epage PDF coc506e05cee7a3a
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
Inhibition of androgen/β-catenin activity as a new treatment approach to prostate cancer Eugine Lee1,4 , Michael J. Garabedian2 , Ramanuj DasGupta1,4,* , Susan K. Logan1,3,4,* Departments of 1 Biochemistry and Molecular Pharmacology, 2 Microbiology, 3 Urology and 4 Stem Cell Biology, New York University School of Medicine, New York, NY 10016 Submitted to Proceedings of the National Academy of Sciences of the United States of America
Androgen receptor (AR) is the major therapeutic target in aggressive prostate cancer. However, targeting AR alone can result in drug resistance and disease recurrence. Therefore, simultaneous targeting of multiple pathways could in principle be an effective new approach to treating prostate cancer. Here we provide proofof-concept that a small molecule inhibitor of nuclear β-catenin activity (called C3) can inhibit both the AR and β-catenin signaling pathways that are often misregulated in prostate cancer. Treatment with C3 ablated prostate cancer cell growth by disruption of both β-catenin/TCF and β-catenin/AR protein interaction, reflecting the fact that TCF and AR have overlapping binding sites on βcatenin. Given that AR interacts with, and is transcriptionally regulated by β-catenin, C3 treatment also resulted in decreased occupancy of β-catenin on the AR promoter and diminished AR and AR/β-catenin target gene expression. Surprisingly, C3 treatment resulted in decreased AR binding to target genes accompanied by decreased recruitment of an AR and β-catenin cofactor, CARM1, providing new insight into the unrecognized function of β-catenin in prostate cancer. Importantly, C3 inhibited tumor growth in an in vivo xenograft model, and blocked renewal of bicalutamideresistant sphere forming cells, indicating the therapeutic potential of this approach.
β-catenin such as c-myc, which acts as an oncogene in prostate cancer (12). Recent studies have identified a small subpopulation of cancer cells, termed “cancer stem cells (CSCs)” or “tumor-initiating cells” based on their ability to self-renew, that play a critical role in both initiation and maintenance of tumors (13). These cells are resistant to conventional chemotherapy and radiation (14) and in prostate cancer, CSCs may survive after androgen ablation therapy, causing CRPC (15). Wnt/β-catenin signaling is highly active in CSCs (16-19), providing an additional rationale to target the Wnt/β-catenin pathway in prostate cancer. In this study, we utilized a recently identified small molecule inhibitor of nuclear β-catenin function, called “inhibitor of βcatenin responsive transcription” (iCRT) (20), to test the hypothesis that inhibition of nuclear β-catenin would repress prostate cancer growth by disrupting both AR and Wnt/β-catenin signaling. The use of inhibitors that specifically target the nuclear function of β-catenin may avoid disruption of cell-cell adherens junctions, which has been implicated in promoting epithelialmesenchymal transition (EMT) and metastasis. We tested the three lead compounds isolated in the previous study, iCRT-3, 5 and -14 (20), for inhibition of prostate cancer cell growth and selected iCRT3 (referred to as C3 from here on) for further analysis based on its high potency. We found that C3 interferes with two different protein-protein interactions in prostate cancer cells, TCF/β-catenin and AR/β-catenin. In addition, C3 inhibited both in vitro and in vivo proliferation of the LNCaP-abl, androgen receptor sensitive but androgen-ligand insensitive cell line, which serves as a model of CRPC. Importantly, C3 inhibited self-renewal of prostate cancer cells that express higher levels of stem cell markers, suggesting that inhibition of nuclear β-catenin function may be an important therapeutic approach to advanced prostate cancer treatment.
Submission PDF
prostate cancer | androgen receptor | Wnt//β-catenin
Introduction Prostate cancer is the most common form of cancer in males and is currently treated with androgen deprivation therapy. While this results in tumor regression, aggressive disease typically recurs, making the treatment of what is then called castrationresistant prostate cancer (CRPC) the major challenge in the field. AR up-regulation is the major determinate in CRPC (1), but conventional anti-androgen drugs fail to block AR activity in CRPCs where they can gain partial AR agonist properties (1). New promising drugs have been reported (2, 3) but they extend life by only 4-5 months (4), hinting that targeting AR activity might not be enough to inhibit tumor growth given the idea that increased crosstalk between distinct signaling pathways causes activation of AR regulatory networks in advanced prostate cancer (5). Therefore, development of drugs that target multiple pathways or can be used sequentially will further improve life expectancy. Growing evidence indicates that the canonical Wnt/β-catenin pathway plays an important role in prostate tumorigenesis (6). Recent studies reveal that Wnt signaling is a significantly mutated pathway in lethal CRPC (7). Additionally, Wnt16B promotes resistant disease, underscoring the importance of targeting the Wnt/β-catenin pathway in advanced disease (8). Synergy between AR and β-catenin pathways has been well documented. AR binds β-catenin directly to stimulate AR mediated gene transcription (9) and importantly, the AR gene itself is a transcriptional target of β-catenin (10). Furthermore, enhanced crosstalk between AR and β-catenin has been observed in in vivo models of CRPC (11). Therefore, hypothetically, inhibitors of nuclear β-catenin would modulate AR and its target genes including the direct targets of www.pnas.org --- ---
Results C3 affects AR and β-catenin signaling by inhibition of proteinprotein interaction
Significance proof-of-concept for the use of new reagents that have multipronged effects to treat prostate cancer
Reserved for Publication Footnotes
PNAS
Issue Date
Volume
Issue Number
1--??
69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136
137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
Fig. 2. C3 inhibits expression of AR and β-catenin target genes. (A and B) C3 inhibits transcription of AR and β-catenin target genes in LNCaP (A) and abl (B) cells, as judged by qPCR. LNCaP cells were androgen-deprived for 48 h and treated for 24 h with DMSO or 20 μM C3 with or without 0.1 nM R1881. abl cells were treated with DMSO or 20 μM C3 for 24 h. (C) Loss of protein expression of AR and β-catenin target genes in the presence of C3. Protein from LNCaP and abl cells treated as described in (A) and (B) was extracted at 4, 24 and 48 h after the initial treatment. AR levels were normalized to corresponding tubulin levels and the fold over 4hr vehicle treated cells are presented on the bottom of the figure. (D) β-catenin knockdown decreases the expression of AR and Wnt/β-catenin target genes in LNCaP and abl cells. Cells were transfected with control or β-catenin siRNA and cultured for 48 h. (E) β-catenin knockdown decreases protein levels of AR and c-myc in LNCaP and abl cells. Cells were transfected as described in (D).
Submission PDF Fig. 1. C3 affects AR and β-catenin signaling by inhibition of protein:protein interaction.(A) Chemical structure of iCRT3 (C3). (B) Co-immunoprecipitation studies reveal that C3 inhibits β-catenin/TCF4 and β-catenin/AR interactions in LNCaP cells. Cells were treated with vehicle or 20 μM C3 for 4 h. (C) LNCaP, abl and VCaP cells express high levels of nuclear β-catenin. Transcription factor II-I (TFII-I) and tubulin antibodies are included as nuclear and cytoplasmic loading controls, respectively. (D) C3 inhibits transcription of AR nascent mRNA. LNCaP cells were androgen-deprived for 48 h and then treated with DMSO or 20 μM C3 with or without 0.1 nM R1881 for 4 h. (E) C3 inhibits βcatenin occupancy on a TCF/LEF binding site within the AR promoter. LNCaP cells were androgen-deprived for 3 days and then treated with vehicle or 100 nM DHT with or without 20 μM C3 for 4 h. (F) β-catenin knockdown inhibits AR expression in abl cells. Cells were transfected with control or β-catenin siRNA and cultured for 48 h. Cells were fixed and incubated with anti-AR (red), anti-β-catenin (green) antibodies, and DAPI solution (blue). Scale bar, 50 μM.(G) Determination of the IC50 of C3 in prostate cancer cells. Cells were transfected with the indicated luciferase reporter constructs; STF16-LUC (βcatenin reporter) and ARE-LUC (AR reporter).
Given that AR interacts with, and is transcriptionally regulated by β-catenin, we reasoned that inhibition of nuclear βcatenin could interfere with both AR and β-catenin signaling. C3 (Fig. 1A) was shown to specifically inhibit Wnt/β-catenininduced transcription by interfering with the interaction between β-catenin and TCF-4 (20). Since AR and TCF-4 have overlapping interaction domains with β-catenin (9, 21), we hypothesized that C3 may also inhibit AR and β-catenin interaction. In support of this idea, C3 robustly inhibited β-catenin/TCF-4 as well as βcatenin/AR interactions in LNCaP cells (Fig. 1B). We examined the levels of nuclear (transcriptionally active) β-catenin in androgen-dependent LNCaP cells, an androgenindependent LNCaP cell derivative called LNCaP-abl (abl) (22), and androgen-dependent VCaP cells, which have amplified AR (23). β-catenin was abundantly expressed in the nucleus of all three cell lines indicating that they may be susceptible to β-catenin inhibition (Fig. 1C and S1A). Importantly, C3 did not disrupt localization of junctional β-catenin at the cell membrane (Fig. S1B), the loss of which is implicated in tumor metastasis. As AR is transcriptionally regulated by β-catenin through TCF/LEF binding sites on AR promoter, we reasoned that loss 2
www.pnas.org --- ---
of TCF/β-catenin interaction would diminish AR mRNA levels in C3 treated cells. To test if transcription of AR is directly targeted by C3, nascent AR mRNA levels were analyzed in LNCaP cells using primers flanking exon/intron junctions. Indeed, C3 treatment decreased nascent AR mRNA levels (Fig. 1D), consistent with the decreased occupancy of β-catenin on TCF/LEF binding sites within the AR promoter shown by chromatin immunoprecipitation (ChIP) (Fig. 1E). Depletion of β-catenin revealed that cells retaining strong β-catenin levels due to incomplete knockdown (Fig. 1F, arrow) clearly have higher levels of AR than adjacent cells with lower levels of β-catenin (Fig. 1F, arrowheads), indicating that the cell specific expression of AR is regulated by β-catenin. We next tested the effect of C3 on transcriptional activities of AR and β-catenin. C3 dramatically reduced expression of both AR and Wnt/β-catenin reporter genes (Fig. S1C), and the overexpression of AR cDNA diminished the inhibitory effect of C3 on AR reporter activity (Fig. S1D), suggesting that C3 may directly influence AR activity. Dose-responsive inhibition of C3 on AR and Wnt/β-catenin transcriptional activities revealed that C3 inhibits the Wnt/β-catenin reporter with an IC50 of 0.14 μM in LNCaP, 0.89 μM in abl and 0.97 μM in VCaP cells (Fig. 1G). The somewhat higher IC50 in abl cells likely reflects the increased activity of nuclear β-catenin as seen in other androgenindependent cell lines and in a CRPC model (11, 24). Also, the higher IC50 in VCaP cells is consistent with the previous report showing ERG-induced Wnt signaling (25). The IC50 needed to inhibit the AR reporter was comparable between LNCaP and abl cells (Fig. 1G, 1.44 and 1.61 μM), but higher in VCaP cells (6.51 μM), reflecting the amplification of AR in VCaP cells (23). AR ligand competition assays were conducted to test if C3 modulates AR activity by affecting ligand binding. Samples incubated with C3 showed that C3 does not affect ligand binding (Fig. Footline Author
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272
273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408
Submission PDF Fig. 3. C3 induces growth arrest and apoptosis in LNCaP and abl cells.(A) Growth curve assays and cell cycle profile of LNCaP and abl cells in the presence of C3: Upper panel; Cells were treated with DMSO or indicated concentration of C3 everyday. Lower panel; Cells were treated with DMSO or 20 μM C3 for 24 h and then subjected to cell cycle analysis. (B) C3 treatment induces apoptosis in LNCaP and abl cells as indicated by the increased expression of cleaved PARP. Cells were treated with DMSO or 20 μM C3 for 72 h. (C and D) C3 does not induce significant growth arrest or apoptosis in HEK293 cells. HEK 293 cells were treated with DMSO, 10 μM or 20 μM C3 everyday (C). To analyze apoptosis, cells were treated with DMSO or 20uM C3 for 72 h (D). (E) β-catenin knockdown inhibits growth of LNCaP and abl cells. Cells were stably infected with lentiviral vectors encoding control shRNA (sh-control) or β-catenin shRNA (sh-β-catenin) and subjected to proliferation assay.
S1E). Examination of AR coactivator recruitment showed that C3 did not induce recruitment of FxxLF coactivator peptide to the AR-ligand binding domain (LBD), whereas di-hydrotestosterone (DHT), and higher concentrations of the AR antagonist, bicalutamide (BIC), induced dose-dependent interaction of the peptide and AR-LBD (Fig. S1F). C3 inhibits AR and Wnt/β-catenin target gene expression Next, we evaluated the impact of C3 on endogenous AR and Wnt/β-catenin target genes. LNCaP and VCaP cells were treated with vehicle or R1881 at the growth promoting concentration of 0.1 nM in the presence or absence of C3, while abl cells, which grow in the absence of androgens, were treated only with C3. AR mRNA was quantified along with well-characterized AR target genes, PSA and NKX3.1. In addition, we examined CDC20, CDK1 and UBE2C, M phase cell cycle regulatory genes that are AR targets in abl cells (26). Levels of the Wnt/β-catenin target gene, c-myc, were also examined. C3 treatment decreased transcription of AR mRNA, and AR and β-catenin target genes in both LNCaP and abl cells (Fig. 2A and B). This inhibitory effect was observed Footline Author
Fig. 4. C3 inhibits AR and β-catenin occupancy on target genes.(A) C3 does not inhibit expression of AR, β-catenin and CARM1 when treated with 100nM DHT in LNCaP cells. Cells were androgen-deprived for 3 days and then treated with vehicle or 20 μM C3 in the presence of 100 nM DHT for 16 h. (B) C3 decreases expression of PSA, UBE2C and C-MYC when treated with 100nM DHT in LNCaP cells. Cells were androgen-deprived for 3 days and then treated with vehicle or 20 μM C3 in the presence of 100 nM DHT for 24 h. (C-F) C3 decreases occupancy of AR, β-catenin, CARM1 and dimethyl-H3R17 on AR binding sites.LNCaP cells were treated as described in (A) and crosslinked at 16 h post treatment.
as early as 4 h after treatment in some of the target genes (Fig. S2A) and persisted up to 48 h after treatment (Fig. S2B). Expression of control genes, E-cadherin and the housekeeping gene GAPDH, remained unaffected (Fig. 2A and B). We also tested transcription of a gene reported to be repressed by AR, MAN1A1 (27). The results shown in Figure 2A and B indicate that MAN1A is derepressed in response to C3 treatment in both LNCaP and abl cells, suggesting that there is altered transcription of genes both positively and negatively regulated by AR in response to C3 treatment. Decreased protein levels of target genes in C3 treated LNCaP, abl and VCaP cells were also observed, consistent with decreased mRNA levels (Fig. 2C and S2D). Treatment of VCaP cells with C3 also inhibited mRNA expression of target genes in the presence of R1881, with the exception of NKX3.1 (Fig. S2C). There are some differences in target gene expression upon C3 treatment between the LNCaP and VCaP cells (Fig. 2A versus S2C), reflecting the possibility that VCaP cells may use a different mechanism to regulate the Wnt pathway than LNCaP cells as suggested by a study showing that the TMPRSS2-ERG fusion gene in VCaP cells up-regulates Wnt pathway components including the receptor, frizzled-4 (25). Next, we tested if the expression of AR and Wnt/β-catenin target genes is similarly affected by β-catenin depletion. RNAi-mediated depletion of βcatenin in LNCaP and abl cells resulted in down-regulation of AR levels, as well as AR and Wnt/β-catenin target gene expression, in a manner similar to that observed upon C3 treatment (Fig. 2D and PNAS
Issue Date
Volume
Issue Number
3
409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476
with si-β-catenin, C3 and si-AR resulted in similar genes being differentially expressed, with the greatest overlap in the si-βcatenin and C3 treatments (Fig S3A’). GSEA analysis on a set of 179 previously annotated AR target genes (26) revealed that while there is a clear separation between the GSEA graphs of siAR and si-β-catenin compared to controls, it is not as strong as for the β-catenin target gene set, suggesting that while some target genes may be commonly regulated by β-catenin and AR, there must be distinct sets of AR-specific target genes that are independent of β-catenin regulation. Consistent with GSEA, cluster analysis indicates that there were some AR target genes similarly affected by si-β-catenin, C3 and AR siRNA and others affected by si-AR and either C3 or si-β-catenin (Fig.S3B’). Given that off target effects (OTEs) are critical factor to consider for small molecule inhibitor, we reanalyzed the top 1,000 genes downregulated in cells treated with si-β-catenin, and compared the effect on their expression by C3 and si-AR treatment (description of the method of analysis is in SI Figure legend 3C and D). Figure S3D shows a fairly similar range of overlap between C3 and si-βcatenin (67-58%) and si-AR and si-β-catenin (74-48%). At face value, this indicates that the upper limit of OTEs for C3 is 33-42%. However, some OTEs may result from intrinsic experimental error, such as biological variability and incomplete knockout of β-catenin, as observed in Figure S2D, given that some of the known Wnt/β-catenin targets such as Fzd7, VEGF, FGF were found amongst the genes that appear to be specifically modulated by C3, not si-β-catenin. Taken together, the GSEA and clustering analysis strongly suggests that C3 targets a similar set of β-catenin target genes compared to si-β-catenin. C3 blocks proliferation of prostate cancer cells by inducing cell death To test if C3 inhibits prostate cell growth, LNCaP, abl and VCaP cells were treated with varying concentrations of C3 or vehicle. Treatment with 20 μM C3 completely abolished growth of LNCaP, abl (Fig. 3A) and VCaP cells (Fig. S4). While 24 h treatment of C3 induced growth arrest of the cells in G0/G1 (Fig. 3A, lower panels), 72 h treatment induced apoptosis shown by increased levels of PARP cleavage (Fig. 3B). To determine if C3 was generally toxic, as opposed to inhibiting specific pathways in individual cell types, we examined the effect of C3 on HEK 293 cells, which require Wnt ligand for activation of the pathway (28). While higher levels of C3 modestly slow cell growth in HEK 293 cells (Fig. 3C), apoptosis was not induced (Fig. 3D). β- catenin depletion significantly reduced growth of LNCaP, abl and VCaP cells (Fig. 3E, S4B and S5), indicating that the proliferation of prostate cancer cells is dependent on β-catenin. C3 inhibits AR and β-catenin occupancy on target genes Since β-catenin can act as an AR coactivator (29, 30) and we found that C3 inhibits AR/β-catenin protein interaction (Fig. 1), we reasoned that C3 might affect AR recruitment to target genes. To test this hypothesis, we examined AR occupancy in the presence and absence of C3 via ChIP. To circumvent the reduced levels of total AR observed in C3 treated cells in the presence of low (0.1nM) androgen concentration (Fig. 2), we treated cells with 100nM DHT which stabilized AR protein and diminished the inhibitory effect of C3 on total AR levels (Fig. 4A). The inhibitory effect of C3 on transcription of AR and Wnt/β-catenin target genes was sustained when C3 was treated with 100nM DHT (Fig. 4B). A comparison of AR and β-catenin recruitment to the PSA and UBE2C enhancers showed that both proteins were recruited in response to DHT treatment but cells treated with C3 and DHT exhibited decreased recruitment of both proteins (Fig. 4C and D). ChIP results on the c-myc promoter showed that while DHT treatment increased β-catenin occupancy, C3 decreased this recruitment to a great extent (Fig. 4D). Consistent with a previous report (31), AR was also recruited to c-myc promoter in response to DHT but the level of recruitment was modest and
Submission PDF Fig. 5. C3 inhibits the self-renewal ability of prostate sphere forming cells.(A) LNCaP and abl spheres show higher levels of stem cell marker expression. Cells were cultured in the same media under adherent or sphere forming conditions.(B) C3 inhibits sphere formation in LNCaP and abl. Upper panel; cells were cultured under sphere forming conditions. Vehicle or 20 μM C3 were added a single time on day 0. Lower panel; representative pictures of the LNCaP spheres were imaged. Scale bar, 100 μM. (C and D) bicalutamide (BIC) blocks proliferation of adherent LNCaP but has no effect on sphere formation. Cells were treated with vehicle, BIC or C3 every two days in adherent conditions (C) or a single time on day 0 in sphere forming conditions (D). One-way ANOVA, *** p
Inhibition of androgen/β-catenin activity as a new treatment approach to prostate cancer Eugine Lee1,4 , Michael J. Garabedian2 , Ramanuj DasGupta1,4,* , Susan K. Logan1,3,4,* Departments of 1 Biochemistry and Molecular Pharmacology, 2 Microbiology, 3 Urology and 4 Stem Cell Biology, New York University School of Medicine, New York, NY 10016 Submitted to Proceedings of the National Academy of Sciences of the United States of America
Androgen receptor (AR) is the major therapeutic target in aggressive prostate cancer. However, targeting AR alone can result in drug resistance and disease recurrence. Therefore, simultaneous targeting of multiple pathways could in principle be an effective new approach to treating prostate cancer. Here we provide proofof-concept that a small molecule inhibitor of nuclear β-catenin activity (called C3) can inhibit both the AR and β-catenin signaling pathways that are often misregulated in prostate cancer. Treatment with C3 ablated prostate cancer cell growth by disruption of both β-catenin/TCF and β-catenin/AR protein interaction, reflecting the fact that TCF and AR have overlapping binding sites on βcatenin. Given that AR interacts with, and is transcriptionally regulated by β-catenin, C3 treatment also resulted in decreased occupancy of β-catenin on the AR promoter and diminished AR and AR/β-catenin target gene expression. Surprisingly, C3 treatment resulted in decreased AR binding to target genes accompanied by decreased recruitment of an AR and β-catenin cofactor, CARM1, providing new insight into the unrecognized function of β-catenin in prostate cancer. Importantly, C3 inhibited tumor growth in an in vivo xenograft model, and blocked renewal of bicalutamideresistant sphere forming cells, indicating the therapeutic potential of this approach.
β-catenin such as c-myc, which acts as an oncogene in prostate cancer (12). Recent studies have identified a small subpopulation of cancer cells, termed “cancer stem cells (CSCs)” or “tumor-initiating cells” based on their ability to self-renew, that play a critical role in both initiation and maintenance of tumors (13). These cells are resistant to conventional chemotherapy and radiation (14) and in prostate cancer, CSCs may survive after androgen ablation therapy, causing CRPC (15). Wnt/β-catenin signaling is highly active in CSCs (16-19), providing an additional rationale to target the Wnt/β-catenin pathway in prostate cancer. In this study, we utilized a recently identified small molecule inhibitor of nuclear β-catenin function, called “inhibitor of βcatenin responsive transcription” (iCRT) (20), to test the hypothesis that inhibition of nuclear β-catenin would repress prostate cancer growth by disrupting both AR and Wnt/β-catenin signaling. The use of inhibitors that specifically target the nuclear function of β-catenin may avoid disruption of cell-cell adherens junctions, which has been implicated in promoting epithelialmesenchymal transition (EMT) and metastasis. We tested the three lead compounds isolated in the previous study, iCRT-3, 5 and -14 (20), for inhibition of prostate cancer cell growth and selected iCRT3 (referred to as C3 from here on) for further analysis based on its high potency. We found that C3 interferes with two different protein-protein interactions in prostate cancer cells, TCF/β-catenin and AR/β-catenin. In addition, C3 inhibited both in vitro and in vivo proliferation of the LNCaP-abl, androgen receptor sensitive but androgen-ligand insensitive cell line, which serves as a model of CRPC. Importantly, C3 inhibited self-renewal of prostate cancer cells that express higher levels of stem cell markers, suggesting that inhibition of nuclear β-catenin function may be an important therapeutic approach to advanced prostate cancer treatment.
Submission PDF
prostate cancer | androgen receptor | Wnt//β-catenin
Introduction Prostate cancer is the most common form of cancer in males and is currently treated with androgen deprivation therapy. While this results in tumor regression, aggressive disease typically recurs, making the treatment of what is then called castrationresistant prostate cancer (CRPC) the major challenge in the field. AR up-regulation is the major determinate in CRPC (1), but conventional anti-androgen drugs fail to block AR activity in CRPCs where they can gain partial AR agonist properties (1). New promising drugs have been reported (2, 3) but they extend life by only 4-5 months (4), hinting that targeting AR activity might not be enough to inhibit tumor growth given the idea that increased crosstalk between distinct signaling pathways causes activation of AR regulatory networks in advanced prostate cancer (5). Therefore, development of drugs that target multiple pathways or can be used sequentially will further improve life expectancy. Growing evidence indicates that the canonical Wnt/β-catenin pathway plays an important role in prostate tumorigenesis (6). Recent studies reveal that Wnt signaling is a significantly mutated pathway in lethal CRPC (7). Additionally, Wnt16B promotes resistant disease, underscoring the importance of targeting the Wnt/β-catenin pathway in advanced disease (8). Synergy between AR and β-catenin pathways has been well documented. AR binds β-catenin directly to stimulate AR mediated gene transcription (9) and importantly, the AR gene itself is a transcriptional target of β-catenin (10). Furthermore, enhanced crosstalk between AR and β-catenin has been observed in in vivo models of CRPC (11). Therefore, hypothetically, inhibitors of nuclear β-catenin would modulate AR and its target genes including the direct targets of www.pnas.org --- ---
Results C3 affects AR and β-catenin signaling by inhibition of proteinprotein interaction
Significance proof-of-concept for the use of new reagents that have multipronged effects to treat prostate cancer
Reserved for Publication Footnotes
PNAS
Issue Date
Volume
Issue Number
1--??
69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136
137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
Fig. 2. C3 inhibits expression of AR and β-catenin target genes. (A and B) C3 inhibits transcription of AR and β-catenin target genes in LNCaP (A) and abl (B) cells, as judged by qPCR. LNCaP cells were androgen-deprived for 48 h and treated for 24 h with DMSO or 20 μM C3 with or without 0.1 nM R1881. abl cells were treated with DMSO or 20 μM C3 for 24 h. (C) Loss of protein expression of AR and β-catenin target genes in the presence of C3. Protein from LNCaP and abl cells treated as described in (A) and (B) was extracted at 4, 24 and 48 h after the initial treatment. AR levels were normalized to corresponding tubulin levels and the fold over 4hr vehicle treated cells are presented on the bottom of the figure. (D) β-catenin knockdown decreases the expression of AR and Wnt/β-catenin target genes in LNCaP and abl cells. Cells were transfected with control or β-catenin siRNA and cultured for 48 h. (E) β-catenin knockdown decreases protein levels of AR and c-myc in LNCaP and abl cells. Cells were transfected as described in (D).
Submission PDF Fig. 1. C3 affects AR and β-catenin signaling by inhibition of protein:protein interaction.(A) Chemical structure of iCRT3 (C3). (B) Co-immunoprecipitation studies reveal that C3 inhibits β-catenin/TCF4 and β-catenin/AR interactions in LNCaP cells. Cells were treated with vehicle or 20 μM C3 for 4 h. (C) LNCaP, abl and VCaP cells express high levels of nuclear β-catenin. Transcription factor II-I (TFII-I) and tubulin antibodies are included as nuclear and cytoplasmic loading controls, respectively. (D) C3 inhibits transcription of AR nascent mRNA. LNCaP cells were androgen-deprived for 48 h and then treated with DMSO or 20 μM C3 with or without 0.1 nM R1881 for 4 h. (E) C3 inhibits βcatenin occupancy on a TCF/LEF binding site within the AR promoter. LNCaP cells were androgen-deprived for 3 days and then treated with vehicle or 100 nM DHT with or without 20 μM C3 for 4 h. (F) β-catenin knockdown inhibits AR expression in abl cells. Cells were transfected with control or β-catenin siRNA and cultured for 48 h. Cells were fixed and incubated with anti-AR (red), anti-β-catenin (green) antibodies, and DAPI solution (blue). Scale bar, 50 μM.(G) Determination of the IC50 of C3 in prostate cancer cells. Cells were transfected with the indicated luciferase reporter constructs; STF16-LUC (βcatenin reporter) and ARE-LUC (AR reporter).
Given that AR interacts with, and is transcriptionally regulated by β-catenin, we reasoned that inhibition of nuclear βcatenin could interfere with both AR and β-catenin signaling. C3 (Fig. 1A) was shown to specifically inhibit Wnt/β-catenininduced transcription by interfering with the interaction between β-catenin and TCF-4 (20). Since AR and TCF-4 have overlapping interaction domains with β-catenin (9, 21), we hypothesized that C3 may also inhibit AR and β-catenin interaction. In support of this idea, C3 robustly inhibited β-catenin/TCF-4 as well as βcatenin/AR interactions in LNCaP cells (Fig. 1B). We examined the levels of nuclear (transcriptionally active) β-catenin in androgen-dependent LNCaP cells, an androgenindependent LNCaP cell derivative called LNCaP-abl (abl) (22), and androgen-dependent VCaP cells, which have amplified AR (23). β-catenin was abundantly expressed in the nucleus of all three cell lines indicating that they may be susceptible to β-catenin inhibition (Fig. 1C and S1A). Importantly, C3 did not disrupt localization of junctional β-catenin at the cell membrane (Fig. S1B), the loss of which is implicated in tumor metastasis. As AR is transcriptionally regulated by β-catenin through TCF/LEF binding sites on AR promoter, we reasoned that loss 2
www.pnas.org --- ---
of TCF/β-catenin interaction would diminish AR mRNA levels in C3 treated cells. To test if transcription of AR is directly targeted by C3, nascent AR mRNA levels were analyzed in LNCaP cells using primers flanking exon/intron junctions. Indeed, C3 treatment decreased nascent AR mRNA levels (Fig. 1D), consistent with the decreased occupancy of β-catenin on TCF/LEF binding sites within the AR promoter shown by chromatin immunoprecipitation (ChIP) (Fig. 1E). Depletion of β-catenin revealed that cells retaining strong β-catenin levels due to incomplete knockdown (Fig. 1F, arrow) clearly have higher levels of AR than adjacent cells with lower levels of β-catenin (Fig. 1F, arrowheads), indicating that the cell specific expression of AR is regulated by β-catenin. We next tested the effect of C3 on transcriptional activities of AR and β-catenin. C3 dramatically reduced expression of both AR and Wnt/β-catenin reporter genes (Fig. S1C), and the overexpression of AR cDNA diminished the inhibitory effect of C3 on AR reporter activity (Fig. S1D), suggesting that C3 may directly influence AR activity. Dose-responsive inhibition of C3 on AR and Wnt/β-catenin transcriptional activities revealed that C3 inhibits the Wnt/β-catenin reporter with an IC50 of 0.14 μM in LNCaP, 0.89 μM in abl and 0.97 μM in VCaP cells (Fig. 1G). The somewhat higher IC50 in abl cells likely reflects the increased activity of nuclear β-catenin as seen in other androgenindependent cell lines and in a CRPC model (11, 24). Also, the higher IC50 in VCaP cells is consistent with the previous report showing ERG-induced Wnt signaling (25). The IC50 needed to inhibit the AR reporter was comparable between LNCaP and abl cells (Fig. 1G, 1.44 and 1.61 μM), but higher in VCaP cells (6.51 μM), reflecting the amplification of AR in VCaP cells (23). AR ligand competition assays were conducted to test if C3 modulates AR activity by affecting ligand binding. Samples incubated with C3 showed that C3 does not affect ligand binding (Fig. Footline Author
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272
273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408
Submission PDF Fig. 3. C3 induces growth arrest and apoptosis in LNCaP and abl cells.(A) Growth curve assays and cell cycle profile of LNCaP and abl cells in the presence of C3: Upper panel; Cells were treated with DMSO or indicated concentration of C3 everyday. Lower panel; Cells were treated with DMSO or 20 μM C3 for 24 h and then subjected to cell cycle analysis. (B) C3 treatment induces apoptosis in LNCaP and abl cells as indicated by the increased expression of cleaved PARP. Cells were treated with DMSO or 20 μM C3 for 72 h. (C and D) C3 does not induce significant growth arrest or apoptosis in HEK293 cells. HEK 293 cells were treated with DMSO, 10 μM or 20 μM C3 everyday (C). To analyze apoptosis, cells were treated with DMSO or 20uM C3 for 72 h (D). (E) β-catenin knockdown inhibits growth of LNCaP and abl cells. Cells were stably infected with lentiviral vectors encoding control shRNA (sh-control) or β-catenin shRNA (sh-β-catenin) and subjected to proliferation assay.
S1E). Examination of AR coactivator recruitment showed that C3 did not induce recruitment of FxxLF coactivator peptide to the AR-ligand binding domain (LBD), whereas di-hydrotestosterone (DHT), and higher concentrations of the AR antagonist, bicalutamide (BIC), induced dose-dependent interaction of the peptide and AR-LBD (Fig. S1F). C3 inhibits AR and Wnt/β-catenin target gene expression Next, we evaluated the impact of C3 on endogenous AR and Wnt/β-catenin target genes. LNCaP and VCaP cells were treated with vehicle or R1881 at the growth promoting concentration of 0.1 nM in the presence or absence of C3, while abl cells, which grow in the absence of androgens, were treated only with C3. AR mRNA was quantified along with well-characterized AR target genes, PSA and NKX3.1. In addition, we examined CDC20, CDK1 and UBE2C, M phase cell cycle regulatory genes that are AR targets in abl cells (26). Levels of the Wnt/β-catenin target gene, c-myc, were also examined. C3 treatment decreased transcription of AR mRNA, and AR and β-catenin target genes in both LNCaP and abl cells (Fig. 2A and B). This inhibitory effect was observed Footline Author
Fig. 4. C3 inhibits AR and β-catenin occupancy on target genes.(A) C3 does not inhibit expression of AR, β-catenin and CARM1 when treated with 100nM DHT in LNCaP cells. Cells were androgen-deprived for 3 days and then treated with vehicle or 20 μM C3 in the presence of 100 nM DHT for 16 h. (B) C3 decreases expression of PSA, UBE2C and C-MYC when treated with 100nM DHT in LNCaP cells. Cells were androgen-deprived for 3 days and then treated with vehicle or 20 μM C3 in the presence of 100 nM DHT for 24 h. (C-F) C3 decreases occupancy of AR, β-catenin, CARM1 and dimethyl-H3R17 on AR binding sites.LNCaP cells were treated as described in (A) and crosslinked at 16 h post treatment.
as early as 4 h after treatment in some of the target genes (Fig. S2A) and persisted up to 48 h after treatment (Fig. S2B). Expression of control genes, E-cadherin and the housekeeping gene GAPDH, remained unaffected (Fig. 2A and B). We also tested transcription of a gene reported to be repressed by AR, MAN1A1 (27). The results shown in Figure 2A and B indicate that MAN1A is derepressed in response to C3 treatment in both LNCaP and abl cells, suggesting that there is altered transcription of genes both positively and negatively regulated by AR in response to C3 treatment. Decreased protein levels of target genes in C3 treated LNCaP, abl and VCaP cells were also observed, consistent with decreased mRNA levels (Fig. 2C and S2D). Treatment of VCaP cells with C3 also inhibited mRNA expression of target genes in the presence of R1881, with the exception of NKX3.1 (Fig. S2C). There are some differences in target gene expression upon C3 treatment between the LNCaP and VCaP cells (Fig. 2A versus S2C), reflecting the possibility that VCaP cells may use a different mechanism to regulate the Wnt pathway than LNCaP cells as suggested by a study showing that the TMPRSS2-ERG fusion gene in VCaP cells up-regulates Wnt pathway components including the receptor, frizzled-4 (25). Next, we tested if the expression of AR and Wnt/β-catenin target genes is similarly affected by β-catenin depletion. RNAi-mediated depletion of βcatenin in LNCaP and abl cells resulted in down-regulation of AR levels, as well as AR and Wnt/β-catenin target gene expression, in a manner similar to that observed upon C3 treatment (Fig. 2D and PNAS
Issue Date
Volume
Issue Number
3
409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476
with si-β-catenin, C3 and si-AR resulted in similar genes being differentially expressed, with the greatest overlap in the si-βcatenin and C3 treatments (Fig S3A’). GSEA analysis on a set of 179 previously annotated AR target genes (26) revealed that while there is a clear separation between the GSEA graphs of siAR and si-β-catenin compared to controls, it is not as strong as for the β-catenin target gene set, suggesting that while some target genes may be commonly regulated by β-catenin and AR, there must be distinct sets of AR-specific target genes that are independent of β-catenin regulation. Consistent with GSEA, cluster analysis indicates that there were some AR target genes similarly affected by si-β-catenin, C3 and AR siRNA and others affected by si-AR and either C3 or si-β-catenin (Fig.S3B’). Given that off target effects (OTEs) are critical factor to consider for small molecule inhibitor, we reanalyzed the top 1,000 genes downregulated in cells treated with si-β-catenin, and compared the effect on their expression by C3 and si-AR treatment (description of the method of analysis is in SI Figure legend 3C and D). Figure S3D shows a fairly similar range of overlap between C3 and si-βcatenin (67-58%) and si-AR and si-β-catenin (74-48%). At face value, this indicates that the upper limit of OTEs for C3 is 33-42%. However, some OTEs may result from intrinsic experimental error, such as biological variability and incomplete knockout of β-catenin, as observed in Figure S2D, given that some of the known Wnt/β-catenin targets such as Fzd7, VEGF, FGF were found amongst the genes that appear to be specifically modulated by C3, not si-β-catenin. Taken together, the GSEA and clustering analysis strongly suggests that C3 targets a similar set of β-catenin target genes compared to si-β-catenin. C3 blocks proliferation of prostate cancer cells by inducing cell death To test if C3 inhibits prostate cell growth, LNCaP, abl and VCaP cells were treated with varying concentrations of C3 or vehicle. Treatment with 20 μM C3 completely abolished growth of LNCaP, abl (Fig. 3A) and VCaP cells (Fig. S4). While 24 h treatment of C3 induced growth arrest of the cells in G0/G1 (Fig. 3A, lower panels), 72 h treatment induced apoptosis shown by increased levels of PARP cleavage (Fig. 3B). To determine if C3 was generally toxic, as opposed to inhibiting specific pathways in individual cell types, we examined the effect of C3 on HEK 293 cells, which require Wnt ligand for activation of the pathway (28). While higher levels of C3 modestly slow cell growth in HEK 293 cells (Fig. 3C), apoptosis was not induced (Fig. 3D). β- catenin depletion significantly reduced growth of LNCaP, abl and VCaP cells (Fig. 3E, S4B and S5), indicating that the proliferation of prostate cancer cells is dependent on β-catenin. C3 inhibits AR and β-catenin occupancy on target genes Since β-catenin can act as an AR coactivator (29, 30) and we found that C3 inhibits AR/β-catenin protein interaction (Fig. 1), we reasoned that C3 might affect AR recruitment to target genes. To test this hypothesis, we examined AR occupancy in the presence and absence of C3 via ChIP. To circumvent the reduced levels of total AR observed in C3 treated cells in the presence of low (0.1nM) androgen concentration (Fig. 2), we treated cells with 100nM DHT which stabilized AR protein and diminished the inhibitory effect of C3 on total AR levels (Fig. 4A). The inhibitory effect of C3 on transcription of AR and Wnt/β-catenin target genes was sustained when C3 was treated with 100nM DHT (Fig. 4B). A comparison of AR and β-catenin recruitment to the PSA and UBE2C enhancers showed that both proteins were recruited in response to DHT treatment but cells treated with C3 and DHT exhibited decreased recruitment of both proteins (Fig. 4C and D). ChIP results on the c-myc promoter showed that while DHT treatment increased β-catenin occupancy, C3 decreased this recruitment to a great extent (Fig. 4D). Consistent with a previous report (31), AR was also recruited to c-myc promoter in response to DHT but the level of recruitment was modest and
Submission PDF Fig. 5. C3 inhibits the self-renewal ability of prostate sphere forming cells.(A) LNCaP and abl spheres show higher levels of stem cell marker expression. Cells were cultured in the same media under adherent or sphere forming conditions.(B) C3 inhibits sphere formation in LNCaP and abl. Upper panel; cells were cultured under sphere forming conditions. Vehicle or 20 μM C3 were added a single time on day 0. Lower panel; representative pictures of the LNCaP spheres were imaged. Scale bar, 100 μM. (C and D) bicalutamide (BIC) blocks proliferation of adherent LNCaP but has no effect on sphere formation. Cells were treated with vehicle, BIC or C3 every two days in adherent conditions (C) or a single time on day 0 in sphere forming conditions (D). One-way ANOVA, *** p
