Plasmon-Activated Water Reduces Amyloid Burden and Improves Memory in Animals with Alzheimer’s Disease

With the great extension of the human lifespan in recent times, many aging diseases have inevitably followed. Dementia is one of the most-commom neurodegenerative aging diseases, in which inflammation-related Alzheimer’s disease (AD) is the most prevalent cause of dementia. Amyloid accumulation in the brain, which occurs before any clinical presentations, might be the first and key step in the development of AD. However, many clinical trials have attempted to remove amyloid from brains of AD patients, but none has so far been successful. Negatively charged plasmon-activated water (PAW) is created by resonantly illuminated gold (Au) nanoparticles (NPs), which reduce the hydrogen-bonded (HB) structure of water. PAW was found to possess anti-oxidative and anti-inflammatory effects. Herein, we report on an innovative strategy to retard the progression of AD by the daily consumption of PAW instead of normal deionized (DI) water. APPswe/PS1dE9 transgenic mice were treated with PAW or DI water from the age of 5 months for the next 9 months. Encouragingly, compared to DI water-treated mice, mice treated with PAW presented better memory performance on a test of novel object recognition and had a significantly lower amyloid burden according to 18F-florbetapir amyloid-PET and phosphorylated (p)-tau burden according to Western blotting and immunohistochemistry measurements. There were no obvious side effects in PAW-treated mice. Collectively, our findings support that PAW was able to reduce the amyloid and p-tau burden and improve memory in an AD mouse model. However, the protein levels of molecules involved in amyloid metabolism and oligomeric amyloid did not change. We propose that the effects of PAW of reducing the amyloid burden and improving memory function cannot be attributed to synthesis/degradation of amyloid-βprotein but probably in preventing aggregation of amyloid-β proteins or other mechanisms, including anti-inflammation. Further applications of PAW in clinical trials to prevent the progression of AD are being designed.


Results
Activity of scavenging free radicals by pAW. In experiments, the drinking water of AD mice was prepared every day using either fresh DI water or PAW. Thus, in the current experiments, we examined the ability of as-prepared and 1-day-aged PAW to scavenge free radicals compared to DI water to verify this ability. Figure S1 demonstrates the scavenging abilities of as-prepared and 1-day-aged PAW, compared to DI water, on active hydroxyl radicals. The four electro-spin resonance (ESR) splitting signals shown in Fig. S1A are characteristic of hydroxyl radicals 16 . Figure S1B shows the corresponding statistically significant results. Compared to DI water, the intensities of free radicals decreased by 36% (p = 0.0078) and 23% (p = 0.0026) in magnitude, respectively, with as-prepared and 1-day-aged PAW. Similarly, the ability of as-prepared and 1-day-aged PAW to scavenge radicals also demonstrated a positive effect on decreasing the corresponding ESR intensities of DPPH stable free radicals, as shown in Fig. S2. Compared to DI water, the intensities of free radicals decreased by 12% (p = 0.0020) and 9% (p = 0.022) in magnitude, respectively, with as-prepared and one-day-aged PAW.
Memory decline was rescued in APP/PS1 mice with PAW treatment. Memory decline is one of the clinical symptoms that occur in AD. To analyze the effects of PAW on memory decline in APP/PS1 mice, we conducted a novel object recognition (NOR) test every 3 months after treatment using 5-month-old APP/PS1 mice (Fig. 1A). Results of the NOR test did not differ between APP/PS1 mice fed PAW or no PAW (63.290% ± 3.284% vs. 63.123% ± 4.511%, p = 1.000) at the beginning. After 3 months of treatment, the performance of APP/PS1 mice given PAW had improved, but did not significantly differ from those mice without PAW (70.818% ± 2.402% vs. 64.200% ± 2.145%, p = 1.305). The recognition index of APP/PS1 mice without PAW had declined at 6 months (70.093 ± 1.153% vs. 56.260 ± 5.290%, p = 0.0178). After 9 months of treatment, the recognition index of APP/ PS1 mice without PAW had significantly declined (67.354 ± 2.575% vs. 49.516 ± 6.216%, p = 0.0008). Wild-type (WT) mice with or without PAW treatment showed no difference from 3 to 9 months of treatment ( Fig. 1B,C). These findings indicated that PAW treatment could prevent memory decline in APP/PS1 mice.

F-Florbetapir animal positron emission tomography (PET). Representative average PET images
(30~60 min after injection) of 18 F-florbetapir in brains of APP/PS1 and WT mice with/without PAW treatment were prepared. Intense and diffusely increased uptake of radioactivity was observed in the cortex, striatum, hippocampus, amygdala, and brain stem regions of APP/PS1 mice (Fig. 2C); the radioactivity uptake was less intense in APP/PS1 mice undergoing PAW treatment (Fig. 2D). 18 F-Florbetapir uptake in the abovementioned brain regions was lowest in WT C57BL/6 mice with/without PAW treatment ( Fig. 2A,B). Of note, radioactivity uptake levels were similar in background regions of all animals, such as the cerebellum, midbrain, and thalamus. Taking into account all data, significant increases in the 18 F-florbetapir uptake ratio were observed in the cortex   Fig. 3H) of APP/PS1 mice under PAW treatment compared to those without treatment. However, radioactivity uptake levels of the cortex and hippocampus in treated animals were still higher than those of control mice (Fig. 3). As expected, no significant difference in 18 F-florbetapir uptake in any brain region was observed between control groups with or without treatment (i.e., WT vs. WT + PAW) ( Table 1). These data suggest that PAW might reduce the formation of senile plaques in APP/PS1 mice.
Amyloid plaque and p-tau burden decreased in the hippocampus in APP/PS1 mice undergoing pAW treatment. The amyloid plaques and p-tau burden in the hippocampus was reported to be a biomarker of AD. To analyze the effects of PAW on the amyloid plaques and p-tau burden, we conducted thioflavin-S staining and immunocytochemistry in APP/PS1 mice at 9 months after treatment. The amyloid plaques and p-tau level had increased at APP/PS1 mice (Fig. 4). Compared to APP/PS1 mice, the p-tau level in APP/PS1 mice treated with PAW had significantly declined (Fig. S3). We further analyzed protein levels of p-tau in the cortex and hippocampus of APP/PS1 mice with or without PAW treatment for 9 months by Western blotting (Fig. 5A). The protein level of p-tau had significantly decreased in the hippocampus but not in the cortex of APP/PS1 mice treated with PAW compared to untreated mice (Fig. 5Ba,c). The ratio of p-tau/tau had also decreased in the hippocampus but not in the cortex (Fig. 5Bb,d). These data suggest that PAW might reduce the p-tau burden in the hippocampus of APP/PS1 mice. Comparison of relative 18F-florbetapir uptake levels in the cortex, striatum, hippocampus, amygdala, midbrain, thalamus, basal forebrain, hypothalamus, brain stem, and olfactory bulb (A~J) expressed as the region/cerebellum ratio for APP/PS1and wild-type mice with/without PAW treatment. The left line shows anatomical structures in the MRI/PET images. The right line shows the color bar of the standardized uptake value ratio (SUVR) compared to the cerebellum (SUVR cer ). (n = 5 or 6 per group, by a repeated-measures ANOVA and post-hoc analysis, *p < 0.05, **p < 0.01, ***p < 0.001).  www.nature.com/scientificreports www.nature.com/scientificreports/ Oligomeric amyloid level was not affected in APP/PS1 mice treated with PAW. To analyze the effect of the oligomeric amyloid level on APP/PS1 mice treated with PAW, we prepared an ELISA experiment. Oligomeric amyloid levels in the hippocampus (458.2 ± 55.31 pg/mg vs. 616.9 ± 81.73 pg/mg, p = 0.622) (Fig. 6A) and cortex (873.2 ± 87.5 pg/mg vs. 768.9 ± 181 pg/mg, p = 0.538) (Fig. 6B) of APP/PS1 mice showed no difference between mice with or those without PAW treatment. Further analyses of protein levels of neprilysin (Fig. 7A,C), PSEN1, BACE1, and APP (Fig. 7B,D) in APP/PS1 and WT mice showed that protein levels did not differ. Collectively, results suggested that the effect of PAW on APP/PS1 mice was not at the protein level of oligomeric amyloid.

Discussion
The real cause of AD is still unclear. Aβ containing 40 or 42 amino acids is the main component of extracellular senile plaques, which are the most important marker of AD pathology. There is compelling evidence supporting the crucial role of amyloid in the pathophysiology of AD, from genetic, in vitro, and in vivo studies 9 . Transgenic animals, like APPswe/PSEN1dE9 (APP/PS1) mice, reveal early and progressive accumulation of amyloid and develop memory decline, similar to symptoms in humans, from about 3 months old 22 . The amyloid PET scan has become the most important biomarker tool for diagnosing AD. However, the burden of amyloid in the brain is not perfectly correlated with the severity of AD, and almost all clinical trials of amyloid-cleaning therapy have failed 7,14 . Most studies showed that although the amyloid burden was reduced, dementic symptoms continued to progress 23 . The amyloid hypothesis was recently challenged 24 . To test the amyloid hypothesis, treatment in many ongoing trials was moved to the early or asymptomatic phase, to try and prevent mental decline 25 . In this study, we tested the effects of PAW on transgenic APP/PS1 mice. Based on prior research, significant amyloid accumulation appeared at 3 months of age, behavior and memory declines occurred a few months later, and abnormal amyloid PET results were found after 9 months of age in APP/PS1 animals 26 . Therefore, AD mice were given PAW from 5 months old, and memory surveys with the NOR test were conducted every 3 months. Results showed that memory continuously and obviously declined in AD mice without PAW, but memory decline was slower in AD mice treated with PAW. The memory performance became significantly different at 6 months after treatment (Fig. 1). Amyloid PET scans showed that the amyloid burden was higher in AD mice without PAW treatment, especially in the cortex, hippocampus, and hypothalamus (Figs 2 and 3). These findings support the close relationships of amyloid burdens in the cortex and hippocampus with memory declines. Amyloid-reduction therapy should potentially be effective. Results also implied that treatment to reduce the amyloid burden might begin in the early stage, before or soon after the accelerating accumulation of amyloid. PAW has been tested in a few diseases, and its antioxidative and anti-inflammatory effects were documented 19,20 . The mechanisms of its reduction of the amyloid burden by PAW could be in multiple manners. The antioxidative and anti-inflammatory actions could suppress APP, β-secretase, and γ-secretase gene expressions and reduce the Aβ formation [27][28][29] . Inhibition of polymerization from monomeric or oilgomeric Aβ is also a possible mechanism 30 . In our previous report, the amount of small water clusters created in PAW showed a slightly negative charged 20 . These small water clusters might prevent polymerization of Aβ in APP/PS1 mice. Enhancement of expressions of amyloid-degradation enzymes, including neprilysin, is another possible mechanism 31 . In this study, oligomeric Aβ protein levels were measured, but they showed no change with PAW treatment in APP/PS1 mice (Fig. 6). The synthesis of the Aβ protein is regulated by neprilysin, PSEN1, BACE1, and the APP. Levels of these proteins also did not change in our present data. We suggest that the effect of PAW on reducing the formation of senile plaques did not occur in the synthesis of the Aβ protein, but in preventing aggregations of Aβ proteins. Further studies on the reduction mechanism of PAW are needed. www.nature.com/scientificreports www.nature.com/scientificreports/ The tau protein which is also the another main component of pathological biomarker, intracellular neurofibrillary tangles, is considered to be another target for the diagnosis and treatment of AD 14 . Interactions between amyloid and tau are well studied 22 . Amyloid enhances tau phosphorylation, and amyloid alters tau splicing 32,33 . In this study, p-tau accumulated in the hippocampus of APP/PS1 mice with a high amyloid burden along with aging. APP/PS1 mice given PAW showed reduced p-tau deposition in the hippocampus compared to APP/PS1 mice without PAW, and such deposition is considered the beginning site of AD pathology, by comparison of APP/ PS1 mice without PAW (Fig. 4). These findings suggest that PAW can reduce p-tau accumulation, which might be attributed to reduced amyloid plaque formation in APP/PS1 mice. However, the detailed mechanisms still need further investigation.
Inflammation occurs in the pathologically vulnerable region of AD brains and contributes to AD pathogenesis. Interleukin (IL)-1 and IL-6 are immunoregulatory cytokines which are overexpressed in the AD brains 34 . Reduction of inflammation in AD brain might help to delay the progression of AD. In our present data, we analyzed the inflammation by measuring protein levels of IL-1β and IL-6 using an ELISA assay. Although the protein levels of IL-1β and IL-6 were not significantly reduced in PAW-treated APP/PS1 mice, decreased values   www.nature.com/scientificreports www.nature.com/scientificreports/ Methods Animals. APPswe/PSEN1dE9 (APP/PS1) transgenic mice were provided by Dr. Rita P-Y Chen of the Institute of Biological Chemistry, Academia Sinica (Taipei, Taiwan) 35 . The genotype of the transgenic mice was analyzed by a polymerase chain reaction (PCR) with Hot-NaOH-extracted genomic DNA from the tail. We used 5-month-old male APP/PS1 mice in this study and their littermates as controls. The mice were individually kept in ventilated cages with feeding racks and water bottles were attached to the front panel of the cage, which allowed animals ad libitum access to food and water. Mice were maintained on a 12-h light/dark cycle (12L/12D; lights on at 07:00 and lights off at 19:00). The food and deionized (DI) water or PAW were freely available, and the water was changed daily. Six (APP/PS1) mice for each group were fed with PAW or DI water for 9 months, beginning from the age of 5-months. All the experiments were approved by the Experimental Animal Review Committee of Taipei Medical University, and all methods were performed in accordance with guidelines of the National Institute of Health, Taiwan.
Full Methods and any associated references are available in the online version of the paper.