Non-genomic mechanisms mediate androgen-induced PSD95 expression

T-BSA rapidly increased PSD95 expression via membrane binding sites for androgen

To assess the impact of membrane-impermeable T-BSA on PSD95 expression, we first analyzed its time- and dose-dependent effects in HT22 cells. Western blotting analysis showed that treatment with 10 nM T-BSA (soluble in DMSO) significantly increased PSD95 protein expression at 30 min and 60 min but did not change the expression levels at 0 min, 5 min and 15 min (Figure 1A and 1B). T-BSA treatment significantly increased PSD95 expression at concentrations of 10 nM and 15 nM compared with DMSO (the vehicle group), 5 nM T-BSA and 20 nM T-BSA groups (Figure 1C and 1D). Given that incubation with 10 nM T-BSA for 30 min significantly increased PSD95 expression, this treatment condition was used in the following experiments.

Figure 1. T-BSA rapidly increased PSD95 expression through a transcription-independent mechanism. (A and B) Time-dependent effects of T-BSA (0 min, 5 min, 15 min, 30 min and 60min) on PSD95 protein levels (n=5). (C and D) Dose-dependent effects of T-BSA (DMSO, 5 nM, 10 nM, 15 nM and 20 nM) on PSD95 protein levels (n=5). (E and F) FITC signals on the HT22 cell plasma membrane (n=5, scale bars = 50 μm). (G and H) Western blotting for PSD95 expression induced by T-BSA in HT22 cells pre-treated with 10 μM Act D or 200 μM CHX for 2 h (n=5). (I and J) Immunofluorescence staining for PSD95 induced by T-BSA in HT22 cells pre-treated with Act D or CHX (n=4, scale bars = 20 μm). (K and L) RT-PCR for PSD95 mRNA level induced by T-BSA (n=5). (n.s.: non-significant; * P < 0.05; ** P < 0.01).

We further verified the presence of membrane androgen-binding sites by using T-BSA conjugated FITC. The non-specific binding of T-BSA-FITC was evaluated using BSA conjugated FITC lacking a testosterone moiety. The binding of BSA-FITC was found to be negligible, so that non-specific binding via the BSA or FITC moiety of the T-BSA-FITC molecule can be excluded (Figure 1E and 1F).

T-BSA increased PSD95 expression in a transcription-independent manner

To determine whether the effect of T-BSA on PSD95 expression depends on a transcription and/or translation process, we measured PSD95 protein level in HT22 cells after pre-treatment with the transcription inhibitor Act D or the translation inhibitor CHX. Western blotting and immunofluorescent staining analysis showed that CHX effectively inhibited the effect of T-BSA on PSD95 expression compared with T-BSA group. However, PSD95 expression had no significant change after treatment with Act D (Figure 1G–1J). Moreover, RT-PCR analysis revealed that T-BSA treatment did not affect the PSD95 mRNA level compared with DMSO treatment (Figure 1K and 1L).

T-BSA affected PSD95 expression via ZIP9 but not membrane-localized AR

Since the results described above suggested that T-BSA increased PSD95 expression through non-genomic action, we next sought to identify whether these effects are mediated by membrane-localized AR and/or ZIP9. The AR antagonist flutamide was used to competitively bind AR to prevent them from functioning normally. Western blotting and immunofluorescent staining analysis revealed that PSD95 protein level in T-BSA treatment did not significantly differ from that in HT22 cells treated with flutamide plus T-BSA treatment (Figure 2A–2D).

Figure 2. T-BSA increased PSD95 expression via ZIP9 rather than membrane-localized AR. (A and B) Western blotting for PSD95 expression induced by T-BSA in HT22 cells pre-treated with flutamide (F, 100 μM) for 1 h (n=5). (C and D) Immunofluorescence staining for PSD95 induced by T-BSA in HT22 cells pretreated with flutamide (n=4, scale bar = 20 μm). (E and F) Western blotting for the knockdown efficiency of ZIP9 protein in HT22 cells transfected with nc-shRNA or ZIP9-shRNA (n=5). (G and H) Western blotting for PSD95 expression induced by T-BSA in HT22 cells pre-treated with nc-shRNA or ZIP9-shRNA (n=5). (I and J) Immunofluorescence staining for PSD95 induced by T-BSA in HT22 cells pre-treated with nc-shRNA or ZIP9-shRNA (n=4, scale bars = 20 μm). (n.s.: not-significant; * P < 0.05; ** P < 0.01).

To determine the role of ZIP9 in the upregulation of PSD95 induced by T-BSA, we then transfected HT22 cells with ZIP9-shRNA or negative control shRNA (nc-shRNA). The protein expression of ZIP9 in ZIP9-shRNA group was markedly lower than that in nc-shRNA group (Figure 2E and 2F). Western blotting and immunofluorescent staining analysis showed that ZIP9-shRNA treatment significantly reduced T-BSA-induced PSD95 levels compared with nc-shRNA plus T-BSA group (Figure 2G–2J). These data suggest that ZIP9 mediates the non-genomic action of T-BSA on PSD95 expression in HT22 cells, while the membrane-localized AR is not involved in this process.

Formation of ZIP9/Gnα11 complex is required in T-BSA-induced PSD95 expression

We used Co-IP to determine the interaction of ZIP9 and Gnα11 detected in HT22 cells following T-BSA treatment (Figure 3A). Western blotting analysis confirmed that Gnα11-shRNA significantly reduced Gnα11 protein level compared with the cells that received the nc-shRNA treatment (Figure 3B and 3C). We then used PLAs which detecting protein-protein interactions in fixed cells [26] to further confirm the direct interaction between ZIP9 and Gnα11. The red color in Figure 3D represented with a strong ZIP9/Gnα11 interaction signals observed in HT22 cells treated with nc-shRNA plus T-BSA. However, knockdown of ZIP9 or Gnα11 expression with ZIP9-shRNA or Gnα11-shRNA significantly reduced this interaction compared with nc-shRNA treatment (Figure 3D and 3E). Meanwhile, treatment with either ZIP9-shRNA or Gnα11-shRNA significantly abolished T-BSA-induced upregulation of PSD95 protein expression levels compared with the nc-shRNA plus T-BSA group (Figure 3F and 3G).

Figure 3. ZIP9/Gnα11 interactions mediate the effect of T-BSA on PSD95 expression. (A) Co-IP for the interaction between ZIP9 and Gnα11 induced by T-BSA. (B and C) Western blotting for the knockdown efficiency of Gnα11 protein in HT22 cells transfected with nc-shRNA or Gnα11-shRNA. (D and E) Duolink^® proximity ligation assay for the interaction between ZIP9 and Gnα11 induced by T-BSA pre-treated with nc-shRNA, ZIP9-shRNA or Gnα11-shRNA. (F and G) Western blotting for PSD95 expression induced by T-BSA in HT22 cells pre-treated with nc-shRNA, ZIP9-shRNA or Gnα11-shRNA. (n = 5; n.s.: no-significant; * P < 0.05; ** P < 0.01).

T-BSA upregulated the phosphorylation levels of Erk1/2 and eIF4E by ZIP9/Gnα11

To determine the downstream molecular mechanisms following activation of ZIP9/Gnα11, Flowmetric Luminex xMAP^® assay was carried out to examine the effect of T-BSA on members in the mitogen-activated protein kinase (MAPK)-related signaling pathways in HT22 cells. T-BSA treatment did not change total Erk1/2 expression level but significantly increased phosphorylated Erk1/2 level compared with that in DMSO-treated HT22 cells. Furthermore, T-BSA treatment has little effect on either total levels or phosphorylated levels of P38 and JNK (Figure 4A–4C). Knockdown of Gnα11 expression with shRNA dramatically inhibited T-BSA-induced increase in phosphorylated-Erk1/2 (p-Erk1/2) level. The role of ZIP9/Gnα11 in T-BSA-induced increase in p-Erk1/2 was further confirmed by western blot and immunofluorescent staining analysis. The results showed that the p-Erk1/2 levels in ZIP9-shRNA plus T-BSA group and Gnα11-shRNA plus T-BSA group were significantly lower than those in nc-shRNA plus T-BSA group (Figure 4D–4H).

Figure 4. T-BSA upregulated the phosphorylation level of Erk1/2 and eIF4E through ZIP9/Gnα11. (A–C) The ratio of phosphorylated/total levels of P38 (A), JNK (B), and Erk1/2 (C) detected in HT22 cells pre-treated with nc-shRNA or Gnα11-shRNA using Flowmetric Luminex xMAP^® assay (n=5). (D–F) Western blotting for the phosphorylated/total levels of Erk1/2 and eIF4E induced by T-BSA pre-treated with nc-shRNA, ZIP9-shRNA, or Gnα11-shRNA (n=5). (G–J) Immunofluorescence staining for the phosphorylation level of Erk1/2 and eIF4E induced by T-BSA pre-treated with nc-shRNA, ZIP9-shRNA or Gnα11-shRNA (n=4, scale bars = 20 μm). (n.s.: non-significant; * P < 0.05; ** P < 0.01).

The finding that CHX treatment removed T-BSA-induced PSD95 expression suggest that this action may involve the regulation of the translation process. The eukaryotic translation initiation factor (eIF) 4E, which involved in the recruitment of the ribosome to the mRNA cap structure, integrates input from Erk1/2 signaling to regulate the translation of a subset of mRNAs [27, 28]. Western blotting and immunofluorescent staining analysis showed that knockdown of ZIP9 and Gnα11 expression significantly reduced phosphorylated-eIF4E (p-eIF4E) level compared with the nc-shRNA plus T-BSA group (Figure 4D–4J). Quantification of eIF4E immunoreactivity revealed that T-BSA treatment upregulated the p-eIF4E by ZIP9/Gnα11 in HT22 cells.

Erk1/2 activation is required for T-BSA-induced eIF4E phosphorylation and PSD95 expression

To assess if Erk1/2 signaling is involved in T-BSA-induced PSD95 expression, we used the specific Erk1/2 inhibitor SCH772984. As expected, western blotting analysis showed that pretreatment with SCH772984 blocked Erk1/2 phosphorylation in T-BSA-treated HT22 cells (Figure 5A and 5B). We thus further determined the effect of SCH772984 on eIF4E phosphorylation and PSD95 expression induced by T-BSA. Western blotting and immunofluorescent staining analysis demonstrated that HT22 cells treated with SCH772984 plus T-BSA exhibited significantly lower eIF4E phosphorylation compared to that in cells treated with T-BSA alone (Figure 5A and 5C, 5E and 5F).

Figure 5. Erk1/2 activation is required in T-BSA-induced eIF4E phosphorylation and PSD95 expression. (A–D) Western blotting for the phosphorylated /total levels of Erk1/2 and eIF4E and the level of synaptic protein PSD95 induced by T-BSA in HT22 cells pre-treated with the Erk1/2 inhibitor SCH772984 (S, 100 nM) for 2 h (n=5). (E–H) Immunofluorescence staining for p-eIF4E and PSD95 induced by T-BSA pre-treated with S (n=4, scale bar = 20 μm). (n.s.: non-significant; * P < 0.05; ** P < 0.01).

Western blotting and immunofluorescent staining analysis demonstrated that HT22 cells treated with SCH772984 plus T-BSA exhibited significantly lower PSD95 protein level compared with that in cells treated the T-BSA alone (Figure 5A and 5D, 5G and 5H). These results revealed that eIF4E phosphorylation and PSD95 expression induced by T-BSA depend on Erk1/2 activation.

Cell culture

Immortalized mouse hippocampal HT22 cells were maintained in Dulbecco’s modified Eagle medium (DMEM) with 10% fetal bovine serum (FBS) and a 0.2% penicillin/streptomycin solution in an H₂O-saturated 5% CO₂ atmosphere at 37°C. The cells were digested with 0.25% trypsin at 37°C from 10 cm culture dishes at 70–80% confluence. The cells were then seeded at 50,000 cells/cm² and cultured for another day in 6-well plates for protein/mRNA assays or were mounted on coverslips in 12-well plates for immunofluorescence staining.

Reagents

T-BSA (cat#: T3392) and fluorescein isothiocyanate-conjugated T-BSA(T-BSA-FITC) (cat#: T5771) were purchased from Sigma-Aldrich (MO, USA). BSA-FITC (cat#: SF063) was purchased from Solarbio (China). Actinomycin D (Act D; cat#: HY-17559) and Cycloheximide (CHX; cat#: HY-12320) were purchased from MedChemExpress (NJ, USA). Flutamide (F; cat#: F0663) and Dimethyl sulfoxide (DMSO; cat#: D0798) were purchased from Tokyo Chemical Industry Co., Ltd (Japan). SCH772984 (cat#: S7101) was purchased from Selleck Chemicals (TX, USA). Rabbit anti-PSD95 (cat#: ab18258), rabbit anti-phospho-eIF4E (S209) (cat#: 76256), rabbit anti-eIF4E (total) (cat#: 33766), rabbit anti-GAPDH (cat#: ab9485) antibodies were purchased from Abcam (MA, USA). Rabbit anti-phospho-Erk1/2 (Thr202/Tyr204) (cat#: 9101) and mouse anti-Erk1/2 (total) (cat#: 9107) antibodies were purchased from cell signaling technology (MA, USA). Rabbit anti-ZIP9 antibody (cat#: GTX31817) was purchased from GeneTex (CA, USA). Mouse anti-Gnα11 (cat#: sc-390382) and normal mouse IgG (cat#: sc-2025) antibodies were purchased from Santa Cruz Biotechnology (MA, USA).

Western blotting

Protein samples were isolated in 200 μl of RIPA buffer supplemented with protease and phosphatase inhibitors (Complete Protease Inhibitor Cocktail, cat#:04693116001 and PhosSTOP, cat#:04906845001, Roche, Switzerland). The homogenates were cleared by centrifugation at 12,000 g for 20 min at 4 °C. Total protein concentration was evaluated using BCA Protein Assay reagent kit (cat#: PC0020, Solarbio, China) and an Infinite F200 (TECAN, Switzerland) plate reader. Protein samples were mixed with 5X Laemmli buffer. After boiling, 25 μg of protein was loaded onto a 10% sodium dodecyl sulphate-polyacrylamide gel for protein separation and electro-transferred onto polyvinylidene fluoride membranes. The membranes were then blocked with 5% milk in TBS-T (0.1% Tween-20) and incubated with the following diluted primary antibodies overnight at 4 °C (rabbit anti-PSD95, 1:1,000; rabbit anti-phospho-Erk1/2 (Thr202/Tyr204), 1:2,000; rabbit anti-ZIP9, 1:1,000; rabbit anti-phospho-eIF4E (S209), 1:1,000; rabbit anti-eIF4E (total), 1:1,000; rabbit anti-GAPDH, 1:2,000; mouse anti-Erk1/2 (total), 1:1,000; mouse anti-Gnα11, 1:1,000). The membranes were then washed five times in TBST for 5 min each, and blotted with the secondary antibody (rabbit IgG (H&L) Antibody Dylight™ 800 Conjugated, cat#: 610-445-002 and mouse IgG (H&L) Antibody Dylight™ 680 Conjugated, cat#: 610-144-002, 1:10,000, ROCKLAND, USA) for 2 h at room temperature. After washing again, reactions were visualized and analyzed using an Odyssey IR fluorescence scanning imaging system (LI-COR, USA).

Cell-surface labelling with T-BSA-FITC

To detect the presence of membrane androgen-binding sites, HT22 cells were incubated with10⁻⁵ M T-BSA-FITC for 20 min at 37 °C. The medium was then aspirated, and the cells were fixed with 1 ml ice-cold 4% paraformaldehyde (PFA) for 10 min. After washing five times for 5 min each in 0.01 M phosphate-buffered saline (PBS), the coverslips were mounted with Vectashield-DAPI, and then gently placed over the slide. BSA-FITC was used as the control for detecting non-specific FITC signals. Images were obtained by an inverse Olympus FV1200 microscope equipped with the corresponding fluorescence system (Olympus, Hamburg, Germany) and were analyzed using Image Pro Plus 6.0 to quantify the fluorescence intensity.

Immunofluorescence staining

The cells were fixed with 1 ml ice-cold 4% PFA for 10 min at room temperature. After washing with 0.01 M PBS, the cultured cells on coverslips were blocked with PBS-T (0.5% Triton X-100) containing 5% donkey serum for 60 min at room temperature, followed by incubation with primary antibodies diluted in 0.01 M PBS overnight at 4 °C (rabbit anti-PSD95, 1:200; rabbit anti-phospho-Erk1/2 (Thr202/Tyr204), 1:200; rabbit anti-phospho-eIF4E (S209), 1:200). After washing, the coverslips were incubated with secondary antibodies (Donkey anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary antibody Alexa Fluor® 594, 1:1,000, cat#: A-21207, Thermo Scientific™) for 2 h at room temperature and mounted with Vectashield-DAPI after a final wash. Images were acquired with an inverse Olympus FV1200 microscope equipped with the corresponding fluorescence system and were analyzed using Image Pro Plus 6.0 to determine the optical density for quantitative analysis.

RNA isolation and reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA from the cells was extracted using the SV Total RNA Isolation System (cat#: Z3100, Promega, Germany), which was used to synthesize cDNA with the GoScript™ Reverse Transcription System (cat#: A5001, Promega, Germany). Total RNA (2 μg) was used in reverse transcription reactions at a total volume of 25 μl with the following three-step incubation: 25 °C for 5 min, 42 °C for 60 min, and 75 °C for 15 min. Quantification of mRNA levels was performed using quantitative PCR with 2×Taq PCR MasterMix (cat#: KT201, TIANGEN, China). Samples were incubated at 94 °C for 3 min, followed by 40 cycles of denaturation at 94 °C for 30 s, annealing at 57 °C for 30 s, and cDNA extension at 72 °C for 45 s. After the amplification cycles, a final extension step at 72 °C was performed for 10 min. The results for the target gene were normalized to the level of GAPDH for quantifying relative mRNA expression levels. The genes were amplified using the following primers: PSD95 forward, 5′TACCAAAGACCGTGCCAACG3′, and reverse 5′CGGCATTGGCTGAGACATCA3′; GAPDH forward, 5′CCGGTGCTGAGTATGTCGTG 3′, and reverse 5′TGGTCATGAGCCCTTCCACA3′.

Knockdown of ZIP9 and Gnα11 with RNA interference

Short hairpin RNA (shRNA) targeting ZIP9 (5′-ATTGTGTTCGTGGCAATAA-3′) and Gnα11 (5′-ACTCACACTTGGTCGATTA-3′) were packaged into lentiviruses (Genechem Inc.) for transfection in the cells at 30% confluence. The cells were maintained in 0.5 ml DMEM containing 2% FBS with ZIP9-shRNA, Gnα11-shRNA, or the negative control LVCON077 (hU6-MCS-Ubiquitin-EGFP-IRES-puromycin) for 12 h, respectively. After replacing the medium with 1 ml DMEM containing 2% FBS, the cells were cultured for another 72 h. The knockdown efficiency was assessed by western blotting as described above.

Co-immunoprecipitation (Co-IP)

After removing the culture medium, cells were washed once with ice-cold 0.01 M PBS and IP Lysis/Wash Buffer was added, followed by incubation on ice for 5 min. The homogenates were cleared by centrifugation at 12,000 g for 20 min at 4 °C. The total protein concentration was evaluated using BCA Protein Assay reagent kit and an Infinite F200 plate reader. The mouse anti-Gnα11 antibody or a control mouse IgG was used for immunoprecipitation (IP) and incubated overnight at 4°C to form the immune complex. The Pierce Protein A/G Magnetic Beads of Pierce™ Classic Magnetic IP/Co-IP Kit (cat#: 88804, Thermo Scientific™) were washed twice with ice-cold IP Lysis/Wash Buffer. The antigen sample/antibody mixture was then added to the tube containing the pre-washed magnetic beads and incubated at room temperature for 1 h with mixing. After washing the beads three times, the eluted proteins were subjected to SDS-PAGE and western blotting with rabbit anti-ZIP9 antibody (1:1,000) and mouse anti-Gnα11 antibody (1:1,000).

Proximity ligation assay (PLA®)

The cells were fixed with 1 ml ice-cold 4% PFA for 10 min. After washing, the cultured cells on coverslips were permeabilized with PBST and then one drop of Duolink II blocking solution of the Duolink^® PLA^® Kit (Sigma-Aldrich, cat#: DUO92008, Germany) was added to each well and incubated for a further 1 h at 37 °C. Subsequently, 20 μl of rabbit anti-ZIP9 antibody (1:200) and mouse anti-Gnα11 antibody (1:200) were added to each coverslip, and the cells were incubated overnight at 4 °C and washed. A total of 40 μl minus and plus PLA probes of the kit were then added to each coverslip and incubation proceeded for 1 h at 37 °C. After washing, 40 μl of ligation-ligase solution was added to each coverslip and incubation continued for 30 min at 37 °C, followed by the addition of 40 μl amplification-polymerase solution and incubation for 100 min at 37 °C in the dark. The coverslips were mounted with Vectashield-DAPI, and then gently placed over the slide. Images were obtained by an inverse Olympus FV1200 microscope and analysed using Image J to determine the number of ligation reactions per DAPI-stained cell.

Flowmetric Luminex xMAP® analysis

Protein samples were isolated in 200 μl of RIPA buffer supplemented with protease and phosphatase inhibitors. The homogenates were cleared by centrifugation at 12,000 g for 20 min at 4 °C, and the total protein concentration was evaluated using BCA Protein Assay reagent kit and an Infinite F200 plate reader. Aliquots of the supernatant were collected and stored at −80 °C until analysis using. Intracellular Bead-Based Multiplex Assays (cat#: 48-619MAG, 2-Plex Phospho/Total Erk Magnetic bead kit; cat#: 48-624MAG, 2-Plex Phospho/Total p38 Magnetic Bead Kit; cat#: 48-622MAG, 2-Plex Phospho/Total JNK Magnetic Bead Kit; cat#: 46-713MAG, Total β-Tubulin Magnetic Bead MAPmate™, Merck Millipore, USA) with Luminex technology. This method enables the simultaneous relative quantitation of multiple phosphorylated or total proteins in cell lysate samples using the Milliplex Multi-Analyte Profiling system.

Statistical analysis

All data were analysed with SPSS 21.0 statistical software (SPSS Inc., Chicago, IL, USA). Direct comparisons between two groups were performed using the two-tailed Student t test. Multiple group comparisons were made with one-way analysis of variance with post-hoc Fisher’s LSD test. The results are expressed as the mean ± SEM. Statistical significance was defined as *P < 0.05 or **P < 0.01, respectively.