Abstract
Background/Aim: We aimed to evaluate the changes of androgen receptor (AR) signaling-related long non-coding RNAs (lncRNAs) in serum extracellular vesicles (EVs) from prostate cancer (PC) patients, in order to identify novel biomarkers for AR axis-targeted therapy (ARAT)-resistance among castration-resistant PC (CRPC) patients. Patients and Methods: EVs were isolated from 2 patients before and after acquiring ARAT-resistance. RNA profiling of EVs was performed by RNA-sequencing. The expression levels of selected lncRNAs in EVs were analyzed by digital droplet PCR (ddPCR) in 58 localized and 14 metastatic PC patients at diagnosis, 7 ARAT-naïve and 6 ARAT-resistant CRPC patients. LncRNA H19 expression in PC tissue was examined using published data. In order to analyze the role of H19, the prognosis was analyzed in PC patients and proteomic analysis was performed in 22Rv1 PC cells. Results: RNA-sequencing revealed that AR-regulated RNAs were most enriched in EVs after acquiring ARAT-resistance. Among them, up-regulation of AR signaling-related lncRNAs (PCAT1, H19, HOXA-11AS, ZEB1-AS1, ARLNC1, PART1, CTBP1-AS and PCA3) was confirmed by ddPCR. H19 contained in EVs (EV-H19) was significantly increased among ARAT-resistant patients compared to ARAT-naïve CRPC or metastatic PC patients. In PC tissue, H19 was negatively correlated with AR protein and AR-activity score and up-regulated in neuroendocrine CRPC tissue with low AR expression. Furthermore, EV-H19 expression was significantly associated with worse outcome to androgen-deprivation therapy. Proteomic analysis demonstrated that H19 knockdown enhanced PC-related protein expression. Conclusion: EV-H19 may negatively correlate with AR-signaling activity and could be a marker to diagnose ARAT-resistance among CRPC patients.
Prostate cancer (PC) is the most commonly diagnosed male malignancy in the United States, with 268,490 new cases and the second leading cause of death with 34,500 deaths in 2022 (1). The five-year overall survival rate among localized and regional PC patients is over 99%, however, and it decreases to 31% among metastatic PC patients (1). Androgen receptor (AR)-signaling is a central axis in PC pathogenesis and androgen-deprivation therapy (ADT) is a common treatment for metastatic PC patients. ADT is temporarily effective; however, PC develops castration-resistant PC (CRPC) following ADT. It has been reported that AR-signaling is still activated in CRPC (2). For this reason, second generation androgen receptor axis-targeted therapy (ARAT; Abiraterone, Enzalutamide, Apalutamide and Darolutamide) is strongly recommended for CRPC patients by the National Comprehensive Cancer Network (NCCN) guidelines (3). Although ARAT improves the prognosis of CRPC, it has limited effect and CRPC patients develop ARAT-resistance (4). Prostate-specific antigen (PSA) is a reliable tumor monitoring marker for PC; however, it sometimes deviates from the disease state in CRPC patients who are receiving ARAT (5). Therefore, it is of great importance to identify novel biomarkers to diagnose ARAT-resistance accurately. AR mutations and splicing variants are one of the known mechanisms of ARAT-resistance (4, 6). Pathological evidence of AR signaling-related gene expression change may support treatment selection; however, serial biopsy of primary or metastatic PC site is invasive and not routinely done.
Extracellular vesicles (EVs) are small vesicles that are secreted from cells to body fluids and contain cellular components such as protein and RNA, thus including the cell information; they are considered as a promising biomarker for various diseases including PC (7). Recent transcriptomic analysis revealed that there was a positive correlation between RNA expression in EVs (EV-RNA) and that in PC tissue (8). Del Re et al. reported that AR splice variant 7 (AR-V7) mRNA contained in plasma EVs, which lacks the androgen binding domain and is in part involved in ARAT-resistance, was a predictive biomarker for ARAT treatment (9). Therefore, EV-RNA can provide PC information including AR-signaling and ARAT-resistance as a less invasive blood-based test.
Long non-coding RNAs (lncRNAs) are non-protein coding RNAs that are over 200 base pairs. In the past few decades, high-throughput RNA-sequencing (RNA-seq) revealed that ten thousand of lncRNAs are present, much more than protein coding mRNAs (10). LncRNAs have been implicated in cancer development such as cell growth, invasion, metastasis, anti-apoptosis, mis-regulated cell cycle, etc. through binding to proteins or sponging to miRNAs (10). In PC, some lncRNAs (e.g., PCAT1, PCAT14, MEG3, MALAT1, HOTAIR) have been reported as potential diagnostic markers or therapeutic targets (11). As for lncRNAs contained in EVs (EV- lncRNAs), only 5 EV-lncRNAs (PCA3, MALAT1 and lncRNA-21p in urine, SAP30L-AS1 and CChLAP1 in plasma) have been reported as potential diagnostic markers for PC (12-15) and EV-lncRNAs to diagnose ARAT-resistance in CRPC patients have not yet been identified.
In this study, we aimed to identify novel EV-based AR signaling-related biomarkers for ARAT-resistance with a special interest in lncRNAs. We performed RNA-sequencing of EVs isolated from serum of the same CRPC patients before and after acquiring ARAT-resistance and confirmed the expression changes of selected lncRNAs by digital droplet PCR (ddPCR) in a cohort of PC patients. We then focused on a particular EV-lncRNA and analyzed its expression, function and clinical significance.
Patients and Methods
Patients. This study was approved by the ethics committee of Fujita Health University (# HM22-032), Asahi University Hospital (# 2019-10-06) and Central Japan International Medical Center (# 2022-012-2). Written informed consent was obtained from all patients. Serum was collected from 2 CRPC patients before ARAT-administration and after acquiring ARAT-resistance for RNA-seq to compare EV-RNA expression (Table I). To confirm EV-RNA expression changes, sera collected between January 2020 and October 2022 from 85 PC patients were used, which included 58 therapy-naïve localized PC (localized PC), 14 therapy-naïve metastatic PC (metastatic PC), 7 ARAT-naïve CRPC (CRPC) and 6 ARAT-resistant CRPC (ARAT-resistant) patients. Patient characteristics are summarized in Table II. Two patients, whose serum was used for RNA-seq, were also included in the confirmation study. To evaluate progression free survival, we compared EV-lncRNA expression in 28 patients who received ADT as a first-line therapy. Serum was centrifuged at 1,500 × g for 10 min to remove cells. Samples were then stored at −80°C until use. CRPC- or ARAT-resistance was diagnosed by the increasing PSA level or radiographic progression with castration level of serum testosterone (<50 ng/dl), respectively.
Patient characteristics of 2 patients subjected to RNA-seq and 8 selected androgen receptor (AR)-related lncRNAs whose expression was upregulated after acquiring AR axis-targeted therapy (ARAT)-resistance.
Patient characteristics.
EV isolation. Highly-purified EVs were prepared by PS-affinity method as we described previously (16, 17). In this study, we isolated EVs using MagCapture™ Exosome Isolation Kit PS Ver. 2 (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) according to the manufacturer’s instructions. Briefly, serum or cell culture medium was centrifuged at 12,500 × g to remove large microvesicles. The supernatant was incubated for 1 h at room temperature with Tim4-conjugated beads. EV-captured beads were directly used for subsequent experiments without releasing EVs.
Total RNA extraction and complementary DNA synthesis. Total RNA was extracted using the miRNeasy mini kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s protocol. EV-RNA isolated from 1,000 μl of serum was used for RNA-seq and that from 500 μl of serum or cell culture medium was used to confirm RNA expression by reverse transcription (RT)-ddPCR or quantitative RT-PCR (qRT-PCR). Complementary DNA (cDNA) synthesis was performed using the iScript cDNA synthesis kit (BIO-RAD, Hercules, CA, USA). For RT-ddPCR, 15 out of 20 μl of the total volume containing total RNA isolated from serum EVs was subjected to cDNA synthesis. For qRT-PCR, 500 ng of RNA from cells or 10 ng of EV-RNA from cell culture medium was subjected to cDNA synthesis after determination of RNA concentration. cDNA was stored at −20°C until use.
RNA-sequencing. RNA-seq was conducted at the TAKARA Bio incorporation (Kusatsu, Japan). Briefly, cDNA was synthesized using random primer and a specific sequence was assigned to the end of the 1st strand cDNA by the SMART method (18). Then, using the specific sequences, samples were amplified by PCR using indexed primers having different tag sequences for each sample. The obtained PCR product was purified by the magnetic bead method using AMPure XP (Beckman Coulter, Brea, CA, USA), and the ribosomal cDNA was cleaved. The uncleaved cDNA was PCR-amplified using adapter-specific primers to form a sequencing library. The library quality was determined by Agilent 2100 BioAnalyzer (Agilent Technologies, Santa Clara, CA, USA). The strand-specific RNA-seq libraries were sequenced using the NovaSeq 6000 platform (Illumina, San Diego, CA, USA). The results of RNA-seq were analyzed using iDEP.96 (http://bioinformatics.sdstate.edu/idep96/) and Metascape software (https://metascape.org/gp/index.html#/main/step1). Based on the result of RNA-seq, 8 AR signaling-related lncRNAs were selected to confirm expression changes (Table I).
Droplet digital PCR. To determine the expression levels of AR and AR-V7 mRNA as well as 8 lncRNAs, Taqman probes were purchased from Thermo Fisher Scientific (Waltham, MA, USA, AR; Hs00171172_m1, AR-V7; AIQJDUH, PCAT1; Hs04275386_s1, H19; Hs00399294_g1, HOXA11-AS; Hs03296531_m1, ZEB1-AS; Hs01398645_g1, ARLNC1; Hs01596342_m1, PART1; Hs00213199_m1, CTBP1-AS; Hs04401800_m1, PCA3; Hs01371939_g1, GAPDH; Hs03929097_g1). RT-ddPCR was performed using QX200 AutoDG Droplet Digital PCR system (BIO-RAD), where 1.1 μl of cDNA was used to generate each droplet. The following conditions were used for PCR: 95 °C × 10 min, 94 °C × 30 s, 60 °C × 1 min (60 cycles), 98 °C × 10 min, 10 °C hold. The fluorescence signal of the PCR reaction was measured using a droplet reader. Absolute quantification was determined using QuantaSoft software (BIO-RAD). Then, RNA copy number contained in EVs was calculated. Since a normalization method for EV-RNA has not yet been established, the expression level of each RNA in EVs was represented as EV-RNA copies in 1 ml of serum (copies/ml) in this study.
Database analysis. Databases of The Cancer Genome Atlas (TCGA) and Weill Cornel Medicine (WCM) were downloaded from cBioPortal for cancer genomics (https://www.cbioportal.org). Based on the database of TCGA, lncRNA H19 expression was compared to AR protein and AR activity score determined by the expression pattern of 20 AR transcriptional targets (19). We used WCM cohort to compare AR mRNA and H19 expression among CRPC specimens with adenocarcinoma phenotype (CRPC-Adeno) and neuroendocrine (NE) phenotype (CRPC-NE).
Cell culture. Human CRPC C4-2B cells were obtained from the MD Anderson Cancer Center (Houston, TX, USA) and human PC 22Rv1 cells were from the European Collection of Authenticated Cell Cultures (ECACC, Salisbury, UK). Cells were cultured in RPMI 1640 medium supplemented with 10 % fetal bovine serum (FBS), 100 units/ml of penicillin and 100 mg/ml of streptomycin in a humidified atmosphere containing 5 % CO2.
Quantitative reverse transcription-PCR. To determine the H19 expression levels in cells, qRT-PCR was performed using Taqman fast advanced master mix (BIO-RAD). The following conditions were used for PCR: 50 °C × 2 min, 95 °C × 20 s, 95 °C × 1 s and 60 °C × 20 s (50 cycles), 4 °C hold. Cellular H19 was normalized to GAPDH mRNA, and a relative fold change was calculated using the comparative Ct method.
Small interfering RNA transfection. Two small interfering RNA (siRNA) for H19 and negative control siRNA were purchased from Dharmacon (Lafayette, CO, USA). The target sequences of the siRNAs were as follows: H19 siRNA1 (GGACGAUGGG CCUGAGCUA); H19 siRNA2 (GGAGCAGCCUUCAAGCAUU); negative control siRNA (UAGCGACUAAACACAUCA). Cells seeded onto 6-well plates and left for 24 h were transfected with siRNA for 4 h at 10 nM using the Lipofectamine™ RNAiMAX transfection reagent (Thermo Fisher Scientific), according to the manufacturer’s instructions. 48 h after transfection, total RNA was extracted in triplicate from cells and EVs were isolated from cell culture medium.
Nano liquid chromatography tandem mass spectrometry (nanoLC-MS/MS). Proteins were extracted using RIPA buffer 48 h after siRNA transfection in triplicate. Trypsin digestion prior to proteomic analysis was performed using S-Trap micro columns (ProtiFi, NY, USA) according to the instruction manual. Desalting of peptides after trypsin digestion was conducted using styrene-divinylbenzene (SDB) stage-Tip. Proteomic analysis was performed as described previously (17). All data were analyzed for protein identification by Proteome Discoverer 2.4 software (Thermo Fisher Scientific). Metascape software was used for enrichment analysis.
Statistical analysis. The student t-test was used to compare two parametric groups and Mann-Whitney U-test for non-parametric groups. One-way ANOVA with Fischer’s LSD was used to compare three parametric groups and Kruskal-Wallis test with Dunn’s multiple comparison test was used for three or more non-parametric groups. Kaplan Meier method and generalized Wilcoxon test were used to evaluate survival. The data analysis was performed using GraphPad Prism 9 software (San Diego, CA, USA). p<0.05 was considered as statistically significant and p<0.10 as a statistically significant trend.
Results
RNA-sequencing and ddPCR revealed that H19 was up-regulated in ARAT-resistant patients. Firstly, we isolated EV-RNA from the serum of 2 CRPC patients before and after acquiring ARAT-resistance, who were initially sensitive to ARAT, and then performed RNA-seq. The results showed that 2958 RNAs were up-regulated and 16 were down-regulated, respectively, by 2-fold or more after acquiring ARAT-resistance compared with ARAT-naïve (Figure 1A). Among up-regulated transcripts, those regulated by AR were the most enriched, when analyzed by the TRRUST database (http://www.grnpedia.org/trrust) (Figure 1B). We then selected 8 AR-related lncRNAs from the up-regulated lncRNAs to confirm their expression changes using ddPCR (Table I). The H19 copy number was significantly elevated in ARAT-resistant PC patients compared to metastatic PC and CRPC patients (Figure 2). The expression levels of PCAT1 and HOXA11-AS were not changed among these patients. AR, AR-V7 mRNA and other lncRNAs were not detected in these samples (data not shown).
Transcriptomic analysis before and after acquiring androgen receptor axis-targeted therapy (ARAT) resistance. (A) Heatmap of RNA-seq among 2 castration-resistant prostate cancer (CRPC) patients before and after acquiring ARAT-resistance. A total of 9323 genes are identified. (B) Upstream transcription factors that up-regulated transcripts by 2 folds or more in 2 patients after acquiring ARAT-resistance are listed in descending order of the number of transcripts each transcription factor regulated.
Androgen receptor (AR)-related lncRNAs expression among prostate cancer (PC) patients evaluated by reverse transcription-droplet digital PCR (RT-ddPCR). EV-H19 was significantly up-regulated in ARAT-resistant PC patients compared with metastatic or CRPC patients (median 100 copies/ml vs. 0 and 0, p=0.0013 and 0.0104). *p<0.05, **p<0.005, #p<0.10 using the Kruskal-Wallis test with Dunn’s multiple comparison test.
H19 is negatively correlated with AR and up-regulated in neuroendocrine CRPC among PC tissues. To evaluate the correlation between H19 and AR expression in PC tissue, we compared the H19 expression with AR protein expression and AR activity score using the TCGA cohort. We found a very weak but negative correlation between H19 and AR protein expression and AR activity score in PC tissue (Figure 3A and B). We also compared AR mRNA and H19 expression among CRPC-Adeno and CRPC-NE tissues using the WCM cohort. AR mRNA and H19 were significantly down-regulated and up-regulated, respectively, in CRPC-NE compared to CRPC-Adeno specimens (Figure 3C).
Correlation between H19 and androgen receptor (AR) expression in prostate cancer (PC) tissue. The correlation of H19 expression with AR protein (A) expression and AR activity score (B) in PC tissue is represented. Data were downloaded from The Cancer Genome Atlas (TCGA) database (n=333). (C) AR mRNA and H19 expression among 34 CRPC-Adeno and 15 CRPC-NE specimens from the Weil Cornel Medicine (WCM) cohort (p<0.0001 and p=0.0009). RNA expression is shown in log2 scale. ***p<0.001 and ****<0.0001 using the Mann-Whitney U-test.
EV-H19 expression associates with worth outcome among patients who received androgen-deprivation therapy. To evaluate the association between EV-H19 expression and PC progression, we performed survival analysis among 28 PC patients who received ADT as a first-line therapy. We found that EV-H19 was positive in 11 patients and that those positive for EV-H19 had a significantly higher probability of disease progression (Figure 4).
Progression free survival of prostate cancer (PC) patients who received androgen deprivation therapy (ADT) as a first-line therapy. Red and blue lines indicate EV-H19 positive (n=11) and negative (n=17) patients, respectively. Median progression free survival was 15.5 months in EV-H19 positive patients and undefined in negative patients (p=0.0498 using the Kaplan Meier method and the generalized Wilcoxon test).
EV-H19 correlates with cellular H19 expression and H19 knockdown enhances the PC phenotype. The 22Rv1 CRPC cells express AR-V7 protein and share some features with NE cells (20, 21). On the other hand, C4-2B cells express AR wild-type protein with low AR-V7 protein expression and are considered to be typical CRPC cells (20, 22). We compared H19 expression among these two lines. H19 was significantly higher in cells and EVs of 22Rv1 cells than in those of C4-2B CRPC cells (Figure 5A). After H19 knockdown in 22Rv1 cells, EV-H19 was also significantly reduced (Figure 5B). We then performed proteomic analysis to investigate H19 function. 114 proteins were significantly up-regulated and 87 significantly down-regulated after H19 knockdown using both siRNAs, respectively (Table III). Among up-regulated proteins, those related to “Prostatic Diseases” and “Metastasis from malignant tumor of prostate” were enriched, when analyzed using the DisGeNET database (https://www.disgenet.org) (Figure 5C and D). H19 knockdown did not affect AR protein expression (Table III).
Effects of H19 knockdown on EV-H19 expression and proteomic changes in 22Rv1 cells. (A) Cells (2.0×105) cultured in serum-free medium were seeded onto 6-well plates. 72 h after seeding, total RNA was extracted from the cells and EVs were isolated from the cell culture medium. H19 was significantly higher in cells and EVs of 22Rv1 cells compared to those of C4-2B cells (n=3, 21.01- and 2.37-fold, p=0.0006 and 0.0323, respectively, student-t test). (B) Total RNA was extracted 48 h after transfection of siRNAs. H19 knockdown in 22Rv1 cells reduced EV-H19 expression (n=3, negative control vs. siRNA1; 0.239-fold, p=0.0144, negative control vs. siRNA2; 0.266-fold, p=0.0170, one-way ANOVA with Fischer’s LSD). (C) Up-regulated proteins after siRNA-mediated H19 knockdown were analyzed using the DisGeNET database, in terms of association with diseases. (D) Proteins regulated by AR are shown in blue letters and those activating AR are in red letters. *p<0.05, ***p<0.001 and ****p<0.0001.
List of proteins that significantly changed expression after H19 knockdown.
Discussion
In the present study, we aimed to identify a novel EV-based biomarker to indicate AR-signaling activity and to diagnose ARAT-resistance. As the result of RNA-seq before and after acquiring ARAT-resistance in the same patients, a quite different expression pattern of transcripts contained in EVs was observed (Figure 1). Among up-regulated lncRNAs, we confirmed that EV-H19 was elevated in ARAT-resistant CRPC patients compared to localized, metastatic and CRPC patients (p=0.0735, 0.0013 and 0.0104, Figure 2).
H19 is the first discovered lncRNA with 2.3 kb total length which is highly expressed in most tissues of the embryogenesis or placenta and declines postnatally (23, 24). In adult tissue, H19 is present in skeletal muscle, cartilage and cardiac muscle (24). H19 has also been reported to be expressed in various types of cancer and regulate oncogenic signaling pathways such as PI3K/Akt, MAPK, Wnt, JAK/STAT and NF-
B signaling (24). In colorectal cancer, H19 expression was significantly higher in cancerous tissue compared to tumor-free surrounding tissue (25). H19 expression in endocrine tumors is controversial. High H19 expression was observed in malignant pancreatic neuroendocrine neoplasms and their liver metastasis but was found to be low in non-malignant tumors (26). In PC, Singh et al. demonstrated that AR suppressed H19 transcription and that H19 was up-regulated in CRPC-NE specimens compared to CRPC-Adeno (27). CRPC-NE is a lethal phenotype of PC associated with relative resistance to AR-directed therapies and is characterized by negative expression of AR and positive expression of NE markers such as neuron-specific enolase (NSE), chromogranin A (CHGA) and synaptophysin (SYP) (21, 27-30). We confirmed that H19 expression was negatively correlated with AR protein and AR activity score in PC specimens (Figure 3A and B). Moreover, AR mRNA and H19 were significantly down-regulated and up-regulated, respectively, in CRPC-NE compared to CRPC-Adeno specimens (Figure 3C), suggesting that higher tissue H19 expression may represent lower AR-signaling activity in PC specimens such as CRPC-NE.
In serum EVs, it has been shown that H19 was significantly increased in bladder, gastric and breast cancer patients compared to control patients and decreased after tumor resection (31-33). Wang et al. also reported that EV-H19 expression was significantly correlated with H19 expression in bladder cancer specimens (31), suggesting that EV-H19 may be associated with tissue H19 expression in cancer. In this study, a similar correlation was observed for H19 expression between PC cells and EVs (Figure 5A and B), supporting the notion that EV-H19 may also reflect PC tissue information. Interestingly, EV-H19 was up-regulated in localized PC patients at diagnosis compared to treatment-naïve metastatic PC patients (p=0.0860, Figure 2A). It is also well known that the AR gene is amplified in advanced PC tissue compared to the primary site and that AR protein is increased in advanced PC (34, 35). These findings suggest that higher EV-H19 may indicate lower AR-signaling activity in PC tissue.
A previous report demonstrated that higher H19 expression in PC tissue was associated with lower progression free survival among patients who received ADT (27). In this study, we also found that EV-H19 expression was associated with worse outcome among patients treated with ADT (Figure 4), suggesting that H19 expression in PC tissue and EVs may explain insensitivity to ADT caused by reduced AR-signaling.
It has also been reported that H19 attenuated AR-signaling without changing AR protein expression, up-regulated NE-related gene expression through H19-bound PRC2 protein complex-mediated methylation and changed the phenotype from CRPC-Adeno to CRPC-NE (27). Several studies demonstrated that repeated ARAT treatment may lead to a phenotypic change from CRPC-Adeno to CRPC-NE (4, 5, 21, 30). In this study, EV-H19 was significantly elevated in ARAT-resistant patients compared to metastatic or CRPC patients (Figure 2). Furthermore, H19 expression, both in cells and EVs, was up-regulated in NE-like CRPC 22Rv1 cells compared to C4-2B CRPC cells (Figure 5A). These findings suggest that EV-H19 up-regulation may be indicative of NE differentiation and could be a marker to diagnose ARAT-resistance among CRPC patients. The NCCN guideline recommends platinum- or taxane-based chemotherapy for CRPC-NE patients. EV-H19 expression could also help select treatment options.
Lastly, we evaluated the effect of H19 knockdown on protein expression in 22Rv1 cells to investigate H19 function. Although the expression level of AR protein was not changed, prostate cancer-related proteins were enriched in up-regulated proteins (Table III, Figure 5C and D). Among them, EGFR and FOLH1 were found to be regulated by AR according to the TRRUST database (Figure 5D, blue letter). EGFR has been reported as not only an AR-regulated protein but also as an activator of AR through PI3K-Akt and MAPK signaling (36, 37). FOLH1 protein is also known as a prostate-specific membrane antigen (PSMA). PSMA is strongly expressed on the cell surface membrane of prostate adenocarcinoma, whose expression is found in 92% of metastatic prostate adenocarcinoma but lost in metastatic NE (38). Sommer et al. reported that AR inhibition using ADT or ARAT induced up-regulation of PSMA protein in LNCaP and C4-2 PC cells with mutant AR (T877A), but not in LAPC4 PC cells with wild-type AR (39). They also showed that PSMA protein levels were up-regulated in PC tissues under ADT compared with treatment-naive PC tissues from the same patients. Radiolabeled ligands of PSMA are currently used in positron emission tomography for the diagnosis of PC (68Ga-PSMA-PET) and in radioligand therapy for CRPC patients (177Lu-PSMA-RLT) (40). Thus, PSMA strongly correlates with AR-signaling and PC phenotypes and is a key diagnostic and therapeutic molecular target for prostate adenocarcinoma. On the other hand, TGFB1, also up-regulated after H19 knockdown (Figure 5D, red letter), has been reported to enhance AR-mediated transactivation via Smad activation without changing AR expression (22, 41). Shiota et al. demonstrated that a higher Gleason score and a worse outcome to ADT were observed in metastatic PC patients with TGFB1 polymorphisms (CT/TT on rs224176 and CA/AA on rs4803455) (42), suggesting that TGFB1 expression may affect the prognosis of PC patients through AR activation. These findings suggest that H19 knockdown enhances AR-signaling activity without changing AR protein expression and that H19 may regulate PC plasticity.
Conclusion
In summary, RNA-seq revealed that AR-regulated genes were enriched in EVs after acquiring ARAT-resistance. Among them, lncRNA H19 was up-regulated in EVs isolated from serum of ARAT-resistant CRPC patients. In PC tissue, H19 was negatively correlated with AR-signaling activity. EV-H19 expression was also associated with worse outcome among PC patients. Furthermore, EV-H19 correlated with cellular H19 expression. Our findings suggest that EV-H19 may negatively reflect AR-signaling activity in PC tissue and could be a novel biomarker to diagnose ARAT-resistance among CRPC patients.
Acknowledgements
We would like to thank Dr. Hiroyuki Tezuka, who is affiliated with the Open Facility Center at Fujita Health University, for his contribution to the cell culture experiment. This study was supported by JSPS KAKENHI Grant Number 21K09437, 21K09340, 23K08790 and research fund of Asahi University.
Footnotes
Conflicts of Interest
The Authors declare that there are no conflicts of interest in this study.
Authors’ Contributions
TK designed the study and experiments. TK, ST, and HE recruited patients. TK performed and analyzed the experiments. KK performed proteomic analysis. TK, KK, KM, TA, YS, HI, and MI participated in planning the study. The first draft of the manuscript was written by TK and MI. All Authors read, commented on, and approved the manuscript.
- Received May 22, 2023.
- Revision received June 28, 2023.
- Accepted July 3, 2023.
- Copyright © 2023 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
