SGC7901 GC cells, purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA), were maintained in RPMI-1640 medium with 10% FBS and 1% antibiotic solution containing penicillin/streptomycin (all from Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA). All cells were cultured in a humidified incubator at 37°C with 5% CO₂. The stable ST6Gal-I overexpression, knockdown and empty vector cell lines were established as previously described (24). In brief, the pcDNA3.1(−)/ST6Gal-I, sh-ST6Gal-I and empty vectors were transfected into SGC7901 cancer cells using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). Limiting dilution was applied to obtain sub-cell line clones after 24 h of transfection. Blasticidin S HCl (1.5 µg/ml) was used to select the low-expressing ST6Gal-I clone and G418 (350 µg/ml) was utilized to select the ST6Gal-I overexpression clone. ST6Gal-I overexpression or knockdown was verified by ST6Gal-I mRNA expression and protein expression analyses.

Total RNA was isolated using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.), and cDNA was synthesized using the PrimerScript^® RT Master Mix kit (Takara Bio, Inc., Otsu, Japan) according to the manufacturer's protocol. qRT-PCR was performed on a Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The PCR condition was as following: 95°C for 5 min, followed by 40 cycles of 95°C for 15 sec, 50°C for 30 sec and 72°C for 30 sec. The sequences of primers used for the real-time PCR assays were as follows: ST6Gal-I: forward, 5′-CCTCTGGGATGCTTGGTATC-3′ and reverse, 5′-GTGCAGGCACTATCGAAGAA-3′; GAPDH: forward, 5′-AGCCTCAAGATCATCAGC-3′ and reverse, 5′-GAGTCCTTCCACGATACC-3′.

We conducted lectin staining to assess ST6Gal-I activity. Briefly, the cells were stained with FITC-conjugated SNA lectin (EY Laboratories, San Mateo, CA, USA), which is specific for 2,6-sialic acids, according to the manufacturer's instructions. Cells were stained for 40 min at 4°C with SNA-FITC at a 1:200 dilution and analyzed by fluorescence-activated cell sorting (FACS; BD Biosciences, Franklin Lakes, NJ, USA). In addition, cells were stained with SNA-FITC at a 1:100 dilution for 4 h for the cell immunofluorescence assay.

To assess the proliferation of the transfected cloned cell lines, a CCK-8 detection kit (Dojindo Laboratories, Kumamoto, Japan) was used according to the manufacturer's instructions. Approximately 3,000 cells were seeded onto a 96-well plate in quintuplicate for 6 h, and complete medium was then replaced with RPMI-1640 medium without fetal bovine serum (FBS) for 0, 24, 48, 72, 96, 120 or 144 h. Next, 10 µl of CCK-8 reagent was added to each well, and the absorbance was measured at 450 nm using a Multiskan Spectrum spectrophotometer (BioTek Instruments, Inc., Winooski, VT, USA). To assess drug resistance, cells were seeded and incubated with different doses of trastuzumab (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) for 24 h and CCK-8 assays were then performed.

To assess the cell cycle, ST6Gal-I overexpression or ST6Gal-I knockdown cells were seeded onto 6-well plates and cultured in medium containing no FBS for 24 h. The cells were harvested, fixed with 70% cold ethanol at 4°C for 2 h, washed with ice cold phosphate-buffered saline (PBS) 3 times and incubated with staining solution (20 mg/ml propidium iodide and 200 µg/ml DNase-free RNase-A in PBS) at room temperature for 30 min at 37°C. The labeled cell cycle populations were quantified by FACSCalibur flow cytometry (BD Biosciences) and cell cycle distributions were analyzed by FlowJo software (Tree Star, Inc., Ashland, OR, USA). In addition to the cell apoptosis assays, cells were incubated with trastuzumab for 24 h, collected for staining, and stained with an FITC-labeled Annexin V and propidium iodide staining kit (BD Biosciences, San Jose, CA, USA) for 60 min at 37°C. Staining controls were prepared with fixed cells single stained and unstained; a positive apoptotic control was obtained by incubating the cells with 1 mM hydrogen peroxide for 15 h (Sigma-Aldrich; Merck KGaA). Cell populations stained by the 2 different reagents were recorded by FACS.

In vitro invasion assays were performed using Transwell-24 plates (Corning Inc., Corning, NY, USA); ~5×10⁴ transfected cells in 150 µl of RPMI-1640 medium without FBS were added to the upper compartment of the Transwell chamber coated with 40 µl of Matrigel (pore size, 8.0 µm; diameter, 6.5 mm), and 350 µl of RPMI-1640 medium supplemented with 20% FBS was added to the lower compartment as a chemoattractant. After incubation for 24 h at 37°C and 5% CO₂, the upper chamber was washed 3 times with PBS, and the cells that did not invade through the pores of the upper surface of the membrane were removed by cotton swabs. The invaded cells were subsequently fixed and stained with 0.1% crystal violet. After being washed with PBS and dried, the cells were visually counted in 4 randomly selected fields with an inverted microscope (Olympus IX71; Olympus Corp., Tokyo, Japan) at ×200 magnification.

Cells were incubated with trastuzumab for 24 h and collected for western blot analysis. The cells were lysed using cell lysis buffer, and protein concentrations were assessed with a BCA assay kit (both from Beyotime Institute of Biotechnology, Shanghai, China). Equal amounts of denatured proteins were separated by 10% SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were blocked with 5% non-fat milk and then incubated overnight with primary antibodies directed against ST6Gal-I (dilution at 1:500; cat. no. sc-6263), ERK (dilution at 1:1,000; cat. no. sc-514302), p-ERK (dilution at 1:500; cat. no. sc-81492), AKT (dilution at 1:1000; cat. no. sc-5298), p-AKT (dilution at 1:500; cat. no. sc-135650), HER2 (dilution at 1:500; cat. no. sc-7301), p-HER2 (dilution at 1:500; cat. no. sc-81507), caspase-3 (dilution at 1:400; cat. no. sc-7272; all from Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) and GAPDH (dilution at 1:2,000; cat. no. ab9484; Abcam, Cambridge, MA, USA), after washed with TBS-Tween (0.1%) three times, all membranes were then incubated with mouse-IgGκ BP-HRP secondary antibody (dilution at 1:2,000; cat. no. sc-516102; Santa Cruz Biotechnology, Inc.) for 1 h at room temperature. Proteins were visualized using an enhanced chemiluminescence (ECL) kit (Amersham Biosciences, Little Chalfont, UK) according to the manufacturer's instructions. The relative amounts of proteins were determined by densitometry using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Cells were lysed, and 500 µg of lysed protein was incubated overnight at 4°C with 50 µl of SNA-conjugated agarose (EY Laboratories). α2,6-Sialylated proteins bound to SNA-agarose beads were precipitated by centrifugation and washed extensively with lysis buffer. Sialylated proteins were released from complexes by boiling in SDS-PAGE sample buffer and immunoblotted for HER2 (Santa Cruz Biotechnology, Inc.).

Statistical analyses were performed using SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA). The data were presented as the mean value ± standard deviation (SD). Statistical comparisons between different groups were performed by one-way analysis of variance (ANOVA) with the Least Significant Difference (LSD) post hoc test and statistical comparisons between two groups were analyzed by t-test comparison. Differences were considered statistically significant when P-values were <0.05.

ST6Gal-I and HER2 are biomarkers associated with the malignant progression and poor prognosis of GC (25–27). We confirmed that HER2, p-HER2 and SNA-HER2 were expressed in SGC7901 GC cell lines, MCF7 breast cancer cell lines, and A549 and H460 lung cancer cell lines (Fig. 1A). To evaluate the role of ST6Gal-I-induced sialylation in regulating GC development, we constructed and transfected ST6Gal-I overexpression or ST6Gal-I knockdown plasmids into SGC7901 cells. After limiting dilution and persistent culture, we obtained several different sub-cell line clones with stable expression. The vector 2 cell line was selected as the control group, the X61-4 cell line was selected as the knockdown group, and the PS7 cell line was selected as the overexpression group. ST6Gal-I expression was further confirmed by PCR and western blotting. Endogenous ST6Gal-I gene and protein expression were stably knocked down in the X61-4 cell line, while the PS7 cell line expressed the ST6Gal-I gene and protein at high levels compared to that in the vector cell line (Fig. 1B-E). Collectively, these data strongly demonstrated that ST6Gal-I was stably overexpressed or knocked down in SGC7901 cells.

To assess the functional consequence of ST6Gal-I upregulation or downregulation in tumor cells, we measured the levels of α2,6-sialylation on the HER2 receptor. FITC-conjugated SNA lectin was used to recognize α2,6-linked sialic acids using flow cytometry and immunofluorescence microscopy. The FACS results indicated that the SNA fluorescence in PS7 cells was stronger than that in the X61-4 and vector groups (Fig. 2A and B). The immunofluorescence results also revealed that PS7 cells expressed significantly higher levels of α2,6-linked sialic acids than the vector cells, while X61-4 cells expressed lower levels of α2,6-linked sialic acids than the vector control cells (Fig. 2C and D). Moreover, X61-4 cells and PS7 cells were incubated with agarose-conjugated SNA lectin. The α2,6-sialylated proteins bound by SNA-agarose were isolated by SDS-PAGE and immunoblotted for HER2. The SNA-HER2 expression in PS7 cells was much higher than that in the vector group, and SNA-HER2 expression in X61-4 cells was lower than that in the vector group (Fig. 2E). Thus, overexpressing ST6Gal-I hypersialylated HER2, while knocking down ST6Gal-I hyposialylated HER2.

To investigate whether ST6Gal-I protected GC cells from serum starvation by regulating cell cycle progression, FACS was used to analyze the cell cycles of PS7 cells, X61-4 cells and vector cells. Representative FACS results for the three group cells are shown in Fig. 3A, and FACS data are summarized in Fig. 3B. Among the PS7 cells, 35.07% were in the S phase, which was much higher than that in X61-4 cells (26.6%) and vector control cells (30.71%). Accordingly, the proportion of PS7 cells remaining in the G2/M phase was 11.66%, compared to 6.74% for X61-4 cells and 8% for vector cells (P<0.01). Serum-starved cells highly expressing ST6Gal-I maintained more cells in the S phase than low ST6Gal-I-expressing cells, suggesting that ST6Gal-I expression promoted GC cells from the G0/G1 phase into the S phase, thus enhancing proliferation. Thus, disturbances in ST6Gal-I-mediated sialylation altered cell proliferation.

To further assess whether ST6Gal-I expression promoted the proliferation of GC cells, CCK-8 assays were conducted. The growth rates of PS7 and X61-4 cloned cells were increased and markedly decreased, respectively, compared to those of vector cells under serum starvation (Fig. 3C). Matrigel-coated Transwell assays were applied to detect the invasion abilities of the cloned cell lines. As shown in Fig. 3D and E, more PS7 clone cells penetrated the Matrigel-coated membrane than vector cells, and fewer ST6Gal-I knockdown cells penetrated the Matrigel-coated membrane than vector cells, indicating that ST6Gal-I-overexpressing cells exhibited a much higher invasion ability than vector cells (P<0.05). Collectively, these results indicated that ST6Gal-I may be an essential factor promoting GC cell proliferation and invasion.

For future therapeutic applications, we investigated whether HER2 sialylation decreases apoptosis and enhances resistance to trastuzumab treatment. FACS analysis and a CCK-8 assay were performed to analyze trastuzumab inhibition in GC cells. Trastuzumab inhibition was dose-dependent in each group, and the inhibition effect of trastuzumab on PS7 cells was lower than that on vector cells; however, X61-4 cells were inhibited at a much higher rate than vector cells. When the drug concentration reached 10 µM, each group exhibited evident cell death. Moreover, FITC-labeled Annexin V and propidium iodide staining were recorded by FACS to indicate cell apoptosis. As shown in Fig. 4B and C, a small population of PS7 cells were apoptotic (including 2.08% early-stage apoptosis and 9.10% late-stage necrosis), whereas a large population of X61-4 cells were apoptotic (76.3% early-stage apoptosis and 0.52% late-stage necrosis) under serum starvation. After incubation with trastuzumab, the number of apoptotic PS7 cells was not significantly increased (11.90% with treatment vs. 11.18% without), but the number of apoptotic X61-4 cells was distinctly increased (88.04% with treatment vs. 76.82% without). In addition, caspase-3 was reportedly the executioner of apoptosis (28). Our western blot results revealed much higher caspase-3 expression in X61-4 cells compared to that in PS7 cells, and trastuzumab treatment increased caspase-3 expression in all three groups compared to that in cells not treated with trastuzumab (Fig. 4D). Collectively, elevated HER2 α2,6 sialylation reduced cell apoptosis and increased resistance to trastuzumab treatment. In addition, HER2 α2,6 sialylation promoted GC cell survival against many antitumor treatments.

To investigate how HER2 sialylation affects the downstream signaling pathways exerting drug resistance, we examined the protein levels of HER2, Akt and ERK. HER2 was sialylated by ST6Gal-I in cancer cells, and the SNA-HER2 level in PS7 cells was distinctly higher than that in X61-4 cells. HER2 sialylation was not attenuated with 10 µM trastuzumab treatment for 24 h, while HER2 sialylation in X61-4 cells was much lower after trastuzumab treatment (Fig. 5A and B). Phosphorylated HER2 expression was significantly higher in X61-4 cells than in PS7 cells, and the phosphorylation levels in the three groups were substantially decreased after trastuzumab treatment. The HER2 protein expression levels in the three groups were similar (Fig. 5A and C). It has been revealed that HER2 activation cοuld initiate the Akt and ERK signaling pathways by increasing Akt and ERK phosphorylation (29). Western blot analyses revealed that the Akt and ERK phosphorylation levels were significantly higher in PS7 cells than in X61-4 cells and vector cells, and trastuzumab treatment did not alter the phosphorylation levels in any of the groups. Furthermore, the total Akt and Erk repression levels were consistent among all the groups (Fig. 5A, D and E). Collectively, our results indicated that overexpressing ST6Gal-I in cancer cells elevated HER2 α2,6 sialylation, activating the downstream cascade, increasing the phosphorylation of Akt and ERK, and inhibiting the effect of trastuzumab treatment.