| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Department of Physiological Chemistry I, Biocenter (Theodor-Boveri Institut), University of Würzburg, Am Hubland, 97074 Würzburg, Germany
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
vß3 known to be involved in melanoma cell survival and growth was identified as an OPN receptor, which points to an OPN-mediated cross-talk between growth factor receptors and this integrin receptor in melanocytes. In summary, we could show that in melanocytes growth factor receptor-induced secretion of OPN can promote antiapoptotic signaling and mediate appropriate interactions with the extracellular matrix in an autocrine way. Our findings suggest a new role of growth factor receptors of the family of receptor tyrosine kinases in processes associated with melanoma development and progression. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A crucial difference between nontransformed and transformed pigment cells is that normal melanocytes cannot survive for longer periods in the dermal environment but melanoma cells can. The survival and continuous growth of transformed cells in the dermal tissue are important steps in melanoma progression (2)
. Survival of pigment cells in the dermis seems to be dependent on the presence of cell surface molecules that regulate the contact between the cells and the dermal ECM, i.e., components of the dermal connective tissue such as collagen. A surface molecule, which can mediate the adhesion of melanoma cells to dermal collagen thereby inducing antiapoptotic signaling, is the integrin receptor vß3 (3
, 4)
. The expression of vß3 has been directly linked to neoplastic progression and tumorigenicity of melanoma (5
, 6)
. Also, bFGF, a growth factor most uniformly expressed by melanoma cells but not by normal melanocytes (7)
, is able to counteract apoptosis induced in the presence of type I collagen. Addition of this melanoma-associated factor to melanocytes cultured in three-dimensional collagen gels, which mimic the conditions in the dermis in vitro, prevented apoptosis (8)
. However, the mechanism by which bFGF exerts its positive effect on melanocyte survival is not clear yet.
The receptor for bFGF is a receptor tyrosine kinase that triggers proliferation in melanocytes as well as in melanoma cells, where it is stimulated in an autocrine way (9 , 10) . Similarly, the receptor tyrosine kinase Xmrk, the overexpression of which is the initiatory event of the formation of highly invasive melanoma in Xiphophorus (11) , induces mitogenic signaling in melanocytes and in melanoma cells (12 , 13) . Remarkably, activation of the Xmrk receptor does not only trigger proliferation, it also stimulates antiapoptotic signaling. We have established a cell system in which Xmrk-specific antiapoptotic and mitogenic signaling can be dissected (Ba/F3 cells; Ref. 14 ). Using this cell system, we identified genes that have an expression that is regulated either by mitogenic or by antiapoptotic signaling triggered by the melanoma-inducing receptor.
One of the identified Xmrk target genes, with an expression that is correlated with antiapoptotic signaling in Ba/F3 cells, is opn. Additional investigation in pigment cells revealed that OPN can promote survival of melanocytes in three-dimensional gels of the ECM component collagen. This attributes to OPN a new and relevant role in processes correlating with pigment cell transformation and tumor progression in melanoma development.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Differential Display Analysis.
For identification of Xmrk target genes, a modified differential display procedure according to Liang et al. (19, 20, 21)
was performed. Amplifications and subsequent analyses were done in parallel from three independent cell cultures for each condition. DNase-digested total RNA (2.5 µg) from unstimulated and 2-h EGF-stimulated BaF Hm cells were used for RT using one-base-anchored oligodeoxythymidylate primers. Subsequent PCR amplifications of 1 µl of the RT reaction with the appropriate oligodeoxythymidylate primer in combination with an arbitrary 10-mer primer using [32P]dCTP were performed. The PCR reactions were separated on a 6% polyacrylamide sequencing gel. Gels were dried and exposed to X-ray films. Differentially displayed bands were excised from the gel, and one-third of a gel slice was directly used for reamplification with the primer combinations of the original differential display PCR. PCR products were cloned, using the SureClone kit (Amersham, United Kingdom) and subsequently sequenced. GCG software was applied to compare sequences to Genbank and European Molecular Biology Laboratory databases. The differential expression of the respective genes was confirmed by Northern blot analysis using probes generated by random hexamer labeling of the gel-purified plasmid insert.
RT-PCR and Northern Blot Analysis.
After stimulation and/or inhibitor treatment, RNA was isolated with Trizol reagent (Life Technologies, Inc., Karlsruhe, Germany) following the instructions of the supplier. First-strand cDNA synthesis was performed by using Superscript RT reverse transcriptase (Life Technologies, Inc.), using 1 µg total RNA and random hexanucleotides (0.5 µM) in 20 µl following the suppliers protocol. RT reaction (0.5 µl) was then used to amplify the respective integrine genes: v 5'-ATCCGACGAGCACTGTTTCT-3' and 5'-TCCATCTCTGACTGCTGGTG-3' (30 cycles); and ß3 5'-GCAGTGTCTGGCTGTGAGTC-3' and 5'-GGGACACGCTCTGTTTCTTC3' (36 cycles). For Northern blot analysis, total RNA was blotted onto Hybond-N nylon membranes (Amersham, United Kingdom). Prehybridization and hybridization were carried out at 42°C in 50% formamide. Filters were hybridized with a 32P-labeled mouse OPN probe that was generated by random priming with [32P]dCTP. The probes used were a 550-bp opn differential display-PCR-fragment or a mouse cDNA probe, which was kindly provided by Dr. L. W. Fisher (22)
.
Western Blot Analysis.
After stimulation for the indicated times, cells were rinsed twice with PBS and lysed in 50 mM Tris (pH 7.5), 150 mM NaCl, 0.5% NP40, 1 mM phenylmethylsulfonyl, 1 mM fluoride Na3VO4, 10 µg/ml leupeptin, and 10 µg/ml aprotinin for 20 min at 4°C. Proteins were separated by SDS-PAGE and analyzed by Western blotting onto nitrocellulose. Filters were blocked for 10 min with 10 mM Tris-Cl (pH 7.9), 0.5% Tween, and 1.5% BSA and were incubated at least for 60 min with the first antibody. Horseradish peroxidase-coupled second antibodies were used for nonradioactive detection. Monoclonal anti-phospho MAPK (clone 12D4) was from Nanotools (Teningen, Germany). Anti-ERK-2 (C-14), antimouse OPN (P-18), antihuman OPN (K-20), anti-v (Q-20), and anti-ß3 (N-20) were from Santa Cruz Biotechnology (Santa Cruz, CA).
Preparation of Collagen Gels and Analysis of Cell Spreading.
Collagen gels were prepared using Cellagen solution DMEM (pH 7.4) containing type I collagen (ICN Biomedicals, Eschwege, Germany). Cells were seeded into 24-well plates (4 x 104/well) and suspended in the Cellagen solution. After gelation, either melan-a HER-mrk CM or FCS (5%) containing medium with the respective factors was added. The concentrations were as follows: 8 nM EGF; 0.3 nM bFGF; 100 nM OPN; and 200 nM TPA. Cell morphology and spreading were examined hourly the first 12 h, and cells were photographed after 48 h.
Adhesion Assay.
The adhesion assay was performed according to Smith et al. (23)
. A 96-well ELISA plastic plate (Greiner, Frickenhausen, Germany) was coated with OPN (100 nM) or BSA (100 nM) overnight at 4°C. The wells were washed twice with PBS and blocked with PBS/1% BSA for 1 h at 37°C. Melan-a cells were preincubated with either anti-v (clone H9.2B8; Pharmingen, BD Biosciences, Germany), anti-ß3 (clone 2C9.G2; Pharmingen, BD Biosciences), an unspecific control antibody or without antibody in DMEM/0.1% BSA for 15 min at RT, and 5 x 104 cells/96-well were allowed to adhere to the substrate at 37°C for 2 h. Nonadhering cells were washed off with PBS, and the adhering cells were fixed in 4% paraformaldehyde/PBS for 1 h at RT and stained with 0.5% toluidine blue in 4% paraformaldehyde for 1 h at RT. After three washes with PBS, the color was solubilized using 1% SDS, and the color intensity was measured with an ELISA reader using a filter of 595 nm.
Determination of Apoptotic Cells.
Cells (2 x 105) suspended in Cellagen solution were seeded on chamber slides (Nunc,), and after gelation, the gels were overlayed with DMEM containing 5% FCS without any additives or supplemented with the indicated factors (concentrations see above). For the blocking experiments, either a neutralizing anti-OPN antibody, kindly provided by Dr. Hsin-Fang Yan-Yen (24)
, was added to the medium at a concentration of 10 µl of serum/ml or anti-v (clone H9.2B8) or anti-ß3 (clone 2C9.G2) was added to the preincubated cells in a concentration of 5 µg/ml. As control, antibody anti-MAP kinase phosphatase 1 (Santa Cruz Biotechnology, Heidelberg, Germany) was used at a concentration of 10 µg/ml. After 4 days, the cells were fixed in formaldehyde for at least 30 min and after washing with PBS incubated in 0.1% Triton X-100 for 30 min. Cells were washed again in PBS, and the DNA was stained with 1 µg/ml bisbenzimide (Hoechst 33342) in PBS. After embedding the fixed cells in mounting medium, they were analyzed using an UV microscope. Condensed and fragmented nuclei were counted in a total of 1000 cells from each sample, and the amount of fragmented nuclei was calculated in percentages.
DNA Fragmentation.
Cells (4 x 106) were seeded in Cellagen on 10-cm culture dishes and cultured for 4 days under different conditions. Cell-containing gels were washed with PBS, and the cells were scraped off the dish. After centrifugation, collagen-containing cell pellets were lysed in 500 µl lysis buffer (0.2% SDS, 100 mM EDTA, and 200 mM NaCl) complemented with proteinase K (100 µg/ml) and then incubated at 80°C for 1 h. After phenol/chloroform extraction using SST tubes (BD Biosciences, Heidelberg, Germany), the DNA was recovered by ethanol precipitation. The DNA was dissolved in water, complemented with RNase A (500 µg/ml) and gel-loading buffer, and analyzed on an agarose gel.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
For a differential display analysis of Xmrk target genes, quiescent BaF Hm cells were treated for 2 h with EGF, and RNAs were analyzed by differential display RT-PCR for differentially expressed genes in EGF-treated and nontreated cells. One of the PCR products was found at particularly high levels after amplification of cDNA from EGF-treated cells. The respective fragment was therefore isolated from the gel, reamplified, cloned, and sequenced. The sequence matched 100% to the mouse opn gene. Differential expression of opn in BaF Hm cells was confirmed by Northern blot analysis. EGF treatment of BaF Hm cells led to a rapid up regulation of OPN, which lasted up to 14 h (Fig. 1A)
. Also mIL-3-induced opn expression in BaF Hm cells, although the expression level was significantly weaker (Fig. 1A)
.
|
1006 and BaF Hm 1075, Ref. 14
). Hm 1006 possesses no tyrosine that can be phosphorylated after receptor stimulation, and it is incapable of activating the MAPK pathway (Fig. 1B)
. Hm 1075 possesses a growth factor receptor binding protein 2 binding site and activates MAPK such as the full-length HER-mrk receptor. Activation of MAPK was detectable by its phosphorylation using anti-phospho MAPK (Fig. 1B)
. The induction of opn expression by Hm 1006 was clearly affected (Fig. 1C)
, and the fact that Hm 1075 efficiently up-regulated the opn RNA level (Fig. 1C)
points to a role of MAPK in regulation of opn expression by HER-mrk. Interestingly, Hm 1006 in contrast to Hm 1075 is inefficient in mediating antiapoptotic signaling in Ba/F3 cells (14)
, confirming an involvement of OPN in cell survival as it was described in this cell system by Lin et al. (24)
.
Xmrk-induced opn Expression in Melanocytes Is Dependent on MAPK Activation.
Since activation of the Xmrk receptor induces melanoma formation in fish, we asked whether the Xmrk kinase also induces opn expression in pigment cells and which role osteopontin plays in these cells. Therefore, we analyzed opn expression after stimulation of the chimeric Xmrk receptor in HER-mrk-expressing mouse melanocytes melan-a Hm. These cells proliferate, dedifferentiate, and show a transformed phenotype in response to receptor stimulation (13)
. Stimulation of melan-a Hm cells with either EGF or with TPA, a general growth factor for melanocytes in vitro, led to strong activation of the MAPKs ERK1 and 2 (Fig. 2A)
. This activation was almost completely abolished in the presence of the MAP/ERK kinase inhibitor PD98059 (Fig. 2A)
.
|
. Nearly the same level of induction could be seen when melan-a Hm cells were stimulated with TPA alone confirming, the fact that opn originally has been identified as a TPA-inducible gene (25)
. Because TPA activates MAPK in melan-a cells (Fig. 2A)
, MAPK might be involved in up regulation of opn expression in melanocytes. Using PD98059, we blocked MAPK activity during HER-mrk stimulation. In the presence of PD98059, no induction of opn could be detected (Fig. 2B)
, pointing to a role of MAPK in HER-mrk-induced opn expression in melanocytes.
OPN Protein Expression Is Induced by Xmrk as well as by the FGF Receptor in Melanocytes and Is Constitutively Enhanced in Melanoma Cells.
The expression of the OPN in melan-a Hm cells was analyzed after stimulation of the HER-mrk receptor using a polyclonal anti-OPN antibody. OPN was almost completely absent in cells that were factor starved (no TPA, no EGF) for 24 h, pointing to a fast turnover of the protein in these cells. After 2 h of receptor stimulation, OPN was clearly detectable as a Mr 55,000 protein, and this protein expression increased even more during the next 16 h (Fig. 3A)
.
|
, indicating that opn is also a target for the signaling induced by the FGF receptor. However, HER-mrk stimulation appeared to be more effective in inducing OPN expression. In contrast to melanocytes where OPN basal expression was almost not detectable, constitutive-enhanced expression of OPN was found in several different human melanoma cell lines (Fig. 3C)
. Interestingly, the lowest expression was seen in WM35 cells (Fig. 3C)
, which are derived from a noninvasive RGP melanoma (26)
.
OPN Contributes to Adhesion of Melanocytes to Type I Collagen.
OPN is known to function both as cytokine as well as cell attachment protein with integrins vß3, vß1 and vß5 acting as OPN receptors (27)
. In addition, OPN can bind to the ECM component collagen, including type I collagen (28)
. Melanocytes, which grow in the epidermis, normally do not get into contact with type I collagen, which is an ubiquitous component of the connective tissue of the dermis. Moreover, melanocytes are not able to adhere properly and spread over a long-term period in three-dimensional type I collagen gels, which simulate the conditions in the dermis in vitro (8)
. Strikingly, bFGF was found to induce spreading of melanocytes in collagen gels, although the underlying mechanism is not clear yet.
In melan-a cells, OPN expression was substantially increased in response to HER-mrk activation and also by stimulation with bFGF. Because OPN is described to act as a mediator between integrin receptors and collagens (29)
, we wondered whether it might increase the capacity of melanocytes to adhere to type I collagen and to spread in three-dimensional type I collagen gels. Wild-type melan-a cells were embedded in type I collagen gels, and the cells were cultured in the presence of different factors. Cells in the presence of serum alone and no additional factors initially showed increased spreading during the first 8 h. However, between 30 and 48 h, cell rounding increased, and the cells lost the spread phenotype (Fig. 4A)
. In contrast, cells cultured in the presence of TPA, which is a general growth factor for melanocytes in vitro, spread much faster, and the spread dentritic phenotype lasted up to 48 h. Also bFGF (0.3 nM), which is described to mediate spreading of human melanocytes in type I collagen gels (8)
, induced this phenotype in melan-a cells. Strikingly, the addition of 100 nM OPN to the medium also increased the fraction of spread cells in the gel (Fig. 4A)
.
|
, pointing to factors in the HER-mrk CM that enhance interactions of the cells with the collagen matrix. That the observed cell spreading was not due to EGF present in the conditioned medium was seen in wild-type melan-a cells cultured in medium containing 8 nM EGF and FCS (Fig. 4A)
. These cells showed the same round phenotype as cells cultured in the presence of FCS alone.
Although OPN and HER-mrk CM could prevent cell rounding induced by collagen over a long-term period, they did not induce an increase in the cell number. TPA and bFGF, in contrast, very efficiently triggered cell growth in the collagen matrix. To verify that OPN, EGF, and factors in HER-mrk CM do not trigger mitogenic signaling in melan-a cells, the growth response of melan-a cells was assessed by counting the cells after 4 days in medium containing the different factors. TPA, as well as bFGF, clearly induced cell growth and led to a 3–4-fold increase. However, neither OPN, EGF, or factors in HER-mrk CM produced a significant mitogenic effect (Fig. 4B)
.
OPN Counteracts Type I Collagen-induced Apoptosis in Melanocytes.
Melan-a cells not only showed inefficient spreading in collagen gels, they also started to apoptose in the absence of any additional factors. Already after 30 h in medium containing only FCS, the first apoptotic cells could be detected, and the number increased dramatically between days 2 and 4. It should be mentioned that this is not a general response of melan-a cells to the absence of defined growth factors since cells cultured under these conditions on conventional culture plates for 6 days, although showing a starved phenotype, do not apoptose (13
, 18)
.
To quantify the antiapoptotic effect of OPN on melanocytes, cells were stained with DAPI after 4 days in collagen (Fig. 5A)
, and the ratio of intact to fragmented nuclei was determined (Fig. 5B)
. The effects quantified by DAPI staining were confirmed by analysis of DNA fragmentation (Fig. 5C)
clearly showing that apoptosis is induced by type I collagen in melan-a cells.
|
. Nuclei were also intact in cells cultured in HER-mrk CM, but the total number of nuclei was lower. Confirming the apoptotic phenotype seen in cells cultured in FCS alone, most of the nuclei were fragmented (Fig. 5A)
, and 80–90% of the cells were apoptotic (Fig. 5B)
. In contrast, TPA almost completely prevented collagen-induced apoptosis and only 5–10% of the cells showed fragmented nuclei. Also, OPN suppressed cell death in the collagen gel with only 40% of the cells undergoing apoptosis. The conditioned medium HER-mrk CM was almost as efficient as TPA in reducing the number of apoptotic cells (13%). Using a neutralizing antibody to OPN (24)
, the protective effect of HER-mrk CM was significantly reduced. Approximately 60% of the cells were apoptotic (Fig. 5B)
, which clearly shows that OPN contributes to the antiapoptotic effect of the HER-mrk CM. Taken together, these data demonstrate that OPN is able to counteract apoptosis induced in melanocytes by collagen.
Integrin vß3 Acts as Receptor for OPN in Melanocytes and Contributes to the Survival Effect Produced by bFGF.
Because it is known that the vitronectin receptor vß3 is also a receptor for OPN (27)
, and vß3 has been linked to the tumorigenicity of malignant melanoma (6)
, we asked whether this integrin might also act as a receptor for OPN in melanocytes.
Using specific primers for the respective integrin subunits, the expression of v as well as of ß3 was readily detectable in melan-a cells by RT-PCR analysis (Fig. 6A)
. Specific antibodies against v and ß3 identified the respective proteins in melan-a cells (Fig. 6B)
. Strikingly, v expression was much higher in melan-a cells than in the human melanoma cell lines IFB and A375. This was not attributable to reduced cross-reactivity of anti-v with the human protein because the antibody used (Q-20) is directed against the human v chain.
|
vß3 as a potential OPN receptor, we made use of neutralizing antibodies against v and ß3 in an OPN adhesion assay. The antibodies H9.2B8 (anti-v) and 2C9.G2 (anti-ß3) specifically detected expression of the integrins on the surface of melan-a cells (data not shown). Incubation of melan-a cells with OPN coated on plastic wells revealed a high tendency of the cells to specifically attach to the recombinant protein (Fig. 6C)
. Attachment of the cells to OPN was almost unaffected when a control antibody was present during the incubation. On the other hand, both neutralizing antibodies anti-v (H9.2B8), as well as anti-ß3 (2C9.G2), were highly effective in blocking the interaction of melan-a cells with OPN, indicating that v, as well as ß3, are involved in the attachment of these cells to OPN. The specificity of the adhesion assay was additionally tested by measuring adhesion of melan-a cells to vitronectin, a natural ligand of the vitronectin receptor vß3 (data not shown).
To examine whether vß3 is involved in the antiapoptotic effect in melan-a cells, the blocking antibodies against v and ß3 were used to interfere with the OPN interaction in the presence of collagen, and apoptotic cells were identified by DAPI staining as described in Fig. 5B
. The presence of the antibodies did not influence apoptosis induced in the absence of any defined factor (Fig. 6D)
. However, the survival effect brought about by OPN was significantly reduced by addition of anti-v with 
80% apoptotic cells compared with 40% in the absence of any antibody (Fig. 6D)
. Also, anti-ß3 had an influence on OPN-mediated cell survival (55% apoptotic cells), although the effect was lower than that of blocking v (Fig. 6D)
.
Because these data showed that OPN-mediated cell survival is correlated with binding to vß3, and we had found that bFGF induces OPN expression in melan-a cells (Fig. 3B)
, we wondered whether the OPN/vß3 interaction also contributes to the survival effect on melanocytes seen and described for bFGF (8)
. Therefore, cells were cultured in the collagen matrix in bFGF containing medium either in the presence of the v- and ß3-blocking antibodies or in the presence of a control antibody. bFGF alone or in the presence of the control antibody very efficiently suppressed apoptosis with only 5–10% apoptotic cells after 4 days in the collagen gel (Fig. 6E)
. However, both blocking antibodies, anti-v as well as anti-ß3, had a significant influence on bFGF-induced cell survival in the collagen matrix, and the number of apoptotic cells increased to 35 and 20%, respectively (Fig. 6E)
. This clearly indicated that vß3 interactions are contributing to the antiapoptotic signaling brought about by bFGF.
OPN Triggers Antiapoptotic Signaling in Human Melanoma Cells.
To finally investigate if OPN is also playing a role in survival of human melanoma cells in dermal collagen, the different human melanoma cell lines were cultured in collagen gels, and the influence of constitutive OPN expression was analyzed by using the OPN-blocking antibody. The various cell lines showed a distinct behavior in the collagen matrix with Bro and Mel26 cells spreading and growing very efficiently during 4 days. Also A375 cells proliferated in the collagen gel but many cells showed a round phenotype. IFB and D10 grew very slowly, and no increase in the cell number of WM35 cells was seen during 4 days. Approximately 10% of WM35 cells were apoptotic when cultured in three-dimensional collagen. None of the other analyzed cell lines showed significant apoptosis either in the absence of any defined factor or in the presence of the control antibody. Addition of the OPN-blocking antibody, however, resulted in an increase in apoptotic cells in all cell lines except WM35 (Fig. 7)
. The highest increase was seen in Bro cells with 25% of the cells undergoing apoptosis. Mel26, A375, IFB, and D10 showed 9–20% more apoptotic cells when the neutralizing OPN antibody was added (Fig. 7)
, which points to a contribution of OPN to the survival of transformed pigment cells in dermal collagen.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have identified OPN as a factor that can transmit survival signals in melanocytes grown in dermal collagen, and its expression is regulated by two receptor tyrosine kinases that play an important role in pigment cell transformation and melanoma progression. opn originally was cloned as a tumor promoter-inducible gene, and it is frequently overexpressed in human tumors (27)
. Its expression is regulated by various factors, including cytokines like IL-3 and granulocyte macrophage colony-stimulating factor in Ba/F3 cells, where it contributes to cellular survival and proliferation induced by these cytokines (24)
. Also, growth factors (30, 31, 32)
and oncogenes such as ras can induce opn expression. Increased expression of opn is associated with ras-transformation of NIH3T3 cells (33)
, and its expression is enhanced in mammary tumors of v-Hras transgenic mice (34)
. This points to a role of MAPK, which acts downstream of ras, in the induction of the opn promoter. It is also in line with our results showing that the HER-mrk-triggered up regulation of opn expression was dependent on intact MAPK signaling in both cell lines Ba/F3, as well as melan-a cells. This also might explain the higher level of opn expression induced by HER-mrk than by the FGF receptor in melan-a cells (Fig. 3)
because HER-mrk stimulates strong MAPK activation that lasts several hours (13)
, whereas bFGF-induced MAPK activation in melanocytes is only transient (35)
.
Activation of both the HER-mrk as well as the FGF receptor not only induced OPN expression but also enabled murine melanocytes to survive in three-dimensional collagen gels. This could be due to a direct activation of antiapoptotic signaling pathways by the receptors. However, using HER-mrk-conditioned medium and an OPN-neutralizing antibody, we clearly could show that OPN in an indirect way contributes to the survival effect produced by the HER-mrk receptor. For the FGF receptor, it has been described that its stimulation by bFGF induces antiapoptotic signaling in human melanocytes, but the mechanism for this effect has not been elucidated yet (8)
. Because we found that bFGF stimulation of melanocytes can induce OPN expression, this might point to a contribution of OPN to the survival effect seen by Alanko et al. (8)
. One function of OPN, once up regulated by either the FGF or the HER-mrk receptor, could be an interaction of the factor with other receptors present on the melanocyte surface. Our studies revealed that in melan-a cells vß3 can act as such a receptor for OPN and that the OPN/vß3 interaction is crucial for the mediation of cell survival in collagen. The fact that blocking ß3 interactions with a specific antibody did not have an equally strong influence on the OPN-produced antiapoptotic effect suggests that other ß-subunits such as ß1 or ß5 might also participate in binding OPN. Because the ability of melan-a cells to adhere to OPN was reduced to 20% in the presence of the ß3 antibody, it nevertheless seems very likely that vß3 is the main receptor for OPN in this melanocyte cell line.
Most striking, vß3 is known to play an important role in melanoma cell survival (3)
, and its expression correlates with melanoma progression from RGP to VGP melanoma and tumorigenicity (5
, 6
, 36
, 37)
. It should be mentioned, however, that for the effects produced on melanoma cells by vß3, not only OPN (38)
but also denatured (proteolytically-cleaved) collagen was identified as the relevant ligand (3
, 4)
. Thereby, the melanoma cells were able to alter the collagen to promote a stable interaction of vß3 with the matrix protein. Because melan-a cells do express a functional vß3 but were not able to spread properly and survive in collagen gels for a long-term period in the absence of defined factors, this particular interaction seems not to occur in untreated melanocytes. Interestingly, it is described that expression of vß3 in the absence of an appropriate ligand, which is the case for melan-a cells in native collagen, induces apoptosis because of the presence of unligated integrins (39)
. This integrin-mediated death (39)
is described to be biologically and biochemically distinct from anoikis, which is characterized as apoptosis induced by the loss of adhesion, per se (40)
.
The presence of OPN induced spreading and survival of melan-a cells in collagen gels, and OPN is known to bind collagen (28)
. This suggests a role for OPN as attachment protein for melanocytes mediating an interaction between collagen and the OPN receptor vß3. A function of OPN such as an adhesive intermediate (bridging molecule) for cellular attachment to matrix components has also been discussed for vascular smooth muscle cells, where addition of exogenous OPN could restore normal cell adhesion of OPN-deficient cells to collagen (41)
. On the other hand, recent studies revealed that in OPN-deficient vascular smooth muscle cells, the expression of receptors necessary for adequate adhesion was affected (29)
. Hence, OPN, instead of acting directly as attachment protein for melanocytes, might also act as an activating ligand for vß3 thus, stimulating intracellular pathways associated with regulation of adhesion, spreading, or survival.
Just recently it has been reported that in B16 mouse melanoma cells, stimulation of vß3 by OPN leads to the up regulation of the matrix metalloproteinases MMP-2 and MT1-MMP (38)
. OPN also enhanced cell migration and ECM invasion of the cells, demonstrating that it can induce all cellular alterations necessary for the most important step in the establishment of a VGP melanoma from RGP lesions.
The interaction of OPN with vß3 not only can influence cell attachment, spreading, and migration but also mediates survival signals. Such a survival signaling induced by vß3 has been described in endothelial cells. Stimulation of vß3 by OPN thereby resulted in activation of nuclear factor 
B through Src- and ras- dependent pathways (42)
. Activation of the nuclear factor B with OPN as stimulating ligand for vß3 was also found in melanoma cells. The antiapoptotic effect of vß3 in melanoma cells after interaction with denatured collagen, however, was linked to an increase in the Bcl-2:Bax ratio (4)
, which is known to correlate with cell survival. Whether the interaction of OPN with vß3 in melanocytes also leads to an induction of bcl-2 expression or triggers other signaling pathways finally leading to survival of the cells in dermal collagen remains to be investigated.
We found that OPN, once expressed after treatment of the cells with growth factors or tumor promoters, can act as a ligand for vß3 in melanocytes in an autocrine way. Moreover, in several human melanoma cell lines, which constitutively express OPN, blocking of OPN interactions in collagen gels enhanced the number of apoptotic cells. Therefore, constitutive expression of OPN might induce a phenotype that exhibits not only increased migratory but also enhanced antiapoptotic activities as it is found for tumor cells. Indeed, overexpression of OPN in rat mammary epithelial cells has been shown to be sufficient to produce a metastatic phenotype (43)
and reduced metastasis of injected B16 melanoma cells was found in OPN-deficient mice (44)
.
In summary, our data show that in melanocytes, activation of growth factor receptors such as the Xmrk and the FGF receptor, both known to be involved in melanoma development (10 , 11) , can result in the production and secretion of OPN. We have shown that OPN, then in an autocrine way, can promote cell attachment, spreading, and survival of melanocytes in three-dimensional collagen matrices. OPN has been implicated in metastasis processes of tumors (43 , 44) , and cutaneous malignant melanoma is notorious for its high metastatic potential once the pigment cells are transformed. Since recent data proved that OPN can trigger migration and invasion and enhances tumor growth of melanoma cells (38) , production of OPN by melanocytes might be a crucial step in establishing cells with the potential to give rise to tumorigenic VGP melanoma. Our data demonstrate that growth factor receptors of the family of receptor tyrosine kinases thereby contribute to this step by up regulating opn expression in melanocytes, which points to a new role of these receptors in processes associated with melanoma development and progression.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Present address: Department of Pathology, University of Würzburg 97080 Würzburg, Germany.
3 To whom correspondence should be addressed, at Department of Physiological Chemistry I Biocenter (Theodor-Boveri Institut), University of Würzburg, Am Hubland, 97074 Würzburg, Germany. Phone: 49-931-8884153; Fax: 49-931-8884150, E-mail: wellbrock{at}biozentrum.uni-wuerzburg.de
4 The abbreviations used are: RGP, radial growth phase; VGP, vertical growth phase; ECM, extracellular matrix; bFGF, basic fibroblast growth factor; Xmrk, Xiphophorus melanoma receptor kinase; OPN, osteopontin; IL-3, interleukin 3; Hm, HER-mrk; HER-mrk CM, Hm-conditioned medium; EGF, epidermal growth factor; TPA, 12-O-tetradecanoylphorbol-13-acetate; RT, reverse transcription; MAPK, mitogen-activated protein kinase; ERK, extracellular signal-regulated kinase; DAPI, 4',6-diamidino-2-phenylindole.
Received 2/26/02. Accepted 6/11/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
v ß 3 rescues melanoma cells from apoptosis in three-dimensional dermal collagen. Proc. Natl. Acad. Sci. USA, 91: 8856-8860, 1994.(v)ß3 promotes M21 melanoma growth in human skin by regulating tumor cell survival. Cancer Res., 59: 2724-2730, 1999. V gene expression in human melanoma tumorigenicity. J. Clin. Investig., 89: 2018-2022, 1992.
(v)ß3 integrin expression. Am. J. Pathol., 153: 1347-1351, 1998. 9 ß 1-mediated adhesion and migration to thrombin-cleaved osteopontin. Exp. Cell Res., 242: 351-360, 1998.[Medline]
v integrins in K1735 melanoma cells. Invasion Metastasis, 18: 1-14, 1998.[Medline]
B-mediated induction of membrane type 1 matrix metalloproteinase in murine melanoma cells. J. Biol. Chem., 276: 44926-44935, 2001.B mediates vß3 integrin-induced endothelial cell survival. J. Cell Biol., 141: 1083-1093, 1998.This article has been cited by other articles:
|
|
 |
|
  G. Song, Y. Ming, Y. Mao, S. Bao, and G. Ouyang Osteopontin Prevents Curcumin-Induced Apoptosis and Promotes Survival Through Akt Activation via {alpha}v{beta}3 Integrins in Human Gastric Cancer Cells Experimental Biology and Medicine, December 1, 2008; 233(12): 1537 - 1545. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. L. Winfield, F. Kirkland, F. I. Ramos-Ceballos, and T. D. Horn Osteopontin Expression in Spitz Nevi Arch Dermatol, August 1, 2007; 143(8): 1076 - 1077. [Full Text] [PDF]
|
|
|
|
 |
|
 T. H. Burdo, M. R. Wood, and H. S. Fox Osteopontin prevents monocyte recirculation and apoptosis J. Leukoc. Biol., June 1, 2007; 81(6): 1504 - 1511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Alonso, L. Tracey, P. Ortiz, B. Perez-Gomez, J. Palacios, M. Pollan, J. Linares, S. Serrano, A. I. Saez-Castillo, L. Sanchez, et al. A High-Throughput Study in Melanoma Identifies Epithelial-Mesenchymal Transition as a Major Determinant of Metastasis Cancer Res., April 1, 2007; 67(7): 3450 - 3460. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Larsson, D. M. Perlman, D. Fan, C. S. Reilly, M. Peterson, C. Dahlgren, Z. Liang, S. Li, V. A. Polunovsky, C. Wahlestedt, et al. Apoptosis resistance downstream of eIF4E: posttranscriptional activation of an anti-apoptotic transcript carrying a consensus hairpin structure Nucleic Acids Res., September 11, 2006; 34(16): 4375 - 4386. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Packer, S. Pavey, A. Parker, M. Stark, P. Johansson, B. Clarke, P. Pollock, M. Ringner, and N. Hayward Osteopontin is a downstream effector of the PI3-kinase pathway in melanomas that is inversely correlated with functional PTEN Carcinogenesis, September 1, 2006; 27(9): 1778 - 1786. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-H. Hsieh, M. M. Juliana, P. H. Hicks, G. Feng, C. Elmets, L. Liaw, and P.-L. Chang Papilloma development is delayed in osteopontin-null mice: implicating an antiapoptosis role for osteopontin. Cancer Res., July 15, 2006; 66(14): 7119 - 7127. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Meierjohann, E. Wende, A. Kraiss, C. Wellbrock, and M. Schartl The oncogenic epidermal growth factor receptor variant xiphophorus melanoma receptor kinase induces motility in melanocytes by modulation of focal adhesions. Cancer Res., March 15, 2006; 66(6): 3145 - 3152. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Guarino, P. Faviana, G. Salvatore, M. D. Castellone, A. M. Cirafici, V. De Falco, A. Celetti, R. Giannini, F. Basolo, R. M. Melillo, et al. Osteopontin Is Overexpressed in Human Papillary Thyroid Carcinomas and Enhances Thyroid Carcinoma Cell Invasiveness J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5270 - 5278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Bao and S. Stromblad Integrin {alpha}v-mediated inactivation of p53 controls a MEK1-dependent melanoma cell survival pathway in three-dimensional collagen J. Cell Biol., November 22, 2004; 167(4): 745 - 756. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Zohar, B. Zhu, P. Liu, J. Sodek, and C. A. McCulloch Increased cell death in osteopontin-deficient cardiac fibroblasts occurs by a caspase-3-independent pathway Am J Physiol Heart Circ Physiol, October 1, 2004; 287(4): H1730 - H1739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P.-L. Chang, M. Cao, and P. Hicks Osteopontin induction is required for tumor promoter-induced transformation of preneoplastic mouse cells Carcinogenesis, November 1, 2003; 24(11): 1749 - 1758. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Furger, A. L. Allan, S. M. Wilson, C. Hota, S. A. Vantyghem, C. O. Postenka, W. Al-Katib, A. F. Chambers, and A. B. Tuck {beta}3 Integrin Expression Increases Breast Carcinoma Cell Responsiveness to the Malignancy-Enhancing Effects of Osteopontin Mol. Cancer Res., September 1, 2003; 1(11): 810 - 819. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Chen, I. Yang, R. Irby, K. H. Shain, H. G. Wang, J. Quackenbush, D. Coppola, J. Q. Cheng, and T. J. Yeatman Regulation of Caspase Expression and Apoptosis by Adenomatous Polyposis Coli Cancer Res., August 1, 2003; 63(15): 4368 - 4374. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
