Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Feb 28.
Published in final edited form as: FEBS Lett. 2019 Jul 21;593(19):2751–2761. doi: 10.1002/1873-3468.13532

MUC1 Negatively Regulates GalNAc Transferase 5 Expression in Pancreatic Cancer

Thomas Caffrey 1, Satish Sagar 1,#, Divya Thomas 1,#, Michelle E Lewallen 2, Michael A Hollingsworth 1, Prakash Radhakrishnan 1,*
PMCID: PMC7048170  NIHMSID: NIHMS1558618  PMID: 31283009

Abstract

Aberrant expression of MUC1 glycoprotein has been associated with pancreatic cancer (PC) progression and metastasis and mediates oncogenic transcriptional regulation of target genes during the disease progression. Here, we demonstrate that MUC1 downregulates the tumor suppressor function associated GalNAc-T5 expression in PC. ChIP-on-chip analysis revealed that MUC1 cytoplasmic tail binds to the regulatory elements of the GALNT5 gene. Additionally, MUC1 increases binding of p53 and c-Jun and decreases the binding of Sp1 in the proximal promoter and exonic regions of GALNT5. We also observed expression of GalNAc-T5 is inversionally proportional to MUC1 expression in human PC. These results demonstrate that MUC1 downregulates the expression of GalNAc-T5 in PC by modifying the promoter occupancy of transcriptional factors through its cytoplasmic domain.

Keywords: MUC1-CT, GalNAc-T5, pancreatic cancer, glycosyltransferase

Introduction

The type 1 transmembrane glycoprotein Mucin-1 (MUC1) is overexpressed and aberrantly glycosylated in several epithelial cancers including more than 90% in human pancreatic cancer (PC) and is considered as a potential prognostic biomarker for PC1,2. Overexpression of MUC1 is detected in more than 60% of Pancreatic Intraepithelial Neoplasia (PanIN) lesions and is positively correlates with increased metastasis and poor patient survival3,4. MUC1 increases PC cell invasiveness and metastasis through the induction of epithelial-to-mesenchymal transition (EMT)5. Also, it enhances the progression of PanIN lesions into pancreatic ductal adenocarcinoma (PDAC) through increasing cell proliferation in transgenic human MUC1-expressing mouse pancreas6. Human MUC1 contains a large extracellular domain comprised of 20 amino acid variable number of tandem repeats (VNTR), the number of repeats varying from 20 to 120 in different individual, and rich in serine and threonine residues. During pot-translational modifications, the full length MUC1 undergoes proteolytic cleavage to produce a larger subunit containing most of the extracellular tandem repeat domain and a smaller subunit containing short extra cellular domain, a trans-membrane domain and a cytoplasmic tail2,5,7. The extra cellular fragment of MUC1 is believed to contribute to the molecular structure of the glycoprotein and configures the adhesive properties of the cell. However, little is known about the MUC1 cytoplasmic tail (MUC1-CT); evidences suggest its role in signal transduction though. MUC1-CT is comprised of 72 amino acids and serves as a target for phosphorylation, which enables its binding to various kinases and adaptor proteins7,8 and subsequent translocation to the nucleus where it participates in regulation of gene transcription9.

Although MUC1-CT possesses neither a known nuclear localization sequence nor DNA binding motif, it indirectly modulates gene expression through direct interaction with transcriptional factors and signaling mediators. For example, binding of MUC1-CT along with Kruppel-like factor 4 (KLF4) on p53 promoter elements down regulates p53 gene expression10. Previous studies from our laboratory have shown that the interactions of MUC1-CT and mutant p53 complex suppress the expression of matrix metalloprotease 1 (MMP1) through its occupancy at proximal promoter regions of MMP111. MUC1-CT also functions in a complex with mutant p53 and β-catenin to up regulate the expression of the tumor enhancer gene connective tissue growth factor (CTGF) through occupancy at various regulatory regions of the CTGF promoter12.

The mucin type O-glycosylation is initiated by a group of polypeptide N-acetylgalactosaminyltransferases (GalNAc-Ts), and ~ 20 different isozymes were identified and cloned to date10. The Golgi resident GalNAc transferases catalyze the addition of the GalNAc residues to serine or threonine amino acids in the peptide backbone to form GalNAcα-ser/thr (Tn antigen), which is a critical aspect of O-glycosylation13. Thus, through regulating the initial step of O-glycan biosynthesis and also determining specific sites of O-glycosylation on proteins, GalNAc-transferases are central for understanding normal and cancer-associated O-glycosylation14. Altered expressions of GalNAc transferases have been associated with disease proegression and an increased expression of GalNAc transferases −3 and −6 were observed in PC15,16. Ectopic expression of GalNAc-T3 enhances PC cell growth and cell survival15. Also, overexpression of GalNAc-T6 and T3 are associated with differentiation and poor prognosis of pancreatic and gastric tumors, respectively16,17. However, the available evidences on the glycosylation machinery and GalNAc transferases are limited to structure of glycoproteins and not to site of protein attachment. Hence current understanding of the consequences of altered expressions of GalNAc-transferases in cancer are highly limited and needs better evaluation. Interestingly, GalNAc transferases are also associated with tumor suppressor functions. Wood et al., have found two inactivating somatic mutations in the GALNT5 gene (L692F and E507D) in breast cancer cells18. Studies have shown that GalNAc-T5 physically interacts with the tumor suppressor exostoses 2 (EXT2) protein19. Mutations in the EXT gene family (EXT1 or EXT2) cause multiple cartilaginous tumors in patients by altering the heparin sulfate biosynthesis20. However, to date none of the studies has assessed the expression of GalNAc-T5 in PC, and there is no documented example available for the glycoprotein MUC1 mediated regulation of GalNAc-T5 expression in any cancers.

Here, we illustrate for the first time that MUC1-CT differentially regulate the expression of polypeptide GalNAc transferases in PC. The results indicated that MUC1-CT, in complex with c-Jun and p53 transcriptional factors at the GALNT5 promoter and exon regions suppresses the expression of GalNAc-T5. Further, it was observed that the expression of GalNAc-T5 is inversely related to level of MUC1 in human pancreatic tumor tissues.

Materials and Methods

Cell lines and culture conditions

Moderate to well differentiated pancreatic cancer cell lines FG, a fast growing clone of Colo-357 PC cells21 and S2–013, an in vitro subline clone of SUIT-2 PC cells derived from liver metastasis22 were used for the study. The cells were cultured in Dulbecco’s Modified Eagles Medium (DMEM) (Corning, USA) supplemented with 10% fetal bovine serum (Valley Biomedical Inc., USA), 100 μg/ml streptomycin and 100 units/ml penicillin (Corning, USA) at 37°C in a 5% CO2 incubator.

Overexpression and knockdown of MUC1

S2–013 cells stably expressing MUC1 (MUC1-F) has been produced using pHBeta Apr 1 neo plasmid as described previously23. S2–013 cells expressing MUC1 cytoplasmic tail (MUC1-CT) was produced by deleting MUC1 tandem repeat domain using restriction enzyme sites Bsm1 and EcoN1 as described previously (Burdict). Briefly, tandem repeat deleted MUC1-CT cDNA was subcloned into the expression vector pHBeta Apr 1 neo plasmid at the BamH1 site. S2–013 cells were then transfected with plasmid DNA using lipofectine. Trasfectants were selected in growth medium containing G418 (800μg/ml). Clones were screened for the expression of MUC1-CT. Stable knockdown of MUC1 in FG cells (FG MUC1-KD) with pSuper-gfp-neo plasmid (OligoEngine, Seattle, WA, USA) was carried out as described previously24. Briefly, after transfection of vectors containing MUC1shRNA or respective control vectors, the medium was replaced with new medium containing 600 μg/ml of G418 for selection of transfectants. FG cell clones in which MUC1 was stably knocked down were maintained at 200 μg/ml of G418-containing medium.

Real-time PCR analysis of glycosyltransferases expression

Total RNA was isolated from pancreatic cancer cell lines using TRI reagent (Molecular Research Center, Cincinnati, OH, USA). Reverse transcription and cDNA synthesis were carried out by using verso cDNA synthesis kit (Thermo Fisher Scientific, Waltham, MA, USA) and real-time PCR was carried out using SYBR Green master mix (Applied Biosystem, Warrington, UK) according to the manufacturer’s instructions and reactions were carried out in a Bio-Rad CFX96 Real-Time system and C1000 Thermal cycler. All experiments were conducted in triplicate. Threshold Ct values were calculated by ΔΔCt method as described previously25. The primers used for real-time PCR are listed in Supplementary Table 2.

Western blotting of GalNAc-T5

NP-40 cell lysis buffer containing protease inhibitors (Roche Diagnostics GmbH, Mannheim, Germany) was used to isolate proteins from pancreatic cancer cells. Equal concentrations of protein lysates (50 μg) were resolved using 4–20% denaturing polyacrylamide gels (Bio-Rad, Hercules, CA, USA), transferred to polyvinylidene fluoride (PVDF) membrane (Millipore EMD, Billerica, MA, USA) and blocked with 5% powdered milk in Tris-buffered saline (pH 7.5) containing 0.1% Tween 20 (TBS-T). The following primary antibodies were used to probe the membrane for the protein expressions: rabbit anti-GalNAc-T5 (Abgent, San Diego, CA, USA, 1:1000), hamster anti-MUC1-CT (CT2, Abcam, USA, 1:1000), and mouse anti-β-actin (Sigma-Aldrich, St. Louis, MO, USA, 1:5000). After incubating with respective secondary antibodies, protein-antibody complex was detected using enhanced chemiluminescence (Bio-Rad, USA).

Immunoprecipitation analysis

S2–013 MUC1-CT cells were lysed in non-denaturing RIPA buffer (Thermo scientific, KS, USA) containing protease inhibitor (Roche Diagnostics GmbH, Mannheim, Germany). 500μg of proteins were incubated with either Hamster anti-MUC1 (5μg/ml) or Armenian Hamster IgG (5μg/ml) overnight at 4°C. Lysates were then incubated with protein G sepharose beeds (GenScript, Piscataway, NJ, USA) for 2 h at room temperature. Beeds were collected, washed three times with RIPA buffer without protease inhibitor, and boiled in SDS loading buffer (1X). Samples were subjected to 4–20% gradient (Bio-rad, USA) SDS-PAGE gel electrophoresis and transferred to PVDF membrane. The membranes were blocked with 5% skimmed milk and incubated with primary antibodies rabbit anti-p53 (EMD Millipore Corporation, Billerica, MA, USA, 1:1000) and rabbit anti-c-Jun (Calbiochem, CA, USA, 1:1000) overnight. After incubating with respective secondary antibodies, protein-antibody complex was detected using enhanced chemiluminescence (Bio-Rad, USA).

Immunohistochemical and Immunofluorescence analysis

De-identified human normal (n=3) and pancreatic tumor tissues (n=25) were obtained from patients who underwent Rapid Autopsy Pancreas Programme with proper consent at UNMC. Standard immunohistochemical procedure was used for staining of normal and primary pancreatic tumor tissues with mouse anti-MUC1 (AR 20.5 specific for extracellular domain, Quest Pharma Tech, Inc., Edmonton, Alberta, CA, 1:500), anti-MUC1-CT (CT2 specific for cytoplasmic tail, Abcam USA 1:500) and rabbit anti-GalNAc-T5 (Abgent, San Diego, CA, USA, 1:100) antibodies. Universal anti-rabbit and mouse HRP conjugated immunohistochemical kit (Vector Laboratories, Burlingame, CA, USA) were used for tissue immunohistochemical analysis. The histological scoring was performed based on stain proportion (0–100%) and intensity (0-negligible, 1-low, 2-moderate, 43high). The histoscore was generated by multiplying the stain proportion score (1= <5%, 2=5–25%, 3=26–50%, 4=51–75%, 5=>75%) with the intensity score (0 – 3) to obtain values between 0–15. For immunofluorescence analysis, S2–013 Neo and S2–013 MUC1 cells were fixed and probed with anti-MUC1-CT mAb CT2 (Abcam USA; 1:200) and anti-GalNAcT5 (Abgent, San Diego, CA, USA, 1:200) for 2 hours at room temperature. Then, the cells were washed and probed with Goat-anti-armenian hamster 488/647 (Abcam, USA) and Donkey anti-rabbit 488 (Jackson Immunoresearch laboratories Inc, West Grove, PA, USA) secondiary antibodies. Cells were washed and mounted with Vectashield mounting medium with DAPI (Vector Laboratories, Burlingame, CA, USA). The images were captured by using Zeiss LSM 710 confocal laser scanning microscope (Carl Zeiss, Inc., Thornwood, NY, USA) (Confocal Laser Scanning Fluorescence Microscope Core Facility, UNMC).

Chromatin immunoprecipitation and real-time PCR analysis

Chromatin immunoprecipitation (ChIP) assays were performed according to the EZ-Magna ChIP kit protocol as described previously12 with minor modifications. Equal concentrations of the following antibodies and matched isotype IgG controls were used for ChIP: hamster anti-MUC1-CT (Abcam, USA), rabbit anti-Sp1 (EMD Millipore Corporation, Billerica, MA, USA), mouse anti-p53 (EMD Millipore Corporation, Billerica, MA, USA), mouse anti-c-Jun (Abcam), Armenian hamster IgG (Jackson Immunoresearch laboratories Inc., West Grove, PA, USA), mouse IgG (Jackson Immunoresearch laboratories Inc., West Grove, PA, USA), and rabbit IgG (EMD Millipore Corporation, Billerica, MA, USA). The purified DNA (2 μl) from either immunoprecipitated chromatin or input chromatin was used for quantitative PCR as mentioned above. The ChIP PCR results were calculated by using ΔΔCt method. The following formula (2 (-ΔΔCt [normalized IP-normalized Mock])) has been used to calculate the assay site IP Fold Enrichment above the sample specific background. Real-time PCR primers used for amplification of GALNT5 gene regulatory regions are listed in Supplementary Table 3.

Mithramycin A treatment and real-time PCR analysis

Pancreatic cancer cell line S2–013 Neo was treated with RNA synthesis inhibitor Mithramycin A (200nM) (Sigma-Aldrich, St. Louis, MO, USA) for 48 hours. After treatment, the cells were washed with sterile PBS. Total RNA isolation, cDNA synthesis and real-time PCR analysis were carried out as described above. The following genes were amplified: GALNT5, VEGF-A and 18S rRNA. The primers sequence used for real-time PCR are listed in Supplementary Table 2.

Pifithrin-α hydrobromide treatment and Western blotting

S2–013 MUC1F and S2–013MUC-CT cells were treated with variying concentration (1, 2.5 and 5μM) of p53 inhibitor Pifithrin-α hydrobromide (Tocris, MN, USA). After 24 hours of treatment, the cells were harvested and total proteins were isolated using RIPA buffer containing protease and phosphatase inhibitors. Wstern blotting analysis was performed as described above using the following primary antibodies: rabbit anti-GalNAc-T5 (Abgent, San Diego, CA, USA, 1:1000), hamster anti-MUC1-CT (CT2, Abcam, USA, 1:1000) and mouse anti-β-actin (Sigma-Aldrich, St. Louis, MO, USA, 1:5000).

Statistical Analysis

The analysis of variance (ANOVA) method with Tukey multiple comparison test was used to compare the mean value of normalized relative levels of GALNT5 mRNA between different pancreatic cancer cell lines and the normalized relative levels of region immunoprecipitated at different GALNT5 promoter regions. A p-value of less than 0.05 was considered as statistically significant.

Results

MUC1 induces differential expression of glycosyltransferases

To determine whether MUC1 regulates the expressions of glycosyltransferases, we examined the mRNA expressions of glycosyltransferases in S2–013 pancreatic cancer cells stably overexpressing MUC1 (S2–013 MUC1-F) or vector-transfected control (S2–013 Neo). Schematics of MUC1 constructs MUC1-F and MUC1-CT used in this study are depicted in Figure1A. Real-time PCR analysis showed a significantly increased expression of MUC1 (58.44 fold), GalNAc-T6 (1.8-fold) and GalNAc-T8 (1.7-fold) in S2–013 MUC1 cells as compared to S2–013 Neo cells. Conversely, a significantly reduced expression of GalNAc-T5 (14.3-fold) and ST6GalNAc1 (2.3-fold) was detected in S2–013 MUC1-F cells as compared to S2–013 Neo cells (Figure 1B). Further, we validated these results by knocking down of endogenous MUC1 in PC cells and determined the expression of aforementioned glycosyltransferases. To achieve this, we transfected MUC1-specific siRNA sequence containing vector and control vector into MUC1-expressing PC (FG) cells, and the loss of MUC1 expression was determined by qRT-PCR. FG cells with MUC1 stably knocked down (FG MUC1KD) show a significantly decreased expression of MUC1 (0.26 fold), GalNAc-T6 (1.3 fold) and GalNAc-T8 (2.3 fold), and a significantly increased expression of GalNAc-T5 (3.8-fold) and ST6GalNAc1 (1.5-fold) as compared to its vector control transfected FG cells (FG Neo) (Figure 1C). These results suggest that MUC1 downregulates the expression of GalNAc-T5 and ST6GalNAc1 and upregulates the expression of GalNAc-T6 and GalNAc-T8 in PC cells.

Figure 1.

Figure 1.

Open in a new tab

MUC1 regulates glycosyltransfereases in PC cells. A. Schematic representation of MUC1 constructs. Full length MUC1-F construct including FLAG tag, Tandem Repeat domain, Transmembrane domain and cytoplasmic tail (top) and MUC1-CT construct excluding tandem repeat domain (bottom). B. Real-time RT PCR analysis of GALNT5, GALNT6, GALNT8 and ST6GalNAc-1 in S2–013 Neo and S2–013 MUC1-F cells. C. Real-time RT PCR analysis of GALNT5, GALNT6, GALNT8 and ST6GalNAc-1 in FG Neo and FG MUC1 KD cells. All values have been normalized to 18S rRNA and relative to respective control cells (S2–013 MUC1-F / S2–013 Neo and FG MUC1-KD / FG Neo). The values are expressed as average of three reactions ± s.e. mean (*, p<0.05; **, p<0.005; ns, non-significant). D. Immunoblot analysis of MUC1-CT and GalNAc-T5 expression in S2–013 Neo, S2–013 MUC1-CT, S2–013 MUC1-F, FG-Neo and FG-MUC1 KD cells. β-actin was used as a loading control. E. Immunoflourescence analysis of protein expression of MUC1 (red color) and GalNAc-T5 (green color) in S2–013 MUC1-F and S2–013 Neo cells. The images were taken at the magnification of 20X.

Next, we determined the protein expression of GalNAc-T5 in MUC1 over expressing and knockdown cells. Prior to this, we confirmed the protein expression of MUC1-CT in S2–013 Neo, S2–013 MUC1-CT, S2–013 MUC1-F, FG Neo and FG-MUC1-KD cells. Reduced expression of MUC1-CT was observed in S2–013 Neo and FG MUC1-KD cells (Figure1D, left panel upper lane 1 and right panel upper lane). However, a prominent MUC1-CT expression was observed in S2–013 MUC1-CT and S2–013 MUC1-F cells at 37KDa with subunits at 24 and 55 KDa (Figure 1D, left panel lane 2 and 3). Our immunofluorescence analysis further confirmed these findings that MUC1-CT majorly expressed in the nucleus whereas S2–013 MUC1-F cells exhibited the expression both in cell membrane as well the nucleus (Supplementary Figure 1). We have analyzed the expression of GalNAc-T5 protein in the same sets of cells and found that S2–013 MUC1-CT and S2–013 MUC1-F cells showed a decreased expression of GalNAc-T5 as compared to vector control cells (Figure 1D). Additionally, we found an increased protein expression of GalNAc-T5 in FG MUC1KD cells as compared to their vector control cells (Figure 1D). We further validated these results by immunofluorescence analysis which showed a reduced expression of GalNAc-T5 in S2–013 MUC1-F cells as compared to S2–013 Neo cells (Figure 1E). These results indicate that aberrant expression of MUC1 downregulates the expression of GalNAc-T5 in PC cells.

MUC1-CT localizes to genomic regions of various GALNTs

To determine the mechanism by which MUC1 regulates the expression of glycosyltransferease in PC cells, we utilized our previous MUC1-CT promoter occupancy data (ChIP-on-chip) from S2–013 MUC1 cells12. ChIP-on-chip results showed that MUC1-CT binds to the proximal promoter and within the coding regions (exon) of various GalNAc and sialyltransferases, including the GALNT5 gene (Supplementary Table 1).

MUC1-CT occupies genomic regions in GALNT5 gene

ChIP-on-chip analysis revealed that MUC1-CT occupies the GALNT5 gene proximal promoter (−808/−210) and exon 1 (312/941) sequences, with the signal peak occurring at −419/−323 and 370/507 of those regions, respectively (Figure 2A). Real-time PCR primers were designed for the amplification of −419/−323 and 370/507 regions, as well as for a MUC1 non-binding sequence ~23 kb upstream as a control. Next, we validated this Chip-on-chip results by chromatin immunoprecipitation (ChIP) assay in S2–013 MUC1-F and S2–013 Neo cells using anti-MUC1-CT antibody to confirm the MUC1-CT binding regions on the GALNT5 gene. ChIP results showed a significant enrichment of MUC1-CT binding in the proximal promoter (**p<0.005) and exon 1 region (*p<0.05) in S2–013-MUC1-F cells as compared to S2–013 Neo cells (Figure 2B). However, no change in MUC1 occupancy of the upstream control region was observed. Our ChIP results indicate that MUC1-CT is localized to the GALNT5 gene proximal promoter and exonic sequences and may thereby contribute to the transcriptional regulation of the GALNT5 gene expression.

Figure 2.

Figure 2.

Open in a new tab

Occupancy of MUC1-CT at the GALNT5 promoter. A (top panel). ChIP-on-chip experiments revealed MUC1-CT occupies the proximal promoter and a region within the first exon regions of GALNT5 gene. Image from Integrated Genome Browser (Affymetrix) shows two experimental MUC1-CT versus IgG ChIP-chip replicates. The degree of enrichment for individual oligonucleotide probes was represented by vertical lines. A (bottom panel). Schematic diagram shows the occupancy of MUC1-CT to proximal promoter (−808/−210 bp) and exon 1 (312/941 bp) sequences of GALNT5 gene, base pairs were numbered based on the start codon (ATG). B. Promoter occupancy of MUC1-CT was confirmed by chromatin immunoprecipitation (ChIP) using hamster anti-MUC1-CT or isotype control IgG, followed by real-time PCR analysis. Enrichment of MUC1 binding was observed in the proximal promoter region (−419/−323 bp) and within the exon 1 (370/507 bp) of GALNT5 gene in S2–013 MUC1 cells, either little or no enrichment was detected in S2–013 Neo cells. The MUC1 non-binding regions remain unchanged. Regions of interest values have been normalized with the enrichment detected using isotype control IgG and are expressed as the average of three reactions ± s.e. mean. Amplification of MUC1-CT non-binding region (~23 kbp upstream of transcription start site) was used as a negative control. [TF, transcription factor] [*, p<0.05; **p<0.005; ns, non-significant].

MUC1-CT alters the binding of transcriptional factors in the GALNT5 gene

Transcriptional factors binding sites (TFBS) in the GALNT5 promoter and within the coding regions were predicted by using MatInspector (genomatix, https://www.genomatix.de/online_help/help_matinspector/matinspector_help.html) and Alibaba 2.1 (gene-regulation, http://gene-regulation.com/pub/programs/alibaba2/) online tools. The transcriptional factors p53, c-Jun and Sp1 are predicted to bind the DNA sequences of GALNT5 gene (Supplementary Figure 2) where MUC1-CT occupies.

Next, we performed ChIP experiment to confirm these transcriptional factors occupancy at the proximal promoter and exonic sequences of GALNT5 gene. We oberserved a significantly enhanced occupancy of p53 in the proximal promoter (−419/−323) (p<0.005) and exon 1 region (370/507) (p<0.05) of the GALNT5 gene in S2–013 MUC1-F cells compared to S2–013 Neo cells and no change in the upstream control region (Figure 3A). Similarly, we observed that occupancy of c-Jun at the GALNT5 promoter (p<0.005) and exon 1 regions (p<0.05) were significantly enhanced in MUC1-expressing S2–013 cells, compared to vector control cells (Figure 3B). Also, c-Jun showed significantly increased binding to an upstream control region (p<0.05). In contrast to our p53 and c-Jun ChIP results, we observed a significantly reduced fold enrichment of Sp1 occupancy at the proximal promoter (−419/−323) (***p<0.0005) and exon 1 region (370/507) (**p<0.005) of GALNT5 in S2–013 MUC1-F cells as compared to S2–013 Neo cells and no change in the upstream control region (Figure 3C).

Figure 3.

Figure 3.

Open in a new tab

MUC1 regulates the p53, c-Jun and Sp1 occupancy of the GALNT5 promoter. A. Chromatin immunoprecipitation reveals a significant enrichment of p53 occupancy at the proximal promoter (−419/−323 bp) and exon 1 (370/507 bp) region in cells overexpressing MUC1 cells but the MUC1 non-binding region remains unchanged. B. MUC1 overexpression increased occupancy of c-Jun at the −419/−323, 370/507 and upstream non-binding control regions was seen in S2–013 MUC1 cells. C. MUC1 overexpression significantly decreased Sp1 occupancy at GALNT5 proximal promoter (−419/−323 bp) and exon 1 (370/507 bp) region (S2–013 MUC1 cells), and the MUC1 non-binding region remains unchanged. Regions of interest values have been normalized with the enrichment detected using isotype control IgG and are expressed as the average of three reactions ± s.e. mean. Amplification of MUC1-CT non-binding region (~23 kb upstream of transcription start site) was used as a negative control. D. Mithramycin A treatment and real-time PCR analysis of GALNT5 mRNA expression. GALNT5-expressing S2–013 Neo cells were treated with Mithramycin A (200 nM) for 48 hours showed reduced levels of GALNT5 (p<0.05) and VEGF-A (p<0.05) compared to untreated control cells. Values have been normalized to 18S rRNA. The values are expressed as average of three reactions ± s.e. mean. [*p<0.05;**p<0.005; ***p<0.0005; ns, non-significant].

To confirm the involvement of Sp1 transcription factor-mediated GALNT5 gene transcription in PC cells, we have treated S2–013 Neo cells with Mithramycin A (200 nM for 48 hours), and observed a significant reduction in the GALNT5 (4.18 fold) and VEGF-A (3.9 fold) expression compared to untreated cells S2–013 Neo cells. Amplification of VEGF-A served as a positive control for the Sp1 transcription factor (Figure 3D). Amplification of 18S rRNA was used as an internal control. These results indicate that the occupancy of MUC1-CT at the GALNT5 regulatory elements displaces the Sp1 transcriptional activator in MUC1-expressing S2–013 cells.

MUC1 suppresses GALNT5 expression through complex with p53 and c-Jun

Our ChIP analysis revealed a significantly enhanced occupancy of p53 and c-Jun in the proximal promoter and exon 1 region of the GALNT5 gene in S2–013 MUC1-F cells. In order to confirm that MUC1-CT translocate to the nucleus through binding with p53 and c-Jun, we have analyzed the binding interactions of MUC1-CT with p53 and c-Jun by immunoprecipitation. MUC1 immunoprecipitates from whole cell lysates of S2–013 MUC1-F and S2–013MUC1-CT cells were probed with anti-p53 and anti-C-Jun antibodies. MUC1 immunoprecipitation of p53 increased 6.9 fold (Figure 4A, upper lane) and c-Jun increased 5.6fold (Figure 4A, middle lane) as compared with their respective input controls in S2–013 MUC1-CT cells. These results revealed an increased binding interactions of MUC1-CT with p53 and C-Jun. Next, we asked whether p53 arbitrate the nuclear translocation of MUC1 and mediate transcriptional regulation of GALNT5 gene. In order to prove this, we have treated MUC1-CT cells with Pifithrin-α hydrobromide, a p53 inhibitor for 24 hours and have analyzed the protein expression of GalNAc-T5. We found a significant increase in the expression of GalNAc-T5 in a concentration dependant manner in Pifithrin-α hydrobromide treated MUC1-CT cells as compard to untreated control cells (Figure 4B). A similar results were obetained in S2–013MUC1-F cells treated with p53 inhibitor (Supplementary Figure 3). These results further confim the hypothesis that MUC1-CT translocates to the nucleus and mediate the transcriptional regulation of GALNT5 gene through complex with p53 and c-Jun.

Figure 4.

Figure 4.

Open in a new tab

MUC1 suppresses GalNAc-T5 expression through complex with p53 and c-Jun. A. Immunoprecipitation of MUC1-CT with anti-MUC1-CT antibody (CT2) in S2–013 MUC1-CT cell lysate and probed with p53, c-Jun and MUC1-CT specific antibodies (lane 1: input, lane 2: IP with Armenian Hamster IgG, lane 3: IP with MUC1-CT). Membrane re-probed with anti-Armenian Hamster IgG-HRP was served as loading control. B. Protein expressions of GalNAc-T5 and MUC1-CT2 in S2–013 MUC1-CT cells treated with varying concentrations of Pifithrin-α hydrobromide (1, 2.5 and 5μM) for 24 hours. C. Immunohistochemical analysis of MUC1 (AR20.5 and CT2) and GalNAc-T5 expression in human normal (n=3) and primary pancreatic tumor tissues (n=25). Original magnification x 20. A p value of less than 0.05 was considered as stastically significant. D. Proposed mechanism of MUC1-CT mediated transcriptional down regulation of GALNT5 gene expression through complex with p53 and c-Jun and displacement of Sp1.

Expressions of MUC1 and GalNAc-T5 in human pancreatic tumor tissues

To find out the extent of correlation between MUC1 and GalNAc-T5 expression in clinical specimens of human PDAC, we performed immunohistochemistry staining in normal pancreas and pancreatic tumor tissues using anti-MUC1 (AR20.5 and CT2) and anti-GalNAc-T5 antibodies. We observed a basal level expression of MUC1 in acinar cells and reduced expression in islet cells of normal pancreas, whereas increased expression of MUC1 was observed in all the cell types in primary pancreatic tumor tissues (Figure 4C). In contrast to MUC1 expression, GalNAc-T5 expression was detected in both acinar and islet cells of normal pancreas, and lower expression of GalNAc-T5 was detected in primary pancreatic tumor tissues. Expression of MUC1-F (by AR20.5) and MUC1-CT (by CT2) was significantly high (p<0.0001 and p<0.0379, respectively) when compared with GalNAc-T5 expression in primary pancreatic tumor tissues (Figure 4C).

Discussion

Pancreatic cancer is projected to become the second most leading cause of cancer-related death in U.S. In a normal pancreas, MUC1 is expressed in basal levels on the luminar surface of the ductal epithelial cells. However, aberrant expression of cell surface epithelial MUC1 enhances PC progression and metastasis and is associated with wrose disease prognosis6,26. MUC1 is a transmembrane glycoprotein which is aberrantly glycosylated in PC. Altered promoter occupancy of MUC1-CT along with transcriptional factors is one of the major hallmarks of MUC1-mediated oncogenic transcriptional regulation of target genes during tumor progression and metastasis27. In this study, for the first time, we demonstrate how MUC1 could regulate the expression of various glycosyltransferases in PC. A recent study demonstrated that gastric cancer patients with elevated levels of GalNAc-T5 showed an increased overall survival28. Also, GalNAc-T5 physically interacts with tumor suppressor proteins exostosins with an unknown function19. Altogether, these previous studies suggest that GalNAc-T5-encoded glycophenotypes may possess a tumor suppressor activity. Therefore, we analyzed the expression of various glycosyltrasnsferases in MUC1 stably-overexpressing and MUC1 stably-knockdown PC cells. Our results demonstrate that MUC1 negatively regulates the expression of mucin-type O-glycan-initiating GalNAc-T5 at the transcriptional and translational levels in PC cells. Also, our ChIP-on-chip and real-time PCR analyses revealed that MUC1-CT occupies the promoter elements of various glycosyltransferase genes, including GALNT5, GALNT6, GALNT8 and ST6GALNAC1 (Supplementary Table 1). The selective up regulation of GalNAc-T6 by MUC1 is significant as GalNAc-T6 is aberrantly expressed in PC16. Park et al. have demonstrated that GalNAc-T6-mediated O-glycosylation of MUC1 enhances mammary tumorigenesis through increasing MUC1 protein stability29. In contrast to this, occupancy of MUC1-CT at the GALNT5 gene proximal promoter and exon regions significantly suppresses GalNAc-T5 expression.

Occupancy of MUC1-CT, along with transcriptional factors, on various gene regulatory elements either upregulate or down regulate the expression of target genes in a context-dependent manner9–11. Additionally, ability of MUC1 in potentiating oncogenic signaling through conjugation with transcription factors such as β-catenin, p53 and c-Jun has been reported27. We therefore investigated whether MUC1-CT influences the occupancy of transcriptional factors that are predicted to bind on the GALNT5 promoter and Exon 1 region such as p53, c-Jun and Sp1 where MUC1-CT binds. Our ChIP-qPCR experiments indicated that occupancy by p53 and c-Jun is significantly increased and occupancy by Sp1 is significantly reduced in the GALNT5 proximal promoter and Exon 1 region of MUC1-CT binding sites in S2–013 MUC1 cells. Similar to our findings, previous studies have reported that occupancy of c-Jun at the promoter elements of p53 and human chorionic gonadotropin alpha and beta genes suppresses their expression30,31. Immunoprecipitation analysis further confirmed the direct binding interaction of MUC1 with p53 and c-Jun. Additionally, inhibition of p53 using Pifithrin-α hydrobromide resulted in a significant increase in the expression of GalNAc-T5 in MUC1 overexpressing S2–013 cells indicate the role of MUC1-p53 complex in mediating the transcriptional regulation of GALNT5. Conversely, binding of Sp1 on the promoter elements activate the transcription of multiple genes, including MUC2 and MUC5Ac32. Mithramycin A is a gene specific selective inhibitor of Sp1 transcription factor. Another important finding of our study is treatment of PC cells with Mithramycin A results in reduced expression of Sp1 and suggests that Sp1 is one of the key transcriptional factors for upregulation of GALNT5 expression in cancer cells. We have validated these findings in PC clinical specimens and found that GalNAc-T5 expression is inversely related to MUC1 expression.

In summary, we have demonstrated that overexpression of MUC1 results in downregulation of GalNAc-T5 expression in PC cells. Furthermore, our mechanistic approach revealed that increased binding of MUC1-CT with p53 and c-Jun complex on the GALNT5 promter and Exon 1 region displaces Sp1 and downregulates the expression of GALNT5 in PC cells (Figure 4D). These novel findings confirm that MUC1-CT exerts oncogenic activity by targeting the tumor suppressor function-associated glycosyltransferase GalNAc-T5 in PC.

Supplementary Material

Supplementary information

Acknowledgments

The authors on this manuscript are, in parts supported by grants from the National Insitutes of Health R01 CA208108, P50CA127297, 5U01CA111294, 5U01CA210240, U54 CA163120 and NE-DHHS/LB506. We would like to thank Dr. Sandy Gendler (Mayo Clinic, Scottsdale, AZ, USA) for consultation on use of CT2 antibody. Also, we thank Drs Premila Leiphrakpam and Joyce Solheim (UNMC, Omaha, NE) for critiquing the manuscript.

Abbreviations

PC

Pancreatic Cancer

MUC1-F

MUC1 FLAG Tagged

MUC1-CT

MUC1 cytoplasmic tail

KD

Knockdown

ChIP

Chromatin Immunoprecipitation

qRT-PCR

quantitative Real-time PCR

Footnotes

Conflict of Interest

The authors have declared no conflict of interest.

References

  • 1.Lan MS, Batra SK, Qi WN, Metzgar RS & Hollingsworth MA Cloning and sequencing of a human pancreatic tumor mucin cDNA. J Biol Chem 265, 15294–15299 (1990). [PubMed] [Google Scholar]
  • 2.Hollingsworth MA & Swanson BJ Mucins in cancer: protection and control of the cell surface. Nat Rev Cancer 4, 45–60, doi: 10.1038/nrc1251 (2004). [DOI] [PubMed] [Google Scholar]
  • 3.Levi E, Klimstra DS, Andea A, Basturk O & Adsay NV MUC1 and MUC2 in pancreatic neoplasia. J Clin Pathol 57, 456–462, doi: 10.1136/jcp.2003.013292 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Qu CF et al. MUC1 expression in primary and metastatic pancreatic cancer cells for in vitro treatment by (213)Bi-C595 radioimmunoconjugate. Br J Cancer 91, 2086–2093, doi: 10.1038/sj.bjc.6602232 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Roy LD et al. MUC1 enhances invasiveness of pancreatic cancer cells by inducing epithelial to mesenchymal transition. Oncogene 30, 1449–1459, doi: 10.1038/onc.2010.526 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Tinder TL et al. MUC1 enhances tumor progression and contributes toward immunosuppression in a mouse model of spontaneous pancreatic adenocarcinoma. J Immunol 181, 3116–3125 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Singh PK & Hollingsworth MA Cell surface-associated mucins in signal transduction. Trends Cell Biol 16, 467–476, doi: 10.1016/j.tcb.2006.07.006 (2006). [DOI] [PubMed] [Google Scholar]
  • 8.Kim KC & Lillehoj EP MUC1 mucin: a peacemaker in the lung. Am J Respir Cell Mol Biol 39, 644–647, doi: 10.1165/rcmb.2008-0169TR (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Mahanta S, Fessler SP, Park J & Bamdad C A minimal fragment of MUC1 mediates growth of cancer cells. PLoS One 3, e2054, doi: 10.1371/journal.pone.0002054 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Wei X, Xu H & Kufe D Human mucin 1 oncoprotein represses transcription of the p53 tumor suppressor gene. Cancer Res 67, 1853–1858, doi: 10.1158/0008-5472.CAN-06-3063 (2007). [DOI] [PubMed] [Google Scholar]
  • 11.Singh PK et al. Phosphorylation of MUC1 by Met modulates interaction with p53 and MMP1 expression. J Biol Chem 283, 26985–26995, doi: 10.1074/jbc.M805036200 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Behrens ME et al. The reactive tumor microenvironment: MUC1 signaling directly reprograms transcription of CTGF. Oncogene 29, 5667–5677, doi: 10.1038/onc.2010.327 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Bennett EP et al. Control of mucin-type O-glycosylation: a classification of the polypeptide GalNAc-transferase gene family. Glycobiology 22, 736–756, doi: 10.1093/glycob/cwr182 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Brockhausen I Mucin-type O-glycans in human colon and breast cancer: glycodynamics and functions. EMBO Rep 7, 599–604, doi: 10.1038/sj.embor.7400705 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Taniuchi K et al. Overexpression of GalNAc-transferase GalNAc-T3 promotes pancreatic cancer cell growth. Oncogene 30, 4843–4854, doi: 10.1038/onc.2011.194 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Li Z et al. Polypeptide N-acetylgalactosaminyltransferase 6 expression in pancreatic cancer is an independent prognostic factor indicating better overall survival. Br J Cancer 104, 1882–1889, doi: 10.1038/bjc.2011.166 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Onitsuka K et al. Prognostic significance of UDP-N-acetyl-alpha-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase-3 (GalNAc-T3) expression in patients with gastric carcinoma. Cancer Sci 94, 32–36 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Wood LD et al. The genomic landscapes of human breast and colorectal cancers. Science 318, 1108–1113, doi: 10.1126/science.1145720 (2007). [DOI] [PubMed] [Google Scholar]
  • 19.Simmons AD et al. A direct interaction between EXT proteins and glycosyltransferases is defective in hereditary multiple exostoses. Hum Mol Genet 8, 2155–2164 (1999). [DOI] [PubMed] [Google Scholar]
  • 20.McCormick C, Duncan G, Goutsos KT & Tufaro F The putative tumor suppressors EXT1 and EXT2 form a stable complex that accumulates in the Golgi apparatus and catalyzes the synthesis of heparan sulfate. Proc Natl Acad Sci U S A 97, 668–673 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Vezeridis MP et al. In vivo selection of a highly metastatic cell line from a human pancreatic carcinoma in the nude mouse. Cancer 69, 2060–2063, doi: (1992). [DOI] [PubMed] [Google Scholar]
  • 22.Iwamura T, Katsuki T & Ide K Establishment and characterization of a human pancreatic cancer cell line (SUIT-2) producing carcinoembryonic antigen and carbohydrate antigen 19–9. Jpn J Cancer Res 78, 54–62 (1987). [PubMed] [Google Scholar]
  • 23.Burdick MD, Harris A, Reid CJ, Iwamura T & Hollingsworth MA Oligosaccharides expressed on MUC1 produced by pancreatic and colon tumor cell lines. J Biol Chem 272, 24198–24202 (1997). [DOI] [PubMed] [Google Scholar]
  • 24.Tsutsumida H et al. RNA interference suppression of MUC1 reduces the growth rate and metastatic phenotype of human pancreatic cancer cells. Clin Cancer Res 12, 2976–2987, doi: 10.1158/1078-0432.CCR-05-1197 (2006). [DOI] [PubMed] [Google Scholar]
  • 25.Tassone F et al. Elevated levels of FMR1 mRNA in carrier males: a new mechanism of involvement in the fragile-X syndrome. Am J Hum Genet 66, 6–15, doi: 10.1086/302720 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Nath S et al. MUC1 induces drug resistance in pancreatic cancer cells via upregulation of multidrug resistance genes. Oncogenesis 2, e51, doi: 10.1038/oncsis.2013.16 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Wen Y, Caffrey TC, Wheelock MJ, Johnson KR & Hollingsworth MA Nuclear association of the cytoplasmic tail of MUC1 and beta-catenin. J Biol Chem 278, 38029–38039, doi: 10.1074/jbc.M304333200 (2003). [DOI] [PubMed] [Google Scholar]
  • 28.He H et al. Clinical significance of polypeptide N-acetylgalactosaminyl transferase-5 (GalNAc-T5) expression in patients with gastric cancer. Br J Cancer 110, 2021–2029, doi: 10.1038/bjc.2014.93 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Park JH et al. Critical roles of mucin 1 glycosylation by transactivated polypeptide N-acetylgalactosaminyltransferase 6 in mammary carcinogenesis. Cancer Res 70, 2759–2769, doi: 10.1158/0008-5472.CAN-09-3911 (2010). [DOI] [PubMed] [Google Scholar]
  • 30.Schreiber M et al. Control of cell cycle progression by c-Jun is p53 dependent. Genes Dev 13, 607–619 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Pestell RG, Hollenberg AN, Albanese C & Jameson JL c-Jun represses transcription of the human chorionic gonadotropin alpha and beta genes through distinct types of CREs. J Biol Chem 269, 31090–31096 (1994). [PubMed] [Google Scholar]
  • 32.Perrais M, Pigny P, Copin MC, Aubert JP & Van Seuningen I Induction of MUC2 and MUC5AC mucins by factors of the epidermal growth factor (EGF) family is mediated by EGF receptor/Ras/Raf/extracellular signal-regulated kinase cascade and Sp1. J Biol Chem 277, 32258–32267, doi: 10.1074/jbc.M204862200 (2002). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary information
NIHMS1558618-supplement-Supplementary_information.docx (11.3MB, docx)

RESOURCES