PhIP exposure in rodents produces neuropathology potentially relevant to Alzheimer’s disease

Abstract

Alzheimer’s disease (AD) is a public health crisis due debilitating cognitive symptoms and lack of curative treatments, in the context of increasing prevalence. Thus, it is critical to identify modifiable risk factors. High levels of meat consumption may increase AD risk. Many toxins are formed during meat cooking such as heterocyclic aromatic amines (HAAs). Our prior studies have shown that HAAs produce dopaminergic neurotoxicity. Given the mechanistic and pathological overlap between AD and dopaminergic disorders we investigated whether exposure to 2-amino-1-methyl-6-phenylimidazo [4,5-b] pyridine (PhIP), a prevalent dietary HAA formed during high-temperature meat cooking, may produce AD-relevant neurotoxicity. Here, C57BL/6 mice were treated with 100 or 200 mg/kg PhIP for 8h or 75 mg/kg for 4 weeks and 16 weeks. PhIP exposure for 8h produced oxidative damage, and AD-relevant alterations in hippocampal synaptic proteins, Amyloid-beta precursor protein (APP), and β-Site amyloid precursor protein cleaving enzyme 1 (BACE1). PhIP exposure for 4 weeks resulted in an increase in BACE1. PhIP exposure for 16 weeks resulted in increased hippocampal oxidative damage, APP, BACE1, Aβ aggregation, and tau phosphorylation. Quantification of intracellular nitrotyrosine revealed oxidative damage in cholinergic neurons after 8h, 4 weeks and 16 weeks of PhIP exposure. Our study demonstrates that increase in oxidative damage, APP and BACE1 might be a possible mechanism by which PhIP promotes Aβ aggregation. Given many patients with AD or PD exhibit neuropathological overlap, our study suggests that HAA exposure should be further studied for roles in mediating pathogenic overlap.

1. INTRODUCTION

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by two hallmark pathologies: accumulation of extracellular plaques containing amyloid-beta (Aβ) protein and intracellular neurofibrillary tangles due to hyperphosphorylation of tau (Rajmohan and Reddy 2017). Based on the age of onset and genetic predisposition, AD is classified into two types: early onset and late-onset (Van Cauwenberghe et al. 2016). Early onset, or familial AD (FAD), is a rare type of AD occurring between the ages of 30 and 65 years and is linked to mutations in Amyloid-beta precursor protein (APP), Presenilin 1 (PSEN1), and Presenilin 2 (PSEN2) (Bekris et al. 2010; Sherrington et al. 1995; St George-Hyslop et al. 1987) (Goate et al. 1991; Sherrington et al. 1996). Late-onset (LOAD), or Sporadic AD, accounts for >95% of cases after the age of 65 years; advanced age and carrying the E4 allele of apolipoprotein (APOE4) is the major risk factor (Bekris et al. 2010; Farrer et al. 1997; Liu et al. 2013; Riedel et al. 2016; Saunders et al. 1993; Van Cauwenberghe et al. 2016). Advanced age is the greatest risk factor. Therefore, as the number and proportion of the U.S. population aged 65 and older continue to increase, the number of Americans with Alzheimer’s or other dementias is speculated to grow from 55 million in 2019 to 88 million by 2050 (2019 Alzheimer’s Disease Facts and Figures).

Although AD was discovered over a century ago, curative and disease modifying treatments remain elusive. Multifactorial pathology and etiology conveying different mechanistic pathways could be the reason for no effective therapies for AD (Gong et al. 2018). Recent studies have determined that intervention of several modifiable risk factors, proper management of comorbidities and maintaining a good lifestyle could reduce the risk of dementia by ~35% (Barnes and Yaffe 2011). Therefore, identifying and preventing modifiable risk factors contributing to AD etiology could be an alternative approach. Diet has been investigated for a role in the etiology of neurodegenerative diseases, as dietary factors are potentially encountered in higher doses and more frequently through one’s life span compared to other environmental contaminants (Francis and Stevenson 2013; Grant 2016; Perrone and Grant 2015). The role of the diet in AD has received recent attention (Grant 2016; Perrone and Grant 2015). In a study on dietary risk factors for the prevalence of AD in the Polish population, it was found that AD dietary pattern was characterized by a high intake of meat, butter, high-fat dairy products, eggs, and refined sugar (Hu et al. 2013). Hence, meat consumption could also be a potential risk factor in AD. While meat itself, often in the context of high saturate fat has been investigated, the potential role of preparation has received limited attention. Here, toxins formed during the preparation of meat have thus far not been examined for AD relevance.

Heterocyclic aromatic amines (HAAs), are primarily formed during high-temperature meat cooking and represent a widespread exposure (Matsumoto et al. 1981; Skog et al. 1998). HAAs are formed through the Maillard reaction between amino acids and sugars, producing pyridine, Pyrimidine or pyrazine, which later reacts with creatine in a heat-dependent reaction (Matsumoto et al. 1981; McGrath 1972; Skog et al. 1998). More than 25 HAAs are formed in cooked red meats (Alaejos and Afonso 2011; Gibis 2016). Of dietary HAAs, 2-amino-1-methyl-6-phenylimidazo[4,5-b] pyridine (PhIP) has been extensively studied. PhIP is the most abundant HCAAs isolated from the crust of cooked meat, and its levels may reach ~15 micrograms/kg uncooked meat (~75% of the genotoxic material) (Felton et al. 1986a; Felton et al. 1986b). PhIP and other HAAs are classified as possible human carcinogens (Group 2A or 2B). To date, these compounds have been primarily investigated as mutagens, with our studies being the first to examine selective neurotoxicity (Agim and Cannon 2018; Cruz-Hernandez et al. 2018; Griggs et al. 2014; Lawana et al. 2019). Recently, our group showed that acute systemic PhIP exposure induces alterations in the nigrostriatal dopamine system in rats (Agim and Cannon 2018). We also found that PhIP and its bioactive phase I metabolite, N-OH-PhIP induced loss of dopaminergic neurons and neurite retraction (Cruz-Hernandez et al. 2018; Griggs et al. 2014). Further, we showed that PhIP induced oxidative damage in dopaminergic neurons (Griggs et al. 2014; Lawana et al. 2019). There is extensive mechanistic and pathological overlap between AD and PD, often resulting in significant symptomatic overlap in patients (Sagar 1987; Xie et al. 2014).Thus, we investigated whether exposure to PhIP, a prevalent dietary HAA formed during high-temperature meat cooking, may also produce AD-relevant neurotoxicity.

2. MATERIALS AND METHODS

2.1. Animals.

C57BL/6 mice were obtained from Envigo (Indianapolis, USA) or bred in house were used for the study. All mice were housed with access to food and water ad libitum, under controlled temperature and 12 h/12 h light-dark cycle. Animal studies were approved by the Purdue Animal Care and Use Committee.

2.2. PhIP treatment and Dose Rationale.

To study the acute, sub-acute, and sub-chronic effects of PhIP, C57BL/6 mice were exposed to PhIP for 8h, 4 weeks, or 16 weeks. Different doses were used according to duration of exposure (higher doses for acute exposure and lower doses for sub-acute and sub-chronic exposure): 1) 6–7 weeks old male and female C57BL/6 mice were treated with 100 or 200 mg PhIP/kg bw for 8h; 2) 8–9 weeks old male C57BL/6 mice were treated with 75 mg PhIP/kg bw for 4 weeks; 3) 8–9 weeks old male C57BL/6 mice were treated with 75 mg PhIP/kg bw for 16 weeks. Here it should be noted, that while exposure durations are similar to those traditionally utilized for sub-acute and sub-chronic studies, such studies usually involve dosing once/day. Here we chose to dose 3x/week due to our literature review.

An extensive review of the literature was performed to choose the doses as reported in our previous study (Agim and Cannon 2018). The doses were chosen from a review of >70 published cancer bioassays (Agim and Cannon 2018), as similar doses have been shown to induce genotoxicity (Inaguma et al. 2003; Li et al. 2013; Metry et al. 2009). Doses used in published cancer studies are 2,000–40,000 times higher than the average human consumption (5 – 50 ng/kg daily). The rationale for such doses in the present study (as well as the many prior reports) is influenced by several factors: 1. Humans consume numerous HAAs (~30 known, likely many more unknown) vs. PhIP alone (Augustsson et al. 1997; Roemer et al. 2016); 2. PhIP content in the food can vary by >500-fold due to differences in meat types, cooking times, and temperatures (Augustsson et al. 1997; Byrne et al. 1998; Keating and Bogen 2004; Knize et al. 1994; Layton et al. 1995; Zimmerli et al. 2001); 3. Rodents are less efficient in bioconversion to toxic metabolites (described in detail below). Previous studies from our lab found that of the two primary metabolic pathways for metabolism of PhIP, the N-hydroxylation product is far more neurotoxic than ring-hydroxylation (Griggs et al. 2014). N-hydroxylation is at least 13-fold less efficient in rodents vs. humans (rodent CYP1A2 converts far more PhIP to 4′-OH-PhIP than to the genotoxic and neurotoxic N-hydroxylated metabolite, N-OH-PhIP), suggesting that far higher doses may be required in rodents vs. humans to produce neurotoxic metabolites (Cheung et al. 2005; Griggs et al. 2014; Turesky et al. 1998). Thus, rodents may require far higher doses of this single representative HAA to elicit neurotoxicity. Even with this strong rationale for the chosen doses, it should be noted that the current study is focused on experimental neuropathology and far lower doses would be required to test potential human risk for neurodegenerative disease.

PhIP (Toronto Research Chemicals, Ontario, Canada; A617000) was dissolved in corn oil (Sigma, St. Louis, USA; C8237) at a final concentration of 0 (control) or 26 mg/ml for doses of 100 or 200 mg/kg bw, respectively. Suspensions were sonicated in an ultrasound water bath (Branson, 1800) at 37°C until a homogenous solution was formed with no precipitates. PhIP was administered by oral gavage. For oral gavage, 20-gauge feeding tubes at 38mm in length (Instech, Pensilvania, USA; FTP-20-38) attached to 1 ml syringes (BD Biosciences, San Jose, USA; 309657) were used. For the 8h time-point, PhIP was administered only once; for the 4 week and 16 week time-points, PhIP was administered three times per week and animals were sacrificed after 8h, 4 weeks and 16 weeks, respectively. Treatment groups were as follows: for the 8h treatment: 1) vehicle-treated, 2) 100 mg PhIP/kg bw, 3) 200 mg PhIP/kg bw (n = 10 per treatment); for the 4 week treatment: 1) vehicle-treated, 2) 75 mg PhIP/kg bw (n = 10 per treatment), and for the 16 week treatment: 1) vehicle-treated 2) 75 mg PhIP/kg bw (n = 10 per treatment).

2.3. Tissue Collection and Preparation.

Mice were sacrificed by decapitation after 8h, 4 weeks or 16 weeks of PhIP treatment and the brains were immediately collected. The brain was divided into two halves sagittally with the help of a sagittal brain matrix (BASi, West Lafayette, USA; BS-Z 2000C). One hemisphere was dissected, and the hippocampus and cortex were separated and flash-frozen in liquid nitrogen and stored at −80°C until further processing for western blot, acetylcholine (ACh) and acetylcholinesterase (AChE) quantification (only in 8h treatment groups). The other hemisphere was post fixed in 4% paraformaldehyde (ACROS Oganics, New Jersey USA; 416780030) for 48 hours at 4°C for histological examination. Brain tissue was cryoprotected by immersion in 30% sucrose (Fisher scientific Hampton, USA; S5–3) in PBS and coronal brain slices of 35 μm thickness were obtained with a sliding microtome (Thermo Scientific, Dreieich, Germany; Microm HM450,). Tissue sections were immersed in cryoprotectant solutions [0.1 M phosphate buffer containing 30% sucrose (w/v) and 30% ethylene glycol (v/v); (Fisher scientific Hampton, USA; BP230)] and stored at −20°C until use.

2.4. Immunohistochemistry.

Tissue sections were washed with PBS and incubated in antigen retrieval buffer, formic acid (89.8%) (Fisher scientific Hampton, USA; A118P-500) for 10 mins (for Aβ1–42) (Christensen et al. 2009), or citrate buffer for 30 min at 37°C (for Iba-1). Endogenous peroxidases were blocked by incubation in a solution of H2O2 (30%) + Methanol (10%) for 30 mins, tissue sections were washed three times for 10 min each, in PBS. Non-specific binding was blocked by incubation in 5% bovine serum albumin (BSA) (Sigma, St. Louis, USA; A6003–5G) in PBS for 1h. Subsequently, sections were incubated with rabbit anti-Aβ1–42 antibody (1:200, Abcam, Cambridge, USA; AB5078P) or rabbit anti-Ionized calcium-binding adaptor molecule 1 (Iba-1) (1:500, Wako, Osaka Japan; WEE4506), diluted in 1% donkey serum in PBS with 0.3%Triton X for 24h. Tissues were washed and incubated with secondary donkey anti-rabbit biotinylated antibody (1:500 Jackson immunoresearch, West Grove, USA; 711-065-152) for 2h at room temperature. Subsequently, sections were washed and incubated with the avidin-biotin complex (ABC Kit; Vector Laboratories, Burlingame, USA; PK-6100) in PBS-Triton X (0.3%) for 2h according to the manufacturer’s instructions. Finally, antibody binding was visualized with the chromogen 3,3’-Diaminobenzidine (DAB) (DAB Peroxidase Substrate Kit; Burlingame, USA; SK-4100) according to the manufacturer’s instructions. Sections were washed with PBS, mounted onto glass slide and dried overnight. Slides were dehydrated by serial dilution in ethanol (75%, 95%, and 100% two times each for 5 mins) and histoclear (National Diagnostic, USA; HS200) two times for 5 minutes each; tissue sections were coverslipped with histomount (National Diagnostic, USA; HS103).

2.5. Immunofluorescence.

For immunofluorescence staining, sections were washed in PBS and treated with antigen retrieval citrate buffer at 37°C for 30 min. Non-specific binding was blocked by incubation with 5% BSA for 1h, and then tissues were incubated with the following primary antibodies: rabbit anti-Glial Fibrillary Acidic Protein (GFAP) (1:500, Dako, Santa Clara, USA; Z0334), rabbit anti-Iba-1 (Iba-1, 1:500), rabbit anti-p-Tau ser 199/202 (1:250, Invitrogen, Carlsbad, USA; 44–768), rabbit anti-Aβ 1–42 (1:200), rabbit anti-PSD-95 (1:200, Abcam, Cambridge, USA; AB18258), rabbit anti-BACE1 (1:500, Cell Signaling, Danvers, USA; 5606s), rabbit anti-nitrotyrosine (1:500, Millipore, Burlington, USA; 06–284), rabbit anti-nitrotyrosine + goat anti-choline acetyltransferase (ChAT) (1:500, Millipore, Burlington, USA; AB144P), all diluted in 1% donkey serum in PBS with 0.3% Triton X for 48 h at 4°C. Tissue was washed with PBS and incubated with the a selected combination of the following secondary antibodies: Donkey Alexa-488 anti-rabbit (1:500, Jackson Immunoresearch, West Grove, USA; 711-545-152), Donkey CY3 anti-rabbit (1:500, Jackson Immunoresearch, West Grove, USA; 711-165-152), Donkey CY3 anti-Goat (1:500, Jackson Immunoresearch, West Grove, USA; 705-165-147), in PBS-Triton X (0.3%). For PSD-95 and BACE 1 staining, IRDDye 800CW anti-rabbit antibody (1:750, LI-COR, Lincoln, USA; 926–32213) was used. All the sections were washed and mounted on glass slides and dried overnight. Tissues were dehydrated by serial dilution 70% ethanol for 5 min each, 95% ethanol 5 min each, 100% ethanol for 5 min each, and histoclear (National Diagnostics, HS-200) for 5 min each. After drying, they were cover slipped with DPX mounting medium (Electron microscopy science, Hatfield, USA; B512).

2.6. Image acquisition and analysis.

High resolution images were obtained using a Nikon A1R inverse scanning confocal microscope (Nikon Instruments, USA), images were acquired with 40X or 60X oil immersion objective. To quantify oxidative damage in the cholinergic neurons in the striatal region, nitrotyrosine intensity was measured in ChAT positive cholinergic neurons. Prior to acquisition, microscope and detector settings were optimized to avoid saturation. Images were analyzed using Image J 1.49v (NIH, open access). For quantification of nitrotyrosine, images were taken from dentate gyrus and CA3 hippocampal region; APP was imaged in the st. pyramidale in CA 1 and CA3 hippocampal region; p-tau and Aβ1–42 were imaged in the st. pyramidale in CA3 hippocampal region; GFAP positive astrocyte intensity and number of Iba-1 positive microglia were quantified in st. radiatum of the CA1 hippocampal region.

Regions of interest (ROI) were drawn around the CA3 / CA1 hippocampal region, cell bodies in dentate gyrus or striatal region. The mean intensity in each ROI was measured and normalized to the average intensity of the control group. Each ROI around the cell body was considered a single data point, as previously described (Agim and Cannon 2018; Wise and Cannon 2016).

For quantification of PSD-95 and BACE1, immune-labeled sections were scanned at 800nm on an Odyssey Infrared Imaging System (Licor Inc., Lincoln, Nebraska) at a resolution of 21 μm similar our previous reports (Agim and Cannon 2018; Cannon et al. 2013; Lee et al. 2015); Intensities of ROIs surrounding the hippocampus (for PSD-95) and mossy fiber projections (for BACE1) were quantified. The mean intensity in each ROI was measured and normalized to the average intensity of the control group.

Light microscopy images were captured on an Olympus System Microscope Model BX53 with an Olympus digital camera model DP72 (12.8 megapixel, color), using the Olympus cellSens Dimensions software (ver.1.14). For quantification of Aβ1–42 images (DAB stained) were obtained in st. Pyramidale in the CA3 region of the hippocampus. For analysis of the images, tissue background staining was subtracted from the area within the same section (Perez-Cruz et al. 2011; Syeda et al. 2018). The threshold was determined by identifying the best signal-to-noise ratio from at least 4–5 images and averaging threshold values from those images. All images were analyzed using that averaged threshold (Perez-Cruz et al. 2011). ROI were drawn in st. pyramidale in CA3 hippocampal region and the percentage of area with Aβ staining was evaluated by using the plugin “measure” in Image J 1.49v (NIH, open access) as previously reported (Rodriguez-Callejas et al. 2019; Syeda et al. 2018).

Iba-1 positive microglia were imaged in the st. radiatum in the CA1 hippocampal region. Microglia were quantified based on their morphological characteristics, according to previous reports. Microglia were classified as inactive (displaying a slight ramified morphology and small rounded soma) or active (hypertrophic soma and with extensively thick and branched processes) (Rodriguez-Callejas et al. 2019; Syeda et al. 2018). For all staining, three slices per animal, 6–10 animals per group were analyzed.

2.7. Western blot.

Tissue was homogenized in RIPA buffer (10% weight/volume). Homogenates were centrifuged at 13,000 rpm for 30 mins at 4 °C. The purified total protein was quantified by the BCA (Thermo fisher scientific, Waltham, USA; BCA Protein Assay Kit, 23225) method. Samples (40 μg of protein) were loaded on 10% SDS-PAGE and run for 2h at 100V, and then proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane (Biorad, Hercules, USA; 1620177) at 100V for 90 mins at 4 °C. Blots were blocked in LI-COR blocking buffer (LI-COR Biosciences, Lincoln, USA; 927–40000) for 1h at room temperature and incubated overnight at 4 °C in the following primary antibodies: Mouse anti-Actin (1:1000, Millipore, Burlington, USA; MAB1501) Chicken anti-GAPDH (1:1000, Abcam, Cambridge, USA; Ab83956), Rabbit anti-p-Tau 199/202 (1:500), Rabbit anti-PSD-95 (1:1000), Rabbit anti-BACE1 (1:1000) and rabbit anti-Vesicular glutamate transporter 1 (V-GLUT1) (1:1000, Invitrogen, Carlsbad, USA; 43–2400,) antibodies. Membranes were washed with PBS, and then incubated in donkey IRDDye 680CW anti-mouse(1:10,000, LICOR, Lincoln, USA; 926–68072),donkey IRDDye 680CW anti-chicken (1:10,000, LICOR, Lincoln, USA; 925–68075), donkey IRDDye 800CW anti-rabbit -secondary antibodies (1:10,000, LICOR, Lincoln, USA; 926–32213), washed again with PBS and scanned using an Odyssey CLx imaging system (LI-COR Biosciences, Lincoln, Lincoln, USA). Densitometry analysis was performed using NIH Image J software. Values were normalized with Actin or GAPDH and expressed as fold increase.

2.8. Acetylcholine (ACh) and Acetylcholinesterase (AChE) estimation.

Ach was analyzed in the hippocampus and cortex, and Ache was analyzed in the cortex using the Invitrogen Molecular Probes Amplex Acetylcholine /Acetylcholinesterase Assay Kit (Invitrogen, Carlsbad, USA; A12217). Briefly, samples were diluted in 1x reaction buffer; samples were incubated in 400 μM Amplex Red reagent containing 2 U/mL HRP, 0.2 U/mL choline oxidase and 1 U/mL acetylcholinesterase/100 μM acetylcholine. Fluorescence was measured in a microplate reader at 530 nm excitation and 590 nm emission.

2.9. Statistical Analysis.

Data from each endpoint were first subjected to the D’Agostino & Pearson normality test using GraphPad Prism 6 (GraphPad, La Jolla, CA). Datasets with a normal distribution were analyzed by unpaired t-test (2 groups) or one-way ANOVA followed by Tukey’s post-hoc test for multiple comparisons (≥3 groups). Data that were not normally distributed were analyzed by the Mann-Whitney test (2 groups) or Kruskal–Wallis nonparametric test, followed by Dunn’s post-hoc test for multiple comparisons (≥3 groups). For all tests, p<0.05 was deemed significant. Data is presented as mean ± standard error of the mean (SEM).

3. RESULTS.

3.1. PhIP exposure increases oxidative damage in ChAT positive neurons in the striatal region.

AD is associated with cholinergic abnormalities and degeneration of cholinergic neurons (Mufson et al. 2008) to investigate if PhIP treatment affects cholinergic neurons, we quantified nitrotyrosine intensity as a marker of oxidative damage in chat positive neurons in the dorsal striatal region. Quantification revealed statistically significant differences in nitrotyrosine from PhIP treatment at 200 mg/kg and 100 mg/kg for 8h compared to the control; this increase was dose dependent (control vs PhIP-100: 100.0 ± 1.838 vs 105.8 ± 2.615, p< 0.01; control vs PhIP-200 100.0 ± 1.838 vs 119.1 ± 2.065, p< 0.001; PhIP-100 vs PhIP-200: 105.8 ± 2.615 vs 119.1 ± 2.065, p< 0.05) (FIG. 1A).

Figure 1. PhIP exposure increases oxidative damage in striatal ChAT positive neurons and ChAT negative cells.

Mice were exposed to PhIP for 8h (100 or 200 mg/kg) (A), 4 weeks (75 mg/kg) (B) and 16 weeks (75 mg/kg) (C). Tissue sections from the control and PhIP treated mice were stained for nitrotyrosine (green) and ChAT (red). Scale bar = 50 μm. To quantify oxidative damage, ROIs were drawn surrounding cell bodies of cholinergic neurons (ChAT), and nitrotyrosine intensity in each ROI was quantified. Cells with no ChAT staining were considered ChAT negative cells. Data were normalized to the mean of the control and reported as percent increase compared to the control group. Data are presented as the mean ± SEM; Data was analyzed using a Kruskal-Wallis non-parametric test, followed by Dunn’s multiple comparisons post hoc test for the 8h treatment group; a Mann-Whitney test was used for both the 4 week and 16-week treatment groups. *p< 0.05, **p< 0.01, ***p< 0.001; n = 120–200 cell bodies from 6–8 animals/treatment (ChAT positive neurons) and n = 600–1000 cell bodies from 6–8 animals/treatment (ChAT negative cells).

Similar effects were seen after PhIP treatment for 4 weeks (control vs PhIP 100 ± 3.378 vs 115 ± 4.350, p< 0.01, p< 0.01) and 16 weeks (control vs PhIP 100 ± 2.460 vs 113.5 ± 1.714, p< 0.001) in ChAT positive neurons (FIG. 1B, 1C). Thes data indicate that cholinergic neurons are susceptible to PhIP induced oxidative damage at acute, sub-acute and sub-chronic exposures.

ChAT-negative cells in this region were susceptible to PhIP induced oxidative damage only on 4 and 16 week exposure: 4 weeks (control vs PhIP 100.0 ± 1.564 vs 114.6 ± 1.437, p< 0.001); 16 weeks (control vs PhIP 100.0 ± 0.9344 vs 118.9 ±1 1.368, p< 0.001). (FIG. 1A).

3.2. PhIP exposure does not alter ACh and AChE.

Oxidative damage in cholinergic neurons (striatal region) was not associated with alterations in ACh level or AChE activity. No significant change was found in Ach levels in the hippocampus or cortex; AChE activity in the cortex also remained unaltered after PhIP treatment for 8h (FIG. 2).

Figure 2. PhIP exposure does not alter Ach and Ache.

Mice were exposed to PhIP for 8h (100 or 200 mg/kg); hippocampus and cortex were obtained. Acetylcholine (ACh) was measured in the hippocampus (A) and cortex (B) and Acetylcholine esterase (AChE) activity (C) was measured in the cortex by the Invitrogen Molecular Probes Amplex Ach/Acetylcholinesterase Assay Kit. ACh and AChE activity were normalized to the amount of protein per sample. The data are presented as the mean ± SEM; n = 6–9 animals/treatment.

3.3. PhIP exposure increases oxidative damage in hippocampus.

Previous research from our lab has shown that PhIP induces oxidative damage in PD relevant brain regions/neurons (Agim and Cannon 2018). Oxidative damage is also an important factor contributing to the pathology of AD (Zhao and Zhao 2013). To quantify oxidative damage, nitrotyrosine intensity was analyzed in the hippocampus. PhIP treatment at 200 mg/kg for 8h, significantly increased nitrotyrosine intensity in the CA3 region (control vs P-200: 100.0 ± 6.833 vs 136.2 ± 12.65, p< 0.05) and cell bodies in the dentate gyrus of the hippocampus (control vs PhIP-100: 100.0 ± 2.386 vs 116.0 ± 3.651, p< 0.01; control vs PhIP-200: 100.0 ± 2.386 vs 123.1 ± 3.588, p< 0.001) (FIG. 3A). No oxidative damage was observed in the hippocampus of PhIP treated mice after 4 weeks (FIG. 3B).

Figure 3. PhIP exposure increases oxidative damage in the hippocampus.

Mice were exposed to PhIP for 8h (100 or 200 mg/kg) (A), 4 weeks (75 mg/kg) (B) and 16 weeks (75 mg/kg) (C). Coronal hippocampal sections from the control and PhIP-treated mice were stained for nitrotyrosine. Scale bar = 50 μm. ROIs were drawn in the CA3 region and cell bodies in the dentate gyrus (DG) of the hippocampus to quantify nitrotyrosine staining. Data is normalized to the average of the control group and reported as percent increase compared to the control group. The data are presented as the mean ± SEM; data was analyzed using a Kruskal-Wallis non-parametric test, followed by Dunn’s multiple comparisons post hoc test for the 8h treatment group, and an unpaired t-test was used for both 4 week and 16-week treatment groups. *p< 0.05, **p< 0.01, ***p< 0.001; n = 45–80 cell bodies from 6–8 animals/treatment.

Similar to the acute exposure, mice treated with PhIP for 16 weeks also had oxidative damage in the CA3 region (control vs PhIP: 100 ± 23.08 vs 183.3 ± 27.10, p< 0.05) and cell bodies in the dentate gyrus of the hippocampus (control vs PhIP: 100.0 ± 3.629 vs 118.4 ± 4.614, p< 0.01) (FIG. 3C).

3.4. PhIP exposure increases synaptic proteins in the hippocampus.

To assess whether PhIP-induced neurotoxicity causes synaptic alterations, we analyzed synaptic proteins such as PSD-95 and V-GLUT1 in the hippocampus. There was a significant increase in PSD-95 in the hippocampus upon PhIP treatment at 200 mg/kg and 100 mg/kg for 8h (control vs PhIP-100: 100.0 ± 2.118 vs 112.4 ± 2.493 p< 0.01; control vs PhIP-200 100.0 ± 2.118 vs 113.2 ± 3.029, p< 0.01) (FIG. 4A)

Figure 4. PhIP exposure increases synaptic proteins in the hippocampus.

Mice were exposed to PhIP for 8h (100 or 200 mg/kg) (A, B), 4weeks (75 mg/kg) (C) and 16 weeks (75mg/kg) (D). Coronal hippocampal sections from control and PhIP treated mice were stained for PSD-95 and scanned by an Odyssey Infrared Imaging System. Scale bar = 1 mm. To quantify PSD-95 staining, ROIs were drawn around the hippocampus. Data is normalized to the average of the control group and reported as percent increase compared to the control group. PSD-95 and V-GLUT1 were measured in the hippocampus of mice exposed to PhIP for 8h by western blot (B). The data are presented as the mean ± SEM. Data was analyzed using One-way ANOVA, followed by Tukey’s multiple comparisons post hoc test for the 8h treatment groups. *p< 0.05, **p< 0.01; n = 6–10 animals/treatment.

We also performed western blot analysis on hippocamapal samples from PhIP 8h exposure groups. Consistent with histological findings, we found a significant increase in PSD-95 at both 200 mg/kg and 100 mg/kg (control vs PhIP-100: 0.7967 ± 0.04018 vs 1.177 ± 0.05067 p< 0.05; control vs PhIP-200: 0.7967± 0.04018 vs 1.173 ± 0.1478, p< 0.05). To confirm increases in synaptic proteins after PhIP exposure, we analysed another synaptic marker, V-GLUT1, and found a similar increase after PhIP treatment at 100 mg/kg and 200 mg/kg for 8h (control vs PhIP-100: 0.7769 ± 0.04921 vs 1.117 ± 0.06972, p< 0.05; control vs PhIP-200: 0.7769 ± 0.04921 vs 1.211 ± 0.09097, p< 0.01) (FIG. 4B).

PhIP exposure at both longer time points (4 weeks and 16 weeks) had no effect on synaptic protein (FIG. 4C–D).

3.5. PhIP exposure does not induce inflammation in the hippocampus.

PhIP exposure did not induce overt inflammation in the hippocampus, as no significant change in GFAP intensity (FIG. 5) or the number of active or total microglia (FIG. 6) was observed in 8h, 4 week, or 16 week treatments.

Figure 5. PhIP exposure does not induce astrocytic activation in the hippocampus.

Mice were exposed to PhIP for 8h (100 or 200 mg/kg) (A), 4 weeks (75 mg/kg) (B) and 16 weeks (75 mg/kg) (C). Coronal hippocampal sections from the control and PhIP treated mice were stained for GFAP. Scale bar = 50μm. GFAP intensity was quantified in the stratum radiatum in the CA1 region of the hippocampus. The data are presented as the mean ± SEM; n = 6–10 animals/treatment.

Figure 6. PhIP exposure does not induce microglial activation in the hippocampus.

Mice were exposed to PhIP for 8h (100 or 200mg/kg) (A), 4 weeks (75mg/kg) (B) and 16 weeks (75mg/kg) (C). Coronal hippocampal sections from the control and PhIP treated mice were stained for Iba-1. Scale bar = 50 μm. Iba-1 positive microglia were quantified in the stratum radiatum in the CA1 region of the hippocampus. Microglia were classified as inactive or active based on morphological phenotype. The data are presented as the mean ± SEM; Data was analyzed using. n = 6–8 animals/treatment.

3.6. PhIP exposure increases BACE1 in the hippocampus.

Upregulation of BACE1 expression is attributed to oxidative stress (Tan et al. 2016), given the observed increase in oxidative stress, we analyzed BACE1 expression in the hippocampal mossy fiber terminals.

Histological analysis of BACE1 provides evidence of significant increase in BACE1 in the hippocampal mossy fiber terminals upon PhIP treatment at 200 mg/kg for 8h (control vs PhIP-200: 100 ± 3.047 vs 115.3 ± 3.584, p< 0.01) (FIG. 7 A), and both 4 weeks (control vs PhIP: 100.0 ± 3.619 vs 113.1±2.626, p< 0.01) (FIG. 7C) and 16 weeks (control vs PhIP: 100.0 ± 2.208 vs 106.7 ± 2.324, p< 0.05) (FIG. 7D). We also performed western blot analysis on hippocampal samples from PhIP 8h exposure groups and, consistent with histological findings, we observed a significant increase in BACE-1 protein in mice exposed to PhIP at 200 mg/kg for 8h (control vs. PhIP-200: 0.5683 ± 0.09266 vs. 1.114 ± 0.1088, p< 0.05) (FIG. 7B).

Figure 7. PhIP exposure increases BACE1 in the hippocampus.

Mice were exposed to PhIP for 8h (100 or 200 mg/kg) (A, B), 4weeks (75mg/kg) (C) and 16 weeks (75 mg/kg) (D). Coronal hippocampal sections from control and PhIP treated mice were stained for BACE1 and scanned by an Odyssey Infrared Imaging System Scale bar = 1 mm. ROIs were drawn around hippocampal mossy fiber for the quantification of BACE 1 staining. Data is normalized to the average of the control group and reported as percent increase compared to the control group. BACE1 was also measured in the hippocampus of mice exposed to PhIP for 8h by western blot. The data are presented as the mean ± SEM; Data was analyzed using One-way ANOVA, followed by Tukey’s multiple comparisons post hoc test for 8h treatment group, and an unpaired t-test was used for both the 4 week and 16-week treatment groups. *p< 0.05, **p< 0.01; n = 6–10 animals /treatment.

3.7. PhIP exposure increases APP and Aβ (1–42) in the hippocampus.

Given the observed changes in BACE1, involved in the processing of APP, after exposure to PhIP for 8h, 4 weeks, and 16 weeks, we performed histological staining for APP. We analyzed APP in the CA1 and CA3 of hippocampus in mice treated with PhIP and found significant increases in APP at 200 mg/kg for 8h (control vs PhIP-200: 100 ± 3.389 vs 113.2 ± 3.861, p< 0.05) and after 16 weeks (control vs PhIP: 100.0 ± 4.804 vs 112.5±3.538, p< 0.05) (FIG. 8A, 8C).

Figure 8. PhIP exposure increases APP and Aβ (1–42) in the hippocampus.

Mice were exposed to PhIP for 8h (100 or 200 mg/kg) (A), 4 weeks (75 mg/kg) (B) and 16 weeks (75 mg/kg) (C, D). Coronal hippocampal sections from control and PhIP treated groups were stained for amyloid precursor protein (APP) Scale bar = 50μm and 10 μm (Aβ 1–42 DAB staining). ROIs were drawn in the CA3 region of the hippocampus for quantification of the APP staining. Coronal hippocampal sections from control and 16 week PhIP treated animals were stained for amyloid-beta protein (Aβ 1–42), ROIs were drawn in the CA3 region of the hippocampus and % area stained with Aβ 1–42, was quantified. For microscopy contrast, neuropathology staining for Aβ was conducted using coulometric development using DAB and shown in brown. Data is normalized to the average of the control group and reported as percent increase compared to the control group. The data are presented as the mean ± SEM; Data was analyzed using one-way ANOVA, followed by Tukey’s multiple comparisons post hoc test for the 8h treatment group and an unpaired t-test was used for 16-week treatment groups. *p< 0.05, ***p< 0.001; n = 5–8 animals/treatment, for DAB staining n = 3 animals/treatment.

No significant differences were found in CA1 region on exposure to PhIP for 8h (control vs PhIP-100: 100.0 ± 3.722 vs 108.9 ± 3.765 control vs PhIP-200: 100.0 ± 3.722 vs 110.4 ± 5.173) and 16 weeks (control vs PhIP: 100.0 ± 5.607 vs 103.9 ± 5.942).

There was no significant change in APP after PhIP treatment for 4 weeks (FIG. 8B). In mice treated with PhIP for 16 weeks, increase in BACE1 and APP was accompanied by a significant increase in Aβ1–42 (For immunofluorescence staining: control vs PhIP: 100.0 ± 4.509 vs 112.5 ± 2.107, p< 0.05; For DAB staining: control vs PhIP: 100.0 ± 24.46 vs 1320 ± 199.7, p< 0.001) (FIG 8D).

3.8. PhIP exposure increases the phosphorylation of tau in the hippocampus.

Given the Aβ-related pathology observed after PhIP exposure, we analyzed phosphorylation of tau (one of the hallmarks of AD) in the CA3 region and dentate gyrus of the hippocampus. There was no change in phosphorylation of tau after PhIP treatment for 8h and 4 weeks (FIG 9A, 9C). p-Tau ser 199/202 was also measured in the hippocampus of mice exposed to PhIP for 8h by western blot and, consistent with histological findings, no change was observed in phosphorylated tau (FIG 9B).

Figure 9. PhIP exposure increases the phosphorylation of tau in the hippocampus.

Mice were exposed to PhIP for 8h (100mg/kg and 200mg/kg) (A, B), 4weeks (75mg/kg) (C) and 16 weeks (75mg/kg) (D Coronal hippocampal tissues sections from control and PhIP treated mice were stained for p-Tau ser 199/202. ROI was drawn in CA3 region of the hippocampus to quantify p-Tau 199/202 staining. Data is normalized to the average of the control group and reported as percent increase compared to the control group. The data are presented as the mean ± SEM; Data was analyzed using unpaired t-test. *p< 0.05, n = 6–10 animals/treatment

However, PhIP treatment for 16 weeks increased phosphorylation of tau in the CA3 region (control vs PhIP: 100.0 ± 4.785 vs 118.1 ± 5.088, p< 0.05) (FIG 9D). There was no change in phosphorylation of tau in dentate gyrus (control vs PhIP: 100.0 ± 5.935 vs 115.2 ± 6.695).

4. DISCUSSION

Specific dietary components have been implicated in the etiology of AD, such as meat, however, the role of toxins formed during food preparation have not been studied. Here we report that exposure to PhIP, a prevalent dietary HAA formed during high-temperature meat cooking, produces AD relevant neurotoxicity. It is expected that these findings will prompt more detailed mechanistic investigations to further examine the potential role of HAA’s and other toxins formed during cooking in AD etiology.

Acute, sub-acute, and sub-chronic PhIP exposure induced oxidative damage in ChAT-positive neurons of the striatum neurons, as evidenced by increased nitrotyrosine levels. In some cases, the increases were modest (<10%), yet statistically significant. Here, in vitro studies may be needed to reveal whether such increases are sufficient to disrupt cholinergic neuron function. Cholinergic neuron function is especially important in AD. Cholinergic neurons of the striatum are predominantly large interneurons that interact with the dopaminergic innervation of the striatum (Clarke and Adermark 2015; Gonzales and Smith 2015; Holt et al. 1996) Striatal cholinergic dysfunction has been reported in AD, PD, Huntington’s disease, and dystonia (Aubert et al. 1992; Calabresi et al. 2000; Court et al. 2000; Eskow Jaunarajs et al. 2015; Gilman et al. 2010; Gotti et al. 2006). Here, oxidative damage in cholinergic neurons was not associated with alterations in ACh levels in the hippocampus or cortex or AChE activity in the cortex after PhIP treatment for 8h. This could be because the hippocampus and cortex are innervated by cholinergic neurons from other regions of the brain (Blusztajn and Rinnofner 2016). It will be important to examine oxidative damage in cholinergic neurons in other brain regions and ACh level and AChE activity on sub-acute and sub-chronic exposure in future studies. Acute and sub-chronic PhIP exposure also induced oxidative damage in the CA3 region and dentate gyrus of the hippocampus. Cruz-Sánchez et al., have shown that the oxidative stress pathway starts in the CA3 and progresses to other hippocampal areas in AD progression (Cruz-Sanchez et al. 2010). This study along with our prior report suggests that systemic PhIP exposures increases oxidative damage in all measured neuronal subpopulations, though DA neurons exhibit heightened levels of damage (Agim and Cannon 2018). Such a widespread effect could explain the presence of overlapping AD and PD neurotoxicity and neuropathology. Data from our prior report focused on the nigrostriatal dopamine system and this report, together, suggest neuronal dysfunction in the absence of overt cell loss. This may explain the lack of glial activation observed in this study.

Increased levels of reactive oxygen species (ROS) and subsequent oxidative damage is an important factor contributing to the initiation and progression of AD (Zhao and Zhao 2013). There are multiple potential sources of the oxidative damage observed here: PhIP itself may be metabolized to reactive species; 4’-OH PhIP and N-OH-PhIP are 2 major PhIP metabolites formed in rodents (Chen et al. 2007; Malfatti et al. 1999; Turteltaub et al. 1999; Wang et al. 2015), with N-OH-PhIP being a genotoxic metabolite capable of forming covalent adducts to DNA and protein (Nguyen and Novak 2007; Peng et al. 2012; Turesky and Le Marchand 2011). In contrast, 4’-OH-PhIP is viewed as a detoxified metabolite with regard to its genotoxicity (Holme et al. 1989). Several studies in various disease models suggest that HAAs have pro-oxidant effects (Haza and Morales 2011; Karim et al. 2003; Sammi et al. 2018). Previous research from our lab has demonstrated that pre-treatment by an antioxidant, N-acetylcysteine, reduces PhIP-mediated oxidative stress in midbrain primary culture and neuromelanin forming SH-SY5Y cells (Griggs et al. 2014; Lawana et al. 2019).

Alterations in hippocampal glutamatergic neurotransmission have been reported in AD brains. Early stages of AD show an increased excitability of pyramidal neurons and an upregulation of glutamatergic presynaptic, whereas, later stages of AD show markedly decreased glutamatergic activity (Bell et al. 2007; Busche et al. 2012; Hascup et al. 2019; Hascup and Hascup 2015). To assess whether PhIP-induced neurotoxicity alters glutamatergic synaptic transmission we analyzed synaptic proteins. Synapses are an early site of pathology in AD (Alonso-Nanclares et al. 2013; Bell et al. 2007; Coleman and Yao 2003) and the strongest pathological correlate of cognitive ability in AD (DeKosky and Scheff 1990; Robinson et al. 2014; Terry et al. 1991). Here, we analyzed PSD-95 and V-GLUT1 in the hippocampus which is predominated by glutamatergic large pyramidal neurons. PSD-95 is the major scaffold protein of the excitatory postsynaptic density, where it functions to cluster proteins such as glutamate receptors on the postsynaptic membrane (Keith and El-Husseini 2008; Okabe 2007). Acute PhIP treatment increased PSD-95 expression in the hippocampus. However, no change was found after PhIP treatment for 4 or 16 weeks. We also analyzed presynaptic glutamatergic protein, V-GLUT1, and found a similar increase after PhIP treatment. Expression level of V-GLUTs reflects the amount of glutamate in vesicles (Herman et al. 2014; Johnson et al. 2004; Wojcik et al. 2004). PSD-95 and VGLUT1 expression are reported to alter in AD and PD patients (Fourie et al. 2014; Kashani et al. 2007; Leuba et al. 2008; Poirel et al. 2018; Preissmann et al. 2012). An increase in PSD-95 and V-GLUT1 observed in the present study indicates a possible increase in glutamatergic signaling upon acute PhIP exposure.

The excessive generation and aggregation of Aβ is a major pathological event in the AD (Gravina et al. 1995; Masters et al. 1985; Murphy and LeVine 2010; Sun et al. 2015). Many studies have reported that oxidative stress increases Aβ generation and aggregation (Apelt et al. 2004; Goldsbury et al. 2008; Muche et al. 2017; Quiroz-Baez et al. 2009; Shen et al. 2008; Swomley et al. 2014). Here, we found that sub-chronic PhIP exposure leads to an increase in Aβ 1–42 in the CA3 region of the hippocampus; Aβ 1–42 is the most toxic and fibrillogenic form of amyloid-beta deposited in the brain (Gravina et al. 1995; Jarrett et al. 1993; Suzuki et al. 1994). Aβ peptides are primarily generated from the cleavage of APP. Importantly, we also observed APP accumulation in the CA3 region of the hippocampus after acute and sub-chronic PhIP exposure.

APP processing and subsequent Aβ generation are dependent on secretase enzymes (De Strooper and Annaert 2000; Esler and Wolfe 2001; Nunan and Small 2000). APP processing is generally divided into two pathways, non-amyloidogenic and amyloidogenic. α-secretase-mediated cleavage of APP leads to the production of non-amyloidogenic products, whereas β-secretase (BACE1)-mediated APP cleavage leads to the production of amyloidogenic products (Aβ) (Babes et al. 1975; Cai et al. 2001; Lichtenthaler 2011, 2012; Swomley et al. 2014).

Oxidative stress has been reported to decrease the activity of α-secretase while promoting the expression and activity of BACE1. There is a significant correlation between BACE1 activity and oxidative markers in AD (Apelt et al. 2004; Borghi et al. 2007; Quiroz-Baez et al. 2009; Tamagno et al. 2008).. We analyzed BACE1 expression in hippocampal mossy fiber terminals, a site of high BACE1 concentration (Hitt et al. 2012; Laird et al. 2005; Zhao et al. 2007). Histological analysis of BACE1 provides evidence of a significant increase in BACE1 upon acute, sub-acute, and sub-chronic PhIP exposure. Further, in vitro mechanistic studies may be better suited to examining how PhIP alters APP and BACE1. It is worth studying if PhIP alters transcription or trafficking and degradation of APP and BACE1.

Aberrant phosphorylation is a key feature of tau isolates from the brains of AD and PD patients (Bancher et al. 1989; Blazquez-Llorca et al. 2011; Iqbal et al. 1986). Tau promotes microtubule assembly and maintains microtubule structure and its activity is regulated by the degree of phosphorylation (Alonso et al. 1994; Kosik 1993; Lindwall and Cole 1984). Sub-chronic PhIP exposure increased the phosphorylation of tau at Ser199/Ser202 in the CA3 region of the hippocampus. Ser199/Ser202 is one of the major site for Phosphorylation of tau in AD brain (Ikura et al. 1998; Kopke et al. 1993). No change was observed in phosphorylation of tau after acute and sub-acute PhIP exposure. Furthermore, in vitro mechanistic studies may be better suited to examine other phosphorylation sites of tau and to study which protein kinases PhIP activates to increase phosphorylation of tau.

Many studies have demonstrated that the normally resistant hippocampal CA3 region becomes hypersensitive in sporadic AD models (Zhang et al. 2013). The CA3 region is critical for storage and retrieval of information, particularly with regard to working memory and alteration in this region might affect the cognition (Florian and Roullet 2004; Kesner 2007). In the present study, changes in APP, Aβ, and p-tau were observed in the CA3 region whereas no significant differences were found in CA1 or dentate gyrus, indicating that CA3 region is more susceptible and first region to be affected on PhIP exposure.

We also observed a difference in response to acute, sub-acute and sub-chronic PhIP exposure. Acute exposure increased glutamatergic synaptic protein, while this effect was not observed in sub-acute and sub-chronic exposure. Possible explanation for this could be difference in the doses, high dose of PhIP (100 or 200 mg/kg) for acute period might be enough to accumulate in brain and cause alterations in glutamatergic signaling. Whereas oral gavage 3 times a week for 4 weeks or 16 weeks, might not be enough to obtain a substantial increase in PhIP accumulation in the brain enough to alter glutamatergic signalling, as most of the PhIP is excreted in urine in the first 6 hours after exposure, the levels in tissues tested, such as liver, kidney and stomach, are also very low after 24h (Turteltaub et al. 1992). Rapid excretion of PhIP decreases bioavailability in organs such as brain (Teunissen et al. 2010). It is worth studying if increase in glutamatergic synaptic protein on acute PhIP exposure leads to excitotoxicity. BACE 1 and APP was found to increase on acute and sub-chronic PhIP exposure, while accumulation of Aβ was observed only in Sub-chronic PhIP exposure. Accumulation of Aβ is dependent upon increase in BACE 1 mediated processing of APP and clearance and degradation of Aβ (Baranello et al. 2015). Although BACE 1 and APP was found to increase on acute exposure, clearance or degradation pathway might be intact and preventing accumulation of Aβ. It is worth noting that, in general, at the 8-hour timepoint effects were dose dependent. However, at later time-points dose-dependence was not tested because a single dose was used for these exposures. These observations suggest potential adaptive responses and the both lower doses and a wider dose range should be included in future studies.

The main limitations of this study that will need to be addressed in future studies are: 1) whether PhIP exposure influences AD-relevant learning and memory endpoints; 2) While no sex differences was observed between male and female responses after acute exposure, it is important to study the effects of long term PhIP exposure on female mice, given higher AD prevalence in women and 3) the use of lower-dose, longer term exposures that bear more relevance to human exposures.

Overall, our data indicates that despite AD and PD being clinically distinct entities, with respect to HAA neurotoxicity, there is a mechanistic and pathological overlap between resultant neurotoxicity and neuropathology. Understanding these overlapping mechanisms may help us to identify neuroprotective strategies that would be beneficial in both AD and PD.

5. CONCLUSIONS

In summary, our study demonstrates that PhIP exposure induces neurotoxicity to amyloid and tau biology. Lack of glial activation suggests that alterations in oxidative stress are early toxic events in response to PhIP exposure. This study also implies that an increase in oxidative stress, APP, and BACE1 might be a possible mechanism by which PhIP promotes Aβ aggregation. Collectively, our study suggests a potential link between a diet enriched in HAAs and AD. Our results can also lead to further research connecting meat consumption and neurodegenerative diseases.

ABBREVIATIONS:

Ach

Acetylcholine

AChE

Acetylcholinesterase

AD

Alzheimer’s disease

APP

Amyloid-beta precursor protein

APOE4

Apolipoprotein

BACE1

β-Site amyloid precursor protein cleaving enzyme 1

BSA

Bovine serum albumin

HAA

heterocyclic aromatic amine

ChAT

Choline acetyltransferase

DAB

3,3’-Diaminobenzidine

FAD

familial AD

GFAP

Glial Fibrillary Acidic Protein

LOAD

Ionized calcium-binding adaptor molecule 1, Iba-1, Late-onset

NT

Nitrotyrosine

PD

Parkinson’s disease

PhIP

2-amino-1-methyl-6-phenylimidazo [4,5-b]pyridine

PSEN1

Presenilin 1

PSD-95

Postsynaptic density protein 95

V-GLUT1

Vesicular glutamate transporter 1