Differential regulation of CaMKIIα interactions with mGluR5 and NMDA receptors by Ca(2+) in neurons

Abstract

Two glutamate receptors, metabotropic glutamate receptor 5 (mGluR5), and ionotropic NMDA receptors (NMDAR), functionally interact with each other to regulate excitatory synaptic transmission in the mammalian brain. In exploring molecular mechanisms underlying their interactions, we found that Ca(2+) /calmodulin-dependent protein kinase IIα (CaMKIIα) may play a central role. The synapse-enriched CaMKIIα directly binds to the proximal region of intracellular C terminal tails of mGluR5 in vitro. This binding is state-dependent: inactive CaMKIIα binds to mGluR5 at a high level whereas the active form of the kinase (following Ca(2+) /calmodulin binding and activation) loses its affinity for the receptor. Ca(2+) also promotes calmodulin to bind to mGluR5 at a region overlapping with the CaMKIIα-binding site, resulting in a competitive inhibition of CaMKIIα binding to mGluR5. In rat striatal neurons, inactive CaMKIIα constitutively binds to mGluR5. Activation of mGluR5 Ca(2+) -dependently dissociates CaMKIIα from the receptor and simultaneously promotes CaMKIIα to bind to the adjacent NMDAR GluN2B subunit, which enables CaMKIIα to phosphorylate GluN2B at a CaMKIIα-sensitive site. Together, the long intracellular C-terminal tail of mGluR5 seems to serve as a scaffolding domain to recruit and store CaMKIIα within synapses. The mGluR5-dependent Ca(2+) transients differentially regulate CaMKIIα interactions with mGluR5 and GluN2B in striatal neurons, which may contribute to cross-talk between the two receptors. We show that activation of mGluR5 with a selective agonist triggers intracellular Ca(2+) release in striatal neurons. Released Ca(2+) dissociates preformed CaMKIIα from mGluR5 and meanwhile promotes active CaMKIIα to bind to the adjacent NMDAR GluN2B subunit, which enables CaMKIIα to phosphorylate GluN2B at a CaMKIIα-sensitive site. This agonist-induced cascade seems to mediate crosstalk between mGluR5 and NMDA receptors in neurons.

Keywords: GluN2B; NR2B; calmodulin; mGluR; nucleus accumbens; striatum.

Figure 1. Interactions between CaMKIIα and mGluR5

A, GST-fusion proteins containing individual intracellular domains of rat mGluR5. B, Interactions of CaMKIIα with mGluR5. Interactions were detected in pull-down assays with immobilized GST-fusion proteins and rat striatal lysates. Note that mGluR5-CT1 selectively pulled down CaMKIIα since mGluR5-CT1 did not precipitate CaMKIV. C, Binding of CaMKIIα to mGluR5. Binding activities between GST or GST-fusion proteins and purified recombinant CaMKIIα were detected in in vitro binding assays. D, GST-fusion proteins containing different mGluR5-CT1 fragments. E, Binding of CaMKIIα to mGluR5-CT1 fragments as detected in in vitro binding assays with GST-fusion mGluR5-CT1 fragments and purified CaMKIIα. Note that CT1b, but not CT1a and CT1c, bound to CaMKIIα. F, CT1b sequences of mGluR5 from the rat, mouse, and human aligned with the corresponding CT region of rat mGluR1a. Amino acids were aligned using clustal W. Dark boxes indicate the regions of homology. A CaM binding site on rodent mGluR5 (K889-K917) is indicated. Pull-down and in vitro binding assays were conducted in the presence of EGTA (1 mM). Proteins bound to GST-fusion proteins in either pull-down or binding assays were visualized with immunoblots (IB) using the specific antibodies as indicated. The experiments were repeated independently for three times.

Figure 2. Ca²⁺/CaM-mediated regulation of CaMKIIα binding to mGluR5

A, CaMKIIα binding to mGluR5-CT1 as detected in in vitro binding assays in the presence or absence of Ca²⁺/CaM. Either CaCl₂ (0.5 mM)/CaM (1 μM) or EGTA (1 mM) without CaCl₂/CaM were added into binding solutions. B, T286-autophosphorylation reduced the binding of CaMKIIα to mGluR5-CT1. CaMKIIα was autophosphorylated prior to binding assays. The T286-autophosphorylated state of CaMKIIα was validated by immunoblots with a T286 phospho-specific antibody. C, A graph of the data from (B). D, CaMKIIα wild-type (WT) but not the T286D phosphomimetic mutant bound to mGluR5-CT1. Binding assays were performed in the absence of Ca²⁺/CaM. E, Binding of WT and T305/306D mutant CaMKIIα to mGluR5-CT1 in the absence or presence of Ca²⁺/CaM. F, Effects of the inhibitory peptide (L290-A309) on the CaMKIIα-mGluR5 binding. G, Effects of the L290-A309 peptide on T286 autophosphorylation of CaMKIIα. H, CaM bound to mGluR5-CT1 in the presence but not absence of Ca²⁺. I, Binding of the T305/306 mutant and CaM to mGluR5-CT1 in the absence or presence of Ca²⁺. Binding assays were performed between CaMKIIα, pCaMKIIα, CaMKIIα mutants, or CaM and immobilized GST or GST-mGluR5-CT1 in the absence or presence of CaCl₂ (0.5 mM), CaM (1 μM), or L290-A309 (100 μM) (F) as indicated. EGTA (1 mM) was added in the assays lacking CaCl₂. Bound CaMKIIα, pCaMKIIα, CaMKII mutant, or CaM proteins were visualized by immunoblots (IB). Data are presented as means ± SEM (n = 3 per group). *p < 0.05 versus CaMKIIα.

Figure 3. Interactions of endogenous CaMKIIα with mGluR5 in rat striatal neurons

A and B, Representative immunoblots showing CaMKIIα and mGluR5 interactions in striatal neurons as detected by coimmunoprecipitation (IP). No precipitating antibody, an irrelevant IgG or an mGluR5 antibody was used in lane 2 (L2), 3 (L3) or 4 (L4), respectively. No striatal proteins and antibodies were used in lane 1 (L1). C, Effects of Ca²⁺ on the association of CaMKIIα with mGluR5 in rat striatal lyses. D, Effects of ionomycin (10 min) on the association of CaMKIIα with mGluR5 in striatal slices. Note that CaMKIIα-mGluR5 interactions were concentration-dependently reduced after adding Ca²⁺ or ionomycin to striatal lyses or slices, respectively. Coimmunoprecipitation was performed with solubilized striatal synaptosomal proteins to detect interactions between CaMKIIα and mGluR5. Precipitated proteins were visualized by immunoblots (IB) with indicated antibodies. Data are presented as means ± SEM (n = 3 per group). *p < 0.05 versus EGTA or vehicle.

Figure 4. Effects of DHPG on interactions between CaMKIIα and mGluR5 in rat striatal neurons

A and B, Concentration-dependent effects of DHPG (10 min) on CaMKIIα-mGluR5 interactions (A) and T286 phosphorylation of CaMKIIα (pCaMKIIα-T286) (B) in striatal slices. Note that CaMKIIα-mGluR5 interactions were reduced, whereas T286 phosphorylation of CaMKIIα was elevated, by DHPG in a concentration-dependent manner. C and D, Time-dependent effects of DHPG on CaMKIIα-mGluR5 (C) and CaM-mGluR5 (D) interactions in striatal slices. DHPG was incubated in slices at 100 μM for different durations (10, 30, or 60 min) prior to sample collections. Note that DHPG time-dependently reduced and enhanced CaMKIIα-mGluR5 and CaM-mGluR5 interactions, respectively. E, Effects of the mGluR5 antagonist MTEP on the DHPG-induced reduction in CaMKIIα-mGluR5 interactions in striatal slices. MTEP (10 μM) was applied 30 min prior to and during DHPG incubation (100 μM, 10 min). Coimmunoprecipitation with solubilized striatal synaptosomal proteins was performed to detect interactions between CaMKIIα and mGluR5 (A, C, and E) or between CaM and mGluR5 (D). Precipitated proteins were visualized by immunoblots (IB) with indicated antibodies. Data are presented as means ± SEM (n = 3–6 per group). *p < 0.05 versus vehicle. +p < 0.05 versus vehicle + DHPG.

Figure 5. Effects of DHPG on CaMKIIα-GluN2B interactions in rat striatal neurons

A, Effects of Ca²⁺ on the association of CaMKIIα with GluN2B in rat striatal lyses. Ca²⁺ was added to lyses to a final concentration of 2.5 μM for 10 min. Note a robust increase in co-precipitating levels of CaMKIIα in Ca²⁺-treated samples. B, Concentration-dependent effects of DHPG (10 min) on CaMKIIα-GluN2B interactions in striatal slices. C, Time-dependent effects of DHPG on CaMKIIα-GluN2B interactions in striatal slices. DHPG was incubated in slices at 100 μM for different durations (10, 30, or 60 min) prior to sample collections. D, Effects of MTEP, KN93, or KN92 on the DHPG-stimulated CaMKIIα-GluN2B interactions in striatal slices. MTEP (10 μM), KN93 (20 μM), or KN92 (20 μM) was applied 30 min prior to and during DHPG incubation (100 μM, 10 min). Coimmunoprecipitation (IP) with solubilized striatal synaptosomal proteins was performed to detect interactions between CaMKIIα and GluN2B. Precipitated proteins were visualized by immunoblots (IB) with indicated antibodies. Data are presented as means ± SEM (n = 3–6 per group). *p < 0.05 versus vehicle. +p < 0.05 versus vehicle + DHPG.

Figure 6. Effects of DHPG on GluN2B phosphorylation in rat striatal slices

A, Concentration-dependent effects of DHPG (10 min) on GluN2B-S1303 phosphorylation. Note that DHPG concentration-dependently increased S1303 phosphorylation. B, Time-dependent effects of DHPG on GluN2B-S1303 phosphorylation. DHPG was incubated in striatal slices at 100 μM for different durations (10, 30, or 60 min). Note a dynamic and reversible change in S1303 phosphorylation following DHPG stimulation. C, Effects of MTEP, KN93, or KN92 on the DHPG-stimulated GluN2B-S1303 phosphorylation. MTEP (10 μM), KN93 (20 μM), or KN92 (20 μM) was applied 30 min prior to and during DHPG incubation (100 μM, 10 min). After drug treatment, slices were collected for preparing synaptosomal proteins for immunoblot analysis of pGluN2B-S1303 proteins with a phospho- and site-specific antibody. Representative immunoblots are shown left to quantification data. Data are presented as means ± SEM (n = 3–6 per group). *p < 0.05 versus vehicle. +p < 0.05 versus vehicle + DHPG.