Fig. 1. Secretion of galectin-3 by detaching and uptake by spreading 549-Gal-3 cells. In panel A, the carcinoma cells were plated in 12-well microtiter plates at 5 × 10⁵ cells/well and allowed to grow for 2 days in the cell incubator. After 48 h, the growth medium was replaced with 200 μl of SFM without and with 2.5 mM EDTA for 5, 10, 30, and 60 min. At the indicated time points, 100 μl of the conditioned medium from each well was assayed for galectin-3 by Western blot analysis, and the cells adhered in the wells fixed, stained with crystal violet, and washed with water. The dye was released from the cells and absorbance at 570 nm determined as described in Materials and methods (B). In panel C, the cells were also allowed to adhere in the 12-well microtiter plate (2.5 × 10⁵ cells/well) in 400 μl of SFM in the presence (lanes 1 and 2) and absence (lanes 3 and 4) of 2.5 mM EDTA. After 6 h of incubation, the conditioned medium was assayed for galectin-3.

Fig. 2. Galectin-3 in the conditioned medium of adherent and non-adherent rounded MDA-MB-435 cells. In panel A, MDA-MB-435 (435 V) cells were plated in 12-well microtiter plates at 5 × 10⁵ cells/well and allowed to grow for 2 days in the cell incubator. The complete medium was replaced with serum free medium (200 μl/well) in the absence or presence of 2 mM EDTA. After 30 min of incubation, galectin-3 in the conditioned media was measured by Western blot. In panel B, the same experiment as panel A was repeated with 435 AS. In panel C, MDA-MB-435 (435V) cells were allowed to adhere and spread in complete medium in 12-well microtiter plates (2.5 × 10⁵ cells/well; 200 μl/well) in the absence (lane 1) or presence of 2.5 mM EDTA (lane 2) for 3 h. At the end of the incubation period, the conditioned medium was centrifuged to pellet cells and the relative concentration of galectin-3 was determined as above. The same number of cells (2.5 × 10⁵ cells/tube) was lysed in 200 μl of RIPA lysis buffer containing protease inhibitors and the relative concentration of galectin-3 was determined by Western blot as above (lane 3).

Fig. 3. Speed of adhesion and spreading as a function of galectin-3 expression. The cells (BT-549, 549-PCN, and 549 Gal-3) were plated in 96-well microtiter plate at 4 × 10⁴ cells/well. In panels A and B, after 10 min, the non-adherent cells were washed off and the adherent cells fixed and stained with crystal violet and photographed. In panel C, the stain was removed and the used to determine the relative numbers of cells adhered as described in Materials and methods. In panel D, the BT-549 cells and its sub-clones were incubated in eppendorf tubes (1 × 10⁶ cells/tube) for 2 h. At the end of the incubation, the tubes were centrifuged and galectin-3 in the conditioned medium determined by Western blot. The lane assignments were 549 Gal-3 (lanes 1–3), BT-549 (lane 5), and 549-PCN (lanes 6 and 7).

Fig. 4. The speed of adhesion and spreading of 435 V/435 AS cells. The cells 435 AS and 435 V were plated in 96-well microtiter plates at 4 × 10⁴ cells/well in quadriplicates. After 10 min of incubation at 37 °C in panel A, the non-adherent cells were washed off and the adherent cells fixed, and stained with crystal violet. The stain was subsequently dissolved in 10% acetic acid and O.D. at 570 nm determined. In panel B, the same experiment as panel A was repeated with the exception that the cells were allowed to adhere for 7 h before fixing and assaying for adherent cells.

Fig. 5. Uptake of galectin-3 by LNCaP and 549 Gal-3 cells tumor cells. The LNCaP prostate cancer cells were plated in fibronectin coated chambered glass slides and allowed to adhere overnight in complete RPMI 1640 medium. The medium was replaced with serum free medium without (control) or with 5 μM galectin-3 and incubated for 30 min at 37°C. The cells were then fixed with cold methanol (permeabilized) or 3% paraformaldehyde (non-permeabilized) and then incubated with rat monoclonal antibodies against galectin-3 and finally with FITC-labeled anti-rat secondary antibodies. In panel B, 549 Gal-3 tumor cells were allowed to adhere and spread in chambered glass slides in the presence of recombinant galectin-3 labeled with rhodamine isothiocynate (5 μM) in the presence of 50 mM sucrose or 50 mM lactose. After 1 h of incubation, the cells were examined by a fluorescence microscope and images were captured digitally.

Fig. 6. Modulation of galectin-3 expression in the conditioned medium by growth factors. In panel A, the 549 Gal-3 cells were detached, washed, and suspended in SFM without growth factors for 10 min at RT. The cells were divided into 7 eppendorf tubes (5 × 10⁵ cells/tube; 500 μl/tube), pelleted, SFM aspirated, and replaced with 200 μl/tube of fresh SFM without growth factors (control #1) and with: cholera toxin; EGF; fetuin; hydrocortisone; insulin; and all the five growth factors (control #2). Panel B is the densitometric scans of the gel in panel A. In panel C, the cells were detached and washed in SFM without growth factors and allowed to adhere to micro-titer wells in: complete medium; SFM containing cholera toxin, EGF, hydrocortisone, and insulin; and in SFM without growth factors.

Fig. 7. Inhibition of cellular adhesion and spreading and galectin-3 uptake by HPBCD in SFM but not complete medium. In panel A, the 549 Gal-3 cells were allowed to adhere to the wells of a 96-well microtiter plate (4 × 10⁴ cells/well) in SFM containing insulin, EGF, hydrocortisone, and cholera toxin or complete medium in the absence and presence of graded doses of HPBCD (0–14 mM). After 2 h incubation, the relative numbers of adherent cells were determined as described in Materials and methods. In panel B, the cells that adhered to wells in SFM without and with 7 mM HPBCD were fixed and stained as described and photographed. In panel C, the cells we allowed to adhere to 12-well plates (2.5 × 10⁵ cells/well; 400 μl/well) in SFM without and with 7 mM of HBCD for 2 h and the conditioned medium assayed for galectin-3.

Fig. 8. Remodeling of adhesion plaques and cytoskeleton by galectin-3. The galectin-3 expressing breast carcinoma cell lines, 549 Gal-3 (A) and 435 V (C), as well as null cell lines, BT-549 (B) and 435 AS (D), were plated in 8-chamber glass slides. At approximately 60% confluence, the cells were fixed briefly in a solution of PBS containing 3% paraformaldehyde and 0.5% Triton X-100. They were then incubated with a mouse monoclonal antibody against human vinculin followed by secondary antibodies labeled with Cy3. The slides were processed and visualized under a fluorescence microscope. The adhered and spread 435 AS cells were also incubated without (E) and with 5 μM galectin-3 in SFM (F) for 10 min. They were then fixed and expression of adhesion plaques (vinculin) was examined as above and images captured digitally.

Fig. 9. Galectin-3 modulates the growth of breast carcinoma cells in Matrigel. The galectin-3 expressing 435 V (A) and its subclone 435 AS (B) in which galectin-3 expression is knocked down (2 × 10⁴ cells/well), were allowed to grow on matrigel in 96-well microtiter plate for 7 days. The cells were then photographed.