Novel senescence associated gene, YPEL3, is repressed by estrogen in ER+ mammary tumor cells and required for tamoxifen-induced cellular senescence

Reverse transcription-PCR

Total RNA was isolated using the e.Z.N.A. Total RNA kit (Omega Bio-Tek) according to the manufacturer’s instructions. RNA quantity was assessed using the Nanodrop with 260/280 ratios ranging from 2 to 2.2. One microgram of total RNA was used as a template for cDNA synthesis using the qScript cDNA SuperMix (Quanta Biosciences). Taqman-based PCR was performed in triplicate using Assay on Demand probe sets (Applied Biosystems) and an Applied Biosystems 7900 Sequence Detection System. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an endogenous control. All Assay on Demand (AOD) probes used in these experiments (YPEL3, pS2, p21, p53, ERα and GAPDH) were validated gene targets. Error bars in each Figure represent 95% confidence interval from a single experiment. All experiments were completed in biological triplicate with similar results.

Cell lines and reagents

MCF-7 and ZR-75.1 cells were derived from breast carcinomas and express wild-type estrogen receptor alpha and p53. T-47D cells were derived from a ductal carcinoma and are wild-type estrogen receptor alpha and express mutant p53. SkBr-3 cells were derived from breast carcinoma and are devoid of estrogen receptor alpha and express mutant p53. All cell lines were purchased from American Type Culture Collection. All cell lines passages in these experiments were grown for no longer than 3 months. MCF-7 and SKBr3 cells were grown in DMEM with 10% fetal bovine serum. ZR-75.1 and T-47D cells were grown in RMPI-1640 with 10% fetal bovine serum. Charcoal/Dextran stripped fetal bovine serum was obtained from Invitrogen. Estrogen and Tamoxifen were obtained from Sigma and MP Biomedicals, LLC, respectively. Chemical resuspension and storage was per manufacturer’s recommendations.

Immunoblotting

Whole-cell extracts and Western blotting were performed as previously described with the following modifications ¹⁴. Cell pellets were lysed in three freeze-thaws using RIPA buffer [25 mmol/L Tris-HCl (pH 7.6), 150 mmol/L NaCl, 1% NP40, 1% sodium deoxycholate] containing a protease inhibitor cocktail (Sigma). For endogenous YPEL3 Western, 10% to 20% gradient Tricine gels were used to separate protein extracts. For p53 and ER-alpha Westerns, 12% sodium dodecyl sulfate polyacrylamide gels were used to separate protein extracts. Proteins were transferred to polyvinylidene difluoride using a Bio-Rad transfer system. Antibodies were obtained from Proteintech (YPEL3), Neomarkers (ER alpha, Ab-10), and Calbiochem (p53, Ab-4 and Ab-6).

Lentiviral production

Lentivirus was produced by the cotransfection of 293FT cells with a pLenti vector and lentiviral packaging mix (Invitrogen) according to manufacturer’s instructions. Lentivirus-containing supernatant was harvested at 48 hours posttransfection, purified by centrifugation, and stored at −80°C. Viral transductions were carried out overnight in the presence of 6 μg/mL Polybrene (Sigma). Following transductions, cells were selected with 750 μg/mL Zeocin or 1μg/mL Puromycin. Plasmids were obtained from Addgene (shCON, shp53), Sigma (shERα, shYPEL3(B)) and Open Biosystems (shYPEL3(A)).

Senescence-associated β-galactosidase staining

Cells were processed with the Senescence β-galactosidase Staining kit (Cell Signaling Technology, Millipore) according to the manufacturer’s instructions and were visualized on an Olympus 1 × 70 fluorescence microscope. Experiments were completed in biological triplicate counting at least 100 cells per plate. Error bars represent the standard error of the means.

ViaCell Counting/MTT Assays

MCF-7 cells or MCF-7 cells transduced with lentivirus expressing a shYPEL3 target were plated in 35 mM dishes. At each time point, three plates of cells were separately trypsinized and subjected to ViCell counting following manufacturer’s recommendations. Each time point represents the average of three biological samples with error bars representing the standard deviation. For MTT assays the cells were seeded (1000 cells/well) into black 96 well tissue culture plates. At the indicated time point, 6 wells per cell type were treated with Cell Quant-Blue MTT reagent (BioAssay) for 2 hours following manufacturer’s dilutions. Sample fluorescence was determined using a Tecan Safine² with an excitation wavelength of 530 nm and emission wavelength of 590 nm.

YPEL3 expression increases in ER positive cells grown in the absence of estrogen

We initially grew MCF-7 (ER+, p53 wt) cells in charcoal stripped serum (CSS) to determine if estrogen was playing a regulatory role in YPEL3 mRNA expression. MCF-7 cells grown in CSS containing media showed a 4.8-fold increase in relative YPEL3 mRNA expression when compared to cells grown in media with 10% fetal bovine serum (Figure 1A). Similarly, we saw an increase in YPEL3 protein in MCF-7 cells grown in CSS (Figure 1B). The induction of YPEL3 was also observed in ZR-75.1 (ER+, p53 wt) and T-47D (ER+, p53 mutant) when these cells were grown in charcoal stripped serum (Supplemental Figure 1). To determine if the elevation in YPEL3 mRNA expression was a result of interactions with the estrogen receptor, we performed a similar experiment using SkBr-3 cells (ER-, p53 mutant). In these cells, growth in charcoal stripped serum did not modulate YPEL3 mRNA expression (Figure 1C).

Estrogen represses YPEL3 expression of endogenous gene

Charcoal:dextran stripping reduces the serum concentration of many hormones and certain growth factors, such as estradiol, cortisol, corticosterone, the B vitamins, T3, T4 and prostaglandins. To determine if the elevation of YPEL3 mRNA expression seen with growth in charcoal stripped serum was specific to estrogen, we added back beta-estradiol to the experimental conditions. pS2, also known as TIFF1, was used as a positive control. pS2 expression has previously been shown to be increased in the presence of estrogen, and this effect is regulated by the estrogen receptor ¹⁵. MCF-7 cells were grown in the presence or absence of estrogen. When beta-estradiol was added back to the growth media at a concentration of 1 nM, YPEL3 mRNA expression was repressed to 30% of that seen in charcoal stripped serum (Figure 2A). As expected, the addition of beta-estradiol to the growth media caused a greater than 2-fold elevation in pS2 mRNA expression (Figure 2A). Addition of ten times more beta-estradiol (10 nM) did not result in a statistically significant further reduction in YPEL3 or increase in pS2 mRNA expression suggesting that estrogen signaling was saturated at 1 nM.

As additional support of estrogen’s repressive role on YPEL3 gene expression, we utilized the selective estrogen receptor modulator, tamoxifen. MCF-7 cells were cultured in the absence of estrogen, with the addition of 1 nM beta-estradiol or with the addition of 1 nM beta-estradiol and 1 μM tamoxifen ¹⁶. A 3-fold increase in YPEL3 mRNA expression was seen when MCF-7 cells were cultured in charcoal stripped serum when compared to YPEL3 expression in complete media (Figure 2B). As expected, this increase was blocked by the addition of beta-estradiol. However, with the simultaneous addition of tamoxifen and beta-estradiol, YPEL3 mRNA expression increased by 185% compared to beta-estradiol alone consistent with the model where YPEL3 expression is regulated through the estrogen receptor (Figure 2B). With growth in charcoal stripped serum, pS2 expression was decreased to 55% of that seen in complete media (Figure 2B). The addition of 1 nM beta-estradiol allowed for the level of expression to return back to that seen in complete media (Figure 2B). The addition of tamoxifen caused a 20% reduction in pS2 mRNA expression (Figure 2B), as expected given its inhibitory effects on the estrogen receptor alpha.

Elevation in YPEL3 mRNA expression upon removal of estrogen in ER+ breast cancer cells is p53 independent and ERα dependent

Towards the goal of determining if YPEL3 mRNA induction following estrogen removal in breast cancer cells was dependent on p53 and estrogen receptor alpha, we used lentiviral transduction of shRNAs targeting each gene. shCON, a scramble shRNA vector that does not target any human genes ¹⁷, was used as a short hairpin control. p53 and estrogen receptor alpha knockdown was confirmed by both RNA and Western blot analysis (Figures 3B, 3C, 3D and 3E). In the case of p53, since it is normally expressed at low levels in MCF-7 cells ¹⁸, shp53 knockdown was confirmed following treatment for 24 hours with 0.5 μg/mL doxorubicin (Figure 3C). MCF-7 cells stably selected for the loss of p53 were grown in complete media or charcoal stripped serum. There was a similar elevation (3-fold) in YPEL3 mRNA expression in control cells and shp53 cells when grown in charcoal stripped serum (Figure 3A). Furthermore, p21 expression was monitored to assess whether growth in charcoal stripped serum induced a p53 response. p21 expression was not modulated by growth in charcoal stripped serum for 48 hours in either shCon or shp53 cells (Figure 3A). Taken with the increase in YPEL3 mRNA seen when T-47D cells, which harbor mutant p53, were grown in charcoal stripped serum (Supplementary Figure 1), these findings demonstrate that wild-type p53 is not playing a significant role in the estrogen repression of YPEL3 gene expression.

Next, we tested if estrogen repression of YPEL3 mRNA expression was dependent upon expression of ER alpha. Attempts to create stable MCF-7 cell lines selected for the loss of estrogen receptor alpha using lentiviral shERα were unsuccessful likely due to the growth dependency of estrogen signaling in MCF-7 cells (K. Miller, personal communication). Instead MCF-7 cells were super-infected (MOI=5) with shERα or shCON and grown in either complete media or charcoal stripped serum. MCF-7 cells infected with shCON virus and grown in the absence of estrogen showed a 2.6-fold elevation in YPEL3 mRNA expression (Figure 3B). Interestingly, MCF-7 cells infected with shERα grown in complete media also demonstrated an elevation (3.1-fold) in YPEL3 mRNA expression (Figure 3D). Thus, cells grown in the absence of estrogen or the absence of estrogen receptor alpha both demonstrated an elevation of YPEL3 mRNA expression suggesting it is the cooperation of estrogen with its receptor, alpha subtype, that is allowing for the observed modulation of YPEL3 mRNA expression. Knockdown of estrogen receptor alpha expression in MCF-7-shERα cells was 35% of MCF-7-shCon cells based on relative mRNA expression (Figure 3B), and confirmed at the protein level by western blot (Figure 3E).

Estrogen removal causes YPEL3 dependent cellular senescence

Elevated expression of YPEL3 causes an induction of cellular senescence in both primary and tumor cell lines ¹¹. We questioned whether the elevated YPEL3 expression seen with growth in charcoal stripped serum was correlated with an elevation in cellular senescence. To test this, MCF-7 cells were grown in DMEM media with 10% fetal bovine serum or in DMEM media containing 10% charcoal stripped fetal bovine serum in the presence or absence of 1 nM beta-estradiol. Cells grown in charcoal stripped serum had a significant induction of cellular senescence as measured by SA-beta-galactosidase activity at a pH of 6 ¹⁹ (Figure 4A). Again, induction of senescence after 6 days of treatment in media containing charcoal strip serum was seen in T-47D cells. Here we monitored senescence based on increases in SA-beta-galactosidase positive cells and the induction of p21, an essential senescence activator in breast cancer cells ³ (Supplementary Figure 2). The addition of 1 nM beta-estradiol to MCF-7 cells grown in charcoal stripped serum brought the level of cellular senescence down to 35% beta-galactosidase positive cells (Figure 4B). Increasing doses of beta-estradiol to 100 nM resulted in an inhibition of beta-galactosidase activity back to levels seen in cells grown in complete media (Supplementary Figure 3). To determine if the cellular senescence seen in MCF-7 cells grown in charcoal stripped serum was dependent upon YPEL3, MCF-7 cells were infected with lentivirus shRNA vectors targeting YPEL3 (shYPEL3(A) and shYPEL3(B)) or an shLacZ vector as a short hairpin control. MCF-7 cells stably selected for the loss of YPEL3 grown in charcoal stripped serum did not demonstrate an elevation in cellular senescence. In contrast, shLacZ expressing MCF-7 cells did show the expected induction of cellular senescence when grown in charcoal stripped serum (Figure 4B). Knockdown of YPEL3 expression was confirmed with relative mRNA expression (Figure 4C). Similar findings were seen with T-47D and ZR-75.1 cell lines (Supplementary Figures 4 and 5). These results point to an essential role for YPEL3 in estrogen-dependent cellular senescence in ER+ breast cancer cells.

Loss of YPEL3 triggers increased cell proliferation

Given that induction of YPEL3 triggers a growth suppression in MCF-7 and U2OS cells ¹¹ we next sought to determine the rate of cell growth in MCF-7 cells expressing shYPEL3. When compared to MCF-7 cells, shYPEL3-expressing MCF-7 cells showed a statistically significant increase in cell numbers when grown over 10 days (Figure 5A). Using an MTT based assay, YPEL3 knockdown for 6 days led to an increase in cell numbers while YPEL3 induction over the same period resulted in a decrease in cell number (Figure 5B). These findings suggest that the while YPEL3 induction triggers growth inhibition through cellular senescence ¹¹, decreased YPEL3 expression enhances cell proliferation.

Estrogen regulated senescence seen with removal of estrogen in ER+ breast cancer cells is p53 independent and ERα dependent

While the increase in YPEL3 mRNA expression seen with the removal of estrogen was not dependent upon the expression of p53, we wanted to determine if the elevation of cellular senescence seen with the removal of estrogen was dependent on p53. In previous studies, we have demonstrated that over expression of YPEL3 alone could induce cellular senescence independent of p53 ¹¹. Again, MCF-7 cells stably selected for the loss of p53 or a control siRNA (shCON) were cultured in complete media or charcoal stripped serum for 6 days. At the completion of 6 days, beta-galactosidase activity at a pH of 6 was measured. MCF-7 shp53 cells grown in the absence of estrogen had a similar increase in cellular senescence when compared to control cells (Figure 6A). This corresponds with the resulting increase in YPEL3 mRNA expression (Figure 3A). These results suggest that both the elevation in YPEL3 mRNA expression and the elevation in cellular senescence seen with estrogen removal are independent of p53 expression.

We also tested whether the expression of estrogen receptor alpha was required for the observed induction of cellular senescence. MCF-7 cells were infected at an MOI=5 with shERα and shCON viral constructs. Cells were then grown in the presence or absence of estrogen. Again, MCF-7 cells cultured in the absence of estrogen had a robust increase in beta-galactosidase activity. Similar to the mRNA expression (Figure 3C), MCF-7 cells transduced with lentivirus expressing shERα and grown in the presence of estrogen had an elevation of cellular senescence, although not to the level seen in control cells grown in the absence of estrogen (Figure 6B). The incomplete response is likely due to the incomplete knockdown of estrogen receptor alpha resulting from our inability to create stable pools of shERα expressing MCF-7 cells (Figure 3D).

Tamoxifen treatment of ER+ breast cancer cells causes induction of YPEL3-dependent cellular senescence

Tamoxifen treatment of ER+ breast tumors has previously been shown to induce a G0/G1 phase cell cycle growth arrest ⁴. MCF-7 cells were exposed to tamoxifen treatment to assess if YPEL3 expression was elevated under these conditions, and further, if YPEL3 was required for the growth arrest seen with tamoxifen treatment. MCF-7 cells stably selected for the loss of YPEL3 (shYPEL3(A) and shYPEL3(B)) or control (shLacZ) were grown in complete media and treated with 0.5 μM tamoxifen or vehicle control (100% ethanol) for 6 days. There was a 2.3- fold induction of YPEL3 mRNA expression in control cells (shLacZ) treated with tamoxifen verses vehicle control (Figure 6C). Similarly, there was a 35% induction of cellular senescence in tamoxifen-treated shLacZ cells that was not observed in either pool of MCF-7 cells expressing shYPEL3 (Figure 6D). Interestingly, while there was some induction of YPEL3 mRNA expression in the shYPEL3 cell lines with tamoxifen treatment, levels did not exceed those observed in shLacZ cells grown in complete media and thus were insufficient to induce cellular senescence (Figures 6C and 6D).