Culture conditions for the PrCa cell lines (C4-2B, LNCaP, PC3) have been previously described4,37,38. NCI-H660 cells were grown following ATCC instructions. PC3 cells were transfected using lentivirus shRNA clones to: Kindlin-2, clone TRCN0000128058 (which targets a coding sequence of K2); and non-targeting scrambled control SHC002 (purchased from Sigma). The lentivirus-mediated shRNA gene knockdown procedures were previously described in39,40. LNCaP cells were used for CRISPR/Cas9-mediated knockout of Kindlin-2 (FERMT2) as previously described41. Culture conditions for the PrCa cell lines 22Rv1 and VCaP were previously described42.
Generation of Kindlin-2 Knockout cell lines using electroporation
The sgRNA pool of 3 guide RNAs (G*A*C*GGGAUAAGGAUGCCAGA, C*G*C*GGUUCAGGUCCGUCACA and A*G*G*CGUGAUGCUUAAGCUGG with their respective Synthego modified EZ scaffolds) targeting FERMT2 and the scramble control were obtained from Synthego. The sgRNA pellets were rehydrated in 1X TE buffer (provided by Synthego) to make a stock of 100 μM. A working solution of 30 μM sgRNA was made (in nuclease free water) fresh before electroporating the cells. For every reaction, the Ribonucleoprotein (RNP) complex was assembled by adding 3 μl of30 μM sgRNA to 0.5 μl of 20 μM Cas9 (provided by Synthego) at a ratio of 9:1 in 3.5 μl resuspension buffer R (Neon Transfection System; Invitrogen Basel, Switzerland) and incubated for 10 min at room temperature.
Electroporation of LNCaP cells was achieved by an implemented electroporation device system according to manufacturer’s instructions (Neon Transfection System; Invitrogen, Basel, Switzerland). The Neon Transfection System 10 μL kit was used for the transfection of human prostate cancer cells. Cells were cultured 48 h before electroporation and harvested at nearly 80% confluency. Cells at a density of 2 × 105 were washed with PBS and resuspended in 5 μl resuspension buffer R (Neon Transfection System; Invitrogen Basel, Switzerland). Within 15 min of resuspension, the cells were added to the tube containing RNP and the cell-RNP complex was electroporated with the Neon Transfection System. Per electroporation, 2 × 105 cells were taken up in a 10 μl Neon tip using the Neon Transfection System pipette (Invitrogen). The electroporation was performed by applying 3 pulses at 1450 Volts for 10 ms to PC3 cells and 2 pulses at 1200 Volts for 20 ms to LNCaP cells. Control cells were incubated with the resuspension buffer without the sgRNA and electroporated at the same settings. After electroporation, the cells were seeded in a 6-well plate by adding 2.5 ml DMEM (Cytiva) with 10% FBS without antibiotic supplements. The cells were cultured for 48–72 h and subsequently proceeded for further analysis. Western blot analysis was used to assess the Kindlin-2 KO efficiency in all cell lines.
The following antibodies (Abs) were used: for the immunoblotting (IB) analysis, rabbit monoclonal Abs against the αVβ3 integrin (13166S, Cell Signaling) and Aurora Kinase A (14475S, Cell Signaling), polyclonal goat Abs against the αVβ6 integrin (AF2389, R&D system) and NgR2 (AF2776, R&D system), rabbit polyclonal Abs against calnexin (CANX, sc11397, Santa Cruz), actin (a2066, Sigma), RhoA (sc-179, Santa Cruz), TSG101 (Abcam, ab30871), mouse monoclonal Abs against RhoA (sc-418, Santa Cruz), NSE (LS-C197136, LSBio), Kindlin-2 (MAB2617, Millipore), were also used. For immunohistochemical analysis, rabbit monoclonal Ab against the β3 integrin (13166S, Cell Signaling), rabbit polyclonal Abs against SYP (PA1-1043, Invitrogen), and NgR2 (PA5-98577, Invitrogen) were also used. Rabbit IgG (I5006, Sigma) was used as negative control. For immunoprecipitation, rabbit polyclonal Ab against NgR2 (PA5-98577, Invitrogen), rabbit monoclonal Ab against the β3 integrin (13166S, Cell Signaling), mouse monoclonal Abs against the β6 (62A1) and β1 integrins (NBP2-52708, Novus) were used. For the adhesion assay, the αVβ3 integrin (LM609, Millipore MAB1976), and the non-immune mouse IgG (02-6502, Thermo Fisher) were used.
PC3 cells were lysed with lysis buffer (50 mM Tris–HCl pH 7.2, 150 mM NaCl, 1% Triton X-100, 1 mM Na3VO4, 1 mM Na4O7P2, 50 mM NaF, 0.01% aprotinin, 4 μg/ml pepstatin, 10 μg/ml leupeptin, 1 mM PMSF, 1 mM CaCl2, 1 mM MgCl2, 1 μM Calpain inhibitor) and pre-clearing was performed by two consecutive incubations with protein G-Sepharose (17061801, Cytivia) at 4 °C for 30 min. Binding to specific Abs was performed by incubation at 4 °C overnight, followed by incubation with protein G-Sepharose at 4 °C for 3 h. After six washes with lysis buffer, immunocomplexes were resuspended in 1X reducing Laemmli buffer and separated by SDS-PAGE. All western blotting films were developed using the Protec Optimax developer system. The film’s images were acquired using a Microtek ArtixScan M2 and processed using the Microtek Scan Wizard Pro V8.20 software.
PrCa cell transfection
PC3 cells were transfected with three different shRNA constructs that target RTN4RL2 (SMARTvector, Dharmacon/Horizon, SO-2914049G, sequences: V3SVHS00_4716901 for shRTN4RL2_1, and V3SVHS00_8801245 for shRTN4RL2_3). As a control, PC3 cells were transfected with a non-targeting scrambled control shRNA (SMARTvector, Dharmacon/Horizon, VSC11707). DU145 cells were transfected with pCMV6-Entry vector carrying RTN4RL2 (Origene, SC310413) or with the empty vector as a control (Origene, PS100001). Cells were plated (2.5 × 105) in a 6-well plate and grown overnight at 37 °C. The next day, cells were washed with PBS and incubated with 1 mL serum-free media at 37 °C for 2 h. For transfection, 4 μg of plasmid was mixed with 12 μL of lipofectamine 2000 (Invitrogen, 11668-019) in 200 μL of serum-free RPMI media. The plasmid-lipofectamine mix was incubated at room temperature for 25 min. The mix was then added drop-wise to the cells and incubated at 37 °C for 6 h. After 6 h, 700 μL of growth medium (without pen-strep) was added to the cells and incubated at 37 °C overnight. After 24 h, the medium was replaced with the growth medium, and cells were incubated for 48 h at 37 °C. For the selection of the transfected cells, the growth medium was supplemented with puromycin (3 μg/mL) for PC3 cells and G418 (0.5 mg/ml) for DU145 cells. The pooled populations of selected cells were maintained in complete media containing 2 μg/mL of puromycin for PC3 cells and 0.5 mg/ml of G418 for DU145 cells.
Downregulation of RTN4RL2 (NgR2) expression was accomplished using siRNA SMARTPool (L-008045-00-0010, Dharmacon/Horizon). For the downregulation of ITGB3 FlexiTube (Qiagen) siRNAs were used. In Fig. 1: siRNA_2, SI00004599 (target sequence, CTCTCCTGATGTAGCACTTAA), siRNA_3, SI00004606 (target sequence, CAAGCTGAACCTAATAGCCAT), and siRNA_4, SI02623159 (target sequence, CACGTGTGGCCTGTTCTTCTA) were used.
Transfection of the siRNA and immunoblotting analysis were performed as previously described14. Briefly, 300,000 cells were transfected with siRNA (final concentration 100 nM) duplexes using oligofectamine at a final concentration of 20 nM. Twenty-four hours after transfection, the siRNAs were removed, and the cells were kept in complete media overnight. This process was repeated for a second time, after which the cells were harvested and analyzed.
The heatmap of gene expression in cell lines was generated using Broad Institute Morpheus software (MA, USA). Statistical analysis was done using GraphPad Prism (CA, USA) and differences between two groups were compared by unpaired student’s t-test.
RNA sequencing of metastatic CRPrCa samples
RNA sequencing analysis of metastatic CRPrCa specimens acquired through rapid autopsy of 98 patients was performed as previously described43. Specimens were classified based on their levels of AR and NE markers. AR-positive and NE-negative (n = 76), AR low NE-negative (n = 13), AR-positive NE-positive (n = 9), AR-negative NE-negative (n = 8), and AR-negative NE-positive (n = 11).
Generation of mice carrying prostate-specific deletions
Mice of genotype PB-Cre4 ptenloxP/loxP rb1loxP/loxP trp53loxP/loxP (TKO), and PB-Cre4 ptenloxP/loxP (SKO) were generated as previously described44. Mice with the same genetic background (mixed C57BL/6 and 129SVJ) were used for the wild-type control.
Generation of NgR2 knockout mice
NgR2-null, and NgR1-null mice have been previously described45,46,47. Wild-type (WT) littermate controls, NgR2−/− and NgR1−/− animals were obtained by mating heterozygous NgR2+/− and NgR1+/− animals, respectively.
Care of animals followed standards established by the Office of Laboratory Animal Welfare, NIH, Department of Health and Human Services. All mice were maintained following the recommendations of the Institutional Animal Care and Use Committee which is a standing committee mandated by Federal law and regulations that ensures the humane and ethical treatment of animals. All experimental protocols were approved by the Animal Care and Use Committees at the institutions where the mice were hosted: Thomas Jefferson University for CB-17 SCID mice; Roswell Park Cancer Center for TKO mice and the Medical University of Innsbruck, Austria for NgR2-null and NgR1-null mice. This study is reported in accordance with ARRIVE guidelines.
Immunohistochemical analysis was performed on tissue sections of 5 TKO (primary tumor and lung metastases), and LuCaP PDXs TMA as previously described13. The tissue sections were incubated overnight at 4 °C with Abs against the β3 integrin subunit (1:25), NgR2 (1:500 for mouse samples and 1:1000 for LuCaP PDXs TMA), SYP (1:200), or the IgG isotype, which was used as the negative control. The following day, the tissue sections were washed with PBST (5 min × 2), followed by PBS (5 min), and incubated with biotinylated goat anti-rabbit IgG (BA-1000, Vector Laboratories) in PBST for 30 min at room temperature. Pictures were acquired using an Olympus BX43 microscope and the imaging processing was performed using the Olympus Sens Entry V2.3 software.
The specificity of our IHC staining was confirmed by staining dorsal root ganglion sections samples from three NgR2 knockout mice (Fig. S1A45) or by preincubating for 1 h the NgR2 Ab (Invitrogen, PA5-98577) with the blocking peptide (DSRGRQGGDAPTEDDYWG, 10 μg/ml, Thermo Fisher) that is specifically recognized by this Ab (Fig. S1B). The preincubated Ab was then used for immunostaining of rat brain samples or primary tumors from a NEPrCa patient (Fig. S1B).
Human subject inclusion criteria
The NEPrCa tissue sample was obtained from the Department of Pathology at Thomas Jefferson University (Philadelphia, PA). The specimen was de-identified and discarded in accordance with guidelines established by the Institutional Review Board (IRB), an administrative body established to protect the rights and welfare of human research subjects recruited to participate in research activities conducted at Jefferson.
LuCaP TMA immunohistochemical assessment and statistical analysis
The immunostaining of each LuCaP was scored as previously reported13.
PC3 cells treated with oligofectamine, non-silencing siRNA, or siRNA specifically targeting RTN4RL2 were plated (104) in 100 µL complete media on 96 well plates (six replicates) and analyzed using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; 5 mg/mL).
Boyden chamber assay
PC3 cells treated with oligofectamine, non-silencing siRNA, or siRNA specifically targeting RTN4RL2 were seeded (5 × 104) on fibronectin-coated (10 µg/ml) Transwell chambers (three replicates) in serum-free media and analyzed as been previously described38.
Anchorage-independent growth assay
Six-well plates were coated with 0.8% agarose to create a bottom layer and 10,000 PC3 cells (parental, scramble, or shRTN4RL2) from each well were resuspended in 2 mL of complete medium (RPMI containing 10% FBS, 100 μg/mL streptomycin, and 100 U/mL penicillin). The protocol has been previously described12. Pictures were acquired using a Nikon Eclipse TS100 microscope and the imaging processing was performed using the NIS Elements F3.0 software.
Adhesion assays were performed as previously described48.
αVβ3 activation assay
Activation assays were performed as previously described49.