- Email: [email protected]
Experimentally induced diabetes worsens neuropathology, but not learning and memory, in middle aged 3xTg mice
Accepted Manuscript Title: Experimentally Induced Diabetes Worsens Neuropathology, but not Learning and Memory, Author: Emi Hayashi Park Bria N. Ozment Chelsea M. Griffith Haiying Zhang Peter R. Patrylo Gregory M. Rose PII: DOI: Reference:
S0166-4328(16)30295-9 http://dx.doi.org/doi:10.1016/j.bbr.2016.05.020 BBR 10199
To appear in:
Behavioural Brain Research
Received date: Revised date: Accepted date:
9-3-2016 6-5-2016 8-5-2016
Please cite this article as: Park Emi Hayashi, Ozment Bria N, Griffith Chelsea M, Zhang Haiying, Patrylo Peter R, Rose Gregory M.Experimentally Induced Diabetes Worsens Neuropathology, but not Learning and Memory,.Behavioural Brain Research http://dx.doi.org/10.1016/j.bbr.2016.05.020 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Experimentally Induced Diabetes Worsens Neuropathology, but not Learning and Memory, in Middle Aged 3xTg Mice
Emi Hayashi Parkb,c,d, Bria N. Ozmenta,c,d, Chelsea M. Griffithb,c, Haiying Zhanga,c, Peter R. Patryloa,b,c and Gregory M. Rosea,b,c
Departments of aAnatomy and bPhysiology, School of Medicine and cNeuroscience Research Center Southern Illinois University Carbondale, IL 62901, USA dThese
authors contributed equally to this work.
Corresponding author: Gregory M. Rose, Ph.D. Department of Anatomy, MC6503 SIU School of Medicine 600 Agriculture Drive Carbondale, IL 62901 USA Email: [email protected]
1
HIGHLIGHTS
Streptozotocin administration to 12-month old 3xTg mice chronically elevated blood glucose levels. Diabetic mice did not show performance deficits in water maze or fear conditioning tasks. Beta-amyloid containing plaque numbers were substantially increased in diabetic mice.
Abstract: Alzheimer’s disease (AD) is the primary cause of dementia in the elderly. The cause of the disease is still unknown, but amyloid plaques and neurofibrillary tangles in the brain are thought to play a role. However, transgenic mouse models expressing these neuropathological features do not show severe or consistent cognitive impairments. There is accumulating evidence that diabetes increases the risk for developing AD. We tested the hypothesis that experimentally induced diabetes would exacerbate cognitive symptoms in a mouse model of AD. Diabetes was induced in 12-month old 3xTg mice using streptozotocin (STZ; 90 mg/kg, i.p., on two successive days). Hyperglycemia was verified by sampling blood glucose levels. Three months after injection (at 15 months of age), the mice were behaviorally tested in the Morris water maze and contextual fear conditioning. Subsequently, the hippocampal region was examined using immunohistochemistry (6E10 antibody for amyloid) and immunoblotting (AT8 antibody for phosphorylated tau). No differences were found in learning or memory between the vehicle-treated control and STZ-treated groups. A significant increase in the number of amyloid-positive plaques was observed in the subiculum of STZ-treated mice; very few plaques were seen in other hippocampal regions in either group. No differences in AT8 load were
2
observed. These results reinforce that amyloid plaques, per se, are not sufficient to cause memory impairments. Further, while diabetes can enhance this aspect of brain pathology, the combination of disrupted glucose metabolism and the transgenes is still not sufficient to cause the severe cognitive impairments associated with clinical AD. Keywords: 3xTg, triple transgenic, Alzheimer, diabetes, streptozotocin
1. Introduction It is common knowledge that Alzheimer’s disease (AD) is a debilitating cognitive disorder and the leading cause of dementia in the elderly. The brains of AD patients contain extracellular senile plaques composed of aggregated beta amyloid (Aβ) and neurofibrillary tangles composed of hyperphosphorylated tau protein; in later stages of the disease, substantial atrophy and neuronal death are observed. There is currently no cure for AD, and marketed symptomatic treatments have, at best, limited and temporary effects. The development of transgenic mouse models containing mutant human genes regulating Aβ synthesis was heralded as a breakthrough because it offered the opportunity to examine the genesis of major aspects of the amyloid plaques, and, in later models, the neurofibrillary tangles composed of hyperphosphorylated tau protein. A reasonable expectation was that the presence of these neuropathologies would result in the profound cognitive deficits that are the major symptom of AD. Unfortunately, this has not turned out to be the case. The cognitive deficits in the AD models are usually mild, are often restricted to a particular measure on a particular behavioral task, and are not always reproducible either within or between
3
laboratories. This outcome is seen despite the presence of heavy plaque loads, particularly in hippocampal and neocortical regions. Further, while many experimental treatments have been shown to improve learning or memory in transgenic AD models, none of these that have progressed to human clinical trials have been successful [1]. In addition to genetic abnormalities, several factors have been identified that likely increase the risk of developing AD; among these is diabetes [2-5]. Type 1 diabetes, which results in the loss of insulin-producing pancreatic beta cells, generally increases the risk of cognitive dysfunction [6]. Type 2 diabetes, a disease traditionally associated with aging, and which results initially in reduced sensitivity to insulin and later in reduced insulin production, is also associated with increased risk of developing dementia, including AD (e.g., [7, 8]). The basis for a relationship between diabetes and AD is still unclear, but could involve chronic hyperglycemia, impaired insulin signaling, or both. Many downstream mechanisms have been proposed, including increased amyloid and tau pathology, elevated brain inflammation, decreased dendritic spine density, reduced synaptic plasticity, and neuron loss [9, 10]. However, treatment with insulin, or insulin receptor sensitizing drugs, has been shown to improve cognitive function in both animal models of AD [9, 11, 12] and in AD patients [13, 14]. In the current study we explored the possibility that a 3-month period of experimentally-induced diabetes would impair hippocampus-dependent learning and memory in 3xTg mice [15] when the treatment was administered prior to the age at which impairments are normally observed [16].
2. Materials and Methods
4
2.1. Animals The subjects in this study were male 3xTg mice raised in our colony from breeding stock provided by Dr. Frank LaFerla [15]. The animals were housed in group cages in a climate controlled vivarium (25° C) at Southern Illinois University Carbondale. Mice had food (Purina LabDiet® 5001, Framingham, MA) and water available ad libitum and were kept on a 14/10-hour light/dark schedule (lights on 0600 – 2000). All experimental treatments and behavioral tests were conducted during the light part of the cycle. Experiments were performed in compliance with the Institutional Animal Care and Use Committee (IACUC) of Southern Illinois University Carbondale according to guidelines set forth by the National Institutes of Health. The mice were 12 months old when the experiments began. This age was chosen because, in our colony, the onset of behavioral impairments occurs later than in the LaFerla lab colony(e.g., [17, 18]; significant deficits in the water maze task are not consistently observed in our 3xTg mice until they are at least 15 months old [16]. 2.2. Induction of Diabetes Diabetes mellitus was induced after overnight fasting by the chemical ablation of pancreatic beta cells using streptozotocin (STZ). Two injections (90 mg/kg, i.p. in sodium citrate buffer, pH 4.5; injection volume 10 ml/kg) were given at a 24 hour interval. Control mice received two injections of the sodium citrate buffer alone on the same schedule. A human OTC glucometer (TRUEtrack, Nipro Diagnostics, Fort Lauderdale, FL) was used to measure blood glucose levels obtained from a tail prick. Hyperglycemia (blood sugar > 250 mg/dL) was confirmed one week following STZ injections in the experimental treatment group. Blood glucose concentrations and body weight were measured weekly for four weeks, and once more prior to sacrifice to confirm that the animals remained
5
hyperglycemic and were not clinically ill (as determined by overall body weight and normal grooming and nesting behaviors) in spite of the presence of diabetes. Cage bedding was changed every other day because of the polyuria present in the diabetic mice. 2.3. Behavioral Testing 2.3.1. Morris Water Maze The Morris water maze is well established test of hippocampus-dependent spatial learning and memory [19]. Mice in this study were tested at 15 months of age, three months after having received STZ or vehicle injections. The pool consisted of a circular tank (1.35 m diameter) painted white and filled with water made opaque by non-toxic white tempera paint. Water temperature was maintained at approximately 21.5 degrees Celsius. For the hidden platform task, a circular hidden platform (10 cm diameter) was located in the “center” of one quadrant of the maze and submerged one cm below the surface of the water. For the visible platform task, an extension was added to the hidden platform so that a portion of the platform was visible above the water level. The four walls of the room were decorated with distinct visual cues to clearly differentiate between them. For the first three consecutive training days (Days 1-3) mice received three daily trials using the visible platform. Visible platform training allows for the assessment of sensorimotor abilities and motivation to escape the water independent of spatial learning ability. On each training trial, an animal was placed into the water at the edge of the tank, facing the wall, with the start locations varying pseudorandomly (N, S, & E quadrants, with the platform in the W quadrant) and allowed to swim until they reached the escape platform. If a mouse did not reach the platform by a maximum of 90 seconds, the animal was hand guided to it. Once on the platform, mice remained there for 15 seconds before being removed for a minimum 20-minute inter-trial interval in their home cage. On
6
Day 4, hidden platform training began and lasted for five consecutive days (Days 4-8), also with three trials/day, a pseudorandom start position for each trial, and a 90 second maximum trial duration. Twenty-four hours after completing hidden platform training, a probe trial (Probe 1) was conducted with no platform in the pool. Mice were allowed to swim for 60 seconds before being removed from the water. A second memory probe test (Probe 2) was performed two weeks later. Performance was recorded using a camera mounted above the maze with an interface to a computerized tracking system (AnyMaze, San Diego Instruments, San Diego, CA). AnyMaze measured swim time and distance for the learning trials, and platform location crossings, time in the annulus-40 (a 40 cm diameter circle centered on the platform location), and total swim distance for the probe trials. 2.3.2. Fear Conditioning After the completion of water maze testing the mice were allowed to rest for one week before undergoing a fear conditioning procedure. Each mouse was placed in a soundattenuating chamber (30 x 25 x 20 cm, l x w x h), the floor of which was made of parallel 2-mm diameter stainless steel rods spaced 8 mm apart (Med Associates, St. Albans, VT). The floor was wired to a scrambled shock generator (also Med Associates). A red light (off during training and context memory testing) was located on one wall of the chamber. A video camera was mounted in front of the clear door of the chamber for video monitoring and recording of the animals’ behavior. On Day 1 of this paradigm, mice were placed in the chamber and allowed to explore for 60 seconds. After this they were exposed to a 2 kHz, 90 dB tone for 30 seconds, which coterminated with a 2-second, 0.6 mA foot shock. Sixty seconds later the auditory cue was repeated for 30 seconds, ending with a second foot shock. The mice remained in the chamber
7
for an additional 90 seconds before being removed and returned to their home cage. A 70% alcohol solution was used to clean the chamber before and after each subject. Twenty-four hours later (Day 2) the mice were returned to their training chamber and their behavior was monitored for 5 minutes. Behavioral freezing (immobility except for respiration) to the context was measured using Freeze Frame video tracking and software (ActiMetrics, Willmette, IL). On Day 3, the chamber was altered to mask the conditioning context by covering the grid floor, adding a drop of citrus scent in a weighing dish underneath the floor, turning on the red light and placing novel objects in the chamber. Mice placed in the chamber demonstrated that they did not recognize the training context of the box by immediately exploring the altered environment. After three minutes of exploration, memory for the association between the auditory cue and the foot shocks given on Day 1 was assessed by monitoring behavioral freezing during continuous delivery of the tone for three additional minutes. Unfortunately, the data from the cued memory assessment were lost to due to a computer problem so are not included in the analysis. 2.4 Aβ Immunohistochemistry Within one week after behavioral testing was completed, the mice were anesthetized with isoflurane and intracardially perfused with cold (4 °C) phosphate-buffered saline (PBS). The brains were removed and bisected along the sagittal sulcus; the hippocampus was dissected from the right half of the brain, snap frozen using dry ice powder and stored at -80°C for later biochemical analysis (see below). The left half of the brain was placed in a solution of cold 4% paraformaldehyde in PBS for 2 days, then placed in a 30% sucrose solution and stored at 4 °C until it was sectioned. Forty-
8
micron-thick sagittal sections through the hippocampal region were cut using a freezing microtome and stored in a cryoprotectant solution [20] until they were processed for immunohistochemistry. Sections were treated with 1% H2O2 in PBS for 30 minutes to quench endogenous peroxidase, followed by a pre-incubation in 5% normal goat serum in PBS with 0.3% Triton X100 for 1 hour. Antigen retrieval was used for Aβ labeling using 50% formic acid in PBS for 30 minutes at room temperature prior to bleaching of endogenous peroxidase. Sections were then incubated overnight with the primary antibody (anti-Aβ 6E10, 1:1000; BioLegend SIG39320, Dedham, MA) in blocking serum (4°C), then rinsed and processed using the reagents and protocol supplied with Vectastain elite ABC kit (Vector Laboratories, Burlingame, CA). Immunoreactive product was visualized by incubation in a solution containing 0.05% diaminobenzidine and 0.003% H2O2. Three 10-minute washes with PBST were performed between all incubations. Sections were mounted on slides, allowed to air-dry, delipidated with butanol and xylene, and then cover slipped. 2.5 p-Tau Western Blotting Hippocampal samples were homogenized by sonication in 700 µl of protein isolation buffer (150 nM NaCl, 0.1% SDS, 0.5% sodium deoxycholate, 1% Igepal (NP-40), 50 mM Tris, pH 8.0, 20 mM NaF + mixed protease and phosphatase inhibitors). The homogenates were then centrifuged at 12,000 RPM for 10 minutes at 4˚C. Supernatants were collected and protein concentrations determined using the bicinchoninic acid method (Pierce BCA Protein Assay Kit, ThermoFisher Scientific, Waltham, MA). Samples (50 μg/ml) were loaded onto 10% SDS-Page gels, transferred to PDVF membranes and probed with a p-Tau antibody (mouse monoclonal Phospho-Tau (Ser396),
9
1:1000; Cell Signaling, Danvers, MA). Beta-actin was used as a control protein (rabbit monoclonal Anti-β-actin, 1:500, Cell Signaling, Danvers, MA). Appropriate secondary antibodies were used (goat -anti-rabbit IRDye 700 DX and goat-anti-mouse IRDye 800 CW, 1:10,000; Rockland, Gilbertsville, PA) for densitometric quantification of protein bands using an Odyssey Licor infrared imaging system and were normalized to β-actin. For comparison between groups the mean for the control group was normalized to one. 2.6. Statistical Analysis All data are presented as mean ± s.e.m. Data analyses were performed using Prism 5 (v. 5.03; GraphPad Software, La Jolla, CA). Differences between multiple means were analyzed by ANOVA. Differences between two means were analyzed using two-tailed t-tests. Statistical significance was defined as p < 0.05.
3. Results 3.1 Effect of STZ Blood Glucose Levels With the expectation that chronically elevated blood glucose would cause mortality, 15 mice received STZ injections while 10 mice received the vehicle. Prior to treatment, glucose levels for control mice were 116.8 ± 4.4 mg/dL (mean ± s.e.m.) and 111.3 +/- 6.8 mg/dL for the group to receive STZ (p > 0.1, t-test). One week after STZ treatment, 13 of 15 mice had blood glucose levels of > 250 mg/dl, meeting our prospective criterion for diabetes. The two remaining mice were eliminated from the study. By three months later, when the surviving mice were 15 months old, one mouse in the control group had died; the mean blood glucose level for the remaining 9 animals was 135.6 +/- 9.4 mg/dL, not significantly different from the initial value (p > 0.1, paired t-test). By contrast, the mean value for the 13 mice in the STZ-treated group was significantly increased to 10
492.9 +/- 38.7 mg/dL (p < 0.001, paired t-test; Fig. 1A). Body weights for Control and STZ mice were comparable prior to treatment (33.0 ± 1.2 vs. 33.1 ± 0.9 gm, p > 0.1, t-test). Body weights in both groups decreased slightly, but not significantly, during the following 3 months (weights prior to behavioral testing: Control – 31.4 ± 0.9 gm; STZ – 31.5 ± 1.3 gm; p > 0.1, paired t-tests; Fig. 1B). Thus, despite having extremely high blood glucose levels, the STZ group did not show increased morbidity or mortality. 3.2 Water Maze Testing The mice were tested in the water maze at 15 months of age, 3 months after the induction of diabetes in the STZ-treated group. Control and STZ-treated animals behaved comparably in all phases of the water maze task. Swim times to the visible platform improved over the 3-day training period in both groups. Mean latencies on Day 3 were: Control – 10.3 ± 2.4 sec; STZ – 8.0 ± 1.3 sec; p > 0.1, t-test). Similarly, mean swim distances to locate the hidden platform decreased significantly over the 5-day training period (p < 0.001, repeated measures ANOVA). However, no difference was observed between groups (p > 0.05 for daily comparisons, Bonferroni post-hoc tests; Fig 2A). Performance during the probe trial given 24 hours later (Probe 1) was also not different between groups (Platform location crosses: Control – 3.78 ± 1.05, STZ – 4.23 ± 0.63, p > 0.1, t-test, Fig 2B; Swim time in Annulus-40: Control – 12.23 ± 1.83 sec , STZ – 11.49 ± 1.51 sec, p > 0.1, t-test, Fig 2C; Swim distance during probe trial: Control – 10.63 ± 0.44 m, STZ – 11.28 ± 0.31, p > 0.1, t-test, Fig 2D). Performance during Probe 2, given 14 days later, produced similar results. Platform location crosses were not different between probes or between groups (Probe 2 platform location crosses: Control – 3.56 ± 0.67, STZ – 2.85 ± 0.59, p > 0.1, ANOVA, Fig 2B). Swim time in the annulus-40 declined for both groups between Probe 1 and Probe 2 (p = 0.01, ANOVA), but the degree was
11
comparable, and there was no difference between groups for Probe 2 (Control – 8.90 ± 1.12 sec , STZ – 7.25 ± 1.01 sec, p > 0.1, ANOVA, Fig 2C). Total swim distances during Probe 2 reached significance (swim distance during Probe 2: Control – 10.76 ± 0.22 m, STZ – 11.87 ± 0.27, p > 0.01, ANOVA, Fig 2D), but there was no significant difference for either group between the two probe trials. 3.3 Fear Conditioning Both Control and STZ-treated mice demonstrated significant context memory 24 hours after training, i.e., an increase in freezing compared to baseline levels (p < 0.001, ANOVA; Fig 3A). The STZ group showed less activity (greater percent freezing) than controls during the 90-second period prior to the first tone-shock delivery (Control – 2.24 ± 1.49, STZ – 17.17 ± 6.71; p < 0.05, t-test with Welch correction for unequal variance), although freezing during the memory test was not different between groups (Control – 57 ± 11.16, STZ – 78.48 ± 6.70, p > 0.09, t-test). Thus, the results were analyzed by computing difference scores (freezing after training minus freezing before training) for each animal. This approach revealed that there was not a significant difference in the amount of freezing to context between the Control and STZ groups (Control – 54.51 ± 10.67, STZ – 61.31 ± 8.22, p > 0.1, t-test). 3.4 Amyloid Immunohistochemistry Amyloid staining, visualized by 6E10 immunohistochemistry, was present in the hippocampus, amygdala, and deep layers of neocortex. Labeling was primarily confined to neuronal cell bodies. Amyloid-containing plaques were occasionally observed in the hippocampus proper, but were more concentrated in the subiculum (Fig 4A). Counts were made of the plaques seen in all sections containing the subiculum, and then averaged to provide a number/section for each mouse. Comparison of the mean number of plaques/section revealed that there were significantly more Aβ-
12
positive plaques in the subiculum of STZ-treated mice compared to Controls (plaques per histological section: Control – 2.3 ± 0.8, STZ – 8.6 ± 1.8, p < 0.01, t-test; Fig 4B). 3.5 p-Tau Western Blotting To assess whether STZ treatment caused changes in levels of phosphorylated tau in the hippocampus, western blotting for p-Tau (Ser396) was performed. Levels of phosphorylated tau protein in the hippocampus were normalized to β-actin, the control mean was normalized to one, and group means were compared. No difference in hippocampal p-tau was found between Control and STZ-treated mice (Control normalized optical density – 1.00 ± 0.10, STZ normalized optical density – 1.15 ± 0.15, p > 0.1, t-test; Fig 5). 4. Discussion The goal of this study was to test the hypothesis that inducing frank diabetes in 3xTg mice would significantly worsen learning and memory deficits. We found that a 3-month period of profoundly elevated blood glucose levels, produced by STZ injections, did not affect the behavioral performance of middle-aged (12 months old at the start of the experiment) 3xTg mice in either the Morris water maze or contextual fear conditioning tasks when the animals were tested at 15 months of age. However, STZ-treated mice had a greater than 3-fold increase in the number of Aβ-containing plaques in the subiculum labeled by 6E10 immunocytochemistry. Hippocampal levels of phosphorylated tau were not significantly altered by STZ treatment. The seminal observation of Alois Alzheimer of profound brain pathology in patients with debilitating cognitive deficits logically gave rise to the hypothesis that what were later identified as Aβ-containing plaques and neurofibrillary tangles containing hyperphosphorylated
13
tau were responsible for the symptoms of AD [21]. In particular, the “amyloid hypothesis” [22] has provided an overarching influence on developing potential therapeutics for the disease. However, further research, and an unfortunate lack of success with clinical drug candidates designed to reduce behavioral symptoms, are leading to a reevaluation of the role of Aβ in AD [23-25]. This line of thinking has been reinforced by the lack of correspondence between the severity of cognitive deficits in AD patients and in the many transgenic mouse models now available to study amyloid and tau pathology. (See [1, 26, 27] for a range of perspectives.) It is certainly premature to abandon the idea that amyloid and tau are involved in important ways in AD. However, at this point it is reasonable to explore other factors that could, in combination with amyloid and tau pathology, contribute to the appearance of behavioral symptoms. Emerging evidence from epidemiological studies suggests that diabetes is a risk factor for developing AD [2, 28, 29] (but see [30]). At this point it is unclear whether hyperglycemia, insulin resistance, or other factors are the most important for this linkage [31, 32]. Systemic STZ, through the destruction of pancreatic beta cells, results in both hyperglycemia and disruption of insulin receptor signaling [33-36]. We found no significant differences in learning or memory measures assessed approximately 3 months after STZ treatment, despite a substantial and prolonged elevation in blood glucose levels. While intraperitoneal administration of STZ to wild-type rodents has been reported to result in deficits in the water maze task, significant weight loss in the treated animals is a confounding factor [37, 38]. Inducing diabetes in transgenic AD mouse models has produced mixed results in behavioral tests. Intraperitoneal STZ alone caused profound deficits in the Barnes maze task in wild-type
14
mice that were debatably worsened in hAPP transgenic animals [39]. In APP-PS1 mice, learning and memory in the water maze were also somewhat worsened in STZ-treated animals [40]. More robust effects of STZ treatment have been reported for the mAβAPP model [32]. Dietary manipulations to induce a type II-like diabetic state in transgenic AD mouse models have also been shown to exacerbate learning and memory impairments (e.g., [41-44]). To our knowledge, the present study is the first to examine the effect of inducing systemic diabetes in the 3xTg model. While treatment with STZ did not affect behavioral measures in our 3xTg mice, it did result in an increased number of amyloid plaques in the subiculum. Enhanced amyloid pathology has also been reported in diabetic AD patients [45], as well as in aged diabetic nonhuman primates [46] and in hAPP [39] and APP/PS1 [47] transgenic mice treated with STZ. (However, reduced amyloid deposition in the brains of APP/PS1 mice after STZ treatment has also been reported [40], as well in APP/PS1 mice crossed with diabetic db/db mice [10].) The mechanism responsible the elevated amyloid plaques is unclear, but could involve increased βamyloid production [48, 49], a reduction in insulin-degrading enzyme which also degrades Aβ [50], or through generation of advanced glycation end products that promotes Aβ aggregation [47, 51, 52]. In our study, levels of phosphorylated tau in the hippocampus were not elevated following STZ treatment. This finding contrasts with those reported examining the effects of STZ in other transgenic mouse models that selectively overexpress human tau (Tg601 [53] and pR5 [54]). While reason for this discrepancy is unclear, it may be due to differences in the mouse models (including the presence of multiple transgenes in 3xTg mice), ages or brain
15
regions examined, or that our data were normalized to an already elevated baseline in our 15month old mice. The results of this study support the emerging consensus that neuropathological hallmarks of AD do not seem to bear a direct relationship to the debilitating cognitive impairments associated with the disease [24, 25]. In addition, our work emphasizes this discrepancy in a transgenic mouse model of AD. Since there is strong evidence that elevated brain amyloid deposition is a risk factor for AD, albeit one that appears 1-2 decades before the onset of dementia [24, 55], it is possible that the mild behavioral deficits seen in transgenic AD models are due merely to the relatively short lives of the animals [56]. However, a stronger case could be made that the lack of substantial neuron loss observed in existing models of AD is the key factor [56, 57]. While neuron numbers or density were not quantified in our study, the combination of STZ treatment and multiple human transgenes was not sufficient to obviously affect these variables. Interestingly, a recent report documented contextual fear conditioning deficits and neuron loss in the subiculum of 5xFAD transgenic mice [58]. Elimination of microglia by inhibiting CSFR1 in these mice ameliorated both neuron loss and the behavioral impairment, but had no effect on brain amyloid levels or plaque load. This result further reinforces the dichotomy between amyloid and cognitive deficits, and suggests a possible path forward to understanding the mechanism(s) responsible for neuronal death in the AD brain. Conflicts of Interest The authors declare no conflict of interest. Acknowledgments
16
This work was supported by a predoctoral fellowship from Southern Illinois University to E.H.P., a REACH award to B.O., and the SIU Neuroscience Research Center. H.Z. received support from Hainan Medical College and the China Scholarship Council.
References [1] Windisch M. We can treat Alzheimer's disease successfully in mice but not in men: failure in translation? A perspective. Neurodegenerative diseases. 2014;13:147-50. [2] Ott A, Stolk RP, van Harskamp F, Pols HA, Hofman A, Breteler MM. Diabetes mellitus and the risk of dementia: The Rotterdam Study. Neurology. 1999;53:1937-42. [3] Akomolafe A, Beiser A, Meigs JB, Au R, Green RC, Farrer LA, et al. Diabetes mellitus and risk of developing Alzheimer disease: results from the Framingham Study. Archives of neurology. 2006;63:1551-5. [4] Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer's disease: an analysis of population-based data. Lancet Neurol. 2014;13:788-94. [5] Cheng G, Huang C, Deng H, Wang H. Diabetes as a risk factor for dementia and mild cognitive impairment: a meta-analysis of longitudinal studies. Intern Med J. 2012;42:484-91. [6] Brands AM, Biessels GJ, de Haan EH, Kappelle LJ, Kessels RP. The effects of type 1 diabetes on cognitive performance: a meta-analysis. Diabetes Care. 2005;28:726-35. [7] Gudala K, Bansal D, Schifano F, Bhansali A. Diabetes mellitus and risk of dementia: A metaanalysis of prospective observational studies. J Diabetes Investig. 2013;4:640-50. [8] Chatterjee S, Peters SA, Woodward M, Mejia Arango S, Batty GD, Beckett N, et al. Type 2 Diabetes as a Risk Factor for Dementia in Women Compared With Men: A Pooled Analysis of 17
2.3 Million People Comprising More Than 100,000 Cases of Dementia. Diabetes Care. 2016;39:300-7. [9] Searcy JL, Phelps JT, Pancani T, Kadish I, Popovic J, Anderson KL, et al. Long-term pioglitazone treatment improves learning and attenuates pathological markers in a mouse model of Alzheimer's disease. Journal of Alzheimer's disease : JAD. 2012;30:943-61. [10] Infante-Garcia C, Ramos-Rodriguez JJ, Galindo-Gonzalez L, Garcia-Alloza M. Long-term central pathology and cognitive impairment are exacerbated in a mixed model of Alzheimer's disease and type 2 diabetes. Psychoneuroendocrinology. 2016;65:15-25. [11] Papadopoulos P, Rosa-Neto P, Rochford J, Hamel E. Pioglitazone improves reversal learning and exerts mixed cerebrovascular effects in a mouse model of Alzheimer's disease with combined amyloid-beta and cerebrovascular pathology. PloS One. 2013;8:e68612. [12] Yamanaka M, Ishikawa T, Griep A, Axt D, Kummer MP, Heneka MT. PPARgamma/RXRalphainduced and CD36-mediated microglial amyloid-beta phagocytosis results in cognitive improvement in amyloid precursor protein/presenilin 1 mice. Journal of Neuroscience. 2012;32:17321-31. [13] Craft S, Asthana S, Cook DG, Baker LD, Cherrier M, Purganan K, et al. Insulin dose-response effects on memory and plasma amyloid precursor protein in Alzheimer's disease: interactions with apolipoprotein E genotype. Psychoneuroendocrinology. 2003;28:809-22. [14] Craft S, Baker LD, Montine TJ, Minoshima S, Watson GS, Claxton A, et al. Intranasal insulin therapy for Alzheimer disease and amnestic mild cognitive impairment: a pilot clinical trial. Archives of Neurology. 2011;69:29-38.
18
[15] Oddo S, Caccamo A, Shepherd JD, Murphy MP, Golde TE, Kayed R, et al. Triple-transgenic model of Alzheimer's disease with plaques and tangles: intracellular Abeta and synaptic dysfunction. Neuron. 2003;39:409-21. [16] Macklin LN, Cai Y, Kanak DJ, Ebersole J, Corbin AE, Womack A, et al. Impaired glucose tolerance precedes neuroanatomical identification of Aβ and hyperphosphorylated tau accumulation in the 3xTg model of Alzheimer’s disease. Society for Neuroscience Abstracts. 2009;528.26. [17] Billings LM, Oddo S, Green KN, McGaugh JL, LaFerla FM. Intraneuronal Abeta causes the onset of early Alzheimer's disease-related cognitive deficits in transgenic mice. Neuron. 2005;45:675-88. [18] Billings LM, Green KN, McGaugh JL, LaFerla FM. Learning decreases A beta*56 and tau pathology and ameliorates behavioral decline in 3xTg-AD mice. Journal of Neuroscience. 2007;27:751-61. [19] Morris RGM, Garrud P, Rawlins JNP, O'Keefe J. Place navigation is impaired in rats with hippocampal lesions. Nature. 1982;297:681-3. [20] Watson RE, Jr., Wiegand SJ, Clough RW, Hoffman GE. Use of cryoprotectant to maintain long-term peptide immunoreactivity and tissue morphology. Peptides. 1986;7:155-9. [21] Small DH, Cappai R. Alois Alzheimer and Alzheimer's disease: a centennial perspective. Journal of Neurochemistry. 2006;99:708-10. [22] Hardy JA, Higgins GA. Alzheimer's disease: The amyloid cascade hypothesis. Science. 1992;256:184-5.
19
[23] Hardy J. The amyloid hypothesis for Alzheimer's disease: a critical reappraisal. Journal of Neurochemistry. 2009;110:1129-34. [24] Jansen WJ, Ossenkoppele R, Knol DL, Tijms BM, Scheltens P, Verhey FR, et al. Prevalence of cerebral amyloid pathology in persons without dementia: a meta-analysis. JAMA. 2015;313:1924-38. [25] Drachman DA. The amyloid hypothesis, time to move on: Amyloid is the downstream result, not cause, of Alzheimer's disease. Alzheimers Dement. 2014;10:372-80. [26] Webster SJ, Bachstetter AD, Nelson PT, Schmitt FA, Van Eldik LJ. Using mice to model Alzheimer's dementia: an overview of the clinical disease and the preclinical behavioral changes in 10 mouse models. Front Genet. 2014;5:88. [27] Onos KD, Sukoff Rizzo SJ, Howell GR, Sasner M. Toward more predictive genetic mouse models of Alzheimer's disease. Brain Res Bull. 2015;122:1-11. [28] Luchsinger JA, Tang MX, Shea S, Mayeux R. Hyperinsulinemia and risk of Alzheimer disease. Neurology. 2004;63:1187-92. [29] Holscher C. Diabetes as a risk factor for Alzheimer's disease: insulin signalling impairment in the brain as an alternative model of Alzheimer's disease. Biochem Soc Trans. 2011;39:891-7. [30] Sherzai D, Sherzai A, Lui K, Pan D, Chiou D, Bazargan M, et al. The association between diabetes and dementia among elderly individuals: a nationwide inpatient sample analysis. J Geriatr Psychiatry Neurol. 2016. [31] Butterfield DA, Di Domenico F, Barone E. Elevated risk of type 2 diabetes for development of Alzheimer disease: a key role for oxidative stress in brain. Biochimica et Biophysica Acta. 2014;1842:1693-706.
20
[32] Wang Y, Wu L, Li J, Fang D, Zhong C, Chen JX, et al. Synergistic exacerbation of mitochondrial and synaptic dysfunction and resultant learning and memory deficit in a mouse model of diabetic Alzheimer's disease. Journal of Alzheimer's disease : JAD. 2015;43:451-63. [33] Lenzen S. The mechanisms of alloxan- and streptozotocin-induced diabetes. Diabetologia. 2008;51:216-26. [34] Rajasekar N, Dwivedi S, Nath C, Hanif K, Shukla R. Protection of streptozotocin induced insulin receptor dysfunction, neuroinflammation and amyloidogenesis in astrocytes by insulin. Neuropharmacology. 2014;86:337-52. [35] Anitha M, Abraham PM, Paulose CS. Striatal dopamine receptors modulate the expression of insulin receptor, IGF-1 and GLUT-3 in diabetic rats: effect of pyridoxine treatment. Eur J Pharmacol. 2012;696:54-61. [36] Kadowaki T, Kasuga M, Akanuma Y, Ezaki O, Takaku F. Decreased autophosphorylation of the insulin receptor-kinase in streptozotocin-diabetic rats. J Biol Chem. 1984;259:14208-16. [37] Mao XY, Cao DF, Li X, Yin JY, Wang ZB, Zhang Y, et al. Huperzine A ameliorates cognitive deficits in streptozotocin-induced diabetic rats. Int J Mol Sci. 2014;15:7667-83. [38] Wang S, Zhou SL, Min FY, Ma JJ, Shi XJ, Bereczki E, et al. mTOR-mediated hyperphosphorylation of tau in the hippocampus is involved in cognitive deficits in streptozotocin-induced diabetic mice. Metab Brain Dis. 2014;29:729-36. [39] Jolivalt CG, Hurford R, Lee CA, Dumaop W, Rockenstein E, Masliah E. Type 1 diabetes exaggerates features of Alzheimer's disease in APP transgenic mice. Exp Neurol. 2009;223:42231.
21
[40] Ramos-Rodriguez JJ, Infante-Garcia C, Galindo-Gonzalez L, Garcia-Molina Y, LechugaSancho A, Garcia-Alloza M. Increased spontaneous central bleeding and cognition impairment in APP/PS1 mice with poorly controlled diabetes mellitus. Mol Neurobiol. 2015. [41] Vandal M, White PJ, Tremblay C, St-Amour I, Chevrier G, Emond V, et al. Insulin reverses the high-fat diet-induced increase in brain Abeta and improves memory in an animal model of Alzheimer disease. Diabetes. 2014;63:4291-301. [42] Knight EM, Martins IV, Gumusgoz S, Allan SM, Lawrence CB. High-fat diet-induced memory impairment in triple-transgenic Alzheimer's disease (3xTgAD) mice is independent of changes in amyloid and tau pathology. Neurobiology of Aging. 2014;35:1821-32. [43] Ho L, Qin W, Pompl PN, Xiang Z, Wang J, Zhao Z, et al. Diet-induced insulin resistance promotes amyloidosis in a transgenic mouse model of Alzheimer's disease. FASEB J. 2004;18:902-4. [44] Cao D, Lu H, Lewis TL, Li L. Intake of sucrose-sweetened water induces insulin resistance and exacerbates memory deficits and amyloidosis in a transgenic mouse model of Alzheimer disease. J Biol Chem. 2007;282:36275-82. [45] Valente T, Gella A, Fernandez-Busquets X, Unzeta M, Durany N. Immunohistochemical analysis of human brain suggests pathological synergism of Alzheimer's disease and diabetes mellitus. Neurobiol Dis. 2010;37:67-76. [46] Okabayashi S, Shimozawa N, Yasutomi Y, Yanagisawa K, Kimura N. Diabetes mellitus accelerates Abeta pathology in brain accompanied by enhanced GAbeta generation in nonhuman primates. PloS One. 2015;10:e0117362.
22
[47] Wang X, Yu S, Hu JP, Wang CY, Wang Y, Liu HX, et al. Streptozotocin-induced diabetes increases amyloid plaque deposition in AD transgenic mice through modulating AGEs/RAGE/NFkappaB pathway. Int J Neurosci. 2014;124:601-8. [48] Macauley SL, Stanley M, Caesar EE, Yamada SA, Raichle ME, Perez R, et al. Hyperglycemia modulates extracellular amyloid-beta concentrations and neuronal activity in vivo. J Clin Invest. 2015;125:2463-7. [49] Chao AC, Lee TC, Juo SH, Yang DI. Hyperglycemia increases the production of amyloid betapeptide leading to decreased endothelial tight junction. CNS Neurosci Ther. 2016;22:291-7. [50] Zhao L, Teter B, Morihara T, Lim GP, Ambegaokar SS, Ubeda OJ, et al. Insulin-degrading enzyme as a downstream target of insulin receptor signaling cascade: implications for Alzheimer's disease intervention. Journal of Neuroscience. 2004;24:11120-6. [51] Kimura T, Takamatsu J, Araki N, Goto M, Kondo A, Miyakawa T, et al. Are advanced glycation end-products associated with amyloidosis in Alzheimer's disease? Neuroreport. 1995;6:866-8. [52] Munch G, Mayer S, Michaelis J, Hipkiss AR, Riederer P, Muller R, et al. Influence of advanced glycation end-products and AGE-inhibitors on nucleation-dependent polymerization of beta-amyloid peptide. Biochimica et Biophysica Acta. 1997;1360:17-29. [53] Elahi M, Hasan Z, Motoi Y, Matsumoto SE, Ishiguro K, Hattori N. Region-specific vulnerability to oxidative stress, neuroinflammation, and tau hyperphosphorylation in experimental diabetes mellitus mice. Journal of Alzheimer's disease : JAD. 2016;51:1209-24.
23
[54] Ke YD, Delerue F, Gladbach A, Gotz J, Ittner LM. Experimental diabetes mellitus exacerbates tau pathology in a transgenic mouse model of Alzheimer's disease. PloS One. 2009;4:e7917. [55] Jack CR, Jr., Knopman DS, Jagust WJ, Petersen RC, Weiner MW, Aisen PS, et al. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol. 2013;12:207-16. [56] Zec RF. Of mice and men and Alzheimer disease: aging versus time. JSM Alzheimer's Disease and Related Dementia. 2015;2:1017-9. [57] Wirths O, Bayer TA. Neuron loss in transgenic mouse models of Alzheimer's disease. Int J Alzheimers Dis. 2010;article ID 723782. [58] Spangenberg EE, Lee RJ, Najafi AR, Rice RA, Elmore MR, Blurton-Jones M, et al. Eliminating microglia in Alzheimer's mice prevents neuronal loss without modulating amyloid-beta pathology. Brain. 2016;139:1265-81.
24
Figure Legends
Fig 1. STZ increases blood glucose but does not affect body weight (A) Mean blood glucose levels in Control and STZ groups were comparable before treatment, but were significantly greater in treated mice 3 months after STZ administration. Data in this and subsequent figures represent mean ± s.e.m.; group sizes: 9 Control, 13 STZ. ***p < 0.001. (B) Mean body weights were not different between groups either at baseline or 3 months after treatment. Fig 2. STZ treatment did not affect spatial learning or memory in the water maze (A) Both Control and STZ-treated mice learned the hidden platform location over the 5-day training period. There was no difference between groups. (B) Platform location crosses during probe trials (platform removed from the pool) 24 hours (Probe 1) and 14 days (Probe 2) after the last training trial. There were no significant differences in performance between groups or between tests for either group. (C) Swim time in Annulus-40 during probe trials 24 hours (Probe 1) and 14 days (Probe 2) after the last training trial. Performance significantly declined over the delay interval in both groups (##p < 0.01) but was not different between groups at either time point. (D) Total swim distance during probe trials 24 hours (Probe 1) and 14 days (Probe 2) after the last training trial. Total swim distance during the 60-second trial was not different between groups for the immediate probe, but STZ-treated mice swam farther than Controls during the delayed probe (**p < 0.01).
25
Fig 3. STZ treatment did not affect contextual fear conditioning (A) Both Control and STZ-treated mice showed significantly increased freezing in the conditioning chamber 24 hours after training (***p < 0.001), indicating strong context memory. However, STZ-treated mice were significantly less mobile prior to training (#p < 0.05). (B) Difference scores (percent freezing after training minus percent freezing before training) revealed that the net increase in freezing was not different between groups. Fig 4. STZ treatment increased amyloid-containing plaques in the subiculum (A) Top: Photomicrograph of the hippocampal formation from an STZ-treated 3xTg mouse immunostained using 6E10 antibody to illustrate the presence of amyloid precursor protein (APP) and amyloid-containing plaques. Intracellular staining is seen in deep neocortex (very top of image) and in CA1 pyramidal neurons and neurons in the adjacent subiculum. Occasional plaques were seen in the hippocampus proper (e.g., in CA3 and CA1 in the image) but were much more common in the subiculum. Bottom: Higher magnification images of the subiculum from Control and STZ-treated mice. Arrows point to individual plaques. Scale bar = 500 µm. (B) Counts showed that mean plaque numbers in sections containing the subiculum were significantly greater in STZ-treated than in Control mice (**p < 0.01). Fig 5. STZ treatment did not affect phosphorylated tau in the hippocampus Normalized densitometric quantification of hippocampal p-Tau (normalized to β-actin) from untreated 3xTg mice and STZ-treated mice revealed no significant difference in pTau between treatment groups (p = 0.42).
26
Figure 1
A
B
1
Figure 2
2
Figure 3
A
B
3
Plaques/section
Figure 4
A B
4
Figure 5
5
S0166-4328(16)30295-9 http://dx.doi.org/doi:10.1016/j.bbr.2016.05.020 BBR 10199
To appear in:
Behavioural Brain Research
Received date: Revised date: Accepted date:
9-3-2016 6-5-2016 8-5-2016
Please cite this article as: Park Emi Hayashi, Ozment Bria N, Griffith Chelsea M, Zhang Haiying, Patrylo Peter R, Rose Gregory M.Experimentally Induced Diabetes Worsens Neuropathology, but not Learning and Memory,.Behavioural Brain Research http://dx.doi.org/10.1016/j.bbr.2016.05.020 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Experimentally Induced Diabetes Worsens Neuropathology, but not Learning and Memory, in Middle Aged 3xTg Mice
Emi Hayashi Parkb,c,d, Bria N. Ozmenta,c,d, Chelsea M. Griffithb,c, Haiying Zhanga,c, Peter R. Patryloa,b,c and Gregory M. Rosea,b,c
Departments of aAnatomy and bPhysiology, School of Medicine and cNeuroscience Research Center Southern Illinois University Carbondale, IL 62901, USA dThese
authors contributed equally to this work.
Corresponding author: Gregory M. Rose, Ph.D. Department of Anatomy, MC6503 SIU School of Medicine 600 Agriculture Drive Carbondale, IL 62901 USA Email: [email protected]
1
HIGHLIGHTS
Streptozotocin administration to 12-month old 3xTg mice chronically elevated blood glucose levels. Diabetic mice did not show performance deficits in water maze or fear conditioning tasks. Beta-amyloid containing plaque numbers were substantially increased in diabetic mice.
Abstract: Alzheimer’s disease (AD) is the primary cause of dementia in the elderly. The cause of the disease is still unknown, but amyloid plaques and neurofibrillary tangles in the brain are thought to play a role. However, transgenic mouse models expressing these neuropathological features do not show severe or consistent cognitive impairments. There is accumulating evidence that diabetes increases the risk for developing AD. We tested the hypothesis that experimentally induced diabetes would exacerbate cognitive symptoms in a mouse model of AD. Diabetes was induced in 12-month old 3xTg mice using streptozotocin (STZ; 90 mg/kg, i.p., on two successive days). Hyperglycemia was verified by sampling blood glucose levels. Three months after injection (at 15 months of age), the mice were behaviorally tested in the Morris water maze and contextual fear conditioning. Subsequently, the hippocampal region was examined using immunohistochemistry (6E10 antibody for amyloid) and immunoblotting (AT8 antibody for phosphorylated tau). No differences were found in learning or memory between the vehicle-treated control and STZ-treated groups. A significant increase in the number of amyloid-positive plaques was observed in the subiculum of STZ-treated mice; very few plaques were seen in other hippocampal regions in either group. No differences in AT8 load were
2
observed. These results reinforce that amyloid plaques, per se, are not sufficient to cause memory impairments. Further, while diabetes can enhance this aspect of brain pathology, the combination of disrupted glucose metabolism and the transgenes is still not sufficient to cause the severe cognitive impairments associated with clinical AD. Keywords: 3xTg, triple transgenic, Alzheimer, diabetes, streptozotocin
1. Introduction It is common knowledge that Alzheimer’s disease (AD) is a debilitating cognitive disorder and the leading cause of dementia in the elderly. The brains of AD patients contain extracellular senile plaques composed of aggregated beta amyloid (Aβ) and neurofibrillary tangles composed of hyperphosphorylated tau protein; in later stages of the disease, substantial atrophy and neuronal death are observed. There is currently no cure for AD, and marketed symptomatic treatments have, at best, limited and temporary effects. The development of transgenic mouse models containing mutant human genes regulating Aβ synthesis was heralded as a breakthrough because it offered the opportunity to examine the genesis of major aspects of the amyloid plaques, and, in later models, the neurofibrillary tangles composed of hyperphosphorylated tau protein. A reasonable expectation was that the presence of these neuropathologies would result in the profound cognitive deficits that are the major symptom of AD. Unfortunately, this has not turned out to be the case. The cognitive deficits in the AD models are usually mild, are often restricted to a particular measure on a particular behavioral task, and are not always reproducible either within or between
3
laboratories. This outcome is seen despite the presence of heavy plaque loads, particularly in hippocampal and neocortical regions. Further, while many experimental treatments have been shown to improve learning or memory in transgenic AD models, none of these that have progressed to human clinical trials have been successful [1]. In addition to genetic abnormalities, several factors have been identified that likely increase the risk of developing AD; among these is diabetes [2-5]. Type 1 diabetes, which results in the loss of insulin-producing pancreatic beta cells, generally increases the risk of cognitive dysfunction [6]. Type 2 diabetes, a disease traditionally associated with aging, and which results initially in reduced sensitivity to insulin and later in reduced insulin production, is also associated with increased risk of developing dementia, including AD (e.g., [7, 8]). The basis for a relationship between diabetes and AD is still unclear, but could involve chronic hyperglycemia, impaired insulin signaling, or both. Many downstream mechanisms have been proposed, including increased amyloid and tau pathology, elevated brain inflammation, decreased dendritic spine density, reduced synaptic plasticity, and neuron loss [9, 10]. However, treatment with insulin, or insulin receptor sensitizing drugs, has been shown to improve cognitive function in both animal models of AD [9, 11, 12] and in AD patients [13, 14]. In the current study we explored the possibility that a 3-month period of experimentally-induced diabetes would impair hippocampus-dependent learning and memory in 3xTg mice [15] when the treatment was administered prior to the age at which impairments are normally observed [16].
2. Materials and Methods
4
2.1. Animals The subjects in this study were male 3xTg mice raised in our colony from breeding stock provided by Dr. Frank LaFerla [15]. The animals were housed in group cages in a climate controlled vivarium (25° C) at Southern Illinois University Carbondale. Mice had food (Purina LabDiet® 5001, Framingham, MA) and water available ad libitum and were kept on a 14/10-hour light/dark schedule (lights on 0600 – 2000). All experimental treatments and behavioral tests were conducted during the light part of the cycle. Experiments were performed in compliance with the Institutional Animal Care and Use Committee (IACUC) of Southern Illinois University Carbondale according to guidelines set forth by the National Institutes of Health. The mice were 12 months old when the experiments began. This age was chosen because, in our colony, the onset of behavioral impairments occurs later than in the LaFerla lab colony(e.g., [17, 18]; significant deficits in the water maze task are not consistently observed in our 3xTg mice until they are at least 15 months old [16]. 2.2. Induction of Diabetes Diabetes mellitus was induced after overnight fasting by the chemical ablation of pancreatic beta cells using streptozotocin (STZ). Two injections (90 mg/kg, i.p. in sodium citrate buffer, pH 4.5; injection volume 10 ml/kg) were given at a 24 hour interval. Control mice received two injections of the sodium citrate buffer alone on the same schedule. A human OTC glucometer (TRUEtrack, Nipro Diagnostics, Fort Lauderdale, FL) was used to measure blood glucose levels obtained from a tail prick. Hyperglycemia (blood sugar > 250 mg/dL) was confirmed one week following STZ injections in the experimental treatment group. Blood glucose concentrations and body weight were measured weekly for four weeks, and once more prior to sacrifice to confirm that the animals remained
5
hyperglycemic and were not clinically ill (as determined by overall body weight and normal grooming and nesting behaviors) in spite of the presence of diabetes. Cage bedding was changed every other day because of the polyuria present in the diabetic mice. 2.3. Behavioral Testing 2.3.1. Morris Water Maze The Morris water maze is well established test of hippocampus-dependent spatial learning and memory [19]. Mice in this study were tested at 15 months of age, three months after having received STZ or vehicle injections. The pool consisted of a circular tank (1.35 m diameter) painted white and filled with water made opaque by non-toxic white tempera paint. Water temperature was maintained at approximately 21.5 degrees Celsius. For the hidden platform task, a circular hidden platform (10 cm diameter) was located in the “center” of one quadrant of the maze and submerged one cm below the surface of the water. For the visible platform task, an extension was added to the hidden platform so that a portion of the platform was visible above the water level. The four walls of the room were decorated with distinct visual cues to clearly differentiate between them. For the first three consecutive training days (Days 1-3) mice received three daily trials using the visible platform. Visible platform training allows for the assessment of sensorimotor abilities and motivation to escape the water independent of spatial learning ability. On each training trial, an animal was placed into the water at the edge of the tank, facing the wall, with the start locations varying pseudorandomly (N, S, & E quadrants, with the platform in the W quadrant) and allowed to swim until they reached the escape platform. If a mouse did not reach the platform by a maximum of 90 seconds, the animal was hand guided to it. Once on the platform, mice remained there for 15 seconds before being removed for a minimum 20-minute inter-trial interval in their home cage. On
6
Day 4, hidden platform training began and lasted for five consecutive days (Days 4-8), also with three trials/day, a pseudorandom start position for each trial, and a 90 second maximum trial duration. Twenty-four hours after completing hidden platform training, a probe trial (Probe 1) was conducted with no platform in the pool. Mice were allowed to swim for 60 seconds before being removed from the water. A second memory probe test (Probe 2) was performed two weeks later. Performance was recorded using a camera mounted above the maze with an interface to a computerized tracking system (AnyMaze, San Diego Instruments, San Diego, CA). AnyMaze measured swim time and distance for the learning trials, and platform location crossings, time in the annulus-40 (a 40 cm diameter circle centered on the platform location), and total swim distance for the probe trials. 2.3.2. Fear Conditioning After the completion of water maze testing the mice were allowed to rest for one week before undergoing a fear conditioning procedure. Each mouse was placed in a soundattenuating chamber (30 x 25 x 20 cm, l x w x h), the floor of which was made of parallel 2-mm diameter stainless steel rods spaced 8 mm apart (Med Associates, St. Albans, VT). The floor was wired to a scrambled shock generator (also Med Associates). A red light (off during training and context memory testing) was located on one wall of the chamber. A video camera was mounted in front of the clear door of the chamber for video monitoring and recording of the animals’ behavior. On Day 1 of this paradigm, mice were placed in the chamber and allowed to explore for 60 seconds. After this they were exposed to a 2 kHz, 90 dB tone for 30 seconds, which coterminated with a 2-second, 0.6 mA foot shock. Sixty seconds later the auditory cue was repeated for 30 seconds, ending with a second foot shock. The mice remained in the chamber
7
for an additional 90 seconds before being removed and returned to their home cage. A 70% alcohol solution was used to clean the chamber before and after each subject. Twenty-four hours later (Day 2) the mice were returned to their training chamber and their behavior was monitored for 5 minutes. Behavioral freezing (immobility except for respiration) to the context was measured using Freeze Frame video tracking and software (ActiMetrics, Willmette, IL). On Day 3, the chamber was altered to mask the conditioning context by covering the grid floor, adding a drop of citrus scent in a weighing dish underneath the floor, turning on the red light and placing novel objects in the chamber. Mice placed in the chamber demonstrated that they did not recognize the training context of the box by immediately exploring the altered environment. After three minutes of exploration, memory for the association between the auditory cue and the foot shocks given on Day 1 was assessed by monitoring behavioral freezing during continuous delivery of the tone for three additional minutes. Unfortunately, the data from the cued memory assessment were lost to due to a computer problem so are not included in the analysis. 2.4 Aβ Immunohistochemistry Within one week after behavioral testing was completed, the mice were anesthetized with isoflurane and intracardially perfused with cold (4 °C) phosphate-buffered saline (PBS). The brains were removed and bisected along the sagittal sulcus; the hippocampus was dissected from the right half of the brain, snap frozen using dry ice powder and stored at -80°C for later biochemical analysis (see below). The left half of the brain was placed in a solution of cold 4% paraformaldehyde in PBS for 2 days, then placed in a 30% sucrose solution and stored at 4 °C until it was sectioned. Forty-
8
micron-thick sagittal sections through the hippocampal region were cut using a freezing microtome and stored in a cryoprotectant solution [20] until they were processed for immunohistochemistry. Sections were treated with 1% H2O2 in PBS for 30 minutes to quench endogenous peroxidase, followed by a pre-incubation in 5% normal goat serum in PBS with 0.3% Triton X100 for 1 hour. Antigen retrieval was used for Aβ labeling using 50% formic acid in PBS for 30 minutes at room temperature prior to bleaching of endogenous peroxidase. Sections were then incubated overnight with the primary antibody (anti-Aβ 6E10, 1:1000; BioLegend SIG39320, Dedham, MA) in blocking serum (4°C), then rinsed and processed using the reagents and protocol supplied with Vectastain elite ABC kit (Vector Laboratories, Burlingame, CA). Immunoreactive product was visualized by incubation in a solution containing 0.05% diaminobenzidine and 0.003% H2O2. Three 10-minute washes with PBST were performed between all incubations. Sections were mounted on slides, allowed to air-dry, delipidated with butanol and xylene, and then cover slipped. 2.5 p-Tau Western Blotting Hippocampal samples were homogenized by sonication in 700 µl of protein isolation buffer (150 nM NaCl, 0.1% SDS, 0.5% sodium deoxycholate, 1% Igepal (NP-40), 50 mM Tris, pH 8.0, 20 mM NaF + mixed protease and phosphatase inhibitors). The homogenates were then centrifuged at 12,000 RPM for 10 minutes at 4˚C. Supernatants were collected and protein concentrations determined using the bicinchoninic acid method (Pierce BCA Protein Assay Kit, ThermoFisher Scientific, Waltham, MA). Samples (50 μg/ml) were loaded onto 10% SDS-Page gels, transferred to PDVF membranes and probed with a p-Tau antibody (mouse monoclonal Phospho-Tau (Ser396),
9
1:1000; Cell Signaling, Danvers, MA). Beta-actin was used as a control protein (rabbit monoclonal Anti-β-actin, 1:500, Cell Signaling, Danvers, MA). Appropriate secondary antibodies were used (goat -anti-rabbit IRDye 700 DX and goat-anti-mouse IRDye 800 CW, 1:10,000; Rockland, Gilbertsville, PA) for densitometric quantification of protein bands using an Odyssey Licor infrared imaging system and were normalized to β-actin. For comparison between groups the mean for the control group was normalized to one. 2.6. Statistical Analysis All data are presented as mean ± s.e.m. Data analyses were performed using Prism 5 (v. 5.03; GraphPad Software, La Jolla, CA). Differences between multiple means were analyzed by ANOVA. Differences between two means were analyzed using two-tailed t-tests. Statistical significance was defined as p < 0.05.
3. Results 3.1 Effect of STZ Blood Glucose Levels With the expectation that chronically elevated blood glucose would cause mortality, 15 mice received STZ injections while 10 mice received the vehicle. Prior to treatment, glucose levels for control mice were 116.8 ± 4.4 mg/dL (mean ± s.e.m.) and 111.3 +/- 6.8 mg/dL for the group to receive STZ (p > 0.1, t-test). One week after STZ treatment, 13 of 15 mice had blood glucose levels of > 250 mg/dl, meeting our prospective criterion for diabetes. The two remaining mice were eliminated from the study. By three months later, when the surviving mice were 15 months old, one mouse in the control group had died; the mean blood glucose level for the remaining 9 animals was 135.6 +/- 9.4 mg/dL, not significantly different from the initial value (p > 0.1, paired t-test). By contrast, the mean value for the 13 mice in the STZ-treated group was significantly increased to 10
492.9 +/- 38.7 mg/dL (p < 0.001, paired t-test; Fig. 1A). Body weights for Control and STZ mice were comparable prior to treatment (33.0 ± 1.2 vs. 33.1 ± 0.9 gm, p > 0.1, t-test). Body weights in both groups decreased slightly, but not significantly, during the following 3 months (weights prior to behavioral testing: Control – 31.4 ± 0.9 gm; STZ – 31.5 ± 1.3 gm; p > 0.1, paired t-tests; Fig. 1B). Thus, despite having extremely high blood glucose levels, the STZ group did not show increased morbidity or mortality. 3.2 Water Maze Testing The mice were tested in the water maze at 15 months of age, 3 months after the induction of diabetes in the STZ-treated group. Control and STZ-treated animals behaved comparably in all phases of the water maze task. Swim times to the visible platform improved over the 3-day training period in both groups. Mean latencies on Day 3 were: Control – 10.3 ± 2.4 sec; STZ – 8.0 ± 1.3 sec; p > 0.1, t-test). Similarly, mean swim distances to locate the hidden platform decreased significantly over the 5-day training period (p < 0.001, repeated measures ANOVA). However, no difference was observed between groups (p > 0.05 for daily comparisons, Bonferroni post-hoc tests; Fig 2A). Performance during the probe trial given 24 hours later (Probe 1) was also not different between groups (Platform location crosses: Control – 3.78 ± 1.05, STZ – 4.23 ± 0.63, p > 0.1, t-test, Fig 2B; Swim time in Annulus-40: Control – 12.23 ± 1.83 sec , STZ – 11.49 ± 1.51 sec, p > 0.1, t-test, Fig 2C; Swim distance during probe trial: Control – 10.63 ± 0.44 m, STZ – 11.28 ± 0.31, p > 0.1, t-test, Fig 2D). Performance during Probe 2, given 14 days later, produced similar results. Platform location crosses were not different between probes or between groups (Probe 2 platform location crosses: Control – 3.56 ± 0.67, STZ – 2.85 ± 0.59, p > 0.1, ANOVA, Fig 2B). Swim time in the annulus-40 declined for both groups between Probe 1 and Probe 2 (p = 0.01, ANOVA), but the degree was
11
comparable, and there was no difference between groups for Probe 2 (Control – 8.90 ± 1.12 sec , STZ – 7.25 ± 1.01 sec, p > 0.1, ANOVA, Fig 2C). Total swim distances during Probe 2 reached significance (swim distance during Probe 2: Control – 10.76 ± 0.22 m, STZ – 11.87 ± 0.27, p > 0.01, ANOVA, Fig 2D), but there was no significant difference for either group between the two probe trials. 3.3 Fear Conditioning Both Control and STZ-treated mice demonstrated significant context memory 24 hours after training, i.e., an increase in freezing compared to baseline levels (p < 0.001, ANOVA; Fig 3A). The STZ group showed less activity (greater percent freezing) than controls during the 90-second period prior to the first tone-shock delivery (Control – 2.24 ± 1.49, STZ – 17.17 ± 6.71; p < 0.05, t-test with Welch correction for unequal variance), although freezing during the memory test was not different between groups (Control – 57 ± 11.16, STZ – 78.48 ± 6.70, p > 0.09, t-test). Thus, the results were analyzed by computing difference scores (freezing after training minus freezing before training) for each animal. This approach revealed that there was not a significant difference in the amount of freezing to context between the Control and STZ groups (Control – 54.51 ± 10.67, STZ – 61.31 ± 8.22, p > 0.1, t-test). 3.4 Amyloid Immunohistochemistry Amyloid staining, visualized by 6E10 immunohistochemistry, was present in the hippocampus, amygdala, and deep layers of neocortex. Labeling was primarily confined to neuronal cell bodies. Amyloid-containing plaques were occasionally observed in the hippocampus proper, but were more concentrated in the subiculum (Fig 4A). Counts were made of the plaques seen in all sections containing the subiculum, and then averaged to provide a number/section for each mouse. Comparison of the mean number of plaques/section revealed that there were significantly more Aβ-
12
positive plaques in the subiculum of STZ-treated mice compared to Controls (plaques per histological section: Control – 2.3 ± 0.8, STZ – 8.6 ± 1.8, p < 0.01, t-test; Fig 4B). 3.5 p-Tau Western Blotting To assess whether STZ treatment caused changes in levels of phosphorylated tau in the hippocampus, western blotting for p-Tau (Ser396) was performed. Levels of phosphorylated tau protein in the hippocampus were normalized to β-actin, the control mean was normalized to one, and group means were compared. No difference in hippocampal p-tau was found between Control and STZ-treated mice (Control normalized optical density – 1.00 ± 0.10, STZ normalized optical density – 1.15 ± 0.15, p > 0.1, t-test; Fig 5). 4. Discussion The goal of this study was to test the hypothesis that inducing frank diabetes in 3xTg mice would significantly worsen learning and memory deficits. We found that a 3-month period of profoundly elevated blood glucose levels, produced by STZ injections, did not affect the behavioral performance of middle-aged (12 months old at the start of the experiment) 3xTg mice in either the Morris water maze or contextual fear conditioning tasks when the animals were tested at 15 months of age. However, STZ-treated mice had a greater than 3-fold increase in the number of Aβ-containing plaques in the subiculum labeled by 6E10 immunocytochemistry. Hippocampal levels of phosphorylated tau were not significantly altered by STZ treatment. The seminal observation of Alois Alzheimer of profound brain pathology in patients with debilitating cognitive deficits logically gave rise to the hypothesis that what were later identified as Aβ-containing plaques and neurofibrillary tangles containing hyperphosphorylated
13
tau were responsible for the symptoms of AD [21]. In particular, the “amyloid hypothesis” [22] has provided an overarching influence on developing potential therapeutics for the disease. However, further research, and an unfortunate lack of success with clinical drug candidates designed to reduce behavioral symptoms, are leading to a reevaluation of the role of Aβ in AD [23-25]. This line of thinking has been reinforced by the lack of correspondence between the severity of cognitive deficits in AD patients and in the many transgenic mouse models now available to study amyloid and tau pathology. (See [1, 26, 27] for a range of perspectives.) It is certainly premature to abandon the idea that amyloid and tau are involved in important ways in AD. However, at this point it is reasonable to explore other factors that could, in combination with amyloid and tau pathology, contribute to the appearance of behavioral symptoms. Emerging evidence from epidemiological studies suggests that diabetes is a risk factor for developing AD [2, 28, 29] (but see [30]). At this point it is unclear whether hyperglycemia, insulin resistance, or other factors are the most important for this linkage [31, 32]. Systemic STZ, through the destruction of pancreatic beta cells, results in both hyperglycemia and disruption of insulin receptor signaling [33-36]. We found no significant differences in learning or memory measures assessed approximately 3 months after STZ treatment, despite a substantial and prolonged elevation in blood glucose levels. While intraperitoneal administration of STZ to wild-type rodents has been reported to result in deficits in the water maze task, significant weight loss in the treated animals is a confounding factor [37, 38]. Inducing diabetes in transgenic AD mouse models has produced mixed results in behavioral tests. Intraperitoneal STZ alone caused profound deficits in the Barnes maze task in wild-type
14
mice that were debatably worsened in hAPP transgenic animals [39]. In APP-PS1 mice, learning and memory in the water maze were also somewhat worsened in STZ-treated animals [40]. More robust effects of STZ treatment have been reported for the mAβAPP model [32]. Dietary manipulations to induce a type II-like diabetic state in transgenic AD mouse models have also been shown to exacerbate learning and memory impairments (e.g., [41-44]). To our knowledge, the present study is the first to examine the effect of inducing systemic diabetes in the 3xTg model. While treatment with STZ did not affect behavioral measures in our 3xTg mice, it did result in an increased number of amyloid plaques in the subiculum. Enhanced amyloid pathology has also been reported in diabetic AD patients [45], as well as in aged diabetic nonhuman primates [46] and in hAPP [39] and APP/PS1 [47] transgenic mice treated with STZ. (However, reduced amyloid deposition in the brains of APP/PS1 mice after STZ treatment has also been reported [40], as well in APP/PS1 mice crossed with diabetic db/db mice [10].) The mechanism responsible the elevated amyloid plaques is unclear, but could involve increased βamyloid production [48, 49], a reduction in insulin-degrading enzyme which also degrades Aβ [50], or through generation of advanced glycation end products that promotes Aβ aggregation [47, 51, 52]. In our study, levels of phosphorylated tau in the hippocampus were not elevated following STZ treatment. This finding contrasts with those reported examining the effects of STZ in other transgenic mouse models that selectively overexpress human tau (Tg601 [53] and pR5 [54]). While reason for this discrepancy is unclear, it may be due to differences in the mouse models (including the presence of multiple transgenes in 3xTg mice), ages or brain
15
regions examined, or that our data were normalized to an already elevated baseline in our 15month old mice. The results of this study support the emerging consensus that neuropathological hallmarks of AD do not seem to bear a direct relationship to the debilitating cognitive impairments associated with the disease [24, 25]. In addition, our work emphasizes this discrepancy in a transgenic mouse model of AD. Since there is strong evidence that elevated brain amyloid deposition is a risk factor for AD, albeit one that appears 1-2 decades before the onset of dementia [24, 55], it is possible that the mild behavioral deficits seen in transgenic AD models are due merely to the relatively short lives of the animals [56]. However, a stronger case could be made that the lack of substantial neuron loss observed in existing models of AD is the key factor [56, 57]. While neuron numbers or density were not quantified in our study, the combination of STZ treatment and multiple human transgenes was not sufficient to obviously affect these variables. Interestingly, a recent report documented contextual fear conditioning deficits and neuron loss in the subiculum of 5xFAD transgenic mice [58]. Elimination of microglia by inhibiting CSFR1 in these mice ameliorated both neuron loss and the behavioral impairment, but had no effect on brain amyloid levels or plaque load. This result further reinforces the dichotomy between amyloid and cognitive deficits, and suggests a possible path forward to understanding the mechanism(s) responsible for neuronal death in the AD brain. Conflicts of Interest The authors declare no conflict of interest. Acknowledgments
16
This work was supported by a predoctoral fellowship from Southern Illinois University to E.H.P., a REACH award to B.O., and the SIU Neuroscience Research Center. H.Z. received support from Hainan Medical College and the China Scholarship Council.
References [1] Windisch M. We can treat Alzheimer's disease successfully in mice but not in men: failure in translation? A perspective. Neurodegenerative diseases. 2014;13:147-50. [2] Ott A, Stolk RP, van Harskamp F, Pols HA, Hofman A, Breteler MM. Diabetes mellitus and the risk of dementia: The Rotterdam Study. Neurology. 1999;53:1937-42. [3] Akomolafe A, Beiser A, Meigs JB, Au R, Green RC, Farrer LA, et al. Diabetes mellitus and risk of developing Alzheimer disease: results from the Framingham Study. Archives of neurology. 2006;63:1551-5. [4] Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer's disease: an analysis of population-based data. Lancet Neurol. 2014;13:788-94. [5] Cheng G, Huang C, Deng H, Wang H. Diabetes as a risk factor for dementia and mild cognitive impairment: a meta-analysis of longitudinal studies. Intern Med J. 2012;42:484-91. [6] Brands AM, Biessels GJ, de Haan EH, Kappelle LJ, Kessels RP. The effects of type 1 diabetes on cognitive performance: a meta-analysis. Diabetes Care. 2005;28:726-35. [7] Gudala K, Bansal D, Schifano F, Bhansali A. Diabetes mellitus and risk of dementia: A metaanalysis of prospective observational studies. J Diabetes Investig. 2013;4:640-50. [8] Chatterjee S, Peters SA, Woodward M, Mejia Arango S, Batty GD, Beckett N, et al. Type 2 Diabetes as a Risk Factor for Dementia in Women Compared With Men: A Pooled Analysis of 17
2.3 Million People Comprising More Than 100,000 Cases of Dementia. Diabetes Care. 2016;39:300-7. [9] Searcy JL, Phelps JT, Pancani T, Kadish I, Popovic J, Anderson KL, et al. Long-term pioglitazone treatment improves learning and attenuates pathological markers in a mouse model of Alzheimer's disease. Journal of Alzheimer's disease : JAD. 2012;30:943-61. [10] Infante-Garcia C, Ramos-Rodriguez JJ, Galindo-Gonzalez L, Garcia-Alloza M. Long-term central pathology and cognitive impairment are exacerbated in a mixed model of Alzheimer's disease and type 2 diabetes. Psychoneuroendocrinology. 2016;65:15-25. [11] Papadopoulos P, Rosa-Neto P, Rochford J, Hamel E. Pioglitazone improves reversal learning and exerts mixed cerebrovascular effects in a mouse model of Alzheimer's disease with combined amyloid-beta and cerebrovascular pathology. PloS One. 2013;8:e68612. [12] Yamanaka M, Ishikawa T, Griep A, Axt D, Kummer MP, Heneka MT. PPARgamma/RXRalphainduced and CD36-mediated microglial amyloid-beta phagocytosis results in cognitive improvement in amyloid precursor protein/presenilin 1 mice. Journal of Neuroscience. 2012;32:17321-31. [13] Craft S, Asthana S, Cook DG, Baker LD, Cherrier M, Purganan K, et al. Insulin dose-response effects on memory and plasma amyloid precursor protein in Alzheimer's disease: interactions with apolipoprotein E genotype. Psychoneuroendocrinology. 2003;28:809-22. [14] Craft S, Baker LD, Montine TJ, Minoshima S, Watson GS, Claxton A, et al. Intranasal insulin therapy for Alzheimer disease and amnestic mild cognitive impairment: a pilot clinical trial. Archives of Neurology. 2011;69:29-38.
18
[15] Oddo S, Caccamo A, Shepherd JD, Murphy MP, Golde TE, Kayed R, et al. Triple-transgenic model of Alzheimer's disease with plaques and tangles: intracellular Abeta and synaptic dysfunction. Neuron. 2003;39:409-21. [16] Macklin LN, Cai Y, Kanak DJ, Ebersole J, Corbin AE, Womack A, et al. Impaired glucose tolerance precedes neuroanatomical identification of Aβ and hyperphosphorylated tau accumulation in the 3xTg model of Alzheimer’s disease. Society for Neuroscience Abstracts. 2009;528.26. [17] Billings LM, Oddo S, Green KN, McGaugh JL, LaFerla FM. Intraneuronal Abeta causes the onset of early Alzheimer's disease-related cognitive deficits in transgenic mice. Neuron. 2005;45:675-88. [18] Billings LM, Green KN, McGaugh JL, LaFerla FM. Learning decreases A beta*56 and tau pathology and ameliorates behavioral decline in 3xTg-AD mice. Journal of Neuroscience. 2007;27:751-61. [19] Morris RGM, Garrud P, Rawlins JNP, O'Keefe J. Place navigation is impaired in rats with hippocampal lesions. Nature. 1982;297:681-3. [20] Watson RE, Jr., Wiegand SJ, Clough RW, Hoffman GE. Use of cryoprotectant to maintain long-term peptide immunoreactivity and tissue morphology. Peptides. 1986;7:155-9. [21] Small DH, Cappai R. Alois Alzheimer and Alzheimer's disease: a centennial perspective. Journal of Neurochemistry. 2006;99:708-10. [22] Hardy JA, Higgins GA. Alzheimer's disease: The amyloid cascade hypothesis. Science. 1992;256:184-5.
19
[23] Hardy J. The amyloid hypothesis for Alzheimer's disease: a critical reappraisal. Journal of Neurochemistry. 2009;110:1129-34. [24] Jansen WJ, Ossenkoppele R, Knol DL, Tijms BM, Scheltens P, Verhey FR, et al. Prevalence of cerebral amyloid pathology in persons without dementia: a meta-analysis. JAMA. 2015;313:1924-38. [25] Drachman DA. The amyloid hypothesis, time to move on: Amyloid is the downstream result, not cause, of Alzheimer's disease. Alzheimers Dement. 2014;10:372-80. [26] Webster SJ, Bachstetter AD, Nelson PT, Schmitt FA, Van Eldik LJ. Using mice to model Alzheimer's dementia: an overview of the clinical disease and the preclinical behavioral changes in 10 mouse models. Front Genet. 2014;5:88. [27] Onos KD, Sukoff Rizzo SJ, Howell GR, Sasner M. Toward more predictive genetic mouse models of Alzheimer's disease. Brain Res Bull. 2015;122:1-11. [28] Luchsinger JA, Tang MX, Shea S, Mayeux R. Hyperinsulinemia and risk of Alzheimer disease. Neurology. 2004;63:1187-92. [29] Holscher C. Diabetes as a risk factor for Alzheimer's disease: insulin signalling impairment in the brain as an alternative model of Alzheimer's disease. Biochem Soc Trans. 2011;39:891-7. [30] Sherzai D, Sherzai A, Lui K, Pan D, Chiou D, Bazargan M, et al. The association between diabetes and dementia among elderly individuals: a nationwide inpatient sample analysis. J Geriatr Psychiatry Neurol. 2016. [31] Butterfield DA, Di Domenico F, Barone E. Elevated risk of type 2 diabetes for development of Alzheimer disease: a key role for oxidative stress in brain. Biochimica et Biophysica Acta. 2014;1842:1693-706.
20
[32] Wang Y, Wu L, Li J, Fang D, Zhong C, Chen JX, et al. Synergistic exacerbation of mitochondrial and synaptic dysfunction and resultant learning and memory deficit in a mouse model of diabetic Alzheimer's disease. Journal of Alzheimer's disease : JAD. 2015;43:451-63. [33] Lenzen S. The mechanisms of alloxan- and streptozotocin-induced diabetes. Diabetologia. 2008;51:216-26. [34] Rajasekar N, Dwivedi S, Nath C, Hanif K, Shukla R. Protection of streptozotocin induced insulin receptor dysfunction, neuroinflammation and amyloidogenesis in astrocytes by insulin. Neuropharmacology. 2014;86:337-52. [35] Anitha M, Abraham PM, Paulose CS. Striatal dopamine receptors modulate the expression of insulin receptor, IGF-1 and GLUT-3 in diabetic rats: effect of pyridoxine treatment. Eur J Pharmacol. 2012;696:54-61. [36] Kadowaki T, Kasuga M, Akanuma Y, Ezaki O, Takaku F. Decreased autophosphorylation of the insulin receptor-kinase in streptozotocin-diabetic rats. J Biol Chem. 1984;259:14208-16. [37] Mao XY, Cao DF, Li X, Yin JY, Wang ZB, Zhang Y, et al. Huperzine A ameliorates cognitive deficits in streptozotocin-induced diabetic rats. Int J Mol Sci. 2014;15:7667-83. [38] Wang S, Zhou SL, Min FY, Ma JJ, Shi XJ, Bereczki E, et al. mTOR-mediated hyperphosphorylation of tau in the hippocampus is involved in cognitive deficits in streptozotocin-induced diabetic mice. Metab Brain Dis. 2014;29:729-36. [39] Jolivalt CG, Hurford R, Lee CA, Dumaop W, Rockenstein E, Masliah E. Type 1 diabetes exaggerates features of Alzheimer's disease in APP transgenic mice. Exp Neurol. 2009;223:42231.
21
[40] Ramos-Rodriguez JJ, Infante-Garcia C, Galindo-Gonzalez L, Garcia-Molina Y, LechugaSancho A, Garcia-Alloza M. Increased spontaneous central bleeding and cognition impairment in APP/PS1 mice with poorly controlled diabetes mellitus. Mol Neurobiol. 2015. [41] Vandal M, White PJ, Tremblay C, St-Amour I, Chevrier G, Emond V, et al. Insulin reverses the high-fat diet-induced increase in brain Abeta and improves memory in an animal model of Alzheimer disease. Diabetes. 2014;63:4291-301. [42] Knight EM, Martins IV, Gumusgoz S, Allan SM, Lawrence CB. High-fat diet-induced memory impairment in triple-transgenic Alzheimer's disease (3xTgAD) mice is independent of changes in amyloid and tau pathology. Neurobiology of Aging. 2014;35:1821-32. [43] Ho L, Qin W, Pompl PN, Xiang Z, Wang J, Zhao Z, et al. Diet-induced insulin resistance promotes amyloidosis in a transgenic mouse model of Alzheimer's disease. FASEB J. 2004;18:902-4. [44] Cao D, Lu H, Lewis TL, Li L. Intake of sucrose-sweetened water induces insulin resistance and exacerbates memory deficits and amyloidosis in a transgenic mouse model of Alzheimer disease. J Biol Chem. 2007;282:36275-82. [45] Valente T, Gella A, Fernandez-Busquets X, Unzeta M, Durany N. Immunohistochemical analysis of human brain suggests pathological synergism of Alzheimer's disease and diabetes mellitus. Neurobiol Dis. 2010;37:67-76. [46] Okabayashi S, Shimozawa N, Yasutomi Y, Yanagisawa K, Kimura N. Diabetes mellitus accelerates Abeta pathology in brain accompanied by enhanced GAbeta generation in nonhuman primates. PloS One. 2015;10:e0117362.
22
[47] Wang X, Yu S, Hu JP, Wang CY, Wang Y, Liu HX, et al. Streptozotocin-induced diabetes increases amyloid plaque deposition in AD transgenic mice through modulating AGEs/RAGE/NFkappaB pathway. Int J Neurosci. 2014;124:601-8. [48] Macauley SL, Stanley M, Caesar EE, Yamada SA, Raichle ME, Perez R, et al. Hyperglycemia modulates extracellular amyloid-beta concentrations and neuronal activity in vivo. J Clin Invest. 2015;125:2463-7. [49] Chao AC, Lee TC, Juo SH, Yang DI. Hyperglycemia increases the production of amyloid betapeptide leading to decreased endothelial tight junction. CNS Neurosci Ther. 2016;22:291-7. [50] Zhao L, Teter B, Morihara T, Lim GP, Ambegaokar SS, Ubeda OJ, et al. Insulin-degrading enzyme as a downstream target of insulin receptor signaling cascade: implications for Alzheimer's disease intervention. Journal of Neuroscience. 2004;24:11120-6. [51] Kimura T, Takamatsu J, Araki N, Goto M, Kondo A, Miyakawa T, et al. Are advanced glycation end-products associated with amyloidosis in Alzheimer's disease? Neuroreport. 1995;6:866-8. [52] Munch G, Mayer S, Michaelis J, Hipkiss AR, Riederer P, Muller R, et al. Influence of advanced glycation end-products and AGE-inhibitors on nucleation-dependent polymerization of beta-amyloid peptide. Biochimica et Biophysica Acta. 1997;1360:17-29. [53] Elahi M, Hasan Z, Motoi Y, Matsumoto SE, Ishiguro K, Hattori N. Region-specific vulnerability to oxidative stress, neuroinflammation, and tau hyperphosphorylation in experimental diabetes mellitus mice. Journal of Alzheimer's disease : JAD. 2016;51:1209-24.
23
[54] Ke YD, Delerue F, Gladbach A, Gotz J, Ittner LM. Experimental diabetes mellitus exacerbates tau pathology in a transgenic mouse model of Alzheimer's disease. PloS One. 2009;4:e7917. [55] Jack CR, Jr., Knopman DS, Jagust WJ, Petersen RC, Weiner MW, Aisen PS, et al. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol. 2013;12:207-16. [56] Zec RF. Of mice and men and Alzheimer disease: aging versus time. JSM Alzheimer's Disease and Related Dementia. 2015;2:1017-9. [57] Wirths O, Bayer TA. Neuron loss in transgenic mouse models of Alzheimer's disease. Int J Alzheimers Dis. 2010;article ID 723782. [58] Spangenberg EE, Lee RJ, Najafi AR, Rice RA, Elmore MR, Blurton-Jones M, et al. Eliminating microglia in Alzheimer's mice prevents neuronal loss without modulating amyloid-beta pathology. Brain. 2016;139:1265-81.
24
Figure Legends
Fig 1. STZ increases blood glucose but does not affect body weight (A) Mean blood glucose levels in Control and STZ groups were comparable before treatment, but were significantly greater in treated mice 3 months after STZ administration. Data in this and subsequent figures represent mean ± s.e.m.; group sizes: 9 Control, 13 STZ. ***p < 0.001. (B) Mean body weights were not different between groups either at baseline or 3 months after treatment. Fig 2. STZ treatment did not affect spatial learning or memory in the water maze (A) Both Control and STZ-treated mice learned the hidden platform location over the 5-day training period. There was no difference between groups. (B) Platform location crosses during probe trials (platform removed from the pool) 24 hours (Probe 1) and 14 days (Probe 2) after the last training trial. There were no significant differences in performance between groups or between tests for either group. (C) Swim time in Annulus-40 during probe trials 24 hours (Probe 1) and 14 days (Probe 2) after the last training trial. Performance significantly declined over the delay interval in both groups (##p < 0.01) but was not different between groups at either time point. (D) Total swim distance during probe trials 24 hours (Probe 1) and 14 days (Probe 2) after the last training trial. Total swim distance during the 60-second trial was not different between groups for the immediate probe, but STZ-treated mice swam farther than Controls during the delayed probe (**p < 0.01).
25
Fig 3. STZ treatment did not affect contextual fear conditioning (A) Both Control and STZ-treated mice showed significantly increased freezing in the conditioning chamber 24 hours after training (***p < 0.001), indicating strong context memory. However, STZ-treated mice were significantly less mobile prior to training (#p < 0.05). (B) Difference scores (percent freezing after training minus percent freezing before training) revealed that the net increase in freezing was not different between groups. Fig 4. STZ treatment increased amyloid-containing plaques in the subiculum (A) Top: Photomicrograph of the hippocampal formation from an STZ-treated 3xTg mouse immunostained using 6E10 antibody to illustrate the presence of amyloid precursor protein (APP) and amyloid-containing plaques. Intracellular staining is seen in deep neocortex (very top of image) and in CA1 pyramidal neurons and neurons in the adjacent subiculum. Occasional plaques were seen in the hippocampus proper (e.g., in CA3 and CA1 in the image) but were much more common in the subiculum. Bottom: Higher magnification images of the subiculum from Control and STZ-treated mice. Arrows point to individual plaques. Scale bar = 500 µm. (B) Counts showed that mean plaque numbers in sections containing the subiculum were significantly greater in STZ-treated than in Control mice (**p < 0.01). Fig 5. STZ treatment did not affect phosphorylated tau in the hippocampus Normalized densitometric quantification of hippocampal p-Tau (normalized to β-actin) from untreated 3xTg mice and STZ-treated mice revealed no significant difference in pTau between treatment groups (p = 0.42).
26
Figure 1
A
B
1
Figure 2
2
Figure 3
A
B
3
Plaques/section
Figure 4
A B
4
Figure 5
5