Characterization and regulation of ADAMTS-16

Characterization and regulation of ADAMTS-16

Alison K Surridgea,1, Ursula R Rodgersa, Tracey E Swinglera, Rose K Davidsona, LaraKevorkiana,2, Rosemary Nortona, Jasmine G Watersa, Mary B Goldringb, Andrew EParkerc,3, and Ian M Clarka,c,⁎aSchool of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, UK.bHospital for Special Surgery, 535 East 70th Street, New York, USA.cRespiratory and Inflammation Department, AstraZeneca Pharmaceuticals, Cheshire, UK.

AbstractThe ADAMTS (a disintegrin and metalloproteinase domain with thrombospondin motifs) familyincludes 19 secreted proteinases in man. ADAMTS16 is a recently cloned gene expressed at highlevels in fetal lung and kidney and adult brain and ovary. The ADAMTS-16 protein currently has noknown function. ADAMTS16 is also expressed in human cartilage and synovium where its expressionis increased in tissues from osteoarthritis patients compared to normal tissues. In this study, weascertained that the full length ADAMTS16 mRNA was expressed in chondrocytes and cloned theappropriate cDNA. Stable over-expression of ADAMTS16 in chondrosarcoma cells led to a decreasein cell proliferation and migration, though not adhesion, as well as a decrease in the expression ofmatrix metalloproteinase-13 (MMP13). The transcription start point of the human ADAMTS16 genewas experimentally identified as 138 bp upstream of the translation start ATG and the basal promoterwas mapped out to − 1802 bp. Overexpression of Egr1 induced ADAMTS16 promoter constructs of− 157/+138 or longer whilst Sp1 induced all ADAMTS16 promoter constructs. Transforming growthfactor beta (TGFβ) stimulated expression of endogenous ADAMTS16 gene expression in chondrocytecell lines.

AbbreviationsADAMTS, a disintegrin and metalloproteinase domain with thrombospondin motif; MMP, matrixmetalloproteinase; RACE, rapid amplification of cDNA ends; TGFβ, transforming growth factorbeta; TIMP, tissue inhibitor of metalloproteinases

KeywordsADAMTS; Metalloproteinase; Chondrocyte; Cartilage; Promoter; Transcription

© 2009 Elsevier B.V.This document may be redistributed and reused, subject to certain conditions.

⁎Corresponding author. Cellular Protease Group, School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, UK. Tel.:+44 1603 592760; fax: +44 1603 592250. [email protected] address: Department of Zoology, University of Cambridge, UK.2Current address: UCB Celltech, Slough, Berkshire, UK.3Current address: OPSONA Therapeutics Ltd, Dublin, Ireland.This document was posted here by permission of the publisher. At the time of deposit, it included all changes made during peer review,copyediting, and publishing. The U.S. National Library of Medicine is responsible for all links within the document and for incorporatingany publisher-supplied amendments or retractions issued subsequently. The published journal article, guaranteed to be such by Elsevier,is available for free, on ScienceDirect.

Sponsored document fromMatrix Biology

Published as: Matrix Biol. 2009 September ; 28(7): 416–424.

Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

http://www.elsevier.com/wps/find/authorsview.authors/supplementalterms1.0

1 IntroductionThe ADAMTS (a disintegrin and metalloproteinase domain with thrombospondin motifs)family includes 19 secreted proteinases in man. These enzymes have a complex domainstructure consisting of at least a signal peptide, pro domain, metalloproteinase domain,disintegrin domain, thrombospondin type I motif and cysteine rich domain (Porter et al.,2005). Phylogenetically, the enzyme family separates into eight clades which to some extentcorrelate with function where assigned (Huxley-Jones et al., 2005). ADAMTS-1, -4, -5, -8, -9and -15 are all capable of degrading aggrecan at a specific set of loci with ADAMTS-5 beingthe major aggrecanase in cartilage destruction, at least in mice (Clark and Parker, 2003; Collins-Racie et al., 2004; Glasson et al., 2005; Stanton et al., 2005). At least three ADAMTS enzymes(ADAMTS-4, -7 and -12) can degrade cartilage oligomeric matrix protein, another componentof cartilage ECM (Dickinson et al., 2003; Liu et al., 2006a,b). Three ADAMTS enzymes,ADAMTS-2, -3 and -14, are procollagen N-propeptidases (PCNPs), which have roles incollagen biosynthesis. ADAMTS-2 acts preferentially on type I collagen and ADAMTS-3 ontype II collagen, whereas the preference of ADAMTS-14 is currently unknown (Colige et al.,2002). Mutation of the ADAMTS2 gene causes Ehlers–Danlos syndrome Type VIIC in man(Colige et al., 1999). ADAMTS-13 cleaves von Willebrand factor with mutations in thisenzyme leading to an inherited thrombotic thrombocytopenia (Tsai, 2007). A form of Weill–Marchesani syndrome is caused by mutation in the ADAMTS10 gene (Dagoneau et al., 2004).

ADAMTS-16 is a recently described member of the ADAMTS gene family (Porter et al.,2005). The cDNA was cloned using a combination of bioinformatics and degenerate RT-PCR,which also identified ADAMTS-13, -14, -15, -17, -18 and -19 (Cal et al., 2002). Amino acidsequence alignment showed a significant percentage of identity between ADAMTS-16 andADAMTS-18 (overall identity, 57%), particularly in the catalytic domain where identityreaches 85% (Porter et al., 2005). Indeed, the zinc-binding motif is identical between these twoproteinases (HESGHNFGMIHD) (Somerville et al., 2003a). ADAMTS-16 and -18 form aphylogenetic clade, with the nearest evolutionary neighbours being ADAMTS-6, -7, -10 and-12.

Whilst the substrates for ADAMTS-16 are currently unknown, a recombinant truncated formof ADAMTS-16 shows weak aggrecanase activity (Zeng et al., 2006) and full lengthrecombinant ADAMTS-16 is capable of cleaving the proteinase inhibitor α2-macroglobulin(Gao et al., 2007).

An initial expression analysis in a selection of human tissues shows high expression ofADAMTS16 mRNA in fetal lung and kidney, and adult brain and ovary (Cal et al., 2002). Inthis latter tissue, ADAMTS16 is expressed predominantly in the parietal granulosa cells of pre-ovulatory follicles. Expression of the gene can be induced by follicle-stimulating hormone andforskolin in granulosa cells, suggesting that the cAMP pathway may be involved in itsregulation in this system (Gao et al., 2007). ADAMTS16 has also recently been geneticallylinked to inherited hypertension (Joe et al., 2009).

We undertook the expression profiling of all ADAMTS genes in cartilage and synovium. Wecompared expression in patients undergoing hip replacement for osteoarthritis (OA) tophenotypically normal tissues from patients undergoing hip replacement following fracture tothe neck of femur. In cartilage, the expression of ADAMTS16 increased in the OA sampleswith a significance of p < 0.001, comparable to the increase in expression of MMP13, acollagenase whose activity is pathognomic with cartilage destruction in OA. This was backedup with a preliminary study in the knee using cartilage from OA patients compared to normalcartilage from post-mortem, where a similar increase in expression of these two genes wasmeasured (Kevorkian et al., 2004). Similarly, in synovium, the expression of ADAMTS16 was

Surridge et al. Page 2


Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

significantly increased in OA with the absolute level of ADAMTS16 mRNA approximately 10-fold higher in synovium than in cartilage (Davidson et al., 2006).

In order to provide some insight into the regulation and function of ADAMTS-16 in the joint,the current study examines the expression of ADAMTS-16 by chondrocytes, the regulation ofthe gene and the consequence of stable over-expression of the gene in chondrosarcoma andimmortalized chondrocyte cell lines.

2 Results2.1 Characterisation of ADAMTS16 and creation of stably expressing cell lines

Using PCR primer pairs to amplify overlapping sections of the ADAMTS16 gene, the entiretranscript was amplified from cDNA reverse transcribed from mRNA purified from animmortalized human chondrocyte cell line, C28/I2 (Loeser et al., 2000) (data not shown). Noevidence for expression of splice variants was found (though it is possible that some truncatedforms of the transcript might not have been detected if they were not amplified at all by selectedprimers). Based on these data, a full length cDNA was cloned and the final cDNA verified bysequencing in both directions. It should be noted that, despite using high fidelity polymerasesand low numbers of amplifications cycles, frequent mutations were introduced by the PCR.The correct version of ADAMTS16 could only be created by amplifying sections of the gene(up to approximately 1000 bp), verifying sequence, then assembling via restriction digestionand ligation into an error-free cDNA.

Initially, both immortalized chondrocytes (C28/I2) and chondrosarcoma (SW1353) cells weretransfected with the full length ADAMTS16 construct (with a C-terminal FLAG tag). However,the C28/I2 transfectants did not grow robustly after selection and experiments were pursuedin the SW1353 line alone. Following cloning by limiting dilution, four different clonalpopulations expressing ADAMTS16 (TS16) were compared to two clonal populationstransfected with vector only control constructs (VO). Expression of ADAMTS16 at the mRNAlevel was confirmed using qRT-PCR (Fig. 1a). The ADAMTS16 expressing clones had anaverage of approximately 95-fold higher expression than the vector only controls. Western blotusing both an anti-ADAMTS-16 antibody and an anti-FLAG antibody showed expression ofa high molecular weight (~ 130 kDa) protein predominantly in the extracellular matrix fractionof the ADAMTS16 transfectants (Fig. 1b). Two bands of similar size were also detected in theconditioned medium and cell lysate fractions (Fig. 1c).

2.2 Influence of ADAMTS-16 on expression levels of other metzincin genesExpression levels of all 19 members of the ADAMTS family, all four members of the TIMPgene family and MMP2, MMP9, MMP13 and MMP28 were determined in cells stably over-expressing ADAMTS16 compared to vector only controls using qRT-PCR. The representativesof the MMP family were chosen because of their expression pattern in tissues of the OA joint(Davidson et al., 2006; Kevorkian et al., 2004). MMP13 expression levels were significantlyreduced in ADAMTS16-expressing clones compared to vector only controls (p < 0.005; Fig.1d) with a mean of approximately 9-fold lower expression. Interestingly, the level of MMP13expression was inversely correlated with that of ADAMTS16 (compare Fig. 1a and d).Expression levels of the other genes analysed did not change significantly between over-expressing cells and vector only controls (data not shown).

2.3 Effects of ADAMTS16 expression on cell phenotypeExpression of ADAMTS-16 by stably expressing clones was confirmed byimmunocytochemistry using anti-FLAG antibodies (Fig. 2a). Protein was detected only inpermeabilized cells, with cytoplasmic and some perinuclear staining. No staining was observed



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

in vector only controls or non-immune IgG controls. Cells stably over-expressingADAMTS16 were larger and more elongated than vector only controls (Fig. 2b) which werenot significantly different from non-transfected SW1353 cells. These differences wereobserved at all stages of confluence. The ability of cells to adhere to extracellular matrixproteins (collagen I, collagen II, fibronectin and vitronectin) was measured and no significantdifference was observed between ADAMTS16 over-expressing cells and vector only controls(data not shown). Cell proliferation was measured by uptake of tritiated thymidine, a directmeasure of DNA replication. This showed that proliferation of over-expressing cells wassignificantly reduced in all four clones in medium containing 0.5% FCS compared to vectoronly controls (p < 0.001 comparing ADAMTS16-expressing cells vs. vector only; Fig. 3).Proliferation was similarly reduced in cells growing in medium containing 10% FCS thoughwith lower significance (p = 0.01). Migration of cells measured using time-lapse microscopyalso showed that ADAMTS16 over-expressing cells had significantly reduced migration onplastic compared to vector only controls (p < 0.005, comparing mean of ADAMTS16-expressing cells vs. vector only; Fig. 4).

2.4 Regulation of ADAMTS16 by growth factors and cytokinesOf a number of growth factors and cytokines tested (IL-1, TNF-α, TGF-β1, IGF-1, IFN-γ, IL-6,IL-6, IL-4, IL-10, IL-11, IL-17, OSM, EGF, BMP-2, BMP-7), only TGFβ was found to regulatelevels of endogenous ADAMTS16 expression in chondrocyte cell lines. TGFβ induced mRNAexpression at the steady state mRNA level in both SW1353 and C28/I2 cells in a dose-dependent manner (Fig. 5).

2.5 Identification of the transcription start site and minimal promoterFig. 6 shows the alignment of 1000 base pairs (bp) of sequence upstream of the translation startsite of ADAMTS16 for human, chimp, macaque and mouse. This alignment revealed a numberof conserved regions (indicated ‘⁎’ in Fig. 6) that potentially indicate regulatory sequences,with no strong TATA sequence. The predicted translation start site (ATG) is conservedamongst the primates, but not in mouse which appears to use a downstream ATG. This lattersequence is conserved in primates, but not in frame with the upstream primate ATG.

The longest PCR products obtained using 5′ RLM-RACE indicate that the transcription startsite lies 138 bp upstream of the ATG in the human gene. Fig. 7a shows the PCR productsobtained by performing 5′ RLM-RACE on RNA from three different sources (C28/I2 cellsstimulated with TGFβ, ovary and HeLa cells). Fig. 7b shows the position of the primer SP2used for RLM-RACE and the predicted transcription start point. This site is in agreement withan mRNA transcript of ADAMTS16 deposited in Genbank (NM_139056). Primer extensionusing primer SP2 also detected a band of approximately the same size as the RACE products(data not shown).

Eight fragments of the 5′ region of ADAMTS16 ranging from +110/+138 to − 1802/+138(marked on Fig. 6) were cloned into a luciferase reporter vector. Analysis of reporter activity(Fig. 8) indicated that the minimal promoter lies between − 13 and + 25, encompassing thetranscription start point identified above. Activity increased greatly in deletions from − 1802to − 412 indicating the presence of a repressive element within this region. Successive deletionsbetween − 412 and − 13 resulted in incremental decreases in activity, suggesting the presenceof a number of transcription factor binding sites in these regions.

2.6 Identification of transcription factors activating the ADAMTS16 promoterThe proximal promoter is GC rich and a search for transcription factor consensus sequences(MatInspector, Genomatix) revealed a number of both Sp1 and Egr1 binding sites (Fig. 9a).These factors have also been implicated in TGFβ signalling to extracellular matrix genes (Chen



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

ncbi-n:NM_139056

et al., 2006; Zhang et al., 2000). Co-transfection of the ADAMTS16 promoter constructs withexpression vectors for each of these factors showed responsiveness to both transcription factors(Fig. 9b and c). Induction of the promoter by Egr1 is reduced by deletion from − 412 to − 157and lost by further deletion to − 42 (with Egr1 repressing smaller constructs). All promoterconstructs were induced by Sp1 and indeed there are further Sp1 sites downstream of thetranscription start point.

3 DiscussionADAMTS-16 is a multi-domain protein encoded by 23 exons. Initially in this study, we verifiedthat the full length transcript was expressed in chondrocyte cell lines and then cloned this cDNAinto an expression construct with a C-terminal FLAG tag. Chondrosarcoma cells stably over-expressing ADAMTS-16 were selected and cloned. Over-expression of ADAMTS16 wasverified at the mRNA level by qRT-PCR and at the protein level by immunocytochemistry andwestern blot using the anti-FLAG antibody and/or an anti-ADAMTS-16 antibody (raisedagainst a C-terminal peptide). Western blot showed an approximately 130 kDa bandpredominantly associated with the extracellular matrix fraction, with an additional highermolecular weight band faintly detected. The full length ADAMTS-16 polypeptide has apredicted molecular weight of 136 kDa, in good agreement with the experimental finding. Adoublet around this molecular weight was also weakly detected in the cell lysate andconditioned medium of the ADAMTS16 transfects. The lower molecular weight bandcorresponds to the major band in the ECM, suggesting that this is a processed form. Two ‘in-house’ anti-peptide antibodies raised against sequences in the propeptide and catalytic domainrespectively recognise a band of approximately 65 kDa in conditioned medium (data notshown), though the specificity of these antibodies remains to be determined.

Whilst stably transfected SW1353 cells could be maintained and expanded, this wasn't the casefor the C28/I2 transformed chondrocytes. In the SW1353 cells, over-expression ofADAMTS-16 lead to a reduction in cell proliferation and this change in phenotype may, inpart, explain why the C28/I2 cells were not successfully expanded. SW1353 cells over-expressing ADAMTS-16 also showed a change in cell shape and a decrease in cell migrationon plastic, though no difference in cell adhesion to a number of extracellular matrix proteins.This does not appear to be the consequence solely of stable transfection since similar cell linesover-expressing MMP-28 display a different phenotype (Rodgers et al., 2009). Interestingly,ADAMTS-1 modifies both cell proliferation and migration dependent on cell type, domaincontent and concentration of the enzyme used (Krampert et al., 2005; Liu et al., 2006c).

Members of the metzincin family are known to process a number of growth factors, cytokinesand signalling molecules in addition to matrix substrates (Somerville et al., 2003b). Since theover-expression of ADAMTS-16 has no effect on expression levels of most of the ADAMTS,TIMP and MMP genes tested, it is unlikely that ADAMTS-16 processes factors that regulatethese genes. However, our initial interests in ADAMTS-16 centred around the fact that in endstage osteoarthritis, the expression of the ADAMTS16 gene is increased in cartilage andsynovium in a way similar to that of MMP-13, a collagenolytic enzyme known to be involvedin the destruction of tissues in the joint [17, 18]. Interestingly, the SW1353 cells over-expressing ADAMTS-16 display reduced expression of MMP13 mRNA, though themechanism or functional significance of this observation is unknown. There is no clear link inthe literature between decreased MMP13 expression and decreased proliferation or migration(as observed for ADAMTS16 over-expressing cells). Indeed one agent, pleiotrophin, hasrecently been reported to induce proliferation and migration of chondrocytes whilst repressingMMP13 expression (Pufe et al., 2007). In OA cartilage, other factors must override therepression of MMP13 expression by increased ADAMTS-16 since both enzymes showincreased expression compared to that in normal cartilage. Given the importance of MMP-13



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

as a collagenase in OA, the mechanistic link between ADAMTS-16 and MMP13 expressionis worthy of further investigation.

Since ADAMTS16 expression was increased in OA tissues, we tested a number of growthfactors and cytokines implicated in this disease for their ability to regulate expression of theADAMTS16 gene in vitro. In both C28/I2 and SW1353 cells, TGFβ induced expression ofADAMTS16 in a dose-dependent fashion. TGFβ has been reported to induce MMP13expression in chondrocytes (Moldovan et al., 1997) and there is evidence to demonstrate a rolefor this growth factor in the development or progression of osteoarthritis (Blaney Davidson etal., 2007).

Sequence alignment shows a high level of conservation between primates and mouse in the 5′upstream region of the ADAMTS16 gene. 5′ RACE gives a transcription start point 138 bpupstream of the translation start point and this is in agreement with both primer extension dataand the end of exon 1 given in Ensembl (www.ensembl.org). Interestingly, the sequence acrossthe transcription start point (− 2/+5, 5′TCAGTAA3′) is a putative AP-1 binding site. This iscontrary to the computer prediction of the transcription start point at 560 bp upstream of thetranslation start point (− 422 from the transcription start point) and the associated promoterreported in Gao et al. (Gao et al., 2007), however it is possible that the gene has greater thanone transcription start point.

A deletion analysis from − 1802 (relative to the transcription start point) showed that constructsexpressed promoter activity only when they encompassed the transcription start point.Promoter activity increased with the increasing length to − 412, a region which is GC rich andcontains many putative binding sites for transcription factors such as Sp1 and Egr1.Interestingly, these factors have previously been implicated in TGFβ signalling, with bothfactors involved in the induction of type I collagen by TGFβ (Chen et al., 2006; Zhang et al.,2000). Forced over-expression of these factors demonstrates that both factors are capable ofinducing the ADAMTS16 promoter. Egr1 induction appears to be mediated mainly by the − 412/− 157 region, with Sp1 induction also mediated through further downstream sequences.

In summary, over-expression of ADAMTS-16 leads to decreased cell proliferation andmigration either directly, or indirectly via changes in gene expression e.g. MMP13. Expressionof ADAMTS16 is induced by TGFβ and we have experimentally identified the transcriptionstart point of the gene, its promoter, and two relevant activating transcription factors. Overall,these data provide new information on the function and regulation of ADAMTS-16 inchondrocyte cell lines.

4 Experimental procedures4.1 PCR and cloning of full length transcript

RNA was extracted from C28/I2 immortalised human chondrocyte cells (Loeser et al., 2000)using TRIzol® reagent (Invitrogen) and adding 0.5× volume of chloroform. The aqueous phasewas recovered after centrifugation at 12,000 rpm for 15 min at 4 °C and added to an equalvolume of isopropanol. RNA was pelleted by centrifugation at 12,000 rpm for 30 min at 4 °Cand washed with 70% ethanol. RNA was eluted into distilled water and quantified using aNanoDrop® spectrophotometer (Nanodrop Technologies). cDNA was synthesized from 1 µgof RNA using SuperscriptTM II reverse transcriptase (Invitrogen) and random hexamers,according to the manufacturer's recommendations. PCRs were performed using 50 ng of cDNAtemplate, 0.5 µM each primer, 200 µM dNTPs and 0.5 U of AccuTaqTM LA polymerase(Sigma-Aldrich) in the manufacturer's buffer, in a PTC-100 thermal cycler (MJ Research).PCR products were visualised on 1% agarose gels stained with ethidium bromide. For cloning,PCR primers at each end of the gene were tagged with four random nucleotides (italics) and



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

http://www.ensembl.org

an EcoRI (5′) and NotI (3′) restriction enzyme recognition site (underlined). The 5′ forwardprimer also contained a consensus Kozak sequence (bold; (Kozak, 1987)). Primer sequenceswere as follows: 5′-ACGTGAATTCGCCGCCACCATGAAGCCCCGCGCG-3′ and 5′-ACGTGCGGCCGCCCAAGTTGGACTTAGAGCAAG-3′ for forward and reverse primersrespectively. PCR products were digested with EcoRI and NotI and ligated into a modifiedversion of the pcDNA4TM expression vector (Invitrogen) expressing an in frame C-terminalFLAG-tag. Sequencing of the cloned gene was performed in both directions usingBigDyeTM v.3.1 terminator sequencing chemistry (Applied Biosytems) on an ABI Prism®

3730 capillary sequencer (Applied Biosystems).

4.2 Creation of stably expressing cell linesSW1353 human chondrosarcoma cells (ATCC) or immortalized C28/I2 chondrocytes werecultured in DMEM plus GlutaMAXTM (Invitrogen) with 10% fetal calf serum (FCS;Invitrogen). 100,000 cells per well were plated into 6-well plates and incubated overnight. TheADAMTS16/pcDNA4-FLAG construct was linearised by digesting with BglII and 1 µg wastransfected into cells using FuGENE®6 transfection reagent (Roche) according to themanufacturer's recommendations. After 48 h, cells were switched to culture medium containing200 µg/ml zeocin (Melford Laboratories). After 7 days, surviving cells were plated into 96-well plates containing three, one and 0.3 cells/well. Wells containing cells were expanded andplated again into 96-well plates containing three, one and 0.3 cells/well to select for clonalcolonies originating from single cells. Cells were maintained in zeocin-containing media.

4.3 Quantitative real-time PCRCells were plated into 6-well plates containing 200,000 cells/well in triplicate. After 24 h, cellswere changed to serum-free media and incubated for a further 24 h. RNA extraction and cDNAsynthesis were performed as before. Primers and fluorescent probes for ADAMTS16 and all ofthe ADAMTSs, MMPs and TIMPs were designed using Primer Express 1.0 software (AppliedBiosystems) and are described elsewhere (Nuttall et al., 2003; Porter et al., 2004). The 18SrRNA gene was used as an endogenous control to normalise for differences in the amount ofRNA between samples. Primers and probe for 18S were obtained from Applied Biosystems.Quantitative PCRs were performed and analysed as described previously (Davidson et al.,2006).

4.4 ImmunocytochemistryStably transfected SW1353 cells were plated into chamber slides at a density of 10,000 cells/well. After incubating for 24 h, cells were switched to serum-free media for a further 24 h.Cells were fixed using 4% paraformaldehyde, blocked with 3% BSA and incubated with a1:5000 dilution of mouse ANTI-FLAG® M2 IgG monoclonal antibody (Sigma) and a 1:1000dilution of FITC-labelled rabbit anti-mouse IgG secondary antibody (Dako). Nuclei werestained with 2.5 µg/ml DAPI. Cells were also incubated with mouse IgG and secondaryantibody as before as a control. Cells were visualised using an AxioPlan 2ie confocalmicroscope with Axiovision software (Carl Zeiss).

4.5 Western blot analysisStably transfected SW1353 cells were plated at 100,000/well of a 6-well plate. After 24 h cellswere switched to serum-free medium for a further 48 h. The conditioned media was collectedand proteins precipitated with 0.5 vol. of 10% (v/v) trichloroacetic acid (TCA) for 1 h on ice.Samples were centrifuged for 15 min at 13,000 rpm, and pellets washed with cold acetone andpelleted for a further 15 min at 13,000 rpm. Protein pellets were air-dried and resuspended in1 × SDS final sample buffer (0.058 M Tris-HCl, pH 6.8; 5% v/v glycerol; 1.7% w/v SDS and0.002% w/v bromophenol blue) containing 50 mM DTT and 6 M urea (FSB/DTT/urea). Cells



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

were detached with 0.5 mM EDTA in PBS. Cell pellets were collected by centrifugation at fullspeed for 5 min, resuspended in FSB/DTT/urea and sonicated prior to analysis. ECM proteinswere harvested by scraping each well with FSB/DTT/urea for approximately 30 s/well. Allsamples were boiled for 5 min and subjected to western blotting. Proteins were separated bySDS-PAGE, followed by transfer to PVDF membrane. ADAMTS-16 was probed using a1:5000 dilution of mouse ANTI-FLAG® M2 IgG monoclonal antibody (Sigma-Aldrich)followed by a 1:1000 dilution of HRP-labelled rabbit anti-mouse IgG secondary antibody(Dako, Denmark). An antibody raised against a 75 amino acid peptide from the C-terminalthrombospondin repeat domain of human ADAMTS-16 (Santa Cruz SC-50490) were used toverify expression. LumiGLO reagent and peroxide (Cell Signalling Technology, Beverley,MA) were used for detection on Kodak Biomax MS film (Sigma-Aldrich) bychemiluminescence.

4.6 Adhesion, proliferation and migration assays96-well culture plates were coated with 50 µg/ml of vitronectin or fibronectin or 100 µg/ml ofcollagens I or II overnight at 4 °C. Plates were blocked with 1% BSA for 1 h at 37 °C beforeadding 30,000 cells/well. After 15 min, plates were washed six times with PBS. Cell adhesionwas measured using a Cell Titer 96® AQueous One Solution Cell Proliferation Assay kit(Promega), according to the manufacturer's recommendations. For proliferation assays, cellswere plated in medium containing 10% FCS at 25,000 cells/well in 24 well plates in replicatesof six and allowed to adhere overnight. Each well was labelled with 0.5 µCi per well [6-3H]-thymidine for 6 h in either medium containing 10% or 0.5% FCS, washed twice with HBSSand fresh medium added containing 3 µM cold thymidine. After overnight incubation, mediumwas aspirated and 300 µl/well 0.25 M ammonia added. Plates were rocked for 2 h at roomtemperature and lysate from each well transferred into scintillant for counting tritium. Formigration assays, cells were plated at 10,000 cells/well in 24-well plates. After 8 h cells weresubjected to time-lapse microscopy on an AxioVert 200 M microscope that was enclosed inan Incubator XL (Carl Zeiss) for temperature and CO2 control. Cells were photographed at15 min intervals for 13 h. Total distance moved (in µm) for 10 cells/well was measured usingAxiovision Software (Carl Zeiss).

4.7 Induction with cytokines and growth factorsSW1353 cells were plated into six-well plates at 200,000 cells per well. After 24 h cells weretransferred into serum-free media. Cells were incubated with each factor in triplicate for 6 hfor Taqman analysis. RNA extraction, cDNA synthesis and TaqMan® real-time PCR wereperformed as described above for analysis of ADAMTS16.

4.8 Promoter alignmentThe 5′ upstream region of human ADAMTS16 was obtained and aligned with chimp, macaqueand mouse using the Ensembl database www.ensembl.org.

4.9 5′ RACE5′ RLM-RACE was performed using a GeneRacerTM kit (Invitrogen) according to themanufacturer's recommendations. The primer 5′TGGAAAAGGGCGAATATGAC3′ (SP2)was used to generate amplified products which were cloned using a TOPO-TA cloning kit(Invitrogen) according to the manufacturer's recommendations. DNA sequencing wasperformed using the M13 universal primers supplied with the kit.

4.10 Cloning of a deletion set and luciferase reporter assaysGenomic DNA was extracted from cultured SW1353 cells using a QIAamp DNA Mini Kit(Qiagen) according to the manufacturer's recommendations. PCR products were obtained using



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent


50 ng of genomic DNA template, 0.5 µM of each primer, (synthesised with a 5′ MluIrecognition site) 200 µM dNTPs and 0.5 U of AccuTaqTM LA polymerase (Sigma-Aldrich) inthe manufacturer's buffer, in a PTC-100 thermal cycler (MJ Research). For fragments of 29 bpand 54 bp, synthetic oligonucleotides were manufactured. Fragments were run on 1% agarosegels stained with ethidium bromide, digested with MluI and cloned into the MluI site of thepGL3-basic luciferase reporter vector (Promega). Insert orientation was verified by sequencingusing pGL3 sequencing primers. SW1353 cells were plated into 24-well plates at 20,000 cells/well. Each pGL3 construct was transfected in triplicate using 200 ng DNA and FuGENE®6transfection reagent (Roche), according to the manufacturer's recommendations. After 24 hcells were lysed and assayed using a Luciferase Reporter Gene Assay kit (Roche) followingthe manufacturer's recommendations. Transfections were performed with at least two differentisolates of each plasmid. The Sp1 cDNA was a kind gift from Professor G. Suske (Marburg,Germany) and was subcloned into the pcDNA4 vector (above). The Egr1 expression constructwas as previously described (Tan et al., 2003).

AcknowledgmentsAKS was funded by a grant from the Arthritis Research Campaign UK.

ReferencesBlaney Davidson E.N. van der Kraan P.M. van den Berg W.B. TGF-beta and osteoarthritis. Osteoarthr.

Cartil. 2007;15:597–604. [PubMed: 17391995]Cal S. Obaya A.J. Llamazares M. Garabaya C. Quesada V. Lopez-Otin C. Cloning, expression analysis,

and structural characterization of seven novel human ADAMTSs, a family of metalloproteinases withdisintegrin and thrombospondin-1 domains. Gene 2002;283:49–62. [PubMed: 11867212]

Chen S.J. Ning H. Ishida W. Sodin-Semrl S. Takagawa S. Mori Y. Varga J. The early-immediate geneEGR-1 is induced by transforming growth factor-beta and mediates stimulation of collagen geneexpression. J. Biol. Chem. 2006;281:21183–21197. [PubMed: 16702209]

Clark I.M. Parker A.E. Metalloproteinases: their role in arthritis and potential as therapeutic targets.Expert Opin. Ther. Targets 2003;7:19–34. [PubMed: 12556200]

Colige A. Sieron A.L. Li S.W. Schwarze U. Petty E. Wertelecki W. Wilcox W. Krakow D. Cohn D.H.Reardon W. Byers P.H. Lapiere C.M. Prockop D.J. Nusgens B.V. Human Ehlers–Danlos syndrometype VII C and bovine dermatosparaxis are caused by mutations in the procollagen I N-proteinasegene. Am. J. Hum. Genet. 1999;65:308–317. [PubMed: 10417273]

Colige A. Vandenberghe I. Thiry M. Lambert C.A. Van Beeumen J. Li S.W. Prockop D.J. Lapiere C.M.Nusgens B.V. Cloning and characterization of ADAMTS-14, a novel ADAMTS displaying highhomology with ADAMTS-2 and ADAMTS-3. J. Biol. Chem. 2002;277:5756–5766. [PubMed:11741898]

Collins-Racie L.A. Flannery C.R. Zeng W. Corcoran C. Annis-Freeman B. Agostino M.J. Arai M.DiBlasio-Smith E. Dorner A.J. Georgiadis K.E. Jin M. Tan X.Y. Morris E.A. LaVallie E.R.ADAMTS-8 exhibits aggrecanase activity and is expressed in human articular cartilage. Matrix Biol.2004;23:219–230. [PubMed: 15296936]

Dagoneau N. Benoist-Lasselin C. Huber C. Faivre L. Megarbane A. Alswaid A. Dollfus H. Alembik Y.Munnich A. Legeai-Mallet L. Cormier-Daire V. ADAMTS10 mutations in autosomal recessive Weill–Marchesani syndrome. Am. J. Hum. Genet. 2004;75:801–806. [PubMed: 15368195]

Davidson R.K. Waters J.G. Kevorkian L. Darrah C. Cooper A. Donell S.T. Clark I.M. Expressionprofiling of metalloproteinases and their inhibitors in synovium and cartilage. Arthritis Res. Ther.2006;8:R124. [PubMed: 16859525]

Dickinson S.C. Vankemmelbeke M.N. Buttle D.J. Rosenberg K. Heinegard D. Hollander A.P. Cleavageof cartilage oligomeric matrix protein (thrombospondin-5) by matrix metalloproteinases and adisintegrin and metalloproteinase with thrombospondin motifs. Matrix Biol. 2003;22:267–278.[PubMed: 12853037]



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Gao S. De Geyter C. Kossowska K. Zhang H. FSH stimulates the expression of the ADAMTS-16 proteasein mature human ovarian follicles. Mol. Hum. Reprod. 2007;13:465–471. [PubMed: 17519324]

Glasson S.S. Askew R. Sheppard B. Carito B. Blanchet T. Ma H.L. Flannery C.R. Peluso D. Kanki K.Yang Z. Majumdar M.K. Morris E.A. Deletion of active ADAMTS5 prevents cartilage degradationin a murine model of osteoarthritis. Nature 2005;434:644–648. [PubMed: 15800624]

Huxley-Jones J. Apte S.S. Robertson D.L. Boot-Handford R.P. The characterisation of six ADAMTSproteases in the basal chordate Ciona intestinalis provides new insights into the vertebrate ADAMTSfamily. Int. J. Biochem. Cell Biol. 2005;37:1838–1845. [PubMed: 15899586]

Joe B. Saad Y. Lee N. Frank B. Achinike O. Luu T. Gopalakrishnan K. Toland E. Farms P. Yerga-Woolwine S. Manickavasagam E. Rapp J. Garrett M. Coe D. Apte S. Rankinen T. Perusse L. EhretG. Ganesh S. Cooper R. Connor A. Rice T. Weder A. Chakravarti A. Rao D. Bouchard C. Positionalidentification of variants of ADAMTS16 linked to inherited hypertension. Hum. Mol. Genet.2009;18:2825–2838. [PubMed: 19423552]

Kevorkian L. Young D.A. Darrah C. Donell S.T. Shepstone L. Porter S. Brockbank S.M. Edwards D.R.Parker A.E. Clark I.M. Expression profiling of metalloproteinases and their inhibitors in cartilage.Arthritis Rheum. 2004;50:131–141. [PubMed: 14730609]

Kozak M. An analysis of 5′-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic AcidsRes. 1987;15:8125–8148. [PubMed: 3313277]

Krampert M. Kuenzle S. Thai S.N. Lee N. Iruela-Arispe M.L. Werner S. ADAMTS1 proteinase is up-regulated in wounded skin and regulates migration of fibroblasts and endothelial cells. J. Biol. Chem.2005;280:23844–23852. [PubMed: 15843381]

Liu C.J. Kong W. Ilalov K. Yu S. Xu K. Prazak L. Fajardo M. Sehgal B. Di Cesare P.E. ADAMTS-7: ametalloproteinase that directly binds to and degrades cartilage oligomeric matrix protein. FASEB J.2006;20:988–990. [PubMed: 16585064]

Liu C.J. Kong W. Xu K. Luan Y. Ilalov K. Sehgal B. Yu S. Howell R.D. Di Cesare P.E. ADAMTS-12associates with and degrades cartilage oligomeric matrix protein. J. Biol. Chem. 2006;281:15800–15808. [PubMed: 16611630]

Liu Y.J. Xu Y. Yu Q. Full-length ADAMTS-1 and the ADAMTS-1 fragments display pro- andantimetastatic activity, respectively. Oncogene 2006;25:2452–2467. [PubMed: 16314835]

Loeser R.F. Sadiev S. Tan L. Goldring M.B. Integrin expression by primary and immortalized humanchondrocytes: evidence of a differential role for alpha1beta1 and alpha2beta1 integrins in mediatingchondrocyte adhesion to types II and VI collagen. Osteoarthr. Cartil. 2000;8:96–105. [PubMed:10772239]

Moldovan F. Pelletier J.P. Hambor J. Cloutier J.M. Martel-Pelletier J. Collagenase-3 (matrixmetalloprotease 13) is preferentially localized in the deep layer of human arthritic cartilage in situ:in vitro mimicking effect by transforming growth factor beta. Arthritis Rheum. 1997;40:1653–1661.[PubMed: 9324020]

Nuttall R.K. Pennington C.J. Taplin J. Wheal A. Yong V.W. Forsyth P.A. Edwards D.R. Elevatedmembrane-type matrix metalloproteinases in gliomas revealed by profiling proteases and inhibitorsin human cancer cells. Mol. Cancer Res. 2003;1:333–345. [PubMed: 12651907]

Porter S. Scott S.D. Sassoon E.M. Williams M.R. Jones J.L. Girling A.C. Ball R.Y. Edwards D.R.Dysregulated expression of adamalysin-thrombospondin genes in human breast carcinoma. Clin.Cancer Res. 2004;10:2429–2440. [PubMed: 15073121]

Porter S. Clark I.M. Kevorkian L. Edwards D.R. The ADAMTS metalloproteinases. Biochem. J.2005;386:15–27. [PubMed: 15554875]

Pufe T. Groth G. Goldring M.B. Tillmann B. Mentlein R. Effects of pleiotrophin, a heparin-bindinggrowth factor, on human primary and immortalized chondrocytes. Osteoarthr. Cartil. 2007;15:155–162. [PubMed: 16949312]

Rodgers U.R. Kevorkian L. Surridge A.K. Waters J.G. Swingler T.E. Culley K. Illman S. Lohi J. ParkerA.E. Clark I.M. Expression and function of matrix metalloproteinase (MMP)-28. Matrix Biol.2009;28:263–272. [PubMed: 19375502]

Somerville R.P. Longpre J.M. Jungers K.A. Engle J.M. Ross M. Evanko S. Wight T.N. Leduc R. ApteS.S. Characterization of ADAMTS-9 and ADAMTS-20 as a distinct ADAMTS subfamily related toCaenorhabditis elegans GON-1. J. Biol. Chem. 2003;278:9503–9513. [PubMed: 12514189]



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Somerville R.P. Oblander S.A. Apte S.S. Matrix metalloproteinases: old dogs with new tricks.GenomeBiology 2003;4:216. [PubMed: 12801404]

Stanton H. Rogerson F.M. East C.J. Golub S.B. Lawlor K.E. Meeker C.T. Little C.B. Last K. Farmer P.J.Campbell I.K. Fourie A.M. Fosang A.J. ADAMTS5 is the major aggrecanase in mouse cartilage invivo and in vitro. Nature 2005;434:648–652. [PubMed: 15800625]

Tan L. Peng H. Osaki M. Choy B.K. Auron P.E. Sandell L.J. Goldring M.B. Egr-1 mediates transcriptionalrepression of COL2A1 promoter activity by interleukin-1beta. J. Biol. Chem. 2003;278:17688–17700. [PubMed: 12637574]

Tsai H.M. Thrombotic thrombocytopenic purpura: a thrombotic disorder caused by ADAMTS13deficiency. Hematol. Oncol. Clin. North Am. 2007;21:609–632. [PubMed: 17666281]

Zeng W. Corcoran C. Collins-Racie L.A. Lavallie E.R. Morris E.A. Flannery C.R. Glycosaminoglycan-binding properties and aggrecanase activities of truncated ADAMTSs: comparative analyses withADAMTS-5, -9, -16 and -18. Biochim. Biophys. Acta 2006;1760:517–524. [PubMed: 16507336]

Zhang W. Ou J. Inagaki Y. Greenwel P. Ramirez F. Synergistic cooperation between Sp1 and Smad3/Smad4 mediates transforming growth factor beta1 stimulation of alpha 2(I)-collagen (COL1A2)transcription. J. Biol. Chem. 2000;275:39237–39245. [PubMed: 11007770]



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Fig. 1.Stable over-expression of ADAMTS16 in SW1353 cells. SW1353 cells were stably transfectedwith either vector only (VO) or ADAMTS16 (TS16) expression constructs. Cells wereharvested and total RNA isolated and subjected to qRT-PCR for expression of (a)ADAMTS16 or (d) MMP13. Data are normalized to 18 S and expressed as mean ± s.e.m. Meanthreshold cycle (Ct) is given above each bar. Extracellular matrix (b) or cell lysate andconditioned medium (c) were harvested (VO vs. TS16) and subjected to western blot analysisusing anti-ADAMTS-16 and/or anti-FLAG primary antibodies.



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Fig. 2.Effect of ADAMTS-16 on cell phenotype. SW1353 cells were stably transfected with eithervector only (VO) or ADAMTS16 (TS16) expression constructs. (a) Cells were fixed in 4%paraformaldehyde and stained with a 1:5000 dilution of mouse anti-FLAG antibody followedby FITC-linked second antibody; nuclei were counterstained with DAPI. (b) Cells werevisualised under phase contrast 72 h after plating on plastic (×10 objective).



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Fig. 3.Proliferation of SW1353 cells stably expressing ADAMTS16. SW1353 cells were stablytransfected with either vector only (VO) or ADAMTS16 (TS16) expression constructs. Cellswere allowed to adhere overnight in DMEM/10% FCS then treated with 3H-thymidine inDMEM/0.5% FCS for 6 h followed by an overnight cold chase. Cell lysates were harvestedand counted for tritium. Data are expressed as mean ± s.e.m., ⁎p < 0.05; ⁎⁎p < 0.01,⁎⁎⁎p<0.001 comparing TS16 with either VO-1 or VO-2.



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Fig. 4.Cell migration of SW1353 cells stably expressing ADAMTS16. SW1353 cells were stablytransfected with either vector only (VO) or ADAMTS16 (TS16) expression constructs. Eighthours after plating in DMEM/10% FCS, cells were subjected to time-lapse microscopy at15 min intervals for 13 h. Distance moved per cell was averaged for 10 cells. Data are expressedas mean ± s.e.m., ⁎p < 0.05; ⁎⁎p < 0.01, ⁎⁎⁎p<0.001 comparing TS16 with VO-1 or VO-2.



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Fig. 5.Induction of ADAMTS16 expression by TGFβ. Confluent C28/I2 cells were serum starved for24 h prior to addition of TGFβ for 6 h followed by harvest of RNA and qRT-PCR forADAMTS16. Data are expressed as mean ± s.e.m., ⁎p < 0.05; ⁎⁎p < 0.01.



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Fig. 6.Sequence alignment for ADAMTS16 putative promoter region. Sequence alignment of 1000 bpupstream of ADAMTS16 ATG from human, macaque, chimp and mouse (www.ensembl.org).⁎Indicates conserved regions. Translation start codon is shaded in grey with transcription startpoint (TSP) also marked. The 5′ positions of the deletion set of promoter–reporter constructsare also shown (arrows).



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent


Fig. 7.Identification of transcription start point in ADAMTS16 gene. 5′ RNA ligase-mediated (RLM)RACE was undertaken using C28/I2, human ovary or HeLa cell cDNA templates. (a) Anagarose gel of the PCR products is shown; (b) following sequencing of RACE products, thetranscription start point is marked as +1, translation start codon is shaded grey and position ofRACE primer is shown (arrow).



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Fig. 8.Promoter activity in the ADAMTS16 gene. SW1353 cells were transiently transfected withpromoter–reporter fragments in pGL3-basic as shown. Luciferase activity (measured asrelative light units) of untransfected cells (control) and transient transfections of empty vector(pGL3) and eight deletion constructs ranging from − 1802/+138 to +110/+138 are plotted asmean ± s.e.m.



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Fig. 9.Activation of the ADAMTS16 promoter by Egr1 and Sp1. SW1353 cells were transientlytransfected with promoter–reporter fragments in pGL3-basic as shown and co-transfected withexpression constructs for Egr1, Sp1 or empty vector. Luciferase activity is plotted as foldcompared to empty vector, mean ± s.e.m.; ⁎p < 0.05; ⁎⁎p < 0.01; ⁎⁎⁎p < 0.001. (a) schematicrepresentation of Egr1 and Sp1 consensus sequences in the ADAMTS16 promoter; (b) inductionof promoter by Egr1; (c) induction of promoter by Sp1.



Sponsored Docum

ent Sponsored D

ocument

Sponsored Docum

ent

Characterization and regulation of ADAMTS-16

Documents