UFH and LMWH were supplied by Sigma (St. Louis, MO, USA). Stock solutions of 10 mg/ml were prepared by dissolving heparin in sterile, RNase- and DNase-free DEPC water (Cayman Chemical Co., Ann Arbor, MI, USA). Human plasma FN (1 mg/ml) was obtained by Millipore Corp. (Billerica, MA, USA). RPMI medium and penicillin-streptomycin were obtained from Biosera (Sussex, UK) and gentamycin was supplied by Invitrogen Life Technologies (Carlsbad, CA, USA). Fetal bovine serum (FBS) was purchased by Gibco Life Technologies (Carlsbad, CA, USA). Fluorescein isothiocyanate (FITC)-conjugated unfractionated heparin (referred to as FITC-Heparin) was obtained from Invitrogen Life Technologies. D-[6-3H(N)]glucosamine hydrochloride was supplied by DuPont de Nemours (Dreiech, Germany). Heparin lyase II (heparinase II, no EC number) from Flavobacterium heparinum, chondroitinase AC II from Arthrobacter aurescens (EC 4.2.2.5), proteinase K and 2X crystallized papain (EC 3.4.22.2) were obtained from Sigma Chemical Co. (St. Louis, MO, USA). Heparin lyases I and III from Flavobacterium heparinum (EC 4.2.2.7 and EC 4.2.2.8, respectively), chondroitinase ABC from Proteus vulgaris (EC 4.2.2.4), keratanase II and the anti-heparitinase stubs antibody clone (3G10) were from Seikagaku Kogyo Co. (Tokyo, Japan).

In this study we used the B6FS human fibrosarcoma cell line (40). The cells were obtained from the Cell Bank of the Karolinska Institute, Stockholm, Sweden and were a kind gift from Dr. Anders Hjerpe (Karolinska Institute). The cells were cultured at 37°C and humidity 5% CO₂ in RPMI supplemented with 10% FBS and antimicrobial agents (100 IU/ml penicillin, 100 µg/ml streptomycin and 0.5% gentamycin). All experiments were conducted under serum-free conditions; specifically after 24 h of serum starvation, the cells were treated with UFH and LMWH (10 µg/ml and 30 µg/ml) for 48 h.

The B6FS cells (450,000 cells per T25 flask) were incubated with siRNA (100 nM) specific for the FAK gene (siFAK), as previously described by Hong et al (41) and with siRNA negative control sequences (siScramble) for 6 h. Specific RNA (Invitrogen Life Technologies) and Lipofectamine^® 2000 (Invitrogen Life Technologies) (1/50 µl medium) were added to Opti-MEM^© (Invitrogen Life Technologies) for 5 min at room temperature. In continuation, diluted Lipofectamine 2000 was mixed with the siRNA for 20 min to induce the formation of liposome-siRNA complexes. The medium containing siRNA or siScramble was removed after 6 h of incubation and fresh RPMI 0% FBS medium supplemented with antibiotics was added. The cells were harvested for the respective experiments, after 48 h of culture.

After 48 h of respective treatments, the harvested B6FS cells were lysed with RIPA buffer and electrophoresed on an 8% polyacrylamide gel. Protein bands were transferred onto nitrocellulose membrane in 10 nM CAPS, pH 11 and 10% methanol. All membranes were blocked and incubated overnight at 4°C with PBS containing 0,1% Tween-20 and 5% v/v low fat milk powder. The respective membranes were incubated with the primary antibodies diluted in PBS containing 0,1% Tween and 1% v/v low fat milk powder, for 1 h at room temperature. The following primary antibodies were used: p-FAK (1:200; MAB1141; Millipore Corp.), FAK (1:200; sc-557; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), nuclear factor-κB (NF-κB) p65 subunit (1:200; (sc-109; Santa Cruz Biotechnology, Inc.) and β-actin (1:2,500; MAB1501; Millipore Corp.). The immune complexes were detected by peroxidase-conjugated anti-mouse, anti-goat and anti-rabbit secondary antibodies (1:10,000; Millipore Corp.), using SuperSignal West Pico Chemiluminescent substrate (Pierce Biotechnology, Inc., Rockford, IL, USA).

For real-time PCR, mRNA was isolated using TRIzol reagent (Invitrogen Life Technologies) according to manufacturer's instructions. RNA (1 µg) was used for cDNA synthesis, using the Takara reverse transcription reagent kit (Takara Bio, Dalian, China). Real-time PCR was performed using the Mx300P cycler (Stratagene; Agilent Technologies, Inc., Santa Clara, CA, USA). The KAPA SYBR^® FAST Universal qPCR kit (Kapa Biosystems, Inc., Wilmington, MA, USA) was used for real-time PCR reactions in a total volume of 20 µl and with suitable specific gene primers (Table I). The PCR conditions used for amplification were: 94°C for 15 min followed by 40 cycles at 94°C for 20 sec, 55°C for 30 sec and 72°C for 30 sec, followed by 72°C for 10 min. Standard curves were run and produced a linear plot of threshold cycle (Ct) against dilution (log). Gene levels were quantified according to the concentrations of a standard curve and are presented as arbitrary units. GAPDH was used as a housekeeping gene, for comparison between samples.

The determination of the HS content was carried out as previously described by Karamanos et al (42). Briefly, in order to determine the amount of HS production by the B6FS cells, we performed metabolic labeling of GAGs by supplementing the cell cultures with D-[6-3H(N)]glucosamine hydrochloride (10 µCi/ml) during the period of 16 h prior to the respective harvesting time. Upon the termination of the incubation period, the cells were harvested and cell-associated proteoglycans (PGs) were extracted with 50 mM Tris-HCl, pH 8.0, containing 1% (v/v) Triton X-100 and 0.1% (w/v) NaCl and the following proteinase inhibitors: phenylmethanesulfonyl flouride, benzamidine hydrochloride and hexanoic acid at final concentrations of 2, 5 and 50 mM, respectively. The collected conditioned medium was concentrated to 1:100 of its original volume on an YM-10 membrane (Amicon/Millipore). The PGs were then precipitated by the addition of 4 vol. of 95% (v/v) ethanol containing 2.5% (w/v) sodium acetate with 40 µl chondroitin sulfate (CSA; 0.2 mg/l) added as a carrier. Following centrifugation (11,000 × g for 10 min at 25°C), the precipitates of Pgs were digested with 2 U/ml proteolytic enzyme papain in 100 mM phosphate buffer (pH 7.0) at 65°C for 60 min. The GAGs liberated in this manner were precipitated by the addition of 10 vol. 1% (w/v) cetylpyridium chloride (CPC) and centrifuged at 10,000 × g for 10 min. The pellets obtained were dissolved in 500 µl of 60 (v/v) propanol-1 containing 0.4% (w/v) CPC. The liberated GAGs were reprecipitated by the addition of 6 vol. of 95% (v/v) ethanol containing 2.5% (w/v) sodium acetate. The precipitates were then washed with ethanol and allowed to dry. For the identification of galactosaminoglycans (GalAGs)., i.e., chondroitin sulfate (CS) and/or dermatan sulfate (DS), the GAG preparation was dissolved in water and digested with an equi-unit mixture (0.2 U/ml) of chondroitinases ABC and AC II. Aliquots from the supernatant were analyzed by reversed polarity high-performance capillary electrophoresis (HPCE), as previously described (42). The determination of HS was carried out in the GAG preparations which were digested with heparin lyases I, II and III in combination (0.05 U/ml) in 20 mM acetate buffer, pH 7.0, containing 1 µmol calcium acetate at 37°C for 90 min (43). In all cases, the amount of GAGs was determined from the integrated peak area of the GAG-derived Δ-disaccharides.

Heparitinase treatments were performed for the digestion of B6FS cell HS chains, as previously described (24,26). In brief, cells seeded in 24-well plates were serum-starved for 24 h and then treated with heparitinase (0.001 U/ml) for 24 and 48 h in 0% FBS medium. The cell extracts treated with heparitinase were electrophoresed on 8% polyacrylamide Tris/glycine gels and transferred onto nitrocellulose membranes. After blocking, the membranes were incubated for 1 h at room temperature with primary antibodies [mouse anti-heparitinase stubs antibody (clone 3G10; 1:500; CF500913; Seigakagu, Tokyo, Japan); goat anti-actin (1:200; sc-1616; Santa Cruz Biotechnology, Inc.) and mouse anti-CSA (1:200; C8035; Sigma)]. The immune complexes were detected following incubation with peroxidase-conjugated anti-goat or anti-mouse antibody, 1:4,000 or 1:2,000, respectively, with the SuperSignal West Pico Chemiluminescent substrate (Pierce Biotechnology, Inc.)

For the cell attachment assay, we used cells transfected with siFAK or cells treated with UHF and LMWH (for 48 h). The cells were detached with 5 mM PBS/EDTA. In continuation 5,000 cells/well were seeded onto a flat-bottom 96-well black plate. The bottom of the wells was coated with FN (5 µg/ml) for 1 h at 37°C. BSA 1% (30 min at room temperature) was added for the blocking of non-specific binding sites. The cells were allowed to adhere for 30 min at 37°C. The number of attached cells was determined using the CyQUANT fluometric assay (Molecular Probes; Invitrogen Life Technologies) according to the manufacturer's instructions.

To investigate the motility of cells, 24-well plates were used. Cells (60,000 cells/well) were seeded for 24 h in RPMI 10% FBS and subsequently treated (as described above) for 48 h. The cell stroma was wounded by scratching with a sterile 10 µl pipette tip. The detached cells were washed twice by using fresh RPMI (serum-free). The wound closure was monitored at 0 and 6 h using a digital camera (Canon Inc., Tokyo, Japan) connected to a microscope (Leica, Mannheim, Germany). The quantification of the wound area was measured using ImageJ software.

The B6FS cells were seeded onto a 96-well plate in RPMI supplemented with 10% FBS for 24 h incubation and subsequently, FITC-Heparin was added at 1 and 10 µg/ml for 24 h, in serum-free RPMI. Following this incubation, FITC-Heparin was removed and the cells were gently washed with RPMI (0% FBS). The concentration and the incubation times were chosen selected following optimization experiments (data not shown). FITC-Heparin internalization was visualized under a fluorescence microscope utilizing Leica DM2500 to acquire images.

The B6FS cells were seeded onto glass coverslips into 24-well plates (65,000 cells/well) and incubated with 10% FBS RPMI for 24 h. Following 24 h of serum starvation, the cells were treated with UFH and LMWH for 48 h. Subsequenlty, the cells were fixed in 5% formldehyde and 2% sucrose in PBS (incubation for 10 min at room temperature). Following 3 washes with PBS, Triton X permeabilizing agent was appled for 10 min at room temperature and washed prior to the addition of fluorescent phalloidin (1:100; Molecular Probes; Invitrogen Life Technologies) for 20 min in the dark. Fluorescent phalloidin was used for the detection of actin filaments. TO-PRO-3 (T3605; Molecular Probes/Thermo Fisher Scientific, Waltham, MA, USA) was then used for the nuclear staining (20 min in the dark). The coverslips were placed onto slides using glycerol and images were collected using a laser-scanning spectral confocal microscope (TCS SP2; Leica), LCS Lite software (Leica) and a 63 Apochromat 1.40 numerical aperture oil objective.

Statistical significance was evaluated by one-way ANOVA analysis with Turkey's post-test, using GraphPad Prism (version 4.0) software. A value of p<0.05 was considered to indicate a statistically significant difference.

Initially, we examined the effects of UFH and LMWH on FN-dependent B6FS cell adhesion. As shown in Fig. 1A, both UFH and LMWH enhanced B6FS cell adhesion (p<0.01, p<0.05); the maximal effect being evident at the concentration of 10 µg/ml. Of note, the molecular weight of the heparin preparations did not modify their effects on cellular function. Thus, UFH and LMWH had the same promoting effect on B6FS cell adhesion. This suggests that the effects of heparin are attributed to its oligosaccharide structure.

In continuation, we analyzed the effects of heparin on B6FS fibrosarcoma cell migration. Subsequent to pre-treatment with UFH and LMWH (10 µg/ml) an enhancement of fibrosarcoma cell migration was observed (p<0.05; Fig. 1B and C). To exclude false-positive results due to traces of heparin-binding growth factors, heat treatment was applied for the utilized heparin preparation. The effects of heparin on cell adhesion/migration were not affected by heat treatment (data not shown).

FAK is a 125-kDa cytoplasmic tyrosine kinase protein which is primarily positioned at adhesion sites and plays a key role in the processes of cell adhesion and migration (44). Importantly, FAK expression has been shown to be closely associated with (even in early reports) aggressive cancer behavior (45). To examine the putative participation of FAK in heparin-dependent adhesion, the B6FS cells were transfected with siRNA specific for the FAK gene as previously described (41). This approach resulted in an efficient downregulation of FAK both at the mRNA (p<0.01) and protein level (p<0.05) (Fig. 2A–C). Our results demonstrated that the FAK-deficient cells exhibited a marked decrease in their basal level ability to adhere to the FN substrate (p<0.05; Fig. 2D). Importantly, both the UFH and LMWH stimulatory effects on B6FS cell adhesion were abolished in the FAK-deficient cells (p<0.05 and p<0.01, respectively) (Fig. 2D and E). Therefore, the effect of heparin on fibrosarcoma cell adhesion was FAK-dependent.

Heparin has been shown to antagonize the function of integrin adhesion receptors (46), which trigger FAK clustering (47), and to have an effect on FAK expression and activation (33,34). Thus, we investigated the possible role of heparin in FAK expression and activation. To this end, the heparin-treated B6FS cell extracts were probed with antibodies against FAK and p-FAK. As shown in Fig. 3A and B, both UFH and LMWH significantly activated FAK (Y397) at 10 µg/ml, whereas no effect of heparin on total FAK protein expression was evident (Fig. 3A and B).

NF-κB has been previoulsy demonstrated to stimulate FAK promoter activity and upregulation (41). Heparin has been suggested to modulate NF-κB translocation to the nucleus and its transcriptional activity (48). In this study, we examined the possible role of NF-κB in heparin-induced FAK activation by using a specific antibody for the p65 subunit of NF-κB. As shown in Fig. 3C and D, UFH and LMWH did not affect the protein expression of the p65 subunit of NF-κB. Thus, p65 does not participate in the heparin-induced upregulation of FAK (Fig. 3C and D).

The phenotype of cancer cells is directly associated with their adhesive and migratory properties, which are of utmost importance for the process of cancer metastasis (49). FAK, has been postulated as the key conduit point which regulates the flow of signals from the ECM to the actin cytoskeleton (50). In this study, we investigated the possible effect of heparin on actin cytoskeleton organization in B6FS fibrosarcoma cells. To this end, rhodamine-conjugated phalloidin staining was used. The results of confocal miscroscopy indicated that B6FS cell actin cytoskeleton organization was enhanced. Thus, both the HMW- and LMWH-treated cells were found to have the fully spread, adhesive phenotype with the cells being stretched by the tensile forces of actin stress fibers (Fig. 4).

In continuation, we hypothesized that heparin affects fibrosarcoma cell motility and cell attachment through its internalization. Indeed, the cellular internalization of heparin has been demonstrated in previous studies (51,52). In this study, FITC-Heparin was utilized to investigate the internalization of heparin in B6FS fibrosarcoma cells. As shown in Fig. 5, FITC-Heparin was taken up by the B6FS cells in a dose-dependent manner, with the majority of cells exhibiting strong staining. The results of immunoflourescence demonstrated that FITC-Heparin was located, not only in the cytoplasmic region, but also in the nucleus of fibrosarcoma cells.

HS exhibits structural similarities with heparin, but does not have as many 'highly sulfated' sequences consisting of tri-sulfated disaccharide units, which are sequences with the highest binding affinity to heparin/HS-binding proteins (53). Biochemical analyses of the metabolically [3H]-labeled CS/DS/HS/PGs produced by the B6FS cells demonstrated that the HS amount was 63.1% of the total B6FS cell-associated GAGs (data not shown). Therefore, we examined the possible involvement of endogenous HS chains on B6FS cell adhesion. Initially, treatment of the cells with heparitinase and concomitant blotting with the monoclonal antibody 3G10, which recognizes a neoepitope generated by heparitinase digestion, demonstrated an efficient removal of HS chains at the 24 h time point (Fig. 6A and B). The specific digestion of cell-associated HS with heparitinase did not affect their adhesion to FN (Fig. 6C). These results suggest that B6FS cell adhesion is independent of their cell-associated HS content.