Supplementary Figures1.Specificity of S100A4-Mediated Gene Transactivation in Tumor And

Supplementary Data

Supplementary FigureS1.Specificity of S100A4-mediated gene transactivation in tumor and immune cells.

(A)Primary bone marrow macrophages and(B) spleen T-cells were either mock-treated (CTL) or treated with the recombinant hS100A4 protein.Expression of SAA1 and SAA3 in macrophages and inT-cells was monitored by qPCR 6 hours later. Transcript levels are presented relative to non-treatedcells. (C) Possible contamination of the recombinant hS100A4 protein with bacterial endotoxin wastested by using Polymixin B. VMR cells were mock treated (CTL) or were treated with hS100A4protein, PMB or LPS as indicated. Expression of SAA3 was then assessed using qRT-PCR. The datawere normalized relative to mock-treated cells. PMB treatment resulted in nearly complete inhibition ofLPS-induced SAA3 expression, but had no significant effect on SAA3 expression in response tohS100A4. The experiments were performed independently two times, each with three replicate samples.(D) Loading controlsfor CM proteinscorresponding to Fig.1E. CM samples were run on 15% SDS-PAGE gels. In the case of a, the upper part of thegel was cut off and stained with Coomassie, while the lower portion of the gel was used forWestern blot analysis. In the case of b, c and d, the gels were blotted in their entirety, and then the upper parts of the Western blot filters were cut off and stained with Coomassie, while the lower parts were used for Western blot analysis.

Supplementary Figure S2.Effects of SAAs on cell adhesion and migration in vitro

SAAs augments the adhesive and migratory properties of tumor cells in vitro. (A)VMR cells were transduced with either empty vector (VMR/CTL), SAA1 (VMR/SAA1) or SAA3(VMR/SAA3) retroviruses. Expression of SAA1 and SAA3 in cell lysates and CM from these cells wasassessed by Western blot analysis. The data verify that transduction of VMR cells with SAA1 or SAA3retroviruses strongly augmented constitutive expression of these proteins.

(B) Adhesion of VMR cellstransducedwitheitheremptyvector (VMR/CTL),SAA1 (VMR/SAA1) or SAA3 (VMR/SAA3)retrovirusestocollagenI and laminin.Adherentcellswerequantifiedon thebasis of crystalvioletstainingandsubsequent absorbancemeasurements(OD550).One of two independent experiments is shown. (C) RecSAA proteins increaseCSML100 cell motilityin monolayerwound-healingassays. After 6.5hours incubation thedifferencein CSML100 migrationwas statisticallysignificant (p=0027 for recSAA1 and p=0.05 for recSAA3). (D) Human colon cancer SW480 cell monolayerswerewounded, thenincubatedwithCM from VMR cellstransducedwitheitheremptyvector(cmVMR) or SAA1(cmVMR/SAA1) and SAA3 (cmVMR/SAA3) retroviruses. Woundclosurewas monitoredovera8 h period.CM from VMR/SAA3 significantly increasedmotilityin monolayerwound-healingassays after 8 hours (p=0,0027). CM from VMR/SAA1 also stimulated cell motility. (E) MEFmonolayerswerewounded, thenincubatedwithCM from VMR cellstransducedwitheitheremptyvector(VMR/CTL),SAA1 (VMR/SAA1)or SAA3 (VMR/SAA3) retroviruses. Woundclosurewas monitoredovera6 h period.At 2h, thedifferencein MEF migrationwas statisticallysignificant(p=0.0251andp=0.0235for CM VMR/SAA1 andCM VMR/SAA3, respectively).Each data point represents the mean of triplicate samples. Four wounded area have been monitored in each sample. (F)3D culturesof MDA-MB-231 cells in Matrigel were stimulated withCM from VMR cells (cmVMR), VMR/SAA1 cells (cmVMR/SAA1), VMR/SAA3 cells (cmVMR/SAA3), orwith cmVMR supplemented with 5µg/ml recombinant human SAA1 (cmVMR+recSAA1). The CM contained 5% FCS.Areas of invasive outgrowth are measured between the two concentric white lines in each picture. A representative image analysis is presented. Quantification of this experiment is presented in Fig.2E.

Supplementary Figure S3. SAA-mediated stimulation of MMP expression and proteolytic activity

(A). Gelatin zymography assay showing that expression of SAA1 and SAA3 in VMR cells (VMR/SAA1 and VMR/SAA3, respectively augments the activity of MMPs in their CM compared to CM from the VMR/CTL cell line. Recombinant MMP2 and MMP9 (rMMP2 and rMMP9) served as positive controls. The upper arrow indicates the position of MMP9 proteolytic activity. The lower arrow indicates the position of MMP2 proteolytic activity. (B). MMP2, MMP3, MMP9 and MMP13 protein levels in CM from VMR cells treated with SAA1 or SAA3 was assessed by Western blot analysis. CM from S100A4 (A4) – treated VMR cells and mock- treated VMR cells (CTL) served as positive and negative controls, respectively. MMP9 protein levels were also increased in CM from VMR/SAA1 and VMR/SAA3 cells. PolymixinB(PMB) was includedintheexperiments where indicated toexclude possibleeffectsof endotoxin. (C). RecS100A4 and recSAA3 proteinsinduceMMP activity inVMR cells.Gelatinandcaseinzymography was used toshow increasedproteolyticactivityintheCM fromVMRcellstreatedwithS100A4 and SAA3 comparedtountreatedcontrols.PolymixinB(PMB) was includedintheexperimentstoexclude possibleeffectsof endotoxinintherecombinantSAA3 samples.TLR4andNFkB inhibitorswere employedtoexaminetheroleof thesesignalingmoleculesinSAA3-inducedinductionof MPP activity.Molecularweightmarkers(kDa) areindicatedon thelefthandsideof thezymographs. Arrows indicate the position of the induced proteolytic activity. (D). Casein zymography assay showing that SAA1 and SAA3 both induce a similar proteolytic activity in VMR cells (marked with an arrow). Representative Westernblot and zymography analysis from two or three independent experiments are presented.

Supplementary Figure S4. S100A4 does not induce SAA1 and SAA3 expression in CSML0 cells, but CSML0 cells respond to SAA3 by increasing proteolytic activity.

(A). Real-time qRT-PCR of transactivation of SAA1 and SAA3 after 6h treatment of VMR and CSML0 tumor cell lines with S100A4.Each data point represents the mean of triplicate samples. (B). Gelatin zymography of 15x concentrated CM from mock-, S100A4- and SAA-treated CSML0 and VMR cells. The size of the molecular weight markers is indicated.Representative zymography analysis from two independent experiments is presented.

Supplementary Figure S5. S100A4 regulates SAA expression via TLR-4 and NF-κB.

(A). Loading control for the Western blot shown in Fig. 4A. After gel electrophoresis the proteins werestained with Coomassie Blue. (B).Western blots of lysates of the cells were then probed with anti-phosphoERK antibodies. The data demonstrate the specificity of the inhibitors employed, as they exerted no effect on EGF-induced ERK phosphorylation.

Supplementary Figure S6. S100A4-mediated transcriptionalregulation of SAA1 and SAA3.

VMR cells were treated with S100A4 in the presence or absence of TLR4, NF-κB and EGFR inhibitors. Transcription of (A) SAA1 and (B) SAA3 relative to mock-treated (CTL) cells was assessed by qPCR.The experiments were performed independently two times, each with three replicate samples.