Cell Biology International ISSN 1065-6995 doi: 10.1002/cbin.10280
RESEARCH ARTICLE
Studies on primary uveal and cutaneous melanoma cell interaction with vitronectin ska and Małgorzata Przybyło* Marcelina Elz_ bieta Janik, Anna Lityn w, ul. Gronostajowa 9, Poland Department of Glycoconjugate Biochemistry, Institute of Zoology, Jagiellonian University, 30-387 Krako
Abstract We have examined the diversity between primary uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells in their interaction with vitronectin, and established the effect of integrins and b1,6-branched N-oligosaccharides on this process. The adhesion level of uveal melanoma cells to vitronectin was at least twice lower than that of cutaneous ones, but all cells tested repaired scratch wounds on vitronectin-coated surfaces with similar speed. Swainsonine treatment, by reducing the amount of b1,6-branches, significantly decreased cell attachment in all cases, but reduction of wound healing efficiency was compromised only in cutaneous melanoma cell. Functional blocking antibodies used in adhesion and migration assays revealed that integrin avb3 was strongly involved in adhesion and migration only in cutaneous melanoma cells, but its role here was less pronounced than that of integrin avb5. However, in uveal melanoma the specific anti-avb5 integrin antibody had no impact on migration speed. Therefore, the anti-a3b1 integrin antibody was used in order to explain the nature of uveal melanoma interaction with vitronectin, which caused a mild decrease in adhesion efficiency and reduced their motility. Expression of avb5 integrin differed between the cell lines, but there was no distinct pattern to distinguish uveal melanoma from cutaneous melanoma. In conclusion, avb5, but not avb3 integrin is heavily involved in uveal melanoma cell interaction with vitronectin. The role of b1,6-branched N-glycans in the adhesion, but not during migration, of all cells to vitronectin has been confirmed. Keywords: avb3 integrin; avb5 integrin; b1,6-branched N-glycans; cutaneous melanoma; uveal melanoma; vitronectin Introduction Most cancer-related deaths are caused by metastasis and there is no doubt that the speed of dissemination determines the tumor mortality rate. Interactions of tumor cells with the primary tumor microenvironment are involved in the acquisition of migrationary and invasive behavior of these cells, and are important determinants of cancer progression toward metastasis. During the growth of solid tumors, the constitution of extracellular matrix (ECM) is modulated by proteins derived from the plasma, like plasmin(ogen), fibronectin, fibrin(ogen) and vitronectin (Vn), as a consequence of tumor-induced vascular permeability. Vn is a 75 kDa glycoprotein that circulates as a soluble molecule in the plasma. During tissue injury or
tumor growth, Vn is recruited from the plasma into ECM and undergoes multimerization mediated by a complex of thrombin–antithrombin III, complement factor C5b-9, or plasminogen activator inhibitor-1 (PAI-1) (Bloemendal et al., 2004). Vn is involved in the control of cell morphology, proliferation, migration, and survival. It also participates in ECM remodeling during tumor growth and angiogenesis by providing a link between plasminregulated matrix-proteolysis and integrin-mediated migration and is quite predominant in directing tumor cell migration during metastatic formation of many highly invasive cancers (van Belle et al., 1999; McGary et al., 2002). The interactions of tumor cells with ECM are mediated by cell surface adhesion receptors. Some of these, members of
Corresponding author: e-mail: [email protected] Abbreviations: CM, cutaneous melanoma; DSL, Datura stramonium agglutinin; GNL, Galanthus nivalis agglutinin; LEL, Lycopersicon esculentum agglutinin; MAL II, Maackia amurensis agglutinin; PHA-L, Phaseolus vulgaris agglutinin; SNA, Sambucus nigra agglutinin; SW, swainsonine; UM, uveal melanoma; Vn, vitronectin
942
Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
integrin family, are implicated in the establishment, metastasis, and progression of different tumors (Moschos et al., 2007; Desgrosellier and Cheresh, 2010; Rathinam and Alahari, 2010). Vn contains an Arginine-Glycine-Aspartic acid (RGD) motifs that binds the av-integrin family of receptors (avb3, avb5, avb1, and avb8) and the platelet receptors aIIb3 (Murphy and Stupack, 2010). In general, integrin avb3 is regarded as a classical Vn receptor; interaction between avb3 integrin and ECM (besides Vn also fibronectin, laminin, collagen, and osteopontin) promotes cell attachment, spreading, and cell motility. Integrin avb3 also plays a critical role in the survival and metastasis process of cancer cells including cutaneous melanoma (CM), and there is a direct relationship between avb3 integrin expression and melanoma progression (van Belle et al., 1999; McGary et al., 2002). avb3 integrin is also involved in liver metastasis (Yun et al., 1997; Kikkawa et al., 2002), probably through interaction with Vn, which is produced by hepatocytes along with other serum components. Moreover, Vn, but not fibronectin, protects human glioma cell lines from apoptotic death (Uhm et al., 1999). The risk factor for melanoma is also associated with increased levels of Vn in the skin (McKeown-Longo and Panetti, 1996). There is accumulating evidence that the integrin binding capacity can be modified by posttranslational modification, for example glycosylation (Guo et al., 2003; Gu and Taniguchi, 2004; Zhang et al., 2004; Zhao et al., 2008; Janik et al., 2010; Link-Lenczowski and Lity nska, 2011). Tumor cell-surface oligosaccharides are remarkably distinct from those of normal counterparts, and is the most common changes, including the presence of more branched and hypersialylated N-linked oligosaccharides, re-expression of fetal type antigens, and prematurely terminated glycans (Li et al., 2010; Borsig, 2011). In melanoma, some of these tumor-associated alterations of cell surface glycosylation are especially important in cancer progression and metastasis (Przybyło et al., 2007; Lity nska et al., 2008; Przybyło et al., 2008; Przybyło and Lity nska, 2011). This study was undertaken to compare primary uveal melanoma (UM) cell lines (92-1, Mel202) and primary CM cell lines (FM55P, IGR-39) in their capacity to adhere to and migrate on Vn. The role of avb3, avb5, and a3b1 integrins in mediating melanoma cell interaction with Vn was studied, and the level of avb5 integrin expressed by each cell line was determined. Finally, we tested whether b1,6-branched oligosaccharides of surface N-glycoproteins influence melanoma cell interaction with Vn by using swainsonine (SW), a specific inhibitor of Golgi amannosidase II, and we also monitored the influence of SW treatment on cell surface glycosylation pattern of melanoma cells using a variety of oligosaccharide-specific lectins. Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
Melanoma cell interaction with vitronectin
Materials and methods
Chemicals Penicillin–streptomycin solution, bovine serum albumin (BSA), swainsonine, RGD peptide, ExtraAvidin-FITC conjugate, and poly-L-lysine were obtained from SigmaAldrich (St. Louis, MO). Fetal bovine serum and phosphatebuffered saline (PBS) were from GibcoBRLTM (Paisley, UK). Vitronectin from human plasma and monoclonal antibodies against a3b1 (clone P1B5), avb3 (clone LM609), and avb5 (clone P1F6) integrins were products of MerckMillipore (Darmstadt, Germany). Rabbit anti-mouse F(ab’)2 FITCconjugated (X0931) antibody as well as mouse IgG1 negative control were purchased from DAKO (Denmark). Biotinylated lectins were from Vector Laboratories (Burlingame, CA) (Table 1, see supplement). All remaining chemicals were of analytical grade, commercially available.
Cell culture and treatment 92-1 (de Waard-Siebinga et al., 1995) and Mel202 (Verbik et al., 1997) cells (i.e. primary UM cell lines) and FM55P (Real et al., 1998) and IGR-39 (Aubert et al., 1980) cells (i.e. primary CM cell lines) were obtained from ESTDAB Melanoma Cell Bank (Tübingen, Germany). The cells were maintained in RPMI-1640 Glutamax-I1 medium (GibcoBRLTM) supplemented with 10% (v/v) heat-inactivated fetal bovine serum, penicillin (100 U/mL) and streptomycin (100 mg/mL). The cells were grown to confluence as a monolayer at 37 C in a humidified atmosphere containing air plus 5% CO2. Experiments were initiated after the cells had reached sub-confluence. In some experiments, cells were treated with swainsonine (10 mg/mL) for 24 h prior to assay, a dose that had no effect on the viability or growth rate, as seen by trypan blue exclusion and 3[4,5-dimethyldiazol-2-yl]-2,5diphenyltetrazolium bromine (MTT), respectively (data not shown).
Cell adhesion assay to Vn Vn (0.7 mg/mL) in PBS were added to 96-well tissue culture black solid plates (Costar) overnight at 4 C, followed by blockage with 1% (m/v) heat-denatured BSA dissolved in PBS for 1 h at 37 C. Sub-confluent melanoma cells were starved in serum-free medium for 30 min and harvested. Cells (5 104) in 100 mL serum-free medium were added to each well and allowed to attach for 1 h. Non-adhered cells were removed by washing twice with PBS and adherent cells were stained with calcein AM (2 mM) for 30 min. The wells were washed three times with PBS and the fluorescence was measured at 495/520 nm with a microtiter plate reader (FLx800TBT Microplate Reader, Bio-Tek Instruments Inc.). Non-specific cell adhesion on BSA-coated wells was 943
M. E. Janik et al.
Melanoma cell interaction with vitronectin
subtracted. A reference of 100% attachment was estimated from the fluorescence of cells in wells coated with 500 mg/mL poly-L-lysine. All data are given as relative percentages of adhesion compared to adhesion on poly-L-lysine (taken as 100%). Three independent experiments were carried out. Prior to the adhesion assay in some experiments, cells in suspension were agitated on a rotator at 37 C for 60 min with monoclonal function-blocking antibodies specific for avb3, avb5, and a3b1 integrins (all at 5 mg/mL), RGD peptide (0.5 mg/mL), and EDTA (10 mM). Control competition with non-specific IgG1 (80 mg/mL) was performed (data not shown). Changes in adhesion in competition experiments and after SW treatment were calculated by comparing untreated (taken as 100% adhesion) and treated cells. Data are given as percentage of adhesion compared to untreated cells (control). Three independent experiments were carried out.
Scrape-wound healing assay Scrape-wound healing assay was done in 6-well culture plates (Lity nska et al., 2006). Briefly, melanoma cells were grown to confluence on Vn-coated surface (0.7 mg/mL), the medium aspirated, and the cell-coated surface scraped with a 200 mL pipette tip in a single stripe. The surface was washed twice with PBS, covered with complete medium and the wounds were allowed to heal for 24 h at 37 C. To investigate the participation of integrins in this process and the impact of b1,6-branched N-oligosaccharides, some experiments wound healing was carried out in complete culture medium containing antibodies specific for avb3, avb5, and a3b1 integrins (all at 5 mg/mL), RGD peptide (0.5 mg/mL), EDTA (10 mM), swainsonine (10 mg/mL), or non-specific IgG1 (80 mg/mL). Migration of cells into wounded areas was examined with an inverted microscope and photographed. The average extent of wound closure was measured at several widths of the wound. Ten measurements in three separate trials were made for all conditions. Changes in cell migration rate after competition and SW treatment were calculated by comparing migration of untreated (taken as 100% migration) and treated cells. All data are given as percentage of migration compared to untreated cells (control).
Flow cytometric analysis Expression of avb5 integrin was assessed by flow cytometry as previously described (Laidler et al., 2000). Briefly, each sample with 1 105 cells was suspended in 50 mL PBS and incubated with anti-avb5 integrin (20 mg/mL) or normal mouse IgG1 (50 mg/mL) as the negative control in the dark for 45 min at 4 C. After incubation, the cells were washed with PBS and incubated with FITC-conjugated anti-mouse 944
IgG (Fab’)2 fragments on ice for 45 min. They were washed again with PBS and fluorescence of 104 cells was measured by a FACScan flow cytometer (BDBiosciences, San Diego, CA). Three independent experiments were carried out. The influence of SW treatment on cell surface glycosylation of the tested cells was assessed by flow cytometry as previously described (Przybyło et al., 2008), with minor modification. Briefly, cells (1 105) were incubated with biotynylated DSL, MAL II, SNA, GNL, PHA-L, and LEL lectins (1:100) in PBS for 45 min on ice. The cells were washed in PBS, and incubated with ExtraAvidin-FITC conjugate on ice for 45 min. Cells were washed again with PBS and fluorescence measured in a FACSCalibur flow cytometer (BD Biosciences, San Diego, CA). Three independent experiments were carried out.
Statistical analysis Results are expressed as mean standard deviation. Statistical analysis was performed with the use of Duncan’s new multiple range test and P < 0.05 were considered statistically significant. The data are the results of at least 3 separate experiments. Results and discussion
Adhesion and migration abilities of melanoma cells Two important steps in the metastatic cascade involve the ability of cancer cells to adhere to ECM components and subsequently migrate through them (Menon and Beningo, 2011; Polacheck et al., 2013). We first analyzed primary UM and CM cells in term of their adhesion to Vn. There was a significant difference in the adhesion level, and attachment of 92-1 and Mel202 cells was at least twice lower than FM55P and IGR-39 cells (Figure 1A). All the cells repaired scratch wounds on Vn-coated surfaces with quite similar speed (Figure 2AB). Primary UM and CM cells have already been shown to differ significantly in their adhesiveness to most abundant ECM proteins, i.e. collagen type IV, fibronectin, and laminin (Laidler et al., 2006; Przybyło et al., 2008). Indeed, UM cells hardly adhered to any of them, whereas CM cells became attached to these ECM proteins with different efficiencies of adhesion. Since tumor cells escape from the primary site and their subsequent dissemination determines the speed of metastasis formation, it could be supposed that the loosened ECM adhesion of primary UM cells permits them to leave their original site in the tissue more readily than primary CM cells. This is in good agreement with the observation that, in UM, metastasis begins at a very early stage of disease. Our previously published data also showed that in contrast to similar UM and CM cell motility on Vn, UM cells repaired scratch Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Figure 1 Studies on human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cell adhesion to vitronecitn. (A) Adhesion properties of melanoma cells. All data are given as percentage of adhesion relative to adhesion on poly-L-lysine (taken as 100%). Cell adhesion to BSAcoated wells served as negative control. (B) Effect of swainsonine treatment on adhesion of melanoma cells to Vn. The extent of cell adhesion in the presence of SW is presented relatively to cell adhesion in the absence of swainsonine that was considered as 100%. (C) Effect of monoclonal antibodies to avb3, avb5, and a3b1 integrins as well as RGD peptide and EDTA on the adhesion of melanoma cells to Vn. The extent of cell adhesion in the presence of the indicated antibody is presented relatively to cell adhesion in the absence of antibody that was considered as 100%. Each result is the mean of three independent experiments performed in triplicate. Error bars indicate standard deviations. * indicates P < 0.05.
wounds on fibronectin-coated surfaces twice as fast as CM cells (Przybyło et al., 2008).
Cell surface expression of Vn receptors on melanoma cells Individual cells can vary in their adhesive properties by selectively expressing different integrins and modulating their integrin specificity and affinity for ligands. Deregulated expression of integrins in invasive melanoma and the functional importance of these adhesion receptors has been seen before. Melanoma cell interaction with Vn may engage several different receptors belonging to the av-integrin family. av-series integrins are implicated in mediating cell adhesion and spreading, cell locomotion, ligand-receptor internalization, management of the extracellular protease cascade, regulation of cancer progression, apoptosis, and angiogenesis (Nemeth et al., 2007; Murphy and Stupack, 2010). We have previously shown that the percentage of Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
avb3 integrin positive cells is relatively low (85% of Mel202 cells expressed avb5 integrin on their cell surface, adhesion was three times lower than for IGR-39 cells, which express a similarly high level of avb5 integrin. 92-1 cell adhesion efficiency was comparable 945
Melanoma cell interaction with vitronectin
M. E. Janik et al.
Figure 2 Repair of wounds in monolayers of human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells on vitronectin-coated surface. The line was scratched with a plastic pipette tip through a confluent monolayer of cells maintained on Vn-coated surfaces. The scratch-wounded cultures were allowed to heal for 24 h in the presence of SW and specific antibodies. (A) Panels show migration of cells into wounded area in the presence of SW and specific antibodies as well as EDTA and RGD peptide after 24 h. (B) The extent of wound closure, quantified by measuring the width at twenty different locations in the wound and the mean value were compared to the width of the original closure (0 h). (C) Effect of swainsonine treatment on repair of scratch wounds in monolayers of melanoma cells. (D) Effect of treatment with antibodies against avb3, avb5, and a3b1 integrin as well as EDTA and RGD peptide on repair of scratch wounds in monolayers of melanoma cells. Changes in migration rate after SW and antibody treatment were calculated by comparing the migration of untreated (taken as 100%) and treated cells. All data are given as percentage of migration versus that of untreated cells (control). Values are means standard deviation of three separate experiments. * indicates P < 0.05.
to that of Mel202 cells, even though the former is four times weaker in avb5 integrin expression than the latter. These results are in line with findings on three other primary UM cell lines, showing that although these cells had a weak expression of avb5, integrin mediates their adhesion to Vn in various degree (Marshall et al., 1998). 946
Integrins are necessary for melanoma cell interaction with Vn Cell adhesion and migration assays in the presence of EDTA and RGD peptide proved that the interaction of cells with VN was exclusively mediated by integrins in an Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
Figure 3 Expression of avb5 integrin on human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells. (A) Panels show flow cytometric analysis of cells stained for avb5 integrin (mAb P1F6). The samples were analyzed in BD FACSan after incubation with anti-mouse IgG FITC conjugate. Colored areas indicate the fluorescence profile of cells after indirect fluorescence staining with mAb P1F6. Open histograms represent background fluorescence in the presence of isotype matched antibody (negative control). The vertical axis shows cell number and the horizontal axis shows log fluorescence intensity. Data from one of three similar experiments are presented. (B) Diagram shows percentage of melanoma cells expressing avb5 integrin. Values are means standard deviation of three separate experiments. * indicates P < 0.05.
RGD-dependent manner (Figures 1C and 2D). EDTA prevented cell adhesion and resulted in melanoma cell detachment from Vn-coated dishes in the wound healing assay, whereas RGD-peptide completely blocked attachment of these cells to Vn as also wound healing. Thus, to assess whether avb3 and avb5 integrins are responsible for mediating primary melanoma cell interaction with Vn, cell adhesion and wound healing were assessed in the Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
Melanoma cell interaction with vitronectin
presence of specific function-blocking anti-integrin monoclonal antibodies. Anti-avb3 integrin antibody slightly inhibited FM55P and IGR-39 cell adhesion (by 30 and 15%, respectively), but had no effect on 92-1 and Mel202 attachment (Figure 1C). In contrast, anti-avb5 integrin antibody markedly reduced adhesiveness of all tested cells – by 75, 80, 55, and 100% in FM55P, IGR-39, 92-1, and Mel202 cells, respectively (Figure 1C). Most epithelialderived tumor cells do not preferentially express avb3 integrin, but rather use avb5 integrin as a vitronectin receptor (Mizejewski, 1999; Wakayama et al., 2007). Nevertheless, the greater adhesion level of CM cells could not be attributed directly to the differences in the cell surface expression of integrins acting as receptors for Vn (i.e. avb3 and avb5 integrins) between CM and UM cells. Incubation of FM55P and IGR-39 cells with anti-avb3 or anti-avb5 integrin antibodies in migration assays resulted in 50 and 55% inhibition of migration speed in both cases, respectively. In contrast, these anti-integrin antibodies had no influence on UM cell migration (Figure 2D). Many studies of cell migration on Vn indicate that, in most cases, cell motility is dependent on the action of avb3 integrin (Felding-Habermann et al., 2002). An unexpected finding was that avb3 and avb5 integrins were not involved in primary UM cell interaction with Vn during migration in contrast to primary CM cells. These findings implicate participation of other receptors than avb3 and avb5 integrins in UM cell interaction with Vn. The integrin receptors for Vn do not include the a3b1 integrin. However, one of the receptors involved in this interaction could be a3b1 integrin in complex with uPAR (Wei et al., 1994; Wei et al., 2001; Zhang et al., 2004; Janik et al., 2010). uPAR and avb3 integrin can influence each other’s expression and function in Vn adhesion and migration (Nip et al., 1995; Khatib et al., 2001; Tarui et al., 2001). Preincubation of 92-1 and Mel202 cells with anti-a3b1 integrin antibody decreased slightly their adhesion efficiency and reduced their motility by 50 and 45%, respectively (Figures 1C and 2D). Because we found that interaction of UM cells with Vn is integrindependent, it is possible that the interaction between a3b1 integrin and uPAR is responsible for this effect. uPAR in this interaction may be a partner rather than the Vn receptor itself. Because cell-Vn adhesion was not completely abolished by function-blocking avb5 integrin antibody, and was slightly blocked by anti-avb3 integrin antibody, it is also suggested that other Vn/RGDdependent integrins contribute to these events. Such candidates are avb1 and avb8 integrins, but it would be difficult to confirm their contribution in melanoma cell interaction with Vn, because specific function blocking antibodies against these integrin heterodimers are not available commercially yet. 947
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Impact of b1,6-branched glycans on melanoma cell interaction with Vn Taking into consideration the altered glycosylation phenotype of numerous types of human cancer cells, the sugarchain structures on the cell surface of the cell lines were investigated using the following lectins: DSL, GNL, LEL, MAL II, and SNA. It has already been shown that both UM cell lines (92-1 and Mel202) and IGR-39 cells possess more glycoproteins bearing b1,6-branched N-oligosaccharides than FM55P cells (Przybyło et al., 2008). Similar to PHAL staining, FM55P cells possess less terminal galactose (positive reaction with DSL), polylactosaminyl units (positive reaction with LEL), as well as high mannose and/or hybrid-type oligosaccharides than the three other cell lines (positive reaction with GNL) (Figure 4). In the case of sialylated oligosaccharides, as detected by MAL II and SNA, all cells showed a similar degree of staining (Figure 4). b1,6-branched complex type N-oligosaccharides and N-acetylglucosaminyltransferase V (GnT-V) are elevated in multiple tumors, including melanoma (Przybyło et al., 2007, 2008; Lity nska et al., 2008; Przybyło and Lity nska, 2011, and references therein). This increase is associated with an increased expression of GnT-V gene (Mgat5). In cutaneous melanoma, we have shown by comparative lectin-binding studies on cell extracts obtained from different primary (WM35) and metastatic (WM9, WM239, A375) cells that acquisition of the metastatic potential is associated with the increased staining and/or the number of proteins that reacted with PHA-L, SNA, and MAL II (Lity nska et al., 2001), and changes in the number of proteins being a substrate for GnT-V are better correlated with melanoma development and progression than with the expression of cell adhesion molecules (Ochwat et al., 2004; Przybyło et al., 2008). We have also shown that b1,6branched complex type N-oligosaccharides present on the surface of these cells are important for their biological properties (Ciołczyk-Wierzbicka et al., 2004; Lity nska et al., 2006; Janik et al., 2010). The possible functional role of b1,6-branched complex type N-oligosaccharides in primary melanoma cell interactions with Vn was assessed using alkaloid SW. Cells were treated with SW for 24 h prior to adhesion assays or SW was added to complete medium during wound healing assays. To establish the effect of SW on glycosylation profile, the SWtreated cells were also probed with previously mentioned lectins prior to flow cytometry analysis. SW acting as a specific inhibitor of Golgi a-mannosidase II, consequently obstructs the oligosaccharide biosynthetic pathway prior to the initiation of GlcNAc b1,6 branching. It results in the accumulation of high mannose and hybrid type structures. The data (Figure 5) confirmed that this alkaloid significantly reduces the presence of b1,6-branched complex type 948
N-oligosaccharides on the cell surface in all the cells. It also decreased the percentage of cells possessing a2,3-linked sialic acids in FM55P and IGR-39 cell lines, a2,6-linked sialic acids in Mel202 cell lines (Figure 4). The percentage of cells bearing high mannose and hybrid type structures was unchanged, but the relative amount of these structures increased (data not shown). Although integrin-mediated interaction is based on binding of a and b subunits to the defined peptide sequence of the ligand, the strength of this binding is modulated by various factors including the status of glycosylation of integrin subunits. Therefore, the altered cell surface glycosylation may contribute to changes in tumor cell adhesion, migration and invasion abilities. Addition of SW decreased attachment of all cells within the range of 20% (IGR-39) to 45% (Mel202), and partially inhibited motility in FM55P (40%) and IGR-39 (50%) cells (Figures 1B and 2C). Therefore, the results suggest that modulation of melanoma cell interaction with Vn by b1,6-branched N-glycans was not directly dependent on the amount of these glycans. Neither suppression of melanoma cells adhesion nor slower rate of their migration speed could be attributed to the reduced cell growth rate or cell viability (data not shown). Thus, in view of these results: (i) UM and CM cell interaction with Vn is exclusively mediated by integrins in an RGD-dependent manner and (ii) swainsonine treatment influences melanoma cell interaction with Vn, except UM cell migration; we assume that these changes might be directly related to changes in the function of integrin receptors and resulted from alteration in integrin glycosylation status. Thus we have investigated the role of b1,6-branched N-glycans in order to gain a deeper understanding of their role in uveal melanoma biology. Earlier work (Przybyło et al., 2008) with PHA-L or SW showed that UM cell adhesion to Vn, as well as migration on fibronecti but not on Vn, are influenced by the presence of b1,6-branched N-oligosaccharides. Expression of b1,6-branched complex type N-oligosaccharides on a3, a5, av, and b1 integrin subunits is associated with the increased cellular motility and reduced substratum adhesion (Gu, and Taniguchi, 2004; Zhao et al., 2008; Janik et al., 2010; Link-Lenczowski and Lity nska, 2011). Little, however, is known about how glycosylation of adhesion molecules, (integrins) influence their function in uveal melanoma, and it is unclear whether the same mechanisms important in cutaneous melanoma are similarly relevant, or if different interactions are required.
Concluding remarks UM and CM, being of the same neuroectodermal origin, are similar in respect of their cellular morphology, expression of melanocytic lineage markers, predisposition to metastatic spread, as well as resistance to the commonly used therapies Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Figure 4 Effect of swainsonine on the following lectins: Datura stramonium (DSL), Maackia amurensis (MAL II), Sambucus nigra (SNA), Galanthus nivalis (GNL), and Lycopersicon esculentum (LEL) binding to the cell surface of human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells. (A) Panels show FACS profiles for lectin-positive cells. Open histograms with colored lines indicate the fluorescence profile of cells untreated with swainsonine and gray areas indicate the fluorescence profile of cells treated with swainsonine after indirect fluorescence staining with lectins. Open histograms represent background fluorescence in the presence of Extravidin FITC conjugate (negative control). Vertical axis shows cell number and horizontal axis shows log fluorescence intensity. Data from one of three similar experiments are presented. The negative control for each line is different in some experiments because the experiments were not run on the same occasion.
Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
949
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Figure 5 Effect of swainsonine treatment on Phaseolus vulgaris (PHA-L) lectin binding on the cell surface of human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells. (A) Panels show FACS profiles for PHA-L-positive cells. Colored areas indicate the fluorescence profile of cells treated or not with swainsonine after indirect fluorescence staining with lectins. Open histograms represent background fluorescence in the presence of Extravidin FITC conjugate (negative control). Vertical axis shows cell number and horizontal axis shows log fluorescence intensity. Data from one of three similar experiments are presented. (B) Diagram shows percentage of PHA-L positive cells. (C) Diagrams show quantification of data from flow cytometric analyses. Fluorescence intensity is shown relatively to untreated cells (set as 100%). Values are means standard deviation of three separate experiments. * indicates P < 0.05.
950
(Ramaiya and Harbour, 2007). Nevertheless, they differ profoundly in several aspects, including their genetic alternation, immune response, metastatic potential, clinical progression, and response to treatments (Belmar-Lopez et al., 2008; Niederkorn, 2009). In contrast to the rapid progress in studies on cell adhesion and migration through protein–protein and protein–carbohydrate interactions performed on CM, mainly metastatic ones, the number and progress of studies on UM is still very small. Our results provide evidence that both melanoma subtypes also differ in terms of integrin receptors involved in the interaction with Vn. avb5 integrin partly mediates interaction of primary UM cells with Vn in vitro, but the classical Vn receptor, i.e. avb3 integrin, does not seem to be involved in these interactions. In contrast, both avb3 and avb5 integrins are necessary for primary CM cells interaction with Vn. Probably UM and CM cells have alternative means of accomplishing a similar task. Moreover, the role of b1,6 branched N-oligosaccharides in UM biology also seems to be important, as it is in the case of CM cells. Integrin avb3 has a prominent function in cancer biology, due to being a marker of angiogenesis and directly contributing to metastasis of several different cancer types, including cutaneous melanoma (Jin and Varner, 2004). Integrin avb3 mediates vascular invasion; its binding to L1 on endothelial cells promotes melanoma cell migration towards blood vessels, and its binding to Vn upregulates MMP-2 expression, which appears to facilitate cell-mediated stroma degradation (Brooks et al., 2010). Ligation of avb3 integrin with Vn triggers signals that helps them remain alive during migration, as well as in the environment of newly colonized organs. UM metastasizes hematogenously, with the liver being predominantly affected, whereas CM cells tend to spread through the lymphatic system, usually first affecting the regional lymph nodes (Chang et al., 1998). Adhesion of CM cells to lymph nodes through the interaction between avb3 integrin and Vn is involved in the process of lymphatic dissemination of these cells (Nip et al., 1992). Thus, our data significantly broaden the current views on the role of integrins and carbohydrates in behavior of primary UM cells. A clear understanding of their role in UM cell biology should be the scope for further investigation, which could open up new avenues for therapeutic intervention. Acknowledgements and funding This research was supported in part by grants from the Institute of Zoology, Jagiellonian University (K/ZDS/ 001953, WRBW/BiNoZ/IZ/5/2009) and from the Ministry of Science and Higher Education (MNiSW) of Poland (N N303 552338). Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
References Aubert C, Rouge F, Galindo JR (1980) Tumorigenicity of human malignant melanocytes in nude mice in relation to their differentiation in vitro. J Natl Cancer Inst 64: 1029–40. Belmar-Lopez C, Mancheno-Corvo P, Saornil MA, Baril P, Vassaux G, Quintanilla M, Martin-Duque P (2008) Uveal vs. cutaneous melanoma. Origins and cause of differences. Clin Trans Oncol 10: 137–42. Bloemendal HJ, de Boer HC, Koop EA, van Dongen AJ, Goldschmeding R, Landman WJM, Logtenberg T, Gebbink MFBG, Voest EE (2004) Acivated vitronectin as a target for anticancer therapy with human antibodies. Cancer Immunol Immunother 53: 799–808. Borsig L (2011) Glycans in cancer. In: Pavão MSG ed. Glycans in Disease and Therapeutics. Biology of Extracellular Matrix. Berlin: Springer-Verlag, pp. 63–81. Brooks SA, Lomax-Browne HJ, Carter TM, Kinch CE, Hall DMS (2010) Molecular interactions in cancer cell metastasis. Acta Histochem 112: 3–25. Chang AE, Karnell LH, Menck HR (1998) The National Cancer Data Base report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases from the past decade. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer 15: 1664–78. Ciołczyk-Wierzbicka D, Amoresano A, Casbarra A, HojaŁukowicz D, Lity nska A, Laidler P (2004) The structure of the oligosaccharides of N-cadherin from human melanoma cell lines. Glycoconj J 20: 483–92. Desgrosellier JS, Cheresh DA (2010) Integrins in cancer: biological implications and therapeutic opportunities. Nat Rev Cancer 10: 9–22. de Waard-Siebinga, Blom DJ, Griffioen M, Schrier PI, Hoogendoorn E, Beverstock G, Danen EH, Jager MJ (1995) Establishment and characterization of an uveal-melanoma cell line. Int J Cancer 62: 155–61. Felding-Habermann B, Fransvea E, O’Toole TEO, Manzuk L, Faha B, Hensler M (2002) Involvement of tumor cell integrin avb3 in hematogenous metastasis of human melanoma cells. Clin Exp Metastasis 19: 427–36. Gu J, Taniguchi N (2004) Regulation of integrin functions by N-glycans. Glycoconj J 21: 9–15. Guo P, Zhang Y, Zhao J, Wang A, Guo H, Zhang X, Chen H (2003) Regulation on the expression and N-glycosylation of integrins by N-acethylglucosaminyltransferase V. Biochim Biophys Res Commun 310: 619–26. Janik ME, Lity nska A, Vereecken P (2010) Cell migration—the role of integrin glycosylation. Biochim Biophys Acta 1800: 545–55. Janik ME, Przybyło M, Pochec E, Pokrywka M, Lity nska A (2010) Effect of a3b1 and avb3 integrin on interaction of melanoma cells with vitronectin. Acta Biochim Pol 57: 55–61. Jin H, Varner J (2004) Integrins: the role in cancer development and as treatment targets. Br J Cancer 90: 561–5.
Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
Melanoma cell interaction with vitronectin
Khatib AM, Nip J, Fallavollita L, Lehmann M, Jensen G, Brodt P (2001) Regulation of urokinase plasminogen activator/plasmin-mediated invasion of melanoma cells by the integrin vitronectin receptor avb3. Int J Cancer 91: 300–8. Kikkawa H, Kaihou M, Horaguchi N, Uchida T, Imafuku H, Takiguchi A, Yamazaki Y, Koike C, Kututo R, Kakiuchi T, Tsukada H, Takada Y, Matsuura N, Oku N (2002) Role of integrin avb3 in the early phase of liver metastasis: PET and IVM analyses. Clin Exp Metastasis 19: 717–25. Laidler P, Gil D, Pituch-Noworolska A, Ciołczyk D, Ksia˛z_ ek D, Przybyło M, Lity nska A (2000) Expression of b1-integrins and N-cadherin in bladder cancer and melanoma cell lines. Acta Biochim Pol 47: 1159–70. Laidler P, Lity nska A, Hoja-Łukowicz D, Łabe˛dź M, Przybyło M, Ciołczyk-Wierzbicka D, Pochec E, Tre˛bacz E, Kremser E (2006) Characterization of glycosylation and adherent properties of melanoma cell lines. Cancer Immunol Immunother 55: 112–8. Li M, Song L, Qin X (2010) Glycan changes: cancer metastasis and anti-cancer vaccines. J Biosci 35: 665–73. Link-Lenczowski P, Lity nska A (2011) Glycans in melanoma screening. Part 2. Towards the understanding of integrin Nglycosylation in melanoma. Biochem Soc Trans 39: 374–7. Lity nska A, Przybyło M, Pochec E, Hoja-Łukowicz D, Ciołczyk D, Laidler P, Gil D (2001) Comparison of the lectin-binding pattern in different human melanoma cell lines. Melanoma Res 11: 205–12. Lity nska A, Przybyło M, Pochec E, Kremser E, Hoja-Łukowicz D, Sulowska U (2006) Does glycosylation of melanoma cells influence their interactions with fibronectin?. Biochimie 88: 527–34. Lity nska A, Przybyło M, Hoja-Łukowicz D, Pochec E, Kremser M, Ciołczyk-Wierzbicka D, Łabe˛dź M, Laidler P (2008) Protein glycosylation as marker of melanoma progression. Curr Cancer Ther Rev 2: 144–8. Marshall JF, Rutherford DC, Happerfield L, Hanby A, McCartney ACE, Newton-Bishop J, Hart IR (1998) Comparative analysis of integrins in vitro and in vivo in uveal and cutaneous melanomas. Br J Cancer 77: 522–9. McGary EC, Lev DC, Bar-Eli M (2002) Cellular adhesion pathways and metastatic potential of human melanoma. Cancer Biol Ther 1: 459–65. McKeown-Longo PJ, Panetti TS (1996) Structure and function of vitronectin. Trends Glycosci Glycotechnol 8: 327–40. Menon S, Beningo KA (2011) Cancer cell invasion in enhanced by applied mechanical stimulation. PLoS One 6(2): e17277. Mizejewski GJ (1999) Role of integrins in cancer: survey of expressing patterns. Proc Soc Exp Biol Med 222: 124–38. Moschos JS, Drogowsky LM, Reppert SL, Kirkwood JM (2007) Integrins and cancer. Oncology (Williston Park) 21: 13–20. Murphy E, Stupack D (2010) Vitronectin-binding integrins in cancer. In: Zent R , Pozzi A ed. Cell-extracellular Matrix Interactions in Cancer. New York: Springer MediaþBussines Media, pp. 137–70.
951
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Nemeth JA, Nakada MT, Trikha M, Lang Z, Gordon MS, Jayson GC, Corringham R, Prabhakar U, Davis HM, Beckman RA (2007) Alpha-v integrins as a therapeutic targets in oncology. Cancer Invest 25: 632–46. Niederkorn JY (2009) Immune escape mechanisms of intraocular tumors. Prog Retin Eye Res 28: 329–47. Nip J, Shibata H, Loskutoff DJ, Cheresh DA, Brodt P (1992) Human melanoma cell derived from lymphatic metastases use integrin alphavbeta3 to adhere to lymph node vitronectin. J Clin Invest 90: 1406–13. Nip J, Rabbani SA, Shibata HR, Brodt P (1995) Coordinated expression of the vitronectin receptor and the urokinase-type plasminogen activator receptor in metastatic melanoma cells. J Clin Invest 95: 2096–103. Ochwat D, Hoja-Łukowicz D, Lity nska A (2004) N-glycoproteins bearing b1-6 branched oligosaccharides from the A375 human melanoma cell line analysed by tandem mass spectrometry. Melanoma Res 14: 479–85. Polacheck WJ, Zervantonakis IK, Kamm RD (2013) Tumor cell migration in complex microenvironments. Cell Mol Life Sci 70: 1335–56. Przybyło M, Martuszewska D, Pochec E, Hoja-Łukowicz D, Lity nska A (2007) Identification of b1-6 branched N-glycans in human melanoma cell lines from different progression stages by tandem mass spectrometry analysis. Biochim Biophys Acta 1770: 1427–35. Przybyło M, Pochec E, Link-Lenczowski P, Lity nska A (2008) b1,6-branching of cell surface glycoproteins may contribute to uveal melanoma progression by up-regulating cell motility. Mol Vis 14: 625–36. Przybyło M, Lity nska A (2011) Glycans in melanoma screening. Part 1. The role of b1,6-branched N-linked oligosaccharides in melanoma. Biochem Soc Trans 39: 370–73. Ramaiya KJ, Harbour JW (2007) Current management of uveal melanoma. Expert Rev Ophthalmol 2: 939–46. Rathinam R, Alahari SK (2010) Important role of integrins in the cancer biology. Cancer Metastasis Rev 29: 223–37. Real LM, Jimenez P, Canton J, Kirkin A, Garcia A, Abril E, Zeuthen J, Ruiz-Cabello F, Garrido F (1998) In vivo and in vitro generation of a new altered HLA phenotype in melanomatumour-cell variants expressing a single HLA-class-I allele. Int J Cancer 75: 317–23. Saida T (2001) Recent advances in melanoma research. J Dermatol Sci 26: 1–13.
952
Tarui T, Mazar AP, Cines DB, Takada Y (2001) Urokinase-type plasminogen activator receptor (CD87) is a ligand for integrins and mediates cell-cell interaction. J Biol Chem 276: 3983–90. Uhm JH, Dooley NP, Kyritsis AP, Rao JS, Gladson Vitronectin CL (1999) A glioma-derived extracellular matrix protein, protects tumor cells from apoptotic death. Clin Cancer Res 5: 1587– 94. van Belle, Elenitsas R, Satyamoorthy K, Wolfe JT, Guerry 4th D, Schuchter L, van Belle TJ, Albelda S, Tahin P, Herlyn M, Elder DE (1999) Progression-related expression of b3 integrin in melanomas and nevi. Hum Pathol 30: 62–7. Verbik DJ, Murray TG, Tran JM, Ksander BR (1997) Melanomas that develop within the eye inhibit lymphocyte proliferation. Int J Cancer 73: 470–8. Wakayama M, Abei M, Kawashima R, Seo E, Fukuda K, Ugai H, Murata T, Tanaka N, Hyodo I, Hamada H, Yokoyama KK (2007) E1A, E1B double-restricted adenovirus with RGD-fiber modification exhibits enhanced oncolysis for CAR-deficient biliary cancers. Clin Cancer Res 13: 3043–50. Wei Y, Waltz DA, Rao D, Drummont RJ, Rosenberg S, Chapman HA (1994) Identification of urokinase receptor as an adhesive receptor for vitronectin. J Cell Biol 276: 8436–44. Wei Y, Eble JA, Wang Z, Kreidberg JA, Chapman HA (2001) Urokinase receptor promotes b1 integrin function through interaction with integrin a3b1. Mol Biol Cell 12: 2975–86. Yun Z, Smith TW, Menter DG, McIntire LV, Nicolson GL (1997) Differential adhesion of metastatic RAW117 large-cell lymphoma cells under static or hydrodynamic conditions: role of integrin avb3. Clin Exp Metastasis 15: 3–11. Zhang Y, Zhao J, Zhang X, Guo H, Liu F, Chen H (2004) Relations of the type and branch of surface N-glycans to cell adhesion, migration and integrin expressions. Mol Cell Biochem 260: 137–46. Zhao Y, Sato Y, Isaji T, Fukuda T, Matsumoto A, Miyoshi E, Gu J, Taniguchi N (2008) Branched N-glycans regulate the biological functions of integrins and cadherins. FEBS J 275: 1939–48. Received 7 October 2013; accepted 20 March 2014. Final version published online 28 April 2014.
Supporting Information Additional supporting information may be found in the online version of this article at the publisher’s web-site.
Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
RESEARCH ARTICLE
Studies on primary uveal and cutaneous melanoma cell interaction with vitronectin ska and Małgorzata Przybyło* Marcelina Elz_ bieta Janik, Anna Lityn w, ul. Gronostajowa 9, Poland Department of Glycoconjugate Biochemistry, Institute of Zoology, Jagiellonian University, 30-387 Krako
Abstract We have examined the diversity between primary uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells in their interaction with vitronectin, and established the effect of integrins and b1,6-branched N-oligosaccharides on this process. The adhesion level of uveal melanoma cells to vitronectin was at least twice lower than that of cutaneous ones, but all cells tested repaired scratch wounds on vitronectin-coated surfaces with similar speed. Swainsonine treatment, by reducing the amount of b1,6-branches, significantly decreased cell attachment in all cases, but reduction of wound healing efficiency was compromised only in cutaneous melanoma cell. Functional blocking antibodies used in adhesion and migration assays revealed that integrin avb3 was strongly involved in adhesion and migration only in cutaneous melanoma cells, but its role here was less pronounced than that of integrin avb5. However, in uveal melanoma the specific anti-avb5 integrin antibody had no impact on migration speed. Therefore, the anti-a3b1 integrin antibody was used in order to explain the nature of uveal melanoma interaction with vitronectin, which caused a mild decrease in adhesion efficiency and reduced their motility. Expression of avb5 integrin differed between the cell lines, but there was no distinct pattern to distinguish uveal melanoma from cutaneous melanoma. In conclusion, avb5, but not avb3 integrin is heavily involved in uveal melanoma cell interaction with vitronectin. The role of b1,6-branched N-glycans in the adhesion, but not during migration, of all cells to vitronectin has been confirmed. Keywords: avb3 integrin; avb5 integrin; b1,6-branched N-glycans; cutaneous melanoma; uveal melanoma; vitronectin Introduction Most cancer-related deaths are caused by metastasis and there is no doubt that the speed of dissemination determines the tumor mortality rate. Interactions of tumor cells with the primary tumor microenvironment are involved in the acquisition of migrationary and invasive behavior of these cells, and are important determinants of cancer progression toward metastasis. During the growth of solid tumors, the constitution of extracellular matrix (ECM) is modulated by proteins derived from the plasma, like plasmin(ogen), fibronectin, fibrin(ogen) and vitronectin (Vn), as a consequence of tumor-induced vascular permeability. Vn is a 75 kDa glycoprotein that circulates as a soluble molecule in the plasma. During tissue injury or
tumor growth, Vn is recruited from the plasma into ECM and undergoes multimerization mediated by a complex of thrombin–antithrombin III, complement factor C5b-9, or plasminogen activator inhibitor-1 (PAI-1) (Bloemendal et al., 2004). Vn is involved in the control of cell morphology, proliferation, migration, and survival. It also participates in ECM remodeling during tumor growth and angiogenesis by providing a link between plasminregulated matrix-proteolysis and integrin-mediated migration and is quite predominant in directing tumor cell migration during metastatic formation of many highly invasive cancers (van Belle et al., 1999; McGary et al., 2002). The interactions of tumor cells with ECM are mediated by cell surface adhesion receptors. Some of these, members of
Corresponding author: e-mail: [email protected] Abbreviations: CM, cutaneous melanoma; DSL, Datura stramonium agglutinin; GNL, Galanthus nivalis agglutinin; LEL, Lycopersicon esculentum agglutinin; MAL II, Maackia amurensis agglutinin; PHA-L, Phaseolus vulgaris agglutinin; SNA, Sambucus nigra agglutinin; SW, swainsonine; UM, uveal melanoma; Vn, vitronectin
942
Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
integrin family, are implicated in the establishment, metastasis, and progression of different tumors (Moschos et al., 2007; Desgrosellier and Cheresh, 2010; Rathinam and Alahari, 2010). Vn contains an Arginine-Glycine-Aspartic acid (RGD) motifs that binds the av-integrin family of receptors (avb3, avb5, avb1, and avb8) and the platelet receptors aIIb3 (Murphy and Stupack, 2010). In general, integrin avb3 is regarded as a classical Vn receptor; interaction between avb3 integrin and ECM (besides Vn also fibronectin, laminin, collagen, and osteopontin) promotes cell attachment, spreading, and cell motility. Integrin avb3 also plays a critical role in the survival and metastasis process of cancer cells including cutaneous melanoma (CM), and there is a direct relationship between avb3 integrin expression and melanoma progression (van Belle et al., 1999; McGary et al., 2002). avb3 integrin is also involved in liver metastasis (Yun et al., 1997; Kikkawa et al., 2002), probably through interaction with Vn, which is produced by hepatocytes along with other serum components. Moreover, Vn, but not fibronectin, protects human glioma cell lines from apoptotic death (Uhm et al., 1999). The risk factor for melanoma is also associated with increased levels of Vn in the skin (McKeown-Longo and Panetti, 1996). There is accumulating evidence that the integrin binding capacity can be modified by posttranslational modification, for example glycosylation (Guo et al., 2003; Gu and Taniguchi, 2004; Zhang et al., 2004; Zhao et al., 2008; Janik et al., 2010; Link-Lenczowski and Lity nska, 2011). Tumor cell-surface oligosaccharides are remarkably distinct from those of normal counterparts, and is the most common changes, including the presence of more branched and hypersialylated N-linked oligosaccharides, re-expression of fetal type antigens, and prematurely terminated glycans (Li et al., 2010; Borsig, 2011). In melanoma, some of these tumor-associated alterations of cell surface glycosylation are especially important in cancer progression and metastasis (Przybyło et al., 2007; Lity nska et al., 2008; Przybyło et al., 2008; Przybyło and Lity nska, 2011). This study was undertaken to compare primary uveal melanoma (UM) cell lines (92-1, Mel202) and primary CM cell lines (FM55P, IGR-39) in their capacity to adhere to and migrate on Vn. The role of avb3, avb5, and a3b1 integrins in mediating melanoma cell interaction with Vn was studied, and the level of avb5 integrin expressed by each cell line was determined. Finally, we tested whether b1,6-branched oligosaccharides of surface N-glycoproteins influence melanoma cell interaction with Vn by using swainsonine (SW), a specific inhibitor of Golgi amannosidase II, and we also monitored the influence of SW treatment on cell surface glycosylation pattern of melanoma cells using a variety of oligosaccharide-specific lectins. Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
Melanoma cell interaction with vitronectin
Materials and methods
Chemicals Penicillin–streptomycin solution, bovine serum albumin (BSA), swainsonine, RGD peptide, ExtraAvidin-FITC conjugate, and poly-L-lysine were obtained from SigmaAldrich (St. Louis, MO). Fetal bovine serum and phosphatebuffered saline (PBS) were from GibcoBRLTM (Paisley, UK). Vitronectin from human plasma and monoclonal antibodies against a3b1 (clone P1B5), avb3 (clone LM609), and avb5 (clone P1F6) integrins were products of MerckMillipore (Darmstadt, Germany). Rabbit anti-mouse F(ab’)2 FITCconjugated (X0931) antibody as well as mouse IgG1 negative control were purchased from DAKO (Denmark). Biotinylated lectins were from Vector Laboratories (Burlingame, CA) (Table 1, see supplement). All remaining chemicals were of analytical grade, commercially available.
Cell culture and treatment 92-1 (de Waard-Siebinga et al., 1995) and Mel202 (Verbik et al., 1997) cells (i.e. primary UM cell lines) and FM55P (Real et al., 1998) and IGR-39 (Aubert et al., 1980) cells (i.e. primary CM cell lines) were obtained from ESTDAB Melanoma Cell Bank (Tübingen, Germany). The cells were maintained in RPMI-1640 Glutamax-I1 medium (GibcoBRLTM) supplemented with 10% (v/v) heat-inactivated fetal bovine serum, penicillin (100 U/mL) and streptomycin (100 mg/mL). The cells were grown to confluence as a monolayer at 37 C in a humidified atmosphere containing air plus 5% CO2. Experiments were initiated after the cells had reached sub-confluence. In some experiments, cells were treated with swainsonine (10 mg/mL) for 24 h prior to assay, a dose that had no effect on the viability or growth rate, as seen by trypan blue exclusion and 3[4,5-dimethyldiazol-2-yl]-2,5diphenyltetrazolium bromine (MTT), respectively (data not shown).
Cell adhesion assay to Vn Vn (0.7 mg/mL) in PBS were added to 96-well tissue culture black solid plates (Costar) overnight at 4 C, followed by blockage with 1% (m/v) heat-denatured BSA dissolved in PBS for 1 h at 37 C. Sub-confluent melanoma cells were starved in serum-free medium for 30 min and harvested. Cells (5 104) in 100 mL serum-free medium were added to each well and allowed to attach for 1 h. Non-adhered cells were removed by washing twice with PBS and adherent cells were stained with calcein AM (2 mM) for 30 min. The wells were washed three times with PBS and the fluorescence was measured at 495/520 nm with a microtiter plate reader (FLx800TBT Microplate Reader, Bio-Tek Instruments Inc.). Non-specific cell adhesion on BSA-coated wells was 943
M. E. Janik et al.
Melanoma cell interaction with vitronectin
subtracted. A reference of 100% attachment was estimated from the fluorescence of cells in wells coated with 500 mg/mL poly-L-lysine. All data are given as relative percentages of adhesion compared to adhesion on poly-L-lysine (taken as 100%). Three independent experiments were carried out. Prior to the adhesion assay in some experiments, cells in suspension were agitated on a rotator at 37 C for 60 min with monoclonal function-blocking antibodies specific for avb3, avb5, and a3b1 integrins (all at 5 mg/mL), RGD peptide (0.5 mg/mL), and EDTA (10 mM). Control competition with non-specific IgG1 (80 mg/mL) was performed (data not shown). Changes in adhesion in competition experiments and after SW treatment were calculated by comparing untreated (taken as 100% adhesion) and treated cells. Data are given as percentage of adhesion compared to untreated cells (control). Three independent experiments were carried out.
Scrape-wound healing assay Scrape-wound healing assay was done in 6-well culture plates (Lity nska et al., 2006). Briefly, melanoma cells were grown to confluence on Vn-coated surface (0.7 mg/mL), the medium aspirated, and the cell-coated surface scraped with a 200 mL pipette tip in a single stripe. The surface was washed twice with PBS, covered with complete medium and the wounds were allowed to heal for 24 h at 37 C. To investigate the participation of integrins in this process and the impact of b1,6-branched N-oligosaccharides, some experiments wound healing was carried out in complete culture medium containing antibodies specific for avb3, avb5, and a3b1 integrins (all at 5 mg/mL), RGD peptide (0.5 mg/mL), EDTA (10 mM), swainsonine (10 mg/mL), or non-specific IgG1 (80 mg/mL). Migration of cells into wounded areas was examined with an inverted microscope and photographed. The average extent of wound closure was measured at several widths of the wound. Ten measurements in three separate trials were made for all conditions. Changes in cell migration rate after competition and SW treatment were calculated by comparing migration of untreated (taken as 100% migration) and treated cells. All data are given as percentage of migration compared to untreated cells (control).
Flow cytometric analysis Expression of avb5 integrin was assessed by flow cytometry as previously described (Laidler et al., 2000). Briefly, each sample with 1 105 cells was suspended in 50 mL PBS and incubated with anti-avb5 integrin (20 mg/mL) or normal mouse IgG1 (50 mg/mL) as the negative control in the dark for 45 min at 4 C. After incubation, the cells were washed with PBS and incubated with FITC-conjugated anti-mouse 944
IgG (Fab’)2 fragments on ice for 45 min. They were washed again with PBS and fluorescence of 104 cells was measured by a FACScan flow cytometer (BDBiosciences, San Diego, CA). Three independent experiments were carried out. The influence of SW treatment on cell surface glycosylation of the tested cells was assessed by flow cytometry as previously described (Przybyło et al., 2008), with minor modification. Briefly, cells (1 105) were incubated with biotynylated DSL, MAL II, SNA, GNL, PHA-L, and LEL lectins (1:100) in PBS for 45 min on ice. The cells were washed in PBS, and incubated with ExtraAvidin-FITC conjugate on ice for 45 min. Cells were washed again with PBS and fluorescence measured in a FACSCalibur flow cytometer (BD Biosciences, San Diego, CA). Three independent experiments were carried out.
Statistical analysis Results are expressed as mean standard deviation. Statistical analysis was performed with the use of Duncan’s new multiple range test and P < 0.05 were considered statistically significant. The data are the results of at least 3 separate experiments. Results and discussion
Adhesion and migration abilities of melanoma cells Two important steps in the metastatic cascade involve the ability of cancer cells to adhere to ECM components and subsequently migrate through them (Menon and Beningo, 2011; Polacheck et al., 2013). We first analyzed primary UM and CM cells in term of their adhesion to Vn. There was a significant difference in the adhesion level, and attachment of 92-1 and Mel202 cells was at least twice lower than FM55P and IGR-39 cells (Figure 1A). All the cells repaired scratch wounds on Vn-coated surfaces with quite similar speed (Figure 2AB). Primary UM and CM cells have already been shown to differ significantly in their adhesiveness to most abundant ECM proteins, i.e. collagen type IV, fibronectin, and laminin (Laidler et al., 2006; Przybyło et al., 2008). Indeed, UM cells hardly adhered to any of them, whereas CM cells became attached to these ECM proteins with different efficiencies of adhesion. Since tumor cells escape from the primary site and their subsequent dissemination determines the speed of metastasis formation, it could be supposed that the loosened ECM adhesion of primary UM cells permits them to leave their original site in the tissue more readily than primary CM cells. This is in good agreement with the observation that, in UM, metastasis begins at a very early stage of disease. Our previously published data also showed that in contrast to similar UM and CM cell motility on Vn, UM cells repaired scratch Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Figure 1 Studies on human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cell adhesion to vitronecitn. (A) Adhesion properties of melanoma cells. All data are given as percentage of adhesion relative to adhesion on poly-L-lysine (taken as 100%). Cell adhesion to BSAcoated wells served as negative control. (B) Effect of swainsonine treatment on adhesion of melanoma cells to Vn. The extent of cell adhesion in the presence of SW is presented relatively to cell adhesion in the absence of swainsonine that was considered as 100%. (C) Effect of monoclonal antibodies to avb3, avb5, and a3b1 integrins as well as RGD peptide and EDTA on the adhesion of melanoma cells to Vn. The extent of cell adhesion in the presence of the indicated antibody is presented relatively to cell adhesion in the absence of antibody that was considered as 100%. Each result is the mean of three independent experiments performed in triplicate. Error bars indicate standard deviations. * indicates P < 0.05.
wounds on fibronectin-coated surfaces twice as fast as CM cells (Przybyło et al., 2008).
Cell surface expression of Vn receptors on melanoma cells Individual cells can vary in their adhesive properties by selectively expressing different integrins and modulating their integrin specificity and affinity for ligands. Deregulated expression of integrins in invasive melanoma and the functional importance of these adhesion receptors has been seen before. Melanoma cell interaction with Vn may engage several different receptors belonging to the av-integrin family. av-series integrins are implicated in mediating cell adhesion and spreading, cell locomotion, ligand-receptor internalization, management of the extracellular protease cascade, regulation of cancer progression, apoptosis, and angiogenesis (Nemeth et al., 2007; Murphy and Stupack, 2010). We have previously shown that the percentage of Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
avb3 integrin positive cells is relatively low (85% of Mel202 cells expressed avb5 integrin on their cell surface, adhesion was three times lower than for IGR-39 cells, which express a similarly high level of avb5 integrin. 92-1 cell adhesion efficiency was comparable 945
Melanoma cell interaction with vitronectin
M. E. Janik et al.
Figure 2 Repair of wounds in monolayers of human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells on vitronectin-coated surface. The line was scratched with a plastic pipette tip through a confluent monolayer of cells maintained on Vn-coated surfaces. The scratch-wounded cultures were allowed to heal for 24 h in the presence of SW and specific antibodies. (A) Panels show migration of cells into wounded area in the presence of SW and specific antibodies as well as EDTA and RGD peptide after 24 h. (B) The extent of wound closure, quantified by measuring the width at twenty different locations in the wound and the mean value were compared to the width of the original closure (0 h). (C) Effect of swainsonine treatment on repair of scratch wounds in monolayers of melanoma cells. (D) Effect of treatment with antibodies against avb3, avb5, and a3b1 integrin as well as EDTA and RGD peptide on repair of scratch wounds in monolayers of melanoma cells. Changes in migration rate after SW and antibody treatment were calculated by comparing the migration of untreated (taken as 100%) and treated cells. All data are given as percentage of migration versus that of untreated cells (control). Values are means standard deviation of three separate experiments. * indicates P < 0.05.
to that of Mel202 cells, even though the former is four times weaker in avb5 integrin expression than the latter. These results are in line with findings on three other primary UM cell lines, showing that although these cells had a weak expression of avb5, integrin mediates their adhesion to Vn in various degree (Marshall et al., 1998). 946
Integrins are necessary for melanoma cell interaction with Vn Cell adhesion and migration assays in the presence of EDTA and RGD peptide proved that the interaction of cells with VN was exclusively mediated by integrins in an Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
Figure 3 Expression of avb5 integrin on human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells. (A) Panels show flow cytometric analysis of cells stained for avb5 integrin (mAb P1F6). The samples were analyzed in BD FACSan after incubation with anti-mouse IgG FITC conjugate. Colored areas indicate the fluorescence profile of cells after indirect fluorescence staining with mAb P1F6. Open histograms represent background fluorescence in the presence of isotype matched antibody (negative control). The vertical axis shows cell number and the horizontal axis shows log fluorescence intensity. Data from one of three similar experiments are presented. (B) Diagram shows percentage of melanoma cells expressing avb5 integrin. Values are means standard deviation of three separate experiments. * indicates P < 0.05.
RGD-dependent manner (Figures 1C and 2D). EDTA prevented cell adhesion and resulted in melanoma cell detachment from Vn-coated dishes in the wound healing assay, whereas RGD-peptide completely blocked attachment of these cells to Vn as also wound healing. Thus, to assess whether avb3 and avb5 integrins are responsible for mediating primary melanoma cell interaction with Vn, cell adhesion and wound healing were assessed in the Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
Melanoma cell interaction with vitronectin
presence of specific function-blocking anti-integrin monoclonal antibodies. Anti-avb3 integrin antibody slightly inhibited FM55P and IGR-39 cell adhesion (by 30 and 15%, respectively), but had no effect on 92-1 and Mel202 attachment (Figure 1C). In contrast, anti-avb5 integrin antibody markedly reduced adhesiveness of all tested cells – by 75, 80, 55, and 100% in FM55P, IGR-39, 92-1, and Mel202 cells, respectively (Figure 1C). Most epithelialderived tumor cells do not preferentially express avb3 integrin, but rather use avb5 integrin as a vitronectin receptor (Mizejewski, 1999; Wakayama et al., 2007). Nevertheless, the greater adhesion level of CM cells could not be attributed directly to the differences in the cell surface expression of integrins acting as receptors for Vn (i.e. avb3 and avb5 integrins) between CM and UM cells. Incubation of FM55P and IGR-39 cells with anti-avb3 or anti-avb5 integrin antibodies in migration assays resulted in 50 and 55% inhibition of migration speed in both cases, respectively. In contrast, these anti-integrin antibodies had no influence on UM cell migration (Figure 2D). Many studies of cell migration on Vn indicate that, in most cases, cell motility is dependent on the action of avb3 integrin (Felding-Habermann et al., 2002). An unexpected finding was that avb3 and avb5 integrins were not involved in primary UM cell interaction with Vn during migration in contrast to primary CM cells. These findings implicate participation of other receptors than avb3 and avb5 integrins in UM cell interaction with Vn. The integrin receptors for Vn do not include the a3b1 integrin. However, one of the receptors involved in this interaction could be a3b1 integrin in complex with uPAR (Wei et al., 1994; Wei et al., 2001; Zhang et al., 2004; Janik et al., 2010). uPAR and avb3 integrin can influence each other’s expression and function in Vn adhesion and migration (Nip et al., 1995; Khatib et al., 2001; Tarui et al., 2001). Preincubation of 92-1 and Mel202 cells with anti-a3b1 integrin antibody decreased slightly their adhesion efficiency and reduced their motility by 50 and 45%, respectively (Figures 1C and 2D). Because we found that interaction of UM cells with Vn is integrindependent, it is possible that the interaction between a3b1 integrin and uPAR is responsible for this effect. uPAR in this interaction may be a partner rather than the Vn receptor itself. Because cell-Vn adhesion was not completely abolished by function-blocking avb5 integrin antibody, and was slightly blocked by anti-avb3 integrin antibody, it is also suggested that other Vn/RGDdependent integrins contribute to these events. Such candidates are avb1 and avb8 integrins, but it would be difficult to confirm their contribution in melanoma cell interaction with Vn, because specific function blocking antibodies against these integrin heterodimers are not available commercially yet. 947
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Impact of b1,6-branched glycans on melanoma cell interaction with Vn Taking into consideration the altered glycosylation phenotype of numerous types of human cancer cells, the sugarchain structures on the cell surface of the cell lines were investigated using the following lectins: DSL, GNL, LEL, MAL II, and SNA. It has already been shown that both UM cell lines (92-1 and Mel202) and IGR-39 cells possess more glycoproteins bearing b1,6-branched N-oligosaccharides than FM55P cells (Przybyło et al., 2008). Similar to PHAL staining, FM55P cells possess less terminal galactose (positive reaction with DSL), polylactosaminyl units (positive reaction with LEL), as well as high mannose and/or hybrid-type oligosaccharides than the three other cell lines (positive reaction with GNL) (Figure 4). In the case of sialylated oligosaccharides, as detected by MAL II and SNA, all cells showed a similar degree of staining (Figure 4). b1,6-branched complex type N-oligosaccharides and N-acetylglucosaminyltransferase V (GnT-V) are elevated in multiple tumors, including melanoma (Przybyło et al., 2007, 2008; Lity nska et al., 2008; Przybyło and Lity nska, 2011, and references therein). This increase is associated with an increased expression of GnT-V gene (Mgat5). In cutaneous melanoma, we have shown by comparative lectin-binding studies on cell extracts obtained from different primary (WM35) and metastatic (WM9, WM239, A375) cells that acquisition of the metastatic potential is associated with the increased staining and/or the number of proteins that reacted with PHA-L, SNA, and MAL II (Lity nska et al., 2001), and changes in the number of proteins being a substrate for GnT-V are better correlated with melanoma development and progression than with the expression of cell adhesion molecules (Ochwat et al., 2004; Przybyło et al., 2008). We have also shown that b1,6branched complex type N-oligosaccharides present on the surface of these cells are important for their biological properties (Ciołczyk-Wierzbicka et al., 2004; Lity nska et al., 2006; Janik et al., 2010). The possible functional role of b1,6-branched complex type N-oligosaccharides in primary melanoma cell interactions with Vn was assessed using alkaloid SW. Cells were treated with SW for 24 h prior to adhesion assays or SW was added to complete medium during wound healing assays. To establish the effect of SW on glycosylation profile, the SWtreated cells were also probed with previously mentioned lectins prior to flow cytometry analysis. SW acting as a specific inhibitor of Golgi a-mannosidase II, consequently obstructs the oligosaccharide biosynthetic pathway prior to the initiation of GlcNAc b1,6 branching. It results in the accumulation of high mannose and hybrid type structures. The data (Figure 5) confirmed that this alkaloid significantly reduces the presence of b1,6-branched complex type 948
N-oligosaccharides on the cell surface in all the cells. It also decreased the percentage of cells possessing a2,3-linked sialic acids in FM55P and IGR-39 cell lines, a2,6-linked sialic acids in Mel202 cell lines (Figure 4). The percentage of cells bearing high mannose and hybrid type structures was unchanged, but the relative amount of these structures increased (data not shown). Although integrin-mediated interaction is based on binding of a and b subunits to the defined peptide sequence of the ligand, the strength of this binding is modulated by various factors including the status of glycosylation of integrin subunits. Therefore, the altered cell surface glycosylation may contribute to changes in tumor cell adhesion, migration and invasion abilities. Addition of SW decreased attachment of all cells within the range of 20% (IGR-39) to 45% (Mel202), and partially inhibited motility in FM55P (40%) and IGR-39 (50%) cells (Figures 1B and 2C). Therefore, the results suggest that modulation of melanoma cell interaction with Vn by b1,6-branched N-glycans was not directly dependent on the amount of these glycans. Neither suppression of melanoma cells adhesion nor slower rate of their migration speed could be attributed to the reduced cell growth rate or cell viability (data not shown). Thus, in view of these results: (i) UM and CM cell interaction with Vn is exclusively mediated by integrins in an RGD-dependent manner and (ii) swainsonine treatment influences melanoma cell interaction with Vn, except UM cell migration; we assume that these changes might be directly related to changes in the function of integrin receptors and resulted from alteration in integrin glycosylation status. Thus we have investigated the role of b1,6-branched N-glycans in order to gain a deeper understanding of their role in uveal melanoma biology. Earlier work (Przybyło et al., 2008) with PHA-L or SW showed that UM cell adhesion to Vn, as well as migration on fibronecti but not on Vn, are influenced by the presence of b1,6-branched N-oligosaccharides. Expression of b1,6-branched complex type N-oligosaccharides on a3, a5, av, and b1 integrin subunits is associated with the increased cellular motility and reduced substratum adhesion (Gu, and Taniguchi, 2004; Zhao et al., 2008; Janik et al., 2010; Link-Lenczowski and Lity nska, 2011). Little, however, is known about how glycosylation of adhesion molecules, (integrins) influence their function in uveal melanoma, and it is unclear whether the same mechanisms important in cutaneous melanoma are similarly relevant, or if different interactions are required.
Concluding remarks UM and CM, being of the same neuroectodermal origin, are similar in respect of their cellular morphology, expression of melanocytic lineage markers, predisposition to metastatic spread, as well as resistance to the commonly used therapies Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Figure 4 Effect of swainsonine on the following lectins: Datura stramonium (DSL), Maackia amurensis (MAL II), Sambucus nigra (SNA), Galanthus nivalis (GNL), and Lycopersicon esculentum (LEL) binding to the cell surface of human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells. (A) Panels show FACS profiles for lectin-positive cells. Open histograms with colored lines indicate the fluorescence profile of cells untreated with swainsonine and gray areas indicate the fluorescence profile of cells treated with swainsonine after indirect fluorescence staining with lectins. Open histograms represent background fluorescence in the presence of Extravidin FITC conjugate (negative control). Vertical axis shows cell number and horizontal axis shows log fluorescence intensity. Data from one of three similar experiments are presented. The negative control for each line is different in some experiments because the experiments were not run on the same occasion.
Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
949
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Figure 5 Effect of swainsonine treatment on Phaseolus vulgaris (PHA-L) lectin binding on the cell surface of human uveal (92-1 and Mel202) and cutaneous (FM55P and IGR-39) melanoma cells. (A) Panels show FACS profiles for PHA-L-positive cells. Colored areas indicate the fluorescence profile of cells treated or not with swainsonine after indirect fluorescence staining with lectins. Open histograms represent background fluorescence in the presence of Extravidin FITC conjugate (negative control). Vertical axis shows cell number and horizontal axis shows log fluorescence intensity. Data from one of three similar experiments are presented. (B) Diagram shows percentage of PHA-L positive cells. (C) Diagrams show quantification of data from flow cytometric analyses. Fluorescence intensity is shown relatively to untreated cells (set as 100%). Values are means standard deviation of three separate experiments. * indicates P < 0.05.
950
(Ramaiya and Harbour, 2007). Nevertheless, they differ profoundly in several aspects, including their genetic alternation, immune response, metastatic potential, clinical progression, and response to treatments (Belmar-Lopez et al., 2008; Niederkorn, 2009). In contrast to the rapid progress in studies on cell adhesion and migration through protein–protein and protein–carbohydrate interactions performed on CM, mainly metastatic ones, the number and progress of studies on UM is still very small. Our results provide evidence that both melanoma subtypes also differ in terms of integrin receptors involved in the interaction with Vn. avb5 integrin partly mediates interaction of primary UM cells with Vn in vitro, but the classical Vn receptor, i.e. avb3 integrin, does not seem to be involved in these interactions. In contrast, both avb3 and avb5 integrins are necessary for primary CM cells interaction with Vn. Probably UM and CM cells have alternative means of accomplishing a similar task. Moreover, the role of b1,6 branched N-oligosaccharides in UM biology also seems to be important, as it is in the case of CM cells. Integrin avb3 has a prominent function in cancer biology, due to being a marker of angiogenesis and directly contributing to metastasis of several different cancer types, including cutaneous melanoma (Jin and Varner, 2004). Integrin avb3 mediates vascular invasion; its binding to L1 on endothelial cells promotes melanoma cell migration towards blood vessels, and its binding to Vn upregulates MMP-2 expression, which appears to facilitate cell-mediated stroma degradation (Brooks et al., 2010). Ligation of avb3 integrin with Vn triggers signals that helps them remain alive during migration, as well as in the environment of newly colonized organs. UM metastasizes hematogenously, with the liver being predominantly affected, whereas CM cells tend to spread through the lymphatic system, usually first affecting the regional lymph nodes (Chang et al., 1998). Adhesion of CM cells to lymph nodes through the interaction between avb3 integrin and Vn is involved in the process of lymphatic dissemination of these cells (Nip et al., 1992). Thus, our data significantly broaden the current views on the role of integrins and carbohydrates in behavior of primary UM cells. A clear understanding of their role in UM cell biology should be the scope for further investigation, which could open up new avenues for therapeutic intervention. Acknowledgements and funding This research was supported in part by grants from the Institute of Zoology, Jagiellonian University (K/ZDS/ 001953, WRBW/BiNoZ/IZ/5/2009) and from the Ministry of Science and Higher Education (MNiSW) of Poland (N N303 552338). Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
M. E. Janik et al.
References Aubert C, Rouge F, Galindo JR (1980) Tumorigenicity of human malignant melanocytes in nude mice in relation to their differentiation in vitro. J Natl Cancer Inst 64: 1029–40. Belmar-Lopez C, Mancheno-Corvo P, Saornil MA, Baril P, Vassaux G, Quintanilla M, Martin-Duque P (2008) Uveal vs. cutaneous melanoma. Origins and cause of differences. Clin Trans Oncol 10: 137–42. Bloemendal HJ, de Boer HC, Koop EA, van Dongen AJ, Goldschmeding R, Landman WJM, Logtenberg T, Gebbink MFBG, Voest EE (2004) Acivated vitronectin as a target for anticancer therapy with human antibodies. Cancer Immunol Immunother 53: 799–808. Borsig L (2011) Glycans in cancer. In: Pavão MSG ed. Glycans in Disease and Therapeutics. Biology of Extracellular Matrix. Berlin: Springer-Verlag, pp. 63–81. Brooks SA, Lomax-Browne HJ, Carter TM, Kinch CE, Hall DMS (2010) Molecular interactions in cancer cell metastasis. Acta Histochem 112: 3–25. Chang AE, Karnell LH, Menck HR (1998) The National Cancer Data Base report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases from the past decade. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer 15: 1664–78. Ciołczyk-Wierzbicka D, Amoresano A, Casbarra A, HojaŁukowicz D, Lity nska A, Laidler P (2004) The structure of the oligosaccharides of N-cadherin from human melanoma cell lines. Glycoconj J 20: 483–92. Desgrosellier JS, Cheresh DA (2010) Integrins in cancer: biological implications and therapeutic opportunities. Nat Rev Cancer 10: 9–22. de Waard-Siebinga, Blom DJ, Griffioen M, Schrier PI, Hoogendoorn E, Beverstock G, Danen EH, Jager MJ (1995) Establishment and characterization of an uveal-melanoma cell line. Int J Cancer 62: 155–61. Felding-Habermann B, Fransvea E, O’Toole TEO, Manzuk L, Faha B, Hensler M (2002) Involvement of tumor cell integrin avb3 in hematogenous metastasis of human melanoma cells. Clin Exp Metastasis 19: 427–36. Gu J, Taniguchi N (2004) Regulation of integrin functions by N-glycans. Glycoconj J 21: 9–15. Guo P, Zhang Y, Zhao J, Wang A, Guo H, Zhang X, Chen H (2003) Regulation on the expression and N-glycosylation of integrins by N-acethylglucosaminyltransferase V. Biochim Biophys Res Commun 310: 619–26. Janik ME, Lity nska A, Vereecken P (2010) Cell migration—the role of integrin glycosylation. Biochim Biophys Acta 1800: 545–55. Janik ME, Przybyło M, Pochec E, Pokrywka M, Lity nska A (2010) Effect of a3b1 and avb3 integrin on interaction of melanoma cells with vitronectin. Acta Biochim Pol 57: 55–61. Jin H, Varner J (2004) Integrins: the role in cancer development and as treatment targets. Br J Cancer 90: 561–5.
Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
Melanoma cell interaction with vitronectin
Khatib AM, Nip J, Fallavollita L, Lehmann M, Jensen G, Brodt P (2001) Regulation of urokinase plasminogen activator/plasmin-mediated invasion of melanoma cells by the integrin vitronectin receptor avb3. Int J Cancer 91: 300–8. Kikkawa H, Kaihou M, Horaguchi N, Uchida T, Imafuku H, Takiguchi A, Yamazaki Y, Koike C, Kututo R, Kakiuchi T, Tsukada H, Takada Y, Matsuura N, Oku N (2002) Role of integrin avb3 in the early phase of liver metastasis: PET and IVM analyses. Clin Exp Metastasis 19: 717–25. Laidler P, Gil D, Pituch-Noworolska A, Ciołczyk D, Ksia˛z_ ek D, Przybyło M, Lity nska A (2000) Expression of b1-integrins and N-cadherin in bladder cancer and melanoma cell lines. Acta Biochim Pol 47: 1159–70. Laidler P, Lity nska A, Hoja-Łukowicz D, Łabe˛dź M, Przybyło M, Ciołczyk-Wierzbicka D, Pochec E, Tre˛bacz E, Kremser E (2006) Characterization of glycosylation and adherent properties of melanoma cell lines. Cancer Immunol Immunother 55: 112–8. Li M, Song L, Qin X (2010) Glycan changes: cancer metastasis and anti-cancer vaccines. J Biosci 35: 665–73. Link-Lenczowski P, Lity nska A (2011) Glycans in melanoma screening. Part 2. Towards the understanding of integrin Nglycosylation in melanoma. Biochem Soc Trans 39: 374–7. Lity nska A, Przybyło M, Pochec E, Hoja-Łukowicz D, Ciołczyk D, Laidler P, Gil D (2001) Comparison of the lectin-binding pattern in different human melanoma cell lines. Melanoma Res 11: 205–12. Lity nska A, Przybyło M, Pochec E, Kremser E, Hoja-Łukowicz D, Sulowska U (2006) Does glycosylation of melanoma cells influence their interactions with fibronectin?. Biochimie 88: 527–34. Lity nska A, Przybyło M, Hoja-Łukowicz D, Pochec E, Kremser M, Ciołczyk-Wierzbicka D, Łabe˛dź M, Laidler P (2008) Protein glycosylation as marker of melanoma progression. Curr Cancer Ther Rev 2: 144–8. Marshall JF, Rutherford DC, Happerfield L, Hanby A, McCartney ACE, Newton-Bishop J, Hart IR (1998) Comparative analysis of integrins in vitro and in vivo in uveal and cutaneous melanomas. Br J Cancer 77: 522–9. McGary EC, Lev DC, Bar-Eli M (2002) Cellular adhesion pathways and metastatic potential of human melanoma. Cancer Biol Ther 1: 459–65. McKeown-Longo PJ, Panetti TS (1996) Structure and function of vitronectin. Trends Glycosci Glycotechnol 8: 327–40. Menon S, Beningo KA (2011) Cancer cell invasion in enhanced by applied mechanical stimulation. PLoS One 6(2): e17277. Mizejewski GJ (1999) Role of integrins in cancer: survey of expressing patterns. Proc Soc Exp Biol Med 222: 124–38. Moschos JS, Drogowsky LM, Reppert SL, Kirkwood JM (2007) Integrins and cancer. Oncology (Williston Park) 21: 13–20. Murphy E, Stupack D (2010) Vitronectin-binding integrins in cancer. In: Zent R , Pozzi A ed. Cell-extracellular Matrix Interactions in Cancer. New York: Springer MediaþBussines Media, pp. 137–70.
951
M. E. Janik et al.
Melanoma cell interaction with vitronectin
Nemeth JA, Nakada MT, Trikha M, Lang Z, Gordon MS, Jayson GC, Corringham R, Prabhakar U, Davis HM, Beckman RA (2007) Alpha-v integrins as a therapeutic targets in oncology. Cancer Invest 25: 632–46. Niederkorn JY (2009) Immune escape mechanisms of intraocular tumors. Prog Retin Eye Res 28: 329–47. Nip J, Shibata H, Loskutoff DJ, Cheresh DA, Brodt P (1992) Human melanoma cell derived from lymphatic metastases use integrin alphavbeta3 to adhere to lymph node vitronectin. J Clin Invest 90: 1406–13. Nip J, Rabbani SA, Shibata HR, Brodt P (1995) Coordinated expression of the vitronectin receptor and the urokinase-type plasminogen activator receptor in metastatic melanoma cells. J Clin Invest 95: 2096–103. Ochwat D, Hoja-Łukowicz D, Lity nska A (2004) N-glycoproteins bearing b1-6 branched oligosaccharides from the A375 human melanoma cell line analysed by tandem mass spectrometry. Melanoma Res 14: 479–85. Polacheck WJ, Zervantonakis IK, Kamm RD (2013) Tumor cell migration in complex microenvironments. Cell Mol Life Sci 70: 1335–56. Przybyło M, Martuszewska D, Pochec E, Hoja-Łukowicz D, Lity nska A (2007) Identification of b1-6 branched N-glycans in human melanoma cell lines from different progression stages by tandem mass spectrometry analysis. Biochim Biophys Acta 1770: 1427–35. Przybyło M, Pochec E, Link-Lenczowski P, Lity nska A (2008) b1,6-branching of cell surface glycoproteins may contribute to uveal melanoma progression by up-regulating cell motility. Mol Vis 14: 625–36. Przybyło M, Lity nska A (2011) Glycans in melanoma screening. Part 1. The role of b1,6-branched N-linked oligosaccharides in melanoma. Biochem Soc Trans 39: 370–73. Ramaiya KJ, Harbour JW (2007) Current management of uveal melanoma. Expert Rev Ophthalmol 2: 939–46. Rathinam R, Alahari SK (2010) Important role of integrins in the cancer biology. Cancer Metastasis Rev 29: 223–37. Real LM, Jimenez P, Canton J, Kirkin A, Garcia A, Abril E, Zeuthen J, Ruiz-Cabello F, Garrido F (1998) In vivo and in vitro generation of a new altered HLA phenotype in melanomatumour-cell variants expressing a single HLA-class-I allele. Int J Cancer 75: 317–23. Saida T (2001) Recent advances in melanoma research. J Dermatol Sci 26: 1–13.
952
Tarui T, Mazar AP, Cines DB, Takada Y (2001) Urokinase-type plasminogen activator receptor (CD87) is a ligand for integrins and mediates cell-cell interaction. J Biol Chem 276: 3983–90. Uhm JH, Dooley NP, Kyritsis AP, Rao JS, Gladson Vitronectin CL (1999) A glioma-derived extracellular matrix protein, protects tumor cells from apoptotic death. Clin Cancer Res 5: 1587– 94. van Belle, Elenitsas R, Satyamoorthy K, Wolfe JT, Guerry 4th D, Schuchter L, van Belle TJ, Albelda S, Tahin P, Herlyn M, Elder DE (1999) Progression-related expression of b3 integrin in melanomas and nevi. Hum Pathol 30: 62–7. Verbik DJ, Murray TG, Tran JM, Ksander BR (1997) Melanomas that develop within the eye inhibit lymphocyte proliferation. Int J Cancer 73: 470–8. Wakayama M, Abei M, Kawashima R, Seo E, Fukuda K, Ugai H, Murata T, Tanaka N, Hyodo I, Hamada H, Yokoyama KK (2007) E1A, E1B double-restricted adenovirus with RGD-fiber modification exhibits enhanced oncolysis for CAR-deficient biliary cancers. Clin Cancer Res 13: 3043–50. Wei Y, Waltz DA, Rao D, Drummont RJ, Rosenberg S, Chapman HA (1994) Identification of urokinase receptor as an adhesive receptor for vitronectin. J Cell Biol 276: 8436–44. Wei Y, Eble JA, Wang Z, Kreidberg JA, Chapman HA (2001) Urokinase receptor promotes b1 integrin function through interaction with integrin a3b1. Mol Biol Cell 12: 2975–86. Yun Z, Smith TW, Menter DG, McIntire LV, Nicolson GL (1997) Differential adhesion of metastatic RAW117 large-cell lymphoma cells under static or hydrodynamic conditions: role of integrin avb3. Clin Exp Metastasis 15: 3–11. Zhang Y, Zhao J, Zhang X, Guo H, Liu F, Chen H (2004) Relations of the type and branch of surface N-glycans to cell adhesion, migration and integrin expressions. Mol Cell Biochem 260: 137–46. Zhao Y, Sato Y, Isaji T, Fukuda T, Matsumoto A, Miyoshi E, Gu J, Taniguchi N (2008) Branched N-glycans regulate the biological functions of integrins and cadherins. FEBS J 275: 1939–48. Received 7 October 2013; accepted 20 March 2014. Final version published online 28 April 2014.
Supporting Information Additional supporting information may be found in the online version of this article at the publisher’s web-site.
Cell Biol Int 38 (2014) 942–952 © 2014 International Federation for Cell Biology
