Proteolytic degradation of SCOP in the hippocampus contributes to activation of MAP kinase and memory

Abstract

Because activation of ERK1/2 MAP kinase (MAPK) is critical for hippocampus-dependent memory, there is considerable interest in mechanisms for regulation of MAPK during memory formation. Here we report that MAPK and CREB-mediated transcription are negatively regulated by SCOP (suprachiasmatic nucleus [SCN] circadian oscillatory protein) and that SCOP is proteolyzed by calpain when hippocampal neurons are stimulated by brain-derived neurotrophic factor (BDNF), KCl depolarization, or NMDA. Moreover, training for novel object memory decreases SCOP in the hippocampus. To determine if hippocampus-dependent memory is influenced by SCOP in vivo, we generated a transgenic mouse strain for the inducible overexpression of SCOP in the forebrain. Overexpression of SCOP completely blocked memory for novel objects. We conclude that degradation of SCOP by calpain contributes to activation of MAPK during memory formation.

Figure 1. SCOP is Expressed in the Hippocampus

SCOP immunoreactivity was detectable in several areas of the rat and mouse hippocampus including area CA1. Coronal sections of the hippocampus were stained with SCOP antibody and visualized using 3-3’ diaminobenzidine as described in Experimental Procedures. The diagrams on the right side identify those areas of the hippocampus shown in the images. Scale bars represent 100 μm.

Figure 2. CRE-Mediated Transcription in Hippocampal Neurons is Inhibited by SCOP

(A) SCOP overexpression inhibits BDNF stimulation of CRE-mediated transcription. Cultured hippocampal neurons were cotransfected with a CRE-luciferase reporter construct and varying amounts of a SCOP expression plasmid or empty vector (total 300 ng). An expression vector for β-galactosidase was cotransfected for normalization of transfection efficiencies. Luciferase activity was measured 6 hr after BDNF (5ng/ml) or vehicle treatment. Data are reported as the percent activity relative to the vector control. Error bars, SEM (n=3). (B) Inhibition of SCOP expression by siRNA stimulates CRE-mediated transcription. Cultured hippocampal neurons were cotransfected with a CRE-luciferase reporter construct with SCOP siRNA, or a scrambled RNA of the same composition. Luciferase activity was measured 24 hr after transfection and were normalized to β-galactosidase activity. Error bars, SEM (n=3).

Figure 3. BDNF Stimulates Rapid Degradation of SCOP in Hippocampal Neurons

(A) SCOP protein in cultured hippocampal neurons (Day 18 in vitro: DIV 18) was monitored by Western analysis various times after treatment with BDNF (50ng/ml). The bar graph shows SCOP protein normalized to ERK-2. Error bars, SEM (n= 3) (B) SCOP protein in cultured hippocampal neurons (DIV18) was monitored by Western analysis 10 min after treatment with KCl (50mM) or NMDA (50μM). SCOP protein was normalized relative to ERK-2. Error bars, SEM (n=3).

Figure 4. Calpain Catalyzes the Proteolysis of SCOP

(A) SCOP protein was monitored by Western analysis after adding Mg²⁺ (10 mM) and/or Ca²⁺ (4 mM) to mouse brain lysates and incubating for three hours at 4°C. (B) Calpeptin inhibited Ca²⁺-stimulated proteolysis of SCOP. The mouse brain extract was pretreated with a caspase-3 inhibitor (10 μM), lactacystin (5 μM) or calpeptin (10 μM) 30 min before adding Ca²⁺. SCOP protein was monitored by Western analysis. (C) Calpain I and II both catalyzed the proteolysis of SCOP in vitro. Calpain I or calpain II was incubated with immunoprecipitated SCOP at 37 °C for one hr. SCOP protein was monitored by Western analysis.

Figure 5. Inhibition of Calpain in Cultured Hippocampal Neurons Blocks SCOP Degradation and Inhibits Activation of MAPK

(A) Calpain inhibitor III inhibited BDNF-induced SCOP degradation in cultured hippocampal neurons. Neurons were pretreated with calpain inhibitor III (25μM) or vehicle 30 min before BDNF (5ng/ml) stimulation and harvested at various times. SCOP protein was monitored by Western analysis. SCOP protein was normalized to actin. Error bar, SEM (n=3). (B) Treatment of hippocampal neurons with calpain inhibitor III increases SCOP. Neurons were pretreated with calpain inhibitor III (25μM) or vehicle and harvested at various times after applying calpain inhibitor III (25μM) or vehicle. SCOP was quantitated by Western analysis. SCOP levels were normalized to ERK-2. (C) Calpain inhibitor III inhibits BDNF stimulation of MAPK activity measured as pERK in cultured hippocampal neurons. Total ERK-2 measured on the same Western blot was used as loading controls. The percentage change in pERK relative to the zero time point is reported. Error bars, SEM (n=3).

Figure 6. Training for Novel Object Memory Decreases SCOP in the Hippocampus

(A) The time spent exploring two novel objects (A and B) was measured for 5 min during training. Preference for one of the original objects (A) and a new object (C) was quantified 24 hr after training. The dotted line represents no object preference. Error bars, SEM (n=8 mice) (B) SCOP and pERK in the hippocampus from trained and untrained mice was analyzed by Western analysis 5 min after training. Untrained mice were handled and placed in the training chamber without novel objects. SCOP and p-ERK were normalized to ERK-2 from the Western blot. Error bars, SEM (n=9).

Figure 7. Overexpression of SCOP in the Hippocampus Blocks Long-Term but not Short-Term Memory for Novel Objects

(A) SCOP in the hippocampus of TRE-SCOP x CamKII rtTA2 transgenic mice was increased by feeding doxycycline (Dox). Upper panels show the SCOP Western blot for each mouse and the bar graphs reports SCOP normalized to actin. Lower panels show the corresponding PCR genotyping of the transgenic mice. (B) Overexpression of SCOP (Dox) in the forebrain did not block short-term memory for novel objects measured 8 min after training. Error bars, SEM ( no DOX, n=8 mice; DOX , n=13). (C) Overexpression of SCOP in the forebrain blocked memory for novel objects measured 24 hr after training. Error bars, SEM (no Dox, n=12 mice; Dox, n=22 mice. (D) Doxycyline-treated mice were allowed to recover for three months after removal of doxycyline, trained for novel objects, and tested 24 hr after training. Error bars, SEM (n=8 mice).

Figure 8. Inhibition of Calpain in vivo Blocks Memory for Novel Objects

A calpain Inhibitor cocktail containing a mixture of calpeptin, calpain inhibitor I and calpain inhibitor III or vehicle was bilaterally infused in area CA1 in the hippocampus. Mice were trained 45 min after infusion. Object preference was measured 24 hr after training (A) or 8 min after training (B). Error bars, SEM (n=10 (24 hr), n=8 (8 min) mice in each group). The dotted line represents performance by chance 50%.

Figure 9. Model for Role of SCOP in MAPK Activation

It is hypothesized that SCOP attenuates MAPK activity in hippocampal neurons by binding to the nucleotide-free form of K-Ras, thereby inhibiting stimulation of the MAPK pathway. Calapin may contribute to ERK activation by degrading SCOP.