Altered expression of genes identified in rats with prostatic chronic inflammation in a prostate spheroid model treated by estradiol/testosterone

Abstract

Rats are the standard model for male reproductive toxicity testing. Rat prostates are physiologically and anatomically different from those of humans. Drug and chemical toxicity testing would benefit from an in vitro model of human prostate cells. Recently, spheroids derived by three-dimensional culture of human cell lines have been used for assessing drug and chemical toxicity in vitro as they mimic in vivo environments more closely than two-dimensional culture. However, forming consistently sized, uniform spheroids is technically challenging for toxicity testing. The purpose of this study was to identify potential genetic markers for assessing prostatic toxicity in spheroids. We formed prostate spheroids using agarose-coated plates seeded with human primary prostate epithelial cells. Prostate spheroids were treated with either 17β-estradiol (E2) or testosterone (T) on days 2–7 of culture. Samples were harvested on culture day 7. qPCR was used to examine gene expression levels previously identified in rats with chronic inflammation exposed to estradiol benzoate, E2 and/or T. Changes in some gene expression levels were observed in the spheroids treated with E2 or T. We found that treatment with 1 nM E2 and/or 10 μM T significantly altered spheroid proliferation and viability, as well as the expression levels of genes including Nanog homeobox (NANOG), C-C motif chemokine ligand 2 (CCL2) and bone morphogenetic protein receptor type 2 (BMPR2). Further studies using biologically active molecules with prostatic toxicity are needed to verify the results and to determine whether gene expression changes in the spheroid are specific to E2 or T treatment.

INTRODUCTION

Currently, animal models are commonly used for prostate toxicity testing according to guidelines of the Organisation for Economic Co-operation and Development (OECD) (http://www.oecd.org/chemicalsafety/testing/39791889.pdf; https://www.oecd.org/env/ehs/testing/E452_2009.pdf). In addition, endocrine responses in the prostate are measured via the Hershberger assay using rats (https://www.epa.gov/sites/production/files/2015-07/documents/final_890.1600_hershberger_assay_sep_10.6.11.pdf). Although the rat prostate has commonly been used as a model for male reproductive toxicity testing, the prostates of rats are physiologically and anatomically different from those of humans (Russell and Voeks, 2003).

Most in vitro cell-based methods for drug screening and toxicity testing utilize two-dimensional (2D) monolayer culture approaches, in which one or more cell types are cultured on flat surfaces (e.g., dishes, plates or slides). Although these methods are both simple and cost-effective, they do not recapitulate the structure of tissues in vivo (Kapałczyńska et al., 2018; Langhans, 2018). Recently, three-dimensional (3D) culture methods, which mimic in vivo environments more closely, have been used for toxicity testing (Pampaloni and Stelzer, 2010; Li and Cui, 2014; Kapałczyńska et al., 2018; Langhans, 2018; Zink et al., 2020). In these approaches, cells from multiple organs are cultured under low-attachment conditions or on culture dishes coated with extracellular matrix (e.g., fibronectin, Matrigel and collagen), and form aggregates known as “spheroids” (Kapałczyńska et al., 2018; Langhans, 2018; Thoms et al., 2018; Zink et al., 2020).

Several studies using prostate spheroids for toxicity testing have previously been published (Mittler et al., 2017; Fontana et al., 2020; Paczkowski et al., 2021). Moreover, in the context of cancer, prostate spheroids using prostate cancer cell lines can be used for cancer drug development and screening (Mittler et al., 2017). Spheroid assays may be used to supplement currently available toxicology testing approaches.

One disadvantage of using spheroids for toxicity testing, however, is the difficulty of forming spheroids of consistent and uniform size and shape between the control and chemical-treated groups (Katt et al., 2016). As an alternative, gene expression profiling of spheroids for potential markers of prostatic diseases may be a more functional and useful method for assessing drug and chemical toxicity. We selected potential genes to be examined from our recent study, which identified differential expression of genes associated with prostatic chronic inflammation in rats following exposure to estradiol benzoate (EB), 17β-estradiol (E2) and/or testosterone (T) (Nakamura et al., 2020a, 2020b). We selected these genes to compare in vivo and in vitro data. Prostate diseases are likely to occur when the ratios of sex hormone levels are imbalanced (Kaufman and Vermeulen, 2005; Prins, 2008; Yao et al., 2011).

We therefore used reverse transcription-quantitative PCR (RT-qPCR) to validate whether the expression levels of genes selected from the in vivo study in rats were also altered in 3D prostate spheroids formed from human primary prostate epithelial cells (HPrEpCs) as a step toward evaluating whether this assay can be used to supplement current toxicity testing approaches.

MATERIALS AND METHODS

Materials

All reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) or Fisher Scientific (Pittsburgh, PA, USA) unless otherwise indicated.

Cell lines, authentication, and 2D culture

Human primary prostate epithelial cells, obtained from an adult male, > 60 years of age, were purchased from Cell Applications, Inc. (cat. # 934-05a; San Diego, CA, USA). Cells were cultured in Human EpiVita Basal Medium containing Epi growth supplement (cat. # 214-500; Cell Applications, Inc.). This media is specifically used for the culture of human primary prostate epithelial cells, which should be maintained at 37°C and 5% CO₂ according to manufacturer’s instructions. DNA profiling-based cell authentication was performed by Biosynthesis, Inc. (Lewisville, TX, USA). The company concluded that “no contamination by another human cell line was detected,” and that no matches were found after searching the American Type Culture Collection (ATCC), the Japanese Collection of Research Bioresources (JCRB), and the German Collection of Microorganisms and Cell Cultures GmbH (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH; DSMZ) databases.

Prostate spheroid formation (3D culture)

Spheroids were formed using the 3D culture method described by Hedlund et al. (1999) and Friedrich et al. (2009). In brief, 50 µL of 1.5% agarose I (Dojindo Molecular Technologies, Rockville, MD, USA) diluted in supplement-free Human Epi Growth Medium was added to each well of a 96-well cell culture plate. The matrix was allowed to dry for 20 min at room temperature (22–25°C), after which 150 µL of a suspension containing 3.3 × 10⁵ HPrEpCs/mL was plated. These plates were incubated at 37°C and 5% CO_2. One prostate spheroid was cultured per well. Samples were collected on day 0 (as 2D cultures), day 1, and day 2, and qPCR was performed to confirm spheroid formation.

Chemical treatments

Chemical treatment of prostate spheroids was performed as described by Prins et al. (2014). On day 2 of culture, the aggregated cells (one per well) were treated with one of four different concentrations of 17β-estradiol (E2; cat. # E8875; 1, 10, 100, and 1000 nM) or testosterone (T; cat. # T1500; 1, 5, 10, and 50 µM). DMSO (Life Technologies, Carlsbad, CA, USA) was used as the vehicle (control). On day 7 of culture, the spheroids were harvested and used for the following assays.

RNA extraction

RNA was extracted from 2D HPrEpC and 3D prostate spheroids (n = 4 per timepoint) using TRIzol reagent (Life Technologies) or using a Single Cell RNA Purification Kit (Norgen Biotek Corp., Thorold, ON, Canada). For RNA extraction from chemically-treated prostate spheroids (n = 6–9 per group), one or two prostate spheroids were pooled and collected as one sample. RNA was extracted using the Single Cell RNA Purification Kit (Norgen Biotek). The concentration of each RNA sample was determined using a DS-11 spectrophotometer (DeNovix Inc., Wilmington, DE, USA).

cDNA synthesis

cDNA was synthesized by reverse transcription (RT) of 0.5 µg of RNA with the Super Script IV VILO Master Mix (Life Technologies) for each sample.

RT-qPCR

Gene expression levels were quantified by RT-qPCR with cDNA templates using the PowerUp SYBR Green Master Mix (Life Technologies) according to manufacturer’s instructions, in a 10 μL reaction volume. Samples were run on an ABI PRISM 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) using an initial denaturation step (50°C for 2 min and 95°C for 2 min), followed by 45 amplification cycles (95°C for 15 sec and 60°C for 1 min). The expression level of each gene of interest was normalized using glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as an internal control. Relative steady-state transcript levels were calculated using threshold cycle (Ct) values, and the relative ratios of the transcript levels in each sample were calculated with the following equation: relative quantity = 2^−ΔΔCt (Livak and Schmittgen, 2001); the values for each control sample were set to “1”. The qPCR reactions for each sample were performed in triplicate. Specific primer pairs used in this study are shown in Table 1.

Table 1. Primer pairs used for qPCR validation.

Genes	Primer sequences	Amplified size (bp)	GenBank Accession/reference
hNANOG	For: 5′-AAT GGT GTG ACG CAG AAG G-3′	139	NM_024865.4
	Rev: 5′-GGT TGC TCC AGG TTG AAT TG-3′		Prins et al., 2014
hOCT4 (POU5F1)	For: 5′-GAC AGG GGG AGG GGA GGA GCT AGG-3′	144	NM_001173531.3
	Rev: 5′-CTT CCC TCC AAC CAG TTG CCC CAA AC-3′
hTP63	For: 5′-GTG AGC CAC AGT ACA CGA ACCG-3′		Prins et al., 2014
	Rev: 5′-GAG CAT CGA AGG TGG AGC TGG-3′
hAR	For: 5′- TGT CCA TCT TGT CGT CTT CG-3′	209	NM_000044.6
	Rev: 5′- ATG GCT TCC AGG ACA TTC AG-3′		Prins et al., 2014
hKRT18	For: 5′- CAG CAG ATT GAG GAG AGC AC-3′	110	NM_000224.3
	Rev: 5′- TCG ATC TCC AAG GAC TGG AC-3′		Prins et al., 2014
hLCK	For: 5′- GGA GCT GGG ACC CCC TAT TT -3′	81	NM_001042771.3
	Rev: 5′- GCC CAT GGT CCC TGA GAT TG -3′
hBMPR2	For: 5′- TGG CAG CAG TAT ACA GAG TGA G -3′	150	XM_011511687.1
	Rev: 5′- TTG ACT TCA CAG TCC AGC GA -3′
hBMP7	For: 5′- GGG TAG CGC GTA GAG CC -3′	128	NM_001719.3
	Rev: 5′- GTT GTC CAG GCT GAA GTC GG -3′
hCCL19	For: 5′- CAC ACC TTG CAT TTC ACC CC -3′	98	NM_006274.2
	Rev: 5′- GCC AAC GGT GAA TGT GTG AG -3′
hNFKB1	For: 5′- CTT AGG AGG GAG AGC CCA C -3′	105	NM_01165412.1
	Rev: 5′- ACA TTT GTT CAG GCC TTC CC ´-3′
hCXCR6	For: 5′- ACC AAT GCC TTG CCA ACA AC -3′	95	XM_011533290.2
	Rev: 5′- GTT GGC CTG CTC TCC TTA CC -3′
hADAM19	For: 5′- TAT GGC TGA GGG CGT GTG AG -3′	145	NM_033274.5
	Rev: 5′- CCT CTT GTC CAT CCA GGC TC -3′
hCCL2	For: 5′- AGA TCT GTG CTG ACC CCA AG -3′	73	NM_002982.4
	Rev: 5′- GGA GTT TGG GTT TGC TTG TCC-3′
hGAPDH	For: 5′- AAG ACG GGC GGA GAG AAA CC -3′	140	NM_001289745.3
	Rev: 5′- CGT TGA CTC CGA CCT TCA CC 3′

Assessment of prostate spheroid area

We measured prostate spheroid areas manually. Specifically, images of chemically treated prostate spheroids were captured with the EVOS XL Core Cell Imaging System (Life Technologies), and spheroid area was measured with ImageJ software (version 1.47t; Rasband, 1997-2016; https://imagej.nih.gov/ij/).

Cell viability assay

To measure prostate spheroid viability, we used the CellTiter-Glo^® 3D cell viability assay (Promega Corporation, Madison, WI, USA). In brief, prostate spheroids in their associated culture medium were transferred to 96-well cell culture microplates (Greiner Bio-One North America Inc., Monroe, NC, USA). The old medium was removed and 100 µL of fresh medium was added to each well. Subsequently, CellTiter-Glo^® 3D reagent was added at a 1:1 volume/volume ratio. The plates were vortexed for 5 min and incubated at room temperature for 30 min. Luminescence was measured with the GloMax^®-Multi Detection System (Promega Corporation). Adenosine triphosphate solution (ATP; Sigma-Aldrich) was used to generate a standard curve.

Statistical analysis

Data are presented as the mean ± standard error of the mean (SEM). Student’s t-tests were performed to compare 2D vs. 3D cultures, and a one-way ANOVA with Tukey’s multiple comparisons test was performed for the chemical treatment study. qPCR data were analyzed using a t-test with Bonferroni adjustment (Hochberg, 1988; Shaffer, 1995). P-values of ≤ 0.05 were considered statistically significant.

RESULTS

Prostate spheroid formation

We produced consistent prostate spheroids using the methods described by Friedrich et al. (2009) and Hedlund et al. (1999) (Fig. 1A). As it has been reported that a small number of stem/progenitor cells exist in primary prostate epithelial cells (Hu et al., 2011), we examined the expression levels of stem cell markers (Takahashi et al., 2007) via qPCR. The transcript levels of Nanog homeobox (NANOG), SRY-box transcription factor 2 (SOX2), and POU class 5 homeobox 1 (POU5F1, formerly named OCT4) genes were confirmed at day seven and were higher compared to day zero (2D). Especially, POU5F1 gene expression levels were significantly increased (Supplemental Fig. 1) (Gu et al., 2007).

Fig. 1

Effect of E2 and T on prostate spheroids formation. (A) Images of developing prostate spheroids. Prostate spheroids from human primary prostate epithelial cells (HPrEpCs) cultured on 1.5% agarose-coated 96-well plates were aggregated on day 1 of culture. The aggregated cells compacted on day 2 of culture, and the formation of spheroids was observed on day 7. Scale bars: 1 mm. (B) Relative fold changes in the area of E2- or T-treated spheroids. Data are expressed as the mean fold change ± the standard deviation (n = 8–9 per group). *p < 0.05 vs. the DMSO-treated group. (C) Viability of spheroids treated with E2 or T. Data are expressed as the mean fold change ± the standard deviation (n = 11–12 per group). †p < 0.001 compared to the DMSO-treated group.

Effects of E2 or T treatment on prostate spheroids

To examine the effect of E2 or T on prostate spheroid proliferation we measured the area of treated prostate spheroids with ImageJ software. The average area of prostate spheroids treated with 1 nM E2 or 10 μM T was significantly higher than that of prostate spheroids treated with DMSO (Fig. 1B). Prostate spheroids treated with 5 μM T also tended to be larger than those treated with DMSO; however, this observation was not statistically significant (Fig. 1B). We also measured the viability of 3D prostate spheroids with the CellTiter-Glo (ATP luciferase) viability assay. In this assay, ATP levels were significantly higher in the prostate spheroids treated with 1 nM E2 or 10 μM T, compared to those treated with DMSO (Fig. 1C). As a result, we used prostate spheroids treated with 1 and 10 nM E2, and 1 and 10 μM T, in further qPCR-based studies of gene expression.

Gene expression profiles of E2- or T-treated prostate spheroids

We measured expression levels of the following genes: C-C motif chemokine ligand 2 (CCL2); C-X-C motif chemokine receptor 6 (CXCR6); nuclear factor kappa B subunit 1 (NFKB1); ADAM metallopeptidase domain 19 (ADAM19); bone morphogenetic protein receptor type 2 (BMPR2), LCK proto-oncogene, Src family tyrosine kinase (LCK), phospholipase C zeta 1 (PLCZ1); bone morphogenetic protein 7 (BMP7); and C-C motif chemokine ligand 2 (CCL19). These genes were identified in previous gene expression studies (Nakamura et al., 2020a, 2020b). We also examined the transcript levels of the following genes: NANOG, POU5F1, tumor protein p63 (TP63), androgen receptor (AR), and keratin 18 (KRT18), which were identified in other published studies (Gu et al., 2007; Prins et al., 2014; Hu et al., 2017) (Fig. 2; Supplemental Fig. 2).

Fig. 2

Relative transcript levels of NANOG, POU5F1, LCK, NFKB1, ADAM19, BMPR2, CCL2, CXCR6 and KRT18 in prostate spheroids treated with E2 or T. Data are expressed as the mean fold change ± the standard error of the mean (n = 4–7 per group). *p < 0.05 compared to the DMSO-treated group.

Estrogen treatment - upregulated genes

Compared to the DMSO-treated controls, the transcript levels of NANOG were significantly higher in the prostate spheroids treated with 1 nM E2. Its levels were higher in the prostate spheroids treated with 10 nM E2 group compared to the control group; however, no statistical significances were observed. The significantly higher NANOG gene expression levels in the 1 nM E2 group was same tendency in the viability assay. In addition, the expression levels of LCK, NFKB1, ADAM19, BMPR2, and POU5F1 were higher in prostate spheroids treated with E2 compared to those treated with DMSO. However, statistical significance was only observed in response to treatment with 1 nM and 10 nM E2 (Fig. 2). The transcript levels of the remaining genes (PLCZ1, BMP7, CCL19, TP63, and AR) tended to be higher in the E2- and T-treated groups, but these effects were not statistically significant (Supplemental Fig. 2).

Downregulated genes

The expression levels of CCL2, CXCR6, and KRT18 genes were lower in the E2- treated groups, compared to the DMSO control group (Fig. 2). CCL2 and KRT18 gene expression levels were significantly downregulated in 1 nM E2 and 10 nM E2 groups, respectively.

Testosterone treatment - upregulated genes

Compared to the DMSO-treated controls, the transcript levels of NANOG, ADAM19, and BMPR2 were significantly higher in the prostate spheroids treated with 1 μM T and 10 μM T. In addition, the expression levels of POU5F1 were higher in prostate spheroids treated with 10 μM T compared to those treated with DMSO. Statistical significance was only observed in the rest of the genes in the T-treated group (Fig. 2). The transcript levels of the remaining genes (PLCZ1, BMP7, CCL19, TP63, and AR) tended to be higher in the T-treated groups, but the effect was not statistically significant (Supplemental Fig. 2).

Downregulated genes

The expression levels of CCL2 and CXCR6 genes were significantly lower in the T-treated groups, compared to the DMSO control group (Fig. 2).

DISCUSSION

The present study profiled the expression of potential markers of prostate diseases, identified in our previous study, in prostate spheroids treated with E2 or T, to identify candidate genes to characterize an in vitro human prostate spheroid model. We confirmed that prostate spheroids were successfully formed from HPrEpCs using the agarose method as described by Hedlund et al. (1999) and Friedrich et al. (2009). The expression levels of NANOG and CCL2 genes were significantly altered in prostate spheroids in response to E2 or T treatment, accompanied by changes in the size and viability of prostate spheroids compared to the control group. The change in BMPR2 expression was similar to the E2+T responses observed in rats dosed with EB (Nakamura et al., 2020a). CCL2, which was identified in our study in rats, was downregulated in the spheroids treated with E2 or T.

NANOG is a known stem cell marker (Takahashi et al., 2007). In addition, CCL2 has been described as a “stemness gene” and is highly expressed in spheroid cultures (Lazennec and Richmond, 2010; Tsuyada et al., 2012; Hu et al., 2017). Interestingly, in contrast, the transcript levels of KRT18, which is known as a differentiation gene (Hu et al., 2017), were found to be downregulated by E2 or T treatment. Many researchers have examined the expression of stem cell markers in prostate spheroids formed from normal, benign prostate epithelial cells, or prostate cancer cell lines (Jeter et al., 2011; Zhao et al., 2013; Chandrasekaran et al., 2020). Our findings suggest that measuring the expression of stemness and differentiation genes in prostate spheroids may help to assess the effects of drugs and/or chemical toxicity on the growth of stem or progenitor cells, such as the effect of such treatments on prostatic glandular epithelium in vivo (Kwon and Xin, 2014).

In our previous study in rats, Ccl2 gene expression was dramatically upregulated in the dorsolateral prostates of rats treated with EB, E and T with chronic prostatic inflammation (Nakamura et al., 2020b). However, transcript levels of the CCL2 gene were downregulated in prostate spheroids treated with E2 or T, indicating that these compounds may attenuate the stem cell-like characteristics of prostate spheroids. The difference in Ccl2 gene expression levels in rat prostates versus CCL2 gene in human spheroids may be due to chronic inflammation in the rat prostates. Prostate epithelial cells can secrete increased levels of CCL2 during inflammation (De Nunzio et al., 2016). Thus, the changes in CCL2 gene expression in the spheroids indicates “stemness” function, without an activated inflammation signaling pathway. In addition, CCL2 gene expression levels are the lower in prostate epithelial cells compared to prostate stromal cells (Fujita et al., 2010), suggesting that the change in CCL2 expression levels modulated by E2 or T treatment may be smaller compared to the control group. Additional studies are needed to elucidate the mechanisms by which E2 and T treatments alter CCL2 expression in prostate spheroids. The protein encoded by the BMPR2 gene plays an important role in the differentiation of mesenchymal stem cells (Guo et al., 2019). However, there are no reports about BMPR2 function in prostate spheroids.

Many of the genes examined in the present study have previously been shown to play important roles in prostate spheroids or prostate cell lines. For example, alterations in NFKB1 signaling pathway genes have been reported to effect the formation of prostate spheroids (Härmä et al., 2010). The CXCR6 gene is reported to play a role in the invasive and migratory properties of prostate cancer cells (Rycaj and Tang, 2017). The CXCR6-CXCL16 signaling pathway also has been examined in the context of a 3D spheroid invasion assay (Cho et al., 2016). To our knowledge, no studies have examined the functions of these genes (LCK and ADAM19) in prostate spheroids and therefore further studies of this nature are needed.

This study found that changes of spheroid size and viability by T treatment increased in a dose-dependent manner except for the 50 μM group. Specifically, statistically significances were observed in spheroids treated with 10 μM T. Song and Khera (2014) reported that the effects of optimal T levels on proliferation of prostate cancer cell lines was less than 2.4 ng/mL, which is on the lower end of adult men’s physiological T levels. They also found that lower than optimal T levels shifted cell proliferation, while higher T levels inhibited cell proliferation. Compared to optimal levels of prostate cancer cell proliferation, T levels observed in our current study were higher. We believe that the differences of characteristics of T responses between cancer cell lines (2D) and spheroids (3D) may be the cause of observed differences in optimal T levels. Further study is necessary to support this claim.

Although the expression levels of certain genes (e.g., LCK, BMPR2, and ADAM19) showed more than a two-fold response to E2 or T treatments, these differences were not statistically significant, given the high variability among samples. This variability may be due to inconsistent prostate spheroid formation in the agarose-coated 96-well plates. Almost 100% of spheroids were formed. However, some spheroids (less than 5%) were not round in shape due to uneven surfaces of agarose-coated 96-well plates. Alternatively, this variability may have arisen from the use of two spheroids per sample for gene expression analyses, which was required to obtain sufficient quantities of RNA. Future studies should evaluate which materials (e.g., low-attachment conditions or Matrigel-coated plates) facilitate the formation of spheres with a consistent level of pluripotency. Moreover, our results should be validated with additional drugs/chemicals known to affect prostate cells using RNA extracted from a single prostate spheroid.

In conclusion, our study found that the expression of the stem cell-related genes, NANOG and CCL2, was altered in prostate spheroids treated with E2 and T. CCL2, NANOG, and KRT18 may be potentially useful markers to characterize prostate spheroids. Spheroid assays with HPrEpCs may supplement in vivo studies. However, we have not yet determined whether these gene expression changes are specific responses to treatment with E2 or T. Further studies are necessary to verify these results using additional test compounds and 3D cell culture models. In addition, we also need to examine if the identified markers can be used to characterize prostaspheres, formed from prostate stem/progenitor cells of prostate epithelial cells, to explore whether this model can be used for assessing drug toxicity on tumorigenesis in the prostate. Improvements in prostate spheroid formation methodology are required to reduce experimental variability.

ACKNOWLEDGMENTS

The authors thank Drs. Dayton Petibone, Richard Beger, and Pierre Alusta for their suggestions pertaining to the preparation of this manuscript. We also thank Dr. John K. Leighton for his suggestions related to the conceptualization of this project. The authors also thank Joanne Berger, FDA Library, for assistance with editing of the manuscript. This study was funded by the National Center for Toxicological Research (NCTR) (E07597.01). The views expressed herein are those of the authors and do not represent the views of the U.S. Food and Drug Administration.

Conflict of interest

The authors declare that there is no conflict of interest.

REFERENCES

•  Cho,  S.W.,  Kim,  Y.A.,  Sun,  H.J.,  Kim,  Y.A.,  Oh,  B.C.,  Yi,  K.H.,  Park,  D.J. and  Park,  Y.J. (2016): CXCL16 signaling mediated macrophage effects on tumor invasion of papillary thyroid carcinoma. Endocr. Relat. Cancer, 23, 113-124.
•  De Nunzio,  C.,  Presicce,  F. and  Tubaro,  A. (2016): Inflammatory mediators in the development and progression of benign prostatic hyperplasia. Nat. Rev. Urol., 13, 613-626.
•  Friedrich,  J.,  Seidel,  C.,  Ebner,  R. and  Kunz-Schughart,  L.A. (2009): Spheroid-based drug screen: considerations and practical approach. Nat. Protoc., 4, 309-324.
•  Fontana,  F.,  Raimondi,  M.,  Marzagalli,  M.,  Sommariva,  M.,  Gagliano,  N. and  Limonta,  P. (2020): Three-Dimensional Cell Cultures as an In Vitro Tool for Prostate Cancer Modeling and Drug Discovery. Int. J. Mol. Sci., 21, 6806.
•  Fujita,  K.,  Ewing,  C.M.,  Getzenberg,  R.H.,  Parsons,  J.K.,  Isaacs,  W.B. and  Pavlovich,  C.P. (2010): Monocyte chemotactic protein-1 (MCP-1/CCL2) is associated with prostatic growth dysregulation and benign prostatic hyperplasia. Prostate, 70, 473-481.
•  Gu,  G.,  Yuan,  J.,  Wills,  M. and  Kasper,  S. (2007): Prostate cancer cells with stem cell characteristics reconstitute the original human tumor in vivo. Cancer Res., 67, 4807-4815.
•  Guo,  L.,  Wang,  R.,  Zhang,  K.,  Yuan,  J.,  Wang,  J.,  Wang,  X.,  Ma,  J. and  Wu,  C. (2019): A PINCH-1-Smurf1 signaling axis mediates mechano-regulation of BMPR2 and stem cell differentiation. J. Cell Biol., 218, 3773-3794.
•  Härmä,  V.,  Virtanen,  J.,  Mäkelä,  R.,  Happonen,  A.,  Mpindi,  J.P.,  Knuuttila,  M.,  Kohonen,  P.,  Lötjönen,  J.,  Kallioniemi,  O. and  Nees,  M. (2010): A comprehensive panel of three-dimensional models for studies of prostate cancer growth, invasion and drug responses. PLoS One, 5, e10431.
•  Hedlund,  T.E.,  Duke,  R.C. and  Miller,  G.J. (1999): Three-dimensional spheroid cultures of human prostate cancer cell lines. Prostate, 41, 154-165.
•  Hu,  W.Y.,  Shi,  G.B.,  Lam,  H.M.,  Hu,  D.P.,  Ho,  S.M.,  Madueke,  I.C.,  Kajdacsy-Balla,  A. and  Prins,  G.S. (2011): Estrogen-initiated transformation of prostate epithelium derived from normal human prostate stem-progenitor cells. Endocrinology, 152, 2150-2163.
•  Hu,  W.Y.,  Hu,  D.P.,  Xie,  L.,  Li,  Y.,  Majumdar,  S.,  Nonn,  L.,  Hu,  H.,  Shioda,  T. and  Prins,  G.S. (2017): Isolation and functional interrogation of adult human prostate epithelial stem cells at single cell resolution. Stem Cell Res. (Amst.), 23, 1-12.
•  Hochberg,  Y. (1988): A sharper Bonferroni procedure for multiple tests of significance. Biomtrika, 75, 800-802.
•  Jeter,  C.R.,  Liu,  B.,  Liu,  X.,  Chen,  X.,  Liu,  C.,  Calhoun-Davis,  T.,  Repass,  J.,  Zaehres,  H.,  Shen,  J.J. and  Tang,  D.G. (2011): NANOG promotes cancer stem cell characteristics and prostate cancer resistance to androgen deprivation. Oncogene, 30, 3833-3845.
•  Kapałczyńska,  M.,  Kolenda,  T.,  Przybyła,  W.,  Zajączkowska,  M.,  Teresiak,  A.,  Filas,  V.,  Ibbs,  M.,  Bliźniak,  R.,  Łuczewski,  Ł. and  Lamperska,  K. (2018): 2D and 3D cell cultures - a comparison of different types of cancer cell cultures. Arch. Med. Sci., 14, 910-919.
•  Katt,  M.E.,  Placone,  A.L.,  Wong,  A.D.,  Xu,  Z.S. and  Searson,  P.C. (2016): In Vitro Tumor Models: Advantages, Disadvantages, Variables, and Selecting the Right Platform. Front. Bioeng. Biotechnol., 4, 12.
•  Kaufman,  J.M. and  Vermeulen,  A. (2005): The decline of androgen levels in elderly men and its clinical and therapeutic implications. Endocr. Rev., 26, 833-876.
•  Kwon,  O.J. and  Xin,  L. (2014): Prostate epithelial stem and progenitor cells. Am. J. Clin. Exp. Urol., 2, 209-218.
•  Langhans,  S.A. (2018): Three-Dimensional in Vitro Cell Culture Models in Drug Discovery and Drug Repositioning. Front. Pharmacol., 9, 6.
•  Lazennec,  G. and  Richmond,  A. (2010): Chemokines and chemokine receptors: new insights into cancer-related inflammation. Trends Mol. Med., 16, 133-144.
•  Li,  Z. and  Cui,  Z. (2014): Three-dimensional perfused cell culture. Biotechnol. Adv., 32, 243-254.
•  Livak,  K.J. and  Schmittgen,  T.D. (2001): Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)). Method. Methods, 25, 402-408.
•  Mittler,  F.,  Obeïd,  P.,  Rulina,  A.V.,  Haguet,  V.,  Gidrol,  X. and  Balakirev,  M.Y. (2017): High-Content Monitoring of Drug Effects in a 3D Spheroid Model. Front. Oncol., 7, 293.
•  Nakamura,  N.,  Davis,  K.,  Yan,  J.,  Sloper,  D.T. and  Chen,  T. (2020a): Increased estrogen levels altered microRNA expression in prostate and plasma of rats dosed with sex hormones. Andrology, 8, 1360-1374.
•  Nakamura,  N.,  Vijay,  V. and  Sloper,  D.T. (2020b): Gene expression profiling in dorsolateral prostates of prepubertal and adult Sprague-Dawley rats dosed with estradiol benzoate, estradiol, and testosterone. J. Toxicol. Sci., 45, 435-447.
•  Pampaloni,  F. and  Stelzer,  E. (2010): Three-dimensional cell cultures in toxicology. Biotechnol. Genet. Eng. Rev., 26, 117-138.
•  Paczkowski,  M.,  Kretzschmar,  W.W.,  Markelc,  B.,  Liu,  S.K.,  Kunz-Schughart,  L.A.,  Harris,  A.L.,  Partridge,  M.,  Byrne,  H.M. and  Kannan,  P. (2021): Reciprocal interactions between tumour cell populations enhance growth and reduce radiation sensitivity in prostate cancer. Commun. Biol., 4, 6.
•  Prins,  G.S. (2008): Endocrine disruptors and prostate cancer risk. Endocr. Relat. Cancer, 15, 649-656.
•  Prins,  G.S.,  Hu,  W.Y.,  Shi,  G.B.,  Hu,  D.P.,  Majumdar,  S.,  Li,  G.,  Huang,  K.,  Nelles,  J.L.,  Ho,  S.M.,  Walker,  C.L.,  Kajdacsy-Balla,  A. and  van Breemen,  R.B. (2014): Bisphenol A promotes human prostate stem-progenitor cell self-renewal and increases in vivo carcinogenesis in human prostate epithelium. Endocrinology, 155, 805-817.
• Rasband, W.S. (1997-2018): ImageJ. U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/.
•  Russell,  P.J. and  Voeks,  D.J. (2003): Animal models of prostate cancer. Methods Mol. Med., 81, 89-112.
•  Rycaj,  K. and  Tang,  D.G. (2017): Molecular determinants of prostate cancer metastasis. Oncotarget, 8, 88211-88231.
•  Shaffer,  J.P. (1995): Multiple hypothesis testing. Annu. Rev. Psychol., 46, 561-584.
•  Song,  W. and  Khera,  M. (2014): Physiological normal levels of androgen inhibit proliferation of prostate cancer cells in vitro. Asian J. Androl., 16, 864-868.
•  Takahashi,  K.,  Tanabe,  K.,  Ohnuki,  M.,  Narita,  M.,  Ichisaka,  T.,  Tomoda,  K. and  Yamanaka,  S. (2007): Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 131, 861-872.
•  Tsuyada,  A.,  Chow,  A.,  Wu,  J.,  Somlo,  G.,  Chu,  P.,  Loera,  S.,  Luu,  T.,  Li,  A.X.,  Wu,  X.,  Ye,  W.,  Chen,  S.,  Zhou,  W.,  Yu,  Y.,  Wang,  Y.Z.,  Ren,  X.,  Li,  H.,  Scherle,  P.,  Kuroki,  Y. and  Wang,  S.E. (2012): CCL2 mediates cross-talk between cancer cells and stromal fibroblasts that regulates breast cancer stem cells. Cancer Res., 72, 2768-2779.
•  Thoms,  S.,  Ali,  A.I.,  Jonczyk,  R.,  Scheper,  T. and  Blume,  C. (2018): Tacrolimus inhibits angiogenesis and induces disaggregation of endothelial cells in spheroids - Toxicity testing in a 3D cell culture approach. Toxicol. In Vitro, 53, 10-19.
•  Yao,  S.,  Till,  C.,  Kristal,  A.R.,  Goodman,  P.J.,  Hsing,  A.W.,  Tangen,  C.M.,  Platz,  E.A.,  Stanczyk,  F.Z.,  Reichardt,  J.K.,  Tang,  L.,  Neuhouser,  M.L.,  Santella,  R.M.,  Figg,  W.D.,  Price,  D.K.,  Parnes,  H.L.,  Lippman,  S.M.,  Thompson,  I.M.,  Ambrosone,  C.B. and  Hoque,  A. (2011): Serum estrogen levels and prostate cancer risk in the prostate cancer prevention trial: a nested case-control study. Cancer Causes Control, 22, 1121-1131.
•  Zink,  D.,  Chuah,  J.K. and  Ying,  J.Y. (2020): Assessing Toxicity with Human Cell-Based In Vitro Methods. Trends Mol. Med., 26, 570-582.
•  Zhao,  H.,  Sun,  N.,  Young,  S.R.,  Nolley,  R.,  Santos,  J.,  Wu,  J.C. and  Peehl,  D.M. (2013): Induced pluripotency of human prostatic epithelial cells. PLoS One, 8, e64503.