Targeting TCTP as a new therapeutic strategy in castration-resistant prostate cancer

Abstract

Heat shock protein 27 (Hsp27) is highly overexpressed in castration-resistant prostate cancer (CRPC) and an antisense inhibitor (OGX-427) is currently in phase II clinical trials. In order to understand mechanisms of action of Hsp27 and find new therapeutic targets specific of CRPC, we screened for Hsp27 client proteins. Here, we report that translationally controlled tumor protein (TCTP) is a new Hsp27 client protein involved in Hsp27 cytoprotection. We found that TCTP expression is absent or weak in normal prostate cells, moderately expressed in 18.5% of treatment naive PC, and becomes uniformly and strongly expressed in 75% of CRPC. To define TCTP function, we developed and worldwide patented a TCTP antisense oligonucleotide (ASO). Interestingly, we found that CRPC progression correlates with TCTP overexpression and loss of P53. TCTP knockdown restored P53 expression and function, suggesting that castration-sensitivity is directly linked to P53 expression. Collectively, these findings provide a new Hsp27 cytoprotection mechanism in CRPC, and preclinical proof-of-concept that combining ASO-mediated TCTP knockdown with castration and/or docetaxel therapy could serve as a novel strategy to treat CRPC, with no or little toxicity for normal prostate cells.

Figure 1

TCTP protein level increases in castration-resistant (CR) prostate cancer cells and human tumors. (a,b) Proteins from normal (PNT2C2), castration sensitive (LNCaP), androgen independent (PC-3), and CR (C4-2) cells were extracted. TCTP, MSL-1, Hsp27, vinculin, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) protein levels were analyzed by western blot. (c) Expression profile of TCTP, Hsp27, and vinculin were analyzed by western blot in CR C4-2 and CS LNCaP xenograft. (d) Average intensity of bands for TCTP and Hsp27 were normalized with vinculin and scoring with Image J software (**P ≤ 0.01). (e) CS (harvested before castration) and CR (harvested 40 days after castration) LNCaP tumors were used to analyze TCTP, Hsp27, and vinculin protein levels by western blot. (f) Average intensity of TCTP and Hsp27 bands was scored with Image J software after normalization with vinculin (**P ≤ 0.01 and ***P ≤ 0.001). (g) Representative microscopic fields of TCTP immunostaining in prostate benign hyperplasia (BPH) and treatment naive prostate cancer (PC) tissue microarray. Micrographs illustrate that 18.5% of treatment naive PC show TCTP immunoreactivity in tumor glands epithelium (T, arrows). In BPH, the epithelium that borders normal glands (N, arrows) is not immunoreactive for TCTP. We can notice that within prostate tumor, normal gland is not immunoreactive. (h) Representative microscopic fields of TCTP immunostaining in castration-treated PC tissue microarray. (i) Mean TCTP staining after castration therapy (CT). Specimens were graded from 0 to +3 intensity, representing the range from no staining to heavy staining by visual scoring and automated quantitative image analysis by proplus image software. Data from 112 samples were used to calculate (average) mean ±SE. All comparisons of stain intensities were made at ×200 magnification. AI, androgen independent; CS, castration sensitive; Hsp27, heat shock protein 27; MSL-1, male-specific lethal-1 homolog; N, normal; OD, optical density; TCTP, translationally controlled tumor protein.

Figure 2

Hsp27 interaction protects TCTP from the ubiquitin-proteasome degradation. (a) LNCaP_Mock and LNCaP_Hsp27 stained with rabbit polyclonal TCTP (green immunofluorescence, IF) and mouse monoclonal Hsp27 (red immunofluorescence, IF) antibodies. Yellow is for colocalized foci (merged) and green or red represent non-colocalization foci. (b) LNCaP_Mock and LNCaP_Hsp27 cell lysates were used to immunoprecipitate (IP) TCTP using rabbit anti-TCTP or rabbit anti-immunoglobulin (IgG) antibodies. The membrane was then immunoblotted with anti-Hsp27 antibody. Total cell lysate (TCL) represents proteins from LNCaP_Mock versus LNCaP_Hsp27 cells, extracted from cultured cells and blotted as control with anti-TCTP or anti-vinculin antibodies. (c) REG cells, stably transfected with Hsp27 (REG_Hsp27) or empty vector (REG_Mock), were harvested and proteins extracted. TCTP, Hsp27, and vinculin protein levels were analyzed by western blot. (d) Proteins were extracted from culture of LNCaP_Mock, LNCaP_Hsp27, and LNCaP treated with OGX-427 or OGX-control and analyzed by western blot with TCTP, Hsp27, and vinculin antibodies. (e) The graph represents TCTP protein levels, after normalization to vinculin protein levels, for each condition by scoring the band's intensity with Image J software. (f) Total RNAs was extracted from culture of LNCaP_Mock, LNCaP_Hsp27, and LNCaP treated with OGX-427 or OGX-control. TCTP and 18S levels were analyzed by quantitative reverse transcription-PCR (qRT-PCR). TCTP mRNA levels were analyzed after normalization to 18S ribosomal RNA (rRNA) levels. Results are expressed as percentage of LNCaP cells transfected with the OGX-control (100%). (g) Protein lysates from LNCaP_Mock and LNCaP_Hsp27 cells were used to immunoprecipitate (IP) TCTP, followed by western blotting with an anti-ubiquitin (Ub) antibody. (h) PC-3 cells were treated with OGX-427 or OGX-control for 2 days and then harvested for protein extraction or pretreated with cycloheximide (10 µg/ml) followed by MG-132 (10 µmol/l) for 48 hours and tested by western blot analysis using an anti-TCTP, anti-Hsp27, and anti-vinculin antibodies. Hsp27, heat shock protein 27; OD, optical density; TCTP, translationally controlled tumor protein.

Figure 3

Cytoprotection induced by Hsp27 is in part mediated by TCTP and TCTP silencing inhibits prostate cancer cells growth and enhances chemotherapy in vitro. (a) LNCaP_Mock and LNCaP_Hsp27 cells were treated with TCTP- or control-siRNA (5 nmol/l), then proteins were extracted and analyzed by western blot. (b) LNCaP_Mock and LNCaP_Hsp27 cells were treated with 5 nmol/l TCTP- or control-siRNA. After 2 days, serum-free media (mimics androgen withdrawal in vitro) and docetaxel was added for 24 hours and cells were analyzed for growth rates using crystal violet dye. The experiment was repeated in triplicate and statistical analysis was done using Statview software (***P ≤ 0.001; **P ≤ 0.01). (c) PC-3 cells were treated with indicated concentrations of TCTP- or control-siRNA, and 2 days after proteins were extracted and analyzed by western blot. (d) Proteins were extracted from PC-3 cells treated with the indicated concentrations of TCTP- or control-ASO. TCTP and vinculin protein levels were analyzed by western blot. (e) PC-3 cells were treated for 1 day with 5 nmol/l of TCTP- or control-siRNA, and for 2 days with 100 nmol/l of TCTP- or control-ASO. Total RNA was extracted and TCTP levels were analyzed by qRT-PCR. After the normalization of TCTP mRNA with 18S rRNA levels, results were analyzed with the 2(-δδC_T) Method. Each sample was analyzed in triplicate. **Differs from PC-3 transfected with control-siRNA (P ≤ 0.01) and ***with control-ASO (P ≤ 0.001) using Statview software. (f) PC-3 cells were treated with 5 nmol/l of TCTP- or control-siRNA. After 2 days, cell growth rates were analyzed using MTT test (***P ≤ 0.001). (g) Cell viability analysis of PC-3 treated with TCTP-ASO combined with docetaxel using MTT test. The experiment was repeated in triplicate. ***Differs from PC-3 transfected with control-ASO ± docetaxel (P ≤ 0.001) and **differs between PC-3 treated with TCTP-ASO monotherapy and TCTP-ASO plus 50 nmol/l docetaxel (P ≤ 0.01) treatments using Statview software. (h) TCTP-ASO efficiency on C4-2 was tested by western blot after 100 nmol/l of TCTP- or control-ASO treatment for 48 hours, the proteins was extracted. TCTP and vinculin levels were analyzed by western blot. (i) C4-2 cell viability was determined using crystal violet dye. TCTP inhibition decreases C4-2 cell survival after 24 hours treatment with docetaxel in serum-free media (mimics androgen withdrawal in vitro). Error bars represent the SE, ***P ≤ 0.001 by Statview software. (j) C4-2 and LNCaP cells were transiently transfected with pDEST-TCTP or pDEST-Mock vectors for 2 days and proteins were extracted. TCTP and vinculin levels were analyzed by western blot. (k,l) LNCaP and C4-2 cells overexpressing TCTP, increased cell survival after combined androgen withdrawal and docetaxel-chemotherapy treatments. After 48 hours, cell viability was determined using the crystal violet assay. Error bars represent the SE, **P ≤ 0.01 and ***P ≤ 0.001 by Statview software. AI, androgen independent; ASO, antisense oligonucleotide; CR, castration resistant; CS, castration sensitive; Hsp27, heat shock protein 27; MTT, 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl tetrazolium; qRT-PCR, quantitative reverse transcription-PCR; rRNA, ribosomal RNA; siRNA, small interference RNA; TCTP, translationally controlled tumor protein.

Figure 4

TCTP silencing induces cell cycle arrest and apoptosis in prostate cancer cells in vitro. (a-d) Flow cytometry was used to quantify the percentage of PC-3 cells in each cell cycle phase, 2 days after siRNA or ASO treatment. The plot represents the mean of phase S and sub G0-G1 fractions from three independent flow samples. The results are expressed in percentages, PC-3 cells transfected with control-siRNA or -ASO representing 100%. (a,b) Rate of PC-3 cells in phase S was determined after (a) TCTP-siRNA and (b) -ASO treatments. * and **differs from PC-3 transfected with controls (P ≤ 0.05 and P ≤ 0.01, respectively). Nonsignificant (NS) difference was found between the phase S fractions of PC-3 treated with TCTP-ASO alone and TCTP-ASO plus chemotherapy. (c,d) Apoptotic determination of PC-3 cells treated with (c) TCTP-siRNA and (d) -ASO. ** and *** differs from PC-3 transfected with control-siRNA and -ASO alone (P ≤ 0.01 and P ≤ 0.001, respectively). A 40% increase (**P ≤ 0.01) of apoptotic PC-3 treated with TCTP-ASO plus docetaxel was found compared to PC-3 treated with TCTP-ASO monotherapy. (e) Effect of TCTP silencing on caspase-3 cleavage and activity. PC-3 cells, 24 hours after transfection with 80 nmol/l of control or caspase-3 siRNA, were treated with, TCTP- or control-ASO (100 nmol/l, 24 hours). Proteins were extracted for western blotting with a caspase-3 antibody that recognizes both full-length and cleaved caspase-3. AI, androgen independent; ASO, antisense oligonucleotide; siRNA, small interference RNA; TCTP, translationally controlled tumor protein.

Figure 5

TCTP-ASO treatment inhibits CR tumor progression, enhances docetaxel chemotherapy, and delays prostate cancer progression in vivo. (a,b) Mice bearing AI PC-3 tumors were randomly selected for treatment with (a) TCTP- or control-ASO alone, or (b) combined with docetaxel. When PC-3 tumors reached 50 mm³, 12.5 mg/kg/mouse of TCTP- or control-ASO were injected intraperitoneally (i.p.) daily for 10 weeks in animals receiving ASO monotherapy and for 13 weeks in animals receiving ASO combined with docetaxel. From days 7–14 docetaxel (see Material and Methods) was administered i.p. three times per week. Tumor volume was measured once weekly and calculated by the formula length × width × depth × 0.5236. Points, mean tumor volume in each experimental group containing 10 mice; bars, SE. *, **, and ***differ from control-ASO (P ≤ 0.05, P ≤ 0.01, and P ≤ 0.001, respectively) by Statview software. (c) Effect of TCTP-ASO treatment on LNCaP tumor growth in vivo after castration. Twenty male mice bearing LNCaP tumors were randomly selected for treatment with TCTP- or control-ASO. Castration was performed when tumors reached a mean volume of 150 mm³. TCTP- or control-ASO were injected (12.5 mg/kg/day) daily from 1 to 9 weeks after castration. Tumor volume was measured weekly. Each point represents the mean tumor volume of 10 mice; bars, SE. *P ≤ 0.05; **P ≤ 0.01, differ from control by Statview software. (d) Photographs of PC-3– and LNCaP-harvested tumors from animals that received i.p. TCTP- or control-ASO. (e) Proteins were extracted from harvested PC-3 tumors treated with TCTP- or control-ASO alone. The effect of TCTP-ASO treatment on caspase-3 cleavage was analyzed by western blot. (f) Body weight of animals treated i.p. with ASOs over the duration of the experiment. t0 = body weight before the first injection. tf = body weight the day of killing. AI, androgen independent; ASO, antisense oligonucleotide; CR, castration resistant; CS, castration sensitive; TCTP, translationally controlled tumor protein.

Figure 6

TCTP silencing inhibits CR tumor growth by inducing P53. (a) Proteins were extracted from CS (harvested before castration) and CR (harvested 40 days after castration) LNCaP tumors. P53 and vinculin protein levels were analyzed by western blot. (b) CR LNCaP tumors treated with TCTP- or control-ASO after castration, were used to analyze P53 and vinculin protein levels by western blot. (c) P53 protein levels were analyzed by western blot after P53- or mock-shRNA treatment. (d) LNCaP cell survival after combined P53- or mock-shRNA and TCTP- or control-ASO treatments was analyzed. After 2 weeks, cell viability was determined using a crystal violet assay. Error bars represent the SE, **P ≤ 0.01 and ***P ≤ 0.001 by Statview software. ASO, antisense oligonucleotide; CR, castration resistant; CS, castration sensitive; shRNA, short hairpin RNA; TCTP, translationally controlled tumor protein.