Adaptive auto-regulation of androgen receptor provides a paradigm shifting rationale for bipolar androgen therapy (BAT) for castrate resistant human prostate cancer

Abstract

Cell culture/xenograft and gene arrays of clinical material document that development of castration resistant prostate cancer (CRPC) cells involves acquisition of adaptive auto-regulation resulting in >25-fold increase in Androgen Receptor (AR) protein expression in a low androgen environment. Such adaptive AR increase paradoxically is a liability in castrated hosts, however, when supraphysiologic androgen is acutely replaced. Cell synchronization/anti-androgen response is due to AR binding to replication complexes (RC) at origin of replication sites (ORS) in early G1 associated with licensing/restricting DNA for single round of duplication during S-phase. When CRPC cells are acutely exposed to supraphysiologic androgen, adaptively increased nuclear AR is over-stabilized, preventing sufficient degradation in mitosis, inhibiting DNA re-licensing, and thus death in the subsequent cell cycle. These mechanistic results and the fact that AR/RC binding occurs in metastatic CRPCs directly from patients provides a paradigm shifting rationale for bipolar androgen therapy (BAT) in patient progressing on chronic androgen ablation. BAT involves giving sequential cycles alternating between periods of acute supraphysiologic androgen followed by acute ablation to take advantage of vulnerability produced by adaptive auto-regulation and binding of AR to RC in CRPC cells. BAT therapy is effective in xenografts and based upon positive results has entered clinical testing.

Figure 1. AR and RC protein expression in human prostate cancer cell lines and antibody validation for IP and IB analysis

(A) IB of AR protein in AR-expressing LNCaP, CWR22Rv1, LAPC4, and VCaP cells compared to AR-non expressing PC-3 prostate cancer cells using an N-terminal specific anti-AR antibody. Nuclear protein extracts from 10⁵ cells were loaded per well for each of the lines. (B) IB of AR in cytosolic (denoted cyto. extract) vs. nuclear (denoted nucl extract) extract from 10⁵ LNCaP cells. (C) IB of Orc1, Orc2, Cdc6, Cdt1, and Mcm2 in nuclear extracts from 10⁵ LNCaP cells. (D) IB for indicated protein in cytosolic (denoted cyto extract) vs. nuclear (denoted nucl extract) extracts and IB of IP using specific antibody against the indicated protein (denoted IP) or using non-specific IgG (denoted IP:IgG) from nuclear extracts of LAPC4 and LNCaP cells. (E) IB of IP using non-specific IgG (denoted IP:IgG) vs. specific antibody against the indicated protein (denoted IP) vs. flow through from the specific IP (denoted IP:FT) from nuclear extracts of LNCaP and PC-3 cells.

Figure 2. AR is characteristically increased in human castrate resistant prostate cancer (CRPC) cell lines and CRPC clinical specimens

(A) Western blots of AR protein in EpCam flow sorted human normal human prostate epithelial cells without culturing (i.e., denoted N-PrEC), localized prostate cancer cells without culturing (i.e., denoted L-PCA) and a series of human prostate cancer xenografts (i.e., PC82 and CWR22) and prostate cancer cell lines (i.e., E006AA and PacMetUT1) from hormonally naïve patients and CRPC cell lines from androgen ablated patients (i.e., LNCaP, LAPC4, VCaP, MDA-PCA-2b). Number below each lane is the relative level of AR protein for the indicated prostate cancer xenograft/cell line normalized to AR expression in N-PrECs. Upper panel are the results of long film exposure time to detect AR in N-PrEC cells so that its densitometry level can be used to normalize AR expression in the indicated samples. Lower panel are the results of short film exposure time to determine the relative AR levels in 4 human CRPC cell lines (i.e., LNCaP and LAPC-4 derived from lymph node metastases and VCaP, and MDA-PCA-2b derived from bone metastases, all from different patients progressing on androgen ablation therapy). (B) Level of AR m-RNA expression in normal prostate tissues from 23 organ donors (denoted Normal Donor), normal prostate tissues from radical prostatectomy specimens from 6 hormone naïve patients (denoted RRP-N), localized prostate cancer from radical prostatectomy specimens from 30 hormone naïve patients (denoted RRP-C), and 18 metastases from 6 castration resistant patients obtained at rapid autopsy (denoted Met).

Figure 3. In Vivo adaptive AR auto-regulatory response of castrate resistant CWR22RH cancer to varying androgen environments

(A) In vivo growth response of PC82 and CWR22 tissue xenografted into intact (n=10) vs. castrated (N=10) mice. For both cancers, all intact animals developed growing tumors, while only a single castrate resistant cancer (i.e. termed CWR22RH) grew in one of the castrates inoculated with CWR22. Mean tumor volumes for the intact PC82 and CWR22 groups vs. the growth of the single CWR22RH tumor which developed in a castrated host are presented. (B) AR level in parental CWR22 growing in intact host vs. CWR22RH grown in intact vs. castrated host. Number below blot is the relative AR expression vs. CWR22 in intact host. (C) Growth of CWR22RH in intact (n=8) vs. castrate (n=8) hosts. (D) Growth of CWR22-Rv1(4X/AR) cells transduced to express 4fold higher AR than parental cells in intact (n=5) vs. castrated (N=5) host.

Figure 4. Growth and AR response of parental LNCaP cells vs. low androgen adapted LNCaP/A- cells to acute exposure to either anti-androgen or androgen supplementation

(A) In vitro growth response of LNCaP cells to acute exposure to 10uM of the anti-androgen, bicalutamide or 10nM of the synthetic androgen, methyltrienolone. (B) Western blot of AR in LNCaP cells after indicated treatment. Numbers below each lane are the relative level of AR in untreated vs. bicalutamide or methyltrienolone treated LNCaP cells. (C) Western blot of AR in LNCaP vs. LNCaP/A- cells. Numbers below each lane are the relative level of AR in LNCaP vs. LNCaP/A- cells. (D) In vitro growth response of LNCaP/A- cells to acute exposure to 10uM of bicalutamide or 10nM of methyltrienolone.

Figure 5. AR binds RC proteins in AR-dependent but not AR-independent human castrate resistant prostate cancer cells

(A) Average base pair (bp) size of DNA in nuclear extract. (B) IB of AR in nuclear extracts (denoted nucl extract) and from co-IP using a non-specific IgG antibody (denoted IgG) vs. antibody specific for either Orc2, Mcm2, or Cdc6 on nuclear extracts from indicated human prostate cancer line (C) IB for Orc2 and Mcm2 from co-IP using an AR specific antibody on nuclear extracts of CWR22Rv1 cells. (D) IB for Orc2 and Mcm2 in nuclear extracts (denoted nucl extract) and from co-IP using an AR specific antibody (denoted AR) vs. a non-specific IgG (denoted IgG) on nuclear extracts from E006AA cells.

Figure 6. AR binding to Orc2 is chemically cross-linkable and present in castration resistant metastatic prostate cancer tissue directly from patients

(A) LNCaP cells were treated for 30 minutes with 0.5mM of the cell-permeable cross-linker dithiobis[succinimdylpropionate] and whole cell extracts made with RIPA buffer. IB for AR and Mcm2 from co-IP using a non-specific IgG (denoted IgG) vs. Orc2 specific antibody (denoted ORC2) on these RIPA buffer whole cell extracts. (B) IB for Orc2 and AR in nuclear extracts from human metastatic prostate cancer tissues obtained from indicated sites at rapid autopsy. Histone H3 (denoted HH3) expression was evaluated as a marker of nuclear extract efficiency. (C) IB of AR from co-IP using non-specific IgG (denoted IP:IgG) vs. Orc2 specific antibodies on nuclear extracts of the metastatic prostate cancer tissue with the highest nuclear AR expression.

Figure 7. AR binds with RCs in early-G₁ in AR-dependent castration-resistant prostate cancer cells

(A) Temporal change in expression detected by IB of Orc2, Cdt1, Mcm2, AR, Orc1, Cyclin D₁, Cdk4, Cdc6, and Cdk2 in LNCaP cells progressing through cycle after release from early-G₁ arrest. Protein lysates from 10⁵ cells were loaded per lane. Expression in asynchronously growing LNCaP cells (denoted Asynchro). (B) IB of AR from nuclear co-IP using non-specific IgG (denoted IgG) vs. specific antibody to either Orc2 or Cdc6 in asynchronously growing LNCaP cells (denoted control) vs. cells arrested in early-G₁ (denoted 0hr) or 8, or 24 hr post-release. (C) IB of AR and ORC2 from nuclear co-IP using non-specific IgG (denoted IgG) vs. specific antibody to Orc2 in asynchronously growing LNCaP cells (denoted bicalutamide -) vs. cells treated for 4 days with 10uM bicalutamide (denoted bicalutamide+). (D) IB of AR and ORC2 from nuclear extracts of bicalutamide untreated (denoted-) vs. treated (denoted+) from (C). (E) IB of AR and ORC2 from nuclear co-IP using specific antibody to Orc2 on LNCaP cells treated for 4 days with 10uM bicalutamide.

Figure 8. Morphologic, Ki-67, and AR Response of casodex resistant LNCaP/A-to Bipolar Androgen Therapy

(A) In vivo growth response of LNCaP/A- cells to bipolar androgen therapy. Castrated male NOG mice were inoculated with LNCaP/A- cells and half (n=10) of the mice received nothing further and the other half of the castrated animals (n=10) received bipolar androgen therapy consisting of implantation with testosterone filled capsules (n=10) which rapidly produce a super-physiologic level of androgen. After 2 weeks of a high level, the serum androgen was then rapidly reduced to a castrate level by removing the T-implants and after 2 additional weeks of low androgen, the animals were re-implanted with T-capsules. H&E histology (B) in untreated castrates vs. (C) castrates treated with bipolar androgen therapy as described in (A). Black arrows indicate cancer mitoses and white arrows indicate dying cancer cells. Cancer cell Ki-67 expression is high in both in untreated castrates (D) vs. castrates treated with bipolar androgen therapy (E). AR is undetectable in cancer cell mitosis, denoted by circle, in untreated castrates (F) vs. AR expression in mitosis, denoted by circle, in castrates treated with bipolar androgen therapy (G).