Heart-type fatty acid binding protein regulates dopamine D2 receptor function in mouse brain

Abstract

Fatty acid binding proteins (FABPs) are essential for energy production and long-chain polyunsaturated fatty acid-related signaling in the brain and other tissues. Of various FABPs, heart-type fatty acid binding protein (H-FABP, FABP3) is highly expressed in neurons of mature brain and plays a role in arachidonic acid incorporation into brain and heart cells. However, the precise function of H-FABP in brain remains unclear. We previously demonstrated that H-FABP is associated with the dopamine D(2) receptor long isoform (D2LR) in vitro. Here, we confirm that H-FABP binds to dopamine D(2) receptor (D2R) in brain extracts and colocalizes immunohistochemically with D2R in the dorsal striatum. We show that H-FABP is highly expressed in acetylcholinergic interneurons and terminals of glutamatergic neurons in the dorsal striatum of mouse brain but absent in dopamine neuron terminals and spines in the same region. H-FABP knock-out (KO) mice showed lower responsiveness to methamphetamine-induced sensitization and enhanced haloperidol-induced catalepsy compared with wild-type mice, indicative of D2R dysfunction. Consistent with the latter, aberrant increased acetylcholine (ACh) release and depolarization-induced glutamate (Glu) release were observed in the dorsal striatum of H-FABP KO mice. Furthermore, phosphorylation of CaMKII (Ca(2+)/calmodulin-dependent protein kinase II) and ERK (extracellular signal-regulated kinase) was significantly increased in the dorsal striatum. We confirmed elevated ERK phosphorylation following quinpirole-mediated D2R stimulation in H-FABP-overexpressing SHSY-5Y human neuroblastoma cells. Together, H-FABP is highly expressed in ACh interneurons and glutamatergic terminals, thereby regulating dopamine D2R function in the striatum.

Figure 1.

H-FABP forms a complex and colocalizes with D2R in the dorsal striatum. A, B, Coimmunoprecipitation of H-FABP and D2R in brain extracts. Extracts were obtained from DS of wild-type (L1; lane 1) and H-FABP KO (L2; lane 2) mice. Extracts were immunoprecipitated (IP) with anti-D2R (A) or anti-H-FABP (B) antibody, and immunoprecipitates were then immunoblotted (IB) with anti-H-FABP (A) or with anti-D2R (B) antibody. As control H-FABP and D2R immunoreactive bands, cell extracts (Input) are shown from DS (L3; lane 3), PFC (L4; lane 4) or NG108-15 cells cotransfected with H-FABP and D2LR (L5; lane 5) constructs. C, Squares in schematic drawing are after Paxinos and Franklin (2001) and indicate dorsal striatum brain regions. D, Confocal images showing colocalization of H-FABP (green) and D2R (red) in the dorsal striatum. H-FABP colocalized with D2R in large aspiny cells likely to be cholinergic neurons. Scale bars, 20 μm. WT, wild-type mice; KO, H-FABP KO mice.

Figure 2.

H-FABP localization in the dorsal striatum. Confocal images showing colocalization of H-FABP (green) and markers of three classical neurotransmitters (acetylcholine, glutamate, and dopamine) or spinophilin (red) in the dorsal striatum. A, Immunoreactivities of H-FABP and VAChT almost completely merge. B, Most H-FABP-containing boutons show VGLUT1 immunoreactivity. C, D, Most H-FABP-positive structures do not show immunoreactivity for either TH or spinophilin. At right in B–D are high-magnification images. Areas circled with dashed lines are cell soma. Scale bars: A, 30 μm; B–D, 10 μm.

Figure 3.

Behavioral responses to METH, haloperidol, and SCH23390 in H-FABP KO mice. A, B, Locomotor activity of H-FABP KO mice was significantly reduced following repeated administration of 0.25 mg/kg METH (A) and 0.5 mg/kg METH (B) compared with wild-type mice. Each bar of points represents the mean ± SEM. ***p < 0.001, versus saline-treated (day 0) wild-type mice; ^## p < 0.01 and ^### p < 0.001, versus saline-treated (day 0) H-FABP KO mice. C, H-FABP KO mice exhibited strong haloperidol-induced catalepsy compared with wild-type mice. Each bar of points represents the mean ± SEM. **p < 0.01, versus vehicle-treated (Veh.) wild-type mice; # p < 0.05, ## p < 0.01, versus vehicle-treated (Veh.) H-FABP KO mice. ^§ p < 0.05, ^§§ p < 0.01 in H-FABP KO versus wild-type mice at the same dosage. D, SCH23390-induced catalepsy did not differ significantly between wild-type and H-FABP KO mice. Each bar of points represents the mean ± SEM. *p < 0.05 and **p < 0.01 versus vehicle-treated (Veh.) wild-type mice; # p < 0.05 and ## p < 0.01 versus vehicle-treated (Veh.) H-FABP KO mice. WT, wild-type mice; KO, H-FABP KO mice.

Figure 4.

Extracellular ACh and Glu concentrations in the dorsal striatum seen in freely moving animals. A, Extracellular ACh changes following intraperitoneal haloperidol (HAL) administration. H-FABP KO mice showed markedly increased haloperidol-induced ACh release compared with wild-type mice at a dosage of 0.5 mg/kg. ^§ p < 0.05, in H-FABP KO mice versus wild-type mice at the same dose. *p < 0.05, versus basal levels in wild-type mice; ^# p < 0.05 and ^## p < 0.01, versus basal levels in H-FABP KO mice. B, Extracellular Glu changes following intraperitoneal haloperidol administration. C, KCl depolarization-induced ACh release in the dorsal striatum. Each bar of points represents the mean ± SEM. ^§ p < 0.05, H-FABP KO mice versus wild-type mice at the same dosage. **p < 0.01, versus basal levels in wild-type mice; ^## p < 0.01, versus basal levels in H-FABP KO mice. D, KCl depolarization-induced Glu release in the dorsal striatum. Each bar of points represents the mean ± SEM. ^§ p < 0.05, in H-FABP KO versus wild-type mice at the same dosage. *p < 0.05 and **p < 0.01, versus basal levels in wild-type mice; ^# p < 0.05 and ^## p < 0.01, versus basal levels in H-FABP KO mice. WT, wild-type mice; KO, H-FABP KO mice.

Figure 5.

CaMKII and ERK phosphorylation is significantly increased in H-FABP KO mice. A, B, Top, representative immunoblots probed with various antibodies are shown. Bottom, Quantitative analyses by densitometry are shown. Each bar represents the mean ± SEM. *p < 0.05 and **p < 0.01 versus wild-type mice. C, Confocal microscopy images of double staining in dorsal striatum for phospho- (p)ERK (green) and PI (red), as well as merged images. Scale bar, 40 μm. WT, Wild-type mice; KO, H-FABP KO mice.

Figure 6.

Effect of quinpirole on H-FABP-overexpressing SHSY-5Y human neuroblastoma cells. A, Schematic diagrams of H-FABP sense (top) and antisense (bottom) orientation plasmids in pcDNA3. H-FABP cDNA is the open reading frame of Mus musculus fatty acid binding protein 3 (NM_010174). B, Representative immunoblots of cell extract proteins from treated (Quin.) and control (D.W.) cells probed with anti-pERK, anti-D2LR, H-FABP and anti-ERK (conventional) antibodies. C, Quantitative analysis of ERK activity was performed by measuring phosphorylated ERK immunoreactivity. After treatment of cells with 10 μm quinpirole or distilled water (D.W.) for 10 min in standard medium, cell extracts (20 μg) were prepared and subjected to immunoblot analysis. Each bar represents the mean ± SEM. **p < 0.01 versus distilled water-treated group; ^## p < 0.01 versus antisense with quinpirole-treated group.

Figure 7.

Schematic representation of altered neurotransmission in the dorsal striatum following H-FABP deletion. Left, The striatal microcircuit is composed of medium spiny neurons (MSNs) that receive input from excitatory corticostriatal glutamatergic projections, dopaminergic nigrostriatal fibers, and cholinergic terminals. D2R is present at the postsynapse (spine) of MSN and in each terminal. However, H-FABP is present only in glutamatergic and cholinergic terminals. H-FABP binds to the third cytoplasmic loop of D2R. Right, Following H-FABP deletion, D2R inhibitory effects are disrupted, resulting in increased Glu and ACh release from respective glutamatergic and cholinergic terminals and CaMKII and ERK activation at MSN postsynapses.