The MUC1 extracellular domain subunit is found in nuclear speckles and associates with spliceosomes

Abstract

MUC1 is a large transmembrane glycoprotein and oncogene expressed by epithelial cells and overexpressed and underglycosylated in cancer cells. The MUC1 cytoplasmic subunit (MUC1-C) can translocate to the nucleus and regulate gene expression. It is frequently assumed that the MUC1 extracellular subunit (MUC1-N) does not enter the nucleus. Based on an unexpected observation that MUC1 extracellular domain antibody produced an apparently nucleus-associated staining pattern in trophoblasts, we have tested the hypothesis that MUC1-N is expressed inside the nucleus. Three different antibodies were used to identify MUC1-N in normal epithelial cells and tissues as well as in several cancer cell lines. The results of immunofluorescence and confocal microscopy analyses as well as subcellular fractionation, Western blotting, and siRNA/shRNA studies, confirm that MUC1-N is found within nuclei of all cell types examined. More detailed examination of its intranuclear distribution using a proximity ligation assay, subcellular fractionation, and immunoprecipitation suggests that MUC1-N is located in nuclear speckles (interchromatin granule clusters) and closely associates with the spliceosome protein U2AF65. Nuclear localization of MUC1-N was abolished when cells were treated with RNase A and nuclear localization was altered when cells were incubated with the transcription inhibitor 5,6-dichloro-1-b-d-ribofuranosylbenzimidazole (DRB). While MUC1-N predominantly associated with speckles, MUC1-C was present in the nuclear matrix, nucleoli, and the nuclear periphery. In some nuclei, confocal microscopic analysis suggest that MUC1-C staining is located close to, but only partially overlaps, MUC1-N in speckles. However, only MUC1-N was found in isolated speckles by Western blotting. Also, MUC1-C and MUC1-N distributed differently during mitosis. These results suggest that MUC1-N translocates to the nucleus where it is expressed in nuclear speckles and that MUC1-N and MUC1-C have dissimilar intranuclear distribution patterns.

Figure 1. MUC1 extracellular domain antibody staining patterns in cultured epithelial cells.

The indicated cells were fixed, permeabilized and stained for immunofluorescence microscopy using B27.29, HMFG1, DF3, or matched isotype control immunoglobulins (Control Ig) antibodies as described in Methods. DAPI was used to identify nuclei. The insets are higher magnification views of selected nuclei in trophoblast cells (Troph) to more clearly reveal the speckled fluorescence pattern. Arrows indicate other examples of nucleus-associated staining and arrow heads indicate examples of cytoplasmic/membrane staining. The asterisk indicates dividing cells in which HMFG1 staining is excluded from the nucleus. Images are representative of 3–8 independent experiments. The white bars represent 10 µm.

Figure 2. MUC1 extracellular domain antibody staining pattern in colonic tissue.

Cryosections of rhesus monkey colonic tissues were processed for staining with B27.29 and HMFG1 antibodies as described in Methods. Nuclei were identified using DAPI. Arrows indicate examples of nucleus-associated staining and arrow heads indicate examples of apical plasma membrane staining of ductal epithelial cells. The white bar represents 10 µm. Images are representative of 3 independent experiments.

Figure 3. Nuclear localization of MUC1 extracellular domain antibody-reactive proteins.

(A) Trophoblasts (troph) and MCF-7 cells were stained with B27.29, HMFG1, or β1-integrin antibodies (each shown in red) and then examined by confocal microscopy. Nuclei were stained using DAPI (blue). The images show the staining patterns roughly midway through the respective z-series. Lateral projections (the plane of view is indicated by the yellow horizontal and vertical lines) of individual z-stack series are shown below and to the right of each image. The bars represent 5 µm. The asterisk indicates cytoplasmic/membrane staining. (B) Western blot analysis of total lysates obtained from Jar, MCF-7, and COS7.MUC1 cells using B27.29, DF3, and HMFG1 antibodies. (C) Western blot analysis of subcellular fractions prepared from Jar, MCF-7, and COS7.MUC1 cells using MUC1 antibodies. (D) Western blot analysis of subcellular fractions using antibodies against marker proteins (GAPDH, U2AF65, β1-integrin, spliceosomes, and Sp1). The abbreviations for the subcellular fractions are, C; cytoplasmic, M; membrane, N; nuclear soluble, Nc; nuclear chromatin, Cs; cytoskeletal. The images are representative of 3–5 independent experiments.

Figure 4. Effect of MUC1 siRNAs and shRNA on nuclear MUC1 expression.

Jar cells were transfected independently with four different MUC1 siRNAs, non-targeting siRNA (NT), or GAPDH siRNA (see Methods). Five days after transfection cells were (A) stained using B27.29 or HMFG1 or (B) lysed and analyzed by Western blotting. In other experiments, Jar cells were stably transfected with MUC1 shRNA as described in Methods and stained using DF3 and HMFG1 antibodies (C) or lysed and analyzed for MUC1 expression by Western blotting (D) and RT-PCR (E). GAPDH was used as a loading control for Western blotting and RT-PCR. NT; non-targeting control. Reagent; transfection reagent alone. Medium; culture medium alone.

Figure 5. Nuclear MUC1 localizes with the spliceosome protein, U2AF65.

(A) Trophoblasts were double-stained stained with HMFG1 (Green) and either antibody against U2AF65, matrin-3, or lamin B1 (all shown in Red). The cells were then examined by confocal microscopy. Representative images from the middle of the respective z-stacks are shown. Where there is overlap of green and red in the merged images, a yellow color is produced. (B) Proximity Ligation Assay of MUC1 and U2AF65. BeWo choriocarcinoma cells were fixed, permeabilized and processed for the proximity ligation assay (PLA) using the indicated combinations of primary antibodies/control immunoglobulin as described in Methods. Illustrated controls consist of non-immune rabbit IgG (rb-IgG; used in place of U2AF65), PLA probes alone (no primary antibodies), and B27.29 plus the nuclear matrix protein, matrin-3. Negative results were also obtained using control mouse Ig (in place of B27.29) and U2AF65 (not shown). This experiment was carried out three times with identical results. The white horizontal bars on the micrographs represent 5 µm. (C) Nuclear speckles were isolated from Jar cells and analyzed by Western blotting using HMFG1, B27.29, or DF3 antibodies. The blot on the right of C shows analysis of the nuclear speckle fraction using antibodies against marker proteins. (D) Nuclear speckles were isolated from COS7.MUC1 cells and analyzed by Western blotting. (E) Jar cells were lysed and analyzed by immunoprecipitation (IP) using antibody DF3 or control mouse Ig as described in Methods. Immunoprecipitates were subjected to Western blotting (IB) and probed using DF3 antibody or antibody against U2AF65, as indicated. The numbers represent molecular mass (kDa).

Figure 6. Effect of RNase A and transcriptional inhibition on the intranuclear distribution of MUC1.

(A) BeWo cells were fixed and permeabilized using methanol and then incubated with RNase A (100 µg/mL) for 2 h as described in Methods. The cells were then stained with antibodies against MUC1 (HMFG1 and B27.29), U2AF65, and spliceosomes as indicated. Nuclei were stained with DAPI. (B) BeWo cells were incubated in the presence of DRB (100 µM) for 2 h and then stained with antibodies against MUC1, U2AF65, and spliceosomes (Spl) as described in Methods. The bar represents 5 µm. Results are representative of 3 independent experiments.

Figure 7. Comparison of the expression of MUC1-N and MUC1-C.

(A) Confocal z-series images of Jar cell nuclei double-stained with antibodies HMFG1 (red) and the CT2 (green). The arrowhead indicates CT2 staining in a nucleolus and the arrow indicates CT2 staining at the nuclear periphery. Note the absence of HMFG1 staining in these locations. (B) Lateral projections along x and y from the same z-series shown in B. The yellow lines indicate the planes of view for the lateral projections. Note the more apical distribution of HMFG1 staining and the more basal staining of CT2 staining. (C) Single section confocal images of a Jar cell in metaphase and stained with HMFG1 and CT2. The white horizontal bars indicate 5 µm. (D) Nuclear speckles (Sp) were isolated from Jar cells and analyzed by native polyacrylamide gel electrophoresis followed by Western blotting using the indicated antibodies. The same antibodies were used to analyze Western blots prepared from Jar membrane (M) fractions. Note the prominent band detected by CT2 in the membrane fraction was not found in nuclear speckles. The numbers represent molecular mass (kDa). Results are representative of 3 independent experiments.