SIRT1 enhances matrix metalloproteinase-2 expression and tumor cell invasion in prostate cancer cells

Abstract

Background: Matrix metalloproteinase-2 (MMP2) has been shown to play an important role in cancer cell invasion and the expression of MMP2 is associated with the poor prognosis of prostate cancer; however, the mechanism of MMP2 expression is largely unknown. SIRT1 is a nicotinamide adenine dinucleotide-dependent histone deacetylase (class III HDAC) that has recently been shown to have implications in regulating cancer cell growth and apoptosis. The purpose of this study is to determine the role of SIRT1 in regulating MMP2 expression and tumor invasion in prostate cancer cells.

Methods: The interfering RNAi was used to knockdown SIRT1 from prostate cancer cells. Immunoblots, RT-PCR, zymographic assays, co-immunoprecipitation, analysis and transwell assays were used to examine the effects of SIRT1 silencing on MMP2 expression and activity, on SIRT1 and MMP2 interaction, and on prostate cancer cell invasion. The immuno-histochemical assay was performed to study SIRT1 expression in prostate cancer tissues.

Results: We show that SIRT1 associates and deacetylates MMP2 and SIRT1 regulates MMP2 expression by controlling MMP2 protein stability through the proteosomal pathway. Thus, we demonstrated a novel mechanism in that MMP2 expression can be regulated at the posttranslational level by SIRT1. Furthermore, we determined that SIRT1 inhibition reduced prostate cancer cell invasion and SIRT1 is highly expressed in advanced prostate cancer tissues.

Conclusions: SIRT1 is an important regulator of MMP2 expression, activity, and prostate cancer cell invasion. Overexpressed SIRT1 in advanced prostate cancer may play an important role in prostate cancer progression.

Figure 1

SIRT1 inhibition down-regulates MMP2 expression without change MMP2 mRNA abundance. A. SIRT1 inhibition decreases MMP2 expression. Prostate cancer cell lines PC3 and LNCaP were treated with SIRT1 inhibitor Sirtinol (50uM) for 12hr and cell lysis were separated on an 8% SDS gel and analyzed with anti-SIRT1, anti-MMP2 and anti-β-actin antibodies. B. SIRT1 knockdown decreases MMP2 expression. Equal amounts of cell lysates from LNCaP and PC3 cells with stably SIRT1 knockdown (siSIRT1) or vector control (siVector) were analyzed by immunoblot analysis with antibodies against SIRT1, MMP2 and β-actin. C. SIRT1 knockdown has no effect on MMP2 mRNA abundance. The mRNA levels of MMP2 were measured by quantitative RT-PCR analysis of 5 µg RNA extracted from SIRT1 siRNA (siSIRT1) or control siRNA (siVector) LNCaP and PC3 cells. The mRNA levels of MMP2 are expressed relative to β-actin transcripts. Each experiment was performed in triplicate and repeated three times. The error bars represent the SEM.

Figure 2

SIRT1 regulates MMP2 stability through the ubiquitin-proteasome pathway. A. RNAi vector- (upper panel) or SIRT1 RNAi- (lower panel) transfected LNCaP cells were treated with cycloheximide (10µg/ml). The cell extracts were prepared after treatment with cycloheximide for 0, 1, 2, 4, 7, and 24 hr and equal amounts of cell extracts were separated by SDS gel electrophoresis. The immunoblot was performed with anti-MMP2 and anti-β-actin antibodies. B. The SIRT1 RNAi-transfected LNCaP cells were pretreated with vehicle (upper panel) or MG132 for 1hr (lower panel), then treated with cycloheximide (10µg/ml) for 0, 2, 4, 7, and 24 hr and the immunoblot was performed with anti-MMP2 and anti-β-actin antibodies.

Figure 3

SIRT1 associates with and deacetylases MMP2 and SIRT1 deacetylase activity is required for MMP2 expression regulation. A. Co- immunoprecipitation was performed using anti-SIRT1 antibodies and IgG as a control with LNCaP cell lysates and immunoblot was performed with anti-MMP2 and anti-SIRT1 antibodies respectively. B. Co-immunoprecipitation was performed using anti-MMP2 antibodies and IgG as a control and immunoblot was performed with anti-SIRT1 and anti-MMP2 antibodies. C. The LNCaP cells were treated with or without CoCl2 at 200 uM for 5 hours, then co- immunoprecipitation was performed using SIRT1 antibody and immunoblot was performed with anti-MMP2 and anti-SIRT1. D. Co-immunoprecipitation was performed with anti-MMP2 antibodies or IgG as a control using equal amounts of LNCaP cell lysates from vehicle treated or SIRT1 inhibitor Sirtinol treated cells. Immunoblot was performed with acetylated lysine antibodies to compare the acetylation level of MMP2 in response to SIRT1 inhibitor treatment. The membrane was erased and re-blotted with MMP2 to determine the co-immunoprecipitation efficiency of MMP2 antibody. E. SIRT1 deacetylase activity is required for SIRT1 effects on MMP2 expression. LNCaP cells was transfected with a SIRT1 expression vector (SIRT1), or a dominant-negative SIRT1 vector (H363Y), or an empty vector (Vector). At 48 hr after the transfection the cells were harvested for immunoblot analysis with anti-MMP2 to determine the level of MMP2 expression, and anti-His tag to determine the expression wt SIRT1 and H363Y SIRT1.

Figure 4

SIRT1 inhibition reduces MMP2 activity and invasion activity. A. SIRT1 knockdown decreases zymographic activity of MMP2. PC3 (2×10⁶) cells were seeded in 10 cm plate with 5 ml medium and incubated for 24 hr. Culture supernatant was harvested and diluted 1:1 with non-reducing Zymogram Sample Buffer. 10 ug of protein from culture supernatant was loaded onto 7.5% SDS-polyacrylamide gels containing 2 mg/ml gelatin. The MMP2 zymographic activity was analyzed. B. SIRT1 knockdown decreases prostate cancer cell invasion. 2.5×10⁴ siVector or siSIRT1 PC3 cells in 500ul were loaded onto matrigel in transwells coated with matrigel. After 24 hr of culture at 37°C, the transwells were removed from 24-well plates and stained with Diff-Quick solution. The invading cells were counted for quantity analysis and the representative imaging was taken under a light microscope. Each experiment was performed in triplicate and repeated three times. The error bars represent the SEM. C. MMP2 inhibition impairs SIRT1 inhibition-induced invasion suppression. PC3 cells were pretreated with vehicle or MMP2 inhibitor (MMP2 inhibitor I, 1.7uM, Millipore) for 1 hr, then the cells were treated with SIRT1 inhibitor Sirtinol (100uM) for 24 hr. 2.5×10⁴ cells in 500 ul were loaded onto transwells coated with matrigel. Cell invasion was analyzed as described in Figure 4B.

Figure 5

SIRT1 is highly expressed in advanced prostate cancer tissue. The prostate cancer tissues array from Cybrdi were examined via standard IHC methods (DAB kit, DAB150, Upstate) using SIRT1 antibody (1:2000, Upstate). This figure is the representative of low, medium and high SIRT1 intensity in prostate cancer tissues with Gleason score 4, 6 and 9.