This study was designed to define the differential mechanism of anti-proliferative response of Sirt1 inhibition in human PCa cells differing in p53 status. Mutations in the tumor suppressor p53 gene are known to occur in several cancer types including PCa. Earlier studies have suggested that p53 mutations occur as a late event during the progression of PCa and were associated with androgen-independence, increased angiogenesis, metastasis, recurrence, and a worse prognosis (19
). Interestingly, recently p53 mutations have been reported to occur in approximately one third of early stage PCa (19
). Further, p53 has been shown to be a downstream target of Sirt1 (14
). Vaziri et al
. found that Sirt1 bound to and decetylated the p53 protein with a specificity for its C-terminal Lys382, reducing the transcriptional activity of p53 (16
). Furthermore, Luo and colleagues reported that mammalian Sir2α physically interacted with and attenuated p53-mediated function. In addition, Sir2α was found to repress p53-dependent apoptosis in response to DNA damage and oxidative stress whereas expression of a Sir2α point mutant increased the sensitivity of cells to the stress response (15
). While it is generally accepted that Sirt1 does indeed interact and deacetylate p53, the biological outcome of this regulation is under debate. A number of studies have shown that Sirt1 does not affect many of the p53-mediated biological activities despite the fact that acetylated p53 has shown to be a target of Sirt1. For example, Kamel et al.
found that while Sirt1 does interact with p53, the Sirt1 protein had little effect on p53-dependent transcription of transfected or endogenous genes and did not affect the sensitivity of thymocytes and splenocytes to radiation induced apoptosis (20
). Solomon and colleagues reported that treatment of cells with EX-527 (a specific small-molecule inhibitor of Sirt1) dramatically increased the acetylation of p53 at lysine 382 following DNA damage, but had no effect on cell growth, viability, or p53-controlled gene expression in cells treated with etoposide (21
). Conversely, Cheng et al
. reported that Sirt1-deficient cells from mice with Sirt1 gene-targeted mutations, exhibited p53 hyperacetylation following DNA damage and increased ionizing radiation-induced thymocyte apoptosis (22
). Another study showed that inhibition of Sirt1 expression with microRNA 34a (miR-34a) resulted in an increase in acetylated p53 and ultimately an increase in apoptosis in colon cancer cells with wild type p53, but not in colon cancer cells lacking p53 (23
). In addition, Stunkel and colleagues found that Sirt1 mRNA knockdown lead to a p53-independent decrease of cell proliferation and induction of apoptosis (24
). These studies suggest that Sirt1 plays a role in the p53 pathway, albeit possibly in a cell-type specific fashion.
In this study, to dissect the role of p53 in Sirt1 inhibition-mediated responses in PCa cells, we employed two isogenic PCa cell lines differing in p53 status viz
. PC3 and PC3-p53 cells. PC3 cells are null for both p53 as well as the androgen receptor (AR) whereas PC3-p53 cells (kind gift from Dr. Munna L. Agarwal at Case Western Reserve University) are PC3 cells stably transfected with wild-type p53 (25
). Thus, the only difference between these two cell lines is their p53 status. We confirmed the status of p53 in these cell lines. As shown by the Western blot analysis, PC3 cells lacked p53 protein whereas PC3-p53 cells were found to have a significant level of p53 protein (data not shown).
In this study, we first determined if p53 status had any effect on Sirt1 inhibition mediated anti-proliferative effects in PC3-p53 cells versus PC3 cells. As shown in and , treatment of cells with the small molecule Sirt1 inhibitor sirtinol (30 and 120 μM; for 24 hours) resulted in a decrease in cell growth and cell viability of both cell types compared to the vehicle (DMSO) control as assessed by Trypan blue assay. Thus, we did not observe a differential response in cell growth and viability in these cells differing in p53 status. We next determined the effect of sirtinol on FoxO1, since we and others have shown that Sirt1 can regulate mammalian FoxO transcription factors through direct binding and/or deacetylation. Our data demonstrated that sirtinol-mediated inhibition of Sirt1 caused an increase in the protein levels of nuclear and cytoplasmic acetylated FoxO1 while having no effect on total FoxO1 protein as shown by the Western blot analysis (). Further, the observed increase in acetylated FoxO1 protein was found to be accompanied by an increase in FoxO1 transcription as determined by a luciferase reporter assay (). Thus our data showed that Sirt1 inhibition-mediated activation of FoxO1 occurs in both PC3 and PC3-p53 cells, irrespective of their p53 status.
Effects of sirtinol treatment on PC3 and PC3-p53 cells
Because sirtinol, at higher concentrations, has been reported to inhibit other sirtuin proteins (i.e. Sirt2), we utilized Sirt1 specific short hairpin RNA (shRNA) to knockdown Sirt1 gene to determine response of Sirt1 specific inhibition in PC3 and PC3-p53 cells. We found that Sirt1 shRNA-mediated inhibition of Sirt1 resulted in a decrease in growth and viability of both the cell types ((3
) and data not shown
). In addition, we also found that shRNA-mediated inhibition of Sirt1 caused a decrease in the clonogenic survival of both PC3-p53 and PC3 cells, irrespective of p53 association ().
Effect of shRNA mediated Sirt1 inhibition on clonogenic survival of PC3-p53 and PC3 cells
Our next aim was to determine the reason of decreased cell growth by Sirt1 inhibition in these isogenic PCa cell lines. Studies have shown that Sirt1 regulates senescence, a permanent state of cell growth arrest after a limited number of cell divisions (8
). Specifically, Huang et al.
demonstrated that enforced Sirt1 expression promoted cell proliferation and antagonized cellular senescence in human diploid fibroblasts, possibly via the activation of ERK/S6K1 signaling (27
). Senescence is associated with a number of specific cellular alterations such as growth inhibition, morphological changes (an increase in cell size and granularity), and the appearance of senescence-associated beta-galactosidase (SA-β-gal) activity. We determined if the observed anti-proliferative response of Sirt1 inhibition was associated with a senescence induction in PC3-p53 and PC3 cells. We found that shRNA-mediated Sirt1 knockdown resulted in a marked increase in SA-β-gal staining () in PC3-p53 cells suggestive of senescence induction. Interestingly, Sirt1 knockdown did not cause an increase in SA-β-gal stain in PC3 cells which lack p53. The induction of senescence was further confirmed by an increase in both forward-angle light scatter (FSC), roughly equivalent to particle size, and side-angle light scatter (SSC), roughly equivalent to granularity (), again only in PC3-p53 cells. The induction of senescence was also confirmed by concurrently using doxorubicin, a positive control for senescence induction (28
) (data not shown).
Effect of shRNA mediated Sirt1 inhibition on senescence in PC3-p53 and PC3 cells
The observed lack of senescence by Sirt1 knockdown in PC3 cells prompted us to question other possible modes of cell growth inhibition. Since apoptosis, another process of inhibiting cell growth, has also been linked to the functionality of Sirt1 (4
), we assessed if shRNA-mediated knockdown of Sirt1 resulted in apoptosis in PC3-p53 and PC3 cells. For this purpose, we determined the extent of poly(ADP-ribose) polymerase (PARP) cleavage, an established indicator of apoptosis in mammalian cells (29
). As shown by Western blot analysis, we found that shRNA-mediated knockdown of Sirt1 resulted in a marked decrease in full length PARP accompanied with an increase in the cleaved PARP in PC3 cells (which lack p53), but not in PC3-p53 cells (). The differential induction of apoptosis in PC3 cells was further confirmed by a terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay that detects DNA fragmentation by labeling the terminal end of nucleic acids () (31
Effect of shRNA mediated Sirt1 inhibition on apoptosis in PC3-p53 and PC3 cells
In summary, our data demonstrated that Sirt1 inhibition resulted in a decrease in cell growth and viability, a decrease in colony formation, an increase in acetylated FoxO1 protein levels as well as an increase in subsequent FoxO1 transcriptional activity in PCa cells regardless of p53 status. The most interesting and novel finding of this study is that Sirt1 inhibition resulted in an increase in senescence (but not apoptosis) in PC3-p53 which possess wild-type p53. On the other hand, Sirt1 inhibition resulted in an induction of apoptosis (but not senescence) in PC3 cells which lack p53. In addition, the results presented here further support the pro-proliferative action of Sirt1 in PCa cells.
Based on our data, it is tempting to suggest that Sirt1 inhibition may have different downstream targets in cells with active p53 versus cells where p53 is inactive (mutated or deleted). In cells with active p53, two scenarios are possible. First, Sirt1 inhibition may inhibit p53 binding and deacetylation either directly or via inhibition of p300/CBP binding and deacetylation thereby activating p53 mediated responses (such as induction of p21/Waf1) leading to cell cycle arrest or apoptosis or senescence in cancer cells. Second, Sirt1 inhibition may inhibit FoxO binding and deacetylation either directly or via inhibition of p300/CBP binding and deacetylation thereby activating FoxO mediated responses (such activation of caspases) leading to apoptosis of cancer cells. In cells with active p53 both events may happen; whereas, in cells without active p53, only the second option is possible. These possible scenarios are depicted in . However, further in-depth studies are needed to dissect the molecular mechanism(s) of the observed differential responses of Sirt1 inhibition in PCa cells, especially in connection with p53.
Proposed model of Sirt1 inhibition-mediated responses in human PCa cells differing in p53 status