LEF1 in androgen-independent prostate cancer: regulation of androgen receptor expression, prostate cancer growth and invasion

Cell culture, dual luciferase assays, cell proliferation and in vitro Matrigel invasion assays

LNCaP cells were maintained in RPMI 1640 (GIBCO, Gaithersburg, MD) with 10% heat-inactivated bovine serum (FBS). The androgen independent LNCaP-AI cells were maintained in RPMI 1640 medium containing 10% charcoal-stripped, heat-inactivated FBS (CSFBS) (Hyclone Laboratories, Inc., Logan, UT) and 5 μg/ml of insulin, as described previously (38). Dual luciferase assay was performed as previously described (38) with luciferase reporter pAR-Luc (with 1.7 kb AR promoter sequence) and pCMV:LEF1 (Origene, MD). Cell proliferation by cell counting, flow cytometry and Matrigel invasion assays were performed as described previously (41). For invasion assays, percent invasion was expressed as the ratio of invading cells over cell number normalized on day 2 growth curve.

Immunofluorescence microscopy

The paraformaldehyde (4%) fixed LNCaP and LNCaP-AI cells were permeablized in methynol:acetone (1:1) for 10 minutes at -20°C. The cells were incubated overnight at 4°C with anti-AR and anti-LEF1 specific antibodies diluted 1:500 in 5% BSA buffer followed with anti-rabbit and anti-mouse IgG-Cy2-conjugated antibodies (Molecular Probes, Portland, OR) (1:300 in 5% BSA) incubation for 45 min at room temperature. Nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI; 10μg/mL; Molecular Probes, Carlsbad, CA) according to manufacturer’s instructions.

Chromatin immunoprecipitation (ChIP) assays

CHIP was performed as described (42) with the following modifications: cross-linking was initiated with 11% formaldehyde solution at room temperature for 10 min and then stopped by addition of 0.125M glycine. The cross-linked chromatin was sonicated with Sonifer 450 microtip (Branson Ultrasonic Corp., Danbury, CT) at power setting 2 for 10 × 25sec on ice, produced DNA fragments with an average size of 1-3kb. For immunoprecipitation, 2μg of anti-AR antibody, anti-LEF1 antibody and IgG were mixed with 25μg of the purified cross-linked chromatin and incubated overnight at 4°C. Immunocomplexes were washed with RIPA buffer and TE buffer. After reversal and recovery of the immunoprecipitated chromatin DNA, the final DNA pellets were dissolved with 100μl water. Immunopurified DNA (4 μl) was used for a PCR with primers described in Table 1.

Gel shift assay

The assay was performed as previously described (43). Nuclear proteins extracted from LNCaP or LNCaP-AI cells were reacted with ³²P-labeled LEF1 binding sequence oligos corresponding to the promoter of AR gene (25), in binding buffer containing 1 μg of poly (dIdC) for 30 min at room temperature. The reaction mixtures were then subjected to electrophoresis and the binding complexes visualized by exposure on x-ray films. Specificity of the LEF binding complexes were established using specific and non-specific competitions (43), and LEF1 antibody.

Western blot and Immunohistochemistry analysis

Whole cell or cell fraction extracts (42) were subjected to SDS-PAGE and transferred to a nitrocellulose membrane for western blot analysis. Blots were incubated with primary antibodies (AR, LEF1, and β-actin Cell Signaling Technology, Inc., Danvers, MA) for 2h at room temperature, washed with TBST, and incubated for 1.5h with the horseradish peroxidase-conjugated secondary antibody (1:5,000, Amersham Biosciences). The protein bands were detected by an enhanced chemiluminesence kit (Amersham Biosciences). Immunohistochemistry for LEF1 and AR was performed using single label IHC on a NexES automated immunostainer as described previously (44).

High levels of LEF1 expression in androgen-independent disease

In Affymetrix microarray with U133A GeneChip (15), the ratio of LEF1 expression between LNCaP-AI (an androgen-independent variant of LNCaP)(35) and LNCaP was 100-fold. To confirm this dramatic change of LEF1 expression and its celluar localization between LNCaP and LNCaP-AI cells, we performed RT-PCR, Western blot analysis and immunofluorescent microscopy. LEF1 showed sharply increased expression in LNCaP-AI at the mRNA level by RT-PCR (Fig. 1 A, Left, panel 1-2) and at the protein level by Western blot analysis (Fig.1 A, Left, panel 3-4). Cell fractionation revealed that LEF1 is localized to the nucleus in LNCaP-AI cells (Fig.1A, Right, panel 1) with nuclear co-localization of AR (Fig. 1A, Right, panel 2) and Histone 4 (Fig. 1A, Right, panel 4), serving as controls for the nuclear localization and β-actin (Fig. 1A, Right, panel 3) as control for the cytoplasmic localization. Immunofluorescent microscopy confirmed nuclear localization of LEF1 in the majority (98%) of LNCaP-AI cells (Fig. 1 B). In contrast, there is minimal LEF1 expression in LNCaP cells. As a control, AR was localized to the nucleus in both LNCaP and LNCaP-AI cells (Fig. 1B, mid panels). We also examined the levels of LEF1 expression in several prostate cell lines, including benign immortalized RC165 and RC170 (45), PC3 (low or negative AR) and DU145 (negative for AR) cells and did not detect LEF1 expression by Western blot analysis (data not shown).

Since LEF1 is specifically increased in LNCaP-AI prostate cancer cells, we examined the expression of LEF1 in androgen-independent cancer using a tissue microarray consisting of 99 cases, including 24 benign, 56 androgen-dependent, and 19 androgen-independent cases of prostate cancer. Androgen-dependent specimens were derived from patients who were diagnosed with prostate cancer by transurethral resection of the prostate (TURP), having high grade (Gleason 8 or higher) and volume of disease. Androgen-independent samples were derived from patients who underwent TURP at least 6 months after surgical orchiectomy. The immunostaining signal is scored as a combination of intensity (0 as negative, 1+ as weak, 2+ as moderate and 3+ as strong expression) and percentage of positive cells (<10% as 1+ and ≥10% as 2+) as in any given case with highest score as 5. Of 24 benign cases, LEF1 is not expressed in luminal cells in all cases and is only expressed in basal cells in certain cases (Fig. 1C, Left). Lef1 is expressed in 9 of 56 (16%) androgen-dependent cases ranging from 20% to 60% positive cells for a given case. We observed a statistically significant (p<0.01) difference of LEF1 expression between androgen-dependent and -independent cases. LEF1 is expressed (Fig. 1C, Right and Fig. 1D, Left) in 13 of 19 (66%) androgen-independent cases. The expression of combined score is statistically significantly higher in androgen-independent than -dependent cases (p=0.016) (Fig. 1D, Right). The IHC was performed without primary antibody and find no staining above background. The antigen peptide was also able to compete off the IHC signal.

Increased growth and invasion of LNCaP-AI is AR-dependent

AR is expressed in both LNCaP and LNCaP-AI cells, with greater levels in LNCaP-AI cells (Fig. 2A, Left). The increase in cell proliferation among LNCaP-AI cells is prominent in androgen free, though androgen can stimulate its growth (Fig. 2B, Left and right, respectively), validating the androgen-independent nature of LNCaP-AI cells. Flow cytometric analysis of LNCaP-AI cells with reduced levels of AR by siRNA knock down showed a 5.1% of decrease of cells in the S-phase cell cycle. Further, we used Matrigel invasion assays to determine the difference of invasion ability between LNCaP and LNCaP-AI cells, both in androgen medium and androgen-free medium. LNCaP and LNCaP-AI cells showed comparable minimal invasion in androgen free medium (Fig. 2C and 2D, left). While androgen increased the invasion ability of both LNCaP and LNCaP-AI cells, LNCaP-AI cells showed greater increase in invasive ability (up to 4-fold) compared to LNCaP cells (1.5 fold) in presence of 10 nM R1881 (Fig. 2D, Left), suggesting that androgen and AR plays a key role for invasion in LNCaP-AI cells. We employed small RNA interference technology to knock down AR in LNCaP (Fig. 2A, Middle) and LNCaP-AI-cells (Fig. 2A, right) followed by invasion assays to confirm AR-dependent nature of invasion. In the presence of 10 nM R1881, there is a 50% reduction in invasion by LNCaP cells and even greater reduction (up to 3 fold) of invasive LNCaP-AI cells (Fig. 2D, Right). There was no statistical difference in invasiveness of both LNCaP and LNCaP-AI cells with (Fig. 2D, Right) or without (Fig. 2D, Left) AR knockdown in androgen-free medium. These results strongly indicate that increased invasiveness of LNCaP-AI cells is dependent on AR.

LEF1 promotes growth and invasion of LNCaP-AI cells

LEF1 is highly expressed in LNCaP-AI but not LNCaP cells (Fig. 1A, left). To determine whether LEF1 is involved in the increased growth and invasion of LNCaP-AI cells, we performed cell proliferation and invasion assays of LNCaP with LEF1 over-expression and of LNCaP-AI cells with decreased LEF1 expression by siRNA knock down. The levels of LEF1 expression in LNCaP cells transfected with pCMV:LEF1 is comparable with LEF1 in LNCaP-AI cells (data not shown). With LEF1 overexpression (Fig. 3A, Left, upper panel), increases were observed in both LNCaP-LEF1 cell proliferation, up to 4.1% in S-phase cell population by flow cytometry (Fig. 3B, Left) and invasion, up to 2.5-fold by Matrigel invasion assays (Fig. 3B, Right). On the other hand, with reduced levels of LEF1 by siRNA (Fig. 3A, Right), LNCaP-AI cell growth was reduced as indicated by a 6% reduction in S-phase cell population in androgen media (Fig. 3C, Left). Similarly, the invasion ability of LNCaP-AI cells was also reduced 3-fold in androgen media (Fig. 3C, Right). To examine the specificity of the LEF1 knockdown and to guard against off target effects, we have repeated this experiment using two additional LEF1 siRNAs and obtained similar effects in cell growth and invasion assay (data not shown). Importantly, these distinct LEF1 siRNAs also substantially reduce AR expression in LNCaP-AI cells. Taken together, the results show that LEF1 is involved in growth and invasion of LNCaP-AI cells.

LEF1 regulates expression of AR in LNCaP-AI cells

AR is expressed at greater level in LNCaP-AI than LNCaP cells. In parallel, LEF1 is expressed at high levels in LNCaP-AI and at minimal levels in LNCaP cells. We determined the relationship between the expression of LEF1 and AR in LNCaP and LNCaP-AI cells by siRNA knockdown of either AR or LEF1, coupled with Western blot analysis to detect LEF1 and AR expression, respectively. The LNCaP-AI cells with LEF1 knocked down by 10 nM siRNA exhibited a complete block in AR expression, shown by RT-PCR at mRNA levels (Fig. 4A, Left) and Western blot analysis at the protein level (Fig. 4A, Middle). In contrast, knockdown of AR did not affect the levels of LEF1 expression in both LNCaP and LNCaP-AI cells (Fig. 4A, Right), indicating that LEF1 regulates expression of AR in LNCaP-AI cells.

We performed ChIP and gel shift assays to examine the direct occupancy of LEF1 on the eight putative LEF1 binding sites of AR promoter of both LNCaP and LNCaP-AI cells. In LNCaP cells, we first confirmed the promoter occupancy of the three reported binding sites (25), sites 2, 3, 4/5 (Fig. 4B, upper panels), validating our ChIP assay data. In LNCaP-AI cells in the absence of androgen, LEF1 was readily recruited to the sites 2, 3, 4/5 of the AR promoter (Fig. 4B, Lower part). After treatment of LNCaP-AI cells with androgen, LEF1 recruitment increased in both sites and intensity on the promoter of AR (Fig. 4B, Lower panels).

The specific binding of LEF1 on the AR promoter region was confirmed by gel shift assays using both wild type and mutant probes. Data indicate that LEF1 binds to site 2 binding region in LNCaP (Fig. 4C, lane 2) and LNCaP-AI cells (Fig. 4C, lane 3). The specificity of binding was confirmed as the complex does not form in the absence of probe (Fig. 4C, lane 1) and in the presence of mutant probes (5’-CCAGACTCTCGACGGCAAAATC-3’) (Fig. 4C, lane 4-6). Binding at sites 3 and 4/5 were also confirmed (data not shown), and there is no complex binding for site 6 (Fig. 4C, lane 7-9).

To correlate the physical occupancy of LEF1 on the AR promoter with transcriptional activation of AR, we performed dual luciferase assays with a luciferase reporter containing a 6 kb AR promoter region in LNCaP cells. Transcriptional activity was increased up to 2-fold with LEF1 overexpression (Fig. 4D). The results of the above studies strongly support the direct regulation of AR expression by LEF1.