Inhibition of Histamine H3 receptor Attenuates Neuroinammation and Cognitive Impairments in Alzheimer’s Disease via activating CREB Pathway

Alzheimer's disease (AD) is an age-related neurodegenerative disease, which characterized by deposition of amyloid-β (Aβ) plaques, neurobrillary tangles, neuronal loss, and accompanied by neuroinammation. Neuroinammatory processes are well acknowledged to contribute to the progression of AD pathology. Histamine H3 receptor (H3R) is a presynaptic autoreceptor regulating histamine release via negative feedback way. H3R antagonist has been reported to have anti-inammatory ecacy. However, whether inhibition of H3R is responsible for the anti-neuroinammation and neuroprotection on APP/PS1 Tg mice remains unclear. the nuclear factor in APP/PS1 Tg an effects of gliosis and cytokine APP/PS1 which reversed by administration of H89.

All animal studies were carried out according to protocols approved by the Institutional Animal Care and Use Committee of Binzhou Medical University, and conducted in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Efforts were made to minimize any pain or discomfort, and the minimum number of animals was used.

Animals
Adult male wild type (WT) and BL/6-Tg (APPswe, PSEN1dE9) (APP/PS1) mice of 9 months (from Jackson Lab, Stock Number: 004462) were used in this study. The mice were housed in a temperatureand humidity-controlled animal facility, which was maintained on a 12 h light/dark cycle, food and water were given ad libitum.

Stereotaxic surgery
Stereotaxic surgery was performed as previously described [43]. Mice were anesthetized with the intraperitoneal injection of 1% chloral hydrate and then immobilized on a stereotactic frame. The gauge guide cannula was implanted into the lateral ventricle (0.2 mm posterior, 1.1 mm lateral and 2.7 mm ventral to the bregma). After surgery, mice were housed individually and allowed to recover for 7 d.

Drug treatments
For in vivo study, a stainless-steel injector connected to a 5-μl syringe was inserted into the guide cannula and extended 1 mm beyond the tip. The administration of the chemicals was administrated on the basis of previous studies [44]. Thioperamide (i.p., 5 mg/kg) or vehicle was administrated daily at 7 days after stereotaxic surgery until the beginning of the behavior tests on day 14. H89 or vehicle (i.c.v., 5 µM of 2 µl) was administrated 0.5 h before thioperamide injection. Thioperamide and H89 were dissolved in saline. At 14 days after stereotaxic surgery, the novel object recognition (NOR) test was carried on for 4 days. At 18 days after stereotaxic surgery, Y maze (YM) test was carried on for 1 day. At 19 days after stereotaxic surgery, morris water maze (MWM) test was carried on for 6 days. During the period of behavior test, thioperamide or vehicle was administrated 0.5 h, whereas H89 or vehicle was administrated 1 h before the test at day 14, day 18 and day 19.

Western blot
Western blot was performed as previously described [45]. Brie y, mice were anaesthetized by i.p. injection of chloral hydrate (400 mg/kg), sacri ced and the brain was quickly removed. The separated tissues were lysed in ice-cold RIPA lysis buffer (R0020, Solarbio), then centrifuged at 14,000×g at 4°C for 20 min, and the protein concentration in the extracts was determined by the Bradford assay (Thermo, Hercules, CA). The precipitates were denatured with SDS sample loading buffer and separated on 10% SDS-PAGE.
Proteins were transferred onto nitrocellulose membranes using a Bio-Rad mini-protein-III wet transfer unit overnight at 4°C. Transfer membranes were then incubated with blocking solution (5% nonfat dried milk dissolved in tris buffered saline tween (TBST) buffer (in mM): 10 Tris-HCl, 150 NaCl, and 0.1% Tween-20) for 2 h at room temperature, and incubated with primary antibody overnight at 4°C. The primary antibodies used in this experiment were phospho-CREB (Cell Signaling Technology, 1:1,000), CREB (Cell Signaling Technology, 1:1,000), p-P65 NF-kB (abcam, 1:1000), P65 NF-kB (abcam, 1:1000) and GAPDH (Kangchen Biotech, 1:3,000). Membranes were washed three times in TBST buffer and incubated with the appropriate secondary antibodies (LI-COR, Odyssey, 1:5,000) for 2 h. Images were acquired with the Odyssey infrared imaging system and analyzed as speci ed in the Odyssey software manual. The results were expressed as the target protein/GAPDH ratio and then normalized to the values measured in the control groups (presented as 100%).

Novel object recognition (NOR)
The NOR test was performed 14 days after stereotaxic surgery as previously described [43,46]. Brie y, mice received 2 d of habituation in a 45 × 45 cm square arena, and on the third day, they were allowed to explore two identical objects made from large Lego bricks for 10 min (training trial). They were returned to their home cage, and 24 h later, a different shape and color object replaced one of the objects and the mice were returned to the arena for 10 min (testing trial). The time spent on each object was then calculated as a percentage of total object exploration.
The Y maze test was performed 18 days after stereotaxic surgery as previously described [43,46]. Brie y, the apparatus for YM was made of gray plastic, with each arm 40 cm long, 12 cm high, 3 cm wide at the bottom and 10 cm wide at the top. The three arms were connected at an angle of 120°. Mice were individually placed at the end of an arm and allowed to explore the maze freely for 8 min. The total arm entries and spontaneous alternation percentage (SA%) were measured. SA% was de ned as a ratio of the arm choices that differed from the previous two choices ('successful choices') to total choices during the run ('total entry minus two' because the rst two entries could not be evaluated). For example, if a mouse made 10 entries, such as 1-2-3-2-3-1-2-3-2-1, there are 5 successful choices in 8 total choices (10 entries minus 2). Therefore, SA% in this case is 62.5%.

Morris water maze (MWM)
The MWM maze test was performed 19 days after stereotaxic surgery as previously described [43,47]. Brie y, the water maze of 1.50 m in diameter and 0.50 m in height was lled with water (20 ± 1°C) to maintain the water surface 1.50 cm higher than the platform (10 cm in diameter). Water was dyed white and the tank was divided into four quadrants by four points: North (N), South (S), East (E), and West (W). The platform was placed at the center of either quadrant and video tracking software was used to automatically track the animals. Learning and memory acquisition lasts for ve days. Animals were put into the water from four points in random order every day until they found the platform and stayed for 10 s within 1 min. If the mice cannot nd the platform within 1 min, they were guided to the platform. Following acquisition test, on the sixth day, learning and memory maintenance test was carried on. The platform was removed, and the mice were placed in water from the opposite quadrant of the platform, and then the times crossing the platform was recorded within 1 min.

Statistical analysis
Results are expressed as mean ± SEM. Statistical analysis was performed by one-way ANOVAs followed by Tukey's post hoc comparisons or two-way ANOVAs followed by Bonferroni post hoc comparisons, using prism software. P value <0.05 was considered statistically signi cant.

Results
Thioperamide decreases the microglial reactivity in APP/PS1 Tg mice We used 8-month-old APP/PS1 Tg mice which have the mutated human APPswe and PSEN1ΔE9, to examine the therapeutic e cacy of thioperamide, a H3R antagonist in AD. Activated microglia may play an important role in the pathogenesis of AD as they cluster around Aβ plaques [48]. Therefore, we examined the activated microglia in the hippocampus and cortex by Iba1 immunostaining. Results showed that the area of Iba1 + -cells in the whole hippocampus, hippocampal CA1, CA3 and DG of APP/PS1 Tg mice increased dramatically compared with the WT mice (from 0.8422 ± 0.1892 to 5.012 ± 0.8202, P <0.001; from 0.9386 ± 0.1078 to 4.403 ± 0.6446, P <0.001; from 0.8750 ± 0.03305 to 6.096 ± 0.4614, P <0.001; from 1.119 ± 0.1738 to 4.730 ± 0.3491, P <0.001, respectively, Fig. 1A-E), which was reversed by administration of thioperamide (to 2.280 ± 0.4046, P <0.01; to 1.987 ± 0.4507, P <0.01; to 1.821 ± 0.2154, P <0.001; to 1.834 ± 0.3491, P <0.001, respectively, Fig. 1A-E). Similarly, we also observed a decrease of Iba1 + -cells in the cortex with administration of thioperamide compared with the vehicletreated APP/PS1 mice (from 3.262 ± 0.6586 to 1.596 ± 0.1891, P <0.01, Fig. 1A, F). Moreover, we examined the number of Iba1 + -cells around the plaques in the hippocampus and cortex. Results showed that thioperamide treatment signi cantly inhibited the clustering microglia around the plaques staining with Thio avin S (from 7.000 ± 0.9487 to 3.400 ± 0.5099 in hippocampus, P <0.05; from 6.400 ± 0.8124 to 2.600 ± 0.4000 in cortex, P <0.01, Fig. 1A, G, H). These data suggested that APP/PS1 Tg mice induced an increased activation of microglia in the hippocampus and cortex. Thioperamide treatment resulted in suppressed activation and decreased clustering microglia around plaques in APP/PS1 Tg mice.
Thioperamide induces a phenotypical switch in astrocytes in APP/PS1 Tg mice There is an A1/A2 nomenclature for astrocytes to characterize pro-in ammatory and anti-in ammatory effects respectively. Accordingly, toxic reactive astrocytes would be termed as "A1 astrocytes" and protective reactive astrocytes as "A2 astrocytes" [50]. Activated microglia secrets pro-in ammatory cytokines to induce A1 astrocytes, which lose most normal astrocytic functions but gain a new neurotoxic function [51]. In order to investigate the effect of thioperamide on the neurotoxic A1 astrocytes in AD, complement 3 (C3), a protein marker of A1 astrocytes was co-stained with GFAP. The results indicated that thioperamide signi cantly decreased the number of neurotoxic C3 + /GFAP + astrocytes in hippocampus (decreased to 51.95 ± 10.27% of vehicle group, P <0.05, Fig. 3C, D). In order to investigate the effect of thioperamide on protective A2 astrocytes, S100A10, a protein marker of A2 astrocytes was co-immunostained with GFAP. Interestingly, the number of A2 type protective astrocytes increased markedly in the thioperamide treated group compared with the vehicle group (increased to 211.1 ± 25.76% of vehicle group, P <0.01, Fig. 3E, F). Taken together, these results indicated that thioperamide induced an A1-to-A2 switch in astrocytes.
Thioperamide decreases pro-in ammatory cytokines production in APP/PS1 Tg mice The activation of microglia and A1 type astrocytes may promote the pathological process through releasing of pro-in ammatory cytokines [51,52], which could lead to neuronal damage and cognitive impairments. Moreover, the in ammatory cytokines in AD lesions are thought to lead to an increased accumulation of Aβ production [12]. Therefore, we investigated expression of pro-in ammatory factors associated with Aβ accumulation, including IL-1β, TNF-α and IL-4 in APP/PS1 Tg mice and WT mice. Our results showed that the levels of IL-1β and TNF-α increased markedly (from 380.1 ± 33.44 to 550.9 ± 26.67 in hippocampus of IL-1β, P <0.01, Fig Interestingly, the levels of in ammatory cytokines IL-1β and TNF-α were dramatically decreased in both hippocampus and cortex in thioperamide-treated APP/PS1 Tg mice (decreased to 390.9 ± 27.33 in hippocampus of IL-1β, P <0.01, Fig. 4A; decreased to 408.9 ± 29.31 in cortex IL-1β, P <0.01, Fig. 4B; decreased to 722.5 ± 54.57 in hippocampus of TNF-α, P <0.05, Fig. 4C; decreased to 727.5 ± 48.21 in cortex of TNF-α, P <0.05, Fig. 4D). Moreover, the anti-in ammatory mediator IL-4 increased signi cantly in both the hippocampus and cortex of thioperamide-treated APP/PS1 mice compared with the APP/PS1 Tg controls (increased to 485.1 ± 44.52 in hippocampus, P <0.05, Fig. 4E; decreased to 490.5 ± 32.10 in cortex, P <0.01, Fig. 4F). All these results suggested that thioperamide could effectively suppress the secretion of in ammatory cytokines in APP/PS1 Tg mice.
Thioperamide up-regulates the phosphorylated CREB in APP/PS1 Tg mice De cits in CREB signaling may be implicated in AD pathology through the detrimental effects of A [53,54]. However, activation of CREB pathway induces anti-in ammatory effects and ameliorated cognitive de cits [55]. Therefore, we tested the p-CREB protein level to clarify whether or not CREB, the H3R downstream signaling, is involved in the thioperamide mediated glial inactivation and anti-in ammatory effect in AD. In consistent with the previous reports, decreased intensity of p-CREB immunostaining was observed either in hippocampus (reduced to 67.05 ± 6.424% of WT, P <0.05, Fig. 5A, B) or in cortex (reduced to 51.07 ± 1.184% of WT, P <0.001, Fig. 5A, C) of APP/PS1 Tg mice compared to the WT mice. As expect, thioperamide up-regulated the p-CREB intensity remarkably (increased to 96.52 ± 5.964% of WT, P <0.05, Fig. 5A, B; increased to 86.70 ± 2.965% of WT, P <0.001, Fig. 5A, C) in APP/PS1 Tg mice. In addition, results of western blot also suggested that the expression of phosphorylated CREB decreased signi cantly (decreased to 60.52 ± 4.572% of WT, P <0.01, Fig. 5D, E) in hippocampus of APP/PS1 Tg mice, which was reversed markedly by thioperamide (increased to 86.70 ± 2.965% of WT, P <0.01, Fig. 5D, E).
It has been well known that the expression of pro-in ammatory cytokines requires NF-κB activation. Therefore, we investigated the effect of thioperamide on the activation of NF-κB. Interestingly, the results indicated that the expression of phosphorylated NF-κB p65 increased signi cantly (increased to 325.8 ± 48.64% of WT, P <0.01, Fig. 5D, F) in hippocampus of APP/PS1 Tg mice compared with the WT mice, which was compromised by administration of thioperamide (decreased to 160.1 ± 31.17% of WT, P <0.05, Fig. 5D, F). Together, all these results showed that AD induced a decreased expression of p-CREB, whereas up-regulated p-NF-κB p65 and treatment with thioperamide signi cantly activated the CREB signaling and suppressed the activation of NF-κB p65.
Activation of CREB is involved in the effects of thioperamide on glial reactivity and in ammatory response in APP/PS1 Tg mice In order to further investigate the involvement of CREB signaling in the effects of thioperamide on the activation of microglia and astrocytes, H89, the inhibitor of PKA/CREB was administrated to inhibit p-CREB. The results showed that the area of Iba1 + -cells in both hippocampus and cortex markedly increased in the thioperamide + H89 group compared with the thioperamide group in APP/PS1 Tg mice (from 2.370 ± 0.2887 to 4.534 ± 0.2449 in hippocampus, P <0.05; from 1.721 ± 0.06341 to 3.333 ± 0.2425 in hippocampus, P <0.01, Fig. 6A, C). Similarly, the inhibitory effect of thioperamide on reactivated astrocytes area was also reversed by administration of H89 in either hippocampus or cortex in APP/PS1 Tg mice (from 1.183 ± 0.2078 to 2.806 ± 0.4070 in hippocampus, P <0.01; from 0.7970 ± 0.1013 to 2.1060 ± 0.2257 in cortex, P <0.05, Fig. 6A, C). Above all, results suggested that activation of CREB was involved in the alleviated reactive glial cells offered by thioperamide in AD.

Thioperamide reduces Aβ burden through CREB activation in APP/PS1 Tg mice
The reactive glial cells in AD induce enhanced in ammatory cytokines release, which contributes to the accumulation of pathologic Aβ [12]. Therefore, we further examined the plaque deposition and soluble Aβ levels to investigate whether the up-regulated p-CREB by thioperamide was involved in the Aβ pathology. We observed a dramatic reduction in plaque burden in the hippocampus and cortex with thio avin-S staining in the thioperamide group compared with the vehicle group in APP/PS1 Tg mice, which was obviously reversed by administration of H89 (Fig. 3A, B). The quantitative analysis indicated that thioperamide down-regulated the area of Aβ-positive plaque burden in both hippocampus and cortex signi cantly (from 7.805 ± 0.8837% to 3.729 ± 0.2948% in hippocampus, P <0.01, Fig. 7C; from 10.33 ± 0.7218% to 4.584 ± 0.4692% in cortex, P <0.01, Fig. 7D). H89 treatment reversed the decreased Aβpositive plaque by thioperamide markedly (increased to 7.339 ± 0.8506% in hippocampus, P <0.05, Fig.  7C; increased to 8.726 ± 1.227% in cortex, P <0.05, Fig. 7D).

Discussion
In this study, we have shown that inhibition of H3R by thioperamide could modulate the gliosis and confers protection from pathological hallmarks and cognitive impairment in a mouse model of AD.
Speci cally, we found that thioperamide induced a lower reactivity of glial cells around Aβ plaques, attenuated neuroin ammation, reduced soluble Aβ, and ultimately resulting in a better cognitive outcome. Importantly, the effect of thioperamide on reactivity of glial cells, neuroin ammation, Aβ deposition and cognitive function in AD were compromised by treatment with H89, a PKA/CREB inhibitor, suggesting a CREB dependent mechanism.
Up to now, no therapy is available to block or slow down AD progression, and the mechanisms of AD are not fully understood [56]. Proliferation and activation of microglia concentrated around amyloid plaques, is a prominent feature of AD, which plays a key role in the pathogenesis of AD [57]. In AD pathogenesis, microglia activation plays a dual role: on one side, acute microglial activation in some experimental paradigms leads to decreased Aβ accumulation by increasing phagocytosis or clearance. In contrast, chronic activation of microglia contributes to neurotoxicity and synapse loss by triggering several proin ammatory cascades [58].
It is reported that whilst H3R are preserved in AD brain, higher density of H3R correlated with more severe dementia [29, 59,60]. H3R agonist (R)-α-methylhistamine (RAMH) induces impaired learning and memory [59], while H3R knockout enhanced cognitive function [61,62]. Recent reports show that H3R are highly expressed not only in neurons but also in microglia and astrocytes to regulate the in ammatory response [63,64]. However, the role of H3R in modulation of reactivity of gliosis in AD has not been reported. To further explore the effect of thioperamide on microgliosis in AD, experiments were carried out by using APP/PS1 Tg mice. Our results showed that APP/PS1 Tg mice induce an obvious activation of microglia, which was reversed by thioperamide treatment. Therefore, this study suggested that inhibition of H3R by thioperamide could inhibit the microgliosis in AD.
In addition to neurotoxicity, activated microglia can enhance the activation of adjacent astrocytes by releasing factors such TNFα and IL-1β that can further magnify neuronal injury [65]. Reactive astrocytes have traditionally been considered a uniform response mechanism of the brain to acute or chronic injury.
However, recent studies have demonstrated that reactive astrocytes could create either detrimental or bene cial conditions for damaged neurons, likely depending on the nature, severity, and duration of the pathological stimulus [66]. Neuroin ammation in AD could induce two different types of reactive astrocytes that we termed protective "A2" and neurotoxic "A1" [66,67]. Moreover, the abovementioned studies have provided evidence that many neurotoxic pathways are directly executed by astrocytes, with microglia perhaps playing a triggering or initiating role [65,66]. Interestingly, we found that thioperamide induced a lower reactivity of astrocytes in either hippocampus or cortex in APP/PS1 Tg mice. More importantly, a reduced expression of C3, a characteristic product of "A1" astrocytes [66] that contributes to AD pathology [68], while increased expression of S100A10, a characteristic product of "A2" astrocytes [69] that confers protection against AD pathology [66] were observed with treatment of thioperamide, indicating that inhibition of H3R contributes to a shift from toxic to protective astrocyte phenotypes in AD mice. Reactive astrocytes recruited to the Aβ plaques likely prolong the neuroin ammatory process by secreting proin ammatory molecules, such as IL-1β and TNF-α, which might lead to pathogenic chronic neuroin ammation and contribute to progression of pathology in AD [50,70,71]. As expect, we also found a lower reactivity of astrocytes around Aβ plaques. As an autoreceptor, H3R could enhance the release of histamine in a feedback way. Consistent with our results, histamine also attenuates astrogliosis in spinal cord injury [72] and cerebral ischemic injury [73], suggesting a modulation of astrocytic function by histaminergic system. Our results demonstrated for the rst time that inhibition of histamine H3R could suppress the activation of astrocytes and promote shift from toxic A1 to protective A2 astrocyte phenotypes in AD.
Deposition of Aβ triggers microgliosis and astrogliosis, both of which contribute to the pathogenesis of AD [7]. Gliosis is responsible for the secretion of pro-and anti-in ammatory cytokines (e.g. TNF-α, IL-1β, IL-4) [74]. More importantly, chronic neuroin ammation, manifested as gliosis and production of proin ammatory cytokines, is an invariant hallmark of AD [7], indicating that an in ammatory response further promoted the injury of the brain in AD. Another result reveals that H3R antagonist GSK247246 could inhibit gliosis and induces anti-neuroin ammatory effect under white matter injury in the brain [75].
Consistent with the results, here we demonstrated that except for activation of glial cells, enhanced levels of IL-1β and TNF-α and decreased level of IL-4 were also observed in AD. H3R inhibition by thioperamide suppressed in ammatory response in AD, as it was demonstrated by down-regulated levels of proin ammatory cytokines IL-1β and TNF-α, and up-regulated level of anti-in ammatory cytokine IL-4.
Conversely, there is report shows that activation but not inhibition of microglial H3R suppresses LPSinduced proin ammatory activities [63]. Concerning that acute moderate activation of microglia promotes anti-in ammation M2 microglia phenotypes, whereas chronic microglial activation exerts proin ammation M1 phenotypes in AD [76]. The differences might be attributed to the regulation of two different polarization of microglia by different factors.
To further understand the molecular mechanism underlying the effect of thioperamide, we examined the protein expressions of p-CREB and p-P65NF-κB. Reduced p-CREB level has been observed in AD mice overexpressing Aβ [53]. Therefore, de cit in CREB signaling may be implicated in AD pathology through the detrimental effects of Aβ [35, 54,77]. In addition, the decreased p-CREB expression is involved in the activation of glial cells and neuroin ammation in CNS [78][79][80]. CREB is also thought to be important for regulation of the in ammatory cytokines expression, and its dysfunction contributes to the pathogenesis of AD [37], implicating CREB might play an important role in mediating the effect of thioperamide on in ammation in AD. Moreover, stimulates of CREB pathway also enhances the microglial M2 polarization [81], and inhibits glia-mediated in ammation [39,82]. It is well acknowledged that inhibition of H3R induces CREB pathway activation [28]. Therefore, we investigated the effect of thioperamide on CREB phosphorylation in AD. As expect, a decreased level of p-CREB was observed in APP/PS1 Tg mice, and it was reversed by thioperamide treatment. H89, an inhibitor of protein kinase A (PKA)/CREB, compromised the alleviated effect of thioperamide on glial reactivity in AD, suggesting that thioperamide inhibited the activity of microglia and astrocytes in AD through activating CREB pathway. Both elevated antiin ammatory factors and decreased in ammatory factors by thioperamide were also reversed by administration of H89, indicating that thioperamide inhibited the in ammatory response through activating CREB signaling.
Expression of pro-in ammatory cytokines requires NF-κB activation and its nuclear translocation to interact with DNA [83]. Study indicates that activation of CREB induces reduced expression of p-NF-κB and decreased production of proin ammatory cytokines in brain injury [84]. We found a signi cant increase of p-P65 NF-κB phosphorylation in APP/PS1 Tg mice, suggesting that Aβ could activate NF-κB signaling pathway. However, thioperamide treatment suppressed the levels of p-P65 NF-κB in APP/PS1 Tg mice. These data indicated that the anti-in ammatory effects of thioperamide might be mediated by inhibition of NF-κB activity.
Glial cells have a variety of functions in the brain, ranging from immune defense against external and endogenous hazardous stimuli, regulation of synaptic formation, calcium homeostasis, and metabolic support for neurons. Their dysregulation contributes to the cognitive decline and pathological development of AD. One of the most important functions of glial cells in AD is the regulation of Aβ levels in the brain [85]. Microglia and astrocytes have been reported to play a central role as moderators of Aβ clearance and degradation [86,87]. In ammation could reduce microglial clearance of Aβ in APP/PS1 Tg mice [88]. Inhibition of activated microglia and astrocytes could suppress the in ammatory response and attenuate the accumulation of Aβ, and ultimately protect mice against synaptic dysfunction and a decline in spatial cognition [89]. Therefore, we further studied the effect of thioperamide on Aβ clearance and related mechanism. Interestingly, we found that thioperamide treatment decreased the level of Aβ either in hippocampus or cortex. More importantly, H89 reversed the effect of thioperamide on Aβ clearance, indicating a CREB dependent mechanism. Furthermore, we also valuated cognitive function based on the NOR test, Y maze test and MWM test. Our results showed that APP/PS1 Tg mice showed memory impairment and could be rescued by thioperamide administration. H89 treatment reversed the ameliorated cognition offered by thioperamide, suggesting CREB pathway was involved in the alleviated effect of thioperamide on cognitive dysfunction.

Conclusions
The present study indicates that H3R antagonist thioperamide improved cognitive impairment in APP/PS1 Tg mice via modulation of the CREB-mediated inhibited gliosis and in ammation, which contributed to Aβ clearance. These results uncovered a novel mechanism behind the therapeutic effect of thioperamide in AD and further provided an experimental basis for starting a clinical trial for H3R antagonists as a treatment for AD.