Prostate cancer androgen receptor activity dictates efficacy of Bipolar Androgen Therapy

Cell culture and reagents

LNCaP and VCaP cell lines were obtained from American Type Culture Collection (ATCC). LAPC4 and 22Rv1 cell lines were a gift from J Isaacs. LNCaP, LAPC4, and 22Rv1 were grown in RPMI 1640 (Gibco; 11835-055) supplemented with 10% fetal bovine serum (Corning), sodium lactate 1.6mM, sodium pyruvate 0.5mM, Lalanine 0.43mM, 1% pen-strep (Gibco). VCaP was grown in DMEM (ATCC 30-2002) supplemented with 10% fetal bovine serum (Corning) and 1% pen-strep (Gibco). Full-length AR cDNA was cloned into the BAMH1/Sal1 site of pLenti-CMV-GFP-Puro (Addgene; 17448, E Campeau and P Kaufman laboratories) and empty vector control was generated through excision of GFP of this plasmid. These vectors, along with pCDH-puro-cMyc vector (Addgene; 46970, J Wang laboratory), pCDH-EF1-FHC empty vector control (Addgene, 64874, R Wood laboratory), piSMARTvector-PGK-TurboGFP-TRE3G-shAR vectors (Horizon; V3IHSPGG_8216343 and V3IHSPGG_7292640) were transfected into 293T cells (ATCC) along with pMD2.G (Addgene; 12259) and psPAX2 (Addgene; 12260) packaging vectors using lipofectamine (Invitrogen) to produce AR-puro, cMyc-puro, control-puro, and Tet-On-shAR-puro lentivirus particles. Two days after transduction with indicated virus, vector-expressing cells were selected with puromycin 1ug/ml for 72 hours. To constitutively express MYC in LAPC4-AR cells, LAPC4-AR cells were transfected with pCDNA3-HA-HA-humanCMYC vector (Addgene; 74164, M Roussel laboratory) using lipofectamine (Invitrogen) then selected for stable expression with G418 (Sigma) for 7 days. Cells were maintained at 37 °C in 5% CO2. They regularly tested negative for mycoplasma contamination using MycSensor PCR Assay kit (Agilent Technologies). R1881 was obtained from Sigma, enzalutamide from Selleck Chem, and testosterone cypionate from Steraloids.

Proliferation, cell viability, clonogenic survival, and cell cycle analyses of cell lines. For proliferation and cell viability assays, cells were plated in triplicate on 6-well (0.1 × 10⁶ cells/well) or 12-well (0.05 × 10⁶ cells/well) plates and incubated with R1881, enzalutamide, or vehicle control for indicated duration. Cells were counted using a hemocytometer with viability assessed by trypan blue exclusion. For clonogenic survival assessment, cells were plated on 6-well plates at low density (1500 cells/well) in 1 ml fresh media and 250 ul conditioned media obtained from a confluent flask of the parental cell line. Colonies were stained with crystal violet (Sigma; 0.5% in 20% methanol) after 10-20 days. For cell cycle analysis, cell pellets were resuspended in cold 70% ethanol at least overnight, then subsequently washed with PBS, and resuspended in 50 ug/ml propidium iodide (Sigma) and 100 ug/ml RNAse (Sigma) and run on the BD FACSCelesta flow cytometer with analysis using FlowJo software 10.4.2.

Western blot analyses

Cells or tissues were lysed with 1x denaturing lysis buffer (Cell Signaling technology) containing protease and phosphatase inhibitors (Roche). Protein concentration was determined using the Pierce BCA Protein Assay kit (Thermo Scientific), and 5-20 ug of lysate was resolved on a SDS-PAGE gel and transferred to a nitrocellulose membrane. Membranes were blocked with 5% milk for 1 hour, then incubated with primary antibody overnight. Primary antibodies used were: anti-cMYC (Abcam, clone Y69, 1:1000 dilution), anti-AR (Santa Cruz, sc7305, 1:1000 dilution), anti-PSA (Dako, A0562, 1:1000 dilution), anti-vinculin (Millipore, clone V284, 1:2000 dilution). Anti-rabbit IgG, HRP-linked Antibody (CST, 7074S, 1:5000 dilution) and Anti-mouse IgG, HRP-linked Antibody (CST, 7076S, 1:5000 dilution) were used as secondary antibodies.

Quantitative real-time polymerase chain reactions

RNA was extracted using the RNeasy kit (Qiagen) and cDNA generated using the high-capacity cDNA reverse transcription kit (ThermoFisher). RT-PCR were performed in triplicate using 500ug cDNA, 10ul TaqMan Gene Expression Master Mix (ThermoFisher), and 1ul 20x TaqMan Gene Expression Assay probe mix for MYC (Hs00153408_m1), AR (Hs00907244_m1), KLK3 (Hs02576345_m1), and ACTB (Hs01060665_g1) (ThermoFisher) on an ABI7500 RT-PCR System (ThermoFisher). Relative gene expression was determined by delta-delta CT.

Clinical trial design and procedures

The COMbination of Bipolar Androgen Therapy and Nivolumab (COMBAT-CRPC; NCT03554317) clinical trial was a single-arm, multicenter, open-label phase II study of BAT in combination with the anti-PD1 agent nivolumab for patients with metastatic CRPC that had progressed on at least one novel androgen receptortarget therapy. This study was approved by the Institutional Review Board at Johns Hopkins, and all accrued patients provided written informed consent. The inclusion and exclusion criteria and pre-specified study end points were previously described [31]. Patients were required to have soft tissue metastases amenable to biopsy to participate. Patients were treated with 3 cycles of BAT (testosterone cypionate 400 mg intramuscular every 28 days) followed by concurrent BAT and nivolumab 480 mg intravenously every 28 days until progression. Paired core-needle tumor biopsies were performed prior to treatment and after 3 cycles of BAT monotherapy. Response was assessed with PSA at each cycle, and CT chest, abdomen, and pelvis and technetium-99 bone scan every 3 cycles.

Chromogenic Immunohistochemistry (IHC) for AR

Tissue sections were baked on a 60°C hotplate for 10 min. and then deparaffinized in xylenes. Tissues were rehydrated in graded alcohol solutions and distilled water (diH₂O). Slides were then rinsed in a solution of diH₂O with 0.1% Tween and placed in a citrate bucket wash for 1 min. Tissue sections were steamed in Target Retrieval Solution (Dako; S1699, Santa Clara, CA), washed in Tris Buffered Saline with Tween (TBST; Sigma), and subjected to a Dual Endogenous Enzyme Block (Dako; S2003). The tissues were incubated with the primary antibody (conditions varied for each antibody) and, rinsed with TBST, and incubated with the secondary antibody, PowerVision Poly-HRP anti-Rabbit IgG (Leica, Deer Park, IL), for 30 min. at RT. Tissues were rinsed with TBST and then aminoethyl carbazole (Immpact AEC; Vector Labs; SK-4205, Burlingame, CA) was applied for 20 min. at RT. Slides were washed with TBST and counterstained with hematoxylin (Mayers; Dako; S3309; diluted 1:4). Slides were then coverslipped with VectaMount AQ Aqueous Mounting Medium (Vector Laboratories; H-5501) and scanned on a Ventana DP200 Digital Whole Slide Scanner (Roche Diagnostics, Rotkreuz, Switzerland) using a 40X microscope objective. Coverslips were removed with diH₂O and AEC was dissolved with graded levels of alcohol. Slides were then steamed and incubated with an SDS-2ME solution as previously described [32]. Slides were placed in TBST for 5 min. and then steamed, after which the dual endogenous block was reapplied and the tissue was incubated with the next antibody. This process was repeated until each tissue had been iteratively stained with the following antibodies:

As a control, after primary antibody removal, for each of the antibody staining rounds in a number of experiments we performed staining by leaving out the primary antibody and performing a round of staining with the secondary antibody only. These experiments showed a complete absence of signals indicating a complete antibody removal.

Quantitative Image Analysis for AR Protein

Image analysis was performed using the MultiPlex module on HALO 3.2 (Indica Labs). Whole slide scans of each staining round for AR and CK8 were imported into HALO and registered with one another. A patientspecific Random Forest classifier was manually trained to classify all regions of each CK8-stained tissue into one of three categories: tumor, non-neoplastic tissue, and background. Annotations created from this classifier were then mapped onto the scans of the AR-stained tissues, using HALO’s “Classify Registered” feature. In the event that the scan of the CK8-stained tissue could not be successfully used for tumor classification (low CK8 expression, degraded tissue or imprecise scan registration), a patient-specific tumor classifier was trained on the scan of the AR-stained tissue. If the AR signals at a dilution of 1:10,000 was too weak for classifier training, scans of the tissues stained for AR at 1:5,000 were utilized to train the classifier. AR positivity thresholds were determined using the real-time tuning window. The MultiPlex analysis was restricted to tumor positive regions (using the CK8-trained classifier) and provided several measurements for each tumor-positive cell within each AR-stained tissue, including nuclear AR optical density (OD), cytoplasmic AR OD, nuclear area (µm²), cytoplasmic area (µm²), and cell area (µm²). Whole cell average AR ODs were determined by calculating an integrated cell OD for each cell and dividing that value by each cell’s respective area.

Immunohistochemistry (IHC) for MYC and Ki67 Proteins

IHC on patient biopsy samples for MYC and Ki67 was performed as previously described [18,33]. IHC on SKCaP PDX tissue was performed by the Sidney Kimmel Comprehensive Cancer Center (SKCCC) IHC Core facility. Primary antibodies were anti-MYC (Epitomics, Clone Y69, 1:200) and anti-Ki67 (AbCam, Clone SP6, 1:200), and tissues were stained with the Ventana Discovery Ultra Autostainer, using the Discovery anti-HQ HRP kit.

Quantitative Image Analysis for MYC and Ki67 Proteins

For quantitative image analysis, whole IHC stained biopsy slides were scanned using a 40X objective on a Ventana DP200 slide scanner. Whole slide scans were imported into HALO 3.1 image analysis platform. The nuclear staining for both MYC and Ki67 were analyzed using the cytonuclear module v 1.6. Tumor regions were delineated by a pathologist (CG-A) using manual annotations, guided by adjacent H&E and androgen receptor stained slides. Positively stained (DAB staining above minimum intensity threshold) for MYC and Ki67 cells were binned into 3 intensity levels. We calculated the percentage of positively stained cells for both Ki67 and MYC proteins, as well as an H-SCORE for MYC using the intensity levels with the following formula (percentage of intensity level 3 cells x 3)+(percentage of intensity level 2 cells x 2)+ (percentage of intensity level 1 cells x 1) to give a total possible score of 300. Percentage positively stained cells were validated on a number of samples by manual counting.

Frozen biopsy specimens for laser capture microdissection (LCM) for bulk RNAseq analysis

Frozen sections were cut onto PEN (polyethylene naphthalate) slides (Leica), stained with hematoxylin and LCM was performed at the SKCCC Core facility using a Leica LMC7000 laser capture microdissection system in the SKCCC Cell Imaging Core. Laser captured tissues from regions highly enriched for tumor cells were collected by manual annotations of regions of interest that outlined tumor nests. RNA was isolated using the ALLPREP RNA/DNA extraction protocol (Qiagen Cat. No. 802804). RNA amounts were measured using a Qubit fluorometer (Invitrogen) and quality were measured using an Agilent Bioanalyzer 2100 with the Pico Chips Kit.

RNA sequencing

For RNA sequencing of patient biopsy samples, purified RNA was provided to the SKCCC Experimental and Computational Genomics Core to carry out their low-input RNA-seq workflow as described previously with some modifications [34]. Briefly, quality of total RNA was measured by the Agilent Bioanlayzer to determine RNA integrity (RIN). Samples with starting input between 100pg-100ng of total RNA and RIN > 7.0 were considered to have sufficient quality to proceed to construction of whole transcriptome sample-barcoded libraries using the Ovation RNA-Seq System V2 according to the manufacturer’s protocols (Nugen).

Quantification of the libraries was performed by qPCR or by the Agilent Bioanalyzer and equimolar concentrations of each library were pooled together, clustered and sequenced on an Illumina Novaseq 6000 platform, with paired end sequencing. The resulting reads were aligned the human reference genome build hg38 using RSEM [35] with the STAR aligner [36] to obtain posterior mean estimate transcripts per million (pmeTPM) as a measure of gene expression. pmeTPM values for each gene were normalized using upper quantile normalization across samples and log2 transformed. For RNA sequencing of cell lines, purified RNA was sent to Admera Health who performed library preparation with polyA selection, sequencing on the Illumina HiSeq 2500 with read length configuration of 2×150bp with 60 million reads per sample, and read mapping to GRCM38. Sequenced libraries were processed with deepTools [37], using STAR [36], for trimming and mapping, and featureCounts [38] to quantify mapped reads. Raw mapped reads were processed in R (Lucent Technologies) with DESeq2 [39] to generate normalized read counts to visualize and cluster as heatmaps using pheatmap [53] and determine differentially expressed genes with greater than 2-fold change and lower than 0.01 adjusted p value. Gene set enrichment analysis was performed using fGSEA [40].

ARA_MW Score Calculation

We calculated the AR activity score using Mann-Whitney ranking (ARA_MW) by adapting a previously reported approach [41]. First, genes were selected based on previous reports [42] and examined for co-linearity based on the pearson correlation coefficient and testing for its significance using the cor and cor.test functions in R, respectively. Having observed no correlation between most genes, we selected 10 genes (KLK2, KLK3, FKBP5, STEAP1, STEAP2, PPAP2A, RAB3B, ACSL3, NKX3-1, TMPRSS2) and applied the adapted gene scoring method to RNAseq results from COMBAT biopsies, and to polyA captured RNAseq results from 266 biopsies from participants in the SU2C/PCF study [24] retrieved from cBioPortal [43]. Briefly, we used the log transformed quantile normalized pmeTPM values and calculated a ranked list of genes for each biopsy to generate a g x b matrix R of ranked genes (with g genes and b biopsies), considering only the top 1500 highest ranked genes by setting r_g,b = r_max + 1, with r_max = 1500. The ARA_MW score was then calculated for the n selected genes as: where U is the Mann-Whitney U statistic calculated as: and R^′ is a subset of R including only the n selected genes in the score. In this manner, the score was independent of sequencing depth, normalization strategy and dataset composition, and was primarily based on the relative abundance of selected genes within each studied sample.

Clinical data from 266 participants in the SU2C/PCF was retrieved from cBioPortal and participants classified based on ARA_MW. Differences between classified patients for each clinical parameter was examined using a Wilcoxon ranked sum test.

ATAC sequencing

Cells were provided to the Single Cell and Transcriptomics Core at Johns Hopkins University who performed nuclei isolation, library preparation, and sequencing with read length configuration of 2×75 with 50 million reads per sample. Briefly, 5×10⁴ cells were washed in PBS and then lysed in 10 mM Tris-HCl, pH 7.4,10 mM NaCl, 3 mM MgCl₂ and 0.1% Igepal CA-630 (all SIGMA). Nuclei were then spun down and then resuspend in 25 μL TD (2x reaction buffer), 2.5 μL TDE1 (Nextera Tn5 Transposase) and 22.5 μL nuclease-free water, incubated for 30 min at 37°C. Purified DNA was PCR-amplified, then cleaned from adaptors. Sequenced libraries were trimmed with Trimmomatic [44], mapped using Bowtie2 [45], and processed with SAM tools [46] to retain open chromatin fragments of less than 150kbp. Coverage files were generated with deepTools [37]. Open chromatin and differentially regulated peaks were detected with MACS2 [47] with a p value < 1 × 10⁻⁷ and a q value of less than 0.1 and DiffBind [48] using a 2-fold change and 0.01 p value as significance thresholds. Bed files were analyzed with Bedtools [49], and visualized alongside coverage files on IGV [50]. Reference H3K27Ac CHIP-seq on human VCAP data was obtained from the ENCODE project [51].

RNA in situ hybridization (RISH)

RISH was performed as previously described [52].

Mouse models and tumor studies

Male NSG mice aged 8-12 weeks were obtained from the Sidney Kimmel Comprehensive Cancer Center (SKCCC) Animal Core Facility and surgically castrated. SKCaP patient-derived xenograft tissue was minced, mixed with Matrigel (BD Biosciences) and implanted subcutaneously on the flank. Testosterone was administered by implantation of a slow-release subcutaneous pellet in the opposite flank. Pellets were assembled using 25 mm sections of Silastic Laboratory tubing, filled with 30 mg testosterone cypionate (Steraloids INC A6960-000), sealed on both ends with Silastic Medical Adhesive Type A, then sterilized. Enzalutamide (Selleck Chem) was administered by oral gavage 10 mg/kg/day in 200 ul 1% carboxymethyl cellulose, 0.1% Tween-80, 5% DMSO. Tumors were measured twice weekly using microcalipers, and tumor volume was calculated using the following formula: 0.5236 x L x W x H. At study completion, mice were euthanized and tumors were extracted. Tumors were flash frozen for subsequent lysis for immunoblot analyses and formalin-fixed for subsequent immunohistochemistry analyses. All mice were housed in the Johns Hopkins animal facility and we have complied with all relevant ethical regulations in accordance with Johns Hopkins Institutional Animal Care and Use Committee.