Transgenic mice expressing a truncated form of CREB-binding protein (CBP) exhibit deficits in hippocampal synaptic plasticity and memory storage

Abstract

Deletions, translocations, or point mutations in the CREB-binding protein (CBP) gene have been associated with Rubinstein-Taybi Syndrome; a human developmental disorder characterized by retarded growth and reduced mental function. To examine the role of CBP in memory, transgenic mice were generated in which the CaMKII alpha promoter drives expression of an inhibitory truncated CBP protein in forebrain neurons. Examination of hippocampal long-term potentiation (LTP), a form of synaptic plasticity thought to underlie memory storage, revealed significantly reduced late-phase LTP induced by dopamine-regulated potentiation in hippocampal slices from CBP transgenic mice. However, four-train induced late-phase LTP is normal. Behaviorally, CBP transgenic mice exhibited memory deficits in spatial learning in the Morris water maze and deficits in long-term memory for contextual fear conditioning, two hippocampus-dependent tasks. Together, these results demonstrate that CBP is involved in specific forms of hippocampal synaptic plasticity and hippocampus-dependent long-term memory formation.

Figure 1.

Generation of CaMKIIα-CBPΔ1 transgenic animals. (A) A mouse CBP cDNA truncation mutant, including amino acids 1-1084, was FLAG-epitope tagged at the amino terminus and cloned into a vector containing intron and exon sequences with splice sites and the SV40 polyadenylation signal. This entire sequence was then cloned downstream of the 8.5 kb mouse CaMKIIα promoter. This construct was then used to generate CaMKIIα-CBPΔ1 transgenic mice via pronuclear injection. (B) Coronal sections from the brains of CBPΔ1 transgenic mice (B1) and a wild-type littermate (B2) were hybridized with a probe specific to the CBPΔ1 transgene. Expression of the transgene is observed in the hippocampus, cortex, striatum, and amygdala. (C) Coronal sections from a CBPΔ1 transgenic mouse (C1) and a wild-type littermate (C2) were Nissl stained to confirm integrity of the hippocampus and other structures. (D) Total RNA was isolated from the hippocampus, cortex, and cerebellum of CaMKIIα-CBPΔ1 transgenic mice and wild-type littermates and analyzed by Northern blot. The membrane was first hybridized to an aminoterminal CBP radiolabeled probe, stripped, and then subject to a β-actin radiolabeled probe. The ethidium bromide-stained RNA gel is also shown, revealing 18S and 28S rRNAs.

Figure 2.

CBPΔ1 inhibits CRE-mediated transcription in tissue culture. The truncated CBP mutant (CBPΔ1) inhibits CRE-mediated transcription by ∼75%. Normalized CRE-luciferase levels reached 53.1 ± 6.8 relative light units (RLUs), but only 11.9 ± 0.9 RLUs in the presence of the truncated CBP mutant, showing that this CBP mutant affects CRE-dependent transcription. ^* indicates p < 0.0001.

Figure 3.

Baseline electrophysiological properties are normal in CBPΔ1 transgenic mice. (A-C) Input-output characteristics were examined by recording field potentials (fEPSPs) in area CA1 resulting from stimuli of increasing intensity delivered to the Schaeffer collaterals in 400 μm hippocampal slices from wild type (A) and CBPΔ1 transgenic mice (B). Plots of fEPSP initial slopes versus the corresponding presynaptic fiber volley amplitudes (C) are similar in slices from wild-type littermates (n = 4 slices from three mice) and CBPΔ1 transgenic mice (n = 7 slices from four mice). (D) Maximum fEPSP slopes are similar in CBPΔ1 transgenic mice (n = 32 slices from 15 mice) and wild-type littermates (n = 26 slices from 16 mice). Sample traces of paired pulse facilitation at 25, 50, 100, 200, and 300 msec are shown in slices from wild type (E) and CBPΔ1 transgenic mice (F). (G) Paired-pulse facilitation is not significantly different between wild type (n = 15 slices from eight mice) and CBPΔ1 transgenic mice (n = 21 slices from nine mice) at interstimulus intervals between 25 and 300 msec. (H) Post-tetanic potentiation is not altered in CBPΔ1 transgenic mice (n = 11 slices from eight transgenic mice and n = 10 slices from eight wild-type mice).

Figure 4.

Long-term potentiation deficits in CBPΔ1 transgenic mice. (A) Long-term potentiation (LTP) at Schaffer collateral synapses in response to four 1 sec, 100 Hz trains administered 5 min apart is normal in CBPΔ1 transgenic mice. Both CBPΔ1 transgenic mice (n = 7 slices from four mice) and wild-type littermates (n = 6 slices from five mice) show stable potentiation. Insets show superimposed sample sweeps from before and 60 min after tetanic stimulation for CBPΔ1 transgenic mice and wild-type littermates. (B) At 60 min following tetanic stimulation, wild-type mice showed potentiation to 165 ± 5% of control, and CBPΔ1 transgenic mice showed potentiation to 161 ± 11% of control. (C) LTP in response to a single 1 sec, 100 Hz train in the presence of a D1 dopaminergic agonist (10 μM chloro-APB hydrobromide) is impaired in CBPΔ1 transgenic mice. Both CBPΔ1 transgenic mice (n = 7 slices from seven mice) and wild-type littermates (n = 5 slices from five mice) show stable potentiation, but CBPΔ1 transgenic hippocampal slices show significantly reduced LTP. (D) At 60 min following tetanic stimulation, wild-type mice showed potentiation to 216 ± 13% of control, whereas CBPΔ1 transgenic mice showed significantly decreased potentiation to 176 ± 8% of control. (E) Long-term potentiation in response to a single 1 sec, 100 Hz train in control saline is not significantly different between slices from CBPΔ1 transgenic mice (n = 5 slices from four mice) and wild-type littermates (n = 4 slices from three mice). (F) At 60 min following tetanic stimulation by a single 1 sec, 100 Hz train in control saline, wild-type mice (133 ± 5% of control) show equivalent potentiation to CBPΔ1 transgenic mice (135 ± 9% of control). ^* indicates p < 0.05. Scale bars: 3 msec, 1 mV.

Figure 5.

CBPΔ1 trangenic mice exhibit impaired spatial learning. (A) Mean escape latency during acquisition of the hidden-platform version of the Morris water maze. CBPΔ1 transgenic mice (n = 9) have a significant overall impairment in acquisition as compared to wild-type littermates (n = 12). (B) Mean percent time spent in each quadrant during the post-training probe trial on day ten. CBPΔ1 transgenic mice (n = 9) spent significantly less time in the target quadrant, where the platform was located during training, than did wild-type littermates (n = 12). (C) Mean escape latency during acquisition of the visible-platform version of the Morris water maze. A different group of mice were tested in the visible-platform experiment than those tested in the hidden-platform experiment. There was no difference observed between CBPΔ1 transgenic mice (n = 8) and wild-type littermates (n = 5) in the visible platform version of the Morris water maze. ^* indicates p < 0.05.

Figure 6.

CBPΔ1 transgenic mice exhibit impaired long-term memory for contextual fear conditioning but normal cued fear conditioning. (A) Training and 24 h long-term memory test during contextual fear conditioning. There was no difference in freezing behavior between CBPΔ1 transgenic mice (n = 18) and wild-type littermates (n = 20) during the first 2.5 min before the shock is presented (preshock) or the 0.5 min after the shock (post-shock). However, CBPΔ1 transgenic mice exhibited a significant decrease in freezing in a 24 h retention test, performed in the same conditioned context, as compared to wild-type littermates. (B) Training and 1 h short-term memory test during contextual fear conditioning. CBPΔ1 transgenic mice (n = 11) and wild-type littermates (n = 11) showed no differences in freezing in a 1 h retention test, performed in the same conditioned context. (C) Training and 24 h long-term memory test during cued fear conditioning. No differences in freezing behavior were observed between CBPΔ1 transgenic mice (n = 14) and wild-type littermates (n = 8) during training or the 24 h retention test. The 24 h test was performed in a novel context for cued fear conditioning. ^* indicates p < 0.05.