immediately after inhalation of buffer and after each consecutive administration of 10 breaths of increasing concentrations of carbachol (0.25%, 0.5%, 1.0%, 2.0%, and 4.0% [wt/vol] in buffered saline). The provocation test was discontinued when SRL increased by more than 400% from the postsaline value or after the highest carbachol concentration had been administered. Bronchial responsiveness was assessed by determining the cumulative carbachol concentration (in breath units) that increased SRL by 400% over the postsaline value (PC400) through interpolation from the dose-response curve. One breath unit was defined as one breath of a 1% (wt/vol) carbachol aerosol solution.
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
MUKHOPADHYAY ET AL 73
FIG 1. A-C, The MMP12 serine 357 variant is associated with clinically worse asthma with relation to increasing inhaled b2-agonist use (Fig 1, A), asthma treatment step (Fig 1, B), and risk of asthma exacerbations over 6 months (Fig 1, C; P for x2 analysis trend). D, Increasing medication use positively correlates with exacerbation risk, implying that current treatment regimens are inadequate overall for effective asthma control.
Evaluation of the efficacy of an MMP-12–specific inhibitor in a sheep model of allergic airway disease. Compounds were dosed either intravenously or through the oral route twice daily the day before challenge and then the following day (day of A suum challenge) 1 hour before challenge and 8 hours after challenge. Increases in airway resistance were measured throughout the day to capture both the early-phase bronchoconstrictor response that peaked approximately 1 hour after allergen challenge and the late-phase bronchoconstrictor response that appeared approximately 5 to 6 hours after allergen challenge. AHR to aerosolized carbachol was measured the following day. After AHR measurements, lungs underwent lavage, and total cell counts were quantified in the bronchoalveolar lavage fluid.
RESULTS Association of gene variant with clinical asthma phenotype Participants with the serine variant showed greater airway lability, which was measured as greater frequency of use of inhaled short-acting b2-agonists (x2 5 7.6 and P 5 .031, test for trend; Fig 1, A, and see Table E3 in this article’s Online Repository at www.jacionline.org). More dramatically, however, when all asthma medication was considered, the carrier frequency for the serine variant also increased in line with asthma treatment step as per the modified BTS guidelines (x2 5 12.7 and P 5 .0004, test for trend; Fig 1, B).13 By using binary logistic regression, the odds ratio for having greater severity of asthma, requiring the regular preventive use of long-acting b-agonists (BTS step 3 and greater), was 2.00 (95% CI, 1.24-3.24; P 5 .004) when comparing those patients carrying the serine variant with those without the serine variant. We also performed a logistic regression analysis of MMP12 genotype on exacerbations of asthma over a period of 6 months. The serine variant was associated with a trend in the increase of frequency of any school absence and oral steroid
use. Overall, the presence of the serine variant nearly doubled the risk of an asthma exacerbation (odds ratio, 1.90; 95% CI, 1.19-3.04; P 5 .008), with a significant increase in the frequency of the serine variant in participants with exacerbations in comparison with those without exacerbations over the 6-month period of study (Fig 1, C). This increase in exacerbation rate in the carriers of the serine variant is seen despite the increased medication being applied to these patients, further highlighting the importance of this variant in overall asthma severity in these children. Indeed, exacerbation rates are highly colinear with the increased medication regimen, which both confirms the use of the treatment step as a measure of severity and reveals a degree of inadequacy in current medications in the management of childhood asthma (Fig 1, D). The association between the MMP12 variant and both exacerbations was partly, although not entirely, due to this collinearity. The serine variant was significantly associated with increased exacerbations within the less severe group not prescribed long-acting b-agonists (BTS step 2 or below; odds ratio, 2.04; 95% CI, 1.2-3.5; P 5 .009). In contrast, no associations were seen between the 282 A/G promoter variant and any of the clinical parameters when analyzed as a single marker or in combination with the serine variant (see Tables E4-E6 in this article’s Online Repository at www.jacionline.org). Pulmonary function was measured reproducibly in 815 patients. Our initial analysis was hypothesis driven, exploring the role of the 2 specific potentially at-risk variants in the MMP12 gene. However, we have also subsequently extensively genotyped the MMP12 locus using a haplotype-tagging approach to determine the specificity of our observation with the serine 357 variant. Nine additional SNPS were typed across the MMP12 locus and 10 kb of flanking sequence in either direction that tagged greater than 90% of the genetic variation represented by HapMap version 2.2 (see Table E7 and Fig E1 in this article’s Online Repository at
74 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
www.jacionline.org). None of these additional variants had any effect on our 2 principal clinical outcomes: asthma treatment step (see Table E8 in this article’s Online Repository at www. jacionline.org) and exacerbation risk (see Table E9 in this article’s Online Repository at www.jacionline.org). Genotyping of published functional SNPs from the adjacent MMP genes MMP1, MMP3, MMP10, and MMP13 did not reveal any additional signal (see Tables E10 and E11 in this article’s Online Repository at _5 years) is www.jacionline.org). Asthma in early childhood (< thought to differ from the disease in older children, but stratification of the analysis on exacerbations or treatment step did not affect the conclusions (see Table E12 in this article’s Online Repository at www.jacionline.org).
Replication of gene variant association findings in COPD severity We sought to replicate our principal finding of an effect of the rs652438 variant on disease severity in a large cohort of patients with COPD in Tayside (see Table E13 in this article’s Online Repository at www.jacionline.org) for whom DNA has recently become available. By using the x2 test (Pearson linear-by-linear method), a significant relationship was observed between the presence of the at-risk variant and disease severity as defined by FEV1 (normalized as percent predicted). In this analysis we observed a relationship between the frequency of the serine variant and decreased lung function (P 5 .016, see Table E14 in this article’s Online Repository at www.jacionline.org). The 282 promoter variant rs2276109 was not associated with lung function in this population (see Table E15 in this article’s Online Repository at www.jacionline.org).
Study of the role of MMP-12 inhibitor in sheep When animals were treated intravenously with the MMP-12 inhibitor MMP408 at 10 mg/kg, there was a complete blockade of the influx of inflammatory cells into the bronchoalveolar cavity in comparison with control levels seen in untreated A suum–challenged sheep (Fig 2, A). To determine its effects on airway resistance, MMP408 at 10 mg/kg was administered intravenously at 10 mg/kg twice daily, the day before A suum antigen challenge, and again 1 hour before challenge. MMP408 significantly attenuated the early-phase allergic airway bronchoconstriction and completely blocked the late-phase allergic airway bronchoconstriction as measured by an increase in RL compared with responses observed in the absence of the inhibitor (Fig 2, B). Oral administration of MMP408 at 10 mg/kg had comparable effects on the late-phase allergic airway bronchoconstriction as observed with intravenous administration (Fig 2, C). The animals received a fourth dose of MMP408 eight hours after A suum challenge to determine the efficacy of the blockade = FIG 2. MMP-12 inhibition by MMP408 reduced total bronchoalveolar lavage inflammation (A), airway resistance (0-8 hours; B and C), and carbachol responsiveness (24 hours; D and E) after A suum antigen challenge in sheep (MMP408 dose: Fig 2, A, B, and D, 10 mg/kg administered intravenously; Fig 2, C and E, 10 mg/kg administered orally). Graphs represent group means 6 SEMs. *P < .05, 5 animals per group, Student paired t test versus test control subjects.
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
on carbachol-induced AHR. AHR to aerosolized carbachol that was measured the following day was inhibited with this dosing regimen after both intravenous and oral administration of MMP408 (Fig 2, D and E).
DISCUSSION This article describes a consistent role for the Asn357Ser variant of MMP12 in the disease pathway of both asthma and COPD; however, the study has several potential limitations. Multiple testing is always an issue with gene association studies, and in this case we tested 2 previously studied SNPs providing a priori evidence for their involvement in disease and then tested for non–hypothesis-driven haplotype-tagging screening of the chromosomal region to test the specificity of the results for these putative functional variants. Correction of the results for the multiple testing of the 2 prespecified SNPs does not abolish the significance of the results for the Asn357Ser variant. Bonferroni correction across the screening SNPs abolished all significant observations, confirming the specificity of the observations for the Asn357Ser variant. Multiple testing across the phenotypes was not applied because of the interdependence of the phenotypes and the fact that the Asn357Ser variant was significantly associated with the majority of the available outcome measures. Another consideration in gene association studies is population stratification. This was a highly homogenous population from a single region of Scotland, and there was no difference in the allele frequencies between the asthmatic and control groups, suggesting that the current observations are highly unlikely to be the result of some inherent population substructure. The data presented in this study do not prove that the Asn357Ser variant is the causal variant in this process, and recent studies have suggested a role for the 282 promoter variant in asthma and COPD susceptibility,21 a concept that is not supported in our current data. This raises the possibility that the genetic architecture of the MMP12 locus might be complex and differ both between populations, disease context, or both. It also raises the possibility that a causal variant is yet to be found and that the differential signals are due to population differences in haplotype diversity. Our data do display a trend toward a protective role for the rs2276109 (282) genotype, as recently suggested,21 and therefore it is also possible that given a larger sample size, we might be able to detect an effect with this variant. The current treatment of asthma rests primarily on the management of overall inflammation through the use of inhaled steroids, smooth muscle relaxation by b2-agonists, and, more recently, specific inflammatory mediator–blocking therapy (montelukast) that is effective in a proportion of patients.13 This work suggests that MMP-12 could represent a more specific inflammatory molecule in human subjects, regulating pathways for acute airway changes associated with allergic airway inflammation and chronic airway remodeling, which could contribute to a more severe clinical asthma phenotype, including a greater risk of exacerbation. There is much recent evidence that genetic variation, possibly interacting with environmental factors, regulates susceptibility to and the severity of childhood diseases associated with atopy. A more complete understanding of these influences might create a composite picture of these interactions, thus providing novel insights into the natural history of childhood diseases associated with atopy and leading to the development of new therapeutic
MUKHOPADHYAY ET AL 75
targets and strategies. The poorly functioning epidermal protein filaggrin, for example, occurs in about 8% of white subjects from the United Kingdom and has recently been shown to contribute to overall increased susceptibility to atopic diseases triggering asthma in childhood.14,16,17 In contrast, we have shown that MMP12 gene variations regulate the clinical asthma phenotype but do not affect the susceptibility to asthma in this current population (see Table E1). These effects at the genetic and clinical levels now require mapping onto the natural history of childhood atopic disease through functional studies to provide an improved understanding of atopic disease in childhood. It might be hypothesized that unless processes such as epithelial filaggrin protein malfunction let in a critical allergen dose, the activation of the immune cascade that sets off airway inflammation does not occur. Differences in MMP-12 function in the airway might become clinically relevant only when such airway inflammation sets in. These complex gene-environment interactions can, however, only be studied through the prospective analysis of much larger datasets of genotyped and appropriately phenotyped subjects. The inflammation-related and regenerative processes within COPD are closely related to those within asthma, and we have referred above to how our work was primarily driven by the previous observation that haplotypes that contain the MMP12 Asn357Ser (rs652438) allele might contribute to lung function decrease in smokers with COPD.11 We have now genotyped a large cohort of patients with COPD in Tayside about whom detailed longitudinal clinical data are being collected over a number of years. We used this resource to explore the possible role of the rs652438 allele on disease severity in patients with COPD (see Table E13). We observed a significant relationship between the presence of this at-risk gene variant and the last measurement of FEV1 (normalized as percent predicted, see Table E14). We believe these findings from a related chronic respiratory disease in a different age group of patients reinforce the role of this at-risk gene variant in airway inflammation. We have not observed associations with pulmonary function in our asthma cohort because, unlike in participants with COPD, pulmonary function (normalized FEV1) is normal in young asthmatic patients in the United Kingdom during the quiescent stage (Table II), which is not the case in asthmatic populations in all countries. This difference in lung function preservation between populations might contribute to the differences seen between the effects of different MMP12 SNPs on asthma in different populations. MMP12 might influence asthma from a number of different aspects. It is thus possible that the 282 promoter variant might have effects on the asthma phenotype that are different from the serine 357 variant. These effects could be regulated through differences in the mechanism of action between the 2 genetic variants, and these questions require further study. Our concurrent development of an MMP-12–specific inhibitor19,20 that blocks allergen-induced airway disease in sheep offers direct hope of the therapeutic targeting of MMP-12 in asthmatic patients. The stage is thus set for early exploratory safety and efficacy trials of specific MMP-12 inhibition in young asthmatic patients. Efficacy outcomes of interest, on the basis of our results and knowledge of MMP-12 function, might include early and late airway responses and carbachol sensitization, methacholine sensitization, or both (over days or weeks) and airway remodeling assessment over a longer period of time (ie, months). Airway remodeling is established during childhood22,23 and acts as a critical determinant for asthma severity throughout
76 MUKHOPADHYAY ET AL
life.24 Thus we speculate that the inhibition of MMP12 might require investigation as a possible therapeutic strategy for asthma and COPD. We acknowledge support from the following colleagues: Rustem Krykbaev (Pfizer Research, Inflammation and Immunology) for cloning of sheep MMP12; Paul Morgan (Pfizer Research, Inflammation and Immunology) and Kristina Cunningham (Pfizer Research, Chemical and Screening Sciences) for in vitro compound specificity assays; Jianchang Li and Yuchan Wu (Pfizer Research, Chemical and Screening Sciences) for compound synthesis and scaleup; Vicky Alexander, Tahmina Ismail, and Inez Murrie (University of Dundee Medical School) for clinical data collection and entry and help with the development of the asthma database in Scotland; and David Crook (Royal Sussex County Hospital) for useful comments on the manuscript.
Clinical implications: The specific inhibition of MMP-12– driven airway inflammation and airway remodeling might represent a new therapeutic strategy for asthma and COPD, justifying exploratory human safety and efficacy trials.
REFERENCES 1. Warner RL, Lukacs NW, Shapiro SD, Bhagarvathula N, Neruso KC, Varani J, et al. Role of metalloelastase in a model of allergic lung responses induced by cockroach allergen. Am J Pathol 2004;165:1921-30. 2. Chandler S, Cossins J, Lury J, Wells G. Macrophage metalloelastase degrades matrix and myelin proteins and processes a tumour necrosis factor-alpha fusion protein. Biochem Biophys Res Commun 1996;228:421-9. 3. Gronski TJ Jr, Martin RL, Kobayashi DK, Walsh BC, Holman MC, Huber M, et al. Hydrolysis of a broad spectrum of extracellular matrix proteins by human macrophage elastase. J Biol Chem 1997;272:12189-94. 4. Lanone S, Zheng T, Zhu Z, Lee CG, Ma B, Chen Q, et al. Overlapping and enzyme-specific contributions of matrix metalloproteinases-9 and -12 in IL-13– induced inflammation and remodelling. J Clin Invest 2002;110:463-74. 5. Agrawal DK, Townley RG. Inflammatory cells and mediators in bronchial asthma. In: Hollinger MA, editor. CRC press series in pharmacology and toxicology. Boca Raton (FL): CRC Press; 1990. 6. Pouladi MA, Robbins CS, Swirski FK, Cundall M, McKenzie AN, Jordana M, et al. Interleukin-13-dependent expression of matrix metalloproteinase-12 is required for the development of airway eosinophilia in mice. Am J Respir Cell Mol Biol 2004;30:84-90. 7. Lavigne MC, Thakker P, Gunn J, Wong A, Miyashiro JS, Wasserman AM, et al. Human bronchial epithelial cells express and secrete MMP-12. Biochem Biophys Res Commun 2004;324:534-46. 8. Molnar J, Sohar I, Kovacs J, Rakonczay Z, Rausch H. Effects of tricyclic compounds on membrane binding of bivalent cations, activities of acetylcholinesterase and some tissue proteases. In Vivo 1993;7:431-4.
J ALLERGY CLIN IMMUNOL JULY 2010
9. Abraham WM. Modeling of asthma, COPD and cystic fibrosis in sheep. Pulm Pharmacol Ther 2008;21:743-54. 10. Shin A, Cai Q, Shu X, Gao YT, Zheng W. Genetic polymorphisms in the matrix metalloproteinase 12 gene (MMP12) and breast cancer risk and survival: the Shanghai Breast Cancer Study. Breast Cancer Research 2005;7:R506-12. 11. Joos L, He JQ, Shepherdson MB, Connett JE, Anthonisen NR, Pare´ PD, et al. The role of matrix metalloproteinase polymorphisms in the rate of decline in lung function. Hum Mol Genet 2002;11:569-76. 12. Jormsjo S, Ye S, Moritz J, Walter DH, Dimmeler S, Zeiher AM, et al. Allelespecific regulation of matrix metalloproteinase-12 gene activity is associated with coronary artery luminal dimensions in diabetic patients with manifest coronary artery disease. Circ Res 2000;86:998-1003. 13. BTS/SIGN. British guidelines on the management of asthma. Thorax 2003; 58(suppl 1):1-94. 14. Palmer CN, Irvine AD, Terron-Kwiatkowski A, Zhao Y, Liao H, Lee SP, et al. Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis. Nat Genetics 2006;38: 441-6. 15. Palmer CN, Doney AS, Ismail T, Lee SP, Murrie I, Macgregor DF, et al. PPARG locus haplotype variation and exacerbations in asthma. Clin Pharmacol Ther 2007; 81:713-8. 16. Palmer CN, Ismail T, Lee SP, Terron-Kwiatkowski A, Zhao Y, Liao H, et al. Filaggrin null mutations are associated with increased asthma severity in children and young adults. J Allergy Clin Immunol 2007;120:64-8. 17. Basu K, Palmer CN, Lipworth BJ, Irwin McLean WH, Terron-Kwiatkowski A, Zhao Y, et al. Filaggrin null mutations are associated with increased asthma exacerbations in children and young adults. Allergy 2008;63:1211-7. 18. Tavendale R, Macgregor DF, Mukhopadhyay S, Palmer CNA. A polymorphism controlling ORMDL3 expression is associated with asthma that is poorly controlled by current medications. J Allergy Clin Immunol 2008;121:860-3. 19. Li W, Li J, Wu Y, Tam SY, Mansour T, Sypek JP, et al. Tricyclic compounds as matrix metalloproteinase inhibitors. WO 2008/057254 A2; international publication date, May 15, 2008. Available at: http://www.wipo.int/pctdb/en/wo. jsp?WO52008057254&IA5US2007022651&DISPLAY5DESC). 20. Li W, Li J, Wu Y, Hotchandani R, Cunningham K, McFadyen I, et al. A selective matrix metalloprotease 12 inhibitor for potential treatment of chronic obstructive pulmonary disease (COPD): discovery of (S)-2-(8 (methoxycarbonylamino)dibenzo[b, d]furan-3-sulfonamido)-3-methylbutanoic acid (MMP408). J Med Chem 2009;52:1799-802. 21. Hunninghake GM, Cho MH, Tesfaigzi Y, Soto-Quiros ME, Avila L, Lasky-Su J, et al. MMP12, lung function, and COPD in high-risk populations. N Engl J Med 2009;361:2599-608. 22. De Blic J, Tillie-Leblond I, Emond S, Mahut B, Dang Duy TL, Scheinmann P. High-resolution computed tomography scan and airway remodeling in children with severe asthma. J Allergy Clin Immunol 2005;116:750-4. 23. Saglani S, Payne DN, Zhu J, Wang Z, Nicholson AG, Bush A, et al. Early detection of airway wall remodeling and eosinophilic inflammation in preschool wheezers. Am J Respir Crit Care Med 2007;176:858-64. 24. James AL, Wenzel S. Clinical relevance of airway remodelling in airway diseases. Eur Respir J 2007;30:134-55.
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
MUKHOPADHYAY ET AL 76.e1
FIG E1. Linkage disequilibrium relationship of haplotype-tagging SNPs across the MMP12 locus. Visualized with Haploview V4.1: Hap Map CEU Phase 2 Release 21. CEU genotypes. The R2 values are shown from the CEU dataset.
76.e2 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E1. Genotype distribution in asthmatic cases and population control subjects Genotype/n SNP
282A/G Asthmatic cases Control subjects Asn357Ser Asthmatic cases Control subjects
AA
AG
GG
n
HWE
MAF
P value
780 1093
216 295
14 27
1010 1415
0.827 0.178
0.121 0.123
.791
905 1285
88 132
4 7
997 1424
0.243 0.076
0.048 0.051
.624
HWE, Hardy-Weinberg equilibrium; MAF, minor allele frequency.
MUKHOPADHYAY ET AL 76.e3
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E2. Two-marker genotype distribution in asthmatic cases Asn357Ser 282A/G
AA AG GG Total
Asn/Asn AA
Asn/Ser AG
Ser/Ser GG
Total
685 200 13 898
75 13 0 88
4 0 0 4
764 213 13 990
282G and serine 357 are resident on different ancestral haplotypes and thus rarely occur as the joint carrier genotype.
76.e4 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E3. Contingency table for MMP12 genotype versus inhaled b2-agonist use on a scale of 0 to 3 (see the Methods section for details) Asn357Ser SNP
Inhaled b2-agonist use 0 1 2 3
Asn/Asn
100 623 142 19
Serine/X
(96.2%) (90.4%) (90.4%) (79.2%)
4 66 15 5
(3.8%) (9.6%) (9.6%) (20.8%)
Total
P value
104 689 157 24
.0305
Total
P value
282A/G SNP
Inhaled b2-agonist use 0 1 2 3
AA
81 540 121 21
(76.4%) (77.6%) (75.6%) (84.0%)
P value relates to trend on x2 analysis by means of linear-by-linear association.
AG/GG
25 156 39 4
(23.6%) (22.4%) (24.4%) (16.0%)
106 696 160 25
.839
MUKHOPADHYAY ET AL 76.e5
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E4. MMP12 genotype and drug treatment Genotype/no. (%) MMP12 282A/G
BTS drug class 0 1 2 3 4 P value (x2)
MMP12 Asn357Ser
AA
AG1GG
18 (2.3) 137 (17.9) 438 (57.2) 105 (13.7) 68 (8.9) Test for trend (1 df)
12 (5.3) 35 (15.4) 129 (56.6) 28 (12.3) 24 (10.5) .832
AA
30 164 508 111 77
(3.4) (18.4) (57.1) (12.5) (8.7)
AG1GG
0 (0.0) 10 (11.0) 49 (53.8) 18 (19.8) 14 (15.4) .000374
76.e6 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E5. MMP12 282A/G genotype and risk of specific asthma exacerbations Clinical outcome
Hospital admission School absence Oral steroid use Any
Odds ratio
95% CI
P value
0.710 0.968 0.796 0.915
0.419-1.203 0.692-1.353 0.536-1.183 0.660-1.267
.203 .847 .259 .591
MUKHOPADHYAY ET AL 76.e7
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E6. MMP12 codon 357 genotype and risk of specific asthma exacerbations Clinical outcome
Hospital admission School absence Oral steroid use Any
Odds ratio
95% CI
P value
0.734 1.535 1.416 1.898
0.324-1.660 0.948-2.485 0.833-2.406 1.185-3.040
.457 .0812 .199 .00768
76.e8 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E7. Linkage disequilibrium relationships between pairs of SNPs in the population of children with asthma rs673163
1.0
rs586701
rs580266
rs12808148
rs476185
rs652438
rs10895367
rs2276109
rs17368814
rs11225445
0 1.0
0.48 0 1.0
0.47 0 0 1.0
0 0 0 0 1.0
0 0.48 0 0 0 1.0
0 0 0.53 0 0 0 1.0
0 0.79 0 0 0 0 0 1.0
0 0.80 0 0 0 0 0 1.0 1.0
0.87 0 0.36 0.52 0 0 0 0 0 1.0
Shown is the R2 value between the markers for this population.
rs673163 rs586701 rs580266 rs12808148 rs476185 rs652438 rs10895367 rs2276109 rs17368814 rs11225445
MUKHOPADHYAY ET AL 76.e9
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E8. MMP12 haplotype-tagging SNPs and drug class: Linear regression P values with age and exposure to cigarette smoke as covariates MMP12 SNP
P value
rs673163 rs586701 rs580266 rs12808148 rs476185 rs652438 rs10895367 rs2276109 rs17368814 rs11225445
.955 .338 .562 .781 .853 .000374 .591 .664 .293 .694
76.e10 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E9. MMP12 haplotype-tagging SNPs and risk of any asthma exacerbation: Binary logistic regression P values with age and exposure to cigarette smoke as covariates MMP12 SNP
P value
rs673163 rs586701 rs580266 rs12808148 rs476185 rs652438 rs10895367 rs2276109 rs17368814 rs11225445
.205 .410 .658 .509 .346 .00768 .829 .591 .447 .210
MUKHOPADHYAY ET AL 76.e11
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E10. MMP gene cluster functional polymorphisms and drug treatment class Reference
rs1799750 rs679620 rs17293607 rs2276109 rs652438 rs2252070
Gene
P value*
MMP1 MMP3 MMP10 MMP12 282A/G MMP12 Asn357Ser MMP13
.991 .516 .932 .832 .000374 .932
*P value (x2), test for trend (1 df).
76.e12 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E11. MMP gene cluster functional polymorphisms and asthma exacerbations Reference
rs1799750 rs679620 rs17293607 rs2276109 rs652438 rs2252070
Gene
Odds ratio
95% CI
P value
MMP1 MMP3 MMP10 MMP12 282A/G MMP12 Asn357Ser MMP13
1.085 0.993 0.954 0.915 1.898 0.955
0.897-1.312 0.818-1.205 0.725-1.255 0.660-1.267 1.185-3.040 0.727-1.256
.400 .942 .736 .591 .00768 .743
MUKHOPADHYAY ET AL 76.e13
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E12. Associations with MMP12 codon 357 variant and exacerbations and drug class stratified by age (<6 and >6 years of age)
Exacerbation <6 y >6 y Drug class (0-2 vs 3-4) <6 y >6 y
Odds ratio
95% CI
P value
7.50 1.71
0.859-65.08 0.818-1.205
.068 .033
1.96 2.02
0.459-8.386 1.215-3.359
.363 .007
Binary logistic regression adjusted for age and exposure to smoke.
76.e14 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E13. Summary characteristics of 989 patients with COPD in Tayside, Scotland Variable
Mean (SD)
Range
No.
Age at diagnosis (y) Follow-up (y) FEV1 at last visit (% of mean predicted value) Smoking pack-years Body mass index (kg/m2)
63.9 (9.4) 4.01 (1.87) 63.0 (20.1)
34.6-88.5 0-8.00 10.3-131.9
989 989 989
41.8 (20.5) 26.4 (5.46)
0-150 15.4-54.2
925 970
MUKHOPADHYAY ET AL 76.e15
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E14. rs652438 genotype and most recent measurement of FEV1 in a longitudinal cohort of patients with COPD in Tayside (n5 986) Percent predicted FEV1/n rs652438 genotype
>80% FEV1
50% to 80% FEV1
30% to 50% FEV1
<30% FEV1
AA AG1GG
194 (94.6) 11 (5.4)
458 (88.9) 57 (11.1)
194 (89.8) 22 (10.2)
41 (82.0) 9 (18.0)
Three subjects did not genotype successfully for rs652438. *Linear-by-linear association: P 5 .013 after adjustment for age, sex, duration of disease, and smoking pack-years.
P value (x2)*
.0161
76.e16 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E15. rs2276109 (282) genotype and most recent measurement of FEV1 in a longitudinal cohort of patients with COPD in Tayside (n 5 982) Percent predicted FEV1/n rs2276109 genotype
>80% FEV1
50% to 80% FEV1
30% to 50% FEV1
<30% FEV1
AA AG1GG
164 (79.6) 42 (20.4)
404 (78.8) 109 (21.2)
160 (74.4) 55 (25.6)
41 (85.4) 7 (14.6)
Four subjects who were genotyped for rs652438 were not successfully genotyped for rs2276109. *Linear-by-linear association: P 5 .251 after adjustment for age, sex, duration of disease, and smoking pack-years.
P value (x2)*
.676
From aThe Department of Pediatrics, Royal Alexandra Children’s Hospital, Brighton and Sussex Medical School, Brighton; bPfizer Research, Inflammation and Immunology, Cambridge; cthe Population Pharmacogenetics Group, Biomedical Research Institute, University of Dundee, Ninewells Hospital and Medical School, Dundee; dPfizer Research, Chemical and Screening Sciences, Cambridge; eTMRC Laboratory, University of Dundee; fPfizer Research, Biologic Therapeutics, Global Biotherapeutics Technologies, Cambridge; gDirectorate of Paediatrics, NHS Tayside, Ninewells Hospital, Dundee; hMt Sinai Medical Center, Department of Research, University of Miami, Miami Beach; and iPfizer Research, Discovery Translational Medicine, Collegeville. *These authors contributed equally to this work. Supported by the Translational Medicine Research Collaboration (INF-DU-031) and Wyeth Research. The BREATHE study of asthma in children was funded by Scottish Enterprise Tayside, the Gannochy Trust (Perth), and the Perth and Kinross City Council. Disclosure of potential conflict of interest: S. Mukhopadhyay receives research support from MSD UK and Wyeth plc and has provided legal consultation/expert witness testimony in cases related to cystic fibrosis. J. Winter provides personal injury reports per annum income @ £20k. J. Sypek, W. Li, K. Page, M. Fleming, J. Brady, M. O’Toole, S. Goldman, S. Tam, C. Williams, and D. K. Miller are employed by Pfizer Research. D. F. Macgregor receives research support from MSD and has provided legal consultation/expert witness testimony in cases related to pediatric respiratory illness and child protection issues. W. Abraham receives research support from Wyeth, the National Institute of Environmental Health Sciences, Parion, JSJ Pharmaceuticals, and Novartis. C. N. A. Palmer receives research support from Wyeth Pharma. The rest of the authors have declared that they have no conflict of interest. Received for publication April 13, 2009; revised March 22, 2010; accepted for publication March 25, 2010. Available online May 24, 2010. Reprint requests: Somnath Mukhopadhyay, FRCPCH, MD, PhD, Chair of Paediatrics, Academic Unit level 10C, Royal Alexandra Children’s Hospital, Eastern Road, Brighton, BN2 5BE, United Kingdom. E-mail: [email protected]. Joseph Sypek, PhD, Inflammation and Immunology, Pfizer Research, 200 Cambridge Park Dr, Cambridge, MA 02140. E-mail: [email protected]. 0091-6749/$36.00 Ó 2010 American Academy of Allergy, Asthma & Immunology doi:10.1016/j.jaci.2010.03.027
70
Results: The odds ratio for having greater asthma severity was 2.00 (95% CI, 1.24-3.24; P 5 .004) when comparing asthmatic patients with at least 1 copy of the serine variant with those with none. The carrier frequency for the variant increased in line with asthma treatment step (P 5 .000). The presence of the variant nearly doubled the odds in favor of asthmatic exacerbations (odds ratio, 1.90; 95% CI, 1.19-3.04; P 5 .008) over the previous 6 months. The serine variant was also associated with increased disease severity in patients with COPD (P 5 .016). Prior administration of an MMP-12–specific inhibitor attenuated the early airway response and completely blocked the late airway response with subsequent Ascaris suum challenge in sheep. Conclusion: Studies on human participants with asthma and COPD show that the risk MMP12 gene variant is associated with disease severity. In allergen-sensitized sheep pharmacologic inhibition of MMP12 downregulates both early and late airway responses in response to allergic stimuli. (J Allergy Clin Immunol 2010;126:70-6.) Key words: Asthma, allergy, chronic obstructive pulmonary disease, metalloproteinase, genetics, exacerbation, polymorphism, animal studies
There is now strong evidence across species implicating both acute allergen-responsive proinflammatory and chronic airway remodeling roles for the enzyme matrix metalloproteinase (MMP)-12 in the lung.1-4 Mice deficient in MMP-12 exhibit a marked overall reduction in their airway inflammation profile after allergen-induced lung injury.1 Infiltration with eosinophils and macrophages is critical to the airway responses to allergens and the subsequent development of airway hyperresponsiveness (AHR).4-6 MMP-12 is a critical driver of this process, regulating IL-13–induced allergic inflammation in the airways, thus controlling the accumulation of eosinophils and macrophages in response to allergen exposure.4,6 MMP-12 also plays a key role in influencing airway remodeling by degrading a wide range of extracellular matrix proteins, including elastin, type IV collagen, fibronectin, laminin, and gelatin. This activity occurs either directly or through the activation of other MMPs or downstream from IL-13.2-4 Although MMP-12 is found predominantly in macrophages, human bronchial epithelial cells also express and secrete MMP-12.7 Tricyclic and other compounds have been known to inhibit MMP-12 and other metalloproteinases, possibly through the pharmacologic roles of forming chelate complexes and acting as electron donors.8 Animal models of allergen-induced lung inflammation are well characterized9; this offers the potential for testing the role of MMP-12 inhibition in such models.
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
Abbreviations used AHR: Airway hyperresponsiveness BTS: British Thoracic Society COPD: Chronic obstructive pulmonary disease MMP: Matrix metalloproteinase RL: Mean pulmonary resistance SNP: Single nucleotide polymorphism SRL: Specific lung resistance
MMP-12 polymorphic variation in the general population appears to influence clinical phenotype in diseases in which MMP-12 could play a prominent role. Two putative functional polymorphisms have been characterized in the MMP12 gene. A common asparagine to serine substitution at codon 357 of the MMP12 gene (1082A/G, rs652438) is associated with worse survival outcomes in patients with breast cancer,10 indicating faster tumor invasion with this allele, possibly because of a greater risk of degradation of the extracellular matrix. Haplotypes containing the MMP12 Asn357Ser (rs652438) allele are associated with lung function decrease in smokers with chronic obstructive pulmonary disease (COPD).11 An A-82G polymorphism occurs at the site of binding of the transcription factor activator protein 1, the A allele being associated with a greater binding affinity for activator protein 1.10,12 This results in higher MMP12 promoter activity.10,12 The A allele is also associated with a clinically adverse outcome, being linked to smaller coronary arterial diameter in diabetic patients.10,12 These putative functional variants would be predicted to influence increased lung remodeling in patients with asthma and might be associated with asthma outcomes and response to treatment. To explore the hypothesis that MMP-12 inhibitors might have therapeutic potential in patients with asthma, we adopted a number of concurrent approaches. We explored the role of the putative functional variants on asthma clinical outcomes that are likely to be affected by acute allergen-responsive proinflammatory (inhaled b2-agonist use and risk of asthma exacerbations) and chronic remodeling (chronic medication needs) responses in children and young adults, as well as disease severity in patients with COPD. In parallel, we developed specific MMP-12 inhibitors and characterized the role of the most specific inhibitor on the regulation of airway inflammation in a well-characterized animal model of allergic airway bronchoconstriction, airway responsiveness, and allergic airway inflammation.
METHODS Population genetic studies in patients with asthma and COPD Children and young adults (3-22 years) with physician-diagnosed asthma recruited into the BREATHE study formed the group of cases in this investigation (Tables I and II). Participants attended primary or secondary clinics in 14 primary care practices and a secondary care asthma clinic in either Tayside or Dumfries. During 2004-2006, we collected mouthwash samples for DNA analysis from 1313 subjects; demographic and anthropometric data were available for 1017 of these participants. A detailed history was obtained, including information on symptoms, treatment, and exacerbations. We measured pulmonary function by means of spirometry (FEV1, forced vital capacity, and peak expiratory flow rate). The asthma prescribing level was determined in accordance with the British Thoracic Society (BTS) guidelines for physician-led management of asthma,13 with some modification as follows:
MUKHOPADHYAY ET AL 71
step 0, no use of inhaled albuterol on demand within the past month; step 1, inhaled albuterol on demand; step 2, regular inhaled steroids plus inhaled albuterol on demand; step 3, regular inhaled salmeterol plus inhaled steroids with inhaled albuterol on demand; and step 4, regular inhaled salmeterol plus inhaled steroids plus oral montelukast with inhaled albuterol on demand. The use of inhaled short-acting b-agonists (bronchodilators) was categorized as follows: 0, rarely or never required; 1, required few times a week but less than once daily; 2, required daily; and 3, multiple doses over a 24-hour period on a regular basis. An asthma exacerbation was defined as the presence of 1 or more of the following 3 indicators during the period of study: (1) asthmarelated hospital admission; (2) 1 or more days of asthma-related absence from school; and (3) prescribed course of oral steroids. All participants were of Northern European origin. Informed consent was provided by the patient and the parent/guardian, as relevant. A control population consisting of a random sample of 1451 specimens from a previous study of 2,454 schoolchildren from the Tayside area was used to assess the possibility that these MMP12 single nucleotide polymorphisms (SNPs) might be involved in susceptibility to asthma. Informed consent and saliva DNA samples were obtained from 989 patients with physician-diagnosed COPD, and clinical phenotypes were obtained from their medical records. All studies were approved by the Tayside Committee on Medical Research Ethics. DNA extraction and genotyping were performed as described in our previously reported studies in patients with asthma.14-18 DNA in the mouthwash specimens was extracted with QIAamp 96 DNA blood kits (Qiagen, Hilden, Germany). The codon 357 variant and the 282 A/G promoter variant were genotyped in both asthmatic children and young adults and population control subjects (see Tables E1 and E2 in this article’s Online Repository at www.jacionline.org). The allele frequency of both variants was similar in the patients with asthma (serine variant, 0.048; 95% CI, 0.039-0.058; n 5 997) and population control subjects (serine variant, 0.051; 95% CI, 0.0430.059; n 5 1431), demonstrating that they are not associated with an increased susceptibility to asthma per se (see Tables E1 and E2). This is consistent with the prior observations regarding the serine variant and lung cancer susceptibility. We therefore examined the association between these variants and the clinical phenotype within the asthmatic population. We have previously extensively validated and tested the above measures of asthma severity and control in this population of young asthmatic patients and have demonstrated associations between other common gene polymorphisms and clinical phenotype in allergy and asthma.14-18 Because the serine variant occurs at a low frequency, subjects carrying 1 or 2 copies of the serine variant were analyzed as one group. Allelic discrimination assays for the 282A/G (rs2276109) and 1082A/G (rs652438) SNPs in the MMP12 gene were carried out using in-house assays in the asthmatic population (details available on request) and Applied Biosystems–catalogued assays in the control population with a 7700 Sequence Detection System (Applied Biosystems, Foster City, Calif). Results obtained with the 2 assay systems were 100% concordant. The genotype distribution of both MMP12 SNPs was in Hardy-Weinberg equilibrium (P > .05), and genotyping call rates ranged from 96% to 97%. Statistical analysis was carried out with SPSS for Windows version 15 software (SPSS, Inc, Chicago, Ill). Differences in the genotype distributions between the groups of cases and control subjects were evaluated by using the x2 test. Logistic regression analysis, controlling for age and exposure to cigarette smoke, was used to determine odds ratios for the occurrence of any exacerbation. Sex was considered as a covariate but was subsequently removed from all models because it was not a significant covariate for exacerbations or severity in the BREATHE study.
Laboratory studies Development of a specific MMP-12 inhibitor. We subsequently proceeded to screen and identify potential candidate compounds with specific inhibitory properties for MMP-12. A unique class of compounds featured with a dibenzofuran core and a zinc-chelating group were prepared, modified, and profiled for their drug-like properties. Compounds were selected for their potency in both enzymatic and cell-based assays, as well as their selectivity against a panel of 8 other closely related MMPs, as well as
72 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE III. Half-maximal inhibitory concentrations for MMP40820 on MMP-12 and other related enzymes
TABLE I. Summary characteristics of 1,017 asthmatic patients and 1451 population control subjects
Sex, male/female (n) Age (y)* Body mass index (kg/m2)*
Asthmatic patients
Control subjects
608/409 10.43 (7.13-13.12) 19.22 (16.09-21.11)
724/727 7.42 (6.51-8.45) 17.11 (15.66-18.09)
*Mean (25th-75th percentile).
TABLE II. Clinical characteristics of asthmatic patients Lung function
Peak expiratory flow rate (% predicted) FEV1 (% predicted) Forced vital capacity (% predicted) BTS step of treatment (n 5 1001) 0 1 2 3 4 Asthma exacerbation in preceding 6 mo (n 5 971) Any hospital admission Any school absence One or more courses of oral steroids Any exacerbation
Mean (25th-75th percentile)
88.0 95.8 92.3 No. 30 176 569 133 93 No.
(77.0-98.4) (85.8-105.7) (83.2-101.1) (%) (3.0) (17.6) (56.8) (13.3) (9.3) (%)
113 301 205 352
(11.2) (30.8) (20.4) (36.2)
No.
811 814 813
TNF-a converting enzyme and aggrecanase 1 and 2. The procedure followed for the development of the active tricyclic compound has already been described in detail.19,20 One highly potent and specific inhibitor selective for MMP-12, (S)-2-(8 (methoxycarbonylamino)dibenzo[b,d]furan-3-sulfonamido)-3-methylbutanoic acid (MMP408; Table III),19,20 was taken forward for testing in an Ascaris suum–sensitized sheep model of allergic airway disease.9 Test and control articles. A suum extract (Greer Diagnostics, Lenoir, NC) was diluted with PBS to a concentration of 82,000 protein nitrogen units/mL and delivered as an aerosol (20 breaths/min 3 20 minutes). This crude preparation has an endotoxin level of 50 EU/mL, an amount that does not have an effect on pulmonary responses in sheep. Carbamylcholine (Carbachol; Sigma Chemical Co, St Louis, Mo) was dissolved in buffered saline at concentrations of 0.25, 0.50, 1.0, 2.0, and 4.0% wt/vol and delivered as an aerosol. Animal preparation. A well-validated animal model of allergic airway disease was used for these studies.9 Sheep weighing 45 to 55 kg that were naturally sensitized to the nematode A suum were used. All sheep had previously been shown to develop both early and late bronchial responses to inhaled A suum antigen. The sheep were conscious and restrained in a modified shopping cart in the prone position with their heads immobilized. After topical anesthesia of the nasal passages with 2% lidocaine, a balloon catheter was advanced through one nostril into the lower esophagus. The animals were intubated with a cuffed endotracheal tube through the other nostril with a flexible fiberoptic bronchoscope as a guide. Airway mechanics. Breath-by breath determination of mean pulmonary resistance (RL) was measured with the esophageal balloon technique. Pleural pressure was estimated with an esophageal balloon catheter (filled with 1 mL of air), which was positioned 5 to 10 cm from the gastroesophageal junction. With the catheter in this position, the end-expiratory pleural pressure ranged from 22 to 25 cm H2O. Once the balloon was placed, it was secured so that it remained in the same position for the duration of the experiment. Lateral pressure in the trachea was measured with a sidehole catheter (internal diameter, 2.5 mm) advanced through and
Enzyme
Human MMP-12 Monkey MMP-12 Dog MMP-12 Rabbit MMP-12 Sheep MMP-12 Mouse MMP-12 Rat MMP-12 Human MMP-1 Human MMP-2 Human MMP-3 Human MMP-7 Human MMP-8 Human MMP-9 Human MMP-13 Human MMP-14 TNF-a converting enzyme Aggrecanase 1 Aggrecanase 2
IC50 (nmol/L)
2 1.5 3.4 6.7 22.3 160 320 >5000 61 351 >5000 41 1300 120 1100 >5000 >5000 >5000
IC50, Half-maximal inhibitory concentration.
positioned distal to the tip of the endotracheal tube. The tracheal and pleural-pressure catheters were connected to a differential pressure transducer (MP45; Validyne, Northridge, Calif) for the measurement of transpulmonary pressure, which was defined as the difference between tracheal and pleural pressure. Airflow was measured by connecting the proximal end of the endotracheal tube to a pneumotachograph (Fleisch No. 1; Dyna Sciences, Inc, Blue Bell, Pa). The transpulmonary pressure and flow signals were recorded on a multichannel physiologic recorder, which was linked to an 80386 DOS Personal Computer (CCI, Inc, Miami, Fla) for online calculation of RL by dividing the change in transpulmonary pressure by the change in flow at midtidal volume (obtained by means of digital integration). The mean of at least 5 breaths free of swallowing artifact was used to obtain RL in centimeters of H2O per liter per second. Immediately after the measurement of RL, thoracic gas volume was measured in a constant-volume body plethysmograph to obtain specific lung resistance (SRL) as follows: SRL 5 RL 3 Thoracic gas volume. Aerosol delivery systems. All aerosols were generated with a disposable medical nebulizer (Raindrop; Puritan Bennett, Lenexa, Kan) that provided an aerosol with a mass median aerodynamic diameter of 3.2 mm, as determined with an Andersen cascade impactor. The nebulizer was connected to a dosimeter system consisting of a solenoid valve and a source of compressed air (20 psi). The output of the nebulizer was directed into a plastic T-piece, one end of which was connected to the inspiratory port of a piston respirator (Harvard Apparatus, Mills, Mass). The solenoid valve was activated for 1 second at the beginning of the inspiratory cycle of the respirator. Aerosols were delivered at a tidal volume of 500 mL and a rate of 20 breaths/min.
Concentration response curves to carbachol aerosol. To assess bronchial responsiveness, cumulative concentration-response curves to carbachol were generated by means of SRL immediately after inhalation of buffer and after each consecutive administration of 10 breaths of increasing concentrations of carbachol (0.25%, 0.5%, 1.0%, 2.0%, and 4.0% [wt/vol] in buffered saline). The provocation test was discontinued when SRL increased by more than 400% from the postsaline value or after the highest carbachol concentration had been administered. Bronchial responsiveness was assessed by determining the cumulative carbachol concentration (in breath units) that increased SRL by 400% over the postsaline value (PC400) through interpolation from the dose-response curve. One breath unit was defined as one breath of a 1% (wt/vol) carbachol aerosol solution.
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
MUKHOPADHYAY ET AL 73
FIG 1. A-C, The MMP12 serine 357 variant is associated with clinically worse asthma with relation to increasing inhaled b2-agonist use (Fig 1, A), asthma treatment step (Fig 1, B), and risk of asthma exacerbations over 6 months (Fig 1, C; P for x2 analysis trend). D, Increasing medication use positively correlates with exacerbation risk, implying that current treatment regimens are inadequate overall for effective asthma control.
Evaluation of the efficacy of an MMP-12–specific inhibitor in a sheep model of allergic airway disease. Compounds were dosed either intravenously or through the oral route twice daily the day before challenge and then the following day (day of A suum challenge) 1 hour before challenge and 8 hours after challenge. Increases in airway resistance were measured throughout the day to capture both the early-phase bronchoconstrictor response that peaked approximately 1 hour after allergen challenge and the late-phase bronchoconstrictor response that appeared approximately 5 to 6 hours after allergen challenge. AHR to aerosolized carbachol was measured the following day. After AHR measurements, lungs underwent lavage, and total cell counts were quantified in the bronchoalveolar lavage fluid.
RESULTS Association of gene variant with clinical asthma phenotype Participants with the serine variant showed greater airway lability, which was measured as greater frequency of use of inhaled short-acting b2-agonists (x2 5 7.6 and P 5 .031, test for trend; Fig 1, A, and see Table E3 in this article’s Online Repository at www.jacionline.org). More dramatically, however, when all asthma medication was considered, the carrier frequency for the serine variant also increased in line with asthma treatment step as per the modified BTS guidelines (x2 5 12.7 and P 5 .0004, test for trend; Fig 1, B).13 By using binary logistic regression, the odds ratio for having greater severity of asthma, requiring the regular preventive use of long-acting b-agonists (BTS step 3 and greater), was 2.00 (95% CI, 1.24-3.24; P 5 .004) when comparing those patients carrying the serine variant with those without the serine variant. We also performed a logistic regression analysis of MMP12 genotype on exacerbations of asthma over a period of 6 months. The serine variant was associated with a trend in the increase of frequency of any school absence and oral steroid
use. Overall, the presence of the serine variant nearly doubled the risk of an asthma exacerbation (odds ratio, 1.90; 95% CI, 1.19-3.04; P 5 .008), with a significant increase in the frequency of the serine variant in participants with exacerbations in comparison with those without exacerbations over the 6-month period of study (Fig 1, C). This increase in exacerbation rate in the carriers of the serine variant is seen despite the increased medication being applied to these patients, further highlighting the importance of this variant in overall asthma severity in these children. Indeed, exacerbation rates are highly colinear with the increased medication regimen, which both confirms the use of the treatment step as a measure of severity and reveals a degree of inadequacy in current medications in the management of childhood asthma (Fig 1, D). The association between the MMP12 variant and both exacerbations was partly, although not entirely, due to this collinearity. The serine variant was significantly associated with increased exacerbations within the less severe group not prescribed long-acting b-agonists (BTS step 2 or below; odds ratio, 2.04; 95% CI, 1.2-3.5; P 5 .009). In contrast, no associations were seen between the 282 A/G promoter variant and any of the clinical parameters when analyzed as a single marker or in combination with the serine variant (see Tables E4-E6 in this article’s Online Repository at www.jacionline.org). Pulmonary function was measured reproducibly in 815 patients. Our initial analysis was hypothesis driven, exploring the role of the 2 specific potentially at-risk variants in the MMP12 gene. However, we have also subsequently extensively genotyped the MMP12 locus using a haplotype-tagging approach to determine the specificity of our observation with the serine 357 variant. Nine additional SNPS were typed across the MMP12 locus and 10 kb of flanking sequence in either direction that tagged greater than 90% of the genetic variation represented by HapMap version 2.2 (see Table E7 and Fig E1 in this article’s Online Repository at
74 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
www.jacionline.org). None of these additional variants had any effect on our 2 principal clinical outcomes: asthma treatment step (see Table E8 in this article’s Online Repository at www. jacionline.org) and exacerbation risk (see Table E9 in this article’s Online Repository at www.jacionline.org). Genotyping of published functional SNPs from the adjacent MMP genes MMP1, MMP3, MMP10, and MMP13 did not reveal any additional signal (see Tables E10 and E11 in this article’s Online Repository at _5 years) is www.jacionline.org). Asthma in early childhood (< thought to differ from the disease in older children, but stratification of the analysis on exacerbations or treatment step did not affect the conclusions (see Table E12 in this article’s Online Repository at www.jacionline.org).
Replication of gene variant association findings in COPD severity We sought to replicate our principal finding of an effect of the rs652438 variant on disease severity in a large cohort of patients with COPD in Tayside (see Table E13 in this article’s Online Repository at www.jacionline.org) for whom DNA has recently become available. By using the x2 test (Pearson linear-by-linear method), a significant relationship was observed between the presence of the at-risk variant and disease severity as defined by FEV1 (normalized as percent predicted). In this analysis we observed a relationship between the frequency of the serine variant and decreased lung function (P 5 .016, see Table E14 in this article’s Online Repository at www.jacionline.org). The 282 promoter variant rs2276109 was not associated with lung function in this population (see Table E15 in this article’s Online Repository at www.jacionline.org).
Study of the role of MMP-12 inhibitor in sheep When animals were treated intravenously with the MMP-12 inhibitor MMP408 at 10 mg/kg, there was a complete blockade of the influx of inflammatory cells into the bronchoalveolar cavity in comparison with control levels seen in untreated A suum–challenged sheep (Fig 2, A). To determine its effects on airway resistance, MMP408 at 10 mg/kg was administered intravenously at 10 mg/kg twice daily, the day before A suum antigen challenge, and again 1 hour before challenge. MMP408 significantly attenuated the early-phase allergic airway bronchoconstriction and completely blocked the late-phase allergic airway bronchoconstriction as measured by an increase in RL compared with responses observed in the absence of the inhibitor (Fig 2, B). Oral administration of MMP408 at 10 mg/kg had comparable effects on the late-phase allergic airway bronchoconstriction as observed with intravenous administration (Fig 2, C). The animals received a fourth dose of MMP408 eight hours after A suum challenge to determine the efficacy of the blockade = FIG 2. MMP-12 inhibition by MMP408 reduced total bronchoalveolar lavage inflammation (A), airway resistance (0-8 hours; B and C), and carbachol responsiveness (24 hours; D and E) after A suum antigen challenge in sheep (MMP408 dose: Fig 2, A, B, and D, 10 mg/kg administered intravenously; Fig 2, C and E, 10 mg/kg administered orally). Graphs represent group means 6 SEMs. *P < .05, 5 animals per group, Student paired t test versus test control subjects.
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
on carbachol-induced AHR. AHR to aerosolized carbachol that was measured the following day was inhibited with this dosing regimen after both intravenous and oral administration of MMP408 (Fig 2, D and E).
DISCUSSION This article describes a consistent role for the Asn357Ser variant of MMP12 in the disease pathway of both asthma and COPD; however, the study has several potential limitations. Multiple testing is always an issue with gene association studies, and in this case we tested 2 previously studied SNPs providing a priori evidence for their involvement in disease and then tested for non–hypothesis-driven haplotype-tagging screening of the chromosomal region to test the specificity of the results for these putative functional variants. Correction of the results for the multiple testing of the 2 prespecified SNPs does not abolish the significance of the results for the Asn357Ser variant. Bonferroni correction across the screening SNPs abolished all significant observations, confirming the specificity of the observations for the Asn357Ser variant. Multiple testing across the phenotypes was not applied because of the interdependence of the phenotypes and the fact that the Asn357Ser variant was significantly associated with the majority of the available outcome measures. Another consideration in gene association studies is population stratification. This was a highly homogenous population from a single region of Scotland, and there was no difference in the allele frequencies between the asthmatic and control groups, suggesting that the current observations are highly unlikely to be the result of some inherent population substructure. The data presented in this study do not prove that the Asn357Ser variant is the causal variant in this process, and recent studies have suggested a role for the 282 promoter variant in asthma and COPD susceptibility,21 a concept that is not supported in our current data. This raises the possibility that the genetic architecture of the MMP12 locus might be complex and differ both between populations, disease context, or both. It also raises the possibility that a causal variant is yet to be found and that the differential signals are due to population differences in haplotype diversity. Our data do display a trend toward a protective role for the rs2276109 (282) genotype, as recently suggested,21 and therefore it is also possible that given a larger sample size, we might be able to detect an effect with this variant. The current treatment of asthma rests primarily on the management of overall inflammation through the use of inhaled steroids, smooth muscle relaxation by b2-agonists, and, more recently, specific inflammatory mediator–blocking therapy (montelukast) that is effective in a proportion of patients.13 This work suggests that MMP-12 could represent a more specific inflammatory molecule in human subjects, regulating pathways for acute airway changes associated with allergic airway inflammation and chronic airway remodeling, which could contribute to a more severe clinical asthma phenotype, including a greater risk of exacerbation. There is much recent evidence that genetic variation, possibly interacting with environmental factors, regulates susceptibility to and the severity of childhood diseases associated with atopy. A more complete understanding of these influences might create a composite picture of these interactions, thus providing novel insights into the natural history of childhood diseases associated with atopy and leading to the development of new therapeutic
MUKHOPADHYAY ET AL 75
targets and strategies. The poorly functioning epidermal protein filaggrin, for example, occurs in about 8% of white subjects from the United Kingdom and has recently been shown to contribute to overall increased susceptibility to atopic diseases triggering asthma in childhood.14,16,17 In contrast, we have shown that MMP12 gene variations regulate the clinical asthma phenotype but do not affect the susceptibility to asthma in this current population (see Table E1). These effects at the genetic and clinical levels now require mapping onto the natural history of childhood atopic disease through functional studies to provide an improved understanding of atopic disease in childhood. It might be hypothesized that unless processes such as epithelial filaggrin protein malfunction let in a critical allergen dose, the activation of the immune cascade that sets off airway inflammation does not occur. Differences in MMP-12 function in the airway might become clinically relevant only when such airway inflammation sets in. These complex gene-environment interactions can, however, only be studied through the prospective analysis of much larger datasets of genotyped and appropriately phenotyped subjects. The inflammation-related and regenerative processes within COPD are closely related to those within asthma, and we have referred above to how our work was primarily driven by the previous observation that haplotypes that contain the MMP12 Asn357Ser (rs652438) allele might contribute to lung function decrease in smokers with COPD.11 We have now genotyped a large cohort of patients with COPD in Tayside about whom detailed longitudinal clinical data are being collected over a number of years. We used this resource to explore the possible role of the rs652438 allele on disease severity in patients with COPD (see Table E13). We observed a significant relationship between the presence of this at-risk gene variant and the last measurement of FEV1 (normalized as percent predicted, see Table E14). We believe these findings from a related chronic respiratory disease in a different age group of patients reinforce the role of this at-risk gene variant in airway inflammation. We have not observed associations with pulmonary function in our asthma cohort because, unlike in participants with COPD, pulmonary function (normalized FEV1) is normal in young asthmatic patients in the United Kingdom during the quiescent stage (Table II), which is not the case in asthmatic populations in all countries. This difference in lung function preservation between populations might contribute to the differences seen between the effects of different MMP12 SNPs on asthma in different populations. MMP12 might influence asthma from a number of different aspects. It is thus possible that the 282 promoter variant might have effects on the asthma phenotype that are different from the serine 357 variant. These effects could be regulated through differences in the mechanism of action between the 2 genetic variants, and these questions require further study. Our concurrent development of an MMP-12–specific inhibitor19,20 that blocks allergen-induced airway disease in sheep offers direct hope of the therapeutic targeting of MMP-12 in asthmatic patients. The stage is thus set for early exploratory safety and efficacy trials of specific MMP-12 inhibition in young asthmatic patients. Efficacy outcomes of interest, on the basis of our results and knowledge of MMP-12 function, might include early and late airway responses and carbachol sensitization, methacholine sensitization, or both (over days or weeks) and airway remodeling assessment over a longer period of time (ie, months). Airway remodeling is established during childhood22,23 and acts as a critical determinant for asthma severity throughout
76 MUKHOPADHYAY ET AL
life.24 Thus we speculate that the inhibition of MMP12 might require investigation as a possible therapeutic strategy for asthma and COPD. We acknowledge support from the following colleagues: Rustem Krykbaev (Pfizer Research, Inflammation and Immunology) for cloning of sheep MMP12; Paul Morgan (Pfizer Research, Inflammation and Immunology) and Kristina Cunningham (Pfizer Research, Chemical and Screening Sciences) for in vitro compound specificity assays; Jianchang Li and Yuchan Wu (Pfizer Research, Chemical and Screening Sciences) for compound synthesis and scaleup; Vicky Alexander, Tahmina Ismail, and Inez Murrie (University of Dundee Medical School) for clinical data collection and entry and help with the development of the asthma database in Scotland; and David Crook (Royal Sussex County Hospital) for useful comments on the manuscript.
Clinical implications: The specific inhibition of MMP-12– driven airway inflammation and airway remodeling might represent a new therapeutic strategy for asthma and COPD, justifying exploratory human safety and efficacy trials.
REFERENCES 1. Warner RL, Lukacs NW, Shapiro SD, Bhagarvathula N, Neruso KC, Varani J, et al. Role of metalloelastase in a model of allergic lung responses induced by cockroach allergen. Am J Pathol 2004;165:1921-30. 2. Chandler S, Cossins J, Lury J, Wells G. Macrophage metalloelastase degrades matrix and myelin proteins and processes a tumour necrosis factor-alpha fusion protein. Biochem Biophys Res Commun 1996;228:421-9. 3. Gronski TJ Jr, Martin RL, Kobayashi DK, Walsh BC, Holman MC, Huber M, et al. Hydrolysis of a broad spectrum of extracellular matrix proteins by human macrophage elastase. J Biol Chem 1997;272:12189-94. 4. Lanone S, Zheng T, Zhu Z, Lee CG, Ma B, Chen Q, et al. Overlapping and enzyme-specific contributions of matrix metalloproteinases-9 and -12 in IL-13– induced inflammation and remodelling. J Clin Invest 2002;110:463-74. 5. Agrawal DK, Townley RG. Inflammatory cells and mediators in bronchial asthma. In: Hollinger MA, editor. CRC press series in pharmacology and toxicology. Boca Raton (FL): CRC Press; 1990. 6. Pouladi MA, Robbins CS, Swirski FK, Cundall M, McKenzie AN, Jordana M, et al. Interleukin-13-dependent expression of matrix metalloproteinase-12 is required for the development of airway eosinophilia in mice. Am J Respir Cell Mol Biol 2004;30:84-90. 7. Lavigne MC, Thakker P, Gunn J, Wong A, Miyashiro JS, Wasserman AM, et al. Human bronchial epithelial cells express and secrete MMP-12. Biochem Biophys Res Commun 2004;324:534-46. 8. Molnar J, Sohar I, Kovacs J, Rakonczay Z, Rausch H. Effects of tricyclic compounds on membrane binding of bivalent cations, activities of acetylcholinesterase and some tissue proteases. In Vivo 1993;7:431-4.
J ALLERGY CLIN IMMUNOL JULY 2010
9. Abraham WM. Modeling of asthma, COPD and cystic fibrosis in sheep. Pulm Pharmacol Ther 2008;21:743-54. 10. Shin A, Cai Q, Shu X, Gao YT, Zheng W. Genetic polymorphisms in the matrix metalloproteinase 12 gene (MMP12) and breast cancer risk and survival: the Shanghai Breast Cancer Study. Breast Cancer Research 2005;7:R506-12. 11. Joos L, He JQ, Shepherdson MB, Connett JE, Anthonisen NR, Pare´ PD, et al. The role of matrix metalloproteinase polymorphisms in the rate of decline in lung function. Hum Mol Genet 2002;11:569-76. 12. Jormsjo S, Ye S, Moritz J, Walter DH, Dimmeler S, Zeiher AM, et al. Allelespecific regulation of matrix metalloproteinase-12 gene activity is associated with coronary artery luminal dimensions in diabetic patients with manifest coronary artery disease. Circ Res 2000;86:998-1003. 13. BTS/SIGN. British guidelines on the management of asthma. Thorax 2003; 58(suppl 1):1-94. 14. Palmer CN, Irvine AD, Terron-Kwiatkowski A, Zhao Y, Liao H, Lee SP, et al. Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis. Nat Genetics 2006;38: 441-6. 15. Palmer CN, Doney AS, Ismail T, Lee SP, Murrie I, Macgregor DF, et al. PPARG locus haplotype variation and exacerbations in asthma. Clin Pharmacol Ther 2007; 81:713-8. 16. Palmer CN, Ismail T, Lee SP, Terron-Kwiatkowski A, Zhao Y, Liao H, et al. Filaggrin null mutations are associated with increased asthma severity in children and young adults. J Allergy Clin Immunol 2007;120:64-8. 17. Basu K, Palmer CN, Lipworth BJ, Irwin McLean WH, Terron-Kwiatkowski A, Zhao Y, et al. Filaggrin null mutations are associated with increased asthma exacerbations in children and young adults. Allergy 2008;63:1211-7. 18. Tavendale R, Macgregor DF, Mukhopadhyay S, Palmer CNA. A polymorphism controlling ORMDL3 expression is associated with asthma that is poorly controlled by current medications. J Allergy Clin Immunol 2008;121:860-3. 19. Li W, Li J, Wu Y, Tam SY, Mansour T, Sypek JP, et al. Tricyclic compounds as matrix metalloproteinase inhibitors. WO 2008/057254 A2; international publication date, May 15, 2008. Available at: http://www.wipo.int/pctdb/en/wo. jsp?WO52008057254&IA5US2007022651&DISPLAY5DESC). 20. Li W, Li J, Wu Y, Hotchandani R, Cunningham K, McFadyen I, et al. A selective matrix metalloprotease 12 inhibitor for potential treatment of chronic obstructive pulmonary disease (COPD): discovery of (S)-2-(8 (methoxycarbonylamino)dibenzo[b, d]furan-3-sulfonamido)-3-methylbutanoic acid (MMP408). J Med Chem 2009;52:1799-802. 21. Hunninghake GM, Cho MH, Tesfaigzi Y, Soto-Quiros ME, Avila L, Lasky-Su J, et al. MMP12, lung function, and COPD in high-risk populations. N Engl J Med 2009;361:2599-608. 22. De Blic J, Tillie-Leblond I, Emond S, Mahut B, Dang Duy TL, Scheinmann P. High-resolution computed tomography scan and airway remodeling in children with severe asthma. J Allergy Clin Immunol 2005;116:750-4. 23. Saglani S, Payne DN, Zhu J, Wang Z, Nicholson AG, Bush A, et al. Early detection of airway wall remodeling and eosinophilic inflammation in preschool wheezers. Am J Respir Crit Care Med 2007;176:858-64. 24. James AL, Wenzel S. Clinical relevance of airway remodelling in airway diseases. Eur Respir J 2007;30:134-55.
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
MUKHOPADHYAY ET AL 76.e1
FIG E1. Linkage disequilibrium relationship of haplotype-tagging SNPs across the MMP12 locus. Visualized with Haploview V4.1: Hap Map CEU Phase 2 Release 21. CEU genotypes. The R2 values are shown from the CEU dataset.
76.e2 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E1. Genotype distribution in asthmatic cases and population control subjects Genotype/n SNP
282A/G Asthmatic cases Control subjects Asn357Ser Asthmatic cases Control subjects
AA
AG
GG
n
HWE
MAF
P value
780 1093
216 295
14 27
1010 1415
0.827 0.178
0.121 0.123
.791
905 1285
88 132
4 7
997 1424
0.243 0.076
0.048 0.051
.624
HWE, Hardy-Weinberg equilibrium; MAF, minor allele frequency.
MUKHOPADHYAY ET AL 76.e3
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E2. Two-marker genotype distribution in asthmatic cases Asn357Ser 282A/G
AA AG GG Total
Asn/Asn AA
Asn/Ser AG
Ser/Ser GG
Total
685 200 13 898
75 13 0 88
4 0 0 4
764 213 13 990
282G and serine 357 are resident on different ancestral haplotypes and thus rarely occur as the joint carrier genotype.
76.e4 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E3. Contingency table for MMP12 genotype versus inhaled b2-agonist use on a scale of 0 to 3 (see the Methods section for details) Asn357Ser SNP
Inhaled b2-agonist use 0 1 2 3
Asn/Asn
100 623 142 19
Serine/X
(96.2%) (90.4%) (90.4%) (79.2%)
4 66 15 5
(3.8%) (9.6%) (9.6%) (20.8%)
Total
P value
104 689 157 24
.0305
Total
P value
282A/G SNP
Inhaled b2-agonist use 0 1 2 3
AA
81 540 121 21
(76.4%) (77.6%) (75.6%) (84.0%)
P value relates to trend on x2 analysis by means of linear-by-linear association.
AG/GG
25 156 39 4
(23.6%) (22.4%) (24.4%) (16.0%)
106 696 160 25
.839
MUKHOPADHYAY ET AL 76.e5
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E4. MMP12 genotype and drug treatment Genotype/no. (%) MMP12 282A/G
BTS drug class 0 1 2 3 4 P value (x2)
MMP12 Asn357Ser
AA
AG1GG
18 (2.3) 137 (17.9) 438 (57.2) 105 (13.7) 68 (8.9) Test for trend (1 df)
12 (5.3) 35 (15.4) 129 (56.6) 28 (12.3) 24 (10.5) .832
AA
30 164 508 111 77
(3.4) (18.4) (57.1) (12.5) (8.7)
AG1GG
0 (0.0) 10 (11.0) 49 (53.8) 18 (19.8) 14 (15.4) .000374
76.e6 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E5. MMP12 282A/G genotype and risk of specific asthma exacerbations Clinical outcome
Hospital admission School absence Oral steroid use Any
Odds ratio
95% CI
P value
0.710 0.968 0.796 0.915
0.419-1.203 0.692-1.353 0.536-1.183 0.660-1.267
.203 .847 .259 .591
MUKHOPADHYAY ET AL 76.e7
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E6. MMP12 codon 357 genotype and risk of specific asthma exacerbations Clinical outcome
Hospital admission School absence Oral steroid use Any
Odds ratio
95% CI
P value
0.734 1.535 1.416 1.898
0.324-1.660 0.948-2.485 0.833-2.406 1.185-3.040
.457 .0812 .199 .00768
76.e8 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E7. Linkage disequilibrium relationships between pairs of SNPs in the population of children with asthma rs673163
1.0
rs586701
rs580266
rs12808148
rs476185
rs652438
rs10895367
rs2276109
rs17368814
rs11225445
0 1.0
0.48 0 1.0
0.47 0 0 1.0
0 0 0 0 1.0
0 0.48 0 0 0 1.0
0 0 0.53 0 0 0 1.0
0 0.79 0 0 0 0 0 1.0
0 0.80 0 0 0 0 0 1.0 1.0
0.87 0 0.36 0.52 0 0 0 0 0 1.0
Shown is the R2 value between the markers for this population.
rs673163 rs586701 rs580266 rs12808148 rs476185 rs652438 rs10895367 rs2276109 rs17368814 rs11225445
MUKHOPADHYAY ET AL 76.e9
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E8. MMP12 haplotype-tagging SNPs and drug class: Linear regression P values with age and exposure to cigarette smoke as covariates MMP12 SNP
P value
rs673163 rs586701 rs580266 rs12808148 rs476185 rs652438 rs10895367 rs2276109 rs17368814 rs11225445
.955 .338 .562 .781 .853 .000374 .591 .664 .293 .694
76.e10 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E9. MMP12 haplotype-tagging SNPs and risk of any asthma exacerbation: Binary logistic regression P values with age and exposure to cigarette smoke as covariates MMP12 SNP
P value
rs673163 rs586701 rs580266 rs12808148 rs476185 rs652438 rs10895367 rs2276109 rs17368814 rs11225445
.205 .410 .658 .509 .346 .00768 .829 .591 .447 .210
MUKHOPADHYAY ET AL 76.e11
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E10. MMP gene cluster functional polymorphisms and drug treatment class Reference
rs1799750 rs679620 rs17293607 rs2276109 rs652438 rs2252070
Gene
P value*
MMP1 MMP3 MMP10 MMP12 282A/G MMP12 Asn357Ser MMP13
.991 .516 .932 .832 .000374 .932
*P value (x2), test for trend (1 df).
76.e12 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E11. MMP gene cluster functional polymorphisms and asthma exacerbations Reference
rs1799750 rs679620 rs17293607 rs2276109 rs652438 rs2252070
Gene
Odds ratio
95% CI
P value
MMP1 MMP3 MMP10 MMP12 282A/G MMP12 Asn357Ser MMP13
1.085 0.993 0.954 0.915 1.898 0.955
0.897-1.312 0.818-1.205 0.725-1.255 0.660-1.267 1.185-3.040 0.727-1.256
.400 .942 .736 .591 .00768 .743
MUKHOPADHYAY ET AL 76.e13
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E12. Associations with MMP12 codon 357 variant and exacerbations and drug class stratified by age (<6 and >6 years of age)
Exacerbation <6 y >6 y Drug class (0-2 vs 3-4) <6 y >6 y
Odds ratio
95% CI
P value
7.50 1.71
0.859-65.08 0.818-1.205
.068 .033
1.96 2.02
0.459-8.386 1.215-3.359
.363 .007
Binary logistic regression adjusted for age and exposure to smoke.
76.e14 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E13. Summary characteristics of 989 patients with COPD in Tayside, Scotland Variable
Mean (SD)
Range
No.
Age at diagnosis (y) Follow-up (y) FEV1 at last visit (% of mean predicted value) Smoking pack-years Body mass index (kg/m2)
63.9 (9.4) 4.01 (1.87) 63.0 (20.1)
34.6-88.5 0-8.00 10.3-131.9
989 989 989
41.8 (20.5) 26.4 (5.46)
0-150 15.4-54.2
925 970
MUKHOPADHYAY ET AL 76.e15
J ALLERGY CLIN IMMUNOL VOLUME 126, NUMBER 1
TABLE E14. rs652438 genotype and most recent measurement of FEV1 in a longitudinal cohort of patients with COPD in Tayside (n5 986) Percent predicted FEV1/n rs652438 genotype
>80% FEV1
50% to 80% FEV1
30% to 50% FEV1
<30% FEV1
AA AG1GG
194 (94.6) 11 (5.4)
458 (88.9) 57 (11.1)
194 (89.8) 22 (10.2)
41 (82.0) 9 (18.0)
Three subjects did not genotype successfully for rs652438. *Linear-by-linear association: P 5 .013 after adjustment for age, sex, duration of disease, and smoking pack-years.
P value (x2)*
.0161
76.e16 MUKHOPADHYAY ET AL
J ALLERGY CLIN IMMUNOL JULY 2010
TABLE E15. rs2276109 (282) genotype and most recent measurement of FEV1 in a longitudinal cohort of patients with COPD in Tayside (n 5 982) Percent predicted FEV1/n rs2276109 genotype
>80% FEV1
50% to 80% FEV1
30% to 50% FEV1
<30% FEV1
AA AG1GG
164 (79.6) 42 (20.4)
404 (78.8) 109 (21.2)
160 (74.4) 55 (25.6)
41 (85.4) 7 (14.6)
Four subjects who were genotyped for rs652438 were not successfully genotyped for rs2276109. *Linear-by-linear association: P 5 .251 after adjustment for age, sex, duration of disease, and smoking pack-years.
P value (x2)*
.676