Increasing levels of wild-type CREB up-regulates several activity-regulated inhibitor of death (AID) genes and promotes neuronal survival

Abstract

Background: CREB (cAMP-response element binding protein) is the prototypical signal-regulated transcription factor. In neurons, it is the target of the synaptic activity-induced nuclear calcium-calcium/calmodulin dependent protein kinase (CaMK) IV signaling pathway that controls the expression of genes important for acquired neuroprotection as well as other long-lasting adaptive processes in the nervous system. The function of CREB as a transcriptional activator is controlled by its phosphorylation on serine 133, which can be catalyzed by CaMKIV and leads to the recruitment of the co-activator, CREB binding protein (CBP). Activation of CBP function by nuclear calcium-CaMKIV signaling is a second regulatory step required for CREB/CBP-mediated transcription.

Results: Here we used recombinant adeno-associated virus (rAAV) to increase the levels of wild type CREB or to overexpress a mutant version of CREB (mCREB) containing a serine to alanine mutation at position amino acid 133 in mouse hippocampal neurons. Increasing the levels of CREB was sufficient to boost neuroprotective activity even under basal conditions (i.e., in the absence of stimulation of synaptic activity). In contrast, overexpression of mCREB increased cell death. The ratio of phospho(serine 133)CREB to CREB immunoreactivity in unstimulated hippocampal neurons was similar for endogenous CREB and overexpressed wild type CREB and, as expected, dramatically reduced for overexpressed mCREB. A gene expression analysis revealed that increased expression of CREB but not that of mCREB in hippocampal neurons led to elevated expression levels of bdnf as well as that of several members of a previously characterized set of Activity-regulated Inhibitor of Death (AID) genes, which include atf3, btg2, gadd45β, and gadd45γ.

Conclusions: Our findings indicate that the expression levels of wild type CREB are a critical determinant of the ability of hippocampal neurons to survive harmful conditions. Increasing the levels of wild type CREB can, even without inducing synaptic activity, increase pro-survival gene expression and strengthen the neurons' neuroprotective shield. The observed degradation of CREB protein following NMDA treatment of hippocampal neurons suggests that the known CREB shut-off associated with extrasynaptic NMDA receptor-induced excitotoxicity is followed by CREB proteolysis.

Figure 1

Immunocytochemical (A) and immunoblot analyses (B) of hippocampal neurons infected with rAAVs containing expression cassettes for hrGFP, CREB or mCREB. CREB and mCREB contain triple Flag epitope tags and were detected with M2 Flag antibody; hrGFP was detected with an antibody to hrGFP. Cell nuclei were counterstained with Hoechst 33258. Representative images are shown. Scale bar, 10 μm.

Figure 2

(A) Immunoblot analysis of expression and phosphorylation on serine 133 of endogenous and overexpressed CREB in uninfected hippocampal neurons and in neurons infected with rAAV-hrGFP, rAAV-CREB and rAAV-mCREB. CREB expression and CREB phosphorylation on serine 133 (pCREB) was assessed using antibodies to CREB and antibodies specific for the serine 133 phosphorylated form of CREB, respectively. (B) Quantitative analysis of the levels of CREB and CREB phosphorylated on serine 133 (pCREB). In uninfected hippocampal neurons and in neurons infected with rAAV-hrGFP, the ratio of pCREB to CREB (pCREB/CREB) was calculated for endogenous CREB; for hippocampal neurons infected with rAAV-CREB and rAAV-mCREB, the ratios were determined for overexpressed CREB and mCREB. Bars represent means ± SEM (n = 3). Statistical significance was determined by One-way ANOVA, Bonferroni post hoc test; statistically significant differences are indicated with an asterisk (***p < 0.001 versus rAAV-hrGFP; N.S., not significant).

Figure 3

Increasing levels of CREB protects against apoptosis and excitotoxicity. (A) Analysis of cell death in uninfected hippocampal neurons and in hippocampal neurons infected with rAAVs-hrGFP, rAAV-CREB or rAAV-mCREB. CREB and mCREB were detected with the M2 Flag antibody; hrGFP was detected with an antibody to hrGFP. Cell death was determined by analyzing Hoechst stained nuclei with shrunken cell body and large round chromatin clumps. For mCREB expression we obtained very weak Flag immunoreactivity in the death-inducing conditions, which is most likely due to compromised cell health following mCREB expression. Representative images are shown. Scale bar, 10 μm. Quantitative analysis of cell death induced by 10 nM staurosporine (B), growth factor withdrawal (C), or bath application of 20 μM NMDA (D). Bars represent means ± SEM (n = 4). Statistical significance was determined by Two-way ANOVA, Bonferroni post hoc test; statistically significant differences are indicated (***p < 0.001, **p < 0.01, *p < 0.05 compared to rAAV-hrGFP infected neurons in groups induced to undergo cell death; ###p < 0.001, ##p < 0.01 compared to rAAV-hrGFP infected neurons in untreated groups).

Figure 4

(A) Immunoblot analysis of expression and phosphorylation on serine 133 of endogenous CREB (endo-CREB; endo-pCREB) and overexpressed CREB (exo-CREB; exo-pCREB) in uninfected hippocampal neurons and in neurons infected with rAAV-hrGFP or rAAV-CREB at the indicated times after treatment with 20 μM NMDA. Expression of endogenous and overexpressed CREB was analyzed with an antibody to CREB; overexpressed CREB was also detected using an anti-Flag antibody. CREB phosphorylation on serine 133 (pCREB) was assessed using antibodies specific for the serine 133 phosphorylated form of CREB. CREB degradation products are indicated by an asterisk. (B) Quantitative analysis of the pCREB/CREB ratios of the immunoblot analyses shown in (A). Means ± SEM are given (n = 4).

Figure 5

QRT-PCR analysis ofatf3,btg2,gadd45β, andgadd45γandbdnfexpression in uninfected hippocampal neurons and in neurons infected with rAAV-hrGFP, rAAV-CREB and rAAV-mCREB. Gene expression analyses were done on DIV 9 (A) and DIV 10 (B), which correspond to five days and six days, respectively, post-infection. Bars represent SEM (n = 3). Statistical significance was determined by One-way ANOVA, Bonferroni post hoc test, statistically significant differences are indicated with an asterisk (***p < 0.001, **p < 0.01, *p < 0.05).