P2X4 purinergic receptors offer a therapeutic target for aggressive prostate cancer

Abstract

Prostate cancer (PCa) remains a leading cause of cancer-related deaths in American men and treatment options for metastatic PCa are limited. There is a critical need to identify new mechanisms that contribute to PCa progression, that distinguish benign from lethal disease, and that have potential for therapeutic targeting. P2X4 belongs to the P2 purinergic receptor family that is commonly upregulated in cancer and is associated with poorer outcomes. We observed P2X4 protein expression primarily in epithelial cells of the prostate, a subset of CD66⁺ neutrophils, and most CD68⁺ macrophages. Our analysis of tissue microarrays representing 491 PCa cases demonstrated significantly elevated P2X4 expression in cancer- compared to benign- tissue spots, in prostatic intraepithelial neoplasia, and in PCa with ERG positivity or with PTEN loss. High level P2X4 expression in benign tissues was likewise associated with the development of metastasis after radical prostatectomy. Treatment with the P2X4-specific agonist cytidine 5′-triphosphate (CTP) increased Transwell migration and invasion of PC3, DU145, and CWR22Rv1 PCa cells. The P2X4 antagonist5-(3-Bromophenyl)-1,3-dihydro-2H-Benzofuro[3,2-e]-1,4-diazepin-2-one (5-BDBD) resulted in a dose-dependent decrease in viability of PC3, DU145, LNCaP, CWR22Rv1, TRAMP-C2, Myc-CaP, BMPC1, and BMPC2 cells and decreased DU145 cell migration and invasion. Knockdown of P2X4 attenuated growth, migration, and invasion of PCa cells. Finally, knockdown of P2X4 in Myc-CaP cells resulted in significantly attenuated subcutaneous allograft growth in FVB/NJ mice. Collectively, these data strongly support a role for the P2X4 purinergic receptor in PCa aggressiveness and identifies P2X4 as a candidate for therapeutic targeting.

Introduction

Prostate cancer (PCa) remains the second leading cause of cancer-related deaths in American men and Black men are two to three times more likely to die from the disease than White men are [1,2]. Currently, we lack effective therapies against distant metastatic PCa for which the 5-year survival is about 30% [1]. There is a critical need to identify new mechanisms that contribute to PCa progression, that distinguish benign from lethal disease, and that have potential for therapeutic targeting.

Chronic inflammation may contribute to prostate carcinogenesis and is associated with aggressive prostate cancer [3]. Exogenous exposures stimulate innate immune cells including macrophages and neutrophils to infiltrate the prostate. These phagocytes release reactive oxygen species and reactive nitrogen species causing DNA damage, cell injury, and cell death [4,5]. Cell stress, cell injury, and cell death subsequently result in the release of danger-associated molecular patterns (DAMPs), including extracellular ATP (eATP), that exacerbate the inflammatory response [6,7]. Increased eATP concentrations act as powerful chemotactic stimuli for host innate immune cells, many of which express P2 purinergic receptors [6,8-10]. P2X purinergic receptors are ionotropic ligand-gated ion channel-type receptors (P2X1–P2X7) and P2Y purinergic receptors are metabotropic G protein-coupled receptors (P2Y1, P2Y2, P2Y4, P2Y6, P2Y11–14) [11,12]. P2 purinergic receptors are differentially expressed across cell types and may sometimes only be detected in pathological conditions [11]. When eATP binds P2 purinergic receptors on immune cells, activation results in the induction of chemokines and inflammatory markers, and the assembly or activation of the inflammasome [13-15]. These events help drive inflammation and further stimulate ATP release from nearby cells. Propagation of inflammation is thought, in part, to upregulate multiple P2 purinergic receptors on both immune and non-immune cells [16]. Activation of P2 purinergic receptors on non-immune cells promotes cell proliferation and mediates cell differentiation promoting migratory and invasive phenotypes [17-19]. For instance, P2X7 and P2Y2 activation enhances the invasiveness of PCa cells [20,21]. Consequently, purinergic signaling has been studied in the context of inflammation and cancer. Significantly greater eATP concentrations were measured in models of chronic inflammation and in the tumor interstitium of mice compared to healthy tissues [22,23]. P2 purinergic receptor overexpression is reported in multiple cancer types and correlates with poorer outcomes [24-28]. P2X4 purinergic receptor expression was identified as the most highly expressed P2 purinergic receptor in PC3, LNCaP, and C4-2B PCa cell lines and P2X4 antagonists had anti-tumorigenic effects in a PCa xenograft model [29]. However, precisely which cells in the tumor and tumor microenvironment express P2 purinergic receptors, their functional significance and their association with PCa outcomes is not fully described.

Herein, we demonstrate elevated P2X4 purinergic receptor expression in the main presumptive precursor to invasive prostatic adenocarcinoma, prostatic intraepithelial neoplasia (PIN), and PCa compared to benign tissues. P2X4 receptor expression was significantly increased in cases with ERG positivity or with PTEN loss and higher P2X4 protein expression in benign regions adjacent to primary PCa was associated with an increased risk of the development of metastasis after radical prostatectomy. We further demonstrate that P2X4 receptor drives PCa cell growth, migration, and invasion in vitro and PCa tumor development in vivo.

Materials and methods

Patient population and clinical samples

All specimens were acquired using protocols approved by the Johns Hopkins University Institutional Review Board. Formalin-fixed paraffin-embedded (FFPE) whole tissue sections (standard tissue slides) from tissue blocks were obtained from organ donor prostates with no indication of cancer (n=4) and from radical prostatectomy specimens (n = 6). Three separate tissue microarray (TMA) sets were used in this study: Prostate Cancer Biorepository Network (PCBN) High Grade Race, Race Disparity [30,31], and Metastasis at Autopsy. TMAs were used in RNA in situ hybridization (ISH) and immunohistochemistry (IHC) experiments. Further details of the patients and TMAs are provided in Supplementary materials and methods and supplementary material, Tables S1 and S2.

Cell lines, cell treatment, and P2X4 knockdown

TRAMPC2, Myc-CaP, LNCaP, CWR22Rv1, VCaP, and HEK293T cells were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). PC3 and DU145 were obtained from the NCI-Frederick (Frederick, MD, USA). LAPC4 and MDA PCa 2b cells were obtained from J.T. Isaacs (Johns Hopkins University). BMPC1, BMPC2, cells were obtained from Angelo De Marzo (Johns Hopkins University) [32]. Details of the cell line growth conditions and authentication, siRNA, shRNA, and CRISPR-cas9 knockdown are given in Supplementary materials and methods.

Details for protein extraction, western blotting, MTT assays, cell migration and invasion assays, ISH, IHC, dual staining and controls are provided in Supplementary materials and methods.

Mice

All procedures were performed under the guidelines of Johns Hopkins Animal Care and Use Committee (ACUC) after a two-week acclimatization period. Details for housing and allograft experiments are provided in Supplementary materials and methods.

Statistical analyses

Data were analyzed and graphed using GraphPad Prism Software (version 7.01; GraphPad Software, Inc., San Diego, CA, USA) and differences were considered significant if p < 0.05.

We pooled men from the “Race Disparity” and “PCBN High Grade Race” TMA sets and estimated the median P2X4 protein intensity per unit area of tumor- or of benign-tissue using negative binomial regression and adjusting for TMA set, race/ethnicity, age, stage, grade, PTEN and ERG status (Tables 1,2 and supplementary material, Table S3). Analyses were conducted using SAS version 9.4. (SAS Institute Inc., Cary, NC, USA). Further details are provided in t in Supplementary materials and methods.

Table 1.

Adjusted median P2X4 protein intensity per unit area measurement stratified by PTEN, ERG, and race.^*

Tissue Type	Cancer			Benign
Gleason	<=3+4	>=4+3	p**	<=3+4	>=4+3	p**
No. cases	178	213		178	213
Median	2.4579	2.4927	0.9016	0.4363	0.6965	0.0006
Race	Black	White	p**	Black	White	p**
No. cases	182	209		182	209
Median	2.77	2.20	0.036	0.53	0.58	0.46
PTEN	Intact	Loss	p**	Intact	Loss	p**
No. cases	302	87		302	87
Median	2.16	3.33	0.0003	0.51	0.67	0.06
ERG	Negative	Positive	p**	Negative	Positive	p**
No. cases	255	134		255	134
Median	1.94	3.34	<.0001	0.52	0.6	0.26
White Patients
PTEN	Intact	Loss	p***	Intact	Loss	p***
No. cases	134	48		134	48
Median	3.13	4.32	0.037	0.84	0.94	0.52
ERG	Negative	Positive	p***	Negative	Positive	p***
No. cases	95	87		95	87
Median	2.3899	4.4341	<.0001	0.8577	0.8814	0.8671
Black Patients
PTEN	Intact	Loss	p***	Intact	Loss	p***
No. cases	168	39		168	39
Median	1.94	3.20	0.0081	0.50	0.68	0.18
ERG	Negative	Positive	p***	Negative	Positive	p***
No. cases	160	47		160	47
Median	1.9704	2.9725	0.0241	0.5101	0.6211	0.3701
PTEN Adjusted
Race	White	Black	p#	White	Black	p#
No. cases	182	207		182	207
Median	2.96	2.43	0.071	0.55	0.62	0.37
ERG Adjusted
Race	White	Black	p##	White	Black	p##
No. cases	182	207		182	207
Median	2.62	2.47	0.59	0.53	0.60	0.31

^*, participants from “Race Disparity” and “PCBN High Grade Race” TMA sets.

P-values (p) estimated from negative binomial regression with the adjustment for

^**TMA set, age, race, grade, and stage

^***adjustment for TMA set, age, grade, and stage

^#TMA set, age, grade, stage, and PTEN status

^##TMA set, age, grade, stage, and ERG status.

Table 2.

Association of P2X4 expression and joint categories of P2X4 expression and PTEN loss with prostate cancer metastasis and biochemical recurrence.^*

	Cases (N)	Person-Years		HR (95% CI) **		p**
Metastasis	Benign
Below median	10	861		1.00
Above median	25	1043	2.57	1.18	5.60	0.02
Below median***	10	861		1.00
Above median***	25	1043	2.67	1.21	5.90	0.02
Below median /PTEN+	7	712		1.00
Below median /PTEN−	3	149	1.01	0.23	4.36	0.99
Above median /PTEN+	18	769	2.39	0.92	6.23	0.08
Above median /PTEN−	7	274	3.36	1.01	11.2	0.05
	Cancer
Below median	12	933		1.00
Above median	23	971	1.88	0.88	4.00	0.10
Below median***	12	933		1.00
Above median***	23	971	1.88	0.88	4.00 998	0.10
Below median /PTEN+	11	831		1.00
Below median /PTEN−	1	101	0.25	0.03	2.23	0.22
Above median /PTEN+	14	649	1.25	0.51	3.05	0.62
Above median /PTEN−	9	322	1.83	0.67	4.97	0.24
Biochemical	Benign
Below median	40	702		1.00
Above median	61	767	1.44	0.96	2.16	0.08
Below median***	40	702		1.00
Above median***	61	767	1.45	0.96	2.17	0.08
Below median /PTEN+	26	607		1.00
Below median /PTEN−	14	95	1.59	0.81	3.14	0.18
Above median /PTEN+	39	567	1.44	0.87	2.39	0.15
Above median /PTEN−	22	200	2.31	1.26	4.24	0.01
	Cancer
Below median	47	729		1.00
Above median	54	740	1.27	0.84	1.91	0.25
Below median***	47	729		1.00
Above median***	54	740	1.19	0.79	1.8	0.41
Below median /PTEN+	34	668		1.00
Below median /PTEN−	13	61	1.84	0.90	3.8	0.09
Above median /PTEN+	31	506	1.311	0.79	2.2	0.30
Above median /PTEN−	23	234	1.82	1.03	3.21	0.04

^*, participants from Race Disparity” and “PCBN High Grade Race” TMA sets. Median was determined based on the P2X4 measurement among all men. Hazard ratio (HR), 95% confidence intervals (CI), and p-values (p) estimated from Cox Hazard regression model adjusting for

^**age, stage, grade, race, and TMA sets

^***age, stage, grade, race, TMA sets, and PTEN status.

Results

P2X4 purinergic receptor is overexpressed in prostate cancer.

We began by examining all 15 human P2 purinergic receptors in an RNAseq dataset (n=25) obtained on PCa and matched benign tissues [33]. We identified P2X4 (HUGO symbol P2RX4) purinergic receptor expression as elevated in cancer compared to benign tissues (p = 0.0045) in both low grade (p = 0.0168) and high grade (Gleason sum ≥ 4 +3) cases (p = 0.0474, supplementary materials, Figure S2A). P2X4 overexpression in PCa was corroborated in publicly available gene datasets, Tomlins (p = 7.44E-5), Lapointe (p = 8.81E-5), Vanaja (p = 4.07E-5), Welsh (p = 1.93E-5), and Wallace (p = 0.002) analyzed via ONCOMINE (supplementary material, Figure S2B-F) [34-39]. P2X4 receptor mRNA expression was also elevated in PIN compared to benign tissues (p = 9.48E-4) in the Tomlins dataset (supplementary material, Figure S2B) [34]. Analysis of TCGA RNA sequencing data using Wanderer [40] further corroborated elevated P2X4 expression in PCa (n = 374) compared to benign tissues (n = 52, p = 9.10E-12, supplementary material, Figure S2G).

P2X4 purinergic receptor protein is primarily expressed in epithelial cells and immune cells in the prostate.

We aimed to determine the cellular localization and distribution of P2X4 receptor expression in normal prostate tissues. IHC analysis performed on FFPE whole tissue sections from normal organ donor prostate tissues with no cancer (n=4, supplementary material, Table S1). PX4 receptor expression was observed primarily in epithelial cells of the prostate gland. Prostate-infiltrating immune cells had more intense P2X4 expression than did epithelium, particularly in regions of inflammation (Figure 1A). Next, we performed P2X4 IHC on FFPE whole tissue sections from a representative tumor-containing block, a matched benign tissue block with inflammation, and a matched benign tissue block without overt inflammation in radical prostatectomy specimens from 3 Black and 3 White patients (supplementary material, Table S1). Similar to organ donor prostates, there was P2X4 receptor protein expression in epithelial cells of benign glands (Figure 1B,C). We observed increased intensity of P2X4 receptor expression in regions of PIN and cancer compared to benign glands (Figure 1B,C). There was also intense P2X4 receptor expression on immune cells in the stromal compartment and in the lumens of glands (Figure 1D). Dual IHC staining indicated that tryptase-positive mast cells do not express P2X4 receptors (Fig. 1E). On the contrary, some CD66ce⁺ myeloid cells, considered in most cases to be neutrophils, and most CD68⁺ macrophages expressed P2X4 receptors (Figure 1E).

Figure 1.

P2X4 purinergic receptor protein expression in prostate tissues. P2X4 purinergic receptor protein is expressed (A) modestly in the epithelium of prostate glands and more intensely on immune cells of organ donor prostates (C) modestly in benign glands and more intensely in prostatic intraepithelial neoplasia (PIN) and cancer in prostatectomy samples and (D) on immune cells (arrow indicates P2X4 positive immune cell). (B) H&E staining confirmed tissue types. (E) Dual staining showed that tryptase positive mast cells (red) do not express P2X4 protein (brown), but some CD66ce positive neutrophils (brown) and most CD68 positive macrophages (brown) express P2X4 receptors (red) (arrow indicates double-stained immune cell). Images captured with 40x objective magnification unless otherwise stated.

P2X4 purinergic receptor expression is elevated in PIN, prostate cancer, and inflamed regions with increased CD68⁺ macrophages.

We quantified P2X4 receptor protein expression in PCa TMA sets (n = 491 cases, supplementary material, Table S2). There was significantly elevated P2X4 receptor expression in cancer regions compared to benign regions (Wilcoxon matched-pairs signed rank analysis, p < 0.0001, Figure 2A), in both low grade (p < 0.0001) and high grade (p < 0.0001) cancer tissues (Figure 2B). P2X4 protein expression was significantly increased in benign tissues from higher grade compared to low grade cases (Table 1). P2X4 expression was increased in cancer versus benign tissues from both Black men (p < 0.0001) and White men (p < 0.0001, Figure 2C). Interestingly, there was significantly higher P2X4 receptor protein expression in cancer tissues from White men compared to cancer tissues from Black men (p = 0.0006, Figure 2C, Table 1).

Figure 2.

Quantitative image analysis of P2X4 purinergic receptor expression. There was elevated P2X4 receptor protein expression in (A) cancer compared to benign tissues (n = 456; p < 0.0001) (B) from both high grade and low grade cases, and (C) in cancer TMA spots from White men compared to those from Black men. (D) There was elevated P2RX4 mRNA expression in prostatic intraepithelial neoplasia (PIN) and cancer while there is a wide range of expression across metastatic sites. (E) There is a positive correlation between P2RX4 mRNA expression and CD68 mRNA expression. (F) There was elevated P2X4 protein expression in tumors with Erg expression and in tumors with PTEN loss. (A–C, E,F) Each dot represents the mean expression per patient (3 or 4 TMA spots per patient). (D) Each dot represents the expression per TMA core. *p <0.05; *** p <0.001; **** p < 0.0001

ISH analysis of P2X4 (P2RX4) mRNA in the PCBN High Grade Race TMA set (n=120) also showed elevated expression in cancer compared to normal appearing benign tissues (p < 0.0001, Figure 2D). Benign tissue spots on the TMA were further classified as normal (normal appearing, non-inflamed) or as containing atrophy or PIN. There was increased P2X4 mRNA expression in PIN compared to normal benign tissues (p = 0.0152, Figure 2D), corroborating the Tomlins data (supplementary material, Figure S2) [34]. There was a variable degree of P2X4 mRNA expression in metastatic tissues collected at autopsy, but overall there was significantly more P2X4 in primary untreated hormone naïve cancer tissues compared to those from castration resistant lesions from metastatic tissues (p < 0.05, Figure 2D, supplementary material, Table S2).

Based on our observation that CD68⁺ macrophages and some CD66ce⁺ neutrophils express P2X4 receptors (Figure 1E), we quantified CD68 mRNA expression and CD66ce protein expression in the PCBN High Grade Race TMA set. Interestingly, we found a positive correlation between P2X4 mRNA and CD68 mRNA expression (r = 0.4084, p < 0.0001, Figure 2E) but not between P2X4 mRNA and CD66ce protein expression (supplementary material, Figure S3A).

P2X4 receptor protein expression is elevated in prostate cancer cases with ERG positivity or PTEN loss.

We stratified cancer cases based on common genetic alterations of PTEN loss, ERG positivity, or TP53 missense mutation. PTEN is the most commonly inactivated tumor suppressor in PCa and PTEN loss is associated with aggressive PCa [41]. ERG expression, observed in about 50% of PCa, is commonly observed in cases with PTEN loss [41]. TP53 alterations occur in about 53.3% of metastatic castration resistant prostate cancer (mCRPC) and is among the most frequently aberrant genes in mCRPC cases [42]. Our analysis revealed that cancer tissues with loss of PTEN protein expression had significantly elevated P2X4 receptor expression compared to normal PTEN expressing cancer tissues (p < 0.0001, Figure 2F, Table 1). Similarly, cancer tissues with ERG protein expression had significantly higher P2X4 receptor expression compared to ERG-negative cancer tissues (p < 0.0001, Figure 2F, Table 1). There was no difference in P2X4 receptor expression between cancer tissues with or without TP53 nuclear accumulation (supplementary material, Figure S3B). PTEN loss and ERG positivity are more frequent in White compared to Black men [30]. To do determine whether the dissimilar frequency of PTEN or ERG alterations between races contributed to increased P2X4 expression observed among White men, we assessed P2X4 expression by PTEN and ERG status separately by race. P2X4 expression was significantly elevated in cases with PTEN loss and ERG positivity in both White and Black men (Table 1). Interestingly, statistical adjustment for PTEN or ERG status each attenuated the difference in P2X4 expression observed between races (Table 1).

Elevated P2X4 purinergic receptor expression in metastatic prostate cancer.

We performed P2X4 IHC analysis in the Metastasis at Autopsy TMA. P2X4 protein was expressed on immune cells in the metastatic niche and in metastatic cancer cells (Figure 3A). We quantified a broad range of expression across patients and between sites within the same patient (Figure 3B). The mean P2X4 receptor protein expression from adrenal gland, lung, lymph node, and pancreas were each significantly elevated compared to prostate benign tissues (harvested at autopsy e.g., the men had not undergone radical prostatectomy) (Figure 3B). Dual staining revealed positive P2RX4 mRNA expression on CD68⁺ macrophages in metastatic tissues (Figure 3C). Similar to primary PCa tissues, there was a positive correlation between P2RX4 mRNA and CD68 mRNA expression in the Metastasis at Autopsy TMA set (r = 0.5563, p < 0.0001, Figure 3D).

Figure 3.

P2X4 purinergic receptor expression in metastatic prostate cancer tissues. (A, B) P2X4 protein expression was detected in cancer and immune cells and in various metastatic sites. (C) P2RX4 mRNA expression (brown, arrows) detected on CD68⁺ (red) macrophages in metastatic tissues. (D) There is a positive correlation between CD68 and P2RX4 mRNA expression in metastatic tissues. *p <0.05; *** p <0.001, **** p < 0.0001

P2X4 purinergic receptor expression in benign tissue is associated with risk of metastasis after radical prostatectomy.

We separately examined the question of whether P2X4 receptor expression in prostatectomy tissues was associated with outcomes in terms of biochemical recurrence and time to metastasis development. Compared to men with P2X4 below the median in benign glands, men with P2X4 protein expression above the median in benign glands had a suggestive increased risk of biochemical recurrence (HR = 1.437, 95% CI: 0.96, 2.16) and a greater than two-fold increased risk of metastasis (HR = 2.57, 95% CI:1.18, 5.60; Table 2). PTEN loss has been shown to be prognostic in prostate cancer [30,43], however, adjustment for PTEN did not change inferences (Table 2). We examined the combined effect of P2X4 receptor expression in benign tissues and PTEN loss in tumor cells. As compared to men with P2X4 below the median and PTEN intact, men with P2X4 above the median and PTEN loss had a greater than two-fold increased risk of biochemical recurrence (HR = 2.31, 95% CI:1.26, 4.24) and a greater than 3-fold increased risk of metastasis (HR = 3.36, 95% CI: 1.01,11.18; Table 2). Interestingly, as compared to men with P2X4 expression in cancer below the median and PTEN intact in cancer tissue, men with P2X4 expression in cancer above the median and PTEN loss had a significantly increased risk of biochemical recurrence (HR = 1.82, 95% CI:1.03, 3.21; Table 2). P2X4 expression in cancer tissue was not significantly associated with metastasis (Table 2).

P2X4 purinergic receptor activation increases prostate cancer cell migration and invasion.

RISH and IHC analysis detected P2X4 receptor expression in human derived PCa cell lines PC3, DU145, LNCaP, CWR22Rv1, VCaP, LAPC4, and MD PCa 2b (supplementary material, Figure S4). Similarly, P2RX4 protein and P2rx4 mRNA expression were detected in mouse derived BMPC1, BMPC2, Myc-CaP, and TRAMP-C2 cell lines (supplementary material, Figure S5). Considering the association between P2X4 receptor expression in benign regions and increased risk of metastasis, we assessed the effect of P2X4 receptor activation on the migratory or invasive phenotype of PCa cells. Treatment of PC3 cells with endogenous broad spectrum P2 receptor ligand ATP (100 μM) or P2X4 selective agonist CTP (100 μM) modestly decreased expression of epithelial markers E-cadherin and cytokeratin 8 by western blotting analysis (supplementary material, Figure S6A). Concurrently, ATP and CTP treatment increased expression of mesenchymal markers vimentin and snail, suggesting that P2X4 activation may promote epithelial to mesenchymal transition (EMT) in PCa cells (supplementary material, Figure S6A). Similar results were seen in E-cadherin and vimentin WB analysis of DU145 cells (Figure 4A). We used CRISPR-cas9 gene editing to knock down P2X4 expression in DU145 cells (supplementary material, Figure S7A). P2X4 knockdown (KD) resulted in increased e-cadherin and decreased vimentin protein expression in DU145 cells (Figure 4B). We performed functional Transwell assays to assess the effect of P2X4 activation on cell migration and invasion. Nucleotide treatment alone was sufficient to significantly increase PC3 and CWR22Rv1 cell migration across Transwell membranes and invasion through matrigel coated Transwell membranes at 24, 48, or 72 h (supplementary material, Figure S6B). Specifically, treatment with CTP for 72 h significantly increased both migration (7.2-fold, p=0.018; 11.6-fold, p=0.04 for CTP at 100 or 400 μM respectively) and invasion (6.0-fold, p=0.045; 8.4-fold, p=0.009 for CTP at 100 or 400 μM respectively) of DU145 cells (Figure 4C). Conversely, treatment with the P2X4 selective antagonist, 5-(3-Bromophenyl)-1,3-dihydro-2H-Benzofuro[3,2-e]-1,4-diazepin-2-one (5-BDBD) (5 or 10 μM) significantly decreased DU145 cell migration (0.2-fold, p=0.001; 0.2-fold, p=0.008 respectively) and invasion (0.2-fold, p<0.001; 0.3-fold, p<0.0001, respectively. Figure 4C). Further, P2X4 knockdown also resulted in reduced cell migration (0.5-fold, n.s.) and invasion (0.3-fold, p=0.001) in DU145 cells (Figure 4C).

Figure 4.

P2X4 purinergic receptor function affects PCa cell motility. (A) Nucleotide treatment and (B) P2X4 knock down in DU145 cells decreased E-cadherin protein expression and increased vimentin protein expression as assessed by western blotting. (C) CTP treatment increased cell migration and invasion while 5-BDBD and P2X4 knock down decreased cell migration and invasion in Transwell assays. *p <0.05 (P = parental, KD = knockdown)

P2X4 purinergic receptor blockade decreases prostate cancer cell viability.

We assessed the effect of P2X4 blockade on PCa cell viability. Treatment with 5-BDBD resulted in a dose-dependent decrease in cell viability (as assessed by MTT assays) of PC3, DU145, LNCaP, CWR22Rv1, TRAMP-C2, Myc-CaP, BMPC1, and BMPC2 cells (Figure 5A and supplementary material, Figure S8A). P2X4 KD attenuated growth rates of LNCaP and DU145 cells (Figure 5B and supplementary material, Figure S8B). Treatment with 5-BDBD induced dose dependent expression of the apoptotic marker cleaved caspase-3 in LNCaP and CWR22Rv1 cells (Figure 5C). Temporary knockdown of P2X4 by siRNA also increased levels of cleaved caspase-3 in PC3 cells (supplementary material, Figure S8C).

Figure 5.

P2X4 purinergic receptor function affects PCa cell viability. (A) P2X4 antagonist 5-BDBD resulted in a dose-dependent decrease in PCa cell viability, *p<0.0001 for each cell line compared to no treatment. Each point represents 12 wells of the respective cell line. Each cell line was assessed in different experiments. (B) P2X4 knockdown in LNCaP cells resulted attenuated cell growth over time. Each point represents 24 wells. * p<0.05 for KD-6 and KD-8 compared to LNCaP; #p<0.05 for KD-6 only compared to LNCaP. (C) Treatment with 5-BDBD resulted in a dose-dependent induction of cleaved caspase-3 in PCa cells.

P2X4 purinergic receptor knockdown attenuates Myc-CaP allograft tumor growth.

We assessed whether P2X4 receptor function affects tumor initiation and progression in vivo. We confirmed P2X4 receptor expression in Myc-CaP allograft, TRAMP-C2 allograft, Hi-Myc prostate tumor, BMPC prostate tumor, and BMPC bone metastasis tissues. We observed robust expression of P2rx4 mRNA in all four models (Figure 6A). We selected an allograft model for the feasibility of genetically manipulating P2X4 expression directly in cancer cells and the benefit of using an immunocompetent host. We used CRISPR-cas9 to knock down P2X4 receptor expression in Myc-CaP cells. Myc-CaP control (CRISPR empty vector) and Myc-CaP P2X4 KD cells in 50% matrigel were subcutaneously allografted into male FVB/NJ mice (10-week old, n=10). Cages were blinded to the investigator that measured tumor growth. Myc-CaP-P2X4 KD cells resulted in significantly attenuated subcutaneous allograft growth in FVB/NJ mice compared to control Myc-CaP cells (Figure 6B). Tumors were palpable or measurable in all 10 mice in the control group, compared to 8 out of 10 in the P2X4 KD group. The tumors that were established in the P2X4 KD group progressed significantly slower than tumors in the control group (p < 0.0001, Figure 6B). H&E staining, ISH to detect human MYC mRNA (human MYC mRNA is specific for the Myc-CaP cells and is not present in host mouse cells), and IHC for androgen receptor (AR) were used to confirm that tumors were developed from allografted Myc-CaP cells (Figure 6C). IHC for P2X4 confirmed P2X4 reduction in allografts generated from Myc-CaP-P2X4 KD cells (Figure 6C). Interestingly, we observed substantial P2X4 protein expression in some regions lacking human c-Myc mRNA and AR protein expression (Fig. 6D). These regions likely represented the host inflammatory cells in the tumor microenvironment (TME), which was confirmed by IHC analysis that identified CD11b⁺ immune cells, likely macrophages or neutrophils, and CD3⁺ T cells in these P2X4 positive non-tumor regions (Figure 6D and supplementary material, Figure S9).

Figure 6.

P2X4 purinergic receptor protein expression in mouse prostate tissues. (A) Robust P2rx4 mRNA expression is detected in mouse prostate tumors. (B) CRISPR knockdown of P2X4 receptor expression in Myc-CaP cells (n=10) resulted in attenuated allograft growth in FVB/NJ mice compared to control Myc-CaP cells (n=10). Gray lines represent tumor volume for individual mice and solid lines represent average tumor volume for the group, **** p < 0.0001. (C) Human MYC ISH and P2X4 and AR IHC analysis confirmed reduced P2X4 expression in Myc-CaP allograft tissue from mice. (D) P2X4 positive regions lacking AR and human MYC stain positive for CD11b and CD3 in Myc-CaP allograft tissue from mice.

Discussion

Herein we report a comprehensive profile of P2X4 receptor expression in normal prostate, primary PCa, metastatic PCa, human and mouse PCa cells, and mouse PCa tissues. We demonstrate P2X4 receptor protein expression in the epithelium and infiltrating immune cells of the prostate and measured elevated P2X4 receptor expression in human PIN and cancer by RNA sequencing analysis, ISH, or IHC. Higher P2X4 protein expression in benign prostatic tissues accompanying the cancer lesions was also associated with an increased risk of metastasis, and above median P2X4 protein expression in benign tissues coupled with PTEN loss was associated with an increased risk of both biochemical recurrence and distant metastasis after prostatectomy.

P2X4 purinergic receptors have primarily been studied for their role in neuropathic and inflammatory pain [44,45]. In fact, a P2X4 antagonist developed for the treatment of neuropathic pain has completed a Phase 1 clinical trial [46,47]. Like other P2X purinergic receptors, P2X4 forms heterotrimers with P2X2, P2X5, and/or P2X6 and is localized at the plasma membrane [48]. P2X4 is unique in that it may also localize to intracellular compartments such as lysosomes [48]. Our IHC analysis demonstrates that P2X4 receptor expression is not limited to the plasma membrane in PCa. While P2 purinergic receptors have been widely investigated in different cancer sites, studies on P2X4 receptor expression and function in cancer have rarely been reported [20,21,24-28]. A recent study identified the P2X4 purinergic receptor as a potential clinical target for patients with PCa [29]. The authors demonstrate that pharmacological inhibition of P2X4 receptors impaired PCa cell growth and decreased PCa cell motility [29]. Our study corroborates these findings using additional cell lines and using genetic manipulation to confirm the specific role of P2X4 purinergic receptors in PCa cell viability, migration and invasion. Further, we report that direct activation of P2X4 receptors alone was sufficient to promote PCa migration and invasion in Transwell assays. Future studies will be aimed at determining the mechanistic pathways through which P2X4 receptors function in PCa cells.

Our dual staining analysis identified some CD66⁺ neutrophils, most CD68⁺ macrophages, but no tryptase⁺ mast cells as P2X4 receptor-expressing immune cells in PCa tissues. We also found a positive correlation between P2RX4 mRNA and CD68 mRNA expression in the PCBN High Grade and Metastasis at Autopsy TMA sets. Considering the strong P2X4 receptor staining we observed in epithelial cells, we believe that this positive correlation is not primarily due to macrophages expressing P2X4 receptors. Rather, there may be higher macrophage infiltration in regions with high P2X4 receptor expression. A study reported that 5-BDBD treatment in a mouse model of ischemic stroke reduced infiltration of total leukocytes (including neutrophils and monocytes) into the brain after stroke [49]. Other studies report P2X4 receptors as involved in macrophage activation, facilitating macrophage phagocytosis and NLRP1/3 inflammasome induction [48,50,51]. High numbers of M2 macrophages are associated with poorer outcomes in men with PCa and are increased in metastatic disease [52,53]. As such, investigating P2X4 receptor function in the context of PCa would be worthwhile.

We report higher P2X4 receptor expression in cases with PTEN loss and cases with high P2X4 expression and PTEN loss have an increased risk of biochemical recurrence and metastasis. Future studies will be aimed at assessing the feasibility of using P2X4 receptor expression in combination with other markers as a prognostic tool. Additionally, the biological associations of P2X4 receptors and PTEN is not yet fully investigated. PTEN is a negative regulator of PI3K resulting in de-phosphorylation of Akt [54]. Phosphorylated Akt (p-Akt), its active form, is associated with poor prognosis in multiple cancers and promotes mesenchymal-like properties and metastasis of PCa cells [55]. Interestingly, a study by Ghalali et al reported crosstalk between PTEN and another Akt regulatory phosphatase, PHLPP that is dependent on P2X4 receptor signaling [54]. Ghalali et al proposed that this crosstalk controls the levels of p-Akt, promotes PCa cell invasion, and is mediated by the P2X4 purinergic receptor. Further, this study showed that the P2X4 receptor is necessary for TGF-β mediated PC3 cell invasiveness [54].

Higher P2X4 receptor expression was also found in cases with ERG positivity, compared to ERG negative cases, within our cohorts. ERG positivity in PCa is the result of gene fusions between the androgen-regulated gene TMPRSS2 (or less commonly other androgen regulated genes) and the transcription factor ERG [56]. One study reported the P2Y2 purinergic receptor among common upregulated genes in TMPRSS2:ERG-expressing PC3 cells [57]. The study found that the collective TMPRSS2:ERG target genes were associated with PC3 cell motility and invasiveness, further corroborating the role for the P2 purinergic receptor family in driving PCa cell invasiveness [57]. However, little is known about the direct relationship between the P2X4 receptor and androgen signaling. One study reported that ATP treatment inhibited growth in androgen-independent PCa cells, but the specific receptor involved was not identified [58]. Another study reports that synthetic testosterone derivatives modulate rat P2X4 receptor channel gating, suggesting interplay between the two pathways [59]. Future studies will be aimed at interrogating the relationship between P2X4 purinergic and androgen signaling.

Our study demonstrated that P2X4 receptors are involved in the tumor initiation and progression of PCa subcutaneous allografts in mice. These data provide a strong proof of concept that P2X4 purinergic receptors, specifically, have a direct role in tumor development. The Myc-CaP allograft model allowed us to assess tumor growth in a fully immunocompetent mouse. Our histological analysis demonstrated that P2X4 receptor expression was robust on immune cells and throughout the TME. Future studies will investigate the role of P2X4 purinergic receptors in the PCa TME.

Overall, our study demonstrates a role for P2X4 purinergic receptors in PCa aggressiveness. The associations of P2X4 receptor expression with common genetic alterations and increased risk of metastasis or biochemical recurrence highlight the P2X4 purinergic receptor as a potential prognostic factor. Further, the current study indicates a functional role for P2X4 in maintaining PCa cell viability, promoting PCa cell motility and invasiveness, and driving PCa tumor development in vivo. These data and the existence of a P2X4 antagonist in-development highlight the P2X4 purinergic receptor as a therapeutic target for the treatment of aggressive PCa.

Supplementary Material

supinfo1

Figure S1. P2X4 targeted shRNA shows specific reduction of (A) human anti-P2X4 antibody signal, (B) mouse anti-P2X4 antibody signal, (C) human P2RX4 RISH probe signal and (D) mouse P2rx4 RISH probe signal

Figure S2. P2RX4 purinergic receptor expression in multiple datasets

Figure S3. P2X4 receptor expression is not associated with CD66ce expression or p53 status

Figure S4. P2RX4 purinergic receptor mRNA and its protein expression is detected in human derived prostate cancer cell lines

Figure S5. P2rx4 purinergic receptor mRNA and its protein expression is detected in mouse derived prostate cancer cell lines

Figure S6. Effects of nucleotide treatment on the expression of epithelial and mesenchymal protein markers and the migratory and invasive abilities of PCa cells

Figure S7. CRISPR knockdown of P2X4 was confirmed by ISH analysis and western blotting

Figure S8. P2X4 purinergic receptor function affects PC3 and DU145 cell viability.

Figure S9. P2X4 positive regions lacking AR and human c-Myc stain positive for CD11b in Myc-CaP allograft tissue from mice

supinfo3

Table S1. Patient characteristics for whole tissue sections

Table S2. Patient characteristics for TMA sets

Table S3. Association between P2X4 expression and prostate cancer metastasis, or biochemical recurrence by race*