Structural organization of the human ESE-1 protein. The ESE-1 protein is depicted from the amino end (N) to the carboxy end (C). The Pointed domain, TAD, SAR domain, AT hook/NLS domain, and ETS domain are shown. Short black lines reveal the positions of NLS1 (aa 238 to 269) and putative NES1 (aa 102 to 112), NES2 (aa 275 to 284), and NLS2 (aa 316 to 323).

IHC staining of endogenous ESE-1 protein in MCF-12A and T-47D cell lines, in human breast cancer, and in normal breast tissue. Sections from paraffin-embedded MCF-12A (A), T-47D (B), human DCIS (C, D, and F), and normal breast tissue (E) blocks were incubated with an anti-ESE-1 primary antibody (A to E) or with rabbit immune serum (F) and then stained with an HRP-conjugated secondary antibody. Panels C (DCIS), E (normal breast), and F (DCIS) represent adjacent sections from the same paraffin-embedded block, whereas panel D shows anti-ESE-1 antibody staining of a separate DCIS sample. The brown stain represents positive anti-ESE-1 antibody staining, and arrowheads indicate representative stained cells in each sample (B to E). The absence of brown stain (A and F) represents the absence of anti-ESE-1 antibody binding (negative control). Stained sections were photographed at a magnification of ×40.

Transiently expressed ESE-1 protein is localized in the nucleus and induces apoptosis in MCF-12A and MCF-10A cells but not in T-47D and SK-BR-3 cells. (A) Fluorescence imaging of transiently expressed GFP-ESE-1 and GFP only in living MCF-12A cells. Representative MCF-12A cells transiently transfected with either the pEGFP-ESE-1 plasmid (panels 1 to 3; four cells shown) or the pEGFP-C3 control plasmid (panels 4 to 6; one cell shown) were imaged by digital deconvolution UV microscopy (magnification, ×100) 12 h after transfection. All transfectants were stained with DAPI or with FM 4-64 prior to microscopy. Panels 1 and 4 show overlays of DAPI (blue) and FM 4-64 (red) fluorescence images. Panels 2 and 5 show overlays of GFP (green) and FM 4-64 (red) fluorescence images. Panels 3 and 6 represent overlays of all three (DAPI, GFP, and FM 4-64) fluorescence signals. Although only four representative cells are shown in the GFP-ESE-1 panels and only one representative cell is shown in the GFP-only panels, a ∼30% transfection efficiency was consistently achieved per 10⁶ cells transfected and ∼10⁵ green fluorescent cells were present following each transfection. Approximately 95% of cells transiently transfected with GFP-ESE-1 showed a nucleus-only GFP-ESE-1 pattern, as shown here, while the other approximately 5% showed a primarily nuclear GFP signal with an associated faint cytoplasmic GFP signal (data not shown). All cells transfected with GFP only showed the pattern displayed in panels 4 to 6. (B) IHC staining of ESE-1 in Ad.ESE-1- and Ad.empty-infected MCF-12A cells. Sections from paraffin-embedded MCF-12A cells infected with either Ad.ESE-1 or Ad.empty were stained with an anti-ESE-1 antibody and then counterstained with an HRP-conjugated secondary antibody. The brown DAB precipitate represents positive anti-ESE-1 antibody staining in the cell nuclei (left panel), and the lack of brown stain (right panel) shows the absence of ESE-1 protein expression in the negative control. Stained sections were photographed at a magnification of ×40. (C) Transiently expressed nuclear GFP-ESE-1 induces apoptosis in MCF-12A and MCF-10A cells but not in T-47D or SK-BR-3 cells. MCF-12A, MCF-10A, T-47D, and SK-BR-3 cells were transfected with either pEGFP-C3 or pEGFP-ESE-1, cultured in complete medium for 28 h, and then stained with an apoptosis-specific annexin V-Alexa Fluor 647 conjugate. FACS analysis was used to quantitate the number of cells positive for both green (GFP-ESE-1 or GFP only) and red (annexin V; apoptosis specific) fluorescence in each transfectant population. The percentage of green and red cells in each GFP-only population was normalized to 1, and the fold increase in apoptotic transient transfectants of MCF-12A, MCF-10A, T-47D, and SK-BR-3 GFP-ESE-1 cells is shown. Three separate transfection experiments were performed for each plasmid, with the same amounts of both plasmid DNA and cells in each experiment, and the resulting data are shown as means and standard deviations. Finally, Western blots corresponding to each transiently transfected cell line are shown immediately below the apoptosis bar graph. From top to bottom, panels show GFP-ESE-1 (anti-GFP Western blot), GFP only (anti-GFP Western blot), and β-tubulin (anti-β-tubulin Western blot) for each cell population. Whole-cell extracts were prepared from each transfectant population 12 h after transfection, and 70 μg of total protein from each extract was used for SDS-PAGE and Western blotting. (D) Fluorescence imaging of transiently expressed GFP-ESE-1 and GFP only in living MCF-10A, T-47D, and SK-BR-3 cells. Representative MCF-10A, T-47D, and SK-BR-3 cells transiently transfected with either the pEGFP-ESE-1 plasmid (panels 1 to 3) or the pEGFP-C3 control plasmid (panels 4 to 6) were imaged for GFP fluorescence by digital deconvolution UV microscopy (magnification, ×100) 12 h after transfection. Broken white lines depict cell outlines. GFP-ESE-1 localized in the nucleus was observed in MCF-10A cells (panel 1, two cells shown), T-47D cells (panel 2, four cells shown), and SK-BR-3 cells (panel 3, three cells shown), whereas GFP-only fluorescence was equally distributed between the nucleus and the cytoplasm in each GFP-only-transfected cell line (panels 4 to 6, one cell shown per transient transfection). Details for transfection efficiencies, numbers of GFP-ESE-1- and GFP-only-positive cells, and subcellular distributions of GFP-ESE-1 and GFP only in each transient transfectant population were as described for panel A.

Stable expression of GFP-ESE-1 transforms MCF-12A cells. (A) RT-PCR strategy. Diagram of GFP-ESE-1 showing the positions of the forward 5′ GFP-specific primer (top arrow) and the reverse 3′ untranslated region-specific primer (bottom arrow) used in RT-PCR assays for GFP-ESE-1 (1,285 bp) and GFP only (169 bp) in stably transfected MCF-12A cell pools. (B) Expression of GFP-ESE-1 and of GFP only in stably transfected MCF-12A cells. The products of RT-PCRs with the primers described above and total RNA purified from each cell population were analyzed by 1% agarose gel electrophoresis. Lane 1 shows a no-template negative RT-PCR control. Two GFP-ESE-1 stable transfectant pools, GFP-ESE-1A (lane 2) and GFP-ESE-1B (lane 3), and two GFP-only stable transfectant pools, GFP A (lane 4) and GFP B (lane 5), were independently generated from MCF-12A cells. Lanes 2 and 3 show the expected GFP-ESE-1-specific 1,285-bp product, and lanes 4 and 5 show the expected GFP-only-specific 169-bp product. Total RNA from each stable transfectant pool was also treated with RNase A (GFP-ESE-1A, GFP-ESE-1B, GFP A, and GFP B in lanes 6 to 9, respectively) prior to RT-PCR. (C) Anchorage-independent colony formation by control MCF-12A and T-47D cells and by stably transfected MCF-12A cell pools. Cells were plated in soft agarose, and typical colonies were photographed 8 days later (magnification, ×40). The 12A and GFP-only panels each show a cluster of three cells (untransfected MCF-12A control cells and MCF-12A cells stably transfected with GFP only, respectively). The T-47D and GFP-ESE-1 panels each show a multicellular colony (T-47D human breast cancer cells and MCF-12A cells stably transfected with GFP-ESE-1, respectively).

Deletion mapping of the GFP-ESE-1 transforming domain. (A) Structural organization and transiently expressed protein localization for ESE-1 internal deletion mutants and SV40 NLS fusion constructs. For each of the seven constructs shown, the ESE-1 structural domains are depicted in the same color scheme at that used in Fig. 1, and each is fused to GFP at its amino terminus (data not shown). Intact GFP-ESE-1 is depicted at the top, followed by constructs with internal deletions of the Pointed domain (GFP-ESE-1ΔPointed), TAD (GFP-ESE-1ΔTAD), SAR domain (GFP-ESE-1ΔSAR), and AT-hook/NLS domain (GFP-ESE-1ΔAT/NLS). The bottom two constructs contain an in-frame insertion of the SV40 NLS between GFP and the ESE-1ΔAT/NLS deletion (GFP-NLS-ESE-1ΔAT/NLS) and between GFP and the intact ESE-1 sequence (GFP-NLS-ESE-1). A representative image of the corresponding GFP fusion localization in transiently transfected MCF-12A cells 12 h after transfection is shown to the right of each construct. Broken white lines depict cell outlines; except for the GFP-ESE-1ΔAT/NLS construct (two cells shown), one representative cell is shown for each construct. (B) Anchorage-independent growth of MCF-12A cell pools stably transfected with GFP-ESE-1 internal deletion or SV40 NLS fusion constructs. Soft-agarose cultures were prepared and photographed as described in the legend to Fig. 4C (magnification, ×40). A representative colony for each GFP-only and GFP-ESE-1 construct depicted in panel A is shown. Small, two- to four-cell clusters were generated in MCF-12A cell pools stably transfected with GFP-only, GFP-ESE-1ΔSAR, GFP-NLS-ESE-1ΔAT/NLS, and GFP-NLS-ESE-1. Large, multicellular colonies were generated in MCF-12A cell pools stably transfected with intact GFP-ESE-1, GFP-ESE-1ΔPointed, GFP-ESE-1ΔTAD, and GFP-ESE-1ΔAT/NLS. (C) Quantitation of anchorage-independent colony formation. Untransfected MCF-12A and T-47D cells and two independent, stably transfected MCF-12A cell populations for each GFP construct were seeded in triplicate soft-agarose cultures, and all colonies visible by ×6 light microscopy were counted 21 days later. Each bar shows the mean and standard deviation number of total colonies per plate for each cell line or transfectant population.

Deletion of the ESE-1 TAD impairs GFP-ESE-1-mediated apoptosis in transiently transfected MCF-12A cells, and the SAR domain, either as GFP-SAR or GFP-NLS-SAR, is insufficient to recapitulate GFP-ESE-1 apoptotic function. (A) Map and amino acid sequence of GFP-SAR and GFP-NLS-SAR. For GFP-SAR, GFP is indicated by a broken green box (not to scale) followed by the in-frame fusion of a 50-aa ESE-1 region (aa 190 to 239) spanning all but the first 2 aa of the SAR domain. The amino acid sequence of the 50-aa region is shown, with the SAR sequence in bold italic type and carboxy-terminal ESE-1-flanking amino acids in plain type. The SV40 NLS sequence and the in-frame insertion site used to generate GFP-NLS-SAR are also shown. (B) Subcellular localization of GFP-SAR and GFP-NLS-SAR in transiently transfected MCF-12A cells. MCF-12A cells transiently transfected with the GFP-SAR or GFP-NLS-SAR expression plasmid were stained with DAPI, and typical cells in each culture were imaged by digital deconvolution UV microscopy (magnification, ×100). The DAPI, GFP, and merge images are as described in the legend to Fig. 3A, and broken white lines represent cell outlines. Panels 1 to 3 show the diffuse cytoplasmic and nuclear localization of GFP-SAR in transiently transfected MCF-12A cells. Panels 4 to 6 shows the exclusive nuclear localization of transiently expressed GFP-SAR that results from the SV40 NLS fusion. (C) Analysis of the roles of the ESE-1 TAD and SAR domain in the apoptotic activity of transiently expressed GFP-ESE-1. MCF-12A cells were transiently transfected with the GFP-only, GFP-ESE-1, GFP-ESE-1ΔTAD, GFP-SAR, or GFP-NLS-SAR expression plasmid. As described in the legend to Fig. 3C, apoptosis-specific annexin V staining in each transfectant population was examined 12 h later. The percentage of green and red cells in GFP-only transient transfectants was normalized to 1, and the fold increase in apoptosis for GFP-ESE-1, GFP-ESE-1ΔTAD, GFP-SAR, and GFP-NLS-SAR transient transfectants is shown. Three separate transfection experiments were performed for each plasmid, with the same amounts of both plasmid DNA and MCF-12A cells in each experiment, and the resulting data are shown as means and standard deviations. Finally, Western blots corresponding to each transient transfectant are shown immediately below the apoptosis bar graph. From top to bottom, panels show GFP-ESE-1 and GFP-ESE-1ΔTAD (anti-GFP Western blot), GFP-SAR and GFP-NLS-SAR (anti-GFP Western blot), GFP only (anti-GFP Western blot), and β-tubulin (anti-β-tubulin Western blot) for each cell population. Whole-cell extracts were prepared from each transfectant population 28 h after transfection, and 50 μg of total protein from each extract was subjected to SDS-PAGE and Western blotting.

Quantitation of colony formation and analysis of subcellular localization of of GFP-SAR and GFP-NLS-SAR in stably transfected MCF-12A cell pools. (A) Anchorage-independent growth of MCF-12A cell pools stably transfected with GFP-SAR or GFP-NLS-SAR. Soft-agarose cultures were prepared for each cell line, and representative colonies were photographed as described in the legends to Fig. 4C and 5B (magnification, ×40). GFP-only and GFP-NLS-SAR stable transfectant pools each showed two- to four-cell clusters, and GFP-ESE-1 and GFP-SAR stable transfectant pools each showed large, multicellular colonies. (B) Quantitation of anchorage-independent colony formation in soft-agarose cultures seeded with MCF-12A cell pools stably transfected with GFP-SAR or GFP-NLS-SAR. Three independent MCF-12A populations stably transfected with the GFP-SAR or GFP-NLS-SAR expression plasmid were seeded in triplicate soft-agarose cultures, and colonies were counted as described in the legend to Fig. 5C. Each bar shows the mean and standard deviation number of total colonies per plate for each transfectant population. (C) Subcellular localization of stably expressed GFP-SAR and GFP-NLS-SAR. Pools of MCF-12A cells stably transfected with the GFP-SAR or GFP-NLS-SAR expression plasmid were stained with DAPI, and typical cells in each culture were imaged by digital deconvolution UV microscopy (magnification, ×100). The DAPI, GFP and merge images are as described in the legend to Fig. 3A, and broken white lines represent cell outlines. Panels 1 to 3 show the diffuse cytoplasmic and nuclear localization of stably expressed GFP-SAR. Panels 4 to 6 show the exclusive nuclear localization that results from fusing the SV40 NLS to GFP-SAR.