Combined treatment with the phenolics (−)-epigallocatechin-3-gallate and ferulic acid improves cognition and reduces Alzheimer-like pathology in mice

“Nutraceuticals” are well-tolerated natural dietary compounds with drug-like properties that make them attractive as Alzheimer's disease (AD) therapeutics. Combination therapy for AD has garnered attention following a recent National Institute on Aging mandate, but this approach has not yet been fully validated. In this report, we combined the two most promising nutraceuticals with complementary anti-amyloidogenic properties: the plant-derived phenolics (−)-epigallocatechin-3-gallate (EGCG, an α-secretase activator) and ferulic acid (FA, a β-secretase modulator). We used transgenic mice expressing mutant human amyloid β-protein precursor and presenilin 1 (APP/PS1) to model cerebral amyloidosis. At 12 months of age, we orally administered EGCG and/or FA (30 mg/kg each) or vehicle once daily for 3 months. At 15 months, combined EGCG–FA treatment reversed cognitive impairment in most tests of learning and memory, including novel object recognition and maze tasks. Moreover, EGCG- and FA-treated APP/PS1 mice exhibited amelioration of brain parenchymal and cerebral vascular β-amyloid deposits and decreased abundance of amyloid β-proteins compared with either EGCG or FA single treatment. Combined treatment elevated nonamyloidogenic soluble APP-α and α-secretase candidate and down-regulated amyloidogenic soluble APP-β, β-C-terminal APP fragment, and β-secretase protein expression, providing evidence for a shift toward nonamyloidogenic APP processing. Additional beneficial co-treatment effects included amelioration of neuroinflammation, oxidative stress, and synaptotoxicity. Our findings offer preclinical evidence that combined treatment with EGCG and FA is a promising AD therapeutic approach.

Despite intense investigation, a synthetic drug that is both safe and effective has not yet been developed for AD. This inconvenient truth has led to the consideration of alternative strategies, including naturally occurring dietary compounds with therapeutic potential (so-called "nutraceuticals"). Another important matter is a recent mandate to develop combination therapy for AD. Because nutraceuticals are generally safe and well-tolerated, these agents are more amenable to combination than designer drugs. In this report, we explore combination therapy with two promising nutraceuticals that have complementary anti-amyloidogenic properties: (Ϫ)-epigallocatechin-3-gallate (EGCG, an ␣-secretase enhancer) (8,9) and ferulic acid (FA, a ␤-secretase modulator) (10).
EGCG is a polyphenol catechin (the ester of epigallocatechin and gallic acid) (11) that is found in high abundance in tea leaves, including green tea, and in carob flour (i.e. Ceratonia siliqua) at lesser abundance. Trace amounts are found in fruits (e.g. apple and cranberries) and nuts (e.g. pecans and hazelnuts) (12) (http://www.ars.usda.gov/nutrientdata). 4 EGCG has garnered attention as a medicinal agent to improve cognition, reduce inflammation, and even to treat cancer. Beneficial effects are often attributed to anti-oxidant, metal-chelating, anti-inflammatory, anti-carcinogenic, and anti-apoptotic properties (13). Substantial quantities pass from the small to the large intestine, where EGCG undergoes degradation by gut flora and absorption. EGCG can be reabsorbed from the intestine through enterohepatic re-circulation (14,15). It has been shown that EGCG crosses the bloodbrain barrier after systemic administration (16), and we were the first to show that EGCG enhances APP cleavage to sAPP-␣ and reduces A␤ abundance both in neuron-like cells and in Tg2576 mouse brains (8). A disintegrin and metalloproteinase domaincontaining protein 10 (ADAM10) activation is mechanistically responsible for EGCG promotion of nonamyloidogenic ␣-secretase APP cleavage (9).
Like EGCG, FA is also a plant-derived compound. Seed plants and leaves (e.g. rice, wheat, and oats), vegetables (e.g. tomatoes and carrots), and fruits (e.g. pineapples and oranges) are the main sources of dietary FA. The compound has both anti-inflammatory and anti-oxidant properties (10,17,18), and we have previously shown that FA is a ␤-secretase modulator that inhibits amyloidogenic APP cleavage. We further reported that FA reverses cognitive/behavioral deficits and mitigates AD-like pathology after 6 months of oral treatment in a transgenic mouse model of cerebral amyloidosis, and it alters amyloidogenic ␤-secretase APP cleavage in mutant APP-overexpressing neuron-like cells (10). Because FA has a relatively low molecular weight, the compound is freely cell-permeable and bioavailable. FA is absorbed in free form via stomach mucosa and is then transported into the hepatic portal vein where it is metabolized in the liver. Of note, oral FA can be recovered in rat plasma after only 5 min, when the ratio of free FA to total FA is markedly increased, but it rapidly gives rise to conjugated FA. Both free and conjugated FA are distributed via the systemic circulation into peripheral tissues (18 -20). Whereas FA is negatively charged at physiological pH due to a hydroxyl moiety (21), the compound has been shown to cross the blood-brain barrier in rodents after peripheral administration (22,23).
Given that both compounds share anti-inflammatory and anti-oxidant properties and have complementary modes of action on APP cleavage, we tested whether combination therapy with EGCG and FA (each at 30 mg/kg) ameliorated learning and memory changes, cerebral amyloidosis, AD-like pathology, and amyloidogenic APP processing in the mutant human APP and presenilin 1 (APP/PS1) transgenic mouse model of cerebral amyloidosis (APP/PS1 mice). We orally administered EGCG/FA (alone or in combination) or vehicle to APP/PS1 mice once daily for 3 months (beginning at 12 months of age) and evaluated animals at 15 months old. Here, our preclinical results show that combination therapy with EGCG plus FA has significant advantages over single treatment with either compound.

Co-treatment with EGCG and FA reverses learning and memory impairment in APP/PS1 mice
APP/PS1 mice reportedly have transgene-associated behavioral impairment as early as 5-7 months of age (24), so we began by assessing baseline cognitive status prior to dosing at 12 months. Data revealed behavioral impairment in novel object recognition, Y-maze, and radial arm water maze (RAWM) tests in our cohort of 12-month-old APP/PS1 mice versus wildtype (WT) littermate controls (data not shown). We then randomly assigned behaviorally impaired APP/PS1 mice (four treatments with n ϭ 8 per group; equal numbers of four males and four females) and WT littermates (same group sizes and distribution as APP/PS1 mice). Single or double treatment (via gavage using a 6202 gastric tube with a rounded silicone-coated tip, Fuchigami Kikai, Kyoto, Japan) was given once daily for 3 months with EGCG and/or FA (each at 30 mg/kg) or vehicle beginning at 12 months in APP/PS1 and in WT littermate controls. This directly administered treatment more precisely delivers the targeted amount of agent compared with ad libitum access to drinking water or chow. At the end of treatment (15 months of age), we conducted a behavioral testing battery.
We initially assessed episodic memory by novel object recognition test, and all eight mouse groups showed similar recognition indices (50.3-52.5%) in the training phase of the test (Fig.  1A, left; Table S1). In the retention phase, one-way analysis of variance (ANOVA) followed by post hoc comparison disclosed statistically significant differences on recognition index between APP/PS1-V mice and the other seven mouse groups (Fig. 1A, right; Table S1; *, p Ͻ 0.05 for APP/PS1-V versus all other mouse groups). Singly-or doubly-treated APP/PS1 mice had significantly increased novel object exploration frequency by 60.8 -71.6% versus APP/PS1-V mice (50.0%). Of note, EGCG/FA combination therapy completely mitigated episodic memory impairment (Fig. 1A, right; Table S1; †, p Ͻ 0.05, APP/ PS1-EGCG/FA versus APP/PS1-EGCG or APP/PS1-FA mice), as there were no significant differences versus any of the WT mouse groups (Fig. 1A, right; Table S1; p Ͼ 0.05).
We moved on to test exploratory activity and spatial working memory in the alternation Y-maze task. One-way ANOVA fol-lowed by post hoc testing revealed statistically significant differences for total arm entries between APP/PS1-V mice and the other seven mouse groups (Fig. 1B, left; Table S2; *, p Ͻ 0.05). This transgene-related behavioral phenotype has been noted in this and other mouse models of cerebral amyloidosis (10, 24 -28), and it may mirror disinhibition caused by cortical and/or hippocampal damage (29). Of interest, APP/PS1 transgene-associated hyperactivity, often taken as anxiety-like behavior in APP/PS1 mice, was completely reversed by combined treatment with EGCG plus FA (Fig. 1B, left; Table S2; †, p Ͻ 0.05, APP/PS1-EGCG/FA versus APP/PS1-EGCG or APP/PS1-FA mice). Combination treatment completely stabilized hyperactivity, as the co-treatment group did not significantly differ from any of the WT mouse groups (Fig. 1B, left; Table S2; p Ͼ 0.05).
Mice instinctively alternate arms in the Y-maze, such that they enter the three arms in sequence more often than by  Combination therapy modifies Alzheimer-like pathology chance alone (50%, see dotted line in Fig. 1B, right; Table S2); this behavioral phenotype is generally interpreted as a measure of spatial working memory (24). As expected, APP/PS1-V mouse behavior revealed less tendency to alternate versus WT controls. One-way ANOVA followed by post hoc testing showed statistically significant differences on Y-maze spontaneous alternation between APP/PS1-V mice and the other seven mouse groups (Fig. 1B, Table S2; p Ͼ 0.05). Therefore, combination treatment fully restored defective spatial working memory in the alternation Y-maze. Finally, we assessed hippocampus-dependent spatial reference learning and memory with the RAWM test. On day 1, overall ANOVA disclosed the main effects of block (p Ͻ 0.001 for both errors and escape latency), genotype (p Ͻ 0.05 for errors and p Ͻ 0.01 for escape latency), and treatment (p Ͻ 0.05 for escape latency). APP/PS1-V mice tended toward increased number of errors and longer escape latencies to reach the visible or hidden platform locations compared with the other seven mouse groups. On day 2, overall ANOVA revealed the main effects of block (p Ͻ 0.001 for both errors and escape latency), genotype (p Ͻ 0.05 for errors and p Ͻ 0.01 for escape latency), and treatment (p Ͻ 0.001 for both errors and escape latency). Repeated-measures ANOVA followed by post hoc assessment disclosed statistically significant differences between APP/PS1-V mice and the other seven mouse groups ( Fig. 1, C and D, left; Tables S3 and S5; *, p Ͻ 0.05 for both errors and escape latency). APP/PS1-V mice had more errors and greater escape latencies to reach the visible or hidden platforms compared with the other seven mouse groups; yet, either singly-or doubly-treated APP/PS1 mice accomplished the task with significantly less errors and shorter latencies, and their behavior did not significantly differ from the four WT mouse groups ( Fig. 1, C and D, left and right; Tables S3-S6; p Ͼ 0.05). There were no significant between-group differences (p Ͼ 0.05) on swim speed, nor did we find thigmotaxis (characteristic of extreme anxiety-like behavior) in any of the mice tested. Thus, behavioral differences in the RAWM test were not due to motivational issues, locomotor impairment, or anxiety. These results show that the 3-month treatment with EGCG and/or FA completely remediates spatial reference learning and memory impairment associated with the APP and PS1 transgenes. For all behavioral tests, sex was included as a categorical covariate in multiple ANOVA models, but we did not find significant main effects or interactive terms with sex (p Ͼ 0.05). Furthermore, we stratified all analyses by sex and did not find any significant differences (p Ͼ 0.05; data not shown).
Importantly, we verified histological findings with biochemical analysis of A␤ species in brain homogenates using sandwich enzyme-linked immunosorbent assay (ELISA). In the TBS-soluble fraction, single-or double-treatment with EGCG/FA revealed significant reductions in A␤

Tandem modulation of APP cleavage by combination therapy
Reduced cerebral amyloidosis in APP/PS1 mice after EGCG/FA treatment could be due to the following: 1) increased brain-to-blood A␤ efflux (6); 2) reduced expression of APP or PS1 transgenes; or 3) altered APP cleavage. We obtained peripheral blood samples from each APP/PS1 mouse group at the time of sacrifice and assayed plasma A␤(1-40) and A␤(1-42) species, but we did not detect between-group differences (data not shown). We then moved on to assess expression of transgene-derived APP or PS1 by Western blotting in brain homogenates. Results showed comparable band densities; therefore, neither single nor dual treatment with EGCG/FA affected transgene-derived APP (Fig. 5A) or PS1 (data not shown) expression.

Combination therapy modifies Alzheimer-like pathology
further reduced sAPP-␤ as well as C99 and P-C99 band densities compared with either treatment alone (Fig. 5, A, C, and D; Table S11; † †, p Ͻ 0.01; † † †, p Ͻ 0.001). Moreover, the 4-kDa monomeric A␤ band (Fig. 5A) and A␤ species between 25 and 75 kDa (putative A␤ oligomers, Fig. 5A) were also reduced in brain homogenates from three APP/PS1 mouse treatment groups. This was further supported by 82E1 sandwich ELISA for A␤ oligomers ( Fig. 5E; **, p Ͻ 0.01 for APP/PS1-V versus treatment groups). It is noteworthy that combined treatment posted additional reductions across most measures of amyloidogenicity compared with EGCG or FA alone ( Fig. 5E; † †, p Ͻ 0.01).
The above data provided evidence that combination therapy promoted nonamyloidogenic and inhibited amyloidogenic APP cleavage. To understand the molecular underpinnings of these findings, we examined ADAM10 (␣-secretase candidate)

Combination therapy alleviates neuroinflammation, oxidative stress, and synaptotoxicity in APP/PS1 mice
In addition to anti-amyloidogenic properties, EGCG and FA have anti-inflammatory and anti-oxidant properties (10,11,13,17,18). Therefore, we moved on to investigate putative effects of each compound on neuroinflammatory processes and oxidative stress in APP/PS1 mice. We first examined ␤-amyloid deposit-associated gliosis (i.e. astrocytosis and microgliosis) by quantitative immunohistochemical analyses alongside brain mRNA expression of the cardinal proinflammatory innate immune cytokines, tumor necrosis factor-␣ (TNF-␣) and interleukin-1␤ (IL-1␤), by quantitative real-time PCR (QPCR). We also assayed brain mRNA expression of two key oxidative stress markers: superoxide dismutase 1 (SOD1) and GSH peroxidase 1 (GPx1). APP/PS1-V mouse brains disclosed hyperplasia and hypertrophy of ␤-amyloid plaque-associated reactive astrocytes and microglia. We observed staining for the astrocytic activation marker, glial fibrillary acidic protein (GFAP), and the structural marker of activated microglia, ionized calciumbinding adapter molecule 1 (Iba1), in glial somata and processes. Numerous minute GFAP-positive astrocytic processes were distributed between neurons, and GFAP expression was intensely co-localized with dystrophic neurites associated with ␤-amyloid deposits (Fig. 7, A-D). All three mouse treatment groups had decreased reactive astrocytes and microglia, and

Combination therapy modifies Alzheimer-like pathology
bination therapy further significantly reduced brain proinflammatory cytokine expression compared with single treatment ( Fig. 9A; Table S13; †, p Ͻ 0.05); brain expression of proinflammatory cytokines was completely blocked by co-treatment and did not significantly differ from any of the WT control groups ( Fig. 9A; Table S13; p Ͼ 0.05). We noted a similar pattern of results for expression of key oxidative stress markers. APP/PS1-V brains had greater mRNA abundance of both SOD1 and GPx1 transcripts, and both were significantly reduced in the three treatment groups (Fig. 9B; Table S13; *, p Ͻ 0.05; **, p Ͻ 0.01). Combination therapy further significantly reduced brain expression of key oxidative stress markers compared with single treatment (Fig. 9B; Table  S13; †, p Ͻ 0.05). Similar to proinflammatory cytokine mRNAs, combined treatment completely inhibited SOD1 and GPx1 mRNAs, which did not significantly differ from the three WT mouse groups (Fig. 9B; Table S13; p Ͼ 0.05). These effects were sex-independent (data not shown). Western blottings on brain homogenates supported the results presented above. In each treatment group, both SOD1 and GPx1 proteins were significantly reduced versus APP/PS1-V mice by densitometry (Fig. 9, C-E; Table S14; ***, p Ͻ 0.001). Strikingly, combination therapy exhibited further reduction compared with EGCG or FA treatment alone (Fig. 9, C-E; Table S14; † † †, p Ͻ 0.001). Together, these results demonstrate that EGCG plus FA combination therapy stabilizes neuroinflammation and oxidative stress in APP/PS1 mouse brains.

Discussion
A key drug development challenge is delivering an agent with few side effects that not only treats symptoms but actually modifies disease. Here, we report that combination therapy

Combination therapy modifies Alzheimer-like pathology
with EGCG and FA completely reverses transgene-associated behavioral deficits in the APP/PS1 transgenic mouse model (see Fig. 1). In concert with learning and memory improvement, pathological findings demonstrate that combined therapy posts additional benefits over single treatments to attenuate parenchymal and vascular ␤-amyloidosis (see Figs. 2-4). In this regard, clinicopathological studies have investigated the relationship between cerebral amyloid pathology and cognitive deficits in human AD (31,32). Similar to our results, numerous reports have linked reversal of transgene-associated behavioral deficits to attenuated cerebral amyloid pathology in mouse models (10, 26 -28, 33-36), although others have shown that cognitive function and cerebral amyloidosis can be uncoupled (37,38). In humans, the elderly with preserved cognitive function reportedly have cerebral amyloid pathology comparable with prodromal or frank AD at autopsy (39, 40). One explanation provided for the discordancy between senile plaques and cognitive impairment in the human AD literature is that soluble, oligomeric forms of A␤ are the true neurotoxic species (41)(42)(43)(44). In support of this hypothesis, others have demonstrated that behavioral deficit occurs in parallel with elevated A␤ oligomers in rodent models (34,38). Interestingly, we show that combination treatment significantly decreases A␤ oligomers (see Fig. 5), which inversely correlates with cognitive function (see Fig. 1). Sequential APP endoproteolysis by the secretases generates amyloidogenic C99 and sAPP-␤ followed by A␤, whereas the nonamyloidogenic pathway generates sAPP-␣ and C83. Our data show significantly increased sAPP-␣ in brain homogenates from APP/PS1 mice treated with EGCG alone or in combination with FA, whereas sAPP-␤, C99, and P-C99 production are reduced, providing evidence that APP cleavage is shifted toward the nonamyloidogenic pathway. Soluble APP-␣ has beneficial properties, including neurotrophism and neuropro-

Combination therapy modifies Alzheimer-like pathology
tection (45,46), and even enhancing long-term potentiation (47). Interestingly, treatment with EGCG alone not only increases nonamyloidogenic processing but also reduces amyloidogenic APP processing (see Fig. 5). Combined treatment significantly enhances protein expression of ␣-secretase candidate ADAM10, whereas BACE1 protein abundance is significantly reduced, even by EGCG treatment alone. This latter result is interesting, because a previous study showed that EGCG can act as a ␤-secretase modulator in a cell-free system (48). Nonetheless, combination therapy further significantly decreases BACE1 protein expression compared with EGCG alone (see Fig. 6). This agrees with our prior report that sAPP-␣ can indirectly modulate ␤-secretase activity and A␤ generation through a negative feedback loop (49), and others have shown that sAPP-␣ can autoregulate APP cleavage (50 -52). Hence, EGCG can inhibit A␤ generation both by directly enhancing ␣-secretase and by indirectly modulating ␤-secretase activity. Collectively, our data show that combination therapy further polarizes APP cleavage toward nonamyloidogenic ␣-secretase cleavage versus either single treatment.
In the AD brain, reactive microglia and astrocytes co-exist in both temporal and spatial proximity to ␤-amyloid deposits, and glial activation leads to neuroinflammation that can cause bystander neuronal injury (53)(54)(55). Both EGCG and FA have additional beneficial properties, including reducing inflammation (10,13,18). Strikingly, in fact, combination therapy confers the greatest additional benefit over single treatments on neuroinflammation. When considering neuroinflammatory biomarkers, results show that combined treatment dampens plaque-associated microgliosis and astrocytosis and completely blocks brain mRNA expression of TNF-␣ and IL-1␤ to baseline levels (see Figs. [7][8][9]. Given the recent intense focus on inflammatory mechanisms in AD (54,55), it remains possible that mitigated neuroinflammation after combination therapy plays a dominant role to reduce AD-like pathology and improve cognition in APP/PS1 mice.
Reactive oxygen species are produced during host defense against pathogens. However, when activated inappropriately or overproduced, oxidative substances can damage DNA, proteins, and lipids, leading to cellular injury and death (56). Oxi-

Combination therapy modifies Alzheimer-like pathology
dative damage has been linked to AD pathogenesis (57,58), and both EGCG and FA have anti-oxidant and free radical scavenging properties (10,11,13,17). Therefore, we explored whether the compounds modulated oxidative stress in APP/PS1 mice. Results show that combined treatment with EGCG and FA completely reduces cerebral mRNA expression of SOD1 and GPx1 to baseline levels in APP/PS1 mice, whereas EGCG or FA single treatment has only modest effects. We actually observe the greatest reductions in SOD1 or GPx1 proteins with combination therapy (see Fig. 9). Synaptic loss is a cardinal feature of the AD brain, and the role of A␤ in synapse loss and dysfunction is well-documented. Importantly, synaptic plasticity is considered to form the cellular basis for learning and memory (59 -61). Dysregulated plasticity leads to synaptic loss, and decreased synapse distribution is a durable predictor of disease progression and behavioral decline (59). Synaptophysin is a pre-synaptic marker that is most commonly used as a surrogate for synaptic density (62). In this regard, we quantified synaptophysin IR in hippocampal CA1 and EC regions of APP/PS1 mice and found significant increases in each APP/PS1 treatment group, with the greatest benefit after combined treatment (see Fig. 10).
Even more than with single treatment approaches, safety is a key issue with combination therapy, and adverse effects must be taken into account. Importantly, single or dual treatment with EGCG and FA for 3 months is well-tolerated, and we did not observe any adverse events in any of the mice included in our study. Based on the chemical database for EGCG by Registry of Toxic Effects of Chemical Substances, the acute oral lethal dose for 50% survival (LD 50 ) is as high as 2,710 mg/kg in QPCRs are shown for tumor necrosis factor-␣ (TNF-␣) or interleukin-1␤ (IL-1␤) proinflammatory cytokines or for key oxidative stress markers superoxide dismutase 1 (SOD1) or GSH peroxidase 1 (GPx1). Data for A and B are expressed as relative fold over WT-V mice. C, Western blots are shown using anti-Cu/Zn SOD polyclonal antibody (pAb SOD1) or by anti-GPx1 polyclonal antibody (pAb GPx1). Western blots included each mouse (n ϭ 8 per group), and quantitative data were averaged. Ten g of protein from each sample was equally loaded per lane. Actin is included as a loading control for each appropriate blot, and densitometry data are shown below each lane. Densitometry data are shown for ratios of SOD1 to actin (D) or for GPx1 to actin (E). Statistical comparisons for A and B are between-groups but within each mRNA species. Statistical comparisons for D and E are within each protein but between-groups. *, p Ͻ 0.05; **, p Ͻ 0.01; ***, p Ͻ 0.001 for APP/PS1-V versus the other treated mice; †, p Ͻ 0.05; † † †, p Ͻ 0.001 for APP/PS1-EGCG/FA versus APP/PS1-EGCG or APP/PS1-FA mice. V, vehicle.

Combination therapy modifies Alzheimer-like pathology
mice. Likewise, FA has a high acute oral LD 50 of 2,370 mg/kg in mice (10), with a human equivalent dose of 2.43 mg/kg for EGCG or FA, which equates to a 146 mg dose for a 60 kg person (63). Moreover, the tolerable daily intake in a human can be extrapolated from rodent LD 50 threshold data (64). Assuming the body surface area calculation factor for interspecies variation (63), the tolerable human daily intake for EGCG is equivalent to 13.18 g, whereas FA is 11.53 g (both assume 60 kg human weight). Thus, the doses (30 mg/kg/day for a mouse: 146 mg for a human) used in this study are quite low, and such advantageous pharmacotoxicity profiles reinforce the notion that combined EGCG/FA administration is safe. Nonetheless, it remains possible that increased ADAM10 and/or decreased BACE1 activity might also affect processing of other substrates, leading to unwanted side effects. Hence, further translational research is warranted to investigate whether long-term EGCG and FA therapy is safe in humans. With respect to the pharmacokinetics of EGCG in humans, the compound is absorbed at Ͻ5% abundance and reaches peak plasma concentration time (t max ) at 2 h after per os administration (65,66). Green tea catechins are primarily absorbed in the intestine (i.e. jejunum and ileum) by paracellular diffusion through epithelial cells (67). Once absorbed, free EGCG is mainly detected in plasma (Ͼ75%) (14,68) and has a half-life (t1 ⁄ 2 ) of ϳ3 h (65, 66). Similar to EGCG, FA is absorbed from the stomach, jejunum, and ileum. Free FA can be found in human plasma just 10 min after oral administration of sodium ferulate (69). Plasma FA concentrations reach t max at 24 min following oral administration, with a t1 ⁄ 2 of 42 min (69). Free FA and its glucuronic conjugate are detected in the plasma, and both free FA and its glycine conjugate are found in urine after ingestion of FA containing wheat bran (70). The urinary excretion of FA in humans plateaus between 7 and 9 h after ingestion (71).
An important question is whether EGCG and FA can cross the blood-brain barrier. Using LC coupled with tandem MS electrospray ionization, EGCG was detected in the rodent brain after systemic administration (16). With respect to FA, standard HPLC approaches have detected the compound in the brain after peripheral administration (22,23). This result has been validated using a more advanced technique: ultra-performance LC coupled to tandem MS (72).
The vast majority (Ͼ 95%) of AD cases are "sporadic" and are characterized by onset at 65 years or later, whereas less than 5% of cases are early-onset autosomal dominant AD. To date, a series of mutations in three different genes (i.e. APP, PS1, and PS2) cause autosomal dominant AD (73). These mutations have been used with varying success to study basic and translational aspects of AD. One of the most widely used tools is transgenic mice expressing one or more of these mutant human transgenes. In this study, we utilized APP/PS1 mice harboring human APP and PS1 mutations. It is important to note that, although APP/PS1 mice demonstrate some of the AD-relevant

Combination therapy modifies Alzheimer-like pathology
phenotypes, including cerebral amyloidosis, neuroinflammation, and cognitive disturbance, they do not recapitulate all AD features and are therefore not a complete model of human AD.
In conclusion, we report that administering 3-month combination therapy with EGCG and FA to the aged APP/PS1 transgenic mouse model of cerebral amyloidosis confers additional benefits over single treatments on improving behavioral deficits, ameliorating cerebral amyloidosis, and reducing A␤ generation. Furthermore, combined treatment potently stabilizes neuroinflammation, alleviates oxidative stress, and mitigates synaptotoxicity. We propose long-term combination therapy with EGCG and FA as an effective disease-modifying therapy for AD.

Ethics statement
All experiments were performed in accordance with the guidelines of the National Institutes of Health, and all animal studies were approved by the Saitama Medical University Institutional Animal Care and Use Committee. Animals were humanely cared for during all experiments, and all efforts were made to minimize suffering.
EGCG and FA were obtained from Sigma. Thirty mg of each reagent was resuspended in 25 l of 100% DMSO and then diluted with distilled water to a final concentration of 0.2% DMSO. All reagents were freshly prepared daily prior to each treatment. We randomly assigned APP/PS1 mice to four treatment groups (n ϭ 8 per condition; four males and four females): EGCG (APP/PS1-EGCG); FA (APP/PS1-FA); EGCG ϩ FA (APP/PS1-EGCG/FA); or vehicle (distilled water containing DMSO at a final concentration of 0.2%; APP/PS1-V). Additionally, WT littermates received the same four treatments (n ϭ 8 per group; four males and four females) as follows: WT-EGCG; WT-FA; WT-EGCG/FA mice; or vehicle (WT-V). To determine whether treatment prevented versus reversed kinetics of cerebral amyloid accumulation, we included eight untreated APP/PS1 mice at 12 months of age (APP/PS1-12M mice, four males and four females) for analysis of ␤-amyloid pathology. After baseline behavioral assessment just before dosing (at 12 months of age), animals were gavaged with EGCG, FA, EGCG plus FA (all at 30 mg/kg), or vehicle once daily for 3 months. Mice were housed in a specific pathogen-free barrier facility under a 12/12-h light/dark cycle with ad libitum access to food and water.

Behavioral analyses
To assess novel object recognition and retention, animals were habituated in a cage for 4 h, and two objects of different shapes were concurrently provided for 10 min. The number of times that the animal explored the familiar object (defined as number of instances where an animal directed its nose 2 cm or less from the object) was counted for the initial 5 min of exposure (training phase). To test memory retention on the following day, one of the familiar objects was replaced with a different shaped novel object, and explorations were recorded for 5 min (retention test). The recognition index, taken as a measurement of episodic memory, is reported as frequency (%) of explorations of the novel versus familiar objects (75).
To measure exploratory activity and spatial working memory, animals were individually placed in one arm of a radially symmetric Y-maze made of opaque gray acrylic (arms, 40 cm long and 4 cm wide; walls, 30 cm tall), and the sequence of arm entries and total number of entries were counted over a period of 8 min, beginning when the animal first entered the central area. Percent alternation was defined as entries into sequentially different arms on consecutive occasions using the following formula: % alternation ϭ number of alternations/(number of total arm entries Ϫ 2) ϫ 100% (24).
To assess spatial reference learning and memory, the RAWM test was conducted over 2 days and consisted of triangular wedges in a circular pool (80 cm diameter) configured to form swim lanes that enclosed a central open space. Mice naïve to the task were placed in the pool and allowed to search for the platform for 60 s. Animals were dropped into a random start arm and allowed to swim until they located and climbed onto the platform (goal) over a period of 60 s. Latency to locate the platform and errors were recorded. Each mouse was given a total of 15 trials. On day 1, the goal alternated between visible and hidden, although on day 2, the goal was always hidden. All data were organized as individual blocks of three trials each. The goal arms remained in the same location for both days, whereas the start arm was randomly altered. All trials were done at the same time of day (Ϯ1 h), during the animals' light phase. To avoid interference with behavioral testing, each treatment was carried out 1 h after concluding behavioral testing (76).

Brain tissue preparation
At 15 months of age, animals were anesthetized with sodium pentobarbital (50 mg/kg) and euthanized by transcardial perfusion with ice-cold physiological saline containing heparin (10 units/ml). Brains were isolated and quartered (sagittally at the level of the longitudinal fissure of the cerebrum and then coronally at the level of the anterior commissure). Left anterior hemispheres were weighed and snap-frozen at Ϫ80°C for Western blotting. Right anterior hemispheres were weighed and immersed in RNA stabilization solution (RNAlater, Applied Biosystems, Foster City, CA) and then snap-frozen at Ϫ80°C for QPCR analysis. Left posterior hemispheres were immersed in 4% paraformaldehyde at 4°C overnight and routinely processed in paraffin. Right posterior hemispheres were weighed and snap-frozen at Ϫ80°C for ELISA.

Image analysis
Images were acquired and quantified using SimplePCI software (Hamamatsu Photonics, Shizuoka, Japan). Images of five 5-m sections through each anatomic ROI (i.e. RSC, EC, and H) were captured based on anatomical criteria (77), and we set a threshold optical density that discriminated staining from background. Selection bias was controlled for by analyzing each ROI in its entirety. For A␤, GFAP, and Iba1 burden analyses, data were reported as the percentage of positive pixels captured divided by the full area captured. For conventional A␤ burden analysis, the A␤(17-24) mAb was used, and its epitope lies within amino acids [18][19][20][21][22].
For ␤-amyloid plaque morphometric analysis, diameters (maximum lengths) of ␤-amyloid plaques were measured, and three mutually exclusive plaque size categories (Ͻ25, 25-50, or Ͼ50 m) were blindly tabulated. For quantitative analysis of CAA, numbers of A␤ antibody-positive cerebral vessels were blindly counted in each ROI. To evaluate synaptophysin IR, images of five 5-m sections through hippocampal CA1 and EC were blindly captured based on anatomical criteria (77) and converted to grayscale. The average optical density of positive signals from each image was quantified as a relative number from zero (white) to 255 (black) and expressed as mean intensity of synaptophysin IR.

ELISA
Brain A␤(1-40) and A␤(1-42) species were detected by a three-step extraction protocol with modifications (78). Brains were homogenized using TissueLyser LT (Qiagen, Valencia, CA) in Tris-buffered saline (TBS: 25 mM Tris-HCl, pH 7.4, 150 mM NaCl) containing protease inhibitor mixture (Sigma) and centrifuged at 18,800 ϫ g for 60 min at 4°C, and supernatants were collected (representing the TBS-soluble fraction). Resulting pellets were treated with TNE buffer (10 mM Tris-HCl, 1% Nonidet P-40, 1 mM EDTA, and 150 mM NaCl) with protease inhibitors and homogenized using TissueLyser LT. Homogenates were then centrifuged at 18,800 ϫ g for 30 min at 4°C, and supernatants were collected (representing the detergent-soluble fraction). Remaining pellets were treated with 5 M guanidine HCl and dissolved by occasional mixing on ice for 30 min and centrifuged at 18,800 ϫ g for 30 min at 4°C. Supernatants were then collected; this is taken as the guanidine HCl-soluble fraction. A␤  and A␤(1-42) species were separately quantified in each sample in duplicate by sandwich ELISA (IBL, Gunma, Japan) (79). A␤ oligomers were quantified in the TNEsoluble fraction in duplicate by human amyloid ␤ oligomers (82E1-specific) assay kit (IBL) (80). All samples fell within the linear range of the standard curve.

QPCR
We quantified TNF-␣, IL-1␤, SOD1, GPx1, and ␤-actin mRNA levels by QPCR analyses (10,26,27). Total RNA was extracted using the RNeasy mini kit (Qiagen), and first-strand cDNA synthesis was carried out using the QuantiTect reverse transcription kit (Qiagen) in accordance with the manufacturer's instructions. We diluted cDNA 1:1 in H 2 O and carried out QPCR for all genes of interest using cDNA-specific TaqMan primer/probe sets (TaqMan Gene Expression Assays, Applied Biosystems) on an ABI 7500 Fast real-time PCR instrument (Applied Biosystems). Each 20-l reaction mixture contained 2 l of cDNA with 1 l of TaqMan Gene Expression Assay reagent, 10 l of TaqMan Fast Universal PCR Master Mix Combination therapy modifies Alzheimer-like pathology (Applied Biosystems), and 7 l of H 2 O. Thermocycler conditions consisted of 95°C for 15 s, followed by 40 cycles of 95°C for 1 s and 60°C for 20 s. Mouse TaqMan probe/primer sets were as follows: TNF-␣ (number Mm00443258_m1); IL-1␤ (number Mm00434228_m1); SOD1 (number Mm01700393_ g1); GPx1 (number Mm00656767_g1); and ␤-actin (number Mm00607939_s1; used as an internal reference control) (Applied Biosystems). Samples that were not subjected to reverse transcription were run in parallel as negative controls to rule out genomic DNA contamination, and a "no template control" was also included for each primer set. The cycle threshold number (C T ) method (81) was used to determine relative amounts of initial target cDNA in each sample. Results were expressed relative to vehicle control WT mice.

Statistical analysis
Data are presented as means with associated standard deviations. A hierarchical analysis strategy was used in which the first step was an overall ANOVA (repeated measures was used for behavioral data) to assess significance of main effects and interactive terms. If significant, post hoc testing was done with Tukey's HSD or Dunnett's T3 methods, where appropriateness was determined based on Levene's test for equality of the variance. In instances of multiple mean comparisons, one-way ANOVA was used, followed by post hoc comparison of the means using Bonferroni's or Dunnett's T3 methods (depending on Levene's test for equality of the variance). The ␣ levels were set at 0.05 for all analyses. All analyses were performed using the Statistical Package for the Social Sciences, release 23.0 (IBM SPSS, Armonk, NY).