Apolipoprotein E4 Stimulates cAMP Response Element-binding Protein Transcriptional Activity through the Extracellular Signal-regulated Kinase Pathway*

Inheritance of the e 4 allele of the apolipoprotein E gene ( APOE4 ) is a major risk factor for the development of Alzheimer’s disease (AD). Although the association between APOE4 and AD is well documented, the mechanism by which apolipoprotein E exerts an isoform-spe-cific effect on neurons in disease is unknown. In this report, we demonstrate that apoE4 stimulates the transcriptional activity of cAMP-response element-binding protein (CREB) by activating the extracellular signal-regulated kinase (ERK) cascade in rat primary hippocampal neurons. In contrast, apoE3 was unable to stimulate CREB transcriptional activity and unable to activate the ERK pathway. Elevation of intracellular Ca 2 1 levels are also involved because treatment with receptor-associated protein, nifedipine, MK801, removal of Ca 2 1 from the medium and dantrolene all served to inhibit calcium elevation and attenuate the activation of CREB. Treatment with an apoE peptide was also found to facilitate transcription of the CREB-dependent genes, c- fos L-type voltage-dependent calcium channels, which is open during strong depolarization, permit the entry of larger amount of Ca 2 1 along the dendrites and into the cell bodies, causing the opening of ryanodine-sensitive Ca 2 1 channel on ER membrane and the influx of larger amount of Ca 2 1 from ER in Ca 2 1 -induced Ca 2 1 release fashion. The Ca 2 1 entry from extracellular space may be negligible, comparing that from ER calcium elevation allows activation of PKA-dependent Rap1-ERK pathway and phosphorylation of CREB.

Alzheimer's disease is a devastating neurological disorder characterized by progressive memory loss and cognitive deficits. To date, four genes have been reported to be associated with Alzheimer's disease phenotypes including the amyloid precursor protein gene on chromosome 21 (1)(2)(3)(4), the presenilin-1 gene on chromosome 14 (5), the presenilin-2 gene on chromosome 1 (6 -7), and the apolipoprotein-E (APOE) 1 gene on chromosome 19 (8 -10). There are three common alleles of the APOE gene: ⑀3 is the most common representing ϳ78% of all APOE alleles, ⑀4 representing 15% and ⑀2 representing 7% (11). Although these alleles give six possible genotypes, the risk of Alzheimer's disease is increased, and the age of onset distribution is earlier, with each dose of ⑀4 (8). While the influence of the APOE-⑀4 genotype was clearly demonstrated, the biological mechanism mediated by apoE4, the translated protein isoform derived from the ⑀4 allele of APOE gene remains speculative.
Several hypotheses have been proposed regarding potential mechanisms where apoE4 enhances the risk for Alzheimer's disease. One interesting hypothesis is that apoE4 itself causes toxicity to hippocampal neurons, but that apoE3 does not (12)(13). A 22-kDa thrombin cleavage fragment of apoE4 known as "truncated apoE4" and an apoE peptide consisting of a tandem repeat of amino acid residues 141-149 of apoE showed higher toxicity to neurons than apoE4 itself (14 -15). Their toxicity was apoE receptor-mediated (14) and involved calcium influx (16). Based on these reports, we started to study the mechanism of apoE toxicity using the synthetic apoE peptide.
We employed the Mercury Pathway Profiling System (CLONTECH) to elucidate the mechanistic pathways activated in the presence of apoE peptide. This system evaluates several cis-acting enhancer elements such as AP1, CRE, HSE, Myc, NFB, and serum response element and among them, transcription from the CRE element was found to be activated. In addition to the apoE peptide, a specific isoform of the apoE protein was found to stimulate the phosphorylation that activates CREB. We then examined the pathway upstream of CREB activation and found that ERK and the apoE receptor are involved in this same pathway. Downstream of CREB, the expression of c-fos and Bcl-2 genes that are modulated by activated CREB, were also increased. Our finding that apoE4 stimulates phosphorylation of CREB through an ERK pathway to increase transcription of specific genes while apoE3 does not, suggests a mechanism of apoE isoform-specific function on neurons.
ApoE Peptide Synthesis-A 30-amino acid apoE peptide, which is a tandem repeat of apoE amino acid residues 141-155 (LRKLRKRLLRDADDL) was synthesized by the solid-phase method as described previously (17) and greater than 95% purity was determined by high performance liquid chromatography and mass spectrometry.
Cell Culture-Primary cultures of rat hippocampal neurons were prepared from Wistar rats at embryonic day 18 as described previously (18). Cells were routinely propagated in Dulbecco's modified Eagle's medium (Sigma) with 10% fetal calf serum (Life Technologies, Inc.). All experiments were performed on cells cultured for 5 to 8 days.
Primary astrocyte cultures were prepared from neonatal apoE-deficient mice as described previously (19) with minor modifications. In brief, apoE-deficient mice were purchased from Taconic Farms (Germantown, NY) and backbred to C57BL/6 for at least six generations, brains were removed from neonatal apoE-deficient mice within 24 h of birth, meninges were removed, and brain tissues were digested at 37°C with Dispase II (Rochem Molecular Biochemicals, Germany) in PBS. The mixture was centrifuged (3000 ϫ g for 10 min), plated in 175-cm 2 culture flasks (two brains/flask), and cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum. After 10 days, cells were incubated for 48 h with cytosine arabinofuranoside (10 g/ml; Wako, Osaka, Japan) to prevent fibroblast overgrowth. Astrocytes were then separated from microglia and oligodendroglia by agitation on a shaking platform (Bioshaker BR-30L, Tautek, Tokyo, Japan) and identified by immunoreactivity with an anti-GFAP antibody. Cultures used for experiments were Ͼ98% astrocytes based on these techniques.
Stimulation of Primary Astrocytes and Quantification of TNF␣-Inhibition of secretion of TNF␣ from primary astrocytes by apoE was quantified as described previously (20). Briefly, primary astrocytes from apoE-deficient mice were plated on 96-well tissue culture dishes at a density of 20,000 cells/well and incubated in serum-free media (Opti-MEM I, Life Technologies, Inc.) containing 1% N2 supplement (Life Technologies, Inc.). The following day, recombinant apoE3, apoE4, or control protein (denatured apoE3 by boiling for 30 min) was added. 24 h later, 100 ng/ml lipopolysaccharide (Sigma) was added. 60 h later, 50 l of medium was removed and TNF␣ secretion measured by Quantikine M Mouse TNF␣ ELISA kit (R & D systems) as described in the manufacturer's protocol.
Promoter-reporter Assay-To detect the pathway activated in the presence of apoE peptide, we utilized the Mercury Pathway Profiling Systems (CLONTECH). Briefly, rat hippocampal neurons were cultured in 6-well dishes, transiently transfected with promoter-reporter plasmids using LipofectAMINE 2000 (Life Technologies, Inc.) as described in the manufacturer's protocol. These plasmids contained the secreted alkaline phosphatase (SEAP) reporter gene downstream of several copies of specific transcription factors binding sequences such as AP1, CRE, HSE, Myc, NFB, and serum response element. Transfected neurons were allowed to recover for 24 h before 2 M apoE peptide was added to the media. Alkaline phosphatase activities in the media, at indicated times after addition of apoE peptide, were measured using Great EscApe SEAP Chemiluminescence Detection Kit (CLON-TECH) following the manufacturer's protocol.
Western Blot Analysis-Cells were cultured in 6-well dishes, washed three times with PBS, and lysed with 200 l of lysis buffer (50 mM Tris-HCl, pH 8.0, 20 mM EDTA, 1% SDS, and 100 mM NaCl). 20-g samples were boiled for 5 min, electrophoresed on 12.5% SDS-polyacrylamide electrophoresis gels and transferred onto Immobilon membrane (Millipore Corp., Bedford, MA). The membrane was incubated in blocking buffer (1 ϫ PBS, 5% nonfat dried milk) for 1 h at room temperature and then probed with a primary antibody in blocking buffer overnight at 4°C. After four washes in PBS containing 0.3% Tween 20, blots were probed with the secondary antibody in blocking buffer for 1 h at room temperature, and washed again in PBS containing 0.05% Tween 20. Detection of signal was performed with an enhanced chemiluminescence detection kit (Amersham International, Little Chalfont, United Kingdom).
Measurement of Intracellular Calcium Levels-Fura-2 AM (Molecular probes, Eugene, OR) and an Argus 50/CA system (Hamamatsu Photonics, Japan) were used to quantify the cytoplasmic free calcium as described in the manufacturer's protocol.

Recombinant ApoE4 and Recombinant ApoE3 Were Equally Active in Suppressing the Secretion of TNF␣ from Astrocytes-
The recombinant apoE4 and apoE3 proteins used for this study were not degraded as determined by Western blot analysis of denatured and reduced apoE proteins (Fig. 1A). To show that these recombinant proteins also retained biological activity, we examined whether they equally inhibited the secretion of TNF␣ from astrocytes as described by Laskowitz et al. (20). Preincubation of primary astrocytes cultures prepared from neonatal apoE-deficient mice with recombinant apoE3 or with recombinant apoE4, prior to stimulation with 100 ng/ml lipopolysaccharide, each decreased the concentration of TNF␣ released into the conditioned medium in a dose-dependent fashion as compared with untreated cells (Fig. 1B). There was no significant difference in TNF␣ levels between cells treated with the same concentration of apoE3 or apoE4. Boiled and denatured apoE3 was used as control protein and it did not suppress the secretion of TNF␣.
ApoE Peptide Facilitates CRE-driven Transcription-To detect the signaling pathway activated in the presence of apoE peptide, we utilized the Mercury Pathway Profiling Systems (CLONTECH). Several cis-acting enhancer elements, such as AP1 (activator protein 1), CRE (cAMP responsive element), HSE (heat shock element), Myc, NFB (nuclear factor of B), and serum response element included in this system were assessed in rat primary hippocampal neurons exposed to 1 M apoE peptide. Among these elements, transcription from the CRE element was found to be activated (Fig. 2).
ApoE Peptide and ApoE4, but Not ApoE3, Facilitated Phosphorylation of CREB at Ser-133-Since apoE peptide stimulated transcription from the CRE element, then the CREB transcription factor could be activated by apoE treatments. Phosphorylation of CREB at Ser-133 leads to its activation as a transcription factor (21,22). To monitor the activation of CREB directly, Western blots of cell lysates from rat hippocampal neurons stimulated with 1 M apoE peptide for the indicated times were probed with anti-CREB and anti-phospho-CREB (p-CREB) antibodies. 15 min after addition of apoE peptide, phospho-CREB levels substantially increased followed by a gradual decrease over the next 12 h while the total amount of CREB did not appear to change (Fig. 3A). In addition to apoE peptide, treatment with 1 M apoE4 protein also appeared to significantly increase phospho-CREB levels by 1 h after treatment, while 1 M apoE3 protein did not appear to change phospho-CREB levels significantly at any time (Fig. 3, B-D). 1. Characterization of recombinant apoE3 and apoE4. 5 ng of recombinant apoE3 and apoE4 were electrophoresed on denaturing and reducing polyacrylamide gels, Western blotted, and probed with anti-apolipoprotein E antibody, demonstrating that both are not degraded (A). Preincubation with recombinant apoE3 or apoE4 inhibited the secretion of TNF␣ from astrocytes prepared from apoE-deficient mice in a dose-dependent fashion as compared with heat-denatured apoE3. The difference between apoE3 and apoE4 was not significant at any concentration (Student's t test). All results are plotted as a mean Ϯ S.E. of four data (B).
Although the molar concentration of apoE4 and apoE peptide added to the media was equivalent, the kinetics (time course) of CREB phosphorylation following treatment with apoE4 protein appeared to be slower than that stimulated by apoE peptide treatment.
ApoE Peptide and ApoE4, but Not ApoE3, Facilitated ERK Activity in Rat Hippocampal Neurons-Since activated CREB is phosphorylated at Ser-133 and apoE treatment increases levels of phosphorylated CREB, then apoE may stimulate a pathway resulting in kinase-mediated phosphorylation of CREB. One candidate for CREB phosphorylation at Ser-133 is through the ERK pathway that is activated by a double phosphorylation of ERK at Thr-202 and Tyr-204. We examined ERK activation by measuring phosphorylated ERK levels following treatment with apoE peptide, apoE3, and apoE4 proteins. Treatment with apoE peptide or with apoE4 protein resulted in significantly increased levels of phosphorylated ERK while apoE3 treatment failed to increase levels (Fig. 4). ApoE peptide increased phospho-ERK levels by 15 min of treatment while apoE4 protein increased them by 1 h of treatment. Like the phosphorylation of CREB, phosphorylation of ERK following apoE4 protein treatment appeared to be slower than that following apoE peptide treatment.  apoE treatment, we examined the effects of RAP which is a competitive blocker of apoE for its receptor; MK801 which is a selective antagonist of the NMDA receptor; and nifedipine which is a selective antagonist of the L-type voltage-dependent calcium channel (LVDCC). Pretreatment with RAP followed by apoE appeared to attenuate the increase in phospho-CREB levels seen with apoE treatment alone suggesting that the apoE receptor is involved in the pathway (Fig. 6A). Similarly, pretreatment with MK801 or with nifedipine appeared to at-tenuate the increase in phospho-CREB levels, suggesting that the NMDA receptor and/or the LVDCC are also involved in the pathway (Fig. 6B).
Activation of MEK and Protein Kinase A Is Involved in the Pathway-To find out which of the possible pathways involving apoE receptors, LVDCC and/or NMDA receptors, contribute to the phosphorylation of CREB, we examined the effect of inhibiting kinases associated with these receptors by treatment with the U0126 inhibitor of MAPK/ERK kinase (MEK) and with the c-AMP-dependent protein kinase inhibitor of protein kinase A (PKA). Pretreatment with one of these selective inhibitors, as well as pretreatment with both, mostly inhibited CREB phosphorylation suggesting that activation of MEK and PKA are involved in this pathway (Fig. 6C).
ERK Cascade Is a Major Pathway Leading to the ApoE Peptide-stimulated Phosphorylation of CREB-To further investigate the apoE stimulation pathway, we examined the effect of inhibiting kinases associated with the receptor signaling by treatment with the SB203580 inhibitor of p38 MAP kinase, KN62 inhibitor of calmodulin-dependent protein kinases (CaMK II/IV), and the U0126 inhibitor of MEK. Of these selective inhibitors, CREB phosphorylation was mostly inhibited by pretreatment with U0126 while pretreatment with KN62 or SB203580 did not appear to have any significant effect on the phosphorylation status of CREB. This result suggests that the ERK cascade is a major pathway leading to the apoE-stimulated phosphorylation of CREB (Fig. 6D).

Elevation of Intracellular Ca 2ϩ following ApoE Peptide Treatment Involves the ApoE Receptor, NMDA Receptor, and L-type Voltage-dependent Ca 2ϩ
Channel-To demonstrate if the elevation of intracellular Ca 2ϩ parallels the phosphorylation of CREB, we measured the concentration of intracellular Ca 2ϩ under unstimulated or apoE peptide-stimulated conditions. Treatment with apoE peptide significantly raised the concentration of intracellular calcium in rat hippocampal neurons. Preincubation with RAP, MK801, or nifedipine significantly attenuated this elevation, suggesting that the apoE receptor, NMDA receptor, and LVDCC are involved in the pathway to calcium elevation as was observed for the activation of CREB (Fig. 7A).
A Major Part of the Intracellular Ca 2ϩ Elevation May Be from an Intracellular Source-To determine whether calcium influx from the extracellular space or from the endoplasmic reticulum (ER) contributes to our apoE-mediated Ca 2ϩ increases, we measured the intracellular Ca 2ϩ elevation in rat primary hippocampal neurons cultured in medium lacking Ca 2ϩ and containing 1 mM EGTA (Fig. 7B). Under these conditions, apoE-stimulated calcium elevation was significantly attenuated suggesting that calcium influx from extracellular sources significantly contributes to the elevation of intracellular Ca 2ϩ . To confirm this finding, we also measured the contribution of calcium release from the endoplasmic reticulum to the increase in intracellular Ca 2ϩ . Rat primary hippocampal neurons were preincubated with 10 M dantrolene, an inhibitor of ryanodine-sensitive Ca 2ϩ channel found on the ER membrane, followed by apoE peptide treatment. Interestingly, preincubation with dantrolene also significantly attenuated the elevation of intracellular Ca 2ϩ . Combining extracellular Ca 2ϩfree conditions and preincubation with dantrolene also attenuated the calcium elevation, even though there was no significant difference between the calcium levels measured under extracellular Ca 2ϩ -free conditions without dantrolene preincubation and the extracellular calcium-free condition with dantrolene preincubation. Taken together, we hypothesize that the calcium influx from extracellular sources triggers the release of Ca 2ϩ from intracellular endoplasmic reticulum sources through its ryanodine-sensitive Ca 2ϩ channels. Calcium influx from extracellular space seems to be essential for signaling the apoE-stimulated elevation of intracellular Ca 2ϩ levels, but contributes only a small amount to the rise in intracellular calcium levels compared with the calcium released from the ER. DISCUSSION Although evidence of the association between the ⑀4 allele of the APOE gene and Alzheimer's disease is overwhelming, the mechanism by which the apoE4 protein isoform influences onset and progression of the disease and its pathology is unknown. Of the many suggested mechanisms, we have focused on reports that apoE3 and apoE4 protein isoforms have differential effects on neuronal plasticity and survival (13,23,24). Compared with the apoE3 protein isoform, apoE4 protein, proteolytic fragments of apoE4 protein, and peptides corresponding to the receptor-binding domain of apoE proteins appear to actively injure, and certainly do not support maintenance of healthy neurites and neuronal cells (13)(14)(15)(16). On the larger scale, the failure of apoE4 protein isoforms and their related fragments to support neuronal plasticity and maintenance may infer a mechanism that underlies the association between APOE gene alleles and disease.
Under many conditions that eventually result in neuronal death, the cell struggles to induce protective mechanisms even though destructive forces inevitably march forward. In this report, we have demonstrated one such scenario where apoE4, but not apoE3, activates an ERK cascade that results in activation of CREB and induction of many different genes including the cell-protective gene, Bcl-2. In prior reports, we showed that overexpression of only Bcl-2 protein could inhibit neuronal death following a toxic insult (25). In this case, treatment of neurons with recombinant apoE4 protein or with synthetic apoE peptide resulted in CREB phosphorylation and induced Bcl-2 expression, events which were not observed following recombinant apoE3 protein treatments. Time course experiments showed that the increase in phospho-CREB levels following apoE4 protein treatments appeared to be slower than that by apoE peptide, although their molar concentrations were the same. This time lag is consistent with previous results where protease inhibitors reduced the neurotoxicity of apoE4 (16) through a presumed mechanism where the full-length apoE4 protein must be digested over time to produce a toxic fragment of apoE4. Despite the increases in phospho-CREB and Bcl-2, we also observed apoE4/apoE peptide-mediated increases in calcium levels that were also observed by others in the context of their association with neuronal death (16,26).
The apoE4/apoE-peptide-mediated increase of intracellular Ca 2ϩ can be generated by calcium influx from the extracellular space and release of calcium from intracellular stores. Significant attenuation of increased intracellular calcium levels was observed when we cultured rat primary hippocampal neurons in the medium lacking Ca 2ϩ and containing EGTA as a Ca 2ϩ chelator, suggesting that calcium influx from extracellular sources follows apoE4/apoE peptide treatments. When we measured the intracellular Ca 2ϩ elevation following preincubation with 10 M dantrolene, an inhibitor of ryanodine-sensitive Ca 2ϩ channel on ER membranes, Ca 2ϩ elevation was also significantly attenuated, suggesting the participation of the calcium influx from ER in the elevation of intracellular Ca 2ϩ .
Interestingly, extracellular Ca 2ϩ -free conditions did not seem to have any additional effect on attenuating the Ca 2ϩ elevation following preincubation with dantrolene, suggesting that calcium influx from the extracellular space may serve to trigger a larger amount of Ca 2ϩ release from the ER through ryanodinesensitive Ca 2ϩ channel in a "Ca 2ϩ -induced Ca 2ϩ -release" fashion. Thus, calcium influx from extracellular space may be less than calcium release from intracellular sources such as the endoplasmic reticulum.
The apoE peptide-mediated increase in phospho-CREB levels and of intracellular Ca 2ϩ levels was inhibited by pretreatment with RAP, MK801, or nifedipine, suggesting that apoE receptors, NMDA receptor, and LVDCC are involved in rat hippocampal neuron responses to apoE. Tolar et al. (16) re-ported that RAP and MK801 attenuated the elevation of intracellular Ca 2ϩ caused by truncated apoE. They also reported that nifedipine did not provide any protection against the apoE-mediated rise in intracellular Ca 2ϩ in rat hippocampal neurons, but data was not shown. Wang et al. (26) reported that RAP, MK801, and diltiazem, another inhibitor of LVDCCs, failed to block the apoE peptide-induced calcium influx in rat primary cultured neurons. Assuming that neurotoxicity is caused by calcium influx following apoE treatment, our data are consistent with the data provided by Tolar et al. (16) in that RAP and MK801 attenuated calcium influx (16). This result suggests that LRP or some other apoE receptors found in neurons may be linked to NMDA receptor. Our finding that preincubation with nifedipine almost completely inhibited the apoE peptide-induced calcium influx is reasonable because calcium entry through NMDA receptors may depolarize the plasma membrane and open LVDCCs that results in larger amount of calcium influx. Compared with the calcium entry from LVDCCs, calcium entry from NMDA receptors may be negligible (27). Nevertheless, this calcium elevation can activate kinases such as ERK and calmodulin kinase II that function to phosphorylate CREB (27).
Further inhibition experiments showed that pretreatment with a MEK inhibitor and/or a PKA inhibitor mostly prevented the apoE-peptide stimulated phosphorylation of CREB, suggesting that activation of both MEK and PKA kinases are involved in the pathway. Ca 2ϩ elevation stimulates CREB phosphorylation by activation of a PKA-dependent, Rap1-MEK-ERK pathway (28). Pretreatment with KN62 or SB203580 did not appear to have any effect on the phosphoryl-FIG. 7. Intracellular calcium levels increase after apoE-peptide treatment. As a measure of intracellular calcium levels, Fura 2 fluorescence in 7-day-old rat hippocampal neuronal cultures increases rapidly after treatment with apoE-peptide (bar labeled a). This apoEpeptide stimulated fluorescence is inhibited by pretreatment with RAP, MK801 at 10 or 100 M, or nifedipine at 10 or 100 M. Peak values of intracellular calcium were observed about 40 s after exposure to apoEpeptide in at least 10 neurons per experiment in each of four separate experiments. Compared with the apoE-peptide alone, significant inhibition of calcium fluorescence at the p Ͻ 0.01 are calculated with a Student's t test and marked by a double asterisk (A). ApoE-peptide induces a rapid elevation of intracellular calcium fluorescence in rat hippocampal neurons (bar a) that is inhibited by culturing in Ca 2ϩ -free growth medium or in the presence of growth medium containing calcium and 10 M dantrolene. Peak concentrations of intracellular calcium in at least 10 neurons for each of four experiments are plotted. Significance of the difference between apoE-peptide alone and treatment conditions were calculated with Student's t test where double asterisks represent p Ͻ 0.01 (B).

FIG. 8. A model of apoE-mediated CREB activation in hippocampal neurons.
When apoE4, but not apoE3, binds to apoE receptor on the surface of hippocampal neuron, NMDA receptors open. Its mechanism is unclear. Activation of NMDA receptor causes small amounts of Ca 2ϩ entry into postsynaptic spines. The resulting depolarization triggers the opening of L-type voltage-dependent calcium channels exist in cytoplasmic membranes of the neurons. The activation of L-type voltage-dependent calcium channels, which is open during strong depolarization, permit the entry of larger amount of Ca 2ϩ along the dendrites and into the cell bodies, causing the opening of ryanodinesensitive Ca 2ϩ channel on ER membrane and the influx of larger amount of Ca 2ϩ from ER in Ca 2ϩ -induced Ca 2ϩ release fashion. The Ca 2ϩ entry from extracellular space may be negligible, comparing that from ER calcium elevation allows activation of PKA-dependent Rap1-ERK pathway and phosphorylation of CREB. ation status of CREB, suggesting that the ERK cascade is a major pathway leading the phosphorylation of CREB following apoE treatment.
Upon binding to the CRE element of a gene's promoter, the transcription factor known as CREB facilitates transcription of many different genes, such as c-fos, Bcl-2, and BDNF (27, 29 -31). CREB that actively binds its DNA element, CRE, is typically phosphorylated on serine at position 133. Activation of CREB by its phosphorylation at serine residue 133 is modulated by a variety of effectors. ␣-amino-3-hydroxy-5-methyl-4isoxazole propionic acid (AMPA) /kainate receptors and NMDA receptors are associated with synapses and LVDCCs are involved in neuronal functions like long-term potentiation (32)(33)(34). Stimulation of these receptors has also been associated with increased intracellular Ca 2ϩ levels, potentiation of kinase activities, and CREB activation.
To clarify if calcium influx caused by apoE treatment causes neurotoxicity and parallels phosphorylation of CREB, we measured the elevation of intracellular Ca 2ϩ caused by apoE peptide in rat hippocampal neurons upon exposure to these receptor antagonists. As others has reported (16), treatment of apoE peptide to rat primary hippocampal neurons caused calcium influx and intracellular Ca 2ϩ levels to rise. Preincubation with RAP, MK801, or nifedipine clearly attenuated the Ca 2ϩ elevation (Fig. 7A), which parallels our data on CREB phosphorylation mediated by apoE peptide, suggesting the involvement of apoE receptor.
We have shown a possible mechanism whereby apoE4 facilitates the expression of CRE-driven genes and apoE3 does not (Fig. 8). We hypothesize that apoE4, which stimulates CREB phosphorylation, may be associated with synaptic plasticity. This apparent isoform specificity of the effects might be relevant to the isoform-specific association of apoE4 as a significant risk factor in late-onset Alzheimer's disease.