Figure 1.
mAb 4C5 is a kappa L- chain dimer.
A. Electrophoretic analysis of mAb 4C5, followed by immunoblotting with an anti-mouse kappa chain, an anti-mouse Fab and an anti-Fcγ antibody. Intact IgG1 immunoglobulin produced by the 2D10 hybridoma cells serves as positive control. Under reducing electrophoresis followed by western blot with the anti-Fab antibody, a single 25 kDa-immunoreactive band is observed, instead of the 25- and 50-kDa bands corresponding to the L- and H-chain, respectively, of an intact IgG1. This 25 kDa band is identical to the band corresponding to the kappa L-chain as shown by western blot with the anti-mouse kappa chain antibody. Under-non reducing electrophoresis followed by western blot with both the anti-kappa and the anti-Fab antibodies, mAb 4C5 is shown to migrate at approximately 50 kDa, and not at 150 kDa as expected for an intact IgG1 molecule. No mAb 4C5 immunoreactivity is detected after electrophoresis under both reducing and non-reducing conditions followed by western blot with an anti-Fcγ antibody. In the case of the IgG1 immunoglobulin under the same conditions a 50 kDa and a 150 kDa band is observed, respectively. B. No radioactivity is detected after northern blot analysis of 4C5 hybridoma-derived RNA with a H-chain radiolabelled probe. RNA derived from the IgG1-producing 2D10 hybridoma cells and the NSO myeloma cells served as positive and negative control, respectively.
Figure 2.
Nucleotide and aminoacid sequences of mAb 4C5.
Nucleotide and amino acid sequences of the L-chain variable region of mAb 4C5.
Figure 3.
Electrophoretic mobility and specificity of antibodies.
A. Left panel: SDS-PAGE of purified antibodies under reducing conditions followed by Coomasie Brilliant Blue-R staining revealed in all cases an approximately 25 kDa band corresponding to the L-chain. Right panel: Under non-reducing conditions the antibodies are shown to migrate as a L-chain dimer. B. Western blot of MDA-MB-453 cell lysates using mAb 4C5, rec-4C5, ch-4C5 and a commercial anti-HSP90α antibody, serving as positive control. In all cases a single 90 kDa immunoreactive band corresponding to HSP90 is observed. C. Immunoprecipitation in MDA-MB-453 cell lysates with anti-HSP90α, followed by immunoblotting with either the murine or the recombinant antibodies. In all cases a single immunoreactive band is observed. D. Reverse immunoprecipitation experiments in MDA-MB-453 cell lysates using mAb 4C5, rec-4C5 and ch-4C5, followed by western blot with anti-HSP90α. In all cases a single immunoreactive band is observed indicating that both recombinant antibodies recognize HSP90.
Figure 4.
ch-4C5 binds to the surface pool of HSP90 and is not internalized by MDA-MB-453 cells.
A. Immunofluoresence labelling of MDA-MB-453 cells using both the rec-4C5 and ch-4C5. The punctuate immunolabeling indicates the surface pool of HSP90. Negative controls were performed using an antibody against the intracellular protein β tubulin (data non shown). Scale bar: 20 µm B. Immunofluoresence detection of intracellular HSP90 in fixed MDA-MB-453 cells, permeabilized with 0.1% Triton X-100. Similarly to the murine antibody, rec-4C5 and ch-4C5 recognize the intracellular pool of HSP90. Scale bar: 20 µm C. For detection of antibody internalization, live cells were incubated with the antibodies for various time intervals, then fixed, permeabilized and fluorescently labelled. No internalization of mAb4C5, rec-4C5 and ch-4C5 is observed even after 24 h of incubation. In contrast the anti-HSP90α antibody could be detected intracellularly after 24 h of incubation. Scale bar: 20 µm D. MDA-MB-453 cell lysates, treated with mAb 4C5, rec-4C5, ch-4C5 and anti-HSP90α were analyzed by western blot using antibodies against ErbB2, Akt, cRaf and HSP90α. Actin served as a loading control. Presence of mAb 4C5, rec-4C5 and ch-4C5 did not affect the levels of the kinases compared to controls. In contrast the cell permeable anti-HSP90α antibody significantly reduced the levels of ErbB2, Akt and cRaf.
Figure 5.
ch-4C5 inhibis cancer cell invasion in vitro.
A. Wound healing assay. Photographs represent phase-contrast images obtained at zero time (left panel) and at 24 h (right panel) after scratch formation, showing MDA-MB-453 cell migration either in control cultures or cultures including anti-HSP90α, mAb 4C5, rec-4C5 or ch-4C5. Scale bar: 200 µm. B. Quantitative effect of antibodies on the closure of the wound. Addition of 200 µg/ml of anti-HSP90α and mAb 4C5 in the culture medium resulted in a 48% and 55% reduction of wound closure, respectively when compared to control cultures that were considered as resulting in 100% wound closure. Addition of 200 µg/ml rec-4C5 and ch-4C5 in the culture medium resulted in a 46% and 46% inhibition of wound closure, respectively. Statistical significance of differences was assessed by Student's t test. The presence of anti-HSP90α, mAb 4C5, rec-4C5 or ch-4C5 had a statistically significant effect on the wound closure (p<0.01 in each case). C. Phase-contrast images obtained at zero time (left panel) and at 24 h after scratch formation (right panel), showing B16 F10 melanoma cell invasion in a wound healing assay in the presence of 200 µg/ml of ch-4C5. Scale bar: 200 µm. D. Quantitative effect of increasing concentrations of ch-4C5 on the invasion level of B16 F10 melanoma cells. Presence of 50 µg/ml ch-4C5 resulted in 15% inhibition of invasion, while addition of 100 µg/ml and 200 µg/ml ch 4C5 resulted in 21% and 43% inhibition of migration, respectively when compared to control cultures that were considered as resulting in 100% wound closure. E. Visualization of dead cells using trypan blue dye. In ch-4C5 treated cultures, the cell death incidence is similar to that observed in the control cultures. In contrast, in the cultures treated with the anti-HSP90α antibody, a much greater number of cells are stained with Trypan blue, indicating an increased incidence of cell death Scale bar, 30 µm. F. Control and ch-4C5 treated cells were fixed permeabilized and stained with fluorescently labelled phalloidin. Scale bar 40 µm G. Higher magnification showing phalloidin staining (F-actin). Ch-4C5 effectively blocks spreading of lamellipodia. Scale bar: 16 µm.
Figure 6.
ch-4C5 inhibits the metastatic deposition of MDA-MB-453 cancer cells into the lungs of SCID mice.
Control or ch-4C5 treated MDA-MB-453 cells were labeled with the fluorescent dye DiO and injected into SCID mice as described in Materials and Methods. Evaluation of metastatic deposits was performed several hours later. A. Representative cryosections of the lungs of control and ch-4C5 treated mice. The arrows show MDA-MB-453 cells stained with DiO present in the lung tissue. A significant decrease in the deposition of cancer cells was observed in the lungs of ch-4C5 treated mice. Scale bar: 100 µm B. Quantitative effect of ch-4C5 on the metastatic deposition of MDA-MB-453 cells into the lungs, showed a 38% inhibition when compared to control animals. The bars represent the average of two independent experiments ±SEM. Statistical significance of differences was tested by Student's t test (p<0.01).