Colocalization and regulated physical association of presynaptic serotonin transporters with A₃ adenosine receptors

Abstract

Activation of A₃ adenosine receptors (A₃ARs) rapidly enhances the activity of antidepressant-sensitive serotonin (5-HT) transporters (SERTs) in vitro, ex vivo, and in vivo. A₃AR agonist stimulation of SERT activity is lost in A₃AR knockout mice. A₃AR-stimulated SERT activity is mediated by protein kinase G1 (PKGI)- and p38 mitogen-activated protein kinase (MAPK)-linked pathways that support, respectively, enhanced SERT surface expression and catalytic activation. The mechanisms by which A₃ARs target SERTs among other potential effectors is unknown. Here we present evidence that A₃ARs are coexpressed with SERT in midbrain serotonergic neurons and form a physical complex in A₃AR/hSERT cotransfected cells. Treatment of A₃AR/SERT-cotransfected Chinese hamster ovary cells with the A₃AR agonist N⁶-(3-iodobenzyl)-N-methyl-5'-carbamoyladenosine (1 μM, 10 min), conditions previously reported to increase SERT surface expression and 5-HT uptake activity, enhanced the abundance of A₃AR/SERT complexes in a PKGI-dependent manner. Cotransfection of SERT with L90V-A₃AR, a hyperfunctional coding variant identified in subjects with autism spectrum disorder, resulted in a prolonged recovery of receptor/transporter complexes after A₃AR activation. Because PKGI and nitric-oxide synthetase are required for A₃AR stimulation of SERT activity, and proteins PKGI and NOS both form complexes with SERT, our findings suggest a mechanism by which signaling pathways coordinating A₃AR signaling to SERT can be spatially restricted and regulated, as well as compromised by neuropsychiatric disorders.

Fig. 1.

Colocalization of A₃AR and SERT in mouse midbrain serotonergic neuron. C56BL/7 mice were perfused and fixed for immunostaining of A₃AR and 5-HT (A–C), A₃AR and SERT protein (D–F), or A₃AR and GAD protein (G–I) in the medial aspects of the dorsal raphe nucleus. A, D, and G, A₃AR staining; B, 5-HT staining; E, SERT staining; C, F, and I, overlap of A₃AR and 5-HT (C) or SERT (F) and lack of costaining with GAD (I). Arrows in A to C identify examples of colocalization of A₃AR and 5-HT in cell bodies and axons. Arrows in D to E identify examples of colocalization of A₃AR and SERT in axons. Scale bar, 10 μM.

Fig. 2.

Physical association of SERT with A₃AR. A and B, CHO cells were transfected with vector (pcDNA) and either myc-A₃AR or HA-hSERT individually or contransfected with either myc-A₃AR or HA-hSERT. Total myc-A₃AR or coIP of myc-A₃AR/HA-SERT complexes was eluted from anti-HA beads and detected by Western blot (WB) analysis using anti-myc antibody. Lane 3, HA-SERT and myc- A3AR were individually transfected and detergent extracts were mixed before coIP; lane 5, sample was from direct cotransfection of HA-SERT and myc-A3AR. C and D, CHO cells were transfected with vector (pcDNA), myc-A₃AR, or HA-hSERT individually or in combination, and complexes were collected on anti-Myc beads. Blots were probed with anti-HA antibody. A and C, signals from coimmunoprecipitation; B and D, signals from total extracts of experiments shown in A and C.

Fig. 3.

Recovery of A₃AR-hSERT complexes is A₃AR-regulated and PKGI-dependent. A and B, CHO cells were cotransfected with myc-A₃AR and HA-SERT and treated with IB-MECA (1 μM) for 0, 10, or 40 min, followed by collection of complexes on anti-HA resin. Samples were blotted with anti myc antibody to detect A₃AR receptor. A, representative immunoblot. B, quantitation from multiple experiments in A (n = 4). C and D, CHO cells were cotransfected with myc-A₃AR and hSERT, and treated with the PKGI membrane-permeant peptide inhibitor DT-2 (0.1 μM) for 10 min followed by incubation with IB-MECA (1.0 μM) for an additional 10 min. Cells then were lysed with 1% Triton X-100 as detailed under Materials and Methods and collected on anti-myc beads. Western blotting was performed with anti-SERT antibody. C, representative immunoblot. D, quantification from multiple experiments in C (n = 3). *, p < 0.05; **, p < 0.01 versus 0 min or vehicle control (one-way ANOVA with Dunnett's multiple comparison test).

Fig. 4.

Elevated A₃AR-SERT recovery and SERT surface expression produced by IB-MECA are blocked by the specific A₃AR antagonist MRS1191 (MRS). A, CHO cells were cotransfected with myc-A₃AR and HA-SERT and treated with IB-MECA (1 μM) ± MRS1191 (1 μM) for 10 min, followed by coIP with anti-HA beads. Samples were blotted with anti-myc antibody. A, representative immunoblot. B, quantitation of multiple experiments from A (n = 4). C, CHO cells were cotransfected with myc-A₃AR and hSERT and treated with IB-MECA (1 μM) ± MRS1191 (1 μM) for 10 min, followed by biotinylation as described under Materials and Methods. Samples were blotted with antibody targeted to hSERT. C, representative experiment. D, quantitation of multiple experiments from C (n = 3). *, p < 0.05 versus vehicle control (one-way ANOVA with Dunnett's multiple comparison test).

Fig. 5.

Elevated SERT surface expression produced by IB-MECA requires PKGI activity. A and B, CHO cells were cotransfected with myc-A₃AR and hSERT and treated with IB-MECA (1 μM) for 10 and 40 min, followed by cell surface biotinylation and blotting with anti-SERT antibody. A, representative immunoblot. B, quantitation from multiple experiments from A (n = 4). C and D, CHO cells were cotransfected with myc-A₃AR and hSERT and treated with DT-2 (0.1 μM) for 10 min followed by incubation with IB-MECA (1.0 μM) for an additional 10 min. Cells were then biotinylated as described under Materials and Methods, lysed with 1% Triton X-100, and collected on Streptavidin beads as detailed under Materials and Methods. Western blots were performed with anti-SERT antibody. C, representative immunoblot. Veh, vehicle. D, quantification from multiple experiments from C (n = 4). *, p < 0.05 versus vehicle control (one-way ANOVA with Dunnett's multiple comparison test).

Fig. 6.

L90V A₃AR exhibits prolonged enhancement of recovery of SERT-A₃AR complexes and SERT surface expression after A₃AR agonist stimulation. A, wild-type or L90V Myc-A₃AR were cotransfected with HA-hSERT and treated with IB-MECA (1 μM) for 0, 10, or 40 min. Complexes were recovered on HA beads, and eluants were probed with anti-myc antibody. B, quantitation of multiple experiments from A (n = 3). *, p < 0.05 versus vehicle control; #, p < 0.05 versus L90V at 40 min (two-way ANOVA with Bonferroni post tests). C, L90V myc-A₃AR+HA-hSERT-transfected CHO cells were treated with MRS1191 (MRS, 1 μM) or DT2 (0.1 μM) for 10 min followed by a 40-min incubation with either vehicle or IB-MECA (1 μM). Blot was immunoprecipitated with HA beads and probed with anti-myc antibody. D, quantification of multiple Western blots from C (n = 4). *, p < 0.05 versus vehicle control (one-way ANOVA with with Dunnett's multiple comparison test). E, CHO cells transfected with wild type A₃AR or L90V A₃AR and hSERT were treated with vehicle or IB-MECA (1 μM) for 10 to 40 min then biotinylated on ice for 40 min. Western blots were performed using anti-SERT antibody. F, quantification of multiple experiments from E (n = 3). *, p < 0.05; **, p < 0.01 versus vehicle control; #, p < 0.05 versus L90V at 40 min (two-way ANOVA with Bonferroni post tests).