Tyrosine phosphorylation of the 2B subunit of the NMDA receptor is necessary for taste memory formation

Abstract

We aimed to test whether tyrosine phosphorylation of the NMDA receptor (NMDAR) in the insular cortex is necessary for novel taste learning. We found that in rats, novel taste learning leads to elevated phosphorylation of tyrosine 1472 of the NR2B subunit of the NMDAR and increases the interaction of phosphorylated NR2B with the major postsynaptic scaffold protein PSD-95. Injection of the tyrosine kinase inhibitor genistein directly into the insular cortex of rats before novel taste exposure prevented the increase in NR2B tyrosine phosphorylation and behaviorally attenuated taste-memory formation. Functionally, tyrosine phosphorylation of NR2B after learning was found to determine the synaptic distribution of the NMDAR, since microinjection of genistein to the insular cortex altered the distribution pattern of NMDAR caused by novel taste learning.

Figure 1.

Novel taste learning induces phosphorylation of Y1472 NR2B and increases levels of NMDAR in the phosphotyrosine fractions of the insular cortex. A, Immunoprecipitation (IP) of tyrosine-phosphorylated fractions from the insular cortex after novel taste learning. Insular cortices of rats were removed bilaterally 1 h after ingestion of a familiar taste (water, black bars) or a novel taste [0.1% saccharin (sacc), gray bars]. Tyrosine-phosphorylated proteins were immunoprecipitated with phosphor-Tyr-specific antibodies (pY) as described in Materials and Methods. The phosphotyrosine content of a band corresponding to 180 kDa was determined from these experiments by immunoblot (IB) (left). The identity of this band was confirmed as NR2B by probing the same immunoprecipitates with anti-NR2B-specific antibody (right). B, Immunoprecipitation of total NR2B content from the insular cortex after novel taste learning. Total NR2B protein was immunoprecipitated from brains of water- and saccharin-receiving rats by means of anti-mouse NR2B antibodies, and was probed for phosphotyrosine content with rabbit polyclonal anti-phosphotyrosine (left) and rabbit anti-NR2B (right). C, NR1 and NR2 subunits of NMDA are elevated in the phosphotyrosine fractions of the insular cortex. Immunoprecipitates of tyrosine-phosphorylated proteins from animals treated with water or saccharin were probed with specific anti-NR1, anti-NR2A, or anti-NR2B antibodies. D, No change in total levels of NMDAR after novel taste learning. The different NMDAR subunits in total tissue lysates of insular cortex were probed 1 h after treatment. E, Novel taste exposure elevates tyrosine phosphorylation of the Fyn kinase target Y1472 NR2B. Phospho-Y1472-specific anti-NR2B antibody (left) and total anti-NR2B antibody were used to probe total tissue lysates of insular cortex of rats 1 h after exposure to water or saccharin. In each panel, a representative immunoblot is shown above a graph depicting mean ± SE for three or four independent experiments. *Saccharin versus water, p < 0.05.

Figure 2.

Novel taste learning increases the association of PSD-95 with phosphorylated NR2B. A, Novel taste learning increases the levels of p-NR2B-associated PSD-95. Insular cortices of rats were removed bilaterally 1 h after ingestion of a familiar taste (water, black bars) or saccharin (sacc; gray bars). The phosphotyrosine content (p-Y) of a band corresponding to 95 kDa was determined from these experiments (left). The identity of this band was confirmed as PSD-95 by probing the same immunoprecipitates with mouse anti-PSD-95-specific antibody (right). B, No change in total PSD-95 levels after exposure to novel taste. PSD-95 protein was immunoprecipitated from insular cortices of rats 1 h after ingestion of water or saccharin and probed with a general phosphotyrosine antibody (left) or total PSD-95 (right). C, Total NR2B content and phospho-NR2B levels were probed in the same PSD-95 immunoprecipitates. In each panel a representative immunoblot is shown above a graph depicting mean ± SE for three or four independent experiments. Other abbreviations are as defined in the legend to Figure 1. *Saccharin versus water, p < 0.05.

Figure 3.

Tyrosine phosphorylation of NR2B is necessary for novel taste learning. A, Genistein inhibits phosphorylation of NR2B in the insular cortex. A 1 μl aliquot of 30 μmol of genistein (Gen) or vehicle (Veh; 0.3% DMSO in saline) was injected bilaterally to the insular cortex of naive rats. The rats were killed 1 h later and tyrosine-phosphorylated proteins were immunoprecipitated with phospho-Tyr-specific antibodies. The tyrosine phosphorylation of the 180 kDa band (left) and total NR2B (right) levels were assayed by using specific antibodies, as described in Materials and Methods. A representative immunoblot is shown above the graph depicting mean ± SE for two independent experiments. *Saccharin versus water, p < 0.05. B, Genistein attenuates novel taste learning. Rats received a bilateral injection of 1 μl aliquot of 30 μmol of genistein or vehicle (0.3% DMSO in saline) 10 min before pre-exposure to the novel taste (saccharin or NaCl) or water (CS); 2 d later, both groups were exposed to 0.1% saccharin or 0.3% NaCl followed by intraperitoneal injection of 0.15 m LiCl, as the US. A third group underwent the procedure of CTA without pre-exposure. The animals were tested for their preference for the novel or the familiar taste 2 d later. Aversion to the novel taste was calculated as described in Materials and Methods. Results are mean ± SE for four independent experiments. *Gen or CTA versus Vehicle, p < 0.05. C, Genistein prevents the increase in NR2B phosphorylation after exposure to a new taste. Bilaterally cannulated animals were injected with either genistein or vehicle 10 min before a 20 min single exposure to either saccharin or water. The animals were killed 1 h after termination of drinking. Tissue was homogenized and immunoprecipitated with anti-phosphotyrosine antibody, and the levels of phosphorylated NR2B were assessed as before. D, Ifenprodil (Ifen) attenuates novel taste learning. Ifenprodil (5 μm) or vehicle was microinjected into the insular cortex 20 min before CTA training (control n = 6, ifenprodil n = 8). The animals were tested for their preference for the novel or the familiar taste 2 d later. Aversion to the novel taste was calculated as described in Materials and Methods. Other abbreviations are as defined in the legend to Figure 1. *Ifenprodil CTA versus Vehicle CTA, p < 0.05.

Figure 4.

Exposure to novel taste affects the localization of the NMDAR. A, Total levels of NR2B are reduced in synaptoneurosomes after novel taste exposure. Synaptoneurosomes were prepared from insular cortices of rats exposed to either water or 0.1% saccharin (sacc) for 1 h. Samples were probed with phospho-Y1472-specific anti-NR2B antibody (left) and total anti-NR2B antibody (middle), and the ratios of phosphorylated to total NR2B were calculated (right). *Saccharin versus water, p < 0.05. B, Total levels of NMDAR are reduced in synaptoneurosomes after exposure to saccharin. Levels of different NMDAR subunits in synaptoneurosomal fractions were assayed by using anti-NR1, anti-NR2A, and anti-NR2B antibodies, as described above. *Saccharin versus water, p < 0.05. C, No change in presynaptic markers after novel taste exposure. Synaptoneurosomal fractions were probed with monoclonal antibodies against the presynaptic proteins synapsin and syntaxin. D, Redistribution of NR2B and NR2A after exposure to novel taste is reversed by genistein. Bilaterally cannulated animals were injected with either vehicle (0.3% DMSO in saline) or genistein (1 μl, 30 μmol) 10 min before ingestion of fluids. Synaptoneurosomes were prepared from the insular cortices and analyzed for NR2B, NR2A, and NR1 content, as described above. *Saccharin versus water, p < 0.05.