Regulation of cytoskeleton organization by sphingosine in a mouse cell model of progressive ovarian cancer

Abstract

Ovarian cancer is a multigenic disease and molecular events driving ovarian cancer progression are not well established. We have previously reported the dysregulation of the cytoskeleton during ovarian cancer progression in a syngeneic mouse cell model for progressive ovarian cancer. In the present studies, we investigated if the cytoskeleton organization is a potential target for chemopreventive treatment with the bioactive sphingolipid metabolite sphingosine. Long-term treatment with non-toxic concentrations of sphingosine but not other sphingolipid metabolites led to a partial reversal of a cytoskeleton architecture commonly associated with aggressive cancer phenotypes towards an organization reminiscent of non-malignant cell phenotypes. This was evident by increased F-actin polymerization and organization, a reduced focal adhesion kinase expression, increased a-actinin and vinculin levels which together led to the assembly of more mature focal adhesions. Downstream focal adhesion signaling, the suppression of myosin light chain kinase expression and hypophosphorylation of its targets were observed after treatment with sphingosine. These results suggest that sphingosine modulate the assembly of actin stress fibers via regulation of focal adhesions and myosin light chain kinase. The impact of these events on suppression of ovarian cancer by exogenous sphingosine and their potential as molecular markers for treatment efficacy warrants further investigation.

Figure 1

Sph treatment increases cytoskeleton organization. Mouse Ovarian Surface Epithelium (MOSE)-E and MOSE–L cells were grown in the absence (Ctrl) or presence of 1.5 µM sphingosine (So), fixed with paraformaldehyde and stained with AlexaFluor⁴⁸⁸-conjugated phalloidin to visualize filamentous actin (A) or with antibodies against ß-tubulin; (B) to visualize the microtubule network; (C) Quantitation of F-actin after 4–24 h treatment with 1.5 µM So, or continuous passaging. *significantly different from MOSE-E, p < 0.05, + different from untreated corresponding control, p < 0.05; (D) MOSE-E and L cells were treated with 1.0 µM enigmol for 3 passages and stained with AlexaFluor⁴⁸⁸-conjugated phalloidin; (E) MOSE-E and –L cells were treated with 500nM S1P for 8 h and stained with AlexaFluor⁴⁸⁸-conjugated phalloidin; (Original magnification X600).

Figure 2

Sph treatment affects protein but not mRNA levels of cytoskeleton genes and regulators in MOSE cells. MOSE-E and MOSE–L cells were passaged three times in the absence or presence of 1.5 µM So. (A) Real-time PCR analyses of changes in mRNA levels of select cytoskeleton genes. *significantly different from MOSE-E, p < 0.05; (B) Representative Western blot of MOSE cells treated with So; and (C) quantitated using γ-tubulin as as housekeeping protein; expressed as percent of MOSE-E levels ± SD. *significantly different from untreated MOSE-E, p < 0.05; + significantly different from corresponding untreated controls, p < 0.05.

Figure 3

Effects of Sph treatment on FAK phosphorylation. MOSE-E and MOSE-L cells were treated with sphingosine (So) or vehicle, and immunostained for FAK phosphorylated on (A) Tyr-397; or (B) Tyr-861; (C) Co-localization of non-phosphorylated FAK and FAK; (D) co-localization of FAK phosphorylated on Tyr-861 and Tyr-397

Figure 4

Increased focal adhesion assembly and maturation by treatment with non-toxic concentrations of Sph. MOSE-E and –L cells were treated with 1.5 µM sphingosine (So) or vehicle (Ctrl); FA were visualized by staining for FAK and foci were counted and measured in cells that were completely in the visual field using the NIS Elements software (Nikon). Data represent mean ± SEM. *significantly different from untreated MOSE-E, p < 0.001; ⁺significantly different from corresponding untreated control, p < 0.001.

Figure 5

Changes in MLCK expression and its target phosphorylation after Sph treatment. (A) Immunofluorescent identification of myosin light chain kinase (MLCK); arrow indicates co-localization with F-actin (left panel); Western blot analysis and quantitation of MLCK in whole cell extracts, normalized to to γ tubulin. A representative blot is shown in the right panel. Data are expressed as percent of MOSE-E ± SEM; *significantly different from MOSE-E, p < 0.05; +significantly different from corresponding untreated control, p < 0.05. (B) Myosin light chain expression (left panels) or myosin light chain phosphorylated at Ser-19 (pMyosin light chain).

Figure 6

Suppression of invasion by Sph treatment. Number of cells able to invade matrigel afte r sphingosine (So) treatment. Date are presented as mean ± SEM. * significantly different from MOSE-E, p < 0.01.