PS1 transgenic mice

Hormones and Behavior 118 (2020) 104640

Contents lists available at ScienceDirect

Hormones and Behavior journal homepage: www.elsevier.com/locate/yhbeh

A dual GLP-1 and Gcg receptor agonist rescues spatial memory and synaptic plasticity in APP/PS1 transgenic mice

T

Zhao-Jun Wanga,1, Yu-Fei Hanb,1, Fang Zhaoa, Guang-Zhao Yangc, Li Yuand, Hong-Yan Caie, ⁎ ⁎ Jun-Ting Yanga, Christian Holscherf, Jin-Shun Qia, , Mei-Na Wua, a

Department of Physiology, Key Laboratory of Cellular Physiology, Ministry of Education, Shanxi Medical University, Taiyuan, PR China Guangzhou Kingmed Diagnostics, Guangzhou, PR China c Department of Cardiovascular Medicine, The First Hospital of Shanxi Medical University, Taiyuan, PR China d Department of Physiology, Changzhi Medical College, Changzhi, PR China e Department of Microbiology and Immunology, Shanxi Medical University, Taiyuan, PR China f Neuroscience research group, Henan university of Chinese medicine, Zhengzhou, PR China b

A R T I C LE I N FO

A B S T R A C T

Keywords: (D-Ser2) oxyntomodulin GLP-l receptor Glucagon receptor APP/PS1 transgenic mice Learning and memory Long-term potentiation PI3K/AKT/GSK3β

Alzheimer's disease (AD) is a neurodegenerative disease that severely affects the health and lifespan of the elderly worldwide. Recently, the correlation between AD and type 2 diabetes mellitus (T2DM) has received intensive attention, and a promising new anti-AD strategy is the use of anti-diabetic drugs. Oxyntomodulin (Oxm) is a peptide hormone and growth factor that acts on neurons in the hypothalamus. OXM activates glucagon-like peptide 1 (GLP-1) and glucagon (Gcg) receptors, facilitates insulin signaling and has neuroprotective effects against Aβ1–42-induced cytotoxicity in primary hippocampal neurons. Here, we tested the effects of the protease-resistant analogue (D-Ser2)Oxm on spatial memory and synaptic plasticity and the underlying molecular mechanisms in the APP/PS1 transgenic mouse model of AD. The results showed that (D-Ser2)Oxm not only alleviated the impairments of working memory and long-term spatial memory, but also reduced the number of Aβ plaques in the hippocampus, and reversed the suppression of hippocampal synaptic long-term potentiation (LTP). Moreover, (D-Ser2)Oxm administration significantly increased p-PI3K/p-AKT1 expression and decreased p-GSK3β levels in the hippocampus. These results are the first to show an in vivo neuroprotective role of (D-Ser2) Oxm in APP/PS1 mice, and this role involves the improvement of synaptic plasticity, clearance of Aβ and normalization of PI3K/AKT/GSK3β cell signaling in the hippocampus. This study suggests that (D-Ser2)Oxm holds promise for the prevention and treatment of AD.

1. Introduction Alzheimer's disease (AD) is the most common cause of dementia, and its incidence increases with age. The main pathological hallmarks of AD are amyloid plaques consisting of extracellular amyloid β (Aβ) and neurofibrillary tangles containing hyperphosphorylated tau in neurons (Congdon and Sigurdsson, 2018; Onyango, 2018). Studies focused on clearing brain Aβ have not yielded an effective treatment for preventing the development of AD, and an alternative therapeutic strategy to improve brain function is essential. Type 2 diabetes mellitus (T2DM) is a major risk factor for AD (Craft, 2007; Hoyer, 1998; Luchsinger et al., 2004). Several epidemiological studies have found that 70–80% of patients with AD have T2DM or show blood glucose or insulin level abnormalities (Janson et al., 2004; Mwamburi and Qiu,

2016). Insulin signaling in the brains of people with AD was found to be severely de-sensitized (Moloney et al., 2010; Steen et al., 2005; Talbot et al., 2012). Nasal application of insulin showed some results in AD and MCI patients (Arnold et al., 2018; Freiherr et al., 2013). However, insulin signaling appears to be further de-sensitized by the treatment (Holscher, 2014b). Currently, another kind of anti-diabetes drug, Glucagon-like peptide-1 (GLP-1) analogues has received attention as a potential treatment for AD. Preclinical tests in animal models of AD show good protective effects and also re-sensitized insulin signaling in the brain (Bomfim et al., 2012; Cai et al., 2014; Hölscher, 2018; LongSmith et al., 2013; McClean and Holscher, 2014b). Encouraged by these results, we tested an analogue of oxyntomodulin (OXM) to analyse its neuroprotective effects. Oxm is a peptide hormone that has agonist properties for both GLP-1 receptors and glucagon (GCG) receptors

⁎

Corresponding authors at: Department of Physiology, Shanxi Medical University, Taiyuan, Shanxi 030001, PR China. E-mail addresses: [email protected] (J.-S. Qi), [email protected] (M.-N. Wu). 1 Contributed equally to this work. https://doi.org/10.1016/j.yhbeh.2019.104640 Received 20 April 2019; Received in revised form 16 November 2019; Accepted 16 November 2019 0018-506X/ © 2019 Elsevier Inc. All rights reserved.

Hormones and Behavior 118 (2020) 104640

Z.-J. Wang, et al.

(Kosinski et al., 2012). Importantly, these receptors are also expressed in many brain regions, including the hippocampus, cerebral cortex, and hypothalamus (Li et al., 2017). Notably, the dual receptor agonist properties of Oxm in treating diabetes are superior to those of a GLP-1 receptor-only agonist (Vrang and Larsen, 2010), which may be due to the concurrent activation of GCG receptors by OXM. Oxm is degraded by DPP-IV, which results in a short half-life in vivo (Wynne et al., 2005). A recent study showed that the DPP-IV resistant analogue (D-Ser2)OXM is a potent drug to treat diabetes and obesity (Pocai, 2012). Indeed, (DSer2)OXM causes appetite suppression and weight loss, but only reduces weight in people who are overweight (Ambery et al., 2018; Field et al., 2010; Shankar et al., 2018). Furthermore, (D-Ser2)OXM not only ameliorates glucose and insulin abnormalities in diabetic animals, but also inhibits the neurotoxicity of Aβ protein in primary cultured hippocampal neurons (Han et al., 2016). However, whether (D-Ser2)OXM can relieve the cognitive impairments and synaptic plasticity deficits in a transgenic model of AD mice has never been tested. It is well known that glycogen synthase kinase 3β (GSK3β) is involved in a variety of cellular processes ranging from glycogen metabolism, insulin signaling, neuronal function, and embryonic development (Amar et al., 2011). Considerable evidence suggests that GSK3β is involved in the common pathology underlying AD and T2DM (Gao et al., 2011). It has been proposed that GSK3β plays a leading role in the cascade of events that culminate in AD, such as the mechanisms underlying learning and memory, the hyperphosphorylation of tau, and the increased production of Aβ from APP. Our previous studies found that the activation of the PI3K/AKT pathway and the inhibition of GSK3β activity may be the molecular mechanisms by which GLP-1/ GCG receptor agonist protects cognitive behavior and LTP in AD mice (Cao et al., 2018; Cai et al., 2014). Upon insulin stimulation, the linear signaling cascade of IR (insulin receptor)/IRSs (insulin receptor substrates)/PI3K (phosphatidylinositol 3 kinase)/ AKT leads to phosphorylation of GSK3, thus GSK3 activity is inhibited. Therefore, the present study examined the neuroprotective role of (D-Ser2)Oxm in the APP/PS1 transgenic mouse model of AD, and then investigated its probable electrophysiological and molecular mechanisms by recording in vivo hippocampal LTP，and measuring Aβ deposition and PI3K/AKT/GSK3β expression levels to explore its potential as a new treatment of AD.

Fig. 1. Schematic diagram of animal treatment and experimental procedures. All mice were injected daily with (D-Ser2)OXM or saline (i.p.) for 2 weeks before behavioral tests, and the injections were kept during all behavioral tests, including Y maze and Morris water maze. Following the behavioral tests, half the mice were used for in vivo hippocampal long-term potentiation (LTP) recording and immunohistochemistry (IHC) and the other half for Western blot.

2.2. Body weight and blood glucose of mice Before each administration of drug, the mice were weighed to determine the injection dose. The blood glucose of mice was measured regularly on day 1, day 7, day 14 and day 21 during experiments. The mice were anesthetized by ether and fixed in a homemade casing, then 1 cm of the tail was cut off and the blood was collected from the caudal vein. Blood glucose was estimated using a glucose meter (OneTouch®, Johnson & Johnson, Shanghai, China).

2.3. Y maze test The spontaneous alternation of the animals was examined in a Y maze to examine the spatial working memory of mice (n = 10 in each group). The Y maze had three radial arms with equal angles (120°) between the arms. Each arm was 30 cm long, 15 cm high and 5 cm wide. A mouse was placed at the junction of the three arms and allowed free movement for 8 min in each session. The entries of the mouse into each arm were recorded, and every entry that differed from the last two entries was considered a successful alternation. The alternation percentage was calculated according to the formula: [(number of alternations) / (total number of arm entries - 2)] × 100(%) (Bak et al., 2017).

2. Materials and methods 2.1. Animals and drug treatments

2.4. Morris water maze task

Two-month-old male APPswe/PS1dE9 (APP/PS1) transgenic mice and wild-type (WT) C57BL/6 J control mice (20–30 g) were purchased from the Institute of Laboratory Animal Sciences (ILAS, China). The mice were housed with sufficient food and water in an animal room with an independent air supply system that was maintained on a 12/ 12 h light-dark cycle at 23 ± 2 °C and 55 ± 5% humidity. Subsequently, the mice were divided into four groups: WT + Saline, WT + (D-Ser2)Oxm, APP/PS1 + Saline and APP/PS1 + (D-Ser2)Oxm. After reaching 9 months of age, the mice began to receive saline (0.1 mL/kg body weight) or (D-Ser2) Oxm (25 nmol/kg body weight) daily by intraperitoneal (i.p.) injection for 2 weeks (Kerr et al., 2010). The daily administration was kept during the following 7 days of behavioral experiments (Fig. 1). After behavioral tests, electrophysiological recordings, immunohistochemistry and western blot were performed. (D-Ser2)Oxm was synthetized by SynPeptide Co. (Shanghai, China) with 97% purity as tested by HPLC and Mass-spec teechniques. All animal experimental procedures were approved by the Institutional Animal Care Committee of Shanxi Medical University and conformed to the guidelines of the National Institutes of Health (NIH) (NIH publication NO. 85-23, revised 1985).

After the Y maze tests, the classic Morris water maze (MWM) test was performed as described previously (Morris, 1984). A circular pool (diameter 120 cm, height 50 cm) was filled with 30 cm deep of tap water (maintained at 23 ± 2 °C), and titanium dioxide was added to ensure that the water was equally white throughout the pool. A white circular platform was hidden 1 cm below the water level. In the hidden platform tests, the mice (n = 12 in each group) were allowed to swim in the water to search for the underwater platform. The tests were performed four times per day for 5 consecutive days. In each trial, a mouse was placed in the water facing the pool wall in one of four equal quadrants designated by computer software. The order of mouse entry into the individual quadrants was randomized by a number table. Subsequently, 24 h after completing the hidden platform trials, each animal underwent a 60 s probe trial to evaluate the retention of the learned task. During the probe test, the platform was removed, and the searching behavior in the target quadrant (where the platform was located during the hidden training) was measured. After the probe test, the visual and motor ability of the mice were assessed by a visible platform test. EthoVision 3.0 software (Noldus Information Technology, Wageningen, Netherlands) was used to collect the movement data with respect to latency, swim path and speed. 2

Hormones and Behavior 118 (2020) 104640

Z.-J. Wang, et al.

Nonspecific sites were blocked with 5% BSA in Tris-buffered saline containing 0.05% Tween-20 (TBST), and the membrane was probed with primary antibody overnight at 4 °C, followed by secondary antibody at room temperature for 2 h. The following primary antibodies (Abcam) were used: anti-phospho-PI3K p85 antibody (1:1300), antiphospho-AKT1 monoclonal antibody (pS473, 1:600), and anti-phosphoGSK3β monoclonal antibody (pY216, 1:750). β-actin (1:3000) was used as a loading control. The corresponding secondary antibodies were antirabbit and anti-rat IgG HRP (1:3000, Boster). Finally, the immunocomplex was visualized using an enhanced chemiluminescence detection kit (Boster). The immunoreactive signals were scanned with a FluorChem Scanner and quantified using Alpha View SA software.

2.5. In vivo hippocampal LTP recording After behavioral tests, half of mice (n = 6 in each group) were randomly selected for electrophysiological experiment. Given the correlation between spatial memory and synaptic plasticity, we recorded in vivo hippocampal field excitatory postsynaptic potentials (fEPSPs). The mice were anesthetized with 5% chloral hydrate (0.07 mL/10 g, i.p.) and mounted onto a stereotaxic apparatus (RWD, Shenzhen, China). The body temperature was monitored and maintained at 37 ± 0.5 °C by a feedback-regulated heating pad (Temperature controller 69,000, RWD, China). A parallel bound stimulating/recording electrode was inserted into the hippocampal Schaffer-collateral/CA1 region (2.0 mm posterior to the bregma and 1.5 mm lateral to the midline for the tip of the recording electrode). The test stimuli were delivered to the Schaffercollateral/commissural pathway every 30 s, and a maximal field excitatory postsynaptic potential (fEPSP) was evoked in the CA1 region by increasing the intensity of the single stimulus gradually. To further observe the LTP of fEPSPs, a baseline fEPSP of 50% of the maximal fEPSP amplitude was selected by adjusting the pulse intensity. The baseline fEPSP was recorded for at least 30 min to ensure steady synaptic transmission. LTP was induced by a high-frequency stimulus (HFS) protocol with three trains of 20 pulses at 200 Hz at an interval of 30 s and an intensity which evoked 80% of the maximum response of fEPSP. Subsequently, fEPSPs were monitored for an additional 1 h to observe the induction and maintenance of LTP. An electronic stimulator (SEN-3301, Japan) and a coupled isolator (ss-102 J, Japan) were used for the pulse stimulation. The signals from the recording electrode were filtered at 1 kHz, amplified and displayed by a multichannel biological signal acquisition/processing system (Chengdu Instruments, China).

2.8. Statistical analysis Data are presented as the mean ± standard deviation (SD). Statistical analyses were conducted using SPSS 22.0 and SigmaPlot 12.0. The escape latency in the MWM task and the amplitude of fEPSP were analyzed using repeated measures analysis of variance (ANOVA). The staining data were analyzed by t-test, while the other data were analyzed by two-way ANOVA. Partial eta squares (ƞp2) are reported as a measure of effect size for main effects and interactions of ANOVAs. ttests was used to probe interactions, and Cohen's dare reported as a measure of effect size for t-tests (calculated with online calculator at www.socscistatistics.com/effectsize/default3.aspx). The statistical significance level was set at P < .05. 3. Results 3.1. (D-Ser2)Oxm prevents the deficits in the spatial working memory of APP/PS1 mice in the Y maze test

2.6. Immunohistochemistry After behavioral tests and in vivo hippocampal LTP recording, the anesthetized mice were perfuse-fixed transcardially with 37 °C warm phosphate-buffered saline (PBS, pH 7.4) and cold 4% paraformaldehyde (PFA, pH 7.4) for immunohistochemical analysis. For frozen section experiments, the left hemisphere of the brain was fixed in 4% PFA for 24 h and incubated in 30% sucrose at room temperature for an additional 24 h. Immediately, the tissues were embedded in OCT frozen embedding medium (Leica Inc., Germany). The coronal sections of the brain were sliced at 25 μm thicknesses using a chilled microtome (Leica, CM1850, Germany) and stored at 4 °C in PBS. Then, the sections were incubated with 5% hydrogen peroxide at room temperature for 15 min, blocked with 5% goat serum (Solarbio, China) for 30 min, and incubated with primary antibodies (anti-Aβ antibody 6E10, concentration 1:500, 803105, BioLegend, USA) overnight at 4 °C, followed by incubation with secondary antibody (peroxidase-conjugated AffiniPure goat anti-rabbit IgG (H + L), concentration 1:200, ZSGB-BIO, China) at 37 °C for 2 h. The DAB method was applied for staining the immunoreactive bands, and the average intensity of the antigen-antibody complex was measured in at least three hippocampal slices per animal and presented as a percentage using the software Image-Pro Plus 6.0 (Media Cybernetics, USA).

Memory dysfunction is a prominent symptom of patients with AD. In this study, the spatial working memory ability of mice was evaluated based on spontaneous alternation in the Y maze. Two-way ANOVA showed that the APP/PS1 gene mutation and Oxm treatment had significant main effect and interaction on the spontaneous alternation of the mice (APP/PS1: F(1, 39) = 17.5, P < .01, ηp2 = 0.33; Oxm: F(1, 2 39) = 15.1, P < .01, ηp = 0.30; APP/PS1 × Oxm: F(1, 39) = 12.3, 2 P < .01, ηp = 0.25). Tukey's post hoc tests showed that, compared with the WT + saline group, the percentage of correct alternation was significantly lower in the APP/PS1 + Saline group (P < .01, d = 2.085), while (D-Ser2) Oxm treatment prevented this reduction in APP/PS1 mice (P < .01, d = 2.075). However, the total arm entries of mice did not show any significant difference among the groups (Fig. 2B), suggesting that the differences in spontaneous alternation among these groups were due to impaired spatial working memory rather than impaired locomotor activity. 3.2. (D-Ser2)Oxm reverses the impairments in spatial learning and memory of APP/PS1 mice in the Morris water maze task The hippocampal-dependent long term spatial learning and memory ability of the mice were evaluated in the MWM task. As expected, the time (escape latency) to find the hidden platform decreased over the 5 consecutive days. As shown in Fig. 3A, no significant differences among the groups were detected in the first 2 days; however, the escape latency on training days 3, 4, and 5 was longer in the APP/PS1 + Saline group than in the WT + Saline group (P < .01). The increase in escape latency was reversed by Oxm treatment (P < .05 vs. APP/ PS1 + Saline). These results indicated that the spatial learning disabilities of APP/PS1 mice could be reversed by chronic i.p. (D-Ser2) Oxm administration. Probe trials without the platform were performed on day 6 to assess the long term spatial memory ability of the mice. Two-way ANOVA showed that the APP/PS1 gene mutation and (D-Ser2) Oxm treatment

2.7. Western blotting After behavioral tests, the remaining mice were used for western blot experiments, which were anesthetized by urethane (1.5 g/kg, i.p.). Brains were immediately removed and the hippocampi were dissected carefully on ice. Then, the hippocampal tissue was homogenized in tissue protein extraction reagent (Boster Inc., Pleasanton, CA, USA) supplemented with complete protease and phosphatase inhibitors (Boster). The homogenates were centrifuged at 13,000 rpm for 15 min at 4 °C. The protein concentrations were measured using a BCA kit (Boster). Equivalent samples of 30 μg of protein extract were separated on 12% SDS-polyacrylamide gels and transferred to PVDF membranes. 3

Hormones and Behavior 118 (2020) 104640

Z.-J. Wang, et al.

Fig. 2. (D-Ser2)Oxm improved working memory of APP/PS1 mice in a spontaneous alternation Y-maze test. (A) Histograms showing the correct spontaneous alternation percentage of the mice, which was higher in APP/PS1 + (D-Ser2) Oxm mice than APP/PS1 + PBS mice (**P < .01). (B) Histograms showing the total arm entries of mice in 8 min. No significant differences were found among the different groups (P > .05, n = 10 for each group).

Fig. 3. (D-Ser2)Oxm treatment reversed the impairments of spatial learning and memory in APP/PS1 mice. (A) Plots showing the changes in the escape latency of mice searching for the hidden platform over 5 consecutive training days. The escape latency was significantly extended in the APP/PS1 group in the 3rd, 4th, and 5th training sessions, and these increases were reversed by (D-Ser2)Oxm treatment (*P < .05, **P < .01 vs. APP/PS1 + saline group, n = 12 for each group). (B) Histograms showing the swimming time percentage of mice in the target quadrant in the probe trial on day 6. APP/PS1 mice treated with (D-Ser2)Oxm spent more time in the target quadrant than APP/PS1 mice treated with saline (**P < .01). (C) The swimming duration (s) of mice to the target did not differ significantly among the groups in the visible platform test (P > .05). (D) The swimming speed (cm/s) of the mice did not differ significantly among the groups (P > .05). (E) Sample swimming traces of mice in the four groups during the probe trial. 4

Hormones and Behavior 118 (2020) 104640

Z.-J. Wang, et al.

significantly affected the time in the target quadrant (APP/PS1: F(1,47) = 590.1, P < .01, ηp2 = 0.93; Oxm: F(1,47) = 287.2, P < .01, ηp2 = 0.87; APP/PS1 × Oxm interaction: F(1,47) = 263.9, P < .01, ηp2 = 0.86, Fig. 3B). Tukey's post hoc test showed that the time in the target quadrant was significantly lower for the APP/PS1 mice (P < .01, d = 0.19), and this decrease was reversed by (D-Ser2) Oxm treatment (P < .01, d = 2.18). These results indicated that (D-Ser2) Oxm ameliorated the spatial cognitive behavioral deficit of APP/PS1 mice. After the probe trial, the visible platform test was performed on day 6 for all mice to exclude visual or motor dysfunction. As shown in Fig. 3C and D, the time to the target did not differ significantly (P > .05) among all groups, thereby suggesting that the altered escape latency and time in the target quadrant were due to the differences in learning and memory impairment rather than in the visual or motor ability of the mice.

LTP is different among groups, with an obvious decrease in the APP/ PS1 mice at 60 min post-HFS. These results showed that the average LTP value was significantly suppressed (P < .01, d = 6.10) in APP/ PS1 mice, and this suppression was significantly reversed by (D-Ser2) Oxm treatment (P < .01, d = 6.04), suggesting that the drug exerted a protective effect on the deficits in hippocampal synaptic plasticity. Furthermore, we examined whether the presynaptic mechanism was altered in these four groups of mice by measuring paired-pulse facilitation (PPF) at an intra-pulse interval of 50 ms before HFS. Two-way ANOVA showed that the APP/PS1 gene mutation and Oxm treatment did not significantly affect PPF (Fig. 4D), suggesting that neither the APP/PS1 gene mutation nor (D-Ser2) Oxm affected presynaptic transmitter release in the hippocampal CA1 region.

3.3. (D-Ser2)Oxm reverses LTP impairments in the hippocampal CA1 region in APP/PS1 mice

Aβ plaques are the primary pathological biomarkers of AD. Fig. 5A shows representative immunohistochemistry images of amyloid plaques in the hippocampus of APP/PS1 mice and WT mice. No Aβ-immunopositive particles were detected in the 9-month-old WT mice; however, numerous large Aβ-immunopositive particles were observed in the hippocampus of the corresponding APP/PS1 mice. Importantly, after (D-Ser2)Oxm treatment, one-way ANOVA showed that the Aβpositive staining area of the hippocampus was significantly smaller in the APP/PS1 + OXM group than in the APP/PS1 + Saline group (Fig. 5B).

3.4. The number of amyloid plaques in the hippocampus of APP/PS1 mice is reduced by (D-Ser2)Oxm treatment

Synaptic plasticity is widely considered the physiological basis of memory formation and consolidation. Hence, we recorded hippocampal CA1 LTP to evaluate the effects of (D-Ser2)Oxm on the severe impairment of synaptic biology in the AD brain. After stably recording the baseline fEPSP for 30 min, LTP was induced by HFS. As shown in Fig. 4A and B, immediately after HFS, the amplitude of the fEPSPs (percentage of baseline) abruptly increased in the four groups，indicating a successful induction of LTP. However, the maintenance of

Fig. 4. (D-Ser2)Oxm treatment reversed the in vivo hippocampal LTP suppression in APP/PS1 mice. (A) Plots representing the time course of fEPSPs before and after high-frequency stimulation (HFS) in the four groups (n = 6 for each group). (B) Histograms showing the amplitude percentage of fEPSPs pre-HFS and 0 min, 30 min, and 60 min post-HFS (*P < .05 and **P < .01). (C) Sample fEPSP traces before and after HFS recorded from mice in the four groups. The dashed lines represent fEPSPs before HFS, and the solid lines represent fEPSPs after HFS. Scale bars, 1 mV and 10 ms. (D) Histograms showing that neither Oxm treatment nor APP/PS1 gene mutation affected PPF (fEPSP2/fEPSP1) in the hippocampal CA1 region. Inset: a sample trace of paired fEPSP, with an obvious increase in the amplitude of the second fEPSP. 5

Hormones and Behavior 118 (2020) 104640

Z.-J. Wang, et al.

Fig. 5. (D-Ser2)Oxm treatment was associated with fewer amyloid plaques in the hippocampus of APP/PS1 mice. (A) Representative immunohistochemistry images of amyloid plaques in the hippocampus of 9-month-old APP/PS1 and WT mice. (B) Histograms showing that after (D-Ser2)Oxm treatment, the percentage area of Aβ in the hippocampus decreased significantly in the APP/PS1 mice (*P < .05, n = 6 for each group).

PS1 × Oxm interaction: F(1,23) = 23.1, P < .01, ηp2 = 0.54), p-AKT1 (APP/PS1: F(1,23) = 4.0, P < .05, ηp2 = 0.17; Oxm: F(1,23) = 36.2, P < .01, ηp2 = 0.64; APP/PS1 × Oxm interaction: F(1,23) = 10.4, P < .01, ηp2 = 0.34), and GSK3β (APP/PS1: F(1,23) = 8.9, P < .01, ηp2 = 0.31; Oxm: F(1,23) = 48.2, P < .01, ηp2 = 0.71; APP/ PS1 × Oxm interaction: F(1,23) = 7.0, P < .05, ηp2 = 0.26). Tukey's post hoc test showed that the expression levels of p-PI3K (P < .01, d = 0.45) and p-AKT1(P < .01, d = 0.52) were lower in the APP/ PS1 + Saline group than in the WT + Saline group, and this decrease was reversed by (D-Ser2) Oxm treatment (p-PI3K: P < .01, d = 1.20; p-AKT1: P < .01, d = 0.45, Fig. 6B and C). By contrast, the level of pGSK3β in the APP/PS1 + Saline group increased as compared to that in

3.5. (D-Ser2)Oxm normalizes the PI3K/AKT1/GSK3β cell signaling pathway Aβ neurotoxicity and insulin resistance demonstrate reductions in the PI3K/AKT1/GSK3β signaling pathway. Therefore, to elucidate the molecular mechanisms underlying the neuroprotective effects of (DSer2)Oxm, we further examined the expression levels of p-PI3K, pAKT1, and p-GSK3β in the hippocampus of APP/PS1 mice by western blotting. As shown in Fig. 6, two-way ANOVA revealed that the APP/ PS1 gene mutation and Oxm treatment significantly affected the expression levels of p-PI3K (APP/PS1: F(1,23) = 12.5, P < .01, ηp2 = 0.39; Oxm: F(1,23) = 34.9, P < .01, ηp2 = 0.64; APP/

Fig. 6. The effects of (D-Ser2)Oxm on the expression of phosphorylated p-PI3K, p-AKT1, and p-GSK3β in the hippocampus of APP/PS1 and WT mice. (A) Representative western blot images of phosphorylated p-PI3K, p-AKT1, p-GSK3β, and β-actin in the four groups (n = 6 for each group). (B and C) Histograms showing the decreases in the levels of p-PI3K and p-AKT1 in the APP/PS1 + saline group were reversed by Oxm treatment (**P < .01). (D) Histograms showing that the increase in the level of p-GSK3β in the APP/PS1 + saline group was decreased by (D-Ser2)Oxm treatment (**P < .01). 6

Hormones and Behavior 118 (2020) 104640

Z.-J. Wang, et al.

Fig. 7. Body weight and blood glucose levels were not changed by (D-Ser2)Oxm. Body weight (A) and blood glucose (B) of the mice did not differ significantly among the different groups during the experiment (P > .05, n = 10 for each group).

synapses in APP/PS1 mice (McClean and Holscher, 2014a; McClean et al., 2015) and amyloid-β-injected rats (Qi et al., 2016). The GLP-1R agonist lixisenatide improved spatial memory in amyloid-β intrahippocampal injected rats (Cai et al., 2017), and reduced the amyloid plaque load and chronic inflammation in APP/PS1 mice (Li et al., 2018). D-Ala2-GIP, a protease resistant analogue of the peptide hormone gastric inhibitory polypeptide (GIP), protected memory formation, increased the number of synapses in the brain, and reduced Aβ deposition in APP/PS1 mice (Duffy and Holscher, 2013; Faivre and Holscher, 2013); DAJ-C4, a novel dual GLP-1/GIP receptor agonist, exhibited neuroprotective effects in the intraventricular injected streptozotocin (STZ)-induced AD rat model, including amelioration of memory disorders (Shi et al., 2017). These findings are in accordance with the present results, as chronic i.p. injection of (D-Ser2) Oxm improved working memory, long-term spatial memory and reduced pathological Aβ in APP/PS1 mice. Importantly, (D-Ser2)Oxm treatment in the study did not affect body weight or blood glucose levels in either APP/PS1 or WT mice during 4–5 weeks of experiments. These negative results suggest that long-term treatment with the dual agonist may be relative safe for those AD patients without T2DM by avoiding hypoglycemia and weight loss. Similarly, our recent work (Li et al., 2018) also showed that GLP-1/ GIP/GCG triagonist, another novel drug candidate for treating T2DM, did not affect the body weight of APP/PS1/Tau transgenic mice. As a glucose-dependent anti-diabetes drug, (D-Ser2)Oxm reduced hyperglycemia in T2DM and did not affect normal blood glucose in AD, which well explains why OXM only reduced body weight in the people who are overweight (Ambery et al., 2018; Field et al., 2010; Shankar et al., 2018). Therefore, (D-Ser2)Oxm may offer a novel and clinically valuable approach to disease management for AD patients, especially people who have T2DM or hyperglycemia at the same time. Learning and memory coding in animals enhances synaptic efficacy via a hippocampal LTP-like mechanism (Alam et al., 2018). Hippocampal LTP is considered one of the major electrophysiological synaptic mechanisms underlying learning and memory (Bliss and Collingridge, 1993). Hippocampal LTP can persist for at least 1 h by modifying preexisting proteins to exert synaptic plasticity (Barria et al., 1997). Therefore, hippocampal LTP might be the electrophysiological basis of the effects of long-term memory (Aidil-Carvalho et al., 2017; Araujo et al., 2017; Arias-Cavieres et al., 2017; Guimaraes Marques et al., 2018; Ma et al., 2017). Moreover, it has been reported that Aβ inhibits LTP both in hippocampal slices (Kapay et al., 2013) and in vivo (Wang et al., 2014; Wu et al., 2015). In the present study, we found that (DSer2)Oxm alone did not affect basic synaptic transmission but significantly alleviated the hippocampal LTP depression in APP/PS1 mice. The protective effect of (D-Ser2)Oxm on in vivo hippocampal LTP strongly supports the behavioral improvement of APP/PS1 mice in the

the WT + Saline group (P < .01, d = 0.18), which was decreased by (D-Ser2)Oxm treatment (P < .01, d = 1.59, Fig. 6D). These results indicated that the prevention of deficits in spatial memory, synaptic plasticity, and amyloid plaques by (D-Ser2)Oxm might be related to activation of the PI3K-AKT1 signaling pathway and inhibition of the GSK3β signaling pathway. 3.6. (D-Ser2)Oxm does not affect body weight and blood glucose of mice We examined whether body weight and blood glucose of mice was affected by i.p. injection of (D-Ser2) Oxm. Two-way ANOVA showed that the APP/PS1 gene mutation and (D-Ser2)Oxm treatment did not significantly change the body weight (Fig. 7A) and blood glucose (Fig. 7B). Thus, (D-Ser2)Oxm did not affect body weight or blood glucose levels in the normal weight, non-glycometabolism disorder mice. 4. Discussion (D-Ser2)Oxm has been shown to have a prolonged half-life and greater DPP-IV resistance compared with native Oxm. A recent study found that Oxm exerts positive effects on hippocampal neurogenesis and gene expression and improves glucose homeostasis in mice fed a high-fat diet, thus demonstrating the potential of (D-Ser2)Oxm to enhance neurogenesis and synaptogenesis and exert protective effects against oxidative damage in the hippocampus and cortex of mice fed a high-fat diet (Pathak et al., 2015). Oxm increases cell viability and protects cells from glutamate toxicity and oxidative stress in a time and dose-dependent manner (Li et al., 2017). (D-Ser2)Oxm also prevented motor impairments in a Parkinson's disease mouse model, protected dopaminergic neurons and reduced chronic inflammation in the brain (Liu et al., 2015). We previously found that (D-Ser2)Oxm can protect hippocampal neurons against Aβ1–42-induced cytotoxicity and that this effect may be related to the regulation of intracellular calcium homeostasis and stabilization of mitochondrial membrane potential (Han et al., 2016). Here, we confirmed for the first time that (D-Ser2)Oxm exerts neuroprotective effects in APP/PS1 mice, a model of AD. In the Y maze test, (D-Ser2)Oxm treatment did not affect the total arm entries but completely reversed the decrease in spontaneous alternation of APP/PS1 mice, suggesting that (D-Ser2)Oxm can effectively prevent the deficits of spatial working memory caused by the APP/PS1 gene mutation without affecting locomotor activity. In the MWM task, (D-Ser2) Oxm effectively decreased the escape latency of APP/PS1 mice in the hidden platform test and increased the duration of swimming in the target quadrant in the probe trials, which demonstrates that (D-Ser2) Oxm can ameliorate the deficits in spatial learning and memory of mice. Our results are in line with previous studies of the GLP-1 analog liraglutide that prevented the decline of spatial memory and loss of 7

Hormones and Behavior 118 (2020) 104640

Z.-J. Wang, et al.

term spatial memory, hippocampal LTP, and Aβ deposition in APP/PS1 transgenic mice. The mechanisms of (D-Ser2)Oxm might involve the activation of insulin and GLP-1 receptors and modulation of the PI3K/ AKT/GSK3β signaling pathway. These experimental findings provide further evidence that long-acting OXM analogues may be an effective treatment for AD.

Y maze test and MWM task, suggesting a mechanism underlying the (DSer2)Oxm-mediated reversal of cognitive impairment in APP/PS1 transgenic mice. Aggregation of amyloid plaques is one of the characteristic pathological hallmarks of AD. The present experimental results did not find any immunopositive staining of Aβ in the hippocampus in 9-month-old WT mice, whereas many Aβ-immunopositive plaques were detected in APP/PS1 mice of the same age. Treatment with (D-Ser2)Oxm reduced the deposition of amyloid plaques in the hippocampus of APP/PS1 mice. Previous studies have also shown a reduction in the number of amyloid plaques in various AD transgenic mouse models upon treatment with a GLP-1R agonist, GLP-1/GIP dual agonist, or GLP-1/GIP/ GCG triagonist (Cao et al., 2018; Holscher, 2014a; Li et al., 2018). We previously also found that (D-Ser2)Oxm significantly antagonizes Aβ1–42-induced phenotypes in cell viability, neuronal early apoptosis, mitochondrial membrane potential, and intracellular calcium concentration (Han et al., 2013). These changes may contribute to the improvement of cognitive behavior and synaptic plasticity in APP/PS1 mice. Furthermore, this protective effect is inhibited by pretreatment with exendin (9-39), a GLP-1 receptor blocker, suggesting that this effect might be related to the activation of GLP-1Rs, regulation of intracellular calcium homeostasis, and stabilization of mitochondrial membrane potential (Rivero-Gutierrez et al., 2018). It is well known that insulin signaling plays an important role in the neuronal function and synaptogenesis. Most recent study showed that the downstream mediators of insulin signaling pathway function as a regulatory hub for aggregation and clearance of unfolded proteins like Aβ and Tau (Gupta et al., 2018). Moreover, several epidemiological studies have shown that the systemic insulin resistance state of T2DM is a major risk factor for age-related cognitive decline and progression from mild cognitive impairment (MCI) to AD (Craft et al., 2013; Freiherr et al., 2013). A recent study suggests that brain insulin resistance directly promotes the development of Aβ and neurofibrillary tangles, and shares the down-regulation of PI3K/AKT1/GSK3β signaling pathway with Aβ (Rad et al., 2018), while inhibition of PI3K/ AKT pathway has been connected to the clearance of Aβ plaques in the hippocampus (Ali and Kim, 2015; Xian et al., 2014). In addition, our previous in vitro study showed that the neuroprotective mechanism of (D-Ser2)Oxm involves the PI3K/AKT/GSK3β signaling pathway (Ali and Kim, 2015; Cao et al., 2018; Cheng et al., 2018). The PI3K/AKT signaling pathway is activated by a range of growth factors such as insulin and IGF-1. These key kinases regulate energy utilization by enhancing mitochondrial synthesis via PGC-1α and Nrf1 transcription factors (Gutierrez-Rodelo et al., 2017; Steen et al., 2005). Reduced energy utilization in the brain of AD patients is a key hallmark of the disease (Edison et al., 2008; Neth and Craft, 2017). GLP-1 and glucagon receptors activate PKA and CREB to enhance gene expression of proteins related to insulin cell signaling (Ayush et al., 2015; Bomfim et al., 2012; Doyle and Egan, 2007; Talbot, 2014). The high levels of senile plaques in APP/PS1 mice with unbalanced PI3K/AKT activity resulted in accelerated impairment of memory capacity and synaptic plasticity. (D-Ser2)Oxm treatment not only reduced these pathological hallmarks but also normalized the impaired PI3K/AKT/GSK3β levels. The normalizing of PI3K/AKT cell signaling demonstrates that insulin signaling in the brain has been normalized by the drug as shown previously with GLP-1R agonists (Bomfim et al., 2012; Long-Smith et al., 2013). The dual GLP-1/GIP receptor agonist DA-JC4 also improved insulin signaling in an i.c.v. STZ rat model of AD (Shi et al., 2017). Interestingly, the (D-Ser2) Oxm treatment almost completely reversed the deficits in behavior or pathology while only partially ameliorating the changes in the PI3K/AKT/GSK3β signaling pathway, suggesting that other signaling pathways may also be involved in the neuroprotective mechanism of (D-Ser2) Oxm such as cAMP/Ras/MAPK-ERK and PKA/ CREB (Chen et al., 2012; Cuellar and Isokawa, 2011). In summary, the present results indicate that (D-Ser2) Oxm can reverse the impairments in short-term spatial working memory, long-

Acknowledgements We thank Peerwith for his linguistic assistance during the preparation of this manuscript. Funding This work was supported by the grants from the National Natural Science Foundation of China (31600865, 31700918); The Higher School Science and Technology Innovation Project of Education Department in Shanxi Province (2015153); The Doctoral Startup Research Fund of Shanxi Medical University (03201404, 03201536); “Sanjin Scholars” of Shanxi Province ([2016]7), Shanxi Province Science Foundation for Excellent Young Scholars (201801D211005), Shanxi Province Science Foundation for Youth (201801D221263), and the Fund for Shanxi Key Subjects Construction, FSKSC, Shanxi “1331 Project” Key Subjects Construction (1331KSC) and Key Laboratory of Cellular Physiology (Shanxi Medical University) in Shanxi Province. References Aidil-Carvalho, M.F., Carmo, A.J.S., Ribeiro, J.A., Cunha-Reis, D., 2017. Mismatch novelty exploration training enhances hippocampal synaptic plasticity: a tool for cognitive stimulation? Neurobiol. Learn. Mem. 145, 240–250. Alam, M.J., Kitamura, T., Saitoh, Y., Ohkawa, N., Kondo, T., Inokuchi, K., 2018. Adult neurogenesis conserves hippocampal memory capacity. J. Neurosci. 38, 6854–6863. Ali, T., Kim, M.O., 2015. Melatonin ameliorates amyloid beta-induced memory deficits, tau hyperphosphorylation and neurodegeneration via PI3/Akt/GSk3beta pathway in the mouse hippocampus. J. Pineal Res. 59, 47–59. Amar, S., Belmaker, R.H., Agam, G., 2011. The possible involvement of glycogen synthase kinase-3 (GSK-3) in diabetes, cancer and central nervous system diseases. Curr. Pharm. Des. 17, 2264–2277. Ambery, P., Parker, V.E., Stumvoll, M., Posch, M.G., Heise, T., Plum-Moerschel, L., Tsai, L.F., Robertson, D., Jain, M., Petrone, M., Rondinone, C., Hirshberg, B., Jermutus, L., 2018. MEDI0382, a GLP-1 and glucagon receptor dual agonist, in obese or overweight patients with type 2 diabetes: a randomised, controlled, double-blind, ascending dose and phase 2a study. Lancet 391, 2607–2618. Araujo, D.J., Toriumi, K., Escamilla, C.O., Kulkarni, A., Anderson, A.G., Harper, M., Usui, N., Ellegood, J., Lerch, J.P., Birnbaum, S.G., Tucker, H.O., Powell, C.M., Konopka, G., 2017. Foxp1 in forebrain pyramidal neurons controls gene expression required for spatial learning and synaptic plasticity. J. Neurosci. 37, 10917–10931. Arias-Cavieres, A., Adasme, T., Sanchez, G., Munoz, P., Hidalgo, C., 2017. Aging impairs hippocampal- dependent recognition memory and LTP and prevents the associated RyR up-regulation. Front. Aging Neurosci. 9, 111. Arnold, S.E., Arvanitakis, Z., Macauley-Rambach, S.L., Koenig, A.M., Wang, H.Y., Ahima, R.S., Craft, S., Gandy, S., Buettner, C., Stoeckel, L.E., Holtzman, D.M., Nathan, D.M., 2018. Brain insulin resistance in type 2 diabetes and Alzheimer disease: concepts and conundrums. Nat. Rev. Neurol. 14, 168–181. Ayush, E.A., Iwasaki, Y., Iwamoto, S., Nakabayashi, H., Kakei, M., Yada, T., 2015. Glucagon directly interacts with vagal afferent nodose ganglion neurons to induce Ca (2+) signaling via glucagon receptors. Biochem. Biophys. Res. Commun. 456, 727–732. Bak, J., Pyeon, H.I., Seok, J.I., Choi, Y.S., 2017. Effect of rotation preference on spontaneous alternation behavior on Y maze and introduction of a new analytical method, entropy of spontaneous alternation. Behav. Brain Res. 320, 219–224. Barria, A., Muller, D., Derkach, V., Griffith, L.C., Soderling, T.R., 1997. Regulatory phosphorylation of AMPA-type glutamate receptors by CaM-KII during long-term potentiation. Science 276, 2042–2045. Bliss, T.V., Collingridge, G.L., 1993. A synaptic model of memory: long-term potentiation in the hippocampus. Nature 361, 31–39. Bomfim, T.R., Forny-Germano, L., Sathler, L.B., Brito-Moreira, J., Houzel, J.C., Decker, H., Silverman, M.A., Kazi, H., Melo, H.M., McClean, P.L., Holscher, C., Arnold, S.E., Talbot, K., Klein, W.L., Munoz, D.P., Ferreira, S.T., De Felice, F.G., 2012. An antidiabetes agent protects the mouse brain from defective insulin signaling caused by Alzheimer’s disease- associated Abeta oligomers. J. Clin. Invest. 122, 1339–1353. Cai, H.Y., Holscher, C., Yue, X.H., Zhang, S.X., Wang, X.H., Qiao, F., Yang, W., Qi, J.S., 2014. Lixisenatide rescues spatial memory and synaptic plasticity from amyloid beta protein-induced impairments in rats. Neuroscience 277, 6–13. Cai, H.Y., Wang, Z.J., Holscher, C., Yuan, L., Zhang, J., Sun, P., Li, J., Yang, W., Wu, M.N., Qi, J.S., 2017. Lixisenatide attenuates the detrimental effects of amyloid beta protein on spatial working memory and hippocampal neurons in rats. Behav. Brain Res. 318, 28–35. Cao, Y., Holscher, C., Hu, M.M., Wang, T., Zhao, F., Bai, Y., Zhang, J., Wu, M.N., Qi, J.S.,

8

Hormones and Behavior 118 (2020) 104640

Z.-J. Wang, et al.

C., O’Neill, C., 2013. The diabetes drug liraglutide ameliorates aberrant insulin receptor localisation and signalling in parallel with decreasing both amyloid-beta plaque and glial pathology in a mouse model of Alzheimer’s disease. NeuroMolecular Med. 15, 102–114. Luchsinger, J.A., Tang, M.X., Shea, S., Mayeux, R., 2004. Hyperinsulinemia and risk of Alzheimer disease. Neurology 63, 1187–1192. Ma, Y., Sun, X., Li, J., Jia, R., Yuan, F., Wei, D., Jiang, W., 2017. Melatonin alleviates the epilepsy-associated impairments in hippocampal LTP and spatial learning through rescue of surface GluR2 expression at hippocampal CA1 synapses. Neurochem. Res. 42, 1438–1448. McClean, P.L., Holscher, C., 2014a. Liraglutide can reverse memory impairment, synaptic loss and reduce plaque load in aged APP/PS1 mice, a model of Alzheimer's disease. Neuropharmacology 76 Pt A 57–67. McClean, P.L., Holscher, C., 2014b. Lixisenatide, a drug developed to treat type 2 diabetes, shows neuroprotective effects in a mouse model of Alzheimer's disease. Neuropharmacology 86C, 241–258. McClean, P.L., Jalewa, J., Holscher, C., 2015. Prophylactic liraglutide treatment prevents amyloid plaque deposition, chronic inflammation and memory impairment in APP/ PS1 mice. Behav. Brain Res. 293, 96–106. Moloney, A.M., Griffin, R.J., Timmons, S., O’Connor, R., Ravid, R., O’Neill, C., 2010. Defects in IGF-1 receptor, insulin receptor and IRS-1/2 in Alzheimer’s disease indicate possible resistance to IGF-1 and insulin signalling. Neurobiol. Aging 31, 224–243. Morris, R., 1984. Developments of a water-maze procedure for studying spatial learning in the rat. J. Neurosci. Methods 11, 47–60. Mwamburi, M., Qiu, W.Q., 2016. Different associations of premorbid intelligence vs. current cognition with BMI, insulin and diabetes in the homebound elderly. Integrative molecular medicine 3, 547–552. Neth, B.J., Craft, S., 2017. Insulin resistance and Alzheimer’s disease: bioenergetic linkages. Front. Aging Neurosci. 9, 345. Onyango, I.G., 2018. Modulation of mitochondrial bioenergetics as a therapeutic strategy in Alzheimer’s disease. Neural Regen. Res. 13, 19–25. Pathak, N.M., Pathak, V., Lynch, A.M., Irwin, N., Gault, V.A., Flatt, P.R., 2015. Stable oxyntomodulin analogues exert positive effects on hippocampal neurogenesis and gene expression as well as improving glucose homeostasis in high fat fed mice. Mol. Cell. Endocrinol. 412, 95–103. Pocai, A., 2012. Unraveling oxyntomodulin, GLP1’s enigmatic brother. J. Endocrinol. 215, 335–346. Qi, L., Ke, L., Liu, X., Liao, L., Ke, S., Liu, X., Wang, Y., Lin, X., Zhou, Y., Wu, L., Chen, Z., Liu, L., 2016. Subcutaneous administration of liraglutide ameliorates learning and memory impairment by modulating tau hyperphosphorylation via the glycogen synthase kinase-3beta pathway in an amyloid beta protein induced alzheimer disease mouse model. Eur. J. Pharmacol. 783, 23–32. Rad, S.K., Arya, A., Karimian, H., Madhavan, P., Rizwan, F., Koshy, S., Prabhu, G., 2018. Mechanism involved in insulin resistance via accumulation of beta-amyloid and neurofibrillary tangles: link between type 2 diabetes and Alzheimer's disease. Drug design, development and therapy 12, 3999–4021. Rivero-Gutierrez, B., Haller, A., Holland, J., Yates, E., Khrisna, R., Habegger, K., Dimarchi, R., D'Alessio, D., Perez-Tilve, D., 2018. Deletion of the glucagon receptor gene before and after experimental diabetes reveals differential protection from hyperglycemia. Molecular metabolism 17, 28–38. Shankar, S.S., Shankar, R.R., Mixson, L.A., Miller, D.L., Pramanik, B., O’Dowd, A.K., Williams, D.M., Frederick, C.B., Beals, C.R., Stoch, S.A., Steinberg, H.O., Kelley, D.E., 2018. Native oxyntomodulin has significant glucoregulatory effects independent of weight loss in obese humans with and without type 2 diabetes. Diabetes 67, 1105–1112. Shi, L., Zhang, Z., Li, L., Holscher, C., 2017. A novel dual GLP-1/GIP receptor agonist alleviates cognitive decline by re-sensitizing insulin signaling in the Alzheimer icv. STZ rat model. Behav. Brain Res. 327, 65–74. Steen, E., Terry, B.M., Rivera, E., Cannon, J.L., Neely, T.R., Tavares, R., Xu, X.J., Wands, J.R., de la Monte, S.M., 2005. Impaired insulin and insulin-like growth factor expression and signaling mechanisms in Alzheimer’s disease - is this type 3 diabetes? J. Alzheimers Dis. 7, 63–80. Talbot, K., 2014. Brain insulin resistance in Alzheimer's disease and its potential treatment with GLP-1 analogs. Neurodegenerative disease management 4, 31–40. Talbot, K., Wang, H.Y., Kazi, H., Han, L.Y., Bakshi, K.P., Stucky, A., Fuino, R.L., Kawaguchi, K.R., Samoyedny, A.J., Wilson, R.S., Arvanitakis, Z., Schneider, J.A., Wolf, B.A., Bennett, D.A., Trojanowski, J.Q., Arnold, S.E., 2012. Demonstrated brain insulin resistance in Alzheimer’s disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. J. Clin. Invest. 122, 1316–1338. Vrang, N., Larsen, P.J., 2010. Preproglucagon derived peptides GLP-1, GLP-2 and oxyntomodulin in the CNS: role of peripherally secreted and centrally produced peptides. Prog. Neurobiol. 92, 442–462. Wang, Z.J., Han, W.N., Yang, G.Z., Yuan, L., Liu, X.J., Li, Q.S., Qi, J.S., 2014. The neuroprotection of Rattin against amyloid beta peptide in spatial memory and synaptic plasticity of rats. Hippocampus 24, 44–53. Wu, M.N., Zhou, L.W., Wang, Z.J., Han, W.N., Zhang, J., Liu, X.J., Tong, J.Q., Qi, J.S., 2015. Colivelin ameliorates amyloid beta peptide-induced impairments in spatial memory, synaptic plasticity, and calcium homeostasis in rats. Hippocampus 25, 363–372. Wynne, K., Park, A.J., Small, C.J., Patterson, M., Ellis, S.M., Murphy, K.G., Wren, A.M., Frost, G.S., Meeran, K., Ghatei, M.A., Bloom, S.R., 2005. Subcutaneous oxyntomodulin reduces body weight in overweight and obese subjects: a double-blind, randomized, controlled trial. Diabetes 54, 2390–2395. Xian, Y.F., Mao, Q.Q., Wu, J.C., Su, Z.R., Chen, J.N., Lai, X.P., Ip, S.P., Lin, Z.X., 2014. Isorhynchophylline treatment improves the amyloid-beta-induced cognitive impairment in rats via inhibition of neuronal apoptosis and tau protein hyperphosphorylation. J. Alzheimers Dis. 39, 331–346.

2018. DA5-CH, a novel GLP-1/GIP dual agonist, effectively ameliorates the cognitive impairments and pathology in the APP/PS1 mouse model of Alzheimer’s disease. Eur. J. Pharmacol. 827, 215–226. Chen, Y., Huang, X., Zhang, Y.W., Rockenstein, E., Bu, G., Golde, T.E., Masliah, E., Xu, H., 2012. Alzheimer’s beta-secretase (BACE1) regulates the cAMP/PKA/CREB pathway independently of beta-amyloid. J. Neurosci. 32, 11390–11395. Cheng, W., Chen, W., Wang, P., Chu, J., 2018. Asiatic acid protects differentiated PC12 cells from Abeta25-35-induced apoptosis and tau hyperphosphorylation via regulating PI3K/Akt/GSK-3beta signaling. Life Sci. 208, 96–101. Congdon, E.E., Sigurdsson, E.M., 2018. Tau-targeting therapies for Alzheimer disease. Nat. Rev. Neurol. 14, 399–415. Craft, S., 2007. Insulin resistance and Alzheimer’s disease pathogenesis: potential mechanisms and implications for treatment. Curr. Alzheimer Res. 4, 147–152. Craft, S., Cholerton, B., Baker, L.D., 2013. Insulin and Alzheimer’s disease: untangling the web. J. Alzheimers Dis. 33 (Suppl. 1), S263–S275. Cuellar, J.N., Isokawa, M., 2011. Ghrelin-induced activation of cAMP signal transduction and its negative regulation by endocannabinoids in the hippocampus. Neuropharmacology 60, 842–851. Doyle, M.E., Egan, J.M., 2007. Mechanisms of action of glucagon-like peptide 1 in the pancreas. Pharmacol. Ther. 113, 546–593. Duffy, A.M., Holscher, C., 2013. The incretin analogue D-Ala2GIP reduces plaque load, astrogliosis and oxidative stress in an APP/PS1 mouse model of Alzheimer's disease. Neuroscience 228, 294–300. Edison, P., Archer, H.A., Gerhard, A., Hinz, R., Pavese, N., Turkheimer, F.E., Hammers, A., Tai, Y.F., Fox, N., Kennedy, A., Rossor, M., Brooks, D.J., 2008. Microglia, amyloid, and cognition in Alzheimer’s disease: an [11C](R)PK11195-PET and [11C]PIB-PET study. Neurobiol. Dis. 32, 412–419. Faivre, E., Holscher, C., 2013. Neuroprotective effects of D-Ala(2)GIP on Alzheimer’s disease biomarkers in an APP/PS1 mouse model. Alzheimers Res. Ther. 5, 20. Field, B.C., Wren, A.M., Peters, V., Baynes, K.C., Martin, N.M., Patterson, M., Alsaraf, S., Amber, V., Wynne, K., Ghatei, M.A., Bloom, S.R., 2010. PYY3-36 and oxyntomodulin can be additive in their effect on food intake in overweight and obese humans. Diabetes 59, 1635–1639. Freiherr, J., Hallschmid, M., Frey 2nd, W.H., Brunner, Y.F., Chapman, C.D., Holscher, C., Craft, S., De Felice, F.G., Benedict, C., 2013. Intranasal insulin as a treatment for Alzheimer's disease: a review of basic research and clinical evidence. CNS drugs 27, 505–514. Gao, C., Holscher, C., Liu, Y., Li, L., 2011. GSK3: a key target for the development of novel treatments for type 2 diabetes mellitus and Alzheimer disease. Rev. Neurosci. 23, 1–11. Guimaraes Marques, M.J., Reyes-Garcia, S.Z., Marques-Carneiro, J.E., Lopes-Silva, L.B., Andersen, M.L., Cavalheiro, E.A., Scorza, F.A., Scorza, C.A., 2018. Long-term potentiation decay and poor Long-lasting memory process in the wild rodents proechimys from Brazil’s Amazon rainforest. Front. Behav. Neurosci. 12, 2. Gupta, S., Singhal, N.K., Ganesh, S., Sandhir, R., 2018. Extending arms of insulin resistance from diabetes to Alzheimer’s disease: identification of potential therapeutic targets. CNS & neurological disorders drug targets. 18, 172–184. Gutierrez-Rodelo, C., Roura-Guiberna, A., Olivares-Reyes, J.A., 2017. Molecular mechanisms of insulin resistance: an update. Gac. Med. Mex. 153, 214–228. Han, W.N., Holscher, C., Yuan, L., Yang, W., Wang, X.H., Wu, M.N., Qi, J.S., 2013. Liraglutide protects against amyloid-beta protein-induced impairment of spatial learning and memory in rats. Neurobiol. Aging 34, 576–588. Han, Y.F., Holscher, C., Wang, Z.J., Zhang, J., Yuan, L., Tong, J.Q., Wang, D.D., Wu, M.N., Qi, J.S., 2016. Neuroprotective effects of a novel antidiabetic drug (D-Ser2)Oxm on amyloid beta protein-induced cytotoxicity. Sheng li xue bao: [Acta physiologica Sinica] 68, 265–275. Holscher, C., 2014a. Drugs developed for treatment of diabetes show protective effects in Alzheimer's and Parkinson's diseases. Acta Physiol Sinica 66, 497–510. Holscher, C., 2014b. First clinical data of the neuroprotective effects of nasal insulin application in patients with Alzheimer's disease. Alzheimer's & dementia: the journal of the Alzheimer's Association 10, S33–S37. Hölscher, C., 2018. Novel dual GLP-1/GIP receptor agonists show neuroprotective effects in Alzheimer’s and Parkinson’s disease models. Neuropharmacol. 136, 251–259. Hoyer, S., 1998. Risk factors for Alzheimer’s disease during aging. Impacts of glucose/ energy metabolism. J. Neural Transm. Suppl. 54, 187–194. Janson, J., Laedtke, T., Parisi, J.E., O'Brien, P., Petersen, R.C., Butler, P.C., 2004. Increased risk of type 2 diabetes in Alzheimer disease. Diabetes 53, 474–481. Kapay, N.A., Popova, O.V., Isaev, N.K., Stelmashook, E.V., Kondratenko, R.V., Zorov, D.B., Skrebitsky, V.G., Skulachev, V.P., 2013. Mitochondria-targeted plastoquinone antioxidant SkQ1 prevents amyloid-beta-induced impairment of long-term potentiation in rat hippocampal slices. J. Alzheimers Dis. 36, 377–383. Kerr, B.D., Flatt, P.R., Gault, V.A., 2010. (D-Ser2)Oxm[mPEG-PAL]: a novel chemically modified analogue of oxyntomodulin with antihyperglycaemic, insulinotropic and anorexigenic actions. Biochem. Pharmacol. 80, 1727–1735. Kosinski, J.R., Hubert, J., Carrington, P.E., Chicchi, G.G., Mu, J., Miller, C., Cao, J., Bianchi, E., Pessi, A., Sinharoy, R., Marsh, D.J., Pocai, A., 2012. The glucagon receptor is involved in mediating the body weight-lowering effects of oxyntomodulin. Obesity 20, 1566–1571. Li, Y., Wu, K.J., Yu, S.J., Tamargo, I.A., Wang, Y., Greig, N.H., 2017. Neurotrophic and neuroprotective effects of oxyntomodulin in neuronal cells and a rat model of stroke. Exp. Neurol. 288, 104–113. Li, T., Jiao, J.J., Holscher, C., Wu, M.N., Zhang, J., Tong, J.Q., Dong, X.F., Qu, X.S., Cao, Y., Cai, H.Y., Su, Q., Qi, J.S., 2018. A novel GLP-1/GIP/Gcg triagonist reduces cognitive deficits and pathology in the 3xTg mouse model of Alzheimer's disease. Hippocampus 28, 358–372. Liu, W., Li, Y., Jalewa, J., Saunders-Wood, T., Li, L., Holscher, C., 2015. Neuroprotective effects of an oxyntomodulin analogue in the MPTP mouse model of Parkinson’s disease. Eur. J. Pharmacol. 765, 284–290. Long-Smith, C.M., Manning, S., McClean, P.L., Coakley, M.F., O’Halloran, D.J., Holscher,

9