CREB expression mediates amyloid β-induced basal BDNF downregulation

Neurobiology of Aging xxx (2015) 1e8

Contents lists available at ScienceDirect

Neurobiology of Aging journal homepage: www.elsevier.com/locate/neuaging

CREB expression mediates amyloid b-induced basal BDNF downregulation Elyse Rosa, Margaret Fahnestock* Department of Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, Ontario, Canada

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 January 2015 Received in revised form 18 April 2015 Accepted 24 April 2015 Available online 30 April 2015

In Alzheimer’s disease, accumulation of amyloid-b (Ab) is associated with loss of brain-derived neurotrophic factor (BDNF), synapses, and memory. Previous work demonstrated that Ab decreases activityinduced BDNF transcription by regulating cyclic adenosine monophosphate response element binding protein (CREB) phosphorylation. However, the specific mechanism by which Ab reduces basal BDNF expression remains unclear. Differentiated, unstimulated human neuroblastoma (SH-SY5Y) cells treated with oligomeric Ab exhibited significantly reduced CREB messenger RNA compared with controls. Phosphorylated and total CREB proteins were decreased in both the cytoplasm and nucleus of Ab-treated cells. However, neither pCREB129 nor pCREB133 levels were altered relative to total CREB levels. The protein kinase A activator forskolin increased pCREB133 levels and prevented Ab-induced basal BDNF loss when administered before Ab but did not rescue BDNF expression when administered later. These data demonstrate a new mechanism for Ab-induced BDNF downregulation: in the absence of cell stimulation, Ab downregulates basal BDNF levels via Ab-induced CREB transcriptional downregulation, not changes in CREB phosphorylation. Thus, Ab reduces basal and activity-induced BDNF expression by different mechanisms. Ó 2015 Elsevier Inc. All rights reserved.

Keywords: Alzheimer’s disease Amyloid-b Brain-derived neurotrophic factor cAMP response element binding protein Human neuroblastoma SH-SY5Y mRNA

1. Introduction Alzheimer’s disease (AD) presents as global cognitive decline with associated memory loss and altered personality. The neuropathological hallmarks of AD include extracellular beta-amyloid (Ab)econtaining plaques, intracellular neurofibrillary tangles formed by hyperphosphorylated tau protein, synaptic loss, and neurodegeneration (Coleman and Flood, 1987; Coleman and Yao, 2003; Hyman et al., 2012; Hyman and Trojanowski, 1997; McKhann et al., 1984; Scheff and Price, 2003). According to the amyloid cascade hypothesis of AD, accumulation of Ab is the driving force behind AD neuropathology, and other pathological correlates are the result of an imbalance between Ab production and clearance (Hardy and Selkoe, 2002). Ab, and particularly Ab42, is prone to oligomerization, leading to the formation of aggregated oligomers of various sizes and conformations (Glabe, 2001; Walsh and Selkoe, 2007). These soluble oligomers are now thought to be the toxic species in AD (Ferreira et al., 2007; Garzon and Fahnestock, 2007; Hardy and Selkoe, 2002; Lacor et al., 2007; Walsh and Selkoe,

* Corresponding author at: Department of Psychiatry and Behavioural Neurosciences, McMaster University, 1280 Main Street West, Hamilton, Ontario, L8S 4K1, Canada. Tel.: 1 905 525 9140. E-mail address: [email protected] (M. Fahnestock). 0197-4580/$ e see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.neurobiolaging.2015.04.014

2007). Despite the well-documented consensus that soluble Ab exerts neurotoxic effects, the mechanism by which Ab exerts its toxicity is poorly understood. Some of the toxic effects of Ab are mediated by its adverse effect on neurotrophic factor expression, deficits of which may lead to synaptic loss and neurodegeneration. Specifically, brain-derived neurotrophic factor (BDNF) is critical for survival of the neurons that degenerate in AD, for synaptic plasticity, and for memory consolidation (Allen et al., 2011; Fahnestock, 2011; Lu, 2003). BDNF messenger RNA (mRNA) (Connor et al., 1997; Garzon et al., 2002; Holsinger et al., 2000; Phillips et al., 1991) and protein (Ferrer et al., 1999; Hock et al., 2000; Michalski and Fahnestock, 2003; Peng et al., 2005) are decreased in AD. BDNF transcript IV (formerly transcript III), 1 of 17 different transcripts resulting from alternative splicing, accounts for approximately half of the total BDNF mRNA found in the cortex (Garzon and Fahnestock, 2007; Pruunsild et al., 2007) and is downregulated in human AD cortical tissue (Garzon et al., 2002), in mouse models of AD (Peng et al., 2009), and in Ab-treated cells (Garzon and Fahnestock, 2007). Furthermore, this loss of BDNF occurs early, before plaque deposition and coinciding with memory deficits in AD transgenic mouse models (Francis et al., 2012) and in individuals with mild cognitive impairment (Peng et al., 2005). There is evidence that Ab-induced BDNF downregulation is mediated by cyclic adenosine monophosphate response element binding protein (CREB). CREB is a transcriptional regulator of BDNF,

2

E. Rosa, M. Fahnestock / Neurobiology of Aging xxx (2015) 1e8

particularly BDNF transcript IV (Pruunsild et al., 2011; Shieh et al., 1998; Tao et al., 1998), and it plays an important role in learning and memory processes. Phosphorylation of CREB at serine-133 by protein kinase A (PKA), Ca2þ-activated calmodulin kinases, or mitogen-activated protein kinase 2 is required for its activation (Walton and Dragunow, 2000) and initiation of BDNF transcription. Conversely, CREB is inactivated via phosphorylation at serine-129 by glycogen synthase kinase-3b (GSK3b) (Grimes and Jope, 2001; Wang et al., 1994). Previous work has demonstrated that Ab decreases CREB activity by activation of GSK3b (DaRocha-Souto et al., 2012) and by inactivation of PKA (Vitolo et al., 2002), both of which may contribute to Ab-induced impairment of activity-dependent BDNF expression (DaRocha-Souto et al., 2012; Tong et al., 2001). However, understanding the regulation of basal levels of BDNF in the absence of stimulation remains critically important, as the dramatic downregulation of BDNF in AD is not solely activitydependent. The main objective of this work is to determine the mechanism of Ab-induced downregulation of basal BDNF expression.

2. Methods

2.2. Quantitative real-time reverse transcription-polymerase chain reaction RNA was extracted from 3.65  105 SH-SY5Y cells as previously described (Rosa et al., 2015). RNA concentration and purity were determined by Multiskan GO and SkanIt software (Thermo Scientific, Nepean, ON, USA) at 260/ 280 nm. One microgram of SH-SY5Y RNA was reverse transcribed with Superscript III, following the manufacturer’s protocol (Invitrogen). Real-time PCR was carried out as described in Rosa et al. (2015) using 300 nM each forward and reverse BDNF primers (forward 50 -AAACATCCGAGGACAAGGTG-30 , reverse 50 AGAAGAGGAGGCTCCAAAGG-30 , Mobix, Hamilton, ON, USA), or 300 nM each forward and reverse b-actin primers (forward 50 -AGCCATGTACGTAGCCATCC-30 , reverse 50 -CTCTCAGCTGTGGTGGTGAA-30 ), or 300 nM each forward and reverse CREB primers (forward 50 CTGCCTCTGGAGACGTACAA-30 , reverse 50 -CAAGCACTGCCACTCTGTTT30 ). BDNF and CREB PCR standards were generated from purified PCR products using the primers listed, whereas b-actin standard was generated from a plasmid obtained from Invitrogen (Garzon and Fahnestock, 2007). Amplifications of samples, standards, and controls (no-RT and no-template controls) were run in triplicate. BDNF mRNA expression and CREB mRNA expression were normalized to the housekeeping gene b-actin.

2.1. SH-SY5Y cell culture and treatment 2.3. Cell viability assay Human neuroblastoma SH-SY5Y cells (ATCC, Manassas, VA, USA) were grown in Dulbecco’s Modified Eagle’s Medium containing 10% fetal bovine serum (Gibco BRL, Carlsbad, CA, USA), 1% L-glutamine (Gibco) and 1% penicillin/streptomycin (Gibco). Cells were incubated at 37  C and 5% CO2 in a 75 cm2 flask and split at a ratio of 2:3 with growth medium every 3e4 days. Cells were differentiated and treated with 5 mM oligomeric Ab as previously described for 24 or 48 hours (Garzon and Fahnestock, 2007). Lyophilized Ab42 peptide (rPeptide, Athens, GA, USA) was dissolved to 1 mM in 100% 1,1,1,3,3,3-hexafluoro-2-propanol (SigmaAldrich) and dried as described previously (Garzon and Fahnestock, 2007). Ab oligomers were prepared 24 hours before treatment by dissolving the biofilm in 100% dimethyl sulfoxide (DMSO) (SigmaAldrich) to obtain a 2 mM solution and sonicating at 37  C for 10 minutes. This solution was diluted 1:10 in Ham’s F-12 (phenol red free; BioSource, Camarillo, CA, USA), vortexed for 30 seconds, and incubated for 24 hours at 4  C. The Ab solution was then diluted 1:40 in Dulbecco’s Modified Eagle’s Medium containing 1% fetal bovine serum, 1% N2 Supplement, 1% L-glutamine, and 1% penicillin/streptomycin (treatment medium), giving a final concentration of 5 mM Ab and 0.25% DMSO. Vehicle-treated cells were exposed to 0.25% DMSO in treatment medium. There was no difference in BDNF expression between DMSO-treated cells and nonvehicle (medium only) treated controls (data not shown, p ¼ 1.00). CT 99021 (6-[[2-[[4-(2,4-dichlorophenyl)-5-(5-methyl-1Himidazol-2-yl)-2-pyrimidinyl]amino]ethyl]amino]-3-pyridinecarbonitrile) and forskolin (Sigma-Aldrich) were solubilized in 100% DMSO to obtain 4 mM and 50 mM solutions, respectively. Stock solutions of both CT 99021 and forskolin were diluted further into treatment medium, giving a final concentration of 2 mM CT 99021 with 0.05% DMSO and 30 mM forskolin with 0.06% DMSO. Cells were treated with either CT 99021 or forskolin beginning 30 minutes before Ab treatment for 48 hours. In separate experiments, forskolin was administered 24 hours after Ab treatment and remained on the cells for the subsequent 24 hours of treatment to determine if forskolin is capable of rescuing BDNF expression after downregulation by Ab. After treatment, GSK3b inactivation or PKA activation (via Western blot) and BDNF mRNA levels (via quantitative real-time reverse transcription-polymerase chain reaction) were compared with cells treated with Ab42 alone and to vehicle-treated groups.

Cell viability was determined by quantifying the release of lactate dehydrogenase (LDH) into conditioned medium. Briefly, 30 mL of conditioned medium was combined with 200 mL of LDH buffer (1M Tris pH 7.4, 1.4 mM Na pyruvate, 3.15 mM nicotinamide adenine dinucleotide), and LDH levels from each sample were quantified using a Multiskan GO microplate reader and SkanIt software (Thermo Scientific). Rabbit muscle LDH type II (SigmaAldrich), diluted 1:5000 and 1:25,000, was used as a positive control, and medium alone was used as a negative control. All samples were read at 340 nm every 30 seconds for 5 minutes. 2.4. Protein extraction In 150 mL of lysis buffer [50 mM Tris pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% Triton, 1 complete EDTA-free protease inhibitor tablet (Roche, Mississauga, ON, USA) and 1 PhosSTOP Phosphatase Inhibitor Cocktail Tablet (Roche) per 10 mL cell lysis buffer], 3.65x105 SH-SY5Y cells were lysed. Cell lysates were centrifuged at 14,000g for 5 minutes at 4  C, supernatants were collected, and protein concentrations determined using the DC Protein Assay (Bio-Rad Laboratories, Mississauga, Ontario, Canada) as described by the manufacturer, before Western blotting. 2.5. Nuclear/cytoplasmic localization of CREB SH-SY5Y cells were harvested using NE-PER Nuclear and Cytoplasmic Extraction Kit (Thermo Scientific) reagents according to manufacturer’s protocols. Briefly, cells were lysed in a cytoplasmic extraction buffer and centrifuged. Nuclear extraction buffer was added to the pellet, which was vortexed and centrifuged (14,000g) at 4  C to extract the remaining nuclear fraction. Protein concentrations of the samples were then determined using the DC Protein Assay (Bio-Rad Laboratories) before Western blotting. 2.6. Western blotting Twelve percent sodium dodecyl sulfateepolyacrylamide gels were used to separate 15e35 mg of total protein under reducing conditions before transferring to polyvinylidene fluoride

E. Rosa, M. Fahnestock / Neurobiology of Aging xxx (2015) 1e8

3

membranes (Bio-Rad, Hercules, CA, USA). Membranes were then blocked with a 1:1 solution of phosphate-buffered saline (PBS) pH 7.4 and Odyssey Blocking Buffer (Cedarlane, Burlington, Ontario, Canada) for 1 hour. After blocking, the blots were probed overnight at 4  C with the following primary antibodies: human CREB (Cell Signaling Technology, Danvers MA, USA; diluted 1:200), phosphorylated CREB (pCREB)-Ser133 (Cell Signaling Technology; diluted 1:200), pCREB-Ser129 (Santa Cruz Biotechnology, Dallas, TX, USA; diluted 1:200), b-catenin (Cell Signaling Biotechnology; diluted 1:500), or alpha-tubulin (Sigma-Aldrich; diluted 1:8000). After washing with PBS containing 0.5% Tween-20 (PBS-T), blots were incubated with the secondary antibodies IRDye 680conjugated goat anti-rabbit and IRDye 800CW-conjugated goat anti-mouse (LI-COR Biosciences, Lincoln, NE, USA; diluted 1:8000) for 1 hour at room temperature, washed with PBS-T, and scanned using an Odyssey Infrared Imaging System (LI-COR Biosciences). Band intensities were quantified by densitometry with local background subtraction using LI-COR Odyssey Software, version 2.0. 2.7. Statistical analysis All statistical analyses were carried out using IBM Statistics 22 software (SPSS, Chicago, IL, USA). A 2-sample t test, assuming equal variances or a 1-way analysis of variance with post hoc Tukey’s test for pairwise comparisons was done, according to the experimental design. Significance was set at p < 0.05, using a 2-tailed critical value. 3. Results 3.1. Oligomeric Ab1-42 downregulates basal BDNF expression BDNF mRNA was significantly downregulated in differentiated SH-SY5Y cells after treatment with 5/mM oligomeric Ab (Fig. 1A; p < 0.001), without affecting cell viability (no difference in LDH released; Fig. 1B; p ¼ 0.65), as we previously reported (Garzon and Fahnestock, 2007). This concentration of Ab has also been shown to decrease cellular signaling without affecting cell viability after cell stimulation with KCl and N-methyl-d-aspartate (Tong et al., 2001, 2004). However, it is important to note that our experimental conditions do not include cell stimulation. 3.2. Ab treatment of unstimulated cells decreases CREB transcription but does not alter CREB phosphorylation or subcellular localization

Fig. 1. Ab significantly downregulates BDNF and CREB mRNA without affecting cell viability. After treatment of differentiated SH-SY5Y cells with 5 mM oligomeric Ab for 48 hours, (A) BDNF mRNA was significantly reduced (Student t test; **p < 0.001). Error bars represent standard error of the mean (SEM) n ¼ 6 per group; Results were replicated over 7 independent experiments. (B) There was no difference in amount of LDH released between control cells (treated with 0.25% dimethyl sulfoxide [DMSO] in treatment medium) and cells treated with 5 mM oligomeric Ab also exposed to 0.25% DMSO (Student t test; p ¼ 0.65). Error bars represent SEM. n ¼ 11e12 per group. (C) CREB mRNA was also significantly reduced (Student t test; **p ¼ 0.009). CREB mRNA

CREB mRNA was significantly downregulated in differentiated SHSY5Y cells treated with 5 mM Ab compared with control cells (Fig. 1C; p ¼ 0.009). To determine whether decreased BDNF transcription following Ab treatment, in the absence of cell stimulation, could also be due to altered CREB phosphorylation, the activating phosphorylation (pCREB133) and inactivating phosphorylation (pCREB129) of CREB were quantified by Western blotting. Phosphorylated CREB at Ser-133 (normalized to total CREB) in cells treated with 5 mM Ab was not significantly different than in control cells (Fig. 2A; p ¼ 0.65). Similarly, phosphorylated CREB at Ser-129 (normalized to total CREB) in cells treated with 5 mM Ab was not significantly different than in control cells (Fig. 2B; p ¼ 0.43). To assess whether Ab treatment of unstimulated cells inactivates CREB by sequestering it outside the nucleus, phosphorylated and expression was normalized to b-actin mRNA for each sample. Error bars represent SEM. n ¼ 6 per group. Abbreviations: Ab, amyloid beta; BDNF, brain-derived neurotrophic factor; CREB, cyclic adenosine monophosphate response element binding protein; LDH, lactate dehydrogenase; mRNA, messenger RNA.

4

E. Rosa, M. Fahnestock / Neurobiology of Aging xxx (2015) 1e8

Fig. 2. Ab treatment does not affect phosphorylation of CREB. After treatment with 5 mM oligomeric Ab there was no difference in (A) pCREB133 levels (Student t test; p ¼ 0.65) or (B) pCREB129 levels (Student t test; p ¼ 0.43). Graphs represent integrated intensity of each target as defined by densitometric counts per mm2. Error bars represent standard error of the mean, n ¼ 18 per group. Abbreviations: Ab, amyloid beta; pCREB; phosphorylated cyclic adenosine monophosphate response element binding protein.

total CREB levels were quantified from both the nuclear and cytoplasmic fractions after Ab treatment and compared with controls. Phosphorylated CREB at Ser-133 was not significantly different in either the cytoplasmic (Fig. 3A; p ¼ 0.55) or the nuclear (Fig. 3B; p ¼ 0.39) fractions after Ab treatment compared with controls. However, total CREB protein was significantly decreased in both the cytoplasmic (Fig. 3C; p ¼ 0.009) and the nuclear (Fig. 3D; p ¼ 0.008) fractions after Ab treatment compared with controls. 3.3. Activation of PKA, but not inactivation of GSK3b, prevented but did not rescue Ab-induced downregulation of BDNF An inhibitor of GSK3b or activator of PKA was added in conjunction with Ab42 treatment to determine if Ab-induced BDNF downregulation could be prevented by manipulating either pathway. CT 99021 is a potent inhibitor of GSK3b that prevents phosphorylation of GSK3b substrates (Bain et al., 2007). Cells were exposed to either 2 mM CT 99021 alone or in combination with Ab42 and compared with both Ab42 alone and vehicle-treated groups. CT 99021-treated SH-SY5Y cells exhibited significantly reduced activation of GSK3b, as indicated by significantly increased total bcatenin levels (normalized to alpha-tubulin), compared with control cells (Fig. 4A; p ¼ 0.012). There was no effect of CT 99021 treatment on BDNF expression in the absence of Ab (Fig. 4B; p ¼ 0.50). Furthermore, CT 99021 inactivation of GSK3b was not sufficient to prevent Ab-induced downregulation of BDNF, as there was no difference between cells treated with Ab and those treated with Ab þ CT 99021 (Fig. 4B; p ¼ 0.87). Forskolin, a selective activator of adenylate cyclase (Seamon and Daly, 1981; Vitolo et al., 2002), was used to activate PKA. After treatment with 30 mM forskolin, differentiated SH-SY5Y cells exhibited significantly increased activation of PKA, as indicated by significantly increased pCREB133 levels (normalized to total CREB), compared with control cells (Fig. 5A; p < 0.001). However, there was no effect of forskolin treatment on BDNF expression in the absence of Ab (Fig. 5B; p ¼ 0.86). Unlike the inactivation of GSK3b, activation of PKA before Ab administration was sufficient to prevent Ab-induced BDNF downregulation. Cells treated with Ab42 alone had significantly lower BDNF mRNA than cells treated with Ab þ forskolin (Fig. 5B; p ¼ 0.01). However, when forskolin was administered 24 hours after Ab administration (a time at which BDNF was downregulated compared with vehicle-treated cells, data not shown; p ¼ 0.006), forskolin was not able to rescue Ab-induced BDNF downregulation. Cells exposed to forskolin after Ab treatment exhibited BDNF levels that were not significantly different from cells treated with Ab42 alone (Fig. 5C; p ¼ 0.654).

4. Discussion In AD, soluble aggregated Ab is thought to be the primary neurotoxic insult leading to synaptic loss and neurodegeneration. However, the mechanisms that lead from Ab aggregation to the pathological and physical symptoms of AD are not clear. In this study, we confirmed that oligomeric Ab significantly downregulates basal BDNF transcription (Garzon and Fahnestock, 2007) and determined the mechanism. We showed that Ab does not alter basal levels of CREB phosphorylation, but rather it induces CREB transcriptional downregulation, resulting in downregulation of BDNF. Thus, in the absence of cell stimulation, Ab reduces basal levels of BDNF by transcriptional downregulation of CREB, whereas previous reports (Tong et al., 2001, 2004; Vitolo et al., 2002) have shown that Ab, in the presence of cell stimulation, inhibits CREB phosphorylation. These data demonstrate a new mechanism for Ab-induced BDNF downregulation, different from its activity-dependent regulation. In this study, we used retinoic acid (RA)edifferentiated human neuroblastoma SH-SY5Y cells, which exhibit neuronal morphology, express BDNF, and its receptor TrkB and are dependent on BDNF for survival (Encinas et al., 2000; Feng et al., 2001; Kaplan et al., 1993). The response of these cells to Ab treatment mirrors that of human cortical neurons (Lambert et al., 1994). Treatment with a sub-toxic dose of Ab allows us to demonstrate that basal BDNF expression is downregulated by Ab specifically and not as a consequence of cell death. BDNF expression under sub-toxic Ab conditions is reduced by half (Fig. 1A), approximating the amount of BDNF decrease found in cortex of Alzheimer’s patients (Holsinger et al., 2000). This degree of BDNF downregulation is consistent with the idea that BDNF reduction still allows neurons to survive, albeit with reduced function, producing synaptic loss and memory dysfunction long before frank cell loss (Fahnestock, 2011). It has been shown previously that RA-differentiated SH-SY5Y cells express all 7 BDNF transcripts tested at similar levels as in human cortical tissue, with BDNF transcript IV accounting for more than half the total BDNF expressed (Garzon and Fahnestock, 2007). BDNF transcript IV is not only the most highly expressed BDNF transcript in RA-induced SH-SY5Y cells and in human cortex, but it is significantly reduced in AD and after Ab treatment (Garzon and Fahnestock, 2007; Garzon et al., 2002). BDNF transcript IV is regulated at least in part through CREB (Pruunsild et al., 2011; Shieh et al., 1998; Tao et al., 1998). The phosphorylation and subsequent activation of CREB result from the activity of several kinase pathways including protein kinase A (PKA), protein kinase C and phosphoinositide 3-kinase/protein kinase B (Walton and Dragunow, 2000). It has been shown that after cell stimulation,

E. Rosa, M. Fahnestock / Neurobiology of Aging xxx (2015) 1e8

5

Fig. 3. Ab does not sequester pCREB133 or total CREB outside the nucleus; Total CREB protein is decreased by Ab treatment. After treatment with 5 mM oligomeric Ab there was no difference in pCREB133 levels in either the (A) cytoplasmic fraction (Student t test; p ¼ 0.55) or (B) nuclear fraction (Student t test; p ¼ 0.39) of differentiated SH-SY5Y cells. Conversely, 5 mM oligomeric Ab treatment resulted in significantly reduced total CREB protein in both the (C) cytoplasmic fraction (Student t test; **p ¼ 0.009) and (D) nuclear fraction (Student t test; **p ¼ 0.008). Graphs represent integrated intensity (counts per mm2) of each target. Below each graph are representative Western blots for both pCREB133 and total CREB protein expression from the nuclear and cytoplasmic fractions for vehicle (V) and Ab (A) treated cells. Error bars represent standard error of the mean. n ¼ 4e5 per group. Abbreviations: Ab, amyloid beta; pCREB; phosphorylated cyclic adenosine monophosphate response element binding protein.

Ab can inactivate PKA in vitro, which decreases pCREB133 (Vitolo et al., 2002). Ab also inhibits the Ras/extracellular signal-regulated kinase and phosphoinositide 3-kinase/protein kinase B

pathways (Tong et al., 2004), thereby decreasing CREB activation by increasing GSK3b activity and pCREB129. Although the inactivation of CREB via phosphorylation may play an important role in the

Fig. 4. Inhibiting GSK3b is not sufficient to prevent Ab-induced BDNF downregulation. Treating differentiated SH-SY5Y cells with 2 mM CT 99021 was sufficient to (A) significantly decrease GSK3b activity as measured by total b-catenin levels (normalized to alpha-tubulin) (Student t test; *p ¼ 0.012). CT 99021 was unable to rescue Ab-induced BDNF downregulation: (B) Ab alone and Ab þ CT significantly downregulated BDNF compared with control cells (1-way analysis of variance and post hoc Tukey’s test **p < 0.001). Error bars represent standard error of the mean n ¼ 5e6 per group. Abbreviations: Ab, amyloid beta; BDNF, brain-derived neurotrophic factor; mRNA, messenger RNA.

6

E. Rosa, M. Fahnestock / Neurobiology of Aging xxx (2015) 1e8

Fig. 5. Activating PKA is sufficient to prevent but not rescue Ab-induced BDNF downregulation. (A) Treating differentiated SH-SY5Y cells with 30 mM forskolin significantly increased PKA activity as measured by phosphorylated CREB Ser-133 levels (normalized to total CREB) (Student t test; **p < 0.001). (B) Ab alone significantly downregulated BDNF compared with control cells (1-way analysis of variance (ANOVA) and post hoc Tukey’s test *p ¼ 0.02), but when forskolin was administered before Ab (Ab þ forskolin), this treatment resulted in levels of BDNF mRNA that were not significantly different from control cells (1-way ANOVA and post hoc Tukey’s test, p ¼ 0.86). Error bars represent standard error of the mean (SEM). n ¼ 5e6 per group. (C) Administration of forskolin 24 hours after Ab did not rescue BDNF levels. Both Ab alone (1-way ANOVA and post hoc Tukey’s test **p < 0.001) and Ab þ forskolin (1-way ANOVA and post hoc Tukey’s test **p ¼ 0.006) groups had significantly reduced BDNF mRNA normalized to b-actin mRNA. Error bars represent SEM. n ¼ 6 per group. Abbreviations: Ab, amyloid beta; BDNF, brainderived neurotrophic factor; mRNA, messenger RNA; pCREB, phosphorylated cyclic adenosine monophosphate response element binding protein.

effect of Ab on stimulated cells in vitro, we show here that the levels of phosphorylated CREB (both pCREB133 and pCREB129) are unaffected by Ab treatment in the absence of cell stimulation. This suggests that Ab-induced basal BDNF downregulation is not mediated by changes in CREB phosphorylation. Another mechanism of transcription factor inactivation is their sequestration outside of the nucleus. For example, the transcription factor splicing factor proline/glutamine-rich is sequestered in the cytoplasm in AD and in mutated tau-transfected SH-SY5Y cells (Ke et al., 2012). Additionally, mutant huntingtin sequesters CREB binding protein, preventing it from entering the nucleus to enhance CREB binding (Choi et al., 2012). We investigated whether Ab could sequester CREB outside the nucleus, but instead we found that the levels of total CREB in both the nucleus and the cytoplasm of Abtreated cells were significantly lower than in control cells. Thus, CREB is not sequestered in the cytoplasm, but rather the amount of total CREB protein in the cell is reduced after treatment with Ab oligomers. This was verified by our demonstration that Ab treatment significantly downregulates CREB mRNA. Reduced CREB, in turn, significantly decreases BDNF transcription. CREB is an essential component of molecular pathways required for learning and memory (Barco et al., 2003). Therefore, downregulation of CREB by Ab is expected to lead to cognitive deficits in AD. Recent findings have shown that CREB transcription is reduced in both AD postmortem hippocampal tissue and in Ab-treated rat hippocampal neurons (Pugazhenthi et al., 2011). Our report is the first to show that this Ab-induced CREB mRNA downregulation is associated with the significant downregulation of basal levels of BDNF expression. It has been suggested that Ab may downregulate CREB transcription via oxidative stress, as preincubation of neurons with the antioxidant N-acetyl cysteine prevented Ab-induced decreases in CREB mRNA (Pugazhenthi et al., 2011). Although others have shown that both the activating phosphorylation of CREB via activation of PKA (Vitolo et al., 2002) and inactivating phosphorylation of CREB via activation of GSK3b (DaRochaSouto et al., 2012) play an important role in Ab-induced toxicity, we show here that in the absence of cell stimulation, Ab has no effect on the phosphorylation of CREB. This importantly distinguishes that basal and activity-induced BDNF downregulation rely on different mechanisms. Furthermore, although inactivating GSK3b and thus reducing pCREB129 has no effect on Ab’s ability to downregulate basal BDNF, increasing the levels of pCREB133 using forskolin before Ab addition can prevent Ab-induced BDNF downregulation. However, if administered after Ab-induced BDNF downregulation, forskolin is unable to rescue BDNF expression. This result supports the view that altering CREB phosphorylation after Ab downregulates CREB transcriptionally is not sufficient to rescue Ab-induced basal BDNF downregulation. BDNF is essential for cognition and memory and is decreased early in the progression of AD. The amount of BDNF decrease in AD is directly correlated with the degree of cognitive decline (Peng et al., 2005). Our current findings reveal a novel mechanism of BDNF downregulation which could lead to new methods to combat BDNF decline in AD. Increasing BDNF levels has been shown to greatly improve learning and memory deficits in animal models (Ando et al., 2002; Blurton-Jones et al., 2009; Fahnestock et al., 2012; Nagahara et al., 2009, 2013). Furthermore, increasing CREB activity via viral delivery of CREB activators, CREB binding protein (Caccamo et al., 2010; Espana et al., 2010) and CREB-regulated transcription cofactor 1 (Espana et al., 2010), has successfully reversed synaptic atrophy and learning and memory impairments in transgenic mice. These findings highlight the possibility that increasing BDNF expression by modulating CREB mRNA levels in AD, even after clinical onset of the disease, could rescue memory impairments and cognitive function.

E. Rosa, M. Fahnestock / Neurobiology of Aging xxx (2015) 1e8

Disclosure statement The authors declare no conflicts of interest. Acknowledgements This study was funded by grants from the Alzheimer Society of Canada and the Canadian Institutes of Health Research (MOP-102723) to Margaret Fahnestock. Elyse Rosa was partially supported by an Ontario Graduate Scholarship. References Allen, S.J., Watson, J.J., Dawbarn, D., 2011. The neurotrophins and their role in Alzheimer’s disease. Curr. Neuropharmacol. 9, 559e573. Ando, S., Kobayashi, S., Waki, H., Kon, K., Fukui, F., Tadenuma, T., Iwamoto, M., Takeda, Y., Izumiyama, N., Watanabe, K., Nakamura, H., 2002. Animal model of dementia induced by entorhinal synaptic damage and partial restoration of cognitive deficits by BDNF and carnitine. J. Neurosci. Res. 70, 519e527. Bain, J., Plater, L., Elliott, M., Shpiro, N., Hastie, C.J., McLauchlan, H., Klevernic, I., Arthur, J.S., Alessi, D.R., Cohen, P., 2007. The selectivity of protein kinase inhibitors: a further update. Biochem. J. 408, 297e315. Barco, A., Pittenger, C., Kandel, E.R., 2003. CREB, memory enhancement and the treatment of memory disorders: promises, pitfalls and prospects. Expert Opin. Ther. Targets 7, 101e114. Blurton-Jones, M., Kitazawa, M., Martinez-Coria, H., Castello, N.A., Muller, F.J., Loring, J.F., Yamasaki, T.R., Poon, W.W., Green, K.N., LaFerla, F.M., 2009. Neural stem cells improve cognition via BDNF in a transgenic model of Alzheimer disease. Proc. Natl. Acad. Sci. U. S. A. 106, 13594e13599. Caccamo, A., Maldonado, M.A., Bokov, A.F., Majumder, S., Oddo, S., 2010. CBP gene transfer increases BDNF levels and ameliorates learning and memory deficits in a mouse model of Alzheimer’s disease. Proc. Natl. Acad. Sci. U. S. A.107, 22687e22692. Choi, Y.J., Kim, S.I., Lee, J.W., Kwon, Y.S., Lee, H.J., Kim, S.S., Chun, W., 2012. Suppression of aggregate formation of mutant huntingtin potentiates CREB-binding protein sequestration and apoptotic cell death. Mol. Cell Neurosci. 49, 127e137. Coleman, P.D., Flood, D.G., 1987. Neuron numbers and dendritic extent in normal aging and Alzheimer’s disease. Neurobiol. Aging 8, 521e545. Coleman, P.D., Yao, P.J., 2003. Synaptic slaughter in Alzheimer’s disease. Neurobiol. Aging 24, 1023e1027. Connor, B., Young, D., Yan, Q., Faull, R.L., Synek, B., Dragunow, M., 1997. Brain-derived neurotrophic factor is reduced in Alzheimer’s disease. Brain Res. Mol. Brain Res. 49, 71e81. DaRocha-Souto, B., Coma, M., Perez-Nievas, B.G., Scotton, T.C., Siao, M., SanchezFerrer, P., Hashimoto, T., Fan, Z., Hudry, E., Barroeta, I., Sereno, L., Rodriguez, M., Sanchez, M.B., Hyman, B.T., Gomez-Isla, T., 2012. Activation of glycogen synthase kinase-3 beta mediates beta-amyloid induced neuritic damage in Alzheimer’s disease. Neurobiol. Dis. 45, 425e437. Encinas, M., Iglesias, M., Liu, Y., Wang, H., Muhaisen, A., Cena, V., Gallego, C., Comella, J.X., 2000. Sequential treatment of SH-SY5Y cells with retinoic acid and brain-derived neurotrophic factor gives rise to fully differentiated, neurotrophic factor-dependent, human neuron-like cells. J. Neurochem. 75, 991e1003. Espana, J., Valero, J., Minano-Molina, A.J., Masgrau, R., Martin, E., GuardiaLaguarta, C., Lleo, A., Gimenez-Llort, L., Rodriguez-Alvarez, J., Saura, C.A., 2010. beta-Amyloid disrupts activity-dependent gene transcription required for memory through the CREB coactivator CRTC1. J. Neurosci. 30, 9402e9410. Fahnestock, M., 2011. Brain-derived neurotrophic factor: the link between amyloidb and memory loss. Future Neurol. 6, 627e639. Fahnestock, M., Marchese, M., Head, E., Pop, V., Michalski, B., Milgram, W.N., Cotman, C.W., 2012. BDNF increases with behavioral enrichment and an antioxidant diet in the aged dog. Neurobiol. Aging 33, 546e554. Feng, X., Jiang, H., Baik, J.C., Edgar, C., Eide, F.F., 2001. BDNF dependence in neuroblastoma. J. Neurosci. Res. 64, 355e363. Ferreira, S.T., Vieira, M.N., De Felice, F.G., 2007. Soluble protein oligomers as emerging toxins in Alzheimer’s and other amyloid diseases. IUBMB Life 59, 332e345. Ferrer, I., Marin, C., Rey, M.J., Ribalta, T., Goutan, E., Blanco, R., Tolosa, E., Marti, E., 1999. BDNF and full-length and truncated TrkB expression in Alzheimer disease. Implications in therapeutic strategies. J. Neuropathol. Exp. Neurol. 58, 729e739. Francis, B.M., Kim, J., Barakat, M.E., Fraenkl, S., Yucel, Y.H., Peng, S., Michalski, B., Fahnestock, M., McLaurin, J., Mount, H.T., 2012. Object recognition memory and BDNF expression are reduced in young TgCRND8 mice. Neurobiol. Aging 33, 555e563. Garzon, D., Yu, G., Fahnestock, M., 2002. A new brain-derived neurotrophic factor transcript and decrease in brain-derived neurotrophic factor transcripts 1, 2 and 3 in Alzheimer’s disease parietal cortex. J. Neurochem. 82, 1058e1064. Garzon, D.J., Fahnestock, M., 2007. Oligomeric amyloid decreases basal levels of brain-derived neurotrophic factor (BDNF) mRNA via specific downregulation of BDNF transcripts IV and V in differentiated human neuroblastoma cells. J. Neurosci. 27, 2628e2635. Glabe, C., 2001. Intracellular mechanisms of amyloid accumulation and pathogenesis in Alzheimer’s disease. J. Mol. Neurosci. 17, 137e145.

7

Grimes, C.A., Jope, R.S., 2001. CREB DNA binding activity is inhibited by glycogen synthase kinase-3 beta and facilitated by lithium. J. Neurochem. 78, 1219e1232. Hardy, J., Selkoe, D.J., 2002. The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 297, 353e356. Hock, C., Heese, K., Hulette, C., Rosenberg, C., Otten, U., 2000. Region-specific neurotrophin imbalances in Alzheimer disease: decreased levels of brain-derived neurotrophic factor and increased levels of nerve growth factor in hippocampus and cortical areas. Arch. Neurol. 57, 846e851. Holsinger, R.M., Schnarr, J., Henry, P., Castelo, V.T., Fahnestock, M., 2000. Quantitation of BDNF mRNA in human parietal cortex by competitive reverse transcription-polymerase chain reaction: decreased levels in Alzheimer’s disease. Brain Res. Mol. Brain Res. 76, 347e354. Hyman, B.T., Phelps, C.H., Beach, T.G., Bigio, E.H., Cairns, N.J., Carrillo, M.C., Dickson, D.W., Duyckaerts, C., Frosch, M.P., Masliah, E., Mirra, S.S., Nelson, P.T., Schneider, J.A., Thal, D.R., Thies, B., Trojanowski, J.Q., Vinters, H.V., Montine, T.J., 2012. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement. 8, 1e13. Hyman, B.T., Trojanowski, J.Q., 1997. Consensus recommendations for the postmortem diagnosis of Alzheimer disease from the National Institute on Aging and the Reagan Institute Working Group on diagnostic criteria for the neuropathological assessment of Alzheimer disease. J. Neuropathol. Exp. Neurol. 56, 1095e1097. Kaplan, D.R., Matsumoto, K., Lucarelli, E., Thiele, C.J., 1993. Induction of TrkB by retinoic acid mediates biologic responsiveness to BDNF and differentiation of human neuroblastoma cells. Eukaryotic Signal Transduction Group. Neuron 11, 321e331. Ke, Y.D., Dramiga, J., Schutz, U., Kril, J.J., Ittner, L.M., Schroder, H., Gotz, J., 2012. Taumediated nuclear depletion and cytoplasmic accumulation of SFPQ in Alzheimer’s and Pick’s disease. PLoS One 7, e35678. Lacor, P.N., Buniel, M.C., Furlow, P.W., Clemente, A.S., Velasco, P.T., Wood, M., Viola, K.L., Klein, W.L., 2007. Abeta oligomer-induced aberrations in synapse composition, shape, and density provide a molecular basis for loss of connectivity in Alzheimer’s disease. J. Neurosci. 27, 796e807. Lambert, M.P., Stevens, G., Sabo, S., Barber, K., Wang, G., Wade, W., Krafft, G., Snyder, S., Holzman, T.F., Klein, W.L., 1994. Beta/A4-evoked degeneration of differentiated SH-SY5Y human neuroblastoma cells. J. Neurosci. Res. 39, 377e385. Lu, B., 2003. BDNF and activity-dependent synaptic modulation. Learn Mem. 10, 86e98. McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., Stadlan, E.M., 1984. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939e944. Michalski, B., Fahnestock, M., 2003. Pro-brain-derived neurotrophic factor is decreased in parietal cortex in Alzheimer’s disease. Brain Res. Mol. Brain Res. 111, 148e154. Nagahara, A.H., Mateling, M., Kovacs, I., Wang, L., Eggert, S., Rockenstein, E., Koo, E.H., Masliah, E., Tuszynski, M.H., 2013. Early BDNF treatment ameliorates cell loss in the entorhinal cortex of APP transgenic mice. J. Neurosci. 33, 15596e15602. Nagahara, A.H., Merrill, D.A., Coppola, G., Tsukada, S., Schroeder, B.E., Shaked, G.M., Wang, L., Blesch, A., Kim, A., Conner, J.M., Rockenstein, E., Chao, M.V., Koo, E.H., Geschwind, D., Masliah, E., Chiba, A.A., Tuszynski, M.H., 2009. Neuroprotective effects of brain-derived neurotrophic factor in rodent and primate models of Alzheimer’s disease. Nat. Med. 15, 331e337. Peng, S., Garzon, D.J., Marchese, M., Klein, W., Ginsberg, S.D., Francis, B.M., Mount, H.T., Mufson, E.J., Salehi, A., Fahnestock, M., 2009. Decreased brainderived neurotrophic factor depends on amyloid aggregation state in transgenic mouse models of Alzheimer’s disease. J. Neurosci. 29, 9321e9329. Peng, S., Wuu, J., Mufson, E.J., Fahnestock, M., 2005. Precursor form of brain-derived neurotrophic factor and mature brain-derived neurotrophic factor are decreased in the pre-clinical stages of Alzheimer’s disease. J. Neurochem. 93, 1412e1421. Phillips, H.S., Hains, J.M., Armanini, M., Laramee, G.R., Johnson, S.A., Winslow, J.W., 1991. BDNF mRNA is decreased in the hippocampus of individuals with Alzheimer’s disease. Neuron 7, 695e702. Pruunsild, P., Kazantseva, A., Aid, T., Palm, K., Timmusk, T., 2007. Dissecting the human BDNF locus: bidirectional transcription, complex splicing, and multiple promoters. Genomics 90, 397e406. Pruunsild, P., Sepp, M., Orav, E., Koppel, I., Timmusk, T., 2011. Identification of ciselements and transcription factors regulating neuronal activity-dependent transcription of human BDNF gene. J. Neurosci. 31, 3295e3308. Pugazhenthi, S., Wang, M., Pham, S., Sze, C.I., Eckman, C.B., 2011. Downregulation of CREB expression in Alzheimer’s brain and in Abeta-treated rat hippocampal neurons. Mol. Neurodegener. 6, 60. Rosa, E., Cha, J., Bain, JR., Fahnestock, M., 2015. Calcitonin gene-related peptide regulation of glial cell-line derived neurotrophic factor in differentiated rat myotubes. J. Neurosci. Res. 93, 514e520. Seamon, K.B., Daly, J.W., 1981. Forskolin: a unique diterpene activator of cyclic AMPgenerating systems. J. Cyclic Nucleotide Res. 7, 201e224. Shieh, P.B., Hu, S.C., Bobb, K., Timmusk, T., Ghosh, A., 1998. Identification of a signaling pathway involved in calcium regulation of BDNF expression. Neuron 20, 727e740. Tao, X., Finkbeiner, S., Arnold, D.B., Shaywitz, A.J., Greenberg, M.E., 1998. Ca2þ influx regulates BDNF transcription by a CREB family transcription factor-dependent mechanism. Neuron 20, 709e726.

8

E. Rosa, M. Fahnestock / Neurobiology of Aging xxx (2015) 1e8

Tong, L., Balazs, R., Thornton, P.L., Cotman, C.W., 2004. Beta-amyloid peptide at sublethal concentrations downregulates brain-derived neurotrophic factor functions in cultured cortical neurons. J. Neurosci. 24, 6799e6809. Tong, L., Thornton, P.L., Balazs, R., Cotman, C.W., 2001. Beta -amyloid-(1-42) impairs activity-dependent cAMP-response element-binding protein signaling in neurons at concentrations in which cell survival is not compromised. J. Biol. Chem. 276, 17301e17306. Vitolo, O.V., Sant’Angelo, A., Costanzo, V., Battaglia, F., Arancio, O., Shelanski, M., 2002. Amyloid beta -peptide inhibition of the PKA/CREB pathway and long-

term potentiation: reversibility by drugs that enhance cAMP signaling. Proc. Natl. Acad. Sci. U. S. A. 99, 13217e13221. Walsh, D.M., Selkoe, D.J., 2007. A beta oligomers - a decade of discovery. J. Neurochem. 101, 1172e1184. Walton, M.R., Dragunow, I., 2000. Is CREB a key to neuronal survival? Trends Neurosci. 23, 48e53. Wang, Q.M., Roach, P.J., Fiol, C.J., 1994. Use of a synthetic peptide as a selective substrate for glycogen synthase kinase 3. Anal. Biochem. 220, 397e402.