Figure 1.
Estrogen receptor promoter binding in PRL-HeLa cells.
GFP-ER was transiently expressed in PRL-HeLa cells and then treated with ethanolic vehicle, 17-β estradiol, EGF or 4-hydroxytamoxifen for 2 hours prior to fixation. Each ligand concentration was 10 nM except for EGF, which was 100 ng/ml. A. Representative images of PRL-HeLa cells treated with EGF. Arrowheads point to visible arrays in two cells. One of the cells has two fluorescent foci/arrays, while the other has one larger focus. B The percentage of each cell population that has visible accumulation at the array in response to each treatment has been calculated and graphed (n>200). Bars indicate Standard Deviations from 3 different experiments. Student t-test was performed for each bar compared to vehicle treatment. *p<0.05.
Figure 2.
Role of AF-1 and AF-2 ER domains in promoter binding.
Representative images of PRL-HeLa cells transiently expressing the indicated deletion mutants and treated with either ethanolic vehicle, E2 or EGF for 2 hours (see Table 1). The inset images are representative of fluorescence images observed in a minority of cells. The size bar is in microns. All images were captured so as to optimize the intensity of the GFP-ER.
Table 1.
Promoter targeting and chromatin status in PRL-HeLa for ER deletion mutants.
Figure 3.
Large-scale chromatin decondensation of the PRL-array.
A. PRL-HeLa cells expressing GFP-ER were treated for different times with ethanol (vehicle), E2 or EGF. After fixing and counterstaining with DAPI, cells were imaged and the array size was quantified using high throughput microscopy. B. This panel represents earlier time points to show the slower effect of EGF in inducing maximal decondensation of the array. Data represent the mean ±SEM of three different experiments. B. PRL-HeLa cells transiently expressing GFP-ER were imaged live at 37°C under constant perfusion of fresh medium (plus 100 ng/ml EGF after time 0). Image stacks were recorded every four minutes and are presented as projections. The value indicates the area of the array (µm2) shown in the image. Addition of EGF caused chromatin decondensation at the array within 30 minutes.
Figure 4.
ER transcriptional activity at the promoter array.
PRL-HeLa cells transiently expressing GFP-ER were treated with E2, EGF, 4-hydroxytamoxifen (4HT) or ethanolic vehicle for the indicated time. Subsequent to ligand treatment the cells were fixed and subjected to an RNA FISH protocol using a biotinylated dsRED2 probe followed by fluorescent-tagged streptavidin. A. Representative images of a single cell for each treatment. The presence of transcripts at the promoter array is identified by accumulated signal above the level for the nucleoplasm. The inset values (red type) represent the amount of transcript at 2 hours, relative to vehicle controls. B. Representative image of PRL-HeLa cells transfected with GFP-ER (green) exemplify the heterogeneity of ER expression levels. The cells show the RNA FISH signal associated with the array in the cell population (red signal) (I). The nuclear GFP-ER mean of fluorescence and RNA FISH array signal were plotted for vehicle- and E2-treated cells. Each symbol represents the measurements from a singe cell (II). C. To quantify FISH signal over 24 hours the total intensity of signal at the array (minus background signal) was determined by cumulative summation of 20 planes. Data represent the mean ±SEM of three different experiments graphed as fold induction over mean time-matched vehicle-treated control cells. D. This panel represents the earlier time points for E2 and EGF treatment. Fold activation between EGF and E2 at 30 minutes was significantly different, with a p value of 0.03.
Figure 5.
EGFR signaling and PRL array responses.
PRL-HeLa cells transiently transfected with GFP-ER were pretreated or not with Tryphostin AG537 for 30 minutes before adding ethanol (vehicle), E2, EGF or E2 and EGF together for 2 hours. A. Promoter targeting: after fixing and counterstaining with DAPI, cells were imaged and the percentage of each cell population that showed visible accumulation at the array in response to treatment was calculated. Student t-test was performed for EGF treatment compared to each other treatments groups (* p<0.05, † p<0.03). B. Large-scale chromatin modification: the same images were acquired using high throughput microscopy and the array size was quantified as described in the Methods. Differences from respective vehicle treatment reaching statistical significance (p<0.05) are labeled by asterisks (*). C. Transcriptional activity: subsequent to ligand treatment, the cells were fixed and subjected to RNA FISH. The FISH signals at the array were quantified in >20 cells for each protein and treatment condition and graphed as the average total array-associated fluorescence ±SEM. Student t-test was performed for each treatment group compared to its control (* p<0.05), or to vehicle († p<0.05).
Figure 6.
MAPK signaling and the role ER-S118 in PRL array responses.
A. PRL-HeLa cells transiently expressing GFP-ER were treated with 50 µM PD98589 or 10 µM UO126 for 30 minutes before adding 10 nM E2 or 100 ng/ml EGF for 2 h. The percentage of each cell population that showed visible accumulation at the array in response to treatment was calculated and reported in the graphs (n>200). Both inhibitors dramatically reduced the number of GFP-ER-targeted arrays in transfected cells (EGF treated only) p<0.05. B. Percentage of cells with a visible array in PRL-HeLa cells transiently transfected with GFP-ER or with the two mutants, GFP-ERS118A and GFP-ERS118E, and treated with ethanol, E2 or EGF for 2 hours. Student t-test was performed for each bar compared to GFP-ER (* p<0.05), or to vehicle († p<0.05).
Figure 7.
Temporal single-cell analyses of ER-S118A and ER-S118E transcriptional response at the PRL-array.
PRL-HeLa cells transiently expressing the two phosphomutants, GFP-ERS118A (panel A and C) or GFP-ERS118E (panel B and D), were treated with either ethanol, E2 (panel A and B) or EGF (panel C and D) for different times. Subsequent to ligand treatment, the cells were fixed and subjected to RNA FISH using a biotinylated dsRED2 probe followed by fluorescent-tagged streptavidin. The FISH signal at the array was determined as described and the value graphed as fold induction over vehicle control cells (solid line). For each cell the area of the array was also determined and plotted as fold induction over vehicle control (dotted line). Data represent the mean ±SEM of three different experiments graphed as fold induction over time-matched vehicle-treated control cells.