Long-lasting glucocorticoid desensitization of NE ARα1-induced increase in sEPSC frequency in CRH neurons via a dynamin-dependent mechanism.

A. Model of retrograde neuronal-glial signaling response triggered in CRH neurons by NE, based on Chen et al., 2019. B. Representative traces from our previous study of the sEPSC response in PVN CRH neurons to NE (100 µM) without pretreatment (Baseline/NE) (upper trace) and in the presence of 1 μM corticosterone (Baseline (CORT)/NE (CORT), 10 min) (lower trace). C. Summary graph of the stress (Restraint) and CORT (CORT) suppression of the NE-induced increase in sEPSC frequency and the rescue of the NE response by blocking endocytosis with the dynamin inhibitor dynasore (80 μM, 10 min) (Dyna+CORT+NE). D. Time histogram of the NE-induced increase in sEPSC frequency in CRH neurons over time in slices after 4 h or 18 h recovery from a 30-min acute restraint stress (4-h and 8-h Stress Recovery), compared to the NE response following no prior stress (Control NE, taken from 21). Complete desensitization of the NE response was seen after a recovery period of 4 h following restraint and the response was partially restored after 18 h of recovery. Panels B and C and the Control NE graph in D were modified from 21 with permission.

Cort causes a rapid increase in cytosolic ARα1b accumulation in a distinct subcellular “hot spot”.

A. In N42 cells transiently transfected with ARα1b-eGFP, ARα1b is mostly localized to the membrane at baseline, but redistributes from the membrane to internal ‘hotspots’ (arrows) after 30 min of NE application (NE) and further after 20 min of application of NE and CORT (NE+Cort). Membrane region in the dashed box is enlarged in the upper insets. B. Time course of increase in intracellular ARα1b-eGFP. Following plateau of the increase in the fluorescence signal by NE (Vehicle), the co-application of CORT caused a rapid, additional increase in cytosolic ARα1b that plateaued within 10 min (*, p<0.05, t-test vs NE). C. Relative increase in cytosolic intensity of ARα1b-eGFP compared to NE alone (NE) following the addition of CORT (NE+Cort). (**, p<0.01, paired t-test vs. NE).

Cort redirects ARα1b trafficking from the rapid recycling endosome to the late endosome.

A. N42 cells expressing ARα1b-eGFP (ARα1b) and Rab5-dsRed (Rab 5) and images of calculated FRET ratios (FRET) after treatments with NE and NE+Cort. Right: Mean (+/- SEM) of FRET signal in cells treated with NE (1 µM) and NE+Cort showing increased FRET between ARα1b and Rab5 induced by NE, but no further increase after subsequent Cort application (2 µM) (repeated measures ANOVA, Tukey’s post-hoc, * p < 0.05 vs baseline; n.s., p=0.7423). B. N42 cells expressing ARα1b-eGFP (ARα1b) and Rab7-dsRed (Rab 7) and images of calculated FRET ratios (FRET) after treatments with NE and NE+Cort. Right: Mean (+/- SEM) of FRET signal in cells treated with NE (1 µM) and NE+Cort, showing increased FRET between ARα1b and Rab7 with NE, and a further increase in FRET ratio with subsequent Cort application (2 µM) (repeated measures ANOVA, Tukey’s post-hoc, * p<0.05). C. Images of calculated FRET ratios of N42 cells expressing ARα1b-GFP and Rab4a-dsRed after NE and Cort treatments. Images in the right column show the areas in the squares at higher magnification. Right: Mean (+/- SEM) of FRET signal showing no significant change in ARα1b/Rab4a interaction after treatment with NE (1 µM) and a decrease in FRET ratio with subsequent Cort application (2 µM) (repeated measures ANOVA, * p<0.05 vs NE). D. Images of calculated FRET ratios of N42 cells expressing ARα1b-GFP and Rab11-dsRed after NE and NE+Cort treatments. Images in the right column show the areas in the squares at higher magnification Right: Mean (+/- SEM) of FRET signal in NE and NE+Cort, normalized to baseline. There were no significant changes to the ARα1b-Rab 11 FRET ratio after application of either NE (1 µM) or NE+Cort (2 µM) (repeated measures ANOVA, p=0.2895).

ARα1b trafficking to lysosomes.

A. ARα1b-eGFP and LAMP1-RFP were co-transfected into N42 cells, and the cells were treated with NE for 30 min, followed by NE + CORT for 20 min. Arrow: CORT-facilitated internalized ARα1b co-localized with LAMP1, a lysosomal marker. B. Time lapse quantification of co-localization by Pearson’s coefficient. C. Co-localization averages of baseline and the last 3 minutes of NE and NE +CORT treatment (Mean +/- SEM, ANOVA, Tukey’s post-hoc, * p<0.05, ** p<0.01).

Reversal of NE- and Cort-induced ARα1b trafficking.

A) ARα1b-EGFP-transfected N42 cells were treated with NE for 30 min, followed by the addition of Cort for 20 min and finally the addition of prazosin for 30 min. Arrows: ARα1b-EGFP hot spot. B) Normalized cytosolic fluorescence (ANOVA, Tukey’s post-hoc, * p<0.05 vs. baseline, # p<0.05 vs. NE). The ARα1b hot spot returned to near baseline after 30 min prazosin co-application (n.s., p=0.35 vs baseline).

CORT regulation of protein S-nitrosylation.

A. Nitrosylation dependence of the NE-induced increase in sEPSCs in PVN CRH neurons. The NE-induced increase in sEPSC frequency was blocked by the nitric oxide synthase (NOS) inhibitor L-NAME and the S-nitrosylation inhibitor NEM, but not by the soluble guanylyl cyclase inhibitor ODQ. B. Summary graph of the effects of L-NAME, NEM, and ODQ on the NE-induced increase in sEPSC frequency. C. Changes over time in the cytosolic intensity of ARα1b-eGFP in N42 cells after treatment with NE and L-NAME (NE+LNAME) and NE, L-NAME and CORT (NE+LNAME+Cort). D. Summary graph of the changes in ARα1b-EGFP fluorescence intensity indicative of internalization in NE+L-NAME and NE+L-NAME+Cort, normalized to the baseline fluorescence intensity (** p<0.01 compared to baseline). Cort and L-NAME co-application was not significantly different from L-NAME alone (p=0.2009, Students t-test). Cytosolic ARα1b measurements in L-NAME and L-NAME+Cort were taken at 40-45 min and compared to measurements in NE alone at 30-35 min. E. Western blot analysis of whole-cell lysate (cytosolic fraction) from N42 cells treated with vehicle or 2 µM Cort followed by TMT switch assay for total nitrosylated protein. Cort-treated cells showed a significant decrease in total nitrosylated protein [t(9) = 3.368, p = 0.0083, Student’s t-test, n=5-6]. F. Images of β-arrestin1/2 nitrosylation in N42 cells treated with NE, Cort, or NE+Cort for 20 min. The biotin-switch assay was performed, followed by proximity ligation assay (PLA) with antibodies to biotin and β-arrestin1/2. Maximum intensity projections of z-stacks. Red dots = PLA signal representing nitrosylated β-arrestin1/2; blue = DAPI. G. Nitrosylation of β-arrestin1/2 by NE, Cort, and NE+Cort relative to vehicle. There were no significant drug effects [One-Way ANOVA, F (3, 24) = 1.861; p = 0.1632 n = 6-8 fields from 4 separate assays]. H. None of the drug treatments altered the nuclear localization of nitrosylated β-arrestin 1/2 [One-Way ANOVA, F (3, 25) = 0.2101; p = 0.8885, n = 6-8 fields from 4 separate assays].

Interaction between ARα1b and GR.

A. Representative images of ARα1b-myc-DDK-transfected N42 cells after the different treatments and following PLA. Lower images are high-magnification of boxes in upper images. Red = PLA signal, blue = DAPI staining of nuclei, green = phalloidin-FITC staining of F-actin B. Summary graph of total numbers of ARα1b-GR interactions under the different drug conditions. There was an overall decrease in interactions with Cort treatment [F(3, 102) = 3.831 P=0.012]. C. Summary graph of the nuclear fraction of ARα1b-GR profiles under the different treatments. Cort treatment significantly increased nuclear localization of the receptor complex relative to vehicle and the other treatment groups [F (3, 103) = 17.30 P<0.0001]. n = 21-31 cells from 3 separate assays. Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons test. * p<0.05, ** p<0.0001.

Interaction between ARα1b and β-arrestin.

Proximity ligation assay with antibodies to tagged ARα1b and β-arrestin1/2. N42 cells were treated with drug for 20 min, followed by proximity ligation assay. A. Representative images. Blue = DAPI, red = PLA signal, green = phalloidin-FITC stain of F-actin. B. Treatments with NE, Cort, and NE+Cort. NE increased the number of ARa1b-b-arrestin interactions per cell; Cort alone had no effect, but blocked the NE-induced increase in ARα1b-β-arrestin interactions [F(3,99), p = 6.157, p = 0.0007], C. Cort treatment increased the nuclear translocation of the ARα1b-β-arrestin1/2 complex [F (3, 99) = 6.314, p = 0.0006], n = 23-30 cells from 3 separate assays. * p < 0.05. ** p<0.01, ***p<0.001. Data were analyzed by one-way ANOVA Followed by Tukey’s multiple comparisons test. Scale bar = 20 µm (main), 5 µm (inset).

Interaction between GR and β-arrestin.

Proximity ligation assay with antibodies to GR and β-arrestin1/2. A. Representative images of the PLA labeling with vehicle (Veh), NE, Cort, and NE + Cort. Lower images are high magnification of boxes in upper images. Blue = DAPI, red = PLA signal, green = phalloidin-FITC stain of F-actin. B. None of the treatments had an effect on the total number of GR-β-arrestin1/2 interactions per cell (one-way ANOVA, [F(3, 32) = 0.3233, p = 0.8085]. C. Cort treatment increased the nuclear fraction of the GR-β-arrestin1/2 complex (one-way ANOVA followed by Dunnett’s multiple comparisons test, [F(3, 33) = 4.058, p = 0.0147], n = 9-10 cells from 3 separate assays, * p < 0.05).