BMP activated Smad signaling strongly promotes migration and invasion of hepatocellular carcinoma cells

Experimental and Molecular Pathology 92 (2012) 74–81

Contents lists available at SciVerse ScienceDirect

Experimental and Molecular Pathology journal homepage: www.elsevier.com/locate/yexmp

BMP activated Smad signaling strongly promotes migration and invasion of hepatocellular carcinoma cells Ulrike Maegdefrau, Anja-Katrin Bosserhoff ⁎ Institute of Pathology, University of Regensburg, Germany

a r t i c l e

i n f o

Article history: Received 24 August 2011 Available online 15 October 2011 Keywords: Nidogen-2 Collagen XVI Bone morphogenetic proteins Hepatocellular carcinoma Smad signaling MAPK/ERK signaling

a b s t r a c t Several of the different bone morphogenetic proteins (BMPs) are involved in development and progression of specific tumors. For hepatocellular carcinoma (HCC) only BMP4 and BMP6 are described to be important for carcinogenesis. However, up to now neither the influence of other BMPs on tumor progression, nor the responsible signaling pathways to mediate target gene expression in HCC are known. In order to characterize BMP expression pattern in HCC cell lines, we performed RT-PCR analysis and revealed enhanced expression levels of several BMPs (BMP4, 6, 7, 8, 9, 10, 11, 13 and 15) in HCC. Thus, we treated HCC cells with the general BMP inhibitors chordin and noggin to determine the functional relevance of BMP overexpression and observed decreased migration and invasion of HCC cells. A cDNA microarray of noggin treated HCC cells was performed to analyze downstream targets of BMPs mediating these oncogenic functions. Subsequent analysis identified collagen XVI as ‘Smad signaling specific’ and nidogen-2 as ‘MAPK/ERK signaling specific’ BMP-target genes. To examine which signaling pathway is mainly responsible for the oncogenic role of BMPs in HCC, we treated HCC cells with dorsomorphin to determine the influence of BMP activated Smad signaling. Interestingly, also migratory and invasive behavior of dorsomorphin treated HCC cells was diminished. In summary, our findings demonstrate enhanced expression levels of several BMPs in HCC supporting enhanced migratory and invasive phenotype of HCC cells mainly via activation of Smad signaling. © 2011 Elsevier Inc. All rights reserved.

Introduction Bone morphogenetic proteins (BMP2 to BMP15) together with TGFß, activin and nodal belong to the transforming growth factor beta (TGFß) superfamily (Piek et al., 1999; Valcourt et al., 2005). Besides their role in embryonic development they are known to be important for tumor progression in different organs (Arnold et al., 1999; Buijs et al., 2010; Hogan, 1996). Very often not only expression of a specific BMP but rather different BMPs together fosters tumor progression (Gobbi et al., 2002; Thawani et al., 2010). For example, enhanced expression levels of BMP4, 6 and 7 could be detected in prostate carcinoma tissues, thereby promoting tumor progression (Bailey et al., 2007; Hamdy et al., 1997; Ye et al., 2007). Increased BMP2, 4 and 6 expression levels in osteo- and chondrosarcomas seems to be involved in growth stimulation of these tumors (Gobbi et al., 2002; Guo et al., 1999). In malignant melanoma cells, we found enhanced expression levels of BMP 4 and 7, whereby treatment of the cells with the general BMP inhibitor chordin leads to decreased migration and invasion of the cells (Rothhammer et al., 2005). Moreover, enhanced expression of BMP2 and 4 results in paracrine effects, as treatment of fibroblast with these BMPs causes enhanced MMP expression ⁎ Corresponding author at: Institute of Pathology, University of Regensburg, FranzJosef-Strauss-Allee 11, D-93053 Regensburg, Germany. Fax: + 49 941 944 6602. E-mail address: [email protected] (A.-K. Bosserhoff). 0014-4800/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.yexmp.2011.10.004

levels to promote tumor progression (Rothhammer et al., 2008). In the case of hepatocellular carcinoma (HCC), previous publications of our group exclusively described enhanced BMP4 and BMP6 expression levels and showed that BMP4 promotes the migratory and invasive behavior of HCC cells (Maegdefrau et al., 2009; Maegdefrau et al., in press). However, expression patterns of additional BMPs in HCC and their functional relevance are unknown so far. BMPs elicit their cellular effects via two distinct signaling pathways, the canonical Smad-dependent and the non-canonical Smadindependent pathway (MAPK/ERK). On the one hand, BMPs can bind to a preformed hetero-oligomeric BMP receptor complex (PFC) and activate the Smad signaling pathway via transphosphorylation of the receptors. On the other hand, binding of BMPs to BMP type I receptor leads to a BMP-induced signaling complex (BISC), which activates the MAPK/ERK signaling (Nohe et al., 2004; Sieber et al., 2009). Activation of the distinct BMP signaling pathways results in an expression of ‘signaling specific’ BMP-target gene sets. The most important direct target gene of an active BMP signaling is Id-1 (inhibitor of DNA binding/differentiation). Id-1 is a family member of helix-loophelix (HLH) proteins, which inhibits the function of HLH transcription factors. It is overexpressed in different cancer types and causes an aggressive, invasive tumor phenotype with improved blood supply through enhanced angiogenesis (Gautschi et al., 2008; Houldsworth et al., 2001; Sikder et al., 2003). However, apart from the direct ‘Smad signaling specific’ BMP-target gene Id-1, additional BMP-

U. Maegdefrau, A.-K. Bosserhoff / Experimental and Molecular Pathology 92 (2012) 74–81

target genes responsible for tumor progression, especially in HCC, are still unknown. Consequently, next to the analysis of the expression pattern and the role of BMPs in human HCC, the aim of this study was further to reveal BMP-target genes. Moreover, we were also interested in investigations on corresponding signaling pathways of the target genes, which are responsible for the biological role of BMPs during hepatocancerogenesis. Materials and methods Primary human hepatocytes Human liver tissue for cell isolation was obtained and experimental procedures were performed according to the guidelines of the charitable state controlled foundation HTCR, with the informed patient's consent. Hepatocytes were isolated using a modified twostep EGTA/collagenase perfusion procedure and cultured as described previously (Weiss et al., 2002). HCC cell lines The human HCC cell lines Hep3B (ATCC HB-8064), and PLC (ATCC CRL-8024) were cultured as described (Hellerbrand et al., 2008). RNA isolation and reverse transcription Total cellular RNA was isolated from cultured cells using the E.Z.N.A. ® Total RNA Kit I (Omega Bio-tek, VWR, Darmstadt, Germany) according to the manufacturer's instructions. cDNAs were generated by reverse transcriptase reaction (500 ng of total RNA) using SuperScript II Reverse Transcriptase Kit (Invitrogen, Groningen, the Netherlands). Expression analysis RT-PCR (reverse transcription-PCR) analysis of BMP2 to BMP15 was performed using specific primers (Table 1). The PCR reaction was performed in a 50 μl reaction volume containing 5 μl 10× Taqbuffer, 1 μl of cDNA, 1 μl of each primer (20 mM), 0.5 μl of dNTPs (10 mM), 0.5 Units of Taq polymerase and 41 μl of water. The amplification reactions were performed by 32 cycles of 1 min at 94 °C, 1 min at 62 °C and a final extension step at 72 °C for 1.5 min. The PCR products were resolved on 1.5% agarose gels. Nidogen-2 and collagen XVI alpha 1 (collagen XVI) total mRNA expression levels were analyzed by quantitative real time-PCR (qRT-

75

PCR) applying Lightcycler technology (Roche, Mannheim, Germany) as described (Rothhammer et al., 2005, 2007), using primers specified for qRT-PCR (Table 1). All experiments were repeated at least three times.

Transfection experiments Cells (2 × 10 5 per well) were seeded into 6-well plates and transfected with 0.5 μg Smad6 expression constructs using lipofectamine plus method (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Subsequent experiments were performed 24 h after transfection. All transfection experiments were repeated at least three times.

Stimulation of HCC cells Cells were seeded into six-well plates (2 × 10 5 HCC cells per well) and were treated with recombinant noggin (150 ng/mL, R&D systems, Wiesbaden, Germany) or dorsomorphin (2 μM, Sigma-Aldrich, Munich, Germany; DMSO treated control cells) for 6 h in normal High Glucose DMEM (PAN Biotech GmbH, Aidenbach, Germany). For inhibition of ERK1/2, cells were treated with UO126 (20 μM, Calbiochem, Nottingham, UK) for 24 h. Subsequently, mRNA was extracted as indicated above.

Migration and invasion assay Migration and invasion assays were performed as previously described (Rothhammer et al., 2004). Briefly, migration was assessed in Boyden chambers containing polycarbonate filters with 8 μm pore size (Costar, Bodenheim, Germany) coated with gelatin. The lower compartment was filled with fibroblast-conditioned medium used as a chemoattractant, and the filter was placed above. Treated HCC cells were harvested by trypsinization and resuspended in DMEM without FCS. Cell suspensions (800 μl) at a density of 3 × 10 4 cells/ml were placed in the upper compartment of the chambers. After incubation at 37 °C for 4 h filters were removed, cells were fixed, stained and counted. For invasion assays 2.5 × 10 5 cells/ml were used and filters were coated with a commercially available reconstituted basement membrane (Matrigel, diluted 1:3 in H2O; Becton Dickinson, Heidelberg, Germany). Each condition was assayed in triplicate, and assays were repeated at least twice.

Table 1 Primers used for quantitative RT-PCR analysis. name

ß-actin BMP2 BMP3 BMP4 BMP5 BMP6 BMP7 BMP8 BMP9 BMP10 BMP11 BMP12 BMP13 BMP14 BMP15 Id1 Coll XVI Nidogen-2

nucleotide sequence forward primer

reverse primer

5’- CTACGTGGCCCTGGACTTCGAGC -3’ 5’- TGGATTCGTGGTGGAAGTGGC -3’ 5’-GCAGATATTGGCTGGAGTGAATGGATTATCTC-3’ 5’- GATTCCCGTCCAAGCTATC -3’ 5’- TAAATCCAGCTCTCATCAGGACTCCTC-3’ 5’- AAGGCTGGCTGGAATTTGACATCACG -3’ 5’- GCCAGCCTGCAAGATAGCCATTTCC -3’ 5’- GTTAACATGGTGGAGCGAGAC -3’ 5’- CTGAGCACACCTTCAACCT -3’ 5’- CCAGTGCCCAGAATAAGCA -3’ 5’- CCGAGACCGTCATTAGCAT -3’ 5’- GCACGCAGAGGAAAGAGA -3’ 5’- CACCGTTGACGCATCTTGA -3’ 5’- CCCACCATTTCTCCTCACCT -3’ 5’- CGCCATCATCTCCAACTAAC -3’ 5’- TGTTACTCACGCCTCAAGGAG -3’ 5’- CCTGGTGCTGACACTACTGC -3’ 5’- CCGAGGTCTTCACGTATAATGC -3’

5’- GATGGAGCCGCCGATCCACACGG -3’ 5’- AGGGCATTCTCCGTGGCAGTA -3’ 5’-AGCGCAAGACTCTACTGTCATGTTAGGGTATA-3’ 5’- TCCATGATTCTTGACAGCC -3’ 5’- GAGATGGCATTTAATTTGGTTGGAGCAC -3’ 5’- GGTAGAGCGATTACGACTCTGTTGTC -3’ 5’- GAGCACCTGATAAACGCTGATCCGG -3’ 5’- GACTCCCTGTTGGACTGCTC -3’ 5’- TGGCAGTTATGGAGATGGC -3’ 5’- CCACCCAATCTCCTTGAAGT -3’ 5’- CAATCTTCAGTGAGCGGATAC -3’ 5’- AACGCAAAGGGAAGTCGC -3’ 5’- CAATGAAGGCAGAGACCTGG -3’ 5’- AGCCTCACACTCCTCAACTC -3’ 5’- GCTCAAGACCACCACTATCT -3’ 5’- TTCAGCGACACAAGATGCG -3’ 5’- AGCCACACTCAGCATCAGC -3’ 5’- GTGAAGTGCACGGGTGTATG -3’

76

U. Maegdefrau, A.-K. Bosserhoff / Experimental and Molecular Pathology 92 (2012) 74–81

Western blot analysis

Results

Protein extracts of cells were prepared as described (Rothhammer et al., 2005). For western blotting of pSmad1,5,8 (9511)/Smad5 (9517) (Cell Signaling Technology, Denvers, MA,USA) and pERK1/2 (4370)/ERK1/2 (9102) (Cell Signaling Technology) 30 μg RIPA-cell lysates were loaded, separated on 10% SDS-PAGE gels and subsequently blotted onto a PVDF membrane. After blocking for 1 h with 5% MP (dry milk)/TBST (100 mM Tris/HCl, 150 mM NaCl with 0,1% tween 20) the membrane was incubated for 16 h at 4 °C with the primary antibodies pSmad1,5,8 (1:1000)/Smad5 (1:1000) or pERK1/2 (1:2000)/ ERK1/2 (1:1000) in 5% BSA (bovine serum albumin)/TBST. Afterwards the membrane was washed three times in TBST, incubated for 1 h with a horseradish peroxidase-conjugated secondary anti-rabbit antibody (1:2000, Cell Signaling Technology) in 5% MP/TBST and then washed again. Finally, immunoreactions were visualized by Amersham ECL Plus™ Western Blotting Detection Reagent (GE Healthcare Europe GmbH, Freiburg, Germany) according to the manufacturer's instructions. All western blots were repeated three times.

BMP expression in hepatocellular carcinoma

Gene array PLC and Hep3B cells were treated with noggin (R&D systems) for 6 h and gene expression was analyzed via Gene 1.0 ST arrays (Affymetrix, Santa Clara, USA). The array was performed in cooperation with the official Affymetrix service provider KFP (Competence Center of Fluorescent Bioanalysis, Regensburg, Germany) and analyzed using the genomatix software (Genomatix Software GmbH, Munich, Germany). For gene clustering we used the functional annotation tools of the bioinformatic program DAVID (david.abcc.ncifcrf.gov). Statistical analysis Results are expressed as mean ± S.E.M (range) or percent. Comparison between groups was made using the Student's unpaired t-test. A p value b0.05 was considered statistically significant (ns: not significant). All calculations were performed using the GraphPad Prism software (GraphPad software Inc, San Diego, USA).

Previously, we have shown enhanced BMP4 and BMP6 expression levels in HCC (Maegdefrau et al., 2009, in press). Here, we further analyzed the mRNA expression pattern of BMP2 to BMP15 in HCC cell lines Hep3B and PLC and revealed enhanced expression levels of several BMPs (BMP4, 6, 7, 8, 9, 10, 11, 13 and 15) in HCC cell lines compared to primary human hepatocytes (PHH), whereas BMP2, 3, 5 and 14 were not differentially expressed (Fig. 1).

Effects of inhibition of BMP activity on migration and invasion of HCC cells To gain insight into the functional role of increased BMP expression levels in HCC, we studied effects of chordin and noggin on HCC cells. Both BMP antagonists interact with BMPs and inhibit their activity by sequestering BMP ligands in the extracellular space and prevent interactions with their membrane receptors (Groppe et al., 2003; Millet et al., 2001; Piccolo et al., 1996). Initially, we confirmed the inhibitory effect of chordin and noggin on BMPs secreted by HCC cells by measuring expression of Id1, a well known BMP responsive gene (Hollnagel et al., 1999) that has been implicated in the early steps of hepatocarcinogenesis (Matsuda et al., 2005). Treatments with either recombinant chordin or noggin significantly inhibited the expression of the target gene Id-1 (Supplementary Fig. 1A). Next, we investigated the effect of the two BMP ligand inhibitors on invasive and migratory potential of Hep3B and PLC cells in vitro. Boyden chamber assays revealed that invasion of HCC cells treated with recombinant chordin or noggin was significantly inhibited as compared to control cells (Fig. 2A). Furthermore, expression of chordin or noggin significantly inhibited the migratory potential of both Hep3B and PLC cells in vitro (Fig. 2B). Proliferation of HCC cells remained unchanged after chordin or noggin treatment (Fig. 2C).

Fig. 1. Expression of BMPs in HCC. A The expression of BMP2 to BMP15 mRNA was analyzed by RT-PCR in HCC cell lines Hep3B and PLC as well as primary human hepatocytes (PHH). Most BMPs (BMP4, 6, 7, 8, 9, 10, 11, 13 and 15) were stronger expressed in HCC cells compared to PHH cells.

U. Maegdefrau, A.-K. Bosserhoff / Experimental and Molecular Pathology 92 (2012) 74–81

Effects of noggin treatment on gene expression in HCC In addition, we aimed to determine BMP-target genes in HCC which could be responsible for the observed strong effects of BMP inhibition on migration and invasion of HCC cells in vitro. Since in both HCC cell lines effects of noggin were stronger than those of chordin we focused our attention on noggin. Following treatment with noggin in both HCC cell lines, 601 genes were reduced more than 1.5 fold (data not shown), and 58 genes were more than two fold reduced in Affymetrix gene expression arrays (Table 2). The functional annotation tools of the bioinformatic program DAVID showed that several of the 58 genes are involved in cell motility processes and are components of the extracellular matrix. Due to this clustering we analyzed the target genes collagen XVI and nidogen-2, which are components of the extracellular matrix and known to be important for migratory and invasive processes, in detail. To confirm array results, we performed qRT-PCR analysis of HCC cell lines treated with noggin and observed a down-regulation of these two target genes by noggin treatment (Fig. 3A and B). Determination of responsible BMP activated signaling pathways for target gene expression Two pathways (MAPK/ERK and Smad signaling) are known to be responsible for regulation of BMP-target genes (Nohe et al., 2004; Sieber et al., 2009). Noggin treatment of HCC cells causes inhibition of both pathways. Thus, to distinguish which BMP activated signaling pathway regulates the corresponding target gene, we treated HCC cells with dorsomorphin as Smad signaling specific inhibitor (Yu et al., 2008). Specificity of dorsomorphin for the Smad signaling pathway could be confirmed performing pSmad/Smad and pERK/ERK western blots of dorsomorphin treated HCC cell lines. As expected, we observed that dorsomorphin treatment leads to an inhibition of pSmad levels in contrast to the pERK levels, which remained unchanged (Supplementary Fig. 1B). Interestingly, qRT-PCR analysis of HCC cells treated with dorsomorphin showed only reduced collagen XVI expression levels, in contrast to nidogen-2, which was not differentially expressed (Fig. 3C and D). Thus, we suggest that the target gene collagen XVI is

77

regulated via the Smad signaling pathway and nidogen-2 via ERK signaling. To ensure that collagen XVI is regulated via the Smad signaling, we transfected HCC cells with the Smad signaling specific inhibitor Smad6 (Imamura et al., 1997) for 24 h. Collagen XVI mRNA expression levels were reduced in Hep3B and PLC cells after Smad6 transfection (Fig. 3C). We also determined nidogen-2 expression in Smad6 transfected cells and observed no significant changes (Fig. 3D). Furthermore, we also treated HCC cells with UO126, an ERK specific inhibitor (Favata et al., 1998). As expected, qRT-PCR analysis of collagen XVI did not show reduced expression levels after UO126 treatment, in contrast to strong decreased expression levels of nidogen-2 (Fig. 3E and F). Influence of Smad signaling specific inhibition of BMP activity on migration and invasion of HCC cells To analyze to what extent the BMP activated Smad signaling pathway effects the migratory and invasive behavior of HCC cells, we again treated HCC cells with dorsomorphin. Thereby, we distinguished the effects of the BMP activated Smad signaling from the ERK signaling cascade. In analogy to noggin treated HCC cells (see Fig. 2A and B) a decreased migratory and invasive phenotype of dorsomorphin treated cells could be observed (Fig. 4A and B). The proliferative behavior of Hep3B and PLC cells was not decreased (Fig. 4C). Thus, we suggest that the Smad signaling pathway and the corresponding target genes might be very important for the functional role of BMPs in HCC. Discussion Growth factors play an important role in physiological process like growth and differentiation of cells. However, dysregulation of growth factor expression is often the reason for tumor development and progression (Aaronson, 1991; Witsch et al., 2010). Bone morphogenetic proteins are known to be important for cancerogenesis of different organs. As described in the introduction, in most cases several distinct BMPs act together to promote the migratory and invasive potential of cancer cells (Gobbi et al., 2002; Thawani et al., 2010). Surprisingly, only few studies exist on the role of BMPs in hepatocellular carcinoma

Fig. 2. Functional role of BMPs in HCC. A & B Invasion (A) and migration (B) assays of noggin and chordin treated HCC cells were performed in the Boyden chamber model. Treatment with chordin and noggin, respectively, leads to a significantly decreased migratory and invasive potential of HCC cells compared to control treated cells. C Chordin and noggin treatment of HCC cells did not influence the proliferative behavior of the cells. (*: p b 0.05).

78

U. Maegdefrau, A.-K. Bosserhoff / Experimental and Molecular Pathology 92 (2012) 74–81

Table 2 Genes regulated more than 2fold by inhibition of BMP (gene regulation 2n). Gene Symbol

Gene name

Regulation [− fold]

ACTG2 AGTR2 ALDH1L1 COL16A1 COL4A2 COL4A4 CTSE DLGAP1 DNAH11 DNALI1 EBF2 EPB41L4B FCRL4 FGF13 GJA5 GNLY GPR128 ID1

actin, gamma 2, smooth muscle, enteric angiotensin II receptor, type 2 aldehyde dehydrogenase 1 family, member L1 collagen, type XVI, alpha 1 collagen, type IV, alpha 2 collagen, type IV, alpha 4 cathepsin E discs, large (Drosophila) homolog-associated protein 1 dynein, axonemal, heavy chain 11 dynein, axonemal, light intermediate chain 1 early B-cell factor 2 erythrocyte membrane protein band 4.1 like 4B Fc receptor-like 4 fibroblast growth factor 13 gap junction protein, alpha 5, 40 kDa granulysin G protein-coupled receptor 128 inhibitor of DNA binding 1, dominant negative helix-loop-helix protein inhibitor of DNA binding 4, dominant negative helix-loop-helix protein interferon, omega 1 integrin, alpha E (antigen CD103, human mucosal lymphocyte antigen 1; alpha polypeptide) potassium large conductance calcium-activated channel, subfamily M, beta member 1 potassium intermediate/small conductance calciumactivated channel, subfamily N, member 3 kallikrein-related peptidase 15 kallikrein-related peptidase 8 low density lipoprotein-related protein 1B (deleted in tumors) mannosidase, alpha, class 1 C, member 1 microtubule-associated protein tau myosin XVA N-acetylglutamate synthase nidogen 2 (osteonidogen) obscurin, cytoskeletal calmodulin and titin-interacting RhoGEF otoconin 90 ovo-like 1(Drosophila) protein kinase C and casein kinase substrate in neurons 1 phosphodiesterase 1B, calmodulin-dependent peptidoglycan recognition protein 3 plasminogen activator, tissue polymerase (RNA) II (DNA directed) polypeptide D PR domain containing 12 protein tyrosine phosphatase, receptor type, C regulator of chromosome condensation (RCC1) and BTB (POZ) domain containing protein 2 Rho-guanine nucleotide exchange factor R-spondin 3 homolog (Xenopus laevis) ryanodine receptor 2 (cardiac) S100 calcium binding protein A5 secretoglobin, family 3A, member 2 solute carrier family 8 (sodium/calcium exchanger), member 1 spermatogenesis and oogenesis specific basic helix-loop-helix 1 spectrin, alpha, erythrocytic 1 (elliptocytosis 2) spectrin repeat containing, nuclear envelope 1 spectrin repeat containing, nuclear envelope 2 transcription factor 4 transcobalamin I (vitamin B12 binding protein, R binder family) transition protein 1 (during histone to protamine replacement) tripartite motif-containing 54 titin Usher syndrome 2A (autosomal recessive, mild)

− 1.591 − 1.302 − 1.343 − 1.193 − 1.302 − 1.675 − 1.302 − 1.354 − 1.354 − 1.312 − 1.314 − 1.184 − 1.428 − 1.411 − 1.46 − 1.214 − 1.163 − 1.237

ID4 IFNW1 ITGAE KCNMB1 KCNN3 KLK15 KLK8 LRP1B MAN1C1 MAPT MYO15A NAGS NID2 OBSCN OC90 OVOL1 PACSIN1 PDE1B PGLYRP3 PLAT POLR2D PRDM12 PTPRC RCBTB2 RGNEF RSPO3 RYR2 S100A5 SCGB3A2 SLC8A1 SOHLH1 SPTA1 SYNE1 SYNE2 TCF4 TCN1 TNP1 TRIM54 TTN USH2A

− 1.322 − 1.389 − 1.457 − 1.235 − 1.474 − 1.357 − 1.509 − 1.634 − 1.397 − 1.267 − 1.607 − 1.305 − 1.522 − 1.407 − 1.261 − 1.425 − 1.343 − 1.498 − 1.685 − 1.568 − 1.283 − 1.527 − 1.466 − 1.43 − 1.246 − 1.329 − 1.749 − 1.453 −1.629 − 1.368 − 1.583 − 1.308 − 1.183 − 1.201 − 1.501 − 1.377 − 1.324 − 1.432 − 1.314 − 1.414

(Maegdefrau et al., 2009, in press; Qiu et al., 2010). Indeed, our previous publications showed that BMP4 and BMP6 expression levels are increased in HCC, but additional BMPs were not yet examined.

Moreover, we only showed a functional influence of enhanced BMP4 levels on migratory and invasive potential of HCC cells, whereby the general role of BMPs in hepatocancerogenesis remained unknown. For the first time, we now analyzed expression of all fourteen BMPs (BMP2-15) in HCC cell lines compared to primary human hepatocytes and showed enhanced expression levels of several BMPs (BMP4, 6, 7, 8, 9, 10, 11, 13 and 15) in HCC. Interestingly, inhibition of BMP expression with the general BMP inhibitors chordin and noggin led to decreased migration and invasion of HCC cells. These findings indicate that BMPs play an important role in hepatocancerogenesis. The target genes which mediate these tumorigenic effects are unknown so far. Until now only Id1 has been described as direct target gene of BMP-dependent Smad signaling (Gautschi et al., 2008; Houldsworth et al., 2001; Sikder et al., 2003). In view of the fact that BMPs not only activate Smad signaling but also MAPK/ERK signaling to exert their effects (Nohe et al., 2004; Sieber et al., 2009), we asked which signaling pathway and which corresponding target genes are responsible for the functional roles of BMPs in HCC. To identify new target genes of BMP signaling a cDNA microarray with HCC cells treated with BMP inhibitor noggin was performed. Remarkably, all regulated genes identified were down-regulated in noggin treated HCC cells (601 genes were reduced more than 1.5 fold and 58 genes were more than two fold reduced) suggesting that BMPs mainly lead to up-regulation of genes involved in tumor progression. Functional annotation tools of the bioinformatic program DAVID showed involvement of these genes in cell motility processes. Next to the direct target gene Id1, which causes an invasive tumor phenotype (Sikder et al., 2003) the program also detected further BMP-target genes like plasminogen activator as well as dynein and transition protein 1, which promote migratory and invasive processes in cells (Adham et al., 2001; Roda et al., 2009; Salvi et al., 2004; Suzuki et al., 2007). Moreover, the functional clustering of the gene list revealed that several genes are components of the extracellular matrix and are important for cell adhesion processes. Collagen XVI, as a minor component of the ECM (extracellular matrix), is strongly detectable in glioblastoma (Bauer et al., 2011). Interestingly, authors showed that enhanced collagen XVI expression leads to an increase in the invasive potential of the cancer cells, whereby proliferation was not affected. Also the target gene nidogen-2 is a component of the ECM (more closely, of the basement membrane). Nidogen-2 belongs to the nidogen family, which is known to have a promigratory activity on Schwann cells in the peripheral nerve system (Lee et al., 2007). Consequently, collagen XVI as well as nidogen-2 are components of matrix structures, which control a large number of cellular activities, like migration, differentiation, and chemotaxis of cells during morphogenesis as well as during cancerogenesis (Engbring & Kleinman, 2003; Kleinman et al., 2003; Lukashev & Werb, 1998). Therefore, we focussed our analysis on these both genes because it is conceivable that deregulation of the genes is also associated with enhanced migratory or invasive potential of HCC cells. We confirmed our array results by qRT-PCR analysis of noggin treated HCC cells. As expected, collagen XVI as well as nidogen-2 expression was reduced in noggin treated HCC cells. Indeed, in one qRTPCR analysis of the noggin treated HCC cell line (Hep3B) expression of collagen XVI was not significantly reduced. In this regard it should be noted that the expression pattern of the different BMPs (for example of BMP9, 14, etc.) varies in each HCC cell line, which possibly causes different collagen XVI expression levels in the cell lines. To identify the BMP activated signaling pathway responsible for BMP-target gene expression, we treated HCC cells with dorsomorphin (2 μM), a Smad signaling specific inhibitor. In contrast to Boergermann and colleagues, which stated that dorsomorphin (5 μM) inhibits the Smad as well as the MAPK signaling in mouse myoblasts (Boergermann et al., 2010), we showed a Smad specific inhibition of the BMP activated pathway in HCC cell lines. Surprisingly, only collagen XVI expression

U. Maegdefrau, A.-K. Bosserhoff / Experimental and Molecular Pathology 92 (2012) 74–81

79

Fig. 3. ‘Signaling specific’ BMP-target gene expression in HCC Affymetrix gene array of noggin (BMP inhibitor) treated HCC cells revealed 58 potential target genes of BMP signaling pathway. After functional clustering of the genes by the program DAVID two target genes (collagen XVI, nidogen-2) were analyzed in detail. QRT-PCR analyses were performed to confirm array results. A & B Reduced mRNA expression level of collagen XVI (A) and nidogen-2 (B) in noggin treated HCC cells compared to control treated HCC cells were observed. C Treatment with dorsomorphin (DM), a Smad specific inhibitor, leads to reduced expression levels of collagen XVI. To prove regulation of collagen XVI by the BMP activated Smad signaling, HCC cells were transfected with Smad6 (inhibitor of Smad signaling) for 24 h. QRT-PCR analysis showed reduced collagen XVI expression (A) in Smad6 transfected HCC cells. D Nidogen-2 mRNA expression levels were not reduced in dorsomorphin treated HCC cells. Moreover, qRT-PCR analysis of Smad6 transfected HCC cells compared to control transfected cells displayed unchanged nidogen-2 expression levels. E & F In contrast to collagen XVI, which is not differentially expressed in HCC cells after 24 h of UO126 (ERK inhibitor (UO)) treatment (E), nidogen-2 mRNA expression was reduced in UO126 treated Hep3B and PLC cells (F). (*: p b 0.05).

levels were reduced after dorsomorphin treatment but not nidogen-2 levels. We confirmed the ‘Smad signaling specific’ BMP-target gene expression, as cells transfected with Smad6 (as a Smad-Inhibitor) also displayed reduced collagen XVI levels. In UO126 (ERK inhibitor) treated cells only nidogen-2 expression was reduced. Therefore, we summarized that collagen XVI expression is mediated via Smad signaling, whereby nidogen-2 depends on activation of the MAPK/ERK signaling pathway. The MAPK/ERK signaling cascade is activated by a large panel of different molecules, in contrast to the Smad signaling, induced only by BMPs. Various publications showed that BMP activated Smad signaling is very important for migratory processes (Ketolainen et al., 2010; Sun et al., 2010; Virtanen et al., 2011). Therefore, we suggest that a BMP mediated activation of Smad signaling has the main impact on functional behaviour of HCC cells. Analogues to noggin treated HCC cells dorsomorphin treated HCC cells displayed an impaired migratory and invasive potential. This actually shows that BMP

activated Smad signaling seem to be the major pathway for functional behaviour of HCC cells. In summary, our studies demonstrate that BMPs are strongly expressed in hepatocellular carcinoma cells and promote migration and invasion of HCC cells. Consequently, BMPs have an important biological function in human HCC and appear as an attractive therapeutic target for this highly aggressive tumor. Conflict of Interest Statement The authors have no conflict of interest. Acknowledgments We are indebted to Ralph Weiskirchen (RWTH Aachen University Hospital, Germany) for providing the Smad6 construct and Susanne Wallner for excellent technical assistance. This work was supported

80

U. Maegdefrau, A.-K. Bosserhoff / Experimental and Molecular Pathology 92 (2012) 74–81

Fig. 4. Smad signaling specific functional role of BMPs. A - C In analogy to chordin and noggin treated HCC cells, dorsomorphin treated HCC cells showed decreased migratory (A) and invasive (B) behavior compared to control (DMSO) treated HCC cells. Proliferation (C) of HCC cells was not decreased after dorsomorphin treatment. (*: p b 0.05).

by grants from the Medical Faculty of the University of Regensburg (ReForM) and the German Research Foundation (DFG). Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.yexmp.2011.10.004. References Aaronson, S.A., 1991. Growth factors and cancer. Science 254, 1146–1153. Adham, I.M., Nayernia, K., Burkhardt-Gottges, E., et al., 2001. Teratozoospermia in mice lacking the transition protein 2 (Tnp2). Mol. Hum. Reprod. 7, 513–520. Arnold, S.F., Tims, E., Mcgrath, B.E., 1999. Identification of bone morphogenetic proteins and their receptors in human breast cancer cell lines: importance of BMP2. Cytokine 11, 1031–1037. Bailey, J.M., Singh, P.K., Hollingsworth, M.A., 2007. Cancer metastasis facilitated by developmental pathways: Sonic hedgehog, Notch, and bone morphogenic proteins. J. Cell. Biochem. 102, 829–839. Bauer, R., Ratzinger, S., Wales, L., et al., 2011. Inhibition of collagen XVI expression reduces glioma cell invasiveness. Cell. Physiol. Biochem. 27, 217–226. Boergermann, J.H., Kopf, J., Yu, P.B., Knaus, P., 2010. Dorsomorphin and LDN-193189 inhibit BMP-mediated Smad, p38 and Akt signalling in C2C12 cells. Int. J. Biochem. Cell Biol. 42, 1802–1807. Buijs, J.T., Petersen, M., Vander, H.G., Vander, P.G., 2010. Bone morphogenetic proteins and its receptors; therapeutic targets in cancer progression and bone metastasis? Curr. Pharm. Des. 16, 1291–1300. Engbring, J.A., Kleinman, H.K., 2003. The basement membrane matrix in malignancy. J. Pathol. 200, 465–470. Favata, M.F., Horiuchi, K.Y., Manos, E.J., et al., 1998. Identification of a novel inhibitor of mitogen-activated protein kinase kinase. J. Biol. Chem. 273, 18623–18632. Gautschi, O., Tepper, C.G., Purnell, P.R., et al., 2008. Regulation of Id1 expression by SRC: implications for targeting of the bone morphogenetic protein pathway in cancer. Cancer Res. 68, 2250–2258. Gobbi, G., Sangiorgi, L., Lenzi, L., et al., 2002. Seven BMPs and all their receptors are simultaneously expressed in osteosarcoma cells. Int. J. Oncol. 20, 143–147. Groppe, J., Greenwald, J., Wiater, E., et al., 2003. Structural basis of BMP signaling inhibition by Noggin, a novel twelve-membered cystine knot protein. J. Bone Joint Surg. Am. 85-A (Suppl 3), 52–58. Guo, W., Gorlick, R., Ladanyi, M., et al., 1999. Expression of bone morphogenetic proteins and receptors in sarcomas. Clin. Orthop. Relat. Res. 175–183. Hamdy, F.C., Autzen, P., Robinson, M.C., Horne, C.H., Neal, D.E., Robson, C.N., 1997. Immunolocalization and messenger RNA expression of bone morphogenetic protein-6 in human benign and malignant prostatic tissue. Cancer Res. 57, 4427–4431. Hellerbrand, C., Amann, T., Schlegel, J., et al., 2008. The novel gene MIA2 acts as a tumour suppressor in hepatocellular carcinoma. Gut 57, 243–251.

Hogan, B.L., 1996. Bone morphogenetic proteins in development. Curr. Opin. Genet. Dev. 6, 432–438. Hollnagel, A., Oehlmann, V., Heymer, J., Ruther, U., Nordheim, A., 1999. Id genes are direct targets of bone morphogenetic protein induction in embryonic stem cells. J. Biol. Chem. 274, 19838–19845. Houldsworth, J., Reuter, V.E., Bosl, G.J., Chaganti, R.S., 2001. ID gene expression varies with lineage during differentiation of pluripotential male germ cell tumor cell lines. Cell Tissue Res. 303, 371–379. Imamura, T., Takase, M., Nishihara, A., et al., 1997. Smad6 inhibits signalling by the TGFbeta superfamily. Nature 389, 622–626. Ketolainen, J.M., Alarmo, E.L., Tuominen, V.J., Kallioniemi, A., 2010. Parallel inhibition of cell growth and induction of cell migration and invasion in breast cancer cells by bone morphogenetic protein 4. Breast Cancer Res. Treat. 124, 377–386. Kleinman, H.K., Philp, D., Hoffman, M.P., 2003. Role of the extracellular matrix in morphogenesis. Curr. Opin. Biotechnol. 14, 526–532. Lee, H.K., Seo, I.A., Park, H.K., et al., 2007. Nidogen is a prosurvival and promigratory factor for adult Schwann cells. J. Neurochem. 102, 686–698. Lukashev, M.E., Werb, Z., 1998. ECM signalling: orchestrating cell behaviour and misbehaviour. Trends Cell Biol. 8, 437–441. Maegdefrau, U., Amann, T., Winklmeier, A., et al., 2009. Bone morphogenetic protein 4 is induced in hepatocellular carcinoma by hypoxia and promotes tumour progression. J. Pathol. 218, 520–529. Maegdefrau, U., Arndt, S., Kivorski, G., Hellerbrand, C., Bosserhoff, A.K., in press. Downregulation of hemojuvelin prevents inhibitory effects of bone morphogenetic proteins on iron metabolism in hepatocellular carcinoma. Lab. Invest. doi:10.1038/ labinvest2011.123. Matsuda, Y., Yamagiwa, S., Takamura, M., et al., 2005. Overexpressed Id-1 is associated with a high risk of hepatocellular carcinoma development in patients with cirrhosis without transcriptional repression of p16. Cancer 104, 1037–1044. Millet, C., Lemaire, P., Orsetti, B., Guglielmi, P., Francois, V., 2001. The human chordin gene encodes several differentially expressed spliced variants with distinct BMP opposing activities. Mech. Dev. 106, 85–96. Nohe, A., Keating, E., Knaus, P., Petersen, N.O., 2004. Signal transduction of bone morphogenetic protein receptors. Cell. Signal. 16, 291–299. Piccolo, S., Sasai, Y., Lu, B., De Robertis, E.M., 1996. Dorsoventral patterning in Xenopus: inhibition of ventral signals by direct binding of chordin to BMP-4. Cell 86, 589–598. Piek, E., Heldin, C.H., Ten Dijke, P., 1999. Specificity, diversity, and regulation in TGFbeta superfamily signaling. FASEB J. 13, 2105–2124. Qiu, H., Yang, B., Pei, Z.C., Zhang, Z., Ding, K., 2010. WSS25 inhibits growth of xenografted hepatocellular cancer cells in nude mice by disrupting angiogenesis via blocking bone morphogenetic protein (BMP)/Smad/Id1 signaling. J. Biol. Chem. 285, 32638–32646. Roda, O., Ortiz-Zapater, E., Martinez-Bosch, N., et al., 2009. Galectin-1 is a novel functional receptor for tissue plasminogen activator in pancreatic cancer. Gastroenterology 136, 1375–1379. Rothhammer, T., Hahne, J.C., Florin, A., et al., 2004. The Ets-1 transcription factor is involved in the development and invasion of malignant melanoma. Cell. Mol. Life Sci. 61, 118–128. Rothhammer, T., Poser, I., Soncin, F., Bataille, F., Moser, M., Bosserhoff, A.K., 2005. Bone morphogenic proteins are overexpressed in malignant melanoma and promote cell invasion and migration. Cancer Res. 65, 448–456.

U. Maegdefrau, A.-K. Bosserhoff / Experimental and Molecular Pathology 92 (2012) 74–81 Rothhammer, T., Bataille, F., Spruss, T., Eissner, G., Bosserhoff, A.K., 2007. Functional implication of BMP4 expression on angiogenesis in malignant melanoma. Oncogene 26, 4158–4170. Rothhammer, T., Braig, S., Bosserhoff, A.K., 2008. Bone morphogenetic proteins induce expression of metalloproteinases in melanoma cells and fibroblasts. Eur. J. Cancer 44, 2526–2534. Salvi, A., Arici, B., De Petro, G., Barlati, S., 2004. Small interfering RNA urokinase silencing inhibits invasion and migration of human hepatocellular carcinoma cells. Mol. Cancer Ther. 3, 671–678. Sieber, C., Kopf, J., Hiepen, C., Knaus, P., 2009. Recent advances in BMP receptor signaling. Cytokine Growth Factor Rev. 20, 343–355. Sikder, H.A., Devlin, M.K., Dunlap, S., Ryu, B., Alani, R.M., 2003. Id proteins in cell growth and tumorigenesis. Cancer Cell 3, 525–530. Sun, Y., Fei, T., Yang, T., et al., 2010. The suppression of CRMP2 expression by bone morphogenetic protein (BMP)-SMAD gradient signaling controls multiple stages of neuronal development. J. Biol. Chem. 285, 39039–39050. Suzuki, S.O., McKenney, R.J., Mawatari, S.Y., et al., 2007. Expression patterns of LIS1, dynein and their interaction partners dynactin, NudE, NudEL and NudC in

81

human gliomas suggest roles in invasion and proliferation. Acta Neuropathol. 113, 591–599. Thawani, J.P., Wang, A.C., Than, K.D., Lin, C.Y., La Marca, F., Park, P., 2010. Bone morphogenetic proteins and cancer: review of the literature. Neurosurgery 66, 233–246. Valcourt, U., Kowanetz, M., Niimi, H., Heldin, C.H., Moustakas, A., 2005. TGF-beta and the Smad signaling pathway support transcriptomic reprogramming during epithelial-mesenchymal cell transition. Mol. Biol. Cell 16, 1987–2002. Virtanen, S., Alarmo, E.L., Sandstrom, S., Ampuja, M., Kallioniemi, A., 2011. Bone morphogenetic protein −4 and −5 in pancreatic cancer-Novel bidirectional players. Exp. Cell Res. Weiss, T.S., Jahn, B., Cetto, M., Jauch, K.W., Thasler, W.E., 2002. Collagen sandwich culture affects intracellular polyamine levels of human hepatocytes. Cell Prolif. 35, 257–267. Witsch, E., Sela, M., Yarden, Y., 2010. Roles for growth factors in cancer progression. Physiology (Bethesda) 25, 85–101. Ye, L., Lewis-Russell, J.M., Kynaston, H., Jiang, W.G., 2007. Endogenous bone morphogenetic protein-7 controls the motility of prostate cancer cells through regulation of bone morphogenetic protein antagonists. J. Urol. 178, 1086–1091. Yu, P.B., Hong, C.C., Sachidanandan, C., et al., 2008. Dorsomorphin inhibits BMP signals required for embryogenesis and iron metabolism. Nat. Chem. Biol. 4, 33–41.