The Glycosyltransferase ST6Gal-I Protects Tumor Cells against Serum Growth Factor Withdrawal by Enhancing Survival Signaling and Proliferative Potential

Abstract

A hallmark of cancer cells is the ability to survive and proliferate when challenged with stressors such as growth factor insufficiency. In this study, we report a novel glycosylation-dependent mechanism that protects tumor cells from serum growth factor withdrawal. Our results suggest that the β-galactoside α-2,6-sialyltransferase 1 (ST6Gal-I) sialyltransferase, which is up-regulated in numerous cancers, promotes the survival of serum-starved cells. Using ovarian and pancreatic cancer cell models with ST6Gal-I overexpression or knockdown, we find that serum-starved cells with high ST6Gal-I levels exhibit increased activation of prosurvival signaling molecules, including pAkt, p-p70S6K, and pNFκB. Correspondingly, ST6Gal-I activity augments the expression of tumor-promoting pNFκB transcriptional targets such as IL-6, IL-8, and the apoptosis inhibitor cIAP2. ST6Gal-I also potentiates expression of the cell cycle regulator cyclin D2, leading to increased phosphorylation and inactivation of the cell cycle inhibitor pRb. Consistent with these results, serum-starved cells with high ST6Gal-I expression maintain a greater number of S phase cells compared with low ST6Gal-I expressors, reflecting enhanced proliferation. Finally, selective enrichment in clonal variants with high ST6Gal-I expression is observed upon prolonged serum deprivation, supporting the concept that ST6Gal-I confers a survival advantage. Collectively, these results implicate a functional role for ST6Gal-I in fostering tumor cell survival within the serum-depleted tumor microenvironment.

Keywords: cell cycle; glycosylation; growth factor; sialyltransferase; β-galactoside α-2,6-sialyltransferase 1 (ST6Gal-I).

FIGURE 1.

Cells with high ST6Gal-I expression are resistant to stress induced by serum withdrawal. A, OV4 cells were stably transduced with a lentivirus encoding ST6Gal-I, and ST6Gal-I OE was confirmed by immunoblotting. Control cells were generated by stable transduction of an EV lentiviral construct. B, OV4 EV and OE cells were stained with SNA-FITC and evaluated by flow cytometry. C, OV4 EV and OE cells were stained with MAA-FITC and evaluated by flow cytometry. D, OV4 EV and OE cells were cultured in 10%, 1%, or 0% FBS-containing medium for 1 week and imaged to observe differences in cell morphology. E, BxPC3 cells were stably transduced with shRNA for ST6Gal-I using a lentivirus, and ST6Gal-I KD was verified by immunoblotting. F, BxPC3 cells were stained with SNA-FITC and evaluated by flow cytometry. G, BxPC3 cells were stained with MAA-FITC and evaluated by flow cytometry. H, EV and KD BxPC3 cells were cultured in 10% or 1% FBS-containing medium for 1 week and then imaged.

FIGURE 2.

ST6Gal-I enhances survival signaling in serum-starved cells. A, OV4 cells were serum deprived for 1 week and then immunoblotted for survival (pAkt and cIAP2) or cell death (Beclin and AIF) markers. Representative blots are shown. B, densitometric analyses of p-Akt, total Akt, cIAP2, AIF, and Beclin. Three distinct blots (from three independently generated cell lysates) were evaluated using ImageJ, and data were plotted as mean ± S.D. *, p < 0.05. For Akt, pAkt and total Akt were each normalized to β-tubulin, and then data were plotted as pAkt/total Akt (p/t Akt). For cIAP2, AIF, and Beclin, values were normalized to β-tubulin. C, BxPC3 EV and KD cells were serum-deprived for 1 week and immunoblotted for survival or cell death markers as in A. D, densitometric analyses of three independent blots for pAkt, total Akt, cIAP2, AIF, and Beclin. Data were normalized to β-tubulin as described in B. *, p < 0.05.

FIGURE 3.

ST6Gal-I levels are enriched in serum-starved cells. A, representative immunoblot showing an increased level of ST6Gal-I in OV4 OE cells upon serum starvation. B, densitometry of three independent blots for ST6Gal-I in OV4 cells. C, representative immunoblot showing an increased level of ST6Gal-I in BxPC3 EV cells upon serum starvation. D, densitometry of three blots for ST6Gal-I in BxPC3 cells. All densitometric values were normalized to β-tubulin. *, p < 0.05.

FIGURE 4.

ST6Gal-I activity promotes the activation of prosurvival molecules upon acute serum deprivation. A, OV4 cells were serum-deprived for 4, 18, or 24 h and then immunoblotted for pAkt, total Akt, p-p70S6K, and total p70S6K. B, densitometric analyses for pAkt and total Akt for three blots. Values for pAkt and total Akt were each normalized to β-tubulin, and then data were plotted as pAkt/total Akt. *, p < 0.05. C, densitometric analyses for p-p70S6K and total p70S6K for three blots. Values for p-p70S6K and total p70S6K were each normalized to β-tubulin, and then data were plotted as p-p70S6K/total p70S6K. *, p < 0.05.

FIGURE 5.

ST6Gal-I knockdown inhibits activation of prosurvival molecules upon acute serum deprivation. A, BxPC3 cells were serum-deprived for 4, 18, or 24 h and then immunoblotted for pAkt, total Akt, p-p70S6K, and total p70S6K. B, densitometric analyses for pAkt and total Akt from three blots. Values for pAkt and total Akt were each normalized to β-tubulin, and then data were plotted as pAkt/total Akt. *, p < 0.05. C, densitometric analyses for p-p70S6K and total p70S6K from three blots. Values for p-p70S6K and total p70S6K were each normalized to β-tubulin, and then data were plotted as p-p70S6K/total p70S6K. *, p < 0.05.

FIGURE 6.

ST6Gal-I activity up-regulates the expression of NFκB pathway proteins. A and B, OV4 (A) and BxPC3 (B) cells were grown in 10% or 1% FBS for 24 h and then immunoblotted for pNFκB and total NFκB (p65). C, densitometry for three OV4 immunoblots. Values for pNFκB and total NFκB were each normalized to β-tubulin. pNFκB and total NFκB were graphed separately (rather than phospho/total) because differences were noted in total NFκB levels. *, p < 0.05. D, densitometry for three BxPC3 immunoblots. Values for pNFκB and total NFκB were each normalized to β-tubulin. *, p < 0.05. E and F, OV4 (E) and BxPC3 (F) cells were serum-starved for 6 or 24 h and then analyzed by qRT-PCR for various prosurvival molecules known to be transcriptionally up-regulated by NFκB. Values represent the averages of at least four independent experiments ± S.D., with each independent experiment performed in triplicate. *, p < 0.05.

FIGURE 7.

ST6Gal-I regulates cyclin D2 expression and pRb phosphorylation. A and B, OV4 (A) and BxPC3 (B) cells were serum-starved for 6 or 24 h, and cyclin D2 mRNA was measured by qRT-PCR. Data are from at least four independent experiments, with each independent experiment performed in triplicate. *, p < 0.05. C and D, OV4 (C) and BxPC3 (D) cells were serum-starved for 24 h, and cell lysates were immunoblotted for cyclin D2. E and F, OV4 (E) and BxPC3 (F) immunoblots of cyclin D2 were evaluated by densitometry (3 blots/cell line). *, p < 0.05. G and H, OV4 (G) and BxPC3 (H) cells were immunoblotted for the downstream cyclin D2 target pRb. I and J, OV4 (I) and BxPC3 (J) immunoblots for p-Rb and total pRb were evaluated by densitometric analyses of three independent blots. Values for p-pRb and total pRb were normalized to β-tubulin and plotted separately because differences were noted in both phosphorylated and total pRb. *, p < 0.05.

FIGURE 8.

ST6Gal-I activity promotes cell cycle progression in serum-starved cells. To assay for cell cycle progression, cells in S phase were labeled with the Click-it EdU reagent (Thermo) and then quantified by flow cytometry. A, representative experiment showing a greater percentage of S phase cells in serum-starved OV OE compared with EV cells. B, three independent flow cytometry experiments measuring the percentage of cells in S phase. *, p < 0.05. C, representative experiment showing a greater percentage of S phase cells in serum-starved BxPC3 EV versus KD cells. D, three independent flow cytometry experiments measuring the percentage of cells in S phase. *, p < 0.05.