Targeting mutant p53 by a SIRT1 activator YK-3-237 inhibits the proliferation of triple-negative breast cancer cells.

Many types of mutations in tumor suppressor p53 are oncogenic through gain-of-function. Therefore, targeting mutant p53 (mtp53) is a promising therapeutic approach to fight against many types of cancers. We report here a small molecule compound YK-3-237 that reduces acetylation of mtp53 and exhibits anti-proliferative effects toward triple-negative breast cancer (TNBC) cells carrying mtp53. YK-3-237 activates SIRT1 enzyme activities in vitro and deacetylation of both mtp53 in a SIRT1-dependent manner. Deacetylation of mtp53 resulted in depletion of mtp53 protein level and up-regulated the expression of WTp53-target genes, PUMA and NOXA. YK-3-237 also induces PARP-dependent apoptotic cell death and arrests the cell cycle at G2/M phase of mtp53 TNBC cells. Taken together, our data suggest that targeting acetylation of mtp53 is a potential target to treat human cancers.


INTRODUCTION
TNBCs are immunohistologically characterized by lack of estrogen receptor (ER) and progesterone receptor (PR) as well as human epidermal growth factor receptor 2 (HER2) amplification, and ~20% of all breast cancers are identified as TNBCs [1]. TNBC tumors are a group of heterogeneous cancers and substantial overlapped with basal-like breast cancer [1]. Although early TNBCs have higher response rates to neoadjuvant chemotherapy, patients with advanced disease show poor response and rapid progression of disease than patients with other types of breast cancers [2]. Currently, there are no therapies that target TNBCs [2,3].
Breast tumors with p53 mutations were mostly classified into basal-like or HER2-amplified subgroup, while luminal subgroup of breast cancers almost exclusively expresses WTp53 [4]. Tumor suppressor protein p53 is a cellular guardian and mutations of p53 are found ~50% of human cancers [5][6][7][8]. In fact, p53 is originally described as a tumor protein (or oncoprotein) and subsequent investigations have revealed that WTp53 has roles as a tumor suppressor and protects cells from malignant transformation [5][6][7][8]. In general, mutation of p53 results in stabilization and accumulation of mtp53 proteins in cells, while WTp53 is tightly regulated by post-translational modification coupled with proteasomal degradation. First reports appeared to support that mutation of p53 simply causes a loss of tumor suppressor function. Recent advances, however, have demonstrated that mtp53 possess pro-oncogenic potential through gain of functions such as transcriptional repression and activation, coaggregation with other tumor suppressors, and so on [5][6][7][8]. Therefore, targeting mtp53 pathways may represent a promising approach in the development of novel anti-cancer agents [4,6]. Because mtp53 has many types of mutations and functions, distinct strategies will be required for therapeutic targeting [6].
Silent information regulator 2 (Sir2) proteins, or surtuins (SIRTs), are evolutionally conserved enzyme family acting as protein deacetylases/ADP ribosyltransferases [9][10][11][12]. At least 7 enzymes of this class have been identified in humans (SIRT1-7). Among these proteins, class I sirtuins (SIRT1-3) and SIRT5 showed deacetylase activity [9][10][11][12]. It has been reported that SIRT1, SIRT2, SIRT3, SIRT4 and SIRT6 appear to act as tumor suppressors in certain conditions [13][14][15][16][17]. Since varied reports on tumor suppressor function of SIRT1 have been reported [18,19], SIRT1 activators could be useful for treatment or prevention of certain type of cancers [9][10][11][12]. In addition to resveratrol, a series of compounds are reported as potent SIRT1 activators [20][21][22][23][24]. More recently, years-long debates on the activation of SIRT1 by these compounds [25][26][27][28] were resolved by two 237. B, relative sensitivities of breast cancer cell lines to YK-3-237. Breast cancer cells were treated with increasing amount of YK-3-237 for ~72 hr and the viable cells were measured by MTT assay. Relative sensitivities were calculated as described in Materials and Methods. The EC 50 value of each cell line was plotted in Sup Figure 1. Data are presented as mean ± SEM from two independent experiments performed in triplicate. C, western blot analysis of breast cancer cell lines. Cell lysates from each cell line were analysed by indicated antibodies. D, effect of YK-3-237 on the level of mtp53. Cells were treated with 1 µM of YK-3-237 for 24 hr and western blot analysis was performed with indicated antibodies. E, YK-3-237 reduces acetyl-mtp53. Cells were treated with increasing concentrations of YK-3-237 for 24 hr and cell lysates were subjected to western blot analysis with indicated antibodies. C~E, β-actin was used as a loading control. independent groups [29,30].

YK-3-237 inhibits the proliferation of TNBC cells
Previously it has been reported that YK-3-237 ( Figure 1A) exhibited potent anti-proliferative activity toward a broad range of NCI cancer cell lines with unknown mechanism [31]. To further determine the anti-proliferative effects of YK-3-237, we performed the cell viability assay with a panel of breast cancer cell lines. Cells were treated with increasing concentration of YK-3-237 up to 72 hr and viable cells were measured by MTT assay. Notably, YK-3-237 exhibited the antiproliferative activities toward most of the breast cancer cell lines tested at submicromolar concentration ( Table  1 and Sup Figure 1). As shown in Figure 1B, YK-3-237 more preferentially inhibited the proliferation of breast cancer cell lines carrying mtp53. As previously reported [4], most of TNBC cell lines in this study are expressing mtp53 (Table 1). Western blot analysis showed that the levels of p53 protein (data not shown) are highly elevated in TNBC cell lines carrying mutations of p53 gene ( Figure  1C). Although cells with WTp53 such as MCF7 and ZR-75-1 expressed detectable levels of p53 protein, the levels of mtp53 protein are much higher than that of WTp53. As expected, expression of ERα was not detected in TNBC cell lines (Sup Figure 2). Notably, no significant difference in the level of SIRT1 protein was observed in our breast cancer cell line panel (Sup Figure 2). To determine the effect of YK-3-237 on the level of mtp53, western blot analysis was further performed with cell lysates from TNBC cells treated with 1 µM of YK-3-237 for 24 hr. We found that YK-3-237 (1 µM) reduced the level of mtp53 protein in all TNBC cell lines tested after 24 hr treatment ( Figure 1D).

YK-3-237 deacetylates mtp53 in TNBC cell lines
Previously it has been reported that the stability of WTp53 is post-translationally regulated by acetylation at K382 residue [33]. Recently, mtp53 has also been reported to be regulated by acetylation [34]. Based on these findings, we further analysed the acetylation status of mtp53 in TNBC cell lines treated with YK-3-237 by western blot analysis. Twenty four hour treatment of YK-3-237 reduced both the acetylation of K382 and the level of mtp53 in a dose-dependent manner in mtp53 TNBC cell lines ( Figure 1E). We observed that treatment of YK-3-237 had little or no significant effect on the level of SIRT1, one of the deacetylases for p53 [35,36], in mtp53 TNBC cell lines upto 10 µM (Sup Figure 3A). The deacetylation of mtp53 was observed as early as 4 hr after treatment of YK-3-237 without significant reduction in mtp53 level (Sup Figure 3B).
Since SIRT1 is a well-known deacetylase for p53 on K382 residue, we further addressed whether YK-3-237 affects SIRT enzyme activity by in vitro SIRT assay with a fluorophore-conjugated peptide substrate. As shown in Figure 2A, YK-3-237 activated SIRT1 enzyme activities in a dose-dependent manner. Under this condition, a SIRT1/2 inhibitor suramin [37] antagonized YK-3-237mediated SIRT1 activation. Interestingly YK-3-237 was more potent to activate SIRT1 activity than resveratrol and maximal activation was observed at 10 µM (Sup Figure 4A). Moreover, YK-3-237 effectively reduced the survival of SUM149PT cells as compared resveratrol in a long-term survival assay (Sup Figure 4B). YK-3-237 also activated the SIRT2 enzyme in vitro and enhanced the deacetylation of α-tubulin (K40) in HS578T cells (Sup Figure 4C and D).
To exclude potential artifacts from the in vitro assay, we further confirmed SIRT1 activation by YK-3-237 using a p53-Luc reporter gene assay ( Figure 2B-C). Since acetylation of WTp53 has been known to induce its stabilization and transcriptional activation [33], we used WTp53 to monitor deacetylase activity of SIRT1 by YK-3-237. MCF7 cells were transfected with p53-Luc reporter plasmid and WTp53 expression vector in the presence or absence of SIRT1 or dominant negative (DN)-SIRT1 expression vector [38] and further treated with YK-3-237. As expected, YK-3-237 repressed the WTp53-mediated SIRT1 enzyme in vitro. SIRT1 enzyme assay was performed as described in the Materials and Methods. Data are presented as mean ± SD. *** P < 0.001. B, the effect of YK-3-237 on the SIRT1-mediated repression of WTp53 transcriptional activity. MCF7 cells were transfected with indicated plasmid DNAs and further treated with YK-3-237 for 24 hr. Luciferase reporter activity was measured as described in the Materials and Methods. Data are presented as mean ± SD. **P < 0.01 and ***P < 0.001. C, SIRT1-KD reverses the YK-3-237-mediated repression of WTp53 transcriptional activity. MCF7 cells were transfected with siRNA and p53-Luc and further treated with YK-3-237 for 24 hr. Relative luciferase activity was determined as in B. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, and *** P < 0.001. D, suramin reduces the YK-3-237-mediated deacetylation of mtp53. HS578T cells, pretreated with suramin for 1 hr, were further treated with YK-3-237 for 23 hr. Western blot analysis was performed as indicated. β-actin was used as a loading control. E, the effect of YK-3-237 on the SIRT1-mediated deacetylation of mtp53. SUM149PT cells were transfected with either control (C)-or SIRT1 (S1)-siRNA for 3 days, re-seeded and further treated with YK-3-237 for 4 hr. Western blot analysis was conducted with indicated antibodies. β-actin was used as a loading control. activation of p53-Luc reporter gene in a dose-dependent manner ( Figure 2B, lanes 1~4). Under this condition, co-expression of SIRT1 reduced the p53-Luc reporter activity induced by WTp53 and co-treatment of YK-3-237 further enhanced the SIRT1-mediated repression of p53-Luc reporter ( Figure 2B, lanes 5~7). On the contrary, DN-SIRT1 did not repress the p53-mediated reporter gene activation. In addition, repression of the p53-Luc reporter by YK-3-237 was limited in the presence of DN-SIRT1 ( Figure 2B, lanes 9 and 10).
The effect of SIRT1 knockdown (SIRT1-KD) on the WTp53-mediated transcription was also tested in MCF7 cells. MCF7 cells were transfected by either control-or SIRT1-siRNA followed by transfection of p53-Luc and further treated with increasing amounts (1~10 µM) of YK-3-237. Under these conditions, SIRT1-KD slightly induced p53-Luc reporter activity ( Figure 2C). While YK-3-237 reduced the p53-Luc reporter activity in control-siRNA-transfected cells, SIRT1-KD antagonized YK-3-237-mediated repression of p53-Luc activity. These results suggest that the repression of WTp53 transcriptional activity by YK-3-237 is, at least in part, dependent on the presence of functional SIRT1.
SIRT1-dependent deacetylation of mtp53 was further investigated in two mtp53 cell lines, HS578T and SUM149PT. First, HS578T cells were pre-treated with 10 µM of suramin for 1 hr and further treated with 1 µM of YK-3-237 for 23 hr. As shown in Figure 2D, pre-treatment of suramin reversed both deacetylation of mtp53 (K382) and reduction of mtp53 protein. Second, the effect of SIRT1-KD on the YK-3-237 activity was determined in SUM149PT cells. After transfection of either controlor SIRT1-siRNA, the cells were further treated with increasing amount of YK-3-237 for 4 hr. Under this condition, knockdown of SIRT1 increased acetyl-mtp53 in SUM149PT cells ( Figure 2E). In addition, knockdown of SIRT1 reversed deacetylation of mtp53 by YK-3-237 ( Figure 2E). Taken together, these results suggest that YK-3-237 activates the SIRT1 enzyme activity in vitro and deacetylates both wild type and mutant p53 in cells in a SIRT1-dependent manner.

YK-3-237 induces apoptotic cell death in TNBC cells
It has been well established that induction of WTp53 target genes arrests cells to G2 phase of cell cycle and finally leads cells to apoptotic cell death [43]. In fact, YK-3-237 induced PARP cleavage, a hallmark of caspasedependent apoptosis, within 24 hr post-treatment ( Figure   4A). TNBC cells treated with YK-3-237 were markedly arrested at G2/M phase ( Figure 4B). In addition, sub-G1 fractions were also drastically increased by YK-3-237 within 24 hr post-treatment in TNBC cells with mtp53 ( Figure 4B).

DISCUSSION
In this study we found that the small molecule YK-3-237 reduced the proliferation of a broad range of breast cancer cell lines carrying mtp53. YK-3-237 reduced the level of mtp53 by removing acetyl group from its K382 residue in a SIRT1-dependent manner. In addition, YK-3-237 activated in vitro enzyme activity of purified human SIRT1 more potently than resveratrol. In cells carrying WTp53, YK-3-237 enhanced SIRT1-mediated repression of transcriptional activation by WTp53. Similar to mtp53-KD, YK-3-237 induced mRNA expression of PUMA and NOXA and caused apoptotic cell death of TNBC cells.
Since more than 50% of human cancers possess mutations of p53, mtp53 represents a potentially important anti-cancer target [4,6]. More importantly, the active oncogenic potential by gain-of-function of mtp53s stresses this approach as a broad range therapeutic opportunity. Many attempts, however, have failed because of the diverse character of mtp53 [6]. Molecules that restore the wild-type conformation of mtp53 such as PRIMA-1 [44], MIRA-1 [45], CP-31398 [46], and P53R3 [47] have effects on specific types of mtp53. More recently, a HSP90 inhibitor 17AAG has been suggested as an alternative approach to reduce the proliferation of a broad range cancer cell lines carrying various types of mtp53 [48]. In this study, inhibiting HSP90 by 17AAG disrupted the molecular chaperone complex to liberate mtp53 for degradation by MDM2.
The mechanism of hyperstabilization of mtp53 is largely unknown. The stabilization of p53 proteins are thought to be regulated by post-translational modifications. For WTp53, acetylation of K382 residue has been proposed as an important determinant to stabilize the WTp53 protein [33]. Although mtp53 is known to be acetylated [34], the roles of these acetylations of mtp53 still need to further investigation. We demonstrated in this study that mtp53 is acetylated at K382 residue in TNBC cell lines and a small molecule compound YK-3-237 reduced this acetylation of mtp53 and depleted the level of mtp53. In addition, YK-3-237 activated SIRT1 enzyme activity in vitro and the deacetylation of either WTp53 or mtp53 was evidently dependent on SIRT1 activity, which was demonstrated by overexpression of wild type or DN-SIRT1, SIRT1-KD, and SIRT1 inhibition by a small molecule inhibitor, regardless of mtp53 types. However, the molecular mechanism of SIRT1 activation by YK-3-237 needs to be further elucidated. As reported recently [29], YK-3-237 could activate in vitro SIRT1 activity in the presence of aminomethylcoumarin (AMC)-tagged peptide derived from p53 protein. Contrarily, YK-3-237 could not activate in vitro SIRT1 activity toward nascent p53 peptide (data not shown). Allosteric SIRT1 activator such as STAC-1 has been known to require specific hydrophobic motifs found in a subset of SIRT1 substrates including PGC-1α and FOXO3a to facilitate SIRT1 activation [29]. To our knowledge, p53 has no such motifs. Notably, SRT1720 was reported to reduce acetyl-p53 in cells [22], in spite of it could not activate in vitro SIRT1 activity toward nascent p53 peptide [28]. How YK-3-237 activates SIRT1 to remove acetyl group from mtp53 in cells needs to be determined in the future.
As we demonstrated, YK-3-237 also activated the in vitro enzyme activity of human SIRT2. Additionally, YK-3-237 reduced the acetyl-α-tubulin (K40) and the level of α-tubulin in cells. Because YK-3-237 was determined not to bind tubulin binding sites in vitro [31], it is less plausible that YK-3-237 exerts its anti-proliferative effect through inhibition of tubulin polymerization. Rather decrease of α-tubulin through deacetylation by SIRT2 may contribute G2/M arrest in a cooperative manner with depletion of mtp53.
In summary, we demonstrated YK-3-237 is a small molecule activator of SIRT1. Activation of SIRT1 by YK-3-237 functionally reduces the level of mtp53 by deacetylation. The roles of SIRT1 in human cancers are controversial [18,19] and both SIRT1 inhibitors and activators (such as resveratrol) have been reported to inhibit growth of human cancer cell lines [19]. Although several SIRT1 activating compounds have been recently developed [20][21][22][23], the anti-cancer effects of these compounds have not been reported. Because SIRT1 has a deacetylase activity against broad range of substrates acting as either tumor promoters or tumor suppressors [18,19], activating its enzyme activity may be beneficial for certain subtypes of cancers. Given increasing interests of SIRT1 in many biological processes including cancer, aging, diabetes, and neuronal diseases, YK-3-237 may represent a valuable tool for basic research and pharmaceutical development.

MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide) assay
Cell viability was measured by MTT assay as described previously [49]. In brief, cells were counted by the Luna Automated Cell Counter (Logos Biosystems, Gyunggi-Do, Korea) and subcultured in 96-well plates. The day after subculture, cells were treated with increasing amount of YK-3-237 and further incubated for ~72 hr in triplicate. To measure the viable cells, 20 µl of 5 mg/ml MTT solution was added per 100 µl of growth media. After 2-4 hr incubation at 37°C, the media were removed and 150 µl/well of DMSO was added to dissolve the formazan. The absorbance was measured by ELx808 microplate reader (BioTek, Winooski, VT). Viable cells are presented as a percent of the control. The half maximal inhibitory concentration (EC 50 ) was calculated by CompuSyn software V1.0 (ComboSyn, Paramus, NJ). The relative sensitivity was calculated by equation of -Log 2 [(EC 50 of each cell line)/(EC 50 of T47D)].

In vitro SIRT assay
The Fluor-de-Lys-SIRT1 and SIRT2 deacetylase assay were performed as manufacturer's instruction (Enzo Life Sciences, Farmingdale, NY). For competition assay, 50 µM of suramin was added to reaction.

p53 reporter gene assay
Plasmid DNAs for reporter gene assay were obtained from following sources: SIRT1 and DN-SIRT1 from Addgene (Cambridge, MA); Expression vector for WTp53 from Dr. B. Vogelstein; and p53-Luc from Promega (Madison, WI). MCF7 cells were transfected with p53-Luc reporter plasmid with expression vector for WTp53 in the presence or absence of SIRT1 expression vector by Lipofectamine 2000 (Invitrogen). Twenty-four hour after transfection, the cells were further treated with YK-3-237 for 24 hr and luciferase activity was measured according to manufacturer's instruction (Promega) using Victor2 plate reader (PerkinElmer, Waltham, MA, USA) at the Genomics and Epigenomics Shared Resource of Georgetown University Medical Center and normalized to β-galactosidase activities.

Cell cycle analysis
Cells were treated with either DMSO or 1 µM of YK-3-237 for 24 hr. The attached cells were harvested by trypsinization and combined with floating cells in the culture media. After washing by phosphatebuffered saline, the cells were fixed with 70% ethanol at -20˚C. Flow cytometric analysis was performed with a FACSCalibur flow cytometer (Becton-Dickinson, CA) at the Flow Cytometry and Cell Sorting Shared Resource at Georgetown University Medical Center.

Statistical analysis
To compare two groups of interest, the two-tailed Student's t-test was applied for statistical analysis. * indicates P < 0.05; ** indicates P < 0.01; and *** indicates P < 0.001.

Conflict of Interest
Patent applications have been filed by Georgetown University on the behalf of the inventors that are listed as authors in this work.