These authors contributed equally to this work.
Functional roles of glycogene and N-glycan in multidrug resistance of human breast cancer cells
Version of Record online: 26 FEB 2013
Copyright © 2013 International Union of Biochemistry and Molecular Biology, Inc.
Volume 65, Issue 5, pages 409–422, May 2013
How to Cite
Ma, H., Miao, X., Ma, Q., Zheng, W., Zhou, H. and Jia, L. (2013), Functional roles of glycogene and N-glycan in multidrug resistance of human breast cancer cells. IUBMB Life, 65: 409–422. doi: 10.1002/iub.1133
- Issue online: 22 APR 2013
- Version of Record online: 26 FEB 2013
- Manuscript Accepted: 10 DEC 2012
- Manuscript Received: 13 NOV 2012
- National Natural Science Foundation of China. Grant Number: 81071415
- Natural Science Foundation of Liaoning Province. Grant Number: 20102052
- Liaoning Excellent Talents in University. Grant Number: LR2011025
- multidrug resistance;
- human breast cancer cell line
Drug resistance is a major problem in cancer chemotherapy. Aberrant glycosylation has been known to be associated with cancer chemoresistance. Aim of this work is to investigate the alterations of glycogene and N-glycan involved in multidrug resistance (MDR) in human breast cancer cell lines. Using real-time polymerase chain reaction (PCR) for quantification of glycogenes, fluorescein isothiocyanate (FITC)-lectin binding for glycan profiling, and mass spectrometry for N-glycan composition, the expression of glycogenes, glycan profiling, and N-glycan composition differed between drug-resistant MCF/ADR cells and the parental MCF-7 line. Further analysis of the N-glycan regulation by tunicamycin (TM) application or PNGase F treatment in MCF/ADR cells showed partial inhibition of the N-glycan biosynthesis and increased sensitivity to chemotherapeutic drugs dramatically both in vitro and in vivo. Using an RNA interference strategy, we showed that the downregulation of MGAT5 in MCF/ADR cells could enhance the chemosensitivity to antitumor drugs both in vitro and in vivo. Conversely, a stable high expression of MGAT5 in MCF-7 cells could increase resistance to chemotherapeutic drugs both in vitro and in vivo. In conclusion, the alterations of glycogene and N-glycan in human breast cancer cells correlate with tumor sensitivity to chemotherapeutic drug and have significant implications for the development of new treatment strategies. © 2013 IUBMB Life, 65(5):409–422, 2013.
Aberrant glycosylation has been known to be associated with various human diseases, particularly with cancer (1). N-glycosylation of protein is increasingly being recognized as one of the most prominent biochemical alterations involved in tumorigenesis and metastasis (2, 3). Both changes in the core structures and increased branching of N-linked oligosaccharides are associated with malignancy and metastasis (4). Thus, alterations to glycans found in glycoproteins derived from cancer are a common feature and can be used as a convenient biomarker.
Multidrug resistance (MDR) frequently contributes to the failure of chemotherapeutic drug treatments in patients diagnosed with various cancers (5). Several mechanisms involving changes in the expression of drug transporting pumps (6), drug metabolizing enzymes (7), changes in apoptosis regulatory pathways (8), and oncogenic signaling have been described to induce MDR. Recently, a number of studies have aimed at identifying those glycans that are expressed specifically by tumor cells and correlate with chemoresistance (9). Fiala et al. (10) reported considerably reduced levels of uridine diphosphate (UDP) sugars in L1210/VCR cells compared to L1210 cells, which reflects the lower contents of glycoproteins and polysaccharides. A different study showed that the activity of β-galactoside α-2,6-sialyltransferase I in human colon cancer cells was lost when they acquired methotrexate resistance and that glycan structures on the cell membrane were altered (11). N-linked glycans on α5β1 integrin in cisplatin-resistant head and neck cancer cell showed reduced β-1,6-N-acetylglucosamine branches compared to the parent line (12). Kudo et al. (13) showed that N-glycan alterations were associated with drug resistance in human hepatocellular carcinoma. Lattová et al. (14) reported that N-glycomic changes in human breast carcinoma MCF-7 and T-lymphoblastoid cells occurred after treatment with herceptin and herceptin/Lipoplex. Our results also showed that glycomic alterations were associated with MDR in human leukemia (15). These studies have detected N-glycan alterations in resistant cancer cell line compared to its parent line, but systemic and complete information is still available on the correlation of N-glycans with drug resistance.
The objective of this study was to examine any differences of glycogenes, glycan profiling, and N-glycan composition between the MCF-7 and MCF/ADR cell lines and to investigate whether glycogenes and N-glycans participate in the regulation of cancer MDR. Meanwhile, we mainly focused on the regulation of N-glycans of cell surface and glycogene expression to further address the important roles of N-glycans in drug resistance of human breast cancer cells both in vitro and in vivo.
Materials and Methods
Human breast cancer cell line MCF-7 (purchased from KeyGEN, China) was cultured at 37 °C under a humidified 5% CO2 in roswell park memorial institute (RPMI) 1640 (Gibco, Rockville, MD) supplemented with antibiotics (1× penicillin/streptomycin 100 U/mL, Gibco, Rockville, MD) and heat-inactivated fetal bovine serum (Gibco, Rockville, MD). Adriamycin (Sigma, St. Louis, MO) was added to MCF-7 cell cultures in stepwise increasing concentrations to develop a drug-resistant cell subline (MCF/ADR). To maintain the adriamycin-resistant phenotype, MCF/ADR cells were cultured with RPMI 1640 supplemented with adriamycin at 1.0 mg/L.
Analysis of Glycogenes
To test the expression profiles of genes related to glycan synthesis, a real time RT-PCR analysis was performed. Total RNA was isolated from two cell lines using an RNeasy Mini Kit (QIAGEN, Valencia, CA), and cDNA was synthesized using QuantiTect Reverse Transcription Kit (QIAGEN, Valencia, CA) from 5 μg of total RNA according to the manufacturer's instruction. Real-time PCR amplification and analysis were performed on 7500 fast Real-time PCR System (Applied Biosystems, Foster City, CA) for 40 cycles (15 sec at 95 °C, 15 sec at 60 °C, and 30 sec at 72 °C). All reactions were performed with QuantiTect SYBR Green PCR Kit (QIAGEN, Valencia, CA) according to the manufacturer's instruction. The sequences of the upstream and downstream primers were as follows: 5′-AGCAGCTCCATGTTACGG-3′ and 5′-GACCAGATTGTCCACCTTT-3′ for MGAT5; 5′-GGTGGGCCCACACAAGCGGCG-3′ and 5′-GACTCAGGTCCTGAGGGAAA-3′ for ALG3; 5′-GTGGGCACAAAAACTACCAT-3′ and 5′-GGCTCTGGGCTCATA AACTG-3′ for ST6GAL1; 5′-AACGCCTCCTCTTCCTGTC-3′ and 5′-TGGGGTA GACAGTCCAGGTG-3′ for FUT1; 5′-CTCATCAGCACCACTCAC-3′ and 5′-TACCACTTACTGCGGACA-3′ for B3GALT1; 5′-GGCCTGACCTAGACTCACT AGTG-3′; 5′-CGCAGTGCGGTCTGCTGGCCAG-3′ for B3GNT8; 5′-GCATAACGA ACCTAACCCTCAG-3′; 5′-GCCCAATGTCCACTGTGATA-3′ for B4GALT2; 5′-G CCGCGTCATCAACGCCATCAA-3′; 5′-CAGGTAGTCGTCGGCGATCCA-3′ for MGAT3; 5′-CCATGTTCGTCATGGGTGTG-3′ and 5′-GGTGCTAAGCAGTTGG TGGTG-3′ for GAPDH. Expression levels of each glycogene were normalized using either the expression level of GAPDH mRNA and compared between MCF-7 and MCF/ADR cell lines. Real-time RT-PCR analysis was performed in triplicate.
Flow Cytometry Assay
MCF-7 and MCF/ADR cell lines were washed thrice with fluorescence-activated cell sorting (FACS) buffer (phosphate buffered saline [PBS] containing 20 g/L bovine serum) and then centrifuged at 1,000 rpm/min for 5 min in a 1.0 mL eppendorf tube for collecting cells. The cells were blocked for 30 min at 37°C in 5% powdered skim milk and then were washed between each step with FACS buffer. Cells were placed in sterile conical tubes in aliquots of 500,000 cells each and stained with one of the five FITC-lectins (GNA, SSA, AAL, BPL, RCA120) at a final concentration of 10 μg/mL for 40 min at 4°C in the dark. Residual unbound FITC-lectin was then discarded by repeat centrifugation of samples at 1000 rpm/min. After removal of supernatant, cells were resuspended in 0.2 mL PBS. Control, which was negative, cells and FITC-lectins were alone. Fluorescence and light scatters were analyzed in a BD Biosciences fluorescence-activated cell sorter (FACSCalibur) equipped with an argon laser tuned at 488 nm and a 635-nm diode, and Cell Quest software was used for acquisition. Ten thousand cells were analyzed for each sample. Three independent assays were carried out using both MCF-7 and MCF/ADR lines.
Membrane Protein Extract
A total of 1 × 107 cells were rinsed with PBS and lysed on a plate with lysis and separation buffer containing a protease inhibitor cocktail. Cell membrane proteins were extracted from the cell suspension using a CelLytic MEM Protein Extraction kit (Sigma, St. Louis, MO). The membrane protein concentration was measured with a Micro BCA Protein Assay kit (Pierce, Rockford, IL) and used for further experiments as described below.
Release of N-Glycans from Cell Membrane Proteins
Dried three 100 μg aliquots of cell membrane proteins were first digested with trypsin (10 μg) and chymotrypsin (10 μg) dissolved in 25 mM ammonium bicarbonate (25 μL) at 37°C for 18 h. The digest was left in a water bath (85°C, 5 min) and after cooling N-linked oligosaccharides were released from peptides by treatment with PNGaseF enzyme (2 μL; 6 U) at 37°C (18 h) followed by Pronase digestion (10 μg) at 37°C (8 h). During the incubation time, the reaction sample was mixed occasionally. The released N-glycans were purified using an Oasis HLB cartridge (60 mg/3 mL; Waters) and then were lyophilized.
The mass spectra were carried out in reflectron positive ion mode with matrix-assisted laser desorption/Ionization time of flight mass spectrometry (MALDI-TOF MS) (Bruker Corp., Billerica, MA). To increase sensitivity and provide more informative fragmentation, the released glycans were permethylated (16, 17) and further characterized by MALDI-TOF MS. For the type of MALDI analysis of the permethylated glycans, 2,5-dihydroxy-benzoic acid (2,5-DHB) was used as the matrix. Values are the mean ± S.D of three permethylated samples from N-glycan samples. All MS spectra were obtained from Na+ adduct ions.
RNA Interference Assay
MCF/ADR cells were incubated in appropriate antibiotic-free medium with 10% fetal bovine serum (Gibco, Rockville, MD), transferred to a six-well tissue culture and incubated at 37°C, in a CO2 incubator to obtain 60–80% confluence. The cell cultures were transfected with MGAT5-specific siRNA Transfection Reagent Complex (Santa Cruz Biotech, sc-40642), which was prepared according to the protocol. Scrambled siRNA was used as the negative control. Transfer cells were cultured and incubated at 37°C for 6 h, followed by incubation with complete medium for additional 24 h. Then cells were harvested and experimented as described for Western blot analysis, in vitro extracellular matrix invasion (ECM) assay, and in vitro drug sensitivity assay. The cell transfection efficiency was 75%, and the survival rate was 89%.
To generate MGAT5-transfected human breast cancer cell line MCF-7, the coding region of wild-type MGAT5 was subcloned into pcDNA3.1 expression vector (Invitrogen, Carlsbad, USA) to generate pcDNA3.1/MGAT5. The plasmid was mixed with Lipofectamine 2000 (Invitrogen, Carlsbad, USA) according to manufacturer's instructions and added to MCF-7 cells. The selection for transfected cells was carried out in a medium containing 800 µg/mL G418 (Calbiochem, Darmstadt, Germany) for at least 4 weeks before the experiments. The stable transfectants were selected and used as a population assigned MCF-7/MGAT5. The empty vector was used as a transfection control and resulting transfeatants were assigned MCF-7/Mock. The cell transfection efficiency was 78%, and the survival rate was 90%.
Western Blot Analysis
Cell membrane proteins or whole cell proteins were electrophoresed under reducing conditions in 10% polyacrylamide gels. The separated proteins were transferred to a polyvinylidene difluoride membrane (Pall Corporation, Port Washington, NY). After blocking with 5% skimmed milk in PBS containing 0.1% Tween 20, the membrane was incubated with anti-CD147, anti-P-gp (1/200 diluted; Santa Cruz Biotech), and anti-MGAT5 (1/500 diluted, Abcam), and then with peroxidase-conjugated anti-rabbit IgG (1/1,000 diluted; Santa Cruz Biotech). A Na+/K+-ATPase antibody (1/200 diluted; Santa Cruz Biotech) or GAPDH (1/200 diluted; Santa Cruz Biotech) was used as a control. Crossreacting bands were detected with Western Lightning Chemiluminescence Plus (Perkin-Elmer LAS).
To inhibit N-linked glycosylation of newly synthesized proteins, MCF/ADR cells were washed once with PBS and cultivated for 12 h in fresh culture media (90% RPMI 1640 supplemented with antibiotics and 10% fetal bovine serum) in the absence or presence of tunicamycin (TM) in a dose-dependent manner (0, 1, 5, or 10 µg/mL). The cells were washed with PBS again and then were determined by Western blot analysis, cytotoxicity assay, and antitumor activity assay in vivo. The cell survival rates were 86%, 89%, 84%, and 82% by trypan blue dye exclusion assay, respectively.
PNGase F Treatment
To remove N-glycans, membrane proteins (200 μg) from MCF/ADR cells were dried in vacuum and redissolved in denaturing buffer (1 mmol/L Tris-HCl, pH 8.6, 1% sodium dodecyl sulfate, and 1% β-mercaptoethanol) and then were heated at 100°C for 3 min. Subsequently, corresponding reaction buffer, 10% Nonidet NP-40, and 25 units of PNGase F (Sigma Aldrich) were added. The probes were incubated at 37°C in a time-dependent manner (0, 8, 16, 24 h). Afterward, the reaction was stopped with 0.5% trifluoroacetic acid (TFA) solution, and the protein was separated in a gel to Western-Blot Analysis.
Additionally, for deglycosylation of membrane protein, intact MCF/ADR cells were incubated with 25 units of PNGase for 0, 8, 16, 24 hours, washed, and subsequently treated as described for drug sensitivity and in vivo antitumor activity assays (18). The cell survival rate was 87%, 85%, 85%, and 81% by trypan blue dye exclusion assay, respectively.
In Vitro Drug Sensitivity Assay
Drug resistance was measured using an MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. Cells (1×104) were plated in 96-well plates (Costar, Charlotte, NC) and incubated with different anticancer drugs paclitaxel (Sigma, St. Louis, MO), vincristine (Sigma, St. Louis, MO), and adriamycin (Sigma, St. Louis, MO) for 48 h. Then cells were treated with 100 µl MTT (5 mg/mL, Sigma). After 4 h incubation at 37°C in 5% CO2, 100 µl dimethyl sulfoxide (DMSO) (Gibco, Rockville, MD) was pipetted to solubilize the formazan product for 30 min at room temperature. Spectrometric absorbance at 490 nm was measured with a microplate reader. Each group contained three wells and was repeated three times. The concentrations required for 50% growth inhibition (IC50 values) were determined by the drug dose that caused 50% cell viability.
In Vivo Antitumor Activity
To investigate whether N-glycans or glycogens are related to tumor cell chemosensitivity in vivo, the antitumor activity of adriamycin was examined in nude mice bearing tumor cell xenografts. Five-week-old nude mice were obtained from Animal Facility of Dalian Medical University and were provided with sterilized food and water. Approximately, 1×107 MCF-7, 1×107 MCF/ADR, or 1×107 MCF/ADR (TM or PNGase F treatment) viable cells were injected subcutaneously into the right flank of each nude mouse. When mice were beared palpable tumors (about 1 week after tumor cell inoculation), MCF-7, MCF/ADR, or MCF-7/ADR (TM or PNGase F treatment) tumor-beared mice were randomly divided into control and treatment groups (n = 6 animals per group). The treatment groups received 4 mg/kg adriamycin i.p. three times a week for 3 weeks, and the control groups received physiological saline alone. Mice were sacrificed and their tumors were isolated, weighed, and photographed. The tumor inhibition rate (IR) was calculated according to the follow equation: IR (%) = ((Wc × Wt)/Wc) × 100%, wherein Wc and Wt represent the mean tumor weight of the control group and treatment group, respectively.
Furthermore, MCF/ADR-control siRNA, MCF/ADR-MGAT5 siRNA, MCF-7/mock, and MCF-7/MGAT5 tumor-bearing mice models were finished according to above-mentioned method.
SPSS13.0 software was used. Each assay was performed at least three times. The data were presented as mean ± SD. Comparisons were done by independent-samples t test while multiple comparisons were carried out using one-way ANOVA to determine the significance of differences. Probability values less than 0.05 was considered to be statistically significant.
Differential Expression of Glycogenes in MCF-7 and MCF/ADR Cell Lines
The biosynthetic pathway of N-glycans highlights importance of glycosyltransferases encoded by glycogenes. To analyze the expression profile of glycogenes in drug sensitive and MDR cells, a real-time RT-PCR analysis was performed. Eight genes (out of 54) were differentially expressed between the two cell lines. MGAT5, ALG3, ST6GAL1, FUT1, B3GALT1, and B3GNT8 glycogenes were expressed at an elevated level (i.e., >3-fold higher) in MCF/ADR compared with MCF-7 cells. Conversely, B4GALT2 and MGAT3 were expressed at a higher level in MCF-7 compared with MCF/ADR cells (i.e., >3-fold higher, Table 1). These data indicated that differential glycogene expressions might be associated with MDR of human breast cancer cells.
Differential FITC-Lectin Binding Profiles of MCF-7 and MCF/ADR Cell Lines Using Flow Cytometry
Fluorescence of MCF-7 and MCF/ADR cells, exposed to a diverse panel of FITC-lectins, suggests that both cells have some lectin-binding natures. Figures 1A and 1B show that the obvious differences in fluorescence intensity for N-glycosylation were evident by comparison between the MCF-7 and the MCF/ADR cell lines as summarized below: 1) increased level of branching of N-glycans in MCF/ADR cells estimated by fluorescence intensity of L-PHA (phytohemagglutinin) (tri- and tetra-antennary complex oligosaccharides), GNA (nonsubstituted α1-6Man), SNA (Siaα2-6Gal/GalNAc), DSA ((GlcNAc)n, polyLacNAc and LacNAc (NA3, NA4)), and BPL (Galβ1-3GalNAc and NA3, NA4), 2) higher terminal fucosylation in MCF/ADR cells as revealed by fluorescence intensity on AAL, 3) incremental increase in LacNAc structures evident from fluorescence intensity of RCA120, and higher signals of E-PHA (NA2 and bisecting GlcNAc) in MCF-7 cells. These results correlated well with the real-time PCR analysis of glycogene expression. High expressions of glycogene were corresponding with high fluorescence intensity of lectins in both cell lines (Fig. 1C).
MALDI-MS Analysis of N-Glycans from MCF-7 and MCF/ADR Cells
MALDI-TOF MS analysis was used to evaluate the N-glycan composition of MCF-7 and MCF/ADR cell lines. Figure 2 shows the MS spectra of N-glycans released from cell membranes and the observed MS signals of the N-glycans (peaks 1–32 in Fig. 3A) and the assigned N-glycan signals as were summarized in Table 2. The observed signal intensities in the mass spectra are presented as a histogram (Fig. 3B), with the estimated monosaccharide compositions. High mannoses analyzed in both cell lines were detected at peak 2, 4, 8, 13, and 17 (Table 2). Several major N-glycans differences of cell membrane derived from MCF-7 and MCF/ADR were detected. Nine peaks, peak 3, 5, 7, 10, 12, 15, 16, 20, 22 were detected exclusively in the drug-resistant MCF/ADR cell line. Peak 11 was detected exclusively in MCF-7 cell line. Although there were many differences regarding the intensities of all peaks in the spectra recorded from pools of MCF-7 and MCF/ADR samples, generally the main analyzed N-glycans in both cell lines were detected at peak 2, 8, 17, and 25. Among those oligosaccharides, peak 2, 8, and 17 increased in drug-resistant MCF/ADR cell line (High Mannoses), and peak 25 increased in MCF-7 cell line. These data indicated that differential N-glycan composition might be associated with tumor MDR.
N-Glycans Alter the Chemosensitivity of MCF/ADR Cells In Vitro
To exam directly whether the N-glycan of MCF/ADR cells influenced their chemosensitivity, the regulation of N-glycosylation was investigated. P-gp and CD147 are N-glycoproteins, which have been analyzed before in total cell membrane extracts were analyzed by Western blot. As illustrated in Fig. 3A, glycoprotein levels of P-gp and CD147 in MCF/ADR cells were much higher than those in MCF-7 cell line.
MCF/ADR cells were treated with TM, an inhibitor of endogenous protein N-glycosylation. In addition, aliquots of membrane proteins extracted from MCF/ADR cells were exposed to exogenous PNGase F to achieve complete deglycosylation. Figure 3A showed that P-gp appeared as a double band—the upper band at 170 kDa and the lower band at 140 kDa. The P-gp (170 kDa) completely disappeared in the TM-treated membranes. Interestingly, a 110 kDa P-gp band appeared after PNGase F treatment. CD147 existed in both HG-CD147 (40–60 kDa) and LG-CD147 (33 kDa). The CD147 (33 kDa) completely disappeared, and the level of CD147 (40–60 kDa) was greatly diminished after TM treatment. However, the CD147 (40–60 kDa) completely disappeared and the 27 kDa band appeared after PNGase F treatment. The 27 kDa band that appeared was consistent with the size of the core protein. These results suggested that the N-glycosylation process in MCF/ADR cells was highly sensitive to inhibition by TM and PNGase F.
To further investigate whether the deglycosylation modulated the sensitivity of the cells to the chemotherapeutic drugs, MCF/ADR cells were treated with TM or PNGase F and thereafter their chemosensitivity were analyzed. As depicted in Fig. 3B, in addition to adriamycin, MCF/ADR cells were also resistant to other chemotherapeutic drugs including vincristine and paclitaxel. The IC50 values for the drugs were greater in MCF/ADR group than in MCF-7 group (Fig. 3B), indicating that MCF/ADR cells awarded MDR characteristics. Under TM and PNGase F treatment, the MCF/ADR cells showed increased sensitivity to these chemotherapeutic drugs compared with the MCF/ADR cells in the absence of TM and PNGase F (Figs. 3B and 3C). Thus, downregulation of N-glycan in MCF/ADR cells resulted in increased chemosensitivity to antitumor drugs.
N-Glycans Effect on Antitumor Activity of Chemotherapeutics on MCF/ADR Cells In Vivo
The in vivo antitumor activity of adriamycin was assessed by intraperitoneal administration using nude mice bearing MCF-7, MCF/ADR, or MCF/ADR (TM or PNGase F treated) tumors as the model animals. The treatments included three injections per week for 3 weeks. Figures 4A and 4B show that a significant reduction of mean tumor weight of deglycosylation MDR tumors was observed, as compared with MCF/ADR groups in the absence of TM and PNGase F. The effect of concomitant application of adriamycin to TM and PNGase F pre-treatment was depended by increasing dose of the TM and PNGase F preapplication. In the TM pretreatment MCF/ADR xenograft model, we observed a significant decrease in tumor growth when compared with the control groups which received physiological saline instead of adriamycin. The IR of adriamycin were 9.94%, 14.45%, 22.29%, and 37.30% in this TM treated group (Fig. 4A). Similar results were obtained in the PNGase F pretreatment MCF/ADR xenograft model. The IR of adriamycin were 9.94%, 16.88%, 27.46%, 40.37% at 3 weeks after implantation, respectively, with the same conditions as Fig. 4A (Fig. 4B). In summary, TM and PNGase F pretreated MCF/ADR cells showed a dose per se reduction of tumor growth compared to the non-pretreated control and this effect was significantly enhanced by concomitant application of adriamycin particularly at high doses of TM and PNGase F pretreatment. Thus, the antitumor activity of chemotherapeutic drugs can be enhanced by deglycosylation in MCF/ADR cells.
Silence of MGAT5 Effects on the Chemosensitivity of MCF/ADR Cells Both In Vitro and In Vivo
Due to the significant increase of MGAT5 mRNA expression in MCF/ADR cells (Table 1), we silenced, by siRNA, MGAT5 to elucidate the direct implication MGAT5 in the chemosensitivity of MCF/ADR cells. As shown in Fig. 5A, MGAT5 expression at protein level was downregulated in MGAT5 transfectants compared with MCF/ADR-control siRNA transfectants. To further evaluate the effect of MGAT5 silencing on cells chemosensitivity, each cell group was treated with adriamycin, paclitaxel, and vincristine, respectively. The MCF/ADR-control siRNA cells had a similar resistance to MCF/ADR cells, but the MCF /ADR-MGAT5 siRNA cells showed increased sensitivity to these chemotherapeutic drugs, as compared with MCF/ADR-control siRNA groups (Fig. 5B). The IC50 values for the drugs were significantly less in the MCF/ADR-MGAT5 siRNA group than those in the MCF/ADR-control siRNA groups. Thus, the downregulation of MGAT5 in MCF/ADR cells resulted in increased chemosensitivity to antitumor drugs in vitro.
Figure 5C shows that a significant reduction of mean tumor weight of MCF/ADR-MGAT5 siRNA tumor was observed, as compared with MCF/ADR-control siRNA group, and the effect of concomitant application of adriamycin. The IR of adriamycin were 9.20%, 10.59%, and 27.87% at 3 weeks after implantation, respectively. Thus, the antitumor activity of chemotherapeutic drugs can be enhanced by downregulation of MGAT5 in MCF/ADR cells.
L-PHA lectin, which specifically recognizes β1,6 branched structures (product of MGAT5), was used to analyze the alterations in the N-glycosylation pattern of glycoproteins. To evaluate whether MGAT5 silencing could modify the N-glycosylation profile in terms of β1,6 branched structures using a flow cytometry, each cell group was binded with L-PHA lectin. Figure 5D shows that the MGAT5 knockdown resulted in a decrease of fluorescence intensity (β1,6 branched structures) compared with the control cells. These results clearly showed that MGAT5 was responsible for the overcoming tumor cells MDR resistance via regulating N-glycosylation profile in terms of β1,6 branched structures in human breast cancer cells.
Overexpression of MGAT5 Influences the Chemosensitivity of MCF-7 Cells Both In Vitro and In Vivo
To explore the effect of MGAT5 on chemosensitivity, a MCF-7 cell line stably expressing MGAT5 was established. It was found that the level of MGAT5 protein was notably increased in MCF-7 transfectants (Fig. 6A). To further evaluate the effect of MGAT5 overexpression on cells chemosensitivity, each cell group was treated with adriamycin, paclitaxel, and vincristine, respectively (Fig. 6B). The IC50 values for the drugs were significantly more in the MCF-7/MGAT5 groups than those in the MCF-7/mock groups. Thus, the overexpression of MGAT5 in MCF-7 cells resulted in decreased chemosensitivity to antitumor drugs in vitro.
Figure 6C shows that a significant increase of mean tumor weight of MCF-7/MGAT5 tumor was observed, as compared with MCF-7/mock group, and the effect of concomitant application of adriamycin. The IR of adriamycin were 37.51%, 38.33%, and 29.46% at 3 weeks after implantation, respectively. Thus, the antitumor activity of chemotherapeutic drugs can be inhibited by overexpression of MGAT5 in MCF/ADR cells.
Figure 6D shows that the MGAT5 overexpression resulted in an increase of fluorescence intensity (β1, 6 branched structures) compared with the MCF-7/mock cells. These results clearly showed that MGAT5 was responsible for the overcoming tumor cells MDR resistance via regulating N-glycosylation profile in terms of β1,6 branched structures in human breast cancer cells.
In this study, we investigated the possible correlation of glycosylation and MDR resistance in parental and drug-resistant human breast cancer cell lines using real time RT-PCR, FITC-lectin binding, and mass spectrometry (MS). Additionally, we demonstrated that the glycogene and N-glycan affected drug resistance of human breast cancer cells both in vitro and in vivo.
Glycogenes, which encode proteins, are involved in glycan synthesis and modification. In this study, we also found that the expression profiles of glycogenes were remodeled between MCF-7 and MCF/ADR using a real-time PCR analysis. Glycogene expressions were highly regulated, with eight (out of 54) glycogenes (at least 3-fold, Table 1) significantly differentially expressed among the two cell lines. The major altering expressions of glycogenes in the two cell lines may be more important as indicators and functional contributors of tumor MDR. For example, ALG3, which is significantly expressed at an elevated level (3.01-fold higher) in MCF/ADR compared with that in MCF-7 cells, encodes Dol-P-Man: Dol-PP-Man5GlcNAc2 α-1,2-mannosyltransferase that converts Man5GlcNAc2-Dol-PP to Man6GlcNAc2-Dol-PP. These data were consistent with the FITC-lectins binding and MS analysis.
The flow cytometry analysis is an effective approach to measure the linkage of FITC-lectin to cell surface carbohydrate not only qualitatively but also quantitatively. Current report (19) revealed lectin binding profiles of SSEA-4 enriched, pluripotent human embryonic stem cell surfaces using flow cytometry assay. Population fluorescence related to FITC-lectin binding to acanthamoebal surface moieties was also confirmed by flow cytometry (20). This study clearly showed the lectin binding profiles of cell surfaces in MCF-7 and MCF/ADR cell lines by flow cytometry consisting of eight FITC-lectins with distinctive binding specificities. The obtained datasets could be statistically compared to identify lectins that show significant differences between the two cell lines (Figs. 2A and 2B). L-PHA, SNA, BPL, AAL, and DSA signals were upregulated in MCF/ADR cells. GNA signal (preferential binding to high mannose glycans) also increased in MCF/ADR cells and a series of high mannose glycans were increased in the MS profiles of the MCF/ADR cells. Similarly, terminal galactose and N-acetylgalactosamine (RCA120), E-PHA were only highly expressed on MCF-7 cell surfaces. Furthermore, we also found that high expression of glycogene was responsible for high fluorescence intensity of FITC-lectin in the two cell lines (Fig. 2C). The result was consistent with the glycogene expression analysis.
MS technology as a novel methodology provides high sensitivity and more rapid glycan analysis (21–23). Hu et al. (24) have investigated novel protein changes involved in drug resistance between leukemia cell line HL-60 and adriamycin-resistant HL-60 (HL-60/ADR) by MALDI-TOF/MS. Moreover, glycoprotein of human ovarian cancer cell line A2780 and its paclitaxel resistant counterpart A2780TC1 was identified by MS (MALDI TOF or LC MS/MS) (25). In this study, we extracted all glycoproteins from the cell membrane, released N-glycans from the total glycoprotein pool by PNGase F digestion, and then performed an N-glycan profiling using MALDI-TOF MS, which allows the characterization of different N-glycan classes. In addition, the abundance ratios of oligosaccharides can be obtained from a comparison of relative ion intensities corresponding to individual glycans. We compared the total N-glycans from MCF-7 and MCF/ADR cell lines and found dramatic differences in N-glycan profiles between these two groups (Fig. 2, Table 2). A major population of N-glycans detected in MCF-7 and MCF/ADR cells corresponded to high-mannose structures, and some high mannoses in particular peak 2, 8, 17 were dramatically increased in drug-resistant MCF/ADR cell line. Moreover, MCF/ADR cells showed a significant increase of fucosylated and sialylated oligosaccharides (peak 3, 5, 10, 12, 16, 20, 22). Therefore, monitoring of the N-glycan profile would be an important step in the prevention of tumor MDR and would increase our understanding of resistance mechanisms.
The combinations of N-glycan and protein, and N-glycan bound protein may be crucial for their functions (26, 27). It is more likely that altered N-glycans associated with several different proteins affect the drug-resistance mechanisms. P-gp is a heavy N-glycosylated transmembrane transporter that appears as a double band (28, 29). Glycosylation of P-gp is important for appropriate protein folding and plasma membrane export. The core-glycosylated P-gp in contrast to the fully glycosylated 170-kDa species of P-gp is inactive and accumulates in the endoplasmic reticulum (30, 31). CD147 is a highly N-glycosylated immunoglobulin superfamily transmembrane protein that is composed of two extracellular Ig domains, which contribute to a highly N-glycosylated HG-CD147 (∼40–60 kDa) and a low glycosylated form, LG-CD147 (∼33 kDa) (32). CD147 is also involved in resistance of tumor cells to some chemotherapeutic agents. Overexpression of CD147 is reported in several MDR cancer cell lines compared to the parental controls (33). Downregulation of CD147 gene expression via RNA interference (RNAi) improved chemosensitivity to paclitaxel in the human ovarian cancer cell line (34). Our previous results showed that the expression of CD147 mediates tumor cells invasion and MDR in hepatocellular carcinoma (35). Here, the results showed that altered N-glycosylations of CD147 and P-gp were found in MCF/ADR cells and further suggested a link between defective N-glycosylation of MCF/ADR cells and drug resistance (Fig. 3). Additionally, findings relative to the role of N-glycan affecting antitumor activity of chemotherapeutic drugs in vivo were observed (Fig. 4). Deglycosylation treatment of MDR tumors prior to adriamycin treatment resulted in a reduced mean tumor weight. Thus, significantly changed N-glycans might be associated with the development of drug resistance in MDR cancer cells.
The glycogenes that were common between the model systems may be of particular interest for further study. MGAT5, which encodes N-acetylglucosaminyltransferase V, has strong activity toward β1, 6 branched N-glycans. It has been reported to be involved in cancer metastasis (36, 37), but a comprehensive understanding of how the glycogene MGAT5 correlates with the MDR of human breast cancer cells is not currently available. Our previous results showed that the silencing of MGAT5 in human hepatocarcinoma MHCC97-H cells inhibited invasion and increased sensitivity to 5-fluorouracil in vitro (38). Here, we have clearly demonstrated that the expression of glycogene MGAT5 was increased in the MDR cells MCF/ADR as compared to the parental cells MCF-7 (Table 1). This result suggested that over-expression of MGAT5 glycogene was associated with MDR in MCF/ADR cells. In addition, we further detected that the silencing of MGAT5 in MCF/ADR cells resulted in increased chemosensitivity to antitumor drugs both in vitro and in vivo (Figs. 5B and 5C). MGAT5 product also decreased remarkably in MCF/ADR-MGAT5 siRNA cells labeled with L-PHA lectin (which looks like a tumor marker, Fig. 5D). Conversely, a stable high expression of MGAT5 in MCF-7 cells could increase resistance to chemotherapeutic drugs both in vitro and in vivo (Fig. 6). These results clearly show that the change in glycogene expression level has impact in the remodeling of cell surface glycoproteins oligosaccharides, which may consequently affect the biological functions of tumor cells such as MDR resistance.
Taken together, we found that the glycogene expressions and the structure of N-glycan on cell-surface glycoproteins are clearly associated with drug resistance of breast cancer cells both in vitro and in vivo. As no complete reversal of drug resistance was obtained by targeting a single factor (N-glycan or MCAT5), this study corroborated that drug resistance was multifactorial event, and the multiple glycomic alterations could be associated with the phenotype. Therefore, the molecular bases of drug resistance remain to be further investigated.
This work was supported by grants from National Natural Science Foundation of China (81071415), from Natural Science Foundation of Liaoning Province (20102052), and supported by Program for Liaoning Excellent Talents in University (LR2011025).
- 192005) Lectin binding profiles of SSEA-4 enriched, pluripotent human embryonic stem cell surfaces. BMC Dev. Biol. 21, 5–15., , , , , , and (