LE300, L-741,626, PG 01037, PD 168568, and SCH 39166 were purchased from Tocris. Antibodies against Akt, phosphor-Akt, AMPKa, phospho-AMPKa, mTOR, phospho-mTOR, SAPK/JNK, phospho-SAPK/JNK, Oct4, Klf4, c-Myc, and b-actin were purchased from Cell Signaling Technology. Unless otherwise stated, all other reagents were purchased from Merck KGaA.

The human prostate cancer cell line PC-3 was obtained from the American Type Culture Collection and cultured in RPMI-1640 medium supplemented with 10% (v/v) heat-inactivated fetal bovine serum (FBS). Cell cultures enriched for CSC-like cells derived from PC-3 cells were prepared according to a previously described protocol (17). Briefly, PC-3 cells were trypsinized, harvested after washing with phosphate-buffered saline (PBS), and then suspended in serum-free DMEM/F12 medium (R&D Systems) supplemented with 100 IU/ml penicillin, 100 µg/ml streptomycin, 10 ng/ml human recombinant epidermal growth factor (hrEGF; R&D Systems), 10 ng/ml human recombinant basic fibroblast growth factor (hrbFGF; R&D Systems), and 2% B27 supplement (Thermo Fisher Scientific, Inc.). These suspended cells were cultured in ultralow attachment culture dishes at a density of 1×10⁶ cells/dish and then allowed to form tumorspheres for 7 days. Single cells obtained via the trypsinization of the tumorspheres were collected and allowed to form an adherent monolayer culture in a regular animal cell culture ware in the abovementioned serum-free medium and maintained in serum-free DMEM/F12 medium supplemented with 100 IU/ml penicillin, 100 µg/ml streptomycin, 10 ng/ml hrEGF, 10 ng/ml hrbFGF, and 2% B27 supplement.

The cells were seeded into a four-chamber plate (Falcon) at a density of 1×10⁴ cells/well, incubated for 72 h, and fixed with 3.7% formaldehyde at room temperature for 10 min. Fixed cells were permeabilized with 0.1% Tween-20 in PBS (pH 7.4) for 5 min, blocked with PBS containing 1% bovine serum albumin (BSA) for 30 min, and incubated with a rabbit monoclonal antibody against DRD1, DRD2, DRD3, DRD4, and DRD5 at 4°C overnight. The cells were washed with PBS and incubated in the presence of goat antirabbit IgG conjugated with Alexa Fluor^™ 594 (Thermo Fisher Scientific, Inc.) for 1 h in the dark. Alexa Fluor^™ 488 Phalloidin (Thermo Fisher Scientific, Inc.) diluted to a ratio of 1:40 using PBS was added to the wells and incubated for another 20 min. The cells were washed with PBS, stained with DAPI for 5 min, and observed under an LSM 700 confocal microscope (Zeiss).

The Institutional Animal Care & Use Committee (IACUC) of Korea University (Seoul, Korea), reviewed and approved the animal study protocol (protocol number: KUIACUC-2021-0028). Six- to seven-week-old female Balb/c nude mice (Charles River Laboratories) were maintained as previously described (21). Cancer cells (5×10⁶ cells/200 µl/mouse) suspended in PBS (pH 7.4) were injected subcutaneously into the right flank of a mice. Tumor volumes were measured 2–3 times weekly using a Vernier caliper and were calculated using the following equation: Tumor volume=Dmin² × Dmax × 0.5 (Dmin: short axis, Dmax: long axis of mass). The experiments were stopped before the tumor volume reached 1,000 mm³ and the mice were euthanized by introducing CO₂ into the euthanasia chamber at a rate sufficient to fill 50% of the chamber's volume per minute. The mice were continuously monitored for visible signs of death, such as cessation of breathing and lack of movement. Even after observing the visible signs of death, CO₂ exposure were continued for an additional 1 min. After then, the tumors were collected from the mice.

PC-3-derived CSC-like cells were seeded into tissue culture plates at a density of 3×10⁴ to 1×10⁵ cells/ml depending on the type of experiment using antibiotic-free DMEM/F12 medium with supplements and growth factors. To silence the expression of the DRD subtype genes, the cells were transfected for 6 h with 50–100 pmol/ml of control siRNA and DRD1, DRD2, DRD3, DRD4, and DRD5 siRNA separately using the siRNA Reagent System (Santa Cruz Biotechnology, Inc.) following the manufacturer's instructions (see Table SI for more information about siRNA).

Total RNA was extracted from cells using a GeneAll kit (GeneAll Biotechnology), and cDNA was prepared and amplified using a One Step RT-PCR kit (SolGent). Table SII shows the primer sequences for the dopamine receptor subtypes (DRD1, DRD2, DRD3, DRD4, and DRD5) and GAPDH. The PCR products were subjected to electrophoresis using 1.5–1.8% agarose gel and visualized after staining with ethidium bromide.

PC-3-derived CSC-like cells were transfected with siRNA against DRD1-DRD5 and seeded in 96-well plates at a density of 5×10⁴ cells/well. After 24 h of incubation, the intracellular cAMP levels were assayed using cAMP-Glo™ assay (Promega) following the manufacturer's instructions. Luminescence was measured using a Centro LB 960 microplate luminometer (Berthold Technologies).

To measure the growth of PC-3-derived CSC-like cells transfected with DRD siRNAs, the WST-8 assay (Biomax) was used. Briefly, 24 h after seeding the cells in a six-well culture plate at a density of 1×10⁴ cells/ml, WST-8 assay was conducted at designated times and absorbance was measured at 450 nm. To determine the growth of DRD2 heterozygous knockout PC-3 cells, the cells were seeded into a six-well culture plate at a density of 1×10⁴ cells/ml and grown in RPMI-1640 medium supplemented with 10% heat-inactivated FBS at 37°C. The cells were trypsinized and harvested at designated times and counted using an automated cell counter (Countess^™ II #AMQX1000).

The insert polycarbonate membranes (8 mm in pore size) of the upper compartments of the 24-well Transwell Boyden chamber were coated with Matrigel^® (BD Biosciences) diluted using a serum-free medium. Depending on the cell type, 300 µl of cells (1.5×10⁵ cells/ml) suspended in RPMI-1640 or DMEM/F12 medium were placed in the upper compartment. The lower compartment was then filled with 500 µl of RPMI-1640 medium supplemented with 10% FBS or serum-free DMEM/F12 medium supplemented with 2% B-27 supplements and growth factors. The cells on the lower side of the insert membrane were fixed with 10% trichloroacetic acid (TCA) and stained with 0.5% crystal violet in 2% ethanol for 2 h. The insert was washed with PBS and air dried. The upper side of the insert was wiped using a cotton swab, and images of the cells that had migrated to the underside of the insert membrane were taken. The dye that stained the cells was extracted with 0.2 ml of 30% acetic acid, and the absorbance was measured at 590 nm.

Protein extracts (20 µg) of cell lysates were resolved on 8–10% SDS-polyacrylamide gels and transferred to Immobilon-P transfer membranes as described previously (22). The membranes were blocked with Tris-buffered saline containing 0.1% Tween-20 (TBST) supplemented with 0.5–2% BSA and probed with primary antibodies. After washing with TBST, the membranes were probed with species-specific horseradish peroxidase (HRP)-conjugated secondary antibodies and developed using an Immobilon western chemiluminescent HRP substrate.

The cells incubated for 6 days in the presence and absence of compounds were fixed with 50% TCA and stained with 0.4% sulforhodamine B (SRB) in 0.1% acetic acid. Excess staining was removed by washing the cells with 1% acetic acid. The morphology of the stained cells was observed using an inverted microscope and photographed (Nikon).

PC-3 cells were seeded into 60-mm dishes at a density of 1.5×10⁵ cells/ml and cultured in an antibiotic-free medium. When the cells were at ~80% confluency, they were transfected with an all-in-one vector (Macrogen, Seoul, Korea) containing the DRD2 sgRNA sequence (5′-GGTATGATGATGATCTGGAGAGG-3′), puromycin resistance gene, and CAS9 expression gene using the TransIT-LT1 transfection reagent (Mirus Bio) following the manufacturer's instructions (see Fig. S1 for the all-in-one vector map). After 48 h, the cells were cultured on RPMI medium supplemented with 100 IU/ml penicillin, 100 µg/ml streptomycin, and 1 µg/ml puromycin for 6 days. During this period, the medium was changed daily. The surviving cells were amplified, and genomic DNA was isolated using Exgene^™ Cell SV mini kit (GeneAll Biotechnology, Seoul, Korea). DNA fragments were then amplified by PCR with the DRD2 primers used in the T7 endonuclease assay and sequenced using the Sanger sequencing method (Cosmogenetech) (See the DNA sequences in Fig. S2).

To validate DRD2 heterozygous knockout in PC-3 cells, genomic DNA isolated from wild-type and all-in-one vector-transfected PC-3 cells were amplified using PCR with the DRD2 primers (forward: 5′-TGTGTTTGCTCATTTGTCCTACC-3′, reverse: 5′-AGGAAACAATCTACCCATTTCGT-3′). The amplified DNA products were incubated at 37°C for 20 min in the presence of T7 endonuclease (Goldbio). The reaction products were then subjected to electrophoresis using 1.2% agarose gel and visualized after staining with ethidium bromide.

Wild-type, mock-transfected, and DRD2^+/− PC-3 cells grown in RPMI medium containing 10% heat-inactivated FBS were trypsinized and cultured at a density of 1×10⁴ cells/well in an ultralow attachment six-well plate containing DMEM/F12 medium supplemented with 2% B27, 10 ng/ml hrEGF, and 10 ng/ml hrbFGF. Round cell clusters of >40 µm were classified as spheres.

GraphPad Prism 5.03 (GraphPad Software, Boston, Massachusetts, USA) was used for all statistical analyses. Statistical differences among experimental groups were analyzed using a unpaired Student's t-test or one-way/two-way ANOVA with Dunnett's post hoc analysis. P<0.05 was considered to indicate a statistically significant difference.

Our previous study showed that cell cultures prepared for enriching CSC-like cells from PC-3 cells possessed several properties characteristic of CSC-like cells, such as the expression of Oct4, Klf4, and Sox2. These PC-3-derived CSC-like cells were more resistant to some agents with anticancer activity such as dasatinib and saracatinib, tyrosine kinase inhibitors, and tirbanibulin, a tubulin polymerization inhibitor, than PC-3 cells (Fig. S3). In addition, the mRNA of all DR subtypes, which were of primary interest in our study, were expressed in PC-3 cells and PC-3-derived CSC-like cells (17). Immunocytochemical analysis revealed that the proteins of the DR subtypes were also expressed in these cells (Fig. 1). Capturing stained images of DRD3 was challenging, as expected from the extremely low expression of its mRNA; however, the presence of all five subtypes was detected in both the cell types.

Thioridazine used in our previous study (17) is a DRD2 antagonist but it also binds and inhibits various other receptors, including serotonin and cholinergic receptors. Thus, more specific antagonists against each DR subtype, such as LE300, L-741,626, PG 01037, PD 168568, and SCH39166, were applied to PC-3-derived CSC-like cells at a concentration of 3 µM. Subsequently, the morphological changes of the cells and the expression of pluripotency markers such as Oct4, Klf4, and c-Myc were analyzed (Fig. 2). L-742,626, a DRD2 antagonist, strongly induced morphological changes in the cells, a strong decrease in Oct4 and Klf4 protein expression, and a strong increase in c-Myc protein expression, indicating the loss of CSC properties. These results are consistent with those of our previous study (17). Meanwhile, PD 168568, a DRD4 antagonist, also induced changes in the morphology and Oct4, Klf4, and c-Myc protein expression in the cells, although the effects were weaker than those induced by L0741,626. PG-01037 induced very weak morphological changes in the cells and a slight decrease in Oct4 protein expression; however, further studies are needed to determine whether DRD2 is involved in maintaining the properties of PC-3-derived CSC-like cells.

To determine whether a specific DRD2 antagonist inhibits tumor growth in vivo, L-741,626 (20 mg/kg) was administered intraperitoneally to nude mice with tumors formed from PC-3-derived CSC-like cells daily. L-741,626 induced 59.5% tumor growth inhibition compared with the control group without significant changes in body weight (Fig. 3).

Because the inhibition of DRD2 in PC-3-derived CSC-like cells by thioridazine, a DRD2 antagonist, appeared to induce a loss of CSC characteristics in our previous study (17), it was necessary to determine whether this effect could be mimicked by the knockdown of any DR subtype. siRNA against all five subtypes was transfected into PC-3-derived CSC-like cells. Semiquantitative RT-PCR revealed that the mRNA of each subtype was effectively downregulated by its corresponding siRNA (Fig. 4).

While DRD2, DRD3, and DRD4 are type II dopamine receptors coupled to the Gs protein, DRD1 and DRD5 are type I dopamine receptors coupled to the G_i protein (23). To determine whether the siRNA-mediated knockdown of DR subtypes is functionally effective, intracellular cAMP concentrations were measured (Fig. 5). The knockdown of type I dopamine receptors suppressed intracellular cAMP concentrations to the control levels, whereas the knockdown of type II dopamine receptors increased intracellular cAMP concentrations.

The growth of PC-3-derived CSC-like cells transfected with DRD2 siRNA decreased substantially with time (Fig. 6). Transfection with DRD4 siRNA also inhibited cell growth, but the effect was less pronounced than that caused by DRD2 siRNA. A slight but statistically significant increase in the growth of cells transfected with DRD1 or DRD3 on day 6 was observed. In contrast to the effects on cell growth, in vitro cell invasion remained unaffected by DRD1 and DRD3 siRNA (Fig. 7). However, it was substantially inhibited by DRD2 siRNA and slightly by DRD4 siRNA.

Our previous study showed that AMPK phosphorylation was inhibited by thioridazine, a DRD2 antagonist (17). To examine whether AMPK phosphorylation is inhibited by DR knockdown, PC-3-derived CSC-like cells were transfected with DR siRNAs. Fig. 8 shows that only DRD2 siRNA significantly inhibited AMPK phosphorylation. mTOR phosphorylation, which was unaffected by thioridazine, was also unaffected by transfection with siRNA against DR subtypes.

Because the suppression of DRD2 activity or expression with a DRD2 antagonist or siRNA changed the characteristics of PC-3-derived CSC-like cells, it was necessary to investigate whether the suppression of DRD2 expression in parental PC-3 cells affected their intrinsic characteristics, such as sphere formation ability. DRD2 heterozygous knockout (DRD2^+/−) PC-3 cells were prepared, and the heterozygotic genetic mutation was validated using the T7 endonuclease digestion method (Fig. 9A). The downregulation of DRD2 expression was observed (Fig. 9B and C). The growth of DRD2^+/− PC-3 cells was substantially slower than that of mock-transfected PC-3 cells. Addition of 30 µM of A769662, an AMPK activator, to DRD2^+/− PC-3 cells partially restored the cell growth (Fig. 9D). The concentration of 30 mM was chosen from the previous study as it showed the highest effects in restoring cell growth (17). The morphology of the DRD2^+/− PC-3 cells differed substantially from that of the mock-transfected PC-3 cells (Fig. 9E). The sphere formation assay, an in vitro method for amplifying and isolating CSCs, revealed that the sphere-forming ability of the DRD2^+/− PC-3 cells was substantially lower than that of the parental and mock-transfected PC-3 cells (Fig. 9F and G). Because AMPK phosphorylation in PC-3-derived CSC-like cells is inhibited by DRD2 siRNA, the effect of heterozygous DRD2 knockout on the phosphorylation of several signal transduction pathways, including AMPK, was examined. As shown in Fig. 10, AMPK phosphorylation in DRD2^+/− PC-3 cells was substantially decreased compared with that in the parental and mock-transfected PC-3 cells, whereas there were no significant differences in mTOR, SAPK/JNK, and Akt phosphorylation in the parental, mock-transfected, and DRD2^+/− PC-3 cells, indicating that AMPK activity or phosphorylation is crucial for maintaining or inducing CSC properties.