UBC Faculty Research and Publications

Sulfatase modifying factor 1 (SUMF1) is associated with Chronic Obstructive Pulmonary Disease Weidner, Julie; Jarenbäck, Linnea; de Jong, Kim; Vonk, Judith M; van den Berge, Maarten; Brandsma, Corry-Anke; Boezen, H. M; Sin, Don; Bossé, Yohan; Nickle, David; Ankerst, Jaro; Bjermer, Leif; Postma, Dirkje S; Faiz, Alen; Tufvesson, Ellen May 2, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12931_2017_Article_562.pdf [ 1.12MB ]
Metadata
JSON: 52383-1.0366295.json
JSON-LD: 52383-1.0366295-ld.json
RDF/XML (Pretty): 52383-1.0366295-rdf.xml
RDF/JSON: 52383-1.0366295-rdf.json
Turtle: 52383-1.0366295-turtle.txt
N-Triples: 52383-1.0366295-rdf-ntriples.txt
Original Record: 52383-1.0366295-source.json
Full Text
52383-1.0366295-fulltext.txt
Citation
52383-1.0366295.ris

Full Text

RESEARCH Open AccessSulfatase modifying factor 1 (SUMF1) isassociated with Chronic ObstructivePulmonary DiseaseJulie Weidner1, Linnea Jarenbäck1, Kim de Jong2, Judith M. Vonk2, Maarten van den Berge3, Corry-Anke Brandsma3,H. Marike Boezen2, Don Sin4, Yohan Bossé5, David Nickle6, Jaro Ankerst1, Leif Bjermer1, Dirkje S. Postma3, Alen Faiz3and Ellen Tufvesson1*AbstractBackground: It has been observed that mice lacking the sulfatase modifying factor (Sumf1) developed anemphysema-like phenotype. However, it is unknown if SUMF1 may play a role in Chronic Obstructive PulmonaryDisease (COPD) in humans. The aim was to investigate if the expression and genetic regulation of SUMF1 differsbetween smokers with and without COPD.Methods: SUMF1 mRNA was investigated in sputum cells and whole blood from controls and COPD patients (allcurrent or former smokers). Expression quantitative trait loci (eQTL) analysis was used to investigate if singlenucleotide polymorphisms (SNPs) in SUMF1 were significantly associated with SUMF1 expression. The association ofSUMF1 SNPs with COPD was examined in a population based cohort, Lifelines. SUMF1 mRNA from sputum cells,lung tissue, and lung fibroblasts, as well as lung function parameters, were investigated in relation to genotype.Results: Certain splice variants of SUMF1 showed a relatively high expression in lung tissue compared to many othertissues. SUMF1 Splice variant 2 and 3 showed lower levels in sputum cells from COPD patients as compared to controls.Twelve SNPs were found significant by eQTL analysis and overlapped with the array used for genotyping of Lifelines.We found alterations in mRNA expression in sputum cells and lung fibroblasts associated with SNP rs11915920 (top hitin eQTL), which validated the results of the lung tissue eQTL analysis. Of the twelve SNPs, two SNPs, rs793391 andrs308739, were found to be associated with COPD in Lifelines. The SNP rs793391 was also confirmed to be associatedwith lung function changes.Conclusions: We show that SUMF1 expression is affected in COPD patients compared to controls, and that SNPs inSUMF1 are associated with an increased risk of COPD. Certain COPD-associated SNPs have effects on either SUMF1gene expression or on lung function. Collectively, this study shows that SUMF1 is associated with an increased risk ofdeveloping COPD.Keywords: Chronic obstructive pulmonary disease, Lung fibroblast, Single nucleotide polymorphism, Sputum, Sulfatasemodifying factor 1* Correspondence: Ellen.Tufvesson@med.lu.se1Respiratory Medicine and Allergology, Department of Clinical Sciences Lund,BMC, D12, Lund University, Skåne University Hospital, 221 84 Lund, SwedenFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Weidner et al. Respiratory Research  (2017) 18:77 DOI 10.1186/s12931-017-0562-5BackgroundIn recent years, Chronic Obstructive Pulmonary Disease(COPD) has risen to the third leading cause of mortalityworld-wide [1]. The disease is irreversible and characterizedby chronic inflammation around the bronchi and bron-chioles leading to fibrosis, tissue destruction, and the deve-lopment of emphysema. Smoking is the main risk factor fordeveloping COPD, although other environmental factorssuch as air pollution can also trigger the development ofthe disease.Several recent studies have sought to uncover geneticcauses of COPD in order to better understand the diseaseand its progression [2–9]. Through genome-wide asso-ciation studies (GWAS) and whole genome sequencing,several genes and single nucleotide polymorphisms (SNPs)have been identified as being associated with COPD [2, 3].To date, the only known single gene mutation related toCOPD is in SERPINA1, which leads to alpha1-antitrypsindeficiency [10]. Multiple cohorts have identified othergenes as being associated with COPD susceptibility, buttheir role in the pathology of the disease remains to beidentified [2].In the lung, the extracellular matrix is important forthe proper formation and maintenance of the struc-ture of the alveoli, highlighting the importance of pro-teoglycans in lung development [11]. Sulfatases act onvarious cellular substrates, including glycosaminogly-cans (GAGs) on proteoglycans, and all sulfatases inthe cell are regulated by a single protein, sulfatasemodifying factor-1 (SUMF1) [12, 13]. SUMF1 modu-lates a very specific and unique post-translationalmodification in the active site of sulfatases [14–17].Mutations in SUMF1 lead to a variety of human dis-eases, including effects in the lungs, where an over-abundance of sulfated GAGs accumulate [18–20]. Todate, there have been no reports on measured GAGsin COPD. Recently, it was observed that a Sumf1−/−mouse developed an emphysema-like phenotype follow-ing an arrest of alveolarization [21, 22]. It is, however,unknown if SUMF1 may be involved in the developmentof COPD.The aim of this study was to examine if SUMF1 isassociated with COPD. Primarily we aimed to inves-tigate the SUMF1 expression in COPD patients. Byusing expression quantitative trait loci (eQTL) ana-lysis, we investigated if SNPs in SUMF1 were asso-ciated with SUMF1 expression in lung tissue, andinvestigated SUMF1 mRNA expression in sputumcells and lung fibroblasts. Thereafter, we examinedwhether there was a genetic association betweenSUMF1 and COPD amongst smokers in a populationbased cohort, and subsequently investigated advancedlung physiology from subjects in the context of thedifferent genotypes.MethodsA flowchart diagram (Fig. 1) provides an overview of allthe analyses performed in this study investigating theassociations between SUMF1 and COPD.Patients in the Lund cohortForty controls and 82 COPD patients, defined accordingto GOLD criteria (forced expiratory volume in 1 second(FEV1)/forced volume capacity (FVC) <0.7), were in-cluded in the Lund cohort (Table 1). All subjects werecurrent smokers or ex-smokers with >15 pack-years, hadnormal levels of alpha-1 antitrypsin, and had no historyof asthma, lung cancer, or any other cardiorespiratorydiseases. They did not suffer from any lower respiratoryinfections within 3 weeks prior to the visit. They wereasked to refrain from inhaled bronchodilators for 8 h forshort acting beta agonists and short acting muscarinicantagonists and 48 h for long acting beta agonists andlong acting muscarinic antagonists before the visit. Allstudy participants performed flow-volume spirometry,body plethysmography (MasterScreen Body, Erich JaegerGmbH), and single breath helium dilution carbon mon-oxide diffusion (MasterScreen Diffusion, Erich JaegerGmbH) after bronchodilation (400 μg salbutamol,Buventol Easyhaler®). Lung function measurements wereperformed according to manufacturer’s instructions andEuropean Respiratory Society/American Thorax Societyrecommendations [23–25]. The reference values usedwere established by Crapo et al. [26] (spirometry), andfrom Quanjer et al. [27] (Body plethysmography andcarbon monoxide diffusion). All subjects signed writteninformed consent and the study was approved by theRegional Ethics Review Board in Lund.Sputum induction and processingSputum was induced by inhalation of 3% saline for5 min, and thereafter 4.5% saline for 2x5 min. After eachstep, patients were asked to try to expectorate sputum.Samples were picked for plugs which were incubatedwith 4 volumes of cold 0.1% dithiothreitol in phosphatebuffered saline. After 30 min incubation in 4 °C,additional 4 volumes of phosphate buffered saline wereadded, and the sample was filtered (60 μm filters). Cellswere pelleted at 1000 × g for 5 minutes (4 °C) and lysedfor future RNA analysis [28].Lung fibroblasts from biopsiesA bronchoscopy was performed in 15 COPD patients.Central lung biopsies were sampled from which fibro-blasts were isolated as previously published [29].RNA extraction and qPCR analysisFor examination of RNA from various body tissues, theHuman Total RNA Master Panel II Lot# 1505145AWeidner et al. Respiratory Research  (2017) 18:77 Page 2 of 14(TakaraBio-Clonetech, Saint-Germain-en-Laye, France)was utilized.For mRNA analyses, RNA was extracted from wholeblood, sputum cells, and lung fibroblasts. cDNA synthesisand quantitative real-time PCR (qPCR) was performed asdescribed previously [29].qPCR analysis in the Lund cohortMultiple protein coding splice variants have been identi-fied for SUMF1, of which the functional role and tissuespecificity remains unknown. We focused on three well-established splice variants in SUMF1 (Splice variants 1(full length), 2 (lacking exon 3) and 3 (lacking exon 8);For primer sequences and NCBI codes see Additionalfile 1: Table S1) that were predicted at the time of thisstudy. All mRNA expressions were normalized againstexpression of the reference genes β-Actin and GAPDH(see Additional file 1: Table S1).Patient selection in the Lung eQTL datasetTo assess associations between the SNPs and SUMF1gene expression in lung tissue (i.e., cis-acting expression(RNA) quantitative trait loci (cis-eQTL) analysis), theSUMF1 mRNA in different organsSUMF1 mRNA in sputum and blood(COPD vs controls) in the Lund cohort(n=122) eQTL analyses in the lung eQTLm dataset (n=512) SNP analyses in the LifeLines cohort (n=1483)SUMF1 mRNA in sputum (n=38) and  lung fibroblasts (n=15)in r F-1 SNP genotypesLung physiology in re  SUMF-1  SNP genotypes(n=122) Gene expression Genetic analysisIn-vivo and in-vitrovalidationAssociation withclinical f eaturesFig. 1 Flowchart diagram providing an overview of the analyses of SUMF1 in relation to COPD done in this studyTable 1 Characteristics of the total Lund cohortControls(n = 40)COPD(n = 82)Sex (male/female) 19/21 46/36Smoking status (current/former) 7/32a 24/58Age (years) 68 (66–70) 67 (62–69)Pack-years 26 (21–36)a 37 (27–48)**BMI (kg/m2) 27 (23–28) 26 (23–29)FEV1 (%predicted) 94 (90–103) 60 (49–72)***FEV1/FVC 0.77 (0.73–0.79) 0.53 (0.44–0.62)***RV (%predicted) 117 (102–128) 144 (116–165)***TLC (%predicted) 106 (99–111) 113 (102–122)*RV/TLC 0.41 (0.38–0.46) 0.47 (0.42–0.54)***VA (%predicted) 90 (86–99) 86 (79–94)*DLCO (%predicted) 76 (69–89) 58 (48–68)***DLCO/VA (%predicted) 88 (78–96) 69 (57–82)***Pulmonary function data is post inhalation of β2 agonist (400 μgsalbutamol). Data presented as median (interquartile range). Pack years isdefined as the equivalent of smoking 1 pack per day for a yearBMI body mass index, RV residual volume, TLC total lung capacity,VA alveolar volume, DLCO diffusion lung capacity*p < 0.05; **p < 0.01 and ***p < 0.001amissing data from 1 patient. * depicts significantly different from controlsWeidner et al. Respiratory Research  (2017) 18:77 Page 3 of 14Lung eQTL consortium was used, including lung tissuesamples obtained from patients at three participatingsites; University of Groningen (GRN), Laval University(Laval) and University of British Columbia (UBC) [6].Tissue was obtained from patients that underwentlung resectional surgery. DNA samples were genotypedwith Illumina Human1M-Duo BeadChip arrays, andgene expression profiles were obtained using a customAffymetrix microarray. Gene expression data is availableon the Gene Expression Omnibus accession numberGSE23546 and platform GPL10379.Imputed SNP data was available for 1,095 of the 1,111subjects, covariate data was missing for another 8subjects. In the current analyses, we included currentand ex-smokers >40 years with ≥5 pack-years. COPDwas defined as an FEV1/FVC ratio <0.7. Non-COPDcontrol was defined as an FEV1/FVC ≥ 0.7. In case lungtissue samples were derived from healthy donors, nodata on FEV1 or FEV1/FVC ratio were available. ForFEV1 and FEV1/FVC, pre-bronchodilator values wereused when post-bronchodilator values were not avail-able. Subjects with other lung diseases such as asthma,cystic fibrosis or interstitial lung diseases were excluded.The final dataset included 512 subjects. Patientsprovided written informed consent and the study wasapproved by the ethics committees of the Institut univer-sitaire de cardiologie et de pneumologie de Québec andthe UBC-Providence Health Care Research InstituteEthics Board for Laval and UBC, respectively. The studyprotocol was consistent with the Research Code of theUniversity Medical Center Groningen and Dutchnational ethical and professional guidelines.First, cohort specific (GRN, Laval and UBC) principalcomponents (PCs) were calculated based on residualsfrom linear regression models on 2-log transformed geneexpression levels (of each probe separately) adjusted forage, gender and smoking status (never/ever/unknown).PCs that explained at least one percent of the total vari-ance were saved and included as covariates in the mainanalysis, these were 14 PCs for GRN and Laval, and 16 forUBC. Second, in each cohort separately, linear regressionanalysis was used to test for association between the SNPsand 2-log transformed gene expression levels. SNPs weretested in an additive genetic model and the models wereadjusted for disease status, age, gender, smoking statusand the cohort specific number of PCs. Finally, SNP effectestimates of the three cohorts were meta-analyzed usingfixed effects models with effect estimates weighted by theinverse of the standard errors.A cis-eQTL was defined as a SNP that was significantlyassociated with expression levels of a probe (gene) withina 50 Kb distance of that SNP. We focused on SNPs whichoverlapped between eQTL imputed database and CytoChip 12, the array used to genotype the Lifelines cohort.Associations between SUMF1 SNPs and COPD in theLifeLines cohortAssociations between SUMF1 SNPs and COPD wasperformed in a Dutch general-population based cohort,the LifeLines cohort study [30]. Subjects with completegenotype and phenotype data (existing data [30]) wereincluded when having smoked at least 5 pack-years andif over 50 years of age. COPD was defined as havingFEV1/FVC < 0.7 and FEV1%predicted < 80, based onQuanjer et al.[24] with pre-bronchodilator spirometryfollowing European Respiratory Society/AmericanThorax Society criteria [24]. Controls were defined ashaving FEV1/FVC ≥ 0.7 and FEV1% predicted > 90.In the Lifelines cohort, genotyping was performed usingIlluminaCytoSNP-12 arrays and SNPs were included thatfulfilled the quality control criteria: genotype call-rate≥95%, minor allele frequency ≥1%, and Hardy-Weinbergequilibrium cut-off p-value ≥10−4. Samples with call ratesbelow 95% were excluded.SUMF1 genotyping in the Lund cohortWhole blood was taken from all subjects in the Lund co-hort and DNA was extracted. All patients were genotypedfor the SUMF1 SNPs identified to be top hits in the eQTLanalysis and Lifelines using Agena iPLEX genotyping.Genotyping was performed at the Mutation AnalysisFacility at Karolinska University Hospital (Huddinge,Sweden) using iPLEX® Gold chemistry and MassARRAY®mass spectrometry system [31] (Agena Bioscience, SanDiego, CA, U.S.A.). Multiplexed assays were designedusing MassARRAY® Assay Design v4.0 Software (AgenaBioscience). Protocol for allele-specific base extension wasperformed according to Agena Bioscience’s recom-mendation. Analytes were spotted onto a 384-elementSpectroCHIP II array (Agena Bioscience) using Nanodis-penser RS1000 (Agena Bioscience) and subsequently ana-lyzed by MALDI-TOF on a MassARRAY® Compact massspectrometer (Agena Bioscience). Genotype calls weremanually checked by two persons individually usingMassARRAY® TYPER v4.0 Software (Agena Bioscience).StatisticsDescriptive statistics are presented as median (interquar-tile range (IQR)). P < 0.05 was considered significant.The differences in gene expression in sputum andblood between controls and COPD patients were ana-lyzed using the Mann-Whitney U-test using GraphPadPrism 5 (Graphpad, La Jolla, CA, USA). In the Lund co-hort the associations between SNPs and gene expressionin sputum and lung fibroblasts as well as in the lungphysiology was tested using the Kruskal-Wallis testincluding Dunn’s Multiple Comparison Post Test (usingGraph Pad Prism 5 software). The eQTL-analyses usingthe lung eQTL-consortium data are described above.Weidner et al. Respiratory Research  (2017) 18:77 Page 4 of 14The association between SNPs and COPD in theDutch cohort was performed using logistic regressionmodels including the SNP in a co-dominant geneticmodel and adjusted for sex, age, and pack years usingSPSS version 22.Finally, associations between the SNPs (in an additivemodel) and lung function parameters were tested usinglinear regression adjusted for COPD, smoking status,and age in the Lund cohort (using SPSS version 22).ResultsDescription of the Lund cohortTable 1 shows the descriptive statistics of the Lund co-hort. An adequate sputum sample, from which RNAcould be extracted, could be obtained from 38 subjects(19 controls and 19 COPD patients, Additional file 1:Table S2) in the Lund cohort. Additional file 1: Table S3shows the descriptive statistics of the 15 COPD patientsin the Lund cohort that performed a bronchoscopy, andfrom which lung fibroblasts were obtained.SUMF1 expression is altered in COPD patients comparedto controlsWe found that SUMF1 mRNA was expressed relativelyhigh in whole lung tissue (Figs. 2a-d), and specifically,Splice variant 3 showed the highest expression in lungtissue compared to all other investigated tissues in thebody (Fig. 2d).To examine if SUMF1 expression was systemic or lungspecific, sputum cells and whole blood from COPD pa-tients and controls from the Lund cohort were examinedfor differences in SUMF1 expression. In sputum cells(Figs. 2e-h), all three splice variants examined were detect-able and showed significantly lower levels in COPDpatients than controls in Splice variant 2 (p = 0.018) andSplice variant 3 (p = 0.0086). While in contrast, in wholeblood there was no significant difference in total SUMF1expression between controls and COPD patients (p = 0.39,Additional file 2: Figure S1), and the three splice variantswere unable to be detected in the majority of individuals.Lung expression quantitative trait loci (eQTL) analysis andlinkage disequilibrium analysisWe next performed an expression quantitative trait loci(eQTL) analysis in lung tissue in order to determinewhether the differential gene expression of SUMF1 wereassociated with genetic polymorphisms. In the threelarge cohorts (Groningen, Laval, and UBC; n = 512)examined, twelve of the SNPs, that overlapped with thearray used to genotype the Lifelines cohort, showedsignificant expression differences (Table 2).A linkage disequilibrium (LD) analysis (HaploView4.2) show the associations between the twelve SUMF1SNPs identified (Fig. 3a).The top hit SNP from the eQTL analysis, rs11915920(Fig. 4a) provided a strong eQTLs (Table 2). For furtherdata presentation in this study, the most significant SNPassociated with gene expression, i.e., rs11915920, is usedfor further data presentation in this study.The SNP rs793391 (Fig. 4b), the most significant SNPfrom the Lifelines cohort (see below), was also a signifi-cant eQTL (Table 2), but to a much smaller extent.SUMF1 SNPs show differences in SUMF1 expression in thelungIn the Lund cohort, the SUMF1 mRNA levels, of totalSUMF1 and the different splice variants, were examinedin sputum cells from controls and COPD patients as wellas in lung fibroblasts from COPD patients in relation tothe SUMF1 genotypes of SNPs rs11915920 and rs793391.Similar trends in SUMF1 mRNA expression were seenin both sputum cells and fibroblasts with SNP rs11915920(Fig. 5). Significant differences were observed among thers11915920 genotypes, with a higher expression level insubjects homozygous for the reference allele (C), inall splice variants in sputum cells (Fig. 5b-d; Splicevariant 1: p = 0.017, Splice variant 2: p = 0.038, Splicevariant 3: p = 0.015). In lung fibroblasts, the expression ofSplice variant 3 was significantly different between thegenotypes (Fig. 5h, p = 0.014). These in vitro findingsvalidate the eQTL analysis where there were also higherlevels of mRNA expression observed in subjects with thereference allele (C) of rs11915920 (Fig. 4a). The top candi-date from our SNP analyses of the Lifelines cohort (seebelow), rs793391, did not show any association withSUMF1 expression in sputum cells or lung fibroblasts(Additional file 2: Figure S2). rs793391 was a much weakercandidate than rs11915920 in the eQTL analysis and thein vitro analysis corroborates these results.Association between SUMF1 SNPs and COPDWe also investigated the association between the SUMF1SNPs associated with eQTLs and COPD in the Dutchcohort, LifeLines (n = 1483, for descriptive statistics see(Additional file 1: Table S4)). Convincingly, the referenceallele (A) of SNP rs793391 was associated with a higherrisk for COPD in the LifeLines cohort (Table 3). Inaddition, the SNP rs308739 was also associated withCOPD. (For allele frequencies, see Table 4). The mostsignificant SNP from the association between SUMF1and COPD, rs793391, was chosen to be followed uppathophysiologically in this study.SNP in SUMF1 is associated with lung functionWhen examining advanced lung physiology in subjectsfrom the Lund Cohort, including controls and COPDpatients, we found that among the rs793391 genotypesthere was an overall difference in FEV1/FVC (p = 0.031),Weidner et al. Respiratory Research  (2017) 18:77 Page 5 of 14A BC DWhole BrainBone MarrowFetal LiverHeartKidneyFetal BrainLiverLungPlacentaProstateSalivary GlandAdrenal GlandTestisSmall IntestineSpleenStomachThymusColonSkeletal MuscleUterus0.00.51.01.52.0Total SUMF1A.U.Whole BrainBone MarrowFetal LiverHeartKidneyFetal BrainLiverLungPlacentaProstateSalivary GlandAdrenal GlandTestisSmall IntestineSpleenStomachThymusColonSkeletal MuscleUterus0.000.020.040.060.080.10SUMF1 splice variant 1A.U.Whole BrainBone MarrowFetal LiverHeartKidneyFetal BrainLiverLungPlacentaProstateSalivary GlandAdrenal GlandTestisSmall IntestineSpleenStomachThymusColonSkeletal MuscleUterus0.000.020.040.060.08SUMF1 splice variant 2A.U.Whole BrainBone MarrowFetal LiverHeartKidneyFetal BrainLiverLungPlacentaProstateSalivary GlandAdrenal GlandTestisSmall IntestineSpleenStomachThymusColonSkeletal MuscleUterus0.0000.0010.0020.0030.0040.005SUMF1 splice variant 3A.U.Controls COPD0.00010.0010.010.1110Total SUMF1A.U.SUMF1 splice variant 1Controls COPD0.00010.0010.010.1110A.U.SUMF1 splice variant 2Controls COPD0.00010.0010.010.1110 M-W: *A.U.SUMF1 splice variant 3Controls COPD0.00010.0010.010.1110M-W: **A.U.E FG HFig. 2 (See legend on next page.)Weidner et al. Respiratory Research  (2017) 18:77 Page 6 of 14FEV1%predicted (p = 0.035), diffusion capacity (DLCO =lung diffusion capacity for carbon monoxide)%predicted(p = 0.027) and alveolar volume (VA)%predicted (p =0.040). Specifically, subjects homozygous for the refer-ence allele of rs793391 had lower FEV1/FVC andFEV1%predicted compared to heterozygous subjects(Fig. 6a and b, respectively). A similar pattern was seenin DLCO%predicted (Fig. 6c) and VA%predicted amongthe different SUMF1 rs793391 genotypes, but not inDLCO/VA%predicted. Interestingly, even after correctionfor COPD, smoking status and age, the association be-tween rs793391 and DLCO%predicted remained signifi-cant, while the association between rs793391 and FEV1/FVC, FEV1%predicted and VA%predicted did not(Additional file 1: Table S5).Neither residual volume, total lung capacity, nor airtrapping index (residual volume/total lung capacity)showed any difference among the different genotypes ofrs793391 (data not shown).The SNP rs11915920, highly significant in the eQTLanalyses, did not have any significant association withmeasured lung function parameters (Additional file 2:Figure S3), neither had the SNP rs308739.DiscussionWe found that SUMF1 was associated with COPD. Pri-marily we showed that SUMF1 is differently expressedin sputum cells from COPD patients and controls. Inaddition, eQTL analysis revealed that several SNPs weresignificantly associated with SUMF1 expression, with thetop hit being SNP rs11915920. This was further verifiedin mRNA from sputum cells and lung fibroblasts, andthe main differences were in SUMF1 Splice variant 3.We also show that two SNPs in SUMF1, rs793391 and308739, were associated with increased risk of COPD ina population based cohort, LifeLines. Finally, we foundthat rs793391 was associated with differences in lungfunction parameters.Our study found, that SUMF1 Splice variant 3 wasmost highly expressed in whole lung tissue as comparedto other tissues examined in the body and showed thebiggest expression effect in lung fibroblasts. Splice vari-ant 3 lacks exon 8 in SUMF1 (Fig. 3b) but, currently, noeffects regarding the protein function or structure of thisvariant have been reported. Additionally, rs11915920,the top hit SNP related to SUMF1 expression in lung tis-sue, is in close proximity to SUMF1 exon 8 (Fig. 3b),and might affect the splicing of exon 8. Perhaps Splicevariant 3 is an important variant of SUMF1 specificallyin the lungs with a yet unknown function. Future studieswill be needed to investigate this possibility.The importance of SUMF1 to the development andmaintenance of alveoli was recently discovered in mice[21, 22]. Although Sumf1 −/− mice have a very shortTable 2 eQTL analysis of SUMF1 SNPs in lung tissue from three large cohorts (Groningen, Laval, and UBC; n = 512)SNP Ref Var eQTL meta-estimate (B) eQTL meta-standard error (SE) eQTL meta-p-valuers11915920 C T −0.110 0.009 6.41E-38rs2819562 C T −0.096 0.009 2.46E-26rs809437 A G −0.081 0.011 2.41E-14rs17030493 T C 0.066 0.013 3.64E-07rs1687863 G A 0.056 0.013 6.97E-06rs1968930 A C 0.054 0.014 7.84E-05rs1688411 T G 0.048 0.014 0.0005rs807785 C T 0.037 0.011 0.0011rs308739 A C −0.060 0.019 0.0019rs1688413 C T 0.035 0.012 0.0028rs17040589 C T −0.050 0.021 0.0199rs793391 A C 0.022 0.011 0.0400Presented are SNPs that were significantly associated with expression levels of a probe (gene) within a 50Kb distance of that SNP and overlapped with the arrayused to genotype the Lifelines cohort. Bold indicates significant valuesRef reference allele, Var variance allele(See figure on previous page.)Fig. 2 SUMF1 expression is altered in COPD patients. A master panel of mRNA from twenty different human tissues was probed for total SUMF1mRNA expression (a) as well as three individual Splice variants 1 (b), 2 (c) and 3 (d). Total SUMF1 mRNA expression (e) as well as Splice variant 1(f), 2 (g) and 3 (h) expression were examined in sputum cells from COPD patients and controls in the Lund cohort. * = p < 0.05 and ** = p < 0.01,A.U. = Arbitrary units, M-W =Mann-Whitney test usedWeidner et al. Respiratory Research  (2017) 18:77 Page 7 of 14lifespan, they have provided a wealth of information re-garding sulfatase activation and function. In these mice,there was an overabundance of sulfated GAGs resultingin inactive sulfatases, leading to an arrest in the alveolar-ization process and an emphysema-like phenotype [22].This emphysema-like phenotype was one of our firsthints that perhaps SUMF1 may play a role in the devel-opment of COPD, which is hallmarked by the develop-ment of emphysema. In addition to the emphysema-likephenotype, many cell and tissue types were found tohave massive GAG accumulation in the Sumf1 −/− mice,but this has not yet been investigated in COPD.We show that DLCO%predicted is independently af-fected by the rs793391, since it is not driven by the dis-ease or smoking status, which is the case for FEV1/FVCand FEV1%predicted (Additional file 1: Table S5). Ourfinding that DLCO%predicted is lower in patients withthe reference allele (A) of rs793391 is in accordance withthe Sumf1 −/− mouse showing a deficient alveolar sept-ation and a subsequent arrest in alveolar formation.Interestingly, no difference in residual volume, air trap-ping index (residual volume/total lung capacity), orDLCO%predicted corrected for alveolar volume (DLCO/VA%predicted) was observed between the SUMF1BUTR 5´3´rs793391rs11915920rs281956256 4 3 2 19 8 7UTR 5´3´ 56 4 2 19 8 7UTR 5´3´ 56 4 3 2 19 7Splicevariant 1Splicevariant 2Splicevariant 3rs308739rs17030493rs809437rs1687863rs1688411rs1688413rs1968930rs807785rs17040589AFig. 3 Linkage Disequilibrium analysis of SUMF1 SNPs. An LD plot (a) shows the 12 SUMF1 SNPs overlapping between the eQTL analysis and thearray used to genotype the Lifelines cohort. A schematic picture (b) showing localization of the 12 SNPs on the SUMF1 mRNA, and the differentsplice variants. Boxes showing exon 1–9, UTR = untranslated regionWeidner et al. Respiratory Research  (2017) 18:77 Page 8 of 14A B AA(228)    AC(231)     CC(53) AA(64)     AC(55)      CC(8) AA(104)    AC(106)    CC(34) AA(60)     AC(70)    CC(11) Fig. 4 eQTL analysis of SUMF1 SNPs. Each set of box plots represents the three different cohorts, combined (ALL) as well as separately(GRN = Groningen, Laval = Laval University and UBC = University of British Columbia), and the corrected expression differences seenbetween the different SNP genotypes. a represents the SNP rs11915920 and b represents the SNP rs793391. Genotype is presented withthe reference/reference genotype to the leftWeidner et al. Respiratory Research  (2017) 18:77 Page 9 of 14SNP rs11915920SUMF1 splice variant 1C/C C/T T/T0.00010.0010.010.1110 K-W: *D:*A.U.Total SUMF1C/C C/T T/T0.00010.0010.010.1110A.U.SUMF1 splice variant 2C/C C/T T/T0.00010.0010.010.1110 K-W: *A.U.SUMF1 splice variant 3C/C C/T T/T0.00010.0010.010.1110 K-W: *D:*A.U.ACBDFHEGTotal SUMF1C/C C/T T/T0.00010.0010.010.11 K-W*D*A.U.SUMF1 splice variant 1C/C C/T T/T0.00010.0010.010.11A.U.SUMF1 splice variant 2C/C C/T T/T0.00010.0010.010.11A.U.SUMF1 splice variant 3C/C C/T T/T0.00010.0010.010.11K-W*D*A.U.Fig. 5 (See legend on next page.)Weidner et al. Respiratory Research  (2017) 18:77 Page 10 of 14genotypes. These findings are also in agreement with theSumf1 −/− mouse, suggesting a developmental perturb-ation of distal alveolar septation rather than a destructiveprocess. Future studies will be performed in order tofocus on extensive lung physiology in larger cohorts todetermine if the clinical phenotype related to SUMF1SNPs holds true.In lung function data, the subjects with homozygousreference genotype of rs793391 (AA) showed impairedlung physiology compared to the respective heterozy-gous genotype. The heterozygous genotype thereby ap-pears to be protective in association with COPD.However, there was no significant differences betweenthe homozygous reference and variance genotypes (AAversus CC in rs793391), which might be due to the lownumber of patients in the Lund cohort that had thehomozygous variance genotype. Unfortunately, onlyflow-volume spirometry was performed in the Lifelinescohort, so we were not able to verify the differences inDLCO%predicted observed in patients with variousrs793391 genotypes in the Lund cohort. Future studieswill be needed to determine if this potentially protectivegenotype holds true for other populations.A recent GWAS identified genetic variants associ-ated with total lung capacity in COPD [4]. Amongseveral SNPs that were identified in patients withprominent emphysema, one was in SUMF1, however,it was not studied futher. This GWAS identified SNPwas neither present in our analysis platforms, nor wasit found to be in LD with either of SNPs described inour study.To our knowledge this is the first study to geneticallyfocus on SUMF1 in the context of COPD. Our resultsindicate that the different SUMF1 SNPs may be respon-sible for different factors in the development of the dis-ease. We showed that several SNPs were associated withSUMF1 expression, however, on a functional level themolecular mechanism and their relationship to COPDremains undiscovered. Alternatively, as all of the SUMF1SNPs from this study were found to be in introns or un-translated regions (none are found in translated exons),there is the possibility that they may act as small RNAprecursors, such as microRNAs. These small RNAs may,in turn, regulate the expression of SUMF1 or anotherunknown gene, but this possibility has yet to beTable 3 Logistic regression models assessing associationsbetween SUMF1 SNPs and COPD (additive model) in theLifeLines cohortLifeLines cohortn = 1483SUMF1 SNP Ref Var OR SE p-valuers793391 A C 1.42 0.13 0.0066rs308739 A C 0.40 0.36 0.010rs807785 C T 0.82 0.13 0.14rs1688411 T G 0.77 0.18 0.16rs1968930 A C 0.78 0.19 0.19rs1687863 G A 0.84 0.15 0.24rs17030493 T C 0.87 0.17 0.39rs1688413 C T 0.90 0.14 0.44rs809437 A G 0.92 0.13 0.55rs17040589 C T 0.88 0.25 0.62rs11915920 C T 0.99 0.12 0.90rs2819562 C T 1.01 0.12 0.96Shown are SNPs that were significant eQTLs and overlapped with the arrayused to genotype the Lifelines cohort. OR = odds ratio, SE = standard error,p-value is from logistic regression models assessing associations between SNPs(additive model) and COPD, adjusted for sex, age, and pack years. Smokingcontrols were defined as an FEV1/FVC > 0.7 and COPD was defined as an FEV1/FVC < 0.7. Ref = reference allele. Var = variance allele. Bold indicatessignificant values(See figure on previous page.)Fig. 5 SUMF1 expression in sputum cells and lung fibroblasts divided by rs11915920 genotype. SUMF1 expression, including the three splicevariants, was examined for SNP rs11915920 in sputum cells (a-d) and lung fibroblasts (e-h) from subjects from the Lund cohort. Both controls andCOPD patients were used for sputum cell analysis and COPD patients for the lung fibroblasts, then divided depending on genotype. Opensymbols = controls, filled symbols = COPD patients. A.U. = Arbitrary units, * = significance at p < 0.05. K-W = Kruskal-Wallis test was used, followedby Dunn’s multiple comparison post tests (=D). Genotype is presented with the reference/reference genotype to the leftTable 4 Genotype and allele frequencies in the LifeLines cohortLifeLines cohortn = 1483SUMF1 SNP Ref Var Ref/Refgenotype n (%)Ref/Vargenotype n (%)Var/Vargenotype n (%)MAFrs793391 A C 669 (45) 644 (43) 170 (12) 0.33rs308739 A C 6 (0.4) 137 (9) 1340 (90) 0.05rs807785 C T 122 (8) 587 (40) 774 (52) 0.28rs1688411 T G 24 (2) 356 (24) 1103 (74) 0.14rs1968930 A C 22 (2) 339 (23) 1122 (76) 0.13rs1687863 G A 54 (4) 455 (31) 974 (66) 0.19rs17030493 T C 35 (2) 395 (27) 1053 (71) 0.16rs1688413 C T 95 (6) 527 (36) 861 (58) 0.24rs809437 A G 103 (7) 597 (40) 783 (53) 0.27rs17040589 C T 9 (1) 164 (11) 1310 (88) 0.06rs11915920 C T 366 (25) 766 (52) 351 (24) 0.49rs2819562 C T 293 (20) 775 (52) 415 (28) 0.46Ref reference allele, Var variance allele, MAF minor allele frequencyWeidner et al. Respiratory Research  (2017) 18:77 Page 11 of 14examined. In contrast to rs11915920, which was stronglyassociated to SUMF1 expression, rs793391 had a uni-form impact on lung function. These findings lead us tobelieve that the different SNPs may have different rolesin the biology of the disease. SUMF1 is a good candidatefor further study into how the genotype of patients af-fects the different phenotypes of COPD on a molecularlevel. Future studies into downstream effects of SUMF1,such as sulfatase activity would need to be undertakenand we can begin to delve deeper into the molecularmechanisms of the disease and work towards better pos-sible treatments for those affected.A limitation of the study is that the different cohortshave been analysed with platforms investigating differentSNPs, and subsequently only twelve of the significantSNPs in the lung tissue dataset were found in theLifelines cohort. Another limitation is the difference inrationale for inclusion in the cohorts. The LifeLines co-hort is a large general-population based study, giving ahigh power. However, most COPD patients have only amild disease, and the possibility of finding relevant genesin a multigenetic disease such as COPD might then bedifficult. This might explain why there is a lack of associ-ation between COPD and several of the different SNPs.This could also explain why there is a strong relation-ship between rs11915920 and SUMF1 expression, but nodirect association to COPD in the population basedLifeLines cohort. Maybe a cohort including patients withmore severe COPD would give a significant associationbetween rs11915920 and COPD. This is suggested froma subanalysis of the Lund cohort, comparing 24 more se-vere COPD patients versus the contrasting 24 clearlyhealthy controls, showing a significant association tors11915920 (data not shown), even though the subjectnumbers were low. This hypothesis needs to be furtherexplored in larger cohorts where more patients with se-vere COPD are included.ConclusionWe provide evidence that expression and genetic regula-tion of SUMF1 differs between smokers with and with-out COPD. SUMF1 is differentially expressed in sputumcells from COPD patients and controls. Through exam-ination of the SUMF1 gene, we found SNPs thatFig. 6 Lung function in COPD patients and controls divided by SNPrs793391 genotypes. FEV1/FVC (a), FEV1 (b) and DLCO %predicted (c)of subjects from the Lund cohort are divided according to thegenotype of rs793391. Open symbols = controls, filled symbols =COPD patients. * = p < 0.05, ** = p < 0.01. K-W = Kruskal-Wallis testwas used, followed by Dunn’s multiple comparison post tests (=D).Genotype is presented with the reference/reference genotype tothe leftWeidner et al. Respiratory Research  (2017) 18:77 Page 12 of 14significantly affect mRNA levels through the use of aneQTL analysis from a lung tissue dataset, which was cor-roborated in vitro by mRNA expression analysis of spu-tum cells and lung fibroblasts from the Lund cohort. Inaddition, some of these SNPs in SUMF1 are associatedwith an increased risk of COPD. Furthermore, the differ-ent SUMF1 SNPs were found to have differential effectsin COPD. Some SNPs, such as rs11915920, had an effecton SUMF1 mRNA expression in tissue, sputum cells,and lung fibroblasts, while the SNP rs793391 was signifi-cantly associated with lung function parameters andthereby COPD.Additional filesAdditional file 1: Table S1. Primer sequences used for qPCR. Table S2.Characteristics of controls and COPD patients who could expectorate anadequate sputum. Table S3. Characteristics of COPD patients fromwhom lung fibroblasts were obtained. Table S4. Characteristics ofsubjects included in the LifeLines cohort. Table S5. Associations betweenSNP rs793391 and FEV1/FVC, FEV1%predicted and DLCO%predicted usinglinear regression models adjusted for current smoking status, age, andCOPD. (DOCX 18 kb)Additional file 2: Figure S1. SUMF1 expression in blood from COPDpatients and controls. Total SUMF1 mRNA expression was examined inwhole blood from COPD patients and controls in the Lund cohort.A.U. = Arbitrary units. Figure S2. SUMF1 expression in sputum cells andlung fibroblasts divided by rs793391 genotypes. SUMF1 expression,including the three splice variants, was examined for the SNP rs793391genotype in sputum cells (A-D) from controls and COPD patients and inlung fibroblasts (E-H) obtained from COPD patients from the Lundcohort. Open symbols = controls, filled symbols = COPD patients,A.U. = Arbitrary units. Genotype is presented with the reference/referencegenotype to the left. Figure S3. Lung function in COPD patients andcontrols divided by SNP rs11915920 genotype. FEV1/FVC (A), FEV1 (B) andDLCO %predicted (C) of subjects from the Lund cohort are dividedaccording to the genotype of rs11915920. Open symbols = controls, filledsymbols = COPD patients. Genotype is presented with the reference/reference genotype to the left. (PPTX 275 kb)AbbreviationsCOPD: Chronic obstructive pulmonary disease; DLCO: Lung diffusion capacityfor carbon monoxide); eQTL: Expression (RNA) quantitative trait loci;FEV1: Forced expiratory volume in 1 second; FVC: Forced volume capacity;GAGs: Glycosaminoglycans; GWAS: Genome-wide association studies;SNP: Single nucleotide polymorphism; SUMF1: Sulfatase modifying factor 1;VA: Alveolar volumeAcknowledgmentsWe would like to thank the staff at the Lung and Allergy research Unit,Skåne University hospital for assistance and Anders Olin for includingpatients in the Lund cohorts. We would like to thank Ida Åberg for excellentexperimental help. The authors would like to thank the staff at theRespiratory Health Network Tissue Bank of the FRQS for their valuableassistance with the lung eQTL dataset at Laval University.FundingEllen Tufvesson and The Lund cohort study was supported by independentresearch grants from the Swedish Heart and Lung Foundation, CrafoordFoundation, Evy and Gunnar Sandberg’s Foundation and the RoyalPhysiographic Society in Lund. K. de Jong is supported by grant number4.113.007 the Lung Foundation Netherlands. The LifeLines cohort study wassupported by the Dutch Ministry of Health, Welfare and Sport, the Ministry ofEconomic Affairs, Agriculture and Innovation, the province of Groningen, theEuropean Union (regional development fund), the Northern NetherlandsProvinces (SNN), the Netherlands Organization for Scientific Research (NWO),University Medical Center Groningen (UMCG), University of Groningen, deNierstichting (the Dutch Kidney Foundation), and the Diabetes Fonds (theDiabetic Foundation). The Vlagtwedde-Vlaardingen cohort study wassupported by the Ministry of Health and Environmental Hygiene of theNetherlands and the Netherlands Asthma Fund (grant 187) and theNetherlands Asthma Fund grant no. 3.2.02.51, the Stichting Astma Bestrijding,BBMRI-NL (Complementiation project), and the European Respiratory SocietyCOPD research award 2011 to H.M. Boezen. Y. Bossé holds a Canada ResearchChair in Genomics of Heart and Lung Diseases. A. Faiz holds a ERS RESPIRE2fellowship.Availability of data and materialsPlease contact author for data request.Authors’ contributionsConcept and Design: JW, LJ, KJ, DP, AF, ET; Acquisition of data: JW, LJ, KJ, DS,YB, DN, JA, LB, AF, ET; Analysis and Interpretation: JW, LJ, KJ, JV, DP, AF, ET;Critical comments: All authors provided critical comments. All authors readand approved the final manuscript.Competing interestsDr. van den Berge reports research grants paid to University fromGlaxoSmithkline, TEVA, and Chiesi, outside the submitted work; Dr. Sinreports grants and personal fees from AstraZeneca, personal fees fromBoehringer Ingelheim, personal fees from Almirall, outside the submittedwork; Dirkje S. Postma: The University of Groningen has received money forProfessor Postma regarding a grant for research from Astra Zeneca, Chiesi,Genentec, GSK and Roche. Fees for consultancies were given to theUniversity of Groningen by Astra Zeneca, Boehringer Ingelheim, Chiesi, GSK,Takeda and TEVA. None of the other authors have any conflict of interest.Consent for publicationNot applicableEthics approval and consent to participateAll subjects signed written informed consent. The Regional Ethics ReviewBoard in Lund approved the study on the Lund cohort (ref. 2008–431). Theethics committees of the Institut universitaire de cardiologie et depneumologie de Québec and the UBC-Providence Health Care ResearchInstitute Ethics Board approved the study for Laval and UBC, respectively.The study was approved by the Medical Ethics Committee of the UniversityMedical Center Groningen, Groningen, The Netherlands (ref. METc 2007/152).Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Respiratory Medicine and Allergology, Department of Clinical Sciences Lund,BMC, D12, Lund University, Skåne University Hospital, 221 84 Lund, Sweden.2University Medical Center Groningen, GRIAC (Groningen Research Institutefor Asthma and COPD), Department of Epidemiology, University ofGroningen, Groningen, The Netherlands. 3University Medical CenterGroningen, Department of Pulmonology, GRIAC (Groningen ResearchInstitute for Asthma and COPD), University of Groningen, Groningen, TheNetherlands. 4Department of Medicine (Respirology), University of BritishColumbia, Centre for Heart Lung Innovation, Vancouver, Canada.5Department of Molecular Medicine, Institut universitaire de cardiologie etde pneumologie de Québec, Laval University, Québec, Canada. 6Geneticsand Pharmacogenomics (GpGx), Merck Research Laboratories, Boston, MA,USA.Received: 16 February 2017 Accepted: 21 April 2017References1. (WHO) WHO. The top 10 causes of death [web page on the Internet.Geneva: WHO. http://www.who.int/mediacentre/factsheets/fs310/en/[updated May 2014. Fact sheet number 310].Weidner et al. Respiratory Research  (2017) 18:77 Page 13 of 142. Kim WJ, Lee SD. Candidate genes for COPD: current evidence and research.Int J Chron Obstruct Pulmon Dis. 2015;10:2249–55.3. Huang Q. Genetic study of complex diseases in the post-GWAS era. J GenetGenomics. 2015;42(3):87–98.4. Lee JH, McDonald ML, Cho MH, Wan ES, Castaldi PJ, Hunninghake GM, et al.DNAH5 is associated with total lung capacity in chronic obstructivepulmonary disease. Respir Res. 2014;15:97.5. Brandsma CA, van den Berge M, Postma DS, Jonker MR, Brouwer S, Pare PD,et al. A large lung gene expression study identifying fibulin-5 as a novelplayer in tissue repair in COPD. Thorax. 2015;70(1):21–32.6. Hao K, Bosse Y, Nickle DC, Pare PD, Postma DS, Laviolette M, et al. LungeQTLs to help reveal the molecular underpinnings of asthma. PLoS Genet.2012;8(11):e1003029.7. Dijkstra AE, Smolonska J, van den Berge M, Wijmenga C, Zanen P,Luinge MA, et al. Susceptibility to chronic mucus hypersecretion, agenome wide association study. PLoS One. 2014;9(4):e91621.8. Chen X, Xu X, Xiao F. Heterogeneity of chronic obstructive pulmonarydisease: from phenotype to genotype. Front Med. 2013;7(4):425–32.9. Bossé Y. Updates on the COPD gene list. Int J Chron Obstruct Pulmon Dis.2012;7:607–31.10. Silverman EK, Sandhaus RA. Clinical practice. Alpha1-antitrypsin deficiency.N Engl J Med. 2009;360(26):2749–57.11. Buono M, Cosma MP. Sulfatase activities towards the regulation of cellmetabolism and signaling in mammals. Cell Mol Life Sci. 2010;67(5):769–80.12. Cosma MP, Pepe S, Annunziata I, Newbold RF, Grompe M, Parenti G, et al.The multiple sulfatase deficiency gene encodes an essential and limitingfactor for the activity of sulfatases. Cell. 2003;113(4):445–56.13. Dierks T, Schmidt B, Borissenko LV, Peng J, Preusser A, Mariappan M,et al. Multiple sulfatase deficiency is caused by mutations in the geneencoding the human C(alpha)-formylglycine generating enzyme. Cell.2003;113(4):435–44.14. Cosma MP, Pepe S, Parenti G, Settembre C, Annunziata I, Wade-Martins R,et al. Molecular and functional analysis of SUMF1 mutations in multiplesulfatase deficiency. Hum Mutat. 2004;23(6):576–81.15. Fraldi A, Biffi A, Lombardi A, Visigalli I, Pepe S, Settembre C, et al. SUMF1enhances sulfatase activities in vivo in five sulfatase deficiencies. Biochem J.2007;403(2):305–12.16. Dickmanns A, Schmidt B, Rudolph MG, Mariappan M, Dierks T, von Figura K,et al. Crystal structure of human pFGE, the paralog of the Calpha-formylglycine-generating enzyme. J Biol Chem. 2005;280(15):15180–7.17. Preusser-Kunze A, Mariappan M, Schmidt B, Gande SL, Mutenda K,Wenzel D, et al. Molecular characterization of the human Calpha-formylglycine-generating enzyme. J Biol Chem. 2005;280(15):14900–10.18. Diez-Roux G, Ballabio A. Sulfatases and human disease. Annu Rev GenomicsHum Genet. 2005;6:355–79.19. Schlotawa L, Ennemann EC, Radhakrishnan K, Schmidt B, Chakrapani A,Christen HJ, et al. SUMF1 mutations affecting stability and activity offormylglycine generating enzyme predict clinical outcome in multiplesulfatase deficiency. Eur J Hum Genet. 2011;19(3):253–61.20. Berger KI, Fagondes SC, Giugliani R, Hardy KA, Lee KS, McArdle C, et al.Respiratory and sleep disorders in mucopolysaccharidosis. J Inherit MetabDis. 2013;36(2):201–10.21. Settembre C, Annunziata I, Spampanato C, Zarcone D, Cobellis G, Nusco E,et al. Systemic inflammation and neurodegeneration in a mouse model ofmultiple sulfatase deficiency. Proc Natl Acad Sci U S A. 2007;104(11):4506–11.22. Arteaga-Solis E, Settembre C, Ballabio A, Karsenty G. Sulfatases aredeterminants of alveolar formation. Matrix Biol. 2012;31(4):253–60.23. Macintyre N, Crapo RO, Viegi G, Johnson DC, van der Grinten CP, Brusasco V,et al. Standardisation of the single-breath determination of carbon monoxideuptake in the lung. Eur Respir J. 2005;26(4):720–35.24. Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al.Standardisation of spirometry. Eur Respir J. 2005;26(2):319–38.25. Wanger J, Clausen JL, Coates A, Pedersen OF, Brusasco V, Burgos F, et al.Standardisation of the measurement of lung volumes. Eur Respir J.2005;26(3):511–22.26. Crapo RO, Morris AH, Gardner RM. Reference spirometric values usingtechniques and equipment that meet ATS recommendations. Am RevRespir Dis. 1981;123(6):659–64.27. Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC.Lung volumes and forced ventilatory flows. Report Working PartyStandardization of Lung Function Tests. European Community for Steel andCoal. Official Statement of the European Respiratory Society. Eur Respir JSuppl. 1993;16:5–40.28. Tufvesson E, Aronsson D, Bjermer L. Cysteinyl-leukotriene levels in sputumdifferentiate asthma from rhinitis patients with or without bronchialhyperresponsiveness. Clin Exp Allergy. 2007;37(7):1067–73.29. Tufvesson E, Nihlberg K, Westergren-Thorsson G, Bjermer L. Leukotrienereceptors are differently expressed in fibroblast from peripheral versuscentral airways in asthmatics and healthy controls. Prostaglandins LeukotEssent Fat Acids. 2011;85(2):67–73.30. Scholtens S, Smidt N, Swertz MA, Bakker SJ, Dotinga A, Vonk JM, et al.Cohort Profile: LifeLines, a three-generation cohort study and biobank. Int JEpidemiol. 2015;44(4):1172–80.31. Jurinke C, van den Boom D, Cantor CR, Koster H. Automated genotypingusing the DNA MassArray technology. Methods Mol Biol. 2002;187:179–92.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Weidner et al. Respiratory Research  (2017) 18:77 Page 14 of 14

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.52383.1-0366295/manifest

Comment

Related Items